Split (1)

train · ~9.11M rows (showing the first 93.5k)

identifier stringlengths 11 32	pdf_url stringlengths 17 4.62k ⌀	lang stringclasses 120 values	error stringclasses 1 value	title stringlengths 2 500 ⌀	source_name stringlengths 1 435 ⌀	publication_year float64 1.9k 2.02k	license stringclasses 3 values	word_count int64 0 1.64M	text stringlengths 1 9.75M
https://openalex.org/W4250124314	https://opus.lib.uts.edu.au/bitstream/10453/148090/2/document.pdf	English	null	Microbial Predation Accelerates Granulation and Modulates Microbial Community Composition	Research Square (Research Square)	2,021	cc-by	13,201	Microbial predation accelerates granulation and modulates microbial community composition Siew Herng Chan1,2†, Muhammad Hafiz Ismail1,3†, Chuan Hao Tan1,4, Scott A. Rice1,3,5* and Diane McDougald1,5* Siew Herng Chan1,2†, Muhammad Hafiz Ismail1,3†, Chuan Hao Tan1,4, Scott A. Rice1,3,5* and Dia Siew Herng Chan1,2†, Muhammad Hafiz Ismail1,3†, Chuan Hao Tan1,4, Scott A. Rice1,3,5* and Diane McDougald1,5* Chan et al. BMC Microbiology (2021) 21:91 https://doi.org/10.1186/s12866-021-02156-8 Chan et al. BMC Microbiology (2021) 21:91 https://doi.org/10.1186/s12866-021-02156-8 Open Access * Correspondence: rscott@ntu.edu.sg; Diane.McDougald@uts.edu.au †Siew Herng Chan and Muhammad Hafiz Ismail contributed equally to this work. * Correspondence: rscott@ntu.edu.sg; Diane.McDougald@uts.edu.au †Siew Herng Chan and Muhammad Hafiz Ismail contributed equally to this work. 1Singapore Centre for Environmental Life Sciences Engineering, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore Full list of author information is available at the end of the article Abstract Background: Bacterial communities are responsible for biological nutrient removal and flocculation in engineered systems such as activated floccular sludge. Predators such as bacteriophage and protozoa exert significant predation pressure and cause bacterial mortality within these communities. However, the roles of bacteriophage and protozoan predation in impacting granulation process remain limited. Recent studies hypothesised that protozoa, particularly sessile ciliates, could have an important role in granulation as these ciliates were often observed in high abundance on surfaces of granules. Bacteriophages were hypothesized to contribute to granular stability through bacteriophage-mediated extracellular DNA release by lysing bacterial cells. This current study investigated the bacteriophage and protozoan communities throughout the granulation process. In addition, the importance of protozoan predation during granulation was also determined through chemical killing of protozoa in the floccular sludge. Results: Four independent bioreactors seeded with activated floccular sludge were operated for aerobic granulation for 11 weeks. Changes in the phage, protozoa and bacterial communities were characterized throughout the granulation process. The filamentous phage, Inoviridae, increased in abundance at the initiation phase of granulation. However, the abundance shifted towards lytic phages during the maturation phase. In contrast, the abundance and diversity of protozoa decreased initially, possibly due to the reduction in settling time and subsequent washout. Upon the formation of granules, ciliated protozoa from the class Oligohymenophorea were the dominant group of protozoa based on metacommunity analysis. These protozoa had a strong, positive- correlation with the initial formation of compact aggregates prior to granule development. Furthermore, chemical inhibition of these ciliates in the floccular sludge delayed the initiation of granule formation. Analysis of the bacterial communities in the thiram treated sludge demonstrated that the recovery of ‘Candidatus Accumulibacter’ was positively correlated with the formation of compact aggregates and granules. (Continued on next page) Background Aerobic granular sludge is a complex, human engineered ecosystem consisting of highly diverse and functional microbial communities that are utilized for specific bio- logical functions [1, 2]. These densely packed biofilm ag- gregates are typically developed from activated floccular sludge. Using laboratory sequencing batch reactors (SBRs), the formation of aerobic granules from flocs has been improved with the concomitant increased under- standing of the effects of operating conditions such as hydrodynamic shear force, settling time, hydraulic reten- tion time and discharging time [3–7]. In contrast to the impact of physical factors, the bio- logical processes that drive granule formation are less well understood. For example, N-acyl-homoserine-lac- tone (AHL) mediated quorum sensing was found to positively correlate with the formation of granules from floccular sludge [8]. Furthermore, the addition of AHLs to the SBR markedly increased the production of EPS, which mediates contact between bacterial cells [8, 9]. Other biological factors such as predation have been demonstrated to enhance biofilm formation for several bacterial species [10–12]. Predation on those free-living bacteria may therefore represent a strong pressure selecting for bacteria that are tightly embedded in aggre- gates of biomass. Bacteriophages are highly abundant in engineered wastewater systems, appear to be active com- ponents of activated sludge systems and are able to in- fect both planktonic and biofilm associated bacterial cells [13–15]. Here, bacteriophage, protozoan and bacterial commu- nities were characterized throughout the aerobic granu- lation process. The dynamics of different bacteriophage families were investigated to elucidate their role in granulation. Additionally, the succession of protozoan communities was tracked during the aerobic granulation process and the inhibition of protozoa was performed to determine the potential role of protozoan predation in driving aerobic granulation. It is hypothesised that protozoan predation can promote the formation of gran- ules through grazing selection pressure and/or via a structural role. Microbial community analysis indicated that there was an increase in the abundance of non-lytic, filamentous Inoviridae bacteriophages during the initi- ation phase of granulation when compact aggregates were formed. In addition, the abundance and diversity of protozoa decreased significantly during the aerobic granulation process. Our results also demonstrated that the absence of protozoa did not negatively affect the for- mation of mature granules, although there was a delay in the formation of compact aggregates in the absence of protozoa. Background Phage-mediated mortality has the potential to influ- ence the treatment performance of a system through controlling the abundance of key functional groups, leading to their utilisation as a biocontrol strategy to lyse filamentous bacteria that are responsible for bulk- ing in activated sludge [16, 17]. In contrast, bacterio- phage predation in wastewater systems has been demonstrated to cause the collapse of reactors [18] and the failure of bacterial biological processes such as phosphorus removal and nitrification [18, 19]. Bac- teriophage predation has recently been suggested to mediate the release of extracellular DNA via the lysis of bacterial cells, which plays a role in providing structural stability to granules [20]. (Continued from previous page) Conclusion: Predation by bacteriophage and protozoa were positively correlated with the formation of aerobic granules. Increases in Inoviridae abundance suggested that filamentous phages may promote the structural formation of granules. Initiation of granules formation was delayed due to an absence of protozoa after chemical treatment. The presence of ‘Candidatus Accumulibacter’ was necessary for the formation of granules in the absence of protozoa. Keywords: Granulation, Protozoa, Bacteriophage, Metagenomics, Activated sludge, Microbial predation Protozoa are abundant in activated floccular sludge sys- tems and play an important role in the predation of sus- pended bacteria, which aids in the clarification of wastewater effluent [21]. In addition, previous studies of aerobic granulation systems demonstrated an abundance of sessile ciliates on the surface of aerobic granules [21– 24]. Electron microscopy of granular surfaces revealed the attachment of bacteria on the stalks of sessile ciliates [24]. Weber et al. [24] further hypothesised that these sessile ciliates may act as nucleating agents for the attachment of bacteria. Taken together, these studies strongly suggest that protozoan predation may have a role in promoting aerobic granulation. However, the role of protozoa in the formation of aerobic granules remains unclear to date. © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Page 2 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology Development and microscopic observations of aerobic granular sludge Activated floccular sludge was used to seed the SBRs, which were operated under conditions optimal for the aerobic granulation process over a period of 11 weeks. The granulation process has five distinct phases: Chan et al. BMC Microbiology (2021) 21:91 Page 3 of 18 Page 3 of 18 floccular, initiation, maturation, maintenance and dis- persal [8]. Here, only three phases of floccular, initiation and maturation phases were observed (Fig. 1a). size from 108.5 ± 6.9 to 193.0 ± 16.7 μm (50th percentile) (Fig. 1b). In addition, the SVI5 also decreased 44% from 112.5 ± 13.2 to 63.0 ± 6.5 mL g−1 (Fig. 1c). g g By week 7, the sludge biomass had entered Phase III of the aerobic granulation process. The mean particle size of the sludge biomass increased 90% from 193.0 ± 16.7 μm in week 6 to 367.0 ± 68.1 μm in week 7 (50th percentile) (Fig. 1b). The particle size and SVI5 of the sludge biomass con- tinued to increase and decrease, respectively, over the remaining weeks. The MLSS of the sludge steadily in- creased from week 7 onwards (Fig. 1c). Over the entire 11 weeks, the reduction in settling time from 120 to 5 min was linked to the appearance of high density and compact sludge particles. This was associated with a mean particle size increase from 51.3 ± 2.2 to 792.4 ± 130.6 μm (Fig. 1b). Similarly, the SVI5 decreased significantly from 190.8 ± 2.5 to 16.0 ± 2.1 mL g−1 (Fig. 1b). In addition, the MLSS of the sludge also increased from 3.9 ± 0.5 in week 6 to 12.7 ± 0.6 mL g−1 by the end of week 11. These observations indi- cated that the sludge biomass was mostly in granular form. During Phase I, the floccular biomass had a mean par- ticle size of 51.3 ± 2.2 μm (50th percentile) (Fig. 1b). Aer- obic granules are typically defined as dense and compact aggregates characterized by a minimum particle size of 100 μm and a SVI5 of 50 mL g−1 or less [25]. Initial de- creases in settling time from 120 to 56 min resulted in a 10.5% average loss of biomass (MLSS decreased from 5.0 ± 0.1 to 4.1 ± 0.1 g L−1) by the end of week 1 (Fig. 1c). Development and microscopic observations of aerobic granular sludge The SVI5 of the floccular sludge increased from 190.8 ± 2.0 to 221.8 ± 5.4 mL g−1, which indicated poor settling of the floccular sludge (Fig. 1b). By Phase II, compact aggregates were observed in the floccular sludge at week 4 and the mean particle size was 96.2 μm (50th percentile) (Fig. 1b). Subsequent decreases in settling time from 56 to 24 min did not result in a decrease in overall biomass until week 4 (MLSS increased from 4.9 ± 0.4 to 5.1 ± 0.4 g L−1) when the sludge biomass en- tered the Phase II. During weeks 4 to 6 of Phase II, the set- tling time was reduced from 24 to 5 min, which resulted in an average of 23.7% loss of biomass (MLSS decreased from 5.1 ± 0.4 to 3.9 ± 0.5 g L−1) (Fig. 1b). This reduction in set- tling time also coincided with an increase in mean particle Microbial community composition of floccular and granular sludge Here, the total genomic DNA of the granular sludge was sequenced to track the diversity and changes in Fig. 1 Development of granules from floccular sludge. a Development of small granules from floccular sludge over 11 weeks based on microscopic visualizations of sludge samples. b Average particle size distribution and SVI5 in 4 SBRs and 50th percentile (filled square) represent the percentage of total particles below the corresponding size distribution and the compactness of sludge particles as measured by SVI5 (open circle), respectively. c Average sludge biomass concentrations represented by both MLSS (filled circle) and MLVSS (filled squares). Error bars represent standard deviations (n = 4). Magnification × 40 (Bar, 100 μm) Fig. 1 Development of granules from floccular sludge. a Development of small granules from floccular sludge over 11 weeks based on microscopic visualizations of sludge samples. b Average particle size distribution and SVI5 in 4 SBRs and 50th percentile (filled square) represent the percentage of total particles below the corresponding size distribution and the compactness of sludge particles as measured by SVI5 (open circle), respectively. c Average sludge biomass concentrations represented by both MLSS (filled circle) and MLVSS (filled squares). Error bars represent standard deviations (n = 4). Magnification × 40 (Bar, 100 μm) Fig. 1 Development of granules from floccular sludge. a Development of small granules from floccular sludge over 11 weeks based on microscopic visualizations of sludge samples. b Average particle size distribution and SVI5 in 4 SBRs and 50th percentile (filled square) represent the percentage of total particles below the corresponding size distribution and the compactness of sludge particles as measured by SVI5 (open circle), respectively. c Average sludge biomass concentrations represented by both MLSS (filled circle) and MLVSS (filled squares). Error bars represent standard deviations (n = 4). Magnification × 40 (Bar, 100 μm) Page 4 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 that the reactors are not statistically different from each other (P = 0.184) (Table S1). bacteria abundance as granulation takes place over 11 weeks of reactor operation. Clustering based on the relative abundance of the microbial communities sug- gested that in the early floccular stages (weeks 0 and 1), the communities were similar across the 4 SBRs (Fig. S1a). Microbial community composition of floccular and granular sludge However, from week 2, the communities between the reactors diverged, as reflected in changes in the community composition, as the reactors under- went granulation. Despite this, PERMANOVA showed The genus ‘Candidatus Accumulibacter’, which is a polyphosphate accumulating organism (PAO) and nitri- fier from the phylum Proteobacteria, was the most abun- dant, with an average increase from 3.6 to 63.53% by week 11 (Fig. 2). ‘Candidatus Competibacter’ and ‘Can- didatus Contendobacter’, glycogen accumulating organ- isms (GAOs), did not change appreciably in abundance, Fig. 2 Total abundance of bacterial populations in 4 SBRs over 11 weeks of granulation. The top 20 abundant bacterial genus based on metagenome (DNA) reads. All read counts are natural log transformed before analysis bundance of bacterial populations in 4 SBRs over 11 weeks of granulation. The top 20 abundant bacterial genus based on (DNA) reads. All read counts are natural log transformed before analysis Fig. 2 Total abundance of bacterial populations in 4 SBRs over 11 weeks of granulation. The top 20 abundant bacterial genus based on metagenome (DNA) reads. All read counts are natural log transformed before analysis Page 5 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 Bacteriophages exert a complex influence over their microbial hosts and additionally may play a structural component of the matrix [26–28]. Therefore, the rela- tionship between granulation and bacteriophage com- munity dynamics were also investigated here. Only DNA bacteriophages were targeted here and their sequences were assembled into viral contigs to study their relative abundance during granulation (Fig. 3a). Microviridae between 0.97 to 3.11% and 1.47 to 3.8%, respectively (Fig. 2). Nitrifiers, such as Nitrospira, progressively de- creased from 16.45 to 6.06% over the course of the ex- periment. There was a peak of Thauera (a denitrifier) at week 1 at 10.98% but reduced to 3.88% by the end of the experiment. The other members of the top 20 genera generally had a lower abundance with Terrimonas at the lowest between 0.3 and 0.97% (Fig. 2). Fig. 3 Total abundance of predator populations in 4 SBRs over 11 weeks of granulation. a Viral sequences were classified accordingly to family level (b) with corresponding Spearman correlation between viral families and 50th percentile sludge particle size. Microbial community composition of floccular and granular sludge c Distance based redundancy analysis (dbRDA) ordination of the effect of viral dynamics on the changes in bacterial community with aerobic granulation. d Top 10 most abundant eukaryotic OTUs with (e) corresponding Spearman correlation with 50th percentile sludge particle size differentiated into the three phases of granulation. Protozoan OTUs are in bold. All read counts are natural log transformed before analysis. Viral analyses were based on viral metagenome (DNA) reads while eukaryotic OTUs were based on Ribotagger tags (RNA). The correlation matrixes were clustered based on Bray Curtis distance with between-group average linkage (UPGMA) method. The correlation matrixes are color-coded where yellow and blue indicate positive and negative correlations, respectively. Where correlations cannot be made, the tile is coloured white Fig. 3 Total abundance of predator populations in 4 SBRs over 11 weeks of granulation. a Viral sequences were classified accordingly to family level (b) with corresponding Spearman correlation between viral families and 50th percentile sludge particle size. c Distance based redundancy analysis (dbRDA) ordination of the effect of viral dynamics on the changes in bacterial community with aerobic granulation. d Top 10 most abundant eukaryotic OTUs with (e) corresponding Spearman correlation with 50th percentile sludge particle size differentiated into the three phases of granulation. Protozoan OTUs are in bold. All read counts are natural log transformed before analysis. Viral analyses were based on viral metagenome (DNA) reads while eukaryotic OTUs were based on Ribotagger tags (RNA). The correlation matrixes were clustered based on Bray Curtis distance with between-group average linkage (UPGMA) method. The correlation matrixes are color-coded where yellow and blue indicate positive and negative correlations, respectively. Where correlations cannot be made, the tile is coloured white Fig. 3 Total abundance of predator populations in 4 SBRs over 11 weeks of granulation. a Viral sequences were classified accordingly to family level (b) with corresponding Spearman correlation between viral families and 50th percentile sludge particle size. c Distance based redundancy analysis (dbRDA) ordination of the effect of viral dynamics on the changes in bacterial community with aerobic granulation. d Top 10 most abundant eukaryotic OTUs with (e) corresponding Spearman correlation with 50th percentile sludge particle size differentiated into the three phases of granulation. Protozoan OTUs are in bold. All read counts are natural log transformed before analysis. Viral analyses were based on viral metagenome (DNA) reads while eukaryotic OTUs were based on Ribotagger tags (RNA). Microbial community composition of floccular and granular sludge There were 4 protozoan members that were correlated to the bacterial members of the reactor community. OTU01 and OTU07 were negatively correlated to ‘Candidatus Competibacter’ and Dechloromonas while OTU05 and OTU10 were posi- tively correlated to Sulfuritalea and Mycobacterium, and Terrimonas, respectively. Most of the phages, except for Microviridae, were positively correlated to each other, with Podoviridae being the connecting node that is negatively correlated to Verrucomicrobium. The effect of protozoan predation on aerobic granulation was investigated via total RNA sequen- cing as metagenomic sequencing did not yield suffi- cient reads for classification and annotation of eukaryotic sequences beyond the class level (Fig. S2). The abundance of the microbial populations was represented by the number of sequencing reads de- tected per OTU. Mean values were calculated for the number of sequences per OTU to represent the abundance in the four SBRs. A total of 10 OTUs represented approximately 95% of all sequencing reads. Within these 10 OTUs, there were 8 proto- zoan OTUs which were mostly represented by the genus Telotrochidium (OTU02), class Oligohymeno- phorea (OTU01, 03 and 04), genus Arcella and order Salpingoecidae (Fig. 3d). While sequencing provided insights into the eukaryotic communities in the sludge during granula- tion, microscopic observations were also performed to determine the presence of protozoa and other eukary- otes. Microscopic observations of sludge have also been utilized in membrane bioreactors to compliment sequen- cing data observations [30]. Swimming ciliates that were most likely Paramecium spp. were observed within the floccular sludge (Fig. 5a), while sessile ciliates were at- tached to the surfaces of the flocs (Fig. 5b). These cili- ates represent the Oligohymenophorea OTUs detected by sequencing (Fig. 3d). Metazoans such as tardigardes (Fig. 5c) and rotifers (Fig. 5d) were frequently observed in the floccular sludge with crawling ciliates such as Aspidisca sp., circling the Phase I flocs (Fig. 5e). These rotifers were likely to be represented by OTU05 and 08 as identified in the sequencing data (Fig. 3d). These ob- servations clearly indicated that the inoculum floccular sludge had a diverse community of protozoa present prior to seeding into the SBRs. Upon the formation of compact aggregates at Phase II, no swimming ciliates or large eukaryotes were observed, although rotifers were still occasionally present. Microbial community composition of floccular and granular sludge The correlation matrixes were clustered based on Bray Curtis distance with between-group average linkage (UPGMA) method. The correlation matrixes are color-coded where yellow and blue indicate positive and negative correlations, respectively. Where correlations cannot be made, the tile is coloured white Page 6 of 18 Page 6 of 18 Chan et al. BMC Microbiology (2021) 21:91 were the most abundant and present in all samples throughout reactor operation at 17 to 99.9%. At the end of the initiation phase (week 7), Podoviridae and Sipho- viridae began to significantly increase in abundance and at week 9, were the most abundant viral families after Microviridae at 11.15% ± 1.66 and 8.34% ± 0.96%, re- spectively. Inoviridae had an increase in abundance to 0.2% ± 0.05% when the sludge developed into compact aggregates (week 5) and peaked at week 9 at 1.46% ± 0.32% (Fig. 3a). There was a positive correlation between the increasing granule particle size and the viral counts of Siphoviridae, Microviridae and Myoviridae (Fig. 3b). Additionally, a distance based redundancy analysis (dbRDA) was performed to identify covariates which have an effect on the changes in bacterial community using viral family abundance [29]. This analysis sug- gested the Microviridae and Inoviridae viral families had an effect on the changes in bacterial community com- position during the initiation phase (weeks 4 to 7) and maturation phase (weeks 8 to 9), respectively (Fig. 3c). the most abundant eukaryotic members in the sludge bio- mass in all reactors. Testate amoeba, including OTU06 and OTU07, were not detected beyond week 5, by which time compact aggregates had formed. Non-metric multi-dimensional scaling (nMDS) visualization of the eukaryotic communities during granu- lation demonstrated a high level of dissimilarity between the flocs at week 0 and granules at week 1 (Fig. S1c). Based on sludge particle size, the determinant of granula- tion, the majority of the eukaryotic OTUs, except for Sal- pingoecidae (OTU10), were positively correlated with the floccular particle size (Fig. 3e). In contrast, both Oligohy- menophorea (OTU01 and 03) demonstrated a strong posi- tive correlation with the particle size during the initiation and maturation phase. The remaining eukaryotic OTUs had a negative correlation during both the initiation and maturation phases (Fig. 3e). A network analysis was undetaken to identify taxa that were possibly interacting with each other over the period of reactor operation (Fig. 4). Microbial community composition of floccular and granular sludge Upon granule formation at Phase III, the frequency of crawling ciliates decreased significantly, while sessile ciliates were frequently p g g The genus Telotrochidium is a group of free swimming peritrichous ciliates while the genus Arcella, and the fam- ily of Salpingoecidae represent testate amoebae and flagel- lates, respectively (Fig. 3d). The class Oligohymenophorea represents a large class of ciliated protozoa. Both OTU05 and 08 represented rotifers, which are metazoan predators of suspended microorganisms (Fig. 3d). During Phase I, the abundance of Telotrochidium (OTU02) decreased sharply by week 2 and was not detected in most reactors in the following weeks. The family Oligohymenophorea OTU01 also demonstrated gradual decline in abundance from week 0 to 3. Both Oligohymenophorea (OTU 03 and 04) were constantly detected during Phase I in all reactors except in reactor 4, where it was absent at week 03. Sal- pingoecidae (OTU10), of the flagellate family, was also constantly present from Phase I to III. However, as com- pact aggregates and granules formed by Phase II and III respectively, Oligohymenophorea (OTU03 and 04) were Chan et al. BMC Microbiology (2021) 21:91 Page 7 of 18 Chan et al. BMC Microbiology Fig. 4 Microbiome network of the SBR showing potential interactions between the bacteria, protozoa and phage communities. The different groups are differentiated by colour. The lines connecting nodes (edges) are coloured depending on whether the correlation is positive (green) or negative (red) Fig. 4 Microbiome network of the SBR showing potential interactions between the bacteria, protozoa and phage communities. The different groups are differentiated by colour. The lines connecting nodes (edges) are coloured depending on whether the correlation is positive (green) or negative (red) observed on the granule surfaces (Fig. 5f and g). The abundance of sessile ciliates, as determined by micros- copy, were also reflected in the sequencing data where there were increases in Oligohymenophorea associated sequences (i.e OTU01, 03 and 04) in most reactors as granules formed. sizing was obtained by quantitative image analysis. The initial mean sludge particle sizes were 84.36 ± 12.41 μm (Fig. 6b) and by week 2, the control sludge mean particle size was 89.61 ± 5.94 μm, while the treated sludge was significantly smaller, 67.02 ± 2.65 μm, than the control sludge (Fig. 6b). By week 7, the treated sludge was 125.42 ± 10.60 μm, which was similar to the control sludge particles, 122.71 ± 23.00 μm (Fig. 6b). Microbial community composition of floccular and granular sludge By week 8, there was a slight decrease in the control sludge (104.60 ± 17.57 μm), while the thiram treated sludge was significantly larger, 119.36 μm ± 6.05 μm (Fig. 6b). Development of aerobic granules from untreated and thiram treated floccular sludge Six mSBRs were seeded with activated floccular sludge and operated under conditions that were optimal for the aerobic granulation process over a period of 8 weeks. To investigate the role of protozoan predation in aerobic granulation, protozoa were removed from the floccular sludge by the addition of 20 mg L−1 thiram to the mSBRs and DMSO was added as a control. The concen- tration of thiram was previously optimized to minimize any negative effects on the viability of bacteria in the floccular sludge (data not shown). Microscopic observa- tions of control floccular sludge indicated that the con- version of floccular into granular sludge began at week 4 (Fig. 6a). Compact aggregates were observed in the initi- ation phase and these aggregates continued to expand in size. The sludge entered the maturation phase at week 6 and remained in this phase until the end of the experi- ment at week 8 (Fig. 6a). In contrast, thiram treated sludge did not initiate granulation until week 6 and only started to mature by week 8 (Fig. 6a). The SVI5 of the treated sludge was significantly higher than the control sludge from weeks 2 to 4 (Fig. 6c), sug- gesting that the thiram treated sludge was less dense and compact and hence required a longer settling time com- pared to the control sludge. However, from week 5 on- wards, the SVI5 for the thiram treated sludge decreased and was not significantly different from the control sludge. Effects of thiram treatment on microbial communities during aerobic granulation The microbial communities in the two sludge types were compared by metacommunity sequencing of the V5 re- gion of the 16S and 18S rRNA genes using the Ribotag- ger method [31]. A total of 30 OTUs, representing approximately 92% of the eukaryotic communities were selected for analysis. Within the inoculum sludge, the eukaryotic communities were dominated mainly by cili- ated protozoa OTUs, e.g. OTUs 01, 02 and 03 (Fig. 7). As the volumes of the mSBRs were too low to allow for particle sizing by the particle size analyser, particle Page 8 of 18 Chan et al. BMC Microbiology (2021) 21:91 Fig. 5 Micrographs of protozoa and metazoa in floccular and granular sludge. a A swimming ciliate, Paramecium spp. and b sessile ciliates. c Metazoa such as tardigardes and d large rotifers from the genus Euchlanis. e Crawling ciliates (circled in white) were commonly sighted. f and g Sessile ciliates attached on the surface of granules. (Bar, 50 μm) Fig. 5 Micrographs of protozoa and metazoa in floccular and granular sludge. a A swimming ciliate, Paramecium spp. and b sessile ciliates. c Metazoa such as tardigardes and d large rotifers from the genus Euchlanis. e Crawling ciliates (circled in white) were commonly sighted. f and g Sessile ciliates attached on the surface of granules. (Bar, 50 μm) communities in the control sludge did not changed dras- tically over time (Fig. S3a). However, eukaryotic commu- nities in the thiram treated sludge diverged over time and were distinctly different from the control sludge from week 1 to 5 (Fig. S3a). This was likely due to the absence several dominant protozoan OTUs including OTU01, 02 and 07. Interestingly, the eukaryotic commu- nities in the control and treated sludge began to con- verge from week 6 onwards, which was likely due to the resurgence of protozoan OTU08 and OTU24 (Fig. S3a). As granulation progressed in the control mSBRs, the abundances of these OTUs were consistent, with Oligo- hymenophorea (OTU01) being the most dominant. Both Oligohymenophorea OTUs 02 and 07 showed a gradual decline in abundance while Oligohymenophorea (OTU26) was not detected beyond week 5. Swimming ciliates from the genus Paramecium (OTU 03) were not detected after week 1 (Fig. 7). In contrast to the swim- ming ciliates, crawling ciliates from the genus Aspidisca (OTU 23) were relatively abundant during granulation. Effects of thiram treatment on microbial communities during aerobic granulation However, these protozoan OTUs were mostly not de- tected after week 1 in the thiram treated sludge (Fig. 7). Interestingly, two flagellate associated OTUs, OTU08 and 24, increased in abundance in the treated sludge from week 4 onwards. As granulation progressed in the control mSBRs, the abundances of these OTUs were consistent, with Oligo- hymenophorea (OTU01) being the most dominant. Both Oligohymenophorea OTUs 02 and 07 showed a gradual decline in abundance while Oligohymenophorea (OTU26) was not detected beyond week 5. Swimming ciliates from the genus Paramecium (OTU 03) were not detected after week 1 (Fig. 7). In contrast to the swim- ming ciliates, crawling ciliates from the genus Aspidisca (OTU 23) were relatively abundant during granulation. However, these protozoan OTUs were mostly not de- tected after week 1 in the thiram treated sludge (Fig. 7). Interestingly, two flagellate associated OTUs, OTU08 and 24, increased in abundance in the treated sludge from week 4 onwards. Bacterial OTUs were also analyzed to determine if the absence of predators had any impact on the bacterial communities during aerobic granulation. Based on nMDS visualization, two distinct clusters were observed which indicated dissimilarities between the control and thiram treated bacterial communities during granulation (Fig. S3b). The bacterial communities remained rela- tively similar from week 2 to week 8 in the control sludge while the bacterial communities in the thiram treated sludge continue to change on a weekly basis (Fig. S3b). In addition, there was no significant difference in the microbial communities between replicates of the Metazoan OTUs representing rotifers, e.g. OTUs 04, 05, 06, 15 and 22, were also detected in relatively high abundance in the control sludge and were present throughout the entire granulation process (Fig. 7). In contrast, these rotifers were only detected at low abun- dance during the first 3 weeks in the treated sludge and were mostly not detected beyond week 4. Eukaryotic Chan et al. BMC Microbiology (2021) 21:91 Page 9 of 18 Fig. 6 Development of granules from untreated and thiram treated floccular sludge. a Micrographs of control and thiram treated floccular sludge treated. b Mean particle size of seed (dark grey, week 0), control (black) and treated sludge (white) over 8 weeks. c SVI5 of seed (black triangle), control (black square) and treated (grey circle) sludge. Error bars represent standard deviation (n = 3) * and **** denote significant differences (One-way ANOVA: P-value ≤0.05 and 0.0001, respectively). Effects of thiram treatment on microbial communities during aerobic granulation Magnification × 40 (Bar, 200 μm) Fig. 6 Development of granules from untreated and thiram treated floccular sludge. a Micrographs of control and thiram treated floccular sludge treated. b Mean particle size of seed (dark grey, week 0), control (black) and treated sludge (white) over 8 weeks. c SVI5 of seed (black triangle), control (black square) and treated (grey circle) sludge. Error bars represent standard deviation (n = 3) * and **** denote significant differences (One-way ANOVA: P-value ≤0.05 and 0.0001, respectively). Magnification × 40 (Bar, 200 μm) was also observed in ‘Candidatus Competibacter’ (OTU12, 17 and 19) and ‘Candidatus Contendobacter’ (OTU09) (Fig. 8). Interestingly, there were several bac- terial OTUs such as Zoogloea (OTU03), Thauera (OTU02), Dechloromonas (OTU07) and Defluviicoccus (OTU18) that increased in abundance from week 1. However, as ‘Candidatus Accumulibacter’ (OTU01) began to gradually increase in abundance from week 5, the abundance of Zoogloea (OTU03), Thauera (OTU02), Dechloromonas (OTU07) decreased. In contrast, OTU18 and 20 continued to gradually increase in abundance from week 5 onwards. ‘Candidatus Accumulibacter’ (OTU01) increased in abundance in the thiram treated sludge as it entered Phase II of granulation, where com- pact aggregates were formed. was also observed in ‘Candidatus Competibacter’ (OTU12, 17 and 19) and ‘Candidatus Contendobacter’ (OTU09) (Fig. 8). Interestingly, there were several bac- terial OTUs such as Zoogloea (OTU03), Thauera (OTU02), Dechloromonas (OTU07) and Defluviicoccus (OTU18) that increased in abundance from week 1. However, as ‘Candidatus Accumulibacter’ (OTU01) began to gradually increase in abundance from week 5, the abundance of Zoogloea (OTU03), Thauera (OTU02), Dechloromonas (OTU07) decreased. In contrast, OTU18 and 20 continued to gradually increase in abundance from week 5 onwards. ‘Candidatus Accumulibacter’ (OTU01) increased in abundance in the thiram treated sludge as it entered Phase II of granulation, where com- pact aggregates were formed. control or thiram sludge due to close clustering in each week (Fig. S3b). In the control sludge, the bacterial communities were dominated mainly by PAOs such as ‘Candidatus Accu- mulibacter’ (OTU01), GAOs such as ‘Candidatus Com- petibacter’ (OTU04 and 05) and Nitrospira (OTU06) throughout 8 weeks of aerobic granulation (Fig. 8). The abundance of other bacterial members such as Zoogloea (OTU03), Thauera (OTU02), Dechloromonas (OTU07), ‘Candidatus Competibacter’ (OTU12, 17 and 19), ‘Can- didatus Contendobacter’ (OTU09), Defluviicoccus (OTU18) and Actinobacteria (OTU20) remained rela- tively consistent during granulation (Fig. 8). Effects of thiram treatment on microbial communities during aerobic granulation In contrast, there was a decrease in the abundance of ‘Candidatus Accumulibacter’ (OTU01) and ‘Candidatus Competibac- ter’ (OTU04 and 05) from week 1 in the thiram treated sludge. The genus Nitrospira (OTU06) also demon- strated decline in abundance from week 1 onwards with no sign of recovery (Fig. 8). The decrease in abundance Discussion The abundance of ‘Candidatus Accumulibacter’ in- creased progressively during the granulation process Page 10 of 18 Chan et al. BMC Microbiology (2021) 21:91 Fig. 7 Abundance of eukaryotic populations in control and treated mini-SBRs over 8 weeks of granulation. There were 3 control and 3 thiram treated mini-SBRs. The number of sequences per OTU in both control and treated were natural log transformed. These 30 OTUs represented approximately 92% of the total eukaryotic sequences Fig. 7 Abundance of eukaryotic populations in control and treated mini-SBRs over 8 weeks of granulation. There were 3 control and 3 thiram treated mini-SBRs. The number of sequences per OTU in both control and treated were natural log transformed. These 30 OTUs represented approximately 92% of the total eukaryotic sequences eukaryotic populations in control and treated mini-SBRs over 8 weeks of granulation. There were 3 control and 3 thiram number of sequences per OTU in both control and treated were natural log transformed. These 30 OTUs represented the total eukaryotic sequences ‘Candidatus Accumulibacter’ and ‘Candidatus Competi- bacter’ were among the most abundant in the granular sludge community [25, 32, 33]. The higher abundance of ‘Candidatus Accumulibacter’ compared to ‘Candidatus Competibacter’ observed during the maturation phase could be partly due to the presence of propionate in the synthetic feed, which ‘Candidatus Accumulibacter’ has together with ‘Candidatus Competibacter’, ‘Candidatus Contendobacter’ and Nitrospira which were present at relatively high abundance (Fig. 2). Their higher abun- dance during granulation over other bacterial members suggests their close association with denser flocs or granules with better settling characteristics. This is simi- lar to other biological nutrient removal systems where Page 11 of 18 Chan et al. BMC Microbiology (2021) 21:91 Fig. 8 Abundance of top 20 abundant bacterial OTUs in control and treated mini-SBRs over 8 weeks of granulation. There were 3 control and 3 thiram treated mini-SBRs. The number of sequences per OTU in both control and treated were natural log transformed before analysis Fig. 8 Abundance of top 20 abundant bacterial OTUs in control and treated mini-SBRs over 8 weeks of granulation. There were 3 control and 3 thiram treated mini-SBRs. The number of sequences per OTU in both control and treated were natural log transformed before analysis been shown to utilize/uptake more efficiently than ‘Can- didatus Competibacter’ [34]. sludge particle size (Fig. 3b) and Microviridae were associ- ated with changes in bacterial composition (Fig. 3c). Discussion y [ ] Other than bacteriophages, both microscopy and se- quencing data analysis demonstrated an abundance of protozoa present during the floccular phase (Fig. 3d and 4). Previous studies also reported a high abundance and diversity of protozoa in floccular sludge [40–42]. While protozoan predation in activated sludge has been sug- gested to facilitate increased biofilm production, a nega- tive or lack of correlation was observed between the ciliated protozoa and particle sizes during the floccular phase. In contrast, there was a positive correlation be- tween the flagellate family Salpingoecidae and the floc- cular particle size (Fig. 3e). Flagellates commonly predominate the activated sludge in the early stages as they consume lesser energy required for growth com- pared other larger protozoa such as ciliates. Interest- ingly, this family of flagellates was continuously detected throughout granulation and did not demonstrate any positive correlation with the particle sizes in the initi- ation or maturation phases. However, as compact aggre- gates formed and expanded during the initiation phase, sequences associated with Oligohymenophorea continued to be detected at relatively high abundance. While these sequences cannot clearly define the types of ciliated protozoa, microscopy analysis indicated that crawling and sessile ciliates increased in abundance. Grazing by crawling ciliates has been reported to stimulate the growth of microcolonies for surface attached biofilms [43]. In addition, activated sludge flocs that were co- cultured with crawling ciliates from the genus Aspidisca demonstrated an increase in floc particle size and com- pactness [44]. Moreover, the motility of crawling ciliates such as Chilodonella can dislodge cells from biofilms [43, 45]. These dislodged cells could then become a food source for the filter-feeding sessile ciliates. As more compact and dense granules formed, crawling ciliates Weber et al. [24] hypothesised that sessile ciliates could also act as nucleating agents for the attachment of bacteria. To investigate the role of protozoan predation in granulation, thiram was added into floccular sludge in this project. The resulting eukaryotic community in the thiram treated sludge was significantly different from the non-treated sludge where OTUs corresponding to proto- zoa were rarely detected by week 2 in the thiram treated sludge (Fig. 7). Importantly, there was very low or no de- tection of ciliated protozoa in the treated sludge. With- out these ciliated protozoa particularly sessile ciliates, bacteria attachment was affected as the significant re- duction of sessile agents contributed towards a loss of nucleating agents. Discussion Inoviridae also seemed to influence changes in the bacterial community (Fig. 3c), in particular for weeks 4 to 8, when the sludge was in the initiation to early maturation stages of granu- lation. Aerobic granule development has been suggested to be similar to biofilms that are surface associated [38], both containing EPS in their structure [26] as well as ad- hesion and cell-cell contact [39]. As for surface- associated biofilms, it is possible that filamentous phages play a role as a structural component in the granule as well. An example would be for the Pf4 filamentous phage in Pseudomonas aeruginosa bio- films where the phage organises the biofilm matrix into a liquid crystal structure that has increased vis- cosity and adhesion [27]. Sequencing analysis demonstrated ciliates from the class Oligohymenophorea were present throughout the granula- tion process (Fig. 3d). The class Oligohymenophorea con- sist of several subclasses such as Peritrichia that represent a distinctive group of sessile ciliates which are hypothe- sised to play an important role in granulation [22, 24, 46]. Sessile ciliates were often observed on the surfaces of acti- vated sludge flocs and aerobic granules [21, 24, 47]. While sessile ciliates were observed microscopically on granules (Fig. 5f and g), there were no sequences from sessile cili- ates that were classified to the genus level. Nonetheless, positive correlation with particle size during the initiation phase for the OTU 01 and 03 representing Oligohymeno- phorea suggests that these two classes of ciliated protozoa could play a more important role during the initiation phase of aerobic granulation rather than in the formation of mature granules. It is also possible that the change from flocs to granules acted as a form of selection pressure on the protozoan community, which also led to a significant reduction in the abundance and diversity of protozoa. For example, the presence of abundant, free-swimming proto- zoa will exert predation pressure on free swimming bac- teria, which is known to promote biofilm formation such as compact aggregates and aerobic granules. However, the formation of biofilms simultaneously increases availability of substratum for colonization of sessile ciliates and sur- face grazing by crawling ciliates while reducing availability of free swimming bacteria for predatory flagellates and free swimming ciliates. Overall, the data suggest that the formation of aggregates favoured the growth of crawling and sessile ciliates while the compact and large granules favoured the colonisation of sessile ciliates. Discussion While network analysis suggested that Podoviridae may be select- ively predating on Verrucomicrobium (Fig. 4), the bacterial targets in flocs and granules of Microviridae, Podoviridae and Siphoviridae remain unclear. Verrucomicrobia are com- mon in many wastewater treatment plants though a specific function has not been attributed to them [35, 36]. As ‘Candidatus Accumulibacter’ is often enriched for their capacity to remove phosphorus, their high abundance can increase their susceptibility to phage attack. It has been re- ported that abundant bacteriophage-like particles coupled with declining abundance of ‘Candidatus Accumulibacter’ and lysed ‘Candidatus Accumulibacter’ cells strongly sug- gested that phage infection was the main cause for ‘Candi- datus Accumulibacter’ mortality [18]. Furthermore, the addition of bacteriophage-containing supernatant to other wastewater sludge also demonstrated similar decline in the abundance of ‘Candidatus Accumulibacter’, resulting in poor phosphorus removal [18]. These observations strongly suggest that phage infection can alter bacterial communities and their functionality of the systems in which they are present. The presence of certain bacteriophage families may be an indication that phage predation and the lysis of spe- cific bacteria are required during the process of granulation. The change in abundance of Microviridae and Podoviridae (which are generally lytic phages) and Siphoviridae (temper- ate phages) were positively correlated with the increase in Although the role of lytic phages in the formation of aerobic granules is still unknown, a recent study demon- strated that bacteriophage-mediated extracellular DNA release was found to be vital for the structural stability of smaller aerobic granules [20]. Extracellular DNA con- taining clustered regularly interspaced short palindromic repeats (CRISPR) spacers was found to be part of the aerobic granular structure and extensive flocculation happened upon treatment with DNase I [20]. In our study here, the sequences associated with lytic phages from Microviridae and Podoviridae were constantly de- tected during the granulation process, suggesting that bacterial hosts associated with these families of lytic phages were present and phage lysis were actively Page 12 of 18 Page 12 of 18 Page 12 of 18 Chan et al. BMC Microbiology (2021) 21:91 are likely outcompeted by sessile ciliates whose growth is favored by an increase in granule surface area. ongoing throughout granulation. The increased abun- dance of Inoviridae (which are generally non-lytic fila- mentous bacteriophages) [37] in the reactor effluent at weeks 5, 7, 8 and 9 (Fig. 3a) coincided with the appear- ance of compact aggregates (Fig. 1). Sequencing batch reactor setup and operation q g p p To characterize the protozoan communities during aer- obic granulation, four independent SBRs were seeded with activated floccular sludge from the Ulu Pandan Wastewater Treatment Plant, Singapore, as previously described [52, 53]. Briefly, each SBR had a final working volume of 2 L and was operated in a 6 h cycle compris- ing two different phases: Phase I - feeding (8 min), an- aerobic (60 min), aerobic (80 min at day 0, with a gradual increase to 95 min by week 5) and anoxic (40 min at day 0, with a gradual increase to 50 min by week 5); Phase II - feeding (2 min), anaerobic (30 min), aerobic (40 min at day 0 and gradual increase to 70 min by week 5) and anoxic (30 min). Each cycle was completed with a settling stage (120 min at day 0, with a gradual decrease to 5 min by the end of week 6) and a 10 min decanting stage. The settling time was maintained at 5 min per cycle from week 6 onwards. While there was a delay in initiation, the thiram treated sludge ultimately granulated by week 8 and we hypothesised that it could be largely attributed to the emergence of ‘Candidatus Accumulibacter’ after week 5. In contrast, the control sludge, which was dominated mainly by ‘Candidatus Accumulibacter’ and ‘Candidatus Competibacter’, entered the initiation and granulation phase earlier at weeks 4 and 6, respectively. These obser- vations further support that the high abundance of ‘Candidatus Accumulibacter’ and ‘Candidatus Competi- bacter’ are strong potential contributors towards the for- mation of aerobic granules. Other than bacteria, it is also likely that protozoan predation plays a partial role in enhancing the formation of aerobic granules by redu- cing the time to initiate compact aggregates formation. A total volume of 1 L of synthetic wastewater was sup- plied to each SBR by Phase II and 1 L of effluent was dis- charged at the completion of each cycle. Synthetic wastewater was prepared as previously described [54, 55]. Dissolved oxygen (DO) levels were maintained at 0.0 mg L−1 during anaerobic phases via intermittent ni- trogen sparging and maintained between 3.0 to 4.0 mg L−1 during aerobic phases by compressed air sparging. Sparging of both nitrogen and air provided complete mixing of the sludge and the hydrodynamic shear force required for aerobic granulation. Discussion It is likely that the absence of proto- zoa was responsible for the delayed increase in particle size. Protozoa have also been demonstrated to excrete growth stimulating products which could potentially in- duce flocculation [48]. For example, the co-incubation of activated sludge bacteria with sludge protozoa composed of attached, crawling ciliates, flagellates and amoeba for Page 13 of 18 Page 13 of 18 Chan et al. BMC Microbiology (2021) 21:91 48 h, resulted in biofilms that had 2000% more biomass than the biofilms that were not exposed to protozoan predation [11]. In addition, polymeric substances such as extrusomes, cellular debris and undigested residues se- creted from protozoa could also facilitate aggregation between bacterial flocs [49]. Hence, the absence of protozoan predation could have reduced the selection pressure aggregation, resulting in less dense flocs that settle poorly. Poor settling of the treated sludge could also have resulted in larger losses of sludge biomass dur- ing discharge. This finding corresponded to previous ob- servations where ciliated protozoa were important for the formation of compact aggregates which leads to granule formation. isolation of specific protozoa and subsequently adding them back in significant amounts in floccular or granu- lar sludge. Future work should address these issues by isolating phages and protozoa from key points in the granular development (e.g. initiation phase) and adding back these phages to either floccular or granular sludge to see their more direct impacts on granulation. Conclusions Predation by bacteriophages and protozoa can influence the diversity and structure of the bacterial community. The alteration of bacterial community composition sub- sequently affected the rate of granulation of floccular sludge. While physical parameters such as settling time have significant effects on promoting granulation, we have also demonstrated the potential role of bacterio- phage and protozoa in promoting granulation through physical means such as bacterial attachment on phage filaments or sessile ciliates. In the absence of protozoan predators, the floccular sludge demonstrated poor compactness and settling. Both ‘Candidatus Accumulibacter’ and ‘Candidatus Competibacter’ were replaced by the genera Thauera and Zoogloea as the dominant members of the bacterial community in the treated floccular sludge in Phase I. The ‘Candidatus Accumulibacter’ could have also been replaced by both genera of Thauera and Dechloromonas, which are PAOs that are also capable of denitrification [50]. The proliferation of Zoogloea was likely due to in- sufficient retention of sludge [32, 51]. Zoogloea are floc- forming bacteria that produce aggregates enveloped in gelatinous matrixes that could have initiated the forma- tion of compact aggregates. However, the initiation phase was delayed in the thiram treated reactors despite the abundance of Zoogloea associated sequences and it is possible that the delay in initiation was due to the ab- sence of protozoa. Total genomic DNA extraction from aerobic granular sludge One milliliter of suspended sludge was pelleted by centrifugation for 10 min at 10,000 g. The total gen- omic DNA was extracted using the sludge pellet with the FastDNA™SPIN Kit for Soil (MP Biomedical, USA) mostly according to the manufacturer’s guide- lines. Homogenization was performed twice in the FastPrep® Instrument for 40 s at a speed setting of 6.0. The extracted genomic DNA was then cleaned up using the Genomic DNA Clean & Concentrator (Zymo Research, USA) according to the manufac- turer’s guidelines. The concentration of the DNA was quantified using the dsDNA HS Assay Kit and the Qubit® 2.0 Fluorometer (Life Technologies, USA) be- fore sequencing on the Illumina HiSeq as 250 bp paired end reads. Viral fraction sampling and concentration Viral fraction sampling and concentration Viral fraction sampling and concentration During SBR operation, the viral fraction was collected and concentrated as previously described [52]. Briefly, effluent from each reactor was discharged into their re- spective containers before transferring into 25 L carboys. Samples were collected from weeks 1 to 10 of the study. The initial filtrate was obtained by passing through a 25 μm filter bag (Puridea, Singapore) to remove any sus- pended biomass before storing at 4 °C. The filtrate was pooled over 4 d to obtain 20 L, at which time, 2 mL of DNase I (200 U/mL) (Calbiochem, USA) was added to the samples to digest any extracellular DNA. Bacteria were then removed by passing the filtered effluent through a 0.2 μm Sartocon Slice Disposable tangential flow filter (TFF) (Sartorius Stedim, Germany). To con- centrate the viral fraction, the permeate was concen- trated using a 100 kDa Sartocon Slice Dispostable TFF (Sartorius Stedim, Germany). In this process, the fluid phase and particles smaller than 100 kDa were removed while the viral fraction remained in the reservoir. To elute the viral fraction, SM buffer (100 mM NaCl, 8 mM MgSO4, 50 mM Tris-Cl at pH 7.5) was added to the phage reservoir until the volume was reduced to a 100 mL. The viral fraction was further concentrated using the Vivaspin 20 100,000 MWCO Centrifugal Concentra- tors (Sartorius Stedium, Germany) by centrifuging at 5000 g for 30 min at 4 °C until a volume between 2 to 3 mL was obtained and stored at -80 °C as 200 μL aliquots. Sequencing batch reactor setup and operation The pH of each SBR was maintained between 6.8 and 8.2 by dosing with 0.1 M HCl and 0.1 M NaOH as required. Both pH and DO levels were monitored by inline probes connected to a programmable logic controller (PLC). The limitations of this study are that while the mech- anistic link between protozoan predation and granula- tion was directly tested, the effect of bacteriophages on granulation is correlative. In addition, the current se- quencing data is unable to define Oligohymenophorea se- quences to the genus or species level. Hence, any true impact on granulation from these ciliates will require Page 14 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 idxstats function of Samtools (v 1.3.1) [64]. The con- tig abundance and LCA-annotated contigs were then analysed using Phyloseq (v 1.22.3) [65] in R. Mixed liquor suspended solids (MLSS) and mixed li- quor volatile suspended solids (MLVSS), were deter- mined using APHA standard engineering methods [56]. Sludge density and compactness was measured by sludge volumetric index at 5 min (SVI5) as described [7]. Sludge particle sizes were determined using a laser diffraction particle size analyser (SALD-3101, Shimadzu, Japan) and their morphology was recorded by light microscopy (Primo Star, Carl Zeiss, Germany). At the end of each cycle study, well-mixed sludge samples of 1 mL were col- lected from each reactor at the end of Phase II anoxic stage. These sludge samples were centrifuged at 8000 g for 5 min and snap frozen in liquid nitrogen prior to storage at −80 °C. idxstats function of Samtools (v 1.3.1) [64]. The con- tig abundance and LCA-annotated contigs were then analysed using Phyloseq (v 1.22.3) [65] in R. Bacteriophage nucleic acid extraction and multiple displacement amplification g The quality of the metagenomic reads was assessed using FastQC (v 0.11.5) before it was adapter and quality trimmed using BBMap (v 36.38) [57]. Contigs were co-assembled using MEGAHIT (v 1.0.6–3) [58] with the meta-sensitive preset mode before ORF pre- diction was done using the meta mode of Prodigal (v 2.6.3) [59]. Using nucleic acid ORF sequences, redun- dancy was removed using cd-hit-est (v 4.6.8) [60] with the options for 95% sequence identity and word length of 10. The non-redundant ORF sequences were then used in a protein homology search using the Blastx function of DIAMOND (v 0.8.22) [61] against the NCBI nr database. Based on the Blastx output, the lowest common ancestor (LCA) annotation for the contigs were performed using MEGAN6 Commu- nity Edition (v 6.8.12) [62]. To obtain the contig abundance table, the metagenome reads were mapped to the co-assembled contigs using Bowtie2 (v 2.2.6) [63] before read coverage was obtained with the Nucleic acids were extracted using the QIAamp MinE- lute Virus Spin kit (Qiagen, Germany) from 200 μL of concentrated viral fraction according to manufacturer’s guidelines. The viral DNA was used for whole genome multiple displacement amplification (MDA) using ran- dom hexamers with the illustra GenomiPhi V2 DNA amplification kit (Cytiva, USA) according to the manu- facturer’s guidelines. The amplified DNA was then puri- fied using the ethanol precipitation method [66]. Briefly, sodium acetate was added and mixed to the amplified sample to a final concentration of 0.3 M at pH 5.2. Two volumes of cold 100% molecular grade ethanol was added and incubated overnight at −20 °C. After incuba- tion, the sample was centrifuged at 15,000 g for 30 min and the supernatant was removed. One mL of 70% etha- nol was then added and incubated at −20 °C for 2 h be- fore centrifuging at 15,000 g for 30 min to pellet DNA. The supernatant was discarded and the pellet air dried for 5 min before resuspension in sterile dH2O. Page 15 of 18 Page 15 of 18 Page 15 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 Mini-sequencing batch reactors setup and operation Mini-sequencing batch reactors setup and operation Mini-SBRs (mSBR) were seeded with activated floccular sludge from the Ulu Pandan Wastewater Treatment Plant, Singapore. For floccular sludge experiments, both controls and treatments were performed in triplicate while granular sludge experiments were performed in duplicate. Each mSBR had a final working volume of 30 mL and was operated in a 6 h cycle: feeding (10 min), an- aerobic (100 min), aerobic (110 min at day 0, with a gradual increase to 120 min by the end of week 1) and anoxic (100 min) phases. Each cycle was completed with a settling stage (30 min at day 0, with a gradual decrease to 20 min by the end of week 1) and a 10 min decanting stage. The settling time was maintained at 20 min per cycle from the end of week 1 onwards. Analysis of the viral fraction (National Institute of Health, USA). For enumeration of protozoa, triplicate 10 μL aliquots were removed from each mSBR and the numbers of protozoa determined using light microscopy (Primo Star, Carl Zeiss, Germany). Samples (1 mL) were collected from each mSBR at the end of Phase II anoxic stage, centrifuged at 8000 g for 5 min and snap frozen in liquid nitrogen prior to storage at -80 °C. The quality of the metavirome reads was assessed using FastQC (v 0.11.5) before adapter and quality trimming using BBMap (v 36.38) [57]. Contig co-assembly was done using the MEGAHIT (v 1.0.6–3) meta-sensitive preset mode [58] before doing ORF prediction using the Prodigal (v 2.6.3) meta mode [59]. The viral ORF se- quences were used in a protein homology search using the DIAMOND (v 0.8.22) Blastp program against the A CLAssification of Mobile genetic Elements (ACLAME) database. Based on the output, the LCA annotation for the viral contigs were obtained using MEGAN6 Com- munity Edition (v 6.8.12) [62]. The metavirome reads were then mapped to the viral contigs using Bowtie2 (v 2.2.6) [63] before obtaining read coverage using the idx- stats function of Samtools (v 1.3.1) [64]. The LCA anno- tated viral contigs and their abundances were then used for downstream analyses. RNA extractions for total RNA sequencing and analysis RNA extractions for total RNA sequencing and analysis Total RNA was extracted from sludge samples using the Soil, Fecal and Plant RNA kit (Zymo Research, USA) as described [68, 69], according to the manufacturer’s guidelines. Extracted RNA underwent a single round of DNase treatment to remove residual DNA (TURBO™ DNase kit; Invitrogen, Singapore). The quality of the ex- tracted RNA was measured by spectrophotometry (Nanodrop; Thermo Scientific, USA). The concentration of RNA and residual DNA was determined by fluorom- etry (Qubit® 2.0 Fluorometer; Invitrogen, USA), using the Qubit® RNA broad range assay kit (Invitrogen, USA) and Qubit® DNA high sensitivity range assay kit respect- ively, following the manufacturer’s guidelines. In addition, the integrity of the RNA was determined using the RNA Analysis ScreenTape and 2200 Tapestation in- strument (Agilent Technologies, Singapore) and re- ported as the RNA Integrity Number (RIN). These RNA samples were subsequently sent for RNA library prepar- ation prior to pooling and sequencing on an Illumina HiSeq 2500 System (Illumina Inc.) using 100 bp paired- end (PE) sequencing as per the manufacturer’s guidelines. Declarations Ethics approval and consent to participate Not applicable. Acknowledgements Not applicable. 8. Tan CH, Koh KS, Xie C, Tay M, Zhou Y, Williams R, Ng WJ, Rice SA, Kjelleberg S. The role of quorum sensing signalling in EPS production and the assembly of a sludge community into aerobic granules. ISME J. 2014;8(6): 1186–97. https://doi.org/10.1038/ismej.2013.240. 8. Tan CH, Koh KS, Xie C, Tay M, Zhou Y, Williams R, Ng WJ, Rice SA, Kjelleberg S. The role of quorum sensing signalling in EPS production and the assembly of a sludge community into aerobic granules. ISME J. 2014;8(6): 1186–97. https://doi.org/10.1038/ismej.2013.240. Funding Th g 11. Rychert K, Neu T. Protozoan impact on bacterial biofilm formation. In: Biological Letters, vol. 47; 2010. p. 3. 11. Rychert K, Neu T. Protozoan impact on bacterial biofilm formation. In: Biological Letters, vol. 47; 2010. p. 3. This research was supported by the Singapore Centre for Environmental Life Sciences Engineering (SCELSE), whose research is supported by the National Research Foundation Singapore, Ministry of Education, Nanyang Technological University and National University of Singapore, under its Research Centre of Excellence Programme. MH Ismail was supported by the National Research Foundation Singapore under its National Research Foundation (NRF) Environmental and Water Technologies (EWT) PhD Scholarship Programme, administered by the Environment and Water Industry Programme Office (EWI). 12. Sun S, Kjelleberg S, McDougald D. Relative contributions of Vibrio polysaccharide and quorum sensing to the resistance of Vibrio cholerae to predation by heterotrophic protists. PLoS One. 2013;8(2):e56338. https://doi. org/10.1371/journal.pone.0056338. 12. Sun S, Kjelleberg S, McDougald D. Relative contributions of Vibrio polysaccharide and quorum sensing to the resistance of Vibrio cholerae to predation by heterotrophic protists. PLoS One. 2013;8(2):e56338. https://doi. org/10.1371/journal.pone.0056338. 13. Brown MR, Baptista JC, Lunn M, Swan DL, Smith SJ, Davenport RJ, et al. Coupled virus - bacteria interactions and ecosystem function in an engineered microbial system. Water Research. 2019;152:264–73. https://doi. org/10.1016/j.watres.2019.01.003. 14. Khan MA, Satoh H, Katayama H, Kurisu F, Mino T. Bacteriophages isolated from activated sludge processes and their polyvalency. Water Research. 2002;36(13):3364–70. https://doi.org/10.1016/S0043-1354(02)00029-5. 14. Khan MA, Satoh H, Katayama H, Kurisu F, Mino T. Bacteriophages isolated from activated sludge processes and their polyvalency. Water Research. 2002;36(13):3364–70. https://doi.org/10.1016/S0043-1354(02)00029-5. Total RNA sequencing and analysis q g y The microbial composition of the floccular and granular sludge was determined by analysis of the se- quence data using the Ribotagger fast tag-based ap- proach [31]. Briefly, universal recognition profiles that target bacteria, Archaea and eukaryotes were selected for each of the hypervariable regions of both 16S and 18S rRNA (e.g. V4, V5, V6 and V7) (Xie et al. [31]). These universal recognition profiles were used to scan the sequencing reads to obtain 33 nucleotides (nt) downstream of the primers (Xie et al. [31]). Each of these 33 nt tags were defined as a ribotag and each ribotag was screened against the SILVA database to map it to a known organism. Hence, each ribotag was used as a signature sequence to represent one oper- ational taxonomic unit (OTU). Here, only the sequen- cing reads from the V5 regions of 18S rRNA were used to represent the abundance of protozoan com- munities. Based on the lowest number of total se- quencing reads within the samples set, these V5 sequencing reads were randomly subsampled based y Synthetic wastewater (15 mL) was fed to each mSBR in Phase II and 15 mL of treated effluent was discharged at the end of the cycle. For the inhibition of eukaryotes, thiram (Sigma Aldrich, Germany) was dissolved in di- methyl sulfoxide (DMSO) to obtain a stock solution of 20 g L−1 for treatment of the floccular sludge. Thiram has been shown to inhibit protozoa with minimal im- pacts on bacterial activities [67]. Based on optimization studies, thiram was added to each reactor once per day after feeding to obtain a final concentration of 20 mg L−1 (data not shown), while DMSO was added to con- trol mSBRs. Both DMSO and thiram treatment of sludge was completed by week 2. Both control and treated mSBRs were operated from weeks 3 to 8 without the addition of DMSO or thiram. To achieve anaerobic and aerobic conditions, nitrogen and compressed air were sparged intermittently into the mSBRs. The average particle diameter of the floccular sludge was determined by analyzing images of sludge, taken in triplicate for each mSBR, on a weekly basis using ImageJ Page 16 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology on a seed value of 100 using the seqtk FASTQ pro- gram (https://github.com/lh3/seqtk). Received: 13 January 2021 Accepted: 8 March 2021 Received: 13 January 2021 Accepted: 8 March 2021 Competing interests p g The authors declare that they have no competing interests. Network analysis Consent for publication Not applicable. Consent for publication Not applicable. The combined abundance table of the bacterial, proto- zoan and viral communities were combined before it was loaded as a phyloseq object using Phyloseq (v 1.22.3) [65] in R. The network analysis was performed using the SpiecEasi package (v 1.1.0) [70]. Briefly, the abundance table was normalised using centered log-ratio transformation before inverse covariance estimation was done. The stability of the network was inferred using the package’s Stability Approach to Regularization Selection (StARS) criterion. The following parameters were used: method = “mb”, lambda.min.ratio = 0.05, nlambda = 100. Statistical analysis Correlation studies for protozoa and bacteriophages were performed by calculating Spearman correlation co- efficient using Prism (Graphpad 6.0). The resulting protozoa matrices were clustered hierarchically based firstly by obtaining the Bray- Curtis dissimilarity matrix and then clustering using the hclust function in vegan (v.5–6) [71] in R. The distance matrices were used for non-metric multi-dimensional scaling (NMDS) to deter- mine the level of similarity or dissimilarity between of samples based no bacteria and eukaryotic communities. Additional file 1. Additional file 1. 7. Liu Y. Wastewater purification: Aerobic granulation in sequencing batch reactors. Boca Raton: Taylor & Francis; 2008. 7. Liu Y. Wastewater purification: Aerobic granulation in sequencing batch reactors. Boca Raton: Taylor & Francis; 2008. Authors’ contributions 9. Tay JH, Liu QS, Liu Y. The role of cellular polysaccharides in the formation and stability of aerobic granules. Lett Appl Microbiol. 2001;33(3):222–6. https://doi.org/10.1046/j.1472-765x.2001.00986.x. 9. Tay JH, Liu QS, Liu Y. The role of cellular polysaccharides in the formation and stability of aerobic granules. Lett Appl Microbiol. 2001;33(3):222–6. https://doi.org/10.1046/j.1472-765x.2001.00986.x. SHC, MHI, CHT and SAR conceived and designed the experimental studies. SHC, MHI and CHT performed most of the experiments. SHC and MHI analysed most of the experiments. SAR and DM edited the paper. All authors had substantial contributions to the paper, were involved in writing, approved the final version, and are accountable for all aspects of the work. 10. Matz C, McDougald D, Moreno AM, Yung PY, Yildiz FH, Kjelleberg S. Biofilm formation and phenotypic variation enhance predation-driven persistence of Vibrio cholerae. P Natl Acad Sci USA. 2005;102(46):16819–24. https://doi. org/10.1073/pnas.0505350102. 10. Matz C, McDougald D, Moreno AM, Yung PY, Yildiz FH, Kjelleberg S. Biofilm formation and phenotypic variation enhance predation-driven persistence of Vibrio cholerae. P Natl Acad Sci USA. 2005;102(46):16819–24. https://doi. org/10.1073/pnas.0505350102. References 1. Beun JJ, Hendriks A, Van Loosdrecht MCM, Morgenroth E, Wilderer PA, Heijnen JJ. Aerobic granulation in a sequencing batch reactor. Water Res. 1999;33(10):2283–90. https://doi.org/10.1016/S0043-1354(98)00463-1. 2. Morgenroth E, Sherden T, van Loosdrecht MCM, Heijnen JJ, Wilderer PA. Aerobic granular sludge in a sequencing batch reactor. Water Res. 1997; 31(12):3191–4. https://doi.org/10.1016/S0043-1354(97)00216-9. 2. Morgenroth E, Sherden T, van Loosdrecht MCM, Heijnen JJ, Wilderer PA. Aerobic granular sludge in a sequencing batch reactor. Water Res. 1997; 31(12):3191–4. https://doi.org/10.1016/S0043-1354(97)00216-9. 3. McSwain BS, Irvine RL, Wilderer PA. The influence of settling time on the formation of aerobic granules. Water Sci Technol. 2004;50(10):195–202. https://doi.org/10.2166/wst.2004.0643. 4. Qin L, Liu Y, Tay J-H. Effect of settling time on aerobic granulation in sequencing batch reactor. Biochem Eng J. 2004;21(1):47–52. https://doi. org/10.1016/j.bej.2004.03.005. 5. Tay JH, Liu QS, Liu Y. The effects of shear force on the formation, structure and metabolism of aerobic granules. Appl Microbiol Biotechnol. 2001;57(1): 227–33. https://doi.org/10.1007/s002530100766. Author details 1 1Singapore Centre for Environmental Life Sciences Engineering, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore. 2Interdisciplinary Graduate School, Nanyang Technological University, Singapore, Singapore. 3School of Biological Sciences, Nanyang Technological University, Singapore, Singapore. 4School of Materials Science and Engineering, Nanyang Technological University, Singapore, Singapore. 5The iThree Institute, University of Technology Sydney, Sydney, Australia. Received: 13 January 2021 Accepted: 8 March 2021 Supplementary Information Supplementary Information pp y The online version contains supplementary material available at https://doi. org/10.1186/s12866-021-02156-8. y The online version contains supplementary material available at https://doi. org/10.1186/s12866-021-02156-8. 6. Winkler MKH, Kleerebezem R, Khunjar WO, de Bruin B, van Loosdrecht MCM. Evaluating the solid retention time of bacteria in flocculent and granular sludge. Water Res. 2012;46(16):4973–80. https://doi.org/10.1016/j.watres.2 012.06.027. Additional file 1. Availability of data and materials Th d d d/ l Aerobic granules dwelling vorticella and rotifers in an SBR fed with domestic wastewater. Separation Purification Technol. 2013; 110:127–31. https://doi.org/10.1016/j.seppur.2013.03.022. 42. Rodriguez-Perez S, Fermoso FG, Arnaiz C. Influence of different anoxic time exposures on active biomass, protozoa and filamentous bacteria in activated sludge. Water Sci Technol. 2016;74(3):595–605. https://doi.org/1 0.2166/wst.2016.247. 22. Lemaire R, Webb RI, Yuan Z. Micro-scale observations of the structure of aerobic microbial granules used for the treatment of nutrient-rich industrial wastewater. ISME J. 2008;2(5):528–41. https://doi.org/10.1038/ismej.2008.12. 43. Dopheide A, Lear G, Stott R, Lewis G. Preferential feeding by the ciliates Chilodonella and Tetrahymena spp. and effects of these protozoa on bacterial biofilm structure and composition. Appl Environ Microbiol. 2011; 77(13):4564–72. https://doi.org/10.1128/aem.02421-10. 23. Schwarzenbeck N, Erley R, Wilderer PA. Aerobic granular sludge in an SBR- system treating wastewater rich in particulate matter. Water Sci Technol. 2004;49(11–12):41–6. https://doi.org/10.2166/wst.2004.0799. 24. Weber SD, Ludwig W, Schleifer KH, Fried J. Microbial composition and structure of aerobic granular sewage biofilms. Appl Environ Microbiol. 2007; 73(19):6233–40. https://doi.org/10.1128/AEM.01002-07. 44. Walczyńska A, Sobczyk M, Fiałkowska E, Pajdak-Stós A, Fyda J, Wiąckowski K. Interaction between a bacterivorous ciliate Aspidisca cicada and a rotifer Lecane inermis: Doozers and Fraggles in aquatic flocs. Microb Ecol. 2018; 75(3):569–81. https://doi.org/10.1007/s00248-017-1036-5. 25. Barr JJ, Cook AE, Bond PL. Granule formation mechanisms within an aerobic wastewater system for phosphorus removal. Appl Environ Microbiol. 2010; 76(22):7588–97. https://doi.org/10.1128/AEM.00864-10. 45. Böhme A, Risse-Buhl U, Küsel K. Protists with different feeding modes change biofilm morphology. FEMS Microbiol Ecol. 2009;69(2):158–69. https://doi.org/10.1111/j.1574-6941.2009.00710.x. 45. Böhme A, Risse-Buhl U, Küsel K. Protists with different feeding modes change biofilm morphology. FEMS Microbiol Ecol. 2009;69(2):158–69. https://doi.org/10.1111/j.1574-6941.2009.00710.x. 26. Liu Y, Wang Z-W. Essential Roles of Extracellular Polymeric Substances in Aerobic Granulation. In: Liu Y, editor. Wastewater Purification: Aerobic Granulation in Sequencing Batch Reactors (1st ed.). Boca Raton: CRC Press; 2008. p. 181–194. 46. Fried J, Lemmer H. On the dynamics and function of ciliates in sequencing batch biofilm reactors. Water Sci Technol. 2003;47(5):189–96. https://doi. org/10.2166/wst.2003.0316. 46. Fried J, Lemmer H. On the dynamics and function of ciliates in sequencing batch biofilm reactors. Water Sci Technol. 2003;47(5):189–96. https://doi. org/10.2166/wst.2003.0316. 47. de Kreuk MK, Kishida N, Tsuneda S, van Loosdrecht MCM. Behavior of polymeric substrates in an aerobic granular sludge system. Water Res. 2010; 44(20):5929–38. https://doi.org/10.1016/j.watres.2010.07.033. 27. Secor PR, Sweere JM, Michaels LA, Malkovskiy AV, Lazzareschi D, Katznelson E, Rajadas J, Birnbaum ME, Arrigoni A, Braun KR, Evanko SP, Stevens DA, Kaminsky W, Singh PK, Parks WC, Bollyky PL. Availability of data and materials Th d d d/ l The datasets generated and/or analysed during the current study are available in the DR-NTU (Data) repository, https://doi.org/10.21979/N9/ TBOI0Y. 15. Fan N, Yang M, Jin R, Qi R. Isolation and Genomic Characterization of an Acinetobacter johnsonii Bacteriophage AJO2 From Bulking Activated 15. Fan N, Yang M, Jin R, Qi R. Isolation and Genomic Characterization of an Acinetobacter johnsonii Bacteriophage AJO2 From Bulking Activated Page 17 of 18 Page 17 of 18 Chan et al. BMC Microbiology (2021) 21:91 Sludge. Front Microbiol. 2019;10(266); doi: https://doi.org/10.3389/fmicb.201 9.00266. Sludge. Front Microbiol. 2019;10(266); doi: https://doi.org/10.3389/fmicb.201 9.00266. 35. Cydzik-Kwiatkowska A, Zielinska M. Bacterial communities in full-scale wastewater treatment systems. World J Microbiol Biotechnol. 2016;32(4):66. https://doi.org/10.1007/s11274-016-2012-9. 16. Liu M, Gill JJ, Young R, Summer EJ. Bacteriophages of wastewater foaming- associated filamentous Gordonia reduce host levels in raw activated sludge. Sci Rep. 2015;5(1):13754. https://doi.org/10.1038/srep13754. 36. Yu K, Zhang T. Metagenomic and metatranscriptomic analysis of microbial community structure and gene expression of activated sludge. PLoS One. 2012;7(5):e38183. https://doi.org/10.1371/journal.pone.0038183. 17. Khairnar K, Chandekar R, Nair A, Pal P, Paunikar WN. Novel application of bacteriophage for controlling foaming in wastewater treatment plant- an eco-friendly approach. Bioengineered. 2016;7(1):46–9. https://doi.org/10.1 080/21655979.2015.1134066. 37. Ploss M, Kuhn A. Kinetics of filamentous phage assembly. Phys Biol. 2010; 7(4):045002. https://doi.org/10.1088/1478-3975/7/4/045002. 38. Ren TT, Yu HQ, Li XY. The quorum-sensing effect of aerobic granules on bacterial adhesion, biofilm formation, and sludge granulation. Appl Microbiol Biotechnol. 2010;88(3):789–97. https://doi.org/10.1007/s00253-01 0-2796-8. 18. Barr JJ, Slater FR, Fukushima T, Bond PL. Evidence for bacteriophage activity causing community and performance changes in a phosphorus-removal activated sludge. FEMS Microbiol Ecol. 2010;74(3):631–42. https://doi.org/1 0.1111/j.1574-6941.2010.00967.x. 39. Liu Y, Tay J-H. State of the art of biogranulation technology for wastewater treatment. Biotechnol Adv. 2004;22(7):533–63. https://doi.org/10.1016/j. biotechadv.2004.05.001. 19. Wanner J, Ruzickova I, Krhutkova O, Pribyl M. Activated sludge population dynamics and wastewater treatment plant design and operation. Water Sci Technol. 2000;41(9):217–25. https://doi.org/10.2166/wst.2000.0210. 40. Madoni P. Protozoa in activated sludge. In: Bitton G, editor. Encyclopedia of Environmental Microbiology. New York: Wiley; 2003. https://doi.org/10.1002/ 0471263397.env057. 20. Wang Y-A, Li W, Zhuang J-L, Liu Y-D, Shapleigh JP. Bacteriophage-mediated extracellular DNA release is important for the structural stability of aerobic granular sludge. Sci Total Environ. 2020;726:138392. https://doi.org/10.1016/j. scitotenv.2020.138392. 41. Madoni P. A sludge biotic index (SBI) for the evaluation of the biological performance of activated sludge plants based on the microfauna analysis. Water Res. 1994;28(1):67–75. https://doi.org/10.1016/0043-1354(94)90120-1. 21. Li J, Ma L, Wei S, Horn H. Availability of data and materials Th d d d/ l Filamentous bacteriophage promote biofilm assembly and function. Cell Host Microbe. 2015;18(5):549– 59. https://doi.org/10.1016/j.chom.2015.10.013. 48. Ratsak CH, Kooi BW, van Verseveld HW. Biomass reduction and mineralization increase due to the ciliate Tetrahymena pyriformis grazing on the bacterium Pseudomonas fluorescens. Water Sci Technol. 1994;29(7):119– 28. https://doi.org/10.2166/wst.1994.0322. 28. Webb JS, Lau M, Kjelleberg S. Bacteriophage and phenotypic variation in Pseudomonas aeruginosa biofilm development. J Bacteriol. 2004;186(23): 8066–73. https://doi.org/10.1128/jb.186.23.8066-8073.2004. 49. Arregui L, Linares M, Pέrez-Uz B, Guinea A, Serrano S. Involvement of crawling and attached ciliates in the aggregation of particles in wastewater treatment plants. Air Soil Water Res. 2008;1:ASWR.S752; doi: https://doi.org/1 0.4137/aswr.s752. 29. McArdle BH, Anderson MJ. Fitting multivariate models to community data: a comment on distance-based redundancy analysis. Ecology. 2001; 82(1):290–7. https://doi.org/10.1890/0012-9658(2001)082[0290:FMMTCD]2. 0.CO;2. 50. Yun G, Lee H, Hong Y, Kim S, Daigger GT, Yun Z. The difference of morphological characteristics and population structure in PAO and DPAOgranular sludges. J Environ Sci. 2019;76:388–402. https://doi.org/10.101 6/j.jes.2018.06.003. 30. Inaba T, Hori T, Sato Y, Aoyagi T, Hanajima D, Ogata A, Habe H. Eukaryotic microbiomes of membrane-attached biofilms in membrane mioreactors analyzed by high-throughput sequencing and microscopic observations. Microbes Environ. 2018;33(1):98–101. https://doi.org/10.1264/jsme2.ME17112. 51. Shao Y, Chung BS, Lee SS, Park W, Lee S-S, Jeon CO. Zoogloea caeni sp. nov., a floc-forming bacterium isolated from activated sludge. Int J Syst Evol Microbiol. 2009;59(3):526–30. https://doi.org/10.1099/ijs.0.65670-0. 31. Xie C, Goi CLW, Huson DH, Little PFR, Williams RBH. RiboTagger: fast and unbiased 16S/18S profiling using whole community shotgun metagenomic or metatranscriptome surveys. BMC Bioinformatics. 2016;17(19):508. https:// doi.org/10.1186/s12859-016-1378-x. 52. Ismail MH. The role of bacteriophages in mixed microbial communities and populations of Pseudomonas aeruginosa. In: Doctor of Philosophy. Singapore: Nanyang Technological University; 2019. p. 255. 32. Gonzalez-Gil G, Holliger C. Dynamics of microbial community structure of and enhanced biological phosphorus removal by aerobic granules cultivated on propionate or acetate. Appl Environ Microbiol. 2011;77(22): 8041–51. https://doi.org/10.1128/AEM.05738-11. 53. Chan SH. Predator-prey interactions in aerobic granulation systems. In: Doctor of Philosophy (IGS). Singapore: Nanyang Technological University; 2018. p. 228. 33. Lemaire R, Yuan Z, Blackall LL, Crocetti GR. Microbial distribution of Accumulibacter spp. and Competibacter spp. in aerobic granules from a lab-scale biological nutrient removal system. Environ Microbiol. 2008;10(2): 354–63. https://doi.org/10.1111/j.1462-2920.2007.01456.x. 54. Zhou Y, Ganda L, Lim M, Yuan Z, Kjelleberg S, Ng W. Free nitrous acid (FNA) inhibition on denitrifying poly-phosphate accumulating organisms (DPAOs). Appl Microbiol Biotechnol. 2010;88(1):359–69. https://doi.org/10.1007/s002 53-010-2780-3. 34. Oehmen A, Yuan Z, Blackall LL, Keller J. Availability of data and materials Th d d d/ l Comparison of acetate and propionate uptake by polyphosphate accumulating organisms and glycogen accumulating organisms. Biotechnol Bioeng. 2005;91(2):162–8. https://doi.org/10.1002/bit.20500. 55. Smolders GJF, van der Meij J, van Loosdrecht MCM, Heijnen JJ. Model of the anaerobic metabolism of the biological phosphorus removal process: stoichiometry and pH influence. Biotechnol Bioeng. 1994;43(6):461–70. https://doi.org/10.1002/bit.260430605. 55. Smolders GJF, van der Meij J, van Loosdrecht MCM, Heijnen JJ. Model of the anaerobic metabolism of the biological phosphorus removal process: stoichiometry and pH influence. Biotechnol Bioeng. 1994;43(6):461–70. https://doi.org/10.1002/bit.260430605. Page 18 of 18 Page 18 of 18 Chan et al. BMC Microbiology (2021) 21:91 Chan et al. BMC Microbiology (2021) 21:91 56. Eaton AD, Franson MAH, Association APH, Association AWW, Federation WE. Standard Methods for the Examination of Water and Wastewater. Washington, DC: American Public Health Association; 2005. 56. Eaton AD, Franson MAH, Association APH, Association AWW, Federation WE. 57. Bushnell B. BBMap short-read aligner, and other bioinformatics tools; 2015. 58. Li D, Luo R, Liu CM, Leung CM, Ting HF, Sadakane K, et al. MEGAHIT v1.0: A fast and scalable metagenome assembler driven by advanced methodologies and community practices. Methods (San Diego, Calif). 2016; 102:3–11. https://doi.org/10.1016/j.ymeth.2016.02.020. 59. Hyatt D, Chen G-L, LoCascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics. 2010;11:119. https://doi.org/10.1186/1471-2105-11-119. 60. Fu L, Niu B, Zhu Z, Wu S, Li W. CD-HIT: accelerated for clustering the next- generation sequencing data. Bioinformatics. 2012;28(23):3150–2. https://doi. org/10.1093/bioinformatics/bts565. 61. Buchfink B, Xie C, Huson DH. Fast and sensitive protein alignment using DIAMOND. Nat Methods. 2015;12(1):59–60. https://doi.org/10.1038/nmeth.3176. 62. Huson DH, Beier S, Flade I, Gorska A, El-Hadidi M, Mitra S, et al. MEGAN Community edition - interactive exploration and analysis of large-scale microbiome sequencing data. PLoS Comput Biol. 2016;12(6):e1004957. https://doi.org/10.1371/journal.pcbi.1004957. 63. Langmead B, Salzberg SL. Fast gapped-read alignment with bowtie 2. Nat Methods. 2012;9(4):357–9. https://doi.org/10.1038/nmeth.1923. 64. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics (Oxford, England). 2009;25(16):2078–9. https://doi.org/10.1093/bioinformatics/btp352. 65. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PloS one. 2013;8(4): e61217. https://doi.org/10.1371/journal.pone.0061217. 66. Green MR, Sambrook J. Precipitation of DNA with Ethanol. Cold Spring Harbor Protocols. 2016;2016(12); doi: https://doi.org/10.1101/pdb.prot093377. 67. Shimeta J, Cook PLM. Testing assumptions of the eukaryotic inhibitor method for investigating interactions between aquatic protozoa and bacteria, applied to marine sediment. Limnol Oceanogr Methods. 2011;9(7): 288–95. Availability of data and materials Th d d d/ l https://doi.org/10.4319/lom.2011.9.288. 68. Law Y, Kirkegaard RH, Cokro AA, Liu X, Arumugam K, Xie C, Stokholm- Bjerregaard M, Drautz-Moses DI, Nielsen PH, Wuertz S, Williams RBH. Integrative microbial community analysis reveals full-scale enhanced biological phosphorus removal under tropical conditions. Sci Rep. 2016;6(1): 25719. https://doi.org/10.1038/srep25719. 69. Feng S, Tan CH, Constancias F, Kohli GS, Cohen Y, Rice SA. Predation by Bdellovibrio bacteriovorus significantly reduces viability and alters the microbial community composition of activated sludge flocs and granules. FEMS Microbiol Ecol. 2017;93(4):fix020-fix; https://doi.org/10.1093/femsec/ fix020. 70. Zachary D. Kurtz, Christian L. Müller, Emily R. Miraldi, Dan R. Littman, Martin J. Blaser, Richard A. Bonneau, Christian von Mering. Sparse and Compositionally Robust Inference of Microbial Ecological Networks. PLoS Comput Biol. 2015;11(5):e1004226. 71. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community Ecology Package. In., 2.5–6 edn; 2019. 71. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community Ecology Package. In., 2.5–6 edn; 2019. Chan et al. BMC Microbiology (2021) 21:91 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under the CCAL, authors retain copyright to the article but users are allowed to download, reprint, distribute and /or copy articles in BioMed Central journals, as long as the original work is properly cited. BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under the CCAL, authors retain copyright to the article but users are allowed to download, reprint, distribute and /or copy articles in BioMed Central journals, as long as the original work is properly cited. BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under the CCAL, authors retain copyright to the article but users are allowed to download, reprint, distribute and /or copy articles in BioMed Central journals, as long as the original work is properly cited.
https://openalex.org/W4252555486	http://publikationen.ub.uni-frankfurt.de/files/42439/amtd-8-13257-2015.pdf	English	null	Technical note: Detection of dimethylamine in the low pptv range using nitrate Chemical Ionization-Atmospheric Pressure interface-Time Of Flight (CI-APi-TOF) mass spectrometry	null	2,015	cc-by	19,423	Detection of DMA in the low pptv range using a nitrate CIMS Detection of DMA in the low pptv range using a nitrate CIMS AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Pa Discussion Paper Atmos. Meas. Tech. Discuss., 8, 13257–13284, 2015 www.atmos-meas-tech-discuss.net/8/13257/2015/ doi:10.5194/amtd-8-13257-2015 © Author(s) 2015. CC Attribution 3.0 License. Atmos. Meas. Tech. Discuss., 8, 13257–13284, 2015 www.atmos-meas-tech-discuss.net/8/13257/2015/ doi:10.5194/amtd-8-13257-2015 © Author(s) 2015. CC Attribution 3.0 License. iscussion Paper \| Discussion Paper \| Discussion Paper AMTD 8, 13257–13284, 2015 This discussion paper is/has been under review for the journal Atmospheric Measurement Techniques (AMT). Please refer to the corresponding ﬁnal paper in AMT if available. mass spectrometry M. Simon1, M. Heinritzi1, S. Herzog1, M. Leiminger1, F. Bianchi2,3, A. Praplan4, J. Dommen2, J. Curtius1, and A. Kürten1 1Institute for Atmospheric and Environmental Sciences, Goethe University of Frankfurt, 60438 Frankfurt am Main, Germany 2Laboratory of Atmospheric Chemistry, Paul-Scherrer-Institute, 5232 Villigen, Switzerland 3Institute for Atmospheric and Climate Science, ETH Zurich, 8092 Zurich, Switzerland 4Helsinki Institute of Physics, University of Helsinki, 00014 Helsinki, Finland 13257 Technical note: Detection of dimethylamine in the low pptv range using nitrate Chemical Ionization-Atmospheric Pressure interface-Time Of Flight (CI-APi-TOF) mass spectrometry M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion scussion Paper \| Discussion Paper \| Discussion Paper \| M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion scussion Paper \| Discussion Paper \| Discussion Paper \| 13257 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper Received: 30 September 2015 – Accepted: 19 November 2015 – Published: 17 December 2015 Received: 30 September 2015 – Accepted: 19 November 2015 – Published: 17 December 2015 Discussion Paper \| Discussion Paper \| Discussion Paper 1 Introduction The gas-phase abundance of amines in the atmosphere received considerable atten- tion recently as amines are potentially an important agent contributing to atmospheric 15 aerosol nucleation events in those regions where amines are emitted. A large variety of diﬀerent amines exists in the atmosphere and various sources of amines are known such as animal husbandry or sewage, nevertheless, the gas phase concentrations of amines are expected to be low due to rapid uptake into acidic aerosols and high solu- bility (Ge et al., 2011). Despite concentrations expected to be typically 10 to 1000 times 20 below atmospheric gas phase ammonia levels, amines such as methyl-, dimethyl- or trimethylamine were postulated to enhance the nucleation of sulfuric acid much more eﬃciently than NH3 (Kurtén et al., 2008). Furthermore, it was found that typical con- centration levels of H2SO4 and NH3 in the boundary layer are too low to explain aerosol formation rates as frequently observed during nucleation events via nucleation mech- 25 anisms such as binary H2SO4-H2O or NH3-ternary nucleation (Kirkby et al., 2011). The gas-phase abundance of amines in the atmosphere received considerable atten- tion recently as amines are potentially an important agent contributing to atmospheric 15 aerosol nucleation events in those regions where amines are emitted. A large variety of diﬀerent amines exists in the atmosphere and various sources of amines are known such as animal husbandry or sewage, nevertheless, the gas phase concentrations of amines are expected to be low due to rapid uptake into acidic aerosols and high solu- Paper \| Discussion Paper \| p p p g bility (Ge et al., 2011). Despite concentrations expected to be typically 10 to 1000 times 20 below atmospheric gas phase ammonia levels, amines such as methyl-, dimethyl- or trimethylamine were postulated to enhance the nucleation of sulfuric acid much more eﬃciently than NH3 (Kurtén et al., 2008). Furthermore, it was found that typical con- centration levels of H2SO4 and NH3 in the boundary layer are too low to explain aerosol formation rates as frequently observed during nucleation events via nucleation mech- 25 anisms such as binary H SO H O or NH ternary nucleation (Kirkby et al 2011) p p p g bility (Ge et al., 2011). AMTD Amines are potentially important for atmospheric new particle formation and there- fore the demand for highly sensitive gas phase amine measurements has emerged in the last several years. Nitrate Chemical Ionization Mass Spectrometry (CIMS) is y p y ( ) routinely used for the measurement of gas phase-sulfuric acid in the sub-pptv range. 5 Furthermore, Extremely Low Volatile Organic Compounds (ELVOCs) can be detected with a nitrate CIMS. In this study we demonstrate that a nitrate CIMS can also be used for the sensitive measurement of dimethylamine ((CH3)2NH, DMA) using the NO− 3 (HNO3)1−2(DMA) cluster ion signals. This observation was made at the CLOUD aerosol chamber, which was also used for calibration measurements. Good linearity 10 between 0 and ∼120 pptv of DMA as well as a sub-pptv detection limit of 0.7 pptv for a 10 min integration time are demonstrated at 278 K and 38 % RH. routinely used for the measurement of gas phase-sulfuric acid in the sub-pptv range. 5 Furthermore, Extremely Low Volatile Organic Compounds (ELVOCs) can be detected with a nitrate CIMS. In this study we demonstrate that a nitrate CIMS can also be used for the sensitive measurement of dimethylamine ((CH3)2NH, DMA) using the NO− 3 (HNO3)1−2(DMA) cluster ion signals. This observation was made at the CLOUD l h b hi h l d f lib i G d li i Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion 3 aerosol chamber, which was also used for calibration measurements. Good linearity 10 between 0 and ∼120 pptv of DMA as well as a sub-pptv detection limit of 0.7 pptv for a 10 min integration time are demonstrated at 278 K and 38 % RH. 3 aerosol chamber, which was also used for calibration measurements. Good linearity 10 between 0 and ∼120 pptv of DMA as well as a sub-pptv detection limit of 0.7 pptv for a 10 min integration time are demonstrated at 278 K and 38 % RH. AMTD Correspondence to: M. Simon (simon@iau.uni-frankfurt.de) 8, 13257–13284, 2015 Published by Copernicus Publications on behalf of the European Geosciences Union. Published by Copernicus Publications on behalf of the European Geosciences Union. 13258 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Pap Discussion Paper Abstract AMTD DMA) that already the presence of a few pptv enhances the aerosol formation rates of 5 sulfuric acid by several orders of magnitude and formation rates that are typical for at- mospheric nucleation events are observed. Kürten et al. (2014) studied the formation of neutral (i.e. uncharged) H2SO4-DMA clusters and showed that the cluster formation process proceeds at or near the kinetic limit. This means that already for the low abun- Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion dances of H2SO4 and DMA (H2SO4 at sub-pptv level, DMA at pptv level) the growth is 10 limited only by the arrival rate of H2SO4 molecules and an eﬃcient acid-base stabiliza- tion prevents even the smallest H2SO4-DMA clusters (i.e. the sulfuric acid dimer) from evaporation. Evidence for the participation of amines in aerosol nucleation near ground has been found (e.g. Mäkelä et al., 2001; Smith et al., 2010; Zhao et al., 2011; Creamean et al., 15 2011; Yu et al., 2012; Chen et al., 2012). However, the extent to which amines are in- deed participating in atmospheric nucleation is still not established. This lack of knowl- edge is to a large degree due to the diﬃculty of measuring amines in real-time at low pptv to sub-pptv mixing ratios. Mass spectrometric methods using chemical ionization p p g found (e.g. Mäkelä et al., 2001; Smith et al., 2010; Zhao et al., 2011; Creamean et al., 15 2011; Yu et al., 2012; Chen et al., 2012). However, the extent to which amines are in- deed participating in atmospheric nucleation is still not established. This lack of knowl- edge is to a large degree due to the diﬃculty of measuring amines in real-time at low pptv to sub-pptv mixing ratios. Mass spectrometric methods using chemical ionization g g (CIMS) have become available for amine measurements. These methods have suﬃ- 20 cient time resolution and a low enough limit of detection for atmospherically relevant mixing ratios. Various amines were detected by positive-ion chemical ionization via am- bient pressure proton transfer (Hanson et al., 2011; Freshour et al., 2014). Protonated ethanol or acetone ions were used as reagent ions by Yu and Lee (2012). 1 Introduction Despite concentrations expected to be typically 10 to 1000 times 20 below atmospheric gas phase ammonia levels, amines such as methyl-, dimethyl- or trimethylamine were postulated to enhance the nucleation of sulfuric acid much more eﬃciently than NH3 (Kurtén et al., 2008). Furthermore, it was found that typical con- centration levels of H2SO4 and NH3 in the boundary layer are too low to explain aerosol formation rates as frequently observed during nucleation events via nucleation mech- 25 anisms such as binary H2SO4-H2O or NH3-ternary nucleation (Kirkby et al., 2011). 13259 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper Discussion Paper Participation of amines in nucleation was studied in the laboratory for the amine- sulfuric acid-water system (e.g. Berndt et al., 2010, 2014; Erupe et al., 2011; Zollner et al., 2012; Almeida et al., 2013; Kürten et al., 2014; Bianchi et al., 2014; Jen et al., 2015; Glasoe et al., 2015). Almeida et al. (2013) showed for dimethylamine ((CH3)2NH, AMTD ussion Paper \| Discussion Paper \| Discussion Paper \| Kürten et al., 2011) and also for the detection of Extremely Low Volatile Organic Com pounds (ELVOC, Ehn et al., 2014). The simultaneous detection of DMA as well as ELVOCs and sulfuric acid by one instrument at high time resolution is very useful as 5 H2SO4 and ELVOC measurements are already performed when aerosol nucleation is studied. The method and the detection scheme are described in detail; absolute concentrations are derived by calculating the DMA mixing ratios from the balance of sources and sinks in the CLOUD chamber. DMA detection limits are discussed. Fur- thermore, the method is also compared with DMA measurements by ion chromatogra- 10 phy (Praplan et al., 2012). ELVOCs and sulfuric acid by one instrument at high time resolution is very useful as 5 H2SO4 and ELVOC measurements are already performed when aerosol nucleation is studied. The method and the detection scheme are described in detail; absolute concentrations are derived by calculating the DMA mixing ratios from the balance of sources and sinks in the CLOUD chamber. DMA detection limits are discussed. Fur- ELVOCs and sulfuric acid by one instrument at high time resolution is very useful as 5 H2SO4 and ELVOC measurements are already performed when aerosol nucleation is studied. The method and the detection scheme are described in detail; absolute concentrations are derived by calculating the DMA mixing ratios from the balance of sources and sinks in the CLOUD chamber. DMA detection limits are discussed. Fur- Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion thermore, the method is also compared with DMA measurements by ion chromatogra- 10 phy (Praplan et al., 2012). thermore, the method is also compared with DMA measurements by ion chromatogra- 10 phy (Praplan et al., 2012). AMTD Negative-ion detection of amines using bisulfate reagent ions has been described recently (Sipilä 25 et al., 2015). Here we describe the detection of gas phase DMA at sub-pptv levels at the CLOUD aerosol chamber at CERN by use of a nitrate Chemical Ionization-Atmospheric Pres- sure interface-Time Of Flight-Mass Spectrometer (CI-APi-TOF-MS, Jokinen et al., (CIMS) have become available for amine measurements. These methods have suﬃ- 20 cient time resolution and a low enough limit of detection for atmospherically relevant mixing ratios. Various amines were detected by positive-ion chemical ionization via am- bient pressure proton transfer (Hanson et al., 2011; Freshour et al., 2014). Protonated ethanol or acetone ions were used as reagent ions by Yu and Lee (2012). Negative-ion d t ti f i i bi lf t t i h b d ib d tl (Si ilä \| Discussion Paper \| ethanol or acetone ions were used as reagent ions by Yu and Lee (2012). Negative ion detection of amines using bisulfate reagent ions has been described recently (Sipilä 25 et al., 2015). Here we describe the detection of gas phase DMA at sub-pptv levels at the CLOUD aerosol chamber at CERN by use of a nitrate Chemical Ionization-Atmospheric Pres- sure interface-Time Of Flight-Mass Spectrometer (CI-APi-TOF-MS, Jokinen et al., ethanol or acetone ions were used as reagent ions by Yu and Lee (2012). Negative ion detection of amines using bisulfate reagent ions has been described recently (Sipilä 25 et al., 2015). Here we describe the detection of gas phase DMA at sub-pptv levels at the CLOUD aerosol chamber at CERN by use of a nitrate Chemical Ionization-Atmospheric Pres- sure interface-Time Of Flight-Mass Spectrometer (CI-APi-TOF-MS, Jokinen et al., Here we describe the detection of gas phase DMA at sub-pptv levels at the CLOUD aerosol chamber at CERN by use of a nitrate Chemical Ionization-Atmospheric Pres- sure interface-Time Of Flight-Mass Spectrometer (CI-APi-TOF-MS, Jokinen et al., 13260 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Di i P \| Di i P \| Di i P \| Di i P \| 2012; Kürten et al., 2014). AMTD Nitrate chemical ionization mass spectrometry is already used frequently for the highly sensitive detection of H2SO4 (Tanner and Eisele, 1991; Kürten et al., 2011) and also for the detection of Extremely Low Volatile Organic Com- pounds (ELVOC, Ehn et al., 2014). The simultaneous detection of DMA as well as ELVOCs and sulfuric acid by one instrument at high time resolution is very useful as 5 H2SO4 and ELVOC measurements are already performed when aerosol nucleation is studied. The method and the detection scheme are described in detail; absolute concentrations are derived by calculating the DMA mixing ratios from the balance of sources and sinks in the CLOUD chamber. DMA detection limits are discussed. Fur- thermore, the method is also compared with DMA measurements by ion chromatogra- 10 phy (Praplan et al., 2012). Discussion Paper \| Discussion Pape 2012; Kürten et al., 2014). Nitrate chemical ionization mass spectrometry is already used frequently for the highly sensitive detection of H2SO4 (Tanner and Eisele, 1991; Kürten et al., 2011) and also for the detection of Extremely Low Volatile Organic Com- pounds (ELVOC, Ehn et al., 2014). The simultaneous detection of DMA as well as 2012; Kürten et al., 2014). Nitrate chemical ionization mass spectrometry is already used frequently for the highly sensitive detection of H2SO4 (Tanner and Eisele, 1991; Kürten et al., 2011) and also for the detection of Extremely Low Volatile Organic Com- pounds (ELVOC, Ehn et al., 2014). The simultaneous detection of DMA as well as AMTD The volume mixing ratio (VMR) of DMA (in pptv) inside 15 the CLOUD chamber can be derived from the following diﬀerential equation: dard temperature and pressure, in this case 293.15 K and 1013 hPa), and the quantity ADMA is the ﬂow rate of DMA. The volume mixing ratio (VMR) of DMA (in pptv) inside 15 the CLOUD chamber can be derived from the following diﬀerential equation: p p ) q y ADMA is the ﬂow rate of DMA. The volume mixing ratio (VMR) of DMA (in pptv) inside 15 the CLOUD chamber can be derived from the following diﬀerential equation: dVMRDMA = ADMA × 1012 pptv −k ll · VMRDMA −kdil · VMRDMA (2) p p ) q y ADMA is the ﬂow rate of DMA. The volume mixing ratio (VMR) of DMA (in pptv) inside 15 the CLOUD chamber can be derived from the following diﬀerential equation: dVMRDMA dt = ADMA Vch × 1012 pptv −kwall · VMRDMA −kdil · VMRDMA. (2) Here, Vch is the chamber volume (2.61×107 cm3, where Vch denotes a physical volume), kwall is the wall loss rate constant for DMA and kdil is the dilution rate constant. The DMA g ( ) ( pp ) the CLOUD chamber can be derived from the following diﬀerential equation: dVMRDMA dt = ADMA Vch × 1012 pptv −kwall · VMRDMA −kdil · VMRDMA. (2) VMRDMA dt = ADMA Vch × 1012 pptv −kwall · VMRDMA −kdil · VMRDMA. dVMRDMA dt = ADMA Vch × 1012 pptv −kwall · VMRDMA −kdil · VMRDMA. (2) (2) Here, Vch is the chamber volume (2.61×107 cm3, where Vch denotes a physical volume), kwall is the wall loss rate constant for DMA and kdil is the dilution rate constant. The kwall is the wall loss rate constant for DMA and kdil is the dilution rate constant. The dilution rate constant can be calculated from the ratio of the clean gas ﬂow rate that 20 is required to replenish the gas taken by the instruments and the chamber volume. In this study, the ﬂow rate of air into the chamber is 160 Lmin−1 at standard temperature and pressure, which yields a dilution rate constant (assuming homogenous mixing) of 1×10−4 s−1. 2.1 CLOUD facility The Cosmics Leaving OUtdoor Droplets Chamber (CLOUD) at CERN is a 26 m3 cylin- drical vessel to study aerosol processes such as new particle formation. The inner 15 surfaces consist of electropolished stainless steel. Care is taken to minimize contami- nation with any condensable substances that may inﬂuence new particle formation. The chamber and its components have been described in detail before (Kirkby et al., 2011; Kupc et al., 2011; Voigtländer et al., 2012). Results are reported from the CLOUD7 p g ) p and CLOUD10-T experiments (October–December 2012 and April–May 2015) in which 20 the aerosol nucleation for the sulfuric acid-water-dimethylamine system was studied (Almeida et al., 2013; Kürten et al., 2014). All measurements were carried out at a tem- perature of 278 K and a relative humidity of 38 % in the chamber. p g ) p and CLOUD10-T experiments (October–December 2012 and April–May 2015) in which 20 the aerosol nucleation for the sulfuric acid-water-dimethylamine system was studied (Almeida et al., 2013; Kürten et al., 2014). All measurements were carried out at a tem- perature of 278 K and a relative humidity of 38 % in the chamber. \| Discussion Paper 13261 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| mber is shown in Fig. 1. amine that is fed into the nted at CLOUD. The gas ed mass ﬂow controllers ottle with clean air before calculated from the frac- ee Fig. 1). When the by- LOUD7 and CLOUD10-T, (1) units of cm3 s−1 (at stan- 3 hPa), and the quantity ) of DMA (in pptv) inside ntial equation: A. (2) notes a physical volume), ution rate constant. The clean gas ﬂow rate that the chamber volume. In at standard temperature homogenous mixing) of point it can be compared AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Discussion Paper 2.2 Gas system and calculated DMA mixing ratios AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Di i P \| Di i P \| Di i P \| Di i P AMTD A schematic drawing of the gas system and the CLOUD chamber is shown in Fig. 1. In order to have precise control over the amount of dimethylamine that is fed into the chamber a specially designed gas system has been implemented at CLOUD. The gas A schematic drawing of the gas system and the CLOUD chamber is shown in Fig. 1. In order to have precise control over the amount of dimethylamine that is fed into the chamber a specially designed gas system has been implemented at CLOUD. The gas system for each individual trace gas includes three calibrated mass ﬂow controllers 5 (MFCs) and several valves for diluting a mixture from a gas bottle with clean air before system for each individual trace gas includes three calibrated mass ﬂow controllers 5 (MFCs) and several valves for diluting a mixture from a gas bottle with clean air before it is fed into the chamber close to the lower mixing fan. The amount of DMA introduced into the chamber can be calculated from the frac- tion B of DMA inside the gas bottle and the MFC ﬂow rates (see Fig. 1). When the by- The amount of DMA introduced into the chamber can be calculated from the frac- tion B of DMA inside the gas bottle and the MFC ﬂow rates (see Fig. 1). When the by- pass valve is closed, which was the case at all times during CLOUD7 and CLOUD10-T, 10 the following amount of DMA enters the chamber: pass valve is closed, which was the case at all times during CLOUD7 and CLOUD10-T, 10 the following amount of DMA enters the chamber: ADMA = MFC1 · MFC3 MFC1 + MFC2 · B. (1) ADMA = MFC1 · MFC3 MFC1 + MFC2 · B. ADMA = MFC1 · MFC3 MFC1 + MFC2 · B. (1) (1) The ﬂow rates (denoted with MFC1, MFC2, and MFC3) have units of cm3 s−1 (at stan- dard temperature and pressure, in this case 293.15 K and 1013 hPa), and the quantity The ﬂow rates (denoted with MFC1, MFC2, and MFC3) have units of cm3 s−1 (at stan- dard temperature and pressure, in this case 293.15 K and 1013 hPa), and the quantity ADMA is the ﬂow rate of DMA. AMTD Alternatively, 10 the wall loss rate can be determined from the decay rate of the signal, which is indicat- ing the DMA mixing ratio in this study (see Sect. 3.1). The factor F describes the addition of DMA to the chamber in units of pptvs . The unknown quantity kwall could in principle be derived from Eq. (3) by calibration experi- ments if a reference instrument for the measurement of DMA were used. Alternatively, 10 the wall loss rate can be determined from the decay rate of the signal, which is indicat- ing the DMA mixing ratio in this study (see Sect. 3.1). ments if a reference instrument for the measurement of DMA were used. Alternatively, 10 the wall loss rate can be determined from the decay rate of the signal, which is indicat- ing the DMA mixing ratio in this study (see Sect. 3.1). It should be noted that the eﬀect of DMA condensation on aerosol particles, which are formed during nucleation experiments, is not taken into account in Eqs. (2) and (3). For the data shown in this study, either no sulfuric acid was generated when the DMA 15 calibration measurements (see Sect. 3.2) were performed, or the condensation sink was so low that it had no eﬀect on the DMA mixing ratio. It should be noted that the eﬀect of DMA condensation on aerosol particles, which are formed during nucleation experiments, is not taken into account in Eqs. (2) and (3). For the data shown in this study, either no sulfuric acid was generated when the DMA 15 calibration measurements (see Sect. 3.2) were performed, or the condensation sink was so low that it had no eﬀect on the DMA mixing ratio. For the data shown in this study, either no sulfuric acid was generated when the DMA 15 calibration measurements (see Sect. 3.2) were performed, or the condensation sink was so low that it had no eﬀect on the DMA mixing ratio. g The assumption that the DMA mixing ratio is at equilibrium inside the pipes once the chamber valve is opened, i.e. that wall loss is negligible for the DMA inlet lines, is The assumption that the DMA mixing ratio is at equilibrium inside the pipes once the chamber valve is opened, i.e. AMTD The wall loss rate is not known a priori but at this point it can be compared wall dil dilution rate constant can be calculated from the ratio of the clean gas ﬂow rate that 20 is required to replenish the gas taken by the instruments and the chamber volume. In this study, the ﬂow rate of air into the chamber is 160 Lmin−1 at standard temperature and pressure, which yields a dilution rate constant (assuming homogenous mixing) of 1×10−4 s−1. The wall loss rate is not known a priori but at this point it can be compared 13262 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper to the one for sulfuric acid which has been experimentally determined as 2.2×10−3 s−1. Assuming that the walls act as a perfect sink the wall loss rate can be assumed to be proportional to the square root of the gas-phase diﬀusion coeﬃcient (Crump and Seinfeld, 1981) and should therefore be faster for DMA because it is a lighter molecule compared to sulfuric acid. Assuming steady-state in Eq. (2) yields Assuming steady-state in Eq. (2) yields VMRDMA = ADMA × 1012 pptv Vch · (kwall + kdil) = MFC1 · MFC3 MFC1 + MFC2 · B × 1012 pptv Vch · (kwall + kdil) = F kwall + kdil . (3) (3) M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion cussion Paper \| Discussion Paper \| Discussion Paper The factor F describes the addition of DMA to the chamber in units of pptvs−1. The unknown quantity kwall could in principle be derived from Eq. (3) by calibration experi- The factor F describes the addition of DMA to the chamber in units of pptvs−1. The unknown quantity kwall could in principle be derived from Eq. (3) by calibration experi- ments if a reference instrument for the measurement of DMA were used. Detection of DMA in the low pptv range using a nitrate CIMS Detection of DMA in the low pptv range using a nitrate CIMS Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion they also allow for the detection of extremely low volatile organic compounds (ELVOCs, 10 see e.g. Ehn et al., 2014). As will be described in the next section dimethylamine ((CH3)2NH, DMA) can mainly be detected at integer mass m/z 170 Th (NO− 3 (HNO3)(DMA) ion) and m/z 233 Th (NO− 3 (HNO3)2(DMA) ion); however, the exact masses of these ion clusters are they also allow for the detection of extremely low volatile organic compounds (ELVOCs, 10 see e.g. Ehn et al., 2014). As will be described in the next section dimethylamine ((CH3)2NH, DMA) can mainly be detected at integer mass m/z 170 Th (NO− 3 (HNO3)(DMA) ion) and m/z 233 Th (NO− 3 (HNO3)2(DMA) ion); however, the exact masses of these ion clusters are (NO3 (HNO3)2(DMA) ion); however, the exact masses of these ion clusters are 170.0419 and 233.0375 Th due to their mass defect. Owing to the high mass resolving 15 power (∼4500 ThTh−1) and the high mass accuracy (better than 10 ppm) of the CI-APi- TOF these ions can be unambiguously identiﬁed in this study if additional information like the isotopic pattern is taken into account. Especially in ﬁeld measurements, where a lot of unknown compounds are potentially present, the high mass resolving power al- lows distinguishing between diﬀerent ion species having the same integer mass which 20 3 3 2 170.0419 and 233.0375 Th due to their mass defect. Owing to the high mass resolving 15 power (∼4500 ThTh−1) and the high mass accuracy (better than 10 ppm) of the CI-APi- TOF these ions can be unambiguously identiﬁed in this study if additional information like the isotopic pattern is taken into account. Especially in ﬁeld measurements, where a lot of unknown compounds are potentially present, the high mass resolving power al- p p y p g g p lows distinguishing between diﬀerent ion species having the same integer mass which 20 minimizes potential interferences. AMTD 13263 13263 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Pa Discussion Paper 2.3 CI-APi-TOF instrument 2.3 AMTD that wall loss is negligible for the DMA inlet lines, is The assumption that the DMA mixing ratio is at equilibrium inside the pipes once the chamber valve is opened, i.e. that wall loss is negligible for the DMA inlet lines, is justiﬁed due to the following reasons. First, the gas lines are conditioned over a duration 20 of at least 24 h before DMA is added to the chamber for the ﬁrst time. During this time the purge valve is open and the chamber valve is closed (Fig. 1). Only the last ∼23 cm between the chamber valve and the point where the DMA enters the chamber are therefore not conditioned. Second, the mixing ratio of DMA inside the gas lines is generally higher than several tens of ppbv even though the DMA inside the chamber is 25 in the pptv-range due to the strong dilution inside the chamber. The high DMA mixing ratio enables a rapid equilibration of the short unconditioned section of the gas lines r \| Discussion Paper \| the chamber valve is opened, i.e. that wall loss is negligible for the DMA inlet lines, is justiﬁed due to the following reasons. First, the gas lines are conditioned over a duration 20 of at least 24 h before DMA is added to the chamber for the ﬁrst time. During this time the purge valve is open and the chamber valve is closed (Fig. 1). Only the last ∼23 cm between the chamber valve and the point where the DMA enters the chamber are therefore not conditioned. Second, the mixing ratio of DMA inside the gas lines is justiﬁed due to the following reasons. First, the gas lines are conditioned over a duration 20 of at least 24 h before DMA is added to the chamber for the ﬁrst time. During this time the purge valve is open and the chamber valve is closed (Fig. 1). Only the last ∼23 cm between the chamber valve and the point where the DMA enters the chamber are therefore not conditioned. Second, the mixing ratio of DMA inside the gas lines is generally higher than several tens of ppbv even though the DMA inside the chamber is 25 in the pptv-range due to the strong dilution inside the chamber. The high DMA mixing ratio enables a rapid equilibration of the short unconditioned section of the gas lines. AMTD The Chemical Ionization-Atmospheric Pressure interface-Time Of Flight (CI-APi-TOF) mass spectrometer has been described recently (Jokinen et al., 2012; Kürten et al., p y ( 2014). The CI-APi-TOF combines an atmospheric pressure chemical ionization source 2014). The CI-APi-TOF combines an atmospheric pressure che based on the design by Eisele and coworkers (Eisele and Tanner, 1993) and a high res- 5 olution atmospheric pressure interface time-of-ﬂight mass spectrometer (Tofwerk AG, Switzerland). The ion source uses a corona discharge for the primary ion generation (Kürten et al., 2011). Nitrate ions (NO− 3 (HNO3)x=0−3) are generally used for the detec- tion of sulfuric acid and sulfuric acid-amine clusters but more recently it was found that AMTD cussion Paper \| Discussion Paper \| Discussion Paper \| Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Detection of DMA in the low pptv range using a nitrate CIMS On the contrary to CLOUD10-T where the CI-APi-TOF was connected to the cham- ber by its own sampling line in CLOUD7 the instrument was connected to the cham- ber with a y-splitter. Therefore, the sampling line losses cannot be easily calculated in the same way as for a straight tube and laminar ﬂow. Instead, the eﬀective length 25 method (Karlsson and Martinsson, 2003) was used after comparing the sulfuric acid concentrations measured by a Chemical Ionization Mass Spectrometer (CIMS) and the CI-APi-TOF simultaneously. Since the CIMS was connected to the CLOUD chamber p On the contrary to CLOUD10-T where the CI-APi-TOF was connected to the cham- ber by its own sampling line in CLOUD7 the instrument was connected to the cham- ber with a y-splitter. Therefore, the sampling line losses cannot be easily calculated in the same way as for a straight tube and laminar ﬂow. Instead, the eﬀective length 25 method (Karlsson and Martinsson, 2003) was used after comparing the sulfuric acid concentrations measured by a Chemical Ionization Mass Spectrometer (CIMS) and the CI-APi-TOF simultaneously. Since the CIMS was connected to the CLOUD chamber \| Discussion Paper \| 13264 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper with its own dedicated sampling line, the loss rate could be quantiﬁed for sulfuric acid. Taking into account the independently determined calibration constant regarding sul- furic acid for the CIMS and the CI-APi-TOF (Kürten et al., 2012; Kürten et al., 2014) an eﬀective length of ∼1.5 m could be determined for the CI-APi-TOF sampling line at a ﬂow rate of 8.5 Lmin−1 (at standard pressure and temperature) for sulfuric acid. These values are used in the next section for deriving the transmission eﬃciency of DMA to the CI-APi-TOF. Discussion Paper \| Discussion Pa 2.4 Ionization process In contrast to the detection of gaseous sulfuric acid (H2SO4) In contrast to the detection of gaseous sulfuric acid (H2SO4) by a proton transfer re- In contrast to the detection of gaseous sulfuric acid (H2SO4) by a proton transfer re- action which leads to the formation of bisulfate ions (HSO− 4 , Eisele and Tanner, 1993), 10 dimethylamine is detected due to its ability to cluster with the NO− 3 (HNO3)x 1 3 primary g ( 2 4) y p action which leads to the formation of bisulfate ions (HSO− 4 , Eisele and Tanner, 1993), 10 dimethylamine is detected due to its ability to cluster with the NO− 3 (HNO3)x = 1−3 primary ions: DMA + NO− 3 (HNO3) →NO− 3 (HNO3)(DMA) (R1) DMA + NO− 3 (HNO3)≥2 →NO− 3 (HNO3)≥2(DMA). (R2) DMA + NO− 3 (HNO3) →NO− 3 (HNO3)(DMA) (R1) DMA + NO− 3 (HNO3)≥2 →NO− 3 (HNO3)≥2(DMA). (R2) DMA + NO− 3 (HNO3) →NO− 3 (HNO3)(DMA) (R1) DMA + NO− 3 (HNO3)≥2 →NO− 3 (HNO3)≥2(DMA). (R2) DMA + NO− 3 (HNO3) →NO− 3 (HNO3)(DMA) DMA + NO− 3 (HNO3)≥2 →NO− 3 (HNO3)≥2(DMA). DMA + NO− 3 (HNO3) →NO− 3 (HNO3)(DMA) DMA + NO− 3 (HNO3)≥2 →NO− 3 (HNO3)≥2(DMA). (R1) (R2) The association Reactions (R1) and (R2) could both occur in the ion-molecule reaction 15 zone of the CI-APi-TOF, however, DMA is mainly detected at integer mass m/z 170 Th (NO− 3 (HNO3)(DMA) ions) and integer mass m/z 233 Th (NO− 3 (HNO3)2(DMA)) (Fig. 2b and c). In addition, a small signal at m/z 296 Th (NO− 3 (HNO3)3(DMA)) is also visible in the mass spectra at high DMA mixing ratios (Fig. 2d). Therefore, in principle it can- The association Reactions (R1) and (R2) could both occur in the ion-molecule reaction 15 zone of the CI-APi-TOF, however, DMA is mainly detected at integer mass m/z 170 Th (NO− 3 (HNO3)(DMA) ions) and integer mass m/z 233 Th (NO− 3 (HNO3)2(DMA)) (Fig. 2b and c). In addition, a small signal at m/z 296 Th (NO− 3 (HNO3)3(DMA)) is also visible in the mass spectra at high DMA mixing ratios (Fig. 2d). Therefore, in principle it can- p g g ( g ) p p not be ruled out that Reaction (R2) is the more important reaction and subsequent 20 fragmentation in the APi section is the reason for the dominant signal at m/z 170 Th. Regarding the identiﬁcation and quantiﬁcation of the mentioned signals it should be noted that although integer masses are mentioned, the CI-APi-TOF mass spectra are evaluated using high-resolution data. The data is processed using tofTools developed b th U i it f H l i ki (J i t l 2010) \| Discussion Paper by the University of Helsinki (Junninen et al., 2010). 25 To quantify the DMA concentration, the sum of the two most intense DMA signals are normalized by the primary ion count rate at m/z 188 Th (NO− 3 (HNO3)2). Tentatively, we 13265 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Pape AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. AMTD ) p ion clusters involving sulfuric acid and ammonia or dimethylamine (see Kirkby et al., 2011; Almeida et al., 2013; Kürten et al., 2014; Bianchi et al., 2014). This leads to the following equation for the DMA concentration: [DMA] = C · T · ln 1 + (CR170 + CR233) CR188 ≈C · T · (CR170 + CR233) CR188 . (4) [DMA] = C · T · ln 1 + (CR170 + CR233) CR188 ≈C · T · (CR170 + CR233) CR188 . (4) (4) M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion i P \| Di i P \| Di i P The factor C (in molecule cm−3) can be derived from calibration measurements using the CLOUD chamber. Generally DMA mixing ratios are reported rather than concen- 10 trations and therefore the calibration factor has a diﬀerent unit than in Eq. (4). However, the derived calibration constant can be converted to molecule cm−3 (see Sect. 3.2). An additional factor T is required to take into account losses of DMA molecules in the CI- APi-TOF sampling line during the transport from the chamber to the instrument; the pa- p g g p p rameters CR170, CR233, and CR188 denote the count rates at the exact masses for the 15 NO− 3 (HNO3)1,2(DMA) and the NO− 3 (HNO3)2 ions, respectively. The factor T has a value of ∼2.5 for CLOUD10-T. During CLOUD7 the factor has a value of ∼4 and was eval- uated based on the eﬀective length method mentioned in the previous section. For the evaluation of T it was assumed that the diﬀusivity of H2SO4 at 278 K and 38 % relative humidity equals 0.07 cm2 s−1 (Hanson and Eisele, 2000), while a value of 0.10 cm2 s−1 20 was assumed for DMA which follows when applying a power dependency of 1.75 for the diﬀusion coeﬃcient regarding the temperature for a reported value (Freshour et al., 2014). To our knowledge the existence of ion clusters containing amines and nitrate \| Discussion Paper \| To our knowledge the existence of ion clusters containing amines and nitrate has been reported only once by Luts et al. (2011). DMA + NO− 3 (HNO3) →NO− 3 (HNO3)(DMA) DMA + NO− 3 (HNO3)≥2 →NO− 3 (HNO3)≥2(DMA). Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion have chosen the NO− 3 (HNO3)2 ion as the reference because it seems likely that this produces more stable cluster ions compared to NO− 3 (HNO3) due to an eﬃcient acid- base stabilization mechanism (1 : 1-ratio between acid and base if the NO− 3 is regarded as a Lewis base). In previous CLOUD studies a similar scheme has been observed for ion clusters involving sulfuric acid and ammonia or dimethylamine (see Kirkby et al., 2011; Almeida et al., 2013; Kürten et al., 2014; Bianchi et al., 2014). This leads to the following equation for the DMA concentration: have chosen the NO− 3 (HNO3)2 ion as the reference because it seems likely that this produces more stable cluster ions compared to NO− 3 (HNO3) due to an eﬃcient acid- base stabilization mechanism (1 : 1-ratio between acid and base if the NO− 3 is regarded as a Lewis base). In previous CLOUD studies a similar scheme has been observed for ion clusters involving sulfuric acid and ammonia or dimethylamine (see Kirkby et al., 2011; Almeida et al., 2013; Kürten et al., 2014; Bianchi et al., 2014). This leads to the following equation for the DMA concentration: have chosen the NO− 3 (HNO3)2 ion as the reference because it seems likely that this produces more stable cluster ions compared to NO− 3 (HNO3) due to an eﬃcient acid- base stabilization mechanism (1 : 1-ratio between acid and base if the NO− 3 is regarded as a Lewis base). In previous CLOUD studies a similar scheme has been observed for ion clusters involving sulfuric acid and ammonia or dimethylamine (see Kirkby et al., AMTD Detection of DMA in the low pptv range using a nitrate CIMS Figure 2 shows high resolution mass spectra for the m/z values corresponding to the NO− 3 (HNO3)0−3(DMA) ions when diﬀerent amounts of DMA were added to the CLOUD chamber (the indicated mixing ratios are discussed further below). Clearly it can be 10 seen that the NO− 3 (DMA) ion never exceeds the background level (Fig. 2a) indicating that this ion is very unstable. For the other signals the intensity increases with increas- ing DMA mixing ratio. M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion AMTD They added diethylamine 25 (CH3CH2NHCH2CH3, DEA) to ions created from ambient air and identiﬁed the clus- ter NO− 3 (HNO3)(DEA) (m/z 198 Th). Additionally, signals at m/z 261 and m/z 334 Th 13266 has been reported only once by Luts et al. (2011). They added diethylamine 25 (CH3CH2NHCH2CH3, DEA) to ions created from ambient air and identiﬁed the clus- ter NO− 3 (HNO3)(DEA) (m/z 198 Th). Additionally, signals at m/z 261 and m/z 334 Th 13266 13266 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Pap Discussion Paper were observed, which were tentatively assigned to the clusters NO− 3 (HNO3)2(DEA) and NO− 3 (HNO3)2(DEA)2. This study is in agreement with the ﬁndings by Luts et al. (2011) regarding the clusters containing one amine molecule; however, we did not observe a cluster containing two DMA molecules. Combining the results of the present study and the one by Luts et al. (2011) indicates that nitrate chemical ionization mass spec- 5 trometry should likely be capable of detecting various other amines and not just DMA or DEA. Fi 2 h hi h l i f h / l di h were observed, which were tentatively assigned to the clusters NO− 3 (HNO3)2(DEA) and NO− 3 (HNO3)2(DEA)2. This study is in agreement with the ﬁndings by Luts et al. (2011) regarding the clusters containing one amine molecule; however, we did not observe a cluster containing two DMA molecules. Combining the results of the present study and the one by Luts et al. (2011) indicates that nitrate chemical ionization mass spec- 5 trometry should likely be capable of detecting various other amines and not just DMA or DEA. Figure 2 shows high resolution mass spectra for the m/z values corresponding to the NO− 3 (HNO3)0−3(DMA) ions when diﬀerent amounts of DMA were added to the CLOUD AMTD , p p q of the diﬀusivity as expected from theory (Crump and Seinfeld, 1981), which would 5 result in a wall loss rate of 2.2 × 10−3 s−1. The fact that the wall loss rate of DMA is consistent with the one found for sulfuric acid implies that the walls act as a perfect sink for DMA which sticks eﬃciently to the available surfaces at these low mixing ratios and rather short exposure time at a temperature of 278 K and a relative humidity of 38 %. p p q of the diﬀusivity as expected from theory (Crump and Seinfeld, 1981), which would 5 result in a wall loss rate of 2.2 × 10−3 s−1. The fact that the wall loss rate of DMA is consistent with the one found for sulfuric acid implies that the walls act as a perfect sink for DMA which sticks eﬃciently to the available surfaces at these low mixing ratios and rather short exposure time at a temperature of 278 K and a relative humidity of 38 %. of the diﬀusivity as expected from theory (Crump and Seinfeld, 1981), which would 5 result in a wall loss rate of 2.2 × 10−3 s−1. The fact that the wall loss rate of DMA is consistent with the one found for sulfuric acid implies that the walls act as a perfect sink for DMA which sticks eﬃciently to the available surfaces at these low mixing ratios and rather short exposure time at a temperature of 278 K and a relative humidity of 38 %. of the diﬀusivity as expected from theory (Crump and Seinfeld, 1981), which would 5 result in a wall loss rate of 2.2 × 10−3 s−1. The fact that the wall loss rate of DMA is consistent with the one found for sulfuric acid implies that the walls act as a perfect sink for DMA which sticks eﬃciently to the available surfaces at these low mixing ratios and rather short exposure time at a temperature of 278 K and a relative humidity of 38 %. Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. AMTD Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Furthermore, this assumption should be justiﬁed by the fact that the calibration lines in 10 Fig. 4 show no steepening when going from low mixing ratios to higher values which is evidence that no wall saturation occurs. However, when measuring higher mixing ratios over a long time it will probably be necessary either to clean the sampling line from time to time or to calibrate with a known amine concentration. The re-evaporation of DMA mentioned above therefore seems to have only a small eﬀect and comes into play only 15 when the ﬂow of DMA into the chamber is shut-oﬀand the DMA concentration reaches very low values. The DMA wall loss rate of 2.8 × 10−3 s−1 for CLOUD10-T is slightly higher due to a diﬀerent conﬁguration of the mixing fans inside the chamber resulting in a diﬀerent thickness of the diﬀusion layer. 20 mentioned above therefore seems to have only a small eﬀect and comes into play only 15 when the ﬂow of DMA into the chamber is shut-oﬀand the DMA concentration reaches very low values. The DMA wall loss rate of 2.8 × 10−3 s−1 for CLOUD10-T is slightly higher due to a diﬀerent conﬁguration of the mixing fans inside the chamber resulting in a diﬀerent thickness of the diﬀusion layer. 20 The DMA wall loss rate of 2.8 × 10−3 s−1 for CLOUD10-T is slightly higher due to a diﬀerent conﬁguration of the mixing fans inside the chamber resulting in a diﬀerent thickness of the diﬀusion layer. 20 thickness of the diﬀusion layer. 20 Using the derived wall loss rate, the DMA mixing ratio can be calculated according to Eq. (3). The error in the targeted mixing ratio during CLOUD7 is calculated based on the 1σ-standard deviations for the parameter τ1 (432.9 ± 47.9 s) from the double- exponential ﬁt which was made with the software IGOR. Furthermore, a 10 % error in y Using the derived wall loss rate, the DMA mixing ratio can be calculated according to Eq. (3). The error in the targeted mixing ratio during CLOUD7 is calculated based on the 1σ-standard deviations for the parameter τ1 (432.9 ± 47.9 s) from the double- exponential ﬁt which was made with the software IGOR. 3.1 DMA wall loss and dilution rate 15 The data shown in Fig. 3 were used to determine the wall loss rate of DMA in the CLOUD chamber experimentally during CLOUD7. The same procedure was repeated for CLOUD10-T. The red line shows the signal for the NO− 3 (HNO3)1(DMA) and the NO− 3 (HNO3)2(DMA) ion normalized by the primary count rate of the NO− 3 (HNO3)2 ion. \| Discussion Paper 3 ( 3)2( ) y p y 3 ( 3)2 Additionally, the setting of the DMA ﬂow into the CLOUD chamber is shown in arbitrary 20 units (shaded area in Fig. 3). When the DMA ﬂow is shut-oﬀa clear decrease in the DMA signal can be seen. This decay is due to two diﬀerent loss processes: (1) wall loss, which is fast, and (2) mainly loss due to dilution, which is a slow process (see Sect. 2.2). However, this latter loss process is inﬂuenced by re-evaporation of DMA from 3 3 Additionally, the setting of the DMA ﬂow into the CLOUD chamber is shown in arbitrary 20 units (shaded area in Fig. 3). When the DMA ﬂow is shut-oﬀa clear decrease in the DMA signal can be seen. This decay is due to two diﬀerent loss processes: (1) wall loss, which is fast, and (2) mainly loss due to dilution, which is a slow process (see Sect. 2.2). However, this latter loss process is inﬂuenced by re-evaporation of DMA from the chamber walls. Fitting the decay with a double-exponential function yields a value 25 for the wall loss rate of 2.3 × 10−3 s−1. Comparison to the wall loss rate of sulfuric 13267 13267 Discussion Paper \| Discussion Paper Discussion Paper \| Discus Discussion Paper acid (2.2 × 10−3 s−1) conﬁrms that the derived wall loss rate for DMA is reasonable. One expects a slightly higher loss rate for DMA because it is the lighter molecule and therefore diﬀuses faster. Using the diﬀusion coeﬃcients and wall loss rates for DMA and sulfuric acid, one can show that the wall loss rate is proportional to the square root acid (2.2 × 10−3 s−1) conﬁrms that the derived wall loss rate for DMA is reasonable. One expects a slightly higher loss rate for DMA because it is the lighter molecule and therefore diﬀuses faster. 3.1 DMA wall loss and dilution rate 15 Using the diﬀusion coeﬃcients and wall loss rates for DMA and sulfuric acid, one can show that the wall loss rate is proportional to the square root AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion AMTD Diﬀerent ﬂow rates of DMA were applied to the chamber during both CLOUD cam- paigns and for certain periods the DMA was completely shut-oﬀ. The periods when the chamber was ﬂushed with clean air for extended times, can be used to determine Diﬀerent ﬂow rates of DMA were applied to the chamber during both CLOUD cam- paigns and for certain periods the DMA was completely shut-oﬀ. The periods when the chamber was ﬂushed with clean air for extended times, can be used to determine the background signal for the NO− 3 (HNO3)1−3(DMA) cluster ions. We believe that this 5 background is caused by electronic noise, since no DMA was detected in the clusters for the nucleation experiments conducted during these periods (Kürten et al., 2014). Since the CI-APi-TOF uses the same clean gas as the CLOUD chamber for the sheath the background signal for the NO− 3 (HNO3)1−3(DMA) cluster ions. We believe that this 5 background is caused by electronic noise, since no DMA was detected in the clusters for the nucleation experiments conducted during these periods (Kürten et al., 2014). Since the CI-APi-TOF uses the same clean gas as the CLOUD chamber for the sheath gas it is also unlikely that there is any source of DMA inside the instrument. the background signal for the NO− 3 (HNO3)1−3(DMA) cluster ions. We believe that this 5 background is caused by electronic noise, since no DMA was detected in the clusters for the nucleation experiments conducted during these periods (Kürten et al., 2014). Since the CI-APi-TOF uses the same clean gas as the CLOUD chamber for the sheath gas it is also unlikely that there is any source of DMA inside the instrument. Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Figure 5 show the time series of the normalized DMA signal (red line) during 10 CLOUD7 and CLOUD10-T together with the calculated DMA mixing ratio (shaded area) according to Eq. (2). It can be seen in Fig. 5a that even at the lowest set point of ∼2.2 pptv DMA the signal is signiﬁcantly elevated compared to background conditions. Further increase of the DMA ﬂow leads to correspondingly higher signals. AMTD Furthermore, a 10 % error in the MFC ﬂow rates is taken into account in the error analysis. 25 er \| Discussion Paper \| the MFC ﬂow rates is taken into account in the error analysis. 25 13268 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Pap Discussion Paper AMTD Figure 5 show the time series of the normalized DMA signal (red line) during 10 CLOUD7 and CLOUD10-T together with the calculated DMA mixing ratio (shaded area) according to Eq. (2). It can be seen in Fig. 5a that even at the lowest set point of ∼2.2 pptv DMA the signal is signiﬁcantly elevated compared to background conditions. Further increase of the DMA ﬂow leads to correspondingly higher signals. p g y g g The data from Fig. 5 and from other periods (not shown) has been averaged over suf- 15 ﬁciently long periods where a constant DMA mixing ratio was applied to the chamber. These averaged normalized signals are shown as a function of the calculated DMA mixing ratio in Fig. 4. A linear ﬁt has been applied to the data from each calibration yielding a correlation coeﬃcient close to 1 (R2 = 0.99). This indicates that the applied The data from Fig. 5 and from other periods (not shown) has been averaged over suf- 15 ﬁciently long periods where a constant DMA mixing ratio was applied to the chamber. These averaged normalized signals are shown as a function of the calculated DMA mixing ratio in Fig. 4. A linear ﬁt has been applied to the data from each calibration yielding a correlation coeﬃcient close to 1 (R2 = 0.99). This indicates that the applied methodology is well suited to quantify DMA at low mixing ratios in the pptv range. 20 methodology is well suited to quantify DMA at low mixing ratios in the pptv range. 20 The slopes of the calibration line from Fig. 4 are a measure of the sensitivity of the nitrate CI-APi-TOF towards DMA. After converting the mixing ratio of DMA into a con- centration (1 pptv corresponds to 2.61 × 107 moleculecm−3 at 278 K and 1 bar), the calibration constant from Eq. (4) can be evaluated as C = 1.48 × 1011 moleculecm−3 gy q y g pp g The slopes of the calibration line from Fig. 4 are a measure of the sensitivity of the nitrate CI-APi-TOF towards DMA. After converting the mixing ratio of DMA into a con- centration (1 pptv corresponds to 2.61 × 107 moleculecm−3 at 278 K and 1 bar), the calibration constant from Eq. (4) can be evaluated as C = 1.48 × 1011 moleculecm−3 11 3 The slopes of the calibration line from Fig. AMTD getic collisions with neutral gas molecules. In order to avoid ambiguity due to changes 5 in the primary ion count rate distribution over time, routinely performed calibration mea- surements are therefore recommended (Freshour et al., 2014). In this case, the method should yield accurate and reproducible results despite the yet unknown details of the ion molecule clustering and declustering processes involved. getic collisions with neutral gas molecules. In order to avoid ambiguity due to changes 5 in the primary ion count rate distribution over time, routinely performed calibration mea- surements are therefore recommended (Freshour et al., 2014). In this case, the method should yield accurate and reproducible results despite the yet unknown details of the ion molecule clustering and declustering processes involved. Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion 3.3 Limit of detection for measurement of DMA 10 10 The data shown in Fig. 5 can be used to determine the limit of detection (LOD) for the DMA measurements with the nitrate CI-APi-TOF. The average background signal together with the 3σ-standard deviation is indicated by the light orange band in Fig. 5a and b. The second step in Fig. 5a (4.4 pptv of DMA) during the CLOUD7 calibration yields an average signal that is outside the orange band. By use of the 3σ-standard 15 deviation the LOD during the CLOUD7 campaign was below 2.6 pptv for a 10 min inte- gration time. Taking into account that wall loss within the CI-APi-TOF sampling line accounts for a factor of 4 reduction in the DMA concentration reaching the instrument, leads to yields an average signal that is outside the orange band. By use of the 3σ-standard 15 deviation the LOD during the CLOUD7 campaign was below 2.6 pptv for a 10 min inte- gration time. Taking into account that wall loss within the CI-APi-TOF sampling line accounts for a factor of 4 reduction in the DMA concentration reaching the instrument, leads to g , a LOD of ∼0.7 pptv for a situation where sampling line losses are negligible. While this 20 cannot be realized in reality, a considerable reduction in the sampling line losses is, however, possible in ﬁeld studies. The CLOUD sampling lines are relatively long (∼1 m in total) and the instrument operated at a low ﬂow rate (∼8.5 Lmin−1). In the ﬁeld there is no restriction on the available amount of air that can be drawn, which will lead to reduced line losses 25 a LOD of ∼0.7 pptv for a situation where sampling line losses are negligible. While this 20 cannot be realized in reality, a considerable reduction in the sampling line losses is, however, possible in ﬁeld studies. The CLOUD sampling lines are relatively long (∼1 m in total) and the instrument operated at a low ﬂow rate (∼8.5 Lmin−1). In the ﬁeld there is no restriction on the available amount of air that can be drawn, which will lead to \| Discussion Paper \| reduced line losses. 25 During the CLOUD10-T calibration (Fig. 5b) the ﬁrst step (5.8 pptv of DMA) is clearly above the LOD. This is supported by the determination of the LOD which was below 1.7 pptv for the CLOUD10-T campaign. AMTD 4 are a measure of the sensitivity of the nitrate CI-APi-TOF towards DMA. After converting the mixing ratio of DMA into a con- centration (1 pptv corresponds to 2.61 × 107 moleculecm−3 at 278 K and 1 bar), the calibration constant from Eq. (4) can be evaluated as C = 1.48 × 1011 moleculecm−3 for CLOUD7 and C = 3.45×1011 molecule cm−3 for CLOUD10-T from the slope of the 25 individual linear ﬁt. Compared to the calibration constant for sulfuric acid this value is about 1 to 1.5 orders of magnitude higher (Kürten et al., 2012; Kürten et al., 2014) and \| Discussion Paper \| for CLOUD7 and C = 3.45×1011 molecule cm−3 for CLOUD10-T from the slope of the 25 individual linear ﬁt. Compared to the calibration constant for sulfuric acid this value is about 1 to 1.5 orders of magnitude higher (Kürten et al., 2012; Kürten et al., 2014) and therefore indicates a lower sensitivity for DMA compared to sulfuric acid. One expla- for CLOUD7 and C = 3.45×1011 molecule cm−3 for CLOUD10-T from the slope of the 25 individual linear ﬁt. Compared to the calibration constant for sulfuric acid this value is about 1 to 1.5 orders of magnitude higher (Kürten et al., 2012; Kürten et al., 2014) and therefore indicates a lower sensitivity for DMA compared to sulfuric acid. One expla- 13269 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper Discussion Paper \| Discussio Discussion Paper nation for this behavior could be that the evaporation rate of the NO− 3 (HNO3)1−3(DMA) clusters is non-negligible. It is still an open question, whether this evaporation occurs inside the ion source, where the temperature is close to the chamber-temperature or within the APi section, where a higher eﬀective temperature is expected due to ener- getic collisions with neutral gas molecules. In order to avoid ambiguity due to changes 5 in the primary ion count rate distribution over time, routinely performed calibration mea- surements are therefore recommended (Freshour et al., 2014). In this case, the method should yield accurate and reproducible results despite the yet unknown details of the ion molecule clustering and declustering processes involved. AMTD nation for this behavior could be that the evaporation rate of the NO− 3 (HNO3)1−3(DMA) clusters is non-negligible. It is still an open question, whether this evaporation occurs inside the ion source, where the temperature is close to the chamber-temperature or within the APi section, where a higher eﬀective temperature is expected due to ener- scussion Paper \| Discussion Paper \| Discussion Paper 3.4 Comparison to previously published DMA mixing ratio One important aspect of the present study is that the DMA mixing ratios are calcu- lated for the CLOUD chamber from the evaluated wall loss rate for DMA and the MFC 5 settings. A comparison to DMA mixing ratios measured directly using an ion chromato- graph (IC, see Praplan et al., 2012) yields generally reasonable agreement for the CLOUD7 data for the time periods when both instruments where operated in parallel with the settings reported here (Fig. 6). The DMA average mixing ratio (averaged for Detection of DMA in the low pptv range using a nitrate CIMS One important aspect of the present study is that the DMA mixing ratios are calcu- lated for the CLOUD chamber from the evaluated wall loss rate for DMA and the MFC 5 settings. A comparison to DMA mixing ratios measured directly using an ion chromato- graph (IC, see Praplan et al., 2012) yields generally reasonable agreement for the CLOUD7 data for the time periods when both instruments where operated in parallel with the settings reported here (Fig. 6). The DMA average mixing ratio (averaged for the time period displayed) was 17 pptv for the IC and 22 pptv for the CI-APi-TOF. During 10 the CLOUD10-T measurements the IC was not available. Note that the displayed DMA mixing ratios from the IC for CLOUD7 have been multiplied with a factor of 1.8 to ac- count for the eﬃciency of the cation trap column. This correction was not considered in previous publications and therefore some of the reported DMA mixing ratios should be scaled up by a factor of 1.8. Some of the high concentration points reported in Almeida 15 et al. (2013) are from CLOUD7 and need to be scaled up, while the low concentration points (≤5pptv) are from CLOUD4 and are correct. The correction of the high con- centration points from CLOUD7 therefore relates to data where nucleation rates were saturated with respect to DMA anyhow, and therefore the change does not aﬀect any of the conclusions or statements. Similarly, Kürten et al. (2014) reported DMA mixing 20 ratios between 5 and 32 pptv during measurements of neutral sulfuric acid-DMA clus- ters. Applying the correction factor these values increase correspondingly. Since the actual DMA mixing ratios were not used in the data analysis by Kürten et al. (2014) the adjustment of the reported DMA values does not imply a signiﬁcant change. M. 3.3 Limit of detection for measurement of DMA 10 Taking the sampling loss of a factor ∼2.5 into reduced line losses. 25 During the CLOUD10-T calibration (Fig. 5b) the ﬁrst step (5.8 pptv of DMA) is clearly above the LOD. This is supported by the determination of the LOD which was below 1.7 pptv for the CLOUD10-T campaign. Taking the sampling loss of a factor ∼2.5 into 13270 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussio account yields an LOD of ∼0.7 pptv, which matches very well with the value obtained for CLOUD7. Discussion Paper account yields an LOD of ∼0.7 pptv, which matches very well with the value obtained for CLOUD7. 3.4 Comparison to previously published DMA mixing ratio Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion with the settings reported here (Fig. 6). The DMA average mixing ratio (averaged for the time period displayed) was 17 pptv for the IC and 22 pptv for the CI-APi-TOF. During 10 the CLOUD10-T measurements the IC was not available. Note that the displayed DMA mixing ratios from the IC for CLOUD7 have been multiplied with a factor of 1.8 to ac- count for the eﬃciency of the cation trap column. This correction was not considered in previous publications and therefore some of the reported DMA mixing ratios should be scaled up by a factor of 1.8. Some of the high concentration points reported in Almeida 15 et al. (2013) are from CLOUD7 and need to be scaled up, while the low concentration points (≤5pptv) are from CLOUD4 and are correct. The correction of the high con- g g g g g the time period displayed) was 17 pptv for the IC and 22 pptv for the CI-APi-TOF. During 10 the CLOUD10-T measurements the IC was not available. Note that the displayed DMA mixing ratios from the IC for CLOUD7 have been multiplied with a factor of 1.8 to ac- count for the eﬃciency of the cation trap column. This correction was not considered in previous publications and therefore some of the reported DMA mixing ratios should be the time period displayed) was 17 pptv for the IC and 22 pptv for the CI-APi-TOF. During 10 the CLOUD10-T measurements the IC was not available. Note that the displayed DMA mixing ratios from the IC for CLOUD7 have been multiplied with a factor of 1.8 to ac- count for the eﬃciency of the cation trap column. This correction was not considered in previous publications and therefore some of the reported DMA mixing ratios should be p p p g scaled up by a factor of 1.8. Some of the high concentration points reported in Almeida 15 et al. (2013) are from CLOUD7 and need to be scaled up, while the low concentration points (≤5pptv) are from CLOUD4 and are correct. 3.4 Comparison to previously published DMA mixing ratio The correction of the high con- centration points from CLOUD7 therefore relates to data where nucleation rates were saturated with respect to DMA anyhow, and therefore the change does not aﬀect any of the conclusions or statements. Similarly, Kürten et al. (2014) reported DMA mixing 20 ratios between 5 and 32 pptv during measurements of neutral sulfuric acid-DMA clus- ters. Applying the correction factor these values increase correspondingly. Since the actual DMA mixing ratios were not used in the data analysis by Kürten et al. (2014) the dj t t f th t d DMA l d t i l i iﬁ t h scaled up by a factor of 1.8. Some of the high concentration points reported in Almeida 15 et al. (2013) are from CLOUD7 and need to be scaled up, while the low concentration points (≤5pptv) are from CLOUD4 and are correct. The correction of the high con- centration points from CLOUD7 therefore relates to data where nucleation rates were saturated with respect to DMA anyhow, and therefore the change does not aﬀect any ussion Paper \| Discussion Paper \| p y g y of the conclusions or statements. Similarly, Kürten et al. (2014) reported DMA mixing 20 ratios between 5 and 32 pptv during measurements of neutral sulfuric acid-DMA clus- ters. Applying the correction factor these values increase correspondingly. Since the actual DMA mixing ratios were not used in the data analysis by Kürten et al. (2014) the adjustment of the reported DMA values does not imply a signiﬁcant change. \| Discussion Paper 13271 Discussion Paper \| Discussion Pap Discussion Paper Detection of DMA in the low pptv range using a nitrate CIMS Detection of DMA in the low pptv range using a nitrate CIMS Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion DMA (or amine measurements in general) the method introduced in this study has the 10 beneﬁt of not being restricted to amine measurements. Nitrate chemical ionization can be used at the same time and with the same instrument for highly-sensitive measure- ments of sulfuric acid and extremely low volatile organic compounds (ELVOCs). These compounds are thought to play an essential role in the formation of new particles. Be- ing capable of measuring DMA together with sulfuric acid and ELVOCs makes nitrate 15 CI an even more versatile tool for studying NPF than previously thought. Future studies will focus on the eﬀect of temperature and RH regarding the sensitivity of nitrate CI towards DMA. Furthermore, the detection of other amines will be tested and the method will be deployed in ﬁeld studies. For this an amine source providing ll d ﬁ d i i di ll h CI APi TOF ld b d i bl ( g p g g CI an even more versatile tool for studying NPF than previously thought. Future studies will focus on the eﬀect of temperature and RH regarding the sensitivity of nitrate CI towards DMA. Furthermore, the detection of other amines will be tested and the method will be deployed in ﬁeld studies. For this an amine source providing well-deﬁned concentrations periodically to the CI-APi-TOF would be desirable (see e.g. 20 Freshour et al., 2014). CI an even more versatile tool for studying NPF than previously thought. Future studies will focus on the eﬀect of temperature and RH regarding the sensitivity of nitrate CI towards DMA. Furthermore, the detection of other amines will be tested and the method will be deployed in ﬁeld studies. For this an amine source providing well-deﬁned concentrations periodically to the CI-APi-TOF would be desirable (see e.g. 20 Freshour et al., 2014). and the method will be deployed in ﬁeld studies. For this an amine source providing well-deﬁned concentrations periodically to the CI-APi-TOF would be desirable (see e.g. 20 Freshour et al., 2014). Acknowledgements. Conclusions 4 AMTD It is demonstrated that dimethylamine (DMA) can be detected at low mixing ratios us- ing nitrate chemical ionization mass spectrometry. DMA is mainly detected in a cluster containing the nitrate ion plus additional nitric acid molecules (NO− 3 (HNO3)1−3(DMA)). Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion 3 3 1 3 Calibration of the CI-APi-TOF used during the CLOUD7 and CLOUD10-T campaign 5 indicates very good linearity in the range between 0 and ∼120 pptv of DMA. The de- tection limit under ideal conditions at the CLOUD chamber was below 1 pptv for an integration time of 10 min at a temperature of 278 K and a relative humidity of 38 %. While there are other techniques yielding similar or even better detection limits for 3 3 1 3 Calibration of the CI-APi-TOF used during the CLOUD7 and CLOUD10-T campaign 5 indicates very good linearity in the range between 0 and ∼120 pptv of DMA. The de- tection limit under ideal conditions at the CLOUD chamber was below 1 pptv for an integration time of 10 min at a temperature of 278 K and a relative humidity of 38 %. While there are other techniques yielding similar or even better detection limits for AMTD C., Franchin, A., Guida, R., Hakala, J., Hansel, A., Heinritzi, M., Henschel, H., Jokinen, T., Junninen, H., 5 Kajos, M., Kangasluoma, J., Keskinen, H., Kupc, A., Kurtén, T., Kvashin, A. N., Laakso- nen, A., Lehtipalo, K., Leiminger, M., Leppä, J., Loukonen, V., Makhmutov, V., Mathot, S., McGrath, M. J., Nieminen, T., Olenius, T., Onnela, A., Petäjä, T., Riccobono, F., Riip- inen, I., Rissanen, M., Rondo, L., Ruuskanen, T., Santos, F. D., Sarnela, N., Schallhart, S., S h it h f R S i f ld J H Si M Si ilä M St hk Y St t F T é A 5 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. , , , , , , , , , , , , , , Schnitzhofer, R., Seinfeld, J. H., Simon, M., Sipilä, M., Stozhkov, Y., Stratmann, F., Tomé, A., 10 Tröstl, J., Tsagkogeorgas, G., Vaattovaara, P., Viisanen, Y., Virtanen, A., Vrtala, A., Wag- ner, P. E., Weingartner, E., Wex, H., Williamson, C., Wimmer, D., Ye, P., Yli-Juuti, T., Carslaw, K. S., Kulmala, M., Curtius, J., Baltensperger, U., Worsnop, D. R., Vehkamäki, H., and Kirkby, J.: Molecular understanding of sulphuric acid-amine particle nucleation in the atmosphere, Nature, 502, 359–363, 2013. 15 10 10 Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion 15 Berndt, T., Stratmann, F., Sipilä, M., Vanhanen, J., Petäjä, T., Mikkilä, J., Grüner, A., Spindler, G., Lee Mauldin III, R., Curtius, J., Kulmala, M., and Heintzenberg, J.: Laboratory study on new particle formation from the reaction OH + SO2: inﬂuence of experimental conditions, H2O vapour, NH3 and the amine tert-butylamine on the overall process, Atmos. Chem. Phys., 10, 7101–7116, doi:10.5194/acp-10-7101-2010, 2010. 0 20 Berndt, T., Sipilä, M., Stratmann, F., Petäjä, T., Vanhanen, J., Mikkilä, J., Patokoski, J., Taipale, R., Mauldin III, R. L., and Kulmala, M.: Enhancement of atmospheric H2SO4/H2O nucleation: organic oxidation products versus amines, Atmos. Chem. Phys., 14, 751–764, doi:10.5194/acp-14-751-2014, 2014. \| Discussion Paper \| Bianchi, F., Praplan, A. P., Sarnela, N., Dommen, J., Kürten, A., Ortega, I. K., Schobes- 25 berger, S., Junninen, H., Simon, M., Tröstl, J., Jokinen, T., Sipilä, M., Adamov, A., Amorim, A., Almeida, J., Breitenlechner, M., Duplissy, J., Ehrhart, S., Flagan, R. AMTD Almeida, J., Schobesberger, S., Kürten, A., Ortega, I. K., Kupiainen-Määttä, O., Praplan, A. P., Adamov, A., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Don- ahue, N. M., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Almeida, J., Schobesberger, S., Kürten, A., Ortega, I. K., Kupiainen-Määttä, O., Praplan, A. P., Adamov, A., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Don- ahue, N. M., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Guida, R., Hakala, J., Hansel, A., Heinritzi, M., Henschel, H., Jokinen, T., Junninen, H., 5 Kajos, M., Kangasluoma, J., Keskinen, H., Kupc, A., Kurtén, T., Kvashin, A. N., Laakso- nen, A., Lehtipalo, K., Leiminger, M., Leppä, J., Loukonen, V., Makhmutov, V., Mathot, S., McGrath, M. J., Nieminen, T., Olenius, T., Onnela, A., Petäjä, T., Riccobono, F., Riip- inen, I., Rissanen, M., Rondo, L., Ruuskanen, T., Santos, F. D., Sarnela, N., Schallhart, S., Schnitzhofer R Seinfeld J H Simon M Sipilä M Stozhkov Y Stratmann F Tomé A 10 8, 13257–13284, 2015 Adamov, A., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Don ahue, N. M., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Guida, R., Hakala, J., Hansel, A., Heinritzi, M., Henschel, H., Jokinen, T., Junninen, H., 5 Kajos, M., Kangasluoma, J., Keskinen, H., Kupc, A., Kurtén, T., Kvashin, A. N., Laakso- nen, A., Lehtipalo, K., Leiminger, M., Leppä, J., Loukonen, V., Makhmutov, V., Mathot, S., McGrath, M. J., Nieminen, T., Olenius, T., Onnela, A., Petäjä, T., Riccobono, F., Riip- inen, I., Rissanen, M., Rondo, L., Ruuskanen, T., Santos, F. D., Sarnela, N., Schallhart, S., Schnitzhofer, R., Seinfeld, J. H., Simon, M., Sipilä, M., Stozhkov, Y., Stratmann, F., Tomé, A., 10 Tröstl, J., Tsagkogeorgas, G., Vaattovaara, P., Viisanen, Y., Virtanen, A., Vrtala, A., Wag- ner, P. E., Weingartner, E., Wex, H., Williamson, C., Wimmer, D., Ye, P., Yli-Juuti, T., Carslaw, K. S., Kulmala, M., Curtius, J., Baltensperger, U., Worsnop, D. R., Vehkamäki, H., and Kirkby, J.: Molecular understanding of sulphuric acid-amine particle nucleation in the atmosphere, Nature, 502, 359–363, 2013. 15 ahue, N. M., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. Detection of DMA in the low pptv range using a nitrate CIMS We thank CERN for supporting CLOUD with important technical and ﬁ- nancial resources and provision of a particle beam from the proton synchrotron. This research received funding from the EC Seventh Framework Programme (Marie Curie Initial Training Net- Acknowledgements. We thank CERN for supporting CLOUD with important technical and ﬁ- nancial resources and provision of a particle beam from the proton synchrotron. This research received funding from the EC Seventh Framework Programme (Marie Curie Initial Training Net- Acknowledgements. We thank CERN for supporting CLOUD with important technical and ﬁ- nancial resources and provision of a particle beam from the proton synchrotron. This research received funding from the EC Seventh Framework Programme (Marie Curie Initial Training Net- work MC-ITN “CLOUD-TRAIN” no. 316662), the German Federal Ministry of Education and 25 Research (project no. 01LK1222A) as well as the Swiss National Science Foundation (project no. 200020_152907). We thank the tofTools team for providing tools for mass spectrometry analysis Acknowledgements. We thank CERN for supporting CLOUD with important technical and ﬁ- nancial resources and provision of a particle beam from the proton synchrotron. This research received funding from the EC Seventh Framework Programme (Marie Curie Initial Training Net- work MC-ITN “CLOUD-TRAIN” no. 316662), the German Federal Ministry of Education and 25 Discussion Paper g g ( g work MC-ITN “CLOUD-TRAIN” no. 316662), the German Federal Ministry of Education and 25 Research (project no. 01LK1222A) as well as the Swiss National Science Foundation (project no. 200020_152907). We thank the tofTools team for providing tools for mass spectrometry analysis. work MC-ITN “CLOUD-TRAIN” no. 316662), the German Federal Ministry of Education and 25 Research (project no. 01LK1222A) as well as the Swiss National Science Foundation (project no. 200020_152907). We thank the tofTools team for providing tools for mass spectrometry analysis. 13272 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Pa Discussion Paper References Almeida, J., Schobesberger, S., Kürten, A., Ortega, I. K., Kupiainen-Määttä, O., Praplan, A. P., Adamov, A., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Don- ahue, N. M., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Guida R Hakala J Hansel A Heinritzi M Henschel H Jokinen T Junninen H AMTD C., Franchin, A., Hakala, J., Hansel, A., Heinritzi, M., Kangasluoma, J., Keskinen, H., Kim, J., Kirkby, J., Laak- sonen, A., Lawler, M. J., Lehtipalo, K., Leiminger, M., Makhmutov, V., Mathot, S., Onnela, A., Petäjä, T., Riccobono, F., Rissanen, M. P., Rondo, L., Tomé, A., Virtanen, A., Viisanen, Y., 30 Williamson, C., Wimmer, D., Winkler, P. M., Ye, P., Curtius, J., Kulmala, M., Worsnop, D. R., Donahue, N. M., and Baltensperger, U.: Insight into acid–base nucleation experiments by 30 13273 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discu Discussion Paper comparison of the chemical composition of positive, negative, and neutral clusters, Environ. Sci. Technol., 48, 13675–13684, doi:10.1021/es502380b, 2014. comparison of the chemical composition of positive, negative, and neutral clusters, Enviro Sci. Technol., 48, 13675–13684, doi:10.1021/es502380b, 2014. Detection of DMA in the low pptv range using a nitrate CIMS p Creamean, J. M., Ault, A. P., Ten Hoeve, J. E., Jacobson, M. Z., Roberts, G. C., and Prather, K. A.: Measurements of aerosol chemistry during new particle formation events at a remote rural mountain site, Environ. Sci. Technol., 45, 8208–8216, doi:10.1021/es103692f, 2011. 10 10 M. Simon et al. Crump, J. G. and Seinfeld, J. H.: Turbulent deposition and gravitational sedimentation of an aerosol in a vessel of arbitrary shape, J. Aerosol Sci., 12, 405–415, 1981. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Ehn, M., Thornton, J. A., Kleist, E., Sipilä, M., Junninen, H., Pullinen, I., Springer, M., Rubach, F., Tillmann, R., Lee, B., Lopez-Hilﬁker, F., Andres, S., Acir, I.-H., Rissanen, M., Jokinen, T., Schobesberger, S., Kangasluoma, J., Kontkanen, J., Nieminen, T., Kurtén, T., 15 Nielsen, L. B., Jørgensen, S., Kjaergaard, H. G., Canagaratna, M., Dal Maso, M., Berndt, T., Petäjä, T., Wahner, A., Kerminen, V.-M., Kulmala, M., Worsnop, D. R., Wildt, J., and Mentel, T. F.: A large source of low–volatility secondary organic aerosol, Nature, 506, 476– 479, 2014. 15 Eisele, F. L. and Tanner, D. J.: Measurement of the gas phase concentration of H2SO4 and 20 methane sulfonic acid and estimates of H2SO4 production and loss in the atmosphere, J. Geophys. Res., 98, 9001–9010, 1993. Erupe M E Viggiano A A and Lee S H : The eﬀect of trimethylamine on atmospheric Eisele, F. L. and Tanner, D. J.: Measurement of the gas phase concentration of H2SO4 and 20 methane sulfonic acid and estimates of H2SO4 production and loss in the atmosphere, J. Geophys. Res., 98, 9001–9010, 1993. Erupe, M. E., Viggiano, A. A., and Lee, S.-H.: The eﬀect of trimethylamine on atmospheric nucleation involving H2SO4, Atmos. Chem. Phys., 11, 4767–4775, doi:10.5194/acp-11-4767- 2011 2011 p y Erupe, M. E., Viggiano, A. A., and Lee, S.-H.: The eﬀect of trimethylamine on atmospheric nucleation involving H2SO4, Atmos. Chem. Phys., 11, 4767–4775, doi:10.5194/acp-11-4767- 2011, 2011. 5 25 \| Discussion Paper \| 2011, 2011. 25 Freshour, N. A., Carlson, K. K., Melka, Y. A., Hinz, S., Panta, B., and Hanson, D. R.: Amine permeation sources characterized with acid neutralization and sensitivities of an amine mass spectrometer, Atmos. Meas. Tech., 7, 3611–3621, doi:10.5194/amt-7-3611-2014, 2014. spectrometer, Atmos. Meas. Tech., 7, 3611 3621, doi:10.5194/amt 7 3611 2014, 2014. Ge, X. L., Wexler, A. AMTD Chen, M., Titcombe, M., Jiang, J., Jen, C., Kuang, C., Fischer, M. L., Eisele, F. L., Siep- mann, J. I., Hanson, D. R., Zhao, J., and McMurry, P. H.: Acid–base chemical reaction model for nucleation rates in the polluted atmospheric boundary layer, P. Natl. Acad. Sci. USA, 109, 5 18713–18718, doi:/10.1073/pnas.1210285109, 2012. for nucleation rates in the polluted atmospheric boundary layer, P. Natl. Acad. Sci. USA, 109, 5 18713–18718, doi:/10.1073/pnas.1210285109, 2012. Creamean, J. M., Ault, A. P., Ten Hoeve, J. E., Jacobson, M. Z., Roberts, G. C., and Prather, K. A.: Measurements of aerosol chemistry during new particle formation events at a remote rural mountain site, Environ. Sci. Technol., 45, 8208–8216, doi:10.1021/es103692f, 2011. 0 5 Detection of DMA in the low pptv range using a nitrate CIMS Jen, C. N., McMurry, P. H., and Hanson, D. R.: Stabilization of sulfuric acid dimers by ammo- nia, methylamine, dimethylamine, and trimethylamine, J. Geophys. Res.-Atmos., 119, 7502– 7514, doi:10.1002/2014JD021592, 2014. 10 M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Jokinen, T., Sipilä, M., Junninen, H., Ehn, M., Lönn, G., Hakala, J., Petäjä, T., Mauldin III, R. L., Kulmala, M., and Worsnop, D. R.: Atmospheric sulphuric acid and neutral cluster measure- ments using CI-APi-TOF, Atmos. Chem. Phys., 12, 4117–4125, doi:10.5194/acp-12-4117- 2012, 2012. 5 15 Junninen, H., Ehn, M., Petäjä, T., Luosujärvi, L., Kotiaho, T., Kostiainen, R., Rohner, U., Go- nin, M., Fuhrer, K., Kulmala, M., and Worsnop, D. R.: A high-resolution mass spectrometer to measure atmospheric ion composition, Atmos. Meas. Tech., 3, 1039–1053, doi:10.5194/amt- 3-1039-2010, 2010. , Karlsson, M. N. A. and Martinsson, B. G.: Methods to measure and predict the transfer function 20 size dependence of individual DMAs, J. Aerosol Sci., 34, 603–625, 2003. Kirkby, J., Curtius, J., Almeida, J., Dunne, E., Duplissy, J., Ehrhart, S., Franchin, A., Gagné, S., Ickes, L., Kürten, A., Kupc, A., Metzger, A., Riccobono, F., Rondo, L., Schobesberger, S., Tsagkogeorgas, G., Wimmer, D., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dom- Karlsson, M. N. A. and Martinsson, B. G.: Methods to measure and predict the transfer function 20 size dependence of individual DMAs, J. Aerosol Sci., 34, 603–625, 2003. Karlsson, M. N. A. and Martinsson, B. G.: Methods to measure and predict the transfer function 20 size dependence of individual DMAs, J. Aerosol Sci., 34, 603–625, 2003. Kirkby, J., Curtius, J., Almeida, J., Dunne, E., Duplissy, J., Ehrhart, S., Franchin, A., Gagné, S., p , , , , Kirkby, J., Curtius, J., Almeida, J., Dunne, E., Duplissy, J., Ehrhart, S., Franchin, A., Gagné, S., Ickes, L., Kürten, A., Kupc, A., Metzger, A., Riccobono, F., Rondo, L., Schobesberger, S., Tsagkogeorgas, G., Wimmer, D., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dom- men, J., Downard, A., Ehn, M., Flagan, R. C., Haider, S., Hansel, A., Hauser, D., Jud, W., 25 Junninen, H., Kreissl, F., Kvashin, A., Laaksonen, A., Lehtipalo, K., Lima, J., Lovejoy, E. R., Makhmutov, V., Mathot, S., Mikkilä, J., Minginette, P., Mogo, S., Nieminen, T., Onnela, A., Pereira, P., Petäjä, T., Schnitzhofer, R., Seinfeld, J. Detection of DMA in the low pptv range using a nitrate CIMS S., and Clegg, S. L.: Atmospheric amines – Part I, a review, Atmos. Environ 45 524 546 doi:10 1016/j atmosenv 2010 10 012 2011a Ge, X. L., Wexler, A. S., and Clegg, S. L.: Atmospheric amines – Part I, a review, Atmos. Environ., 45, 524–546, doi:10.1016/j.atmosenv.2010.10.012, 2011a. 30 30 , , , j , Ge, X. L., Wexler, A. S., and Clegg, S. L.: Atmospheric amines – Part II, ther- modynamic properties and gas/particle partitioning, Atmos. Environ., 45, 561–577, doi:10.1016/j.atmosenv.2010.10.013, 2011b. 13274 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper Discussion Paper \| Discus Discussion Paper Glasoe, W. A., Volz, K., Panta, B., Freshour, N., Bachman, R., Hanson, D. R., McMurry, P. H., and Jen, C.: Sulfuric acid nucleation: an experimental study of the eﬀect of seven bases, J. Geophys. Res.-Atmos., 120, 1933–1950, doi:10.1002/2014JD022730, 2015. AMTD Hanson, D. R. and Eisele, F.: Diﬀusion of H2SO4 in humidiﬁed nitrogen: hydrated H2SO4, J. Phys. Chem. A, 104, 1715–1719, 2000. 5 y Hanson, D. R., McMurry, P. H., Jiang, J., Tanner, D., and Huey, L. G.: Ambient pressure proton transfer mass spectrometry: detection of amines and ammonia, Environ. Sci. Technol., 45, 8881–8888, doi:10.1021/es201819a, 2011. Detection of DMA in the low pptv range using a nitrate CIMS H., Sipilä, M., Stozhkov, Y., Stratmann, F., Tomé, A., Vanhanen, J., Viisanen, Y., Vrtala, A., Wagner, P. E., Walther, H., Weingartner, E., Wex, H., Winkler, P. M., Carslaw, K. S., Worsnop, D. R., Baltensperger, U., and Kulmala, M.: 30 Role of sulphuric acid, ammonia and galactic cosmic rays in atmospheric aerosol nucleation, Nature, 476, 429–435, 2011. Kirkby, J., Curtius, J., Almeida, J., Dunne, E., Duplissy, J., Ehrhart, S., Franchin, A., Gagné, S., Ickes, L., Kürten, A., Kupc, A., Metzger, A., Riccobono, F., Rondo, L., Schobesberger, S., Tsagkogeorgas, G., Wimmer, D., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dom- aper \| Discussion Paper \| g g g , , , , , , , , , , , , men, J., Downard, A., Ehn, M., Flagan, R. C., Haider, S., Hansel, A., Hauser, D., Jud, W., 25 Junninen, H., Kreissl, F., Kvashin, A., Laaksonen, A., Lehtipalo, K., Lima, J., Lovejoy, E. R., Makhmutov, V., Mathot, S., Mikkilä, J., Minginette, P., Mogo, S., Nieminen, T., Onnela, A., Pereira, P., Petäjä, T., Schnitzhofer, R., Seinfeld, J. H., Sipilä, M., Stozhkov, Y., Stratmann, F., Tomé, A., Vanhanen, J., Viisanen, Y., Vrtala, A., Wagner, P. E., Walther, H., Weingartner, E., g g g , , , , , , , , , , , , men, J., Downard, A., Ehn, M., Flagan, R. C., Haider, S., Hansel, A., Hauser, D., Jud, W., 25 Junninen, H., Kreissl, F., Kvashin, A., Laaksonen, A., Lehtipalo, K., Lima, J., Lovejoy, E. R., Makhmutov, V., Mathot, S., Mikkilä, J., Minginette, P., Mogo, S., Nieminen, T., Onnela, A., Pereira, P., Petäjä, T., Schnitzhofer, R., Seinfeld, J. H., Sipilä, M., Stozhkov, Y., Stratmann, F., Tomé, A., Vanhanen, J., Viisanen, Y., Vrtala, A., Wagner, P. E., Walther, H., Weingartner, E., Wex, H., Winkler, P. M., Carslaw, K. S., Worsnop, D. R., Baltensperger, U., and Kulmala, M.: 30 Role of sulphuric acid, ammonia and galactic cosmic rays in atmospheric aerosol nucleation, Nature 476 429 435 2011 25 \| Discussion Paper \| 30 13275 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Detection of DMA in the low pptv range using a nitrate CIMS Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussio Discussion Paper Kulmala, M., Vehkamäki, H., Petäjä, T., Dal Maso, M., Lauri, A., Kerminen, V.-M., Birmili, W., and McMurry, P. H.: Formation and growth rates of ultraﬁne atmospheric particles: a review of observations, J. Aerosol Sci., 35, 143–176, 2004. AMTD Kulmala, M., Kontkanen, J., Junninen, H., Lehtipalo, K., Manninen, H. E., Nieminen, T., Petäjä, T., Sipilä, M., Schobesberger, S., Rantala, P., Franchin, A., Jokinen, T., Järvi- 5 nen, E., Äijälä, M., Kangasluoma, J., Hakala, J., Aalto, P. P., Paasonen, P., Mikkilä, J., Van- hanen, J., Aalto, J., Hakola, H., Makkonen, U., Ruuskanen, T., Mauldin III, R. L., Duplissy, J., Vehkamäki, H., Bäck, J., Kortelainen, A., Riipinen, I., Kurtén, T., Johnston, M. V., Smith, J. N., Ehn, M., Mentel, T. F., Lehtinen, K. E. J., Laaksonen, A., Kerminen, V.-M., Worsnop, D. R.: Direct observations of atmospheric aerosol nucleation, Science, 339, 943–946, 2013. 10 p p Sipilä, M., Sarnela, N., Jokinen, T., Junninen, H., Hakala, J., Rissanen, M. P., Praplan, A., Simon, M., Kürten, A., Bianchi, F., Dommen, J., Curtius, J., Petäjä, T., and Worsnop, D. R.: Bisulfate – cluster based atmospheric pressure chemical ionization mass spectrometer for high-sensitivity (< 100 ppqV) detection of atmospheric dimethyl amine: proof-of-concept and ﬁrst ambient data from boreal forest, Atmos. Meas. Tech., 8, 4001–4011, doi:10.5194/amt-8- 4001-2015, 2015. Detection of DMA in the low pptv range using a nitrate CIMS 10 M. Simon et al. Kupc, A., Amorim, A., Curtius, J., Danielczok, A., Duplissy, J., Ehrhart, S., Walther, H., Ickes, L., Kirkby, J., Kurten, A., Lima, J. M., Mathot, S., Minginette, P., Onnela, A., Rondo, L., and Wagner, P. E.: A ﬁbre-optic UV system for H2SO4 production in aerosol chambers causing minimal thermal eﬀects, J. Aerosol Sci., 42, 532–543, doi:10.1016/j.jaerosci.2011.05.001, 2011. 15 M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion i P \| Di i P \| Di i P 15 Kürten, A., Rondo, L., Ehrhart, S., and Curtius, J.: Performance of a corona ion source for measurement of sulfuric acid by chemical ionization mass spectrometry, Atmos. Meas. Tech., 4, 437–443, doi:10.5194/amt-4-437-2011, 2011. Kürten, A., Rondo, L., Ehrhart, S., and Curtius, J.: Calibration of a chemical ionization mass spectrometer for the measurement of gaseous sulfuric acid, J. Phys. Chem. A, 116, 6375– 20 6386, doi:10.1021/jp212123n, 2012. 20 jp Kürten, A., Jokinen, T., Simon, M., Sipilä, M., Sarnela, N., Junninen, H., Adamov, A., Almeida, J., Amorim, A., Bianchi, F., Breitenlechner, M., Dommen, J., Donahue, N. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Hakala, J., Hansel, A., Heinritzi, M., Hutterli, M., Kangasluoma, J., Kirkby, J., Laaksonen, A., Lehtipalo, K., Leiminger, M., 25 Makhmutov, V., Mathot, S., Onnela, A., Petäjä, T., Praplan, A. P., Riccobono, F., Rissa- nen, M. P., Rondo, L., Schobesberger, S., Seinfeld, J. H., Steiner, G., Tomé, A., Tröstl, J., Winkler, P. M., Williamson, C., Wimmer, D., Ye, P., Baltensperger, U., Carslaw, K. S., Kul- mala, M., Worsnop, D. R., and Curtius, J.: Neutral molecular cluster formation of sulfuric acid-dimethylamine observed in real-time under atmospheric conditions, P. Natl. Acad. Sci. 30 USA, 111, 15019–15024, doi:/10.1073/pnas.1404853111, 2014. 25 \| Discussion Paper \| 30 13276 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Pape Discussion Paper Kurtén, T., Loukonen, V., Vehkamäki, H., and Kulmala, M.: Amines are likely to enhance neu- tral and ion-induced sulfuric acid-water nucleation in the atmosphere more eﬀectively than ammonia, Atmos. Chem. Phys., 8, 4095–4103, doi:10.5194/acp-8-4095-2008, 2008. AMTD Makela, J. M., Yli-Koivisto, S., Hiltunen, V., Seidl, W., Swietlicki, E., Teinila, K., Sillanpaa, M., Koponen, I. K., Paatero, J., Rosman, K., and Hameri, K.: Chemical composition of aerosol during particle formation events in boreal forest, Tellus B, 53, 380–393, doi:10.1034/j.1600- 0889.2001.530405.x, 2001. 5 Detection of DMA in the low pptv range using a nitrate CIMS Luts, A., Parts, T.-E., Hõrrak, U., Junninen, H., and Kulmala, M.: Composition of negative air ions as a function of ion age and selected trace gases: mass- and mobility distribution, J. Aerosol Sci., 42, 820–838, 2011. 10 M. Simon et al. Ouyang, H., He, S., Larriba-Andaluz, C., and Hogan Jr., C. J.: IMS–MS and IMS–IMS inves- tigation of the structure and stability of dimethylamine–sulfuric acid nanoclusters, J. Phys. Chem. A, 119, 2026–2036, doi:10.1021/jp512645g, 2015. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Praplan, A. P., Bianchi, F., Dommen, J., and Baltensperger, U.: Dimethylamine and ammonia measurements with ion chromatography during the CLOUD4 campaign, Atmos. Meas. Tech., 15 5, 2161–2167, doi:10.5194/amt-5-2161-2012, 2012. p , , , , , , p g , y measurements with ion chromatography during the CLOUD4 campaign, Atmos. Meas. Tech., 15 5, 2161–2167, doi:10.5194/amt-5-2161-2012, 2012. Riccobono, F., Schobesberger, S., Scott, C. E., Dommen, J., Ortega, I. K., Rondo, L., Almeida, J., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Hansel, A., Junninen, H., Kajos, M., Keskinen, H., Kupc, A., Kürten, A., Kvashin, A. N., Laaksonen, A., Lehtipalo, K., 20 M kh t V M th t S Ni i T O l A P täjä T P l A P S t F D 15 measurements with ion chromatography during the CLOUD4 campaign, Atmos. Meas. Tech., 15 5, 2161–2167, doi:10.5194/amt-5-2161-2012, 2012. Riccobono, F., Schobesberger, S., Scott, C. E., Dommen, J., Ortega, I. K., Rondo, L., Almeida, J., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Hansel, A., Junninen, H., , , , Riccobono, F., Schobesberger, S., Scott, C. E., Dommen, J., Ortega, I. K., Rondo, L., Almeida, J., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Hansel, A., Junninen, H., Riccobono, F., Schobesberger, S., Scott, C. E., Dommen, J., Ortega, I. K., Rondo, L., Almeida, J., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Downard, A., Dunne, E. M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Hansel, A., Junninen, H., Kajos, M., Keskinen, H., Kupc, A., Kürten, A., Kvashin, A. Detection of DMA in the low pptv range using a nitrate CIMS N., Laaksonen, A., Lehtipalo, K., 20 Makhmutov, V., Mathot, S., Nieminen, T., Onnela, A., Petäjä, T., Praplan, A. P., Santos, F. D., Schallhart, S., Seinfeld, J. H., Sipilä, M., Spracklen, D. V., Stozhkov, Y., Stratmann, F., Tomé, A., Tsagkogeorgas, G., Vaattovaara, P., Viisanen, Y., Vrtala, A., Wagner, P. E., Wein- gartner, E., Wex, H., Wimmer, D., Carslaw, K. S., Curtius, J., Donahue, N. M., Kirkby, J., Kulmala, M., Worsnop, D. R., and Baltensperger, U.: Oxidation products of biogenic emis- 25 sions contribute to nucleation of atmospheric particles, Science, 344, 717–721, 2014. Kajos, M., Keskinen, H., Kupc, A., Kürten, A., Kvashin, A. N., Laaksonen, A., Lehtipalo, K., 20 Makhmutov, V., Mathot, S., Nieminen, T., Onnela, A., Petäjä, T., Praplan, A. P., Santos, F. D., Schallhart, S., Seinfeld, J. H., Sipilä, M., Spracklen, D. V., Stozhkov, Y., Stratmann, F., Tomé, A., Tsagkogeorgas, G., Vaattovaara, P., Viisanen, Y., Vrtala, A., Wagner, P. E., Wein- gartner, E., Wex, H., Wimmer, D., Carslaw, K. S., Curtius, J., Donahue, N. M., Kirkby, J., Kulmala M Worsnop D R and Baltensperger U : Oxidation products of biogenic emis- 25 20 25 Discussion Paper Sipilä, M., Sarnela, N., Jokinen, T., Junninen, H., Hakala, J., Rissanen, M. P., Praplan, A., Simon, M., Kürten, A., Bianchi, F., Dommen, J., Curtius, J., Petäjä, T., and Worsnop, D. R.: Bisulfate – cluster based atmospheric pressure chemical ionization mass spectrometer for high-sensitivity (< 100 ppqV) detection of atmospheric dimethyl amine: proof-of-concept and ﬁrst ambient data from boreal forest, Atmos. Meas. Tech., 8, 4001–4011, doi:10.5194/amt-8- 4001-2015, 2015. 30 13277 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discu Discussion Paper Smith, J. N., Barsanti, K. C., Friedli, H. R., Ehn, M., Kulmala, M., Collins, D. R., Scheck- man, J. H., Williams, B. J., and McMurry, P. H.: Observations of aminium salts in atmospheric nanoparticles and possible climatic implications, P. Natl. Acad. Sci. USA, 107, 6634–6639, doi:10.1073/pnas.0912127107, 2010. AMTD ussion Paper \| Discussion Paper \| Discussion Paper Viggiano, A. A., Seeley, J. V., Mundis, P. L., Williamson, J. S., and Morris, R. A.: Rate Con- 5 stants for the reactions of XO− 3 (H2O)n (X = C, HC, and N) and NO− 3 (HNO3)n with H2SO4: implications for atmospheric detection of H2SO4, J. Phys. Chem. A, 101, 8275–8278, 1997. Viggiano, A. A., Seeley, J. V., Mundis, P. L., Williamson, J. S., and Morris, R. A.: Rate Con- 5 stants for the reactions of XO− 3 (H2O)n (X = C, HC, and N) and NO− 3 (HNO3)n with H2SO4: implications for atmospheric detection of H2SO4, J. Phys. Chem. A, 101, 8275–8278, 1997. Voigtländer, J., Duplissy, J., Rondo, L., Kürten, A., and Stratmann, F.: Numerical simulations of mixing conditions and aerosol dynamics in the CERN CLOUD chamber, Atmos. Chem. Phys., 12, 2205–2214, doi:10.5194/acp-12-2205-2012, 2012. 10 Yu, H. and Lee, S. H.: Chemical ionisation mass spectrometry for the measurement of atmo- implications for atmospheric detection of H2SO4, J. Phys. Chem. A, 101, 8275–8278, 1997. Voigtländer, J., Duplissy, J., Rondo, L., Kürten, A., and Stratmann, F.: Numerical simulations of mixing conditions and aerosol dynamics in the CERN CLOUD chamber, Atmos. Chem. Phys., 12, 2205–2214, doi:10.5194/acp-12-2205-2012, 2012. 10 Voigtländer, J., Duplissy, J., Rondo, L., Kürten, A., and Stratmann, F.: Numerical simulations of mixing conditions and aerosol dynamics in the CERN CLOUD chamber, Atmos. Chem. Phys., 12, 2205–2214, doi:10.5194/acp-12-2205-2012, 2012. 10 Yu, H. and Lee, S. H.: Chemical ionisation mass spectrometry for the measurement of atmo- spheric amines, Environ. Chem., 9, 190–201, doi:10.1071/EN12020, 2012. 10 M. Simon et al. M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion cussion Paper \| Discussion Paper \| Discussion Paper \| 13278 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper AMTD 8, 13257–13284, 2015 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion on Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| Figure 1. CLOUD chamber and gas system for delivering DMA to the chamber. Three mass ﬂow controllers (MFC1 to MFC3) and several valves are used to control the ﬂow rates. The ﬁgure indicates a setting where the bypass and the purge valve are closed while the other valves are open. M. Simon et al. Figure 1. CLOUD chamber and gas system for delivering DMA to the chamber. Three mass ﬂow controllers (MFC1 to MFC3) and several valves are used to control the ﬂow rates. The ﬁgure indicates a setting where the bypass and the purge valve are closed while the other valves are open. \| Discussion Paper \| 13279 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper \| Figure 2. High resolution mass spectra for narrow ranges of m/z values corresponding to NO− 3 (HNO3)0−3(DMA) ions. The colors indicate diﬀerent mixing ratios that were established in the CLOUD chamber during calibration measurements and the dashed vertical lines show the exact mass of the cluster ions. AMTD 8, 13257–13284, 2015 M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion iscussion Paper \| Discussion Paper \| Discussion Paper M. Simon et al. er \| Discussion Paper \| Figure 2. High resolution mass spectra for narrow ranges of m/z values corresponding to NO− 3 (HNO3)0−3(DMA) ions. The colors indicate diﬀerent mixing ratios that were established in the CLOUD chamber during calibration measurements and the dashed vertical lines show the exact mass of the cluster ions. 13280 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper AMTD 8, 13257–13284, 2015 30x10 - 25 20 15 10 5 0 DMA fow into chamber (a.u.) 05:00 18.10.2012 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 Time [UTC] 5 6 7 8 9 0.001 2 3 4 5 6 7 8 9 0.01 DMA signal [cps normalized by NO3 -(HNO3)2] Signal (10 min average) Signal (30 sec average) Double exp. ft (τ1 = 432s, τ2 = 8640s) DMA fow Figure 3. Decay of the normalized cluster ion signal indicating the DMA concentration with 10 min time resolution (red line) and 30 s time resolution (light red line). The DMA ﬂow (grey line and area) into the chamber is shut-oﬀat ∼06:32 UTC and turned on again (at a lower setpoint) around 12:54 UTC. Using a double-exponential ﬁt (black line) the decaying signal can be well represented. The ﬁrst inverse decay constant represents the wall loss rate (1/τ1 = 2.3×10−3 s−1), while the second decay (1/τ2 = 1.2×10−4 s−1) represents a source term due to slow re-evaporation of DMA from the chamber walls superimposed by dilution. 30x10 - 25 20 15 10 5 0 DMA fow into chamber (a.u.) 05:00 18.10.2012 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 Time [UTC] 5 6 7 8 9 0.001 2 3 4 5 6 7 8 9 0.01 DMA signal [cps normalized by NO3 -(HNO3)2] Signal (10 min average) Signal (30 sec average) Double exp. ft (τ1 = 432s, τ2 = 8640s) DMA fow 0 5 0 DMA fow into chamber (a.u.) M. Simon et al. Figure 3. Decay of the normalized cluster ion signal indicating the DMA concentration with 10 min time resolution (red line) and 30 s time resolution (light red line). The DMA ﬂow (grey line and area) into the chamber is shut-oﬀat ∼06:32 UTC and turned on again (at a lower setpoint) around 12:54 UTC. Using a double-exponential ﬁt (black line) the decaying signal can be well represented. The ﬁrst inverse decay constant represents the wall loss rate (1/τ1 = 2.3×10−3 s−1), while the second decay (1/τ2 = 1.2×10−4 s−1) represents a source term due to slow re-evaporation of DMA from the chamber walls superimposed by dilution. per \| Discussion Paper \| 13281 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. AMTD Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper \| AMTD AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion ssion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| 8, 13257–13284, 2015 4x10 -3 3 2 1 0 DMA signal [cps normalized by NO3 -(HNO3)2] 150 135 120 105 90 75 60 45 30 15 0 DMA setpoint [pptv] CLOUD7 Linear ft (R 2=0.99) CLOUD10T Linear ft (R 2=0.99) Figure 4. Calibration curves for the average DMA signals as a function of the DMA mixing ratio during CLOUD7 (red symbols) and CLOUD10-T (blue symbols). The linear ﬁt for the CLOUD7 calibration follows the expression y = 9.13×10−5+4.41×10−5 pptv−1×x. The expression for the linear ﬁt of the CLOUD10-T (blue) calibration follows y = 14.35 × 10−5 + 2.91 × 10−5 pptv−1 × x. Error bars for the DMA setpoint values are based on a ±10 % uncertainty for each of the MFC ﬂow rate settings and the standard deviation of the ﬁt parameter for kwall. The errors for the measured DMA signals are based on the standard deviation of the mean. 4x10 -3 3 2 1 0 DMA signal [cps normalized by NO3 -(HNO3)2] 150 135 120 105 90 75 60 45 30 15 0 DMA setpoint [pptv] CLOUD7 Linear ft (R 2=0.99) CLOUD10T Linear ft (R 2=0.99) M. Simon et al. Figure 4. Calibration curves for the average DMA signals as a function of the DMA mixing ratio during CLOUD7 (red symbols) and CLOUD10-T (blue symbols). The linear ﬁt for the CLOUD7 calibration follows the expression y = 9.13×10−5+4.41×10−5 pptv−1×x. The expression for the linear ﬁt of the CLOUD10-T (blue) calibration follows y = 14.35 × 10−5 + 2.91 × 10−5 pptv−1 × x. Error bars for the DMA setpoint values are based on a ±10 % uncertainty for each of the MFC ﬂow rate settings and the standard deviation of the ﬁt parameter for kwall. The errors for the measured DMA signals are based on the standard deviation of the mean. per \| Discussion Paper \| 13282 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. AMTD Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper \| series of the normalized cluster ion signal indicating the DMA concentration olution (red lines) and 30 s time resolution (light red line) during the CLOUD7 10-T (b) calibration. The grey lines and areas indicate the targeted DMA mixing e MFC setting for the gas system. The average background signals including d deviation are shown by the horizontal orange lines and the light orange bands. illustrate the averaged periods. AMTD Discussion Paper \| Figure 5. Time series of the normalized cluster ion signal indicating the DMA concentration with 10 min resolution (red lines) and 30 s time resolution (light red line) during the CLOUD7 (a) and CLOUD10-T (b) calibration. The grey lines and areas indicate the targeted DMA mixing ratios due to the MFC setting for the gas system. The average background signals including the 3σ-standard deviation are shown by the horizontal orange lines and the light orange bands. The black lines illustrate the averaged periods. 13283 AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper Discussion Paper \| Discussion Paper \| AMTD AMTD 8, 13257–13284, 2015 Detection of DMA in the low pptv range using a nitrate CIMS M. Simon et al. Title Page Abstract Introduction Conclusions References Tables Figures ◀ ▶ ◀ ▶ Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion sion Paper \| Discussion Paper \| Discussion Paper \| Discussion Paper \| 35 30 25 20 15 10 5 0 DMA concentration [pptv] 21:00 23.10.2012 09:00 24.10.2012 21:00 09:00 25.10.2012 21:00 Time [UTC] IC signal CI-APi-TOF signal Figure 6. Comparison between the measured ion chromatograph signal (IC) and the calculated CI-APi-TOF signal in pptv. The CI-APi-TOF signal is determined by the raw ion counts multiplied by the calibration factor C which is deﬁned by the slope of the linear ﬁt for CLOUD7 (red line of Fig. 4). Both signals show in general a reasonable agreement but the IC signal is lower on average and point-to-point variability is large. The average values over the entire time period displayed are 17 pptv for the IC and 22 pptv for the CI-APi-TOF. 35 30 25 20 15 10 5 0 DMA concentration [pptv] 21:00 23.10.2012 09:00 24.10.2012 21:00 09:00 25.10.2012 21:00 Time [UTC] IC signal CI-APi-TOF signal M. Simon et al. Figure 6. Comparison between the measured ion chromatograph signal (IC) and the calculated CI-APi-TOF signal in pptv. The CI-APi-TOF signal is determined by the raw ion counts multiplied by the calibration factor C which is deﬁned by the slope of the linear ﬁt for CLOUD7 (red line of Fig. 4). Both signals show in general a reasonable agreement but the IC signal is lower on average and point-to-point variability is large. The average values over the entire time period displayed are 17 pptv for the IC and 22 pptv for the CI-APi-TOF. per \| Discussion Paper \| 13284
https://openalex.org/W2794967078	https://www.intechopen.com/citation-pdf-url/58396	English	null	Chaos on Set-Valued Dynamics and Control Sets	InTech eBooks	2,018	cc-by	15,947	Abstract The aim of this chapter is threefold. First, we show some advances in complexity dynamics of set-valued discrete systems in connection with the Devaney’s notion of chaos. Secondly, we start to explore some relationships between control sets for the class of linear control systems on Lie groups with chaotic sets. Finally, through several open problems, we invite the readers to give a contribution to this beauty theory. Keywords: chaos, set-valued maps, dynamic, Devaney, control sets Keywords: chaos, set-valued maps, dynamic, Devaney, control sets Heriberto Román-Flores and Víctor Ayala Additional information is available at the end of the chapter Additional information is available at the end of the chapter http://dx.doi.org/10.5772/intechopen.72232 http://dx.doi.org/10.5772/intechopen.72232 http://dx.doi.org/10.5772/intechopen.72232 Chapter 6 Chapter 6 Chaos on Set-Valued Dynamics and Control Sets1 Heriberto Román-Flores and Víctor Ayala 1. Introduction nt classes of real problems are modelled by a discrete dynamical system Relevant classes of real problems are modelled by a discrete dynamical system (1) xnþ1 ¼ f xn ð Þ , n ¼ 0, 1, 2, … (1) xnþ1 ¼ f xn ð Þ , n ¼ 0, 1, 2, … where X; d ð Þ is a metric space and f : X ! X is a continuous function. The basic goal of this theory is to understand the nature of the orbit O x; f ð Þ ¼ f n x ð Þ= n ¼ 0; 1; 2; … f g for any state x ∈X, as n becomes large and, in general this is a hard task. The study of orbits says us how the initial states are moving in the base space X and, in many cases, these orbits present a chaotic structure. In 1989 in [1], Devaney isolates three main conditions which determine the essential features of chaos. where X; d ð Þ is a metric space and f : X ! X is a continuous function. The basic goal of this theory is to understand the nature of the orbit O x; f ð Þ ¼ f n x ð Þ= n ¼ 0; 1; 2; … f g for any state x ∈X, as n becomes large and, in general this is a hard task. The study of orbits says us how the initial states are moving in the base space X and, in many cases, these orbits present a chaotic structure. In 1989 in [1], Devaney isolates three main conditions which determine the essential features of chaos. Definition 1 Let X be a metric space and f : X ! X a continuous map. Hence, f . Definition 1 Let X be a metric space and f : X ! X a continuous map. Hence, f . a. is transitive if for any couple of non-empty open subsets U and V of X there exists a natural number k such that f k U ð Þ ∩V 6¼ ∅. a. is transitive if for any couple of non-empty open subsets U and V of X there exists a natural number k such that f k U ð Þ ∩V 6¼ ∅. b. is periodically dense if the set of periodic points of f is a dense subset of X. 1. Introduction It is clear that f exact ) f mixing ) f weakly mixing ) f transitive. Remark 3 Also, other concepts very useful in this work are the following: i) f is weakly mixing iff for any non-empty open subsets U and V of X there exists a natural number k such that f k U ð Þ ∩V 6¼ ∅and f k V ð Þ ∩V 6¼ ∅. ii) f is mixing iff given two non-empty open subsets U and V of X there exists a natural number k such that f n U ð Þ ∩V 6¼ ∅for all n ≥k. iii) f is exact iff given a non-empty open subsets U there exists a natural number k such that f k U ð Þ ¼ X. It is clear that f exact ) f mixing ) f weakly mixing ) f transitive. Remark 3 Also, other concepts very useful in this work are the following: i) f is weakly mixing iff for any non-empty open subsets U and V of X there exists a natural number k such that f k U ð Þ ∩V 6¼ ∅and f k V ð Þ ∩V 6¼ ∅. ii) f is mixing iff given two non-empty open subsets U and V of X there exists a natural number k such that f n U ð Þ ∩V 6¼ ∅for all n ≥k. iii) f is exact iff given a non-empty open subsets U there exists a natural number k such that f k U ð Þ ¼ X. It is clear that f exact ) f mixing ) f weakly mixing ) f transitive. It is worth to point out that sensitivity dependence on initial conditions was widely under- stood as being the central idea in chaos for many years. However, in a surprising way, Banks et al. has proved that transitivity and periodically density imply sensitivity dependence (for details see [3]). Furthermore, for continuous functions on real intervals, Vellekoop and Berglund in [4] show that transitivity by itself is sufficient to get chaos. This last result is not necessarily true in other type of metric spaces (see Example 4.1 in [5]). However, sometimes we need to know information about the collective dynamics, i.e. how are moved subsets of X via iteration or dynamics induced by f. 1. Introduction 1Partially supported by Conicyt, Chile through Regular Fondecyt Projects no. 1151159 and no. 1150292 respectively. 1Partially supported by Conicyt, Chile through Regular Fondecyt Projects no. 1151159 and no. 1150292 respectively. artially supported by Conicyt, Chile through Regular Fondecyt Projects no. 1151159 and no. 1150292 respectively © 2018 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, Chaos Theory 98 Chaos Theory 98 c. has sensitive dependence on initial conditions if there is a positive number δ (a sensitivity constant) such that for every point x ∈X and every neighbourhood N of x there exists a point y ∈N and a non-negative integer number n such that d f n x ð Þ; f n y ð Þ ð Þ ≥δ. Next, we mention a remarkable characterisation of transitive maps. In fact, as a consequence of the Birkhoff Transitivity Theorem (see [2] for details), it is possible to prove. Next, we mention a remarkable characterisation of transitive maps. In fact, as a consequence of the Birkhoff Transitivity Theorem (see [2] for details), it is possible to prove. Proposition 2 Let X be a complete metric space which is also perfect (closed and without isolated points). If f : X ! X is continuous, then f is transitive if and only if there exists at least one orbit O x; f ð Þ dense in X. O x; f ð Þ dense in X. Remark 3 Also, other concepts very useful in this work are the following: i) f is weakly mixing iff for any non-empty open subsets U and V of X there exists a natural number k such that f k U ð Þ ∩V 6¼ ∅and f k V ð Þ ∩V 6¼ ∅. ii) f is mixing iff given two non-empty open subsets U and V of X there exists a natural number k such that f n U ð Þ ∩V 6¼ ∅for all n ≥k. iii) f is exact iff given a non-empty open subsets U there exists a natural number k such that f k U ð Þ ¼ X. 1. Introduction For example, if X denotes an ecosystem and x ∈X, then, by using radio telemetry elements, we can obtain information about the movement of x in the ecosystem X. In this form, it is possible to build an individual displacement function f : X ! X. Of course, this function could be chaotic or not. Eventually, we could also be interested to get information about the collective dynamics induced by f, means, to follow the dynamics of a group of individuals. Thus, in a natural way the following question appears: what is the relationship between individual and collective dynamics? This is the main topic of this chapter. iven the system (1), consider the set-valued discrete system associated to f defined by Anþ1 ¼ f An ð Þ, n ¼ 0, 1, 2, … (2) (2) where f is the natural extension of f to the metric space K X ð Þ; H ð Þ of the non-empty compact subsets of X endowed with the Hausdorff metric H induced by the original distance d of X. In a more general set up, this work is strictly related with the following fundamental question: what is the relationship between individual and collective chaos? As a partial response to this question, in this chapter we search the transitivity of a continuous function f on X in relation to the transitivity of its extension f to K X ð Þ: Our main result here establishes that f transitive implies f transitive. That is to say, collective chaos implies individ- ual chaos under the dynamics of f . Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 99 On the other hand, we propose a new approach to this problem: to study the dynamics induced by f on the subextension Kc X ð Þ of K X ð Þ: Precisely, on the class of non-empty compact-convex subsets of X. We prove that the induced dynamics is less chaotic than the original one! Finally, we mention that some relevant problems in the theory of control systems can be also approached by the theory of set-valuated map. In fact, to any initial state x of the system, one can associate its reachable set A x ð Þ. 1. Introduction In other words, A x ð Þ contains all the possible states of the manifold that starting from x you can reach in non-negative time by using the admissible control functions U of the system. The aim of this section is twofold. First of all, to apply to the class of linear control systems on Lie groups, the existent relationship between control sets of an affine control system Σ on a Riemannian manifold M with chaotic sets of the shift flow induced by Σ on M U, [6]. In particular, we are looking for the consequences of this relation on the controllability property. At the very end, we propose a challenge to the readers to motivate the research on this topic through some open problem relatives to the mentioned relationship. 2. Preliminaries In this section, we mention some notions and fundamental results we use through the chapter. 2.1. Extensions If X; d ð Þ is a metric space and f : X ! X continuous, then we can consider the space K X ð Þ; H ð Þ of all non-empty and compact subsets of X endowed with the Hausdorff metric induced by d and f : K X ð Þ ! K X ð Þ, f A ð Þ ¼ f A ð Þg, the natural extension of f to K X ð Þ. Also, we denote by Kc X ð Þ ¼ A ∈K X ð Þ=A is convex f g. If A ∈K X ð Þ we define the “e -dilatation of A” as the set N A; e ð Þ ¼ x ∈X= d x; A ð Þ < e f g, where d x; A ð Þ ¼ inf a ∈Ad x; a ð Þ. The Hausdorff metric on K X ð Þ is given by H A; B ð Þ ¼ inf e > 0= A⊆N B; e ð Þ and B⊆N A; e ð Þ f g: The Hausdorff metric on K X ð Þ is given by H A; B ð Þ ¼ inf e > 0= A⊆N B; e ð Þ and B⊆N A; e ð Þ f g: We know that K X ð Þ; H ð Þ is a complete (separable, compact) metric space if and only if X; d ð Þ is a complete (separable, compact) metric space, respectively, (see [3, 7, 8]). Also, if A ∈K X ð Þ, the set B A; e ð Þ ¼ B ∈K X ð Þ=H A; B ð Þ < e f g denotes the ball centred in A and radius e in the space K X ð Þ; H ð Þ. Furthermore, given a continuous function I; d ð Þ ! f I; d ð Þ on a real interval I, we also consider the extension Kc Ið Þ; H ð Þ ! f c Kc Ið Þ; H ð Þ, where f c is the restriction f Kc Ið Þ. 2.2. Baire spaces In this section, we review some properties of Baire spaces. In this section, we review some properties of Baire spaces. In this section, we review some properties of Baire spaces. Definition 4 A topological space X is a Baire space if for any given countable family of closed sets An : n ∈N f g covering X, then int An ð Þ 6¼ ∅for at least one n. Definition 4 A topological space X is a Baire space if for any given countable family of closed sets An : n ∈N f g covering X, then int An ð Þ 6¼ ∅for at least one n. Chaos Theory 100 Definition 5 In any Baire space X, 1. D ⊂X is called nowhere dense if int cl D ð Þ ð Þ ¼ ∅: 2. Any countable union of nowhere dense sets is called a set of first category. 3. Any set not of first category is said to be of second category. 4. The complement of a set of first category is called a residual set. Remark 6 It is important to note that: a. Any complete metric space is a Baire space. b. Every residual set is of second category in X. c. Every residual set is dense in X. d. The complement of a residual set is of first category. e. If B is of first category and A⊆B, then Ais of first category. (For details, see [8–10]) In particular, if X ¼ I is an interval, then C X ð Þ and C X; R ð Þ, endowed with the respectiv supremum metrics, are Baire spaces. In a Baire space X, we say that “most elements of X” verify the property (P) if the set of all x ∈ Definition 5 In any Baire space X, Definition 5 In any Baire space X, 1. D ⊂X is called nowhere dense if int cl D ð Þ ð Þ ¼ ∅: 2. Any countable union of nowhere dense sets is called a set of first category. 3. Any set not of first category is said to be of second category. 4. The complement of a set of first category is called a residual set. Remark 6 It is important to note that: a. Any complete metric space is a Baire space. b. Every residual set is of second category in X. c. Every residual set is dense in X. 1. D ⊂X is called nowhere dense if int cl D ð Þ ð Þ ¼ ∅: c. In this section, we review some properties of Baire spaces. Every residual set is dense in X. d. The complement of a residual set is of first category. d. The complement of a residual set is of first category. e. If B is of first category and A⊆B, then Ais of first category. In particular, if X ¼ I is an interval, then C X ð Þ and C X; R ð Þ, endowed with the respective supremum metrics, are Baire spaces. In a Baire space X, we say that “most elements of X” verify the property (P) if the set of all x ∈X that do not verify property (P) is of first category in X. In this form, sets of second category can be regarded as “big” sets. A relevant area of the real analysis is to estimate the “size” of some sets associated to a continuous interval function f such as the set P fð Þ of periodic points of f , or the set F fð Þ of fixed points of f . Typically, continuous interval functions have a first category set of periodic points (see [11]) and, in particular, a first category set of fixed points. It has also been recently proved that a typical continuously differentiable interval function has a finite set of fixed points and a countable set of periodic points (see [12] and references therein). It is also well- known that theclassofnowhere differentiablefunctions ND Ið Þ is a residual set in C Ið Þ (see [13,14]). Also, a special class of functions in C Ið Þ is the class CN L Ið Þ of all continuous functions whose graphs “cross no lines” defined in a negative way as follows (see [10]): Definition 7 Let f : a; b ½ ! a; b ½ a continuous map and L : R ! R a function whose graph is a straight line. We say that L crosses f (or f crosses L) if there exists x0 ∈a; b ½ and δ > 0 such that f x0 ð Þ ¼ L x0 ð Þ and either. (a) L x ð Þ ≤f x ð Þ for all x ∈x0 δ; x0 ½ ∩a; b ½ and L x ð Þ ≥f x ð Þ for all x ∈x0; x0 þ δ ½ ∩a; b ½ ; or. 2.3. The dynamics of control theory In Section 7, we propose some challenges through the relationship between the notion of chaotic sets in the Devaney sense and control sets for the class of Linear Control Systems on Lie Groups, [15]. In particular, we explicitly show some results concerning the controllability property in terms of chaotic dynamics. In the sequel, we follow the relevant book The Dynamics of Control by Colonius and Kliemann, [6]. Let M be a d dimensional smooth manifold. By an affine control system Σ in M, we understand the family of ordinary differential equations: Σ : _x tð Þ ¼ X x tð Þ ð Þ þ X m j¼1 uj tð ÞYj x tð Þ ð Þ, u ¼ u1; …; um ð Þ ∈U (3) (3) where X, Yj, j ¼ 0, 1, …, m are arbitrary C∞vector fields on M: The set U ⊂L∞R; Ω ⊂Rm ð Þ is the class of restricted admissible control functions where Ω ⊂Rm with 0 ∈intΩ, is a compact and convex set. Assume Σ satisfy the Lie algebra rank condition, i.e. Assume Σ satisfy the Lie algebra rank condition, i.e. for any x ∈M ) SpanLA X; Y1; …; Ym x ð Þ ¼ d: Of course, LA means the Lie algebra generated by the vector fields through the usual notion of Lie bracket. Furthermore, the ad -rank condition for Σ is defined as follows: Of course, LA means the Lie algebra generated by the vector fields through the usual notion of Lie bracket. Furthermore, the ad -rank condition for Σ is defined as follows: Of course, LA means the Lie algebra generated by the vector fields through the u Lie bracket. Furthermore, the ad -rank condition for Σ is defined as follows: for any x ∈M ) Span adi Yj : j ¼ 1; …; m and i ¼ 0; 1; … x ð Þ ¼ d: For each u ∈U and each initial value x ∈M, there exists an unique solution w t; x; u ð Þ defined on an open interval containing t ¼ 0, satisfying w 0; x; u ð Þ ¼ x. Since we are concerned with dynamics on Lie Groups, without loss of generality we assume that the vector fields X, Y1, …, Ym are completes. In this section, we review some properties of Baire spaces. (b) L x ð Þ ≥f x ð Þ for all x ∈x0 δ; x0 ½ ∩a; b ½ and new L x ð Þ ≤f x ð Þ for all x ∈x0; x0 þ δ ½ ∩a; b ½ . The following result can be found in [10]: (a) L x ð Þ ≤f x ð Þ for all x ∈x0 δ; x0 ½ ∩a; b ½ and L x ð Þ ≥f x ð Þ for all x ∈x0; x0 þ δ ½ ∩a; b ½ ; or. (b) L x ð Þ ≥f x ð Þ for all x ∈x0 δ; x0 ½ ∩a; b ½ and new L x ð Þ ≤f x ð Þ for all x ∈x0; x0 þ δ ½ ∩a; b ½ . The following result can be found in [10]: The following result can be found in [10]: Theorem 8 ([10]) The set CN L Ið Þ ¼ f ∈C Ið Þ= f crosses no lines f g is residual in C Ið Þ. The set CN L Ið Þ will play an important role in the next sections. Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 101 2.3. The dynamics of control theory Then, we obtain a mapping Φ satisfying the cocycle property Φ : R M U ! M, t; x; u ð Þ ↦Φ t; x; u ð Þ and Φ t þ s; x; u ð Þ ¼ Φ t; Φ s; x; u ð Þ; Θsu ð Þ for all t, s ∈R, x ∈M, u ∈U: Where, for any t ∈R, the map Θt is the shift flow on U defined by Θs u ð Þ tð Þ≔u t þ s ð Þ: Hence, Φ is a skew-product flow. The topology here is given by the product topology between the topology of the manifold and the weak* topology on U: It turns out the following results. It turns out the following results. Lemma 9 [6] Consider the set U equipped with the weak* topology associated to L∞R; Rm ð Þ ¼ L1 R; Rm ð Þ∗ as a dual vector space. Therefore, Lemma 9 [6] Consider the set U equipped with the weak* topology associated to L∞R; Rm ð Þ ¼ L1 R; Rm ð Þ∗ as a dual vector space. Therefore, 1. U; d ð Þ is a compact, complete and separable metric space with the distance given by d u1; u2 ð Þ ¼ X ∞ n¼1 1 2n Ð R < u1 tð Þ u2 tð Þ; vn tð Þ > dt 1 þ Ð R < u1 tð Þ u2 tð Þ; vn tð Þ > dt : Here, vn : n ∈N f g ⊂L1 R; Rm ð Þ is a dense set of Lebesgue integrable functions. Chaos Theory 102 2. The map Θ : R U ! U defines a continuous dynamical systems on U. Its periodic points are dense and the shift is topologically mixing (and then topologically transitive). 2. The map Θ : R U ! U defines a continuous dynamical systems on U. Its periodic points are dense and the shift is topologically mixing (and then topologically transitive). 2. The map Θ : R U ! U defines a continuous dynamical systems on U. Its periodic points are dense and the shift is topologically mixing (and then topologically transitive). 3. The map Φ defines a continuous dynamical system on M U: 3. Definition 10 A non-empty set C ⊂M is called a control set of (3) if. i. for every x ∈M there exists u ∈U such that w t; x; u ð Þ : t ≥0 f g ⊂C ii. for every x ∈C, C ⊂cl A x ð Þ ð Þ iii. C is maximal with respect to the properties ið Þ and ii ð Þ: A x ð Þ denotes the states that can be reached from x by Σ in positive time and cl its closure A x ð Þ ¼ y ∈M : ∃u ∈U and t > 0 with y ¼ w t; x; u ð Þ f g: A x ð Þ denotes the states that can be reached from x by Σ in positive time and cl its closure A x ð Þ ¼ y ∈M : ∃u ∈U and t > 0 with y ¼ w t; x; u ð Þ f g: Moreover, for an element x ∈M, the set of points that can be steered to x through a Σ-trajectory in positive time is denoted by A∗x ð Þ ¼ ∪ τ>0 y ∈M : ∃u ∈U; e ¼ wτ,u x ð Þ : Finally, we mention that the Lie algebra rank condition warranty that the system is locally accessible, which means that for every τ > 0, Finally, we mention that the Lie algebra rank condition warranty that the system is locally accessible, which means that for every τ > 0, int A ≤τ x ð Þ ð Þ and int A∗ ≤τ x ð Þ are non empty, for any x ∈M: int A ≤τ x ð Þ ð Þ and int A∗ ≤τ x ð Þ are non empty, for any x ∈M: 2.3. The dynamics of control theory The map Φ defines a continuous dynamical system on M U: On the other hand, the completely controllable property of Σ, i.e. the possibility to connect any two arbitrary points of M through a Σ-trajectory in positive time, is one of the most relevant issue for any control system. But, few systems have this property. A more realistic approach comes from a Kliemann notion introduced in [16]. Definition 10 A non-empty set C ⊂M is called a control set of (3) if. Definition 10 A non-empty set C ⊂M is called a control set of (3) if. 3. f transitive implies f transitive Proof: Let A, B be two non-empty open sets in X. Due to Lemma 13, e A ð Þ and e B ð Þ are non- empty open sets in K X ð Þ. Thus, by transitivity of f , there exists some k ∈N such that Proof: Let A, B be two non-empty open sets in X. Due to Lemma 13, e A ð Þ and e B ð Þ empty open sets in K X ð Þ. Thus, by transitivity of f , there exists some k ∈N such that f k e A ð Þ ð Þ ∩e B ð Þ ¼ f k e A ð Þ ð Þ ∩e B ð Þ 6¼ ∅ and, from Lemma 14, we obtain e f k A ð Þ ∩e B ð Þ ¼ e f k A ð Þ ∩B 6¼ ∅ e f k A ð Þ ∩e B ð Þ ¼ e f k A ð Þ ∩B 6¼ ∅ which implies f k A ð Þ ∩B 6¼ ∅and, consequently, f is a transitive function. which implies f k A ð Þ ∩B 6¼ ∅and, consequently, f is a transitive function. 3. f transitive implies f transitive As we explain, in terms of the original dynamics and its extensions a natural question arises: what are the relations between individual and collective chaos? As a partial response to this question, in the sequel, we show that the transitivity of the extension f implies the transitivity of f : For that, we need to describe some previous results. Lemma 11 [5] Let A be a non-empty open subset of X. If K ∈K X ð Þ and K ⊂A, then there exists e > 0 such that N K; e ð Þ ⊂A:. Definition 12 Let A ⊂X be. Then the extension of A to K X ð Þ is given by e A ð Þ ¼ K ∈K X ð Þ= f Definition 12 Let A ⊂X be. Then the extension of A to K X ð Þ is given by e A ð Þ ¼ K ∈K X ð Þ= K ⊂A f g. Remark 13 e A ð Þ ¼ ∅⇔A ¼ ∅:. Remark 13 e A ð Þ ¼ ∅⇔A ¼ ∅:. Lemma 14 [5] Let A ⊂X be, A 6¼ ∅, an open subset of X. Then, e A ð Þ is a non-empty open subset of K X ð Þ. Lemma 14 [5] Let A ⊂X be, A 6¼ ∅, an open subset of X. Then, e A ð Þ is a non-empty open subset of K X ð Þ. Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 103 103 Lemma 15 [5] If A, B ⊂X, then: i) e A ∩B ð Þ ¼ e A ð Þ ∩e B ð Þ, ii) f e A ð Þ ð Þ⊆e f A ð Þ ð Þ, and iii) f p ¼ f p, for every p ∈N. Lemma 15 [5] If A, B ⊂X, then: i) e A ∩B ð Þ ¼ e A ð Þ ∩e B ð Þ, ii) f e A ð Þ ð Þ⊆e f A ð Þ ð Þ, and iii) f p ¼ f p, for every p ∈N. Now, we are in a position to prove the following results Theorem 16 Let f : X ! X be a continuous function. Then, f transitive implies f transitive. Theorem 16 Let f : X ! X be a continuous function. Then, f transitive implies f transitive. 4. Two examples Now we show that, in general, the converse of Theorem 15 is not true. Example 4.1 (Translations of the circle). If λ ∈R is an irrational number and we define Tλ : S1 ! S1 by Tλ eiθ ¼ ei θþ2πλ ð Þ, then it was shown by Devaney [1] that each orbit Tn λ eiθ =n ∈N is dense in S1 and, due Proposition 2, Tλ is transitive. Nevertheless, Tλ has no periodic points and, because Tλ is isometric, it does not exhibit sensitive dependence on initial conditions either. If K ∈K S1 , because Tλ preserves diameter, then diam K ð Þ ¼ diam Tλ n K ð Þ , for all n ∈N. If K ∈K S1 , because Tλ preserves diameter, then diam K ð Þ ¼ diam Tλ n K ð Þ , for all n ∈N. Now, let K ∈K S1 such that diam K ð Þ ¼ 1, and let e > 0 sufficiently small. Then If K ∈K S , because Tλ preserves diameter, then diam K ð Þ diam Tλ K ð Þ , for all n Now, let K ∈K S1 such that diam K ð Þ ¼ 1, and let e > 0 sufficiently small. Then Now, let K ∈K S1 such that diam K ð Þ ¼ 1, and let e > 0 sufficiently small. Then F ∈U ¼ B K; e ð Þ ) diam F ð Þ ≈1 G ∈V ¼ B 1 f g; e ð Þ ) diam G ð Þ ≈0: Thus, diam Tλ n F ð Þ ≈1 ∀n ∈N and, consequently, Tλ n U ð Þ ∩V ¼ ∅for all n ∈N, which implies that Tλ is not transitive on K S1 . Thus, diam Tλ n F ð Þ ≈1 ∀n ∈N and, consequently, Tλ n U ð Þ ∩V ¼ ∅for all n ∈N, which implies that Tλ is not transitive on K S1 . Example 4.2 Define the “tent” function f : 0; 1 ½ ! 4. Two examples However, considering the we-topology on K X ð Þ generated by the sets e A ð Þ with A an open set in X, we obtain the following complementary result, see [5]: Theorem 18 For a continuous map f : X ! X the following conditions are equivalent: Theorem 18 For a continuous map f : X ! X the following conditions are equivalent: i) f is transitive in X d ð Þ ii) f is transitive in the we topology i) f is transitive in X; d ð Þ, ii) f is transitive in the we-topology. 4. Two examples 0; 1 ½ as f x ð Þ ¼ 2x if 0 ≤x ≤1=2 and f x ð Þ ¼ 2 1 x ð Þ if 1=2 ≤x ≤1. Example 4.2 Define the “tent” function f : 0; 1 ½ ! 0; 1 ½ as f x ð Þ ¼ 2x if 0 ≤x ≤1=2 and f x ð Þ ¼ 2 1 x ð Þ if 1=2 ≤x ≤1. It is not difficult to show that f is an exact function on [0,1]. In fact, intuitively we can see that, after each iteration, the number of tent in the graphics is increasing, whereas the base of each tent is decreasing and they are uniformly distributed over the interval 0; 1 ½ . Chaos Theory 104 Chaos Theory Thus, if U is an arbitrary non-empty open subset of 0; 1 ½ , then U contains an open interval J and, after certain number of iterations, there exists a tent, with height equal to one, whose base is contained in J, which implies that f U ð Þ ¼ 0; 1 ½ and, according to Remark 3, f is an exact mapping and, consequently, f is a mixing function. Thus, if U is an arbitrary non-empty open subset of 0; 1 ½ , then U contains an open interval J and, after certain number of iterations, there exists a tent, with height equal to one, whose base is contained in J, which implies that f U ð Þ ¼ 0; 1 ½ and, according to Remark 3, f is an exact mapping and, consequently, f is a mixing function. The conclusions in Examples 4.1 and 4.2 come from the next result, Banks [17] in 2005. Theorem 17 If f : X ! X is continuous, then the following conditions are equivalent: Theorem 17 If f : X ! X is continuous, then the following conditions are equivalent: i) f is weakly mixing, ii) f is weakly mixing, iii) f is transitive. i) f is weakly mixing, ii) f is weakly mixing, iii) f is transitive. Hitherto, we have used the strong topology induced by the H-metric on K X ð Þ. 5. Sensitivity and periodic density of f Let f : X ! X be a continuous function and let f be its corresponding extension to the hyper- space K X ð Þ. Then, the study of sensitivity of f in the base space in relation to the sensitivity of f on K X ð Þ has been very exhaustively analysed in the last years. Román and Chalco published the first result in this direction [18] in 2005, where the authors prove Theorem 19 f sensitively dependent implies f sensitively dependent. We finish this section assuming the existence of a dense set of periodic points for f, we sh this section assuming the existence of a dense set of periodic points for f, we have Theorem 24 Let X; d ð Þ be a compact metric space and f : X ! X a continuous function. If f : X ! X has a dense set of periodic points then f : K X ð Þ ! K X ð Þ has the same property. Proof: Let K ∈K X ð Þ and e > 0. Then there exists a e=2-net covering K, That is to say, there are x1, …, xp in K such that K ⊂B x1; e=2 ð Þ∪…∪B xp; e=2 : Because f has periodic density, there are yi ∈X and ni ∈N such that: yi ∈B xi; e=2 ð Þ , ∀i ¼ 1, …, p and f ni yi ¼ yi , ∀i ¼ 1, …, p: Now, take G ¼ y1; …; yp n o : By construction, we have H K; G ð Þ < e and, moreover, f n1n2…np yi ¼ yi, for all i ¼ 1, …, p. Therefore, f n1n2…np G ð Þ ¼ G, which implies that f has periodic density. Now, take G ¼ y1; …; yp n o : By construction, we have H K; G ð Þ < e and, moreover, f n1n2…np yi ¼ yi, for all i ¼ 1, …, p. Therefore, f n1n2…np G ð Þ ¼ G, which implies that f has periodic density. The converse of this theorem is no longer true (for a counterexample, see Banks [17]). How- ever, to find conditions on f warranting the existence of a dense set of periodic points for f is a very hard problem which still remains open. Theorem 19 f sensitively dependent implies f sensitively dependent. Proof: If f has sensitive dependence, then there exists a constant δ > 0 such that for every K ∈K X ð Þ and every e > 0 there exists G ∈B K; e ð Þ and n ∈N such that H f n K ð Þ; f n G ð Þ ð Þ ≥δ. Now, let x ∈X be and e > 0. Then, taking K ¼ x f g ∈K X ð Þ, we have that there exists G ∈B x f g; e ð Þ and n ∈N such that H f n x f g ð Þ; f n G ð Þ ð Þ ¼ H f n x ð Þ; f n G ð Þ ð Þ ≥δ. Thus, H f n x ð Þ; f n G ð Þ ð Þ ¼ supy ∈G d f n x ð Þ; f n y ð Þ ð Þ ≥δ and, due to the compactness of G and the continuity of f , there exists y0 ∈G such that H f n x ð Þ; f n G ð Þ ð Þ ¼ d f n x ð Þ; f n y0 ≥δ. But, G ∈B x; e ð Þ implies G ⊂B x; e ð Þ and, consequently, y0 ∈B x; e ð Þ. This proves that f is sensi- tively dependent (with constant δ). The reverse of this theorem is not true. In fact, recently Sharma and Nagar [19] show an example where X; d ð Þ is sensitive but K X ð Þ; H ð Þ is not. Now, in order to overcome that short- coming, the authors in [19] introduce the following notion of sensitivity: Definition 20 (Stronger sensitivity [19]). Let f : X ! X be a continuous function. Then f is strongly sensitive if there exists δ > 0 such that for each x ∈X and each e > 0, there exists n0 ∈N such that for every n ≥n0, there is a y ∈X with d x; y ð Þ < e and d f n x ð Þ; f n y ð Þ ð Þ > δ. Theorem 19 f sensitively dependent implies f sensitively dependent. Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 105 Obviously, the notion of stronger sensitivity is more restrictive than sensitivity, and the authors in [19] obtain the following results: Obviously, the notion of stronger sensitivity is more restrictive than sensitivity, and the authors in [19] obtain the following results: Theorem 21 If f : X ! X is a continuous function and K X ð Þ; H; f is strongly sensitive then X; d; f ð Þ is strongly sensitive. In the compact case, it is possible to obtain a characterization as follows. the compact case, it is possible to obtain a characterization as follows. Theorem 22 Let X; d ð Þ be a compact metric space and f : X ! X a continuous function. Then K X ð Þ; H; f is strongly sensitive if and only if X; d; f ð Þ is strongly sensitive. Theorem 22 Let X; d ð Þ be a compact metric space and f : X ! X a continuous function. Then K X ð Þ; H; f is strongly sensitive if and only if X; d; f ð Þ is strongly sensitive. In connection with these results, recently Subrahmomian ([20], 2007) has been shown that most of the important sensitive dynamical systems are all strongly sensitive (the author here calls them cofinitely sensitive). Hence, we can say that for most cases, sensitivity is equivalent in both cases X; d ð Þ and K X ð Þ; H ð Þ. It turns out that, strongly sensitivity and sensitivity are equivalent on the class of interval functions, which implies that Theorem 23 If f : I ! I is a continuous function, the following conditions are equivalent. a) I; d; f ð Þ is sensitive, b) K Ið Þ; H; f is sensitive. Theorem 23 If f : I ! I is a continuous function, the following conditions are equivalent. I; d; f ð Þ is sensitive, b) K Ið Þ; H; f is sensitive. 6. The dynamics on the Kc Ið Þ; H ð Þ extension In the previous sections, we have studied the diagram K X ð Þ; H ð Þ ! f K X ð Þ; H ð Þ ↑ ↑ X; d ð Þ ! f X;d ð Þ (4) (4) Chaos Theory 106 Chaos Theory 106 and the chaotic relationships between f and f . However, in the setting of mathematical model- ling of many real-world applications, it is necessary to take into account additional consider- ations such as vagueness or uncertainty on the variables. This implies the use of interval parameters and, consequently, to deal with interval systems. That is, it is necessary to consider an interval X ¼ I and to study the following new diagram: Kc Ið Þ; H ð Þ ! f c Kc Ið Þ; H ð Þ ↑ ↑ I; d ð Þ ! f I;d ð Þ (5) (5) along with the analysis of the connection between their respective dynamical relationships. Here f c denotes the restriction of f to Kc Ið Þ, the class of all compact subintervals of I. For A ¼ a; b ½ , B ¼ c; d ½ ∈Kc Ið Þ, the Hausdorff metric can be explicitly computed as H A; B ð Þ ¼ max a c j j; b d j j f g: (6) (6) The aim of this section is to show that the Devaney complexity of the extension f c on Kc Ið Þ is less or equal than the complexity of f on the base space I. More precisely, f c is never transitive for any continuous function f ∈C Ið Þ. Also, we will show that f c has no dense set of periodic points for most functions f ∈C Ið Þ: Finally, we prove that f c has no sensitive dependence for most functions f ∈C Ið Þ. As a motivation, we present the following examples. As a motivation, we present the following examples. Example 6.1 Consider the “tent” function f : 0; 1 ½ ! 0; 1 ½ defined by f x ð Þ ¼ 2x if 0 ≤x ≤1 2 2 1 x ð Þ if 1 2 ≤x ≤1: 8 > < > : Then it is well known that f is D-chaotic on 0; 1 ½ (see [1]). Thus, f c is not transitive on Kc 0; 1 ½ ð Þ. Thus, we obtain that 1 4 ∈K ) 3 4 ∈f K ð Þ ) f K ð Þ 6¼ K. Thus, we obtain that 1 4 ∈K ) 3 4 ∈f K ð Þ ) f K ð Þ 6¼ K. On the other hand, 3 4 ∈f K ð Þ ) 3 4 ∈f n K ð Þ, ∀n ≥2 ) f n K ð Þ 6¼ K, ∀n ≥1 and, consequently, f c has no periodic points in the ball B 1 8 ; 3 8 ; 1 4 ⊆Kc 0; 1 ½ ð Þ; H ð Þ, which implies that f c has no dense set of periodic points on Kc 0; 1 ½ ð Þ; H ð Þ. and, consequently, f c has no periodic points in the ball B 1 8 ; 3 8 ; 1 4 ⊆Kc 0; 1 ½ ð Þ; H ð Þ, which implies that f c has no dense set of periodic points on Kc 0; 1 ½ ð Þ; H ð Þ. Lemma 25 f c transitive on Kc a; b ½ ð Þ implies f transitive on a; b ½ . Proof. Let U, V non-empty open subsets of X ¼ a; b ½ . We can choose x ∈U, y ∈V and e > 0 such that B x; e ð Þ ⊂U and B y; e ð Þ ⊂V. Now, in Kc a; b ½ ð Þ consider the open balls B x f g; e ð Þ and B y f g; e ð Þ with respect to the H-metric. Due to the transitivity of f c on Kc a; b ½ ð Þ, there exists n ∈N such that f n c B x f g; e ð Þ ð Þ ∩B y f g; e ð Þ 6¼ ∅. Therefore, there exists an interval J ∈B x f g; e ð Þ such that f n c Jð Þ ¼ f n Jð Þ ∈B y f g; e ð Þ. However, J ⊂B x; e ð Þ and, analogously, f n Jð Þ ⊂B y; e ð Þ, which implies that f n B x; e ð Þ ð Þ ∩B y; e ð Þ 6¼ ∅and, consequently, f n U ð Þ ∩V 6¼ ∅. Thus, f c is not transitive on Kc 0; 1 ½ ð Þ. Example 6.1 shows a function f which is transitive on the base space X ¼ 0; 1 ½ and f is also transitive on the total extension K 0; 1 ½ ð Þ, but f c is not transitive on the subextension Kc 0; 1 ½ ð Þ. The following example shows a function f : 0; 1 ½ ! 0, 1 with a dense set of periodic points, and where the total extension of f to K 0; 1 ½ ð Þ also has a dense set of periodic points, whereas f c does not have a dense set of periodic points on Kc 0; 1 ½ ð Þ. Example 6.2. Let X ¼ 0; 1 ½ and consider the “logistic” function f : 0; 1 ½ ! 0; 1 ½ defined by f x ð Þ ¼ 4x 1 x ð Þ. It is well known that f is D-chaotic on 0; 1 ½ (see [1]). Moreover, f is a mixing function. Thus, in particular, f has a dense set of periodic points and, therefore, f also has a dense set of periodic points on the total extension K 0; 1 ½ ð Þ) (see Theorem 24). However, f c has no a dense set of periodic points on Kc X ð Þ. However, f c has no a dense set of periodic points on Kc X ð Þ. In order to see this, we claim that the open ball B 1 8 ; 3 8 ; 1 8 in Kc 0; 1 ½ ð Þ; H ð Þ does not contain periodic points of f c. In fact, if K ¼ c; d ½ ∈B 1 8 ; 3 8 ; 1 8 , then c 1 8 < 1 8 and d 3 8 < 1 8, which implies that 0 < c < 1 4 and 1 4 < d < 1 2. In fact, if K ¼ c; d ½ ∈B 1 8 ; 3 8 ; 1 8 , then c 1 8 < 1 8 and d 3 8 < 1 8, which implies that 0 < c < 1 4 and 1 4 < d < 1 2. Therefore, Therefore, f p c B 0; 1 ½ ; 1 10 ∩B 0 f g; 1 10 ¼ ∅, ∀p ∈N: Thus, f c is not transitive on Kc 0; 1 ½ ð Þ. 6. The dynamics on the Kc Ið Þ; H ð Þ extension Moreover, because f is a mixing function on 0; 1 ½ , then f is transitive on K 0; 1 ½ ð Þ (see [17]). Also, we observe that x ¼ 2 3 is a fixed point of f . On the other hand, it is clear that if K is a compact and convex subset of X ¼ 0; 1 ½ , then f K ð Þ is also a compact and convex subset of X. Consequently, if we let Kc 0; 1 ½ ð Þ denote the class of all closed subintervals of 0; 1 ½ , then we can consider f c as a mapping f c : Kc 0; 1 ½ ð Þ ! Kc 0; 1 ½ ð Þ. We recall that Kc 0; 1 ½ ð Þ is a closed subspace of K 0; 1 ½ ð Þ (see [21]). Now, considering the open balls B 0; 1 ½ ; 1 10 and B 0 f g; 1 10 in Kc 0; 1 ½ ð Þ, we have. K ∈B 0; 1 ½ ; 1 10 ) 2 3 ∈K which implies 2 3 ∈f p c K ð Þ, ∀p ∈N:. K ∈B 0; 1 ½ ; 1 10 ) 2 3 ∈K which implies 2 3 ∈f p c K ð Þ, ∀p ∈N:. K ∈B 0; 1 ½ ; 1 10 ) 2 3 ∈K which implies 2 3 ∈f p c K ð Þ, ∀p ∈N:. On the other hand, if F ∈B 0 f g; 1 10 , then F ⊂0; 1=10 ½ . Consequently, H f p c K ð Þ; F ≥17 30 for every K ∈B 0; 1 ½ ; 1 10 and F ∈B 0 f g; 1 10 . Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 107 Thus, f c is not transitive on Kc 0; 1 ½ ð Þ. Thus, f n B a; b ½ ; e ð Þ ð Þ ∩B a; e ð Þ ¼ ∅, ∀n ∈N: Consequently, f is not transitive on Kc a; b ½ ð Þ. Thus, f c is not transitive on Kc 0; 1 ½ ð Þ. And f is a transitive function on a; b ½ . Chaos Theory 108 Chaos Theory It is well-known that if X ¼ I is an interval, then most functions f ∈C Ið Þ has no dense orbits, that is to say, there exists a residual set D ⊂C Ið Þ such that every function f ∈D has no point whose orbit is dense in I (see [22]) and, consequently, most functions f ∈C Ið Þ are not transitive. From Lemma 24, we can conclude that f c is not transitive for most functions f ∈C Ið Þ. The next theorem provides a stronger result. Theorem 26 Let f : a; b ½ ! a; b ½ be continuous. Then f c is not transitive on Kc a; b ½ ð Þ. Proof. By Schauder Theorem, f has at least one fixed point p ∈a; b ½ . Case 1. Suppose that p ∈a; b ð Þ and let r ¼ max p a; b p f g. Without loss of generality, we can suppose that r ¼ p a and, because a < b, it is clear that r > 0. Now, let r0 ¼ b p > 0 and let e ¼ r0 2. If we consider the open balls B a; b ½ ; e ð Þ, B a f g; e ð Þ ∈ Kc a; b ½ ð Þ, it follows that K ∈B a; b ½ ; e ð Þ ) p ∈K ) p ∈f n K ð Þ for any n ∈N. On the other hand, On the other hand, F ∈B a f g; e ð Þ ) H F; a f g ð Þ < e ) F ⊂a, a þ e : Because r0 < r we get Because r0 < r we get H f n K ð Þ; F ≥p a e ¼ r r0 2 > 0 for each K ∈B a; b ½ ; e ð Þ, F ∈B a; e ð Þ and for any n ∈N. Thus, f n B a; b ½ ; e ð Þ ð Þ ∩B a; e ð Þ ¼ ∅, ∀n ∈N: for each K ∈B a; b ½ ; e ð Þ, F ∈B a; e ð Þ and for any n ∈N. Consequently, f is not transitive on Kc a; b ½ ð Þ. we get d x0; f n x ð Þ ð Þ < e, for all x ∈K: (7) (7) d x0; f n x ð Þ ð Þ < e, for all x ∈K: d x0; f n x ð Þ ð Þ < e, for all x ∈K: Because f n K ð Þ ¼ f n K ð Þ ¼ f n K ð Þ ¼ K and f n is continuous on K then, by the Schauder’s Fixed Point Theorem, there exists xp ∈K such that f n xp ¼ xp. Thus, xp is a periodic point of f and, due to (7), we obtain d x0; xp < e. Hence, f has periodic density on I. □ Theorem 28 Let I ¼ a; b ½ be a compact interval in R. Then f c does not have a dense set of periodic points in Kc Ið Þ, for most functions f ∈C Ið Þ. Proof. The proof is based on an exhaustive analysis of the behaviour of the fixed points of f . We connect this analysis with an adequate residual set in C Ið Þ. The analysis of each fixed point of f is fundamental to decide whether the function f allows or not an extension f c that has a dense set of periodic points. More precisely, the behaviour of each fixed point will imply only two (mutually exclusive) options: A. f c does not have a dense set of periodic points, or. A. f c does not have a dense set of periodic points, or. B. f ∈CN L Ið Þ ½ c, which is a set of first category in C Ið Þ. Towards this end, let f : a; b ½ ! a; b ½ be a continuous function. By the Schauder’s Fixed Point Theorem, f has at least one fixed point p ∈a; b ½ . The proof is divided in. Case 1: f has no fixed points in a; b ð Þ. Case 1: f has no fixed points in a; b ð Þ. In this case, we have the following three subcases: 1iÞ p ¼ a is the unique fixed point of f . 1iÞ p ¼ a is the unique fixed point of f . Consequently, f is not transitive on Kc a; b ½ ð Þ. Case 2. Suppose that f has no fixed points in a; b ð Þ. From the continuity of f , we have that f x ð Þ > x for all x ∈a; b ð Þ or f x ð Þ < x for all x ∈a; b ð Þ. This clearly implies that f is not a transitive function, and consequently, due to Lemma 24, f c is not transitive on Kc a; b ½ ð Þ. An important question to answer is what about the size of the set of periodic points of f c. It is clear that there are some functions f ∈C Ið Þ with a dense set of periodic points on I, and such that their extensions f c also has a dense set of periodic points on Kc Ið Þ (for instance, f x ð Þ ¼ x). Therefore, an analogous result to Theorem 26, but for periodic density of f c, cannot be obtained. However, as we will see, most functions f ∈C Ið Þ do not have an extension f c with a dense set of periodic points on Kc Ið Þ. To prove it, we need the following lemma. Lemma 27 Let I be a compact interval in R, and f : I ! I be a continuous function. If we suppose that f c has periodic density on Kc Ið Þ, then f has periodic density on I. Proof. If x0 ∈I and e > 0 then x0 f g ∈Kc Ið Þ and, consequently, there exists K ∈Kc Ið Þ and n ∈N such that Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 109 a. H x0 f g; K ð Þ < e b f n K ð Þ ¼ K a. H x0 f g; K ð Þ < e b. f n c K ð Þ ¼ K. Combining a. and b. we get Combining a. and b. Consequently, f is not transitive on Kc a; b ½ ð Þ. We have, either We have, either f x ð Þ > x , ∀x ∈a; b ð Þ ) x < f x ð Þ < f 2 x ð Þ < … < f n x ð Þ < … , or f x ð Þ < x , ∀x ∈a; b ð Þ ) x > f x ð Þ > f 2 x ð Þ > … > f n x ð Þ > … : In both cases it follows that f has no periodic points in a; b ð Þ. In both cases it follows that f has no periodic points in a; b ð Þ. In both cases it follows that f has no periodic points in a; b ð Þ. 1iiÞ p ¼ b is the unique fixed point of f . 1iiÞ p ¼ b is the unique fixed point of f . This case is analogous to the case 1iÞ. 1iiiÞ p ¼ a and p ¼ b are the unique fixed points of f . 1iiiÞ p ¼ a and p ¼ b are the unique fixed points of f . This case is also analogous to the cases 1iÞ and 1iiÞ. This case is also analogous to the cases 1iÞ and 1iiÞ. This case is also analogous to the cases 1iÞ and 1iiÞ. Therefore, in case 1 the function f does not have a dense set of periodic points in a; b ½ . Due to Lemma 24, f c does not have a dense set of periodic points in Kc a; b ½ ð Þ. Chaos Theory 110 Chaos Theory Case 2: f has at least one fixed point p ∈a; b ð Þ. Case 2: f has at least one fixed point p ∈a; b ð Þ. Case 2: f has at least one fixed point p ∈a; b ð Þ. We have the following subcases: 2iÞ ∃q ∈a; b ð Þ , q 6¼ p such that f q ð Þ ¼ p. 2iÞ ∃q ∈a; b ð Þ , q 6¼ p such that f q ð Þ ¼ p. Without loss of generality, suppose that q ∈a; p ð Þ. Consequently, f is not transitive on Kc a; b ½ ð Þ. Then, taking 0 < e < min qa 2 ; pq 2 , we can consider the open ball B q e; q þ e ½ ; e ð Þ in the space Kc a; b ½ ð Þ. If J ¼ c; d ½ ∈B q e; q þ e ½ ; e ð Þ, from (6) we have c q e ð Þ j j < e and d q þ e ð Þ j j < e c q e ð Þ j j < e and d q þ e ð Þ j j < e which implies that a < c < q and q < d < p and, consequently, q ∈J whereas p∉J. Thus, q ∈J ) f q ð Þ ¼ p ∈f Jð Þ ) f Jð Þ 6¼ J : (8) which implies that a < c < q and q < d < p and, consequently, q ∈J whereas p∉J. Thus, q ∈J ) f q ð Þ ¼ p ∈f Jð Þ ) f Jð Þ 6¼ J : (8) q ∈J ) f q ð Þ ¼ p ∈f Jð Þ ) f Jð Þ 6¼ J : (8) On the other hand, p ∈f Jð Þ implies that On the other hand, p ∈f Jð Þ implies that On the other hand, p ∈f Jð Þ implies that p ∈f n Jð Þ, ∀n ≥2 ) f n Jð Þ 6¼ J, ∀n ≥2 , (9) (9) and, consequently, f c has no periodic points in the ball B q e; q þ e ½ ; e ð Þ⊆Kc a; b ½ ð Þ; H ð Þ, which implies that f c does not have a dense set of periodic points on Kc a; b ½ ð Þ; H ð Þ. and, consequently, f c has no periodic points in the ball B q e; q þ e ½ ; e ð Þ⊆Kc a; b ½ ð Þ; H ð Þ, which implies that f c does not have a dense set of periodic points on Kc a; b ½ ð Þ; H ð Þ. a, q 6¼ p, is the unique point such that f a ð Þ ¼ p. Consequently, f is not transitive on Kc a; b ½ ð Þ. 2iiÞ q ¼ a, q 6¼ p, is the unique point such that f a ð Þ ¼ p. Without loss of generality, we can suppose that f x ð Þ > p, for all x ∈a; p ð Þ. t loss of generality, we can suppose that f x ð Þ > p, for all x ∈a; p ð Þ. Without loss of generality, we can suppose that f x ð Þ > p, for all x ∈a; p ð Þ. Now, in addition to hypothesis 2iiÞ, we have two subcases: Now, in addition to hypothesis 2iiÞ, we have two subcases: ia1Þ f does not cross the line y ¼ p and f x ð Þ > p for all x ∈a; p ð Þ. In this situation, f x ð Þ ≥p for all x ∈a, b. Thus, choosing q ∈a; p ð Þ and 0 < e < max qa 2 ; pq 2 , we can consider the open ball B q f g; e ð Þ to have In this situation, f x ð Þ ≥p for all x ∈a, b. Thus, choosing q ∈a; p ð Þ and 0 < e < max qa 2 ; pq 2 , we can consider the open ball B q f g; e ð Þ to have K ¼ c; d ½ ∈B q f g; e ð Þ ) K ⊂a; p ð Þ: (10) (10) n f n z ð Þ > p , ∀z ∈K, ∀n ∈N, (11) From our hypothesis, we obtain From our hypothesis, we obtain (11) which implies that f n K ð Þ 6¼ K, ∀n ∈N. Consequently, f c has no periodic points in the ball B q f g; e ð Þ. In other words, f c does not have a dense set of periodic points in Kc Ið Þ. which implies that f n K ð Þ 6¼ K, ∀n ∈N. Consequently, f c has no periodic points in the ball B q f g; e ð Þ. In other words, f c does not have a dense set of periodic points in Kc Ið Þ. 2iia2Þ f does not cross the line y ¼ p and f x ð Þ < p for all x ∈a; p ð Þ. Consequently, f is not transitive on Kc a; b ½ ð Þ. In this case, we have the following subcases: In this case, we have the following subcases: 2iva1Þ f does not cross the line y ¼ p and f x ð Þ > p and f x ð Þ > p for all x ∈a; b ð Þ p f g 2iva1Þ f does not cross the line y ¼ p and f x ð Þ > p and f x ð Þ > p for all x ∈a; b ð Þ p f g. va1Þ f does not cross the line y ¼ p and f x ð Þ > p and f x ð Þ > p for all x ∈a; b ð Þ p f g. This case is analogous to the case 2iia1Þ and the same is true for 2iva2Þ when f does not cross the line y ¼ p and f x ð Þ < p for all x ∈a; b ð Þ p f g which is analogous to the case 2iia2Þ Finally, there only remains two subcases: 2ivb1Þ f crosses the line y ¼ p and f x ð Þ > p in a; p ð Þ and f x ð Þ < p in p; b ð Þ, and. 2ivb2Þ f crosses the line y ¼ p and f x ð Þ < p in a; p ð Þ and f x ð Þ > p in p; b ð Þ. 2ivb1Þ f crosses the line y ¼ p and f x ð Þ > p in a; p ð Þ and f x ð Þ < p in p; b ð Þ, and. 2ivb2Þ f crosses the line y ¼ p and f x ð Þ < p in a; p ð Þ and f x ð Þ > p in p; b ð Þ. It is clear that in both cases f ∈CN L Ið Þ ½ c. Thus, as a direct consequence of the analysis of the behaviour of the set of fixed points of f , it turns out that the unique cases in which f could have an extension f c with a dense set of periodic points on Kc Ið Þ are when there exists a fixed point p of f such that f crosses the line y ¼ p at x ¼ p. Consequently, f is not transitive on Kc a; b ½ ð Þ. does not cross the line y ¼ p and f x ð Þ < p for all x ∈a; p ð Þ. In this case, f x ð Þ ≥p for all x ∈a; b ½ . Thus, choosing q ∈p; b ð Þ and 0 < e < max qp 2 ; bq 2 n o , we can consider the open ball B q f g; e ð Þ to obtain K ¼ c; d ½ ∈B q f g; e ð Þ ) K ⊂p; b ð Þ: (12) K ¼ c; d ½ ∈B q f g; e ð Þ ) K ⊂p; b ð Þ: (12) Again, from our hypothesis, we get Again, from our hypothesis, we get Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 111 f n z ð Þ < p , ∀z ∈K, ∀n ∈N, (13) (13) which implies that f n K ð Þ 6¼ K, ∀n ∈N and, consequently, f c has no periodic points in the ball B q f g; e ð Þ. In other words, f c does not have a dense set of periodic points in Kc Ið Þ. It is clear that, in this case, f ∈CN L Ið Þ ½ c which, due to Theorem 8 and Remark 6, is a set of first category in C a; b ½ ð Þ. 2iiiÞ q ¼ b, q 6¼ p, is the unique point such that f b ð Þ ¼ p. 2iiiÞ q ¼ b, q 6¼ p, is the unique point such that f b ð Þ ¼ p. iiÞ q ¼ b, q 6¼ p, is the unique point such that f b ð Þ ¼ p. 2iiiÞ q ¼ b, q 6¼ p, is the unique point such that f b ð Þ ¼ p. This case is analogous to case 2iiÞ and, consequently, if f does not cross the line y ¼ p then f c does not have a dense set of periodic points in Kc Ið Þ, whereas if f crosses the line y ¼ p, then f ∈CN L Ið Þ ½ c. 2ivÞ q1 ¼ a and q2 ¼ b, q1, q2 6¼ p, are the unique points such that f a ð Þ ¼ f b ð Þ ¼ p. 7.1. Linear control systems on lie groups Let G be a connected d dimensional Lie group with Lie algebra g. A linear control system ΣL on G is an affine system determined by ΣL : _x tð Þ ¼ X x tð Þ ð Þ þ X m j¼1 uj tð ÞYj x tð Þ ð Þ, u ¼ u1; …; um ð Þ ∈U (14) (14) where X is linear, that is, its flow X t ð Þt ∈R is a one-parameter group of G-automorphism, the control vectors Yj, j ¼ 1, …, m are invariant vector fields, as elements of g. The restricted class of admissible control U is the same as before. Certainly, the drift vector field X is complete and the same is true for every invariant vector field Yj, j ¼ 1, …, m. As usual, we assume that ΣL satisfy the Lie algebra rank condition, i.e. for any x ∈M ) SpanLA X; Y1; …; Ym x ð Þ ¼ d: for any x ∈M ) SpanLA X; Y1; …; Ym x ð Þ ¼ d: The system is said to be controllable if A eð Þ ¼ A is G: The system is said to be controllable if A eð Þ ¼ A is G: The class of systems ΣL is huge and contains many relevant algebraic systems as the classical linear and bilinear systems on Euclidean spaces [6], and the class of invariant systems on Lie groups, [24]. Furthermore, according to the Jouan Equivalence Theorem [25], ΣL is also rele- vant in applications. It approaches globally any affine non-linear control system Σ on a Riemannian manifold when the Lie algebra of the dynamics of Σ is finite dimensional. One can associate to X a derivation D of g defined by DY ¼ X; Y ½ eð Þ, Y ∈g:Indeed, the Jacobi identity shows D X; Y ½ ¼ DX; Y ½ þ X; DY ½ is in fact a derivation. The relation between wt and D is given by the formula wt exp Y ¼ exp etDY , for all t ∈R, Y ∈g: Consider the generalised eigenspaces of D defined by gα ¼ X ∈g : D α ð ÞnX ¼ 0 for some n ≥1 f g where α ∈Spec D ð Þ. Consequently, f is not transitive on Kc a; b ½ ð Þ. In other words, we obtain HDS Ið Þ ¼ f ∈C Ið Þ= f c has a dense set of periodic points in Kc Ið Þ ) HDS Ið Þ⊆CN L Ið Þ ½ c, But, CN L Ið Þ is a residual set in C Ið Þ, therefore from Remark 6, we conclude that HDS Ið Þ is of first category in C Ið Þ. Equivalently, f c does not have a dense set of periodic points, for most functions f ∈C Ið Þ, which ends the proof. Finally based on the following result, Theorem 29 ([23]) For most functions f ∈C Ið Þ, the set of all points where f is sensitive is dense in the set of all periodic points of f . we show an analogous result for the sensitivity property, as follows. Theorem 30 For most functions f ∈C Ið Þ, the extension f c ∈C Kc Ið Þ ð Þ is not sensitive. Theorem 30 For most functions f ∈C Ið Þ, the extension f c ∈C Kc Ið Þ ð Þ is not sensitive. Theorem 30 For most functions f ∈C Ið Þ, the extension f c ∈C Kc Ið Þ ð Þ is not sensitive Proof. This is a direct consequence of Theorem 28 and Theorem 29. Proof. This is a direct consequence of Theorem 28 and Theorem 29. Chaos Theory 112 Chaos Theory 7. Control sets of linear systems and chaotic dynamics The aim of this section is twofold. First of all, to start to apply to the class of linear control systems on Lie groups, the existent relationship between control sets of an affine control system Σ on a Riemannian manifold M with chaotic sets of the shift flow induced by Σ on M U, [6]. In particular, we are looking for the consequences of this relation on the controlla- bility property The second part is intended to motivate the research on this topic to writing down some open problems relatives to this relationship. 7.1. Linear control systems on lie groups Then, gα; gβ h i ⊂gαþβ when α þ β is an eigenvalue of D and zero otherwise. Therefore, it is possible to decompose g as g ¼ gþ ⊕g0 ⊕g, where Therefore, it is possible to decompose g as g ¼ gþ ⊕g0 ⊕g, where Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 113 Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 113 113 g ¼ gþ ⊕g0 ⊕g, where gþ ¼ ⊕ α:Re α ð Þ>0 gα, g0 ¼ ⊕ α:Re α ð Þ¼0 gα and g ¼ ⊕ α:Re α ð Þ<0 gα: g ¼ gþ ⊕g0 ⊕g, where gþ ¼ ⊕ α:Re α ð Þ>0 gα, g0 ¼ ⊕ α:Re α ð Þ¼0 gα and g ¼ ⊕ α:Re α ð Þ<0 gα: Actually, gþ, g0, g are Lie algebras and gþ, g are nilpotent. Denote by Gþ, G and G0 the connected and closed Lie subgroups of G with Lie algebras gþ, g and g0 respectively. Despite the fact that for an invariant system the global controllability property is local, this class has been studied for more than 50 years, see [24] and the references there in. The important point to note here is: for an invariant system the reachable set from the identity is a semigroup. However, in [26] the authors show that this is not the case for a linear system which turns the problem more complicated. Therefore, we would like to explore the men- tioned connection between control sets and the Devaney and Colonius-Kliemann ideas. This section is the starting point for the ΣL class. We begin with a fundamental result. Theorem 31 Assume the system ΣL satisfy the Lie algebra rank condition. Therefore, there exists a control set Ce ¼ cl A eð Þ ð Þ ∩A∗eð Þ which contains the identity element e in its interior. Here, A∗eð Þ is the set of states of G that can be sent by ΣL to e in positive time. which contains the identity element e in its interior. Here, A∗eð Þ is the set of states of G that can be sent by ΣL to e in positive time. For a proof in a more general set up, see [6]. 7.1. Linear control systems on lie groups Recently, we were able to establish some algebraic, topological, and dynamical conditions on ΣL to study uniqueness and boundness of control sets and it consequences on controllability : But, the state of arts is really far from being complete. In order to approach this problem for ΣL, as in [27] we assume here that G has finite semisimple centre, i.e. all semisimple Lie subgroups of G have finite center. We notice that any nilpotent and solvable Lie group, and any semisimple Lie group with finite centre has the finite semisimple centre property. But also, the product between groups with finite semisimple centre have the same property. We also assume that A is open. This is true if for example, the system satisfy the ad -rank condition. About the uniqueness and boundness of control sets of a linear systems, we know few things [27]. Theorem 32 Let ΣL a linear control system on the Lie group G: 1. If G ¼ GG0Gþ is decomposable, Ce is the only control set with non-empty interior. In particular, this is true for any solvable Lie group. 2. Suppose that G is semisimple or nilpotent, it turns out that 2. Suppose that G is semisimple or nilpotent, it turns out that if cl AG ð Þ, cl A∗ Gþ and G0 are compact sets C is bounded: 3. If G is a nilpotent simply connected Lie group, it follows that C is bounded ⇔cl AG ð Þ and cl A∗ Gþ are compact sets and D is hyperbolic: C is bounded ⇔cl AG ð Þ and cl A∗ Gþ are compact sets and D is hyperbolic: Furthermore, it is possible to determine algebraic sufficient conditions to decide when C is bounded. Actually, in a forthcoming paper we show that Furthermore, it is possible to determine algebraic sufficient conditions to decide when C is bounded. Actually, in a forthcoming paper we show that Chaos Theory 114 Theorem 33 Let ΣL be a linear control system on the Lie group G: Assume that G is decomposable and Gþ,0 is a normal subgroup of G. Hence, cl G ∩A ð Þ is compact. Theorem 33 Let ΣL be a linear control system on the Lie group G: Assume that G is decomposable and Gþ,0 is a normal subgroup of G. Hence, cl G ∩A ð Þ is compact. 7.1. Linear control systems on lie groups A analogous result is obtained for Gþ ∩A assuming that G,0 is normal. Of course, Gþ,0 is a normal subgroup of G if and only if gþ ⊕g0 is an ideal of g. On the other hand, gþ ⊕g0 and gþ ⊕g0 are ideals of g ⇔gþ; g0 ¼ 0 and gþ; g ½ ⊂g0: 7.2. Chaos and control sets We start with an explicitly relationship between chaotic subsets of M U and the Σ-control sets. Theorem 34 Let ℭ⊂M U and the canonical projection πM : M U ! M: Hence, πM ℭ ð Þ ¼ x ∈M : there exists u ∈U with x; u ð Þ ∈ℭ f g We start with an explicitly relationship between chaotic subsets of M U and the Σ-control sets. Theorem 34 Let ℭ⊂M U and the canonical projection πM : M U ! M: Hence, πM ℭ ð Þ ¼ x ∈M : there exists u ∈U with x; u ð Þ ∈ℭ f g is compact and its non-void interior consists of locally accessible points. Then, is compact and its non-void interior consists of locally accessible points. Then, 1. ℭis a maximal topologically mixing set if and only if there exists a control C such that ℭ¼ cl x; u ð Þ ∈M U : w t; x; u ð Þ ∈int C ð Þ for every t ∈R In this case, C is unique and int C ð Þ ¼ int πM ℭ ð Þ ð Þ, cl C ð Þ ¼ cl πM ℭ ð Þ ð Þ. ℭ¼ cl x; u ð Þ ∈M U : w t; x; u ð Þ ∈int C ð Þ for every t ∈R 2. The periodic points of Φ are dense in ℭ. 2. The periodic points of Φ are dense in ℭ. 2. The periodic points of Φ are dense in ℭ. 3. Φ restrict to ℭis topologically mixing, topologically transitive and has sensitive depen- dence on initial conditions. 3. Φ restrict to ℭis topologically mixing, topologically transitive and has sensitive depen- dence on initial conditions. In order to apply this fundamental result for a non-controllable linear control system, the boundness property of its control set is crucial. Theorem 35 Let ΣL be a linear control system on a Lie group G. Any condition. 7.2. Chaos and control sets Let us assume that C is a bounded control set with non-empty interior of ΣL and define ℭ¼ π1 M C ð Þ ¼ cl C UC ð Þ where UC ¼ u ∈U : exist x ∈C with w t; x; u ð Þ ∈int C ð Þ for every t ∈R UC ¼ u ∈U : exist x ∈C with w t; x; u ð Þ ∈int C ð Þ for every t ∈R : The Lie group G is finite dimensional and UC is a closed subset of the compact class of admissible control U ⊂L∞R; Ω ⊂Rm ð Þ with the weak* topology. Since the projection is a contin- uous map, it turns out that πM ℭ ð Þ is compact and ℭ, C are uniquely defined. On the other hand, we are assuming that ΣL satisfy the Lie algebra rank condition, hence the system is locally accessible at any point of the state space. Therefore, we are in a position to apply Theorem 32, first, for some classes of controllable linear systems, as follows. Theorem 35 Let ΣL be a linear control system on a Lie group G. Any condition. 1. G is compact, or 2. G is Abelian, or 2. G is Abelian, or Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 115 3. G has the finite semisimple centre property and the Lyapunov spectrum of D is 0 f g implies that the skew flow Φ is chaotic in G U. 3. G has the finite semisimple centre property and the Lyapunov spectrum of D is 0 f g implies that the skew flow Φ is chaotic in G U. Proof. Under the hypothesis in 1 ð Þ, any control set is bounded. Furthermore, if G is compact, the Lie algebra rank condition assures that the linear control system ΣL is controllable on G, see [15]. Hence, Φ is topologically mixing, topologically transitive and the periodic points of Φ are dense in G U, which give us the desired conclusion. It is well known that any Abelian Lie group is a product G ¼ Rm Tn between the Euclidean space Rm and the torus Tn ¼ S1 … S1 (n times), for some m, n ∈N: In this case, ΣL is also controllable [15]. 7.2. Chaos and control sets If G is nilpotent and D has only eigenvalues with non-negative real parts then the only chaotic set ℭ¼ π1 M C ð Þ in G U is open Proof. If G is decomposable, we know that there exists just one control set: the one which contains the identity element. Hence, ℭ¼ π1 M Ce ð Þ is the only chaotic set of Φ on G U which proves 1 ð Þ: To prove 2 ð Þ and 3 ð Þ, we observe that the Lyapunov spectrum condition on the derivation D associated to the drift vector field X is equivalent to the control set Ce be closed or open, respectively. Since the projection πG : G U ! G is a continuous map with the weak* topology on U, the lifting π1 G Ce ð Þ is both closed and open, respectively. 7.2. Chaos and control sets Indeed, since the automorphism group of Tn is discrete, any linear vector field on the torus is trivial. But, we are assuming the Lie algebra condition on G which coincides with the Kalman rank condition in Rm: And, on the compact part, we apply 1 ð Þ: Hence, the skew flow Φ is chaotic in G U. In fact, π1 M C ð Þ ¼ G U and the hypothesis of the compacity on the projection in Theorem 32 is not necessary for the lifting, see Proposition 4.3.3 in [6]. The same is true for 3 ð Þ: Actually, for this more general set up, we recently prove that the system is also controllable, [28, 29]. In the sequel, we use some topological properties of Ce to translate these properties to its associated chaotic set ℭ, as follows. Theorem 36 Let ΣL be a linear control system on a Lie group G: It holds. 1. If G ¼ GG0Gþ there exists one and only one chaotic set ℭ¼ π1 M Ce ð Þ in G U given by ℭ¼ cl x; u ð Þ : w t; x; u ð Þ ∈int Ce ð Þ for every t ∈R ⊂M U 1. If G ¼ GG0Gþ there exists one and only one chaotic set ℭ¼ π1 M Ce ð Þ in G U given by ℭ¼ cl x; u ð Þ : w t; x; u ð Þ ∈int Ce ð Þ for every t ∈R ⊂M U 2. If G is nilpotent and D has only eigenvalues with non-positive real parts, then the only chaotic set ℭ¼ π1 M C ð Þ in G U is closed 2. If G is nilpotent and D has only eigenvalues with non-positive real parts, then the only chaotic set ℭ¼ π1 M C ð Þ in G U is closed 2. If G is nilpotent and D has only eigenvalues with non-positive real parts, then the only chaotic set ℭ¼ π1 M C ð Þ in G U is closed 3. If G is nilpotent and D has only eigenvalues with non-negative real parts then the only chaotic set ℭ¼ π1 M C ð Þ in G U is open 3. 7.3. Challenge In this very short section, we would like to invite the readers to work on the relationship between chaotic and control sets. We suggest to go further in this research through some specific examples on low-dimensional Lie groups. For that, we give some relevant information Chaos Theory 116 Chaos Theory 116 about two groups of dimension three: the simply connected nilpotent Heisenberg Lie group H and the special linear group SL 2; R ð Þ. We finish by computing an example on H. 1. The nilpotent Lie algebra h ¼ R3; þ; ; , has the basis E12; E23; E13 f g with E12; E23 ½ ¼ E13: Here, Eij denotes the real matrix of order 3 with zero everywhere except 1 in the position ij: The associated Heisenberg Lie group has the matrix representation 1. The nilpotent Lie algebra h ¼ R3; þ; ; , has the basis E12; E23; E13 f g with E12; E23 ½ ¼ E13: Here, Eij denotes the real matrix of order 3 with zero everywhere except 1 in the position ij: The associated Heisenberg Lie group has the matrix representation G ¼ g ¼ 1 x z 0 1 y 0 0 1 0 B @ 1 C A : x; y; z ∈R 8 > < > : 9 > = > ; ! w:g! x;y;z ð Þ R3: As invariant vector fields, the basis elements of g has the following description E12 ¼ ∂ ∂x , E23 ¼ ∂ ∂y þ x ∂ ∂z and E13 ¼ ∂ ∂z : The canonical form of any g-derivation is given by The canonical form of any g-derivation is given by The canonical form of any g-derivation is given by D ¼ a d 0 b e 0 c f a þ e 0 B @ 1 C A : a, b, c, d, e, f ∈R: Any linear vector field X reads as Any linear vector field X reads as Any linear vector field X reads as X x; y; z ð Þ ¼ ax þ dy ð Þ ∂ ∂x þ bx þ ey ð Þ ∂ ∂y þ b 2 x2 þ d 2 y2 þ cx þ fy þ a þ e ð Þz ∂ ∂z : 2. Author details Heriberto Román-Flores and Víctor Ayala* *Address all correspondence to: vayala@ucn.cl Instituto de Alta Investigación, Universidad de Tarapacá, Sede Esmeralda, Iquique, Chile 7.3. Challenge The vector space g ¼ sl 2; R ð Þ of all real matrices of order three and trace zero is the Lie algebra of the Lie group G ¼ SL 2; R ð Þ ¼ det1 1 ð Þ. Let us consider the following generators of g: Y1 ¼ 0 1 1 0 , Y2 ¼ 0 1 0 0 and Y3 ¼ 1 0 0 1 : The Lie group G is semisimple, then any g derivation is inner which means that there exists an invariant vector field Y such that ad Y ð Þ represents : Thus, a general form of a derivation reads as g g p ; ð Þ ð Þ g g Y1 ¼ 0 1 1 0 , Y2 ¼ 0 1 0 0 and Y3 ¼ 1 0 0 1 : The Lie group G is semisimpl g derivation is inner which means that there exists an invariant vector field Y such that ad Y ð Þ represents : Thus, a general form of a derivation reads as α ad Y1 þ β ad Y2 þ γ ad Y3 : Example 7.1 On the Heisenberg Lie group, consider the system ΣL : g tð Þ ¼ X g tð Þ ð Þ þ u1 tð ÞE12 g tð Þ ð Þ þ u2 tð ÞE23 g tð Þ ð Þ, u ¼ u1; u2 ð Þ ∈U (15) (15) where X is determined by the derivation D ¼ ad E12 ð Þ ¼ E32: Since the group is nilpotent, it has the semisimple finite centre property. The Lyapunov spectrum of D reduces to zero. Finally, the reachable set from the identity A is open. In fact, the ad-rank condition is obviously true because D E12 ð Þ ¼ E13. It turns out that the skew flow Φ is chaotic in H U: Chaos on Set-Valued Dynamics and Control Sets http://dx.doi.org/10.5772/intechopen.72232 117 References [1] Devaney RL. An Introduction to Chaotic Dynamical Systems. Redwood City: Addison- Wesley; 1989 [2] Robinson C. Dynamical Systems: Stability, Symbolic Dynamics, and Chaos. New York: CRC Press; 1999 [3] Banks J, Brooks J, Cairns G, Stacey P. On the Devaney’s definition of chaos. The American Mathematical Monthly. 1992;99:332-334 [4] Vellekoop M, Berglund R. On intervals, transitivity=chaos. The American Mathematical Monthly. 1994;101:353-355 [5] Román-Flores H. A note on transitivity in set-valued discrete systems. Chaos, Solitons & Fractals. 2003;17:99-104 [6] Colonius F, Kliemann C. The Dynamics of Control, Systems & Control: Foundations & Applications. Boston, MA: Birkäuser Boston, Inc.; 2000 [7] Román-Flores H, Barros LC, Bassanezi RCA. Note on the Zadeh’s extensions. Fuzzy Sets and Systems. 2001;17:327-331 [8] Román-Flores H, Chalco-Cano Y. A note on dynamics of interval extensions of interval function. 2015 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS) held jointly with 2015 5th World Conference on Soft Computing (WConSC), 2015 [9] Dugundji J. Topology. Boston: Allyn and Bacon Inc.; 1966 [9] Dugundji J. Topology. Boston: Allyn and Bacon Inc.; 1966 [10] Thomson B, Bruckner J, Bruckner A. Elementary Real Analysis. Prentice Hall: Upper Saddle River; 2001 [11] Agronski S, Bruckner A, Laczkovich M. Dynamics of typical continuous functions. Jour- nal of the London Mathematical Society. 1989;40:227-243 [12] Alikhani-Koopaei A. On the set of fixed points and periodic points of continuously differentiable functions. International Journal of Mathematics and Mathematical Sciences. 2013. Article ID: 929475 [13] Banach S. Über die Baire’sche Kategorie gewisser Funktionenmengen. Studia Math. 1931;3:174-179 Chaos Theory 118 [14] Mazurkiewicz S. Sur les fonctions non-dérivables. Studia Mathematica. 1932;3:92-94 [15] Ayala V, Tirao J. Linear control systems on lie groups and controllability. American Mathematical Society, Series: Symposia in Pure Mathematics. 1999;64:47-64 [16] Kliemann W. Analysis of nonlinear stochastic systems. Schiehlen W, Wedig W, eds. In: Analysis and Estimation of Stochastic Mechanical Systems: Springer-Verlag; 1988 [17] Banks J. Chaos for induced hyperspace maps. Chaos, Solitons & Fractals. 2005;25:681-685 [18] Román-Flores H, Chalco-Cano Y. Robinson’s chaos in set-valued discrete systems. Chaos, Solitons & Fractals. 2005;25:33-42 [19] Sharma P, Nagar A. Inducing sensitivity on hyperspaces. Topology and its Applications. 2010;157:2052-2058 [20] Subramonian TK. Stronger forms of sensitivity for dynamical systems. Nonlinearity. 2007;20:2115-2126 [21] Klein E, Thompson A. Theory of Correspondences. New York: Wiley-Interscience; 1984 [22] Jones SH. Topical survey: Applications of the Baire category theorem. Real Analysis Exchange. 1997;23:363-394 [23] Bernardes N. References On the predictability of discrete dynamical systems II. Proceedings of American Mathematical Society. 2005;133:3473-3483 [24] Sachkov Y. Survey on controllability of invariant control systems on solvable lie groups. AMS Proceedings of Symposia in Pure Mathematics. 1999;64:297-317 [25] Jouan P. Equivalence of control systems with linear systems on lie groups and homoge- neous spaces. ESAIM: Control Optimization and Calculus of Variations. 2010;16:956-973 [26] Ayala V, San Martin L. Controllability properties of a class of control systems on lie groups. Lecture Notes in Control and Information Sciences. 2001;258:83-92 [27] Ayala V, Da Silva A, Zsigmond G. Control sets of linear systems on lie groups. Nonlinear Differential Equations and Applications. 2017;24:8, 99.1-99.15 [28] Ayala V, Da Silva A. Controllability of linear systems on Lie groups with finite semisimple center. SIAM Journal on Control and Optimization. 2017;55(2):1332-1343 [29] Ayala V, Da Silva A. Control systems on flag manifolds and their chain control sets. Discrete and Continuous Dynamical Systems. 2017;37(5):2301-2313
https://openalex.org/W2127367165	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0016378&type=printable	English	null	Prostaglandin E2 Synthesizing Enzymes in Rheumatoid Arthritis B Cells and the Effects of B Cell Depleting Therapy on Enzyme Expression	PloS one	2,011	cc-by	6,954	Abstract Introduction: B cells may play an important role in promoting immune activation in the rheumatoid synovium and can produce prostaglandin E2 (PGE2) when activated. In its turn, PGE2 formed by cyclooxygenase (COX) and microsomal prostaglandin E2 synthase 1 (MPGES1) contributes to the rheumatoid arthritis (RA) pathological process. Therapeutic depletion of B cells results in important improvement in controlling disease activity in rheumatoid patients. Therefore we investigated the expression of PGE2 pathway enzymes in RA B cells and evaluated the effects of B cell depleting therapy on their expression in RA tissue. Methods: B cells expressing MPGES1 and COX-2 were identified by flow cytometry in in vitro stimulated and control mononuclear cells isolated from synovial fluid and peripheral blood of RA patients. Synovial biopsies were obtained from 24 RA patients before and at two consecutive time points after rituximab therapy. Expression of MPGES1, COX-1 and COX-2, as well as interleukin (IL)-1b and IL-6, known inducers of MPGES1, was quantified in immunostained biopsy sections using computerized image analysis. Results: Expression of MPGES1 or COX-2 was significantly upregulated upon stimulation of B cells from blood and synovial fluid while control cells displayed no detectable enzymes. In synovial biopsy sections, the expression of MPGES1, COX-1 or COX-2 was resistant to rituximab therapy at 8 or 16 weeks after start of treatment. Furthermore expression of IL-1b in the synovial tissue remained unchanged, while IL-6 tended to decrease after therapy. Conclusions: Therapy with B cell depleting agents, although efficient in achieving good clinical and radiographic response in RA patients, leaves important inflammatory pathways in the rheumatoid synovium essentially unaffected. Citation: Gheorghe KR, Thurlings RM, Westman M, Boumans MJ, Malmstro¨m V, et al. (2011) Prostaglandin E2 Synthesizing Enzymes in Rheumatoid Arthritis B Cells and the Effects of B Cell Depleting Therapy on Enzyme Expression. PLoS ONE 6(1): e16378. doi:10.1371/journal.pone.0016378 Citation: Gheorghe KR, Thurlings RM, Westman M, Boumans MJ, Malmstro¨m V, et al. (2011) Prostaglandin E2 Synthesizing Enzymes in Rheumatoid Arthritis B Cells and the Effects of B Cell Depleting Therapy on Enzyme Expression. PLoS ONE 6(1): e16378. doi:10.1371/journal.pone.0016378 Editor: Mehrdad Matloubian, University of California San Francisco, United States of America Editor: Mehrdad Matloubian, University of California San Francisco, United States of America Received October 6, 2010; Accepted December 13, 2010; Published January 27, 2011 ceived October 6, 2010; Accepted December 13, 2010; Published January 27, 2011 Copyright: 2011 Gheorghe et al. Abstract This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2011 Gheorghe et al. This is an open-access article distributed under the terms of the Creative Commons Att unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This study was supported by The Swedish Research Council, The Swedish Rheumatism Association, King Gustaf V 80 years Foundation, The Swedish Society of Medicine, Marianne and Marcus Wallenberg foundation, Karolinska Institutet Foundation and European Commission Sixth Framework Program Autocure, the Center for Translational Molecular Medicine (TRACER), and the Dutch Arthritis Association. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors have read the journal’s policy and have the following conflicts: Profesor Tak has served as a consultant for Roche. Associate bsson is a member of the board of directors of NovaSAID. This does not alter the authors’ adherence to all the PLoS ONE policies on ls Competing Interests: The authors have read the journal’s policy and have the following conflicts: Profesor Tak has served as a consultant for Roche. Associate Professor Per-Johan Jakobsson is a member of the board of directors of NovaSAID. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials. * E-mail: per-johan.jakobsson@ki.se Prostaglandin E2 Synthesizing Enzymes in Rheumatoid Arthritis B Cells and the Effects of B Cell Depleting Therapy on Enzyme Expression Karina Roxana Gheorghe1, Rogier M. Thurlings2, Marie Westman1, Maartje J. Boumans2, Vivianne Malmstro¨ m1, Christina Trollmo1, Marina Korotkova1,3, Per-Johan Jakobsson1, Paul-Peter Tak2 1 Rheumatology Unit, Department of Medicine, Karolinska Institute, Karolinska University Hospital Solna, Stockholm, Sweden, 2 Division of Clinical Immunology and Rheumatology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands, 3 Actar AB Inc., Stockholm, Sweden 1 Rheumatology Unit, Department of Medicine, Karolinska Institute, Karolinska University Hospital Solna, Stockholm, Sweden, 2 Rheumatology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands, 3 Actar AB Inc., Stockholm, Swe PLoS ONE \| www.plosone.org ation: Gheorghe KR, Thurlings RM, Westman M, Boumans MJ, Malmstro¨m V, et al. (2011) Prostaglandin E2 Synthesizing Enzymes in ls and the Effects of B Cell Depleting Therapy on Enzyme Expression. PLoS ONE 6(1): e16378. doi:10.1371/journal.pone.0016378 January 2011 \| Volume 6 \| Issue 1 \| e16378 Treatment and clinical evaluation COX-2 derived PGE2 also plays a central role in the humoral responses, since blocking this pathway substantially decreases antibody production [17]. PGE2 regulates B cell proliferation and activation [18] as well as survival [19]. This implies a possible role for PGE2 as a mediator of B cell immune responses in RA. To investigate this hypothesis, we studied the expression of PGE2- related enzymes in SF and peripheral blood (PB)-derived B cells of RA patients. Furthermore, we hypothesised that depleting B cells could change synovial immune interactions, reduce cytokine levels and decrease disease activity in the inflamed joint. These effects can in turn affect the PGE2 biosynthetic pathway and further contribute to decline local inflammation and clinical benefit. In this sense, it has been reported that B cells are essential in sustaining PGE2 production by lung macrophages [20]. Therefore, we analysed the in vivo effects of B cell depletion by examination of serial synovial tissue biopsies obtained before and at two consecutive points after rituximab treatment. The patients were treated with two infusions of 1,000 mg rituximab (Roche, Woerden, The Netherlands) at day 1 and day 14. Methylprednisolon premedication was omitted to be able to study the specific effects of rituximab. The clinical baseline characteristics of the cohort were described previously [7]. The DAS28 was measured monthly after treatment until week 24. Response to treatment was considered according to the European League Against Rheumatism (EULAR) response criteria [23]. Responders were defined as those patients that had a good or moderate response during at least two consecutive study visits according to EULAR criteria. B Cell Depleting Therapy and PGE2 Pathway an inflammation-induced enzyme overexpressed in synovial tissue of RA patients [10]. MPGES1 is mostly found in fibroblast-like synoviocytes (FLS) and macrophages. Cyclooxygenase (COX) enzymes known as COX-1 and COX-2 are also widely expressed in the inflamed synovium. COX-1 is present in intimal lining layer and synovial sublining mononuclear cells and FLS [10,11]. COX- 2 has a similar localization, but is also highly expressed by endothelial cells [11]. Furthermore, whereas COX-1 expression is independent of the inflammatory status in the joint tissue, COX-2 is markedly upregulated at sites of inflammation [12]. Proin- flammatory cytokines present in the rheumatic milieu, such as tumor necrosis factor (TNF), interleukin (IL) 1b [13] and IL-6 [14] are prominent inducers of MPGES1. In turn, by interacting with FLS, PGE2 promotes release of IL-6 [15] and matrix metallopro- teinase-1 [16], thereby further sustaining a pathogenic circle. Patients and treatment protocol Twenty-four patients were included from a study on the mechanism of action of rituximab in RA that was reported previously [7]. Patients had active RA (Disease Activity Score evaluated in 28 joints (DAS28$3.2) [22] despite previous treatment with combination(s) of disease-modifying antirheumatic drugs (DMARDs) and/or TNF-blocking agents. During the study they were treated with stable dosages of methotrexate; stable treatment with non-steroidal anti-inflammatory drugs (NSAIDs) and prednisone (when taken orally in a dosage up to 10 milligrams) was allowed. The study was approved by the local Medical Ethical Committee of the Academic Medical Center/University of Amsterdam and all patients gave their written consent before participation in the study. Synovial biopsy and immunohistochemistry y p y y Synovial tissue was obtained using arthroscopy- guided synovial biopsy as described previously [24]. Biopsies were collected from the same affected joint before treatment as well as 4 weeks and 16 weeks after initiation of therapy. Immunohistochemical staining was performed on serial cryosections using MPGES1 antiserum, rabbit polyclonal anti-COX-1 (Cayman Chemical), mouse mono- clonal anti-COX-2 (Cayman Chemical) antibodies, mouse anti- human CD20 (L26; DakoCytomation, Glostrup, Denmark), mouse anti-human IL-6 (B-E8; Millipore Chemicon, Billerica, MA) and mouse anti-human IL-1 (2D8; Immunokontact, Abingdon, United Kingdom). The procedure has been published earlier [25]. For surface marker staining we used the following mouse monoclonal antibodies: anti-CD22 (CLB-B-Ly; Central laboratory of The Netherlands Red Cross Blood Transfusion Centre, Amsterdam) for B cells, anti-CD3 (SK7; Becton Dickinson) for T cells, anti-CD138 (B-B4; Immunotech, Marseille, France) for plasma cells and anti-CD68 (EBM11; Dako, Glostrup, Denmark) for macrophages, as previously described [7]. Out of the 24 patients included, we have analyzed biopsy sections at all three time points in 16 patients. Due to synovial sections of insufficient quality, samples from one patient taken at week 4 and from two patients at week 16 were not included in the analysis. Additionally two patients had un-assessable tissue sections in the baseline biopsy. Three patients where we could only assess biopsy at one time point were excluded from the analysis. Introduction mediators [5] thereby promoting the extent and direction of immune responses [6]. The observation that therapeutic B cell depletions using rituximab treatment disrupts synovial lymphoid neogenesis and decreases macrophages infiltration supports the notion that B cells orchestrate synovial inflammation in RA [7] Rheumatoid arthritis (RA) is a chronic autoimmune disease that features persistent synovial inflammation and proliferation along with infiltration of predominantly T lymphocytes, plasma cells and macrophages. A central role for the B lymphocytes in the pathogenesis of RA is supported by the presence of autoantibodies, which are locally produced in the inflamed synovium and may promote tissue inflammation and destruction by forming immune complexes [1]. Moreover, a significant percentage of RA patients display ectopic lymphoid structures in the synovial membrane [2,3] that could sustain T and B cell interaction [4]. Finally, effector B cells produce cytokines and other immunological notion that B cells orchestrate synovial inflammation in RA [7]. In the rheumatoid joint, the synovial fluid (SF) contains a variety of cytokines, chemokines, growth factors and lipid-derived mediators, which potentially mediate B cells effector functions. Of the prostaglandins, high levels are reached by prostaglandin E2, (PGE2) which plays a prominent role in the rheumatoid pathogenic process by promoting tissue damaging and autoim- munity [8,9]. Microsomal prostaglandin E2 synthase (MPGES) 1 catalyses its formation from cyclooxygenase-derived PGH2 and is PLoS ONE \| www.plosone.org 1 January 2011 \| Volume 6 \| Issue 1 \| e16378 January 2011 \| Volume 6 \| Issue 1 \| e16378 Materials and Methods Cell preparation and flow cytometric analysis Cell preparation and flow cytometric analysis For immunoflu- January 2011 \| Volume 6 \| Issue 1 \| e16378 PLoS ONE \| www.plosone.org 2 PLoS ONE \| www.p PLoS ONE \| B Cell Depleting Therapy and PGE2 Pathway B Cell Depleting Therapy and PGE2 Pathway orescence analysis of double stained synovial biopsy sections, samples were incubated with primary antiserum against MPGES1 and mouse monoclonal antibody against CD55 (Serotec, Oxford, UK), CD20 (Dako Cytomation) and CD138 (B-A38; Diaclone, Besancon, France), followed by addition of anti-rabbit Alexa 594 and anti-mouse Alexa 488 conjugated mouse monoclonal antibodies (both from Molecular Probes, Eugene, Oregon). orescence analysis of double stained synovial biopsy sections, samples were incubated with primary antiserum against MPGES1 and mouse monoclonal antibody against CD55 (Serotec, Oxford, UK), CD20 (Dako Cytomation) and CD138 (B-A38; Diaclone, Besancon, France), followed by addition of anti-rabbit Alexa 594 and anti-mouse Alexa 488 conjugated mouse monoclonal antibodies (both from Molecular Probes, Eugene, Oregon). Activated B cells from synovial fluid display a higher expression of MPGES1 and COX2 enzymes compared to peripheral blood RA B cells originating from SF or PB expressed minimal levels of MPGES1 or COX-2 after incubation without stimuli, but could significantly upregulate their expression after in vitro stimulation with Staphylococcus aureus Cowan strain I and pokeweed mitogen (Fig. 1 A, B). The expression was significantly higher in the B cell population from SF compared to PB (p = 0.002). Unstimulated B cells displayed no COX-2 expression before culture either in SF or PB (data not shown), while low MPGES1 was detected in B cells in some of the SF samples (median 0, range 0-10.2%). Confirming our flow cytometric data, we detected by immunofluorescence MPGES1 positive B cells in activated RA SFMC (Fig. 1D), but not in unstimulated cells (Fig. 1C). Cell preparation and flow cytometric analysis y y SF and PB mononuclear cells (MC) from 10 RA patients were collected by gradient centrifugation using Ficoll-Paque (Phar- macia, Uppsala, Sweden) and stored in liquid nitrogen until assayed. Cells were cultured in RPMI medium containing 2 mM glutamine, 100 IU/mL penicillin, 100 IU/ml strepto- mycin and 5% human serum, at 37uC in a humidified atmosphere containing 5% CO2. To activate B cells, Staphylo- coccus aureus Cowan strain I and pokeweed mitogen (both from Sigma-Aldrich, Sweden) were added to the cultures at a final concentration of 10 mg/ml and 0.1 mg/ml, respectively, and cells were incubated for 16 hours. Detection of intracellular enzymes by flow cytometry was performed both in incubated cells and in unstimulated cells before culture, by using anti- MPGES1 and COX-2 antibodies according to previously described protocol [10]. Briefly, SFMC or PBMC were washed in 5% human serum, followed by staining of surface marker with anti-CD19-PerCP mouse monoclonal antibody (Becton Dickinson, San Jose, CA). Subsequently cells were fixed in 4% paraformaldehyde and incubated with rabbit polyclonal antise- rum raised against MPGES1 [21] and mouse monoclonal anti- COX-2 (CX229; Cayman Chemical, Ann Harbor, MI) antibodies in saponin containing phosphate buffer, followed by addition of secondary FITC-coupled anti-rabbit or anti-mouse antibodies. B cells were identified based on scatter properties and CD19 expression and were then analyzed for expression of MPGES1 and COX-2. Due to technical issues in using COX-1 antibody for flow cytometry, we did not further pursue analysis of COX-1 in these cells. Stained sections were quantitatively evaluated by computer- assisted image analysis on a Polyvar II microscope and expressed as percentage of positive staining versus total counterstained area. IL-1 and IL-6 staining was quantified as the integrated optical density, using a Leica DM-RXA light microscope [26]. Double immunofluorescence was performed on SFMC cultured on chamber slides with or without activation stimuli, using rabbit MPGES1 antiserum and a mixture of mouse monoclonal anti-CD19 and anti-CD20 antibodies (Dako Cyto- mation), using a protocol published earlier [21]. B Cell Depleting Therapy and PGE2 Pathway MPGES1 expression does not co-localize with B cells in rheumatoid synovium aggregates did not display any specific COX-2 expression. In some patients we could however detect COX-1 expressing cells within the lymphoid inflammatory infiltrates, possibly coinciding with localization of CD20 positive B cells. Both chromogenic and double fluorescence staining of synovial biopsies from several RA patients revealed that MPGES1 positive cells and B cells were not co-localized in the same tissue areas. As illustrated in Fig. 2, CD20 positive B cells and CD138 plasma cells have different areas of distribution compared to MPGES1 expressing cells, with virtually no overlapping. Indeed, B cells mostly accumulate inside lymphoid structures in synovial sections that display follicular synovitis, while MPGES1 positive cells can be found in intimal lining layer and synovial sublining regions, in agreement with our previous study [21]. Also, in RA biopsies not showing ectopic lymphoid neogenesis, the pattern of detected B cells is variable, from scattered isolated cells to extremely scarce (data not shown), whereas strong MPGES1 expression is virtually the rule. Clinical response to rituximab treatment The DAS28 did not decrease yet at 4 weeks, but there was a statistically significant decrease in DAS28 from week 8 with a maximum mean decrease at week 20 (compared to baseline: mean (SD) decrease of 0.96 (1.1) at week 8 and 1.85 (1.4) at week 20; both p,0.001). Of the 24 patients studied, 4 patients had a good response according to EULAR response criteria and 15 patients had a moderate response, while 5 patients did not fulfil response criteria [7]. Statistical analysis doi:10.1371/journal.pone.0016378.g001 January 2011 \| Volume 6 \| Issue 1 \| e16378 PLoS ONE \| www.plosone.org 3 B Cell Depleting Therapy and PGE2 Pathway Statistical analysis For comparison of paired synovial biopsy samples, Wilcoxon- matched paired test for nonparametric data was used. Statistical analysis of flow cytometry data was performed using Mann- Whitney test or Wilcoxon test for paired samples and Bonferroni corrections were applied for multiple comparisons. Spearman rank correlation test for non-parametrical samples was applied for correlation analysis. Figure 1. Expression of MPGES1 and COX-2 in synovial fluid (SF) and peripheral blood (PB)-derived B cells. In rheumatoid arthritis patients, SF B cells stimulated in vitro expressed higher levels of MPGES1 and COX-2 than PB B cells. A, Percentage of B cells that are MPGES1 positive in unstimulated and stimulated SFMC and PBMC. B, Percentage of B cells that are COX-2 positive in unstimulated and stimulated SFMC and PBMC. * = P,0.01, * = P,0.001. C, D Immunofluorescence staining of unstimulated (C) and activated (D) RA synovial fluid cells showing CD19 positive B cells (red), MPGES1 positive cells (green) and double stained cells (yellow). Original magnification 250x. doi:10.1371/journal.pone.0016378.g001 Figure 1. Expression of MPGES1 and COX-2 in synovial fluid (SF) and peripheral blood (PB)-derived B cells. In rheumatoid arthritis patients, SF B cells stimulated in vitro expressed higher levels of MPGES1 and COX-2 than PB B cells. A, Percentage of B cells that are MPGES1 positive in unstimulated and stimulated SFMC and PBMC. B, Percentage of B cells that are COX-2 positive in unstimulated and stimulated SFMC and PBMC. = P,0.01, * = P,0.001. C, D Immunofluorescence staining of unstimulated (C) and activated (D) RA synovial fluid cells showing CD19 positive B cells (red), MPGES1 positive cells (green) and double stained cells (yellow). Original magnification 250x. doi:10.1371/journal.pone.0016378.g001 Figure 1. Expression of MPGES1 and COX-2 in synovial fluid (SF) and peripheral blood (PB)-derived B cells. In rheumatoid arthritis patients, SF B cells stimulated in vitro expressed higher levels of MPGES1 and COX-2 than PB B cells. A, Percentage of B cells that are MPGES1 positive in unstimulated and stimulated SFMC and PBMC. B, Percentage of B cells that are COX-2 positive in unstimulated and stimulated SFMC and PBMC. = P,0.01, *** = P,0.001. C, D Immunofluorescence staining of unstimulated (C) and activated (D) RA synovial fluid cells showing CD19 positive B cells (red), MPGES1 positive cells (green) and double stained cells (yellow). Original magnification 250x. B Cell Depleting Therapy and PGE2 Pathway weeks or 16 weeks after therapy initiation compared to baseline (Fig. 3). synovial enzyme expression was in our study not related to the clinical outcome or the inflammatory cell load in the synovial tissue. Subsequently we analyzed the cells expressing MPGES1 before and after treatment and showed that the enzyme is still present in CD55 positive FLS 16 weeks after rituximab treatment start (Fig. 5). On an individual response level, 7 patients showed an initial decrease in MPGES1 positive staining at 4 weeks, followed by a subsequent increase at 16 weeks, while 4 patients followed a different pattern with enhanced expression at 4 weeks and a decrease thereafter. In most of the patients we observed a large variability in the synovial MPGES1 response to rituximab between the different time points. A trend towards decreased COX-2 expression at 4 weeks was observed, albeit non-significant (P = 0.09) which was similarly maintained at 16 weeks (P = 0.1). COX-1 expression displayed however variable changes over time in individual patients, with no identifiable trend. Analysis of subgroups of patients not taking NSAID or GC medication showed no significant differences between groups and no consistent change in enzyme expression over time. B cell depletion has limited effect on MPGES1-inducing cytokines The absent effect of rituximab on MPGES1 expression suggests a similar limited effect on the expression of MPGES1 inducing cytokines. Therefore we analyzed the change in synovial IL-1b and IL-6 expression after treatment. Before treatment IL-6 and IL-1b were present on cellular membranes and diffusely in the synovial extracellular matrix. Rituximab treatment induced a trend towards a decrease in IL-6 (P = 0.062) expression at week 16 but not in IL-1b (P = 0.12) (Fig. 6). Next, we investigated if the synovial expression of these enzymes followed a specific pattern in responders versus non-responder patients and if the change in enzyme expression at 4 or 16 weeks is correlated with the change in inflammatory cell numbers in the tissue. As shown in Fig. 4, no specific response pattern could be observed in either of the groups, and no statistical significant difference was detected between responders and non-responders at baseline or at any of the later time points studied. Furthermore, the change in cell surface marker expression for B cells, T cells, plasma cells and lining and sublining macrophages did not correlate with the difference in enzyme expression at 4 weeks or 16 weeks following therapy start (data not shown). Thus the variability in the B cell depletion therapy has limited effects on expression of MPGES1 and COX enzymes in RA synovium B cell depletion therapy has limited effects on expression of MPGES1 and COX enzymes in RA synovium Expression of COX enzymes was detected in intimal and sublining macrophages as well as in FLS (Fig. 2 B, C), as previously described [10]. In addition, cells surrounding vessels were positively stained for COX-2, while inflammatory lymphoid Analysis of MPGES1, COX-1 and COX-2 in synovial tissue showed no statistically significant change in their expression at 4 Figure 2. MPGES1 is not expressed in B cell-rich areas in RA synovium. Frozen sections of synovial tissue showing diaminobenzidine (brown) staining of MPGES1 (A), COX-2 (B), COX-1 (C) and CD20 B cells (D), CD138 plasma cells (E) (hematoxiline counterstained). Insets represent higher power of image A and D. Original magnification 80x and 200x (insets). (F) and (G) Immunofluorescence pictures of double stained MPGES1 (red) and CD20+ B cells (F) and CD138+ plasma cells (G) (green). Magnification 400x. doi:10.1371/journal.pone.0016378.g002 Figure 2. MPGES1 is not expressed in B cell-rich areas in RA synovium. Frozen sections of synovial tissue showing diaminobenzidine (brown) staining of MPGES1 (A), COX-2 (B), COX-1 (C) and CD20 B cells (D), CD138 plasma cells (E) (hematoxiline counterstained). Insets represent higher power of image A and D. Original magnification 80x and 200x (insets). (F) and (G) Immunofluorescence pictures of double stained MPGES1 (red) and CD20+ B cells (F) and CD138+ plasma cells (G) (green). Magnification 400x. doi:10.1371/journal.pone.0016378.g002 PLoS ONE \| www.plosone.org January 2011 \| Volume 6 \| Issue 1 \| e16378 January 2011 \| Volume 6 \| Issue 1 \| e16378 4 B Cell Depleting Therapy and PGE2 Pathway Discussion In this study we demonstrate that MPGES1 and COX-2 enzymes are upregulated in SF B cells upon activation while synovial tissue B cells do not express these enzymes. Furthermore, after B cell depletion therapy MPGES1, COX-1 and COX-2 levels in synovium of RA patients are essentially unaffected 4 weeks or 16 weeks after therapy despite clinical improvement in the majority of the studied patients. Also IL-1b and IL-6, strong inducers of MPGES1, did not change significantly. Figure 3. Minimal influence of rituximab treatment on expression of MPGES1, COX-1 and COX-2 in synovial tissue. Immunohistochemical staining of frozen biopsy sections from rheumatoid arthritis patients shows diaminobenzidine staining (brown) of MPGES1 (A–C), COX-2 (E-G) and COX-1 (I–K) before treatment, 4 weeks and 16 weeks after treatment (hematoxiline counterstained). Graphs depict image analysis of MPGES1 (D), COX-2 (H) and COX-1 (L) expression in stained sections from patients’ biopsies at the different time points. Original magnification 150x. doi:10.1371/journal.pone.0016378.g003 Figure 3. Minimal influence of rituximab treatment on expression of MPGES1, COX-1 and COX-2 in synovial tissue. Immunohistochemical staining of frozen biopsy sections from rheumatoid arthritis patients shows diaminobenzidine staining (brown) of MPGES1 (A–C), COX-2 (E-G) and COX-1 (I–K) before treatment, 4 weeks and 16 weeks after treatment (hematoxiline counterstained). Graphs depict image analysis of MPGES1 (D), COX-2 (H) and COX-1 (L) expression in stained sections from patients’ biopsies at the different time points. Original magnification 150x. doi:10.1371/journal.pone.0016378.g003 January 2011 \| Volume 6 \| Issue 1 \| e16378 PLoS ONE \| www.plosone.org 5 B Cell Depleting Therapy and PGE2 Pathway Figure 4. The expression of the PGE2 biosynthetic enzymes does not differ between responder and non-responder patients. Graph display immunohistochemical analysis of stained biopsy sections for MPGES1, COX-1 and COX-2 expression in individual responder/non-responde patients at baseline and at 4 weeks and 16 weeks after initiation of rituximab treatment. doi:10.1371/journal.pone.0016378.g004 Figure 4. The expression of the PGE2 biosynthetic enzymes does not differ between responder and non-responder patients. Graphs display immunohistochemical analysis of stained biopsy sections for MPGES1, COX-1 and COX-2 expression in individual responder/non-responder patients at baseline and at 4 weeks and 16 weeks after initiation of rituximab treatment. doi:10.1371/journal.pone.0016378.g004 In a previous study COX-2 was upregulated in activated PB B cells of healthy volunteers [27]. We found the same effect in SF and PB B cells from RA patients. In the synovial tissue we observed MPGES1 expression in intimal macrophages, scattered sublining macrophages and FLS. January 2011 \| Volume 6 \| Issue 1 \| e16378 Discussion A similar distribution was observed for COX-2 with additionally positive staining around vessels. This is in line with a previous study describing MPGES1 localization in synovial tissue [21]. Indeed cells expressing MPGES1 in the synovium are mostly fibroblasts and macrophag- es. Also, we showed here that in RA synovial tissue MPGES1 is not present in B cell-rich areas such as the lymphoid aggregates. Despite failing to detect MPGES1 positive B cells in the rheumatoid tissue, we showed that SF and PB B cells from RA patients are able of upregulating MPGES1 and COX-2 upon in vitro activation while unstimulated B cells do not readily express these enzymes. These differences between in vitro activated SF B cells and synovial tissue B cells suggest that activation signals that are required to upregulate MPGES1 in B cells are not present in RA synovium. Moreover, we report here that activated B cells from SF have a higher expression level of MPGES1 and COX-2 than those from PB, suggesting that the local synovial compart- ment has a different repertoire of immune cells or a particular PLoS ONE \| www.plosone.org January 2011 \| Volume 6 \| Issue 1 \| e16378 6 B Cell Depleting Therapy and PGE2 Pathway Figure 5. MPGES1 expression in synovial lining fibroblasts before and 16 weeks after initiation of rituximab therapy. Double immunofluorescence pictures show the presence of MPGES1 (red) expression in CD55 positive fibroblasts (green) in the rheumatoid tissue before rituximab initiation (A) and 16 weeks later (B). Original magnification 500x. Arrows point to double stained cells. doi:10.1371/journal.pone.0016378.g005 Figure 5. MPGES1 expression in synovial lining fibroblasts before and 16 weeks after initiation of rituximab therapy. Double immunofluorescence pictures show the presence of MPGES1 (red) expression in CD55 positive fibroblasts (green) in the rheumatoid tissue before rituximab initiation (A) and 16 weeks later (B). Original magnification 500x. Arrows point to double stained cells. doi:10.1371/journal.pone.0016378.g005 activation state. In fact, the proportion of B cell subsets, such as memory or naive B cells may vary between PB and SF with some phenotypes being prone to express these enzymes more than others. As such, the difference in enzyme expression that we detected may simply reflect the different abilities of the B cell subsets to respond to activation stimuli. macrophages and more heterogeneously, plasma cells [7]. In line with these changes, the inducible prostaglandin synthesis could have been affected. Discussion Of importance, our study showed that although achieving clinical improvement in a large percentage of the patients studied, rituximab did not change the local expression of MPGES1 and COX. Moreover, the variation in enzyme expression between the different time points did not reflect the change in synovial inflammatory cell populations, such as B cells, T cells, plasma cells and macrophages. In line, we found no clear cut decrease in the local expression of IL-1b and IL-6, even though these are produced by B cells, T cells and macrophages. Similarly, we have previously reported that anti-TNF agents do not suppress expression of MPGES1 or COX-2 in the rheumatoid synovium [10]. Taken together, these data indicate that important inflammatory pathways are relatively unaffected despite rituximab mediated B-cell depletion and indirect decrease in other inflammatory cells. Possibly, rituximab exerts relatively little effects on activation of the more resident cell populations in the We showed that B cell depleting therapy exerts little effect on the PGE2 pathway enzymes, in agreement with our observation that B cells do not express these enzymes in the synovial tissue. However, B cells are important contributors to the inflammatory milieu in RA also by virtue of their capacity to activate T cells and secrete cytokines [28]. Since COX-2 and MPGES1 expression is inflammation-induced, a reduction of the B cell load in the rheumatoid tissue could in theory be followed by a decrease in antibody and cytokine formation and a reduced interaction with other immune cells, leading to a decrease in their activity and infiltration. Indeed a recent study of the same cohort showed that rituximab induces a decrease in the number of synovial T cells, Figure 6. Rituximab treatment exerts limited effects on the synovial tissue expression of IL-1b and IL-6. Graphs show image analysis of positive stained sections for (A) IL-1b and (B) IL-6 before and at consecutive time points after initiation of rituximab therapy. At week 16 a trend towards decrease occurred in IL-6 but not in IL-1b. doi:10.1371/journal.pone.0016378.g006 Figure 6. Rituximab treatment exerts limited effects on the synovial tissue expression of IL-1b and IL-6. Graphs show image analysis of positive stained sections for (A) IL-1b and (B) IL-6 before and at consecutive time points after initiation of rituximab therapy. At week 16 a trend towards decrease occurred in IL-6 but not in IL-1b. B Cell Depleting Therapy and PGE2 Pathway B Cell Depleting Therapy and PGE2 Pathway synovium, like FLS, dendritic cells, mast cells and CD163+ macrophages. It should be noted that we cannot completely exclude the possibility of a delayed effect of rituximab on MPGES1 and COX enzymes that may become evident after 16 weeks of treatment. Furthermore, it is noteworthy that PGE2 is capable of upregulating its own formation in an autocrine manner [33,34], and thus local MPGES1 expressing cells can provide a positive feedback. Here we showed that expression of PGE2 related enzymes is most likely not a simple result of the local number of inflammatory cells, but of the interplay of mediators. Furthermore, it is noteworthy that PGE2 is capable of upregulating its own formation in an autocrine manner [33,34], and thus local MPGES1 expressing cells can provide a positive feedback. Here we showed that expression of PGE2 related enzymes is most likely not a simple result of the local number of inflammatory cells, but of the interplay of mediators. Despite almost complete B cell depletion in the periphery, persistence of synovial B cells is seen in a subset of patients [29], which also correlates with infiltration with other inflammatory cells [7]. While the decrease in synovial plasma cells can predict the response to B cell depleting agents [7], persistence of plasma cell infiltration is associated with residual synovial inflammation [30]. It is hypothesized that the synovial milieu harbours molecules able to rescue B cells and promote their survival [29]. In this sense, it is worth mentioning the ability of PGE2 to promote survival pathways and support viability of B cells [31]. Thus the persistence of an active PGE2 pathway despite rituximab treatment may contribute to later relapse. In conclusion, we demonstrated in this study that rituximab therapy has minimal influence on synovial expression of enzymes involved in the PGE2 pathway, despite clinical response in most RA patients. Furthermore, recent reports have demonstrated that optimal antibody production by B cells needs COX-2-derived PGE2 [17], implying that blocking this pathway may possibly lead to reduced antibody formation in synovial tissue and thus less joint damage. References (2006) B lymphocytes are critical for lung fibrosis control and prostaglandin E2 regulation in IL-9 transgenic mice. Am J Respir Cell Mol Biol 34: 573–580. 8. Sheibanie AF, Khayrullina T, Safadi FF, Ganea D (2007) Prostaglandin E2 exacerbates collagen-induced arthritis in mice through the inflammatory interleukin-23/interleukin-17 axis. Arthritis Rheum 56: 2608–2619. 21. Westman M, Korotkova M, af Klint E, Stark A, Audoly LP, et al. (2004) Expression of microsomal prostaglandin E synthase 1 in rheumatoid arthritis synovium. Arthritis Rheum 50: 1774–1780. 9. Yao C, Sakata D, Esaki Y, Li Y, Matsuoka T, et al. (2009) Prostaglandin E2-EP4 signaling promotes immune inflammation through Th1 cell differentiation and Th17 cell expansion. Nat Med 15: 633–640. 22. Prevoo ML, van ’t Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, et al. (1995) Modified disease activity scores that include twenty-eight-joint counts. Development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 38: 44–48. p 10. Korotkova M, Westman M, Gheorghe KR, af Klint E, Trollmo C, et al. (2005) Effects of antirheumatic treatments on the prostaglandin E2 biosynthetic pathway. Arthritis Rheum 52: 3439–3447. 23. van Gestel AM, Haagsma CJ, van Riel PL (1998) Validation of rheumatoid arthritis improvement criteria that include simplified joint counts. Arthritis Rheum 41: 1845–1850. p y 11. Siegle I, Klein T, Backman JT, Saal JG, Nusing RM, et al. (1998) Expression of cyclooxygenase 1 and cyclooxygenase 2 in human synovial tissue: differential elevation of cyclooxygenase 2 in inflammatory joint diseases. Arthritis Rheum 41: 122–129. 24. Kraan MC, Reece RJ, Smeets TJ, Veale DJ, Emery P, et al. (2002) Comparison of synovial tissues from the knee joints and the small joints of rheumatoid arthritis patients: Implications for pathogenesis and evaluation of treatment. Arthritis Rheum 46: 2034–2038. 12. Sano H, Hla T, Maier JA, Crofford LJ, Case JP, et al. (1992) In vivo cyclooxygenase expression in synovial tissues of patients with rheumatoid arthritis and osteoarthritis and rats with adjuvant and streptococcal cell wall arthritis. J Clin Invest 89: 97–108. 25. Ulfgren AK, Lindblad S, Klareskog L, Andersson J, Andersson U (1995) Detection of cytokine producing cells in the synovial membrane from patients with rheumatoid arthritis. Ann Rheum Dis 54: 654–661. 13. Stichtenoth DO, Thoren S, Bian H, Peters-Golden M, Jakobsson PJ, et al. (2001) Microsomal prostaglandin E synthase is regulated by proinflammatory cytokines and glucocorticoids in primary rheumatoid synovial cells. J Immunol 167: 469–474. References 1. Steiner G, Smolen J (2002) Autoantibodies in rheumatoid arthritis and their clinical significance. Arthritis Res 4 Suppl 2: S1–5. 15. Inoue H, Takamori M, Shimoyama Y, Ishibashi H, Yamamoto S, et al. (2002) Regulation by PGE2 of the production of interleukin-6, macrophage colony stimulating factor, and vascular endothelial growth factor in human synovial fibroblasts. Br J Pharmacol 136: 287–295. 1. Steiner G, Smolen J (2002) Autoantibodies in rheumatoid arthritis and their clinical significance. Arthritis Res 4 Suppl 2: S1–5. 2. Cantaert T, Kolln J, Timmer T, van der Pouw Kraan TC, Vandooren B, et al. (2008) B lymphocyte autoimmunity in rheumatoid synovitis is independent of ectopic lymphoid neogenesis. J Immunol 181: 785–794. 16. Kunisch E, Jansen A, Kojima F, Loffler I, Kapoor M, et al. (2009) Prostaglandin E2 differentially modulates proinflammatory/prodestructive effects of TNF- alpha on synovial fibroblasts via specific E prostanoid receptors/cAMP. J Immunol 183: 1328–1336. 3. Klimiuk PA, Goronzy JJ, Bjor nsson J, Beckenbaugh RD, Weyand CM (1997) Tissue cytokine patterns distinguish variants of rheumatoid synovitis. Am J Pathol 151: 1311–1319. 4. Takemura S, Klimiuk PA, Braun A, Goronzy JJ, Weyand CM (2001) T cell activation in rheumatoid synovium is B cell dependent. J Immunol 167: 4710–4718. 17. Ryan EP, Pollock SJ, Murant TI, Bernstein SH, Felgar RE, et al. (2005) Activated human B lymphocytes express cyclooxygenase-2 and cyclooxygenase inhibitors attenuate antibody production. J Immunol 174: 2619–2626. 5. Lund FE, Randall TD. Effector and regulatory B cells: modulators of CD4(+) T cell immunity. Nat Rev Immunol 10: 236–247. 18. Murn J, Alibert O, Wu N, Tendil S, Gidrol X (2008) Prostaglandin E2 regulates B cell proliferation through a candidate tumor suppressor, Ptger4. J Exp Med 205: 3091–3103. 6. Dorner T, Kinnman N, Tak PP. Targeting B cells in immune-mediated inflammatory disease: A comprehensive review of mechanisms of action and identification of biomarkers. Pharmacol Ther 125: 464–475. 19. Mongini PK, Inman JK, Han H, Fattah RJ, Abramson SB, et al. (2006) APRIL and BAFF promote increased viability of replicating human B2 cells via mechanism involving cyclooxygenase 2. J Immunol 176: 6736–6751. 7. Thurlings RM, Vos K, Wijbrandts CA, Zwinderman AH, Gerlag DM, et al. (2008) Synovial tissue response to rituximab: mechanism of action and identification of biomarkers of response. Ann Rheum Dis 67: 917–925. 20. Arras M, Louahed J, Simoen V, Barbarin V, Misson P, et al. B Cell Depleting Therapy and PGE2 Pathway Together, these data may provide one explanation for the fact that rituximab treatment, similar to TNF blockade, does not induce complete remission in the majority of the RA patients and suggest that blocking the PGE2 pathway by targeting MPGES1 may lead to novel therapeutical strategies and complement current anti-rheumatic therapy by providing additional benefit in controlling the inflammatory process in the rheumatoid joint. The majority of the patients included in this study received concomitant medication with NSAIDs, which can decrease the formation of PGE2 in synovial fluid, albeit not completely, and even affect COX-2 production, as seen in ostheoarthritis [32]. Despite representing an inherent confounding factor in such clinical studies, NSAIDs do not alter MPGES1 expression. In addition, once this medication is discontinued, COX activity may resume and account, together with MPGES1, for PGE2 production. Author Contributions Conceived and designed the experiments: PJJ PPT. Performed the experiments: KRG RMT MW MJB. Analyzed the data: KRG MW MJB MK VM CT PJJ PPT. Contributed reagents/materials/analysis tools: RMT. Wrote the paper: KRG RMT MW MJB VM CT MK PJJ PPT. In some patients it is possible that the remaining synovial B cells may enhance PGE2 pathway in local fibroblasts and macrophages. Discussion doi:10.1371/journal.pone.0016378.g006 January 2011 \| Volume 6 \| Issue 1 \| e16378 PLoS ONE \| www.plosone.org 7 7 B Cell Depleting Therapy and PGE2 Pathway synovial membrane of patients with osteoarthritis: differential proinflammatory cytokine profile between celecoxib and aceclofenac. Ann Rheum Dis 65: 998–1005. 32. Alvarez-Soria MA, Largo R, Santillana J, Sanchez-Pernaute O, Calvo E, et al. (2006) Long term NSAID treatment inhibits COX-2 synthesis in the knee 31. Lee H, Trott JS, Haque S, McCormick S, Chiorazzi N, et al. (2010) A Cyclooxygenase-2/Prostaglandin E2 Pathway Augments Activation-Induced Cytosine Deaminase Expression within Replicating Human B Cells. J Immunol. 33. Hinz B, Brune K, Pahl A (2000) Cyclooxygenase-2 expression in lipopolysac- charide-stimulated human monocytes is modulated by cyclic AMP, prostaglan- din E(2), and nonsteroidal anti-inflammatory drugs. Biochem Biophys Res Commun 278: 790–796. 30. Teng YK, Levarht EW, Toes RE, Huizinga TW, van Laar JM (2009) Residual inflammation after rituximab treatment is associated with sustained synovial plasma cell infiltration and enhanced B cell repopulation. Ann Rheum Dis 68: 1011–1016. 28. Duddy ME, Alter A, Bar-Or A (2004) Distinct profiles of human B cell effector cytokines: a role in immune regulation? J Immunol 172: 3422–3427. 29. Jacobi AM, Dorner T (2010) Current aspects of anti-CD20 therapy in rheumatoid arthritis. Curr Opin Pharmacol 10: 316–321. References 26. Haringman JJ, Vinkenoog M, Gerlag DM, Smeets TJ, Zwinderman AH, et al. (2005) Reliability of computerized image analysis for the evaluation of serial synovial biopsies in randomized controlled trials in rheumatoid arthritis. Arthritis Res Ther 7: R862–867. 14. Uematsu S, Matsumoto M, Takeda K, Akira S (2002) Lipopolysaccharide- dependent prostaglandin E(2) production is regulated by the glutathione- dependent prostaglandin E(2) synthase gene induced by the Toll-like receptor 4/ MyD88/NF-IL6 pathway. J Immunol 168: 5811–5816. 27. Bernard MP, Phipps RP (2007) CpG oligodeoxynucleotides induce cyclooxy- genase-2 in human B lymphocytes: implications for adjuvant activity and antibody production. Clin Immunol 125: 138–148. PLoS ONE \| www.plosone.org January 2011 \| Volume 6 \| Issue 1 \| e16378 January 2011 \| Volume 6 \| Issue 1 \| e16378 8 January 2011 \| Volume 6 \| Issue 1 \| e16378 synovial membrane of patients with osteoarthritis: differential proinflammatory cytokine profile between celecoxib and aceclofenac. Ann Rheum Dis 65: 998–1005. PLoS ONE \| www.plosone.org B Cell Depleting Therapy and PGE2 Pathway B Cell Depleting Therapy and PGE2 Pathway 28. Duddy ME, Alter A, Bar-Or A (2004) Distinct profiles of human B cell effector cytokines: a role in immune regulation? J Immunol 172: 3422–3427. y g J 29. Jacobi AM, Dorner T (2010) Current aspects of anti-CD20 therapy in rheumatoid arthritis. Curr Opin Pharmacol 10: 316–321. 33. Hinz B, Brune K, Pahl A (2000) Cyclooxygenase-2 expression in lipopolysac- charide-stimulated human monocytes is modulated by cyclic AMP, prostaglan- din E(2), and nonsteroidal anti-inflammatory drugs. Biochem Biophys Res Commun 278: 790–796. p 30. Teng YK, Levarht EW, Toes RE, Huizinga TW, van Laar JM (2009) Residual inflammation after rituximab treatment is associated with sustained synovial plasma cell infiltration and enhanced B cell repopulation. Ann Rheum Dis 68: 1011–1016. 34. Pino MS, Nawrocki ST, Cognetti F, Abruzzese JL, Xiong HQ, et al. (2005) Prostaglandin E2 drives cyclooxygenase-2 expression via cyclic AMP response element activation in human pancreatic cancer cells. Cancer Biol Ther 4: 1263–1269. 31. Lee H, Trott JS, Haque S, McCormick S, Chiorazzi N, et al. (2010) A Cyclooxygenase-2/Prostaglandin E2 Pathway Augments Activation-Induced Cytosine Deaminase Expression within Replicating Human B Cells. J Immunol. 32. Alvarez-Soria MA, Largo R, Santillana J, Sanchez-Pernaute O, Calvo E, et al. (2006) Long term NSAID treatment inhibits COX-2 synthesis in the knee PLoS ONE \| www.plosone.org January 2011 \| Volume 6 \| Issue 1 \| e16378 9
https://openalex.org/W2891465278	https://publications.goettingen-research-online.de/bitstream/2/68943/1/document.pdf	English	null	Frustrated endocytosis controls contractility-independent mechanotransduction at clathrin-coated structures	Nature communications	2,018	cc-by	15,439	ARTICLE Frustrated endocytosis controls contractility- independent mechanotransduction at clathrin- coated structures Francesco Baschieri 1, Stéphane Dayot1,5, Nadia Elkhatib1, Nathalie Ly1, Anahi Capmany2, Kristine Schauer2, Timo Betz3, Danijela Matic Vignjevic 2, Renaud Poincloux 4 & Guillaume Montagnac1 Francesco Baschieri 1, Stéphane Dayot1,5, Nadia Elkhatib1, Nathalie Ly1, Anahi Capmany2, Kristine Schauer2, Timo Betz3, Danijela Matic Vignjevic 2, Renaud Poincloux 4 & Guillaume Montagnac1 It is generally assumed that cells interrogate the mechanical properties of their environment by pushing and pulling on the extracellular matrix (ECM). For instance, acto-myosin- dependent contraction forces exerted at focal adhesions (FAs) allow the cell to actively probe substrate elasticity. Here, we report that a subset of long-lived and ﬂat clathrin-coated structures (CCSs), also termed plaques, are contractility-independent mechanosensitive signaling platforms. We observed that plaques assemble in response to increasing substrate rigidity and that this is independent of FAs, actin and myosin-II activity. We show that plaque assembly depends on αvβ5 integrin, and is a consequence of frustrated endocytosis whereby αvβ5 tightly engaged with the stiff substrate locally stalls CCS dynamics. We also report that plaques serve as platforms for receptor-dependent signaling and are required for increased Erk activation and cell proliferation on stiff environments. We conclude that CCSs are mechanotransduction structures that sense substrate rigidity independently of cell contractility. 1 Inserm U1170, Gustave Roussy Institute, Université Paris-Saclay, Villejuif, France. 2 Institut Curie, CNRS UMR144, PSL Research University, Centre Universitaire, Paris, France. 3 Institute of Cell Biology, Center of Molecular Biology of Inﬂammation, Cells-in-Motion Cluster of Excellence, University of Münster, Münster, Germany. 4 Institut de Pharmacologie et Biologie Structurale, IPBS, Université de Toulouse, CNRS, UPS, Toulouse, France. 5Present address: Institut Curie, Inserm U830, PSL Research University, Centre Universitaire, Paris, France. Correspondence and requests for materials should be addressed to F.B. (email: francesco.baschieri@gustaveroussy.fr) or to G.M. (email: guillaume.montagnac@gustaveroussy.fr) 1 NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y were large and long-lived when cells were grown on glass, reﬂecting the mostly static nature of clathrin-coated plaques (Fig. 1b, c, Supplementary Fig. 2a and Supplementary Movie 1). Similar results were obtained when cells were seeded on 31 kPa gels (Fig. 1b, c, Supplementary Fig. 2a and Supplementary Movie 1). However, the proportion of long-lived CCSs dramati- cally dropped in HeLa cells cultured on softer gels (0.1 and 5 kPa; Fig. 1b, c, Supplementary Fig. 2a and Supplementary Movie 1). Frustrated endocytosis controls contractility- independent mechanotransduction at clathrin- coated structures Sur- prisingly, plaque formation on stiff environments is independent of cell contractility but is the consequence of a frustrated endo- cytosis process whereby αvβ5-integrin prevents CCSs budding by anchoring the structure to the substrate. We further report that receptor clustering at clathrin-coated plaques potentiates intracellular signaling and increases cell proliferation. In sum- mary, we propose that clathrin-coated plaques are mechan- osensitive structures instructing the cell about the rigidity of its environment. Substrate rigidity sensing is described as an active process involving actin-dependent contraction forces transmitted to the ECM at FAs4–6. Thus, we tested whether acto-myosin-generated forces and FAs are required for plaques formation. Inhibiting myosin-II activity or actin polymerization with blebbistatin or cytochalasin D respectively, did not prevent the formation of large and static CCSs in HeLa cells grown on glass (Fig. 1d, e) and large, ﬂat clathrin lattices were still detected by scanning electron microscopy in cells treated with blebbistatin (Fig. 1f). In addition, interfering with FAs assembly using Talin1-speciﬁc siRNAs did not prevent but rather increased the formation of long-lived, large CCSs (Supplementary Fig. 2h–j). These results demonstrate that plaque formation does not rely on cell contractility and suggest that plaques represent a new type of mechanosensitive structures. Integrin αvβ5 is required for plaque assembly. In the classical mechanosensation model, cell-generated forces are transmitted through integrins binding the ECM at FAs. Although they assemble in a contractility-independent manner, plaques have been proposed to adhere to the substratum16,17. Integrin β5 was reported to be highly enriched at plaques10,18. β5 interacts with αv-integrin to form the high-afﬁnity vitronectin receptor19,20. We observed that αvβ5 strongly colocalized with CCSs at the ventral plasma membrane of HeLa cells cultured on glass (Fig. 2a). Close inspection revealed that αvβ5 was highly enriched in plaques as compared to CCPs (Fig. 2a, b). The degree of association between αvβ5 and AP-2-marked CCSs signiﬁcantly decreased on softer gels (Fig. 2c and Supplementary Fig. 3a). We next tested whether αvβ5 was required for plaque formation on hard surfaces. Knockdown of either αv or β5 integrins using speciﬁc siRNAs resulted in a complete loss of large and static CCSs (Fig. 2d–f, Supplementary Fig. 4a–d and Supplementary Movie 2). We did not observed any effect of β1 or β3 integrins depletion on CCS dynamics (Supplementary Fig. 4e, f). Inhibiting β5 in HepG2 cells also precluded the formation of large CCSs (Supplementary Fig. 4g). Frustrated endocytosis controls contractility- independent mechanotransduction at clathrin- coated structures Similar to HeLa cells, HepG2 and Caco-2 cell lines harbored both dynamic CCPs and large, static CCSs when cultured on glass or on 31 kPa gels while only short-lived CCPs were detectable on softer gels (Supplementary Fig. 2b–e). However, we did not measure any elasticity-dependent modulation of CCSs dynamics in MDA-MB-231 cell line that was previously reported to only harbor CCPs15 (Supplementary Fig. 2f, g). Together, our results demonstrate that clathrin-coated plaques are mechanosensitive structures that some cell types assemble in response to stiff substrates. C ells constantly probe the extracellular milieu in order to adapt to the changing conditions of the environment. Besides chemical signals sensed by speciﬁc receptors, cells also respond to mechanical stimuli with important consequences for cell migration, proliferation and differentiation1–3. It is gen- erally accepted that cells probe mechanical features of the micro- environment by applying forces on it4–6. Contractile forces gen- erated by the acto-myosin network and transmitted to the sub- strate at integrin-rich cell adhesions endow these adhesions to grow and mature into focal adhesions (FAs), in a matrix rigidity- dependent manner7,8. In turn, FAs maturation has profound consequences for the cell as it modulates signaling pathways regulating migration, survival and proliferation. Clathrin-coated structures (CCSs) are mostly described to control the uptake of cell-surface receptors, including some integrins. However, it is now clear that in some conditions, CCSs can also serve as integrin-dependent adhesion structures9. Many cell types, including HeLa cells, display two distinct types of CCSs: cano- nical, dynamic clathrin-coated pits (CCPs) and long-lived, large and ﬂat clathrin lattices also called plaques. Although plaques have been widely described and shown to be enriched in signaling receptors and integrins10–12, it is still not clear how they form and what is their function. CCSs have mostly been studied in cells growing on glass which is an extremely stiff substrate. A whole range of tissue rigidity is encountered in vivo with some tissues being very soft (Young’s modulus, E ≈0.1 kPa) like the brain or fat tissues, while some other are stiffer like muscles (≈30 kPa)13. Here, we set out to investigate CCSs dynamics on substrates of controlled elasticity. We report that clathrin-coated plaques assemble as a consequence of increasing substrate rigidity. Frustrated endocytosis controls contractility- independent mechanotransduction at clathrin- coated structures In addition, plaque formation was rescued by expressing a siRNA-resistant β5-encoding construct in β5-depleted HeLa cells (Fig. 2g). Of note, MDA-MB-231 cells that mostly harbor dynamic CCSs (Supplementary Fig. 2e, f) displayed drastically lower levels of αv and β5 integrins as compared to HeLa cells (Supplementary Fig. 4h). Overexpression of αv and β5 integrins in these cells resulted in a ~4-fold increase in long-lived CCSs (Supplementary Fig. 4i). Together, our results demon- strate that αvβ5-integrin is required for plaque formation on rigid environments. NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications Results b g CCS dynamics in genome-edited HeLa cells expressing endogenous GFP-tagged μ2-adaptin seeded on the indicated collagen- aged by spinning disk microscopy every 5 s for 5 min. c Quantiﬁcation of the dynamics of CCSs observed as in b (*P < 0.001, as ndition, one-way analysis of variance—ANOVA. n = 3). d HeLa cells seeded on collagen-coated glass were treated with Blebbistatin dicated for 30 min before being ﬁxed and stained for α-adaptin. Scale bar: 10 µm. Higher magniﬁcations of boxed regions are shown dynamics of CCSs observed as in d (P < 0.01, P < 0.05, ANOVA. n = 3). f EM micrographs of unroofed HeLa cells that were cultu for 30 min with blebbistatin before being ﬁxed and processed. Clathrin-coated plaques are highlighted in green. Scale bar: 200 n sed as mean ± SD Glass 0.1 kPa 5 kPa 31 kPa a Distance (20 μm) Glass 0.1 kPa 5 kPa 31 kPa Time (5 min) b b Glass 0.1 kPa 5 kPa 31 kPa a Ti ( i ) b a 100 80 60 40 20 0 Percentage of total CCSs 0.1 kPa Glass 31 kPa 5 kPa Dynamic Static * c d d c Blebbistatin Cytochalasin D d 100 80 60 40 20 0 Percentage of total CCSs Dynamic Static * Control Cytochalasin D Blebbistatin 0 e f e e Fig. 1 Clathrin-coated plaques are mechanosensitive structures. a HeLa cells were seeded on collagen-coated glass or polyacrylamide gels of indicated stiffness and ﬁxed 24 h later before being stained for α-adaptin. Scale bar: 15 µm. Higher magniﬁcations of boxed regions are shown. b Kymographs showing CCS dynamics in genome-edited HeLa cells expressing endogenous GFP-tagged μ2-adaptin seeded on the indicated collagen-coated substrate and imaged by spinning disk microscopy every 5 s for 5 min. c Quantiﬁcation of the dynamics of CCSs observed as in b (P < 0.001, as compared to 0.1 kPa condition, one-way analysis of variance—ANOVA. n = 3). d HeLa cells seeded on collagen-coated glass were treated with Blebbistatin or Cytochlalasin D as indicated for 30 min before being ﬁxed and stained for α-adaptin. Scale bar: 10 µm. Higher magniﬁcations of boxed regions are shown. e Quantiﬁcation of the dynamics of CCSs observed as in d (P < 0.01, P < 0.05, ANOVA. n = 3). Results Clathrin-coated plaques are sensitive to substrate rigidity. When HeLa cells were grown on collagen-coated glass, ventral plasma membrane CCSs marked with the α-adaptin subunit of the clathrin adaptor AP-2 appeared as a mix of dot-like, diffraction-limited structures corresponding to CCPs, and large, heterogeneous structures corresponding to plaques, as previously reported11,12,14 (Fig. 1a). Strikingly, cells seeded on soft (0.1 kPa) collagen-coated polyacrylamide gels only showed dot-like CCSs suggesting that plaques cannot form in these conditions (Fig. 1a). Similar results were obtained with cells cultured on 5 kPa gels (Fig. 1a). However, cells seeded on 31 kPa gels showed a mix of diffraction-limited CCPs and larger structures potentially corre- sponding to plaques (Fig. 1a). Super-resolution STED microscopy analyses further conﬁrmed the presence of many large CCSs in cells grown on glass or on 31 kPa gels while only dot-like structures were detected on 0.1 and 5 kPa gels (Supplementary Fig. 1a). Scanning electron microscopy analyses of unroofed cells conﬁrmed the presence of large, ﬂat clathrin-coated plaques at the adherent plasma membrane of cells cultured on glass or on 31 kPa gels (Supplementary Fig. 1b). Importantly, such large and ﬂat clathrin lattices were mostly absent in cells seeded on 0.1 or 5 kPa gels (Supplementary Fig. 1b). We next performed live cell ima- ging of genome-edited HeLa cells engineered to express GFP- tagged, endogenous μ2-adaptin subunit of AP-2. Many CCSs Integrins are generally recruited at CCSs through binding to speciﬁc adaptors of the Phosphotyrosine Binding (PTB) NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 2 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y t i i t i f il F i t β5 i t i h d t d ti f th ti f l li 100 80 60 40 20 0 Percentage of total CCSs Dynamic Static * Control Cytochalasin D Blebbistatin 100 80 60 40 20 0 Percentage of total CCSs 0.1 kPa Glass 31 kPa 5 kPa * Dynamic Static * Blebbistatin Cytochalasin D Glass 0.1 kPa 5 kPa 31 kPa a Distance (20 μm) Glass 0.1 kPa 5 kPa 31 kPa Time (5 min) b c d f e Clathrin-coated plaques are mechanosensitive structures. a HeLa cells were seeded on collagen-coated glass or polyacrylamide g s and ﬁxed 24 h later before being stained for α-adaptin. Scale bar: 15 µm. Higher magniﬁcations of boxed regions are shown. Results f EM micrographs of unroofed HeLa cells that were cultured on glass and treated for 30 min with blebbistatin before being ﬁxed and processed. Clathrin-coated plaques are highlighted in green. Scale bar: 200 nm. All results are expressed as mean ± SD domain-containing protein family. For instance, β5-integrin has been shown to interact with the clathrin adaptors Numb and Dab221. Both Numb and Dab2 are present at plaques (Supple- mentary Fig. 5a) and inhibiting their expression using speciﬁc siRNAs led to a strong reduction of the size of CCSs as well as a moderate reduction of the proportion of long-lived structures (Supplementary Fig. 5b–e). In these experiments, we noticed a strong delocalization of αvβ5 to peripheral structures that were positive for the FA marker phospho-Focal Adhesion Kinase (FAK) (Supplementary Fig. 5f). A similar phenotype was NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y Percentage of total CCSs Dynamic Static siControl siαv-1 siαv-2 siβ5-2 siβ5-1 ** b αvβ5/α-adaptin fluorescence intensity ratio Plaques CCPs * 100 80 60 40 20 0 1 0.8 0.6 0.4 0.2 0 0.8 0.6 0.4 0.2 0 αvβ5/α-adaptin Pearson’s correlation coefficient Glass 31 kPa * 5 kPa 0.1 kPa Distance (20 μm) siControl siβ5-1 siβ5-2 Time (5 min) * * α-Adaptin β5-GFP siβ5-1 a α-Adaptin αvβ5 c e f siControl siβ5-1 d g g. 2 αvβv5 integrin localizes to plaques and is required for their assembly. a HeLa cells were seeded on collagen-coated glass and ﬁxed 2 eing stained for α-adaptin and αvβ5-integrin. Scale bar: 10 µm. Higher magniﬁcations of boxed regions are shown. Arrows point to clathrin- rrowheads point to CCPs. b Quantiﬁcation of αvβ5 enrichment at plaques versus CCPs (P < 0.005, two tailed Student’s t-test. n = 3; 100 xperiment were counted.). c Quantiﬁcation of colocalization (Pearson’s coefﬁcient) between αvβ5 and α-adaptin in cells cultured on th ollagen-coated substrate (P < 0.005; P < 0.001, one-way analysis of variance—ANOVA. n = 3). d HeLa cells treated with control (upp peciﬁc (lower panel) siRNAs were seeded on collagen-coated glass and ﬁxed 24 h later before being stained for α-adaptin. Scale bar: 15 µm howing CCS dynamics in genome-edited HeLa cells treated with the indicated siRNA, seeded on collagen-coated glass, and imaged by icroscopy every 5 s for 5 min. Results f Quantiﬁcation of the dynamics of CCSs observed as in e and treated with the indicated siRNAs (P < 0.00 ). g HeLa cells treated with β5-speciﬁc siRNAs were transfected with a siRNA-resistant β 5-GFP encoding construct and then ﬁxed 24 h la ained for α-adaptin. Scale bar: 10 µm. The star marks a cell that is not transfected by β5-GFP. Higher magniﬁcations of boxed regions esults are expressed as mean ± SD b αvβ5/α-adaptin fluorescence intensity ratio Plaques CCPs * 1 0.8 0.6 0.4 0.2 0 a α-Adaptin αvβ5 αvβ5 b a a α-Adaptin Percentage of total CCSs Dynamic Static siControl siαv-1 siαv-2 siβ5-2 siβ5-1 100 80 60 40 20 0 f f 0.8 0.6 0.4 0.2 0 αvβ5/α-adaptin Pearson’s correlation coefficient Glass 31 kPa * 5 kPa 0.1 kPa c f Distance (20 μm) siControl siβ5-1 siβ5-2 Time (5 min) e e c Percentage of total CCSs Time (5 min) siControl siβ5-1 d d * * α-Adaptin β5-GFP siβ5-1 g g * α-Adaptin siβ g * β5-GFP Fig. 2 αvβv5 integrin localizes to plaques and is required for their assembly. a HeLa cells were seeded on collagen-coated glass and ﬁxed 24 h later before being stained for α-adaptin and αvβ5-integrin. Scale bar: 10 µm. Higher magniﬁcations of boxed regions are shown. Arrows point to clathrin-coated plaques; arrowheads point to CCPs. b Quantiﬁcation of αvβ5 enrichment at plaques versus CCPs (P < 0.005, two tailed Student’s t-test. n = 3; 100 structures per experiment were counted.). c Quantiﬁcation of colocalization (Pearson’s coefﬁcient) between αvβ5 and α-adaptin in cells cultured on the indicated collagen-coated substrate (P < 0.005; P < 0.001, one-way analysis of variance—ANOVA. n = 3). d HeLa cells treated with control (upper panel) or β5- speciﬁc (lower panel) siRNAs were seeded on collagen-coated glass and ﬁxed 24 h later before being stained for α-adaptin. Scale bar: 15 µm. e Kymographs showing CCS dynamics in genome-edited HeLa cells treated with the indicated siRNA, seeded on collagen-coated glass, and imaged by spinning disk microscopy every 5 s for 5 min. f Quantiﬁcation of the dynamics of CCSs observed as in e and treated with the indicated siRNAs (P < 0.001, ANOVA. n = 3). g HeLa cells treated with β5-speciﬁc siRNAs were transfected with a siRNA-resistant β 5-GFP encoding construct and then ﬁxed 24 h later before being stained for α-adaptin. Results Similar to experiments performed with trypsin, plaques seemingly disassembled into dot-like structures evoking CCPs (Fig. 3d). Thus, acute inhibition of αvβ5 binding to the substrate may allow the clathrin machinery to bud, leading to plaques dissolution. Accordingly, we observed that overexpressed, GFP-tagged β5- integrin accumulated into mCherry-tagged transferrin receptor (TfR)-positive vesicles after but not before treating cells with Cilengitide (Supplementary Fig. 8d). This demonstrates that the integrin was internalized upon adding the drug, in agreement with our hypothesis. Plaques are signaling platforms for different receptors. It has been proposed that plaques are a hub for the recruitment/sorting of many signaling receptors11,12,30. Prolonged CCS lifetime has been associated with sustained signaling and CCSs have been envisaged as signaling platforms, independently of their role in endocytosis31. Thus, long-lived clathrin-coated plaques may modulate signaling pathways. In agreement with this hypothesis, we observed that the inhibition of plaque formation in β5- integrin-depleted cells cultured on glass resulted in a reduced steady-state signaling in the Erk pathway (Fig. 4a, b). However, β5 knockdown did not modulate Erk activity on a soft gel that do not allow plaque formation (Fig. 4c, d). This suggested that plaque assembly, but not simply β5 expression, is required to tune Erk activity. Surprisingly, AP-2 or clathrin heavy chain (CHC) knockdown did not modulate Erk activation status on a rigid surface (Supplementary Fig. 9a). Because in these latter condi- tions, αvβ5 shuttling to FAs leads to the enlargement of these structures (Supplementary Fig. 6f, g) and because FAs are also known to signal in the Erk pathway32, we aimed at testing whether FAs could mask a potential role for plaques in regulating Erk activity. We ﬁrst observed that AP-2 or CHC knockdown induced a strong accumulation of phosphotyrosines at the enlarged FAs (Supplementary Fig. 9b–d), in a β5-integrin- dependent manner (Supplementary Fig. 6f–g). In agreement with our hypothesis, these results suggest that the signaling activity of FAs is increased in the absence of CCSs. In addition, while the inhibition of FA formation in Talin1-depleted cells only slightly inhibited steady-state Erk activity, co-knockdown of Talin1 and AP-2 or CHC strongly reduced it (Supplementary Fig. 9e, f). Together, our results suggest that Erk activity is mostly controlled by CCSs in HeLa cells but that in the absence of CCSs, αvβ5-mediated FA enlargement maintains steady-state Erk activity. Results Scale bar: 10 µm. The star marks a cell that is not transfected by β5-GFP. Higher magniﬁcations of boxed regions are shown. All results are expressed as mean ± SD observed upon inhibiting CCSs formation with AP-2 or CHC siRNAs (Supplementary Fig. 5f). This suggested that αvβ5 may shuttle to FAs in the absence of clathrin-coated plaques. αvβ5 is only poorly associated with FAs in control cells and, accordingly, β5-depletion did not modulate FAs number and size, nor the activation status of FAK (Supplementary Fig. 6a–e). However, preventing CCSs formation with AP-2 siRNAs resulted in larger FAs and this was dependent on β5-integrin expression (Supple- mentary Fig. 6f, g). Thus, in the absence of CCSs, αvβ5 shuttling to FAs induces an overgrowth of these adhesion structures. Together, our data suggest that Dab2/Numb-dependent recruit- ment of αvβ5 at CCSs results in the formation of clathrin-coated plaques and that in the absence of plaques, αvβ5 is targeted to and modulates the dynamics of FAs. Plaque formation is a consequence of frustrated endocytosis. We reasoned that αvβ5 engagement with the ECM may prevent CCSs budding, leading to the formation of plaques. The main αvβ5 ligand is vitronectin, an ECM component that is present in NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 4 4 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y that immobilizing the TfR, which has a strong afﬁnity for AP-2, may stall endocytosis and result in plaque formation. Indeed, while CCSs of β5-depleted cells were dot-liked and short-lived on the anti-mCherry antibodies-coated glass, cells expressing TfR- mCherry displayed many large and long-lived structures (Fig. 3e–g and Supplementary Movie 6). We further controlled that TfR-mCherry was present at CCSs in these conditions (Supplementary Fig. 8e). Of note, TfR-mCherry-expressing cells seeded on a coverslip functionalized with an irrelevant antibody where not able to assemble plaques upon β5-depletion (Fig. 3g). In addition, immobilizing the TfR on soft gels resulted in the accumulation of long-lived and large CCSs, although to a lesser extend as compared to the glass condition (Supplementary Fig. 8f). Thus, strong receptor engagement with the substrate is sufﬁcient to prevent CCSs budding and to lead to plaque formation. Overall, our results demonstrate that plaques assemble as a consequence of frustrated endocytosis. the serum and that was described as the serum spreading factor allowing cells to adhere on glass but also binding a plethora of other ECM components, including collagen22,23. Results We observed an increased density of large and stable CCSs when the glass was ﬁrst coated with vitronectin before allowing HeLa cells to adhere in the presence of complete medium (Supplementary Fig. 7a, b). However, vitronectin coating did not result in more static CCSs on the soft environment (Supplementary Fig. 7c, d). In addition, most large and long-lived CCSs were detected at cell/ECM con- tact areas rather than at non-adherent regions of the ventral plasma membrane when cells were cultured on ring-shaped, collagen-coated micropatterns (Supplementary Fig. 7e–g). These data suggest that local and strong engagement of αvβ5 with the ECM may be required for plaque assembly. To test this hypothesis, cells seeded on glass were incubated in the presence of trypsin. We reasoned that trypsin-mediated αvβ5 proteolysis may lead to the disassembly of clathrin-coated plaques. Indeed, we observed a fast dissolution of plaques upon trypsin treatment (Fig. 3a). Dot-like structures seemingly emanating from dissol- ving plaques were frequently observed before they abruptly vanished (Fig. 3a, arrows). These observations suggest that CCPs may bud in the vicinity of disassembling plaques, potentially participating in the dissolution of these structures. Along this line, we observed that auxillin bursts, a feature of budding CCPs24, were signiﬁcantly more frequent at disassembling plaques as compared to plaques of untreated cells (Fig. 3b, c and Supple- mentary Movie 3). Because trypsin treatment is not speciﬁc towards αvβ5, we next monitored plaque dynamics in cells treated with Cilengitide, a drug that binds the RGD-binding cleft of this integrin and outcompetes ECM ligands25–27. CCSs of HeLa cells grown on glass and treated with Cilengitide were highly dynamic as compared to untreated cells (Supplementary Fig. 8a, b and Supplementary Movie 4), demonstrating that direct binding of αvβ5 to the substrate is required to maintain plaques. Moreover, we observed a fast dissolution of plaques when monitoring CCS dynamics immediately after adding Cilengitide in the medium (Fig. 3d, Supplementary Fig. 8c and Supplementary Movie 5). Similar to experiments performed with trypsin, plaques seemingly disassembled into dot-like structures evoking CCPs (Fig. 3d). Thus, acute inhibition of αvβ5 binding to the substrate may allow the clathrin machinery to bud, leading to plaques dissolution. Accordingly, we observed that overexpressed, GFP-tagged β5- integrin accumulated into mCherry-tagged transferrin receptor (TfR)-positive vesicles after but not before treating cells with Cilengitide (Supplementary Fig. 8d). Results This demonstrates that the integrin was internalized upon adding the drug, in agreement with our hypothesis. Based on a theoretical model, it has been proposed that i i dh i h i d d f i the serum and that was described as the serum spreading factor allowing cells to adhere on glass but also binding a plethora of other ECM components, including collagen22,23. We observed an increased density of large and stable CCSs when the glass was ﬁrst coated with vitronectin before allowing HeLa cells to adhere in the presence of complete medium (Supplementary Fig. 7a, b). However, vitronectin coating did not result in more static CCSs on the soft environment (Supplementary Fig. 7c, d). In addition, most large and long-lived CCSs were detected at cell/ECM con- tact areas rather than at non-adherent regions of the ventral plasma membrane when cells were cultured on ring-shaped, collagen-coated micropatterns (Supplementary Fig. 7e–g). These data suggest that local and strong engagement of αvβ5 with the ECM may be required for plaque assembly. To test this hypothesis, cells seeded on glass were incubated in the presence of trypsin. We reasoned that trypsin-mediated αvβ5 proteolysis may lead to the disassembly of clathrin-coated plaques. Indeed, we observed a fast dissolution of plaques upon trypsin treatment (Fig. 3a). Dot-like structures seemingly emanating from dissol- ving plaques were frequently observed before they abruptly vanished (Fig. 3a, arrows). These observations suggest that CCPs may bud in the vicinity of disassembling plaques, potentially participating in the dissolution of these structures. Along this line, we observed that auxillin bursts, a feature of budding CCPs24, were signiﬁcantly more frequent at disassembling plaques as compared to plaques of untreated cells (Fig. 3b, c and Supple- mentary Movie 3). Because trypsin treatment is not speciﬁc towards αvβ5, we next monitored plaque dynamics in cells treated with Cilengitide, a drug that binds the RGD-binding cleft of this integrin and outcompetes ECM ligands25–27. CCSs of HeLa cells grown on glass and treated with Cilengitide were highly dynamic as compared to untreated cells (Supplementary Fig. 8a, b and Supplementary Movie 4), demonstrating that direct binding of αvβ5 to the substrate is required to maintain plaques. Moreover, we observed a fast dissolution of plaques when monitoring CCS dynamics immediately after adding Cilengitide in the medium (Fig. 3d, Supplementary Fig. 8c and Supplementary Movie 5). NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications Results A total of 90 structures from three independent exp TICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467- 10 20 30 40 130 140 150 160 a a 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 a 50 60 70 80 90 100 110 120 170 180 190 200 210 220 230 240 Distance (2 μm) Auxillin Time (180 s) μ2- Adaptin Merge b d e b d 200 225 250 275 300 325 350 375 400 425 450 475 d μ2-Adaptin-GFP e e TfR-mCherry Percentage of total CCSs g 100 80 60 40 20 0 Percentage of total CCSs Dynamic Static Control Ab siβ5-1 * Control Ab mCherry Ab mCherry Ab siControl g 4 3 2 1 0 Number of auxillin flashes at plaques per minuite Before trypsin * After trypsin c Distance (40 μm) TfR-mcherry expressing cell Time (2.5 min) Non-TfR-mcherry expressing cell μ2-Adaptin TfR f f g Percentage of total CCSs g Time (2.5 min) siControl siControl Fig. 3 Clathrin-coated plaques assemble as a consequence of frustrated endocytosis. a Genome-edited HeLa cells expressing endogenous mCherry-tagged μ2-adaptin, seeded on collagen-coated glass, were treated with trypsin and imaged by spinning disk microscopy every 5 s. Time after trypsin addition is indicated in seconds. Arrowheads point to dot-like structure emanating from the disassembling plaques. Scale bar: 1 μm. b Kymographs showing plaque disassembly dynamics and concomitant GFP-auxillin bursts in HeLa cell treated and imaged as in a. Note that loss of μ2-adaptin-associated ﬂuorescence is correlated with auxillin ﬂashes. c Quantiﬁcation of the number of auxillin ﬂashes per plaque and per minute in HeLa cells before and after incubation with trypsin. Results are expressed as mean ± SD (P < 0.001, Mann–Whitney rank sum test. A total of 90 structures from three independent experiments was quantiﬁed). d Genome-edited HeLa cells expressing endogenous mCherry-tagged μ2-adaptin, seeded on collagen-coated glass, were treated with Cilengitide and imaged by spinning disk microscopy every 5 s. Time after Cilengitide addition is indicated in seconds. Arrowheads point to dot-like structure emanating from the disassembling plaques. Scale bar: 1 μm. e Genome-edited HeLa cells treated with β5-speciﬁc siRNA and transfected with a construct encoding mCherry-tagged TfR were seeded on anti-mCherry antibodies-coated glass and imaged 24 h later by spinning disk microscopy every 5 s. Scale bar: 10 μm. Results Based on a theoretical model, it has been proposed that integrin adhesion strength is reduced on soft environments because local elastic recoil upon stochastic integrin/ligand unbinding would lower chances to rebind, independently of forces28,29. Such a mechanism may explain why plaques only assemble on relatively stiff but not on more elastic substrates. We observed that the kinetics of plaque disassembly upon Cilengitide treatment were directly correlated with the stiffness of the substrate as plaques disassembled faster on softer environments (Supplementary Fig. 8c). Because Cilengitide acts by outcompet- ing integrin ligands, these results suggest that αvβ5 adhesion strength is reduced on soft substrates. y To test more directly whether clathrin-coated plaques regulate Erk signaling, we took advantage of the possibility to rescue plaque formation in β5-depleted cells with the antibody-coating system described above. In these experimental conditions, cells are seeded on a layer of antibodies deposited onto a Poly-L-lysine coated glass. Poly-L-lysine is known to inhibit FAs maturation33 and indeed, we could conﬁrm that HeLa cells plated on such a substrate displayed fewer and smaller FAs (Supplementary Fig. 10a–c). Cells expressing TfR-mCherry and Erk-GFP were treated with control or β5-integrin-speciﬁc siRNAs and plated onto control or anti-mCherry antibody-coated glass. We observed that Erk-GFP nuclear enrichment, which reﬂects Erk activation, was similar for control cells plated on irrelevant or on anti- mCherry-coated surfaces (Fig. 4e, f). Conﬁrming our western-blot Our data suggest that plaque assembly results from a frustrated endocytosis process whereby strong receptor-mediated anchoring to the stiff substrate prevents CCSs budding. To further test this hypothesis, we made use of a construct encoding the transferrin receptor (TfR) carrying a mCherry tag in its extracellular domain. β5-integrin-depleted HeLa cells expressing or not TfR-mCherry were plated on anti-mCherry antibody-coated glass. Results We reasoned 5 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y 100 80 60 40 20 0 Percentage of total CCSs Dynamic Static Control Ab siβ5-1 Control Ab mCherry Ab mCherry Ab siControl 4 3 2 1 0 Number of auxillin flashes at plaques per minuite Before trypsin * After trypsin Distance (40 μm) TfR-mcherry expressing cell Time (2.5 min) Non-TfR-mcherry expressing cell μ2-Adaptin TfR Distance (2 μm) Auxillin Time (180 s) μ2- Adaptin Merge 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 200 225 250 275 300 325 350 375 400 425 450 475 190 200 210 220 230 240 a b d μ2-Adaptin-GFP TfR-mCherry e c f g Clathrin-coated plaques assemble as a consequence of frustrated endocytosis. a Genome-edited HeLa cells expressing endogenous mC aptin, seeded on collagen-coated glass, were treated with trypsin and imaged by spinning disk microscopy every 5 s. Time after tryps ed in seconds. Arrowheads point to dot-like structure emanating from the disassembling plaques. Scale bar: 1 μm. b Kymographs sh embly dynamics and concomitant GFP-auxillin bursts in HeLa cell treated and imaged as in a. Note that loss of μ2-adaptin-associated ﬂ ted with auxillin ﬂashes. c Quantiﬁcation of the number of auxillin ﬂashes per plaque and per minute in HeLa cells before and after in . Results are expressed as mean ± SD (P < 0.001, Mann–Whitney rank sum test. Results Results are expressed as mean ± SD from four independent experiments (P < 0.05, one-way analysis of variance—ANOVA). c Western-blot analysis of phospho-Erk (P-Erk) levels in HeLa cells growing on collagen- coated, 0.1 kPa polyacrylamide gels and treated with control or integrin β5-speciﬁc siRNAs, as indicated. Total-Erk was used as a loading control (representative image of three independent experiments) d Densitometry analysis of bands obtained in western-blots as in c. Results are expressed as mean ± SD from three independent experiments. e GFP-Erk-expressing HeLa cells treated or not with a β5-speciﬁc siRNA and transfected or not with TfR- mCherry, as indicated, were seeded on anti-mCherry antibodies-coated glass and imaged 24 h later. Scale bar: 10 µm. A color-coded scale for low and high signal intensity is shown. f Quantiﬁcation of GFP-Erk nuclear enrichment index in cells as in e and cultured on control antibody- or anti-mCherry antibody- coated glass, as indicated, and treated or not with indicated siRNAs. Results are expressed as mean ± SD (P < 0.01, P < 0.001, ANOVA. ControlAb-siCtr: 114 cells from n = 5 independent experiments; mChAb-SiCtrl: 119 cells from n = 5 independent experiments; ControAb-siβ5-1: 146 cells from n = 5 independent experiments; mChAb-Siβ5-1: 120 cells from n = 5 independent experiments; ControlAb-siβ5-1 + siAP-2: 69 cells from n = 3 independent experiments; mChAb-siβ5-1 + siAP-2: 68 cells from n = 3 independent experiments; ControlAb-siβ5-1 + siCHC: 63 cells from n = 3 independent experiments; mChAb-siβ5-1 + siCHC: 67 cells from n = 3 independent experiments) (Supplementary Fig. 11b, c). Collectively, these data strongly suggest that plaques serve as signaling platforms for different receptors. In addition to EGFR and HGFR, many other signaling receptors are known to be recruited at plaques11,12, and we propose that this is critical to regulate the observed steady-state Erk activity. Indeed, we observed that EGFR and HGFR are still recruited to immobilized TfR-induced plaques in the absence of β5-integrin and that phosphotyrosine staining still accumulate at these artiﬁcial structures upon stimulation (Fig. 5e–g). Thus, our model is that αvβ5 controls plaque formation on stiff environ- ments but that it is the subsequent recruitment of different signaling receptors at plaques that is the cause of Erk activity modulation. structures corresponding to plaques were strongly marked with an anti-phosphotyrosine antibody upon EGF stimulation (Fig. 5a, b). Results ControlAb-siCtr: 114 cells from n = 5 independent experiments; mChAb-SiCtrl: 119 cells from n = 5 independent experiments; ControAb-siβ5-1: 146 cells from n = 5 independent experiments; mChAb-Siβ5-1: 120 cells from n = 5 independent experiments; ControlAb-siβ5-1 + siAP-2: 69 cells from n = 3 independent experiments; mChAb-siβ5-1 + siAP-2: 68 cells from n = 3 independent experiments; ControlAb-siβ5-1 + siCHC: 63 cells from n = 3 independent experiments; mChAb-siβ5-1 + siCHC: 67 cells from n = 3 independent experiments) NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y ARTICLE P-Erk Total-Erk siβ5-1 siβ5-2 siControl 35 35 100 80 60 40 20 0 Average band intensity (% of control) siControl siβ5-1 siβ5-2 * 1 0.8 0.6 0.4 0.2 0 GFP-ERK nuclear enrichment index siCtr siβ5-1 siβ5-1 + * ** Control Ab mCherry Ab Control Ab mCherry Ab Control Ab mCherry Ab Control Ab mCherry Ab siAP2 siCHC * P-Erk siControl siβ5-1 siβ5-2 0.1 kPa Total-Erk 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 0 20 40 60 80 100 120 140 160 GFP-ERK DAPI siControl siβ5-1 siControl siβ5-1 Control antibody coating Anti-mCherry antibody coating High Low a e b c d f P-Erk Total-Erk siβ5-1 siβ5-2 siControl 35 35 GFP ERK a e a e P-Erk Total-Erk 35 35 100 80 60 40 20 0 Average band intensity (% of control) siControl siβ5-1 siβ5-2 * * GFP-ERK DAPI Control antibody coating Anti-mCherry antibody coating b GFP-ERK b DAPI 1 0.8 0.6 0.4 0.2 0 GFP-ERK nuclear enrichment index siCtr siβ5-1 siβ5-1 + * ** Control Ab mCherry Ab Control Ab mCherry Ab Control Ab mCherry Ab Control Ab mCherry Ab siAP2 siCHC * P-Erk siControl siβ5-1 siβ5-2 0.1 kPa Total-Erk 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 0 20 40 60 80 100 120 140 160 Control antibody coating Anti mCherry antibody coating High Low c d f P-Erk siControl siβ5-1 siβ5-2 0.1 kPa Total-Erk 35 35 c d c Fig. 4 Clathrin-coated plaques regulate stiffness-dependent Erk signaling. a Western-blot analysis of phospho-Erk (P-Erk) levels in HeLa cells growing on collagen-coated glass and treated with control or β5-speciﬁc siRNAs as indicated (representative image of four independent experiments). Total-Erk was used as a loading control. b Densitometry analysis of bands obtained in western-blots as in a. Results f Kymograph showing CCS dynamics in the region corresponding to the boxed area in e. Note that the cell on the left is not transfected by the TfR-mCherry construct and only display dynamic CCSs. g Quantiﬁcation of the dynamics of CCSs observed as in e in genome-edited HeLa cells expressing the TfR-mCherry construct and treated as indicated (P < 0.001, one-way analysis of variance—ANOVA. n = 3). Results are expressed as mean ± SD Together, our data demonstrate that plaques serve as signaling platforms. analysis, β5-depleted cells showed a reduction of Erk accumula- tion in the nucleus when plated on the control substrate (Fig. 4e, f). However, nuclear Erk accumulation was restored when plaque formation was allowed by plating cells on anti-mCherry anti- body-coated glass (Fig. 4e, f). This results suggest that clathrin- coated plaques regulate steady-state Erk activity independently of β5-integrin. Accordingly, we observed that CHC and AP-2 were required to restore Erk nuclear levels in these conditions (Fig. 4f). We next aimed at investigating the mechanisms of plaque- regulated Erk activation. Conﬁrming previous ﬁndings34,35, we observed that the epidermal growth factor receptor (EGFR) accumulates at plaques upon stimulation (Supplementary Fig. 10d) and the hepatocyte growth factor receptor (HGFR) behaved similarly (Supplementary Fig. 10e). Large AP-2-positive NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 6 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y P-Erk Total-Erk siβ5-1 siβ5-2 siControl 35 35 100 80 60 40 20 0 Average band intensity (% of control) siControl siβ5-1 siβ5-2 * 1 0.8 0.6 0.4 0.2 0 GFP-ERK nuclear enrichment index siCtr siβ5-1 siβ5-1 + * ** Control Ab mCherry Ab Control Ab mCherry Ab Control Ab mCherry Ab Control Ab mCherry Ab siAP2 siCHC * P-Erk siControl siβ5-1 siβ5-2 0.1 kPa Total-Erk 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 0 20 40 60 80 100 120 140 160 GFP-ERK DAPI siControl siβ5-1 siControl siβ5-1 Control antibody coating Anti-mCherry antibody coating High Low a e b c d f Fig. 4 Clathrin-coated plaques regulate stiffness-dependent Erk signaling. a Western-blot analysis of phospho-Erk (P-Erk) levels in HeLa cells growing on collagen-coated glass and treated with control or β5-speciﬁc siRNAs as indicated (representative image of four independent experiments). Total-Erk was used as a loading control. b Densitometry analysis of bands obtained in western-blots as in a. Results Results are expressed as mean ± SD from four independent experiments (P < 0.05, one-way analysis of variance—ANOVA). c Western-blot analysis of phospho-Erk (P-Erk) levels in HeLa cells growing on collagen- coated, 0.1 kPa polyacrylamide gels and treated with control or integrin β5-speciﬁc siRNAs, as indicated. Total-Erk was used as a loading control (representative image of three independent experiments) d Densitometry analysis of bands obtained in western-blots as in c. Results are expressed as mean ± SD from three independent experiments. e GFP-Erk-expressing HeLa cells treated or not with a β5-speciﬁc siRNA and transfected or not with TfR- mCherry, as indicated, were seeded on anti-mCherry antibodies-coated glass and imaged 24 h later. Scale bar: 10 µm. A color-coded scale for low and high signal intensity is shown. f Quantiﬁcation of GFP-Erk nuclear enrichment index in cells as in e and cultured on control antibody- or anti-mCherry antibody- coated glass, as indicated, and treated or not with indicated siRNAs. Results are expressed as mean ± SD (P < 0.01, P < 0.001, ANOVA. NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications Results Smaller structures corresponding to CCPs were also positive for the anti-phosphotyrosine staining but to a lesser extend (Fig. 5a, b). Anti-phosphotyrosine staining was absent from both large and small CCSs when cells were ﬁrst treated with Geﬁtinib, a drug that inhibits EGFR kinase activity (Fig. 5a, b). Thus, our data suggest that plaques may serve as signaling platforms in the EGFR pathway. Accordingly, EGF-induced Erk activation was reduced upon plaques ablation in β5-integrin-depleted cells (Fig. 5c, d). Of note, cell-surface EGFR levels where not signiﬁcantly modulated in these later conditions, excluding the possibility that reduced Erk activity is due to a down-modulation of the surface-exposed receptor (Supplementary Fig. 11a). In addition, a speciﬁc enrichment of phosphotyrosine staining at plaques was also observed upon stimulation with HGF Plaques regulate cell proliferation. Our observations suggest that plaques are signaling platforms that assemble in response to 7 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y 1 0.8 0.6 0.4 0.2 0 P-Tyr/α-adaptin fluorescence intensity ratio Plaques * EGF + Gef. EGF Starved CCPs EGF α-Adaptin P-Tyr Starved EGF EGF + Gefitinib * * siControl 0 30 60 120 0 30 60 120 0 30 60 120 si β5-1 si β5-2 Time (min): P-Erk Total-Erk 35 35 Band intensity (A.U.) 0.0 0.5 1.0 1.5 2.0 2.5 siControl 0 min 30 min 60 min 120 min * * si β5-1 si β5-2 Time after EGF addition (s): TfR-mCherry EGFR-GFP TfR-mCherry HGFR-GFP Time after HGF addition (s): 0 570 0 470 P-Tyr TfR-mCherry a b c d e f g oated plaques locally regulate receptor-dependent Erk signaling. a HeLa cells seeded on collagen-coated glass were starved for 4 h and ot with 10 ng/ml EGF for 5 min alone or added after 30 min preincubation with 10 µM Geﬁtinib. Cells were then ﬁxed and stained for α- sphotyrosines. The arrowhead points to one CCP and the star marks a plaque. Scale bar: 2 μm. b Quantiﬁcation of phosphotyrosines plaques or CCPs in the indicated conditions. Results are expressed as mean ± SD (P < 0.05, P < 0.001, one-way analysis of variance— Number of structures analyzed per condition: Starved 160 plaques; EGF: 452 plaques, EGF + Geﬁtinib: 301 plaques, CCPs/EGF: 200 pits). nalysis of phospho-Erk (P-Erk) levels in starved HeLa cells treated with control or β5-speciﬁc siRNAs as indicated, and stimulated for the ith 10 ng/ml EGF. Results Total-Erk was used as a loading control (representative image of three independent experiments). d Densitometry s obtained in western-blots as in c. Results are expressed as mean ± SD (P < 0.05, ANOVA. n = 3). e HeLa cells treated with β5-speciﬁc nsfected with plasmids encoding for TfR-mCherry and EGFR-GFP and seeded on anti-mCherry antibodies-coated glass. Cells were serum- nd then treated with 10 ng/ml EGF for 470 s. Scale bar: 0.5 μm. Arrowheads point to plaques positive for EGFR-GFP. f HeLa cells treated siRNA were transfected with plasmids encoding for TfR-mCherry and HGFR-GFP and seeded on anti-mCherry antibodies-coated glass. m-starved for 4 h and then treated with 50 ng/ml HGF for 570 s. Scale bar: 0.5 μm. Arrowheads point to plaques positive for HGFR-GFP. ated with β5-speciﬁc siRNA were transfected with plasmids encoding for TfR-mCherry and seeded on anti-mCherry antibodies-coated serum-starved for 4 h and then treated with 10 ng/ml EGF for 5 min prior to ﬁxation and staining for phosphotyrosines. Scale bar: 1.5 μm. 1 0.8 0.6 0.4 0.2 0 P-Tyr/α-adaptin fluorescence intensity ratio ** * EGF + Gef. EGF Starved EGF α-Adaptin P-Tyr Starved EGF EGF + Gefitinib * * a b b a Plaques CCPs siControl 0 30 60 120 0 30 60 120 0 30 60 120 si β5-1 si β5-2 Time (min): P-Erk Total-Erk 35 35 Band intensity (A.U.) 0.0 0.5 1.0 1.5 2.0 2.5 siControl 0 min 30 min 60 min 120 min * * si β5-1 si β5-2 c d c TfR-mCherry HGFR-GFP Time after HGF addition (s): 0 570 f Time after EGF addition (s): TfR-mCherry EGFR-GFP 0 470 e P-Tyr TfR-mCherry g GF addition (s): 570 f g e 0 0 TfR-mCherry HGFR-GFP EGFR-GFP P-Tyr Fig. 5 Clathrin-coated plaques locally regulate receptor-dependent Erk signaling. a HeLa cells seeded on collagen-coated glass were starved for 4 h and then treated or not with 10 ng/ml EGF for 5 min alone or added after 30 min preincubation with 10 µM Geﬁtinib. Cells were then ﬁxed and stained for α- adaptin and phosphotyrosines. The arrowhead points to one CCP and the star marks a plaque. Scale bar: 2 μm. b Quantiﬁcation of phosphotyrosines accumulation at plaques or CCPs in the indicated conditions. Results are expressed as mean ± SD (P < 0.05, P < 0.001, one-way analysis of variance— ANOVA. n = 3. NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y Clathrin-coated pla have been consistently observed in diverse cell types but overall received little attention as compared to canonical C possibly reﬂecting the lack of speciﬁc markers for these structu Yet, past investigations have pointed to a possible role of pla in cell adhesion because of the close association of t 2 1.5 1 0.5 0 P-Tyr/α-adaptin fluorescence intensity ratio Blebbistatin Control Cytochalasin Blebbistatin Cytochalasin P-Erk Total-Erk Glass 0.1 kPa Glass 0.1 kPa Collagen Vitronectin 35 35 Average band intensity (% of vitronectin-glass) 0.1 kPa * * Glass 0.1 kPa Glass Collagen Vitronectin * P-Erk Total-Erk siControl siβ5-1 siβ5-2 siControl + Blebbistatin 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 ** * siControl +Blebbistatin Glass Vitronectin Collagen 0 1 2 Cell number Days after seeding *** siControl 0 1 2 Days after seeding siβ5-1 siβ5-2 siαv-1 siαv-2 * 0 1 2 0 1.0.105 1.5.105 2.0.105 2.5 .105 100 80 60 40 20 0 100 80 60 40 20 0 6.105 5.105 4.105 3.105 2.105 1.105 0 siβ5-1 + TfR-mCherry/control Ab siβ5-1 + TfR-mCherry/anti-mCherry Ab * Days after seeding Cell number Cell number 6.105 5.105 4.105 3.105 2.105 1.105 0 c e g d f h i Glass Vitronectin Collagen 0 1 2 Cell number Days after seeding *** 6.105 5.105 4.105 3.105 2.105 1.105 0 g g P-Erk Total-Erk siControl siβ5-1 siβ5-2 siControl + Blebbistatin 35 35 e g c e P-Erk Total-Erk Glass 0.1 kPa Glass 0.1 kPa Collagen Vitronectin 35 35 c P-Erk Total-Erk Gl 0.1 Gl 0.1 35 35 Average band intensity (% of vitronectin-glass) 0.1 kPa * * Glass 0.1 kPa Glass Collagen Vitronectin * P-Erk Total-Erk si siβ siβ s B 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 ** * siControl Blebbistatin Glass Vitronectin Collagen 0 1 2 Cell number Days after seeding * 100 80 60 40 20 0 100 80 60 40 20 0 6.105 5.105 4.105 3.105 2.105 1.105 0 d f 5 Total-Erk 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 * siControl +Blebbistatin Collagen 0 1 2 Cell number Days after seeding 100 80 60 40 20 0 6.105 5.105 4.105 3.105 2.105 1.105 0 f Average band intensity (% of control) siControl siβ5-1 siβ5-2 ** * siControl lebbistatin 100 80 60 40 20 0 f Average band intensity (% of vitronectin-glass) 0.1 kPa * * Glass 0.1 kPa Glass Collagen Vitronectin * 100 80 60 40 20 0 d f d Days after seeding siControl 0 1 2 Days after seeding siβ5-1 siβ5-2 siαv-1 siαv-2 * Cell number 6.105 5.105 4.105 3.105 2.105 1.105 0 h mation of does not assembly on rigid environments results in increased Erk activity and sustained cell proliferation + 0 1 2 0 1.0.105 1.5.105 2.0.105 2.5 .105 siβ5-1 + TfR-mCherry/control Ab siβ5-1 + TfR-mCherry/anti-mCherry Ab * Days after seeding Cell number i 0 1 2 0 1.0.105 1.5.105 2.0.105 2.5 .105 siβ5-1 + TfR-mCherry/control Ab siβ5-1 + TfR-mCherry/anti-mCherry Ab * Days after seeding Cell number i i h assembly on rigid environments results in increased Erk activity and sustained cell proliferation. NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y 2 1.5 1 0.5 0 P-Tyr/α-adaptin fluorescence intensity ratio Blebbistatin Control Cytochalasin * α-Adaptin P-Tyr Blebbistatin Cytochalasin Merge P-Erk Total-Erk Glass 0.1 kPa Glass 0.1 kPa Collagen Vitronectin 35 35 Average band intensity (% of vitronectin-glass) 0.1 kPa * * Glass 0.1 kPa Glass Collagen Vitronectin * P-Erk Total-Erk siControl siβ5-1 siβ5-2 siControl + Blebbistatin 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 ** * siControl Blebbistatin Glass Vitronectin Collagen 0 1 2 Cell number Days after seeding *** 100 80 60 40 20 0 100 80 60 40 20 0 6.105 5.105 4.105 3.105 2.105 1.105 0 a b c e g d f 2 1.5 1 0.5 0 P-Tyr/α-adaptin fluorescence intensity ratio Blebbistatin Control Cytochalasin * α-Adaptin P-Tyr Blebbistatin Cytochalasin Merge P-Erk Total-Erk Glass 0.1 kPa Glass 0.1 kPa Collagen Vitronectin 35 35 Average band intensity (% of vitronectin-glass) 0.1 kPa * * Glass 0.1 kPa Glass Collagen Vitronectin * P-Erk Total-Erk siControl siβ5-1 siβ5-2 siControl + Blebbistatin 35 35 Average band intensity (% of control) siControl siβ5-1 siβ5-2 ** * siControl +Blebbistatin Glass Vitronectin Collagen 0 1 2 Cell number Days after seeding *** siControl siβ5-1 siβ5-2 siαv-1 siαv-2 * 1 0 105 1.5.105 2.0.105 2.5 .105 100 80 60 40 20 0 100 80 60 40 20 0 6.105 5.105 4.105 3.105 2.105 1.105 0 siβ5-1 + TfR-mCherry/control Ab siβ5-1 + TfR-mCherry/anti-mCherry Ab * Cell number Cell number 6.105 5.105 4.105 3.105 2.105 a b c e g d f h i 2 1.5 1 0.5 0 P-Tyr/α-adaptin fluorescence intensity ratio Blebbistatin Control Cytochalasin * α-Adaptin P-Tyr Blebbistatin Cytochalasin Merge a b 2 1.5 1 0.5 0 P-Tyr/α-adaptin fluorescence intensity ratio Blebbistatin Control Cytochalasin * b b a vitronectin-coated glass, a condition that favors the formation of plaques (Fig. 6c, d). However, vitronectin coating does not overcome the need for a stiff environment in order for plaques to assemble (Supplementary Fig. 7a–d) and thus, it did not modulate Erk activation status on 0.1 kPa gels (Fig. 6c, d). In agreement with the role of Erk in controlling cell proliferation, we noticed that cells proliferated faster on vitronectin- as compared to collagen-coated glass (Fig. 6g) and inhibiting plaque formation with β5-speciﬁc or αv-speciﬁc siRNAs reduced proliferation rate (Fig. 6h). In addition, rescuing plaque formation in β5-integrin- depleted cells through immobilizing the TfR on the substrate resulted in an increased proliferation rate (Fig. 6i). NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y assembly on rigid environments results in increased Erk activity and sustained cell proliferation. vitronectin-coated glass, a condition that favors the formation of plaques (Fig. 6c, d). However, vitronectin coating does not overcome the need for a stiff environment in order for plaques to assemble (Supplementary Fig. 7a–d) and thus, it did not modulate Erk activation status on 0.1 kPa gels (Fig. 6c, d). In agreement with the role of Erk in controlling cell proliferation, we noticed that cells proliferated faster on vitronectin- as compared to collagen-coated glass (Fig. 6g) and inhibiting plaque formation with β5-speciﬁc or αv-speciﬁc siRNAs reduced proliferation rate (Fig. 6h). In addition, rescuing plaque formation in β5-integrin- depleted cells through immobilizing the TfR on the substrate resulted in an increased proliferation rate (Fig. 6i). Thus, plaque NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y Thus, plaque assembly on rigid environments results in increased Erk act and sustained cell proliferation. Discussion Overall, we showed that clathrin-coated plaques are mech osensitive signaling platforms that assemble on rigid substrat a consequence of frustrated endocytosis. Results Number of structures analyzed per condition: Starved 160 plaques; EGF: 452 plaques, EGF + Geﬁtinib: 301 plaques, CCPs/EGF: 200 pits). c Western-blot analysis of phospho-Erk (P-Erk) levels in starved HeLa cells treated with control or β5-speciﬁc siRNAs as indicated, and stimulated for the indicated time with 10 ng/ml EGF. Total-Erk was used as a loading control (representative image of three independent experiments). d Densitometry analysis of bands obtained in western-blots as in c. Results are expressed as mean ± SD (P < 0.05, ANOVA. n = 3). e HeLa cells treated with β5-speciﬁc siRNAs were transfected with plasmids encoding for TfR-mCherry and EGFR-GFP and seeded on anti-mCherry antibodies-coated glass. Cells were serum- starved for 4 h and then treated with 10 ng/ml EGF for 470 s. Scale bar: 0.5 μm. Arrowheads point to plaques positive for EGFR-GFP. f HeLa cells treated with β5-speciﬁc siRNA were transfected with plasmids encoding for TfR-mCherry and HGFR-GFP and seeded on anti-mCherry antibodies-coated glass. Cells were serum-starved for 4 h and then treated with 50 ng/ml HGF for 570 s. Scale bar: 0.5 μm. Arrowheads point to plaques positive for HGFR-GFP. g HeLa cells treated with β5-speciﬁc siRNA were transfected with plasmids encoding for TfR-mCherry and seeded on anti-mCherry antibodies-coated glass. Cells were serum-starved for 4 h and then treated with 10 ng/ml EGF for 5 min prior to ﬁxation and staining for phosphotyrosines. Scale bar: 1.5 μm. Arrowheads point to plaques positive for P-Tyr forces36,37. Indeed, we observed that phosphorylated-Erk levels were reduced on soft as compared to hard environment (Fig. 6c, d) and inhibiting myosin-dependent contractility using blebbis- tatin did not modulate Erk activation status (Fig. 6e, f). Thus, Erk activation on rigid environments depends on the presence of clathrin-coated plaques (Fig. 4) but not on cell contractility. Accordingly, Erk activity was increased when cells were seeded on substrate rigidity. Because their formation does not require a functional acto-myosin machinery (Fig. 1), plaques may represent a distinct, contractility-independent mechanotransduction sys- tem. Along this line, blebbistatin or cytochalasin treatments did not prevent phosphotyrosine accumulation at plaques upon EGF stimulation (Fig. 6a, b). Erk activity has been reported to be mechanosensitive in a manner that is independent of acto-myosin NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 8 ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y Fig. 6 Signaling at plaques is contractility-independent and regulate cell proliferation. a HeLa cells on collagen-coated glass were treated for 1 h with 10 µM Blebbistatin or 10 µM Cytochalasin D prior to be ﬁxed and stained for α-adaptin (red) and phosphotyrosines (P-Tyr, green). Scale bar: 3 μm. b, Fig. 6 Signaling at plaques is contractility-independent and regulate cell proliferation. a HeLa cells on collagen-coated glass were treated for 1 h with 10 µM Blebbistatin or 10 µM Cytochalasin D prior to be ﬁxed and stained for α-adaptin (red) and phosphotyrosines (P-Tyr, green). Scale bar: 3 μm. b, Quantiﬁcation of phosphotyrosines accumulation at plaques in cells treated as in a, as indicated (P < 0.05, one-way analysis of variance—ANOVA. Control: 402 plaques from n = 3 independent experiments; Blebbistatin: 303 plaques from n = 3 independent experiments; Cytochalasin: 302 plaques from n = 3 independent experiments). c Western-blot analysis of phospho-Erk (P-Erk) levels in HeLa cells cultured on collagen- or vitronectin-coated glass or 0.1 kPa polyacrylamide gels, as indicated. Total-Erk was used as a loading control (representative image of three independent experiments). d Densitometry analysis of bands obtained in western-blots as in c (P < 0.05, ANOVA. n = 3). e Western-blot analysis of phospho-Erk (P-Erk) levels in HeLa cells growing on collagen-coated glass and treated with control or integrin β5-speciﬁc siRNAs and incubated or not with 10 µM Blebbistatin for 1 h, as indicated. Total-Erk was used as a loading control (representative image of three independent experiments). f Densitometry analysis of bands obtained in western-blots as in e (P < 0.05, P < 0.01, ANOVA. n = 5). g Equal numbers of HeLa cells were plated on non-coated glass (open squares), or on collagen-coated (purple open circles) or vitronectin-coated glass (black open circles), as indicated. Cells were harvested and counted 24 and 48 h after plating (P < 0.001, ANOVA. n = 3). h HeLa cells treated with control (circles), β5-speciﬁc (triangles and crosses), or αv-speciﬁc siRNAs (diamonds and hexagons) for 48 h were seeded on vitronectin-coated glass in equal numbers. 24 and 48 h later, cells were harvested and counted (P < 0.05, ANOVA. n = 3). i HeLa cells treated with β5- speciﬁc siRNA were transfected with a plasmid encoding TfR-mCherry and seeded in equal numbers on glass coated with either anti-mCherry (black circles) or control antibodies (open circles). NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y 23966) at 1 mg/ml was used to transfect 50% conﬂuent cells in a 6 well plate according to the following protocol: 2 µg of DNA were added to 100 µl of OptiMEM, followed by addition of 4 µl of PEI, vortex and incubation for 10 minutes at RT prior to add the mix to the cells. Plaques have also been proposed to have a role in signal transduction11,12 although direct evidences have been missing because of the lack of tools to speciﬁcally impair plaque assembly. How plaques regulate signaling is not clear yet. Others have proposed that CCSs are platforms for signaling31, independently of endocytosis. The capacity of CCSs to compartmentalize the plasma membrane is certainly an important, underestimated factor that will need to be investigated in the future. In that respect, large CCSs such as plaques may be more efﬁcient in transducing the signal. Antibodies and drugs. Mouse monoclonal anti-clathrin heavy chain (CHC – Cat. Nr. 610500) antibody and mouse monoclonal anti-FAK (Cat. Nr. 610088) antibody were obtained from BD Transduction Laboratories (Becton Dickinson France SAS, Le Pont-De-Claix, France). Rabbit polyclonal anti-actin was from Sigma (Cat. Nr. A5060). Rabbit polyclonal anti-α-adaptin antibodies (M300), rabbit polyclonal DAB2 (Cat. Nr. sc-13982), Goat polyclonal anti-Talin1 (C20) were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Integrin αvβ5 (Cat. Nr. MAB1961) and P-Tyrosine (Cat. Nr. 05-321) were obtained from Millipore. Mouse monoclonal anti-α-adaptin (Cat. Nr. ab2807) was purchased from Abcam. Numb (Cat.Nr. 2756), Integrin β5 (Cat. Nr. 3629), Tot-ERK1/2 (Cat. Nr. 9102) and P- ERK1/2 (Cat. Nr. 9101) were purchased from Cell Signalling. Tot-ERK1/2 (Cat. Nr. 13-6200) was purchased from Thermo Fisher. Antibodies used for western-blot analyses were diluted at 1:1000 in PBS-0.1% Tween-5% BSA or 5% non-fat dried milk. For immunoﬂuorescences, antibodies were diluted 1:200 in PBS-0.3% BSA. HRP-conjugated anti-mouse and anti-rabbit antibodies for western-blot were from Jackson ImmunoResearch Laboratories (West Grove, PA, USA) and were used at a dilution of 1:3000. Alexa-conjugated antibodies as well as Cy3 and Cy5-conjugated antibodes were from Molecular Probes (Invitrogen) and were used at a dilution of 1:200. For expansion microscopy, the following secondary antibodies were used: Donkey anti-Mouse IgG (H + L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor® 488, Thermo Fisher (Cat. Nr. A21202), Goat Anti-Rabbit IgG H&L, Alexa Fluor® 568, AbCam (Cat. Nr. ab175471), CF®633 Donkey Anti-Mouse IgG (H + L), highly cross-adsorbed, Biotium (Cat. Nr. NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y pLenti-MetGFP was a gift from David Rimm (Addgene plasmid #37560). mEmerald-Alpha-V- Integrin was a gift from Michael Davidson (Addgene plasmid #53985). cells engineered to expressed an endogenous GFP-tagged μ2 subunit (a gift from D. Drubin, University of California-Berkeley, California, USA) were grown in DMEM Glutamax supplemented with 10% foetal calf serum at 37 °C in 5% CO2. All cell lines have been tested for mycoplasma contaminations. For most experiments, cells were grown on substrates coated with a 50 µg/ml solution of collagen-I (Thermo Fisher Scientiﬁc—Cat. Nr. A10483-01) unless otherwise stated. DNA sequence encoding full-length β5-Integrin was obtained by PCR by using the cDNA of human β5-Integrin, a gift from Raymond Birge (Addgene plasmid #14996), as a template. PCR fragments with engineered ﬂanking restriction sites (XhoI/BamHI) were subcloned into the multi-cloning sites of pEGFP-N1 (Clontech) to encode an in-frame fusion protein with the carboxy-terminal EGFP-tag (pEGFP-N1- Integrin β5). SiRNA-resistant Integrin β5 was obtained by site-directed mutagenesis of the β5-encoding cDNA at the following positions: A69T, T72G, C73G, C75G (silent mutations—resistant to siβ5-1). mCherry-TfR was a gift from Michael Davidson (Addgene plasmid #55144). GFP-Erk2 was a gift from Dr.Hesso Farhan. EGFR- GFP was a gift from Alexander Sorkin (Addgene plasmid # 32751). pLenti-MetGFP was a gift from David Rimm (Addgene plasmid #37560). mEmerald-Alpha-V- Integrin was a gift from Michael Davidson (Addgene plasmid #53985). q encoding full-length β5-Integrin was obtained by PCR by using the cDNA of human β5-Integrin, a gift from Raymond Birge (Addgene plasmid #14996), as a template. PCR fragments with engineered ﬂanking restriction sites (XhoI/BamHI) were subcloned into the multi-cloning sites of pEGFP-N1 (Clontech) to encode an in-frame fusion protein with the carboxy-terminal EGFP-tag (pEGFP-N1- Integrin β5). SiRNA-resistant Integrin β5 was obtained by site-directed mutagenesis of the β5-encoding cDNA at the following positions: A69T, T72G, C73G, C75G (silent mutations—resistant to siβ5-1). mCherry-TfR was a gift from Michael Davidson (Addgene plasmid #55144). GFP-Erk2 was a gift from Dr.Hesso Farhan. EGFR- GFP was a gift from Alexander Sorkin (Addgene plasmid # 32751). pLenti-MetGFP was a gift from David Rimm (Addgene plasmid #37560). mEmerald-Alpha-V- Integrin was a gift from Michael Davidson (Addgene plasmid #53985). Plasmids were transfected 24 h after cell plating using either Lipofectamine 3000 according to the manufacturer’s instructions or electroporating cells in suspension using AMAXA nucleofector Kit V according to the manufacturer’s instructions. Alternatively, linear PEI (MW 25.000—Polysciences Cat. Nr. NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y 20124) and were used at a dilution of 1:100. The anti-mCherry and P-FAK Tyr576 (western-blot) antibodies were obtained from the recombinant antibody platform of the Curie Institute, Paris and was used at a dilution of 1:1000. P-FAK Tyr397 (IF) was a gift from Monique Arpin and was used at a dilution of 1:200. Alexa Fluor® 488 EGF complex was obtained from Thermo Fisher (Cat. Nr. E-13345). Rat tail Collagen-I (Cat. Nr. A10483-01) and Vitronectin (Cat. Nr. A14700) were purchased from GIBCO. Human recombinant EGF (Cat. Nr. E9644) and HGF (Cat. Nr. 1404) were pur- chased from Sigma. Blebbistatin (Cat. Nr. B0560), Geﬁtinib (Cat. Nr. CDS022106) Because the formation of these structures does not rely on a functional acto-myosin network, we propose that plaques represent an alternative, contractility-independent rigidity sensing mechan- ism. The use of different strategies to sense the environment’s elasticity may reﬂect the importance of this mechanical parameter that has fundamental biological consequences, ranging from cell proliferation to differentiation. Because αvβ5-integrin and vitro- nectin levels, together with tissue rigidity, are often modulated in different physio-pathological conditions38–42, CCSs-dependent mechanotransduction may have a universal role in regulating the cell response to changing environmental conditions. Cell lines and constructs. HeLa cells (a gift from P. Chavrier, Institut Curie, Paris, France; ATCC CCL-2), genome-edited HeLa cells engineered to expressed an endogenous GFP-tagged or mCherry-tagged μ2 subunit, HepG2 cells (ATCC HB- 8065), Caco-2 cells (ATCC HTB-37), MDA-MB-231 cells (a gift from P. Chavrier, Institut Curie, Paris, France; ATCC HTB-26), or genome-edited MDA-MB-231 Discussion Overall, we showed that clathrin-coated plaques are mechan- osensitive signaling platforms that assemble on rigid substrates as a consequence of frustrated endocytosis. Clathrin-coated plaques have been consistently observed in diverse cell types but have overall received little attention as compared to canonical CCPs, possibly reﬂecting the lack of speciﬁc markers for these structures. Yet, past investigations have pointed to a possible role of plaques in cell adhesion because of the close association of these NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 9 9 NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y 24 and 48 h later, cells were harvested and counted (*P < 0.05, ANOVA. n = 3). All results are expressed as mean ± SD mean ± SD structures with the substrate16, and because they were shown to be enriched in integrins, and in particular in β5-integrin10. We report here that αvβ5-integrin is required for plaque formation and can be considered as a marker of these structures although it can also be found at FAs in some cases. αvβ5 engagement with the substrate prevents the clathrin machinery to bud. This is reminiscent of the β1-integrin-mediated frustrated endocytosis of clathrin-coated tubes found on collagen ﬁbers9. However, in this later case, frustration was only transient while plaques can be extremely long-lived. It has recently been reported that actin polymerization can participate in plaque disassembly upon LPA- receptor activation12. It is possible that actin polymerization provides an extra force needed for the clathrin machinery to overcome αvβ5-mediated adhesion and to bud at plaques. Along this line, inhibiting actin polymerization resulted in a slight increase in plaque size (Fig. 1e). Together, these data suggest that a force balance between endocytosis and adhesion regulates the dynamics of these structures. The equilibrium between these forces is modulated by the rigidity of the substratum so that plaques can only assemble on relatively stiff substrates. cells engineered to expressed an endogenous GFP-tagged μ2 subunit (a gift from D. Drubin, University of California-Berkeley, California, USA) were grown in DMEM Glutamax supplemented with 10% foetal calf serum at 37 °C in 5% CO2. All cell lines have been tested for mycoplasma contaminations. For most experiments, cells were grown on substrates coated with a 50 µg/ml solution of collagen-I (Thermo Fisher Scientiﬁc—Cat. Nr. A10483-01) unless otherwise stated. DNA sequence encoding full-length β5-Integrin was obtained by PCR by using the cDNA of human β5-Integrin, a gift from Raymond Birge (Addgene plasmid #14996), as a template. PCR fragments with engineered ﬂanking restriction sites (XhoI/BamHI) were subcloned into the multi-cloning sites of pEGFP-N1 (Clontech) to encode an in-frame fusion protein with the carboxy-terminal EGFP-tag (pEGFP-N1- Integrin β5). SiRNA-resistant Integrin β5 was obtained by site-directed mutagenesis of the β5-encoding cDNA at the following positions: A69T, T72G, C73G, C75G (silent mutations—resistant to siβ5-1). mCherry-TfR was a gift from Michael Davidson (Addgene plasmid #55144). GFP-Erk2 was a gift from Dr.Hesso Farhan. EGFR- GFP was a gift from Alexander Sorkin (Addgene plasmid # 32751). ARTICLE The following siRNAs were used: β5-1, 5′-GCUCGCAGGUCUCAACA UA-3′; β5-2, 5′-GGUCUAAAGUGGAGUUGUC-3′; μ2-adaptin, 5′-AAGUGGA UGCCUUUCGGGUCA-3′; Clathrin heavy chain (CHC), 5′-GCUGGGAA AACUCUUCAGATT-3′; αv-1, 5′-CCUCUGACAUUGAUUGUUA-3′; αv-2, 5′-C CGAAACAAUGAAGCCUUA-3′; DAB2, 5′-GAGCAUGAACAUCCAGAU AATT-3′; Numb, 5′-GAUAGUCGUUGGUUCAUCATT-3′; Integrin β1 ON-TARGET plus SMART pool (Dharmacon L-004506-00), Integrin β3 siGENOME Human ITGB3 siRNA (Dharmacon M-004124-02); Talin1, 5′-AC AAGAUGGAUGAAUCAAUUUU-3′; non-targeting siRNAs (siControl), ON- TARGET plus Non-Targeting SMART pool siRNAs (Dharmacon D-001810-01). Antibody coating. Coverslips were incubated in a sterile solution of sodium bicarbonate (0.1 M, pH 9.5) for 1 h at 37 °C. Coverslips were then incubated with poly-L-lysine 0.01% (Sigma, diluted in water) for 1 h at 37 °C, washed once with PBS, dried and incubated overnight at 37 °C with the desired antibody (mCherry antibody or anti-rabbit A647 from Invitrogen; ﬁnal concentration of 1.25 µg/ml) diluted in the bicarbonate solution. Acrylamide gels were ﬁrst coated with Protein G (Sigma) diluted in sodium bicarbonate (0.1 M, pH 9.5) to a ﬁnal concentration of 5 µg/ml for 1 h at 37 °C, then treated with poly-L-lysine 0.01% (Sigma, diluted in water) for 1 h at 37 °C, washed once with PBS, dried and incubated overnight at 37 °C with the desired antibody. Acrylamide gels of controlled stiffness. Coverslips or ﬂuorodishes were incu- bated with APTMS (3‐aminopropyltrimethoxysilane) for 15 min at RT, then washed extensively with water and incubated for 30 min with Glutaraldehyde 0.5% in PBS and washed again with water. Acrylamide 40% and bis-acrylamide 2% were mixed (respectively 5% and 0.04% for 0.1 kPa gels, 7.5% and 0.06% for 5 kPa gels, 18% and 0.4% for 31.7 kPa gels, and 16% and 0.96% for 80 kPa) with PBS, APS 10% and TEMED. 9 µl of this solution were polymerized on the treated glass. Gels were washed with PBS, followed by a 30 min incubation with 300 μl 50 mM Hepes pH 7.5 + 100 μl sulfo-sampah (1 mg/ml in 50 mM Hepes pH 7.5) + 100 μl EDC (10 mg/ml in 50 mM Hepes pH 7.5). Gels were subsequently cross-linked under UV light for 10 min, washed and incubated with 50 µg/ml collagen-I for 1 h at 37 °C. Elasticity of the different gels was controlled by Atomic Force Microscopy as indicated in Betz et al.44 The generation of 80 kPa gels was performed according to a previously published protocol45. Reverse transcription and polymerase chain reaction. Total RNA was isolated from HeLa cells by using the Qiagen RNeasy kit. ARTICLE ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y ﬂuorescence intensity of α-adaptin and of αvβ5 or of P-Tyr staining. Values were subsequently normalized to the average ﬂuorescence of the α-adaptin channel and normalized intensity values were used for the analysis. At least 100 structures (50 for the HGF stimulated cells) per condition and per experiments were counted in three independent experiments. Data are expressed as mean ± SD. For the data in Supplementary Figure 9, phosphotyrosine (P-Tyr)-positive structures were segmented in ImageJ and mean ﬂuorescence intensity was measured in at least 15 cells per condition in three independent experiments. Data are expressed as mean ± SD. For measuring ERK-GFP intensity in the nucleus, cells cytoplasm and nuclei were manually segmented, GFP-associated integrated intensity was measured in the cytoplasm and in the nucleus and the nucleus/cytosol ratio was calculated for every cell. At least 15 cells per condition were measured in at least three independent experiments. Results are expressed as the relative enrichment over the no-enrichment condition (nucleus/cytosol signal = 1) ± SD. and Cytochalasin D (Cat. Nr. 8273) were purchased from Sigma. Latrunculin A (Cat. Nr. ab144290) was purchased from AbCam. Blebbistatin, Cytochalasin D, and Geﬁtinib were used at a ﬁnal concentration of 10 μM and Latrunculin A at 5 µM. Cilengitide was purchased from Selleckchem (Cat. Nr. S7077) and used at a ﬁnal concentration of 10 μM. For western-blot experiments, cells were lysed in ice-cold MAPK buffer (100 mM NaCl, 10 nM EDTA, 1% IGEPAL® CA-630, 0.1% SDS, 50 mM TRIS–HCl pH 7.4) supplemented with protease and phosphatase inhibitors. RNA interference. For siRNA depletion, 200,000 cells were plated in six well plates. After 24 h, cells were treated with the indicated siRNA (30 nM) using RNAimax (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruction. Protein depletion was maximal after 72 h of siRNA treatment as shown by immunoblotting analysis with speciﬁc antibodies. To deplete CHC or µ2-adaptin, cells were transfected once as described above and then a second time, 48 h later, with the same siRNAs. In this case, cells were analyzed 96 h after the ﬁrst trans- fection. ARTICLE 24 h after plating, medium was aspirated and replaced with 4 ml of stabilization buffer (3 mM EGTA, 5 mM MgCl2, 70 mM KCl, 30 mM HEPES pH 7.4). To unroof cells, 1 ml of 2% PFA was pipetted directly on the cells with a pipette kept perpendicular to the glass at 1 mm from the surface. The coverslips were then immediately transferred to 2% glutaraldehyde for 30 min and then left in Cacodylate buffer. For expansion microscopy43, cells grown on coverslips were immunostained and then treated for 10 min with glutaraldehyde 0.25%. The coverslips were washed and incubated in monomer solution for 1 min (1 × PBS, 2 M NaCl, 2.5% (w/w) acrylamide, 0.15% (w/w) N,N′-methylenebisacrylamide, 8.625% (w/w) sodium acrylate). 120 µl drops of Gel solution (1.3 µl of TEMED, 1.3 µl of APS 10%, 197.4 µl of monomer solution) were deposited on paraﬁlm and the coverslips were deposited on top. After 30 min, gels were moved to the digestion solution (Tris- acetate-EDTA (TEA), 0.5% Triton-×100, 0.8 M guanidine HCl, Proteinase K—8 U/ ml added fresh before use) for 30 more minutes. Gel were then manually separated from the coverslips and incubated in ddH2O for at least 2 h, until full expansion (4.167 expansion factor). All quantiﬁcations on immunoﬂuorescence images were done with FIJI after background subtraction. To measure the size of CCSs on expansion microscopy images, α-adaptin-positive structures were manually delimited and the area was measured using ImageJ. Dot-like objects with area lower or equal to 0.0175 µm2 (radius 0.075 µm) were classiﬁed as clathrin-coated pits (CCPs), whereas irregularly shaped objects with an area larger than 0.0175 µm2 were classiﬁed as clathrin-coated plaques. The area of the adherent side of the cell was also measured in order to calculate the percentage of adherent surface occupied by plaques. Data are expressed as mean ± SD. Scanning electron microscopy observations of unroofed HeLa cells were performed as previously described46. Brieﬂy, ﬁxed HeLa cells were dehydrated in a series of increasing ethanol concentrations. Critical point was dried using carbon dioxide in a Leica EMCPD300. After coating with 2 nm platinum, cells were examined with a FEI Quanta FEG250 scanning electron microscope. Spinning disk microscopy of live cells. ARTICLE Data are expressed as mean ± SD. 43 ll l d Immunoﬂuorescence microscopy and ﬂuorescence quantiﬁcation. Cells were ﬁxed in ice-cold methanol unless stated otherwise and processed for immuno- ﬂuorescence microscopy by using the indicated antibodies. For anti- Phosphotyrosine staining (P-Tyr), cells were brieﬂy extracted for 30 s using 0.1% Triton prior to ﬁxation. Depending on experiments imaged either through a ×100 1.40NA UPlanSApo objective lens of a wide-ﬁeld IX73 microscope (Olympus) equipped with an Orca-Flash2.8 CMOS camera (Hamamatsu) and steered by CellSens Dimension software (Olympus), or by spinning microscopy (see description below). Surface levels of EGFR were measured by incubating the cells on ice with A488-EGF for 1 h. Cells were subsequently ﬁxed with PFA 4% and images were taken. Total ﬂuorescence was measured on at least 50 cells per con- dition in three independent experiments. Data are expressed as mean ± SD. For expansion microscopy43, cells grown on coverslips were immunostained and then treated for 10 min with glutaraldehyde 0.25%. The coverslips were washed and incubated in monomer solution for 1 min (1 × PBS, 2 M NaCl, 2.5% (w/w) acrylamide, 0.15% (w/w) N,N′-methylenebisacrylamide, 8.625% (w/w) sodium acrylate). 120 µl drops of Gel solution (1.3 µl of TEMED, 1.3 µl of APS 10%, 197.4 µl of monomer solution) were deposited on paraﬁlm and the coverslips were deposited on top. After 30 min, gels were moved to the digestion solution (Tris- acetate-EDTA (TEA), 0.5% Triton-×100, 0.8 M guanidine HCl, Proteinase K—8 U/ ml added fresh before use) for 30 more minutes. Gel were then manually separated from the coverslips and incubated in ddH2O for at least 2 h, until full expansion (4.167 expansion factor). All quantiﬁcations on immunoﬂuorescence images were done with FIJI after background subtraction. To measure the size of CCSs on expansion microscopy images, α-adaptin-positive structures were manually delimited and the area was measured using ImageJ. Dot-like objects with area lower or equal to 0.0175 µm2 (radius 0.075 µm) were classiﬁed as clathrin-coated pits (CCPs), whereas irregularly shaped objects with an area larger than 0.0175 µm2 were classiﬁed as clathrin-coated plaques. The area of the adherent side of the cell was also measured in order to calculate the percentage of adherent surface occupied by plaques. Data are expressed as mean ± SD. Electron microscopy of unroofed cells. HeLa cells were plated on Poly-L-lysine coated coverslips in 12 well plates. ARTICLE cDNA was prepared from 100 ng total RNA by using the high capacity cDNA reverse transcription kit (Applied Biosystems) according to the manufacturer’s instructions. Q-PCR was performed by using Fast SYBR green PCR MasterMix (Applied Biosystems). Reverse tran- scription and polymerase chain reaction (RT-PCR) were run on a QuantStudio™7 Flex Real-Time PCR system (Applied Biosystems). Expression of each gene was normalized to the expression of GAPDH (primers: Fw: 5′-CTTTTGCGTCGC CAGCCGAG-3′; Rev 5′- CCAGGCGCCCAATACGACCA-3′). Speciﬁc primers for Integrin β5 were designed using the free online tool PrimerExpress® (Fw: 5′-CT GGAACAACGGTGGAGATT-3′; Rev: 5′-TACCCCATCTTGGCAGGTAG-3′). The relative amount of integrin β5 cDNA was normalized to the cDNA relative amount of the housekeeping gene GAPDH. Values obtained in the control were set to 1 and the other values expressed as percentage of control. The reaction was performed three times and the data are expressed as mean ± SD. Micropatterns. Coverslips were cleaned by washing in 70% ethanol and sub- sequent irradiated under UV light for 5 min. The activated side of the coverslips was then covered with 0.1 mg/ml PLL-g-PEG (Surface Solutions, Zurich) for 1 h at RT. Coverslips were then washed twice in water and ring-shaped areas were exposed to deep UV during 5 min using a photomask. Coverslips were recovered and coated with Alexa-568-labeled Collagen-I at 50 µg/ml overnight at 37 °C. The following day, 50-80,000 HeLa cells were plated on micropatterns in complete medium supplemented with penicillin and streptomycin. Cells were either ﬁxed or imaged 5–6 h after plating. Eight to ten cells per experiment were measured per experiments in three independent experiments. Data are expressed as mean ± SD. Immunoﬂuorescence microscopy and ﬂuorescence quantiﬁcation. Cells were ﬁxed in ice-cold methanol unless stated otherwise and processed for immuno- ﬂuorescence microscopy by using the indicated antibodies. For anti- Phosphotyrosine staining (P-Tyr), cells were brieﬂy extracted for 30 s using 0.1% Triton prior to ﬁxation. Depending on experiments imaged either through a ×100 1.40NA UPlanSApo objective lens of a wide-ﬁeld IX73 microscope (Olympus) equipped with an Orca-Flash2.8 CMOS camera (Hamamatsu) and steered by CellSens Dimension software (Olympus), or by spinning microscopy (see description below). Surface levels of EGFR were measured by incubating the cells on ice with A488-EGF for 1 h. Cells were subsequently ﬁxed with PFA 4% and images were taken. Total ﬂuorescence was measured on at least 50 cells per con- dition in three independent experiments. NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications Methods Cell lines and constructs. HeLa cells (a gift from P. Chavrier, Institut Curie, Paris, France; ATCC CCL-2), genome-edited HeLa cells engineered to expressed an endogenous GFP-tagged or mCherry-tagged μ2 subunit, HepG2 cells (ATCC HB- 8065), Caco-2 cells (ATCC HTB-37), MDA-MB-231 cells (a gift from P. Chavrier, Institut Curie, Paris, France; ATCC HTB-26), or genome-edited MDA-MB-231 10 NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications ARTICLE Butcher, D. T., Alliston, T. & Weaver, V. M. A tense situation: forcing tumour progression. Nat. Rev. Cancer 9, 108–122 (2009). 14. Saffarian, S., Cocucci, E. & Kirchhausen, T. Distinct dynamics of endocytic clathrin-coated pits and coated plaques. PLoS Biol. 7, e1000191 (2009). 15. Montagnac, G. et al. αTAT1 catalyses microtubule acetylation at clathrin- coated pits. Nature 502, 567 (2013). p 16. Maupin, P. & Pollard, T. D. Improved preservation and staining of HeLa cell actin ﬁlaments, clathrin-coated membranes, and other cytoplasmic structures by tannic acid-glutaraldehyde-saponin ﬁxation. J. Cell. Biol. 96, 51–62 (1983). Stimulated-emission-depletion microscopy. Image acquisitions were performed with a ×100 oil immersion objective (NA 1.4) through gated Continuous Wave (gCW) Stimulated-emission-depletion (STED) imaging (TCS SP8-3×; Leica Microsystems) with optimized parameters for Alexa Fluor 568 detection. Samples (zoom 4, pixel size = 14 nm) were excited with a 575 nm wavelength of a super- continuum laser and a 660-nm laser for depletion. For Alexa Fluor 568, 30% AOTF, conventional scanner (400 Hz, Line Average 2, Accumulation 4) and 50% of depletion lasers were used. Fluorescence (585–630 nm) was collected with a hybrid detector (Gain 30%) in the gated mode (0.5–6 ns) and a pinhole for 1 Airy Unit. Deconvolution of raw data from STED imaging was obtained through image processing with Huygens professional 4.5.1 software. 17. Tawil, N., Wilson, P. & Carbonetto, S. Integrins in point contacts mediate cell spreading: factors that regulate integrin accumulation in point contacts vs. focal contacts. J. Cell. Biol. 120, 261–271 (1993). 18. Vassilopoulos, S. et al. Actin scaffolding by clathrin heavy chain is required for skeletal muscle sarcomere organization. J. Cell. Biol. 205, 377–393 (2014). 19. Cheresh, D. A., Smith, J. W., Cooper, H. M. & Quaranta, V. A novel vitronectin receptor integrin (αvβx) is responsible for distinct adhesive properties of carcinoma cells. Cell 57, 59–69 (1989). 20. Smith, J. W., Vestal, D. J., Irwin, S. V., Burke, T. A. & Cheresh, D. A. Puriﬁcation and functional characterization of integrin alpha v beta 5. An adhesion receptor for vitronectin. J. Biol. Chem. 265, 11008–11013 (1990). Cell proliferation assays. Cells were seeded at a density of 100,000 per well in 12 wells plates coated with 50 µg/ml collagen-I, or 0.5 µg/ml vitronectin, or uncoated. Twenty-four and forty-eight hours later, cells were harvested and counted. All conditions were plated in duplicates in at least three independent experiments. Data are expressed as mean ± SD. 21. ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y dynamic structures continuously nucleate and disappear during the movie. For this reason, all quantiﬁcations of CCS dynamics represent the relative number of static or dynamic events detectable at the plasma membrane at a given time point. At least 1000 CCSs from at least ﬁve cells per conditions and per experiments were tracked in 3–5 independent experiments. Data are expressed as mean ± SD. 7. Choquet, D., Felsenfeld, D. P. & Sheetz, M. P. Extracellular matrix rigidity causes strengthening of integrin–cytoskeleton linkages. Cell 88, 39–48 (1997). 8. Balaban, N. Q. et al. Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates. Nat. Cell Biol. 3, 466 (2001). 9. Elkhatib, N. et al. Tubular clathrin/AP-2 lattices pinch collagen ﬁbers to support 3D cell migration. Science 356, https://doi.org/10.1126/science. aal4713 (2017). 9. Elkhatib, N. et al. Tubular clathrin/AP-2 lattices pinch c p p p To measure Auxillin bursts, cells were imaged at 1 frame every 2 s for 3 min. Then, trypsin to a ﬁnal concentration of 0.0375% was added and imaging was immediately started until disappearance of plaques. Ninety structures (plaques) from a total of six cells were analyzed from two independent experiments. Data are expressed as mean ± SD. , p g support 3D cell migration. Science 356, https://doi.org/10.1126/science. aal4713 (2017). 10. De Deyne, P. G. et al. The vitronectin receptor associates with clathrin-coated membrane domains via the cytoplasmic domain of its beta5 subunit. J. Cell. Sci. 111, 2729–2740 (1998). 11. Grove, J. et al. Flat clathrin lattices: stable features of the plasma membrane. Mol. Biol. Cell 25, 3581–3594 (2014). Total internal reﬂection ﬂuorescence microscopy. For total internal reﬂection ﬂuorescence microscopy (TIRF), HeLa cells transfected with the indicated plasmids were imaged through a ×100 1.49 NA TIRF objective lens on a Nikon TE2000 (Nikon France SAS, Champigny sur Marne, France) inverted microscope equipped with a QuantEM EMCCD camera (Roper Scientiﬁc SAS, Evry, France/Photo- metrics, AZ, USA), a dual output laser launch, which included 491 and 561 nm 50 mW DPSS lasers (Roper Scientiﬁc), and driven by Metamorph 7 software (MDS Analytical Technologies, Sunnyvale, CA, USA). A motorized device driven by Metamorph allowed the accurate positioning of the illumination light for evanes- cent wave excitation. 12. Leyton-Puig, D. et al. Flat clathrin lattices are dynamic actin-controlled hubs for clathrin-mediated endocytosis and signalling of speciﬁc receptors. Nat. Commun. 8, 16068 (2017). ( ) 13. ARTICLE Calderwood, D. A. et al. Integrin β cytoplasmic domain interactions with phosphotyrosine-binding domains: A structural prototype for diversity in integrin signaling. Proc. Natl Acad. Sci. USA 100, 2272–2277 (2003). integrin signaling. Proc. Natl Acad. Sci. USA 100, 2272–2277 (2 22. Hayman, E. G., Pierschbacher, M. D., Suzuki, S. & Ruoslahti, E. Vitronectin— a major cell attachment-promoting protein in fetal bovine serum. Exp. Cell Res. 160, 245–258 (1985). Statistical analyses. Statistical analyses in Figs. 1, 2c, e, 3g, 4d, f, 6b, f, g, h, i, Supplementary Figs. 2, 4, 5d, 6b, c, e, 7d, g, 8, 9d, 11a have been performed using one-way analysis of variance (ANOVA) followed by all pairwise multiple com- parison procedure (Holm–Sidak method). Statistical analyses in Figs. 4b, 5b, d, 6d, h and Supplementary Figs. 6g, 7b, 9f have been performed using ANOVA followed by all pairwise multiple comparison procedure (Neuman–Keuls method). Mann–Whitney rank sum test was used in Fig. 3c and Supplementary Data Fig. 5c. Student’s t-test was used in Fig. 2b and Supplementary Figs. 10 and 11c. All data are presented as mean of at least three independent experiments ± SD. All statistical analyses were performed using SigmaPlot software. 23. Gebb, C., Hayman, E. G., Engvall, E. & Ruoslahti, E. Interaction of vitronectin with collagen. J. Biol. Chem. 261, 16698–16703 (1986). 24. Massol, R. H., Boll, W., Grifﬁn, A. M. & Kirchhausen, T. A burst of auxilin recruitment determines the onset of clathrin-coated vesicle uncoating. Proc. Natl Acad. Sci. USA 103, 10265–10270 (2006). 25. Dechantsreiter, M. A. et al. N-methylated cyclic RGD peptides as highly active and selective αVβ3 integrin antagonists. J. Med. Chem. 42, 3033–3040 (1999). Mann–Whitney rank sum test was used in Fig. 3c and Supplementary Data Fig. 5c. Student’s t-test was used in Fig. 2b and Supplementary Figs. 10 and 11c. All data are presented as mean of at least three independent experiments ± SD. All statistical analyses were performed using SigmaPlot software. 26. Marinelli, L., Gottschalk, K.-E., Meyer, A., Novellino, E. & Kessler, H. Human integrin αvβ5: homology modeling and ligand binding. J. Med. Chem. 47, 4166–4177 (2004). 27. Xiong, J.-P. et al. Crystal structure of the complete integrin αVβ3 ectodomain plus an α/β transmembrane fragment. J. Cell. Biol. 186, 589–600 (2009). Data availability 28. Qian, J. & Gao, H. Soft matrices suppress cooperative behaviors among receptor-ligand bonds in cell adhesion. PLoS ONE 5, e12342 (2010). The authors declare that all data supporting the ﬁndings of this study are available within the article and its supplementary information ﬁles or from the corresponding author upon reasonable request. 29. Du, J. et al. Integrin activation and internalization on soft ECM as a mechanism of induction of stem cell differentiation by ECM elasticity. Proc. Natl Acad. Sci. USA 108, 9466–9471 (2011). Received: 26 December 2017 Accepted: 27 August 2018 30. Lampe, M., Pierre, F., Al-Sabah, S., Krasel, C. & Merriﬁeld, C. J. Dual single- scission event analysis of constitutive transferrin receptor (TfR) endocytosis and ligand-triggered β2-adrenergic receptor (β2AR) or Mu-opioid receptor (MOR) endocytosis. Mol. Biol. Cell 25, 3070–3080 (2014). y 31. Garay, C. et al. Epidermal growth factor–stimulated Akt phosphorylation requires clathrin or ErbB2 but not receptor endocytosis. Mol. Biol. Cell 26, 3504–3519 (2015). ARTICLE Cells were imaged at 5 s intervals for the indicated time using a spinning disk microscope (Andor) based on a CSU-W1 Yokogawa head mounted on the lateral port of an inverted IX-83 Olympus microscope equipped with a ×60 1.35NA UPLSAPO objective lens and a laser combiner system, which included 491 and 561 nm 100 mW DPSS lasers (Andor). Images were acquired with a Zyla sCMOS camera (Andor). The system was steered by IQ3 software (Andor). For CCS dynamics quantiﬁcation, we measured the lifetime of CCSs using the TrackMate plugin of ImageJ47. Tracks below 5 s of duration (detected on only 1 frame) were discarded. Measured individual lifetimes were pooled into two groups: the “dynamic” group corresponding to structures with a lifetime below the duration of the movie (5 min) and the “static” group with a lifetime of 5 min. Of note, the relative percentage of dynamic versus static structures depends on the duration of the movie because static structures are only counted once while To analyze focal adhesions (FAs), FAK-positive structures were segmented in ImageJ excluding objects with radius smaller than 75 nm. At least 160 FAs were counted for each condition from nine cells per experiments from three independent experiments. Data are expressed as mean ± SD. Fluorescence enrichment index of αvβ5 or of P-Tyr at CCPs or plaques was measured by manually delimitating α-adaptin-positive structures in order to measure the 11 NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications References 1. Pelham, R. J. Jr. & Wang, Y. Cell locomotion and focal adhesions are regulated by substrate ﬂexibility. Proc. Natl Acad. Sci. USA 94, 13661–13665 (1997). 32. Moreno-Layseca, P. & Streuli, C. H. Signalling pathways linking integrins with cell cycle progression. Matrix Biol. 34, 144–153 (2014). 33. Nardone, G. et al. YAP regulates cell mechanics by controlling focal adhesion assembly. Nat. Commun. 8, 15321 (2017). 2. Georges, P. C. & Janmey, P. A. Cell type-speciﬁc response to growth on soft materials. J. Appl. Physiol. 98, 1547–1553 (2005). 3. Engler, A. J., Sen, S., Sweeney, H. L. & Discher, D. E. Matrix elasticity directs stem cell lineage speciﬁcation. Cell 126, 677–689 (2006). 34. Rappoport, J. Z. & Simon, S. M. Endocytic trafﬁcking of activated EGFR is AP-2 dependent and occurs through preformed clathrin spots. J. Cell. Sci. 122, 1301–1305 (2009). 4. Chen, C. S. Mechanotransduction–a ﬁeld pulling together? J. Cell. Sci. 121, 3285–3292 (2008). 35. Joffre, C. et al. A direct role for Met endocytosis in tumorigenesis. Nat. Cell Biol. 13, 827–837 (2011). 5. Geiger, B., Spatz, J. P. & Bershadsky, A. D. Environmental sensing through focal adhesions. Nat. Rev. Mol. Cell Biol. 10, 21–33 (2009). 36. Paszek, M. J. et al. Tensional homeostasis and the malignant phenotype. Cancer Cell 8, 241–254 (2005). 6. Moore, S. W., Roca-Cusachs, P. & Sheetz, M. P. Stretchy proteins on stretchy substrates: the important elements of integrin-mediated rigidity sensing. Dev. Cell. 19, 194–206 (2010). 37. Driscoll, T. P. et al. Cytoskeletal to nuclear strain transfer regulates YAP signaling in mesenchymal stem cells. Biophys. J. 108, 2783–2793 (2015). 12 NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunicatio ARTICLE NATURE COMMUNICATIONS \| DOI: 10.1038/s41467-018-06367-y Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ 45. Cozzolino, A. M. et al. Modulating the substrate stiffness to manipulate differentiation of resident liver stem cells and to improve the differentiation state of hepatocytes. Stem Cells Int. 2016, 5481493 (2016). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 46. Cougoule, C. et al. Three-dimensional migration of macrophages requires Hck for podosome organization and extracellular matrix proteolysis. Blood 115, 1444–1452 (2010). 47. Tinevez, J.-Y. et al. TrackMate: an open and extensible platform for single- particle tracking. Methods 115, 80–90 (2017). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. Author contributions 38. Jaalouk, D. E. & Lammerding, J. Mechanotransduction gone awry. Nat. Rev. Mol. Cell Biol. 10, 63 (2009). F.B. designed and performed experiments, analyzed results, and wrote the manuscript. S.D., N.E., and N.L. performed experiments. N.L. generated CRISPR knock-in cell lines. A.C. and K.S. designed experiments related to micropatterning technology. T.B. designed experiments and provided assistance in generating polyacrylamide gels. D.M.V. provided reagents and designed experiments. R.P. designed and performed electron microscopy analysis. G.M. supervised the study, designed experiments, and wrote the manuscript. F.B. designed and performed experiments, analyzed results, and wrote the manuscript. S.D., N.E., and N.L. performed experiments. N.L. generated CRISPR knock-in cell lines. A.C. and K.S. designed experiments related to micropatterning technology. T.B. designed experiments and provided assistance in generating polyacrylamide gels. D.M.V. 39. Levental, K. R. et al. Matrix crosslinking forces tumor progression by enhancing integrin signaling. Cell 139, 891–906 (2009). 40. Schittenhelm, J. et al. Comparing the expression of integrins αvβ3, αvβ5, αvβ6, αvβ8, ﬁbronectin and ﬁbrinogen in human brain metastases and their corresponding primary tumors. Int. J. Clin. Exp. Pathol. 6, 2719–2732 (2013). p p g g p y y g provided reagents and designed experiments. R.P. designed and performed electron microscopy analysis. G.M. supervised the study, designed experiments, and wrote the manuscript. 41. Bai, S. Y. et al. Integrin αvβ5 as a biomarker for the assessment of non-small cell lung cancer metastasis and overall survival. Clin. Respir. J. 9, 457–467 (2015). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 018-06367-y. 43. Chozinski, T. J. et al. Expansion microscopy with conventional antibodies and ﬂuorescent proteins. Nat. Methods 13, 485–488 (2016). Competing interests: The authors declare no competing interests. Additional information 42. Hao, W. et al. Vitronectin: a promising breast cancer serum biomarker for early diagnosis of breast cancer in patients. Tumor Biol. 37, 8909–8916 (2016). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- 018-06367-y. Competing interests: The authors declare no competing interests. 44. Betz, T., Koch, D., Lu, Y. B., Franze, K. & Kas, J. A. Growth cones as soft and weak force generators. Proc. Natl Acad. Sci. USA 108, 13420–13425 (2011). Acknowledgements We wish to thank Drs J. Ivaska, P. Chavrier, A. Benmerah, and C. Albiges-Rizo for critical comments on the manuscript. We thank the imaging facilities of Gustave Roussy, Institut Curie, and Institut Imagine for help with image acquisition. We thank Isabelle Fourqueaux (TRI imaging facility, CMEAB, Toulouse) for her help with sample pre- paration for scanning electron microscopy. Core funding for this work was provided by the Gustave Roussy Institute and the Inserm and additional support was provided by grants from ATIP/Avenir Program, la Fondation ARC pour la Recherche sur le cancer, Le Groupement des Entreprises Françaises dans la LUtte contre le Cancer (GEFLUC), and from the Agence Nationale de la Recherche (ANR-15-CE15-0005-03) to GM. F.B. was supported by a fellowship from La Ligue Nationale contre le Cancer. © The Author(s) 2018 NATURE COMMUNICATIONS \| (2018) 9:3825 \| DOI: 10.1038/s41467-018-06367-y \| www.nature.com/naturecommunications 13
https://openalex.org/W3023892943	https://www.repo.uni-hannover.de/bitstream/123456789/9945/1/Kemarrec%20et%20al%202020%2c%20On%20Estimating%20the%20Hurst%20Parameter%20from%20Least-Squares%20Residuals.pdf	English	null	On Estimating the Hurst Parameter from Least-Squares Residuals. Case Study: Correlated Terrestrial Laser Scanner Range Noise	Mathematics	2,020	cc-by	14,840	Received: 27 February 2020; Accepted: 26 April 2020; Published: 29 April 2020 Received: 27 February 2020; Accepted: 26 April 2020; Published: 29 April 2020 Abstract: Many signals appear fractal and have self-similarity over a large range of their power spectral densities. They can be described by so-called Hermite processes, among which the ﬁrst order one is called fractional Brownian motion (fBm), and has a wide range of applications. The fractional Gaussian noise (fGn) series is the successive diﬀerences between elements of a fBm series; they are stationary and completely characterized by two parameters: the variance, and the Hurst coeﬃcient (H). From physical considerations, the fGn could be used to model the noise of observations coming from sensors working with, e.g., phase diﬀerences: due to the high recording rate, temporal correlations are expected to have long range dependency (LRD), decaying hyperbolically rather than exponentially. For the rigorous testing of deformations detected with terrestrial laser scanners (TLS), the correct determination of the correlation structure of the observations is mandatory. In this study, we show that the residuals from surface approximations with regression B-splines from simulated TLS data allow the estimation of the Hurst parameter of a known correlated input noise. We derive a simple procedure to ﬁlter the residuals in the presence of additional white noise or low frequencies. Our methodology can be applied to any kind of residuals, where the presence of additional noise and/or biases due to short samples or inaccurate functional modeling make the estimation of the Hurst coeﬃcient with usual methods, such as maximum likelihood estimators, imprecise. We demonstrate the feasibility of our proposal with real observations from a white plate scanned by a TLS. Keywords: terrestrial laser scanner; stochastic model; B-spline approximation; Hurst exponent; fractional Gaussian noise; generalized Hurst estimator Article Gaël Kermarrec Geodetic Institute, Leibniz Universität Hannover, Nienburger Str. 1, 30167 Hannover, Germany; kermarrec@gih.uni-hannover.de; Tel.: +49-511-7621-4736 Received: 27 February 2020; Accepted: 26 April 2020; Published: 29 April 2020 Mathematics 2020, 8, 674; doi:10.3390/math8050674 www.mdpi.com/journal/mathematics 1. Introduction Physically, the range or distance measurements are phase diﬀerences, so that a power law spectral density of the correlated range noise is hardly plausible ([7], [8]). This correlation structure was empirically proven in a few recent real case analyses see e g [9] or [10] are expected to be temporally correlated. Physically, the range or distance measurements are phase diﬀerences, so that a power law spectral density of the correlated range noise is hardly plausible ([7], [8]). This correlation structure was empirically proven in a few recent real case analyses, see, e.g., [9] or [10]. The authors approximated simple scanned objects with a Gauss–Helmert model [11], assuming pre-deﬁned geometric primitives such as circle, ellipsoid, and plane. The correlation parameters were estimated by ﬁtting the residuals of the approximation with an autoregressive function of the ﬁrst order (AR(1)). This methodology has, however, drawbacks: (i) the exponential covariance function restricts the description of the correlation structure to short range dependency and may not be a physically adequate modelization, (i) the exponential covariance function restricts the description of the correlation structure to short range dependency and may not be a physically adequate modelization, (ii) empirically the covariance can be problematic. Two autocovariance functions sharing a common principal irregular term won’t yield asymptotically the same best linear predictor (see [12], chapter 3) for some examples. Using the popular Gaussian function can lead to overoptimistic predictions. Additionally, antipersistent data are diﬃcult to distinguish from uncorrelated data and the correlations can be mistaken for noise ﬂuctuations around zero, (iii) the methodology could be made more general: it is based on a calibrated object scanned in a controlled environment. (iii) the methodology could be made more general: it is based on a calibrated object scanned in a controlled environment. In this contribution, we propose to address these drawbacks and to derive a general methodology to assess the correlation structure of the TLS range measurements. We will base ourselves on the physical expectation that the TLS range noise should have a long-range dependency (LRD) and heavy tailed distribution. Our proposal to extract the correlation structure is applicable to every kind of object, without being restricted to predeﬁned objects or calibration scenarios. It is extendable to other kinds of observations, such as residuals from a geodetic coordinate time series [13]. 1. Introduction Terrestrial laser scanners (TLS) capture a large amount of 3D points rapidly, with high precision and spatial resolution [1]. These scanners are used for applications as diverse as modeling architectural and engineering structures, and high-resolution mapping of terrain, vegetation, and other landscape features. The recorded point clouds can be processed and analyzed with dedicated software. In engineering geodesy, this processing allows for the computation of deformation magnitudes. Unfortunately, these latter are negatively aﬀected when noisy and scattered point clouds (PC) are used. Additionally, no rigorous statistical test for deformation can be performed with the raw PC [2]. These drawbacks can be circumvented by approximating the PC with mathematical surfaces [3]. Besides norms such as L1 or L∞[4], a widely used criterion is the sum of squares of the orthogonal distances from the data points to the parametric surface. Exemplarily, regression B-spline enjoys special attention to approximate point clouds from TLS: B-splines basis functions have a closed form expression, are polynomial, and, thus, particularly easy to compute (see [5] for one of the ﬁrst articles related to that topic in geodesy). The setup of speciﬁc statistical tests with conﬁdence intervals is based Mathematics 2020, 8, 674; doi:10.3390/math8050674 www.mdpi.com/journal/mathematics 2 of 23 Mathematics 2020, 8, 674 on the estimated parameters, or on the approximated surface points. Exemplarily, the congruence test can be used to test for deformation ([2]) and is known to be the most powerful test in Gauss–Markov models with normally distributed random deviations and a correctly speciﬁed stochastic model. The setting of a realistic variance covariance matrix (VCM) of the raw observations of the TLS is done prior to this test [6]. As for every sensor recording millions of points in a few minutes, the measurements of TLS are expected to be temporally correlated. Physically, the range or distance measurements are phase diﬀerences, so that a power law spectral density of the correlated range noise is hardly plausible ([7], [8]). This correlation structure was empirically proven in a few recent real case analyses, see, e.g., [9] or [10]. The authors approximated simple scanned objects with a Gauss–Helmert model [11], assuming pre-deﬁned geometric primitives such as circle, ellipsoid, and plane. The correlation parameters were estimated by ﬁtting the residuals of the approximation with an autoregressive function of the ﬁrst As for every sensor recording millions of points in a few minutes, the measurements of TLS are expected to be temporally correlated. 1. Introduction We choose to model the noise of TLS range with a stationary LRD noise: the fractional Gaussian noise (fGn), which is entirely deﬁned by the Hurst parameter (abbreviated by H) and the variance. It has the main advantage that the autocorrelation function can be easily estimated without computation burden [14]. Fractal time series or signals such as the fGn have been found in many domains, including biology [15], medicine (EEG [16,17]), ﬁnance (stock market analysis [18]), geology, and traﬃc analysis [19]. Various statistical techniques have been proposed to estimate H and each has shortcomings and advantages: they may perform better in the presence of noise, for short samples, or for H close to a given value, may have slow convergence, etc. (see, e.g., [20–24]). There exist three families of estimation methods: the time domain (e.g., Rescaled Range R/S estimator [25], the detrended ﬂuctuation analysis method [26]), the frequency domain (periodogram [27], the Whittle estimator [28]), and the wavelet space [29], which was shown to provide an unbiased, eﬃcient, and robust estimator. We will use the residuals of the B-splines approximation of the TLS point clouds to assess the correlation structure of TLS range. We conjecture that although (i) additional white noise and (ii) possible model misspeciﬁcation could introduce additional frequencies in the residuals, these latter still contain enough information to estimate H, provided that an adequate ﬁltering is performed. Besides the simulated observations from a TLS, we will evaluate our methodology and compare the performance of three diﬀerent estimators for H using real observations from a white plane. We ﬁrstly disregard the correlations of the polar angles; a similar methodology as presented in this present study could be used to that aim. The remainder of this paper is structured as follows: the ﬁrst section provides a brief summary of the mathematical concepts of least-squares and stochastic modeling. The second section introduces the 3 of 23 Mathematics 2020, 8, 674 concepts of fGn, Hurst exponent, and ﬁltering. The third section describes the results of simulations for two speciﬁc cases: a plane and a Gaussian curve. We conclude with a real case study and some recommendations. 2.1. Functional Model Free-form curves and surface ﬁttings are ﬂexible tools to approximate PC without being restricted by the use of geometric primitives, such as circle, planes, or cylinders. Possible applications of surface approximation include the testing of deformation [30] or the reduction of a huge amount of points to a simpler form. In this contribution, we make use of B-spline surfaces. Their properties and advantages over other functions, such as control and ﬂexibility, are exemplarily described in [31]. Readers interested in more details on how spline ﬁtting works should refer, e.g., to [32,33], and more speciﬁcally for geodetic applications to [5] or [31]. Such surfaces satisfy the strong convex hull property and have a ﬁne and local shape control so that they were shown to be adequate for approximating noisy and scattered PC (see, e.g., [34]). For the sake of shortness, we shortly introduce the main concept, focusing on least-squares (LS) approach to determine the model parameters, called control points (CP). We start with nobs polar observations from a TLS expressed in vector form lPOLAR of size (3nobs). The observations are made of two angles, HA and VA, and one range ρ to which a VCM Σll_POLAR is associated. This matrix describes the variance and possible correlation between the observations [11] and is focus of our contribution. 2.1.1. First Step: From Polar to Cartesian The ﬁrst step of the approximation is the transformation of the PC coordinates vector from polar lPOLAR into Cartesian lCART. The VCM has to be transformed by the error propagation law. The VCM Σll_CART reads: Σll_CART = FΣll_POLARFT (1) (1) The matrix F contains the derivatives of the point coordinates with respect to the range and angles and is given for one point i by: The matrix F contains the derivatives of the point coordinates with respect to the range and angles and is given for one point i by: 2.1.2. Second Step: The Approximation 2.1.3. Third Step: LS Solution Approximating a PC with a B-spline surface is ﬁnding the coordinates of the CP so that the distance of the data points to the approximated surface is minimized. This step can be performed by solving the LS problem, for which the minimum in the LS sense of the zero-mean error term v is searched: min p∈R3 v = Ap −lCART 2 Σll_CART. p is the matrix of CP to be estimated and is of size (3(n + 1)(m + 1)), A −(3nobs, 3(n + 1)(m + 1))— is called the design or mass matrix. It contains the evaluation of the B-spline functions at the parameters. Interested readers should refer to [35] for the description of the design matrix. The estimated coordinates of the control points are expressed by the unbiased generalized LS estimator (GLSE [11]): 1 ^pGLSE = ATΣ−1 ll_CARTA −1ATΣ−1 ll_CARTlCART (3) (3) If the VCM Σll_CART is the identity matrix (equal variance for all coordinates), the ordinary LS estimator (OLSE) is obtained: ^pOLSE = ATA −1ATlCART. We further note that the LS estimator is unbiased E ^pGLSE = E ^pOLSE = p so that ^p can be computed either with the OLSE and GLSE solution; thanks to the unbiasedness of the LS estimator, the expectation E of the estimated coordinates of the CP are not aﬀected by the choice of Σll_CART. However, the OLSE is not the most eﬃcient within the class of linear unbiased estimators anymore when Σll_CART deviates from the true (and unknown) VCM. Consecutively, hypothesis tests such as the global test, outlier tests, or congruence tests become invalid [36]. It is one of the main reasons why assessing the correlation structure of the raw measurements is an actual research topic. The number of control points has an impact on the LS solution on the ﬁtted surface. It can be either ﬁxed a priori or iteratively adjusted in the context of model selection [37]. As the impact of model misspeciﬁcations is interesting for our purpose, we will make use of the ﬁrst strategy. 2.2. The Residuals of the LS Surface Approximation 2.2. The Residuals of the LS Surface Approximation 2.2.1. The Cartesian Residuals 2.2.1. The Cartesian Residuals We call ^vCART = A ^p −lCART the residuals of the LS adjustment. ^ lCART = H ^p are the adjusted observations with H being the Hat matrix, H = A ATΣ−1 ll_CARTA −1ATΣ−1 ll_CART. The VCM of the adjusted residuals reads: We call ^vCART = A ^p −lCART the residuals of the LS adjustment. ^ lCART = H ^p are the adjusted observations with H being the Hat matrix, H = A ATΣ−1 ll_CARTA −1ATΣ−1 ll_CART. The VCM of the adjusted residuals reads: Σ^vCART ^vCART=(I −H)TΣ−1 ll_CART(I −H) (4) Σ^vCART ^vCART=(I −H)TΣ−1 ll_CART(I −H) (4) 2.1.2. Second Step: The Approximation The Cartesian PC can be approximated mathematically by means of a linear combination of basis functions, such as B-splines. In its parametric formulation, the B-spline surface s(t, f) is a tensor product surface and can be expressed as s(t, f) = n X i=0 m X j=0 Ni,pb(t)Nj,qb( f)pi,j, (2) (2) where (t, f) ∈[0, 1] × [0, 1] are the parameters in the two directions so that a B-spline surface maps the unit square to a rectangular surface patch. The basis function Ni,p and Nj,q are composite curves of degree p and q polynomials, respectively, with joining points at knots in the interval h ui, ui+p+1 and h vj, vj+q+1 . They can be evaluated by means of a recurrence relationship [32]. To summarize, the surface is deﬁned by: • a set of n + 1 CP in the direction t and m + 1 CP in the direction f, • a knot vector of h + 1 knots in the t-direction, U = [u0, . . . , uh], • a knot vector of k + 1 knots in the f-direction V = [v0, . . . , vk], • a knot vector of k + 1 knots in the f-direction V = [v0, . . . , vk], 4 of 23 Mathematics 2020, 8, 674 • the degree pb of the basis functions in the t-direction, and the degree qb in the f-direction. • the degree pb of the basis functions in the t-direction, and the degree qb in the f-direction. In this contribution, we will take a degree of pb = qb = 3 for the B-splines functions (cubic B-splines). We solve the determination of an optimal knot vector using the knot placement technique as described in [33]. The Cartesian point cloud is parametrized in advance with the equidistant parametrization, justiﬁed by the simple structure of the objects under consideration in this contribution. 3.1. Variance Raw observations from TLS are the three polar coordinates of the recorded points. They are made of a range and two angles in the vertical and horizontal direction. These observations are known to have diﬀerent noise properties ([38–40]): the noise of angles is widely assumed to be Gaussian with a variance taken from manufacturer datasheets. The noise of the range measurements has a slightly diﬀerent structure. Its variance can be considered as a constant; the manufacturer datasheets provide diﬀerent values depending on, e.g., the approximated distance to the scanned object and/or to the properties of the surface (roughness, color). Alternatively, the variance can be modeled as following a point-wise power law intensity model [41,42]. In this contribution, we simulate diﬀerent point clouds with a noise variance close to what is expected in a real case experiment: ( σHA = σVA = 0.0001 σρ = 0.005m , where σHA = σVA are the standard In this contribution, we simulate diﬀerent point clouds with a noise variance close to what is expected in a real case experiment: ( σHA = σVA = 0.0001 σρ = 0.005m , where σHA = σVA are the standard deviations for the HA and VA and σρ for the range, respectively. We intentionally chose a case where range and angle have diﬀerent variances to simulate a more general scenario. deviations for the HA and VA and σρ for the range, respectively. We intentionally chose a case where range and angle have diﬀerent variances to simulate a more general scenario. 3.2. Correlation Structure for Range Measurements with I being the identity matrix. g y We further deﬁned the a posteriori variance factor as g y We further deﬁned the a posteriori variance factor as We further deﬁned the a posteriori variance factor as ˆσ2 0 = ^vCARTTΣ−1 ll_CART ^vCART nobs −3(n + 1)(m + 1), (5) (5) This factor can be used to judge the goodness of ﬁt of the LS adjustment by means of a global test [36]. The a priori VCM of the estimates is given by Σ^p ^p= ATΣ−1 ll_CARTA −1. 5 of 23 Mathematics 2020, 8, 674 2.2.2. The Polar Residuals In this contribution, we propose to extract the correlation structure of the TLS range from the LS residuals of the B-spline approximation. We answer the drawback raised in the introduction by being independent of calibrated objects, i.e., our methodology should be applicable in every environment. As mentioned in Section 2.2.1, the LS adjustment gives access to the Cartesian residuals. To assess the noise of the raw TLS observations (range), we transform the Cartesian residuals into polar ^vPOLAR = ^vHA, ^vVA, ^vr . These latter have a VCM Σ^vPOLAR ^vPOLAR obtained similarly to Equation (1), p g O We will further focus on ^vr and conjecture that this ﬁnal vector still contains enough information to give us access to the approximate correlation structure of Σll_r, deﬁned as the VCM of the range observations. 3.2.1. What is a fGn? To deﬁne a fGn, the understanding of the LRD concept is mandatory. This property of a process is linked with the slow decay of the autocorrelation ρ to zero so it is a non-summable function, i.e., if the average value of its partial sums does not converge, see [43]. More precisely, ρ(τ) ≈cτ−δ, (6) (6) ρ(τ) ≈cτ−δ, with τ the time lag, c a positive constant, and 0 < δ < 1. As τ increases, the dependence between the observations stays strong, which implies a fat tailed autocorrelation function. Exemplarily, for a stationary process δ = 0.3, the autocorrelation for lag 100 will stay at 0.15, whereas for a Markovian process, the autocorrelation would be practically zero for lags 10 times less. This important property is the reason why such processes are said to have a “long-term memory” and it is one of the major reasons why we wish to model the correlation structure of TLS range observations with such a process. Intuitively, the high rate of measurements induces a long dependency between the observations: the autocorrelation may decay quickly at the origin—e.g., between the ﬁrst and the second observation—but stays for a long dependency much higher than 0. The autocorrelation will be similar between the ﬁrst observation and the 100th or the 200th. For a stationary process, the LRD can be related to a parameter called the Hurst exponent H, deﬁned as a measure self-similarity. A stochastic process XH(t) is self-similar if XH(t) has the same distribution as λ−HXH(λt), where λ is a scale parameter. Concretely, the process will appear statistically identical under rescaling of the time axis by a given factor and XH(t)∞λH; it lacks any characteristic time scale. This characteristic allows interpretation of H as a measure of the strength of dependence between the time points, or more loosely, how much space the signal “ﬁlls in”. p y p g A self-similar process with stationary increment XH(t + 1) −XH(t) has an autocorrelation CH(τ) i b A self-similar process with stationary increment XH(t + 1) −XH(t) has an autocorrelation CH(τ) given by 1 CH(τ) = 1 2 \|τ + 1\|2H −2\|τ\|2H + \|τ −1\|2H , (7) (7) so that for τ →∞,C(τ) →H(2H −1)τ2H−2 , meaning that the process has a long-range dependency, see Equation (6) and [44]. 3.2. Correlation Structure for Range Measurements In a ﬁrst approximation, the range measurements are considered to be uncorrelated, i.e., one observation recorded at time t is not dependent on the observation recorded at t + τ, τ being the interval between two measurements. τ is also called time lag; it is related to the scanning rate of the observations and depends on the setting. Exemplarily, the resolution for a TLS Z+F 5016 can be varied from preview to ultrahigh up to extremely high, and low to high quality: these choices impact the scanning rate, and thus the scanning time of an object. The assumption of uncorrelatedness is overoptimistic: range measurements are based on phase diﬀerences, which are inherently inﬂuenced by, e.g., the propagation of the signal through a random media, but also by the point spacing on the surface. In this contribution, the correlation structure of the range will be modeled as a fractional Gaussian noise (fGn). This assumption is justiﬁed by the physically based expectation of the author that the range noise is stationary and that its power spectrum will follow a power law [7]. The validation of this model with real data is shortly shown in Section 4.5. More extensive works using TLS observations will be performed in a next step based on the proposed methodology. fGn has the beneﬁcial properties that it is characterized by its variance and a single parameter called the Hurst exponent H. We will here shortly introduce the concept of fGn; interested readers can refer to [14] for more information. Mathematics 2020, 8, 674 6 of 23 3.2.1. What is a fGn? From these deﬁnitions, one can deﬁne the Hermite process of ﬁrst order called the fractional Brownian Motion (fBm, [14]) as a generalization of a Brownian motion for which H = 1 2. It is a non-stationary process with stationary increments and possesses the long-term memory, also called persistency or positive correlations when H > 1 2. When H < 1 2, the process has short term memory, or similarly anti-persistency or negative correlations; the autocorrelation decays fast enough so that their sums converge to a ﬁnite value. Both processes are described by a fractal dimension D, which is related to the Hurst exponent by D = 2 −H for a fBm [44]. Successive increments ςH of a fBm are called fGn: Successive increments ςH of a fBm are called fGn: (8) ςH(t) = XH(t + 1) −XH(t), (8) A fGn is, thus, a zero mean stationary process, deﬁned as the stationary increment of fBm. 2 A fGn is, thus, a zero mean stationary process, deﬁned as the stationary increment of fBm. 2 The fGn is fully characterized by the Hurst exponent and the variance σ2 ςH. The corresponding distribution is completely speciﬁed by its autocovariance function given by Equation (7). H can be related to the power-law spectrum P( f)∞1 f β , with f, β being the frequency and the power law of the process, respectively. Exemplarily β = 0 corresponds to a white noise, β = 1 is a pink noise and β = 2 is the Brownian noise. For a fBm, H is related to β by H = β−1 2 , with 1 < β < 3 and for a fGn, H = β+1 2 with −1 < β < 1 [45]. Using real observations, it is important to check if the noise is fGn or fBm: using the Matlab built-in function to estimate the Hurst exponent can lead to a misinterpretation of the results when not accounted for. 7 of 23 Mathematics 2020, 8, 674 The diﬀerence between a fBm and a fGn can be visualized in Figure 1, where fBm (Figure 1, right top) and fGn (Figure 1, left top) versus time with diﬀerent H are simulated. They are given with their corresponding power spectral densities (PSD), which decay linearly in a logplot (Figure 1, right bottom). 3.2.2. Generation of fGn In this study, we are focusing on fractal stationary noise, i.e., fGn. From the previous section, and per the deﬁnition of the fGn, it can be generated by diﬀerentiating a fBm. Matlab (2018) provides a function called wfbm, which returns a fBm for a given Hurst parameter. This function uses the wavelet method from [46], which may bias the estimation of the Hurst parameter towards the wavelet method (described in the following section). Alternatively, we propose to use the function called ﬀGn, a freely available function in the Matlab ﬁle exchange section. The ﬀGn function has the main advantage of being based on the circulant embedding method for persistent noise, resulting in a reproduction of its exact autocovariance [47]. To test the function, we assessed the standard deviation with which the Hurst parameter can be reproduced. We generated 10,000 realizations of short time series of lengths (1) 400 and (2) 1000. Focusing in this contribution on persistent fGn, the Hurst parameter was varied in the range of [1/2-1] by steps of 0.05. H was estimated using the three methods presented in the following sections. The standard deviation was found for all three methods to be between 0.01 for case (1) and 0.005 for case (2), highlighting the good performance and stability of the chosen function for noise generation. 3.2.1. What is a fGn? We visualize the aforementioned “ﬁlls in” property of the process, i.e., small H“ﬁlls in” signiﬁcantly more space than H = 0.9, which is related with a higher fractal dimension D. Figure 1. (right,top): Two realizations of a fGn with H = 0.6 (top) and H = 0.2 (bottom). (left,top): Two realizations of a fBm with H = 0.6 (top) and H = 0.9 (bottom). (right,bottom): The four corresponding PSD are plotted. Time series of 1000 observations were generated. Figure 1. (right,top): Two realizations of a fGn with H = 0.6 (top) and H = 0.2 (bottom). (left,top): Two realizations of a fBm with H = 0.6 (top) and H = 0.9 (bottom). (right,bottom): The four corresponding PSD are plotted. Time series of 1000 observations were generated. Generalized Hurst Exponent (GHE) The generalized Hurst exponent was introduced in [50] and used for ﬁnance market analysis in [51] and [52]. It is a generalization of the approach proposed by [25]. The generalized Hurst exponent was introduced in [50] and used for ﬁnance market analysis in [51] and [52]. It is a generalization of the approach proposed by [25]. The generalized Hurst exponent measures the LRD in the time domain. It is evaluated by using the qth-order moments of the distribution of increments: Kq(t) = D \|XH(t+τ)−XH(t)\| qE D \|XH(t)\| qE . D E τ is varied between 1 and τmax (usually taken to 20). Acting on τmax allows accounting for the speciﬁcity of the observations—for example, to force the estimator to focus on high frequencies—as described in [24]. ⟨·⟩stays for the average operator. Hq is related to Kq(t): Kq(τ)∞τqHq, which allows the computation of Hq ∼ log Kq(τ) q log(τ) , (9) (9) as an average over a set of values corresponding to diﬀerent τ. If Hq is not constant by varying q, the process is called multifractal, whereas Hq = H characterizes an monofractal process [53]. In this contribution, and because we are not interested in the behavior of ﬁnancial time series to predict the evolution of speciﬁc markets, we only estimate H1, which describes the scaling behavior of the absolute values of the increments. H1 reaches the value 1 2 for a Gaussian noise. H2 would correspond to an estimation in the frequency domain. 3.2.3. How to Estimate the Hurst Parameter Many methods have been proposed to estimate the Hurst exponent. They can be classiﬁed in three families: estimation in the time domain, frequency domain and wavelet domain. Intuitively, 8 of 23 Mathematics 2020, 8, 674 whereas the ﬁrst family investigates the power-law relationship between a speciﬁc statistical property of the time-series and a time aggregation of it, the two latter examine if the spectrum or energy of the time-series follows power-law behavior. Inside these three families, diﬀerent estimators have been proposed and constantly improved (see [48] and the references inside). We do not aim to review them all, which is exemplarily done in [49]. In this contribution, we focus on three estimators, which belong to each class: the generalized Hurst estimator, the Whittle estimator, and the wavelet estimator. The Whittle estimator was chosen for its capability to perform well when the number of observations is reduced; this case can occur when the TLS scanned lines are short, due to, e.g., the measurement conﬁguration and the scanned object. The selection of the generalized Hurst estimator is justiﬁed by the fact that the noise of the angle or functional misspeciﬁcations may aﬀect the power spectral density of the B-spline residuals at low frequency; this estimator focuses on the middle-high frequency part of the PSD and can be adapted with additional parameters. The wavelet estimator is said to be the less biased Hurst exponent estimator provided that a huge amount of observations is available (asymptotic behavior). These challenges of estimating the correlation structure accurately are similar to the ones of the geodetic coordinate time series analysis (see [13] for further references on that topic). The chosen estimators have to account for these speciﬁcities. In the following, we will shortly review the three methods under consideration. A good understanding of their properties is mandatory to derive a meaningful methodology to extract an unknown H from B-spline residuals. Whittle Likelihood Estimator (WhiE) Method The Maximum Likelihood Estimator (MLE) is not a graphical method but is a purely numerical one. Thus, more than just the asymptotic self-similarity is assumed [53]; the MLE requires at least an assumption about the form of the LRD (such as a noise coming from fBm or Autoregressive integrated moving average ARIMA). If this assumption holds, it is often considered to be the best obtainable estimator; the estimates are asymptotically unbiased, and the estimator is asymptotically eﬃcient and fast to compute. Unfortunately, MLE performs poorly if the assumption is incorrect or for short samples [54]. Exact maximum likelihood inference can be performed for Gaussian data ([55]) by evaluation the log-likelihood (10) l(H) = −log(\|CH\|) −XT HC−1 H XH, l(H) = −log(\|CH\|) −XT HC−1 H XH, (10) 9 of 23 Mathematics 2020, 8, 674 where XH denotes the column vector of length n of observations and CH is a fully populated VCM, which components are given using Equation (7). \|CH\| is here the determinant of the matrix. By maximizing the likelihood function, one obtains an optimal choice for H: ˆH = argmax(l(H)), with 0 < H < 1. To approximate Equation (10), matrix inversions are necessary. They can be avoided using the Whittle estimator [28], which aims to provide faster estimation with only a slight inaccuracy. In that case, the Whittle likelihood in its discretized form is given by lW(H) = − X ω∈Ω  log ef(ω, H) + I(ω) ef(ω, H)  , (11) (11) with Ωthe set of discrete Fourier frequencies, ef(ω, H) the continuous-time process spectral density and I(ω) the periodogram I(ω)∞ NP j=1 XH,je−ijω2. The same notation as in [54] was adopted. j Whittle estimator assumes a priori that the power spectrum of the underlying process of the dataset is known. Moreover, to be applicable to fGn, the mean of the time series has to be subtracted beforehand [56]. As aforementioned, the Whittle estimator should only be used if a time-series has already been shown by other methods to be consistent with a speciﬁc process, e.g., a fGn. Thus, it is not an adequate method to detect LRD. Wavelet Estimator (WE) Since H describes the level of statistical self-similarity of a time series or spatial process, the exponent can be found by averaging squared values of the wavelet coeﬃcients Ej = 1 nj nj X k=1 dX(j, k)2, (12) (12) where dX(j, k) are the detailing coeﬃcients deﬁned as dX(j, k) = ∞ R −∞ ψj,k(t)XH(t)dt, with ψj,k = J where dX(j, k) are the detailing coeﬃcients deﬁned as dX(j, k) = ∞ R −∞ ψj,k(t)XH(t)dt, with ψj,k = 2−j 2 ψ 2−jt −k , ψ the mother wavelet. XH(t) = P k JP j=1 dX(j, k)ψj,k(t) + approx, with J, the numbe 2−j 2 ψ 2−jt −k , ψ the mother wavelet. XH(t) = P k JP j=1 dX(j, k)ψj,k(t) + approx, with J, the number of j decomposition level and approx the approximating component—not of interest for our purpose. Ej at scale j can be shown to obey the scaling law: j decomposition level and approx the approximating component—not of interest for our purpose. Ej at scale j can be shown to obey the scaling law: Ej ∼2αj. (13) (13) Ej ∼2αj. The Hurst exponent is obtained by ﬁtting a line to the linear part of log2 Ej versus j in order to obtain the slope α. Diﬀerently to the power spectrum method, Ej contains here the information about the power carried at each time scale j. It was found to be robust even if the LRD is not equivocal [57] but performs poorly for short sample. Similarly to the power law β, α is linked to H diﬀerently for a fBm and fGn, with H = α−1 2 , H = α+1 2 , respectively. Wavelet based estimator are implemented in Matlab under wfbmesti. The values are based on the estimation of the Hurst exponent for a fBm and have, thus, to be applied to the cumulative sum of a fGn. 4. Simulations and Results In this section, we will combine all the mathematical developments presented in the previous sections: surface ﬁtting and Hurst exponent estimation. We recall that our aim is to estimate the Hurst exponent of the range measurements from the residuals of a B-spline approximation. In order to work in a controlled framework, we use in a ﬁrst step simulated TLS observations. A short real data analysis highlights the potential of the proposed methodology, which will be pursued in further dedicated contributions. Additional Remarks Periodicity and noise in the time series biased strongly the identiﬁcation of LRD; the estimators are misleading and can detect LRD erroneously, or on the contrary ﬁnd a Gaussian noise with H = 0.5 [20]. Frequency or wavelet-based estimators depend strongly on short-memory and necessitates strategies to alleviate these eﬀects. The estimators have to be enabled to focus on the long-range correlation in case of additional Gaussian noise of unknown variance. One possible way to face this challenging 10 of 23 Mathematics 2020, 8, 674 situation will be proposed in this contribution; the ﬁltering of the noise with a low pass Butterworth ﬁlter. Detailed simulations in Section 4 will explain the reasons of this choice. 3.3. Butterworth Filter Butterworth ﬁlters can be designed as bandpass, lowpass, or high pass ﬁlters. They are called maximally ﬂat ﬁlters as for a given order they have the sharpest roll-oﬀpossible without inducing peaking in the Bode plot. The Bode plot is a log–log graph where the gain in decibels is plotted against the logarithm of the angular frequency. An example is shown exemplarily in Figure 2 for diﬀerent order of the Butterworth ﬁlter. We note that the Butterworth ﬁlter changes from pass band to stop-band by achieving pass band ﬂatness. This is done at the expense of wide transition bands. This property, sometimes considered as the main disadvantage of Butterworth ﬁlter, turned out to be the main reason for using such a ﬁlter for our application. A great ﬂexibility is given in locating the cutoﬀfrequency, i.e., the values of the elements of the Butterworth ﬁlter are more practical and less critical than many other ﬁlter types. Interested readers should refer to [58] or [59]. Figure 2. Bode plot for a lowpass Butterworth ﬁlter with a cutoﬀfrequency of 300 Hz (0.6π rad/sample for data sampled at 1000 Hz). Diﬀerent orders were simulated. (bottom): the phase response; (top): the magnitude. Figure 2. Bode plot for a lowpass Butterworth ﬁlter with a cutoﬀfrequency of 300 Hz (0.6π rad/sample for data sampled at 1000 Hz). Diﬀerent orders were simulated. (bottom): the phase response; (top): the magnitude. 4.1. Simulation of TLS Observations The ﬁrst step towards analyzing the correlation structure of the range residuals as described in Section 2 is to simulate TLS observations. In this contribution, we choose two diﬀerent surfaces with increasing complexity: a plane and a Gaussian surface. The plane has the equation z = −3x + 15y + 7. The distance between the origin of the coordinates and the centre of the plane is 7 m. The coeﬃcients of the plane were chosen without any search for optimal scanning condition in order to test our methodology in complex cases. The representation of the plane is shown in Figure 3 (left bottom). The TLS is placed at the origin of the axes, see Figure 3 (right). 11 of 23 Mathematics 2020, 8, 674 Figure 3. (left,top): Simulated plane. (left,bottom): Simulated Gaussian surface. (right): Origin of the laser scanner in Cartesian and polar coordinates. We call θ = VA, ϕ = HA. Figure 3. (left,top): Simulated plane. (left,bottom): Simulated Gaussian surface. (right): Origin of the laser scanner in Cartesian and polar coordinates. We call θ = VA, ϕ = HA. The Gaussian surface has the equation z = 1 2π52 e−1 2 ( x2 52 + y 52 ) and is h i For each surface, the PC were generated by varying x ∈ h −1 1 i , y ∈ h −1 1 i . Two samplings were chosen: case (i) 400 observations and case (ii) 1000 observations per scanning line, resulting in PC of size 400400 and 10001000, respectively. These cases are chosen to study the impact of the density of the PC on the estimation of the Hurst parameter. 4.2. Noise Simulation The simulated Cartesian coordinates were backwards transformed into polar coordinates [VA, HA, r] and noise component wise: • to the vertical and horizontal angles is added a Gaussian noise with a standard deviation of 0.0001◦generated with the Matlab function randn, • to the range r is added a fGn noise with a standard deviation of 0.005 m. We generated noise vectors with three diﬀerent Hurst exponents: 0.6 (nearly Gaussian), 0.7, and 0.9 (strong LRD). 4.3. Estimation of the Hurst Exponent from the Residuals In the following, we will compare the estimates of H with the three previously described estimators. An application to a real case scenario is presented in Section 4.5, as well as a generalization of the results in Section 4.6. We start with the approximation of the plane (Section 4.1, Figure 3, left bottom). We approximate the PC with a cubic B-spline surface and ﬁx the numbers of CP to estimate to 4, which is justiﬁed by the simple geometry of the simulated object [60]. Intentionally, we are not searching from an optimal functional model, which could be based on an iterative method using information criteria [37]. We take a reference value of Hre f = 0.6 for the simulated noise. This reference value is close to H = 0.5 (white noise) and is challenging to estimate accurately. Interestingly, this scenario is diﬃcult to ﬁt with B-splines due to the unfavorable orientation of the plane in space; it leads to a so-called strong “border eﬀect” in the B-spline surface approximation, which is not solved entirely by a higher knot multiplicity. This eﬀect can be visually seen in the plot of the residuals by a strong increase of the variance at the beginning of each line (Figure 5 left). The correlation structure of the residuals is not dependent on the stationarity (or not) of the residual’s variance. Consequently, we allowed ourselves to disregard the corresponding inaccurately approximated ﬁrst epochs of each lines; exactly the same results as those presented in the next section were found. We interpret this eﬀect as being due to the self-similarity property of the noise and, thus, we did not intend to suppress it. Figure 5. (left): Residuals of the plane adjustment versus time (top: case 1, the whole residuals, middle: case 2, the ﬁrst 10,000 values, bottom: case 3 the ﬁrst 1000 values corresponding to one scanning line). The x-axis corresponds to the time, exemplarily in (s), whereas the y-axis is the residuals, exemplarily in (m). (right): The corresponding PSD as a log–log plot. F is given in Hz and the PSD in dB/Hz. No additional angle noise. Figure 5. (left): Residuals of the plane adjustment versus time (top: case 1, the whole residuals, middle: case 2, the ﬁrst 10,000 values, bottom: case 3 the ﬁrst 1000 values corresponding to one scanning line). Line Wise Noise We did not generate one noise vector for the whole observations. Instead, we added to each scanning line an independent noise vector; see Figure 4 for an illustration of the chosen strategy. We generated as many noise vectors as scanning lines, which size depend on the chosen sampling (case (i) or (ii)). Figure 4. Explanation of the concept of line wise noise. One noise vector is added to each line for a constant x independently. In this example, from ti = 1 to ti = 10 is added one noise vector. A new one is added starting from ti = 11. The same procedure is repeated for as many lines as the point clouds contain. Figure 4. Explanation of the concept of line wise noise. One noise vector is added to each line for a constant x independently. In this example, from ti = 1 to ti = 10 is added one noise vector. A new one is added starting from ti = 11. The same procedure is repeated for as many lines as the point clouds contain. 12 of 23 Mathematics 2020, 8, 674 Thus, the noise is not added as a whole to the observations. We justify this line wise strategy by the fact that TLS observations are recorded in such a way that the elapsed time between the last observation of one line and the ﬁrst of the following line is much longer than the time between two observations inside one line. Using this modelization, we are able to place ourselves in the context of a more general and potentially time varying correlation structure during the scanning process, answering the challenge (iii) mentioned in the introduction. This eﬀect can be caused, e.g., by changing object properties or atmospheric conditions. 4.3. Estimation of the Hurst Exponent from the Residuals The x-axis corresponds to the time, exemplarily in (s), whereas the y-axis is the residuals, exemplarily in (m). (right): The corresponding PSD as a log–log plot. F is given in Hz and the PSD in dB/Hz. No additional angle noise. 4.3.1. Impact of Model Misspeciﬁcation: No Noise Angle In order to highlight the impact of both the model misspeciﬁcation and the angle noise on the estimation, we ﬁrstly noised the simulated range only. On the contrary to real data analysis, the simulation framework allows for this ﬂexibility. The obtained range residuals of the adjustment are shown in Figure 5. They are plotted 13 of 23 Mathematics 2020, 8, 674 • case 1: as a whole in Figure 5 (left top), i.e., each 1000 observations correspond to one line; • case 2: the ﬁrst 10 lines (Figure 5 left middle) and; • case 3: only the ﬁrst line in Figure 5 (left bottom). • case 1: as a whole in Figure 5 (left top), i.e., each 1000 observations correspond to one line; • case 2: the ﬁrst 10 lines (Figure 5 left middle) and; • case 3: only the ﬁrst line in Figure 5 (left bottom). • case 3: only the ﬁrst line in Figure 5 (left bottom). Although the whole residuals may visually appear as white noise with a slight variance increase in the between t = 4e5 and t = 7e5 (unit of time) due to the scanning conﬁguration (Figure 3, left top), we identify repetitive pattern for each approximated scanned line (Figure 3, left middle); they are likely to inﬂuence negatively the estimators of the Hurst parameter that are acting in the frequency domain such as the WhiE. The PSD for the three cases (1–3) corresponding to Figure 5 (left) are plotted in Figure 5 (right). We note that it has strong similarities with the one of the simulated noise for the whole range residuals. The expected power law decay is kept nearly intact, which is beneﬁcial to the wavelet and WhiE estimator. Additional frequencies for −2.5 < log10( f) < 1 are visible, which we link with the aforementioned repetitive pattern due to model misspeciﬁcations. For decreasing sample size (Figure 5, right bottom), the low frequency domain of the analyzed residuals from the ﬁrst scanning line does not follow exactly the one of the original noise. It is possible to compensate for that eﬀect using the GHE by decreasing τmax; this corresponds to down-weighting the impact of the low frequencies in the estimation of H. For the sake of convenience and without lack of generality, we will carry our explanation with the residuals of the ﬁrst line. 4.3.1. Impact of Model Misspeciﬁcation: No Noise Angle We computed the Hurst exponent for case 1 and 3 with the chosen three methods. Case 2 is of minor interest and was only presented to show the pattern of the residuals. For case 1, the whole residuals are considered in the estimation of H, leading in a longer time series, whereas for case 3 we take the mean of the H estimated over smaller samples. For case 3, the standard deviation as well as the min/max values of H can be given (one Hurst exponent is estimated for each line). The results are presented in Table 1. We added the estimation of the Hurst exponent from the original generated noise for comparison purpose. Table 1. Estimation of the Hurst parameters from the residuals for case 1 and case 3, for the three estimators under consideration. We give additionally the standard deviation of the estimation, when available. Case without additional angle noise, Hre f = 0.6. Table 1. Estimation of the Hurst parameters from the residuals for case 1 and case 3, for the three estimators under consideration. We give additionally the standard deviation of the estimation, when available. Case without additional angle noise, Hre f = 0.6. GHE WhiE WE Case 1 H 0.60 (std 8×10−4) 0.54 (std 0.14) 0.61 Noise 0.60 (std 4×10 −4) 0.54 (std 0.12) 0.60 Case 3 mean(H) min/max std(H) 0.61 0.51/0.72 0.03 0.7 0.5/0.9 0.14 0.60 0.35/0.72 0.07 Noise mean(H) min/max std(H) 0.59 0.55/0.65 0.02 0.7 0.5/0.9 0.14 0.60 0.37/0.70 0.06 Table 1 shows that from the three estimators, the WhiE performs worst. This holds true particularly for case 3, for which the Hurst exponent for both the simulated noise and the residuals are overestimated; this eﬀect was expected due to the small samples under consideration (1000 observations) and is related in the literature as the main drawback of this estimator—under the assumption that the noise is fGn. For case 1 (whole residuals), the WhiE has a better performance regarding case 3 due to the frequency Mathematics 2020, 8, 674 14 of 23 averaging but remains a poorer estimate compared with the values given by the GHE and the WE; both estimators provide the true H. The GHE is less aﬀected by the sample size than the wavelet estimator, i.e., in case 3 the standard deviation of H for the WE is higher than for GHE for both the noise and the residuals. 4.3.1. Impact of Model Misspeciﬁcation: No Noise Angle From this simulation and without additional noise on the angles, the preference goes towards the GHE when the H exponent has to be evaluated for each line, i.e., for small samples. This is a nice result when temporal variations of the parameter are expected ([61]), since they will be detected with a higher trustworthiness than with other estimators. g Similar results are obtained for Hre f = 0.7 and Hre f = 0.9, and are not presented for the sake of shortness and readability of this article. 4.3.2. Impact of Model Misspeciﬁcation and Noise Angle Before Filtering After Filtering Case 1: cutoﬀlog10 fc = −0.6 Case 3: cutoﬀlog10 fc = −0.4 GHE WhiE WE GHE WhiE WE Case 1 H 0.71 (std 0.01) 0.53 (std 0.3) 0.71 0.87 (std 0.01) 0.63 (std 0.3) 0.98 Noise 0.89 (std 3×10−4) 0.7 (std 0.2) 0.90 Case 3 mean(H) min/max std(H) 0.71 0.56/0.86 0.05 0.67 0.50/0.96 0.15 0.71 0.51/0.92 0.08 0.89 0.76/0.96 0.03 3 1.52/4.58 1.51 0.96 Noise mean(H) min/max std(H) 0.86 0.75/0.95 0.03 0.64 0.51/0.75 0.20 0.83 0.74/0.98 0.08 The ﬁrst remark to draw from Table 2 is the stronger diﬃculty to estimate the Hurst parameter from the true fGn for small samples (case 3) than for longer sample (case 1). The mean values are slightly below the true one of Hre f = 0.9 for the GHE and WE estimators, with a higher standard deviation than in the previous case with Hre f = 0.6. Clearly, the WhiE performs poorly and systematically underestimates the true parameter. The second remark is the impossibility to extract the correct, or a value close to the correct Hurst exponent, independently of the case under consideration. The noise of the angles, as well as the noise induced by the ﬁtting, leads to a strong underestimation of H close to 0.7. The decrease towards H = 1 2 (a white noise) is due to the increase of white noise in the signal. As previously, the WhiE estimates poorly H (0.53 for case 1). It is shown to be thus strongly aﬀected by additional white noise on the residuals. The analysis of the PSD (Figure 7) for case 1 and 3 highlights the impact of additional white noise. From Equation (9) and Equation (13), we notice that the GHE and WE need both low and high frequencies to perform the approximation of the Hurst exponent with trustworthiness; it seems advantageous to ﬁlter the high frequency noise of the residuals. In this contribution, we propose to apply a lowpass Butterworth ﬁlter of ﬁrst order on the residuals from the cutoﬀfrequency at which the PSD kicks towards white noise. This choice is not justiﬁed by empirical ﬁndings and we propose in the following to detail the reasons why we opted for the Butterworth ﬁlter. Figure 7. The PSD of the residuals for case 1 (top) and case 3 (bottom). The Hurst exponent for the simulated noise is Href = 0.9. 4.3.2. Impact of Model Misspeciﬁcation and Noise Angle In a second step, we added a noise with a standard deviation of 1×10−4◦to the angle components. In order to be able to visually identify the diﬀerence between the slope of the PSD for the noise to the one of the residuals—aﬀected by additional white noise—we consider the case Hre f = 0.9 (see Figure 6). This is a challenging exponent to estimate, since the corresponding process is close to a ﬂicker noise. Results for other H are similar when the same methodology is applied. Figure 6. The PSD of the residuals for case 1 (top), case 2 (middle), and case 3 (bottom). The Hurst exponent for the simulated noise is Href = 0.9. A plane was approximated with 1000 observations per line. Log–log plot. Case with additional angle noise. Figure 6. The PSD of the residuals for case 1 (top), case 2 (middle), and case 3 (bottom). The Hurst exponent for the simulated noise is Href = 0.9. A plane was approximated with 1000 observations per line. Log–log plot. Case with additional angle noise. From Figure 6, the impact of the additional noise coming from the angles and propagating in the range residuals can be clearly identiﬁed in the high frequency domain, i.e., from log10( f) >−0.6 for case 1, and log10( f) >−0.4 for case 3. This corresponds to a noise at −40 dB/Hz for case 1, −45 dB/Hz for case 2, and −57 dB/Hz for case 3, approximately. The corresponding Hurst exponents were estimated and the results are presented in Table 2. As previously, we also give the results obtained for the simulated noise. 15 of 23 Mathematics 2020, 8, 674 Table 2. Estimation of the Hurst parameters from the residuals for case 1 and case 3, with the three estimators under consideration. We additionally give the standard deviation of the estimation, when available. Additional angle noise, Hre f = 0.9. 4.3.2. Impact of Model Misspeciﬁcation and Noise Angle A plane is approximated with 1000 observations per line. Case with additional angle noise. The red curve corresponds to the PSD of the simulated noise, the blue curve to the PSD of the residuals and the green to the ﬁltered residuals. Figure 7. The PSD of the residuals for case 1 (top) and case 3 (bottom). The Hurst exponent for the simulated noise is Href = 0.9. A plane is approximated with 1000 observations per line. Case with additional angle noise. The red curve corresponds to the PSD of the simulated noise, the blue curve to the PSD of the residuals and the green to the ﬁltered residuals. Mathematics 2020, 8, 674 16 of 23 16 of 23 Why a Butterworth Filter A sharp low pass ﬁltering would lead to an abrupt decrease of the PSD from the cutoﬀfrequency of the ﬁlter; this eﬀect is here unwanted as the estimation of H necessitates the whole range of frequencies, which would not be given any more. We prefer, thus, the “smooth and gentle” Butterworth ﬁlter of ﬁrst order; it allows a continuous decrease of the high frequencies from the cutoﬀfrequency. This leads to a ﬁltering shown Figure 7 (green line), where the PSD of the ﬁltered signal has the same decrease as the reference noise from log10( f) = −2 for case 1 and log10( f) = 0 for case 3. The ﬁltering leads to an estimate of the Hurst exponent of 0.87 (std 0.01) and 0.89 (std 0.03) with the GHE for case 1 and 3, respectively (see Table 2). τmax was ﬁxed to 20, as proposed in the literature [51]. Increasing τmax leads to a slight decrease of H of 0.2 for τmax = 40 with an increase of the std to 0.03, whereas τmax =5 is linked with an increase of H of 0.3 by a decrease of the std to 0.008. Thus, a balance has to be found to ﬁx τmax optimally. A deep analysis of the PSD, i.e., the amount of power at low frequencies is necessary; whereas τmax can be used to ﬁlter unwanted low frequencies due to model misspeciﬁcation, the cutoﬀfrequency acts on the high frequency domain of the PSD. With the chosen cutoﬀfrequency, the WE overestimates H. Using a cutoﬀfrequency of log10( fc) =−0.35, instead of log10( fc) =0.4, for case 3 yields H = 0.88. Similar results are obtained for case 1 with the WE by increasing log10( fc) to −0.55. Unfortunately, the GHE decreases to 0.83 in both cases. However, considering that (1) no prior knowledge of the Hurst exponent was available, (2) additional white noise aﬀects the residuals, and (3) model misspeciﬁcation are present, this remains a good approximation of the true H of 0.9. Indeed, this value of the Hurst parameter is known to be challenging to estimate since it is close to the limit between fGn and fBm. Why First Order? The answer is strongly related to the previous one: as shown in Figure 2 from the Bode plot, the ﬂatness of the ﬁlter is of main importance to ensure smooth transition in the PSD of the ﬁltered signal. 4.3.3. Sensitivity Analysis: Impact of the CutoﬀFrequency In this section, we propose to analyze the sensitivity of the estimated H exponent regarding the chosen cutoﬀfrequency. Figure 8 summarizes the results for case 1 (top) and 3 (bottom) by varying log10( f) from to −0.7 to −0.25. Figure 8. Sensitivity analysis of the estimated H from the residuals of the B-spline surface ﬁtting by varying the cutoﬀfrequency. A plane is estimated with 1000 observations per line and the simulated noise is fGn with Href =0.9. Case 1 corresponds to the whole residuals, case 3 to the ﬁrst line. Case with additional angle noise. The red curve corresponds to the GHE, the blue curve to the WE. Figure 8. Sensitivity analysis of the estimated H from the residuals of the B-spline surface ﬁtting by varying the cutoﬀfrequency. A plane is estimated with 1000 observations per line and the simulated noise is fGn with Href =0.9. Case 1 corresponds to the whole residuals, case 3 to the ﬁrst line. Case with additional angle noise. The red curve corresponds to the GHE, the blue curve to the WE. Figure 8. Sensitivity analysis of the estimated H from the residuals of the B-spline surface ﬁtting by varying the cutoﬀfrequency. A plane is estimated with 1000 observations per line and the simulated noise is fGn with Href =0.9. Case 1 corresponds to the whole residuals, case 3 to the ﬁrst line. Case with additional angle noise. The red curve corresponds to the GHE, the blue curve to the WE. With great evidence, the GHE is much less sensitive to the cutoﬀfrequency than the WE. A linear dependency can be found, with a variation of H from 0.88 to 0.78 for the chosen range of cutoﬀ frequencies and from 0.87 to 0.80 for case 1 and 3, respectively. H, estimated with WE, has a much 17 of 23 Mathematics 2020, 8, 674 higher range of values—from 1 to 0.68 and to 1.15 to 0.8 for case 1 and 3, respectively. From these results, and considering that we placed ourselves intentionally in a challenging estimating scenario with a strong Hurst exponent, we recommend using the GHE instead of the WE when the residuals are ﬁltered and small samples are considered. 4.3.4. Small Samples In this section, we place ourselves in case (i) as described in Section 4.1. and generate smaller samples of 400 observations per line. We chose three values for H: 0.6, 0.7, and 0.9, and apply our methodology to ﬁlter the residuals from additional white noise and/or results from model misspeciﬁcations. We identify the cutoﬀfrequency fc visually by plotting (1) the PSD of the whole residuals for case 1 and (2) the PSD of 10 randomly chosen lines for case 3 and averaging the identiﬁed cutoﬀfrequencies. The Hurst exponent is estimated with the GHE; due to their asymptotic properties, the WE and the WhiE are known to perform poorly for small samples [62]. For Hre f = 0.6, the PSD is nearly similar to a white noise (see Figure 1). This leads to a stronger diﬃculty to identify the PSD kick. Nevertheless, we were able to identify with a high conﬁdence the correct cutoﬀfrequency and a value of H =0.60 could be estimated for case 1 for log10( fc) =−0.5. We link this result with the use of a Butterworth ﬁlter of the ﬁrst order and the low sensitivity of the GHE to a misspeciﬁcation of the cutoﬀfrequency. The same cutoﬀwas used for the two other simulated Hre f . This result strongly conﬁrms the feasibility of extracting the Hurst exponent from residuals of regression B-splines, in the presence of both functional misspeciﬁcations and additional unknown noise. The cutoﬀfrequency depends on the noise angle variance, as illustrated in Table 3. Increasing the σHA = σVA to 0.001◦instead of 0.0001◦yields a diﬀerent cutoﬀfrequency. We intentionally do not present this result in order not to overload the readers with simulation results that lead to similar conclusions. Table 3. Estimation of the Hurst parameters from the residuals for case 1 and case 3, with the GHE and with additional angle noise for two standard deviations (1×10−4◦and 1×10−3◦). Hre f = 0.6, 0.7, and 0.9. The cutoﬀfrequencies ( fc) are visually determined. Table 3. Estimation of the Hurst parameters from the residuals for case 1 and case 3, with the GHE and with additional angle noise for two standard deviations (1×10−4◦and 1×10−3◦). Hre f = 0.6, 0.7, and 0.9. The cutoﬀfrequencies ( fc) are visually determined. Table 3. 4.3.4. Small Samples Estimation of the Hurst parameters from the residuals for case 1 and case 3, with the GHE and with additional angle noise for two standard deviations (1×10−4◦and 1×10−3◦). Hre f = 0.6, 0.7, and 0.9. The cutoﬀfrequencies ( fc) are visually determined. Std Noise Angle 1 × 10−4◦ Std Noise Angle 1 × 10−3◦ Case 1 Cutoﬀ log10(f) = −0.5 Case 3 Cutoﬀ log10(f) = −0.25 Case 1 Case 3 Hre f = 0.6 0.60 (std 0.05) 0.63 (std 0.03) Hre f = 0.7 0.68 (std 0.02) 0.70 (std 0.03) log10( fc) = −0.5 0.66 (std 0.08) log10( fc) = −0.4 0.71 (std 0.07) log10( fc) = −0.25 0.66 (std 0.03) log10( fc) = −0.55 0.72 (std 0.07) Hre f = 0.9 0.88 (std 0.02) 0.87 (std 0.04) 4.4. Result for Gaussian Surface 4.4. Result for Gaussian Surface 4.4. Result for Gaussian Surface The second example corresponds to a simulated Gaussian surface (case (ii), Section 4.1.). Ten CP in the two directions were estimated with B-splines of order three. The stationary reference noise was simulated with Hre f = 0.7 and 1000 observations per line. Similarly to the previous simulations, we do not aim to optimally ﬁt the surfaces so that the impact of potential misspeciﬁcation can be considered. In Figure 9 (left), the PSD of the residuals together with the PSD of the simulated noise are shown; Figure 9 (right) represents the residuals for case 1 and 3 respectively, following the previous section. This latter ﬁgure highlights the lack of repetitive patterns in the residuals plotted per line (case 3). Only a steady increase of the variance towards the middle of the surface can be seen, which is coherent with the Gaussian form of the surface (Figure 2 left bottom). This behavior does not aﬀect the estimation of the Hurst parameter, which was 0.72 (std 1×10−3) for case 1 and 0.71 (std 0.01) for 18 of 23 Mathematics 2020, 8, 674 case 3 with or without ﬁltering. From the PSD, a low additional white noise from log10( fc)= 0 could be identiﬁed, which did not aﬀect the determination of H. We interpret this lack of additional white noise in the residuals as coming from the goodness of the surface approximation, i.e., the B-splines themselves are acting as a low pass ﬁlter so that no additional noise coming from the angles could drift into the residuals in that case. However, we were able to decrease the estimated Hurst parameter for case 1 to 0.69 (std 5×10−4) by decreasing τmax to 10, i.e., decreasing the impact of the low frequencies. In this case, the B-spline LS system ﬁlters the low frequencies domain strongly, which could be accounted for by acting on τmax. Figure 9. (left): Residuals of the B-spline approximation for a Gaussian surface. Case 1 (top); the whole residuals and case 3: the 1000 ﬁrst observations corresponding to one scanning line. (right): The corresponding PSD (red the original noise, blue the residuals). Figure 9. (left): Residuals of the B-spline approximation for a Gaussian surface. Case 1 (top); the whole residuals and case 3: the 1000 ﬁrst observations corresponding to one scanning line. (right): The corresponding PSD (red the original noise, blue the residuals). 4.4. Result for Gaussian Surface This correction highlights the potential of our methodology to identify and ﬁlter model misspeciﬁcation from the LS residuals. It is and remains based on a visual analysis of the PSD and an understanding of the residuals as prior to the estimation of the Hurst parameter. It is not recommended to use a bandpass Butterworth (or any other ﬁlter such as a notch ﬁlter) to ﬁlter speciﬁc frequencies. This was shown to strongly aﬀect the determination of the Hurst exponent by creating an artiﬁcial decrease of frequencies amplitude in the middle of the frequency range, where a regular decrease is of main importance for the determination of H. 4.5. Application to Real Data We propose to apply the proposed methodology to a real case scenario. Unfortunately, the true correlation structure is unknown; the development of a model based on a physical explanation of the TLS correlation is beyond the scope of this paper and led to further studies. A white plane of size 1 m*1 m was scanned at a distance of 10 m with a Z+F 2016F using the scanning modus “extremely high”, with which 1 Mio. per s can be recorded. The scanning conﬁguration is presented in Figure 10 (left); it is optimal and corresponds to the simulated data with no tilt and the TLS pointing in the direction of the z-axis. The obtained point cloud was pre-processed to avoid edge eﬀects and outliers, and cut using a free software. We ﬁnally approximated the data with a cubic B-spline, following the methodology presented in Section 2. The residuals for one scanning line are plotted in Figure 10 (right, top), together with the corresponding PSD (Figure 10, right, bottom, blue line). We visually identiﬁed a cutoﬀfrequency of log10( fc) =0, which we used to ﬁlter the residuals with a Butterworth ﬁlter (Figure 10, bottom, yellow line). As for the simulations, the results obtained with the three Hurst estimators proposed in this contribution diﬀer. Without ﬁltering, we found values of 0.85 for the GHE with τmax =20 (which was chosen due to the lack of additional low frequencies from inaccurate functional model), 1.01 (i.e., ﬂicker noise) for the WhiE, and 0.61 for the WE. This last result highlights that the WE is aﬀected by white noise—the value found was close to 0.5—and by the small number of observations used (900 per line). The tendency for the WhiE to overestimate the Hurst parameter with respect to the GHE (Table 1) is additionally shown. Using the visually identiﬁed 19 of 23 Mathematics 2020, 8, 674 cutoﬀfrequency, the Hurst estimator was increased to 0.88 for the GHE, but stayed constant for the WE; this estimator is deﬁnitively not a relevant choice for the case study under consideration. A high value of 1.61 was found for the WhiE, which seems not usable with the ﬁltered residuals, i.e., the WhiE being a spectral estimator is aﬀected by the strong decrease of the PSD at high frequencies. Figure 10. (left): The schematic scanning conﬁguration: no tilt and a distance of 10 m to the center of the coordinate system. 4.5. Application to Real Data Right, top: residuals of the B-spline approximation for the plane under consideration. (right,bottom): The corresponding PSD. The blue line is a reference fGn with Hre f = 0.9, the red line corresponds to the residuals and the yellow one to the ﬁltered residuals with log10( fc) = 0. Figure 10. (left): The schematic scanning conﬁguration: no tilt and a distance of 10 m to the center of the coordinate system. Right, top: residuals of the B-spline approximation for the plane under consideration. (right,bottom): The corresponding PSD. The blue line is a reference fGn with Hre f = 0.9, the red line corresponds to the residuals and the yellow one to the ﬁltered residuals with log10( fc) = 0. From Figure 10 (right bottom) the plausibility of considering the noise of the TLS range as being fGn is conﬁrmed; the blue line corresponds to a reference fGn of 0.9 and is nearly parallel to the yellow one from the ﬁltered residuals. From Figure 10 (right bottom) the plausibility of considering the noise of the TLS range as being fGn is conﬁrmed; the blue line corresponds to a reference fGn of 0.9 and is nearly parallel to the yellow one from the ﬁltered residuals. This short case study validates the proposed methodology for a real case scenario: it is feasible to estimate the Hurst parameter from the range residuals of a plane scanned with a TLS. Further studies will be carried out in a next future to investigate more deeply the correlation structure. This latter is expected to depend on, e.g., the scanning rate, distance from the plane to the TLS, or atmospheric conditions. igu e B-spline app 5. Conclusions 5. Conclusions In this contribution, we have developed and validated an innovative yet simple strategy to extract the correlation structure of the underlying observation noise from the residuals of a B-spline In this contribution, we have developed and validated an innovative yet simple strategy to extract the correlation structure of the underlying observation noise from the residuals of a B-spline surface approximation. This determination is neither based on least-squares estimation or collocation, nor parametric, and has the main advantages of being easy to use and computationally eﬃcient. 5. Conclusions In this contribution, we have developed and validated an innovative yet simple strategy to extract the correlation structure of the underlying observation noise from the residuals of a B-spline In this contribution, we have developed and validated an innovative yet simple strategy to extract the correlation structure of the underlying observation noise from the residuals of a B-spline surface approximation. This determination is neither based on least-squares estimation or collocation, nor parametric, and has the main advantages of being easy to use and computationally eﬃcient. y g p surface approximation. This determination is neither based on least-squares estimation or collocation, nor parametric, and has the main advantages of being easy to use and computationally efficient. Our case study dealt with TLS raw observations, having in mind to analyze the correlations of the range observations to perform more rigorous and trustworthy statistical tests for deformation of scanned objects. This is a highly relevant application for avoiding and/or quantifying the potential Our case study dealt with TLS raw observations, having in mind to analyze the correlations of the range observations to perform more rigorous and trustworthy statistical tests for deformation of scanned objects. This is a highly relevant application for avoiding and/or quantifying the potential risk related to the deformation of structures such as dams or bridges. Moreover, knowledge of the correlation structure could serve to predict future deformations. risk related to the deformation of structures such as dams or bridges. Moreover, knowledge of the correlation structure could serve to predict future deformations. The range measurements of TLS observations are known to be temporally correlated. We guess from physical consideration that the power spectral density of the noise could be represented by a power law. The framework of LRD allows description of such kinds of noise accurately. igu e B-spline app 5. Conclusions In this study, we chose to model the correlation structure of the TLS range measurements by a stationary persistent fGn. The fGn is widely used to describe all kinds of noise in various domains and can be fully described by means of its Hurst exponent (related to the fractal dimension) and its variance. There exist various estimators for the Hurst exponent. In this contribution, we compared the performance of one of each family: the generalized Hurst estimator, the Whittle likelihood estimator and the wavelet estimator. We simulated small and longer samples, as well as observations noise with different Hurst exponents. Our goal was to determine as accurately as possible, from the B-splines range residuals, the reference parameter. Regression B-spline surface fitting can be applied to nearly every noisy and scattered point cloud, without limitation to specific surfaces such as circle or plane. The range measurements of TLS observations are known to be temporally correlated. We guess from physical consideration that the power spectral density of the noise could be represented by a power law. The framework of LRD allows description of such kinds of noise accurately. In this study, we chose to model the correlation structure of the TLS range measurements by a stationary persistent fGn. The fGn is widely used to describe all kinds of noise in various domains and can be fully described by means of its Hurst exponent (related to the fractal dimension) and its variance. There exist various estimators for the Hurst exponent. In this contribution, we compared the performance of one of each family: the generalized Hurst estimator, the Whittle likelihood estimator and the wavelet estimator. We simulated small and longer samples, as well as observations noise with diﬀerent Hurst exponents. Our goal was to determine as accurately as possible, from the B-splines range residuals, the reference parameter. Regression B-spline surface ﬁtting can be applied to nearly every noisy and scattered point cloud, without limitation to speciﬁc surfaces such as circle or plane. Even if they are structurally correlated, the residuals of the approximation still contain information about the correlation and noise structure of the raw observations. Even if they are structurally correlated, the residuals of the approximation still contain information about the correlation and noise structure of the raw observations. Unfortunately, as in every approximation model, misspecifications are likely to arise. 4.6. Summary: A Methodology to Extract the Hurst Parameter from the B-Spline Residuals In this section, we summarize our methodology to extract the Hurst parameter of the underlying noise from TLS range measurements from LS residuals (Figure 11). We recall that the noise is simulated line wise as a fGn with a given Hurst parameter varying from 0.5–1 (persistent correlations). Working with real observations this assumption has to be tested by analyzing the stationarity of the time series as well as the power law of its PSD. We start with the raw polar observations, which are to be transformed into Cartesian coordinates. After having parametrized the point cloud, a B-spline surface approximation is performed. The choice of the order of the B-splines is left to the user (e.g., cubic B-splines), as well as the method to ﬁx the knot vector optimally or/and the number of CP to estimate. The residuals of the approximation are transformed backwards into polar coordinates; only the range residuals are further analyzed. They are plotted as a whole and line wise to visually identify the potential impact of model misspeciﬁcation (low frequencies, repetitive pattern). These patterns could act—as a snow ball eﬀect—on the determination of the Hurst parameter. As an important tool to understand the structure of the residuals, the PSD is plotted against the frequencies (a log–log plot should be used for a better visualization). Additional white noise or model misspeciﬁcation can slide into the frequency domain; they are identiﬁed and ﬁltered with a low pass Butterworth ﬁlter of ﬁrst order. We recommend the use of the generalized Hurst estimator. This latter was shown to be robust to slight uncertainties in the determination of the cutoﬀfrequency, as well as less sensitive to small samples eﬀect, compared with the wavelet estimator. Thus, temporal variations of the Hurst exponent can be analyzed by making a line wise analysis of the Hurst parameters. 20 of 23 21 of 25 Mathematics 2020, 8, 674 Mathematics 2020, 8, x FO Figure 11 Summary of the methodology to extract the Hurst parameter from the range residuals of a Figure 11. Summary of the methodology to extract the Hurst parameter from the range residuals of a B-spline approximation from a TLS point cloud. Fi 11 S f th th d l t t t th H t t f th id l f Figure 11. 4.6. Summary: A Methodology to Extract the Hurst Parameter from the B-Spline Residuals Summary of the methodology to extract the Hurst parameter from the range residuals of a B-spline approximation from a TLS point cloud. igu e B-spline app 5. Conclusions They introduce additional frequencies in the residuals, which affect the determination of the Hurst parameter. Simulating a plane, we identified unwanted white noise as strongly affecting the estimation. A low pass Butterworth filter of the first order applied to the residuals was able to correct the bias induced by an unwanted additional white noise. The generalized Hurst estimator was shown to be robust against slight over or underestimation of the cutoff frequency of the filter. The Whittle likelihood performs badly in estimating H , which was linked to the potential non-stationarity of the residuals, i.e., the assumption that the residuals should be fGn is mandatory for this estimator. Unfortunately, as in every approximation model, misspeciﬁcations are likely to arise. They introduce additional frequencies in the residuals, which aﬀect the determination of the Hurst parameter. Simulating a plane, we identiﬁed unwanted white noise as strongly aﬀecting the estimation. A low pass Butterworth ﬁlter of the ﬁrst order applied to the residuals was able to correct the bias induced by an unwanted additional white noise. The generalized Hurst estimator was shown to be robust against slight over or underestimation of the cutoﬀfrequency of the ﬁlter. The Whittle likelihood performs badly in estimating H, which was linked to the potential non-stationarity of the residuals, i.e., the assumption that the residuals should be fGn is mandatory for this estimator. The wavelet estimator performs ideally in absence of white noise and could be shown to be sensitive to the choice of the cutoﬀ frequency. We interpreted this behavior as being linked with non-averaging, compared with the GHE. 21 of 23 Mathematics 2020, 8, 674 Simulating a Gaussian surface, the impact of model misspeciﬁcation in the low frequency domain was highlighted and ﬁltered adequately with a high pass Butterworth ﬁlter to improve the determination of the Hurst exponent. For both simulated cases, similar conclusions were drawn; the Hurst exponent can be well determined with the GHE, provided that a preﬁltering of the residuals with the smooth Butterworth ﬁlter of ﬁrst order is performed. The cutoﬀfrequency could be visually identiﬁed from the PSD of the residuals (line wise or as a whole). The feasibility of the proposed methodology was conﬁrmed using real data from a plane scanned with a TLS with the “extremely high” resolution. igu e B-spline app 5. Conclusions This powerful way to identify the noise structure from the residuals paves the way for a deeper study of the correlation dependency of TLS range measurements, independent of speciﬁc calibration procedures. Due to the high accuracy and precision of the determination of the fractal parameter, potential atmospheric parameters could be deduced from the B-spline residuals, as well as sensor characteristics. This analysis will be the topic of a later study with real data. The estimation of the range variance remains to be solved. A proposal could be based on the calibration of the LS system with white noise. Funding: The publication of this article was funded by the Open Access fund of Leibniz Universität Hannover. The author gratefully acknowledge the funding by the Deutsche Forschungsgemeinschaft under the label KE 2453/2-1. Conﬂicts of Interest: The author declares no conﬂict of interest. Conﬂicts of Interest: The author declares no conﬂict of interest. References 22 of 23 Mathematics 2020, 8, 674 18. Vandewalle, N.; Ausloos, M. Coherent and random sequences in ﬁnancial ﬂuctuations. Phys. A: Stat. Mech. Appl. 1997, 246, 254–459. [CrossRef] 19. Abry, P.; Sellan, F. The wavelet-based synthesis for fractional Brownian motion proposed by F. Sellan and Y. Meyer: remarks and fast implementation. Appl. Comput. Harmon. Anal. 1996, 3, 377–383. 20. Bardet, J.-M.; Lang, G.; Oppenheim, G.; Philippe, A.; Stoev, S.; Taqqu, M.S. Semi-parametric estimation of the long-range dependence parameter: A survey. In Theory and Applications of Long-Range Dependence; Doukhan, P., Oppenheim, G., Taqqu, M., Eds.; Springer: Berlin/Heidelberg, Germany, 2003; pp. 557–577. 21. Taqqu, M.S. Fractional Brownian motion and long-range dependence. In Theory and Applications of Long-Range Dependence; Doukhan, P., Oppenheim, G., Taqqu, M., Eds.; Birkhäuser: Boston, MA, USA, 2001. 22. Stolojescu, C.; Isar, A. A comparison of some Hurst parameter estimators. In Proceedings of the 2012 13th International Conference on Optimization of Electrical and Electronic Equipment (OPTIM), Brasov, Romania, 24–26 May 2012; pp. 1152–1157. 23. Krakovská, H.; Krakovská, A. Fractal Dimension of Self-Aﬃne Signals: Four Methods of Estimation. arXiv 2016, arXiv:1611.06190. 24. Cannon, M.J.; Percival, D.B.; Caccia, D.C.; Raymond, G.M.; Bassingthwaighte, J.B. Evaluating scaled windowed variance methods for estimating the Hurst coeﬃcient of time series. Physica A 1997, 241, 606–626. [CrossRef] 25. Hurst, H.E. Long-term storage capacity of reservoirs. Trans. Am. Soc. Civ. Eng. 1951, 116, 770. 26. Peng, C.K.; Buldyrev, S.V.; Havlin, S.; Simons, M.; Stanley, H.E.; Goldberger, A.L. Mosaic organization of DNA nucleotides. Phys. Rev. E 1994, 49, 1685–1689. [CrossRef] 27. Geweke, J.; Porter-Hudak, S. The estimation and application of long memory time series models. J. Time Ser. Anal. 1983, 4, 221–238. [CrossRef] 28. Whittle, P. Estimation and information in stationary time series. Ark. Mat. 1953, 2, 423–434. [CrossRef] 29. Abry, P.; Flandrin, P.; Taqqu, M.S.; Veitch, D. Self-similarity and long-range dependence through the wavelet lens. In Theory and Applications of Long-Range Dependence; Doukhan, P., Oppenheim, G., Taqqu, M., Eds.; Birkhäuser: Boston, MA, USA, 2003; pp. 527–556. 30. Kermarrec, G.; Alkhatib, H.; Bureick, J.; Kargoll, B. Impact of mathematical correlations on the statistic of the congruency test case study: B-splines surface approximation from bridge observations. In Proceedings of the 4th Joint International Symposium on Deformation Monitoring (JISDM), Athens, Greece, 15–17 May 2019. 1. Bureick, J.; Alkhatib, H.; Neumann, I. Robust spatial approximation of laser scanner points clouds by me of free-form curve approaches in deformation analysis. J. Appl. Geod. References 1. Vosselman, G.; Maas, H.G. Airborne and Terrestrial Laser Scanning; CRC Press: Boca Raton, FL, USA, 2010. 1. Vosselman, G.; Maas, H.G. Airborne and Terrestrial Laser Scanning; CRC Press: Boca Raton, FL, USA, 2010. 2. Pelzer, H. Zur Analyse Geodätischer Deformationsmessungen; Verlag der Bayer. Akad. d. Wiss: Munchen, Germany, 1971. 2. Pelzer, H. Zur Analyse Geodätischer Deformationsmessungen; Verlag der Bayer. Akad. d. Wiss: Munchen, Germany, 1971. y 3. Lee, S.Y.; Wolberg, G.; Shin, S.Y. Scattered data interpolation with multilevel B-splines. IEEE Trans. Vis. Comput. Graph. 1997, 3, 228–244. [CrossRef] 4. Aigner, M.; Jüttler, B. Distance regression by Gauss–Newton-type methods and iteratively re-w least-squares. Computing 2009, 86, 73–87. [CrossRef] 5. Koch, K.R. Fitting free-form surfaces to laserscan data by NURBS. AVN Allg. Vermess.-Nachr. 2009, 116, 134–140. 6. Kermarrec, G.; Neumann, I.; Alkhatib, H.; Schön, S. The stochastic model for Global Navigation Satellite Systems and terrestrial laser scanning observations: A proposal to account for correlations in least squares adjustment. J. Appl. Geod. 2019, 13, 93–104. [CrossRef] 7. Wheelon, A.D. Electromagnetic Scintillation: Part I Geometrical Optics; Cambridge University Press: Cambridge, UK, 2001. 8. Ishimaru, A. Wave Propagation and Scattering in Random Media; IEEE Press and Oxford University Press: New York, NY, USA, 1997. . Kauker, S.; Holst, C.; Schwieger, V.; Kuhlmann, H.; Schön, S. Spatio-temporal correlations of terrestrial l scanning. AVN Allg. Vermess.-Nachr. 2016, 6, 170–182. 0. Kauker, S.; Schwieger, V. A synthetic covariance matrix for monitoring by terrestrial laser scanning. J. A Geod. 2017, 11, 77–87. [CrossRef] 1. Koch, K.-R. Parameter Estimation and Hypothesis Testing in Linear Models; Springer: Berlin, Germany, 199 2. Stein, M.L. Interpolation of Spatial Data: Some Theory for Kriging; Springer: New York, NY, USA, 1999. 13. Montillet, J.-P.; Bos, M.S. Geodetic Time Series Analysis in Earth Sciences; Springer: Cham, Switzerland, 2020. J y p g 14. Mandelbrot, B.B.; Van Ness, J.W. Fractional Brownian motion, fractional noises and applications. SIAM Rev. 1968, 10, 422–437. [CrossRef] 15. Sims, D.W.; Southall, E.J.; Humphries, N.; Hays, G.C.; Bradshaw, C.J.A.; Pitchford, J.W.; James, A.; Ahmed, M.Z.; Brierley, A.; Hindell, M.A.; et al. Scaling laws of marine predator search behaviour. Nature 2008, 451, 1098–1102. [CrossRef] [PubMed] 16. Keshner, M.S. 1/f noise. Proc. IEEE 1982, 70, 212–218. [CrossRef] 16. Keshner, M.S. 1/f noise. Proc. IEEE 1982, 70, 17. Accardo, A.; Aﬃnito, M.; Carrozzi, M.; Bouquet, F. Use of the fractal dimension for the analysis of electroencephalographic time series. Biol. Cybern. 1997, 77, 339–350. References 2016, 10, 27–35. [CrossRef] 32. De Boor, C.A. Practical Guide to Splines; Revised ed.; Springer: New York, NY, USA, 2001. 33. Piegl, L.; Tiller, W. The NURBS Book; Springer Science & Business Media: Berlin, Germany, 1997. 34. Bracco, C.; Giannelli, C.; Sestini, A. Adaptive scattered data ﬁtting by extension of local approximations to hierarchical splines. Comput. Aided Geom Des 2017, 52–53, 90–105. [CrossRef] 35. Zhao, X.; Kermarrec, G.; Kargoll, B.; Alkhatib, H.; Neumann, I. Inﬂuence of the simpliﬁed stochastic model of TLS measurements on geometry-based deformation analysis. J. Appl. Geod. 2019, 13, 199–214. [CrossRef] 36. Teunissen, P.J.G. Testing Theory: An Introduction; VSSD Publishing: Delft, The Netherlands, 2000. 37. Alkhatib, H.; Kargoll, B.; Bureick, J.; Paﬀenholz, J.A. Statistical evaluation of the B-Splines approximation of 3D point clouds. In Proceedings of the FIG-Kongresses, Istanbul, Türkey, 6–11 May 2018. 8. Soudarissanane, S.; Lindenbergh, R.; Menenti, M.; Teunissen, P. Scanning geometry: Inﬂuencing facto the quality of terrestrial laser scanning points. ISPRS 2011, 66, 389–399. [CrossRef] 39. Boehler, W.; Marbs, A. 3D Scanning instruments. In Proceedings of the CIPA WG6 International Workshop on Scanning for Cultural Heritage Recording, Corfu, Greece, 1–2 September 2002. 40. Zhao, X.; Alkhatib, H.; Kargoll, B.; Neumann, I. Statistical evaluation of the inﬂuence of the uncertainty budget on B-spline curve approximation. J. Appl. Geod. 2017, 11, 215–230. [CrossRef] 41. Wujanz, D.; Burger, M.; Mettenleiter, M.; Neitzel, F. An intensity-based stochastic model for terrestrial laser scanners. ISPRS J. Photogramm. Remote Sens. 2017, 125, 146–155. [CrossRef] 42. Wujanz, D.; Burger, M.; Tschirschwitz, F.; Nietzschmann, T.; Neitzel, F.; Kersten, T.P. Determination of intensity-based stochastic models for terrestrial laser scanners utilising 3D-point clouds. Sensors 2018, 18, 2187. [CrossRef] [PubMed] 43. Beran, J. Statistics for Long Memory Processes; Chapman and Hall: New York, NY, USA, 1994. 44. Mandelbrot, B.B. The Fractional Geometry of Nature; W.H. Freeman: New York, NY, USA, 1983. 23 of 23 Mathematics 2020, 8, 674 45. Eke, A.; Hermán, P.; Bassingthwaighte, J.; Raymond, G.; Percival, D.B.; Cannon, M.J.; Ikrényi, C.; Balla, I. Physiological time series: distinguishing fractal noises from motions. Pﬂügers Arch—Eur. J. Physiol. 2000, 439, 403–415. [CrossRef] [PubMed] Abry, P.; Veitch, D. Wavelet analysis of long-range dependent traﬃc. IEEE Trans. Inf. Theory 1998, 44, 2–15 [CrossRef] 47. Perrin, E.; Harba, R.; Jennane, R.; Iribarren, I. Fast and exact synthesis for 1-D fractional Brownian motion and fractional Gaussian noises. IEEE Signal Process. Lett. 2002, 9, 382–384. [CrossRef] 48. References Liu, Y.; Liu, Y.; Wang, K.; Jiang, T.; Yang, L. Modiﬁed periodogram method for estimating the Hurst exponent of fractional Gaussian noise. Phys. Rev. E 2009, 80, 066207. [CrossRef] 49. Jeong, H.D.J.; Lee, J.S.R.; McNickle, D.; Pawlikowski, P. Distributed Steady-State Simulation of Telecommunication Networks with Self-Similar Teletraﬃc. Simul. Model. Pract. Theory 2005, 13, 233–256. [CrossRef] 50. Barabasi, A.L. Vicsek Multifractality of self-aﬃne fractals. Phys. Rev. A 1991, 44, 2730–2733. [CrossRef] [PubMed] 51. Sensoy, A. Generalized Hurst Exponent approach to eﬃciency in MENA markets. Phys. Rev A 2013, 392, 5019–5026. [CrossRef] 52. Taqq, M.S.; Teverovsky, V. On estimating the intensity of long-range dependence in ﬁnite and inﬁnite variance time series. In A Practical Guide to Heavy Tails; Feldman, R.E., Aﬂer, R.J., Taqqu, M.S., Eds.; Birkhäuser: Boston, MA, USA, 1998; pp. 177–217. 53. Di Matteo, T.; Aste, T.; Dacorogna, M.M. Long-term memories of developed and emerging markets: Using the scaling analysis to characterize their stage of development. J. Bank. Finance 2005, 29, 827–851. [CrossRef] 54. Sykulski, A.M.; Olhede, S.C.; Guillaumin, A.P.; Lilly, J.M.; Early, J.J. The debiased Whittle likelihood. Biometrika 2019, 106, 251–266. [CrossRef] 55. Brockwell, P.J.; Davis, R.A. Time Series: Theory and Methods; Springer: New York, NY, USA, 1991. 56. Chang, Y.C. Eﬃciently Implementing the Maximum Likelihood Estimator for Hurst Exponent. Math. Probl. Eng. 2014, 2014, 490568. [CrossRef] 57. Tarnopolski, M. On the relationship between the Hurst exponent, the ratio of the mean square successive diﬀerence to the variance, and the number of turning points. Phys. A: Stat. Mech. Appl. 2016, 461, 662–673. [CrossRef] 8. Butterworth, S. On the Theory of Filter Ampliﬁers. Wirel. Eng. 1930, 7, 536–541. 59. Zumbahlen, H. CHAPTER 8—Analog Filters. In Linear Circuit Design Handbook; Zumbahlen, H., Ed.; Newnes: New South Wales, Australia, 2008; pp. 581–679. 60. Kermarrec, G.; Alkhatib, H.; Neumann, I. On the Sensitivity of the Parameters of the Intensity-Based Stochastic Model for Terrestrial Laser Scanner. Case Study: B-Spline Approximation. Sensors 2018, 18, 2964. [CrossRef] 61. Garcin, M. Estimation of time-dependent Hurst exponents with variational smoothing and application to forecasting foreign exchange rates. Phys. A: Stat. Mech. Appl. 2017, 483, 462–479. [CrossRef] 62. Kirichenko, L.; Radivilova, T.; Deineko, Z. Comparative analysis for estimating of the Hurst exponent for stationary and nonstationary time series. Inf. Technol. Knowl. 2011, 5, 371–388. © 2020 by the author. Licensee MDPI, Basel, Switzerland. References This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4289362069	https://zenodo.org/records/1491883/files/30_IJRG18_A10_1788.pdf	English	null	CHALLENGES OF POTABLE WATER SUPPLY MANAGEMENT IN BAFUT SUB-DIVISION, NORTH WEST REGION OF CAMEROON	Zenodo (CERN European Organization for Nuclear Research)	2,018	cc-by	9,425	Keywords: Challenges; Potable Water; Quality; Quantity; Supply Management; Water Sources Keywords: Challenges; Potable Water; Quality; Quantity; Supply Management; Water Sources ds: Challenges; Potable Water; Quality; Quantity; Supply Management; Water Sour Cite This Article: Modeste Lumnwi, Wilson Yetoh Fantong and Samuel Ndonwi Ayonghe. (2018). “CHALLENGES OF POTABLE WATER SUPPLY MANAGEMENT IN BAFUT SUB- DIVISION, NORTH WEST REGION OF CAMEROON.” International Journal of Research - Granthaalayah, 6(10), 256-273. https://doi.org/10.5281/zenodo.1491883. Abstract Drinking water if not monitored could be a vehicle for diseases of public health concern in Bafut Sub-Division Cameroon. The present study sought to assess the challenges faced by community water supply schemes based on the evaluation of quality and quantity. This study was a cross sectional community survey marked by field survey involving mapping, discharge measurement, potable water samples collection (springs, streams, rain, wells and boreholes) and laboratory analysis. Twelve pipe born water supply schemes and thirty-eight water sources were appraised within two intervals (March-July 2015 and October 2017-March 2018). Chemical analysis was done using an Ion Chromatography and microbial analyses using the presumptive test method and standard plate count. Hospital registers were reviewed on prevalence of water borne diseases. The results showed that major challenges of water supply included; dilapidated pipes, spring inputs and climate variability. Hydro chemical faces were sodium bicarbonate and calcium chloride water types. A total Coliform count of 43 to >1100 per 100ml was recorded for March and July 2015 and 0-236Tcfu in March 2018. Escherichia coli and Salmonella typhi were probably responsible for high rate of water borne diseases. There is need to develop standardize integrated water management plan in this area. [Lummi et. al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 Science 1. Introduction Water is an essential commodity for the existence of life on Earth, but the expanding human population and climate change have caused its demand to exceed supply and quantitative depletion [256] Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) (Received: September 18, 2018 - Accepted: October 27, 2018) in many parts of the world (Patricia, 2014, Scott and Richard, 2015). Fresh water which is often used for drinking now occupies less than 3% of water out of 326 trillion gallons of water on Earth (Michel, 2017). As developmental projects in developing countries, provision and access to safe drinking water has increased over the last decade but approximately one billion people still lack safe drinking water throughout the world. Sustainable Development Goal (SDG) six ("Ensure availability and sustainable management of water and sanitation for all targets to achieve universal and equitable access to safe and affordable drinking water for all by 2030. In Cameroon as at 2006, 70% of the population had access to safe drinking water and a supply coverage of 88% in urban areas and 47% in rural areas (Ako et al., 2010). in many parts of the world (Patricia, 2014, Scott and Richard, 2015). Fresh water which is often used for drinking now occupies less than 3% of water out of 326 trillion gallons of water on Earth (Michel, 2017). As developmental projects in developing countries, provision and access to safe drinking water has increased over the last decade but approximately one billion people still lack safe drinking water throughout the world. Sustainable Development Goal (SDG) six ("Ensure availability and sustainable management of water and sanitation for all targets to achieve universal and equitable access to safe and affordable drinking water for all by 2030. In Cameroon as at 2006, 70% of the population had access to safe drinking water and a supply coverage of 88% in urban areas and 47% in rural areas (Ako et al., 2010). It is important to investigate the water quality so as to endorse a scheme for drinking and sustainable management (CAWST, 2009). Human needs for water vary per daily consumption of water and investigating on water quantity will improve on the usage and valorization of this natural resource. 1. Introduction al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 Despite this situation, only few recent studies have been carried out on the microbial quality of drinking-water, restricted to major cities of Douala (Ndjama et al., 2008; Ako et al., 2009), Yaounde (Kuitcha et al., 2010; Ateba et al., 2012) and Dschang (Katte et al., 2003; Temgoua, 2011). These studies demonstrated exceptionally high and unacceptable levels of bacteriological contamination of most drinking-water supplies. It is thus important to assess rural water supply schemes where demography and climate variability are real. ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) Despite this situation, only few recent studies have been carried out on the microbial quality of drinking-water, restricted to major cities of Douala (Ndjama et al., 2008; Ako et al., 2009), Yaounde (Kuitcha et al., 2010; Ateba et al., 2012) and Dschang (Katte et al., 2003; Temgoua, 2011). These studies demonstrated exceptionally high and unacceptable levels of bacteriological contamination of most drinking-water supplies. It is thus important to assess rural water supply schemes where demography and climate variability are real. The study was limited to Bafut Sub Division situated between longitude 10o 06N and latitude 6o ‘10E North West Region of Cameroon (Figure 1). The study area is characterized by a humid sub- Equatorial climate with nine months of rainfall from November to March with an annual rainfall of 2400 mm which recharges groundwater and a dry season from March to October (CDP, 2011). The population of Bafut has been on an increase from the year 1953 with a population of over 19,000, proceeded by 80,305 in 1987 to 129,363 inhabitants in 2011 following a census made from council development plan with support from the National Community Driven Development Program (PNDP) (CDP, 2011). The study area has a population growth rate of about 2.9 % per annum. This population increase triggered this work, which sought to determine the quantity and quality of water available for this growing population and to determine whether the service delivery of catchments are satisfactory. Developmental issues such as water supply and management for these growing population are handled by development associations who financed water projects through Water Maintenance Committee (WMC) (CDP, 2011). 1. Introduction Groundwater quantities are greatly influenced by wet and dry periods, with water tables rising when the climate is wetter and falling when climate is dry. One effect of rising and falling water tables is rising and falling of stream levels, springs discharges and groundwater in wells and boreholes (Gereta, 2004). The negative impacts from the variations are mostly on the vulnerable population of women and children who are forced to travel long distances to fetch for water. WHO (2014) defines an improved drinking-water source as a source that by the nature of its construction and design, adequately protects the source from outside contamination, by fecal matter in particular (WHO, 2014). Drinking water from these sources should have a high probability of being free of fecal indicator bacteria and, if possible, should meet the WHO quality guidelines for drinking-water. Conversely, unimproved drinking-water sources are vulnerable to permanent or temporary water-quality deterioration and often do not meet the WHO guideline values for safe drinking water. Insufficient access to safe water supply and adequate sanitation has been reported as the root cause of many diseases that afflict Africa and a contributing factor to high infant and maternal mortality rates (Ako et al., 2010). Thus the performance of most water resources especially at the community level are dwindling because of unsustainability, mainly due to poor financial management and lack of trust (Strasser, 2000). Therefore, rural water supply schemes should deliver the expected service to users for a reasonable period of time in terms of quality, quantity, accessibility, coverage, affordability and continuity (Harvey, 2004). Changing climatic conditions affect water quality and quantity, thus leaving the study area at risk of water insecurity (Ngwa, 1982). Water related diseases represent about two-thirds of all the diseases in Cameroon and are responsible for approximately 50 % of the cases of deaths recorded (Katte et al., 2003). The combination of low access to safe water and poor sanitation conditions is made evident by Cameroon’s recent history of cholera outbreaks and the burden of disease attributed to diarrheal illness in the country. Despite Cameroon’s greatest cholera outbreaks, majority of water quality studies conducted in Cameroon have mostly been centered in the Far North Region and not much done in the North West region particularly in Bafut sub division (Djaouda et al., 2014, Takem, 2010, Nola et al., 2013). [257] Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [Lummi et. 1. Introduction However there has been limited studies carried out on the quality and quantity of potable water consumed by the population, thus creating the necessity to have a close look on how operational water supply schemes are management and the quality of water they supply. Figure 1: Location of Bafut Sub- Division (Adapted from CAMGIS Bamenda, 2015) Figure 1: Location of Bafut Sub- Division (Adapted from CAMGIS Bamenda, 201 Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [258] [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) Study Design The present study assesses potable water resources and their suitability for drinking with the challenges faced in water supply management. This study was carried out in two intervals, January- August 2015 and October 2017-March 2018. During the first interval the study was centered in South East Bafut which is composed of the most densely populated areas with administrative units and within the second interval the study cut across the entire Sub-Division. Sample collection, preservation and transportation was done following American Public Health Association standards, in sterilized plastic bottles (1500ml and 500ml) after thorough rinsing directly from the respective water sources (taps, wells, streams, boreholes and rain) (APHA, 1998. A cloistered cold flask with ice blocks was used for sample preservation and transported to “Laboratoire d’Analyse Geochemie et l’eau” (LAGE) of the Institute of Geology and Mining Research (IRGM) Yaounde. Samples for microbial analyses were analyzed within 24 hours after transportation as used in a study carried out by Omam and Ayonghe, (2015) assessing the potability of some sachet water brands sold in Cameroon. All samples collected were labeled appropriately for physico-chemical and microbial analyses. A questionnaire design containing both structured and non-structured questions as used by Henry, (2007) in his study assessing the Challenges of Water Supply in Ghana. A stratified purposive and random sampling technique as used by Regassa (2016) was adopted as a method of administering the questionnaires. Figure 2: Showing locations of sample collection points Figure 2: Showing locations of sample collection points Figure 2: Showing locations of sample collection points Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [259] [Lummi et. al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 Four hundred Questionnaires were administered to nine communities in South East Bafut (Mbebali, Mankaha, Bujung, Niko, Njinteh, Nem, Ntabuwe, Agyati and Swie) and 40 to members of WMC to have their perception on prevailing water conditions and demand (Figure 2). Twelve pipe born water supply schemes and 38 other supplementary potable water sources were appraised for qualitative analyses amongst which six schemes where assessed for quantitative analyses from South East Bafut. ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. Discharge Analysis Discharge measurements were carried out for 30 days (March, 2015) using a 20 liters bucket and a stop watch per seconds to obtain the mean daily discharge. Population water demand was derived from questionnaires administered based on the amount of water each individual in a household uses daily as used by Henry, (2007). Supply coverage was determined from discharge and population demand. Physico-Chemical Analysis y y Physical parameters analyzed included water temperature, atmospheric temperature, electrical conductivity and pH, measured insitu using a PT154ET thermometer, Cond 330i/set and WTW 315i pH meter respectively. Thirty samples were collected in March (24) and July (6), 2015 for physico-chemical analyses using an Ion Chromatography (ICS) (Small et al., 1975) and fourty two in March 2018 for microbial assessment. The identities of the cations and anions were determined by their retention time and their concentration by a calibration curve. The device used was ICS- 900 and ICS-1100 Dionex model. The charge balance for reliability of chemical measurement calculated was within the acceptable limit of ±5% (Dominico et al., 1990). Study Design al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) Four hundred Questionnaires were administered to nine communities in South East Bafut (Mbebali, Mankaha, Bujung, Niko, Njinteh, Nem, Ntabuwe, Agyati and Swie) and 40 to members of WMC to have their perception on prevailing water conditions and demand (Figure 2). Twelve pipe born water supply schemes and 38 other supplementary potable water sources were appraised for qualitative analyses amongst which six schemes where assessed for quantitative analyses from South East Bafut. Analyses for Microbial Quality y Q y During the first interval, a presumptive test was carried out on 12 samples in March 2015 and 6 samples in July, 2015 to detect the Most Probable Number (MPN) of bacteria present in each water sample (Benson, 2002) in the Life Science Laboratory of the University of Buea. A medium was prepared using lactose broth following manufacturers guide and distributed with a pipette into 103 Durham tubes with 9 tubes representing a sample; 3 Durham tubes of Double Strength Lactose Broth (DSLB) and 6 of single strength lactose broth (SSLB) and autoclaved for 15minutes. A pipette was used to drop 10ml of water sample into each DSLB tubes and 1ml into a set of 3 Durham tubes; 0.1ml of water drop into a set of 3tubes SSLB per sample. They were incubated at 35oC for 24 hours and examined to obtain the number of tubes with 10% or more gas formation. The MPN of coliform was determined based on the American Public Health Association standards manual (APHA, 1998). Cultural analysis using standard plate count (SPC) and the light microscope was used for microbial and bacteriological analyses using MacConkey and SS Agar medium during the second interval of the study (March 2018). Fourty two potable water schemes were sampled (15 springs, 13 taps, 6 streams, 1 borehole, 5 wells and 2 rain water sources) from 31communities. Stock culture and subculture on nutrient agar were done to obtain fresh colonies for microscopic analysis. Preparation of nutrient agar was done following manufacturer’s instructions and APHA manual of 1978 and 1992. Violet red bile lactose agar and Macconkey agar was used to isolate Echerichia coli from other gram negative pathogens present in water samples that show the presence of coliform, while SS agar was used to isolate Salmonella. The mediums were prepared under a Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [260] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] , ( ) (Received: September 18, 2018 - Accepted: October 27, 2018) Bunsen and allowed to cool at 47o C and inoculated into corresponding labeled cultured plates of water samples and incubated for 24-48 hours at 35o C -37o C (Druce et al., 1957). Samples were also centrifuge and inoculated on three nutrient agar used. Analyses for Microbial Quality Colony forming unit (cfu/ml) per milliliters was calculated by multiplying the average number of colonies per plate by the dilution factor of the inoculated sample. The average number of colonies per plate was derived by dividing the total number of colonies of all plates that were inoculated by the same dilution volume and dividing the sum by the two plates used. For quality control and reliability microscopy gram control was done on all isolated slides from centrifuge samples and on isolated colonies. Microscopic analysis was done after cultural quantification of bacteria on gram stained slides. Slides were examined on a light microscope using 100x oil immersion objectives to differentiate gram negative from gram positive bacteria. The appearance of pink color signifies a positive result for Escherichia coli and blackish color positive for Salmonella (Njunda et al, 2013). Hospital registers were reviewed from 2003-2017 on the prevalence of water borne diseases and causative agents in within the months of October 2017-December 2017. Statistical Analysis of Data SPSS version 22, epi info version 9.1 and Pearson correlation coefficient were used to establish the relationship between water variables and the Pipers plot to determine water type (Piper, 1944). Climatic data was obtained from the Meteorological Station at Bafut airport to establish rainfall trends. Challenges of Water Supply g pp y The results indicated that the population uses Gravitational (98%) and electrical pumps (2%) water canalization systems. The study established two levels of factors causing water unavailability in the study area; namely internal (factors within the utility) and external (factors outside the utility). Within the utility, there was poor governance which was portrayed by, unavailability of performance reports and lack of incentives to workers (low salaries), low tariffs collection which was aggravated by illegal connections rated at 15% and faults (dilapidated pipes) which eventually led to high non-revenue water (Table 1) Table 1: Challenges of potable water supply Causes for break down in water supply Percentage Poor slope gradient 2.5 Dilapidated pipes and activities carried out at catchment area 42.5 Leakages and increasing population with partitioning 37 illegal home connection 15 The study established two levels of factors challenging potable water supply mainly internal which is within the management of the utility and external which could be natural factors. Within the supply scheme there is poor management marked by Farming and logging carried out within catchment areas, dilapidated pipes (remarkably in Niko, Njinteh, Nsem Ntabuwe communities), water partitioning, leakages from reservoirs, slope gradient, pump failure and illegal home connections. Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [261] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) This was aggravated by a lack of incentives to caretakers of the schemes sponsored by the unavailability of performance and financial reports. Reservoir chambers and tanks accounted for poor sanitation as most of the reservoirs and chambers (about 85%) were not regularly cleaned, and about 75% of the catchments were located close to toilets while some catchment areas served as grazing land for cattle. Despite these factors, the study identified external factors that impinge water availability in the community such as, population growth pegged at 2.9% annually and natural factors such as climate variability. The mean annual rainfall of Bafut observed a 20% decrease from 2825mm in 1995 to 2250mm in 2015, creating an impact on ground water resources as it could lead to a decrease in water level (Figure 2). Figure 3: Plot of total annual rainfall from 1995 to 2014 Annual rainfall Annual rainfall Figure 3: Plot of total annual rainfall from 1995 to 2014 Water Discharge g The results for the analysis of water discharge on daily bases indicated that 83.3% of the six catchments within the study area experienced an increase in water discharge particularly Mbebali, Mankaha/ Bujung and Niko/Njinteh from the period of installation till present, while 16.6% of the supply schemes experience a decrease in discharge rates mainly from Ntabuwe (Table 2). Table 2: Discharge rate of water supply schemes Community Installed capacity Calculated mean litters per day Rate of change % change Water per person in a day(liters) Total population using supply scheme Mbebali 18,144 64,800 46,656 72% 6.82 9,500 Mankaha/Bujung 16,416 36,288 19,872 54.8% 14.52 2,500 Niko/ Njinteh 19,872 43,200 23,328 54% 12.17 3,550 Nsem 16,416 34,560 18,144 52.5% 17.28 2,000 Agyati 2,592 4,320 1,728 40% 0.39 11,000 Ntabuwe 19,872 13,824 -6,048 -43.8% 9.21 1,500 Total 93,312 196,992 103,680 52.6% 6.5 1,000 Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [262] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] , (Received: September 18, 2018 - Accepted: O (Received: September 18, 2018 - Accepted: October 27, 2018) In a whole the study area observed a 52.6% (103,680 l/day) increase in discharge per day from 1981 to 2015 with about 6.5 liters of water available for an individual per day (Table 2). Comparing daily water demand (Table 3) and discharge, daily water demand (206,6282.5 L) was greater than daily catchment discharge (196,992 L). Table 3: Demand and supply measurement Community Calculated Present daily demand in litters % demand Mean daily supply Supply coverage (%) Mbebali 505,970 24.5 64,800 12.8 Mankaha/Bujung 198,375 9.6 36,288 18.2 Niko/ Njinteh 238,737.5 11.6 43,200 1.8 Nsem 160,000 7.7 34,560 21.6 Agyati 888,800 43 4,320 0.48 Ntabuwe 74,400 3.6 13,824 18.6 Total 2066282.5 100 196992 9.5 Table 3: Demand and supply measurement Agyati community had the highest demand (43%). Infact this was in line with a supply coverage of 9.5%, and greatest supply coverage covered by Nsem scheme (21.6%). Agyati community had the highest demand (43%). Infact this was in line with a supply coverage of 9.5%, and greatest supply coverage covered by Nsem scheme (21.6%). Hydrochemistry of Water Sources Physico- chemical analysis showed that water temperature varied from 21 to 23o C with a mean temperature of 22.2o C, 22.15oC for springs and 22.51oC for taps. pH values ranged between 5.75 - 7.75 with a mean of 6.78, indicating acidity (Table 4). Electrical conductivity varied between 15.7µs/cm to 171µS/cm and TDS between 0.4 to 23.6mg/l. The concentrations of ions were minimal in water samples which could be as a result of low mineralization from short duration of water rock interaction. In decreasing order, the relative concentration of cations stood at Mg2+>Ca2+>Na2+>k+>NH4+ while that of anions stood at HCO- 3>Cl- >F->SO4->NO3->PO-4. The mean concentration of magnesium stood at 0.5meq/l ranging from 0.02meq/l to 2.2meq/l, with highest concentration from Mankaha water source (2.2meq/l). Bicarbonate ranged from 0.13meq/l to 1.9meq/l having a mean concentration of 0.72meq/l. Sodium ion concentration ranged from 0.7mg/l to 1.7mg/l (Table 4) in tap water and 0.6mg/l to 1.9mg/l in spring water with a mean concentration of 1.31mg/l. Table 4: Physical and chemical parameters of water samples Samples pH EC TDS Na+ NH4+ K+ Mg2+ Ca2+ F- Cl- NO3- PO3- 4 SO24- HCO- 3 Unit μS/cm mg/L Mg/l Mg/l Mg/l Mg/l Mg/l Mg/l Mg/l Mg/l Mg/l Mg/l Mg/l AGT 6.75 15.7 0.4 0.7 0.06 1.07 0.25 0.09 0.04 0.03 0.02 0.02 0.04 8.1 AGS 6.74 17.3 0.8 0.77 0.04 1.16 0.39 0.22 0.08 0.23 0.07 ND 0.07 8.8 MKS 7.43 168 0.4 3.5 0.03 3.65 25.94 36.21 0.06 0.09 0.55 0.12 0.04 116.3 MKT 7.59 171 0.4 1.72 0.08 1.81 13.06 18.29 0.08 0.06 0.84 0.17 0.06 114.3 MMS 6.35 42.8 1.4 1.91 0 1.45 1.91 1.72 0.04 0.1 0.02 0.08 0.09 26.3 MMT 7.63 130 1.2 1.72 0.03 1.35 10.52 10.93 0.07 0.08 0.82 0.08 0.05 84.7 NSS 7.75 88 3.6 0.74 0.02 1.01 6.55 9.12 0.07 0.07 1.44 0.06 0.08 56.4 NST 6.6 78.4 2.4 0.89 0.08 0.9 5.76 8.01 0.07 0.11 1.55 0.04 0.06 49.5 NTS 6.88 24.4 4 0.68 0.03 0.61 1.38 1.1 0.03 0.06 0 0.05 0.17 14.4 Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [263] [Lummi et. Hydrochemistry of Water Sources al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 NTT 6.57 30 2.4 0.75 0.08 0.88 1.82 1.39 0.05 0.06 0.12 ND 0.22 17 NNS 6.58 29.7 23.6 1.51 0.03 1.23 1.70 1.82 0.09 0.24 0.14 0.33 0.12 16.8 NNT 6.44 28.1 1.2 0.83 0.88 1.82 1.39 0.05 0.06 0.08 0.12 0.03 0.12 16.9 ND= None detectable NTT 6.57 30 2.4 0.75 0.08 0.88 1.82 1.39 0.05 0.06 0.12 ND 0.22 17 NNS 6.58 29.7 23.6 1.51 0.03 1.23 1.70 1.82 0.09 0.24 0.14 0.33 0.12 16.8 NNT 6.44 28.1 1.2 0.83 0.88 1.82 1.39 0.05 0.06 0.08 0.12 0.03 0.12 16.9 ND= None detectable The mean concentration of sodium was more in spring water (1.51mg/l) than in the tap water (1.09mg/l) with Bujung/Mankaha springs (3.5mg/l) indicating the highest concentrations. Chloride ranged from 0.03mg/l (Agyati tap, Mankah/Bujung tap, Ntabuwe spring and Ntabuwe tap) to 0.1mg/l (Mankaha/Bujung) with a mean of 0.1 mg/l mean while Nitrate had a mean concentration of 0.47 mg/l. Correlation values indicated a positive relationship between HCO3- with Ca, Mg, Na and K, values of 0.91, 0.92, 0.72 and 0.7 respectively. The dominant water types were Ca+- Cl-, and Na+ + K+ - HCO3 (Figure 3). Figure 4: Piper’s diagram showing various water types Figure 4: Piper’s diagram showing various water types Figure 4: Piper’s diagram showing various water types Table 5: Correlation of parameters pH EC TDS WT Ca2+ Mg2+ Na+ K+ NH4 HCO- 3 Cl- SO2- 4 NO3 PO3- 4 F- pH 1 EC 0.82** 1 TDS -0.11 -0.26 1 WT 0.54 0.69* -0.48 1 Ca2+ 0.66 0.89** -0.23 0.485 1 Mg2+ 0.68* 0.91 -0.23 0.526 0.99 1 Na+ 0.457 0.704 -0.21 0.45 0.84 0.85 1 K+ 0.42 0.64* -0.18 0.38 0.84 0.83 0.88 1 NH4 -0.16 -0.21 -0.13 -0.42 -0.22 -0.19 -0.21 0.14 1 HCO3- 0.807 0.99 - 0.267 0.679* 0.909 0.92 0.72** 0.661 -0.2 1 Cl- 0.02 -0.28 0.59* -0.49 -0.20 -0.21 -0.027 -0.64 0.11 -0.28 1 SO24- -0.38 -0.45 0.23 -0.46 -0.41 -0.39 -0.38 0.17 -0.45 -0.18 - 0.16 1 Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [264] [Lummi et. Hydrochemistry of Water Sources al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 NO3 0.66* 0.576* -0.13 0.46 0.4 0.39 0.04 - 0.006 -0.2 0,56 - 0.08 - 0.43 1 PO34- 0.65* 0.9** -0.29 0.62* 0.78** 0.77 0.64* 0.56 -0.24 0.9** - 0.37 0.41 0.4 1 F- 0.58* 0.36 0.24 0.12 0.26 0.26 0.24 0.32 0.075 0.35 0.6* - 0.78 0.32 0.24 1 ** 0.05 level of significance and * 0.01 level of significance [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) Microbial Quality Q y The microbial analysis indicated that all water points in March/July 2015 had detectable levels of coliform count between 9 to >1100 MPN (Table 7) with springs having greater contamination than taps. Tcfu varied seasonally, with an increase observed in the wet season than during the dry season especially from Agyati, Ntabuwe, and Mbebali (Table 7). Unlike the first interval 58.3% of the water sources remained contaminated during the second interval of the study and 41.7% of the samples were free from coliform contamination in the dry season, remarkably Mbebali/Nchum/Manji, Nsem and Ntabuwe schemes (Table 7). Table 7: Seasonal variations in Total Coliform counts from water sources Samples MPN for March2015 MPN for July2015 Tcfu March 2018 Mbebali/Nchum/Manji Tap 460 >1100 0 Niko/Njinteh Tap >1100 >1100 1 Bujung/Mankaha Tap 1100 1100 2 Nsem Tap >1100 >1100 0 Ntabuwe Tap 93 >1100 0 Agyati Tap Mbebali/Nchum/Manji Spring Niko/Njinteh Spring Bujung/Mankaha Spring Nsem Spring Ntabuwe Spring Agyati Spring 43 >1100 460 >1100 >1100 210 9 1100 6 0 3 6 4 0 17 e 7: Seasonal variations in Total Coliform counts from water sources Table 7: Seasonal variations in Total Coliform counts from water sources l MPN f M h2015 MPN f J l 2015 T f M h Bacterial growth was visible within 48 hours on the Macconkey, SS and Voilet red bile agar. wth was visible within 48 hours on the Macconkey, SS and Voilet red bile agar Bacterial growth was visible within 48 hours on the Macconkey, SS and Voilet g y g During the second interval 25% of the samples were within WHO guideline of zero coliform per ml of water (Table 8). The mean colony forming unit per ml was 10.26cfu/ml. it was also noticed that total plate count recorded at least 1cfu/ml on 61.9% of the samples. E.coli was the most abundant bacterial found in water samples and 16.6% for salmonella. Notably Mankwi borehole observed 236 Tcfu and rain water 16-20 Tcfu/ml. g y g During the second interval 25% of the samples were within WHO guideline of zero coliform per ml of water (Table 8). The mean colony forming unit per ml was 10.26cfu/ml. it was also noticed that total plate count recorded at least 1cfu/ml on 61.9% of the samples. E.coli was the most abundant bacterial found in water samples and 16.6% for salmonella. Microbial Quality al., Vol.6 (Iss.10): October 2018] At least one of the causative agents which results to water-borne diseases (Salmonella typhi, Escherichia historic, Escherichia coli and Coli Bacilli) were identified in 87% of patients, E. coli 40% and Salmonella typhi 39%. Prevalence of Salmonella typhi was eminent in women (66.8%) mostly from Mbebali, Nchum and Manji (Figure 5). At least one of the causative agents which results to water-borne diseases (Salmonella typhi, Escherichia historic, Escherichia coli and Coli Bacilli) were identified in 87% of patients, E. coli 40% and Salmonella typhi 39%. Prevalence of Salmonella typhi was eminent in women (66.8%) mostly from Mbebali, Nchum and Manji (Figure 5). At least one of the causative agents which results to water-borne diseases (Salmonella typhi, Escherichia historic, Escherichia coli and Coli Bacilli) were identified in 87% of patients, E. coli 40% and Salmonella typhi 39%. Prevalence of Salmonella typhi was eminent in women (66.8%) mostly from Mbebali, Nchum and Manji (Figure 5). y = 32.269x - 133.19 -200 -100 0 100 200 300 400 500 600 700 800 9002003200420052006200720082009201020112012201320142015 Typhoid Intestinal infections Linear (Typhoid ) Figure 5: Water borne diseases causative agents and trend y = 32.269x - 133.19 -200 -100 0 100 200 300 400 500 600 700 800 9002003200420052006200720082009201020112012201320142015 Typhoid Intestinal infections Linear (Typhoid ) 12 15 0 3 247 1 15 20 3 6 14 1 10 6 1 3 7 2 60 85 17 39 72 13 0 50 100 150 200 250 300Niko/Njibujang/ NjintehAgyatiNsemBujung/MankahaMbebali/Nchum/ManjiNtabuwe Number of counts SALMONELLA TYPHI E HISTORIC E COLI E BASILLI Figure 5: Water borne diseases causative agents and trend The health consequences may also be felt by adults but as common in rural settings of Cameroon, most people go for traditional herbs in case of (waterborne) infections. Also, most people might have adapted to this poor water quality such that the most vulnerable are children and likely new users. Adult (+30year) were more inclined to all causative agents particularly Salmonella typhi (54%) and infants within the ages of 0-5years (23% of patients) had Coli bacilli and Salmonella typhi infection. Typhoid had an increasing trend of occurrence from 7.4% in 2010 to 43.3% in 2015. Coli basilli, Escherichia historic and Escherichia coli observed a decreasing trend due to a transit of test to basically salmonella typhi from 2007-2015. Microbial Quality Notably Mankwi borehole observed 236 Tcfu and rain water 16-20 Tcfu/ml. Computed results of data on water-borne diseases from 2 Health Units, in the study area, suggested the presence of intestinal pathogens in water, which may account for diarrhoea, dysentery through intestinal infections in patients. The most common of these illnesses was intestinal infections, mainly affecting children. With regards to patient’s record reviewed (1879 patients) from 2003 to 2015 from the Government Hospital at Manji and from the Presbyterian hospital Nsem. Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [265] [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 , p , ) Table 8: Bacteriological quality of water second interval March 2018 Community/water resource TCfu E.coli Salmonella typhi Mforya HC tap 2 † Manka N spring 0 † Mankwi spring 0 GHS Obang spring 3 † Akossia tap 5 † Niko tap 1 † † Nsem tap 0 Mundum spring 60 † Manka tap 2 † Akossia spring 0 Mankwi tap 0 Nta ya New spring 1 † Ntabuwe tap 0 Tingho HC spring 3 † Nchum tap 0 † Rain water 1 16 † Nta ya old spring 7 † Njimbee spring 0 Agyati tap 6 † † Mundum tap 1 † Mankwi borehole 236 † Mambu Bafcul well 0 Mforya stream 3 † Obang stream 5 † Mforya well 0 Mbakong HC well 3 † Adiemukong tap 0 Tingho stream 17 † Mambu nechu well 0 Nchoho tap 0 Obang Alabam stream 3 † Nchoho stream 0 Njimbee Plan well 0 Mbakong big water 5 † † Tingho tap 5 † Niko spring 3 † † Nsem spring 4 † Ntabuwe spring 2 † Mankaha spring 6 † Nchum spring 0 † Agyati spring 17 † † Rain water 2 20 † Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [266] [Lummi et. al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) (Received: September 18, 2018 - Accepted: October 27, 2018) DOI: 10.5281/zenodo.1491883 At least one of the causative agents which results to water-borne diseases (Salmonella typhi, Escherichia historic, Escherichia coli and Coli Bacilli) were identified in 87% of patients, E. coli 40% and Salmonella typhi 39%. Prevalence of Salmonella typhi was eminent in women (66.8%) mostly from Mbebali, Nchum and Manji (Figure 5). [Lummi et. 4. Discussion The population within the study area depends on pipe-borne water (97%) from springs for domestic activities (98.5%) such as cooking, bathing and drinking. This is similar to the results of Fitsum and Fikirte, (2014), in rural Ethopia who equally found that the communities either use rivers or unprotected springs for their domestic water supplies. Most of the water supply schemes within the study area were constructed about 20 years ago financed by the community and None Governmental Associations (SATA-HEVETAS and Plan International). Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [267] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) The challenges of potable water supply management included; Burst pipes, unplanned development, slope gradients, old existing pipe line and low finance. Henry, (2007), in a study of Assessing the Challenges of Water Supply in Ghana observed that bad pumps and low slope gradients, low input from springs and change in land use wherein forested areas became areas of inhabitation, accordingly led to water crises in Ghana. Challenges of potable water supply was also triggered by variations in rainfall from 2,825mm in 1995 to 2,250mm in 2015, which could lead to a drop in the water table thereby reducing discharge from springs. This is in conformity with Ngounou et al., (2005), in their study of climate variability and impacts on an alluvial aquifer that precipitation has a positive impact on aquifer water level. In addition, Ayonghe, (2001), observed that seasonal variability of climate is one of the most challenging pressures facing hydrological systems and water resources. Although WHO (2012) and UNICEF, (2012) guideline for water, states that an individual should have a daily water consumption need of 20 litters per day within 1km from household, this study indicated a low supply coverage of 9.5% and 6.5 liters, indicating less quantity of water and outreach to the entire population. Following these guidelines, none of the supply system could meet up to 20 liters of water per individual in a day. This is similar to the results of Henry, (2007) in Ghana with a daily water need of 8 liters per person in a day. 4. Discussion However this concentration was far above the WHO threshold of 250mg/l making it safe for drinking. Concentration of nitrate in the study area was found to be comparatively low, ranged from 0 to 1.55mg/l from Nsem tap (NST) and 0 to 0.9mg/l for ammonium. The source of nitrates can usually be traced to agricultural activities and human wastes as observed in the study of Oslon, (2012). The major chemical elements found in volcanic rocks particularly basalts are magnesium, calcium, sodium and potassium (Hem, 1989), giving reasons for the dominant water types being sodium/potassium water type, magnesium water type and bicarbonate water type. These water samples were made up of two hydro chemical facies, including sodium bicarbonate and calcium chloride groundwater (Figure 4). This signifies that they are from granitic and basaltic rocks rich in sodium potassium, calcium and magnesium from a natural origin (Prasad et al., 2009). This study recorded the presence of coliform in all water sources during the first interval for March and July 2015 indicating contamination. WHO drinking water guideline standardizes zero coliform per 100 ml (WHO, 2012), thus rendering water sources unsuitable for drinking. This is similar to the results of Okonko et al., (2008) indicating the presence of coliforms in potable water. There was a significant increase in the number of coliform found in tap water within the months of July and March 2015. This seasonal variation could be as a result of increase runoff and infiltration into springs coupled with burst pipes during water distribution. During the second interval of the study, coliform was also observed in the water samples, with higher concentrations from springs than taps. The presence of coliform confirmed by the detection of Escherichia coli served as an indication of faecal contamination in the water. This could be attributed to the location of latrines close to the catchment area and reservoir tanks. This is similar to the studies of Egwari and Aboaba, (2002), on the environmental impact of bacteriological quality of domestic water supply in Lagos, that organism contained in waste water such as those from latrine are responsible for health defects in children. The detection of high total coliform count in water samples could also be attributed to climate based on a mean water temperature of 22.5oC, which could boost bacteria growth in water. 4. Discussion With regards to the physico-chemical properties the 24 water samples had a pH range of 6.7-7.75 implying they are acidic, which according to Ebbing, (1990), a pH of less than 7 are acidic and solutions with a pH greater than 7 are basic. In addition, according to Taiwo et al., (2012) the pH range of these water samples falls within the WHO, (2012) guideline for drinking water quality of 6.5 to 8.0. The concentration of calcium and magnesium showed a correlation value of 0.997, implying that they come from a similar source, possibly igneous rocks particularly basalts found in the study area. This correlation value also explains the abundance of magnesium (0.02meq/l - 2.2meq/l) and calcium (0.002meq/l -1.81meq/l) in the water sources because these are the most abundant cations found in most igneous rock types with high solubility. This is in accordance with Marskak, (2007), who illustrated that igneous rocks are often dominated by magnesium and calcium minerals. Sodium concentration varied from 0.7 to 1.9 mg/l while that of potassium varied from 0.61mg/l for Ntabuwe spring (NTS) to 1.82 mg/l for Niko/Njinteh tap (NNT). Potassium had a positive correlation with sodium (r=0.88) signifying that their concentrations are directly related. This ties with Pidwirny, (2007), whose study showed that felsic magma such as rhyolites and granites are usually rich in sodium and potassium. This direct relationship indicates the presence of sodium and potassium bearing minerals from rhyolites and granites in the study area. The concentration of bicarbonate, ranged from 8.1 for AGT to 116.3 mg/l for MKS. This exceptional high concentration could be attributed to the dissociation of common salt deposits in geologic formations exemplified in the work of Witzel, (2001). The results also indicated that chemical elements were significantly correlated with HCO-3, which could be an indication that the aquifer system may have experienced various processes such as ion exchange, and water-rock interaction (Nally et al., 2004). [268] Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) Chloride concentration ranged from 0.03 to 0.23 mg/L, with highest concentration from Agyati spring (AGS). Chloride in groundwater comes from both natural and anthropogenic sources such as the use of inorganic fertilizers, landfill leachates, septic tank effluents, animal feeds, leaching during rainfall, industrial effluents, irrigation drainage, and seawater intrusion (Malin and Carl, 1992). 5. Conclusions and Recommendations Water supply in the study area was grossly inadequate due to multiple reasons; population growth which the old pipelines cannot support, dilapidated pipes, activities carried out in the catchment area, unplanned developments of springing communities, illegal connections, dilapidated pumps and less input from springs. Natural factors such as climate variability accounted for shortage in water supply. This shortage was in conformity with a 9.5% supply coverage and an average of 6.5 liters of water per individual. Water quality assessment per chemical and physical parameters was generally lower than the WHO recommended limits for potable water. The main water type was Sodium Bicarbonate for ground water. The ionic concentrations in the water was largely based on the geology of the area, rain fall input and atmospheric conditions which is largely influenced by climate. Springs and taps had a Total Coliform count of 43 to >1100 per 100ml in March and July, 2015 with more counts in springs than taps in March. The concentration of bacterial was greater in the rain season than the dry season probably due to runoff infiltration in the springs. Nearness of catchment area to latrines and laundry carried out at some of the taps/springs coupled with water temperatures could be accountable for high levels of coliform counts in the water sources. Escherichia coli and Salmonella typhi were the main bacteria found in water samples which accounts for the high prevalence of water borne diseases particularly typhoid which affects hild Springs and taps had a Total Coliform count of 43 to >1100 per 100ml in March and July, 2015 with more counts in springs than taps in March. The concentration of bacterial was greater in the rain season than the dry season probably due to runoff infiltration in the springs. Nearness of catchment area to latrines and laundry carried out at some of the taps/springs coupled with water temperatures could be accountable for high levels of coliform counts in the water sources. p g Escherichia coli and Salmonella typhi were the main bacteria found in water samples which accounts for the high prevalence of water borne diseases particularly typhoid which affects children. Water supply schemes were found to be contaminated during the first interval (March- August 2015) and the second interval (March 2018). 4. Discussion This is similar to the study of Mengnjo et al., (2013), that warm climate and water temperature increase bacteria growth in unconfined aquifers. The unprotected structural nature of the catchment also provides a suitable ground for bacterial infection. The detection of Escherichia coli and Salmonella in the water samples, with Salmonella observed on 7 samples was remarkable. At least one causative agent for water borne diseases (Salmonella typhi, Escherichia historic, Escherichia coli and Coli bacilli) were identified in 87% of patients’ laboratory results. Escherichia coli recorded 40% and Salmonella typhi 39% which could be responsible for water borne diseases such as cholera, typhoid and diarrhea. The prevalence of typhoid which results from Salmonella typhi was eminent in women (66.8%) due to their exposure to water for domestic activities. At the third World Water Forum in Kyoto (2003) it was recognized Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [269] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) (Received: September 18, 2018 - Accepted: October 27, 2018) that, in addition to being prime users of “domestic water”, women used water in their key role in food production and that women and children are most vulnerable to water-related disasters. High prevalence of typhoid was renowned in Mbebali, Nchum and Manji communities where the water catchment has been encroached by inhabitants. Infants within the ages of 0-5years (23% of patients) had coli bacilli and Salmonella typhi infection, this could be attributed to feeding during pregnancy from the mothers as reported in WHO, (2011). Egwari and Aboaba, (2002), in their study on environmental impact of bacteriological quality of domestic water supply in Lagos, explained that children less than 5 years old are the most vulnerable to water-borne diseases. Although rain water can be one of the alternative sources of drinking water due to its low level of contamination and mineral content as compared to water from fountains could be a vehicle to potential health risk (Carolyne and Kui, 2015). However this study indicates that Escherichia coli contamination was present in rain water which could be attributed to dust particles absorbed by water particles in the atmosphere. References [1] Ako, A, Shimada. J, Eyong G.E And Fantong W. (2010). Access to Potable Water and Sanitation in Cameroon Within The Context Of Millennium Development Goals (Mdgs). Journal Of Water Science And Technology, 61(5):1317-39. gy ( ) [2] Ako Aa, Nkeng Eg, Takem, G.E. (2009). Water Quality and Occurrence of Waterborne Diseases in The Douala 4th District, Cameroon. Water Sci. Technol., 59(12): 2321-2329. gy [2] Ako Aa, Nkeng Eg, Takem, G.E. (2009). Water Quality and Occurrence of Waterborne Diseases in The Douala 4th District, Cameroon. Water Sci. Technol., 59(12): 2321-2329. [3] American Public Health Association. (1985). Standard Methods for The Examination of Water and Waste Water. Section 908 And 909.16th Edition,Apha , Washington, D.C [3] American Public Health Association. (1985). Standard Methods for The Examination of Water and Waste Water. Section 908 And 909.16th Edition,Apha , Washington, D.C p g [4] Bafut Council Office. (1964): Files on The Classification of Cases at Njinteh. Bafut Sub-Division Bamenda Cameroon [4] Bafut Council Office. (1964): Files on The Classification of Cases at Njinteh. Bafut Sub-Division Bamenda Cameroon [5] Benson, H.J. (2002). Microbiological Application. Laboratory Manual in General Microbiology. 8th Edition, Mc. Graw Hill Companies, New York. [6] Carolyne, P and Kui, J. M. (2015). Consumption of Rain Water Harvesting in Terms of Water Quality. International Journal of Geomate. Vol 9, No. 2 [7] Centre for Affordable Water and Sanitation Technology. (Cawst, 2009): Introduction to Drinking Water Quality Testing. Journal of Drinking Water Quality Research Foundation United States. [7] Centre for Affordable Water and Sanitation Technology. (Cawst, 2009): Introduction to Drinking Water Quality Testing. Journal of Drinking Water Quality Research Foundation United States. [8] Council Development Plan. (Cdp, 2011). An Elaborate Report with The Support of The National Community Driven Development Programme (Pndp) [ ] gy ( , ) g Water Quality Testing. Journal of Drinking Water Quality Research Foundation United States. [8] Council Development Plan. (Cdp, 2011). An Elaborate Report with The Support of The National Community Driven Development Programme (Pndp). y g [8] Council Development Plan. (Cdp, 2011). An Elaborate Report with The Support of The National Community Driven Development Programme (Pndp). [9] Djaouda, M.; Njiné, T.; Liang, S.; Ebang Menye, D.; Gaké, B.; Zébazé Togouet, S.H.; Nola, M. (2014). Bacteriological Quality of Well Waters in Garoua, North Cameroon. Water Qual. Expo. Health 2014, 6, 161–176. [10] Domenico, P.A., Schwartz, F.W. (1990). Physical and Chemical Hydrology. John Wiley and Sons. 411:485-497. 5. Conclusions and Recommendations The Microbial quality of the water supply schemes were poor and not suitable for drinking and a public health concern following WHO guideline for drinking water. There is need for a Standardized Water Resource Management Policy and an implement of Integrated Water Resources Management (IWRM) plan. [270] Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) Acknowledgements Our utmost gratitude to Laboratoire d’Analyse Geochemie et l’Eau (LAGE) for the chemical analyses of water samples and the Life Science Laboratory of the University of Buea. Partial funding towards the realisation of this study was provided by the USAID/PERIPERI U grant through the UBuea Consortium. References (1992)."Population and Water Resources: A Delicate Balance”: Washington, Dc: Population Reference Bureau. [24] Marshak, Stephen (2007). Essentials of Geology, 2nd Ed, W. W. Norton. 545p [25] Mengnjo, J.W, Ayonghe, S.N, Engome, R.W, Takeshi, O, Fantong, W.Y (2013). Sources of Bacteriological Contamination of Shallow Groundwater and Health Effects in The Ndop Plain North West Cameroon. Journal of Environmental Science and Water Resources. Vol. 2(4) Wudpecher Journals. [26] Michel, W. (2017). Intercomparison Of Regional Scale Hydrological Models and Climate Change Impact Project For 12 Large River Basins Worldwide. [27] Nally, W.H and Methta, A.J. (2004). Sediment Transportation and Deposition in Estuary. Vol 11 Pp 78-91 [28] Neba, A. (1999). Modern Geography of The Republic of Cameroon. Neba Publishers P.O Box 778 Bamenda North West Province. [29] Ndenecho, E.N (2007). Upstream Water Resource Management Strategy and Stakeholders Participation. Bamenda, Cameroon: Agwecam Printers. 250p [30] Ndjama J, Kamgang K.B.V, Sigha N.L, Ekodeck G, Awah T.M (2008). Water Supply, Sanitation and Health Risk in Douala, Cameroon. African Journal of Environmental and Technology, Vol2(12):422-429. [31] Nforba, M., Kabeyene, V and Suh, C. (2011). Regolith Geochemistry and Mineralogy of The Mbalam Itabirite-Hosted Iron Ore District, South Eastern Cameroon. Open Journal of Geology, Vol. 1 No. 2, 2011, Pp. 17-36. , , p [32] Ngwa Nebasina Emmanuel. (1982). Rural Geography of The Bafut Chiefdom. Yaounde Cameroon. p [32] Ngwa Nebasina Emmanuel. (1982). Rural Geography of The Bafut Chiefdom. Yaounde Cameroon. [33] Ngounou, N.B, Mudry, J., Sigha, L.N., Njitchoua, R and Naah, E. (2005). Climate Variability and [ ] g ( ) g p y [33] Ngounou, N.B, Mudry, J., Sigha, L.N., Njitchoua, R and Naah, E. (2005). Climate Variability and Impacts on An Alluvial Aquifer In A Semi-Arid Climate, The Logone-Chari Plain (South Of Lake Chad). Regional Impacts of Climate Change – Impact Assessment and Decision Making. Publish. 296 Isbni-901 1502-13-9. Pp94-100. [34] Njunda, L.A., Jules, C.N., Nsagha,D.S., Ngene, B.A And Tebit, E.K. (2013). Bacteriological Analysis of Drinking Water in The Cameroon Development Corporation (Cdc) Penda Mboko. American Journal for Integrated Health Care,02:26-31 [35] Nola, M.; Nougang, M.E.; Noah Ewoti, O.V.; Moungang, L.M.; Krier, F.; Chihib, N.-E. (2014) Detection of Pathogenic Escherichia Coli Strains in Groundwater in The Yaoundé Region (Cameroon, Central Africa). Water Environ. Journal.2013, 27, 328–337. [36] Okonko, I.O., Adejoye, O.D., Ogunnusi, T.A., Fajobi, E.A and Shittu, O.B. (2008). References [11] Druce, R.G, Nancy, B.B, Elson, K.J and Harcombe, S.B. (1957). The Determination of The Coli Aerogenes Content of Milk and Dairy Equipment by Plating on Violet Red Bile Agar Incubated At 30o. Journal of Applied Bacteriology. Vol 20 No 1 f pp gy [12] Egwari, L.O., Iwuanyanwu, S., Ojelabi, C.I., Uzonchukwu, O and Effiok, W.W. (2005). Bacteriology of Sachet Water Sold in Lagos, Nigeria. East African Medical Journal.85: 235-240. Ü [13] Fewtrell, L., Prüss-Üstün A, Bos R, Gore F, Bartram J. (2007). Water, Sanitation and Hygiene: Quantifying the Health Impact at National and Local Levels In Countries With Incomplete Water Supply And Sanitation Coverage. Environmental Burden of Disease Series, Who, Geneva. No. 15 [14] Fitsum, D. K and Fikirte, D. T. (2014). Challenges of Potable Water Supply System in Rural Ethiopia: The Case of Gonji Kolela Woreda, West Gojjam Zone, Ethiopia. Natural Resources and Conservation. Horizon Research Publishing. Journal of Natural Resources and Conservation Vol. 2(4), Pp. 59 - 69 [15] Gereta, E.J. (2004). The Importance of Water Quality and Quantity in The Tropical Ecosystem Tanzania. Norwegian University of Science and Technology Trondheim. [16] Gleick, P.H. (1996). Water Resources. In Encyclopaedia of Climate and Weather. Oxford University Press, New York Vol 2. [17] Harvey, R. (2007). Community-Managed Water Supplies in Africa: Sustainable or Dispensable? Community Development Journal Vol 42 No 3 Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [271] [Lummi et. al., Vol.6 (Iss.10): October 2018] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 (Received: September 18, 2018 - Accepted: October 27, 2018) [18] Hem H.D (1989), Study and Interpretation of The Chemical Characteristics of Natural Waters. (3rd Ed). U.S. Geological Survey Water-Supply Paper 22 54. [19] Henry, W. D. (2007) Assessing the Challenges of Water Supply in Urban Ghana: The Case of North Teshie. Royal Institute of Technology (Kth) Stockholm, Sweden. [20] Katte Vy, Fonteh Mf, Guemuh Gn (2003). Domestic Water Quality in Urban Centres in Cameroon: A Case Study of Dschang In the West Province. Afr. Water Journal. Pilot Edition: 43-54. [21] Kuitcha D, Kamgang Kbv, Sigha Nl, Lienou G, Ekodeck Ge (2008). Water Supply, Sanitation and Health Risks in Yaounde, Cameroon. Afr. J. Environ. Sci. Technol., 2(11): 379-386. [22] Kuitcha D, Ndjama J, Tita Am, Lienou G, Kamgang Kbv, Ateba Bh, Ekodeck Ge (2010). Bacterial Contamination of Water Points of The Upper Mfoundi Watershed, Yaounde, Cameroon. Afr. Journal. Microbiol. Res., 4(7): 568574. [23] Malin, Falkenmark And Carl Widstrand. References Microbial and Physiochemical Analysis of Different Water Samples Used for Domestic Purposes in Abeokuta And Ojota, Lagos State, Nigeria. African Journal of Biochemistry. j g g f f y [37] Omam, C.M and Ayonghe, S.N. (2015). An Assessment of The Potability Of Some Sachet Water Brands Sold in Cameroon. Journal of The Cameroon Academy of Science, Vol.12 No.3 [38] Olson R. J, D.L. Hughes, J.A. Gore, And M.P. Brossett. (2010). Candidate Reference Conditions. Rapid Bio Assessment of Stream Health. Crc Press, Boca Raton, Florida. Chapter 4. Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [272] ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 ISSN- 2350-0530(O), ISSN- 2394-3629(P) DOI: 10.5281/zenodo.1491883 [Lummi et. al., Vol.6 (Iss.10): October 2018] (Received: September 18, 2018 - Accepted: October 27, 2018) [39] Patricia, B. (2014). Water Our Most Important Natural Resource. The Green Dandelion. University of Rochester. [40] Pidwirnymicheal. (2006). "Characteristics of Igneous Rocks". Fundamentals of Physical Geography.2nd Edition. [41] Piper, A.M. (1984). A Graphical Proceduce In the Geochemical Interpretation of Water Analysis Trans. Am. Geophysical Union, In Ground Water Quality Series: Vol 25, 914-1923 [42] Regassa A. E. (2016): Income Determinants of Irish Potato (Solanum Tuberosum L.) Growers: The Case of West Arsi Zone of Oromia Regional State, Ethopia. Net Journal of Agricultural Science, 4(1): 1-8. [43] Scott, J And Richard,G.T. (2015). Intensive Rainfall Recharges Tropical Ground Water. Environmental Research Letters. Science Daily. [44] Small, H., Stevens, T.S and Bauman, W.C. (1975). Novel Ion Exchange Chromatography Method Using Conductimetric Detection. Analytical Chemistry 47:1801-1809. [45] Strasser, Balz. (2000). Keeping the Water Flowing: A Sociological and Financial Analysis of Financing Infrastructure, Operation and Maintenance of A Rural Water Supply Scheme: The Case Study Of Kedjom Ketinguh In The Northwest Province Of Cameroon. Wageningen University y j g g g y [46] Sophocleous, Marios (2002). "Interactions Between Groundwater and Surface Water: The State of The Science". Hydrogeology Journal 10: 52–67. [47] Takem, G.E.; Chandrasekharam, D.; Ayonghe, S.N.; Thambidurai, P. (2010). Pollution Characteristics of Alluvial Groundwater from Springs and Bore Wells in Semi-Urban Informal Settlements of Douala, Cameroon, Western Africa. Environ. Earth Sci. 2010, 61, 287–298. [48] Taiwo. A.A., Idowu, O., Lanre-Iyanda, Y.A., Jolaoso, A. A., Jiboku, O.O., T.O. And Taiwo, A.G. (2012). Physico Chemical and Bacteriological Analyses of Sachet Water Samples in Abeokuta Metropolis. Global Advance Research Journal of Agricultural Science, 1(1): 001-006. [ ] , , , y , , , , , , , (2012). *Corresponding author. E-mail address: samayonghe @yahoo.com References Physico Chemical and Bacteriological Analyses of Sachet Water Samples in Abeokuta Metropolis. Global Advance Research Journal of Agricultural Science, 1(1): 001-006. p [49] Temgoua E (2011). Chemical and Bacteriological Analysis of Drinking Water from Alternative Sources in The Dschang Municipality, Cameroon. Journa Environ. Prot., 2: 620-628 [50] Unicef/Who (2012). Progress on Drinking Water and Sanitation: Joint Monitoring Program for Water Supply and Sanitation. Unicef, Newyork. [51] World Health Organization. Preventing Diarrhoea Through Better Water, Sanitation and Hygiene: Exposures and Impacts in Low and Middle-Income Countries; Who: Geneva, Switzerland, 2014; Pp. 1–48. [52] Who and Unicef (2000): Global Water Supply and Sanitation Assessment 2000 Report World Health Organization and United Nations Children's Fund, 2000. [53] World Health Organization (2011): Taking Sex and Gender into Account In Emerging Infectious Disease Programme: An Analytical Framework. All Rights Reserved. Who Press, 20 Avenue Appia, 1211 Geneva 27, Switzerland Http://www.granthaalayah.com ©International Journal of Research - GRANTHAALAYAH [273]
https://openalex.org/W4310712854	https://www.guncelretina.com/upload/article/GRD-2022-0000389_Publish.pdf	Turkish	null	Preoperative / Intraoperative and Postoperative Intravitreal Applications and Preoperative Evaluation in Diabetic Vitrectomy (Advantages of Early Vitrectomy / Late Vitrectomy Approaches and Current Recommendations)	Güncel retina	2,023	cc-by	5,617	Giriş nan olgularda da kullanılmaktadır. Bu derlemede diyabetik vit- rektomide preoperatif, intraoperatif ve postoperatif intravitreal uygulamalardan ameliyat öncesi değerlendirmeden (erken vitrek- tomi / geç vitrektomi yaklaşımlarının avantajları ve güncel öneri- ler) bahsedilecektir. Günümüzde diyabetik retinopati tedavisinde birçok medikal ve lazer tedavileri uygulanabilmesine rağmen bu tedavilerin ye- terli olmadığı diyabetik maküler ödemde (DMÖ) ve vitre içi he- moraji (VH), traksiyonel retina dekolmanı (TRD) gibi proliferatif diyabetik retinopatiye (PDR) sekonder gelişen komplikasyonlar nedeniyle diyabetik vitrektomi uygulanabilmektedir. PDR’nin di- agnostik bulgusu olan neovaskülarizasyonun gelişiminde vasküler endotelial büyüme faktörü (VEGF) önemli bir yere sahiptir. Bazı çalışmalarda anti-VEGF ajanların pre-proliferatif ve proliferatif DRP’nin gerilemesine ve VH’nin açılmasına katkıda bulunduğu gösterilmiştir.[1-3] ABSTRACT Treatment modalities in diabetic retinopathies, advances in surgical techniques, and a better understanding of the physiopathology of diabetic retinopathy continue to change the way we approach the disease. In addition to anti-VEGF agents, steroid agents are also used in the treatment of diabetic retinopathy. These intravitreal agents are used in the medical treatment of diabetic retinopathy as well as in cases where diabetic vitrectomy is planned/performed. Costs and non-adherence to treatment may highlight the option of early vitrectomy in patients with proliferative diabetic retinopathy Keywords: Diabetic retinopathy, Intravitreal injection, Vitrectomy ÖZ Diyabetik retinopatilerdeki tedavi modaliteleri, cerrahi tekniklerdeki gelişmeler, diyabetik retinopati fizyopatolojisinin daha iyi anlaşılması, hastalığa yaklaşım şeklimizi değiştirmeye devam etmektedir. Diyabetik retinopati tedavisinde Anti-VEGF ajanların yanısıra steroid yapıda ajanlar da kullanılmaktadır. Bu intravitreal ajanlar diyabetik retinopatinin medikal tedavisinde kullanıldığı gibi diyabetik vitrektomi planlanan/yapılan olgularda da kullanılmaktadır. Maliyetler, tedaviye uyumsuzluk, proliferatif diyabetik retinopatili hastalarda erken vitrektomi seçeneğini ön plana çıkarabilir. Anahtar Kelimeler: Diyabetik retinopati, vitrektomi, inravitreal enjeksiyon Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler) Preoperative / Intraoperative and Postoperative Intravitreal Applications and Preoperative Evaluation in Diabetic Vitrectomy (Advantages of Early Vitrectomy / Late Vitrectomy Approaches and Current Recommendations) Muhammed BATUR / ORCID No: 0000-0003-3748-4646, Veysi YILDIZ / ORCID No: 0000-0002-4467-2365, *Sinan EKİNCİ / ORCID No: 0000-0003-4783-5016 Doç Dr., Van Yüzüncü Yıl Üniversitesi Tıp Fakültesi Göz Hastalıkları AD, Van, Türkiye Uzm Dr., Batman Eğitim ve Araştırma Hastanesi Göz Hastalıkları Kliniği, Batman, Türkiye * Araştırma Görevlisi Doktor, Van Yüzüncü Yıl Üniversitesi Tıp Fakültesi Göz Hastalıkları AD, Van, Türkiye Geliş Tarihi/Received: 12.07.2022 Kabul Tarihi/Accepted: 25.09.2022 DOI: 10.37783/CRJ-0371 Yazışma Adresi/Address for Correspondence: Muhammed BATUR, Van Yüzüncü Yıl Üniversitesi Tıp Fakültesi Hastanesi, Göz Hastalıkları AD, Zeve Kampüsü, 65080, Tuşba, Van, Türkiye Tel./Phone: +90 532 5647956, E-posta/E-mail: muhammedbatur@gmail.com Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler) Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler) 2 Batur M, Yıldız V, Ekinci S. Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler). Güncel Retina 2023; 7 (4): 257-262 Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler) Preoperative / Intraoperative and Postoperative Intravitreal Applications and Preoperative Evaluation in Diabetic Vitrectomy (Advantages of Early Vitrectomy / Late Vitrectomy Approaches and Current Recommendations) Muhammed BATUR / ORCID No: 0000-0003-3748-4646, Veysi YILDIZ / ORCID No: 0000-0002-4467-2365, *Sinan EKİNCİ / ORCID No: 0000-0003-4783-5016 Doç Dr., Van Yüzüncü Yıl Üniversitesi Tıp Fakültesi Göz Hastalıkları AD, Van, Türkiye Uzm Dr., Batman Eğitim ve Araştırma Hastanesi Göz Hastalıkları Kliniği, Batman, Türkiye * Araştırma Görevlisi Doktor, Van Yüzüncü Yıl Üniversitesi Tıp Fakültesi Göz Hastalıkları AD, Van, Türkiye Geliş Tarihi/Received: 12.07.2022 Kabul Tarihi/Accepted: 25.09.2022 DOI: 10.37783/CRJ-0371 Yazışma Adresi/Address for Correspondence: Muhammed BATUR, Van Yüzüncü Yıl Üniversitesi Tıp Fakültesi Hastanesi, Göz Hastalıkları AD, Zeve Kampüsü, 65080, Tuşba, Van, Türkiye Tel./Phone: +90 532 5647956, E-posta/E-mail: muhammedbatur@gmail.com yona neden olarak sekonder retina dekolmanı gelişebilmektedir. [15,18] Triamsinolon asetonid (TA) suda çözünmeyen bir steroid olup antiinflamatuar, antianjiojenik ve kan-retina bariyerini sta- bilize edici etkilerinden dolayı postoperatif inflamasyonun kont- rol altına alınmasını sağlayarak PVR gibi birçok komplikasyonun önüne geçebilir.[18,19] Tavşanlar üzerinde yapılmış deneysel bir çalışmada intravitreal TA’nın PVR, traksiyonel retina dekolmanı gelişimini engellediği ve intravitreal neovaskülarizasyonu inhibe ettiği gösterilmiştir.[20] Yapılan diğer bir çalışmada ise vitrektomi sırasında intravitreal TA kullanımının vitreusun görünürlüğünü arttırarak epiretinal membranın soyulmasını ve özellikle PVD ge- lişmemiş olgularda vitreusun ayrılmasını kolaylaştırdığı gösteril- miştir.[21] Vitrektomi sırasında kullanılan TA’nın cerrahi ile büyük kısmı temizlenmektedir, fakat bir kısım TA’nın retina yüzeyinde birkaç gün kaldığı ve postoperatif inflamasyonda yeterli miktar- da azalmayı sağladığı gösterilmiştir.[22] Bilindiği gibi kortikoste- roidler yatkınlığı olan bireylerde göz içi basıncı yükseltebilirler. Yapılan bir çalışmada pars plana vitrektomi sırasında kullanılan TA’nın göz içi basıncında anlamlı bir yükselmeye neden olmadığı gösterilmiştir. Bu durum PPV sırasında kullanılan TA’nın büyük kısmının cerrahi sonrasında temizlenmesiyle açıklanmıştır.[23] mi öncesi yapılan intravitreal bevacizumab enjeksiyonunun pos- toperatif VH gelişme oranını etkilemediğini ifade eden çalışmalar da vardır.[8,9] Liang ve arkadaşlarının yaptığı çalışmada vitrektomi sonunda yani hava-sıvı değişimi sonrasında yapılan intravitre- al ranibizumab’ın postoperatif VH ve maküler ödem insidansını azaltarak daha iyi bir görme keskinliğini netice verdiğini ve pre- operatif yapılan intravitreal anti-VEGF’in cerrahi esnasında göz içinden temizlenerek etkisinin azaldığını belirtmişlerdir.[10] Yapılan bir derleme ve meta-analiz çalışmasında diyabetik vitrektomi öncesi anti-VEGF enjeksiyonu yapılan ve yapılmayan gruplar karşılaştırılmış, anti-VEGF enjeksiyonu yapılan grupta iatrojenik retinal yırtık gelişimi, endodiatermi kullanımı, cerrahi süresi ve intraoperatif kanama oranının daha az olduğu gösteril- miştir. Yine aynı çalışmada postoperatif olarak kanama emilim zamanının ve tekrar VH gelişme oranının daha az olduğu, sonuç görme keskinliğinin daha iyi olduğu, tekrar retina dekolman geli- şimi ve reoperasyon açısından bir fark olmadığı tespit edilmiştir. [11] Bilindiği üzere anti-VEGF enjeksiyonu sonrası retinal vasküler sızıntı, retinal kalınlık ve ödem gerilemektedir. Bu durumun retina dokusunda traksiyona karşı direnç oluşturmasıyla ve intraoperatif kanama riskinin azalması ile iyi bir cerrahi görüntü sağlanmakta böylece iatrojenik retinal yırtık riskini azalttığı düşünülmektedir. TA’nın cerrahi ile büyük kısmının temizlenmesi göz içi basın- cının yükselmemesi açısından avantaj sağlarken etkisinde azal- maya neden olmaktadır. Deksametazon implant, bilindiği gibi intravitreal uygulanmakta ve 6 aya kadar etki gösterebilmektedir. Literatürde dirençli DMÖ’de vitrektomi ile deksametazon implant kombinasyonunun uygulandığı çalışmalar mevcuttur. Vitrektomi ile cerrahi sırasında uygulanan deksametazon implantının pos- toperatif intravitreal enjeksiyon sayısında azalma sağladığı ve di- rençli DMÖ’de makula kalınlığının azalmasına katkıda bulunduğu gösterilmiştir.[24,25] Öte yandan PDR’de anti-VEGF enjeksiyonu sonrası fibrozisin artmasına bağlı olarak traksiyonel retina dekolmanı sıklığında artış olabilmektedir.[12,13] Anti VEGF sonrası vitreustaki sitokin seviyeleri değişmekte ve bu da TRD’de ve vitreoretinal fibroziste kötüleşmeye yol açabilmektedir. VEGF, Plasental büyüme fak- törü (PIGF) and VE-cadherin konsantrasyonu önemli düzeyde düşerken, insülin benzeri büyüme faktörü-1 (IGF-1) konsantras- yonu artmaktadır.[14] Bu nedenle anti-VEGF enjeksiyonu sonrası retinal fibrozis agreve olmadan vitrektomi yapılması önerilmek- tedir. Yapılan bir çalışmada şiddetli PDR’li 211 hastaya intravit- real bevacizumab (IVB) enjeksiyonu yapılmış, 11 (%5,2) olguda TRD’de progresyon görülmüş veya yeni TRD gelişmiş. Enjeksiyon sonrası TRD görülme zamanı ortalama 13 gün (3-31 gün) olarak ifade edilmiştir.[15] Li ve arkadaşlarının çalışmasında PDR’de vit- rektomi öncesi ranibizumab enjeksiyonu yapılmış, aköz hümörde intraoküler fibrozisin göstergesi olan bağ doku büyüme faktörü (CTGF) ve VEGF sevilerine bakılmıştır. Enjeksiyon sonrası 1. günde aköz hümör VEGF seviyesinde anlamlı düşüş izlenmiş, enjeksiyon sonrası 3. günde ise aköz hümörde VEGF sevilerinde ciddi düşüş görülürken CTGF sevilerinde enjeksiyon öncesine göre önemli farklılık izlenmemiştir.[7] Bu çalışmada anti-VEGF enjeksiyonu sonrası 1-3 gün TRD’de progresyona neden olmadığı veya yeni TRD oluşumunu indüklemediği gösterilmiş ve vitrekto- mi için güvenli aralık olduğu ifade edilmiştir.[7] Aynı şekilde başka bir çalışmada bevacizumab’ın 1. günde etki göstermeye başladığı gösterilmiştir.[16] 8 Batur M, Yıldız V, Ekinci S. Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler). Güncel Retina 2023; 7 (4): 257-262 Diyabetik vitrektomide preop/ intraop intravitreal uygulamalar Diyabetik vitrektomi cerrahisi öncesi intravitreal anti-VEGF enjeksiyonu etkili ve güvenilir bir tedavidir.[4,5] PDR’de görülen fibrovasküler membranlar retinaya genellikle sıkı bağlanmakta ve cerrahi esnasında çekme veya soyma manevralarında ciddi kanama görülebilmektedir. Bu durum cerrahiyi zorlaştırmakta ve cerrahi süresini uzatmaktadır. Cerrahi öncesi yapılan anti-VEGF enjeksiyonu fibrovasküler membranın vasküler komponentinde gerileme yaparak retinadan ayrılmasını kolaylaştırarak intraope- ratif ve postoperatif erken dönemde VH riskini azaltarak cerrahiyi daha güvenilir yapmaktadır.[6,7] Bunun aksine diyabetik vitrekto- Diyabetik retinopati tedavisinde bevasizumab, ranibizumab ve aflibersept gibi Anti-VEGF ajanların yanısıra dexametazon implant ve Triamsinolon asetonid gibi steroid yapıda ajanlar da kullanılmaktadır. Bu intravitreal ajanlar diyabetik retinopatinin medikal tedavisinde kullanıldığı gibi diyabetik vitrektomi planla- 257 Pars Plana Vitrektominin Rolü Dünya çapında 463 milyondan fazla diyabet hastası vardır ve bunların yaklaşık %1.4’ünde proliferatif diyabetik retinopa- ti (PDR) vardır.[35] Diyabetlilerin %60’ının PDR geliştireceği ve bunların yaklaşık yarısının ciddi görme kaybı yaşayacağı tahmin edilmektedir.[36] Tedavi modaliteleri, cerrahi tekniklerdeki geliş- meler, diyabetik retinopati ve PDR’nin fizyopatolojisinin daha iyi anlaşılması, hastalığa yaklaşım şeklimizi değiştirmeye devam etmektedir. Maliyetler, tedaviye uyumsuzluk, PDR riski altındaki hastalarda erken vitrektomi üzerinde inceleme yapılmasını gerekli kılmıştır. Kötü göze vitrektomi ve PRP yapılan PDR’li 120 gözün ret- rospektif bir incelemesinde, vitrektomize gözlerde vitrektomi uy- gulanmamış gözlere kıyasla daha iyi görsel ve anatomik sonuçlar gösterilmiştir.[48] Bu durum özellikle 50 yaş altı vitreusu tamamen ayrılmayan hastalarda belirgindir. Elli yaş altında, son ortalama görme keskinliği post-vitrektomi gözlerde 20/80, PRP ile tedavi edilen gözlerde 20/400 olarak ölçülmüştür. Vitrektomize gözlerin %88’inde, vitrektomi uygulanmamış gözlerin %24’ünde görme keskinliği (GK)’nde artış izlenmiştir. Vitrektomi uygulanan grup- ta %8, vitrektomi yapılmayan grupta %36 el hareketi veya daha az GK ile sonuçlanmıştır. İki grup arasındaki bir diğer çarpıcı fark, 8 yıllık takip süresince çoklu tedavi gerektiren göz sayısıdır. Vit- rektomi uygulanan gözlerin %16’sında ek lazer, %12’sinde yeni- den ameliyat ve %40’ında katarakt görülmüştür. Vitrektomi uygu- lanmayanların %72’si ek lazer, %60’ında vitrektomi gerekmiş ve bunların %72’si TRD, %16’sı TRRD, %16’sı inoperabl olarak kabul edilmiş, %24’ünde katarakt gelişmiştir. Vitrektominin geleneksel tedaviye kıyasla faydaları, daha genç hastalar kadar dramatik ol- masa da, 50 yaş üstü grupta da kaydedilmiştir. Bu büyük olasılıkla yaşlı hastalar arasında daha yüksek total vitreus dekolmanı insi- dansından kaynaklanmaktadır. Elli yaş üstü grupta, vitrektomi uy- gulanmış gözlerin ortalama postoperatif GK’sı 20/80 iken, vitrek- tomi uygulanmamış gözlerde 20/200’dir. Vitrektomi uygulanmış gözlerin %14’ü, vitrektomi uygulanmamış gözlerin %36’sı, 20/200 veya daha kötü VA ile sonuçlanmıştır. Vitrektomi uygulanmamış hastalarda ayrıca %70 ek lazer, %28 vitrektomi, %27 TRD, %11 inoperabl RD ve %16 katarakt progresyonu gösterilmiştir. Vitrek- tomi uygulanan gözlerde ek lazer %8 ve yeniden ameliyat gerekti- ren %8 olarak bildirilmiştir. Bu verilerden, 50 yaş altı grupta vit- rektomi uygulanmadığında 4 kat daha fazla gözde el hareketi veya daha az GK oluştuğunu ve 50 yaş üstü grupta 2.5 kat daha fazla gözde, 2/200 veya daha az GK ile sonuçlandığını görüyoruz.[48] Khan’ın Hindistan’da yaptığı 10 yıllık bir takip çalışması benzer bulgular göstermiştir.[41] PDR için, Diyabetik Retinopati Çalışması (DRS) sonuçların- dan bu yana panretinal fotokoagülasyon (PRP) standart olarak uygulanmaktadır. Arka Hyaloid ve Arka Vitreus Dekolmanının Rolü daha yavaş olduğu ve ranibizumab enjeksiyonu sayısının daha yüksek olduğu ifade edilmiştir. Çalışmanın 3 yıllık sonuçlarında ise tedaviye anatomik cevap ve toplam enjeksiyonu sayısı bakı- mından gruplar arasında fark olmadığı belirtilmiştir.[32] Chen ve arkadaşları vitrektomize ve non-vitrektomize gözlerde DMÖ te- davisinde intravitreal ranibizumab tedavisini karşılaştırmışlar, 6 aylık takipte intravitreal ranibizumab tedavisi her iki grupta et- kili olmakla birlikte non-vitrektomize gözlerde daha iyi görme keskinliği, görsel kazanç ve santral foveal kalınlıkta (SFK) daha fazla azalma olduğu görülmüştür. Ayrıca vitrektomize gözlerde iyileşmenin daha yavaş ve enjeksiyonu sayısının daha fazla olduğu gözlenmiştir ( 6 aylık tedavide 5.05 – 4.1 enjeksiyon).[33] PDR’li gözler, prognozları ve potansiyel komplikasyonları farklı olan iki kategoriye ayrılabilir. Genç diyabetiklerde daha sık görülen, yapışık veya kısmen yapışık arka hyaloidli gözler ve to- tal arka vitreus dekolmanı (PVD) olanlar. PDR ve total PVD’li gözlerde, neovasküler damarlar sadece retina yüzeyinde büyüye- bilir ve bu gözlerde yeni damarlara traksiyon uygulayacak vitreus yoktur. Bu nedenle, bu gözler nadiren kanar ve TRD veya TRRD geliştirmez. Tersine, yapışık veya kısmen ayrılmış arka hyaloid olan gözler vitreus hemorajisi ve TRD’ye ilerler. Üç yıl boyunca izlenen 403 gözden oluşan bir seride, total PVD’si olan hiçbir göz- de retinopati progresyonu görülmezken, PVD’si olmayan gözlerin %44’ünde progresyon ve kısmi PVD’si olan gözlerin %100’ünde vitreus hemorajisi (VH) veya TRD’ye progresyon olmuştur.[46] Total PVD, diyabetik retinopatili gözleri uzun süre stabilize eder. Bu durum erken vitrektomi yapılmasının değerlendirilmesi ve gündeme alınmasını gerekli kılmıştır. Yapılan çalışmalarda aynı sonuçlar hyaloidi çıkarılan vitrektomi yapılmış gözlerde de görül- müştür ve bu gözler maküler ödem veya TRD gelişmeden onlarca yıl stabil kalabildiği belirtilmiştir.[47] Vitrektomize gözlerde dexametazon implantın etkinliğinin araştırıldığı bir çalışmada 17 gözün 10 (%58.8) tanesinde 6 ay bo- yunca devam eden SFK’da azalma ve görme keskinliğinde iyileşme görülmüştür, 7 (%41.2) gözde 6 aydan önce 2. defa dexametazon implant ihtiyacı olmuştur.[34] 259 Batur M, Yıldız V, Ekinci S. Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler). Güncel Retina 2023; 7 (4): 257-262 Diyabetik vitrektomide postop intravitreal uygulamalar yg Vitrektomize gözlerde intravitreal uygulanan ilaçların farma- kokinetiğinde değişiklikler olmaktadır.[26] Hayvan çalışmalarında vitrektomize gözlerde intravitreal bevacizumab, ranibizumab ve TA’nın temizlenme oranlarının daha hızlı olduğu gösterilmiştir.[27- 29] Makak gözleriyle yapılmış bir çalışmada intravitreal aflibersept ve ranibizumab’ın yarı ömürlerinin non-vitrektomize gözlerde sırasıyla 2.2 ve 2.3 gün olduğu, lensektomi ile birlikte vitrektomi yapılmış gözlerde ise sırasıyla 1.5 ve 1.4 gün olduğu gösterilmiştir. [27] Başka bir çalışmada intravitreal bevacizumab‘ın yarı-ömrü ise lensektomi ile birlikte vitrektomi yapılmış gözlerde 1.5 gün ola- rak ölçülmüştür.[28] Tavşanlar üzerinde intravitreal TA ile yapılmış bir çalışmada vitrektomize gözlerde TA konsantrasyonunun hızlı bir şekilde azaldığı ve yarı ömrünün non-vitrektomize gözlerde 2.8 gün iken vitrektomize gözlerde 1.5 gün olduğu gösterilmiştir. [29] Öte yandan vitrektomize gözlerde intravitreal bevacizumab ve ranibizumabın farmakokinetiğinde değişme olmadığını gösteren çalışmalar da mevcuttur. Bu çalışmalarda lens koruyucu vitrekto- mi yapılmış ve lensektominin dahil edildiği vitrektomide intravit- real uygulanan ajanların ön kamaraya geçişi ve trabeküler ağ yolu ile eliminasyonun artmasıyla ilaçların yarılanma ömürlerinde azalma olabileceği ifade edilmiştir.[30,31] Vitrektomize gözlerde intravitreal uygulanan ilaçların farma- kokinetiğinde değişiklikler olmaktadır.[26] Hayvan çalışmalarında vitrektomize gözlerde intravitreal bevacizumab, ranibizumab ve TA’nın temizlenme oranlarının daha hızlı olduğu gösterilmiştir.[27-l 29] Makak gözleriyle yapılmış bir çalışmada intravitreal aflibersept ve ranibizumab’ın yarı ömürlerinin non-vitrektomize gözlerde sırasıyla 2.2 ve 2.3 gün olduğu, lensektomi ile birlikte vitrektomi yapılmış gözlerde ise sırasıyla 1.5 ve 1.4 gün olduğu gösterilmiştir. Diyabetik traksiyonel retina dekolmanı nedeniyle vitrektomi öncesi anti-VEGF uygulanan hastalarda endotamponad olarak silikon kullanımı ihtiyacı azalmaktadır. Yapılan bir çalışmada vit- rektomi öncesi intravitreal bevacizumab yapılan hastalarda %26 oranında silikon kullanılmışken, intravitreal bevacizumab yapıl- mayan hastalarda %70 oranında silikon kullanıldığı tespit edilmiş- tir (P = .039).[17] Vitrektomize gözlerde uygulanan intravitreal ilacın temiz- lenme hızında artma ile ilacın etki süresinde kısalma olabilir ve bunun sonucu olarak daha sık intravitreal enjeksiyon gerekebilir. DRCR.net çalışmasında vitrektomize ve non-vitrektomi gözlerde ranimizumab ile erken/geç lazer tedavisi kombinasyonu karşılaş- tırılmış ve 1 yıllık sonuçlarda vitrektomize gözlerde iyileşmenin Vitrektomi sonrası inflamasyon, PVR gibi birtakım kompli- kasyonların önemli bir nedenidir. Postopertatif inflamatuar hüc- relerden salgılanan kimyasal mediatörler vitreoretinal arayüzeyde bazı değişikliklere ve retinal glial hücreler ile retina pigment epite- li hücrelerinde aktivasyona neden olur. Bu hücrelerde aktivasyona sekonder oluşan proliferasyon preretinal membranda kontraksi- 258 nucun önemli prognostik göstergesidir. Sklerotik damarlar, yaygın maküler tutulum ve düşük kaliteli IVFA vakalarında iskeminin ipuçları olarak kullanılabilir. de PRP, PVD’si olmayan gözlerde tamamen koruyucu değildir. Bunların çoğunda 6. ayda VH gelişmekte ve Protokol S’deki gibi mükemmel takibe rağmen, %19’u düzenli takiplere ve anti VEGF enjeksiyonlarına rağmen vitrektomi gerektirmektedir.[39,43,50] Maküler OCT, iç retinaya ve dış retinanın bütünlüğüne ne- ovasküler yapılara odaklanarak maküler morfolojinin daha fazla detayını sağlayabilir. OCT ile doğrulanan makula tutulumu has- tanın cerrahi adaylığını etkileyebilir. OCT, TRD’yi traksiyonel re- tinoskizisten ayırt etmek için de kullanılabilir. Preoperatif OCT bulguları da prognostik değer taşır. Shah ve arkadaşları, maküler TRD’lerdeki dış retinal tabakalardaki bozulmanın, koryokapilla- risten anormal difüzyona sekonder olabileceğini ve cerrahi ona- rıma rağmen düşük görme ile sonuçlanabileceğini öne sürdüler. [54] Dooley ve arkadaşları tarafından preoperatif santral maküler kalınlık ve subretinal sıvının da OCT’de önemli bir prognostik gösterge olduğu kaydedilmiştir.[55] Vitrektomideki ilerlemelere rağmen, TRD’li gözlerde %3- 34’ünde GK kaybı, %7-22’sinde anatomik başarısızlık ve %29- 37’lik iyatrojenik retinal hasar insidansı mevcuttur.[51,52] Tedavi yükü sadece kaybedilen zaman, seyahat ve iş gücü değil, aynı zamanda devletlerin maliyetidir. Lin ve arkadaşları, ABD’de PDR için erken vitrektomi maliyetini PRP ve ranibizumab ile karşılaştırdı.[53] İki yıllık vitrektomi maliyeti, PRP maliyeti ile karşılaştırılabilir, ancak Protokol S’de 5 yılda görüldüğü gibi, PRP gözlerinin %50’si ek tedavi gerektirir.[39] Vitrektomisiz tedavi edi- len gözler birden fazla ek tedavi gerektirirken, kombine vitrektomi ve PRP ile tedavi edilen gözler minimum yardımcı tedavi gerek- tirir ve uzun yıllar stabil kalır.[41,47] Bu vitrektomize gözlerde tek- rar kanama nadirdir, hyaloid çıkarıldıktan sonra TRD’ye ilerleme mümkün değildir ve diyabetik makula ödemi gelişimi de nadirdir. B-scan ultrasonografi, özellikle vitreus kanaması gibi durum- larda tanıda yardımcıdır. USG-B ayrıca cerrahi planlama için de faydalıdır, dekolmanın topografisinin görselleştirilmesine ve int- raoperatif olarak posterior vitreus dekolmanını taklit edebilen vitreoskizisin tanımlanmasına izin verir. Optik sinir ve retina ile arka kutup arasındaki arka vitreus ilişkisi bir USG-B taramasında açıkça görülebilir. Şiddetli NPDR ve PDR’si olan hastaları, tam PVD’si olan göz- ler ve kısmi PVD’si olan veya PVD olmayan gözler olmak üzere iki gruba ayırmalıyız. Total PVD’li gözler daha az agresif PRP veya uyumluysa anti-VEGF ile tedavi edilebilir. PVD veya kısmi PVD si olmayan gözler, hyaloid çıkarma ve PRP ile vitrektomi- den fayda görecektir. Bu gruptaki vitrektomi, vakaların büyük bir bölümünde gözleri stabilize edebilir ve bu hastaların maliyetini ve tedavi yükünü azaltabilir. Muayene, görüntüleme ve teşhis Ameliyat öncesi GK, ameliyat sonrası sonuçlarla ilişkilidir; Williams ve arkadaşları, ameliyat öncesi görmenin 5/200 veya daha iyi olmasının, ameliyat sonrası görsel sonucun iyileşmesi için en güçlü prognostik faktör olduğunu bulmuşlardır. Ön segment bulguları prognostik olabilir; ameliyat öncesi iris neovaskülarizasyonu (NVI) ve katarakt dikkatlice değerlendiril- melidir. Tedavi planlaması açısından bunlar göz önüne alınmalıdır. Açıklama ç Yazarlar bu çalışmada herhangi bir çıkar çatışması olmadığını bildirmektedir. Ayrıntılı bir binoküler dilate muayene çok önemlidir. Bu mu- ayene, cerrahi planlama ve yaklaşım için gerekli olan en yüksek düzeyde stereoskopik ayrıntı sağlar. Arka vitreusun yapısına ve optik sinirden tutunma noktalarına ve retina dekolmanlarına dik- kat edilmelidir. Günümüzün elektronik tıbbi kayıtlarında, fundus çizimleri yaygın uygulamadan düşmüştür, ancak vitreoretinal ya- pıların ayrıntılı değerlendirilmesi ve çizilmesi, cerrahi bir yaklaşı- mın planlanmasında yardımcı olabilir. Fundus fotoğrafı kolaylıkla elde edilse de kapsamlı bir oftalmoskopik muayeneden elde edilen 3 boyutlu detayların yerini tutamaz. Sonuç Diyabetik vitrektomi öncesi yapılan intravitreal anti-VEGF enjeksiyonunun postoperatif VH gelişme oranını etkilemediğini ifade eden çalışmalar olmakla birlikte genel olarak cerrahi öncesi 1-3. günlerde intravitreal anti-VEGF enjeksiyonu intraoperatif ve postoperatif erken dönemde VH riskini azaltarak etkili ve güveni- lir bir cerrahi tedavi olanağı sağlamaktadır. Vitrektomize gözlerde uygulanan intravitreal anti-VEGF ve steroidlerde temizlenme hızında artma ile ilacın etki süresinde kısalma olabilir ve bunun sonucu olarak daha sık intravitreal en- jeksiyon gerekebilmektedir. Bu tedavi yöntemi sadece gözleri sta- bilize etmekle ve görmeyi korumakla kalmayacak, aynı zamanda hem hasta hem de toplum için maliyeti düşürürken görme engelli hasta sayısını da azaltacaktır. Erken vitrektomi, özellikle ileri evre hastalığı olan genç diyabetik hastalar için yararlı, uygun maliyetli, uzun vadeli stabilize edici bir tedavidir. Özellikle üretken yılların- da sayısız hasta için körlüğün önlenmesine yardımcı olabilir. Ameliyat Öncesi Planlama Preoperatif detaylı bir muayene , intravenöz floresein anjiyog- rafi (IVFA), optik koherens tomografi (OCT) ve USG-B taraması yapılmalıdır. Tam PRP’ye rağmen ek tedavi gerekliliğinin devam etmesi, lazer tedavisinin görme alanı kusurları ve gece görüşünde azalma gibi potansiyel yan etkileri nedeniyle DR’nin anti-VEGF ile takip ve tedavisi popülerlik kazanmıştır. Ancak aylık enjeksiyonlar, za- man kaybı, seyahat zorluğu ve tedavi maliyeti gibi sebeplerle teda- viye uyumsuzluk oranı göz ardı edilemeyecek düzeydedir. PDR’li gözlerde tedavi uyumsuzluğuna bağlı progresyon, retina dekol- manı ve geri dönüşü olmayan görme kaybıyla birlikte neovasküler glokom erken vitrektomi seçeneğini özellikle genç diyabetiklerde ön plana çıkarabilir. Pars Plana Vitrektominin Rolü Tam PRP’ye rağmen, gözlerin %5’inde vitrek- tomi gerektiren komplikasyonlar gelişmektedir.[37] ETDRS, lazerin görme kaybını %50 oranında azalttığını ancak yarısının ek teda- viler gerektirdiğini göstermiştir.[38] Protokol S’nin PRP kolunda; PRP ile tedavi edilen gözlerin %45’inde ek lazer, %46’sında vitreus hemorajisi, %19’unda vitrektomi ve %4’ünde neovasküler glokom geliştiği bildirilmiştir. [39,40] Khan tarafından yapılan çalışmada, tam PRP’den sonra gözlerin %89’unun ek PRP’ye ve %33’ünün vitrektomiye ihtiyacı olduğu vurgulanmıştır.[41] y y ğ g Lazer tedavisinin görme alanı kusurları ve gece görüşünde azalma gibi potansiyel yan etkileri nedeniyle DR’nin anti-VEGF ile takip ve tedavisi popülerlik kazanmıştır. Ancak aylık enjeksi- yonlara devam etmek çeşitli nedenlerle zor olabilir. Zaman kaybı, seyahat zorluğu ve tedavi maliyetinin de dikkate alınması gerekir. Hastaların %50’ye varan oranında, 5 yıllık bir dönemde 6 aydan fazla takipte randevusuna gelmemektedir.[42] PDR’li gözlerde at- lanan enjeksiyonlar veya kaçırılan randevular, sıklıkla hastalığın traksiyonel retina dekolmanı (TRD), kombine traksiyonel ve yırtıklı retina dekolmanı (TRRD) veya geri dönüşü olmayan gör- me kaybıyla birlikte neovasküler glokoma progresyonuna neden olabilir.[39,43,44] Ayrıca anti-VEGF tedavisinin potansiyel kompli- kasyonlarının da dikkate alınması gerekir. Protokol S’de 5 yılda %0.3’lük endoftalmi insidansı, traksiyonel dekolman alanlarının potansiyel progresyonu ve %19’unun 5 yılda vitrektomi gerektir- diği vurgulanmıştır.[39] Diyabetik retinopati tedavisindeki ilerle- melere rağmen, PDR için ideal tedavi, özellikle yaşlı diyabetiklere göre daha yüksek körlük insidansına sahip olan ilerlemiş hastalığı olan genç diyabetikler arasında hala belirsizliğini korumaktadır.[45] Tedavide aksamalar diyabetik hastalar arasında sık gözlenir ve PDR hızla TRD’ye ilerleyebilir. Tedavide aksama nedenleri arasın- da komorbit hastalıklar, tedaviye uyumsuzluk, ekonomik nedenler ve daha genç yaş yer almaktadır.[43,49,50] Anti-VEGF ile tedavi edi- len gözlerde takiplerin aksamasının sonuçları daha yıkıcı olabilse 259 Batur M, Yıldız V, Ekinci S. Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler). Güncel Retina 2023; 7 (4): 257-262 betic Retinopathy. J Ophthalmol. 2018;2018:4927259. betic Retinopathy. J Ophthalmol. 2018;2018:4927259. proliferative diabetic retinopathy. Graefes Arch Clin Exp Opht- halmol. 2008;246(12):1699-1705. betic Retinopathy. J Ophthalmol. 2018;2018:4927259. 19- Sakamoto T, Miyazaki M, Hisatomi T, Nakamura T, Ueno A, Itaya K, et al. Triamcinolone-assisted pars plana vitrectomy im- proves the surgical procedures and decreases the postoperative blood-ocular barrier breakdown. Graefes Arch Clin Exp Opht- halmol. 2002;240(6):423-429. 4- Russo A, Longo A, Avitabile T, Bonfiglio V, Fallico M, Boscia F, et al. Incidence and Risk Factors for Tractional Macular Deta- chment after Anti-Vascular Endothelial Growth Factor Agent Pretreatment before Vitrectomy for Complicated Proliferative Diabetic Retinopathy. J Clin Med. 2019;8(11):1960. 20- Tano Y, Chandler D, Machemer R. Treatment of intraocular pro- liferation with intravitreal injection of triamcinolone acetonide. Am J Ophthalmol. 1980;90(6):810-816. 5- Hernández-Da Mota SE, Nuñez-Solorio SM. Experience with intravitreal bevacizumab as a preoperative adjunct in 23-G vit- rectomy for advanced proliferative diabetic retinopathy. Eur J Ophthalmol. 2010;20(6):1047-1052. 21- Meyer CH. Current treatment approaches in diabetic macular edema. Ophthalmologica. 2007;221(2):118-31. 6- Yeh PT, Yang CM, Lin YC, Chen MS, Yang CH. Bevacizumab pretreatment in vitrectomy with silicone oil for severe diabetic retinopathy. Retina. 2009;29(6):768-774. 22- Ueno A, Enaida H, Hata Y, Hisatomi T, Nakamura T, Mochizuki Y, et al. Long-term clinical outcomes and therapeutic benefits of triamcinolone-assisted pars plana vitrectomy for proliferative vit- reoretinopathy: a case study. Eur J Ophthalmol. 2007;17(3):392- 398. 7- Li S, Yang Y, Zou J, Zeng J, Ding C. The efficacy and safety of intravitreal injection of Ranibizumab as pre-treatment for vitrec- tomy in proliferative diabetic retinopathy with vitreous hemorr- hage. BMC Ophthalmol. 2022;10;22(1):63. 23- Enaida H, Hata Y, Ueno A, Nakamura T, Hisatomi T, Miyazaki M, et al. Possible benefits of triamcinolone-assisted pars plana vitrectomy for retinal diseases. Retina. 2003;23(6):764-70.f 8- Lo WR, Kim SJ, Aaberg TM Sr, Bergstrom C, Srivastava SK, Yan J, et al. 3rd. Visual outcomes and incidence of recurrent vitreous hemorrhage after vitrectomy in diabetic eyes pretreated with be- vacizumab (avastin). Retina. 2009;29(7):926-931. 24- Altun A, Kanar HS, Aki SF, Arsan A, Hacisalihoglu A. Effective- ness and Safety of Coadministration of Intravitreal Dexametha- sone Implant and Silicone Oil Endotamponade for Proliferative Diabetic Retinopathy with Tractional Diabetic Macular Edema. J Ocul Pharmacol Ther. 2021;37(2):131-137. 9- Sato T, Morita S, Bando H, Sato S, Ikeda T, Emi K. Early vitreous hemorrhage after vitrectomy with preoperative intravitreal beva- cizumab for proliferative diabetic retinopathy. Middle East Afr J Ophthalmol. 2013;20(1):51-55. h 25- Flaxel CJ, Edwards AR, Aiello LP, Arrigg PG, Beck RW, Bressler NM, et al. Factors associated with visual acuity outcomes after vitrectomy for diabetic macular edema: diabetic retinopathy cli- nical research network. Retina. 2010;30(9):1488-1495. 10- Liang X, Zhang Y, Wang JX, Wang LF, Huang WR, Tang X. Intra- vitreal ranibizumab injection at the end of vitrectomy for diabe- tic vitreous hemorrhage (Observational Study). Medicine (Balti- more). 2019;98(20):e15735. 26- Stefánsson E. Physiology of vitreous surgery. Graefes Arch Clin Exp Ophthalmol. 2009;247(2):147-163. 11- Zhao LQ, Zhu H, Zhao PQ, Hu YQ. A systematic review and meta-analysis of clinical outcomes of vitrectomy with or without intravitreal bevacizumab pretreatment for severe diabetic retino- pathy. Br J Ophthalmol 2011;95:1216-22. 27- Niwa Y, Kakinoki M, Sawada T, Wang X, Ohji M. Ranibizumab and Aflibercept: Intraocular Pharmacokinetics and Their Effects on Aqueous VEGF Level in Vitrectomized and Nonvitrectomi- zed Macaque Eyes. Invest Ophthalmol Vis Sci. 2015;56(11):6501- 6505. 12- Moradian S, Ahmadieh H, Malihi M, Soheilian M, Dehghan MH, Azarmina M. Intravitreal bevacizumab in active progressive proliferative diabetic retinopathy. Graefes Arch Clin Exp Opht- halmol. 2008;246(12):1699-1705. 28- Kakinoki M, Sawada O, Sawada T, Saishin Y, Kawamura H, Ohji M. Effect of vitrectomy on aqueous VEGF concentration and pharmacokinetics of bevacizumab in macaque monkeys. Invest Ophthalmol Vis Sci. 2012;53(9):5877-5880. 13- Van Geest RJ, Lesnik-Oberstein SY, Tan HS, Mura M, Goldsch- meding R, Van Noorden CJ, et al. A shift in the balance of vascu- lar endothelial growth factor and connective tissue growth factor by bevacizumab causes the angiofibrotic switch in proliferative diabetic retinopathy. Br J Ophthalmol. 2012;96(4):587-590. 29- Chin HS, Park TS, Moon YS, Oh JH. Difference in clearance of intravitreal triamcinolone acetonide between vitrectomized and nonvitrectomized eyes. Retina. 2005;25(5):556-560. 30- Ahn J, Kim H, Woo SJ, Park JH, Park S, Hwang DJ, et al. Pharma- cokinetics of intravitreally injected bevacizumab in vitrectomi- zed eyes. J Ocul Pharmacol Ther. 2013;29(7):612-618. 14- Ozer F, Tokuc EO, Bal Albayrak MG, Akpinar G, Kasap M, Kara- bas VL. Comparison of before versus after intravitreal bevacizu- mab injection, growth factor levels and fibrotic markers in vitre- ous samples from patients with proliferative diabetic retinopathy. Graefes Arch Clin and Exp Ophthalmol (2022) 260:1899–1906 h 31- Ahn SJ, Ahn J, Park S, Kim H, Hwang DJ, Park JH, et al. Int- raocular pharmacokinetics of ranibizumab in vitrectomized versus nonvitrectomized eyes. Invest Ophthalmol Vis Sci. 2014;55(1):567-573. Erratum in: Invest Ophthalmol Vis Sci. 2015;56(3):1473-1474. 261 Batur M, Yıldız V, Ekinci S. Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler). Güncel Retina 2023; 7 (4): 257-262 Kaynaklar 1- Ip MS, Domalpally A, Sun JK, Ehrlich JS. Long-term effects of therapy with ranibizumab on diabetic retinopathy severity and baseline risk factors for worsening retinopathy. Ophthalmology. 2015;122(2):367-374. 1- Ip MS, Domalpally A, Sun JK, Ehrlich JS. Long-term effects of therapy with ranibizumab on diabetic retinopathy severity and baseline risk factors for worsening retinopathy. Ophthalmology. 2015;122(2):367-374. 2- Adamis AP, Altaweel M, Bressler NM, Cunningham ET Jr, Davis MD, Goldbaum M, at al. Macugen Diabetic Retinopathy Study Group. Changes in retinal neovascularization after pegaptanib (Macugen) therapy in diabetic individuals. Ophthalmology. 2006;113(1):23-28. IVFA, spektral OCT ve B-scan ultrasonografi de preoperatif değerlendirmelerde değerli olabilir. IVFA, spektral OCT ve B-scan ultrasonografi de preoperatif değerlendirmelerde değerli olabilir. IVFA, ameliyat öncesi planlama için önemlidir. Makula tutu- lumu TRD’de IVFA yorumunu zorlaştırsa da, makula perfüzyonu- nun değerlendirilmesi denenmelidir. Maküler iskemi, görsel so- IVFA, ameliyat öncesi planlama için önemlidir. Makula tutu- lumu TRD’de IVFA yorumunu zorlaştırsa da, makula perfüzyonu- nun değerlendirilmesi denenmelidir. Maküler iskemi, görsel so- 3- Moradian S, Ahmadieh H, Malihi M, Soheilian M, Dehghan MH, Azarmina M. Intravitreal bevacizumab in active progressive 260 interruptions. Am J Ophthalmol 2019;204:13–18. 35. International Diabetes Federation. Diabetes facts & figures. 2019. Available at: https://www.idf.org/aboutdiabetes/what-is-diabe- tes/facts-figures.html. [Accessed 7 December 2020]. 45. Liao M, Wang X, Yu J, Meng X, Liu Y, Dong X, et al. Characte- ristics and outcomes of vitrectomy for proliferative diabetic reti- nopathy in young versus senior patients. BMC Ophthalmol 2020; 20:416. 36. Shani M, Eviatar T, Komaneshter D, Vinker S. Diabetic retino- pathy-incidence and risk factors in a community setting – a lon- gitudinal study. Scand J Prim Healthcare 2018; 36:237–241. 46. Ono R, Kakehashi A, Yamagami H, Sugi N, Kinoshita N, Saito T, et al. Prospective assessment of proliferative diabetic retinopathy with observations of posterior vitreous detachment. Int Ophthal- mol 2005; 26(1):15–19. 37. The Diabetic Retinopathy Study Research Group. Photocoagu- lation treatment of proliferative diabetic retinopathy. Clinical application of Diabetic Retinopathy Study (DRS) findings, DRS report number 8. Ophthalmology. 1981; 88:583–600. 47. Luliano L, Corbelli E, Bandello F, Codenotti M. Protective effect of vitrectomy on the course of diabetic retinopathy: a case report. Eur J Ophthalmol 2020; Online ahead of print. 38. Flynn HW, Chew EY, Simons BD, Barton FB, Remaley NA, Ferris FL, Pars plana vitrectomy in the early treatment Diabetic Reti- nopathy. ETDRS Report Number 17. The Early Treatment Dia- betic Retinopathy Study Research Group. Ophthalmology. 1992; 99:1351–1357. 48. Berrocal MH, Acaba-Berrocal L. Early pars plana vitrectomy for proliferative diabetic retinopathy: update and review of current literature. Curr Opin Ophthalmol. 2021;32(3):203-208 39. Gross JG, Glassman AR, Liu D, et al., Diabetic Retinopathy Cli- nical Research Network. Five-year outcomes of panretinal pho- tocoagulation vs intravitreous ranibizumab for proliferative dia- betic retinopathy: a randomized clinical trial. JAMA Ophthalmol 2018; 136:1138–1148. 49. Green M, Tien T, Ness S. Predictors of lost to follow-up in pa- tients being treated for proliferative diabetic retinopathy. Am J Ophthalmol 2020;216:18–27. 50. Obeid A, Gao X, Ali FS, Talcott K E, Aderman CM, Hyman L, et al. Loss to follow-up in patients with proliferative diabetic retino- pathy after panretinal photocoagulation or intravitreal anti-VE- GF injections. Ophthalmology 2018; 125(9):1386–1392. 40. Writing Committee for the Diabetic Retinopathy Clinical Re- search Network. Panretinal Photocoagulation vs Intravitreous Ranibizumab for Proliferative Diabetic Retinopathy: A Rando- mized Clinical Trial. JAMA. 2015;314(20):2137–2146. 51. Abdelhadi AM, Helaly HA, Abuelkeir A. Evaluation of retinal detachment after diabetic vitrectomy: causes and ways of mana- gement. Clin Ophthalmol 2020;14:53–60. 41. 15- Arevalo JF, Maia M, Flynn HW Jr, Saravia M, Avery RL, Wu L, et al. Tractional retinal detachment following intravitreal beva- cizumab (Avastin) in patients with severe proliferative diabetic retinopathy. Br J Ophthalmol. 2008;92(2):213-216. 32- Bressler SB, Melia M, Glassman AR, Almukhtar T, Jampol LM, Shami M, et al. Diabetic Retinopathy Clinical Research Network. Ranibizumab plus prompt or deferred laser for diabetic macular edema in eyes with vitrectomy before anti-vascular endothelial growth factor therapy. Retina. 2015;35(12):2516-28. 16- Avery RL, Pearlman J, Pieramici DJ, Rabena MD, Castellarin AA, Nasir MA, et al. Intravitreal bevacizumab (Avastin) in the treatment of proliferative diabetic retinopathy. Ophthalmology. 2006;113(10):1695.e1-15. 33- Chen YY, Chen PY, Chen FT, Chen YJ, Wang JK. Comparison of efficacy of intravitreal ranibizumab between non-vitrectomized and vitrectomized eyes with diabetic macular edema. Int Opht- halmol. 2018;38(1):293-299. 17- Uzel MM, Citirik M, Ilhan C, Inanc M. The effect of Bevacizu- mab pretreatment on the choice of endotamponade in diabetic tractional retinal detachment. Ophthalmic Surg Lasers Imaging Retina. 2016;47:924–929 34- Özdemir HB, Hasanreisoğlu M, Yüksel M, Ertop M, Gürelik G, Özdek Ş. Effectiveness of Intravitreal Dexamethasone Implant Treatment for Diabetic Macular Edema in Vitrectomized Eyes. Turk J Ophthalmol. 2019;49(6):323-327. 18- Cui J, Chen H, Lu H, Dong F, Wei D, Jiao Y, et al. Efficacy and Safety of Intravitreal Conbercept, Ranibizumab, and Triamcino- lone on 23-Gauge Vitrectomy for Patients with Proliferative Dia- 261 Batur M, Yıldız V, Ekinci S. Diyabetik Vitrektomide Preoperatif / İntraoperatif ve Postoperatif İntravitreal Uygulamalar ve Ameliyat Öncesi Değerlendirme (Erken Vitrektomi / Geç Vitrektomi Yaklaşımlarının Avantajları ve Güncel Öneriler). Güncel Retina 2023; 7 (4): 257-262 Khan R, Surya J, Rajalakshmi R, Rani PK, Anantharaman G, Go- palakrishnan M, et al. Need for vitreous surgeries in proliferative diabetic retinopathy in 10 years follow-up: India Retinal Disease Study (IRDS) group. Report no. 2. Ophthalmic Res 2021;64:432– 439 52. Storey PP, Ter-Zakarian A, Philander SA, De Koo LO, George M, Humayun MS, et al. Visual and anatomical outcomes after dia- betic traction and traction-rhegmatogenous retinal detachment repair. Retina 2018; 38(10):1913–1919. 42. Suresh R, Yu HJ, Thoveson A, Swisher J, Apolinario M, Zhou B, et al. Loss to follow-up among patients with proliferative diabetic retinopathy in clinical practice. Am J Ophthalmol 2020; 215:66– 71. 53. Lin J, Chang JS, Yannuzzi NA, Smiddy WE. Cost evaluation of early vitrectomy versus panretinal photocoagulation and intra- vitreal ranibizumab for proliferative diabetic retinopathy. Opht- halmology 2018; 125(9):1393–1400 43. Obeid A, Su D, Patel SN, Uhr JH, Borkar D, Gao X, et al. Outco- mes of eyes lost to follow-up with proliferative diabetic retinopat- hy that received panretinal photocoagulation versus intravitreal anti-vascular endothelial growth factor. Ophthalmology. 2019; 126:407–413. 54. Shah VA, Brown JS, Mahmoud TH. Correlation of outer retinal microstucture and foveal thickness with visual acuity after pars plana vitrectomy for complications of proliferative diabetic reti- nopathy. Retina. 2012;32(9):1775 55. Dooley I, Laviers H, Papavasileiou E, Mckechnie C, Zambarakji H. Spectral domain ocular coherence tomography findings pre- and post vitrectomy with fibrovascular membrane delamination for proliferative diabetic retinopathy. Eye (Lond). 2016;30(1):34-39 44. Wubben TJ, Johnson MW, Anti-VEGF Treatment Interruption Study Group. Anti-vascular endothelial growth factor therapy for diabetic retinopathy: consequences of inadvertent treatment Doç. Dr. Muhammed BATUR Doç. Dr. Muhammed BATUR Doç. Dr. Muhammed BATUR 2005 yılında Ankara Üniversitesi Tıp Fakültesi’nden mezun olduktan sonra, aynı yıl Yüzün- cü Yıl Üniversitesi Tıp Fakültesi Göz Hastalıkları Anabilim Dalı’nda uzmanlık eğitimine baş- ladı ve 2010 yılında Göz Hastalıkları Uzmanı oldu. Van Eğitim ve Araştırma Hastanesi’nde 2010-2012 yılları arasında mecburi hizmet görevini tamamladı. 2012 tarihinde Yüzüncü Yıl Üniversitesi Tıp Fakültesi Göz Hastalıkları Anabilim Dalında yardımcı doçent olarak çalış- maya başladı, 2018 yılında doçent ünvanını aldı ve 2017 yılından itibaren Göz Hastalıkları AD Başkanlığı görevini sürdürmektedir. 2009 yılında Ankara Eğitim ve Araştırma Hastane- si’nde üveit ve oküloplastik cerrahi konusunda gözlemci, 2010 yılında Almanya Teubingen Üniversitesi Göz Hastalıkları Kliniğinde vitreoretinal cerrahi konusunda gözlemci, 2011 yı- lında Kartal Eğitim ve Araştırma Hastanesi Göz Bankası Tıbbi Müdür eğitimi, 2015 yılında Ankara Gülhane Askeri Tıp Akademisi’nde vitreoretinal cerrahi eğitimi, 2016 yılında İrlan- da, Dublin’de dil eğitimlerini aldı. 2012 yılında Van Eğitim ve Araştırma Hastanesi Göz Ban- kası’nı kurdu ve kuruluşundan beri Göz Bankası Tıbbi Müdür görevini devam etmektedir. Daha önce Türk Oftalmoloji Derneği Genç Oftalmologlar Grubu Yönetim Kurulu Üyeliği yaptı ve halen TOD Diyarbakır Şubesi Yönetim Kurulu Üyesi ve Şube Başkanlığı görevini yapmaktadır. Van Yüzüncü yıl Üniversitesi Sağlık Bilimleri Enstitüsü Sinirbilim Doktora Programı kurucularından olup halen öğretim üyeliği görevi devam etmektedir. 262
https://openalex.org/W1952992482	https://europepmc.org/articles/pmc4526085?pdf=render	English	null	Continuous expression of CD83 on activated human CD4+ T cells is correlated with their differentiation into induced regulatory T cells	Molecular Medicine Reports	2,015	cc-by	5,395	DOI: 10.3892/mmr.2015.3796 Abstract. CD83 is a widely recognized surface marker for mature dendritic cells, which are essential for priming naïve CD4+ T cells into effector cells. However, CD83 is also expressed on activated CD4+ T cells, which remains an enigma in T‑cell mediated immunity. Therefore, the identification of the biological features and regulation of the expression of CD83 on activated CD4+ T cells is important in understanding the function of CD83 in the adaptive immune response. The present study revealed a time‑dependent manner of the expression of CD83 on anti‑CD3/CD28‑stimulated human CD4+ T cells, which is characterized by the maximum expression at day 2 and a significant decrease at day 3. The reduced expression is not a result of a reduced rate of cell proliferation. The activation of interleukin‑2 and secretion of interferon‑γ accumulated progressively from day 1 to 3. Of note, sustained expression of CD83 was observed when CD4+ T cells were induced by transforming growth factor-β to differentiate into CD4+CD25+ forkhead box P3+ regulatory T (iTreg) cells. Confocal immunofluorescence microscopy analysis demonstrated that CD83 was highly co‑localized with CD25 on activated CD4+ T cells. In conclusion, the findings of the present study suggested that the continuous expression of CD83 on activated human CD4+ T cells is correlated with their differentiation into iTreg cells. MOLECULAR MEDICINE REPORTS 12: 3309-3314, 2015 MOLECULAR MEDICINE REPORTS 12: 3309-3314, 2015 Correspondence to: Dr Liwen Chen or Professor Shihe Guan, Department of Laboratory Medicine, The Second Hospital of Anhui Medical University, 678 Furong Road, Hefei, Anhui 230601, P.R. China E‑mail: lw_ch@hotmail.com E-mail: shiheguan@126.com Continuous expression of CD83 on activated human CD4+ T cells is correlated with their differentiation into induced regulatory T cells Departments of 1Laboratory Medicine and 2Medical Research Center, The Second Hospital of Anhui Medical University, Hefei, Anhui 230601; 3Department of Medical Oncology, The First Hospital of Anhui Medical University, Hefei, Anhui 230022, P.R. China Received July 29, 2014; Accepted April 15, 2015 Received July 29, 2014; Accepted April 15, 2015 E‑mail: lw_ch@hotmail.com Materials and methods Lymphocyte purification and cell culture. Using Ficoll‑Hypaque density gradient centrifugation at 900 x g, mononuclear cells were isolated from the blood of healthy donors who provided written informed consent. This study was approved by the Ethics Committee of the Second Hispital of Anhui Medical University (Hefei, China). The mononuclear cell suspension (8 ml) was added into a T‑25 culture flask and incubated at 37˚C with 5% CO2 for 2 h. The cells were gently agitated, the non‑adherent cells were aspirated, and adherent monocytes and B cells were discarded. Alternatively, untouched CD4+ T cells were purified from mononuclear cells using a CD4+ T-cell isolation kit (cat. no. 130-096-533) and magnetic columns (cat. no. 130-042-306) (both from Miltenyi Biotech, Bergisch Gladbach, Germany). This procedure routinely provided >95% pure CD4+ T cells. Non‑adherent lymphocytes or purified CD4+ cells were cultured in RPMI‑1640 containing 10% fetal bovine serum, 10 mM 4-(2-hydroxyethyl)-1-piper azineethanesulfonic acid and 1% penicillin‑streptomycin (all purchased from Invitrogen Life Technologies, Carlsbad, CA, USA) at 1x106 cells/ml in 24‑well plates, in the pres ence of pre-coated agonistic murine anti human CD3 monoclonal antibody (mAb; clone UCHT-1; cat. no. 555329; 0.5 µg/ml) and soluble agonistic murine anti human CD28 mAb (clone ANC28.1/5D10; cat. no. 177-820; 1.0 µg/ml), purchased from BD Biosciences, Franklin Lakes, NJ, USA and Ancell, Bayport, MN, USA, respectively. To assess the importance of TGF‑β regulation on the expression of CD83 on CD4+ T cells, a final concentration of 2 ng/ml exogenous TGF‑β (Sigma‑Aldrich, St. Louis, MO, USA) was added to the cultures at day 0. All cells were harvested 1, 2 or 3 days following incubation for flow cytometry and confocal immu nofluorescence microscopy analysis. Confocal immunofluorescence microscopy. Following stimu lation with anti‑CD3/CD28 for 2 days, purified CD4+ T cells were washed twice with ice‑cold phosphate‑buffered saline (PBS) and fixed in 4% (w/v) paraformaldehyde (Sigma‑Aldrich) in PBS for 15 min at room temperature (20˚C). The cells were subsequently stained with FITC‑conjugated anti‑CD25 mAb and PE‑conjugated anti‑CD83 mAb, and incubated on ice for 30 min in the dark. Fluorescently‑labeled cells were observed and recorded by an examiner, in a blinded manner, using a Zeiss LSM 410 inverted laser scan microscope (Carl Zeiss, Oberkochen, Germany) equipped with Ar488, Kr568 and HeNe633 lasers. Non‑specific fluorescence was assessed by incubating cells with FITC‑ or PE‑conjugated isotype controls. Statistical analysis. CHEN et al: CONTINUOUS EXPRESSION OF CD83 ON ACTIVATED HUMAN CD4+ T CELLS 3310 canonical CD3/CD28 signal and transforming growth factor (TGF)‑β on the expression of CD83 on CD4+ T cells as well as on their differentiation into CD4+CD25+ forkhead box (Fox) P3+-induced regulatory T (iTreg) cells were investigated. stimulation, and the cells were stained with phycoerythrin (PE)‑conjugated anti‑CD83, fluorescein isothiocyanate (FITC)‑conjugated anti‑CD4 or FITC‑conjugated anti‑CD25 (All from BioLegend, San Diego, CA, USA) and incubated on ice for 30 min in the dark. The controls were prepared using FITC‑ or PE‑conjugated isotype controls. The intracellular staining of TGF‑β‑treated CD4+ T cells with anti‑Foxp3 mAb was performed using an anti‑human Foxp3 staining set (eBioscience, San Diego, CA, USA) according to the manu facturer's instructions. The cells were subsequently analyzed using a flow cytometer (Beckman Coulter Epics XL; Beckman Coulter, Miami, FL, USA) by using a two‑color acquisition method and the data were analyzed by using FlowJo 7.6.1 software (Treestar, Inc., Ashland, OR, USA). Materials and methods Values presented in the figures are repre sentative of at least three independent experiments and are expressed as the mean± standard derivation. Comparisons between the expression levels of CD83 at days 1, 2 and 3 were determined by Student's t‑test with SPSS 13.0 software (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference. Introduction CD83 is a type I transmembrane glycoprotein with a highly glycosylated N‑terminal ectodomain and a short C‑terminal intracellular domain. Being absent from the majority of resting cells, CD83 is predominantly induced on the surface of mature dendritic cells (mDCs) and activated T and B lympho cytes (1‑3). Increasing evidence has demonstrated the significant regulatory roles of CD83 in the central and peripheral immune system (4-14). CD83 was demonstrated to be essential for the lineage commitment of CD4+ T cells in the thymus (4). A previous study has demonstrated that membrane CD83 (mCD83) promotes the expression of major histocompatability complex (MHC) class II and CD86 on mDCs by inhibiting membrane‑associated RING‑CH1 (MARCH1)‑dependent ubiquitination and degradation of the two target molecules (5). The co‑stimulatory effects of mCD83 on mDCs per se for CD4+ T cells have remained controversial (4,6‑8). However, soluble CD83 (sCD83), which is produced predominantly by ectodomain shedding, has clear suppressive effects in vitro and in vivo (9‑14). A previous study by our group demonstrated that sCD83 suppresses T‑cell proliferation and the secretion of interleukin (IL)‑2 and interferon (IFN)‑γ through prosta glandin E2 (PGE2) produced by monocytes (15). A previous study demonstrated that native or forced expression of CD83 confers an immunosuppressive function to CD4+ T cells (16). However, a previous study using short hairpin (sh)RNA‑mediated gene silencing of CD83 on CD4+ T cells revealed a reduced proliferation and lower production of IL‑2 and IL‑17 by the CD4+ T cells, indicating that CD83 serves as a positive co‑stimulator for CD4+ T cells (17). It was noted that genetic manipulation may cause unintended effects to the target cells. On the other hand, modified expres sion of CD83 is likely paralleled by concurrent changes of co‑stimulatory molecules on CD4+ T cells, since CD83 is an important regulator of MHC class II and the expression of CD86 (5). Therefore, the biological and functional definition of the expression of CD83 on CD4+ T cells remains to be elucidated. Key words: CD83, differentiation, activation, co-localization, regulation In the present study, the expression of CD83 on CD4+ T cells was assessed. The effects of stimulation with a Results Time‑dependent expression of CD83 on CD4+ T cells is stimulated by anti‑CD3/CD28. To assess the time‑depen dent kinetics of the expression of CD83 on CD4+ T cells, non‑adherent lymphocytes isolated from mononuclear cells were stimulated with agonistic anti‑CD3/CD28 and the expression of CD83 on CD4+ T cells was determined by flow cytometry 1, 2 and 3 days following stimulation. As shown in Fig. 1A, CD83 was rarely expressed on resting CD4+ T cells (0 days), whereas the percentages of CD83‑positive cells in the total CD4+ T cells were significantly upregulated at day 1 (7.69%) and markedly increased to 21.61% by day 2 (Fig. 1B and C). However, the activation-induced expression of CD83 on CD4+ T cells decreased significantly to 2.30% at day 3 (Fig. 1D). By contrast, unstimulated CD4+ T cells demonstrated less severe increases in the expression of CD83 compared with those in the stimulated group on days 1 and 2, and on day 3, CD83 were similar to those in the stimulated group (Fig. 1E‑H). The differences in the percentage of CD83+ cells (with regard to the total CD4+ T-cell population) between the anti‑CD3/CD28‑stimulated and the unstimulated group were statistically significant (P<0.05) at days 1 and 2 (Fig. 1E). In vitro cell proliferation assay and cytokine detection. As mentioned above, purified CD4+ T cells were collected and stimulated with agonistic anti‑CD3/CD28 at 105/ml in 96‑well plates. Determination of cell proliferation was performed 1, 2 and 3 days later using a Cell Counting kit‑8 (Dojindo, Kumamoto, Japan), according to the manufacturer's instruc tions. Cell viability was determined by measuring the absorbance at 450 nm using a multiwell scanning spectropho tometer (KHB ST-360; Kehua Bio-Engineering, Shanghai, China). The experiments were repeated in triplicate wells. The levels of IL‑2 and IFN‑γ in CD4+ T‑cell culture supernatants were measured in duplicate for each of the serial aliquots using a Human IL-2 Quantikine ELISA kit (cat. no. D2050) and Human IFN-γ Quantikine ELISA kit (cat. no. DIF50), purchased from R&D Systems (Minneapolis, MN, USA), according to the manufacturer's instructions. Flow cytometric analysis. Non‑adherent lymphocytes or puri fied CD4+ T cells were collected at 0, 1, 2 and 3 days following MOLECULAR MEDICINE REPORTS 12: 3309-3314, 2015 3311 Figure 1. Expression of CD83 on non‑adherent lymphocytes. (A) Freshly isolated (0 d) non‑adherent lymphocytes were directly stained with FITC‑labeled CD4 and PE‑labeled CD83 and analyzed by flow cytometry. Results (B‑D) Alternatively, non‑adherent lymphocytes were stimulated with anti‑CD3/CD28 or (F‑H) left unstimulated for 1‑3 days. The cells were subsequently stained with FITC‑labeled anti‑CD4 and PE‑labeled anti‑CD83, and analyzed by flow cytometry. Dot plots are representative of three independent experiments. (E) Values are expressed as the mean ± standard deviation of the percentages of CD83‑positive cells in total CD4+ T cells of three independent experiments (P<0.05, compared with the unstimulated group on the same day). FITC, fluorescein isothiocyanate; PE, phycoerythrin; CD, cluster of differentiation; neg.con, negative control. A B C D E F G H B C D B C D C D A B A D C D B E F G H F E H G Figure 1. Expression of CD83 on non‑adherent lymphocytes. (A) Freshly isolated (0 d) non‑adherent lymphocytes were directly stained with FITC‑labeled CD4 and PE‑labeled CD83 and analyzed by flow cytometry. (B‑D) Alternatively, non‑adherent lymphocytes were stimulated with anti‑CD3/CD28 or (F‑H) left unstimulated for 1‑3 days. The cells were subsequently stained with FITC‑labeled anti‑CD4 and PE‑labeled anti‑CD83, and analyzed by flow cytometry. Dot plots are representative of three independent experiments. (E) Values are expressed as the mean ± standard deviation of the percentages of CD83‑positive cells in total CD4+ T cells of three independent experiments (P<0.05, compared with the unstimulated group on the same day). FITC, fluorescein isothiocyanate; PE, phycoerythrin; CD, cluster of differentiation; neg.con, negative control. Figure 2. Time‑dependent expression of CD83, proliferation and the production of IL‑2 and IFN‑γ in CD4+ T cells. Following purity identification, the purified CD4+ T cells were either (A) directly stained with PE‑labeled CD83 or (B‑D) stimulated with anti‑CD3/CD28 for 1‑3 days, followed by staining with PE‑labeled CD83. All samples were analyzed by flow cytometry and the typical flow cytometric histograms indicating the percentages of CD83‑positive cells are shown. (E) Additionally, cell proliferation at days 1, 2 and 3 were determined using Cell Counting Kit 8. (F) The levels of IL‑2 and IFN‑γ in the supernatants from 1-, 2- and 3‑day cultures of activated CD4+ T cells were analyzed by ELISA. Values are expressed as the mean ± standard deviation of three independent experiments (*P<0.01; P<0.05). PE, phycoerythrin; CD, cluster of differentiation; IFN, interferon; IL, interleukin; neg.con, negative control. A B C D E F B F A B C A D C B C B A D E F D E F F D E Figure 2. Results The time‑dependent kinetics were consistent with those observed in non‑adherent lymphocytes. These data suggested that CD83 was signifi cantly upregulated; however, it was transiently presented on CD4+ T cells activated by the canonical CD3/CD28 signal. T cells (Fig. 3C). These results suggested that there was a significant correlation between the expression levels of CD83 and CD25 on anti‑CD3/CD28‑stimulated CD4+ T cells. TGF‑β restores the expression of CD83 on CD4+ T cells. Induced CD4+CD25+FoxP3+ T cells represent an important sub‑type of CD4+ T cells, termed iTreg cells, which are instru mental in the maintenance of immunological tolerance (18). The highly co‑localized expression of CD83 and CD25 on CD4+ T cells suggests that the expression of CD83 is closely associated with the differentiation of CD4+ T cells into iTreg cells upon activation. The present study aimed to determine whether the sudden decrease in the expression of CD83 on day 3 was due to the differentiation of CD4+ T cells, since the canonical CD3/CD28 signal rarely drives the generation of iTreg cells. With this suggestive evidence, purified CD4+ T cells were stimulated with anti‑CD3/CD28 in the presence or absence of TGF‑β, an immunosuppressive cytokine, which controls the balance between iTreg cells and pathogenic effector T cells (19). The expression of CD83 on CD4+ T cells was detected 3 days later using flow cytometry. In addition, the differentiation of CD4+ T cells triggered by TGF‑β signaling was identified by co‑expression analysis of FoxP3 and CD25. The experiments demonstrated that TGF‑β signaling increased the percentage of CD83‑postive cells in total CD4+ T cells to 12.1% on day 3, while stimulation without TGF‑β only resulted in 1.78% CD38-positive cells (Fig. 4A‑C). The differ ence was statistically significant according to Student's t‑test (P=0.0006; Fig. 4D). Furthermore, it was observed that, in contrast to anti‑CD3/CD28 stimulation, the combination with TGF‑β induced a significant increase in the co‑expression of FoxP3 and CD25 on CD4+ T cells on day 3 (Fig. 4E‑G). In conclusion, these findings demonstrated that the decreased surface expression of CD83 on activated CD4+ T cells can be restored by TGF‑β, which simultaneously drives the differen tiation of CD4+ T cells into iTreg cells. Decreased expression of CD83 is not caused by a reduced proliferation or activation of CD4+ T cells. Results Time‑dependent expression of CD83, proliferation and the production of IL‑2 and IFN‑γ in CD4+ T cells. Following purity identification, the purified CD4+ T cells were either (A) directly stained with PE‑labeled CD83 or (B‑D) stimulated with anti‑CD3/CD28 for 1‑3 days, followed by staining with PE‑labeled CD83. All samples were analyzed by flow cytometry and the typical flow cytometric histograms indicating the percentages of CD83‑positive cells are shown. (E) Additionally, cell proliferation at days 1, 2 and 3 were determined using Cell Counting Kit 8. (F) The levels of IL‑2 and IFN‑γ in the supernatants from 1-, 2- and 3‑day cultures of activated CD4+ T cells were analyzed by ELISA. Values are expressed as the mean ± standard deviation of three independent experiments (*P<0.01; P<0.05). PE, phycoerythrin; CD, cluster of differentiation; IFN, interferon; IL, interleukin; neg.con, negative control. 3312 CHEN et al: CONTINUOUS EXPRESSION OF CD83 ON ACTIVATED HUMAN CD4+ T CELLS Figure 3. Surface co‑localization of CD83 and CD25 on activated CD4+ T cells. Purified CD4+ T cells were stimulated with anti‑CD3/CD28 for 2 days, depos ited on slides, stained with (A) PE‑labeled anti‑CD83 and (B) FITC‑labeled anti‑CD25 and visualized by confocal microscopy. (C) The images of PE‑labeled anti‑CD83 and FITC‑labeled anti‑CD25 were merged to reveal an overlap of the two proteins. Images were captured from a x63 objective of a Zeiss LSM 410 microscope. Images are representative of three separate experiments (Scale bar, 5 µM). FITC, fluorescein isothiocyanate; PE, phycoerythrin; CD, cluster of differentiation. A B C A A C B Figure 3. Surface co‑localization of CD83 and CD25 on activated CD4+ T cells. Purified CD4+ T cells were stimulated with anti‑CD3/CD28 for 2 days, depos ited on slides, stained with (A) PE‑labeled anti‑CD83 and (B) FITC‑labeled anti‑CD25 and visualized by confocal microscopy. (C) The images of PE‑labeled anti‑CD83 and FITC‑labeled anti‑CD25 were merged to reveal an overlap of the two proteins. Images were captured from a x63 objective of a Zeiss LSM 410 microscope. Images are representative of three separate experiments (Scale bar, 5 µM). FITC, fluorescein isothiocyanate; PE, phycoerythrin; CD, cluster of differentiation. Purified CD4+ T cells were used to confirm the time‑dependent surface expression of CD83. As shown in Fig. 2A‑D, anti‑CD3/CD28 stimulation of CD4+ T cells led to an upregulation of CD83 on days 1 and 2 and substantially low levels of CD83 on CD4+ T cells on day 3. Results The present study next determined whether the decreased expression of CD83 on day 3 was a result of reduced proliferation or activation of target CD4+ T cells. Purified CD4+ T cells were stimulated with anti‑CD3/CD28 for 1, 2 and 3 days, and cell proliferation as well as the levels of IL‑2 and IFN‑γ in culture supernatants were analyzed. This experiment paralleled the detection of the expression of CD83 on CD4+ T cells as mentioned above. By contrast to fluctuated surface expression of CD83, the prolif eration of CD4+ T cells stimulated by anti‑CD3/CD28 was continuously increased from days 1 to 3 (Fig. 2E). Similarly, progressively increased secretion of IL‑2 and IFN‑γ was observed in CD4+ T cells (Fig. 2F). Therefore, the notable decrease in the surface expression of CD83 at day 3 was not a result of reduced proliferation or the activation of CD4+ T cells. Co‑localization of CD83 and CD25 on activated CD4+ T cells. Previous studies have reported that the mRNA expression of CD83 was predominantly in the naturally occur ring CD4+CD25+ T (nTreg) cells, which rapidly expressed large quantities of surface CD83 upon activation (16). It was therefore of interest to determine whether CD83 is co‑localized with CD25 on activated CD4+ T cells. Purified CD4+ T cells were stimulated with anti‑CD3/CD28 for 2 days, double‑labeled with CD25‑FITC/CD83‑PE and observed by immunofluorescence microscopy. The results demonstrated that CD83 and CD25 were accumulated on the surface of activated CD4+ T cells (Fig. 3A and B). Furthermore, merging of the two images revealed a high degree of co‑localization between CD83 and CD25 on the surface of the activated CD4+ Discussion A growing body of in vitro and in vivo evidence has demonstrated an immunosuppressive role of sCD83 in T cell‑mediated immunity (9‑14). However, the effect of MOLECULAR MEDICINE REPORTS 12: 3309-3314, 2015 MOLECULAR MEDICINE REPORTS 12: 3309-3314, 2015 3313 Figure 4. TGF‑β restored the expression of CD83 in CD4+ T cells. Purified human CD4+ T cells were (A) left unstimulated or (B and C) stimulated with anti‑CD3/CD28 in the presence or absence of TGF‑β. Following 3 days of incubation, all samples were collected, stained with PE‑labeled anti‑CD83 and ana lyzed by flow cytometry. The typical flow cytometric histograms indicating the percentages of CD83‑positive cells are shown in A, B and C. (D) Quantification of the number of CD83‑positive cells in the total CD4+ T cell population. Values are expressed as the mean ± standard deviation of four independent experi ments and P‑values were calculated by Student's t‑test. Additionally, the (E) unstimulated, (F) anti‑CD3/CD28‑ and (G) anti‑CD3/CD28/TGF‑β‑treated CD4+ T cells were stained with FITC‑labeled anti‑CD25, followed by intracellular FoxP3‑PE staining and analyzed by flow cytometry. The data demonstrated representative data for the donors in A‑C and E‑G. FITC, fluorescein isothiocyanate; PE, phycoerythrin; CD, cluster of differentiation, TGF, transforming growth factor; Fox, forkhead box. A B C D E F G A B C D B E F G E F F G E Figure 4. TGF‑β restored the expression of CD83 in CD4+ T cells. Purified human CD4+ T cells were (A) left unstimulated or (B and C) stimulated with anti‑CD3/CD28 in the presence or absence of TGF‑β. Following 3 days of incubation, all samples were collected, stained with PE‑labeled anti‑CD83 and ana lyzed by flow cytometry. The typical flow cytometric histograms indicating the percentages of CD83‑positive cells are shown in A, B and C. (D) Quantification of the number of CD83‑positive cells in the total CD4+ T cell population. Values are expressed as the mean ± standard deviation of four independent experi ments and P‑values were calculated by Student's t‑test. Additionally, the (E) unstimulated, (F) anti‑CD3/CD28‑ and (G) anti‑CD3/CD28/TGF‑β‑treated CD4+ T cells were stained with FITC‑labeled anti‑CD25, followed by intracellular FoxP3‑PE staining and analyzed by flow cytometry. The data demonstrated representative data for the donors in A‑C and E‑G. FITC, fluorescein isothiocyanate; PE, phycoerythrin; CD, cluster of differentiation, TGF, transforming growth factor; Fox, forkhead box. Discussion Upon T‑cell receptor (TCR)‑mediated cell activation, naïve CD4+ T cells differentiate into at least four major lineages (Th1, Th2, Th17 and iTreg cells), which can be distinguished by their specialized expression profiles, unique cytokine production and effector functions (22). The present study suggested that differentiation of CD4+ T cells, which follows cell activation, may be responsible for the downregulation of CD83 on CD4+ T cells stimulated by the canonical CD3/CD28 signal. By using a CD83 reporter mouse expressing enhanced green fluorescent protein under the control of the CD83 promoter, a previous study has demonstrated that CD83 was predominantly identi fied in CD4+CD25+ and in CD4 memory cells (21). Similarly, an in vitro study revealed that the mRNA expression levels of CD83 were differentially expressed in nTreg cells, which rapidly expressed large quantities of surface CD83 upon activation (16). In this context, the present study examined the spatial positions of CD83 and CD25 on activated CD4+ T cells by confocal microscopy. Of note, co‑localization of CD83 and CD25 on the surface of activated CD4+ T cells was observed. CD25 is the α‑chain of the IL‑2 receptor, and an IL‑2 signal is essential for the differentiation, expansion and function of Tregs (23,24). The function of highly co‑localized presenta tion of CD83 and CD25 remains to be elucidated; however, it may be associated with the stabilization of surface CD25 on target cells by CD83. A previous study has demonstrated that the transmembrane domain of CD83 inhibited the activity mCD83 expressed on antigen‑presenting cells (APCs), including dendritic cells and B lymphocytes, remains a matter of controversy (4,6‑8). Similarly, CD83 expressed on the surface of CD4+ T cells poses a novel challenge to elucidate the biological and functional behavior of mCD83. The present study demonstrated that CD83 was expressed in a time‑depen dent manner on activated human CD4+ T cells, which reached the maximum at day 2 and decreased significantly on day 3. These time‑dependent kinetics of the expression of CD83 are consistent with observations using CD4+ T cells isolated from BALB/c mice and stimulated with anti‑CD3 and IL‑2 in the presence of irradiated CD4‑depleted splenocytes as APCs (16). The decreased expression of CD83 at day 3 was not a result of reduced proliferation or activation of CD4+ T cells, in that IL‑2 and IFN‑γ production was sustained until day 3. CHEN et al: CONTINUOUS EXPRESSION OF CD83 ON ACTIVATED HUMAN CD4+ T CELLS 3314 4. Fujimoto Y, Tu L, Miller AS, Bock C, Fujimoto M, Doyle C, Steeber DA and Tedder TF: CD83 expression inﬂuences CD4+ T cell development in the thymus. Cell 108: 755‑767, 2002. of membrane-associated ring finger (C3HC4) 1, E3 ubiquitin protein ligase (MARCH1), a member of the MARCH family of membrane‑bound E3 ubiquitin ligases, which ubiquitinate and downregulate cell surface molecules (5). 5. Tze LE, Horikawa K, Domaschenz H, et al: CD83 increases MHC II and CD86 on dendritic cells by opposing IL‑10‑driven MARCH1‑mediated ubiquitination and degradation. J Exp Med 208: 149‑165, 2011. Numerous studies have demonstrated that the presence of TGF‑β at the onset of cell cultures can drive naïve CD4+ T cells to differentiate into iTreg cells (19,25). The present study therefore investigated the effects of TGF‑β regulation on the expression of CD83 on CD4+ T cells. The findings support the evidence that the addition of TGF‑β can restore the surface expression of CD83 on purified CD4+ T cells stimulated with anti‑CD3/CD28 for 3 days. In addition, the combined TGF‑β stimulation drove CD4+ T cells towards a phenotype of CD4+CD25+FoxP3+ iTreg cells. As mentioned above, the close association between CD83 and iTreg cells was also demonstrated to be present in nTreg cells, which develop in the thymus and constitutively express CD25 (16,21). These obser vations may have important implications for CD83 regulation on Treg cells. Therefore, it is conceivable that the sustained presentation of CD83 on CD4+ Treg sub‑sets is required for Treg induction, differentiation, survival or functional mainte nance. Defining the factors which regulate the expression of CD83 and understanding the mechanisms of CD83 regulation on regulatory T cells may facilitate the development of Treg cells and novel therapeutic strategies in immune intervention. 6. Prechtel AT, Turza NM, Theodoridis AA and Steinkasserer A: CD83 knockdown in monocyte‑derived dendritic cells by small interfering RNA leads to a diminished T cell stimulation. J Immunol 178: 5454‑5464, 2007. 7. Kretschmer B, Lüthje K, Ehrlich S, Osterloh A, Piedavent M, Fleischer B and Breloer M: CD83 on murine APC does not function as a costimulatory receptor for T cells. Immunol Lett 120: 87‑95, 2008. 8. Pinho MP, Migliori IK, Flatow EA and Barbuto JA: Dendritic cell membrane CD83 enhances immune responses by boosting intracellular calcium release in T lymphocytes. J Leukoc Biol 95: 755‑762, 2014. , 9. CHEN et al: CONTINUOUS EXPRESSION OF CD83 ON ACTIVATED HUMAN CD4+ T CELLS Dudziak D, Nimmerjahn F, Bornkamm GW and Laux G: Alternative splicing generates putative soluble CD83 proteins that inhibit T cell proliferation. J Immunol 174: 6672‑6676, 2005. 10. Bock F, Rössner S, Onderka J, et al: Topical application of soluble CD83 induces IDO‑mediated immune modulation, increases Foxp3+ T cells and prolongs allogeneic corneal graft survival. J Immunol 191: 1965‑1975, 2013. , 11. Starke C, Steinkasserer A, Voll RE and Zinser E: Soluble human CD83 ameliorates lupus in NZB/W F1 mice. Immunobiology 218: 1411‑1415, 2013. gy 12. Yuan Y, Wan L, Chen Y, et al: Production and characterization of human soluble CD83 fused with the fragment crystallizable region of human IgG1 in Pichia pastoris. Appl Microbiol Biotechnol 97: 9409‑9417, 2013. In conclusion, the results of the present study suggested that CD83 is presented in a time‑dependent manner on CD4+ T cells stimulated by the canonical CD3/CD28 signal. The maximum expression of CD83 expression was observed at day 2 and was followed by a marked decrease at day 3. Of note, the addition of TGF‑β at the onset of stimulation, which drove the differentiation of CD4+CD25+FoxP3+ Tregs, restored the expression of CD83 on day 3. Furthermore, CD83 was highly co‑localized with CD25, as observed by fluorescence microscopy. Therefore, the present study suggested that the continuous expression of CD83 on activated human CD4+ T cells was correlated with their differentiation into iTreg cells. Establishing the functional connection between the expression of CD83 and iTregs may facilitate the further elucidation of iTregs and their clinical application. , 13. Eckhardt J, Kreiser S, Döbbeler M, et al: Soluble CD83 ameliorates experimental colitis in mice. Mucosal Immunol 7: 1006‑1018, 2014. , 14. Guo Y, Li R, Song X, et al: The Expression and characterization of functionally active soluble CD83 by pichia Pastoris using high‑density fermentation. PLoS One 9: e89264, 2014. g y 15. Chen L, Zhu Y, Zhang G, Gao C, Zhong W and Zhang X: CD83‑stimulated monocytes suppress T‑cell immune responses through production of prostaglandin E2. Proc Natl Acad Sci USA 108: 18778‑18783, 2011. 16. Reinwald S, Wiethe C, Westendorf AM, Breloer M, Probst‑Kepper M, Fleischer B, Steinkasserer A, Buer J and Hansen W: CD83 expression in CD4+ T cells modulates inﬂammation and autoimmunity. J Immunol 180: 5890‑5897, 2008. 17. Su LL, Iwai H, Lin JT and Fathman CG: The transmembrane E3 ligase GRAIL ubiquitinates and degrades CD83 on CD4+ T cells. CHEN et al: CONTINUOUS EXPRESSION OF CD83 ON ACTIVATED HUMAN CD4+ T CELLS J Immunol 183: 438‑444, 2009. 18. Kumar S, Naqvi RA, Ali R, Rani R, Khanna N and Rao DN: CD4+CD25+ T regs with acetylated FoxP3 are associated with immune suppression in human leprosy. Mol Immunol 56: 513‑520, 2013. Acknowledgements 19. Chen W, Jin W, Hardegen N, Lei KJ, Li L, Marinos N, McGrady G and Wahl SM: Conversion of peripheral CD4+CD25‑ naive T cells to CD4+CD25+ regulatory T cells by TGF‑beta induction of transcription factor Foxp3. J Exp Med 198: 1875‑1886, 2003. This study was supported by the National Natural Science Foundation of China (no. 81171662), the Natural Science Foundation of Anhui Province (no. 1408085MH169) and the Scientific Research Foundation for the Doctoral Program of the Second Hospital of Anhui Medical University (no. 2012BKJ011). p p p 20. Kretschmer B, Kühl S, Fleischer B and Breloer M: Activated T cells induce rapid CD83 expression on B cells by engagement of CD40. Immunol Lett 136: 221‑227, 2011. 21. Lechmann M, Shuman N, Wakeham A and Mak TW: The CD83 reporter mouse elucidates the activity of the CD83 promoter in B, T and dendritic cell populations in vivo. Proc Natl Acad Sci USA 105: 11887‑11892, 2008. , 22. Yamane H and Paul WE: Early signaling events that underlie fate decisions of naive CD4(+) T cells toward distinct T‑helper cell subsets. Immunol Rev 252: 12‑23, 2013. Discussion Therefore, these findings demonstrated the fine‑tuning of activation‑induced expression of CD83 on CD4+ T cells. Of note, non‑CD4 cells in non‑adherent lymphocytes expressed considerable quantities of surface CD83. It was suggested that this synchronous presentation of CD83 originates from CD8+ T cells and remaining B cells upon activation (20,21). On the other hand, unstimulated CD4+ T cells also expressed certain quantities of surface CD83, which may be explained by the spontaneous activation of undepleted monocytes in non‑adherent lymphocytes, which in turn stimulate and acti vate CD4+ T cells. CHEN et al: CONTINUOUS EXPRESSION OF CD83 ON ACTIVATED HUMAN CD4+ T CELLS , 3. Stein MF, Lang S, Winkler TH, et al: Multiple interferon regulatory factor and NF‑κB sites cooperate in mediating cell‑type‑ and maturation‑specific activation of the human CD83 promoter in dendritic cells. Mol Cell Biol 33: 1331‑1344, 2013. 1. Zhou LJ, Schwarting R, Smith HM and Tedder TF: A novel cell‑surface molecule expressed by human interdigitating reticulum cells, Langerhans cells and activated lymphocytes is a new member of the Ig superfamily. J Immunol 149: 735‑742, 1992. g p y , 2. Zhou LJ and Tedder TF: CD14+ blood monocytes can differ entiate into functionally mature CD83+ dendritic cells. Proc Natl Acad Sci USA 93: 2588‑2592, 1996. References 1. Zhou LJ, Schwarting R, Smith HM and Tedder TF: A novel cell‑surface molecule expressed by human interdigitating reticulum cells, Langerhans cells and activated lymphocytes is a new member of the Ig superfamily. J Immunol 149: 735‑742, 1992. 1. Zhou LJ, Schwarting R, Smith HM and Tedder TF: A novel cell‑surface molecule expressed by human interdigitating reticulum cells, Langerhans cells and activated lymphocytes is a new member of the Ig superfamily. J Immunol 149: 735‑742, 1992. , 23. Yu A and Malek TR: Selective availability of IL‑2 is a major determinant controlling the production of CD4+CD25+Foxp3+ T regulatory cells. J Immunol 177: 5115‑5121, 2006. g y 24. Barron L, Dooms H, Hoyer KK, Kuswanto W, Hofmann J, O'Gorman WE and Abbas AK: Cutting edge: mechanisms of IL‑2‑dependent maintenance of functional regulatory T cells. J Immunol 185: 6426‑6430, 2010. g p y 2. Zhou LJ and Tedder TF: CD14+ blood monocytes can differ entiate into functionally mature CD83+ dendritic cells. Proc Natl Acad Sci USA 93: 2588‑2592, 1996. , 3. Stein MF, Lang S, Winkler TH, et al: Multiple interferon regulatory factor and NF‑κB sites cooperate in mediating cell‑type‑ and maturation‑specific activation of the human CD83 promoter in dendritic cells. Mol Cell Biol 33: 1331‑1344, 2013. , 25. Rao PE, Petrone AL and Ponath PD: Differentiation and expansion of T cells with regulatory function from human peripheral lymphocytes by stimulation in the presence of TGF‑β. J Immunol 174: 1446‑1455, 2005.
https://openalex.org/W3090943904	https://ojs.sites.ufsc.br/index.php/mixsustentavel/article/download/4280/3220	Portuguese	null	RESUMO DE DISSERTAÇÃO: IDENTIFICAÇÃO DE FATORES QUE CONTRIBUEM PARA O USO DA BICICLETA COMO TRANSPORTE URBANO	Mix Sustentável	2,020	cc-by	925	1. RESUMO mobilidade nas cidades, entretanto, os espaços dedica- dos a ela ainda são poucos e os ciclistas se vêem numa disputa com veículos muito maiores e mais pesados. Andar de bicicleta é expor-se ao mundo. Deixar as pernas embalarem o movimento, fechar os olhos e sentir o vento no rosto. O corpo parece flutuar no espaço... A paisagem corre rápida ao redor. Na pele sente-se atravessar os raios de sol, filtrados pelas árvores. Folhas caídas estalam sob o peso das rodas. Sensações se misturam: cheiros, sons, luzes, cores. Uma pedalada e muitos metros se foram. Muitas são as questões a serem respondidas sobre os ciclistas e suas atitudes. O conhecimento sobre seu com- portamento e preferências gera poder para planejadores e governos criarem ações eficazes no que concerne o in- centivo do uso da bicicleta como meio de transporte. Concentração e força nas subidas são recompensadas com descidas emocionantes e uma vista monumental. Um prazer lúdico e quase infantil toma conta... Os diferentes fatores que motivam o uso da bicicleta vêm sendo amplamente investigados no meio acadêmico, no entanto, a análise da percepção dos ciclistas e sua re- lação com os espaços ainda é um tema pouco explorado. Por que andar de bicicleta? Optou-se por realizar o estudo na cidade de Florianópolis, devido ao crescente interesse por este meio de transporte na cidade, déficit de pesquisas sobre percepção de ciclistas no município e melhor acesso do pesquisador aos dados. Desde a crise do petróleo de 1979 se discute o uso da bicicleta como meio de transporte. As questões eco- nômicas e ambientais motivaram o início dos debates. Recentemente, a saturação das vias das cidades por auto- móveis reforçou a importância da bicicleta, agora sugeri- da como alternativa de mobilidade. Uma boa mobilidade pressupõe acesso amplo e democrático da população ao espaço urbano, contribuindo para sua inclusão social. 163 Buscando identificar os fatores que contribuem para o uso da bicicleta como meio de transporte, foi realiza- do um experimento com 20 ciclistas, que desenharam os mapas de seus trajetos, descreveram as características encontradas em cada trecho e responderam a uma entre- vista com 52 perguntas. Quando vivemos em uma cidade, buscamos usufruir de suas opções realizando trocas culturais, econômicas e so- ciais. Para efetiválas, precisamos entrar em contato com ou- tras pessoas e outros lugares e fazemos isso através da mo- vimentação no espaço e da comunicação remota. DISSERTAÇÕES DISSERTAÇÕES Mix Sustentável \| Florianópolis \| v.6 \| n.4 \| p.163-164 \| ago. \| 2020 1. RESUMO Certas trocas não podem ser desassociadas do espaço físico das cidades e certos grupos só se mantêm ativos através de encontros pessoais e frequentes de seus membros. Estas relações pessoais são dependentes diretas da mobilidade. Através da análise do conteúdo das respostas, os dados obtidos foram categorizados, quantificados, reagrupados e representados em gráficos e mapas de maneira a melhor proceder a análise. Os fatores objetivos incluem infraestrutura na origem, destino, integração modal e trajetos; e características do terreno e condições ambientais. Em relação a infraestrutu- ra, contribuem para o maior uso da bicicleta a existência de bicicletários bem localizados, abundantes, seguros e que acomodem os variados tipos de bicicleta; a integração modal para transposição de barreiras físicas, como morro e baía; a existência de rotas ciclísticas exclusivas; a continui- dade das rotas exclusivas ou compartilhadas; sinalização que reforce a prioridade dos ciclistas; adequação da largura Frente a este panorama, o transporte por bicicleta pa- rece solucionar muitas questões. Em comparação com os automóveis, a bicicleta requer menos espaço urbano e menores investimentos em infraestrutura. Ela é mais aces- sível economicamente, menos poluente e mais democrá- tica, pois pode ser utilizada por pessoas de praticamente todas as idades e condições sociais. As características da bicicleta parecem ser extremamente vantajosas para a Mix Sustentável \| Florianópolis \| v.6 \| n.4 \| p.163-164 \| ago. \| 2020 Mix Sustentável \| Florianópolis \| v.6 \| n.4 \| p.163-164 \| ago. \| 2020 Identificação de fatores que contribuem para o uso da bicicleta como transporte urbano J. M. de Oliveira & A. D. Neto das pistas para acomodar a oscilação da bicicleta, principal- mente nas subidas; transições bem projetadas e sinalizadas entre ciclovia e vias; eliminação das faixas de tráfego que alargam e afinam; remoção de placas, obstáculos, valas e tachas mal posicionadas e melhora na qualidade e limpeza dos pisos. Em relação às condições ambientais, contribuem para o uso da bicicleta as rotas protegidas do vento; a boa drenagem das vias e ciclovias; a disponibilidade de integra- ção modal para os dias de chuva; vias arborizadas e som- breadas e baixos níveis de poluição do ar. Os fatores subjetivos incluem a percepção da seguran- ça; o uso desde a infância; o preparo físico gradual; a apre- ciação do prazer de pedalar; a percepção de vantagem em relação a gastos financeiros, tempo e autonomia; e a acei- tação da sociedade. Mix Sustentável \| Florianópolis \| v.6 \| n.4 \| p.163-164 \| ago. \| 2020 1. RESUMO Para fins práticos, os resultados desta pesquisa podem ser utilizados para o aprimoramento das rotas ciclísticas atuais e projeto de novas rotas; adequação da infraestrutu- ra de apoio à origem e destino, priorização na implantação de integração modal e direcionamento de programas de educação e incentivo ao uso da bicicleta. Para os ciclistas, espera-se que a publicação desta pes- quisa possa expor suas necessidades e os fatores que pre- cisam ser levados em conta em projetos de incentivo a este meio de transporte. 164 Mix Sustentável \| Florianópolis \| v.6 \| n.4 \| p.163-164 \| ago. \| 2020
https://openalex.org/W4387580858	https://link.springer.com/content/pdf/10.1007/s12186-023-09337-8.pdf	English	null	Practical nurse students’ misconceptions about infection prevention and control	Vocations and learning	2,023	cc-by	9,221	Vocations and Learning (2024) 17:143–164 https://doi.org/10.1007/s12186-023-09337-8 Vocations and Learning (2024) 17:143–164 https://doi.org/10.1007/s12186-023-09337-8 RESEARCH RESEARCH Extended author information available on the last page of the article Practical nurse students’ misconceptions about infection prevention and control Received: 25 November 2022 / Accepted: 18 September 2023 / Published online: 12 October 2023 © The Author(s) 2023 Keywords Nurse · Student · Infection prevention and control · Misconception · Mental model · Conceptual change Background and aim Although the healthcare system is meant to prevent and treat diseases, and to care for patients, it repeatedly fails to achieve this goal. Treatments can have side-effects that need to be treated in turn, and procedures can be performed carelessly. An exam- ple of the latter is healthcare associated infections (HAI) (Cieslak et al., 2009). The World Health Organization (WHO, 2016) defines HAI as an infection that occurs in a patient during the process of care, which was either not present or incubat- ing at the time of admission. HAI is a serious threat to public health. For instance, about 8.9 million episodes of HAI are estimated to occur in EU countries yearly; 1% of the cases are lethal (Suetens et al., 2018). Of these infections 50% could have been prevented by consistent and careful application of professional hand hygiene (WHO, 2016). Nevertheless, it was found that a majority of the healthcare work- ers (HCWs) do not adhere to these procedures. Improvements of the HAI situation must come from healthcare practice and education, but these are difficult to accom- plish. The recent experiences during the Covid-19 pandemic made the benefits of hygiene practices very clear. The public campaigns that emphasised wearing surgi- cal masks and washing and disinfecting hands led to a reduction in incidence of res- piratory and gastrointestinal virus infections. However, despite the increased use of hand disinfectants HAIs did not decrease (Dapper et al., 2022). Dapper et al. (2022) hypothesised that increased use of hand disinfectants helped protect the staff but did not influence the protection of the patients. Infection prevention and control (IPC) and hand hygiene are mainly taught as skills in nurse training programmes (Kısacık et al., 2021; Korhonen et al., 2019). Such training programmes mostly focus on good practice, whilst challenging the pre-existing beliefs of students receives less atten- tion; however, these beliefs warrant more consideration. In this article, we focus on these beliefs in order to discover how far they correspond with professional IPC knowledge. Abstract When teaching infection prevention and control (IPC), nursing education tends to focus on skills and fostering good practice rather than challenging students’ think- ing. Therefore, students’ misconceptions about IPC receive less attention than they deserve. The purpose of the study was to make an inventory of student nurses’ mis- conceptions about IPC before instruction and to make these misconceptions visible to teachers. The study was conducted in one vocational institute in Finland and is based on the answers of 29 practical nurse students before IPC training. The stu- dents took an online test requiring them to justify their answers to two multiple- true–false questions: 1) What is the main route of transmission between patients in healthcare facilities, and 2) What is the most effective and easiest manner to prevent the spreading of pathogens, e.g., multi-resistant bacteria in long-term care facilities? Analysis of the students’ written justifications resulted in three mental models: 1) the Household Hygiene Model manifesting lay knowledge learned in domestic situ- ations, 2) the Mixed Model consisting of lay knowledge, enriched with some profes- sional knowledge of IPC, and 3) the Transmission Model manifesting a professional understanding of IPC. The first two mental models were considered to be miscon- ceptions. Only one of the participants showed a professional understanding (i.e., the Transmission Model). To conclude, student nurses manifested systematic patterns of misconceptions before instruction. Unless the students are confronted with their misconceptions of IPC during instruction, it is likely that these misconceptions will impede their learning or make learning outcomes transient. Keywords Nurse · Student · Infection prevention and control · Misconception · Mental model · Conceptual change Keywords Nurse · Student · Infection prevention and control · Misconception · Mental model · Conceptual change 0123456789) 1 3 012341 456789) 3 R. Eronen et al. 144 Conceptual change and infection prevention and control When student nurses start their studies, they have a basic knowledge of how to take care of their own personal and domestic hygiene as this is a general learning requirement in life (Boshuizen & Marambe, 2020; Moll et al., 1992; Vosniadou et al., 2001). A wealth of knowledge is already accumulated during early child- hood by observing parents at home, e.g., from the revulsion parents show when confronted with unpleasant smelling food or bodily secretions (Curtis & Biran, 2001), and by learning from caretakers in nurseries or from teachers in schools (Boshuizen & Marambe, 2020). This familiarity with domestic hygiene situations results in a form of lay knowledge such as having a “gut feeling” for things, sub- stances and odours which are “disgusting” (Curtis & Biran, 2001), and that wash- ing one’s hands prevents the spreading of “bugs” and sicknesses. Some of this 1 3 Practical nurse students’ misconceptions about infection… 145 knowledge causes physical reactions such as feeling the urge to withdraw when touching slimy materials or knowing where to cough and how to clean one’s hands. knowledge causes physical reactions such as feeling the urge to withdraw when touching slimy materials or knowing where to cough and how to clean one’s hands. In the course of vocational education, student nurses need to construct profes- sional knowledge based on official guidelines of IPC. IPC is a scientific approach and a set of practical procedures which prevent patients from being harmed by avoid- able HAIs (WHO, 2016). HAIs are caused by lapses in IPC (Cieslak et al., 2009). Vocational education does seem to influence student nurses’ knowledge of IPC, but during clinical training discrepancies between theory and practice may appear (Boshuizen et al., 2020; Cox et al., 2014; Korhonen et al., 2019). There is evidence that student nurses’ knowledge and the ability to apply IPC are not sufficient (AL- Rawajfah & Tubaishat, 2015; Kelcíkova et al., 2012). The nursing curricula should prepare students to understand and apply IPC knowledge in a professional way in a range of clinical situations (Cox et al., 2014). It is possible that existing preconcep- tions may interfere with the construction of professional knowledge. There is a strong line of earlier research on the misconceptions, preconceptions, alternative conceptions, naïve theories or mental models of learners which educators see as barriers to knowledge restructuring (Guzzetti et al., 1993). Conceptual change and infection prevention and control This lay knowl- edge has proven to be difficult to modify (Vosniadou & Brewer, 1992). In the con- text of IPC, nurses may have a preconception emphasising the importance of self- protection (Dapper et al., 2022; Jansson et al., 2016; Jeong & Kim, 2016). Unknown patients may be considered “dirty” (Curtis & Biran, 2001), and nurses need to pro- tect themselves from the patients’ “bugs” with, e.g., gloves. In other domains, stud- ies about the persistence of faulty mental models, e.g., of the cardiovascular sys- tem have demonstrated that not all medical students reach the necessary scientific level of understanding about the circulatory system even after instruction (Ahopelto et al., 2011; Mikkilä-Erdmann et al., 2012; Södervik et al., 2019). According to Chi (2013), a mental model is an organised collection of an individual’s beliefs. It can be an internal representation of a concept, or an interrelated system of concepts. Individuals use mental models to solve problems or answer questions (Vosniadou & Brewer, 1992). It has been argued that students’ prior misconceptions are often hard to trans- form into scientific knowledge because students attempt to reconstruct professional knowledge based on their misconceptions (Boshuizen & Marambe, 2020; Meren- luoto & Lehtinen, 2004). This process can result in the enrichment of existing mis- conceptions—instead of a transformation—because some but not all false beliefs are altered based on the instruction (Vosniadou et al., 2001). This enrichment occurs at an easier level of learning (Merenluoto & Lehtinen, 2004), whereas Conceptual Change requires a more profound revision of students’ mental models (Vosniadou et al., 2001). Before this occurs, students might have mixed, or fragmented concep- tions about the topic. When students are taught to be mindful of their existing misconceptions instead of being merely taught scientific knowledge, the result may lead to belief revi- sion (Chi, 2013; Vosniadou et al., 2001). This revision of existing misconception may result in the abandonment of previous misconceptions and the acquisition of enriched knowledge, which according to Södervik et al. (2019) occurs gradually. 3 3 146 R. Eronen et al. An alternative line of inquiry stresses that students do not need to abandon their lay conceptualisations (Linn, 2008; Lundholm, 2018; Solomon, 1983). Instead, naïve models based on everyday knowledge and scientific models learned in educa- tional settings can coexist in the minds of learners. Conceptual change and infection prevention and control The scientific models are learnt in a meaning-making process involving a perpetual process concerning the interpre- tation of isolated pieces of knowledge and reflections regarding a coherent whole (Halldén, 1993; Halldén et al., 2009). Importantly, in this view on the conceptual change process, the role of education is to offer more powerful ideas and discussion of context (private context of one’s own home versus professional context) so that students can build their meta-knowledge by learning to identify differences between these two ways of thinking (Lundholm, 2022) (See Fig. 1).f In education, students are expected to see the difference between conceptions generated by everyday experiences and perceptions, and the concepts and evidence created by scientific communities. While concepts refer to classifications and the set of knowledge the person associates with the concept’s name, conceptions are sys- tems of explanation (White, 1994) (see Fig. 1). Lay thinking and scientific thinking differ in many important ways. First, lay thinking is based on personal values, which are likely to be influenced by cultural norms and stem from personal preferences and cultural influences. For example, cultural norms about hygiene and cleanliness dictate what is edible and what is not, when one washes oneself, or what can be touched with bare hands. Instead, professional thinking needs to be built on profes- sional values which include that absolute priority is given to the patient’s safety; IPC procedures are viewed as a means of ensuring patient safety. Second, in lay thinking, causality tends to be understood as a simple linear relation (e.g., use of gloves blocks pathogen spreading) without taking into consideration an emerging process with multiple feedback loops as in the scientific model. Third, lay thinking Fig. 1 Common sense context versus professional/scientific context based on Lundholm (2018, 2022) Fig. 1 Common sense context versus professional/scientific context based on Lundholm (2018, 2022) Practical nurse students’ misconceptions about infection… 147 relies on personal experience, whereas scientific thinking relies on a review of the accumulated empirical evidence (see Fig. 1). Students also need to learn the context of applicability i.e., an awareness regarding which system of explanation and related IPC procedures to use under different circumstances (Halldén, 1999; Lundholm, 2018); students need to know that in household cleaning activities a pair of gloves can be put on without applying a hand rub in advance, while in healthcare facilities this is mandatory. Students’ values and identities are part of this process. Little by little they must learn to think like a nurse. Finally, learning to think like a professional is likely to be moderated by motiva- tional factors such as control beliefs and the influence of significant others (see Pin- trich et al., 1993). It is important to know what kind of lay conceptions the students bring into vocational education so that teachers are able to identify lay conceptions and assist the students in making comparisons between the two ways of thinking. There are essentially two types of values: the values of the discipline and the values and emotions that the students hold (see Fig. 1). Therefore, the purpose of the present study is to make an inventory of practi- cal nurse students’ misconceptions about IPC. The following research questions are addressed: 1) What kind of misconceptions of IPC are held by students? 2) How prevalent are these misconceptions? To make the misconceptions “visible”, the students were asked to answer some multiple-true–false questions and to justify their answers. Based on these written justifications, the research team created mental models to describe the typical ways students thought about IPC. Context of the study The study was conducted in the context of practical nurse education in Finland. Practical nurse education leads to a vocational, upper secondary degree requiring 120 ECTS. The studies take approximately three years to complete (Finnish National Agency for Education, 2018). Practical nurses are the largest group of healthcare workers in Finland (Virtanen, 2020). They typically work in basic public healthcare or private social care, including long-term care and day-care facilities. They also work in specialised public medical care, e.g., inpatient wards, hospitals, ambulances, or homecare depending on their specialisation (Finnish National Agency for Educa- tion, 2018). During their studies practical nurse students acquire a wide range of basic com- petences related to nursing and care including observing the vital functions and symptoms of patients. The students also obtain a medication administration licence allowing them to dispense and administer medications via natural routes, and by subcutaneous and intramuscular injections when instructed to do so. Regarding IPC, the curriculum states that “students follow the principles of aseptic working meth- ods” (Finnish National Agency for Education, 2018). 1 3 R. Eronen et al. 148 Participants Participants were recruited from a vocational school in Finland immediately prior to the outbreak of the COVID-19 pandemic. The first-year adult students had just begun their studies in practical nursing, and the first author was the teacher of the 30-h compulsory course on IPC. The study was conducted during the course but providing the material for the study was voluntary. Of a total of 90 students, 67 students (74%) (54 females, 81%, 13 males, 19%) provided their written consent to participate in the study. However, only 29 partic- ipants (24 females, 83%, 5 males, 17%) supplied complete answers (i.e., answers including written justifications) to the questions this study focused on, resulting in a sample size of 29. The participants’ age varied from 18 to 55. Eight participants were under 25 years of age, thirteen were 26–40, and eight were over 40. Materials and procedures As a part of their studies, the students took a test on IPC before the start of the course. This test, containing multiple-true–false questions together with prompts to justify one’s responses, was administered during the first lesson using an on-line tool (Webropol). It was possible to complete the survey at home for those students who were absent. The link to the test was sent via the school’s learning management sys- tem with the possibility to complete it via a computer or a mobile device. The time to complete the test ranged from 12 to 61 min. The present study is based on an examination of two questions from a more extensive questionnaire developed explicitly for the purpose of this study. The valid- ity and reliability of the entire questionnaire will be reported elsewhere. Students were asked to indicate whether each of the four answers provided for both questions were true or false (see Appendix 1). A point was given for every correct answer (a point for each true positive or true negative) with a maximum of eight points. The first question ‘What is the main route of transmission between patients in healthcare facilities?’ was inspired by the study conducted by Jeong and Kim (2016); the second question ‘What is the most effective and easiest manner to prevent the spreading of pathogens (e.g., multiresistant bacteria) in long-term care facilities?’ was derived from the study by Caniza et al. (2007). The questions were translated into Finnish and partly modified based on current, national, and interna- tional guidelines. The present study is based on an analysis of the written justifica- tions given for the two multiple-true–false questions. 1 3 Data coding and analysis The data analysis was based on a method initiated by Patel and Groen (1986) who investigated clinical reasoning in medicine. This method has been described very succinctly by Van de Wiel et al. (2000). To enable comparison at a conceptual level, students’ answers were rewritten as concept maps (Tversky, 2011). These concept maps consisted of nodes and links between nodes. The nodes represent the differ- ent concepts applied in a justification, and the links between nodes represent the relationship between the concepts. To enable comparison between maps some stand- ard abbreviations or elements were used: P for patient, HCW for healthcare worker, arrow for connecting (causal) link, X connected to arrow for breaking that link, att for attributes such as “has features of…, or instr for instrument such as “with their hands”. The reliability of the methodology and validity of the outcomes was monitored through a combination of independent coding and a discussion of any differences. f Analysis proceeded in chronological order as follows: The research team read through the justifications of questions 1 and 2. A subset of 1. The research team read through the justifications of questions 1 and 2. A subset of 19 protocols (i.e., cases) was selected and visually represented as concept maps. 1. The research team read through the justifications of questions 1 and 2. A subset o i 19 protocols (i.e., cases) was selected and visually represented as concept map 2. The fourth author created two extreme prototypical models and visualised them as concept maps (see Appendix 2). The Household Hygiene Model was created bottom-up, i.e., based on the students’ justifications. The term was inspired by Moll et al., (1992, p 133) who used the term ‘household funds of knowledge’ to refer to the ‘historically accumulated and culturally developed bodies of knowl- edge and skills essential for household […] functioning’. The Transmission Model was created using a top-down approach based on scientific knowledge of IPC (WHO, 2016).i 3. Then the first author attempted to categorise the cases into these two extremes. To solve discrepancies, the research team created a third model called the Mixed Model, which contained features of both of the two previous models. 4. These three models were translated into categorisation criteria (see Table 1).i 5. Using these criteria, the first author categorised the 19 cases into one of the three mental models. Ethical approval The study was granted ethical approval, (code 27/2019) by the Ethical review board of the Finnish university involved in the study. The principal of the voca- tional school granted the research permit. The participants were informed about the research project, and informed consent was obtained at the beginning of the first 1 3 1 3 Practical nurse students’ misconceptions about infection… 149 lesson. Permission via email was sought from those participants who did not attend the first lesson. Participation in the study was voluntary. Participants or non-partici- pants were not advantaged or disadvantaged in any way. lesson. Permission via email was sought from those participants who did not attend the first lesson. Participation in the study was voluntary. Participants or non-partici- pants were not advantaged or disadvantaged in any way. Data coding and analysis The responses to the multiple-true–false questions were ignored at this phase.i p 6. Authors 1, 2 and 3 discussed the decisions made by the first author in phase 5. This resulted in transferring some cases to another category. 7. The fourth author worked independently as a second coder. This resulted in disa- greement regarding four cases. 8. The team concluded that referring to “hand hygiene” instead of “hand disinfec- tion” was a sufficient inclusion criterium for the Mixed Model. This resulted in agreement regarding all cases. 8. The team concluded that referring to “hand hygiene” instead of “hand disinfec- tion” was a sufficient inclusion criterium for the Mixed Model. This resulted in agreement regarding all cases. 1 3 150 R. Eronen et al. 9. To check for coder agreement the remaining ten cases were categorised by the first and fourth author independently. This resulted in complete agreement without the necessity for further discussions. Results: the Three Mental Models Based on the analyses of the students’ justifications to questions 1 and 2, three mental models were formed. These were defined as 1) the Household Hygiene Model, 2) the Mixed Model, and 3) the Transmission Model (see Table 1). The three models found showed that there were two large subclasses of misconcep- tions: household conceptions and household conceptions enriched with scientific professional knowledge. The background knowledge of IPC and use of terminology in the present study requires some clarification. The main route of transmission is the HCWs’ direct contact with patients by means of undisinfected hands (WHO, 2009). Other routes of transmission include airborne transmission (via dust) and drop- let (respiratory) transmission. Hand hygiene is considered a general lay expression for cleaning one’s hands because there is no definition of how it should be applied. Hand disinfection means rubbing alcohol-based liquid to disinfect the healthcare workers’ hands (Pittet & Boyce, 2003). Hand disinfection (“hand rub”) is considered a pro- fessional expression when used in this context. Hand disinfection is the easi- est, cheapest, and most favourable way to prevent the spreading of pathogens from the HCWs’ hands within healthcare. Improving HCWs’ hand disinfection according to official guidelines reduces the spreading of pathogens between vul- nerable patients and therefore prevents HAIs (WHO, 2009). In contrast, washing one’s hands (hand wash) is an action whereby the hands are cleaned using soap and water when they are visibly soiled. Several authors (AL-Rawajfah & Tubaishat, 2015; Caniza et al., 2007; Jeong & Kim, 2016; Ward, 2013) consider handwashing sufficient. In this study, hand wash- ing was not considered to be a professional manner for healthcare workers to clean their hands at work. Finnish student nurses are taught according to the guidelines provided by the local authorities and the WHO, which state that the healthcare worker’s hands should be disinfected using an alcohol-based hand rub unless visibly soiled (Sairaalahygienia- ja infektiontorjuntayksikkö, 2023; WHO, 2009). Gloves are disposable, and non-sterile. They are used during nursing proce- dures where there is a risk of contact with any kind of secretion (Pittet & Boyce, 2003). Appropriate use of gloves is poorly understood among HCW (e.g., when, and how to put on and take off gloves correctly). Using gloves in situations when they are not indicated can lead to cross-contamination between patients, and a waste of resources (WHO, 2009). Household hygiene model Instances of the Household Hygiene Model represent the clearest misconceptions students have about IPC. In this model, the knowledge is based on lay knowledge about domestic hygiene learned from everyday situations. It is typical of this model that the HCW is not mentioned as the pathogen-spreading link between patients. Instead, the patients and their visitors, surfaces, and secretions are reported as the main route of transmission. These students speak about hand hygiene or hand wash- ing, but they do not mention hand disinfection. They use everyday language instead of professional language, with many of them emphasising the use of gloves in patient contacts. The mental model of Student 12 (see Fig. 2) is presented, because it is closest to the prototypical model of the Household Hygiene Model (see Appendix 2). The original justifications for Question 1 and 2 are presented on the left side of the fig- ure, and on the right side the justifications are visualised in the form of a concept map. Student 12 scored six out of eight points on the multiple-true–false test by answering most questions correctly but failing to indicate that the main transmission route is the healthcare worker’s hands. The student was of the opinion that surfaces are the main route of transmission. When the student’s justifications were analysed, it seemed that the student had no idea of IPC in healthcare and believed that “poor cleaning” and “poor air qual- ity” are the main routes of transmission, instead of the HCW’s hands. Perhaps by poor air quality, the student was referring to airborne transmission. By stating that “all relatives and healthcare workers prevent the spreading of pathogens by wash- ing their hands”, the student was referring to basic household knowledge. This is clearly incorrect: both relatives and HCWs are instructed to prevent the spreading of pathogens by disinfecting their hands by using hand rub in healthcare settings.i Student 64’s mental model was also classified as an instance of the House- hold Hygiene Model. The student scored two out of eight points on the multiple- true–false questions, answering only two items correctly, namely 1) antibiotics are not the most effective manner to prevent infections, and 2) hand disinfection is the ig. 2 First example of the Household Hygiene Model. Results: the Three Mental Models 1 3 Practical nurse students’ misconceptions about infection… 151 Table 1 The Three Mental Models and their criteria HOUSEHOLD HYGIENE MODEL (Lay knowledge) MIXED MODEL (Between lay and scientific knowl- edge) TRANSMISSION MODEL (Scientific knowledge) Main route of transmission - No mention of the nurse as the pathogen-spreading link between the patients - Emphasis on the use of disposable gloves in patient contact - Patients and visitors, surfaces, and secretions are reported as the main route of transmission Hand hygiene - No mention of hand washing, hand disinfection, hand hygiene OR - refers to “hand hygiene” or “hand wash” BUT -does not mention “hand disinfection” Language - Uses everyday language e.g., “bugs” Main route of transmission - The nurse is the pathogen-spreading link between patients (scientific) AND - Patients and visitors, surfaces, and secretions are reported as the main route of transmission (lay) Hand hygiene - Disinfection or hand rub or at least hand hygiene is mentioned Language - Uses everyday language e.g., “bugs” (lay) or some professional phrases (scientific) Main route of transmission - The nurse is the main pathogen-spreading link between patients Hand hygiene - Hand disinfection mentioned Language - Uses professional vocabulary e.g., pathogens and Standard Precautions 1 3 R. Eronen et al. 152 Fig. 2 First example of the Household Hygiene Model. (ID 12 = student pseudonym, Q1-2 = justi- fications, P = patient, X = preventing spreading of pathogens, HW = hand wash, HCW = healthcare worker, → = affects, causes) Household hygiene model (ID 12 = student pseudonym, Q1-2 = justi- fications, P = patient, X = preventing spreading of pathogens, HW = hand wash, HCW = healthcare worker, → = affects, causes) 3 Practical nurse students’ misconceptions about infection… 153 most effective manner to prevent the spreading of infections. The student specified that surfaces touched by patients who do not attend to hygiene sufficiently spread bacteria (see Fig. 3) which is correct, but it is not the main route of transmission; it is the HCWs’ hands that are the main route of transmission. The student also stated that “Infections can be prevented with vaccinations” which is also correct, but again not the correct answer to the question regarding the main route of transmission (WHO, 2009). It remains unclear what exactly the student meant by stating that “the right kind of protection can prevent the spreading of diseases”. Perhaps the student meant using personal protective equipment, meaning gloves, gowns, and masks. As in the case of Student 12, it is assumed that Student 64 had not internalised how critical the healthcare worker’s role is in IPC, because in the written accounts other people were held responsible for the spreading of pathogens.i The final example of the Household Hygiene Model (Student 63) is interesting, because the student (see Fig. 4) appears to have some working experience with the elderly. In the response, the student emphasises the use of gloves in IPC, instead of hand disinfection, which is typical of the household hygiene model. This student seems to focus on self-protection and resorts to intuitive-emotional decision-making (Aarkrog & Wahlgren, 2022) rather than protecting the patients. Similar to Student 64, Student 63 had the same two answers correct, with antibiotics not being the most effective manner to prevent infections and hand disinfection being the easiest way to prevent infections from spreading.i Student 63 justifies how surfaces and patients are the main route of transmission: “All patients may not necessarily control themselves and spreading their own secre- tions, while moving they touch the mouth (saliva) and then the surfaces of the corri- dor etc.” The student may be referring to droplet transmission. There is some idea of what patients and healthcare workers do, as evidenced by the comment “staff mainly wash their hands regularly and use gloves”. Claiming that “preventing vaccinations with protective gloves” is a clear misunderstanding of the purpose of gloves (Pittet & Boyce, 2003). Gloves should be used only when there is risk of contaminating the Fig. 3 Second example of the Household Hygiene Model (Student 64). 3 (ID 64 = student pseudonym, Q1-2 = justifications, P = patient, X = preventing spreading of pathogens, PPE = personal protective equipment, → = affects, causes) Fig. 3 Second example of the Household Hygiene Model (Student 64). (ID 64 = student pseudonym, Q1-2 = justifications, P = patient, X = preventing spreading of pathogens, PPE = personal protective equipment, → = affects, causes) 1 3 154 R. Eronen et al. xample of the Household Hygiene Model (Student 63). (ID 63 = student pseudonym, Q1-2 = justifications, P = patient, HCW = healthcare f ects causes) ig. 4 Third example of the Household Hygiene Model (Student 63). (ID 63 = student pseudonym, Q1-2 = justifications, P = patient, HCW = healt worker, → = affects, causes) 1 3 Practical nurse students’ misconceptions about infection… 155 HCW’s hands with secretions. It also seems that the student has no awareness of the purpose of disinfecting a nurse’s hands.i HCW’s hands with secretions. It also seems that the student has no awareness of the purpose of disinfecting a nurse’s hands.i Ten of the 29 (34%) mental models provided by the students were classified as instances of the Household Hygiene Model. Seven of these students scored 5–6 points out of a possible 8 from the multiple-true–false questions, and three students scored between 2 to 4 points. The mixed model In the Mixed Model, the students expressed some understanding of the critical role of the nurse in IPC. They expressed that the nurse is the main pathogen spreading link between the patients which is correct, but patients and visitors, surfaces and secretions were also reported as the main route of transmission as in the Household Hygiene Model. Hand disinfection or a hand rub, central in IPC, is also presented in this mental model. Everyday language such as “bugs” instead of professional phrases such as pathogens, bacteria or virus are typically used. The mental model of Student 51 is the most characteristic instance of the Mixed Model (see Fig. 5). The student scored only two points out of eight on the multi- ple-true–false questions, stating that all the alternatives represent the main route of transmission, and all the alternatives are the most effective and easiest manners to prevent spreading of pathogens. The student indicated professional knowledge and awareness of the critical role of the nurse in IPC stating that “Staff should clean their hands before and after touch- ing the patient. Protective gloves should be changed (between procedures/ when the patient changes).” It is not clear what Student 51 meant by stating “If the secretions are not handled properly, the contact surfaces are contaminated”. In any case, this statement is interpreted as household hygiene knowledge because it is not the main route of transmission. Whilst the knowledge of vaccinations and antibiotics is cor- rect in the sense that they can prevent infections, it is not the correct answer to the question regarding the most effective way to prevent infections. Stating that “Hand rub and the use of gloves always” emphasises the use of gloves, but there is no indication of the situations in which they are needed. Gloves should be used only in contact with secretions, not always (Pittet & Boyce, 2003). Student 51’s comments thus show some professional knowledge and some lay knowledge, which is typical for the Mixed Model.i Eighteen of the 29 (62%) students’ mental models were classified as instances of the Mixed Model. Six of these students scored a maximum of 8 points from the mul- tiple-true–false questions. The remaining students scored between 2 and 6 points. The transmission model The Transmission Model is based on scientific knowledge of IPC. In this model, the nurse is understood as the main pathogen spreading link between the patients (WHO, 2016), and it is acknowledged that the patients need to be protected against the spreading of pathogens by the HCWs’ hands. Hand disinfection, the cornerstone 3 3 R. Eronen et al. 156 ample of the Mixed Model (Student 51). (ID 51 = student pseudonym, Q1-2 = justifications, P = patient, X = preventing spreading of pathogens, HR = hand rub, d hygiene HCW=healthcare worker →=affects causes) g. 5 Example of the Mixed Model (Student 51). (ID 51 = student pseudonym, Q1-2 = justifications, P = patient, X = preventing spreading of pathogens, HR = hand rub, H=hand hygiene HCW=healthcare worker →=affects causes) f the Mixed Model (Student 51). (ID 51 = student pseudonym, Q1-2 = justifications, P = patient, X = preventing spreading of pathogens, HR = hand rub, e HCW=healthcare worker →=affects causes) Fig. 5 Example of the Mixed Model (Student 51). (ID 51 = student pseudonym, Q1-2 = justifications, P = patient, X = preventing spreading of pathogens, HR = han HH = hand hygiene, HCW = healthcare worker, → = affects, causes) 1 3 Practical nurse students’ misconceptions about infection… 157 of IPC, is expected to be mentioned (Pittet & Boyce, 2003). In addition, the use of professional language is a characteristic of this model (see Table 1). of IPC, is expected to be mentioned (Pittet & Boyce, 2003). In addition, the use of professional language is a characteristic of this model (see Table 1). Only one of the concept maps (Student 18) was categorised as an instance of the Transmission Model. The student scored a maximum of 8 points on the multiple- true–false questions. The student manifested professional knowledge in a written explanation by stat- ing that the nurse is the pathogen spreading link “with poor hand hygiene health- care workers spread diseases to other patients”, and the best way to prevent spread- ing of pathogens is hand disinfection “because hand disinfection is the most efficient and cheapest way to prevent the spreading of pathogens”. The student also used professional language (e.g., “pathogens”) (see Fig. 6). Discussion The Transmission Model focuses on protecting the patient and is based on the scientific finding that the healthcare worker is the main route of transmission; hand disinfection plays a crucial role in disrupting this trans- mission route. Nevertheless, it should be noted that the Transmission Model is only a starting point for the HCW to learn very detailed IPC related routines for perform- ing nursing procedures in compliance with the guidelines (Heininger et al., 2021; Purssell & Gould, 2022). It can be concluded that all the students except one demonstrated beliefs that could be classified as household conception or mixed conceptions that have self- protection as a common value. A majority of the students also understood the role of healthcare workers in spreading disease and the importance of hand disinfection as a means to prevent this. At the same time, they believed that wearing gloves can also serve that role without necessarily being aware of the potential danger that gloves can also be transmitters. The study design makes it impossible to draw conclusions on how many students entertain specific lay beliefs. The extent of the students’ jus- tifications for the answers to the multiple-true–false questions varied widely, which is a limitation of the study. Some of the participants provided elaborated responses, whereas others provided very short replies. However, this study provides evidence that a very small minority of the students at this stage of education have developed a scientific understanding of IPC in healthcare. These findings have implications for nursing education.ii Our findings are in line with findings in healthcare practice. Researchers of compliance to hand hygiene standards in healthcare institutions have hypothesised that a preference for prioritising self-protection can explain part of the disappoint- ing effects of promotional campaigns. Evidence for this is found in the discrepancy between hand rub ‘before patient contact’ (serving the patient) and ‘after patient contact’ (serving self). During the COVID-19 pandemic both increased dramatically but despite this the gap was not reduced (Israel et al., 2020). The results of this study can be interpreted along the lines of transition from lay to scientific understanding. In this view, transformations of earlier knowledge and integration of scientific knowledge has to take place. In this process, knowledge fragmentation and formation of synthetic conceptions may take place (Vosniadou & Skopeliti, 2014), however, emotional or motivational factors may complicate these transitions (Curtis & Biran, 2001; Pintrich et al., 1993). Discussion The purpose of this study was to make an inventory of the misconceptions of practi- cal nurse students about IPC at the beginning of their studies, and to discover (1) what kind of IPC misconceptions are held by students, and (2) how prevalent they are. The Three Mental Models that were created based on the students’ justifications indicated that every third student maintained beliefs that were corresponding to a Household Hygiene Model. Patients and visitors, surfaces and secretions were seen as the main routes of transmission and hence should be cleaned. The use of gloves was emphasised, whereas neither the HCW nor hand disinfection were mentioned. This set of beliefs seems to serve a self-protection goal (Dapper et al., 2022; Jansson et al., 2016; Jeong & Kim, 2016). The Mixed Model was observed by almost two thirds of the students. It included the same kind of household beliefs, for instance claiming that bacteria spread easily from surfaces, but was enriched with some sci- entific knowledge of IPC, i.e., knowledge about the role of the HCWs in spreading disease and the importance of using hand rub before and after dealing with a patient. It is important that teachers are aware of how their students encounter hygiene in everyday life and what transgression of their in-built standards might mean to them. Fig. 6 Example of the Transmission Model (Student 18). (ID 18 = student pseudonym, Q1-2 = justi- fications, P = patient, X = preventing spreading of pathogens, HH = hand hygiene, HCW = healthcare worker, → = affects, causes) Fig. 6 Example of the Transmission Model (Student 18). (ID 18 = student pseudonym, Q1-2 = justi- fications, P = patient, X = preventing spreading of pathogens, HH = hand hygiene, HCW = healthcare worker, → = affects, causes) Fig. 6 Example of the Transmission Model (Student 18). (ID 18 = student pseudonym, Q1-2 = justi- fications, P = patient, X = preventing spreading of pathogens, HH = hand hygiene, HCW = healthcare worker, → = affects, causes) 3 158 R. Eronen et al. Teachers should not confuse almost perfect test results with understanding and prac- ticing IPC, not even as an indicator that a student is on the right tract. Only one student expressed beliefs that are covered by the Transmission Model, reflecting sci- entific and discipline knowledge. Discussion The results are also in the line with earlier findings (Chi, 2013; Vosniadou & Skopeliti, 2014) that scientific mental models hardly exist in a “pure” form but include various perspectives simul- taneously (e.g., Solomon, 1983). In the present study, it could even be claimed that the Household Hygiene Model represents a rather sophisticated set of knowledge and skills that can in no way be called naïve. It is the context and goal that define its applicability (WHO, 2009). We agree with the view of Lundholm (2018) and Hall- dén (1999) that the lay model and the scientific model may coexist in the minds of students and HCWs provided that students and HCWs are able to activate the scien- tific/professional model in a professional context (see Fig. 1). As Lundholm (2018) 1 3 Practical nurse students’ misconceptions about infection… 159 and Halldén (1999) emphasise: it is the learners who should become aware of their own mental models and build meta-knowledge about the relation and differences between the two, especially in regard to context demands and applicability. and Halldén (1999) emphasise: it is the learners who should become aware of their own mental models and build meta-knowledge about the relation and differences between the two, especially in regard to context demands and applicability. Finally, the decision by an HCW to follow or not to follow IPC can also be con- sidered a social dilemma. In a social dilemma, 1) the payoff for everyone to defect, i.e., act in self-interest, is higher than the payoff for acting in the interest of the col- lective regardless of what the others do, but 2) all individuals receive a lower payoff if all defects. There is a temptation for HCWs to discard IPC and to relapse into lay conceptions in the workplace simply because following IPC requires constant cogni- tive effort without immediate payoff to the individual. The temptation is even higher if the other HCWs are serving as poor role models (Cox et al., 2014; Oh, 2021). The study by Harring and Lundholm (2018) indicated that students with knowledge of social dilemmas showed more willingness to take personal responsibility in the con- text of mitigating climate change. Research on the incorporation of social dilemmas into teaching could be a direction for future research, as well as research on emo- tional and motivational complicating factors. Discussion The practical implication of the study is that developing an awareness of students’ mental models in different situations related to hygiene, and their appropriateness to situation-specific demands, should be an explicit goal of nursing education. If the mental model of nurses does not involve the nurse as the pathogen-spreading link between patients, there is no reason for nurses to disinfect their hands or to learn about standard procedures. This awareness is referred to as “meta-knowledge” in Fig. 1. The ability to be aware of one’s thinking and to monitor one’s thinking in daily practice is important – even lifesaving, because in this context it is particularly easy to relapse into deeply-rooted thinking habits. We believe that the three mental models presented in this study can serve as reference points in class discussions on students’ understandings of IPC. Appendix 1 The multiple-true–false questions = correct answer, = incorrect answer 1 3 1 3 R. Eronen et al. 160 Declarations Ethics approval and consent to participate The study was granted ethical approval by the Ethical Review Board of the Finnish university involved in the study. The principal of the vocational school granted the research permit. Consent for publication The students who volunteered to participate in this study were debriefed about the study beforehand and signed an Informed Consent Form. Each authors has read and approved the final version of the paper. Competing interests The authors have no competing interests to declare that are relevant to the content of this article. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permis- sion directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ licenses/by/4.0/. Appendix 2 The two prototypical concept maps; Household Hygiene Model on the left and Transmission Model on the right side (P = patient, X = preventing transmission of pathogens, HD = hand disinfection, HH = hand hygiene, HW = hand wash, HCW = healthcare worker, → = affects, causes) Code availability Not applicable. 3 1 Practical nurse students’ misconceptions about infection… 161 Authors’ contributions The first author conceptualised, planned, collected the data, and wrote the original draft. All authors participated in the conceptualisation of the study, analysis of the results, reviewing, and editing the article. All authors read and approved the final manuscript. Funding Open Access funding provided by University of Turku (including Turku University Central Hospital). A total of EUR 3000 from the OPPI doctoral programme. Availability of data and materials As explained to the participants, the data will not be disclosed availab to any third party. References Vocational Qualifications in Social and Health Care. https://eperusteet.opintopolku.fi/eperusteet-service/api/dokumentit/6671197 i Guzzetti, B. J., Snyder, T. E., Glass, G. V., & Gamas, W. S. (1993). Promoting conceptual change in science: A comparative meta-analysis of instructional interventions from reading education and sci- ence education. Reading Research Quarterly, 28(2), 117–159. https://doi.org/10.2307/747886 Halldén, O. (1993). Leaners’ conceptions of the subject matter being taught A case from learning history. International Journal of Educational Research, 19(3), 317–325. Halldén, O. (1999). Conceptual change and contextualisation. In W. Schnotz, S. Vosniadou, & M. Car- retero (Eds.), New perspectives on conceptual change (pp. 53–65). Pergamon. perspectives on conceptual change (pp. 53–65). Perg Halldén, O., Scheja, M., & Haglund, L. (2009). The contextuality of knowedge: An intentional approach to meaning making and conceptual change. In S. Vosniadou (Ed.), International handbook of research on conceptual change (pp. 537–560). Routledge. Harring, N., & Lundholm, C. (2018). Does knowledge about social dilemmas generate cynical citizens? [Paper presentation]. American Political Science Association Annual Meeting, August 30-September 2, Boston, USA. Heininger, S. K., Baumgartner, M., Zehner, F., Burgkart, R., Söllner, N., Berberat, P. O., & Gartmeier, M. (2021). Measuring hygiene competence: the picture-based situational judgement test HygiKo. BMC Medical Education, 21(1), 410. https://doi.org/10.1186/s12909-021-02829-y Israel, S., Harpaz, K., Radvogin, E., Schwartz, C., Gross, I., Mazeh, H., Cohen, M. J., & Benenson, S. (2020). Dramatically improved hand hygiene performance rates at time of coronavirus pandemic. Clinical Microbiology and Infection, 26(11), 1566–1568. https://doi.org/10.1016/j.cmi.2020.06.002 Jansson, M. M., Syrjälä, H. P., Ohtonen, P. P., Meriläinen, M. H., Kyngäs, H. A., & Ala-Kokko, T. I. (2016). Simulation education as a single intervention does not improve hand hygiene practices: a randomized controlled follow-up study. American Journal of Infection Control, 44(6), 625–630. https://doi.org/10.1016/j.ajic.2015.12.030 p g j j Jeong, S. Y., & Kim, K. M. (2016). Influencing factors on hand hygiene behavior of nursing students based on theory of planned behavior: a descriptive survey study. Nurse Education Today, 36, 159– 164. https://doi.org/10.1016/j.nedt.2015.09.014f p g j Kelcíkova, S., Skodova, Z., & Straka, S. (2012). Effectiveness of hand hygiene education in a basic nurs- ing school curricula. Public Health Nursing, 29(2), 152–159. https://doi.org/10.1111/j.1525-1446. 2011.00985.x Kısacık, Ö. G., Ciğerci, Y., & Güneş, Ü. (2021). Impact of the fluorescent concretization intervention on effectiveness of hand hygiene in nursing students: a randomized controlled study. Nurse Education Today, 97, 104719. https://doi.org/10.1016/j.nedt.2020.104719 Korhonen, A., Vuori, A., Lukkari, A., Laitinen, A., Perälä, M., Koskela, T., & Pölkki, T. (2019). References Aarkrog, V., & Wahlgren, B. (2022). Goal orientation and decision-making in education. Vocations and Learning, 15(1), 71–86. https://doi.org/10.1007/s12186-021-09278-0 Ahopelto, I., Mikkilä-Erdmann, M., Olkinuora, E., & Kääpä, P. (2011). A follow-up study of medical students’ biomedical understanding and clinical reasoning concerning the cardiovascular system. Advances in Health Sciences Education, 16, 655–668. https://doi.org/10.1007/s10459-011-9286-3 AL-Rawajfah, O. M., & Tubaishat, A. (2015). Nursing students’ knowledge and practices of standard precautions: a Jordanian web-based survey. Nurse Education Today, 35(12), 1175–1180. https://doi. org/10.1016/j.nedt.2015.05.011 Boshuizen, H. P. A., & Marambe, K. N. (2020). Misconceptions in medicine, their origin and develop- ment in education and working life. International Journal of Educational Research, 100, 101536. https://doi.org/10.1016/j.ijer.2020.101536 Boshuizen, H. P. A., Vosniadou, S., & Lehtinen, E. (2020). Conceptual changes for and during work- ing life. International Journal of Educational Research, 104, 101682. https://doi.org/10.1016/j.ijer. 2020.101682 Caniza, M. A., Maron, G., Moore, E. J., Quintana, Y., & Liu, T. (2007). Effective hand hygiene education with the use of flipcharts in a hospital in El Salvador. Journal of Hospital Infection, 65(1), 58–64. https://doi.org/10.1016/j.jhin.2006.08.011 1 3 R. Eronen et al. 162 Chi, M. T. H. (2013). Two kinds and four sub-types of misconceived knowledge, ways to change it, and the learning outcomes. In S. Vosniadou (Ed.), International Handbook of Research on Conceptual Change (pp. 49–70). Routledge. https://doi.org/10.4324/9780203154472.ch3 g pp g p g Cieslak, P. R., Lee, L. E., Papafragkou, E., & An, N. (2009). Recurring norovirus outbreaks in a long- term residential treatment facility — Oregon, 2007. MMWR Morbidity and Mortality Weekly Cieslak, P. R., Lee, L. E., Papafragkou, E., & An, N. (2009). Recurring norovirus outbreaks in a long- term residential treatment facility — Oregon, 2007. MMWR Morbidity and Mortality Weekly Report, 58(25), 694–698. https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5825a2.htm Cox, J. L., Simpson, M. D., Letts, W., & Cavanagh, H. M. A. (2014). Putting it into practice: Infection control professionals’ perspectives on early career nursing graduates’ microbiology and infection control knowledge and practice. Contemporary Nurse, 49(1), 83–92. https://doi.org/10.1080/10376 178.2014.11081957 Curtis, V., & Biran, A. (2001). Dirt, disgust, and disease: Is hygiene in our genes? Perspectives in Biology and Medicine, 44(1), 17–31. https://doi.org/10.1353/pbm.2001.0001l Dapper, L., Dick, A., Nonnenmacher-Winter, C., & Günther, F. (2022). Influence of public health and infection control interventions during the severe acute respiratory syndrome coronavirus 2 pan- demic on the in-hospital epidemiology of pathogens: In hospital versus community circulating pathogens. Antimicrobial Resistance & Infection Control, 11(1), 140. https://doi.org/10.1186/ s13756-022-01182-zi Finnish National Agency for Education. (2018). References Increas- ing nursing students’ knowledge of evidence-based hand-hygiene: A quasi-experimental study. Nurse Education in Practice, 35, 104–110. https://doi.org/10.1016/j.nepr.2018.12.009 Linn, M. C. (2008). Teaching for conceptual change: Distinguish or extinguish ideas. In S. Vosniadou (Ed.), International Handbook of Research on Conceptual Change (pp. 694–722). Routledge. 1 1 3 3 Practical nurse students’ misconceptions about infection… 163 Lundholm, C. (2018). Conceptual change and the complexity of learning. In T. G. Amin & O. Levrini (Eds.), Converging Perspectives on Conceptual Change: Mapping an Emerging Paradigm in the Learning Sciences (pp. 34–42). Routledge. https://doi.org/10.4324/9781315467139 Lundholm, C. (2022). Conceptual change and teaching - focusing on social science and climate edu- cation. Keynote Speach Held at the 12th Conference on Conceptual Change, 8, 24–27. Zwolle, The Netherlands. Merenluoto, K., & Lehtinen, E. (2004). Number concept and conceptual change: Towards a systemic model of the processes of change. Learning and Instruction, 14(5 SPEC.ISS), 519–534. https:// doi.org/10.1016/j.learninstruc.2004.06.016 Mikkilä-Erdmann, M., Södervik, I., Vilppu, H., Kääpä, P., & Olkinuora, E. (2012). First-year medi- cal students’ conceptual understanding of and resistance to conceptual change concerning the central cardiovascular system. Instructional Science, 40(5), 745–754. https://doi.org/10.1007/ s11251-012-9212-yf Moll, L. C., Amanti, C., Neff, D., & Gonzalez, N. (1992). Funds of knowledge for teaching: Using a qualitative approach to connect homes and classrooms. Theory into Practice, 31(2), 132–141. https://doi.org/10.1080/00405849209543534 Oh, H. S. (2021). Knowledge, perception, and performance of hand hygiene and their correlation among nursing students in Republic of Korea. Healthcare, 9(7), 913. https://doi.org/10.3390/ healthcare9070913 Patel, V. L., & Groen, G. J. (1986). Knowledge based solution strategies in medical reasoning. Cognitive Science, 10(1), 91–116. https://doi.org/10.1016/S0364-0213(86)80010-6. 10.1016/ S0364-0213(86)80010-6. Pintrich, P. R., Marx, R. W., & Boyle, R. A. (1993). Beyond cold conceptual change: the role of moti- vational beliefs and classroom contextual factors in the process of conceptual change. Review of Educational Research, 63(2), 167. https://doi.org/10.2307/1170472 Pittet, D., & Boyce, J. M. (2003). Revolutionising hand hygiene in health-care settings: guidelines revisited. The Lancet Infectious Diseases, 3(5), 269–270. https://doi.org/10.1016/S1473- 3099(03)00601-7 Purssell, E., & Gould, D. (2022). Teaching health care students hand hygiene theory and skills: a systematic review. International Journal of Environmental Health Research, 32(9), 2065–2073. https://doi.org/10.1080/09603123.2021.1937580 p g Sairaalahygienia- ja infektiontorjuntayksikkö. (2023). Hoitoon liittyvien infektioiden torjunnan perus- teet Varsinais-Suomen hyvinvointialueella. Varsinais-Suomen hyvinvointialue /Tyks Sairaalahy- gienia- ja infektiontorjuntayksikkö. https://hoito-ohjeet.fi/OhjepankkiVSSHP/Suositus hoitoon liittyvien infektioiden torjunnasta.pdf y j p Södervik, I., Mikkilä-Erdmann, M., & Chi, M. T. H. (2019). Conceptual change challenges in medicine during professional development. References International Journal of Educational Research, 98(August 2018), 159–170. https://doi.org/10.1016/j.ijer.2019.07.003 ( g ) p g j j Solomon, J. (1983). Learning about energy: How pupils think in two domains. European Journal of Science Education, 5(1), 49–59. https://doi.org/10.1080/0140528830050105 Suetens, C., Latour, K., Kärki, T., Ricchizzi, E., Kinross, P., Moro, M. L., Jans, B., Hopkins, S., Hansen, S., Lyytikäinen, O., Reilly, J., Deptula, A., Zingg, W., Plachouras, D., & Monnet, D. L. (2018). Prevalence of healthcare-associated infections, estimated incidence and composite anti- microbial resistance index in acute care hospitals and long-term care facilities: results from two european point prevalence surveys, 2016 to 2017. Eurosurveillance, 23(46), 1–17. https://doi. org/10.2807/1560-7917.ES.2018.23.46.1800516 g Tversky, B. (2011). Visualizing Thought. Topics in Cognitive. Science, 3(3), 499–535. https://doi.org/ 10.1111/j.1756-8765.2010.01113.x Van de Wiel, M. W. J., Boshuizen, H. P. A., & Schmidt, H. G. (2000). Knowledge restructuring in expertise development: evidence from pathophysiological representations of clinical cases by students and physicians. European Journal of Cognitive Psychology, 12(3), 323–356. https://doi. org/10.1080/09541440050114543 Virtanen, A. (2020). THL - Tilastoraportti 02/2020 Sosiaali- ja terveydenhuollon ammattioikeudet 2010 – 2018. https://www.julkari.fi/bitstream/handle/10024/139099/Tr02_20.pdf?sequence=5& isAllowed=y y Vosniadou, S., & Brewer, W. F. (1992). Mental models of the earth: a study of conceptual change in childhood. Cognitive Psychology, 24(4), 535–585. https://doi.org/10.1016/0010-0285(92) 90018-W 1 3 R. Eronen et al. 164 Vosniadou, S., & Skopeliti, I. (2014). Conceptual change from the framework theory side of the fenc Science and Education, 23(7), 1427–1445. https://doi.org/10.1007/s11191-013-9640-3 Vosniadou, S., Ioannides, C., Dimitrakopoulou, A., & Papademetriou, E. (2001). Designing learning environments to promote conceptual change in science. Learning and Instruction, 11(4–5), 381– 419. https://doi.org/10.1016/S0959-4752(00)00038-4 Ward, D. J. (2013). The barriers and motivators to learning infection control in clinical placements: inter- views with midwifery students. Nurse Education Today, 33(5), 486–491. https://doi.org/10.1016/j. nedt.2012.05.024 White, R. T. (1994). Conceptual and conceptional change. Learning and Instruction, 4, 117–121. https:// doi.org/10.1016/0959-4752(94)90022-1 WHO. (2009). WHO Guidelines on hand hygiene in health care. In World Health Organization (Vol. 30, Issue 1). World Health Organization Press. https://apps.who.int/iris/bitstream/handle/10665/70126/ WHO_IER_PSP_2009.07_eng.pdf;sequence=1 yg g Issue 1). World Health Organization Press. https://apps.who.int/iris/bitstream/handle/10665/70126/ WHO_IER_PSP_2009.07_eng.pdf;sequence=1 WHO. (2016). Health care without avoidable infections: The critical role of infection prevention and con- trol. World Health Organization. https://www.who.int/publications/i/item/health-care-without-avoid able infections the critical role of infection prevention and control able-infections-the-critical-role-of-infection-prevention-and-control Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Riikka Eronen1 · Laura Helle1 · Tuire Palonen1 · Henny P. A. Boshuizen1,2 Riikka Eronen1 · Laura Helle1 · Tuire Palonen1 · Henny P. A. Boshuizen1,2 * Riikka Eronen riikka.m.eronen@utu.fi Laura Helle laura.helle@utu.fi Tuire Palonen tuire.palonen@utu.fi Henny P. A. Boshuizen Els.Boshuizen@ou.nl 1 Department of Teacher Education, University of Turku, Turku, Finland 2 Open University of the Netherlands, Faculty of Educational Sciences, Heerlen, The Netherlands * Riikka Eronen riikka.m.eronen@utu.fi Laura Helle laura.helle@utu.fi Tuire Palonen tuire.palonen@utu.fi Henny P. A. Boshuizen Els.Boshuizen@ou.nl 1 Department of Teacher Education, University of Turku, Turku, Finland 2 Open University of the Netherlands Faculty of Educational Sciences Heerlen The Netherlands 1 Department of Teacher Education, University of Turku, Turku, Finland 2 Open University of the Netherlands, Faculty of Educational Sciences, Heerlen, The Netherlands 1 3
https://openalex.org/W2136859124	https://www.arkat-usa.org/get-file/20331/	English	null	Methyl derivatives of tetracyclic psoralen analogues: antiproliferative activity and interaction with DNA	ARKIVOC	2,004	cc-by	7,835	Abstract A number of new tetracyclic psoralen derivatives were studied. The fourth ring is constituted by cyclopentane (4, 7 and 9), cyclohexane (11, 15 and 18) or benzene (12, 16 and 20) fused to either 4',5' or 3,4 photoreactive double bond of tricyclic furocoumarin moiety. The photoantiproliferative activity of all compounds, tested on two human tumor cell lines (HeLa and HL-60), appeared from 8 to 22 times higher than that of the well-known photochemotherapeutic drug 8-methoxypsoralen (8-MOP) in HeLa, slightly higher in HL-60. Interestingly, the evaluation of skin phototoxicity on guinea pigs evidenced a decrease in erythema induction for all compounds with respect to the drug. As regards the molecular mechanism of action, the ability to photoadd to DNA is demonstrated by isolation and characterization of the photoadducts and by the ability to give rise to interstrand cross-links for the difunctional derivatives. Keywords: Tetracyclic psoralen analogues, antiproliferative activity, DNA interaction Keywords: Tetracyclic psoralen analogues, antiproliferative activity, DNA interaction ARKIVOC 2004 (v) 131-146 ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Vincenzo Tortorella Dedicated to Professor Vincenzo Tortorella on the occasion of his “Fuori Ruolo” status (received 22 Dec 03; accepted 20 Feb 04; published on the web 04 Mar 04) Dedicated to Professor Vincenzo Tortorella on the occasion of his “Fuori Ruolo” status (received 22 Dec 03; accepted 20 Feb 04; published on the web 04 Mar 04) Lisa Dalla Via, a Eugenio Uriarte,b Lourdes Santana,a Sebastiano Marciani Magno,a and Ornella Giaa a Department of Pharmaceutical Sciences, University of Padova, via Marzolo 5, 35131 Padova, Italy b Department of Organic Chemistry, University of Santiago de Compostela, Spain E-mail: ornellamaria.gia@unipd.it Methyl derivatives of tetracyclic psoralen analogues: antiproliferative activity and interaction with DNA Lisa Dalla Via, a Eugenio Uriarte,b Lourdes Santana,a Sebastiano Marciani Magno,a and Ornella Giaa Introduction Inside the wide field of photochemotherapy the psoralens play a significant role. In particular, skin diseases characterized by hyperproliferation, such as psoriasis, are effectively treated by means of PUVA therapy (psoralen plus UVA light).1-3 Moreover, psoralens are successfully employed for the treatment of cutaneous T-cell lymphoma, a T-cell mediated disorder, by an extracorporeal photochemotherapy called photopheresis.4,5 Nevertheless, the existence of some undesired side-effects limits the therapeutic use of psoralens. Indeed, both short-term (erythema, ©ARKAT USA, Inc Page 131 ISSN 1424-6376 ISSN 1424-6376 ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Vincenzo Tortorella hyperpigmentation) and long-term (benign keratoses, premalignant keratoses, skin cancers) undesired effects can occur.2 Between psoralens, 8-methoxypsoralen (8-MOP) constitutes the drug most widely employed, but also 5-methoxypsoralen and 4,5',8-trimethylpsoralen hold a noticeable role in photochemotherapy, even though to a lesser extent.6,7 The psoralen tricyclic moiety, due to its planarity, is able to intercalate inside base pairs of DNA. Upon UVA irradiation (365 nm) a photoaddition reaction between the 4’,5’ furan side and/or the 3,4 pyrone side double bonds and the 5,6 double bond of a pyrimidine base, generally thymine, can take place. The ability of the furan side monoadduct to absorb at 365 nm can allow the formation of diadducts involving two bases belonging to the opposite DNA strands.8 The psoralen tricyclic moiety, due to its planarity, is able to intercalate inside base pairs of DNA. Upon UVA irradiation (365 nm) a photoaddition reaction between the 4’,5’ furan side and/or the 3,4 pyrone side double bonds and the 5,6 double bond of a pyrimidine base, generally thymine, can take place. The ability of the furan side monoadduct to absorb at 365 nm can allow the formation of diadducts involving two bases belonging to the opposite DNA strands.8 Many psoralen derivatives have been synthesized up to now with the aim of weakening the undesired side effects. Inside the different lines of research, the condensation of a fourth benzenic or cyclohexenyl ring to the tricyclic psoralen chromophore led to some tetracyclic moieties endowed with interesting photobiological properties.9-15 In particular, it was shown that the presence of a fourth aromatic ring caused the disappearance of the skin photosensitizing potency, evaluated as erythema induction. Furthermore, for some benzopsoralen derivatives a noteworthy increase in photoantiproliferative activity on human tumor cell lines, with respect to 8-MOP, was demonstrated. Introduction As regards the tetrahydrobenzopsoralen moiety, the most interesting derivatives showed a cytotoxic activity higher with respect to that exerted by the reference drug along with a significant decrease of skin phototoxicity.14,15 The condensation of a cyclopentane ring to the tricyclic psoralen nucleus was also reported.16 In detail, the synthesized cyclopentenepsoralen bearing the cyclopentane ring fused to the 4',5' double bond appears to show a slight improvement in photobiological properties, i.e. increased photoantiproliferative activity on human tumor cell line and diminished skin photosensitization, with respect to the drug 8-MOP.16 Moreover, in previous studies it was demonstrated that the introduction of methyl groups in the psoralen cromophore appeared to increase significantly its ability to photobind to DNA and, in some cases, to improve the photobiological properties.17-19 In this paper we studied the photobiological behaviour of a series of structurally related psoralen derivatives endowed with a tetracyclic nucleus and carrying two methyl groups in different positions. In detail, the derivatives taken into consideration are characterized by the presence of a cyclopentane (4, 7 and 9), a cyclohexane (11, 15 and 18) or a benzene (12, 16 and 20) ring fused to the tricyclic psoralen cromophore. The fourth ring is condensed at the level of the furan or pyrone photoreactive double bond. Furthermore, a methyl group is linked in the 8 position along with a further methyl at the level of the 4', 5' or 4 position of the tricyclic psoralen structure. The photoantiproliferative activity on human tumor cell lines, the skin phototoxicity on guinea pigs and the photoaddition to DNA were evaluated in comparison with the well-known drug 8- MOP. Furthermore, the isolation and characterization of furan photoadducts with thymine were reported. In detail, the derivatives taken into consideration are characterized by the presence of a cyclopentane (4, 7 and 9), a cyclohexane (11, 15 and 18) or a benzene (12, 16 and 20) ring fused to the tricyclic psoralen cromophore. The fourth ring is condensed at the level of the furan or pyrone photoreactive double bond. Furthermore, a methyl group is linked in the 8 position along with a further methyl at the level of the 4', 5' or 4 position of the tricyclic psoralen structure. The photoantiproliferative activity on human tumor cell lines, the skin phototoxicity on guinea pigs and the photoaddition to DNA were evaluated in comparison with the well-known drug 8- MOP. Introduction Furthermore, the isolation and characterization of furan photoadducts with thymine were reported. Page 132 ©ARKAT USA, Inc ISSN 1424-6376 ISSN 1424-6376 ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Vincenzo Tortorella Chemistry The pyrone side derivative 9 was obtained by the treatment of the hydroxycoumarin 5 with 2,3-dibromopropene, and following cyclization of the achieved product 8 in 40% overall yield. Compound 18 was also prepared through a different synthetic route with respect to that already described22 starting from 17 by reaction with the N,N-diethylaniline as base, in 57% yield. Chemistry The studied compounds (4, 7, 9, 11, 12, 15, 16, 18, 20) were obtained starting from 2- methylresorcinol (1), as shown in Scheme 1. y ( ), 3 (90.1%) 5 6 7 4 (30%) 8 (77.7%) 9 (52%) O O O O O O O O O H O O O O O O O Br O O O O H OH O O H O O O O H O O O O O O Br O O O O O O O O O O O Br O O O O O O O O O O O O O O O O O O O c d b a d f e e e 2 1 13 10 14 17 (75.9%) 18 (57.1%) 11 15 19 (81.7%) 12 16 20 (52%) Scheme 1. Reagents: (a) ethylacetoacetate, H2SO4; (b) 2-chlorocyclopentanone; (c) NaOH K2CO3 , 2,3-dibromopropene; (e) N,N-diethylaniline; (f) 2,3-dichloro-5,6-dicyan benzoquinone, toluene. The yields are reported in parentheses. 12 16 Scheme 1. Reagents: (a) ethylacetoacetate, H2SO4; (b) 2-chlorocyclopentanone; (c) NaOH; (d) K2CO3 , 2,3-dibromopropene; (e) N,N-diethylaniline; (f) 2,3-dichloro-5,6-dicyano-1,4- benzoquinone, toluene. The yields are reported in parentheses. ISSN 1424-6376 Page 133 ©ARKAT USA, Inc ISSN 1424-6376 Page 133 Page 133 ©ARKAT USA, Inc Page 133 ©ARKAT USA, Inc ISSN 1424-6376 ISSN 1424-6376 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 The preparation of 7, 11, 12, 15, 16 and 18, in accordance with a general approach to the psoralen skeleton, has been described elsewhere.9,20,21,22 The preparation of 7, 11, 12, 15, 16 and 18, in accordance with a general approach to the psoralen skeleton, has been described elsewhere.9,20,21,22 Pechmann condensation of compound 1 and ethyl acetoacetate afforded the corresponding hydroxycoumarin 2, which through a Williamson reaction with 2-chlorocyclopentanone gave the oxoether 3 in 90% yield. Cyclization of 3 in strongly alkaline solution afforded the ciclopentenefurocoumarin 4 in 30% yield. The pyrone side derivative 9 was obtained by the treatment of the hydroxycoumarin 5 with 2,3-dibromopropene, and following cyclization of the achieved product 8 in 40% overall yield. Compound 20 was synthesised from hydroxycoumarin 13 in a 4-step reaction: the reaction of the latter with 2,3-dibromopropene, in the same manner as 8 from 5, gave the resulting ether 17 which then was oxidized with DDQ in toluene to give 19 in 62% overall yield. Finally, the cyclization of 19 with N,N-diethylaniline produced the benzoderivative 20 in 52% yield. Photobiological activity Comparing the above results with the chemical molecular structure of the tetracyclic methyl derivatives, it is evident that the extension of the tricyclic psoralen moiety, due to the condensation of a fourth ring, appears to be always relevant to increase the photoantiproliferative activity. Nevertheless, the ability to inhibit cell growth seems not to be correlated with the chemical structure of the fourth ring. In particular, both the steric hindrance caused by lack of aromaticity in the hydrogenated rings (cyclohexane for 11, 15 and 18, cyclopentane for 4, 7 and 9) and the delocalization of the electron density due to a tetracyclic aromatic system (12, 16 and 20) similarly affect the antiproliferative ability of the psoralen chromophore. An analogous consideration has to be made with regards to the role played by the methyl substituents. Indeed, it seems that their position (4, 4’ or 5’ of the psoralen moiety) do not constitute a critical factor for the cytotoxicity. In detail, it could be stated that the presence of a methyl substituent both at the pyrone side double bond and at the furan side double bond does not compromise the capacity of these site to exert their photoreactivity. The induction of skin phototoxicity, determined by evaluating the appearance of erythema, a marker of cutaneous sensitization, was investigated and the results obtained are shown in Table 1. It can be noted that, unlike in the case of the antiproliferative activity, the chemical structure of the fourth condensed ring is crucial in determining the appearance of this undesired side effect. In particular, the benzopsoralen derivatives 12, 16 and 20 appear to be devoid of photosensitizing effect, even if they are applied on skin at a concentration 3.8 times higher with respect to that of the reference drug. As regards the derivatives characterized by the condensation of a cyclohexenyl ring it is noteworthy that the presence of the fourth ring at the pyrone side of the tricyclic psoralen moiety, along with the presence of a methyl substituent at 4’ position of the furan side (compound 15), constitutes the most interesting tetrahydrobenzopsoralen structure, since it does not induce skin phototoxicity. By contrast, with analogues 11 and 18 the appearance of a certain skin photosensitization was scored, even though it is significantly lower if compared with the effect induced by 8-MOP. Photobiological activity The antiproliferative activity of the new methyl derivatives, cyclopentene- (4, 7 and 9) tetrahydrobenzo- (11, 15 and 18) and benzo-psoralens (12, 16 and 20) was evaluated on HeLa and HL-60 human tumor cell lines. The photochemotherapeutic drug 8-MOP was taken into account as reference compound. The results obtained after exposure to UVA light (0.793 J cm-2 at 365 nm) are reported in Table 1 and expressed as IC50 values, i.e. the concentration of compound (µM) able to induce 50% cell death with respect to the control culture. Table 1. Cell growth inhibition and skin phototoxicity in guinea pigs in the presence of examined compounds and 8-MOP as reference drug Table 1. Cell growth inhibition and skin phototoxicity in guinea pigs in the presence of examined compounds and 8-MOP as reference drug Table 1. Cell growth inhibition and skin phototoxicity in guinea pigs in the presence of examined compounds and 8-MOP as reference drug IC50 (µM) of cell lines Skin phototoxicity Compd HeLa HL-60 dose (mmol/cm2) erythema intensitya 4 0.65±0.15 2.4±0.4 0.08 + + - 7 1.05±0.26 3.3±0.6 0.08 + + - 9 0.66±0.16 2.8±0.6 0.08 + + - 11 0.59±0.13 1.7±0.2 0.08 + - - 15 0.45±0.05 2.7±0.3 0.08 - - - 18 0.65±0.07 4.3±0.4 0.08 + - - 12 0.63±0.06 1.8±0.2 0.19 - - - 16 0.47±0.07 2.0±0.2 0.19 - - - 20 0.65±0.07 1.4±0.1 0.19 - - - 8-MOPb 10±3 5.4±0.7 0.05 + + + a + + +, strong with edema; + + -, strong without edema; + - -, mild; - - - absent. b taken from from ref. 14. ©ARKAT USA, Inc Page 134 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 All the methyl derivatives are able to exert a noticeable antiproliferative activity towards both tumor cell lines. In detail, in the HeLa cells the IC50 values of the new compounds appear to be from 8 to 22 times lower with respect to that calculated for 8-MOP. As regards the human promyelocytic leukemia cell line, HL-60, it appears to be less sensitive toward the treatment with the new compounds. Indeed, the IC50 values are only slightly lower in comparison with that of the reference drug. It is interesting to note that, notwithstanding the remarkable antiproliferative activity exerted upon treatment with UVA light, none of the considered compounds show cytotoxic effect in the dark (data not shown). Photobinding to DNA It is well known that in the ground state the psoralen tryciclic moiety forms a preliminary complex with DNA in which the planar chromophore undergoes intercalation inside two base pairs.23 After UVA irradiation, the intercalated ligand photobinds covalently to DNA giving rise to monoadducts and diadducts. Similar behavior was also already observed for various tetracyclic psoralen derivatives carrying both a benzene10,11,14,15 or a cyclohexane11,14,15 as fourth ring. In detail, in these previous studies the ability of benzopsoralen and tetrahydrobenzopsoralen derivatives to photobind to DNA bases was widely demonstrated by the evaluation of the amount of radiolabelled compounds photoadded to the macromolecule10,11,14 and by the isolation and characterization of the monoadducts.15 In particular, it was clearly shown that the condensation of a cyclohexenyl or benzenic ring to a photoreactive double bond affects its photoaddition ability towards the DNA bases in a very different way. In detail, the steric hindrance caused by lack of aromaticity in the hydrogenated ring does not prevent the ability to photoaddition, while the delocalization of the electron density of the photoreactive double bond of the furocoumarin moiety in a tetracyclic aromatic system impedes its involvement in the cycloaddition.11,14,15 In this connection, the investigation was also undertaken into the photoaddition process of the methyl tetrahydrobenzo and benzopsoralen derivatives. In detail, the irradiation (365 nm) of an aqueous solution of DNA in the presence of the considered compounds, followed by precipitation and acid hydrolysis, as described in the Experimental Section, allowed us to isolate a photoproduct characterized by a strong violet fluorescence upon exposure to 365 nm UV light. This characteristic is typical of furocoumarins having a 3,4-ethylenic bond within the pyrone ring and thus is usually retained consistent with the molecular structure of the furan side monoadduct.24 Furthermore, the UV absorption spectra of an ethanol solution of the isolated fluorescent products showed the disappearance of the peculiar furocoumarin band around 300 nm and an evident absorption at 330 nm which gradually disappears as the irradiation time is increased to 254 nm (spectra not shown). This behavior further confirms the above assumption; indeed it was already established that the C4-cycloadducts undergo breakage upon 254 nm irradiation, yielding the parent compounds, i.e. the psoralen and the DNA base. As already demonstrated for psoralens and also for benzo and tetrahydrobenzopsoralen derivatives,14,15 inside DNA bases, thymine constitutes by far the preferred target of the photoreaction. Photobiological activity Finally, the compounds 4, 7 and 9, characterized by the condensation of a cyclopentene ring at pyrone or furan side, induce skin phototoxicity. Nevertheless, it has to be underlined that also in this case the undesired side effect is diminished with respect to that exerted by the reference drug. ©ARKAT USA, Inc ISSN 1424-6376 Page 135 ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Vincenzo Tortorella ISSN 1424-6376 Photobinding to DNA The mass spectra of the photoproduct of the tetrahydrobenzopsoralen 11, where the major peak appears at m/z=395, and of the benzopsoralen 16 which shows the major peak at m/z=391, are consistent with a thymine-11 and a thymine-16 photoadduct, respectively (spectra not shown). As regards the cyclopentene-psoralen tetracyclic moiety, a previous study reports some indications on its ability to photobind covalently to DNA.16 In this connection, it appeared of interest to investigate the ability of 4, 7 and 9 to photoadd to the macromolecule. In detail, the capacity of the methyl cyclopentenepsoralen derivatives to give rise to a covalent photoadduct ©ARKAT USA, Inc Page 136 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 with DNA bases was demonstrated by the isolation and characterization of the furan side monoadduct obtained after UVA irradiation of an aqueous solution of DNA and the above compounds as described in the Experimental Section. After TLC separation a fluorescent band appears. The UV spectrum of an ethanol solution of this photoproduct shows the typical characteristics of the C4-cycloadducts between the furan double bond of the furocoumarin and the 5,6 double bond of a pyrimidine base, i.e. an evident absorption at 330 nm and the disappearance of the peculiar furocoumarin band around 300 nm, present in the spectrum of 4, 7 and 9. Photoreversion experiments provide further evidence on the photoadduct formation. In Figure 1, the UV absorption spectra of the photoproduct, obtained by the photoreaction between 4 and DNA, before (line a) and after increasing periods of irradiation at 254 nm (lines b-f), is shown as an example. p wavelength (nm) 240 280 320 360 400 absorbance 0.0 0.2 0.4 0.6 0.8 1.0 a f Figure 1. UV absorption spectra of an ethanol solution of furan cycloadduct obtained from DNA and 4 before (line a) and after irradiation at 254 nm (10, 30, 60, 120, 180 min, lines b-f, respectively). wavelength (nm) 240 280 320 360 400 absorbance 0.0 0.2 0.4 0.6 0.8 1.0 a f absorbance Figure 1. UV absorption spectra of an ethanol solution of furan cycloadduct obtained from DNA and 4 before (line a) and after irradiation at 254 nm (10, 30, 60, 120, 180 min, lines b-f, respectively). Page 137 ©ARKAT USA, Inc Page 137 ©ARKAT USA, Inc ISSN 1424-6376 Issue in Honor of Prof. Photobinding to DNA Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 The analysis of the fluorescent photoproducts by mass spectrometry confirms also for cyclopentenepsoralen derivatives 4, 7 and 9 the preference to covalent photoadd to pyrimidine bases, and in particular to thymine. Indeed, the mass spectra show the presence of a major peak at m/z=381 which is consistent with a thymine-cyclopentenepsoralen cycloadduct (spectra not shown). In particular, the results obtained allow us to affirm that, similarly to what was demonstrated for the tetrahydrobenzosporalen moiety, the cyclopentane ring fused at a photoreactive double bond did not seem to compromise its DNA photoaddition capacity. Cross-linking The furan side monoadduct possesses the important property to absorb at 365 nm. This ability allows the formation of interstrand cross-links. As already stated, the condensation of the aromatic fourth ring renders the tetracyclic benzopsoralens monofunctional molecules, indeed both 16 and 20 are unable to form interstrand cross-links (data not shown). Regarding the tetrahydrobenzopsoralen derivatives 11, 15 and 18, Figure 2a shows the results obtained by denaturation-renaturation experiments, in comparison with 8-MOP. Figure 2a. Cross-linking of compounds 11, 15, 18 and 8-MOP to double stranded DNA (nucleotide/drug=75) as a function of irradiation time. Figure 2a. Cross-linking of compounds 11, 15, 18 and 8-MOP to double stranded DN ( l id /d 75) f i f i di i i Figure 2a. Cross-linking of compounds 11, 15, 18 and 8-MOP to double stranded DNA (nucleotide/drug=75) as a function of irradiation time. ©ARKAT USA, Inc Page 138 Page 138 ISSN 1424-6376 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 From the behaviors shown in Figure 2a it can be noted that a noticeable difference appears between the ability to induce cross-links exerted by 15 and 18 with respect to that of 11. In particular, the two tetrahydrobenzopsoralens characterized by the condensation of the fourth ring to the pyrone side of the psoralen chromophore (compounds 15 and 18) show a negligible capacity to induce cross-links. Also, the derivative carrying the fourth condensed ring at the 4’,5’ furan side double bond (compound 11) is able to exert a much higher capacity, even considerably higher in comparison with that induced by the reference drug 8-MOP. These overall results appear to be in agreement with previous studies where the ability to induce cross-links of furan side14 and pyrone side15 tetrahydrobenzopsoralens was evaluated. Denaturation-renaturation experiments was performed also for cyclopentenepsoralen derivatives 4, 7 and 9. The results shown in Figure 2b indicate for these compounds a remarkable ability to give rise to cross-links with double stranded DNA. Indeed, in the presence of both 4, 7 and 9, the double helix shows a renaturation capacity higher with respect to that induced by 8- MOP. Figure 2b. Cross-linking of compounds 4, 7, 9 and 8-MOP to double stranded DNA (nucleotide/drug=75) as a function of irradiation time. Cross-linking m in 0 2 4 6 8 1 0 1 2 1 4 1 6 log % non renaturated fraction 0 .8 1 .0 1 .2 1 .4 1 .6 1 .8 2 .0 8-M O P 7 4 9 m in 0 2 4 6 8 1 0 1 2 1 4 1 6 log % non renaturated fraction 0 .8 1 .0 1 .2 1 .4 1 .6 1 .8 2 .0 8-M O P 7 4 9 log % non renaturated fraction 8-M O P 7 Figure 2b. Cross-linking of compounds 4, 7, 9 and 8-MOP to double stranded DNA (nucleotide/drug=75) as a function of irradiation time. Furthermore, similarly to tetrahydrobenzopsoralens 11, 15 and 18, also for the cyclopentene derivatives this capacity appears dependent from the position of the fourth ring. In detail, Page 139 ©ARKAT USA, Inc ©ARKAT USA, Inc ISSN 1424-6376 ISSN 1424-6376 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 compound 4, characterized by the condensation of the cyclopentane at the level of furan side, is able to induce cross-links more efficiently with respect to the congeners carrying the cyclopentane ring condensed at the 4’,5’ double bond (7 and 9). Taken together, this results seem to indicate that the steric hyndrance, due to the presence of a saturated ring at the level of 3,4 photoreactive double bond of the furan side monoadduct, hindered the subsequent monoaddition. On the contrary, once the furan side monoadduct takes place, the presence of a fourth ring at the 4’,5’ level probably does not affect significantly the subsequent photoaddition involving the 3,4 pyrone side. Comparing the above data with those reported in Table 1, it is possible to conclude that, notwithstanding the fact that cross-links undoubtedly constitute lethal damage for the cell, there is not a linear relationship between the capacity to give rise to this molecular event and the ability to induce cell death. Consequently, it is reasonable to assume that they are not the leading molecular consequence accountable for the cellular cytotoxicity derived from UVA treatment in the presence of these derivatives. Similarly, despite a previous hypothesis of a correlation between furocoumarin capacity for crosslinking DNA and skin phototoxicity,25 the obtained results (see Fig. 2a and 2b and Table 1) appear in conflict with this assumption and indicate, rather, the existence of complex mechanism(s) responsible for this undesired side effect. ISSN 1424-6376 Conclusions The photobiological behaviour of cyclopentene- (4, 7 and 9), tetrahydrobenzo- (11, 15 and 18) and benzo-psoralens (12, 16 and 20), carrying two methyl groups in different positions of the psoralen cromophore, was investigated. In particular, for all the tested compounds a noticeable and comparable photoantiproliferative activity was scored. Interestingly, in HeLa cells they appears to exert an antiproliferative activity from 8 to 22 times greater with respect to that of 8- MOP. On the other hand, evaluation of skin phototoxicity brought out significant differences depending on the type of the fourth ring condensed to the psoralen moiety. In particular, the following order in inducing skin photosensitisation is detected: 8-MOP>cyclopentene- psoralens>tetrahydrobenzo-psoralens>benzo-psoralens. In this connection, it is noteworthy that in our experimental conditions the derivatives 12, 16 and 20 appear unable to induce the appearance of any erythema, even when tested at a concentration higher than that of 8-MOP. As regards the ability to interact with DNA, the tetracyclic derivatives seem to act like the parent compound. The capacity to photoreact with the macromolecule was demonstrated, and in particular, a preference toward the pyrimidine base thymine was highlighted. As regards the capacity to behave as bifunctional molecules, both tetrahydrobenzopsoralens and cyclopentenepsoralens are able to induce interstrand cross-links. Nevertheless, the capacity to photoreact with the opposite DNA strands appears strongly dependent on the photoreactive side at which the condensation of the fourth ring occurs. In detail, condensation at the level of the ©ARKAT USA, Inc ISSN 1424-6376 Page 140 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 pyrone double bond strongly prevents the formation of cross-links as compared with condensation on the furan side. Furthermore, it can be underlined that this inhibitory effect is more pronounced for tetrahydrobenzopsoralens than for cyclopentenepsoralens and it could be attributable to the wider steric hindrance of the cyclohexane, so that it could influence more strongly the correct superimposition of the photoreactive double bonds during the bifunctional photoaddition. In conclusion, the study of the photobiological behavior of structurally correlated tetracyclic psoralen derivatives allowed us to highlight some relationships between molecular events and structural properties and this knowledge could constitute a preliminary goal in the rational planning of new more promising photochemotherapeutic agents. Experimental Section 1H NMR (CDCl3): 2.48 (s, 3H, Me-C8), 2.57 (s, 3H, Me-C4), 2.59 (m, 2H, CH2CH2CH2), 2.78 (m, 2H, CH2-C5′), 2.90 (m, 2H, CH2-C4′), 6.24 (s, 1H, H-3), 7.41 (s, 1H, H-5). 13C NMR (CDCl3): 8.9 (Me-C8), 19.7 (Me-C4), 23.1 (CH2CH2CH2), 25.6 (CH2-C4′), 27.6 (CH2-C5′), 110.6 (C8), 111.3 (C3), 113.2 (C5), 116.2, 121.7, 123.3, 148.9, 153.8, 161.4, 162.1, 164.9 (C2). MS m/z (%): 254(M+, 80), 226(M+-CO, 100), 225(98), 199(18), 183(15), 149(56). Anal. C16H14O3: C, H. 2922, 2850, 1706, 1558, 1480, 1396, 1125, 1092. 1H NMR (CDCl3): 2.48 (s, 3H, Me-C8), 2.57 (s, 3H, Me-C4), 2.59 (m, 2H, CH2CH2CH2), 2.78 (m, 2H, CH2-C5′), 2.90 (m, 2H, CH2-C4′), 6.24 (s, 1H, H-3), 7.41 (s, 1H, H-5). 13C NMR (CDCl3): 8.9 (Me-C8), 19.7 (Me-C4), 23.1 (CH2CH2CH2), 25.6 (CH2-C4′), 27.6 (CH2-C5′), 110.6 (C8), 111.3 (C3), 113.2 (C5), 116.2, 121.7, 123.3, 148.9, 153.8, 161.4, 162.1, 164.9 (C2). MS m/z (%): 254(M+, 80), 226(M+-CO, 100), 225(98), 199(18), 183(15), 149(56). Anal. C16H14O3: C, H. 7-(β-Bromoallyloxy)-3,4-cyclopentene-8-methylcoumarin (8). This compound was prepared from 5 (1.0 g, 4.62 mmol)2, K2CO3 (958 mg, 6.93 mmol) and 2,3-dibromopropene (716 µL, 1.38 g, 6.93 mmol) in the same way as 3. The crude product was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding pure 8 (1.2 g, 77.7%). Mp 140oC. IR: 2919, 1716, 1611, 1373, 1282, 1109, 803. 1H NMR (CDCl3): 2.19 (m, 2H, CH2CH2CH2), 2.38 (s, 3H, Me-C8), 2.90 (t, J = 7.5 Hz, 2H, CH2-C4), 3.03 (t, J = 7.6 Hz, 2H, CH2-C3), 4.72 (s, 2H, CH2O), 5.71 (d, J = 2.1 Hz, 1H, C=CH), 6.01 (d, J = 2.1 Hz, 1H, C=CH), 6.77 (d, J = 8.6 Hz, 1H, H-6), 7.23 (d, J = 8.6 Hz, 1H, H-5). 13C NMR (CDCl3): 8.61 (Me-C8), 22.56 (CH2CH2CH2), 30.40 (CH2-C4), 32.07 (CH2-C3), 72.02 (CH2O), 108.09 (C6), 113.26 (C8), 114.81, 118.00 (CH2=C), 122.47 (C5), 124.94, 126.52, 153.40, 156.40, 157.63, 160.58 (C2). MS m/z (%): 336([M+2]+, 12), 335(M+, 12), 255(76), 215(29), 187(100), 128(15). Anal. C16H15BrO3: C, H. 3,4-Cyclopentene-5′,8-dimethylfuro[3,2-g]coumarin (9). A solution of 8 (218 mg, 0.65 mmol) in N,N-diethylaniline (5 mL) was refluxed for 40 h. The solution was cooled, and ethyl ether was added (35 mL). The precipitate was filtered, washed with 1N NaOH (3 x 25 mL) and 6N HCl (2 x 25 mL), the extract was dried (Na2SO4), and the solvent was evaporated under reduced pressure. Experimental Section General Procedures. Melting points are uncorrected and were determined in a Reichert Kofler thermopan or in capillary tubes in a Büchi 510 apparatus. IR spectra were recorded in a Perkin- Elmer 1640FT spectrometer (KBr disks, υ in cm-1). 1H-NMR (300 MHz) and 13C-NMR (75.4 MHz) spectra were recorded in a Bruker AMX spectrometer, using TMS as internal standard (chemical shifts in δ values, J in Hz). Mass spectrometry was carried out on a Kratos MS-50 or on a Varian AT-711 spectrometer. Elemental analyses were performed by a Perkin-Elmer 240B microanalyser and were within ±0.4% of calculated values in all cases. Flash chromatography (FC) was performed on silica gel (Merck 60, 230-400 mesh); analytical TLC was performed on precoated silica gel plates (Merck 60 F254, 0.25 mm). 4,8-Dimethyl-7-(2′-oxocyclopentyloxy)coumarin (3). To a solution of 2 (500 mg, 2.62 mmol)26 in dry acetone (30 mL) were added 2-chlorocyclopentanone (525 µL, 623 mg, 5.25 mmol) and K2CO3 (726 mg, 5.25 mmol), and the reaction mixture was refluxed for 24 h. The precipitate was filtered out and the solvent concentrated under reduced pressure. The crude product was purified by FC using hexane/ethyl acetate (3:1) as eluent, yielding pure 3 (644 mg, 90.1%). Mp 187oC. IR: 2922, 1722, 1605, 1288, 1119. 1H NMR (CDCl3): 2.09 (m, 2H + 2H, H-4′ + H-5′), 2.31 (s, 3H, Me-C8), 2.39 (d, J = 1.1 Hz, 3H, Me-C4), 2.47 (m, 2H, H-3′), 4.70 (t, J = 7.8 Hz, 1H, H-1′), 6.14 (d, J = 1.1 Hz, 1H, H-3), 6.94 (d, J = 8.8 Hz, 1H, H-6), 7.39 (d, J = 8.8 Hz, 1H, H-5). 13C NMR (CDCl3): 8.40 (Me-C8), 17.26 (Me-C4), 18.70 (C4′), 29.67 (C5′), 35.24 (C3′), 79.97 (C1′), 109.52 (C3), 112.12 (C6), 114.35 (C8), 115.15, 122.28 (C5), 152.64, 152.74, 158.69, 161.52 (C2), 213.41 (C2′). MS m/z (%): 273([M+1]+, 16), 272(M+, 92), 201(32), 190(90), 162(100), 115(26). Anal. C16H16O4: C, H. 4′,5′-Cyclopentene-4,8-dimethylfuro[3,2-g]coumarin (4). A mixture of 3 (510 mg, 1.87 mmol) and 1M NaOH (30 mL) was refluxed for 24 h, cooled, and acidified with 3M HCl. The precipitate was recovered by filtration and washed with water. The crude product was purified by FC using hexane/ethyl acetate (19:1) as eluent, yielding pure 4 (143 mg, 30%). Mp 153oC. IR: ©ARKAT USA, Inc ISSN 1424-6376 Page 141 Page 141 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 2922, 2850, 1706, 1558, 1480, 1396, 1125, 1092. Experimental Section The mixture was cooled, the precipitate collected, the solvent evaporated under reduced pressure, and the resulting residue purified by FC using CHCl3 as eluent, giving pure 19 (85 mg, 81.7%). Mp: 128oC. IR: 2921, 1726, 1608, 1470, 1283, 1115, 891, 766. 1H NMR (CDCl3): 2.38 (s, 3H, Me-C8), 4.72 (s, 2H, CH2O), 5.72 (d, J = 1.9 Hz, 1H, C=CH), 6.03 (d, J = 1.9 Hz, 1H, C=CH), 6.80 (d, J = 8.8 Hz, 1H, H-6), 7.49 (m, 1H, CH-CHC4), 7.78 (m, 1H + 1H, CH-CHC3 + H-5), 7.98 (d, J = 8.0 Hz, 1H, CH-C3), 8.34 (d, J = 8.0 Hz, 1H, CH-C4). 13C NMR (CDCl3): 8.95 (Me-C8), 72.33 (CH2O), 108.68 (C6), 112.36, 115.67, 118.39 (C=CH2), 120.25, 120.84 (CH-C4), 121.63 (CH- CHC3), 127.01, 128.18 (C5), 130.75 (CH-CHC4), 135.12 (CH-C3), 135.57, 150.66, 157.49, 161.78 (C2). MS m/z (%): 347 ([M+2]+, 2), 345 (M+, 12), 344 (13), 265 (60), 225 (100), 171 (17). Anal. C17H13BrO3: C, H. 3,4-Cyclohexene-5′,8-dimethylfuro[3,2-g]coumarin (18). This compound was prepared from 17 (2.5 g, 7.1 mmol) and N,N-diethylaniline (55 mL), in the same way as 9 from 8. The residue was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding 18 (1.1 g, 57.1%). 3,4-Benzo-7-(β-bromoallyloxy)-8-methylcoumarin (19). A solution of 17 (106 mg, 0.30 mmol) and DDQ (137 mg, 0.60 mmol) in toluene (15 mL) was refluxed for 5 h. The mixture was cooled, the precipitate collected, the solvent evaporated under reduced pressure, and the resulting residue purified by FC using CHCl3 as eluent, giving pure 19 (85 mg, 81.7%). Mp: 128oC. IR: 2921, 1726, 1608, 1470, 1283, 1115, 891, 766. 1H NMR (CDCl3): 2.38 (s, 3H, Me-C8), 4.72 (s, 2H, CH2O), 5.72 (d, J = 1.9 Hz, 1H, C=CH), 6.03 (d, J = 1.9 Hz, 1H, C=CH), 6.80 (d, J = 8.8 Hz, 1H, H-6), 7.49 (m, 1H, CH-CHC4), 7.78 (m, 1H + 1H, CH-CHC3 + H-5), 7.98 (d, J = 8.0 Hz, 1H, CH-C3), 8.34 (d, J = 8.0 Hz, 1H, CH-C4). 13C NMR (CDCl3): 8.95 (Me-C8), 72.33 (CH2O), 108.68 (C6), 112.36, 115.67, 118.39 (C=CH2), 120.25, 120.84 (CH-C4), 121.63 (CH- CHC3), 127.01, 128.18 (C5), 130.75 (CH-CHC4), 135.12 (CH-C3), 135.57, 150.66, 157.49, 161.78 (C2). MS m/z (%): 347 ([M+2]+, 2), 345 (M+, 12), 344 (13), 265 (60), 225 (100), 171 (17). Anal. C17H13BrO3: C, H. 3,4-Cyclohexene-5′,8-dimethylfuro[3,2-g]coumarin (18). Experimental Section The crude product was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding pure 9 (85 mg, 52.0%). Mp 213oC. IR: 3068, 2923, 1716, 1593, 1173, 1101, 807. 1H NMR (CDCl3): 2.20 (m, 2H, CH2CH2CH2), 2.48 (d, J = 1.0 Hz, 3H, Me-C5′), 2.56 (s, 3H, Me- C8), 2.91 (t, J = 7.5 Hz, 2H, CH2-C4), 3.08(t, J = 7.6 Hz, 2H, CH2-C3), 6.37 (d, J = 1.0Hz, 1H, H-4′), 7.28 (s, 1H, H-5). 13C NMR (CDCl3): 8.69 (Me-C8), 14.20 (Me-C5′), 22.44 (CH2CH2CH2), 30.54 (CH2-C4), 32.40 (CH2-C3), 102.50 (C4′), 108.92 (C8), 112.09 (C5), 114.77, 124.96, 125.25, 149.41, 154.93, 156.98, 157.12, 160.79 (C2). MS m/z (%): 255([M+1]+, 17), 254(M+, 97), 226(100), 199(48), 183(26), 153(12). Anal. C16H14O3: C, H. 7-(β-Bromoallyloxy)-3,4-cyclohexene-8-methylcoumarin (17). This compound was prepared from 13 (1.1 g, 4.8 mmol)3, K2CO3 (1.0 g, 7.2 mmol) and 2,3-dibromopropene (747 µL, 1.4 g, 7.2 mmol) in the same way as 3. The crude product was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding pure 17 (1.27 g, 75.9%). Mp: 135-136oC. IR: 3071, 2935, 1708, 1605, 1114, 755. 1H NMR (CDCl3): 1.82 (m, 4H, CH2(CH2)2CH2), 2.35 (s, 3H, Me-C8), 2.56 (m, 2H, CH2-C4), 2.72 (m, 2H, CH2-C3), 4.71 (s, 2H, CH2O), 5.76 (d, J = 1.6 Hz, 1H, C=CH), 6.01 (d, J = 1.6 Hz, 1H, C=CH), 6.75 (d, J = 8.8 Hz, 1H, H-6), 7.35 (d, J = 8.8 Hz, 1H, H-5). 13C NMR (CDCl3): 8.78 (Me-C8), 21.86 (CH2-CH2C3), 22.11 (CH2-CH2C4), 24.31 (CH2-C3), 25.65 (CH2-C4), 72.36 (CH2O), 108.32 (C6), 114.80 (C8), 115.04, 118.36 (CH2=C), 121.26, 121.35 (C5), 127.03, 147.64, 151.57, 157.47, 162.60 (C2). MS m/z (%): 350 ([M+2]+, 4), 349 (M+, 22), 269 (89), 229 (100), 201 (58), 187 (40). Anal. C17H17BrO3: C, 58.44; H, 4.87. ©ARKAT USA, Inc Page 142 Page 142 ISSN 1424-6376 ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Vincenzo Tortorella 3,4-Cyclohexene-5′,8-dimethylfuro[3,2-g]coumarin (18). This compound was prepared from 17 (2.5 g, 7.1 mmol) and N,N-diethylaniline (55 mL), in the same way as 9 from 8. The residue was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding 18 (1.1 g, 57.1%). 3,4-Benzo-7-(β-bromoallyloxy)-8-methylcoumarin (19). A solution of 17 (106 mg, 0.30 mmol) and DDQ (137 mg, 0.60 mmol) in toluene (15 mL) was refluxed for 5 h. Experimental Section Vincenzo Tortorella dose of 0.793 J cm-2. After irradiation, the medium containing the compounds was removed, and the cells were incubated in complete F-12 medium for 24 hours. dose of 0.793 J cm-2. After irradiation, the medium containing the compounds was removed, and the cells were incubated in complete F-12 medium for 24 hours. dose of 0.793 J cm-2. After irradiation, the medium containing the compounds was removed, and the cells were incubated in complete F-12 medium for 24 hours. HL-60 cells (105) were seeded into each well of a 24-well cell culture plate. After incubation for 24 hours, various concentrations of the test agents were added in complete medium. The cells were kept in the dark for 1 hour, irradiated with a UVA dose of 0.793 J cm-1 and then incubated for a further 24 hours. HL-60 cells (105) were seeded into each well of a 24-well cell culture plate. After incubation for 24 hours, various concentrations of the test agents were added in complete medium. The cells were kept in the dark for 1 hour, irradiated with a UVA dose of 0.793 J cm-1 and then incubated for a further 24 hours. In the case of the experiments carried out in the dark, the cells, both HeLa and HL-60 were seeded (105) and incubated for 24 hours. Then the test agents were added and the cells were incubated for a further 24 hours. A trypan blue assay was performed to determine cell viability. Cytotoxicity data were expressed as IC50 values, i.e., the concentrations of the test agent inducing 50% reduction in cell numbers compared with control cultures. Skin phototoxicity. Skin phototoxicity was tested on depilated albino guinea pigs (outbred Dunkin-Hartley strain), as previously reported.27 An ethanol solution of each new compound was applied topically to the skin up to 50 µg/cm2. For 8-MOP the concentration used was 10 µg/cm2. The animals were then kept in the dark for 45 min and the treated skin was irradiated with 20 kJ m-2 of UVA; erythema was scored after 48 h. Nucleic acid. Salmon testes DNA was purchased from Sigma Chemical Company (Cat. D- 1626). Its hypochromicity, determined according to Marmur and Doty,28 was over 35%. The DNA concentration was determined using extinction coefficient 6600 M-1cm-1 at 260 nm. Preparation of adducts. Experimental Section This compound was prepared from 17 (2.5 g, 7.1 mmol) and N,N-diethylaniline (55 mL), in the same way as 9 from 8. The residue was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding 18 (1.1 g, 57.1%). 3,4-Benzo-5′,8-dimethylfuro[3,2-g]coumarin (20). This compound was prepared from 19 (1.0 g, 2.9 mmol) and N,N-diethylaniline, in the same way as 9 from 8. The crude product was purified by FC using hexane/ethyl acetate (6:1) as eluent, yielding pure 20 (397 mg, 52.0%). Mp: 230-231oC. IR: 2925, 1722, 1610, 1440, 1115, 850. 1H NMR (CDCl3): 2.45 (d, J = 1.2 Hz, 3H, Me-C5′), 2.55 (s, 3H, Me-C8), 6.60 (d, J = 1.2 Hz, 1H, H-4′), 7.47 (m, 1H, CH-CHC3), 7.74 (m, 2H, CH-CHC4), 7.94 (s, 1H, H-5), 8.11 (d, J = 8.0 Hz, 1H, CH-C4), 8.32 (d, J = 8.0 Hz, 1H, CH-C3). 13C NMR (CDCl3): 8.56 (Me-C8), 14.01 (Me-C5′), 101.80 (C4′), 110.00 (C5), 110.10 (C8), 113.65 (C4a), 120.87 (C6), 121.40 (CH-C4), 124.62 (C3), 127.88 (CH-CHC3), 130.35 (CH-C3), 134.50 (CH-CHC4), 135.00 (C4), 147.10 (C8a), 154.98 (C7), 156.80 (C5′), 161.47 (C2). MS m/z (%): 265 ([M+1]+, 15), 264 (M+, 100), 235 (19), 207 (10), 178 (17), 152 (10). Anal. C17H12O3: C, H. Photobiological methods. Cell cultures. HL-60 and HeLa cells were grown in RPMI 1640 (Sigma Chemical Co.) supplemented with 15% heat-inactivated fetal calf serum (Biological Industries) and Nutrient Mixture F-12 [HAM] (Sigma Chemical Co.) supplemented with 10% heat-inactivated fetal calf serum (Biological Industries), respectively. 100 U/mL penicillin, 100 µg/mL streptomycin and 0.25 µg/mL amphotericin B (Sigma Chemical Co.) were added to both media. The cells were cultured at 37°C in a moist atmosphere of 5% carbon dioxide in air. Irradiation procedure. Irradiations were performed by means of Philips HPW 125 lamps equipped with a Philips filter emitting over 90% at 365 nm. Irradiation intensity was checked on a UV-X radiometer (Ultraviolet Products Inc., Cambridge, UK) for each experimental procedure. Inhibition growth assays. HeLa cells (105) were seeded into each well of a 24-well cell culture plate. After incubation for 24 h, the medium was replaced with an equal volume of Dulbecco′s modified Eagle medium (DMEM, Sigma Chemical Co.) without phenol red, and various concentrations of the test agent were added. One hour later the cells were irradiated with a UVA ©ARKAT USA, Inc ISSN 1424-6376 Page 143 ISSN 1424-6376 ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Experimental Section Data were expressed in terms of non-renaturated fraction of irradiated compound-DNA complex relative to irradiated DNA, as suggested by Blais et al.29 In detail: ( ) ( ) DNA native R DNA modified R x 100 % N N A A A A NRF − − = where AN and AR are, respectively, the optical densities measured at 260 nm of DNA before heat denaturation and after renaturation. The calculated NRF% values were expressed as log. where AN and AR are, respectively, the optical densities measured at 260 nm of DNA before heat denaturation and after renaturation. The calculated NRF% values were expressed as log. Acknowledgments The authors are grateful to Mr. Mariano Schiavon for his skilful technical assistance. Experimental Section Volumes of concentrated solutions of the examined compound were added to salmon testes DNA in ETN solution (1.5x10-3M) to achieve a DNA/compound ratio of about 80. The mixture was irradiated in a glass dish with four Philips HPW 125 lamps, arranged two above and two below the dish, at a distance of 7 cm, for 120 min at room temperature. After irradiation the DNA was precipitated with NaCl (up to 1 M concentration) and cool ethanol (2 volumes), the precipitated DNA was collected, washed with 80% ethanol, dried and then dissolved in a measured volume of buffer. The final solution was made 0.5 N with HCl, heated at 100 °C for 2 h, neutralised and extracted exhaustively with CHCl3. After this procedure the organic layers were collected, dried under high vacuum and dissolved in ethanol and the adduct was separated on TLC plates and eluted with 100% ethanol. UV spectra were recorded on a Perkin Elmer model Lambda 5 spectrophotometer. The 1H and 13C assignment was obtained by utilising HMQC and HMBC spectra on a Bruker Avance DMX600 instrument. Mass spectrometry measurements were performed on a ElectroSpray Ionization (ESI) Time of Flight (ToF) instrument (mod. Mariner, Perseptive-Biosystem) by dissolving the samples in water/acetonitrile/formic acid (50:49:1) solution. Photoreversal of adducts. Ethanol solution of the adduct (ca. 20 µg/mL) was irradiated in quartz cuvettes with a mineral lamp (254 nm). The photosplitting reaction was followed spectrophotometrically. Page 144 ©ARKAT USA, Inc ISSN 1424-6376 Issue in Honor of Prof. Vincenzo Tortorella ARKIVOC 2004 (v) 131-146 Evaluation of interstrand cross-links in vitro. Evaluation of cross-links was carried out by measuring the renaturation capacity of cross-linked double helix after thermal denaturation. Aliquots of aqueous solutions of salmon testes DNA (Sigma Co.), alone and at [nucleic acid]/[drug]=75, were introduced into calibrated glass tubes, immersed in a thermostatically controlled bath, and then irradiated for various periods of time. After irradiation the samples were thermally denatured (95°C for 15 min) and quickly cooled in ice. The renaturation capacity of DNA, due to cross-link formation, was investigated by recording absorbance at 260 nm. References 1. Parrish, J. A.; Fitzpatrick, T. B.; Tannenbaum, L.; Pathak, M.A. N. Engl. J. Med. 1974, 291, 1207. 1. Parrish, J. A.; Fitzpatrick, T. B.; Tannenbaum, L.; Pathak, M.A. N. Engl. J. Med. 1974, 291, 1207. 2. Lowe, N. J.; Chizhevsky, V.; Gabriel, H. Clin. Dermatol. 1997, 15, 745. 3. Bethea, D.; Fullmer, B.; Syed, S.; Seltzer, G.; Tiano, J.; Rischko, C.; Gillespie, L.; Brown, D.; Gasparro, F. P. J. Dermatol. Sci. 1999, 19, 78. 3. Bethea, D.; Fullmer, B.; Syed, S.; Seltzer, G.; Tiano, J.; Rischko, C.; Gillespie, L.; Brown, D.; Gasparro, F. P. J. Dermatol. Sci. 1999, 19, 78. 4. Edelson, R. L. Am. N. Y. Acad. Sci. 1991, 636, 154. 4. Edelson, R. L. Am. N. Y. Acad. Sci. 1991, 636, 154. 5. Van Iperen, H. P.; Beijersbergen van Henegouwen G. M. J. J. Photochem. Photobiol B: Biol. 1997, 39, 99. 5. Van Iperen, H. P.; Beijersbergen van Henegouwen G. M. J. J. Photochem. Photobiol B: Biol. 1997, 39, 99. 6. McNeely, W.; Goa, K. L. Drugs 1998, 56, 667. 7. Pathak, M. A.; Fitzpatrick, T. B. J. Photochem. Photobiol. B. 1992, 14, 3. 8. Song, P.-S.; Tapley, K. J. Jr. Photochem. Photobiol. 1979, 29, 1177. 9. Rodighiero, P.; Palumbo, M.; Marciani Magno, S.; Manzini, P.; Gia, O.; Piro, R.; Guiotto, A. J. Heterocycl. Chem. 1986, 44, 1405. 10. Palumbo, M.; Rodighiero, P.; Gia, O.; Guiotto, A.; Marciani Magno, S. Photochem. Photobiol. 1986, 44, 1. 11. Gia, O.; Mobilio, S.; Palumbo, M.; Pathak, M. A. Photochem. Photobiol. 1993, ISSN 1424-6376 ©ARKAT USA, Inc Page 145 ©ARKAT USA, Inc ARKIVOC 2004 (v) 131-146 Issue in Honor of Prof. Vincenzo Tortorella 12. Gia, O.; Conconi, M. T.; Uriarte, E. Farmaco. 1997, 5, 289. 13. Chilin, A.; Marzano, C.; Guiotto, A.; Manzini, P.; Baccichetti, F.; Carlassare, F.; Bordin, F. J. Med. Chem. 1999, 42, 2936. 14. Dalla Via, L.; Gia, O.; Marciani Magno, S.; Santana, L.; Teijera, M.; Uriarte E. J. Med. Chem. 1999, 42, 4405. 15. Dalla Via, L.; Uriarte, E.; Quezada, E.; Dolmella, A.; Ferlin, M. G.; Gia, O. J. Med. Chem. 2003, 46, 3800. 16. Dalla Via, L.; Gia, O.; Viola, G.; Bertoloni, G.; Santana, L.; Uriarte, E. Farmaco. 1998, 53, 638. 17. Dall′Acqua, F.; Marciani, S.; Vedaldi, D.; Rodighiero, G. Biochim. Biophys. Acta 1974, 353, 267. 18. Palumbo, M.; Baccichetti, F.; Antonello, C.; Gia, O.; Capozzi, A.; Marciani Magno, S. Photochem. Photobiol. 1990, 52, 533. 19. 29. Blais, J.; Vigny, P.; Moron, J.; Bisagni, E. Photochem. Photobiol. 1984, 39, 145. References Gia, O.; Uriarte, E.; Zagotto, G.; Baccichetti, F.; Antonello, C.; Marciani Magno, S. J. Photochem. Photobiol. B: Biol. 1992, 14, 95. 20. Garazd, M. M.; Garazd, Ya. L.; Shilin, S. V.; Panteleimonova, T. N.; Khilya, V. P. Chem. Nat. Compd. 2002, 38, 230. 21. Confalone, P. N.; Confalone, D. L. Tetrahedron 1983, 39, 1265. 22. Garazd, Ya. L.; Ogorodniichuk, A. S.; Garazd, M. M.; Khilya, V. P. Chem. Nat. Compd. 2002, 38, 424. 23. Dall′Acqua, F.; Terbojevich, M.; Marciani S.; Vedaldi D.; Recher M. Chem.-Biol. Interact. 1978, 21, 103. 24. Caffieri, S.; Rodighiero, P.; Vedaldi, D.; Dall′Acqua, F. Photochem. Photobiol. 1985, 42, 361. 25. Dall′Acqua, F.; Vedaldi, D.; Baccichetti, F.; Rodighiero, G.; Gennaro, A. QSAR In Design Of Bioactive Compounds; M. Kuchar: Barcelona, 1984. 26. Miranda, R.; Santana, L.; Uriarte, E.; Zagotto, G. Spectrochim. Acta 1994, 50A, 27. Gia, O.; Anselmo, A; Conconi, M. T.; Antonello, C.; Uriarte, E.; Caffieri, S. J. Med. Chem. 1996, 39, 4489. 28. Marmur, J.; Doty, P. J. Mol. Biol. 1962, 5, 109. 29. Blais, J.; Vigny, P.; Moron, J.; Bisagni, E. Photochem. Photobiol. 1984, 39, 145. ©ARKAT USA, Inc ISSN 1424-6376 Page 146
https://openalex.org/W4210684762	https://journals.plos.org/plosone/article/file?type=printable&id=10.1371/journal.pone.0078218	English	null	Correction: Catheter-Based Renal Sympathetic Denervation Significantly Inhibits Atrial Fibrillation Induced by Electrical Stimulation of the Left Stellate Ganglion and Rapid Atrial Pacing	PloS one	2,013	cc-by	7,327	Abstract Funding: This study was supported by the International Science and Technology Cooperation Project of the Ministry of Science and Technology of China (2011DFA32860) and the Shanghai Science and Technology Commission Research Projects (10411956900). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Funding: This study was supported by the International Science and Technology Cooperation Project of the Ministry of Science (2011DFA32860) and the Shanghai Science and Technology Commission Research Projects (10411956900). The funders had no collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * E-mail: zhouqina_5217@sina.com * E-mail: zhouqina_5217@sina.com . These authors contributed equally to this work. . These authors contributed equally to this work. interventional therapy for patients with resistant hypertension [7–10]. Recently, several studies have demonstrated that the efficacy of RSD on AF. Linz et al. [11] reported that RSD reduced blood pressure, atrial effective refractory period (ERP) shortening and inducibility of vagally mediated atrial fibrillation in a model of obstructive sleep apnea. Furthermore, they also showed that electrical baroreflex stimulation significantly short- ened atrial refractoriness, thereby causing increased AF induc- ibility. In contrast, RSD did not significantly modulate atrial refractoriness or AFinducibility [12]. Zhao et al. [13] showed that RSD could decrease episodes of AF during short-time rapid atrial pacing, which might correlate with decreased activity of Yuemei Hou1., Jialu Hu2., Sunny S. Po3, Huan Wang4, Ling Zhang4, Feng Zhang4, Kun Wang4, Qina Zhou4* , Jialu Hu2., Sunny S. Po3, Huan Wang4, Ling Zhang4, Feng Zhang4, Kun Wang4, 1 Department of Cardiovascular Diseases, The 6th People’s Hospital affiliated to Shanghai Jiaotong University, Shanghai, China, 2 Department of Cardiovascular Diseases, Zhongshan Hospital Fudan University, Shanghai, China, 3 Heart Rhythm Institute and Department of Medicine, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, United States of America, 4 Arrhythmia Research Lab, First Affiliated Hospital of Xinjiang Medical University, Urumqi, China Abstract Background: Sympathetic activity involves the pathogenesis of atrial fibrillation (AF). Renal sympathetic denervation (RSD) decreases sympathetic renal afferent nerve activity, leading to decreased central sympathetic drive. The aim of this study was to identify the effects of RSD on AF inducibility induced by hyper-sympathetic activity in a canine model. Methods: To establish a hyper-sympathetic tone canine model of AF, sixteen dogs were subjected to stimulation of left stellate ganglion (LSG) and rapid atrial pacing (RAP) for 3 hours. Then animals in the RSD group (n = 8) underwent radiofrequency ablation of the renal sympathetic nerve. The control group (n = 8) underwent the same procedure except for ablation. AF inducibility, effective refractory period (ERP), ERP dispersion, heart rate variability and plasma norepinephrine levels were measured at baseline, after stimulation and after ablation. Results: LSG stimulation combined RAP significantly induced higher AF induction rate, shorter ERP, larger ERP dispersion at all sites examined and higher plasma norepinephrine levels (P,0.05 in all values), compared to baseline. The increased AF induction rate, shortened ERP, increased ERP dispersion and elevated plasma norepinephrine levels can be almost reversed by RSD, compared to the control group (P,0.05). LSG stimulation combined RAP markedly shortened RR-interval and standard deviation of all RR-intervals (SDNN), Low-frequency (LF), high-frequency (HF) and LF/HF ratio (P,0.05). These changes can be reversed by RSD, compared to the control group (P,0.05). Conclusions: RSD significantly reduced AF inducibility and reversed the atrial electrophysiological changes induced by hyper-sympathetic activity. Citation: Hou Y, Hu J, Po SS, Wang H, Zhang L, et al. (2013) Catheter-Based Renal Sympathetic Denervation Significantly Inhibits Atrial Fibrillation Induced by Electrical Stimulation of the Left Stellate Ganglion and Rapid Atrial Pacing. PLoS ONE 8(11): e78218. doi:10.1371/journal.pone.0078218 Editor: German E. Gonzalez, University of Buenos Aires, Faculty of Medicine. Cardiovascular Pathophysiology Institute, Argentina Received March 10, 2013; Accepted September 10, 2013; Published November 6, 2013 Copyright: 2013 Hou et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2013 Hou et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Introduction Atrial fibrillation (AF) is a complex arrhythmia with multiple mechanisms. Electrical, structural and neural remodelling contribute to the AF substrate. Sympathetic activity constitutes an important factor in the pathogenesis of AF [1–5]. Strategies aiming at reducing sympathetic activity potentially protect against the initiation and maintenance of AF. Ablation of bilateral stellate and thoracic sympathetic ganglia can signifi- cantly reduced paroxysmal atrial tachyarrhythmia episodes induced by sympathetic discharges in dogs [6]. Catheter-based renal sympathetic denervation (RSD) is an effective and November 2013 \| Volume 8 \| Issue 11 \| e78218 1 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org Renal Denervation Inhibits Atrial Fibrillation Renal Denervation Inhibits Atrial Fibrillation Figure 1. Representative images showing a stellate ganglion and renal hilum. A shows a star-shaped stellate ganglion behind the subclavian artery. B shows the renal hilum before dissection. The renal artery and renal sympathetic nerve enter the kidney through the renal hilum. doi:10.1371/journal.pone.0078218.g001 Figure 1. Representative images showing a stellate ganglion and renal hilum. A shows a star-shaped stellate ganglion behind the subclavian artery. B shows the renal hilum before dissection. The renal artery and renal sympathetic nerve enter the kidney through the renal hilum. doi:10.1371/journal.pone.0078218.g001 Figure 1. Representative images showing a stellate ganglion and renal hilum. A shows a star-shaped stellate ganglion behind the subclavian artery. B shows the renal hilum before dissection. The renal artery and renal sympathetic nerve enter the kidney through the renal hilum. doi:10 1371/journal pone 0078218 g001 Rapid Atrial Pacing (RAP) The right femoral arteries were used for recording of blood pressure (via a pressure transducer). The right external jugular vein was cannulated for catheter insertion into the right atrium (RA) to record right atrial potentials and to induce rapid atrial pacing. A left-sided thoracotomy was performed at the fourth intercostal space. Multi-electrode catheters were sutured to the left atrial appendage (LAA), left superior pulmonary vein (LSPV) and left inferior pulmonary vein (LIPV) to record pulmonary vein (PV) and atrial potentials. Continuous RAP in all the dogs was delivered (600 bpm, twice-threshold current, 0.5 ms in duration) at the right atrium (RA) for 3 hours. Ethics Statement This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee of the First Affiliated Hospital of Xinjiang Medical University (Permit Number: IACUC-20110325009), and con- formed to the guidelines of the Association for Assessment and Accreditation of Laboratory Care (AAALAC). All surgery was performed under sodium pentovarbital anesthesia, and all efforts were made to minimize suffering [14]. A Canine Model of Acute Atrial Fibrillation Induced by the Left Stellate Ganglion Electrical Stimulation Combined with Rapid Atrial Pacing both renin–angiotensin–aldosterone system (RAAS) and renal sympathetic nerve. However, the AF models in those three studies were all associated with increased vagal activity. The effects of RSD on atrial electrophysiological and AF inducibility remains controversial and the mechanisms underlying the effects remains unclear. Whether RSD could reduce AF inducibility induced by hyper-sympathetic activity is unknown. Electrical stimulation of the left stellate ganglion (LSG) combined rapid atrial pacing (RAP) for 3 hours was used to establish a canine model of acute AF mediated by hyper- sympathetic activity. In the present study, we developed an animal model by delivering rapid atrial pacing in the presence of a hyper- sympathetic tone induced by stimulation of left stellate ganglion (LSG) to evaluate the effects of RSD on AF inducibility, atrial electrophysiological changes and cardiac autonomic activity. The Electrical Stimulation of LSG Anaesthesia was performed as described above. Vertical paramedian incision was made in the supraclavicular fossa. Behind the subclavian artery and vertebral artery, and in the adipose tissue in front of the seventh cervical vertebra, a star- shaped SG was visible on the left side (Figure 1A). The adipose tissue surrounding the SG was bluntly dissected with a glass dissecting needle to expose its branch and the cardiac sympathetic nerve. The LSG was then stimulated by GRASS S88 Nerve and Muscle Stimulator (Astro-Med Inc, USA) at a gradual level of 2 V to 10 V (20 Hz, 2-ms pulse width) for a period of 30 seconds. The stimulation threshold of SG is defined as the current required to produce a rise of 20% or more in systolic blood pressure (SBP) or heart rate [15], once the stimulation threshold was established, the LSG was then continuously stimulated (20 Hz, 2-ms pulse width, threshold voltage) for 3 hours. Animal and Groups Setting Sixteen adult mongrel dogs weighing 18 to 22 kg were anesthetized with sodium pentobarbital (20 mg/kg) and ventilated with room air by a positive pressure respirator. Ketamine (2 mg/ kg) was used for induction of general anesthesia and analgesic. Core body temperature was maintained at 36.561.5uC. Standard ECG leads were continuously recorded to determine heart rate and rhythm. Sixteen dogs were randomly divided into two groups. The control group (n = 8) underwent rapid atrial pacing (RAP) and the left satellite ganglion (LSG) sitimulatioin for 3 hours but without renal sympathetic nerve ablation. The renal sympathetic denervation (RSD) group (n = 8) underwent RAP and LSG sitimulatioin for 3 hours and followed by catheter-based radiofre- quency ablation of the renal sympathetic nerve. Electrophysiological Measurements Electrophysiological measurements were made every one hour immediately after RAP and LSG stimulation and after cather- based RSD and sham ablation. Electrophysiological recordings including effective refractory period (ERP) and ERP dispersion were measured before pacing and stimulation as baseline values, after pacing and stimulating for 3 hours (stimulation) and after ablation of LSG or bilateral renal sympathetic nerve. Total duration of reading was 10 minutes. Results RSD Significantly Reduced Systolic Blood Pressure (SBP) but not Heart Rate (HR) Catheter -based Renal Sympathetic Denervation Blood samples were collected from the femoral artery into a tube containing EDTA, and immediately centrifuged at 3000 rpm for 10 min at 4uC, and then finally stored at 280uC till analysis.The plasma norepinephrine level was determined by high-performance liquid chromatography (HPLC) with YWG- C18 column and electrochemical detection (Waters 2465, Milford, MA) [17]. y p After completion of all the above stimulation and measurement, laparotomy was performed through a midline incision. A peritoneal window was created to expose the renal hilum (Figure 1B). Steerable 6 F radiofrequency catheter (HL-75F, Sichuan, P. R. China) was introduced into the right and left renal artery from the femoral artery. The tip of the catheter was positioned under direct vision to make sure that it was accurately introduced into the renal artery on the both sides. In the control group, after the catheter was accurately introduced into the renal artery on the both sides, no further procedures were performed for 30 minutes (sham ablation). In the RSD group, we evaluated the effects of the renal sympathetic nerve by electrical stimulation (5 V, 20 Hz, 2-ms, Grass S88 nerve stimulator, USA) at the proximal renal artery for 30 seconds before and after bliateral RSD. Each discrete radiofrequency ablation (of approximately 6 to 8 watts) lasted 60 seconds, and total four ablations were performed both longitudinally and circumferentially within each renal artery. The left RSD was performed at first, followed by 4 ablations of the right RSD immediately. The electrical stimulation and ablation of the renal sympathetic nerve lasted for approx- imately 30 minutes. Statistical Analyses Qualitative data were expressed as a ratio and measurement data were expressed as means with SD. One-Sample Kolmogorov- Smirnov Test was used to test the normality of the measurement data. ANOVA for repeated measures was used to compare the changes at baseline, after stimulation and ablation. LSD was used for Post Hoc multiple comparisons. The chi-square test was used to compare the AF induction rate. Values of P,0.05 were considered statistically significant. Programmed stimulation at the atrial and PV sites was performed using a cardiac programmable stimulator (Lead-2000 EP CONTROL, Sichuan, P. R. China). The ERP was defined as the longest S1S2 interval that failed to produce a response. It was measured at an atrial pacing cycle length of 300 ms and the S1–S2 intervals were decreased from 200 ms to refractoriness initially by decrements of 5 ms (S1:S2 = 8:1, twice-threshold current, 0.5 ms in duration). ERP dispersion was defined as the coefficient of variation (standard deviation/mean) of the ERP at all 4 sites (LSPV, LIPV, LAA and RA). Histology Non-ablated renal ganglion in the segments which were dissected from the renal arteries with their perivascular tissue in the 8 control group dogs for histological control. After the completion of all the electrophysiological measurement after the ablation, the renal arteries with their perivascular tissues from the RSD group and control group were immediately dissected and fixed in formaldehyde. Two months later, hematoxylin-eosin (HE) and silver staining and immunohistochemical staining were performed. Multiple tissue blocks were sampled from the sites of the renal artery with or without ablation. The sections were stained with routine staining for structural examination. The slides were also silver stained and immunostained with antibodies to tyrosine hydroxylase (TH) for the sympathetic nerves, according to methods described previously [18,19]. The primary antibodies used in this study were sheep poly-clonal anti-TH (1:50 dilution; Thermo Scientific, USA, MJ1473853,). The secondary antibody was polink-2 plus polymer HRP detection system (For Goat Primary Antibody, GBI, USA). The silver staining was used for staining nerve fibers and nerve endings [20]. Light microscopy was used to examine the sections. Measurement of AF Induction Rate To calculate mean AF induction rate, AF was induced 6 times with S1S1 stimuli at 600 bpm (2 ms in duration, fourfold threshold current). AF induction rate was defined as (the relative ratio of successful induction frequency to total frequency of stimulation)6100%. AF was defined as irregular atrial rates .500 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org November 2013 \| Volume 8 \| Issue 11 \| e78218 2 Renal Denervation Inhibits Atrial Fibrillation mainly by sympathetic innervation of the heart; and LF/HF ratio reflects sympathovagal balance. beats/min and a duration .5 seconds associated with irregular atrioventricular conduction [16]. AF lasting .30 minutes was considered sustained. RSD Reduced AF Induction Rate When compared with those in the baseline condition, LSG stimulation with RAP significantly increased the AF induction rate at the LAA, RA, LSPV and LIPV sites by 49.98%, 47.92%, 50.00%, 43.75% in the control group, and by 41.67%, 52.08%, 50.04%, 40.09% in the RSD group, respectively. No significant difference in AF induction rate was found under baseline and after stimulation between the control group and RSD group. After ablation, RSD markedly decreased AF induction rate at the LAA, RA, LSPV and LIPV sites by 56.17%,36.83%, 41.84%, 41.92%, respectively compared to the control group (P,0.05), (Fig. 2). Effects of RSD on ERP with that in the LSG stimulation condition and control group [the control group (150.98619.66) mmHg verse the RSD group (114.74623.29) mmHg, P,0.05; within RSD group: stimulation (167.36629.31) mmHg verse ablation (114.74623.29) mmHg, P,0.05; Table 1). When compared with that at the respective baselines, LSG stimulation with RAP in 3 hours induced a pronounced ERP shortening at the LAA, RA, LSPV and LIPV sites by 14.67 ms, 19.17 ms, 18.14 ms, 21.52 ms in the control group, respectively (Fig. 3; P,0.05), and by 14.61 ms, 18.11 ms, 17.86 ms, 23.31 ms in the RSD group, respectively (Fig. 3; P,0.05). No significant difference in ERP was found at each site between the control group and RSD group under baseline and stimulation conditions. However, the ERP shortening can be reversed by RSD in the RSD group, when compared with that in the control group (Fig. 3; P,0.05). Analysis of Heart Rate Variability (HRV) PowerLab (ML866/P, FE132 Bio Amp, ADInstruments, Australia) was used for ECG measurement. Three self-adhesive ECG electrodes are administered for recording. All signals were then analyzed using the HRV Module of LabChart \ro V7 software for HRV time domain analysis and frequency domain analysis. To examine effects of denervation on HR, we compared HR between stimulation and denervation, between the control group and RSD group. We found that LSG stimulation induced mild elevation of HR in both the control group and RSD group when compared with that without simulation, however, no statistical significance was seen (P.0.05). By contrast, RSD induced mild reduction of HR when compared with that in LSG stimulation condition and the control group(P.0.05). HRV measurements included the following cardiovascular parameters: range of duration of RR-intervals (milliseconds, ms); SDNN: standard deviation of all RR-intervals(ms), which reflects overall variation in the heart beat series; pNN50: number of successive difference of intervals which differ by more than 50 ms, as a proportion of total beat cycles, which sensitively reflects the parasympathetic activity. High-frequency (HF) components (0.15– 0.4 Hz) are driven mainly via parasympathetic innervation of the heart; Low-frequency (LF) components (0.04–0.15 Hz) are driven LSG stimulation significantly elevated SBP by approximately 1- fold in both the control group and RSD group when compared with that without LSG stimulation (the control group: baseline: 142.17625.45 mmHg verse stimulation: 165.68634.59 mmHg, P,0.05; the RSD group: baseline: 139.02633.41 mmHg verse stimulation: 167.36629.31 mmHg, P,0.05; Table 1). Whereas, RSD significantly reduced the SBP in RSD group when compared PLOS ONE \| www.plosone.org November 2013 \| Volume 8 \| Issue 11 \| e78218 3 Renal Denervation Inhibits Atrial Fibrillation Table 1. Effect of RSD on HR and SBP(mean+/2standard deviation). Control Group RSD Group Baseline Stimulation Ablation Baseline Stimulation Ablation HR(bpm) 158.43639.72 163.82643.04 160.54638.61 159.90631.94 167.22641.06 153.24636.72 SBP(mmHg) 142.17625.45 165.68634.59* 150.98619.66 139.02633.41 167.36629.31* 114.74623.29#{ Note: RSD = Renal sympathetic denervation, HR = Heart rate, SBP = Systolic blood pressure. P,0.05 indicated significance between stimulation and baseline conditions, #P,0.05 indicated significance between ablation and stimulation conditions. {P,0.05 indicated significance between the control group and RSD group. doi:10.1371/journal.pone.0078218.t001 RSD Reduced ERP Dispersion When compared with that of the respective baselines, LSG stimulation with RAP in 3 hrs markedly increased ERP dispersion within the 4 sites in both the control group and RSD group [(the Figure 2. Changes in AF induction rate at RA, LA, LSPV and LIPV sites. RA = the right atrium, LA = the left atrium, LSPV = the left superior pulmonary vein, LIPV = the left inferior pulmonary vein, RSD = the renal sympathetic denervation. P,0.05 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g002 Figure 2. Changes in AF induction rate at RA, LA, LSPV and LIPV sites. RA = the right atrium, LA = the left atrium, LSPV = the left superior pulmonary vein, LIPV = the left inferior pulmonary vein, RSD = the renal sympathetic denervation. P,0.05 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g002 November 2013 \| Volume 8 \| Issue 11 \| e78218 November 2013 \| Volume 8 \| Issue 11 \| e78218 PLOS ONE \| www.plosone.org 4 Renal Denervation Inhibits Atrial Fibrillation Figure 3. Changes in ERP at different sites. RA = the right atrium, LA = the left atrium, LSPV = the left superior pulmonary vein, LIPV = the left inferior pulmonary vein, RSD = the renal sympathetic denervation. P,0.05 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g003 Figure 3. Changes in ERP at different sites. RA = the right atrium, LA = the left atrium, LSPV = the left superior pulmonary vein, LIPV = the left inferior pulmonary vein, RSD = the renal sympathetic denervation. P,0.05 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g003 Effects of RSD on Time and Frequency Domain HRV Parameters control group: baseline: 5.89 ms verse stimulation 15.25 ms, P,0.05; the RSD group: baseline: 7.80 ms verse stimulation 14.78 ms, P,0.05); Fig. 4]. No significant difference in ERP dispersion was found between the control group and RSD group under baseline and stimulation conditions. An increase in ERP dispersion can be reversed by RSD in the RSD group, compared with that in the control group [14.58 ms verse 5.97 ms, P,0.05); Fig. 4]. Time domain parameters: When compared with the respective baseline, the RR-interval, SDNN and pNN50 were significantly shortened by LSG stimulation with RAP for 3 hours respectively by 27.6%, 55.8%, 60.8% in the control group, and by 31.7%, 48.7%, 53.3% in the RSD group, respectively (P,0.05, Table 2). These shortening can be completely reversed by RSD in the RSD group, when compared with those under LSG stimulation conditions and compared with those in the control group (P,0.05, Table 2). Figure 4. Changes in ERP dispersion (dERP) within different sites. RA = the right atrium, LA = the left atrium, LSPV = the left superior pulmonary vein, LIPV = the left inferior pulmonary vein, RSD = the renal sympathetic denervation. P,0.05 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g004 Frequency domain parameters: When compared with the respective baseline, the LF, HF and LF/HF ratio were significantly increased by LSG stimulation with RAP for 3 hours by 2.43-, 1.39- , 0.45-fold in the control group, and by 2.27-, 1.34-, 0.40- fold in the RSD group, respectively (P,0.05, Table 2). These increase induced by LSG stimulation was completely reversed by RSD in the RSD group, when compared with those in the control group (P,0.05, Table 2). Major Findings In the present study, AF inducibility was facilitated and atrial electrophysiological changes (shortening of ERP, increasing of ERP dispersion) were enhanced by LSG stimulation and RAP. Cardiac hyper-sympathetic activity was manifested by shortened RR-interval, increased LF, HF and LF/HF ratio and elevated plasma norepinephrine levels after LSG stimulation. RSD reversed the altered cardiac sympathetic activity as well as AF inducibility and electrophysiological changes caused by LSG stimulation and RAP. To compare effects of ablation on histological structures of the renal artery and renal nerves, we performed HE staining in the sections of the renal artery of the rats with or without ablation. HE staining of cells was uniformly arranged around the wall of normal renal arteries without ablation in the control group (Fig. 7A). However, after ablation in the RSD group, the ablated renal artery tissue was replaced by dense scar tissue that stained purple Figure 5. Plasma norepinephrine level. RSD = the renal sympa- thetic denervation. #P,0.001 indicated significant difference between stimulation and baseline conditions. {P,0.001 indicated significant difference between ablation and stimulation conditions. P,0.001 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g005 Plasma Norepinephrine Level The plasma norepinephrine levels were significantly elevated by LSG stimulation with RAP for 3 hours, by 6.38 -fold in the control group and by 8.4-fold in the RSD group, respectively (P,0.001, Fig. 5). No significant difference in the plasma norepinephrine level was found between the control group and RSD group under baseline and stimulation conditions. The elevated plasma norepi- nephrine levels were significantly reduced by RSD in the RSD group, when compared with that in the control group [(2099.51694.31)ng/L verse (605.34699.23)ng/L, P,0.001); Fig. 5]. Figure 4. Changes in ERP dispersion (dERP) within different sites. RA = the right atrium, LA = the left atrium, LSPV = the left superior pulmonary vein, LIPV = the left inferior pulmonary vein, RSD = the renal sympathetic denervation. P,0.05 indicated significant difference between the control group and the renal sympathetic denervation group. g p doi:10.1371/journal.pone.0078218.g004 g p doi:10.1371/journal.pone.0078218.g004 November 2013 \| Volume 8 \| Issue 11 \| e78218 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 5 Renal Denervation Inhibits Atrial Fibrillation ble 2. Effects of RSD on time and frequency domain HRV parameters (mean+/2standard deviation). Control Group RSD Group Baseline Stimulation Ablation Baseline Stimulation Ablation RR-Interval(ms) 364.25669.35 263.27644.05* 284.74637.82 359.28640.56 244.51625.40* 380.24669.47#{ SDNN(ms) 94.31611.19 41.6869.93* 43.5767.30 98.3667.98 50.4168.55* 106.34610.02#{ pNN50 34.4864.58 13.5163.06* 10.4664.81 33.3063.19 15.5464.02* 25.5162.93#{ LF(ms2) 253.566106.52 871.706233.28* 790.786101.49 261.52697.71 855.286311.06* 290.786221.49#{ HF(ms2) 408.626179.01 967.096363.51* 933.026172.63 435.736173.55 1017.546431.32* 410.026272.63#{ LF/HF 0.6260.38 0.9060.41* 0.8560.15 0.6060.47 0.8460.35* 0.6060.29#{ Note: RSD = Renal sympathetic denervation. P,0.05 indicated significant difference between stimulation and baseline conditions, #P,0.05 indicated significant difference between ablation and stimulation conditions, {P,0.05 indicated significance between the control group and RSD group. doi:10.1371/journal.pone.0078218.t002 Histological Studies and it seemed that base membrane of the artery wall was disrupted and disorganized and become loose (Fig. 7C). In addition, the ganglionic cells after the ablation morphologically became contracted and characterized by vacuolar degeneration (Fig. 7D), when compared with that in non-ablated renal ganglion in the control group (Fig. 7B). To verify whether radiofrequency ablation of the renal sympathetic nerve was successfully made, we performed HE staining and silver staining on the renal arteries. Silver staining of the nerve innervating the renal artery without ablation in the control group showed the nerve fiber distribution and normal structure of ganglionic cells, (Fig. 6A, 6B), and immunostaining results showed that TH-positive staining of renal sympathetic nerves without ablation (Fig. 6C). Brown structures indicated positively stained nerve structures. Intact (unablated) ganglion cells were observed in the middle of the section. November 2013 \| Volume 8 \| Issue 11 \| e78218 The Role of Sympathetic Nerve Activity in AF and Acute Atrial Electrophysiological Changes doi:10.1371/journal.pone.0078218.g006 action potential duration (APD) and ERP, increases ERP dispersion, induces early after depolarisations (EADs), decreases wavelet re-entry and increases spatial heterogeneity of atrial electrical activity. atrial electrophysiological changes. Therefore, in the present study, we used this acute AF model induced by LSG stimulation combined with RAP for 3 hours to observe the effects of RSD on AF inducibility and cardiac autonomic nerve activity. The indirect effect of LSG stimulation or sympathetic denervation on sympathetic activity was reflected by the changes of HR and significant SBP, HRV parameters and plasma norepinephrine levels. Cardiac hyper-sympathetic activity was manifested by shortened RR-interval shortened, increased LF, HF and LF/HF ratio and elevated plasma norepinephrine levels after LSG stimulation. AF inducibility was facilitated and atrial electrophys- iological changes (shortening of ERP, increasing of ERP disper- sion) was aggravated by hyper-sympathetic nerve activity. The Role of Sympathetic Nerve Activity in AF and Acute Atrial Electrophysiological Changes Previous studies have provided evidence that sympathetic nerve activity involved in the initiation and/or maintenance of AF. Infusion of isoprenaline (isoproterenol, ISO), a b-receptor agonist, induced AF in 5% of patients without history of AF and in 84% of patients with paroxysmal AF in a dose-dependent manner [3]. Several b-blockers have been shown to be effective in suppression of AF in patients with various heart diseases [21]. Chronic RAP increases the innervation of the atrial sympathetic nervous system [22,23]. Furthermore, atrial sympathetic hyper-innervation was associated with chronic AF in patients [24]. Swissa and Tan et al. have reported that sympathetic hyper-innervation induced either by electrical stimulation of SG or nerve growth factor infusion can induce atrial nerve sprouting and a significantly higher incidence of paroxysmal AF and paroxysmal atrial tachycardia in dogs [15,25]. By contrast, reduction of cardiac sympathetic outflow by cryoablation of the bilateral stellate ganglia and T2–T4 thoracic ganglia can effectively eliminate both paroxysmal atrial tachyar- rhythmia in dogs with pacing-induced heart failure [26]. Figure 5. Plasma norepinephrine level. RSD = the renal sympa- thetic denervation. #P,0.001 indicated significant difference between stimulation and baseline conditions. {P,0.001 indicated significant difference between ablation and stimulation conditions. P,0.001 indicated significant difference between the control group and the renal sympathetic denervation group. doi:10.1371/journal.pone.0078218.g005 A recent study from our institute showed that LSG stimulation facilitated AF inducibility and aggravated atrial electrical atrial electrophysiological changes in the first 2 to 4 hours [27]. The inhibition of sympathetic nerve activity by left stellate ganglion- ectomy can reduce the AF induction and reverses the process of November 2013 \| Volume 8 \| Issue 11 \| e78218 November 2013 \| Volume 8 \| Issue 11 \| e78218 PLOS ONE \| www.plosone.org 6 Renal Denervation Inhibits Atrial Fibrillation Figure 6. Sliver staining and tyrosine hydroxylase staining of the renal sympathetic nerves. A and B shows an example of silver staining of the renal artery nerves without ablation. C shows an example of tyrosine hydroxylase -positive staining of renal sympathetic nerves without ablation. Brown structures indicate positively stained nerve structures. doi:10.1371/journal.pone.0078218.g006 Figure 6. Sliver staining and tyrosine hydroxylase staining of the renal sympathetic nerves. A and B shows an example of silver staining of the renal artery nerves without ablation. C shows an example of tyrosine hydroxylase -positive staining of renal sympathetic nerves without ablation. Brown structures indicate positively stained nerve structures. Study Limitations Although we show Although we showed evidence that renal nerve stimulation or ablation could activate or suppress sympathetic activity by indirectly observe changes of HR and SBP, direct neural firing from the SG or renal sympathetic nerve was not recorded in this study. The spectral index LF, obtained from systolic arterial pressure variability (LFSAP) which can furnished additional information on the sympathetic vasomotor control and reflect sympathetic activity was not tested. Since renal arteriography was not performed before and after catheter-based renal sympathetic denervation in our study, whether the renal artery had obvious stenosis after ablation or not was unclear. In addition, anaesthesia is known to interfere with the cardiovascular autonomic control. Furthermore, we need to investigate long-term changes of AF induction, electrophysiological data, and concentrations of nor- epinephrine in the hypothalamic after RSD. RSD was attempted to prevent or terminate arrhythmia. In 2012, Ukena [33] et al. reported the first-in-man experience in two patients with chronic heart failure and electrical storm, in whom catheter-based renal sympathetic denervation was recom- mended and performed after other treatment options failed. They suggest that RDN is feasible even in cardiac unstable patients. Linz et al [11] first reported that renal denervation reduced the inducibility of obstructive sleep apnea-associated AF. A pro- nounced AERP shortening induced by vagally mediated negative tracheal pressure was modulated by RSD or atenolol. Recently, their study showed that electrical baroreflex stimulation signifi- cantly shortened atrial refractoriness, thereby causing increased AF inducibility. In contrast, RSD did not significantly modulate atrial refractoriness or AFinducibility [12]. Zhao et al [13] demonstrated that episodes and duration of AF could be reduced by RSD during 7-hour RAP in dogs. The reduction of AF inducibility might be associated with decreased activity of renin- angiotensin-aldosterone system. Notably, AF in all the three studies associates with a substantially increase in vagal activity, leading to inconsistent effects of RSD. It is not clear if the inhibitory effect of RSD on AF is related to inhibition of the excitatory reflexs of vagal activity. Acknowledgments We thank Dr Yan Yan of Department of Cardiovascular Diseases, Zhongshan Hospital Fudan University for assistance and support. We thank Dr Yan Yan of Department of Cardiovascular Diseases, Zhongshan Hospital Fudan University for assistance and support. In the present study, RSD significantly reduced AF inducibility and reversed the atrial electrophysiological changes.The results strongly suggested that RSD had a potent inhibitory effect on AF induced by hyper-sympathetic activity. The mechanism may be that the excitatory reflexes on central sympathetic outflow including cardiac sympathetic activity was inhibited by afferent renal nerve denervation. Conclusions Hyper-sympathetic activity may facilitate the initiation of AF and acute atrial electrophysiological changes. RSD significantly reduced AF inducibility and reversed the atrial electrophysiolog- ical changes induced by hyper-sympathetic activity. The mecha- nism maybe that excitatory reflexes on cardiac sympathetic outflow was inhibited by afferent renal nerve denervation. Author Contributions Conceived and designed the experiments: YH QZ. Performed the experiments: QZ HW LZ. Analyzed the data: QZ JH. Contributed reagents/materials/analysis tools: FZ KW. Wrote the paper: QZ YH. Revised and proofreaded the manuscript, guided the experiments: SSP. Conceived and designed the experiments: YH QZ. Performed the experiments: QZ HW LZ. Analyzed the data: QZ JH. Contributed reagents/materials/analysis tools: FZ KW. Wrote the paper: QZ YH. Revised and proofreaded the manuscript, guided the experiments: SSP. The ventricular response rate in atrial fibrillation is often a manifestation of sympathetic state. We found that the shortened 8. Ahmed H, Neuzil P, Skoda J, Petru J, Sediva L, et al. (2012) Renal sympathetic denervation using an irrigated radiofrequency ablation catheter for the 7. Esler MD, Krum H, Sobotka PA, Schlaich MP, Schmieder RE, et al. (2010) Renal sympathetic denervation in patients with treatment-resistant hypertension (the symplicity htn-2 trial): a randomised controlled trial. Lancet 376: 1903– 1909. 1. Patterson E, Lazzara R, Szabo B, Liu H, Tang D, et al. (2006) Sodium-calcium exchange initiated by the Ca2+ transient:an arrhythmia trigger within pulmonary veins. J Am Coll Cardiol 47: 1196–1206. 2. Sharifov OF, Fedorov VV, Beloshapko GG, Glukhov AV, Yushmanova AV, et al. (2004) Roles of adrenergic and cholinergic stimulation in spontaneously atrial fibrillation in dogs. J Am Coll Cardiol 43: 483–490. 3. Oral H, Crawford T, Frederick M, Gadeela N, Wimmer A, et al. (2008) Inducibility of paroxysmal atrial fibrillation by isoproterenol and its relation to the mode of onset of atrial fibrillation. J Cardiovasc Electrophysiol 19: 466–470. 4. Chen SA, Hsieh MH, Tai CT, Tsai CF, Prakash VS, et al. (1999) Initiation of atrial fibrillation by ectopic beats originating from the pulmonary veins:elec- trophysiological characteristics, pharmacological response, and effect s of radiofrequency ablation. Circulation 100: 1879–1886. Catheter-based Renal Sympathetic Denervation as a Potential Strategy for AF Management The sophisticated network of afferent and efferent sensory, chemo- and baroreceptor nerve fibers is distributed as a network in the adventitia of the renal artery and throughout the kidney [28]. The sympathetic nerves to the kidneys terminate in the blood vessels, the juxtaglomerular apparatus, and the renal tubules [29]. Altering the signals from the kidney to the hypothalamus is expected to impact peripherally, including on arterial resistance, The following electrophysiological mechanisms have been proposed to account for AF: sympathetic stimulation shortens Figure 7. HE staining of the renal artery and renal nerves structures with or without ablation. (A) Non-ablated renal artery in control group. (B) Non-ablated renal ganglion in control group. (C) The ablated renal artery tissue was replaced by dense scar tissue that stained purple. (D) In the ablated renal artery, the ganglioinic cells become contracted morphologically and vacuolar degeneration was observed. doi:10.1371/journal.pone.0078218.g007 Figure 7. HE staining of the renal artery and renal nerves structures with or without ablation. (A) Non-ablated renal artery in control group. (B) Non-ablated renal ganglion in control group. (C) The ablated renal artery tissue was replaced by dense scar tissue that stained purple. (D) In the ablated renal artery, the ganglioinic cells become contracted morphologically and vacuolar degeneration was observed. doi:10.1371/journal.pone.0078218.g007 November 2013 \| Volume 8 \| Issue 11 \| e78218 PLOS ONE \| www.plosone.org 7 Renal Denervation Inhibits Atrial Fibrillation peripheral and central chemoreceptors, sympathetic activity of the kidney and the heart itself [30]. Recently, therapeutic catheter- based RSD has been applied in a controlled randomized trial for the treatment of resistant hypertension with a significant reduction of systolic blood pressure, diastolic pressure [7–10]. RSD was also shown to reduce resting HR in patients with resistant hypertension [31]. The reduction of central sympathetic stimulation associated with nerve stimulation from the carotid sinus may result in substantially reduced renal sympathetic efferent signaling [32]. In our study, we found that no significant difference in SBP was found after stimulation between the control group and RSD group and RSD decreased SBP after ablation. RSD significantly reversed the shortened RR-interval, increased LF, HF and LF/HF ratio and elevated plasma norepinephrine levels by LSG stimulation. These findings suggested that RSD significantly inhibited cardiac sympathetic activity. Therefore, our results confirmed the hypothesis that RSD can reduce not only local sympathetic activity (efferent renal sympathetic nerve activity and afferent renal nerve activity) but also whole-body sympathetic activities, includ- ing cardiac sympathetic activity. RR-interval (elevated ventricular rate) induced by LSG stimula- tion can be prolonged by RSD. The ventricular rate in AF has been observed to fall following renal denervation in resistant hypertension (Felix Mahfoud, personal communication). While this may represent changes in myocardial work and stress associated with blood pressure declines, and indicates reductions in direct cardiac sympathetic signaling [34]. Consistently, HRV analysis in this study showed that RSD markedly reversed the abnormity of the time and frequency domain parameters (SDNN, pNN50, LF, HF and LF/HF ratio) induced by hyper-sympathetic nerve activity. These results further demonstrated that RSD modulated the cardiac autonomic activity which innervated the atrium and facilitated AF. 1. Patterson E, Lazzara R, Szabo B, Liu H, Tang D, et al. (2006) Sodium-calcium exchange initiated by the Ca2+ transient:an arrhythmia trigger within pulmonary veins. J Am Coll Cardiol 47: 1196–1206. 6. Ogawa M, Tan AY, Song J, Kobayashi K, Fishbein MC, et al. (2009) Cryoablation of stellate ganglia and atrial arrhythmia in ambulatory dogs with pacing-induced heart failure. Heart Rhythm 6: 1772–1779. p y J 2. Sharifov OF, Fedorov VV, Beloshapko GG, Glukhov AV, Yushmanova AV, et al. (2004) Roles of adrenergic and cholinergic stimulation in spontaneously atrial fibrillation in dogs. J Am Coll Cardiol 43: 483–490. 5. Ku¨hlkamp V, Schirdewan A, Stangl K, Homberg M, Ploch M, et al. (2000) Use of metoprolol CR/XL to maintain sinus rhythm after conversion from persistent atrial fibrillation:a randomized, double-blind, placebo-controlled study. J Am Coll Cardiol 36: 139–146. 6. Ogawa M, Tan AY, Song J, Kobayashi K, Fishbein MC, et al. (2009) Cryoablation of stellate ganglia and atrial arrhythmia in ambulatory dogs with pacing-induced heart failure. Heart Rhythm 6: 1772–1779. 7. Esler MD, Krum H, Sobotka PA, Schlaich MP, Schmieder RE, et al. (2010) Renal sympathetic denervation in patients with treatment-resistant hypertension (the symplicity htn-2 trial): a randomised controlled trial. Lancet 376: 1903– 1909. 8. Ahmed H, Neuzil P, Skoda J, Petru J, Sediva L, et al. (2012) Renal sympathetic denervation using an irrigated radiofrequency ablation catheter for the Renal Denervation Inhibits Atrial Fibrillation management of drug-resistant hypertension. JACC Cardiovasc Interv 5: 758– 765. management of drug-resistant hypertension. JACC Cardiovasc Interv 5: 758– 765. 22. Chang CM, Wu TJ, Zhou S, Doshi RN, Lee MH, et al. (2001) Nerve sprouting and sympathetic hyperinnervation in a canine model of atrial fibrillation produced by prolonged right atrial pacing. Circulation 103: 22–25. 9. Prochnau D, Lucas N, Kuehnert H, Figulla HR, Surber R (2012) Catheter- based renal denervation for drug-resistant hypertension by using a standard electrophysiology catheter. EuroIntervention 7: 1077–1080. ´ 23. Jayachandran JV, Sih HJ, Winkle W, Zipes DP, Hutchins GD, et al. (2000) Atrial fibrillation produced by prolonged rapid atrial pacing is associated with heterogeneous changes in atrial sympathetic innervation. Circulation 101: 1185– 1191. p y gy 10. Witkowski A, Prejbisz A, Florczak E, Ka˛dziela J, S´liwin´ski P et al. (2011) Effects of renal sympathetic denervation on blood pressure, sleep apnea course, and glycemic control in patients with resistant hypertension and sleep apnea. Hypertension 58: 559–565. 24. Gould PA, Yii M, McLean C, Finch S, Marshall T, et al. (2006) Evidence for increased atrial sympathetic innervation in persistent human atrial fibrillation. Pacing Clin Electrophysiol 29: 821–829. yp 11. Linz D, Mahfoud F, Schotten U, Ukena C, Neuberger HR, et al. (2012) Renal sympathetic denervation suppresses postapneic blood pressure rises and atrial fibrillation in a model for sleep apnea. Hypertension 60: 172–178. g p y 25. Swissa M, Zhou S, Paz O, Fishbein MC, Chen LS, et al. (2005) Canine model of paroxysmal atrial fibrillation and paroxysmal atrial tachycardia. Am J Physiol Heart Circ Physiol 289: H1851–7. 12. Linz D, Mahfoud F, Schotten U, Ukena C, Neuberger HR, et al. (2013) Effects of Electrical Stimulation of Carotid Baroreflex and Renal Denervation on Atrial Electrophysiology. J Cardiovasc Electrophysiol. 2013 Apr 12. 26. Ogawa M, Tan AY, Song J, Kobayashi K, Fishbein MC, et al. (2009) Cryoablation of stellate ganglia and atrial arrhythmia in ambulatory dogs with pacing-induced heart failure. Heart Rhythm 6: 1772–1779. 13. Zhao Q, Yu S, Zou M, Dai Z, Wang X, et al. (2012) Effect of renal sympathetic denervation on the inducibility of atrial fibrillation during rapid atrial pacing. J Interv Card Electrophysiol 35: 119–125. pacing-induced heart failure. Heart Rhythm 6: 1772–1779. 27. Qina Z, Yuemei H (2012) Stellate ganglion electrical stimulation for establishing a canine model of acute atrial fibrillation mediated by sympathetic nerve. Heart 98 supplement 2: E30. J p y 14. Renal Denervation Inhibits Atrial Fibrillation Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG (2010) Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol. 2010 8: e1000412. 28. Vonend O, Marsalek P, Russ H, Wulkow R, Oberhauser V, et al. (2003) Moxonidine treatment of hypertensive patients with advanced renal failure. J Hypertension 21: 1709–1717. 15. Tan AY, Zhou S, Ogawa M, Song J, Chu M, et al. (2008) Neural mechanisms of paroxysmal atrial fibrillation and paroxysmal atrial tachycardia in ambulatory canines. Circulation 118: 916–925. 29. Barajas L (1978) Innervation of the renal cortex. Fed Proc 37: 1192–1201. 30. Schlaich MP, Lambert E, Kaye DM, Krozowski Z, Campbell DJ, et al. (2004) Sympathetic augmentation in hypertension: role of nerve firing norepinephrine reuptake, and angiotensin neuromodulation. Hypertension 43: 169–175. 16. Lu Z, Scherlag BJ, Lin J, Niu G, Fung KM, et al. (2008) Atrial fibrillation begets atrial fibrillation: autonomic mechanism for atrial electrical remodeling induced by short-term rapid atrial pacing.Circ Arrhythm Electrophysiol 1: 184–192. 31. Ukena CMF, Kindermann I, Barth C, Barth C, Lenski M, et al. (2011) Cardiorespiratory response to exercise after renal sympathetic denervation in patients with resistant hypertension. J Am Coll Cardiol 58: 1176–1182. 17. Shi Z, Chen AD, Xu Y, Chen Q, Gao XY, et al. (2009) Long-term administration of tempol attenuates postinfarct ventricular dysfunction and sympathetic activity in rats. Pflugers Arch 458: 247–257. y p y g 18. Rager G, Lausmann S, Gallyas F (1979) An improved silver stain for developing nervous tissue. Stain Technol 54: 193–200. 32. Heusser K, Tank J, Engeli S, Diedrich A, Menne J, et al. (2010) Carotid baroreceptor stimulation, sympathetic activity,baroreflex function, and blood pressure in hypertensive patients. Hypertension 55: 619–626. 19. Cao JM, Chen LS, KenKnight BH, Ohara T, Lee MH, et al. (2000) Nerve sprouting and sudden cardiac death. Circ Res 86: 816–821. sprouting and sudden cardiac death. Circ Res 86: 816–821 33. Ukena C, Bauer A, Mahfoud F, Schreieck J, Neuberger HR, et al. (2012) Renal sympathetic denervation for treatment of electrical storm: first-in-man experience. Clin Res Cardiol 101: 63–67. 20. Ishida Y, Goto N, Chiba K, Souma Y, Goto J (2008) Comparison of preparation methods for the evaluation of nerve axons. Okajimas Folia Anat Jpn 84: 133– 135. 34. Sobotka PA, Mahfoud F, Schlaich MP, Hoppe UC, Bo¨hm M, et al. (2011) Sympatho-renal axis in chronic disease. Clin Res Cardiol 100: 1049–1057. 21. References 5. Ku¨hlkamp V, Schirdewan A, Stangl K, Homberg M, Ploch M, et al. (2000) Use of metoprolol CR/XL to maintain sinus rhythm after conversion from persistent atrial fibrillation:a randomized, double-blind, placebo-controlled study. J Am Coll Cardiol 36: 139–146. p y J 2. Sharifov OF, Fedorov VV, Beloshapko GG, Glukhov AV, Yushmanova AV, et al. (2004) Roles of adrenergic and cholinergic stimulation in spontaneously atrial fibrillation in dogs. J Am Coll Cardiol 43: 483–490. 6. Ogawa M, Tan AY, Song J, Kobayashi K, Fishbein MC, et al. (2009) Cryoablation of stellate ganglia and atrial arrhythmia in ambulatory dogs with pacing-induced heart failure. Heart Rhythm 6: 1772–1779. g 3. Oral H, Crawford T, Frederick M, Gadeela N, Wimmer A, et al. (2008) Inducibility of paroxysmal atrial fibrillation by isoproterenol and its relation to the mode of onset of atrial fibrillation. J Cardiovasc Electrophysiol 19: 466–470. J p y 4. Chen SA, Hsieh MH, Tai CT, Tsai CF, Prakash VS, et al. (1999) Initiation of atrial fibrillation by ectopic beats originating from the pulmonary veins:elec- trophysiological characteristics, pharmacological response, and effect s of radiofrequency ablation. Circulation 100: 1879–1886. PLOS ONE \| www.plosone.org November 2013 \| Volume 8 \| Issue 11 \| e78218 8 Renal Denervation Inhibits Atrial Fibrillation Lehnart SE, Wehrens XHT, Marks AR (2004) Calstabin deficiency, ryanodine receptors, and sudden cardiac death. Biochem Biophys Res Commun 322: 1267–1279. PLOS ONE \| www.plosone.org November 2013 \| Volume 8 \| Issue 11 \| e78218 PLOS ONE \| www.plosone.org 9
W2469709749.txt	http://hipatiapress.com/hpjournals/index.php/hse/article/download/2062/pdf	en		Filipinos in the U.S.: Historical, Social, and Educational Experiences	Historia social y de la educación	2,016	cc-by	8,406	Instructions for authors, subscriptions and further details: http://hse.hipatiapress.com Filipinos in the U.S.: Historical, Social, and Educational Experiences Susan J. Paik1, Shirlie Mae Mamaril Choe1, Matthew A. Witenstein2 1) Claremont Graduate University (USA) 2) University of San Diego (USA) Date of publication: June 23rd, 2016 Edition period: June 2016 – October 2016 To cite this article: Paik, S.J., Choe, S.M.M., & Witenstein, M.A. (2016). Filipinos in the U.S.: Historical, Social, and Educational Experiences. Social and Education History 5(2), 134-160. doi:10.17583/hse.2016.2062 To link this article: http://dx.doi.org/10.17583/hse.2016.2062 PLEASE SCROLL DOWN FOR ARTICLE The terms and conditions of use are related to the Open Journal System and to Creative Commons Attribution License (CC-BY). HSE – Social and Education History Vol. 5 No. 2 June 2016 pp.134160 Filipinos in the U.S.: Historical, Social, and Educational Experiences Susan J. Paik Claremont Graduate University (USA) Shirlie Mae Mamaril Choe Claremont Graduate University (USA) Matthew A. Witenstein University of San Diego (USA) Abstract _________________________________________________________________ The purpose of this article is to better understand current Filipino American communities and their educational experiences by examining the historical and social context of their immigration to the United States. Filipinos are the second largest Asian American group with a distinct immigration history that has been complicated by centuries of colonization by Spain and later the United States. Based on an adapted model of incorporation and literature review, the article examines government policies, societal reception, co-ethnic communities, as well as other barriers and opportunities, which influenced their acculturation both before and after the Immigration Act of 1965. Government and societal reception of Filipinos have ranged from hostile to mostly neutral positions through four waves of immigration. Earlier Filipino communities were stronger often serving their economic, cultural and social needs, but have become more dispersed over time as Filipinos have become less reliant on their co-ethnic networks. Despite their overall success, there is research showing mixed educational achievement levels across later generations. The Filipino community is steadily growing and more research and support are needed for Filipino American students. Historical contexts can provide a comprehensive lens to understand current educational issues. Key words: Filipino American, immigration, history, communities, education 2016 Hipatia Press ISSN: 2014-3567 DOI: 10.17583/hse.2016.2062 HSE – Social and Education History Vol. 5 No. 2 February 2016 pp. 134-160 Filipinos en USA: Experiencia Histórica, Social y Educativa. Susan J. Paik Claremont Graduate University (USA) Shirlie Mae Mamaril Choe Claremont Graduate University (USA) Matthew A. Witenstein University of San Diego (USA) Resumen _________________________________________________________________ El propósito de este artículo es conocer mejor las actuales comunidades filipino-estadounidenses y sus experiencias educativas a través del análisis del contexto histórico y social de su inmigración a los Estados Unidos. Los filipinos son el segundo grupo cultural asiático-estadounidense más grande, con una historia inmigratoria distinta que se ha complicado debido a siglos de colonización por parte de España y más tarde Estados Unidos. Basado en un modelo adaptado de incorporación y revisión de la literatura, el artículo examina las políticas gubernamentales, la recepción social, las comunidades co-étnicas, así como otras barreras y oportunidades que influyeron en su aculturación, tanto antes como después de la Ley de Inmigración de 1965. Las posturas gubernamentales y la recepción social de las personas filipinas han oscilado desde posiciones hostiles a actitudes más neutrales a lo largo de cuatro olas de inmigración. Las primeras comunidades filipinas se mantuvieron fuertes cubriendo sus necesidades económicas, culturales y sociales pero con el tiempo se han vuelto más dispersas ya que los filipinos y filipinas se muestran menos dependientes de sus redes co-étnicas. A pesar de su éxito global, hay investigaciones que muestran que han alcanzado niveles educativos mixtos a lo largo de las generaciones posteriores. La comunidad filipina está creciendo de manera constante y es necesario realizar más investigaciones y aportar apoyo a los y las estudiantes filipino-estadounidenses. Estos contextos históricos pueden proporcionar instrumentos de análisis para entender los problemas educativos actuales. Palabras clave: Filipino-estadounidense, inmigración, historia, comunidades, educación 2016 Hipatia Press ISSN: 2014-3567 DOI: 10.17583/hse.2016.2062 136 Paik, Choe & Witenstein– Filipinos in U.S. A ccording to the U.S. Census 2010, there are almost 18,000,000 Asian Americans or roughly 6% of the U.S. population. After Chinese Americans, Filipinos constitute the second largest Asian group at 3.4 million, representing approximately 18% of the adult Asian population in the U.S. (Hoeffel, Rastogi, Kim, & Hasan, 2012; Pew Research Center, 2013; U.S. Census Bureau, 2010). Filipino Americans have had a long history in the U.S., tracing back as early as the 1500s in present-day Morro Bay, California (Nadal, 2011). For centuries, Filipino Americans have clearly had a long presence in American society, with unique historical, social, and educational experiences. However, few research studies have focused strictly on Filipino Americans; more often they are included as part of broader examinations of Asian Americans, assuming a homogeneity of histories and current experiences. Historically, mainstream American society has had the tendency to treat all persons of Asian ancestry alike, making an Asian American panethnic identity instrumental in addressing common issues and challenges facing different Asian American ethnic groups (Chan, 1991; Espiritu, 1993). Unfortunately, given the heterogeneity of Asian American communities, the “Asian American” label has deemphasized the distinctive experiences of specific groups and masked key differences including but not limited to educational attainment, socioeconomic status, language ability, regional residency, and gender (Lee, 2006). It is important to acknowledge the complex ways in which these factors may distinguish the educational and later career experiences of specific Asian populations such as Filipino Americans (Museus & Maramba, 2011). Moreover, their complex colonial history involving centuries of Spanish rule and later American occupation, warrants a closer look at this community. Their distinct immigration patterns both before and after the Immigration Act of 1965 also underscore their unique story. Currently, there are few studies that examine the linkages between the historical and social experiences of Filipino Americans with present-day educational patterns and issues. An exploration of factors such as societal reception, government support, co-ethnic communities, settlement patterns, language ability, class status, education, occupation, and time of arrival to the U.S. would help to understand success outcomes for this group. This HSE – Social and Education History, 5(2) 137 type of historical analysis of the immigration experiences of Filipino Americans would provide a long-lens perspective and insight that is limited in contemporary educational research (Paik, Kula, Saito, Rahman, & Witenstein, 2014). Therefore, the purpose of this article is to provide a historical examination of Filipino immigration experiences to better understand Filipino communities and their current outcomes in education and related issues. This article specifically: 1) employs a model of incorporation adapted by Paik et al. (2014) to understand the experiences of Filipino American communities, 2) presents demographic, educational, and sociocultural background and context on Filipinos, 3) provides historical context on waves of Filipino immigration both before and after the Immigration Act of 1965, 4) systematically analyzes historical experiences based on the theoretical framework and through the review of the literature, and 5) discusses findings and implications as it relates to Filipino immigration experiences and educational outcomes. Theoretical Framework: Modes of Incorporation Paik and colleagues (2014) adapted Portes and Rumbaut’s (1990, 2001) early work on modes of incorporation to examine the immigration experiences of Asian communities in the U.S. and their impact on later group outcomes (e.g. educational and occupational). The original model looked at the reception of immigrant groups by government policy, society, and co-ethnic communities. Paik et al. (2014) added factors related to other barriers and opportunities such as time of arrival; location and settlement patterns; class status, occupation, and educational level; and language abilities. The model is adapted to the historical immigration experiences of Asian communities, including Filipino Americans, capturing the changing migration patterns and demographic features of the communities based on policies specifically targeting immigrants from Asia in the pre and post-1965 eras (Paik et al., 2014; Portes, Fernández-Kelly, & Haller, 2009; Solorzano & Villalpando, 1998). The adapted modes of incorporation focuses attention on educational outcomes, providing historical, social, and political contexts 138 Paik, Choe & Witenstein– Filipinos in U.S. for understanding the broad impact of Filipino immigrant experiences. Brief descriptions of the framework’s factors are below and will be described further for Filipino American communities in the following sections: Government Policy The reception of immigrants through government policy may be categorized as “receptive”, “indifferent”, or “hostile” (Portes & Rumbaut, 1990; 2001). Prior to the implementation of the 1965 Immigration Act, most Filipino immigrants along with other Asian immigrants met “hostile” policies, which blocked or severely curtailed immigration to the U.S. After 1965, policies were either “indifferent” (or neutral) – allowing for open immigration without any assistance – or “receptive” – offering assistance or incentive programs for immigration. Unique to Filipinos than any other Asian group, they experienced hostile, indifferent, and receptive policies throughout their immigration history (Paik et al., 2014). Societal Reception Societal reception is a key aspect of the immigration experience. Public perception and prejudices against immigrant populations may strongly influence the types of employment and institutions that are open to new immigrants. Upon entry into the U.S., immigrants may face “prejudiced”, “not prejudiced”, or “neutral” reception. All Asian immigrants experienced “prejudiced” reception upon arrival pre- and post-1965; there was no group who experienced “unprejudiced” reception. However, Filipinos and South Asians were the only two Asian groups who had more receptivity or “neutral” reception after 1965 (Paik et al., 2014). Co-ethnic Communities Co-ethnic communities are members of the same ethnic or national origin group often living in geographic proximity to one another (Portes & HSE – Social and Education History, 5(2) 139 Rumbaut, 1990; Zhou & Bankston III, 1998). Large ethnic clusters of residential and business districts may be found in metropolitan areas, known as ethnic towns (or enclaves), or suburban areas, also known as ethnoburbs (Li, 2009). According to Portes and Rumbaut (1990, 2001), immigrants may encounter one of three types of communities upon entering the host country: “weak”, “strong”, or “dispersed”. The strength or weakness of a co-ethnic community is determined by the concentration or disbursement of laborers, professionals, or entrepreneurs. The lack of or presence of these workers in ethnically concentrated communities may greatly influence the types of opportunities available to new immigrants in terms of education, jobs, or other key resources (Portes & Rumbaut, 2001, p. 48). The strongest communities are highly concentrated in numbers and offer the most opportunities and resources, while the weakest communities are smaller and less skilled, limiting their resources and opportunities. Dispersed communities, such as Filipinos who arrived post-1965, are less reliant on their ethnic communities for resources in general. Their earlier pre-1965 communities were stronger, but became more dispersed over time. While sense of community is still important for them, their ability to speak English upon arrival, their educational background, and professional skills allowed them to live in less ethnically concentrated areas, allowing them to navigate mainstream America more easily than most new Asian immigrants. Other Barriers and Opportunities Paik and colleagues’ (2014) addition of other barriers and opportunities to modes of incorporation highlight six other factors that are critical to understanding immigration experiences, particularly Filipino immigration. As previously mentioned, the reception of Asian immigrants in the U.S. differed depending on their time of arrival. Immigrants that arrived before 1965 typically faced more challenges and barriers, particularly in terms of government policies, than those that arrived after 1965 (Chan, 1991; Takaki, 1998). The immigrant experience is also influenced by location or settlement patterns. Immigrants to places such as California or Hawaii have greater access to large co-ethnic communities and face less challenges to 140 Paik, Choe & Witenstein– Filipinos in U.S. acculturation than those that are dispersed in regions with limited numbers of co-ethnic populations (Chan, 1991; Portes & Rumbaut, 1990, 2001; Takaki, 1998). Demographic factors such as class status (SES), occupation, and educational levels upon arrival to the U.S. are also significant factors in the immigrant experience. In general, higher levels of education, more professional skills, or higher-class status afford immigrants with greater resources and opportunities. Lastly, language ability, or English-language fluency, may help to reduce challenges and ease acculturation into mainstream America. These additional barriers and opportunities for Filipinos will be further delineated in the following sections. Background, Cultural Context and Educational Trends Filipino American communities comprise a diverse mix of individuals in terms of levels of acculturation, number of generations living in America, socioeconomic background, geographical origin, dialect/languages, and even culture (Lai & Arguelles, 1998). The following section will provide demographic background, cultural context, and educational trends that influence the current population. Current Demographics and Geographic Distribution As previously stated, with a population of over 3.4 million, Filipino Americans comprise the second largest Asian American group in the U.S. This figure includes single ethnicity individuals, as well as individuals who identify with one or more groups (Hoeffel, Rastogi, Kim, & Hasan, 2012). According to the 2010 U.S. Census, there are nearly 2,650,000 individuals identifying solely as Filipino American, an increase of 38.9% since the 2000 Census, while the overall number including mixed ethnicity increased 44.1% (Hoeffel et al., 2012). Over 65% of the Filipino American population resides on the West Coast, with the largest proportion living in California at nearly 1.5 million; additionally, at least 100,000 Filipino Americans live in each of the following states: Hawaii, Illinois, Texas, Washington, New Jersey, New HSE – Social and Education History, 5(2) 141 York, Nevada, and Florida (Hoeffel et al., 2012; National Federation of Filipino American Associations, 2011; U.S. Census Bureau, 2000). Moreover, Filipino Americans are the largest Asian American group in Alaska, Arizona, California, Hawaii, Idaho, Montana, Nevada, New Mexico, Washington, and Wyoming (Hoeffel et al., 2012). Cultural Context Past studies have often cited the family as a crucial component of Filipino American identity and cultural, with a strong emphasis on the larger extended family, or clan (Agbayani-Siewert, 1994; Chan, 1992; Cimmarusti, 1992). The Filipino extended family typically includes a network of blood relations, family members related by marriage, godparents, and close family friends (Agbayani-Siewert, 1994). The compadrazgo system ritually bonds godparents to their godchildren and the children’s parents through socioemotional and economic support (Salvador, Omizo, & Kim, 1997). Additionally, compared to other Asian families, the Filipino family structure is described as primarily egalitarian in nature (Chao & Tseng, 2002). The egalitarian family structure encourages any family member, female or male, to advance the family status and resources through educational and financial success (Okamura & Agbayani, 1997). The behavior of Filipino Americans may also be impacted by three primary cultural values that are reflective of the strong emphasis placed on relationship building especially within the extended family network (Salvador et al., 1997). Respect, especially of one’s elders, is often acted out through the concept of utang na loob or debt of gratitude. Utang na loob evokes feelings of reciprocal obligation, especially in terms of repaying acts of kindness (Agbayani-Siewert, 1994; Nadal, 2011; Salvador et al., 1997). This sense of gratitude may manifest in children feeling familial pressure to succeed in school as a way of “repaying” parental sacrifices to provide educational opportunities. Individuals who do not follow this code of gratitude are viewed as disrespectful and may be ostracized, which violates another key value known as pakikisama or harmony (Agbayani-Siewert, 1994; Bankston, 2006; Cimmarusti, 1992). Maintaining the appearance of 142 Paik, Choe & Witenstein– Filipinos in U.S. harmonious social relations (or group solidarity) requires people to always act in a respectful manner towards each other, even if it means suffering in silence rather than initiating a conflict (Cimmarusti, 1992; Salvador et al., 1997). Lastly, the emphasis on not bringing hiya or shame to the family reinforces the need to maintain harmonious relations and meet obligations (Agbayani-Siewert, 1994; Nadal, 2011). These cultural concepts may influence critical life choices like an individual’s educational or career decision-making process, where a person may feel obligated to pursue a major or career chosen by their parents or other elders. Choosing against elders’ wishes may not be in line with concepts of pakikisama or utang na loob. Educational Trends In general, Filipino American educational achievement levels are high – Filipino adults aged 25 and older have a higher rate of educational attainment (47%) compared to the national population (28%) overall (Pew Research Center, 2013). However, Museus and Maramba (2011) pointed out that disparities exist in the Filipino American educational experience, especially when looking at regions where the largest numbers of Filipino Americans reside. Their research revealed educational attainment of Filipino Americans has a bimodal character and that underrepresentation continues to exist at four-year post-secondary institutions in states where Filipino Americans are concentrated. Additionally, Zhou and Xiong (2005) found that postsecondary achievement levels varied by generation with first generation students achieving significantly higher than second. Though limited studies are available on the K-12 educational experiences of Filipino Americans, existing studies suggest that the mixed character of educational success is also reflected in K-12 education. In particular, one study of ten U.S. urban communities found that while high achievement among Filipino American K-12 public school students existed in six areas, patterns of underachievement were found in four other California and Hawaii cities, homes to large and long-established Filipino communities (Museus & Maramba, 2011; Ogilvie, 2008). HSE – Social and Education History, 5(2) 143 Immigration History The complex nature of Filipino Americans in the U.S. is greatly influenced by the immigration policies that have excluded and invited immigrants from the Philippines at different times in history (Lai & Arguelles, 1998). The changing patterns of Filipino immigration makes it challenging to generalize the population which is comprised both of long-standing communities eight generations old to the continuous influx of present day immigrants throughout the United States. The following section traces key waves of Filipino immigration. First Wave of Immigration The Philippines has a long history of overseas migration stemming back to the times of Spanish colonization from 1521 to 1898 (Bonus & Maramba, 2013). Filipinos, then known as “Manilamen”, first arrived in North America as a result of the Manila-Acapulco galleon trade between 1565 to 1815 (Bonus & Maramba, 2013; Cordova, Cordova, & Acena, 1983). The earliest documentation of a Filipino American settlement is the Louisiana fishing village of St. Malo dating as early as 1843; however, other scholars have noted Filipinos in the U.S in the late 1700s as many escaped brutal conditions on the Spanish galleons and formed communities in the region (Lai & Arguelles, 1998; Lee, 2015; Nadal, 2011). During this long period of Spanish colonial rule, other Filipino seafarers, forced into service on Philippine-made Spanish vessels, landed in California, British Columbia, Washington, Alaska, and Hawaii (Cordova et al., 1983). Second Wave of Immigration American annexation of the Philippines following the end of the SpanishAmerican War in 1898 enabled steady immigration of Filipinos to the 144 Paik, Choe & Witenstein– Filipinos in U.S. United States (Cordova et al., 1983). Filipinos were not considered citizens; yet, as members of a U.S. territory, their status as American nationals made them exempt from early immigration laws that prohibited other Asian immigration (Chan, 1991; Takaki, 1989). The first formal group to arrive during this period were the pensionados – students whose postsecondary education in the U.S. were subsidized by the Philippine territorial government (Cordova et al., 1983; Lai & Arguelles, 1998). Under the Pensionado Act of 1903, which lasted until 1938, pensionados entered into a contract by which for every year of education, they were obligated to work for the Philippine colonial government. Approximately 14,000 students took advantage of this program, many of whom returned to the Philippines. The pensionados were also followed by self-supporting students seeking greater economic opportunities in the United States (Cordova et al., 1983). These students were often encouraged by American teachers in the Philippines, who were part of the new educational system set up in the country to assist Filipinos in learning the English language and familiarize them with American culture (Agoncillo, 1990). Among the pensionados and self-supporting students that remained in the U.S., few were able to find acceptance in mainstream white communities and often ended up working menial jobs not commensurate with their education level (Nadal, 2011). However, the largest group to arrive in the U.S. during this time was primarily young male laborers under the age of thirty (Lai & Arguelles, 1998; Takaki, 1998). Filipinos from poor rural communities were first recruited in large numbers to work in Hawaiian sugar and pineapple plantations in 1906. These workers were recruited as a response to the loss of Japanese and Chinese laborers, whose immigration had been severely curtailed by exclusionary immigration laws in the late 1800s and early 1900s (Lai & Arguelles, 1998). As demands for cheap labor grew, Filipinos were also recruited to fill jobs on California farms and in the Alaskan fishing industry. Based on these early immigration patterns, Filipinos, like the larger Asian American community, were concentrated in a few states, particularly California and Hawaii. About 45,000 had moved to California from Hawaii and the Philippines during the 1920s, and by 1934, there were 119,470 Filipinos working and living in Hawaii (Lai & Arguelles, 1998). HSE – Social and Education History, 5(2) 145 Third Wave of Immigration By the late 1920s and early 1930s, Filipinos faced increasing hostilities from the white majority population as a result of diminished opportunities stemming from the Great Depression (Cordova et al., 1983; Lai & Arguelles, 1998). Demands to exclude Filipinos from U.S. immigration policies grew as negative public perception of Filipinos as economic threats and social deviants heightened (Chan, 1991; Lai & Arguelles, 1998). The TydingsMcDuffie Act of 1934 transitioned the Philippines from an American territory to commonwealth, guaranteeing independence within 10 years. It stripped Filipinos of their status as nationals, and restricted their annual immigration quota to fifty individuals (Cordova et al., 1983, Takaki, 1998). After the Tydings-McDuffie Act was implemented, societal pressures further called for repatriation of Filipinos living in the U.S. resulting in the Filipino Repatriation Act of 1935 (Lee, 2015). During World War II, Filipinos were recruited to serve in the U.S. military. Though immigration during this time was minimal, exceptions were made to enable the military recruitment of thousands of Philippineborn Filipinos, particularly into the U.S. Navy (Lai & Arguelles, 1998). This became a major influence in later Filipino American immigration patterns. Immediately following the War, Asian immigration restrictions slightly eased with the Luce Celler Act of 1946 raising Filipino immigration quota from 50 to 100 and also allowing them to become naturalized citizens (Kang, 2012). Fourth Wave of Immigration The largest-scale migration from the Philippines occurred after the passage of the 1965 Immigration and Nationality Act, which abolished the national origins quota system and gave preference to family members and certain skilled workers (Bankston, 2006; Pew Research Center, 2013). Like their earlier counterparts, many Filipinos immigrated to the United States seeking better employment opportunities, though many also came to escape political persecutions under the dictatorship of Ferdinand Marcos (Pew Research Center, 2013). Many Filipinos immigrating to the United States in the 1960s 146 Paik, Choe & Witenstein– Filipinos in U.S. were highly trained professionals primarily recruited to fulfill shortages in fields such as healthcare, science, and engineering. This occupational-based recruitment continued through the years resulting in the stereotype of Filipinos occupying primarily high status professional occupations in science or medicine, particularly nursing (Bankston, 2006). Moreover, with the 1990 Immigration Act limiting the number of visas to family members, more Filipinos have utilized employment-based preferences to enter the U.S. (Lai & Arguelles, 1998), impacting the demographic characteristics of newer immigrants. Furthermore, more than one-fourth of all Filipinos in the U.S. have immigrated since 2000 (Terrazas & Batalova, 2010). Modes of Incorporation, Barriers, & Opportunities for Filipino Americans Using the modes of incorporation theoretical framework (Paik et al., 2014), the following section will review the literature and provide further analysis of the immigration experiences of Filipino Americans. As described in the previous section, the history of Filipinos in America have been influenced by government policies, societal reception, availability of co-ethnic communities, and other demographic factors such as settlement patterns, class and occupational levels, education, and fluency in English. Government Policy Through the four waves of immigration, Filipinos have experienced different levels of support from the U.S. government. Given their unique and long history, they are the only Asian group to experience hostile, indifferent, and receptive policies at various times (Paik et al., 2014). Some of these policies include the Pensionado Act of 1903, Tydings-McDuffie Act of 1934, Filipino Repatriation Act of 1935, and Luce Celler Act of 1946. U.S. recruitment efforts pre- and post-1965 also differed in terms of their overall support and receptivity. As with all newly arrived Asian immigrants in the pre-1965 era, the first HSE – Social and Education History, 5(2) 147 waves of Filipino immigrants was also largely met with hostile policies blocking immigration to the U.S. (Paik et al., 2014). Though some Filipinos came to the Americas during the Spanish colonial era, mass immigration of Filipinos to the U.S. did not occur until after the Philippines became a U.S. territory at the end of the Spanish American War in 1898 (Lai & Arguelles, 1998; Lee, 2015). In light of the immigration restrictions placed on Chinese, Japanese, and Koreans during the late 1800s to early 1900s, Filipinos were welcomed and heavily recruited to work, particularly in the agricultural sector (Takaki, 1998). As previously mentioned, their status as American nationals (prior to the Tydings-McDuffie act of 1934) allowed for relatively free entry into the U.S. though it did not confer other rights reserved for citizens (Lee, 2015). Filipinos found themselves barred from living in certain neighborhoods, unable to own property, and included in various states’ antimiscegenation laws (Lai & Arguelles, 1998; Lee, 2015; Matsouka & Ryujin, 1991). These simultaneous inclusionary and exclusionary government policies highlighted the unequal status of Filipinos in America during the second and third waves of immigration. Despite the fact that Americans in the Philippines touted the U.S. as a place to achieve economic success and encouraged Filipinos to see themselves as part of America, Filipinos arriving to the U.S. during the early 1900s may have found most government policies stacked against them. The Pensionado Act of 1903 was an exception as it was seen as one of the more receptive policies allowing students (pensionados) to study higher education in the U.S. The Philippine government supported the students with the agreement that they would return and work for the colonial government (Cordova et al., 1983; Lai & Arguelles, 1998). Self-supporting students experienced more indifferent (or neutral) support. They were allowed to enter, but they did not receive any support from the U.S. or the Philippines (Cordova et al., 1983). Several hostile policies developed in the 1930’s. The Tydings-McDuffie Act of 1934 (also known as the Philippine Independence Act) was considered a hostile policy. By granting the Philippines its independence from the U.S., Filipinos were effectively barred from entering the country with limited exemptions; an immigration quota up to 50 was allowed per year (Cordova et al., 1983; Takaki, 1998). The Filipino Repatriation Act of 1935 soon called for Filipinos to return to the Philippines, providing oneway support for Filipinos and their U.S. born children (Lee, 2015). The 148 Paik, Choe & Witenstein– Filipinos in U.S. program only lasted for three years and was not successful in eliminating undesirable Filipino immigrants (Lee, 2015). In 1946, the Luce-Celler Act on the surface looked more receptive, but the policy was still hostile in nature as it only slightly raised Filipino immigration from 50 to 100 (Kang, 2012). The fourth wave of immigration saw indifferent (or neutral) and even more receptive policies and support with the implementation of the 1965 Immigration Act lifting the national origins restriction. The newer immigrants were primarily educated professional and technical workers specifically recruited to fill a shortage in fields like healthcare (e.g., nursing, engineering, etc.). However, as positions were filled, restrictions were once again placed on Filipinos and other immigrants from Asia, Latin America and the Caribbean (Lai & Arguelles, 1998). Many Filipinos also experienced downward occupational mobility as they found the validity of their professional degrees and licenses earned in the Philippines unrecognized by U.S. institutions (Chan, 1991; Lai & Arguelles, 1998). Societal Reception The restrictive government policies, particularly in the first half of the twentieth century, were a reflection of prejudiced reception of Filipino Americans that grew with rising economic hardships and the advent of the Great Depression. Initially welcomed as a cheap source of labor, the increasing Filipino American population of the 1920s and ’30s became seen as an economic and social threat as competition for jobs increased. The fight for limited opportunities fueled racist and nativist sentiments calling for the expulsion of Filipinos from the U.S. (Chan, 1991; Lee, 2015). Violent attacks against Filipinos were justified by portrayals of Filipinos as uncivilized criminals and sexual deviants. The Filipino Repatriation Act of 1935 confirmed the unwelcome receptivity during the pre-1965 era (Lee, 2015). By comparison, post-1965 reception has been more neutral. Many Filipino Americans have found economic success due to their overall high professional skills and have been able to comfortably integrate into American society. Their ability to speak English has also been helpful in HSE – Social and Education History, 5(2) 149 their assimilation. However, with the growing influx of Filipinos, immigration trends show more low-/semi-skilled and skilled workers, attracting immigrants who subsequently occupy low to middle wage positions with few opportunities for career advancement (Lai & Arguelles, 1998). Co-ethnic Communities During the pre-1965 era, Filipinos had strong co-ethnic communities as they were highly dependent on each other for support and survival (Paik et al., 2014). As previously stated, early Filipino immigrants to the United States were primarily men recruited to work in the agriculture sector or canneries. In the first waves of Filipino immigration, the gender ratio was greatly imbalanced. There were 2,500 Filipina women compared to the overall population of 42,500 Filipinos living in California in 1930; while in 1934 Hawaii, only 8,952 were women compared to the total population of 119,470 Filipinos (Lai & Arguelles, 1998; Lee, 2015). Recruiters discouraged the immigration of women possibly as a way of discouraging the establishment of permanent Filipino American communities. However, Filipinas in the early twentieth century played an instrumental role in supporting families, building communities, and propagating Filipino culture (Lee, 2015). Because Filipinos were prevented from settling in mainstream American neighborhoods, Filipino communities – “Little Manilas” or “Filipinotowns” – were established in cities like Los Angeles, San Francisco, Seattle, and New York (Paik et al., 2014). These co-ethnic communities had eventually developed into thriving businesses, which catered to the needs of Filipino residents and families for many decades (Lee, 2015). After 1965, Filipino communities have become more dispersed throughout the United States, with many living away from significant Filipino immigrant centers (Portes & Rumbaut, 2001). Fewer Filipino Americans lived in urban ethnic enclaves; many more moved to the suburban areas (Nadal, 2011). Despite the lack of co-ethnic community support, newer Filipino immigrants have been able to adapt more readily to American society due to the continued political, economic, and social 150 Paik, Choe & Witenstein– Filipinos in U.S. influence of the U.S. in the Philippines in concert with their ability to speak English and their typically high levels of education and professional skills (Espiritu, 1996). These skills have allowed them to eventually rely less on their own ethnic networks for support (Paik et al., 2014). However, Filipinos still continue to have organizations to help advance the welfare of Filipinos and Filipino Americans (Espiritu, 2003). Many of these organizations help to maintain social and economic ties to the Philippines, particularly to regional hometowns; other organizations promote Filipino culture, especially amongst children of immigrants and later generations, and offer a range of social and cultural activities (Espiritu, 2003; Reisch, 2008). Through these organizations, Filipino Americans primarily derive co-ethnic social and cultural support. Other Barriers and Opportunities Time of arrival has clearly impacted the types of barriers and opportunities encountered by Filipino immigrants throughout the four waves of immigration. Filipinos arriving to the U.S. at the turn of the twentieth century found easy entrance to the U.S. due to their status as nationals. Unfortunately, they encountered a hostile and racist social climate, which at times led to violent encounters like the Watsonville, California riots in 1928 which led to murder (Chan, 1991; Lee, 2015). Severe immigration quotas curtailed opportunities until the passage of the 1965 Immigration Act when the restrictions were lifted. Compared to second and third wave immigrants, many recent Filipino immigrants were primarily recruited to work in occupations, such as professional and technical fields. As a result, they have been characterized as highly educated and arriving to the U.S. with higher socioeconomic status (Bankston, 2006). As for their location and settlement areas, Filipino Americans continue to be concentrated in the West, with the highest proportion (43%) living in California (Hoeffel et al., 2012; Pew Research Center, 2013). Lastly, the combination of having been educated in an American-style educational system and familiarity with the English language enables today’s Filipino immigrants to be less reliant on the HSE – Social and Education History, 5(2) 151 support of co-ethnic communities for jobs or other socioeconomic resources (Lee, 2015; Portes & Rumbaut, 2001). From Historical Immigration to Present-Day Communities and Educational Experiences Filipino immigrants’ experiences with American government policies, societal reception, and co-ethnic communities were strongly influenced by other barriers and opportunities. An examination of the waves of Filipino immigration to the U.S. reveals the significant impact that time of arrival had on these factors. Filipinos who arrived prior to 1965 were primarily laborers with little education who were recruited to work on farms and canneries along the West Coast. They were negatively received by mainstream society, which viewed them as economic and social threats. Government policies intensified these experiences by enacting a series of barriers ultimately leading to a strict restriction of immigrants from the Philippines with the passage of the Tydings-McDuffie Act in 1934. The Luce Celler Act in 1946 appeared to be more receptive, but still had limitations and controlled Filipino immigration and naturalization. These restrictive government policies alongside the negative societal prejudice severely limited economic and social mobility. While most policies were hostile, neutral and more receptive policies were offered to Filipino students to study in the U.S., including governmentsupported students as well as self-supported students. With immigration restrictions lifted, post-1965 immigrants also encountered much more neutral and receptive government support. Often recruited to work in professional and technical fields, newer immigrants generally came from higher socioeconomic brackets and had higher levels of education than their predecessors. However, these immigrants encountered employment barriers such as new licensure requirements, extensive recertification processes, or workplace racism (Lee, 2015). As a result, many professionals arriving in the U.S. experienced downward occupational mobility, such as Philippinetrained nurses who found themselves working as nurse’s aides or lab technicians rather than nurses (Espiritu & Wolf, 2001). Many immigrants take supplementary courses or pursue additional degrees in order to meet 152 Paik, Choe & Witenstein– Filipinos in U.S. licensure requirements to work in their fields despite extensive training prior to arriving in the U.S. Despite the dispersed nature of Filipino co-ethnic communities, they were historically a significant source of support for newly arrived Filipinos. Particularly in the years prior to 1965, the co-ethnic community often served not only as a space of safety against the hostile dominant society, but provided other important resources such as job leads, spaces for social and cultural interactions, or access to goods and services that were typically closed to Filipino immigrants outside of these ethnic enclaves (Lee, 2015). In the fourth wave of immigration, increased professional skills, higher education, and ability to speak English enabled newer Filipino immigrants to enter dominant culture neighborhoods and the labor market with more ease. The less hostile societal reception mitigated the need for the economic and social buffer previously provided by co-ethnic communities; post-1965 immigrants were able to acculturate at a much quicker pace and became more dispersed over time (Portes & Rumbaut, 2001). New immigrants arriving from countries like the Philippines often move to places near family, friends or other co-ethnics, helping to establish urban ethnic communities and ethnoburbs in several states. The Filipino American population resides throughout the United States, with the vast majority residing in the West (Lai & Arguelles, 1998). The historical legacy of labor recruitment in California and Hawaii helped to establish large Filipino American communities and networks that exist to this day. A large Filipino community also developed in Illinois, partly due to the communities established by early pensionados studying in places like the University of Illinois, but also thanks to the military (naval) recruitment connection which saw many Filipinos settle in places like Chicago, San Diego, and Norfolk (Lai & Arguelles, 1998; Pak, Maramba, & Hernandez, 2014; Yang, 2011). Job prospects within the tourist and retail industry in states like Nevada and New Jersey, have also attracted many newer immigrants seeking better economic opportunities (Lai & Arguelles, 1998). Immigrants tend to reflect the socioeconomic and educational backgrounds of their sponsors; therefore, the constant influx of individuals ranging from low-wage job workers to professionals account for the economic diversity of the current Filipino American population (Espiritu & Wolf, 2001; Lai & Arguelles, 1998). Changes in the demographic HSE – Social and Education History, 5(2) 153 composition of Filipino immigrants over the four waves of immigration has characterized today’s Filipinos as having higher levels of assimilation, acculturation, and socioeconomic status compared to other immigrants (Bankston, 2006; Espiritu & Wolf, 2001; Pew Research Center, 2013; Wolf, 1997). Not surprisingly, Filipinos naturalize at a much higher rate than the US average and have the highest percentage of naturalized immigrants among Asian American groups (Reeves & Bennett, 2004; U.S. Census Bureau, 2000). A long history and significant presence in the United States has not shielded Filipinos from facing continued societal prejudices such as being perceived as perpetual foreigners (Pak et al., 2014). English proficiency, a long history of exposure to American culture, and ability to adapt economically have not necessarily helped with acceptance into mainstream society. Second generation Filipino Americans have reported lower levels of self-esteem and higher levels of depression compared to other immigrant groups (Espiritu & Wolf, 2001). Filipino American youth still straddle two worlds, which inevitably influence their social, emotional, and academic skills. Outcomes such as education, ethnic identity, acculturation, and related issues need to be further explored (Paik et al., 2014). Despite the overall high educational attainment levels, research highlighting disparities in educational progress do exist particularly for later generations of Filipino Americans (Espiritu & Wolf, 2001; Museus & Maramba, 2011). Overall, immigrant children tend to perform better than children of native-born Americans; therefore, levels of acculturation may have an impact on later educational achievements (Portes & Rumbaut, 1990, 2001). A study by Eng et al. (2008) found a negative relationship between higher levels of acculturation and poorer academic performance among Filipino American adolescents, signaling a need for further examination of this phenomenon. Like many students of color, Filipino Americans also experience barriers to success in the postsecondary sector (Buenavista, Jayakumar, & Misa-Escalante, 2009). Postsecondary experiences of Filipino Americans continue to be mixed. Often categorized with other Asian American groups, Filipinos are sometimes considered a model minority and are excluded from support programs in colleges and universities. This label has a negative effect due to 154 Paik, Choe & Witenstein– Filipinos in U.S. the fact that it masks challenges and risk factors experienced by students in need of support. Filipino youth are not devoid of such experiences and more research is needed to understand why educational levels vary across students and the needs to support this group. Conclusion and Implications A review of the historical context of immigration among Filipino Americans offers a clearer understanding of their social, economic, and educational outcomes and mobility. The adapted modes of incorporation (Paik et al., 2014) provides context for these experiences through an examination of the government policies, societal reception, co-ethnic communities, and other barriers and opportunities experienced by Filipino Americans through four waves of immigration. Historically, the Filipino community has undergone a long and complicated history in the U.S., which have undoubtedly influenced their co-ethnic communities over time. Facing extremely negative attitudes and mostly exclusionary policies in the early twentieth century to mostly neutral positions after 1965, the Filipino American community has adapted and persevered. Though generations of Filipino Americans had established communities in the U.S. through the first three waves of immigration, the most significant influx of Filipino immigrant came in the fourth wave after the passage of the Immigration and Nationality Act of 1965. Many post-1965 immigrants initially arrived to address shortages in professional fields such as the healthcare industry. Additionally, many also utilized the family reunification clause of the Immigration Act to reunite with family members living in the U.S. and to pursue better economic opportunities unavailable in the Philippines. These four waves of immigration contribute to the current diversity of Filipino Americans and reveal generational differences. Future researchers may want to further explore the implications for generational status on social, economic, and educational outcomes as it may lead to a greater understanding of the mixed nature of achievement among Filipino Americans. Furthermore, rates of acculturation may vary not only among immigrant groups but among families. Family dynamics may be influenced HSE – Social and Education History, 5(2) 155 by differing rates of acculturation between parents and children, thereby impacting school experiences and educational outcomes. Generational status and levels of acculturation may also influence ethnic identity development among successive generations of immigrants and their children. However, it is unclear whether the continued influx of new immigrants would reinforce a transnational identity and strong ethnic community with sustained ties to the Philippines, or if ties would lessen and Filipino Americans become more dispersed over time. It may also be prudent to ask the question “what influence would these generational changes and levels of acculturation have on later achievement outcomes?” These considerations are beyond the scope of this paper, but more research is needed to disaggregate these findings. This article provides a comprehensive model for analyzing historical experiences of the Filipino American experience. Exploring a community’s historical context is a critical exercise that provides a greater understanding of immigrant group experiences and later achievement outcomes. The diverse composition of American schools and communities warrant a greater understanding of students’ familial and ethnic backgrounds. Social and historical contextual factors may help educators to work more effectively with parents, community members, and other key stakeholders to create positive home-school-neighborhood partnerships for student educational success. Researchers and policymakers also benefit from understanding historical context because it may help anticipate and effectively address potential issues, as well as predict possible outcomes and trends. As evidenced by the experiences of Filipino Americans over four waves of immigration, the experiences of immigrant families and communities are not stagnant and evolve over time. Consequently, history provides a long-lens perspective to understanding and addressing current social and educational concerns. References Agbayani-Siewert, P. (1994). Filipino American culture and family: Guidelines for practitioners. Families in Society, 75, 429–438. Retrieved from http://psycnet.apa.org/psycinfo/1995-10914-001 156 Paik, Choe & Witenstein– Filipinos in U.S. Agoncillo, T. A. (1990). History of the Filipino people (8th ed.). Quezon City: Garotech. Bankston, C. L. (2006). Filipino Americans. In P. G. Min (Ed.), Asian Americans: Trends and issues (pp. 180–203). Thousand Oaks, CA: Pine Forge Press. Bonus, R., & Maramba, D. C. (2013). Introduction: Filipino Americans as “Others.” In D. C. Maramba & R. Bonus (Eds.), The “other” students: Filipino Americans, education, and power (pp. 15 – 38). Charlotte, NC: Information Age Publishing, Inc. Buenavista, T. L., Jayakumar, U. M., & Misa-Escalante, K. (2009). Contextualizing Asian American education through critical race theory: An example of U.S. Pilipino college student experiences. New Directions for Institutional Research, 2009 (142), 69–81. doi: http://doi.org/10.1002/ir.297 Chan, S. (1991). Asian Americans: An interpretive history (Twayne Pub). Boston, MA. Chan, S. (1992). Families with Pilipino roots. In E. W. Lynch & M. J. Hanson (Eds.), Developing Cross-Cultural Competence (pp. 259–300). Baltimore: Paul H. Brookes Publishing Co. Chao, R., & Tseng, V. (2002). Parenting of Asians. In M. H. Bornstein (Ed.), Handbook of Parenting: Volume 4 Social Conditions and Applied Parenting (Second, p. 59093). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Cimmarusti, R. A. (1992). Family preservation practice based upon multisystems approach. Child Welfare, 71(3), 241–255. Retrieved from http://eric.ed.gov/?id=EJ444916 Cordova, F., Cordova, D. L., & Acena, A. A. (1983). Filipinos, forgotten Asian Americans: A pictorial essay, 1764-circa-1963. Dubuque: Kendall/Hunt Publishing Company. Eng, S., Kanitkar, K., Cleveland, H. H., Herbert, R., Fischer, J., & Wiersma, J. D. (2008). School achievement differences among Chinese and HSE – Social and Education History, 5(2) 157 Filipino American students: acculturation and the family. Educational Psychology, 28(5), 535–550. doi:10.1080/01443410701861308 Espiritu, Y. L. (1993). Asian American panethnicity: Bridging institutions and identities. Philadelphia: Temple University Press. Espiritu, Y. L. (1996). Colonial oppression, labour importation, and group formation: Filipinos in the United States. Ethnic and Racial Studies, 19(1), 29–48. doi:10.1080/01419870.1996.9993897 Espiritu, Y. L. (2003). Home bound Filipino American lives across cultures, communities, and countries (Vols. 1–1 online resource (xi, 1–271 pages)). Berkeley: University of California Press. Retrieved from EBSCOhost http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nleb k&db=nlabk&AN=90500 Espiritu, Y. L., & Wolf, D. L. (2001). The paradox of assimilation: Children of Filipino immigrants in San Diego. In R. G. Rumbaut & A. Portes (Eds.), Ethnicities: Children of immigrant America (pp. 157–186). Berkeley, CA: University of California Press. Hoeffel, E. M., Rastogi, S., Kim, M. O., & Hasan, S. (2012). The Asian population: 2010 (No. March). U.S. Department of Commerce Economics and Statistics Administration, U.S. Census Bureau. Retrieved from http://www.census.gov/prod/cen2010/briefs/c2010br-11.pdf Kang, H. K. (2012). Re-imagining citizenship, re-imagining social work: US immigration policies and social work practice in the era of AZ SB1070. Advances in Social Work, 13(3), 510–526. Retrieved from https://journals.iupui.edu/index.php/advancesinsocialwork/article/view/2 057 Lai, E. Y. P., & Arguelles, D. (Eds.). (1998). The New Face of Asian Pacific America: Numbers, Diversity, and Change in the 21st Century (Paperback edition). San Francisco: UCLA Asian American Studies Center Press. Lee, E. (2015). The Making of Asian America: A History. New York: Simon and Schuster. Lee, S. J. (2006). Additional complexities: social class, ethnicity, generation, and gender in Asian American student experiences. Race, Ethnicity and Education, 9(1), 17–28. doi:10.1080/13613320500490630 158 Paik, Choe & Witenstein– Filipinos in U.S. Li, W. (2009). Ethnoburb the new ethnic community in urban America (Vols. 1–1 online resource (xvii, 1–214 pages). Honolulu: University of Hawai’i Press. Retrieved from ebrary http://site.ebrary.com/id/10388468 Matsouka, J. K., & Ryujin, D. H. (1991). Asian American immigrants: A comparison of the Chinese, Japanese, and Filipinos. J. Soc. & Soc. Welfare, 18, 123. Retrieved from http://scholarworks.wmich.edu/cgi/viewcontent.cgi?article=1992&contex t=jssw Museus, S. D., & Maramba, D. C. (2011). The impact of culture on Filipino American students’ sense of belonging. The Review of Higher Education, 34(2), 231–258. Nadal, K. L. (2011). Filipino American psychology a handbook of theory, research, and clinical practice (Vols. 1–1 online resource (xviii, 1–318 pages): illustrations). Hoboken, N.J.: Wiley. Retrieved from 123Library http://www.123library.org/book_details/?id=16722 National Federation of Filipino American Associations. (2011). Retrieved from http://naffaausa.org/us-2010census-filipinos-in-the-u-s-increasedby-38-nevada-has-fastest-growing-population/ Ogilvie, A. B. (2008). Filipino-American K-12 public school students: A national survey. Washington, DC: The National Federation of Filipino American Associations. Okamura, J. Y., & Agbayani, A. R. (1997). Pamantasan: Filipino American higher education. In M. P. P. Root (Ed.), Filipino Americans: Transformation and identity (pp. 183–197). Thousand Oaks, CA: Sage. Paik, S. J., Kula, S. M., Saito, L. E., Rahman, Z., & Witenstein, M. A. (2014). Historical Perspectives on Diverse Asian American Communities: Immigration, Incorporation, and Education. Teachers College Record, 116(11). Retrieved from https://www.tcrecord.org/library/abstract.asp?contentid=17512 Pak, Y. K., Maramba, D. C., & Hernandez, X. J. (2014). Asian Americans in Higher Education: Charting New Realities: AEHE Volume 40, Number 1. John Wiley & Sons. Pew Research Center. (2013). The rise of Asian Americans. Retrieved from http://www.pewsocialtrends.org/2012/06/19/the-rise-of-asian-americans/ Portes, A., Fernández-Kelly, P., & Haller, W. (2009). The adaptation of the immigrant second generation in America: Theoretical overview and HSE – Social and Education History, 5(2) 159 recent evidence. Journal of Ethnic and Migration Studies, 35(7), 1077– 1104. doi: 10.1080/13691830903006127 Portes, A., & Rumbaut, R. G. (1990). Immigrant America: a portrait. Berkeley: University of California Press. Retrieved from http://catdir.loc.gov/catdir/toc/fy0709/89020444.html Portes, A., & Rumbaut, R. G. (2001). Legacies: the story of the immigrant second generation. Berkeley: University of California Press. Retrieved from http://site.ebrary.com/id/10058840 Reeves, T. J., & Bennett, C. E. (2004). We the people: Asians in the United States (Vol. 17). Washington, DC: U.S. Census Bureau. Retrieved from http://www.m.advancingjustice-aajc.org/sites/aajc/files/censr-17_0.pdf Reisch, M. (2008). From Melting Pot to Multiculturalism: The Impact of Racial and Ethnic Diversity on Social Work and Social Justice in the USA. British Journal of Social Work, 38(4), 788–804. doi:10.1093/bjsw/bcn001 Salvador, D. S., Omizo, M. M., & Kim, B. S. K. (1997). Bayanihan: Providing effective counseling strategies with children of Filipino ancestry. Journal of Multicultural Counseling and Development, 25(3), 201–209. doi: 10.1002/j.2161-1912.1997.tb00330.x Solorzano, D. G., & Villalpando, O. (1998). Critical race theory, marginality, and the experience of students of color in higher education. Sociology of Education: Emerging Perspectives, 211–224. Takaki, R. (1989). Dollar a day, dime a dance. In Strangers from a Different Shore: A History of Asian Americans (pp. 315–354). New York, NY: Penguin Books. Takaki, R. (1998). Strangers From a Different Shore: A History of Asian Americans. Boston, MA: Little, Brown and Company. Terrazas, A., & Batalova, J. (2010). Filipino immigrants in the United States (pp. 4–4). Washington, DC: Migration Policy Institute. Retrieved from http://www.adapttech.it/old/files/document/6505MPI_IMMIGRAZIONE. pdf U.S. Census Bureau. (2000). Summary files 1 through 4 tables. Available at American Factfinder: Washington, DC: U.S. Census Bureau. Retrieved from http://factfinder2.census.gov. Washington, DC: U.S. Census Bureau 160 Paik, Choe & Witenstein– Filipinos in U.S. Wolf, D. L. (1997). Family Secrets: Transnational Struggles among Children of Filipino Immigrants. Sociological Perspectives, 40(3), 457–482. doi: 10.2307/1389452 Yang, P. Q. (2011). Asian immigration to the United States. Cambridge, UK; Polity. Zhou, M., & Bankston III, C. L. (1998). Straddling the gap: Bicultural conflicts and gender role changes. In Growing up American: How Vietnamese children adapt to life in the United States (pp. 160–184). New York: Russell Sage Foundation. Zhou, M., & Xiong, Y. S. (2005). The multifaceted American experiences of the children of Asian immigrants: Lessons for segmented assimilation. Ethnic and Racial Studies, 28(6), 1119–1152. doi: 10.1080/01419870500224455 Susan J. Paik: Associate Professor. School of Educational Studies. Claremont Graduate University (USA) Shirlie Mae Mamaril Choe: PhD Candidate. School of Educational Studies. Claremont Graduate University (USA) Matthew A. Witenstein: Clinical Assistant Professor. School of Leadership and Education Sciences, Department of Leadership. University of San Diego (USA) Contact Address: susan.paik@cgu.edu
https://openalex.org/W4312527118	https://convergenceseditorial.com.br/index.php/revistafisiologia/article/download/4878/7760	English	null	Effects of different blood flow restriction pressure levels on muscular hemodynamics	Revista brasileira de fisiologia do exercício/Revista Brasileira de Fisiologia do Exercício	2,022	cc-by	8,131	How to cite: Carvalho RF, Gomes PSC, Fernandes-Júnior ML, Meirelles CM. Effects of different blood flow restriction pressure levels on muscular hemodynamics. Rev Bras Fisiol Exerc 2021;20(6):604-617. doi: 10.33233/rbfex.v20i6.4878 How to cite: Carvalho RF, Gomes PSC, Fernandes-Júnior ML, Meirelles CM. Effects of different blood flow restriction pressure levels on muscular hemodynamics. Rev Bras Fisiol Exerc 2021;20(6):604-617. doi: 10.33233/rbfex.v20i6.4878 604 Efeito de diferentes níveis de restrição de fluxo sanguíneo sobre a hemodinâmica muscular Effects of different blood flow restriction pressure levels on muscular hemodynamics Ramon Franco Carvalho1 , Paulo Sergio Chagas Gomes1 ,Márcio Lopes Fernandes Júnior2 , Claudia Mello Meirelles3 1. Instituto de Educação Física e Desportos, Universidade do Estado do Rio de Janeiro, RJ, Brazil 2. Universidade Estácio de Sá, Campus Duque de Caxias 2, Duque de Caxias, RJ, Brazil 3. Seção de Pesquisa e Extensão, Escola de Educação Física do Exército, Rio de Janeiro, RJ, Brazil on muscular hemodynamics. Rev Bras Fisiol Exerc 2021;20(6):604-617. doi: 10.33233/rbfex.v20i6.4878 Original article Revista Brasileira de Fisiologia do Exercício 2675-1372 Ex Efeito de diferentes níveis de restrição de fluxo sanguíneo sobre a hemodinâmica muscular Ramon Franco Carvalho1 , Paulo Sergio Chagas Gomes1 ,Márcio Lopes Fernandes Júnior2 , Claudia Mello Meirelles3 Original article Effects of different blood flow restriction pressure levels on muscular hemodynamics ABSTRACT d i Introduction: Resistance exercise with blood flow restriction (BFR) is an effective method to promote muscle strength gains and hypertrophy. However, little is known about the effects of different BFR levels on hemodynamic responses. Objective: To verify whether the different blood flow restriction pressures applied to the upper limb cause acute changes in vascular microcirculation in young, healthy male adults. Methods: Ten young male visited the laboratory on four occasions. In the first visit, after 10- min rest in supine position, the brachial artery occlusion pressure (AOP) was identified with a Doppler ultrasound. Thereafter, the participants were submitted to a protocol consisting of 1 min for baseline measurements, 2 min of BFR, and 2 min after cuff deflation. It was used a cuff placed on the proximal portion of the forearm and inflated with pressures equivalents to 30% (30BFR), 50% (50BFR) 80% (80BFR), or 100% (100BFR) of the AOP in a random order in separate days. Measurements of tissue saturation index (TSI), oxyhemoglobin, deoxyhemoglobin, and total hemoglobin were collected continuously using near- infrared spectrometry. Results: A two-way ANOVA with repeated measures demonstrated: 1) a significant decrease in TSI in all conditions, with higher decay in 100BFR; 2) a significant increase in oxyhemoglobin in all conditions, but 100BFR; 3) a similar increase in deoxyhemoglobin in all conditions; 4) a significant increase in total hemoglobin in all conditions, mainly in both 30BFR and 50BFR. Conclusion: The relative pressures adopted demonstrated that the hemodynamic changes do not occur linearly with the pressure level imposed by the inflated cuff. Keywords: spectroscopy, near-infrared; vascular closure devices; resistance training. Correspondence: Paulo Sergio Chagas Gomes, PhD, Universidade do Estado do Rio de Janeiro, Instituto de Educação Física e Desportos, Rua São Francisco Xavier, 524, 8o Andar, Bloco F, Sala 8104, Maracanã, 20550-900 Rio de Janeiro RJ. paulo.gomes@uerj.br Revista Brasileira de Fisiologia do Exercício ISSN Online: 2675-1372 RBFEx ISSN Online: 2675-1372 RBFEx Introduction Resistance exercise (RE) with restricted blood flow (BFR) is an effective method to promote strength gains [1-3] as well as muscle hypertrophy [2,4,5]. This method consists of using an inflated cuff at the proximal extremity of the limbs during the performance of an activity with relatively low resistance overload, ranging from 10 to 50% of 1RM [6-8]. The purpose of restricting the influx of arterial blood to the limb is to cause a more significant metabolic stress and stimulate the mechanisms of muscle hypertro- phy, such as additional recruitment of motor units, cell swelling, the release of ana- bolic hormones, altered production of myokines, and reactive oxygen species [9-11]. Although the magnitude of the responses to strength gains are lower than the ones obtained with traditional high resistive loads strength training routines, resistance training with blood flow restriction (BFR) may be a more appropriate strategy in populations that are unable to mobilize high overload, such as the elderly [12], and people recovering from musculoskeletal injury or surgery [13]. When inflating a cuff in the proximal region of the upper or lower limb, ve- nous blood is easily occluded, generating blood storage in the venules and thus pre- venting the removal of metabolites from muscle contraction. This procedure can pre- vent venous blood return to the limb but still allow the entry of arterial blood, even if in a limited way [14]. Understanding the impact of different percentages of blood flow restriction on muscle hemodynamics can help clarify the best relationship between metabolic stress and the lowest health risk associated with blood flow restriction [14]. Also, the scientific literature indicates that high-pressure loads promote a higher level of discomfort [15]. Previous studies [16,17] carried out in healthy young subjects at rest observed that the reduction in blood flow occurs in a staggered and non-linear manner due to increased pressure load. Using ultrasound in the Doppler mode, Mouser et al. [17] observed that 10% of the pressure at the artery occlusion pressure (AOP) applied by a cuff is enough to significantly reduce absolute and relative blood flow speed in the brachial artery when compared to resting condition. This flow reduction remained similar until 40% of the artery occlusion pressure (AOP) when a further significant drop in blood flow was observed and remained up to 80% (absolute blood flow) or 90% (relative blood flow) when the last phase of fall occurred. RESUMO d ã Introdução: O exercício contrarresistência com restrição do fluxo sanguíneo (RFS) é um método eficaz para ganho de força e hipertrofia muscular. No entanto, pouco se sabe sobre os efeitos dos diferentes níveis de RFS nas respostas hemodinâmicas. Objetivo: Verificar se as diferentes pressões de restrição ao fluxo sanguíneo aplicadas no membro superior causam alterações na microcirculação vascular em adul- tos jovens saudáveis do sexo masculino. Métodos: Dez jovens do sexo masculino visitaram o laboratório em quatro ocasiões. Na primeira visita, após 10 min de repouso em decúbito dorsal, a pressão de oclusão da artéria braquial (POA) foi identificada através de ultrassom com Doppler. Posteriormente, os parti- cipantes foram submetidos a um protocolo que consistia de 1 min para as medidas basais, 2 min de RFS e 2 min após a liberação da restrição sanguínea. Foi utilizado um manguito colocado na porção proxi- mal do antebraço e inflado com pressões equivalentes a 30% (30RFS), 50% (50RFS) 80% (80RFS) ou 100% (100RFS) do POA em ordem aleatória em dias separados. As medições do índice de saturação do tecido (IST), oxiemoglobina, desoxihemoglobina e hemoglobina total foram coletadas continuamente usando espectrometria de infravermelho próximo. Resultados: Uma ANOVA de duas vias com medidas repetidas demonstrou 1) uma diminuição significativa no IST em todas as condições, com maior queda em 100RFS; 2) um aumento significativo na oxihemoglobina em todas as condições, exceto 100RFS; 3) um aumento semelhante na desoxihemoglobina em todas as condições; 4) um aumento significativo na hemoglobina total em todas as condições, principalmente em 30RFS e 50RFS. Conclusão: As pressões relativas adotadas demonstraram que as alterações hemodinâmicas não ocorrem linearmente com o nível de pressão impos- to pelo manguito insuflado. alavras-chave: espectroscopia de luz próxima ao infravermelho; dispositivos de oclusão vascular; treinamento e força Palavras-chave: espectroscopia de luz próxima ao infravermelho; dispositivos de oclusão vascular; treinamento de força. Received: August 11, 2021; Accepted: December 2, 2021. Received: August 11, 2021; Accepted: December 2, 2021. 605 Rev Bras Fisiol Exerc 2021;20(6):604-617 Introduction Introduction Despite the importan- ce of this finding in blood flow, the study, as mentioned earlier, did not observe the impact on hypoxia. The literature has shown that the intracellular deviation of blood plasma and cellular hypoxia generated by flow restriction significantly influences the mechanisms associated with increased muscle strength and hypertrophy [18]. Muscle hemodynamics measurements can also be performed by near-infrared spectroscopy (NIRS), which is widely used in research to monitor acute and chronic muscle perfusion changes under different settings [19]. 606 Rev Bras Fisiol Exerc 2021;20(6):604-617 Kilgas et al. [20] showed that 30 seconds under BFR did not change muscle hemodynamics at pressures lower than 60% AOP, assessed by a NIRS probe placed on the forearm of ten healthy men. Less is known about higher periods of BFR, as em- ployed in typical resistance exercise protocols. Blood flow reduction seems to occur staggered and not linear or parallel by increased pressure levels exerted externally by a cuff. With this shortcoming in mind, it is necessary to identify the impact of different pressure level ranges on local he- modynamic responses, especially in cell hypoxia. This knowledge may contribute to a better understanding of the physiological responses, allowing a safer and more efficient prescription method. Thus, the present study aimed to verify whether the different blood flow res- triction pressures applied to the upper limb cause acute changes in vascular micro- circulation in young, healthy male adults. Methods Study sample Ten young college male students volunteered for the present study (age: 26 ± 5 years, biceps skinfold: 3.4 ± 1.1 mm, systolic blood pressure: 122.9 ± 7.1 mmHg, dias- tolic blood pressure: 81, 4 ± 7.5 mmHg, resting heart rate: 69.3 ± 5.7 bpm; body mass index: 24.7 ± 1.1 kg/m²). All participants were normotensive and healthy based on the Physical Activity Readiness Questionnaire (Par-Q) evaluation, and nobody was involved in any systematic physical training practice in the last six months. All of them signed the informed consent form before starting the tests. The Research Ethics Committee of the President Antônio Carlos University approved this study (CAAE: 83463517.7.0000.5156), based on the principles of the Declaration of Helsinki. Determination of brachial artery blood flow occlusion pressure (AOP) The AOP was determined with an ultrasound scanner (Logic e, General Elec- tric - GE Healthcare, Milwaukee, WI, USA) equipped with Doppler. A 10 cm-wide cuff was positioned at the most proximal portion of the right arm. A 40 mm-ultrasound transducer was placed on the anteromedial face of the right arm. The transducer was positioned perpendicular to its axis, 5 to 10 cm above the antecubital fold. The cuff pressure was progressively slowly released until the first sign of flow was observed in the brachial artery procedure was repeated two or three more times to confirm the pressure level of the cuff, operationally defined as the AOP. This procedure was per- formed in all visits and lasted approximately one minute. Study design The study was characterized by a randomized controlled trial model, and par- ticipants attended the research laboratory on four separate occasions with two to se- ven-day between trial intervals. All participants were instructed not to consume any drink or food, like caffeine and alcohol, that would affect hemodynamic responses and not to practice any physical activity 24 hours before the test. Also, all visits took place within the same time of day, with a maximum variation of one hour more or less to avoid the effect of the circadian cycle on blood pressure responses. At each visit, participants were tested under one of the four experimental treatments. The subjects were submitted to different percentages of AOP: 30%, 50%, 80%, and 100% (30BFR, 50BFR, 80BFR, and 100BFR, respectively). Upon reaching the laboratory, the volunteers rested on a stretcher in the supi- ne position for 10 minutes. At the end of this period, the pressure level representing the AOP was identified using ultrasound equipment in Doppler mode. This procedu- re took between 40 to 60 seconds. After 20 minutes of recovery in the supine position, 607 Rev Bras Fisiol Exerc 2021;20(6):604-617 the subjects were submitted, in random order, to one of the experimental treatments, in order to have the hemodynamic variables monitored for five minutes, as follows: one minute to obtain baseline measurements, two minutes with the cuff inflated in the proximal portion of the right upper limb and two minutes of observation with the cuff deflated. The NIRS measurements O2Hb, HHb, tHb, and TSI, were collected continuously during the five min test procedure. Figure 1 shows the procedures per- formed. Caption = BFR - blood flow restriction; the arrows indicate the start and/or end of each phase (Base- line, BFR and flow released) Figure 1 - Experimental design. Treatment conditions (30%, 50%, 80% and 100% of brachial artery blood flow occlusion pressure) were randomly assigned for each subject on = BFR - blood flow restriction; the arrows indicate the start and/or end of each phase (Bas BFR and flow released) f Caption = BFR - blood flow restriction; the arrows indicate the start and/or end of each phase (Base- line, BFR and flow released) Figure 1 - Experimental design. Treatment conditions (30%, 50%, 80% and 100% of brachial artery blood flow occlusion pressure) were randomly assigned for each subject ) - Experimental design. Study design Treatment conditions (30%, 50%, 80% and 100% of brachial artery blood lusion pressure) were randomly assigned for each subject , ) Figure 1 - Experimental design. Treatment conditions (30%, 50%, 80% and 100% of bra flow occlusion pressure) were randomly assigned for each subject Determination of brachial artery blood flow occlusion pressure (AOP) Near infrared spectroscopy (NIRS) measurements Monitoring of the muscle hemodynamics was performed using near-infrared spectroscopy (NIRS). This is a non-invasive optical technique that measures changes in the relative concentration of oxyhemoglobin (O2Hb) and deoxyhemoglobin (HHb) in arterioles, venules, and capillaries [19]. The electrons of the hemoglobin chromo- phores can absorb light near the infrared region at different peaks according to the presence or absence of oxygen bound to the hemoglobin molecule [21]. In this way, using the Lambert-Beer law, it is possible to calculate changes in the concentration of the chromophores of interest, such as O2Hb or HHb. Changes in the concentrations of oxygenated hemoglobin (O2Hb), deoxygenated (HHb), total hemoglobin (tHb = 608 Rev Bras Fisiol Exerc 2021;20(6):604-617 O2Hb + HHb), and the tissue saturation index (TSI) were measured continuously, in random order, for 5 min in all experimental conditions (30BFR, 50BFR, 80BFR and 100BFR), using a near-infrared continuous-wave spectrometer (NIRS; PortaMon, Ar- tinis Medical Systems BV, Zetten, Netherlands). The total hemoglobin concentration (tHb) was obtained by adding the concentration of O2Hb with HHb and is an indirect indicator of blood volume. The TSI is a direct indication of the percentage of oxy- genated hemoglobin and was obtained through the following equation: TSI (%) = (O2Hb/tHb) x 100. The sensor was positioned in the most distal position on the belly of the bi- ceps brachii muscle. The sensor was surrounded by a plastic film, attached to the skin by tape, and covered with a dark towel to avoid distortion of the signal caused by sweat and ambient light. Data were collected using dedicated OxySoft software version (OxySoft Ver. 2.1.1-2.1.6 Artinis Medical Systems BV, Zetten, Netherlands) with a sampling frequency of 10 Hz. Results Post-hoc analysis identified the study’s power at 0.83. For this result, an effect size of 0.40 was considered, an error α: 0.05, for a sample size of 10 participants, in four conditions (of blood flow restriction), three measures repeated over time (ba- seline, restriction of blood flow and after the release of arterial flow), a correlation between repeated measures of 0.8 and non-sphericity correction of 1. After the blood flow was released, a significant difference was observed in TSI between the 30BFR and 50BFR (p = 0.012) and 30BFR and 100BFR (p = 0.006). The tHb showed significant difference between the 30BFR and 50BFR (p = 0.037), between 50BFR and 100BFR (p = 0.002), and between 80BFR and 100BFR (p = 0.007). In addi- tion, a significant difference was observed in O2Hb between the 30BFR and 100BFR (p = 0.000), 50BFR and 100BFR (p = 0.000) and between 80BFR and 100BFR (p = 0.007) conditions. Finally, HHb showed significant difference between 30BFR and 50BFR (p = 0.032), as well as between 30BFR and 80BFR (p = 0.007). Figure 2 shows the results of each dependent variable evaluated in this study. For comparisons within groups, there was a difference in O2Hb between all conditions (baseline, blood flow restric- tion, and blood flow released), except for 100BFR (baseline vs. blood flow restriction; p = 0.999) and 30BFR (baseline vs. blood flow released; p = 0.699). For tHB there was a difference for all combinations, except 50BFR (p = 0.991) and 80BFR (p = 0.995) betwe- en baseline and blood flow released. In HHb conditions, only there was no difference between baseline and blood flow released to 30BFR (p = 0.258) and 100BFR (p =0.225). Finally, there was a significant difference in all conditions over time to TSI. The effect size varied from very small to huge in the most diverse combina- tions between groups, according to Sawilowsky’s classification [25]. The following Tables I to III show the results of all effect sizes related to TSI, tHb, O2Hb, and HHb. In Table I, it was possible to observe that the most significant clinical impacts be- tween the measurements obtained during and after blood flow restriction occurred in the TSI and HHb measurements in all restriction conditions. In the tHB variable, the 100BFR condition had the lowest clinical impact, while in the O2Hb variable, the 50BFR and 80BFR conditions had a huge effect. Statistical analysis NIRS variables values at baseline were obtained by averaging the 15 s before blood flow restriction. Measurements were obtained at the end of a 2-min period of blood flow restriction and 30 s after deflation of the cuff. All measurements obtained during and after blood flow restriction were normalized by the baseline obtained on the same day to reduce the influence of the measurements collected on different days. After testing the assumptions of normality and sphericity using Shapiro-Wilk and Mauchly tests, respectively, a two-way ANOVA with repeated measures was used to determine a significant interaction difference between treatments and time con- ditions. Where significant F was observed, Sidak’s post hoc test was applied to analy- ze possible differences in the dependent variables among conditions (30BFR, 50BFR, 80BFR, and 100BFR) within each phase (BFR and blood flow release). The level of sig- nificance adopted in this study was 0.05. Also, the effect size (ES) was used to identify the clinical effect through the magnitude of the difference [22,23]. Except for the variable O2Hb in groups 30BFR, 50BFR, and 100BFR during the blood flow restriction phase, TSI in group 30BFR, and tHB in 100BFR, all other varia- bles showed normal distribution. However, the ANOVA test was used in all analyses because it is robust enough to be used even when normality is not observed [24]. The Greenhouse-Geisser correction was used to compare TSI between the conditions due to the violation of sphericity. All analysis were performed using commercially available SPSS statistical sof- tware (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 21.0. Ar- monk, NY: IBM Corp.). 609 Rev Bras Fisiol Exerc 2021;20(6):604-617 Results Table II shows the clinical impact of the difference between groups during the period of blood flow restriction. The 100BFR condition had larger effect sizes than all other conditions for the variables. O2Hb, tHb, and TSI, indicating that this condition is the one that generated the most significant impact on tHb and muscle oxygenation while the cuff was inflated. On the other hand, the impact of changes in tHb and muscle oxygenation between 30BFR and 50BFR were the smallest. Table III shows the clinical impact of the difference between treatments after the period of blood flow restriction. All conditions of blood flow restrictions showed a large effect size between the TSI variable, indicating that each change in the res- triction range causes a great clinical impact on muscle oxygenation. Muscle volume measured indirectly by tHb indicated a little clinical impact on the change observed 610 Rev Bras Fisiol Exerc 2021;20(6 Rev Bras Fisiol Exerc 2021;20(6):604-617 between 50BFR and 80BFR, but the other changes at each restriction range change occurred with greater impact. occurred with greater impact. All values during BFR were statistically different from baseline in each pressure level. All values pos- t-BFR were statistically different from during BFR in each pressure level. 1 = different from 30BFR; 2 = different from 50BFR; 3 = different from 80BFR. All differences for p < 0.05 Figure 2 - Oxyhemoglobin (O2Hb), desoxyhemoglobin (HHb), total hemoglobin (tHB) and tissue sa- turation index (TSI) modifications from baseline at the different body flow restriction pressure levels (30%, 50% 80% and 100% BFR) during blood flow restriction (BFR) and after flow release (Post) es during BFR were statistically different from baseline in each pressure level. All values pos- ere statistically different from during BFR in each pressure level. 1 = different from 30BFR; 2 = t from 50BFR; 3 = different from 80BFR. All differences for p < 0.05 All values during BFR were statistically different from baseline in each pressure level. All values pos- t-BFR were statistically different from during BFR in each pressure level. 1 = different from 30BFR; 2 = different from 50BFR; 3 = different from 80BFR. Results All differences for p < 0.05 Figure 2 - Oxyhemoglobin (O2Hb), desoxyhemoglobin (HHb), total hemoglobin (tHB) and tissue sa- turation index (TSI) modifications from baseline at the different body flow restriction pressure levels (30%, 50% 80% and 100% BFR) during blood flow restriction (BFR) and after flow release (Post) ; p - Oxyhemoglobin (O2Hb), desoxyhemoglobin (HHb), total hemoglobin (tHB) and tissue sa- index (TSI) modifications from baseline at the different body flow restriction pressure levels % 80% and 100% BFR) during blood flow restriction (BFR) and after flow release (Post) Table I - Effect size (ES) of the dependent variables TSI, O2Hb, HHb e tHb for repeated measures (BFR vs Post-BFR) between treatments, based on the criteria proposed by Sawilowsky [25] Table I - Effect size (ES) of the dependent variables TSI, O2Hb, HHb e tHb for repeated measures (BFR vs Post-BFR) between treatments, based on the criteria proposed by Sawilowsky [25] TSI O2Hb HHb tHb ES Classif ES Classif ES Classif ES Classif 30BFR -2.02 Huge 1.86 Very large 3.72 Huge 2.85 Huge 50BFR -3.58 Huge 2.44 Huge 7.85 Huge 6.49 Huge 80BFR -5.09 Huge 2.35 Huge 6.93 Huge 4.85 Huge 100BFR -7.76 Huge -1.67 Very large 4.56 Huge 0.94 Large TSI = Tissue Saturation Index; O2Hb = Oxyhemoglobin; HHb = Deoxyhemoglobin; tHb = Total Hemo- globin; Classif: classification Effect size (ES) of the dependent variables TSI, O2Hb, HHb e tHb for repeated measures (BFR BFR) between treatments, based on the criteria proposed by Sawilowsky [25] 611 Rev Bras Fisiol Exerc 2021;20(6):604-617 Table II - Effect size (ES) of the dependent variables TSI, O2Hb, HHb and tHb for treatment compari- sons during blood flow restriction, based on the criteria proposed by Sawilowsky [25] TSI O2Hb HHb tHb ES Classif ES Classif ES Classif ES Classif 30BFR- 50BFR 0.26 Medium 0.06 Very small -0.95 Large -0.40 Small 30BFR- 80BFR 0.36 Medium -0.01 Very small -0.99 Large -0.46 Small 30BFR- 100BFR 1.55 Very large 1.99 Very large -0.30 Small 1.15 Large 50BFR- 80BFR 0.01 Very small -0.07 Very small -0.14 Very small -0.12 Very small 50BFR- 100BFR 1.25 Very large 1.88 Very large 0.84 Large 1.90 Very large 80BFR- 100BFR 2.05 Huge 1.93 Very large 0.91 Large 1.77 Very large TSI = Tissue Saturation Index; O2Hb = Oxyhemoglobin; HHb = Deoxyhemoglobin; tHb = Total Hemo- globin; Classif: classification Table III - Effect size (ES) of the dependent variables TSI, O2Hb, HHb and tHb for condition compari- sons, after blood flow release, based on the criteria proposed by Sawilowsky [25 TSI O2Hb HHb tHb ES Classif ES Classif ES Classif ES Classif 30BFR- 50BFR -1.85 Very large 0.86 Large 1.28 Very large 1.44 Large 30BFR- 80BFR -1.18 Large 0.57 Medium 1.62 Very large 1.18 Large 30BFR- 100BFR -2.26 Huge -1.32 Very large 0.53 Medium -0.75 Medium 50BFR- 80BFR 0.65 Large -0.37 Small 0.45 Small -0.02 Very small 50BFR- 100BFR -0.79 Large -2.37 Huge -0.27 Small -2.15 Huge 80BFR- 100BFR -1.31 Very large -2.12 Huge -0.55 Medium -1.83 Very large TSI = Tissue Saturation Index; O2Hb = Oxyhemoglobin; HHb = Deoxyhemoglobin; tHb = Total Hemo- globin; Classif: classification Discussion Thi t d h d th t diff t l l f bl d fl t i ti i th Table II - Effect size (ES) of the dependent variables TSI, O2Hb, HHb and tHb for treatment compari- sons during blood flow restriction, based on the criteria proposed by Sawilowsky [25] TSI O2Hb HHb tHb ES Classif ES Classif ES Classif ES Classif 30BFR- 50BFR 0.26 Medium 0.06 Very small -0.95 Large -0.40 Small 30BFR- 80BFR 0.36 Medium -0.01 Very small -0.99 Large -0.46 Small 30BFR- 100BFR 1.55 Very large 1.99 Very large -0.30 Small 1.15 Large 50BFR- 80BFR 0.01 Very small -0.07 Very small -0.14 Very small -0.12 Very small 50BFR- 100BFR 1.25 Very large 1.88 Very large 0.84 Large 1.90 Very large 80BFR- 100BFR 2.05 Huge 1.93 Very large 0.91 Large 1.77 Very large TSI = Tissue Saturation Index; O2Hb = Oxyhemoglobin; HHb = Deoxyhemoglobin; tHb = Total Hemo- globin; Classif: classification - Effect size (ES) of the dependent variables TSI, O2Hb, HHb and tHb for treatment compari- ring blood flow restriction, based on the criteria proposed by Sawilowsky [25] Table III - Effect size (ES) of the dependent variables TSI, O2Hb, HHb and tHb for condition compari- sons, after blood flow release, based on the criteria proposed by Sawilowsky [25 , , p p y y [ TSI O2Hb HHb tHb ES Classif ES Classif ES Classif ES Classif 30BFR- 50BFR -1.85 Very large 0.86 Large 1.28 Very large 1.44 Large 30BFR- 80BFR -1.18 Large 0.57 Medium 1.62 Very large 1.18 Large 30BFR- 100BFR -2.26 Huge -1.32 Very large 0.53 Medium -0.75 Medium 50BFR- 80BFR 0.65 Large -0.37 Small 0.45 Small -0.02 Very small 50BFR- 100BFR -0.79 Large -2.37 Huge -0.27 Small -2.15 Huge 80BFR- 100BFR -1.31 Very large -2.12 Huge -0.55 Medium -1.83 Very large TSI = Tissue Saturation Index; O2Hb = Oxyhemoglobin; HHb = Deoxyhemoglobin; tHb = Total Hemo- globin; Classif: classification Results All differences for p < 0.05 Figure 2 - Oxyhemoglobin (O2Hb), desoxyhemoglobin (HHb), total hemoglobin (tHB) and tissue sa- turation index (TSI) modifications from baseline at the different body flow restriction pressure levels (30%, 50% 80% and 100% BFR) during blood flow restriction (BFR) and after flow release (Post) All values during BFR were statistically different from baseline in each pressure level. All values pos- t-BFR were statistically different from during BFR in each pressure level. 1 = different from 30BFR; 2 = different from 50BFR; 3 = different from 80BFR. Discussion TSI is a direct indicator of the percentage of oxygenated hemoglobin in the tissue directly below the sensor. The present study observed a reduction in TSI con- centration during inflated cuff, which indicates that the oxygenated blood supply is less than muscle demand. The uptake of muscle oxygen can be influenced, among other factors, by the ability of the microcirculation to provide the necessary oxygen to the tissue [19]. The reduction in TSI has already been observed in other studies of blood flow restriction associated or not with the practice of physical exercise [25,26]. Kilgas et al. [20] observed a significant reduction in TSI compared to the con- trol condition in four different pressure loads (60%, 80%, 100%, and 120% of the AOP) associated with the handgrip exercise. The authors identified a more significant re- duction in TSI as the pressure level increased, but with no difference between 60 and 80% (submaximal loads) and between 100 and 120% of the AOP (maximum and supra-maximum, respectively). Although the present study did not associate blood flow restriction with exercise, there was also a tendency to reduce the TSI as the pressure level increased, with no significant difference between submaximal loads (baseline> 30BFR = 50BFR = 80BFR > 100BFR). Both studies used a 10 cm wide cuff. Kilgas et al. [20] observed a significant reduction in TSI compared to the con- trol condition in four different pressure loads (60%, 80%, 100%, and 120% of the AOP) associated with the handgrip exercise. The authors identified a more significant re- duction in TSI as the pressure level increased, but with no difference between 60 and 80% (submaximal loads) and between 100 and 120% of the AOP (maximum and supra-maximum, respectively). Although the present study did not associate blood flow restriction with exercise, there was also a tendency to reduce the TSI as the pressure level increased, with no significant difference between submaximal loads (baseline> 30BFR = 50BFR = 80BFR > 100BFR). Both studies used a 10 cm wide cuff. A hyperemic rebound effect allowed the TSI indicators to remain higher than the resting condition even after 30 seconds of withdrawal of the cuff pressure. This result is reinforced by the clinical difference observed through the effect size ob- tained in the multiple comparisons between the conditions in the present study. Discussion This study showed that different levels of blood flow restriction in the upper limb do not promote linear changes in the percentage of tissue oxygenation and total hemoglobin. This finding agrees with previous studies that also identified that total hemoglobin reduction is not linear with pressure load. The differential of the current study was, in addition to observing the beha- vior of total hemoglobin (indirectly), having followed the impact on cellular hypoxia. 612 Rev Bras Fisiol Exerc 2021;20(6):604-617 Identifying non-linear behavior in cellular hypoxia is important because this seems to be a stimulus condition for muscle hypertrophy mechanisms [18]. The current results demonstrate that it is unnecessary to exert high-pressure loads to significantly impact hypoxia, allowing the participant to reach a possible hypertrophic stimulation with loads between 30 and 50% of total occlusion without experiencing the significant discomfort generated by heavier loads. The main findings were as follows. Tissue saturation index (TSI) decreased under all conditions, significant for 100BFR compared to 80BFR during the blood flow restriction phase. After the release of blood flow, a significant increase was observed in all conditions, indicating a rebound effect, and for loads of 100BFR and 50BFR, those showed more significant effect than 30BFR. The tHb value is higher in moderate blood flow restriction loads, mainly 50BFR and 80BFR compared to more extreme pressure level (100BFR). The oxygenated hemoglobin increased significantly with submaximal pressure load. However, after the release of blood flow, O2Hb aug- mented for 100BFR conditions while the other groups decreased. Furthermore, the muscle oxygenation returned to baseline condition for 30BFR. Finally, the deoxyge- nated hemoglobin was higher in medium and high blood pressure loads (50BFR and 80BFR) when compared to lower blood flow restriction loads (30BFR) after the rele- ase of the blood flow. TSI is a direct indicator of the percentage of oxygenated hemoglobin in the tissue directly below the sensor. The present study observed a reduction in TSI con- centration during inflated cuff, which indicates that the oxygenated blood supply is less than muscle demand. The uptake of muscle oxygen can be influenced, among other factors, by the ability of the microcirculation to provide the necessary oxygen to the tissue [19]. The reduction in TSI has already been observed in other studies of blood flow restriction associated or not with the practice of physical exercise [25,26]. Discussion The effect size was considered very small to medium between the 30BFR, 50BFR, and 80BFR conditions, but very large to huge when these intermediates were compared to 100BFR. Thus, the 100BFR had a more significant impact on the TSI compared to the other conditions. 613 Rev Bras Fisiol Exerc 2021;20(6):604-617 In practical terms, the similar lower oxygen saturation between the 30BFR, 50BFR, and 80BFR experimental treatments indicates that this flow restriction mar- gin appears to have a similar impact on cellular hypoxia. Disregarding exercise, a restriction between 30 and 80% of the AOP could have a similar impact on the hyper- trophic mechanisms associated with the more metabolic environment. Previous stu- dies have shown that simple exposure to blood flow restriction without exercise can promote hypertrophic stimuli that would reduce the impact of atrophy caused by an injury to the muscle-tendon structure [13]. Thus, the lower pressure level (30%) may be more comfortable and safer for most people, particularly older and untrained individuals, providing similar benefits to an 80% arterial pressure level restriction. On the other hand, higher pressures that allow total or close to AOP would probably promote a higher hypoxic ambient despite being more uncomfortable. Such a more favorable milieu would potentiate mechanisms such as cell swelling [10] and the re- lease of growth hormone (GH) [11]. Although Hunt et al. [27] have observed that the deformation of the brachial artery occurs at approximately 110 mmHg of pressure with the use of an 11 cm wide cuff, the reduction in blood flow occurs early in order to change the arterial diame- ter. On average, men experienced a reduction in blood flow with 60 mmHg. In another study by Mouser et al. [17], the venous system was impacted with pressure loads of 10 to 30% of AOP. Notwithstanding, the artery would only be im- pacted with pressure loads higher than 60%. This study was carried out with a 5 cm cuff, half the width of the cuff in the present study. The literature has shown that cuffs with a smaller width require a higher-pressure level to cause a similar impact to a broader cuff [28]. The present study observed an increase in tHB in all pressures used in relation to the baseline and being more significant in the conditions 50BFR and 80BFR com- pared to 100BFR (baseline> 30 BFR = 100BFR > 50BFR = 80BFR). Discussion When considering the results of these previous studies with the current observations, the 50BFR condition was performed with an average pressure level of 68 mmHg. It is possible to assume that 50BFR and 80BFR must have interfered equally in the blood flow, as noted by the small effect size between these conditions. This assumption is supported by another study by Mouser et al. [16]. The authors identified blood flow reduction up to 50% of arterial occlusion, followed by stabilizing the flow up to 90% before another sudden drop. This abrupt reduction in blood flow in the last 10% before reaching the point of AOP should explain why 100BFR had a lower tHb than intermediate pressure loads (50BFR and 80BFR). After the cuff deflated, blood flow observed by the tHb concen- tration returned to rest at 50BFR and 80BFR, but not at 30BFR and 100BFR. The concentration of HHb is an important indicator of oxidative metabolism in muscles [29]. The increase in the concentration of HHb in all restriction conditions performed in the present study indicates the hypoxia generated by the mechanical restriction. This result agrees with what was observed in a previous study [30]. Al- though there was no difference between groups during the restriction phase, the effect size indicated that pressure levels between 50BFR and 80BFR had a more sig- 614 Rev Bras Fisiol Exerc 2021;20(6):604-617 nificant clinical impact concerning all conditions and with very little practical dif- ference between them. Furthermore, all groups had a substantial clinical difference and statistically significantly higher HHb values than baseline. The significance of this finding is that hypoxia is an important signal to sti- mulate some mechanisms of strength and muscle mass increase [11,31]. Thus, it is possible to assume that, at least at rest, the 30BFR level has the same impact on hypo- xia and muscle oxidative metabolism as on higher pressure levels, thus reducing dis- comfort and cardiovascular risk. On the other hand, after blood flow release, HHb values reduced in all groups, but only at 30BFR and 100BFR did the levels return to baseline condition after 30 seconds. Finally, a more pronounced increase in O2Hb levels in the 30BFR, 50BFR and 80BFR than the 100BFR condition was observed during the cuff inflation phase. The very high clinical impact between the 100BFR and the other pressure levels confir- med this distinct behavior between maximum and submaximal pressure loads. Discussion On the other hand, observing a minimal effect size between submaximal loads demons- trated that the clinical implications generated with an arterial restriction level of 30 to 80% of the AOP at rest are practically insignificant. The O2Hb concentration during blood flow restriction in the 100BFR condi- tion was similar to the baseline. Besides, a reduction in O2Hb concentrations in the 30BFR, 50BFR, and 80BFR conditions was observed when the cuff pressure was rele- ased. Nevertheless, these values did not return to the baseline condition within 30 seconds of free blood flow. On the other hand, in the 100BFR condition, the O2Hb concentration increased, indicating a possible rebound effect due to the action of some vasodilating substances, such as nitric oxide [32]. After the flow is released, blood moves more turbulently, increasing shear stress stimulating the production and release of nitric oxide, promoting local vasodilation [33,34]. Shear stress is in- fluenced by blood flow speed, which is altered according to the pressure imposed by the cuff and blood flow release by removing the pressure exerted by the cuff [35]. A higher concentration of O2Hb accompanies this increase in blood flow. These results are opposite with those observed in previous studies. Such stu- dies observed a reduction in O2Hb during blood flow restriction [36], possibly due to the difference in the site of signal capture between the studies. The difference observed in the results during the blood restriction phase may be explained by the positioning of the NIRS probe about the site of compression exerted by the cuff. Bopp et al. [36] positioned the probe on the subject’s forearm immediately after the cuff, restricting the blood, while in the present study, the NIRS probe was placed on the arm, and the cuff was placed on the forearm. This procedure was done to avoid inter- ference in the vascular walls due to deformation by the inflated cuff. The NIRS device captures hemoglobin concentrations (oxy and deoxygeated) to a depth of 1.5 cm below the transmitter/receiver. Thus, the relative O2Hb concen- tration is measured in the small blood vessels (arterioles, venules, and capillaries) that cross this region captured by the equipment. In the 100BFR condition, the blood 615 Rev Bras Fisiol Exerc 2021;20(6):604-617 flow must have been interrupted or close to it, even in the deepest regions, and it must have kept the O2Hb concentration in the arteries located before the inflated cuff. Discussion The equipment should not have picked up the blood in the most profound ves- sels. Some limitations do apply to the present study. We implemented BFR during rest, and different muscle hemodynamic behaviors may be expected during resistan- ce exercise. Besides, our findings are limited to the upper limbs and may not entirely represent blood flow restriction involving a larger muscle mass. Therefore, further studies are needed to confirm a possible relationship between acute muscle hemo- dynamics caused by different blood flow restriction pressure levels and hypertrophic markers secondary to resistance exercise with blood flow restriction. Author´s contributions Conception of the study: Gomes PSC, Meirelles CM; Study design: Carvalho RF, Gomes PSC, Fernandes Junior ML, Meirelles CM; Data collection: Carvalho RF, Fernandes Junior ML; Statistical analysis: Car- valho RF, Gomes PSC; Writing of the document: Carvalho RF, Gomes PSC, Fernandes Junior ML, Meirel- les CM; Final review of the manuscript: Meirelles CM; Writing the English version of the manuscript: Gomes PSC. Funding source o RF - PhD scholarship holder from the Carlos Chagas Filho Foundation for Research in the Rio de Janeiro - FAPERJ (Proc.: E-26/201.705/2017) Conclusion In conclusion, this study revealed that pressure levels between 30 and 50% of the brachial artery blood flow occlusion are sufficient to cause hypoxia in the occlu- ded muscle, in the same magnitude as higher pressure loads (up to 80%). Conflict of interest Conflict of interest All authors declare that there is no conflict of interest regarding this study and manuscript. Conflict of interest All authors declare that there is no conflict of interest regarding this study and manuscript. Funding source Gomes PS - Productivity in Research Scholarship (PQ2) from the National Council for Scientific and Technological Development from Brazil - CNPq; PROCIÊNCIA Scholarship sponsored by the State Uni- versity of Rio de Janeiro Carvalho RF - PhD scholarship holder from the Carlos Chagas Filho Foundation for Research in the State of Rio de Janeiro - FAPERJ (Proc.: E-26/201.705/2017) References 1. Lixandrão ME, Ugrinowitsch C, Laurentino G, Libardi CA, Aihara AY, Cardoso FN, et al. Effects of exercise intensity and occlusion pressure after 12 weeks of resistance training with blood-flow restric- tion. Eur J Appl Physiol 2015;115(12):2471-80. doi: 10.1007/s00421-015-3253-2 2. Lixandrão ME, Ugrinowitsch C, Berton R, Vechin FC, Conceição MS, Damas F, et al. Magnitude of muscle strength and mass adaptations between high-load resistance training versus low-load resis- tance training associated with blood-flow restriction: a systematic review and meta-analysis. Sports Med 2018;48(2):361-78. doi: 10.1007/s40279-017-0795-y 3. Martín-Hernández J, Marín PJ, Menéndez H, Ferrero C, Loenneke JP, Herrero AJ. Muscular adapta- tions after two different volumes of blood flow-restricted training. Scand J Med Sci Sports 2013;23(2):1- 7. doi: 10.1111/sms.12036 3. Martín-Hernández J, Marín PJ, Menéndez H, Ferrero C, Loenneke JP, Herrero AJ. Muscular adapta- tions after two different volumes of blood flow-restricted training. Scand J Med Sci Sports 2013;23(2):1- 7. doi: 10.1111/sms.12036 4. Vechin FC, Libardi CA, Conceição MS, Damas FR, Lixandrão ME, Berton RPB, et al. Comparisions between low-intensity resistance training with blood flow restriction and high-intensity resistance 4. Vechin FC, Libardi CA, Conceição MS, Damas FR, Lixandrão ME, Berton RPB, et al. Comparisions between low-intensity resistance training with blood flow restriction and high-intensity resistance 616 Rev Bras Fisiol Exerc 2021;20(6):604-617 training on quadriceps muscle mass and strength in elderly. J Strength Cond Res 2015;29(4):1071-6. doi: 10.1088/0022-3727/8/4/003 training on quadriceps muscle mass and strength in elderly. J Strength Cond Res 2015;29(4):1071-6. doi: 10.1088/0022-3727/8/4/003 5. Yasuda T, Ogasawara R, Sakamaki M, Ozaki H, Sato Y, Abe T. Combined effects of low-intensity blood flow restriction training and high-intensity resistance training on muscle strength and size. Eur J Appl Physiol 2011;111(10):2525-33. doi: 10.1007/s00421-011-1873-8 6. Takarada Y, Nakamura Y, Aruga S, Onda T, Miyazaki S, Ishii N. Rapid increase in plasma growth hor- mone after low-intensity resistance exercise with vascular occlusion. J Appl Physiol. 2000;88(1):61-5. doi: 10.1152/jappl.2000.88.1.61 j 7. Takarada Y, Sato Y, Ishii N. Effects of resistance exercise combined with vascular occlusion on mus- cle function in athletes. Eur J Appl Physiol 2002;86(4):308-14. doi: 10.1007/s00421-001-0561-5 8. Takarada Y, Tsuruta T, Ishii N. Cooperative effects of exercise and occlusive stimuli on muscu- lar function in low-intensity resistance exercise with moderate vascular occlusion. Jpn J Physiol 2004;54(6):585-92. doi: 10.2170/jjphysiol.54.585 9. Laurentino GC, Ugrinowitsch C, Roschel H, Aoki MS, Soares AG, Neves Junior M, et al. Streng- th training with blood flow restriction diminishes myostatin gene expression. References Med Sci Sports Exerc 2012;44(3):406-12. doi: 10.1249/MSS.0b013e318233b4bc 10. Loenneke JP, Fahs CA, Rossow LM, Abe T, Bemben MG. The anabolic benefits of venous blood flow restriction training may be induced by muscle cell swelling. Med Hypotheses 2012;78(1):151-4. doi: 10.1016/j.mehy.2011.10.014 11. Reeves GV, Kraemer RR, Hollander DB, Clavier J, Thomas C, Francois M, et al. Comparison of hor- mone responses following light resistance exercise with partial vascular occlusion and moderately difficult resistance exercise without occlusion. J Appl Physiol 2006;101:1616-22. doi: 10.1152/japplphy- siol.00440.2006 12. Lopes KG, Bottino DA, Farinatti P, Souza MGC, Maranhão PA, Araujo CMS, et al. Strength training with blood flow restriction – a novel therapeutic approach for older adults with sarcopenia? A case report. Clin Interv Aging 2019;14:1461-9. doi: 10.2147/CIA.S206522 p g g 13. Takarada Y, Takazawa H, Ishii N. Applications of vascular occlusion diminish disuse atrophy of knee extensor muscles. Med Sci Sports Exerc 2000;32(12):2035-9. doi: 10.1097/00005768-200012000-00011 p 14. Loenneke JP, Thiebaud RS, Abe T, Bemben MG. Blood flow restriction pressure recommendations: The hormesis hypothesis. Med Hypotheses 2014;82(5):623-6. doi: 10.1016/j.mehy.2014.02.023 15. Mattocks KT, Jessee MB, Counts BR, Buckner SL, Mouser JG, Dankel SJ, et al. The effects of upper body exercise across different levels of blood flow restriction on arterial occlusion pressure and per- ceptual responses. Physiol Behav 2017;171:181-6. doi: 10.1016/j.physbeh.2017.01.015 16. Mouser JG, Dankel SJ, Jessee MB, Mattocks KT, Buckner SL, Counts BR, et al. A tale of three cuffs: the hemodynamics of blood flow restriction. Eur J Appl Physiol 2017;117(7):1493-9. doi: 10.1093/icvts/ ivx022 17. Mouser JGACJ, Black CD, Bemben DA, Bemben MG. Brachial blood flow under relative levels of blood flow restriction is decreased in a nonlinear fashion. Clin Physiol Funct Imaging 2018;38(3):425- 30. doi: 10.1111/cpf.12432 p 18. Schoenfeld BJ. Potential mechanisms for a role of metabolic stress in hypertrophic adaptations to resistance training. Sports Med 2013;43(3):179-94. doi: 10.1007/s40279-013-0017-1 19. Gerovasili V, Dimopoulos S, Tzanis G, Anastasiou-Nana M, Nanas S. Utilizing the vascular occlusion technique with NIRS technology. Int J Ind Ergon 2010;40(2):218-22. doi: 10.1016/j.ergon.2009.02.004 20. Kilgas MA, McDaniel J, Straves J, Pollock BS, Singer TJ, Elmer SJ. Limb blood flow and tissue perfu- sion during exercise with blood flow restriction. Eur J Appl Physiol 2019;119(2):377-87. doi: 10.1007/ s00421-018-4029-2 21. Pereira MIR, Gomes, PSC, Bhambhani, YN. A brief review of the use of near infrared spectrosco- py with particular interest in resistance exercise. Sports Med 2007;37:615-24. doi: 10.2165/00007256- 200737070-00005. 22. Espírito-Santo HA, Daniel F. References Calcular e apresentar tamanhos do efeito em trabalhos científicos (1): As limitações do p < 0,05 na análise de diferenças de médias de dois grupos. Rev Port Inv Comp Soc 2015;1(1):3-16. doi: 10.7342/ismt.rpics.2015.1.1.14 23. Espírito-Santo HA, Daniel F. Calcular e apresentar tamanhos do efeito em trabalhos científicos (3): Guia para reportar os tamanhos do efeito para análises de regressão e ANOVAs Calculating and repor- ting effect sizes on scientific papers (3): Guide to report regression models and ANOVA. Rev Port Inv Comp Soc 2018;4(1):43-60. doi: 10.7342/ismt.rpics.2018.4.1.72 24. Vincent WJ, Weir JP. Statistics in Kinesiology. 4th edition ed. [S. l.]: Human Kinetics, Inc., 2011. E-book. 24. Vincent WJ, Weir JP. Statistics in Kinesiology. 4th edition ed. [S. l.]: Human Kinetics, Inc., 2011. E-book. 25. Sawilowsky SS. New effect size rules of thumb. J Mod Appl Stat Methods 2009;8(2):article26. doi: 10.22237/jmasm/1257035100 25. Sawilowsky SS. New effect size rules of thumb. J Mod Appl Stat Methods 2009;8(2):article26. doi: 10.22237/jmasm/1257035100 26. Padilla J, Johnson BD, Newcomer SC, Wilhite DP, Mickleborough TD, Fly AD, et al. Normalization of 617 Rev Bras Fisiol Exerc 2021;20(6):604-617 flow-mediated dilation to shear stress area under the curve eliminates the impact of variable hypere- mic stimulus. Cardiovasc Ultrasound 2008;6(1):44. doi: 10.1186/1476-7120-6-44. flow-mediated dilation to shear stress area under the curve eliminates the impact of variable hypere- mic stimulus. Cardiovasc Ultrasound 2008;6(1):44. doi: 10.1186/1476-7120-6-44. 27. Hunt JEA, Stodart C, Ferguson RA. The influence of participant characteristics on the relationship between cuff pressure and level of blood flow restriction. Eur J Appl Physiol 2016;116(7):1421-32. doi: 10.1007/s00421-016-3399-6 28. Crenshaw AG, Hargens AR, Gershuni DH, Rydevik B. Wide tourniquet cuffs more effective at lower inflation pressures. Acta Orthop Scand 1988;59(4):447-51. doi: 10.3109/17453678809149401. 29. Ryan TE, Brophy P, Lin C, Hickner RC, Neufer PD. Assessment of in vivo skeletal muscle mitochon- drial respiratory capacity in humans by near-infrared spectroscopy: a comparison with in situ measu- rements. J Physiol 2014;592(15):3231-41. doi: 10.1113/jphysiol.2014.274456 30. Soares RN, McLay KM, George MA, Murias JM. Differences in oxidative metabolism modulation in- duced by ischemia/reperfusion between trained and untrained individuals assessed by NIRS. Physiol Reports 2017;5(19):1–7. doi: 10.14814/phy2.13384 31. Moritani T, Sherman WM, Shibata M, Matsumoto T, Shinohara M. Oxygen availability and motor unit activity in humans. Eur J Appl Physiol and Occup Physiol 1992;64(6):552-6. doi: 10.1007/BF00843767 32. Green DJ, Dawson EA, Groenewoud HMM, Jones H, Thiissen DHJ. Is flow-mediated dilation ni- tric oxide mediated? A meta-analysis. Hypertension 2014;63(2):376-82. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. References doi: 10.1161/HYPERTENSIO- NAHA.113.02044 33. Doshi SN, Naka KK, Payne N, Jones CJH, Ashton M, Lewis MJ, Goodfellow J. Flow-mediated dila- tation following wrist and upper arm occlusion in humans: the contribution of nitric oxide. Clin Sci 2001;101(6):629-35. doi: 10.1042/cs1010629 34. Uematsu M, Ohara Y, Navas JP, Nishida K, Murphy TJ Alexander RW, et al. Regulation of endothelial cell nitric oxide synthase mRNA expression by shear stress. Am J Physiol - Cell Physiol 1995;269(6):38- 46. doi: 10.1152/ajpcell.1995.269.6.c1371 35. Gnasso A, Carallo C, Irace C, Franceschi MS, Mattioli PL, Motti C, Cortese C. Association between wall shear stress and flow-mediated vasodilation in healthy men. Atherosclerosis 2001;156(1):171-6. doi: 10.1016/S0021-9150(00)00617-1 36. Bopp CM, Townsend DK, Barstow TJ. Characterizing near-infrared spectroscopy responses to fore- arm post-occlusive reactive hyperemia in healthy subjects. Eur J Appl Physiol 2011;111(11):2753-61. doi: 10.1007/s00421-011-1898 This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
https://openalex.org/W3130777506	https://www.frontiersin.org/articles/10.3389/fcimb.2021.676650/pdf	English	null	Analysis of the blood microbiome in a porcine model of fecal-induced peritonitis	bioRxiv (Cold Spring Harbor Laboratory)	2,021	cc-by	10,332	Abbreviations: BSI, bloodstream infection; IL, interleukin; KEGG, Kyoto Encyclopedia of Genes and Genomes; MNP, mannose-binding lectin(MBL)-coated magnetic nanoparticles; OTU, operational taxonomic unit; PICRUSt, Phylogenetic Investigation of Communities by Reconstruction of Unobserved States. Kang jookang@unist.ac.kr Jae Hyuk Lee hyukmd@gmail.com †These authors have contributed equally to this work and share ﬁrst authorship ‡These authors have contributed equally to this work and share last authorship Specialty section: This article was submitted to Microbiome in Health and Disease, a section of the journal Frontiers in Cellular and Infection Microbiology Specialty section: This article was submitted to Microbiome in Health and Disease, a section of the journal Frontiers in Cellular and Infection Microbiology Specialty section: This article was submitted to Microbiome in Health and Disease, a section of the journal Frontiers in Cellular and Infection Microbiology Received: 08 March 2021 Accepted: 12 August 2021 Published: 30 August 2021 Keywords: blood microbiome, peritonitis, porcine (pig) model, bloodstream infection (BSI), dysbiosis ORIGINAL RESEARCH published: 30 August 2021 doi: 10.3389/fcimb.2021.676650 INTRODUCTION Received: 08 March 2021 Accepted: 12 August 2021 Published: 30 August 2021 Bloodstream infection (BSI) is deﬁned as a medical condition, in which viable bacteria or fungi are present in the bloodstream (Viscoli, 2016). BSI is a major threat to human health, as it can cause sepsis and organ dysfunction (Cecconi et al., 2018). A survey of the incidence of BSI in America and Europe during the years 1974-2008 reported rates between 80 and 189 per 100,000 individuals per year; this number has increased in recent years (Laupland, 2013; Laupland et al., 2020). Furthermore, many cases progress to critical conditions. BSI is estimated to cause 79,000–94,000 Edited by: Kun Lu, University of North Carolina at Chapel Hill, United States Edited by: Kun Lu, University of North Carolina at Chapel Hill, United States 1 Department of Biomedical Engineering, College of Information and Biotechnology, Ulsan National Institute of Science and Technology (UNIST), Ulsan, South Korea, 2 Department of Emergency Medicine, Seoul National University Bundang Hospital, Seongnam-si, South Korea, 3 Center for Genomic Integrity, Institute for Basic Science, Ulsan, South Korea University of North Carolina at Chapel Hill, United States Reviewed by: Ping Li, Zhejiang Gongshang University, China Megan R Kiedrowski, University of Alabama at Birmingham, United States Reviewed by: Ping Li, Zhejiang Gongshang University, China Megan R Kiedrowski, University of Alabama at Birmingham, United States Reviewed by: Ping Li, Zhejiang Gongshang University, China Megan R Kiedrowski, University of Alabama at Birmingham, United States Recent studies have suggested the existence of a blood microbiome in the healthy host. However, changes in the blood microbiome upon bloodstream infection are not known. Here, we analyzed the dynamics of the blood microbiome in a porcine model of polymicrobial bacteremia induced by fecal peritonitis. Surprisingly, we detected bacterial populations in the bloodstream even before the infection, and these populations were maintained over time. The native blood microbiome was notably taxonomically different from the fecal microbiome that was used to induce peritonitis, reﬂecting microbial tropism for the blood. Although the population composition after the infection was similar to that of the native blood microbiome, new bacterial strains entered the bloodstream upon peritonitis induction as clinical symptoms relevant to sepsis developed. This indicates that the bacteria detected in the blood before peritonitis induction were derived from the blood rather than a contamination. Comparison of the functional pathways enriched in the blood and fecal microbiomes revealed that communication and stress management pathways are essential for the survival of the blood microbiome. Correspondence: Taejoon Kwon tkwon@unist.ac.kr Joo H. Kang jookang@unist.ac.kr Jae Hyuk Lee hyukmd@gmail.com †These authors have contributed equally to this work and share ﬁrst authorship ‡These authors have contributed equally to this work and share last authorship Correspondence: Taejoon Kwon tkwon@unist.ac.kr Joo H. Animal Experiments High-throughput sequencing is an alternative technique for detecting microbes in the blood, even without culturing (Grumaz et al., 2016). However, this highly sensitive method raises some questions about BSI, i.e., on the existence of the blood microbiome [reviewed in (Castillo et al., 2019)]. Although the bloodstream is considered to be a sterile environment, recent evidence suggests that it may contain bacteria (or a microbiome), which may also colonize other organs. According to a 1969 study, metabolically active bacteria might be present in the blood (Tedeschi et al., 1969), and recent studies propose that bacteria may use the bloodstream as a transport system. For example, bacteria have been identiﬁed in the blood and adipose tissue samples from patients with type 2 diabetes (Massier et al., 2020), and in the liver of patients with non-alcoholic fatty liver disease (Sookoian et al., 2020). Furthermore, Porphyromonas gingivalis derived from chronic periodontitis is thought to contribute to Alzheimer’s disease (Dominy et al., 2019), leading to a speculation that the human microbiome can disseminate to other organs via the bloodstream. p Six domestic pigs (Sus scrofa domesticus), weighing approximately 45–55 kg each, were used, as described previously (Park et al., 2019). Autologous feces were collected 1 day before the experiment and preserved overnight at room temperature. The pigs were anesthetized by an intramuscular administration of zolazepam (zoletil, 5 mg/kg; Virbac, Carros, France). The animals were scrubbed with povidone-iodine soap and shaved, and monitoring devices, including an electrocardiograph, pulse oximeter, and a temperature probe, were attached. Then, the animals were intubated using an endotracheal tube and connected to a mechanical ventilator (Drager Fabius GS, Lubeck, Germany) providing an inhalation agent (sevoﬂurane; Baxter Inc., Deerﬁeld, IL) to maintain adequate ventilation with anesthesia. A sterile surgical drape for the abdomen was applied after meticulous dressing with povidone-iodine. Under the guidance of ultrasound, two 6-Fr arterial catheters (Merit Medical, South Jordan, UT) were inserted into the two femoral arteries to allow invasive blood pressure monitoring and repetitive blood sampling for blood culture. The feces (1 g/kg) that had been collected the previous day were diluted in 5% dextrose saline (10 g/dL) and warmed at 37°C for 1 hour in a water bath. A midline surgical incision was made in the abdomen, and the feces were introduced into the abdominal cavity. Citation: Hyun H, Lee MS, Park I, Ko HS, Yun S, Jang D-H, Kim S, Kim H, Kang JH, Lee JH and Kwon T (2021) Analysis of Porcine Model of Fecal-Induced Peritonitis Reveals the Tropism of Blood Microbiome. Front. Cell. Infect. Microbiol. 11:676650. doi: 10.3389/fcimb.2021.676650 August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Longitudinal Analysis of Blood Microbiome Hyun et al. deaths per year in North America and 157,000 deaths per year in Europe (Goto and Al-Hasan, 2013). The likelihood of detecting bacteria by blood culture also gradually increased after the induction. Here, we used the same blood samples as those obtained in a previously reported study of the porcine fecal-induced peritonitis model (Park et al., 2019) to investigate the role of the blood microbiome in bacteremia. Surprisingly, we detected many bacterial cells in the pig blood even before fecal induction. These bacteria likely constitute the blood microbiome. In addition, the bacterial species identiﬁed by blood culture were not the dominant species detected by 16S rRNA gene sequencing, although their increases over time tended to be the similar over time. The data presented herein provide new insights into the relationship between BSI and the blood microbiome. p Blood culture is a well-established method of detecting BSI, but the blood culture ﬁndings are not always clinically relevant. According to a recent study, 42.6% of 2,659 patients with suspected sepsis had a positive blood culture result, whereas the remaining 1,526 patients (56.4%) were blood culture- negative (Nannan Panday et al., 2019). False positives caused by contamination are a concern when testing for BSI. For example, according to some studies, only 51% of blood culture-positive samples represent actual BSI, 41% are a result of contamination, and 8% have unknown clinical signiﬁcance (Weinstein et al., 1997; Pien et al., 2010). The expected sensitivity and speciﬁcity of blood culture ﬁndings vary depending on the experimental conditions, including collection time, skin preparation prior to sampling, sampling site, and sample volume (Lamy et al., 2016). Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Taxonomic Analysis PCR amplicons obtained using the Illumina V34 PCR primers were selected using ipcress (provided in exonerate version 2.2) after concatenating paired-end reads (Slater and Birney, 2005). Next, taxonomic information was assigned to each paired read using the Ribosomal Database Project (RDP) classiﬁer (version 2.11) (Wang et al., 2007) and the RDP database (release 11.5) (Cole et al., 2014), at a conﬁdence score > 0.8. The cut-off was determined by sequencing of the ZymoBIOMICS Microbial Community Standard, as described in Supplementary Figure 1. Operational taxonomic units (OTUs) were deﬁned by clustering concatenated V34 amplicons at 99% identity using VSEARCH (version 2.13.6) (Rognes et al., 2016) (Supplementary Data 1). Ambiguous clusters with > 5% of reads assigned to a different genus than the seed genus assigned to the same cluster were removed. OTUs with low abundance, i.e., accounting for less than 0.4167% of all OTUs in all analyzed samples (N=6), were excluded from the taxonomic analysis. After deﬁning genus-level clusters, additional taxonomic levels (from the genus to the phylum, including the family, order, and class) were deﬁned. OTUs that were not present at the initial time point (i.e., before the induction of fecal peritonitis) were Fluorescence In Situ Hybridization (FISH) for Native Blood Microbiome Mannose-binding lectin (MBL)-coated magnetic nanoparticles (MNPs; 2 mg/mL) were added to 3 mL of the blood–DNA/RNA Shield solution (Cat # R1150, Zymo Research), and the samples were incubated for 20 min at room temperature. Captured bacteria were harvested using N52 magnets (BYO88-N52; KJ Magnetics, Pipersville, PA) and washed with PBS to remove other blood components. The captured bacteria were stored in the DNA/RNA Shield reagent at –20°C before use. The bacterial samples enriched by the human recombinant mannose-binding lectin(hrMBL)-coated magnetic nanoparticles (MNPs) were ﬂuorescently stained with 4′,6- diamidino-2-phenylindole (DAPI), and Cy3-labeled DNA FISH probes (5’-CTTGTACACACCGCCCGTCACACC-3’) targeting universal bacteria-speciﬁc ribosomal RNA sequences for quantitating the blood microbiome in the control blood sample from a porcine model. To magnetically concentrate the sample, we adopted a sinusoidal-shaped polydimethylsiloxane (PDMS) microﬂuidic device (300 µm × 200 µm; width × height) to capture the MNP-bound bacterial cells by locating a magnet (BYO88-N52, KJ Magnetics, PA, USA) underneath the device. Then, FISH reagents were sequentially injected by a syringe pump at a ﬂow rate of 10 mL/min into the microﬂuidic channel. For ﬁxation and permeabilization, the bacterial samples magnetically sequestered in the device were treated in the order of 24% ethanol(v/v) in 1X Tris-Buffered Saline (TBST) with 5 mM CaCl2 (5 min), washing buffer (3 min), 99% methanol (5 min), and washing buffer (3 min). 1X TBST supplemented with 5 mM CaCl2 was used for the washing buffer. Then, DNA FISH probes suspended in a hybridization solution (0.5 µM) were incubated with the samples (1 hour at 45°C), followed by a DAPI staining (30 min) and washing by 2X Saline Sodium Citrate Buffer (SSC) buffer. Finally, FISH images were obtained with a confocal microscope (LSM 780 Conﬁguration 16 NLO multi-photon confocal microscope, Zeiss, Germany) with DAPI and Cy3 ﬂuorescence ﬁlter sets. Animal Experiments Using the aseptic technique, blood samples (10 mL) were abstracted via the femoral arterial catheter at 1 or 2 hour intervals, and split between a bottle containing DNA/RNA Shield reagent (Cat # R1150; Zymo Research, Irvine, CA), and a pair of bottles for aerobe and anaerobe blood culture (BD BACTEC, Becton Dickinson, NJ). The pairs of blood culture bottles were then immediately placed in a blood culture system (BD BACTEC). The ﬁnal microbiological report on blood culture ﬁndings was obtained 5 days after the experiment from the Department of Laboratory Medicine (Seoul National University Hospital, Seoul, Republic of Korea). After the induction with feces, the pigs were monitored for 12 hours. The primary goal was to maintain the mean arterial pressure over 65 mmHg, with maximal ﬂuid (balanced crystalloid solution) and vasopressor (norepinephrine, vasopressin, and Nonetheless, most of the above studies analyzed the blood microbiome at a single time point, making it difﬁcult to rule out the possibility of contamination. Suppose the microbiota is stably maintained in the bloodstream. In that case, it should be detectable over time, like other microbiomes in the body. Concordance between data for different time points could support the existence of the blood microbiome. However, this type of data is difﬁcult to collect for human samples. Furthermore, these data should be assessed alongside blood culture results. Even in cases of BSI, a limited number of bacterial cells are present in the blood (approximately 0.1 to 100 cells per 1 mL of infected blood) (Lamy et al., 2016). If in fact the blood microbiome exists, it is questionable why these bacteria are not detected by blood culture. Hence, a controlled experimental environment is required to evaluate the relationship between BSI and the blood microbiome. Recently, we have developed a porcine model of fecal-induced peritonitis (Park et al., 2019). In the model, we observed symptoms of organ dysfunction approximately 7 hours (median) after introducing feces into the pig abdomen. August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 2 Longitudinal Analysis of Blood Microbiome Hyun et al. epinephrine) support. All procedures were approved by SNUBH IACUC (BA1804-246/040-01). eight cycles of adapter-ligation PCR were performed. The ﬁnal library was sequenced using Illumina MiSeq (Illumina, San Diego, CA) in a 2 × 250 bp conﬁguration. DNA Extraction From ZymoBIOMICS Microbial Community Standard DNA Extraction From ZymoBIOMICS Microbial Community Standard Total genomic DNA was extracted from 20 mL ZymoBIOMICS Microbial Community Standard (Cat # D6300, Zymo Research) using the ZymoBIOMICS DNA Miniprep Kit (Cat # D4300, Zymo Research). Library Preparation y p Beads with the captured bacteria were incubated at 37°C for 1 hour in a solution of lysozyme (10 mg/mL; Cat # 10837059001; Roche, Basel, Switzerland) in 10 mM Tris-HCl (pH 8.0). The beads were then transferred to lysis buffer (10 mM Tris-HCl [pH 7.4], 10 mM EDTA, and 2% SDS) containing 0.5 mg/mL proteinase K (Cat # B-2008; GeNetBio, Daejeon, Republic of Korea), and incubated overnight at 37°C. Genomic DNA was extracted from cell lysates by adding a phenol–chloroform– isoamyl alcohol mixture, followed by overnight incubation at –20°C. Next, DNA was precipitated with 0.6 volumes of isopropanol and 0.1 volume of 3 M sodium acetate (Cat # SR2006-050-55; Biosesang, Seongnam, Republic of Korea). After washing with 70% ethanol, the pellet was resuspended in 100 mL TE buffer for 30 min at 37°C. Then, 1 mL RNase A (Cat # B-2007, GeNetBio) was added, and the samples were incubated for 30 min at 37°C to remove RNA contamination. DNA was puriﬁed using Zymo DNA Clean & Concentrator-5 Kit (Cat # D4014, Zymo Research) and eluted in 20 mL RNase-free water. Fecal Sample Preparation for Microbiome Analysis Autologous feces (3 mL) from each pig were diluted in dextrose saline (10 g/mL) in a tube of DNA/RNA Shield reagent (Cat # R1150, Zymo Research) and stored at –20°C before analysis. Total genomic DNA was extracted using the MoBio PowerFecal® DNA Isolation Kit (Cat # 12830-50, MO BIO Laboratories, Carlsbad, CA) and FastPrep-24™(MP Biomedicals, LLC, Irvine, CA). DNA Extraction From ZymoBIOMICS Microbial Community Standard Total genomic DNA was extracted from 20 mL ZymoBIOMICS Microbial Community Standard (Cat # D6300, Zymo Research) using the ZymoBIOMICS DNA Miniprep Kit (Cat # D4300, Zymo Research). DNA Extraction From ZymoBIOMICS Microbial Community Standard Total genomic DNA was extracted from 20 mL ZymoBIOMICS Microbial Community Standard (Cat # D6300, Zymo Research) using the ZymoBIOMICS DNA Miniprep Kit (Cat # D4300, Zymo Research). Sequencing of the V34 Region of the 16S rRNA Gene Because all samples (except the fecal samples) contained a limited amount of DNA, RT-Q-PCR was performed using MIC qPCR (BioMolecular Systems, Upper Coomera, QLD) to determine the number of ampliﬁcation cycles before saturation (typically, 20–28 cycles). After an initial PCR with V34 primers, August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Longitudinal Analysis of Blood Microbiome Hyun et al. identiﬁed after selecting clusters with no reads from the initial samples. Changes in the initial microbiome after fecal induction were analyzed by using SourceTracker2 with default options (Knights et al., 2011). The proportions of bacterial species from unknown sources for the ﬁrst three time points (T02, T04, and T05) and the last three time points (T10, T11, and T12) were compared using Wilcoxon test. in the fecal microbiome, while Proteobacteria was the most abundant phylum in the blood. The bacterial composition of four out of six animals tested (P1120, P1126, P1211, and P1219; Figure 1) was relatively constant, with few perturbations (e.g., at 11 h post-induction in P1120 and 6 h post-induction in P1219). The microbiome composition in the two other animals (P1016 and P1103) showed some ﬂuctuations during the early induction stage, but it stabilized 4 h after the induction. Pathway Analysis Using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2) g To conﬁrm the presence of microbiome before the fecal induction, we also performed a DNA-FISH experiment using the bead-captured bacterial cells and a universal probe (5′-CTTGTACACACCGCCCGTCACACC-3′), which hybridizes to the 16S rRNA sequences of 98% bacterial species available in the Genome Taxonomy Database (GTDB; release 89) (Parks et al., 2018) (Supplementary Figure 3). Although the blood culture ﬁndings for all samples were negative, we veriﬁed that the blood samples obtained before peritonitis induction contained bacteria as observed in the FISH images. These observations suggest the presence of the blood microbiome in the porcine bacteremia model even before obtaining a positive blood culture result. Differences in biological pathways enriched in different microbiome populations were investigated using PICRUSt2 (version 2.2.0-b) (Douglas et al., 2020), by determining the relative enrichment of biological pathways in each sample. Kyoto Encyclopedia of Genes and Genomes (KEGG) orthologs and pathways inferred at the genus level (99% OTUs) were entered into the PICRUSt2 metagenome pipeline. Results of the KEGG ortholog analysis were assigned to KEGG pathways based on PICRUSt2-appended default ﬁles. KEGG ortholog enrichment values without pathway information were discarded, and the remaining values were summed. If KEGG orthologs belonged to more than two KEGG pathways, they were added in individually. The proportion of each pathway between the blood microbiome over time and the fecal microbiome used to induce peritonitis were compared using one-sample Wilcoxon test. Characterization of the Initial Blood Microbiome in the Porcine Model It is possible that the microbiome detected in the blood samples may have reﬂected the native blood microbiome or contamination of the arterial catheter used for sampling or contamination by the skin microbiome, as discussed previously (Trautner and Darouiche, 2004; Horiba et al., 2018; Okuda et al., 2018). Further, because the number of bacterial cells in the samples was small, the observed microbiome could represent an uncontrollable low biomass contamination from an unknown source, known as the “KitOme” (Stinson et al., 2019). If the majority of identiﬁed bacteria came from an accidental contamination, one would expect to not see any discernible patterns in the blood microbiome proﬁles. However, if the bacteria were blood microbiome related to peritonitis induction, we would detect new types of bacteria that entered the bloodstream. Blood Microbiome in the Porcine Model Before Peritonitis Induction A B D E F C FIGURE 1 \| (A–F) Percentage of operational taxonomic units (OTUs) from each phylum identiﬁed in the blood of pigs with fecal-induced peritonitis. The proportion of each phylum in the blood microbiome was maintained throughout the induction period, which was different from the observations for the fecal microbiome. Although Proteobacteria was the major phylum identiﬁed in the blood, its percentage contribution varied in each of the six animals tested. The proportions at the genus level are shown in Supplementary Figure 2. A B D C C D F E F FIGURE 1 \| (A–F) Percentage of operational taxonomic units (OTUs) from each phylum identiﬁed in the blood of pigs with fecal-induced peritonitis. The proportion of each phylum in the blood microbiome was maintained throughout the induction period, which was different from the observations for the fecal microbiome. Although Proteobacteria was the major phylum identiﬁed in the blood, its percentage contribution varied in each of the six animals tested. The proportions at the genus level are shown in Supplementary Figure 2. E F F FIGURE 1 \| (A–F) Percentage of operational taxonomic units (OTUs) from each phylum identiﬁed in the blood of pigs with fecal-induced peritonitis. The proportion of each phylum in the blood microbiome was maintained throughout the induction period, which was different from the observations for the fecal microbiome. Although Proteobacteria was the major phylum identiﬁed in the blood, its percentage contribution varied in each of the six animals tested. The proportions at the genus level are shown in Supplementary Figure 2. trends. We then set the ﬁrst three time points as T00, T02, and T04, and the last three time points as T09, T10, and T11, and we used Wilcoxon test to compare the two groups. We detected a signiﬁcant difference between the early and late time points in only one animal (Supplementary Figure 4). in its proportion as the dilution factor increased. However, when we used the most dilute sample as the sink (mimicking low biomass at the initial time point), we observed no changes in the population. Similarly, for the S. bongori study (Salter et al., 2014), when we used 107 cells or 103 cells as the sink, we found that the populations were different from those in the blood microbiome. Blood Microbiome in the Porcine Model Before Peritonitis Induction Because the number of bacteria in the blood is estimated to be 0.1–100 colony-forming units (CFUs)/mL, even in BSI (Lamy et al., 2016), we expected to observe similar numbers for the blood microbiome in the porcine bacteremia model. First, we enriched bacteria present in 3 mL blood using opsonin-coated magnetic beads (Kang et al., 2014). We then extracted the genomic DNA and performed 16S rRNA gene sequencing using Illumina V34 primers. For the 12 time points analyzed, we obtained 116,062.39 paired reads, on average, per sample from each of the six animals (median, 102,654 reads; minimum, 38,160 reads; maximum, 333,130 reads; all reads are available at the ENA under the accession ID PRJEB39083). We next clustered the reads to deﬁne OTUs and performed taxonomic analysis (from the phylum to genus level) using the RDP classiﬁer (Wang et al., 2007) (See Materials and Methods for details; Supplementary Data 2 for genus and Supplementary Data 3 for phylum). To distinguish between the two possibilities, we used SourceTracker2 (Knights et al., 2011) and compared the blood microbiomes before and after peritonitis induction (Figure 2; Supplementary Data 4). We assumed that the initial time point would be the “sink” for “the native blood microbiome” and attempted to identify their trends over time. In all animals, the bacterial population detected at the initial time point (sink) decreased gradually until approximately 6–8 hours after peritonitis induction and then remained constant until 12 hours after the induction. This trend matched the physiological symptoms of sepsis (Park et al., 2019). Therefore, we speculated that the blood microbiome was altered approximately 6–8 hours after the fecal induction. We compared the proportion of species from unknown sources at the early time points with those at the late time points and observed signiﬁcant differences in ﬁve out of six animals with Wilcoxon test (Figure 2G). By contrast, when we used 12-hour samples as the sink, we did not observe any Surprisingly, we observed many bacterial species at the initial time point, i.e., even before the induction of fecal peritonitis. Furthermore, the bacterial populations did not change much over 12 h (Figure 1, phylum level; Supplementary Figure 2, genus level). Firmicutes and Bacteroidetes were the most abundant phyla August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Longitudinal Analysis of Blood Microbiome Hyun et al. Blood Microbiome in the Porcine Model Before Peritonitis Induction We hence concluded that the blood microbiome that we detected herein does not represent a contamination associated with sample preparation or “noise” (KitOme), even for the microbiome detected at the initial time point, which may be the “native blood microbiome” of each animal. To validate the above ﬁndings ruling out low biomass contamination, we also analyzed two public datasets: a serially diluted mock community standard (Karstens et al., 2019) and a serially diluted Salmonella bongori culture (Salter et al., 2014) (Supplementary Figure 5). As expected, when we used the mock community standard as the sink, we observed a linear reduction August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 5 Longitudinal Analysis of Blood Microbiome Hyun et al. A B D E F G C FIGURE 2 \| SourceTracker2 analysis indicating that the population of novel bacterial strains not present at the initial time point increased throughout the induction period. (A–F) The microbiome observed at the initial time point is designated as the “source”. The proportions of species that originated from the source were then tracked throughout the induction period. The proportion of unknown strains increased gradually until approximately 7 to 8 h after the induction before being maintained. (G) Wilcoxon test was used to determine the signiﬁcance between ﬁrst three time points (2, 4, 5 hours after induction) and the last three time points (10, 11, 12 hours after induction), and meaningful increase of an “unknown” microbiome was observed in ﬁve of six animals. * denotes the p-value of the Wilcoxon test is less than 0.05. N.S., Not Signiﬁcant. B A D E G FIGURE 2 \| SourceTracker2 analysis indicating that the population of novel bacterial strains not present at the initial time point increased throughout the induction period. (A–F) The microbiome observed at the initial time point is designated as the “source”. The proportions of species that originated from the source were then tracked throughout the induction period. The proportion of unknown strains increased gradually until approximately 7 to 8 h after the induction before being maintained. (G) Wilcoxon test was used to determine the signiﬁcance between ﬁrst three time points (2, 4, 5 hours after induction) and the last three time points (10, 11, 12 hours after induction), and meaningful increase of an “unknown” microbiome was observed in ﬁve of six animals. Blood Microbiome in the Porcine Model Before Peritonitis Induction * denotes the p-value of the Wilcoxon test is less than 0.05. N.S., Not Signiﬁcant. Detection of Altered Blood Microbiome After Peritonitis Induction observed at the initial time point, thus obtaining newly emerged OTUs for each sample. We noted gradual changes in bacterial populations as BSI progressed (Figure 3). Although SourceTracker2 analysis revealed noticeable changes in the blood microbiome after peritonitis induction, the composition of the overall bacterial population was relatively consistent. We speculated that the pre-existing blood microbiome might mask small changes in the blood microbiome caused by peritonitis induction. To test this, we computationally discarded OTU clusters containing any OTUs At the phylum level, the blood and fecal microbiomes harbored different amounts of Proteobacteria and Firmicutes. Proteobacteria was the most abundant phylum in the bloodstream of all animals but was not dominant in the feces. Firmicutes showed the opposite trend. We observed this difference regardless of computational August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Longitudinal Analysis of Blood Microbiome Hyun et al. A B D E F C FIGURE 3 \| (A–F) Percentage of newly emerged operational taxonomic units (OTUs) from each phylum in the blood of pigs with fecal-induced peritonitis. The microbiome observed at the initial time point was discarded computationally from data for all other time points, and the proportion at each time point was re-calculated. A B B D C C D E F FIGURE 3 \| (A–F) Percentage of newly emerged operational taxonomic units (OTUs) from each phylum in the blood of pigs with fecal-induced peritonitis. The microbiome observed at the initial time point was discarded computationally from data for all other time points, and the proportion at each time point was re-calculated. F E F FIGURE 3 \| (A–F) Percentage of newly emerged operational taxonomic units (OTUs) from each phylum in the blood of pigs with fecal-induced peritonitis. The microbiome observed at the initial time point was discarded computationally from data for all other time points, and the proportion at each time point was re-calculated. Biological Pathways Enriched in the Blood Microbiome In the analysis presented above, we conﬁrmed that new bacterial phyla have emerged into the bloodstream after peritonitis induction, and the populations of the native blood microbiome are notably different in each animal. So we utilized OTUs subtracting the native blood microbiome for this analysis to identify pathways essential for bacteria to survive in the bloodstream. Pathways related to ABC transporters, two- component systems, and oxidative phosphorylation were enriched in the blood microbiome. By contrast, pathways related to purine metabolism, pyrimidine metabolism, and ribosome expression were enriched to a lesser extent than observed in the feces. To validate these ﬁndings, we used Wilcoxon test to reveal pathways that are signiﬁcantly enriched or depleted. The ABC transporters and two-component systems constitute essential mechanisms that allow bacteria to appropriately respond to environmental signals (Messenger and Barclay, 1983; Levy, 2000; Gebhard, 2012; Mattos-Graner and Duncan, 2017). For example, some bacteria utilize ABC transporters to regulate acid–base balance and metal iron homeostasis (Messenger and Barclay, 1983; Gebhard, 2012). Furthermore, bacteria can use these transporters to defend themselves against antimicrobial peptides and proteins in the bloodstream (Levy, 2000; Gebhard, 2012). Two-component systems, each composed of a histidine kinase and a response regulator, are major bacterial signaling pathways that sense environmental cues (Mascher et al., 2006; Mattos-Graner and Duncan, 2017), such as pH (Gao and Lynn, 2005; Liu and Burne, 2009) and oxidative stress (Ortiz de Orué Lucana et al., 2012). They are also tightly linked to bacterial responses to the host A B D E F C FIGURE 4 \| (A–F) Pathways differentially enriched in the blood microbiome. Pathways related to ABC transporters, two-component systems, and oxidative phosphorylation were over-represented in the blood microbiome compared with those in the fecal microbiome. By contrast, pathways related to central metabolism, such as purine and pyrimidine metabolism, and ribosome-related pathways, were under-represented. Wilcoxon test was used to check the signiﬁcance of the enrichment and depletion of each pathway. denotes the p-value of the Wilcoxon test is less than 0.05. A B B A C D D C E F F E FIGURE 4 \| (A–F) Pathways differentially enriched in the blood microbiome. Pathways related to ABC transporters, two-component systems, and oxidative phosphorylation were over-represented in the blood microbiome compared with those in the fecal microbiome. By contrast, pathways related to central metabolism, such as purine and pyrimidine metabolism, and ribosome-related pathways, were under-represented. Biological Pathways Enriched in the Blood Microbiome ﬁltering of the intrinsic microbiome population. Interestingly, Bacteroidetes, the representative gut microbiome phylum (Crespo- Piazuelo et al., 2019), was detected at all-time points, and its relative abundance gradually increased over time; this was not observed in the absence of the background blood microbiome (the light blue line in Figure 3). However, the composition of the blood microbiome did not change, even when we discarded the background populations (i.e., the microbiome detected before the fecal induction) from the analysis. ﬁltering of the intrinsic microbiome population. Interestingly, Bacteroidetes, the representative gut microbiome phylum (Crespo- Piazuelo et al., 2019), was detected at all-time points, and its relative abundance gradually increased over time; this was not observed in the absence of the background blood microbiome (the light blue line in Figure 3). However, the composition of the blood microbiome did not change, even when we discarded the background populations (i.e., the microbiome detected before the fecal induction) from the analysis. Genes and pathways enriched in a particular population are more relevant to microbiome function than bacterial composition (Gevers et al., 2014; Goodrich et al., 2014). In the porcine bacteremia model used in the current study, the blood is a unique environment for bacteria. It contains high levels of inﬂammatory cytokines and immune cells, with a unique composition of chemical compounds, such as lactic acid August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 7 Longitudinal Analysis of Blood Microbiome Hyun et al. (Park et al., 2019). Therefore, we performed PICRUSt2 analysis to identify the putative functions of the blood microbiome (Figure 4). In the analysis presented above, we conﬁrmed that new bacterial phyla have emerged into the bloodstream after peritonitis induction, and the populations of the native blood microbiome are notably different in each animal. So we utilized OTUs subtracting the native blood microbiome for this analysis to identify pathways essential for bacteria to survive in the bloodstream. Pathways related to ABC transporters, two- component systems, and oxidative phosphorylation were enriched in the blood microbiome. By contrast, pathways related to purine metabolism, pyrimidine metabolism, and ribosome expression were enriched to a lesser extent than observed in the feces. To validate these ﬁndings, we used Wilcoxon test to reveal pathways that are signiﬁcantly enriched or depleted. (Park et al., 2019). Therefore, we performed PICRUSt2 analysis to identify the putative functions of the blood microbiome (Figure 4). DISCUSSION Here, we analyzed the blood microbiome in a bacteremia-induced porcine model. Microbiota transmission between organs, presumably via the bloodstream, has been reported in previous studies (Dominy et al., 2019; Massier et al., 2020; Sookoian et al., 2020). However, based on these studies, it is unclear how the blood microbiome is maintained because these studies provide only a “snapshot” view of a single point in time. By monitoring the blood microbiome over time, we here conﬁrmed that each animal maintains a relatively consistent blood microbial population. Furthermore, by identifying potential pathways enriched in this population, we revealed that the bloodstream bacteria might have adapted to respond to the blood environment by using ABC transporters and two-component systems. On the other hand, the bacteria may not grow under these unfavorable conditions and, hence, pathways related to nucleotide biosynthesis may be suppressed under these conditions. The bacteremia model used herein yielded detectable bacteria in the blood after the induction (conﬁrmed by culturing) (Park et al., 2019). We, therefore, used these samples as a “positive control” to detect the blood microbiome. We observed that the composition of the blood microbiome changed gradually after peritonitis induction by autologous feces, with an over-representation of Bacteroidetes from the gut microbiota (Crespo-Piazuelo et al., 2019) slightly increasing over time (Figure 3). The gut microbiome can enter the bloodstream when the host is immunocompromised (Taur and Pamer, 2013). Hence, the observed over-representation could indicate a septic symptom of peritonitis. When we systematically traced the microbiota source, we found that new populations were introduced gradually into the bloodstream 4–6 h after the peritonitis induction (Figure 2), mirroring the clinical symptoms, such as increased cytokine production (Figure 6). However, it is not clear whether these newly introduced bacteria induced the septic symptoms or entered the blood because of sepsis. The biggest challenge to monitoring bacteria in the bloodstream is their low number compared with the microbiome at other body sites; such a low number means that even a minor contamination can have a major effect on the detection results (Eisenhofer et al., 2019; Karstens et al., 2019). To overcome this, we here selectively enriched bacteria from the blood using opsonin-coated MNPs (Kang et al., 2014) and then identiﬁed them by 16S rRNA gene sequencing. Biological Pathways Enriched in the Blood Microbiome Wilcoxon test was used to check the signiﬁcance of the enrichment and depletion of each pathway. denotes the p-value of the Wilcoxon test is less than 0.05. Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org August 2021 \| Volume 11 \| Article 676650 Longitudinal Analysis of Blood Microbiome Hyun et al. immune system (Barrett and Hoch, 1998; Kawada-Matsuo and Komatsuzawa, 2017). Therefore, it is likely that cells in the microbiome would utilize these systems to survive. by sequencing in the current study may have also been dormant. This may explain why we observed a discrepancy between the in vitro blood culture and the sequencing data. It is nonetheless surprising that similar (possibly dormant) bacterial populations were maintained in the blood over time, even after peritonitis induction, because the biological function of dormant bacteria is not well known. Further studies are required to understand their roles in the bloodstream. Conversely, purine metabolism, pyrimidine metabolism, and ribosome expression were suppressed in the blood microbiome, which may limit cell proliferation and growth (Samant et al., 2008; Polymenis and Aramayo, 2015; Shaffer et al., 2017). Because the bloodstream is a harsh environment, cells therein may downregulate essential metabolic functions to survive. Simultaneously, reducing cell proliferation or maintaining low metabolic activity (e.g., dormancy) may enable bacteria to escape immune surveillance in the blood (Lennon and Jones, 2011; Rittershaus et al., 2013). Another common source of contamination in blood microbiota studies is the skin microbiome. According to one study, Firmicutes (55.6% of relative abundance), Bacteroidetes (20.8%), Actinobacteria (13.3%), and Proteobacteria (5.1%) are representative resident phyla in the porcine skin microbiota (McIntyre et al., 2016). Among them, Staphylococcus is the dominant genus within the skin microbiota of animals (Kloos et al., 1976) and humans (Byrd et al., 2018; O’Sullivan et al., 2019). If contamination with the skin microbiome had occurred, we should have observed these bacteria consistently, even in blood culture, because we sampled the blood via a catheter. However, these bacteria were not a major component of the blood microbiome in the current study, except one case (P1219), which showed a moderate amount of Firmicutes (Figure 1). Considering the above, we concluded that the skin microbiome contamination was reasonably controlled to analyze the blood microbiome in the current study. Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org DISCUSSION Nonetheless, it could have been difﬁcult to distinguish the actual bacteria from bacterial DNA from the debris circulating in the bloodstream because the sequencing-based method is destructive (i.e., preparation for analysis involves bacterial cell lysis). Therefore, after capturing the bacteria with MNPs, we performed RNA-FISH targeting the common region within the bacterial 16S rRNA; this conﬁrmed that bacteria were indeed present in the bloodstream, even before the induction of peritonitis (Supplementary Figure 3). Because of cell membrane integrity and a high abundance of ribosomal RNAs inside the cell, and because it is unlikely that DNA debris from as similar bacterial species can be observed over time, we concluded that the signal from the dead bacteria is not prominent in the analysis. Surprisingly, the bacterial species identiﬁed by standard clinical laboratory culture testing were not the dominant species identiﬁed by culture-free bacterial population analysis, even though we observed a moderate association for the species and the time of detection (Figure 5). We speculate that culturable bacteria comprise only a small portion of the total blood microbiota. Further, the blood microbiome may enact a homeostatic mechanism that maintains the community members. The composition of the commensal microbiome is preserved upon exposure to exogenous bacteria (Littman and Pamer, 2011; Kamada et al., 2013; Khan and Shekhar, 2019), and the blood microbiome may operate a similar protective mechanism. Because the blood microbiome may not be metabolically active relative to other commensal bacteria, further study is required to examine this possibility. Dormant bacteria, which are metabolically suppressed and not immediately culturable, are quite common in the blood (Potgieter et al., 2015), and bacterial species that we identiﬁed August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 9 Longitudinal Analysis of Blood Microbiome Hyun et al. A B D E F C FIGURE 5 \| (A–F) Bacteria detected by blood culture. E. coli and other species (Streptococcus dysgalactiae, Enterococcus hirae, Streptococcus alactolyticus, Streptococcus gallolyticus, and Bacillus cereus) conﬁrmed previously by blood culture (Park et al., 2019) were also detected by blood microbiome sequencing from each sample. A B D C A D C D E F FIGURE 5 \| (A–F) Bacteria detected by blood culture. E. Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org DISCUSSION coli and other species (Streptococcus dysgalactiae, Enterococcus hirae, Streptococcus alactolyticus, Streptococcus gallolyticus, and Bacillus cereus) conﬁrmed previously by blood culture (Park et al., 2019) were also detected by blood microbiome sequencing from each sample. F F E cted by blood culture. E. coli and other species (Streptococcus dysgalactiae, Enterococcus hirae, Streptococcus alactolyticus, Bacillus cereus) conﬁrmed previously by blood culture (Park et al., 2019) were also detected by blood microbiome sequencing from FIGURE 5 \| (A–F) Bacteria detected by blood culture. E. coli and other species (Streptococcus dysgalactiae, Enterococcus hirae, Streptococcus alactolyticus, Streptococcus gallolyticus, and Bacillus cereus) conﬁrmed previously by blood culture (Park et al., 2019) were also detected by blood microbiome sequencing from each sample. As reported in our previous study (Park et al., 2019), of the 82 bacterial species identiﬁed by blood culture, 83% were Escherichia coli (57.3%) or Streptococcus (25.7%) species, which were also identiﬁed by sequencing (Figure 5). For example, we identiﬁed E. coli and Streptococcus dysgalactiae in animals P1016 and P1120 at the late stages after the peritonitis induction, and we observed a signiﬁcant increase in their relative abundance at the time point that coincided with blood culture positivity. In animal P1126, we detected Streptococcus alactolyticus only 7–8 h after the induction, and we detected E. coli at a later time point. Similarly, we observed a slight increase in the relative Streptococcus abundance 7–8 h after the induction in animal P1120. Moreover, we detected E. coli in The porcine peritonitis model used herein provides a unique opportunity to study the blood microbiome. Because of the large body size of pigs, the dynamics of the blood microbiome can be analyzed over time by multiple sampling, which is challenging when using a small animal model, such as a mouse. Although the animals used in the current study were not raised in an aseptic environment, the multiple sampling approach and computational methods for population comparisons used herein made it possible to comprehensively characterize the blood microbiota. Hence, this model may provide an excellent platform for developing new diagnostic techniques to detect BSI (Grumaz et al., 2016; Wanda et al., 2017; Horiba et al., 2018). August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 10 Longitudinal Analysis of Blood Microbiome Hyun et al. DATA AVAILABILITY STATEMENT The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcimb.2021. 676650/full#supplementary-material The datasets presented in this study can be found in the European Nucleotide Archive (ENA) under the accession ID PRJEB39083. DISCUSSION This could be attributed to that the sequencing might have detected bacterial DNA contained in white blood cells (Thwaites and Gant, 2011) or cell-free bacterial DNA (Wanda et al., 2017; August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Longitudinal Analysis of Blood Microbiome Hyun et al. ACKNOWLEDGMENTS We thank You Hwan Jo, Doyun Kim, Hyunglan Chang, and Hyuksool Kwon from the Seoul National University Bundang Hospital for helping with the animal experiments. ETHICS STATEMENT Previously, we also reported that all animals showed symptoms relevant to sepsis, which developed gradually 5–6 h after the fecal induction (Park et al., 2019). Here, in addition to comparison with the blood culture result, we compared the emergence time points of the altered blood microbiome with those of pro-inﬂammatory host responses. Using SourceTracker2 analysis, we found that the levels of proinﬂammatory cytokines (interleukin (IL)-1b and IL-6) in the blood increased as the new microbiome emerged in the blood (Figure 6). We calculated the Pearson correlation coefﬁcient p-value for those two parameters (the newly emerging bacteria and IL-1b or IL-6), and observed signiﬁcant correlations between those for all animals except for P1113. Although we cannot at this point conclude whether the new blood microbiome plays a role in developing a septic symptom in the porcine peritonitis model induced by fecal inoculation, the data indicate some involvement of the blood microbiome. The animal study was reviewed and approved by Seoul National University Bundang Hospital IACUC (BA1804-246/040-01). FUNDING This research was supported by grants from the National Research Foundation of Korea (2017M3A9E2062138 to TK, 2017M3A9E2062136 to JHK, 2017M3A9E2062210 to JHL); and by the Basic Science Research Program, through the National Research Foundation of Korea, funded by the Ministry of Education (2018R1A6A1A03025810 to TK), and partially by the Future-leading Project Research Fund of UNIST (1.210034.01 to TK and JHK). AUTHOR CONTRIBUTIONS HH performed the microbiome experiments and data analysis, assisted by SY and HSK. MSL prepared the blood microbiome samples and performed the FISH experiment. IP performed the animal experiments, assisted by D-HJ and SK. HK, JHK, JHL, and TK conceived the study and designed the experiments. JK, JL, and TK analyzed the data. All authors contributed to the article and approved the submitted version. Here, we reported changes in the composition of the blood microbiome in a porcine bacteremia model. By analyzing the blood microbiome in the same individual over time, we showed that the bacterial population remains relatively consistent, even after peritonitis induction. However, at the same time, we found that new bacterial populations entered the bloodstream, with the dynamic patterns similar to those observed during a physiological response to BSI (e.g., cytokine level increase). Further, by analyzing population-enriched pathways, we conﬁrmed that sensing mechanisms, such as ABC transporters and two-component systems, are upregulated in the blood microbiome. Conversely, central nucleotide metabolism, essential for cell proliferation and growth, was suppressed in these bacteria, which probably helps the blood microbiome to survive the harsh bloodstream environment and escape immune surveillance. Finally, the current study indicates that further investigations of the blood microbiome are required to improve the current diagnostic approaches for BSI. DISCUSSION A B D E F C FIGURE 6 \| (A–F) Association between cytokine induction and introduction of new bacterial species into the bloodstream. Measurement of IL-1b and IL-6 levels in the blood revealed that most animals began to develop septic responses between 4 and 8 h after fecal induction. The proportion of newly emerging bacteria was determined by SourceTracker2 (Figure 2) by comparing the altered microbiome with the microbiome detected before peritonitis induction. The original cytokine data were published previously (Park et al., 2019). Pearson correlation coefﬁcient p-value was calculated to verify the correlation between the newly emerging bacteria and cytokine markers. A A B B B C D D C E F F E F FIGURE 6 \| (A–F) Association between cytokine induction and introduction of new bacterial species into the bloodstream. Measurement of IL-1b and IL-6 levels in the blood revealed that most animals began to develop septic responses between 4 and 8 h after fecal induction. The proportion of newly emerging bacteria was determined by SourceTracker2 (Figure 2) by comparing the altered microbiome with the microbiome detected before peritonitis induction. The original cytokine data were published previously (Park et al., 2019). Pearson correlation coefﬁcient p-value was calculated to verify the correlation between the newly emerging bacteria and cytokine markers. Camargo et al., 2020). Although microbial DNA can induce inﬂammation and other host responses, it is a false positive in the analysis of the blood microbiome. Another explanation may be dormant bacteria, which cannot be cultured under test conditions (Lennon and Jones, 2011; Rittershaus et al., 2013; Potgieter et al., 2015). Because dormant bacteria can become reactivated depending on the environmental conditions, they can contribute to the blood microbiome function even if they are not immediately culturable. Further investigation is required to explain these discrepancies. animal P1219 at an early stage after the induction, and the sequencing data supported a high proportion of E. coli throughout the experiment of P1219. On the other hand, based on the microbiome sequencing data, bacterial species conﬁrmed by blood culture were not the most abundant in the bloodstream, even 12 h after the peritonitis induction. Cole, J. R., Wang, Q., Fish, J. A., Chai, B., McGarrell, D. M., Sun, Y., et al. (2014). Ribosomal Database Project: Data and Tools for High Throughput rRNA Analysis. Nucl. Acids Res. 42, D633–D642. doi: 10.1093/nar/gkt1244 Crespo-Piazuelo, D., Migura-Garcia, L., Estellé, J., Criado-Mesas, L., Revilla, M., Castelló, A., et al. (2019). Association Between the Pig Genome and Its Gut Microbiota Composition. Sci. Rep. 9, 8791. doi: 10.1038/s41598-019- 45066-6 Castillo, D. J., Rifkin, R. F., Cowan, D. A., and Potgieter, M. (2019). The Healthy Human Blood Microbiome: Fact or Fiction? Front. Cell. Infect. Microbiol. 9, 148. doi: 10.3389/fcimb.2019.00148 REFERENCES 191, 7353–7362. doi: 10.1128/JB.01053-09 Eisenhofer, R., Minich, J. J., Marotz, C., Cooper, A., Knight, R., and Weyrich, L. S. (2019). Contamination in Low Microbial Biomass Microbiome Studies: Issues and Recommendations. Trends Microbiol. 27, 105–117. doi: 10.1016/j.tim.2018.11.003 tresses. J. Bacteriol. 191, 7353–7362. doi: 10.1128/JB.01053-09 Mascher, T., Helmann, J. D., and Unden, G. (2006). Stimulus Perception in ascher, T., Helmann, J. D., and Unden, G. (2006). Stimulus P Mascher, T., Helmann, J. D., and Unden, G. (2006). Stimulus Perception in Bacterial Signal-Transducing Histidine Kinases. Microbiol. Mol. Biol. Rev. 70, 910–938. doi: 10.1128/MMBR.00020-06 Gao, R., and Lynn, D. G. (2005). Environmental pH Sensing: Resolving the VirA/ VirG Two-Component System Inputs for Agrobacterium Pathogenesis. J. Bacteriol. 187, 2182–2189. doi: 10.1128/JB.187.6.2182-2189.2005 Bacterial Signal-Transducing Histidine Kinases. Microbiol. Mol. Biol. Rev. 70, 910–938. doi: 10.1128/MMBR.00020-06 Massier, L., Chakaroun, R., Tabei, S., Crane, A., Didt, K. D., Fallmann, J., et al. (2020). Adipose Tissue Derived Bacteria are Associated With Inﬂammation in Obesity and Type 2 Diabetes. Gut 69 (10), 1796–1806. doi: 10.1136/gutjnl- 2019-320118. gutjnl-2019-320118. Gebhard, S. (2012). ABC Transporters of Antimicrobial Peptides in Firmicutes Bacteria Gebhard, S. (2012). ABC Transporters of Antimicrobial Peptides in Firmicutes Bacteria - Phylogeny, Function and Regulation: Antimicrobial Peptide Transport in Firmicutes. Mol. Microbiol. 86, 1295–1317. doi: 10.1111/mmi.12078 p p - Phylogeny, Function and Regulation: Antimicrobial Peptide Transport in Firmicutes. Mol. Microbiol. 86, 1295–1317. doi: 10.1111/mmi.12078 Mattos-Graner, R. O., and Duncan, M. J. (2017). Two-Component Signal Transduction Systems in Oral Bacteria. J. Oral Microbiol. 9, 1400858. doi: 10.1080/20002297.2017.1400858 Gevers, D., Kugathasan, S., Denson, L. A., Vázquez-Baeza, Y., Van Treuren, W., Ren, B., et al. (2014). The Treatment-Naive Microbiome in New-Onset Crohn’s Disease. Cell Host Microbe 15, 382–392. doi: 10.1016/j.chom.2014.02.005 McIntyre, M. K., Peacock, T. J., Akers, K. S., and Burmeister, D. M. (2016). Initial Characterization of the Pig Skin Bacteriome and Its Effect on In Vitro Models of Wound Healing. PloS One 11, e0166176. doi: 10.1371/journal.pone.0166176 Goodrich, J. K., Waters, J. L., Poole, A. C., Sutter, J. L., Koren, O., Blekhman, R., et al. (2014). Human Genetics Shape the Gut Microbiome. Cell 159, 789–799. doi: 10.1016/j.cell.2014.09.053 Goto, M., and Al-Hasan, M. N. (2013). Overall Burden of Bloodstream Infection and Nosocomial Bloodstream Infection in North America and Europe. Clin. Microbiol. Infec. 19, 501–509. doi: 10.1111/1469-0691.12195 Messenger, A. J., and Barclay, R. (1983). Bacteria, Iron and Pathogenicity. Biochem. Educ. 11, 54–63. doi: 10.1016/0307-4412(83)90043-2 Nannan Panday, R. S., Lammers, E. M. J., Alam, N., and Nanayakkara, P. REFERENCES W. B. (2019). An Overview of Positive Cultures and Clinical Outcomes in Septic Patients: A Sub-Analysis of the Prehospital Antibiotics Against Sepsis (PHANTASi) Trial. Crit. Care 23, 182. doi: 10.1186/s13054-019-2431-8 Grumaz, S., Stevens, P., Grumaz, C., Decker, S. O., Weigand, M. A., Hofer, S., et al. (2016). Next-Generation Sequencing Diagnostics of Bacteremia in Septic Patients. Genome Med. 8, 73. doi: 10.1186/s13073-016-0326-8 Horiba, K., Kawada, J., Okuno, Y., Tetsuka, N., Suzuki, T., Ando, S., et al. (2018). Comprehensive Detection of Pathogens in Immunocompromised Children With Bloodstream Infections by Next-Generation Sequencing. Sci. Rep. 8, 3784. doi: 10.1038/s41598-018-22133-y Okuda, K., Yoshii, Y., Yamada, S., Chiba, A., Hironaka, I., Hori, S., et al. (2018). Detection of Bacterial DNA From Central Venous Catheter Removed From Patients by Next Generation Sequencing: A Preliminary Clinical Study. Ann. Clin. Microbiol. Antimicrob. 17, 44. doi: 10.1186/s12941-018-0297-2 Kamada, N., Chen, G. Y., Inohara, N., and Núñez, G. (2013). Control of Pathogens and Pathobionts by the Gut Microbiota. Nat. Immunol. 14, 685–690. doi: 10.1038/ni.2608 Ortiz de Orué Lucana, D., Wedderhoff, I., and Groves, M. R. (2012). ROS- Mediated Signalling in Bacteria: Zinc-Containing Cys-X-X-Cys Redox Centres and Iron-Based Oxidative Stress. J. Signal Transduct. 2012, 1–9. doi: 10.1155/ 2012/605905 Kang, J. H., Super, M., Yung, C. W., Cooper, R. M., Domansky, K., Graveline, A. R., et al. (2014). An Extracorporeal Blood-Cleansing Device for Sepsis Therapy. Nat. Med. 20, 1211–1216. doi: 10.1038/nm.3640 O’Sullivan, J. N., Rea, M. C., O’Connor, P. M., Hill, C., and Ross, R. P. (2019). Human Skin Microbiota is a Rich Source of Bacteriocin-Producing Staphylococci That Kill Human Pathogens. FEMS Microbiol. Ecol. 95. doi: 10.1093/femsec/ﬁy241 Karstens, L., Asquith, M., Davin, S., Fair, D., Gregory, W. T., Wolfe, A. J., et al. (2019). Controlling for Contaminants in Low-Biomass 16s rRNA Gene Sequencing Experiments. mSystems 4, e00290–e00219. doi: 10.1128/ mSystems.00290-19 Park, I., Lee, J. H., Jang, D.-H., Kim, D., Chang, H., Kwon, H., et al. (2019). Characterization of Fecal Peritonitis–Induced Sepsis in a Porcine Model. J. Surg. Res. 244, 492–501. doi: 10.1016/j.jss.2019.06.094 Kawada-Matsuo, M., and Komatsuzawa, H. (2017). Role of Streptococcus Mutans Two-Component Systems in Antimicrobial Peptide Resistance in the Oral Cavity. Jpn. Dent. Sci. Rev. 53, 86–94. doi: 10.1016/j.jdsr.2016.12.002 Parks, D. H., Chuvochina, M., Waite, D. W., Rinke, C., Skarshewski, A., Chaumeil, P.-A., et al. (2018). A Standardized Bacterial Taxonomy Based on Genome Phylogeny Substantially Revises the Tree of Life. Nat. Biotechnol. 36, 996–1004. doi: 10.1038/nbt.4229 Khan, R., and Shekhar, S. (2019). REFERENCES Commensal Bacteria: An Emerging Player in Defense Against Respiratory Pathogens. Front. Immunol. 10, 9. doi: 10.3389/ ﬁmmu.2019.01203 Pien, B. C., Sundaram, P., Raoof, N., Costa, S. F., Mirrett, S., Woods, C. W., et al. (2010). The Clinical and Prognostic Importance of Positive Blood Cultures in Pien, B. C., Sundaram, P., Raoof, N., Costa, S. F., Mirrett, S., Woods, C. W., et al. (2010). The Clinical and Prognostic Importance of Positive Blood Cultures in Adults. Am. J. Med. 123, 819–828. doi: 10.1016/j.amjmed.2010.03.021 Kloos, W. E., Zimmerman, R. J., and Smith, R. F. (1976). Preliminary Studies on the Characterization and Distribution of Staphylococcus and Micrococcus Species on Animal Skin. Appl. Environ. Microbiol. 31, 53. doi: 10.1128/ aem.31.1.53-59.1976 Polymenis, M., and Aramayo, R. (2015). Translate to Divide: Сontrol of the Cell Cycle by Protein Synthesis. MIC 2, 94–104. doi: 10.15698/mic2015.04.198 Knights, D., Kuczynski, J., Charlson, E. S., Zaneveld, J., Mozer, M. C., Collman, R. G., et al. (2011). Bayesian Community-Wide Culture-Independent Microbial Source Tracking. Nat. Methods 8, 761–763. doi: 10.1038/nmeth.1650 Potgieter, M., Bester, J., Kell, D. B., and Pretorius, E. (2015). The Dormant Blood Microbiome in Chronic, Inﬂammatory Diseases. FEMS Microbiol. Rev. 39, 567–591. doi: 10.1093/femsre/fuv013 Rittershaus, E. S. C., Baek, S.-H., and Sassetti, C. M. (2013). The Normalcy of Dormancy: Common Themes in Microbial Quiescence. Cell Host Microbe 13, 643–651. doi: 10.1016/j.chom.2013.05.012 Lamy, B., Dargère, S., Arendrup, M. C., Parienti, J.-J., and Tattevin, P. (2016). How to Optimize the Use of Blood Cultures for the Diagnosis of Bloodstream Infections? A State-Of-the Art. Front. Microbiol. 7, 697. doi: 10.3389/ fmicb.2016.00697 Rognes, T., Flouri, T., Nichols, B., Quince, C., and Mahé, F. (2016). VSEARCH: A Versatile Open Source Tool for Metagenomics. PeerJ 4, e2584. doi: 10.7717/ peerj.2584 Laupland, K. B. (2013). Incidence of Bloodstream Infection: A Review of Population- Based Studies. Clin. Microbiol. Infection 19, 492–500. doi: 10.1111/1469-0691.12144 Salter, S. J., Cox, M. J., Turek, E. M., Calus, S. T., Cookson, W. O., Moffatt, M. F., et al. (2014). Reagent and Laboratory Contamination can Critically Impact Sequence-Based Microbiome Analyses. BMC Biol. 12, 87. doi: 10.1186/s12915- 014-0087-z Laupland, K. B., Pasquill, K., Dagasso, G., Parﬁtt, E. C., Steele, L., and Schonheyder, H. C. (2020). Population-Based Risk Factors for Community- Onset Bloodstream Infections. Eur. J. Clin. Microbiol. Infect. Dis. 39, 753–758. doi: 10.1007/s10096-019-03777-8 Lennon, J. T., and Jones, S. E. (2011). Microbial Seed Banks: The Ecological and Evolutionary Implications of Dormancy. Nat. Rev. Microbiol. 9, 119–130. REFERENCES Castillo, D. J., Rifkin, R. F., Cowan, D. A., and Potgieter, M. (2019). The Healthy Human Blood Microbiome: Fact or Fiction? Front. Cell. Infect. Microbiol. 9, 148. doi: 10.3389/fcimb.2019.00148 Barrett, J. F., and Hoch, J. A. (1998). Two-Component Signal Transduction as a Target for Microbial Anti-Infective Therapy. Antimicrob. Agents Chemother. 42, 1529–1536. doi: 10.1128/AAC.42.7.1529 Cecconi, M., Evans, L., Levy, M., and Rhodes, A. (2018). Sepsis and Septic Shock. Lancet 392, 75–87. doi: 10.1016/S0140-6736(18)30696-2 Byrd, A. L., Belkaid, Y., and Segre, J. A. (2018). The Human Skin Microbiome. Nat. Rev. Microbiol. 16, 143–155. doi: 10.1038/nrmicro.2017.157 Cole, J. R., Wang, Q., Fish, J. A., Chai, B., McGarrell, D. M., Sun, Y., et al. (2014). Ribosomal Database Project: Data and Tools for High Throughput rRNA Analysis. Nucl. Acids Res. 42, D633–D642. doi: 10.1093/nar/gkt1244 Camargo, J. F., Ahmed, A. A., Lindner, M. S., Morris, M. I., Anjan, S., Anderson, A. D., et al. (2020). Next-Generation Sequencing of Microbial Cell-Free DNA for Rapid Noninvasive Diagnosis of Infectious Diseases in Immunocompromised Hosts. F1000Res. 8, 1194. doi: 10.12688/ f1000research.19766.4 Crespo-Piazuelo, D., Migura-Garcia, L., Estellé, J., Criado-Mesas, L., Revilla, M., Castelló, A., et al. (2019). Association Between the Pig Genome and Its Gut Microbiota Composition. Sci. Rep. 9, 8791. doi: 10.1038/s41598-019- 45066-6 August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 12 12 Longitudinal Analysis of Blood Microbiome Hyun et al. Dominy, S. S., Lynch, C., Ermini, F., Benedyk, M., Marczyk, A., Konradi, A., et al. (2019). Porphyromonas Gingivalis in Alzheimer’s Disease Brains: Evidence for Disease Causation and Treatment With Small-Molecule Inhibitors. Sci. Adv. 5, eaau3333. doi: 10.1126/sciadv.aau3333 Levy, O. (2000). Antimicrobial Proteins and Peptides of Blood: Templates for Novel Antimicrobial Agents. Blood 96, 2664–2672. doi: 10.1182/ blood.V96.8.2664 Littman, D. R., and Pamer, E. G. (2011). Role of the Commensal Microbiota in Normal and Pathogenic Host Immune Responses. Cell Host Microbe 10, 311– 323. doi: 10.1016/j.chom.2011.10.004 Douglas, G. M., Maffei, V. J., Zaneveld, J. R., Yurgel, S. N., Brown, J. R., Taylor, C. M., et al. (2020). PICRUSt2 for Prediction of Metagenome Functions. Nat. Biotechnol. 38, 685–688. doi: 10.1038/s41587-020-0548-6 Liu, Y., and Burne, R. A. (2009). Multiple Two-Component Systems Modulate Alkali Generation in Streptococcus Gordonii in Response to Environmental Liu, Y., and Burne, R. A. (2009). Multiple Two-Component Systems Modulate Alkali Generation in Streptococcus Gordonii in Response to Environmental Stresses. J. Bacteriol. REFERENCES doi: 10.1038/nrmicro2504 Samant, S., Lee, H., Ghassemi, M., Chen, J., Cook, J. L., Mankin, A. S., et al. (2008). Nucleotide Biosynthesis Is Critical for Growth of Bacteria in Human Blood. PloS Pathog. 4, e37. doi: 10.1371/journal.ppat.0040037 August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org 13 Longitudinal Analysis of Blood Microbiome Hyun et al. With Bloodstream Infection (BSI) Using the Karius Plasma Next Generation Sequencing (NGS) Test. Open Forum Infect. Dis. 4, S613–S613. doi: 10.1093/ oﬁd/ofx163.1613 With Bloodstream Infection (BSI) Using the Karius Plasma Next Generation Sequencing (NGS) Test. Open Forum Infect. Dis. 4, S613–S613. doi: 10.1093/ oﬁd/ofx163.1613 Shaffer, C. L., Zhang, E. W., Dudley, A. G., Dixon, B. R. E. A., Guckes, K. R., Breland, E. J., et al. (2017). Purine Biosynthesis Metabolically Constrains Intracellular Survival of Uropathogenic Escherichia Coli. Infect. Immun. 85, e00471–e00416. doi: 10.1128/IAI.00471-16 Wang, Q., Garrity, G. M., Tiedje, J. M., and Cole, J. R. (2007). Naive Bayesian Classiﬁer for Rapid Assignment of rRNA Sequences Into the New Bacterial Taxonomy. Appl. Environ. Microbiol. 73, 5261–5267. doi: 10.1128/ AEM.00062-07 Slater, G., and Birney, E. (2005). Automated Generation of Heuristics for Biological Sequence Comparison. BMC Bioinf. 6, 31. doi: 10.1186/1471-2105-6-31 Sookoian, S., Salatino, A., Castaño, G. O., Landa, M. S., Fijalkowky, C., Garaycoechea, M., et al. (2020). Intrahepatic Bacterial Metataxonomic Signature in non-Alcoholic Fatty Liver Disease. Gut 69, 1483–1491. doi: 10.1136/gutjnl-2019-318811 Weinstein, M. P., Towns, M. L., Quartey, S. M., Mirrett, S., Reimer, L. G., Parmigiani, G., et al. (1997). The Clinical Signiﬁcance of Positive Blood Cultures in the 1990s: A Prospective Comprehensive Evaluation of the Microbiology, Epidemiology, and Outcome of Bacteremia and Fungemia in Adults. Clin. Infect. Dis. 24, 584–602. doi: 10.1093/clind/24.4.584 Stinson, L. F., Keelan, J. A., and Payne, M. S. (2019). Identiﬁcation and Removal of Contaminating Microbial DNA From PCR Reagents: Impact on Low-Biomass Microbiome Analyses. Lett. Appl. Microbiol. 68, 2–8. doi: 10.1111/lam.13091 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Taur, Y., and Pamer, E. G. (2013). The Intestinal Microbiota and Susceptibility to Infection in Immunocompromised Patients. Curr. Opin. Infect. Dis. 26, 332– 337. doi: 10.1097/QCO.0b013e3283630dd3 Tedeschi, G. G., Amici, D., and Paparelli, M. (1969). Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org REFERENCES Incorporation of Nucleosides and Amino-Acids in Human Erythrocyte Suspensions: Possible Relation With a Diffuse Infection of Mycoplasms or Bacteria in the L Form. Nature 222, 1285–1286. doi: 10.1038/2221285a0 Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Thwaites, G. E., and Gant, V. (2011). Are Bloodstream Leukocytes Trojan Horses for the Metastasis of Staphylococcus Aureus? Nat. Rev. Microbiol. 9, 215–222. doi: 10.1038/nrmicro2508 Trautner, B. W., and Darouiche, R. O. (2004). Catheter-Associated Infections: Pathogenesis Affects Prevention. Arch. Intern. Med. 164, 842. doi: 10.1001/ archinte.164.8.842 Copyright © 2021 Hyun, Lee, Park, Ko, Yun, Jang, Kim, Kim, Kang, Lee and Kwon. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Viscoli, C. (2016). Bloodstream Infections: The Peak of the Iceberg. Virulence 7, 248–251. doi: 10.1080/21505594.2016.1152440 Wanda, L., Rufﬁn, F., Hill-Rorie, J., Hollemon, D., Seng, H., Hong, D., et al. (2017). Direct Detection and Quantiﬁcation of Bacterial Cell-Free DNA in Patients August 2021 \| Volume 11 \| Article 676650 Frontiers in Cellular and Infection Microbiology \| www.frontiersin.org
https://openalex.org/W4304187520	https://discovery.ucl.ac.uk/id/eprint/10157436/2/Chau_pharmaceutics-14-02162.pdf	English	null	Therapeutic Application of an Ag-Nanoparticle-PNIPAAm-Modified Eggshell Membrane Construct for Dermal Regeneration and Reconstruction	Pharmaceutics	2,022	cc-by	9,775	Therapeutic Application of an Ag-Nanoparticle-PNIPAAm- Modiﬁed Eggshell Membrane Construct for Dermal Regeneration and Reconstruction Emily Briggs 1,2, Rosemond A. Mensah 1, Kapil D. Patel 1,3,4 , Nandin-Erdene Mandakhbayar 4, Nik San Sharifulden 1, Zalike Keskin Erdogan 1 , Lady V. Barrios Silva 1, Kawther Salim 1, Hae-Won Kim 4,5,6,7, Linh T. B. Nguyen 1 and David Y. S. Chau 1,5,6,7,* 1 Eastman Dental Institute, University College London, Royal Free Hospital, Rowland Hill Street, London NW3 2PF, UK 2 Department of Materials, Henry Royce Institute, The University of Manchester, Rumford Street, Manchester M13 9PL, UK 3 School of Cellular and Molecular Medicine, University of Bristol, Queens Road, Bristol BS8 1QU, UK 4 Institute of Tissue Regeneration Engineering, Dankook University, Cheonan 31116, Korea 5 UCL Eastman-Korea Dental Medicine Innovation Centre, Dankook University, Cheonan 31116, Korea 6 Department of Nanobiomedical Science, Dankook University, Cheonan 31116, Korea 7 BK21 NBM Global Research Centre for Regenerative Medicine, Dankook University, Cheonan 31116, Korea * Correspondence: d.chau@ucl.ac.uk 3 School of Cellular and Molecular Medicine, University of Bristol, Queens Road, Bristol BS8 1QU, UK 4 Institute of Tissue Regeneration Engineering, Dankook University, Cheonan 31116, Korea 5 UCL Eastman-Korea Dental Medicine Innovation Centre, Dankook University, Cheonan 31116, Korea 6 Department of Nanobiomedical Science, Dankook University, Cheonan 31116, Korea 7 BK21 NBM Global Research Centre for Regenerative Medicine, Dankook University, Cheonan 31116, Korea * Correspondence: d.chau@ucl.ac.uk Abstract: Current therapeutic treatments for the repair and/or replacement of damaged skin following disease or traumatic injury is severely limited. The chicken eggshell membrane (ESM) is a unique material: its innate physical and mechanical characteristics offer optimal barrier properties and, as a naturally derived extract, it demonstrates inherent biocompatibility/biodegradability. To further enhance its therapeutic and clinical potential, the ESM can be modified with the thermo-responsive polymer, poly(N- isopropylacrylAmide) (PNIPAAm) as well as the incorporation of (drug-loaded) silver nanoparticles (AgNP); essentially, by a simple change in temperature, the release and delivery of the NP can be targeted and controlled. In this study, ESM samples were isolated using a decellularization protocol, and the physical and mechanical characteristics were profiled using SEM, FT-IR, DSC and DMA. PNIPAAm was successfully grafted to the ESM via amidation reactions and confirmed using FT-IR, which demonstrated the distinctive peaks associated with Amide A (3275 cm−1), Amide B (2970 cm−1), Amide I (1630 cm−1), Amide II (1535 cm−1), CH2, CH3 groups, and Amide III (1250 cm−1) peaks. Therapeutic Application of an Ag-Nanoparticle-PNIPAAm- Modiﬁed Eggshell Membrane Construct for Dermal Regeneration and Reconstruction Confirmation of the incorporation of AgNP onto the stratified membrane was confirmed visually with SEM, qualitatively using FT-IR and also via changes in absorbance at 380 nm using UV-Vis spectrophotometry during a controlled release study for 72 h. The biocompatibility and cytotoxicity of the novel constructs were assessed using human dermal fibroblast (HDFa) and mouse dermal fibroblast (L929) cells and standard cell culture assays. Metabolic activity assessment (i.e., MTS assay), LDH-release profiles and Live/Dead staining demonstrated good attachment and spreading to the samples, and high cell viability following 3 days of culture. Interestingly, longer-term viability (>5 days), the ESM-PNIPAAm and ESM-PNIPAAm (AgNP) samples showed a greater and sustained cell viability profile. In summary, the modified and enhanced ESM constructs were successfully prepared and characterized in terms of their physical and mechanical profiles. AgNP were successfully loaded into the construct and demonstrated a desirable release profile dependent on temperature modulation. Fibroblasts cultured on the extracted ESM samples and ESM-PNIPAAm demonstrated high biocompatibility in terms of high cell attachment, spreading, viability and proliferation rates. As such, this work summarizes the development of an enhanced ESM-based construct which may be exploited as a clinical/therapeutic wound dressing as well as a possible application as a novel biomaterial scaffold for drug development. pharmaceutics pharmaceutics pharmaceutics pharmaceutics pharmaceutics 1. Introduction Damage to the skin can result from several insults, including burns, traumatic injury, cancer, and infection. Physiologically, wound healing is a multi-parameter and complex process which is dependent on the synchronised activation of several cell types and cy- tokine mediators [1]. Furthermore, pathologies such as diabetes mellitus or peripheral vascular disease can result in chronic wounds, which can further develop and/or increase the potential of infection and additional complications [2,3]. During aberrant tissue re- pair, impaired angiogenesis and nutrient delivery result in persistent infection and the continuous activation of inﬂammatory cells in the wound bed. As such, patients suffer from non-healing open wounds (ulcers), which require continuous clinical management. Annually, the cost of managing such wounds is estimated to be £5.3 billion in the UK [4] and further studies have reported that the frequency of such wounds to be increasing at a rate of 12% per year, highlighting the critical requirement for strategies to improve healing rates in chronic dermal wounds [5]. Traditional dressings, made of cotton gauze, fail to provide a moist environment and often become adherent to the wound bed [6]. Modern dressings have been developed to im- prove healing by facilitating regeneration by providing a permeable and moist environment. A range of materials have been explored, including natural and synthetic hydrocolloids [7], semi-permeable hydro ﬁlms [8], and drug-loaded therapeutic dressings [9,10]. The eggshell membrane (ESM) is a highly collagenous, thin, ﬁbrous membrane found sandwiched between the calciﬁed shell and albumin of chicken eggs (Figure 1a). Consisting of an inner and an outer layer, the membranes are arranged to form a semi-permeable bar- rier [11]. Recent studies suggest the material’s physical properties and inherent biocompati- bility make it an ideal candidate to enhance tissue regeneration in corneal/ophthalmologic application [12]. In a preliminary study, Yang et al. report the direct application of the ESM to the site of skin grafts enhanced pain management and wound protection [13]. In addition, both Jun et al. [14] and Guarderas et al. [15] reported (native) ESM accelerates wound closure and regeneration in the early stages of wound healing. Interestingly, as ESM is composed of collagenous material, it may be possible to further enhance the me- chanical properties using traditional crosslinking methods, e.g., glutaraldehyde, NHS-EDC, transglutaminase and physical techniques [16], and is currently an area of research interest. In contrast, the membrane can additionally be modiﬁed to further enhance its thera- peutic potential. Citation: Briggs, E.; Mensah, R.A.; Patel, K.D.; Mandakhbayar, N.-E.; Sharifulden, N.S.; Erdogan, Z.K.; Silva, L.V.B.; Salim, K.; Kim, H.-W.; Nguyen, L.T.B.; et al. Therapeutic Application of an Ag-Nanoparticle- PNIPAAm-Modiﬁed Eggshell Membrane Construct for Dermal Regeneration and Reconstruction. Pharmaceutics 2022, 14, 2162. https://doi.org/10.3390/ pharmaceutics14102162 Academic Editor: Xiangyang Shi Received: 17 September 2022 Accepted: 8 October 2022 Published: 11 October 2022 Citation: Briggs, E.; Mensah, R.A.; Patel, K.D.; Mandakhbayar, N.-E.; Sharifulden, N.S.; Erdogan, Z.K.; Silva, L.V.B.; Salim, K.; Kim, H.-W.; Nguyen, L.T.B.; et al. Therapeutic Application of an Ag-Nanoparticle- PNIPAAm-Modiﬁed Eggshell Membrane Construct for Dermal Regeneration and Reconstruction. Pharmaceutics 2022, 14, 2162. https://doi.org/10.3390/ pharmaceutics14102162 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Keywords: drug-delivery; biomaterial; crosslinking; thermoresponsive; transglutaminase; wound dressing; eggshell membrane https://www.mdpi.com/journal/pharmaceutics Pharmaceutics 2022, 14, 2162. https://doi.org/10.3390/pharmaceutics14102162 2 of 13 Pharmaceutics 2022, 14, 2162 1. Introduction Herein, we report the modiﬁcation of the ESM with the stimuli-responsive hydrogel polymer poly(N-isopropylacrylAmide) (PNIPAAm), alongside the incorporation of drug-loaded nanoparticles (NP). PNIPAAm is a widely studied thermoresponsive hy- drogel: the key attribute being its lower critical solution temperature (LCST) of ~32 ◦C [17] which, accordingly, causes the polymer to undergo a phase transition to a hydrophobic globule conformation at temperatures above its LCST. As such, surfaces coated with PNI- PAAm can therefore alternate between a hydrophilic and a hydrophobic state depending upon changes to the external temperature, causing adherent agents, such as antibacterial silver NP (AgNP) [18], to become detached and consequently released into the environ- ment [19,20]. However, despite being studied as a vehicle for drug delivery systems, applications that demonstrate the use of PNIPAAm-based hydrogels systems in wound healing are limited. Although widely available as food waste, few articles report the complete characterisa- tion of the ESM as a potential material for biomedical applications. Furthermore, to date, no studies have proposed the application of the ESM grafted with an AgNP-loaded PNIPAAm hydrogel. The work herein summarises the isolation and characterisation of a variety of modiﬁed ESM samples in terms of their physical, chemical, biological and drug delivery proﬁles in the context of potential therapeutic applications to (dermal) wound healing. 3 of 13 Pharmaceutics 2022, 14, 2162 Figure 1. Schematic outing the fabrication of the ESM‐PNIPAAm (AgNP). (a) Diagram of chicken (gallus gallus) eggshell membrane. Inset shows the image of Raw ESM extracted with acetic acid (0.5 M) at room temperature (19 °C). (b) PNIPAAm is grafted to the surface of the ESM via the amidation reaction between the NH2‐rich ESM and the COOH‐terminated PNIPAAm hydrogel. The AgNPs are “loaded” onto the conjugated membranes at 4 °C (<LCST of PNIPAAm). Upon in‐ cubation at 37 °C, the hydrogel undergoes phase transition, and the NP are encapsulated within the crosslinked gel. In principle, when the temperature is lowered (<LCST), the hydrogel becomes hy‐ drophilic, and the NP are release into the environment. 2 Materials and Methods Figure 1. Schematic outing the fabrication of the ESM-PNIPAAm (AgNP). (a) Diagram of chicken (gallus gallus) eggshell membrane. Inset shows the image of Raw ESM extracted with acetic acid (0.5 M) at room temperature (19 ◦C). (b) PNIPAAm is grafted to the surface of the ESM via the amidation reaction between the NH2-rich ESM and the COOH-terminated PNIPAAm hydrogel. 1. Introduction The AgNPs are “loaded” onto the conjugated membranes at 4 ◦C (<LCST of PNIPAAm). Upon incubation at 37 ◦C, the hydrogel undergoes phase transition, and the NP are encapsulated within the crosslinked gel. In principle, when the temperature is lowered (<LCST), the hydrogel becomes hydrophilic, and the NP are release into the environment. igure 1. Schematic outing the fabrication of the ESM‐PNIPAAm (AgNP). (a) Diagram of chicken gallus gallus) eggshell membrane. Inset shows the image of Raw ESM extracted with acetic acid 0.5 M) at room temperature (19 °C). (b) PNIPAAm is grafted to the surface of the ESM via the midation reaction between the NH2‐rich ESM and the COOH‐terminated PNIPAAm hydrogel. he AgNPs are “loaded” onto the conjugated membranes at 4 °C (<LCST of PNIPAAm). Upon in‐ ubation at 37 °C, the hydrogel undergoes phase transition, and the NP are encapsulated within the rosslinked gel. In principle, when the temperature is lowered (<LCST), the hydrogel becomes hy‐ rophilic, and the NP are release into the environment. i l d h d Figure 1. Schematic outing the fabrication of the ESM-PNIPAAm (AgNP). (a) Diagram of chicken (gallus gallus) eggshell membrane. Inset shows the image of Raw ESM extracted with acetic acid (0.5 M) at room temperature (19 ◦C). (b) PNIPAAm is grafted to the surface of the ESM via the amidation reaction between the NH2-rich ESM and the COOH-terminated PNIPAAm hydrogel. The AgNPs are “loaded” onto the conjugated membranes at 4 ◦C (<LCST of PNIPAAm). Upon incubation at 37 ◦C, the hydrogel undergoes phase transition, and the NP are encapsulated within the crosslinked gel. In principle, when the temperature is lowered (<LCST), the hydrogel becomes hydrophilic, and the NP are release into the environment. 2.1. Membrane Isolation 2.1. Membrane Isolation The ESM was derived from large chicken eggs; the eggshell was removed using an optimised decellularization protocol, as previously reported [12]. Eggs were submerged in acetic acid (AA, 0.5 M) at room temperature (~19 ◦C) for 44 h before being removed from the acid and rinsed in distilled water. Remaining residual shell (calcium carbonate) was manually removed. The albumin and yolk were removed, and the remaining membrane carefully rinsed. Membranes were refrigerated (4 ◦C) in distilled water in order to avoid dehydration. 2.2. Membrane Modiﬁcation ESM-PNIPAAm (AgNP) Fabrication. The ESM-PNIPAAm was fabricated as previ- ously reported by Nguyen et al. [21], outlined in Figure 1b. An activation solution was prepared by adding MES hydrate (0.195 g, 0.05 M), NHS (0.138 g, 0.06 M) and EDC (0.46 g, 0.12 M) to 20 mL of deionised water. The pH of the solution (Solution A) was conﬁrmed to be 6 using pH indicator strips (PANPEHA). Carboxyl-terminated PNIPAAm-COOH (2 g, 10 w/v) was mixed into 20 mL of Solution A for 3 h on a magnetic stirrer to activate the solution (Solution B). The ESM was cut into discs using an AcuPunch®® (5 mm), submerged in 20 mL of solution B in a petri dish, and continuously stirred (50 rpm) at 4 ◦C for 24 h. ESM-PNIPAAm discs were removed from solution B and inserted into a 24-well plate with 200 µL of distilled water to avoid dehydration. To load the AgNP, the water was removed, and 100 µL of silver dispersion was added dropwise in each well using a micropipette, covered with foil, then shaken at 4 ◦C for 24 h. Modiﬁed samples were then dried for 3 h in a fume hood and stored at 4 ◦C. The grafting of the ESM-PNIPAAm was conﬁrmed using Fourier transform infrared red (FT-IR) spectroscopy as well as scanning electronic microscopy (SEM) analysis. Enzymatic Crosslinking. ESM samples were biologically crosslinked using the enzyme transglutaminase (TG). Stock transglutaminase solution was prepared by dissolving 30 mg of TG powder “Meat Glue” (Special Ingredients Ltd., Chesterﬁeld, UK) to 30 mL of PBS. Raw ESM samples were submerged in the solution and incubated (37 ◦C) for 24 h. The crosslinked membranes (TG-ESM) were then rinsed before being stored at 4 ◦C in PBS. Materials and Methods Large free‐range brown 2. Materials and Methods a ge, ee a ge, o i e egg ( i i a ai , Ga us ga us o e i u ) were purchased from a local supermarket (Sainsbury’s, London, UK). Acetic acid (AA) glacial, 99.7+%) was purchased from Alfa Aesar (36289, Heysham, UK). Sodium hydrox‐ de (NaOH, 98%) (71687), N‐(3‐Dimethylaminopropyl)‐N’‐ethylcarbodiimide hydrochlo‐ de (EDC, crystalline) (E6383), N‐Hydroxysuccinimide (NHS, 98%) (130672), Lugol’s so‐ ution (Potassium iodide solution) (32922), Poly(N‐isopropylacrylAmide) (724459), Large, free-range, brown chicken eggs (British Blacktail, Gallus gallus domesticus) were purchased from a local supermarket (Sainsbury’s, London, UK). Acetic acid (AA) (glacial, 99.7%) was purchased from Alfa Aesar (36289, Heysham, UK). Sodium hydroxide (NaOH, 98%) (71687), N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC, crys- talline) (E6383), N-Hydroxysuccinimide (NHS, 98%) (130672), Lugol’s solution (Potassium iodide solution) (32922), Poly(N-isopropylacrylAmide) (724459), penicillin/streptomycin (P/S) (P4333), MES hydrate (99.5%) (M2933) and silver dispersion solution (0.02 mg/mL in aqueous buffer) (730785) were purchased from Sigma-Aldrich (Poole, UK). Dulbecco’s mod- iﬁed Eagle medium (DMEM) (21331046), foetal bovine serum (FBS) (A4766801), Alamar- BlueTM (DAL1025), paraformaldehyde (PFA, 4%), Live/DeadTM cytotoxicity kit (L3224), phosphate-buffered saline (PBS, pH 7.2) (20012019) and Human Dermal Fibroblast (HDFa) (C0125C) cells were purchased from GIBCO (Thermo Fisher, Paisley, UK). CellTiter 96®® AQueous One Solution Cell Proliferation assay (i.e., MTS assay) (G3582) and the CytoTox 96®® Non-radioactive Cytotoxicity Assay (i.e., LDH assay) (G1780) were purchased from Promega (Southampton, UK). 4 of 13 Pharmaceutics 2022, 14, 2162 [where S = change in stress, e = change in strain]. 2.3. Material Characterisation Morphological Analysis. A scanning electron microscope (SEM) (Zeiss EVO HD, Jena, Germany) was used to image and analyse the inner and outer surfaces of the ESM with and without AgNP. Before examination, specimens were cut into 12 mm discs and ﬁxed in PFA (200 µL, 4%) for 48 h. Samples were then coated (95% gold and 5% palladium, Polaron E5000 Sputter Coater, Quorum Technologies, Laughton, UK). 500× and 5000× magniﬁcations were used to visualise the surface morphology of the samples. Biochemical Analysis. Fourier transform infrared spectrophotometer (FT-IR) (Spec- trum One, Perkin Elmer, Llanstrisant, UK) was used to characterise the biochemical compo- sition of the ESM samples, and to conﬁrm the fabrication of the ESM-PNIPAAm. Samples were scanned in the within the 400 to 4000 cm−1 range, at room temperature (19 ◦C), and calibrated by taking the initial background absorbance. Time Base (Spectrum) software was used to process the spectra. Mechanical Analysis. Tensile strength tests were performed on saturated rectangular samples (~10 × 17 mm, pre-soaked overnight in PBS) at room temperature (19 ◦C) using tensile ﬁlm clamp as part of the Dynamic Mechanical Analysis (DMA850, TA Instruments, New Castle, NSW, USA) setup. The TRIOS software was used to determine both the ulti- mate tensile strength (UTS), and % elongation at break. Young’s Modulus was extrapolated from the linear slope of the stress-strain curves as stated below: Young′sModulus (MPa) = S (MPa) e (%) [where S = change in stress, e = change in strain]. [where S = change in stress, e = change in strain]. 5 of 13 Pharmaceutics 2022, 14, 2162 Thermal Analysis. Differential scanning calorimetry (DCS25, TA Instruments, New Castle, NSW, USA) was used to track variations in the heat capacity of membrane samples. Weighed samples were inserted into Tzero®® Pans and Lids. A control empty pan was used as a reference and runs were performed at least in triplicate. Pans were heated from 0 ◦C to 250 ◦C, at a ramp rate of 20 ◦C/min, under a continuous ﬂowrate of nitrogen gas. TRIOS software was used to analyse and report the data. Contact Angle Measurements. The response of the raw and modiﬁed ESM samples to water was determined by measuring the contact angles (CA) using the sessile drop method/optical contact angle proﬁling. 2.3. Material Characterisation In brief, a droplet of distilled water (~2 µL) was deposited on the sample surface and the CA measured at room temperature (~19 ◦C), using a CAM 200 optical angle meter (KSV Instruments Ltd., Helsinki, Finland). Drug Release Proﬁle of ESM-PNIPAAm (AgNP). The drug release proﬁle and ther- moresponsive properties of the ESM-PNIPAAm (AgNP) was analysed by measuring the absorbance of AgNP within PBS solutions containing the modiﬁed membranes, at 380 nm. In total, 12 mm (disc diameter) samples were immersed in 1.5 mL PBS, at either 4 ◦C or 37 ◦C, to demonstrate the effects of temperatures above and below the LCST. The solutions were removed at hourly intervals, for a total of 6 h, to measure the ‘initial burst’ release proﬁle, and then at 24, 48, and 72 h to determine the ‘absolute’ release proﬁle. Samples were decanted into cuvettes (accurate for minimum 285 nm wavelength and over) and analysed using an UV-Visible spectra (UNICAM UV 500 Spectrophotometer, Spectronic, London, UK). The concentration of NP released was quantiﬁed against a calibration curve of known concentrations. Vision 1 was used to analyse the data. 3.1. Material Characterisation Morphology. Following extraction via acid decellularization, the raw ESM was analysed using SEM; images are shown in Figure 2. The images demonstrate that distinct structural patterns between the inner and outer surface of the membrane exist. The inner surface shows a flat, continuous morphology in which multiple circular bulges are visible upon higher magnification (Figure 2a,b). In contrast, the outer surface displays a dense unorganised fibrillar network containing microporous structures. (Figure 2c,d). The diameter of these fibres was measured as 1.778 ± 0.778 µm. Understanding the structure of the egg, the structural properties of the two surfaces correspond to their position on the eggshell. For example, the fibrils in the outer layer pay a critical role in early development, as nucleation sites on the membrane attract calcium salts, forming the eggshell [22]. The deposition of the AgNP on both the inner and outer layers of the ESM are observed in Figure 2e–h. REVIEW 7 of 15 Figure 2. SEM images of the extracted ESM samples. (a,b) Inner layer of the membrane at magnifi‐ cation of 500× and 5000× respectively. (c,d) Inner layer of the membrane at magnification of 500×, and 5000× respectively. (e,f) inner layer of ESM‐PNIPAAm (AgNP) at magnification of 500×, and 5000× respectively. (g,h) outer layer of ESM‐PNIPAAm (AgNP) at magnification of 500×, and 5000×, respectively. Figure 2. SEM images of the extracted ESM samples. (a,b) Inner layer of the membrane at magnification of 500× and 5000× respectively. (c,d) Inner layer of the membrane at magnification of 500×, and 5000× respectively. (e,f) inner layer of ESM-PNIPAAm (AgNP) at magnification of 500×, and 5000× respectively. (g,h) outer layer of ESM-PNIPAAm (AgNP) at magnification of 500×, and 5000×, respectively. Figure 2. SEM images of the extracted ESM samples. (a,b) Inner layer of the membrane at magnifi‐ cation of 500× and 5000× respectively. (c,d) Inner layer of the membrane at magnification of 500×, and 5000× respectively. (e,f) inner layer of ESM‐PNIPAAm (AgNP) at magnification of 500×, and 5000× respectively. (g,h) outer layer of ESM‐PNIPAAm (AgNP) at magnification of 500×, and 5000×, respectively. Figure 2. SEM images of the extracted ESM samples. (a,b) Inner layer of the membrane at magnification of 500× and 5000× respectively. (c,d) Inner layer of the membrane at magnification of 500×, and 5000× respectively. (e,f) inner layer of ESM-PNIPAAm (AgNP) at magnification of 500×, and 5000× respectively. 2.5. Statistical Analysis Quantitative results are expressed as mean and standard deviation (SD). Experiments were performed in triplicate unless otherwise stated. Signiﬁcance of each dataset was determined using a student’s t-test with the statistical signiﬁcance indicated by (), which corresponds to a p < 0.05, () which corresponds to a p < 0.01 and (*), which corre- sponds to a p < 0.001. For the physical characterisation of membrane, modiﬁed samples were compared to the raw ESM independently. The data were analysed using Microsoft Excel 2019 and presented using Origin 2020. Microscopy images were developed using ImageJ software. 2.4. Biological Characterisation Cytotoxicity analyses. Human dermal ﬁbroblasts (adult) (HDFa) were cultured in T-75 ﬂasks (Corning Life Sciences, UK) with 13 mL of DMEM (Life Technologies Ltd., Paisley, UK) supplemented with 10% FBS, 2 mM L-glutamine (LG), and 100 U/mL penicillin, and 100 µg/mL streptomycin (P/S) under standard humidiﬁed cell culture conditions (37◦C and 5% CO2). Routine cell culture involved a standard trypsinisation protocol (i.e., 1% (v/v) trypsin-EDTA) every 3 days including gentle rinsing with PBS (~10 mL). Cells at passage 8 (P8) were used in subsequent experiments. For Live/Dead staining, L929 mouse ﬁbroblast cells were employed. L929 cells were cultured as described above. p y Sample membranes were cut into 5 mm (diameter) discs before being placed into 96- well plates (Corning CostarTM, Thermo Fisher, Paisley, UK) and UV sterilised (Steristorm 2537a) for 20 min. Thereafter, samples were soaked, in PBS, for 1 min for equilibration. HDFa cells were seeded on the inner side of the membranes (Raw ESM, ESM-PNIPAAm, ESM-PNIPAAm (AgNP), I-ESM and TG-ESM) at a density of 1 × 102 cells/mL in 150 µL of the complete growth media and incubated accordingly (37 ◦C and 5% CO2). Each sample group had 6 replicates and groups of tissue culture plastic (TCP) were included in each plate as a control. The metabolic activity of the cells was evaluated using the CellTiter®® 96 Aqueous One Solution Cell Proliferation assay (Promega, Southampton, UK) according to the manu- facturer’s protocol. In summary, following 1, 3, 5, 7, 10, and 14-days incubation, 50 µL of the culture media was removed from each well before being transferred into a new 96-well plate and retained for the LDH assay. For the proliferation assay, 20 µL of CellTiter One reagent was added to each well and incubated at 37 ◦C for 90 min whilst wrapped in alu- minium foil. Following incubation, the supernatant solution was transferred to a new plate and read at 490 nm using a Tecan Inﬁnite M200 microplate reader (Tecan, Switzerland). Lactate dehydrogenase (LDH) release from the cells was quantiﬁed using the CytoTox 96®® Non-radioactive Cytotoxicity Assay kit (Promega, Southampton, UK). In total, 50 µL of Reagent A was added to 50 µL of media suspension in each well (transferred to new plate as previously described), which was then incubated and covered in aluminium foil at ~19 ◦C for 30 min. Thereafter, 25 µL of stop solution was added to each well. 2.4. Biological Characterisation The Pharmaceutics 2022, 14, 2162 6 of 13 absorbance was the immediately read using a Tecan Inﬁnite M200 microplate reader (Tecan, Männedorf, Switzerland). absorbance was the immediately read using a Tecan Inﬁnite M200 microplate reader (Tecan, Männedorf, Switzerland). To support the quantitative assays, Live/DeadTM staining was used to demonstrate the biocompatibility of the raw and modiﬁed membrane samples (ESM-PNIPAAm and ESM- PNIPAAm (AgNP)). A total of 100 µL of L929 (P11) cells, at a density of 1 × 104 cells/mL, were seeded onto membrane samples (12 mm discs) in 24-well plates. At the relevant time point, the media was discarded, and the samples were rinsed with PBS. The stain was prepared by adding 20 µL of EthD-1 (2 mM) stock solution to 10 mL PBS, combined with 5 mL Calcein AM (4 mM) stock solution. After 3 and 5-days incubation, 100 µL of the stain was added to each sample and incubated (~19 ◦C) for 20 min. The viability of the cells was observed using ﬂuorescence microscopy (LEICA Instruments, Milton Keynes, UK) on Image Capture Pro software. 3.1. Material Characterisation Unlike the raw ESM, spectra of the modiﬁed membranes show a shift of the AAmide II peak (1535 cm−1) at a higher intensity and additional smaller peaks within the wavelength region of 1300–1450 cm−1, indicating CH2 and CH3 groups and C-H bending. Figure 3. Fourier transform infrared spectrophotometer (FT‐IR) spectra. Representative FT‐IR sum‐ marising the chemical bonding structure in the raw ESM, as well as alterations results from modifi‐ cation protocols, i.e., raw ESM and modified ESM (ESM‐PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ ESM). Vertical lines (a, b, c, d, e and f) identify the distinctive peaks associated with the bands: Amide A (3275 cm−1), AAmide B (2970 cm−1), AAmide I (1630 cm−1), AAmide II (1535 cm−1), CH2, CH3 groups, and Amide III (1250 cm−1) respectively. Unlike the raw ESM, spectra of the modified membranes show a shift of the AAmide II peak (1535 cm−1) at a higher intensity and additional smaller peaks within the wavelength region of 1300–1450 cm−1, indicating CH2 and CH3 groups and C‐H bending. Figure 3. Fourier transform infrared spectrophotometer (FT-IR) spectra. Representative FT-IR summarising the chemical bonding structure in the raw ESM, as well as alterations results from modiﬁcation protocols, i.e., raw ESM and modiﬁed ESM (ESM-PNIPAAm, ESM-PNIPAAm (AgNP), TG-ESM). Vertical lines (a, b, c, d, e and f) identify the distinctive peaks associated with the bands: Amide A (3275 cm−1), AAmide B (2970 cm−1), AAmide I (1630 cm−1), AAmide II (1535 cm−1), CH2, CH3 groups, and Amide III (1250 cm−1) respectively. Unlike the raw ESM, spectra of the modiﬁed membranes show a shift of the AAmide II peak (1535 cm−1) at a higher intensity and additional smaller peaks within the wavelength region of 1300–1450 cm−1, indicating CH2 and CH3 groups and C-H bending. 3.2. Thermal Properties Thermal behaviour of the raw and modified membranes were assessed using a DSC‐ based protocol. The general thermograms and thermal profiles are summarised in Figure 4 and Table 1, respectively. All membrane samples show a distinct endothermic decom‐ position peak at ~115 °C. Similar endothermic peaks were reported by Torres et al. [24], associating this to the decomposition of collagen within the membrane, resulting from the rupture of hydrogen bonds and the rearrangement of the helical structure into random chain configuration. 3.1. Material Characterisation (g,h) outer layer of ESM-PNIPAAm (AgNP) at magnification of 500×, and 5000×, respectively. 7 of 13 Pharmaceutics 2022, 14, 2162 Biochemical Properties. Chemical bonding in the raw ESM samples were observed using FT-IR within the 400–4000 cm−1 range. Referring to Figure 3, the spectra summarise the functional groups within the raw samples within which Amide A and B bands can be identiﬁed. The Amide A band (Figure 3a) characterised as the broad peak at 3275 cm−1 corresponds to hydrogen bonds and C-H stretching. The smaller peak observable at 2910 cm−1 is characteristic of typical Amide B bands (Figure 3b), corresponding to the C-H bonds found in the =C-H functional group. Similar spectra and biochemical characteristics have previously been reported in the literature [12,23–26]. Amide band I (Figure 3c), Amide band II (Figure 3d) and Amide band III (Figure 3g), visualised as peaks 1630 cm−1, 1535 cm−1, 1250 cm−1 respectively, are characteristic of the membrane’s high protein content, speciﬁcally collagen. The Amide I and II bands represent major bands of the protein infrared spectrum and can be associated with C=O stretching, N-H bending and C-N stretching vibrations [23]. REVIEW 8 of 15 Figure 3. Fourier transform infrared spectrophotometer (FT‐IR) spectra. Representative FT‐IR sum‐ marising the chemical bonding structure in the raw ESM, as well as alterations results from modifi‐ cation protocols, i.e., raw ESM and modified ESM (ESM‐PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ ESM). Vertical lines (a, b, c, d, e and f) identify the distinctive peaks associated with the bands: Amide A (3275 cm−1), AAmide B (2970 cm−1), AAmide I (1630 cm−1), AAmide II (1535 cm−1), CH2, CH3 groups, and Amide III (1250 cm−1) respectively. Unlike the raw ESM, spectra of the modified membranes show a shift of the AAmide II peak (1535 cm−1) at a higher intensity and additional smaller peaks within the wavelength region of 1300–1450 cm−1, indicating CH2 and CH3 groups and C‐H bending. Figure 3. Fourier transform infrared spectrophotometer (FT-IR) spectra. Representative FT-IR summarising the chemical bonding structure in the raw ESM, as well as alterations results from modiﬁcation protocols, i.e., raw ESM and modiﬁed ESM (ESM-PNIPAAm, ESM-PNIPAAm (AgNP), TG-ESM). Vertical lines (a, b, c, d, e and f) identify the distinctive peaks associated with the bands: Amide A (3275 cm−1), AAmide B (2970 cm−1), AAmide I (1630 cm−1), AAmide II (1535 cm−1), CH2, CH3 groups, and Amide III (1250 cm−1) respectively. 3.1. Material Characterisation Table 1 shows that there is no significant difference between the % mass loss, onset temperature, peak temperature, or enthalpy for the raw and modified samples, indicating that such modifications do not alter the thermal properties of the Additionally, FT-IR was implemented to characterise the modiﬁed membranes (ESM- PNIPAAm, ESM-PNIPAAm (AgNP), ESM-I and TG-ESM) (Figure 3). The technique con- ﬁrmed the successful grating of PNIPAAm to the ESM, evident by a small shift and increased Amide II peak (1535 cm−1) intensity. This indicates conjugation of C=O, C-N and increased N-H stretching within the modiﬁed membrane [27]. This is further supported by the presence of CH3 (Figure 3e) and CH2 (Figure 3f) peaks visible between 1300–1450 cm−1 on the spectra of the grafted membrane not visible on that of the raw ESM. A similar spectrum can be observed for the TG-ESM and I-ESM in which CH3 and CH2 peaks can be observed. Interestingly, the FT-IR proﬁle of the modiﬁed membrane (ESM-PNIPAAm Pharmaceutics 2022, 14, 2162 8 of 13 (AgNP)) has a proﬁle matching that of ESM-PNIPAAm, suggesting that “loading” AgNP to the gel does not further alter the chemical composition. (AgNP)) has a proﬁle matching that of ESM-PNIPAAm, suggesting that “loading” AgNP to the gel does not further alter the chemical composition. 3.2. Thermal Properties Thermal behaviour of the raw and modiﬁed membranes were assessed using a DSC- based protocol. The general thermograms and thermal proﬁles are summarised in Figure 4 and Table 1, respectively. All membrane samples show a distinct endothermic decom- position peak at ~115 ◦C. Similar endothermic peaks were reported by Torres et al. [24], associating this to the decomposition of collagen within the membrane, resulting from the rupture of hydrogen bonds and the rearrangement of the helical structure into random chain conﬁguration. Table 1 shows that there is no signiﬁcant difference between the % mass loss, onset temperature, peak temperature, or enthalpy for the raw and modiﬁed samples, indicating that such modiﬁcations do not alter the thermal properties of the membranes. EVIEW 9 of 15 Figure 4. DSC thermogram of the raw and modified ESM. Distinct endothermic decomposition peak at approximately 160 °C, associated with the decomposition of collagen (Ramp 20 °C/min, 0 to 250 °C). Table 1. Thermal properties of the raw and modified ESM. All values are expressed as mean ± SD Figure 4. DSC thermogram of the raw and modified ESM. Distinct endothermic decomposition peak at approximately 160 ◦C, associated with the decomposition of collagen (Ramp 20 ◦C/min, 0 to 250 ◦C). Table 1. Thermal properties of the raw and modiﬁed ESM. All values are expressed as mean ± SD f 5 ( < 0 05) Figure 4. DSC thermogram of the raw and modified ESM. Distinct endothermic decomposition peak at approximately 160 °C, associated with the decomposition of collagen (Ramp 20 °C/min, 0 to 250 C) Figure 4. DSC thermogram of the raw and modified ESM. Distinct endothermic decomposition peak at approximately 160 ◦C, associated with the decomposition of collagen (Ramp 20 ◦C/min, 0 to 250 ◦C). ) Table 1. Thermal properties of the raw and modified ESM. All values are expressed as mean ± SD for n = 5 (p < 0.05). Table 1. Thermal properties of the raw and modiﬁed ESM. All values are expressed as mean ± SD for n = 5 (p < 0.05). for n 5 (p 0.05). 3.2. Thermal Properties Sample Type % Mass Loss Onset Temp (°C) Peak Temp (°C) Enthalpy (J/g) Raw ESM 73.607 ± 7.624 100.67 ± 0.707 113.817 ± 2.586 1838.167 ± 145.858 ESM‐PNIPAAm 62.724 ± 8.874 99.544 ± 1.221 114.122 ± 2.102 1750.52 ± 176.214 ESM‐PNIPAAm (AgNP) 67.9 ± 1.718 101.41 ± 2 111.493 ± 1.44 2041.933 ± 72.403 TG‐ESM 73.64 ± 6.399 100.592 ± 1.253 112.2 ± 2.032 1745.067 ± 342.617 3 3 Mechanical Properties Sample Type % Mass Loss Onset Temp (◦C) Peak Temp (◦C) Enthalpy (J/g) Raw ESM 73.607 ± 7.624 100.67 ± 0.707 113.817 ± 2.586 1838.167 ± 145.858 ESM-PNIPAAm 62.724 ± 8.874 99.544 ± 1.221 114.122 ± 2.102 1750.52 ± 176.214 ESM-PNIPAAm (AgNP) 67.9 ± 1.718 101.41 ± 2 111.493 ± 1.44 2041.933 ± 72.403 TG-ESM 73.64 ± 6.399 100.592 ± 1.253 112.2 ± 2.032 1745.067 ± 342.617 The tensile properties of b i d f DMA fili 3.3. Mechanical Properties obtained from DMA profiling. Representative stress‐strain curves of the raw and modi‐ fied membranes are shown in Figure 5a,b. The membranes show a predominantly elastic behaviour (Figure 5b), in line with previous publications reporting the mechanical behav‐ iour of collagen‐based materials [28,29], the ESM shows linear and non‐linear regions. Ac‐ cordingly, within the non‐linear (toe) region, a minimal amount of stress is required ini‐ tially to deform the membrane as collagen fibres require alignment in the direction of the stress [30]. Within the second region, the membrane’s stiffness increases with elongation until the UTS is reached and the membrane ruptures (failure) The tensile properties of the membranes were analysed from the stress-strain curves obtained from DMA proﬁling. Representative stress-strain curves of the raw and modiﬁed membranes are shown in Figure 5a,b. The membranes show a predominantly elastic be- haviour (Figure 5b), in line with previous publications reporting the mechanical behaviour of collagen-based materials [28,29], the ESM shows linear and non-linear regions. Accord- ingly, within the non-linear (toe) region, a minimal amount of stress is required initially to Pharmaceutics 2022, 14, 2162 O 9 of 13 5 deform the membrane as collagen ﬁbres require alignment in the direction of the stress [30]. Within the second region, the membrane’s stiffness increases with elongation until the UTS is reached and the membrane ruptures (failure). membrane is a seemingly strong material and further modification, particularly crosslink‐ ing, improved these properties. Figure 5. Tensile profile of the raw and modified ESM. (a) The general stress‐strain profile of the ESM. (b) representative stress‐strain curves of the raw and modified ESM (ESM‐PNIPAAm (AgNP) and TG‐ESM). Figure 5. Tensile proﬁle of the raw and modiﬁed ESM. (a) The general stress-strain proﬁle of the ESM (b) representative stress-strain curves of the raw and modiﬁed ESM (ESM-PNIPAAm (AgNP) and TG-ESM). Figure 5. Tensile profile of the raw and modified ESM. (a) The general stress‐strain profile of the ESM. (b) representative stress‐strain curves of the raw and modified ESM (ESM‐PNIPAAm (AgNP) and TG‐ESM). Figure 5. Tensile proﬁle of the raw and modiﬁed ESM. (a) The general stress-strain proﬁle of the ESM. (b) representative stress-strain curves of the raw and modiﬁed ESM (ESM-PNIPAAm (AgNP) and TG-ESM). The mechanical measurements, including the Young’s modulus, UTS and % elongation at break, are summarised in Table 2. The tensile properties of b i d f DMA fili 3.3. Mechanical Properties It can be noted from the data that signiﬁcant differences between the % elongation at break of the raw ESM and ESM-PNIPAAm (p < 0.05) and raw ESM and TG-ESM (p < 0.005) exists. Similarly, a signiﬁcant difference between the Young’s modulus of the raw ESM, TG-ESM (p < 0.005) and the raw ESM and ESM-PNIPAAm (p < 0.05) can be seen. A signiﬁcant difference between the UTS of the raw ESM and TG-ESM (p < 0.005) was also observed. No signiﬁcant difference was observed between the mechanical properties of the raw ESM and ESM-I. As such, it can be proposed such modiﬁcations increase the tensile strength of the membrane, due to alterations in the ﬁbrillar Pharmaceutics 2022, 14, 2162 10 of 13 10 of 13 structure of the material. The tensile properties indicate the membrane is a seemingly strong material and further modiﬁcation, particularly crosslinking, improved these properties. R REVIEW 11 of 15 Table 2. Mechanical proﬁle of the raw and modiﬁed membrane. All values expressed as mean ± SD, n = 5 ( p < 0.05, ** p < 0.005). Table 2. Mechanical profile of the raw and modified membrane. All values expressed as mean ± SD, Table 2. Mechanical proﬁle of the raw and modiﬁed membrane. All values expressed as mean ± SD, n = 5 (* p < 0.05, ** p < 0.005). Sample Type % Elongation UTS (MPa) Young’s Modulus (MPa) Raw ESM 15.248 ± 2.819 0.419 ± 0.035 2.242 ± 0.309 ESM-PNIPAAm 27.045 ± 8.352 () 0.348 ± 0.124 2.854 ± 0.192 () TG-ESM 37.678 ± 1.739 () 0.612 ± 0.051 () 3.892 ± 0.190 (*) 3.4. Drug Release Proﬁle—ESM-PNIPAAm Table 2. Mechanical profile of the raw and modified membrane. All values expressed as mean ± SD, n = 5 ( p < 0.05, ** p < 0.005). % Elongation UTS (MPa) Young’s Modulus (MPa) 15.248 ± 2.819 0.419 ± 0.035 2.242 ± 0.309 27.045 ± 8.352 () 0.348 ± 0.124 2.854 ± 0.192 () 37.678 ± 1.739 () 0.612 ± 0.051 () 3.892 ± 0.190 (*) 3 4 D R l P fil ESM PNIPAA 3.4. Drug Release Proﬁle—ESM-PNIPAAm 3 4 Drug Release Profile—ESM‐PNIPAAm Ultraviolet-visible (UV-Vis) spectroscopy was used to monitor the release of AgNP from the ESM-PNIPAAm (AgNP)-modiﬁed membrane. Samples were assessed at 380 nm, which corresponded to the absorbance wavelength of the commercially available AgNP (i.e., 380–405 nm). The tensile properties of b i d f DMA fili 3.3. Mechanical Properties To establish the thermoresponsive properties of PNIPAAm, the release of AgNP was monitored at 4 ◦C and 37 ◦C (Figure 6). A signiﬁcant increase of AgNP was observed within the ﬁrst 8 h from the ESM-PNIPAAm (AgNP) at 4 ◦C. The spectra indicate that at temperatures below the LCST (4 ◦C), the NP are entirely released within 1–8 h, after which the concentration plateaus. 3.4. Drug Release Profile ESM PNIPAAm Ultraviolet‐visible (UV‐Vis) spectroscopy was used to monitor the release of AgNP from the ESM‐PNIPAAm (AgNP)‐modified membrane. Samples were assessed at 380 nm, which corresponded to the absorbance wavelength of the commercially available AgNP (i.e., 380–405 nm). To establish the thermoresponsive properties of PNIPAAm, the release of AgNP was monitored at 4 °C and 37 °C (Figure 6). A significant increase of AgNP was observed within the first 8 h from the ESM‐PNIPAAm (AgNP) at 4 °C. The spectra indi‐ cate that at temperatures below the LCST (4 °C), the NP are entirely released within 1–8 h, after which the concentration plateaus. Figure 6. UV‐Vis spectra of drug release from ESM‐PNIPAAm (AgNP) at temperatures above (37° C) and below (4° C) the LCST. Samples were assessed at 380 nm, which corresponded to the absorb‐ ance wavelength of the commercially available AgNP (i.e., 380–405 nm). ESM‐PNIPAAm (AgNP) samples at 4 °C show an initial burst release profile within the first 6 h, in which most of the NPs are released. Samples at 37 °C show little to no release of NP, relating to the thermoresponsive prop‐ erties of PNIPAAm. Figure 6. UV-Vis spectra of drug release from ESM-PNIPAAm (AgNP) at temperatures above (37◦C) and below (4◦C) the LCST. Samples were assessed at 380 nm, which corresponded to the absorbance wavelength of the commercially available AgNP (i.e., 380–405 nm). ESM-PNIPAAm (AgNP) samples at 4 ◦C show an initial burst release proﬁle within the ﬁrst 6 h, in which most of the NPs are released. Samples at 37 ◦C show little to no release of NP, relating to the thermoresponsive properties of PNIPAAm. Figure 6. UV‐Vis spectra of drug release from ESM‐PNIPAAm (AgNP) at temperatures above (37° C) and below (4° C) the LCST. Samples were assessed at 380 nm, which corresponded to the absorb‐ ance wavelength of the commercially available AgNP (i.e., 380–405 nm). The tensile properties of b i d f DMA fili 3.3. Mechanical Properties ESM‐PNIPAAm (AgNP) samples at 4 °C show an initial burst release profile within the first 6 h, in which most of the NPs are released. Samples at 37 °C show little to no release of NP, relating to the thermoresponsive prop‐ erties of PNIPAAm. Figure 6. UV-Vis spectra of drug release from ESM-PNIPAAm (AgNP) at temperatures above (37◦C) and below (4◦C) the LCST. Samples were assessed at 380 nm, which corresponded to the absorbance wavelength of the commercially available AgNP (i.e., 380–405 nm). ESM-PNIPAAm (AgNP) samples at 4 ◦C show an initial burst release proﬁle within the ﬁrst 6 h, in which most of the NPs are released. Samples at 37 ◦C show little to no release of NP, relating to the thermoresponsive properties of PNIPAAm. 3.5. In Vitro Cytotoxicity Sample is significantly different to all other samples within group with p < 0.05, ** p < 0.005, and # p < 0.001). Figure 7. Human dermal ﬁbroblasts (adult) (HDFa) cell viability, proliferation and adhesion assay. (a) Cell metabolic activity of HDFa cells cultured on TCP, raw and modiﬁed (ESM-PNIPAAm, ESM- PNIPAAm (AgNP), TG-ESM) membranes over fourteen days. Data are represented as mean ± SD (n = 4). (b) LDH release of HDFa cells cultured on TCP, raw and modiﬁed membranes (ESM- PNIPAAm, ESM-PNIPAAm (AgNP), TG-ESM) over fourteen days. Data are represented as mean ± SD (n = 4). (c) Cell viability and cytotoxicity was demonstrated by Live/DeadTM staining of cells seeded on the raw ESM, PNIPAAm adn ESM-PNIPAAm (AgNP) on days 3 and 5 under the ﬂuorescence microscope. The stain causes live cells to ﬂuoresce green and the dead cells to ﬂuoresce red. Fluorescence imaging of cells seeded on ESM-PNIPAAm and ESM-PNIPAAm (AgNP) show a sustained cell viability after 3 and 5 days of incubation with the samples. In contrast, cells seeded on the raw ESM showed a decreased cell viability between 3 and 5 days. Sample is signiﬁcantly different to all other samples within group with * p < 0.05, ** p < 0.005, and # p < 0.001). Figure 7. Human dermal fibroblasts (adult) (HDFa) cell viability, proliferation and adhesion assay. (a) Cell metabolic activity of HDFa cells cultured on TCP, raw and modified (ESM‐PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ESM) membranes over fourteen days. Data are represented as mean ± SD (n = 4). (b) LDH release of HDFa cells cultured on TCP, raw and modified membranes (ESM‐ PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ESM) over fourteen days. Data are represented as mean ± SD (n = 4). (c) Cell viability and cytotoxicity was demonstrated by Live/DeadTM staining of cells seeded on the raw ESM, PNIPAAm adn ESM‐PNIPAAm (AgNP) on days 3 and 5 under the fluo‐ rescence microscope. The stain causes live cells to fluoresce green and the dead cells to fluoresce red. Fluorescence imaging of cells seeded on ESM‐PNIPAAm and ESM‐PNIPAAm (AgNP) show a sustained cell viability after 3 and 5 days of incubation with the samples. In contrast, cells seeded on the raw ESM showed a decreased cell viability between 3 and 5 days. Sample is significantly different to all other samples within group with * p < 0.05, ** p < 0.005, and # p < 0.001). 3.5. In Vitro Cytotoxicity Cell mitochondrial activity and cell death of the HDFa cells were assessed using the MTS and LDH assays, respectively, following culture on the raw and modiﬁed membrane samples for up to fourteen days (Figure 7a,b). 11 of 13 5 Pharmaceutics 2022, 14, 2162 Pharmaceutics 2022, 14, x FOR P Figure 7. Human dermal fibroblasts (adult) (HDFa) cell viability, proliferation and adhesion assay. (a) Cell metabolic activity of HDFa cells cultured on TCP, raw and modified (ESM‐PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ESM) membranes over fourteen days. Data are represented as mean ± SD (n = 4). (b) LDH release of HDFa cells cultured on TCP, raw and modified membranes (ESM‐ PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ESM) over fourteen days. Data are represented as mean ± SD (n = 4) (c) Cell viability and cytotoxicity was demonstrated by Live/DeadTM staining of cells Figure 7. Human dermal ﬁbroblasts (adult) (HDFa) cell viability, proliferation and adhesion assay. (a) Cell metabolic activity of HDFa cells cultured on TCP, raw and modiﬁed (ESM-PNIPAAm, ESM- PNIPAAm (AgNP), TG-ESM) membranes over fourteen days. Data are represented as mean ± SD (n = 4). (b) LDH release of HDFa cells cultured on TCP, raw and modiﬁed membranes (ESM- PNIPAAm, ESM-PNIPAAm (AgNP), TG-ESM) over fourteen days. Data are represented as mean ± SD (n = 4) (c) Cell viability and cytotoxicity was demonstrated by Live/DeadTM staining of Figure 7. Human dermal fibroblasts (adult) (HDFa) cell viability, proliferation and adhesion assay. (a) Cell metabolic activity of HDFa cells cultured on TCP, raw and modified (ESM‐PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ESM) membranes over fourteen days. Data are represented as mean ± SD (n = 4). (b) LDH release of HDFa cells cultured on TCP, raw and modified membranes (ESM‐ PNIPAAm, ESM‐PNIPAAm (AgNP), TG‐ESM) over fourteen days. Data are represented as mean ± SD (n = 4). (c) Cell viability and cytotoxicity was demonstrated by Live/DeadTM staining of cells seeded on the raw ESM, PNIPAAm adn ESM‐PNIPAAm (AgNP) on days 3 and 5 under the fluo‐ rescence microscope. The stain causes live cells to fluoresce green and the dead cells to fluoresce red. Fluorescence imaging of cells seeded on ESM‐PNIPAAm and ESM‐PNIPAAm (AgNP) show a sustained cell viability after 3 and 5 days of incubation with the samples. In contrast, cells seeded on the raw ESM showed a decreased cell viability between 3 and 5 days. 3.5. In Vitro Cytotoxicity Figure 7. Human dermal ﬁbroblasts (adult) (HDFa) cell viability, proliferation and adhesion assay. (a) Cell metabolic activity of HDFa cells cultured on TCP, raw and modiﬁed (ESM-PNIPAAm, ESM- PNIPAAm (AgNP), TG-ESM) membranes over fourteen days. Data are represented as mean ± SD (n = 4). (b) LDH release of HDFa cells cultured on TCP, raw and modiﬁed membranes (ESM- PNIPAAm, ESM-PNIPAAm (AgNP), TG-ESM) over fourteen days. Data are represented as mean ± SD (n = 4). (c) Cell viability and cytotoxicity was demonstrated by Live/DeadTM staining of cells seeded on the raw ESM, PNIPAAm adn ESM-PNIPAAm (AgNP) on days 3 and 5 under the ﬂuorescence microscope. The stain causes live cells to ﬂuoresce green and the dead cells to ﬂuoresce red. Fluorescence imaging of cells seeded on ESM-PNIPAAm and ESM-PNIPAAm (AgNP) show a sustained cell viability after 3 and 5 days of incubation with the samples. In contrast, cells seeded on the raw ESM showed a decreased cell viability between 3 and 5 days. Sample is signiﬁcantly different to all other samples within group with * p < 0.05, ** p < 0.005, and # p < 0.001). L929 cells were seeded on samples of the raw ESM, ESM-PNIPAAm and ESM- PNIPAAm (AgNP) and stained with a Live/DeadTM cytotoxicity assay kit to determine Pharmaceutics 2022, 14, 2162 12 of 13 12 of 13 the in vitro viability and cytotoxicity of the dressing. The effect of the sample materials on cell viability was qualitatively validated using ﬂuorescence microscopy (Figure 7c) in which the control groups (TCP) exhibited very high cell viability and contained minimal dead cells after 3 and 5 days, while after 3 days of incubation with each of the test samples, most cells remained viable, although the cell number was not as high as the control. Interestingly, after 5 days, the ESM-PNIPAAm and ESM-PNIPAAm (AgNP) samples showed a sustained cell viability. Acknowledgments: The authors would like to thank Nicola Mordan, George Georgiou and Graham Palmer for their technical support. 4. Conclusions An optimized protocol for isolating the ESM in its native state using an acid-based pro- tocol has been outlined in this study. Physico-mechanical characterisation techniques were used to determine the ESM’s deﬁned properties. Overall, the generated data suggests that the ESM can be employed in regenerative applications, particularly to improve regeneration following dermal injuries and trauma. Moreover, a thermoresponsive PNIPAAm polymer was successfully added to the membrane and addition AgNP subsequently loaded. The drug release proﬁle of the fabricated dressing indicated a “burst” release of NP within the ﬁrst eight hours at temperatures below its LCST (4 ◦C). In addition to having regenerative applications, the proposed material could also be used for drug delivery applications in the future when release and retention proﬁles are optimized. p p Biocompatibility of the ESM-PNIPAAm (AgNP) dressing was also validated using cell culture, and a minimal adverse effect was observed along with sustained cell viability. Thus, these results suggest that the ESM, particularly when combined with the PNIPAAm hydrogel, is effective at improving wound healing and regeneration following dermal injury. It is even more relevant to use the ESM because it is a high-quality, readily available waste material. Author Contributions: E.B.: Data curation, Formal analysis, Investigation, Methodology, Writing— original draft; R.A.M.: Data curation, Formal analysis, Investigation, Methodology, Writing—original draft; K.D.P.: Data curation, Formal analysis, Investigation, Methodology, Writing—original draft; N.-E.M.: Investigation, Methodology, Writing—review and editing; N.S.S.: Data curation, Formal anal- ysis, Investigation, Methodology, Writing—review and editing; Z.K.E.: Investigation, Methodology, Writing—review and editing; L.V.B.S.: Data curation, Formal analysis, Investigation, Methodol- ogy, Writing—review and editing; K.S.: Data curation, Formal analysis, Investigation, Methodol- ogy, Writing—original draft; H.-W.K.: Funding acquisition, Investigation, Methodology, Resources, Supervision, Writing—review and editing; L.T.B.N.: Conceptualization: Investigation, Method- ology, Resources, Supervision, Writing—review and editing; D.Y.S.C.: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Super- vision, Writing—review and editing. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by grants from the National Research Foundation of Korea [2018K1A4A3A01064257, 2018R1A2B3003446 and 2019R1A6A1A11034536], the London Interdisci- plinary Bioscience PhD Consortium (LIDo) and the Biotechnology and Biological Science Research Council (BBSRC) [BB/M009513/1], the Rosetrees Trust [Seedcorn Award], the Stoneygate Trust and the Turkish Ministry of National Education for their YLSY programme. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. 4. Conclusions Data Availability Statement: Not applicable. Data Availability Statement: Not applicable. Data Availability Statement: Not applicable. Acknowledgments: The authors would like to thank Nicola Mordan, George Georgiou and Graham Palmer for their technical support. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 13 of 13 13 of 13 Pharmaceutics 2022, 14, 2162 References 1. Eming, S.A.; Martin, P.; Tomic-Canic, M. Wound repair and regeneration: Mechanisms, signaling, and translation. Sci. Transl. Med. 2014, 6, 265sr6. [CrossRef] [PubMed] [ ] [ ] 2. Jeffcoate, W.J.; Harding, K.G. Diabetic foot ulcers. Lancet 2003, 361, 1545–1551. [CrossRef] 3. Lipsky, B.A.; Berendt, A.R.; Cornia, P.B.; Pile, J.C.; Peters, E.J.G.; Armstrong, D.G.; Deery, H.G.; Embil, J.M.; Joseph, W.S.; Karchmer, A.W.; et al. 2012 infectious diseases society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. Clin. Infect. Dis. 2012, 54, e132–e1773. [CrossRef] [PubMed] f 4. Guest, J.F.; Ayoub, N.; McIlwraith, T.; Uchegbu, I.; Gerrish, A.; Weidlich, D.; Vowden, K.; Vowden, P. Health economic burden that wounds impose on the National Health Service in the UK. BMJ Open 2015, 5, e009283. [CrossRef] [PubMed] p p 5. Guest, J.F.; Ayoub, N.; McIlwraith, T.; Uchegbu, I.; Gerrish, A.; Weidlich, D.; Vowden, K.; Vowden, P. Health economic burden that different wound types impose on the UK’s National Health Service. Int. Wound J. 2017, 14, 322–330. [CrossRef] 6. Boateng, J.S.; Matthews, K.H.; Stevens, H.N.E.; Eccleston, G.M. Wound Healing Dressings and Drug Delivery Systems: A Review. J. Pharm. Sci. 2008, 97, 2892–2923. [CrossRef] 7. Ramshaw, J.A.M.; Werkmeister, J.A.; Glattauer, V. Collagen-based Biomaterials. Biotechnol. Genet. Eng. Rev. 1996, 13, 335–382. [CrossRef] 8. Dhivya, S.; Padma, V.V.; Santhini, E. Wound dressings—A review. BioMedicine 2015, 5, 22. [CrossRef] y g 9. Wang, F.; Wang, M.; She, Z.; Fan, K.; Xu, C.; Chu, B.; Chen, C.; Shi, S.; Tan, R. Collagen/chitosan based two-compartment and bi-functional dermal scaffolds for skin regeneration. Mater. Sci. Eng. C 2015, 52, 155–162. [CrossRef] 10. Williams, C. Algosteril calcium alginate dressing for moderate/high exudate. Br. J. Nurs. 1999, 8, 313–317. [CrossRef] 11. Tsai, W.T.; Yang, J.M.; Lai, C.W.; Cheng, Y.H.; Lin, C.C.; Yeh, C.W. Characterization and adsorption properties of eggshells and eggshell membrane. Bioresour. Technol. 2006, 97, 488–493. [CrossRef] g g g g 11. Tsai, W.T.; Yang, J.M.; Lai, C.W.; Cheng, Y.H.; Lin, C.C.; Yeh, C.W. Characterization and adsorption properties of eggshells and eggshell membrane. Bioresour. Technol. 2006, 97, 488–493. [CrossRef] gg , , [ ] 12. Mensah, R.A.; Jo SBin Kim, H.; Park, S.-M.; Patel, K.D.; Cho, K.J.; Cook, M.T.; Kirton, S.B.; Hutter, V.; Sidney, E.L.; Alves-Lima, D. g p gy g p 30. Duboeuf, F.; Liebgott, H.; Basarab, A.; Brusseau, E.; Delachartre, P.; Vray, D. Static mechanical assessment of elastic Young’s modulus of tissue mimicking materials used for medical imaging. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology, Lyon, France, 22–26 August 2007. References Th h ll b A t ti l bi t i l f l d h li J Bi t A l 2021 36 912 929 [C R f] gg [ ] 12. Mensah, R.A.; Jo SBin Kim, H.; Park, S.-M.; Patel, K.D.; Cho, K.J.; Cook, M.T.; Kirton, S.B.; Hutter, V.; Sidney, E.L.; Alves-Lima, D. The eggshell membrane: A potential biomaterial for corneal wound healing. J. Biomater. Appl. 2021, 36, 912–929. [CrossRef] [ ] .; Jo SBin Kim, H.; Park, S.-M.; Patel, K.D.; Cho, K.J.; Cook, M.T.; Kirton, S.B.; Hutter, V.; Sidney, E.L.; Alves-L 13. Yang, J.Y.; Chuang, S.-S.; Yang, W.-G.; Tsay, P.-K. Egg membrane as a new biological dressing in split-thickness skin graft donor sites: A preliminary clinical evaluation. Chang Gung Med. J. 2003, 26, 153–159. p y g g 14. Jun, H.J.; Oh, K.-H.; Yoo, J.; Han, W.-G.; Chang, J.; Jung, H.H.; Choi, J. A new patch material for tympanic membrane perforation by trauma: The membrane of a hen egg shell. Acta Oto-Laryngologica 2013, 134, 250–254. [CrossRef] y gg y g g 15. Guarderas, F.; Leavell, Y.; Sengupta, T.; Zhukova, M.; Megraw, T.L. Assessment of Chicken-Egg Mem Wound Healing. Adv. Ski. Wound Care 2016, 29, 131–134. [CrossRef] 16. Adamia, K.; Sionkowska, A. Current methods of collagen cross-linking: Review. Int. J. Biol. Macromol. 2020, 161, 550–560. [CrossRef] 17. Brand, F.; Dautzenberg, H. Structural Analysis in Interpolyelectrolyte Complex Formation of Sodium Poly(styrenesulfonate) and Diallyldimethylammonium Chloride−Acrylamide Copolymers by Viscometry. Langmuir 1997, 13, 2905–2910. [CrossRef] 18. Sondi, I.; Salopek-Sondi, B. Silver nanoparticles as antimicrobial agent: A case study on E. coli as a model for Gram-negative bacteria. J. Colloid Interface Sci. 2004, 275, 177–182. [CrossRef] 19. Kumashiro, Y.; Yamato, M.; Okano, T. Cell Attachment–Detachment Control on Temperature-Responsive Thin Surfaces for Novel Tissue Engineering. Ann. Biomed. Eng. 2010, 38, 1977–1988. [CrossRef] 20. Prime, K.L.; Whitesides, G.M. Adsorption of proteins onto surfaces containing end-attached oligo(ethylene oxide): A model system using self-assembled monolayers. J. Am. Chem. Soc. 1993, 115, 10714–10721. [CrossRef] 21. Nguyen, L.T.B.; Odeleye, A.O.O.; Chui, C.; Baudequin, T.; Cui, Z.; Ye, H. Development of thermo-resp macrocarriers conjugated with Poly(N-isopropyl acrylamide) for cell culture. Sci. Rep. 2019, 9, 3477. [C j g y p py y p 22. Solomon, S.E. The eggshell: Strength, structure and function. Br. Poult. Sci. 2010, 51 (Suppl. S1), 52–59. [CrossRef] [PubMed] 22. Solomon, S.E. The eggshell: Strength, structure and function. Br. Poult. Sci. 2010, 51 (Suppl. S1), 52–59. ] axer, A.; Fratzl, P. Erratum: Collagen ﬁbril orientation in the human corneal stroma and its implicatio nvestigate Ophthalmology and Visual Science. Investig. Ophthalmol. Vis. Sci. 1997, 38, 121–129. 28. Buehler, M.J.; Wong, S.Y. Entropic elasticity controls nanomechanics of single tropocollagen molecules. Biophys. J. 2007, 93, 37–43. [CrossRef] References [CrossRef] [PubMed] 23. Ji, Y.; Yang, X.; Ji, Z.; Zhu, L.; Ma, N.; Chen, D.; Jia, X.; Tang, J.; Cao, Y. DFT-Calculated IR Spectrum Amide I, II, and III Band gg g pp 23. Ji, Y.; Yang, X.; Ji, Z.; Zhu, L.; Ma, N.; Chen, D.; Jia, X.; Tang, J.; Cao, Y. DFT-Calculated IR Spectrum Amide I, II, and III Band Contributions of N-Methylacetamide Fine Components. ACS Omega 2020, 5, 8572–8578. [CrossRef] [PubMed] 24. Torres, F.G.; Troncoso, O.P.; Piaggio, F.; Hijar, A. Structure-property relationships of a biopolymer network: The eggshell membrane. Acta Biomater. 2010, 6, 3687–3693. [CrossRef] J g J J g J p Contributions of N-Methylacetamide Fine Components. ACS Omega 2020, 5, 8572–8578. [CrossRef] [PubMed] 24. Torres, F.G.; Troncoso, O.P.; Piaggio, F.; Hijar, A. Structure-property relationships of a biopolymer network: The eggshell Contributions of N-Methylacetamide Fine Components. ACS Omega 2020, 5, 8572–8578. [CrossRef] [PubMed] 24 T F G T O P Pi i F Hij A St t t l ti hi f bi l t k Th h 25. Baláž, M. Eggshell membrane biomaterial as a platform for applications in materials science. Acta Biomater. 2014, 10, 3827–3843. [CrossRef] 26. Bellairs, R.; Boyde, A. Scanning electron microscopy of the shell membranes of the hen’s egg. Z. Zellforsch Mikrosk. Anat 1969, 96, 237–249. [CrossRef] 27. Gupta, K.C.; Khandekar, K. Temperature-responsive cellulose by ceric(IV) ion-initiated graft copolymerization of N- isopropylacrylamide. Biomacromolecules 2003, 4, 758–765. [CrossRef] 28. Buehler, M.J.; Wong, S.Y. Entropic elasticity controls nanomechanics of single tropocollagen molecules. Biophys. J. 2007, 93, 37–43. [CrossRef] 29. Daxer, A.; Fratzl, P. Erratum: Collagen ﬁbril orientation in the human corneal stroma and its implication in keratoconus (Investigate Ophthalmology and Visual Science. Investig. Ophthalmol. Vis. Sci. 1997, 38, 121–129. 30. Duboeuf, F.; Liebgott, H.; Basarab, A.; Brusseau, E.; Delachartre, P.; Vray, D. Static mechanical assessment of elastic Young’s modulus of tissue mimicking materials used for medical imaging. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology, Lyon, France, 22–26 August 2007.
https://openalex.org/W2413006966	https://bmccancer.biomedcentral.com/track/pdf/10.1186/s12885-016-2388-9	English	null	Time course decomposition of cell heterogeneity in TFEB signaling states reveals homeostatic mechanisms restricting the magnitude and duration of TFEB responses to mTOR activity modulation	BMC cancer	2,016	cc-by	14,770	* Correspondence: nbrady7@jhu.edu; abrady9@jhu.edu †Equal contributors 2Systems Biology of Cell Death Mechanisms, German Cancer Research Center (DKFZ) and BioQuant, University of Heidelberg, Heidelberg, Germany 1Lysosomal Systems Biology, German Cancer Research Center (DKFZ) and BioQuant, University of Heidelberg, Heidelberg, Germany Full list of author information is available at the end of the article Marin Zapata et al. BMC Cancer (2016) 16:355 DOI 10.1186/s12885-016-2388-9 Marin Zapata et al. BMC Cancer (2016) 16:355 DOI 10.1186/s12885-016-2388-9 Time course decomposition of cell heterogeneity in TFEB signaling states reveals homeostatic mechanisms restricting the magnitude and duration of TFEB responses to mTOR activity modulation Paula Andrea Marin Zapata1, Carsten Jörn Beese1†, Anja Jünger1,2†, Giovanni Dalmasso1, Nathan Ryan Brady2,3,4* and Anne Hamacher-Brady1,4* © 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Background lysosomal stress [7], suggesting TFEB concentration changes may contribute to the regulation of TFEB sig- naling. However, the relationship between mTOR ac- tivity states and temporal changes in TFEB subcellular localization and concentration has not been elucidated. To that end, we performed time course analysis over 15 hours of TFEB levels and localization by quantita- tive Western blotting and imaging cytometry. We acti- vated mTOR by fresh addition of fully-supplemented medium (FM), or inhibited mTOR by Torin1 treat- ment [13] or nutrient deprivation [14]. We report that overall cellular TFEB levels transiently decrease in re- sponse to small increases in mTOR activation, and transiently increase in response to mTOR inhibition. Both Western blot and population-averaged imaging results displayed high variability, suggesting that het- erogeneous TFEB responses within the cell population may cache important information on these complex dynamics. We therefore analyzed single-cell imaging cytometry data using spanning-tree progression analysis of density-normalized events (SPADE) agglomerative clus- tering [15], as a basis for unbiased and quantitative detec- tion of spatial and temporal dynamics of subpopulations. Using unsupervised clustering, we identified three TFEB phenotype subpopulations, with low, medium and high nuclear TFEB concentrations. We found that total cel- lular TFEB levels and subcellular localization are dir- ectly under control of a short-term rheostat controlled by mTOR. mTOR inhibition rapidly activates TFEB in a fraction of cells, for a limited duration, with distinct TFEB subpopulation re-inactivation dynamics in response to Torin1 vs. nutrient deprivation. Moreover, time course subpopulation analysis identified a correlation between TFEB protein levels and nuclear localization, and revealed differences between HeLa and MCF7 cells in the sensi- tivity of TFEB to mTOR regulation. Finally, subpopula- tion analysis revealed that in response to mTOR inhibition, maximal nuclear localization of TFEB is ne- gatively regulated by the proteasome, independently of TFEB concentration. Autophagy, a process of lysosomal degradation essential for cellular homeostasis, is transcriptionally regulated by Transcription Factor EB (TFEB) [1–3], which coordinates the expression of genes involved in lysosome biogenesis, autophagy and endocytosis [1, 2, 4]. Under normal growth conditions TFEB is transiently recruited to the lysosomes through its interaction with active RAG GTPases at the lysosomal membrane [5]. Active RAG GTPases also recruit the anabolic kinase complex mTOR, which phos- phorylates TFEB at serine S211 to promote its dissociation from the lysosome and binding with 14-3-3 protein family members, which retain TFEB in the cytoplasm and inhibit its transcriptional activity [5–7]. Abstract Background: TFEB (transcription factor EB) regulates metabolic homeostasis through its activation of lysosomal biogenesis following its nuclear translocation. TFEB activity is inhibited by mTOR phosphorylation, which signals its cytoplasmic retention. To date, the temporal relationship between alterations to mTOR activity states and changes in TFEB subcellular localization and concentration has not been sufficiently addressed. Methods: mTOR was activated by renewed addition of fully-supplemented medium, or inhibited by Torin1 or nutrient deprivation. Single-cell TFEB protein levels and subcellular localization in HeLa and MCF7 cells were measured over a time course of 15 hours by multispectral imaging cytometry. To extract single-cell level information on heterogeneous TFEB activity phenotypes, we developed a framework for identification of TFEB activity subpopulations. Through unsupervised clustering, cells were classified according to their TFEB nuclear concentration, which corresponded with downstream lysosomal responses. Results: Bulk population results revealed that mTOR negatively regulates TFEB protein levels, concomitantly to the regulation of TFEB localization. Subpopulation analysis revealed maximal sensitivity of HeLa cells to mTOR activity stimulation, leading to inactivation of 100 % of the cell population within 0.5 hours, which contrasted with a lower sensitivity in MCF7 cells. Conversely, mTOR inhibition increased the fully active subpopulation only fractionally, and full activation of 100 % of the population required co-inhibition of mTOR and the proteasome. Importantly, mTOR inhibition activated TFEB for a limited duration of 1.5 hours, and thereafter the cell population was progressively re-inactivated, with distinct kinetics for Torin1 and nutrient deprivation treatments. (Continued on next page) © 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Page 2 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 (Continued from previous page) Conclusion: TFEB protein levels and subcellular localization are under control of a short-term rheostat, which is highly responsive to negative regulation by mTOR, but under conditions of mTOR inhibition, restricts TFEB activation in a manner dependent on the proteasome. We further identify a long-term, mTOR-independent homeostatic control negatively regulating TFEB upon prolonged mTOR inhibition. Abstract These findings are of relevance for developing strategies to target TFEB activity in disease treatment. Moreover, our quantitative approach to decipher phenotype heterogeneity in imaging datasets is of general interest, as shifts between subpopulations provide a quantitative description of single cell behaviour, indicating novel regulatory behaviors and revealing differences between cell types. Keywords: Transcription Factor EB (TFEB), Mammalian target of rapamycin (mTOR), Autophagy, Lysosomes, Proteasome, Systems biology, Subpopulation dynamics, Single cell, Multispectral imaging cytometry Immunofluorescence and fluorescence microscopy Immunofluorescence and fluorescence microscopy Fifty thousand cells were plated per well of an 8 well μ- slide microscopy chamber (ibidi) 24 hours before treat- ment. Following drug treatments, cells were fixed with 4 % paraformaldehyde in PBS for 15 minutes, permeabilized Immunoblotting Six hundred thousand cells/well were plated in 6-well plates, 24 hours prior drug treatment. Following drug treatments whole cell lysates were prepared of adherent and floating cells with RIPA lysis buffer containing 1X EDTA-free protease inhibitor cocktail (Roche) and 2X PhosphoSTOP (Roche). Dosed protein samples were separated on pre-cast 4–12 % Bis-Tris gels (Invitrogen) and transferred to nitrocellulose using the iBlot dry blot- ting system (Invitrogen). Blocked membranes were incu- bated with primary antibodies against TFEB (#101532; Santa Cruz), LAMP1 (# H4A3-s; Hybridoma Bank), LC3 (#2775; Cell Signaling), 4E-BP1 (#9452; Cell Signaling), phospho-4E-BP1 (#9459S; Cell Signaling), p70-S6K1 rabbit IgG (#9202S; Cell Signaling), phospho-p70-S6K1 (#9205S; Cell Signaling) and GAPDH (#25778; Santa Cruz). HRP-conjugated anti-rabbit IgG (#213110-01; GeneTex) and anti-mouse IgG (#213111-01; GeneTex) antibodies were used as secondary antibodies. For immunodetection membranes were incubated with peroxide and luminol so- lution (1:1) and analyzed with a chemiluminescence imager (Intas). Protein bands were quantified using the gel analysis tool of ImageJ and normalized to the loading control GAPDH. Blots shown are representative of at least three in- dependent experiments. Materials Cell culture reagents were obtained from Invitrogen, Sigma, Lonza and PAN Biotech. Methanol-free parafor- maldehyde was obtained from Alfa Aesar. Torin1 was pur- chased from Merck, DMSO from Genaxxon Biosciences and U0126 was from Biovision. Hoechst 33342 was pur- chased from ImmunoChemistry. Cloning Entry Clones were obtained from the German cDNA Consortium of the German Cancer Research Center. N-terminally tagRFP-tagged clone of 14-3-3 protein isoform YWHAG, RFP-YWHAG, was generated using the Gateway Cloning System (Life Technologies). TFEB wild type was cloned using forward primer: 5′-gtaAAGC TTcgatggcgtcacgcatagggttgcgcatg-3′ and reverse primer 5′- tacGGTACCttacagcacatcgccctcctccat-3′ and inserted into pEGFP (Invitrogen) generating TFEB with N- terminal GFP fusion, GFP-TFEB. Cell culture and treatments The human cervical cancer cell line HeLa Kyoto and the human breast cancer cell line MCF7 (obtained from CLS Cell lines service, Heidelberg) were cultured in DMEM (1 g/L D-glucose, 0.11 g/L sodium pyruvate), supplemented with 2 mM L-Glutamine, 10 % Fetal Bovine Serum, non-essential amino acids and penicillin/streptomycin/ amphotericin B. Cells were routinely tested for myco- plasma contamination using Hoechst 33342. Transient transfections were performed using jetPRIME (Polyplus) according to the manufacturer’s instructions. Transfection complexes were removed after 6 hours and experiments performed at 24 hours of expression. Nutrient deprivation (ND) was introduced using glucose-containing HBSS (Life Technologies; no. 14025), supplemented with penicillin/ streptomycin/amphotericin B. For drug treatments, cells were incubated in FM or HBSS, containing one or a com- bination of the following reagents: Torin1 (2 μM), U0126 (10 μM), epoxomicin (1 μM), and actinomycin D (1 μg/ml). Co-treatments with epoxomicin, actinomycin D or DMSO included a pre-treatment period (Fig. 7c-e). Cells were pretreated with Epox, ActD or vehicle control (DMSO) for 1 hour, and subsequently treated with FM supple- mented with Torin1 in combination with the respective pretreatment reagent for 1 hour. For pre-treatments the drugs were directly added to the culture medium, without addition of fresh FM. Background Upon amino acid starva- tion, RAG GTPases are inactivated [8] resulting in the loss of lysosomal recruitment of TFEB and mTOR. Conse- quently, the cytoplasmic pool of TFEB becomes dephos- phorylated, leading to the dissociation from 14-3-3 proteins and ultimately to nuclear accumulation of TFEB. Besides amino acid starvation, pharmacological inhibition of mTOR and lysosomal stresses result in TFEB de- phosphorylation and nuclear accumulation [7, 9]. In the nucleus, TFEB activates the transcription of the CLEAR network (Coordinated Lysosomal Expression and Regula- tion), which is composed of at least 471 direct targets, in- cluding a battery of lysosomal and autophagy genes [1]. Abnormalities in autophagic processes can lead to neurodegenerative diseases and cancer [10]. Moreover, recent studies have identified TFEB and other family members as key players for metabolic reprogramming in pancreatic cancer [11, 12]. Thus, TFEB presents an attractive target for manipulating the cellular autopha- gic capacity in disease treatment. To date, studies on TFEB have primarily focused on the role of mTOR- mediated regulation of nuclear-cytoplasmic TFEB shut- tling. Intriguingly, transcription of TFEB-controlled autophagosomal and lysosomal genes is increased in cells overexpressing TFEB [1, 2, 6] and overall cellular TFEB protein levels are reduced following TFEB activa- tion via long-term (15 hours) chloroquine-induced Page 3 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Multispectral imaging cytometry Flow cytometry coupled to high resolution imaging was performed using the ImageStreamX cytometer operated with INSPIRE 4.1.501.0 software (Amnis), using a 40X air objective. Methods with 0.3 % Triton X-100 in PBS for 10 minutes and blocked with 3 % BSA in 0.3 % Triton X-100/PBS for 1 hour. Cells were then incubated with primary antibodies against LAMP1 (Hybridoma Bank; #H4A3-s), TFEB (Cell Signaling; #4240S), or p-4E-BP1 (Cell Signaling; #2855S) in 0.3 % Triton X-100/PBS at 4 °C overnight. Fluorescence staining was performed using anti-rabbit Alexa Fluor 488 or 594 secondary antibodies (Life Technologies; #A11008, #A11012) in 0.3 % Triton X-100/PBS at room temperature for 1 hour. Fluorescence microscopy was performed with a DeltaVision microscope system (Applied Precision) using a 60x oil immersion objective (Olympus) and a digital CCD camera (Hamamatsu Photonics). Following acquisition, im- ages were deconvolved with Softworks V3.5.1 (Applied Precision) to increase spatial resolution. Images were prepared using ImageJ (rsbweb.nih.gov/ij/). Representative images shown are total intensity projections (Z-axis scans). with 0.3 % Triton X-100 in PBS for 10 minutes and blocked with 3 % BSA in 0.3 % Triton X-100/PBS for 1 hour. Cells were then incubated with primary antibodies against LAMP1 (Hybridoma Bank; #H4A3-s), TFEB (Cell Signaling; #4240S), or p-4E-BP1 (Cell Signaling; #2855S) in 0.3 % Triton X-100/PBS at 4 °C overnight. Fluorescence staining was performed using anti-rabbit Alexa Fluor 488 or 594 secondary antibodies (Life Technologies; #A11008, #A11012) in 0.3 % Triton X-100/PBS at room temperature for 1 hour. Fluorescence microscopy was performed with a DeltaVision microscope system (Applied Precision) using a 60x oil immersion objective (Olympus) and a digital CCD camera (Hamamatsu Photonics). Following acquisition, im- ages were deconvolved with Softworks V3.5.1 (Applied Precision) to increase spatial resolution. Images were prepared using ImageJ (rsbweb.nih.gov/ij/). Representative images shown are total intensity projections (Z-axis scans). Data processing All d All data processing was performed using the IDEAS v6.0 software (Amnis). For each treatment and time point a total of 10000 cells were collected. Following compensa- tion, cells were gated as single (based on the area and as- pect ratio of the bright filed mask) and in-focus (based on the Gradient RMS of the bright filed image). With the exception of 15 hours, at least 2000 cells were ana- lyzed after gating. The nuclear, cytoplasmic and cellular masks of gated cells were calculated based on the follow- ing morphological and logical operations: Cell, default mask for TFEB channel OR 5-pixel erosion of default bright field mask; Nucleus, 70 % Threshold mask on Hoechst channel; Cytoplasm: Cell AND NOT Nucleus. The features “Intensity Cell, Nucleus and Cytoplasm” were calculated as the total intensities (background subtracted) in their respective masks. The features “Con- centration Cell, Nucleus and Cytoplasm” were calculated by dividing the intensity features by the area of their re- spective masks (in μm2). The feature “Nuclear percent- age” was calculated as the ratio between the features “Intensity Nucleus” and “Intensity Cell”, multiplied by 100. The feature “Max Contour Position” was calculated with a build in function available in IDEAS software [16]. The feature “Mean Pixel Nu/Cyto” was calculated based on masks which underestimated the nuclear and cytoplasmic compartments in order to avoid including cytoplasmic pixels in the nuclear signal or including background or nuclear pixels in the cytoplasmic signal. Underestimated masks were obtained by morphological erosion of the original masks. These feature values were exported to.fcs-files for further processing with the clus- tering software SPADE V2.0 (Spanning-tree Progression Analysis of Density-normalized Events) [15]. The num- ber of clusters and combination of input features was optimized as presented on the Results Section. The remaining SPADE input parameters were set to default values (arcsinh with cofactor = 5, neighborhood size = 5, Data representation l Mean population responses were obtained by averaging the single-cell data from a specific treatment, time point and repetition. All subpopulations were identified according to the classification model obtained from FM and Torin1 data in Fig. 4 (Refer to Additional file 1: Figure S1 for further explanation of the classification work flow). Statistical comparisons Statistical comparisons were performed with Student’s two-tailed t-test or the non-parametric test Wilcoxon- rank-sum (two-sided). The latter was used in inter-cluster comparisons of non-normally distributed variables, in- cluding the features “Mean Pixel Nuc/Cyto” (Figs. 4b, d and 5b) and the discrete variable “LAMP1 Max Contour Position” (Fig. 6d). Data collection Two hundred fifty thousand cells per well of a 12-well plate were plated on the day before drug treatments. Fol- lowing drug treatments, cells were trypsinized, harvested Page 4 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 by centrifugation at 800 g for 5 minutes at 4 °C and fixed with 4 % paraformaldehyde for 15 minutes at room temperature. For detection of endogenous TFEB, cells were immunostained as stated above. Nuclei were labeled with 1 μg/mL Hoechst 33342 in PBS for 10 minutes. Compensation controls were generated from single-color control cells. Endogenous TFEB was immunostained with an antibody against TFEB and Alexa Fluor 594. Lysosomes were immunostained with an antibody against LAMP1 and Alexa Fluor 647. For measurements cells were resus- pended in PBS. Fluorescence signal of Hoechst 33342 was excited using the 405 nm laser and detected in channel 1 (420–480 nm). Alexa Fluor 594 was excited using the 561 nm laser and the fluorescence signal was detected in channel 4 (595–642 nm). local density approximation factor = 1.5, max allowable cells in pooled down-sampled data = 50000, fixed num- ber of cells remained = 20000, Algorithm: K-means). Clustering results were exported to.fcs files and subse- quently converted to .txt files using the IDEAS software. Text files were imported into MATLAB R2014a for data representation and further analysis. Regulation of TFEB localization and protein levels by mTOR We first established conditions for suppressing mTOR activity with the specific inhibitor Torin1 [13], and in- creasing mTOR activity by the renewed addition of fully-supplemented medium (FM) (illustrated in Fig. 1a). Torin1-mediated TFEB activation has been reported for concentrations ranging from 0.25 μM [5, 6, 9] to 2 μM [7]. Thus, we treated HeLa cells with 0.25 to 2 μM Torin1 for 1.5 and 3.0 hours, and determined the phos- phorylation state of the mTOR substrates 4E-BP1 and p70-S6K1 by Western blot. Maximal inhibition of 4E- BP1 and p70-S6K1 phosphorylation was achieved in response to 2 μM after 3 hour treatment (Fig. 1b, c). High-resolution imaging further demonstrated that at 3 hours of treatment with 2 μM Torin1, immunofluores- cence detection of phosphorylated 4E-BP1 was fully sup- pressed, and nuclear accumulation of TFEB was potently induced (Fig. 1d). Interestingly, total cellular TFEB im- munofluorescence appeared strongly increased under Torin1 treatment, suggesting that mTOR inhibition in- creased TFEB protein levels. Of note, the addition of fresh FM resulted in an increased immunofluorescence signal of phosphorylated 4E-BP1, indicating mTOR activation by the replenished metabolic substrates and growth factors present in fresh FM. Consistent with increased mTOR Page 5 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Fig. 1 Characterization of the effect of Torin1 and fresh nutrients on mTOR and endogenous TFEB. a Schematic representation of the effects of Torin1 and fresh fully-supplemented medium (FM) on the regulation of TFEB by mTOR. b Dose-response of the effect of Torin1 on mTOR activity. HeLa cells were treated with FM containing the indicated concentrations of Torin1, or kept in culture medium (non-treated, NT), for 1.5 or 3 hours, and phosphorylation of the mTOR substrates 4E-BP1 and p70-S6K1 was measured by Western blotting. c Quantification of the ratio of phosphorylated 4E-BP1 (p-4E-BP1) to total 4E-BP1. Graphs display mean values of three independent experiments normalized to NT values. Error bars denote mean ± SD of three independent experiments. Statistical significance was tested vs. NT conditions (Student’s two-tailed t-test; , p ≤0.01; , p ≤0.001). d Immunofluorescence of TFEB and p-4E-BP1, as a read-out for mTOR activity, in response to FM and Torin1. HeLa cells were kept in culture medium (non-treated, NT), treated with fresh FM, or with FM supplemented with Torin1 (2 μM) for 3 hours and immunostained for TFEB and p-4E-BP1. Regulation of TFEB localization and protein levels by mTOR To reveal varying intensity levels the look-up-table ‘Fire’ (ImageJ) was applied to grey scale images, representing intensity values ranging from low (dark purple) to high (white) as displayed in color scale bar. Scale bars, 20 μm Fig. 1 Characterization of the effect of Torin1 and fresh nutrients on mTOR and endogenous TFEB. a Schematic representation of the effects of Torin1 and fresh fully-supplemented medium (FM) on the regulation of TFEB by mTOR. b Dose-response of the effect of Torin1 on mTOR activity. HeLa cells were treated with FM containing the indicated concentrations of Torin1, or kept in culture medium (non-treated, NT), for 1.5 or 3 hours, and phosphorylation of the mTOR substrates 4E-BP1 and p70-S6K1 was measured by Western blotting. c Quantification of the ratio of phosphorylated 4E-BP1 (p-4E-BP1) to total 4E-BP1. Graphs display mean values of three independent experiments normalized to NT values. Error bars denote mean ± SD of three independent experiments. Statistical significance was tested vs. NT conditions (Student’s two-tailed t-test; , p ≤0.01; , p ≤0.001). d Immunofluorescence of TFEB and p-4E-BP1, as a read-out for mTOR activity, in response to FM and Torin1. HeLa cells were kept in culture medium (non-treated, NT), treated with fresh FM, or with FM supplemented with Torin1 (2 μM) for 3 hours and immunostained for TFEB and p-4E-BP1. To reveal varying intensity levels the look-up-table ‘Fire’ (ImageJ) was applied to grey scale images, representing intensity values ranging from low (dark purple) to high (white) as displayed in color scale bar. Scale bars, 20 μm Fig. 1 Characterization of the effect of Torin1 and fresh nutrients on mTOR and endogenous TFEB. a Schematic representation of the effects of Torin1 and fresh fully-supplemented medium (FM) on the regulation of TFEB by mTOR. b Dose-response of the effect of Torin1 on mTOR activity. HeLa cells were treated with FM containing the indicated concentrations of Torin1, or kept in culture medium (non-treated, NT), for 1.5 or 3 hours, and phosphorylation of the mTOR substrates 4E-BP1 and p70-S6K1 was measured by Western blotting. c Quantification of the ratio of phosphorylated 4E-BP1 (p-4E-BP1) to total 4E-BP1. Graphs display mean values of three independent experiments normalized to NT values. Error bars denote mean ± SD of three independent experiments. Statistical significance was tested vs. NT conditions (Student’s two-tailed t-test; , p ≤0.01; **, p ≤0.001). Regulation of TFEB localization and protein levels by mTOR d Immunofluorescence of TFEB and p-4E-BP1, as a read-out for mTOR activity, in response to FM and Torin1. HeLa cells were kept in culture medium (non-treated, NT), treated with fresh FM, or with FM supplemented with Torin1 (2 μM) for 3 hours and immunostained for TFEB and p-4E-BP1. To reveal varying intensity levels the look-up-table ‘Fire’ (ImageJ) was applied to grey scale images, representing intensity values ranging from low (dark purple) to high (white) as displayed in color scale bar. Scale bars, 20 μm activity, TFEB was predominantly retained in the cyto- plasm under fresh FM conditions. is only mildly up-regulated, at levels below the sensitivity of Western blot analysis. However, at time points of 5 and 15 hours the reduction to 4E-BP1 phosphorylation suggests a progressive and significant reduction to mTOR activity. Time course quantification of TFEB response to modulations in mTOR activity by Western blot analysis d HeLa cells were treated with FM containing 2 μM Torin1. At the indicated time points, levels of TFEB and phosphorylated 4E-BP1 (p-4E-BP1) were analyzed by Western blotting. e Quantified values for TFEB, normalized to loading control GAPDH, shown relative to time point ‘0’. f Quantified values for p-4E-BP1, normalized to total 4E-BP1, shown relative to time point ‘0’. Error bars denote mean ± SD of three independent experiments. Statistical significances were tested vs. time point ‘0’ (Student’s two-tailed t-test; , p ≤0.05; **, p ≤0.001) Fig. 2 Quantitative Western blot analysis of TFEB protein levels in response to mTOR activity modulations. a HeLa cells were treated with fresh FM to enhance mTOR activity. At the indicated time points, levels of TFEB and phosphorylated 4E-BP1 (p-4E-BP1) were analyzed by Western blotting. Lanes for time point ‘0’ originate from same membrane as later time points. b Quantified values for TFEB, normalized to loading control GAPDH, shown relative to time point ‘0’. c Quantified values for p-4E-BP1, normalized to total 4E-BP1, shown relative to time point ‘0’. d HeLa cells were treated with FM containing 2 μM Torin1. At the indicated time points, levels of TFEB and phosphorylated 4E-BP1 (p-4E-BP1) were analyzed by Western blotting. e Quantified values for TFEB, normalized to loading control GAPDH, shown relative to time point ‘0’. f Quantified values for p-4E-BP1, normalized to total 4E-BP1, shown relative to time point ‘0’. Error bars denote mean ± SD of three independent experiments. Statistical significances were tested vs. time point ‘0’ (Student’s two-tailed t-test; , p ≤0.05; *, p ≤0.001) To gauge the sensitivity of imaging cytometry (ISX) for assessing TFEB subcellular localization we compared the nuclear/cytoplasmic ratios of TFEB (Fig. 3a), and TFEB and Hoechst intensity profiles (Additional file 2: Figure S2), between ISX and high-resolution wide field imaging (WF) data-sets. Extended sets of representative ISX images for each condition are presented in Additional file 3: Figure S3. Similar intensity profiles were obtained with both techniques for all conditions. Furthermore, the nuclear/cytoplasmic ratios obtained for WF were higher than for ISX measurements. However, qualitatively similar FM and Torin1 responses were obtained. Both imaging approaches reported reduced nuclear localization in response to 3 hour FM treatment (8 % reduction for WF and 19 % for ISX) and increased nuclear localization upon treatment with Torin1 (38 % in- crease for WF and 47 % for ISX). Time course quantification of TFEB response to modulations in mTOR activity by Western blot analysis Time course quantification of TFEB response to modulations in mTOR activity by Western blot analysis y Torin1 (2 μM) treatment on the other hand initially re- sulted in stable TFEB protein levels, which increased sig- nificantly after 3 hours (Fig. 2d, e), similar to as observed by imaging (Fig. 1d). Similar to following long-term (15 hour) mTOR inhibition with chloroquine [7], follow- ing 5 and 15 hours of Torin1 treatment TFEB levels were reduced (Fig. 2d, e), albeit with a high degree of variation between experiments. Importantly, Torin1 inhibition of 4E-BP1 phosphorylation was maintained also at 15 hours (Fig. 2f). These findings suggest that total TFEB levels are oppositely regulated by Torin1 and FM treatments during the initial 3-hour treatment period, followed by a pro- longed TFEB recovery to initial levels. However, for most time points, the variability of immunoblotting data was too high to infer dynamic behavior of TFEB. To quantitatively investigate the effect of mTOR activity modulation on TFEB protein levels we treated HeLa cells with either fresh FM alone or containing 2 μM Torin1 over a time course of 15 hours, and measured levels of TFEB and phosphorylated 4E-BP1 by quantita- tive Western blot. Following the addition of fresh FM, TFEB protein levels were significantly decreased between 0.5, 1 and 3 hours, followed by a prolonged recovery to basal levels (Fig. 2a, b). In parallel, 4E-BP1 phosphoryl- ation was stable in the first hour and then decreased over time, significantly at 5 and 15 hours (Fig. 2a, c). Notably, in contrast to imaging results (Fig. 1d), a sig- nificant increase in 4E-BP1 phosphorylation at 3 hours was not detected after addition of FM (Fig. 2c), indicat- ing that in response to the addition of fresh FM mTOR Marin Zapata et al. BMC Cancer (2016) 16:355 Page 6 of 19 Fig. 2 Quantitative Western blot analysis of TFEB protein levels in response to mTOR activity modulations. a HeLa cells were treated with fresh FM to enhance mTOR activity. At the indicated time points, levels of TFEB and phosphorylated 4E-BP1 (p-4E-BP1) were analyzed by Western blotting. Lanes for time point ‘0’ originate from same membrane as later time points. b Quantified values for TFEB, normalized to loading control GAPDH, shown relative to time point ‘0’. c Quantified values for p-4E-BP1, normalized to total 4E-BP1, shown relative to time point ‘0’. Time course quantification of TFEB response to modulations in mTOR activity by Western blot analysis We thus conclude that the ISX approach is more sensitive. Moreover, higher population sampling permits more robust quantitative analysis of relative changes induced by conditions, and, importantly, allows for improved significance testing. Mean population time course quantification of TFEB response to modulations in mTOR activity by multispectral imaging cytometry Multispectral imaging cytometry for quantification of endogenous TFEB in cell populations Multispectral imaging cytometry for quantification of endogenous TFEB in cell populations Agglomerative clustering analysis of single cell multispectral imaging cytometry data identifies underlying subpopulation dynamics and elucidates temporal TFEB regulation Agglomerative clustering analysis of single cell multispectral imaging cytometry data identifies underlying subpopulation dynamics and elucidates temporal TFEB regulation As Western blot and population-averaged ISX analyses report bulk population dynamics of TFEB, we hypothe- sized that cell-to-cell heterogeneity in TFEB signaling may contribute to time point variability for both ap- proaches, and thereby contain relevant information on TFEB dynamics. Therefore, we sought to quantify subpop- ulation TFEB responses from single cell multispectral im- aging cytometry data using SPADE-based agglomerative clustering [15]. Upon treatment with Torin1, at 0.5 hours the nuclear/ cytoplasmic ratio increased (from 1.4 to 1.9), peaking at 1 hour, and following 1.5 hours was gradually reduced to a final distribution of 1.4 at 15 hours, similar to time point 0 (Fig. 3b). Consistent with Western blot findings, Torin1 increased cellular TFEB concentration (Fig. 3c), significantly at 1.5 (45 %) and 3 hours (38 %), after which concentrations were reduced to approximately initial (t = 0) levels. Multispectral imaging cytometry for quantification of endogenous TFEB in cell populations TFEB exerts its activity in the nucleus, and thus spatial dynamics of TFEB signaling contain relevant informa- tion. We therefore performed single-cell analysis of TFEB subcellular localization and protein levels in cell populations using the imaging cytometer, ImageStreamX (ISX) [17]. For each obtained single-cell image, the nu- clear and cytoplasmic compartments were segmented, and based on these masks and the total fluorescence intensity of endogenous TFEB, two normalized features were calculated to report spatial concentration states. The first feature, “Mean Pixel Nuc/Cyto”, reflects the nuclear/cytoplasmic ratio of TFEB concentration, and was calculated as the ratio of the mean pixels from each compartment. The second feature, referred to as “Concentration”, was determined by normalizing the total intensity of TFEB to the cell area (described in Methods). Marin Zapata et al. BMC Cancer (2016) 16:355 Page 7 of 19 Fig. 3 (See legend on next page.) Fig. 3 (See legend on next page.) Marin Zapata et al. BMC Cancer (2016) 16:355 Page 8 of 19 (See figure on previous page.) Fig. 3 Multispectral imaging cytometry quantification of TFEB localization and levels in response to mTOR activity modulations. Cells were kept in culture medium (NT, non-treated), or treated with fresh FM or FM supplemented with Torin1 (2 μM). Following, cells were immunostained for TFEB and nuclei labelled with Hoechst 33342. a Representative fluorescence images and quantified TFEB nuclear localization in HeLa cells at 3 hours of treatment, measured with high-resolution wide field imaging (WF, left panels) or with the multispectral imaging cytometer ImageStreamX (ISX, right panels). Graphed values represent the mean ± SD nuclear/cytoplasmic ratio of 25 to 30 randomly selected cells of one representative experiment from three independent repetitions. Statistical significances were tested vs. NT control (Student’s two-tailed t-test; , p ≤0.001; n.s., p > 0.05). b-e Time course of mean population response of TFEB subcellular localization and protein levels for treatments with Torin1 or fresh FM, in HeLa and MCF7 cell lines. Concentrations are shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Statistical significances were tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01) Time course quantification of mean TFEB responses to FM and Torin1 treatments Time course quantification of mean TFEB responses to FM and Torin1 treatments Time course quantification of mean TFEB responses to FM and Torin1 treatments Furthermore, these results indicate that changes in TFEB protein levels correlate with significant shifts of TFEB between nuclear and cytoplasmic compartments. Next, we assessed the mean population responses in HeLa cells treated under the conditions and time points reported in Fig. 2. Initially (t = 0), TFEB displayed a slightly higher concentration in the nuclear compart- ment, with a nuclear/cytoplasmic ratio of 1.4 (Fig. 3b). Upon treatment with fresh FM, at 0.5 hours the nuclear/ cytoplasmic ratio rapidly decreased (from 1.4 to 1.0), and then gradually increased back to initial levels during the later time points. Consistent with Western blot find- ings, fresh FM induced a rapid, 14 % decrease in mean total cell TFEB concentrations within 0.5 hours, which was maintained up to 15 hours (Fig. 3c). Analytical framework for subpopulation analysis of TFEB distribution in time course datasets Our framework for subpopulation identification consists of five main steps: (I) feature extraction, (II) data merge, (III) clustering, (IV) phenotypes assessment, and (V) time course distribution analysis (Fig. 4a). In the first step, multiple quantitative features, including subcellular localization and total protein levels, are calculated for each cell from all treatments and time points. In the sec- ond step, the extracted features from all conditions are merged together, to ensure that clustering is not influ- enced by time points and treatments, and thus is un- biased. In the third step, the clustering algorithm SPADE is used to split the cells into a given number of groups (clusters), which should represent different phenotypes. In the fourth step, the clustering outcome is evaluated based on several criteria to assess its biological soundness, and the clustering step is iteratively repeated to establish a combination of features and cluster number (if clusters exist) for which the results satisfy all evaluation criteria. Finally, in the fifth step, we trace back the dynamic distribution of the population among the obtained clusters. Specifically, for each treatment and time point, we determine the percentage of cells belonging to each cluster. Assuming that the clusters represent biologically-meaningful phenotypes, the redistribution We further evaluated the effect of fresh FM and Torin1 treatments in MCF7 cells. Consistent with the findings in HeLa cells, within 1 hour of treatment with fresh FM the TFEB nuclear/cytoplasmic ratio slightly but non-significantly decreased from 1.8 to 1.7 (Fig. 3d) and overall TFEB protein levels were reduced by 5 % (Fig. 3e). Conversely, Torin1 treatment increased the nuclear/ cytoplasmic ratio to 2.2 (Fig. 3d) and increased TFEB levels by 20 % (Fig. 3e). As in HeLa cells, TFEB nuclear localization and cellular concentration increased transi- ently in response to Torin1 and, after approximately 1 hour, decreased gradually. Of note, in MCF7 cells, at 15 hours of Torin1 treatment TFEB concentration decreased below the initial values. Taken together, ISX-based analysis of mean population responses support Western blot findings, wherein mTOR inhibition by Torin1 increases TFEB protein levels in HeLa (Figs. 2e and 3c) as well as in MCF7 cells (Fig. 3e). Conversely, addition of fresh FM, to mildly in- crease mTOR activity (Fig. 1d), led to a slight, but sig- nificant, reduction in TFEB levels (Figs. 2b and 3c, e). Marin Zapata et al. BMC Cancer (2016) 16:355 Page 9 of 19 Fig. Analytical framework for subpopulation analysis of TFEB distribution in time course datasets 4 (See legend on next page.) Fig. 4 (See legend on next page.) Page 10 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 (See figure on previous page.) Fig. 4 Clustering-based analysis of subpopulation dynamics of TFEB activity in response to mTOR activity modulation by fresh FM and Torin1. Subpopulation analysis of ISX multispectral imaging cytometry datasets from Fig. 3. a Schematic representation of the analysis workflow. Extracted feature values from all treatments and time points were merged and analyzed using SPADE software for the identification of phenotypically similar clusters. Clustering results were iteratively checked until finding a combination of features and number of clusters that yielded biologically sound and reproducible results. Finally, evolution of subpopulations was observed by tracing the percentage of cells belonging to each cluster for each treatment and time point. b Cluster phenotypes in HeLa cells. Bars represent the mean among all cells in each cluster ± SD (including FM, Torin1, and all repetitions and time points). The number of cells equals 48774, 42994 and 23793 for cluster 1, 2 and 3, respectively. Statistical significances were tested between clusters on 1000 randomly selected cells (two-sided Wilcoxon-rank-sum test; , p ≤0.001). c Cumulative frequency distribution for selected features that were excluded during the generation of the clusters. Bars on the top right corners display the mean value among all cells in each cluster ± SD. Statistical significances were tested between clusters on 1000 randomly selected cells (Student’s two-tailed t-test; , p ≤0.001). d Clusters phenotype in MCF7 cells. Bars represent the mean among all cells in each cluster ± SD. Statistical significances were tested between clusters on 1000 randomly selected cells (two-sided Wilcoxon-rank-sum test; , p ≤0.001). e Mean TFEB protein levels for the three clusters. Bars represent the mean among all cells in each cluster ± SD. Statistical significances were tested between clusters on 1000 randomly selected cells (Student’s two-tailed t-test; *, p ≤0.001). f-g Subpopulation dynamics for the indicated treatments and cell lines. Reported values represent the mean among three independent experiments ± SD. Regions shaded in grey highlight different stages in TFEB dynamic response. R1: short-term rheostatic response, R2: long-term response We characterized the predicted clusters based on the frequency distribution and mean values of a subset of features that were not included in the cluster generation. Analytical framework for subpopulation analysis of TFEB distribution in time course datasets TFEB nuclear percentage and cellular concentration features (described in Methods) yielded normal distribu- tions within predicted clusters, with statistically-different means (Fig. 4c), further indicating that the predicted clus- ters represent biologically-meaningful subpopulations. Interestingly, the “Active” cluster contained the highest total cellular TFEB concentration. We confirmed this positive correlation between TFEB nuclear localization and TFEB protein levels through statistical analysis, with a correlation coefficient of 0.53 (See Additional file 1: Figure S1, panel II). To test whether TFEB protein levels and localization were correlated independently of mTOR, we increased cellular TFEB concentration by ectopic expression of GFP-TFEB, and quantified the percentage of activated cells (mainly nuclear TFEB) by fluorescence microscopy (Additional file 6: Figure S6). The amount of activated cells was significantly increased by GFP-TFEB expression compared to endogenous TFEB levels. Furthermore, the effect of overexpression was par- tially reversed by enhanced sequestration of TFEB in the cytosol through overexpression of 14-3-3 isoform ɣ (YWHAG), for which TFEB has a high binding affinity [6]. These results suggest that increased cellular TFEB protein levels can trigger nuclear localization and override regula- tion by mTOR, and that this effect is partially dependent on 14-3-3 protein levels. of cells among the different clusters should then indi- cate the development of subpopulations. For this analysis to be valid, we utilize the dynamics of the subpopulation response to evaluate the consistency, i.e. biological soundness, of the clustering results (step IV) based on the following criteria: – Criterion 1: The temporal evolution of the percentage of cells in each cluster should be consistent among repetitions, to assure the reproducibility of subpopulation dynamics. (See Additional file 4: Figure S4 and Additional file 5: Figure S5a). g – Criterion 2: The distribution of cells in each cluster should follow independent dynamics. We assume that if the distribution of cells in two or more clusters are affected in the same way by the treatments, this would indicate that the clusters are redundant (See Additional file 5: Figure S5b). Time course quantification of TFEB mean population and subpopulation responses to nutrient deprivation p p p p We then investigated a physiological perturbation on mTOR and TFEB, by subjecting HeLa cells to nutrient deprivation [6, 7]. First, we calculated the mean popula- tion responses for TFEB subcellular localization and concentration (Fig. 5a). Nutrient deprivation transiently increased the TFEB nuclear/cytoplasmic ratio (from 1.2 to maximally 1.7), peaking at 1 hour, subsequently decreasing up to 5 hours (to 1.3), and then increasing significantly again at 15 hours (to 1.6). Consistent with the findings for Torin1, concomitant to inducing nuclear translocation, nutrient deprivation induced a significant increase in the cellular concentration of TFEB, peaking at 1 hour with an increase of 41 %, and then decreasing gradually. Of note, while nuclear localization was increased again at 15 hours, TFEB protein levels stayed low. In contrast, Torin1 treatment induced a differential subpopulation response, and did not drive 100 % of the population towards a single activation state (Fig. 4f). Within 1.5 hours Torin1 reduced the “Inactive” subpop- ulation to marginal levels (from 45 to 2 %), slightly re- duced the “Medium” subpopulation (from 46 to 41 %), and increased the “Active” subpopulation from 9 to 57 % (region R1). Following 1.5 hours, the “Inactive” and “Medium” subpopulations increased, concurrent to a de- crease in “Active” subpopulation, and by 15 hours all subpopulations were similarly re-distributed to time point 0 levels, with 47 %, 45 % and 8 % of the cells in the “Inactive”, “Medium” and “Active” subpopulations, respectively (region R2). Next, we evaluated the TFEB subpopulation response to nutrient deprivation. Cells were classified into “Ac- tive”, “Medium” and “Inactive” phenotypes using the classification model obtained above for FM and Torin1 treatments (see Additional file 1: Figure S1 for a sche- matic of the classification workflow). Consistent with FM and Torin1, the mean features obtained from the nutrient deprivation data set showed that the “Active” cluster, displayed the highest TFEB concentration, while the “Inactive” cluster contained the lowest TFEB levels (Fig. 5b). Similar, but blunted, subpopulation redistributions were observed in MCF7 cells (Fig. 4g). FM increased the per- centage of cells in the “Inactive” subpopulation (from 57 to 67 %), concomitantly decreasing the “Medium” (from 33 to 26 %) and “Active” subpopulations (from 11 to 6 %), while Torin1 decreased the “Inactive” fraction (to 14 %) and increased the “Active” (to 42 %) and “Medium” sub- populations (to 44 %). Identification of three TFEB activation phenotypes/ subpopulations We applied this framework to identify subpopulations in the response to FM and Torin1 treatments. The features used for this analysis included the nuclear/cytoplasmic ratio, and areas, concentrations and total intensities in the segmented compartments (cellular, nuclear and cyto- plasmic masks). After evaluating different feature combinations and num- ber of clusters, we determined that our evaluation criteria were satisfied by the single feature “Mean Pixel Nuc/Cyto”, i.e. nuclear/cytoplasmic ratios, and a total of three, statisti- cally different clusters. Based on the nuclear/cytoplasmic ratios, the three clusters were classified as “Inactive”, mod- erately active (denoted as “Medium”), and “Active”. The “Active” cluster has the highest nuclear localization and the “Inactive” cluster has the lowest nuclear localization, i.e. highest cytoplasmic retention (Fig. 4b). We identically applied our analysis framework to de- fine TFEB subpopulations in the MCF7 cells time course data (the workflow for defining cell line-specific subpop- ulations is represented in Additional file 1: Figure S1). Similar to HeLa cells, statistically-different clusters were obtained, and the cluster with the highest nuclear localization (“Active”) had the highest TFEB concentra- tion, while the “Inactive” cluster displayed the lowest TFEB levels (Fig. 4d, e). Page 11 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Time course subpopulation responses to FM and Torin1 treatments Importantly, the mean population response of the nuclear/cytoplasmic ratio could be accurately predicted based on the subpopulation distributions (see Additional file 7: Figure S7), confirming the consistency of our ana- lysis at all time points. The temporal impact of fresh FM and Torin1 on the population distribution among predicted activity states was calculated for HeLa cells (Fig. 4f). Results from three independent experiments are shown in Additional file 4: Figure S4. At time point 0, cells were equally dis- tributed between the “Inactive” (45 %) and “Medium” (46 %) subpopulations, while the fraction of “Active” cells amounted to only 9 % (Fig. 4f). Thus, under basal metabolism TFEB activity is inactive-to-moderately ac- tive for most cells. In response to fresh FM, at 0.5 hours the “Inactive” subpopulation fraction increased from 45 % to nearly 100 % of the cell population, concurrent with a decrease in the “Medium” subpopulation (from 46 to 6 %) and a depletion of the “Active” subpopulation (region R1). Beginning at 1 hour, the “Inactive” subpopula- tion decreased, concomitant with an increased “Medium” subpopulation. At 15 hours a distribution of 21 %, 76 % and 3 % for the “Inactive”, “Medium” and “Active” sub- populations was reached, respectively (region R2). The de- crease of the “Inactive” subpopulation is consistent with the reduction in mTOR activity detected 5 and 15 hours post FM treatment (Fig. 2c). Overall, the results from HeLa and MCF7 cells show that subpopulation analysis reveals highly accurate cell type-, condition-, and time-dependent phenotype dy- namics. The above findings reveal that mTOR max- imally induced cytoplasmic TFEB retention in all HeLa cells, but fractionally in MCF7 cells. Conversely in both cell types, Torin1 induced maximal TFEB nuclear con- centration only in a fraction of cells, and following 1.5 hours of treatment, TFEB began a re-localization to the cytoplasm, suggesting an early rheostat con- trol by mTOR followed by TFEB re-inactivation uncoupled from mTOR activity. Time course quantification of TFEB mean population and subpopulation responses to nutrient deprivation As observed in HeLa cells, TFEB activation was gradually reversed starting 1.5 hours after Torin1 treatment, nearly reaching the initial subpopula- tion distribution after 15 hours (region R2). Notably, dif- fering from HeLa cell results, FM treatment did not inactivate 100 % of the population, suggesting that TFEB activity is less sensitive to mTOR activation in MCF7 cells. This is in accordance with the recently described lower mTOR control over TFEB in pancreatic cancer cells [11]. Subsequently, we calculated the temporal impact of nu- trient deprivation on the distribution of TFEB subpopula- tions (Fig. 5c). Within 1 hour, nutrient deprivation induced an initial activation wave (region R1), which reduced the “Inactive” subpopulation (from 80 to 18 %) and increased the “Medium” and “Active” subpopulations from 17 % and 3 % to 56 % and 25 %, respectively. Following 1 hour, TFEB Page 12 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Fig. 5 Mean population and subpopulation response of TFEB to nutrient deprivation. HeLa cells were kept in culture medium (NT, non-treated) or subjected to nutrient deprivation. At the indicated time points, cells were immunostained for TFEB and LAMP1, and Nuclei were labelled with Hoechst 33342. Following, cells were analyzed by ISX multispectral imaging cytometry. a Time course of the mean population response of TFEB subcellular localization and protein levels. Concentrations are shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Unless specified by horizontal lines, statistical significance was tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01). b Mean TFEB subcellular localization and protein levels for the three activation phenotypes in a cell population subjected to nutrient deprivation. Bars represent the mean among all cells in each cluster ± SD (from all repetitions and time points). Statistical significance was tested between clusters on 1000 randomly selected cells (two-sided Wilcoxon-rank-sum test; , p ≤0.001). c Dynamics of the distribution of cells among the three activation phenotypes in a cell population subjected to nutrient deprivation. Reported values represent the mean among three independent experiments ± SD. Shaded regions highlight different stages in TFEB dynamic response. R1: first activation wave, R2: re-inactivation, R3: second activation wave Fig. 5 Mean population and subpopulation response of TFEB to nutrient deprivation. Time course quantification of TFEB mean population and subpopulation responses to nutrient deprivation HeLa cells were kept in culture medium (NT, non-treated) or subjected to nutrient deprivation. At the indicated time points, cells were immunostained for TFEB and LAMP1, and Nuclei were labelled with Hoechst 33342. Following, cells were analyzed by ISX multispectral imaging cytometry. a Time course of the mean population response of TFEB subcellular localization and protein levels. Concentrations are shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Unless specified by horizontal lines, statistical significance was tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01). b Mean TFEB subcellular localization and protein levels for the three activation phenotypes in a cell population subjected to nutrient deprivation. Bars represent the mean among all cells in each cluster ± SD (from all repetitions and time points). Statistical significance was tested between clusters on 1000 randomly selected cells (two-sided Wilcoxon-rank-sum test; , p ≤0.001). c Dynamics of the distribution of cells among the three activation phenotypes in a cell population subjected to nutrient deprivation. Reported values represent the mean among three independent experiments ± SD. Shaded regions highlight different stages in TFEB dynamic response. R1: first activation wave, R2: re-inactivation, R3: second activation wave Single cell correlation of TFEB activity states, LAMP1 concentration and lysosomal positioning in response to nutrient deprivation Single cell correlation of TFEB activity states, LAMP1 concentration and lysosomal positioning in response to nutrient deprivation was re-inactivated, moving 60 % of the cells to the “Inactive” phenotype after 5 hours, and leading to full loss of the “Active” subpopulation (region R2). Finally, between 5 and 15 hours TFEB subpopulation dynamics shifted again towards increasing “Medium” and “Active” phenotypes (region R3). We subsequently sought to gain insight into functional relevance of TFEB activity subpopulations during the re- sponse to nutrient deprivation. Therefore, we simultan- eously monitored responses of endogenous TFEB and TFEB-controlled lysosomal marker LAMP1 [1, 2] by multi- spectral imaging cytometry (Fig. 6a). Nutrient deprivation significantly increased the cellular concentration of LAMP1 for all time points up to 5 hours, peaking at Thus, similar to Torin1 treatment, nutrient deprivation rapidly activated TFEB, increasing its nuclear localization. However, in contrast to Torin1 treatment, the subsequent reduction to TFEB nuclear localization was more rapid, and then reversed. Page 13 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Fig. 6 Single cell correlation of nutrient deprivation-induced TFEB activity and downstream lysosomal response by multispectral imaging cytometry. HeLa cells were kept in culture medium (NT, non-treated) or subjected to nutrient deprivation. At the indicated time points, cells were immunostained for TFEB and LAMP1, and nuclei labelled with Hoechst 33342. Following, cells were analyzed by ISX multispectral imaging cytometry. a Bright field and fluorescence images of representative cells for the indicated treatments. b Time course of the mean population response of LAMP1 protein levels, shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Statistical significance was tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01; , p ≤0.001). c Mean LAMP1 concentration for the three TFEB activation phenotypes. Bars represent the mean among all cells in each cluster ± SD (from all repetitions and time points). Statistical significance was tested between clusters on 1000 randomly selected cells (Student’s two-tailed t-test; *, p ≤0.001; n.s. p > 0.05). d Representative fluorescence images for different ranges of the feature “LAMP1 Max Contour Position” to assess lysosomal positioning. e Time course of the mean population response of “LAMP1 Max Contour Position”, shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Statistical significance was tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01). Single cell correlation of TFEB activity states, LAMP1 concentration and lysosomal positioning in response to nutrient deprivation f Mean “LAMP1 Max Contour Position” for the three TFEB activation phenotypes. Bars represent the mean among all cells in each cluster ± SD (from all repetitions and time points). Statistical significance was tested between clusters on 1000 randomly selected cells (two-sided Wilcoxon-rank-sum test; , p ≤0.001). g Percentage of cells with values of “LAMP1 Max Contour Position” within the ranges specified in (d), presented separately for each TFEB activation phenotype Fig. 6 Single cell correlation of nutrient deprivation-induced TFEB activity and downstream lysosomal response by multispectral imaging cytometry. HeLa cells were kept in culture medium (NT, non-treated) or subjected to nutrient deprivation. At the indicated time points, cells were immunostained for TFEB and LAMP1, and nuclei labelled with Hoechst 33342. Following, cells were analyzed by ISX multispectral imaging cytometry. a Bright field and fluorescence images of representative cells for the indicated treatments. b Time course of the mean population response of LAMP1 protein levels, shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Statistical significance was tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01; , p ≤0.001). c Mean LAMP1 concentration for the three TFEB activation phenotypes. Bars represent the mean among all cells in each cluster ± SD (from all repetitions and time points). Statistical significance was tested between clusters on 1000 randomly selected cells (Student’s two-tailed t-test; *, p ≤0.001; n.s. p > 0.05). d Representative fluorescence images for different ranges of the feature “LAMP1 Max Contour Position” to assess lysosomal positioning. e Time course of the mean population response of “LAMP1 Max Contour Position”, shown relative to time point ‘0’. Reported values represent the mean among three independent experiments ± SD. Statistical significance was tested vs. time point ‘0’, which corresponds to the NT control (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01). f Mean “LAMP1 Max Contour Position” for the three TFEB activation phenotypes. Bars represent the mean among all cells in each cluster ± SD (from all repetitions and time points). Statistical significance was tested between clusters on 1000 randomly selected cells (two-sided Wilcoxon-rank-sum test; , p ≤0.001). Single cell correlation of TFEB activity states, LAMP1 concentration and lysosomal positioning in response to nutrient deprivation g Percentage of cells with values of “LAMP1 Max Contour Position” within the ranges specified in (d), presented separately for each TFEB activation phenotype 3 hours with an increase of 30 % and decreasing thereafter (Fig. 6b). The delay in the peak in LAMP1 concentration relative to TFEB is consistent with TFEB-mediated lysosomal biogenesis. Notably, although at 15 hours TFEB accumulated in the nucleus again (Fig. 5a, c), no increase in TFEB (Fig. 5a) or LAMP1 (Fig. 6b) levels was observed at this time point. Importantly, LAMP1 concentration was significantly higher for the “Active” cluster (~17 % increase relative to “Medium” 3 hours with an increase of 30 % and decreasing thereafter (Fig. 6b). The delay in the peak in LAMP1 concentration relative to TFEB is consistent with TFEB-mediated lysosomal biogenesis. Notably, although at 15 hours TFEB accumulated in the nucleus again (Fig. 5a, c), no increase in TFEB (Fig. 5a) or LAMP1 (Fig. 6b) levels was observed at this time point. Importantly, LAMP1 concentration was significantly higher for the “Active” cluster (~17 % increase relative to “Medium” and “Inactive” clusters), demonstrating enhanced lysosomal content and thus higher TFEB downstream signaling for the phenotype classified as “Active” (Fig. 6c). In addition, we analyzed the subcellular positioning of lysosomes, which is intricately linked to the cellular nutrient state [18]. To that end, we employed the “LAMP1 Max Contour Position” feature, defined as the location of the contour in the cell with the highest LAMP1 intensity concentration [16]. As shown in Page 14 of 19 Page 14 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Fig. 6d and Additional file 8: Figure S8, values close to 0 indicate a centered (perinuclear) lysosomal distribution, while values close to 1 indicate a peripheral (cytoplasmic) lysosomal distribution. The mean population response (Fig. 6e) demonstrates that within 0.5 hours, as expected [18], nutrient deprivation significantly redistributed lyso- somes to the nuclear region. This perinuclear lysosomal distribution was maintained up to 5 hours, and after 15 hours lysosomes re-localized towards the cell periph- ery. Notably, the three clusters displayed statistically dif- ferent mean “LAMP1 Max Contour Position” values (Fig. 6f) and clear differences between lysosomal position- ing distributions (Fig. 6g), indicating that cells in the “In- active” cluster contain the most peripheral lysosomes and cells in the “Active” cluster contain the most perinuclear lysosomes. Discussion Here we investigated the relationship between modula- tions to mTOR activity and the consequent changes to localization and concentration of TFEB in HeLa and MCF7 cells. We report the novel findings that mTOR exerts a rapid, time-limited rheostat control on TFEB subcellular localization and protein levels (see Fig. 8). During the period of 0.5 to 1.5 hours following perturb- ation, mTOR activation decreased TFEB protein levels and increased TFEB cytoplasmic retention. Conversely, in response to mTOR inhibition by either Torin1 or nu- trient deprivation, during this period, TFEB protein levels rapidly increased concurrent to enhanced accumu- lation of TFEB in the nucleus. Surprisingly, this effect was limited in duration to a period of 3 hours, and dur- ing the period of 3 to 15 hours, TFEB concentrations and subcellular distributions returned towards basal levels for all conditions, evidencing homeostatic regula- tory mechanisms dependent and independent of mTOR. While we show that mTOR activation maximally in- hibits TFEB, mTOR inhibition only fractionally activated TFEB. Furthermore, our findings indicate that ERK sig- naling exerts a negligible inhibitory effect on TFEB under nutrient deprivation and Torin1 conditions, and instead indicate a role for proteasome degradation path- ways in the regulation of TFEB subcellular localization (see Fig. 8) independent of TFEB levels. Maximal activation of TFEB under conditions of mTOR inhibition is negatively regulated by the proteasome, independently of TFEB concentration Maximal activation of TFEB under conditions of mTOR inhibition is negatively regulated by the proteasome, independently of TFEB concentration ERK signaling does not impact TFEB activation in response to Torin1 or nutrient deprivation Finally, we sought to identify mechanisms which prohibit maximal TFEB activation. We first explored a possible contribution of ERK kinases, which were previously shown to negatively regulate TFEB [4]. We treated HeLa cells with the MEK1 and MEK2 inhibitor U0126 [4, 9] alone or in co-treatment with Torin1 or nutrient dep- rivation for 1.5 and 3 hours. We found that U0126 did not significantly impact Torin1- or nutrient deprivation- induced increases in the fraction of “Active” cells (Fig. 7a), nor in mean TFEB concentrations (Fig. 7b). Of note, treat- ment with fresh FM containing U0126 (Fig. 7a) led to a lesser deactivation of TFEB than treatment with fresh FM alone (Fig. 4f). Together, these results indicate that ERK signalling does not restrain the early phase of TFEB acti- vation under the conditions tested here. Maximal TFEB activation by Torin1 is negatively regulated by the proteasome As increased TFEB activation correlated with increased TFEB protein levels (see Additional file 1: Figure S1, panel II), we next sought to determine whether the ini- tial changes in TFEB protein levels could be attributed to proteasomal degradation or protein synthesis. There- fore, we treated HeLa cells with the proteasome inhibi- tor epoxomicin (1 μM) or the transcriptional inhibitor Single cell correlation of TFEB activity states, LAMP1 concentration and lysosomal positioning in response to nutrient deprivation Consistent with the inter-cluster differences, “LAMP1 Max Contour Position” negatively correlated with TFEB nuclear/cytoplasmic ratio in single cells, with a correlation coefficient of −0.25 (see Additional file 1: Figure S1, panel III). Overall, LAMP1 concentration dy- namics and spatial distribution within subpopulations indicate that the TFEB-defined subpopulations are of physiological relevance for the cellular lysosomal state. actinomycin D (1 μg/mL), alone or in combination with Torin1 for 1 hour (Fig. 7c). Neither epoxomicin nor acti- nomycin D had a significant effect on TFEB levels for all tested conditions, indicating that, within the period of rapid TFEB concentration changes, proteasomal degrad- ation and transcriptional regulation do not play a signifi- cant role. We thus speculate that TFEB changes might be regulated via lysosomal degradation. However, as im- pairment of lysosomal function activates TFEB [9], an unbiased assessment of the involvement of lysosomal degradation on TFEB levels is not readily possible and requires future investigation. Of note, while epoxomicin had no impact on the nuclear/cytoplasmic ratio under control conditions, nu- clear TFEB localization was significantly increased under epoxomicin and Torin1 co-treatment (Fig. 7d). Subpopu- lation analysis demonstrates that this effect was due to a significant increase in the subpopulation with “Active” TFEB (to 96 %) (Fig. 7d), and was not related to changes in total TFEB levels (Fig. 7e). Subpopulation analysis reveals accurate TFEB signaling behavior We achieved highly accurate analysis of spatial and temporal TFEB dynamics by subpopulation analysis. To decompose cellular heterogeneity into discrete Marin Zapata et al. BMC Cancer (2016) 16:355 Page 15 of 19 Fig. 7 Effect of ERK, proteasome and transcriptional inhibition on mTOR inhibition-mediated TFEB activation. a HeLa cells were treated with fresh FM supplemented with U0126 (10 μM) alone or in co-treatment with Torin1 (2 μM), or subjected to nutrient deprivation (ND) alone or in combination with U0126. Following immunostaining for TFEB and labeling of nuclei with Hoechst 33342 cells were analyzed by multispectral imaging cytometry. For visualization, the data for FM supplemented with U0126 is included both for Torin1 (left) and for ND (right) treatments. Bars report mean subpopulation distributions among three independent experiments ± SD. Statistical significance was tested based on the “Active” subpopulation vs. non-treated control (NT), unless otherwise indicated by horizontal lines (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01; , p ≤0.001; n.s., p > 0.05). b Mean population response of TFEB concentration in HeLa cells treated as in (a), shown relative to the non-treated levels (NT). Unless indicated by horizontal lines, statistical significance was tested vs. non-treated control, NT (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01; , p ≤0.001; n.s., p > 0.05). c Mean population response of TFEB concentration (relative to the non-treated levels, NT) in HeLa cells treated with Epox (1 μM), ActD (1 μg/mL) or vehicle control (DMSO) alone or in combination with Torin1 (2 μM). Bars report the mean among four independent experiments ± SD. Statistical significances were tested with Student’s two-tailed t-test (, p ≤0.05; , p ≤0.01; , p ≤0.001; n.s., non-significant). d Mean population response of TFEB subcellular localization and subpopulation distributions for HeLa cells treated as in (c). Bars report the mean among four independent experiments ± SD. Statistical significances were tested with Student’s two-tailed t-test (, p ≤0.01; *, p ≤0.001; n.s., non-significant). Subpopulations were compared based on the “Active” phenotype. e Mean TFEB concentration for the three activation phenotypes in a subset of cells treated with Torin1 alone or in co-treatment with epoxomicin, as indicated in (c). Bars represent the mean among all cells in each cluster for the indicated treatments ± SD Fig. 7 Effect of ERK, proteasome and transcriptional inhibition on mTOR inhibition-mediated TFEB activation. mTOR activity modulations induce distinct time-evolving TFEB subpopulation redistributions Quantifying TFEB subpopulation redistributions over time, we found that with the addition of fresh FM, which mildly increased mTOR activity (Fig. 1d), TFEB hetero- geneity was rapidly lost in HeLa cells (Fig. 4f). In contrast, consistent but smaller population shifts were measured in MCF7 cells (Fig. 4g), suggesting a cell line-dependent lower sensitivity to mTOR activation, possibly related to altered nuclear import of TFEB [11]. Notably, subpopula- tion analysis revealed that at 15 hours post FM treatment, a small percentage of HeLa cells switched to the less active “Medium” phenotype. This subpopulation shift may reflect the consumption of nutrients and/or growth fac- tors, as indicated by decreased 4E-BP1 phosphorylation (Fig. 2c). Fig. 8 Proposed dynamic regulatory network of TFEB. TFEB is under long-term (dark blue area) and short-term (light blue area) control, regulating the strength and time span of its activation, i.e., nuclear localization. The long-term control limits the duration of TFEB activation upon sustained mTOR inhibition, displaying prominent mTOR-independent negative regulation of TFEB. The short-term control acts as a rheostat, which is highly sensitive to inactivation by mTOR, but tightly controls TFEB activation by multiple mechanisms, directly or indirectly regulated by mTOR. Under conditions of sustained mTOR inhibition (Fig. 2d, f), starting at 3 hours, TFEB is gradually inactivated (Fig. 4f, region R2), evidencing the action of other inhibitors with slower kinetics than mTOR. In contrast to Torin1 mTOR inhibition, nutrient deprivation displays a fast reactivation kinetics (Fig. 5c, region R2) followed by a second activation wave (Fig. 5c, region R3), suggesting that TFEB rheostat is sensitive to autophagy feedback on mTOR activity via nutrient recycling. Proteasome inhibition by epoxomicin enhances TFEB activation (Fig. 7d), suggesting that the proteasome mediates the degradation of a “positive regulator” of TFEB, labeled here with a question mark. The effect of epoxomicin on TFEB activation is only detectable under conditions of mTOR inhibition (Fig. 7d), suggesting that the “positive regulator” is not degraded under conditions of high mTOR activity (i.e., mTOR activity inhibits the degradation of the “positive regulator”) In contrast to FM, in both HeLa and MCF7 cells, Torin1 rapidly depleted the “Inactive” subpopulation and increased “Active” and “Medium” subpopulations (Fig. 4f, g). Strikingly, while Torin1 inhibition of mTOR was maintained for 15 hours (Fig. mTOR activity modulations induce distinct time-evolving TFEB subpopulation redistributions 2f), TFEB re-inactiva- tion began 1.5 hours following Torin1 treatment and re- quired 15 hours to inactive approximately 50 % of the population, compared to 100 % inactivation by fresh FM at 0.5 hours. These findings evidence an additional mTOR-independent, negative regulatory mechanism, en- gaged similarly in HeLa and MCF7 cells under conditions of prolonged mTOR inhibition. Putative mechanisms may involve phosphorylation by GSK3 [12], multiple serine phosphorylations [4, 22], and/or other post-translational modifications such as SUMOylation [23]. y Notably, nutrient deprivation-mediated activation of TFEB differed from Torin1 treatment. While Torin1 strongly induced the “Active” subpopulation and depleted the “Inactive” subpopulation (Fig. 4f), nutrient deprivation mostly increased the “Medium” subpopulation without depleting the “Inactive” subpopulation (Fig. 5c). Further, distinct from Torin1 treatment, between 1.5 and 5 hours (R2 period) TFEB was rapidly inactivated, and between 5 and 15 hours (R3 period) both “Active” and “Medium” subpopulations increased, suggesting a second activation wave. subpopulations of TFEB activation states we used ag- glomerative clustering [19–21] of single-cell imaging datasets, and established criteria to optimize the number of phenotypes and input features. We identified “Active”, “Medium” and “Inactive” TFEB subpopulations which significantly and reproducibly correlated with TFEB signaling. Consistent with TFEB activation of lysosomal biogenesis [4], the “Active” subpopulation reported higher lysosomal content than “Medium” and “Inactive” subpop- ulations (Fig. 6c). Moreover, “Active”, “Medium” and “In- active” subpopulations reported significant differences in lysosomal positioning (Fig. 6f, g), an indicator of the cellu- lar metabolic state. In response to nutrient deprivation, the “Active” subpopulation was enriched with perinuclear lysosomes, which report a starvation response, while the subpopulations of TFEB activation states we used ag- glomerative clustering [19–21] of single-cell imaging datasets, and established criteria to optimize the number of phenotypes and input features. We identified “Active”, “Medium” and “Inactive” TFEB subpopulations which significantly and reproducibly correlated with TFEB signaling. Consistent with TFEB activation of lysosomal biogenesis [4], the “Active” subpopulation reported higher lysosomal content than “Medium” and “Inactive” subpop- ulations (Fig. 6c). Moreover, “Active”, “Medium” and “In- active” subpopulations reported significant differences in lysosomal positioning (Fig. 6f, g), an indicator of the cellu- lar metabolic state. In response to nutrient deprivation, the “Active” subpopulation was enriched with perinuclear lysosomes, which report a starvation response, while the We suggest this alternating behavior was due to nega- tive feedback of autophagy on mTOR, as we previously predicted by agent-based modeling [24]. Subpopulation analysis reveals accurate TFEB signaling behavior The short-term control acts as a rheostat, which is highly sensitive to inactivation by mTOR, but tightly controls TFEB activation by multiple mechanisms, directly or indirectly regulated by mTOR. Under conditions of sustained mTOR inhibition (Fig. 2d, f), starting at 3 hours, TFEB is gradually inactivated (Fig. 4f, region R2), evidencing the action of other inhibitors with slower kinetics than mTOR. In contrast to Torin1 mTOR inhibition, nutrient deprivation displays a fast reactivation kinetics (Fig. 5c, region R2) followed by a second activation wave (Fig. 5c, region R3), suggesting that TFEB rheostat is sensitive to autophagy feedback on mTOR activity via nutrient recycling. Proteasome inhibition by epoxomicin enhances TFEB activation (Fig. 7d), suggesting that the proteasome mediates the degradation of a “positive regulator” of TFEB, labeled here with a question mark. The effect of epoxomicin on TFEB activation is only detectable under conditions of mTOR inhibition (Fig. 7d), suggesting that the “positive regulator” is not degraded under conditions of high mTOR activity (i.e., mTOR activity inhibits the degradation of the “positive regulator”) “Inactive” was enriched with mixed and peripheral lyso- somes, which indicate normal growth conditions [18]. We therefore conclude that subpopulation analysis of TFEB distribution and concentration reveals the time course response of TFEB to metabolic stress. Subpopulation analysis reveals accurate TFEB signaling behavior a HeLa cells were treated with fresh FM supplemented with U0126 (10 μM) alone or in co-treatment with Torin1 (2 μM), or subjected to nutrient deprivation (ND) alone or in combination with U0126. Following immunostaining for TFEB and labeling of nuclei with Hoechst 33342 cells were analyzed by multispectral imaging cytometry. For visualization, the data for FM supplemented with U0126 is included both for Torin1 (left) and for ND (right) treatments. Bars report mean subpopulation distributions among three independent experiments ± SD. Statistical significance was tested based on the “Active” subpopulation vs. non-treated control (NT), unless otherwise indicated by horizontal lines (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01; , p ≤0.001; n.s., p > 0.05). b Mean population response of TFEB concentration in HeLa cells treated as in (a), shown relative to the non-treated levels (NT). Unless indicated by horizontal lines, statistical significance was tested vs. non-treated control, NT (Student’s two-tailed t-test; , p ≤0.05; , p ≤0.01; , p ≤0.001; n.s., p > 0.05). c Mean population response of TFEB concentration (relative to the non-treated levels, NT) in HeLa cells treated with Epox (1 μM), ActD (1 μg/mL) or vehicle control (DMSO) alone or in combination with Torin1 (2 μM). Bars report the mean among four independent experiments ± SD. Statistical significances were tested with Student’s two-tailed t-test (, p ≤0.05; , p ≤0.01; , p ≤0.001; n.s., non-significant). d Mean population response of TFEB subcellular localization and subpopulation distributions for HeLa cells treated as in (c). Bars report the mean among four independent experiments ± SD. Statistical significances were tested with Student’s two-tailed t-test (, p ≤0.01; , p ≤0.001; n.s., non-significant). Subpopulations were compared based on the “Active” phenotype. e Mean TFEB concentration for the three activation phenotypes in a subset of cells treated with Torin1 alone or in co-treatment with epoxomicin, as indicated in (c). Bars represent the mean among all cells in each cluster for the indicated treatments ± SD Page 16 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Fig. 8 Proposed dynamic regulatory network of TFEB. TFEB is under long-term (dark blue area) and short-term (light blue area) control, regulating the strength and time span of its activation, i.e., nuclear localization. The long-term control limits the duration of TFEB activation upon sustained mTOR inhibition, displaying prominent mTOR-independent negative regulation of TFEB. During pharmacological mTOR inhibition the proteasome negatively regulates nuclear localization of TFEB g y g Notably, we show that under Torin1 treatment and during early phases of nutrient deprivation TFEB protein levels correlate with subcellular localization. This is con- sistent with increased transcription of autophagy and lysosomal genes following TFEB overexpression [1, 2, 6]. Using specific inhibitors, we ruled out that initial (within 1 hour) increases and decreases in TFEB levels were due to protein translation or proteasomal activities (Fig. 7c), indicating that transcriptional feedback [25] is not rele- vant during the first hour of treatment. These results further suggest that TFEB may be targeted for lysosomal degradation, thereby forming a negative feedback circuit. However, lysosomal stress potently activates TFEB [9] and therefore putative lysosomal targeting of TFEB re- quires further investigation. Additional file 2: Figure S2. Additional file 2: Figure S2. Comparison of imaging cytometry (ISX) and wild-field microscopy (WF) approaches to measure endogenous TFEB subcellular distribution. Representative WF and ISX immunofluorescence images of endogenous TFEB in HeLa cells treated with fresh full medium (FM, 0.5 hours), Torin1 (1 hour) or left untreated (NT). Graphs represent normalized intensity profiles for Hoechst (purple) and TFEB (orange) along the indicated red line. Both measurement approaches show similar intensity profiles for each experimental condition. (JPG 2766 kb) Additional file 3: Figure S3. Representative imaging cytometry images of endogenous TFEB under different treatments with indicated nuclear and cellular contours. (a) Non- treated: Most cells display slightly higher cytosolic concentration, and some cells display similar concentrations in nuclear and cytosolic compartments. (b) Fresh full medium: Cells display higher cytosolic than nuclear concentration. (c) Torin1: Cells display higher nuclear than cytosolic concentration. (JPG 4456 kb) Additional file 4: Figure S4. Positive clustering example. To demonstrate the reproducibility of the clustering outcome, we separately present the curves from three independent experiments (rows 1 to 3) and the combined mean response ± SEM (row 4), which corresponds to the subpopulation dynamics presented in Fig. 4f. The result was obtained using three clusters and the feature “Mean Pixel Nuc/Cyto” as input. Importantly, treatments with FM and Torin1 induced a clear redistribution of the cell population among the different phenotypes (clusters). This distribution was consistent among the three repetitions and displayed independent dynamics for each cluster, thus adhering to our first and second evaluation criteria, respectively. (JPG 1499 kb) Additional file 4: Figure S4. Positive clustering example. To demonstrate the reproducibility of the clustering outcome, we separately present the curves from three independent experiments (rows 1 to 3) and the combined mean response ± SEM (row 4), which corresponds to the subpopulation dynamics presented in Fig. 4f. The result was obtained using three clusters and the feature “Mean Pixel Nuc/Cyto” as input. Importantly, treatments with FM and Torin1 induced a clear redistribution of the cell population among the different phenotypes (clusters). This distribution was consistent among the three repetitions and displayed independent dynamics for each cluster, thus adhering to our first and second evaluation criteria, respectively. (JPG 1499 kb) Additional files Additional file 1: Figure S1. Work flow for classification of cell subpopulations. Initially, cells subjected to FM or Torin1 treatments are classified into three groups/clusters (denoted as activation phenotypes) using agglomerative clustering on the base-clustering-feature “Mean Pixel Nuc/Cyto” (See Fig. 4a). The resulting classification criteria, consisting of thresholds on the base-clustering-feature, constitute our data-based model for cell classification. This model is estimated separately for HeLa and MCF7 cells, and thus, is cell line specific. The activation phenotypes are further characterized by identifying additional phenotypic differences between the cell groups. To this end, the three clusters are statistically compared based on a set of features which were not used in the generation of the cell classification model. Besides identifying features which are specific to each activation phenotype, significant differences between the cell groups report correlations between the evaluated features and the base clustering feature (exemplified by the correlation coefficients and correlation plots on the top right corners of the grey panels). Finally, the FM and Torin1 data-based model is used as a basis for cell classification in response to other treatments such as nutrient deprivation, and inhibition of ERK, proteasome or protein translation (model extrapolation). (JPG 7051 kb) Importantly, we found that mTOR inactivation failed to maximally activate TFEB in all cells of a cell population (Fig. 4f, g). Our findings exclude ERK signaling as a limiting factor (Fig. 7a). Instead, we identified a novel TFEB inhibi- tory role for the proteasome, under conditions of mTOR in- hibition. Selectively in Torin1-treated cells, epoxomicin- mediated inhibition of the proteasome leads to “Active” nu- clear localization of TFEB in nearly 100 % of cells (Fig. 7d), without altering TFEB protein levels (Fig. 7c, e). The mech- anism by which the proteasome negatively regulates TFEB remains to be determined. We propose an indirect pathway, in which mTOR activity inhibits proteasomal degradation of a positive TFEB regulator. Thereby, maximal activation of TFEB nuclear translocation requires inhibition of mTOR and proteasome activities (see Fig. 8). Alternatively, prote- asome inhibition might activate TFEB via induction of pro- teotoxic stress [26], which might be exacerbated under conditions of decreased mTOR activity [27]. mTOR activity modulations induce distinct time-evolving TFEB subpopulation redistributions Here we report that lysosomal content and perinuclear clustering were maximal during the first 5 hours of nutrient deprivation (Fig. 6b, e), indicative of maximal autophagy activation, Page 17 of 19 Page 17 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 which could provide recycled nutrients sufficient to re- activate mTOR and thereby inactivate TFEB. Subse- quently, after 5 hours, depletion of autophagy recycled nutrients could lead to renewed inactivation of mTOR, which is in agreement with the observed re-activation of TFEB at 15 hours of nutrient deprivation. Interestingly, at this time point TFEB nuclear localization is not paral- leled with increases in either TFEB or LAMP1 protein levels. Moreover, also at 15 hours, lysosomes localized at the cell periphery (Fig. 6e). These observations might be due to depletion of intracellular substrates following prolonged nutrient deprivation. Future studies will explore whether peripheral lysosomal redistribution reflects a cel- lular shift toward exploiting endocytic nutrient sources. control. Our findings suggest that the rapid rheostatic re- sponse, mediated by mTOR, allows the cell to quickly adapt to metabolic changes, while the long-term, mTOR- independent homeostatic response controls the magni- tude and duration of TFEB activation, and presumably limits excessive autophagy. Our findings also suggest that TFEB may be targeted by lysosomes, and that under con- ditions of mTOR inhibition the proteasome regulates the early response of TFEB localization, uncoupled from changes to TFEB levels. As TFEB is a central player in cancer [11, 12], our approach to time series analyses of magnitude and dynamics of subpopulation shifts enables biomarker assessment of cell line sensitivity and respon- siveness. We propose our approach as a useful general framework for identifying and quantifying information contained within heterogeneous imaging datasets. During pharmacological mTOR inhibition the proteasome negatively regulates nuclear localization of TFEB Abbreviations ActD, actinomycin D; Epox, epoxomicin; FM, fresh, fully-supplemented medium; ISX, ImageStreamX imaging flow cytometer; LAMP1, lysosomal- associated membrane protein 1; mTOR, mammalian target of rapamycin; ND, nutrient deprivation; NT, non-treated; SPADE, spanning-tree progression analysis of density-normalized events; TFEB, transcription factor EB; WF, wide field imaging 10. Levine B, Kroemer G. Autophagy in the pathogenesis of disease. Cell. 2008;132:27–42. doi:10.1016/j.cell.2007.12.018. 11. Perera RM et al. Transcriptional control of autophagy-lysosome function drives pancreatic cancer metabolism. Nature. 2015;524:361–5. doi:10.1038/ nature14587. 12. Marchand B, Arsenault D, Raymond-Fleury A, Boisvert FM, Boucher MJ. Glycogen synthase kinase-3 (GSK3) inhibition induces prosurvival autophagic signals in human pancreatic cancer cells. J Biol Chem. 2015;290:5592–605. doi:10.1074/jbc.M114.616714. Acknowledgements This work was supported by the German Cancer Research Center (DKFZ), through SBCancer within the Helmholtz Alliance on Systems Biology funded by the Initiative and Networking Fund of the Helmholtz Association (NRB); and the e:Bio grant #0316191 (LysoSys) of the Federal Ministry of Education and Research (BMBF), Germany (AH-B). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. 13. Thoreen CC, et al. An ATP-competitive mammalian target of rapamycin inhibitor reveals rapamycin-resistant functions of mTORC1. J Biol Chem. 2009;284:8023–32. doi:10.1074/jbc.M900301200. 14. Kim D, et al. MTOR interacts with Raptor to form a nutrient-sensitive complex that signals to the cell growth machinery. Cell. 2002;110:163–75. doi:10.1016/S0092-8674(02)00808-5. 15. Qiu P, et al. Extracting a cellular hierarchy from high-dimensional cytometry data with SPADE. Nat Biotechnol. 2011;29:886–91. doi:10.1038/Nbt.1991. Data sets will be shared upon request. 16. Amnis. Image Data Exploration and Analysis Software (IDEAS) User’s Manual 4.0. 2010. Received: 17 December 2015 Accepted: 26 May 2016 Received: 17 December 2015 Accepted: 26 May 2016 References (a) Schematic representation of the feature “MAX Contour Position” used to quantify lysosomal positioning. (b) Representative LAMP1 immunofluorescence images for different ranges of the feature Additional file 8: Figure S8. Extended representative data set of lysosomal positioning. (a) Schematic representation of the feature “MAX Contour Position” used to quantify lysosomal positioning. (b) Representative LAMP i fl i f diff f h f Additional file 8: Figure S8. Extended representative data set of lysosomal positioning. (a) Schematic representation of the feature “MAX Contour Position” used to quantify lysosomal positioning. (b) Representative LAMP1 immunofluorescence images for different ranges of the feature “LAMP1 MAX Contour Position” in HeLa cells treated as in Fig. 6. (JPG 3911 kb) 7. Roczniak-Ferguson A, et al. The transcription factor TFEB links mTORC1 signaling to transcriptional control of lysosome homeostasis. Sci Signal. 2012;5:ra42. doi:10.1126/scisignal.2002790. LAMP1 immunofluorescence images for different ranges of the feature “LAMP1 MAX Contour Position” in HeLa cells treated as in Fig. 6. (JPG 3911 kb) LAMP1 immunofluorescence images for different ranges of the feature “LAMP1 MAX Contour Position” in HeLa cells treated as in Fig. 6. (JPG 3911 kb) 8. Zoncu R, et al. mTORC1 Senses Lysosomal Amino Acids Through an Inside-Out Mechanism That Requires the Vacuolar H+−ATPase. Science. 2011;334:678–83. doi:10.1126/science.1207056. 9. Settembre C, et al. A lysosome-to-nucleus signalling mechanism senses and regulates the lysosome via mTOR and TFEB. Embo J. 2012;31:1095–108. doi:10.1038/emboj.2012.32. 9. Settembre C, et al. A lysosome-to-nucleus signalling mechanism senses and regulates the lysosome via mTOR and TFEB. Embo J. 2012;31:1095–108. doi:10.1038/emboj.2012.32. Author details 1 1Lysosomal Systems Biology, German Cancer Research Center (DKFZ) and BioQuant, University of Heidelberg, Heidelberg, Germany. 2Systems Biology of Cell Death Mechanisms, German Cancer Research Center (DKFZ) and BioQuant, University of Heidelberg, Heidelberg, Germany. 3Department of Surgery, Heidelberg University Hospital, Heidelberg, Germany. 4W. Harry Feinstone Department of Molecular Microbiology & Immunology, Johns Hopkins University Bloomberg School of Public Health, 615 N. Wolfe St., Baltimore, MD 21205, USA. Additional file 6: Figure S6. Nuclear localization of TFEB is influenced by total levels of cellular TFEB. (a) HeLa cells were subjected or not to transfections with the indicated constructs and treated 24 hours post transfection with 2 μM Torin1 for 3 hours or left non-treated (NT). Representative images demonstrate the subcellular distribution of TFEB fluorescence for endogenous TFEB (TFEB immunofluorescence, IF), or transiently overexpressed GFP-TFEB, expressed alone or coexpressed with RFP-tagged 14-3-3 protein isoform YWHAG. The look-up-table ‘Fire’ (ImageJ) was applied to grey scale images of TFEB or GFP-TFEB fluorescence, representing ranging from high (white) to low (dark purple) intensity values, as displayed in color scale bar. Scale bars (white line), 20 μm. (b) Quantification of the number of cells with mainly nuclear TFEB fluorescence. At least 30 cells were scored per condition and experiment in three independent experiments. Statistical significance was tested using two-tailed Student’s t-test (, p ≤0.01; ***, p ≤0.001). (JPG 1702 kb) Conclusions Overall, we demonstrate that TFEB levels and subcellular distribution undergo distinct short-term and long-term Page 18 of 19 Page 18 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 Competing interests Additional file 5: Figure S5. Negative clustering examples. (a) Example of a clustering outcome dissatisfying criterion 1, i.e., reproducibility of the dynamic distribution of cells among clusters. The result was obtained using three clusters with the following input features: area cell, concentration cell, and “Mean Pixel Nuc/Cyto”. In this case, the evolution in time of the percentage of cells in clusters 1 (blue) and 2 (black) is not reproducible. (b) Example of a clustering outcome dissatisfying criterion 2, i.e., non-redundant dynamics. The result was obtained using four clusters with the input feature “Mean Pixel Nuc/Cyto”. In this case, clusters 1 (black) and 3 (blue) follow similar dynamic responses to all treatments, indicating that the two clusters are redundant. (JPG 2839 kb) p g The authors declare that they have no competing interests. Availability of data and materials Data sets will be shared upon request. Consent for publication Not applicable. Consent for publication Not applicable. References 1. Palmieri M, et al. Characterization of the CLEAR network reveals an integrated control of cellular clearance pathways. Hum Mol Genet. 2011;20:3852–66. doi:10.1093/Hmg/Ddr306. 2. Sardiello M, et al. A gene network regulating lysosomal biogenesis and function. Science. 2009;325:473–7. doi:10.1126/science.1174447. 2. Sardiello M, et al. A gene network regulating lysosomal biogenesis and function. Science. 2009;325:473–7. doi:10.1126/science.1174447. 2. Sardiello M, et al. A gene network regulating lysosomal biogenesis and function. Science. 2009;325:473–7. doi:10.1126/science.1174447. 3. Hamacher-Brady A. Autophagy regulation and integration with cell signaling. Antioxid Redox Signal. 2012;17:756–65. doi:10.1089/ars.2011.4410. 4. Settembre C, et al. TFEB Links Autophagy to Lysosomal Biogenesis. Science. 2011;332:1429 33 doi:10 1126/science 1204592 Additional file 7: Figure S7. Mean population response of nuclear/ cytoplasmic ratio predicted from subpopulation distributions. Dots and error bars: mean population response ± SD found experimentally, as presented in Fig. 3b, d. Dotted lines: mean population response predicted based on subpopulation distributions. The predicted mean population response (F) was estimated as follows F = ( αIFI + αMFM + αAFA )/100, where FI, FM and FA are the mean “Mean Pixel Nuc/Cyto” of the “Inactive”, “Medium” and “Active” clusters, respectively, and αI, αM and αA are the percentage of cells in each cluster. FI, FM and FA are constants (presented in Fig. 4b, d), while and αI, αM and αA are functions of time given by the curves in Fig. 4f, g. (JPG 657 kb) function. Science. 2009;325:473–7. doi:10.1126/science.1174447. 3. Hamacher-Brady A. Autophagy regulation and integration with cell signaling. Antioxid Redox Signal. 2012;17:756–65. doi:10.1089/ars.2011.4410. b l k h l ; 3. Hamacher-Brady A. Autophagy regulation and integration with cell signaling. Antioxid Redox Signal. 2012;17:756–65. doi:10.1089/ars.2011.4410. 4. Settembre C, et al. TFEB Links Autophagy to Lysosomal Biogenesis. Science 2011;332:1429–33. doi:10.1126/science.1204592. 5. Martina JA, Puertollano R. Rag GTPases mediate amino acid-dependent recruitment of TFEB and MITF to lysosomes. J Cell Biol. 2013;200:475–91. doi:10.1083/jcb.201209135. 6. Martina JA, Chen Y, Gucek M, Puertollano R. MTORC1 functions as a transcriptional regulator of autophagy by preventing nuclear transport of TFEB. Autophagy. 2012;8:903–14. doi:10.4161/Auto.19653. Additional file 8: Figure S8. Extended representative data set of lysosomal positioning. (a) Schematic representation of the feature “MAX Contour Position” used to quantify lysosomal positioning. (b) Representative LAMP1 immunofluorescence images for different ranges of the feature “LAMP1 MAX Contour Position” in HeLa cells treated as in Fig. 6. (JPG 3911 kb) Additional file 8: Figure S8. Extended representative data set of lysosomal positioning. 27. Chou SD, Prince T, Gong JL, Calderwood SK. mTOR is essential for the proteotoxic stress response, HSF1 activation and heat shock protein synthesis. PLoS One. 2012;7:e39679. doi:10.1371/journal.pone.0039679. Marin Zapata et al. BMC Cancer (2016) 16:355 Authors’ contributions 17. Basiji DA, Ortyn WE, Liang L, Venkatachalam V, Morrissey P. Cellular image analysis and imaging by flow cytometry. Clin Lab Med. 2007;27:653–70. doi:10.1016/j.cll.2007.05.008. NRB and AH-B conceived the study. PAMZ performed data analysis and prepared figures. CJB, AJ and PAMZ performed experiments, data analysis and prepared figures. GD supported analysis. PAMZ, NRB and AH-B wrote the manuscript. All authors edited and approved the manuscript. 18. Korolchuk VI, et al. Lysosomal positioning coordinates cellular nutrient responses. Nat Cell Biol. 2011;13:453–60. doi:10.1038/ncb2204. 18. Korolchuk VI, et al. Lysosomal positioning coordinates cellular nutrient responses. Nat Cell Biol. 2011;13:453–60. doi:10.1038/ncb2204. Page 19 of 19 Marin Zapata et al. BMC Cancer (2016) 16:355 19. Bendall SC, et al. Single-cell mass cytometry of differential immune and drug responses across a human hematopoietic continuum. Science. 2011;332:687–96. doi:10.1126/science.1198704. 20. Gibbs Jr KD, et al. Decoupling of tumor-initiating activity from stable immunophenotype in HoxA9-Meis1-driven AML. Cell Stem Cell. 2012;10:210–7. doi:10.1016/j.stem.2012.01.004. 21. Guo G, et al. Mapping cellular hierarchy by single-cell analysis of the cell surface repertoire. Cell Stem Cell. 2013;13:492–505. doi:10.1016/j.stem. 2013.07.017. 21. Guo G, et al. Mapping cellular hierarchy by single-cell analysis of the cell surface repertoire. Cell Stem Cell. 2013;13:492–505. doi:10.1016/j.stem. 2013.07.017. 22. Humphrey SJ, Azimifar SB, Mann M. High-throughput phosphoproteomics reveals in vivo insulin signaling dynamics. Nat Biotechnol. 2015. doi:10.1038/ nbt.3327. 22. Humphrey SJ, Azimifar SB, Mann M. High-throughput phosphoproteomics reveals in vivo insulin signaling dynamics. Nat Biotechnol. 2015. doi:10.1038/ nbt.3327. 23. Miller AJ, Levy C, Davis IJ, Razin E, Fisher DE. Sumoylation of MITF and its related family members TFE3 and TFEB. J Biol Chem. 2005;280:146–55. doi:10.1074/jbc.M411757200. 24. Borlin CS, Lang V, Hamacher-Brady A, Brady NR. Agent-based modeling of autophagy reveals emergent regulatory behavior of spatio-temporal autophagy dynamics. Cell Commun Signal. 2014;12:56. doi:10.1186/s12964- 014-0056-8. 25. Settembre C, et al. TFEB controls cellular lipid metabolism through a starvation-induced autoregulatory loop. Nat Cell Biol. 2013;15:647–58. doi:10.1038/Ncb2718. 26. Santaguida S, Vasile E, White E, Amon A. Aneuploidy-induced cellular stresses limit autophagic degradation. Gene Dev. 2015;29:2010–21. doi:10.1101/gad. 269118.115. 27. Chou SD, Prince T, Gong JL, Calderwood SK. mTOR is essential for the proteotoxic stress response, HSF1 activation and heat shock protein synthesis. PLoS One. 2012;7:e39679. doi:10.1371/journal.pone.0039679. Authors’ contributions • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step:
https://openalex.org/W2754514951	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0184827&type=printable	English	null	Quantifiable effects of regular exercise on zinc status in a healthy population—A systematic review	PloS one	2,017	cc-by	6,653	Anna Chu1, Trishala Varma1, Peter Petocz2, Samir Samman1,3* 1 Department of Human Nutrition, University of Otago, Dunedin, New Zealand, 2 Department of Statistics, Macquarie University, Sydney, New South Wales, Australia, 3 School of Life and Environmental Sciences, University of Sydney, Sydney, New South Wales, Australia * samir.samman@otago.ac.nz Editor: Jacobus P. van Wouwe, TNO, NETHERLANDS Received: May 8, 2017 Accepted: August 31, 2017 Published: September 20, 2017 Copyright: © 2017 Chu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Abstract a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Zinc is an essential mineral of which its functions have potential implications on exercise performance and beneficial adaptations of physical activity. While the effects of aerobic exercise on zinc metabolism acutely have been well described, the effect of long-term exer- cise training on zinc status remains unclear. The present review aims to determine the effects of exercise training on markers of zinc status in an apparently healthy adult popula- tion. We conducted a systematic literature search on PubMed, Scopus, SPORTDiscus and Cochrane Library from inception to 28 January 2016 to identify interventional or cohort stud- ies that investigated the effects of exercise training on indices of zinc status. Pairwise com- parisons of mean differences in within-group change were calculated and summarised visually in forest plots. Six studies satisfied the inclusion criteria for the systematic review, of which 5 studies included data on changes in serum zinc concentrations and 3 studies pro- vided changes in dietary zinc intake. Two comparisons showed significantly higher increase of serum zinc concentrations in the exercise group compared to control, while one compari- son reported significantly lower change in serum zinc for the exercising group. The exercise groups consumed significantly higher dietary zinc compared to controls in two comparisons. The present review revealed an incomplete evidence base in evaluating the effect of long- term exercise training on markers of zinc status. Further well-designed investigations are required to elucidate the relationship for establishment of dietary recommendation in popu- lations who are continuing exercise interventions. OPEN ACCESS Citation: Chu A, Varma T, Petocz P, Samman S (2017) Quantifiable effects of regular exercise on zinc status in a healthy population—A systematic review. PLoS ONE 12(9): e0184827. https://doi. org/10.1371/journal.pone.0184827 Citation: Chu A, Varma T, Petocz P, Samman S (2017) Quantifiable effects of regular exercise on zinc status in a healthy population—A systematic review. PLoS ONE 12(9): e0184827. https://doi. org/10.1371/journal.pone.0184827 RESEARCH ARTICLE a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * samir.samman@otago.ac.nz Quantifiable effects of regular exercise on zinc status in a healthy population—A systematic review Anna Chu1, Trishala Varma1, Peter Petocz2, Samir Samman1,3* Effects of regular exercise on zinc status pathways and corresponding downstream effects on immune function [4,5] and redox homeo- stasis [6,7], with potential implications for performance [8] and related metabolic benefits of exercise [9]. Exercise and physical activity have been the cornerstone of lifestyle recommendations for the healthy population and those with chronic diseases [10,11]. The public health recommen- dations for physical activity are based on the benefits of exercise on metabolic, musculoskeletal and neuromotor health [12]. Metabolic adaptations of exercise can be modulated by nutri- tional status, such as the availability of macronutrients, specifically protein and carbohydrates [13]. The role of adequate micronutrient status in supporting the beneficial adaptations of exercise has gained research attention [14], particularly with respect to the effects of physical activity on the zinc status and subsequent consequences on exercise performance and meta- bolic effects [15]. A number of cross-sectional studies have investigated the association between exercise/ physical activity and zinc status, in particular the zinc status in athletic and control popula- tions. While some studies showed lower serum zinc concentration in athletes [16,17], other papers report no significant differences in zinc status between athletes and controls [18,19]. The lack of conformity in the results may be driven by factors other than physical activity lev- els, for example differences in dietary habits between the populations. The current evidence is inconclusive in determining the relationship between exercise and zinc status in cross-sec- tional data, however the examination of longitudinal and cohort studies may be able to eluci- date the potential relationship. We have previously reported significant acute fluctuations in serum zinc concentrations as a result of a single bout of aerobic exercise [20,21]. The changes in zinc metabolism were pro- posed to be influenced by the events that occur during exercise and recovery, including leakage of zinc ions from damaged myocytes and exercise-induced inflammatory processes that follow aerobic exercise. Preliminary evidence suggests that training status is a modulating factor for the acute effects of exercise on zinc metabolism [20], but it is currently unclear whether exer- cise training itself imposes adaptations of zinc status and/or metabolism during exercise. Aero- bic exercise has been shown to significantly increase systemic zinc concentration immediately following the exercise bout with decline of zinc concentration to below the baseline values in the hours after aerobic exercise; it is uncertain whether the acute exercise-induced changes in zinc metabolism are sustained in the long term. Therefore, the current review aims to deter- mine the long term effects of exercise training on markers of zinc status in an apparently healthy adult population, as identified by interventional or cohort trials. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Introduction Zinc is an essential trace element with numerous metabolic functions [1]. While zinc is found ubiquitously in the human body, a significant portion of zinc is located within the musculo- skeletal system [2]. In the context of exercise, zinc provides structural integrity and supports catalytic functions for metalloenzymes, such as carbonic anhydrase, superoxide dismutase (SOD) and lactate dehydrogenase [3]. Furthermore, zinc regulates intracellular signalling Funding: The authors received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 1 / 13 Search strategy A literature search of PubMed, Scopus, SPORTDiscus and the Cochrane Library electronic databases were conducted from inception to the 28th of January 2016. The search terms included: zinc, exercis, athlet, physical activity, train. Related terms and MeSH terms were used where appropriate. Searches were limited to human subjects and the English language. Fig 1 shows the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) flowchart describing the electronic search outcomes and selection process [22]. Review questions, search strategies and inclusion criteria for this review were prospectively specified and registered with PROSPERO at http://www.crd.york.ac.uk/PROSPERO/ (CRD42015026336). Study inclusion criteria. Interventional and cohort studies that were published in peer- reviewed journals and examined the effects of exercise training on zinc status were included. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 2 / 13 Effects of regular exercise on zinc status Fig 1. PRISMA diagram showing the systematic review process. https://doi.org/10.1371/journal.pone.0184827.g001 Fig 1. PRISMA diagram showing the systematic review process. https://doi.org/10.1371/journal.pone.0184827.g001 Exercise training was defined as an intervention, over a defined period of time, which involves more than one bout of continuous physical activity [23]. Males and females, aged between 18 and 65 years, who were apparently healthy and not diagnosed with any major health condition or illness were included in the present review. The included studies must include a control group and also report indices of zinc status, such as dietary zinc intake, zinc-related enzymes and/or concentrations of zinc measured in serum/plasma, urine, erythrocytes, hair and nail, before and after the exercise training period. Two investigators independently reviewed each citation, and where appropriate, each full report, to determine whether the study satisfy the inclusion criteria. Results The electronic database search identified a total of 4919 citations following removal of dupli- cates. After initial screening of titles and abstracts, 4802 citations were excluded as they were irrelevant to the current review. Of the remaining 117 full texts assessed for eligibility, six stud- ies met the inclusion criteria. Fig 1 shows the details of study selection and reasons for full text exclusion. Risk of bias assessment All included studies were assessed for risk of biases that are applicable for non-randomised interventional studies as recommended by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) guidelines [24]. Two independent reviewers assessed the potential risk of biases including recruitment bias, valid measurement of primary out- comes (serum zinc concentration and dietary zinc intake), incomplete accounting for partici- pants in control or exercising groups, and selective outcome reporting. The ratings of risk included low, unclear and high. Studies with exercise group matched with control group in age, sex and body composition represent a low risk of bias in recruitment. Valid measurement of serum zinc was determined by the use of trace element free collection tubes, appropriate method of analysis i.e. atomic absorption spectroscopy or inductively coupled plasma mass spectroscopy, and timing of blood collection. Similarly, low risk of bias for the valid measure- ment of dietary zinc was achieved by the use of appropriate method of collection (food record or 24 h recall), reporting of energy or macronutrient intake and consideration of supplemental intake. Data on risk of bias in individual studies were entered into Review Manager 5.3 [25]. Data extraction and quality assessment of included studies Two investigators extracted published data from all included studies independently, and any differences were resolved by discussion. Data extracted included descriptive information, such as the study authors, year of publication, country where the study was conducted, the number, sex and other characteristics of the control and study populations. The duration, intensity and mode of exercise performed in the study were detailed in the data extraction process. Primary outcomes extracted were plasma/serum zinc concentration and dietary zinc intake, before and following exercise training. Plasma and serum concentrations were grouped to represent sys- temic zinc concentration; in this report the term “serum” will be used to represent both serum and/or plasma. Other potential markers of zinc status, including zinc concentrations in urine and erythrocyte, and activity or concentration of Cu-Zn superoxide dismutase were extracted as secondary outcomes. 3 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status Statistical analyses Changes in the primary outcome measures of serum zinc concentration and dietary zinc intake were calculated by subtracting final from baseline values, with calculated estimation of SD of change using an assumed correlation of 0.7 between the two values [26]. The differences between exercise and control groups were calculated by subtracting pre-post change in the exercise group from pre-post change in the control group. Standard error of between-group differences was estimated from calculation using SD of change and number of participants in the two groups. Pairwise comparisons of mean differences in change between the exercise and control groups were calculated in Review Manager 5.3 [25] and summarised visually in forest plots. Meta-analysis for the effect of exercise training on zinc status was deemed inappropriate due to the variations in the exercise training reported. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status Table 1. Characteristics of included studies. Study (author, year) Study type1 Study group Control group Study n Study Age (y)2 Control n Control Age (y) Sex (M/F) Exercise training Indices of zinc status Azizbeigi et al. 2013 I, R Untrained men Untrained men 10 21.1 ± 2.1 10 23.3 ± 2.5 M Progressive resistance exercise training 3 x/week for 8 weeks Dietary, SOD3 Cordova & Navas 1998 C Spanish League of Volleyball players Moderately trained university students 12 25.9 ± 2.6 12 22.3 ± 1.2 M Volleyball season training, 5 h/d, 7 x/week, approximately 8 weeks Dietary, serum, urine Fogelholm et al. 1991 C Sailors Bank clerks 14 28 ± 0.27 11 33 ± 0.6 M Transatlantic sailing for 20 d Serum Fogelholm 1992 I, NR University students University students 21 24 ± 0.6 18 26 ± 0.6 F Progressive aerobic exercise training for 24 weeks, from 2 x/week to 6 x/ week, 30–45 min/d at 60– 80% HRR Serum, erythrocyte Lukaski et al. 1990 C Varsity swimmers Non-training university students 13 NS 15 NS M Competitive swimming season for 24 weeks, not specified Dietary, plasma, SOD Varsity swimmers Non-training university students 16 NS 13 NS F Competitive swimming season for 24 weeks, not specified Dietary, plasma, SOD Peake et al. 2003 I, NR Well-trained distance runners Sedentary males 10 28 ± 7 7 21 ± 0 M 16% increase in running training volume over 4 weeks Plasma 1 C, cohort; I, interventional; NR, not randomised; R, randomised; 2 presented as mean ± SD; NS, not specified; 3 SOD, superoxide dismutase https://doi.org/10.1371/journal.pone.0184827.t001 Table 1. Characteristics of included studies. outcomes reported in the included studies comprise of zinc concentration in serum [28,30– 32], urine [30] and erythrocyte [28], dietary zinc intake [27,30,31], and erythrocyte Cu,Zn- SOD [27,31]. Other zinc-dependent enzymes, such as lactate dehydrogenase or carbonic anhy- drase, were not reported in the included studies. Study characteristics The characteristics of studies included in the current review are described in Table 1. Three of the included studies were prescribed exercise interventions [27–29] whereby two studies assigned specific exercise interventions to sedentary [28] or athletic populations [29]; another study randomised untrained men into either an exercise group or a control group [27]. The remaining three cohort studies examined athletes before and after a period of time that was rel- evant for their sport, i.e. competition season [30,31] and a 20 day sailing race [32]. The types of exercise in the included studies were heterogeneous; five of the included studies included exer- cise that is mostly in aerobic activity [28–32] while one study included a resistance exercise program [27]. The duration of the exercise training was between 20 days and 6 months. The number of total participants in each study ranged from 20 to 75. The majority of the studies examined males [27,29,30,32], with the exception of two studies [28,31]. The relevant zinc PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 4 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Change in dietary zinc intake Three studies [27,30,31] provided data on change in dietary zinc intake for four comparisons. The results from the included studies did not include statistical differences of the change in dietary zinc intake between the exercise and control group. Forest plot summarising the calcu- lated between group mean difference ± 95% CI of dietary zinc intake is shown in Fig 3. The mean differences of change ranged from 0.07 to 9.5 mg/d; the exercising groups consumed sig- nificantly higher dietary zinc compared to controls in two out of four comparisons. Two stud- ies reported dietary zinc intake of participants with comparisons to the recommended intakes. Lukaski et al. [31] reported dietary zinc intake of control females only to be lower than 67% of RDI at both preseason and postseason. Fogelholm [28] reported that 21% of those in the exercising group were consuming dietary zinc intakes below the Nordic recommendations during the intervention; similarly, 27% of the control group presented with low dietary zinc intake. In the only study that reported zinc density [31], no significant changes were noted in zinc density within the diets before and after the competitive swimming season in both male and female participants. Effects of regular exercise on zinc status Fig 2. Pairwise comparisons of the change in serum zinc concentration (μmol/L) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g002 Fig 2. Pairwise comparisons of the change in serum zinc concentration (μmol/L) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g002 Fig 2. Pairwise comparisons of the change in serum zinc concentration (μmol/L) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g002 Fig 2. Pairwise comparisons of the change in serum zinc concentration (μmol/L) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g002 Change in serum zinc concentration Five studies [28–32] provided data on changes in serum zinc concentrations for six compari- sons. Statistical significance of between-group changes in serum zinc concentration were not reported for three studies [29–31], while one study reported non-significant increases in serum zinc concentration in both exercise and control groups [28]. One study found a signifi- cantly higher change in serum zinc concentration in the exercising group compared to the control group [32]. Fig 2 shows a forest plot that summaries the mean difference ± 95% CI in serum zinc concentration between the exercise and control groups. The mean differences of change ranged from -0.20 to 2.40 μmol/L; two out of six comparisons elicited significantly higher increase of serum zinc concentrations in the exercise group compared to control [31,32], while one comparison reported significantly lower change in serum zinc for the exercising group [30]. Two studies reported serum zinc concentration with comparison to the reference range. No participants presented with plasma zinc concentration below the reference range in Lukaski et al.’s study. A small percentage of participants displayed low serum zinc concentration in another study [28], with 5% of the exercise group presenting with low serum zinc at all time points and 6% of the control group only at the end of the intervention period. 5 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 https://doi.org/10.1371/journal.pone.0184827.g003 Risk of bias assessment Fig 4 shows the risk of biases summary for all studies. The support for judgements is presented in S1 Table. The majority of the studies demonstrated low risks of bias for recruitment with control group matched in age, sex and BMI with the exercise group [27,28,30–32]. All included studies scored low or unclear risks of bias for valid measurement of serum zinc, with the excep- tion of one study scoring high risk of bias, where non-fasting blood samples were collected for the analysis of serum zinc [30]. Similarly, all of the studies presented with low risk of bias for valid measurement of dietary zinc. Most of the studies reported follow up data for all partici- pants thereby scoring low risk of bias for incomplete accounting of participants. The majority of the studies reported all relevant outcomes with the exception of one study that failed to report measured dietary intake data [29] and another study which did not present longitudinal changes of dietary zinc intake over the intervention period [28]. Other outcomes (RBC and urinary zinc concentrations and, erythrocyte Cu, Zn-SOD) Results for other zinc outcomes are presented in Table 2. One study reported significant increases in erythrocyte zinc concentration of the exercise group compared to the control group [28]. Another study reported urinary zinc losses, however no statistical significance was reported [30]. Two studies [27,31] reported data on changes in erythrocyte Cu, Zn-SOD activ- ity for three comparisons (Table 2). Significant increase in SOD activity following exercise was reported in the exercise group compared to the control group in one study [27]. Fig 3. Pairwise comparisons of the change in dietary zinc intake (mg/d) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g003 Fig 3. Pairwise comparisons of the change in dietary zinc intake (mg/d) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g003 Fig 3. Pairwise comparisons of the change in dietary zinc intake (mg/d) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi.org/10.1371/journal.pone.0184827.g003 Fig 3. Pairwise comparisons of the change in dietary zinc intake (mg/d) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). Fig 3. Pairwise comparisons of the change in dietary zinc intake (mg/d) between exercise and control groups in interventional trials. Data are presented as mean difference (95% CI). https://doi org/10 1371/journal pone 0184827 g003 6 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status rocyte zinc concentration, urinary zinc loss and Cu,Zn SOD activity in the included studies. Table 2. Change in erythrocyte zinc concentration, urinary zinc loss and Cu,Zn SOD activity in the included studies. Study (author, year) Outcomes Change in control group Change in exercise group Units Statistical significance of between group difference Cordova & Navas 1998 Urinary zinc excretion -15 ± 208.68 145 ± 173.36 μg/day NR Fogelholm 1992 Erythrocyte zinc concentration -0.01 ± 0.003 0.05 ± 0.004 μmol/g Hb < 0.001 Azizbeigi et al. 2013 Erythrocyte Cu, Zn-SOD -48.73 ± 232.65 127.53 ± 154.66 U/g Hb 0.014 Lukaski et al. 1990 M -145 ± 379.66 795 ± 387.31 U/g Hb NR Lukaski et al. 1990 F -197 ± 347.33 1566 ± 413.22 U/g Hb NR Hb, haemoglobin; NR, not reported; SOD, superoxide dismutase https://doi.org/10.1371/journal.pone.0184827.t002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Discussion The current review of interventional and cohort studies has identified incomplete evidence for the effects of exercise training on zinc status in an apparently healthy adult population. In total, six studies satisfied the inclusion criteria for the present analysis. The limited evidence suggests that exercise is associated with higher dietary zinc intake and erythrocyte SOD activ- ity; while minimal differences were observed in serum zinc concentrations. We have revealed distinct gaps in the current literature regarding the effects of exercise training on zinc status with implications on establishing dietary zinc requirements in populations that exercise regularly. The studies included in the current review consisted of exercise interventions that are dis- similar in mode, duration and intensity. Three out of six studies included exercise that is pre- dominately aerobic in nature, for example running and swimming [28,29,31]. A mix of aerobic and anaerobic activities were used in two studies [30,32], while another study utilised resistance exercise training only [27]. The heterogeneity in exercise interventions in the included studies contributed to the variance of the outcomes measured. Further, while changes in early markers of exercise training are evident within hours following the first exercise ses- sion [33], repeated and progressive exercise bouts are required to develop and maintain benefi- cial adaptations of exercise. The duration of exercise interventions described in the current analysis ranged from 3 to 24 weeks; therefore, there may be substantial differences in the extent of exercise adaptations presented by the included studies. Combined with the small number of studies in the current review, a quantitative summation of the changes in zinc status, beyond pairwise analyses, was deemed inappropriate. 7 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status Based on six comparisons for exercise-induced changes in serum zinc co estimated mean difference of change for two comparisons indicated that ex Fig 4. Risk of bias summary judgements on each risk of bias item for each includ symbols represent low risk of bias for the specific criteria for the included study. Yellow unclear risk of bias and red (-) symbols denote high risk of bias. Support for judgements Table. https://doi.org/10.1371/journal.pone.0184827.g004 Effects of regular Fig 4. Risk of bias summary judgements on each risk of bias item for each included study. Green (+) symbols represent low risk of bias for the specific criteria for the included study. Effects of regular exercise on zinc status changes in serum zinc concentration as a result of exercise interventions. The limitations of serum zinc concentration as a marker of zinc status in humans are well described [34]; changes in a multitude of factors, such as inflammation, hormones and age can affect the relationship between zinc status and serum zinc concentration. As such, the majority of studies reported other measures of zinc status, such as erythrocyte and urinary zinc concentration, to describe changes in zinc metabolism under exercise training. The most commonly measured zinc outcome, other than serum zinc concentration, was erythrocyte Cu, Zn-SOD activity. SOD plays a key role in antioxidative activity by providing catalytic function for the disassociation of the radical oxygen species, superoxide, into oxygen or hydrogen peroxide. In erythrocytes where oxygen exchange occurs with haemoglobin, sig- nificant intracellular oxidative stress is induced by exercise acutely through increases in oxygen consumption [35]. As a result, exercise training per se can induce significant increases in SOD activity, with erythrocyte SOD correlating positively to training status [36]. It is unclear from the current evidence whether dietary zinc intake and/or baseline zinc status may influence the exercise-induced increases in erythrocyte SOD activity. Further, while copper status may influ- ence the enzymatic activity of SOD, only two of the included studies reported copper intake of the study populations. Future studies will benefit from the inclusion of copper status to further elucidate the relationship between zinc status and SOD activity. Moreover, the examination of other zinc-dependent enzymes, such as lactate dehydrogenase or carbonic anhydrase, may provide insights into the effects of regular exercise and zinc intake on the enzymatic functions of zinc. Exercise also impacts on immunity, both acutely and chronically. Following a bout of pro- longed high intensity exercise, acute decreases in multiple components of immune function are evident including the reductions in B lymphocyte production of immunoglobulins and antigen-presenting capacity [37]. Whilst the greatest effects of exercise on immune functions are observed following high intensity exercise, prolonged moderate intensity exercise may also exert clinically significant changes in immunity [38]. The acute fluctuations in immune func- tions contribute to the increased susceptibility to potential pathogens in the hours following exercise. As one of the functions of zinc is to support innate and adaptive immune function, the provision of zinc deficient diets has been associated with lowered immunity in humans [39]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Discussion Yellow (?) symbols represent unclear risk of bias and red (-) symbols denote high risk of bias. Support for judgements is presented in S1 Table. Fig 4. Risk of bias summary judgements on each risk of bias item for each included study. Green (+) symbols represent low risk of bias for the specific criteria for the included study. Yellow (?) symbols represent unclear risk of bias and red (-) symbols denote high risk of bias. Support for judgements is presented in S1 Table. https://doi.org/10.1371/journal.pone.0184827.g004 Based on six comparisons for exercise-induced changes in serum zinc concentration, the estimated mean difference of change for two comparisons indicated that exercise induces sig- nificant increase in serum zinc; while one comparison suggested that serum zinc changes are lower in the exercise group compared to control. The current evidence suggests equivocal 8 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status All included studies within this review were low in number of participants and therefore contributed to the limited power of the comparisons in determining the effect of exercise training on markers of zinc status. Further, four out of six included studies recruited men only; similar level of sex bias were noted previously [21], highlighting the sex bias that exists in health research [41]. Moreover, the current review included mostly participants who were highly trained at baseline, for example volleyball players and swimmers. Therefore, the results of this analysis may not be applicable for untrained populations who are initiating exercise training protocols. The implications of the effects on zinc metabolism from initiating exercise training as part of a lifestyle program in patients with chronic diseases, such as diabetes, may be an important consideration particularly in populations at risk of marginal zinc status. This review is the first to determine the effects of exercise training on markers of zinc status in an adult population. The current report extends our knowledge of the acute effects of exer- cise on markers of zinc metabolism [20,21]. In the previous reports we described acute fluctua- tions in serum zinc concentrations in the hours following a bout of aerobic activity, thereby providing the basis to determine the potential long term effects of repeated bouts of exercise on markers of zinc status. One of the strengths of the present review is the determination of risk of biases as per the recommendations by the GRADE working party [24]. In our evalua- tion, a number of risk of biases were identified, including factors that have implications in the valid measurement of serum zinc concentration, and the risk of selective reporting, particu- larly in regards to dietary zinc intake. As serum zinc concentrations are affected by numerous factors, including diurnal fluctuations, fasting status and inflammation [1], it is important for investigators to control for potential confounding variables in the determination of the rela- tionship between exercise and serum zinc concentrations. Further, the omission of dietary zinc intake at baseline and following exercise interventions in some studies does not allow for the determination of the effect of exercise on dietary intake and the potential modulation of diet on exercise-induced changes in zinc metabolism. Future studies should consider including additional information regarding zinc density, i.e. amount of dietary zinc compared to energy intake, thereby furthering the understanding of the relationship between energy and zinc intake. The present review revealed an incomplete evidence base in evaluating the effect of long term exercise training on markers of zinc status. Further, the strength of the presented evi- dence is limited by the majority of study designs reported being non-randomised or cohort studies. The limited evidence suggests that dietary zinc intake may increase in those who are physically active, by homeostatic adjustments for increases in energy expenditure. Additional studies, with different study populations such as those participating in moderate levels of phys- ical activity, are required to extend the evidence to the general population. Future evaluation on the effects of zinc intake on physical performance will present implications for clinical sports nutrition practice. Exercise training was found to be associated with increases in SOD activity as part of the adaptations of exercise, consistent with the current literature [35]. As the turnover of erythrocyte SOD requires sufficient zinc status to support the adaptations of exer- cise, the current dietary recommendations for populations who are initiating or continuing exercise intervention should be to consume dietary zinc levels, to at least the Recommended Daily Intake [42]. S1 Table. Risk of bias assessment of included studies. (DOCX) However, the evidence for zinc status in modulating exercise-induced immune changes is less clear. The included study by Peake et al. failed to find significant relationships between plasma zinc levels and immune outcomes following a period of increased training in runners [29]. Further investigations into the relationships between zinc status, immune function and exercise are required. On the basis of four comparisons, dietary zinc intake appears to be higher in exercise groups compared to control in two comparisons; therefore, we deemed the effects of exercise training on dietary zinc intake to be equivocal. In a study by Lukaski et al., while dietary zinc intake increased in male swimmers over the competition period, zinc density (mg/MJ) remained unchanged [31]. This suggests that higher total zinc intake derived from increased amount of food consumed. In the maintenance of homeostasis, exercise and associated energy expenditure can impact on food and energy intake [40]. It is difficult to elucidate the influence of changing dietary patterns, as a result of exercise, on zinc status and related outcomes. The majority of the included studies did not provide details on the sources of zinc intake or its potential impact on gastrointestinal zinc absorption and zinc status. Further, a number of studies collected dietary data but failed to report quantitative values for dietary zinc intake before and following exercise intervention. The selective reporting of outcomes limited the number of comparisons and evidence available in the present review. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 9 / 13 Author Contributions Conceptualization: Anna Chu, Samir Samman. Data curation: Anna Chu, Trishala Varma. Formal analysis: Anna Chu, Trishala Varma, Peter Petocz, Samir Samman. Methodology: Anna Chu, Trishala Varma, Peter Petocz, Samir Samman. Software: Anna Chu. Conceptualization: Anna Chu, Samir Samman. Data curation: Anna Chu, Trishala Varma. Formal analysis: Anna Chu, Trishala Varma, Peter Petocz, Samir Samman. Supervision: Samir Samman. Writing – original draft: Anna Chu. Writing – review & editing: Peter Petocz, Samir Samman. Writing – review & editing: Peter Petocz, Samir Samman. 13. Smiles WJ, Hawley JA, Camera DM. Effects of skeletal muscle energy availability on protein turnover responses to exercise. J Exp Biol. 2016; 219: 214–225. https://doi.org/10.1242/jeb.125104 PMID: 26792333 Supporting information 10 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status References Does physical exercise modify antioxidant requirements? Nutr Res Rev. 2008; 21: 3–12. https://doi.org/10.1017/S0954422408018076 PMID: 19079851 15. Ganapathy S, Volpe S. Zinc, exercise, and thyroid hormone function. Crit Rev food Sci . . .. 1999; 37– 41. 16. Arikan S, Akkus H, Halifeoglu I, Baltaci AK, Arikan S., Akkus H., et al. Comparison of plasma leptin and zinc levels in elite athletes and sedentary people. Cell Biochem Funct. 2008; 26: 655–658. https://doi. org/10.1002/cbf.1480 PMID: 18570187 17. Haralambie G. Serum zinc in athletes in training. Int J Sport Med. 1981; 17. Haralambie G. Serum zinc in athletes in training. Int J Sport Med. 1981; 18 N i l RJ L i MG B l E M i i d t t i i l d 18. Nuviala RJ, Lapieza MG, Bernal E. Magnesium, zinc, and copper status in women involved in different sports. Int J Sport Nutr. 1999; 9: 295–309. PMID: 10576863 19. Crespo R, Relea P, Lozano D, Macarro-Sanchez M, Usabuaga J, Villa L, et al. Biochemical markers of nutrition in elite-marathon runners. J Sport Med Phys Fit. 1995; 35: 268–72. 20. Chu A, Petocz P, Samman S. Immediate Effects of Aerobic Exercise on Plasma/Serum Zinc Levels: A Meta-analysis. Med Sci Sports Exerc. 2016; 48: 726–733. https://doi.org/10.1249/MSS. 0000000000000805 PMID: 26540263 21. Chu A, Petocz P, Samman S. Plasma/Serum Zinc Status During Aerobic Exercise Recovery: A Sys- tematic Review and Meta-Analysis. Sports Med. 2017; 47: 127–134. https://doi.org/10.1007/s40279- 016-0567-0 PMID: 27260681 22. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP a, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009; 6: e1000100. https://doi.org/10.1371/journal.pmed. 1000100 PMID: 19621070 23. Boule´ NG, Kenny GP, Haddad E, Wells GA, Sigal RJ. Meta-analysis of the effect of structured exercise training on cardiorespiratory fitness in Type 2 diabetes mellitus. Diabetologia. 2003; 46: 1071–1081. https://doi.org/10.1007/s00125-003-1160-2 PMID: 12856082 24. Guyatt GH, Oxman AD, Vist G, Kunz R, Brozek J, Alonso-Coello P, et al. GRADE guidelines: 4. Rating the quality of evidence—study limitations (risk of bias). J Clin Epidemiol. 2011; 64: 407–415. https://doi. org/10.1016/j.jclinepi.2010.07.017 PMID: 21247734 25. The Cochrane Collaboration. Review Manager (RevMan). Copenhagen: The Nordic Cochrane Centre; 2014. 26. The Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions. 5.1.0. Hig- gins J, Green S, editors. 2011. 27. Azizbeigi K, Azarbayjni MA, Peeri M, Agha-Alinejad H, Stannard S. References 1. Samman S. Zinc. Nutr Diet. 2007; 64: S131–S134. https://doi.org/10.1111/j.1747-0080.2007.00200.x 2. Jackson M. Physiology of zinc: general aspects. In: Mills C, editor. Zinc in Human Biology. New York, NY: Springer-Verlag; 1989. pp. 1–14. 3. Vallee BL, Falchuk KH. The biochemical basis of zinc physiology. Physiol Rev. 1993; 73: 79–118. PMID: 8419966 4. Haase H, Rink L. Functional significance of zinc-related signaling pathways in immune cells. Annu Rev Nutr. 2009; 29: 133–52. https://doi.org/10.1146/annurev-nutr-080508-141119 PMID: 19400701 5. Foster M, Samman S. Zinc and regulation of inflammatory cytokines: implications for cardiometabolic disease. Nutrients. 2012; 4: 676–94. https://doi.org/10.3390/nu4070676 PMID: 22852057 6. Oteiza PI. Zinc and the modulation of redox homeostasis. Free Radic Biol Med. 2012; 53: 1748–1759. https://doi.org/10.1016/j.freeradbiomed.2012.08.568 PMID: 22960578 7. Foster M, Samman S. Zinc and redox signaling: perturbations associated with cardiovascular disease and diabetes mellitus. Antioxid Redox Signal. 2010; 13: 1549–73. https://doi.org/10.1089/ars.2010. 3111 PMID: 20568953 8. Lukaski HC. Low dietary zinc decreases erythrocyte carbonic anhydrase activities and impairs cardiore- spiratory function in men during exercise. Am J Clin Nutr. 2005; 81: 1045–1051. PMID: 15883427 9. Chu A, Foster M, Samman S. Determinants of zinc transport in humans: zinc status, exercise, inflam- mation and chronic diseases. In: Ostojic S, editor. Human health and nutrition: new research. New York: Nova Science; 2015. pp. 17–48. 10. Eckel RH, Jakicic JM, Ard JD, De Jesus JM, Houston Miller N, Hubbard VS, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: A report of the American College of cardiology/American Heart Association task force on practice guidelines. Circulation. 2014; 129: 76–99. https://doi.org/10.1161/01.cir.0000437740.48606.d1 11. Colberg SR, Sigal RJ, Fernhall B, Regensteiner JG, Blissmer BJ, Rubin RR, et al. Exercise and type 2 diabetes: the American College of Sports Medicine and the American Diabetes Association: joint posi- tion statement. Diabetes Care. 2010; 33: e147–67. https://doi.org/10.2337/dc10-9990 PMID: 21115758 12. Garber CE, Blissmer B, Deschenes MR, Franklin B a., Lamonte MJ, Lee IM, et al. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: Guidance for prescribing exercise. Med Sci Sports Exerc. 2011; 43: 1334– 1359. https://doi.org/10.1249/MSS.0b013e318213fefb PMID: 21694556 13. Smiles WJ, Hawley JA, Camera DM. Effects of skeletal muscle energy availability on protein turnover responses to exercise. J Exp Biol. 2016; 219: 214–225. https://doi.org/10.1242/jeb.125104 PMID: 26792333 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 11 / 13 Effects of regular exercise on zinc status 14. Margaritis I, Rousseau A. References The Effect of Progressive Resis- tance Training on Oxidative Stress and Antioxidant Enzyme Activity in Erythrocytes in Untrained Men. Int J Sport Nutr Exerc Metab. 2013; 23: 230–238. PMID: 23239675 28. Fogelholm M. Micronutrient status in females during a 24-week fitness-type exercise program. Ann Nutr Metab. 1992; 36: 209–218. PMID: 1471858 29. Peake JM, Gerrard DF, Griffin JFT. Plasma zinc and immune markers in runners in response to a mod- erate increase in training volume. Int J Sports Med. 2003; 24: 212–216. https://doi.org/10.1055/s-2003- 39094 PMID: 12740742 30. Co´rdova A, Navas FJ. Effect of training on zinc metabolism: changes in serum and sweat zinc concen- trations in sportsmen. Ann Nutr Metab. 1998; 42: 274–282. PMID: 9812018 31. Lukaski HC, Hoverson BS, Gallagher SK, Bolonchuk WW. Physical training and copper, iron, and zinc status of swimmers. Am J Clin Nutr. 1990; 51: 1093–9. PMID: 2349924 32. Fogelholm GM, Lahtinen PK. Nutritional evaluation of a sailing crew during a transatlantic race. Scand J Med Sci Sports. 1991; 1: 99–103. https://doi.org/10.1111/j.1600-0838.1991.tb00278.x 33. Hawley JA, Hargreaves M, Joyner MJ, Zierath JR. Integrative Biology of Exercise. Cell. 2014; 159: 738–749. https://doi.org/10.1016/j.cell.2014.10.029 PMID: 25417152 34. Lowe NM, Dykes FC, Skinner A-L, Patel S, Warthon-Medina M, Decsi T, et al. EURRECA-Estimating zinc requirements for deriving dietary reference values. Crit Rev Food Sci Nutr. 2013; 53: 1110–23. https://doi.org/10.1080/10408398.2012.742863 PMID: 23952091 35. Nikolaidis MG, Jamurtas AZ. Blood as a reactive species generator and redox status regulator during exercise. Arch Biochem Biophys. 2009; 490: 77–84. https://doi.org/10.1016/j.abb.2009.08.015 PMID: 19712664 36. Speich M, Pineau a, Ballereau F. Minerals, trace elements and related biological variables in athletes and during physical activity. Clin Chim Acta. 2001; 312: 1–11. PMID: 11580904 37. Gleeson M. Immune function in sport and exercise. J Appl Physiol. 2007; 103: 693–9. https://doi.org/10. 1152/japplphysiol.00008.2007 PMID: 17303714 12 / 13 PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 Effects of regular exercise on zinc status 38. Diment BC, Fortes MB, Edwards JP, Hanstock HG, Ward MD, Dunstall HM, et al. Exercise intensity and duration effects on in vivo immunity. Med Sci Sports Exerc. 2015; 47: 1390–1398. https://doi.org/ 10.1249/MSS.0000000000000562 PMID: 25380472 39. Beck FW, Prasad AS, Kaplan J, Fitzgerald JT, Brewer GJ. Changes in cytokine production and T cell subpopulations in experimentally induced zinc-deficient humans. Am J Physiol. 1997; 272: E1002– E1007. PMID: 9227444 40. Manore MM. Weight Management for Athletes and Active Individuals: A Brief Review. Sport Med. Springer International Publishing; 2015; 45: 83–92. PLOS ONE \| https://doi.org/10.1371/journal.pone.0184827 September 20, 2017 References https://doi.org/10.1007/s40279-015-0401-0 PMID: 26553496 41. Kim AM, Tingen CM, Woodruff TK. Sex bias in trials and treatment must end. Nature. 2010; 465: 688– 9. https://doi.org/10.1038/465688a PMID: 20535184 42. National Health and Medical Research Council. Nutrient Reference Values for Australia and New Zea- land Including Recommended Dietary Intakes. 2006. 13 / 13
https://openalex.org/W4317722007	https://www.frontiersin.org/articles/10.3389/fpls.2022.1102201/pdf	English	null	Computational analysis and expression profiling of potassium transport-related gene families in mango (Mangifera indica) indicate their role in stress response and fruit development	Frontiers in plant science	2,023	cc-by	8,800	OPEN ACCESS OPEN ACCESS EDITED BY Galal Bakr Anis, Agricultural Research Center, Egypt REVIEWED BY Saira Bano, University of Sargodha, Pakistan Wei Li, Agricultural Genomics Institute at Shenzhen (CAAS), China CORRESPONDENCE Farrukh Azeem azeuaf@hotmail.com SPECIALTY SECTION This article was submitted to Plant Biotechnology, a section of the journal Frontiers in Plant Science RECEIVED 18 November 2022 ACCEPTED 21 December 2022 PUBLISHED 23 January 2023 Lin Tan 1, Muhammad Waqas 2, Abdul Rehman 2, Muhammad Abdul Rehman Rashid 2, Sajid Fiaz 3, Hamid Manzoor 4 and Farrukh Azeem 2 Lin Tan 1, Muhammad Waqas 2, Abdul Rehman 2, Muhammad Abdul Rehman Rashid 2, Sajid Fiaz 3, Hamid Manzoor 4 and Farrukh Azeem 2* 1Haikou Experimental Station, Chinese Academy of Tropical Agricultural Sciences, Haikou, Hainan, China, 2Department of Bioinformatics and Biotechnology, Government College University, Faisalabad, Pakistan, 3Department of Plant Breeding and Genetics, The University of Haripur, Haripur, Pakistan, 4Institute of Molecular Biology and Biotechnology, Bahauddin Zakariya University, Multan, Pakistan CITATION Tan L, Waqas M, Rehman A, Rashid MAR, Fiaz S, Manzoor H and Azeem F (2023) Computational analysis and expression proﬁling of potassium transport-related gene families in mango (Mangifera indica) indicate their role in stress response and fruit development. Mango (Mangifera indica) fruit is known for its taste, health beneﬁts, and drought tolerance. Potassium (K+) is one of the most abundant ions in a plant cell. It is important for various biological functions related to plant growth, development, and ﬂowering/fruiting. It signiﬁcantly contributes to fruit yield, quality, and drought tolerance in plants. However, molecular mechanisms comprising K+ transport in mango are least known. In the present study, 37 members of K+ transport-related genes (PTGs) were identiﬁed in mango, which include 22 K+ transporters (16 HAKs, 1 HKT, and 6 KEAs) and 15 K+ channels (6 TPKs and 8 Shakers). All PTGs were predicted to be expressed at the plasma membrane and possess characteristic motifs and domains. Phylogenetic analysis identiﬁed a strong kinship of PTGs among Oryza sativa, Arabidopsis thaliana, Cicer arietinum, Malus domestica, and M. indica. The promoter analysis identiﬁed 60 types of cis-elements related to various biological processes. RNA-seq-based expression proﬁling identiﬁed that MiTPK1.2, MiHAK1, MiHAK2.1, HAK6.1, and MiAKT1.1 were most upregulated in roots and that MiKEA2, MiAKT2, and MiAKT1 were upregulated in leaves. Moreover, MiAKT6, MiHAK1.1, MiKAT2, MiKAT2.1, MiHKT1, MiTPK1.1, MiHAK7, and MiHAK12 were highly expressed during the ﬁve growth stages of mango fruit. mango, potassium, channel, transporter, fruit development, drought stress TYPE Original Research PUBLISHED 23 January 2023 DOI 10.3389/fpls.2022.1102201 TYPE Original Research PUBLISHED 23 January 2023 DOI 10.3389/fpls.2022.1102201 TYPE Original Research PUBLISHED 23 January 2023 DOI 10.3389/fpls.2022.1102201 Computational analysis and expression proﬁling of potassium transport-related gene families in mango (Mangifera indica) indicate their role in stress response and fruit development Computational analysis and expression proﬁling of potassium transport-related gene families in mango (Mangifera indica) indicate their role in stress response and fruit development 1 Introduction Moreover, the carrier-like KUP/HAK/KT family (Azeem et al., 2022), HKT uniporters and symporters (Sahi et al., 2022), and KEA antiporters (Wang et al., 2017) also manifested their involvement in this mechanism. The KUP/HAK/KT family may regulate K+ with great afﬁnity (Ye et al., 2013). The assimilation and homeostasis of K+ and sodium ions depend on the HKT proteins. In plants, there are two kinds of HKTs, namely, HKT1- like and HKT 2-like. The Na+ uniporters make up class I HKTs, whereas Na+ and K+ symporters make up class II HKTs (Riedelsberger et al., 2021). The KEA proteins are probably similar to bacterial KefC K+/H+ antiporters (Waters et al., 2013; Assaha et al., 2015). Most KEAs are shown to control the pH value of stroma and thylakoids at the chloroplast membrane (Sun et al., 2018). In endomembrane cells, KEA 4, 5, and 6 help to maintain pH and K+ homeostasis in a balanced manner (Sze and Chanroj, 2018; Zhu et al., 2018). Pore domains (PDs) are used to calculate the K+ channels, which are heterodimeric proteins with trans-membrane sections. Functional multimeric proteins are associated with four PDs to make a navigation pathway of channels. P domain of the K+ channels contains a highly conserved motif, i.e., “GYGD/E”. In Arabidopsis thaliana, 15 K+-selective channels, including one K+ inward rectiﬁer (Kir- like), ﬁve tandem-pore K+ channels (TPK), and nine voltage- gated ion channels, are grouped into three families based on their physiography. Additionally, K+ transporters are divided into three families—the Trk/HKT family of high-afﬁnity K+ transporters, the KEA (K+/H+ antiprotons) family of K+ efﬂux antiporters, and the KUP/HAK/KT family of K+ uptake permeases—which collectively have 13 members (1 member) (Aranda-Sicilia et al., 2012). K+ acquisition is one of the most important issues covering organic agriculture, and it is because most organic sources of K+ are poorly soluble, limiting plant growth (Council, 1993). Due to its role in protein synthesis, ionic stability, photosynthesis, stress tolerance, translocation of photosynthates, and the initiation of several plant enzymes, K+ is an essential component for plant maturation and ﬁnal production (Hossain et al., 2021; Ul Hassan et al., 2021). K+ plays a unique role in the generation of starch, blooming, and fruit output (Farooq et al., 2014). However, increased K+ rates in plants can result in an imbalance in magnesium and calcium nutrition (Nahar et al., 2015). 1 Introduction development (Bahadur et al., 2019; Wang et al., 2022). Foliar application of K+ is also practiced to improve fruit yield. Therefore, to amend the K+ utilization for mango, there is a need to acknowledge K+ transport mechanisms. Mango (Mangifera indica) is a widespread, evergreen, and one of the most dominant tropical fruits worldwide, being the sixth most cultivated fruit after bananas (Musa acuminata), watermelons (Citrullus lanatus), apples (Malus domestica), oranges (Citrus sinensis), and grapes (Vitis vinifera) (Ngamchuachit et al., 2015). Its bright color, distinctive quality, unique taste, and nutritional value have promoted its consumption. The mango fruit trade continues to grow and develop in the food service and dime markets (Celestin, 2019). Mango can be used as a pudding, fresh juice, and extracted products; reﬁned into jam; or used as a jelly bean. With extensive cultivation of this fruit, India is known as a major producer with 25 million tonnes in 2020 (FAOSTAT, 2022), as compared to other countries like Pakistan, China, Indonesia, Malawi, and Mexico (Galán Saúco, 2002). Mango trees can withstand drought conditions, but these may affect the overall quality of the fruit (Cleveland, 2012). Since peak fruit development occurs in the dry season, water requirement is critical (Singh and Kushwaha, 2006; Snowden et al., 2014). The standard ranking of export mangoes (ﬂavor, size, shape, and color) can contribute an additional 30%–50% to the payout (Laurio, 2021). Farmers are encouraged to increase the irrigation for this crop because of its tremendous impact on proﬁtable yield. However, water scarcity and the enormous expenditure of energy required to raise irrigation water during the peak season have threatened fruit production and cultivation (Zikki, 2020). For this reason, solid agronomic and irrigation practices are applied at the farm level to help plants survive drought stress. However, investigations into mango drought stress tolerance are relatively scarce. Different proteins in plants that control cellular K+ absorption and distribution include both channels and transporters. Voltage-dependent channel proteins mediating K+ transport include shaker-like channels, voltage-independent tandem-pore K+ (TPK) channels, and two-pore channels (TPCs) (Voelker et al., 2010). Five subgroups are used to classify the Shaker family: weak inward rectifying channels, inward rectifying channels with KAT-like characteristics, outward rectifying channels, inward rectifying channels with AKT-like characteristics, and the silent/regulatory subunit (González et al., 2015). OPEN ACCESS The current study is the ﬁrst comprehensive report on K+ transport system in tropical fruits. Therefore, it will provide the foundation knowledge for the functional characterization of K+ genes in mango and related plants. Front. Plant Sci. 13:1102201. doi: 10.3389/fpls.2022.1102201 Front. Plant Sci. 13:1102201. doi: 10.3389/fpls.2022.1102201 COPYRIGHT © 2023 Tan, Waqas, Rehman, Rashid, Fiaz, Manzoor and Azeem. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Plant Science 01 frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 2.6 Cis-regulatory elements of K+ transporting gene family in M. indica For determining, the cis-acting regulatory elements of the K+ transporting genes in M. indica, an approximately 2,000-bp promoter region was retrieved for each gene from the NCBI gene page. After the promoter sequence was retrieved, the PlantCare database (http://bioinformatics.psb.ugent.be/ webtools/plantcare/html/) was employed to predict the cis- acting regulatory elements in promoter regions of K+ transporting genes. The PlantCare results were downloaded and opened in the Excel sheets that were used to anticipate the cis-acting regulatory elements in TBtools software (Lescot et al., 2002). 2 Material and methods 2.1 Identiﬁcation and sequence analysis of K+ transport-related gene families in M indica 2 Material and methods 2.1 Identiﬁcation and sequence analysis of K+ transport-related gene families in M indica To predict the exons and introns of the K+ transporting genes, genomic DNA and cDNA sequences of all the members of K+ transporting genes of M. indica were retrieved from the NCBI database in FASTA format. Exons and introns in these genes were predicted using an online tool named Gene Structure Display Server (GSDS) 2.0 (http://gsds.gao-lab.org/). Motifs of the K+ transporting gene family were identiﬁed by an online software called MEME (https://meme-suite.org/meme/tools/ meme). To predict the conserved motifs, the following parameters were selected: one occurrence per sequence (oops), number of motifs of 10, motifs with at least a width of 10, and utmost width of 50. Motif logos were also made by MEME (Guo et al., 2007; Bailey et al., 2009). For the identiﬁcation of PTG families in M. indica, PTG sequences from A. thaliana, Cicer arietinum, and Cajanus cajan were used to perform a BLAST search in the NCBI (https://www. ncbi.nlm.nih.gov) and Phytozome (https://phytozome-next.jgi.doe. gov/) databases (Johnson et al., 2008; Goodstein et al., 2012). An E- value 1e−10 was used as the cutoff value BLAST search. After retrieving all the sequencing, we utilized SMART and pfam tools to verify the accuracy of K+ transporting gene family in M. indica and deleted the sequences that lacked the conserved domains. To enhance the accuracy of our analysis, we deleted genes having common start positions in the genome representing the same locus or splice variants. 2.3 Physiochemical properties of K+ transporting gene family members in M. indica The physicochemical properties of PTG family members were predicted by using a web-based tool ProtParam (https:// web.expasy.org/protparam/) with default options. Gene ID, chromosomal locations, number of nucleotides in genomic DNA and mRNA, and number of exons were identiﬁed by using the NCBI gene database (https://www.ncbi.nlm.nih.gov/ gene). The subcellular localization of K+ transporting proteins was predicted by using a web-based tool ProtCamp 9.0 (ProtComp—Predict the sub-cellular localization for Plant proteins (softberry.com), and then, these locations were also veriﬁed by another online tool, CELLO v.2.5 (http://cello.life. nctu.edu.tw/), by using protein sequences of K+ transporting gene family (Garg et al., 2016; Mehanny et al., 2022). 2.2 Multiple sequence analysis and phylogenetic analysis of K+ transporting gene family To predict the conserved domains of the K+ transporting gene family in M. indica, the NCBI-CDD database (https://www. ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) was used in which all the sequences were searched against the Pfam database. These results were also veriﬁed by the SMART tool (http://smart.embl- heidelberg.de/). Then, conserved domains of the K+ transporting gene family were visualized by the TB tool, using the results of the Pfam database. The MEGA7 software with default parameters was used for multiple sequence alignment of PTG sequences of ﬁve different species, i.e., A. thaliana, M. domestica, Oryza sativa, C. arietinum, and M. indica. Moving forward after multiple sequence alignment (MSA) of 198 protein sequences of the PTG families, with the use of the IQ TREE web server (http://iqtree.cibiv.univie.ac.at/), a phylogenetic tree was built based upon the maximum likelihood method. The tree was designed and visualized by using a web- based tool ITOL (https://itol.embl.de/) (Schultz et al., 2000). 1 Introduction For plant growth, the main sources are the utilization of chemical K+ fertilizer and the disintegration of K+-minerals (Etesami et al., 2017; Bahadur et al., 2019). The application of appropriate K+ fertilizer to mango trees increases fruit yield and value. Nevertheless, unconstrained fertilizer usage can result in ﬁnancial loss to farmers, and excessive use of non-renewable resources is raising concerns about large-scale sustainable The current study was designed to identify key protein families involved in K+ transport. For this purpose, a combination of bioinformatics approaches was used to characterize potential PTGs, and expression proﬁling was conducted using next- generation sequencing (NGS) data available at the National Center for Biotechnology Information (NCBI). Frontiers in Plant Science 02 frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 Frontiers in Plant Science frontiersin.org 2.7 Chromosomal distribution of K+ transporting genes The gene location (initial and terminating position) of K+ transporting genes present on the chromosomes was identiﬁed Frontiers in Plant Science 03 frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 10.3389/fpls.2022.1102201 by the NCBI gene database. The chromosome number for each member of the PTG family was also identiﬁed from the NCBI database. The chromosomal location was visualized by a Ritchie lab tool, Phenograms (http://visualization.ritchielab.org/ phenograms/plot). indica, such as the total number of amino acids (aa), molecular weights (MW), aliphatic index, gravy, and hypothetical isoelectric points (pI), were identiﬁed using the web-based tool Expasy- protparam. Subcellular localization was also identiﬁed using cello 2.0. The length of protein ranged from 340aa to 1208aa. Molecular weight varied from 39.05 to 130.77 kDa, and the pI values start from 4.92 to 9.47 in which 16 of the proteins were acidic and 20 were basic. Interestingly, the MiHAK1.1 protein possessed a pI value of 7.01. GRAVY values of AKT and KAT-like proteins were negative (hydrophilic), while the rest of the proteins were hydrophobic, and subcellular localization showed that the location of these proteins is in the plasma membrane (Table 1). 2.8 Gene expression proﬁle of K+ transporting genes using NGS data and qRT-PCR For evaluation of the gene expression conﬁguration of the K+ transporting genes under abiotic stress (drought stress with bio- project), genome and gffﬁles were downloaded from the NCBI-SRA [Home—SRA—NCBI (nih.gov)] database after the genome, and gff ﬁles were retrieved from the NCBI-genome database. The galaxy server [Galaxy (usegalaxy.org)] was utilized to obtain the fragments per kilobase of transcript per million mapped reads (FPKM) values, and the expression pattern was shown as a heat map using TB tools. Drought stress was applied to 1-year-old mango plants (Chaunsa) grown in soil-ﬁlled pots (one plant/pot) under an ambient environment (12-h photoperiod, 50%–60% relative humidity, 25° C ± 1°C day/night temperature). Drought stress was imposed by watering each pot (n = 6) with 1 L of 200 g/L PEG-6000. Leaves were collected 2, 4, and 7 days after the start of treatment. To validate the expression proﬁles of NGS data, real-time RT-qPCR was used. For this purpose, RNA was extracted from mango leaves using the TRIzol reagent and was quantiﬁed using NanoDrop 2000 (Thermo Fisher Scientiﬁc, Waltham, MA, USA). With the use of 1 µg of RNA and the Maxima H-minus First-Strand cDNA synthesis kit, the RNA was reverse transcribed, and cDNA was stored at −20° C. An iTaq Universal SYBR Green Super-Mix and a qRT-PCR detection equipment (CFX96 Touch RT PCR Detection System, Bio-Rad Labs, Hercules, CA, USA) were used to perform the qRT- PCR. Gene-speciﬁc primers were mapped using an online program “Oligo Calculator” at mcb.berkeley.edu/labs/krantz/tools/ oligocalc.html (accessed 26 July 2022), and the speciﬁcity of primers was veriﬁed using NCBI-primer BLAST algorithm (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) (retrieved on 26 July 2022). Actin (LOC123192663) was used as a reference gene (Yao et al., 2022). The 2−DDCT method is used to calculate respective gene expression levels on the basis of three biological replicates (Livak and Schmittgen, 2001). Frontiers in Plant Science 3.2 Phylogenetic analysis of K+ transport-related proteins in M. indica To predict the functional properties as well as phylogenetic relationships of the K+ transport-related genes, using protein sequences of PTGs of O. sativa, A. thaliana, C. arietinum, M. domestica, and M. indica, a phylogenetic tree was constructed. The members of the tree were divided into ﬁve groups: i) HAK, ii) HKT, iii) KEA, iv) TPK, and v) Shakers (AKT and KAT, and GORK and SKOR). HAK, HKT, and KEA are transporters, and TPK and Shakers are Channels. In phylogenetic analysis, some orthologous pairs (MiHAK1/MdHAK1.2 and MdHAK6.2/MiHAK6.1), co- orthologous groups (MdHAK1.1/CarHAK1/MiHAK1.1, MdHAK7.1/MdHAK7.2/MiHAK7, and MdHAK5.4/CarHAK5/ MiHAK5), and paralogous groups (MiHAK8/MiHAK8.1, MiHAK2.1/MiHAK2, and MiHAK5.3/MiHAK5.1/MiHAK5.2) of M. indica were identiﬁed in the HAK family among A. thaliana, O. sativa, and M. domestica. In the HKT family, one orthologous pair was identiﬁed in A. thaliana (MiHKT1/ATHKT1). In the KEA family, two orthologous pairs (MiKEA6/CarKEA6 and MiKEA4/ OsKEA4) and two co-orthologous groups (MdKEA3.1/MdKEA3.2/ MiKEA3 and MdKEA5.1/CarKEA5/MiKEA5) were identiﬁed in the genome of M. indica. In the AKT and KAT sub-families, paralogous (MiAKT1/MiAKT1.1 and MiKAT2/MiKAT2.1) and orthologues groups (AtAKT6/MiAKT6, MiAKT2/MdAKT2.1, and MdKAT3.1/ MdKAT3.2/MiKAT3) were identiﬁed in M. indica. In the GORK and SKOR sub-families, one orthologous pair (MiSKOR1/ CarGORK) was identiﬁed; in the TPK family, some orthologous and co-orthologous groups were identiﬁed (MdTPK4.1/MdTPK4.2/ MiTPK4, MiTPK5/CarTPK5, and MiTPK1.2/MdTPK1.3); some paralogous groups were also identiﬁed (MiTPK6/MiTPK6.1 and MiTPK1.1/MiTPK1) in the M. indica genome (Figure 1). 3.4 Conserved domains and chromosomal distribution PTGs Gene expression is controlled by cis-regulatory sequences, such as enhancers and promoters, which play crucial roles in modulating the development and physiology of an organism. In the current study, we have identiﬁed 60 types of cis-regulatory elements in the 2,000-bp promoter sequences of 37 PTGs in M. indica. The promoter sequence of each gene was found to be rich in these regulatory elements. Most abundant of them all were found to be TATA-Box (TATA-box is capable of deﬁning the direction of transcription and also indicates which DNA strand should be read by the transcriptional machinery), CAAT-box (CAAT-box serves as a marker for the binding site of the transcription factor for RNA), O2-site (assists in regulating two transcription factors, those associated with the metabolism of carbon and amino acids as well as the resistance to abiotic stress), AT-rich element (a replication complex is formed at this speciﬁc site and DNA synthesis is initiated), CAT-box (cis-acting regulatory element related to meristem expression), ARE (regulatory element essential for anaerobic induction), MRE (MYB binding site involved in light responsiveness), MBS (MYB binding site involved in drought-response), BOX-4 (conserved DNA module involved in light responsiveness), TCA-element (regulatory element involved in salicylic acid responsiveness), ABRE (Abscisic Acid-Responsive Element), and P-box (gibberellin-responsive element); in addition to these, other regulatory elements were also identiﬁed, which were not present in the promoter sequence of all the genes and were found to be present in a very small number but could have a potentially signiﬁcant role in the life of M. indica plant, which includes Circadian, MREG-box, LTRGCN4_motif, TCT- motif, CGTCA-motif, TGACG-motif, G-Box, GA-motif, chs- CMA2aAT1-motif, ATCT-motif, LAMP-element, GATA-motif, Box III, AE-box, LS7TCCC-motif, TC-rich repeats, chs-Unit 1, m1ATC-motif, A-box, 3-AF1 binding site, TGA-element, TATC-box, I-box, GARE-motif, GT1-motif, AuxRR-core, ACECCAAT-box, SARECAG-motif, AT-rich sequence, GC- motif, RY-element, GTGGC-motif, chs-CMA1a, Box II, Gap- The conserved domains present in the PTG proteins in M. indica were analyzed using NCBI-CDD. After this, further analysis was performed in phases and visualized by using the software Tbtools. All the members of PTGs possess conserved domains like (the K_trans superfamily, ions_trans_2, Na_H_Exchanger, KHA, FRQI superfamily, PLN00149, and PLN00151). Not only the most conserved domains are visualized but less conserved domains are also shown and can be visualized. 3.1 Computational identiﬁcation and characterization of K+ transport- related genes A total of 37 PTGs were identiﬁed in M. indica genome (Table 1). Physiochemical characteristics of K+ transporter in M. To acquire an understanding of the structural properties of PTGs, we explored the intron and exon architecture of these Frontiers in Plant Science 04 frontiersin.org frontiersin.org Chr. No. Exon Protein length (aa) Molecular weight (Da) pI Gravy Aliphatic index Subcellular localization 10 3 498 56,848.8 9.47 0.298 104 Plasma membrane 9 2 9 753 83,293.7 7.01 0.489 111 Plasma membrane 2 5 8 781 87,042 8.59 0.336 108.92 Plasma membrane 14 8 776 86,449.4 8.15 0.377 109.11 Plasma membrane 10 12 793 88,856.8 6.92 0.375 108.15 Plasma membrane 10 8 794 88,971.3 8.53 0.366 109.58 Plasma membrane 8 5 9 837 93,080 5.78 0.373 107.35 Plasma membrane 8 4 10 848 94,175.8 5.66 0.314 105.61 Plasma membrane 2 Unknown 9 775 86,717.8 7.53 0.389 110.76 Plasma membrane 3 14 9 784 87,470.5 8.58 0.448 111.39 Plasma membrane 14 8 775 86,321.2 8.15 8.15 109.25 Plasma membrane 2 12 7 804 89,457.1 8.57 0.301 108.93 Plasma membrane 2 8 8 779 87,506.1 8.04 0.221 105.56 Plasma membrane 8 8 772 86,120.4 6.72 0.279 106 Plasma membrane 6 5 8 780 87,931.8 7.54 0.233 104.79 Plasma membrane 4 15 10 737 81,919.4 8.68 0.522 112.55 Plasma membrane 5 12 11 793 88738.3 6.54 0.358 107.53 Plasma membrane 7 8 21 1,208 130,774 4.92 0.093 105.38 Plasma membrane 13 20 573 61,984.8 6.09 0.635 120.75 Plasma membrane 4 17 20 579 62,521.5 5.39 0.659 124.58 Plasma membrane 7 4 19 806 88,152.1 5.47 0.283 109.98 Plasma membrane 9 9 20 577 62,700.8 6.33 0.628 122.43 Plasma membrane 2 8 3 347 43,193.9 8.82 0.26 107 Plasma membrane 6 2 426 47,442.5 8.96 0.094 97.65 Plasma membrane 1 3 353 39,058.5 5.74 0.265 111.53 Plasma membrane (Continued) ght (Da) pI Gravy Aliphatic index localization 8.31 0.129 99.94 Plasma membrane 6.1 0.136 105.96 Plasma membrane 8.94 0.154 98.11 Plasma membrane 8.4 0.264 107.53 Plasma membrane 6.07 −0.058 −0.058 Plasma membrane 7.33 −0.156 94.43 Plasma membrane 7.86 −0.105 95.61 Plasma membrane 8.18 −0.091 96.52 Plasma membrane 7.36 −0.084 97.62 Plasma membrane 6.49 −0.162 95.27 Plasma membrane 6.23 −0.238 88.61 Plasma membrane 6.44 −0.167 90.77 Plasma membrane Tan et al. 10.3389/fpls.2022.1102201 Superfamily, which revealed that some proteins function distantly (Figure 3). genes in M. indica. 3.1 Computational identiﬁcation and characterization of K+ transport- related genes The results of this analysis demonstrated that the number of exons varies from 2 to 20. In HKTs, there were only two exons. In HAKs, the number of exons ranged from 7 to 12. Similarly, 19–21, 2–5, and 10–13 exons were predicted in KEA, TPK, and Shaker families, respectively (Figure 2A). In some regions, CDS are less in concentration and illustrated in green color in the legend. Similarly, conserved motifs were predicted and visualized by using MEME suite. A total of 10 conserved motifs were predicted in all proteins. Among these, motifs 1–8 and 10 were present in all proteins. Motif 9 was detected in only KEAs, TPKs, and Shakers. This motif possesses the characteristic GYGD motif, which acts as a selectivity ﬁlter for K+ (Figure 2B). In M. indica, 36 of 37 genes were mapped on 14 (out of 18) chromosomes (Supplementary Figure 1). The locus of MiHAK8.1 gene was not found. As of now, this gene has not been assigned to any linkage groups on scaffolds. Hence, its location is not displayed on the map. There were no PTGs on chromosomes 3, 7, 11, and 16. Chromosome 8 contains the maximum number (5) of genes, while only one gene was present on chromosomes 1, 2, 15, and 18. Almost 40% of members (13 out of 36) were present on chromosomes 8, 12, and 13 collectively (Supplementary Figure 1). 3.4 Conserved domains and chromosomal distribution PTGs Interestingly, MiTPK1.2, MiHAK1, MiHAK2.1, HAK6.1, and MiAKT1.1 were most upregulated in roots, and MiKEA2, MiAKT2, and MiAKT1 were upregulated in leaves. box, MBSIHD-Zip 1, MSA-like, Sp1, ACA-motif, TGA-box, L- box, and WUN-motif (Figure 4). box, MBSIHD-Zip 1, MSA-like, Sp1, ACA-motif, TGA-box, L- box, and WUN-motif (Figure 4). 3.4 Conserved domains and chromosomal distribution PTGs The least conserved domains are from EFhs FIGURE 1 Phylogenetics analysis of K+ transporting proteins in Mangifera indica (Mi), Oryza sativa (Os), Arabidopsis thaliana (At), Cicer arietinum (Car), and Malus domestica (Md). Tree was generated by MEGA-7, using 1,000 bootstrap values with the neighbor- joining method. FIGURE 1 Phylogenetics analysis of K+ transporting proteins in Mangifera indica (Mi), Oryza sativa (Os), Arabidopsis thaliana (At), Cicer arietinum (Car), and Malus domestica (Md). Tree was generated by MEGA-7, using 1,000 bootstrap values with the neighbor- joining method. FIGURE 1 Phylogenetics analysis of K+ transporting proteins in Mangifera indica (Mi), Oryza sativa (Os), Arabidopsis thaliana (At), Cicer arietinum (Car), and Malus domestica (Md). Tree was generated by MEGA-7, using 1,000 bootstrap values with the neighbor- joining method. 07 Frontiers in Plant Science frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 A B FIGURE 2 Genomic and proteomic features of PTGs. (A) Intron/exon architecture of PTGs in Mangifera indica. The green lines indicate introns. The upstream and downstream regions of genes are represented in red color. The exons are represented by green shapes. (B) Prediction of conserved motifs PTG proteins from M. indica. The conserved motifs are shown in this ﬁgure along with their motif details. B FIGURE 2 Genomic and proteomic features of PTGs. (A) Intron/exon architecture of PTGs in Mangifera indica. The green lines indicate introns. The upstream and downstream regions of genes are represented in red color. The exons are represented by green shapes. (B) Prediction of conserved motifs PTG proteins from M. indica. The conserved motifs are shown in this ﬁgure along with their motif details. FIGURE 3 Most conserved domain of K+ transporting proteins in Mangifera indica. FIGURE 3 Most conserved domain of K+ transporting proteins in Mangifera indica. FIGURE 3 Most conserved domain of K+ transporting proteins in Mangifera indica. 08 Frontiers in Plant Science frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 MiKEA3, MiKEA5, MiHAK8, MiHAK5.2, MiHAK5.1, MiKAT3, and MiTPK4), and upregulated in root and downregulated in leaves (MiHAK5, MiTPK6.1, MiAKT1.1, MiTPK1.2, MiHAK1, MiHAK6.1, and MiHAK2.1), while the rest of the genes were non-responsive (Figure 5A). Interestingly, MiTPK1.2, MiHAK1, MiHAK2.1, HAK6.1, and MiAKT1.1 were most upregulated in roots, and MiKEA2, MiAKT2, and MiAKT1 were upregulated in leaves. MiKEA3, MiKEA5, MiHAK8, MiHAK5.2, MiHAK5.1, MiKAT3, and MiTPK4), and upregulated in root and downregulated in leaves (MiHAK5, MiTPK6.1, MiAKT1.1, MiTPK1.2, MiHAK1, MiHAK6.1, and MiHAK2.1), while the rest of the genes were non-responsive (Figure 5A). 3.6 Expression proﬁling based on NGS data analysis of drought stress response and fruit development To check the gene expression proﬁle of PTGs in M. indica leaﬂets and roots under drought stress, the gene expression data of M. indica were downloaded from the NCBI-SRA database (Figure 5A). The differentially regulated genes were either upregulated in both leaves and roots (MiKAT2.1, MiHAK2, MiTPK5, MiHKT1, MiHKT7, MiTPK1.1, MiHAK12, MiHAK1.1, MiAKT6, and MiHKT1), upregulated in leaves and downregulated in the root (MiAKT1, MiKEA2, MiAKT2, RNA-seq data were also used to analyze the gene expression proﬁle of PTGs during the ﬁve stages of fruit ripening in M. indica. SRA data for the bioProject (PRJNA797728) were retrieved from the NCBI-SRA database. Gene expression proﬁles were divided into two groups, i.e., those genes that were upregulated in ﬁve different stages of fruit ripening (MiAKT6, MiHAK1.1, MiKAT2, MiKAT2.1, MiHKT1, MiTPK1.1, MiHAK7, and MiHAK12) and those genes that FIGURE 4 Cis-regulatory elements within 2,000-bp promoter region of PTGs. FIGURE 4 Cis-regulatory elements within 2,000-bp promoter region of PTGs. 09 Frontiers in Plant Science frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 A B FIGURE 5 (A) Proﬁles of PTG expression in leaves and roots of Mangifera indica. The expression values [log2(FPKM)] are shown by the colored bar. Red denotes high expression, yellow denotes no expression, and blue represents low expression. (B) Proﬁles of PTG expression during fruit development. The expression values [log2(FPKM)] are shown by the colored bar. Red denotes high expression, whereas blue denotes low expression. B A B FIGURE 5 (A) Proﬁles of PTG expression in leaves and roots of Mangifera indica. The expression values [log2(FPKM)] are shown by the colored bar. Red denotes high expression, yellow denotes no expression, and blue represents low expression. (B) Proﬁles of PTG expression during fruit development. The expression values [log2(FPKM)] are shown by the colored bar. Red denotes high expression, whereas blue denotes low expression. were highly downregulated in leaves and fruit ripening according to FPKM values (MiHAK1, MiHAK5.1, MiHAK5, MiHAK5.2, and MiTPK1.2), and the rest of the genes showed non-responsive or non-signiﬁcant variations of expression (Figure 5B). plant growth and stress conditions were applied, such variations are potentially associated with sampling time in a day, plant age, or even genotypic differences. 4.2 Potassium transport system is involved in drought stress response in M. indica by the fact that various sequence features were common in M. indica and other species. For example, the transmembrane domains of MiHKT1.1 and AtHKT1;1 are eight. Similarly, the presence of the consensus motif GVVYGDLGTSPLY is a characteristic feature of HAK transporters (Cheng et al., 2018). This motif was present with a minor modiﬁcation as GVVYGDLG(I/T)SPLY. This variation of a motif is also present in HAK proteins in A. thaliana and Camellia sinensis (Yang et al., 2020). KEA members of both M. indica and A. thaliana species showed conservation of “FLLFxxGLE and GEFAFVxxxxA” motifs. Previously, it has been established that a divergence in the amino acid sequences of proteins is related to the functional divergence of proteins and vice versa (Sangar et al., 2007). Among K+ channels, the presence of ANK, Ion_trans_2, and KHA domains in the Shaker proteins, similar to Arabidopsis K+ channels, suggests their functional similarities (Keisham et al., 2018). Likewise, the occurrence of the signiﬁcantly conserved residues, i.e., GYGD and RSXpS/pTXP, in MiTPKs indicates potential functional similarities in M. indica and other plants (Ge et al., 2020; Siddique et al., 2021). by the fact that various sequence features were common in M. indica and other species. For example, the transmembrane domains of MiHKT1.1 and AtHKT1;1 are eight. Similarly, the presence of the consensus motif GVVYGDLGTSPLY is a characteristic feature of HAK transporters (Cheng et al., 2018). This motif was present with a minor modiﬁcation as GVVYGDLG(I/T)SPLY. This variation of a motif is also present in HAK proteins in A. thaliana and Camellia sinensis (Yang et al., 2020). KEA members of both M. indica and A. thaliana species showed conservation of “FLLFxxGLE and GEFAFVxxxxA” motifs. Previously, it has been established that a divergence in the amino acid sequences of proteins is related to the functional divergence of proteins and vice versa (Sangar et al., 2007). Among K+ channels, the presence of ANK, Ion_trans_2, and KHA domains in the Shaker proteins, similar to Arabidopsis K+ channels, suggests their functional similarities (Keisham et al., 2018). Likewise, the occurrence of the signiﬁcantly conserved residues, i.e., GYGD and RSXpS/pTXP, in MiTPKs indicates potential functional similarities in M. indica and other plants (Ge et al., 2020; Siddique et al., 2021). As one of the most abundant cations in plant cells, K+ performs a dominant role in various biological processes of plants (Ragel et al., 2019). 4.1 Potassium transport system is highly conserved in M indica and other plants denotes signiﬁcant differences between environmental stressors and the control (p 0.05), whereas denotes extremely signiﬁcant differences (p 0.001). FIGURE 6 K+ transporting gene family relative qRT-PCR in response to drought. To produce an objective average value, the experiment was triplicated. In untreated plants, each gene had a default expression value of 1. Bars have been placed on each column to illustrate the standard error. denotes signiﬁcant differences between environmental stressors and the control (p 0.05), whereas *denotes extremely signiﬁcant differences (p 0.001). FIGURE 6 K+ transporting gene family relative qRT-PCR in response to drought. To produce an objective average value, the experiment was triplicated. In untreated plants, each gene had a default expression value of 1. Bars have been placed on each column to illustrate the standard error. denotes signiﬁcant differences between environmental stressors and the control (p 0.05), whereas **denotes extremely signiﬁcant differences (p 0.001). 4.1 Potassium transport system is highly conserved in M indica and other plants Real-time ampliﬁcation (qRT-PCR) was performed to conﬁrm the expression of nine selected PTGs in M. indica leaves. These genes were selected based on their expression values in drought stress response (Figure 5A) and higher expression during fruit development (Figure 5B). In response to drought stress, nine genes, including MiAKT6, MiHAK1.1, MiKAT2.1, MiHKT1, MiTPK1.1, MiHAK1, MiHAK5.1, MiHAK5, and MiHAK5.2, were selected for qPCR-based quantiﬁcation for 7 days, and samples were taken at the second, fourth, and seventh days. Under drought stress, MiHAK1 was regulated differently. Although according to RNA-seq analysis MiHAK1 was downregulated in response to drought stress, it was found to be upregulated up to 3.5-fold in qPCR. Similarly, MiHAK1.1, MiHAK5.2, MiTPK1.1, MiKAT2.1, and MiAKT6 were signiﬁcantly upregulated after drought stress. On the contrary, MiHAK5, MiHAK5.1, and MiHKT1 were signiﬁcantly downregulated (Figure 6). There was a signiﬁcant correlation between the results of qPCR and RNA-seq data. However, some variations were also observed. Though similar The current study identiﬁed 37 potential genes of the K+ transport system in M. indica genome. The number of PTGs is more or less similar in major plant groups represented by V. vinifera, Glycine max, Triticum aestivum, Vigna radiata, C. cajan, and A. thaliana (Cuéllar et al., 2010; Isayenkov et al., 2011; Chen et al., 2017; Azeem et al., 2018; Azeem et al., 2021a; Azeem et al., 2021b; Ge et al., 2020; Siddique et al., 2021; Cai et al., 2021). It not only indicates the evolutionary conservation of PTGs in these plants but also predicts the existence of similar genetic mechanisms for K+ in other fruit trees. A total of 202 PTG protein sequences from ﬁve species—A. thaliana, M. indica, M. domestica, O. sativa, and C. arietinum—were divided into six groups according to phylogenetic analyses (HAK, HKT, KEA, AKT and KAT, GORK and SKOR, and TPK). In contrast, the phylogenetic relationship analysis showed that PTGs were more closely related to MdPTGs than AtPTGs. This outcome conﬁrmed the ﬁnding that apple and mango also showed a closer association than Arabidopsis. It is also supported Frontiers in Plant Science 10 frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 FIGURE 6 K+ transporting gene family relative qRT-PCR in response to drought. To produce an objective average value, the experiment was triplicated. In untreated plants, each gene had a default expression value of 1. Bars have been placed on each column to illustrate the standard error. Publisher’s note In the current study, the K+ transport system was characterized in M. indica. There are 37 potential members of PTGs in this plant. Among these, 22 genes can be classiﬁed as K+ transporters, and the remaining 15 genes are K+ channels. Moreover, the analysis performed including conserved domain and TM domain prediction, motif analysis, phylogenetic analysis, and genetic structure display, and cis-regulatory elements predicted in the promoter regions revealed the close relation of PTGs in several plant species. Differentially expressed genes MiHKT7, MiHAK5.1, MiHAK5, MiTPK1.1, and MiHAK12 were responsive to drought stress. Moreover, MiHKT7, MiHAK5.1, MiHAK5, MiTPK1.1, and MiHAK12 genes were also differentially expressed during different growth stages of mango fruit development. To gain depth of information about the K+ transport system, these genes can be selected for further genomic and biotechnological study and to improve stress response and fruit quality in mango. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Funding This research was funded by the Hainan Provincial Natural Science Foundation of China (grant number 321RC617). Conﬂict of interest The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. 4.2 Potassium transport system is involved in drought stress response in M. indica The inﬂux/efﬂux of K+ (mediated by channel and transporter proteins) regulates its concentration in the plant body. K+ channels and transporters are important contributors to plant growth and development (Sharma et al., 2013). In the current study, members of HAK, HKT, TPK, and Shaker families were differentially regulated by drought stress. These results were more or less in agreement with RNA-seq and qPCR analyses. Previously, the role of potassium channels and transporters has been documented in the drought stress response of various plants (Ahmad et al., 2016a; Ahmad et al., 2016b; Chen et al., 2017; Hassan et al., 2017; Cai et al., 2019; Qi et al., 2019; Singh et al., 2021). The overexpression of the OsAKT1 and OsHAK1 signiﬁcantly affects potassium nutrition and Frontiers in Plant Science 11 frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 Data availability statement drought stress tolerance of rice (Ahmad et al., 2016a; Chen et al., 2017). Our results also proposed similar roles of K+ transporters. Overexpression of the potassium channel TPKb in small vacuoles confers osmotic and drought tolerance to rice (Ahmad et al., 2016b). Taken together, these ﬁndings suggest a potential role of PTGs in the stress response of the M. indica tree. The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material. SUPPLEMENTARY FIGURE 1 Distribution of PTGs on M. indica on chromosomes. The 37 PTGs were mapped onto all chromosomes except 3, 7, 11, and 16. Each gene is represented by a speciﬁc color. 4.3 M. indica fruit development involves differential regulation of PTGs Conceptualization: FA. Data curation: TL and MW. Formal analysis: TL and AR. Funding acquisition: FA and TL. Investigation: TL and MR. Methodology, MW and AR. Project administration: FA. Resources: FA. Software: MW and AR. Supervision: FA. Validation: FA, TL, and HM. Visualization: MW. Writing—original draft: TL. Writing—review and editing: FA and HM. All authors contributed to the article and approved the submitted version. The essential nutrients (N, P, and K) promote an increase in biomass yield by improving root growth, activating cellular enzymes, ameliorating photosynthesis, conserving energy, and positively affecting other fundamental processes in the plant body. Potassium sources can inﬂuence the K+ release and fruit yield of M. indica (Wang et al., 2022). It emphasizes the importance of potassium uptake and transport system from root to leaves and fruit. The current study reports signiﬁcant differential regulation of K+ transport during ﬁve stages of mango fruit development. Previously, it is reported that supplemental foliar applications of K+ can improve fruit quality in Cucumis melo (Lester et al., 2005), Solanum lycopersicum (Liu et al., 2021), C. sinensis (Wu et al., 2021), and M. indica (Sarker and Rahim, 2013). The ﬁndings of the current study complement the potential role of PTGs in improving mango fruit growth, development, and quality. References FAOSTAT (2022). Production of mangoes, mangosteens, and guavas in 2020, Crops/Regions/World list/Production quantity (pick lists) (UN Food and Agriculture Organization, Corporate Statistical Database (FAOSTAT). (Rome, Italy: Viale delle Terme di Caracalla) Ahmad, I., Devonshire, J., Mohamed, R., Schultze, M., and Maathuis, F. J. M. (2016a). Overexpression of the potassium channel TPKb in small vacuoles confers osmotic and drought tolerance to rice. New Phytol. 209, 1040–1048. doi: 10.1111/ nph.13708 Farooq, M., Hussain, M., and Siddique, K. H. (2014). Drought stress in wheat during ﬂowering and grain-ﬁlling periods. Crit. Rev. Plant Sci. 33 (4), 331–349. doi: 10.1080/07352689.2014.875291 Ahmad, I., Mian, A., and Maathuis, F. J. M. (2016b). Overexpression of the rice AKT1 potassium channel affects potassium nutrition and rice drought tolerance. J. Exp. Bot. 67, 2689–2698. doi: 10.1093/jxb/erw103 Galán Saúco, V. (2002). “Mango production and world market: Current situation and future prospects,” in VII international mango symposium. ISHS Acta Horticulturae. 645, 107–116. Aranda-Sicilia, M. N., Cagnac, O., Chanroj, S., Sze, H., Rodrı́guez-Rosales, M. P., and Venema, K. (2012). Arabidopsis KEA2, a homolog of bacterial KefC, encodes a K+/H+ antiporter with a chloroplast transit peptide. Biochim. Biophys. Acta -Biomembranes. 1818 (9), 2362–2371. doi: 10.1016/j.bbamem.2012.04.011 Garg, V. K., Avashthi, H., Tiwari, A., Jain, P. A., Ramkete, P. W., Kayastha, A. M., et al. (2016). MFPPI–multi FASTA ProtParam interface. Bioinformation 12 (2), 74. doi: 10.6026/97320630012074 Assaha, D. V., Mekawy, A. M. M., Ueda, A., and Saneoka, H. (2015). Salinity- induced expression of HKT may be crucial for na+ exclusion in the leaf blade of huckleberry (Solanum scabrum mill.), but not of eggplant (Solanum melongena l.). Biochem. Biophys. Res. Commun. 460 (2), 416–421. doi: 10.1016/j.bbrc. 2015.03.048 Ge, M., Zhang, L., Ai, J., Ji, R., He, L., Liu, C., et al (2020). Effect of heat shock and potassium sorbate treatments on gray mold and postharvest quality of ‘XuXiang’ kiwifruit. Food Chem. 324, 126891. doi: 10.1016/j.foodchem.2020.126891. Azeem, F., Ahmad, B., Atif, R. M., Ali, M. A., Nadeem, H., Hussain, S., et al. (2018). Genome-wide analysis of potassium transport-related genes in chickpea (Cicer arietinum l.) and their role in abiotic stress responses. Plant Mol. Biol. Rep. 36, 451–468. doi: 10.1007/s11105-018-1090-2 González, W., Valdebenito, B., Caballero, J., Riadi, G., Riedelsberger, J., Martı́nez, G., et al. (2015). K2P channels in plants and animals. Pﬂügers. Archiv- European. J. Physiol. 467 (5), 1091–1104. doi: 10.1007/s00424-014-1638-4 Goodstein, D. M., Shu, S., Howson, R., Neupane, R., Hayes, R. D., Fazo, J., et al. (2012). References Phytozome: a comparative platform for green plant genomics. Nucleic Acids Res. 40 (D1), D1178–D1186. doi: 10.1093/nar/gkr944 Azeem, F., Hussain, M., Hussain, S., Zubair, M., Nadeem, H., Ali, M. A., et al. (2021a). Genome-wide analysis and expression proﬁling of potassium transport related genes in solanum tuberosum. Pakistan J. Agric. Sci. 58, 81–94. doi: 10.3390/ plants11010002 Guo, A.-Y., Zhu, Q.-H., Chen, X., and Luo, J.-C. (2007). GSDS: a gene structure display server. Yi. chuan=. Hereditas. 29 (8), 1023–1026. doi: 10.1360/yc-007-1023 Azeem, F., Ijaz, U., Ali, M. A., Hussain, S., Zubair, M., Manzoor, H., et al. (2021b). Genome-wide identiﬁcation and expression proﬁling of potassium transport-related genes in vigna radiata under abiotic stresses. Plants 11 (1), 2. doi: 10.3390/plants11010002 Hasanuzzaman, M., Bhuyan, M. B., Nahar, K., Hossain, M. S., Mahmud, J. A., Hossen, M. S., et al. (2018). Potassium: a vital regulator of plant responses and tolerance to abiotic stresses. Agronomy. 8, 31. doi: 10.3390/agronomy8030031 Hossain, A., Skalicky, M., Brestic, M., Maitra, S., Ashraful Alam, M., Syed, M. A., et al. (2021). Consequences and mitigation strategies of abiotic stresses in wheat (Triticum aestivum l.) under the changing climate. Agronomy 11 (2), 241. doi: 10.3390/agronomy11020241 Azeem, F., Zameer, R., Rashid, M. A. R., Rasul, I., Ul-Allah, S., Siddique, M. H., et al. (2022). Genome-wide analysis of potassium transport genes in gossypium raimondii suggest a role of GrHAK/KUP/KT8, GrAKT2. 1 and GrAKT1. 1 in response to abiotic stress. Plant Physiol. Biochem. 170, 110–122. doi: 10.1016/ j.plaphy.2021.11.038 Hassan, M. U., Aamer, M., Chattha, M. U., Aman Ullah, M., Sulaman, S., Nawaz, M., et al. (2017). The Role of Potassium in Plants under Drought Stress: Mini Review. J. Basic Appl. Sci. 13, 268–271. doi: 10.6000/1927-5129.2017.13.44. Bahadur, I., Maurya, R., Roy, P., and Kumar, A. (2019). “Potassium-solubilizing bacteria (KSB): a microbial tool for K-solubility, cycling, and availability to plants,” in Plant growth promoting rhizobacteria for agricultural sustainability (Springer), 257–265. doi: 10.1007/978-981-13-7553-8_13 Isayenkov, S., Isner, J.-C., and Maathuis, F. J. M. (2011). Rice two-pore k+ channels are expressed in different types of vacuoles. Plant Cell 23, 756–768. doi: 10.1105/tpc.110.081463 Bailey, T. L., Boden, M., Buske, F. A., Frith, M., Grant, C. E., Clementi, L., et al. (2009). MEME SUITE: tools for motif discovery and searching. Nucleic Acids Res. 37 (suppl_2), W202–W208. doi: 10.1093/nar/gkp335 Johnson, M., Zaretskaya, I., Raytselis, Y., Merezhuk, Y., McGinnis, S., and Madden, T. L. (2008). NCBI BLAST: a better web interface. Nucleic Acids Res. 36 (suppl_2), W5–W9. References doi: 10.1093/nar/gkn201 Cai, K., Gao, H., Wu, X., Zhang, S., Han, Z., Chen, X., et al. (2019). The ability to regulate transmembrane potassium transport in root is critical for drought tolerance in barley. Int. J. Mol. Sci. 20, 4111. doi: 10.3390/ijms20174111 Keisham, M., Mukherjee, S., and Bhatla, S. C. (2018). Mechanisms of sodium transport in plants–progresses and challenges. Int. J. Mol. Sci. 19, 647. doi: 10.3390/ ijms19030647 Cai, K., Zeng, F., Wang, J., and Zhang, G. (2021). Identiﬁcation and characterization of HAK/KUP/KT potassium transporter gene family in barley and their expression under abiotic stress. BMC Genomics 22 (1), 1–14. doi: 10.1186/ s12864-021-07633-y Laurio, M. V. O. (2021). Integrating vibratory membrane-based water recovery systems for sustainable food and beverage production (Glassboro, New Jersey: Rowan University). Celestin, A. (2019). Bridging the gap between smallholder farmers and market access through agricultural value chain development in Haiti. Agronomy. Lescot, M., Déhais, P., Thijs, G., Marchal, K., Moreau, Y., Van de Peer, Y., et al. (2002). PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences. Nucleic Acids Res. 30 (1), 325– 327. doi: 10.1093/nar/30.1.325 Cheng, X., Liu, X., Mao, W., Zhang, X., Chen, S., Zhan, K., et al. (2018). Genome- wide identiﬁcation and analysis of HAK/KUP/KT potassium transporters gene family in wheat (Triticum aestivum l.). Int. J. Mol. Sci. 19, 3969. doi: 10.3390/ ijms19123969 Lester, G. E., Jifon, J. L., and Rogers, G. (2005). Supplemental foliar potassium applications during muskmelon fruit development can improve fruit quality, ascorbic acid, and beta-carotene contents. J. Am. Soc Hortic. Sci. 130, 649–653. doi: 10.21273/JASHS.130.4.649 Chen, G., Liu, C., Gao, Z., Zhang, Y., Jiang, H., Zhu, L., et al. (2017). OsHAK1, a high-afﬁnity potassium transporter, positively regulates responses to drought stress in rice. Front. Plant Sci. 8, 1885. doi: 10.3389/fpls.2017.01885 Liu, J., Hu, T., Feng, P., Yao, D., Gao, F., and Hong, X. (2021). Effect of potassium fertilization during fruit development on tomato quality, potassium uptake, water and potassium use efﬁciency under deﬁcit irrigation regime. Agric. Water Manage. 250, 106831. doi: 10.1016/j.agwat.2021.106831 Cleveland, R. P. (2012). Second report on the state of plant genetic resources for food and agriculture in guyana. Agronomy Science. (EC Demerara: National Agricultural Research and Extension Institute Mon Repos) Livak, K. J., and Schmittgen, T. D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2–DDCT method. Methods 25, 402– 408. Supplementary material The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/ fpls.2022.1102201/full#supplementary-material Frontiers in Plant Science 12 frontiersin.org frontiersin.org 10.3389/fpls.2022.1102201 Tan et al. References Roles of tandem-pore k+ channels in plants–a puzzle still to be solved. Plant Biol. 12, 56–63. doi: 10.1111/j.1438-8677.2010.00353.x Riedelsberger, J., Miller, J. K., Valdebenito-Maturana, B., Piñeros, M. A., González, W., and Dreyer, I. (2021). Plant HKT channels: an updated view on structure, function and gene regulation. Int. J. Mol. Sci. 22 (4), 1892. doi: 10.3390/ ijms22041892 Wang, J., Ding, Z., AL-Huqail, A. A., Hui, Y., He, Y., Reichman, S. M., et al. (2022). Potassium source and biofertilizer inﬂuence K release and fruit yield of mango (Mangifera indica l.): A three-year ﬁeld study in sandy soils. Sustainability 14 (15), 9766. doi: 10.3390/su14159766 Sahi, N., Mostajeran, A., and Ghanadian, M. (2022). Changing in the production of anticancer drugs (vinblastine and vincristine) in catharanthus roseus (L.) g. don by potassium and ascorbic acid treatments. Plant. Soil Environ. 68 (1), 18–28. doi: 10.17221/121/2021-PSE Wang, C., Yamamoto, H., Narumiya, F., Munekage, Y. N., Finazzi, G., Szabo, I., et al. (2017). Fine-tuned regulation of the K+/H+ antiporter KEA 3 is required to optimize photosynthesis during induction. Plant J. 89 (3), 540–553. doi: 10.1111/ tpj.13405 Sangar, V., Blankenberg, D. J., Altman, N., and Lesk, A. M. (2007). Quantitative sequence-function relationships in proteins based on gene ontology. BMC Bioinf. 8, 294. doi: 10.1186/1471-2105-8-294 Waters, S., Gilliham, M., and Hrmova, M. (2013). Plant high-afﬁnity potassium (HKT) transporters involved in salinity tolerance: structural insights to probe differences in ion selectivity. Int. J. Mol. Sci. 14 (4), 7660–7680. doi: 10.3390/ ijms14047660 Sarker, B. C., and Rahim, M. (2013). Yield and quality of mango (Mangifera indica l.) as inﬂuenced by foliar application of potassium nitrate and urea. Bangladesh. J. Agric. Res. 38, 145–154. doi: 10.3329/bjar.v38i1.15201 Schultz, J., Copley, R. R., Doerks, T., Ponting, C. P., and Bork, P. (2000). SMART: a web-based tool for the study of genetically mobile domains. Nucleic Acids Res. 28 (1), 231–234. doi: 10.1093/nar/28.1.231 Wu, S., Zhang, C., Li, M., Tan, Q., Sun, X., Pan, Z., et al. (2021). Effects of potassium on fruit soluble sugar and citrate accumulations in cara cara navel orange (Citrus sinensis l. osbeck). Sci. Hortic. (Amsterdam). 283, 110057. doi: 10.1016/j.scienta.2021.110057 Sharma, T., Dreyer, I., and Riedelsberger, J. (2013). The role of k+ channels in uptake and redistribution of potassium in the model plant arabidopsis thaliana. Front. Plant Sci. 4. doi: 10.3389/fpls.2013.00224 Yang, T., Lu, X., Wang, Y., Xie, Y., Ma, J., Cheng, X., et al. (2020). References doi: 10.1006/meth.2001.1262 Council, N. R. (1993). Sustainable agriculture and the environment in the humid tropics (Washington, DC: The National Academies Press). doi: 10.17226/1985. Cuéllar, T., Pascaud, F., Verdeil, J.-L., Torregrosa, L., Adam-Blondon, A.-F., Thibaud, J.-B., et al. (2010). A grapevine shaker inward k(+) channel activated by the calcineurin b-like calcium sensor 1-protein kinase CIPK23 network is expressed in grape berries under drought stress conditions. Plant J. 61, 58–69. doi: 10.1111/ j.1365-313X.2009.04029.x Mehanny, M., Kroniger, T., Koch, M., Hoppstädter, J., Becher, D., Kiemer, A. K., et al. (2022). Yields and immunomodulatory effects of pneumococcal membrane vesicles differ with the bacterial growth phase. Adv. Healthcare. Mater. 11 (5), 2101151. doi: 10.1002/adhm.202101151 Nahar, K., Hasanuzzaman, M., Ahamed, K. U., Hakeem, K. R., Ozturk, M., and Fujita, M. (2015). “Plant responses and tolerance to high temperature stress: role of exogenous phytoprotectants,” in Crop production and global environmental issues (Switzerland: Springer International Publishing), 385–435. Etesami, H., Emami, S., and Alikhani, H. A. (2017). Potassium solubilizing bacteria (KSB):: Mechanisms, promotion of plant growth, and future prospects a review. J. Soil Sci. Plant Nutr. 17 (4), 897–911. doi: 10.4067/S0718- 95162017000400005 Frontiers in Plant Science 13 frontiersin.org Tan et al. 10.3389/fpls.2022.1102201 rootstock to scion in grafted cucumber. Int. J. Mol. Sci. 19 (9), 2648. doi: 10.3390/ ijms19092648 Ngamchuachit, P., Sivertsen, H. K., Mitcham, E. J., and Barrett, D. M. (2015). Inﬂuence of cultivar and ripeness stage at the time of fresh-cut processing on instrumental and sensory qualities of fresh-cut mangos. Postharvest. Biol. Technol. 106, 11–20. doi: 10.1016/j.postharvbio.2015.03.013 Sze, H., and Chanroj, S. (2018). Plant endomembrane dynamics: studies of K+/H + antiporters provide insights on the effects of pH and ion homeostasis. Plant Physiol. 177 (3), 875–895. doi: 10.1104/pp.18.00142 Qi, J., Sun, S., Yang, L., Li, M., Ma, F., and Zou, Y. (2019). Potassium uptake and transport in apple roots under drought stress. Hortic. Plant J. 5, 10–16. doi: 10.1016/j.hpj.2018.10.001 Ul Hassan, M., Rasool, T., Iqbal, C., Arshad, A., Abrar, M., Abrar, M. M., et al. (2021). Linking plants functioning to adaptive responses under heat stress conditions: a mechanistic review. J. Plant Growth Regul. 41, 2596"C2613. doi: 10.1007/s00344-021-10493-1 Ragel, P., Raddatz, N., Leidi, E. O., Quintero, F. J., and Pardo, J. M. (2019). Regulation of k+ nutrition in plants. Front. Plant Sci. 10. doi: 10.3389/ fpls.2019.00281 Voelker, C., Gomez-Porras, J., Becker, D., Hamamoto, S., Uozumi, N., Gambale, F., et al. (2010). References HAK/KUP/KT family potassium transporter genes are involved in potassium deﬁciency and stress responses in tea plants (Camellia sinensis l.): expression and functional analysis. BMC Genomics 21, 556. doi: 10.1186/s12864-020-06948-6 Siddique, M. H., Babar, N. I., Zameer, R., Muzammil, S., Nahid, N., Ijaz, U., et al. (2021). Genome-wide identiﬁcation, genomic organization, and characterization of potassium transport-related genes in cajanus cajan and their role in abiotic stress. Plants 10, 2238. doi: 10.3390/plants10112238 Yao, R., Huang, X., Cong, H., Qiao, F., Cheng, Y., and Chen, Y. (2022). Selection and identiﬁcation of a reference gene for normalizing real-time PCR in mangos under various stimuli in different tissues. Horticulturae 8, 882. doi: 10.3390/ horticulturae8100882 Singh, S., Kumar, V., Parihar, P., Dhanjal, D. S., Singh, R., Ramamurthy, P. C., et al. (2021). Differential regulation of drought stress by biological membrane transporters and channels. Plant Cell Rep. 40, 1565–1583. doi: 10.1007/s00299-021- 02730-4 Ye, C.-Y., Yang, X., Xia, X., and Yin, W. (2013). Comparative analysis of cation/ proton antiporter superfamily in plants. Gene 521 (2), 245–251. doi: 10.1016/ j.gene.2013.03.104 Singh, K., and Kushwaha, C. (2006). Diversity of ﬂowering and fruiting phenology of trees in a tropical deciduous forest in India. Ann. Bot. 97 (2), 265– 276. doi: 10.1093/aob/mcj028 Zhu, X., Pan, T., Zhang, X., Fan, L., Quintero, F. J., Zhao, H., et al. (2018). K+ efﬂux antiporters 4, 5, and 6 mediate pH and k+ homeostasis in endomembrane compartments. Plant Physiol. 178 (4), 1657–1678. doi: 10.1104/pp.18.01053 Snowden, M. C., Ritchie, G. L., Simao, F. R., and Bordovsky, J. P. (2014). Timing of episodic drought can be critical in cotton. Agron. J. 106 (2), 452–458. doi: 10.2134/agronj2013.0325 Zikki, K. A. F. (2020). Quality of bell pepper (Capsicum annuum l.) affected by drought condition: a thesis presented in the partial fulﬁlment of the requirements for the degree of master of philosophy at Massey university, palmerston north, new Zealand (New Zealand: Massey University). Sun, J., Cao, H., Cheng, J., He, X., Sohail, H., Niu, M., et al. (2018). Pumpkin CmHKT1; 1 controls shoot na+ accumulation via limiting na+ transport from Frontiers in Plant Science 14 frontiersin.org
https://openalex.org/W4241196805	https://pure.mpg.de/pubman/item/item_3281274_5/component/file_3285813/2020MS002101.pdf	English	null	The German Climate Forecast System: GCFS	null	2,020	cc-by	10,633	© 2021. The Authors. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. Citation: Plain Language Summary Information about the expected departure from the “normal” climatic conditions of an upcoming season would be tremendously valuable for many sectors of society. In Germany, three institutes join their expertise to build a climate forecast system using the Earth system model of the Max Planck Institute for Meteorology. This model describes the atmosphere, land and rivers as well as the ocean and sea ice. The model describes their interactions and is well designed for climate studies on a much longer timescale than a season. Max Planck Institute for Meteorology, Universität Hamburg and the German Meteorological Service Deutscher Wetterdienst have developed the methods those are necessary for such a forecast system and operationally perform the seasonal predictions. This paper compares two versions of our forecast system. The forecast quality during different seasons is particularly investigated. The expectation that the second model system is much better than the first system is not entirely fulfilled. We discuss possible reasons and suggest a stronger focus on the model quality for interannual variability for future model development. Received 12 MAR 2020 Accepted 29 DEC 2020 10.1029/2020MS002101 RESEARCH ARTICLE Kristina Fröhlich1 , Mikhail Dobrynin1,2 , Katharina Isensee1 , Claudia Gessner3,4 , Andreas Paxian1, Holger Pohlmann1,5 , Helmuth Haak5 , Sebastian Brune2 , Barbara Früh1 , and Johanna Baehr2 Kristina Fröhlich1 , Mikhail Dobrynin1,2 , Katharina Isensee1 , Claudia Gessner3,4 , Andreas Paxian1, Holger Pohlmann1,5 , Helmuth Haak5 , Sebastian Brune2 , Barbara Früh1 , and Johanna Baehr2 The German Climate Forecast System: GCFS Key Points: • A climate forecast system is developed on the basis of a coupled climate model 1Deutscher Wetterdienst, Offenbach, Germany, 2CEN, Universität Hamburg, Hamburg, Germany, 3Goethe Universität Frankfurt, Frankfurt, Germany, 4Now at ETH, Zurich, Switzerland, 5Max Planck Institute for Meteorology, Hamburg, Germany • The comparison of the first to the second version shows improvements in North Atlantic Oscillation forecasts. Forecast skill in tropical regions is not improved • The comparison of the first to the second version shows improvements in North Atlantic Oscillation forecasts. Forecast skill in tropical regions is not improved Abstract Seasonal prediction is one important element in a seamless prediction chain between weather forecasts and climate projections. After several years of development in collaboration with Universität Hamburg and Max Planck Institute for Meteorology, the Deutscher Wetterdienst performs operational seasonal forecasts since 2016 with the German Climate Forecast System, now in Version 2 (GCFS2.0). Here, the configuration of the previous system GCFS1.0 and the current GCFS2.0 are described and the performance of the two systems is compared over the common hindcast period of 1990– 2014. In GCFS2.0, the forecast skill is improved compared to GCFS1.0 during boreal winter, especially for the Northern Hemisphere where the Pearson correlation has increased for the North Atlantic Oscillation index. Overall, a similar performance of GCFS2.0 in comparison to GCFS1.0 is assessed during the boreal summer. Future developments for climate forecasts need a stronger focus on the performance of interannual variability in a model system. Correspondence to: K. Fröhlich, kristina.froehlich@dwd.de Citation: Fröhlich, K., Dobrynin, M., Isensee, K., Gessner, C., Paxian, A., Pohlmann, H., et al. (2021). The german climate forecast system: GCFS. Journal of Advances in Modeling Earth Systems, 13, e2020MS002101. https://doi. org/10.1029/2020MS002101 Received 12 MAR 2020 Accepted 29 DEC 2020 Fröhlich, K., Dobrynin, M., Isensee, K., Gessner, C., Paxian, A., Pohlmann, H., et al. (2021). The german climate forecast system: GCFS. Journal of Advances in Modeling Earth Systems, 13, e2020MS002101. https://doi. org/10.1029/2020MS002101 2.1. The Seasonal Forecast System Performing forecasts with a climate model requires the development of an appropriate workflow. This in- cludes the provision with initial conditions from reference data, the selection of suited methods of assimila- tion, as well as the generation of an ensemble and the production of hindcasts and forecasts. As an example, Figure 1 sketches the workflow of the seasonal forecast system GCFS2.0 from the generation of the initial conditions to the hindcast and forecast ensembles. This figure and a simple description of the system can also be found under https://www.dwd.de/EN/ourservices/seasonals_forecasts/project_description.html. These tasks will be explained in the following subsections. Details about the two Earth System Model con- figurations are provided and discussed afterwards, summarized in Table 1. Journal of Advances in Modeling Earth Systems Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 This balanced model state serves as a basis for any study on introduced anthropogenic changes within the following centuries. With increasing model resolution, a challengingly long computational time is required to achieve an equilibrium model state. At Universität Hamburg, assimilation and ensemble generation methods are developed for the use in GCFS. DWD adapts and maintains the whole system for operational performance and issues the seasonal forecasts. In the following, we will describe the configuration of the two versions GCFS1.0 and GCFS2.0, analyze both systems concerning the representation and prediction skill of the North Atlantic Oscillation (NAO), the El Niño Southern Oscillation (ENSO), surface temperature and geopotential height at 500 hPa and compare them with the focus on differences in the model climate and the respective hindcast skill. We will discuss GCFS' strengths and weaknesses and possible approaches for the future development and conclude with our main findings. 1. Introduction Over the last two decades, seasonal climate predictions have evolved from a scientific research topic into full-fledged operational systems. Today, seasonal prediction systems are most operationally run at large weather centers like ECMWF (European Center for Medium-Range Weather Forecasts), see Johnson et al. (2019) or the British Met Office (MacLachlan et al. 2015), issuing real-time seasonal forecasts. Thirteen global producing centers (GPC's) currently submit their long-range forecasts to the World Meteorological Organization WMO lead center in South Korea (https://www.wmolc.org/). Since 2011, the German Meteor- ological Service Deutscher Wetterdienst (DWD), Universität Hamburg (UHH) and the Max Planck Institute for Meteorology (MPI-M) have been developing a system for seasonal climate prediction (Baehr et al., 2015). Since October 2016, the German Climate Forecast System (GCFS) operationally produces seasonal predic- tions, which are published every month at DWD's homepage. In 2017, DWD became the 13th GPC of the WMO's multimodel ensemble for long-range seasonal forecasts. At the Max Planck Institute, the Earth-System Model (MPI-ESM; Giorgetta et al., 2013; Mauritsen et al., 2018) is developed. MPI-ESM is tuned over many model-years under pre-industrial conditions (referring to climate conditions of 1850), until it is decided that the best match of the known state of the Earth's climate system is found while keeping the balance of the atmosphere's radiation (Mauritsen et al., 2012). 1 of 17 FRÖHLICH ET AL. 2.2. Assimilation Continuous nudging is used to bring the model's state close to the (re)analyzed state of the climate system. This means a separate simulation runs continuously over the years and months under nudging conditions. Its restart files produced at the end of each month serve as the initial conditions for the retrospective (so- called hindcasts) and current forecasts. The model's atmosphere is nudged towards vorticity, divergence, temperature, and mean sea level pressure of the (re)analysis data. The model's ocean is nudged towards temperature, salinity, and sea ice. The methods are the same as described in Baehr et al. (2015). Assimilation for both versions starts in 1979. Data requirements for initial conditions differ for hindcast and real-time forecast. For the hindcast production both GCFS1.0 and GCFS2.0 use ERA-Interim (Dee et al., 2011) as the atmos- pheric initial conditions. For real-time forecasts, initial conditions are taken from the analyses of the ECM- WF weather forecast model IFS. Ocean initial states for GCFS1.0 are provided by 3D ocean temperature and salinity of the ECMWF ocean reanalysis ORAS4 (Mogensen et al., 2012) and sea ice concentration from the National Snow and Ice Data Center NSIDC (Fetterer et al., 2002). As reanalysis data are usually not available close to the forecast start date, a special product, the near real-time analysis system ORAS4, provided data for GCFS1.0 forecasts. Pragmatical considerations for sea-ice data in terms of availability and consistency led to the choice of ORAS5 data (Zou et al., 2017) for all oceanic variables in GCFS2.0. ORAS5 now also provides the near re- al-time data for GCFS2.0 forecasts. To start the assimilation run for GCFS2.0 in its very beginning at 1979, we use the decadal assimilation with MPI-ESM-HR from the German decadal climate prediction project MiKlip (Pohlmann et al., 2019; Polkova et al., 2019). The decadal MPI-ESM-HR assimilation begins in 1960, accordingly GCFS2.0 starts in 1979 with an ocean and a land surface already adapted to permanently nudged atmospheric and oceanic conditions. 2 of 17 FRÖHLICH ET AL. Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 1. Workflow of GCFS2.0 seasonal forecasts. Figure 1. Workflow of GCFS2.0 seasonal forecasts. 2.3. Ensemble Generation Ensemble members are generated by applying perturbations both to atmosphere and ocean. In the ocean, the ensemble is initialized through bred vectors in all vertical levels (Baehr & Piontek, 2014). The initial ensemble is generated by applying the lagged-day initialization. This means that every member starts on a different day of the preceding month. For instance, on May 1, 1990, the 15 GCFS1.0 members start with con- ditions taken from days between 1 and 30 April. At the end of this first GCFS1.0 hindcast (April 30, 1991), the ocean restart files of these ensemble members serve as perturbations on the assimilation for the next May hindcast. After approximately two years the bred vectors have lost their memory of the initial lagged perturbations (Baehr et al., 2015). With the increase of GCFS2.0's ensemble size to 30, the duration of every individual hindcast and forecast (and therefore the breeding time) is reduced to 6 months. Within the atmosphere we apply a simple perturbation of a physical parameter: the perturbed horizontal diffusion coefficient is imposed on vorticity, divergency and temperature at the uppermost level at 0.01 hPa. It forces slightly different atmospheric conditions of the applied member. This small and simple change has a long known impact on the atmosphere model dynamics and was therefore the first choice for an ensemble FRÖHLICH ET AL. FRÖHLICH ET AL. 3 of 17 3 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 generation. 2.3. Ensemble Generation The perturbation value varies every fifth member in the hindcast ensemble and every tenth member in the forecast ensemble of GCFS2 0 Component Subcomponent GCFS1.0 GCFS2.0 Model Atmosphere/ECHAM ECHAM 6.1.06p4: T63L47 ECHAM 6.3.04p1; T127L95 Land surface vegetation/JSBACH Bulk soil moisture Five layer soil moisture Ocean/MPIOM MPIOM 1.6.1: GR15 MPIOM 1.6.3; TP04 Coupler/OASIS3 MCT Once a day Hourly External forcing data Historical CMIP5 1981–2005 CMIP6 1981–2014 Scenarios CMIP5 RCP 4.5 from 2006 onwards CMIP6 constant 2014 values from 2015 onwards Ensemble generation Atmosphere Perturbation of uppermost atmospheric layer Perturbation of uppermost atmospheric layer Ocean Bred vectors over 12 months Bred vectors over 6 months Assimilation Atmosphere Nudging of ERA-Interim variable fields Nudging of ERA-Interim variable fields Ocean Nudging ORAS4 and NSIDC variable fields Nudging of ORAS5 variable fields Hindcast Period 1981–2014 1990–2017 Forecast duration 12 months 6 months Ensemble member 15 30 Forecast Forecast duration 12 months 6 months Ensemble member 30 50 Table 1 Overview of GCFS1.0 and GCFS2.0 Configuration generation. The perturbation value varies every fifth member in the hindcast ensemble and every tenth member in the forecast ensemble of GCFS2.0. generation. The perturbation value varies every fifth member in the hindcast ensemble and every tenth member in the forecast ensemble of GCFS2.0. generation. The perturbation value varies every fifth member in the hindcast ensemble and every tenth member in the forecast ensemble of GCFS2.0. 2.5. Forecasts The production of the forecasts starts at the beginning of every month, with the exact day depending on the availability of analysis data of atmosphere, ocean, and sea-ice, as explained in the subsection “Assimilation.” GCFS1.0 forecasts ran with 30 ensemble members, having all the same start date at the first of each month, with a forecast duration of 1 year. With GCFS2.0, 50 ensemble members are integrated over half a year, again all members starting on the first day of the month. 2.4. Hindcasts Hindcasts provide necessary climate statistics for the climate forecasts and the assessment of the mod- el's predictive skill in both deterministic and probabilistic scores. This means that for each of the 12 start months of a year, an ensemble prediction is performed for at least 25 - or more - years of the recent past. Hindcast production inherently consumes a large amount of computational resources. In GCFS1.0, the hindcast ensemble of 15 members covers the period from 1981 to 2015. The hindcast dura- tion for each start date was 12 months. The size of the GCFS2.0 ensemble has been increased from 15 to 30. The doubling of the ensemble size to- gether with a higher resolution leads to a considerable increase in computational cost. Therefore, the period of hindcasts and the hindcast duration had to be shortened. The 30 members of GCFS2.0 hindcasts start in 1990 and end in 2017, while performing 6 months of retrospective forecasts for each calendar month in all years. In this way, more than 25 years of hindcasts are achieved. 2.7. Evaluation Metrics of Seasonal Hindcasts A simple but instructive first measure is the difference between the reanalysis and each hindcast data set for a given variable. Decreasing differences between the simulations and a reference data set are usually considered as a success in a correct description of climate processes which is expected to result in improved prediction skill. To further evaluate the predictive skill of each forecast system, deterministic as well as probabilistic meas- ures are used. As a conventional deterministic score we apply the Pearson correlation or so-called anomaly correlation co- efficient (ACC) on the time series of interannual variations of the ensemble mean hindcasts averaged over 3 months. Here, bias corrected hindcast data in cross-validation are used for the ACC. As such, the anomaly for each member is created by subtracting the model climatology where the corresponding member and year is excluded. The significance of the ACC is tested with a t-statistic. To test the significance of differenc- es in the two ACC's we use a method described by (Siegert et al., 2017), with GCFS1.0 (rG1EI) and GCFS2.0 (rG2EI), also taking into account the correlation between the two versions (rG2G1): T r r n n r n n R r r G EI G EI G G G EI G EI 2 2 1 2 1 1 2 2 1 2 1 3 1 4 1          ( ) ( )( ) ( ) ( rG G 2 1 3) , (1) (1) 2 2 2 1 2 1 2 1 2 1 2 (1 ) 2 . G EI G EI G G G EI G EI G G R r r r r r r      (2) 2 2 2 1 2 1 2 1 2 1 2 (1 ) 2 . G EI G EI G G G EI G EI G G R r r r r r r      (2) Here, significance is tested by a two-sided test with the threshold of 2.074 for α = 2.5%. As probabilistic measure for this study, we chose the fair ranked probability skill score RPSS (Ferro, 2014). The fair RPSS is calculated over all defined events or categories (usually 3) of the ensemble hindcasts. The RPSS shows the improvement of using probabilistic forecasts versus using a climatological value. 2.7. Evaluation Metrics of Seasonal Hindcasts The score is called fair as it is adapted to the finite ensemble size. The perfect score is 1, values below 0 denote that the climatology performs better than the respective hindcast ensemble. Further metrics to estimate the reliabil- ity, resolution, and sharpness of the ensemble system are also calculated for GCFS but not shown here for lack of space. For more details on skill measures and scores we refer to Wilks (1995). Journal of Advances in Modeling Earth Systems Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 model-climate of the chosen reference hindcast period. A bias corrected forecast is obtained by adding the observational based climate of the same reference period to each ensemble member anomaly. model-climate of the chosen reference hindcast period. A bias corrected forecast is obtained by adding the observational based climate of the same reference period to each ensemble member anomaly. A probabilistic outlook checks how many members cluster in a defined event category. For seasonal fore- casts, terciles out of the climatological reference period commonly define the cold/normal/warm or dry/ normal/wet events. Seasonal forecasts are usually averaged over 3 months. The first month is discarded in most cases to ac- count for an initialization shock so that the first forecast period starts with the second to the fourth forecast month, and so on. Therefore, a DJF forecast is issued in November, predicting anomalies for the upcoming December, January and February. Likewise, the May forecasts give the outlook for the months June, July and August. To account for the moderate skill and huge uncertainty in seasonal predictions, information about the fore- cast quality is provided along with the forecast itself. Uncertainty estimates are provided via probabilistic representations. 2.6. Communication of Seasonal Predictions Seasonal predictions are provided as anomaly forecasts with respect to a defined model climate. For in- stance, ensemble mean anomalies are created by subtracting the lead-time dependent ensemble-mean FRÖHLICH ET AL. Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 sea-ice model. Coupling between ocean and atmosphere was set to once a day for GCFS1.0. The land and vegetation model JSBACH also hosts a hydrological runoff model. In GCFS1.0, MPI-ESM uses the external forcing like greenhouse gases, ozone, and aerosols based on phase 5 of the Coupled Model Intercomparison Project CMIP5 (Taylor et al., 2012) for historical data and future scenarios. Together with the prescribed solar irradiance these data also account for the solar cycle. Stratospheric aerosol data provide information about volcanoes (Giorgetta et al., 2013). The historical period within CMIP5 ended in 2005. Therefore, all simulations of GCFS1.0 starting from 2006 onwards used the RCP4.5 scenario for the external forcing. The version GCFS2.0 is based on the MPI-ESM-HR (Mauritsen et al., 2018; Müller et al., 2018) with a T127 spectral resolution in the atmosphere corresponding to approximately 70 km at around 50°N, with 95 levels covering the same vertical column up to 0.01 hPa and an ocean resolution of nominally 0.4° in the horizon- tal and, similar to GCFS1.0, 40 levels in the vertical. A number of developments in the atmospheric process- es like radiation, clouds and convection has been applied while ocean physical processes remained largely unchanged. A complete description of all model developments of the new and higher resolved Earth-Sys- tem Model is provided by Mauritsen et al. (2018). We highlight here the increase of the coupling frequency between ocean and atmosphere, which in GCFS2.0 takes places on an hourly basis. Further, the land and vegetation model JSBACH in MPI-ESM-HR includes vertical soil moisture transport (Hagemann & Stacke, 2015). For the low-resolution version MPI-ESM-LR it has been shown, that this contributes to an improve- ment of European summer temperatures (Bunzel et al., 2017). For GCFS2.0, we use external forcing from phase 6 of the Coupled Model Intercomparison Project CMIP6 (Eyring et al., 2016), where the historical period has been extended until 2014. Scenario data were not ready for use for MPI-ESM-HR before mid- 2018, therefore the external forcing has been set constant starting in 2015 up to present time. Table 1 gives an overview of both systems. Journal of Advances in Modeling Earth Systems To give an idea of the different model behavior purely due to the changed model configuration and phys- ics, Figure 2 shows biases of 2 m temperature of the two climate model versions CMIP5 MPI-ESM-LR and CMIP6 MPI-ESM-HR with respect to ERA-Interim for the time range 1990–2014 obtained from historical experiments. The so-called historical experiments are uninitialized model simulations, starting from 1850 until 2014, where the model climate is controlled by the changing external fields like greenhouse gases, aerosol and ozone. Displayed are ensemble means from the 10 member ensemble for December/January/ February DJF in Figure 2 (top) and June/July/August JJA Figure 2 (bottom). Over land, the warm bias of the Amazon basin is considerably reduced in the CMIP6 model version, which we attribute to the improved soil moisture behavior. Seasonal differences also show up, for instance the reduced cold boreal summer bias in northern Russia and northern Africa in the new and higher resolved model system. The bias pattern over northern America also changes between the model versions and the two different seasons, but no clear error reductions are visible there. Europe exhibits a cold bias in the CMIP6 version for DJF and JJA, while the CMIP5 version produced a warm DJF bias. The European cold JJA bias is slightly reduced in the CMIP6 version. Another cold bias region is visible in the North Atlantic. The pattern remains similar between the two versions and appears to be stronger in winter. The two ocean grids GR15 and TP04 (see Table 1) have a very similar resolution setting over the North Atlantic. Therefore, there is no visible effect of an improvement due to a better resolved ocean at the ocean surface. As shown in Gutjahr et al. (2019), the MPI ocean model can represent the pathway of the North Atlantic Current much better at even higher horizontal eddy resolving resolution including improved physics with regard to vertical mixing. In the Southern Hemisphere, a warm bias is visible around the coast of Antarctica for both model versions and both seasons. The bias is slightly reduced in the CMIP6 version, as cloud errors have been alleviated (compare with Figure B6 in Mauritsen et al. (2018)) and possibly as well due to the higher resolution which was also shown in Müller et al. (2018), there Figure 2. Still, sea-ice around Antarctica is not well represent- ed, particularly too little ice in summer (not shown). 2.8. Model Configuration The first version of GCFS1.0 was based on MPI-ESM-LR (Giorgetta et al., 2013), with an atmosphere reso- lution of T63 (corresponding to approximately 150 km at around 50°N), 47 levels reaching up to 0.01 hPa and an ocean resolution of nominally 1.5° in the horizontal with 40 levels in the vertical down to approx- imately 5,000 m (Jungclaus et al., 2013). The ocean model is connected to a dynamic and thermodynamic 5 of 17 5 of 17 FRÖHLICH ET AL. Journal of Advances in Modeling Earth Systems Journal of Advances in Modeling Earth Systems Although some of these biases can be addressed during data assimilation, most will re-emerge in the sea- sonal hindcasts and forecasts. Another striking features of the CMIP6 version for both seasons are the much warmer tropical oceans com- pared to the reanalysis and the reduced cold tongue in the tropical Pacific in Figures 2b and 2d. Although FRÖHLICH ET AL. 6 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 2. Difference of the 2 m temperature between the historical experiment of CMIP5 MPI-ESM-LR (left) and CMIP6 MPI-ESM-HR (right) to ERA-Interim for DJF (top) and JJA (bottom) between 1990 and 2014. Figure 2. Difference of the 2 m temperature between the historical experiment of CMIP5 MPI-ESM-LR (left) and CMIP6 MPI-ESM-HR (right) to ERA-Interim for DJF (top) and JJA (bottom) between 1990 and 2014. rature between the historical experiment of CMIP5 MPI-ESM-LR (left) and CMIP6 MPI-ESM-HR (right) to ERA-Inter n 1990 and 2014. Figure 2. Difference of the 2 m temperature between the historical experiment of CMIP5 MPI-ESM-LR (left) and CMIP6 MPI-ESM-HR (right) to ERA-Interim for DJF (top) and JJA (bottom) between 1990 and 2014. this error pattern has not disappeared completely, it is much smaller for the DJF months than for the JJA months. An issue with the ocean mixing has been reported in Mauritsen et al. (2018). A “bugfix” was implemented and tuned already for the low resolution version CMIP6 MPI-ESM-LR, but for the high resolution version MPI-ESM-HR it was decided to leave out this further tuning connected with the ocean correction. To inves- tigate the impact of the ocean code correction onto the new MPI-ESM-HR climate for seasonal forecasts an experiment was set up. Here, MPI-ESM-HR is run for approximately 100 years under pre-industrial control conditions with correct ocean mixing in order to allow a spin-up for the ocean. Afterwards, a historical experiment simulates the time range from 1850 up to 2014. However, the comparison between the original historical run and the bug-fixed version did not show the desired result of a cooler tropical belt. The revised ocean mixing in MPI-ESM-HR revealed a small improvement (not shown here) but no change of error pattern in the tropical Pacific similar to the change seen in Figure 2. Except for Arctic regions, temperature differences between these two MPI-ESM-HR versions remain mostly below 1 K and are hard to recognize. 3.1. Comparison Setup With May and November we use two of the 12 start months to investigate the hindcast performance of the two forecast systems. To have a common hindcast period of GCFS1.0 and GCFS2.0 the reference time of 1990–2014 is chosen. We focus our assessment on temperature and geopotential fields. Mean sea level pressure is used for the evaluation of the NAO prediction. The reference dataset is the ERA-Interim reanal- ysis except for the evaluation of the Niño3.4 regions, where the NOAA Optimum Interpolation Sea Surface Temperature Version 2 NCEP OIv2 is used. Journal of Advances in Modeling Earth Systems The reason for the tropical warm bias is thus suspected to derive from the revised atmospheric parameter- izations in radiation and cloud cover. Required diagnostics and tuning experiments are currently beyond available resources. 7 of 17 FRÖHLICH ET AL. Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 3. Temperature differences of (a) GCFS1.0 ensemble mean and (b) GCFS2.0 ensemble mean to ERA-Interim of JJA hindcasts during 1990–2014. Figure 3. Temperature differences of (a) GCFS1.0 ensemble mean and (b) GCFS2.0 ensemble mean to ERA-Interim of JJA hindcasts during 1990–2014. Figure 3. Temperature differences of (a) GCFS1.0 ensemble mean and (b) GCFS2.0 ensemble mean to ERA-Interim o 3.2. JJA Mean State and Hindcast Skill However, in some regions it is replaced by a warm bias in GCFS2.0, for example, over North America. The error in the Amazon basin is strongly reduced in GCFS2.0, which we attribute to a feature of the new soil moisture parameterization. The cold tongue in the tropical Pacific, pres- ent in GCFS1.0, completely vanished in GCFS2.0. This seems to be due to a strong impact of ORAS5-nudg- ing. Additionally, the tropical Pacific now shows a strong warm bias in the upwelling region west of the South American and South African coasts. The calculation of the Gaussian weighted, globally averaged RMSE, separately over land and ocean, for both forecast versions is shown in Table 2. This table confirms that a redistribution of the error patterns results in similar RMSE values in GCFS2.0 during boreal summer. We investigate the two systems with respect to their forecast skill in terms of ACC and fair RPSS in Figure 5. Figure 5a shows, that for central Europe no hindcast skill in GCFS2.0 temperature can be expected, while the situation is better in North America, the Mediterranean and some Asian regions, where significant correlation values appear, represented by dots. In the middle troposphere, the ACC for the GCFS2.0 geopo- tential height (Figure 5b) shows negative values over Europe. Again, western US, Greenland and Central Asia exhibit significant positive skill outside the tropical regions. The middle panels, Figures 5c and 5d, show the change in skill from GCFS1.0 to GCFS2.0. For the inter- model comparison GCFS2.0 is mapped onto the coarse grid of GCFS1.0. If regions appear in reddish colors GCFS2.0 is superior over its predecessor GCFS1.0. With the applied method from Siegert et al. (2017) sig- nificant patterns also appear where the difference in the two correlations (rG2EI − rG1EI) is very weak. These places represent regions, where the correlation between GCFS1.0 and GCSF2.0 rG1G2 is very high so that even small differences matter. At the surface relevant improvements for the temperature correlation can be seen over Alaska, the west Siberian Plain, the Amazon and the western tropical Pacific region. A slight improvement is gained in northern Europe. At 500 hPa the correlation for the geopotential height is now stronger over Greenland and parts of Siberia as well as over parts of Antarctica. 3.2. JJA Mean State and Hindcast Skill We present the mean state of the models by looking at the vertical structure of the atmosphere as well as at the surface and the level of 500 hPa. We consider the ensemble mean and time mean of the respective 3 month hindcasts over the time period of 1990–2014. Figure 3 visualizes in a latitude-height-plot the bias of temperature of GCFS1.0 (left) and GCFS2.0 (right) with respect to ERA-Interim. The vertical and the latitudinal temperature structure in the two panels is quite different. In stratospheric levels around/above 35 km a dipole feature of the boreal summer strato- spheric temperature bias of GCFS1.0 (Figure 3a) is visible. This feature is weakened in GCFS2.0 (Figure 3b), where a higher vertical resolution exists, a new ozone data base for CMIP6 is used and the non-orographic gravity waves are newly tuned. However, the middle atmosphere is characterized by a cold bias in the North- ern Hemisphere summer and a strong warm bias in the Southern Hemisphere winter. The comparison shows that the tropospheric cold bias in the lower layers in GCFS1.0 during the JJA months is reduced in GCFS2.0, revealing almost a bias-free Southern Hemisphere. Still, the extra-tropical cold biases in 200 hPa indicate a wrong position of the tropopause in both versions. In the GCFS2.0 Northern Hemisphere, a con- siderable warm bias in the middle troposphere extends from the northern polar latitudes into the tropics. The maximum of the positive bias is placed above the North Pole while in the lowermost layers the cold bias FRÖHLICH ET AL. 8 of 17 8 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 4. Difference of the 2 m temperature hindcasts (a) GCFS1.0 and (b) GCFS2.0 to ERA-Interim for JJA hindcasts in 1990–2014. Figure 4. Difference of the 2 m temperature hindcasts (a) GCFS1.0 and (b) GCFS2.0 to ERA-Interim for JJA hind erence of the 2 m temperature hindcasts (a) GCFS1.0 and (b) GCFS2.0 to ERA-Interim for JJA hindcasts in 1990–2014. of the Arctic remains. The warm bias Northern Hemisphere region is accompanied by a positive humidity bias (not shown here). JJA hindcast temperature biases at 2 m height are displayed in Figure 4. The cold bias of Northern Hemi- sphere land masses in GCSF1.0, which can also be seen in the previously mentioned 10 member ensemble of Baehr et al. (2015), is greatly reduced. 3.2. JJA Mean State and Hindcast Skill The negative skill over Europe is a pattern which has not changed much in comparison to GCFS1.0, but some significant improvements can be stated over Scandinavia. Forecast system RMSE [K] land RMSE [K] ocean GCFS1.0 0.95 0.55 GCFS2.0 0.97 0.55 Table 2 Gaussian Weighted Globally Averaged RMSE of 2 m Temperature for JJA 9 of 17 9 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 The difference in the probabilistic measure RPSS between the two model systems is shown in the bottom p els of Figure 5, again for temperature on the left in Figure 5e and geopotential height on the right in Figur A b f th d l d i hi hli ht th d i h th b bili ti hi d t f GCFS2 0 gure 5. Top panel: Anomaly correlation coefficient (ACC) of hindcasts for June, July, August with respect to ERA-Interim during 1990–2014: (a) 2 m mperature and (b) geopotential height at 500 hPa pressure level. Dots represent significant values at α = 5%. Middle panel: the difference of ACC skill tween GCFS1.0 and GCFS2.0 for (c) 2 m temperature and (d) geopotential height at 500 hPa. Dots represent significant values at α = 2.5% derived from quation 2. Bottom panel: difference of GCFS1.0 and GCFS2.0 for the Fair Ranked Probability Skill Score RPSS for (e) 2 m temperature and (f) geopotential ight at 500 hPa. Figure 5. Top panel: Anomaly correlation coefficient (ACC) of hindcasts for June, July, August with respect to ERA-Interim during 1990–2014: (a) 2 m temperature and (b) geopotential height at 500 hPa pressure level. Dots represent significant values at α = 5%. Middle panel: the difference of ACC skill between GCFS1.0 and GCFS2.0 for (c) 2 m temperature and (d) geopotential height at 500 hPa. Dots represent significant values at α = 2.5% derived from Equation 2. Bottom panel: difference of GCFS1.0 and GCFS2.0 for the Fair Ranked Probability Skill Score RPSS for (e) 2 m temperature and (f) geopotential height at 500 hPa. The difference in the probabilistic measure RPSS between the two model systems is shown in the bottom pan- els of Figure 5, again for temperature on the left in Figure 5e and geopotential height on the right in Figure 5f. 3.2. JJA Mean State and Hindcast Skill As before, the red-colored regions highlight the domains where the probabilistic hindcasts of GCFS2.0 for all event categories are better than the GCFS1.0 ensemble. The differences between the systems are stronger at the surface than in 500 hPa. GCFS2.0 probabilistic hindcasts have improved over the North and Baltic sea FRÖHLICH ET AL. 10 of 17 The difference in the probabilistic measure RPSS between the two model systems is shown in the bottom pan- els of Figure 5, again for temperature on the left in Figure 5e and geopotential height on the right in Figure 5f. As before, the red-colored regions highlight the domains where the probabilistic hindcasts of GCFS2.0 for all event categories are better than the GCFS1.0 ensemble. The differences between the systems are stronger at the surface than in 500 hPa. GCFS2.0 probabilistic hindcasts have improved over the North and Baltic sea FRÖHLICH ET AL. FRÖHLICH ET AL. 10 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 6. Analysis of blocking frequency in JJA during 1990–2014 for (a) ERA-Interim reanalysis, (b) GCFS2.0 ensemble mean, (c) ACC for GCFS1.0 vs. ERA- Interim, and (d) ACC for GCFS2.0 vs. ERA-Interim. Dots represent significant values at the 5% significance level. Figure 6. Analysis of blocking frequency in JJA during 1990–2014 for (a) ERA-Interim reanalysis, (b) GCFS2.0 ensemble mean, (c) ACC for GCFS1.0 vs. ERA- Interim, and (d) ACC for GCFS2.0 vs. ERA-Interim. Dots represent significant values at the 5% significance level. and its surroundings, in the Sahel zone and again over the Amazon region. Degradation is seen for the central tropical Pacific, tropical Atlantic and the Indonesian Archipelago. For geopotential height at 500 hPa proba- bilistic hindcasts are now slightly worse around the tropics, while the North Atlantic shows neutral to slightly improved behavior. The structure and hindcast skill of one of the prominent European summer features, blocking, is shown in Figure 6. Blocking is diagnosed from daily values of geopotential height at 500 hPa of the hindcasts started in May by using a combination of two methods based on Tibaldi and Molteni (1990) and Barriopedro et al. (2010). The ERA-Interim reanalysis (Figure 6a) shows that the European blocking has its maximum over northern Scandinavia. The region where events are present for more than 15% of the summer days during the considered time range extends from Greenland to the Ural Mountains and from Spitsbergen down south to Poland and Ukraine. Please note, we compare the seasonal frequency of summer blocking events but not their correct timing during the summer. While the general shape and the location of the central core of European summer blockings match quite well between reanalysis and forecast systems, the extension and especially the amplitude of GCFS2.0 blockings (Figure 6 b) are smaller. Panels c) and d) of Figure 6 show the ACC of the two versions with dots representing significant values at the 5% significance level. In gener- al, GCFS2.0 better describes the northern and western blocking events, while GCFS1.0 performed better in south-eastern Europe. Over central Europe, the situation is not improved by GCFS2.0. The underestimation of this phenomenon is, however, not unexpected, as blocking processes are known to evolve properly only at horizontal grid resolutions of about 40 km as shown in, for example, Jung et al. (2012). FRÖHLICH ET AL. A further reason of the underestimation of blocking could arise from the cold bias in the North-Atlantic, as has been described for winter blockings by Scaife et al. (2011). This cold bias is considerably reduced in GCFS2.0 but still present. The and its surroundings, in the Sahel zone and again over the Amazon region. Degradation is seen for the central tropical Pacific, tropical Atlantic and the Indonesian Archipelago. For geopotential height at 500 hPa proba- bilistic hindcasts are now slightly worse around the tropics, while the North Atlantic shows neutral to slightly improved behavior. The structure and hindcast skill of one of the prominent European summer features, blocking, is shown in Figure 6. Blocking is diagnosed from daily values of geopotential height at 500 hPa of the hindcasts started in May by using a combination of two methods based on Tibaldi and Molteni (1990) and Barriopedro et al. (2010). The ERA-Interim reanalysis (Figure 6a) shows that the European blocking has its maximum over northern Scandinavia. The region where events are present for more than 15% of the summer days during the considered time range extends from Greenland to the Ural Mountains and from Spitsbergen down south to Poland and Ukraine. Please note, we compare the seasonal frequency of summer blocking events but not their correct timing during the summer. While the general shape and the location of the central core of European summer blockings match quite well between reanalysis and forecast systems, the extension and especially the amplitude of GCFS2.0 blockings (Figure 6 b) are smaller. Panels c) and d) of Figure 6 show the ACC of the two versions with dots representing significant values at the 5% significance level. In gener- al, GCFS2.0 better describes the northern and western blocking events, while GCFS1.0 performed better in south-eastern Europe. Over central Europe, the situation is not improved by GCFS2.0. The underestimation of this phenomenon is, however, not unexpected, as blocking processes are known to evolve properly only at horizontal grid resolutions of about 40 km as shown in, for example, Jung et al. (2012). A further reason of the underestimation of blocking could arise from the cold bias in the North-Atlantic, as has been described for winter blockings by Scaife et al. (2011). This cold bias is considerably reduced in GCFS2.0 but still present. The 11 of 17 FRÖHLICH ET AL. FRÖHLICH ET AL. Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 7. Temperature differences of (a) GCFS1.0 ensemble mean and (b) GCFS2.0 ensemble mean to ERA-Interim with respect to DJF hindcasts during 1990–2014. Figure 7. Temperature differences of (a) GCFS1.0 ensemble mean and (b) GCFS2.0 ensemble mean to ERA-Interim with respect to DJF hindcasts during 1990–2014. Figure 7. Temperature differences of (a) GCFS1.0 ensemble mean and (b) GCFS2.0 ensemble mean to ERA-Interim w 1990–2014. ocean resolution of 0.4° of GCFS2.0 is still not high enough to resolve the processes the North Atlantic current and to place it at the correct position in the North Atlantic. 3.3. DJF Mean State and Hindcast Skill erence of the T2m hindcasts (a) GCFS1.0 and (b) GCFS2.0 to ERA-Interim for DJF hindcasts between in 1990–2014. While ACC values over Europe in DJF are not apparent, the skill in the Arctic region, over Greenland and northern America shows significant values up to 0.8 (Figure 9a), whereas in GCFS1.0 the skill did not ex- ceed values around 0.4 (not shown here). However, directly south of Greenland appears an oceanic region with negative skill. This is a new pattern in GCFS2.0 and is most probably not a model feature but could stem from assimilating the ocean reanalysis data ORAS5 from ECMWF. The same negative Pearson correla- tion appears in the DJF hindcast skill of ECMWF's forecast model System 5 (Johnson et al. , 2019), see their Figure 19a. The authors also discuss ORAS5 as the reason of skill degradation. At the level of 500 hPa again the strongest positive values are found in the tropics. Apart from this, regions with considerable positive and significant skill are found over the eastern North Pacific extending into North America, as well as over Greenland and Arctic regions. Also, the North Atlantic shows a patch of significant positive skill. The middle panels of Figure 9 highlight as before the change in skill in the ACC between the two versions GCFS1.0 and GCFS2.0, with dots over the significant regions. Again, significant regions also appear, where the difference in the two correlations (rG2EI − rG1EI) is very weak but the correlation between GCFS1.0 and GCSF2.0 rG1G2 is very high. Considerable skill has been gained at the surface, represented by the 2 m tem- perature (Figure 9c). Greenland and large parts of Eurasia benefit from the new version. Europe partly gains skill, especially for the very north and the southern regions and partly loses skill, especially over central Europe. A substantial gain in skill is further evident for the geopotential height in the storm track level in the North Atlantic, Arctic regions (Figure 9d), as well as over Eurasia. The significant improvement over central Europe is gained by replacing negative ACC values with weak positive ones. The bottom panels in Figure 9 present the difference between the two systems in terms of the probabilistic hindcast skill score RPSS as previously shown in Figure 5. Skill differences are stronger at the surface for temperature than for the middle troposphere in 500 hPa. 3.3. DJF Mean State and Hindcast Skill As before, the ensemble and time means of GCFS1.0 and GCFS2.0 are compared by using the corresponding ERA-Interim period of December, January and February (DJF) 1990–2014, computed from the November start date. The year is related to the start of the season, namely December. Again, we show first the vertical structure of the temperature differences in Figure 7 for the DJF period. For GCFS1.0 (Figure 7a), a strong cold bias is seen in the upper troposphere extending with weaker values up into the stratosphere. In the GCFS2.0 troposphere (Figure 7b), the cold bias is considerably reduced. In tropical and subtropical latitudes, a minor warm bias is now present, with the maximum at the tropopause. The middle atmosphere of the GCFS1.0 Southern Hemisphere is overly warm up until 35 km, while above a strong cold bias is visible. In contrast, the winter stratosphere in the Northern Hemisphere at the same height appears to be too warm. This dipole structure is strongly reduced in GCFS2.0. However, the warm bias in the winter polar stratosphere extends now from about 100 hPa up to 1 hPa. The polar vortex is too strongly decelerated due to resolved and parameterized wave activity. At the surface, the bias distribution has also changed as is displayed in Figure 8. The tropical oceans now show a general warm bias for GCFS2.0 (8b), as seen before in the historical experiment (Figure 2). In the Pacific, the cold tongue again vanishes as simulations start from an analyzed climate state. Other strong biases from the GCFS1.0 are now also considerably reduced, such as the warm bias over Europe, the Am- azon region and southern Africa or the cold bias over northern Africa. For boreal winter also the globally averaged RMSE is reduced in GCFS2.0, as shown in Table 3. With these improvements in the higher resolved version GCFS2.0 has a better representation of the jet in the storm track region (see Figure 10 from Müller et al. (2018)). This leads to a good skill pattern in ACC of 2 m temperature as seen in Figure 9a and also for the geopotential height at 500 hPa pressure level (Figure 9b). 12 of 17 FRÖHLICH ET AL. Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 8. Difference of the T2m hindcasts (a) GCFS1.0 and (b) GCFS2.0 to ERA-Interim for DJF hindcasts between in 1990–2014. 3.3. DJF Mean State and Hindcast Skill During DJF GCFS2.0 probabilistic hindcasts are more skillful for eastern and southern Europe as well as eastern Russia. The skill amendment over Europe is seen also at 500 hPa. A region where GCFS2.0 forecasts are worse than its predecessor is the central tropical Pacific. The strongest gain in skill is seen over central Africa extending into the Indian Ocean. Im- provements over the ocean are also visible for the subtropical south-eastern Pacific and the Agulhas Basin. The improved conditions in the Northern Hemisphere show up as well in a skillful prediction of the NAO index in GCFS2.0 for the upcoming bore- al winter season (Figure 10). The NAO skill does not directly suffer from the SST problems in ORAS5 reanalysis mentioned above, because it aris- es from a different SST-region in the North Atlantic, and also depends on northern hemispheric sea ice, snow cover and stratospheric temperatures. The NAO index has been calculated using an empirical orthogonal func- tion (EOF) from mean sea level pressure as in Dobrynin et al. (2018). The Forecast system RMSE [K] land RMSE [K] ocean GCFS1.0 1.41 0.59 GCFS2.0 1.36 0.56 Table 3 Gaussian Weighted Globally Averaged RMSE of 2 m Temperature for DJF 13 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 monthly values have been normalized by the standard deviation of the monthly NAO index time series f 1950 to 2000. Comparing to the previous version GCFS1.0 with a not significant NAO prediction skill of 0 the GCFS2.0 increased the NAO skill up to 0.40, which is statistically significant at the 5% significance le although the difference in NAO skill between the two systems is not significant. Further, the NAO skill for gure 9. Top panel: Anomaly correlation coefficient (ACC) of hindcasts for December, January, February with respect to ERA-Interim: (a) 2 m temperatur d (b) geopotential height at 500 hPa pressure level. Dots represent significant values at α = 5%. Middle panel: the difference of ACC skill between GCFS1. d GCFS2.0 for c) 2m temperature and d) geopotential height at 500 hPa. Significance is represented by dots at α = 5% derived from Equation 2. Bottom pa e difference of GCFS1.0 and GCFS2.0 for the Fair Ranked Probability Skill Score for (e) 2 m temperature and (f) geopotential height at 500 hPa. Figure 9. 3.3. DJF Mean State and Hindcast Skill Top panel: Anomaly correlation coefficient (ACC) of hindcasts for December, January, February with respect to ERA-Interim: (a) 2 m temperature and (b) geopotential height at 500 hPa pressure level. Dots represent significant values at α = 5%. Middle panel: the difference of ACC skill between GCFS1.0 and GCFS2.0 for c) 2m temperature and d) geopotential height at 500 hPa. Significance is represented by dots at α = 5% derived from Equation 2. Bottom panel: the difference of GCFS1.0 and GCFS2.0 for the Fair Ranked Probability Skill Score for (e) 2 m temperature and (f) geopotential height at 500 hPa. monthly values have been normalized by the standard deviation of the monthly NAO index time series from 1950 to 2000. Comparing to the previous version GCFS1.0 with a not significant NAO prediction skill of 0.21, the GCFS2.0 increased the NAO skill up to 0.40, which is statistically significant at the 5% significance level, although the difference in NAO skill between the two systems is not significant. Further, the NAO skill for a re- duced 15-member GCFS2.0 ensemble is still considerably higher than in GCFS1.0. The correlation means over all combinations for 15 out of 30 members in GCFS2.0 is 0.34, with a range between 0.01 and 0.62 dependent monthly values have been normalized by the standard deviation of the monthly NAO index time series from 1950 to 2000. Comparing to the previous version GCFS1.0 with a not significant NAO prediction skill of 0.21, the GCFS2.0 increased the NAO skill up to 0.40, which is statistically significant at the 5% significance level, although the difference in NAO skill between the two systems is not significant. Further, the NAO skill for a re- duced 15-member GCFS2.0 ensemble is still considerably higher than in GCFS1.0. The correlation means over all combinations for 15 out of 30 members in GCFS2.0 is 0.34, with a range between 0.01 and 0.62 dependent FRÖHLICH ET AL. 14 of 17 of Advances in Modeling Earth Systems 10.1029/2020MS002101 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 10. NAO time series of ERA-Interim, model ensemble mean and the single ensemble members averaged over December, January, February during 1990–2014, (a) GCFS1.0 and (b) GCFS2.0. Time series are normalized by the standard deviation. The labeled years refer to January of the considered DJF. Figure 10. 3.4. ENSO Hindcasts The assessment of ENSO hindcasts in Figure 11 for the Niño3.4 region shows a strong seasonal dependence of skill in GCFS2.0. However, when considering all 12 start months predictive skill is reduced in GCFS2.0. The general structure of the difficult hindcast start months March, April, May and June can also be seen in GCFS1.0 and is also known in other models as “spring predictability barrier” (see e.g., Wang-Chun Lai et al. (2017)), making ENSO predictions difficult for forecast systems in general. This feature has not been improved in GCFS2.0. However, GCFS2.0 performs with a comparable forecast quality to GCFS1.0 in all other months from July to December. Johnson et al. (2019) also report a warm bias of the ECMWF's seasonal forecast model System 5 during JJA hindcasts in the eastern Pacific basin (their Figure 1d). The warm structure of System 5 is stronger pro- nounced north of the equator and much more confined to the South American coast. Their ENSO predictive quality improved with the higher resolution of the model. 3.3. DJF Mean State and Hindcast Skill NAO time series of ERA-Interim, model ensemble mean and the single ensemble members averaged over December, January, February during 1990–2014, (a) GCFS1.0 and (b) GCFS2.0. Time series are normalized by the standard deviation. The labeled years refer to January of the considered DJF. on the selected members. This highlights that improvements of the model dynamics and physics have large potential for better long-range forecasts in the mid-latitudes as shown by Scaife et al. (2014). on the selected members. This highlights that improvements of the model dynamics and physics have large potential for better long-range forecasts in the mid-latitudes as shown by Scaife et al. (2014). on the selected members. This highlights that improvements of the model dynamics and physics have large potential for better long-range forecasts in the mid-latitudes as shown by Scaife et al. (2014). 4. Conclusions To summarize, we have shown that the second version of the GCFS has some improvements over its pre- decessor. However, a version change does not necessarily lead to improvement everywhere and for every variable, which is certainly true here. An overall similar behavior of GCFS2.0 in comparison to GCFS1.0 has been assessed during JJA. Skill degradation of GCFS2.0 in JJA forecasts is prominent in the tropical Pacific and in the skill for ENSO. During DJF, hindcast quality is improved in GCFS2.0 due to the increased resolu- tion and revised physical parameterizations, especially for Northern Hemisphere, where, for example, NAO skill for the winter months increased. With this version, we have again learned that increasing the model resolution is not per se a solution to many forecast problems as it requires a lot of intense work and evaluation of the model physics on the new grid. Similar challenges are observed for the new system 5 at ECMWF (Johnson et al., 2019), where many issues arise with the new resolution. Scaife et al. (2019) even recommend to invest more into the ensemble size, vertical resolution or ocean resolution than in increasing the atmospheric horizontal resolution. As can be seen from our results, any changes to the model grid need careful adaptation of the model physics. For a future system, a comprehensive performance testing is needed depending on different time-scale ap- plications of this model (Schmidt et al., 2017), besides taking into account a well-tuned climate and climate sensitivity of the Earth-System Model, which is the key for climate experiments and climate projections. In this way, processes active in different seasons or timescales can be accounted for during the model tuning. FRÖHLICH ET AL. 15 of 17 Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Figure 11. Anomaly correlation of the SST forecasts for the ENSO 3.4 region for (a) GCFS1.0 and (b) GCFS2.0, both calculated with respect to the NCEP reanalysis for all 12 start months and all 6 lead months in 1990–2014. Figure 11. Anomaly correlation of the SST forecasts for the ENSO 3.4 region for (a) GCFS1.0 and (b) GCFS2.0, both calculated with respect to the NCEP reanalysis for all 12 start months and all 6 lead months in 1990–2014. Data Availability Statement Hindcast data used for this study and scripts to generate the figures are available by request to the first au- thor under https://doi.org/10.5281/zenodo.3697080. GCFS2.0 data used for this study are available on the C3S climate data store https://climate.copernicus.eu/seasonal-forecasts. Acknowledgments We would like to thank very much all reviewers for their valuable and helpful comments. Large parts of this work have been performed under the Copernicus Service Contracts C3S-433 and C3S-330. The authors thank M. Giorgetta and H. Schmidt from MPI-M Hamburg for their support. The version of ECHAM6 in MPI-ESM is https://svn.zmaw.de/ svn/echam6/tags/echam-6.3.04p1, the version of MPIOM in MPI-ESM is https://svn.zmaw.de/svn/mpiom/tags/ mpiom-1.6.3. Acknowledgments We would like to thank very much all reviewers for their valuable and helpful comments. Large parts of this work have been performed under the Copernicus Service Contracts C3S-433 and C3S-330. The authors thank M. Giorgetta and H. Schmidt from MPI-M Hamburg for their support. The version of ECHAM6 in MPI-ESM is https://svn.zmaw.de/ svn/echam6/tags/echam-6.3.04p1, the version of MPIOM in MPI-ESM is https://svn.zmaw.de/svn/mpiom/tags/ mpiom-1.6.3. References Baehr, J., Fröhlich, K., Botzet, M., Domeisen, D. I. V., Kornblueh, L., Notz, D., et al. (2015). The prediction of surface temperature in the new seasonal prediction system based on the mpi-esm coupled climate model. Climate Dynamics, 44(9-10), 2723–2735. https://doi. org/10.1007/s00382-014-2399-7 Large parts of this work have been performed under the Copernicus Service Contracts C3S-433 and C3S-330. The authors thank M. Giorgetta and H. Schmidt from MPI-M Hamburg for their support. The version of ECHAM6 in MPI-ESM is https://svn.zmaw.de/ svn/echam6/tags/echam-6.3.04p1, the version of MPIOM in MPI-ESM is https://svn.zmaw.de/svn/mpiom/tags/ mpiom-1.6.3. g Baehr, J., & Piontek, R. (2014). Ensemble initialization of the oceanic component of a coupled model through bred vectors at season- al-to-interannual timescales. Geoscientific Model Development, 7(1), 453–461. https://doi.org/10.5194/gmd-7-453-2014 Baehr, J., & Piontek, R. (2014). Ensemble initialization of the oceanic component of a coupled model through bred vectors at season- al-to-interannual timescales. Geoscientific Model Development, 7(1), 453–461. https://doi.org/10.5194/gmd-7-453-2014 Barriopedro, D., Garcia-Herrera, R., & Trigo, R. (2010). Application of blocking diagnosis methods to general circulation models. Part i: A novel detection scheme. Climate Dynamics, 35(7-8), 1373–1391. https://doi.org/10.1007/s00382-010-0767-5 al-to-interannual timescales. Geoscientific Model Development, 7(1), 453–461. https://doi.org/10.5194/gmd-7-453-2014 Barriopedro, D., Garcia-Herrera, R., & Trigo, R. (2010). Application of blocking diagnosis methods to general circulation models. Part i: A novel detection scheme. Climate Dynamics, 35(7-8), 1373–1391. https://doi.org/10.1007/s00382-010-0767-5 f p , ( ), p g g Barriopedro, D., Garcia-Herrera, R., & Trigo, R. (2010). Application of blocking diagnosis methods to general circulation models. novel detection scheme. Climate Dynamics, 35(7-8), 1373–1391. https://doi.org/10.1007/s00382-010-0767-5 f p ( ) p g g Barriopedro, D., Garcia-Herrera, R., & Trigo, R. (2010). Application of blocking diagnosis methods to general circulation models. Part i: A novel detection scheme. Climate Dynamics, 35(7-8), 1373–1391. https://doi.org/10.1007/s00382-010-0767-5 f p ( ) p g g Barriopedro, D., Garcia-Herrera, R., & Trigo, R. (2010). Application of blocking diagnosis methods to general circulation models. Part i: A novel detection scheme. Climate Dynamics, 35(7-8), 1373–1391. https://doi.org/10.1007/s00382-010-0767-5 y p g Bunzel, F., Müller, W. A., Dobrynin, M., Fröhlich, K., Hagemann, S., Pohlmann, H., et al. (2017). Improved seasonal prediction of eu- ropean summer temperatures with new five-layer soil-hydrology scheme. Geophysical Research Letters, 45(1), 346–353. https://doi. org/10.1002/2017GL076204 Bunzel, F., Müller, W. A., Dobrynin, M., Fröhlich, K., Hagemann, S., Pohlmann, H., et al. (2017). Improved seasonal prediction of eu- ropean summer temperatures with new five-layer soil-hydrology scheme. Geophysical Research Letters, 45(1), 346–353. https://doi. org/10.1002/2017GL076204 Dee, D. P., Uppala, S. M., Simmons, A. Journal of Advances in Modeling Earth Systems 10.1029/2020MS002101 Johnson, S. J., Stockdale, T. N., Ferranti, L., Balmaseda, M. A., Molteni, F., Magnusson, L., et al. (2019). Seas5: The new ecmwf seasonal forecast system. Geoscientific Model Development, 12, 1087–1117. https://doi.org/10.5194/gmd-12-1087-2019 Jungclaus, J. H., Fischer, N., Haak, H., Lohmann, K., Marotzke, J., Matei, D., et al. (2013). Characteristics of the ocean simulations in the Max Planck Institute Ocean Model (MPIOM) the ocean component of the mpi-earth system model. Journal of Advances in Modeling Earth Systems, 5(2), 422–446. https://doi.org/10.1002/jame.20023 Jung, T., Miller, M., Palmer, T., Towers, P., Wedi, N., Achuthavarier, D., et al. (2012). High-resolution global climate simulations with the ECMWF Model in Project Athena: Experimental design, model climate, and seasonal forecast skill. Journal of Climate, 25, 3155–3177. https://doi.org/10.1175/JCLI-D-11-00265.1 p g MacLachlan, C., Arribas, A., Peterson, K., Maidens, A., Fereday, D., Scaife, A., et al. (2015). Global Seasonal forecast system versio (GloSea5): A high resolution seasonal forecast system. Quarterly Journal of the Royal Meteorological Society, 141, 1072–1084. http MacLachlan, C., Arribas, A., Peterson, K., Maidens, A., Fereday, D., Scaife, A., et al. (2015). Global Seasonal forecast system version 5 (GloSea5): A high resolution seasonal forecast system. Quarterly Journal of the Royal Meteorological Society, 141, 1072–1084. https:// doi.org/10.1002/qj.2396 Mauritsen, T., Bader, J., Becker, T., Behrens, J., Bittner, M., Brokopf, R., et al. (2018). Developments in the MPI-M Earth System Model version 1.2 (MPI-ESM1.2) and its response to increasing CO2. Journal of Advances in Modeling Earth Systems, 11, 998–1038. https://doi. org/10.1029/2018MS001400 g Mauritsen, T., Stevens, B., Roeckner, E., Crueger, T., Esch, M., Giorgetta, M. A., et al. (2012). Tuning the climate of a global model. Journal of Advances in Modeling Earth Systems, 4, 1–18. Mogensen, K., Balmaseda, M. A., & Weaver, A. (2012). The NEMOVAR ocean data assimilation system as implemented in the ECMWF ocean analysis for system 4. (Technical Memorandum No. 66). Vol. 206. Shinfield Park, Reading: ECMWF. Müller, W. A., Jungclaus, J. H., Mauritsen, T., Baehr, J., Bittner, M., Budich, R., et al. (2018). A higher-resolution version of the Max Planck Institute Earth System Model (MPI-ESM1.2-HR). Journal of Advances in Modeling Earth Systems, 10(7), 1383–1413. https://doi. org/10.1029/2017MS001217 Pohlmann, H., Müller, W. A., Bittner, M., Hettrich, S., Modali, K., Pankatz, K., & Marotzke, J. (2019). Realistic quasi−biennial oscillation variability in historical and decadal hindcast simulations using CMIP6 forcing. Geophysical Research Letters, 46, 14118–14125. https:// doi.org/10.1029/2019GL084878 Polkova, I., Brune, S., Kadow, C., Romanova, V., Gollan, G., Baehr, J., et al. (2019). References J., Berrisford, P., Poli, P., Kobayashi, S., et al. (2011). The era-interim reanalysis: Configuration and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological Society, 137(656), 553–597. https://doi. org/10.1002/qj.828 Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., Kobayashi, S., et al. (2011). The era-interim reanalysis: Configuration and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological Society, 137(656), 553–597. https://doi. org/10.1002/qj.828 Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., Kobayashi, S., et al. (2011). The era-interim reanalysis: Configura and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological Society, 137(656), 553–597. https:// org/10.1002/qj.828 Dobrynin, M., Domeisen, D. I. V., Müller, W. A., Bell, L., Brune, S., Bunzel, F., et al. (2018). Improved teleconnection-based dynamical seasonal predictions of boreal winter. Geophysical Research Letters, 45, 3605–3614. https://doi.org/10.1002/2018GL077209 Dobrynin, M., Domeisen, D. I. V., Müller, W. A., Bell, L., Brune, S., Bunzel, F., et al. (2018). Improved teleconnection-based dynam seasonal predictions of boreal winter. Geophysical Research Letters, 45, 3605–3614. https://doi.org/10.1002/2018GL077209 Dobrynin, M., Domeisen, D. I. V., Müller, W. A., Bell, L., Brune, S., Bunzel, F., et al. (2018). Improved teleconnection-based dynamical seasonal predictions of boreal winter. Geophysical Research Letters, 45, 3605–3614. https://doi.org/10.1002/2018GL077209 Eyring, V., Bony, S., Meehl, G. A., Senior, C. A., Stevens, B., Stouffer, R. J., & Taylor, K. E. (2016). Overview of the coupled model inte parison project phase 6 (cmip6) experimental design and organization. Geoscientific Model Development, 9(5), 1937–1958. https org/10.5194/gmd-9-1937-2016 Ferro, C. (2014). Fair scores for ensemble forecasts. Quarterly Journal of the Royal Meteorological Society, 140, 1917–1923. Fetterer, F., Knowles, K., Meier, W., & Savoie, M. (2002). Sea ice index. Retrieved from http://nsidc.org/data/docs/noaa/g02135_ Giorgetta, M. A., Jungclaus, J., Reick, C. H., Legutke, S., Bader, J., Böttinger, M., et al. (2013). Climate and carbon cycle changes from 1850 to 2100 in mpi-esm simulations for the coupled model intercomparison project phase 5. Journal of Advances in Modeling Earth Systems, 5(3), 572–597. https://doi.org/10.1002/jame.20038 Gutjahr, O., Putrasahan, D., Lohmann, K., Jungclaus, J. H., von Storch, J.-S., Brüggemann, N., et al. (2019). Max planck institute earth sys- tem model (mpi-esm1.2) for the high-resolution model intercomparison project (highresmip). Geoscientific Model Development, 12(7), 3241–3281. https://doi.org/10.5194/gmd-12-3241-2019 Hagemann, S., & Stacke, T. (2015). Impact of the soil hydrology scheme on simulated soil moisture memory. Climate Dynamics, 44, 1731–1750. https://doi.org/10.1007/s00382-014-2221-6 FRÖHLICH ET AL. 16 of 17 Journal of Advances in Modeling Earth Systems Initialization and ensemble generation for decad- al climate predictions: A comparison of different methods. Journal of Advances in Modeling Earth Systems, 11, 149–172. https://doi. org/10.1029/2018MS001439 Scaife, A., Arribas, A., Blockley, E., Brookshaw, A., Clark, R., Dunstone, N., et al. (2014). Skillful long-range prediction of european north american winters. Geophysical Research Letters, 41(7), 2514–2519. https://doi.org/10.1002/2014GL059637 mer, R., Davis, P., Dunstone, N., Gordon, M., et al. (2019). Does increa Scaife, A., Camp, J., Comer, R., Davis, P., Dunstone, N., Gordon, M., et al. (2019). Does increased atmospheric resolution improve seasonal climate predictions? Atmospheric Science Letters, 20, 1–10. https://doi.org/10.1002/asl.922 Scaife, A., Camp, J., Comer, R., Davis, P., Dunstone, N., Gordon, M., et al. (2019). Does increas climate predictions? Atmospheric Science Letters, 20, 1–10. https://doi.org/10.1002/asl.922 Scaife, A., Copsey, D., Gordon, Harris, C., Hinton, T., Keeley, S., et al. (2011). Improved atlantic winter blocking in a climate model. Geo- physical Research Letters, 38(23), L23703. https://doi.org/10.1029/2011GL049573 Scaife, A., Copsey, D., Gordon, Harris, C., Hinton, T., Keeley, S., et al. (2011). Improved a physical Research Letters, 38(23), L23703. https://doi.org/10.1029/2011GL049573 Schmidt, G. A., Bader, D., Donner, L. J., Elsaesser, G. S., Golaz, J.-C., Hannay, C., et al. (2017). Practice and philosophy of climate m tuning across six us modeling centers. Geoscientific Model Development, 10(9), 3207–3223. https://doi.org/10.5194/gmd-10-3207-20 Siegert, S., Bellprat, O., Ménégoz, M., Stephenson, D. B., & Doblas-Reyes, F. J. (2017). Detecting Improvements in Forecast Correlation Skill: Statistical Testing and Power Analysis. Monthly Weather Review, 145(2), 437–450. https://doi.org/10.1175/MWR-D-16-0037.1 Taylor, K. E., Stouffer, R. J., & Meehl, G. A. (2012). An overview of CMIP5 and the experiment design. Bulletin of the American Meteoro ical Society, 93(4), 485–498. https://doi.org/10.1175/bams-d-11-00094.1 y p g baldi, S., & Molteni, F. (1990). On the operational predictability of blocking. Tellus A, 42(3), 343–365. https://doi. org/10.1034/j.1600-0870.1990.t01-2-00003.x Wang-Chun Lai, A., Herzog, M., & Graf, H. (2017). ENSO forecasts near the spring predictability barrier and possible reasons for the re- cently reduced predictability. Journal of Climate, 31, 815–838. https://doi.org/10.1175/JCLI-D-17-0180.1 Wilks, D. (1995). Statistical methods in the atmospheric sciences. San Diego: Academic Press. Zou, H., Balmaseda, M., & Mogensen, C. (2017). The new eddy-permitting ORAP5 ocean reanalysis: description, evaluation and uncertain- ties in climate signals. Climate Dynamics, 49, 791–811. 17 of 17 FRÖHLICH ET AL.
https://openalex.org/W2943189923	https://europepmc.org/articles/pmc6633201?pdf=render	English	null	Strategies for genetic inactivation of long noncoding RNAs in zebrafish	RNA	2,019	cc-by	7,868	© 2019 Lavalou et al. This article, published in RNA, is available under a Creative Commons License (Attribution 4.0 International), as described at http://creativecommons.org/licenses/by/4.0/. REPORT REPORT ABSTRACT The number of annotated long noncoding RNAs (lncRNAs) continues to grow; however, their functional characterization in model organisms has been hampered by the lack of reliable genetic inactivation strategies. While partial or full deletions of lncRNA loci disrupt lncRNA expression, they do not permit the formal association of a phenotype with the encoded tran- script. Here, we examined several alternative strategies for generating lncRNA null alleles in zebrafish and found that they often resulted in unpredicted changes to lncRNA expression. Removal of the transcription start sites (TSSs) of lncRNA genes resulted in hypomorphic mutants, due to the usage of either constitutive or tissue-specific alternative TSSs. Deletions of short, highly conserved lncRNA regions can also lead to overexpression of truncated transcripts. In contrast, knock-in of a polyadenylation signal enabled complete inactivation of malat1, the most abundant vertebrate lncRNA. In summary, lncRNA null alleles require extensive in vivo validation, and we propose insertion of transcription termination sequences as the most reliable approach to generate lncRNA-deficient zebrafish. Keywords: CRISPR-Cas9; hypomorph; long noncoding RNAs; poly(A) signal; zebrafish PERRINE LAVALOU, HELENE ECKERT, LOUISE DAMY, FLORIAN CONSTANTY, SARA MAJELLO, ANGELO BITETTI, ANTOINE GRAINDORGE, and ALENA SHKUMATAVA PERRINE LAVALOU, HELENE ECKERT, LOUISE DAMY, FLORIAN CONSTANTY, SARA MAJELLO, ANGELO BITETTI, ANTOINE GRAINDORGE, and ALENA SHKUMATAVA nstitut Curie, PSL Research University, CNRS UMR3215, INSERM U934, 75005 Paris, France Institut Curie, PSL Research University, CNRS UMR3215, INSERM U934, 75005 Paris, France INTRODUCTION tion, including full or partial deletion of the lncRNA locus, deletion and subsequent replacement of the lncRNA locus by a reporter gene (Nakagawa et al. 2012; Sauvageau et al. 2013), deletion of the lncRNA transcription start site (TSS) and upstream regulatory regions (Fitzpatrick et al. 2002; Zhang et al. 2012) and sequence inversions (Fig.1; Bitetti et al. 2018). Although commonly used, these lncRNA inac- tivation strategies have several caveats and limitations. Full deletions of lncRNA loci, which often span several kilo- bases, or lncRNA replacement by a reporter gene are inva- sive and might lead to phenotypes that are caused by removal of regulatory DNA motifs. Deletions of lncRNA TSS and upstream promoter regions may result in usage of alternative TSSs or cryptic promoters and/or impact the expression of neighboring genes. A less invasive and more accurate approach is to inactivate lncRNAs by inte- grating a premature polyadenylation [poly(A)] cassette. This strategy has been successfully implemented in several recent mouse lncRNA studies (Fig. 1; Bond et al. 2009; Grote et al. 2013; Anderson et al. 2016; Ballarino et al. 2018). Whereas lncRNA locus deletion and partial lncRNA gene inversion strategies have been applied in zebrafish to genetically inactivate lncRNAs (Kok et al. 2015; Hosono et al. 2017; Bitetti et al. 2018; Goudarzi Thousands of lncRNAs have been identified in multiple vertebrate species (Necsulea et al. 2014; Hezroni et al. 2015), but their biological functions remain mostly un- known. To study lncRNAs in vivo, genetic mutants have been generated in model animals, primarily using a mouse model (Leighton et al. 1995; Marahrens et al. 1997; Ripoche et al. 1997; Sado et al. 2001; Sleutels et al. 2002; Bond et al. 2009; Eissmann et al. 2012; Nakagawa et al. 2012, 2014; Zhang et al. 2012; Grote et al. 2013; Li et al. 2013; Sauvageau et al. 2013; Han et al. 2014, 2018; Goff and Rinn 2015; Lai et al. 2015; Amândio et al. 2016; Anderson et al. 2016; Ip et al. 2016; Kotzin et al. 2016; Isoda et al. 2017; Kleaveland et al. 2018), but have also more recently been reported in zebrafish (Kok et al. 2015; Hosono et al. 2017; Bitetti et al. 2018; Goudarzi et al. 2019). Corresponding author: alena.shkumatava@curie.fr INTRODUCTION Genetic inactivation of lncRNAs is less straightforward than for coding genes, where deletion of an exon or a point mutation in the open reading frame (ORF) often leads to stop codons or frame-shift mutations and subsequent loss of function. Several complementary strategies have been implemented to achieve genetic loss of lncRNA func- Thousands of lncRNAs have been identified in multiple vertebrate species (Necsulea et al. 2014; Hezroni et al. 2015), but their biological functions remain mostly un- known. To study lncRNAs in vivo, genetic mutants have been generated in model animals, primarily using a mouse model (Leighton et al. 1995; Marahrens et al. 1997; Ripoche et al. 1997; Sado et al. 2001; Sleutels et al. 2002; Bond et al. 2009; Eissmann et al. 2012; Nakagawa et al. 2012, 2014; Zhang et al. 2012; Grote et al. 2013; Li et al. 2013; Sauvageau et al. 2013; Han et al. 2014, 2018; Goff and Rinn 2015; Lai et al. 2015; Amândio et al. 2016; Anderson et al. 2016; Ip et al. 2016; Kotzin et al. 2016; Isoda et al. 2017; Kleaveland et al. 2018), but have also more recently been reported in zebrafish (Kok et al. 2015; Hosono et al. 2017; Bitetti et al. 2018; Goudarzi et al. 2019). Genetic inactivation of lncRNAs is less straightforward than for coding genes, where deletion of an exon or a point mutation in the open reading frame (ORF) often leads to stop codons or frame-shift mutations and subsequent loss of function. Several complementary strategies have been implemented to achieve genetic loss of lncRNA func- p g Article is online at http://www.rnajournal.org/cgi/doi/10.1261/rna. 069484.118. Freely available online through the RNA Open Access option. 897 RNA (2019) 25:897–904; Published by Cold Spring Harbor Laboratory Press for the RNA Society Lavalou et al. Reporter gene insertion and locus deletion reporter gene Inversion lncRNA locus Full deletion Partial deletion Promoter/TSS deletion PolyA insertion pA FIGURE 1. Strategies for genetic inactivation of lncRNAs in animals. rogations in zebrafish. We generated a ∼280 base pair (bp) deletion of the most conserved region of the 5.5 kb sequence, hereafter referred as cyranoΔCR (Fig. 2A,B; Ulitsky et al. 2011). Interestingly, we detected elevated lev- els of the residual truncated transcript in homozygous cyranoΔCR zebrafish embryos and across cyranoΔCR adult tissues apart from the brain (Fig. 2C,D; Supplemental Fig. 1A). INTRODUCTION These results suggest that removal of a relatively small region of a lncRNA may have an unexpected effect on the transcript levels, potentially leading to its unintend- ed overexpression. TSS deletion of the cyrano locus results in hypomorphic zebrafish mutants Next, we tested if deleting the sequences surrounding and containing lncRNA TSS elements is a reliable alternative strategy for zebrafish lncRNA genetic inactivation. To this end, we generated a minimally invasive cyranoΔTSS mutant allele by removing sequences containing the cyrano TSS (0 to +84) (Fig. 2E). Although cyrano transcript levels were reduced in cyranoΔTSS fish, the transcript was still ro- bustly detectable by RNA blot analysis and qRT-PCR, re- sulting in a hypomorphic cyranoΔTSS mutant (Fig. 2F,G). The 5′ RACE (rapid amplification of cDNA ends) analysis demonstrated that in the absence of the two main TSSs usually used in WT animals, an alternative upstream TSS maintains cyrano expression in cyranoΔTSS mutant zebra- fish (Supplemental Fig. 1B–D). FIGURE 1. Strategies for genetic inactivation of lncRNAs in animals. et al. 2019), analyses of complementary lncRNA silencing approaches including the minimally invasive insertion of the poly(A) sequences have not yet been carried out. Here, we examined the efficiency of several strategies for CRISPR-Cas9-mediated inactivation of lncRNAs in zebrafish. Careful evaluation of lncRNA zebrafish mutants demonstrated that caution is required when analyzing each individual mutant allele. When genetically manipu- lating lncRNA loci, we found that usage of constitutive or tissue-specific alternative TSSs, overexpression or destabi- lization of truncated lncRNA transcripts commonly take place in vivo, minimizing or confounding the effect of the intended genetic intervention. In contrast, using our mini- mally invasive knock-in of a premature polyadenylation signal into the malat1 locus diminished malat1 transcripts to undetectable levels, effectively establishing a malat1 null allele in zebrafish. Notably, neither the cyranoΔCR mutant, with removal of the highly conserved miR-7 site (Ulitsky et al. 2011), nor the cyranoΔTSS mutant fish exhibited obvious morpholo- gical defects. This observation is consistent with recent zebrafish and mouse studies (Kleaveland et al. 2018; Goudarzi et al. 2019) and is in contrast to previous studies that used a morpholino-based knockdown approach to in- activate cyrano (Ulitsky et al. 2011; Sarangdhar et al. 2018). Deletion of the conserved region of the lncRNA cyrano leads to overexpression of the truncated transcript Vertebrate conservation plots based on the eight-genome alignment indicate the location of conserved sequences. (B) The cyranoΔCR mutant allele showing the deletion of the most conserved region of the transcript (dotted, blue line) in zebrafish. Position of the qRT-PCR product is indicated. (C) cyrano expres- sion in WT and homozygous cyranoΔCR embryos detected by qRT-PCR at 2 h postfertilization (hpf), 24 and 72 hpf. (D) cyrano expression across WT and homozygous cyranoΔCR adult tissues detected by qRT-PCR. (E) The cyranoΔTSS zebrafish allele showing deletion of the sequence around the TSS (dotted, blue line). Indicated are positions of the 5′ RACE primer, qPCR prod- uct, RNA blot probe and alternative TSS. (F) cyrano expression in 72 hpf WT and homozygous cyranoΔTSS embryos detected by an RNA blot. 18S rRNA was used as a reference gene. (G) cy- rano expression in 72 hpf WT and homozygous cyranoΔTSS embryos detected by qRT-PCR. eef1α1l1 was used as a reference gene in all qRT-PCR experiments. Each dot represents an in- dividual biological replicate. Data are presented as mean ± S.E.M.; (∗) P < 0.05, n.s., not signifi- cant, unpaired t-tests. Deletion of the conserved region of the lncRNA cyrano leads to overexpression of the truncated transcript To test if the usage of alternative TSSs is a prevalent cellular mechanism to maintain lncRNA gene expression, we exam- ined the effect of TSS deletions on additional lncRNAs in zebrafish. We generated a lnc-sox4aΔTSS mutant allele by removing ∼200 bp surrounding the lnc-sox4a TSS (−43 to +157) (Fig. 3A,B). lnc-sox4a (chr19:29,161,676- 29,270,573; Zv9/danRer7) (Ulitsky et al. 2011) is highly ex- pressed in the zebrafish ovary and was successfully abol- ished in lnc-sox4aΔTSS embryos and across lnc-sox4aΔTSS adult tissues (Fig. 3C,D). However, lnc-sox4a was robustly expressed in the adult lnc-sox4aΔTSS brain at levels compa- rable to WT (Fig. 3D). The 5′ RACE analysis confirmed that a tissue-specific alternative TSS, located in an intron 70 kb A small fraction of zebrafish lncRNAs are conserved in mammals, representing a promising set of candidates for functional interrogation (Ulitsky et al. 2011; Hezroni et al. 2015). The conserved regions of lncRNAs are usually rela- tively short, ranging between 50–300 nucleotides (nt) (Ulitsky et al. 2011; Hezroni et al. 2015) and can be effi- ciently targeted for CRISPR-Cas9-mediated deletions in zebrafish, offering a minimally invasive strategy for func- tional inactivation (Fig. 1). To examine the effect of this strategy on lncRNA expression, we chose the deeply con- served lncRNA cyrano (Ulitsky et al. 2011) for genetic inter- RNA (2019) Vol. 25, No. 8 898 Generation of lncRNA null alleles in zebrafish 1kb cyrano Vertebrate conservation CAGE H3K4me3 RNAseq 3,5 0 1 119 1098 1 qPCR alternative TSS qPCR 5’ RACE RNA blot probe cyrano 18S rRNA A B C D E F G 280bp 84bp 2hpf 24hpf 72hpf 0.00 0.02 0.04 0.06 0.08 0.10 wild type cyranoΔCR n.s. n.s. n.s. Relative expression of cyrano Testis 0.00 0.02 0.04 0.06 Relative expression of cyrano n.s. Heart Skin Kidney n.s. wild type cyranoΔCR Relative expression of cyrano 0.00 0.01 0.02 72hpf 0.0 0.5 1.0 1.5 2.0 wild type cyranoΔTSS Relative expression of cyrano (fold change) cyranoΔTSS wild type n.s. cyranoΔCR cyranoΔTSS FIGURE 2. Genetic perturbations of the lncRNA cyrano in zebrafish result in overexpression and hypomorphic alleles. (A) Gene architecture of the lncRNA cyrano. Shown are the corre- sponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from wild-type (WT) zebrafish. Vertebrate conservation plots based on the eight-genome alignment indicate the location of conserved sequences. Deletion of the conserved region of the lncRNA cyrano leads to overexpression of the truncated transcript wild type cyranoΔCR Relative expression of cyrano 0.00 0.01 0.02 D C alternative TSS qPCR 5’ RACE RNA blot probe cyrano 18S rRNA E F G 84bp 72hpf 0.0 0.5 1.0 1.5 2.0 wild type cyranoΔTSS Relative expression of cyrano (fold change) cyranoΔTSS wild type n.s. cyranoΔTSS alternative TSS qPCR 5’ RACE RNA blot probe cyrano 18S rRNA E F G 84bp 72hpf 0.0 0.5 1.0 1.5 2.0 wild type cyranoΔTSS Relative expression of cyrano (fold change) cyranoΔTSS wild type n.s. cyranoΔTSS FIGURE 2. Genetic perturbations of the lncRNA cyrano in zebrafish result in overexpression and hypomorphic alleles. (A) Gene architecture of the lncRNA cyrano. Shown are the corre- sponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from wild-type (WT) zebrafish. Vertebrate conservation plots based on the eight-genome alignment indicate the location of conserved sequences. (B) The cyranoΔCR mutant allele showing the deletion of the most conserved region of the transcript (dotted, blue line) in zebrafish. Position of the qRT-PCR product is indicated. (C) cyrano expres- sion in WT and homozygous cyranoΔCR embryos detected by qRT-PCR at 2 h postfertilization (hpf), 24 and 72 hpf. (D) cyrano expression across WT and homozygous cyranoΔCR adult tissues detected by qRT-PCR. (E) The cyranoΔTSS zebrafish allele showing deletion of the sequence around the TSS (dotted, blue line). Indicated are positions of the 5′ RACE primer, qPCR prod- uct, RNA blot probe and alternative TSS. (F) cyrano expression in 72 hpf WT and homozygous cyranoΔTSS embryos detected by an RNA blot 18S rRNA was used as a reference gene (G) cy alternative TSS 5’ RACE cyrano 18S rRNA E F 84bp cyranoΔTSS wild type cyranoΔTSS E G 72hpf 0.0 0.5 1.0 1.5 2.0 wild type cyranoΔTSS Relative expression of cyrano (fold change) n.s. G Together, our data showed that in the absence of the main TSS, al- ternative TSSs can be used in a tissue-specific manner, generating hypomorphic mutants, and minimiz- ing the effect of the intended gene inactivation. FIGURE 2. Genetic perturbations of the lncRNA cyrano in zebrafish result in overexpression and hypomorphic alleles. (A) Gene architecture of the lncRNA cyrano. Shown are the corre- sponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from wild-type (WT) zebrafish. Deletion of the conserved region of the lncRNA cyrano leads to overexpression of the truncated transcript (B) The cyranoΔCR mutant allele showing the deletion of the most conserved region of the transcript (dotted, blue line) in zebrafish. Position of the qRT-PCR product is indicated. (C) cyrano expres- sion in WT and homozygous cyranoΔCR embryos detected by qRT-PCR at 2 h postfertilization (hpf), 24 and 72 hpf. (D) cyrano expression across WT and homozygous cyranoΔCR adult tissues detected by qRT-PCR. (E) The cyranoΔTSS zebrafish allele showing deletion of the sequence around the TSS (dotted, blue line). Indicated are positions of the 5′ RACE primer, qPCR prod- uct, RNA blot probe and alternative TSS. (F) cyrano expression in 72 hpf WT and homozygous cyranoΔTSS embryos detected by an RNA blot. 18S rRNA was used as a reference gene. (G) cy- rano expression in 72 hpf WT and homozygous cyranoΔTSS embryos detected by qRT-PCR. eef1α1l1 was used as a reference gene in all qRT-PCR experiments. Each dot represents an in- 1kb cyrano Vertebrate conservation CAGE H3K4me3 RNAseq 3,5 0 1 119 1098 1 qPCR A B C D E 280bp 2hpf 24hpf 72hpf 0.00 0.02 0.04 0.06 0.08 0.10 wild type cyranoΔCR n.s. n.s. n.s. Relative expression of cyrano Testis 0.00 0.02 0.04 0.06 Relative expression of cyrano n.s. Heart Skin Kidney n.s. wild type cyranoΔCR Relative expression of cyrano 0.00 0.01 0.02 cyranoΔCR We generated an additional lncRNA mutant by removing ∼390 bp surrounding the lnc-pou2af1 TSS (−74 to +315) (Fig. 4A,B). Similar to the lnc-sox4aΔTSS allele, the level of lnc-pou2af1 (chr15:16770170-16773 603; Zv9/danRer7) was abolished in lnc-pou2af1ΔTSS embryos and in a sub- set of tested lnc-pou2af1ΔTSS adult tis- sues (Fig. 4C; Supplemental Fig. 3A). However, in skin, kidney, intestine and testis, expression of lnc-pou2af1 was robustly detected in lnc- pou2af1ΔTSS fish (Fig. 4D,E). The 5′ RACE analysis showed that several al- ternative TSSs, located ∼1 kb up- stream of the main TSS, were used in the lnc-pou2af1ΔTSS animals in a tis- sue-specific manner (Fig. 4B; Sup- plemental Fig. 3B,C). Expression of lnc-pou2af1 from alternative TSSs generated new tissue-specific lncRNA exons at the 5′ of the transcript (Fig. 4B; Supplemental Fig. 3C,D). A B C 2hpf 24hpf 72hpf 0.00 0.02 0.04 0.06 0.08 0.10 wild type cyranoΔCR n.s. n.s. n.s. Relative expression of cyrano D 280bp Testis 0.00 0.02 0.04 0.06 Relative expression of cyrano n.s. Heart Skin Kidney n.s. Insertion of a polyadenylation signal resulted in a malat1 null allele in zebrafish Given the evidence that usage of al- ternative TSSs may be a common cel- lular mechanism to confer lncRNA expression, we tested if knock-in of a poly(A) signal into a lncRNA locus can be applied in zebrafish as a mini- mally invasive alternative to generate lncRNA null alleles. This approach has been successfully used to inacti- vate lncRNAs in mice (Grote et al. 2013; Anderson et al. 2016; Isoda et al. 2017; Ballarino et al. 2018). The malat1 locus produces one of the most abundant lncRNAs in vertebrate genomes (Ulitsky et al. 2011; Hezroni et al. 2015). Because malat1 is a mono-exonic lncRNA of ∼7.5 kb and its locus contains multiple TSSs and clustered enhancers forming a so-called super-en- hancer (Pérez-Rico et al. 2017), any deletion strategy of the locus, including TSS removal, has a strong potential sequences. (B) The gion of the transcript d. (C) cyrano expres- t 2 h postfertilization ranoΔCR adult tissues ion of the sequence primer, qPCR prod- WT and homozygous erence gene. (G) cy- tected by qRT-PCR. dot represents an in- 0.05, n.s., not signifi- Given the evidence that usage of al- ternative TSSs may be a common cel- lular mechanism to confer lncRNA expression, we tested if knock-in of a poly(A) signal into a lncRNA locus can be applied in zebrafish as a mini- mally invasive alternative to generate lncRNA null alleles. This approach has been successfully used to inacti- vate lncRNAs in mice (Grote et al. 2013; Anderson et al. 2016; Isoda et al. 2017; Ballarino et al. 2018). downstream from the main TSS (Fig. 3B; Supplemental Fig. 2A,B), was used only in the lnc-sox4aΔTSS animals and main- tained lncRNA expression specifically in the adult brain (Fig. 3D). While homozygous lnc-sox4aΔTSS fish were viable and fertile, our alternative strategy to eliminate lnc-sox4a expression by deleting the last exon failed to generate ho- mozygous fish (Supplemental Fig. 2C,D). 899 www.rnajournal.org Lavalou et al. A B C H3K4me3 RNAseq lnc-sox4 60 1 36 0 0 3.5 CAGE brain specific TSS 5kb 5’ RACE 2 5’ RACE 1 qPCR 200bp D Ovary Testis Brain Skin 0 0.4 1.2 1.8 wild type lnc-sox4ΔTSS Relative expression of lnc-sox4 (x102) n.s. n.s. 72hpf 0.0 0.5 1.0 1.5 Relative expression of lnc-sox4 (x104) wild type lnc-sox4ΔTSS * lnc-sox4ΔTSS FIGURE 3. Presence of a tissue-specific alternative TSS leads to a brain-specific rescue of lnc- sox4a expression. Insertion of a polyadenylation signal resulted in a malat1 null allele in zebrafish (A) The lnc-sox4a locus in zebrafish (chr19:29,161,676-29,270,573). Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from WT zebrafish. (B) The lnc-sox4aΔTSS mutant allele showing deletion of the sequence around the TSS (dotted, blue line). Indicated are positions of the 5′ RACE primers, qPCR primers, and alternative TSS. (C) lnc-sox4a expression in 72 h post- fertilization (hpf) WT and homozygous lnc-sox4aΔTSS embryos detected by qRT-PCR. (D) lnc- sox4a expression across adult WT and homozygous lnc-sox4aΔTSS zebrafish tissues detected by qRT-PCR. eef1α1l1 was used as a reference gene in all qRT-PCR experiments. Each dot rep- resents an individual biological replicate. Data are presented as mean ± S.E.M.; (∗) P < 0.05, (∗∗) P < 0.01, n.s., not significant, unpaired t-tests. A H3K4me3 RNAseq lnc-sox4 60 1 36 0 0 3.5 CAGE 5kb multiple mutant lncRNA alleles. Because common strategies for ge- netic inactivation of lncRNAs often do not allow distinguishing between functions mediated by the lncRNA transcript and those mediated by overlapping DNA regulatory motifs, the generation and interpretation of lncRNA null alleles can be challeng- ing. Here, we compared zebrafish lncRNA mutant alleles generated us- ing several alternative and commonly applied CRISPR-Cas9 strategies for lncRNA inactivation. A B B C 200bp 72hpf 0.0 0.5 1.0 1.5 Relative expression of lnc-sox4 (x104) wild type lnc-sox4ΔTSS * lnc-sox4ΔTSS D Ovary Testis Brain Skin 0 0.4 1.2 1.8 wild type lnc-sox4ΔTSS Relative expression of lnc-sox4 (x102) n.s. n.s. C 72hpf 0.0 0.5 1.0 1.5 Relative expression of lnc-sox4 (x104) wild type lnc-sox4ΔTSS * D x4 We demonstrated that relatively small deletions of conserved regions of lncRNAs, which represent attractive target sequences to eliminate or diminish lncRNA functions (Bitetti et al. 2018; Kleaveland et al. 2018), might result in unexpected changes in lncRNA levels, such as overexpres- sion of the remaining transcript, as demonstrated for cyrano. One possi- bility is that deletion of the conserved region of cyrano, which removed a highly conserved and extensively paired site to miR-7 (Ulitsky et al. 2011), stabilized the cyrano transcript in zebrafish. Alternatively, deletion of this region of cyrano in zebrafish might ve caused transcriptional up-regulation. For example, if letion of this region abrogated cyrano function, cells ght have boosted transcription of the locus in an attempt restore cyrano activity. Insertion of a polyadenylation signal resulted in a malat1 null allele in zebrafish Deletion of the conserved region mouse cyrano does not lead to increased lncRNA levels eaveland et al. 2018), which suggests that cyrano regula- n has diverged between fish and mammals. A better un- rstanding of cyrano regulation and function will help entify the source of this ectopic effect on the remaining cRNAtranscript observedin fish and how this effect might mplicate interpretation of the deletion results. h d h h l f d Skin n.s. fic rescue of lnc- 270,573). Shown 4me3 ChIP-Seq TSS mutant allele d are positions of ion in 72 h post- RT-PCR. (D) lnc- issues detected ts. Each dot rep- ; (∗) P < 0.05, (∗∗) FIGURE 3. Presence of a tissue-specific alternative TSS leads to a brain-specific rescue of lnc- sox4a expression. (A) The lnc-sox4a locus in zebrafish (chr19:29,161,676-29,270,573). Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from WT zebrafish. (B) The lnc-sox4aΔTSS mutant allele showing deletion of the sequence around the TSS (dotted, blue line). Indicated are positions of the 5′ RACE primers, qPCR primers, and alternative TSS. (C) lnc-sox4a expression in 72 h post- fertilization (hpf) WT and homozygous lnc-sox4aΔTSS embryos detected by qRT-PCR. (D) lnc- sox4a expression across adult WT and homozygous lnc-sox4aΔTSS zebrafish tissues detected by qRT-PCR. eef1α1l1 was used as a reference gene in all qRT-PCR experiments. Each dot rep- resents an individual biological replicate. Data are presented as mean ± S.E.M.; (∗) P < 0.05, (∗∗) P < 0.01, n.s., not significant, unpaired t-tests. FIGURE 3. Presence of a tissue-specific alternative TSS leads to a brain-specific rescue of lnc- to affect cis regulatory elements (Fig. 5A). Therefore, we applied our improved protocol for the efficient targeted knock-in to insert a 131 bp SV40 poly(A) signal into the malat1 locus in zebrafish (Fig. 5B; see Materials and Methods; Supplemental Fig. 4A). The targeted knock-in of the poly(A) sequence completely abolished malat1 ex- pression in zebrafish embryos and in all examined adult tis- sues (Fig. 5C,D; Supplemental Fig. 4B,C). Despite effective inactivation of malat1, malat1poly(A) zebrafish were viable and fertile and displayed no obvious morpho- logical defects. The lack of overall morphological abnor- malities is consistent with previously reported Malat1−/− mice (Eissmann et al. 2012; Nakagawa et al. 2012; Zhang et al. 2012) and is in contrast to morpholino-based malat1 inactivation in zebrafish (Wu et al. 2018). Insertion of a polyadenylation signal resulted in a malat1 null allele in zebrafish Taken together, compared to lncRNA deletion strategies, poly(A) signal in- sertion was the most efficient and least invasive approach in zebrafish. to affect cis regulatory elements (Fig. 5A). Therefore, we applied our improved protocol for the efficient targeted knock-in to insert a 131 bp SV40 poly(A) signal into the malat1 locus in zebrafish (Fig. 5B; see Materials and Methods; Supplemental Fig. 4A). The targeted knock-in of the poly(A) sequence completely abolished malat1 ex- pression in zebrafish embryos and in all examined adult tis- sues (Fig. 5C,D; Supplemental Fig. 4B,C). Despite effective inactivation of malat1, malat1poly(A) zebrafish were viable and fertile and displayed no obvious morpho- logical defects. The lack of overall morphological abnor- malities is consistent with previously reported Malat1−/− mice (Eissmann et al. 2012; Nakagawa et al. 2012; Zhang et al. 2012) and is in contrast to morpholino-based malat1 inactivation in zebrafish (Wu et al. 2018). Taken together, compared to lncRNA deletion strategies, poly(A) signal in- sertion was the most efficient and least invasive approach in zebrafish. Moreover, we showed that the removal of TSS and up- stream regulatory regions, a commonly used approach considered to be straightforward to interpret, can result in the presence of eitherconstitutiveor tissue-specific alter- native TSSs that preclude efficient inactivation of lncRNAs and result in hypomorph mutant animals. Although not shown in this study, usage of temporal-specific alternative TSSs might also contribute to the maintenance of lncRNA expression at specific developmental stages, complicating the analysis and interpretation of TSS mutant alleles in an- imal models. Interestingly, a recent study reported that a 326 bp deletion removing cyrano’s TSS leads to loss of the lncRNA expression (Goudarzi et al. 2019). The Generation of lncRNA mutant alleles in zebrafish FIGURE 4. Usage of tissue-specific alternative TSSs maintains lnc-pou2af1 expression in a sub- set of adult tissues. (A) The lnc-pou2af1 locus in zebrafish. Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from WT zebrafish. (B) The lnc-pou2af1ΔTSS mutant allele showing deletion of the se- quence aroundthe TSS (dotted, blue line). Indicated are positions of the 5′ RACE andqPCR prim- ers and alternative TSS. Red box represents a new exon generated from the alternative TSS. (C–E) lnc-pou2af1 expression across a subset of adult WT and homozygous lnc-pou2af1ΔTSS zebrafish tissues detected by qRT-PCR. eef1α1l1 was used as a reference gene in all qRT-PCR experi- ments. Each dot represents an individual biological replicate. Data are presented as mean ± S. E.M.; (∗) P < 0.05, (∗∗) P < 0.01, (∗∗∗) P < 0.001, n.s., not significant, unpaired t-tests. All lncRNA mutant alleles were generated using CRISPR/Cas9-mediated genome editing. To generate lnc-sox4aΔTSS, lnc- pou2af1ΔTSS, cyranoΔTSS, and cyranoΔCR alleles, two sgRNAs (9 ng each, Supple- mental Table 1) and 150 ng in vitro tran- scribed Cas9 mRNA were coinjected into the one-cell stage AB zebrafish embryos (Hwang et al. 2013). To generate lnc-sox4- aΔ3′exon allele, two sgRNAs (100 ng each, difference observed between the cyranoΔTSS alleles may be a consequence of the larger deletion used by Goudarzi et al. potentially leading to a more effective down-regulation of cyrano. In addition, the choice of the lncRNA detection method as well as the developmental timing of detection are important. Our data show that in TSS deletion alleles, lncRNA expression is often abolished at early embryonic stages and robustly reestablished later during development by tissue-specific alternative TSSs. These collective observations underscore the necessity to carefully validate TSS deletion alleles. Supplemental Table 1) and Cas9 protein (50 ng/µL, a gift of the Concordet Lab, Muséum d’Histoire Naturelle, Paris) were coin- jected into the one-cell stage AB zebrafish embryos (Hwang et al. 2013). sgRNAs and Cas9 mRNA were generated as de- scribed previously (Hwang et al. 2013), using the codon-optimized plasmid JDS246 for the Cas9 mRNA synthesis (Addgene #43861), purified with RNeasyMini Kit (Qiagen). Genomic DNA was extract- ed as described previously (Bitetti et al. 2018) and used for geno- typing by PCR, DNA sequencing and mapping of genetic amplification product. The genotyping primers are listed in Sup- plemental Table 2. Generation of lncRNA mutant alleles in zebrafish All zebrafish were bred and maintained at Institut Curie, Paris. Animal care and use for this study were performed in accor- dance with the recommendations of the European Community (2010/63/UE) for the care and use of laboratory animals. Experi- mental procedures were specifically approved by the ethics com- mittee of Institut Curie CEEA-IC #118 (project CEEA-IC 2017-017) in compliance with the international guidelines. Zebrafish were staged using standard procedures (Kimmel et al. 1995). Importantly, our improved protocol for efficient targeted knock-in in zebrafish enabled examination of the effect of a poly(A) signal insertion into the most abundant and en- hancer-dense lncRNA locus. We demonstrate that this min- imally invasive genome editing strategy, previously shown to be successful for lncRNA inactivation in mice (Grote et al. 2013; Anderson et al. 2016; Isoda et al. 2017; Ballarino et al. 2018), is a highly effective strategy in zebrafish. Given the ease of our knock-in approach, which combines the use of a single-strand oligo as a template for homolo- gous recombination and inhibition of nonhomologous end joining, we anticipate that the insertion of a poly(A) se- quence will become a widespread strategy for generating lncRNA mutant alleles in zebrafish. Furthermore, the DISCUSSION The identification of lncRNAs in model vertebrates, their comparative genomics analyses and recent progress in ge- nome editing technologies has led to the generation of RNA (2019) Vol. 25, No. 8 900 Generation of lncRNA null alleles in zebrafish 500bp H3K4me3 RNAseq lnc-pou2af1 A B qPCR 5’ RACE alternative TSS 3.5 0 CAGE 17 1 38 0 389bp C D E Brain Heart Ovary 0 2 4 6 wild type lnc-pou2af1ΔTSS *** * * Relative expression of lnc-pou2af1 (x104) Kidney Intestine Testis 0 5 10 wild type lnc-pou2af1ΔTSS n.s. n.s. ** Relative expression of lnc-pou2af1 (x104) Skin 0 20 40 Relative expression of lnc-pou2af1 (x104) * lnc-pou2af1ΔTSS FIGURE 4. Usage of tissue-specific alternative TSSs maintains lnc-pou2af1 expression in a sub- set of adult tissues. (A) The lnc-pou2af1 locus in zebrafish. Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), and RNA-seq tracks from WT zebrafish. (B) The lnc-pou2af1ΔTSS mutant allele showing deletion of the se- quence aroundthe TSS (dotted, blue line). Indicated are positions of the 5′ RACE andqPCR prim- ers and alternative TSS. Red box represents a new exon generated from the alternative TSS. (C–E) lnc-pou2af1 expression across a subset of adult WT and homozygous lnc-pou2af1ΔTSS zebrafish tissues detected by qRT-PCR. eef1α1l1 was used as a reference gene in all qRT-PCR experi- ments. Each dot represents an individual biological replicate. Data are presented as mean ± S. E.M.; (∗) P < 0.05, (∗∗) P < 0.01, (∗∗∗) P < 0.001, n.s., not significant, unpaired t-tests. knock-in strategy can be used for ge- netic tagging of lncRNAs with self- cleaving ribozymes, which has been demonstrated to perturb lncRNA ex- pression in mouse embryonic stem cells (Tuck et al. 2018) but has not been tested yet in model organisms. Taken together, evaluation of several independent lncRNA mutant alleles in zebrafish indicates that a combination of complementary lncRNA inactivation approaches and their careful analyses are required for robust and accurate lncRNA function- al interrogation. knock-in strategy can be used for ge- netic tagging of lncRNAs with self- cleaving ribozymes, which has been demonstrated to perturb lncRNA ex- pression in mouse embryonic stem cells (Tuck et al. 2018) but has not been tested yet in model organisms. DISCUSSION A Taken together, evaluation of several independent lncRNA mutant alleles in zebrafish indicates that a combination of complementary lncRNA inactivation approaches and their careful analyses are required for robust and accurate lncRNA function- al interrogation. B D Skin 0 20 40 Relative expression of lnc-pou2af1 (x104) * E Kidney Intestine Testis 0 5 10 wild type lnc-pou2af1ΔTSS n.s. n.s. Relative expression of lnc-pou2af1 (x104) 389bp C Brain Heart Ovary 0 2 4 6 wild type lnc-pou2af1ΔTSS * * * Relative expression of lnc-pou2af1 (x104) D 1 E 1 C 1 RNA blots Total RNA was isolated using TRIzol (Invi- trogen), separated on 1% agarose gels containing 0.8% formaldehyde, and trans- ferred to nylon membrane (Nytran SPC, GE Healthcare) by capillary action. Blots were hybridized with α-UTP 32P-labeled RNA probes at 68°C in ULTRAhyb buffer (Ambion) as recommended by the manu- facturer. RNA probe template was ampli- fied from zebrafish brain cDNA by PCR using the primers listed in Supplemental Table 3 (the sequence of the T7 promoter is underlined) and in vitro transcribed (RNA Maxiscript, Ambion) in the presence of α-UTP32P. For each replicate, RNA iso- lated from 30–100 embryos or tissues from three to six adult fish was used. The gel blots and hybridizations in Figure 5C were performed in biological triplicates. The hybridizations in Figures 2F and 5D were performed once. C FIGURE 5. Effective inactivation of the lncRNA malat1 in zebrafish by insertion of a premature polyadenylation signal. (A) The malat1 locus in zebrafish. Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), RNA-seq, and H3K27ac ChIP-Seq (Pérez-Rico et al. 2017) tracks from WT zebrafish. (B) Generation of the malat1poly(A) allele by targeted knock-in of the poly(A) signal. The hybridization site of the RNA blot probe is indicated as a gray box. ha, homology arms. (C) malat1 expression in WT and homozygous malat1poly(A) embryos detected by an RNA blot. 18S rRNA was used as a loading reference. hpf, hours postfertilization. (D) malat1 expression across WT and homozy- gous malat1poly(A) adult zebrafish tissues detected by RNA blot. 18S rRNA was used as a load- ing reference. (131 bp) into the malat1 locus. Briefly, one-cell stage embryos were injected with a single guide RNA (100 ng, Supplemental Table 1), Cas9 protein (50 ng/µL, a gift of the Concordet labora- tory, Muséum d’Histoire Naturelle, Paris), a morpholino against xrcc4 to suppress NHEJ (nonhomologous end joining) (3 ng/µL, Gene Tools LLC, Supplemental Table 1), and a 191 nt single- strand DNA oligo with 30 bp homology arms flanking both sides of the SV40 poly(A) sequence (200 ng, designed and manufac- tured by Ultramer IDT, Supplemental Table 1). Genomic DNA was extracted as described previously (Bitetti et al. 2018), and poly(A) insertion was detected by PCR using primers listed in Supplemental Table 2, DNA sequencing and mapping of genetic amplification product. Generation of the malat1poly(A) allele by CRISPR/ Cas9-mediated homologous recombination in zebrafish The CRISPR/Cas9-mediated knock-in protocol was optimized as described in Supplemental Figure 4A. Zebrafish malat1poly(A) mu- tant was generated by insertion of a single SV40 poly(A) signal 901 www.rnajournal.org www.rnajournal.org Lavalou et al. A A C B FIGURE 5. Effective inactivation of the lncRNA malat1 in zebrafish by insertion of a premature polyadenylation signal. (A) The malat1 locus in zebrafish. Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), RNA-seq, and H3K27ac ChIP-Seq (Pérez-Rico et al. 2017) tracks from WT zebrafish. (B) Generation of the malat1poly(A) allele by targeted knock-in of the poly(A) signal The hybridization site of the A C B FIGURE 5. Effective inactivation of the lncRNA malat1 in zebrafish by insertion of a premature polyadenylation signal. (A) The malat1 locus in zebrafish. Shown are the corresponding CAGE (Nepal et al. 2013; Haberle et al. 2014), H3K4me3 ChIP-Seq (Ulitsky et al. 2011), RNA-seq, and H3K27ac ChIP-Seq (Pérez-Rico et al. 2017) tracks from WT zebrafish. (B) Generation of the malat1poly(A) allele by targeted knock-in of the poly(A) signal. The hybridization site of the RNA blot probe is indicated as a gray box. ha, homology arms. (C) malat1 expression in WT and homozygous malat1poly(A) embryos detected by an RNA blot. 18S rRNA was used as a loading reference. hpf, hours postfertilization. (D) malat1 expression across WT and homozy- gous malat1poly(A) adult zebrafish tissues detected by RNA blot. 18S rRNA was used as a load- ing reference. ment (TURBO DNA-free Ambion). For in- dividual replicates, RNA isolated from 30–100 embryos or tissues from one to six adult fish was used. cDNA was produced with SuperScript IV reverse transcriptase (Invitrogen) and amplified with PowerUp SYBR Green PCR Master Mix (ThermoFisher Scientific) using prim- ers listed in Supplemental Table 3. For each biological replicate, qRT-PCRs were performed in technical triplicate. The eef1α1l1 (eukaryotic translation elonga- tion factor 1 alpha 1, like 1) was used as a reference gene (McCurley and Callard 2008). A C B B RNA ligase-mediated and oligo-capping rapid amplification of cDNA ends (5′′′′′ RACE) TSS usage was determined by rapid amplification of cDNA ends (RACE) according to manufacturer’s instruction (GeneRacer kit, Life Technology). Gene specific primers listed in Supplemental Table 3 were used to amplify lncRNA 5′ RACE products through PCR and nested PCR, subcloned into the PCR BLUNT II TOPO vector (Invitrogen), and transformed in the NEB TOP- 10 cells. A minimum of 12 colonies were sequenced, and the sequences were aligned to the corresponding lncRNA genomic locus. REFERENCES Ip JY, Sone M, Nashiki C, Pan Q, Kitaichi K, Yanaka K, Abe T, Takao K, Miyakawa T, Blencowe BJ, et al. 2016. Gomafu lncRNA knockout mice exhibit mild hyperactivity with enhanced responsiveness to the psychostimulant methamphetamine. Sci Rep 6: 27204. doi:10.1038/srep27204 Amândio AR, Necsulea A, Joye E, Mascrez B, Duboule D. 2016. Hotair is dispensible for mouse development. PLoS Genet 12: e1006232. doi:10.1371/journal.pgen.1006232 p Anderson KM, Anderson DM, McAnally JR, Shelton JM, Bassel- Duby R, Olson EN. 2016. Transcription of the non-coding RNA upperhand controls Hand2 expression and heart development. Nature 539: 433–436. doi:10.1038/nature20128 Isoda T, Moore AJ, He Z, Chandra V, Aida M, Denholtz M, Piet van Hamburg J, Fisch KM, Chang AN, Fahl SP, et al. 2017. Non-coding transcription instructs chromatin folding and com- partmentalization to dictate enhancer-promoter communication and T cell fate. Cell 171: 103–119 e118. doi:10.1016/j.cell.2017 .09.001 Ballarino M, Cipriano A, Tita R, Santini T, Desideri F, Morlando M, Colantoni A, Carrieri C, Nicoletti C, Musarò A, et al. 2018. Deficiency in the nuclear long noncoding RNA Charme causes myogenic defects and heart remodeling in mice. EMBO J 37: e99697. doi:10.15252/embj.201899697 Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. 1995. Stages of embryonic development of the zebrafish. Dev Dyn 203: 253–310. doi:10.1002/aja.1002030302 Bitetti A, Mallory AC, Golini E, Carrieri C, Carreño Gutiérrez H, Perlas E, Pérez-Rico YA, Tocchini-Valentini GP, Enright AJ, Norton WHJ, et al. 2018. MicroRNA degradation by a conserved target RNA regulates animal behavior. Nat Struct Mol Biol 25: 244–251. doi:10.1038/s41594-018-0032-x Kleaveland B, Shi CY, Stefano J, Bartel DP. 2018. A network of non- coding regulatory RNAs acts in the mammalian brain. Cell 174: 350–362 e317. doi:10.1016/j.cell.2018.05.022 Bond AM, Vangompel MJ, Sametsky EA, Clark MF, Savage JC, Disterhoft JF, Kohtz JD. 2009. Balanced gene regulation by an em- bryonic brain ncRNA is critical for adult hippocampal GABA circuit- ry. Nat Neurosci 12: 1020–1027. doi:10.1038/nn.2371 Kok FO, Shin M, Ni CW, Gupta A, Grosse AS, van Impel A, Kirchmaier BC, Peterson-Maduro J, Kourkoulis G, Male I, et al. 2015. Reverse genetic screening reveals poor correlation between morpholino-induced and mutant phenotypes in zebrafish. Dev Cell 32: 97–108. doi:10.1016/j.devcel.2014.11.018 Eissmann M, Gutschner T, Hämmerle M, Günther S, Caudron- Herger M, Gross M, Schirmacher P, Rippe K, Braun T, Zörnig M, et al. 2012. Loss of the abundant nuclear non-coding RNA MALAT1 is compatible with life and development. RNA Biol 9: 1076–1087. ACKNOWLEDGMENTS We thank all members of the Shkumatava laboratory and Ines Drinnenberg for useful discussions. This work was supported by grants from the European Research Council (FLAME-337440), ATIP-Avenir, La Fondation Bettencourt Schueller, ANR-11- LABX-0044_DEEP, and ANR-10-IDEX-0001-02, as well as PSL and La Ligue Nationale Contre Le Cancer doctoral fellowships to P.L. Haberle V, Li N, Hadzhiev Y, Plessy C, Previti C, Nepal C, Gehrig J, Dong X, Akalin A, Suzuki AM, et al. 2014. Two independent tran- scription initiation codes overlap on vertebrate core promoters. Nature 507: 381–385. doi:10.1038/nature12974 Han P, Li W, Lin CH, Yang J, Shang C, Nuernberg ST, Jin KK, Xu W, Lin CY, Lin CJ, et al. 2014. A long noncoding RNA protects the heart from pathological hypertrophy. Nature 514: 102–106. doi:10.1038/nature13596 Author contributions: P.L. developed the protocol for the tar- geted knock-in in zebrafish and contributed to the design, gener- ation, and analysis of the lnc-sox4aΔTSS, lnc-sox4aΔ3’exon, and malat1poly(A) alleles. H.E. contributed to the design, generation, and analysis of the lnc-pou2af1ΔTSS and the maintenance and analyses of lncRNA alleles. L.D and F.C. contributed to lncRNA expression analyses and the maintenance of lncRNA alleles. S. M. contributed to the design, generation, and analyses of the cy- rano alleles. A.B. contributed to the design, generation, and anal- yses of the malat1poly(A) allele. A.G. contributed to the design and generation of the cyrano alleles. P.L. and A.S. wrote the final ver- sion of the manuscript. A.S. conceived and supervised the study. Han X, Luo S, Peng G, Lu JY, Cui G, Liu L, Yan P, Yin Y, Liu W, Wang R, et al. 2018. Mouse knockout models reveal largely dis- pensable but context-dependent functions of lncRNAs during de- velopment. J Mol Cell Biol 10: 175–178. doi:10.1093/jmcb/ mjy003 Hezroni H, Koppstein D, Schwartz MG, Avrutin A, Bartel DP, Ulitsky I. 2015. Principles of long noncoding RNA evolution derived from direct comparison of transcriptomes in 17 species. Cell Rep 11: 1110–1122. doi:10.1016/j.celrep.2015.04.023 Hosono Y, Niknafs YS, Prensner JR, Iyer MK, Dhanasekaran SM, Mehra R, Pitchiaya S, Tien J, Escara-Wilke J, Poliakov A, et al. 2017. Oncogenic role of THOR, a conserved cancer/testis long non-coding RNA. Cell 171: 1559–1572 e1520. doi:10.1016/j.cell .2017.11.040 Received November 2, 2018; accepted April 27, 2019. Hwang WY, Fu Y, Reyon D, Maeder ML, Kaini P, Sander JD, Joung JK, Peterson RT, Yeh JR. 2013. Heritable and precise zebrafish ge- nome editing using a CRISPR-Cas system. SUPPLEMENTAL MATERIAL Total RNA was isolated from zebrafish embryos and adult tissues by TRIzol extraction (Invitrogen) followed by DNase treat- Supplemental material is available for this article. RNA (2019) Vol. 25, No. 8 902 Generation of lncRNA null alleles in zebrafish velopment in the mouse. Dev Cell 24: 206–214. doi:10.1016/j .devcel.2012.12.012 ACKNOWLEDGMENTS PLoS One 8: e68708. doi:10.1371/journal.pone.0068708 Received November 2, 2018; accepted April 27, 2019. REFERENCES doi:10.4161/rna.21089 Kotzin JJ, Spencer SP, McCright SJ, Kumar DBU, Collet MA, Mowel WK, Elliott EN, Uyar A, Makiya MA, Dunagin MC, et al. 2016. The long non-coding RNA Morrbid regulates Bim and short-lived myeloid cell lifespan. Nature 537: 239–243. doi:10 .1038/nature19346 Fitzpatrick GV, Soloway PD, Higgins MJ. 2002. Regional loss of im- printing and growth deficiency in mice with a targeted deletion of KvDMR1. Nat Genet 32: 426–431. doi:10.1038/ng988 Lai KM, Gong G, Atanasio A, Rojas J, Quispe J, Posca J, White D, Huang M, Fedorova D, Grant C, et al. 2015. Diverse phenotypes and specific transcription patterns in twenty mouse lines with ab- lated lincRNAs. PLoS One 10: e0125522. doi:10.1371/journal .pone.0125522 Goff LA, Rinn JL. 2015. Linking RNA biology to lncRNAs. Genome Res 25: 1456–1465. doi:10.1101/gr.191122.115 Goudarzi M, Berg K, Pieper LM, Schier AF. 2019. Individual long non- coding RNAs have no overt functions in zebrafish embryogenesis, viability and fertility. Elife 8: e40815. doi:10.7554/eLife.40815 Leighton PA, Ingram RS, Eggenschwiler J, Efstratiadis A, Tilghman SM. 1995. Disruption of imprinting caused by deletion of the H19 gene region in mice. Nature 375: 34–39. doi:10 .1038/375034a0 Grote P, Wittler L, Hendrix D, Koch F, Währisch S, Beisaw A, Macura K, Bläss G, Kellis M, Werber M, et al. 2013. The tissue-specific lncRNA Fendrr is an essential regulator of heart and body wall de- 903 www.rnajournal.org Lavalou et al. control element. Genes Dev 11: 1596–1604. doi:10.1101/gad .11.12.1596 control element. Genes Dev 11: 1596–1604. doi:10.1101/gad .11.12.1596 Li L, Liu B, Wapinski OL, Tsai MC, Qu K, Zhang J, Carlson JC, Lin M, Fang F, Gupta RA, et al. 2013. Targeted disruption of Hotair leads to homeotic transformation and gene derepression. Cell Rep 5: 3– 12. doi:10.1016/j.celrep.2013.09.003 Sado T, Wang Z, Sasaki H, Li E. 2001. Regulation of imprinted X-chro- mosome inactivation in mice by Tsix. Development 128: 1275– 1286. Marahrens Y, Panning B, Dausman J, Strauss W, Jaenisch R. 1997. Xist-deficient mice are defective in dosage compensation but not spermatogenesis. Genes Dev 11: 156–166. doi:10.1101/gad .11.2.156 Sarangdhar MA, Chaubey D, Srikakulam N, Pillai B. 2018. Parentally inherited long non-coding RNA Cyrano is involved in zebrafish neurodevelopment. Nucleic Acids Res 46: 9726–9735. doi:10 .1093/nar/gky628 McCurley AT, Callard GV. 2008. Characterization of housekeeping genes in zebrafish: male-female differences and effects of tissue type, developmental stage and chemical treatment. BMC Mol Biol 9: 102. REFERENCES doi:10.1186/1471-2199-9-102 Sauvageau M, Goff LA, Lodato S, Bonev B, Groff AF, Gerhardinger C, Sanchez-Gomez DB, Hacisuleyman E, Li E, Spence M, et al. 2013. Multiple knockout mouse models reveal lincRNAs are required for life and brain development. Elife 2: e01749. doi:10.7554/eLife .01749 Nakagawa S, Ip JY, Shioi G, Tripathi V, Zong X, Hirose T, Prasanth KV. 2012. Malat1 is not an essential component of nuclear speckles in mice. RNA 18: 1487–1499. doi:10.1261/rna.033217.112 Sleutels F, Zwart R, Barlow DP. 2002. The non-coding air RNA is re- quired for silencing autosomal imprinted genes. Nature 415: 810–813. doi:10.1038/415810a Nakagawa S, Shimada M, Yanaka K, Mito M, Arai T, Takahashi E, Fujita Y, Fujimori T, Standaert L, Marine JC, et al. 2014. The lncRNA Neat1 is required for corpus luteum formation and the es- tablishment of pregnancy in a subpopulation of mice. Development 141: 4618–4627. doi:10.1242/dev.110544 Tuck AC, Natarajan KN, Rice GM, Borawski J, Mohn F, Rankova A, Flemr M, Wenger A, Nutiu R, Teichmann S, et al. 2018. Distinctive features of lincRNA gene expression suggest widespread RNA-in- dependent functions. Life Sci Alliance 1: e201800124. doi:10 .26508/lsa.201800124 p Necsulea A, Soumillon M, Warnefors M, Liechti A, Daish T, Zeller U, Baker JC, Grützner F, Kaessmann H. 2014. The evolution of lncRNA repertoires and expression patterns in tetrapods. Nature 505: 635–640. doi:10.1038/nature12943 Ulitsky I, Shkumatava A, Jan CH, Sive H, Bartel DP. 2011. Conserved function of lincRNAs in vertebrate embryonic development de- spite rapid sequence evolution. Cell 147: 1537–1550. doi:10 .1016/j.cell.2011.11.055 Nepal C, Hadzhiev Y, Previti C, Haberle V, Li N, Takahashi H, Suzuki AM, Sheng Y, Abdelhamid RF, Anand S, et al. 2013. Dynamic regulation of the transcription initiation landscape at sin- gle nucleotide resolution during vertebrate embryogenesis. Genome Res 23: 1938–1950. doi:10.1101/gr.153692.112 Wu M, Zhang S, Chen X, Xu H, Li X. 2018. Expression and function of lncRNA MALAT-1 in the embryonic development of zebrafish. Gene 680: 65–71. doi:10.1016/j.gene.2018.09.037 Pérez-Rico YA, Boeva V, Mallory AC, Bitetti A, Majello S, Barillot E, Shkumatava A. 2017. Comparative analyses of super-enhancers reveal conserved elements in vertebrate genomes. Genome Res 27: 259–268. doi:10.1101/gr.203679.115 Zhang B, Arun G, Mao YS, Lazar Z, Hung G, Bhattacharjee G, Xiao X, Booth CJ, Wu J, Zhang C, et al. 2012. The lncRNA Malat1 is dis- pensable for mouse development but its transcription plays a cis-regulatory role in the adult. Cell Rep 2: 111–123. REFERENCES doi:10 .1016/j.celrep.2012.06.003 Ripoche MA, Kress C, Poirier F, Dandolo L. 1997. Deletion of the H19 transcription unit reveals the existence of a putative imprinting RNA (2019) Vol. 25, No. 8 904
https://openalex.org/W64971346	https://ccforum.biomedcentral.com/track/pdf/10.1186/ccf-1999-1501	English	null	Microbiologic correlations with serum TNF-α	Critical care	1,999	cc-by	1,113	PublisherInfo PublisherName : BioMed Central PublisherLocation : London PublisherImprintName : BioMed Central PublisherInfo PublisherName : BioMed Central PublisherLocation : London PublisherImprintName : BioMed Central PublisherInfo PublisherName : BioMed Central PublisherLocation : London PublisherImprintName : BioMed Central Comments This interesting paper illustrates the difficulties with developing a clinical trial to investigate novel agents for treating sepsis. The paper is of clinical value in that it provides information on the causative organism in 270 patients with microbiologically documented sepsis and, importantly, the diagnosis of sepsis was made in accordance with strict criteria. The finding that obligate anaerobes were cultured from 10% of patients suggests that antibiotics directed at this group of organisms should be added to "blind" antibiotic strategies for septic patients. Introduction It is a matter of considerable practical importance to understand the possible microbiological causes of sepsis in order to use sensible antibiotics, often in the absence of positive cultures. Only about 60% of clinical sepsis is associated with positive microbiological findings. Understanding of the mediators in sepsis has led to the recognition that TNF-α has a pivotal role and thus therapeutic strategies to reduce or neutralize TNF-α have been developed. Soluble TNF-α receptors have a role in binding and neutralizing the mediator and a trial of TNF-receptor fusion protein is currently underway. Data gathered during this trial to analyze the causative organisms in over 400 patients with septic shock. Aims To determine if there are any differences in the severity of sepsis depending on the organism identified, and also to examine whether different types of organism provoke a different response in terms of TNF-α levels in circulation. Microbiologic correlations with serum TNF-α ArticleInfo ArticleID : 4096 ArticleDOI : 10.1186/ccf-1999-1501 ArticleCitationID : 1501 ArticleSequenceNumber : 33 ArticleCategory : Paper Report ArticleFirstPage : 1 ArticleLastPage : ArticleHistory : RegistrationDate : 1999–8–25 OnlineDate : 1999–8–25 ArticleCopyright : Current Science Ltd1999 ArticleGrants : ArticleContext : 130541111 4 Adrian Mellor Keywords Keywords Sepsis, serum tumour necrosis factor alpha Methods Patients were recruited to this study as part of an international multicenter dose-finding, double-blind, placebo controlled clinical trial. In all, 498 patients were enrolled in the study to evaluate lenercept, a fusion protein consisting of IgG and the p55 component of the TNF receptor. Patients were then stratified into one of two groups, severe sepsis with early septic shock, and late sepsis. A total of 444 patients were included in this analysis. Criteria for inclusion included objective signs of acute infection, at least three signs of systemic inflammatory response syndrome and at least two organ dysfunctions. Patients were enrolled if they fulfilled the criteria for organ dysfunction in the 12 h prior to study-drug medication. APACHE III scores were collected and TNF-α levels assayed using standard techniques. Blood cultures were taken at baseline and on day 3 (two or more sets from two or more sites). Coagulase-negative Staphylococci, Bacillus, Corynebacterium or Propionibacterium species were considered to be contaminants unless isolated from two cultures. Infection was considered "microbiologically documented" when more than one organism that was considered to be clinically significant, was obtained from a clinically relevant sample (blood and other cultures) from day -3 to day 1 (day 1 being the day of study-drug administration). Results Of the 444 patients, 311 were from the US and 133 from Europe. A total of 247 (56%) were stratified into the severe sepsis/early septic shock group and 197 (44%) into the late septic shock group. All patients were followed for 28 days or until death. The mean age of the patients was 58 years (range 18-92) and 60% were male. Sixty one percent had microbiologically documented infection and 31% had bacteremia. The proportion of Gram-negative, Gram-positive and mixed infections was the same in each group. Thirteen patients had a viral/parasitic mixed infection. E. coli was the single most common isolate. Obligate anaerobes such as Bacteroides fragilis and other Bacteroides and Fusobacterium species were isolated in 29 cases. Staphylococcus aureus was the most common Gram-positive organism with other Staphylococci occurring less frequently. Streptococcus pneumoniae was the most frequently isolated Streptococcus species, there were a few cases of S. milleri and of Streptococcus from Lancefield groups B and G. There was a number of cases of infection with S. pyogenes (seven patients). There were 15 cases of fungal infections with Candida species being most common. The respiratory tract and abdomen were the most common sites for sepsis (46% and 27% respectively). Baseline TNF-α levels were available for 409 patients. Mean serum TNF-α levels were 41 pg/ml (range 5-3310 pg/ml) with severe sepsis and 58 pg/ml (range 5-1870 pg/ml) in patients with late sepsis. Patients with positive cultures had significantly higher TNF-α levels and Gram-negative infections resulted in the highest TNF-α concentrations. In the late septic shock group, there was a significantly higher level of TNF-α in patients with a gram-negative infection, compared to infections with gram-positive. Discussion Discussion Pure Gram-positive infections were only slightly less common than pure Gram-negative. These findings are consistent with other studies. The finding of obligate anaerobes in blood and other cultures is interesting. The cell wall of these organisms differs from that of enterobacteriae and the endotoxin they produce is considerably less toxic. If these organisms were responsible for the sepsis in these patients then there are implications for antibiotic therapy. The fact that other organisms such as Candida, M. tuberculosis and L. monocytogenes appeared in cultures is surprising. The patients in this study were selected according to rigorous criteria which suggests that, when specific therapies aimed at TNF are available, they are likely to be used in a wide range of patients. This also leads to the conclusion that septic patients are a heterogeneous group and that it may well be difficult to identify novel agents of particular benefit. No significant correlation could be made between disease severity or presentation and serum TNF-α concentrations, however, TNF-α levels were higher in patients with microbiologically documented infection. Furthermore there were significant TNF-α differences between patients with Gram-positive and Gram negative infections. This suggests firstly that the distinction between "clinically documented" and "microbiologically documented" is a real one that may indicate a difference in biological response, and secondly, Gram-positive and Gram-negative infections produce different host responses and may require different therapeutic interventions. Additional information Full details of the p55 TNF receptor protein factor trial can be obtained as follows:Abraham E, Glauser MP, Butler T, etal. p55 tumour necrosis factor fusion protein in the treatment of patients with severe sepsis and septic shock. A randomised multicenter trial. JAMA 1997; 227: 1531-1538. References 1. Cohen J, Abraham E: Microbiologic findings and correlations with serum tumour necrosis factor alpha in patients with severe sepsis and septic shock. J Infect Dis. 1999, 180: 116-121.
https://openalex.org/W2896033276	https://bmcpublichealth.biomedcentral.com/track/pdf/10.1186/s12889-018-6085-2	English	null	Differentials in prevalence and correlates of metabolic risk factors of non-communicable diseases among women in sub-Saharan Africa: evidence from 33 countries	BMC public health	2,018	cc-by	8,956	Yaya et al. BMC Public Health (2018) 18:1168 https://doi.org/10.1186/s12889-018-6085-2 Yaya et al. BMC Public Health (2018) 18:1168 https://doi.org/10.1186/s12889-018-6085-2 Open Access Differentials in prevalence and correlates of metabolic risk factors of non- communicable diseases among women in sub-Saharan Africa: evidence from 33 countries Sanni Yaya1* , Michael Ekholuenetale2 and Ghose Bishwajit1 Abstract Background: Even with the widespread recognition of non- communicable diseases (NCDs) in sub-Saharan Africa region, yet, sufficient evidence-based surveillance systems to confirm the prevalence and correlates of these diseases is lacking. In an attempt to understand the problem of NCDs in resource-constrained settings, this study was conducted to establish the pattern of the risk factors of NCDs in sub-Sahara Africa region. Methods: The current Demographic and Health Survey (DHS) data sets from 33 countries in sub-Sahara Africa region were used in this study. The individual woman component of DHS 2008–2016 was used. The outcome variables include anemia, hypertension and body mass index (underweight, overweight and obesity). BMI was categorized into; underweight (BMI < 18.5 kg/m2), normal (BMI 18.5–24.9 kg/m2), overweight (BMI 25.0–29.9 kg/m2) and obesity (BMI ≥30 kg/m2). Hemoglobin level: anemic < 12.0 g/dL (< 120 g/L) for women. Hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg. Binary and multinomial logistic regression models were used to investigate the correlates of the variables. Results: The percentage of hypertension was highest among women in Lesotho with about 17.3% and lowest among women in Burundi (1.0%). Anemia was prevalent among sub-Saharan Africa women; where more than half of the women from several countries were anemic with Gabon (60.6%) reporting the highest prevalence. The percentage of obesity in sub-Saharan Africa showed that Lesotho (19.9%), Gabon (18.9%) and Ghana (15.6%) were the prominent countries with obese women, while Madagascar (1.1%) had the minimum obese women. Body mass index was significantly associated with hypertension and anemia. The behavioural or modifiable factors of hypertension and body mass index were; smoking, fruits, vegetables and alcohol consumption. While the non-modifiable significant factors include; age, residence, religion, education, wealth index, marital status, employment and number of children ever born. However, anemia shared similar factors except that smoking and vegetable consumption were not statistically significant. In addition, involvement in exercise was associated with anemia and hypertension. (Continued on next page) * Correspondence: sanni.yaya@uOttawa.ca 1School of International Development and Global Studies, University of Ottawa, 120 University Private, Ottawa, ON K1N 6N5, Canada Full list of author information is available at the end of the article * Correspondence: sanni.yaya@uOttawa.ca 1School of International Development and Global Studies, University of Ottawa, 120 University Private, Ottawa, ON K1N 6N5, Canada Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Background increase in the occurrence of NCDs [10]. Evidence has re- vealed that women had approximately double the obesity prevalence of men [4]. Consequently, the problem of obesity and overweight among women is increasing at a rapid rate in developing countries including sub-Saharan Africa region. However, in multi-country studies, the prevalence of over- weight and obesity is higher in urban areas than in rural areas of developing countries [4, 7]. Non-communicable diseases (NCDs) have recently be- come a prominent cause of mortality, which accounts for nearly two-third of deaths worldwide, while more than three-quarters of NCD-related mortality occur in resource-constrained settings [1]. In sub-Saharan Africa (SSA), the problem of NCDs and their risk factors are on the increase with several causes which may emerge in in- dividuals as anemia, increased blood lipids and obesity, raised blood glucose, elevated blood pressure amongst others [2–4]. These risk factors have been associated with increased disability or morbidity and mortality. Healthy lifestyles and NCD control were the goals of Global Minis- terial Conference, which prepared the platform for the United Nations (UN) Summit on NCD and subsequently, the declaration later in the year 2011 [5]. During the Moscow Declaration on NCDs in 2011, that proceeded from the ministerial conference, encompassed a responsi- bility from governments to create public policies that give equitable health supporting conditions that make commu- nities, families and individuals to have healthy behaviors, as they give precedence to NCD prevention and control, maintaining other health care objectives and to reinforced the commitment of multiple sectors to address NCDs risk factors such as hypertension, anemia, overweight and obesity. Over the past decades, the prevalence of hypertension has been on the rise, from estimated 600 million in 1980 to approximately 1 billion in 2010 due to population growth, ageing and lifestyle [11]. The problem of hyper- tension and other prevalent NCDs has continued un- abated in recent years, and it is the key risk factor for numerous cardiovascular diseases [2, 12]. Recent study reported that the prevalence of hypertension in African most populous country formed a crucial proportion of the total burden in Africa due to large population pres- ently estimated over 170 million [2]. Considering a re- gion with constrained health-care resources, the dynamic disease burden brings issues for health care providers, policy makers and progress toward achieving developmental goals [10]. (Continued from previous page) (Continued from previous page) Conclusion: The problem of NCDs and associated factors remains high among women of reproductive age in sub-Sahara region. The findings of this study suggest that promotion of regular positive health care-seeking behaviour, screening and early treatment are essential to mitigate the burden of NCDs. Furthermore, preventive interventions of NCDs risk factors should be strengthened among key population through behavior change communication with support from government and stakeholders in health care. Keywords: Non-communicable diseases, Hypertension, Obesity, Blood pressure, Risk factors © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Yaya et al. BMC Public Health (2018) 18:1168 Page 2 of 13 Background Hypertension in sub-Saharan Africa has been associated with numerous factors in- cluding urbanization, lifestyles, diets, socio-economic status and physical inactivity [13–16]. p y y NCDs is attributable to numerous risk factors including anemia, which is a prominent nutritional deficiency problem affecting women [17], and a risk factor for poor pregnancy outcome leading to problems that affect the life of foetus and mother alike [17, 18]. Anemia is an age-long public-health problem, particularly among the disadvantaged population in developing countries [19–21]. The prevalence of anemia is a vital health indicator for the measurements of maternal health outcomes [20]. Study on the magnitude of anemia and its determinants in vulnerable groups, such as women of reproductive age is essential for evidence-based intervention modalities, particularly in developing countries, where women suffer from micronutrient deficiencies and shortage of food [21]. The majority of women in resource-constrained settings are at higher risk of anemia due to increased level of macronutrient deficiencies, dietary iron deficiency and infections such as malaria and hook- worm infestation amongst others [22]. Anemia has Obesity and overweight are prominent determinants of metabolic disorders including hypertension, diabetes, cancer and cardiovascular diseases [6–8]. An increase in overweight and obesity due to sedentary lifestyles or physical inactivity and feeding pattern specifically highly processed diet has af- fected community health in an unprecedented manner, by contributing to the risk of metabolic and cardiovascular dis- eases [9]. Similarly, several social, economic and demo- graphic changes have occurred in resource-constrained settings; such as industrialization and urbanization which has influenced the lifestyles of people in developing coun- tries in the past few decades. Obesity and overweight are major public health issues, particularly among women [4]. The effects of obesity and overweight on the sexual and re- productive life of women are more critical, especially during pregnancy. Sharp changes have been observed in many countries in Africa which are currently confronted with obesity and overweight among women, together with large Yaya et al. BMC Public Health (2018) 18:1168 Page 3 of 13 multi-project MEASURE program as MEASURE DHS+. Since 1984, more than 130 nationally representative household-based surveys have been completed under the DHS project in about 70 countries. Many of the countries have conducted multiple DHS to establish trend data that enable them to gauge progress in their programs. Background Countries that participate in the DHS pro- gram are primarily countries that receive USAID assist- ance; however, several non-USAID supported countries have participated with funding from other donors such as UNICEF, UNFPA or the World Bank. DHS are de- signed to collect data on fertility and reproductive health, child health, family planning and HIV/AIDS. Due to the subject matter of the survey, women of reproductive age (15–49) are the main focus of the survey. Women eligible for an individual interview are identified through the households selected in the sample. Therefore, all DHS sur- veys utilize a minimum of two questionnaires-a House- hold Questionnaire and a Women’s Questionnaire. remained more prevalent among women in developing countries [3], while lactating mothers are most prone to blood loss during childbirth and maternal iron de- pletion [21]. p Based on the worrisome prevalence in hypertension, anemia, overweight and obesity, and the possible increase in chronic diseases in sub-Saharan Africa countries, data on the correlates of the risk factors of NCDs is crucial and ur- gently needed to inform policy and programme interven- tions to address the problem in Africa. Evidence-based information on the prevalence and correlates of hyperten- sion, anemia, overweight and obesity is paramount to under- stand the size of the problem, identify key population and develop effective intervention. Notably, a number of studies that examined hypertension, anemia, overweight and obesity prevalence in Africa centered on single country. However, the present study explored the prevalence and correlates of metabolic risk factors of NCDs in several sub-Saharan Africa countries and organized the results to aid understanding. Data extraction The risk factors of NCDs considered in this study; were anemia, hypertension and BMI (underweight, overweight and obesity). DHS grouped non-pregnant “anemic” women at: Hb level < 12.0 g/dl and non-pregnant “not anemic” women at: Hb level ≥12.0 g/dl. To adjust for anemia during pregnancy, women who were pregnant at the time of the surveys were categorized as “anemic” at: Hb level < 11.0 g/dl and “not anemic” at: Hb level ≥ 11.0 g/dl [6, 17]. BMI was calculated as the ratio of weight in kilograms (kg) to the square of height in me- ters (m2). BMI was categorized into; underweight (BMI < 18.5 kg/m2), normal (BMI 18.5–24.9 kg/m2), over- weight (BMI 25.0–29.9 kg/m2) and obesity (BMI ≥30 kg/ m2) [6, 9]. Hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pres- sure (DBP) ≥90 mmHg [6, 10]. The systolic and diastolic blood pressures were confirmed by repeating the mea- surements three times and taking the average. Data extracted for this study involved women of reproduct- ive age (15–49 years) in nationally representative Demo- graphic and Health Surveys conducted in 33 countries in sub-Sahara Africa region, 2008–2016. The study involved the following countries; Benin, Burkina-Faso, Burundi, Cameroon, Chad, Comoros, Congo, Cote d’Ivoire, Demo- cratic Republic of Congo, Ethiopia, Gabon, Gambia, Ghana, Guinea, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome & Principe, Senegal, Sierra Leone, Tanzania, Togo, Uganda, Zambia and Zimbabwe (See details in Table 1). DHS data is publicly available and can be accessed from MEASURE DHS database at http://dhsprogram.com/data/available-da- tasets.cfm. DHS are usually implemented by the National Population Commission (NPC) with financial and technical assistance by ICF International provisioned through the USAID-funded MEASURE DHS program. DHS involved multi-stage stratified cluster design based on a list of enu- meration areas (EAs), which are systematically selected units from localities and constitute the Local Government Areas (LGAs). Details of the sampling procedure have been re- ported in previously [23]. Independent variables Several factors including socio-demographic, economic and other maternal determinants were measured in this study: age (years) of women were categorized to 15–19/20–24/25– 29/30–34/35–39/40–44/45–49; place of residence: rural/ urban; educational attainment: no formal education/pri- mary/secondary/higher education; religious background: Christianity/Islam/other religion or no religion; marital sta- tus: never married/married or living with partner; currently working: yes/no; number of children ever born: nulliparous/ 1–4/> 4; alcohol consumption: yes/no; smoking: yes/no; fruits consumption: low (< 2 days/week)/moderate (2– 3 days/week)/high (≥4 days/week); vegetable consumption: low (< 2 days/week)/moderate (2–3 days/week)/high DHS program was established by the United States Agency for International Development (USAID) in 1984. It was designed as a follow-up to the World Fertility Survey and the Contraceptive Prevalence Survey pro- jects. It was first awarded in 1984 to Westinghouse Health Systems (which subsequently evolved into part of OCR Macro). The project has been implemented in overlapping five-year phases; DHS-I ran from 1984 to1990; DHS-II from 1988 to1993; and DHS-III from 1992 to1998. In 1997, DHS was folded into the new Yaya et al. Independent variables BMC Public Health (2018) 18:1168 Page 4 of 13 Table 1 Weighted percentage of high blood pressure, anemia and body mass index by countries in sub-Saharan Africa Country Survey year Number of women (n) High blood pressure (%) Anemia (%) Body mass index Underweight (%) Overweight (%) Obese (% Benin 2012 16,599 4.4 41.4 6.0 20.3 7.2 Burkina-Faso 2010 17,087 48.8 14.8 8.2 2.9 Burundi 2016–17 17,269 1.0 39.3 17.7 6.6 1.8 Cameroon 2011 15,426 39.5 6.7 22.1 10.6 Chad 2014–15 17,719 18.0 9.6 2.3 Comoros 2012 5329 6.6 25.4 12.3 Congo 2012 10,819 54.2 13.6 17.6 8.5 Cote d’Ivoire 2012 10,060 53.7 7.1 19.3 6.6 Democratic Republic of Congo 2013–14 18,827 38.4 13.2 13.0 3.3 Ethiopia 2016 15,683 23.6 21.1 6.3 1.5 Gabon 2012 8422 60.6 6.7 26.1 18.9 Gambia 2013 10,233 59.0 15.7 16.1 7.3 Ghana 2014 9396 5.2 42.4 5.9 25.1 15.6 Guinea 2012 9142 49.1 11.5 14.9 4.6 Kenya 2014 31,079 9.4 8.6 22.7 10.1 Lesotho 2014 6621 17.3 27.3 4.1 25.5 19.9 Liberia 2013 9239 7.0 18.6 8.2 Madagascar 2009 17,375 35.3 25.5 5.2 1.1 Malawi 2015–16 16,592 32.7 6.7 16.1 5.4 Mali 2013 10,424 51.4 10.8 12.6 5.0 Mozambique 2011 13,537 54.0 7.9 12.7 4.0 Namibia 2013 1018 20.7 13.1 19.0 13.0 Niger 2012 11,160 45.8 13.9 13.5 3.7 Nigeria 2013 38,948 10.6 17.6 7.6 Rwanda 2014–15 13,497 19.2 6.1 18.4 3.9 Sao Tome & Principe 2009 2615 43.1 7.2 23.2 11.8 Senegal 2011 15,688 54.3 20.8 15.4 5.8 Sierra Leone 2013 16,658 44.8 8.6 13.9 4.9 Tanzania 2015–16 13,266 44.8 8.7 18.9 9.9 Togo 2013–14 9480 48.1 6.5 20.3 11.1 Uganda 2016 18,506 31.7 8.1 17.2 7.2 Zambia 2013–14 16,411 9.6 16.6 6.7 Zimbabwe 2015 9955 26.8 5.8 23.4 12.5 ercentage of high blood pressure, anemia and body mass index by countries in sub-Saharan Africa poorer, middle, richer and richest. The computation of wealth scores variable was conducted by DHS and has previ- ously been reported [24]. poorer, middle, richer and richest. The computation of wealth scores variable was conducted by DHS and has previ- ously been reported [24]. (≥4 days/week). In addition, the wealth scores were mea- sured using principal components analysis approach based on a list of household assets, which include, number of household members, wall and roof materials, floor types, ac- cess to potable water and sanitation, type of cooking fuel, ownership of television, radio, motorcycle, refrigerator amongst others. Independent variables Based on the weighted wealth scores, households were grouped into wealth quintiles; poorest, Determinants of high blood pressure and anemia Determinants of high blood pressure and anemia The HBP regression model results applies only to 5 countries in sub-Saharan Africa region which include; Benin, Burundi, Ghana, Kenya and Lesotho where the data was captured. The results showed that women at higher age intervals were more likely to have HBP, com- pared with women aged 15–19 years. Rural women had 39% reduction in HBP, compared with women in the urban residence (OR = 0.61; 95%CI = 0.56–66). Educated women had higher odds of HBP, compared with women with no formal education. Women with no Christianity or Islamic background had 30% reduction in HBP, com- pared with women with Christianity background (OR = 0.70; 95%CI = 0.56–0.88). More so, women who are cur- rently married or living with a partner were 3.39 times as likely to have HBP, compared with women who were never married (OR = 3.39; 95%CI = 3.02–3.81). Women with higher wealth index or those employed had increase in HBP, compared with poorest or unemployed women. Women who had ever given birth had significant in- crease in the odds of HBP, compared with women with no childbirth. Further, alcohol, fruit and vegetable con- sumption showed significant reduction in HBP. Mean- while, involvement in smoking and exercise were more likely to be associated with HBP. Body mass index was significantly associated with HBP; underweight women had 34% reduction in HBP compared with women of normal body mass index (OR = 0.66; 95%CI = 0.54–0.82). Conversely, overweight women were 2.44 times as likely to have HBP, compared with women with normal body weight (OR = 2.44; 95%CI = 2.19–2.72); and obese women were 5.34 times as likely to have HBP compared with normal body mass index women (OR = 5.34; 95%CI = 4.75–5.99) (Details are shown in Table 3). Ethical considerations We did the analyses using publicly available data from demographic health surveys. Ethical procedures were the responsibility of the institutions that commissioned, Yaya et al. BMC Public Health (2018) 18:1168 Page 5 of 13 women, while Madagascar (1.1%) had the minimum obese women (See details in Table 1). funded, or managed the surveys. All DHS surveys are approved by ICF international as well as an Institutional Review Board (IRB) in respective country to ensure that the protocols are in compliance with the U.S. Depart- ment of Health and Human Services regulations for the protection of human subjects. The results from Table 2 show the distribution of women’s characteristics by the risk factors of noncom- municable diseases. High blood pressure increased with advances in age. Similarly, overweight and obesity was higher among older respondents. Compared with rural women, the urban women had higher proportion of high blood pressure (7.2% vs 4.5%), overweight (21.7% vs 12.9%) and obesity (11.8% vs 3.9%). The converse was true for the proportion of women with anemia and underweight. Furthermore, the pattern of the risk factors of noncommunicable diseases was similar across socio- economic characteristics. Women with higher education and higher wealth index had higher proportion of high blood pressure, overweight and obesity. The converse was however true among women with anemia and underweight. The details of women’s characteristics are presented in Table 2. Statistical analyses Prevalence of the risk factors of NCDs was reported by percentages. To adjust for data representation, the survey module (svyset) was used for all analyses to account for sample weight. Correlation matrix was used to conduct multicollinearity diagnostics to examine association be- tween explanatory variables using a cut-off minimum of 0.6 known to cause multicollinearity [25]. All explanatory variables were retained for analysis due to lack of collin- earity. Furthermore, variables that were statistically signifi- cant in the unadjusted model were added in the multivariable regression models to adjust for the effect of confounding. Binary and multinomial logistic regression models were used to investigate the factors associated with anemia, hypertension [14] and body mass index [4]. The level of statistical significance was set at 5%. All data analyses were conducted using Stata 13.0 (Statacorp, Col- lege Station, Texas, United States of America). Results The description of high blood pressure (HBP), anemia and anomalous body mass index (BMI) by countries in sub-Saharan Africa was presented in Table 1. Notably, only 5 countries in sub-Saharan Africa region reported data on blood pressure, which include; Benin, Burundi, Ghana, Kenya and Lesotho. The percentage of HBP was highest among women in Lesotho with about 17.3% and lowest among women in Burundi (1.0%). More so, the results showed that anemia is prevalent among sub-Saharan Africa women; where more than half of the women from Gabon (60.6%), Gambia (59.0%), Senegal (54.3%), Congo (54.2%), Mozambique (54.0%), Cote d’Ivoire (53.7%) and Mali (51.4%) were reportedly anemic. However, Rwanda (19.2), Namibia (20.7%) and Ethiopia (23.6%) had the least anemic women. Further, Madagascar (25.5%), Ethiopia (21.1%), Senegal (20.8%), Chad (18.0%) and Burundi (17.7%) reported the highest underweight women, while Lesotho (4.1%) had the low- est percentage of underweight women. In addition, Gabon (26.1%), Comoros (25.4%), Ghana (25.1%), Zimbabwe (23.4%), Sao Tome & Principe (23.2%), Kenya (22.7%), Cameroon (22.1%) and Benin (20.3%) had the leading overweight women and Madagascar (5.2%) had the minimum overweight women in sub-Saharan Africa countries. The percentage of obesity in sub-Saharan Af- rica showed that Lesotho (19.9%), Gabon (18.9%) and Ghana (15.6%) were the prominent countries with obese Yaya et al. Results BMC Public Health (2018) 18:1168 Page 7 of 13 Table 2 The distribution of respondents’ characteristics (Continued) Variable n (%) High blood pressure (%) Anemia (%) Body mass index (%) Normal (%) Underweight Overweight Obesity Exercise No 27,424 (80.4) 6.3 31.8 57.0 10.6 21.2 11.2 Yes 6692 (19.6) 9.3 26.5 57.5 11.1 19.9 11.5 Fruit consumption Low 7047 (38.4) 5.2 29.7 56.8 12.9 19.1 11.2 Moderate 5128 (27.9) 3.9 31.1 56.3 8.4 20.7 14.5 High 6174 (33.6) 4.8 33.6 50.9 5.5 25.2 18.5 Vegetable consumption Low 5405 (29.0) 5.5 31.3 55.8 11.5 20.2 12.4 Moderate 5156 (27.7) 3.7 31.9 55.9 9.7 20.9 13.5 High 8062 (43.3) 4.9 31.2 52.8 7.2 23.1 16.9 Table 2 The distribution of respondents’ characteristics (Continued) obese relative to normal body mass index, compared with women aged 15–19 years. While rural women were 1.25 times as likely to be underweight, compared with urban women (RR = 1.25; 95%CI = 1.22–1.28); the rural dwellers had 51% (RR = 0.49; 95%CI = 0.48–0.50) and 72% (RR = 0.28; 95%CI = 0.27–0.29) reduction in overweight and obesity relative to women with normal body mass index, compared with women from urban residence. In addition, educated women, those from households with higher wealth index or employed had significant reduction in underweight, but had higher risk of overweight and obes- ity relative to women with normal body mass index, com- pared with women with no formal education, from poorest households or unemployed. Also, Table 3 shows factors associated with anemia among women of reproductive age. The age of women was signifi- cantly associated with anemia among women. Furthermore, rural women were 1.08 times as likely to have anemia, com- pared with women in urban residence (OR = 1.08; 95%CI = 1.06–1.10). Educated women, those from households with higher wealth index or those currently working had signifi- cant reduction in the odds of anemia, compared with women without formal education, those from poorest households or the unemployed. The religious background of women was significantly associated with anemia. Married women or those living with a partner were 1.16 times as likely to have anemia, compared with unmarried women (OR = 1.16; 95%CI = 1.13–1.18). Childbirth was positively associated with anemia, compared with nulliparous women. Results BMC Public Health (2018) 18:1168 Page 6 of 13 Table 2 The distribution of respondents’ characteristics Variable n (%) High blood pressure (%) Anemia (%) Body mass index (%) Normal (%) Underweight Overweight Obesity Age 15–19 99,944 (21.2) 1.3 41.5 72.8 18.6 7.5 1.1 20–24 86,128 (18.3) 3.0 41.4 73.1 10.7 13.1 3.1 25–29 82,221 (17.5) 4.6 40.6 66.9 9.2 17.7 6.2 30–34 67,361 (14.3) 5.5 40.2 61.8 8.6 20.3 9.4 35–39 56,964 (12.1) 7.5 41.9 58.3 8.8 21.4 11.5 40–44 42,935 (9.1) 10.6 42.6 55.7 9.5 21.8 13.0 45–49 34,863 (7.4) 14.1 40.4 54.9 10.2 21.7 13.2 Residence Urban 171,897 (36.5) 7.2 39.9 57.8 8.6 21.7 11.8 Rural 299,361 (63.5) 4.5 41.8 70.1 13.1 12.9 3.9 Educational level None 154,399 (32.8) 3.4 46.8 69.3 14.1 12.6 3.9 Primary 157,612 (33.4) 6.1 39.2 66.4 11.0 16.2 6.4 Secondary 139,247 (29.6) 5.8 38.5 62.7 9.8 18.5 9.0 Higher 19,945 (4.2) 9.4 30.8 49.3 5.9 27.1 17.7 Religion Christianity 292,827 (65.6) 5.6 37.2 65.3 10.0 17.2 7.4 Islam 127,841 (28.7) 5.0 48.3 65.6 14.5 14.3 5.6 Others/no religion 25,420 (5.7) 4.0 44.0 69.2 13.1 12.8 4.8 Marital status Never married 122,194 (25.9) 2.1 38.5 69.8 16.3 10.9 3.1 Married/currently living with a partner 349,063 (74.1) 6.9 42.0 64.0 9.8 18.0 8.2 Wealth index Poorest 93,862 (19.9) 3.3 45.5 72.2 16.3 9.5 2.1 Poorer 88,301 (18.7) 4.9 43.4 71.3 12.8 12.5 3.3 Middle 88,733 (18.8) 5.4 42.0 68.8 11.2 15.0 5.0 Richer 93,190 (19.8) 6.6 39.7 63.0 9.5 19.2 8.3 Richest 107,172 (22.7) 7.0 36.3 54.5 7.9 23.4 14.2 Working Not currently 187,389 (41.3) 5.1 41.9 67.0 14.1 13.8 5.1 Currently working 266,513 (58.7) 5.6 40.6 64.4 9.4 18.0 8.2 Number of children Nulliparous 124,666 (26.5) 1.8 39.5 69.6 16.4 11.1 2.9 1–4 224,224 (47.7) 6.9 40.9 64.4 9.3 18.0 8.3 4+ 121,526 (25.8) 7.0 43.6 63.6 10.3 18.1 8.1 Alcohol consumption No 81,066 (76.1) 5.6 35.5 63.5 12.3 16.7 7.4 Yes 25,492 (23.9) 1.6 27.6 62.6 14.4 14.7 8.3 Currently smoke No 433,319 (97.6) 5.4 40.9 65.4 10.8 16.7 7.2 Yes 10,866 (2.4) 7.5 40.1 62.6 21.0 11.1 5.3 Yaya et al. Results Consumption of alcohol was significantly associated with re- duction in anemia (OR = 0.69; 95%CI = 0.66–0.72), while involvement in exercise had 23% reduction in anemia, com- pared with women who do not exercise (OR = 0.77; 95%CI = 0.68–0.87). High consumption of fruit was positively asso- ciated with anemia. Underweight women were 1.09 times as likely to have anemia compared with women with normal body mass index (OR = 1.09; 95%CI = 1.05–1.12). Mean- while, overweight and obese women had 16% (OR = 0.84; 95%CI = 0.82–0.86) and 29% (OR = 0.71; 95%CI = 0.68– 0.74) reduction in the odds of anemia, compared with women with normal body mass index. Respondent’s religious background was associated with body mass index. Furthermore, married women or those living with partners had 35% reduction in underweight, compared with those never married (RR = 0.65; 95%CI = 0.64–0.67); however, married women or those currently liv- ing with partners were 1.81 (RR = 1.81; 95%CI = 1.76–1.86) and 2.91 (RR = 2.91; 95%CI = 2.77–3.04) as likely to be over- weight and obesity, compared with never married women. Parity was significantly associated with body mass index; women who had given birth had significant reduction in the risk of underweight, compared with nulliparous women; but women who had given birth had higher risk of overweight and obesity, compared with the nulliparous women relative to the normal body mass index. The consumption of alcohol was associated with body mass index. Also, involvement in smoking increased the risk of underweight (RR = 2.03; 95%CI = 1.92–2.16), but reduced the risk of overweight (RR = 0.70; 95%CI = 0.65–0.75) and obesity (RR = 0.78; 95%CI = 0.70–0.86). Moderate and high fruits and vegetable con- sumption, lower the risk of underweight, but increased the risk of overweight and obesity, compared with low consumption level relative to normal body mass index women. Details are presented in Table 4. Determinants of body mass index Multinomial logistic regression model was used to obtain the relative-risk ratio estimates of the factors associated with underweight, overweight and obesity relative to nor- mal body mass index. Results showed a decrease in rela- tive risk ratios of higher age interval for underweight women relative to normal body mass index, compared with women aged 15–19 years. On the contrary, higher age categories were more likely to be overweight and Yaya et al. Discussion reproductive activities such as child bearing with increase in sedentary lifestyles. The content of alcohol or “hot drinks” and nicotine obtain from smoking, especially for heavy users could be responsible for the breakdown of body cells leading to reduction in the risk of overweight and obesity as found in this study [30, 31]. Notwithstand- ing, smoking and alcohol consumption are known to have negative long-term health effect to the body. In this study, we extensively explored the prevalence of prominent risk factors of NCDs which include hyperten- sion, anemia, underweight, overweight and obesity among women of reproductive age in sub-Sahara Africa countries. Furthermore, we investigated the correlates of the risk factors of NCDs with the aim of understanding the pattern of the problem and how best to prevent and control it. The findings of this study revealed that a large number of women in sub-Sahara Africa countries have suffered from hypertension, anemia, underweight, over- weight and obesity. The most striking findings in our study were the very high prevalence in hypertension, anemia, underweight, overweight and obesity among women of reproductive age in sub-Sahara Africa coun- tries. These findings are similar to the reports from pre- vious studies in developing countries [2, 4, 22, 26, 27]. Consumption of fruits, vegetables and rural residence were significantly associated with reduction in the odds of hypertension among women in sub-Saharan Africa. This findings is similar to previous report [32]. The factors as- sociated with increase in the odds of hypertension were; high body mass index, smoking, large number of children ever born, high wealth index, marriage or living with a partner, education and older age. Similar findings have been made by previous studies in developing countries [10, 12, 26, 32]. Reduction in the risk of hypertension among rural dwellers could be due to differences in life styles such as dietary patterns and physical activity. Fruits and vegetables contain micronutrients and vitamins that aid metabolic activities of the body and also help in body circulatory system [33, 34]. More so, the association be- tween older age and increased risk of hypertension could be due to the biological effect of increased arterial resist- ance as a result of arterial thickening as one grows older [35, 36]. In addition, our findings of overweight and obesity as factors of hypertension agree with the general view that adiposity is linked to hypertension [37–39]. Determinants of body mass index BMC Public Health (2018) 18:1168 Page 8 of 13 Table 3 Factors associated with high blood pressure and anemia Variable High blood pressure OR 95% CI Age 15–19 1.00 20–24 2.32 1.88–2.87* 25–29 3.61 2.96–4.40* 30–34 4.40 3.61–5.37* 35–39 6.14 5.04–7.46* 40–44 8.95 7.37–10.87* 45–49 12.45 10.25–15.12* Residence Urban 1.00 Rural 0.61 0.56–0.66* Educational level None 1.00 Primary 1.85 1.66–2.07* Secondary 1.76 1.57–1.97* Higher 2.96 2.51-3.48* Religion Christianity 1.00 Islam 0.90 0.77–1.04 Others/no religion 0.70 0.56–0.88* Marital status Never married 1.00 Married/currently living with a partner 3.39 3.02–3.81* Wealth index Poorest 1.00 Poorer 1.50 1.30–1.73* Middle 1.66 1.45-1.91* Richer 2.04 1.78-2.33* Richest 2.20 1.94-2.51* Currently working 1.11 1.02–1.20* Number of children Nulliparous 1.00 1–4 4.02 3.53–4.57* 4+ 4.02 3.49-4.63* Alcohol consumption 0.27 0.22–0.32* Smoking 1.41 1.17-1.70* Exercise 1.52 1.35–1.71* Fruit consumption Low 1.00 Moderate 0.74 0.57–0.97* High 0.91 0.72–1.14 Vegetable consumption Table 3 Factors associated with high blood pressure and anemia Variable High blood pressure Anemia OR 95% CI OR 95% CI Age 15–19 1.00 1.00 20–24 2.32 1.88–2.87* 1.01 0.97-1.03 25–29 3.61 2.96–4.40* 0.96 0.94-0.99* 30–34 4.40 3.61–5.37* 0.95 0.92-0.98* 35–39 6.14 5.04–7.46* 1.02 0.98-1.05 40–44 8.95 7.37–10.87* 1.05 1.01-1.09* 45–49 12.45 10.25–15.12* 0.96 0.92-0.99* Residence Urban 1.00 1.00 Rural 0.61 0.56–0.66* 1.08 1.06-1.10* Educational level None 1.00 1.00 Primary 1.85 1.66–2.07* 0.73 0.72-0.75* Secondary 1.76 1.57–1.97* 0.71 0.70-0.73* Higher 2.96 2.51-3.48* 0.51 0.48-0.54* Religion Christianity 1.00 1.00 Islam 0.90 0.77–1.04 1.58 1.54–1.61* Others/no religion 0.70 0.56–0.88* 1.33 1.28-1.38* Marital status Never married 1.00 1.00 Married/currently living with a partner 3.39 3.02–3.81* 1.16 1.13-1.18* Wealth index Poorest 1.00 1.00 Poorer 1.50 1.30–1.73* 0.92 0.89-0.94* Middle 1.66 1.45-1.91* 0.87 0.84-0.89* Richer 2.04 1.78-2.33* 0.79 0.77-0.81* Richest 2.20 1.94-2.51* 0.68 0.66-0.70* Currently working 1.11 1.02–1.20* 0.95 0.93-0.97* Number of children Nulliparous 1.00 1.00 1–4 4.02 3.53–4.57* 1.06 1.04-1.08* 4+ 4.02 3.49-4.63* 1.18 1.15-1.21* Alcohol consumption 0.27 0.22–0.32* 0.69 0.66-0.72* Smoking 1.41 1.17-1.70* 0.97 0.92-1.02 Exercise 1.52 1.35–1.71* 0.77 0.68-0.87* Fruit consumption Low 1.00 1.00 M d 0 74 0 57 0 97* 1 07 0 96 1 19 Table 3 Factors associated with high blood pressure and anemia High blood pressure Anemia OR 95% CI OR 95% CI 1.00 1.00 2.32 1.88–2.87* 1.01 0.97-1.03 3.61 2.96–4.40* 0.96 0.94-0.99* 4.40 3.61–5.37* 0.95 0.92-0.98* 6.14 5.04–7.46* 1.02 0.98-1.05 8.95 7.37–10.87* 1.05 1.01-1.09* 12.45 10.25–15.12* 0.96 0.92-0.99* 1.00 1.00 0.61 0.56–0.66* 1.08 1.06-1.10* 1.00 1.00 1.85 1.66–2.07* 0.73 0.72-0.75* 1.76 1.57–1.97* 0.71 0.70-0.73* 2.96 2.51-3.48* 0.51 0.48-0.54* 1.00 1.00 0.90 0.77–1.04 1.58 1.54–1.61* 0.70 0.56–0.88* 1.33 1.28-1.38* 1.00 1.00 3.39 3.02–3.81* 1.16 1.13-1.18* 1.00 1.00 1.50 1.30–1.73* 0.92 0.89-0.94* 1.66 1.45-1.91* 0.87 0.84-0.89* 2.04 1.78-2.33* 0.79 0.77-0.81* 2.20 1.94-2.51* 0.68 0.66-0.70* 1.11 1.02–1.20* 0.95 0.93-0.97* 1.00 1.00 4.02 3.53–4.57* 1.06 1.04-1.08* 4.02 3.49-4.63* 1.18 1.15-1.21* 0.27 0.22–0.32* 0.69 0.66-0.72* 1.41 1.17-1.70* 0.97 0.92-1.02 1.52 1.35–1.71* 0.77 0.68-0.87* 1.00 1.00 0.74 0.57–0.97* 1.07 0.96-1.19 0.91 0.72–1.14 1.20 1.08–1.33* 1.20 1.08–1.33* Yaya et al. Determinants of body mass index BMC Public Health (2018) 18:1168 Page 9 of 13 Table 3 Factors associated with high blood pressure and anemia (Continued) Variable High blood pressure Anemia OR 95% CI OR 95% CI Low 1.00 1.00 Moderate 0.66 0.51–0.86* 1.03 0.92-1.15 High 0.89 0.70–1.11 0.99 0.90–1.11 Body mass index Normal 1.00 1.00 Underweight 0.66 0.54–0.82* 1.09 1.05-1.12* Overweight 2.44 2.19-2.72* 0.84 0.82-0.86* Obesity 5.34 4.75-5.99* 0.71 0.68-0.74* significant at p < 0.05; OR = Odds ratio Table 3 Factors associated with high blood pressure and anemia (Continued) Discussion The associations between hypertension and metabolic, socio-demographic, economic and behavioural factors revealed epidemiological transition in resource-constrained settings usually explained by economic development leading to urbanization with in- creased sedentary life style, deviations in dietary habits, so- cial stress and smoking. Sub-Saharan Africa countries are likely to be undergoing similar development. This study revealed that living in rural residence, smok- ing and alcohol consumption were associated with reduc- tion in the risk of overweight and obesity. Conversely, high consumption of fruits and vegetables were positively associated with overweight and obesity. In addition, higher number of children ever born, older age, involvement in marriage or living with a partner, women’s employment, high socio-economic status such as wealth index and edu- cation were also associated with higher risk of overweight and obesity. Our findings were consistent with the results of previous studies on the factors associated overweight and obesity [8, 9, 28]. The increase in the risk of over- weight and obesity among women who are employed, have formal education and higher wealth index showed that as the economy and individuals income improve, they tend to have sedentary lifestyles, reducing their physical activity levels. Urbanization and adoption of Western life- styles could also contribute to dietary shift to inappropri- ate food choices such as frequent intake of “fast foods” including sugar and fat leading to increase in body weight [29]. More so, the impact of marriage on overweight and obesity could be as a result of women’s sexual and Socio-economic status of women was significantly as- sociated with anemia among women. Women from rural residence, high number of children ever born, married Yaya et al. Discussion BMC Public Health (2018) 18:1168 Page 10 of 13 Table 4 Factors associated with underweight, overweight and obesity Variable Underweight Overweight Obesity RR 95% CI RR 95% CI RR 95% CI Age 15–19 1.00 1.00 1.00 20–24 0.57 0.55–0.59 1.73 1.66–1.80* 2.71 2.47–2.96* 25–29 0.53 0.52–0.56* 2.57 2.47–2.67* 5.94 5.46–6.46* 30–34 0.54 0.52–0.56* 3.19 3.06–3.31* 9.78 9.01–10.62* 35–39 0.59 0.57–0.62* 3.56 3.42–3.70* 12.72 11.72–13.82* 40–44 0.66 0.63–0.70* 3.80 3.64–3.97* 15.00 13.80–16.31* 45–49 0.73 0.69–0.76* 3.84 3.67–4.02* 15.45 14.18–16.83* Residence Urban 1.00 1.00 1.00 Rural 1.25 1.22–1.28* 0.49 0.48–0.50* 0.28 0.27–0.29* Educational level None 1.00 1.00 1.00 Primary 0.82 0.79–0.84* 1.34 1.30–1.37* 1.69 1.63–1.77* Secondary 0.77 0.75–0.79* 1.62 1.58–1.66* 2.52 2.42–2.63* Higher 0.59 0.55–0.64* 3.02 2.88–3.16* 6.34 5.98–6.72* Religion Christianity 1.00 1.00 1.00 Islam 1.44 1.41–1.48* 0.83 0.81–0.85* 0.75 0.73–0.78* Others/no religion 1.24 1.18–1.30* 0.70 0.67–0.74* 0.61 0.56–0.66* Marital status Never married 1.00 1.00 1.00 Married/currently living with a partner 0.65 0.64–0.67* 1.81 1.76–1.86* 2.91 2.77–3.04* Wealth index Poorest 1.00 1.00 1.00 Poorer 0.79 0.77–0.82* 1.34 1.29–1.39* 1.64 1.52–1.76* Middle 0.72 0.69–0.75* 1.66 1.60–1.73* 2.55 2.38–2.73* Richer 0.67 0.65–0.69* 2.33 2.25–2.41* 4.58 4.29–4.89* Richest 0.64 0.62–0.67* 3.28 3.17–3.39* 9.09 8.54–9.68* Currently working 0.69 0.68–0.71* 1.35 1.32–1.38* 1.69 1.64–1.75* Number of children Nulliparous 1.00 1.00 1.00 1–4 0.61 0.59–0.63* 1.76 1.71–1.81* 3.09 2.95–3.24* 4+ 0.69 0.66–0.71* 1.79 1.74–1.84* 3.04 2.89–3.19* Alcohol consumption 1.19 1.13–1.25* 0.89 0.85–0.94* 1.13 1.06–1.20* Smoking 2.03 1.92–2.16* 0.70 0.65–0.75* 0.78 0.70–0.86* Exercise 1.03 0.93–1.14 0.93 0.86–1.01 1.02 0.92–1.13 Fruit consumption Low 1.00 1.00 1.00 Moderate 0.66 0.56–0.79* 1.09 0.96–1.25 1.30 1.12–1.52* High 0.48 0.40–0.57* 1.47 1.31–1.66* 1.84 1.59–2.12* Vegetable consumption Table 4 Factors associated with underweight, overweight and obesity Yaya et al. BMC Public Health (2018) 18:1168 Page 11 of 13 Table 4 Factors associated with underweight, overweight and obesity (Continued) Variable Underweight Overweight Obesity RR 95% CI RR 95% CI RR 95% CI Low 1.00 1.00 1.00 Moderate 0.85 0.71–1.01 1.03 0.90–1.18 1.09 0.92–1.28 High 0.66 0.56–0.78* 1.21 1.07–1.37* 1.44 1.24–1.66* RR Relative-risk ratio; *significant at p < 0.05; Table 4 Factors associated with underweight, overweight and obesity (Continued) and those from lower socio-economic class had higher odds of anemia. The findings of this study is consistent with previous results [17, 40]. Other studies have argued that the association between the place residence and health problems in developing countries is driven mainly by higher individual-level and community-level socio- economic status. Availability of data and materials Data for this study were sourced from Demographic and Health surveys (DHS) and available here: https://www.dhsprogram.com/data/available- datasets.cfm Discussion These findings could also be due to those women from lower socio-economic status lack the finance to purchase quality food as well as sufficient quantity of foods for proper feeding. The results also showed that fruits consumption does not help to cure anemia. However, high body mass index was associated with reduction in the odds of anemia among women in sub-Saharan Africa. It is possible the women with high body mass index are usually among the well-off and those in the advantaged group, who are able to afford quality and adequate food for proper feeding [17, 40]. to focus on reducing anemia, hypertension, overweight and obesity. Interventions that focus on the modifiable determinants of overweight and obesity can help to prevent the increas- ing burden of non-communicable disease in sub-Saharan Africa. Hence, future obesity prevention programmes in resource-constrained transitioning settings will need to foremost, consider the family level with a target on well-off women. Development of community-based programmes influencing dietary patterns, physical activity, alcohol use and smoking, behavior change communication about the effect of hypertension, anemia and high body mass index, could help to reduce chronic disease associated with the metabolic risk factors of non-communicable diseases. This will also involve a multi-dimensional strategy and will re- quire intervention at the individual and population levels. Health policy makers and government need to focus on widespread interventions through community health exten- sion programmes. Authors’ contributions SY, ME and GB contributed to conception and design of the study. ME and GB contributed to acquisition, analysis and interpretation of data. All authors were involved in drafting the manuscript and have given final approval of the version to be published. Acknowledgments The authors thank the MEASURE DHS project for their support and for free access to the original data. Conclusion The goal of this study was to explore the prevalence and correlates of the risk factors of NCDs in sub-Sahara Af- rica region. In presenting an evidence-based context for government and other health policy makers on the strat- egies to reduce this burden in low-resource settings, de- tailed up-to-date information on the metabolic risk factors of non-communicable diseases has been provided in order to match this with available resources. A major strategy to prevent or control the occurrence of NCDs is Ethics approval and consent to participate Ethics approval for this study was not required since the data is secondary and is available in the public domain. More details regarding DHS data and ethical standards are available at: http://goo.gl/ny8T6X. Abbreviations B B d BMI: Body mass index; DHS: Demographic Health Survey; EAs: Enumeration areas; Hb: Hemoglobin; IRB: Institutional Review Board; NCDs: Non- communicable diseases; SSA: Sub-Saharan Africa; UN: United Nations Strength and limitations This study used nationally representative secondary data and the findings are generalizable for the women of repro- ductive age in sub-Saharan Africa countries. Sample size was considerably large that were collected from 33 coun- tries in sub-Saharan Africa which allowed widespread reporting of the prevalence of the risk factors of non-communicable diseases such as hypertension, anemia, underweight, overweight and obesity. However, a major drawback is that cross-sectional study design used cannot adequately establish causality [41, 42]. More so, since the study utilized secondary data, we were unable to measure vital micronutrients such as vitamin B12, folate and vitamin A. Caution must be taken in the generalization of the find- ings on high blood pressure as only 5 countries in sub-Saharan Africa region reported on the data. Funding The authors have no support or funding to report. Competing interests 16. Helelo TP, Gelaw YA, Adane AA. Prevalence and associated factors of hypertension among adults in Durame town, southern Ethiopia. Fuchs FD, editor. PLoS One. 2014;9(11):e112790. Competing interests Sanni Yaya is an Editorial Board member for BMC Public Health. 17. Chowdhury HA, Ahmed KR, Jebunessa F, Akter J, Hossain S, Shahjahan M. Factors associated with maternal anaemia among pregnant women in Dhaka city. BMC Womens Health 2015 Dec [cited 2018 Apr 20];15(1). Available from: http://bmcwomenshealth.biomedcentral.com/articles/10. 1186/s12905-015-0234-x Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 18. Soofi S, Khan GN, Sadiq K, Ariff S, Habib A, Kureishy S, et al. Prevalence and possible factors associated with anaemia, and vitamin B 12 and folate deficiencies in women of reproductive age in Pakistan: analysis of national- level secondary survey data. BMJ Open. 2017;7(12):e018007. 1School of International Development and Global Studies, University of Ottawa, 120 University Private, Ottawa, ON K1N 6N5, Canada. 2Department of 1School of International Development and Global Studies, University of Ottawa, 120 University Private, Ottawa, ON K1N 6N5, Canada. 2Department of Epidemiology and Medical Statistics, Faculty of Public Health, College of Medicine, University of Ibadan, Ibadan, Nigeria. 19. Çelik Kavak E, Kavak SB. The association between anemia prevalence, maternal age and parity in term pregnancies in our city. Perinatal Journal. 2017;25(1):6–10. Received: 27 April 2018 Accepted: 1 October 2018 Received: 27 April 2018 Accepted: 1 October 2018 20. Alaofè H, Burney J, Naylor R, Taren D. Prevalence of anaemia, deficiencies of iron and vitamin a and their determinants in rural women and young children: a cross-sectional study in Kalalé district of northern Benin. Public Health Nutr. 2017;20(07):1203–13. References 1. World Health Organization. Global action plan for the prevention and control of noncommunicable diseases: 2013-2020. [Internet]. 2013 [cited 2018 Apr 24]. Available from: http://apps.who.int/iris/bitstream/10665/ 94384/1/9789241506236_eng.pdf 1. World Health Organization. Global action plan for the prevention and control of noncommunicable diseases: 2013-2020. [Internet]. 2013 [cited 2018 Apr 24]. Available from: http://apps.who.int/iris/bitstream/10665/ 94384/1/9789241506236_eng.pdf 21. Lakew Y, Biadgilign S, Haile D. Anaemia prevalence and associated factors among lactating mothers in Ethiopia: evidence from the 2005 and 2011 demographic and health surveys. BMJ Open. 2015;5(4):–e006001. 2. Akinlua JT, Meakin R, Umar AM, Freemantle N. Current prevalence pattern of hypertension in Nigeria: a systematic review. Reboldi G, editor. PLoS One. 2015;10(10):e0140021. 2. Akinlua JT, Meakin R, Umar AM, Freemantle N. Current prevalence pattern of hypertension in Nigeria: a systematic review. Reboldi G, editor. PLoS One. 2015;10(10):e0140021. 22. Daru J, Zamora J, Fernández-Félix BM, Vogel J, Oladapo OT, Morisaki N, et al. Risk of maternal mortality in women with severe anaemia during pregnancy and post partum: a multilevel analysis. Lancet Glob Health. 2018;6(5):e548–54. 3. Kassa GM, Muche AA, Berhe AK, Fekadu GA. Prevalence and determinants of anemia among pregnant women in Ethiopia; a systematic review and meta-analysis. BMC Hematology. 2017 Dec [cited 2018 Apr 20];17(1). Available from: http://bmchematol.biomedcentral.com/articles/10.1186/ s12878-017-0090-z 23. Guide_to_DHS_Statistics_29Oct2012_DHSG1.pdf [Internet]. [cited 2018 Apr 10]. Available from: https://www.dhsprogram.com/pubs/pdf/DHSG1/Guide_ to_DHS_Statistics_29Oct2012_DHSG1.pdf 24. Rutstein SO, Staveteig S. Making the Demographic and Health Surveys Wealth Index Comparable, vol. 43; 2012. 4. Amugsi DA, Dimbuene ZT, Mberu B, Muthuri S, Ezeh AC. Prevalence and time trends in overweight and obesity among urban women: an analysis of demographic and health surveys data from 24 African countries, 1991 – 2014. BMJ Open. 2017 Oct;7(10):e017344. 4. Amugsi DA, Dimbuene ZT, Mberu B, Muthuri S, Ezeh AC. Prevalence and time trends in overweight and obesity among urban women: an analysis of demographic and health surveys data from 24 African countries, 1991 – 2014. BMJ Open. 2017 Oct;7(10):e017344. 25. Midi H, Sarkar SK, Rana S. Collinearity diagnostics of binary logistic regression model. Journal of Interdisciplinary Mathematics. 2010 Jun;13(3):253–67. 26. Abebe SM, Berhane Y, Worku A, Getachew A. Prevalence and associated factors of hypertension: a Crossectional community based study in Northwest Ethiopia. Li Y, editor. PLoS One. 2015;10(4):e0125210. 5. WHO \| First global ministerial conference on healthy lifestyles and NCDs control [Internet]. WHO. [cited 2018 Apr 24]. Available from: http://www. who.int/nmh/events/moscow_ncds_2011/en/ 5. WHO \| First global ministerial conference on healthy lifestyles and NCDs control [Internet]. Consent for publication bl h No consent to publish was needed for this study as we did not use any details, images or videos related to individual participants. In addition, data used is available in the public domain. Page 12 of 13 Page 12 of 13 Yaya et al. BMC Public Health (2018) 18:1168 References WHO. [cited 2018 Apr 24]. Available from: http://www. who.int/nmh/events/moscow_ncds_2011/en/ 27. Adeloye D, Basquill C. Estimating the prevalence and awareness rates of hypertension in Africa: a systematic analysis. Schnabel RB, editor. PLoS One. 2014;9(8):e104300. 6. Anderson AK. Prevalence of Anemia, Overweight/Obesity, and Undiagnosed Hypertension and Diabetes among Residents of Selected Communities in Ghana. Int J Chronic Dis [Internet]. 2017 [cited 2018 Apr 20];2017. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5574264/ 28. Lanas F, Bazzano L, Rubinstein A, Calandrelli M, Chen C-S, Elorriaga N, et al. Prevalence, distributions and determinants of obesity and central obesity in the southern cone of America. Song Y, editor. PLoS One. 2016;11(10):e0163727. 7. Biadgilign S, Mgutshini T, Haile D, Gebremichael B, Moges Y, Tilahun K. Epidemiology of obesity and overweight in sub-Saharan Africa: a protocol for a systematic review and meta-analysis. BMJ Open. 2017 Nov;7(11):e017666. 29. Burgoine T, Forouhi NG, Griffin SJ, Wareham NJ, Monsivais P. Associations between exposure to takeaway food outlets, takeaway food consumption, and body weight in Cambridgeshire, UK: population based, cross sectional study. BMJ. 2014;348:g1464. 8. Sartorius B, Veerman LJ, Manyema M, Chola L, Hofman K. Determinants of obesity and associated population Attributability, South Africa: empirical evidence from a National Panel Survey, 2008-2012. Zeeb H, editor. PLoS One 2015 Jun 10;10(6):e0130218. 30. Krukowski RA, Bursac Z, Little MA, Klesges RC. The relationship between body mass index and post-cessation weight gain in the year after quitting smoking: a cross-sectional study. Bammann K, editor. PLoS One. 2016;11(3):e0151290. 9. Mkuu RS. Prevalence and Predictors of Overweight and Obesity Among Kenyan Women. Prev Chronic Dis. 2018 [cited 2018 Apr 20];15. Available from: https://www.cdc.gov/pcd/issues/2018/17_0401.htm 31. Piirtola M, Jelenkovic A, Latvala A, Sund R, Honda C, Inui F, et al. Association of current and former smoking with body mass index: a study of smoking discordant twin pairs from 21 twin cohorts. PLoS One. 2018;13(7):e0200140. 10. Price AJ, Crampin AC, Amberbir A, Kayuni-Chihana N, Musicha C, Tafatatha T, et al. Prevalence of obesity, hypertension, and diabetes, and cascade of care in sub-Saharan Africa: a cross-sectional, population-based study in rural and urban Malawi. The Lancet Diabetes & Endocrinology. 2018 Mar;6(3):208–22. 10. Price AJ, Crampin AC, Amberbir A, Kayuni-Chihana N, Musicha C, Tafatatha T, et al. Prevalence of obesity, hypertension, and diabetes, and cascade of care in sub-Saharan Africa: a cross-sectional, population-based study in rural and urban Malawi. The Lancet Diabetes & Endocrinology. 2018 Mar;6(3):208–22. 11. World Health Organization. Global status report on noncommunicable diseases 2010. References Geneva: World Health Organization; 2011. 32. Bjertness MB, Htet AS, Meyer HE, Htike MMT, Zaw KK, Oo WM, et al. Prevalence and determinants of hypertension in Myanmar - a nationwide cross-sectional study. BMC Public Health. 2016 Dec [cited 2018 Apr 20];16(1). Available from: http://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-016-3275-7. 11. World Health Organization. Global status report on noncommunicable diseases 2010. Geneva: World Health Organization; 2011. 33. de Paula TP, Kramer CK, Viana LV, Azevedo MJ. Effects of individual micronutrients on blood pressure in patients with type 2 diabetes: a systematic review and meta-analysis of randomized clinical trials. Sci Rep. 2017 Dec [cited 2018 Aug 2];7(1). Available from: http://www.nature.com/ articles/srep40751 12. Guwatudde D, Mutungi G, Wesonga R, Kajjura R, Kasule H, Muwonge J, et al. The epidemiology of hypertension in Uganda: findings from the National non-Communicable Diseases Risk Factor Survey. Kokubo Y, editor. PLoS One. 2015 Sep 25;10(9):e0138991. 34. Dakshinamurti K, Dakshinamurti S. Blood pressure regulation and micronutrients. Nutr Res Rev. 2001;14(1):3–44. 13. Hendriks ME, Wit FWNM, Roos MTL, Brewster LM, Akande TM, de Beer IH, et al. Hypertension in sub-Saharan Africa: cross-sectional surveys in four rural and urban communities. Atashili J, editor. PLoS One. 2012;7(3):e32638. 35. Pinto E. Blood pressure and ageing. Postgrad Med J. 2007 Feb;83(976):109–14. 35. Pinto E. Blood pressure and ageing. Postgrad M 36. Demisse AG, Greffie ES, Abebe SM, Bulti AB, Alemu S, Abebe B, et al. High burden of hypertension across the age groups among residents of Gondar city in Ethiopia: a population based cross sectional study. BMC Public Health. 2017 Aug 9 [cited 2018 Aug 2];17. Available from: https://www.ncbi. nlm.nih.gov/pmc/articles/PMC5551023/ 14. Nahimana M-R, Nyandwi A, Muhimpundu MA, Olu O, Condo JU, Rusanganwa A, et al. A population-based national estimate of the prevalence and risk factors associated with hypertension in Rwanda: implications for prevention and control. BMC Public Health. 2018 Dec [cited 2018 Apr 20];18(1). Available from: http://bmcpublichealth.biomedcentral. com/articles/10.1186/s12889-017-4536-9 37. Gnatiuc L, Alegre-Díaz J, Halsey J, Herrington WG, López-Cervantes M, Lewington S, et al. Adiposity and Blood Pressure in 110 000 Mexican Adults. Hypertens Dallas Tex 1979. 2017;69(4):608–614. 15. Musinguzi G, Nuwaha F. Prevalence, Awareness and control of hypertension in Uganda. Ashton N, editor. PLoS One. 2013;8(4):e62236. Page 13 of 13 42. Yaya S, Utham OA, Adjiwanou A, Bishwajit G. Exposure to tobacco use in pregnancy and its determinants among sub-Saharan Africa women: analysis of pooled cross-sectional surveys. J Matern Fetal Neonatal Med. 2018:1–231 https://doi.org/10.1080/14767058.2018.1520835. Guimezap JT, et al. Four year trends in adiposity and its association with hypertension in serial groups of young adult university students in urban Cameroon: a time-series study. BMC Public Health. 2017 Dec [cited 2018 Aug 2];17(1). Available from: http://bmcpublichealth.biomedcentral.com/ articles/10.1186/s12889-017-4449-7 39. Doll S, Paccaud F, Bovet P, Burnier M, Wietlisbach V. Body mass index, abdominal adiposity and blood pressure: consistency of their association across developing and developed countries. Int J Obes Relat Metab Disord J Int Assoc Study Obes. 2002;26(1):48–57. 40. Lebso M, Anato A, Loha E. Prevalence of anemia and associated factors among pregnant women in southern Ethiopia: a community based cross- sectional study. Clarke SL, editor. PLoS One. 2017;12(12):e0188783. 41. Yaya S, Utham OA, Amouzou A, Bishwajit G. Mass media exposure and its impact on malaria prevention behaviour among adult women in sub- Saharan Africa: results from malaria indicator surveys. Global health research and policy. 2018;3(1):20 https://doi.org/10.1186/s41256-018-0075-x. 42. Yaya S, Utham OA, Adjiwanou A, Bishwajit G. Exposure to tobacco use in pregnancy and its determinants among sub-Saharan Africa women: analysis of pooled cross-sectional surveys. J Matern Fetal Neonatal Med. 2018:1–231 https://doi.org/10.1080/14767058.2018.1520835. 41. Yaya S, Utham OA, Amouzou A, Bishwajit G. Mass media exposure and its impact on malaria prevention behaviour among adult women in sub- Saharan Africa: results from malaria indicator surveys. Global health research and policy. 2018;3(1):20 https://doi.org/10.1186/s41256-018-0075-x. Yaya et al. BMC Public Health (2018) 18:1168 Yaya et al. BMC Public Health (2018) 18:1168 38. Choukem S-P, Kengne A-P, Nguefack M-L, Mboue-Djieka Y, Nebongo D, Guimezap JT, et al. Four-year trends in adiposity and its association with hypertension in serial groups of young adult university students in urban Cameroon: a time-series study. BMC Public Health. 2017 Dec [cited 2018 Aug 2];17(1). Available from: http://bmcpublichealth.biomedcentral.com/ articles/10.1186/s12889-017-4449-7 38. Choukem S-P, Kengne A-P, Nguefack M-L, Mboue-Djieka Y, Nebongo D, Guimezap JT, et al. Four-year trends in adiposity and its association with hypertension in serial groups of young adult university students in urban Cameroon: a time-series study. BMC Public Health. 2017 Dec [cited 2018 Aug 2];17(1). Available from: http://bmcpublichealth.biomedcentral.com/ articles/10.1186/s12889-017-4449-7 39. Doll S, Paccaud F, Bovet P, Burnier M, Wietlisbach V. Body mass index, abdominal adiposity and blood pressure: consistency of their association across developing and developed countries. Int J Obes Relat Metab Disord J Int Assoc Study Obes. 2002;26(1):48–57. 39. Doll S, Paccaud F, Bovet P, Burnier M, Wietlisbach V. Body mass index, abdominal adiposity and blood pressure: consistency of their association across developing and developed countries. Int J Obes Relat Metab Disord Int Assoc Study Obes. 2002;26(1):48–57. 40. Lebso M, Anato A, Loha E. Prevalence of anemia and associated factors among pregnant women in southern Ethiopia: a community based cross- sectional study. Clarke SL, editor. PLoS One. 2017;12(12):e0188783. 41. Yaya S, Utham OA, Amouzou A, Bishwajit G. Mass media exposure and its impact on malaria prevention behaviour among adult women in sub- Saharan Africa: results from malaria indicator surveys. Global health research and policy. 2018;3(1):20 https://doi.org/10.1186/s41256-018-0075-x. 42. Yaya S, Utham OA, Adjiwanou A, Bishwajit G. Exposure to tobacco use in pregnancy and its determinants among sub-Saharan Africa women: analysis of pooled cross-sectional surveys. J Matern Fetal Neonatal Med. 2018:1–231 https://doi.org/10.1080/14767058.2018.1520835.
https://openalex.org/W1546791830	https://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=1850&context=faculty_scholarship	English	null	The Rise and Demise of the Technology-Specific Approach to the First Amendment	Social Science Research Network	2,002	public-domain	81,351	The Rise and Demise of the Technology-Specific Approach to the The Rise and Demise of the Technology-Specific Approach to the First Amendment First Amendment Follow this and additional works at: https://scholarship.law.upenn.edu/faculty_scholarship Part of the Communications Law Commons, Constitutional Law Commons, Digital Communications and Networking Commons, Economic Policy Commons, Economic Theory Commons, First Amendment Commons, Law and Economics Commons, Political Theory Commons, Science and Technology Law Commons, and the Science and Technology Policy Commons University of Pennsylvania Carey Law School University of Pennsylvania Carey Law School Penn Carey Law: Legal Scholarship Repository Penn Carey Law: Legal Scholarship Repository All Faculty Scholarship Faculty Works 2003 The Rise and Demise of the Technology-Specific Approach to the The Rise and Demise of the Technology-Specific Approach to the First Amendment First Amendment Christopher S. Yoo University of Pennsylvania Carey Law School Author ORCID Identifier: Christopher S. Yoo 0000-0003-2980-9420 Faculty Works * Associate Professor of Law, Vanderbilt University Law School. This article beneﬁted from workshops conducted at the 2001 Annual Meeting of the Law and Society Association, the Vanderbilt University Law School, and the Washington and Lee School of Law. I would like to thank Ed Baker, Mark Brandon, Rebecca Brown, Jim Chen, Allison Danner, Neal Devins, John Goldberg, Steve Heyman, Ron Krotoszynski, Lili Levi, Tom McCoy, David McGowan, Blake Morant, Bob Rasmussen, and Marty Redish for their input on earlier drafts of this article, as well as Rob Mahini, Geni Phipps, Rob Schmoll, and Paul Werner for their expert research assistance. I should also disclose that I served as a law clerk for the courts that decided Arkansas Educational Television Commission v. Forbes, 523 U.S. 666 (1998); National Endowment for the Arts v. Finley, 524 U.S. 569 (1998); and Time Warner Entertainment Co. v. FCC, 93 F.3d 957 (D.C. Cir. 1996). The views contained in this Article are my own, as are any errors. Repository Citation Repository Citation Yoo, Christopher S., "The Rise and Demise of the Technology-Specific Approach to the First Amendment" (2003). All Faculty Scholarship. 851. https://scholarship.law.upenn.edu/faculty_scholarship/851 Repository Citation Repository Citation Yoo, Christopher S., "The Rise and Demise of the Technology-Specific Approach to the First Amendment" (2003). All Faculty Scholarship. 851. https://scholarship.law.upenn.edu/faculty_scholarship/851 https://scholarship.law.upenn.edu/faculty_scholarship/851 This Article is brought to you for free and open access by the Faculty Works at Penn Carey Law: Legal Scholarship Repository. It has been accepted for inclusion in All Faculty Scholarship by an authorized administrator of Penn Carey Law: Legal Scholarship Repository. For more information, please contact biddlerepos@law.upenn.edu. The Rise and Demise of the Technology-Speciﬁc Approach to the First Amendment CHRISTOPHER S. YOO* INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 I. THE PRIMARY FEATURES AND RATIONALES UNDERLYING THE BROADCAST MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 A. LICENSING OF BROADCAST STATIONS . . . . . . . . . . . . . . . . . . . . 255 B. NEGATIVE CONTENT RESTRICTIONS . . . . . . . . . . . . . . . . . . . . . . 260 C. AFFIRMATIVE PROGRAMMING OBLIGATIONS . . . . . . . . . . . . . . . . 263 II. THE RISE AND (IMPLICIT) DEMISE OF THE SCARCITY DOCTRINE . . . . . 266 A. “AN INSIGHT MORE FUNDAMENTAL THAN WE CAN USE”: THE ANALYTICAL CRITIQUE OF SCARCITY . . . . . . . . . . . . . . . . . . . . 267 1. The Analytical Emptiness of Scarcity . . . . . . . . . . . . . . 267 2. Existing Regulations as a Constitutional Baseline . . . . . . 269 a. The Amount of Spectrum Allocated to Broadcasting . 272 b. The Allocation of Stations to Particular Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 c. The Allocation of Individual Stations to Particular Licensees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 B. “SCARCITY IS NOT ENTIRELY A THING OF THE PAST”: THE TECHNOLOGICAL CRITIQUE OF SCARCITY . . . . . . . . . . . . . . . . . . 279 245 246 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 THE GEORGETOWN LAW JOURNAL C. WITH A WHIMPER, NOT A BANG: THE DOCTRINAL COLLAPSE OF THE SCARCITY DOCTRINE . The Rise and Demise of the Technology-Speciﬁc Approach to the First Amendment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting . . . . . . 284 2. Scarcity after Metro Broadcasting . . . . . . . . . . . . . . . . . 288 a. The Refusal to Extend Scarcity to Other Media . . . . 288 b. The Tacit Abandonment of Scarcity with Respect to Broadcasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 III. PERVASIVENESS AND ACCESSIBILITY: NEW RATIONALES FROM PACIFICA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 A. THE THEORETICAL CRITIQUE OF PACIFICA . . . . . . . . . . . . . . . . . . 293 B. THE DOCTRINAL REJECTION OF PACIFICA . . . . . . . . . . . . . . . . . . 298 1. The Refusal to Extend Paciﬁca to Other Media . . . . . . . 298 2. Paciﬁca’s Questionable Vitality with Respect to Broadcasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 C. THE TECHNOLOGICAL EVISCERATION OF PACIFICA . . . . . . . . . . . . 303 IV. BEYOND SCARCITY AND PACIFICA: THE TURN TO CIVIC REPUBLICANISM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306 A. CRITIQUES FROM OUTSIDE THE MODEL: THE ROLE OF AUTONOMY . 311 1. Theories of Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . The Rise and Demise of the Technology-Speciﬁc Approach to the First Amendment 312 a. Deontological Theories of Autonomy . . . . . . . . . . . 312 b. Democratic Theories of Autonomy . . . . . . . . . . . . . 313 c. Hybrid Theories of Autonomy . . . . . . . . . . . . . . . . . 314 2. Fiss on Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 3. Sunstein on Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . 318 a. The Role of History . . . . . . . . . . . . . . . . . . . . . . . . 319 b. The Turn to Communitarianism . . . . . . . . . . . . . . . 320 c. The Appeal to Practical Reason . . . . . . . . . . . . . . . 322 d. The Assault on Preferences . . . . . . . . . . . . . . . . . . 322 B. CRITIQUES FROM INSIDE THE MODEL: TURNING THEORY INTO PRACTICE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 246 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL C. WITH A WHIMPER, NOT A BANG: THE DOCTRINAL COLLAPSE OF THE SCARCITY DOCTRINE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting . . . . . . 284 2. Scarcity after Metro Broadcasting . . . . . . . . . . . . . . . . . 288 a. The Refusal to Extend Scarcity to Other Media . . . . 288 b. The Tacit Abandonment of Scarcity with Respect to Broadcasting . . . . . . The Rise and Demise of the Technology-Speciﬁc Approach to the First Amendment . . . . . . . . . . . . . . . . . . . . . . . 290 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 247 1. Implementation of the Fiss-Sunstein Approach . . . . . . . 325 a. Afﬁrmative Programming Obligations . . . . . . . . . . . 326 b. The Distinction Between Political and Nonpolitical Speech . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329 c. State Action Doctrine . . . . . . . . . . . . . . . . . . . . . . . 331 2. Institutional Guideposts . . . . . . . . . . . . . . . . . . . . . . . . 334 a. Democratic Process as Artifact: The Metaphor of the Town Meeting . . . . . . . . . . . . . . . . . . . . . . . . . 334 b. Jeffersonian Democracy as a Baseline . . . . . . . . . . 336 c. Institutional Faith in the FCC . . . . . . . . . . . . . . . . . 337 C. CRITIQUES OF FISS’S AND SUNSTEIN’S EMPIRICAL AND TECHNOLOGICAL FOUNDATIONS . . . . . . . . . . . . . . . . . . . . . . . 342 1. Empirical Assumptions About Television’s Role in Democracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342 2. Technological Change and the Limits of Civic Republicanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344 V. POSSIBLE EXPLANATIONS FOR THE PERSISTENCE OF BROADCAST-STYLE REGULATION . . . . . . . . . . . . . . . . . . . . . . . . . . 346 A. THE NASCENT STATE OF FIRST AMENDMENT DOCTRINE . . . The Rise and Demise of the Technology-Speciﬁc Approach to the First Amendment . . . . . . 346 B. THE EMERGENCE OF THE ECONOMICS OF PROPERTY RIGHTS . . . . . . 348 C. PROGRESSIVE FAITH IN AGENCY EXPERTISE . . . . . . . . . . . . . . . . 349 D. STARE DECISIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 E. PUBLIC CHOICE THEORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352 CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 V. POSSIBLE EXPLANATIONS FOR THE PERSISTENCE OF BROADCAST-STYLE REGULATION . . . . . . . . . . . . . . . . . . . . . . . . . . 346 A. THE NASCENT STATE OF FIRST AMENDMENT DOCTRINE . . . . . . . . . 346 B. THE EMERGENCE OF THE ECONOMICS OF PROPERTY RIGHTS . . . . . . 348 C. PROGRESSIVE FAITH IN AGENCY EXPERTISE . . . . . . . . . . . . . . . . 349 D. STARE DECISIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351 E. PUBLIC CHOICE THEORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352 CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 1. See Christopher S. Yoo, Vertical Integration and Media Regulation in the New Economy, 19 YALE J. ON REG. 171, 178 n.23, 286–90 (2002). 2. See infra notes 60–67, 80–89, 104–07 and accompanying text (detailing the relevant regulatory provisions governing broadcasting). A third model focuses on the problems associated with vertical integration in media-related industries. Unlike the Telephone and Broadcast Models, the problems that represented the focus of this regulatory approach were not unique to any particular transmission technology. On the contrary, regulators have long expressed concern about vertical integration when regulating the structure of both the broadcast and telephone industries. See generally United States v. Am. Tel. & Tel. Co., 552 F. Supp. 131 (D.D.C. 1982), aff’d, 460 U.S. 1001 (1983); FEDERAL COMMUNICATIONS COMMISSION, REPORT ON CHAIN BROADCASTING (1941). The most distinctive feature of this model, however, is the increasing reliance on access requirements as a remedy to the problems of vertical integration. Because these access requirements were primarily developed in the context of cable regulation, it seems appropriate to refer to this approach as the “Cable Model.” 47 U.S.C. §§ 532, 534, 548(c)(2)(B) (2000). ( g) 5. E.g., FCC v. League of Women Voters, 468 U.S. 367, 377 (1984) (“[W]e have recognized that ‘differences in the characteristics of new media justify differences in the First Amendment standards applied to them.’” (quoting Red Lion Broad. Co. v. FCC, 395 U.S. 367, 386 (1969))); S.E. Promotions, Ltd. v. Conrad, 420 U.S. 546, 557 (1975) (“Each medium of expression . . . must be assessed for First 4. Id. at 97 (Jackson, J., concurring). ( )( )( ) ( ) 3. 336 U.S. 77 (1949). INTRODUCTION Congress and the Federal Communications Commission (FCC) have histori- cally relied on models of media regulation that were shaped by the distinct physical and economic characteristics of the communications media governed by the Communications Act of 1934.1 Originally, person-to-person communica- tions were available only through wire-based technologies, which were character- THE GEORGETOWN LAW JOURNAL 248 [Vol. 91:245 ized by the high ﬁxed costs that have traditionally been associated with natural monopolies. As a result, such communications were governed by the “Tele- phone Model,” which incorporates the system of common carriage obligations and rate regulation that has represented the traditional regulatory response to natural monopoly. Mass communications, by contrast, originally employed the electromagnetic spectrum as its sole means of transmission. As a result, regula- tory authorities relied on what they perceived to be the unique physical qualities of spectrum-based communications in developing the system of administrative licensing, content restrictions, and afﬁrmative programming obligations associ- ated with the “Broadcast Model” of regulation.2 This technologically balkanized approach to regulation remained coherent only so long as each type of communications was available solely through a distinct means of transmission. The emergence of cable television, however, began to cause this tidy regulatory division to unravel because it allowed mass programming to reach consumers via wire-based technologies previously dedi- cated to person-to-person communications. The arrival of wireless telephony completed the collapse of this scheme by making it possible to receive person-to- person communications over the spectrum. Furthermore, the impending shift of all networks to packet switched technologies promises to cause all of the distinctions based on the means of conveyance and the type of speech conveyed to collapse entirely. Indeed, the FCC has spent much of the last half-century struggling with the policy implications of technological convergence. At the same time that Congress and the FCC were developing their technology- speciﬁc approach to telecommunications policy, the courts were engaging in the parallel process of developing a technology-speciﬁc approach to the First Amendment. Expanding on the sentiment reﬂected in Justice Robert Jackson’s declaration in Kovacs v. Cooper3 that each means of communications represents a “law unto itself,”4 the Supreme Court suggested that the First Amendment might apply differently to each communications medium.5 Although one might 2003] 249 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT have expected a plethora of technology-speciﬁc First Amendment standards to emerge, in actuality the Court restricted itself to two. Amendment purposes by standards suited to it. . . .”); see also Reno v. ACLU, 521 U.S. 844, 868 (1997) (“[E]ach medium of expression . . . may present its own problems.”); Metromedia, Inc. v. City of San Diego, 453 U.S. 490, 501 (1981) (same); FCC v. Paciﬁca Found., 438 U.S. 726, 748 (1978) (“We have long recognized that each medium of expression presents special First Amendment problems.”); Erzoznik v. City of Jacksonville, 422 U.S. 205, 220 (1975) (same). Many courts cite Joseph Burstyn, Inc. v. Wilson, 343 U.S. 495, 503 (1952), for the proposition that “[e]ach method tends to present its own peculiar problems.” As then-Chief Judge Harry Edwards has pointed out, such citations ignore that in the next sentence the Burstyn Court emphasized, “[b]ut the basic principles of freedom of speech and the press, like the First Amendment’s command, do not vary.” Id.; see also Action for Children’s Television v. FCC, 58 F.3d 654, 673 n.7 (D.C. Cir. 1995) (en banc) (Edwards, C.J., dissenting) (quoting the same phrase). 12. For a detailed review of these decisions, see infra notes 35–36, 41–59, 68–72, 90–98 and accompanying text. g 6. See, e.g., Miami Herald Publ’g Co. v. Tornillo, 418 U.S. 241 (1974). 11. LEE C. BOLLINGER, IMAGES OF A FREE PRESS 71 (1991). p y g 13. The unsettled nature of the First Amendment standard to be applied to cable is well illustrated by the composition of recent Supreme Court decisions. See, e.g., Playboy Entm’t Group v. FCC, 529 U.S. 8. 395 U.S. 367 (1969). , , ( ) ( ) ( , , g) (q g p ) 6. See, e.g., Miami Herald Publ’g Co. v. Tornillo, 418 U.S. 241 (1974). 7. 319 U.S. 190 (1943). 8. 395 U.S. 367 (1969). 9. 438 U.S. 726 (1978). 10. Id. at 748. 11. LEE C. BOLLINGER, IMAGES OF A FREE PRESS 71 (1991). 12. For a detailed review of these decisions, see infra notes 35–36, 41–59, 68–72, 90–98 and accompanying text. 13. The unsettled nature of the First Amendment standard to be applied to cable is well illustrated by the composition of recent Supreme Court decisions. See, e.g., Playboy Entm’t Group v. FCC, 529 U.S. 10. Id. at 748. INTRODUCTION The ﬁrst, most strongly associated with newspapers and other forms of print, accords an almost prohibi- tive degree of protection against governmental interference.6 The second, devel- oped in the context of broadcasting, has been more permissive of governmental regulation. In a series of decisions highlighted by NBC v. United States7 and Red Lion Broadcasting Co. v. FCC,8 the Court announced that the physical scarcity of the electromagnetic spectrum justiﬁed according broadcasters a lesser degree of First Amendment protection. The Court has relied on this so-called scarcity doctrine to uphold a wide range of structural and content regulations. The Court’s subsequent decision in FCC v. Paciﬁca Foundation9 held that the unique pervasiveness and accessibility of broadcasting provided a further basis for subjecting broadcast regulations to a lower level of First Amendment scrutiny.10 The scheme that has emerged has been aptly dubbed a “virtual celebration of public regulation”11 that has largely obscured the extent to which the current regulatory approach to broadcasting represents a constitu- tional anomaly. Had the Court failed to recognize these two grounds for upholding more intrusive regulation of broadcasting, it is almost certain that the principal features of the broadcasting model would not have withstood constitutional scrutiny.12 The possibility of variable First Amendment standards has had a dramatic impact on other media as well. Until courts resolved whether a new form of communications was more like broadcasting or more like traditional media (such as print), private parties and regulators could do little more than speculate about the constitutional propriety of any particular regulatory provision. The Supreme Court’s continuing struggle with the proper First Amendment standard to be applied to cable television nearly ﬁfty years after its emergence under- scores the signiﬁcance of the costs associated with such uncertainty.13 8. 395 U.S. 367 (1969). 250 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL Similar problems have threatened to impede the ongoing transition to digital television technologies. As the conversion of the music industry from analog- format vinyl albums to digital-format compact discs demonstrates, digital tech- nologies allow for far more efﬁcient storage and transmission of information. They also allow content to be copied without any material signal degradation. As a result, the conversion to digital technology promises to revolutionize the quantity, quality, and variety of television services available to the typical American consumer. Although many do not realize it, the conversion to digital television is already well underway. Denver Area Educ. Telecomms. Consortium, 518 U.S. at 813–14 (Thomas, J., concurring in the judgment in part and dissenting in part). 15. 47 U.S.C. § 309(j)(14)(A) (2000). The Act does contain an escape clause that will permit stations to retain both channels if 85% of the households in their broadcast area have not purchased digital receivers by that time. Id. § 309(j)(14)(B)(iii). 17. ADVISORY COMMITTEE ON PUBLIC INTEREST OBLIGATIONS OF DIGITAL TELEVISION BROADCASTERS, CHARTING THE DIGITAL BROADCASTING FUTURE (1998) [hereinafter GORE COMMISSION REPORT]. y j 16. See FCC, Remedial Steps for Failure to Comply with Digital Television Constr. Schedule, Order and Notice of Proposed Rulemaking, 17 F.C.C.R. 9962 (2002). Our First Amendment distinctions between media . . . placed cable in a doctrinal wasteland in which regulators and cable operators alike could not be sure whether cable was entitled to the substantial First Amendment protections afforded the print media or was subject to the more onerous obligations shouldered by the broadcast media. 18. See Children’s Television Obligations of Digital Television Broadcasters, Notice of Proposed Rulemaking, 15 F.C.C.R. 22946 (2000); Pub. Interest Obligations of TV Broad. Licensees, Notice of Inquiry, 14 F.C.C.R. 21663 (1999); FEDERAL COMMUNICATIONS COMMISSION, REPORT TO CONGRESS ON THE PUBLIC INTEREST OBLIGATIONS OF TELEVISION BROADCASTERS AS THEY TRANSITION TO DIGITAL TELEVISION (2001) [hereinafter REPORT TO CONGRESS], available at http://www.fcc.gov/Speeches/Kennard/Statements/ 2001/stwek106.doc. 803 (2000) (5-4 decision); Turner Broad. Sys. v. FCC, 520 U.S. 1890 (1997) [hereinafter Turner II] (5-4 decision); Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727 (1996) (fragmented decision producing six largely mutually inconsistent opinions). As Justice Thomas eloquently put it: 14. Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programming, Eighth Annual Report, 17 F.C.C.R. 1244, 1283 ¶ 82 (2002). INTRODUCTION FCC regulations required that all television stations supplement their current analog signal with a digital signal by May 1, 2002.14 The governing statute also calls for all stations to cease transmitting analog signals altogether and to broadcast solely in digital by December 31, 2006.15 Although compliance has been less than perfect,16 it is clear that digital television is in the process of becoming a reality. Despite the ongoing deployment of digital television, little progress has been made in determining whether and how digital broadcasting should be regulated. The ﬁrst concrete step was the appointment of the President’s Advisory Commit- tee on Public Interest Obligations of Digital Television Broadcasters (commonly known as the “Gore Commission”), which issued its Final Report in December 1998.17 The FCC has initiated regulatory proceedings based largely on the Report’s recommendations,18 but as of yet, none of those proceedings has reached completion. Moreover, FCC resolution of the regulatory scheme to be applied to digital television will represent only the ﬁrst round in what is almost 803 (2000) (5-4 decision); Turner Broad. Sys. v. FCC, 520 U.S. 1890 (1997) [hereinafter Turner II] (5-4 decision); Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727 (1996) (fragmented decision producing six largely mutually inconsistent opinions). As Justice Thomas eloquently put it: Our First Amendment distinctions between media . . . placed cable in a doctrinal wasteland in which regulators and cable operators alike could not be sure whether cable was entitled to the substantial First Amendment protections afforded the print media or was subject to the more onerous obligations shouldered by the broadcast media. Denver Area Educ. Telecomms. Consortium, 518 U.S. at 813–14 (Thomas, J., concurring in the judgment in part and dissenting in part). 14. Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programmi Eighth Annual Report, 17 F.C.C.R. 1244, 1283 ¶ 82 (2002). 18. See Children’s Television Obligations of Digital Television Broadcasters, Notice of Proposed Rulemaking, 15 F.C.C.R. 22946 (2000); Pub. Interest Obligations of TV Broad. Licensees, Notice of Inquiry, 14 F.C.C.R. 21663 (1999); FEDERAL COMMUNICATIONS COMMISSION, REPORT TO CONGRESS ON THE PUBLIC INTEREST OBLIGATIONS OF TELEVISION BROADCASTERS AS THEY TRANSITION TO DIGITAL TELEVISION (2001) [hereinafter REPORT TO CONGRESS], available at http://www.fcc.gov/Speeches/Kennard/Statements/ 2001/stwek106.doc. 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 251 certain to be a somewhat protracted legal battle. 19. The emerging ﬁxed wireless technologies include multipoint distribution services (MDS), instructional television ﬁxed services (ITFS), and Interactive Video Data Services (IVDS). 20. See infra subsection II.A.2. 21. See infra sections II.B, III.C. 22. See infra sections II.C, III.B. 23. See, e.g., CASS R. SUNSTEIN, DEMOCRACY AND THE PROBLEM OF FREE SPEECH (paperback ed. 1995) [hereinafter SUNSTEIN, DEMOCRACY AND FREE SPEECH]; CASS R. SUNSTEIN, THE PARTIAL CONSTITUTION (1993) [hereinafter SUNSTEIN, PARTIAL CONSTITUTION]; Cass R. Sunstein, Television and the Public Interest, 88 CAL. L. REV. 499 (2000). 24. See, e.g., OWEN M. FISS, LIBERALISM DIVIDED (1996) [hereinafter FISS, LIBERALISM DIVIDED]; OWEN M. FISS, THE IRONY OF FREE SPEECH (1996) [hereinafter FISS, IRONY OF FREE SPEECH]; Owen M. Fiss, The Censorship of Television, 93 NW. U. L. REV. 1215 (1999). 24. See, e.g., OWEN M. FISS, LIBERALISM DIVIDED (1996) [hereinafter FISS, LIBERALISM DIVIDED]; OWEN M. FISS, THE IRONY OF FREE SPEECH (1996) [hereinafter FISS, IRONY OF FREE SPEECH]; Owen M. Fiss, The Censorship of Television, 93 NW. U. L. REV. 1215 (1999). 23. See, e.g., CASS R. SUNSTEIN, DEMOCRACY AND THE PROBLEM OF FREE SPEECH (paperback ed. 1995) [hereinafter SUNSTEIN, DEMOCRACY AND FREE SPEECH]; CASS R. SUNSTEIN, THE PARTIAL CONSTITUTION (1993) [hereinafter SUNSTEIN, PARTIAL CONSTITUTION]; Cass R. Sunstein, Television and the Public Interest, 88 CAL. L. REV. 499 (2000). 19. The emerging ﬁxed wireless technologies include multipoint distribution services (MDS), instructional television ﬁxed services (ITFS), and Interactive Video Data Services (IVDS). 20 S i f b i II 2 19. The emerging ﬁxed wireless technologies include multipoint distribution services (MDS), instructional television ﬁxed services (ITFS), and Interactive Video Data Services (IVDS). 20. See infra subsection II.A.2. 21 S i f i II III 28. See Fiss, supra note 24, at 1217; Sunstein, supra note 23, at 527–31. Because this Article is organized around the primary rationales upon which the FCC and the courts relied, I omit discussion of other theories that attempt to justify upholding the Broadcast Model of regulation. Most notably, I do not discuss attempts to reconceptualize broadcast regulation as a quid pro quo for using the spectrum free of charge. For examples of such attempts, see Reed Hundt & Karen Kornbluh, Renewing the Deal Between Broadcasters and the Public: Requiring Clear Rules for Children’s Educational Television, 9 HARV. J.L. & TECH. 11, 17 (1996); Logan, supra note 26, at 1725–45; Gretchen Craft Rubin, Quid Pro Quo: What Broadcasters Really Want, 66 GEO. WASH. L. REV. 686, 687–90 (1998); Phil Weiser, Promoting Informed Deliberation and a First Amendment Doctrine for a Digital Age: Toward a New Regulatory Regime for Broadcast Regulation, in DELIBERATION, DEMOCRACY, AND THE MEDIA 11, 14–18 (Simone Chambers & Anne Costain eds., 2000). In omitting discussion of these arguments, I do not mean to suggest that I believe that they are unimportant. On the contrary, it is my hope that I will have the opportunity to address these theories in my subsequent work. 26. For other proposals invoking democratic principles to justify the existing regime of broadcast regulation, see Charles W. Logan, Jr., Getting Beyond Scarcity: A New Paradigm for Assessing the Constitutionality of Broadcast Regulation, 85 CAL. L. REV. 1687, 1718–25 (1997); Jonathan Weinberg, Broadcasting and Speech, 81 CAL. L. REV. 1101, 1193–1203 (1993); see also Ashutosh Bhagwat, Of Markets and Media: The First Amendment, the New Mass Media, and the Political Components of Culture, 74 N.C. L. REV. 141, 176–93 (1995) (arguing that democratic principles do not justify the type of content-based regulation traditionally associated with the Broadcast Model, but provide support for structural regulation); Ronald J. Krotoszynski, Jr. & A. Richard M. Blaiklock, Enhancing the Spectrum: Media Power, Democracy, and the Marketplace of Ideas, 2000 U. ILL. L. REV. 813, 863–64 (same). Other scholars have employed similar arguments as support for proposals to reform our copyright system. See, e.g., Yochai Benkler, Free as the Air to Common Use: First Amendment Constraints on Enclosure of the Public Domain, 74 N.Y.U. L. REV. 354 (1999); Neil Weinstock Netanel, Copyright and a Democratic Civil Society, 106 YALE L.J. 283 (1996). For my response, see Christopher S. Yoo, Copyright and Democracy: A Cautionary Note, 53 VAND. L. REV. 1933 (2000). 27. See FISS, LIBERALISM DIVIDED, supra note 24, at 18, 149–50; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 107–14. 25. Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969). INTRODUCTION Under the technology-speciﬁc approach to the First Amendment, the constitutionality of digital television regulations cannot ﬁnally be resolved until the courts address the constitutional standard that will be applied to this medium. Until that occurs, lingering questions about the proper scope of regulation threaten the reliance interests of broadcasters, programmers, and viewers alike in ways that can forestall the realization of the new technology’s potential beneﬁts. Similar uncertainty is likely to surround the deployment of other spectrum-based technologies that are waiting in the wings, such as third-generation wireless (3G), which promises to bring the power of the Internet into handheld devices, and an associated series of ﬁxed wireless technologies.19 In addition to these instrumental considerations, the Court’s technology- speciﬁc approach to the First Amendment suffers from certain debilitating conceptual, technological, and doctrinal shortcomings, only some of which have been recognized by commentators. In particular, I argue that courts and policy- makers have yet to recognize the manner in which the scarcity doctrine sustains and is sustained by the overriding culture of regulation that surrounds broadcast- ing. Put simply, because the amount of spectrum available at any moment is itself a product of regulation, any reliance on spectrum scarcity in effect allows regulation to serve as the constitutional justiﬁcation for other regulations.20 The impending arrival of a wide range of new broadcast technologies, such as digital transmission, improved ﬁltering, program storage, video-on-demand, spread spectrum, and packet switching, promises to alter the underlying constitu- tional analysis in fundamental ways.21 Finally, courts and commentators have largely overlooked how the Supreme Court has been distancing itself from the key decisions that serve as the foundation for the technology-speciﬁc approach to the First Amendment. Although the Court has stopped short of repudiating these precedents outright, it appears increasingly reluctant to extend these principles to other media or to continue applying them to broadcasting itself.22 Given the ongoing collapse of the rationales for treating media differently under the First Amendment, supporters of the Broadcast Model have begun to turn to alternative justiﬁcations for upholding its constitutionality. Most notably, a group of scholars led by Cass Sunstein23 and Owen Fiss24 have drawn on the 252 THE GEORGETOWN LAW JOURNAL [Vol. INTRODUCTION 91:245 language in Red Lion, opining that “[i]t is the right of the viewers and listeners, not the right of the broadcasters, which is paramount,”25 to elaborate a new, largely instrumental basis rooted in the principles of civic republicanism for upholding the constitutionality of the existing regime of broadcast regulation.26 At some points, their theories offer a sweeping reconceptualization of the First Amendment regarding all media rather than an attempt to rationalize regulatory intervention concerning broadcasting in particular.27 In other words, their solu- tion to the impending collapse of the distinction between broadcasting and other language in Red Lion, opining that “[i]t is the right of the viewers and listeners, not the right of the broadcasters, which is paramount,”25 to elaborate a new, largely instrumental basis rooted in the principles of civic republicanism for upholding the constitutionality of the existing regime of broadcast regulation.26 language in Red Lion, opining that “[i]t is the right of the viewers and listeners, not the right of the broadcasters, which is paramount,”25 to elaborate a new, largely instrumental basis rooted in the principles of civic republicanism for upholding the constitutionality of the existing regime of broadcast regulation.26 At some points, their theories offer a sweeping reconceptualization of the First Amendment regarding all media rather than an attempt to rationalize regulatory intervention concerning broadcasting in particular.27 In other words, their solu- tion to the impending collapse of the distinction between broadcasting and other media is not to ﬁght it, but rather to argue that such a reuniﬁcation should center on the broadcast precedents rather than the precedents following the print tradition. The more recent statements of their theories retreat from this notion and instead suggest that the unique role that television plays in the public discourse justiﬁes differential treatment under the Constitution.28 These myriad developments suggest the need for a critical reexamination of the historical and contemporary justiﬁcations underlying the Court’s technology- speciﬁc approach to the First Amendment. Part I describes three distinguishing features of the Broadcast Model and analyzes the largely overlooked fact that were it not for the different constitutional standard applied to broadcast regula- tion, each would represent an archetypical violation of conventional First 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 253 Amendment principles. It then reviews the precedents to determine the justiﬁca- tions upon which the FCC and courts have relied to uphold these features. INTRODUCTION From that, only two rationales emerge: (1) the physical scarcity of the electromagnetic spectrum and (2) the uniquely pervasive and accessible nature of the broadcast medium. Each of the succeeding Parts examines one of the two rationales for the technology-speciﬁc approach to the First Amendment identiﬁed in Part I. Part II evaluates the scarcity doctrine from an analytical, technological, and doctrinal standpoint. The analytical discussion begins with a review of the economic analysis ﬁrst offered by Ronald Coase29 that has become one of the most established critiques of scarcity. I then offer a novel conceptual criticism that focuses on the tendency of the scarcity doctrine to permit regulation to serve as the constitutional justiﬁcation for more regulation. The basic problem stems from the Supreme Court’s increasing willingness, as demonstrated in two recent cases, to blindly accept other forms of regulation as part of the constitutional baseline when evaluating the constitutionality of any particular restriction. As a result, the scarcity doctrine largely ignores that scarcity is primarily the product of other regulatory decisions made by Congress and the FCC. Thus, relying on scarcity effectively allows regulation to serve as the constitutional justiﬁcation for additional regulation. This effect transforms the technology-speciﬁc First Amendment from a transitional ambiguity into a theory that allows the overrid- ing culture of regulation to become self-reinforcing. The other contribution of Part II is to evaluate whether recent technological and doctrinal developments have undermined scarcity as a constitutional justiﬁ- cation. With respect to technology, I conclude that the arrival of a wide range of new television technologies promises to render the scarcity doctrine an empiri- cal nullity—if it has not done so already. With respect to recent judicial decisions, I suggest that the scarcity doctrine may not have as much vitality as many believe. Although the doctrine has never been explicitly repudiated, a close reading of subsequent decisions reveals that the Court has severely limited its scope. Not only has the Supreme Court consistently refused to extend it to other media, what has gone largely unrecognized is that the Court’s most recent broadcast cases have exhibited an extreme reluctance to rely on the scarcity doctrine even with respect to broadcasting. Together these decisions provide reason to believe that the Court may be closer to abandoning the doctrine than is generally thought. 29. R.H. Coase, The Federal Communications Commission, 2 J.L. & ECON. 1 (1959). THE GEORGETOWN LAW JOURNAL the courts have essentially foreclosed the possibility that Paciﬁca will be extended to any other media and have raised questions about its continuing applicability to broadcasting. A technological critique of Paciﬁca provides additional reasons to question whether it can support the technology-speciﬁc approach to the First Amendment. The emergence of the V-chip and the impend- ing arrival of video-on-demand promise to render unconstitutional any attempts to regulate television based on its supposed pervasiveness and accessibility. Part IV examines the extent to which the civic republican visions of Cass Sunstein and Owen Fiss can provide an alternative constitutional justiﬁcation for upholding the Broadcast Model. My analysis yields three core criticisms. First, the civic republican theories fail to come to grips with the major tradition in our First Amendment jurisprudence that holds autonomy to be an important free speech value. Second, even if one were to acquiesce to ignoring autonomy, the civic republican theories ultimately prove to be quite problematic from the standpoint of implementation. Not only are they too incompletely speciﬁed to provide any useful guidance on how to resolve free speech issues, they also fail to engage in any serious comparative institutional analysis of the relative merits of government intervention over private ordering. As a result, they fail to offer any response to the long tradition of distrust of governmental regulation of speech that characterizes the enactment history and doctrinal development of the First Amendment. Finally, the civic republican theories do not advance any plausible solution to certain technological obstacles to their theories. Thus, even if one were to accept all of the arguments offered by Sunstein and Fiss, it remains difﬁcult, if not impossible, to see how their theories would bring about the world that they seek. The failure of the justiﬁcations for enforcing a technology-speciﬁc vision of the First Amendment makes its continuing existence something of a puzzle. Part V explores possible reasons for the adoption and persistence of the technology- speciﬁc approach. Upon close analysis, it appears that the Broadcast Model is in large part a reﬂection of the state of judicial thinking at the time the broadcast industry emerged. Although that provides an explanation for why courts chose a technology-speciﬁc approach in the ﬁrst instance, it fails to explain the doc- trine’s persistence. I then explore public choice explanations for the continua- tion of the doctrine. THE GEORGETOWN LAW JOURNAL Although public choice theory offers a plausible descriptive explanation for the persistence of the technology-speciﬁc approach to the First Amendment, it fails to provide any normative justiﬁcation for sustaining it. INTRODUCTION Part III addresses the second traditional justiﬁcation for according broadcast- ing a lesser degree of constitutional protection than other media, which is the unique pervasiveness and accessibility of broadcasting that formed the basis of the Court’s decision in Paciﬁca. A close analysis of Paciﬁca’s reasoning reveals that the opinion suffers from deep analytical ﬂaws that place it in conﬂict with the overall sweep of the Court’s First Amendment jurisprudence. As a doctrinal matter, moreover, a review of the most recent indecency decisions reveals that 254 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 30. The Court recently summarized its broadcast precedents as follows: The scarcity of broadcast frequencies thus required the establishment of some regulatory mechanism to divide the electromagnetic spectrum and assign speciﬁc frequencies to particu- lar broadcasters. In addition, the inherent physical limitation on the number of speakers who may use the broadcast medium has been thought to require some adjustment in traditional First Amendment analysis to permit the Government to place limited content restraints, and impose certain afﬁrmative obligations, on broadcast licensees. 34. Id. at 759–60 (citation omitted); see also Riley v. Nat’l Fed’n of Blind, 487 U.S. 781, 801 (1988) (“[W]hen a State enacts a statute requiring periodic licensing of speakers, at least when the law is I. THE PRIMARY FEATURES AND RATIONALES UNDERLYING THE BROADCAST MODEL This Part describes the three principal regulatory features of the Broadcast Model: (1) the licensing of broadcast stations, (2) the imposition of content- based restrictions on certain types of programming, and (3) the imposition of afﬁrmative speech obligations. It also explores why courts have sustained these features despite the apparent archetypical violation of the First Amendment constituted by each feature. A review of the relevant precedents reveals that the FCC and the courts have relied exclusively on two rationales to justify sustain- 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 255 ing broadcast regulations that otherwise would not have survived constitutional scrutiny. The ﬁrst rationale is the physical scarcity of the electromagnetic spectrum.30 The second rationale is the uniquely pervasive and accessible nature of broadcasting upon which the Court relied in Paciﬁca. f ( 33. 486 U.S. 750, 760 (1988). § ( )( ) ( g ) ( ); § ( )( )( ) ( ) 32. See Thomas v. Chicago Park Dist., 534 U.S. 316, 320 (2002) (noting that “the core abuse against which [the First Amendment] was directed was the scheme of licensing laws implemented by the monarch and Parliament . . . in 16th- and 17[th]-century England” and attributing opposition to licensing to Blackstone); see also City of Lakewood v. Plain Dealer Publ’g Co., 486 U.S. 750, 760 (1988) (“Indeed, a law requiring the licensing of printers has historically been declared the archetypal censorship statute.” (citing 4 WILLIAM BLACKSTONE, COMMENTARIES 152)); Lovell v. City of Grifﬁn, 303 U.S. 444, 451 (1938) (“The struggle for the freedom of the press was primarily directed against the power of the licensor. It was against that power that John Milton directed his assault by his ‘Appeal for the Liberty of Unlicensed Printing.’ . . . While this freedom from previous restraint upon publication cannot be regarded as exhausting the guaranty of liberty, the prevention of that restraint was a leading purpose in the adoption of the constitutional provision.”); Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713 (1931) (acknowledging that the struggle against “the legislative power of the licenser” has animated attempts to protect the freedom of the press since the days of Blackstone); Respublica v. Oswald, 1 U.S. (1 Dall.) 319, 325, 328 n. (1788) (noting that the impermissibility of “any attempt to fetter the press by the institution of a licenser” had been settled since the reign of William the Third and had been recognized by Blackstone). See generally Philip Hamburger, The Development of the Law of Seditious Libel and the Control of the Press, 37 STAN. L. REV. 661 (1985) (discussing the history of the English licensing statute). A. LICENSING OF BROADCAST STATIONS Since the enactment of the Communications Act of 1934, the FCC has had the unquestioned authority to license broadcast stations in accordance with the “public interest, convenience, and necessity.”31 That the FCC is permitted to license broadcast speakers represents something of a First Amendment anomaly. The Supreme Court has repeatedly recognized that licensing of media speakers has been regarded as the quintessential threat to the freedom of speech since the days of Blackstone and John Milton.32 As the Court observed in City of Lakewood v. Plain Dealer Publishing Co.,33 schemes that require licenses to be periodically renewed pose problems similar to those posed by prior restraints, in that they allow licensors to “measure their probable content or viewpoint by speech already uttered. A speaker in this position is under no illusion regarding the effect of the ‘licensed’ speech on the ability to continue speaking in the future.”34 The applicability of this observation to broadcasting is evident when one considers that during the early days of broadcast regulation, the Federal 31. 47 U.S.C. § 307(c)(1) (initial grants) (2000); id. § 309(k)(1)(A) (renewals). 32. See Thomas v. Chicago Park Dist., 534 U.S. 316, 320 (2002) (noting that “the core abuse against which [the First Amendment] was directed was the scheme of licensing laws implemented by the monarch and Parliament . . . in 16th- and 17[th]-century England” and attributing opposition to licensing to Blackstone); see also City of Lakewood v. Plain Dealer Publ’g Co., 486 U.S. 750, 760 (1988) (“Indeed, a law requiring the licensing of printers has historically been declared the archetypal censorship statute.” (citing 4 WILLIAM BLACKSTONE, COMMENTARIES 152)); Lovell v. City of Grifﬁn, 303 U.S. 444, 451 (1938) (“The struggle for the freedom of the press was primarily directed against the power of the licensor. It was against that power that John Milton directed his assault by his ‘Appeal for the Liberty of Unlicensed Printing.’ . . . While this freedom from previous restraint upon publication cannot be regarded as exhausting the guaranty of liberty, the prevention of that restraint was a leading purpose in the adoption of the constitutional provision.”); Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713 (1931) (acknowledging that the struggle against “the legislative power of the licenser” has animated attempts to protect the freedom of the press since the days of Blackstone); Respublica v. Oswald, 1 U.S. 30. The Court recently summarized its broadcast precedents as follows: f g y g g 33. 486 U.S. 750, 760 (1988). 30. The Court recently summarized its broadcast precedents as follows: The scarcity of broadcast frequencies thus required the establishment of some regulatory mechanism to divide the electromagnetic spectrum and assign speciﬁc frequencies to particu- lar broadcasters. In addition, the inherent physical limitation on the number of speakers who may use the broadcast medium has been thought to require some adjustment in traditional First Amendment analysis to permit the Government to place limited content restraints, and impose certain afﬁrmative obligations, on broadcast licensees. Turner Broad. Sys. v. FCC, 512 U.S. 622, 637–38 (1994) (citations omitted). 31. 47 U.S.C. § 307(c)(1) (initial grants) (2000); id. § 309(k)(1)(A) (renewals). 32. See Thomas v. Chicago Park Dist., 534 U.S. 316, 320 (2002) (noting that “the core abuse against which [the First Amendment] was directed was the scheme of licensing laws implemented by the monarch and Parliament . . . in 16th- and 17[th]-century England” and attributing opposition to licensing to Blackstone); see also City of Lakewood v. Plain Dealer Publ’g Co., 486 U.S. 750, 760 (1988) (“Indeed, a law requiring the licensing of printers has historically been declared the archetypal censorship statute.” (citing 4 WILLIAM BLACKSTONE, COMMENTARIES 152)); Lovell v. City of Grifﬁn, 303 U.S. 444, 451 (1938) (“The struggle for the freedom of the press was primarily directed against the power of the licensor. It was against that power that John Milton directed his assault by his ‘Appeal for the Liberty of Unlicensed Printing.’ . . . While this freedom from previous restraint upon publication cannot be regarded as exhausting the guaranty of liberty, the prevention of that restraint was a leading purpose in the adoption of the constitutional provision.”); Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713 (1931) (acknowledging that the struggle against “the legislative power of the licenser” has animated attempts to protect the freedom of the press since the days of Blackstone); Respublica v. Oswald, 1 U.S. (1 Dall.) 319, 325, 328 n.* (1788) (noting that the impermissibility of “any attempt to fetter the press by the institution of a licenser” had been settled since the reign of William the Third and had been recognized by Blackstone). See generally Philip Hamburger, The Development of the Law of Seditious Libel and the Control of the Press, 37 STAN. L. REV. 661 (1985) (discussing the history of the English licensing statute). 33 486 U S 750 760 (1988) Turner Broad. Sys. v. FCC, 512 U.S. 622, 637–38 (1994) (citations omitted). 35. See Reuel E. Schiller, Free Speech and Expertise: Administrative Censorship and the Birth of the Modern First Amendment, 86 VA. L. REV. 1, 46 (2000). See generally David J. Brennan, Printing in England and Broadcasting in Australia: A Comparative Study of Regulatory Impulse, 22 ADELAIDE L. REV. 63 (2000) (comparing the regulation of printing in England from 1484 to 1695 with the regulation of broadcasting in Australia during the 20th Century, and explaining how special control over each medium arose). 36. As a result, it is somewhat surprising that the scholarly literature on broadcast regulation has made only passing reference to the Court’s licensing jurisprudence and has done so largely in terms supportive of the constitutionality of broadcast licensing. See Reed E. Hundt, The Public’s Airways: What Does the Public Interest Require of Television Broadcasters?, 45 DUKE L.J. 1089, 1114–15 (1996); Harry Kalven, Broadcasting, Public Policy, and the First Amendment, 10 J.L. & ECON. 15, 48 (1967); Logan, supra note 26, at 1743; Matthew L. Spitzer, An Introduction to the Law and Economics of the V-Chip, 15 CARDOZO ARTS & ENT. L.J. 429, 476 (1997); Weinberg, supra note 26, at 1108, 1113, 1131, 1134. 41. See KENNETH CULP DAVIS, ADMINISTRATIVE LAW § 12 (1951); HENRY J. FRIENDLY, THE FEDERAL ADMINISTRATIVE AGENCIES: THE NEED FOR BETTER DEFINITION OF STANDARDS 55 (1962); DAVID SCHOEN- BROD, POWER WITHOUT RESPONSIBILITY 100 (1993); Ernest Gellhorn, Returning to First Principles, 36 AM. U. L. REV. 345, 347 (1987); Gary Lawson, Delegation and the Constitution, CATO REV. BUS. & GOVT. REG., No. 2-2, at 29 (1999); Glen O. Robinson, Title I, The Federal Communications Act: An directly aimed at speech, it is subject to First Amendment scrutiny to ensure that the licensor’s discretion is suitably conﬁned.” (citing City of Lakewood, 486 U.S. at 755–56)); cf. Bantam Books, Inc. v. Sullivan, 372 U.S. 58, 68–71 (1963) (recognizing that a regulatory scheme that permitted an administrative agency to threaten speakers with post hoc punishment for speech they had already uttered was tantamount to a system of prior restraint). 37. FW/PBS v. City of Dallas, 493 U.S. 215, 225–26 (1990); City of Lakewood, 486 U.S. at 769–70; Niemotko v. Maryland, 340 U.S. 268, 271 (1951). 39. Id. at 758. 38. 486 U.S. at 757–58. 42. See Poulos v. New Hampshire, 345 U.S. 395, 403–05 (1953); Cox v. New Hampshire, 312 U.S. 569, 575–76 (1941). Essay on Origins and Regulatory Purpose, in A LEGISLATIVE HISTORY OF THE COMMUNICATIONS ACT OF 1934, at 3, 14 (Max D. Paglin ed., 1989); cf. Mistretta v. United States, 488 U.S. 361, 416 (1989) (Scalia, J., dissenting) (dissenting from the majority’s decision to uphold the constitutionality of the Sentencing Guidelines by ﬁnding that Congress had provided the Sentencing Commission with sufﬁcient guidance to curtail excessive legislative discretion). Commentators recognized this problem from the start. See Louis G. Caldwell, The Standard of Public Interest, Convenience or Necessity as Used in the Radio Act of 1927, 1 AIR L. REV. 295, 296 (1930) (“‘Public interest, convenience or necessity’ means about as little as any phrase that the drafter of the Act could have used . . . .”). It is true that the Court has regarded the public interest standard as constituting an “intelligible principle” sufﬁcient to satisfy the demands of the nondelegation doctrine. See, e.g., NBC v. United States, 319 U.S. 190, 226 (1943). Those decisions, however, simply addressed the limitations on the structure of government embodied in the Constitution and did not consider whether the First Amendment compelled greater speciﬁcity. A. LICENSING OF BROADCAST STATIONS (1 Dall.) 319, 325, 328 n.* (1788) (noting that the impermissibility of “any attempt to fetter the press by the institution of a licenser” had been settled since the reign of William the Third and had been recognized by Blackstone). See generally Philip Hamburger, The Development of the Law of Seditious Libel and the Control of the Press, 37 STAN. L. REV. 661 (1985) (discussing the history of the English licensing statute). 256 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL Radio Commission (FRC) initially required stations to seek renewal every six months. The ever-present threat of nonrenewal allowed the FRC to exercise direct control over the content of broadcast programming.35 The subsequent lengthening of the license term mitigates, but does not eliminate, this effect.36 Consequently, the Court has long required that all licensing schemes contain clear standards that cabin the licensing authority’s discretion.37 As the Court noted in City of Lakewood, such standards reduce the likelihood of the self- censorship that occurs when the inability to discern the line between the permissible and the impermissible leads speakers to restrict themselves to speech favored by the licensing authority.38 Furthermore, clear standards en- hance judicial review by “provid[ing] the guideposts that . . . allow courts quickly and easily to determine whether the licensor is discriminating against disfavored speech.”39 Without such guideposts, it is “far too easy” for licensing ofﬁcials to justify their actions through “post hoc rationalizations” and “the use of shifting or illegitimate criteria.”40 Measured against these criteria, the broadcasting regime clearly fails. Commen- tators have long criticized the basic licensing standard provided by the Commu- nications Act of 1934 (that is, the public interest) as the epitome of analytical emptiness.41 The FCC could have construed the statute in a manner that 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 257 provided sufﬁcient standards to constrain its discretion.42 Indeed, the FCC may be obliged to do so as a matter of administrative law.43 Unfortunately, the FCC has never applied the public interest standard in a manner that imposed any meaningful constraints. 43. See KENNETH CULP DAVIS, DISCRETIONARY JUSTICE: A PRELIMINARY INQUIRY 55–57, 219–20 (1969); Lisa Schultz Bressman, Disciplining Delegation After Whitman v. American Trucking Ass’ns, 87 CORNELL L. REV. 452, 480–81, 484–85 (2002). A. LICENSING OF BROADCAST STATIONS The ad hoc approach initially followed by the FCC was denounced by both congressional observers and distinguished commentators for its arbitrariness and its susceptibility to manipulation for political purposes.44 L l if h d d 45 l d l if b l i provided sufﬁcient standards to constrain its discretion.42 Indeed, the FCC may be obliged to do so as a matter of administrative law.43 Unfortunately, the FCC has never applied the public interest standard in a manner that imposed any meaningful constraints. The ad hoc approach initially followed by the FCC was denounced by both congressional observers and distinguished commentators for its arbitrariness and its susceptibility to manipulation for political purposes.44 Later attempts to clarify these standards45 employed a multifactor balancing test that was far too malleable to foster consistency or predictability in the decision- Later attempts to clarify these standards45 employed a multifactor balancing test that was far too malleable to foster consistency or predictability in the decision- Essay on Origins and Regulatory Purpose, in A LEGISLATIVE HISTORY OF THE COMMUNICATIONS ACT OF 1934, at 3, 14 (Max D. Paglin ed., 1989); cf. Mistretta v. United States, 488 U.S. 361, 416 (1989) (Scalia, J., dissenting) (dissenting from the majority’s decision to uphold the constitutionality of the Sentencing Guidelines by ﬁnding that Congress had provided the Sentencing Commission with sufﬁcient guidance to curtail excessive legislative discretion). Commentators recognized this problem from the start. See Louis G. Caldwell, The Standard of Public Interest, Convenience or Necessity as Used in the Radio Act of 1927, 1 AIR L. REV. 295, 296 (1930) (“‘Public interest, convenience or necessity’ means about as little as any phrase that the drafter of the Act could have used . . . .”). It is true that the Court has regarded the public interest standard as constituting an “intelligible principle” sufﬁcient to satisfy the demands of the nondelegation doctrine. See, e.g., NBC v. United States, 319 U.S. 190, 226 (1943). Those decisions, however, simply addressed the limitations on the structure of government embodied in the Constitution and did not consider whether the First Amendment compelled greater speciﬁcity. 42. See Poulos v. New Hampshire, 345 U.S. 395, 403–05 (1953); Cox v. New Hampshire, 312 U.S. 569, 575–76 (1941). 42. See Poulos v. New Hampshire, 345 U.S. 395, 403–05 (1953); Cox v. New Hampshire, 312 U.S. 569, 575–76 (1941). 43. A. LICENSING OF BROADCAST STATIONS See KENNETH CULP DAVIS, DISCRETIONARY JUSTICE: A PRELIMINARY INQUIRY 55–57, 219–20 (1969); Lisa Schultz Bressman, Disciplining Delegation After Whitman v. American Trucking Ass’ns, 87 CORNELL L. REV. 452, 480–81, 484–85 (2002). 43. See KENNETH CULP DAVIS, DISCRETIONARY JUSTICE: A PRELIMINARY INQUIRY 55–57, 219–20 (1969); Lisa Schultz Bressman, Disciplining Delegation After Whitman v. American Trucking Ass’ns, 87 CORNELL L. REV. 452, 480–81, 484–85 (2002). 44. See AD HOC ADVISORY COMM. ON ALLOCATIONS TO THE SEN. COMM. ON INTERSTATE AND FOREIGN COMMERCE, 85TH CONG., ALLOCATION OF TV CHANNELS, (Comm. Print 1958); COMMISSION ON ORGANIZA- TION OF THE EXECUTIVE BRANCH OF THE GOVERNMENT [HOOVER COMMISSION], COMMITTEE ON INDEPENDENT 44. See AD HOC ADVISORY COMM. ON ALLOCATIONS TO THE SEN. COMM. ON INTERSTATE AND FOREIGN COMMERCE, 85TH CONG., ALLOCATION OF TV CHANNELS, (Comm. Print 1958); COMMISSION ON ORGANIZA- TION OF THE EXECUTIVE BRANCH OF THE GOVERNMENT [HOOVER COMMISSION], COMMITTEE ON INDEPENDENT REGULATORY COMMISSIONS, TASK FORCE REPORT app. N, at 40 (1949); FRIENDLY, supra note 41, at 72; EMMETTE S. REDFORD, PRESIDENT’S ADVISORY COMMITTEE ON GOVERNMENT ORGANIZATION, THE PRESIDENT AND THE REGULATORY COMMISSION 27 (1960); Bernard Schwartz, Comparative Television and the Chancellor’s Foot, 47 GEO. L.J. 655, 689–90, 694 (1959); Louis L. Jaffe, The Scandal in TV Licensing, HARPER’S WKLY., Sept. 1957, at 77, 79. REGULATORY COMMISSIONS, TASK FORCE REPORT app. N, at 40 (1949); FRIENDLY, supra note 41, at 72; EMMETTE S. REDFORD, PRESIDENT’S ADVISORY COMMITTEE ON GOVERNMENT ORGANIZATION, THE PRESIDENT AND THE REGULATORY COMMISSION 27 (1960); Bernard Schwartz, Comparative Television and the Chancellor’s Foot, 47 GEO. L.J. 655, 689–90, 694 (1959); Louis L. Jaffe, The Scandal in TV Licensing, HARPER’S WKLY., Sept. 1957, at 77, 79. p 45. Policy Statement on Comparative Broad. Hearings, Public Notice, 1 F.C.C.2d 393 (1965). Speciﬁcally, the 1965 Policy Statement identiﬁed seven criteria to guide its decisionmaking: 1. whether the applicant owned other media properties; 2. whether the license holder would act as station manager; 3. the programming proposed by the applicant; 4. the applicant’s past broadcast record; 5. whether the applicant proposed to operate the station part-time or twenty-four hours a day; 6. the applicant’s character; 7. other factors. 46. See Amendment of Comm’n’s Rules to Allow Selection from Among Competing Applicants for New AM, FM, &Television Stations by Random Selection (Lottery), Notice of Proposed Rulemaking, 4 F.C.C.R. 2256, 2259 ¶¶ 24–28 (1989) [hereinafter Lottery NPRM]; Robert A. Anthony, Towards Simplicity and Rationality in Comparative Broadcast Licensing Proceedings, 24 STAN. L. REV. 1, 39 (1971); Henry Geller, The Comparative Renewal Process in Television: Problems and Suggested Solutions, 61 VA. L. REV. 471, 501–02 (1975); Mathew L. Spitzer, Multicriteria Choice Processes: An Application of Public Choice Theory to Bakke, the FCC, and the Courts, 88 YALE L.J. 717, 748–51 (1979); see also ROGER NOLL ET AL., ECONOMIC ASPECTS OF TELEVISION REGULATION 112–14 (1973). Thomas Hazlett recounts two particularly dramatic illustrations of the malleability of the FCC’s licensing criteria. In one case, the full Commission did not offer any reasons whatsoever for its licensing decision and simply directed its Ofﬁce of Opinions and Review to justify its decision post hoc. In another case, the Ofﬁce of Opinions and Review drafted a hundred-page opinion supporting a particular licensing decision, only to redraft it into an equally lengthy opinion justifying the opposite result after the full Commission abruptly changed its mind. Thomas W. Hazlett, Digitizing “Must- Carry” Under Turner Broadcasting v. FCC (1997), 8 SUP. CT. ECON. REV. 141, 177–78 (2000). 51. See, e.g., Action for Children’s Television v. FCC, 59 F.3d 1249, 1266 (D.C. Cir. 1995) (Tatel, J., dissenting); Cmty.-Serv. Broad. of Mid-Am., Inc. v. FCC, 593 F.2d 1102, 1116 (D.C. Cir. 1978); Ill. Citizens Comm. for Broad. v. FCC, 515 F.2d 397, 409 (D.C. Cir. 1974) (Bazelon, C.J., dissenting from rehearing en banc); Yale Broad., 478 F.2d at 605–06 (Bazelon, C.J., dissenting from denial of rehearing en banc); Brandywine-Main Line Radio, Inc. v. FCC, 473 F.2d 16, 78 n.60 (D.C. Cir. 1972) (Bazelon, C.J., dissenting); Inquiry into Section 73.190 of Comm’n’s Rules & Regulations Concerning Alterna- tives to Gen. Fairness Doctrine Obligations of Broad. Licensees, Report, 102 F.C.C.2d 145 (1985); LUCAS A. POWE, JR., AMERICAN BROADCASTING AND THE FIRST AMENDMENT 175–82 (1987); ITHIEL DE SOLA POOL, TECHNOLOGIES OF FREEDOM 121–27 (1983); David L. Bazelon, FCC Regulation of the Telecommu- nications Press, 1975 DUKE L.J. 213, 215–16, 235–36, 244–51. y 47. 310 U.S. 88 (1940). 50. See Louis L. Jaffe, WHDH: The FCC and Broadcasting License Renewals, 82 HARV. L. REV. 1693, 1695 (1969). 48. Id. at 97. 49. Quoted in Yale Broad. Co. v. FCC, 478 F.2d 594, 605 n.22 (D.C. Cir. 1973) (Bazelon, C.J., dissenting from denial of rehearing en banc). 7. other factors. Id. at 394–99. By its own terms, the 1965 Policy Statement applied only to the issuance of new licenses. Id. at 393 n.1. The FCC later made it the operative standard for license renewals as well. Seven League Prods., Inc. (WIII), 1 F.C.C.2d 1597, 1598 ¶ 5 (1965) (mem. op. & order); see also Greater Boston Television Corp. v. FCC, 444 F.2d 841, 857 (D.C. Cir. 1970). 258 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 making process.46 The FCC’s grant of the vast majority of renewal applications did not preclude the existence of real harms stemming from the lack of clear standards. As the Court noted in Thornhill v. Alabama47: The power of the licensor . . . is pernicious not merely by reason of the censure of particular comments but by the reason of the threat to censure comments on matters of public concern. It is not merely the sporadic abuse of power by the censor but the pervasive threat inherent in its very existence that constitutes the danger to freedom of discussion.48 Charles Whitehead, who was Director of Telecommunications Policy during the Nixon Administration, put it even more bluntly when he said, “[t]he main value of the sword of Damocles is that it hangs, not that it drops.”49 Confronted with such plastic criteria, applicants for broadcast licenses have had little choice but to tailor their speech to what they perceived to be the FCC’s preferences.50 The reported cases and academic commentary cite numerous historical examples of such self-censorship.51 Indeed, it now appears that the chill caused by the lack of clear standards was far from accidental. In delivering his celebrated “Vast Wasteland” speech, then-FCC Chairman Newton Minow 2003] 259 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT asked rhetorically “Why should you want to know how close you can come to the edge of the cliff?” and admonished broadcasters against “playing brinkman- ship with the public interest.”52 The entire enterprise has become so disrepu- table that it has drawn criticism from commentators sympathetic to the Broadcast Model53 and from the FCC itself.54 After the D.C. 55. See Bechtel v. FCC, 10 F.3d 875, 878–86 (D.C. Cir. 1993) [hereinafter Bechtel II]; Flagstaff Broad. Found. v. FCC, 979 F.2d 1566, 1571 (D.C. Cir. 1992); Bechtel v. FCC, 957 F.2d 873, 881 (D.C. Cir. 1992) [hereinafter Bechtel I]. The D.C. Circuit’s principal concern was that these standards had been promulgated through a policy statement that had never been subjected to the rigors of the notice and comment process. As a result, the FCC had never provided a sufﬁcient administrative justiﬁcation for the criteria that it chose. 53. See Hundt & Kornbluh, supra note 28, at 12, 13; Weinberg, supra note 26, at 1115, 1204. 54. See, e.g., Formulation of Policies & Rules Relating to Broad. Renewal Applicants, Third Further Notice of Inquiry and Notice of Proposed Rulemaking, 4 F.C.C.R. 6363, 6363–64 ¶ 6 (1989); Amendment of Comm’n’s Rules to Allow Selection form Among Competing Applicants for New AM, FM & Television Stations by Random Selection Lottery, Notice of Proposed Rulemaking, 4 F.C.C.R. 2256, 2259 ¶¶ 24–29 (1989). In fact, the FCC has proposed abolishing the comparative renewal process altogether. See Formulation of Policies Relating to Broad. Renewal Applicant, Stemming from Compara- tive Hearing Process, Report and Order, 66 F.C.C.2d 419, 429 ¶ 22 (1977); Formulation of Policies Relating to Broad. Renewal Applicant, Stemming from Comparative Hearing Process, Notice of Inquiry, 88 F.C.C.2d 120, 121 n.1 (1981). 59. Implementation of Sections 204(a) & 204(c) of Telecomms. Act of 1996 (Broad. License Renewal Procedures), Order, 11 F.C.C.R. 6363, 6364 ¶ 5 (1996). 56. See 47 U.S.C. § 309(l) (2000). Although the statute only required auctions for applications ﬁled on or after July 1, 1997, Congress gave the FCC the option of using auctions to resolve the backlog of disputes that had accrued following the invalidation of the FCC’s previous licensing criteria in Bechtel II. See id. In addition, the statute speciﬁcally exempted noncommercial educational and public broadcast stations from the auction mandate. As a result, the FCC has continued to rely on comparative hearings in allocating certain licenses, including those for low power FM radio. Creation of Low Power Radio Serv., Report and Order, 15 F.C.C.R. 2205, 2258–59 ¶ 136 (2000). 52. Newton M. Minow, Address to the National Association of Broadcasters (May 9, 1961), reprinted in NEWTON N. MINOW & CRAIG L. LAMAY, ABANDONED IN THE WASTELAND: CHILDREN, TELEVISION AND THE FIRST AMENDMENT 185, 192 (1995). p 57. See 47 U.S.C. § 309(k) (2000). 57. See 47 U.S.C. § 309(k) (2000). 58. Speciﬁcally, the statute requires the FCC ﬁrst to consider: (1) whether the station has served the public interest, (2) whether the station has committed any serious violations, and (3) whether the station’s nonserious violations, taken together, constitute a pattern of abuse. Id. § 309(k)(1). Only if the renewal applicant fails to satisfy these three criteria and the FCC fails to ﬁnd any mitigating factors that would justify the imposition of lesser sanctions can the FCC compare the incumbent’s application with that of other applicants. Id. § 309(k)(3); see also Lili Levi, Not with a Bang but a Whimper: Broadcast License Renewal and the Telecommunications Act of 1996, 29 CONN. L. REV. 243, 279–80 (1996). , ( ) 53. See Hundt & Kornbluh, supra note 28, at 12, 13; Weinberg, supra note 26, at 1115, 1204. 28, at 12, 13; Weinberg, supra note 26, at 1115, 1204. 7. other factors. Circuit eventually struck down the entire regime as arbitrary and capricious,55 Congress subsequently enacted legislation completely eliminating the FCC’s discretion over new li- censes56 and severely limiting its discretion with respect to renewals.57 Al- though the revised renewal process reduced the scope of the FCC’s licensing discretion, it still requires the FCC to base its renewal decisions on a series of highly subjective inquiries, including the perpetuation of the requirement that the FCC ﬁnd renewal to be in the public interest.58 The FCC has ruled that existing policy statements and case law will serve as the primary guide to its public interest analysis.59 As a result, it is hard to regard the FCC’s licensing 57. See 47 U.S.C. § 309(k) (2000). 58. Speciﬁcally, the statute requires the FCC ﬁrst to consider: (1) whether the station has served the public interest, (2) whether the station has committed any serious violations, and (3) whether the station’s nonserious violations, taken together, constitute a pattern of abuse. Id. § 309(k)(1). Only if the renewal applicant fails to satisfy these three criteria and the FCC fails to ﬁnd any mitigating factors that would justify the imposition of lesser sanctions can the FCC compare the incumbent’s application with that of other applicants. Id. § 309(k)(3); see also Lili Levi, Not with a Bang but a Whimper: Broadcast License Renewal and the Telecommunications Act of 1996, 29 CONN. L. REV. 243, 279–80 (1996). 260 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 scheme as providing the type of clear standards needed to withstand conven- tional First Amendment scrutiny. The Supreme Court has nonetheless relied on the scarcity doctrine to uphold the constitutionality of the scheme. As the Court observed in its seminal decision in NBC v. United States,60 the unique physical characteristics of broadcasting dictated that only a ﬁxed number of speakers could broadcast at any particular time, and private ordering had proven ill-suited to apportioning opportunities to speak.61 As a result, the Court concluded that government allocation was essential if the spectrum was to be developed as a resource.62 Consistent with the spirit of the day,63 the Court also concluded that the realities of modern governance made it impossible for such licensing to be governed by criteria any clearer than the public interest standard. 7. other factors. When faced with a “a ﬁeld of regulation which was both new and dynamic,” it was inevitable that Congress would give the FCC powers that were “not niggardly[,] but expansive.”64 As a result, the Court regarded the public interest standard to be “as concrete as the complicated factors for judgment in such a ﬁeld of delegated authority per- mit.”65 It is only by using the scarcity doctrine to frame the issue as a Hobson’s choice between allowing broadcasting to ﬂounder in chaos or imposing public interest licensing that the Court’s decision seem explicable. As the Court later noted in Red Lion Broadcasting Co. v. FCC,66 “It would be strange if the First Amendment, aimed at protecting and furthering communications, prevented the Government from making radio communication possible by requiring licenses to broadcast and by limiting the number of licenses so as not to overcrowd the spectrum.”67 60. 319 U.S. 190 (1943). 61. Id. at 212–13. 62. Id. at 226. 63. See JAMES LANDIS, THE ADMINISTRATIVE PROCESS 95–103, 132–40 (1938) (noting that ﬂexible standards were necessary to facilitate judicial review of administrative adjudications and to preserve the traditional roles of administrative adjudications as fact-ﬁnding proceedings, while according deference to the lawmaking power of Article III courts). 64. 319 U.S. at 219. 65. Id. at 216 (internal quotation marks omitted). 66. 395 U.S. 367 (1969). 67. Id. at 389; see also id. at 376–77 (“Without government control, the medium would be of little use because of the cacophony of competing voices, none of which could be clearly and predictably heard. Consequently the Federal Radio Commission was established to allocate frequencies among competing applicants in a manner responsive to the public ‘convenience, interest, or necessity.’” (footnote omitted)). 68. 47 U.S.C. § 326 (2000). ( ) 67. Id. at 389; see also id. at 376–77 (“Without government control, the medium would be of little use because of the cacophony of competing voices, none of which could be clearly and predictably heard. Consequently the Federal Radio Commission was established to allocate frequencies among competing applicants in a manner responsive to the public ‘convenience, interest, or necessity.’” (footnote omitted)). 63. See JAMES LANDIS, THE ADMINISTRATIVE PROCESS 95–103, 132–40 (1938) (noting that ﬂexible standards were necessary to facilitate judicial review of administrative adjudications and to preserve the traditional roles of administrative adjudications as fact-ﬁnding proceedings, while according deference to the lawmaking power of Article III courts). 66. 395 U.S. 367 (1969). 65. Id. at 216 (internal quotation marks omitted). 60. 319 U.S. 190 (1943). 61. Id. at 212–13. 62. Id. at 226. 63. See JAMES LANDIS, THE ADMINISTRATIVE PROCESS 95–103, 132–40 (1938) (noting that ﬂexible standards were necessary to facilitate judicial review of administrative adjudications and to preserve the traditional roles of administrative adjudications as fact-ﬁnding proceedings, while according deference to the lawmaking power of Article III courts). 64. 319 U.S. at 219. 65 d 216 (i l i k i d) 68. 47 U.S.C. § 326 (2000). 62. Id. at 226. 64. 319 U.S. at 219. 72. See, e.g., Paciﬁca Found. v. FCC, 438 U.S. 726, 758–61 (1978) (discussing George Carlin’s “Filthy Words” monologue); Yale Broad. Co. v. FCC, 478 F.2d 594, 603–04 (D.C. Cir. 1973) (Bazelon, C.J., dissenting from denial of rehearing en banc) (considering songs with drug-oriented lyrics); Inﬁnity Broad. Corp., 2 F.C.C.R. 2705 (1987) (mem. op. & order) (concerning “The Howard Stern Show”); Paciﬁca Found., 2 F.C.C.R. 2698 (1987) (mem. op. & order) (ruling on a broadcast of a play about two gay men dying of AIDS). For overviews of the FCC’s actions, see POWE, supra note 51, at 165–90; THOMAS G. KRATTENMAKER & LUCAS A. POWE, JR., REGULATING BROADCAST PROGRAMMING 104–19 (1994); Lili Levi, The Hard Case of Broadcast Indecency, 20 N.Y.U. REV. L. & SOC. CHANGE 49, 85–112, 117–19, 132–36 (1992–93). 71. FRC THIRD ANNUAL REPORT, supra note 70, at 34, 36; FRC SECOND ANNUAL REPORT, supra note 70, at 160, 169; Howard A. Shelanski, The Bending Line Between Conventional “Broadcast” and Wireless “Carriage,” 97 COLUM. L. REV. 1048, 1054–57 (1997) (describing the “public interest limitation on content” required by both the Radio Act of 1927 and Telecommunications Act of 1934, and providing examples of denials of license renewals subsequent to each Act). For fulsome and colorful recounts of the most salient cases in this regard, see POWE, supra note 51, at 13–30. types of advertising. Id. §§ 1304, 1307. 70. See Red Lion Broad. Co. v. FCC, 395 U.S. 367, 394 (1969) (“In applying [the public interest] standard the Commission for 40 years has been choosing licensees based in part on their program proposals.”); NBC v. United States, 319 U.S. 190, 217 (1943) (“Since the very inception of federal regulation by radio, comparative considerations as to the services to be rendered have governed the application of the standard of ‘public interest, convenience, or necessity.’”); KFKB Broad. Ass’n v. FRC, 47 F.2d 670, 672 (D.C. Cir. 1931) (“[T]he commission is necessarily called upon to consider the character and quality of the service to be rendered.”); 3 FRC ANNUAL REPORT 3 (1929) [hereinafter FRC THIRD ANNUAL REPORT] (“[T]he kind of service rendered by a station must be a means of appraising its relative standing and must be considered by the commission in making assignments.”); 2 FRC ANNUAL REPORT 161 (1928) [hereinafter FRC SECOND ANNUAL REPORT] (“[T]he commission believes it is entitled to consider the program service rendered by the various applicants, to compare them, and to favor those which render the best service.”). 75. Id. at 538; accord Turner Broad. Sys. v. FCC, 512 U.S. 622, 641 (1994) (“At the heart of the First Amendment lies the principle that each person should decide for him or herself the ideas and beliefs deserving of expression, consideration, and adherence.”). 73. See, e.g., McIntyre v. Ohio Elections Comm’n, 514 U.S. 334, 345–46 (1995); R.A.V. v. City of St. Paul, 505 U.S. 377, 382 (1992); Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105, 116 (1991); Regan v. Time, Inc., 468 U.S. 641, 648–49 (1984). 74 447 U S 530 (1980) 69. 18 U.S.C. § 1464 (2000). Other direct content-based restrictions include prohibitions of certain types of advertising. Id. §§ 1304, 1307. g p 76. Texas v. Johnson, 491 U.S. 397, 414 (1989) (“If there is a bedrock principle underlying the First Amendment, it is that the government may not prohibit the expression of an idea simply because 74. 447 U.S. 530 (1980). B. NEGATIVE CONTENT RESTRICTIONS The second principal regulatory feature of the Broadcast Model is the prohibi- tion of certain categories of programming. Although the Communications Act of 1934 speciﬁcally prohibits censorship,68 the overall regulatory regime does, 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 261 in fact, authorize some content-based restrictions of speech. The most signiﬁ- cant direct restriction is the statute forbidding the broadcast of indecent or profane speech.69 In addition, the federal licensing authorities have long regu- lated content indirectly by treating program content as one of their primary licensing criteria.70 The FRC initially refused to relicense broadcasters who transmitted speech that it perceived to be of low-value.71 The FCC continued the practice, primarily focusing on speech that it perceived to be indecent.72 The Supreme Court has repeatedly held that content-based restrictions on speech are presumptively invalid.73 As the Court noted in Consolidated Edison Co. v. Public Service Commission,74 “[t]o allow a government the choice of permissible subjects for public debate would be to allow that government control over the search for political truth.”75 Content restrictions based upon the supposed offensiveness of speech are particularly problematic.76 Rather than 73. See, e.g., McIntyre v. Ohio Elections Comm’n, 514 U.S. 334, 345–46 (1995); R.A.V. v. City of St. Paul, 505 U.S. 377, 382 (1992); Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105, 116 (1991); Regan v. Time, Inc., 468 U.S. 641, 648–49 (1984). 74 447 U S 530 (1980) 75. Id. at 538; accord Turner Broad. Sys. v. FCC, 512 U.S. 622, 641 (1994) (“At the heart of the First Amendment lies the principle that each person should decide for him or herself the ideas and beliefs deserving of expression, consideration, and adherence.”). g p ) 76. Texas v. Johnson, 491 U.S. 397, 414 (1989) (“If there is a bedrock principle underlying the First Amendment, it is that the government may not prohibit the expression of an idea simply because 262 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 providing a basis for restricting speech, “the fact that society may ﬁnd speech offensive” instead provides “a reason for according it constitutional protec- tion.”77 The prohibition of content-based speech restrictions applies with equal force in the context of licensing. 80. See KFKB Broad. Ass’n v. FRC, 47 F.2d 670, 672 (D.C. Cir. 1931); FRC SECOND ANNUAL REPORT, supra note 70, at 161, 170. See generally Erwin G. Krasnow & Jack N. Goodman, The “Public Interest” Standard: The Search for the Holy Grail, 50 FED. COMM. L.J. 605, 629 (1997) (“Scarcity, of course, has always been the underlying raison d’eˆtre for broadcast regulation. Because one person’s transmission is another’s interference, Congress concluded that the federal government has the duty both to select who may and who may not broadcast . . . .”). 82. Id. at 216–17 (reasoning that if content were an impermissible basis for awarding licenses, “how could the Commission choose between two applicants for the same facilities . . . ?”). y p y ) 77. Paciﬁca Found. v. FCC, 438 U.S. 726, 745 (1978); accord NBC v. FCC, 516 F.2d 1101, 1179 (D.C. Cir. 1975) (Bazelon, C.J., dissenting from the order vacating the previous order granting rehearing en banc) (“I seem to recall that it is controversial speech and not the right to assert that one’s speech is not really controversial which should be protected.”). society ﬁnds the idea itself offensive or disagreeable.”); Carey v. Population Servs. Int’l, 431 U.S. 678, 701 (1977) (“At least where obscenity is not involved, we have consistently held that the fact that protected speech may be offensive to some does not justify its suppression.”); Street v. New York, 394, U.S. 576, 592 (1969) (“It is ﬁrmly settled that under our Constitution the public expression of ideas may not be prohibited merely because the ideas are themselves offensive to some of the hearers.”). y g , , ( ) 79. Forsyth County v. Nationalist Movement, 505 U.S. 123, 134 (1992) (internal quotation marks omitted) (quoting FCC v. League of Women Voters, 468 U.S. 364, 383 (1984)). y 81. 319 U.S. 190, 216 (1943). p y p ) 78. City of Lakewood v. Plain Dealer Publ’g Co., 486 U.S. 750, 760 (1988). p ) d v. Plain Dealer Publ’g Co., 486 U.S. 750, 760 (1988) 83. See Paciﬁca Found. (WBAI-FM), 56 F.C.C.2d 94, 96–97 ¶¶ 8–9 (1975) (mem. op. & order), on reconsideration, 59 F.C.C.2d 892 (1976), rev’d, 556 F.2d 9 (D.C. Cir. 1977), rev’d, 438 U.S. 726 (1978); see also Report on Broad. of Violent, Indecent, & Obscene Material, 51 F.C.C.2d 418, 419–20, 423 (1975); cf. Sonderling Broad. Corp. (WGLD-FM), 41 F.C.C.2d 777, 782 ¶ 16 (1973) (mem. op. & order), aff’d sub nom., Ill. Citizens Comm. for Broad. v. FCC, 515 F.2d 397 (D.C. Cir. 1974); Jack Straw Mem. Found. (KRAB-FM), 29 F.C.C.2d 334, 336 ¶¶ 5–6 (1971) (initial decision of hearing examiner); E. Educ. Radio (WUHY-FM), 24 F.C.C.2d 408, 410–12 ¶ 8 (1970) (notice of apparent liability). 89. Paciﬁca Found., 2 F.C.C.R. 2698, 2699 ¶¶ 10–11 (1987) (mem. op. & order); see also Action for Children’s Television v. FCC, 58 F.3d 654, 664–69 (D.C. Cir. 1995) (en banc); Indus. Guidance on Comm’n’s Case Law Interpreting 18 U.S.C. § 1464 & Enforcement Policies Regarding Broad. Indecency, Policy Statement, 16 F.C.C.R. 7999, 8000 ¶ 4 (2001). B. NEGATIVE CONTENT RESTRICTIONS Although the Court emphasized the narrowness of its holding,88 the rationale outlined in Paciﬁca represented a justiﬁcation for extending a lower level of First Amendment protection to broadcasting that was completely independent of the scarcity doctrine. Supreme Court acceptance of the pervasiveness and accessibility rationales prompted a dramatic shift in the manner in which the FCC justiﬁed the constitutionality of the negative content restrictions that it imposed. In time, the FCC would explicitly abandon any attempt to justify its indecency restrictions on the scarcity doctrine and would opt instead to rely solely on the rationales announced in Paciﬁca.89 y y ( ) 90. See FEDERAL COMMUNICATIONS COMMISSION, PUBLIC SERVICE RESPONSIBILITIES OF BROADCAST LICENS- EES (1946) (emphasizing the importance of public affairs and local programming), reprinted in 88. Id. at 750. 87. Id. at 749–50. 84. 438 U.S. 726 (1978). 85. Id. at 748. B. NEGATIVE CONTENT RESTRICTIONS Even when the state is justiﬁed in imposing periodic licensing, “the Constitution requires that the [licensing authority] establish neutral criteria to insure that the licensing decision is not based on the content or viewpoint of the speech being considered.”78 Even if the criteria applied do not discriminate explicitly on the basis of content, the Court will nonetheless treat them as content-based if the licensing authority “must necessarily examine the content of the message that is conveyed” when deciding whether to issue the license.79 Given the explicitly content-based nature of the direct prohibitions on inde- cent broadcasts and the FCC’s licensing criteria, the broadcast regulatory regime appears to contradict conventional First Amendment principles. The courts and the regulatory authorities initially relied exclusively on the scarcity doctrine to uphold attempts to restrict low-value speech.80 As the Court noted in NBC v. United States, because “[t]he facilities of radio are limited and therefore precious,” the FCC was justiﬁed in preventing “wasteful use” of broadcast frequencies.81 In addition, the public interest standard necessarily required the FCC to award licenses to the applicant who would provide the best service to the community. Simply put, the Court could not conceive how the FCC could identify the best applicant without considering the content of the service to be transmitted.82 In the 1970s, the FCC began to articulate what it acknowledged was a new 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 263 rationale for upholding content-based restrictions of broadcast speech83 that eventually received the approval of the Supreme Court in FCC v. Paciﬁca Foundation.84 Calling the reasons for the First Amendment distinction between broadcasting and other media “complex,” the Court argued that a lower degree of First Amendment protection was justiﬁed, in part, because broadcasting was “uniquely pervasive,” in that it often “confronts the citizen . . . in the privacy of the home, where the individual’s right to be left alone plainly outweighs the First Amendment rights of an intruder.”85 In addition, “[b]ecause the broadcast audience is constantly tuning in and out, prior warnings cannot completely protect the listener or viewer from unexpected program content.”86 Finally, the Court relied on the fact that broadcasting is “uniquely accessible to children.”87 Although the Court emphasized the narrowness of its holding,88 the rationale outlined in Paciﬁca represented a justiﬁcation for extending a lower level of First Amendment protection to broadcasting that was completely independent of the scarcity doctrine. 94. Federal law requires broadcasters to carry political advertisements by candidates for federal ofﬁce. 47 U.S.C. § 312(a)(7) (2000). The statute also strictly limits the amount that the broadcaster can charge for doing so. Id. § 315(b)(1). Broadcasters may refuse to carry political advertisements by state and local candidates. See CBS, Inc. v. Democratic Nat’l Comm., 412 U.S. 94 (1973) [hereinafter CBS v. DNC]. Should broadcasters choose to accept any such advertisements, however, they must do so on a nondiscriminatory basis. 47 U.S.C. § 315(a) (2000). 93. Revision of Programming & Commercialization Policies, Ascertainment Requirements & Pro- gram Log Requirements for Commercial Television Stations, Report and Order, 98 F.C.C.2d 1076, 1093 ¶ 36, 1095–96 ¶¶ 40–43 (1984), on reconsideration, 104 F.C.C.2d 358 (1986), aff’d in part and remanded in part sub nom. Action for Children’s Television v. FCC, 821 F.2d 741 (D.C. Cir. 1987); Syracuse Peace Council, 2 F.C.C.R. 5043 (1987) (mem. op. & order) (repealing the Fairness Doctrine), aff’d, 867 F.2d 654 (D.C. Cir. 1989); Repeal of the Personal Attack and Personal Editorial Rules, Order, 15 F.C.C.R. 20697 (2000) (repealing the rules for both broadcast and cable television operators). THE GEORGETOWN LAW JOURNAL placed particular emphasis on the obligation to provide news, public affairs, and other nonentertainment programming.91 The most celebrated of these obliga- tions is the now-defunct Fairness Doctrine, which required that broadcasters cover controversial issues of public importance in a balanced manner.92 The FCC has since abolished the bulk of its afﬁrmative programming require- ments.93 Only a handful of afﬁrmative broadcasting obligations remain in force today. For example, the federal election statute places broadcasters under an obligation to carry political advertisements for certain candidates for public ofﬁce.94 Similarly, the FCC has construed the Children’s Television Act of 1990 (CTA)95 as requiring that all broadcasters provide at least three hours of DOCUMENTS OF AMERICAN BROADCASTING 151, 198–208 (Frank J. Kahn ed., 2d ed. 1973); FRC THIRD ANNUAL REPORT, supra note 70, at 34 (requiring broadcasters to provide “a well-rounded program, in which entertainment, consisting of music both classical and lighter grades, religion, education and instruction, important public events, discussion of public questions, weather, market reports, and news, and matters of interest to all members of the family ﬁnd a place”); En Banc Programming Inquiry, Report and Statement of Policy, 44 F.C.C. 2303 (1960) (identifying fourteen “major elements usually necessary to meet the public interest”). The fourteen elements listed by the 1960 En Banc Programming Inquiry were: 1. opportunity for local self-expression 8. political broadcasts 9. agricultural programs 2. development and use of local talent 10. news programs 3. programs for children 11. weather and market reports 4. religious programs 12. sports programs 5. educational programs 13. service to minority groups 6. public affairs programs 14. entertainment programs 7. editorials by licensees 91. Id. See Formulation of Policies & Rules Relating to Broad. Renewal Applicants, Competing Applicants, & Other Participants to Comparative Renewal Process and to Prevention of Abuses of Renewal Process, Second Further Notice of Inquiry and Notice of Proposed Rulemaking, 3 F.C.C.R. 5179, 5191 ¶ 84 (1988); Formulation of Policies & Rules Relating to Broad. Renewal Applicants, Competing Applicants, & Other Participants to Comparative Renewal Process & to Prevention of Abuses of Renewal Process, Third Further Notice of Inquiry and Notice of Proposed Rulemaking, 4 F.C.C.R. 6363, 6368 n.11 (1989) (citing Radio Station WABZ, Inc., 90 F.C.C.2d 818, 840–42 (1982), aff’d sub nom. Victor Broad., Inc. v. FCC, 722 F.2d 756, 762, 764–65 (D.C. Cir. 1983)). 91. Id. See Formulation of Policies & Rules Relating to Broad. y 95. 47 U.S.C. § 303b(a)(2) (2000). C. AFFIRMATIVE PROGRAMMING OBLIGATIONS Finally, federal regulators have long ruled that licensees are required to do more than just refrain from uttering certain types of dispreferred speech. Licensees also bear an afﬁrmative obligation to carry certain types of preferred speech. Early efforts speciﬁed a broad range of programming that the FRC and FCC expected broadcasters to include.90 In subsequent decisions, the FCC 264 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 THE GEORGETOWN LAW JOURNAL Renewal Applicants, Competing Applicants, & Other Participants to Comparative Renewal Process and to Prevention of Abuses of Renewal Process, Second Further Notice of Inquiry and Notice of Proposed Rulemaking, 3 F.C.C.R. 5179, 5191 ¶ 84 (1988); Formulation of Policies & Rules Relating to Broad. Renewal Applicants, Competing Applicants, & Other Participants to Comparative Renewal Process & to Prevention of Abuses of Renewal Process, Third Further Notice of Inquiry and Notice of Proposed Rulemaking, 4 F.C.C.R. 6363, 6368 n.11 (1989) (citing Radio Station WABZ, Inc., 90 F.C.C.2d 818, 840–42 (1982), aff’d sub nom. Victor Broad., Inc. v. FCC, 722 F.2d 756, 762, 764–65 (D.C. Cir. 1983)). Competing Applicants, & Other Participants to Comparative Renewal Process & to Prevention of Abuses of Renewal Process, Third Further Notice of Inquiry and Notice of Proposed Rulemaking, 4 F.C.C.R. 6363, 6368 n.11 (1989) (citing Radio Station WABZ, Inc., 90 F.C.C.2d 818, 840–42 (1982), aff’d sub nom. Victor Broad., Inc. v. FCC, 722 F.2d 756, 762, 764–65 (D.C. Cir. 1983)). 92. See Applicability of the Fairness Doctrine in Handling of Controversial Issues of Public Importance, 29 Fed. Reg. 10415 (July 9, 1964); Editorializing by Broad. Licensees, 13 F.C.C. 1246, 1257–58 (1949). 93. Revision of Programming & Commercialization Policies, Ascertainment Requirements & Pro- gram Log Requirements for Commercial Television Stations, Report and Order, 98 F.C.C.2d 1076, 1093 ¶ 36, 1095–96 ¶¶ 40–43 (1984), on reconsideration, 104 F.C.C.2d 358 (1986), aff’d in part and remanded in part sub nom. Action for Children’s Television v. FCC, 821 F.2d 741 (D.C. Cir. 1987); Syracuse Peace Council, 2 F.C.C.R. 5043 (1987) (mem. op. & order) (repealing the Fairness Doctrine), aff’d, 867 F.2d 654 (D.C. Cir. 1989); Repeal of the Personal Attack and Personal Editorial Rules, Order, 15 F.C.C.R. 20697 (2000) (repealing the rules for both broadcast and cable television operators). 93. Revision of Programming & Commercialization Policies, Ascertainment Requirements & Pro- gram Log Requirements for Commercial Television Stations, Report and Order, 98 F.C.C.2d 1076, 1093 ¶ 36, 1095–96 ¶¶ 40–43 (1984), on reconsideration, 104 F.C.C.2d 358 (1986), aff’d in part and remanded in part sub nom. Action for Children’s Television v. FCC, 821 F.2d 741 (D.C. Cir. 1987); Syracuse Peace Council, 2 F.C.C.R. 5043 (1987) (mem. op. & order) (repealing the Fairness Doctrine), aff’d, 867 F.2d 654 (D.C. Cir. 1989); Repeal of the Personal Attack and Personal Editorial Rules, Order, 15 F.C.C.R. THE GEORGETOWN LAW JOURNAL 20697 (2000) (repealing the rules for both broadcast and cable television operators). 93. Revision of Programming & Commercialization Policies, Ascertainment Requirements & Pro- gram Log Requirements for Commercial Television Stations, Report and Order, 98 F.C.C.2d 1076, 1093 ¶ 36, 1095–96 ¶¶ 40–43 (1984), on reconsideration, 104 F.C.C.2d 358 (1986), aff’d in part and remanded in part sub nom. Action for Children’s Television v. FCC, 821 F.2d 741 (D.C. Cir. 1987); Syracuse Peace Council, 2 F.C.C.R. 5043 (1987) (mem. op. & order) (repealing the Fairness Doctrine), aff’d, 867 F.2d 654 (D.C. Cir. 1989); Repeal of the Personal Attack and Personal Editorial Rules, Order, 15 F.C.C.R. 20697 (2000) (repealing the rules for both broadcast and cable television operators). 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 265 children’s programming each week.96 The FCC has, however, proposed expand- ing the afﬁrmative programming obligations demanded of digital broadcasters. Speciﬁcally, it envisions requiring digital broadcasters to provide more local, issue-oriented programming; a larger number of public service announcements; additional children’s programming; mandatory rights of reply; and greater coverage of political campaigns, including free air time for political candi- dates.97 Moreover, the CTA requirement that the FCC consider whether any renewal applicant has served the public interest leaves open the possibility that the FCC may subject broadcasters to additional afﬁrmative programming obliga- tions in the future.98 The Court has traditionally cast a jaundiced eye towards attempts to impose afﬁrmative speech obligations on other media.99 For example, in Miami Herald Publishing Co. v. Tornillo,100 the Court struck down a statute requiring newspa- pers to provide a right of reply as an impermissible intrusion into their editorial judgment and control.101 The Court implicitly reafﬁrmed this principle in Turner Broadcasting System v. FCC (Turner I),102 which involved a First Amendment challenge to a statute requiring local cable operators to carry all local broadcast stations within their service area (commonly known as “must- carry”). In upholding the must-carry requirements, the Court suggested that it would have held otherwise if the statute had attempted to specify the content of the speech that cable operators were obligated to carry.103 The Court’s decisions in Turner I and Tornillo leave little doubt that applying these general First Amendment principles to the type of afﬁrmative program- ming requirements historically imposed on broadcasters would lead to their invalidation. 97. See REPORT TO CONGRESS, supra note 18, at 5–10, 16, 24–26; GORE COMMISSION REPORT, supra note 17, at 48, 56–59; Hundt, supra note 36, at 1096; William E. Kennard, “What Does $70 Billion Buy You Anyway?”: Rethinking Public Interest Requirements at the Dawn of the Digital Age, Remarks at the Museum of Television and Radio (Oct. 10, 2000), available at http://www.fcc.gov/Speeches/Kennard/ 2000/spwek023.html. 96. Policies & Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10662 ¶ 5, 10718–19 ¶ 120 (1996). A broadcaster could provide less than three hours per week if they were able to demonstrate a commitment to education and informing children that was the equivalent of three hours. Id. 99. For an excellent discussion of the cognitive and dignitary harms associated with imposing afﬁrmative content obligations on media, see Martin H. Redish & Kirk J. Kaludis, The Right of Expressive Access in First Amendment Theory: Redistributive Values and the Democratic Dilemma, 93 NW. U. L. REV. 1083, 1114–17 (1999). 103. Id. at 655. 104. E.g., CBS, Inc. v. FCC, 453 U.S. 367, 396 (1981) (upholding a “limited right to ‘reasonable’ access” for declared political candidates in an election); Red Lion Broad. Co. v. FCC, 395 U.S. 367, 98. Levi, supra note 58, at 278–80. 102. 512 U.S. 622 (1994). 103. Id. at 655. 101. Id. at 258. 394 (1969) (upholding the right to equal access for a public ﬁgure who was “personally attacked” over that same broadcast medium); cf. Consol. Edison Co. v. Pub. Serv. Comm’n, 447 U.S. 530, 543 (1980) (using scarcity to hold Red Lion inapplicable to compelled speech with respect to other media); Pac. Gas & Elec. v. Pub. Util. Comm’n, 475 U.S. 1, 10 n.6 (1986) (plurality opinion) (distinguishing the mail from broadcast on scarcity grounds and declaring unconstitutional a policy that prohibited utility companies from including political ﬂyers in monthly billing statements). 105 395 U S 367 (1969) 107. Policies and Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10731–32 ¶¶ 155–156 (1996). p g p 105. 395 U.S. 367 (1969). THE GEORGETOWN LAW JOURNAL The Supreme Court has nonetheless relied on the scarcity doc- trine to uphold the imposition of afﬁrmative programming obligations with respect to broadcasting.104 As the Court held in Red Lion Broadcasting Co. v. 266 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 FCC,105 the inherent physical limitations of the spectrum justiﬁed requiring broadcasters to serve as a proxy for other speakers because “as far as the First Amendment is concerned those who are licensed stand no better than those to whom licenses are refused.”106 The FCC’s most recent children’s television decision also invoked Paciﬁca as a justiﬁcation for upholding the constitutional- ity of the requirement that all broadcasters provide three hours of children’s programming on the ground that the government’s interest in ensuring that children are exposed to educational programming is at least as signiﬁcant as its interest in protecting them from exposure to indecent material.107 Thus, even though each of the key features of the Broadcast Model of regulation represents an archetypical violation of conventional First Amend- ment principles, the Court has nonetheless relied on two key rationales to uphold each feature’s constitutionality: (1) the scarcity doctrine associated with NBC and Red Lion and (2) the unique pervasiveness and accessibility of broadcasting associated with Paciﬁca. As a result, the constitutionality of the Broadcast Model depends entirely on the continuing viability of these rationales as justiﬁcations for having a technology-speciﬁc First Amendment. The next two Parts will consider each of these rationales in turn, focusing on the manner in which analytical, technological, and doctrinal developments have systemati- cally called both of them into question. 106. Id. at 389. 105. 395 U.S. 367 (1969). 106. Id. at 389. II. THE RISE AND (IMPLICIT) DEMISE OF THE SCARCITY DOCTRINE As the foregoing discussion has shown, the scarcity doctrine has represented the principal justiﬁcation for extending a lesser degree of First Amendment protection to broadcasting. This Part will provide my own assessment of the scarcity doctrine. Section A will provide an analytical critique of scarcity by analyzing two theoretical problems with the doctrine. The ﬁrst is the now- standard economic argument that, to the extent that scarcity is meaningful at all, it applies with equal force to all media and thus does not serve to differentiate broadcasting. The second is a novel critique advanced for the ﬁrst time in this Article that focuses on the Court’s growing tendency, when assessing the constitutionality of a particular regulatory provision, to treat all other features of the regulatory regime as ﬁxed. The problem with doing so is that those other features are frequently themselves the product of regulation. As a result, treating 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 267 these other regulatory features as part of the constitutional baseline threatens to allow regulation to serve as the constitutional justiﬁcation for more regulation. Section B will offer a technological critique of the scarcity doctrine. Although commentators have long noted that the development of cable television and other alternative television technologies eliminated the spectrum as a constraint on the number of people who can speak, the Court has largely disregarded the point, opting instead to view broadcasting as a universe unto itself. The impend- ing arrival of a series of new broadcast technologies, including digital transmis- sion, program storage, video-on-demand, spread spectrum, and packet switching, holds the promise of eliminating spectrum as a physical constraint even if broadcasting is viewed in isolation from other media. Once the economic limitations endemic to all markets become more important than the physical limitations of the spectrum, the scarcity doctrine will collapse as a basis for distinguishing broadcasting from other media. In the face of such a withering attack, it is somewhat surprising that the scarcity doctrine has persisted. Section C chronicles its doctrinal history, begin- ning with the move towards its abandonment during the 1980s and ending with the Court’s surprising reafﬁrmation of the doctrine in Metro Broadcasting, Inc. v. FCC.108 Since that decision was handed down, the courts have shown increasing signs of discomfort with the scarcity doctrine. II. THE RISE AND (IMPLICIT) DEMISE OF THE SCARCITY DOCTRINE Not only have the courts consistently refused to extend it to other media, they have even appeared reluctant to continue to apply the doctrine to broadcasting. It appears that the scarcity doctrine may well be in the process of dissipating with a whimper, rather than a bang. 108. 497 U.S. 547, 566–67 (1989). 109. Kalven, supra note 36, at 30. 110. BOLLINGER, supra note 11, at 89. 111. The leading critiques of scarcity include KRATTENMAKER & POWE, supra note 72, at 204–19; POWE, supra note 51, at 200–08; MATTHEW L. SPITZER, SEVEN DIRTY WORDS AND SIX OTHER STORIES 9–18 p 110. BOLLINGER, supra note 11, at 89. 109. Kalven, supra note 36, at 30. 111. The leading critiques of scarcity include KRATTENMAKER & POWE, supra note 72, at 204–19; POWE, supra note 51, at 200–08; MATTHEW L. SPITZER, SEVEN DIRTY WORDS AND SIX OTHER STORIES 9–18 THE GEORGETOWN LAW JOURNAL [I]t is a commonplace of economics that almost all resources used in the economic system (and not simply radio and television frequencies) are limited in amount and scarce, in that people would like to use more than exists. Land, labor, and capital are all scarce, but this of itself, does not call for government regulation.112 It is of no consequence that the Court believed that the amount of spectrum available was absolutely limited. For example, the amount of land and the number of Rembrandts is ﬁxed and ﬁnite; yet, that fact has never been thought to require the government to decide who should use those resources and for what purpose. The typical solution to the problems of allocation is the creation of well-deﬁned property rights in spectrum. Once that is done, the market can allocate the spectrum through a price mechanism without the need for any administrative involvement whatsoever.113 Nor does the potential for interference serve to distinguish broadcasting from other forms of communication. The same problems would arise if more than one person tried to speak in the same place at the same time. Indeed, interfer- ence is a potential problem for all goods. As Ronald Coase has noted, “the use of a piece of land simultaneously for growing wheat and as a parking lot would produce similar results.”114 Such problems, moreover, are easily solved without governmental allocation. The solution is simply to provide owners with a cause of action to enforce their property rights against trespassers.115 The implications of Coase’s critique were so sweeping that contemporary commentators and policymakers could not bring themselves to take it seriously. The reaction of First Amendment scholar and fellow University of Chicago professor Harry Kalven is fairly representative. Calling Coase’s argument “an 115. In the words of Thomas Hazlett: The interference problem is widely recognized as one of deﬁning separate frequency “proper- ties”; it is logically unconnected to the issue of who is to harvest those frequencies. To confuse the deﬁnition of spectrum rights with the assignment of spectrum rights is to believe that, to keep intruders out of (private) backyards, the government must own (or allocate) all the houses. It is a public policy non sequitur . . . . Thomas W. Hazlett, The Rationality of U.S. Regulation of the Broadcast Spectrum, 33 J.L. & ECON. (1986); Coase, supra note 29, at 13–14; Mark S. Fowler & Daniel L. Brenner, A Marketplace Approach to Broadcast Regulation, 60 TEX. L. REV. 207, 221–26 (1982). For a more comprehensive enumeration of the critiques of the scarcity rationale, see J. Gregory Sidak, Telecommunications in Jericho, 81 CAL. L. REV. 1209, 1231 n.63 (1993) (collecting commentary). 112. Coase, supra note 29, at 14; see also KRATTENMAKER & POWE, supra note 72, at 204 (“‘Scarce resource’ is a redundant phrase. Every resource is scarce, be it oil, gas, clean water, trees, or iron ore. A ‘nonscarce resource’ is a contradiction in terms.”). g y 112. Coase, supra note 29, at 14; see also KRATTENMAKER & POWE, supra note 72, at 204 (“‘Scarce resource’ is a redundant phrase. Every resource is scarce, be it oil, gas, clean water, trees, or iron ore. A ‘nonscarce resource’ is a contradiction in terms.”). 1. The Analytical Emptiness of Scarcity As noted above, the FCC and the Supreme Court have long relied on the scarcity doctrine to justify according less First Amendment protection to broad- casting than to other media. The Supreme Court believed that broadcasting was unique in that the number of available channels was strictly limited. Because of this limitation, it was essential for the government to become directly involved in allocating opportunities to speak. Commentators have long recognized, however, that “there is a devastating— even embarrassing—deﬁciency in this analysis,”110 in that the limited nature of the spectrum as a resource does not serve to distinguish broadcasting from any other medium of communication.111 As Ronald Coase observed: 268 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL (1986); Coase, supra note 29, at 13–14; Mark S. Fowler & Daniel L. Brenner, A Marketplace Approach to Broadcast Regulation, 60 TEX. L. REV. 207, 221–26 (1982). For a more comprehensive enumeration of the critiques of the scarcity rationale, see J. Gregory Sidak, Telecommunications in Jericho, 81 CAL. L REV 1209 1231 n 63 (1993) (collecting commentary) (1986); Coase, supra note 29, at 13–14; Mark S. Fowler & Daniel L. Brenner, A Marketplace Approach to Broadcast Regulation, 60 TEX. L. REV. 207, 221–26 (1982). For a more comprehensive enumeration of the critiques of the scarcity rationale, see J. Gregory Sidak, Telecommunications in Jericho, 81 CAL. L. REV. 1209, 1231 n.63 (1993) (collecting commentary). 112. Coase, supra note 29, at 14; see also KRATTENMAKER & POWE, supra note 72, at 204 (“‘Scarce resource’ is a redundant phrase. Every resource is scarce, be it oil, gas, clean water, trees, or iron ore. A ‘nonscarce resource’ is a contradiction in terms.”). 113. Coase, supra note 29, at 14. 114. Id. 115. In the words of Thomas Hazlett: The interference problem is widely recognized as one of deﬁning separate frequency “proper- ties”; it is logically unconnected to the issue of who is to harvest those frequencies. To confuse the deﬁnition of spectrum rights with the assignment of spectrum rights is to believe that, to keep intruders out of (private) backyards, the government must own (or allocate) all the houses. It is a public policy non sequitur . . . . Thomas W. Hazlett, The Rationality of U.S. Regulation of the Broadcast Spectrum, 33 J.L. & ECON. 133, 138 (1990); see also KRATTENMAKER & POWE, supra note 72, at 207 (“To prevent chaos (interference) in broadcasting or publishing, . . . requires not a commission, but a system of property rights. It follows . . . that if one decides to create a commission, it need do no more than deﬁne and protect property rights (that is, allocate spectrum among certain users and deﬁne and punish interfer- ence).”). 113. Coase, supra note 29, at 14. 120. See, e.g., BOLLINGER, supra note 11, at 87–90; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 92, 110–12; Ronald J. Krotoszynski, Jr., Into the Woods: Broadcasters, Bureaucrats, and Children’s Television Programming, 45 DUKE L.J. 1193, 1247 (1996); Logan, supra note 26, at 1701–05; Weinberg, supra note 26, at 1106; Note, The Message in the Medium: The First Amendment on the Information Superhighway, 107 HARV. L. REV. 1062, 1070–77 (1994). The lone exceptions appear to be former FCC Chairmen Newt Minow and Reed Hundt. See MINOW & LAMAY, supra note 52, at 131 (1995); Reed E. Hundt, A New Paradigm for Broadcast Regulation, 15 J.L. & COMMERCE 527, 542–43 (1996). 118. See R.H. Coase, Comment on Thomas W. Hazlett: Assigning Property Rights to Radio Spectrum Users: Why Did FCC License Auctions Take 67 Years?, 41 J.L. & ECON. 577, 579 (1998); R.H. Coase, The Problem of Social Cost, 3 J.L. & ECON. 1, 15–17 (1960) [hereinafter Coase, Social Cost]; Thomas W. Hazlett, The Wireless Craze, The Unlimited Bandwidth Myth, The Spectrum Auction Faux Pas, and the Punchline to Ronald Coase’s “Big Joke”: An Essay on Airwave Allocation Policy, 14 HARV. J.L. & TECH. 335, 343 (2001). 119. This group included such future Chicago School mainstays as Aaron Director, Milton Fried- man, John McGee, and George Stigler. See The Fire of Truth: A Remembrance of Law and Economics at Chicago, 1932–1970, 26 J.L. & ECON. 163, 220–21 (Edmund W. Kitch ed., 1983). THE GEORGETOWN LAW JOURNAL 133, 138 (1990); see also KRATTENMAKER & POWE, supra note 72, at 207 (“To prevent chaos (interference) in broadcasting or publishing, . . . requires not a commission, but a system of property rights. It follows . . . that if one decides to create a commission, it need do no more than deﬁne and protect property rights (that is, allocate spectrum among certain users and deﬁne and punish interfer- ence).”). 2003] 269 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT insight more fundamental than we can use,” Kalven declared it to be “so radical by today’s views that although I am persuaded of its correctness, I am not clear how it can be used in public discussion.”116 As a result, Kalven rejected it as a theoretically interesting nonstarter. Rather than take it seriously, Kalven thought it would be more constructive to explore policy options within a context that took the existing broadcast regime as given.117 Kalven was not alone in this regard. Academics and policymakers alike initially greeted Coase’s argument with ridicule.118 Even the group of University of Chicago economists who would become some of Coase’s biggest champions unanimously rejected his argument at ﬁrst blush.119 Subsequent scholars have not been so reticent, and Coase’s critique of the scarcity doctrine has now become the conventional wisdom. In fact, it is now so broadly accepted that even those scholars who are sympathetic to the Broadcast Model tend to abjure any reliance on the scarcity doctrine.120 117. Id. at 32 (“The key task is to explore what policy can be worked out for the independence of broadcasting if we continue to license commercial broadcasting and do not auction the licenses.”) (emphasis in original). 116. Kalven, supra note 36, at 30. 116. Kalven, supra note 36, at 30. 117. Id. at 32 (“The key task is to explore what policy can be worked out for the independence of broadcasting if we continue to license commercial broadcasting and do not auction the licenses.”) (emphasis in original). 118. See R.H. Coase, Comment on Thomas W. Hazlett: Assigning Property Rights to Radio Spectrum Users: Why Did FCC License Auctions Take 67 Years?, 41 J.L. & ECON. 577, 579 (1998); R.H. Coase, The Problem of Social Cost, 3 J.L. & ECON. 1, 15–17 (1960) [hereinafter Coase, Social Cost]; Thomas W. Hazlett, The Wireless Craze, The Unlimited Bandwidth Myth, The Spectrum Auction Faux Pas, and the Punchline to Ronald Coase’s “Big Joke”: An Essay on Airwave Allocation Policy, 14 HARV. J.L. & TECH. 335, 343 (2001). 119. This group included such future Chicago School mainstays as Aaron Director, Milton Fried- man, John McGee, and George Stigler. See The Fire of Truth: A Remembrance of Law and Economics at Chicago, 1932–1970, 26 J.L. & ECON. 163, 220–21 (Edmund W. Kitch ed., 1983). 120. See, e.g., BOLLINGER, supra note 11, at 87–90; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 92, 110–12; Ronald J. Krotoszynski, Jr., Into the Woods: Broadcasters, Bureaucrats, and Children’s Television Programming, 45 DUKE L.J. 1193, 1247 (1996); Logan, supra note 26, at 1701–05; Weinberg, supra note 26, at 1106; Note, The Message in the Medium: The First Amendment on the Information Superhighway, 107 HARV. L. REV. 1062, 1070–77 (1994). The lone exceptions appear to be former FCC Chairmen Newt Minow and Reed Hundt. See MINOW & LAMAY, supra note 52, at 131 (1995); Reed E. Hundt, A New Paradigm for Broadcast Regulation, 15 J.L. & COMMERCE 527, 542–43 (1996). 127. Id. at 761; see also id. at 766 (arguing that the provision in question “would not signiﬁcantly restore editorial rights of cable operators”). y 126. 518 U.S. at 761. 125. “Public access channels” are channels that municipalities require cable operators to set aside for use by the local community. See id. at 734. Perhaps the most celebrated portrayal of public access programming is the recurring sketch on Saturday Night Live entitled Wayne’s World. “Leased access channels” are a federal regulatory response to the growth of vertical integration in the cable industry. The requirement grew out of the concern that if cable operators and large cable networks were under the same corporate umbrella, those cable operators would have an incentive to exclude unafﬁliated programmers in ways that would harm competition. For example, it is said that Time Warner was able to forestall NBC’s ﬁrst attempt to set up a news network in direct competition with CNN simply by having all of the cable operators under its control refuse to carry the new network. As a result, Congress enacted legislation requiring all cable systems to offer part of their channel capacity for commercial lease by unafﬁliated networks. See Yoo, supra note 1, at 223–24. 124. 518 U.S. 727 (1996). 121. For a related argument, see Stuart Minor Benjamin, The Logic of Scarcity: Idle Spectrum as a First Amendment Violation, 52 DUKE L.J. 1, 38–45 (2002). 122 521 S 844 (199 ) 123. Id. at 868–69. 122. 521 U.S. 844 (1997). 121. For a related argument, see Stuart Minor Benjamin, The Logic of Scarcity: Idle Spectrum as a First Amendment Violation, 52 DUKE L.J. 1, 38–45 (2002). 122. 521 U.S. 844 (1997). 123. Id. at 868–69. 124. 518 U.S. 727 (1996). 125. “Public access channels” are channels that municipalities require cable operators to set aside for use by the local community. See id. at 734. Perhaps the most celebrated portrayal of public access programming is the recurring sketch on Saturday Night Live entitled Wayne’s World. “Leased access channels” are a federal regulatory response to the growth of vertical integration in the cable industry. The requirement grew out of the concern that if cable operators and large cable networks were under the same corporate umbrella, those cable operators would have an incentive to exclude unafﬁliated programmers in ways that would harm competition. For example, it is said that Time Warner was able to forestall NBC’s ﬁrst attempt to set up a news network in direct competition with CNN simply by having all of the cable operators under its control refuse to carry the new network. As a result, Congress enacted legislation requiring all cable systems to offer part of their channel capacity for commercial lease by unafﬁliated networks. See Yoo, supra note 1, at 223–24. 2. Existing Regulations as a Constitutional Baseline There is another fundamental conceptual problem associated with using scarcity to justify holding broadcasting to a lower standard of First Amendment scrutiny. In concluding that the electromagnetic spectrum was scarce, the Court took as given the background factors that determined the total amount of spectrum allocated to broadcasting. The problem with this analysis is that the amount of spectrum allocated to broadcasting was itself purely a product of regulation. Without recognizing that it was doing so, the Court in effect allowed these other regulatory provisions to become part of the constitutional baseline used to determine whether a particular regulation violated the First Amend- 270 [Vol. 91:245 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL ment.121 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 271 leased access channels as part of the constitutional baseline because those channels had not previously been subject to intrusive regulation. Conversely, the past imposition of successful regulation on cable operators became a justiﬁcation for changing the relevant constitutional baseline in ways that tended to permit additional regulation in the future. In addition, the plurality also emphasized that cities had generally required that public access channels be supervised by an “access channel manager,” an entity that is often a governmental actor and typically overseen by a local supervisory board. In the plurality’s view, the presence of an access channel manager indicated that less First Amendment harm would result from striking down a statute giving cable operators greater editorial control over those channels.128 This reasoning treated the existence of access channel managers as part of the relevant baseline for evaluating the extent to which the provision in question intrudes on the First Amendment. The plurality’s reasoning is tanta- mount to saying that the presence of a governmental censor obviates the need for private discretion over public access channels. This last observation underscores the manner in which the plurality’s analysis begs an important question. Before it could properly rely on the legacy of public access regulation as an appropriate constitutional baseline, it should have considered whether the various elements of that legacy were themselves consti- tutional. In other words, before the plurality could rely on the presence of an access channel manager as support for its constitutional position, it should have evaluated the constitutionality of the use of access channel managers. Courts, however, are understandably loath to treat a constitutional challenge to one particular provision as an open invitation to consider the constitutionality of other aspects of a regulatory scheme. It is far more common for them to assume the propriety of the other elements not being challenged without formally resolving the issue.129 Although the Court’s desire to circumscribe the number of statutory provisions under review is understandable, this approach raises the serious danger of allowing regulation to become self-reinforcing. Simply put, it permits regulation, if imposed for a long enough time, to become part of the relevant constitutional baseline that in turn justiﬁes other forms of regulation. Such reasoning is valid if and only if the other features of the regulatory regime that help form the constitutional baseline are themselves constitutional. The Denver plurality, however, failed to address this question. 129. See id. at 821 n.6 (Thomas, J., concurring in the judgment in part and dissenting in part) (noting that the constitutionality of leased access and public access were not at issue in that case). 128. Id. at 761–62. ment.121 ment.121 The Supreme Court followed this analytical approach in two recent decisions involving emerging communications media. For example, in Reno v. ACLU,122 the Court based its refusal to allow First Amendment challenges to regulation of the Internet on the lower level of scrutiny applied to broadcasting and on the ground that the Internet had never “been subject to the type of government supervision and regulation that has attended the broadcast industry.”123 The implication of this reasoning is clear: Had the Internet, like broadcasting, been subject to longstanding regulation, that fact alone would have been a consider- ation supporting the constitutionality of additional regulation. The existence of other regulations played an even more speciﬁc role in the plurality opinion in Denver Area Educational Telecommunications Consortium, Inc. v. FCC.124 In that case, the Court struck down a statutory provision allowing local cable operators to refuse to carry indecent programming on public access channels while upholding a parallel provision authorizing cable operators to refuse to carry indecent programming on their leased access channels.125 The plurality based its distinction between the two provisions in part on the different regulatory legacies surrounding public and leased access. That leased access channels had not historically been subject to signiﬁcant municipal regulation justiﬁed regarding editorial control over those channels as part of the cable operators’ First Amendment prerogatives. Granting them the right to refuse to carry indecent programming on those channels thus repre- sented a restoration of their constitutional rights.126 In contrast, public access channels had historically been subject to much more intrusive regulation. Because providing cable operators with greater control over their public access channels “d[id] not restore to cable operators editorial rights that they once had, . . . the countervailing First Amendment interest is nonexistent, or at least much diminished.”127 Thus, the plurality’s reasoning justiﬁed editorial control over 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 130. The discussion that follows is based in part on several excellent histories of the FCC’s spectrum allocation decisions. See HOUSE COMM. ON INTERSTATE AND FOREIGN COMMERCE, NETWORK BROADCASTING, H.R. REP. NO. 85-1297, at 17–37 (1958) [hereinafter BARROW REPORT]; ALLOCATION OF TV CHANNELS, REPORT OF THE AD HOC ADVISORY COMM. ON ALLOCATIONS TO THE HOUSE COMM. ON INTERSTATE AND FOREIGN COMMERCE, 85TH CONG. 17–32 (1958) [hereinafter BOWLES REPORT]; ERWIN G. KRASNOW ET AL., THE POLITICS OF BROADCAST REGULATION 176–91 (3d ed. 1982); Henry Geller, A Modest Proposal for Modest Reform of the Federal Communications Commission, 63 GEO. L.J. 705, 707–09 (1975); Hazlett, supra note 115, at 143–63; Thomas L. Schuessler, Structural Barriers to the Entry of Additional Television Networks: The Federal Communications Commission’s Spectrum Management Policies, 54 S. CAL. L. REV. 875, 886–981 (1981); Note, The Darkened Channels: UHF Television and the FCC, 75 HARV. L. REV. 1578, 1578–93 (1962). g 132. 1 FRC ANNUAL REPORT 13 (1927); see also POOL, supra note 51, at 115, 141 (“Use of new, shorter wave bands would not only have obsoleted existing radio sets but would also have required more expensive multiband sets and transmitters . . . . The market notion that, when a resource is scarce, one provides more of it in higher priced ways and thereby restricts the demand, did not ﬁt within a populist notion of cheap broadcasting.”). p p p g ) 133. Hazlett, supra note 115, at 155–56. 131. Hazlett, supra note 115, at 155 (citing HARRY J. LEVIN, THE INVISIBLE RESOURCE 20–21 (1971)). ( p ) 142. See BOWLES REPORT, supra note 130, at 76 (noting testimony of FCC Commissioner Hyde that the refusal to reallocate television stations to the UHF band stemmed from “the fact that these stations were constructed, the investments made, [and] the public accustomed to listening to them”); Schuessler, supra note 130, at 909–10. 137. See Allocation of Frequencies to Various Classes of Non-Gov’tal Servs. in Radio Spectrum from 30 Kilocycles to 30,000,000 Kilocycles, 39 F.C.C. 68, 129–30 (1945) [hereinafter Allocation of Frequencies]; Public Release, 39 F.C.C. 16, 16 (1939); BARROW REPORT, supra note 130, at 18–19, 21; Geller, supra note 130, at 707–09; Schuessler, supra note 130, at 888. 141. See Second Report on Deintermixture, 13 Rad. Reg. (P & F) 1571, 1577–78 ¶ 18 (1956); Television Broad. Serv., Third Notice of Further Proposed Rule Making, 16 Fed. Reg. 3072, 3074 (F.C.C. Apr. 7, 1951). 134. POOL, supra note 51, at 141, 152–53; Fowler & Brenner, supra note 111, at 222 n.70; Sidak, supra note 111, at 1230. 136. Frequency Allocation to Servs. in Frequency Bands from 30,000 kc to and Including 300,000 kc, Commission Order No. 19, 4 F.C.C. 30 (1939); Frequency Allocation to Servs. in Frequency Bands from 30,000 kc to and Including 300,000 kc, Report of the Commission, 4 F.C.C. 582 (1939). The FCC reduced the allocation to eighteen channels in 1941. Broad. Servs. Other than Standard Broad., 6 Fed. Reg. 2282, 2283 (May 6, 1941). 140. Amendments to Comm’n’s Rules & Regulations Governing Sharing of Television Channels & Assignment of Frequencies to Television & Non-Govt. Fixed & Mobile Servs., 39 F.C.C. 336 (1948). See generally BARROW REPORT, supra note 130, at 18–19, 21; Schuessler, supra note 130, at 890. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT This mode of analysis provides another reason to question the analytical coherence of the scarcity doctrine. In arguing that the number of channels is strictly limited, the scarcity doctrine accepts as a relevant constitutional baseline those decisions that determined how many channels are available for broadcast- ing in the ﬁrst instance. In so doing, the scarcity doctrine elides the fact that the 272 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 amount of spectrum allocated to broadcasting is itself the product of a series of regulatory decisions concerning: (1) the total amount of spectrum allocated to television broadcasting, (2) the manner in which stations were allocated to particular communities, and (3) the manner in which those stations are allocated to particular licensees. An examination of each of these regulatory decisions reveals that the FCC resolved these questions in a manner that limited the supply of and heightened the demand for broadcast channels.130 Scarcity is thus revealed to be a direct product of FCC regulation. As a result, to rely on scarcity to uphold the constitutionality of the Broadcast Model is to permit the overrid- ing culture of regulation to become its own constitutional justiﬁcation. a. The Amount of Spectrum Allocated to Broadcasting. The manner in which the Federal Radio Commission (FRC) initially determined the amount of spec- trum that would be available for broadcasting provides one of the most telling examples of how the scarcity of channels is a direct result of regulatory decisionmaking. In its ﬁrst signiﬁcant action concerning broadcasting, the FRC refused to follow the example set by European nations, which assigned more spectrum to broadcasting despite having fewer radio stations to accommodate, and rejected the recommendation of an international conference that had desig- nated additional bands of the spectrum for radio broadcasting.131 The FRC’s stated reason for doing so was to avoid rendering existing radio sets obsolete.132 Reliance on this concern was, to put it mildly, remarkably shortsighted. The one-time costs associated with changing receiving equipment at such a nascent stage in the industry’s development were slight in comparison to the long-term beneﬁts that would ﬂow from having a greater range of programming options.133 Furthermore, preserving existing radio sets only served to heighten the problems of scarcity. 135. Steven Phipps, “Order Out of Chaos”: A Reexamination of the Historical Basis for the Scarcity of Channels Concept, 45 J. BROAD. & ELEC. MEDIA 57, 67 (2001). q , p , 139. Id. at 129; Schuessler, supra note 130, at 886 n.41, 887. 138. Allocation of Frequencies, supra note 137, at 130. 138. Allocation of Frequencies, supra note 137, at 130. 139 Id at 129; Schuessler supra note 130 at 886 n 41 887 9. Id. at 129; Schuessler, supra note 130, at 886 n.41, 8 145. Sidak, supra note 111, at 1233. Most notably, although such spectrum-based services as multichannel multipoint distribution services (MMDS) and instructional television ﬁxed services (ITFS) have proven to be abject failures, the FCC has refused to reallocate that spectrum for other uses. See Amendment of Part 2 of Comm’n’s Rules to Allocate Spectrum Below 3 GHz for Mobile and Fixed Servs., First Report and Order, 16 F.C.C.R. 17222, 17333–38 ¶¶ 19–30 (2001) (refusing to reallocate spectrum devoted to ITFS and MMDS and limiting newly authorized mobile use to incumbent licensees). In addition, large numbers of assignments for noncommercial, educational television stations also remain unused. See Deletion of Noncommercial Reservation of Channel 16, 482–88 MHz, Pittsburgh, Pa., 11 F.C.C.R. 11700, 11708 ¶ 18 (1996) (mem. op. & order) (noting that the FCC had never eliminated a noncommercial allotment even when vacant for a long period or proven not to be economically viable). 143. These disadvantages included 20 to 25% more electric power consumption by UHF stations than VHF stations, UHF stations are harder to tune in, and UHF stations are subject to greater interference from terrain and buildings. In addition, at the time most receivers could not receive UHF signals. See 1 FEDERAL COMMUNICATIONS COMMISSION NETWORK INQUIRY SPECIAL STAFF, NEW TELEVISION NETWORKS 69–76 (1980) [hereinafter NEW TELEVISION NETWORKS]; Note, The Darkened Channels, supra note 130, at 1580. 147. As Yochai Benkler has pointed out, the FCC replicated these mistakes when rolling out digital television. Despite the drastically different characteristics of digital broadcasting, the FCC simply assigned the same amount of spectrum to digital stations that it had previously assigned to analog stations. See Yochai Benkler, Siren Songs and Amish Children: Autonomy, Information, and Law, 76 N.Y.U. L. REV. 23, 98–100 (2001). y g p p g 149. See 47 C.F.R. §§ 2.106, 73.127, 73.293, 73.665 (2000) (authorizing broadcasters to transmit subsidiary communications services); id. §§ 73.295(a), 73.667(a) (deﬁning subsidiary communications services). 144. See Geller, supra note 130, at 708–09; Note, The Darkened Channels, supra note 130, at 1580–93. ( ) 148. See Amendment of Comm’n’s Rules Concerning Use of Subsidiary Communications Authoriza- tions, 48 Fed. Reg. 28,445, 28,447 ¶ 15 (June 22, 1983); Amendment of Comm’n’s Rules Concerning Use of Subsidiary CommunicationsAuthorizations, 47 Fed. Reg. 36,235, 36,238 ¶ 14 (proposedAug. 19, 1982). 146. See infra notes 173–74 and accompanying text. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Improved receiver technology can serve as a substitute input for spectrum because better tuners allow stations to be spaced closer together and to operate at lower power without causing any reduction in 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 273 quality.134 The decision to safeguard existing receiver technology exacerbated scarcity by keeping in place a large number of low quality, often homemade, receivers that were difﬁcult to tune accurately and were not properly shielded against interference.135 Federal regulators repeated these mistakes when apportioning spectrum for television. When the FCC ﬁrst allocated spectrum for television broadcasting in 1937, it set aside sufﬁcient spectrum in the VHF band for 19 television channels.136 The FCC soon recognized that this amount of spectrum was insufﬁcient to support a competitive national television service. As a result, it concluded that the best long-range solution was for television to reside entirely in the UHF band.137 Because technical obstacles to UHF broadcasting re- mained, the FCC decided in 1945 to permit the deployment of television in the VHF band on an interim basis, while exhorting the industry to act quickly to solve the problems associated with UHF.138 Because television’s tenure in VHF was intended to be temporary, the FCC allowed the number of VHF channels allocated to be reduced from nineteen to thirteen139 and eventually to twelve.140 When the moment arrived for VHF broadcasters to move into the UHF spectrum, however, the FCC ﬂinched. Even though the agency continued to acknowledge that television would best be served if it were shifted entirely into the UHF band,141 once again the prospect of forcing incumbent broadcasters and viewers to abandon their investments in existing equipment prevented the FCC from making spectrum more broadly available.142 Instead, the FCC opted to overlay UHF assignments on top of the existing VHF assignments without 274 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 150. Promoting Efﬁcient Use of Spectrum Through Elimination of Barriers to Dev. of Secondary Markets, Comments of 37 Concerned Economists, WT Docket No. 00-230 (F.C.C. 2001), available at http://www.aei.org/ct/cthazlett010207.pdf. THE GEORGETOWN LAW JOURNAL forcing any station to change frequencies. The problem is that UHF stations confront several operating disadvantages when compared with VHF stations.143 The decision to intermix UHF and VHF stations inevitably caused UHF to fail as a service.144 The scarcity created by these initial allocation decisions was perpetuated and aggravated by the FCC’s disinclination to reallocate spectrum to broadcasting from other uses, even when those uses had proven unviable145 and despite dramatic increases in total amount of usable spectrum.146 Other regulatory decisions compounded scarcity still further. For example, one might have expected technological innovations to reduce the amount of spectrum required by each channel until eventually new channels could be introduced. FCC policy eliminated any incentive for broadcasters to search for such savings by forcing television and FM broadcasters to adhere to a ﬁxed table of allocations and by greatly restricting the way in which they could use any spectrum that was conserved.147 The FCC eventually realized that its regulatory scheme did not provide any incentive for broadcasters to use spectrum efﬁciently.148 Although it eventually liberalized its rules, it did so in a way that channeled any additional spectrum towards subsidiary communications services, such as paging and data transmission, rather than towards additional broadcasting options.149 2003] 275 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT It is not yet clear how much the FCC’s position on spectrum ﬂexibility is likely to change in the near future. A group of distinguished economists, including several Nobel laureates, recently ﬁled a statement with the FCC asking it to permit licensees to reallocate spectrum to different uses.150 A recent report authored by the FCC’s Spectrum Policy Task Force has proposed adding some ﬂexibility to the FCC’s current command-and-control approach to spec- trum regulation.151 Thus far, policymakers have yet to allow such spectrum markets to emerge. Scarcity is thus partly the result of the manner in which the FCC has determined how much of the spectrum may be used for broadcasting and its general reluctance to revisit that determination. As will be discussed in the succeeding subsections, other regulatory decisions regarding the number of stations that should be assigned to each community and how those stations should be assigned to particular individuals had an equally strong inﬂuence on restricting the supply of and increasing the demand for spectrum. b. The Allocation of Stations to Particular Communities. In addition to setting the total amount of spectrum dedicated to broadcasting, the FCC also had to develop some means for assigning particular channels to particular communities around the country. The FCC followed allocation principles that attempted to assign at least two television stations to as many communities as possible.152 Although the FCC did so in an attempt to disperse control of television broadcasting as broadly as possible, the ﬁnal allocation plan actually exacer- bated the problems of scarcity signiﬁcantly by making it essentially inevitable that television would be dominated by three large networks. b. The Allocation of Stations to Particular Communities. In addition to setting the total amount of spectrum dedicated to broadcasting, the FCC also had to develop some means for assigning particular channels to particular communities around the country. The FCC followed allocation principles that attempted to assign at least two television stations to as many communities as possible.152 Al h h h FCC did i di l f l i i Although the FCC did so in an attempt to disperse control of television broadcasting as broadly as possible, the ﬁnal allocation plan actually exacer- bated the problems of scarcity signiﬁcantly by making it essentially inevitable that television would be dominated by three large networks. Understanding why this is the case requires an appreciation of one of the 151. Amendment of Section 3.606 of Comm’n’s Rules & Regulations, Sixth Report and Order, 41 F.C.C. 148, 167 ¶ 63 (1952). (1) to provide at least one television service to all parts of the United States; 151. FCC SPECTRUM POLICY TASK FORCE, REPORT 5–6 (Nov. 2002), available at http://hraunfoss.fcc.gov/ edocs_public/attachmatch/doc-228542A1.pdf. (recommending expansion of exclusive use and com- mons models with reservation of command-and-control regulation to situations involving public interest objectives and treaty obligations). j y g 152. Speciﬁcally, the FCC assigned television stations according to the followin THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT FCC SPECTRUM POLICY TASK FORCE, REPORT 5–6 (Nov. 2002), available at http://hraunfoss.fcc.gov/ edocs_public/attachmatch/doc-228542A1.pdf. (recommending expansion of exclusive use and com- mons models with reservation of command-and-control regulation to situations involving public interest objectives and treaty obligations). j y g ) 152. Speciﬁcally, the FCC assigned television stations according to the following priorities: (1) to provide at least one television service to all parts of the United States; (2) to provide each community with at least one television broadcast station; (3) to provide a choice of at least two television services to all parts of the United States; (4) to provide each community with at least two television broadcast stations; and (5) to assign any channels which remain unassigned under the foregoing priorities to the various communities depending on the size of the population of each community, the geographical location of such community, and the number of television services available to such community from television stations located in other communities. Amendment of Section 3.606 of Comm’n’s Rules & Regulations, Sixth Report and Order, 41 F.C.C. 148, 167 ¶ 63 (1952). THE GEORGETOWN LAW JOURNAL [Vol. 91:245 276 basic economic qualities of television programming.153 Perhaps the most singu- lar economic feature of the broadcast industry is its cost structure, in that the creation of television programming requires the incurrence of large, upfront, ﬁrst-copy costs, whereas the costs of reproducing and distributing additional copies are relatively minor. This cost structure causes average cost to decline over all relevant volumes, as the large upfront investment is amortized over an increasingly large number of viewers. When faced with such a declining cost structure, efﬁciency increases with every additional viewer reached. Networks also possess certain advantages in dealing with advertisers. Using a centralized purchasing agent makes it easier for the advertiser to reach its goals and provides the program provider greater ﬂexibility in compensating advertisers if a particular program does not do as well as expected. In addition, increasing use of satellites to distribute programming puts a premium on the ability to distrib- ute the same program to a broad geographic area.154 These considerations give programming that is able to reach a national audience a decisive economic advantage over programming that is only distrib- uted locally. The problem is that by their very nature, individual stations can only reach limited geographic areas. 153. The discussion that follows is based in part on Yoo, supra note 1, at 213–17, 232–37; and Christopher S. Yoo, Rethinking the Commitment to Free, Local Television: A Public Goods Analysis 27–35 (Sept. 28, 2002) (unpublished manuscript, available at http://ssrn.com/abstract_id333702) (last visited Feb. 10, 2003). 154. BRUCE M. OWEN & STEVEN S. WILDMAN, VIDEO ECONOMICS 53–54 (1992). 155 NEW TELEVISION NETWORKS supra note 143 at 68 153. The discussion that follows is based in part on Yoo, supra note 1, at 213–17, 232–3 Christopher S. Yoo, Rethinking the Commitment to Free, Local Television: A Public Goods An 27–35 (Sept. 28, 2002) (unpublished manuscript, available at http://ssrn.com/abstract_id333702 visited Feb. 10, 2003). 154. BRUCE M. OWEN & STEVEN S. WILDMAN, VIDEO ECONOMICS 53–54 (1992). 155. NEW TELEVISION NETWORKS, supra note 143, at 68. 156. See supra note 143 and accompanying text. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT As a result, the only practical way for them to gain the beneﬁts associated with national distribution is to afﬁliate with a network with a national reach. Two conclusions follow: First, local stations have a natural tendency to afﬁliate with networks whenever possible. Second, the number of independent voices is determined by the number of available networks rather than the geographic dispersion of television stations. Because diversity in communication depends on the number of networks, the problem with the broad dispersion of television stations becomes clear. The FCC’s allocation plan dictated that a substantial part of the country would be served by only three commercial television signals. A study conducted by the FCC in 1980 revealed that although ninety-two percent of U.S. households could receive at least three commercial television signals, only sixty-four percent of U.S. households could receive a fourth television channel.155 This meant that a fourth network would necessarily operate at an extreme disadvan- tage in terms of national coverage. The problems confronting a fourth network were exacerbated further because even when a fourth commercial station was available, it was not infrequently a UHF station, which, as noted earlier, would face substantial technical disadvantages relative to VHF stations.156 As a result, a fourth network would only be able to reach thirty-four percent of the country with a signal that was comparable in quality to those provided by the other three 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 277 networks.157 Thus, despite evidence that sufﬁcient demand existed to support as many as six networks,158 the FCC’s system for dispersing television stations geographically choked off the supply side of the equation by precluding a fourth network from emerging.159 The FCC’s decision was particularly regrettable because at the time it made its principal allocation decisions, it had before it proposals that would have substantially alleviated this problem. For example, an emerging fourth televi- sion network founded by the DuMont Corporation proposed that, instead of attempting to assign two stations to every possible community, the FCC should focus on assigning four VHF stations to as many major markets as possible. 157. NEW TELEVISION NETWORKS, supra note 143, at 78, 81 tbl.15. Expansion of the analysis to include stations that were authorized but not operational does not materially affect the analysis. Id. at 78, 81 tbl.14 (noting that even when nonoperational stations are included, a fourth network would reach only 36% of the nation with comparable signals). 165. KRASNOW ET AL., supra note 130, at 176–82; Schuessler, supra note 130, at 941–67; Note, The Darkened Channels, supra note 130, at 1583–93. 164. Geller, supra note 130, at 708; Note, The Darkened Channels, supra note 130, at 1593. 165. KRASNOW ET AL., supra note 130, at 176–82; Schuessler, supra note 130, at 941–67; Note, The Darkened Channels, supra note 130, at 1583–93. 164. Geller, supra note 130, at 708; Note, The Darkened Channels, supra note 130, at 1593. 158. See 35 FCC ANNUAL REPORT 135, 137 (1969); NOLL ET AL., supra note 46, at 116–20. 159. Rolla Edward Park, New Television Networks, 6 BELL J. ECON. 607, 607–08, 614–16 (1975). y p g ) 158. See 35 FCC ANNUAL REPORT 135, 137 (1969); NOLL ET AL., supra note 46, at 116–20. 159 Rolla Edward Park New Television Networks 6 BELL J ECON 607 607 08 614 16 (19 163. Id. at 908 n.180, 926 n.273. 157. NEW TELEVISION NETWORKS, supra note 143, at 78, 81 tbl.15. Expansion of the analysis to include stations that were authorized but not operational does not materially affect the analysis. Id. at 78, 81 tbl.14 (noting that even when nonoperational stations are included, a fourth network would reach only 36% of the nation with comparable signals). 158. See 35 FCC ANNUAL REPORT 135, 137 (1969); NOLL ET AL., supra note 46, at 116–20. 159. Rolla Edward Park, New Television Networks, 6 BELL J. ECON. 607, 607–08, 614–16 (1975). 160. See Schuessler, supra note 130, at 891, 921–26, 929 tbl.10, 938–39 & tbl.16. 161. Id. at 906–13; Note, The Darkened Channels, supra note 130, at 1579–80. 162. Schuessler, supra note 130, at 909–10. 163. Id. at 908 n.180, 926 n.273. 164. Geller, supra note 130, at 708; Note, The Darkened Channels, supra note 130, at 1593. 165. KRASNOW ET AL., supra note 130, at 176–82; Schuessler, supra note 130, at 941–67; Note, The Darkened Channels, supra note 130, at 1583–93. 160. See Schuessler, supra note 130, at 891, 921 26, 929 tbl.10, 938 39 & tbl.16. 161. Id. at 906–13; Note, The Darkened Channels, supra note 130, at 1579–80. 158. See 35 FCC ANNUAL REPORT 135, 137 (1969); NOLL ET AL., supra note 46, at 116 20. 159. Rolla Edward Park, New Television Networks, 6 BELL J. ECON. 607, 607–08, 614–16 (197 160. See Schuessler, supra note 130, at 891, 921–26, 929 tbl.10, 938–39 & tbl.16. 161. Id. at 906–13; Note, The Darkened Channels, supra note 130, at 1579–80. p g ) See 35 FCC ANNUAL REPORT 135, 137 (1969); NOLL ET AL., supra note 46, at 116–20. 162. Schuessler, supra note 130, at 909–10. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT The DuMont plans would have increased the population receiving four or more channels to nearly ninety-ﬁve percent and increased the percentage of the population receiving technically comparable signals to ninety-three percent— levels that would have greatly increased the feasibility of a fourth network.160 Another proposal, backed by CBS as well as DuMont, relied on a policy known as “deintermixture” to alleviate the problems of scarcity. Deintermixture would have required every city to be completely devoted to either VHF or UHF stations. Doing so would have mitigated UHF’s disadvantage vis-a`-vis VHF by obviating the need for UHF stations to compete directly with VHF stations.161 The FCC unfortunately rejected both of these proposals. It did so in part because adopting either one would have required incumbent broadcasters to shift to UHF and in so doing abandon their investments in their existing VHF facilities.162 History has proven the FCC incorrect. The DuMont network folded three years later163 and UHF deployment was largely a failure.164 Although the FCC continued to recognize that deintermixture remained the best solution to this problem, it pursued the policy without much ardor over the next decade, until it eventually abandoned it in 1962.165 In the end, UHF television did not become viable until the emergence of cable television in the late 1970s. And even then, the FCC’s initial response was to adopt policies that had the perverse effect of perpetuating scarcity. The FCC’s initial reaction to cable television was to attempt to retard its development out HF stations’ viability. AW JOURNAL THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 278 of fear that cable would eliminate the UHF stations’ viability.166 Ironically, cable would ultimately prove to be UHF’s savior rather than its scourge because cable is what ﬁnally allowed UHF stations to achieve technical parity with VHF stations.167 It was thus no accident that new broadcast networks began to emerge shortly after cable became well established.168 The FCC thus has a long history of allocating channels in ways that exacerbated the problems of spec- trum scarcity. The unfortunate reality is that the FCC continued to do so even when the technological means for alleviating the problems of spectrum scarcity already existed. c. The Allocation of Individual Stations to Particular Licensees. 166. See, e.g., Amendment of Subpart L, Part 91, to Adopt Rules & Regulations to Govern Grant of Authorizations in Bus. Radio Serv. for Microwave Stations to Relay Television Signals to Community Antenna Sys., Second Report and Order, 2 F.C.C.2d 725, 774–77 ¶¶ 123–127, 778 ¶ 130 (1966). 167. Stanley M. Besen & Robert W. Crandall, The Deregulation of Cable Television, 44 LAW & CONTEMP. PROBS., Winter 1981, at 96–97; Hazlett, supra note 118, at 419–20. 168. For useful overviews of this era of cable regulation, see Besen & Crandall, supra note 167, at 93–124. 171. See POOL, supra note 51, at 141 (“Such licensing was the cause not the consequence of scarcity.”). 170. Id. at 143–47. 169. See, e.g., Hazlett, supra note 115, at 136. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT decisions, however, largely shields courts from any consideration of these underlying problems. decisions, however, largely shields courts from any consideration of these underlying problems. 176. Although the vast majority of U.S. households have a signiﬁcant number of broadcast televi- sion options, over 98% of all U.S. cities have only one daily newspaper. See Eli M. Noam & Robert N. Freeman, The Media Monopoly and Other Myths, 29 TELEVISION Q. 18, 22 (1997). Compare Broadcast Station Totals as of September 30, 2001 (F.C.C. Oct. 30, 2001), available at http://www.fcc.gov/Bureaus/ Mass_Media/News_Releases/2001/nrmm0112.txt (reporting that there are presently 1676 broadcast television stations in the U.S.), with WORLD ALMANAC 276 (2002) (reporting that there are presently 1480 daily newspapers in the U.S.). 172. Red Lion Broad. Co. v. FCC, 395 U.S. 367, 396 (1969). 173. See Laurence H. Winer, The Signal Cable Sends—Part I: Why Can’t Cable Be More Like Broadcasting?, 46 MD. L. REV. 212, 238 (1987) (quoting congressional testimony that the amount of usable spectrum had increased by approximately 20% each year for the last several decades). 174. See id. at 238–39; POOL, supra note 51, at 152–53; Fowler & Brenner, supra note 111, at 222–23; Sidak, supra note 111, at 1230. 175. Compare Review of Rules & Policies Concerning Network Broad. by Television Stations: Elimination or Modiﬁcation of Section 73.658(c) of Comm’n’s Rules, Notice of Proposed Rulemaking, 3 F.C.C.R. 5681, 5685 ¶ 17 (1988) (reporting that the average U.S. household could receive 3.9 over-the-air television stations in 1980), with 1998 Biennial Regulatory Review—Review of Comm’n’s Broad. Ownership Rules and Other Rules Adopted Pursuant to Section 202 of Telecomms. Act of 1996, Biennial Review Report, 15 F.C.C.R. 11058, 11064 ¶ 9 (2000) (reporting that as of 2000 the average U.S. household could receive thirteen over-the-air television stations). 176. Although the vast majority of U.S. households have a signiﬁcant number of broadcast televi- sion options, over 98% of all U.S. cities have only one daily newspaper. See Eli M. Noam & Robert N. Freeman, The Media Monopoly and Other Myths, 29 TELEVISION Q. 18, 22 (1997). Compare Broadcast Station Totals as of September 30, 2001 (F.C.C. Oct. 30, 2001), available at http://www.fcc.gov/Bureaus/ Mass_Media/News_Releases/2001/nrmm0112.txt (reporting that there are presently 1676 broadcast television stations in the U.S.), with WORLD ALMANAC 276 (2002) (reporting that there are presently 1480 daily newspapers in the U.S.). 175. Compare Review of Rules & Policies Concerning Network Broad. by Television Stations: Elimination or Modiﬁcation of Section 73.658(c) of Comm’n’s Rules, Notice of Proposed Rulemaking, 3 F.C.C.R. 5681, 5685 ¶ 17 (1988) (reporting that the average U.S. household could receive 3.9 over-the-air television stations in 1980), with 1998 Biennial Regulatory Review—Review of Comm’n’s Broad. Ownership Rules and Other Rules Adopted Pursuant to Section 202 of Telecomms. Act of 1996, Biennial Review Report, 15 F.C.C.R. 11058, 11064 ¶ 9 (2000) (reporting that as of 2000 the average U.S. household could receive thirteen over-the-air television stations). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Not only did federal regulatory policy exacerbate the problems of scarcity by restricting the supply of broadcast spectrum, it further compounded the problem by allocating spectrum to individual license holders in a manner guaranteed to stimulate excess demand. This is because the federal government has always given away initial licenses and renewed existing licenses for free.169 It is an economic truism, however, that demand will outstrip supply whenever any good is given away for free. The ordinary solution to such shortages is to employ a price mechanism because any increase in price will simultaneously stimulate addi- tional supply and reduce demand until the two reach equilibrium. Indeed, the longstanding existence of vibrant markets in which broadcast stations (and their accompanying licenses) are bought and sold suggests that a price mechanism would likely be quite effective in balancing demand with supply.170 In light of this, the shortages associated with scarcity appear to be the direct result of the government’s commitment to price licenses at zero. As a result, scarcity emerges as an example of the technique discussed above with respect to Reno and Denver, in which the Court incorporates the features of the existing regulatory regime into its constitutional baseline when determin- ing whether a First Amendment violation has occurred.171 The inevitable effect of this type of reasoning is to allow regulation to become self-reinforcing by permitting it to serve as a constitutional justiﬁcation for additional regulation. The problem is that the Court never considered whether the other regulatory decisions that formed the baseline for scarcity could not themselves withstand constitutional scrutiny. Confronting the issue in terms of particular licensing 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 279 172. Red Lion Broad. Co. v. FCC, 395 U.S. 367, 396 (1969). B. “SCARCITY IS NOT ENTIRELY A THING OF THE PAST”172: THE TECHNOLOGICAL CRITIQUE OF SCARCITY At the same time that academic criticism has undermined the theoretical underpinnings of the scarcity doctrine, technological change has worked to erode its empirical foundations. Previous commentary focusing on technologi- cal change has raised two core criticisms. First, although the scarcity doctrine implicitly treats the total amount of available spectrum as ﬁxed and unchanging, in reality, technological progress has steadily expanded the range of the electro- magnetic spectrum available for commercial use. The FCC’s technical staff has recognized that improvements in engineering have historically caused a steady increase in the amount of usable spectrum.173 In addition, other developments have allowed us to make more efﬁcient use of the spectrum already available. Improvements in channel spacing, the use of lower power, improved receiver technology, and other management techniques have further loosened the natural restriction imposed by the radio spectrum.174 As a result, the number of over-the-air television stations that the average U.S. household can receive has more than tripled over the last twenty years.175 Indeed, studies indicate that the average household has more options for broadcast television than for daily newspapers.176 Second, the scarcity doctrine was further undercut by the arrival of alterna- tive television technologies, such as cable television and direct broadcast satel- lite systems (DBS), capable of providing large numbers of channels without being subject to the type of constraints faced by broadcasters. These technolo- [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 280 gies are now essentially universally available.177 In fact, cable, DBS, and the other multichannel video program distributors (MVPDs) have eclipsed over-the- air broadcasting as the nation’s primary source of television programming. The FCC estimates that approximately eighty percent of U.S. households subscribe to cable, DBS, or similar MVPD.178 As a result, these alternative technologies have, in effect, eliminated the scarcity of the spectrum as a constraint to television-based communications.179 The Supreme Court, however, has a long history of viewing each technology as a universe unto itself.180 Its broadcasting precedents are no exception, as the Court has consistently refrained from treating the arrival of alternative televi- sion technologies as a basis for revisiting the scarcity doctrine.181 It thus seems that any attempt to overrule scarcity on empirical grounds is most likely to succeed if it arises within the context of broadcasting simpliciter without involving any alternative means of transmission. 177. Cable is now available in 97% of all U.S. households. Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programming, Eighth Annual Report, 17 F.C.C.R. 1244, 1330 tbl.B-1 (2002). DBS is available to any home with a clear line of sight to the southern sky. Id. at 1299–1300 ¶ 122. 183. Advanced Television Sys. and Their Impact Upon Existing Television Broad. Serv., Fourth Report and Order, 11 F.C.C.R. 17771, 17772 ¶ 5 (1996), modiﬁed, 12 F.C.C.R. 3388 (1997). 178. Id. at 1282 ¶ 79, 1338 tbl.C-1. It is arguable that the current level of MVPD penetration understates the true demand for television because the current availability of free, over-the-air broadcast- ing relieves many who are willing and able to pay for television programming from having to do so. f p y g 182. For an excellent, nontechnical overview of many of these emerging technologies, see BRUCE M. OWEN, THE INTERNET CHALLENGE TO TELEVISION 245–325 (1999). 181. See League of Women Voters v. FCC, 468 U.S. 364, 376 n.11 (1984); Red Lion Broad. Co. v. FCC, 395 U.S. 367, 396 (1969). For a more complete discussion of the Court’s reluctance to reconsider the scarcity doctrine in League of Women Voters notwithstanding the growth of cable television, see infra note 203 and accompanying text. 179. See Yoo, supra note 1, at 206–08, 228–29. 180. See id. at 285–86 (discussing the Court’s historical tendency to assume that patents confer monopoly power without inquiring whether substitute technologies exist). 1330 tbl.B-1 (2002). DBS is available to any home with a clear line of sight to the southern sky. Id. at 1299–1300 ¶ 122. 178. Id. at 1282 ¶ 79, 1338 tbl.C-1. It is arguable that the current level of MVPD penetration understates the true demand for television because the current availability of free, over-the-air broadcast- ing relieves many who are willing and able to pay for television programming from having to do so. 179. See Yoo, supra note 1, at 206–08, 228–29. 180. See id. at 285–86 (discussing the Court’s historical tendency to assume that patents confer monopoly power without inquiring whether substitute technologies exist). 181. See League of Women Voters v. FCC, 468 U.S. 364, 376 n.11 (1984); Red Lion Broad. Co. v. FCC, 395 U.S. 367, 396 (1969). For a more complete discussion of the Court’s reluctance to reconsider the scarcity doctrine in League of Women Voters notwithstanding the growth of cable television, see infra note 203 and accompanying text. 182. For an excellent, nontechnical overview of many of these emerging technologies, see BRUCE M. OWEN, THE INTERNET CHALLENGE TO TELEVISION 245–325 (1999). 183. Advanced Television Sys. and Their Impact Upon Existing Television Broad. Serv., Fourth Report and Order, 11 F.C.C.R. 17771, 17772 ¶ 5 (1996), modiﬁed, 12 F.C.C.R. 3388 (1997). B. “SCARCITY IS NOT ENTIRELY A THING OF THE PAST”172: THE TECHNOLOGICAL CRITIQUE OF SCARCITY It is this limitation that makes the current emergence of new broadcast technologies so potentially transformative.182 For the ﬁrst time, the Court will be forced to entertain an empirical attack to the scarcity doctrine that has traction even if evaluated entirely from within the conﬁnes of broadcasting. The most sweeping technological development is the advent of digital television. Although digital television is often associated exclusively with high deﬁnition television (HDTV), what many people do not realize is that digital broadcasters have another option. Rather than transmitting the higher quality pictures associ- ated with HDTV, digital broadcasters can instead use the greater efﬁciency of digital transmission to increase the number of channels transmitted. The FCC estimates that if the resolution of the television picture is left at its current levels, digital broadcasters can send ﬁve or more standard deﬁnition digital signals in the same amount of spectrum needed to send a single high deﬁnition digital signal.183 The prospect of a ﬁvefold increase in the number of channels that any household can receive makes it clear that economic limits have 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 281 surpassed the physical limits posed by the electromagnetic spectrum as the relevant constraint.184 Waiting in the wings are a number of other technologies that should further enhance our ability to use the spectrum. Personal video recorders (PVRs) employ hard drives and other computer-based technologies to store up to 320 hours of digital programming.185 Although industry leaders TiVo and ReplayTV have largely abandoned attempts to sell PVRs as freestanding units, this technol- ogy is in the process of being incorporated into other devices.186 Widespread deployment of PVRs promises to allow broadcasters to increase the efﬁciency of spectrum use by using a greater proportion of the broadcast day for transmis- sion without incurring the degradation in quality associated with current analog technologies. The Chairman of the FCC recently expressed his enthusiastic support for the technology, even going so far as to dub it, “God’s machine.”187 Equally promising are the cluster of emerging “spread spectrum” technolo- gies. Some of these technologies, including a popular one known as ultra- wideband (UWB) that serves as the basis for many wireless local area network (LAN) systems, operate by sending an extremely short duration pulse over a much broader range of the spectrum than is usual for spectrum-based devices. 188. See Revision of Part 15 of Comm’n’s Rules Regarding Ultra-Wideband Transmission Sys., First Report and Order, 17 F.C.C.R. 7435, 7439 ¶ 7 (2002) [hereinafter UWB First Report and Order]; Revision of Part 15 of Comm’n’s Rules Regarding Ultra-Wideband Transmission Sys., Notice of Proposed Rulemaking, 15 F.C.C.R. 12086, 12088 n.13 (2000) [hereinafter UWB NPRM]. The FCC recently authorized the marketing and operation of UWB devices. In so ruling, it limited communica- tions services based on UWB to indoor activities and peer-to-peer operation. See Revision of Part 15 of Comm’n’s Rules Regarding Ultra-Wideband Transmission Sys., First Report and Order, 17 F.C.C.R. 7435, 7437 ¶ 5 (2002). 186. See Anna Wilde Mathews, ReplayTV Will Exit Direct Set-Top Sales, Focus on Licensing Software to Others, WALL ST. J., Nov. 28, 2000, at B10; Nick Wingﬁeld, Sony Agrees to License Video-Recording, WALL ST. J., Oct. 19, 2001, at B6. Interestingly, Microsoft recently abandoned its efforts to market a PVR. See Rebecca Buckman, Microsoft Corp. Disbands Division Making Ultima- teTV, WALL ST. J., Jan. 23, 2002, at B2l. 189. See, e.g., LAWRENCE LESSIG, CODE: AND OTHER LAWS OF CYBERSPACE 184 (1999); Tom W. Bell, The Common Law in Cyberspace, 97 MICH. L. REV. 1746, 1766 (1999); Yochai Benkler, Overcoming Agoraphobia: Building the Commons of the Digitally Networked Environment, 11 HARV. J.L. & TECH. 287, 325–26, 394–97 (1998). For a discussion of this proposal appearing in the popular press, see Yochai Benkler & Lawrence Lessig, Will Technology Make CBS Unconstitutional?, NEW REPUBLIC, 184. See Yoo, supra note 1, at 213, 227–28. This fact makes it quite ironic that the Gore Commission and the FCC have used the arrival of digital television as a basis for calling for more intrusive regulation. See supra note 97 and accompanying text. Properly understood, digital television weakens, rather than strengthens, the traditional grounds for regulatory intervention. g g g y 185. See David P. Hamilton, VCRs: Still Standing, WALL ST. J., Mar. 5, 2002, at R8. 187. Jim Krane, FCC Chief Is Big Fan of Digital Video Recorders, PHILA. INQUIRER, Jan. 12, 2003, at E2. B. “SCARCITY IS NOT ENTIRELY A THING OF THE PAST”172: THE TECHNOLOGICAL CRITIQUE OF SCARCITY The diffusion of the signal allows UWB to transmit at such low power that it is almost indistinguishable from background noise. As a result, it allows more users to operate in the same spectrum without causing interference.188 Indeed, some have claimed that the deployment of such technologies will obviate the need for spectrum rights altogether.189 I am personally skeptical that such 282 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 spread spectrum technologies can support the type of spectrum commons that these scholars envision. Although the airwaves may seem uncrowded initially, as the relevant technologies develop, the absence of well-deﬁned property rights will eventually lead to the overuse and underinvestment associated with any commons.190 This is particularly true with UWB, because widescale use of low power emissions would gradually raise the level of background noise that spectrum users will need to overcome. In addition, such problems are likely to be extremely hard to solve because interference in the spectrum is cumulative, and it is likely to be extremely difﬁcult to identify the source of emissions that are fairly close to the level of background noise. Fortunately, other spread spectrum technologies exist that do not suffer from these ﬂaws. For example, the FCC recently authorized the deployment of “software-deﬁned radio” (SDR),191 which is the ﬁrst spread spectrum technology to operate as a mass media.192 Unlike conventional radio, which employs transmitters that are hard- wired to transmit and receive on a single frequency, SDR employs computer- based technology to allow transmissions to shift dynamically among frequencies.193 As a result, the audience can receive seamless service despite the fact that a particular broadcast may hop among different channels at different times of the day. The added ﬂexibility provided by SDR promises to improve efﬁcient use of spectrum in several ways. First, by enabling transmissions to adapt dynamically to local conditions, SDR makes it possible to maximize use of the available bandwidth. Second, it enhances incentives for conserving spectrum by making it easier for any savings to be transferred to alternative uses. Third, because the key elements of SDR equipment are based in software, not hardware, the shift to SDR can minimize the problems associated with nonuniform standards that vary from service to service and from country to country. g p g 194. See SDR First Report & Order, supra note 191, at 17374 ¶ 5; Authorization and Use of Software Deﬁned Radios, Notice of Proposed Rulemaking, 15 F.C.C.R. 24442, 24443 ¶¶ 3–4, 24446–48 ¶¶ 12–15 (2000); Inquiry Regarding Software Deﬁned Radios, Notice of Inquiry, 15 F.C.C.R. 5930, 5933–36 ¶¶ 10–17 (2000). Dec. 14, 1998, at 12, 15 (“If spectrum can be shared, does the Constitution permit the state to silence the many so that CBS can speak?”). For a related proposal that would charge spectrum fees based on congestion, see Eli Noam, Spectrum Auctions: Yesterday’s Heresy, Today’s Orthodoxy, Tomorrow’s Anachronism, 41 J.L. & ECON. 765 (1998) (advocating a “pay as you go,” license-free spectrum). 193. Because SDR uses speciﬁc channels, it is quite different from other spread spectrum technolo- gies that disperse transmissions over a broader range in a nonchannelized fashion, such as UWB. 192. The FCC has made it clear that UWB does not encompass long-range spread spectrum technologies. See UWB First Report and Order, supra note 188, at 7437 ¶ 5; UWB NPRM, supra note 188, at 12088 n.13. 191. See Authorization and Use of Software Deﬁned Radios, First Report and Order, 16 F.C.C.R. 17373 (2001) [hereinafter SDR First Report & Order]; Hazlett, supra note 118, at 444–46. p 191. See Authorization and Use of Software Deﬁned Radios, First Report and Order, 16 F.C.C.R. 191. See Authorization and Use of Software Deﬁned Radios, First Report and Order, 16 F.C.C.R. 17373 (2001) [hereinafter SDR First Report & Order]; Hazlett supra note 118 at 444–46 195. See generally Yoo, supra note 1, at 289 (“The impending shift of all networks to packet- switched technology promises to cause all of the distinctions based on the means of conveyance and the type of speech to collapse entirely.”). 196. In fact, the various technologies may move beyond being substitutes to being complements. For example, it is easy to envision a combined system that completely devoted the broadcast frequencies to bringing content into the home while having the telephone line completely devoted to the return path. 197. See GORE COMMISSION REPORT, supra note 17, at 20 & n.12 (declaring Red Lion “the operative ruling in this area” and citing cases relying on it); Policies and Rules Concerning Children’s Television THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 283 television broadcasting is years from becoming a reality. Nonetheless, it is apparent that SDR has the potential to enhance the efﬁciency with which we can use the spectrum in some fairly dramatic ways. And lurking over the entire scene is the impending conversion of broadcast- ing to packet-switched technologies. Unlike conventional communications me- dia, which typically transmit information in a single, continuous stream, packet- switched networks divide information into smaller aggregations of data known as “packets.” The packets are then sent to their destination through the most efﬁcient route and are reassembled. The conversion of broadcasting to packet- switched technologies should improve the efﬁciency of broadcast transmission in several ways. First, packet-switched networks can take full advantage of digital compression and other techniques developed to enhance the efﬁciency of computer networks. Dividing information into packets also allows for more efﬁcient network use because different packets can be routed through whatever paths are least congested at the time of transmission. The use of packet- switched networks can thus enhance the ability of ﬂexible technologies like SDR to use the available spectrum in the most efﬁcient manner possible. But perhaps the most important insight for the purposes of this Article is that packet-switched networks treat all different forms of communication and all means of transmission as essentially fungible.195 The technology follows the same process of breaking digital information down into individual packets regardless of whether it is conveying a television program, a telephone conversa- tion, or an e-mail message. In addition, it makes no difference in the end if the packets associated with a particular communication arrived via terrestrial broad- casting, satellite broadcasting, coaxial cable, the traditional twisted pair associ- ated with telephony, or all of the above. As a result, the eventual conversion of television to packet switched technologies will render any remaining distinc- tions between the various media technologies meaningless because all of them will in essence become substitutes for one another.196 Once that day arrives, any continued effort to draw distinctions among media will clearly become sense- less. B. “SCARCITY IS NOT ENTIRELY A THING OF THE PAST”172: THE TECHNOLOGICAL CRITIQUE OF SCARCITY Lastly, the software-oriented nature of SDR can also greatly facilitate the introduction of new, more spectrum-efﬁcient technologies by lowering the costs of changing over receiving and transmitting equipment.194 Admittedly, SDR is still in its nascent stages and deployment of SDR technology for 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 197. See GORE COMMISSION REPORT, supra note 17, at 20 & n.12 (declaring Red Lion “the operative ruling in this area” and citing cases relying on it); Policies and Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10727–32 ¶¶ 149–151 (1996) (drawing a similar conclusion). 198. 497 U.S. 547 (1989). 199. Red Lion Broad. Co. v. FCC, 395 U.S. 367, 396–97 (1969). 200. Id. at 399. 201. 412 U.S. 94 (1973). 202. Id. at 102 (plurality opinion). C. WITH A WHIMPER, NOT A BANG: THE DOCTRINAL COLLAPSE OF THE SCARCITY DOCTRINE Despite these analytical and conceptual problems, policymakers continue to regard the scarcity doctrine as beyond judicial reproach.197 My own review of 284 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 the precedents suggests that the courts have not been completely oblivious to the analytical and empirical shortcomings of the scarcity doctrine and have even signaled willingness to consider overruling it. The initial push in that direction was ultimately blunted by an unusual conﬂuence of issues in Metro Broadcast- ing, Inc. v. FCC.198 Since then, however, the Supreme Court has once again begun to distance itself from the scarcity doctrine by refusing to extend it to other media and appearing to avoid relying on it even with respect to broadcast- ing. These developments suggest that there may be more reason to believe that the courts may be ready to abandon the doctrine than is commonly acknowl- edged. 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting Even while offering the most celebrated statement of the scarcity doctrine, the Court’s opinion in Red Lion recognized the possibility that improvements in technology might undermine its empirical basis. When considering the argu- ment that scientiﬁc progress had rendered scarcity obsolete, the Court conceded that “[a]dvances in technology, such as microwave transmission, have led to more efﬁcient utilization of the frequency spectrum.”199 The Court, however, saw technological change as a double-edged sword. Although scientiﬁc ad- vances tended to improve our ability to employ the spectrum, those advances simultaneously created additional demand for it. The Court thus saw the obsoles- cence of scarcity as a question of which of these two effects would dominate in the long run. The Court declined to resolve the tension between these forces, concluding that in the absence of a concrete factual record, it was “unwise to speculate on the future allocation of that space.”200 In postponing a decision on this issue rather than rejecting it out of hand, the Court implicitly recognized the possibility that scarcity might one day become a thing of the past. The Court’s subsequent decisions have been even more explicit in acknowledg- ing the possibility that technology might undercut scarcity as a basis for upholding the constitutionality of broadcast regulation. For example, in CBS v. Democratic National Committee,201 a plurality of the Court explicitly acknowl- edged that “the broadcast industry is dynamic in terms of technological change” and that “solutions adequate a decade ago are not necessarily so now, and those acceptable today may well be outmoded 10 years hence.”202 Justice Douglas put the matter even more directly: “Scarcity may soon be a constraint of the past, thus obviating the concerns expressed in Red Lion. It has been predicted that it 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 285 may be possible within 10 years to provide television viewers 400 channels through the advances of cable television.”203 This line of authority culminated in FCC v. Id. at 129. 208. For a detailed and nuanced overview of the events that followed, see Neal Devins, Congress, the FCC, and the Search for the Public Trustee, 56 LAW & CONTEMP. PROBS., Aut. 1993, at 152–65, 177–78; see also Susan Low Bloch, Orphaned Rules in the Administrative State: The Fairness Doctrine and Other Orphaned Progeny of Interactive Deregulation, 79 GEO. L.J. 59, 59–62, 66–87 (1987); Matthew L. Spitzer, The Constitutionality of Licensing Broadcasters, 64 N.Y.U. L. REV. 990, 1009–12 (1989). 209. Inquiry into Section 73.1910 of Comm’n’s Rules & Regulations Concerning Gen. Fairness Doctrine Obligations of Broad. Licensees, Report, 102 F.C.C.2d 145, 204–17 ¶¶ 97–122 (1985) [hereinafter 1985 Fairness Report]. The Report also concluded that the Fairness Doctrine was constitu- 203. Id. at 158 n.8 (Douglas, J., concurring in the judgment). 204. 468 U.S. 364 (1984). 205. Id. at 376 n.11 (citing Fowler & Brenner, supra note 111, at 221–26). 206. Id. 207. Landmark Communications, Inc. v. Virginia, 435 U.S. 829, 843 (1978); see also Sable Communications of Cal., Inc. v. FCC, 492 U.S. 115 (1989): To the extent that the federal parties suggest that we should defer to Congress’ conclusion about an issue of constitutional law, our answer is that while we do not ignore it, it is our task 204. 468 U.S. 364 (1984). 204. 468 U.S. 364 (1984). 205. Id. at 376 n.11 (citing Fowler & Brenner, supra note 111, at 221–26). 206. Id. 207. Landmark Communications, Inc. v. Virginia, 435 U.S. 829, 843 (1978); see also Sable Communications of Cal., Inc. v. FCC, 492 U.S. 115 (1989): 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting League of Women Voters,204 in which the Court recognized that “[c]ritics, including the incumbent Chairman of the FCC, charge that with the advent of cable and satellite television technology, communities now have access to such a wide variety of stations that the scarcity doctrine is obsolete.”205 Perhaps reluctant to open itself to the political heat that would accompany invalidating most of the extant regime of broadcast regula- tion, the Court declined to resolve the issue, holding instead that “[w]e are not prepared . . . to reconsider our longstanding approach without some signal from Congress or the FCC that technological developments have advanced so far that some revision of the system of broadcast regulation may be required.”206 Thus, notwithstanding its previous admonition that “[d]eference to a legislative ﬁnd- ing cannot limit judicial inquiry when First Amendment rights are at stake,”207 the Court felt the need to defer to the political branches. Although the propriety of the Court’s decision not to resolve the issue is open to question, the League of Women Voters footnote nonetheless established a road map for the repudiation of the scarcity doctrine in the future. In light of the deregulatory bent of the Reagan Administration in general and the FCC in particular, the FCC did not wait long to take the Supreme Court up on its invitation.208 The year after the Court’s decision in League of Women Voters, the FCC issued a study reviewing the constitutionality of the Fairness Doctrine. Known as the 1985 Fairness Report, this study concluded that the increase in the number of television stations and the emergence of cable television and other new television technologies had undercut scarcity as a basis for giving broadcasting a lesser degree of First Amendment protection than other media.209 To the extent that the federal parties suggest that we should defer to Congress’ conclusion about an issue of constitutional law, our answer is that while we do not ignore it, it is our task in the end to decide whether Congress has violated the Constitution. This is particularly true where the Legislature has concluded that its product does not violate the First Amendment. THE GEORGETOWN LAW JOURNAL 286 [Vol. 91:245 Strong congressional support for the Fairness Doctrine led the FCC to stop short of repealing it. tionally suspect in that it actually lessened the amount of diverse views available to the public. Faced with the obligation to provide balanced coverage whenever controversial issues are discussed, the FCC found that all too often broadcasters avoided the problem altogether simply by declining to address controversial issues at all. Id. at 159–88 ¶¶ 26–68. The Report further concluded that the Fairness Doctrine inhibited the expression of unorthodox views and gave the government too much power to inﬂuence the content of broadcast programming. Id. at 188–94 ¶¶ 69–76. In so ruling, the FCC provided a basis under which a court could reconsider the constitutionality of the Fairness Doctrine without calling the broader sweep of broadcast regulation into question. See Red Lion Broad. Co. v. FCC, 395 U.S. 367, 393 (1969) (noting that the Court would reconsider the constitutionality of the Fairness Doctrine “if experience with the administration of those doctrines indicates that they have the net effect of reducing rather than enhancing the volume and quality of coverage”); League of Women Voters, 468 U.S. at 378 n.12 (offering a similar observation). 210. 1985 Fairness Report, supra note 209, at 148 ¶ 7; see also id. 227 ¶ 144, 246 47 ¶¶ 174 176. 211. See Branch v. FCC, 824 F.2d 37, 50 (D.C. Cir. 1987); Meredith Corp. v. FCC, 809 F.2d 863, 867 (D.C. Cir. 1987); Telecomms. Research & Action Ctr. v. FCC, 801 F.2d 501, 509 n.5 (D.C. Cir. 1986); Bloch, supra note 208, at 74 n.56; Spitzer, supra note 208, at 1009. p , 213. See Syracuse Peace Council, 2 F.C.C.R. 5043, 5053–54 ¶¶ 66–72 (1987) (mem. op. & order), aff’d, 867 F.2d 654 (D.C. Cir. 1989). 214. Id. at 5058 ¶ 99. 215. Id. at 5053 ¶ 65. p p 212. See Meredith Corp., 809 F.2d at 874. 213. See Syracuse Peace Council, 2 F.C.C.R. 5043, 5053–54 ¶¶ 66–72 (1987) (mem. op. & ord aff’d, 867 F.2d 654 (D.C. Cir. 1989). , ( g ) 210. 1985 Fairness Report, supra note 209, at 148 ¶ 7; see also id. 227 ¶ 144, 246–47 ¶¶ 174–176 214. Id. at 5058 ¶ 99. 215. Id. at 5053 ¶ 65. 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting In light of the Fairness Doctrine’s standing as “a longstanding administrative policy and central tenet of broadcast regulation” and the exis- tence of “proposals pending before Congress to repeal the doctrine,” the FCC concluded that it would be inappropriate to repeal the Fairness Doctrine out- right. Much as the Supreme Court had avoided deciding the constitutional issue in League of Women Voters by deferring to the political branches, the FCC sidestepped the issue as well, opting instead to “afford Congress an opportunity to review the Fairness Doctrine in light of the evidence adduced in this proceeding.”210 That the FCC stopped short of reaching the constitutional issue did not stop many contemporary observers from regarding the 1985 Fairness Report as the type of signal that would justify the abandonment of the scarcity doctrine.211 In 1987, the FCC removed any remaining doubts about its intentions. In response to a D.C. Circuit decision ordering it to address the constitutional issues the FCC had avoided in the 1985 Fairness Report,212 the FCC repealed the Fairness Doctrine.213 In the course of doing so, the FCC reafﬁrmed its attack on the scarcity doctrine, concluding that “the dramatic transformation in the telecommu- nications marketplace provides a basis for the Court to reconsider its application of diminished First Amendment protection to the electronic media.”214 Even more importantly, the FCC explicitly indicated that it intended its action to represent the type of signal envisioned by the Supreme Court in League of Women Voters.215 The FCC’s bold action appeared to set the stage for a landmark renunciation of the scarcity doctrine that would bring the Broadcast Model to an abrupt end. Somewhat surprisingly, however, the Supreme Court did not overturn the 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 287 scarcity doctrine in its next major broadcasting decision. To the contrary, in its 1990 decision in Metro Broadcasting, Inc. v. FCC,216 a decision better known for its equal protection implications, the Court reafﬁrmed scarcity as a basis for applying a lower First Amendment standard to broadcasting.217 Although the decision was closely divided on other issues, the dissent refused to challenge the majority’s endorsement of the scarcity doctrine and instead accepted the notions that “First Amendment concerns support limited but inevitable Government regulation of the peculiarly constrained broadcasting spectrum” and that, as such, “measures adopted to further the interest in diversity of broadcasting viewpoints are . . . 221. Adarand Constructors, Inc. v. Pen˜a, 515 U.S. 200, 227 (1995) (overruling Metro Broadcasting only “[t]o the extent that [it] is inconsistent” with the holding “that all racial classiﬁcations . . . must be analyzed by a reviewing court under strict scrutiny”). 216. 497 U.S. 547 (1990). 217. Id. at 566–67 (“We have long recognized that ‘[b]ecause of the scarcity of [electromagnetic] frequencies, the Government is permitted to put restraints on licensees in favor of others whose views should be expressed on this unique medium.’” (quoting Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969))) (alterations in original). g 218. Id. at 616 (O’Connor, J., joined by Rehnquist, C.J., Scalia & Kennedy, JJ., dissenting) (citing Red Lion, 395 U.S. at 389–90). 220. See Devins, supra note 208, at 179. ) 219. 488 U.S. 469 (1989). 216. 497 U.S. 547 (1990). 217. Id. at 566–67 (“We have long recognized that ‘[b]ecause of the scarcity of [electromagnetic] frequencies, the Government is permitted to put restraints on licensees in favor of others whose views should be expressed on this unique medium.’” (quoting Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969))) (alterations in original). 218. Id. at 616 (O’Connor, J., joined by Rehnquist, C.J., Scalia & Kennedy, JJ., dissenting) (citing Red Lion, 395 U.S. at 389–90). 219. 488 U.S. 469 (1989). 220. See Devins, supra note 208, at 179. 221. Adarand Constructors, Inc. v. Pen˜a, 515 U.S. 200, 227 (1995) (overruling Metro Broadcasting only “[t]o the extent that [it] is inconsistent” with the holding “that all racial classiﬁcations . . . must be analyzed by a reviewing court under strict scrutiny”). 216. 497 U.S. 547 (1990). 216. 497 U.S. 547 (1990). 217. Id. at 566–67 (“We have long recognized that ‘[b]ecause of the scarcity of [electromagnetic] frequencies, the Government is permitted to put restraints on licensees in favor of others whose views should be expressed on this unique medium.’” (quoting Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969))) (alterations in original). 218. Id. at 616 (O’Connor, J., joined by Rehnquist, C.J., Scalia & Kennedy, JJ., dissenting) (citing Red Lion, 395 U.S. at 389–90). 219 488 U S 469 (1989) 222. See Reno v. ACLU, 521 U.S. 844, 870 (1997); Turner Broad. Sys. v. FCC, 512 U.S. 622, 637–39 (1994) [hereinafter Turner I]. For earlier refusals to extend the doctrine to other media, see Pac. Gas & Electric Co. v. Public Utilities Comm’n, 475 U.S. 1, 10 n.6 (1986) (plurality opinion); Consolidated Edison Co. v. Public Utilities Comm’n, 447 U.S. 530, 542–43 (1980); Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 74 (1983) (overturning a ban on the mailing of unsolicited contracep- tive advertisements on the ground that “[o]ur decisions have recognized that the special interest of the Federal Government in regulation of the broadcast media does not readily translate into a justiﬁcation for regulation of other means of communication”). p g g ) 224. See Preferred Communications, 476 U.S. at 495 (declining “to express any more detailed views on the proper resolution of the First Amendment question”); id. at 496 (Blackmun, J., concurring) (joining the majority opinion subject to the understanding that “it leaves open the question of the proper standard for judging First Amendment challenges” to the cable regulations in question). 223. See Leathers v. Medlock, 499 U.S. 439, 444 (1991); City of Los Angeles v. Preferred Communications, Inc., 476 U.S. 488, 493–94 (1986); cf. FCC v. Midwest Video Corp., 440 U.S. 689, 707 (1979) (noting that cable operators exercise “a signiﬁcant amount of editorial discretion regarding what their programming will include”). THE GEORGETOWN LAW JOURNAL scarcity doctrine as a formal matter, it did nothing to resolve the burgeoning questions about its continuing validity. 1. A False Start in the Abandonment of Scarcity: From League of Women Voters to Metro Broadcasting no[t] contrary to the First Amendment.”218 Coming after the FCC’s attempt to answer the Supreme Court’s request for a signal that would justify revisiting the scarcity doctrine, this reafﬁrmance came as something of a surprise. Even more striking was the Court’s decision to apply an intermediate level of equal protection scrutiny to the minority preference programs at issue in Metro Broadcasting, as well as its reliance on a nonreme- dial governmental interest in sustaining the programs. Just the previous year in City of Richmond v. J.A. Croson Co.,219 the Court had indicated that it would likely subject such programs to strict scrutiny and would hold any attempts to justify race-conscious measures as remedies for past discrimination to a fairly stringent evidentiary standard. The answer to these puzzles lies with Justice Byron White, one of two justices to vote with the majority in both Croson and Metro Broadcasting. Justice Brennan’s papers reveal that White made a reafﬁr- mation of Red Lion the price of his vote in Metro Broadcasting.220 Dominated as it was by the larger battle over afﬁrmative action, the majority opinion in Metro Broadcasting can hardly be regarded as a ringing endorsement of Red Lion. Here was a clear case of the equal protection tail wagging the First Amendment dog. And yet, somewhat ironically, it is the broadcast implications that have proven more enduring. When the Court later overruled Metro Broad- casting, it left the decision’s endorsement of scarcity intact.221 In so doing, the Court brought the reexamination of the scarcity doctrine prompted by the League of Women Voters footnote to an unsatisfying and somewhat indetermi- nate end. Although there can be little question that the Court reafﬁrmed the 288 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 230. See id. at 674–75 (O’Connor, J., joined in relevant part by Scalia, Thomas, & Ginsburg, JJ., concurring in part & dissenting in part) (“As the Court explains in Parts I, II-A and II-B of its opinion, which I join, cable programmers and operators stand in the same position under the First Amendment as do more traditional media.”). The primary disagreement between the majority and the dissent centered on whether must-carry represented structural rather than content regulation and whether must-carry was sufﬁciently narrowly tailored to survive the appropriate level of scrutiny. See id. at 675–78, 682–85 (O’Connor, J., concurring in part & dissenting in part). For an excellent review of the reasons for regarding must-carry as content based, see Winer, supra note 228, at 25–45. pp y g p 228. Id. Some commentators have interpreted Turner I’s refusal to question the continuing vitality of scarcity as representing an implicit reafﬁrmation of the doctrine with respect to broadcasting. See Laurence H. Winer, The Red Lion of Cable, and Beyond?—Turner Broadcasting v. FCC, 15 CARDOZO ARTS & ENT. L.J. 1, 21–22 (1997). The structure and tone of the majority opinion suggest that the Court reserved the issue as a matter of judicial restraint and not to signal its support for the doctrine. 229 512 U S t 639 g p 227. See Turner I, 512 U.S. at 637 (“[T]he rationale for applying a less rigorous standard of First Amendment scrutiny to broadcast regulation, whatever its validity in the cases elaborating it, does not apply in the context of cable regulation”) (emphasis added). 229. 512 U.S. at 639. 225. See Quincy Cable TV, Inc. v. FCC, 768 F.2d 1434, 1448–50 (D.C. Cir. 1985); Preferred Communications, Inc. v. City of Los Angeles, 754 F.2d 1396, 1404 (9th Cir. 1985), aff’d on other grounds, 476 U.S. 488 (1986); Omega Satellite Prods. Co. v. City of Indianapolis, 694 F.2d 119, 127 (7th Cir. 1982); Home Box Ofﬁce, Inc. v. FCC, 567 F.2d 9, 44–46 (D.C. Cir. 1977). 226. See Turner I, 512 U.S. at 63 & n.5 (citing Telecomms. Action & Research Ctr. v. FCC, 801 F.2d 501, 508–09 (D.C. Cir. 1986)); BOLLINGER, supra note 11, at 87–90; POWE, supra note 51, at 197–209; SPITZER, supra note 111, at 7–18; Coase, supra note 29, at 12–27; Winer, supra note 173, at 218–40; Note, The Message in the Medium, supra note 120, at 1072–74. 2. Scarcity After Metro Broadcasting The reafﬁrmation of the scarcity doctrine in Metro Broadcasting blunted the momentum in favor of its outright abandonment initiated by the League of Women Voters footnote. Metro Broadcasting did not, however, mark an end to the judiciary’s struggles with the doctrine’s analytical and technological shortcom- ings. As the following subsection demonstrates, courts have severely restricted the doctrine’s scope by consistently rejecting calls to extend it to other media. In addition, courts have even begun to curtail the scarcity doctrine with respect to broadcasting. While some lower court judges have continued to call for its repudiation, the Supreme Court has appeared to respond with avoidance, opting to rely on other principles to justify applying a lower level of First Amendment scrutiny even when a simple citation to the scarcity doctrine would have sufﬁced. The Supreme Court’s scrupulous refusal to offer any endorsement of the scarcity doctrine suggests that the Court may be closer to abandoning the doctrine than is generally recognized. a. The Refusal to Extend Scarcity to Other Media. The only time that the Supreme Court has found occasion to mention the scarcity doctrine since Metro Broadcasting was in the process of rejecting requests to extend it to other media.222 The Court offered its most extensive discussion of the extent to which the scarcity doctrine would apply to other electronic media in Turner I. At issue was a First Amendment challenge to the so-called “must carry” statute, which required local cable operators to carry for free the signals of all full-power broadcast stations within their service area. Even though the Court had recog- nized that cable television was protected speech,223 the First Amendment standard that would apply to cable television remained an open question.224 2003] 289 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 225. See Quincy Cable TV, Inc. v. FCC, 768 F.2d 1434, 1448–50 (D.C. Cir. 1985); Preferred Communications, Inc. v. City of Los Angeles, 754 F.2d 1396, 1404 (9th Cir. 1985), aff’d on other grounds, 476 U.S. 488 (1986); Omega Satellite Prods. Co. v. City of Indianapolis, 694 F.2d 119, 127 (7th Cir. 1982); Home Box Ofﬁce, Inc. v. FCC, 567 F.2d 9, 44–46 (D.C. Cir. 1977). 226. See Turner I, 512 U.S. at 63 & n.5 (citing Telecomms. Action & Research Ctr. v. FCC, 801 F.2d 501, 508–09 (D.C. Cir. 1986)); BOLLINGER, supra note 11, at 87–90; POWE, supra note 51, at 197–209; SPITZER, supra note 111, at 7–18; Coase, supra note 29, at 12–27; Winer, supra note 173, at 218–40; Note, The Message in the Medium, supra note 120, at 1072–74. 227. See Turner I, 512 U.S. at 637 (“[T]he rationale for applying a less rigorous standard of First Amendment scrutiny to broadcast regulation, whatever its validity in the cases elaborating it, does not apply in the context of cable regulation”) (emphasis added). 228. Id. Some commentators have interpreted Turner I’s refusal to question the continuing vitality of scarcity as representing an implicit reafﬁrmation of the doctrine with respect to broadcasting. See Laurence H. Winer, The Red Lion of Cable, and Beyond?—Turner Broadcasting v. FCC, 15 CARDOZO ARTS & ENT. L.J. 1, 21–22 (1997). The structure and tone of the majority opinion suggest that the Court reserved the issue as a matter of judicial restraint and not to signal its support for the doctrine. 229. 512 U.S. at 639. 230. See id. at 674–75 (O’Connor, J., joined in relevant part by Scalia, Thomas, & Ginsburg, JJ., concurring in part & dissenting in part) (“As the Court explains in Parts I, II-A and II-B of its opinion, which I join, cable programmers and operators stand in the same position under the First Amendment as do more traditional media.”). The primary disagreement between the majority and the dissent centered on whether must-carry represented structural rather than content regulation and whether must-carry was sufﬁciently narrowly tailored to survive the appropriate level of scrutiny. See id. at 675–78, 682–85 (O’Connor, J., concurring in part & dissenting in part). For an excellent review of the reasons for regarding must-carry as content based, see Winer, supra note 228, at 25–45. 225. See Quincy Cable TV, Inc. v. FCC, 768 F.2d 1434, 1448–50 (D.C. Cir. 1985); Preferred Communications, Inc. v. City of Los Angeles, 754 F.2d 1396, 1404 (9th Cir. 1985), aff’d on other grounds, 476 U.S. 488 (1986); Omega Satellite Prods. Co. v. City of Indianapolis, 694 F.2d 119, 127 (7th Cir. 1982); Home Box Ofﬁce, Inc. v. FCC, 567 F.2d 9, 44–46 (D.C. Cir. 1977). 226. See Turner I, 512 U.S. at 63 & n.5 (citing Telecomms. Action & Research Ctr. v. FCC, 801 F.2d 501, 508–09 (D.C. Cir. 1986)); BOLLINGER, supra note 11, at 87–90; POWE, supra note 51, at 197–209; SPITZER, supra note 111, at 7–18; Coase, supra note 29, at 12–27; Winer, supra note 173, at 218–40; Note, The Message in the Medium, supra note 120, at 1072–74. 227. See Turner I, 512 U.S. at 637 (“[T]he rationale for applying a less rigorous standard of First 225. See Quincy Cable TV, Inc. v. FCC, 768 F.2d 1434, 1448–50 (D.C. Cir. 1985); Preferred Communications, Inc. v. City of Los Angeles, 754 F.2d 1396, 1404 (9th Cir. 1985), aff’d on other grounds, 476 U.S. 488 (1986); Omega Satellite Prods. Co. v. City of Indianapolis, 694 F.2d 119, 127 (7th Cir. 1982); Home Box Ofﬁce, Inc. v. FCC, 567 F.2d 9, 44–46 (D.C. Cir. 1977). 226. See Turner I, 512 U.S. at 63 & n.5 (citing Telecomms. Action & Research Ctr. v. FCC, 801 F.2d 501, 508–09 (D.C. Cir. 1986)); BOLLINGER, supra note 11, at 87–90; POWE, supra note 51, at 197–209; SPITZER, supra note 111, at 7–18; Coase, supra note 29, at 12–27; Winer, supra note 173, at 218–40; Note, The Message in the Medium, supra note 120, at 1072–74. 227. See Turner I, 512 U.S. at 637 (“[T]he rationale for applying a less rigorous standard of First Amendment scrutiny to broadcast regulation, whatever its validity in the cases elaborating it, does not apply in the context of cable regulation”) (emphasis added). 228. Id. Some commentators have interpreted Turner I’s refusal to question the continuing vitality of scarcity as representing an implicit reafﬁrmation of the doctrine with respect to broadcasting. See Laurence H. Winer, The Red Lion of Cable, and Beyond?—Turner Broadcasting v. FCC, 15 CARDOZO ARTS & ENT. L.J. 1, 21–22 (1997). The structure and tone of the majority opinion suggest that the Court reserved the issue as a matter of judicial restraint and not to signal its support for the doctrine. 229 512 U S at 639 236. See, e.g., Time Warner Entm’t Co. v. FCC, 105 F.3d 723, 724 n.2 (D.C. Cir. 1997) (Williams, J., dissenting from denial of rehearing en banc); Action for Children’s Television v. FCC, 58 F.3d 654, 672–76 (D.C. Cir. 1995) (en banc) (Edwards, C.J., dissenting); Forbes v. Ark. Educ. Television Communications Network Found., 22 F.3d 1423, 1431 (8th Cir. 1994) (en banc) (McMillian, J., concurring in part & dissenting in part), rev’d, 523 U.S. 666 (1998); Ark. AFL-CIO v. FCC, 11 F.3d 1430, 1442 n.12 (8th Cir. 1993) (en banc) (noting the views of Judges Bowman, Beam, and Loken); id. at 1443 (R. Arnold, C.J., concurring in the judgment). 235. See Yoo, supra note 1, at 285–89. There is another way in which recent cases promise to prevent the Broadcast Model from expanding to other media. The Reno Court also rejected extending the Broadcast Model to the Internet in part on the ground that the Internet has never been subject to the type of government supervision and regulation that has surrounded the broadcast industry. Reno, 521 U.S. at 868–89. This argument is, in effect, the converse of the arguments reviewed in subsection II.A.2.a. Just as the presence of existing regulation makes other forms of regulation more constitution- ally palatable, the absence of regulation has the opposite effect. Because it is difﬁcult to foresee how a new medium would be subject to regulation for a sufﬁciently long time to become part of the relevant baseline, it is possible that this language from Reno will serve to restrict the technology-speciﬁc approach to the First Amendment even further. 233. Id. at 868. 234. Id. at 870. 231. Turner I, at 669–73 (Stevens, J., concurring in part & concurring in the judgment). 232. 521 U.S. 844 (1997). 237. 509 U.S. 418 (1993). 232. 521 U.S. 844 (1997). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Just as numerous courts of appeals had held earlier,225 the Court ﬂatly rejected the government’s contention that cable television should be subject to the lower level of scrutiny that applied to broadcasting. In so holding, the Court acknowledged the longstanding criticism of the scarcity doctrine leveled by scholars and courts alike.226 In the end, however, the Court did not regard a case about cable to be an appropriate occasion to reconsider the application of the doctrine to broadcasting. “Whatever [scarcity’s] validity”227 in the broadcasting context, the Court held that it did not apply to cable.228 As the Court explained: [C]able television does not suffer from inherent limitations that characterize the broadcast medium. Indeed, given the rapid advances in ﬁber optics and digital compression technology, soon there may be no practical limitation on the number of speakers who may use the cable medium. Nor is there any danger of physical interference between two cable speakers attempting to share the same channel. In light of these fundamental technological differ- ences between broadcast and cable transmission, application of the more relaxed standard of scrutiny adopted in Red Lion . . . is inapt when determin- ing the First Amendment validity of cable regulation.229 This conclusion commanded overwhelming support on the Court, garnering the votes of four of the Justices in the majority, as well as all four dissenting Justices.230 The only member of the Court not to address the issue explicitly was Justice Stevens, and his separate opinion did not indicate any disagreement 290 [Vol. 91:245 [Vol. 91:245 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL with the proposition.231 231. Turner I, at 669–73 (Stevens, J., concurring in part & concurring in the judgment). 232. 521 U.S. 844 (1997). 233. Id. at 868. 234. Id. at 870. 235. See Yoo, supra note 1, at 285–89. There is another way in which recent cases promise to prevent the Broadcast Model from expanding to other media. The Reno Court also rejected extending the Broadcast Model to the Internet in part on the ground that the Internet has never been subject to the type of government supervision and regulation that has surrounded the broadcast industry. Reno, 521 U.S. at 868–89. This argument is, in effect, the converse of the arguments reviewed in subsection II.A.2.a. Just as the presence of existing regulation makes other forms of regulation more constitution- ally palatable, the absence of regulation has the opposite effect. Because it is difﬁcult to foresee how a new medium would be subject to regulation for a sufﬁciently long time to become part of the relevant baseline, it is possible that this language from Reno will serve to restrict the technology-speciﬁc approach to the First Amendment even further. 236. See, e.g., Time Warner Entm’t Co. v. FCC, 105 F.3d 723, 724 n.2 (D.C. Cir. 1997) (Williams, J., dissenting from denial of rehearing en banc); Action for Children’s Television v. FCC, 58 F.3d 654, 672–76 (D.C. Cir. 1995) (en banc) (Edwards, C.J., dissenting); Forbes v. Ark. Educ. Television Communications Network Found., 22 F.3d 1423, 1431 (8th Cir. 1994) (en banc) (McMillian, J., concurring in part & dissenting in part), rev’d, 523 U.S. 666 (1998); Ark. AFL-CIO v. FCC, 11 F.3d 1430, 1442 n.12 (8th Cir. 1993) (en banc) (noting the views of Judges Bowman, Beam, and Loken); id. at 1443 (R. Arnold, C.J., concurring in the judgment). 237. 509 U.S. 418 (1993). with the proposition.231 The Court similarly refused to extend the scarcity doctrine to the Internet in Reno v. ACLU.232 The Court began by recognizing that scarcity was one of the “special justiﬁcations” for upholding more intrusive regulation of broadcasting than would be permissible with regard to other types of speakers.233 Because the Internet “provides relatively unlimited, low-cost capacity for communica- tion of all kinds,” it “can hardly be considered a ‘scarce’ expressive commod- ity.”234 Together, Turner I and Reno v. ACLU appear to close the door on any possible extensions of the scarcity doctrine. The emergence of a media environ- ment dominated by technological convergence, in which wire-based communica- tions and spectrum-based communications are ready substitutes, it is essentially impossible that any one communications medium can be regarded sufﬁciently scarce so as to justify receiving separate constitutional treatment.235 b. The Tacit Abandonment of Scarcity with Respect to Broadcasting. In addi- tion to precluding the possibility that the scarcity doctrine will be extended to other media, the courts have also distanced themselves from the doctrine with respect to broadcasting. A steady stream of separate opinions authored by lower court judges have continued to call for the doctrine’s abandonment.236 Even more telling has been the Supreme Court’s apparent reluctance to rely on the doctrine in its recent broadcasting decisions. When confronted with content- based restrictions on programming that, in earlier times, most likely would have been justiﬁed in terms of scarcity, the Court has assiduously avoided doing so and has instead relied on other legal principles to justify subjecting the restric- tion in question to a form of intermediate scrutiny. For example, in United States v. Edge Broadcasting Co.,237 the Court invoked 237. 509 U.S. 418 (1993). 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 291 the principles of commercial speech to apply a form of intermediate First Amendment scrutiny to uphold a federal restriction on the broadcast of lottery advertisements rather than following the more straightforward path taken by the district court of basing its decision on the scarcity doctrine.238 Similarly, in Arkansas Educational Television Commission v. Forbes,239 the Court invoked the public forum doctrine rather than the scarcity doctrine to justify applying a lesser form of First Amendment scrutiny when sustaining a public broadcaster’s decision to exclude an independent candidate for Congress from a televised debate. 243. See Greater New Orleans Broad. Ass’n v. United States, 866 F. Supp. 975, 980 (E.D. La. 1994), aff’d, 69 F.3d 1296, 1302 n.7 (5th Cir. 1995), vacated, 519 U.S. 801 (1996) (remanding for reconsidera- tion in light of 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484 (1996)). 238. Id. 239. 523 U.S. 666 (1998). 240. Id. at 673–75. 241. See Nat’l Endowment for Arts v. Finley, 524 U.S. 569, 615 (1998) (Souter, J., dissenting) (citing Forbes for the proposition that “scarcity of air time does not justify viewpoint-based exclu- sion”). 242. 527 U.S. 173 (1999). 240. Id. at 673 75. 241. See Nat’l Endowment for Arts v. Finley, 524 U.S. 569, 615 (1998) (Souter, J., dissenting) (citing Forbes for the proposition that “scarcity of air time does not justify viewpoint-based exclu- sion”). 238. Id. 239. 523 U.S. 666 (1998). 240. Id. at 673–75. 241. See Nat’l Endowment for Arts v. Finley, 524 U.S. 569, 615 (1998) (Souter, J., dissenting) (citing Forbes for the proposition that “scarcity of air time does not justify viewpoint-based exclu- sion”). 242. 527 U.S. 173 (1999). 243. See Greater New Orleans Broad. Ass’n v. United States, 866 F. Supp. 975, 980 (E.D. La. 1994), aff’d, 69 F.3d 1296, 1302 n.7 (5th Cir. 1995), vacated, 519 U.S. 801 (1996) (remanding for reconsidera- tion in light of 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484 (1996)). 240. Id. at 673–75. with the proposition.231 Even more telling was that the Court discussed its broadcasting prece- dents solely to underscore the importance of preserving the broadcasters’ editorial discretion without acknowledging the aspects of those cases supporting subjecting broadcasters to more intrusive regulation than other media speak- ers.240 Indeed, at least one Justice has subsequently construed Forbes as reject- ing scarcity as a basis for regulating broadcast content.241 Although it is possible to construe Edge Broadcasting and Forbes as exhibit- ing some discomfort with the scarcity doctrine, such a conclusion is anything but foregone. It was sufﬁcient in those cases for the Court to rely solely on one doctrinal basis for applying a less restrictive First Amendment standard. Given that the restrictions in question passed the level of scrutiny applied, the Court was under no obligation to consider whether some alternative basis might also serve to sustain it. Therefore, as a formal matter, Edge Broadcasting and Forbes are ultimately ambiguous as to the fate of the scarcity doctrine. The same cannot be said about the Court’s most recent broadcasting case, Greater New Orleans Broadcasting Ass’n v. United States,242 in which the Court applied commercial speech doctrine to invalidate certain federal restric- tions on casino advertising. After holding that the restrictions in question did not satisfy the less restrictive standard imposed by the commercial speech doctrine, the presumption of constitutionality enjoyed by all federal legislation put the Court under an obligation to consider whether the restriction in question might be upheld on other grounds. Had the scarcity doctrine retained any vitality, it would have been appropriate for the Court to apply it to the facts of this case. Indeed, both the district court and the court of appeals relied in part on the scarcity doctrine when initially considering the case.243 The Court’s failure to mention the scarcity doctrine in Greater New Orleans is thus considerably 292 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 more telling than its failure to do so in Edge Broadcasting and Forbes. To the contrary, the Court’s silence in Greater New Orleans is deafening. The Court’s apparent reluctance to rely on the scarcity doctrine in these cases thus raises the distinct possibility that, having stopped short of overruling the scarcity doctrine outright, the Court is nonetheless distancing itself from it. Given its conceptual and empirical inﬁrmities, I would certainly welcome any indication that the doctrine is losing its vitality. 245. See Sinclair Broad. Group, Inc. v. FCC, 284 F.3d 148, 161–62, 167–69 (D.C. Cir. 2002); Fox Television Stations, Inc. v. FCC, 280 F.3d 1027, 1045–46 (D.C. Cir. 2002); Radio-Television News Dirs. Ass’n v. FCC, 184 F.3d 872, 877 n.3 (D.C. Cir. 1999); Tribune Co. v. FCC, 133 F.3d 61, 69 (D.C. Cir. 1998); Time Warner Entm’t Co. v. FCC, 93 F.3d 957, 975–77 (D.C. Cir. 1996); Branch v. FCC, 824 F.2d 37, 49–50 (D.C. Cir. 1987); Telecomms. Research & Action Ctr. v. FCC, 801 F.2d 501, 508–09 (D.C. Cir. 1986). 244. See Rodriguez de Quijas v. Shearson/American Express, Inc., 490 U.S. 477, 484 (1989) (“If a precedent of this Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the Court of Appeals should follow the case which directly controls, leaving this Court the prerogative of overruling its decisions.”); Agostini v. Felton, 521 U.S. 203, 237 (1997) (“We do not acknowledge, and we do not hold, that other courts should conclude our more recent cases have, by implication, overruled an earlier precedent.”). 246. See, e.g., Jonathan S. Adelstein, The Last DJ?: Finding a Voice on Media Ownership, Remarks at the Future of Music Coalition Policy Summit (Jan. 6, 2003), available at http://www.fcc.gov/Speeches/ Adelstein/2003/spjsa301.html (calling for the preservation and extension of regulations based on Red Lion). 247. Policies & Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10731–32 ¶¶ 155–156 (1996). with the proposition.231 That said, although the Supreme Court can avoid the illogic of the scarcity doctrine simply by ignoring it, lower courts cannot.244 Even judges who are well aware of the doctrine’s shortcom- ings acknowledge that they remain bound by it until the Supreme Court overrules it.245 Until that happens, the lower courts may have no choice but to apply the scarcity doctrine to all new media that employ the electromagnetic spectrum as their primary means of transmission. In addition, regulatory authori- ties remain free to call for the type of intrusive regulation associated with Red Lion without having to acknowledge the scarcity doctrine’s inherent ﬂaws.246 It is thus clear that, until the Court ﬁnally inters the doctrine, scarcity will continue to exert an unfortunate distorting effect on lower court and regulatory decisionmaking. p p y g 252. Paciﬁca, 438 U.S. at 748. If anything, the Court’s emphasis on the home in Paciﬁca seems singularly misplaced considering that the complainant encountered the program at issue while driving in his car. See id. at 730. 250. FCC v. Paciﬁca Found., 438 U.S. 726, 748 (1978) (“We have long recognized that . . . of all forms of communication, it is broadcasting that has received the most limited First Amendment protection.”). III. PERVASIVENESS AND ACCESSIBILITY: NEW RATIONALES FROM PACIFICA Perhaps sensing the growing weakness of the scarcity doctrine, in recent years, the FCC has increasingly turned to the principles announced in Paciﬁca as the primary support for the constitutionality of the Broadcast Model. For example, the FCC invoked Paciﬁca as a basis for sustaining the constitutional- ity of requiring broadcasters to provide three hours of children’s programming each week.247 Even more telling is the FCC’s abandonment of any attempt to use the scarcity doctrine to justify its indecency restrictions and its decision to 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 293 rely solely on the rationales announced in Paciﬁca.248 The recent upsurge in Paciﬁca’s importance makes it appropriate to evaluate the extent to which the rationales announced in that decision can justify according less than full First Amendment protection to emerging media. This Part begins by critiquing the reasoning of the Paciﬁca opinion, both in terms of logical coherence and consistency with the precedents that existed at the time. It will then trace subsequent judicial discussions of Paciﬁca. Finally, this Part will analyze the impact that technological change has had on Paciﬁca. In sum, each of these considerations provides ample reason to question Paciﬁca’s vitality as a precedent. 251. See supra note 85 and accompanying text. y y 249. See, e.g., KRATTENMAKER & POWE, supra note 72, at 196–202; POOL, supra note 51, at 134; POWE, supra note 51, at 209–15; STEVEN H. SHIFFRIN, THE FIRST AMENDMENT, DEMOCRACY, AND ROMANCE 80–82 (1990); SPITZER, supra note 111, at 119–30; C. Edwin Baker, The Evening Hours During Paciﬁca Standard Time, 3 VILL. SPORTS & ENT. L.J. 45, 45 (1996); Thomas G. Krattenmaker & L.A Powe, Jr., Televised Violence: First Amendment Principles and Social Science Theory, 64 VA. L. REV 1123, 1229 (1978); Levi, supra note 72, at 139–40; Glen O. Robinson, The Electronic First Amendment: An Essay for the New Age, 47 DUKE L.J. 899, 946–51 (1998); Frederick Schauer, Categories and the First Amendment: A Play in Three Acts, 34 VAND. L. REV. 265, 287–96 (1981); Laurence H. Winer, The Signal Cable Sends, Part II—Interference from the Indecency Cases?, 55 FORDHAM L. REV. 459, 493–502 (1987). 248. Paciﬁca Found., Inc., 2 F.C.C.R. 2698, 2699 ¶¶ 10–11 (1987) (mem. op. & order); see also Action for Children’s Television v. FCC, 58 F.3d 654, 664–69 (D.C. Cir. 1995) (en banc); Indus. Guidance on Comm’n’s Case Law Interpreting 18 U.S.C. § 1464 & Enforcement Policies Regarding Broad. Indecency, Policy Statement, 16 F.C.C.R. 7999, 8000 ¶ 4 (2001). p ) 251. See supra note 85 and accompanying text. A. THE THEORETICAL CRITIQUE OF PACIFICA Critical analysis of the Court’s Paciﬁca opinion reveals that its rationales are deeply ﬂawed. Although the criticism of Paciﬁca is voluminous,249 for the purposes of this Article, it sufﬁces to focus on the extent to which the Paciﬁca opinion can justify according broadcasting a lesser degree of First Amendment protection. Unfortunately, the opinion itself provides little guidance in this regard. Although the tone of Paciﬁca suggests that the Court was simply relying on the established justiﬁcations for applying a lower First Amendment standard to broadcasting,250 the rationales upon which the Court relied were actually quite novel. Unfortunately, the Court did not offer much in the way of articulat- ing or defending these new rationales, and its meager explanations raised more questions than they answered. As noted above,251 the Court’s ﬁrst rationale for upholding the restriction on indecent speech was that broadcasting is “uniquely pervasive,” in that it acts as an “intruder” that “confronts the citizen . . . in the privacy of the home.”252 It is 294 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 hard to see how these considerations distinguish broadcasting from other media. Other media, such as books and the mail, are similarly pervasive and enter the home as easily, and yet the Court had previously struck down attempts to ban offensive speech transmitted over those media.253 In addition, it is far from clear that broadcasting can properly be regarded as an intruder. Viewers and listeners must purchase radio and television sets,254 and those sets must, of course, be turned on for people to be subject to such “intrusion.” As Justice Brennan noted, “switching on and listening to communications transmitted over the public airwaves” represents an afﬁrmative decision “to take part . . . in an ongoing public discourse.”255 The Court also attempted to explain the supposed pervasiveness of broadcast- ing by arguing that “prior warnings cannot completely protect the listener or viewer from unexpected program content” and reasoning that “[t]o say that one may avoid further offense by turning off the radio when he hears indecent language is like saying that the remedy for an assault is to run away after the ﬁrst blow.”256 The Court had never previously regarded listeners’ and viewers’ inability to obtain complete protection against exposure to unwanted speech as constitutionally signiﬁcant. 255. Paciﬁca, 438 U.S. at 764–65 (Brennan, J., dissenting). 253. See Butler v. Michigan, 352 U.S. 380 (1957) (books); Carey v. Population Servs. Int’l, 431 U.S. 678 (1977) (mail); see also KRATTENMAKER & POWE, supra note 72, at 220; POWE, supra note 51, at 210; SPITZER, supra note 111, at 120; Harry T. Edwards & Mitchell N. Berman, Regulating Violence on Television, 89 NW. U. L. REV. 1487, 1496 (1995). 257. Erzoznik v. City of Jacksonville, 422 U.S. 205, 210–11 (1975) (quoting Cohen v. California, 403 U.S. 15, 21 (1971)). 254. As Thomas Krattenmaker and Scot Powe have noted, “radios and televisions are not forced upon citizens, but in fact are considered to be among the most valued household purchases. Intruders they are not.” KRATTENMAKER & POWE, supra note 72, at 220. Powe’s initial statement of this criticism was considerably more sardonic: 256. Id. at 748–49. POWE, supra note 51, at 210; see also SPITZER, supra note 111, at 120 (criticizing the notion that radio is “uniquely persuasive” because printed publications must also be ordered and purchased before they are delivered to one’s home). A. THE THEORETICAL CRITIQUE OF PACIFICA To the contrary, prior to Paciﬁca the Court had consistently held that when viewers encounter material they ﬁnd offensive, “the burden normally falls upon the viewer to ‘avoid further bombardment of their sensibilities simply by averting their eyes.’”257 In the words of one lower federal court that the Supreme Court later cited with approval, the “short, though regular, journey from mail box to trash can . . . is an acceptable burden, 253. See Butler v. Michigan, 352 U.S. 380 (1957) (books); Carey v. Population Servs. Int’l, 431 U.S. 678 (1977) (mail); see also KRATTENMAKER & POWE, supra note 72, at 220; POWE, supra note 51, at 210; SPITZER, supra note 111, at 120; Harry T. Edwards & Mitchell N. Berman, Regulating Violence on Television, 89 NW. U. L. REV. 1487, 1496 (1995). ( ) 254. As Thomas Krattenmaker and Scot Powe have noted, “radios and televisions are not forced upon citizens, but in fact are considered to be among the most valued household purchases. Intruders they are not.” KRATTENMAKER & POWE, supra note 72, at 220. Powe’s initial statement of this criticism was considerably more sardonic: Whatever else the Court means, it is not true that the FBI or CIA breaks into millions of American homes to deposit the latest Sony radios in bedrooms and living areas. To the best of my knowledge, Americans bring radios and television sets into their homes because they desire them. . . . If homeowners truly believed that radio or television was an intruder, I would expect to see sets out on the streets for garbage collection. Instead, when I read my morning paper I see numbers of full-page ads for these very appliances, suggesting that the merchants believe, contrary to what the Court might think, that Americans desire radios and televisions. POWE, supra note 51, at 210; see also SPITZER, supra note 111, at 120 (criticizing the notion that radio is “uniquely persuasive” because printed publications must also be ordered and purchased before they are delivered to one’s home). 255. Paciﬁca, 438 U.S. at 764–65 (Brennan, J., dissenting). 258. Lamont v. Comm’r of Motor Vehicles, 269 F. Supp. 880, 883 (S.D.N.Y. 1967), aff’d, 386 F.2d 449 (2d Cir. 1967), quoted with approval by Bolger v. Youngs Drug Prods. Co., 463 U.S. 60, 72 (1983). 259. As Justice Brennan noted in dissent: Whatever minimal discomfort suffered by a listener who inadvertently tunes into a program he ﬁnds offensive during the brief interval before he can simply extend his arm and switch stations or ﬂick the “off” button, it is surely worth the candle to preserve the broadcaster’s right to send, and the right of those interested to receive, a message entitled to full First Amendment protection. ( ) 262. Id. at 736. In support of this proposition, the Court quoted language from a previous case emphasizing the importance of “leaving ‘with the homeowner himself’ the power to decide ‘whether distributors of literature may lawfully call at a home.’” Id. (quoting Martin v. City of Struthers, 319 U.S. 141, 148 (1943)). Paciﬁca, 438 U.S. at 765–66 (Brennan, J., dissenting). In the Court of Appeals decision that led to Paciﬁca, Chief Judge Bazelon similarly observed that “having elected to receive public air waves, the scanner who stumbles onto an offensive program is in the same position as the unsuspecting passers-by in Cohen and Erzoznik; he can avert his attention by changing channels or turning off the set.” Paciﬁca Found. v. FCC, 556 F.2d 9, 26 (D.C. Cir. 1977) (Bazelon, C.J., concurring), rev’d, 438 U.S. 726 (1978). 260. Packer Corp. v. Utah, 285 U.S. 105, 110 (1932); accord Lehman v. City of Shaker Heights, 481 U.S. 298, 302 (1974); see also Schauer, supra note 249, at 294 (“Turning off a radio is much easier than averting your eyes from someone who is in the same room. Just try it sometime.”). 261. 397 U.S. 728 (1970). A. THE THEORETICAL CRITIQUE OF PACIFICA 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 295 at least so far as the Constitution is concerned.”258 This reasoning would seem to apply a fortiori to broadcast speech.259 In upholding limitations on potentially offensive advertising on streetcars, the Court explicitly distinguished broadcast- ing on the grounds that unlike “the billboard or street card placard” at issue in those cases, “[t]he radio can be turned off.”260 This conclusion draws further support from the reasoning of the Court’s opinion in Rowan v. United States Post Ofﬁce Department.261 In upholding a statute allowing people to have their names removed from mailing lists of those sending erotically arousing or sexually provocative materials, the Rowan Court emphasized the importance of making each “householder the exclusive and ﬁnal judge of what will cross his threshold.”262 Consequently, a statute that enhanced each homeowners’ ability to control what speech entered their homes was no This conclusion draws further support from the reasoning of the Court’s opinion in Rowan v. 258. Lamont v. Comm’r of Motor Vehicles, 269 F. Supp. 880, 883 (S.D.N.Y. 1967), aff’d, 386 F.2d 449 (2d Cir. 1967), quoted with approval by Bolger v. Youngs Drug Prods. Co., 463 U.S. 60, 72 (1983). 259. As Justice Brennan noted in dissent: Whatever minimal discomfort suffered by a listener who inadvertently tunes into a program he ﬁnds offensive during the brief interval before he can simply extend his arm and switch stations or ﬂick the “off” button it is surely worth the candle to preserve the broadcaster’s 263. Id. 264. Id. at 737. Paciﬁca, 438 U.S. at 765–66 (Brennan, J., dissenting). In the Court of Appeals decision that led to Paciﬁca, Chief Judge Bazelon similarly observed that “having elected to receive public air waves, the scanner who stumbles onto an offensive program is in the same position as the unsuspecting passers-by in Cohen and Erzoznik; he can avert his attention by changing channels or turning off the set.” Paciﬁca Found. v. FCC, 556 F.2d 9, 26 (D.C. Cir. 1977) (Bazelon, C.J., concurring), rev’d, 438 U.S. 726 (1978). 260. Packer Corp. v. Utah, 285 U.S. 105, 110 (1932); accord Lehman v. City of Shaker Heights, 481 U.S. 298, 302 (1974); see also Schauer, supra note 249, at 294 (“Turning off a radio is much easier than ti f h i i th J t t it ti ”) A. THE THEORETICAL CRITIQUE OF PACIFICA United States Post Ofﬁce Department.261 In upholding a statute allowing people to have their names removed from mailing lists of those sending erotically arousing or sexually provocative materials, the Rowan Court emphasized the importance of making each “householder the exclusive and ﬁnal judge of what will cross his threshold.”262 Consequently, a statute that enhanced each homeowners’ ability to control what speech entered their homes was no more problematic than “a radio or television viewer twist[ing] the dial to cut off an offensive or boring communication and thus bar its entering his home.”263 The Court limited its holding by emphasizing that cutting off potentially offensive speech was permissible only to the extent that “the mailer’s right to communicate is circumscribed only by an afﬁrmative act of the addressee giving notice that he wishes no further mailings from that mailer.”264 The negative implication from this reasoning is clear: Although the possibility that some people may ﬁnd particular speech offensive may justify upholding mea- sures that enhance individuals’ ability to screen out potentially offensive mate- rial on a targeted basis, the potential for offensiveness does not justify the total suppression of the potentially offensive speech, which would also deprive each homeowners’ ability to control what speech entered their homes was no more problematic than “a radio or television viewer twist[ing] the dial to cut off an offensive or boring communication and thus bar its entering his home.”263 The Court limited its holding by emphasizing that cutting off potentially offensive speech was permissible only to the extent that “the mailer’s right to communicate is circumscribed only by an afﬁrmative act of the addressee giving notice that he wishes no further mailings from that mailer.”264 The negative implication from this reasoning is clear: Although the possibility that some people may ﬁnd particular speech offensive may justify upholding mea- sures that enhance individuals’ ability to screen out potentially offensive mate- rial on a targeted basis, the potential for offensiveness does not justify the total suppression of the potentially offensive speech, which would also deprive Paciﬁca, 438 U.S. at 765–66 (Brennan, J., dissenting). 258. Lamont v. Comm’r of Motor Vehicles, 269 F. Supp. 880, 883 (S.D.N.Y. 1967), aff’d, 386 F.2d 449 (2d Cir. 1967), quoted with approval by Bolger v. Youngs Drug Prods. Co., 463 U.S. 60, 72 (1983). 259. As Justice Brennan noted in dissent: 263. Id. 264. Id. at 737. 269. The facts underlying Paciﬁca make it a somewhat inapposite case for invoking the accessibility of broadcast speech to children as a rationale for upholding the regulation. Some parts of the Court’s opinion leave the impression that the child exposed to Carlin’s monologue was a ﬁrst grader. See id. (expressing particular solicitude for “those [children] too young to read”). In fact, the child at issue was actually ﬁfteen years old. See POWE, supra note 51, at 186; Winer, supra note 249, at 490 n.171. This was well above the age that the FCC thought at the time merited protection from indecent program- ming. See Action for Children’s Television v. FCC, 852 F.2d 1332, 1341–42 (D.C. Cir. 1988); KRATTENMAKER & POWE, supra note 72, at 110 & n.32. In addition, the program in question was broadcast at two o’clock when broadcasters could reasonably expect most children to be in school. See Paciﬁca, 438 U.S. at 729. 266. Paciﬁca, 438 U.S. at 766 (Brennan, J., dissenting); see also Paciﬁca Found. v. FCC, 556 F.2d 9, 27 (D.C. Cir. 1977) (Bazelon, C.J., concurring) (distinguishing a statute that enabled a person to require removal of his name from all future mailing lists of a solicitor because the statute “empower[ed] the homeowner to determine what mail he will not receive, [and] avoided the constitutional problems involved in vesting the power to make any discretionary evaluation of the material in a government ofﬁcial”). 265. See Carey v. Population Servs. Int’l, 431 U.S. 678, 701–02 (1977) (noting that the possibility that some people might ﬁnd mailed contraceptive advertisements offensive and embarrassing “d[id] not justify the total suppression of advertising concerning contraceptives”). Compare Ginsberg v. New York, 390 U.S. 629, 641 (1968) (upholding statute barring sale of indecent materials to minors), with Butler v. Michigan, 352 U.S. 380, 383–84 (1957) (overturning blanket ban on indecent materials affecting adults and children alike). The Court would later conﬁrm this intuition in Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 71–72 (1983). See infra notes 286–88 and accompanying text. 267. Rowan, 397 U.S. at 737. 273. As Krattenmaker and Powe note, “To hold broadcasting . . . to the level of hypothetical ﬁve-year olds who may be listening is indeed to invite the risk that the adult population may be reduced in its thinking to ideas ﬁt for a child.” Krattenmaker & Powe, supra note 249, at 1280. 268. Paciﬁca, 438 U.S. at 749. g 274. Paciﬁca, 438 U.S. at 750 n.28. 270. See POWE, supra note 51, at 209; Krattenmaker & Powe, supra note 249, at 1233. 271 352 U S 380 (1957) 272. Id. at 383. 271. 352 U.S. 380 (1957). 271. 352 U.S. 380 (1957). 272. Id. at 383. 270. See POWE, supra note 51, at 209; Krattenmaker & Powe, supra note 249, at 1233. 271. 352 U.S. 380 (1957). 275. Marcy Strauss, Redeﬁning the Captive Audience Doctrine, 19 HASTINGS CONST. L.Q. 85, 93 (1991). 282. Paciﬁca, 438 U.S. at 761 (Powell, J., joined by Blackmun, J., concurring in part & concurring in the judgment). ( ) (p y p ) 280. See Paciﬁca, 438 U.S. at 729 (noting that only Chief Justice Burger and Justice Rehnquist joined the part of the opinion advancing the view that indecency represented low-value speech); id. at 761–62 (Powell, J., joined by Blackmun, J., concurring in part & concurring in the judgment) (rejecting Justice Stevens’s low-value speech rationale and relying solely on the “unique characteristics of the broadcast media”). 276. See Krattenmaker & Powe, supra note 249, at 1239. 278. 438 U.S. at 743 (plurality opinion). Indeed, Justice Stevens equated indecent speech with obscene speech when he opined that “[t]hese words offend for the same reasons that obscenity offends.” Id. at 746. 281. Cohen v. California, 403 U.S. 15, 21 (1971). 277. See KRATTENMAKER & POWE, supra note 72, at 198–201. 279. That Justice Stevens ﬁrst advanced this position in a decision involving adult movie theaters further underscores the assertion that he did not see it as applying only to broadcasting. See Young v. Am. Mini Theatres, Inc., 427 U.S. 50, 70 (1976) (plurality opinion). A. THE THEORETICAL CRITIQUE OF PACIFICA In the Court of Appeals decision that led to Paciﬁca, Chief Judge Bazelon similarly observed that “having elected to receive public air waves, the scanner who stumbles onto an offensive program is in the same position as the unsuspecting passers-by in Cohen and Erzoznik; he can avert his attention by changing channels or turning off the set.” Paciﬁca Found. v. FCC, 556 F.2d 9, 26 (D.C. Cir. 1977) (Bazelon, C.J., concurring), rev’d, 438 U.S. 726 (1978). 260. Packer Corp. v. Utah, 285 U.S. 105, 110 (1932); accord Lehman v. City of Shaker Heights, 481 U.S. 298, 302 (1974); see also Schauer, supra note 249, at 294 (“Turning off a radio is much easier than [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 296 access to those who wish to view such material.265 As Justice Brennan’s dissent in Paciﬁca pointed out, the effect of such a blanket governmental prohibition is to take the discretion away from homeowners and to place it instead in the hands of the government.266 In addition, the clear suggestion of the Court’s reasoning in Rowan is that the proper response of a person encountering unwanted broadcast speech is simply to “twist the dial to cut off [the] offensive or boring communication.”267 g The other rationale proffered by the Paciﬁca majority—that broadcasting is “uniquely accessible to children”268—fares little better.269 It is hard to see how broadcasting is any more accessible to children than newspapers, books, or the mail.270 Moreover, the Court had cautioned in Butler v. Michigan271 that in attempting to protect children, the government could not “reduce the adult population to reading only what is ﬁt for children.”272 And yet, banning indecent speech because children might be listening had precisely that effect.273 The Court’s observation that adults interested in hearing such materials could purchase recordings or go to live performances274 has been criticized as “disin- genuous,” in that the alternatives offered by the Court are far from comparable 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 275. Marcy Strauss, Redeﬁning the Captive Audience Doctrine, 19 HASTINGS CONST. L.Q. 85, 93 (1991). 276. See Krattenmaker & Powe, supra note 249, at 1239. 277. See KRATTENMAKER & POWE, supra note 72, at 198–201. 278. 438 U.S. at 743 (plurality opinion). Indeed, Justice Stevens equated indecent speech with obscene speech when he opined that “[t]hese words offend for the same reasons that obscenity offends.” Id. at 746. 279. That Justice Stevens ﬁrst advanced this position in a decision involving adult movie theaters further underscores the assertion that he did not see it as applying only to broadcasting. See Young v. Am. Mini Theatres, Inc., 427 U.S. 50, 70 (1976) (plurality opinion). 280. See Paciﬁca, 438 U.S. at 729 (noting that only Chief Justice Burger and Justice Rehnquist joined the part of the opinion advancing the view that indecency represented low-value speech); id. at 761–62 (Powell, J., joined by Blackmun, J., concurring in part & concurring in the judgment) (rejecting Justice Stevens’s low-value speech rationale and relying solely on the “unique characteristics of the broadcast media”). 281. Cohen v. California, 403 U.S. 15, 21 (1971). 282 Paciﬁca 438 U S at 761 (Powell J joined by Blackmun J concurring in part & concurring 278. 438 U.S. at 743 (plurality opinion). Indeed, Justice Stevens equated indecent speech with obscene speech when he opined that “[t]hese words offend for the same reasons that obscenity offends.” Id. at 746. 279. That Justice Stevens ﬁrst advanced this position in a decision involving adult movie theaters further underscores the assertion that he did not see it as applying only to broadcasting. See Young v. Am. Mini Theatres, Inc., 427 U.S. 50, 70 (1976) (plurality opinion). 280. See Paciﬁca, 438 U.S. at 729 (noting that only Chief Justice Burger and Justice Rehnquist joined the part of the opinion advancing the view that indecency represented low-value speech); id. at 761–62 (Powell, J., joined by Blackmun, J., concurring in part & concurring in the judgment) (rejecting Justice Stevens’s low-value speech rationale and relying solely on the “unique characteristics of the broadcast media”). 281. Cohen v. California, 403 U.S. 15, 21 (1971). 282. Paciﬁca, 438 U.S. at 761 (Powell, J., joined by Blackmun, J., concurring in part & concurring in the judgment). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 297 in terms of affordability or ease of access.275 In addition, such reasoning would justify upholding the restrictions on the sale of indecent magazines struck down in Butler, as well as virtually any other technology-speciﬁc restriction on indecent speech.276 In sum, none of Paciﬁca’s proffered rationales can sensibly be read as being limited to broadcasting. The key to unraveling this mystery lies in an earlier portion of the Paciﬁca opinion.277 Justice Stevens began the constitutional discussion of his Paciﬁca opinion by stating that indecency constituted low-value speech that “surely lie[s] at the periphery of First Amendment concern” and, as a result, was not worthy of constitutional protection.278 What is most notable about this rationale for the purposes of this Article is that it turned entirely on the content of the speech and did not depend in any way on the medium by which the speech was conveyed.279 Unfortunately for Justice Stevens, two of the ﬁve Justices that comprised his majority refused to join this portion of his opinion.280 Employing words that echoed Justice Harlan’s famous observation that “it is . . . often true that one man’s vulgarity is another’s lyric,”281 Justice Powell, joined by Justice Blackmun, instead concluded that determining the relative value of particular speech “is a judgment for each person to make, not one for the judges to impose upon him.”282 It was only after Justice Stevens’s initial attempt to obtain ﬁve votes failed that he turned to the alternative rationales discussed above. As a result, Justice Stevens’s opinion reads like a non sequitur, ﬁrst suggesting that indecent speech should not receive the full protection of the First Amendment regardless of the medium on which it is conveyed, but then immediately seeming to contradict itself by grounding the decision on the principle that broadcasting is different. Given Justice Stevens’s ambivalence about the media-speciﬁc nature of his opinion, it should come as no surprise that the rationales he offered to limit his holding to broadcasting were somewhat half-hearted. To say that the opinion is explicable in this manner, however, is not to say that it is justiﬁable. Read 298 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 critically, Paciﬁca falls short of providing an alternative basis for according to broadcasting a lesser degree of First Amendment protection. 283. Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 75 (1983). 284. Sable Communications of Cal., Inc. v. FCC, 492 U.S. 115, 124 (1989). 285. Reno v. ACLU, 521 U.S. 844, 886 (1997). 286. Id. at 870, 875–76; Sable, 492 U.S. at 128; Bolger, 463 U.S. at 74. 287. Bolger, 463 U.S. at 74. 288. Id. at 72; see also Consol. Edison Co. of N.Y., Inc. v. Pub. Serv. Comm’n, 447 U.S. 530, 542 (1980) (reasoning that recipients of unwanted mailings may “‘effectively avoid further bombardment of their sensibilities simply by averting their eyes’” (quoting Cohen v. California, 403 U.S. 15, 21 (1971))). 289. 518 U.S. 727 (1996). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Lastly and most importantly for the purposes of this Article, even if read as broadly as possible, Paciﬁca does not justify the full range of regulation associated with the Broadcast Model. At most, concerns about pervasiveness and accessibility to children would justify the imposition of a narrow range of negative content restrictions, likely limited to such areas as indecency and violence. The Paciﬁca rationales would not justify broadcast licensing, negative content restrictions in other areas, and afﬁrmative programming obligations that constitute the other key features of the Broadcast Model. As a result, it is conceptually ill-suited to serve as the foundation for applying an alternative justiﬁcation for the type of intrusive regulation that characterizes the Broadcast Model of regulation. 290. Pub. L. No. 102-385, § 10(a)–(c), 106 Stat. 1460, 1486. Subsection (a) gave cable operators the discretion to refuse to carry indecent programming on leased access channels. 47 U.S.C. § 531(a) (2000). Subsection (c) gave cable operators similar authority over public, educational, and governmen- tal access channels. § 10(c), 106 Stat. at 1486. Subsection (b) also required cable operators to segregate any indecent programming carried on leased access channels on a single channel and to block its transmission unless the viewer indicates otherwise. § 10(b), 106 Stat. at 1486. The decision holding subsection (b) unconstitutional was the only part of the Court’s decision that commanded a clear majority of the Court. Denver Area Educ. Telecomms. Consortium, 518 U.S. at 753–60; id. at 779 (O’Connor, concurring in part & dissenting in part) (joining Part III); id. at 781 (Kennedy, J., joined by Ginsburg, J., concurring in part, concurring in the judgment in part, & dissenting in part) (same). 1. The Refusal to Extend Paciﬁca to Other Media Given the analytical problems with the rationales underlying Paciﬁca, the Court has unsurprisingly and repeatedly refused to extend it to other media, including mailed contraceptive advertisements,283 telephony,284 and the Inter- net.285 The Court so held even though each of those media appears to be as pervasive and accessible to children as broadcasting. Instead, these opinions noted Paciﬁca’s acknowledgement of the narrowness of its holding and empha- sized that, if allowed to stand, the restrictions at issue would limit adults to what was appropriate for children.286 As Justice Marshall eloquently quipped, “[t]he level of discourse reaching a mailbox simply cannot be limited to that which would be suitable for a sandbox.”287 It was more appropriate to expect recipi- ents of unwanted material simply to discard it.288 Thus, as was the case with the scarcity doctrine, the Court’s reluctance to extend Paciﬁca to other media arguably signaled its unease with it. The single deviation from the Court’s refusal to extend Paciﬁca to other media appeared in Justice Breyer’s plurality opinion in Denver Area Educa- tional Telecommunications Consortium, Inc. v. FCC,289 which resolved a First Amendment challenge to three provisions of the Cable Television Consumer Protection and Competition Act of 1992 that restricted the transmission of 2003] 299 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT indecent speech.290 The plurality argued that the Court’s decision to hold the scarcity doctrine inapplicable to cable in Turner I did not foreclose the possibil- ity that Paciﬁca might justify holding cable television to a lower standard of First Amendment protection.291 The plurality proceeded to contradict a long line of lower court decisions holding Paciﬁca inapplicable to cable292 and concluded that all of the considerations that underlay Paciﬁca—pervasiveness, invasion of the home, ineffectiveness of warnings, accessibility to children—applied with equal force to cable television.293 Thus, in applying Paciﬁca to cable, the Denver plurality raised questions about the correctness of the statements in Paciﬁca emphasizing the narrowness of its holding294 and asserting that it was limited to broadcasting.295 Indeed, the plurality implied that Paciﬁca might apply to all media when it indicated that it regarded the question “whether . . . 298. Id. at 742 (plurality opinion) (“[W]e believe it unwise and unnecessary to pick . . . one speciﬁc set of words now”); see also id. at 768 (Stevens, J., concurring) (ﬁnding it “unwise to take a categorical approach to the resolution of novel First Amendment questions arising in an industry as dynamic as this”); id. at 775 (Souter, J., concurring) (acknowledging the plurality’s “unwillingness to announce a deﬁnitive categorical analysis in this case”). 292. See Cruz v. Ferre, 755 F.2d 1415, 1420–21 (11th Cir. 1985); Cmty. Television of Utah, Inc. v. Wilkinson, 611 F. Supp. 1099, 1109–16 (D. Utah 1985), aff’d sub nom. Jones v. Wilkinson, 800 F.2d 989, 991 (10th Cir. 1986), aff’d mem., 480 U.S. 926 (1987); Cmty. Television of Utah, Inc. v. Roy City, 555 F. Supp. 1164, 1167–69 (D. Utah 1982); see also Action for Children’s Television v. FCC, 58 F.3d 654, 660 (D.C. Cir. 1995) (en banc) (suggesting in dicta that Paciﬁca was inapplicable to cable). 293 518 U S 744 45 ( l li i i ) 291. Denver Area Educ. Telecomms. Consortium, 518 U.S. at 748 (“The Court’s distinction in Turner [I]. . . between cable and broadcast television, relied on the inapplicability of the spectrum scarcity problem to cable. While the distinction was relevant in Turner [I] to the justiﬁcation for structural regulations at issue there . . . , it has little to do with a case that involves the effects of television viewing on children.”). 297. See id. at 781, 784–87 (Kennedy, J., joined by Ginsburg, J., concurring in part, concurring in the judgment in part, & dissenting in part); id. at 817–18 (Thomas, J., joined by Rehnquist, C.J., and Scalia, J., concurring in the judgment in part & dissenting in part). 295. Id. at 748; see also id. at 762 (Powell, J., concurring in part & concurring in the judgment). 296. Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727, 755 (1996) (plurality opinion). 294. Paciﬁca, 438 U.S. at 750. 295. Id. at 748; see also id. at 762 (Powell, J., concurring in part & concurring in the judgment). 296. Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727, 755 (1996) (plurality 293. 518 U.S. at 744–45 (plurality opinion). 295. Id. at 748; see also id. at 762 (Powell, J., concurring in part & concurring in the judgment). 95. Id. at 748; see also id. at 762 (Powell, J., concurrin 299. See id. at 733 (stating that regulations must be “appropriately tailored to achieve [a] basic, legitimate objective”); id. at 741 (noting that regulations must be “address extraordinary problems” and be “appropriately tailored to resolve those problems without imposing an unnecessarily great restriction on speech”); id. at 743 (noting that regulations must “properly address[ ] an extremely important problem, without imposing, in light of the relevant interests, an unnecessarily great restriction on speech”); id. (stating that regulations must be a “sufﬁciently tailored response to an extraordinarily important problem”). Equally puzzling was the Denver plurality’s willingness to strike down a parallel provision governing public access channels. For the problems with this portion of the plurality’s analysis, see supra notes 124–29 and accompanying text. 300. 529 U.S. 803 (2000). 301. Telecommunications Act of 1996, Pub. L. No. 104-104, § 505, 110 Stat. 56, 136. 302. Playboy, 529 U.S. at 811–14. 303. Id. at 813 (quoting Cohen v. California, 403 U.S. 15, 21 (1971) and citing Erzoznik v. City of Jacksonville, 422 U.S. 205, 210–11 (1975)). 304. I do not mean to suggest that the Court did not place any emphasis on the extent to which these two media intruded into the home. In fact, the Court speciﬁcally noted that the case turned on the difference from broadcasting that, “[c]able systems have the capacity to block unwanted channels on a household-by-household basis.” Id. at 815. Nor do I mean to suggest that accessibility to children played no part in the Court’s analysis. On the contrary, the Court regarded the feasibility of using technology to restrict minors’ access to indecent material to be an important consideration. Id. at 814. A close reading of the opinion reveals, however, that the Court believed that these considerations were only relevant in evaluating whether the statute in question represented the least restrictive means available. They played no role in determining the level of First Amendment protection that cable received. 300. 529 U.S. 803 (2000). 1. The Refusal to Extend Paciﬁca to Other Media Paciﬁca does, or does not, impose some lesser standard of review where indecent speech is at issue” as still being open.296 The plurality, however, ignored the sharp criticism leveled by ﬁve of the Court’s members297 and refused to resolve which constitutional standard would be applied.298 In doing so, the Court articulated an assortment of standards, all of which appeared to be 292. See Cruz v. Ferre, 755 F.2d 1415, 1420–21 (11th Cir. 1985); Cmty. Television of Utah, Inc. v. Wilkinson, 611 F. Supp. 1099, 1109–16 (D. Utah 1985), aff’d sub nom. Jones v. Wilkinson, 800 F.2d 989, 991 (10th Cir. 1986), aff’d mem., 480 U.S. 926 (1987); Cmty. Television of Utah, Inc. v. Roy City, 555 F. Supp. 1164, 1167–69 (D. Utah 1982); see also Action for Children’s Television v. FCC, 58 F.3d 654, 660 (D.C. Cir. 1995) (en banc) (suggesting in dicta that Paciﬁca was inapplicable to cable). 293 518 U S 744 45 ( l li i i ) 300 [Vol. 91:245 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL less stringent than strict scrutiny.299 301. Telecommunications Act of 1996, Pub. L. No. 104-104, § 505, 110 Stat. 56, 136. , , ( )) 304. I do not mean to suggest that the Court did not place any emphasis on the extent to which these two media intruded into the home. In fact, the Court speciﬁcally noted that the case turned on the difference from broadcasting that, “[c]able systems have the capacity to block unwanted channels on a household-by-household basis.” Id. at 815. Nor do I mean to suggest that accessibility to children played no part in the Court’s analysis. On the contrary, the Court regarded the feasibility of using technology to restrict minors’ access to indecent material to be an important consideration. Id. at 814. A close reading of the opinion reveals, however, that the Court believed that these considerations were only relevant in evaluating whether the statute in question represented the least restrictive means available. They played no role in determining the level of First Amendment protection that cable received. 303. Id. at 813 (quoting Cohen v. California, 403 U.S. 15, 21 (1971) and citing Erzoznik v. City of Jacksonville, 422 U.S. 205, 210–11 (1975)). 300. 529 U.S. 803 (2000). 302. Playboy, 529 U.S. at 811–14. 308. See WGBH Educ. Found., 69 F.C.C.2d 1250, 1254–55 ¶ 10 (1978) (mem. op. & order); see also Rahall Broad. of Ind., Inc., 94 F.C.C.2d 1162 (1983) (mem. op. & order) (“The Paciﬁca decision . . . affords the commission no general prerogative to intervene in any case where words similar or identical to those in Paciﬁca are broadcast.”); Paciﬁca Found., 95 F.C.C.2d 750 (1983) (mem. op. & order) (“The opinion of the Court [in Paciﬁca] speciﬁcally stated that it was not ruling that ‘an occasional expletive . . . would justify any sanction . . . .’Applying these principles to the instant case, it is clear that petitioner failed to make a prima facie case . . . .”); Serv. Broad. Corp., 46 Rad. Reg. 2d (P & F) 413 (1979) (“While it is true that we may impose sanctions on a licensee who has aired obscene, indecent, or profane language, we have no factual evidence that this is the case here . . . . [T]he Commission does not attempt to regulate the rhetorical quality of language spoken on the air.”). 310. In a troika of decisions handed down on the same day, the FCC made three substantive changes to its indecency standards. First, it made clear that in addition to the repeated use of expletives, indecency covered other forms of sexually and scatologically oriented speech. See Inﬁnity Broad. Corp., 2 F.C.C.R. 2705, 2706 ¶ 9 (1987) (mem. op. & order) (ruling that speech can be indecent even in the absence of expletives if it includes sexual innuendo which, in context, is susceptible of only one meaning); Regents of Univ. of Cal., 2 F.C.C.R. 2703 (1987) (mem. op. & order) (ruling that speech can be indecent even in the absence of expletives when it describes or depicts sexual organs or activities); p g q y g g p 309. See Inﬁnity Broad. Corp., 3 F.C.C.R. 930, 930 ¶ 4 (1987) (mem. op. & order) (on reconsidera- tion), rev’d on other grounds sub nom. Action for Children’s Television v. FCC, 852 F.2d 1332 (D.C. Cir. 1988) [hereinafter ACT I]. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 301 standard.305 In the process, Justice Breyer appears to have completely aban- doned the language in his Denver plurality indicating that some lower standard may apply. If anything, Justice Breyer appears to chide the majority for even suggesting that the applicable standard was at issue.306 It thus seems that, despite the disagreement in Denver on this very point just ﬁve years earlier, the Court is now unanimous in believing that cable television should receive full First Amendment protection. Playboy thus seems to foreclose the possibility that Paciﬁca will have any applicability to cable television or any other medium outside of broadcasting. 305. Id. at 836, 846 (Breyer, J., dissenting). 306. See id. at 836 (Breyer, J., dissenting) (arguing that “[t]his case involves the application, not the elucidation of First Amendment principles” and the application of “established First Amendment law; . . . [t]he basic, applicable First Amendment principles are not at issue”); id. at 846 (noting that “[t]his disagreement is not about . . . basic First Amendment principle”). 307. For excellent reviews of the FCC’s early enforcement practices with respect to indecent speech, see Levi, supra note 72, at 86–112; KRATTENMAKER & POWE, supra note 72, at 104–14; Robinson, supra note 249, at 949–59. 305. Id. at 836, 846 (Breyer, J., dissenting). less stringent than strict scrutiny.299 The Denver plurality’s attempt to revive Paciﬁca ultimately proved short- lived. The Court’s subsequent decision in United States v. Playboy Entertain- ment Group, Inc.300 appears to have foreclosed the possibility that Paciﬁca could serve as a new foundation for according certain media less than full First Amendment protection. Playboy involved a First Amendment challenge to a statute requiring cable operators either to “fully scramble or otherwise fully block” channels that are “primarily dedicated to sexually-oriented program- ming” or else to limit their transmission to hours when children were unlikely to be viewing.301 In sharp contrast to the language in Denver suggesting that some other standard might apply, the Playboy Court squarely held that all attempts to regulate indecency on cable television are subject to strict scrutiny.302 In so holding, the court explicitly abandoned Paciﬁca’s concern with the ineffective- ness of warnings and returned instead to the view that individuals are expected to protect themselves from transient exposure to offensive material “‘simply by averting [our] eyes.’”303 Moreover, because cable and broadcast television are largely indistinguishable in the extent to which they invade the home and are accessible to children, the Court’s decision to extend full First Amendment protection to cable in Playboy can only be regarded as a rejection of the proposition that either of those considerations justiﬁed permitting more intru- sive regulation of speech.304 What is perhaps most startling about the Playboy decision is the complete- ness of the majority’s victory on this point. Writing on behalf of the four dissenters, Justice Breyer also accepted strict scrutiny as the unquestioned 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 307. For excellent reviews of the FCC’s early enforcement practices with respect to indecent speech, see Levi, supra note 72, at 86–112; KRATTENMAKER & POWE, supra note 72, at 104–14; Robinson, supra note 249, at 949–59. 305. Id. at 836, 846 (Breyer, J., dissenting). 306. See id. at 836 (Breyer, J., dissenting) (arguing that “[t]his case involves the application, not the elucidation of First Amendment principles” and the application of “established First Amendment law; . . . [t]he basic, applicable First Amendment principles are not at issue”); id. at 846 (noting that “[t]his disagreement is not about . . . basic First Amendment principle”). 307. For excellent reviews of the FCC’s early enforcement practices with respect to indecent speech, see Levi, supra note 72, at 86–112; KRATTENMAKER & POWE, supra note 72, at 104–14; Robinson, supra note 249, at 949–59. 308. See WGBH Educ. Found., 69 F.C.C.2d 1250, 1254–55 ¶ 10 (1978) (mem. op. & order); see also Rahall Broad. of Ind., Inc., 94 F.C.C.2d 1162 (1983) (mem. op. & order) (“The Paciﬁca decision . . . affords the commission no general prerogative to intervene in any case where words similar or identical to those in Paciﬁca are broadcast.”); Paciﬁca Found., 95 F.C.C.2d 750 (1983) (mem. op. & order) (“The opinion of the Court [in Paciﬁca] speciﬁcally stated that it was not ruling that ‘an occasional expletive . . . would justify any sanction . . . .’Applying these principles to the instant case, it is clear that petitioner failed to make a prima facie case . . . .”); Serv. Broad. Corp., 46 Rad. Reg. 2d (P & F) 413 (1979) (“While it is true that we may impose sanctions on a licensee who has aired obscene, indecent, or profane language, we have no factual evidence that this is the case here . . . . [T]he Commission does not attempt to regulate the rhetorical quality of language spoken on the air.”). 309. See Inﬁnity Broad. Corp., 3 F.C.C.R. 930, 930 ¶ 4 (1987) (mem. op. & order) (on reconsidera- tion), rev’d on other grounds sub nom. Action for Children’s Television v. FCC, 852 F.2d 1332 (D.C. Cir. 1988) [hereinafter ACT I]. 310. In a troika of decisions handed down on the same day, the FCC made three substantive changes to its indecency standards. First, it made clear that in addition to the repeated use of expletives, indecency covered other forms of sexually and scatologically oriented speech. See Inﬁnity Broad. Corp., 2 F.C.C.R. 2705, 2706 ¶ 9 (1987) (mem. op. & order) (ruling that speech can be indecent even in the absence of expletives if it includes sexual innuendo which, in context, is susceptible of only one meaning); Regents of Univ. of Cal., 2 F.C.C.R. 2703 (1987) (mem. op. & order) (ruling that speech can be indecent even in the absence of expletives when it describes or depicts sexual organs or activities); 2. Paciﬁca’s Questionable Vitality with Respect to Broadcasting At the same time the Supreme Court was shutting the door on the possibility that Paciﬁca might apply to other media, FCC and judicial decisions began to raise doubts about Paciﬁca’s vitality with respect to broadcasting as well. Almost from the time the decision was handed down, the FCC has taken steps to distance itself from Paciﬁca.307 Following the announcement of the decision, the FCC suggested that it was limiting Paciﬁca to its facts and restricted its application only to repeated use of the seven expletives at issue in that case.308 As a result, the FCC did not bring any indecency enforcement actions for nearly a decade.309 310 At the same time the Supreme Court was shutting the door on the possibility that Paciﬁca might apply to other media, FCC and judicial decisions began to raise doubts about Paciﬁca’s vitality with respect to broadcasting as well. Almost from the time the decision was handed down, the FCC has taken steps to distance itself from Paciﬁca.307 Following the announcement of the decision, the FCC suggested that it was limiting Paciﬁca to its facts and restricted its application only to repeated use of the seven expletives at issue in that case.308 the FCC suggested that it was limiting Paciﬁca to its facts and restricted its application only to repeated use of the seven expletives at issue in that case.308 As a result, the FCC did not bring any indecency enforcement actions for nearly As a result, the FCC did not bring any indecency enforcement actions for nearly a decade.309 Even when FCC interest in indecency enforcement revived in 1987,310 the THE GEORGETOWN LAW JOURNAL 302 [Vol. 91:245 agency continued to distance itself from the notion that broadcasting was subject to a lower level of First Amendment scrutiny. Notably, the FCC’s renewed interest in indecency enforcement coincided with its explicit abandon- ment of scarcity as a constitutional justiﬁcation for restricting indecent broad- casts.311 Even more dramatic was the FCC’s rejection of the notion that Paciﬁca justiﬁed subjecting indecency restrictions on broadcasting to a lower First Amendment standard than other media.312 Instead, the FCC maintained that the indecency restrictions that applied to broadcasting were “consistent with the First Amendment protection applicable to print.”313 After several false starts,314 a series of D.C. p g y 316. ACT I, 852 F.2d at 1343 n.18; ACT II, 932 F.2d at 1509; ACT III, 58 F.3d at 659–60; ACT IV, 59 F.3d at 1252; id. at 1263 (Edwards, C.J., concurring with reservations). But see ACT IV, 59 F.3d at 1262 (noting but declining to rely on the district court’s contention that broadcasting received a lower degree of First Amendment protection than print). Paciﬁca Found., Inc., 2 F.C.C.R. 2698, 2699–700 ¶ 13 (1987) (mem. op. & order) (ruling that speech can be indecent in the absence of expletives if it describes or depicts sexual or excretory functions in a manner patently offensive under contemporary community standards). Second, the FCC signaled a greater willingness to apply its indecency regulations to programs broadcast after ten p.m. See Paciﬁca, 2 F.C.C.R. at 2699–700 ¶ 16, 2701 ¶ 25; Regents of Univ. of Cal., 2 F.C.C.R. at 2704 ¶ 6. Third, the FCC appeared to extend its indecency decision to cover children over the age of twelve. See Paciﬁca, 2 F.C.C.R. at 2699–700 ¶ 16; Regents of Univ. of Cal., 2 F.C.C.R. at 2704 n.10; Inﬁnity, 2 F.C.C.R. at 2707 n.14. 317. See ACT IV, 59 F.3d at 1254 (noting that United States v. Evergreen Media Corp., 832 F. Supp. 1183 (N.D. Ill. 1983), represented the only judicial challenge to an FCC indecency action); Robert Corn-Revere, Eye of the Beholder, BROAD. & CABLE, Apr. 16, 2001, at 30 (same). 314. ACT I, 852 F.2d 1332 (D.C. Cir. 1988) (vacating and remanding the FCC’s decision to regulate indecent programs broadcast after ten p.m.); Action for Children’s Television v. FCC, 932 F.2d 1504, 1509 (D.C. Cir. 1991) [hereinafter ACT II] (vacating and remanding subsequent FCC order implement- ing statute calling for complete ban on broadcast indecency). 315. ACT I, 852 F.2d at 1337–40 (upholding the FCC’s revised deﬁnition of indecency); Action for Children’s Television v. FCC, 58 F.3d 654, 664–69 (1995) (en banc) [hereinafter ACT III] (upholding construing indecency as also protecting children aged twelve through seventeen and upholding prohibi- tion of the broadcast of indecent speech between six a.m. and ten p.m.); Action for Children’s Television v. FCC, 59 F.3d 1249, 1259–62 (D.C. Cir. 1995) [hereinafter ACT IV] (upholding FCC’s procedure for enforcing the indecency statutes). 311. Paciﬁca, 2 F.C.C.R. at 2699 ¶ 11. y f 318. Evergreen, 832 F. Supp. at 1184. 2707 n.14. 311. Paciﬁca, 2 F.C.C.R. at 2699 ¶ 11. 312. Id. 313. Id. 314. ACT I, 852 F.2d 1332 (D.C. Cir. 1988) (vacating and remanding the FCC’s decision to regulate indecent programs broadcast after ten p.m.); Action for Children’s Television v. FCC, 932 F.2d 1504, 1509 (D.C. Cir. 1991) [hereinafter ACT II] (vacating and remanding subsequent FCC order implement- ing statute calling for complete ban on broadcast indecency). 315. ACT I, 852 F.2d at 1337–40 (upholding the FCC’s revised deﬁnition of indecency); Action for Children’s Television v. FCC, 58 F.3d 654, 664–69 (1995) (en banc) [hereinafter ACT III] (upholding construing indecency as also protecting children aged twelve through seventeen and upholding prohibi- tion of the broadcast of indecent speech between six a.m. and ten p.m.); Action for Children’s Television v. FCC, 59 F.3d 1249, 1259–62 (D.C. Cir. 1995) [hereinafter ACT IV] (upholding FCC’s procedure for enforcing the indecency statutes). 316. ACT I, 852 F.2d at 1343 n.18; ACT II, 932 F.2d at 1509; ACT III, 58 F.3d at 659–60; ACT IV, 59 F.3d at 1252; id. at 1263 (Edwards, C.J., concurring with reservations). But see ACT IV, 59 F.3d at 1262 (noting but declining to rely on the district court’s contention that broadcasting received a lower degree of First Amendment protection than print). 317. See ACT IV, 59 F.3d at 1254 (noting that United States v. Evergreen Media Corp., 832 F. Supp. 1183 (N.D. Ill. 1983), represented the only judicial challenge to an FCC indecency action); Robert Corn-Revere, Eye of the Beholder, BROAD. & CABLE, Apr. 16, 2001, at 30 (same). 318. Evergreen, 832 F. Supp. at 1184. 324. See Playboy, 529 U.S. at 824 (“It is no response that voluntary blocking requires a consumer to take action, or may be inconvenient, or may not go perfectly every time.”); see also Reno, 521 U.S. at 876–77 (noting that, as an alternative to the Child Decency Act, “the District Court found that ‘despite its limitations, currently available user-based software suggests that a reasonably effective method by which parents can prevent their children from accessing sexually explicit and other material which parents may believe is inappropriate for their children will soon be widely available.’” (emphasis in 321. Indus. Guidance, 16 F.C.C.R. at 8000 ¶ 3. The FCC’s issuance of this industry guidance appears to have signaled renewed interest in indecency enforcement, as the FCC issued Notice of Apparent Liability to radio stations for broadcasting a cleaned-up version of Eminem’s song, “The Real Slim Shady,” as well as Sarah Jones’s anti-misogynist rap protest, “Your Revolution.” See KBOO Found., 16 F.C.C.R. 10731 (2001) (notice of apparent liability for forfeiture); Bill McConnell, FCC Hip-Hop Deep in #@!, BROAD. & CABLE, July 9, 2001, at 28. Those proceedings may provide another opportunity for judicial conﬁrmation that broadcasting should receive full First Amendment protection. Although the FCC eventually rescinded its action with respect to Eminem, see Citadel Broad. Co., 17 F.C.C.R. 483, 485–86 ¶¶ 8–9 (2002) (mem. op. & order), its action against Sarah Jones’s work appears destined to end up in the courts, see Bill McConnell, Rapper Sues FCC, BROAD. & CABLE, Feb. 4, 2002, at 16. 323. Playboy, 529 U.S. at 816; Reno, 521 U.S. at 877; Denver Area Educ. Telecomms. Consortium, Inc. v. FCC, 518 U.S. 727, 756 (1996) (plurality opinion); Sable, 492 U.S. at 128; Rowan, 397 U.S. at 736–37. 322. United States v. Playboy Entm’t Group, Inc., 529 U.S. 803, 823–26 (2000); Reno v. ACLU, 521 U.S. 844, 867, 869–70 (1997); Sable Communications of Cal., Inc. v. FCC, 492 U.S. 115, 127–28 (1989); Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 74 (1983). 319. See Indus. Guidance on Comm’n’s Case Law Interpreting 18 U.S.C. § 1464 & Enforcement Policies Regarding Broad. Indecency, Policy Statement, 16 F.C.C.R. 7999, 8016 n.23 (2001) [hereinaf- ter Indus. Guidance]. 2. Paciﬁca’s Questionable Vitality with Respect to Broadcasting Circuit opinions eventually sustained the FCC’s new indecency initiatives.315 In these cases, however, the court agreed that strict scrutiny applied to regulations restricting indecency on broadcasting in the same way that it applied to similar restrictions on other media.316 Since that time, there has been only one judicial challenge to the merits of an FCC indecency determination,317 and that case held that restrictions on the broadcast of indecent material are subject to strict scrutiny.318 The FCC settled that case after the district court denied its motion to summarily dispose of the constitutional challenge, agreeing as a condition of that settlement to publish more deﬁnitive industry guidance regarding its enforcement policies with re- 314. ACT I, 852 F.2d 1332 (D.C. Cir. 1988) (vacating and remanding the FCC’s decision to regulate indecent programs broadcast after ten p.m.); Action for Children’s Television v. FCC, 932 F.2d 1504, 1509 (D.C. Cir. 1991) [hereinafter ACT II] (vacating and remanding subsequent FCC order implement- ing statute calling for complete ban on broadcast indecency). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 2003] 303 spect to broadcast indecency.319 When the FCC ﬁnally issued industry guidance seven years later,320 it again recognized that all attempts to restrict broadcast indecency were subject to strict scrutiny.321 The FCC’s and the courts’ willing- ness to subject restrictions on broadcast indecency to strict scrutiny cannot easily be squared with Paciﬁca’s suggestion that such programming is subject to a lower First Amendment standard. Instead, it suggests that, although Paciﬁca continues to inﬂuence the way the applicable legal principles play out in each particular factual context, it no longer serves as a justiﬁcation for deviating from the strict scrutiny standard that applies to all other media. 320. This occurred despite the FCC’s promise to issue such guidance within nine months. See Corn-Revere, supra note 317, at 30. 327. 47 U.S.C. § 303(x) (2000). 328. The statute gave the broadcast industry one year in which to implement a voluntary rating system. If it failed to do so, the FCC would promulgate one itself. Telecommunications Act of 1996, Pub. L. No. 104-104, § 551(e)(1)(A), 110 Stat. 56, 139–42 (originally codiﬁed at 47 U.S.C. § 303(w)). The major networks subsequently agreed to a rating system. The system initially adopted by the industry employed six basic ratings based on age: TV-Y for young children, TV-Y7 for children 7 and younger, TV-G for general audiences, TV-PG to suggest parental guidance, TV-14 for kids 14 and older, and TV-MA for mature audiences only. The industry later agreed to supplement the basic ratings with a series of content codes: S for sex, V for violence, L for foul language, D for sexual innuendo, and FV for fantasy violence in programs for children 7 and older. See Implementation of Section 551 of Telecomms. Act of 1996, Report and Order, 13 F.C.C.R. 8232 (1998); Paige Albiniak, Ratings Get Revamped: Networks, Except for NBC, Agree to Add Content Labels, BROAD. & CABLE, July 14, 1997, at 4. All of the broadcast networks have agreed to implement the age-based ratings and all except NBC have agreed to use the content codes. Among cable networks, only Black Entertainment Television, QVC, and the Home Shopping Network have refused to rate their programs. See Bill McConnell, Non-Violent TV, BROAD. & CABLE, Feb. 7, 2000, at 40. The scholarly consensus, even among those who advocate self-regulation by the broadcast industry, concedes that the ratings are properly regarded as p y g 326. Compare supra notes 257–67, 283–88, 303 and accompanying text (noting that Cohen, Erzoznik, Rowan, Bolger, Sable, Reno, and Playboy did not regard the possibility of transient exposure as sufﬁcient to justify restrictions on speech), with supra notes 86, 256, 293 and accompanying text (noting the contrary conclusion drawn by Paciﬁca and the Denver plurality opinion). original)); Sable, 492 U.S. at 128–29 (refusing to uphold legislation denying adults access to “dial-a- porn” when it found credit card, access code, and scrambling rules a “less restrictive means, short of a total ban, to achieve the government’s interest in protecting minors”). g p g ) 325. As noted earlier, MVPDs have now captured well over 80% of all U.S. households. See supra note 178 and accompanying text. THE GEORGETOWN LAW JOURNAL Once such means exist, the proper balance of First Amendment interests places the burden on parents and viewers to take the appropriate steps. These doctrinal considerations underscore the potential constitutional signiﬁ- cance of technologies that increase viewers’ ability to control what appears on their television sets. The emerging dominance of multichannel video providers, such as cable television and DBS,325 has greatly enhanced the ability of audiences to tailor their viewing environment to their personal preferences. The protection from exposure to unwanted programming is not complete, however, because ﬁltering by changing the channel still involves some ﬂeeting contact with unwanted material. Although the weight of Supreme Court authority suggests that such transient exposure lacks constitutional signiﬁcance, the issue is not completely free from doubt.326 Regardless of the ﬁnal resolution of that debate, the emergence of technolo- gies that allow viewers to screen out unwanted programming promises to offer a more deﬁnitive interment of the rationales in Paciﬁca for according a lower degree of First Amendment protection. The most important of these is the development of the “V-chip,” which if properly programmed can block the display of programs receiving particular content ratings. The Telecommunica- tions Act of 1996 started in motion the steps necessary to make the V-chip universal, ﬁrst by requiring that all television sets over thirteen inches manufac- tured in January 2000 or later contain a V-chip,327 and second by essentially requiring the industry to establish a rating system.328 The bulk of the scholar- 3 7. 7 U.S.C. § 303( ) ( 000). 328. The statute gave the broadcast industry one year in which to implement a voluntary rating system. If it failed to do so, the FCC would promulgate one itself. Telecommunications Act of 1996, Pub. L. No. 104-104, § 551(e)(1)(A), 110 Stat. 56, 139–42 (originally codiﬁed at 47 U.S.C. § 303(w)). The major networks subsequently agreed to a rating system. The system initially adopted by the industry employed six basic ratings based on age: TV-Y for young children, TV-Y7 for children 7 and younger, TV-G for general audiences, TV-PG to suggest parental guidance, TV-14 for kids 14 and older, and TV-MA for mature audiences only. The industry later agreed to supplement the basic ratings with a series of content codes: S for sex, V for violence, L for foul language, D for sexual innuendo, and FV for fantasy violence in programs for children 7 and older. g y 327. 47 U.S.C. § 303(x) (2000). original)); Sable, 492 U.S. at 128–29 (refusing to uphold legislation denying adults access to “dial-a- porn” when it found credit card, access code, and scrambling rules a “less restrictive means, short of a total ban, to achieve the government’s interest in protecting minors”). C. THE TECHNOLOGICAL EVISCERATION OF PACIFICA In addition to the analytical and doctrinal shortcomings of the Paciﬁca opinion discussed above, two technological developments raise further ques- tions about Paciﬁca’s continuing vitality. The Supreme Court has repeatedly emphasized that Paciﬁca does not apply when viewers are unlikely to encounter indecent speech by accident and when prior warnings are likely to be effec- tive.322 The Court has also emphasized that prohibitions of indecent speech cannot stand when alternative means exist that enable individual viewers to control what they see and hear.323 This is true even if individual viewers must take afﬁrmative steps to protect themselves and their families from inadvertent exposure to such speech and if the means of self-protection are not foolproof.324 304 [Vol. 91:245 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL original)); Sable, 492 U.S. at 128–29 (refusing to uphold legislation denying adults access to “dial-a- porn” when it found credit card, access code, and scrambling rules a “less restrictive means, short of a total ban, to achieve the government’s interest in protecting minors”). 325. As noted earlier, MVPDs have now captured well over 80% of all U.S. households. See supra note 178 and accompanying text. 326. Compare supra notes 257–67, 283–88, 303 and accompanying text (noting that Cohen, Erzoznik, Rowan, Bolger, Sable, Reno, and Playboy did not regard the possibility of transient exposure as sufﬁcient to justify restrictions on speech), with supra notes 86, 256, 293 and accompanying text (noting the contrary conclusion drawn by Paciﬁca and the Denver plurality opinion). 327. 47 U.S.C. § 303(x) (2000). 328. The statute gave the broadcast industry one year in which to implement a voluntary rating system. If it failed to do so, the FCC would promulgate one itself. Telecommunications Act of 1996, Pub. L. No. 104-104, § 551(e)(1)(A), 110 Stat. 56, 139–42 (originally codiﬁed at 47 U.S.C. § 303(w)). The major networks subsequently agreed to a rating system. The system initially adopted by the industry employed six basic ratings based on age: TV-Y for young children, TV-Y7 for children 7 and younger, TV-G for general audiences, TV-PG to suggest parental guidance, TV-14 for kids 14 and older, and TV-MA for mature audiences only. The industry later agreed to supplement the basic ratings with a series of content codes: S for sex, V for violence, L for foul language, D for sexual innuendo, and FV for fantasy violence in programs for children 7 and older. See Implementation of Section 551 of Telecomms. Act of 1996, Report and Order, 13 F.C.C.R. 8232 (1998); Paige Albiniak, Ratings Get Revamped: Networks, Except for NBC, Agree to Add Content Labels, BROAD. & CABLE, July 14, 1997, at 4. All of the broadcast networks have agreed to implement the age-based ratings and all except NBC have agreed to use the content codes. Among cable networks, only Black Entertainment Television, QVC, and the Home Shopping Network have refused to rate their programs. See Bill McConnell, Non-Violent TV, BROAD. & CABLE, Feb. 7, 2000, at 40. The scholarly consensus, even among those who advocate self-regulation by the broadcast industry, concedes that the ratings are properly regarded as THE GEORGETOWN LAW JOURNAL See Implementation of Section 551 of Telecomms. Act of 1996, Report and Order, 13 F.C.C.R. 8232 (1998); Paige Albiniak, Ratings Get Revamped: Networks, Except for NBC, Agree to Add Content Labels, BROAD. & CABLE, July 14, 1997, at 4. All of the broadcast networks have agreed to implement the age-based ratings and all except NBC have agreed to use the content codes. Among cable networks, only Black Entertainment Television, QVC, and the Home Shopping Network have refused to rate their programs. See Bill McConnell, Non-Violent TV, BROAD. & CABLE, Feb. 7, 2000, at 40. The scholarly consensus, even among those who advocate self-regulation by the broadcast industry, concedes that the ratings are properly regarded as 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 305 ship on the V-chip has focused on the constitutionality of Congress’s decision to mandate its installation and, in effect, to require program ratings.329 What has largely escaped scholarly attention is that, at least with respect to television, widescale deployment of the V-chip will render all attempts to restrict the broadcast of indecent programming unconstitutional. As newer televisions dis- place older sets, the V-chip should effectively prevent those who do not wish to be exposed to indecent speech from encountering such speech by accident. In addition, as the Court held in Playboy, the technological ability to block unwanted channels on a household-by-household basis “enables the Govern- ment to support parental authority without affecting the First Amendment interests of speakers and willing listeners.”330 As a result, as recognized in the only portion of the Denver opinion that commanded a majority of the Court, the V-chip constitutes a less restrictive means sufﬁcient to render any ban on indecent speech unconstitutional.331 In addition, a new technology known as video-on-demand (VOD) promises to reduce the danger of unwanted exposure to indecent speech even further. When fully deployed, VOD will give viewers total control over the program- ming that enters their homes. As a result, it promises to transform television from a “push” technology, in which control over what speech will be conveyed resides with media companies, into—at least in part—a “pull” technology, in which individuals decide for themselves which programs they would like to see. Unfortunately, widescale deployment of VOD is not yet feasible. In the mean- time, MVPDs are in the process of deploying a related technology known as near video-on-demand (NVOD) that provides some of the same functionality. y y p 331. Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727, 756 (1996) (noting that the impending arrival of the V-chip suggests that the segregation and blocking requirement imposed by statute was not sufﬁciently tailored); see also Edwards & Berman, supra note 253, at 1513–14 (noting that the V-chip promised to “‘empower’ parents to regulate all aspects of their children’s viewing”). ; y, p , 330. United States v. Playboy Entm’t Group, Inc., 529 U.S. 803, 815 (2000). f ( ) 329. See J.M. Balkin, Media Filers, the V-Chip and the Foundations of Broadcast Regulation, 45 DUKE L.J. 1131, 1157 (1996); Edwards & Berman, supra note 253, at 1514–15; Spitzer, supra note 36, at 454–90; Polivy, supra note 328, at 1773–79. the product of a state mandate. See Angela J. Campbell, Self-Regulation and the Media, 51 FED. COMM. L.J. 711, 755–56 (1999); Spitzer, supra note 36, at 436–54; Denise R. Polivy, Note and Comment, Virtue by Machine: A First Amendment Analysis of the V-Chip Provisions of the Telecommunications Act of 1996, 29 CONN. L. REV. 1749, 1761–73 (1997). the product of a state mandate. See Angela J. Campbell, Self-Regulation and the Media, 51 FED. COMM. L.J. 711, 755–56 (1999); Spitzer, supra note 36, at 436–54; Denise R. Polivy, Note and Comment, Virtue by Machine: A First Amendment Analysis of the V-Chip Provisions of the Telecommunications Act of 1996, 29 CONN. L. REV. 1749, 1761–73 (1997). 329. See J.M. Balkin, Media Filers, the V-Chip and the Foundations of Broadcast Regulation, 45 DUKE L.J. 1131, 1157 (1996); Edwards & Berman, supra note 253, at 1514–15; Spitzer, supra note 36, at 454–90; Polivy, supra note 328, at 1773–79. 330. United States v. Playboy Entm’t Group, Inc., 529 U.S. 803, 815 (2000). 331. Denver Area Educ. Telecomms. Consortium v. FCC, 518 U.S. 727, 756 (1996) (noting that the impending arrival of the V-chip suggests that the segregation and blocking requirement imposed by statute was not sufﬁciently tailored); see also Edwards & Berman, supra note 253, at 1513–14 (noting that the V-chip promised to “‘empower’ parents to regulate all aspects of their children’s viewing”). 332. Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programming, Eighth Annual Report, 17 F.C.C.R. 1244, 1264–65 ¶¶ 40–41, 1271 ¶ 54, 1286–87 ¶ 92, 1292–93 ¶ 103, 1318–21 ¶¶ 187–90, 1322–23 ¶ 194 (2002). 333. OWEN, supra note 182, at 8, 10–11, 218–20. 334. See Bill McConnell, Critics Are Out for (No) Blood on TV, BROAD. & CABLE, Jan. 27, 2003, at 1; Siobhan McDonough, Too Much Indecency Allowed on Airwaves, FCC Commissioner Says, SAN DIEGO UNION-TRIB., Nov. 22, 2002, at A11. 335. Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390 (1969). THE GEORGETOWN LAW JOURNAL Unlike pure VOD, which is a point-to-point service that allows a content provider to deliver a particular program to a particular customer on request, NVOD is a point-to-multipoint service in which customers can select from among a broad range of viewing choices offered by the content provider. Providing the same range of options to all viewers allows NVOD to use bandwidth more efﬁciently than VOD. It remains an interactive, pull-oriented service that gives the audience greater control of what is viewed. Current NVOD options are relatively limited, consisting mainly of a small number of pay-per-view offerings. The FCC’s most recent annual assessment of the televi- 306 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 sion industry indicates that cable and DBS providers are investing large amounts of capital in an effort to make available a wider range of NVOD services.332 As the television programming received increasingly becomes the product of the desires of each individual household, the justiﬁcations for governmental intervention to prevent accidental contact with unwanted speech weaken. In so arguing, I do not mean to overstate my case. It is quite possible that some viewers may prefer that television be a predominantly passive experience.333 As a result, it is not yet possible to determine precisely where along the push-pull spectrum television will ultimately settle. In time, however, the increase in viewers’ ability to control what they see will eventually alter the constitutional analysis by undermining the basic reasons announced in Paciﬁca for subjecting broadcasting to a different level of First Amendment scrutiny. It thus appears that the pervasiveness and accessibility rationales of Paciﬁca are in the process of being overrun by technology, as well as by theory and doctrine in almost precisely the same manner as the scarcity doctrine. As of today, however, Paciﬁca remains good law, and regulatory authorities remain free to advance policies based on it. This danger may be more real than is generally recognized: Three of the ﬁve current FCC Commissioners have recently issued public calls for greater restrictions on indecent programming.334 It is thus clear that until the Supreme Court formally abandons Paciﬁca, that decision is likely to continue to be invoked as a basis for allowing the level of First Amendment scrutiny to vary technology by technology. 332. Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programming, Eighth Annual Report, 17 F.C.C.R. 1244, 1264–65 ¶¶ 40–41, 1271 ¶ 54, 1286–87 ¶ 92, 1292–93 ¶ 103, 1318–21 ¶¶ 187–90, 1322–23 ¶ 194 (2002). IV. BEYOND SCARCITY AND PACIFICA: THE TURN TO CIVIC REPUBLICANISM The collapse of the traditional justiﬁcations for holding broadcasting to a lower First Amendment standard has led scholars to search for new constitu- tional bases for the Broadcast Model. Drawing inspiration from the statement in Red Lion contending that “[i]t is the right of the viewers and listeners, not the right of the broadcasters, which is paramount,”335 these scholars have offered a new basis for holding broadcasting to a lesser level of First Amendment scrutiny rooted in the needs of democratic self-governance. This Part analyzes such efforts through the work of Owen Fiss and Cass Sunstein, who have offered the most elaborate theories of this kind. According to Fiss, the conventional wisdom holding that the First Amend- ment protects the autonomy of speakers for its own sake is simply wrong. Rather, he argues, the true central value of the First Amendment is the promo- 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 307 tion of the robust public debate needed to support democratic decisionmak- ing.336 As Fiss puts it, “[a]utonomy is protected not because of its intrinsic value, . . . but rather as a means or instrument of collective self-determination. We allow people to speak so others can vote.”337 The circumstances that existed tion of the robust public debate needed to support democratic decisionmak- ing.336 As Fiss puts it, “[a]utonomy is protected not because of its intrinsic value, . . . but rather as a means or instrument of collective self-determination. We allow people to speak so others can vote.”337 The circumstances that existed when the First Amendment came of age, however, caused courts and commenta- tors to confuse the true end of the First Amendment with the means used to promote that end. At that time, access to speech venues was easy and the social structure relatively decentralized. In such a world, courts could safeguard the proper functioning of the democratic process simply by protecting the au- tonomy of speakers, because doing so typically represented the best available means for fostering robust public debate.338 The identity of these interests eventually came to be enshrined in the doctrinal rule against content regulation, which Fiss acknowledges emerged as “the cornerstone of the Free Speech Tradition.”339 When the focus shifts from the street corner speaker to CBS, Fiss argues, the opportunities for communication become much more limited. 336. FISS, LIBERALISM DIVIDED, supra note 24, at 13–17, 36–38. 337. Id. at 13. 338. Id. at 12–13, 37–38. 339. Id. at 12. 340. Id. at 15–16. 341. Id. at 38. 342. Id. at 10, 23, 43. 343. Id. at 13–14, 19. 343. Id. at 13–14, 19. IV. BEYOND SCARCITY AND PACIFICA: THE TURN TO CIVIC REPUBLICANISM In such a world, an autonomy-based First Amendment that conceives of free speech as a shield around the speaker necessarily sanctions the crowding out of other speech.340 Even more problematic in Fiss’s eyes is the growing inequality in the distribu- tion of power that he believes characterizes the social structure of contemporary society.341 In such a world, protecting autonomy will only induce what Charles Lindblom called “circularity,” in which both markets and governmental pro- cesses tend to reﬂect and reinforce the existing social structure.342 In such a world, the congruence between the autonomy principle and the public debate principle disappears. Quite the contrary, the two principles often become antago- nistic because the protection of autonomy may cause the public debate to become decidedly one-sided and insufﬁciently robust.343 As a result, Fiss argues that if the First Amendment is truly to promote democratic self-determination, fundamental jurisprudential changes are re- quired. Because autonomy can no longer serve as a proxy for the promotion of robust public debate, we as a society must promote public debate more directly. In addition, the law should recognize that private entities can threaten free speech values to the same extent as the government. As a result, Fiss argues that it is appropriate to look to the state not just as an enemy of speech, but rather as a possible means for mitigating the obstacles to robust public debate created by 336. FISS, LIBERALISM DIVIDED, supra note 24, at 13–17, 36–38. 337. Id. at 13. 338. Id. at 12–13, 37–38. 339. Id. at 12. 340. Id. at 15–16. 341. Id. at 38. 342. Id. at 10, 23, 43. 343. Id. at 13–14, 19. THE GEORGETOWN LAW JOURNAL [Vol. 91:245 308 [Vol. 91:245 private ordering.344 Fiss does not offer much detail about the circumstances under which such governmental intervention would be appropriate and the form that such intervention would take, aside from offering a fairly general endorse- ment of the existing scheme of broadcast regulation.345 He simply suggests that the government could regulate speech in a manner similar to a parliamentarian at a town meeting.346 A more complete deﬁnition of what constitutes robust public debate would be helpful, but is not strictly necessary. Concerns about the quality of public discourse under the current regime provide enough impetus for action.347 Sunstein offers a similar critique of the existing autonomy-oriented approach to the First Amendment. 344. Id. at 19–20, 39–41. 345. Id. at 149. 346. Id. at 85, 101, 117–19, 153; FISS, IRONY OF FREE SPEECH, supra note 24, at 21–24. 347. FISS, LIBERALISM DIVIDED, supra note 24, at 16, 26. 348. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 62–71; Sunstein, supra note 23, at 514–17. 349. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at xvi–xix, 18–20. 350. Id. at 3–8. 351. Id. at 58–67. 352. Id. at 18; Sunstein, supra note 23, at 520. 351. Id. at 58–67. 352. Id. at 18; Sunstein, supra note 23, at 520. IV. BEYOND SCARCITY AND PACIFICA: THE TURN TO CIVIC REPUBLICANISM Although Sunstein identiﬁes certain features that cause broadcast markets to fail,348 he focuses the bulk of his analysis on noneconomic considerations. Speciﬁcally, Sunstein argues that when properly understood, the First Amendment is designed to ensure that citizens are exposed to the types of communication they need to participate in a deliberative democracy.349 Indeed, the history of the First Amendment up until the 1970s can be seen as an ongoing debate between this vision of free speech and an alternate vision that viewed the First Amendment as protecting individual autonomy. In the 1980s, however, the autonomy vision prevailed and freedom of speech has ever since become increasingly identiﬁed with the preservation of free economic markets.350 In Sunstein’s opinion, the results of this shift have been catastrophic. Television news has alternately become either overly sensationalized or unduly focused on such trivial matters as entertainment personalities and “human impact” anec- dotes. Coverage of political campaigns has largely been reduced to a monistic focus on “horse race” issues, and statements by political candidates have deteriorated into mere “sound bites” rather than a serious discussion of the issues.351 The reasons for the destitution of the media are clear to Sunstein. A First Amendment that equates speech with economic markets strives to satisfy what individuals want. The choices we make as individuals, however, do not necessar- ily coincide with what we need as citizens actively engaged in a public discourse.352 In addition, Sunstein argues that, while the economic approach largely takes individual preferences as exogenous, individual media preferences are in reality the product of the type of programming that is currently being 2003] 309 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT broadcast.353 Consequently, Sunstein argues that using the media’s ability to satisfy existing preferences as the measure of current performance is ultimately somewhat circular, in that it effectively allows the status quo to serve as its own justiﬁcation.354 Taking existing preferences as exogenous also suffers from a different problem, in that it takes existing distribution of wealth as given. 361. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 35, 43, 81–88. p 363. Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting); JOHN STUART MILL, On Liberty, in THREE ESSAYS 5, 26–44 (Richard Wollheim ed., Oxford Univ. Press 1975) (1859). 362. Id. at 21, 89; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 222. 360. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221. 358. Id. at 9, 122–23, 130–37. 360. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221. 353. Sunstein, supra note 23, at 74. 354. Id. at 19, 71–74. 355. Id. at 28–34, 178–79. 356. Id. at xvi–xix. 357. Id. at 20–21. 358. Id. at 9, 122–23, 130–37. 359. Id. at 21. 360. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221. 361. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 35, 43, 81–88. 362. Id. at 21, 89; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 222. 363. Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting); JOHN STUART MILL, On Liberty, in THREE ESSAYS 5, 26–44 (Richard Wollheim ed., Oxford Univ. Press 1975) (1859). IV. BEYOND SCARCITY AND PACIFICA: THE TURN TO CIVIC REPUBLICANISM As such, it is analogous to the willingness to regard the current distribution of common law entitlements as a neutral constitutional baseline that was rejected during the New Deal.355 Sunstein’s solution to this state of affairs is to return to a “Madisonian” vision of the First Amendment that focuses not on promoting individual autonomy, but rather on promoting the broad communication about matters of public concern needed to facilitate democratic decisionmaking.356 To accomplish this, the government should be permitted to foster broad and deep attention to public issues, as well as public exposure to an appropriate diversity of views.357 In addition, Sunstein argues that his approach would require distinguishing be- tween political speech, which would receive the highest degree of protection, and nonpolitical speech, which necessarily plays a lesser role in fostering democratic deliberation.358 Sunstein offers little guidance as to how to deploy such a program, calling the remedies “obscure” and “far from clear.”359 Instead, Sunstein advocates a “frankly experimental approach”360 and tentatively pro- poses a series of possible remedial measures.361 Uncertainty about the precise steps to be taken should not prevent the government from taking action. When confronted with a system that is so clearly imperfect, something must be done.362 A number of common features unite both Fiss’s and Sunstein’s visions of free speech. Both view speech in largely instrumental terms, valuing it for its contribution to other, more fundamental principles rather than as an end in and of itself. In addition, both hold similar views about the fundamental principle that free speech is supposed to promote. In contrast to other instrumental theories, such as the one advanced by Holmes and Mill that favors protecting speech to promote the search for truth,363 Fiss and Sunstein agree that the true meta-value underlying our commitment to free speech is the proper functioning 310 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 of the democratic process.364 Both generally endorse the existing scheme of broadcast regulation, possibly augmented by the restoration of mandatory rights of reply, greater support for public television, additional limitations to campaign contributions, and improved coverage of political issues, perhaps encompassing free air time for politicians.365 Both endorse adopting a less hostile attitude towards governmental interference with private speech. 364. Fiss’s embrace of the instrumental vision of free speech is open and unqualiﬁed. FISS, LIBERALISM DIVIDED, supra note 24, at 36–38. Sunstein is more circumspect, but only on the margin. Although Sunstein reserves the possibility of other values underlying free speech, SUNSTEIN, DEMOC- RACY AND FREE SPEECH, supra note 23, at 129–30, he would accommodate those values only by adjusting the lower level of protection he would accord to nonpolitical speech rather than by adjusting the higher level of protection that his approach would extend to political speech. Id. at 129–30, 135–36, 147–48. 365. FISS, LIBERALISM DIVIDED, supra note 24, at 19, 22, 149, 152; FISS, IRONY OF FREE SPEECH, supra note 24, at 56; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 35, 43, 81–88; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221–22; Owen M. Fiss, Money and Politics, 97 COLUM. L. REV. 2470 (1997); Sunstein, supra note 23, at 525. 367. FISS, LIBERALISM DIVIDED, supra note 24, at 19, 152–53; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 178–79. 368. FISS, LIBERALISM DIVIDED, supra note 24, at 155–57; FISS, IRONY OF FREE SPEECH, supra note 24, at 68, 82–83; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 46–48. 370. FISS, LIBERALISM DIVIDED, supra note 24, at 17, 19, 51, 150; FISS, IRONY OF FREE SPEECH, supra note 24, at 64 (arguing that licensing is irrelevant to the scope of regulatory authority); SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 108–10 (same). 366. FISS, LIBERALISM DIVIDED, supra note 24, at 18–23, 38–41; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 34. , ; , , p , 369. See Fiss, supra note 24, at 1217; Sunstein, supra note 23, at 527–31. 371. See supra notes 347, 362 and accompanying text. 372. FISS, LIBERALISM DIVIDED, supra note 24, at 146; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 81; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 220. 373. See generally ALEXANDER MEIKLEJOHN, FREE SPEECH AND ITS RELATION TO SELF-GOVERNMENT (1948); ALEXANDER MEIKLEJOHN, POLITICAL FREEDOM (1960); Harry Kalven, The New York Times Case: A Note on “The Central Meaning of the First Amendment,” 1964 SUP. CT. REV. 191; Alexander Meiklejohn, The First Amendment Is an Absolute, 1961 SUP. CT. REV. 245. 374. 395 U.S. 367 (1969). 375. 376 U.S. 254 (1964). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT close analysis, Fiss and Sunstein, for the most part, simply assume away the conﬂict between the individual and the collective will. To the extent that they engage this conﬂict at all, they do so in terms inconsistent with the values of democracy they purport to support. Moving to a critique from inside the civic republican model, section B argues that even if one accepts the instrumental vision of speech proposed by Fiss and Sunstein, their proposals are too incompletely articulated to provide much concrete policy guidance. The result is an approach to regulation that is essentially ad hoc and all too dependent on direct normative value judgments that Fiss and Sunstein fail to articulate clearly, let alone defend. Finally, section C argues that Fiss and Sunstein have failed to come to grips with certain limitations to their theories imposed by recent technological developments. Simply put, the increase in the number of available media outlets and the impending arrival of video-on-demand promise to prevent television from playing the transformative role that Fiss and Sunstein envision. IV. BEYOND SCARCITY AND PACIFICA: THE TURN TO CIVIC REPUBLICANISM In their view, if an obstacle to democratic self-governance exists, it makes little difference whether the obstacle is the result of private ordering rather than state action.366 As a result, both Fiss and Sunstein call for eliminating the current doctrine’s presump- tion against content-based restrictions of speech367 and compelled speech.368 Both argue that the unique role that television plays in the public discourse justiﬁes the application of a different constitutional standard.369 In particular, their general acceptance of government intervention leads them to ﬁnd the licensing of speakers largely unproblematic.370 Finally, although both profess uncertainty about the speciﬁc remedies to be applied, both concur that the current state of affairs is so dire that something must be done.371 This Part will analyze Fiss’s and Sunstein’s theories and assess their potential as an alternative basis for upholding the constitutionality of the Broadcast Model. In particular, I will focus on what I believe to be the three major shortcomings of their proposals. Section A offers a critique from outside the civic republican model by evaluating the extent to which Fiss and Sunstein respond to arguments that view autonomy to be a central free speech value. The role of individual autonomy and the proper way to reconcile it with collective decisionmaking have represented one of the dominant theoretical questions in First Amendment theory over the last half century. My concern is that, upon 2003] 311 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 376. IMMANUEL KANT, GROUNDING FOR THE METAPHYSICS OF MORALS 428 (James W. Ellington trans., Hackett 2d ed. 1983) (1785). For a helpful analysis of Kant’s theory of free speech, see Christina E. Wells, Reinvigorating Autonomy: Freedom and Responsibility in the Supreme Court’s First Amendment Jurisprudence, 32 HARV. C.R.-C.L. L. REV. 159 (1997). 377. See BRUCE ACKERMAN, SOCIAL JUSTICE IN THE LIBERAL STATE (1980); JOHN RAWLS, A THEORY OF JUSTICE § 40, at 251–57 (1971); DAVID A. J. RICHARDS, TOLERATION AND THE CONSTITUTION 165–74 (1986); Charles Fried, The New First Amendment Jurisprudence: A Threat to Liberty, 59 U. CHI. L. REV. 225, 233 (1992); John Rawls, Justice as Fairness: Political Not Metaphysical, 14 PHIL. & PUB. AFF. 223, 227, 232–34, 240–44 (1985); Thomas Scanlon, A Theory of Freedom of Expression, 1 PHIL. & PUB. AFF. 204, 213–22 (1972). For general discussions on the distinction between ascriptive and descriptive visions of liberty and autonomy, see Richard H. Fallon, Jr., Two Senses of Autonomy, 46 STAN. L. REV. 875, 877–78 (1994); Robert Post, Meiklejohn’s Mistake Individual Autonomy and the Reform of Public Discourse, 64 U. COLO. L. REV. 1109, 1129–33 (1993). A. CRITIQUES FROM OUTSIDE THE MODEL: THE ROLE OF AUTONOMY The central premise of the work of both Fiss and Sunstein is that the First Amendment exists to serve the democratic process. They view speech as a means towards promoting this value rather than an end in and of itself. To the extent that other First Amendment values exist, they are subordinate to this fundamental commitment. To the extent that autonomy conﬂicts with the needs of the democratic process, it simply must give way.372 This is a powerful vision of free speech that traces back to Alexander Meiklejohn and Harry Kalven.373 It appears to have played a role in such landmark decisions as Red Lion Broadcasting Co. v. FCC374 and New York Times Co. v. Sullivan.375 This vision of the First Amendment has also proven to be quite controversial. Free speech theorists have long disputed the extent to which promoting the democratic process constitutes the central value of the First Amendment. A number of powerful theories have emerged that value speech as an end unto itself rather than as a means for promoting other values. This section evaluates Fiss’s and Sunstein’s attempts to come to grips with these alternative, autonomy-based visions of free speech. It begins by describ- ing the major autonomy-based theories, including those theories incorporating hybrid approaches that attempt to combine both descriptive and ascriptive visions of free speech. It then analyzes the speciﬁc ways in which Fiss and 312 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 Sunstein attempt to reconcile their positions with autonomy. Upon close analy- sis, it becomes clear that their primary analytical gambit is to assume the problem away by positing that individual choices and the outcome of the collective decisionmaking processes will tend to converge. As a result, both of their theories to fail to engage the conﬂict between the individual and the collective will that has long represented one of the foremost problems of liberal political theory. 385. MARTIN H. REDISH, FREEDOM OF EXPRESSION: A CRITICAL ANALYSIS 19–29 (1984). He has since published articles using his theory to critique the work of Fiss and Sunstein. See Redish & Kaludis, supra note 99, at 1107–10, 1119–21; Martin H. Redish & Gary Lippman, Freedom of Expression and the Civic Republican Revival in Constitutional Theory: The Ominous Implications, 79 CAL. L. REV 267, 284–90 (1991). 378. RONALD M. DWORKIN, FREEDOM’S LAW 199–200 (1996). 379. Id. at 200; see also id. at 7 (arguing that freedom of speech represents an indispensable aspect of “treat[ing] all those subject to its dominion as having equal moral and political status”). 380. Id. at 200. 381. Id.; see also id. (arguing that “withholding an opinion from us on the ground that we are not ﬁt to hear and consider it” thus deprives us of “our dignity, as individuals”). 382. Id. 383. Id. 384. Id. at 205. 385. MARTIN H. REDISH, FREEDOM OF EXPRESSION: A CRITICAL ANALYSIS 19–29 (1984). He has since published articles using his theory to critique the work of Fiss and Sunstein. See Redish & Kaludis, supra note 99, at 1107–10, 1119–21; Martin H. Redish & Gary Lippman, Freedom of Expression and the Civic Republican Revival in Constitutional Theory: The Ominous Implications, 79 CAL. L. REV 267, 284–90 (1991). 378. RONALD M. DWORKIN, FREEDOM’S LAW 199–200 (1996). 379. Id. at 200; see also id. at 7 (arguing that freedom of speech represents an indispensable aspect of “treat[ing] all those subject to its dominion as having equal moral and political status”). 380. Id. at 200. 381. Id.; see also id. (arguing that “withholding an opinion from us on the ground that we are not ﬁt to hear and consider it” thus deprives us of “our dignity, as individuals”). 382. Id. 1. Theories of Autonomy This subsection presents the major autonomy-based approaches to the First Amendment. It begins by describing the two major variants of such theories, which include those that treat autonomy as an indispensable attribute of indi- vidual dignity and those that believe autonomy’s status as a free speech value follows inexorably from our commitment to a democratic form of government. It then outlines a number of hybrid models that combine autonomy with descriptive approaches to the First Amendment. a. Deontological Theories of Autonomy. First and foremost, Fiss and Sunstein view the First Amendment in terms of whether individuals are able to gain exposure to the speech needed to participate in democratic self-governance in a meaningful way. Viewed in this manner, freedom of speech is largely a descrip- tive concept that refers to an empirical condition that must be achieved. Their work contrasts directly with scholars who have drawn on the Kantian precept that all individuals be respected as ends unto themselves376 to construct theories that view free speech as an irreducible attribute of personal sovereignty. Under these theories, autonomy is not a condition that is attained, but rather an entitlement that is ascribed to people based either on moral grounds ﬂowing from the need to respect each individual as an independent moral agent or on political grounds resulting from their status as constituent members of the democratic body politic. The result is an ascriptive vision of the First Amend- ment that respects autonomy regardless of whether it furthers any particular instrumental value.377 The most forceful statement of this position appears in the work of Ronald Dworkin. Dworkin rejects views that “treat[ ] free speech as 2003] 313 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 384. Id. at 205. 391. See FREDERICK SCHAUER, FREE SPEECH: A PHILOSOPHICAL ENQUIRY 35–46, 60–72, 85–86 (1982) (critiquing free speech arguments based on democracy and individuality and ﬁnding it unnecessary to accord primacy to either one); Steven Shiffrin, The First Amendment and Economic Regulation: Away from a General Theory of the First Amendment, 78 NW. U. L. REV. 1212 (1983) (reviewing and rejecting First Amendment theories based on promoting democratic processes and individual autonomy in favor of an eclectic approach to the First Amendment); Geoffrey R. Stone, Autonomy and Distrust, 64 U. COLO. L. REV. 1171 (1993) (advocating a theory of free speech that combines concern for autonomy with an instrumental distrust of government intervention); see also Daniel A. Farber & Philip P. Frickey, Practical Reason and the First Amendment, 34 UCLA L. REV. 1615 (1987) (reviewing the work of Baker and Redish and advocating the rejection of foundation theories in favor of resolving First Amendment disputes through practical reason). 387. Redish & Lippman, supra note 385, at 276; see also id. at 273–74 (arguing that “baseline free speech principles are so centrally intertwined with the values that underlie our political structure— popular sovereignty and self–determination—that abandoning the former would be impossible without weakening or abandoning the latter”). 386. REDISH, supra note 385, at 21; see also Redish & Kaludis, supra note 99, at 1121 (“Absent personal intellectual autonomy, the individual members of society cannot make truly free choices. Absent the individual citizens’ ability to make such free choices, the concept of a democratic society is rendered incoherent.”). 388. Post, supra note 377, at 1115. 389. Id. at 1128. 390. Id. at 1132. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT important instrumentally, that is, not because people have any intrinsic moral right to say what they wish, but because allowing them to do so will produce good effects for the rest of us.”378 Instead, Dworkin argues in favor of a First Amendment that values autonomy in general, and speech in particular, because it is an “essential and ‘constitutive’ feature of a just political society that government treat all its adult members . . . as responsible moral agents.”379 Dworkin’s dignitary vision of free speech has two aspects. The ﬁrst aspect focuses on people’s moral responsibility to “mak[e] up their own minds about what is good or bad in life or in politics, or what is true and false in matters of justice or faith.”380 The existence of this obligation implies that “[g]overnment insults its citizens, and denies their moral responsibility, when it decrees that they cannot be trusted to hear opinions that might persuade them to dangerous or offensive convictions.”381 The second aspect focuses on the more active role of the individual as speaker. From this perspective, moral responsibility carries with it “a responsibility not only to form convictions of one’s own, but to express these to others.”382 A government that “disqualiﬁes some people from exercising” their responsibility to form and communicate their own convictions “on the ground that their convictions make them unworthy participants” forfeits much of its claim to legitimacy.383 The vision of a “liberal society committed to individual moral responsibility” thus requires that the rejection of “any censor- ship on grounds of content,” even if it involves speech that we loathe.384 b. Democratic Theories of Autonomy. Other scholars have argued that respect for autonomy as an independent value is necessarily implicit in our commitment to democracy. For example, Martin Redish argues that “the concept of democ- racy itself is ultimately premised on a belief that individuals are capable of exercising control over decisions that directly affect their lives and morally deserve to do so.”385 Autonomy would thus seem to go hand in hand with democracy as a matter of deﬁnition because in the absence of a commitment to autonomy, it is difﬁcult to see why society would adopt a democratic system of 314 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT technique for arriving at better social judgments through democratic proce- dures,”392 arguing instead that freedom of expression at the same time encom- passed the value of self-fulﬁllment.393 Similarly, after offering a powerful synthesis of the existing descriptive and ascriptive visions of free speech, Richard Fallon argues that each is simultaneously fundamental, yet irreconcil- able and irreducible to a common metric. As a result, Fallon calls for a balancing approach.394 In so arguing, Fallon explicitly rejects Fiss’s and Sun- stein’s attempts to reject autonomy as a value. As Fallon notes, despite their best efforts, Fiss and Sunstein “cannot sensibly advocate enlightened democracy at the expense of autonomy; to do so would get the order of values backwards.”395 C. Edwin Baker advances the most multifaceted and nuanced hybrid argu- ment. For individuals, Baker views free speech in largely deontological terms. Like Dworkin, Baker argues that “the key ethical postulate” underlying the First Amendment is that “respect for individual integrity and autonomy requires the recognition that a person has the right to use speech to develop herself or to inﬂuence or interact with others in a manner that corresponds to her values.”396 As a result, Baker rejects as inherently illegitimate any collective practice that is inconsistent with treating each person as morally independent.397 C. Edwin Baker advances the most multifaceted and nuanced hybrid argu- ment. For individuals, Baker views free speech in largely deontological terms. Like Dworkin, Baker argues that “the key ethical postulate” underlying the First Amendment is that “respect for individual integrity and autonomy requires the recognition that a person has the right to use speech to develop herself or to inﬂuence or interact with others in a manner that corresponds to her values.”396 As a result, Baker rejects as inherently illegitimate any collective practice that is inconsistent with treating each person as morally independent 397 As a result, Baker rejects as inherently illegitimate any collective practice that is inconsistent with treating each person as morally independent.397 At the same time, Baker augments his dignitary vision of individual au- tonomy with arguments based in the democratic process that are reminiscent of Redish’s and Post’s. 399. Id. at 50; see also id. at 49 (reasoning that “the practices of democratic decision making . . . can often be understood as properly implementing equal respect for persons as autonomous agents”); C. Edwin Baker, The Media that Citizens Need, 147 U. PA. L. REV. 317, 327–28 (1998) (arguing that “popular participation or, at least, real opportunity for participation is crucial . . . to maintain a popular sense of government . . . . [This] legitimizing practice must include participatory democracy—only this process recognizes both people’s right to choose [autonomy] and people’s equality as to this ﬁght”). 396. C. EDWIN BAKER, HUMAN LIBERTY AND FREEDOM OF SPEECH 59 (1989). Baker elaborates as follows: “Respect for people as autonomous agents implies that people should be viewed as responsible for, and given maximal liberty in, choosing how to use their bodies to develop and express themselves; and should be given an equal right to try to inﬂuence the nature of their collective worlds.” Id. at 58–59. g 397. Id. at 48–49. 393. Id. at 4–7, 8–11, 14–15 (acknowledging both individual self-fulﬁllment and participation in decisionmaking as free speech values while declining to settle on a single foundation value). 394. Fallon, supra note 377, at 899–901. 398. Id. at 49. 400. BAKER, supra note 396, at 50. 392. THOMAS I. EMERSON, TOWARD A GENERAL THEORY OF THE FIRST AMENDMENT 14 (1966). 395. Id. at 884. THE GEORGETOWN LAW JOURNAL [Vol. 91:245 government in the ﬁrst place.386 Indeed, Redish argues: “Since a belief in societal self-determination underlies our entire political system and constitu- tional structure, however, we cannot reject that belief without simultaneously rejecting the American form of government. Such logic would necessarily constitute a rejection of the ﬁrst amendment, rather than an interpretation of it.”387 Robert Post offers a similar argument that focuses less on the direct relation- ship between autonomy and democracy and more on the role speech plays in the democratic process. In Post’s view, speech allows “democracy . . . to reconcile individual autonomy with collective self-determination” by “instill[ing] in citi- zens a sense of participation, legitimacy, and identiﬁcation” sufﬁcient to induce them to support the substantive outcomes of the political process even when they disagree with such outcomes.388 This constitutive vision of speech necessar- ily “presupposes that those participating in public discourse are free and autono- mous” because, without such autonomy, it is impossible to see how the public discourse could possibly mediate between the individual and the collective will.389 As a result, Post argues that democratically based theories of free speech entail an ascription of autonomy for its own sake, and all attempts to regulate speech on the grounds that individuals cannot fully participate in the political process “contradict the central premise of our democratic enterprise.”390 c. Hybrid Theories of Autonomy. It is particularly telling that even scholars who, like Fiss and Sunstein, are willing to treat the promotion of healthy democratic processes as a First Amendment value have generally regarded it as only one of several competing values.391 For example, Thomas Emerson re- jected the notion that freedom of expression amounted to nothing more than “a 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 315 401. Id. at 200–31; C. Edwin Baker, Turner Broadcasting: Content-Based Regulation of Persons and Presses, 1994 SUP. CT. REV. 57, 66–72. 402. See BAKER, supra note 396, at 225–49; Baker, supra note 401, at 80–81. 403. Baker, supra note 399, at 327–40. 404. FISS, LIBERALISM DIVIDED, supra note 24, at 13, 36–37. 405. Id. at 114. 406. Id. at 13. THE GEORGETOWN LAW JOURNAL Although Baker views individual speech in largely ascriptive terms, he refuses to extend this view to corporate speech on the grounds that liberty is a personal attribute that does not apply to collective institutions.401 However, Baker does make an exception for media corporations, in light of the unique role envisioned for the Fourth Estate by the Press Clause.402 With respect to the press, Baker argues that the government should promote what he calls “Com- plex Democracy,” which is an intermediate position that attempts to capture the best elements of what he calls “Liberal Pluralist Democracy,” which treats values as exogenous to politics and looks to the democratic process to mediate among the varying conceptions of the good, and what he terms “Republican Democracy” of the type advocated by Fiss and Sunstein.403 As the existence of these alternative theories demonstrates, the vision of free speech upon which Fiss and Sunstein rely is far from self-evident. Many theorists would object to their attempt to place democratic decisionmaking at the center of the First Amendment, and many of those who would not object would nonetheless repudiate the instrumental, descriptive vision of free speech that they propose. Equally telling is that most of those theorists who do not reﬂexively object to governmental attempts to regulate speech in ways that promote collective self-determination still recognize that the needs of the democratic process must be balanced against the needs of individual autonomy. The attractiveness of Fiss’s and Sunstein’s theories thus turns largely on their ability to come to grips with these autonomy-based visions of free speech. The balance of this section will examine Fiss’s and Sunstein’s attempts to do so. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Under this view, “[d]emocratic decision making gives each person the same potential say in results, a say that properly would represent the person’s autonomous choice or commitment.”398 Thus, the “normally accepted account of our constitutionalism” necessarily “treats certain values—human dignity, respect for individuals[’] equality and autonomy—as fundamental and directs that democracy must operate within the constraint of respect for these values.”399 The very existence of a democratic system of government, therefore, presupposes certain “fundamental constitutional restraints on democratic choice” that are based in autonomy.400 316 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 407. Id. at 15. 408. Id. at 146. 409. Robert Post, Equality and Autonomy in First Amendment Jurisprudence, 95 MICH. L. REV. 1517, 1525 (1997) (arguing that Fiss “cannot simply rule out a priori the independent constitutional value of individual autonomy”). 410. FISS, LIBERALISM DIVIDED, supra note 24, at 13. 411. Post, supra note 409, at 1526. 412. Id. at 1524. 413. Id. at 1527; see also Post, supra note 377, at 1115–16. 414. FISS, LIBERALISM DIVIDED, supra note 24, at 17–18. 415. Id. at 19–22. 2. Fiss on Autonomy It should now be clear that the democratically oriented vision of free speech upon which Fiss relies represents one of the most highly contested issues in free speech theory. Unfortunately, Fiss fails to offer much in the way of theoretical justiﬁcation for it. For the most part, Fiss simply posits his vision of free speech, at some points categorically averring that it represents the dominant perspec- tive404 and at other points calling it “almost axiomatic.”405 To the extent that Fiss engages autonomy-based arguments at all, he discards them in a somewhat conclusory manner. He simply declares that autonomy is “protected not because of its intrinsic value, . . . but rather as a means or instrument of collective self-determination.”406 Fiss later makes explicit what this argument already clearly implied: “In fact autonomy adds nothing, and if need be, might have to be sacriﬁced, to make certain that public debate is sufﬁciently rich to permit 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 409. Robert Post, Equality and Autonomy in First Amendment Jurisprudence, 95 MICH. L. REV. 1517, 1525 (1997) (arguing that Fiss “cannot simply rule out a priori the independent constitutional value of individual autonomy”). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 317 true collective self-determination.”407 To the extent that the speech people choose fails to reﬂect democratic values, little is lost by failing to respect that choice.408 Autonomy thus falls away as an independent value not by virtue of any sustained analysis, but rather as a byproduct of the central value that Fiss simply declares to be at the heart of the First Amendment. It seems quite problematic to dispose of one of the central questions of free speech theory through the use of a simple ipse dixit.409 Fiss’s refusal to engage the larger debate surrounding the role of autonomy in the First Amendment is made more troubling by his central commitment to democracy. As noted earlier, many scholars who have begun from the same starting point have found autonomy to be a necessary concomi- tant rather than a potential obstacle to democracy. Equally debatable is the particular vision of democracy embedded in Fiss’s work. By Fiss’s own admission, his vision of the way speech interacts with the political process is entirely listener-based, perhaps best captured in the pithy statement, “[w]e allow people to speak so others can vote.”410 As Robert Post has pointed out, this “offers a strikingly passive image of the democratic citizen”411 that does not take democracy’s participatory dimensions into ac- count. The omission is important because “[i]ndividual citizens can identify with the creation of a collective will only if they believe that collective decisionmaking is in some way connected to their own individual self- determination.”412 Active participation thus provides the means through which people reconcile themselves with collective decisions with which they person- ally disagree.413 In fact, Fiss’s summary rejection of autonomy is part of a far more fundamen- tal desire to reorder the basic constitutional relationship between the individual and the state. Fiss claims not to disturb the sharp dichotomy between state and citizen presupposed by classical liberalism.414 Upon closer inspection, it be- comes clear that this disclaimer cannot be taken at face value. Under the classic conception of liberalism, liberty is equated with freedom from governmental interference. Fiss’s vision of free speech demands more. He claims that the state bears an obligation to provide each individual with the means to exercise that liberty in a meaningful way.415 As a result, it is possible to construe Fiss’s argument as another strand of the 318 THE GEORGETOWN LAW JOURNAL [Vol. 421. See also infra subsection IV.B.1.c (discussing Fiss’s views on the state action doctrine). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 91:245 broader constitutional debates surrounding the vision of due process inspired by Charles Reich’s seminal article The New Property416 and the clash between formal and substantive equality. Indeed, at times, Fiss explicitly frames his work in precisely this manner.417 It thus seems apt to describe Fiss’s proposal as calling for a shift from what might be termed formal liberty to substantive liberty. That he would ﬁnd such a vision of free speech attractive is unsurpris- ing. He is an admirer of Goldberg v. Kelly,418 and he has also argued against interpreting the Equal Protection Clause simply as a limit against governmental interference.419 Situating his work in this manner only underscores its weak- ness, however. As Fiss himself recognizes, the Court’s equal protection and due process jurisprudence has already rejected the vision of the relationship between the individual and the state that he proposes.420 Although it remains open for Fiss to propose some basis for a different outcome with respect to free speech, at this point he has not done so.421 f ( g 422. See infra notes 424–27, 439–41, 445–47 and accompanying text. 420. On equal protection, see FISS, LIBERALISM DIVIDED, supra note 24, at 99. On due process, see Owen Fiss, A Life Lived Twice, 100 YALE L.J. 1117, 1123 (1991); Tributes to Robert M. Cover, supra note 418, at 1720. f , , 423. See infra notes 450–52 and accompanying text. g y g g 419. FISS, LIBERALISM DIVIDED, supra note 24, at 99; Owen M. Fiss, The Forms of Justice, 93 HARV. L. REV. 1, 23 (1979); Panel Presentation: Equality in Education, 74 F.R.D. 269, 276–81 (1977) (remarks of Owen Fiss). 416. Charles A. Reich, The New Property, 73 YALE L.J. 733 (1964). 418. 397 U.S. 254, 265 (1970) (suggesting that the Due Process Clause might recognize an inherent right to a basic subsistence level of governmental support); see also Owen M. Fiss, Reason in All Its Splendor, 56 BROOK. L. REV. 789, 792, 804 (1990) (lauding Goldberg v. Kelly as a “triumph of . . . substantive rationality,” “a magniﬁcent achievement,” and “a monument to our own little enlighten- ment”); Tributes to Robert M. Cover, 96 YALE L.J. 1699, 1720 (1987) (remarks of Owen Fiss) (calling Goldberg v. Kelly “the culmination of the Golden Age of American Law” and “bemoaning its demise”). , p y, ( ) 417. FISS, LIBERALISM DIVIDED, supra note 24, at 35–36, 99–102. For a discussion about Sunstein’s attempt to raise a similar argument, see infra notes 660–61 and accompanying text. 8, 0 421. See also infra subsection IV.B.1.c (discussing Fiss’s views on the state action doctrine). 416. Charles A. Reich, The New Property, 73 YALE L.J. 733 (1964). 417. FISS, LIBERALISM DIVIDED, supra note 24, at 35–36, 99–102. For a discussion about Sunstein’s attempt to raise a similar argument, see infra notes 660–61 and accompanying text. 418. 397 U.S. 254, 265 (1970) (suggesting that the Due Process Clause might recognize an inherent right to a basic subsistence level of governmental support); see also Owen M. Fiss, Reason in All Its Splendor, 56 BROOK. L. REV. 789, 792, 804 (1990) (lauding Goldberg v. Kelly as a “triumph of . . . substantive rationality,” “a magniﬁcent achievement,” and “a monument to our own little enlighten- ment”); Tributes to Robert M. Cover, 96 YALE L.J. 1699, 1720 (1987) (remarks of Owen Fiss) (calling Goldberg v. Kelly “the culmination of the Golden Age of American Law” and “bemoaning its demise”). 419. FISS, LIBERALISM DIVIDED, supra note 24, at 99; Owen M. Fiss, The Forms of Justice, 93 HARV. L. REV. 1, 23 (1979); Panel Presentation: Equality in Education, 74 F.R.D. 269, 276–81 (1977) (remarks of Owen Fiss). 420. On equal protection, see FISS, LIBERALISM DIVIDED, supra note 24, at 99. On due process, see Owen Fiss, A Life Lived Twice, 100 YALE L.J. 1117, 1123 (1991); Tributes to Robert M. Cover, supra note 418, at 1720. 421. See also infra subsection IV.B.1.c (discussing Fiss’s views on the state action doctrine). 422. See infra notes 424–27, 439–41, 445–47 and accompanying text. 423. See infra notes 450–52 and accompanying text. 432. As Jack Balkin has famously quipped, “Sunstein’s ‘Madisonian’ theory of the First Amendment is about as Madisonian as Madison, Wisconsin: It is a tribute to a great man and his achievements, but bears only a limited connection to his views.” J.M. Balkin, Populism and Progressivism as Constitu- 424. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at xvii (quoting 17 JAMES MADISON, Report of 1800, in THE PAPERS OF JAMES MADISON 341 (David Mattern et al., 1991). f y, , ( ) 431. See John O. McGinnis, The Once and Future Property-Based Vision of the First Amendment, 63 U. CHI. L. REV. 49 (1996); John O. McGinnis, The Partial Republican, 35 WM. & MARY L. REV. 1751, 1760–62 (1994). 430. See David A.J. Richards, Constitutional Legitimacy, the Principle of Free Speech, and the Politics of Identity, 74 CHI.-KENT L. REV. 779, 796–800 (1999). 428. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at xvii. p f , ( , ) 425. 274 U.S. 357, 372 (1927) (Brandeis, J., concurring), discussed in SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 26–28. 426. 395 U.S. 367 (1969), discussed in SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 48–49. 429. See, e.g., DWORKIN, supra note 378, at 200. 427. 376 U.S. 254 (1964), discussed in SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 38–41. 3. Sunstein on Autonomy Sunstein offers a somewhat more elaborate justiﬁcation for rejecting au- tonomy as a free speech value. His argument in favor of a First Amendment focused on promoting democratic processes is quite complex, invoking the intellectual precepts of history, communitarianism, and practical reason.422 In addition, he criticizes the coherence of the willingness of autonomy-based visions of free speech to regard individual preferences as pre-political.423 In this section, I will critically assess each of these arguments. A close analysis reveals that Sunstein’s historical arguments lack a substantial foundation. Even more problematic is his invocation of communitarianism and practical reason because the manner in which he applies each theory has the effect of simply assuming away the conﬂict between autonomy and democracy. Most troubling is his attempt to undermine preferences, which I ﬁnd to be fundamentally inconsistent 2003] 319 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 426. 395 U.S. 367 (1969), discussed in SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 48–49. p f ( ) 425. 274 U.S. 357, 372 (1927) (Brandeis, J., concurring), discussed in SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 26–28. 428. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra 438. FISS, LIBERALISM DIVIDED, supra note 24, at 20; see also FISS, IRONY OF FREE SPEECH, supra note 24, at 72 (characterizing Red Lion as a “stray”). In his later work, Fiss ﬁnds an endorsement of Red Lion in Justice Breyer’s concurrence in the judgment in Turner II, as well as general support for his approach in the plurality opinion in Denver and Justice Stevens’s dissent in Arkansas Educational Television Commission v. Forbes, 523 U.S. 666 (1998). Fiss, supra note 24, at 1220–26, 1229–31, 1234–35. It should be noted that none of the opinions that Fiss cites commanded a majority of the Court. A close reading of those opinions, moreover, reveals that they are more consistent with the hybrid approach discussed above, which accords weight to both autonomy and instrumental concerns, than it is with the purely instrumental approach that Fiss advocates. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT with most democratic forms of government. None of these approaches offers a satisfactory reconciliation of the conﬂict between the individual and the collec- tive will, which represents perhaps the central problem lying at the heart of liberal and democratic theory. a. The Role of History. Sunstein initially attempts to rebut theories that view the First Amendment as a bulwark that protects individual autonomy by invok- ing history. He argues that viewing the First Amendment in terms of promoting collective self-determination follows from James Madison’s recognition of the need for people to “freely examine public characters and measures” if they are to exercise their democratic obligations in a meaningful manner.424 He ﬁnds further support in such notable opinions as Justice Brandeis’s concurrence in Whitney v. California425 as well as the opinions of the Court in Red Lion Broadcasting Co. v. FCC426 and New York Times Co. v. Sullivan.427 A review of the historical record reveals that this so-called “Madisonian” vision of the First Amendment, which Sunstein uses as a trope throughout his work, is more Sunstein’s creation than Madison’s. Sunstein’s historical discus- sion consists of a single quotation from Madison, without any discussion of Madison’s broader writings.428 An examination of the scholarship studying Madison’s views on free speech reveals that Sunstein’s claim is quite controver- sial. Although some share Sunstein’s vision,429 others have interpreted Madi- son’s views on free speech as ﬁtting better with the deontological approach to free speech discussed above.430 In addition, another burgeoning branch of the commentary argues that Madison espoused a view of free speech closely related to the Lockean theory of property rights.431 In either case, the result is a view that is much more attuned with the protection of speech as an individual or natural right than it is with the instrumental promotion of democracy. The invocation of a single line of Madison’s writings is thus insufﬁcient to establish the principle that Sunstein seeks.432 320 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL Nor does the series of landmark decisions cited by Sunstein enhance the historical pedigree of his theory. As an initial matter, his reliance on precedent is somewhat curious in light of his recognition that the judicial consensus gener- ally supports an autonomy-oriented vision of the First Amendment.433 In any event, when read closely, the opinions that Sunstein cites fall short of establish- ing the primacy of deliberative democracy as a First Amendment value. 436. Compare DWORKIN, supra note 378, at 202–09 (criticizing Justice Brennan’s “almost exclusive [reliance] on the instrumental justiﬁcation in his opinion” because it limited First Amendment protec- tion to “cases involving libel of ‘public ofﬁcials’ rather than extending protection to all libel defen- dants”), with Kalven, supra note 373, at 209 (praising the New York Times Court for “returning to the essence of the First Amendment . . . found in its limitations on seditious libel” and being “carried along by a momentum of insight about the democratic necessities of free speech”); see also William J. Brennan, Jr., The Supreme Court and the Meiklejohn Interpretation of the First Amendment, 79 HARV. L. REV. 1, 14–16 (1965). 437. Farber & Frickey, supra note 391, at 1634–35. 435. See DWORKIN, supra note 378, at 201; see also Vincent Blasi, The First Amendment and the Ideal of Civic Courage: The Brandeis Opinion in Whitney v. California, 29 WM. & MARY L. REV. 653, 684–85 (1988) (offering an alternative reading of Brandeis’s concurrence in Whitney). 433. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 3–5. tional Categories, 104 YALE L.J. 1935, 1955 (1995) (reviewing SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23). EIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 3–5 p ) 433. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 3–5. 433. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 3 5. 434. Whitney v. California, 274 U.S. 357, 372 (1927) (Brandeis, J., concurring). tional Categories, 104 YALE L.J. 1935, 1955 (1995) (reviewing SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23). 433. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 3–5. 434 Whitney v California 274 U S 357 372 (1927) (Brandeis J concurring) tional Categories, 104 YALE L.J. 1935, 1955 (1995) (reviewing SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23). 433. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 3–5. 434. Whitney v. California, 274 U.S. 357, 372 (1927) (Brandeis, J., concurring). 435. See DWORKIN, supra note 378, at 201; see also Vincent Blasi, The First Amendment and the Ideal of Civic Courage: The Brandeis Opinion in Whitney v. California, 29 WM. & MARY L. REV. 653, 684–85 (1988) (offering an alternative reading of Brandeis’s concurrence in Whitney). 436. Compare DWORKIN, supra note 378, at 202–09 (criticizing Justice Brennan’s “almost exclusive [reliance] on the instrumental justiﬁcation in his opinion” because it limited First Amendment protec- tion to “cases involving libel of ‘public ofﬁcials’ rather than extending protection to all libel defen- dants”), with Kalven, supra note 373, at 209 (praising the New York Times Court for “returning to the essence of the First Amendment . . . found in its limitations on seditious libel” and being “carried along by a momentum of insight about the democratic necessities of free speech”); see also William J. Brennan, Jr., The Supreme Court and the Meiklejohn Interpretation of the First Amendment, 79 HARV. L. REV. 1, 14–16 (1965). 439. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 19. It should be emphasized that Sunstein regards the deliberative process as an integral part of the means by which substantive values are selected. The implication is that one cannot bypass the democratic process and derive moral truth directly. The difference in emphasis is critical. As Martin Redish and Gary Lippman have pointed out, “[i]f one believes in the imposition of externally derived moral truth through governmental decisionmak- ing, the concept of democracy is rendered at best a nuisance and at worst a serious social harm.” Redish & Lippman, supra note 385, at 278. pp p 440. Cass R. Sunstein, Beyond the Republican Revival, 97 YALE L.J. 1539, 1550, 1554 (1988) (footnote omitted). 442. See Post, supra note 377, at 1115 (calling any postulated fusion of the individual and collective wills “unconvincing under modern conditions of heterogeneity”). 441. See Baker, supra note 399, at 334; Stephen A. Gardbaum, Broadcasting, Democracy, and the Market, 82 GEO. L.J. 373, 385 (1993). 443. Gardbaum, supra note 441, at 386–87. 444. Id. at 387. 443. Gardbaum, supra note 441, at 386–87. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 321 to communitarian principles. He posits that when political processes are function- ing properly, the interests of the collective and the individual will tend to be the same. Indeed, Sunstein suggests that so long as the machinery of democracy is in good working order, it may be appropriate to deﬁne the outcome of the deliberative process as “political truth.”439 Sunstein’s earlier writings provide a more fulsome elaboration of this theme: The republican commitment to universalism amounts to a belief in the possibility of mediating different approaches to politics, or different con- ceptions of the public good, through discussion and dialogue. The pro- cess of mediation is designed to produce substantively correct outcomes, understood as such through the ultimate criterion of agreement among politi- cal equals.440 Thus, in Sunstein’s perfect polity, there is no conﬂict between autonomy and the needs of the democratic process because, in a properly constructed delibera- tive democracy, these interests tend to converge.441 The communitarian aspects of Sunstein’s argument thus propose to solve the conﬂict between his theory and autonomy simply by positing the convergence of the individual and the collective will. On a theoretical level, Sunstein assumes away the conﬂict with autonomy in much the same way as does Fiss. On a more empirical level, the heterogeneity of modern society provides ample reason to doubt that the consensus that Sunstein envisions will in fact emerge.442 The communitarian aspects of Sunstein’s argument thus propose to solve the conﬂict between his theory and autonomy simply by positing the convergence of the individual and the collective will. On a theoretical level, Sunstein assumes away the conﬂict with autonomy in much the same way as does Fiss. On a more empirical level, the heterogeneity of modern society provides ample reason to doubt that the consensus that Sunstein envisions will in fact emerge.442 It is not even clear, moreover, that such consensus would be normatively desirable. This is because “consensus is not generally the condition of au- tonomy; rather, autonomy is more typically exhibited in divergence of substan- tive . . . views” in which different people “afﬁrm very different conceptions of the public and private good.”443 Under this view, dissensus “is the goal and characteristic symptom of a genuinely free society” rather than “a problem in need of . . . regulation.”444 It is not even clear, moreover, that such consensus would be normatively desirable. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Al- though there are clearly strains of Sunstein’s vision in Justice Brandeis’s concurrence in Whitney, the opinion’s emphasis on the “belie[f] that the ﬁnal end of the state was to make men free to develop their faculties” and that the Founders “valued liberty both as an end and as a means”434 is far more suggestive of a hybrid approach that seeks to balance the interests of autonomy and democracy than accord primacy to the latter.435 The Court’s opinion in New York Times Co. v. Sullivan suffers from similar limitations. Although the opinion has been both lauded and criticized for adopting the type of democracy- enhancing instrumental approach that Sunstein favors,436 other commentators have disagreed, noting that the opinion depended as much on an analogy to seditious libel and concerns about a possible “chilling effect.”437 In the end, the only authority that clearly stands in Sunstein’s corner is Red Lion, and even Fiss has conceded that that decision represents “something of a freak” that has never grown to cover other media.438 b. The Turn to Communitarianism. Sunstein also attempts to resolve the conﬂict between autonomy and the needs of the democratic process by turning 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 445. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 298 n.13 (citing RAWLS, supra note 377). 446. Id. at 133. 447. Id. at 133, 141, 142–43, 148, 175–76. 448. Id. 449. See Farber & Frickey, supra note 391, at 1641–43. 450. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 19–20, 73–74. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT This is because “consensus is not generally the condition of au- tonomy; rather, autonomy is more typically exhibited in divergence of substan- tive . . . views” in which different people “afﬁrm very different conceptions of the public and private good.”443 Under this view, dissensus “is the goal and characteristic symptom of a genuinely free society” rather than “a problem in need of . . . regulation.”444 322 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 c. The Appeal to Practical Reason. Sunstein further seeks to support his democratically oriented view of the First Amendment by invoking the methodol- ogy of practical reason. Sunstein describes his methodology as similar to the notion of reﬂective equilibrium developed by John Rawls,445 in which “theory [is] adjusted to conform to [considered] judgments, and vice versa, until we reach a state of equilibrium.”446 He employs this approach to reject a number of countervailing theories, including autonomy, as inconsistent with certain consid- ered judgments that he regards as unquestionably correct.447 Although Sunstein’s statement of the methodology of practical reason is satisfactory, his application of it leaves much to be desired. Adherents of practical reason would doubtless ﬁnd the level of abstraction at which Sunstein speaks and his resort to ﬁrst principles quite troublesome. There is, moreover, little reﬂection of the Rawlsian reﬂective equilibrium in Sunstein’s application. Rather than oscillating back and forth between the general and the speciﬁc until fusion is achieved, Sunstein’s approach treats certain preferred case outcomes as ﬁxed and uses them to support or invalidate particular theories.448 Such an Although Sunstein’s statement of the methodology of practical reason is satisfactory, his application of it leaves much to be desired. Adherents of practical reason would doubtless ﬁnd the level of abstraction at which Sunstein speaks and his resort to ﬁrst principles quite troublesome. There is, moreover, little reﬂection of the Rawlsian reﬂective equilibrium in Sunstein’s application. y, p , 450. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 19–20, 73–74. 449. See Farber & Frickey, supra note 391, at 1641–43. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Rather than oscillating back and forth between the general and the speciﬁc until fusion is achieved, Sunstein’s approach treats certain preferred case outcomes as ﬁxed and uses them to support or invalidate particular theories.448 Such an approach is hardly faithful to the notions of pragmatism, which rely on sensitiv- ity to complex webs of values and a search for the best legal answer in light of both history and context.449 In addition, his methodology makes his substantive results little more than a product of which case outcomes he chooses to hold constant and which case outcomes he chooses to critique. It would be as easy to hold the Court’s well-established hostility towards content-based regulation constant and use that insight to invalidate Sunstein’s theory. As a result, Sunstein’s turn to pragmatism amounts to little more than the assumption of the results that his theory is trying to prove. d. The Assault on Preferences. Finally, Sunstein attempts to refute the impor- tance of autonomy by attacking the notion that individual preferences are pre-political. Instead, Sunstein argues that individual preferences are the prod- uct of the existing set of social and legal rules. Because preferences are largely the product of the speech that already exists and the social structure that created it, Sunstein argues that any attempt to justify current outcomes in terms of those preferences is ultimately circular, in that it simply tends to validate the status quo.450 True autonomy lies in the realization of those preferences that individu- als would have held had they been exposed to higher quality programming that incorporated a broader range of viewpoints. Consequently, it is not necessarily a violation of autonomy to deviate from current preferences because “the inclu- 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 323 sion of better options . . . does not displace a freely produced desire.”451 On the contrary, “[i]f more and better choices are made available, the outcome may well be to promote autonomy, rightly understood.”452 This argument touches on an issue that has long pervaded liberal political theory. It inheres in the work of John Rawls, who turns to idealized choices made behind a “veil of ignorance” to justify the imposition of outcomes consistent with the collective good without violating the central tenets of democracy and liberalism.453 The work of John Stuart Mill exhibits a deep ambivalence over the choice between actual and idealized preferences. g g p ) 457. ISAIAH BERLIN, Two Concepts of Liberty, in FOUR ESSAYS ON LIBERTY 118, 131–34 (1969), cited in Fallon, supra note 377, at 886 n.65. 456. Robert C. Post, Racist Speech, Democracy, and the First Amendment, 32 WM. & MARY L. REV. 267, 284 (1991) (arguing that “the state undermines the raison d’eˆtre of its own enterprise to the extent that it itself coercively forms the ‘autonomous wills’ that democracy seeks to reconcile into public opinion”); see also Post, supra note 377, at 1133 (“[O]ne cannot but be struck by the sharp anomaly of regulating democratic elections on the premise that voters are not autonomous and free.”). p 454. JOHN STUART MILL, UTILITARIANISM 8–17 (George Sher ed., Hackett Publ’g Co. 1979) (1861). See generally Robin L. West, Liberalism Rediscovered: A Pragmatic Deﬁnition of the Liberal Vision, 46 U. PITT. L. REV. 673, 689–90 (1985). , ( ) 455. MILL, On Liberty, supra note 363, at 124, quoted in Post, supra note 377, at 1131 n.90. 458. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 91. 453. RAWLS, supra note 377, at 136–42. 451. Id. at 74. 452. Id. 453. RAWLS, supra note 377, at 136–42. 454. JOHN STUART MILL, UTILITARIANISM 8–17 (George Sher ed., Hackett Publ’g Co. 1979) (1861). See generally Robin L. West, Liberalism Rediscovered: A Pragmatic Deﬁnition of the Liberal Vision, 46 U. PITT. L. REV. 673, 689–90 (1985). 455. MILL, On Liberty, supra note 363, at 124, quoted in Post, supra note 377, at 1131 n.90. 456. Robert C. Post, Racist Speech, Democracy, and the First Amendment, 32 WM. & MARY L. REV. 267, 284 (1991) (arguing that “the state undermines the raison d’eˆtre of its own enterprise to the extent that it itself coercively forms the ‘autonomous wills’ that democracy seeks to reconcile into public opinion”); see also Post, supra note 377, at 1133 (“[O]ne cannot but be struck by the sharp anomaly of regulating democratic elections on the premise that voters are not autonomous and free.”). 457. ISAIAH BERLIN, Two Concepts of Liberty, in FOUR ESSAYS ON LIBERTY 118, 131–34 (1969), cited in Fallon, supra note 377, at 886 n.65. 458. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 91. Liberty, supra note 363, at 124, quoted in Post, supra note 377, at 1131 n.90. 451. Id. at 74. 452. Id. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT At times, Mill seems to deﬁne liberty as the realization of the desires that a person would hold if exposed to a more complete range of experiences.454 At other times, Mill strongly criticized this perspective, arguing that it essentially treated the labor- ing classes as children or savages unﬁt for the privileges of freedom. As a result, Mill contended that such an approach was inconsistent with the principles of freedom and democracy.455 That said, the attack on preferences strikes me as profoundly antidemocratic. If one regards an individual’s personality as a social construct subject to improvement by the state, there would seem to be little reason for the state to accord any particular respect to the outcomes of the democratic process.456 Indeed, there is a disturbingly illiberal overtone to Sunstein’s critique of prefer- ences. As Isaiah Berlin has argued, acknowledging a distinction between what people actually want and what they ought to want risks justifying coercion in the name of the “true” or “real” interests of the person being coerced.457 It also opens the door to the imposition of an eruditiocracy, in which the preferences of the elite classes are simply imposed on others. This danger does not appear to trouble Sunstein. He notes, “Of course it is possible or even likely that the well-educated will disproportionately enjoy high-quality broadcasting. But this is precisely because they have been educated to do so, and high-quality educa- tion is not something to be disparaged.”458 It is this aspect that has led some 324 [Vol. 91:245 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL critics to condemn Sunstein’s proposal as “indefensible.”459 This dilemma can be illustrated by the following thought experiment.460 Suppose that a group of voters cast their votes in a racially discriminatory manner and that this group is sufﬁciently large to control the outcome of an election. Suppose further that their racist views result directly from the narrow- ness of the viewpoints to which they have been exposed. It is hard to imagine that any democratic system would sanction disregarding the actual votes cast and entering results that more accurately reﬂect the voters’ “true” or idealized preferences. This is so even though the normative value at issue—the elimina- tion of racial discrimination—represents perhaps the most morally attractive and strongest justiﬁcation for intervention. 459. See Burt Neuborne, Blues for the Left Hand: A Critique of Cass Sunstein’s Democracy and the Problem of Free Speech, 62 U. CHI. L. REV. 423, 443 (1995). 460. I borrow the outlines of this hypothetical example from Ronald J. Krotoszynski, Jr., Back to the Briarpatch: An Argument in Favor of Constitutional Meta-Analysis in State Action Determinations, 94 MICH. L. REV. 302, 323 n.109 (1995). 461. See Fallon, supra note 377, at 885. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT If democratic theory will not allow us to overturn outcomes of elections in the name of promoting idealized preferences over actual preferences, the close connection between speech and votes makes it hard to see why such intervention would be permitted with respect to speech. Even when addressed at the level of speech rather than outcomes, this type of intervention still smacks of the Rousseauean notion of being forced to be free.461 Lastly, any theory that seeks to promote idealized preferences will confront severe implementation problems. It must offer some basis for identifying those preferences with sufﬁcient conﬁdence to justify subjecting individuals to such coercion. It is the difﬁculties surrounding any attempt to articulate such a theory that is the focus of the next section of this Article. 461. See Fallon, supra note 377, at 885. 1. Implementation of the Fiss-Sunstein Approach 1. Implementation of the Fiss-Sunstein Approach As noted earlier, Fiss and Sunstein offer only minimal guidance as to the substance of their idealized vision of the democratic process.462 Fiss invokes the concept of “robust public debate” as if the concept were self-explanatory. Simply put, the validity of government intervention turns on “whether the intervention in fact enriches rather than impoverishes the debate.”463 His failure to offer much elaboration on what such a concept requires does not trouble him. He ﬂatly states that a benchmark deﬁnition specifying what robust public debate requires would be helpful but is not strictly necessary.464 Fiss recognizes that fashioning particular remedies will present “no easy question.”465 This is particu- to offer much elaboration on what such a concept requires does not trouble him. He ﬂatly states that a benchmark deﬁnition specifying what robust public debate requires would be helpful but is not strictly necessary.464 Fiss recognizes that fashioning particular remedies will present “no easy question.”465 This is particu- larly so because of the omnipresent danger that regulation designed to enrich public debate may in fact impoverish it.466 As a result, the problems of balancing the various interests are likely to be “excruciating.”467 Still, Fiss ultimately places his faith in “the deliberate and incremental methods of the law” and takes comfort in “the old notion that it is easier to identify an injustice than to explain what is justice.”468 to offer much elaboration on what such a concept requires does not trouble him. 472. Id. at 21; see also id. (noting that the appropriate remedy for the public’s failure to pay sufﬁcient attention to public affairs “is far from clear”). 471. Id. at 123, 130–37. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 325 FCC. Upon close analysis, it becomes clear that none of them provide much assistance in determining the needs of a properly functioning democratic pro- cess. 470. Id. at 20–21. 469. See SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 20 (asserting that conditions for a properly functioning deliberative democracy include “adequate information; a norm of political equal- ity . . . ; an absence of strategic manipulation of information, perspectives, processes, or outcomes; and a broad public orientation toward reaching right answers rather than serving self-interest”). 470 Id 20 21 B. CRITIQUES FROM INSIDE THE MODEL: TURNING THEORY INTO PRACTICE To this point, I have focused on a theoretical critique that is largely external to what Fiss and Sunstein propose. This Section, in contrast, takes Fiss’s and Sunstein’s theoretical assumptions as given in an attempt to offer an internal critique of their work. It begins by exploring a number of typical constitutional questions that would naturally arise out of an attempt to extend the Broadcast Model to another medium of communication. I conclude that the framework that Fiss and Sunstein have proffered is too incompletely articulated to support a principled theory of free speech. The primary problem is that their theories fail to offer any basis for making the types of decisions and tradeoffs necessar- ily required by their democratically oriented vision of free speech. This section then considers three institutional models that Fiss and Sunstein suggest can provide some guidance as to the type of communications environ- ment needed to promote collective self-determination: the New England town meeting, Jeffersonian democracy, and the history of broadcast regulation by the 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 468. Id. at 26. 467. Id. at 25. 462. See supra notes 345, 359–61 and accompanying text. 463. FISS, LIBERALISM DIVIDED, supra note 24, at 26. 464. Id. at 16. 465. Id. 466. Id. at 23–24. 467. Id. at 25. 468. Id. at 26. 469. See SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 20 (asserting that conditions for a properly functioning deliberative democracy include “adequate information; a norm of political equal- ity . . . ; an absence of strategic manipulation of information, perspectives, processes, or outcomes; and a broad public orientation toward reaching right answers rather than serving self-interest”). 470. Id. at 20–21. 471. Id. at 123, 130–37. 472. Id. at 21; see also id. (noting that the appropriate remedy for the public’s failure to pay sufﬁcient attention to public affairs “is far from clear”). 463. FISS, LIBERALISM DIVIDED, supra note 24, at 26. 462. See supra notes 345, 359–61 and accompanying text. 1. Implementation of the Fiss-Sunstein Approach He ﬂatly states that a benchmark deﬁnition specifying what robust public debate requires would be helpful but is not strictly necessary.464 Fiss recognizes that fashioning particular remedies will present “no easy question.”465 This is particu- larly so because of the omnipresent danger that regulation designed to enrich public debate may in fact impoverish it.466 As a result, the problems of balancing the various interests are likely to be “excruciating.”467 Still, Fiss g p p y q p larly so because of the omnipresent danger that regulation designed to enrich public debate may in fact impoverish it.466 As a result, the problems of balancing the various interests are likely to be “excruciating.”467 Still, Fiss ultimately places his faith in “the deliberate and incremental methods of the law” and takes comfort in “the old notion that it is easier to identify an injustice than to explain what is justice.”468 For his part, Sunstein offers a bit more guidance as to what his idealized vision of the democratic process requires. Although in a perfect world Sunstein would insist on a more complete set of preconditions,469 in the end he identiﬁes two minimum requirements for a deliberative democracy to function properly: (1) a broad and deep attention to public issues and (2) public exposure to an appropriate diversity of views.470 In addition, Sunstein argues that the commit- ment to deliberative democracy requires extending a greater degree of protec- tion to political speech than to nonpolitical speech.471 Although such guidance is helpful, Sunstein recognizes that it leaves many questions unanswered and that the particular remedies to be applied “remain[ ] obscure.”472 As a result, Sunstein calls for the government to be “frankly , 472. Id. at 21; see also id. (noting that the appropriate remedy for the public’s failure to pay sufﬁcient attention to public affairs “is far from clear”). 326 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 experimental” in working out the details.473 While the remedies may be unclear, what is clear to Sunstein is that “the current system is worse than imperfect; it creates extremely serious obstacles to a well-functioning system of free expres- sion.”474 As a result, if a particular reading of the First Amendment bars such experimentation, then it is the First Amendment that must give way.475 The problem with such ad hoc approaches is that they fail to provide much guidance as to how to implement the systems that they envision. y 478. See generally Kalven, supra note 373; Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 IND. L.J. 1 (1971); Vincent Blasi, The Checking Value in First Amendment Theory, 1977 AM. B. FOUND. RES. J. 523; Lillian BeVier, The First Amendment and Political Speech: An Inquiry into the Substance and Limits of Principle, 30 STAN. L. REV. 299 (1978). q y f p 479. See supra notes 345, 351, 365–68 and accompanying text. 473. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221. 474. Id. at 222; see also SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 89. 475. See SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 81 (arguing that “the First Amendment should not operate as a talismanic or reﬂexive obstacle to our efforts to experiment with different strategies for achieving free speech goals”); SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 220 (asking rhetorically why the Constitution should “bar a democratic decision to experiment with new methods for achieving their Madisonian goals”). 476. Fallon, supra note 377, at 884. 477. Id. at 885; see also Post, supra note 377, at 1112 (asking, without “a standard by which the quality of the thinking process of the community can be assessed[,] [h]ow . . . could it be known whether public discourse is actually meeting the common needs of all the members of the body politic?”); Redish & Kaludis, supra note 99, at 1109 (arguing that Fiss and Sunstein “must establish some workable standard by which . . . redistributive decisions are to be made”). 478. See generally Kalven, supra note 373; Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 IND. L.J. 1 (1971); Vincent Blasi, The Checking Value in First Amendment Theory, 1977 AM. B. FOUND. RES. J. 523; Lillian BeVier, The First Amendment and Political Speech: An Inquiry into the Substance and Limits of Principle, 30 STAN. L. REV. 299 (1978). 479. See supra notes 345, 351, 365–68 and accompanying text. 476. Fallon, supra note 377, at 884. 477. Id. at 885; see also Post, supra note 377, at 1112 (asking, without “a standard by which the quality of the thinking process of the community can be assessed[,] [h]ow . . . could it be known whether public discourse is actually meeting the common needs of all the members of the body politic?”); Redish & Kaludis, supra note 99, at 1109 (arguing that Fiss and Sunstein “must establish some workable standard by which . . . redistributive decisions are to be made”). 473. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221. 476. Fallon, supra note 377, at 884. 487. See Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programming, Eighth Annual Report, 17 F.C.C.R. 1244, 1288 ¶ 96 (2002) (estimating VCR ownership at 90%); 2002 INTERNATIONAL TELEVISION & VIDEO ALMANAC 14 (47th ed., Tracy Stevens ed., 2002) (estimating VCR ownership at 86%); U.S. CENSUS BUREAU, STATISTICAL ABSTRACT OF THE UNITED STATES 1998, at 573 480. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 84; see also SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221 (“There is a strong case for public promotion of high-quality programming for children or for incentives, imposed by government on broadcasters, to provide such programming.”). 481. See Newton M. Minow, Address to the National Association of Broadcasters (May 9, 1961) (commonly known as the “Vast Wasteland” speech), reprinted in MINOW & LAMAY, supra note 52, at 185–96. 485. Frank Ahrens, That’s All Folks; Saturday Morning Tradition Fades as Networks Bow Out on Kids’Shows, SEATTLE TIMES, Jan. 26, 2002, at A3. 1. Implementation of the Fiss-Sunstein Approach As Richard Fallon has pointed out, when speech is viewed in purely instrumental terms, “[c]laims of positive liberty are often at stake on both sides of debates about regulating speech.”476 As a result, any such theory “would need to specify how competing claims to positive liberty ought to be weighed.”477 The need for some articulation of this vision is further underscored by the use of the same analytical starting point as Fiss and Sunstein by scholars as diverse as Harry Kalven, Robert Bork, Vincent Blasi, and Lillian BeVier to develop radically different visions of free speech.478 A brief review of three constitutional issues likely to be raised by any attempt to extend the Broadcast Model to another medium should help illustrate just how difﬁcult these implementation issues will be. a. Afﬁrmative Programming Obligations. One of the centerpieces of Fiss’s and Sunstein’s proposals is the continuation and expansion of afﬁrmative programming obligations. As noted earlier, both tentatively propose a regulatory system that includes requiring greater and more appropriate coverage of politi- cal campaigns, greater support for children’s television, and the reinstitution of compulsory rights of reply.479 The difﬁculty is that in the absence of a clear articulation of what the democratic process requires, we lack a benchmark for determining whether and how much of any particular type of programming is needed. Consider, for example, the call for greater support for children’s television. 1. Implementation of the Fiss-Sunstein Approach 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 327 Sunstein denigrates the current system as “clearly unsuccessful in terms of both quality and quantity” and avers that a “strong case” exists for requiring broadcast- ers to do more.480 The long tradition of decrying the quality of television programming gives this assertion a certain surface credibility.481 But a more critical examination reveals that Sunstein’s claim may not be as plausible as it ﬁrst seems.482 Many observers insist that the quantity and quality of educational programming available to children has never been better.483 The Corporation for Sunstein denigrates the current system as “clearly unsuccessful in terms of both quality and quantity” and avers that a “strong case” exists for requiring broadcast- ers to do more.480 The long tradition of decrying the quality of television programming gives this assertion a certain surface credibility.481 But a more critical examination reveals that Sunstein’s claim may not be as plausible as it ﬁrst seems.482 Many observers insist that the quantity and quality of educational programming available to children has never been better.483 The Corporation for Public Broadcasting has increasingly redirected its resources towards children’s television. As a result, PBS has been able to augment its offerings with a growing cadre of new shows, including such hits as “Between the Lions” and “Zoboomafoo.” In addition, channels available on cable and other MVPDs, such as Nickelodeon, the Disney Channel, the ABC Family Channel, and new cable offerings such as Noggin, which is a joint venture between Nickelodeon and the Children’s Television Workshop, are providing an increasing variety of high-quality educational programming targeted at children, including such criti- cally acclaimed shows as “Blue’s Clues,” “Bear in the Big Blue House,” and “Dora the Explorer.”484 In fact, Nickelodeon’s success in developing these programs has led CBS to give Nickelodeon control of its entire Saturday morning schedule, the day part that has long been the central focus for chil- dren’s programming. ABC has similarly delegated control of its Saturday morning programming to Disney and NBC has leased three hours of its Satur- day morning lineup to the Discovery Channel.485 These educational options are augmented still further by the growing array of channels, such as the Discovery Channel, Animal Planet, and CNN, that offer news and documentary features targeted towards children.486 The increasing diffusion of video cassette record- ers (VCRs) has expanded parents’ educational programming options even fur- ther because roughly ninety percent of U.S. 482. For a skeptical evaluation of Sunstein’s dismal assessment of contemporary news coverage, see infra note 493 and accompanying text. 483. See, e.g., Tim Goodman, Glory Days for Kids’ TV, S.F. CHRON., Apr. 23, 2001, at E1 (observing that children’s television has become “one of the most competitive arenas in the industry” characterized by “a remarkable raising of standards and a mind-boggling array of options”). 486. Goodman, supra note 483, at E1. 1. Implementation of the Fiss-Sunstein Approach households own a VCR487 and Sunstein denigrates the current system as “clearly unsuccessful in terms of both quality and quantity” and avers that a “strong case” exists for requiring broadcast- ers to do more.480 The long tradition of decrying the quality of television programming gives this assertion a certain surface credibility.481 But a more critical examination reveals that Sunstein’s claim may not be as plausible as it ﬁrst seems.482 Many observers insist that the quantity and quality of educational programming available to children has never been better.483 The Corporation for Public Broadcasting has increasingly redirected its resources towards children’s television. As a result, PBS has been able to augment its offerings with a growing cadre of new shows, including such hits as “Between the Lions” and “Zoboomafoo.” In addition, channels available on cable and other MVPDs, such as Nickelodeon, the Disney Channel, the ABC Family Channel, and new cable offerings such as Noggin, which is a joint venture between Nickelodeon and the Children’s Television Workshop, are providing an increasing variety of high-quality educational programming targeted at children, including such criti- cally acclaimed shows as “Blue’s Clues,” “Bear in the Big Blue House,” and “Dora the Explorer.”484 In fact, Nickelodeon’s success in developing these programs has led CBS to give Nickelodeon control of its entire Saturday morning schedule, the day part that has long been the central focus for chil- dren’s programming. ABC has similarly delegated control of its Saturday morning programming to Disney and NBC has leased three hours of its Satur- day morning lineup to the Discovery Channel.485 These educational options are augmented still further by the growing array of channels, such as the Discovery Channel, Animal Planet, and CNN, that offer news and documentary features targeted towards children.486 The increasing diffusion of video cassette record- ers (VCRs) has expanded parents’ educational programming options even fur- ther because roughly ninety percent of U.S. households own a VCR487 and 328 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL educational video cassettes are available through local video rental outlets and public libraries. All of these outlets have the added beneﬁt of being less susceptible to the types of advertising pressure that can lead to the systematic underprovision of children’s programming in the ﬁrst place.488 And even those who agree with Sunstein’s point about the inadequacy of children’s television must confront the issue of how much additional program- ming is required. 490. There are other, equally thorny deﬁnitional issues lurking within the children’s television issue. The FCC limits its deﬁnition to shows that have “serving the educational and informational needs of children aged 16 and under as a signiﬁcant purpose.” Id. at ¶ 84. Presumably, Sunstein’s theory would have to justify ignoring programming oriented towards the entire family despite the contribution of that programming to the education of children. See 1983 Children’s Television Order, supra note 488, at 646–47 ¶ 31. In addition, the FCC chose not to limit its deﬁnition to programming that furthers children’s “cognitive and intellectual development,” opting instead to include programs that promote children’s “social and emotional development” as well. 1996 Children’s Television Order, supra note 488, at 10701 ¶ 87. Such a distinction runs the simultaneous risk of being unmanageable and being susceptible to manipulation. tbl.915 (1998) (estimating VCR ownership at 82%); TV DIMENSIONS 2000, at 137–38, 140 (Ed Papazian ed., 2000) (estimating VCR ownership at 91%). 488. The FCC’s most recent assessment of the sufﬁciency of the current level of children’s television focuses solely on the amount of children’s programming provided by commercial television stations. See Policies & Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10676–79 ¶¶ 37–41 (1996) [hereinafter 1996 Children’s Television Order]. As a result, it ignores the signiﬁcant contributions to the children’s television environment being offered by PBS, cable, and other nonbroadcast outlets. The FCC’s reasons for doing so result primarily from the legislative intent expressed in enacting the Children’s Television Act of 1990. Id. at 10680–81 ¶ 43; Policies & Rules Concerning Children’s Television Programming, Report and Order, 6 F.C.C.R. 2111, 2116 ¶ 30 (1991). As a policy and constitutional matter, however, it does not seem sensible to ignore the contribution to the educational and information needs of children provided by alternative sources of programming. See Children’s Television Programming & Advertising Practices, Report and Order, 96 F.C.C.2d 634, 645–46 ¶¶ 29–30 (1983) [hereinafter 1983 Children’s Television Order]. It is true that less afﬂuent households may not be able to afford cable programming. As I discuss in a later portion of this Article, this problem can be redressed more effectively through other means. See infra section V.E. p y g 489. 1996 Children’s Television Order, supra note 488, at 10718–19 ¶ 120. 491. Compare 47 U.S.C. § 315(a) (2000) (guaranteeing equal access to all candidates), with id. 47 U.S.C. § 312(a)(7) (2000) (limiting reasonable access to federal candidates). tbl.915 (1998) (estimating VCR ownership at 82%); TV DIMENSIONS 2000, at 137–38, 140 (Ed Papazian ed., 2000) (estimating VCR ownership at 91%). 1. Implementation of the Fiss-Sunstein Approach Current law makes it easier for stations to obtain renewal if they provide at least three hours of “core” children’s programming each week.489 The level of detail with which Sunstein has articulated his theory provides little basis for determining whether three hours is in fact enough and how much more, if any, would be required.490 The level of detail with which Sunstein has articulated his theory provides little basis for determining whether three hours is in fact enough and how much more, if any, would be required.490 Similar problems surround Sunstein’s call for free air time for political candidates. Any such scheme would immediately raise a host of practical questions. How much free air time and at what point in the campaign? Should the mandate apply to presidential elections, all federal elections, or state and local ofﬁce as well?491 Indeed, one could easily argue that state and local candidates present the more compelling case for such support. Presumably, each of these decisions would be made with respect to the particular demands of an idealized democratic process. But the lack of deﬁnition in Sunstein’s proposal makes it impossible to determine how these issues should be resolved. 491. Compare 47 U.S.C. § 315(a) (2000) (guaranteeing equal access to all candidates), with id. 47 U.S.C. § 312(a)(7) (2000) (limiting reasonable access to federal candidates). 329 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT In addition, Sunstein seems to envision a particular format for this speech, suggesting that it should be used for substantial speeches on substantive issues and should not be used for “sound bites” or “infotainment.”492 The source of these particular limitations is far from clear. Indeed, the preference for particular formats threatens to skew the debate in substantive ways. As the original Kennedy-Nixon debates and the subsequent wrangling over debate formats during every presidential election since has demonstrated, the choice of format will typically favor particular candidates. It is equally hard to understand precisely what to make of Sunstein’s infotainment criticism.493 Two of the most signiﬁcant media events of the 1992 presidential campaign—Ross Perot’s appear- ance on “The Larry King Show” and Bill Clinton’s stint as a saxophone player on “The Arsenio Hall Show”—appear to fall outside of the type of speech that Sunstein seeks to promote. Yet both cases underscore the extent to which the format can play an integral role in shaping democratic outcomes. 494. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 141, 146, 150; see also id. at 178 (criticizing free speech theories based on the marketplace of ideas on the ground that the proponents are unable to describe what a properly functioning marketplace of ideas would look like). 492. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 60–61, 82, 85; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221; Sunstein, supra note 23, at 553–54. 492. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 60–61, 82, 85; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 221; Sunstein, supra note 23, at 553–54. 493. See Sunstein, supra note 23, at 527 (denigrating MSNBC as “soft news”). 494. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 141, 146, 150; see also id. at 178 (criticizing free speech theories based on the marketplace of ideas on the ground that the proponents are unable to describe what a properly functioning marketplace of ideas would look like). , p , ; , p , 493. See Sunstein, supra note 23, at 527 (denigrating MSNBC as “soft news”). 1. Implementation of the Fiss-Sunstein Approach In addition, the existence of multiple afﬁrmative programming obligations inevitably raises the prospect that policymakers will have to trade off two forms of high-value speech against each other. The omnipresent reality of limited resources as well as the mandates of administrative law will inevitably force the regulatory authorities to offer a more reasoned explanation of how to resolve these competing claims in a rational manner. The demands of Sunstein’s vision of the First Amendment will be no less insistent. Sunstein’s failure to provide any basis for resolving such questions is particularly surprising in light of his frequent dismissal of competing theories as too ad hoc.494 Such criticism would seem to apply with equal force to his own work. b. The Distinction Between Political and Nonpolitical Speech. Both Fiss and Sunstein place the greatest importance on speech that promotes collective self-determination. As a result, any implementation of their theories raises two questions. First, precisely what types of speech merit greater solicitude under the First Amendment? Second, what degree of protection does that greater solicitude entail? I will consider each question in turn. Neither Fiss nor Sunstein offer a satisfactory description of what types of speech warrant the highest degree of First Amendment protection. Fiss is quite vague on this point. Although he constantly emphasizes the importance of promoting robust public debate, he never offers much in the way of explanation of what types of speech fall within its scope. As a result, we are left to infer his position from the examples that he provides. Unfortunately, even these ex- amples are not easily synthesized. Apparently the photographs of Robert Map- 330 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 plethorpe, which unabashedly offer vivid depictions of the sexual practices of the gay community, qualify as a contribution to robust public debate,495 while the cross burning banned by the ordinance at issue in R.A.V. v. City of St. Paul496 does not.497 Fiss’s discussions of these particular examples only serve to confuse the issue further. 505. See ERWIN CHEMERINSKY, CONSTITUTIONAL LAW § 11.3.6.1, at 868 (1997); R. Polk Wagner, Note, The Medium Is the Mistake: The Law of Software for the First Amendment, 51 STAN. L. REV. 387, 394 (1999). 504. 418 U.S. 405, 410–11 (1974) (holding that conduct falls within the ambit of the First Amendment if “[a]n intent to convey a particularized message was present, and in the surrounding circumstances the likelihood was great that the message would be understood by those who viewed it”). 503. Id. at 131, 154. 500. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 9 11, 121 22. 501. Id. at 130 (emphasis omitted); SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 236. 495. FISS, LIBERALISM DIVIDED, supra note 24, at 91–94. 496. 505 U.S. 377 (1992). 497. FISS, LIBERALISM DIVIDED, supra note 24, at 111–20. 498. Id. at 94. 499. Id. at 116–17. Equally curious is that at one point Fiss ﬁnds political content in “The Love Boat” because it projects a particular view of the world. And yet, immediately thereafter he calls the choice between “The Love Boat” and “Fantasy Island” trivial. Compare id. at 15, with id. at 17. 500. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 9–11, 121–22. 501. Id. at 130 (emphasis omitted); SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 236. 502. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 152–53. 503. Id. at 131, 154. 504. 418 U.S. 405, 410–11 (1974) (holding that conduct falls within the ambit of the First Amendment if “[a]n intent to convey a particularized message was present, and in the surrounding circumstances the likelihood was great that the message would be understood by those who viewed it”). 505. See ERWIN CHEMERINSKY, CONSTITUTIONAL LAW § 11.3.6.1, at 868 (1997); R. Polk Wagner, Note, The Medium Is the Mistake: The Law of Software for the First Amendment, 51 STAN. L. REV. 387, 394 (1999). ( p ) p 502. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 152–53. 496. 505 U.S. 377 (1992). 500. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 9–11, 121–22. 501. Id. at 130 (emphasis omitted); SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 236. 499. Id. at 116–17. Equally curious is that at one point Fiss ﬁnds political content in “The Love Boat” because it projects a particular view of the world. And yet, immediately thereafter he calls the choice between “The Love Boat” and “Fantasy Island” trivial. Compare id. at 15, with id. at 17. 95. FISS, LIBERALISM DIVIDED, supra note 24, at 91–94. 497. FISS, LIBERALISM DIVIDED, supra note 24, at 111–20. y p 500. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 9–11, 121–22. 1. Implementation of the Fiss-Sunstein Approach On the one hand, Fiss argues that the political content of the Mapplethorpe exhibit stemmed from the fact that it represented an “angry protest” by the gay community and that the shocking nature of some of the pictures represented an intrinsic part of that protest.498 On the other hand, with respect to R.A.V., Fiss regards the shocking nature of the speech as a justiﬁca- tion for its regulation rather than its protection.499 Although it is undoubtedly possible to reconcile these positions, Fiss does not offer any means for doing so. Anyone attempting to turn his theory into a coherent scheme of regulation is left guessing about what speech should be regarded as worthy of the highest levels of First Amendment concern. Sunstein offers a little more guidance in identifying the types of speech that will obtain the most protection under his theory. Sunstein would reserve the highest degree of First Amendment protection for “political speech,”500 which Sunstein offers a little more guidance in identifying the types of speech that will obtain the most protection under his theory. Sunstein would reserve the highest degree of First Amendment protection for “political speech,”500 which he deﬁnes as speech “both intended and received as a contribution to public deliberation about some issue.”501 As a result, his deﬁnition would encompass “all art and literature that have the characteristics of social commentary,” including Ulysses, Bleak House, and the Mapplethorpe exhibit.502 The key to keeping this distinction from losing all meaning is to eschew focusing on whether particular speech has political consequences.503 This argument necessar- ily suggests that it is the intent, rather than the reception, that gives his deﬁnition analytical traction. Sunstein offers a little more guidance in identifying the types of speech that will obtain the most protection under his theory. Sunstein would reserve the highest degree of First Amendment protection for “political speech,”500 which he deﬁnes as speech “both intended and received as a contribution to public deliberation about some issue.”501 As a result, his deﬁnition would encompass “all art and literature that have the characteristics of social commentary ” Even as an exercise in pure line drawing, the distinction proffered by Sunstein is somewhat questionable. Indeed, it bears a striking similarity to the test announced in Spence v. 507. Id. at 59–62, 81–82, 85. 511. See, e.g., JOHN LOCKE, SECOND TREATISE OF GOVERNMENT §§ 87–89 (C.B. Macpherson ed., Hackett Publ’g Co. 1980) (1690); BERLIN, supra note 457, at 124. 508. Id. at 134. 506. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 9, 122–23. 508. Id. at 134. 509. See, e.g., J.M. Balkin, Some Realism About Pluralism: Legal Realist Approaches to the First Amendment, 1990 DUKE L.J. 375, 394–414. See generally Steven Shiffrin, The Politics of the Mass Media and the Free Speech Principle, 69 IND. L.J. 689, 689–91 (1994) (collecting commentary). 510. Fiss disclaims any intent “to deny altogether the distinction between state and citizen presup- posed by classical liberalism.” FISS, LIBERALISM DIVIDED, supra note 24, at 18. Sunstein is even more lavish in his praise for the doctrine, arguing that “[i]n fact there should be enthusiastic agreement that the First Amendment is aimed only at governmental action, and that private conduct raises no constitutional question.” SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 204; see also id. at 71, 160; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 36 (“The constitutional text aims at ‘Congress,’ not at the owners of newspapers and radio stations.”). 511. See, e.g., JOHN LOCKE, SECOND TREATISE OF GOVERNMENT §§ 87–89 (C.B. Macpherson ed., Hackett Publ’g Co. 1980) (1690); BERLIN, supra note 457, at 124. 509. See, e.g., J.M. Balkin, Some Realism About Pluralism: Legal Realist Approaches to the First Amendment, 1990 DUKE L.J. 375, 394–414. See generally Steven Shiffrin, The Politics of the Mass Media and the Free Speech Principle, 69 IND. L.J. 689, 689–91 (1994) (collecting commentary). Media and the Free Speech Principle, 69 IND. L.J. 689, 689–91 (1994) (collecting commentary). 510. Fiss disclaims any intent “to deny altogether the distinction between state and citizen presup- posed by classical liberalism.” FISS, LIBERALISM DIVIDED, supra note 24, at 18. Sunstein is even more lavish in his praise for the doctrine, arguing that “[i]n fact there should be enthusiastic agreement that the First Amendment is aimed only at governmental action, and that private conduct raises no constitutional question.” SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 204; see also id. at 71, 160; SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 36 (“The constitutional text aims at ‘Congress,’ not at the owners of newspapers and radio stations.”). 511 See e g JOHN LOCKE SECOND TREATISE OF GOVERNMENT §§ 87 89 (C B Macpherson ed 506. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 9, 122–23. 507. Id. at 59–62, 81–82, 85. 513. FISS, LIBERALISM DIVIDED, supra note 24, at 18; Fiss, supra note 24, at 1223, 1236. 514. See, e.g., Am. Mfrs. Mut. Ins. Co. v. Sullivan, 526 U.S. 40, 53 (1999) (“We have never held that the mere availability of a remedy for wrongful conduct, even when the private use of that remedy serves important public interests, so signiﬁcantly encourages the private activity as to make the State responsible for it.”); Tulsa Prof. Collection Servs., Inc. v. Pope, 485 U.S. 478, 485 (1988) (“Private use of state-sanctioned private remedies or procedures does not rise to the level of state action.”); San Francisco Arts & Athletics, Inc. v. United States Olympic Comm., 483 U.S. 522, 543–44 (1987) (noting that the fact that “[a]ll corporations act under charters granted by a government” does not deprive them of “their essentially private character”); Flagg Bros. v. Brooks, 436 U.S. 149, 160 n.10 (1978) (“It would intolerably broaden, beyond the scope of any of our previous cases, the notion of state action under the Fourteenth Amendment to hold that the mere existence of a body of property law in a State, whether decisional or statutory, itself amounted to ‘state action’ even though no process or state ofﬁcials were ever involved in enforcing that body of law.”). Other decisions have spurned the notion that the receipt of beneﬁts conferred by the government such as licenses or subsidies was sufﬁcient to turn a private actor into a state actor. See San Francisco, 483 U.S. at 544; Blum v. Yaretsky, 457 U.S. 991, 1005, 1011 (1982); Jackson v. Metropolitan Edison Co., 419 U.S. 345, 350 (1974); CBS v. DNC, 412 U.S. 94, 114–21 (1973) (plurality opinion); Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 175, 177 (1972). 512. Lugar v. Edmondson Oil Co., 457 U.S. 922, 936 (1982); accord Brentwood Acad. v. Tenn. Secondary Sch. Athletic Ass’n, 531 U.S. 288, 295 (2001); NCAA v. Tarkanian, 488 U.S. 179, 191 (1988). THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 between the individual and the collective will by “preserv[ing] an area of individual freedom.”512 Fiss’s and Sunstein’s stated commitment to limiting the First Amendment to state action insinuates that they will provide a theory that will allow courts to distinguish private action from state action based on the needs of the democratic process. A review of their proposals reveals that no such principled distinction emerges. between the individual and the collective will by “preserv[ing] an area of individual freedom.”512 Fiss’s and Sunstein’s stated commitment to limiting the First Amendment to state action insinuates that they will provide a theory that will allow courts to distinguish private action from state action based on the needs of the democratic process. A review of their proposals reveals that no such principled distinction emerges. Fiss argues that the affairs of broadcasters and the state are sufﬁciently intertwined to justify regarding the former as state actors. Broadcasters receive the beneﬁts of the state’s general laws of contract, property, corporations, and taxation. In addition, broadcasters depend upon the licenses they receive from the federal government that give them the exclusive right to use a portion of the spectrum, and public broadcasters in particular depend upon federal subsidies. Most importantly, Fiss argues that broadcasting serves the important public function of educating the citizenry. As a result, broadcasters are properly regarded as a hybrid of the public and private.513 The problem is that none of the criteria proposed by Fiss provides a prin- cipled basis for distinguishing between state and private action. Consider, for example, his argument that reliance on background principles of private law is sufﬁcient to support a ﬁnding of state action. Any such rule would swallow the doctrine whole because it would in effect bring everything within the ambit of state action.514 Equally expansive is Fiss’s suggestion that because broadcasters 513. FISS, LIBERALISM DIVIDED, supra note 24, at 18; Fiss, supra note 24, at 1223, 1236. 514. See, e.g., Am. Mfrs. Mut. Ins. Co. v. Sullivan, 526 U.S. 40, 53 (1999) (“We have never held that the mere availability of a remedy for wrongful conduct, even when the private use of that remedy serves important public interests, so signiﬁcantly encourages the private activity as to make the State responsible for it.”); Tulsa Prof. Collection Servs., Inc. v. Pope, 485 U.S. ( ) 513. FISS, LIBERALISM DIVIDED, supra note 24, at 18; Fiss, supra note 24, at 1223, 1236. 1. Implementation of the Fiss-Sunstein Approach Washington504 for distinguishing between expres- sive and nonexpressive conduct that has been criticized as indeterminate.505 A 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 331 more fundamental problem is that Sunstein’s test appears to bear little relation to his theory of free speech. When viewed from the perspective of promoting collective self-determination, Sunstein’s admonition against focusing on the effects of particular speech appears quite strange. On the contrary, one might think that his desire to promote the democratic process might lead him to focus solely on the impact of the speech and to ignore individual motivation alto- gether. With respect to the second question, even after it is determined that particular speech merits the greatest degree of First Amendment protection, it is hard to determine precisely what this additional solicitude would entail. Again, it is Sunstein who provides the more complete articulation. In his view, attempts to regulate political speech carry the strongest presumption of unconstitutionality and require a showing of likely, immediate, and grave harm.506 Although this statement accords well enough with conventional doctrine, it is hard to recon- cile with the speciﬁc policy measures that Sunstein proposes. For example, one of the reforms that Sunstein advocates is the improvement of news coverage, particularly with respect to elections.507 Sunstein’s clear endorsement of govern- mental interference with such a core political process raises seemingly intrac- table questions about what it means to treat regulations of political speech as presumptively unconstitutional. Indeed, his willingness to do so is ironic consid- ering that one of his primary arguments for protecting political speech is the greater likelihood of government partiality.508 c. State Action Doctrine. In calling for governmental intervention to redress distortions caused by private speech choices, Fiss and Sunstein must also confront the state action doctrine. Although a number of free speech theorists have called for an end to the public-private distinction,509 Fiss and Sunstein claim to be more circumspect and eschew any desire to abandon the doctrine altogether.510 Their reasons for doing so are understandable. The state action doctrine has long been central to the relationship between the individual and the state in classic liberal thought511 and has played a critical role in mediating 332 THE GEORGETOWN LAW JOURNAL 521. See, e.g., City of Los Angeles v. Preferred Communications, 476 U.S. 488, 494 (1986); FCC v. Midwest Video Corp., 440 U.S. 689, 707 (1979); CBS v. DNC, 412 U.S. 94, 116–21 (1973) (plurality opinion); cf. Licensee Responsibility to Review Records Before Their Broad., 31 F.C.C.2d 377, 379 ¶ 7 (1971) (mem. op. & order); En Banc Programming Inquiry, Report and Statement of Policy, 44 F.C.C. 2303, 2308–09 (1960); FCC, REPORT ON CHAIN BROADCASTING, supra note 2, at 65–66. enlarge the jurisdictional reach of the First Amendment; it was not intended to expand its substantive scope. Thus, to the extent that general background principles of state private law were insufﬁcient to justify ﬁnding state action prior to Gitlow, it should similarly be insufﬁcient even after the Court construed the First Amendment as applying against the states. 520. Id. at 205. pp y g g 515. FISS, LIBERALISM DIVIDED, supra note 24, at 18. THE GEORGETOWN LAW JOURNAL 478, 485 (1988) (“Private use of state-sanctioned private remedies or procedures does not rise to the level of state action.”); San Francisco Arts & Athletics, Inc. v. United States Olympic Comm., 483 U.S. 522, 543–44 (1987) (noting that the fact that “[a]ll corporations act under charters granted by a government” does not deprive them of “their essentially private character”); Flagg Bros. v. Brooks, 436 U.S. 149, 160 n.10 (1978) (“It would intolerably broaden, beyond the scope of any of our previous cases, the notion of state action under the Fourteenth Amendment to hold that the mere existence of a body of property law in a State, whether decisional or statutory, itself amounted to ‘state action’ even though no process or state ofﬁcials were ever involved in enforcing that body of law.”). Other decisions have spurned the notion that the receipt of beneﬁts conferred by the government such as licenses or subsidies was sufﬁcient to turn a private actor into a state actor. See San Francisco, 483 U.S. at 544; Blum v. Yaretsky, 457 U.S. 991, 1005, 1011 (1982); Jackson v. Metropolitan Edison Co., 419 U.S. 345, 350 (1974); CBS v. DNC, 412 U.S. 94, 114–21 (1973) (plurality opinion); Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 175, 177 (1972). As an aside, such an interpretation of the state action doctrine would represent a historical anomaly. As noted earlier, the text of the First Amendment is directed only at Congress. The First Amendment was not construed as applying to the states until the Court incorporated it by way of the Fourteenth Amendment in Gitlow v. New York, 268 U.S. 652 (1925). Prior to Gitlow, it would have been nonsensical to argue that background principles of contract, property, or corporations law were sufﬁcient to support state action because at that time the First Amendment was construed as a limit on the federal government. The incorporation of the First Amendment effected by Gitlow was intended to 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT p 518. Id. at 36–37; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 72, 209. 519. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 160. g 517. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 44–45. 516. Jackson, 419 U.S. at 352–54 (citing Nebbia v. New York, 291 U.S. 50 enlarge the jurisdictional reach of the First Amendment; it was not intended to expand its substantive scope. Thus, to the extent that general background principles of state private law were insufﬁcient to justify ﬁnding state action prior to Gitlow, it should similarly be insufﬁcient even after the Court construed the First Amendment as applying against the states. 515. FISS, LIBERALISM DIVIDED, supra note 24, at 18. 516. Jackson, 419 U.S. at 352–54 (citing Nebbia v. New York, 291 U.S. 502, 536 (1934)). 517. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 44–45. 518. Id. at 36–37; SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 72, 209. 519. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 160. 520. Id. at 205. 525. See, e.g., DeBauche v. Trani, 191 F.3d 499, 506–09 (4th Cir. 1999); Belluso v. Turner Communications Corp., 633 F.2d 393, 398–400 (5th Cir. 1980); Kuzco v. W. Conn. Broad. Co., 566 F.2d 384, 387 (2d Cir. 1977); Mass. Universalist Convention v. Hildreth & Rogers Co., 183 F.2d 497, 501 (1st Cir. 1950) (per curiam); McIntire v. William Penn Broad. Co., 151 F.2d 597, 601 (3rd Cir. 1945); Mehdi v. Boyce, 931 F. Supp. 268, 270 (S.D.N.Y. 1996), aff’d mem., 125 F.3d 844 (2d Cir. 1997); Tilton v. Capital Cities/ABC Inc., 827 F. Supp. 674, 682 (N.D. Okla. 1993); Rokus v. ABC, Inc., 616 F. Supp. 110 (S.D.N.Y. 1984); Levitch v. CBS, Inc., 495 F. Supp. 649, 655–56 (S.D.N.Y. 1980); Cent. N.Y. Right to Life Fed’n v. Radio Station WIBX, 479 F. Supp. 8, 11 (N.D.N.Y. 1979); Moro v. Telemundo Incorporato, 387 F. Supp. 920, 925 (D.P.R. 1974); Smothers v. CBS, Inc., 351 F. Supp. 622, 627 (C.D. Cal. 1972); Post v. Payton, 323 F. Supp 799, 803–04 (E.D.N.Y. 1971). A four-to-two plurality of the Supreme Court took the same position in CBS v. DNC. Compare 412 U.S. at 115 n.14, 116, 120–21 (plurality opinion) (Burger, C.J., joined by Rehnquist & Stewart, JJ.) (concluding that broadcast- ers were not state actors), id. at 139–40 (Stewart, J., concurring) (same), and id. at 150 (Douglas, J., concurring in the judgment) (same), with id. at 174–81 (Brennan, J., joined by Marshall, J., dissenting) (concluding that broadcasters were state actors). Three other Justices reserved the question, thus depriving the Court of a majority opinion on this issue. Id. at 147 (White, J., concurring in part); id. at 148 (Blackmun, J., joined by Powell, J., concurring in part). j y g p 526. E.g., MEIKLEJOHN, FREE SPEECH, supra note 373, at 22–27; MEIKLEJOHN, POLITICAL FREEDOM, supra note 373, at 24–27; Meiklejohn, supra note 373, at 259–61; see also Harry Kalven, Jr., The THE GEORGETOWN LAW JOURNAL 334 [Vol. 91:245 question was itself a state agency.522 Without such independence, it is impos- sible for the media to serve as the check against governmental abuse envisioned by their role as the Fourth Estate.523 The implication is that the media have an adversarial relationship with the government that is analogous to the role of public defenders, who are not regarded as state actors, although they are in fact government employees.524 As a result, it comes as no surprise that courts have almost invariably concluded that broadcasters are not state actors.525 ( ) 524. See Polk v. Dodson, 454 U.S. 312, 324–25 (1981) (holding that a public defender did not act under “color of state law”). 522. Ark. Educ. Television Comm’n v. Forbes, 523 U.S. 666, 673 (1998); see also FCC v. League of Women Voters, 468 U.S. 364, 378 (1984). 523. See David A. Anderson, The Origins of the Press Clause, 30 UCLA L. REV. 455, 460–62, 488–93, 524, 533–34 (1983); Blasi, supra note 478, at 541–42; Potter Stewart, “Or of the Press,” 26 HASTINGS L.J. 631, 633–34 (1975). THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 333 perform a public function, they are state actors for First Amendment purposes. It is doubtful that such a rule can preserve the bastion of individual freedom envisioned by liberalism. As Fiss himself concedes, this principle would encom- pass the print media as well as “all corporations, unions, universities, and political organizations.”515 In addition, any attempt to identify entities that serve public functions would be tantamount to a return to the Lochner era’s now discredited attempt to allow governmental greater regulation of industries “af- fected with a public interest.”516 Sunstein’s attempts to reconcile his theory with the state action doctrine are similarly unsuccessful. At some points, Sunstein appears to echo Fiss’s sugges- tion that because broadcasters beneﬁt from the protection conferred by property law and hold licenses issued by the government, they are state actors.517 At other points, Sunstein candidly concedes that allowing the role of the govern- ment in enforcing property, contract, and tort law to render private action into state action would render the doctrine a nullity.518 The “real question” for Sunstein is whether the action in question “violates . . . any . . . constitutional provision. . . . It is a question about the meaning of the Constitution, not about state action.”519 Thus, in a First Amendment case, Sunstein argues that whether an action taken by a private entity constituted state action would turn largely on whether the restriction in question was content-neutral or content-based.520 The effect of this proposal is to collapse the state action inquiry into the merits of the underlying constitutional claim, thereby eliminating it as an independent ele- ment of jurisprudence. In short, neither Fiss nor Sunstein are able to reconcile their theories with the separation between state and individual envisioned by classical liberalism and embodied in the distinction between state and private action. Their claims of fealty to the state action doctrine notwithstanding, the positions that they advocate would swallow the doctrine whole. Indeed, an analysis of the role that broadcasters and other media entities play in the democratic process reveals the inherent contradiction in regarding them as state actors. What is perhaps most striking is that the Court has underscored the importance of editorial independence521 even when the broadcaster in THE GEORGETOWN LAW JOURNAL 522. Ark. Educ. Television Comm’n v. Forbes, 523 U.S. 666, 673 (1998); see also FCC v. League Women Voters, 468 U.S. 364, 378 (1984). 2. Institutional Guideposts 2. Institutional Guideposts In the face of such profound implementation problems, the ad hoc, directional advice offered by Fiss and Sunstein appears insufﬁcient. Rather than provide more concrete substantive guidance, both Fiss and Sunstein turn to certain institutional structures to help give content to the mandate to promote robust public debate. First, Fiss suggests that one can draw inspiration from the role of the parliamentarian at a New England town meeting. Second, Fiss suggests that one can gain insights into the needs of the democratic process by comparing the current state of the world to the speech environment of the Jeffersonian era. Lastly, both Fiss and Sunstein suggest that certain features of the federal government provide greater reason to be conﬁdent in the government’s ability to manage the uncertainties of this process. I will discuss each in turn. a. Democratic Process as Artifact: The Metaphor of the Town Meeting. Fiss draws on the work of Alexander Meiklejohn526 to suggest that the state may 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 335 play the role of the parliamentarian in a classic New England town meeting by simply ordering the agenda and metering the extent to which people participate in the dialogue.527 Fiss believes that this agenda will evolve “organically” in a manner external to the public debate itself.528 He further suggests that in its role as parliamentarian, the government would do more than just apply a neutral procedural principle, like temporal priority. This role requires the state be “sensitive to the excesses of advocacy and the impact of such excesses on the fullness of debate.”529 As a result, Fiss envisions the state guarding against repetitious speech by saying, in effect, “We have heard that point several times now,” or “Let’s hear from the other side.”530 In addition, Fiss argues that the state can limit “[u]gly, hateful speech” that may silence others.531 Thus, the state as parliamentarian will have to exercise content-based judgments in setting its agenda. The First Amendment, in Fiss’s view, only bars the government from making content-based judgments that are intended to affect substantive out- comes; it does not extend to content-based judgments made to “protect the integrity of the deliberative process.”532 There is reason to doubt the validity of the rigid division between agenda setting and substantive outcomes upon which Fiss’s theory depends. Agenda setting and procedure are as much part of self-determination as is the substance of the debate. 535. As Kenneth Karst has noted, “even the repetition of speech conveys the distinctive message that an opinion is widely shared,” which is of “great importance in an ‘other-directed society’ where opinion polls are self-fulﬁlling prophecies.” As a result, it is impossible for the state to know when “‘everything worth saying’ has been said.” Kenneth L. Karst, Equality as a Central Principle in the First Amendment, 43 U. CHI. L. REV. 20, 40 (1975); see also Bhagwat, supra note 26, at 182–83 (arguing that “mass media plays an essential role in inculcating, or perhaps reinforcing, basic political values through sheer repetition of the message that the basic American political institutions are ﬂawless and ideal”). 534. FISS, LIBERALISM DIVIDED, supra note 24, at 15 (noting that even such shows as “The Love Boat” project “a view of the world . . . which in turn tends to deﬁne and order our options and choices”). Concept of the Public Forum, 1965 SUP. CT. REV. 1, 23–25 (emphasizing the distinction between parliamentary rules and rules governing content and arguing that “concessions on [the parliamentary] front should not be taken as relevant to . . . questions of control of content”). 528. FISS, LIBERALISM DIVIDED, supra note 24, at 118. 532. Id. at 153. 533. Post, supra note 409, at 1539 (internal quotation marks omitted). 529. Id. at 118. 530. Id. at 153; see also id. at 85 (arguing that a parliamentarian may “requir[e] some to shut up so others can speak” without engaging in censorship). 531. Id. at 118. Concept of the Public Forum, 1965 SUP. CT. REV. 1, 23–25 (emphasizing the distinction between parliamentary rules and rules governing content and arguing that “concessions on [the parliamentary] front should not be taken as relevant to . . . questions of control of content”). q 527. FISS, IRONY OF FREE SPEECH, supra note 24, at 21–24; FISS, LIBERALISM DIVIDED, supra note 24, at 85, 101, 118–19, 153. 2. Institutional Guideposts In other words, “Just what is a political issue is itself a political issue.”533 Consider what Fiss regards as the easiest example: repetitious speech. What is strange about this position is that Fiss himself recognizes that main- stream entertainment-oriented speech is highly political, in that it shapes politi- cal values by reinforcing messages.534 It thus follows that orthodox and repetitive speech can be as much a part of a robust public discourse as dissident speech.535 530. Id. at 153; see also id. at 85 (arguing that a parliamentarian may “requir[e] some to shut up so others can speak” without engaging in censorship). 534. FISS, LIBERALISM DIVIDED, supra note 24, at 15 (noting that even such shows as “The Love Boat” project “a view of the world . . . which in turn tends to deﬁne and order our options and choices”). 535. As Kenneth Karst has noted, “even the repetition of speech conveys the distinctive message that an opinion is widely shared,” which is of “great importance in an ‘other-directed society’ where opinion polls are self-fulﬁlling prophecies.” As a result, it is impossible for the state to know when “‘everything worth saying’ has been said.” Kenneth L. Karst, Equality as a Central Principle in the First Amendment, 43 U. CHI. L. REV. 20, 40 (1975); see also Bhagwat, supra note 26, at 182–83 (arguing that “mass media plays an essential role in inculcating, or perhaps reinforcing, basic political values through sheer repetition of the message that the basic American political institutions are ﬂawless and ideal”). 336 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL Even more troubling is the problem of “ugly, hateful speech.” As noted earlier, the Court has repeatedly disparaged its usefulness as a constitutional category.536 For example, in Hustler Magazine v. Falwell,537 the Court ex- pressed doubt as to the existence of a “principled standard” to distinguish between speech that enhances the public discourse and “outrageous” speech that does not.538 Similarly, as the Court observed in Cohen v. California,539 true political speech “may often appear to be only verbal tumult, discord, and even offensive utterance.”540 Indeed, some degree of “verbal cacophony” is a “neces- sary side effect[ ]” of “a society as diverse and populous as ours” and “is, in this sense not a sign of weakness but of strength.”541 In fact, as the Court noted in Texas v. q g y g 544. Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting); see also supra notes 76–77, 265 and accompanying text (citing cases holding that the fact that speech often excites anger is typically a reason to protect it, not restrict it). 545. FISS, LIBERALISM DIVIDED, supra note 24, at 37–38; see also id. at 12–14, 18, 49–50 (noting the shift in paradigmatic speaker envisioned by the First Amendment from the street corner speaker to CBS). ( ) 543. Id. at 408–09 (quoting Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949)). 536. See supra notes 76–77, 265 and accompanying text. 537. 485 U.S. 46 (1988). 538. Id. at 55. 539. 403 U.S. 15 (1971). 540. Id. at 24–25. 541. Id. at 25. 542. 491 U.S. 397 (1989). 543. Id. at 408–09 (quoting Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949)). 544. Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting); see also supra notes 76–77, 265 and accompanying text (citing cases holding that the fact that speech often excites anger is typically a reason to protect it, not restrict it). 545. FISS, LIBERALISM DIVIDED, supra note 24, at 37–38; see also id. at 12–14, 18, 49–50 (noting the shift in paradigmatic speaker envisioned by the First Amendment from the street corner speaker to CBS). 536. See supra notes 76–77, 265 and accompanying text. 537. 485 U.S. 46 (1988). 538. Id. at 55. 539. 403 U.S. 15 (1971). 540. Id. at 24–25. 541. Id. at 25. 542. 491 U.S. 397 (1989). 543 Id at 408 09 (quoting Terminiello v City of Chicago 337 U S 1 4 (1949)) 536. See supra notes 76–77, 265 and accompanying text. 537. 485 U.S. 46 (1988). 538. Id. at 55. 539. 403 U.S. 15 (1971). 540. Id. at 24–25. 541. Id. at 25. 542. 491 U.S. 397 (1989). 543 Id at 408 09 (quoting Terminiello v City of Chicago 337 U 542. 491 U.S. 397 (1989). 541. Id. at 25. 542. 491 U.S. 397 (1989). 543. Id. at 408–09 (quoting Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949)). 544. Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting); see also supra notes 76–77, 265 and accompanying text (citing cases holding that the fact that speech often excites anger is typically a reason to protect it, not restrict it). 545. FISS, LIBERALISM DIVIDED, supra note 24, at 37–38; see also id. at 12–14, 18, 49–50 (noting the shift in paradigmatic speaker envisioned by the First Amendment from the street corner speaker to THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 337 partisan and highly dependent on political patronage for their economic sur- vival.546 To the extent that the sober reﬂection commonly associated with the Constitutional Convention and the Federalist Papers existed at all, it was a product of the homogeneity of the Founders’ backgrounds as well as the disenfranchisement of blacks, women, and landless males.547 It thus seems far from clear that Jeffersonian democracy represents a normatively attractive baseline for measuring the quality of current speech. It is also possible to dispute Fiss’s dismal assessment of the current environ- ment. A review of the television listings reveals that more news and information programming is available than ever before. Not only are CNN and a growing cohort of competitors providing news programming around the clock, news magazines such as “Dateline” and “20/20” dot the primetime lineups of the major broadcast networks as well. Documentaries on a wide variety of subjects are available twenty-four hours a day on cable television, and cable is spawning a growing number of local news channels. The radio dial crackles with political speech, as talk radio has come to dominate the AM band. Indeed, the experience with talk radio provides some evidence that attempts to promote balanced reporting may in fact be inconsistent with the quest for robust political speech. An empirical study by Thomas Hazlett and David Sosa suggests that the repeal of the Fairness Doctrine was the trigger that led to the talk radio explosion.548 The Jeffersonian world that Fiss would use as his baseline thus has more in common with the modern, discordant world of talk radio and sound bites than it does with the world of balanced reporting that he envisions. In addition, it is hard to see how those dissatisﬁed with the current speech environment can base their complaints on the total quantity of political speech being uttered. More qualitative complaints ultimately must be based on a more ﬁne-tuned assess- ment of what constitutes good and bad political speech. As the foregoing discussion demonstrates, the mere invocation of Jeffersonian democracy cannot substitute for Fiss’s failure to articulate a basis for drawing such a distinction. c. Institutional Faith in the FCC. Finally, both Fiss and Sunstein draw com- fort from the past history of broadcast regulation. Fiss acknowledges that the same social forces that are controlling private media entities will exert a similar inﬂuence over the government as well. 546. See David L. Lange, The Role of the Access Doctrine in the Regulation of the Mass Media: A Critical Review and Assessment, 52 N.C. L. REV. 1, 12–14 (1973–1974). 547. See L.A. Powe, Jr., Scholarship and Markets, 56 GEO. WASH. L. REV. 172, 183 (1987). 548. Thomas W. Hazlett & David W. Sosa, Was the Fairness Doctrine a “Chilling Effect”? Evidence from the Postderegulation Radio Market, 26 J. LEGAL STUD. 279 (1997). 549. FISS, LIBERALISM DIVIDED, supra note 24, at 23–26, 43; see also Fiss, supra note 24, at 1218 (describing Richard Nixon’s attempt to use the broadcast licensing process to suppress criticism of his Administration). f g 549. FISS, LIBERALISM DIVIDED, supra note 24, at 23–26, 43; see also Fiss, supra note 24, at 1218 (describing Richard Nixon’s attempt to use the broadcast licensing process to suppress criticism of his Administration). 2. Institutional Guideposts Johnson,542 “a principal ‘function of free speech under our system of government is to invite dispute. It may indeed best serve its high purpose when it induces a condition of unrest, creates dissatisfaction with conditions as they are, or even stirs people to anger.’”543 That is why Holmes warned that “we should be eternally vigilant against attempts to check the expression of opinions that we loathe and believe to be fraught with death.”544 b. Jeffersonian Democracy as a Baseline. Fiss further suggests that the Jeffer- sonian era can provide an appropriate benchmark for measuring the quality of the current level of public discourse. His assumption is that “[i]n a Jeffersonian democracy, . . . where the dominant social unit is the individual and power is distributed equally, autonomy might well enhance public debate and thus promote collective self-determination.”545 If true, the diffuse nature of Jefferson- ian society can provide a basis for evaluating the efﬁcacy of the current speech environment. The farther the present deviates from that baseline, the more justiﬁable is greater government control over speech. The problem is that the baseline that Fiss invokes does not support the type of speech universe that he apparently envisions. Fiss’s ideal is a balanced explora- tion of the issues. The public discourse in Jefferson’s era was anything but. Fairness and responsibility did not emerge as journalistic values until well into twentieth century. In Jefferson’s day, the printers placed the greatest import on passionate commitment to ideological debate. As a result, the press was highly 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 548. Thomas W. Hazlett & David W. Sosa, Was the Fairness Doctrine a “Chilling Effect”? Evidence from the Postderegulation Radio Market, 26 J. LEGAL STUD. 279 (1997). 546. See David L. Lange, The Role of the Access Doctrine in the Regulation of the Mass Media: A Critical Review and Assessment, 52 N.C. L. REV. 1, 12–14 (1973–1974). C itical eview and ssessment, 5 N.C. . V. , ( 973 97 ). 547. See L.A. Powe, Jr., Scholarship and Markets, 56 GEO. WASH. L. REV. 172, 183 (1987). 559. Id. at 89; see also SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 222 (same). 560. See, e.g., SCHAUER, supra note 391, at 81; MARK G. YUDOF, WHEN GOVERNMENT SPEAKS 51–66, 179–99 (1983); Blasi, supra note 478, at 527–44; Ronald A. Cass, The Perils of Positive Thinking: Constitutional Interpretation and Negative First Amendment Theory, 34 UCLA L. REV. 1405, 1426–27 (1987); Redish & Kaludis, supra note 99, at 1110–11; Stone, supra note 391, at 1173–75. 550. FISS, LIBERALISM DIVIDED, supra note 24, at 43. 551. Id. at 153. 552. Fiss, supra note 24, at 1223. 553. FISS, LIBERALISM DIVIDED, supra note 24, at 25–26, 44. 554. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 36. 555. Id. at 134–35, 146. 556. Id. at 89. 557. Id. 558. Id. at 92, 104–05. 559. Id. at 89; see also SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 222 (same). 560. See, e.g., SCHAUER, supra note 391, at 81; MARK G. YUDOF, WHEN GOVERNMENT SPEAKS 51–66, 179–99 (1983); Blasi, supra note 478, at 527–44; Ronald A. Cass, The Perils of Positive Thinking: Constitutional Interpretation and Negative First Amendment Theory, 34 UCLA L. REV. 1405, 1426–27 (1987); Redish & Kaludis, supra note 99, at 1110–11; Stone, supra note 391, at 1173–75. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 339 checking the excesses of the government.561 The existence of these alternative traditions indicates that any instrumental vision of speech cannot simply assert the benevolence of the state a priori. Such an approach begs the equally important question “[w]hether government intervention would make the unfair market better or worse than it already is.”562 The ambiguity as to whether governmental intervention will promote or hinder democracy is well illustrated by the work of Vincent Blasi. A comparison of his approach with the approach taken by Fiss and Sunstein reveals many similarities. Blasi accepts viewing speech almost entirely in instrumental terms and concurs that the primary value to be promoted is the proper functioning of the democratic process.563 He nonetheless provides a powerful theoretical argument opposing the type of remedies that Fiss and Sunstein propose. Speciﬁ- cally, Blasi argues that state power is not simply one of many forces shaping speech, but rather that it poses a more serious threat to speech than the abuse of private power.564 Drawing on themes advanced by Holmes,565 Blasi argues that the state is unique in its intrusive investigatory powers, its monopoly over legitimized violence, and the ease with which its resources can be mobilized.566 The ambiguity as to whether governmental intervention will promote or hinder democracy is well illustrated by the work of Vincent Blasi. A comparison of his approach with the approach taken by Fiss and Sunstein reveals many similarities. Blasi accepts viewing speech almost entirely in instrumental terms and concurs that the primary value to be promoted is the proper functioning of the democratic process.563 He nonetheless provides a powerful theoretical argument opposing the type of remedies that Fiss and Sunstein propose. 561. See supra note 523 and accompanying text. 562. Shiffrin, supra note 509, at 714; see also Neuborne, supra note 459, at 439 (claiming that “although Prof. Sunstein vests the government with powerful discretionary tools to regulate speech, he cannot assure that those tools would be any less subject to the harmful inﬂuences that he claims currently distort the laissez-faire speech market”). Shiffrin appears to have distanced himself from this conclusion in his more recent work. See STEVEN H. SHIFFRIN, DISSENT, INJUSTICE, AND THE MEANINGS OF AMERICA 115–20 (1999). 565. See OLIVER WENDELL HOLMES, The Path of the Law, in COLLECTED LEGAL PAPERS 167 (1920). 566 Bl i 478 538 39 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT If so, government intervention may be more likely to reinforce the existing distortions to the public discourse than to ameliorate them.549 Nonetheless, Fiss chooses to be optimistic, arguing that 338 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 “our historical experience with the activist state in the sixties” justiﬁes “believ- [ing] that the elements of independence possessed by the state are real and substantial.”550 In addition, institutional arrangements, such as the use of indepen- dent agencies like the FCC, should help ensure that government intervention is not the product of political manipulation.551 Fiss ﬁnds that the FCC has avoided these pitfalls for the most part, noting that “[a]lthough regulators have looked at the content of programs when awarding broadcast licenses, they have only done so to make certain that there is sufﬁcient coverage of public issues and to enforce well-deﬁned boundaries regarding depictions of sexuality.”552 And on those occasions when the FCC fails to promote robust public debate, the courts act as the ultimate guarantor of constitutional rights.553 Sunstein similarly acknowledges that “a central principle of American consti- tutionalism is that the most serious risks to liberty come from government.”554 At the same time, “a sensible view of government’s incentives” suggests that there is no reason to be “especially suspicious of government regulation of nonpolitical speech.”555 And even with respect to political speech, Sunstein similarly suggests that the success of past regulatory efforts makes it “plausible to think that government decisions can be made in a relatively nonpartisan way.”556 In particular, he joins Fiss in lauding the government’s role in fostering the creation of local news, children’s television, and public television,557 al- though he parts company with Fiss in questioning the propriety of involving the federal courts in this enterprise.558 In light of this track record, Sunstein suggests that “[w]e have no basis for doubting that much larger improvements could be brought about in the future.”559 What is missing from these observations is anything more than a token appearance of the First Amendment’s traditional suspicion of governmental interference with speech. 561. See supra note 523 and accompanying text. 562. Shiffrin, supra note 509, at 714; see also Neuborne, supra note 459, at 439 (claiming that “although Prof. Sunstein vests the government with powerful discretionary tools to regulate speech, he cannot assure that those tools would be any less subject to the harmful inﬂuences that he claims currently distort the laissez-faire speech market”). Shiffrin appears to have distanced himself from this conclusion in his more recent work. See STEVEN H. SHIFFRIN, DISSENT, INJUSTICE, AND THE MEANINGS OF AMERICA 115–20 (1999). 563. Blasi, supra note 478, at 558. 564. Id. at 538. 565. See OLIVER WENDELL HOLMES, The Path of the Law, in COLLECTED LEGAL PAPERS 167 (1920). 566. Blasi, supra note 478, at 538–39. 567. Id. at 540. 568. Id. at 538–41. 569. Id. at 541–42. 563. Blasi, supra note 478, at 558. 561. See supra note 523 and accompanying text. 568. Id. at 538–41. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Such suspicion is deeply rooted in the history of the Founding and has been buttressed by the struggles with communist and anti-war speech that led to many of the seminal decisions in the First Amendment canon.560 As noted earlier, there is also an equally substantial body of scholar- ship that regards the institutional press as a “Fourth Estate” charged with 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 569. Id. at 541–42. 566. Blasi, supra note 478, at 538–39. 564. Id. at 538. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Speciﬁ- cally, Blasi argues that state power is not simply one of many forces shaping speech, but rather that it poses a more serious threat to speech than the abuse of private power.564 Drawing on themes advanced by Holmes,565 Blasi argues that the state is unique in its intrusive investigatory powers, its monopoly over legitimized violence, and the ease with which its resources can be mobilized.566 In addition, ofﬁcial power carries with it a moral and symbolic signiﬁcance that makes it more menacing than private power.567 As a result, Blasi argues that exercises of state power raise particularly signiﬁcant reasons for concern.568 The reality of this threat causes Blasi to search for a competing institution that can serve as a counterweight to the Leviathan. And the private actor that he sees best situated to offset the power of the government is the “well-organized, well-ﬁnanced, professional critics” in the institutional press.569 Allowing the government to curb what it perceives as the excesses of the private media, however, would compromise the latter’s ability to serve as this institutional check. Blasi’s work demonstrates the ease with which one can draw different conclusions from the same premises advanced by Fiss and Sunstein. In particu- lar, Blasi’s analysis underscores the danger of simply assuming a priori that governmental intervention will be an improvement. Fiss and Sunstein present THE GEORGETOWN LAW JOURNAL 340 [Vol. 91:245 little analytical justiﬁcation for their faith in the innocuousness of governmental action. For the most part, they are content to draw comfort from what they perceive as the success of past regulatory efforts.570 Their enthusiasm for the past is hard to reconcile with the actual performance of the FCC, however. The works of Louis Jaffe, Bernard Schwartz, Henry Friendly, Glen Robinson, Ithiel de Sola Pool, Scot Powe, Matthew Spitzer, and Thomas Hazlett, among others, have documented countless examples of political abuse in the regulatory pro- cess.571 Although contrary voices have occasionally arisen,572 the weight of the empirical record underscores the constitutional hazards surrounding Fiss’s and Sunstein’s positions. Consider the Fairness Doctrine,573 which has talismanic signiﬁcance for both Fiss and Sunstein. The historical evidence now suggests that the Fairness Doctrine has been widely manipulated for political ends.574 Perhaps most ironic are the revelations about the circumstances giving rise to Red Lion, which sustained the constitutionality of the Fairness Doctrine and represents perhaps the central precedent supporting Fiss’s and Sunstein’s positions. 570. See supra notes 549–61 and accompanying text. 571. See FRIENDLY, supra note 41, at 72; POOL, supra note 51, at 119–28, 135; POWE, supra note 51, at 69–74, 83–84, 112–16, 121–29, 131–41; Thomas W. Hazlett, Physical Scarcity, Rent Seeking, and the First Amendment, 97 COLUM. L. REV. 905, 933–34 (1997); Jaffe, supra note 44, at 79; Robinson, supra note 46, at 238; Schwartz, supra note 44, at 694; Spitzer, supra note 208, at 1048–52. 572. See BOLLINGER, supra note 11, at 115 (“[O]ne of the more interesting features of the broadcast regulation experience has been the absence of egregious abuses by the FCC. The commission has, on the whole, been extraordinarily circumspect in the exercise of its powers.”). 573. See supra notes 92, 209–14 and accompanying text (describing the Fairness Doctrine and its repeal). 575. See FRED W. FRIENDLY, THE GOOD GUYS, THE BAD GUYS AND THE FIRST AMENDMENT 6–34 (1975); POWE, supra note 51, at 112–16; Hazlett, supra note 571, at 933–34; Jaffe, supra note 50, at 1700; Spitzer, supra note 208, at 1051–52. 576. See FISS, LIBERALISM DIVIDED, supra note 24, at 44; Cass R. Sunstein, Interest Groups in American Public Law, 38 STAN. L. REV. 29, 63 (1985). p 574. See KRATTENMAKER & POWE, supra note 72, at 248–49; Hazlett, supra note 115, at 168–69. 577. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 134. 570. See supra notes 549–61 and accompanying text. 573. See supra notes 92, 209–14 and accompanying text (describing the Fairness Doctrine and its repeal). 570. See supra notes 549–61 and accompanying text. 571. See FRIENDLY, supra note 41, at 72; POOL, supra note 51, at 119–28, 135; POWE, supra note 51, at 69–74, 83–84, 112–16, 121–29, 131–41; Thomas W. Hazlett, Physical Scarcity, Rent Seeking, and the First Amendment, 97 COLUM. L. REV. 905, 933–34 (1997); Jaffe, supra note 44, at 79; Robinson, supra note 46, at 238; Schwartz, supra note 44, at 694; Spitzer, supra note 208, at 1048–52. 572. See BOLLINGER, supra note 11, at 115 (“[O]ne of the more interesting features of the broadcast regulation experience has been the absence of egregious abuses by the FCC. The commission has, on the whole, been extraordinarily circumspect in the exercise of its powers.”). 573. See supra notes 92, 209–14 and accompanying text (describing the Fairness Doctrine and its repeal). 574. See KRATTENMAKER & POWE, supra note 72, at 248–49; Hazlett, supra note 115, at 168–69. 575. See FRED W. FRIENDLY, THE GOOD GUYS, THE BAD GUYS AND THE FIRST AMENDMENT 6–34 (1975); POWE, supra note 51, at 112–16; Hazlett, supra note 571, at 933–34; Jaffe, supra note 50, at 1700; Spitzer, supra note 208, at 1051–52. 576. See FISS, LIBERALISM DIVIDED, supra note 24, at 44; Cass R. Sunstein, Interest Groups in American Public Law, 38 STAN. L. REV. 29, 63 (1985). 577. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 134. 571. See FRIENDLY, supra note 41, at 72; POOL, supra note 51, at 119–28, 135; POWE, supra note 51, at 69–74, 83–84, 112–16, 121–29, 131–41; Thomas W. Hazlett, Physical Scarcity, Rent Seeking, and the First Amendment, 97 COLUM. L. REV. 905, 933–34 (1997); Jaffe, supra note 44, at 79; Robinson, supra note 46, at 238; Schwartz, supra note 44, at 694; Spitzer, supra note 208, at 1048–52. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT It is now acknowledged that Red Lion arose out of a campaign orchestrated by the Kennedy and Johnson Administrations to use the Fairness Doctrine to suppress political criticism from the right.575 Thus, the prime example of the type of regulatory measures that Fiss and Sunstein support demonstrates the political corruption that can occur when the government interferes with private speech. Indeed, there is something quite puzzling about Fiss’s and Sunstein’s willing- ness to rely on the FCC. Both have candidly recognized the imperfections of the administrative process.576 Sunstein’s faith in the FCC is even more ironic when he defends his two-tiered vision of the First Amendment because “the premise of distrust of government is strongest when politics is at issue.”577 In light of this recognition, it is difﬁcult to see how he can rely on the FCC as an appropriate mechanism for effecting the increases in the amount and quality of 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 341 information regarding political campaigns that represent one of the central elements of his program for reform. p g There thus seems to be little theoretical or historical reason to draw comfort from the federal government’s ability to regulate the broadcast industry. In the end, Fiss embraces government intervention simply because he sees no other alternative. Fiss turns to the state “because it is the only hope, the only means to correct the distorting inﬂuence of social structure on public debate.”578 Al- though Fiss acknowledges the risk of serious governmental misconduct, “[t]he hope against hope is that in the ﬁnal analysis we will be better off than under a regime of autonomy.”579 Sunstein similarly defends his theory simply by argu- ing that the current state of affairs is so unacceptable that something must be done. According to Sunstein, the remedies he proposes are justiﬁed simply because “the current system is worse than imperfect; it creates extremely serious obstacles to a well-functioning system of free expression.”580 If the Constitution bars such experimentation, then it seems clear to Sunstein that something is wrong with the Constitution: “[W]hy should the Constitution bar a democratic decision to experiment with new methods for achieving their Madiso- nian goals?”581 The question suggests an approach in which the exigencies of the democratic process provide their own constitutional justiﬁcation. 581. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 220; see also SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 81 (“The most important point is that the First Amendment should not operate as a talismanic or reﬂexive obstacle to our efforts to experiment with different strategies for achieving free speech goals.”). 580. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 222; see also SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 89 (“The absence of continuous government supervision should not obscure the point. With respect to attention to public issues, the present system badly disserves Madisonian goals.”). g p g ) 582. Shiffrin, supra note 509, at 717. 578. FISS, LIBERALISM DIVIDED, supra note 24, at 25. 579. Id. 580. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 222; see also SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 89 (“The absence of continuous government supervision should not obscure the point. With respect to attention to public issues, the present system badly disserves Madisonian goals.”). 581. SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 220; see also SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 81 (“The most important point is that the First Amendment should not operate as a talismanic or reﬂexive obstacle to our efforts to experiment with different strategies for achieving free speech goals.”). 582. Shiffrin, supra note 509, at 717. 578. FISS, LIBERALISM DIVIDED, supra note 24, at 25. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT Such a vision turns the First Amendment into simply the residuum after the basic regulatory questions are answered and would eliminate it as an independent check on the government. There thus seems to be little theoretical or historical reason to draw comfort from the federal government’s ability to regulate the broadcast industry. In the end, Fiss embraces government intervention simply because he sees no other alternative. Fiss turns to the state “because it is the only hope, the only means to correct the distorting inﬂuence of social structure on public debate.”578 Al- though Fiss acknowledges the risk of serious governmental misconduct, “[t]he hope against hope is that in the ﬁnal analysis we will be better off than under a regime of autonomy.”579 Sunstein similarly defends his theory simply by argu- ing that the current state of affairs is so unacceptable that something must be done. According to Sunstein, the remedies he proposes are justiﬁed simply because “the current system is worse than imperfect; it creates extremely serious obstacles to a well-functioning system of free expression.”580 If the Constitution bars such experimentation, then it seems clear to Sunstein that There thus seems to be little theoretical or historical reason to draw comfort from the federal government’s ability to regulate the broadcast industry. In the end, Fiss embraces government intervention simply because he sees no other alternative. Fiss turns to the state “because it is the only hope, the only means to correct the distorting inﬂuence of social structure on public debate.”578 Al- alternative. Fiss turns to the state “because it is the only hope, the only means to correct the distorting inﬂuence of social structure on public debate.”578 Al- though Fiss acknowledges the risk of serious governmental misconduct, “[t]he hope against hope is that in the ﬁnal analysis we will be better off than under a regime of autonomy.”579 Sunstein similarly defends his theory simply by argu- ing that the current state of affairs is so unacceptable that something must be done. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT According to Sunstein, the remedies he proposes are justiﬁed simply because “the current system is worse than imperfect; it creates extremely serious obstacles to a well-functioning system of free expression.”580 If the C tit ti b h i t ti th it l t S t i th t g y p Constitution bars such experimentation, then it seems clear to Sunstein that something is wrong with the Constitution: “[W]hy should the Constitution bar a democratic decision to experiment with new methods for achieving their Madiso- nian goals?”581 The question suggests an approach in which the exigencies of the democratic process provide their own constitutional justiﬁcation. Such a vision turns the First Amendment into simply the residuum after the basic regulatory questions are answered and would eliminate it as an independent check on the government. In sum, Fiss and Sunstein have no substantive answer to concerns about governmental abuse. Their proposals are driven entirely by an acute sense of the existing problems. The narrowness of their focus leads them to stop short of offering any serious analysis of whether the solutions that they propose will in fact redress the problems that they perceive. In the absence of a more principled justiﬁcation for believing why state intervention may create a better state of affairs, placing speech within governmental control is little better than a “river- boat gamble.”582 Indeed, the concerns about governmental abuse that underlie the creation and development of the First Amendment and the demonstrated track record of the FCC and Congress in broadcast regulation suggest that as gambles go, Fiss’s and Sunstein’s face particularly long odds. 342 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 585. See, e.g., Samuel Issacharoff & Pamela S. Karlan, The Hydraulics of Campaign Finance Reform, 77 TEX. L. REV. 1705, 1727 (1999); Rev. Jesse Jackson, Reclaiming Our Youth from Violence, 36 B.C. L. REV. 913, 921 (1995); James G. Wilson, Noam Chomsky and Judicial Review, 44 CLEV. ST. L. REV. 439, 469 (1996). As one congressional candidate put it, “[T]he single most important step we need to take right now is to turn off the darn television . . . a device that spews meaningless garbage. We have a generation that has been trained not to think critically about the information that bombards them.” Quoted in Daniel M. Warner, To Hell on the Railroads: Why Our Technology and Law Encourage a Degrading Culture, 26 TRANSP. L.J. 361, 409 n.145 (1999). 583. See SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 20; Fiss, supra note 24, at 1217–18. 584. Sunstein does raise the question whether the democratic process may be better enhanced by policies affecting the print media, but ultimately calls it an as yet unanswerable empirical question. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 112–13. C. CRITIQUES OF FISS’S AND SUNSTEIN’S EMPIRICAL AND TECHNOLOGICAL FOUNDATIONS Even if the theoretical and practical objections detailed above were somehow overcome, Fiss and Sunstein must address certain empirical and technological obstacles that stand in the way of the outcome that they seek. Speciﬁcally, they do not provide an adequate foundation for their empirical assumptions about the role that television plays in democratic self-governance or explain how their theories can accommodate the convergent technological environment that char- acterizes modern communications. 1. Empirical Assumptions About Television’s Role in Democracy 1. Empirical Assumptions About Television’s Role in Democracy Fiss and Sunstein argue that the unique role that television plays in the public discourse justiﬁes regulations designed to ensure that all households continue to have access to free, over-the-air programming.583 Concluding that television currently represents the communications medium with the greatest inﬂuence on democratic self-governance as a descriptive matter, however, says nothing about whether that state of affairs is normatively desirable. Indeed, there appears to be good reason to be skeptical of claims that television is the medium best suited to fostering democratic discourse because television by its nature does not lend itself to deep and penetrating attention to public issues.584 Some would even argue that efforts to encourage greater availability of programming that en- hances democratic deliberation are fundamentally misguided and that the best way to promote self-governance would be for viewers to turn their televisions off.585 Indeed, Sunstein’s willingness to take as given the public’s existing preference for television as a medium stands in stark contrast to his refusal to respect individual preferences for particular types of programming. If Sunstein were correct that preferences for particular types of programs are simply socially constructed misperceptions, it is hard to see how preferences for the medium as a whole would be entitled to any greater respect. In fact, there is a danger that the policy outcomes that Fiss and Sunstein promote will reify television’s role in the political process and give it unjustiﬁed permanence. For example, the central role that broadcast television currently 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 343 plays in the political process leads both Fiss and Sunstein to offer their support for the longstanding policy known as “must-carry,” which requires all local cable operators to include all full-power local broadcast stations in their basic cable packages.586 Thus, although other cable programmers must typically pay for carriage, must-carry allows broadcasters to receive carriage for free. Must- carry essentially represents a large cross-subsidy from the cable industry to the broadcasting industry. In such a situation, it is difﬁcult to determine whether the privileged position that broadcasters enjoy is the cause or the consequence of must-carry. In other words, positing that free, over-the-air television occupies a special place in the political process in turn justiﬁes regulations that cement the role of broadcasting. 586. See SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 260–71; Fiss, supra note 24, at 1228–31. 586. See SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 260–71; Fiss, supra note 24, at 1228–31. 587. For a review of these alternative technologies, see Annual Assessment of Status of Competition in Mkt. for Delivery of Video Programming, Eighth Annual Report, 17 F.C.C.R. 1244, 1277–82 ¶¶ 68–77, 1291–92 ¶¶ 99–102 (2002). 588. See supra notes 153–54 and accompanying text. 589. See Yoo, supra note 153, at 18–21, 48, 102–03, 113–14. 590. 17 U.S.C. § 122 (2000); 47 U.S.C. § 338(a)(1) (2000). 591. See Yoo, supra note 153, at 22–27, 49–50, 103, 115–20. 592. See Advanced Television Sys. & Their Impact Upon Existing Television Broad. Serv., Fifth Report and Order, 12 F.C.C.R. 12809, 12811–12 ¶ 5, 12820–21 ¶¶ 27–29 (1997). 592. See Advanced Television Sys. & Their Impact Upon Existing Television Broad. Serv., Fifth Report and Order, 12 F.C.C.R. 12809, 12811–12 ¶ 5, 12820–21 ¶¶ 27–29 (1997). 593. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 22; see also id. at 20 (“Indeed, mere availability of [serious coverage of public issues] may not be enough if few citizens take advantage of it, and if most viewers and readers are content with programming and news accounts that do not deal well or in depth with public issues.”). 594. FISS, LIBERALISM DIVIDED, supra note 24, at 40. 595. Id. at 151. Fiss also notes that this explains why cable public access channels are an insufﬁcient response to the needs of robust public debate. Id. 596. See Fiss, supra note 24, at 1217; Sunstein, supra note 23, at 527–31. But see Owen Fiss, In Search of a New Paradigm, 104 YALE L.J. 1613, 1614–15 (1995) (acknowledging that his television- based paradigm may already be obsolete). 597. PEW RESEARCH CENTER FOR THE PEOPLE AND THE PRESS, INTERNET SAPPING BROADCAST NEWS AUDIENCE, 1, 5–6, 12–13 (2000) (reporting a 50% drop in consumption of broadcast network news and a corresponding growth in importance of the internet as a source of news between 1993 and 2000). The most recent Biennial News Consumption Survey indicates that these trends have begun to level off. 1. Empirical Assumptions About Television’s Role in Democracy At the same time, must-carry stiﬂes the development of alternative media by placing a drag on the revenues earned by broadcasting’s primary competitors. This same dynamic is currently being replayed with respect to direct broad- cast satellite (DBS) systems. One would think that policymakers would wel- come DBS as the solution to a range of policy problems. The natural monopoly characteristics of local cable service have long been a major focus of Congress and the FCC. Other technologies had vied with cable in the market for multichan- nel video program distribution, but all of them came up short.587 DBS thus represents the ﬁrst technology capable of breaking local cable monopolies. In addition, DBS represents the ﬁrst video technology with a national footprint. As a result, it is in a better position than any other technology to exploit the cost efﬁciencies that accompany national distribution.588 Rather than embrace DBS, however, policymakers have opted to regulate it to preserve free, over-the-air broadcasting.589 As of the beginning of 2002, DBS companies that wish to carry programming from the major networks are subject to the must-carry require- ments that are quite similar to those imposed on local cable operators.590 This result not only frustrates DBS’s ability to realize the efﬁciencies associated with national distribution, but it also further entrenches broadcasting’s role in our society. The deployment of digital television appears to have fallen into the same pattern.591 Concerns about the need to preserve free, over-the-air television have led Congress and the FCC to deploy digital television by doubling the amount of spectrum given to each incumbent broadcast station.592 This has the inevi- 344 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 [Vol. 91:245 table effect of further cementing in place the role that broadcasting plays in the democratic process. In the long run, Fiss’s and Sunstein’s position on promoting access to television may be tautological. Positing the role that free, over-the-air television plays in the democratic process justiﬁes certain regulatory steps to preserve the broadcast industry, and those regulations in turn become the primary means by which the role that broadcasting plays is created and preserved. Although answering why television represents the best medium for fostering democratic discourse may render such a result defensible, Fiss and Sunstein leave this question unaddressed. 593. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 22; see also id. at 20 (“Indeed, mere availability of [serious coverage of public issues] may not be enough if few citizens take advantage of it, and if most viewers and readers are content with programming and news accounts that do not deal well or in depth with public issues.”). 597. PEW RESEARCH CENTER FOR THE PEOPLE AND THE PRESS, INTERNET SAPPING BROADCAST NEWS AUDIENCE, 1, 5–6, 12–13 (2000) (reporting a 50% drop in consumption of broadcast network news and a corresponding growth in importance of the internet as a source of news between 1993 and 2000). The most recent Biennial News Consumption Survey indicates that these trends have begun to level off. 2. Technological Change and the Limits of Civic Republicanism Fiss and Sunstein recognize that it is not enough for broadcasters simply to provide greater quantities of the types of programming that they believe pro- mote good citizenship. To promote the deliberative process in the manner Fiss and Sunstein envision, people actually have to watch that programming. In Sunstein’s words: “It is also important to ensure not merely that diversity is available, but also that a signiﬁcant part of the citizenry is actually exposed to diverse views about public issues.”593 Fiss is even more succinct: “The issue is not market failure but market reach.”594 In Fiss’s view, an unwatched program is no better than an unread book buried in a library.595 In other words, Fiss and Sunstein’s real concern is audience failure, not market failure. The problem is not that viewers cannot get what they want, but rather that they do not want the right things. Their policy prescription thus goes far beyond ensuring that better programming options are available. Steps must be taken to ensure that individu- als actually avail themselves of those options. In the end, however, even this concern may not be able to withstand the mounting tide of technological change. Notwithstanding Fiss’s and Sunstein’s stated belief that television will retain its unique social role for the foreseeable future,596 a recent study suggests that more than thirty percent of U.S. citizens no longer regard television as their primary source of news and information.597 In fact, this possibility has forced Sunstein to revise his views by adding a new g g p y In fact, this possibility has forced Sunstein to revise his views by adding a new 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 345 section to the paperback version of his book republic.com acknowledging that the emergence of the Internet and its accompanying ability to allow individuals to tailor their information environment may in fact be promoting rather than impeding democracy.598 In addition, Fiss and Sunstein do not adequately confront the growth of cable television and other alternative means of delivering television programming. As noted earlier, the average U.S. household receives thirteen over-the-air chan- nels, and approximately eighty percent of all U.S. ( p ) p g p 605. Press Release, FCC, Furchtgott-Roth Seeks to Clarify and Separate Big Government’s Interest from the Public Interest in the Debate Over the Debates, at http://ftp.fcc.gov/Speechs/Furchtgott_Roth/ Statements/2000/sthfr057.doc (last visited Feb. 10, 2003). PEW RESEARCH CENTER FOR THE PEOPLE AND THE PRESS, PUBLIC’S NEWS HABITS LITTLE CHANGED BY SEPT. 11, at 2–3, 6–7 (2002). 604. See Kennard, Remarks at the Museum of Television and Radio, supra note 97; Press Release, FCC, FCC Commissioner Susan Ness Decries Decisions of NBC and Fox Networks Not to Air the First Presidential Debate (Sept. 29, 2000), available at http://www.fcc.gov/Speeches/ Ness/States/2000/stsn032.html. 598. CASS SUNSTEIN, REPUBLIC.COM 203–12 (paperback ed. 2002). y p 602. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 76; see also SUNSTEIN, supra note 598, at 3–22 (criticizing growth of “daily me,” in which people receive customized news via Internet). 603. Gardbaum, supra note 441, at 383–84. 599. See supra notes 175, 178 and accompanying text. 600. For a humorous (but accurate) description of channel surﬁng, see Balkin, supra note 432, at 1935–42. PEW RESEARCH CENTER FOR THE PEOPLE AND THE PRESS, PUBLIC’S NEWS HABITS LITTLE CHANGED BY SEPT. 11, at 2–3, 6–7 (2002). 598. CASS SUNSTEIN, REPUBLIC.COM 203–12 (paperback ed. 2002). 599. See supra notes 175, 178 and accompanying text. 600. For a humorous (but accurate) description of channel surﬁng, see Balkin, supra note 432, at 1935–42. 601. Cass Sunstein, The First Amendment in Cyberspace, 104 YALE L.J. 1757, 1786 (1995). 602. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 76; see also SUNSTEIN, supra note 598, at 3–22 (criticizing growth of “daily me,” in which people receive customized news via Internet). 603. Gardbaum, supra note 441, at 383–84. 604. See Kennard, Remarks at the Museum of Television and Radio, supra note 97; Press Release, FCC, FCC Commissioner Susan Ness Decries Decisions of NBC and Fox Networks Not to Air the First Presidential Debate (Sept. 29, 2000), available at http://www.fcc.gov/Speeches/ Ness/States/2000/stsn032.html. 605. Press Release, FCC, Furchtgott-Roth Seeks to Clarify and Separate Big Government’s Interest from the Public Interest in the Debate Over the Debates, at http://ftp.fcc.gov/Speechs/Furchtgott_Roth/ Statements/2000/sthfr057.doc (last visited Feb. 10, 2003). 599. See supra notes 175, 178 and accompanying text. 601. Cass Sunstein, The First Amendment in Cyberspace, 104 YALE L.J. 1757, 1786 (1995). THE GEORGETOWN LAW JOURNAL Sunstein concedes that people cannot be compelled to watch public affairs programming,606 but suggests that “careful judgments at the level of implemen- tation” can prevent unwilling viewers from turning off their televisions.607 He hopes some members of the general public will become accidentally exposed to higher-quality programming and thereby acquire a greater taste for it.608 Given the ease with which channels can be changed, it is hard to see how any degree of care in implementation could ensure that audiences receive sufﬁcient expo- sure to the type of programming that Fiss and Sunstein believe is essential. The impending arrival of video-on-demand promises to exacerbate the problem by giving individuals still greater control over their viewing environment.609 Noth- ing prevents unwilling viewers from switching to alternative programming or simply tuning out altogether.610 Thus, even if Fiss and Sunstein were able to convince Congress, the FCC, and the courts to sanction their proposals, techno- logical developments will likely make it impossible for television to serve Fiss’s and Sunstein’s purposes. So what if the less-educated members of the public don’t want to view the highbrow material? It will be an uplifting, educational experience. They might even learn to like it. After all, mass taste is just a matter of socialized preference. And, if the less educated turn off the set, that’s OK, too. Perhaps they watch too much TV anyway. V. POSSIBLE EXPLANATIONS FOR THE PERSISTENCE OF BROADCAST-STYLE REGULATION The collapse of the traditional bases for according a lower level of First Amendment scrutiny to broadcasting and the inability of civic republican theories to offer a revisionist justiﬁcation for doing so raises the question why the current jurisprudential state of affairs has persisted. In this Part, I will sketch out some possible explanations. 609. See supra note 332 and accompanying text. 610. Sunstein contradictorily notes that to the extent that the programming is of low quality and does not promote democratic values, little would be lost if viewers simply tuned out. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 89; see also id. at 77 (suggesting it “would not necessarily be bad” if attempts to promote democratically-oriented programming caused people to turn off their televi- sions). This observation conﬂicts with Sunstein’s acknowledgement that, for his system to work, there must be some means for generating a shared civic culture. It is also vulnerable to charges of elitism and paternalism, aptly captured in Burt Neuborne’s rendering of Sunstein’s argument: 608. Id. at 89–90. 606. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 22, 70. 607 Id 76 77 Neuborne, supra note 459, at 440. 2. Technological Change and the Limits of Civic Republicanism households subscribe to cable television or some other MVPD.599 Although this increase in channel capacity (rendered all the more accessible by the ubiquitous presence of the remote control600) has unquestionably increased the diversity of the programming available, it has simultaneously reduced television’s ability to play the transfor- mative role that Fiss and Sunstein envision. Sunstein responds to this possibility with abject horror. The possibility that an individual could “design his own communications universe” in which that person “could see those things that he wanted to see, and only those things” and could “screen out ideas, facts, or accounts that [that person] ﬁnd[s] disturb- ing”601 could lead to the “elimination of a shared civic culture, which contem- plates a degree of commonality among the citizenry.”602 His reaction underscores the “complex, indeed, ambivalent, attitude towards diversity of opinions” re- ﬂected in most civic republican and communitarian conceptions of politics 603 Sunstein responds to this possibility with abject horror. The possibility that an individual could “design his own communications universe” in which that person “could see those things that he wanted to see, and only those things” and could “screen out ideas, facts, or accounts that [that person] ﬁnd[s] disturb- ing”601 could lead to the “elimination of a shared civic culture, which contem- plates a degree of commonality among the citizenry.”602 His reaction underscores the “complex, indeed, ambivalent, attitude towards diversity of opinions” re- ﬂected in most civic republican and communitarian conceptions of politics.603 He loathes the homogenizing effects of large media institutions, and yet his theory depends on large, centralized conduits through which to reach the market. Indeed, it is this ambivalence that underlies the regulatory outcry surrounding the initial decision by NBC and Fox not to air the ﬁrst presidential debate during the 2000 election cycle.604 One might think that in an era of supposed scarcity and limited diversity, it would be anathema to dedicate one of the few major networks to programming that was completely duplicative of what was on the other major networks.605 And yet that is precisely what the FCC pressured NBC and Fox to do. 346 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL 606. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 22, 70. 607. Id. at 76–77. 608. Id. at 89–90. 609. See supra note 332 and accompanying text. 610. Sunstein contradictorily notes that to the extent that the programming is of low quality and does not promote democratic values, little would be lost if viewers simply tuned out. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 89; see also id. at 77 (suggesting it “would not necessarily be bad” if attempts to promote democratically-oriented programming caused people to turn off their televi- sions). This observation conﬂicts with Sunstein’s acknowledgement that, for his system to work, there must be some means for generating a shared civic culture. It is also vulnerable to charges of elitism and paternalism, aptly captured in Burt Neuborne’s rendering of Sunstein’s argument: So what if the less-educated members of the public don’t want to view the highbrow material? It will be an uplifting, educational experience. They might even learn to like it. After all, mass taste is just a matter of socialized preference. And, if the less educated turn off the set, that’s OK, too. Perhaps they watch too much TV anyway. Neuborne, supra note 459, at 440. 607. Id. at 76–77. A. THE NASCENT STATE OF FIRST AMENDMENT DOCTRINE One possible explanation for the rise of the technology-speciﬁc First Amend- ment is the embryonic state of First Amendment doctrine at the time NBC v. 610. Sunstein contradictorily notes that to the extent that the programming is of low quality and does not promote democratic values, little would be lost if viewers simply tuned out. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 89; see also id. at 77 (suggesting it “would not necessarily be bad” if attempts to promote democratically-oriented programming caused people to turn off their televi- sions). This observation conﬂicts with Sunstein’s acknowledgement that, for his system to work, there must be some means for generating a shared civic culture. It is also vulnerable to charges of elitism and paternalism, aptly captured in Burt Neuborne’s rendering of Sunstein’s argument: 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 347 United States and the other seminal broadcasting decisions were issued.611 During the 1930s and early 1940s, neither the FRC nor the courts believed that commercially oriented entertainment merited any First Amendment protection. For example, the Supreme Court in 1915 held that ﬁlm fell outside the ambit of the First Amendment on the ground that “the exhibition of moving pictures is a business, pure and simple, originated and conducted for proﬁt, like other spectacles, not to be regarded . . . as part of the press of the country or as organs of public opinion.”612 The FRC sounded a similar theme in declaring that “[e]ntertainment such as music is not ‘speech’ in the sense in which it is used in the [F]irst [A]mendment to the Federal Constitution.”613 Indeed, the FRC denigrated all forms of direct advertising as “usually offensive to the listening public.”614 Formal recognition that motion pictures fell within the ambit of the First Amendment did not occur until 1948.615 Commercial speech did not receive recognition until 1976.616 In addition, the courts’ appreciation for the particular dangers of licensing was not fully formed during the 1930s and 1940s. Although the Supreme Court had recognized the dangers of licensing when it acted as a prior restraint,617 it had only begun to acknowledge the manner in which subsequent punishment can also suppress speech.618 Indeed, it is telling that in a decision roughly contemporaneous with the Court’s landmark decision in Near v. Minnesota ex rel. Olson, the D.C. p 613. FRC SECOND ANNUAL REPORT, supra note 70, at 161; see also id. at 168 (criticizing playing of phonograph records over the air in part because practice was driven solely by business motive of facilitating advertising); Hazlett, supra note 115, at 152 (quoting Herbert Hoover’s statement at the Fourth National Radio Conference in 1925 as saying that “no one can raise a cry of deprivation of free speech if he is compelled to prove that there is something more than naked commercial selﬁshness in his purpose”). 615. See United States v. Paramount Pictures, Inc. 334 U.S. 131, 166 (1948) (“We have no doubt that moving pictures, like newspapers and radio, are included in the press whose freedom is guaranteed by the First Amendment.”); see also Joseph Burstyn, Inc. v. Wilson, 343 U.S. 495, 502–03 (1952) (formally overruling Mutual Film). ( ) ( 612. Mut. Film Corp. v. Indus. Comm’n, 236 U.S. 230, 244 (1915). 614. FRC SECOND ANNUAL REPORT, supra note 70, at 168. 619. KFKB Broad. Ass’n v. FRC, 47 F.2d 670, 672 (D.C. Cir. 1931); accord Trinity Methodist Church, S. v. FRC, 62 F.2d 850, 851, 853 (D.C. Cir. 1932). 611. See NBC v. United States, 319 U.S. 190 (1943); Trinity Methodist Church, S. v. FRC, 62 F.2d 850 (D.C. Cir. 1932); KFKB Broad. Ass’n v. FRC, 47 F.2d 670 (D.C. Cir. 1931). ( y g ) 616. See Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 770 (1976). ( ) 618. See Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713–16 (1931); Weinberg, supra note 26, at 1137. ( ) 617. See, e.g., Lovell v. City of Grifﬁn, 303 U.S. 444 (1938); Respublica v. Oswald, 1 U.S. (1 Dall.) 319, 325, 328 n. (1788). 611. See NBC v. United States, 319 U.S. 190 (1943); Trinity Methodist Church, S. v. FRC, 62 F.2d 0 (D.C. Cir. 1932); KFKB Broad. Ass’n v. FRC, 47 F.2d 670 (D.C. Cir. 1931). 620. See Martin H. Redish, The Content Distinction in First Amendment Analysis, 34 STAN. L. REV. 113, 123–25 (1981); Paul B. Stephan III, The First Amendment and Content Discrimination, 68 VA. L. REV. 203, 218–27 (1982). 621. See Note, Indirect Censorship of Radio Programs, 40 YALE L.J. 967, 968 (1931) (noting that “the power to revoke or refuse the renewal of a license is in many cases so effective a means of censorship as to make unconvincing any legalistic distinction between ‘previous restraint’ and a refusal to renew a license because of the character of past program”). 622. NBC v. United States, 319 U.S. 190, 226 (1943). 623. See, e.g., id. at 216–17; BRUCE M. OWEN, TELEVISION ECONOMICS 139 (1975); POOL, supra note 51, at 142, 146; POWE, supra note 51, at 201; 624. See Coase, Social Cost, supra note 118; Garrett Hardin, The Tragedy of the Commons, 162 SCIENCE 1243 (1968). 625. See supra notes 116–19 and accompanying text. A. THE NASCENT STATE OF FIRST AMENDMENT DOCTRINE Circuit upheld the FRC’s decision to refuse to renew a license on the grounds that the First Amendment represented nothing more than a prohibition of prior restraints and did not prevent the FRC from exercising its “undoubted right to take note of appellant’s past conduct” when deciding whether to renew a license.619 And the content distinction and compelled speech 615. See United States v. Paramount Pictures, Inc. 334 U.S. 131, 166 (1948) (“We have no doubt that moving pictures, like newspapers and radio, are included in the press whose freedom is guaranteed by the First Amendment.”); see also Joseph Burstyn, Inc. v. Wilson, 343 U.S. 495, 502–03 (1952) (formally overruling Mutual Film). 616. See Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 770 (1976). 617. See, e.g., Lovell v. City of Grifﬁn, 303 U.S. 444 (1938); Respublica v. Oswald, 1 U.S. (1 Dall.) 319, 325, 328 n.* (1788). 618. See Near v. Minnesota ex rel. Olson, 283 U.S. 697, 713–16 (1931); Weinberg, supra note 26, at 1137. 619. KFKB Broad. Ass’n v. FRC, 47 F.2d 670, 672 (D.C. Cir. 1931); accord Trinity Methodist Church, S. v. FRC, 62 F.2d 850, 851, 853 (D.C. Cir. 1932). 348 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 did not emerge as major First Amendment concerns until much later in the century.620 Although it is plausible that these early courts failed to appreciate the First Amendment dangers posed by the broadcast approach to regulation, such an argument faces its share of problems. Contemporaneous commentators were quick to point out the obvious dangers of the courts’ decisions.621 The Supreme Court’s opinion in NBC signaled its recognition of the problem when it acknowl- edged that First Amendment concerns might arise if the FCC were “to choose among applicants upon the basis of their political, economic or social views, or upon any other capricious basis.”622 In any event, even assuming that the courts’ failure to appreciate the dangers to free speech imposed by the broadcast regime was justiﬁed in the 1930s and 1940s, there is little justiﬁcation for continuing that ignorance into the present day. Regardless whether the early judicial decisions were justiﬁed when they were rendered, it should be clear that the state of the doctrine at that time can no longer justify adhering to them now. 620. See Martin H. Redish, The Content Distinction in First Amendment Analysis, 34 STAN. L. REV. 113, 123–25 (1981); Paul B. Stephan III, The First Amendment and Content Discrimination, 68 VA. L. REV. 203, 218–27 (1982). 626. Hazlett, supra note 115, at 143–45, 149–52, 160, 162–63. 627. 309 U.S. 134 (1940). 628. Id. at 142. 629. Id. 630. Id. at 138. 631. Id. 632. NBC v. United States, 319 U.S. 190 (1943). 633. Id. at 219. THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 349 disagree with this assessment. Thomas Hazlett’s careful history of the circum- stances surrounding the enactment of the Radio Act of 1927 suggests that Congress was well aware of the possibility that a system of property rights could obviate the need for governmental allocation and that just such a regime was in fact in the process of evolving.626 Subsequent developments have rendered this debate moot. The FCC’s recent success with auctions and the longstanding existence of a robust market for broadcast stations has since validated the idea that spectrum licenses can be allocated effectively through nonadministrative means. No matter how the debate over the state of economic thinking at the time of the enactment of the Communications Act of 1934 is resolved, continued adherence to the technology- speciﬁc approach to the First Amendment can no longer be defended based on the lack of alternatives to administrative licensing. B. THE EMERGENCE OF THE ECONOMICS OF PROPERTY RIGHTS Another possibility is that Congress thought that it had no choice but to adopt the type of administrative licensing scheme associated with broadcast regula- tion. Indeed, many scholars have suggested that Congress’s decision to adopt the Broadcast Model followed from its belief that administrative licensing represented the only viable alternative to abandoning the spectrum as a re- source.623 In many ways, the claim seems plausible. Coase’s seminal work on the ability of a property rights regime to resolve many matters that were previously thought to require regulation and Garret Hardin’s pathbreaking article, The Tragedy of the Commons, did not appear until several decades after these basic principles had been established.624 The skeptical reaction that greeted Coase’s proposal conﬁrms its perceived novelty at the time.625 It would arguably be somewhat anachronistic therefore to expect regulatory authorities that were taking the ﬁrst steps to regulate broadcasting to have appreciated the possibility of nonadministrative allocation. At the same time, there are those who would 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT C. PROGRESSIVE FAITH IN AGENCY EXPERTISE The New Deal era’s faith in administrative discretion provides another pos- sible explanation for the courts’ willingness to condone broadcast-style regula- tion. Contemporary faith in technocracy is perhaps best reﬂected in the Supreme Court’s decision in FCC v. Pottsville Broadcasting Co.,627 which was authored by Landis prote`ge´ and former New Dealer Felix Frankfurter. According to Frankfurter, modern administrative agencies like the FCC are “a response to the felt need of governmental supervision over economic enterprise.”628 The dyna- mism of modern industrial society left both Congress and the courts poorly situated to undertake such a task.629 As a result, rather than being subject to stringent legal constraints, agencies should be given broad mandates like the public interest standard, which Frankfurter regarded “as concrete as the compli- cated factors for judgment in such a ﬁeld of delegated authority permit.”630 It provided “a supple instrument for the exercise of discretion by the expert body which Congress has charged to carry out its legislative policy.”631 Frankfurter offered similar arguments in his landmark opinion in NBC.632 When faced with “a ﬁeld of regulation which was both new and dynamic,” Congress found it best to give the FCC “not niggardly but expansive powers” to promote radio communications.633 As Congress had learned in areas “far less ﬂuid and dynamic than radio,” the best course was “to deﬁne broad areas for regulation and to establish standards for judgment adequately related in their 350 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 application to the problems to be solved.”634 Although Congress did not give the FCC “unfettered discretion,” it also did not attempt to offer an itemized catalogue of statutory directions. Instead, Congress left it to the FCC to design the appropriate measures.635 When the FCC’s decision was submitted for review, the Court declined to second-guess the FCC’s determinations so long as the agency acted within its statutory authority and supported its ﬁndings with evidence.636 To the extent that litigants wished to challenge the policy of a particular regulation, the Court declared that it had “neither technical compe- tence nor legal authority to pronounce upon the wisdom of the course taken by the Commission.”637 Any such arguments were more properly directed to the FCC itself.638 Such a conclusion may be unsurprising as a matter of contemporary adminis- trative law, but what is more surprising is the Court’s willingness to incorporate such administrative deference into constitutional law. 642. See generally Thomas W. Merrill, Capture Theory and the Courts:1967–1983, 72 CHI.-KENT L. REV. 1039 (1997); Reuel E. Schiller, Enlarging the Administrative Polity: Administrative Law and the Changing Deﬁnition of Pluralism, 1945–1970, 53 VAND. L. REV. 1389 (2000); Richard B. Stewart, The Reformation of American Administrative Law, 88 HARV. L. REV. 1669 (1975); Keith Werhan, The Neoclassical Revival in Administrative Law, 44 ADMIN. L. REV. 567 (1992). 640. See also CBS, Inc. v. Democratic Nat’l Committee, 412 U.S. 94, 102–03 (1973) (plurality opinion) (arguing that, although stopping short of true deference, “in evaluating the First Amendment claims of respondents, we must afford great weight to . . . the experience of the Commission”). 639. Id. at 227 (emphasis added). 634. Id. at 219–20. 635. Id. at 219. 636. Id. at 224. 637. Id. 638. Id. at 224. 639. Id. at 227 (emphasis added). 640. See also CBS, Inc. v. Democratic Nat’l Committee, 412 U.S. 94, 102–03 (1973) (plurality opinion) (arguing that, although stopping short of true deference, “in evaluating the First Amendment claims of respondents, we must afford great weight to . . . the experience of the Commission”). 641. Schiller, supra note 35, at 2–3, 21–51. 642. See generally Thomas W. Merrill, Capture Theory and the Courts:1967–1983, 72 CHI.-KENT L. REV. 1039 (1997); Reuel E. Schiller, Enlarging the Administrative Polity: Administrative Law and the Changing Deﬁnition of Pluralism, 1945–1970, 53 VAND. L. REV. 1389 (2000); Richard B. Stewart, The Reformation of American Administrative Law 88 HARV L REV 1669 (1975); Keith Werhan The 641. Schiller, supra note 35, at 2–3, 21–51. 634. Id. at 219–20. 635. Id. at 219. 636. Id. at 224. 637. Id. 638. Id. at 224. 639. Id. at 227 (emphasis added). 640. See also CBS, Inc. v. Democratic Nat’l Committee, 412 U.S. 94, 102–03 (1973) (plurality opinion) (arguing that, although stopping short of true deference, “in evaluating the First Amendment claims of respondents, we must afford great weight to . . . the experience of the Commission”). 648. Quill Corp. v. North Dakota, 504 U.S. 298, 317 (1992); accord Allied-Signal, Inc. v. Director, Div. of Taxation, 504 U.S. 768, 785–86 (1992); Hilton v. S.C. Pub. Rys. Comm’n, 502 U.S. 197, 202 (1991); United States v. Mason, 412 U.S. 391, 399–400 (1971); United States v. Title Ins. & Trust Co., 265 U.S. 472, 486 (1924); see also Hubbard v. United States, 514 U.S. 695, 714 (1995) (dicta) (“Stare decisis has special force when legislators or citizens ‘have acted in reliance on a previous decision, for in this instance overruling the decision would dislodge settled rights and expectations or require an extensive legislative response.’” (quoting Hilton v. S.C. Pub. Ry. Comm’n, 502 U.S. 197, 202 (1991)). g p (q g y , , ( )) 649. See, e.g., Agostini v. Felton, 521 U.S. 203, 235 (1997); Seminole Tribe v. Florida, 517 U.S. 44, 63 (1996); Payne, 501 U.S. at 828; Thomas v. Washington Gas Light Co., 448 U.S. 261, 272 n.18 (1980) (plurality opinion); United States v. Scott, 437 U.S. 82, 101 (1978); Edelman v. Jordan, 415 U.S. 651, 671 (1974); Glidden Co. v. Zdanok, 370 U.S. 530, 543 (1962); Smith v. Allwright, 321 U.S. 649, 647. Arizona v. Rumsey, 467 U.S. 203, 212 (1984); accord Dickerson v. United States, 530 U.S. 428, 443 (2000); United States v. IBM Corp., 517 U.S. 843, 856 (1996); Adarand Constructors, Inc. v. Pen˜a, 515 U.S. 200, 231 (1995); Patterson v. McLean Credit Union, 491 U.S. 164, 172 (1989); Welch v. Texas Dept. of Highways & Pub. Transp., 483 U.S. 468, 479 (1987). 643. See, e.g., FCC v. WNCN Listeners Guild, 450 U.S. 582, 596 (1981); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 810 (1978). 645. Payne v. Tennessee, 501 U.S. 808, 828 (1991). C. PROGRESSIVE FAITH IN AGENCY EXPERTISE In a brief passage near the end of the opinion, the NBC Court equated satisfaction of the public interest standard with First Amendment compliance. The Court stated, “The standard [the Communications Act of 1934] provided for the licensing of stations was the ‘public interest, convenience, or necessity.’ Denial of a station license on that ground, if valid under the Act, is not a denial of free speech.”639 The implication is clear: Proof of compliance with the statutory mandate by deﬁnition necessar- ily constituted proof of constitutionality. In effect, the First Amendment disap- peared as an independent source of judicial review.640 Coming during the highpoint of jurisprudential faith in administrative exper- tise, this holding reﬂects its time. Such administrative control of speech was once quite common.641 To modern commentators, however, it should seem inappropriate for a court to regard compliance with the requirements of adminis- trative law as necessarily constituting compliance with the mandates of the First Amendment. Experience with the administrative process and growing concerns about agency capture have shattered faith in technocracy that characterized the Progressive era.642 In addition, although the Court has continued to defer to the 2003] 351 THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT FCC’s judgments about how best to promote the public interest,643 courts now recognize the impropriety of deferring to agencies on matters of constitutional law.644 D. STARE DECISIS 644. See Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1988) (declining to defer when agency interpretation of statute raised serious First Amend- ment concerns); see also Solid Waste Agency v. U.S. Army Corps of Eng’rs, 531 U.S. 159, 173–74 (2001) (5-4 decision) (declining to defer when agency interpretation of a statute raised serious constitutional questions). 643. See, e.g., FCC v. WNCN Listeners Guild, 450 U.S. 582, 596 (1981); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 810 (1978). 644. See Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1988) (declining to defer when agency interpretation of statute raised serious First Amend- ment concerns); see also Solid Waste Agency v. U.S. Army Corps of Eng’rs, 531 U.S. 159, 173–74 (2001) (5-4 decision) (declining to defer when agency interpretation of a statute raised serious constitutional questions). 645. Payne v. Tennessee, 501 U.S. 808, 828 (1991). 646. Id. at 827. 647. Arizona v. Rumsey, 467 U.S. 203, 212 (1984); accord Dickerson v. United States, 530 U.S. 428, 443 (2000); United States v. IBM Corp., 517 U.S. 843, 856 (1996); Adarand Constructors, Inc. v. Pen˜a, 515 U.S. 200, 231 (1995); Patterson v. McLean Credit Union, 491 U.S. 164, 172 (1989); Welch v. Texas Dept. of Highways & Pub. Transp., 483 U.S. 468, 479 (1987). 648. Quill Corp. v. North Dakota, 504 U.S. 298, 317 (1992); accord Allied-Signal, Inc. v. Director, Div. of Taxation, 504 U.S. 768, 785–86 (1992); Hilton v. S.C. Pub. Rys. Comm’n, 502 U.S. 197, 202 (1991); United States v. Mason, 412 U.S. 391, 399–400 (1971); United States v. Title Ins. & Trust Co., 265 U.S. 472, 486 (1924); see also Hubbard v. United States, 514 U.S. 695, 714 (1995) (dicta) (“Stare decisis has special force when legislators or citizens ‘have acted in reliance on a previous decision, for in this instance overruling the decision would dislodge settled rights and expectations or require an extensive legislative response.’” (quoting Hilton v. S.C. Pub. Ry. Comm’n, 502 U.S. 197, 202 (1991)). 649. See, e.g., Agostini v. Felton, 521 U.S. 203, 235 (1997); Seminole Tribe v. Florida, 517 U.S. 44, 63 (1996); Payne, 501 U.S. at 828; Thomas v. Washington Gas Light Co., 448 U.S. 261, 272 n.18 (1980) (plurality opinion); United States v. Scott, 437 U.S. 82, 101 (1978); Edelman v. Jordan, 415 U.S. 651, 671 (1974); Glidden Co. v. Zdanok, 370 U.S. 530, 543 (1962); Smith v. Allwright, 321 U.S. 649, 646. Id. at 827. 643. See, e.g., FCC v. WNCN Listeners Guild, 450 U.S. 582, 596 (1981); FCC v. Nat’l Citizens Comm. for Broad., 436 U.S. 775, 810 (1978). 644. See Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1988) (declining to defer when agency interpretation of statute raised serious First Amend- ment concerns); see also Solid Waste Agency v. U.S. Army Corps of Eng’rs, 531 U.S. 159, 173–74 (2001) (5-4 decision) (declining to defer when agency interpretation of a statute raised serious constitutional questions). 645. Payne v. Tennessee, 501 U.S. 808, 828 (1991). 646. Id. at 827. 665 & n.10 (1944); St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 93 (1936) (Stone and Cardozo, JJ., concurring in result); Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 405–10 (1932) (Brandeis, J., dissenting). 650. Planned Parenthood v. Casey, 505 U.S. 833, 854 (1992) (citing Burnet, 285 U.S. at 412 (Brandeis, J., dissenting)); see also Vasquez v. Hillery, 474 U.S. 254, 266 (1986) (recognizing propriety of overruling cases in light of newly ascertained facts and experience). 651. See generally Matthew Spitzer, Dean Krattenmaker’s Road Not Taken: The Political Economy of Broadcasting in the Telecommunications Act of 1996, 29 CONN. L. REV. 353, 370–71 (1996); Hazlett, supra note 571, at 941–42; Thomas W. Hazlett, Explaining the Telecommunications Act of 1996: Comment on Thomas G. Krattenmaker, 29 CONN. L. REV. 217, 235–36 (1996); Robinson, supra note 249, at 919. 652. See supra notes 136–49 and accompanying text. D. STARE DECISIS Another possible justiﬁcation for the courts’ continued willingness to uphold the constitutionality of the Broadcast Model is stare decisis. Although not an “inexorable command,”645 stare decisis “is the preferred course because it promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process.”646 As a result, the Court has stated that “any departure from the doctrine of stare decisis demands special justiﬁcation.”647 In addition, the Court has recognized that stare decisis is particularly strong when a precedent has “engendered substantial reliance and has become part of the basic framework of a sizable industry.”648 In light of these considerations, it is understandable why the Court would be loath to disturb its prior holdings in this area. Not only would overruling cases detract from the integrity of the rule of law, it would also disrupt a sizable industry erected in reliance on the Court’s prior decisions. At the same time, other considerations diminish stare decisis as a justiﬁcation for the Court’s refusal to overrule its broadcast precedents. For example, the Court has recognized that stare decisis is at its “weakest” in matters of constitu- tional law because only constitutional amendment can correct interpretive errors.649 In addition, the Court has found it appropriate to overrule cases when THE GEORGETOWN LAW JOURNAL 352 [Vol. 91:245 the “facts have so changed, or come to be seen so differently, as to have robbed the old rule of signiﬁcant application or justiﬁcation.”650 The constitutional basis of the precedents in question, combined with the dramatic technological transformation the broadcast industry, thus leaves the Court’s key broadcasting decisions open to reconsideration. Thus, concerns about stare decisis are not likely to be the reason preventing the Court from revisiting these precedents. Even if stare decisis did explain the persistence of the current broadcasting regime, it would not justify extending such an erroneous regime to other media. 653. Rubin, supra note 28, at 691–97. g g y p 651. See generally Matthew Spitzer, Dean Krattenmaker’s Road Not Taken: The Political Economy of Broadcasting in the Telecommunications Act of 1996, 29 CONN. L. REV. 353, 370–71 (1996); Hazlett, supra note 571, at 941–42; Thomas W. Hazlett, Explaining the Telecommunications Act of 1996: Comment on Thomas G. Krattenmaker, 29 CONN. L. REV. 217, 235–36 (1996); Robinson, supra note 249, at 919. 665 & n.10 (1944); St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 93 (1936) (Stone and Cardozo, JJ., concurring in result); Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 405–10 (1932) (Brandeis, J., dissenting). 650. Planned Parenthood v. Casey, 505 U.S. 833, 854 (1992) (citing Burnet, 285 U.S. at 412 (Brandeis, J., dissenting)); see also Vasquez v. Hillery, 474 U.S. 254, 266 (1986) (recognizing propriety of overruling cases in light of newly ascertained facts and experience). E. PUBLIC CHOICE THEORY Perhaps the most persuasive explanation for the Broadcast Model’s persis- tence has its roots in public choice theory. A growing body of scholarship has argued that the current structure of broadcast regulation is the product of rent seeking.651 Unlike other users of spectrum-based technologies, broadcasters receive their licenses for free. In addition, as noted earlier, the FCC has adopted policies that have restricted entry in a manner that has enhanced the value of those licenses still further.652 By creating such rents, Congress was able to protect politicians against any adverse impact that television might have over elections. This scheme also allowed policymakers to manipulate the regulatory process to distribute beneﬁts to favored constituencies. The tremendous beneﬁts that broadcasters receive from this arrangement led them not to challenge the imposition of such regulations in most cases. Broadcasters are all too aware that the elimination of such public interest obligations would destroy any justiﬁca- tion for continuing to receive such beneﬁts.653 The events surrounding the allocation of additional spectrum for digital television demonstrates the rent-seeking behavior surrounding broadcast regula- tion. During the early 1990s, the broadcast industry engaged in an extensive lobbying campaign to persuade policymakers to implement the transition to digital broadcasting by giving a second free channel to all existing television stations. 659. See Pub. Interest Obligations of TV Broad. Licensees, Notice of Inquiry, 14 F.C.C.R. 21633, 21648 ¶ 35 (1999); Kennard, Remarks at the Museum of Television and Radio, supra note 97; see also Statement by FCC Chairman William Kennard on EchoStar Announcement of Political Candidate Airtime (F.C.C. Oct. 2, 2000), available at http://www.fcc.gov/Speeches/Kennard/Statements/2000/ stwek079.html. 660. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 58; see also SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 215 (same). 658. See Kennard, Remarks at the Museum of Television and Radio, supra note 97; Hundt, supra note 36, at 1096 (calling for strengthening broadcasters obligations to provide children’s programming and free air time for political candidates as well as additional restrictions on indecent and violent programming). 657. See GORE COMMISSION REPORT, supra note 17, at 46–47, 56–59; REPORT TO CONGRESS, supra note 18, at 24–26. 656. See Policies & Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10718–23 ¶¶ 120–29 (1996); Christopher Stern, TV Makes History at the White House, BROAD. & CABLE, Mar. 4, 1996, at 5. 654. See Hazlett, supra note 571, at 939–40. 654. See Hazlett, supra note 571, at 939–40. p 655. Quoted in id. at 942 (internal quotation marks omitted). 654. See Hazlett, supra note 571, at 939–40. 655. Quoted in id. at 942 (internal quotation marks omitted). 656. See Policies & Rules Concerning Children’s Television Programming, Report and Order, 11 F.C.C.R. 10660, 10718–23 ¶¶ 120–29 (1996); Christopher Stern, TV Makes History at the White House, BROAD. & CABLE, Mar. 4, 1996, at 5. 657. See GORE COMMISSION REPORT, supra note 17, at 46–47, 56–59; REPORT TO CONGRESS, supra note 18, at 24–26. 658. See Kennard, Remarks at the Museum of Television and Radio, supra note 97; Hundt, supra note 36, at 1096 (calling for strengthening broadcasters obligations to provide children’s programming and free air time for political candidates as well as additional restrictions on indecent and violent programming). 659. See Pub. Interest Obligations of TV Broad. Licensees, Notice of Inquiry, 14 F.C.C.R. 21633, 21648 ¶ 35 (1999); Kennard, Remarks at the Museum of Television and Radio, supra note 97; see also Statement by FCC Chairman William Kennard on EchoStar Announcement of Political Candidate Airtime (F.C.C. Oct. 2, 2000), available at http://www.fcc.gov/Speeches/Kennard/Statements/2000/ stwek079.html. 660. SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 58; see also SUNSTEIN, PARTIAL CONSTITUTION, supra note 23, at 215 (same). 654. See Hazlett, supra note 571, at 939–40. 655 Quoted in id at 942 (internal quotation marks omitted) 654. See Hazlett, supra note 571, at 939 40. 655. Quoted in id. at 942 (internal quotation marks omitted). 664. See Ellen P. Goodman, Digital Television and the Allure of Auctions: The Birth and Stillbirth of DTV Legislation, 59 FED. COMM. L.J. 517, 533 (1997) (reporting estimates placing value of spectrum given away for digital broadcasting as being between $11 billion and $70 billion as well as placing estimates of value of analog channels projected to be returned as being between $20 billion and $132 billion). By way of comparison, the high-cost and low-income support programs for telephony cost approximately $3.2 billion in 2001. FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE, UNIVERSAL SERVICE MONITORING REPORT 1-38 tbl. 1.11 (2002) available at http://www.fcc.gov/Bureaus/Common- _Carrier/Reports/FCC-State_Link/Monitor/mrs02-0.pdf. Applying the risk-free discount rate of roughly 5%, the net present value of such a program is approximately $67 billion, which compares favorably with the $31–$202 billion in spectrum currently given to the broadcast industry for free. E. PUBLIC CHOICE THEORY This arrangement was placed in jeopardy when a bipartisan coalition of senators, led by then-Majority Leader Bob Dole, condemned the “license giveaway” as “corporate welfare” and held up passage of the Telecommunica- 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 353 tions Act of 1996 until it was agreed that the FCC would not issue any digital television licenses before Congress enacted spectrum reform.654 In the words of one FCC ofﬁcial, broadcasters then began “tripping all over themselves to give up their First Amendment rights.”655 In exchange for the cancellation of the Dole agreement, the broadcast industry immediately capitulated on two issues that it had long resisted: the imposition of quantitative guidelines for children’s television and the creation of a rating system.656 Since then, governmental actors have continued to pressure broadcasters to offer free air time to political candidates and to adopt a “voluntary” code of conduct reinforcing their commit- ment to public interest broadcasting.657 Indeed, both Reed Hundt and William Kennard were quite outspoken during their tenures as Chairmen of the FCC in demanding a greater commitment to the public interest in return for the spec- trum given away in the 1996 Act.658 Responding to such pressure, two of the major networks and many local broadcasters provided candidates with free airtime during the last election.659 To say that public choice theory can explain the persistence of the Broadcast Model is not to say that it can justify it. Simply put, the decision to give a second television channel to all incumbent broadcasters was a public policy disaster. The traditional justiﬁcation for doing so is to preserve the beneﬁts of free, over-the-air television to those households that cannot afford to pay for television. Sunstein offers a particularly strident version of this argument when he ridicules systems that base the ability to speak on people’s willingness to pay as “a bizarre parody of democratic aspirations” and “the stuff of science ﬁction rather than self-government.”660 Although the rhetoric of Sunstein’s declaration is alluring, the issue is not as simple as he suggests. In fact, it is quite common for our society to allocate any number of goods that are essential to democratic citizenship, including other media such as newspapers, employment, shelter, 354 [Vol. 91:245 [Vol. 91:245 [Vol. 661. Indeed, Sunstein’s sweeping statement can be appropriately regarded as calling for the same reordering of the relationship between the individual and the state sought by Fiss, in which the state is seen as having an afﬁrmative obligation to provide all individuals with the minimum requirements of personhood. See supra notes 415–21 and accompanying text. Although Sunstein recognizes the problem, SUNSTEIN, DEMOCRACY AND FREE SPEECH, supra note 23, at 269–70, aside from the ipse dixit quoted above, he fails to offer an explanation as to why this issue should be resolved any differently with respect to free speech than it was with respect to equal protection and due process. p p p q p p 662. The Lifeline program subsidizes individual households’ ability to pay for their monthly telephone bills. The Link-Up program helps defray the one-time costs associated with obtaining initial phone service. See Ross C. Ericksson et al., Targeted and Untargeted Subsidy Schemes: Evidence from Postdivestiture Efforts to Promote Universal Telephone Service, 41 J.L. & ECON. 477, 481–82 (1998). 663. Id. at 498. p 665. See Yoo, supra note 153, at 72–76. E. PUBLIC CHOICE THEORY 91:245 THE GEORGETOWN LAW JOURNAL and even basic sustenance, through a price mechanism.661 Even if one were to accept that all households should have access to television, the untargeted nature of the particular subsidy chosen renders it singularly cost ineffective. The basic problem is that giving away broadcast spectrum for free effectively subsidizes all viewers, not just those who need the subsidy to obtain access. Such untargeted programs are thus unnecessarily wasteful. By way of comparison, the FCC has addressed similar concerns with respect to telephone service through direct subsidy programs.662 A recent empirical study of these programs concluded that targeted subsidies are up to ﬁve times more effective than untargeted subsidies in increasing household access.663 Indeed, given the comparability in cost of local telephone service and cable service, the telephone subsidy programs provide a useful benchmark for estimating the cost of an analogous subsidy for television. Even under conserva- tive assumptions, the cost of a similarly targeted program for television would be considerably less than the costs associated with giving spectrum away to digital broadcasters.664 In addition, the use of untargeted subsidies unnecessarily distorts secondary markets. On the supply side, the decision to give away broadcast spectrum for free inevitably raises the cost of spectrum for other uses. In short, there is nothing “free” about free over-the-air television. The public bears the costs by paying higher fees for cellular telephony, 3G wireless, and other spectrum- based technologies.665 On the demand side, altering the relative prices of the various spectrum-based services makes broadcasting artiﬁcially appealing from an economic standpoint. Because these prices do not reﬂect the true costs of these goods, these differences will inevitably cause consumers to deviate from 2003] THE TECHNOLOGY-SPECIFIC FIRST AMENDMENT 355 the most efﬁcient product mix.666 This effect is exacerbated by other regulatory features, such as must-carry, that are designed to promote free, over-the-air television at the expense of cable and other forms of pay television. By reducing the revenue generated by cable systems, must-carry causes the price of cable access to rise. The untargeted nature of the cross subsidy implicit in must-carry thus has the counterproductive effect of thwarting the very desire to increase access to television that underlies the subsidy program’s origins.667 Finally, giving away spectrum is also singularly ineffective as a means for promoting particular types of programming. 667. See id. at 499. Cross subsidies that penalize one subsector of an industry to beneﬁt another are also somewhat problematic from the standpoint of fairness, in that rarely is the penalized subsector responsible for creating the problem being redressed. Doing so makes about as much as sense as taxing cable operators to pay for the construction of new public schools. To the extent that general concerns of public welfare form the basis for the subsidy program, those subsidies should be ﬁnanced out of general revenues. E. PUBLIC CHOICE THEORY It would have been more cost effective for the government to auction the spectrum and use the revenue gathered to purchase a major television network and devote its entire output to children’s television and high-quality political speech.668 To make matters worse, by giving licenses only to incumbent licensees, the government doubled the amount of spectrum dedicated to television without increasing the level of competition in the industry or diversifying the ownership base of broadcast stations.669 Thus, although public choice theory may explain why such an arrangement represents a good deal for broadcasters, it fails to provide a reason for courts or policymakers to maintain the existing system of regulation. 666. See Ericksson et al., supra note 662, at 478 (summarizing basic criticisms of untargeted subsidies). 669. Spitzer, supra note 651, at 365–66; Thomas G. Krattenmaker, The Telecommunications Act of 1996, 29 CONN. L. REV. 123, 163–64 (1996). 668. Robinson, supra note 249, at 922 & n.83 (noting that in 1995 CBS was sold for $5.4 billion and in 1996 Turner Broadcasting was sold for $6.7 billion). CONCLUSION The same forces collapsing the technological distinctions embodied in current media policy are achieving the same effect on the technology-speciﬁc approach to the First Amendment. Not only do we now have a better understanding of the conceptual shortcomings of the rationales traditionally cited as supporting the constitutionality of the Broadcast Model, we must also candidly acknowledge the potentially transformative impact of technological innovations such as digital television, the V-chip, PVRs, spread spectrum, and video-on-demand. Furthermore, technological convergence is making it increasingly possible to convey virtually any type of communication through virtually any means of transmission. Indeed, once television networks complete the conversion to data packet switching, a single communication may be transmitted through several media simultaneously. As a result, the collapse of the technology-driven ap- proach to the First Amendment appears inevitable. The Supreme Court’s appar- 356 [Vol. 91:245 THE GEORGETOWN LAW JOURNAL [Vol. 91:245 ent reluctance to rely on the traditional rationales for applying a lower First Amendment standard to broadcasting may well signal their growing apprecia- tion of this reality. The attempts by Fiss and Sunstein to employ civic republican theory to provide revisionist rationales for the Broadcast Model in the end fail to justify upholding its constitutionality. Not only do they fail to come to grips with the First Amendment’s traditional respect for individual autonomy and traditional suspicion of government intervention, they also fail to offer a sufﬁcient articula- tion of how their theories will work in practice or how they will overcome the technological realities surrounding television. It is thus difﬁcult, if not impos- sible, to see how their theories would lead to the world that they envision. The Broadcast Model is a regulatory scheme in search of its own justiﬁcation. As such, it can properly be regarded as an example of the tendency for regulation to persist long after the reasons underlying its creation have since fallen away.670 I do not mean to suggest, however, that theoretical inelegance represents the only problem with the technology-speciﬁc approach to the First Amendment. On the contrary, as the decades-long ordeal to settle the First Amendment standard to govern cable television and the current uncertainty surrounding the constitutional framework that will be applied to digital televi- sion demonstrate, the problems resulting from the uncertainty created by the technology-speciﬁc First Amendment are real. 670. See STEPHEN BREYER, REGULATION AND ITS REFORM 286–87 (1982); 2 ALFRED E. KAHN, THE ECONOMICS OF REGULATION 126–27 (1971); Richard A. Posner, Natural Monopoly and Its Regulation, 21 STAN. L. REV. 548, 611–15 (1969). 671. See supra subsection I.A.2. CONCLUSION Even more sinister than the costs associated with this uncertainty is the novel way that I have identiﬁed in which regulation can serve as a constitutional justiﬁcation for additional regulation.671 Thus, the persistence of the Broadcast Model creates more than just transitional ambiguity; it also threatens to funda- mentally alter constitutional outcomes in ways that tend to reinforce an overrid- ing culture of regulation for its own sake. There would thus appear to be no valid basis for continuing to adhere to the technology-speciﬁc vision of the First Amendment inspired by the Broadcast Model. On the contrary, the analysis contained in this Article suggests that little would be lost and much would be gained from abandoning it.
https://openalex.org/W4389200318	https://link.springer.com/content/pdf/10.1007/s13177-023-00379-9.pdf	English	null	Design Problem of Travel Sections for Shared Autonomous Vehicles Considering the Immaturity of Autonomous Vehicle Technology	International journal of intelligent transportation systems research/International journal of ITS research	2,023	cc-by	11,090	International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 https://doi.org/10.1007/s13177-023-00379-9 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 https://doi.org/10.1007/s13177-023-00379-9 Abstract Most studies of shared autonomous vehicles (SAVs) have assumed established automated driving technology. In contrast, this study assumes that automated driving technology is not yet mature; it proposes two types of models to optimize the number of SAVs and the sections they travel. The first model (i.e., the ‘actual depot model’) does not consider the travel distance from the depot to the service start and end points. The second (i.e., the ‘virtual depot model’) considers the as-described travel distance. We showed in the actual depot model that under mild assumptions, the continuity condition of the SAV travel segments can be described by a simple constraint condition. The models were then applied to a hypothetical network; the performances of the models confirmed that the evaluation indicators for the virtual depot model tended to be better than the indicators for the actual depot model. Keywords Shared Autonomous Vehicles (SAVs) · Lane design problem · Mixed integer programming Design Problem of Travel Sections for Shared Autonomous Vehicles Considering the Immaturity of Autonomous Vehicle Technology Hiroshi Shimamoto1 Accepted: 8 November 2023 / Published online: 30 November 2023 © The Author(s) 2023 Accepted: 8 November 2023 / Published online: 30 November 2023 © The Author(s) 2023 * Hiroshi Shimamoto shimamoto@cc.miyazaki-u.ac.jp 1 Introduction fully automated driving technology has not yet been estab- lished, accidents may occur due to machine error; thus, a new legal liability framework is needed [2]. As discussed below, studies of SAVs generally assume mature automated driving technology. However, until the technology matures, implementation plans should consider the safety of SAVs through methods such as limiting the travel lanes of SAVs to sections where magnetic markers and sensors are installed. The fourth Industrial Revolution, centered around the Inter- net of Things and artificial intelligence (AI), is expected to bring substantial innovation to the transportation sec- tor. Ridesharing services such as Uber and Lyft are widely deployed overseas, and several municipalities in Japan are exploring the use of AI in on-demand shared-ride cabs. In the future, based on advancements in autonomous driv- ing technology, shared autonomous vehicles (SAVs) are expected to provide mobility services at lower costs. The SAV is a new concept, constructed on the basis of book- ing type, sharing system, and level of integration with other transportation modes [1, 2]. Because there is no driver, SAVs offer several advantages over traditional car-sharing vehicles driven by humans [3]: (i) SAVs are able to self- drive to travelers anywhere, (ii) SAVs can continuously work and seamlessly accomplish travelers’ requests, and (iii) SAVs can self-drive to parking locations after travel- ers have been dropped off at destinations. However, because Matching between users, an important element of SAVs, has been extensively studied. Individual users and vehicle behavior are matched using a space–time extended network (STEN) as the system optimal condition, as described in previous studies (e.g., Agatz et al. [4], Najmi et al. [5], Aiko et al. [6], Shimamoto [7]. However, because these problems are formulated as mixed-integer programming problems, in which vehicle and user movements are represented by binary variables, their solutions tend to be computationally expen- sive. Therefore, to reduce computational costs, some models treat users and vehicles as aggregated quantities and repre- sent them as real variables. For example, Shimamoto et al. [8] proposed an assignment model of three subjects, com- prising the solo driver, ridesharing driver, and rider under circumstances where a ridesharing service is available. * Hiroshi Shimamoto shimamoto@cc.miyazaki-u.ac.jp 1 Department of Civil and Environmental Engineering, University of Miyazaki, 1‑1, Gakuen Kibanadai Nishi, Miyazaki 889‑2192, Japan 1 Introduction Regarding studies of SAVs, Shimamoto [7] compared ridesharing systems with manually driven and autonomous vehicles via driver and rider matching using a STEN model; 1 Department of Civil and Environmental Engineering, University of Miyazaki, 1‑1, Gakuen Kibanadai Nishi, Miyazaki 889‑2192, Japan (0121 3456789) 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 82 in sections where magnetic markers and other devices are installed. Accordingly, we propose a strategy that optimizes the number of SAVs and the SAV travel sections, then evalu- ate the relationship between the level of SAV service and the social benefits of reducing the vehicle number (Fig. 1). We propose two models to achieve our goals: one model does not consider the travel distance from the depot to the service start and end points, whereas the second model considers this as-described travel distance. In the second model, the depot and the nodes providing the service must be connected by SAV travel sections to allow the SAVs to complete their travel. We show that the continuous condition of the SAV travel sections can be described by simple constraint condi- tions under mild assumptions. Finally, we apply the proposed model to a hypothetical network, which enables evaluation of the model characteristics. they confirmed that a ridesharing service involving autono- mous vehicles would equalize opportunities for user activity but increase total travel time. Li et al. [3] proposed a time- dependent SAV system design problem by jointly optimizing the fleet size, parking infrastructure deployment, and daily system operation for long-term infrastructure planning. Seo et al. [9] proposed an optimization model for an SAV system as a multi-objective optimization problem that simultane- ously describes traveler behavior and optimizes SAV equip- ment. Maruyama et al. [10] further modified the model of Seo et al. [9] to consider an integrated system of SAVs and bus rapid transit (BRT). In addition to describing traveler behavior and optimizing SAV equipment, their model can optimize BRT routes and schedules; however, the computa- tional cost of this model tends to be very high due to the large number of binary variables used to describe the BRT routes [9]. Note that these earlier studies implicitly assumed that SAV systems used mature automated technology. 1 Introduction The contributions of this study to the related literature are as follows: The optimal lane design problem has also been exten- sively studied in transportation planning on a variety of topics other than SAV lanes, such as transit priority provi- sion (e.g., use of a BRT lane) (Mesbah et al. [11]), wireless charging lanes for electric vehicles (Mubark et al. [12], Tran et al. [13]), and dedicated connected autonomous vehicle lanes (Lin et al. [14], Tani et al. [15]). However, the above- mentioned studies did not explicitly consider the continuous condition of the lane. In the present study, because SAVs are assumed to only travel on specified sections as described later, they cannot return to the depot on their own if the SAV travel sections are not contiguous. Thus, if the continu- ous condition of the SAV travel sections is not considered, many infeasible solutions may be generated, resulting in an increase in computation time. Furthermore, studies regard- ing dedicated lanes for SAVs [14–17] tend to focus on the effect of increased capacity due to shorter distances between vehicles, and/or on changes in driving behavior (e.g., accel- eration and lane changing); they do not focus on lane limita- tions due to the immaturity of automated driving technology. • We formulate the lane continuity condition as a simple constraint to reduce the computational cost for lane deter- mination. • We propose a model describing user choice between pri- vate cars and SAVs, then numerically clarify the relation- ship between private car ownership cost and the SAV penetration rate. The remainder of this paper is organized as follows. Section 2 describes the assumptions, notations, and formu- lations of the proposed models. Section 3 then provides a numerical example. Finally, Section 4 draws conclusions and establishes directions for future research. Fig. 1 Concept of this research 2.1.1 Outline of the Model As discussed thus far, the introduction of SAVs has two effects: an increase in link capacity and a decrease in the number of vehicles due to vehicle sharing. Unlike previous studies, the present study assumes a situation in which auto- mated driving technology is immature and SAVs travel only This study assumes a society in which automated vehicles can only travel in designated sections due to the immatu- rity of automated driving technology. We constructed a Fig. 1 Concept of this research Reduction in the number of vehicles through seat sharing Reduction of road congestion and total travel cost Necessity of magnetic marker installation for the safe operation of SAVs Determination of appropriate number of SAVs Necessity of designation of SAV travel section Budget restrictions for the decision maker Introduction of SAVs Effect Issue Social benefit SAV Level of Service Increase in capacity due to 1. shorter distances b/w vehicles 2. change in drivers behavior Introduction of SAVs Increase in capacity due to 1. shorter distances b/w vehicles 2. change in drivers behavior Budget restrictions for the decision maker Social benefit 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 83 model to determine the sections where SAVs travel and the number of SAVs that satisfy the system optimal condi- tions, where all travelers travel either in manually driven vehicles or SAVs. Hereinafter, we refer to a human-driven vehicle (HDV) and its driver as the ‘HDV driver’; the per- son traveling in an SAV as the ‘SAV passenger’; and both ‘HDV driver’ and ‘SAV passenger’ as ‘travelers’. Note that, as mentioned above, several studies regard- ing autonomous vehicles (AVs) have considered changes in capacity with changes in AV penetration due to differ- ences in driving behavior between HDVs and AVs. How- ever, we included the second assumption above to focus on the effect of SAV introduction on reducing the number of vehicles on the road. p g Because the SAVs must travel from the depot to the ser- vice start point and from the service end point to the depot, the total travel distance varies according to the locations of the depot and the service start and end points. Therefore, we constructed two models: the ‘virtual depot model’, which does not consider the travel distance from the depot to the service start and end points, and the ‘actual depot model’, which does consider the as-described travel dis- tance. 2.1.2 Assumptions of the Model Additionally, the decision-maker may designate SAV travel sections to minimize the total system cost, includ- ing total travel time. Nevertheless, we included the fourth assumption above because it is difficult to appropriately set the unit cost of designating SAV travel sections. First, we describe assumptions common to the virtual and actual depot models. The following assumptions are made regarding the flow of travelers and SAVs. • Travelers either drive an HDV or use SAVs; HDV driv- ers travel alone. In addition to the assumptions described above, the fol- lowing assumptions are made in the actual depot model: • HDV drivers and SAV passengers are not allowed to swap during the journey. • Only one SAV depot is assumed, and its location is given exogenously. • All traveler demands (e.g., departure and arrival loca- tions, and departure and arrival time constraints) are known. The departure and arrival time constraints are established in a manner that requires travelers to depart earlier than the departure time constraint and arrive before the arrival time constraint. • Both directions between adjacent nodes are simultaneously designated as SAV travel sections (unless they are one-way). Note that the second assumption above is included to express in a simple equation the constraint that SAVs can only travel the specified sections and reach the depot after providing service. The formulation of the constraint is described in Section 2.5. • The total number of HDVs and SAVs cannot exceed link capacity. Note that the third assumption above, which corre- sponds to the absence of considering the early arrival pen- alty, is included to avoid an unrealistic solution of making a detour to satisfy the arrival time constraint. 2.1.1 Outline of the Model As indicated below, the actual depot model requires the location of the depot as a component of the input data. By comparing the solutions of the two models, it is pos- sible to evaluate the effect of specifying depot location on model performance efficiency. The following assumptions are made regarding the costs to HDV drivers, SAV passengers, and SAV vehicles. • The decision-maker, an SAV operator who desires to reduce congestion, determines the number of SAVs to be deployed and the sections where SAVs can travel. • The cost of designating the SAV travel sections is pro- portional to the link length. • The decision-maker designates the SAV travel sections within a budget constraint. • HDVs can travel all sections, including the designated SAV travel sections. • HDVs can travel all sections, including the designated SAV travel sections. Note that the third assumption above considers that the cost of installing a device, such as magnetic markers, is proportional to the length of the SAV travel section/s. Note that the third assumption above considers that the cost of installing a device, such as magnetic markers, is proportional to the length of the SAV travel section/s. Additionally, the decision-maker may designate SAV travel sections to minimize the total system cost, includ- ing total travel time. Nevertheless, we included the fourth assumption above because it is difficult to appropriately set the unit cost of designating SAV travel sections. • SAVs can travel only on designated sections. 2.2 Space–time Extended Network (STEN) To represent the flows of travelers and SAVs, an ‘original network’ (i.e., a graphical representation of an actual road network), shown in Fig. 2(a), is converted into two types of networks: a ‘travelers’ network’ and an ‘SAV network,’ as shown in Fig. 2(b). Note that HDV drivers’ and ‘SAV passen- gers’ cannot be interchanged during a journey, as described in the previous section. Thus, the ‘travelers’ network’ is further categorized into two network types: an ‘HDV driver Next, the following assumptions are made regarding SAV behavior. • SAVs can travel only on designated sections. • The driving behavior (e.g., acceleration, lane changing, and distance between vehicles) of SAVs is identical to the behavior of HDVs. 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 84 84 Interna (a) Example of an ‘original network’ (b) Extended network (c) The Space-Time time Extended Network (STEN) A B C A B A B C HDV driver network SAV network A B C C SAV passenger network Travelers' network Depot node A B C A B C A B C Depot node HDV driver network SAV passenger network SAV network Depot node In-vehicle link Waiting link Access link Egress link Time Period Time Period Time Period network’ and an ‘SAV passenger network’. The centroid in the ‘SAV network’ is represented as a dummy node (i.e., an SAV depot node). The virtual depot model creates access and egress links from SAV depot nodes to all nodes in the ‘origi- nal network’, whereas the actual depot model creates access and egress links from SAV depot nodes only to nodes with depots in the ‘original network’. The network in Fig. 2(b) is then extended to the STEN, [18] as shown in Fig. 2(c).i (a) Example of an ‘original network’ (b) Extended network A B A B C HDV driver network SAV network A B C C SAV passenger network Travelers' network Depot node A node in the STEN is defined as a node in the two- dimensional network for each time period and is denoted as (i, t) , where i is the node number in the two-dimensional network and t is the time period of the node. This study defines three node types: an ‘origin node’, a ‘destination node’ where demand is generated and concentrated, and an ‘intermediate node’ corresponding to all other nodes. 2.2 Space–time Extended Network (STEN) Three types of links are also defined: an ‘in-vehicle link’, an ‘access/egress link,’ and a ‘waiting link’. An ‘in-vehicle link’ represents travel by vehicles between two nodes. When both tail and head nodes of an in-vehicle link are expressed as (i, t1 ) and (j, t2 ) , respectively, t2 = t1 + cij , where cij is the travel time between i and j . Note that the interaction between the travelers' network and the SAV network is described by the relationship between flows in the HDV driver network and SAV passenger network and corresponding flows in the SAV network. A ‘waiting link’ connects two adjacent nodes (i, t) and (i, t + Δt) , representing travelers or SAVs waiting at node i from t to t + Δt . Finally, an ‘access/egress link’ represents access from or egress to an origin or depot node. Depot node (b) Extended network (b) Extended network (c) The Space-Time time Extended Network (STEN) A B C A B C A B C Depot node HDV driver network SAV passenger network SAV network Depot node In-vehicle link Waiting link Access link Egress link Time Period Time Period Time Period A B C A B C HDV driver network SAV passenger network Depot node Time Period Time Period 2.4.1 In‑vehicle Links Based on the above assumptions, the cost of an in-vehicle varies among HDV drivers, SAV passengers, and SAVs, as given by (1). ⎧ (1) ca = ⎧ ⎪ ⎨ ⎪⎩ 훼1ta + 훽TCta ∀a ∈AD 1 훼1ta ∀a ∈AP 1 훽TCta ∀a ∈ASAV 1 (1) aX 2,nt Waiting link of type X in the STEN that leads out of node n at time step t 2.3.1 Notations Related to the STEN D Set of drivers P Set of SAV passengers SAV Set of SAVs T Set of travelers (D ∪P) X Type of STEN, consisting of ‘travelers’ network’ (T) or ‘SAV network’ (SAV) Δt Duration of time step NT Set of time steps AX 1 Set of in-vehicle links of type X AX 2 Set of waiting links of type X SAV passenger network SAV passenger network (c) The Space-Time time Extended Network (STEN) A B C Depot node SAV network Depot node In-vehicle link Waiting link Access link Egress link Time Period Depot node (c) The Space-Time time Extended Network (STEN) Fig. 2 Network transformation 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 85 International Journal of Intelligent Transportation Systems Research (2024) AX 3 Set of access links of type X AX 4 Set of egress links of type X A Set of all links IX Set of nodes of type X I Set of all nodes R Set of origin nodes S Set of destination nodes ia Tail node of link a ∈A ja Header node of link a ∈A OUT(i) Set of links that lead out of node i ∈I IN(i) Set of links that lead into node i ∈I qrs Travel demand between r and s d1, d2 Nodes representing SAV origins and destinations, respectively 2.3.2 Notations Related to the Correspondence Between the STEN and the ‘Original Network’ L Set of links in the original network N Set of nodes in the original network n(i) Identity (ID) of node i ∈I in the original network aX 1,lt In-vehicle link of type X in the STEN that leads out link l at time step t aX 2,nt Waiting link of type X in the STEN that leads out of node n at time step t 2.3.3 Notations Related to Decision Variables xs a Passenger flow of link a with destination s ya SAV flow of a ∈A CPn Capacity of a waiting link on node n in the original network 2.4.3 Access/Egress Links (14) The cost of owning an HDV (vehicle price and maintenance cost per day) is included in the access links of the HDV driver, as described in Eq. (4). The costs for egress links are all 0, as shown in Eq. (5). (15) ∑ s∈S xs aP 1,lt ≤휅yaSAV 1,lt , ∀l ∈L, t ∈NT (15) (4) ca = { 휃HDV (∀a ∈AD 3 ) 0 (∀a ∈AP 3 ∪ASAV 3 ) (5) ca = 0(∀a ∈AD 4 ∪AP 4 ∪ASAV 4 ) (4) ca = { 휃HDV (∀a ∈AD 3 ) 0 (∀a ∈AP 3 ∪ASAV 3 ) (5) ca = 0(∀a ∈AD 4 ∪AP 4 ∪ASAV 4 ) (16) yaSAV 1,lt ≤CRl ⋅휙t, ∀l ∈L, t ∈NT (17) ∑ l∈L tl휙l ≤BGT (4) AD 3 ) ∪ASAV 3 ) (16) yaSAV 1,lt ≤CRl ⋅휙t, ∀l ∈L, t ∈NT (4) (16) (5) (17) ∑ l∈L tl휙l ≤BGT (5) ca = 0(∀a ∈AD 4 ∪AP 4 ∪ASAV 4 ) ∑ l∈L tl휙l ≤BGT (5) ∑ l∈L tl휙l ≤BGT (17) Note that the cost of owning an SAV is not included in the access links because the number of SAVs, a decision variable, is explicitly included in the objective function, as shown in the next section. xs a ≥0, ∀a ∈AT, s ∈S (18) ya ≥0, ∀a ∈ASAV ya ≥0, ∀a ∈ASAV 2.5 Formulation z ≥0 (20) 2.4.2 Waiting Links Because the early arrival penalty is not considered, the cost of a waiting link is given as follows: xs a Passenger flow of link a with destination s (2) cs a = { ws a (∀a ∈AD 2 ∪AP 2 ) 0 (∀a ∈ASAV 2 ) (2) where 1 3 1 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 86 (3) (11) ∑ a∈OUT(d2) ya − ∑ a∈IN(d2) ya = −z ink in the travelers’ network the cost in the SAV network fore, in Section 2.5, the cost ed as cs a for each destination, ork is described as ca. icle price and maintenance ess links of the HDV driver, for egress links are all 0, as n SAV is not included in the f SAVs, a decision variable, ve function, as shown in the determines the number of (3) (4) (5) (11) ∑ a∈OUT(d2) ya − ∑ a∈IN(d2) ya = −z (12) ∑ a∈OUT(i) ya − ∑ a∈IN(i) ya = 0, ∀i ∈{ISAV −d1 −d2} (13) ∑ s∈S xs aD 1,lt + yaSAV 1,lt ≤CRl, ∀l ∈L, t ∈NT (14) ∑ s∈S xs aD 2,nt + yaSAV 2,nt ≤CPn, ∀n ∈N, t ∈NT (15) ∑ s∈S xs aP 1,lt ≤휅yaSAV 1,lt , ∀l ∈L, t ∈NT (16) yaSAV 1,lt ≤CRl ⋅휙t, ∀l ∈L, t ∈NT (17) ∑ l∈L tl휙l ≤BGT (18) xs a ≥0, ∀a ∈AT, s ∈S (19) ya ≥0, ∀a ∈ASAV (20) z ≥0 (21) 휙l = {0, 1}, ∀l ∈L (3) ws a = { 훼2Δt (n(s) ≠n(ja )) 0 (n(s) = n(ja )) ws a = { 훼2Δt (n(s) ≠n(ja )) 0 (n(s) = n(ja )) (11) (3) Note that the cost of a waiting link in the travelers’ network is destination-dependent, whereas the cost in the SAV network is destination-independent. Therefore, in Section 2.5, the cost in the travelers’ network is described as cs a for each destination, whereas the cost in the SAV network is described as ca. (12) ∑ a∈OUT(i) ya − ∑ a∈IN(i) ya = 0, ∀i ∈{ISAV −d1 −d2} (12) (13) ∑ s∈S xs aD 1,lt + yaSAV 1,lt ≤CRl, ∀l ∈L, t ∈NT (13) 2.5.1 Virtual Depot Model 휙l = {0, 1}, ∀l ∈L 휙l = {0, 1}, ∀l ∈L (21) The virtual depot model, which determines the number of SAVs and the SAV travel sections without considering the travel distance from the depot to the service start and end points, can be formulated as a mixed-integer linear program- ming problem that minimizes the total cost: The first, second, and third terms in Eq. (6) represent the sum of the total travel time and total travel cost of travelers, the total travel cost of the SAVs, and the cost of introducing SAVs, respectively. Constraints (7)–(9) and (10) − (12) represent the flow conservation laws for travelers and SAVs on the STEN, respectively. Constraints (13) − (15) represent capac- ity constraints. Constraint (13) states that the number of HDVs and SAVs in the in-vehicle link must be less than or equal to the link capacity, which describes the interac- tion between HDV and SAVs. Constraint (14) states that the number of HDVs and SAVs at a node must be less than or equal to the node capacity. Constraint (15) states that the flow on an in-vehicle link of the SAV passenger network is less than or equal to the total SAV capacity, which describes the interaction between SAV passengers and SAVs. Constraint (16) states that SAVs can only travel in designated sections; SAVs cannot travel if a link is not designated as an SAV travel section, and otherwise, the maximum number of SAVs can be the link capacity. Note that HDVs are allowed to travel on all sections because (6) min 퐱,퐲,z,훟Z = ∑ s∈S ∑ a∈AT cs axs a + ∑ a∈ASAV caya + 휃SAVz (6) Subject to: ∑ a∈OUT(r) xs a − ∑ a∈IN(r) xs a = qrs, ∀r ∈R, s ∈S (7) (8) ∑ a∈OUT(s) xs a − ∑ a∈IN(s) xs a = − ∑ r∈R qrs, ∀s ∈S ∑ a∈OUT(s) xs a − ∑ a∈IN(s) xs a = − ∑ r∈R qrs, ∀s ∈S (8) (9) ∑ a∈OUT(i) xs a − ∑ a∈IN(i) xs a = 0, ∀i ∈{IT −R −S} (9) (10) ∑ a∈OUT(d1) ya − ∑ a∈IN(d1) ya = z (10) 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 87 no constraints are imposed on the HDV travel sections. 2.5.1 Virtual Depot Model Constraint (17) represents the budget constraint whereby the designated cost of the SAV travel sections must be less than or equal to the limited budget. Constraints (18)–(20) represent the non-negative conditions for traveler flow, SAV flow, and number of SAVs, respectively. Finally, con- straint (21) states that the design variables for the SAV travel sections are binary variables. where ndepo is the depot node in the original network and Δl m is a binary variable with a value of 1 if a pair of nodes on links l and m are identical. Constraint (22) states that for a forward link to be desig- nated as the travel section, at least one of the forward links directly upstream of that link must be designated as an SAV travel section, representing the continuous condition of SAV travel sections, starting at the depot node (Fig. 3). Constraint (23) represents the requirement for the design variables to be binary. Constraint (24) converts the presence or absence of travel sections on "forward links" into the presence or absence of travel sections on links in the original network. 2.5.2 Actual Depot Model For SAVs to complete their travel, the depot and the nodes providing service must be connected by the SAV travel sec- tions. However, because the virtual depot model does not impose such constraints, it may generate a large number of infeasible solutions, reducing computational efficiency. Here, we discuss how to formulate the continuous condition of the SAV travel sections. Therefore, the actual depot model can be formulated as a problem in which the continuous condition of the SAV travel sections is added to the constraints of the virtual depot model, as follows: (25) min 퐱,퐲,z,훟Z = ∑ s∈S ∑ a∈AT cs axs a + ∑ a∈ASAV caya + 휃SAVz (25) In the original network, there are two links between adjacent nodes in different directions, with the exception of one-way streets. However, as assumed in Section 2.1.2, both directions between adjacent nodes are simultane- ously designated as SAV travel sections; thus, there is no requirement for distinguishing the direction of links between the nodes. The following procedure is defined to create a set of the forward links. Note that the forward and backward links are defined as links in a direction away from or closer to the depot node, respectively. subject to Eqs. (7)–(20) and Eqs. (22)–(24). Note that SAVs may not reach the depot from some SAV travel sections that satisfy the continuous condition if there is a one-way street in the original network. How- ever, the impact of the one-way streets is expected to be limited because the original network generally has few one-way sections. Step 1: Compute the minimum length The two proposed models were applied to a hypotheti- cal network to characterize the models. We solved the mixed-integer linear programming problem using MAT- LAB 2022 (MathWorks, Natick, MA, USA) and Gurobi 9.5.0. (Gurobi Optimization LLC, Beaverton, OR, USA). Section 3.1 describes the study network and computa- tional conditions. Section 3.2 compares the solutions of the virtual depot model and the actual depot model when the HDV ownership cost is not considered. Finally, Section 3.3 presents a sensitivity analysis of the HDV ownership cost. Compute the minimum length from the depot node to all other nodes, where 휛i is defined as the minimum length from the depot node to node i. 3.1 Test Network and Computational Conditions and maintenance costs) was set to 휃SAV = 5000(JPY∕day) , based on the daily cost ( 4406.4JPY∕yen ) estimated using values shown in Table 2(a). The HDV cost was set to two ways: 휃HDV = 1500(JPY∕day) and 휃HDV = 2500(JPY∕day) , referring to the estimated maintenance cost ( 1666.7JPY∕yen ) and the purchase and maintenance cost ( 2488.5JPY∕yen ) of HDVs, which were estimated using values shown in Table 2(b). Note that 휃HDV = 1500 corresponds to the case where only the maintenance cost of an HDV is considered, while 휃HDV = 2500 corresponds to the case where both pur- chase and maintenance costs are considered. Step 2: Generate a set of forward links 5 Setting departure and arrival time constraints 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 Fig. 4 The Sioux Falls network Table 2 Parameters for SAVs and HDVs SAV, Shared autonomous vehicle; HDV, Human-driven vehicle (a) Parameters for SAVs Vehicle Price Durable life Maintenance cost 500 million JPY 5 years 50,000 JPY/month (b) Parameters for HDVs Vehicle Price Durable life Maintenance cost 300 million JPY 10 years 50,000 JPY/month 56 Departure time limit Arrival time limit Time window Deviation in departure time Base time period Shortest travel time between origin and destination Fig. 5 Setting departure and arrival time constraints Shortest travel time between origin and destination Deviation in departure time Time window Departure time limit Base time period Departu Base time period Arrival time limit Departure time limit Fig. 4 The Sioux Falls network Fig. 5 Setting departure and arrival time constraints 3.1.1 Test Network We used the Sioux Falls network [19], as shown in Fig. 4. The travel time is given by [da∕2] × Δt , where da is the link distance (km) and [x] is the largest integer smaller than or equal to x . We assumed that Δt = 15 and the link capacities were set to half of the original values. Additionally, the loca- tion of the depot in the actual depot model was specified, such that two cases were compared: a case where the depot is located in a suburban area (Node 1) and a case where the depot is located in a central area (Node 10). Step 2: Generate a set of forward links 3 Continuous condition of the SAV travel section 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 88 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 Fig. 4 The Sioux Falls network Table 1 Parameters for setting the travel cost per distance Duration of time step Fuel cost Fuel consumption Average speed 15 min 140 JPY/l 12km/l 30km/h Table 2 Parameters for SAVs and HDVs SAV, Shared autonomous vehicle; HDV, Human-driven vehicle (a) Parameters for SAVs Vehicle Price Durable life Maintenance cost 500 million JPY 5 years 50,000 JPY/month (b) Parameters for HDVs Vehicle Price Durable life Maintenance cost 300 million JPY 10 years 50,000 JPY/month Departure time limit Arrival time limit Time window Deviation in departure time Base time period Shortest travel time between origin and destination Fig. Step 2: Generate a set of forward links Remove backward links from the original network and generate a set of forward links L′ . Note that we define ‘for- ward links’ as links such that 휛i(l) ≤휛j(l), ∀l ∈L , and ‘back- ward links’ as links such that 𝜛i(l) > 𝜛j(l), ∀l ∈L. ( ) j( ) The continuous condition of the SAV travel sections is represented by the following simple constraints: ʞ ʞ ʞ For a forward link to be designated as a travel section, at least one of the forward links directly upstream of that link must be designated as the SAV travel section. Depot node represented by the following simple constraints: (22) 휑m ≤ ∑ n∈IN(im) 휑n, ∀m ∈L, im ≠ndepo (23) 휑m = {0, 1}, ∀m ∈L (24) 휙l = Δl m휑m, ∀l ∈L, m ∈L ʞ ʞ ʞ For a forward link to be designated as a travel section, at least one of the forward links directly upstream of that link must be designated as the SAV travel section. Depot node Fig. 3 Continuous condition of the SAV travel section (22) 휑m ≤ ∑ n∈IN(im) 휑n, ∀m ∈L, im ≠ndepo (23) 휑m = {0, 1}, ∀m ∈L (24) 휙l = Δl m휑m, ∀l ∈L, m ∈L ʞ ʞ ʞ For a forward link to be designated as a travel section, at least one of the forward links directly upstream of that link must be designated as the SAV travel section. Depot node Fig. 3 Continuous condition of the SAV travel section (22) 휑m ≤ ∑ n∈IN(im) 휑n, ∀m ∈L, im ≠ndepo (22) For a forward link to be designated as a travel section, at least one of the forward links directly upstream of that link must be designated as the SAV travel section. 휑m = {0, 1}, ∀m ∈L 휑m = {0, 1}, ∀m ∈L (23) 휙l = Δl m휑m, ∀l ∈L, m ∈L (24) 휙l = Δl m휑m, ∀l ∈L, m ∈L Fig. 3 Continuous condition of the SAV travel section Fig. 3.1.4 Valuation Indicators the SAV passenger ratio and the degree of improvement in the objective function, because this increase led to an expanded area in which SAVs can travel. This study utilized three valuation indicators: the number of SAVs, the SAV passenger ratio, and the degree of improve- ment in the objective function value. The number of SAVs and the SAV passenger ratio are indicators of the extent of SAV penetration, whereas the degree of improvement in the objective function value is an indicator used to evaluate the efficiency of SAV introduction. Note that the SAV passenger ratio and degree of improvement in the objective function value are defined as follows: In the virtual depot model, when the depot was located at Node 1, all indicator values were 0, indicating that the solution without SAV introduction and SAV travel sections was optimal. In contrast, when the depot was located at Node 10, the solution in which SAVs are introduced was optimal, but the number of SAVs stabilized after the budget limit reached 20. All indicator values in the virtual depot model were significantly lower than indicator values in the actual depot model. This difference is possibly because the virtual depot model does not consider the cost of travel from the depot to the service start points; it may also be related to the wider range of feasible solutions for the SAV travel section. (26) rSAV = ∑ s∈S ∑ a∈AD 1 taxs a ∑ s∈S ∑ a∈AD 1 ∪AP 1 taxs a (26) (27) IR = 1 −ZOPT ZHDV Figures 7(a) and (b) and 8(a) and (b) show the SAV travel sections for the virtual depot model and the actual depot model (assuming the depot is located at Node 10) when the budget limit was 10 and 50, respectively. When the budget limit was 10, the general locations of the SAV travel sections were similar for both models. However, in the actual depot model, the SAVs can return to the depot only by traveling on the SAV travel sections; in the virtual depot model, the SAVs could not return to the depot only by traveling on the SAV travel sections. When the budget limit was 50, the SAV travel sections were more extensive in the virtual depot model than in the actual depot model. 3.1.4 Valuation Indicators However, as in the case when the budget limit was 10, the SAVs could not return to the depot only by traveling on SAV travel sections in the virtual depot model. (27) where ZOPT and ZHDV represent objective function values with and without consideration of SAVs, respectively. 3.2 A Case without Considering the Ownership Cost of Human‑driven Vehicles (HDVs) This section compares the solutions between the two models when the cost of HDV ownership, 휃HDV , is 0. Figure 6(a)–(c) show the relationships among the num- ber of SAVs, the SAV passenger ratio, and the degree of improvement in the objective function value and the budget constraint specifying the HDV travel sections, respectively. In the actual depot model, the number of SAVs increased with the budget constraint but stabilized after the budget limit reached 70. The SAV passenger ratio and the degree of improvement in the objective function value increased up to the budget limit of 90. Increasing the budget limit above 70 did not increase the number of SAVs; however, increasing the budget limit to 90 increased 3.1.2 Parameter Settings Because only the static OD demand is published as benchmark data in the Sioux Falls network, we consider the conversion of static demand into dynamic demand. As shown in Fig. 5, we assume that there is a specific variation between the base- line time period and the departure time limit. The arrival time limit is then set to ‘departure time limit + minimum travel time between ODs + time window’. In the following section, vari- ation in the departure time limit is represented by a Poisson distribution with a mean value 휆= 1 , and the time window is uniformly set to 4 × Δt (1 h (h)) for all OD pairs. The parameters of the proposed models are the value of time ( 훼1 and 훼2 ), the capacity of SAVs ( 휅 ), the running cost of the vehicle ( 훽TC ), and the ownership cost of SAVs and HDVs ( 휃SAV , 휃HDV ). The value of times is assumed to be 훼1 = 33.33(JPY∕min) and 훼2 = 66.6(JPY∕min) ; the SAV capacity is assumed to be 휅= 3 . The travel cost per distance ( 훽TC ) was calculated by converting the vehicle running cost into hourly units using the values shown in Table 1, resulting in 훽TC = 2.92(JPY∕min) . The SAV ownership cost (purchase 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 89 3.3 Sensitivity Analysis of the Cost of HDV Ownership a VD model, b AD model ) b ( AD model 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 (a) le d o m D V Fig. 7 Illustration of SAV travel section (when the budget limit is 10). a VD model, b AD model ig. 7 Illustration of SAV travel section (when the budget limit is 10). 3.3 Sensitivity Analysis of the Cost of HDV Ownership a VD model, b AD model (a) VD model (b) AD model 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 Fig. 8 Illustration of SAV travel section (when the budget limit is 50). 3.3 Sensitivity Analysis of the Cost of HDV Ownership a VD model, b AD model (a) ) b ( le d o m D V AD model 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 Fig. 7 Illustration of SAV travel section (when the budget limit is 10). 3.3 Sensitivity Analysis of the Cost of HDV Ownership This section analyzes the impact of the cost of HDV own- ership. As stated previously, two cases of the cost of HDV ownership are assumed: a case in which only maintenance (a) Number of SAVs (b) SAV passenger ratio (c) Degree of improvement in the objective function value 0 2000 4000 6000 8000 10000 12000 10 20 30 40 50 60 70 80 90 100 Number of SAVs Budget limit for spesifying SAV travel secon VD AD(Depo:1) AD(Depo:10) 0 2 4 6 8 10 12 14 16 18 20 10 20 30 40 50 60 70 80 90 100 SAV passenger ratio(%) Budget limit for spesifying SAV travel section VD AD(Depo:1) AD(Depo:10) 0 1 2 3 10 20 30 40 50 60 70 80 90 100 Degree of improvement of obj. func. value(%) Budget limit for spesifying SAV travel secon VD AD(Depo:1) AD(Depo:10) Fig. 6 Relationships between valuation indicators and the budget constraint specifying the HDV travel section. VD: virtual depot model; AD: actual depot model. a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value 0 2000 4000 6000 8000 10000 12000 10 20 30 40 50 60 70 80 90 100 Number of SAVs Budget limit for spesifying SAV travel secon VD AD(Depo:1) AD(Depo:10) (c) Degree of improvement in the objective function value 0 1 2 3 10 20 30 40 50 60 70 80 90 100 Degree of improvement of obj. func. value(%) Budget limit for spesifying SAV travel secon VD AD(Depo:1) AD(Depo:10) (b) SAV passenger ratio 0 2 4 6 8 10 12 14 16 18 20 10 20 30 40 50 60 70 80 90 100 SAV passenger ratio(%) Budget limit for spesifying SAV travel section VD AD(Depo:1) AD(Depo:10) (c) Degree of improvement in the objective function value (c) Degree of improvement in the objective function value (b) SAV passenger ratio (a) Number of SAVs Fig. 6 Relationships between valuation indicators and the budget constraint specifying the HDV travel section. VD: virtual depot model; AD: actual depot model. 3.3 Sensitivity Analysis of the Cost of HDV Ownership a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 90 90 International Journal of Intelligent Transportation Systems Research (2024) 22:81 (a) ) b ( le d o m D V AD model 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 Fig. 7 Illustration of SAV travel section (when the budget limit is 10). 3.3 Sensitivity Analysis of the Cost of HDV Ownership costs are considered ( 휃HDV = 1500 ) and a case in which both purchase and maintenance costs are considered ( 휃HDV = 2500 ). Note that the computational time for the actual depot model was ~ 2 min (min), whereas the compu- tational time for the virtual depot model with 휃HDV = 1500 and a budget limit of 50 was > 72 h, using a computer with an Intel Xeon W-2102 central processing unit, with 2.90 GHz and 32.0 GB RAM. Therefore, only the actual depot model is considered in this section. The significantly higher computational cost of the virtual depot model, com- pared with the actual depot model, may be attributed to the simple constraint limitation of the feasible solution region of SAV travel sections in the actual depot model (Eq. (22)). Figure 10(a)–(c) show the relationships among the num- ber of SAVs, the SAV passenger ratio, and the degree of improvement in the objective function value with the budget constraint specifying the HDV travel sections, respectively, when the depot is located at Node 10. The number of SAVs was always larger when the cost of ownership was 2500 than when the cost of ownership was 1500, and the differ- ence between the two was larger than when the depot was located at Node 1. The degree of improvement in the objec- tive function value also tended to increase with increasing cost of ownership. Figure 9(a)–(c) show the relationships among the num- ber of SAVs, the SAV passenger ratio, and the degree of improvement in the objective function value with the budget constraint specifying the HDV travel sections, respectively, when the depot is located at Node 1. As stated previously, the solution that did not introduce SAVs and SAV travel sections was optimal when the cost of HDV ownership was (a) Number of SAVs (b) SAV passenger ratio (c) Degree of improvement in the objective function value 0 500 1000 1500 2000 2500 10 20 30 40 50 60 70 80 90 100 Number of SAVs Budget limit for spesifying SAV travel secon 0 1500 2500 0 2 4 6 10 20 30 40 50 60 70 80 90 100 SAV passenger rao(%) Budget limit for spesifying SAV travel secon 0 1500 2500 0 0.2 0.4 0.6 0.8 1 10 20 30 40 50 60 70 80 90 100 Degree of improvement of obj. func. 3.3 Sensitivity Analysis of the Cost of HDV Ownership When 휃HDV was 1,500 and 2,500, there was no significant difference in the number of SAVs and the SAV passenger ratio for the same budget limit; however, both indicators were always larger when 휃HDV = 2500 . In contrast, there was a significant difference in the degree of improvement in the objective function value when the cost of ownership was 1500 and 2000. This is presumably because a greater cost of HDV ownership led to a greater effect of reduction in the number of HDVs associated with SAV introduction. costs are considered ( 휃HDV = 1500 ) and a case in which both purchase and maintenance costs are considered ( 휃HDV = 2500 ). Note that the computational time for the actual depot model was ~ 2 min (min), whereas the compu- tational time for the virtual depot model with 휃HDV = 1500 and a budget limit of 50 was > 72 h, using a computer with an Intel Xeon W-2102 central processing unit, with 2.90 GHz and 32.0 GB RAM. Therefore, only the actual depot model is considered in this section. The significantly higher computational cost of the virtual depot model, com- pared with the actual depot model, may be attributed to the simple constraint limitation of the feasible solution region of SAV travel sections in the actual depot model (Eq. (22)). Figure 9(a)–(c) show the relationships among the num- ber of SAVs, the SAV passenger ratio, and the degree of improvement in the objective function value with the budget constraint specifying the HDV travel sections, respectively, when the depot is located at Node 1. As stated previously, the solution that did not introduce SAVs and SAV travel sections was optimal when the cost of HDV ownership was 0. When 휃HDV was 1,500 and 2,500, there was no significant difference in the number of SAVs and the SAV passenger ratio for the same budget limit; however, both indicators were always larger when 휃HDV = 2500 . In contrast, there was a significant difference in the degree of improvement in the objective function value when the cost of ownership was 1500 and 2000. This is presumably because a greater cost of HDV ownership led to a greater effect of reduction in the number of HDVs associated with SAV introduction. 3.3 Sensitivity Analysis of the Cost of HDV Ownership a VD model, b AD model (a) VD model (b) AD model 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 Fig. 8 Illustration of SAV travel section (when the budget limit is 50). a VD model, b AD model (b) AD model 1 8 4 5 6 3 2 15 19 17 18 7 12 11 10 16 9 20 23 22 14 13 24 21 3 1 2 6 8 9 11 5 15 12 23 13 21 16 19 17 20 18 54 55 50 48 29 51 49 52 58 24 27 32 33 36 7 35 40 34 41 44 57 45 72 70 46 67 69 65 25 28 43 53 59 61 56 60 66 62 68 63 76 73 30 71 42 64 75 39 74 37 38 26 4 14 22 47 10 31 (b) AD model (a) Fig. 8 Illustration of SAV travel section (when the budget limit is 50). a VD model, b AD model 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 91 0. 3.3 Sensitivity Analysis of the Cost of HDV Ownership value(%) Budget limit for spesifying SAV travel secon 0 1500 2500 Fig. 9 The relationship between the valuation indicators and the budget constraint specifying the HDV travel section when the depot is assumed to be Node 1. a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value (c) Degree of improvement in the objective function value 0 0.2 0.4 0.6 0.8 1 10 20 30 40 50 60 70 80 90 100 Degree of improvement of obj. func. value(%) Budget limit for spesifying SAV travel secon 0 1500 2500 0 500 1000 1500 2000 2500 10 20 30 40 50 60 70 80 90 100 Number of SAVs Budget limit for spesifying SAV travel secon 0 1500 2500 (b) SAV passenger ratio 0 2 4 6 10 20 30 40 50 60 70 80 90 100 SAV passenger rao(%) Budget limit for spesifying SAV travel secon 0 1500 2500 (c) Degree of improvement in the objective function value (b) SAV passenger ratio (a) Number of SAVs (a) (b) Fig. 9 The relationship between the valuation indicators and the budget constraint specifying the HDV travel section when the depot is assumed to be Node 1. a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value (a) Number of SAVs (b) SAV passenger ratio (c) Degree of improvement in the objective function value 0 2000 4000 6000 8000 10000 12000 14000 10 20 30 40 50 60 70 80 90 100 Number of SAVs Budget limit for spesifying SAV travel secon 0 1500 2500 0 2 4 6 8 10 12 14 16 18 20 10 20 30 40 50 60 70 80 90 100 SAV passenger rao(%) Budget limit for spesifying SAV travel secon 0 1500 2500 0 2 4 6 8 10 10 20 30 40 50 60 70 80 90 100 Degree of improvement of obj. func. value(%) Budget limit for spesifying SAV travel secon 0 1500 2500 Fig. 10 The relationship between the valuation indicators and the budget constraint specifying the HDV travel section when the depot is assumed to be Node 10. 4 Conclusion Acknowledgements This research was partially supported by JSPS KAKENHI Grant Number 21H01459, Japan and Committee on Advanced Road Technology, Ministry of Land, Infrastructure, Trans- port and Tourism, Japan (Project leader: Prof. Shoshi Mizokami, Kumamoto Gakuen University). This study proposed a model for determining the optimal numbers of SAVs and SAV travel sections, assuming a situation in which travelers either drive an HDV or use SAVs. Two types of models were formulated as mixed- integer programming problems: a virtual depot model that does not consider the travel distance from the depot to the service start and end points, and an actual depot model that considers travel distance from the depot to the service start and end points. In particular, the actual depot model introduced the continuous condition for SAV travel sections with a simple constraint under mild assumptions about the SAV travel sections. Declarations Conflict of Interest The author declare that they have no conflict of interest Open Access This article is licensed under a Creative Commons Attri- bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The proposed models were applied to a hypothetical net- work, and the following results were confirmed: • Evaluation indicators for the virtual depot model tended to be higher than evaluation indicators for the actual depot model.f • In the actual depot model, the effect of SAV introduc- tion was greater when the cost of HDV ownership was considered.f 3.3 Sensitivity Analysis of the Cost of HDV Ownership a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value (b) SAV passenger ratio 0 2 4 6 8 10 12 14 16 18 20 10 20 30 40 50 60 70 80 90 100 SAV passenger rao(%) Budget limit for spesifying SAV travel secon 0 1500 2500 0 2000 4000 6000 8000 10000 12000 14000 10 20 30 40 50 60 70 80 90 100 Number of SAVs Budget limit for spesifying SAV travel secon 0 1500 2500 (c) Degree of improvement in the 0 2 4 6 8 10 10 20 30 40 50 60 70 80 90 100 Degree of improvement of obj. func. value(%) Budget limit for spesifying SAV travel secon 0 1500 2500 20 30 40 50 60 70 80 90 10 Budget limit for spesifying SAV travel secon (c) Degree of improvement in the objective function value (a) Number of SAVs Fig. 10 The relationship between the valuation indicators and the budget constraint specifying the HDV travel section when the depot is assumed to be Node 10. a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value Fig. 10 The relationship between the valuation indicators and the budget constraint specifying the HDV travel section when the depot is assumed to be Node 10. a Number of SAVs, b SAV passenger ratio, c Degree of improvement in the objective function value 1 3 92 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 the inclusion of binary variables. Future work includes the application of efficient solution algorithms such as Bender’s decomposition algorithm, which decomposes binary and real variables (Mesbah et al. [8]). Finally, all indicators were larger when the depot was located at Node 10 than when it was located at Node 1, indicating that the effect of introducing SAV travel sections is greater when the depot is located in a central area. Abbreviations SAV: Shared autonomous vehicle; HDV: Human driv- ing vehicle; OD: Origin and destination; STEN: Space-time extended network References • In the actual depot model, the effect of SAV introduction depended on the depot location. 1. Etminan, R.G., Patel, R.K., Kermanshachi, S., Rosenberger, J.M., Weinreich, D., Foss, A.: Integration of shared autonomous vehicles (SAVs) into existing transportation services: A focus group study. Transp. Res. Interdiscip. Perspect. 12, 100481 (2021) • The computation time of the virtual depot model was much greater than computation time of the actual depot model. 2. Narayanan, S., Chaniotakis, E., Antoniou, E.: Shared autonomous vehicle services: A comprehensive review. Transp. Res. Part C 111, 255–293 (2020) With regard to the second point above, when a manual driving vehicle is owned, its purchase and maintenance costs are not considered. Thus, the introduction of SAV travel sec- tions should be considered before a drastic change occurs, such as the replacement of all internal combustion engine vehicles with electric vehicles. 3. Li, Y., Long, J., Yu, M.: A time-dependent shared autonomous vehicle system design problem. Transp. Res. Part C 124, 102956 (2021) 4. Agatz, N.A.H., Erera, A., Savelsbergh, M.W.P., Wang, X.: Dynamic ride-sharing: a simulation study in metro Atlanta. Transp. Res. Part B 45, 1450–1464 (2011) There were several limitations in this study. First, because driving behavior is assumed to be identical between SAVs and HDVs, this study did not consider the effect of increased link capacity due to the increased SAV penetration. Therefore, this study may have underestimated the effect of SAV introduction. Second, because several assumptions are made in the virtual depot model, the model can be extended to a more general framework. Finally, although the proposed model can be solved by a commercial solver, the computational cost is high due to 5. Najmi, A., Rey, D., Rashidi, T.H.: Novel dynamic formulations for real-time ride-sharing system. Transp. Res. Part E 108, 122– 140 (2017) 6. Aiko, S., Thaithatkul, P., Asakura, Y.: Incorporating user prefer- ence into optimal vehicle routing problem of integrated sharing transport system. Asian Transp. Stud. 5(1), 98–116 (2018) 7. Shimamoto, H.: Optimal matching problem for ride-sharing con- sidering users’ schedule rearrangement. Int. J. Intell. Transp. Syst. Res. 18(3), 391–399 (2020) 8. Shimamoto, H., Matsuo, Y.: Development of time-space network assignment model considering ride sharing – comparison of mode 1 3 International Journal of Intelligent Transportation Systems Research (2024) 22:81–93 93 share between trip based model and activity based model. JSTE J. Traffic Eng. 9(2), A_44-A_53 (2023). (in Japanese) 17. References He, S., Ding, F., Li, C., Qi, Y.: Impact of connected and autono- mous vehicle dedicated lane on the freeway traffic efficiency. Eur. Transp. Res. Rev. 14, 12 (2022) fi g p 9. Seo, T., Asakura, Y.: Multi-objective linear optimization problem for strategic planning of shared autonomous vehicle operation and infrastructure design. IEEE Trans. Intell. Transp. Syst. 23, 3816–3828 (2022) p ( ) 18. Bell, M. G. H., Iida, Y.: Transportation Network Analysis, Chap- ter 1. Wiley (1997) 19. Transportation Networks for Research Core Team. Transportation Networks for Research. https://github.com/bstabler/Transporta tionNetworks. Accessed 24 Nov 2023 10. Maruyama, R., Seo, T.: Integrated public transportation sys- tem with shared autonomous vehicles and fixed-route transits: Dynamic traffic assignment-based model with multi-objective optimization. Int. J. Intell. Transp. Syst. Res. 21, 99–114 (2023) Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. p p y 11. Mesbah, M., Sarvi, M., Ouveysi, I., Currie, G.: Optimization of transit priority in the transportation network using a decomposi- tion methodology. Transp. Res. Part C 19, 363–373 (2011) 12. Mubark, M., Üster, H., Abdelghany, K., Khodayar, M.: Strategic network design and analysis for in-motion wireless charging of electric vehicles. Transp. Res. Part E 145, 102179 (2021) Hiroshi Shimamoto is an Associ- ate Professor at University of Miyazaki. He received his Doc- tor of Engineering Degree from Kyoto University in 2006. His research interests include trans- portation network analysis, traf- fic data analysis and transit models. Hiroshi Shimamoto is an Associ- ate Professor at University of Miyazaki. He received his Doc- tor of Engineering Degree from Kyoto University in 2006. His research interests include trans- portation network analysis, traf- fic data analysis and transit models. 13. Tran, C.Q., Keyvan-Ekbatani, M., Ngoduy, D., Watling, D.: Dynamic wireless charging lanes location model in urban net- works considering route choices. Transp. Res. Part C 139, 103652 (2022) 14. Lin, Y., Jia, H., Zou, B., Miao, H., Wu, R., Tian, J., Wang, G.: Multiobjective Environmentally sustainable optimal design of dedicated connected autonomous vehicle lanes. Sustainability 13, 3454 (2021) 15. Tani, R., Sumalee, A., Uchida, K.: Travel time reliability-based optimization problem for CAVs dedicated lanes. Transp. A 18(3), 1596–1600 (2022) 16. Hamad, K., Alozi, A.R.: Shared vs. dedicated lanes for auto- mated vehicle deployment: A simulation-based assessment. Int. J. Transp. Sci. Technol. 11, 205–215 (2022) 1 3
https://openalex.org/W3129420242	https://eprints.soton.ac.uk/447441/1/MCSE3052101.pdf	English	null	Using Jupyter for Reproducible Scientific Workflows	Computing in science & engineering	2,021	cc-by-sa	8,240	This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publicat The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 Marijan Beg Abstract—Literate computing has emerged as an important tool for computational studies and open science, with growing folklore of best practices. In this work, we report two case studies – one in computational magnetism and another in computational mathematics – where a dedicated software was exposed into the Jupyter environment. This enabled interactive and batch computational exploration of data, simulations, data analysis, and workﬂow documentation and outcome in Jupyter notebooks. In the ﬁrst study, Ubermag drives existing computational micromagnetics software through a domain-speciﬁc language embedded in Python. In the second study, a dedicated Jupyter kernel interfaces with the GAP system for computational discrete algebra and its dedicated programming language. In light of these case studies, we discuss the beneﬁts of this approach, including progress towards more reproducible and re-usable research results and outputs, notably through the use of infrastructure such as JupyterHub and Binder. conclusions. For a publication to be scientiﬁcally valid, it must present the methodology rigorously, so that readers can follow the “recipe” and Published by the IEEE Computer Society Using Jupyter for reproducible scientiﬁc workﬂows Marijan Beg Faculty of Engineering and Physical Sciences, University of Southampton, University Road, SO17 1BJ Southampton, United Kingdom Juliette Belin Logilab, 104 Boulevard Auguste Blanqui, 75013 Paris, France Thomas Kluyver European XFEL GmbH, Holzkoppel 4, 22869 Schenefeld, Germany Alexander Konovalov School of Computer Science, University of St Andrews, Jack Cole Building, North Haugh, KY16 9SX St Andrews, United Kingdom Min Ragan-Kelley Simula Research Laboratory, Martin Linges vei 25, 1364 Fornebu, Norway Nicolas M. Thi´ery Laboratoire de Recherche en Informatique, Universit´e Paris-Saclay, CNRS, 91405 Orsay, France Hans Fangohr Max Planck Institute for the Structure and Dynamics of Matter, Luruper Chaussee 149, 22761 Hamburg, Germany European XFEL GmbH, Holzkoppel 4, 22869 Schenefeld, Germany Faculty of Engineering and Physical Sciences, University of Southampton, University Road, SO17 1BJ Southampton, United Kingdom Marijan Beg Faculty of Engineering and Physical Sciences, University of Southampton, University Road, SO17 1BJ Southampton, United Kingdom Juliette Belin Logilab, 104 Boulevard Auguste Blanqui, 75013 Paris, France Thomas Kluyver European XFEL GmbH, Holzkoppel 4, 22869 Schenefeld, Germany Alexander Konovalov School of Computer Science, University of St Andrews, Jack Cole Building, North Haugh, KY16 9SX St Andrews, United Kingdom Min Ragan-Kelley Simula Research Laboratory, Martin Linges vei 25, 1364 Fornebu, Norway Nicolas M. Thi´ery Laboratoire de Recherche en Informatique, Universit´e Paris-Saclay, CNRS, 91405 Orsay, France Hans Fangohr Max Planck Institute for the Structure and Dynamics of Matter, Luruper Chaussee 149, 22761 Hamburg, Germany European XFEL GmbH, Holzkoppel 4, 22869 Schenefeld, Germany Faculty of Engineering and Physical Sciences, University of Southampton, University Road, SO17 1BJ Southampton, United Kingdom INTRODUCTION Research usually results in a publication that presents and shares the obtained ﬁndings and 1 © IEEE © IEEE Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. 1 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-perm This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 reproduce the results. If this criterion is met, the publication is considered reproducible. Re- producible publications are more easily re-usable and thus provide a signiﬁcant opportunity to make (often tax-payer funded) research more impactful. However, the reproducibility of computational work is usually hindered not only by a lack of data or meta-data but also by a lack of details on the procedure and tools used: In the last decade, literate computing has emerged as an important tool for computational studies and open science, with an ever-growing set of best practices. In this paper, we review and expand some of these best practices in the context of two case studies: computational magnetism and mathematics. This is based on the experience of enabling and applying Jupyter environments in these ﬁelds as a part of the OpenDreamKit (https://opendreamkit.org/) project. reproduce the results. If this criterion is met, the publication is considered reproducible. Re- producible publications are more easily re-usable and thus provide a signiﬁcant opportunity to make (often tax-payer funded) research more impactful. INTRODUCTION However, the reproducibility of computational work is usually hindered not only by a lack of data or meta-data but also by a lack of details on the procedure and tools used: 1) The source code of the software used is not available. To be able to run computational studies from the Jupyter environment, it is necessary to either have the simulation and/or analysis code exposed to a general-purpose programming language sup- ported by Jupyter, or have a dedicated Jupyter kernel for the computational libraries. Although the main topic of this work is the overview of features and capabilities of the Jupyter environ- ment for reproducible workﬂows, we begin by discussing how a computational library can be exposed to Jupyter as a necessary prerequisite. 2) Information on the computing environment, such as the hardware, operating system, supporting libraries, and (if required) code compilation details is not revealed. p 3) The exact procedure which led to the results reported in the publication is not shared. This should include the set of parameters used, the simulation and data analysis pro- cedure, and any additional data cleaning, processing, and visualization. Ideally, these are shared as open-source code and analy- sis scripts used to perform the simulation and to read, analyze, and visualize the re- sulting data. This way, the entire process can be repeated by re-running simulation and/or analysis scripts. A human-readable document detailing the computational steps taken, despite being “better-than-nothing”, is still insufﬁcient to ensure reproducibility, and keeping a detailed log of all steps taken during a computational study is often impos- sible. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. Prerequisite: Exposing computational libraries to the Jupyter environment Computational studies often use existing com- putational (legacy) tools. These could be executa- bles called from the command line or libraries that are used within a programming language. For the approach suggested here, these computational tools need to be accessible to scientists from a general-purpose programming language sup- ported by Jupyter (such as Python). For some do- mains, such as pure mathematics research, there are domain-speciﬁc languages with enough power to be used directly as the programming language in notebooks (e.g., Singular and GAP). In other areas, exposing computational tools to a general- purpose programming language is the key to inte- grating them into researchers’ custom code. A key beneﬁt of making computational tools available in a general-purpose programming language is that the computation can be driven ﬂexibly using the control structures provided by that language. For example, a simulation can conveniently be repeated with a range of parameters through a for- loop, rather than having to change a conﬁguration ﬁle for each value and trigger execution of the simulation manually. Reproducibility is a challenging question and spans a range of different topics. In this work, we focus on one of them. We describe the features and capabilities of the Jupyter environment that, in our view, make it a highly productive environ- ment for computational science and mathematics, while facilitating reproducibility. The topic of bitwise reproducibility is outside the scope of this work: even with the same hardware and same software, it may be difﬁ- cult to reproduce computational results to be bitwise identical. This can originate from the non- associativity of ﬂoating-point operations com- bined with parallel execution or from compiler optimizations. Bitwise reproducibility is not al- ways required to be achieved. Making the computational capability acces- sible from a general-purpose programming lan- 2 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 guage supported through a Jupyter kernel such as Python may be trivial – for example, if the required code is already a Python library. 2. Easily shareable Jupyter notebooks can be converted to other ﬁle formats, such as HTML, LaTeX and PDF. This is useful because someone working on a notebook can share it with collaborators, super- visors, or management without asking them to install any additional software. 1. One study – one document The notebook allows us to carry out an entire study within a single notebook and provides a complete and executable record of the process. It is possible to put the interpretation of the results into the same document, immediately below the graphical, tabular or text-based output that needs to be described. The “one study – one document” approach has immediate advantages: If the computational tool uses a programming language that Jupyter does not support, another possibility is to implement a Jupyter kernel for that language so that the computational library can be exposed to the Jupyter environment (as done for GAP and SageMath for example). Over time, scientiﬁc communities tend to accumulate functions and classes that are used repeatedly, and occasionally, through organic changes or a systematic restructuring of those computational capabilities, a domain-speciﬁc lan- guage is created, which is embedded in a general- purpose programming language such as Python. Depending on the design of this language, its existence and joint use by researchers of that domain can help to unify and improve computa- tional tasks in the community, avoid duplication of work, support transfer of knowledge and re- producibility. Examples of such domain-speciﬁc languages include Ubermag in magnetism, Sage- Math in pure mathematics, and the atomic simu- lation environment in chemistry [2]. • Scientists can be more efﬁcient as they do not have to search for parts of the study (scripts, data ﬁles, plots) when trying to understand the data and authoring the associated paper. data and authoring the associated paper. • The study is more easily reproducible (see item 6, below). However, putting all the code, data, and nar- rative into a single notebook could substantially affect the notebook’s readability. Thus, it is nec- essary to decide which parts of the code should be in libraries and imported in the notebook. Prerequisite: Exposing computational libraries to the Jupyter environment When the computational functionality is locked into an executable, one can create an interface layer so that functionality can be accessed via a Python function or class [1]: input parameters will then be translated into conﬁguration ﬁles, the exe- cutable called, outputs retrieved, and ﬁnally, the results returned. and many more) are supported through other Jupyter Notebook kernels. In this work, we sug- gest using a Jupyter research environment from which computational studies can be driven and conducted efﬁciently. In this section, we discuss the beneﬁts of using the Jupyter environment for reproducible scientiﬁc workﬂows. Features of the Jupyter research environment Project Jupyter is a set of open-source soft- ware projects for interactive and exploratory com- puting emerging from IPython. The central com- ponent offered by Jupyter is the Jupyter Notebook – a web-based interactive computing platform. It allows users to create data- and code-driven narra- tives that combine live (re-executable) code, equa- tions, narrative text, interactive dashboards, and other rich media. Jupyter Notebook documents provide a complete and executable record of a computation that can be shared with others in a way that has not been possible before [3]. Within the Jupyter Notebook, all libraries available in Python can be imported and combined ﬂexibly. Other languages (such as Julia, R, Haskell, Bash, Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. 3. Interactive execution or as batch job Using a Jupyter notebook often involves in- teractively editing it, executing cells, inspecting computed outputs, modifying commands, and re- executing, while understanding the computational research question. Once a useful processing se- quence has been found, the researcher often wants to repeat that, potentially with different input data. For such scenarios, a notebook can be executed from the command line (using the nbconvert tool), treating the notebook like a script or a batch job. As the notebook executes 3 Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 5, below). in batch mode, it computes the output cells, including images and other multimedia, as if it were executed interactively, and the outputs are stored into the notebook ﬁle for later analysis and inspection. Execution of notebooks as a script is a convenient way to use the computational power of a high-performance computing facility where such notebook jobs can be submitted to the batch queue. 5. Executable interactive documents in the cloud (Binder) The open-source Binder project [4] and Binder instances such as myBinder offer cus- tomized computational environments in the cloud on-demand, in which notebooks can be exe- cuted interactively. To use the free myBinder service, one needs to create a publicly read- able git-repository containing Jupyter notebooks and a speciﬁcation of the software required to execute these notebooks. This speciﬁca- tion follows existing standards, such as a Python-style requirements.txt ﬁle, conda environment.yml ﬁle, or Dockerfile. The myBinder service is invoked when a URL is requested containing the path to the GitHub repository. The myBinder service searches that repository for the software speciﬁcation, creates a suitable container, adds a Jupyter server to the container, and exposes that server to the user. Figure 1 offers an artistic illustration of a typical scenario for using Binder in the research workﬂow. Other use cases include: Where input data needs to be varied, two solutions are available: nbparameterise and papermill. 3. Interactive execution or as batch job With these tools, assignments in the ﬁrst cells of a notebook can be modiﬁed before the notebook is executed as a script. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. 4. Static and interactive software documention The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to pub The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 Figure 1. An artistic depiction of a scenario in which a researcher shares her computational workﬂow with others in the Jupyter environment, taking advantage of the Binder project. Licensed under “Creative Commons License, Attribution-ShareAlike” (enables reuse and adaptations, as long as credits to the original author are kept). © CC BY-SA Figure 1. An artistic depiction of a scenario in which a researcher shares her computational workﬂow with others in the Jupyter environment, taking advantage of the Binder project. Licensed under “Creative Commons License, Attribution-ShareAlike” (enables reuse and adaptations, as long as credits to the original author are kept). © CC BY-SA 6. Reproducibility – combining data, code, and software environment ing commands, and outputs, demonstrating the process leading to the result obtained with that notebook. By sharing the notebooks in a public repository, a DOI can be assigned via Zenodo to preserve the repository’s content permanently and make it citable. Reproducibility of scientiﬁc results is a cor- nerstone of our interpretation of science: only results that can be reproduced are accepted as proven insight. We see an emerging trend that journals and research councils increasingly (and justiﬁably) ask for details on how published re- sults can be reproduced, or at least expect authors to provide that information if a reader requests. p y • Notebooks that create central ﬁgures and state- ments of publications will likely need under- lying libraries. To re-execute the notebook, we need a way to specify a computational envi- ronment containing these libraries and Binder provides that possibility. Although specifying exact versions of underlying libraries is rec- ommended, Binder does not guarantee that this would lead to the same computational environment at any point in the future, and therefore, it cannot entirely address the issue of so-called software collapse where the underly- ing libraries and interfaces become deprecated, compilers and compiler optimization methods It is often impossible to truly document an entire computational workﬂow, software require- ments, hardware used, and other parameters within a conventional manuscript submission. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. 4. Static and interactive software documention Writing research software documentation is a particular challenge in academia. Small teams may not see the need to document their research code, as they can learn about it directly from one another. Jupyter notebooks offer an efﬁcient method for creating documentation. The popular Sphinx documentation software can use Jupyter note- books as the documentation source with the nbsphinx plugin, and create HTML and PDF documents. Demos and tutorials written in note- books can complement reference documentation in Sphinx’s default reStructuredText input format. Notebooks have several beneﬁts for extended examples in documentation: • Providing a computational environment for workshops or teaching purposes: participants are given the URL to invoke the service, and are presented a Jupyter session, in which they ﬁnd the notebooks the presenter/teacher has prepared. No software installation (other than having a modern web browser) is required for participants. • It takes less time to create documentation as the author can type commands and explana- tions into the same document, and the outputs that the commands produce (text and images) appear immediately in the notebook. • Providing interactive documentation: Given Binder-compatible speciﬁcations, documenta- tion can be presented as an executable note- book through myBinder, allowing the person reading the documentation to interactively ex- plore the the software’s behaviour (the com- mand can be modiﬁed and run again). • After changing the user interface or computa- tional algorithms, re-executing the documen- tation notebooks will often show where the documentation needs changing. • Demonstrating and disseminating small com- putational studies: Jupyter notebooks can be used to document computational processes. For example, for dissemination or to demonstrate reproducibility, as we explain in item 6, below. • Tools like nbval can automatically re-execute the notebooks and raise test errors if the ex- ecution fails, or the computed outputs have changed. This means continuous integration can be used to check the documentation and warn developers if changes in the code affect the illustrated behaviour. The related Voil`a project can execute note- books (for example on myBinder) and hide all code cells, making an interactive dashboard to display and explore data without the source code. • Using Binder, the documentation notebook can be executed interactively by the user (see item 4 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. 4. Static and interactive software documention The Jupyter-based research environment can help because it makes the process of publishing repro- ducible computational results easily achievable: • The “one document – one study” model au- tomatically records all parameters, process- 5 5 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to public The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 Figure 2. An artistic illustration of a conﬁgurable JupyterHub where a lecturer provides a customized software environment to support their teaching. JupyterHub can be accessed and used through a web browser and does not require local installation of any software. Institutional computing and storage resources are used, and users have to authenticate themselves. Licensed under “Creative Commons License, Attribution-ShareAlike” (enables reuse and adaptations, as long as credits to the original author are kept). © CC BY-SA Figure 2. An artistic illustration of a conﬁgurable JupyterHub where a lecturer provides a customized software environment to support their teaching. JupyterHub can be accessed and used through a web browser and does not require local installation of any software. Institutional computing and storage resources are used, and users have to authenticate themselves. Licensed under “Creative Commons License, Attribution-ShareAlike” (enables reuse and adaptations, as long as credits to the original author are kept). © CC BY-SA change, etc. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. 7. Remote access to institutional compute resources – JupyterHub 7. Remote access to institutional compute resources – JupyterHub • By publishing the notebooks reproducing cen- tral results together with software environment speciﬁcations for Binder in an open repository, anyone with Internet access and a browser can inspect and re-execute these notebooks and thus reproduce the publication. The discussion above assumed that notebooks were running on the user’s computer. The Jupyter- Hub software allows institutional provision of Jupyter Notebook services. It allows users of an institution to authenticate with their organi- zational credentials and access a Jupyter environ- ment running on the institution’s infrastructure. Typically, any ﬁles and folders the user is allowed to access will also be made available to them through JupyterHub, including access to shared data and folders where they can save their note- books. A key beneﬁt of being able to reproduce a publication in this way is that the study can be modiﬁed and extended easily: reproducibility enables re-usability. This can provide efﬁciency gains for science overall as it allows scientists to focus on new insights rather than having to spend time re-creating known knowledge as a starting point of their new study. The institution generally predeﬁnes the soft- ware environment in which the notebook server executes. However, the technology is available to use the software speciﬁcation as for Binder to create a customized computing environment on- 6 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publicati The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 we may create different results from when we execute all cells in order. demand. A vital point of the user experience is that only a web browser is required to access the JupyterHub and to carry out computational work using these resources remotely. Figure 2 shows an artistic illustration of the scenario where an instructor works with their institution to provide students with a customized software environment. Other use cases of JupyterHub installations in- clude research facilities and universities provid- ing access to their (high-performance) computing resources through Jupyter notebooks, where tra- ditionally ssh or remote desktops may have been used. (d) Sustainability of myBinder.org ( ) y f y g Since 2016 (and at time of writing), a federation of Binder instances is operated as a service avail- able on the world wide web at mybinder.org. The federation is operated by the Jupyter team, in collaboration with the Turing Institute and GESIS (Leibniz Institute for Social Sciences). Comput- ing resources are sponsored by Google Cloud, OVHCloud, the Turing Institute, and GESIS. The federation serves approximately 25,000 Binder instances on a typical weekday, with the Google Cloud instance serving approximately 70% of this trafﬁc. These sponsorships are mostly re- newed annually and can result in members of the federation halting the operation due to periods without funding. We hope that the sustainability of the Binder federation will improve if more ﬁnancially-stable members join, for example, as 8. Blending script and GUI-driven exploration methods The IPyWidgets Jupyter extension pro- vides selection menus, sliders, radio buttons, and other GUI-like graphical interaction widgets to Jupyter notebooks. The Notebook allows embed- ding such graphical widgets inside the notebook, and users can combine the usual scripted analysis with activation of such widgets where desired. They can be used, for example, to vary the input parameter values and explore a data set or computational results. Although less reproducible than typed commands, widgets can be useful for rapid feedback on different possibilities. (c) Rapid development of Jupyter ecosystem Improvements to Project Jupyter and the sur- rounding software ecosystem appear at a rapid rate. For instance, for the issues described in items (a) and (b), contributions providing solu- tions have already emerged, and there is no space here to introduce more of the multitude of high- productivity tools that have been created. It is challenging to follow all the developments and ﬁnd the most appropriate tool for a given task. Conferences such as JupyterCon help disseminate new contributions and help to avoid duplication of development efforts. (b) Opening Jupyter Notebook ( ) p g py Among the feedback we receive from some users who come across Jupyter notebooks for the ﬁrst time is that the way a Jupyter server is started is “strange”. Users who are not used to the command prompt may ﬁnd it unusual to open an application that way, instead of “double-clicking”. 8. Blending script and GUI-driven exploration methods 7. Remote access to institutional compute resources – JupyterHub There is a practical solution to this. When the exploratory phase is completed, the best practice is to restart the kernel to ensure the notebook’s state is forgotten and then execute all cells from top to bottom. This ensures that the results in the notebook are from running the cells in order, and this version of the notebook should be saved and shared. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. 9. Potential disadvantages Above we focused on the features and ca- pabilities of the Jupyter research environment to support computational workﬂows in science. Here, we want to discuss some downsides that have come up either in our work or as feedback from users of Jupyter-based computational tools we developed. Computational magnetism (1.1 s) L= 8.75, m_init_vortex Running OOMMF ... (1.5 s) L= 8.75, m_init_flower Running OOMMF ... (1.1 s) L= 9.0, m_init_vortex Running OOMMF ... (1.5 s) L= 9.0, m_init_flower Running OOMMF ... (1.1 s) ]: L_array = np.linspace(8, 9, 5) vortex_energies, flower_energies = [], [] for L in L_array: vortex = minimise_system_energy(L, m_init_vortex) flower = minimise_system_energy(L, m_init_flower) vortex_energies.append(vortex.table.data.tail(1)['E'][0]) flower_energies.append(flower.table.data.tail(1)['E'][0]) import matplotlib.pyplot as plt plt.figure(figsize=(6, 4)) plt.plot(L_array, vortex_energies, 'o-', label='vortex') plt.plot(L_array, flower_energies, 'o-', label='flower') plt.xlabel(r'$L (l_{ex}$)') plt.ylabel(r'$E$ (J)') plt.grid() plt.legend(); L= 8.4, m_init_vortex Running OOMMF ... (1.7 s) L= 8.4, m init flower Running OOMMF ... (1.1 s) ]: from scipy.optimize import bisect def energy_difference(L): vortex = minimise_system_energy(L, m_init_vortex) flower = minimise_system_energy(L, m_init_flower) return (vortex.table.data.tail(1)['E'][0] - flower.table.data.tail(1)['E'][0]) cross_section = bisect(energy_difference, 8.4, 8.6, xtol=0.02) print(f'\nThe energy crossing occurs at L = {cross_section}lex' The Object-Oriented MicroMagnetic Frame- work (OOMMF) [5] is a micromagnetic simula- tion tool, initially developed during the 1990s at the National Institute of Standards and Technol- ogy (NIST). It solves non-linear time-dependent partial differential equations using the ﬁnite- difference method. It is probably the most widely used and most trusted simulation tool in the com- putational magnetism community. It was written in C++, wrapped with Tcl, and driven through conﬁguration ﬁles that follow the Tcl syntax. The typical computational workﬂow the user must follow to simulate a particular problem is to write a conﬁguration ﬁle. After that, the user runs OOMMF by providing the conﬁguration ﬁle to the OOMMF executable. When the OOMMF run is complete, results are saved in OOMMF- speciﬁc ﬁle formats. Finally, the user analyzes the result ﬁles. L= 8.4, m_init_vortex Running OOMMF ... (1.7 s) L= 8.4, m_init_flower Running OOMMF ... (1.1 s) L= 8.6, m_init_vortex Running OOMMF ... (1.6 s) L= 8.6, m_init_flower Running OOMMF ... (1.1 s) L= 8.5, m_init_vortex Running OOMMF ... (1.7 s) L= 8.5, m_init_flower Running OOMMF ... (1.1 s) L= 8.45, m_init_vortex Running OOMMF ... (1.6 s) L= 8.45, m_init_flower Running OOMMF ... (1.1 s) L= 8.425, m_init_vortex Running OOMMF ... (1.8 s) L= 8.425, m_init_flower Running OOMMF ... (1.2 s) L= 8.4375, m_init_vortex Running OOMMF ... (1.6 s) L= 8.4375, m_init_flower Running OOMMF ... (a) Undeﬁned notebook state Top-to-bottom arrangement of cells in a notebook implies that they should be executed in that order. One of the Jupyter Notebook’s key features is that the code cells can be executed in an arbitrary order – the user can select (and modify) any cell and then execute it. This can be useful while exploring a data set or a property of computation, or even to debug the cell’s code. The execution order used in a notebook is not stored when the notebook is saved. Therefore, it is critical to remember that, by executing cells out of order, 7 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 a part of the European Open Science Cloud initiative. a part of the European Open Science Cloud initiative. L= 8.0, m_init_vortex Running OOMMF ... (2.2 s) L= 8.0, m_init_flower Running OOMMF ... (1.1 s) L= 8.25, m_init_vortex Running OOMMF ... (1.8 s) L= 8.25, m_init_flower Running OOMMF ... (1.1 s) L= 8.5, m_init_vortex Running OOMMF ... (1.7 s) L= 8.5, m_init_flower Running OOMMF ... (1.1 s) L= 8.75, m_init_vortex Running OOMMF ... (1.5 s) L= 8.75, m_init_flower Running OOMMF ... (1.1 s) L= 9.0, m_init_vortex Running OOMMF ... (1.5 s) L= 9.0, m_init_flower Running OOMMF ... (1.1 s) : L_array = np.linspace(8, 9, 5) vortex_energies, flower_energies = [], [] for L in L_array: vortex = minimise_system_energy(L, m_init_vortex) flower = minimise_system_energy(L, m_init_flower) vortex_energies.append(vortex.table.data.tail(1)['E'][0]) flower_energies.append(flower.table.data.tail(1)['E'][0]) import matplotlib.pyplot as plt plt.figure(figsize=(6, 4)) plt.plot(L_array, vortex_energies, 'o-', label='vortex') plt.plot(L_array, flower_energies, 'o-', label='flower') plt.xlabel(r'$L (l_{ex}$)') plt.ylabel(r'$E$ (J)') plt.grid() plt.legend(); Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. Computational magnetism L= 8.0, m_init_vortex Running OOMMF ... (2.2 s) L= 8.0, m_init_flower Running OOMMF ... (1.1 s) L= 8.25, m_init_vortex Running OOMMF ... (1.8 s) L= 8.25, m_init_flower Running OOMMF ... (1.1 s) L= 8.5, m_init_vortex Running OOMMF ... (1.7 s) L= 8.5, m_init_flower Running OOMMF ... (1.1 s) L= 8.75, m_init_vortex Running OOMMF ... (1.5 s) L= 8.75, m_init_flower Running OOMMF ... (1.1 s) L= 9.0, m_init_vortex Running OOMMF ... (1.5 s) L= 9.0, m_init_flower Running OOMMF ... (1.1 s) [6]: L_array = np.linspace(8, 9, 5) vortex_energies, flower_energies = [], [] for L in L_array: vortex = minimise_system_energy(L, m_init_vortex) flower = minimise_system_energy(L, m_init_flower) vortex_energies.append(vortex.table.data.tail(1)['E'][0]) flower_energies.append(flower.table.data.tail(1)['E'][0]) import matplotlib.pyplot as plt plt.figure(figsize=(6, 4)) plt.plot(L_array, vortex_energies, 'o-', label='vortex') plt.plot(L_array, flower_energies, 'o-', label='flower') plt.xlabel(r'$L (l_{ex}$)') plt.ylabel(r'$E$ (J)') plt.grid() plt.legend(); L= 8.4, m_init_vortex Running OOMMF ... (1.7 s) L= 8.4, m_init_flower Running OOMMF ... (1.1 s) L= 8.6, m_init_vortex Running OOMMF ... (1.6 s) L= 8.6, m_init_flower Running OOMMF ... (1.1 s) L= 8.5, m_init_vortex Running OOMMF ... (1.7 s) L= 8.5, m_init_flower Running OOMMF ... (1.1 s) L= 8.45, m_init_vortex Running OOMMF ... (1.6 s) L= 8.45, m_init_flower Running OOMMF ... (1.1 s) L= 8.425, m_init_vortex Running OOMMF ... (1.8 s) L= 8.425, m_init_flower Running OOMMF ... (1.2 s) L= 8.4375, m_init_vortex Running OOMMF ... (1.6 s) L= 8.4375, m_init_flower Running OOMMF ... (1.2 s) The energy crossing occurs at L = 8.4375lex [7]: from scipy.optimize import bisect def energy_difference(L): vortex = minimise_system_energy(L, m_init_vortex) flower = minimise_system_energy(L, m_init_flower) return (vortex.table.data.tail(1)['E'][0] - flower.table.data.tail(1)['E'][0]) cross_section = bisect(energy_difference, 8.4, 8.6, xtol=0.02) print(f'\nThe energy crossing occurs at L = {cross_section}lex' gure 3. Running computational magnetism simu- tions through Python in a Jupyter notebook allows e use of the Python scientiﬁc stack and results in a elf-contained record combining narrative, code, and sults. Computational magnetism complements the- oretical and experimental methods to support research in magnetism. For example, it is used to develop sensors as well as data storage and information processing devices. It is used both in academia and industry to explain experimental observations, design experiments, improve device and product-designs virtually, and verify theoret- ical predictions. L= 8.0, m_init_vortex Running OOMMF ... (2.2 s) L= 8.0, m_init_flower Running OOMMF ... (1.1 s) L= 8.25, m_init_vortex Running OOMMF ... (1.8 s) L= 8.25, m_init_flower Running OOMMF ... (1.1 s) L= 8.5, m_init_vortex Running OOMMF ... (1.7 s) L= 8.5, m_init_flower Running OOMMF ... Computational magnetism (1.2 s) The energy crossing occurs at L = 8.4375lex [7]: from scipy.optimize import bisect def energy_difference(L): vortex = minimise_system_energy(L, m_init_vortex) flower = minimise_system_energy(L, m_init_flower) return (vortex.table.data.tail(1)['E'][0] - flower.table.data.tail(1)['E'][0]) cross_section = bisect(energy_difference, 8.4, 8.6, xtol=0.02) print(f'\nThe energy crossing occurs at L = {cross_section}*lex' One of the speciﬁc goals of a computational micromagnetic study is parameter-space explo- ration. More precisely, the user repeats the simu- lation for different values of input parameters by changing them in the conﬁguration ﬁle. It is often difﬁcult to automate this, and it is challenging for the user to keep a log of all steps performed in the entire micromagnetic study. Besides, postprocess- ing and analysis of results is performed outside OOMMF, using techniques and scripts that are mostly developed by the user, or carried out man- ually. Consequently, it is hard to track, record, and convey the exact simulation procedure. Without this information, resulting publications are gener- ally not reproducible. Figure 3. Running computational magnetism simu- lations through Python in a Jupyter notebook allows the use of the Python scientiﬁc stack and results in a self-contained record combining narrative, code, and results. Figure 3. Running computational magnetism simu- lations through Python in a Jupyter notebook allows the use of the Python scientiﬁc stack and results in a self-contained record combining narrative, code, and results. To address this situation, we developed a 8 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publicatio The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 Python interface to the OOMMF executable. This allows us to conduct computational magnetism simulations from within the Jupyter notebook to capitalize on the beneﬁts of this environment. A Jupyter notebook solving standard problem 3 can be found in the repository accompanying this work [M. Beg et al. Using Jupyter for repro- ducible scientiﬁc workﬂows. GitHub: https://gith ub.com/marijanbeg/2021-paper-jupyter-reprodu cible-workﬂows, DOI: 10.5281/zenodo.4382225 (2021)]. We show the two most relevant code cells inside the Jupyter notebook in Figure 3. Computational studies in mathematical research Many of the leading open-source mathemat- ical software systems (including GAP, LinBox, PARI/GP, OSCAR, SageMath, and Singular) have been made inter-operable with the Jupyter ecosys- tem through bespoke or general-purpose kernels (C++, Python, Julia, . . . ). Focusing on one of these systems, for the sake of concreteness, we il- lustrate how this supports sharing and publishing reproducible computational studies in mathemati- cal research together with the underlying research code. To demonstrate the use of Ubermag, we use standard problem 3 as an example. Standard problem 3 is a standardized problem posed by the micromagnetic community to test, validate, and compare different simulation tools. It describes a magnetic cube of edge length L with two different magnetization states that can occur as local energy minima, called the ﬂower state and the vortex state. The main question of standard problem 3 is “For what edge length L have the ﬂower state and the vortex state the same energy?” GAP is an open-source system for discrete computational algebra, with particular empha- sis on computational group theory. It is used routinely by mathematicians in these ﬁelds and beyond to support teaching and research, notably through computational exploration. It provides a domain-speciﬁc language, also called GAP, and a runtime system with a command-line interface. It can also be used as a library by other systems such as SageMath or OSCAR. In the conventional OOMMF workﬂow, it is necessary to run the micromagnetic simulations for different edge lengths and different initial magnetization states. After every simulation, the total energy is recorded and saved within a tab- separated data ﬁle. Finally, one extracts the mag- netic energy values from all the saved ﬁles and plots them as a function of edge length for both magnetization states. From the plot, an estimation of the energy crossing would be made. GAP has been developed for decades by a community of researchers, teachers, and research software engineers. It has an established mecha- nism for user-contributed extensions, called pack- ages, which may be submitted for the redistri- bution with the system, and a formal refereeing process. The current release of GAP (4.11.0) includes 152 packages that serve different pur- poses, from providing data libraries and extending the system’s infrastructure for testing and writ- By using our Python interface to OOMMF integrated into a Jupyter notebook, we can loop over different input parameters to obtain this crossing in a plot. Computational magnetism We developed a set of Python libraries we refer to as Ubermag, which expose the compu- tational capabilities of OOMMF so that it can be controlled from Python. These Python libraries provide a domain-speciﬁc language to deﬁne a micromagnetic problem [1]. A micromagnetic model, deﬁned using the domain-speciﬁc lan- guage, is not aware of the particular simulation tool that will perform the actual micromagnetic simulation, and it is only used to describe the model. When a simulation is required, the model is translated into the OOMMF conﬁguration ﬁle, the OOMMF executable is called, and the output ﬁles are read. By exposing the micromagnetic simulation capabilities to Python and driving the research from Jupyter Notebook, we have available all the beneﬁts of the Jupyter research environment. Ubermag and the Jupyter environment sim- plify the efforts to make computational mag- netism publications reproducible. For each ﬁgure in the publication, one notebook can be provided (ﬁnd examples in Refs. [6], [7]). Using Binder, the community can inspect and re-run all the calculations in the cloud and make the publication reproducible. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. Conclusions In this article, we discuss some of the chal- lenges researchers in computational science and mathematics experience in their everyday work. We focus on making computational exploration and workﬂows more efﬁcient, more reproducible, and re-usable. We demonstrate the beneﬁts of this approach by showing computational magnetism and computational mathematics use cases. We believe that Project Jupyter and its ecosystem, including JupyterHub and Binder, which allow no-installation browser-based use of notebooks and remote compute resources, can contribute sig- niﬁcantly towards more efﬁcient computational workﬂows, reproducibility and re-usability in sci- ence. These conclusions are part of a widespread trend among researchers in the computational community advocating for the use of literate computing – for example using Jupyter – for enhancing reproducible research. Let us illustrate this with the publication in Ref. [8], which presents a polynomial-time algo- rithm for solving a major problem in computa- tional group theory, which remained open since 1999 [9]. An essential addition to the paper is the author’s GAP implementation of the algorithm. The authors published this implementation in the publicly hosted repository. At once, this ensures long term archival through the Software Heritage project, and with a small additional step, it makes it citable through Zenodo. The repository contains an interactive narrative document – a Jupyter notebook using the GAP Jupyter kernel [10] – combining text, mathematics, inputs, and outputs, and may even be viewed as a slideshow (one could, of course, have separate notebooks for different purposes). Following best practices for organizing re- producible computational studies (see e.g., Ref. [11]), the code is not written in the notebook itself but loaded from external source ﬁles. These are text ﬁles that can be easily managed with version control, reused from multiple Jupyter notebooks, and tested using the GAP automated testing setup. Also, the authors made the repos- itory Binder-ready. Any user (e.g., readers or referees of the paper) can run the notebook and reproduce its execution on Binder itself or – with additional expertise to install the required assets – on their own computing resource. To achieve this, the authors followed the template in Ref. [12], which also brings in continuous integration to automatically check the code against several past, current, and development releases of GAP, and produce coverage reports on how thoroughly the tests exercise the code. It boils down to cre- Computational studies in mathematical research Furthermore, we can make use of the Python scientiﬁc stack, in particular, a root- ﬁnding method such as bisect from scipy. 9 9 Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 ating a tst directory with the test ﬁles, and adapting the conﬁguration ﬁles .travis.yml and .codecov.yml for Travis CI and Codecov services, respectively. ing documentation, to adding new functionality and sharing research codes that underpin their authors’ publications. The latter scenario may require speciﬁc expertise and motivation from a working mathematician who uses GAP, and not everyone will be able to invest efforts in sharing their code in this way. Furthermore, it is not always justiﬁable to organize a supplemen- tary code for a paper as a new GAP package. Instead, authors can combine Jupyter research environments with additional services and parts of the infrastructure for GAP packages to share re- producible computational studies while following good code development practices from the start. Bringing Jupyter interfaces to command-line based computational mathematics tools makes it possible to interface it with numerous JavaScript libraries, notably for visualization. For example, GAP packages Francy and JupyterViz extend the GAP Jupyter kernel [10] with interactive widgets and plotting tools, which can be tried from their Binder-ready repositories. Acknowledgments This work was ﬁnancially supported by the OpenDreamKit Horizon 2020 European Research Infrastructure project (676541) and the EPSRC Programme grant on Skyrmionics (EP/N032128/1). Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. REFERENCES 1. M. Beg, R. A. Pepper, and H. Fangohr, “User interfaces for computational science: A domain speciﬁc language for OOMMF embedded in Python”, AIP Advances, vol. 7, no. 5, p. 056025, 2017. 2. A. H. Larsen, J. J. Mortensen, J. Blomqvist, I. E. Castelli, R. Christensen, M. Dułak, J. Friis, M. N. Groves, B. Hammer, C. Hargus, E. D. Hermes, P. C. Jennings, P. B. Jensen, J. Kermode, J. R. Kitchin, E. L. 10 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/MCSE.2021.3052101 Kolsbjerg, J. Kubal, K. Kaasbjerg, S. Lysgaard, J. B. Maronsson, T. Maxson, T. Olsen, L. Pastewka, A. Pe- terson, C. Rostgaard, J. Schiøtz, O. Sch¨utt, M. Strange, K. S. Thygesen, T. Vegge, L. Vilhelmsen, M. Walter, Z. Zeng, and K. W. Jacobsen, “The atomic simulation environment—a python library for working with atoms”, Journal of Physics: Condensed Matter, vol. 29, no. 27, p. 273002, 2017. erKer nel/ https://gap-packages.github.io/ JupyterKernel/, Feb 2019, GAP package. 11. A. Rule, A. Birmingham, C. Zuniga, I. Altintas, S.-C. Huang, R. Knight, N. Moshiri, M. H. Nguyen, S. B. Rosenthal, F. P´erez, and P. W. Rose, “Ten simple rules for writing and sharing computational analyses in Jupyter notebooks”, PLOS Computational Biology, vol. 15, no. 7, pp. 1–8, 2019. 12. A. Konovalov, “Template for publishing reproducible GAP experiments in Jupyter notebooks runnable on Binder,” Feb 2020. [Online]. Available: https://doi.org/ 10.5281/zenodo.3662155 3. T. Kluyver, B. Ragan-Kelley, F. P´erez, B. Granger, M. Bussonnier, J. Frederic, K. Kelley, J. Hamrick, J. Grout, S. Corlay, P. Ivanov, D. Avila, S. Abdalla, C. Willing and Jupyter Development Team, “Jupyter Notebooks – a publishing format for reproducible com- putational workﬂows”, Positioning and Power in Aca- demic Publishing: Players, Agents and Agendas, 87-90, 2016. 4. Project Jupyter, M. Bussonnier, J. Forde, J. Freeman, B. Granger, T. Head, C. Holdgraf, K. Kelley, G. Nalvarte, A. Osheroff, M. Pacer, Y. Panda, F. Perez, B. Ragan Kel- ley, and C. Willing, “Binder 2.0 - Reproducible, inter- active, shareable environments for science at scale”, Proceedings of the 17th Python in Science Conference, pp. Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org. REFERENCES 113 – 120, 2018. 5. M. J. Donahue and D. G. Porter, “OOMMF User’s Guide, Version 1.0”, Interagency Report NISTIR 6376 National Institute of Standards and Technology, Gaithersburg, MD, 1999. 6. M. Beg, R. A. Pepper, D. Cort´es-Ortu˜no, B. Atie, M.- A. Bisotti, G. Downing, T. Kluyver, O. Hovorka, and H. Fangohr, “Stable and manipulable Bloch point,” Sci- entiﬁc Reports, vol. 9, no. 1, p. 7959, 2019. Code repository for reproducibility at https://github.com/m arijanbeg/2019-paper-bloch-point-stability. 7. M. Albert, M. Beg, D. Chernyshenko, M.-A. Bisotti, R. L. Carey, H. Fangohr, and P. J. Metaxas, “Frequency- based nanoparticle sensing over large ﬁeld ranges us- ing the ferromagnetic resonances of a magnetic nan- odisc”, Nanotechnology, vol. 27, no. 45, p. 455502, 2016. Code repository for reproducibility at https://gi thub.com/maxalbert/paper-supplement-nanoparticle-s ensing. 8. A. Borovik and S¸ . Yalc¸ınkaya, “Adjoint representations of black box groups PSL2(Fq),” J. Algebra, vol. 506, pp. 540–591, 2018. 9. L. Babai and R. Beals, A polynomial-time theory of black box groups I, ser. London Mathematical Society Lecture Note Series. Cambridge University Press, 1999, vol. 1, pp. 30–64, 1998. 10. M. Pfeiffer and M. Martins, “JupyterKernel, Version 1.3”, h t t p s : / / g a p - p a ck a g e s . g i t h u b. i o / J u p y t 11 Copyright (c) 2021 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.
https://openalex.org/W3195143548	https://europepmc.org/articles/pmc8479897?pdf=render	English	null	Direct oral anticoagulants versus low molecular weight heparins for the treatment of cancer-associated thrombosis: a cost-effectiveness analysis	Thrombosis journal	2,021	cc-by	7,505	© The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Direct oral anticoagulants versus low molecular weight heparins for the treatment of cancer-associated thrombosis: a cost-effectiveness analysis Kaidireyahan Wumaier1, Wenqian Li1, Naifei Chen2 and Jiuwei Cui2* Kaidireyahan Wumaier1, Wenqian Li1, Naifei Chen2 and Jiuwei Cui2* Wumaier et al. Thrombosis Journal (2021) 19:68 https://doi.org/10.1186/s12959-021-00319-1 Wumaier et al. Thrombosis Journal (2021) 19:68 https://doi.org/10.1186/s12959-021-00319-1 Correspondence: cuijw@jlu.edu.cn 2Department of Cancer center, the First Hospital of Jilin University, Changchun, Jilin 130021, China Full list of author information is available at the end of the article Abstract Background: Recently, direct oral anticoagulants (DOACs) have been included in guidelines for the treatment of cancer-associated thrombosis (CAT) to be extended to suitable cancer patients. The purpose of this study was to compare the cost-effectiveness of using DOACs and low molecular weight heparins (LMWHs) for treating CAT from the perspective of the Chinese healthcare system. Methods: A Markov model was constructed to estimate the cost-effectiveness of the two strategies with a 6-month and 5-year time horizon. Input parameters were either sourced from the clinical trial, published literature. The primary outcome of the model was reported as incremental cost-effectiveness ratios (ICERs). Sensitivity analyses were performed to test model uncertainty. Results: The 6-month cost of DOACs was $ 654.65 with 0.40 quality adjusted life-years (QALYs) while the 6-month cost of LMWHs was $USD 1719.31 with 0.37 QALYs. Similarly, treatment with DOACs had a lower cost ($USD 657.85 vs. $USD 1716.56) and more health benefits (0.40 QALYs vs. 0.37 QALYs) than treatment with LMWHs in a subgroup of patients with gastrointestinal malignancy. We found treatment with DOACs would result in a large reduction in cost ($USD 1447.22 vs. $USD 3374.70) but a small reduction in QALYs (3.07 QALYs vs. 3.09 QALYs) compared with LMWHs over a 5-year time frame, resulting in an ICER of $USD 112895.50/QALYs. Sensitivity analysis confirmed the robustness of the results. Conclusion: As compared to LMWHs, DOACs can be a cost-saving anticoagulant choice for the treatment of CAT in the general oncology population and gastrointestinal malignancy population. Keywords: Cost-effectiveness, Direct oral anticoagulants, Low molecular weight heparin, Venous thromboembolism, Cancer-associated thrombosis * Correspondence: cuijw@jlu.edu.cn Materials and methods Overview of the model A Markov model also called a state transition model, is a commonly used approach in decision analysis to simu- late disease progression in a defined period of time. The advantage of Markov models is that Markov models can model risks over time, which enables extrapolation to the future and reduces the number of simplifying as- sumptions required. Markov models have been used ex- tensively in the medical literature, and offer an appealing framework for modeling medical decision making, with potential powerful applications in decision support sys- tems and health economics analysis. In cost-effectiveness research, Markov models are made to analyze competing treatment strategies available to a patient that can change that patient’s health state. A Markov model has a time-horizon, which is separated into fixed time periods referred to as cycles. During each of these cycles, the co- hort may transition between a finite number of health states according to appropriate probabilities. Costs and effects are typically incorporated into these models as a mean value per state per cycle. It is thus possible to cal- culate the expected cost and expected outcome of each option under evaluation. For a given option, the ex- pected cost (outcome) is the sum of the costs (out- comes) of each consequence weighted by the probability of that consequence. Cancer-associated thrombosis (CAT) events impose a significant economic burden on the healthcare system. Compared to cancer patients without VTE, cancer pa- tients with VTE have been shown to have three times as many all-cause hospitalizations, more days spent in the hospital, and a significantly higher number of outpatient visits [11], which results in significant healthcare costs among a cancer population of patients. Mean total hospitalization costs were 2.5-times ($17,089) higher among cancer patients with VTE compared to patients without VTE and accounted for 62% of the VTE-related total healthcare costs [12]. Nevertheless, the previous study has shown that VTE-related costs among cancer patients vary according to the type of anticoagulant treatment used [13]. Taking into account all the above, appropriate anticoa- gulation is of utmost importance for both clinical and economic reasons among patients with cancer. Low mo- lecular weight heparins (LMWHs) have been recom- mended as the standard treatment of VTE in patients with malignancies for many years [14, 15]. Introduction anticoagulants should be carefully considered in the decision-making process by balancing the clinical bene- fits and the related costs. Earlier studies found that the cost-effectiveness results for DOACs were uneven in dif- ferent countries as compared with those for LMWHs [24–27], reflecting that cost-effectiveness may depend heavily on country-specific health system organizations and economics. Venous thromboembolic (VTE), which encompasses the diagnoses of both deep vein thrombosis (DVT) and pul- monary embolism (PE), is a common complication of malignancy associated with serious mortality, morbidity, and health economic consequences [1–3]. Patients with cancer are significantly more likely to de- velop VTE than in individuals without this disease, a ninefold increased risk is reported in such patients as compared with the normal population [4]. Of all cancer patients, while up to 50% have VTE at autopsy, with VTE being the second cause of death after cancer [5, 6]. In addition, VTE is associated with a variety of adverse consequences including an increased risk of VTE recur- rence, major bleeding in cancer patients. The statistics revealed that the risk of recurrent VTE and bleeding was approximate 10–20 and 10% annually [7–9]. Moreover, VTE also has a negative impact on the quality of life of patients with malignancies [10]. Given these concerns, we developed cost-effectiveness analyses on the use of the DOACs versus LMWHs in tumor patients with VTE from the Chinese healthcare system, which provides evidence of its clinical and finan- cial benefit for decision-making. © The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Page 2 of 10 Page 2 of 10 Wumaier et al. Thrombosis Journal (2021) 19:68 Materials and methods Overview of the model Nevertheless, the implementation and adherence between recommen- dations and clinical behavior are suboptimal due to de- creased patient satisfaction, decreased adherence rates, and increased cost of LMWHs [16]. Overcoming some of these disadvantages of LMWHs, the so-called direct oral anticoagulants (DOACs) have been recently intro- duced: dabigatran, rivaroxaban apixaban, and edoxaban [17–20], which represent a convenient and patient- centric anticoagulation strategy. Most notably, more re- cent recommendations from guidelines on the use of DOACs for the treatment of VTE to be extended to suit- able cancer patients [21, 22], with emerging data sup- porting their safety and efficacy in the care of cancer patients [23]. However, the use of these new oral We constructed a Markov model using proprietary software (TreeAge Pro 2011 Software, Williamstown, MA) concerning a hypothetical reference case, which was similar to the approach adopted in previously pub- lished studies [24, 26–29]. A hypothetical cancer popula- tion of 64-year-old, 70 kg, and with VTE event receiving treatment with DOACs or LMWHs was considered for the model. A 1-month cycle length with a 6-month and 5-year time horizon was used. The 6-month time hori- zon was chosen based on the applied data period from the randomized controlled trials (RCTs) and the 5-year time horizon was chosen as it is commonly used to Wumaier et al. Thrombosis Journal (2021) 19:68 Page 3 of 10 Page 3 of 10 reflect important clinical and economic impacts of DOACs for CAT and general cancer survival [26, 30–32]. The Markov model, consisting of 10 health states, as depicted in Fig. 1, included on anticoagulant treatment, off anticoagulant treatment, recurrent pulmonary embol- ism (rPE), recurrent deep vein thrombosis (rDVT), intra- cranial hemorrhage (ICH), non-ICH major bleeding (MB), clinically relevant non-major bleeding (CRNMB), PE-related death, MB-related death, and death by any case. Specifically, patients entered the model following a VTE event, at the beginning of the decision tree, the reflect important clinical and economic impacts of DOACs for CAT and general cancer survival [26, 30–32]. patients would receive one of the four following agents for the treatment of VTE in cancer patients: apixaban, rivaroxaban, edoxaban, or LMWHs that dosages for agents were based on their respective trials, and then they either remained in their current on-treatment state, moved to an event state, transitioned to an off- treatment state or died, during the course of 1- monthly cycles. Parameters of the model input the costs for inpatient admissions due to major bleeding events [40]. Costs for other states were also based on values in the previously published literature. All the costs were calculated and reported in US dollars (USD) with the average exchange rate in 2020 (¥ = $0.14). Also, a discount rate of 5% was used, as recommended by Chinese guidelines for pharmacoeconomic evaluations [45] each year. All costs are reported in Table 1. The clinical inputs for various event probabilities used in the model are summarized in Table 1. The probabil- ities of events of DOACs and LMWHs during the first 6 months were obtained from 4 good quality RCTs includ- ing the Hokusai VTE Cancer trial [33], Select-D [34], the Caravaggio study [35], and ADAM VTE trial [36] in which each of the DOACs were directly compared with LMWHs for the treatment of VTE focused on patients with active cancer. All studies had the primary efficacy outcome (recurrent VTE) and the primary safety out- come (major bleeding). In each study, patients were followed for at least 6 months. DOACs were shown to be noninferior to dalteparin for recurrent VTE and major bleeding. Bleeding was more common in patients with GI malignancies taking edoxaban and rivaroxaban compared with dalteparin [33, 34]. In contrast, apixaban was not associated with an increased risk of bleeding compared with dalteparin in the ADAM and Caravaggio trials [35, 36]. Transition probabilities (TP) for 7–12 months were derived directly from the Hokusai-VTE study and the same estimates were extrapolated to the time horizon beyond 12 months (Supplemental Table 1). The time-varying TP for recurrent VTE when off anti- coagulant treatment was estimated from a large popula- tion study of cancer patients [37]. The probability of events of bleeding seen in patients with gastrointestinal malignancy was derived from the randomized controlled trials as above, including the Hokusai-VTE [33], Select- D [34], Caravaggio [35] trials. The event rates were translated into monthly transition probabilities with the following formula: Tp = 1 −(1 −p)^(1/n) (with Tp = monthly probability of events, p = event probability as reported in the literature, and n = number of months). y p The quality adjusted life-years (QALYs; duration times utility) was incorporated in the model by using the values of utility. Evidence from previously published lit- erature was used to determine the various utility values. Analysis We assessed the cost-effectiveness of treatment with DOACs compared to LMWHs among patients with CAT. In addition, given the increased rate of bleeding seen in patients with gastrointestinal malignancy on edoxaban [33] and rivaroxaban [47], a subgroup cost- effectiveness analysis was performed on this patient population. The primary outcome measure of this study is the in- cremental cost-effectiveness ratio (ICER), which is the ratio of incremental cost and incremental effect between the two groups. According to the world health organization (WHO) recommendation, When the ICER was less than three times the gross domestic product (GDP) per capita, cost-effectiveness would be considered [48]. We used three times the per-capita GDP of China in 2020 ($10,142.58) with willingness to pay (WTP) thresholds of US $30,427.74 per QALY as the cost- effectiveness threshold. p ) The cost analysis was evaluated from the healthcare system perspective setting in China. In analysis, it in- cludes patients’ direct medical costs related to drugs and complications, without considering indirect costs and direct non-medical costs. The daily drug acquisition costs of DOACs and LMWHs were collected from pub- lic databases [38]. An average of the edoxaban, rivaroxa- ban, and apixaban total drug costs was used for the DOACs arm. The wording DOACs refer to apixaban, edoxaban, or rivaroxaban, where did not include dabiga- tran as dabigatran was not used in any study. Costs for enoxaparin were used for this model due to its wide- spread use in China, although the clinical trials in cancer patients have used LMWHs such as dalteparin. Monthly costs (each cycle) were derived from 30-day prescrip- tions of the drugs at the labeled dosing frequency. The cost of symptomatic DVT or PE considered both the diagnosis and hospitalization costs incurred for such events [39]. The resource use in managing a major bleeding event was based on a Chinese study analyzing To explore the effect of parameter uncertainty in the model, we performed one-way sensitivity analysis (OWSA) and probabilistic sensitivity analysis (PSA). In OWSA, the value range of each parameter was either based on the reported or estimated 95% CIs in the refer- enced studies or determined by assuming a 20% change from the point estimate in the base-case analysis. The 10 most influencing parameters were presented in a tor- nado diagram. PSA was performed using a Monte Carlo simulation with 1000 iterations. Materials and methods Overview of the model Each state was associated with a cost and utility weighting to calculate the total costs and quality adjusted life-years (QALYs) of patients simu- lated in the model. The Markov model, consisting of 10 health states, as depicted in Fig. 1, included on anticoagulant treatment, off anticoagulant treatment, recurrent pulmonary embol- ism (rPE), recurrent deep vein thrombosis (rDVT), intra- cranial hemorrhage (ICH), non-ICH major bleeding (MB), clinically relevant non-major bleeding (CRNMB), PE-related death, MB-related death, and death by any case. Specifically, patients entered the model following a VTE event, at the beginning of the decision tree, the Fig. 1 Model diagram. Abbreviations: DOACs, direct oral anticoagulants; LMWHs, low molecular weight heparins; onAC, on anticoagulation treatment with no event; rDVT, recurrent deep-vein thrombosis; rPE, recurrent pulmonary embolism; ICH, intracranial hemorrhage; CRNMB, clinically relevant non-major bleeding; offAC, anticoagulant treatment discontinuation; offDVT, risk of deep-vein thrombosis while off-treatment; offPE: risk of pulmonary embolism while off-treatment Fig. 1 Model diagram. Abbreviations: DOACs, direct oral anticoagulants; LMWHs, low molecular weight heparins; onAC, on anticoagulation treatment with no event; rDVT, recurrent deep-vein thrombosis; rPE, recurrent pulmonary embolism; ICH, intracranial hemorrhage; CRNMB, clinically relevant non-major bleeding; offAC, anticoagulant treatment discontinuation; offDVT, risk of deep-vein thrombosis while off-treatment; offPE: risk of pulmonary embolism while off-treatment Page 4 of 10 Page 4 of 10 Wumaier et al. Thrombosis Journal (2021) 19:68 Parameters of the model input As the literature on the utility of various events in can- cer patients with VTE events is scarce, thus, most data were obtained from VTE patients without cancer [42, 46]. The base utility was considered to be 0. 95 and oral anticoagulant treatments were assumed not to change the utility value [42]. The utility inputs of the direct ef- fects of the treatment drugs and the series of clinical events are reported in Table 1. Analysis The distributions as- sumed for the input parameters were gamma (cost), beta (utility weights and TP), and log-normal (RR). All the analyses were performed in TreeAge Pro 2011. Page 5 of 10 Wumaier et al. Thrombosis Journal (2021) 19:68 Table 1 Parameters for model input with a cycle length of 1 month Parameters Base case Ranges Distribution Source 1-6 month transition probabilities (tp) a, % DOACs Recurrent DVT 0.33 0.26–0.40 β [33–36] Recurrent PE 0.54 0.43–0.65 β [33–36] Fatal PE 4.45 3.56–5.34 β [33–36] MB 0.70 0.56–0.84 β [33–36] Fatal MB 0.28 0.22–0.34 β [33–36] ICH 0.02 0.016–0.024 β [33–36] Fatal ICH 0 0 β [33–36] Major GI bleeding 1.17 0.94–1.40 β [33–36] CRNMB 1.81 1.45–2.17 β [33–36] Death of any case 4.46 3.57–5.35 β [33–36] Treatment discontinuation 2.91 2.33–3.49 β [33–36] LMWHs Recurrent DVT 0.61 0.49–0.73 β [33–36] Recurrent PE 0.74 0.59–0.89 β [33–36] Fatal PE 2.24 1.79–2.69 β [33–36] MB 0.47 0.38–0.56 β [33–36] Fatal MB 1.26 1.01–1.51 β [33–36] ICH 0.08 0.06–0.10 β [33–36] Fatal ICH 5.45 4.36–6.54 β [33–36] Major GI bleeding 0.56 0.45–0.67 β [33–36] CRNMB 1.09 0.87–1.31 β [33–36] Death of any case 4.52 3.62–5.42 β [33–36] Treatment discontinuation 4.73 3.78–5.68 β [33–36] OffDVT 1.90 1.52–2.28 β [37] OffPE 2.03 1.63–2.44 β [37] Costs, $ DOACs 1st month 207.34 165.87–248.81 γ [38] DOACs 2nd month onwards 825.89 660.71–991.07 γ [38] LMWHs (enoxaparin) 1st month 149.91 119.93–179.89 γ [38] LMWH (enoxaparin)2nd month onwards 412.94 365.30–547.94 γ [38] DVT 693 329–941 γ [39] PE 1121 448–1793 γ [39] MB 654 603–704 γ [40] ICH 3834 2684–4984 γ [41] CRNMB 8.25 5.77–10.72 γ [41] Utilities β Base 0.95 0.76–1.00 β [42] Table 1 Parameters for model input with a cycle length of 1 month Source Wumaier et al. Discussion To our knowledge, this is the first study to evaluate the cost-effectiveness of all available DOACs simultaneously compared with LMWHs for treatment in patients with cancer-associated VTE. In the present study, an eco- nomic analysis of four potentially competing treatment agents in different treatment durations (5-year and 6- month time horizon, respectively) was undertaken from a Chinese healthcare payer perspective, including three new oral anticoagulants (edoxaban, rivaroxaban, and apixaban) and one low molecular weight heparin (enoxa- parin). We conclude that, in Chinese, DOACs can be a reasonable anticoagulant choice for many patients with cancer-associated VTE. Main results The results of the cost-effectiveness analysis are summa- rized in Table 2. During the first 6 months, the cost of the DOACs treatment was $ 654. 65 and $ 1719.31 for the LMWHs treatment. The effectiveness of the DOACs treatment was 0. 40 QALYs; for the LMWHs treatment it was 0. 37 QALYs. The estimated ICER was $ 32,922.16 and in favor of DOACs. In the outcome ana- lysis of the 60-month time frame, the cost of the DOACs strategy was $ 1447. 22 and $ 3374. 70 with LMWHs. The QALYs associated with DOACs was 3. 07; for the LMWHs it was 3. 09 QALYs. The ICER of DOACs com- pared to LMWHs was $ 112,895.50 per QALY. In the subgroup analysis of those patients with gastrointestinal malignancy, the results showed that DOACs were the preferred strategy over LMWHs with low cost ($ 657.85 vs. $1716.56) and high QALYs (0.40 QALYs vs. 0.37 QALYs). Analysis Thrombosis Journal (2021) 19:68 Page 6 of 10 Table 1 Parameters for model input with a cycle length of 1 month (Continued) Parameters Base case Ranges Distribution Source MB 0.65 0.52–0.78 β [43] ICH 0.33 0.26–0.40 β [43] CRNMBa 0.91 0.73–1.00 β [44] DVT Deep vein thrombosis, PE Pulmonary embolism, MB Major bleeding, ICH Intracranial hemorrhage, CRNMB Clinically relevant non-major bleeding, DOACs Direct oral anticoagulants, LMWHs Low molecular weight heparins, offDVT Risk of deep-vein thrombosis while off-treatment, offPE Risk of pulmonary embolism while off-treatment aUpper and lower bounds estimated to vary ±20% of the mean value for these input parameters estimates bUtility associated with DOACs treatment was assumed to be same as base Table 1 Parameters for model input with a cycle length of 1 month (Continued) Results bleeding (MB) events in treatment with DOACs com- pared to Enoxaparin at five years intervals, respectively. Considering the estimated results for PSA, the majority of simulations showed that the treatment with DOACs was more cost-effective than the equivalent duration of LMWHs treatment. Overall, these analyses suggest that the model outcomes were robust. The scatterplot of these repetitions is shown in Fig. 3. bleeding (MB) events in treatment with DOACs com- pared to Enoxaparin at five years intervals, respectively. Considering the estimated results for PSA, the majority of simulations showed that the treatment with DOACs was more cost-effective than the equivalent duration of LMWHs treatment. Overall, these analyses suggest that the model outcomes were robust. The scatterplot of these repetitions is shown in Fig. 3. Sensitivity analysis A wide bar indicates that the associated variable has a large potential effect on the expected value of the model. This variable with the widest bar (potentially the most critical uncertainty) is plotted on the top. DOACs, direct oral anticoagulants; LMWHs, low molecular weight heparins. CRNMB: clinically relevant non-major bleeding; offAC, anticoagulant treatment discontinuation; ICER, incremental cost-effectiveness ratio; QALYs: quality adjusted life-years (QALYs) Fig. 2 One-way sensitivity analyses (Tornado diagram) over 6-month (left) and 5-year time horizon (right). In the graph, a horizontal bar is generated for each variable being analyzed. Incremental cost is displayed on the horizontal axis, so each bar represents the selected node’s range of incremental values generated by varying the related variable. A wide bar indicates that the associated variable has a large potential effect on the expected value of the model. This variable with the widest bar (potentially the most critical uncertainty) is plotted on the top. DOACs, direct oral anticoagulants; LMWHs, low molecular weight heparins. CRNMB: clinically relevant non-major bleeding; offAC, anticoagulant treatment discontinuation; ICER, incremental cost-effectiveness ratio; QALYs: quality adjusted life-years (QALYs) $19398) as well as similar QALYs gained (0. 34 QALYs vs 0. 35 QALYs) for a 6- month time horizon [27]. Other studies for a 6-month time horizon [25], from the Netherlands, found that rivaroxaban was the most cost-effective treatment choice compared with LMWHs. uncertainties most of the conclusions remained the same. OWSA indicated that the major bleeding in treat- ment with DOACs had a high influence on ICER. The potential reason for this is that the more frequent events of MB with DOACs compared with LMWHs. MB events are very burdensome and frequently severely disabling, leading to low QALYs despite it is modest decrease. Re- cent studies found that DOACs may be associated with low MB events for the treatment of cancer-associated VTE in Asian patients than in non-Asian patients [56]. Furthermore, a significant decrease in GI bleeding risk was observed with DOACs [56]. So DOACs can be a more cost-effective treatment compared to LMWHs in Chinese patients with CAT. Further prospective studies are needed to confirm these findings. Moreover, the Our analysis suggests that DOACs is a cost-saving treatment option with only a modest decrease in QALYs as compared to LMWHs over 5 years. The ICER for DOACs vs. Sensitivity analysis In the sensitivity analysis, we assessed the robustness of the model over a 6- month time horizon and a 5-year time horizon. Tornado diagrams illustrating the 10 most influential variables in descending order of influence are depicted in Fig. 2. According to the OWSA, the 6- month analysis showed that the most sensitive parame- ters included the utilities of DOACs and LMWHs. The ICER was particularly sensitive to non-ICH major In our cost-effectiveness analysis of different DOACs vs LMWHs for the treatment of CAT over a 6-month time frame, our results showed that DOACs are cost- effective, which also has been found in subgroup ana- lysis. DOACs remained more effective and less costly than LMWHs under most of the scenarios explored by sensitivity analysis. The one-way sensitivity analysis re- vealed that the utility of DOACs and LMWHs had the greatest impact on the results. The explanation for the differences seen in the sensitivity analysis is the higher cost of new oral anticoagulants and lower costs of enox- aparin and managing VTE events in China in compari- son with those in developed countries [49–55], lead the strategies will approach equipoise in which case differ- ences in patient preference between injection and oral therapy will become the major determinant between strategies. The PSA demonstrated the robustness of the results, as most of the points in the PSA scatter plot were located in the upper right zone. An economic com- parison of edoxaban and LMWHs in the US showed a lower cost of treatment with rivaroxaban ($ 6061 vs Table 2 Cost-effectiveness analyses Costs ΔCosts QALYs ΔQALYs ICER 6-month time horizon DOACs 654.65 −1064.66 0.40 0.03 −32,922.16 LMWHs 1719.31 0.37 5-year time horizon DOACs 1447.22 −1927.48 3.07 −0.02 112,895.50 LMWHs 3374.70 3.09 Subgroup 0.28 DOACs 657.85 −1058.71 0.40 0.03 −32,821.83 LMWHs 1716.56 0.37 DOACs Direct oral anticoagulants, LMWHs Low molecular weight heparins, ICER Incremental cost-effectiveness ratio, QALYs Quality adjusted life-years Page 7 of 10 Wumaier et al. Thrombosis Journal (2021) 19:68 Fig. 2 One-way sensitivity analyses (Tornado diagram) over 6-month (left) and 5-year time horizon (right). In the graph, a horizontal bar is generated for each variable being analyzed. Incremental cost is displayed on the horizontal axis, so each bar represents the selected node’s range of incremental values generated by varying the related variable. Sensitivity analysis effectiveness ratios; VTE: Venous thromboembolic; DVT: Deep vein thrombosis; PE: Pulmonary embolism; RCTs: Randomized controlled trials; rPE: Recurrent pulmonary embolism; rDVT: Recurrent deep vein thrombosis; ICH: Intracranial hemorrhage; MB: Non-ICH major bleeding; CRNMB: Clinically relevant non-major bleeding; TP: Transition probabilities; USD: US dollars; offAC: Anticoagulant treatment discontinuation; offDVT: Risk of deep-vein thrombosis while off-treatment; offPE: Risk of pulmonary embolism while off- treatment; WTP: Willingness to pay; OWSA: One-way sensitivity analysis; PSA: Probabilistic sensitivity analysis; QALYs: Quality adjusted life-years; GDP: Gross domestic product; WHO: World health organization effectiveness ratios; VTE: Venous thromboembolic; DVT: Deep vein thrombosis; PE: Pulmonary embolism; RCTs: Randomized controlled trials; rPE: Recurrent pulmonary embolism; rDVT: Recurrent deep vein thrombosis; ICH: Intracranial hemorrhage; MB: Non-ICH major bleeding; CRNMB: Clinically relevant non-major bleeding; TP: Transition probabilities; USD: US dollars; offAC: Anticoagulant treatment discontinuation; offDVT: Risk of deep-vein thrombosis while off-treatment; offPE: Risk of pulmonary embolism while off- treatment; WTP: Willingness to pay; OWSA: One-way sensitivity analysis; PSA: Probabilistic sensitivity analysis; QALYs: Quality adjusted life-years; GDP: Gross domestic product; WHO: World health organization Several cost-effectiveness analyses have been con- ducted in different countries [24, 25], but none of them evaluated all DOACs simultaneously, with majority of them focusing on single DOACs only. However, only one study from a US payer perspective was found that compared the cost-effectiveness of different DOACs (Edoxaban+Rivaroxaban) compared with LMWHs [26], have demonstrated DOACs were cost-saving options. The other previous study in the Brazilian population showed that edoxaban is a cost-saving alternative to LMWH for the management of CAT with incremental cost and QALYs increases were $ 16,654.27 and 3.2, re- spectively [24]. Acknowledgements Not surprisingly, as with all cost-effectiveness analyses, there are some uncertainties and limitations associated with our analysis. Firstly, given the absence of local data, clinical and safety estimates were derived from different randomized controlled treatment results in multiple countries rather than the Chinese or the Asian popula- tion specifically, and they may not reflect real-world ob- servations. Secondly, data of utilities that are specially aimed at patients with cancer remain scarce. we extrapo- lated the most utility values from the general medical patients to cancer patients with VTE, which may be overestimated. Future studies are needed to directly as- sess the utility of cancer patients. Whereas, the sensitiv- ity analyses suggest that the results were robust and unlikely to be significantly affected by variations in util- ity variables. Thirdly, Not all relevant costs were in- cluded, the current analysis only included direct medical costs, without considering information about indirect and direct non-medical costs, which may underestimate the total treatment cost per patient. Finally, The model uses a Chinese societal perspective, however, the costs of both DOACs and LMWHs will vary depending on which country or specific health system is evaluating the use of these agents, and this could affect the transfer of cost- effectiveness results from one country to another in a healthcare context. g This work is part of a Master thesis of the Master’s Program in Clinical Research, Cancer Center, the First Hospital of Jilin University, Changchun, Jilin Province, China. Additional file 1. . Additional file 1. . Author details 1 1Jilin University, Changchun, China. 2Department of Cancer center, the First Hospital of Jilin University, Changchun, Jilin 130021, China. 1Jilin University, Changchun, China. 2Department of Cancer center, the First Hospital of Jilin University, Changchun, Jilin 130021, China. Received: 23 July 2021 Accepted: 5 September 2021 Received: 23 July 2021 Accepted: 5 September 2021 Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information files. Authors’ contributions JC designed, conducted and wrote the paper; KW analyzed the results and made the figure, and was a major contributor in writing the manuscript; WL and NC performed data extraction and assisted in writing paper. The author(s) read and approved the final manuscript. Conclusion In conclusion, this economic evaluation has shown that DOACs were estimated to be a cost-saving option when compared to LMWHs for the treatment of CAT in Chinese patients, both the 60-month (extrapolated) and the 6-month (data-driven) horizons. We believe the re- sults of this study would be an important addition to in- form the limited data about the economic impact of VTE among cancer patients. Funding g This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. DOACs: Direct oral anticoagulants; CAT: Cancer-associated thrombosis; LMWHs: Low molecular weight heparins; ICERs: Incremental cost- Consent for publication Not applicable. Consent for publication Not applicable. Consent for publication Not applicable. References 1. Cohen A, Katholing A, Rietbrock S, Bamber L, Martinez C. Epidemiology of first and recurrent venous thromboembolism in patients with active cancer. A population-based cohort study. Thromb Haemost. 2017;117(1):57–65. https://doi.org/10.1160/TH15-08-0686. 2. Faiz AS, Khan I, Beckman MG, Bockenstedt P, Heit JA, Kulkarni R, et al. Characteristics and risk factors of Cancer associated venous thromboembolism. Thromb Res. 2015;136(3):535–41. https://doi.org/10.101 6/j.thromres.2015.06.036. 3. Kourlaba G, Relakis J, Mylonas C, Kapaki V, Kontodimas S, Holm MV, et al. The humanistic and economic burden of venous thromboembolism in cancer patients: a systematic review. Blood Coagul Fibrinolysis. 2015;26(1): 13–31. https://doi.org/10.1097/MBC.0000000000000193. Sensitivity analysis LMWHs was $112,895.50, which is far higher than the threshold of US $30,427.74 (three times GDP per capita of China in 2020). Additionally, this uncertainty and variation surround- ing the model inputs were evaluated in our sensitivity analysis and demonstrated that despite these Fig. 3 Probabilistic sensitivity analysis of the cost-effectiveness using Monte Carlo simulation over 6-month (left) and 5-year time horizon (right). The vertical axis represents incremental cost in USD, horizontal axis represents incremental effectiveness in QALYs, blue spots represents 1000 draws of the probabilistic analysis and the the slope of a line intersecting the origin of the plot represents the willingness to pay (WTP) limit. Values on the right side of the WTP line are considered cost-effective Fig. 3 Probabilistic sensitivity analysis of the cost-effectiveness using Monte Carlo simulation over 6-month (left) and 5-year time horizon (right). The vertical axis represents incremental cost in USD, horizontal axis represents incremental effectiveness in QALYs, blue spots represents 1000 draws of the probabilistic analysis and the the slope of a line intersecting the origin of the plot represents the willingness to pay (WTP) limit. Values on the right side of the WTP line are considered cost-effective Fig. 3 Probabilistic sensitivity analysis of the cost-effectiveness using Monte Carlo simulation over 6-month (left) and 5-year time horizon (right). The vertical axis represents incremental cost in USD, horizontal axis represents incremental effectiveness in QALYs, blue spots represents 1000 draws of the probabilistic analysis and the the slope of a line intersecting the origin of the plot represents the willingness to pay (WTP) limit. Values on the right side of the WTP line are considered cost-effective Wumaier et al. Thrombosis Journal (2021) 19:68 Page 8 of 10 Page 8 of 10 favorable pharmacoeconomic profile was robust in the probabilistic sensitivity analysis. Competing interests h h d l h Competing interests The authors declare that they have no competing interests. Supplementary Information h l l The online version contains supplementary material available at https://doi. org/10.1186/s12959-021-00319-1. Declarations Ethics approval and consent to participate Not applicable. Abbreviations O C N Engl J Med. 2018;379(1):95–6. https://doi.org/10.1056/ NEJMc1806646. 14. Lyman GH, Bohlke K, Khorana AA, Kuderer NM, Lee AY, Arcelus JI, et al. Venous thromboembolism prophylaxis and treatment in patients with cancer: american society of clinical oncology clinical practice guideline update 2014. J Clin Oncol. 2015;33(6):654–6. https://doi.org/10.1200/JCO.2 014.59.7351. 34. Young AM, Marshall A, Thirlwall J, Chapman O, Lokare A, Hill C, et al. Comparison of an Oral factor Xa inhibitor with low molecular weight heparin in patients with Cancer with venous thromboembolism: results of a randomized trial (SELECT-D). J Clin Oncol. 2018;36(20):2017–23. https://doi. org/10.1200/JCO.2018.78.8034. 15. Farge D, Bounameaux H, Brenner B, Cajfinger F, Debourdeau P, Khorana AA, et al. International clinical practice guidelines including guidance for direct oral anticoagulants in the treatment and prophylaxis of venous thromboembolism in patients with cancer. Lancet Oncol. 2016;17(10):e452– 66. https://doi.org/10.1016/S1470-2045(16)30369-2. 35. Agnelli G, Becattini C, Meyer G, Muñoz A, Huisman MV, Connors JM, et al. Apixaban for the treatment of venous thromboembolism associated with Cancer. N Engl J Med. 2020;382(17):1599–607. https://doi.org/10.1056/ NEJMoa1915103. 16. Etxeandia-Ikobaltzeta I, Zhang Y, Brundisini F, Florez ID, Wiercioch W, Nieuwlaat R, et al. Patient values and preferences regarding VTE disease: a systematic review to inform American Society of Hematology guidelines. Blood Adv. 2020;4(5):953–68. https://doi.org/10.1182/bloodadvances.201 9000462. 36. McBane RD 2nd, et al. Apixaban and dalteparin in active malignancy- associated venous thromboembolism: the ADAM VTE trial. J Thromb Haemost. 2020;18(2):411–21. https://doi.org/10.1111/jth.14662. 37. Chee C, et al. Predictors of venous thromboembolism recurrence and bleeding among active cancer patients: a population-based cohort study. Blood. 2014;123(25):3972–8. https://doi.org/10.1182/blood-2014-01-549733. 17. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12):1139–51. https://doi.org/10.1056/NEJMoa0905561. 38. Chinese drug price of drug centralized bid procurement. Chinese drug price of drug centralized bid procurement Accessed 19 February 2021 19 February 2021; Available from: Available from: https://db.yaozh.com/ya opinzhongbiao. 18. Granger CB, Alexander JH, McMurray JJV, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365(11):981–92. https://doi.org/10.1056/NEJMoa1107039. 39. Du J, Wu B. New Oral anticoagulants for Thromboprophylaxis in patients with Cancer receiving chemotherapy: an economic evaluation in a Chinese setting. Clin Drug Investig. 2020;40(7):653–63. https://doi.org/10.1007/s402 61-020-00926-2. 19. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883–91. https://doi.org/10.1056/NEJMoa1009638. 20. Abbreviations O C Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis. Blood. 2002;100(10):3484–8. https://doi.org/10.1182/blood-2002-01-0108. 8. Prandoni P, Lensing AWA, Piccioli A, Bernardi E, Simioni P, Girolami B, et al. Recurrent venous thromboembolism and bleeding complications during anticoagulant treatment in patients with cancer and venous thrombosis. Blood. 2002;100(10):3484–8. https://doi.org/10.1182/blood-2002-01-0108. 27. Connell NT, Connors JM. Cost-effectiveness of edoxaban versus dalteparin for the treatment of cancer-associated thrombosis. J Thromb Thrombolysis. 2019;48(3):382–6. https://doi.org/10.1007/s11239-019-01903-z. 9. Di Minno MND, et al. Direct oral anticoagulants for the treatment of acute venous thromboembolism in patients with cancer: a meta-analysis of randomised controlled trials. Eur Respir J. 2017;50(3). https://doi.org/10.11 83/13993003.01097-2017. 28. Howlett J, Benzenine E, Fagnoni P, Quantin C. Are direct oral anticoagulants an economically attractive alternative to low molecular weight heparins in lung cancer associated venous thromboembolism management? J Thromb Thrombolysis. 2020;50(3):642–51. https://doi.org/10.1007/s11239-020-0204 7-1. 10. Marin-Barrera L, Muñoz-Martin AJ, Rios-Herranz E, Garcia-Escobar I, Beato C, Font C, et al. A Case-Control Analysis of the Impact of Venous Thromboembolic Disease on Quality of Life of Patients with Cancer: Quality of Life in Cancer (Qca) Study. Cancers (Basel). 2019;12(1). https://doi.org/1 0.3390/cancers12010075. 29. Al Mukdad M, Al-Badriyeh D, Elewa HF. Cost-effectiveness Evaluations Among the Direct Oral Anticoagulants for the Prevention and Treatment of Venous Thromboembolism: Systematic Review. Clin Appl Thromb Hemost. 2019;25:1076029619849103. 11. Khorana AA, et al. Health care costs associated with venous thromboembolism in selected high-risk ambulatory patients with solid tumors undergoing chemotherapy in the United States. Clinicoecon Outcomes Res. 2013;5:101–8. https://doi.org/10.2147/CEOR.S39964. 30. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019; 69(1):7–34. https://doi.org/10.3322/caac.21551. 12. Connolly GC, Dalal M, Lin J, Khorana AA. Incidence and predictors of venous thromboembolism (VTE) among ambulatory patients with lung cancer. Lung Cancer. 2012;78(3):253–8. https://doi.org/10.1016/j.lungcan.2 012.09.007. 31. Miller KD, Siegel RL, Lin CC, Mariotto AB, Kramer JL, Rowland JH, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin. 2016; 66(4):271–89. https://doi.org/10.3322/caac.21349. 32. Lui FH, Tuan B, Swenson SL, Wong RJ. Ethnic disparities in gastric cancer incidence and survival in the USA: an updated analysis of 1992-2009 SEER data. Dig Dis Sci. 2014;59(12):3027–34. https://doi.org/10.1007/s10620-014-3275-3. 13. Streiff M, Milentijevic D, McCrae KR, Laliberté F, Lejeune D, Lefebvre P, et al. Healthcare resource utilization and costs associated with venous thromboembolism in cancer patients treated with anticoagulants. J Med Econ. 2019;22(11):1134–40. https://doi.org/10.1080/13696998.2019.1620752. 33. Raskob GE, Büller HR, Segers A. Edoxaban for Cancer-associated venous thromboembolism. Abbreviations O C Page 9 of 10 Page 9 of 10 Wumaier et al. Thrombosis Journal (2021) 19:68 Wumaier et al. Thrombosis Journal (2021) 19:68 4. Mulder F, et al. Venous thromboembolism in cancer patients: a population- based cohort study. Blood. 2021;137(14):1959–69. https://doi.org/10.1182/ blood.2020007338. 4. Mulder F, et al. Venous thromboembolism in cancer patients: a population- based cohort study. Blood. 2021;137(14):1959–69. https://doi.org/10.1182/ blood.2020007338. 22. Streiff MB, Abutalib SA, Farge D, Murphy M, Connors JM, Piazza G. Update on guidelines for the Management of Cancer-Associated Thrombosis. Oncologist. 2021;26(1):e24–40. https://doi.org/10.1002/onco.13596. 5. Khorana AA, et al. Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy. J Thromb Haemost. 2007;5(3): 632–4. https://doi.org/10.1111/j.1538-7836.2007.02374.x. 23. Kimpton M, Carrier M. Efficacy and safety of Xa inhibitors for the treatment of cancer-associated venous thromboembolism. Expert Opin Drug Saf. 2019; 18(4):313–20. https://doi.org/10.1080/14740338.2019.1601699. 5. Khorana AA, et al. Thromboembolism is a leading cause of death in cancer patients receiving outpatient chemotherapy. J Thromb Haemost. 2007;5(3): 632–4. https://doi.org/10.1111/j.1538-7836.2007.02374.x. 6. Kuderer NM, Ortel TL, Francis CW. Impact of venous thromboembolism and anticoagulation on cancer and cancer survival. J Clin Oncol. 2009;27(29): 4902–11. https://doi.org/10.1200/JCO.2009.22.4584. 24. Lopes DG, Tamayo A, Schipp B, Siepmann T. Cost-effectiveness of edoxaban vs low-molecular-weight heparin and warfarin for cancer-associated thrombosis in Brazil. Thromb Res. 2020;196:4–10. https://doi.org/10.1016/j. thromres.2020.08.014. 7. Hutten BA, Prins MH, Gent M, Ginsberg J, Tijssen JGP, Büller HR. Incidence of recurrent thromboembolic and bleeding complications among patients with venous thromboembolism in relation to both malignancy and achieved international normalized ratio: a retrospective analysis. J Clin Oncol. 2000;18(17):3078–83. https://doi.org/10.1200/JCO.2 000.18.17.3078. 7. Hutten BA, Prins MH, Gent M, Ginsberg J, Tijssen JGP, Büller HR. Incidence of recurrent thromboembolic and bleeding complications among patients with venous thromboembolism in relation to both malignancy and achieved international normalized ratio: a retrospective analysis. J Clin Oncol. 2000;18(17):3078–83. https://doi.org/10.1200/JCO.2 000.18.17.3078. 25. de Jong LA, van der Velden AWG, Hulst M, Postma MJ. Cost-effectiveness analysis and budget impact of rivaroxaban compared with dalteparin in patients with cancer at risk of recurrent venous thromboembolism. BMJ Open. 2020;10(11):e039057. https://doi.org/10.1136/bmjopen-2020-039057. 26. Li A, Manohar PM, Garcia DA, Lyman GH, Steuten LM. Cost effectiveness analysis of direct oral anticoagulant (DOAC) versus dalteparin for the treatment of cancer associated thrombosis (CAT) in the United States. Thromb Res. 2019;180:37–42. https://doi.org/10.1016/j.thromres.2019.05.012. 8. Prandoni P, Lensing AWA, Piccioli A, Bernardi E, Simioni P, Girolami B, et al. Abbreviations O C Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369(22):2093–104. https://doi.org/10.1056/NEJMoa1310907. 40. Zhou JS, Bao-Jin LI. Cost-effectiveness analysis of endoscope and Omprazole in the treatment of acute hemorrhage of upper digestive tract. Lingnan J Emerg Med. 2013. 21. Key NS, Khorana AA, Kuderer NM, Bohlke K, Lee AYY, Arcelus JI, et al. Venous thromboembolism prophylaxis and treatment in patients with Cancer: ASCO clinical practice guideline update. J Clin Oncol. 2020;38(5): 496–520. https://doi.org/10.1200/JCO.19.01461. 41. Yang L, Wu J. Cost-effectiveness of rivaroxaban compared with enoxaparin plus warfarin for the treatment of hospitalised acute deep vein thrombosis in China. BMJ Open. 2020;10(7):e038433. https://doi.org/10.1136/bmjopen-2 020-038433. Page 10 of 10 Wumaier et al. Thrombosis Journal (2021) 19:68 Wumaier et al. Thrombosis Journal (2021) 19:68 42. Locadia M, Bossuyt PM, Stalmeier PF, Sprangers MA, van Dongen C, Middeldorp S, et al. Treatment of venous thromboembolism with vitamin K antagonists: patients’ health state valuations and treatment preferences. Thromb Haemost. 2004;92(6):1336–41. https://doi.org/10.1160/TH04-02-0075. 42. Locadia M, Bossuyt PM, Stalmeier PF, Sprangers MA, van Dongen C, Middeldorp S, et al. Treatment of venous thromboembolism with vitamin K antagonists: patients’ health state valuations and treatment preferences. Thromb Haemost. 2004;92(6):1336–41. https://doi.org/10.1160/TH04-02-0075. 43. Büller H, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med. 2013;369(15):1406–15. https://doi. org/10.1056/NEJMoa1306638. 43. Büller H, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med. 2013;369(15):1406–15. https://doi. org/10.1056/NEJMoa1306638. 44. Preblick R, et al. Cost-effectiveness of edoxaban for the treatment of venous thromboembolism based on the Hokusai-VTE study. Hosp Pract (1995). 2015;43(5):249–57. 44. Preblick R, et al. Cost-effectiveness of edoxaban for the treatment of venous thromboembolism based on the Hokusai-VTE study. Hosp Pract (1995). 2015;43(5):249–57. 45. Wang X, et al. Guideline for postmarketing Chinese medicine pharmacoeconomic evaluation. Chin J Integr Med. 2015;21(6):473–80. https://doi.org/10.1007/s11655-014-1749-y. 45. Wang X, et al. Guideline for postmarketing Chinese medicine pharmacoeconomic evaluation. Chin J Integr Med. 2015;21(6):473–80. https://doi.org/10.1007/s11655-014-1749-y. 46. Jankovic SM, Milovanovic DR. Cost and utility of a low-molecular-weight heparin and unfractionated heparin for treatment of deep venous thrombosis in a Balkan country: a model analysis. Exp Clin Cardiol. 2006; 11(2):111–6. 46. Jankovic SM, Milovanovic DR. Cost and utility of a low-molecular-weight heparin and unfractionated heparin for treatment of deep venous thrombosis in a Balkan country: a model analysis. Exp Clin Cardiol. 2006; 11(2):111–6. 47. Abbreviations O C Marshall A, Levine M, Hill C, Hale D, Thirlwall J, Wilkie V, et al. Treatment of cancer-associated venous thromboembolism: 12-month outcomes of the placebo versus rivaroxaban randomization of the SELECT-D trial (SELECT-D: 12m). J Thromb Haemost. 2020;18(4):905–15. https://doi.org/10.1111/jth.14 752. 48. Bertram M, et al. Cost-effectiveness thresholds: pros and cons. Bull World Health Organ. 2016;94(12):925–30. https://doi.org/10.2471/BLT.15.164418. 49. Sullivan S, et al. Measuring the outcomes and pharmacoeconomic consequences of venous thromboembolism prophylaxis in major orthopaedic surgery. PharmacoEconomics. 2003;21(7):477–96. https://doi. org/10.2165/00019053-200321070-00003. 50. McCullagh L, Walsh C, Barry M. Value-of-information analysis to reduce decision uncertainty associated with the choice of thromboprophylaxis after total hip replacement in the Irish healthcare setting. PharmacoEconomics. 2012;30(10):941–59. https://doi.org/10.2165/11591510-000000000-00000. 51. Wolowacz S, et al. Economic evaluation of dabigatran etexilate for the prevention of venous thromboembolism after total knee and hip replacement surgery. Clin Ther. 2009;31(1):194–212. https://doi.org/10.1016/j. clinthera.2009.01.001. 52. Botteman M, Caprini J, Stephens JM, Nadipelli V, Bell CF, Pashos CL, et al. Results of an economic model to assess the cost-effectiveness of enoxaparin, a low-molecular-weight heparin, versus warfarin for the prophylaxis of deep vein thrombosis and associated long-term complications in total hip replacement surgery in the United States. Clin Ther. 2002;24(11):1960–86; discussion 1938. https://doi.org/10.1016/s0149-2 918(02)80091-1. 53. Diamantopoulos A, Lees M, Wells P, Forster F, Ananthapavan J, McDonald H. Cost-effectiveness of rivaroxaban versus enoxaparin for the prevention of postsurgical venous thromboembolism in Canada. Thromb Haemost. 2010; 104(4):760–70. https://doi.org/10.1160/TH10-01-0071. 54. Ryttberg L, Diamantopoulos A, Forster F, Lees M, Fraschke A, Björholt I. Cost-effectiveness of rivaroxaban versus heparins for prevention of venous thromboembolism after total hip or knee surgery in Sweden. Expert Rev Pharmacoecon Outcomes Res. 2011;11(5):601–15. https://doi.org/10.1586/ erp.11.65. 55. Gordois A, Posnett J, Borris L, Bossuyt P, Jonsson B, Levy E, et al. The cost- effectiveness of fondaparinux compared with enoxaparin as prophylaxis against thromboembolism following major orthopedic surgery. J Thromb Haemost. 2003;1(10):2167–74. https://doi.org/10.1046/j.1538-7836.2003.003 96.x. 56. Chen DY, Tseng CN, Hsieh MJ, Lan WC, Chuang CK, Pang ST, et al. Comparison between non-vitamin K antagonist Oral anticoagulants and low-molecular-weight heparin in Asian individuals with Cancer-associated venous thromboembolism. JAMA Netw Open. 2021;4(2):e2036304. https:// doi.org/10.1001/jamanetworkopen.2020.36304. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2615601597	https://europepmc.org/articles/pmc5438574?pdf=render	English	null	The use of lipid-lowering therapy and effects of antihyperglycaemic therapy on lipids in subjects with type 2 diabetes with or without cardiovascular disease: a pooled analysis of data from eleven randomized trials with insulin glargine 100 U/mL	Cardiovascular diabetology	2,017	cc-by	6,948	Cardiovascular Diabetology Cardiovascular Diabetology Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 DOI 10.1186/s12933-017-0548-0 © The Author(s) 2017. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. The use of lipid‑lowering therapy and effects of antihyperglycaemic therapy on lipids in subjects with type 2 diabetes with or without cardiovascular disease: a pooled analysis of data from eleven randomized trials with insulin glargine 100 U/mL Markolf Hanefeld1,2, Louise Traylor3, Ling Gao4 and Wolfgang Landgraf5 Abstract Background: Dyslipidaemia is a major contributor to the increased risk of cardiovascular disease (CVD) associated with type 2 diabetes (T2D). This study aimed to characterize the extent of lipid-lowering therapy use and its impact on lipid and glycaemic outcomes in people with T2D uncontrolled on oral agents who were enrolled in insulin glargine 100 units/mL (Gla-100) randomized controlled trials (RCTs). Methods: A post hoc patient-level pooled analysis of eleven RCTs (≥24 weeks’ duration) comparing Gla-100 (±oral antidiabetes drugs [OADs]) with OADs alone in people with T2D was performed. Baseline and Week 24 or study end- point lipid status (low-density lipoprotein cholesterol [LDL-C], high-density lipoprotein cholesterol [HDL-C], non-high- density lipoprotein cholesterol [non-HDL-C] and triglycerides) and indices of glycaemic control (glycosylated hae- moglobin, fasting plasma glucose [FPG]) were examined in patient groups according to treatment received and CVD status. Lipid-lowering therapy was provided at the discretion of physicians at baseline and throughout the studies. Results: Of the 4768 participants included in the analysis, 41% (n = 1940) received lipid-lowering therapy. Only 51% of participants with CVD (1885/3672) were treated with lipid-lowering therapy; these participants had significantly lower levels of LDL-C, HDL-C and non-HDL-C, and higher levels of triglycerides versus patients not treated with lipid- lowering therapy at baseline and study endpoint (P < 0.001 for all). Antihyperglycaemia therapy resulted in decreases in glycosylated haemoglobin (−1.4 to −1.6%) and FPG (−68.9 to −75.3 mg/dL) at Week 24. Furthermore, slight improvements in non-HDL-C (−3.9 to −9.1 mg/dL) and triglyceride levels (−25.8 to −51.2 mg/dL) were observed. Similar changes were seen irrespective of lipid-lowering therapy or CVD status. Conclusions: In a T2D cohort included in Gla-100 clinical studies, many participants with T2D and CVD did not receive lipid-lowering therapy, and for most categories of lipid the levels were outside the optimal range. Even in patients treated with antihyperglycaemic therapy but not lipid-lowering therapy, there were modest improvements in non-HDL-C and triglyceride levels in all participants with T2D and CVD. There is a need for increased implementation of guideline recommendations such as American College of Cardiology/American Heart Association for the manage- ment of dyslipidaemia in patients with T2D. Correspondence: hanefeld@gwtonline‑zks.de 1 Study Center Metabolic Vascular Medicine, GWT-TU Dresden GmbH/ UKD, Medical clinic III, Fiedler Str. Abstract 34, 01307 Dresden, Germany Full list of author information is available at the end of the article Correspondence: hanefeld@gwtonline‑zks.de 1 Study Center Metabolic Vascular Medicine, GWT-TU Dresden GmbH/ UKD, Medical clinic III, Fiedler Str. 34, 01307 Dresden, Germany Full list of author information is available at the end of the article Background recommends an algorithm which takes into account a range of factors when deciding on the need for lipid- lowering therapy. These include age, race, smoking pres- ence of CVD or history of major cardiovascular events and diabetes status, blood pressure, LDL and cholesterol levels [19, 21]. Current recommendations for the general population suggest that statins should be introduced if the 10-year risk of CVD is greater than 7.5% [19]. For patients with diabetes, statin use is recommended for all patients between the ages of 40 and 75 years and with LDL-C 70–189 mg/dL. High intensity statin use is recommended for patients with diabetes and 10-year risk of CVD above 7.5% or with CVD [19, 22]. Reflecting this, recent out- come studies investigating intensified glucose control require control of major risk factors (e.g. lipids, blood pressure) as a precondition. However, despite detailed guidelines, data indicate that the majority of patients at high risk of CVD, including those with diabetes, are fail- ing to attain lipid-goals [15, 23, 24]. Dyslipidaemia is recognized as a major risk factor for car- diovascular disease (CVD), which is a significant cause of morbidity and mortality in type 2 diabetes (T2D) [1–3]. Reducing hyperglycaemia in patients with diabetes has been shown to decrease onset and progression of micro- vascular complications, although the impact on cardio- vascular complications varies depending on individual risk, quality of overall risk factor control and antidiabe- tes drugs used [3–10]. The use of antidiabetes therapies (including insulin and oral antidiabetes drugs) has been shown to affect lipid levels [11–14]. This is often related to a therapy’s interaction with lipid metabolism, which can result in varied effects on cardiovascular complica- tions. In this complex network of interactions, low-den- sity lipoprotein cholesterol (LDL-C) is a key player as a coronary risk factor. Data from the 2005–2008 National Health and Nutrition Examination Survey (NHANES) indicate that more than one-third (33.5%) of adults in the US age ≥20 years—equivalent to 71 million people—have elevated levels of LDL-C [15]. However, less than half of this number (48.1%) receives treatment, and only 33.2% have their cholesterol under control [16]. Increased LDL-C is the most recognized form of dyslipidaemia; however, high levels of triglycerides and low levels of high-density lipoprotein cholesterol (HDL-C) may also be harmful [17], particularly in T2D. Background Although reducing hyperglycaemia has been shown to decrease onset and progression of microvascular com- plications, the impact on CV complications is unclear. The aim of this patient-level pooled analysis was to char- acterize the extent of lipid-lowering therapy use and the degree of cholesterol control in patients with T2D with and without CVD enrolled in head-to-head randomized trials, together with lipid and glycaemic outcomes follow- ing initiation of antihyperglycaemia therapy. Diabetes is commonly associated with a phenotype of mixed dyslipidaemia. This is characterized by low HDL-C and high triglyceride levels, with often normal or mod- estly elevated LDL-C levels [1]. However, normal LDL-C levels in diabetes may be misleading, as there is generally an increase in the number of small, dense atherogenic LDL and cholesterol-enriched remnant particles [1, 18]. Therefore, in patients with diabetes, non-HDL choles- terol may be a stronger predictor of CVD than LDL cho- lesterol or triglycerides, as levels correlate highly with atherogenic lipoproteins [18]. Page 2 of 9 Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Keywords: Type 2 diabetes, Cardiovascular disease, Lipid control, Statins, Clinical trials Statistical analysesh The pooled analysis was performed by lipid-lowering therapy categories (treatment vs. no treatment) and CVD groups based on standardized patient-level data generated from the identified studies. Demographic and baseline clinical characteristics of age, gender, weight, body mass index (BMI), duration of diabetes, HbA1c and FPG were summarized by lipid-lowering therapy cat- egories and CVD groups. They were also examined for lipid-lowering therapy categories and CVD groups using t-tests for continual variables and Chi square test for gender. HbA1c and FPG at baseline and Week 24 were reported descriptively and compared between lipid-low- ering therapy categories and CVD groups using analysis of covariance (ANCOVA) models. These included age, duration of diabetes, baseline BMI, HbA1c and FPG as covariates. They also used gender, lipid-lowering ther- apy category (treatment vs. no treatment), CVD group, study and randomized treatment arm for type of insulin or oral antihyperglycaemia drug group as fixed factors. Lipid variables at baseline and Week 24 were reported descriptively and compared between lipid-lowering therapy categories and CVD groups using ANCOVA models. These included age, duration of diabetes, base- line BMI, HbA1c, FPG and the corresponding baseline lipid value as covariates. They also used gender, lipid- lowering therapy category (treatment vs. no treatment), CVD group, study, and randomized treatment arm for type of insulin or oral antihyperglycaemia drug group as factors. The lipid data were log transformed to be more normal prior to the model estimation; adjusted means have been antilogged and presented in the original scale as the geometric means. Lipid-lowering therapy included statins (n = 1751; 88%), fibrates (n = 218; 11%), and other agents (n = 185; 10%). The participants with CVD who were treated with lipid-lowering therapy group (n = 1885) consisted of: 51% (n = 964) receiving Gla-100 as glucose-lowering therapy, 16% (n = 299) receiving other insulins (insulin detemir or premixed insulin), 14% (n = 263) receiving NPH insulin, and 19% (n = 359) receiving oral antihy- perglycaemia drugs only. The participants with CVD at baseline who were not treated with lipid-lowering ther- apy group (n = 1787) consisted of: 53% (n = 950) receiv- ing Gla-100, 10% (n = 176) receiving other insulins, 17% (n = 303) receiving NPH insulin, and 20% (n = 358) receiving oral antihyperglycaemia drugs. Patient characteristics In total, data from 4768 participants were analysed, of whom 41% (n = 1940) received lipid-lowering therapy. Of the total number of participants, 40% (n = 1885) had diagnosed CVD at baseline and received lipid-lowering therapy at the discretion of their physicians through- out the 6-month study duration. Also, of the total num- ber of participants 37% (n = 1787) had CVD and did not receive any lipid-lowering therapy during the study period. Furthermore, 22% (n = 1041) of the total num- ber of participants did not have CVD and did not receive lipid-lowering therapy during the study period (Table 2). Overall, only 51% (1885/3672) of the total number of study participants with CVD were treated with lipid-low- ering therapy. Key baseline characteristics by lipid-lower- ing treatment status are presented in Table 2. There were significant differences between treated and untreated patients across all characteristics examined (Table 2). Compared with untreated patients, treated patients were older, more likely to be male, and had a greater body weight, longer duration of diabetes and higher fast- ing C-peptide level (P < 0.001 for all). Compared with untreated patients without CVD, untreated patients with CVD were more likely to be female, have a longer duration of diabetes, weigh more and have a higher BMI (P < 0.01 for all). The following parameters were analysed in participants with both baseline and endpoint values available: • • Glycosylated haemoglobin (HbA1c) and FPG. – – Assessed at baseline and endpoint/Week 24. • • Lipid parameters, including low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein choles- terol (HDL-C), non-high-density lipoprotein choles- terol (non-HDL-C) and triglycerides. – – Assessed at baseline and endpoint/Week 24. Eligible studies A total of 11 studies conducted from 1999 to 2008 ful- filled the inclusion criteria (Table 1). Study and patient selection Eligible studies were randomized controlled trials con- ducted using insulin glargine 100 units/mL (Gla-100). Trials investigated Gla-100 used alone or in combination with other agents versus comparators, or compared treat- ment initiation support methods for Gla-100. Therapies were titrated to a fasting plasma glucose (FPG) target of ≤100 mg/dL (≤5.6 mmol/L) for duration of ≥24 weeks. Lipid measurements were available for all included trials. Lipid-lowering therapy was provided at the discretion of participating physicians at baseline and throughout the duration of the studies. Baseline and Week 24/end- point lipid status and indices of glycaemic control were examined in the various patient groups according to lipid-lowering treatment received and presence of risk factors. Five main subpopulations were investigated: all participants who received lipid-lowering therapy; all participants who did not receive lipid-lowering therapy; participants diagnosed with CVD at baseline and who Until recently, the majority of guidelines for choles- terol treatment and CVD prevention have focused on targets for lipids, broadly set at: <100 mg/dL for LDL-C; <150 mg/dL for triglycerides; and >50 mg/dL for HDL-C [17]. However, as in the majority of branches of medicine, up-to-date guidelines focus on a personalized approach rather than a blanket recommendation for all patients. Current American College of Cardiology/American Heart Association (ACC/AHA) guidelines recommend the use of a risk-assessment algorithm to calculate 10-year risk of CVD [19, 20]. The cholesterol-treatment guideline Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Page 3 of 9 Results Eligible studies received lipid-lowering therapy; participants with CVD who did not receive any lipid-lowering therapy; partici- pants without CVD and who did not receive lipid-lower- ing therapy. Statistical analysesh The participants without CVD at baseline who were not treated with lipid- lowering therapy group (n = 1041) consisted of: 54% (n = 564) receiving Gla-100, 7% (n = 69) receiving other insulins, 24% (n = 248) receiving NPH insulin, and 15% (n = 160) receiving oral antihyperglycaemia drugs. Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Page 4 of 9 Table 1 Summary of included studies DET insulin detemir, GLIM glimepiride, LIS insulin lispro, MET metformin, OAD oral antidiabetes drug, PIO pioglitazone, ROS rosiglitazone, SITA sitagliptin, SU sulfonylurea TZD thiazolidinedione Study Phase Treatment Number of subjects randomized/ treated Treatment period, weeks Insulin titration schedule EASIE [35] 3b/4 Gla-100 + MET vs. SITA + MET 515/501 24 Twice weekly 4020 [36] 3b Gla-100 + SU or MET vs. PIO + SU or MET 389/382 24 extended to 48 Weekly 4022 [37] 3b Gla-100 + SU or MET vs. TZD + SU + MET 337/334 24 extended to 48 Weekly L2T3 [38] 4 Gla-100 + OADs vs. DET + OADs 973/964 24 Every 2 days IN-SIGHT [39] 3b Gla-100 + current OADs vs. current OADs 405/400 24 Daily 4001 [40] 3b Morning vs. bedtime Gla-100 + morning GLIM vs. NPH insulin bedtime + morning GLIM 700/697 24 Weekly 4013 [41] 3b Gla-100 bedtime + morning GLIM vs. NPH insulin bedtime + morning GLIM 528/481 24 Weekly 4002 [42] 3b Gla-100 bedtime + OADs vs. NPH insulin bed- time + OADs 764/756 24 Weekly 4014 [43] 4 Gla-100 + SU + MET vs. ROS + SU + MET 219/217 24 Weekly 4021 [44] 3b Gla-100 + SU + MET vs. LIS 75/25 + SU + MET 212/212 24 Weekly 4041 [45] 4 Gla-100 with group education + OADs vs. Statistical analysesh Gla-100 with individual education + OADs 121/121 24 Self-titration, then investigator reviewed at each visit Table 1 Summary of included studies Table 2 Patient demographics and baseline characteristics Data presented represent mean (standard deviation) unless otherwise specified HbA1c glycosylated haemoglobin, BMI body mass index, CVD cardiovascular disease, T2D type 2 diabetes, FPG fasting plasma glucose Statistically significant differences (P < 0.01) between with-CVD and without-CVD groups Characteristic Lipid-lowering drug treatment No lipid-lowering drug treatment P value Total (n = 1940) With CVD (n = 1885) Total (n = 2828) With CVD (n = 1787) Without CVD (n = 1041) Treated vs non- treated (total populations) Male, % 58 58 50 48* 53 <0.001 Age, years 57.9 (9.0) 57.9 (9.0) 55.7 (10.1) 57.1* (9.7) 53.1 (10.4) <0.001 Weight, kg 89.4 (19.1) 89.6 (19.1) 86.5 (21.2) 88.5* (20.3) 83.3 (22.2) <0.001 BMI, kg/m2 31.4 (5.4) 31.5 (5.4) 30.9 (6.1) 31.7* (5.8) 29.6 (6.5) 0.004 T2D duration, years 9.1 (6.3) 9.1 (6.3) 8.5 (5.8) 8.7* (5.9) 8.0 (5.5) <0.001 HbA1c, % 8.75 (1.03) 8.75 (1.03) 8.89 (1.08) 8.84* (1.06) 8.98 (1.11) <0.001 FPG, mg/dL 196 (56) 196 (55) 200 (57) 200 (57) 201 (58) 0.01 FPG, mmol/L 10.9 (3.1) 10.9 (3.1) 11.1 (3.2) 11.1 (3.2) 11.2 (3.2) 0.01 Fasting C-peptide, nmol/L 1.18 (0.59) 1.19 (0.60) 1.10 (0.60) 1.15 (0.59) 1.01 (0.60) <0.001 Table 2 Patient demographics and baseline characteristics Data presented represent mean (standard deviation) unless otherwise specified HbA1c glycosylated haemoglobin, BMI body mass index, CVD cardiovascular disease, T2D type 2 diabetes, FPG fasting plasma glucose * Statistically significant differences (P < 0.01) between with-CVD and without-CVD groups reduced from baseline to Week 24 similarly across all three patient subgroups (Fig. 1a). Participants who received lipid-lowering therapy had a lower baseline FPG than those who did not (196 mg/ dL vs. 200 mg/dL, respectively; P = 0.01). There was no statistical difference in adjusted endpoint FPG (128.8 mg/ dL vs. 127.6 mg/dL; P = 0.2222) or change from baseline to adjusted endpoint FPG (−69.7 mg/dL vs. −70.9 mg/ reduced from baseline to Week 24 similarly across all three patient subgroups (Fig. 1a). Lipid status At baseline, participants treated with lipid-lowering ther- apy (compared with untreated patients) had lower lev- els of LDL-C (99.8 mg/dL vs. 119.2 mg/dL, respectively; P < 0.001), non-HDL-C (142.6 mg/dL vs. 156.6 mg/ dL; P < 0.001) and HDL-C (43.9 mg/dL vs. 45.5 mg/dL; P < 0.001), and higher levels of triglycerides (240.6 mg/dL vs. 200.3 mg/dL; P < 0.001). Results were similar at Week 24 for LDL-C (98.5 mg/dL vs. 120.1 mg/dL; P < 0.001), non-HDL-C (133.4 mg/dL vs. 151.6 mg/dL; P < 0.001), HDL-C (44.2 mg/dL vs. 46.0 mg/dL; P < 0.036), and tri- glycerides (189.2 mg/dL vs. 166.0 mg/dL; P = 0.1223). In the groups stratified by CVD status and lipid-lower- ing therapy, non-HDL-C and triglyceride levels slightly improved, while LDL-C and HDL-C levels remained almost unchanged following therapy with antihypergly- caemia drugs, irrespective of receipt of lipid-lowering therapy for these groups (Fig. 2a–d). There were significant differences in baseline character- istics between treated and untreated participants in this analysis. These were likely driven by the higher number of patients with CVD in the treated participants group, given that CVD is more common in males, increases with duration of diabetes, and is associated with elevated serum C-peptide [25, 26]. We also found significant dif- ferences in baseline characteristics between untreated patients with and without CVD. The higher proportion of females in the untreated CVD group may be driven by under-recognition of CVD in women leading to later and less aggressive treatment. This may also have contributed to the fact that patients in this group were older and had a longer duration of diabetes [27, 28]. Fig. 1 HbA1c (a) and FPG (b) at baseline and Week 24 ±CVD or lipid-lowering therapy. Data presented represent mean (standard deviation). P < 0.05 between patients treated with lipid-lowering therapy versus not treated for HbA1c at baseline and change to Week 24, differences were not significant in FPG Fig. 1 HbA1c (a) and FPG (b) at baseline and Week 24 ±CVD or lipid-lowering therapy. Data presented represent mean (standard deviation). P < 0.05 between patients treated with lipid-lowering therapy versus not treated for HbA1c at baseline and change to Week 24, differences were not significant in FPG Although there were significant differences in baseline glycaemic control for participants who were treated with lipid-lowering therapy versus those who were not, these differences were numerically small (0.14%), and their clinical relevance is questionable. Discussion In this pooled analysis of head-to-head randomized tri- als of Gla-100 in T2D, approximately 41% of all partici- pants were treated with lipid-lowering therapy. Of those patients with CVD at baseline, only 51% of participants received lipid-lowering therapy. Following initiation of antihyperglycaemia treatment, HbA1c was reduced from baseline to study endpoint in participants with CVD— both with and without lipid-lowering therapy—and in untreated participants without CVD. Those participants receiving lipid-lowering therapy had a lower baseline HbA1c, and achieved smaller reductions in HbA1c from baseline to endpoint. Baseline FPG was lower in lipid- lowering therapy-treated participants and reduced in all CVD/treatment groups examined from baseline to end- point. In terms of lipid profiles, all forms of cholesterol were lower in participants treated with lipid-lowering therapy at baseline, while triglyceride levels were higher; this pattern persisted to Week 24. There were reduc- tions in triglyceride levels from baseline to Week 24 in all CVD/lipid-lowering treatment groups, as well as slight reductions in non-HDL-C; LDL-C and HDL-C levels were not substantially changed during therapy.hif Glycaemic control Overall, participants who received lipid-lowering ther- apy had a lower baseline HbA1c than those who did not (8.75% vs. 8.89%, respectively; P < 0.001). However, they had a higher adjusted HbA1c at endpoint (7.34% vs. 7.22%; P < 0.0001) and a smaller change from baseline to endpoint (−1.50% vs. −1.61%; P < 0.0001). In terms of CVD status and lipid-lowering therapy, HbA1c was Participants who received lipid-lowering therapy had a lower baseline FPG than those who did not (196 mg/ dL vs. 200 mg/dL, respectively; P = 0.01). There was no statistical difference in adjusted endpoint FPG (128.8 mg/ dL vs. 127.6 mg/dL; P = 0.2222) or change from baseline to adjusted endpoint FPG (−69.7 mg/dL vs. −70.9 mg/ Page 5 of 9 Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 dL; P = 0.2222). When stratified by CVD status and lipid-lowering therapy, FPG was reduced across the three examined groups from baseline to Week 24 (Fig. 1b). Reductions were marginally greater in participants who did not receive lipid-lowering therapy, regardless of their CVD status. Lipid status The slight advantages in glycaemic control at baseline in those participants treated with lipid-lowering therapy may be indicative of a more stringent prior treatment regimen or greater patient engagement with healthcare goals prior to trial entry. This may also account for the use of lipid-lowering ther- apy in these participants. Overall, all three CVD/treat- ment groups showed a reduction in HbA1c of around 1.5%. The relevance of the difference in HbA1c change between treated and untreated participants, although sig- nificant, is slight (0.11%) and unlikely to be clinically rel- evant. Reductions in HbA1c can be considered a positive Fig. 1 HbA1c (a) and FPG (b) at baseline and Week 24 ±CVD or lipid-lowering therapy. Data presented represent mean (standard deviation). P < 0.05 between patients treated with lipid-lowering therapy versus not treated for HbA1c at baseline and change to Week 24, differences were not significant in FPG Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Page 6 of 9 Fig. 2 Lipid status at baseline and Week 24 ±CVD or lipid-lowering therapy: LDL-C (a), Non-HDL-C (b), HDL-C (c), and triglycerides (d). Data pre- sented represent mean (standard deviation). P < 0.05 between patients treated with lipid-lowering therapy versus not treated for all lipid param- eters at baseline and change to Week 24, except for triglycerides which were significant (P < 0.05) at baseline only Fig. 2 Lipid status at baseline and Week 24 ±CVD or lipid-lowering therapy: LDL-C (a), Non-HDL-C (b), HDL-C (c), and triglycerides (d). Data pre- sented represent mean (standard deviation). P < 0.05 between patients treated with lipid-lowering therapy versus not treated for all lipid param- eters at baseline and change to Week 24, except for triglycerides which were significant (P < 0.05) at baseline only outcome in relation to CVD, as elevated HbA1c has been shown to have a strong correlation with increased CVD risk in patients with diabetes [29, 30].h treatment were above this optimal goal regardless of CVD status [17]. Elevated triglycerides may be overlooked in patients with well controlled LDL-C [1, 31], and baseline triglycerides were high (>200 mg/dL) in both treated and untreated participants in our study [17]. Following ther- apy with antihyperglycaemia agents, triglyceride levels fell to within the ‘borderline high’ category (150–199 mg/ dL) for both treated and untreated participants. Lipid status This is in keeping with previous reports of reductions in triglycer- ides being associated with antihyperglycaemia treatment, and with the use of insulin therapy in cases of severely elevated triglycerides [11–14]. The fact that this has been seen with treatments other than insulin suggests that this is the result of reduced glycaemia rather than being a direct effect of the treatment. There is a strong evidence base for statin therapy in all patients with diabetes and between the ages of 40 and 75 years [19, 22]. However in our population, which had a median age of 56.5 years, less than half of all par- ticipants were receiving lipid-lowering therapy. Of notice, AHA and ESC guidelines recommend that all patients with clinical CVD receive lipid-lowering therapy, regard- less of their diabetes status [19], with current guidelines indicating the need for high intensity statin therapy for all patients with CVD and T2D [22]. Despite this, only 51% of participants with CVD and diabetes in our study received lipid-lowering therapy. f As noted previously, mixed dyslipidaemia (high triglyc- erides/low HDL-C) is common in patients with diabetes [1]. This pattern was observed in our study population, with HDL-C levels being consistently below the optimal level of 50 mg/dL [17]. In high-risk individuals with con- trolled LDL-C but high triglyceride levels, a non-HDL- C target of <130 mg/dL has been recommended [18], although guidelines indicate that there is insufficient evi- dence from randomized controlled trials to determine Reducing LDL-C is the major focus of CVD prevention. Under current ACC/AHA guidelines, which concentrate on statin treatment based on a CVD risk-assessment algorithm, there is no longer a LDL-C goal [19, 22]. How- ever, previous guidelines (1998–2008) set <100 mg/dL as the optimal LDL-C level in high-risk groups, indicating that participants treated with lipid-lowering therapy in our study had, according to current guidelines, optimal LDL-C at baseline and Week 24, while those without Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Page 7 of 9 either LDL-C or non-HDL-C treatment targets [19]. In our study, mean non-HDL-C was above the <130 mg/ dL target in both treatment categories at baseline and at Week 24, despite improvements from baseline to Week 24 being observed. data suggest a need for a greater awareness of the risks of CVD in patients with diabetes and for more widespread implementation of guideline recommendations for the management of dyslipidaemia in these individuals. Funding This study was funded by Sanofi. Ethics approval and consent to participate Individual randomized trials were approved by local ethic committees at the time of conduction and were conducted in accordance with the declaration of Helsinki. In conclusion, this post hoc pooled analysis of ran- domized controlled trials of people with T2D dem- onstrates a modest improvement in non-HDL-C and triglyceride levels following antihyperglycaemia therapy in study participants with both T2D and CVD. Despite guideline recommendations, many participants with T2D and CVD did not receive lipid-lowering therapy. In addition, participants had lipid levels outside the opti- mal range for most categories of lipid control. These Lipid status Overall, the studies included in this analysis covered patients from a large range of ethnicities and geographi- cal locations including centres in North America, South America, Europe, Africa, Asia and Australia. However, as is often the case with clinical trials, there was an apparent bias towards Caucasian patients and economically devel- oped Western countries, Although it was beyond the scope of this study, a sub analysis by location and ethnic- ity would be of interest, although it is likely that numbers of patients in each subcategory may be too low to obtain statistically significant results. Furthermore, how closely these correspond to real-world outcomes is unclear. Previous studies suggest quite large differences between countries with regards to utilization of various treat- ments, such as lipid-lowering therapies [32, 33]. Higher income countries will tend to have higher coverage of screening and treatment, therefore reducing the burden of disease. Access to disease management programmes has also been shown to help patients achieve greater LDL-C reductions and control rates, which again is likely to result in differences between countries with regards to real-world outcomes [34].h Authors’ contributions MH interpreted the data, contributed to drafting the manuscript. LT provided statistical analysis, interpreted the data, and contributed to drafting the manu- script. LG provided statistical analysis, interpreted the data, and contributed to drafting the manuscript. MV interpreted the data, contributed to drafting the manuscript. WL developed the concept for the analysis, interpreted the data, contributed to drafting the manuscript. All authors read and approved the final manuscript. Competing interests MH speaker and advisory board honoraria from Sanofi. LT employee of Sanofi US, Inc., and a Sanofi shareholder. LG consultant for Sanofi US, Inc. MV employee of Sanofi, France, and a Sanofi shareholder. WL employee of Sanofi, Germany, and a Sanofi shareholder. Acknowledgements The main strengths of our study are that the data included are patient-level data, and that the popula- tion is derived from prospective randomized trials with defined and consistent titration regimens and treatment targets. The pooling of patient data increases statistical power and reduces variability. The main limitation of our study is the age of the data (the most recent of the stud- ies included was conducted in 2008); in addition, statin intensity data was not available. However, viewing this data alongside current recommendations highlights the need for better lipid control in the T2D population [17, 19, 22]. The authors acknowledge the contribution of Maya Vincent to the content of the manuscript. The authors received writing/editorial support in the prepara- tion of this manuscript provided by Michael van der Veer, PhD, of Excerpta Medica, funded by Sanofi. Author details 1 1 Study Center Metabolic Vascular Medicine, GWT-TU Dresden GmbH/UKD, Medical clinic III, Fiedler Str. 34, 01307 Dresden, Germany. 2 University Hospital Carl Gustav Carus, Dresden, Germany. 3 Sanofi Us Inc., 55 Corporate Dr, Bridgewater, NJ 08807, USA. 4 Analysta Inc., Somerset, NJ, USA. 5 Sanofi, K703, Industriepark Höchst, 65926 Frankfurt Am Main, Germany. Availability of data and materialsi The data that support the findings of this study are available from Sanofi but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are how- ever available from the authors upon reasonable request and with permission of Sanofi. We observed no interaction (or only minor interaction) between glucose-lowering treatment and LDL-C/HDL-C levels. However, there was a tendency of improved effect of lipid-lowering therapy in patients with CVD after 24 weeks of treatment for hyperglycaemia. Abbreviations ACC/AHA: American College of Cardiology/American Heart Association; ANCOVA: Analysis of covariance; HbA1c: Glycosylated haemoglobin; BMI: Body mass index; CVD: cardiovascular disease; DET: insulin detemir; FPG: fasting plasma glucose; GLIM: Glimepiride; HDL-C: high-density lipoprotein cholesterol; Gla-100: insulin glargine 100 units/mL; LIS: insulin lispro; LDL-C: low density lipoprotein cholesterol; MET: Metformin; NHANES: National Health and Nutrition Examination Survey; Non-HDL-C: non-high-density lipoprotein cholesterol; OAD: oral antihyperglycaemia drug; PIO: Pioglitazone; ROSR: rosiglitazone; SITA: Sitagliptin; SU: sulfonylurea; TZD: thiazolidinedione; T2D: Type 2 diabetes; US: United States. References 1. Miller M. Dyslipidemia and cardiovascular risk: the importance of early prevention. QJM. 2009;102:657–67. 1. Miller M. Dyslipidemia and cardiovascular risk: the importance of early prevention. QJM. 2009;102:657–67. 23. Jones PH, Nair R, Thakker KM. Prevalence of dyslipidemia and lipid goal attainment in statin-treated subjects from 3 data sources: a retrospective analysis. J Am Heart Assoc. 2012;1:e001800. 2. Nelson RH. Hyperlipidemia as a risk factor for cardiovascular disease. Prim Care. 2013;40:195–211. 2. Nelson RH. Hyperlipidemia as a risk factor for cardiovascular disease. Prim Care. 2013;40:195–211. 24. Gyberg V, De Bacquer D, De Backer G, Jennings C, Kotseva K, Mellbin L, et al. Patients with coronary artery disease and diabetes need improved management: a report from the EUROASPIRE IV survey: a registry from the Euro Observational Research Programme of the European Society of Cardiology. Cardiovasc Diabetol. 2015;14:133. 3. Dailey G, Wang E. A review of cardiovascular outcomes in the treatment of people with type 2 diabetes. Diabetes Ther. 2014;5:385–402. 3. Dailey G, Wang E. A review of cardiovascular outcomes in the treatment of people with type 2 diabetes. Diabetes Ther. 2014;5:385–402. 4. Gaede P, Vedel P, Larsen N, Jensen GV, Parving HH, Pedersen O. Multi- factorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med. 2003;348:383–93. 4. Gaede P, Vedel P, Larsen N, Jensen GV, Parving HH, Pedersen O. Multi- factorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med. 2003;348:383–93. gy 25. Towfighi A, Zheng L, Ovbiagele B. Sex-specific trends in midlife coronary heart disease risk and prevalence. Arch Intern Med. 2009;169:1762–6. 5. Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, et al. Association of glycaemia with macrovascular and microvascular compli- cations of type 2 diabetes (UKPDS 35): prospective observational study. BMJ. 2000;321:405–12. 26. Fox CS, Sullivan L, D’Agostino RB Sr, Wilson PW. Framingham heart study. The significant effect of diabetes duration on coronary heart disease mortality: the Framingham heart study. Diabetes Care. 2004;27:704–8. 6. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treat- ment and risk of complications in patients with type 2 diabetes (UKPDS 33. Lancet. 1998;352:837–53. 27. Maas AHEM, Appelman YEA. Gender differences in coronary heart dis- ease. Neth Heart J. 2010;18:598–602. 28. Daly CA, Clemens F, Sendon JL, Tavazzi L, Boersma E, Danchin N, et al. References The clinical characteristics and investigations planned in patients with stable angina presenting to cardiologists in Europe: from the Euro Heart Survey of Stable Angina. Eur Heart J. 2005;26:996–1010. 7. Gerstein HC, Miller ME, Byington RP, Goff DC Jr, Bigger JT, Buse JB, et al. Action to control cardiovascular risk in diabetes study group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545–59. g 29. Borg R, Kuenen JC, Carstensen B, Zheng H, Nathan DM, Heine RJ, et al. HbA1c and mean blood glucose show stronger associations with car- diovascular disease risk factors than do postprandial glycemia or glucose variability in persons with diabetes: the A1C-Derived Average Glucose (ADAG) study. Diabetologia. 2011;54(1):69–72. 8. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient- centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38:140–9. 30. Eeg-Olofsson K, Cederholm J, Nilsson PM, Zethelius B, Svensson AM, Gud- björnsdóttir S, et al. New aspects of HbA1c as a risk factor for cardiovascu- lar diseases in type 2 diabetes: an observational study from the Swedish National Diabetes Register (NDR). J Intern Med. 2010;268:471–82. 9. Zinman B, Lachin JM, Inzucchi SE. Empagliflozin, cardiovascular out- comes, and mortality in type 2 diabetes. N Engl J Med. 2015;373:2117–28. y yp g 10. Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, Nauck 10. Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, Nauck MA, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375:311–22. MA, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375:311–22. 31. Jacobson TA, Case CC, Roberts S, Buckley A, Murtaugh KM, Sung JC, et al. Characteristics of US adults with the metabolic syndrome and therapeu- tic implications. Diabetes Obes Metab. 2004;6:353–62. 11. Chiquette E, Ramirez G, DeFronzo R. A meta-analysis comparing the effect of thiazolidinediones on cardiovascular risk factors. Arch Intern Med. 2004;164:2097–104. 32. Gitt AK, Sonntag F, Jannowitz C, Weizel A, Karmann B, Schaefer JR, Pittrow D, Hildemann SK. Better lipid target achievement for secondary preven- tion through disease management programs for diabetes mellitus and coronary heart disease in clinical practice in Germany. Curr Med Res Opin. 2016;32(3):417–26. 12. References Hanefeld M, Cagatay M, Petrowitsch T, Neuser D, Petzinna D, Rupp M. Acarbose reduces the risk for myocardial infarction in type 2 dia- betic patients: meta-analysis of seven long-term studies. Eur Heart J. 2004;25:10–6. 33. Wong HK, Ong KL, Cheung CL, Cheung BM. Utilization of glucose, blood pressure, and lipid lowering medications among people with type II dia- betes in the United States, 1999–2010. Ann Epidemiol. 2014;24(7):516–21. 13. Bianchi C, Miccoli R, Daniele G, Penno G, Del Prato S. Is there evidence that oral hypoglycemic agents reduce cardiovascular morbidity/mortal- ity? Yes. Diabetes Care. 2009;32(Suppl 2):S342–8. 34. Wang T, Xu Y, Xu M, Wang W, Bi Y, Lu J, Dai M, Zhang D, Ding L, Xu B, Sun J, Zhao W, Jiang Y, Wang L, Li Y, Zhang M, Lai S, Wang L, Ning G. Awareness, treatment and control of cardiometabolic disorders in Chinese adults with diabetes: a national representative population study. Cardiovasc Diabetol. 2015;26(14):28. 14. Poonuru S, Pathak SR, Vats HS, Pathak RD. Rapid reduction of severely elevated serum triglycerides with insulin infusion, Gemfibrozil and Niacin. Clin Med Res. 2011;9:38–41. 15. CDC. Vital signs: prevalence, treatment, and control of high levels of low- density lipoprotein cholesterol. United States, 1999–2002 and 2005–2008. MMWR. 2011;60:109–14. 35. Aschner P, Chan J, Owens DR, Picard S, Wang E, Dain MP, et al. Insulin glargine versus sitagliptin in insulin-naive patients with type 2 diabetes mellitus uncontrolled on metformin (EASIE): a multicentre, randomised open-label trial. Lancet. 2012;379:2262–9. 16. Wong ND, Patao C, Wong K, Malik S, Franklin SS, Iloeje U. Trends in control of cardiovascular risk factors among US adults with type 2 diabetes from 1999 to 2010: comparison by prevalent cardiovascular disease status. Diabetes Vasc Dis Res. 2013;10(6):505–13. 36. Meneghini LF, Traylor L, Schwartz SL. Improved glycemic control with insulin glargine versus pioglitazone as add-on therapy to sulfonylurea or metformin in patients with uncontrolled type 2 diabetes mellitus. Endocr Pract. 2010;16:588–99. 17. National Institute of Health (NIH). Detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) executive summary. 2001. http://www.nhlbi.nih.gov/files/docs/guidelines/atp3x- sum.pdf. Accessed 19 Jan 2016. 37. Hollander P, Sugimoto D, Vlajnic A, Kilo C. Combination therapy with insulin glargine plus metformin but not insulin glargine plus sulfonylurea provides similar glycaemic control to triple oral combination therapy in patients with type 2 diabetes uncontrolled with dual oral agent therapy. J Diabetes Complicat. 2015;29:1266–71. 18. Peters AL. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Received: 15 February 2017 Accepted: 5 May 2017 Received: 15 February 2017 Accepted: 5 May 2017 Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Page 8 of 9 22. American Diabetes Association. Standards of medical care in diabe- tes—2016. Diabetes Care. 2016;39(suppl 1):S1–106. Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 References Clinical relevance of non-HDL cholesterol in patients with diabetes. Clin Diabetes. 2008;26:3–7. 19. Stone NJ, Robinson JG, Lichtenstein AH, Bairey Merz CN, Blum CB, Eckel RH, et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults a report of the American College of Cardiology/American Heart Association task force on practice guidelines. Circulation. 2014;129(25 Suppl 2):S1–45. 38. Swinnen SG, Dain MP, Aronson R, Davies M, Gerstein HC, Pfeiffer AF, et al. A 24-week, randomized, treat-to-target trial comparing initiation of insu- lin glargine once-daily with insulin detemir twice-daily in patients with type 2 diabetes inadequately controlled on oral glucose-lowering drugs. Diabetes Care. 2010;33:1176–8. 20. Goff DC Jr, Lloyd-Jones DM, Bennett G, Coady S, D’Agostino RB, Gibbons R, et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association task force on practice guidelines. Circulation. 2014;129(25 Suppl 2):S49–73. 39. Gerstein HF, Yale J-F, Harris SB, Issa M, Stewart J, Dempsey E. A rand- omized trial of adding insulin glargine vs. avoidance of insulin in people with type 2 diabetes on either no oral glucose-lowering agents or submaximal doses of metformin and/or sulphonylureas. The Canadian INSIGHT (implementing new strategies with insulin glargine for hypergly- caemia treatment) study. Diabet Med. 2006;23:736–42. 21. American College of Cardiology/American Heart Association (ACC/AHA). ASCVD Risk Estimator. 2014. http://tools.acc.org/ASCVD-Risk-Estimator/. Accessed 19 Jan 2016. Hanefeld et al. Cardiovasc Diabetol (2017) 16:66 Page 9 of 9 Page 9 of 9 Page 9 of 9 41. Eliaschewitz FG, Calvo C, Valbuena H, Ruiz M, Aschner P, Villena J, et al. Therapy in type 2 diabetes: insulin glargine vs. NPH insulin both in com- bination with glimepiride. Arch Med Res. 2006;37:495–501. 42. Riddle MC, Rosenstock J, Gerich J, Insulin Glargine 4002 Study Investiga- tors. The treat-to-target trial: randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care. 2003;26:3080–6. 43. Rosenstock J, Sugimoto D, Strange P, Stewart JA, Soltes-Rak E, Dailey G. Triple therapy in type 2 diabetes: insulin glargine or rosiglitazone added to combination therapy of sulfonylurea plus metformin in insulin-naive patients. Diabetes Care. 2006;29:554–9. 43. Rosenstock J, Sugimoto D, Strange P, Stewart JA, Soltes-Rak E, Dailey G. Triple therapy in type 2 diabetes: insulin glargine or rosiglitazone added to combination therapy of sulfonylurea plus metformin in insulin-naive patients. Diabetes Care. 2006;29:554–9. 44. HOE901_4021 [data on file]. ClinicalTrials.gov Identifier: NCT01336751. 45. Yki-Järvinen H, Juurinen L, Alvarsson M, Bystedt T, Caldwell I, Davies M, et al. Initiate insulin by aggressive titration and education (INITIATE): a randomized study to compare initiation of insulin combination therapy in type 2 diabetic patients individually and in groups. Diabetes Care. 2007;30:1364–9. 41. Eliaschewitz FG, Calvo C, Valbuena H, Ruiz M, Aschner P, Villena J, et al. Therapy in type 2 diabetes: insulin glargine vs. NPH insulin both in com- bination with glimepiride. Arch Med Res. 2006;37:495–501. 45. Yki-Järvinen H, Juurinen L, Alvarsson M, Bystedt T, Caldwell I, Davies M, et al. Initiate insulin by aggressive titration and education (INITIATE): a randomized study to compare initiation of insulin combination therapy in type 2 diabetic patients individually and in groups. Diabetes Care. 2007;30:1364–9. 40. Fritsche A, Schweitzer MA, Häring HU, 4001 Study Group. Glime- piride combined with morning insulin glargine, bedtime neutral protamine hagedorn insulin, or bedtime insulin glargine in patients with type 2 diabetes. A randomized, controlled trial. Ann Intern Med. 2003;138(12):952–9. 42. Riddle MC, Rosenstock J, Gerich J, Insulin Glargine 4002 Study Investiga- tors. The treat-to-target trial: randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care. 2003;26:3080–6. p 44. HOE901_4021 [data on file]. ClinicalTrials.gov Identifier: NCT01336751. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step:
https://openalex.org/W2898300073	https://europepmc.org/articles/pmc6265913?pdf=render	English	null	Cataract Preventive Role of Isolated Phytoconstituents: Findings from a Decade of Research	Nutrients	2,018	cc-by	32,059	Received: 6 September 2018; Accepted: 15 October 2018; Published: 26 October 2018 Abstract: Cataract is an eye disease with clouding of the eye lens leading to disrupted vision, which often develops slowly and causes blurriness of the eyesight. Although the restoration of the vision in people with cataract is conducted through surgery, the costs and risks remain an issue. Abstract: Cataract is an eye disease with clouding of the eye lens leading to disrupted vision, which often develops slowly and causes blurriness of the eyesight. Although the restoration of the vision in people with cataract is conducted through surgery, the costs and risks remain an issue. Botanical drugs have been evaluated for their potential efﬁcacies in reducing cataract formation decades ago and major active phytoconstituents were isolated from the plant extracts. The aim of this review is to ﬁnd effective phytoconstituents in cataract treatments in vitro, ex vivo, and in vivo. A literature search was synthesized from the databases of Pubmed, Science Direct, Google Scholar, Web of Science, and Scopus using different combinations of keywords. Selection of all manuscripts were based on inclusion and exclusion criteria together with analysis of publication year, plant species, isolated phytoconstituents, and evaluated cataract activities. Scientists have focused their attention not only for anti-cataract activity in vitro, but also in ex vivo and in vivo from the review of active phytoconstituents in medicinal plants. In our present review, we identiﬁed 58 active phytoconstituents with strong anti-cataract effects at in vitro and ex vivo with lack of in vivo studies. Considering the beneﬁts of anti-cataract activities require critical evaluation, more in vivo and clinical trials need to be conducted to increase our understanding on the possible mechanisms of action and the therapeutic effects. Keywords: cataract; phytoconstituents; lens; preclinical models; drug discovery Vuanghao Lim 1,2 , Edward Schneider 2, Hongli Wu 3,4,* and Iok-Hou Pang 3,4,* Vuanghao Lim 1,2 , Edward Schneider 2, Hongli Wu 3,4,* and Iok-Hou Pang 3,4,* 1 Integrative Medicine Cluster, Advanced Medical and Dental Institute, Universiti Sains Malaysia, Bertam, Kepala Batas, Penang 13200, Malaysia; vlim@usm.my p g y y 2 Botanical Research Institute of Texas (BRIT), 1700 University Drive, Fort Worth, TX 76107-3400, USA; eschneider@brit.org 2 Botanical Research Institute of Texas (BRIT), 1700 University Drive, Fort Worth, TX 76107-3400, USA; eschneider@brit.org 2 Botanical Research Institute of Texas (BRIT), 1700 University Drive, Fort Worth, TX 76107-3400, USA; eschneider@brit.org g 3 Department of Pharmaceutical Sciences, System College of Pharmacy, University of North Texas Health Science Center, Fort Worth, TX 76107, USA g 3 Department of Pharmaceutical Sciences, System College of Pharmacy, University of North Texas Health Science Center, Fort Worth, TX 76107, USA 3 Department of Pharmaceutical Sciences, System College of Pharmacy, University of North Texas Healt Science Center, Fort Worth, TX 76107, USA 4 North Texas Eye Research Institute, University of North Texas Health Science Center, Fort Worth, TX 76107, USA 4 North Texas Eye Research Institute, University of North Texas Health Science Center, Fort Worth, TX 76107, USA * Correspondence: Hongli.Wu@unthsc.edu (H.W.); iok-hou.pang@unthsc.edu (I.-H.P.); Tel.: +1-817-735-7617 (H.W.); +1-817-735-2960 (J.-H.P.) nutrients nutrients nutrients nutrients 1. Introduction The ocular lens is located at the anterior segment of the eye that, together with the cornea, provides the refractive power of the eye. The mature lens is composed of a core of primary lens ﬁber cells, layers of secondary lens ﬁber cells, and one layer of anterior lens epithelial cells, which covers the anterior surface of the lens [1]. The major function of the lens is to maintain transparency so that the light can be properly focused on the retina. Unfortunately, the delicate balance required for lens transparency can be easily disturbed by oxidative stress, aging, and UV radiation, and cataracts develop as a result [1]. Cataracts are the most common cause of vision loss in people over the age of 40 and are the leading cause of blindness in the world [2]. Cataracts are deﬁned as lens opaciﬁcation that prevents a Nutrients 2018, 10, 1580; doi:10.3390/nu10111580 www.mdpi.com/journal/nutrients 2 of 41 Nutrients 2018, 10, 1580 sharply deﬁned image from reaching the retina. As a result, cataract patients have clouded, blurred, or dim visions, which signiﬁcantly affect their daily life. According to a report from the World Health Organization, nearly 40 million people are blind worldwide, almost half of them are due to cataract [3]. Although cataract-related vision loss can be corrected by replacement with synthetic lenses, cataract surgery is a costly procedure and may develop complications like infectious endophthalmitis, posterior capsule rupture during surgery, post-operative macular edema, and posterior capsule opacity (also called posterior capsule opaciﬁcation). In developing countries, many cataract patients cannot have their vision restored due to ﬁnancial concerns or lack of medical resources. Therefore, identifying a safe compound that can reduce the incidence or delay the onset of cataract is an important step in ﬁnding new treatments for cataract. There have been many compounds evaluated for their potential efﬁcacies in reducing cataract formation. In this article, we focus on active ingredients derived from plants. Phytoconstituents are a trove of often structurally complicated compounds with interesting biological functions. They themselves or their derivatives have always been important sources of pharmacologically active agents. To provide a comprehensive review of potentially useful anti-cataract phytoconstituents, we searched, selected, and extracted the appropriate information from published literature according to the following procedures. 1. Introduction We feel that, by listing the comprehensive collection of phytocontituents in one place, this manuscript serves as an overview and perhaps an inspiration to prompt additional studies in this important research area. Collaborative efforts between phytochemists and cataract researchers are promisingly fruitful. 2.1. Literature Search Literature search of articles published from January 2008 to December 2017 was performed. We searched the databases of Pubmed, Science direct, Google Scholar, Web of Science and Scopus using different combinations of keywords: lens epithelial cells, sodium selenite-, ultraviolet radiation-, steroid induced, oxygen-, H2O2-induced opacity/cataract, congenital/juvenile cataract, transgenic/knockout mice with cataract, diabetic cataract, spontaneous cataract, isolated phytoconstituents, medicinal plants. 2.2. Study Selection The selection of the manuscripts was based on the following inclusion criteria: isolation of phytoconstituents from plants, and selection of phytoconstituent(s) with the most potent anti-cataractogenesis activities. Exclusion of manuscripts from this review involves synthesized/commercialized compounds, phytoconstituents screening without isolated compounds, activities with extracts only and isolated phytoconstituents without names or articles that did not meet the inclusion criteria. The selection process is summarized in Figure 1. 3 of 41 3 of 42 Nutrients 2018, 10, 1580 Nutrients 2018, 10, x FOR Figure 1. Flow chart of review process in article selection. Records identified through various databases (n = 170) Screening Excluded due to duplication (n = 49) Records screened (n = 121) Eligibility Excluded due to criteria (n = 14) Studies included in review (n = 107) Figure 1. Flow chart of review process in article selection. Records identified through various databases (n = 170) Excluded due to duplication (n = 49) Records screened (n = 121) Excluded due to criteria (n = 14) Figure 1. Flow chart of review process in article selection. Figure 1. Flow chart of review process in article selection. 3.1. Experimental Cataract Models 3.1. Experimental Cataract Models There are a large number of in vitro and in vivo models that mimic certain aspects of the pathophysiological features of human cataracts. They have been used to demonstrate the potential therapeutic effects of phytochemicals. In this section, we describe the most commonly used models in order to aid the understanding and appraisal of results. During our literature search, most of the phytochemicals were tested in in vitro or ex vivo models only. Only a dozen or so were assessed in in vivo cataract models. Nevertheless, for completion’s sake, we list both in vitro and in vivo models. In vitro models discussed are hydrogen peroxide (H2O2)-, xylose-, galactose-induced lens opacity, aldose reductase (AR) activity assay, and advanced glycation end products (AGE) formation. In vivo models include sodium selenite-, ultraviolet (UV) radiation-, and steroid-induced cataracts. These models have been widely used to study the mechanisms of cataract and serve as the screening platform of anti-cataract therapies with the long-term goal to treat cataract in humans. There are a large number of in vitro and in vivo models that mimic certain aspects of the pathophysiological features of human cataracts. They have been used to demonstrate the potential therapeutic effects of phytochemicals. In this section, we describe the most commonly used models in order to aid the understanding and appraisal of results. During our literature search, most of the phytochemicals were tested in in vitro or ex vivo models only. Only a dozen or so were assessed in in vivo cataract models. Nevertheless, for completion’s sake, we list both in vitro and in vivo models. In vitro models discussed are hydrogen peroxide (H2O2)-, xylose-, galactose-induced lens opacity, aldose reductase (AR) activity assay, and advanced glycation end products (AGE) formation. In vivo models include sodium selenite-, ultraviolet (UV) radiation-, and steroid-induced cataracts. These models have been widely used to study the mechanisms of cataract and serve as the screening platform of anti-cataract therapies with the long-term goal to treat cataract in humans. 3.2. In Vitro Models 3.2. In Vitro Models 3.2. In Vitro Models 3.2. In Vitro Models 3 2 1 Oxidative Stress Model 3.2.1. Oxidative Stress Model 2.3. Data Extraction 2.3. Data Extraction All the selected manuscripts were analyzed for year of publication, plant species, family, part of plant, solvent extraction, isolation method, isolated phytoconstituents, anti-cataract activities (in vitro, ex vivo or in vivo), route of administration (in vivo), dose or concentration for IC50, treatment duration (in vivo) and isolated phytoconstituent(s) with the strongest activity(s), as well as their structural formula. The extracted data are presented in Tables 1 and 2 throughout this article. All the selected manuscripts were analyzed for year of publication, plant species, family, part of plant, solvent extraction, isolation method, isolated phytoconstituents, anti-cataract activities (in vitro, ex vivo or in vivo), route of administration (in vivo), dose or concentration for IC50, treatment duration (in vivo) and isolated phytoconstituent(s) with the strongest activity(s), as well as their structural formula. The extracted data are presented in Tables 1 and 2 throughout this article. 3.2. In Vitro Models 3.2. In Vitro Models Xylose-Induced Lens Opacity Another in vitro diabetic cataract model is xylose-induced lens opacity. Glucose, galactose, and xylose are all known to induce cataract. Among these three sugars, xylose is the most effective molecule in producing cataracts due to the fact that it is the preferred substrate of AR in the lens. Kinoshita and colleagues ﬁrst established xylose-induced lens in 1974 [12]. They cultured rat lenses in 4 mL of medium containing 30 mM xylose for six days. They observed a progressive development of lens opacity accompanied by increased osmotic stress and lens swelling [12]. 3.2.2. Diabetic Cataract Cataract is a major cause of visual impairment in patient with diabetes mellitus. Both clinical and basic research studies have indicated the strong association between diabetes and cataract formation [8,9]. The molecular mechanisms that may be involved in diabetic cataract include polyol pathway ﬂux, increased formation of AGEs, osmotic stress, and elevated oxidative stress. H O Induced Cataract H2O2-Induced Cataract To establish the model, rat or porcine lenses are dissected and cultured in TC-199 medium containing 200 to 1000 µM H2O2 with ﬁnal osmolarity of 298 ± 2 mOsm/L. The lens are usually harvested after 24 to 96 h to induce cataracts [6,7]. Aldose Reductase (AR) Activity The polyol pathway, also known as sorbitol-AR pathway, is a two-step process that converts glucose to fructose. AR, the ﬁrst and rate-limiting enzyme in the pathway, reduces glucose to sorbitol using nicotinamide adenine dinucleotide phosphate (NADPH) as a cofactor. Sorbitol is then converted to fructose by sorbitol dehydrogenase (SDH) [10]. As a sugar alcohol, sorbitol does not diffuse across cell membranes readily. When accumulating intracellularly, it produces osmotic stress on cells by driving water into the lens that may eventually cause diabetic cataract. Therefore, one of the possibilities to prevent the onset of diabetic cataract is to use AR inhibitor (ARI) [11]. The in vitro ARI assay was used to evaluate if the compound can inhibit the polyol pathway. Brieﬂy, the reaction mixture contains 50 µM potassium phosphate buffer pH 6.2, 0.4 mM lithium sulfate, 5 µM 2-mercaptoethanol, 10 µM DL-glyceraldehyde, 0.1 µM NADPH, and freshly-prepared AR enzyme. The reaction is initiated by the addition of NADPH at 37 ◦C. The AR activity is determined indirectly by a spectrophotometer that measures NADPH absorption [10]. H O Induced Cataract H2O2-Induced Cataract H2O2 Induced Cataract It is widely accepted that oxidative stress is the major factor for the development of cataracts. Hydrogen peroxide (H2O2) is the major reactive oxygen species (ROS). H2O2 is mainly generated in vivo by the detoxification of superoxide (O2−) radical by superoxide dismutase (SOD) through the dismutation reaction [4,5]. Alternatively, H2O2 can be produced by a number of oxidase enzymes including monoamine oxidases and peroxisomal pathway for β-oxidation of fatty acids In the lens It is widely accepted that oxidative stress is the major factor for the development of cataracts. Hydrogen peroxide (H2O2) is the major reactive oxygen species (ROS). H2O2 is mainly generated in vivo by the detoxiﬁcation of superoxide (O2−) radical by superoxide dismutase (SOD) through the dismutation reaction [4,5]. Alternatively, H2O2 can be produced by a number of oxidase enzymes including monoamine oxidases and peroxisomal pathway for β-oxidation of fatty acids. In the lens, 4 of 41 Nutrients 2018, 10, 1580 H2O2 can also be generated by the photochemical reaction [4]. Most human tissues, including the lens, are exposed to some level of H2O2, with the mitochondria being the major site for production. Previous studies have shown the strong association between H2O2 overproduction and cataract development. Cataract patients had elevated H2O2 in both the aqueous body and lens ranging from seven- to 30-fold higher than normal [5]. Lens organ ex vivo culture with H2O2 in the medium is a common experimental model of cataract. This type of cataract is characterized by loss of GSH and increased protein oxidation. To establish the model, rat or porcine lenses are dissected and cultured in TC-199 medium containing 200 to 1000 µM H2O2 with ﬁnal osmolarity of 298 ± 2 mOsm/L. The lens are usually harvested after 24 to 96 h to induce cataracts [6,7]. H2O2 can also be generated by the photochemical reaction [4]. Most human tissues, including the lens, are exposed to some level of H2O2, with the mitochondria being the major site for production. Previous studies have shown the strong association between H2O2 overproduction and cataract development. Cataract patients had elevated H2O2 in both the aqueous body and lens ranging from seven- to 30-fold higher than normal [5]. Lens organ ex vivo culture with H2O2 in the medium is a common experimental model of cataract. This type of cataract is characterized by loss of GSH and increased protein oxidation. 3.3.2. Selenite-Induced Cataract Selenite-induced cataract is an effective, rapid, and reproducible model of nuclear cataracts. Selenite cataract is usually produced either by a single dose (19–30 µM/kg body weight) or repeated smaller dosage of sodium selenite (40–50 nmol/g body weight) subcutaneous injection to suckling rat of 10–14 days of age [20]. It has been proposed that selenite treatment leads to altered metabolism in lens epithelium, including loss of small antioxidant molecules such as glutathione (GSH), decreased rate of epithelial cell differentiation, and increased DNA oxidation damage. Such extensive alterations to the epithelium leads to disrupted calcium homeostasis and calcium accumulation in the nucleus of the lens. Increased calcium activates calcium dependent protease m-calpain (calpain II) which results in rapid proteolysis, precipitation of crystallins, and eventually cataract development in rodent lenses [21,22]. Formation of Advanced Glycation End (AGE) Products Formation of Advanced Glycation End (AGE) Products Another important factor that is involved with the pathogenesis of diabetic cataract is the formation of AGEs. In diabetic patients with cataract, the elevated glucose starts forming covalent adducts with the lens proteins through a non-enzymatic process called glycation [14,15]. This process is known as one of the most important forms of post-translational modiﬁcation of proteins under hyperglycemic conditions. Many studies have shown that protein glycation-induced AGEs play a pivotal role in diabetic cataract formation. Therefore, AGE formation assay is used to examine the potential anti-cataract potential of tested compounds. To determine the amount of AGEs, a reaction mixture containing 10 mg/mL of bovine serum albumin and 0.5 M fructose and glucose are mixed with tested compounds. After 15 days of incubation, the ﬂuorescent intensity is measured using a spectroﬂuorometric detector with an excitation wavelength of 350 nm and an emission wavelength of 450 nm [16,17]. 3.3. In Vivo Models Commonly used in vivo models represent speciﬁc pathogenesis aspects of human cataract. For example, the diabetic cataract rodent model focuses on mechanisms involved in diabetes-related cataract; selenite-induced cataract addresses oxidative damage-induced cataract; the UV- and steroid-induced models represent their respective associated pathological changes. In various studies, drug effects in these in vivo models correlate well with the pharmacodynamics properties shown in appropriate in vitro models. 3.3.1. Diabetic Cataract The in vivo diabetic cataract model can be established by using streptozotocin (STZ). After intraperitoneal (i.p.) or intravenous (i.v.) injection, STZ enters the pancreatic β-cell through the glucose transporter 2 transporter (Glut-2) resulting in hyperglycemia [16]. Moreover, STZ is also a source of free radicals that may lead to DNA oxidative damage and subsequent β-cell death. STZ can be administered as a single high dose (e.g., 160 to 240 mg/kg) or as multiple low doses (e.g., 40 mg/kg for 5 days) [18]. Another commonly used diabetic cataract model is AR transgenic mice. The ubiquitous transgenic and lens-speciﬁc AR transgenic mice were developed to further prove that polyol accumulation is responsible for diabetic cataract. In both models, sorbitol accumulates in the lens, causing osmotic swelling, and eventually leading to accelerated diabetic cataract formation [10,19]. Galactose-Induced Lens Opacity Compared with glucose, galactose has higher afﬁnity with AR and its reduction product galactitol is more difﬁcult to be metabolized by sorbitol dehydrogenase than sorbitol. Therefore, high galactose is more likely to induce sugar cataract than high glucose itself [13]. There are several methods available to establish galactosemic cataract. For example, rat galactosemic cataract can be induced by 30% or 50% galactose diet. Glactose-induced lens opacity can also be achieved by daily intraperitoneal injection of 30–50% galactose solution or daily retrobulbar injection of 20% galactose solution. Another cost efﬁcient way to induce galactosemic cataract is to feed rat 10% galactose solution for 18 days. For in vitro lens culture, 30 mM galactose is added in the culture medium for 72 h incubation [13]. Nutrients 2018, 10, 1580 5 of 41 3.4.1. 1-O-Galloyl-β-D-glucose (β-Glucogallin) Molecular formula: C13H16O10 (332.262 g/mol), Melting point: 214–216 ◦C. ( g ) g p β-Glucogallin isolated from the aqueous fruit extract of Emblica ofﬁcinalis Gaertn. (emblic, Indian gooseberry) or Phyllanthus emblica Linn. (Euphorbeaceae) (gooseberry) [29] shows potent activity against human AR in vitro with an IC50 of 17 µM [30]. Treatment with this compound prevented the sorbitol accumulation by 73% (30 µM) in transgenic human AR expressing lenses ex vivo [30]. This result substantiated the in vitro assay using shared substrate glyceraldehyde at IC50 of 58 µM. Treatment with β-Glucogallin produced a signiﬁcant decrease of sorbitol levels in macrophages [31]. Computational molecular docking studies exhibited favorable binding to the active site of between human AR and β-glucogallin. This corroborates the inhibition result of sorbitol production under hyperglycemic conditions in earlier experiments [30]. 3.4. Anti-Cataract Phytoconstituents Based on our literature strategy listed above, the following phytoconstituents are listed in alphabetical order. They have been shown to possess potential anti-cataract efﬁcacy according to the described study models. 3.3.3. UV-Induced Cataract UV radiation is a major contributor to the pathogenesis of cataract. The strong energy in the UV light can directly cause a DNA lesion in the lens by inducing thymine dimer formation. More importantly, UV can induce cataract formation by the generation of ROS that indirectly induce oxidative 6 of 41 Nutrients 2018, 10, 1580 damage to DNA by disturbing cell proliferation in the lens epithelium, altering kinetic properties of enzymes in the energy metabolism, increasing insoluble and decreasing soluble protein, and disturbing the sodium potassium balance, leading to aberrant water balance in the lens [23]. It has been widely accepted that cataract formation is related to oxidative stress induced by continued intraocular penetration of UV light and consequent photochemical generation of ROS such as superoxide and singlet oxygen and their oxidant derivatives such as hydrogen peroxide and hydroxyl radical [24]. Sprague-Dawley rats or mice are exposed to 8 kJ/m2 UV-B radiation for 15 min to induce cataracts [25]. 3.3.4. Steroid-Induced Cataract As the steroid hormones, glucocorticoids (GCs) have strong anti-inﬂammatory effects. By binding with the glucocorticoid receptor (GR), GCs have the ability to inhibit all stages of the inﬂammatory response [26]. Due to its strong anti-inﬂammatory effects, GCs are widely used in the management of many clinical conditions, including autoimmune disorders, allergies, and asthma, and they also play important roles in chemotherapy and preventing the rejection after solid organ transplantation. However, prolonged use of GCs is associated with the development of posterior subcapsular and nuclear cataracts [26]. The chick embryo has been used to establish an experimental model to study the response of the lens to GCs. When dexamethasone (0.02 µmol/egg) is administered, the lenses of chicken embryos become cataract within 48 h. More recently, the mammalian lens has also been used to establish the steroid-induced cataract models. For example, Brown-Norway rats given a daily 1% prednisolone acetate instillation of a total volume of 1.0 mg/kg or a daily intramuscular injection of 0.8–1.0 mg/kg prednisolone acetate for 10 months successfully induced morphological changes similar to those found in human steroid-induced cataracts [27,28]. Molecular formula: C24H24O11 (488.44 g/mol). A new method of enzyme assay-guided high-performance liquid chromatography microfractionation and elution-extrusion counter-current chromatography of roots ethanolic extract of Nardostachys chinensis (spikenard) afforded six secondary metabolites with 1,5-di-O-caffeoylquinic acid as the most potent inhibitor against RLAR activity (IC50 = 2.98 µM). The compound was reported as the ﬁrst time isolated from the plant [35]. Molecular formula: C16H12O6 (300.26 g/mol). Bioassay-guided fractionation of Knoxia valerianoides (hongdaji) methanolic root extract afforded eight secondary metabolites with 1,3,6-trihydroxy-2-methoxymethylanthraquinone showing the highest inhibition against AGE formation at IC50 value of 52.7 µM. The same phytoconstituent also exhibited strong inhibitory activity against RLAR with IC50 value of 3.0 µM [36]. Molecular formula: C34H28O22 (788.57 g/mol). Molecular formula: C34H28O22 (788.57 g/mol). 1,2,3,6-tetra-O-galloyl-β-D-glucose was isolated from the methanolic seeds extract of Cornus ofﬁcinalis (cornus tree, shan zhu yu) after repeated Sephadex column chromatography. Appeared as an off-white amorphous powder [37], 1,2,3,6-tetra-O-galloyl-β-D-glucose showed the most potent inhibitory activity (IC50 = 0.70 µM) compared to other secondary metabolites. In addition, AGE formation was also reduced to IC50 value of 1.99 µM. This compound was further evaluated for its inhibitory effect on ex vivo cataractogenesis activity using rat lenses induced with xylose 20 mM. Treatment with 1,2,3,6-tetra-O-galloyl-β-D-glucose signiﬁcantly reduced the opacities of the lenses after two days at the concentration of 80 µM [38]. Molecular formula: C25H24O12 (516.45 g/mol). 1,3-di-O-caffeoylquinic acid has been isolated from Artemisia iwayomogi (haninjin) and Xanthium strumarium (rough cocklebur) as inhibitor for rat lens AR (RLAR), recombinant human AR (RHAR) and advanced glycation end-product (AGE) inhibitory activities. The compound inhibited RLAR with IC50 values of 0.22–1.90 µM [32,33]. This result was supported by inhibition of RHAR at IC50 of 0.81 µM. In AGE inhibitory activity, 1,3-di-O-caffeoylquinic acid suppressed at IC50 of 24.85 µM [33]. 7 of 41 Nutrients 2018, 10, 1580 3.4.3. 1,5-Di-hydroxy-1,5-di-[(E)-3-(4-hydroxyphenyl)-2-propenoic]-3-pentanonyl Ester (DHDP) Molecular formula: C23H22O9 (442.41 g/mol). A novel polyphenolic inhibitor of AR, DHDP was isolated from Lysimachia christinae (gold coin grass, jinqiancao) using AR afﬁnity-based ultraﬁltration-HPLC proﬁling method. The reversible inhibitory activity of RHAR was recorded at IC50 value of 194.7 µM with sorbitol content of 1002.3 µg/g of lens weight. The effect of DHDP was further investigated in in silico using computer simulation of binding by molecular docking. DHDP was predicted to block the AR active site by binding and preventing the formation of product [34]. Molecular formula: C13H8O6 (260.19 g/mol). Molecular formula: C13H8O6 (260.19 g/mol). Several xanthones have been isolated from the ethanolic extract of Swertia mussotii Franch (yinchen) as inhibitors for RLAR activity. The most potent inhibition was shown by 1,3,5,8-tetrahydroxyxanthone with IC50 of 0.0886 µM [39]. The compound appeared in slight yellow powder with 98.6% purity. Molecular formula: C20H24O9 (406.42 g/mol). The leaves and twigs of Osteomeles schwerinae C. K. Schneid. (hu xi xiao shi ji) were examined for their possible inhibitory activity on RLAR. Four secondary metabolites have been isolated from the CHCl3-MeOH fraction of the EtOH extract and found that 3′,5′-dimethoxy-(1,1′-biphenyl)-3,4-diol 3-O-β-D-glucopyranoside to be the most potent inhibitor against RLAR activity at IC50 value of 3.8 µM. The phytoconstituent was obtained as a brownish powder [47]. 3.4.8. 2′′,4′′-O-Diacetylquercitrin Molecular formula: C25H24O13 (532.11 g/mol), Melting point: 187 ◦C. 2′′,4′′-O-Diacetylquercitrin was isolated from Melastoma sanguineum (red melastome, fox-tongued melatsome) as a yellow amorphous powder. This compound exhibited the strongest inhibition against RLAR and AGE activities among all the isolated phytoconstituents. IC50 inhibitory activities for RLAR and AGE were recorded at 0.077 µM and 11.46 µM, respectively. Compared to the positive standards, aminoguanidine (IC50 = 965.9 µM, AGE) and 3,3-tetramethyleneglutaric acid (IC50 = 28.8 µM, RLAR), 2′′,4′′-O-Diacetylquercitrin inhibited 87 (AGE) and 374 (RLAR) times more efﬁcaciously [40]. Nutrients 2018, 10, 1580 8 of 41 3.4.9. 3-Isomangostin Molecular formula: C24H26O6 (410.46 g/mol), Melting point: 182–183 ◦C. Three main constituents of dichloromethane extract from root bark of Garcinia mangostana Linn (mangosteen) were isolated from the hexane/methanol fraction. The result of the study indicated that 3-isomangostin possessed the highest RLAR inhibitory activity with at IC50 value of 3.28 µM. The presence of cyclization of the prenyl group at the position-two carbon with xanthone derivative enhanced the structure-activity relationship [41]. 3.4.12. 3,5-Di-O-caffeoylquinic Acid Molecular formula: C25H24O12 (516.45 g/mol), Melting point: 184–187 ◦C. 25 24 12 ( g/ ), g p Methanolic extract of the stems and leaves of Erigeron annuus (annual ﬂeabane, daisy ﬂeabane) afforded 16 secondary metabolites. 3,5-di-O-caffeoylquinic acid appeared as pale-yellow powder and isolated from the ethyl acetate-soluble fraction after repeated column chromatography [48–50]. The same constituent was also isolated from Aster koraiensis (Korean starwart) [51], Xanthium strumarium (clotbur, common cocklebur) [33], Artemisia iwayomogi (haninjin) [32] and Artemisia montana [52]. 3,5-di-O-caffeoylquinic acid was reported as the most signiﬁcant inhibitory activities against AGEs, RLAR and ex vivo xylose-induced lens opacity assays from all isolated constituents. It attenuates AGE formation with IC50 values ranging from 6 µM to 32 µM, and inhibits RLAR with IC50 values of 0.2 to 5 µM. These ﬁndings are further substantiated by its ability in inhibition of galactitol accumulation at an IC50 of 153 µM [33] and prevention of xylose-induced opacity of lenses at a concentration of 10 µM [48]. 3.4.13. 4-O-Butylpaeoniﬂorin and Palbinone 3.4.11. 3′,5′-Dimethoxy-(1,1′-biphenyl)-3,4-diol 3-O-β-D-glucopyranoside 3.4.11. 3′,5′-Dimethoxy-(1,1′-biphenyl)-3,4-diol 3-O-β-D-glucopyranoside Molecular formula: C16H18O4 (274.316 g/mol), Melting point: 135 ◦C. Gigantol is a bibenzyl-type phenolic compound presents in most herbs of Orchidaceae family [42]. It has been isolated from the stems of various Dendrobium genus such as Dendrobium aurantiacum var. denneanum (die qiao shi hu) [43,44] and Dendrobium chrysotoxum Lindl (fried-egg orchid) [45,46] for anti-cataract activities. As a white solid, gigantol suppresses the damage of rat lenses both in vitro and in vivo in galactose-induced cataractogenesis. The delay in lens turbidity was caused by the inhibition of AR and inducible nitric oxide synthase mRNA expression at an IC50 of 239.4 µM (65.7 µg/mL) and 32.0 µM (8.8 µg/mL), respectively [43]. Gigantol isolated from Dendrobium chrysotoxum Lindl interpolated into the DNA base pairs in AR gene with a binding constant of 1.85 × 103 L/mol, thus, suppressed the gene expression [46]. 3.4.10. 3′,4-Dihydroxy-3,5′-dimethoxy-bibenzyl (Gigantol) Molecular formula: C16H18O4 (274.316 g/mol), Melting point: 135 ◦C. Molecular formula: C15H10O5 (270.24 g/mol), Melting point: 297–298 ◦C. Molecular formula: C15H10O5 (270.24 g/mol), Melting point: 297–298 ◦C. 15 10 5 ( g ), g p Genistein appears as colorless plates and isolated from the roots of Pueraria lobata (kudzu, Japanese arrowroot) [57,58] and stem bark of Maackia amurensis (Amur maackia) [59]. Both plants are native to Eastern Asia and used as traditional medicine in China, Korea, and Japan. Genistein shows a signiﬁcant dose-dependent inhibition on RLAR activity (IC50 = 9.48 µM) compared to the positive control, TMG (3,3-tetramethyleneglutaric acid) (IC50 = 28.70 µM). Nevertheless, IC50 was recorded higher at 57.1 µM for the same activity compared to quercetin IC50 = 10.1 µM) [59]. In an ex vivo lens opacity study genistein suppressed xylose-induced lens opacity at 5 µg/mL (18.5 µM). Further analysis with human lens epithelia cells (LECs; HLE-B3 cells) found that the expression of TGF-β2, αβ-crystallin, and ﬁbronectin mRNAs were reduced, suggesting genistein is protective against lens opacity with antioxidative effects [60]. It is proposed that the chemical moiety with free hydroxyl group at C-7 of genistein attributes to the inhibitory of AR [59]. 3.4.13. 4-O-Butylpaeoniﬂorin and Palbinone Molecular formula of 4-O-butylpaeoniflorin: C27H36O11 (536.22 g/mol), Melting point: 173–175 ◦C Molecular formula of Palbinone: C22H30O4 (359.47 g/mol), Melting point: 254–255 ◦C. Molecular formula of 4-O-butylpaeoniflorin: C27H36O11 (536.22 g/mol), Melting point: 173–175 ◦C. Molecular formula of Palbinone: C22H30O4 (359.47 g/mol), Melting point: 254–255 ◦C. Both 4-O-butylpaeoniﬂorin and Palbinone were isolated from methanolic extract of the cortex of Paeonia suffruticosa (tree peony, mudan, moutan) with highest inhibitory activities of RLAR (palbinone) and AGE (4-O-butylpaeoniﬂorin) compared to other isolated phytoconstituents [53]. Palbinone appeared as red needles with [α]D −223.8◦(CHCl3) and absorbed UV at 237 (log ε: 3.2) and 387 9 of 41 Nutrients 2018, 10, 1580 nm (log ε: 3.0) [54]. Isolated from the butanol fraction. 4-O-butylpaeoniﬂorin was found as an optically active white foam, [α]25 D –7.8 (c 0.14, MeOH) and later conﬁrmed as an extraction artifact after HPLC analysis. Palbinone inhibits RLAR at an IC50 value of 11.4 µM. It was suggested that the absence of ring E, side chain of ring D together with double bonds and a conjugated carbonyl group on the ring D played the inhibitory properties. Unlike palbinone, 4-O-butylpaeoniﬂorin inhibited (IC50 = 10.8 µM) for AGE activity. The chemical moiety of hydroxy groups in the benzoyl connected to the sugar unit complement the activity [53]. Molecular formula: C36H62O8 (622.87 g/mol). g 20(S)-Ginsenoside Rh2 is classiﬁed under triterpene glycosides and isolated from the root of Panax ginseng C. A. Meyer, (ginseng). It has been used traditionally in East Asia for many years ago with many main active constituents, ginsenosides have been isolated. In RHAR inhibitory activity, 20(S)-Ginsenoside Rh2 showed the most potent inhibitor with an IC50 of 147.4 µM among all other isolated ginsenosides. It was suggested that the moiety of hydroxyl group at the carbon-20 enhanced the AR activity relationship [61]. Molecular formula: C25H24O12 (516.45 g/mol). Caffeoylquinic acid analog, 4,5-Di-O-trans-caffeoyl-D-quinic acid isolated from Hydrangea macrophylla var. thunbergii (bigleaf hydrangea) and Ilex paraguariensis (Yerba mate) showed the strongest inhibitory activity against RLAR at IC50 value of 0.29 µM [55]. Inhibitory effect of quinic acid with two caffeoyl groups assisted the potency. Molecular formula: C17H20O9 (368.33 g/mol) Bioassay-guided isolation of root methanolic extract of Aralia continentalis Kitag. (dong bei tu dang gui) produced 18 secondary metabolites. 5-O-Feruloly quinic acid was isolated from the ethyl acetate fraction as an amorphous white powder. It had a highest inhibitory activity of RLAR at IC50 value of 14.2 µM among all other phytoconstituents [56]. Molecular formula: C19H24O9 (396.38 g/mol). Molecular formula: C19H24O9 (396.38 g/mol). Basilicumin was isolated from Ocimum basilicum (basil). It exhibits potent inhibitory activity against AR (AKR1B1) and aldehyde reductase (AKR1A1) compared to the second phytoconstituent isolated, ocimunone. Basilicumin inhibited AKR1A1 at IC50 value of 0.78 µM and 2.1 µM for AKR1B1 activity. It was suggested that coumarin and glucose scaffold in basilicumin moiety enhance the activity [64]. 3.4.20. Caffeic Acid Molecular formula: C9H8O4 (180.16 g/mol), Melting point: 223–225 ◦C. Caffeic acid appears as white amorphous powder and classiﬁed as a hydroxycinnamic acid [65]. In an attempt to ﬁnd potential cataractogenesis inhibitors from plants, caffeic acid has been isolated from a few plants with potential activity. Isolation of caffeic acid for RLAR activity has been shown from methanolic extract of Dipsacus asper (xuduan), Erigeron annuus (L.) Pers., and Phellinus linteus (black hoof mushroom, meshima, song gen, meshimakobu, sanghwang) with the highest activity among all isolated secondary metabolites at IC50 values of 16.7 µM to 55 µM. Comparable results were observed for RHAR (IC50 = 55 to 210 µM) and AGE activities (IC50 = 7.6 µM) [66–68]. Interestingly, no inhibitory effects were observed for caffeic acid isolated from Perilla frutescens L. [69] and Prunella vulgaris L. [70]. Molecular formula: C45H72O10 (773.04 g/mol). The isolation of Capsofulvesin A from ethanolic extract of Capsosiphon fulvescens (one of the green algae) showed the strongest RLAR inhibitory activity among all other secondary metabolites, albeit moderate activity at IC50 value of 52.5 µM. However, the constituent did not show any inhibition against AGE activity [72]. 3.4.18. Acteoside Molecular formula: C29H36O15 (624.58 g/mol), Melting point: 143–146 ◦C. Isolated as yellowish amorphous powder from methanolic extract of Abeliophyllum distichum (forsythia) and leaves and stem ethanolic extracts of Brandisia hancei (laijiangteng), acteoside showed the 10 of 41 10 of 41 Nutrients 2018, 10, 1580 highest RLAR inhibitory activities at IC50 values ranging 0.83 µM and 1.39 µM compared to four other isolated phenolic glycosides from each plant, respectively. The isolation was conducted by high-speed counter current chromatography using a solvent system of ethyl acetate:n-butanol:water [62]. Isolated acteoside from Brandisia hancei showed potent AGE inhibitory activity with an IC50 value of 5.11 µM [63]. 3.4.19. Basilicumin [7-(3-hydroxypropyl)-3-methyl-8-β-O-D-glucoside-2H-chromen-2-one] 3.4.23. Caryatin-3′ methyl ether-7-O-β-D-glucoside 3.4.23. Caryatin-3′ methyl ether-7-O-β-D-glucoside 3.4.22. Capsofulvesin A [((2S)-l-O-(6Z,9Z,12Z,15Zoctadecatetraenoyl)-2-O-(4Z,10Z,13Zhexadecatetraen 3-O-β-D-galactopyranosyl Glycerol)] 22. Capsofulvesin A [((2S)-l-O-(6Z,9Z,12Z,15Zoctadecatetraenoyl)-2-O-(4Z,10Z,13Zhexadecatetraenoyl -β-D-galactopyranosyl Glycerol)] Molecular formula: C45H72O10 (773.04 g/mol). Molecular formula: C25H32O12 (524.51 g/mol). Molecular formula: C25H32O12 (524.51 g/mol). The repeated column chromatography of methanolic ﬂower bud extract of Cananga odorata Hook. F. and Thomson generated 25 secondary metabolites and they were tested for RLAR inhibitory activity. The result of the study indicated that among the isolated constituents, canangafruiticoside E possessed the highest activity (IC50 = 0.8 µM) [71]. Molecular formula: C24H26O12 (506.14 g/mol). Molecular formula: C24H26O12 (506.14 g/mol). The bark of the pecan tree (Carya illinoinensis (Wangenh) K. Koch) has shown good inhibition of AR activity with few compounds have been isolated. Among them, caryatin-3′ methyl ether-7-O-β-D-glucoside exhibits the most powerful activity in suppressing the lens AR levels in diabetic cataract rats [73]. The catechol moiety on the B ring of caryatin-3′ methyl ether-7-O-β-D-glucoside was suggested to inhibit AR in comparison to the activity of other compounds isolated. In addition, the potent AR activity was also supported by the presence of neighboring O-methyl group in phenolics and an OH group at C-4′ [73–75]. Caryatin-3′ methyl 11 of 41 Nutrients 2018, 10, 1580 ether-7-O-β-D-glucoside is physically yellow amorphous powder with UV (MeOH) λmax absorption at 350, 330, and 260 nm [73]. ether-7-O-β-D-glucoside is physically yellow amorphous powder with UV (MeOH) λmax absorption at 350, 330, and 260 nm [73]. Molecular formula: C33H38N4O6 (586.67 g/mol). g C-Phycocyanin (C-PC), a prominent phytoconstituent found in the stromal surface of thylakoid membranes of Spirulina platensis (a blue-green algae) is a biliprotein that functions to capture light energy to chlorophyll A [76–78]. As C-PC is miscible in water but not alcohol and esters, most of the isolations of C-PC use water extraction method [79]. C-PC attenuates selenite-induced cataractogenesis both in vitro and in vivo rat model [78,80]. In vitro study showed C-PC recorded low degree of opaciﬁcation at 200 µg C-PC with 100 µM sodium selenite [78]. The puriﬁed C-PC was active toward the in vivo selenite mediated cataractogenesis showing only slight opaciﬁcation at 200 mg/kg [78]. Same concentration was observed for naphthalene- and galactose-induced cataract rat models [81]. The protective effect of C-PC in these models were proven from the increment of glutathione, soluble proteins, and water content levels of the lens [79]. Histology study indicated the protection of the lens from oxidative damage. Restoration of lenticular micro-architecture was found with C-PC treated group [77]. C-PC maintains the lens transparency by transcriptional regulation of crystallin, redox genes, and apoptotic cascade mRNA expression [80]. Furthermore, C-PC was suggested to possess protective effects on human LEC by abrogating D-galactose-induced apoptosis through the mitochondrial pathway (p53 and Bcl-2 family protein expression) and unfolded protein response pathway (GRP78 and CHOP expression) [82]. Molecular formula: C27H31O17 (627.52 g/mol). Anthocyanin delphinidin 3-O-β-galactopyranoside-3′-O-β-glucopyranoside was isolated from the methanolic extract of the air-dried fruit pericarp of Litchi chinensis Sonn (lychee). This fruit is a tropical and subtropical edible fruit native to Southeast Asia. Delphinidin 3-O-β-galactopyranoside-3′-O-β-glucopyranoside exhibits the most signiﬁcant inhibitory activity in RLAR assay with an IC50 value of 0.23 µg/mL (0.37 µM) compared to the positive control, tetramethylene glutaric acid (IC50 = 0.48 µg/mL) [83]. Molecular formula: C25H20O9 (464.43 g/mol). Davallialactone was isolated as yellow amorphous powder from the active ethyl acetate fraction of fruiting body of Phellinus linteus. Davallialactone possessed the most potent inhibitory against RLAR and RHAR among all the isolated compounds with IC50 values of 0.33 µM and 0.56 µM, respectively. The inhibitory activities were nine times (RLAR) and 11 times (RHAR) compared to that of quercetin (IC50 = 2.91 µM; RLAR and IC50 = 6.27 µM; RHAR) [67]. 3.4.26. Delphinidin 3-O-β-galactopyranoside-3′-O-β-glucopyranoside 3.4.25. Davallialactone Molecular formula: C25H20O9 (464.43 g/mol). 3.4.24. C-Phycocyanin (C-PC) Molecular formula: C33H38N4O6 (586.67 g/mol). 3.4.27. Desmethylanhydroicaritin Molecular formula: C20H18O6 (356.36 g/mol), Melting point: 220–222 ◦C. The isolation of repeated chromatography of the CH2Cl2 fraction over a silica-gel column and Sephadex LH20 from root methanolic extract of Sophora ﬂavescens (kushen) afforded desmethylanhydroicaritin. Desmethylanhydroicaritin exerted remarkable inhibitory activity of RLAR with IC50 value of 0.95 µM. Comparable results were observed in RHAR and AGE inhibitions where IC50 values were observed at 0.45 µM and 294.6 µM, respectively. The presence of prenyl and lavandulyl groups enhanced the RLAR and RHAR inhibitory activities. The 3-hydroxyl group at prenylated ﬂavonoids was suggested for the structural contribution for inhibition of AGE formation [84]. 12 of 41 Nutrients 2018, 10, 1580 3.4.30. Geraniin Molecular formula: C41H28O27 (952.64 g/mol), Melting point: 360 ◦C. The anti-cataract activities of Nephelium lappaceum (rambutan) [89] and Geranium thunbergii [85] lead to the isolation of geraniin with good yield. Geraniin was isolated from the ethanolic rind extract of Nephelium lappaceum as the major bioactive compound. This compound exhibits better AR activity with an IC50 of 0.15 µM at approximately 40% higher compared to quercetin IC50 = 5.76 µM [89]. Geraniin isolated from Geranium thunbergii shows slightly higher concentration of IC50 (8.54 µM) in the same activity, however, using rat lens as the source of enzyme [85]. In AGE assay, the activity of geraniin was 96% of inhibition after incubation time of seven days at the concentration of 20 µg/mL (21 µM). Galactitol accumulation in rat lenses incubated with high galactose was inhibited at 39.9% by geraniin with 507.5 µg/lens wet weight (g). It was concluded that geraniin isolated from both plants is a promising agent in the prevention or treatment of diabetic complications. 3.4.31. Hipolon Molecular formula: C12H12O4 (220.22 g/mol), Melting point: 237.5–238.5 ◦C. Three inhibitors have been isolated from ethanolic extract of Phellinus merrillii (willow) fruiting body and identiﬁed as hispidin, hispolon, and inotilone. Hipolon showed highest inhibition against RLAR activity (IC50 = 9.47 µM) among the three suggesting that phenolic chemical moiety enhanced the activity [90]. 3.4.28. Ellagic Acid Molecular formula: C14H6O8 (302.19 g/mol), Melting point: ≥350 ◦C. Molecular formula: C14H6O8 (302.19 g/mol), Melting point: ≥350 ◦C. During a search for possible cataractogenesis activities for isolated ellagic acid, three plants from Korea were found with most potent inhibitions. Ellagic acid isolated from Phellinus linteus, Geranium thunbergii (Thunberg’s geranium), and Syzygium cumini (L.) Skeels (jambolan, Java plum, black plum) inhibits RLAR activity with IC50 value ranging from 0.12 µM to 6.9 µM [67,85–87]. The compound was also effective in the inhibition of AGE formation (IC50 = 26.0 µM). In RHAR assay, the activity of ellagic acid (IC50 = 1.37 µM) from Phellinus linteus was more potent than that of quercetin (IC50 = 6.27 µM) [67]. This was substantiated by its inhibition (42.5%) of galactitol accumulation in rat lenses incubated in high glucose with 485.6 µg/lens wet weight [85]. Molecular formula: C20H18NO4+ (336.36 g/mol), Melting point: 187 ◦C. Molecular formula: C20H18NO4+ (336.36 g/mol), Melting point: 187 ◦C. The bioassay-guided isolation of the rhizome of Coptis chinensis Franch (Chinese goldthread) afforded seven secondary metabolites with epiberberine exhibited the highest inhibitory of RLAR activity. The IC50 of the reported value was 100 µM. Conversely, epiberberine showed a comparable result against RHAR with IC50 value of 168.1 µM. The chemical moiety of dioxymethylene (ring D) and its oxidized form (ring A) was suggested to enhance the AR inhibitory activities, albeit in moderate effects [88]. 3.4.37. Kaempferol Molecular formula: C15H10O6 (286.23 g/mol), Melting point: 276–278 ◦C. Kaempferol isolated from Litsea japonica (Thunb.) Juss. (hamabiwa) showed the most potent against RLAR inhibitory activity with IC50 value of 1.10 µM among all phytoconstituents isolated [98]. The same constituent was also isolated from Agrimonia pilosa Ledeb (hairy agrimony, hangul), Allium victorialis (victory onion), and Paulownia coreana with IC50 values of 15.2 µM (RLAR) [99], 1.10 µM (RLAR) [100], and 45.58 µM (RHAR) [66], respectively. The cataract prevention was further supported by inhibition of AGE activity at IC50 of 36.01 µM from Allium victorialis [100]. 3.4.36. Isorhamnetin-3-glucoside Molecular formula: C22H22O12 (478.406 g/mol), Melting point: 168–172 ◦C. Cochlospermum religiosum (silk-cotton tree, buttercup tree) has been reported to possess anticataract activity [96]. Puriﬁcation of hot 95% ethanolic leaves extract of C. religiosum yielded isorhamnetin-3-glucoside. This bioactive compound was obtained as yellow needles and identiﬁed as ﬂavonoids with yellowish orange color in alkali, pink in Mg-HCl and reaction with Fe3+ gives olive green color. Isorhamnetin-3-glucoside at the concentration of 25 µg/mL (52 µM) inhibited further formation of vacuoles and opacity on sodium selenite-induced lens opacity of rat pups. The antioxidant property of isorhamnetin-3-glucoside was suggested to complement its anticataract activity [97]. 3.4.32. Hirsutrin Molecular formula: C21H20O12 (462.40 g/mol), Melting point: 156–157 ◦C. Hirsutrin was isolated together with six nonanthocyanin and ﬁve anthocyanin compounds from Zea mays L. (corn) for anti-cataractogenesis activity. Isolation of hirsutrin was conducted through bioassay-guided fractionation of ethanolic extract from the kernel of Zea mays L. using repeated column chromatography from ethyl acetate fraction. Hirsutrin showed the highest inhibitory activity in RLAR with an IC50 value of 4.78 µM and inhibitory constant (Ki) at 7.21 × 10−7 M from secondary plots of Lineweaver-Burk plots for RHAR assay. Further inhibition by hirsutrin on galactitol formation in rat 13 of 41 Nutrients 2018, 10, 1580 lens (33.8% inhibition) and erythrocytes (15.7 µM, 32.5% inhibition) supported the efﬁcacy of hirsutrin as the most effective AR inhibitors compared to all isolated compounds [91]. lens (33.8% inhibition) and erythrocytes (15.7 µM, 32.5% inhibition) supported the efﬁcacy of hirsutrin as the most effective AR inhibitors compared to all isolated compounds [91]. 3.4.33. Hopeafuran cular formula: C28H18O7 (466.43 g/mol), Melting point: 131–134 ◦C. Molecular formula: C28H18O7 (466.43 g/mol), Melting point: 131–134 ◦C. Hopeafuran, classiﬁed under oligostilbenoids was isolated from the bark of Shorea roxburghii (white meranti) and exhibits the highest RLAR inhibitory activity compared to other isolated secondary metabolites from the same plant. This phytoconstituent inhibits the AR enzyme at an IC50 value of 6.9 µg/mL (14.8 µM) [92]. 3.4.34. Hypolaetin 7-O-[6′′′-O-acetyl-β-D-allopyranosyl-(1→2)]-6′′-O-acetyl-β-D-glucopyranoside Molecular formula: C31H34O19 (710.59 g/mol). Hypolaetin 7-O-[6′′′-O-acetyl-β-D-allopyranosyl-(1→2)]-6′′-O-acetyl-β-D-glucopyranoside was isolated from Sideritis brevibracteata (Da˘g çayı) [93] and appeared as yellow powder [94,95]. This plant is native to Turkey and widely used as an herbal tea in folk medicine. Isolated hypolaetin has shown the most potent inhibitory activity of AR with IC50 value of 0.66 µM [93]. Molecular formula: C29H36O16 (640.58 g/mol). Molecular formula: C29H36O16 (640.58 g/mol). Isocampneoside II is an active phenylethanoid glycoside isolated from acetone-H2O (7:3, v/v) seeds extract of Paulownia coreana (kiri, paotong) at room temperature for 72 h. Paulownia coreana is long cultivated in Eastern Asia, particularly Korea and has been used traditionally in medicines for certain ailments [66]. A total of nine potential inhibitors have been isolated from this plant, however Isocampneoside II is the most potent inhibitor in anti-cataract activities. This compound signiﬁcantly and uncompetetively inhibited RHAR activity with an IC50 value of 9.72 µM [66]. 3.4.35. Isocampneoside II .4.35. Isocampneoside II Molecular formula: C15H10O6 (286.24 g/mol), Melting point: >320 ◦C. Molecular formula: C15H10O6 (286.24 g/mol), Melting point: >320 ◦C. The potential anti-cataract effect of luteolin is well known [109,110]. As yellow crystalline, luteolin has been isolated from various plants including Platycodon grandiﬂorum (balloon ﬂower, Chinese bellﬂower) [111], Vitex negundo (Chinese chastetree, horseshoe vitex) [112], Artemisia montana [52], Perilla frutescens (L.) (perilla, Korean perilla) [69,113,114] and Sinocrassula indica (Chinese crassula) [115]. The selenite-induced oxidative stress treated group with luteolin (isolated from Vitex negundo) demonstrated 80% transparency of the lenses with minor cortical vacuolization and opacity suggesting that the anticataractogenic effect was supported by the antioxidant property based on signiﬁcant decrease in various antioxidant activities tested [112]. In comparison to the isolated luteolin from different botanicals, luteolin from Platycodon grandiﬂorum was identiﬁed as the highest inhibition with an IC50 of 0.087 µM (RLAR) [111]. The IC50 value increases slightly higher to 0.45 µM (Sinocrassula indica) in the same activity [116]. However, isolated luteolin from the same species, Perilla frutescens (L.) of different parts showed different values in RLAR (seeds, IC50 0.6 µM [113]; IC50 1.89 µM) [114] and RHAR (leaves, IC50 6.34 µM) [69], almost 10 and 3.5 times higher. Luteolin (Artemisia montana) was found to suppress RLAR activity at an IC50 0.19 µM [52]. Luteolin from Platycodon grandiﬂorum (Jacq.) exhibited comparable potent inhibitory effect of AGE (IC50 = 16.6 µM) with chlorogenic acid methyl ester (IC50 = 12.9 µM) [111]. Molecular formula: C30H48O4 (472.69 g/mol). Molecular formula: C30H48O4 (472.69 g/mol). Lucidumol A is a new triterpenoid isolated from the ethanolic extract of the fruiting body of Ganoderma lucidum (lingzhi mushroom, reishi mushroom) from a thorough fractionation process [102]. Obtained as a white amorphous powder [103], lucidumol A suppressed the strongest AR activity with an IC50 of 19.1 µM compared to all other reported isolated secondary metabolites including ganoderic acid Df (IC50 = 22.8 µM) [104], ganoderic acid C2 (IC50 = 43.8 µM) [105], ganoderol B (IC50 = 110.1 µM) [106], and others. 3.4.38. Kakkalide Molecular formula: C28H32O15 (608.549 g/mol), Melting point: 251–253 ◦C. Kakkalide was isolated from Viola hondoensis W. Becker et H Boss (ri ben qiu guo jin cai) for its AR inhibitory activity. This plant is widely distributed in southern Korea and has been used 14 of 41 Nutrients 2018, 10, 1580 to as traditional medicine in the form of expectorant. AR activity-guided isolation using column chromatography on a silica gel and gel ﬁltration column afforded kakkalide. Kakkalide signiﬁcant inhibited AR from Sprague-Dawley rat lenses at an IC50 of 0.34 µg/mL (0.56 µM), more potent than that of the positive control, tetramethylene glutaric acid (IC50 = 0.48 µg/mL) [101]. 3.4.40. Lupeol Molecular formula: C30H50O (426.72 g/mol), Melting point: 120–122 ◦C. Lupeol, a pentacyclic triterpenoid has been isolated from the ethanolic flower extract of Musa sp. var. Nanjangud rasa bale (banana) [107] and methanolic leaf extract of Vernonia cinereal (purple ﬂeabane) [108] with anti-cataractogenesis activities. Repeated silica gel chromatography after fractionation from both plants yielded lupeol as white needles. It inhibits human recombinant AR activity at IC50 of 1.53 µg/mL (3.6 µM) [52]. Similar inhibition was observed for AGE with inhibition in the range of 79–82% [107]. The potent activity of lupeol was substantiated with in vivo study using selenite-induced cataract formation in Sprague-Dawley rat pups. Lupeol attenuated the formation of vacuoles and opacity of rat pups lenses at the concentration of 25 µg/g in a dose-dependent manner in selenite-induced cataractogenesis [108]. 3.4.41. Luteolin (2-(3,4-dihydroxyphenyl)-5,7-dihydroxy-4-chromomenone) 3.4.41. Luteolin (2-(3,4-dihydroxyphenyl)-5,7-dihydroxy-4-chromomenone) Molecular formula: C21H20O11 (488.38 g/mol), Melting point: >195 ◦C. Molecular formula: C21H20O11 (488.38 g/mol), Melting point: >195 ◦C. Luteolin-7-O-β-D-glucopyranoside was isolated from the leaves extract of Stauntonia hexaphylla (Thunb.) Decne. (stauntonia vine), traditionally used as folk medicine in China, Japan and Korea. Luteolin-7-O-β-D-glucopyranoside showed the highest potent inhibitory activity of RLAR at IC50 15 of 41 15 of 41 Nutrients 2018, 10, 1580 value of 7.34 µM among other isolated secondary metabolites. Inhibition was reported 2.4 times higher compared to that of quercetin. This data was substantiated by inhibition of AGE at IC50 value of 117.8 µM and it was suggested that the presence of sugar position in ﬂavonoids enhances the activity [117]. 3.4.45. Mumeic Acid-A Molecular formula: C24H24O10 (472.44 g/mol). In an attempt to obtain inhibitors of RLAR from Prunus mume (Japanese apricot), mumeic acid-A was found to be the most potent inhibitor from all isolated secondary metabolites. The IC50 concentration of mumeic acid-A (IC50 = 0.4 µM) was almost twice that of chlorogenic acid (IC50 = 0.7 µM) as the positive control [119,120]. Molecular formula: C26H26O12 (530.50 g/mol). Molecular formula: C26H26O12 (530.50 g/mol) Methyl-3,5-di-O-caffeoylquinate or also known as 3,5-di-O-caffeoylquinic acid methyl ester was isolated from the ﬂowers of Erigeron annuus [68] and fruits of Xanthium strumarium [33] with highest inhibitory activity towards RLAR among all isolated secondary metabolites. Phytochemical analysis of ethyl acetate-soluble fraction of Erigeron annuus methanolic ﬂower extract afforded methyl-3,5-di-O-caffeoylquinate with yellow gum appearance at percentage yield of 0.0075% [68]. Isolated methyl-3,5-di-O-caffeoylquinate from both plants suppressed RLAR activity at IC50 values of 0.3 to 0.81 µM. Signiﬁcant results (most potent) were also observed with further assays in RHAR and galactitol accumulation in rat lenses ex vivo from Xanthium strumarium at IC50 value of 0.67 µM and 117 µg/lens wet weight, respectively [33]. 3.4.46. Puerariafuran Molecular formula: C16H12O5 (360.32 g/mol), Melting point: 294–296 ◦C. The roots of Pueraria lobata has for long been used in Far Eastern Asia countries as traditional medicine [57]. The isolation of root extract affords puerariafuran, a new 2-arylbenzofuran inhibited RLAR with an IC50 value of 22.2 µM, much lower than the positive control, 3,3-Tetramethyleneglutaric acid (IC50 = 28.8 µM). This data was substantiated with prevention of xylose-induced lens opacity in a dose-dependent manner, with the highest dose at 15 µM [121]. 3.4.47. Quercetin-3-O-β-D-glucoside 3.4.43. Magnoﬂorine Molecular formula: C20H24NO4+ (342.41 g/mol). Magnoﬂorine was isolated from Tinospora cordifolia (heart-leaved moonseed, guduchi, giloy) [118] and Coptidis rhizome (coptis root, huang lian) [88] for inhibitory activities against AR. Identiﬁcation of magnoﬂorine was conducted with spectroscopic analysis and compared with the literature for both plants. Appeared as yellow powder, this compound exhibited lowest concentration of maximum RLAR activity showing IC50 value at 3.6 µM from isolation of Tinospora cordifolia. Further analysis showed that magnoﬂorine inhibited 72.3% of galactose-induced polyol accumulation [118]. Nevertheless, the isolated magnoﬂorine from Coptidis rhizome possessed marginal inhibition against RLAR with 18% inhibition at a concentration of 146 µM. At this point, it is not clear if the very signiﬁcant differences in efﬁcacies and potencies were due to technical differences in isolation and/or biological assay. 3.4.49. Rhetsinine Molecular formula: C19H17N3O2 (319.36 g/mol), Melting point: 196 ◦C. Evodia rutaecarpa Bentham (Rutaceae) (wu zhu yu) is part of the Kampo-herbal medicine in Japan and has been used to relive digestion as well as painkiller. Various compounds have been isolated from this plant especially rhetsinine showed potent inhibitory activity against RLAR at an IC50 value of 24.1 µM. Rhetsinine was also reported to inhibit sorbitol accumulation in human erythrocyte by almost 79.3% at 100 µM [124]. 3.4.50. Rosmarinic Acid Molecular formula: C18H16O8 (360.32 g/mol), Melting point: 171–175 ◦C. g g p Rosmarinic acid, a conjugation of caffeic acid and 3,4-dihydroxyphenyllactic acid is mainly isolated from the family of Lamiaceae with good anti-cataract activity. Isolated rosmarinic acid from Salvia grandifolia (da ye shu wei cao) recorded the lowest IC50 (0.30 µM) in RLAR activity [125] compared to other plants. The IC50 values of rosmarinic acid isolated from other plants were 2.77 µM (Prunella vulgaris L.; woundwort, self-heal) [70], 5.38 µM (Colocasia esculenta (L.) Schott; taro) [126], and 11.2 µM (seeds, Perilla frutescens L.) [113], respectively. In RHAR assay, the activity of rosmarinic acid were shown at 2.77 µM (leaves, Perilla frutescens L.) [69] and 18.6 µM (Prunella vulgaris L.) with potent activity on galactitol accumulation in rat lenses, but low inhibitor of AGE (20.7%) for Prunella vulgaris L. [70]. 3.4.51. Scopoletin Molecular formula: C10H8O4 (192.16 g/mol), Melting point: 204–205 ◦C. The bioassay-guided fractionation of methanol extract of Magnolia fargesii (Shin-i) air-dried buds yielded ﬁve compounds with scopoletin showing the most potent in RLAR, AGE and xylose-induced lens opacity assays [127]. Scopoletin signiﬁcantly inhibits AGE formation with an IC50 value of 2.9 µM, approximately 327 times more potent compared to the positive control (IC50 = 961 µM). Similarly, in the RLAR activity, scopoletin showed marked inhibitory activity with an IC50 value of 22.5 µM. This was substantiated by the suppression of lens opacity to 72.9% (25 µM) after three days of xylose treatment [127]. A lower IC50 concentration of RLAR activity was observed for scopoletin isolated from Angelica gigas (dang gui, Korean angelica) with an IC50 value of 2.6 µM [128] showing the most potent activity among all isolated secondary metabolites. However, scopoletin from methanolic young leaves of Artemisia montana showed higher IC50 value at 64.5 µM for the same activity [52,129]. Molecular formula: C21H20O11 (448.38 g/mol), Melting point: 177–183 ◦C Molecular formula: C21H20O11 (448.38 g/mol), Melting point: 177–183 ◦C Quercitrin, a glycosylation of quercetin at C-3, was isolated from few botanicals including Smilax china L. (China root), Agrimonia pilosa Ledeb, Allium victorialis var. platyphyllum, and Melastoma sanguineum using various isolation methods. It was tested for RLAR and AGE inhibitory assays and found that quercitrin possesses signiﬁcant inhibitory actions for both activities among all isolated phytoconstituents in these plants. The IC50 values were reported at 0.17 to 0.56 µM (RLAR) and 42.0 to 58.0 µM (AGE) [99,100,123]. In contrast, quercitrin from Melastoma sanguineum showed IC50 value of 0.16 µM (RLAR) and 25.11 µM (AGE), respectively [40]. Molecular formula: C21H20O12 (464.38 g/mol). Molecular formula: C21H20O12 (464.38 g/mol). Molecular formula: C21H20O12 (464.38 g/mol). g Quercetin-3-O-β-D-glucoside was isolated from Petasites japonicus (butterbur, fuki, sweet coltsfoot) and Stauntonia hexaphylla as inhibitor for RLAR and AGE activities. It inhibited RLAR activity at IC50 Nutrients 2018, 10, 1580 16 of 41 values between 2.21 and 10.4 µM [117,122]. In contrast, its inhibition of AGE formation required a much higher concentration (IC50 = 1 mM) [117]. values between 2.21 and 10.4 µM [117,122]. In contrast, its inhibition of AGE formation required a much higher concentration (IC50 = 1 mM) [117]. 3.4.54. Syringic Acid Molecular formula: C9H10O5 (198.17 g/mol), Melting point: 206–208 ◦C. Syringic acid is a phenolic compound and a naturally occurring O-methylated trihydroxybenzoic acid monomer extracted from Herba dendrobii (shi hu). Herba dendrobii, found in the stem of many orchid species of the Dendrobium genus, has been used to improve vision centuries ago [136]. Syringic acid at medium dose (79.97%) isolated from Herba dendrobii improves survival of high-concentration D-galactose-injured human LEC with inhibition ratio of 20.3%. Rat lens turned clear to Grade 0 after 90 days of treatment. Syringic acid inhibits AR activity in a dose-dependent manner with an IC50 value of 213.17 µg/mL (1075.7 µM). Data suggest that syringic acid downregulates the expression of mRNA of AR [136]. However, the AR inhibition by syringic acid isolated from Magnolia ofﬁcinalis was weaker with less than 10% of inhibition [137]. 3.4.53. Sulfuretin and Butein Molecular formula of sulfuretin: C15H10O5 (270.24 g/mol), Melting point: 295–303 ◦C. Molecular formula of butein: C15H12O5 (272.25 g/mol), Melting point: 216 ◦C. The AR and AGE guided isolation of ethanolic bark extract of Rhus verniciﬂua (lacquer tree) produced nine secondary metabolites with sulferetin and butein as the most potent phytoconstituents for AGE and RHAR, respectively. Sulferetin was isolated as white to off-white crystalline powder [132] and inhibited against AGE activity at IC50 value of 124 µM, 11 times lower than aminoguanidine (IC50 = 1450 µM). The RHAR inhibitory activity of butein was reported at IC50 = 0.7 µM [133]. The efﬁcacies of both phytoconstituents have been suggested on the structure activity relationships of catechol moiety of the B ring and 4′-hydroxyl at the A ring for butein [134] and hydroxyl groups of ﬂavones at the 3′-, 4′-, 5′-, and 7-positions for sulferetin [135]. Molecular formula: C22H22O10 (446.40 g/mol), Melting point: 243 ◦C. Swertisin appears as pale yellow powdery crystals and isolated from Enicostemma hyssopifolium (najajihva, chota chirayita) methanol extract after repeated column chromatography over silica gel. This compound reacts with ferric chloride and turned greenish brown color as a conﬁrmation test for ﬂavonoids. RLAR activity was signiﬁcantly inhibited by swertisin at an IC50 value of 0.71 µg/mL (1.6 µM; 82.3% inhibition at 10 µg/mL) indicating a higher inhibition compared to the other compound isolated, swertiamarin (IC50 = 7.59 µg/mL). This compound was also found to suppress polyol accumulation (41.7%) in lenses cultured in a galactitol medium [138]. 3.4.52. Semilicoisoﬂavone B Molecular formula: C20H16O6 (352.34 g/mol), Melting point: 131–134 ◦C. Molecular formula: C20H16O6 (352.34 g/mol), Melting point: 131–134 ◦C. Semilicoisoﬂavone B is mostly found in roots and rhizomes of licorice species (Glycyrrhiza sp.) [130]. In searching for potential AR inhibitors, 10 secondary metabolites have been isolated 17 of 41 17 of 41 Nutrients 2018, 10, 1580 from bioactivity-guided isolation of Glycyrrhiza uralensis with semilicoisoﬂavone B showed the most potent inhibition of RLAR and RHAR activities. Both inhibition rates were recorded at IC50 values of 1.8 and 10.6 µM, respectively. Unlike γ,γ-dimethylallyl type prenylated isoﬂavonoids, semilicoisoﬂavone B containing γ,γ-dimethylchromene ring on the aromatic ring inhibited AR more strongly. In kinetic analysis of AR inhibition, semilicoisoﬂavone B did not bind to any substrate and NADPH binding regions of RHAR. Ex vivo analysis showed that this compound highly inhibited sorbitol accumulation in rat lenses incubated with high glucose by 47.0% [131]. 3.4.55. Swertisin Molecular formula: C22H22O10 (446.40 g/mol), Melting point: 243 ◦C. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. 3.4.56. Valoneic Acid Dilactone Molecular formula: C21H10O13 (470.29 g/mol), Melting point: 177–183◦C. The repeated column chromatography and preparative HPLC of seed methanolic extract of Syzygium cumini (L.) Skeels lead to the isolation of six phytoconstituents with valoneic acid dilactone showed the highest activity against RLAR inhibitory activity at IC50 value of 0.075 µM [87]. Valoneic acid dilactone were the ﬁrst constituents from this plant reported to possess RLAR inhibitory activity. 18 of 41 19 of 42 18 of 41 19 of 42 Nutrients 2018, 10, 1580 Nutrients 2018, 10 Act Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR 1-O-galloyl-β-D-glucose (β-Glucogallin) Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR 1-O-galloyl-β-D-glucose (β- Glucogallin) NA NA NA 17.00 μM [30] NA NA NA 1,3-di-O-caffeoylquinic acid 24.85 μM [32] NA NA 0.810 μM [33] NA NA 0.22 μM [32] 1,5-Di-hydroxy-1,5-di-[(E)- 3-(4-hydroxyphenyl)-2- propenoic]-3-pentanonyl ester (DHDP) NA NA NA 194.67μ M [34] NA NA NA 1,5-di-O-caffeoylquinic acid NA NA NA NA NA NA 2.98 μM [35] 1 3 6 trihydroxy 2 O H O H O O O O H OH OH O H O H O CH2OCH3 OH O H O O OH O OH O OH O O O H O H OH OH O COOH O OH OH O O H O H O O O OH OH O H OH OH NA NA NA 17.00 µM [30] NA NA NA 1,3-di-O-caffeoylquinic acid Active Ingredient Structure AGE ARI GLWW RHAR BLAR HLAR RLAR 1-O-galloyl-β-D-glucose (β- Glucogallin) NA NA NA 17.00 μM [30] NA NA NA 1,3-di-O-caffeoylquinic acid 24.85 μM [32] NA NA 0.810 μM [33] NA NA 0.22 μM [32] 1,5-Di-hydroxy-1,5-di-[(E)- 3-(4-hydroxyphenyl)-2- propenoic]-3-pentanonyl ester (DHDP) NA NA NA 194.67μ M [34] NA NA NA 1,5-di-O-caffeoylquinic acid NA NA NA NA NA NA 2.98 μM [35] 1,3,6-trihydroxy-2- methoxymethylanthraquin 52.72 NA NA NA NA NA 3.04 μM O H O H O O O O H OH OH O H O H O CH2OCH3 OH O H O O OH O OH O OH O O O H O H OH OH O COOH O OH OH O O H O H O O O OH OH O H OH OH 24.85 µM [32] NA NA 0.810 µM [33] NA NA 0.22 µM [32] 1,5-Di-hydroxy-1,5-di-[(E)-3- (4-hydroxyphenyl)-2-propenoic] -3-pentanonyl ester (DHDP) 1-O-galloyl-β-D-glucose (β- Glucogallin) NA NA NA 17.00 μM [30] NA NA NA 1,3-di-O-caffeoylquinic acid 24.85 μM [32] NA NA 0.810 μM [33] NA NA 0.22 μM [32] 1,5-Di-hydroxy-1,5-di-[(E)- 3-(4-hydroxyphenyl)-2- propenoic]-3-pentanonyl ester (DHDP) NA NA NA 194.67μ M [34] NA NA NA 1,5-di-O-caffeoylquinic acid NA NA NA NA NA NA 2.98 μM [35] 1,3,6-trihydroxy-2- methoxymethylanthraquin one 52.72 μM [36] NA NA NA NA NA 3.04 μM [36] O H O H O O O O H OH OH O H O H O O H OH CH2OCH3 OH O H O O OH O OH O OH O O O H O H OH OH O COOH O OH OH O O H O H O O O OH OH O H OH OH NA NA NA 194.67µM [34] NA NA NA 1,5-di-O-caffeoylquinic acid 1-O-galloyl-β-D-glucose (β- Glucogallin) NA NA NA 17.00 μM [30] NA NA NA 1,3-di-O-caffeoylquinic acid 24.85 μM [32] NA NA 0.810 μM [33] NA NA 0.22 μM [32] 1,5-Di-hydroxy-1,5-di-[(E)- 3-(4-hydroxyphenyl)-2- propenoic]-3-pentanonyl ester (DHDP) NA NA NA 194.67μ M [34] NA NA NA 1,5-di-O-caffeoylquinic acid NA NA NA NA NA NA 2.98 μM [35] 1,3,6-trihydroxy-2- methoxymethylanthraquin one 52.72 μM [36] NA NA NA NA NA 3.04 μM [36] O H O H O O O O H OH O H O O O H OH CH2OCH3 OH O H O O OH O OH O OH O O O H O H OH OH O COOH O OH OH O O H O H O O O OH OH O H OH OH NA NA NA NA NA NA 2.98 µM [35] 1,3,6-trihydroxy-2- methoxymethylanthraquinone 1-O-galloyl-β-D-glucose (β- Glucogallin) NA NA NA 17.00 μM [30] NA NA NA 1,3-di-O-caffeoylquinic acid 24.85 μM [32] NA NA 0.810 μM [33] NA NA 0.22 μM [32] 1,5-Di-hydroxy-1,5-di-[(E)- 3-(4-hydroxyphenyl)-2- propenoic]-3-pentanonyl ester (DHDP) NA NA NA 194.67μ M [34] NA NA NA 1,5-di-O-caffeoylquinic acid NA NA NA NA NA NA 2.98 μM [35] 1,3,6-trihydroxy-2- methoxymethylanthraquin one 52.72 μM [36] NA NA NA NA NA 3.04 μM [36] O H O O O O H OH O H O O O H OH CH2OCH3 OH O H O O OH O OH O OH O O O H O H OH OH O COOH O OH OH O O H O H O O O OH OH O H OH OH 52.72 µM [36] NA NA NA NA NA 3.04 µM [36] O H O H O O O O H OH OH O H O H O H O H O O O O H OH O H O H O O O O H OH O H O O H O H O OH O H O H O COOH O OH OH 1-O-galloyl-β-D-glucose (β-Glucogallin) 1-O-galloyl-β-D-gluco Glucogallin) g y β g Glucogallin) O H H O O H OH O COOH O OH OH O O H O H OH OH O COOH O OH OH O O H O H OH OH O COOH OH O OH O OH O O O H O H OH OH O O O OH O OH O O H O H OH OH COOHOH H 1,5-Di-hydroxy-1,5-di-[(E 1,5-Di-hydroxy-1,5-di-[(E 3-(4-hydroxyphenyl)-2 1,5-Di-hydroxy-1,5-di-[(E)-3- (4-hydroxyphenyl)-2-propenoic] -3-pentanonyl ester (DHDP) 1,5-Di-hydroxy-1,5-di-[( 3-(4-hydroxyphenyl)-2 propenoic]-3-pentanon 3 (4 hydroxyphenyl) 2 propenoic]-3-pentanon ester (DHDP) propenoic]-3-pentanony ester (DHDP) O OH O OH O O H OH O O O OH O OH O O H O O OH O OH O OH O O H O O OH O H OH O O H O O OH O O O H O H OH O O O O H O O H O O O OH O H OH O O H O H O O O OH OH O H OH OH O O H O H O O OH OH O H OH OH p p ] p ester (DHDP) ( ) 1,5-di-O-caffeoylquinic acid 1,5-di-O-caffeoylquinic 1,5-di-O-caffeoylquinic O OH O O H O H O OH OH OH OH O CH2OCH3 OH O O H O H OH OH OH OH O CH2OCH3 OH O H OH OH O O H OH CH2OCH3 OH O O O H OH CH2OCH3 OH , y q 1 3 6 trihydroxy 2 1,3,6-trihydroxy-2- methoxymethylanthraqu 1,3,6-trihydroxy-2- methoxymethylanthraquinone 1,3,6-trihydroxy-2- methoxymethylanthraqui one 19 of 41 19 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR 1,2,3,6-tetra-O-galloyl-β-D-glucose Nutrients 2018, 10, x FOR PEER REVIEW 20 of 42 1,2,3,6-tetra-O-galloyl-β-D- glucose 1.99 μM [38] NA NA NA NA NA 0.70 μM [38] 1,3,5,8- Tetrahydroxyxanthone NA NA NA NA NA NA 0.0886 μM [39] 2″,4″-O-Diacetylquercitrin 11.46 μM [40] NA NA NA NA NA 0.077 μM [40] 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O O O O H O-gal gal-O gal gal O OH OH O O O H O H O O O CH3 O H O O C H3 CH3 O O OH O O H H3CO CH3 CH3 C H3 CH3 O OH O OH OH O H 1.99 µM [38] NA NA NA NA NA 0.70 µM [38] 1,3,5,8-Tetrahydroxyxanthone Nutrients 2018, 10, x FOR PEER REVIEW 20 of 42 1,2,3,6-tetra-O-galloyl-β-D- glucose 1.99 μM [38] NA NA NA NA NA 0.70 μM [38] 1,3,5,8- Tetrahydroxyxanthone NA NA NA NA NA NA 0.0886 μM [39] 2″,4″-O-Diacetylquercitrin 11.46 μM [40] NA NA NA NA NA 0.077 μM [40] 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O O O O H O-gal gal-O gal gal O OH OH O O O H O H O O O CH3 O H O O C H3 CH3 O O OH O O H H3CO CH3 CH3 C H3 CH3 O OH O OH OH O H NA NA NA NA NA NA 0.0886 µM [39] 2′′,4′′-O-Diacetylquercitrin Nutrients 2018, 10, x FOR PEER REVIEW 20 of 42 1,2,3,6-tetra-O-galloyl-β-D- glucose 1.99 μM [38] NA NA NA NA NA 0.70 μM [38] 1,3,5,8- Tetrahydroxyxanthone NA NA NA NA NA NA 0.0886 μM [39] 2″,4″-O-Diacetylquercitrin 11.46 μM [40] NA NA NA NA NA 0.077 μM [40] 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O O O O H O-gal gal-O gal gal O OH OH O O O H O H O O O CH3 O H O O C H3 CH3 O O OH O O H H3CO CH3 CH3 C H3 CH3 O OH O OH OH O H 11.46 µM [40] NA NA NA NA NA 0.077 µM [40] 3-Isomangostin 1,2,3,6-tetra-O-galloyl-β-D- glucose 1.99 μM [38] NA NA NA NA NA 0.70 μM [38] 1,3,5,8- Tetrahydroxyxanthone NA NA NA NA NA NA 0.0886 μM [39] 2″,4″-O-Diacetylquercitrin 11.46 μM [40] NA NA NA NA NA 0.077 μM [40] 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O O O O H O-gal gal-O gal gal O OH OH O O O H O H O O O CH3 O H O O C H3 CH3 O O OH O O H H3CO CH3 CH3 C H3 CH3 O OH O OH OH O H NA NA NA NA NA NA 3.48 µM [41] Active Ingredient ts 2018, 10, x FOR PEER Structure O-gal O O O O O H O-gal gal-O gal gal O O O O O H O-gal gal-O gal gal O O O O O H g gal-O gal gal O O O O O H gal-O gal gal OH O HO H O OH O OH OH O H O OH O OH OH O H O OH O OH OH O H OH O OH O OH O H OH OH OH OH OH 2″,4″-O-Diacetylquerc 2″,4″-O-Diacetylquerc 2′′,4′′-O-Diacetylquercitrin 2″,4″-O-Diacetylquerc 2″,4″-O-Diacetylquerc 11 46 11.46 11.46 µM μM [40] 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] C H3 O O OH O O H H3CO CH3 C H3 CH3 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O OH O O H H3CO CH CH3 C H3 CH3 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O OH O O H H3CO CH3 CH3 C H3 CH3 3-Isomangostin 3-Isomangostin NA NA NA NA NA NA 3.48 μM [41] O O OH O O H H3CO CH3 CH3 C H3 CH3 NA NA NA NA NA NA 3.48 µM [41] OH O H3CO CH C H3 CH3 OH O H3CO CH3 C H3 CH3 O O OH O O H H3CO CH3 C H3 CH3 O O OH O O H H3CO CH3 CH3 C H3 CH3 3 I 3-Isomangostin 3-Isomango H3CO 20 of 41 20 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR 3′,5′-dimethoxy-(1,1′-biphenyl) -3,4-diol 3-O-β-D-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 21 of 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D- glucopyranoside NA NA NA NA NA NA 3.80 μM [47] 3,5-di-O-caffeoylquinic Acid 6.06 μM [48] NA 153 g [33] 1.34 μM [33] NA NA 0.19 μM [33] 4-O-butylpaeoniflorin 10.80 μM [53] NA NA NA NA NA 36.20 μM [53] O H OH O O O H OH O OH HOOC O H O O O O O O C H3 O O OH O H O H O H C H3 OCH3 OCH3 O H O O CH3 OH O H OH NA NA NA NA NA NA 3.80 µM [47] 3,5-di-O-caffeoylquinic Acid Nutrients 2018, 10, x FOR PEER REVIEW 21 of 4 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D- glucopyranoside NA NA NA NA NA NA 3.80 μM [47] 3,5-di-O-caffeoylquinic Acid 6.06 μM [48] NA 153 g [33] 1.34 μM [33] NA NA 0.19 μM [33] 4-O-butylpaeoniflorin 10.80 μM [53] NA NA NA NA NA 36.20 μM [53] O H OH O O O H OH O OH HOOC O H O O O O O O C H3 O O OH O H O H O H C H3 OCH3 OCH3 O H O O CH3 OH O H OH 6.06 µM [48] NA 153 g [33] 1.34 µM [33] NA NA 0.19 µM [33] 4-O-butylpaeoniﬂorin 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D- glucopyranoside NA NA NA NA NA NA 3.80 μM [47] 3,5-di-O-caffeoylquinic Acid 6.06 μM [48] NA 153 g [33] 1.34 μM [33] NA NA 0.19 μM [33] 4-O-butylpaeoniflorin 10.80 μM [53] NA NA NA NA NA 36.20 μM [53] O H OH O O O H OH O OH HOOC O H O O O O O O C H3 O O OH O H O H O H C H3 OCH3 OCH3 O H O O CH3 OH O H OH 10.80 µM [53] NA NA NA NA NA 36.20 µM [53] Active Ingredient ts 2018, 10, x FOR PEER Active Ingredient ts 2018, 10, x FOR PEER Structure Active Ingredient Structure 3′,5′-dimethoxy-(1,1′-biphenyl) -3,4-diol 3-O-β-D-glucopyranoside 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D- glucopyranoside OCH3 OCH3 O H O O CH3 OH O H OH 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D- glucopyranoside OCH3 OCH3 O H O O CH3 OH O H OH 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D- glucopyranoside OCH3 OCH3 O H O O CH3 OH O H OH 3′,5′-dimethoxy-(1,1′-biphenyl) -3,4-diol 3-O-β-D-glucopyranoside 3′,5′-dimethoxy-(1,1′- biphenyl)-3,4-diol 3-O-β-D glucopyranoside biphenyl)-3,4-diol 3-O-β-D glucopyranoside biphenyl) 3,4 diol 3 O β D glucopyranoside O H OH O O O H OH OH HOOC O H O O H OH O O O H OH O OH HOOC O H O O H O O O H OH O OH HOOC O H O 6.06 μM [48] 6.06 μM [48] 6.06 µM [48] 6.06 μM [48] O H HOOC O H HOOC OH 4-O-butylpaeoniflorin 10.80 μM [53] NA NA NA NA NA 36.20 μM [53] O O O O O O C H O O OH O H O H O H C H3 4-O-butylpaeoniflorin 10.80 μM [53] NA NA NA NA NA 36.20 μM [53] O O O O O C H3 O O OH O H O H O H C H3 4-O-butylpaeoniﬂorin 4-O-butylpaeoniflorin 10.80 μM [53] NA NA NA NA NA 36.20 μM [53] O O O O O C H3 O O OH O H O H O H C H3 10.80 µM [53] NA NA NA NA NA 36.20 µM [53] O H O OH 10.80 10.80 μM [53] 10.80 µM [53] 10.80 10.80 µM 4-O-butylpaeonifl 4-O-butylpaeoniﬂorin 4-O-butylpaeonifl C H3 21 of 41 21 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR 4,5-Di-O-trans-caffeoyl-D-quinic acid Nutrients 2018, 10, x FOR PEER REVIEW 22 of 42 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH O OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA NA NA NA 0.29 µM [55] 5-O-Feruloly quinic acid Nutrients 2018, 10, x FOR PEER REVIEW 22 of 42 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA NA NA NA 14.19 µM [56] 5,7,4′-trihydroxyisoﬂavone (Genistein) Nutrients 2018, 10, x FOR PEER REVIEW 22 of 42 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA NA NA NA 9.48 µM [60] 20(S)-Ginsenoside Rh2 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA 147.40 µM [61] NA NA NA Acteoside 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H O OH O O H O H OH O O H OH O OH 5.11 µM [63] NA NA NA NA NA 0.83 µM [63] Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR 4,5-Di-O-trans-caffeoyl-D-quinic acid Nutrients 2018, 10, x FOR PEER REVIEW 22 of 42 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O OH O H O OH OH OH O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA NA NA NA 0.29 µM [55] 5-O-Feruloly quinic acid Nutrients 2018, 10, x FOR PEER REVIEW 22 of 42 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH O OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA NA NA NA 14.19 µM [56] 5,7,4′-trihydroxyisoﬂavone (Genistein) Nutrients 20 8, 0, O EE EVIEW o 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA NA NA NA 9.48 µM [60] 20(S)-Ginsenoside Rh2 4,5-Di-O-trans-caffeoyl-D- quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H COOH OH O O OH O O H O H O O H OH O O H OH O OH NA NA NA 147.40 µM [61] NA NA NA Acteoside 4,5 Di O trans caffeoyl D quinic acid NA NA NA NA NA NA 0.29 μM [55] 5-O-Feruloly quinic acid NA NA NA NA NA NA 14.19 μM [56] 5,7,4′-trihydroxyisoflavone (Genistein) NA NA NA NA NA NA 9.48 μM [60] 20(S)-Ginsenoside Rh2 NA NA NA 147.40 μM [61] NA NA NA Acteoside 5.11 μM [63] NA NA NA NA NA 0.83 μM [63] O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 OH OCH 3 O O HOOC O H O H O H OH O H O H OH O O H OH O OH 5.11 µM [63] NA NA NA NA NA 0.83 µM [63] Active Ingredient utrients 2018, 10, x FOR PE Nutrients 2018, 10, x FOR P quinic acid Active Ingredient utrients 2018, 10, x FOR PE Nutrients 2018, 10, x FOR P quinic acid Structure COOH OH COO O H 4,5-Di-O-trans-caffeoyl-D-quinic acid 4,5-Di-O-trans-caffeoyl-D- quinic acid COOH OH O O OH O O H O H O O H OH 4,5-Di-O-trans-caffeoyl-D- quinic acid COOH O O OH O O H O H O O H OH 4,5-Di-O-trans-caffeoyl-D- quinic acid OH O O OH O O H O H O O H OH 4,5-Di-O-trans-caffeoyl-D- quinic acid OH O O H O OH O O H O H OH 5-O-Feruloly quinic acid OH OCH 3 O O HOOC O H O H O H O H OH COOH OH COOH O O OH OCH 3 O O HOOC O H O H OH OCH 3 O O HOOC O H O H O H OCH 3 O O HOOC O H O H O H OCH 3 O HOOC O H O H O O H O O H 5 O Feruloly quin 5-O-Feruloly quinic acid 5-O-Feruloly qui 5-O-Feruloly quin ′ h d fl 5 O Feruloly quin 5-O-Feruloly quinic acid 5-O-Feruloly qui 5-O-Feruloly quin ′ h d fl O O H O O H OH O O O H OH O OH OH O OH O H C H3 OH O OH OO 5 7 4′ trihydroxyiso 5,7,4′-trihydroxyiso 5,7,4′-trihydroxyisoﬂavone (Genistein) 5,7,4′-trihydroxyisof (Genistein) y y (Genistein) 5 7 4′ trihyd 5,7,4′-trihy (Genistein) (Ge O H CH3 CH3 OH CH3 CH3 C H3 OH O OH O H CH3 CH3 OH CH3 CH3 CH C H3 OH O H CH3 CH3 OH CH3 CH3 CH3 Glc O C H3 O H CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O C H3 CH3 CH3 OH CH3 CH3 CH3 CH3 C H3 Glc-O 20(S) Gi i 20(S)-Ginsenoside Rh2 20(S)-Ginsenosi CH3 CH3 C H3 Glc-O O OH O H OH OH CH3 C H3 Glc-O O O OH O H OH OH OH O H CH3 CH3 C H3 O O O OH O H O OH OH OH O H CH3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 O O O O OH O H O OH OH O OH OH OH O H O H CH3 c OC H c-O C 3 OH Ac Acteoside Acteos 22 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont.NA Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Basilicumin [7-(3-hydroxypropyl) -3-methyl-8-β-O-D-glucoside-2H -chromen-2-one] Nutrients 2018, 10, x FOR PEER REVIEW 23 of 42 Basilicumin [7-(3- hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O O O H OH OH O H CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH NA NA NA NA 2.09 µM NA NA Caffeic acid Nutrients 2018, 10, x FOR PEER REVIEW 23 of 42 Basilicumin [7-(3- hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O O O H OH OH O H CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH 7.56 µM [68] NA NA 210.28µM [66] NA NA 16.71 µM [65] Canangafruiticoside E Basilicumin [7-(3- hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O O O H OH OH O H CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 µM [71] Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetraenoyl) -2-O-(4Z,10Z,13Zhexadecatetraenoyl) -3-O-β-D-galactopyranosyl glycerol)] 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA N Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16 μM Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 (7 Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52 μM Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 [6 O OH O H O H O O O OH OH O H O H O C H3 O O H O CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH NA NA NA NA NA NA 52.53 µM [72] Davallialactone hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O O H CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH NA NA NA 0.56 µM [67] NA NA 0.33 µM [67] Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Basilicumin [7-(3-hydroxypropyl) -3-methyl-8-β-O-D-glucoside-2H -chromen-2-one] Nutrients 2018, 10, x FOR PEER REVIEW 23 of 42 Basilicumin [7-(3- hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H O H O C H3 O O H O O O O H OH OH O H CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH NA NA NA NA 2.09 µM NA NA Caffeic acid Nutrients 2018, 10, x FOR PEER REVIEW 23 of 42 Basilicumin [7-(3- hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O O O H OH OH O H CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH 7.56 µM [68] NA NA 210.28µM [66] NA NA 16.71 µM [65] Canangafruiticoside E Basilicumin [7-(3- hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O O O H OH OH CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 µM [71] Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetraenoyl) -2-O-(4Z,10Z,13Zhexadecatetraenoyl) -3-O-β-D-galactopyranosyl glycerol)] 8-β-O-D-glucoside-2H- chromen-2-one] μ Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16 μM Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 (7 Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52 μM Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 [6 O OH O H O H O O O OH OH O H O H O C H3 O O H O CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH NA NA NA NA NA NA 52.53 µM [72] Davallialactone hydroxypropyl)-3-methyl- 8-β-O-D-glucoside-2H- chromen-2-one] NA NA NA NA 2.09 μM NA NA Caffeic acid 7.56 μM [68] NA NA 210.28μ M [66] NA NA 16.71 μM [65] Canangafruiticoside E Glc=β-D-glucopyranoside NA NA NA NA NA NA 0.80 μM (71) Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] NA NA NA NA NA NA 52.53 μM [72] Davallialactone NA NA NA 0.56 μM [67] NA NA 0.33 μM [67] O OH O H O H O O O OH OH O H OH O H O C H3 O O H O O CH3 O CHO O-Glc OH O O H O H O O O CH3 O O CH3 O O H OH OH OH NA NA NA 0.56 µM [67] NA NA 0.33 µM [67] Active Ingredient Nutrients 2018, 10, x FOR Basilicumin [ chromen-2-one 8-β-O-D-glucosid h 2 O O H O O O O H OH OH O H CH3 yl- O O H O O O O H OH OH O H CH3 O O O H O CH3 O O O CH3 O OH O H CH3 Basilicumin [7-(3-hydroxypropyl) -3-methyl-8-β-O-D-glucoside-2H -chromen-2-one] Basilicumin [7-(3 hydroxypropyl)-3-me 8-β-O-D-glucoside-2 chromen-2-one] Basilicumin hydroxypropyl 8-β-O-D-gluco chromen-2 chromen 2 one] O OH O H O OH O H O H OH O O O O O O Caffeic acid O H O CHO O-Glc O O H O H e E Glc=β-D-glucopyranoside N )-l-O- catetr O CHO O-Glc OH O O H O H O Glc=β-D-glucopyranoside O CHO O-Glc OH O O H O H Glc=β-D-glucopyranoside Glc=β-D-glucopyranoside O CHO O-Gl OH O H Glc=β-D-glucopyranoside O CHO O-Glc OH O H O O O CHO O O H O Glc=β-D-glucopyranoside Glc=β-D-glucopyranoside O Glc=β-D-glucopyranoside esin A [((2S)-l-O- ,15Zoctadecatetr oyl)-2-O- OH O CH3 O O CH3 O OH OH , , , aenoyl)-2-O- Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- lactopyranosyl glycerol)] NA O O O C H3 O O O CH3 O CH3 O O H OH OH OH A [((2S)-l-O- octadecatetr 2-O- xadecatetrae O-β-D- O O O CH3 O O CH3 O O H OH OH OH S)-l-O- etraenoyl) atetraenoyl) nosyl (( ) adecatetr O- ecatetrae -D- O O O CH3 O O CH3 O O H OH OH OH ctadecatetr 2-O- adecatetrae -β-D- yl glycerol)] O O O CH3 O CH3 O O H OH OH OH Capsofulvesin A [((2S)- (6Z,9Z,12Z,15Zoctadeca aenoyl)-2-O- (4Z 10Z 13Zh d aenoyl)-2-O (4Z,10Z,13Zhexad noyl)-3-O-β galactopyranosyl g Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- Capsofulvesin A [((2S)-l-O- (6Z,9Z,12Z,15Zoctadecatetraenoyl) -2-O-(4Z,10Z,13Zhexadecatetraenoyl) -3-O-β-D-galactopyranosyl glycerol)] p (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- (6Z,9Z,12Z,15Zoctadecatetr aenoyl)-2-O- (4Z,10Z,13Zhexadecatetrae noyl)-3-O-β-D- galactopyranosyl glycerol)] ae l)] O O O OH O H O C H3 O O H OH ne NA OH OH O H OH O H O O O OH OH O H O H O C H3 O O O OH O H O C H3 O O O OH OH O H OH O H O C H3 Davallialactone Davallialact H O OH 23 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Delphinidin 3-O-β-galactopyranoside -3′-O-β-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 24 of 42 Delphinidin 3-O-β- galactopyranoside-3′-O-β- glucopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 μM [83] Desmethylanhydroicaritin 294.60 μM [84] NA NA 0.45 μM [84] NA NA 0.95 μM [84] Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O + OH OH O-gal O-glc OH O H O O OH OH O O H O H O O OH OH OH OH O H CH3 C H3 Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 µM [83] Desmethylanhydroicaritin Nutrients 2018, 10, x FOR PEER REVIEW 24 of 42 Delphinidin 3-O-β- galactopyranoside-3′-O-β- glucopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 μM [83] Desmethylanhydroicaritin 294.60 μM [84] NA NA 0.45 μM [84] NA NA 0.95 μM [84] Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O + OH OH O-gal O-glc OH O H O O OH OH O O H O H O O OH OH OH OH O H CH3 C H3 294.60 µM [84] NA NA 0.45 µM [84] NA NA 0.95 µM [84] Ellagic acid Nutrients 2018, 10, x FOR PEER REVIEW 24 of 42 Delphinidin 3-O-β- galactopyranoside-3′-O-β- glucopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 μM [83] Desmethylanhydroicaritin 294.60 μM [84] NA NA 0.45 μM [84] NA NA 0.95 μM [84] Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O + OH OH O-gal O-glc OH O H O O OH OH O O H O H O O OH OH OH OH O H CH3 C H3 26.0 µM [86] NA 42.47% [85] NA NA 1.37 µM [67] 0.12 µM [87] Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Delphinidin 3-O-β-galactopyranoside -3′-O-β-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 24 of 42 Delphinidin 3-O-β- galactopyranoside-3′-O-β- glucopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 μM [83] Desmethylanhydroicaritin 294.60 μM [84] NA NA 0.45 μM [84] NA NA 0.95 μM [84] Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O + OH OH O-gal O-glc OH O H O O OH OH O O H O H O O OH OH OH OH O H CH3 C H3 Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 µM [83] Desmethylanhydroicaritin Nutrients 2018, 10, x FOR PEER REVIEW 24 of 42 Delphinidin 3-O-β- galactopyranoside-3′-O-β- glucopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 μM [83] Desmethylanhydroicaritin 294.60 μM [84] NA NA 0.45 μM [84] NA NA 0.95 μM [84] Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O + OH OH O-gal O-glc OH O H O O OH OH O O H O H O O OH OH OH OH O H CH3 C H3 294.60 µM [84] NA NA 0.45 µM [84] NA NA 0.95 µM [84] Ellagic acid Nutrients 2018, 10, x FOR PEER REVIEW 24 of 42 Delphinidin 3-O-β- galactopyranoside-3′-O-β- glucopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside NA NA NA NA NA NA 0.37 μM [83] Desmethylanhydroicaritin 294.60 μM [84] NA NA 0.45 μM [84] NA NA 0.95 μM [84] Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O + OH OH O-gal O-glc OH O H O O OH OH O O H O H O O OH OH OH OH O H CH3 C H3 26.0 µM [86] NA 42.47% [85] NA NA 1.37 µM [67] 0.12 µM [87] Structure Glc= β-glucopyranoside, Gal= β-galactopyranoside O + OH OH O-gal O-glc OH O H Glc= β-glucopyranoside, Gal= β-galactopyranoside Glc= β-glucopyranoside, Gal= β-galactopyranoside O + OH OH O-gal O-glc OH O H Glc= β-glucopyranoside, Gal= β-galactopyranoside O + OH OH O-gal O-glc OH O H CH3 C H3 Delphinidin 3-O-β-galactopyranoside -3′-O-β-glucopyranoside Delphinidin 3-O-β galactopyranoside-3′- glucopyranoside Delphinidin 3-O-β galactopyranoside-3′-O glucopyranoside galactopyranoside-3′- glucopyranoside Glc= β-glucopyranoside Gal= β-galactopyranosid Glc= β-glucopyranoside, Gal= β-galactopyranoside Glc= β-glucopyranoside Gal= β-galactopyranosid Gal= β-galactopyranosid CH3 C O OH OH OH OH O H CH3 C H3 O OH OH OH OH O H CH3 C H3 O OH OH OH OH O H 294.60 μM [84] 294.60 μM [84] 294.60 µM [84] 294.60 μM [84] Desmethylanhydroica Desmethylanhydroicaritin Desmethylanhydroica Desmethylanhydroica Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O O OH OH O O H O H O Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O O OH OH O O H O H O Ellagic acid Ellagic acid 26.0 μM [86] NA 42.47% [85] NA NA 1.37 μM [67] 0.12 μM [87] O O OH OH O O H O H O 26.0 µM [86] NA 42.47% [85] NA NA 1.37 µM [67] 0.12 µM [87] O O H 26.0 μM 26.0 μM 26.0 μM [86] 26.0 µM [86] Ellagic acid Ellagic a O 24 of 41 24 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Epiberberine Nutrients 2018, 10, x FOR PEER REVIEW 25 of 42 Epiberberine NA NA NA 168.10 μM [88] NA NA 100.07 μM [88] Geraniin 21.00 μM 96% * [89] 0.15 μM [89] 39.87% [85] NA NA NA NA Hipolon NA NA NA NA NA NA 9.47 μM [90] O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O N + O O H3CO H3CO CH3 O OH O H NA NA NA 168.10 µM [88] NA NA 100.07 µM [88] Geraniin Nutrients 2018, 10, x FOR PEER REVIEW 25 of 42 Epiberberine NA NA NA 168.10 μM [88] NA NA 100.07 μM [88] Geraniin 21.00 μM 96% * [89] 0.15 μM [89] 39.87% [85] NA NA NA NA Hipolon NA NA NA NA NA NA 9.47 μM [90] O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O N + O O H3CO H3CO CH3 O OH OH O H 21.00 µM 96% * [89] 0.15 µM [89] 39.87% [85] NA NA NA NA Hipolon Epiberberine NA NA NA 168.10 μM [88] NA NA 100.07 μM [88] Geraniin 21.00 μM 96% * [89] 0.15 μM [89] 39.87% [85] NA NA NA NA Hipolon NA NA NA NA NA NA 9.47 μM [90] OH OH O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O N + O O H3CO H3CO CH3 O OH OH O H NA NA NA NA NA NA 9.47 µM [90] Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Epiberberine Nutrients 2018, 10, x FOR PEER REVIEW 25 of 42 Epiberberine NA NA NA 168.10 μM [88] NA NA 100.07 μM [88] Geraniin 21.00 μM 96% * [89] 0.15 μM [89] 39.87% [85] NA NA NA NA Hipolon NA NA NA NA NA NA 9.47 μM [90] O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O N + O O H3CO H3CO CH3 O OH O H NA NA NA 168.10 µM [88] NA NA 100.07 µM [88] Geraniin , , Epiberberine NA NA NA 168.10 μM [88] NA NA 100.07 μM [88] Geraniin 21.00 μM 96% * [89] 0.15 μM [89] 39.87% [85] NA NA NA NA Hipolon NA NA NA NA NA NA 9.47 μM [90] O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O N + O O H3CO H3CO CH3 O OH OH O H 21.00 µM 96% * [89] 0.15 µM [89] 39.87% [85] NA NA NA NA Hipolon Epiberberine NA NA NA 168.10 μM [88] NA NA 100.07 μM [88] Geraniin 21.00 μM 96% * [89] 0.15 μM [89] 39.87% [85] NA NA NA NA Hipolon NA NA NA NA NA NA 9.47 μM [90] OH O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O N + O O H3CO H3CO CH3 O OH OH O H NA NA NA NA NA NA 9.47 µM [90] Structure + Structure + N + O O H3CO H3CO N + O O H3CO OH O H O O H3CO Epiberberine Epiberbe p O O OH OH O H O O OH O H O H O H O H O H O O O OH O H OH O O O O O O O OH OH O H O O OH O H O H O H O H O H O O O O H OH OH O H O H OH O O O O O O O OH OH O H O O O H O H O H O H O O O O H OH OH O H O H OH O O O O O O H Geraniin Geran H O H H O H OH O H CH3 O OH CH3 O OH CH3 O OH OH O H Hipolon Hipol O H 25 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H NA Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Hirsutrin p [90] Hirsutrin NA NA 33.78% [91] NA NA NA 4.78 μM [91] O O OH O H OH O O OH OH OH CH3 OH OH O H NA NA 33.78% [91] NA NA NA 4.78 µM [91] Hopeafuran Nutrients 2018, 10, x FOR PEER REVIEW 26 of 42 Hopeafuran NA NA NA NA NA NA 14.80 μM [92] Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl-(1 →2)]-6′′-O-acetyl-β-D- glucopyranoside NA 0.66 μM [93] NA NA NA NA NA Isocampneoside II NA NA NA 9.72 μM [66] NA NA NA OH O O H OH O H OH O O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH NA NA NA NA NA NA 14.80 µM [92] Hypolaetin 7-O-[6′′′-O-acetyl- β-D-allopyranosyl-(1→2)]-6′′- O-acetyl-β-D-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 26 of 42 Hopeafuran NA NA NA NA NA NA 14.80 μM [92] Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl-(1 →2)]-6′′-O-acetyl-β-D- glucopyranoside NA 0.66 μM [93] NA NA NA NA NA Isocampneoside II NA NA NA 9.72 μM [66] NA NA NA OH O O H OH O H OH O O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH NA 0.66 µM [93] NA NA NA NA NA Isocampneoside II Nutrients 2018, 10, x FOR PEER REVIEW 26 of 4 Hopeafuran NA NA NA NA NA NA 14.80 μM [92] Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl-(1 →2)]-6′′-O-acetyl-β-D- glucopyranoside NA 0.66 μM [93] NA NA NA NA NA Isocampneoside II NA NA NA 9.72 μM [66] NA NA NA OH O O H OH O H OH O O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH NA NA NA 9.72 µM [66] NA NA NA Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Hirsutrin [90] Hirsutrin NA NA 33.78% [91] NA NA NA 4.78 μM [91] O O OH O H OH O O OH OH OH CH3 OH OH O H NA NA 33.78% [91] NA NA NA 4.78 µM [91] Hopeafuran Nutrients 2018, 10, x FOR PEER REVIEW 26 of 42 Hopeafuran NA NA NA NA NA NA 14.80 μM [92] Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl-(1 →2)]-6′′-O-acetyl-β-D- glucopyranoside NA 0.66 μM [93] NA NA NA NA NA Isocampneoside II NA NA NA 9.72 μM [66] NA NA NA OH O O H OH O H OH O O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH NA NA NA NA NA NA 14.80 µM [92] Hypolaetin 7-O-[6′′′-O-acetyl- β-D-allopyranosyl-(1→2)]-6′′- O-acetyl-β-D-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 26 of 42 Hopeafuran NA NA NA NA NA NA 14.80 μM [92] Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl-(1 →2)]-6′′-O-acetyl-β-D- glucopyranoside NA 0.66 μM [93] NA NA NA NA NA Isocampneoside II NA NA NA 9.72 μM [66] NA NA NA OH O O H OH O H OH O O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH NA 0.66 µM [93] NA NA NA NA NA Isocampneoside II Nutrients 2018, 10, x FOR PEER REVIEW 26 of 4 Hopeafuran NA NA NA NA NA NA 14.80 μM [92] Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl-(1 →2)]-6′′-O-acetyl-β-D- glucopyranoside NA 0.66 μM [93] NA NA NA NA NA Isocampneoside II NA NA NA 9.72 μM [66] NA NA NA OH O O H OH O H OH O O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH NA NA NA 9.72 µM [66] NA NA NA Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Hirsutrin [90] Hirsutrin NA NA 33.78% [91] NA NA NA 4.78 μM [91] O O OH O H OH O O OH OH OH CH3 OH OH O H NA NA 33.78% [91] NA NA NA 4.78 µM [91] Nutrients 2018, 10, x FOR PEER REVIEW 26 Nutrients 2018, 10, x FOR PEER REVIEW 26 o Nutrients 2018, 10, x FOR PEER REVIEW 26 OH O O OH O H OH O O OH OH OH CH3 OH OH OH Hirsutrin Hirsutr NA NA OH OH O O H OH O H OH O OH O O H O H OH O OH O O H O H OH O OH O A OH Hopeafuran Hopeafu Hopeafur NA NA NA O H O OH OH O OH OH O O H OCOCH3 O O O O H O H O-Ac O OH OH O OH OH O OH O H O H OCOCH3 O O O O H O H O-Ac O OH O OH OH O OH O H O H OCOCH3 O O O O H O H Hypolaetin 7-O-[6′′′-O acetyl-β-D-allopyranosyl →2)]-6′′-O-acetyl-β-D- Hypolaetin 7-O-[6′′′-O-acetyl- β-D-allopyranosyl-(1→2)]-6′′- O-acetyl-β-D-glucopyranoside Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl- →2)]-6′′-O-acetyl-β-D- glucopyranoside acetyl-β-D-allopyranosyl →2)]-6′′-O-acetyl-β-D- glucopyranoside Hypolaetin 7-O-[6′′′-O acetyl-β-D-allopyranosyl →2)]-6′′-O-acetyl-β-D- Hypolaetin 7-O-[6′′′-O-acetyl- β-D-allopyranosyl-(1→2)]-6′′- O-acetyl-β-D-glucopyranoside Hypolaetin 7-O-[6′′′-O- acetyl-β-D-allopyranosyl- →2)]-6′′-O-acetyl-β-D- glucopyranoside acetyl-β-D-allopyranosyl →2)]-6′′-O-acetyl-β-D- glucopyranoside 0.66 μM 0.66 μM [93] 0.66 µM [93] μM [93] NA NA NA O H OH OH O O OH O-caffeoyl O H O O C H3 OH OH OH O O OH O-caffeoyl O H O O O H C H3 OH OH OH O O OH O-caffeoyl O H O O OH OH O H C H3 OH I id Isocampneoside II Isocampneosid I id Isocampneoside II Isocampneosid NA NA 26 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H NA Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Kaempferol p [66] Kaempferol 36.01 μM [100] NA NA 45.58 μM [66] NA NA 1.10 μM [98,100] O O H OH OH O OH OH O OH OH O H 36.01 µM [100] NA NA 45.58 µM [66] NA NA 1.10 µM [98,100] Kakkalide Nutrients 2018, 10, x FOR PEER REVIEW 27 of 42 Kakkalide NA NA NA NA NA NA 0.56 μM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7- dione] NA NA NA NA 19.10 μM [102] NA NA Lupeol NA NA NA 3.60 μM [107] NA NA NA CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 O O O O OH OCH3 H3CO OH OH OH O O OH OH O H CH3 C H2 CH3 CH3 CH3 CH3 O H NA NA NA NA NA NA 0.56 µM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7-dione] Nutrients 2018, 10, x FOR PEER REVIEW 27 of 4 Kakkalide NA NA NA NA NA NA 0.56 μM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7- dione] NA NA NA NA 19.10 μM [102] NA NA Lupeol NA NA NA 3.60 μM [107] NA NA NA CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 O O O O OH OCH3 H3CO OH OH OH O O OH OH O H CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H NA NA NA NA 19.10 µM [102] NA NA Lupeol Nutrients 2018, 10, x FOR PEER REVIEW 27 of 4 Kakkalide NA NA NA NA NA NA 0.56 μM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7- dione] NA NA NA NA 19.10 μM [102] NA NA Lupeol NA NA NA 3.60 μM [107] NA NA NA CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 O O O O OH OCH3 H3CO OH OH OH O O OH OH O H CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H NA NA NA 3.60 µM [107] NA NA NA Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Kaempferol p [66] Kaempferol 36.01 μM [100] NA NA 45.58 μM [66] NA NA 1.10 μM [98,100] O O H OH OH O OH OH O OH OH O H 36.01 µM [100] NA NA 45.58 µM [66] NA NA 1.10 µM [98,100] Kakkalide Nutrients 2018, 10, x FOR PEER REVIEW 27 of Kakkalide NA NA NA NA NA NA 0.56 μM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7- dione] NA NA NA NA 19.10 μM [102] NA NA Lupeol NA NA NA 3.60 μM [107] NA NA NA CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 O O O O OH OCH3 H3CO OH OH OH O O OH OH O H CH3 C H2 CH3 CH3 CH3 CH3 NA NA NA NA NA NA 0.56 µM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7-dione] Nutrients 2018, 10, x FOR PEER REVIEW 27 of Kakkalide NA NA NA NA NA NA 0.56 μM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7- dione] NA NA NA NA 19.10 μM [102] NA NA Lupeol NA NA NA 3.60 μM [107] NA NA NA CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 O O O O OH OCH3 H3CO OH OH OH O O OH OH O H CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H NA NA NA NA 19.10 µM [102] NA NA Lupeol Nutrients 2018, 10, x FOR PEER REVIEW 27 of Kakkalide NA NA NA NA NA NA 0.56 μM [101] Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7- dione] NA NA NA NA 19.10 μM [102] NA NA Lupeol NA NA NA 3.60 μM [107] NA NA NA CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 O O O O OH OCH3 H3CO OH OH OH O O OH OH O H CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H NA NA NA 3.60 µM [107] NA NA NA O O H OH OH O OH O OH OCH3 O OH OCH3 H3CO OH O O O O O OH OCH3 H3CO OH OH OH O O OH OH O H O O O H3CO OH OH OH O O OH OH O H OH C H3 O O O OH OH OH O O OH OH O H Kakkalide Kakkal CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O CH3 OH C H3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 CH3 OH C H3 C H3 CH3 CH3 O O C H3 CH3 Lucidumol A [(24S)-24,25 Dihydroxylanost 8 ene 3 Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7-dione] Lucidumol A [(24S)-24,2 Dihydroxylanost-8-ene-3, dione] Dihydroxylanost 8 ene 3, dione] Lucidumol A [(24S)-24,25 Dih d l t 8 3 Lucidumol A [(24S)-24,25- Dihydroxylanost-8-ene-3,7-dione] Lucidumol A [(24S)-24,25 Dihydroxylanost-8-ene-3, dione] Dihydroxylanost 8 ene 3, dione] CH3 O CH3 C H3 CH3 CH3 C H2 CH3 CH3 CH3 CH3 C H2 CH3 CH3 CH3 CH3 O H CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H Lupeol Lup O H CH3 27 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Luteolin (2-(3,4-dihydroxyphenyl)-5,7- dihydroxy-4-chromomenone) Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) 16.60 μM [111] NA NA 6.34 μM [52] NA NA 0.087 μM [111] C H3 CH3 O H O OH OH O OH O H 16.60 µM [111] NA NA 6.34 µM [52] NA NA 0.087 µM [111] Luteolin-7-O-β-D-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 28 of Luteolin-7-O-β-D- glucopyranoside 117.80 μM [117] NA NA NA NA NA 7.34 μM [117] Magnoflorine NA NA NA NA NA NA 3.60 μM [118] Methyl-3,5-di-O- caffeoylquinate NA NA 117 g [33] 0.67 μM [33] NA NA 0.30 μM [33] N + CH3 CH3 H3CO O H O H H3CO O H O-caffeoyl O H O-caffeoyl H 3COOC O O H O OH OH O O O H O H O H OH 117.80 µM [117] NA NA NA NA NA 7.34 µM [117] Magnoﬂorine Nutrients 2018, 10, x FOR PEER REVIEW 28 of Luteolin-7-O-β-D- glucopyranoside 117.80 μM [117] NA NA NA NA NA 7.34 μM [117] Magnoflorine NA NA NA NA NA NA 3.60 μM [118] Methyl-3,5-di-O- caffeoylquinate NA NA 117 g [33] 0.67 μM [33] NA NA 0.30 μM [33] N + CH3 CH3 H3CO O H O H H3CO O H O-caffeoyl O H O-caffeoyl H 3COOC O O H O OH OH O O O H O H O H OH O NA NA NA NA NA NA 3.60 µM [118] Methyl-3,5-di-O-caffeoylquinate Nutrients 2018, 10, x FOR PEER REVIEW 28 of Luteolin-7-O-β-D- glucopyranoside 117.80 μM [117] NA NA NA NA NA 7.34 μM [117] Magnoflorine NA NA NA NA NA NA 3.60 μM [118] Methyl-3,5-di-O- caffeoylquinate NA NA 117 g [33] 0.67 μM [33] NA NA 0.30 μM [33] N + CH3 CH3 H3CO O H O H H3CO O H O-caffeoyl O H O-caffeoyl H 3COOC O O H O OH OH O O O H O H O H OH O NA NA 117 g [33] 0.67 µM [33] NA NA 0.30 µM [33] Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Luteolin (2-(3,4-dihydroxyphenyl)-5,7- dihydroxy-4-chromomenone) Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) 16.60 μM [111] NA NA 6.34 μM [52] NA NA 0.087 μM [111] C H3 CH3 O H O OH OH O OH O H 16.60 µM [111] NA NA 6.34 µM [52] NA NA 0.087 µM [111] Luteolin-7-O-β-D-glucopyranoside Nutrients 2018, 10, x FOR PEER REVIEW 28 of 42 Luteolin-7-O-β-D- glucopyranoside 117.80 μM [117] NA NA NA NA NA 7.34 μM [117] Magnoflorine NA NA NA NA NA NA 3.60 μM [118] Methyl-3,5-di-O- caffeoylquinate NA NA 117 g [33] 0.67 μM [33] NA NA 0.30 μM [33] N + CH3 CH3 H3CO O H O H H3CO O H O-caffeoyl O H O-caffeoyl H 3COOC O O H O OH OH O O O H O H O H OH 117.80 µM [117] NA NA NA NA NA 7.34 µM [117] Magnoﬂorine Nutrients 2018, 10, x FOR PEER REVIEW 28 of 4 Luteolin-7-O-β-D- glucopyranoside 117.80 μM [117] NA NA NA NA NA 7.34 μM [117] Magnoflorine NA NA NA NA NA NA 3.60 μM [118] Methyl-3,5-di-O- caffeoylquinate NA NA 117 g [33] 0.67 μM [33] NA NA 0.30 μM [33] N + CH3 CH3 H3CO O H O H H3CO O H O-caffeoyl O H O-caffeoyl H 3COOC O O H O OH OH O O O H O H O H OH NA NA NA NA NA NA 3.60 µM [118] Methyl-3,5-di-O-caffeoylquinate Nutrients 2018, 10, x FOR PEER REVIEW 28 of 4 Luteolin-7-O-β-D- glucopyranoside 117.80 μM [117] NA NA NA NA NA 7.34 μM [117] Magnoflorine NA NA NA NA NA NA 3.60 μM [118] Methyl-3,5-di-O- caffeoylquinate NA NA 117 g [33] 0.67 μM [33] NA NA 0.30 μM [33] N + CH3 CH3 H3CO O H O H H3CO O H O-caffeoyl O H O-caffeoyl H 3COOC O O H O OH OH O O O H O H O H OH NA NA 117 g [33] 0.67 µM [33] NA NA 0.30 µM [33] Structure 3 O OH OH O OH O H 16.60 16.60 µM 117.80 117.80 µM 117.80 μM [117] 117.80 μM [117] Luteolin-7-O-β-D-glucopyranoside Luteolin-7-O-β-D- glucopyranoside Luteolin 7 O β D glucopyranoside Luteolin 7 O β D glucopyranoside N + CH3 CH3 H3CO O H O H N + CH3 CH3 H3CO O H O H H3CO N + CH3 CH3 H3CO O H O H H3CO Magnoflor Magnoﬂorine Magnoflor Magnoflor 3 O H O ff l O H O-caffeoyl H 3COOC O H O-caffeoyl O H O-caffeoyl H 3COOC O H O-caffeoyl O H O-caffeoyl H 3COOC O ff O-caffeoyl O-caffeoyl Methyl-3,5-di-O- Methyl-3,5-di-O- caffeoylquinate Methyl-3,5-di-O-caffeoylquinate Methyl-3,5-di-O- caffeoylquinate O-caffeoyl O-caffeoyl 28 of 41 28 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H NA H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Mumeic acid-A caffeoylquinate NA NA [33] [33] NA NA [33] Mumeic acid-A NA NA NA NA NA NA 0.40 μM [119] O H O-caffeoyl O O O O OH OH O OCH3 O H O H NA NA NA NA NA NA 0.40 µM [119] Palbinone Nutrients 2018, 10, x FOR PEER REVIEW 29 of 42 Palbinone >500 μM [53] NA NA NA NA NA 11.40 μM [53] Puerariafuran NA NA NA NA NA NA 22.20 μM [57,121] Quercetin-3-O-β-D- glucoside >1000 μM [117] NA NA NA NA NA 2.21 μM [122] O OH H3CO O H CHO O O OH O H O O OH OH OH OH OH OH O H OH O O CH3 CH3 CH3 CH3 C H3 >500 µM [53] NA NA NA NA NA 11.40 µM [53] Puerariafuran Nutrients 2018, 10, x FOR PEER REVIEW 29 of 42 Palbinone >500 μM [53] NA NA NA NA NA 11.40 μM [53] Puerariafuran NA NA NA NA NA NA 22.20 μM [57,121] Quercetin-3-O-β-D- glucoside >1000 μM [117] NA NA NA NA NA 2.21 μM [122] O OH H3CO O H CHO O O OH O H O O OH OH OH OH OH OH O H OH O O CH3 CH3 CH3 CH3 C H3 NA NA NA NA NA NA 22.20 µM [57,121] Quercetin-3-O-β-D-glucoside Nutrients 2018, 10, x FOR PEER REVIEW 29 of 42 Palbinone >500 μM [53] NA NA NA NA NA 11.40 μM [53] Puerariafuran NA NA NA NA NA NA 22.20 μM [57,121] Quercetin-3-O-β-D- glucoside >1000 μM [117] NA NA NA NA NA 2.21 μM [122] O OH H3CO O H CHO O O OH O H O O OH OH OH OH OH OH O H OH O O CH3 CH3 CH3 CH3 C H3 >1000 µM [117] NA NA NA NA NA 2.21 µM [122] O O O O OH OH O OCH3 O H O H OH Mumeic acid-A Mumeic NA NA O H OH O O CH3 CH3 CH3 CH3 C H3 O H O O CH3 CH3 CH3 CH3 C H3 O H O O CH3 CH3 CH3 CH3 C H3 00 μM [53] >500 µM [53] 500 μM [53] [53] PalbinonePalb Pal CH3 3 O OH O H CHO O OH H3CO O H CHO O OH H3CO O H A NA NA P i Puerariafuran Puerari Puerar O O OH O H O O OH OH OH OH OH O O OH O H O O OH OH OH OH OH O O OH O H O O OH OH OH OH OH OH OH >1000 >1000 μM [117] >1000 μM [117] >1000 µM [117] Quercetin-3-O-β- Quercetin-3-O-β-D-glucoside Quercetin-3-O-β- glucoside 29 of 41 29 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Quercitrin (Quercetin 3-O-α-L-rhamnoside) Quercitrin (Quercetin 3-O- α-L-rhamnoside) 4.20 μM [100] NA NA NA NA NA 0.17 μM [40] O O O OH OH OH CH3 OH OH O OH O H OH OH 4.20 µM [100] NA NA NA NA NA 0.17 µM [40] Rhetsinine Nutrients 2018, 10, x FOR PEER REVIEW 30 of Rhetsinine NA NA NA NA NA NA 24.10 μM [124] Rosmarinic acid NA NA 532.38g [70] 2.77 μM [69] NA NA 0.30 μM [125] Scopoletin 2.93 μM [127] NA NA NA NA NA 2.59 μM [128] Semilicoisoflavone B NA NA NA 10.60 μM NA NA 1.80 μM O H OH O OH OH O OH O O H3CO O H O O O H NA NA NA NA NA NA 24.10 µM [124] Rosmarinic acid Nutrients 2018, 10, x FOR PEER REVIEW 30 of 4 Rhetsinine NA NA NA NA NA NA 24.10 μM [124] Rosmarinic acid NA NA 532.38g [70] 2.77 μM [69] NA NA 0.30 μM [125] Scopoletin 2.93 μM [127] NA NA NA NA NA 2.59 μM [128] Semilicoisoflavone B NA NA NA 10.60 μM [131] NA NA 1.80 μM [131] O H OH O OH OH O OH O O H3CO O H O O O H O O OH CH3 NA NA 532.38g [70] 2.77 µM [69] NA NA 0.30 µM [125] Scopoletin Nutrients 2018, 10, x FOR PEER REVIEW 30 of 42 Rhetsinine NA NA NA NA NA NA 24.10 μM [124] Rosmarinic acid NA NA 532.38g [70] 2.77 μM [69] NA NA 0.30 μM [125] Scopoletin 2.93 μM [127] NA NA NA NA NA 2.59 μM [128] Semilicoisoflavone B NA NA NA 10.60 μM [131] NA NA 1.80 μM [131] O H OH O OH OH O OH O O H3CO O H O O O H O O OH OH CH3 CH3 2.93 µM [127] NA NA NA NA NA 2.59 µM [128] Semilicoisoﬂavone B Nutrients 2018, 10, x FOR PEER REVIEW 30 of 42 Rhetsinine NA NA NA NA NA NA 24.10 μM [124] Rosmarinic acid NA NA 532.38g [70] 2.77 μM [69] NA NA 0.30 μM [125] Scopoletin 2.93 μM [127] NA NA NA NA NA 2.59 μM [128] Semilicoisoflavone B NA NA NA 10.60 μM [131] NA NA 1.80 μM [131] O H OH O OH OH O OH O O H3CO O H O O O H O O OH OH CH3 CH3 NA NA NA 10.60 µM [131] NA NA 1.80 µM [131] Active Ingredient O O O OH OH OH CH3 OH OH O OH O H Quercitrin (Quercetin 3-O-α-L-rhamnoside) Quercitrin (Quer α-L-rhamno ents 2018, 10, x FOR PEER R Nutrients 2018, 10, x FOR PEER OH Rhetsinine Rhetsin Rhetsin Rhe NA NA NA A O H O OH OH O OH O O H OH O OH OH O OH O O H OH O OH OH O OH O O H OH O OH O OH O OH OH Rosmarinic acid Rosmarinic a Rosmarin Rosmarin OH H3CO H3CO O H3CO O H O O H3CO O H O ScopoletinSco Scop O O H O O O H O O H O O O H O O H O O OH OH CH3 CH3 O O H O O OH OH CH3 CH3 Semilicoisofla Semilicoisoﬂavone B Semilicoisofla OH OH 30 of 41 30 of 41 Nutrients 2018, 10, 1580 Table 1. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Summary of relevant ex vivo and in vivo activities of phytoconstituents. Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. T bl 2 S f l t i d i i ti iti f h t tit t t t Doses (IC50/EC50) AR S l it G l t I Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. ( / ) Table 2 Su a y of ele a t ex i o a d i i o acti ities of phytoco stitue ts t t Doses (IC50/EC50) AR S l it G l t I Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite-Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose-Induced Lens Opacity Ref 1-O-galloyl-β-D-glucose (β-Glucogallin) Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H Ex vivo: 30.00 µM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl-β- D-glucose Nutrients 2018, 10, x FOR PEER REVIEW 32 of 42 Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H NA NA NA NA Ex vivo: 80.00 µM [38] 3,5-di-O-caffeoyl-epi-quinic Acid Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H NA NA NA NA Ex vivo: 10.00 µM [48] 5,7,4′-trihydroxyisoﬂavone (Genistein) Nutrients 2018, 10, x FOR PEER REVIEW 32 of 42 Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H NA NA NA NA Ex vivo: 18.50 µM [60] Isorhamnetin-3-glucoside Nutrients 2018, 10, x FOR PEER REVIEW 33 of 42 Isorhamnetin-3- glucoside NA Ex vivo: 52.25 μM NA NA NA [97] O O H OH O OGlu OH OCH3 CH NA Ex vivo: 52.25 µM NA NA NA [97] Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite-Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose-Induced Lens Opacity Ref 1-O-galloyl-β-D-glucose (β-Glucogallin) Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H Ex vivo: 30.00 µM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl-β- D-glucose Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H NA NA NA NA Ex vivo: 80.00 µM [38] 3,5-di-O-caffeoyl-epi-quinic Acid Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H NA NA NA NA Ex vivo: 10.00 µM [48] 5,7,4′-trihydroxyisoﬂavone (Genistein) Nutrients 2018, 10, x FOR PEER REVIEW 32 of 42 Table 2. Summary of relevant ex vivo and in vivo activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H NA CH Active Ingredient Structure IC50 Values AGE ARI GLWW RHAR BLAR HLAR RLAR Sulfuretin μ [131] [131] Sulfuretin 124.00 μM [133] NA NA 1.30 μM [133] NA NA NA O O OH OH CH3 CH3 O O OH O H O H 124.00 µM [133] NA NA 1.30 µM [133] NA NA NA Syringic Acid Nutrients 2018, 10, x FOR PEER REVIEW 31 o Syringic Acid NA NA NA NA NA NA 1081.1 μm [136] Swertisin (C-glycosidic flavonoid) Glu=glucose NA NA NA NA NA NA 1.60 μm [138] Valoneic acid dilactone NA NA NA NA NA NA 0.075 μM [87] O O OH H3CO Glu OH COOH H 3CO O H OCH 3 O O O OH O O H O H O O H O H OH O OH NA NA NA NA NA NA 1081.1 µm [136] Swertisin (C-glycosidic ﬂavonoid) Nutrients 2018, 10, x FOR PEER REVIEW 31 of 4 Syringic Acid NA NA NA NA NA NA 1081.1 μm [136] Swertisin (C-glycosidic flavonoid) Glu=glucose NA NA NA NA NA NA 1.60 μm [138] Valoneic acid dilactone NA NA NA NA NA NA 0.075 μM [87] O O OH H3CO Glu OH COOH H 3CO O H OCH 3 O O O OH O O H O H O O H O H OH O OH Glu=glucose NA NA NA NA NA NA 1.60 µm [138] Valoneic acid dilactone Nutrients 2018, 10, x FOR PEER REVIEW 31 of Syringic Acid NA NA NA NA NA NA 1081.1 μm [136] Swertisin (C-glycosidic flavonoid) Glu=glucose NA NA NA NA NA NA 1.60 μm [138] Valoneic acid dilactone NA NA NA NA NA NA 0.075 μM [87] O O OH H3CO Glu OH COOH H 3CO O H OCH 3 O O O OH O O H O H O O H O H OH O OH NA NA NA NA NA NA 0.075 µM [87] StructureOH O O OH O H O H 124.00 124.00 µM [133] 124.00 μM [133] 124.00 µM [133] Sulfuretin Sulfur O H COOH H 3CO O H OCH 3 COOH H 3CO O H OCH 3 COOH H 3CO O H OCH 3 COOH COOH COOH Syringic Acid Syringic A Syringic Ac Syringic A Glu=glucose O O OH H3CO Glu OH Glu=glucose O O OH H3CO Glu OH Glu glucose Glu=glucose O O OH H3CO Glu OH Swertisin (C-glycosidic flavonoid) Swertisin (C-glycosidic ﬂavonoid) Swertisin (C-glycosidic flavonoid) Swertisin (C-glycosidic flavonoid) Swertisin (C-glycosidic flavonoid) Swertisin (C-glycosidic ﬂavonoid) Swertisin (C-glycosidic flavonoid) Swertisin (C-glycosidic flavonoid) G Glu Glu Glu=glucos Glu=glucose Glu=glucose Glu=glucos O O O OH O O H O H O O H O H OH O OH O O O OH O O H O H O O H O H OH O OH O O O OH O O H O H O O H O H OH O OH Valoneic acid dilac Valoneic acid dilacto Valoneic acid dilactone Valoneic acid dilact 31 of 41 Nutrients 2018, 10, 1580 Nutrients 2018 10 x Table 2. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Cont. Ex vivo: 52.2 Ex vivo: 52.2 Ex vivo: 52. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite- Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose- Induced Lens Opacity Ref 1-O-galloyl-β-D- glucose (β- Glucogallin) Ex vivo: 30.00 μM NA NA NA NA [29] 1,2,3,6-Tetra-O-galloyl- β-D-glucose NA NA NA NA Ex vivo: 80.00 μM [38] 3,5-di-O-caffeoyl-epi- quinic Acid NA NA NA NA Ex vivo: 10.00 μM [48] 5,7,4′- trihydroxyisoflavone (Genistein) NA NA NA NA Ex vivo: 18.50 μM [60] O O O O O H O-gal gal-O gal gal O H OH O O O H OH O OH HOOC O H O O O H OH O OH O H O H O O O O H OH OH O H O H NA NA NA NA Ex vivo: 18.50 µM [60] Isorhamnetin-3-glucoside Nutrients 2018, 10, x FOR PEER REVIEW 33 of 42 Isorhamnetin-3- glucoside NA Ex vivo: 52.25 μM NA NA NA [97] O O H OH O OGlu OH OCH3 CH NA Ex vivo: 52.25 µM NA NA NA [97] O H O H O O O O H OH OH O H O H Ex O H O H O O O O H OH OH O H O H O-gal O H O O O O H OH O H nt) E O H O H O O O O H OH OH O H O H 1-O-galloyl-β-D-glucose (β-Glucogallin) 1-O-galloyl-β-D- glucose (β- 1-O-galloyl-β glucose (β- glucose (β Glucogallin) 1-O-galloyl-β glucose (β O O O O O H O-gal gal O O O O O H O-gal gal-O gal gal O O O O O H gal-O gal gal O O O O O H O-gal gal gal 1,2,3,6-Tetra-O-galloyl 1,2,3,6-Tetra-O-galloyl-β- D-glucose 1,2,3,6-Tetra-O-ga β-D-glucose 1,2,3,6-Tetra-O-galloyl β-D-glucose 1,2,3,6-Tetra-O-ga NA NA O H eoyl-epi- Acid gal-O gal O H OH O O O H OH OH HOOC O H O O-caffeoyl-epi- uinic Acid O H OH O O O H OH O OH HOOC O H O O O H epi-quinic eoyl-epi- Acid O H OH O O O H OH O OH HOOC O H O O-caffeoyl-epi- uinic Acid O H OH O O O H OH O OH HOOC O H O gal-O gal O H OH O O O H OH OH HOOC O H O O H OH O O O H OH O OH HOOC O H O O O H O H OH O O O H OH O OH HOOC O H O O H OH O O O H OH O OH HOOC O H O 3,5-di-O-caffeoyl-epi- quinic Acid NA gal-O gal O H OH O O O H OH OH HOOC O H O 3,5-di-O-caffeoyl-epi- quinic Acid NA O H OH O O O H OH O OH HOOC O H O O O H 3,5-di-O-caffeoyl-epi-quinic Acid 3,5-di-O-caffeoyl-epi- quinic Acid NA O H OH O O O H OH O OH HOOC O H O 3,5-di-O-caffeoyl-epi- quinic Acid NA O H OH O O O H OH O OH HOOC O H O ga OH OH 3,5-di-O-caffeoyl-e quinic Acid 3,5-di-O-caffeoyl-epi-quinic Acid 3,5-di-O-caffeoyl-epi- quinic Acid 3,5-di-O-caffeoyl-e quinic Acid NANA NA NA N OOH OH O O OH O OH O O OH O O O H OH O OH EVIEW , , trihydroxyisoflav (Genistein) 5,7,4′- trihydroxyisoflavone 5,7,4′-trihydroxyisoﬂavone (Genistein) 5,7,4′- trihydroxyisoflav (Genistein) N i FOR P NA NA NA NA OH O OH OH O O H OH O OGlu OH OCH3 52.25 NA Ex vivo: 52.25 µM Isorhamnetin-3-glucoside Isorhamnetin-3- glucoside Isorhamnetin-3-glucoside Isorhamnetin-3- glucoside Ex NA OGlu 32 of 41 32 of 41 Nutrients 2018, 10, 1580 Table 2. 4. Discussion and Outlook Despite the success in surgical replacement of the cataractous lens with an artiﬁcial intraocular lens, discovery of pharmacological prevention and treatment of this blinding disorder has been an earnest, continuous effort in ophthalmology research. In this review manuscript, we summarize ﬁndings of phytoconstitutents and their pharmacological effects as potential anti-cataract agents. The large number of interesting compounds is exciting. It raises hope that clinically useful medication may have a good chance to be derived from this sizable collection of chemicals with diverse structural scaffolds. Many of the compounds have potent and efﬁcacious in vitro pharmacological activities that are consistent with potential anti-cataract effects. For example, 1,2,3,6-tetra-O-galloyl-β-D-glucose inhibits AGE formation with an IC50 of 2 M. Both 1,3,5,8-tetrahydroxyxanthone and 2′′,4′′-O-diacetylquercitrin inhibit AR with IC50 values below 0.1 M. However, a major limitation of the listed compounds is that although they have been shown to have the appropriate biological actions in a variety of in vitro or ex vivo assays, many of them were not tested in animal cataract models. Additionally, a few have been evaluated in only one animal model. Without relevant in vivo data, it is obviously very difﬁcult to develop the compounds into meaningful treatments for cataract patients. p p g p In addition to the lack of in vivo data, there are other challenges facing development of anti-cataract pharmaceuticals. For example, cataract medication has to compete with the very successfully and generally affordable (as least in developed countries) surgical procedure. Moreover, pharmacological prevention of cataract formation is expected to require a long-term, likely multi-year, administration of medicine, which, to some, is undesirable. Overcoming these challenges necessitates careful considerations of drug safety, convenience of administration, and cost. These concerns may have previously prohibited the development of certain agents. Nonetheless, we feel that phytoconstituents are advantageous compared to conventional synthetic drugs. Many societies have been using plant products from where some of the ingredients are derived for centuries, indicating long-term safety and acceptance. The development path and clinical use will be similar to vitamins and phytochemicals such as lutein and zeaxanthine. If proven safe, cost-effective, and most importantly, efﬁcacious in preventing or reversing cataract formation, phytoconstituents can be a revolutionary approach in the treatment of cataract. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. Table 1. Summary of relevant in vitro anti-cataract activities of phytoconstituents. y p y St t IC50 Valu Structure IC50 Valu AGE ARI GLWW RHAR AGE ARI GLWW RHAR OH O H O H OH O H μM μM Constituent Name (Class of Constituent) Structure Doses (IC50/EC50) AR Transgenic Mice Selenite-Induced AR Rat Lens Galactose-Induced Lens Opacity Xylose-Induced Lens Opacity Ref Lupeol Isorhamnetin 3 glucoside NA Ex vivo: 52.25 μM NA NA NA [97] Lupeol NA In vivo: 126.15 μM NA NA NA [108] Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) NA Ex vivo: 6.98 μM NA NA NA [112] Puerariafuran NA NA NA NA Ex vivo: 15.00 μM [121] Scopoletin NA NA NA NA Ex vivo: 25.00 μM [127] OH O OGlu O H3CO O H O O OH H3CO O H CHO CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H O OH OH O OH O H NA In vivo: 126.15 µM NA NA NA [108] Luteolin (2-(3,4-dihydroxyphenyl)-5,7- dihydroxy-4-chromomenone) glucoside NA μM NA NA NA [97] Lupeol NA In vivo: 126.15 μM NA NA NA [108] Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) NA Ex vivo: 6.98 μM NA NA NA [112] Puerariafuran NA NA NA NA Ex vivo: 15.00 μM [121] Scopoletin NA NA NA NA Ex vivo: 25.00 μM [127] OH O OGlu O H3CO O H O O OH H3CO O H CHO CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H O OH OH O OH O H NA Ex vivo: 6.98 µM NA NA NA [112] Puerariafuran glucoside μM Lupeol NA In vivo: 126.15 μM NA NA NA [108] Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) NA Ex vivo: 6.98 μM NA NA NA [112] Puerariafuran NA NA NA NA Ex vivo: 15.00 μM [121] Scopoletin NA NA NA NA Ex vivo: 25.00 μM [127] OH O OGlu O H3CO O H O O OH H3CO O H CHO CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H O OH OH O OH O H NA NA NA NA Ex vivo: 15.00 µM [121] Scopoletin Lupeol NA In vivo: 126.15 μM NA NA NA [108] Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) NA Ex vivo: 6.98 μM NA NA NA [112] Puerariafuran NA NA NA NA Ex vivo: 15.00 μM [121] Scopoletin NA NA NA NA Ex vivo: 25.00 μM [127] OH O OGlu O H3CO O H O O OH H3CO O H CHO CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H O OH OH O OH O H NA NA NA NA Ex vivo: 25.00 µM [127] Syringic acid Nutrients 2018, 10, x FOR PEER REVIEW 34 of 42 Syringic acid NA NA NA Ex vivo: 1075.70 μM In vivo: 2% syringic acid eye drop (131,197.80 μM) NA [136] COOH H3CO OH OCH3 NA NA NA Ex vivo: 1075.70 µM In vivo: 2% syringic acid eye drop (131,197.80 µM) NA [136] Constituent Name (Class of Constituent) glucoside StructureOG H O OG H O H O CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H CH3 C H2 CH3 CH3 CH3 CH3 C H3 CH3 O H CH3 C H2 CH3 CH3 CH3 C H3 CH3 O H CH3 2 CH3 CH3 CH3 C H3 CH3 O H OH Lupeol Lupeo Lupeo Lupeo Lupeo NANA A NA O H H O H CH3 CH3 O OH OH O OH O H O OH OH O OH O H O OH OH O OH O H O OH O OH O H Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- h ) Luteolin (2-(3,4-dihydroxyphenyl)-5,7- dihydroxy-4-chromomenone) Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) Luteolin (2-(3,4- dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) dihydroxyphenyl)-5,7- dihydroxy-4- chromomenone) x vivo: 6.98 N x vivo: 6.98 μM N Ex vivo: 6.98 µM x vivo: 6.98 μM i o μM NA NANA NA NA O OH O H CHO O OH H CO O H CHO O OH H3CO O H CHO O OH H3CO O H NA NANA NA NA A NANA NA P i f Puerariafuran Puerariafura Puerariafura H CO H3CO 3 H3CO H3CO O H3CO O H O O H3CO O H O REVIEW NA NANA A NA S l Scopoletin Scopolet ie t 2018 10 O O H O O O H O COOH H3CO OH OCH3 NA NA A NA A NA Syringic acid Syringic ac OH Nutrients 2018, 10, 1580 33 of 41 5. Conclusions Despite the success in surgical replacement of the cataractous lens with artiﬁcial intraocular lens, pharmacological prevention and treatment of this blinding disorder have been an earnest, continuous effort in ophthalmology research. In this review manuscript, we summarize ﬁndings of 56 entries of phytoconstitutents and their pharmacological effects as potential anti-cataract agents. The large number of interesting compounds is exciting. It raises hope that clinically useful medication may have a good chance to be derived from this sizable collection of chemicals with diverse structural scaffolds. Many of the compounds have potent and efﬁcacious in vitro pharmacological activities that are consistent with potential anti-cataract effects. For example, 1,2,3,6-tetra-O-galloyl-β-D-glucose inhibits AGE formation with an IC50 of 2 µM. Additionally, both 1,3,5,8-tetrahydroxyxanthone and 2′′,4′′-O-diacetylquercitrin inhibit AR with IC50 values below 0.1 µM. However, a major limitation of the listed compounds is that although they have been shown to have the appropriate biological actions in a variety of in vitro or ex vivo assays, many of them were not tested in animal cataract models. And a few have been evaluated in only one animal model. Without relevant in vivo data, it is obviously very difﬁcult to develop the compounds into meaningful treatments for cataract patients. We feel that, by listing the comprehensive collection of phytocontituents in one place, this manuscript serves as an overview and perhaps an inspiration to prompt additional studies in this important research area. Collaborative efforts between phytochemists and cataract researchers are promisingly fruitful. p y p g y In addition to the lack of in vivo data, there are other challenges facing development of anti-cataract pharmaceuticals. For example, cataract medication has to compete with the very successfully and generally affordable (as least in developed countries) surgical procedure. Moreover, 34 of 41 Nutrients 2018, 10, 1580 pharmacological prevention of cataract formation is expected to require a long-term, likely multi-year, administration of medicine, which, to some, is undesirable. Overcoming these challenges necessitates careful considerations of drug safety, convenience of administration, and cost. These concerns may have previously prohibited the development of certain agents. Nonetheless, we feel that phytoconstituents are advantageous compared to conventional synthetic drugs. Many societies have been using plant products where some of the ingredients are derived from for centuries, indicating long-term safety and acceptance. The development path and clinical use will be similar to vitamins and phytochemicals such as lutein and zeaxanthine. 5. Conclusions If proven safe, cost-effective, and most importantly, efﬁcacious in preventing or reversing cataract formation, phytoconstituents can be a revolutionary approach in the treatment of cataract. Author Contributions: V.L. and H.W. drafted the review. E.S. provided key reviewing. I.-H.P. critically revised the drafted manuscript. All authors were involved in writing the paper and had ﬁnal approval of the submitted and published versions. Funding: Funding in part supported by the Frick Foundation, BrightFocus Foundation for Macular Degenera Grant No. M2015180), California Table Grape Commission and Bridging Grant (304/CIIPT/6316033). Acknowledgments: This review was supported in part by a grant from the Community Solutions Program, a program of the Bureau of Educational and Cultural Affairs (ECA) of the United States Department of State, implemented by IREX (International Research and Exchanges Board) and from the Department of Pharmaceutical Science, University of North Texas, System College of Pharmacy. The views expressed are the author’s own and do not represent the Community Solutions Program, the U.S. Department of State, or IREX. VL would like to thank Universiti Sains Malaysia for providing research leave to complete the Community Solutions Program. We also sincerely appreciate the effort of the staffs in BRIT and UNTHSC for the support provided to us. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations Abbreviations AGE Advanced glycation end-product AR Aldose reductase ARI Aldose reductase inhibition BLAR Bovine lens aldose reductase GC Glucocorticoid GSH Glutathione HLAR Human lens aldose reductase LEC Lens epithelial cell NADPH Nicotinamide adenine dinucleotide phosphate RHAR Recombinant human aldose reductase RLAR Rat lens aldose reductase ROS Reactive oxygen species SDH Sorbitol dehydrogenase STZ Streptozotocin UV Ultraviolet References Abbreviations AGE Advanced glycation end-product AR Aldose reductase ARI Aldose reductase inhibition BLAR Bovine lens aldose reductase GC Glucocorticoid GSH Glutathione HLAR Human lens aldose reductase LEC Lens epithelial cell NADPH Nicotinamide adenine dinucleotide phosphate RHAR Recombinant human aldose reductase RLAR Rat lens aldose reductase ROS Reactive oxygen species SDH Sorbitol dehydrogenase STZ Streptozotocin UV Ultraviolet References 1. Lou, M.F. Redox regulation in the lens. Prog. Retin. Eye Res. 2003, 22, 657–682. [CrossRef] 2. Liu, Y.C.; Wilkins, M.; Kim, T.; Malyugin, B.; Mehta, J.S. Cataracts. Lancet 2017, 390, 600–612. [CrossRef] 3. Fischel, J.D.; Lipton, J.R. Cataract surgery and recent advances: A review. Nurs. Stand. 1996, 10, 39–43. [CrossRef] [PubMed] 4. Spector, A. Oxidative stress-induced cataract: Mechanism of action. FASEB J. 1995, 9, 1173–1182. [CrossRef] [PubMed] 5. Spector, A.; Garner, W.H. Hydrogen peroxide and human cataract. Exp. Eye Res. 1981, 33, 673–681. [CrossRef] 6. Cui, X.L.; Lou, M.F. The effect and recovery of long-term H2O2 exposure on lens morphology and biochemistry. Exp. Eye Res. 1993, 57, 157–167. [CrossRef] [PubMed] AGE Advanced glycation end-product AR Aldose reductase ARI Aldose reductase inhibition BLAR Bovine lens aldose reductase GC Glucocorticoid GSH Glutathione HLAR Human lens aldose reductase LEC Lens epithelial cell NADPH Nicotinamide adenine dinucleotide phosphate RHAR Recombinant human aldose reductase RLAR Rat lens aldose reductase ROS Reactive oxygen species SDH Sorbitol dehydrogenase STZ Streptozotocin UV Ultraviolet References 1. Lou, M.F. Redox regulation in the lens. Prog. Retin. Eye Res. 2003, 22, 657–682. [CrossRef] 2. Liu, Y.C.; Wilkins, M.; Kim, T.; Malyugin, B.; Mehta, J.S. Cataracts. Lancet 2017, 390, 600–612. [CrossRef] 3. Fischel, J.D.; Lipton, J.R. Cataract surgery and recent advances: A review. Nurs. Stand. 1996, 10, 39–43. [CrossRef] [PubMed] 4. Spector, A. Oxidative stress-induced cataract: Mechanism of action. FASEB J. 1995, 9, 1173–1182. [CrossRef] [PubMed] 5. Spector, A.; Garner, W.H. Hydrogen peroxide and human cataract. Exp. Eye Res. 1981, 33, 673–681. [CrossRef] 6. Cui, X.L.; Lou, M.F. The effect and recovery of long-term H2O2 exposure on lens morphology and biochemistry. Exp. Eye Res. 1993, 57, 157–167. [CrossRef] [PubMed] . Lou, M.F. Redox regulation in the lens. Prog. Retin. Eye Res. 2003, 22, 657–682. [CrossRef] . Liu, Y.C.; Wilkins, M.; Kim, T.; Malyugin, B.; Mehta, J.S. Cataracts. Lancet 2017, 390, 600–612. [CrossRe Fischel, J.D.; Lipton, J.R. Cataract surgery and recent advances: A review. Nurs. Stand. 1996, 10, 39–43 [CrossRef] [PubMed] Spector, A. Oxidative stress-induced cataract: Mechanism of action. FASEB J. 1995, 9, 1173–1182. [CrossRef [PubMed] 5. Spector, A.; Garner, W.H. Hydrogen peroxide and human cataract. Exp. Eye Res. 1981, 33, 673–681. [CrossRef] 6. Cui, X.L.; Lou, M.F. The effect and recovery of long-term H2O2 exposure on lens morphology and biochemistry. Exp. Eye Res. 1993, 57, 157–167. [CrossRef] [PubMed] 5. Spector, A.; Garner, W.H. Hydrogen peroxide and human cataract. Exp. Eye Res. 1981, 33, 673–681. [CrossRef] 6. Cui, X.L.; Lou, M.F. The effect and recovery of long-term H2O2 exposure on lens morphology and biochemistry. Exp. Eye Res. 1993, 57, 157–167. [CrossRef] [PubMed] 35 of 41 Nutrients 2018, 10, 1580 7. Zigler, J.S., Jr.; Jernigan, H.M., Jr.; Garland, D.; Reddy, V.N. The effects of “oxygen radicals” generated in the medium on lenses in organ culture: Inhibition of damage by chelated iron. Arch. Biochem. Biophys. 1985, 241, 163–172. [CrossRef] 8. Caird, F.I.; Hutchinson, M.; Pirie, A. Cataract and Diabetes. Br. Med. J. 1964, 2, 665–668. [CrossRef] [PubMed] 9. Chodos, J.B.; Habegger-Chodos, H.E. Cataract formation in human and experimental diabetes. Part I. Surv. Ophthalmol. 1960, 5, 129–159. [PubMed] 8. Caird, F.I.; Hutchinson, M.; Pirie, A. Cataract and Diabetes. Br. Med. J. 1964, 2, 665–668. [CrossRef] [P 9. Chodos, J.B.; Habegger-Chodos, H.E. Cataract formation in human and experimental diabetes. Part I. Surv. Ophthalmol. 1960, 5, 129–159. [PubMed] 10. References Snow, A.; Shieh, B.; Chang, K.C.; Pal, A.; Lenhart, P.; Ammar, D.; Ruzycki, P.; Palla, S.; Reddy, G.B.; Petrash, J.M. Aldose reductase expression as a risk factor for cataract. Chem. Biol. Interact. 2015, 234, 247–253. [CrossRef] [PubMed] 11. Hayman, S.; Kinoshita, J.H. Isolation and Properties of Lens Aldose Reductase. J. Biol. Chem. 1965, 240, 877–882. [PubMed] 12. Obazawa, H.; Merola, L.O.; Kinoshita, J.H. The effects of xylose on the isolated lens. Investig. Ophthalmol. 1974, 13, 204–209. 13. Ai, Y.; Zheng, Z.; O’Brien-Jenkins, A.; Bernard, D.J.; Wynshaw-Boris, T.; Ning, C.; Reynolds, R.; Segal, S.; Huang, K.; Stambolian, D. A mouse model of galactose-induced cataracts. Hum. Mol. Genet. 2000, 9, 1821–1827. [CrossRef] [PubMed] 4. Franke, S.; Dawczynski, J.; Strobel, J.; Niwa, T.; Stahl, P.; Stein, G. Increased levels of advanced glycation products in human cataractous lenses. J. Cataract Refract. Surg. 2003, 29, 998–1004. [CrossRef] 15. Ramalho, J.S.; Marques, C.; Pereira, P.C.; Mota, M.C. Role of glycation in human lens protein structure change. Eur. J. Ophthalmol. 1996, 6, 155–161. [CrossRef] [PubMed] 16. Abiko, T.; Abiko, A.; Ishiko, S.; Takeda, M.; Horiuchi, S.; Yoshida, A. Relationship between autoﬂuorescence and advanced glycation end products in diabetic lenses. Exp. Eye Res. 1999, 68, 361–366. [CrossRef] [PubMed] 17. Joon, T.L.; Foo, K.; Panagiotopoulos, S.; Jerums, G.; Taylor, H.R. In vivo assessment of an animal model of diabetic cataract: Medical intervention studies. Dev. Ophthalmol. 1994, 26, 57–62. [PubMed] 18. Patil, M.A.; Suryanarayana, P.; Putcha, U.K.; Srinivas, M.; Reddy, G.B. Evaluation of neonatal streptozotocin induced diabetic rat model for the development of cataract. Oxid. Med. Cell Longev. 2014, 2014, 463264. [CrossRef] [PubMed] 19. Lee, A.Y.; Chung, S.S. Contributions of polyol pathway to oxidative stress in diabetic cataract. FASEB J. 1999, 13, 23–30. [CrossRef] [PubMed] 20. Shearer, T.R.; David, L.L.; Anderson, R.S. Selenite cataract: A review. Curr. Eye Res. 1987, 6, 289–300. [CrossRef] [PubMed] 21. Shearer, T.R.; David, L.L.; Anderson, R.S.; Azuma, M. Review of selenite cataract. Curr. Eye Res. 1992, 11, 357–369. [CrossRef] [PubMed] 22. Shearer, T.R.; Ma, H.; Fukiage, C.; Azuma, M. Selenite nuclear cataract: Review of the model. Mol. Vis. 1997, 3, 8. [PubMed] 23. Dillon, J. UV-B as a pro-aging and pro-cataract factor. Doc. Ophthalmol. 1994, 88, 339–344. [CrossRef] [PubMed] 24. Wolff, S.P. Cataract and UV radiation. Doc. Ophthalmol. 1994, 88, 201–204. [CrossRef] [PubMed] , p , , [ ] [ ] 25. Kronschlager, M.; Lofgren, S.; Yu, Z.; Talebizadeh, N.; Varma, S.D.; Soderberg, P. References Caffeine eye drops protect [C f] [ b d] p 25. Kronschlager, M.; Lofgren, S.; Yu, Z.; Talebizadeh, N.; Varma, S.D.; Soderberg, P. Caffeine eye drops protect against UV-B cataract. Exp. Eye Res. 2013, 113, 26–31. [CrossRef] [PubMed] 25. Kronschlager, M.; Lofgren, S.; Yu, Z.; Talebizadeh, N.; Varma, S.D.; Soderberg, P against UV-B cataract. Exp. Eye Res. 2013, 113, 26–31. [CrossRef] [PubMed] 26. James, E.R. The etiology of steroid cataract. J. Ocul. Pharmacol. Ther. 2007, 23, 403–420. [CrossRef] [P 27. Shui, Y.B.; Kojima, M.; Sasaki, K. A new steroid-induced cataract model in the rat: Long-term prednisolone applications with a minimum of X-irradiation. Ophthalmic Res. 1996, 28, 92–101. [CrossRef] [PubMed] 28. Shui, Y.B.; Vrensen, G.F.; Kojima, M. Experimentally induced steroid cataract in the rat: A scanning electron microscopic study. Surv. Ophthalmol. 1997, 42, S127–S132. [CrossRef] 29. Variya, B.C.; Bakrania, A.K.; Patel, S.S. Emblica ofﬁcinalis (Amla): A review for its phytochemistry, ethnomedicinal uses and medicinal potentials with respect to molecular mechanisms. Pharmacol. Res. 2016, 111, 180–200. [CrossRef] [PubMed] 30. Puppala, M.; Ponder, J.; Suryanarayana, P.; Reddy, G.B.; Petrash, M.; LaBarbera, D.V. The isolation and characterization of β-glucogallin as a novel aldose reductase inhibitor from Emblica ofﬁcinalis. PLoS ONE 2012, 7, e31399. [CrossRef] [PubMed] 36 of 41 36 of 41 Nutrients 2018, 10, 1580 31. Chang, K.C.; Lafﬁn, B.; Ponder, J.; Énzsöly, A.; Németh, J.; Labarbera, D.V.; Petrash, J.M. Beta-glucogallin reduces the expression of lipopolysaccharide-induced inﬂammatory markers by inhibition of aldose reductase in murine macrophages and ocular tissues. Chem. Biol. Interact. 2013, 202, 283–287. [CrossRef] [PubMed] 32. Lee, Y.K.; Hong, E.Y.; Whang, W.K. Inhibitory Effect of Chemical Constituents Isolated from Artemisia iwayomogi on Polyol Pathway and Simultaneous Quantiﬁcation of Major Bioactive Compounds. BioMed Res. Int. 2017, 2017. [CrossRef] [PubMed] 33. Yoon, H.N.; Lee, M.Y.; Kim, J.K.; Suh, H.W.; Lim, S.S. Aldose reductase inhibitory compounds from Xanthium strumarium. Arch. Pharm. Res. 2013, 36, 1090–1095. [CrossRef] [PubMed] 34. Wang, Z.; Hwang, S.H.; Lim, S.S. Characterization of DHDP, a novel aldose reductase inhibitor isolated from Lysimachia christinae. J. Funct. Foods 2017, 37, 241–248. [CrossRef] 35. Paek, J.H.; Lim, S.S. Preparative isolation of aldose reductase inhibitory compounds from Nardostachys chinensis by elution–extrusion counter-current chromatography. Arch. Pharm. Res. 2014, 37, 1271–1279. [CrossRef] [PubMed] 36. Yoo, N.H.; Jang, D.S.; Lee, Y.M.; Jeong, I.H.; Cho, J.H.; Kim, J.H.; Kim, J.S. References Anthraquinones from the Roots of Knoxia valerianoides inhibit the formation of advanced glycation end products and rat lens aldose reductase in vitro. Arch. Pharm. Res. 2010, 33, 209–214. [CrossRef] [PubMed] 37. Duan, D.; Li, Z.; Luo, H.; Zhang, W.; Chen, L.; Xu, X. Antiviral compounds from traditional Chinese medicines Galla Chinese as inhibitors of HCV NS3 protease. Bioorg. Med. Chem. Lett. 2004, 14, 6041–6044. [CrossRef] [PubMed] 38. Lee, J.; Jang, D.S.; Kim, N.H.; Lee, Y.M.; Kim, J.; Kim, J.S. Galloyl glucoses from the seeds of Cornus ofﬁcinalis with inhibitory activity against protein glycation, aldose reductase, and cataractogenesis ex vivo. Biol. Pharm. Bull. 2011, 34, 443–446. [CrossRef] [PubMed] 39. Zheng, H.H.; Luo, C.T.; Chen, H.; Lin, J.N.; Ye, C.L.; Mao, S.S.; Li, Y.L. Xanthones from Swertia mussotii as multitarget-directed antidiabetic agents. Chem. Med. Chem. 2014, 9, 1374–1377. [CrossRef] [PubMed] 40. Lee, I.S.; Kim, I.S.; Lee, Y.M.; Lee, Y.; Kim, J.H.; Kim, J.S. 2′′,4′′-O-diacetylquercitrin, a novel advanced glycation end-product formation and aldose reductase inhibitor from Melastoma sanguineum. Chem. Pharm. Bull. (Tokyo) 2013, 61, 662–665. [CrossRef] [PubMed] y 41. Fatmawati, S.; Ersam, T.; Shimizu, K. The inhibitory activity of aldose reductase in vitro by constituents of Garcinia mangostana Linn. Phytomedicine 2015, 22, 49–51. [CrossRef] [PubMed] 42. Fan, Y.; Han, H.; He, C.; Yang, L.; Wang, Z. Identiﬁcation of the metabolites of gigantol in rat urine by ultra-performance liquid chromatography combined with electrospray ionization quadrupole time-of-ﬂight tandem mass spectrometry. Biomed. Chromatogr. 2014, 28, 1808–1815. [CrossRef] [PubMed] 43. Fang, H.; Hu, X.; Wang, M.; Wan, W.; Yang, Q.; Sun, X.; Gu, Q.; Gao, X.X.; Wang, Z.T.; Gu, L.Q.; et al. Anti-osmotic and antioxidant activities of gigantol from Dendrobium aurantiacum var. denneanum against cataractogenesis in galactosemic rats. J. Ethnopharmacol. 2015, 172, 38–46. [CrossRef] [PubMed] 44. Yang, L.; Wang, Z.; Xu, L. Phenols and a triterpene from Dendrobium aurantiacum var. denneanum (Orchidaceae). Biochem. Syst. Ecol. 2006, 34, 658–660. [CrossRef] 45. Hu, J.; Fan, W.; Dong, F.; Miao, Z.; Zhou, J. Chemical components of Dendrobium chrysotoxum. Chin. J. Chem. 2012, 30, 1327–1330. [CrossRef] 46. Wu, J.; Li, X.; Wan, W.; Yang, Q.; Ma, W.; Chen, D.; Hu, J.M.; Chen, C.Y.O.; Wei, X.Y. Gigantol from Dendrobium chrysotoxum Lindl. binds and inhibits aldose reductase gene to exert its anti-cataract activity: An in vitro mechanistic study. J. Ethnopharmacol. 2017, 198, 255–261. [CrossRef] [PubMed] 47. Lee, I.S.; Jung, S.H.; Lee, Y.M.; Choi, S.J.; Sun, H.; Kim, J.S. References Phenolic Compounds from the Leaves and Twigs of Osteomeles schwerinae That Inhibit Rat Lens Aldose Reductase and Vessel Dilation in Zebraﬁsh Larvae. J. Nat. Prod. 2015, 78, 2249–2254. [CrossRef] [PubMed] 48. Jang, D.S.; Yoo, N.H.; Kim, N.H.; Lee, Y.M.; Kim, C.S.; Kim, J.; Kim, J.H.; Kim, J.S. 3,5-Di-O-caffeoyl-epi-quinic acid from the leaves and stems of Erigeron annuus inhibits protein glycation, aldose reductase, and cataractogenesis. Biol. Pharm. Bull. 2010, 33, 329–333. [CrossRef] [PubMed] 49. Pauli, G.F.; Poetsch, F.; Nahrstedt, A. Structure assignment of natural quinic acid derivatives using proton nuclear magnetic resonance techniques. Phytochem. Anal. 1998, 9, 177–185. [CrossRef] 50. Kim, H.; Lee, Y.S. Identiﬁcation of new dicaffeoylquinic acids from Chrysanthemum morifolium and their antioxidant activities. Planta Med. 2005, 71, 871–876. [CrossRef] [PubMed] 37 of 41 37 of 41 Nutrients 2018, 10, 1580 51. Lee, J.; Lee, Y.M.; Lee, B.W.; Kim, J.H.; Kim, J.S. Chemical constituents from the aerial parts of Aster koraiensis with protein glycation and aldose reductase inhibitory activities. J. Nat. Prod. 2012, 75, 267–270. [CrossRef] [PubMed] 52. Jung, H.A.; Islam, M.D.N.; Kwon, Y.S.; Jin, S.E.; Son, Y.K.; Park, J.J.; Sohn, H.S.; Choi, J.S. Extraction and identiﬁcation of three major aldose reductase inhibitors from Artemisia montana. Food Chem. Toxicol. 2011, 49, 376–384. [CrossRef] [PubMed] 53. Do, T.H.; Tran, M.N.; Lee, I.S.; Yun, M.L.; Jin, S.K.; Jung, H.J.; Lee, S.M.; Na, M.K.; Bae, K. Inhibitors of aldose reductase and formation of advanced glycation end-products in Moutan cortex (Paeonia suffruticosa). J. Nat. Prod. 2009, 72, 1465–1470. 54. Kadota, S.; Terashima, S.; Basnet, P.; Kikuchi, T.; Namba, T.; Palbinone, A. Novel Terpenoid from Peaonia albiﬂora; Potent Inhibitory Activity on 3α-Hydroxysteroid Dehydrogenase. Chem. Pharm. Bull. 1993, 41, 487–489. [CrossRef] [PubMed] 55. Liu, J.; Nakamura, S.; Zhuang, Y.; Yoshikawa, M.; Mohamed, G.; Hussein, E.; Matsuo, K.; Matsuda, H. Medicinal Flowers. XXXX. 1) Structures of Dihydroisocoumarin Glycosides and Inhibitory Effects on Aldose Reducatase from the Flowers of Hydrangea macrophylla var. thunbergii. Chem. Pharm. Bull. 2013, 61, 655–661. [CrossRef] [PubMed] 56. Jung, H.J.; Jung, H.A.; Kang, S.S.; Lee, J.H.; Cho, Y.S.; Moon, K.H.; Choi, J.S. Inhibitory activity of aralia continentalis roots on protein tyrosine phosphatase 1B and rat lens aldose reductase. Arch. Pharm. Res. 2012, 35, 1771–1777. [CrossRef] [PubMed] 57. Jang, D.S.; Kim, J.M.; Lee, Y.M.; Kim, Y.S.; Kim, J.H.; Kim, J.S. Puerariafuran, a new inhibitor of advanced glycation end products (AGEs) isolated from the roots of Pueraria lobata. Chem. Pharm. Bull. 2006, 54, 1315–1317. References [CrossRef] [PubMed] 58. Kinjo, J.E.; Furusawa, J.I.; Baba, J.; Takeshita, T.; Yamasaki, M.; Nohara, T. Studies on the constituents of Pueraria lobata. III. Isoﬂavonoids and related compounds in the roots and the voluble stems. Chem. Pharm. Bull. (Tokyo) 1987, 35, 4846–4850. [CrossRef] 59. Kim, S.B.; Hwang, S.H.; Wang, Z.; Yu, J.M.; Lim, S.S. Rapid identiﬁcation and isolation of inhibitors of rat lens aldose reductase and antioxidant in Maackia amurensis. BioMed Res. Int. 2017, 2017. [CrossRef] [PubMed] 60. Kim, Y.S.; Kim, N.H.; Jung, D.H.; Jang, D.S.; Lee, Y.M.; Kim, J.M.; Kim, J.S. Genistein inhibits aldose reductase activity and high glucose induced TGF beta2 expression in human lens epithelial cells Eur J Pharmacol 59. Kim, S.B.; Hwang, S.H.; Wang, Z.; Yu, J.M.; Lim, S.S. Rapid identiﬁcation and isolation of inhibitors of rat lens aldose reductase and antioxidant in Maackia amurensis. BioMed Res. Int. 2017, 2017. [CrossRef] [PubMed] 60. Kim, Y.S.; Kim, N.H.; Jung, D.H.; Jang, D.S.; Lee, Y.M.; Kim, J.M.; Kim, J.S. Genistein inhibits aldose reductase activity and high glucose-induced TGF-beta2 expression in human lens epithelial cells. Eur. J. Pharmacol. 2008, 594, 18–25. [CrossRef] [PubMed] 61. Fatmawati, S.; Ersam, T.; Yu, H.; Zhang, C.; Jin, F.; Shimizu, K. 20(S)-Ginsenoside Rh2 as aldose reductase inhibitor from Panax ginseng. Bioorg. Med. Chem. Lett. 2014, 24, 4407–4409. [CrossRef] [PubMed] 62. Li, H.M.; Kim, J.K.; Jang, J.M.; Cui, C.B.; Lim, S.S. Analysis of the inhibitory activity of Abeliophyllum distichum leaf constituents against aldose reductase by using high-speed counter current chromatography. Arch. Pharm. Res. 2013, 36, 1104–1112. [CrossRef] [PubMed] 63. Yu, S.Y.; Lee, I.S.; Jung, S.H.; Lee, Y.M.; Lee, Y.R.; Kim, J.H.; Sun, H.; Kim, J.S. Caffeoylated phenylpropanoid glycosides from Brandisia hancei inhibit advanced glycation end product formation and aldose reductase in vitro and vessel dilation in larval zebraﬁsh in vivo. Planta Med. 2013, 79, 1705–1709. [CrossRef] [PubMed] 64. Aslam, H.; Tehseen, Y.; Maryam, K.; Uroos, M.; Siddiqui, B.S.; Hameed, A.; Iqbal, J. Bioorganic Chemistry Identiﬁcation of new potent inhibitor of aldose reductase from Ocimum basilicum. Bioorg. Chem. 2017, 75, 62–70. 65. Koo, D.C.; Baek, S.Y.; Jung, S.H.; Shim, S.H. Aldose reductase inhibitory compounds from extracts of Dipsacus asper. Biotechnol. Bioprocess Eng. 2013, 18, 926–931. [CrossRef] 66. Kim, J.K.; Lee, Y.S.; Kim, S.H.; Bae, Y.S.; Lim, S.S. Inhibition of aldose reductase by phenylethanoid glycoside isolated from the seeds of Paulownia coreana. Biol. Pharm. Bull. 2011, 34, 160–163. [CrossRef] [PubMed] 67. References Structural requirements of ﬂavonoids and related compounds for aldose reductase inhibitory activity. Chem. Pharm. Bull. (Tokyo) 2002, 50, 788–795. [CrossRef] [PubMed] 75. Matsuda, H.; Morikawa, T.; Toguchida, I.; Yoshikawa, M. Structural requirements of ﬂavonoids and related compounds for aldose reductase inhibitory activity. Chem. Pharm. Bull. (Tokyo) 2002, 50, 788–795. [CrossRef] [PubMed] 76. Abdo, S.M.; Hetta, M.H.; Samhan, F.A.; El Din, R.A.S.; Ali, G.H. Phytochemical and antibacterial s ﬁve freshwater algal species. Asian J. Plant Sci. Res. 2012, 11, 109–116. 77. Kumari, R.P.; Anbarasu, K. Protective role of C-phycocyanin against secondary changes during sodium selenite mediated cataractogenesis. Nat. Products Bioprospect. 2014, 4, 81–89. [CrossRef] [PubMed] 78. Kumari, R.P.; Sivakumar, J.; Thankappan, B.; Anbarasu, K. C-phycocyanin modulates selenite-induced cataractogenesis in rats. Biol. Trace Elem. Res. 2013, 151, 59–67. [CrossRef] [PubMed] 79. Liu, Q.; Huang, Y.; Zhang, R.; Cai, T.; Cai, Y. Medical Application of Spirulina platensis Derived C-Phycocyanin. Evid.-Based Complement Altern. Med. 2016, 2016. [CrossRef] 80. Kumari, R.P.; Ramkumar, S.; Thankappan, B.; Natarajaseenivasan, K.; Balaji, S.; Anbarasu, K. Transcriptional regulation of crystallin, redox, and apoptotic genes by C-Phycocyanin in the selenite-induced cataractogenic rat model. Mol. Vis. 2015, 21, 26. [PubMed] 1. Kothadia, A.D.; Shenoy, A.M.; Shabaraya, A.R.; Rajan, M.S.; Viradia, U.M.; Patel, N.H. Evaluation of Cata Preventive Action of Phycocyanin. Int. J. Pharm. Sci. Drug Res. 2011, 3, 42–44. 82. Ou, Y.; Yuan, Z.; Li, K.; Yang, X. Phycocyanin may suppress d-galactose-induced human lens epithelial cell apoptosis through mitochondrial and unfolded protein response pathways. Toxicol. Lett. 2012, 215, 25–30. [CrossRef] [PubMed] 83. Lee, S.J.; Park, W.H.; Park, S.D.; Moon, H.I. Aldose reductase inhibitors from Litchi chinensis Sonn. J. Enzyme Inhib. Med. Chem. 2009, 24, 957–959. [CrossRef] [PubMed] 84. Jung, H.A.; Yoon, N.Y.; Kang, S.S.; Kim, Y.S.; Choi, J.S. Inhibitory activities of prenylated ﬂavonoids from Sophora ﬂavescens against aldose reductase and generation of advanced glycation endproducts. J. Pharm. Pharmacol. 2008, 60, 227–236. [CrossRef] [PubMed] 85. Choi, S.J.; Kim, J.K.; Jang, J.M.; Lim, S.S. Inhibitory Effect of the Phenolic Compounds from Geranium thunbergii on Rat Lens Aldose Reductase and Galactitol Formation. Korean J. Med. Crop. Sci. 2012, 20, 222–230. [CrossRef] 86. Kim, J.M.; Jang, D.S.; Lee, Y.M.; Yoo, J.L.; Kim, Y.S.; Kim, J.H.; Kim, J.S. Aldose-reductase- and protein-glycation-inhibitory principles from the whole plant of Duchesnea chrysantha. Chem. Biodivers. 2008, 5, 352–356. [CrossRef] [PubMed] 87. Sawant, L.; Singh, V.K.; Dethe, S.; Bhaskar, A.; Balachandran, J.; Mundkinajeddu, D.; Agarwal, A. References Lee, Y.S.; Kang, Y.H.; Jung, J.Y.; Kang, I.J.; Han, S.N.; Chung, J.S.; Shin, H.K.; Lim, S.S. Inhibitory constituents of aldose reductase in the fruiting body of Phellinus linteus. Biol. Pharm. Bull. 2008, 31, 765–768. [CrossRef] [PubMed] 68. Jang, D.S.; Yoo, N.H.; Lee, Y.M.; Yoo, J.L.; Kim, Y.S.; Kim, J.S. Constituents of the ﬂowers of Erigeron annuus with inhibitory activity on the formation of advanced glycation end products (AGEs) and aldose reductase. Arch. Pharm. Res. 2008, 31, 900–904. [CrossRef] [PubMed] 69. Paek, J.H.; Shin, K.H.; Kang, Y.; Lee, J.; Lim, S.S. Rapid Identiﬁcation of Aldose Reductase Inhibitory Compounds from Perilla frutescens. BioMed Res. Int. 2013, 2013. [CrossRef] [PubMed] 38 of 41 38 of 41 Nutrients 2018, 10, 1580 70. Li, H.M.; Kim, J.K.; Jang, J.M.; Kwon, S.O.; Cui, C.B.; Lim, S.S. The inhibitory effect of Prunella vulgaris L. on aldose reductase and protein glycation. J. Biomed. Biotechnol. 2012, 2012. [CrossRef] [PubMed] 71. Matsumoto, T.; Nakamura, S.; Fujimoto, K.; Ohta, T.; Ogawa, K.; Yoshikawa, M.; Matsuda, H. Structure of constituents isolated from the ﬂower buds of Cananga odorata and their inhibitory effects on aldose reductase. J. Nat. Med. 2014, 68, 709–716. [CrossRef] [PubMed] 72. Islam, M.N.; Choi, S.H.; Moon, H.E.; Park, J.J.; Jung, H.A.; Woo, M.H.; Woo, H.C.; Choi, J.S. The inhibitory activities of the edible green alga Capsosiphon fulvescens on rat lens aldose reductase and advanced glycation end products formation. Eur. J. Nutr. 2014, 53, 233–242. [CrossRef] [PubMed] 73. Abdallah, H.M.; Salama, M.M.; Abd-Elrahman, E.H.; El-Maraghy, S.A. Antidiabetic activity of phenolic compounds from Pecan bark in streptozotocin-induced diabetic rats. Phytochem. Lett. 2011, 4, 337–341. [CrossRef] 74. Chethan, S.; Dharmesh, S.M.; Malleshi, N.G. Inhibition of aldose reductase from cataracted eye lenses by ﬁnger millet (Eleusine coracana) polyphenols. Bioorg. Med. Chem. 2008, 16, 10085–10090. [CrossRef] [PubMed] 75. Matsuda, H.; Morikawa, T.; Toguchida, I.; Yoshikawa, M. Structural requirements of ﬂavonoids and related d f ld d i hibi i i Ch Ph B ll (T k ) 2002 50 788 795 [C R f] 74. Chethan, S.; Dharmesh, S.M.; Malleshi, N.G. Inhibition of aldose reductase from cataracted eye lenses by ﬁnger millet (Eleusine coracana) polyphenols. Bioorg. Med. Chem. 2008, 16, 10085–10090. [CrossRef] [PubMed] 74. Chethan, S.; Dharmesh, S.M.; Malleshi, N.G. Inhibition of aldose reductase from cataracted eye lenses by ﬁnger millet (Eleusine coracana) polyphenols. Bioorg. Med. Chem. 2008, 16, 10085–10090. [CrossRef] [PubMed] 75. Matsuda, H.; Morikawa, T.; Toguchida, I.; Yoshikawa, M. References Aldose reductase and protein tyrosine phosphatase 1B inhibitory active compounds from Syzygium cumini seeds. Pharm. Biol. 2015, 53, 1176–1182. [CrossRef] [PubMed] 88. Jung, H.A.; Yoon, N.Y.; Bae, H.J.; Min, B.S.; Choi, J.S. Inhibitory activities of the alkaloids from Coptidis Rhizoma against aldose reductase. Arch. Pharm. Res. 2008, 31, 1405–1412. [CrossRef] [PubMed] 89. Palanisamy, U.D.; Ling, L.T.; Manaharan, T.; Appleton, D. Rapid isolation of geraniin from Nephelium lappaceum rind waste and its anti-hyperglycemic activity. Food Chem. 2011, 127, 21–27. [CrossRef] 90. Huang, G.J.; Hsieh, W.T.; Chang, H.Y.; Huang, S.S.; Lin, Y.C.; Kuo, Y.H. α-glucosidase and aldose reductase inhibitory activities from the fruiting body of Phellinus merrillii. J. Agric. Food Chem. 2011, 59, 5702–5706. [CrossRef] [PubMed] 39 of 41 Nutrients 2018, 10, 1580 91. Kim, T.H.; Kim, J.K.; Kang, Y.H.; Lee, J.Y.; Kang, I.J.; Lim, S.S. Aldose Reductase Inhibitory Activity of Compounds from Zea mays L. BioMed Res. Int. 2013, 2013, 1–8. 92. Morikawa, T.; Chaipech, S.; Matsuda, H.; Hamao, M.; Umeda, Y.; Sato, H.; Tamura, H.; Kon’i, H.; Ninomiya, K.; Yoshikawa, M.; et al. Antidiabetogenic oligostilbenoids and 3-ethyl-4-phenyl-3,4-dihydroisocoumarins from the bark of Shorea roxburghii. Bioorg. Med. Chem. 2012, 20, 832–840. [CrossRef] [PubMed] 93. Güvenç, A.; Okada, Y.; Akkol, E.K.; Duman, H.; Okuyama, T.; Çali¸s, I. Investigations of anti-inﬂammatory, antinociceptive, antioxidant and aldose reductase inhibitory activities of phenolic compounds from Sideritis brevibracteata. Food Chem. 2010, 118, 686–692. [CrossRef] 94. Rodriguez-Lyon, M.L.; Diaz-Lanza, A.M.; Bernabé, M.; Villaescusa-castillo, L. Flavone glycosides containing acetylated sugars from Sideritis hyssopifolia. Magn. Reson. Chem. 2000, 38, 684–687. [CrossRef] 95. Lenherr, A.; Mabry, T.J. Acetylated allose-containing ﬂavonoid glucosides from Stachys anisochila. Phytochemistry 1987, 26, 1185–1188. [CrossRef] 96. Patel, D.K.; Prasand, S.K.; Kumar, R.; Hemalatha, S. Cataract: A major secondary complication of diabetes, its epidemiology and an overview on major medicinal plants screened for anticataract activity. Asian Pac. J. Trop. Dis. 2011, 1, 323–329. [CrossRef] 97. Devi, V.G.; Rooban, B.N.; Sasikala, V.; Sahasranamam, V.; Abraham, A. Isorhamnetin-3-glucoside alleviates oxidative stress and opaciﬁcation in selenite cataract in vitro. Toxicol. In Vitro 2010, 24, 1662–1669. [CrossRef] [PubMed] 98. Lee, I.S.; Kim, Y.J.; Jung, S.H.; Kim, J.H.; Kim, J.S. Flavonoids from Litsea japonica Inhibit AGEs Formation and Rat Lense Aldose Reductase in Vitro and Vessel Dilation in Zebraﬁsh. Planta Med. 2017, 83, 318–325. [CrossRef] [PubMed] 99. Kim, S.B.; Hwang, S.H.; Suh, H.W.; Lim, S.S. Phytochemical analysis of Agrimonia pilosa Ledeb, its antioxidant activity and aldose reductase inhibitory potential. Int. J. Mol. Sci. References 2017, 18, 379. [CrossRef] [PubMed] 100. Kim, Y.S.; Jung, D.H.; Lee, I.S.; Choi, S.J.; Yu, S.Y.; Ku, S.K.; Kim, M.H.; Kim, J.S. Effects of Allium victorialis leaf extracts and its single compounds on aldose reductase, advanced glycation end products and TGF-β1 expression in mesangial cells. BMC Complement Altern. Med. 2013, 13, 251. [CrossRef] [PubMed] 101. Chung, I.M.; Kim, M.Y.; Park, W.H.; Moon, H.I. Aldose reductase inhibitors from Viola hondoensis W. Becker et H Boss. Am. J. Chin. Med. 2008, 36, 799–803. [CrossRef] [PubMed] 102. Chen, B.; Tian, J.; Zhang, J.; Wang, K.; Liu, L.; Yang, B.; Bao, L.; Liu, H. Triterpenes and meroterpenes from Ganoderma lucidum with inhibitory activity against HMGs reductase, aldose reductase and α-glucosidase. Fitoterapia 2017, 120, 6–16. [CrossRef] [PubMed] 103. Yang, M.; Wang, X.; Guan, S.; Xia, J.; Sun, J.; Guo, H.; Guo, D. Analysis of Triterpenoids in Ganoderma lucidum Using Liquid Chromatography Coupled with Electrospray Ionization Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2007, 18, 927–939. [CrossRef] [PubMed] 104. Fatmawati, S.; Shimizu, K.; Kondo, R. Ganoderic acid Df, a new triterpenoid with aldose reductase inhibitory activity from the fruiting body of Ganoderma lucidum. Fitoterapia 2010, 81, 1033–1036. [CrossRef] [PubMed] 104. Fatmawati, S.; Shimizu, K.; Kondo, R. Ganoderic acid Df, a new triterpenoid with aldose reductase inhibitory activity from the fruiting body of Ganoderma lucidum. Fitoterapia 2010, 81, 1033–1036. [CrossRef] [PubMed] 105. Fatmawati, S.; Shimizu, K.; Kondo, R. Inhibition of aldose reductase in vitro by constituents of Ganoderma 105. Fatmawati, S.; Shimizu, K.; Kondo, R. Inhibition of aldose reductase in vitro by constituents of Ganoderma lucidum. Planta Med. 2010, 76, 1691–1693. [CrossRef] [PubMed] 106. Fatmawati, S.; Kondo, R.; Shimizu, K. Structure-activity relationships of lanostane-type triterpenoids from Ganoderma lingzhi as α-glucosidase inhibitors. Bioorg. Med. Chem. Lett. 2013, 23, 5900–5903. [CrossRef] [PubMed] 107. Ramu, R.; Shirahatti, P.S.; Zameer, F.; Ranganatha, L.V.; Prasad, M.N.N. Inhibitory effect of banana (Musa sp. var. Nanjangud rasa bale) ﬂower extract and its constituents Umbelliferone and Lupeol on α-glucosidase, aldose reductase and glycation at multiple stages. S. Afr. J. Bot. 2014, 95, 54–63. [CrossRef] 108. Asha, R.; Devi, V.G.; Abraham, A. Chemico-Biological Interactions Lupeol, a pentacyclic triterpenoid isolated from Vernonia cinerea attenuate selenite induced cataract formation in Sprague Dawley rat pups. Chem. Biol. Interact. 2016, 245, 20–29. [CrossRef] [PubMed] 109. Chirumbolo, S. Anticataractogenic Activity of Luteolin. Chem. Biodivers. 2016, 13, 343–344. [CrossRef] [PubMed] 110. Vit, P.; Jacob, T.J. 110. Vit, P.; Jacob, T.J. Putative Anticataract Properties of Honey Studied by the Action of Flavonoids on a Lens Culture Model. J. Heal. Sci. 2008, 54, 196–202. [CrossRef] References Putative Anticataract Properties of Honey Studied by the Action of Flavonoids on a Lens Culture Model. J. Heal. Sci. 2008, 54, 196–202. [CrossRef] 40 of 41 40 of 41 Nutrients 2018, 10, 1580 111. Jang, D.S.; Lee, Y.M.; Jeong, I.H.; Kim, J.S. Constituents of the ﬂowers of Platycodon grandiﬂorum with inhibitory activity on advanced glycation end products and rat lens aldose reductase in vitro. Arch. Pharm. Res. 2010, 33, 875–880. [CrossRef] [PubMed] 111. Jang, D.S.; Lee, Y.M.; Jeong, I.H.; Kim, J.S. Constituents of the ﬂowers of Platycodon grandiﬂorum with inhibitory activity on advanced glycation end products and rat lens aldose reductase in vitro. Arch. Pharm. Res. 2010, 33, 875–880. [CrossRef] [PubMed] 112. Rooban, B.N.; Sasikala, V.; Devi, V.G.; Sahasranamam, V.; Abraham, A. Prevention of selenite induced oxidative stress and cataractogenesis by luteolin isolated from Vitex negundo. Chem. Biol. Interact. 2012, 196, 30–38. [CrossRef] [PubMed] 113. Ha, T.J.; Lee, J.H.; Lee, M.H.; Lee, B.W.; Kwon, H.S.; Park, C.H.; Shim, K.B.; Kim, H.T.; Baek, I.Y.; Jang, D.S. Isolation and identiﬁcation of phenolic compounds from the seeds of Perilla frutescens (L.) and their inhibitory activities against α-glucosidase and aldose reductase. Food Chem. 2012, 135, 1397–1403. [CrossRef] [PubMed] 114. Lee, J.; Rodriguez, J.P.; Quilantang, N.G.; Lee, M.H.; Cho, E.J.; Jacinto, S.D.; Lee, S. Determination of ﬂavonoids from Perilla frutescens var. japonica seeds and their inhibitory effect on aldose reductase. Appl. Biol. Chem. 2017, 60, 155–162. [CrossRef] 115. Morikawa, T.; Xie, H.; Wang, T.; Matsuda, H.; Yoshikawa, M. Bioactive constituents from Chinese natural medicines. XXXII. Aminopeptodase N and aldose reductase inhibitors from Sinocrassula indica: Structures of Sinocrassosides B4, B5, C1 and D1-D3. Chem. Pharm. Bull. (Tokyo) 2008, 56, 1438–1444. [CrossRef] [PubMed] 116. Matsuda, H.; Morikawa, T.; Tao, J.; Ueda, K.; Yoshikawa, M. Bioactive constituents of Chinese natural medicines. VII. Inhibitors of degranulation in RBL-2H3 cells and absolute stereostructures of three new diarylheptanoid glycosides from the bark of Myrica rubra. Chem. Pharm. Bull. (Tokyo) 2002, 50, 208–215. [CrossRef] [PubMed] 117. Hwang, S.H.; Kwon, S.H.; Kim, S.B.; Lim, S.S. Inhibitory Activities of Stauntonia hexaphylla Leaf Constituents on Rat Lens Aldose Reductase and Formation of Advanced Glycation End Products and Antioxidant. BioMed Res. Int. 2017, 2017, 4273257. [CrossRef] [PubMed] 118. Patel, M.B.; Mishra, S. Isoquinoline alkaloids from Tinospora cordifolia inhibit rat lens aldose reductase. Phyther. Res. 2012, 26, 1342–1347. [CrossRef] [PubMed] 119. References Nakamura, S.; Fujimoto, K.; Matsumoto, T.; Nakashima, S.; Ohta, T.; Ogawa, K.; Tamura, H.; Matsuda, H.; Yoshikawa, M. Structures of Acylated Sucroses and Inhibitory Effects of Constituents on Aldose Reducatase from the Flower Buds of Prunus mume. Chem. Pharm. Bull. 2013, 61, 445–451. 120. Nakamura, S.; Fujimoto, K.; Matsumoto, T.; Ohta, T.; Ogawa, K.; Tamura, H.; Matsuda, H.; Yoshikawa, M. Structures of acylated sucroses and an acylated ﬂavonol glycoside and inhibitory effects of constituents on aldose reductase from the ﬂower buds of Prunus mume. J. Nat. Med. 2013, 67, 799–806. [CrossRef] [PubMed] 121. Kim, N.H.; Kim, Y.S.; Lee, Y.M.; Jang, D.S.; Kim, J.S. Inhibition of aldose reductase and xylose-induced lens opacity by puerariafuran from the roots of Pueraria lobata. Biol. Pharm. Bull. 2010, 33, 1605–1609. [CrossRef] [PubMed] 122. Stobiecki, M.; Kachlicki, P. Isolation and identiﬁcation of ﬂavonoids. Sci. Flavonoids 2006, 27, 47–69. 123. Lee, H.E.; Kim, J.A.; Whang, W.K. Chemical constituents of smilax China l. stems and their inhibitory activities against glycation, aldose reductase, α-glucosidase, and lipase. Molecules 2017, 22, 451. [CrossRef] [PubMed] 124. Kato, A.; Yasuko, H.; Goto, H.; Hollinshead, J.; Nash, R.J.; Adachi, I. Inhibitory effect of rhetsinine isolated from Evodia rutaecarpa on aldose reductase activity. Phytomedicine 2009, 16, 258–261. [CrossRef] [PubMed] 125. Kang, J.; Tang, Y.; Liu, Q.; Guo, N.; Zhang, J.; Xiao, Z.; Chen, R.; Shen, Z. Isolation, modiﬁcation, and aldose 124. Kato, A.; Yasuko, H.; Goto, H.; Hollinshead, J.; Nash, R.J.; Adachi, I. Inhibitory effect of rhetsinine isolated from Evodia rutaecarpa on aldose reductase activity. Phytomedicine 2009, 16, 258–261. [CrossRef] [PubMed] J J y from Evodia rutaecarpa on aldose reductase activity. Phytomedicine 2009, 16, 258–261. [CrossRef] [PubMed] 125. Kang, J.; Tang, Y.; Liu, Q.; Guo, N.; Zhang, J.; Xiao, Z.; Chen, R.; Shen, Z. Isolation, modiﬁcation, and aldose reductase inhibitory activity of rosmarinic acid derivatives from the roots of Salvia grandifolia. Fitoterapia 2016, 112, 197–204. [CrossRef] [PubMed] 125. Kang, J.; Tang, Y.; Liu, Q.; Guo, N.; Zhang, J.; Xiao, Z.; Chen, R.; Shen, Z. Isolation, modiﬁcation, and aldose reductase inhibitory activity of rosmarinic acid derivatives from the roots of Salvia grandifolia. Fitoterapia 2016, 112, 197–204. [CrossRef] [PubMed] 126. Li, H.M.; Hwang, S.H.; Kang, B.G.; Hong, J.S.; Lim, S.S. Inhibitory effects of Colocasia esculenta (L.) Schott constituents on aldose reductase. Molecules 2014, 19, 13212–13224. [CrossRef] [PubMed] 127. References Lee, J.; Kim, N.H.; Nam, J.W.; Lee, Y.M.; Jang, D.S.; Kim, Y.S.; Nam, S.H.; Seo, E.K.; Yang, M.S.; Kim, J.S. Scopoletin from the ﬂower buds of Magnolia fargesii inhibits protein glycation, aldose reductase, and cataractogenesis Ex Vivo. Arch. Pharm. Res. 2010, 33, 1317–1323. [CrossRef] [PubMed] 128. Park, H.Y.; Kwon, S.B.; Heo, N.K.; Chun, W.J.; Kim, M.J.; Kwon, Y.S. Constituents of the stem of Angelica gigas with rat lens aldose reductase inhibitory activity. J. Appl. Biol. Chem. 2011, 54, 194–199. [CrossRef] 129. Abou Assi, R.; Darwis, Y.; Abdulbaqi, I.; Khan, A.A.; Lim, V.; Laghari, M.H. Morinda citrifolia (Noni): A comprehensive review on its industrial uses, pharmacological activities, and clinical trials. Arab. J. Chem. 2017, 10, 691–707. [CrossRef] 41 of 41 41 of 41 Nutrients 2018, 10, 1580 130. Wang, Q.; Qian, Y.; Wang, Q.; Yang, Y.F.; Ji, S.; Song, W.; Qiao, X.; Guo, D.; Liang, H.; Ye, M. Metabolites identiﬁcation of bioactive licorice compounds in rats. J. Pharm. Biomed. Anal. 2015, 115, 515–522. [CrossRef] [PubMed] 131. Lee, Y.S.; Kim, S.H.; Jung, S.H.; Kim, J.K.; Pan, C.H.; Lim, S.S. Aldose Reductase Inhibitory Compoun Glycyrrhiza uralensis. Biol. Pharm. Bull. 2010, 33, 917–921. [CrossRef] [PubMed] 132. Lee, Y.R.; Hwang, J.K.; Koh, H.W.; Jang, K.Y.; Lee, J.H.; Park, J.W.; Park, B.H. Sulfuretin, a major ﬂavonoid isolated from Rhus verniciﬂua, ameliorates experimental arthritis in mice. Life Sci. 2012, 90, 799–807. [CrossRef] [PubMed] 133. Lee, E.H.; Song, D.G.; Lee, J.Y.; Pan, C.H.; Um, B.H.; Jung, S.H. Inhibitory Effect of the Compounds Isolated from Rhus verniciﬂua on Aldose Reductase and Advanced Glycation Endproducts. Biol. Pharm. Bull. 2008, 31, 1626–1630. [CrossRef] [PubMed] 134. Lim, S.S.; Jung, S.H.; Ji, J.; Shin, K.H.; Keum, S.R. Inhibitory effects of 2′-hydroxychalcones on rat lens aldose reductase and rat platelet aggregation. Chem. Pharm. Bull. 2000, 48, 1786–1789. [CrossRef] [PubMed] 135. Lou, H.; Yuan, H.; Yamazaki, Y.; Sasaki, T.; Oka, S. Alkaloids and Flavonoids from Peanut Skins. Planta Med. 2001, 67, 345–349. [CrossRef] [PubMed] 136. Wei, X.; Chen, D.; Yi, Y.; Qi, H.; Gao, X.; Fang, H.; Gu, Q.; Wang, L.; Gu, L. Syringic acid extracted from Herba dendrobii prevents diabetic cataract pathogenesis by inhibiting aldose reductase activity. Evid.-Based Complement Altern. Med. 2012, 2012. [CrossRef] [PubMed] 137. Yan, R.Y.; Liu, H.L.; Zhang, J.Y.; Yang, B. Phenolic glycosides and other constituents from the bark of Magnolia ofﬁcinalis. J. Asian Nat. Prod. Res. 2014, 16, 400–405. [CrossRef] [PubMed] 138. Patel, M.B.; Mishra, S.M. References Aldose Reductase Inhibitory Activity of a Cglycosidic Flavonoid Derived from Enicostemma hyssopifolium. Inf. J. Complement. Integr. Med. 2009, 6, 1553–3840. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2889639319	http://www.scielo.br/pdf/bjb/v79n3/1519-6984-bjb-1519-6984181476.pdf	English	null	Volatilization of ammonia in systems of treatment of swine manure with aquatic macrophytes	Brazilian Journal of Biology	2,019	cc-by	6,483	Abstract The usage of aquatic plants represents an alternative in the treatment of residues originating from swine. In these systems, one of the N removal methods is the ammonium (NH4 +) uptake and volatilization of ammonia (NH3). In this way, the objective of this work was to evaluate the volatilization rates of NH3 in waste treatment systems swine fluids (SSF) with aquatic macrophytes, as well as the concentration of NH4 + present in the swine fluids. The experiment was carried out at Campus II/UNOESTE. The treatment systems were composed of 16 boxes of PVC and characterized as: T1 = Control sample 50% of SSF/50% of water; T2 = 50% SSF/50% water + Eichhornia crassipes (Mart.) Solms; T3 = 50% SSF/50% water + Pistia stratiotes L.; T4 = 50% SSF/50% water + Salvinia auriculata Aubl. The design was randomized blocks, with 4 treatments and 4 replicates. The hydrogen potential (pH) and the NH4 + content of the effluent were analyzed weekly, and the volatilization of NH3 by means of collectors installed in each treatment unit. The presence of aquatic macrophytes promoted the reduction of NH4 + concentration and of the pH values of swine fluids, and this resulted in the reduction of NH3 volatilization rates to the environment, with emphasis on the system with Eichhornia crassipes (Mart.) Solms, which presented the lowest rate of volatilization. 3 The presence of aquatic macrophytes promoted the reduction of NH4 + concentration and of the pH values of swine fluids, and this resulted in the reduction of NH3 volatilization rates to the environment, with emphasis on the system with Eichhornia crassipes (Mart.) Solms, which presented the lowest rate of volatilization. Keywords: greenhouse gases, phytoremediation, wastewater, environmental management. Resumo A utilização de plantas aquáticas representa uma alternativa no tratamento de resíduos oriundos da suinocultura. Nestes sistemas, uma das formas de remoção de nitrogênio (N) é a absorção de amônio (NH4 +) pelas plantas, entretanto, também ocorre a volatilização de amônia (NH3). Dessa forma, o objetivo do trabalho foi avaliar as taxas de volatilização de NH3 em sistemas de tratamentos de dejetos líquidos de suínos (DLS) com macrófitas aquáticas, bem como a concentração de NH4 + presente nos dejetos. O experimento foi realizado em área de ambiente aberto no Campus II/UNOESTE. Os sistemas de tratamento foram constituídos de 16 caixas de PVC e caracterizados como: T1 = Testemunha 50% de DLS/50% de água; T2 = 50% de DLS/50% de água + Eichhornia crassipes (Mart.) Solms; T3 = 50% de DLS/50% de água + Pistia stratiotes L.; T4 = 50% de DLS/50% de água + Salvinia auriculata Aubl. O delineamento adotado foi em blocos casualizados, com 4 tratamentos e 4 repetições. Foram analisados o potencial hidrogeniônico (pH) e o teor de NH4 + do efluente semanalmente, e a volatilização de NH3 por meio de coletores instalados em cada unidade de tratamento. A presença das macrófitas aquáticas proporcionou a redução da concentração de NH4 + e dos valores de pH dos dejetos líquidos de suínos, e isto resultou na redução das taxas de volatilização de NH3 ao meio ambiente, com destaque ao sistema com Eichhornia crassipes (Mart.) Solms, que apresentou a menor taxa de volatilização. Palavras-chave: gases do efeito estufa, fitorremediação, efluentes, manejo ambiental. Received: June 16, 2017 – Accepted: January 11, 2018 – Distributed: August 8, 2019 (With 7 figures) Received: June 16, 2017 – Accepted: January 11, 2018 – Distributed: August 8, 2019 (With 7 figures) Brazilian Journal of Biology https://doi.org/10.1590/1519-6984.181476 Original Article https://doi.org/10.1590/1519-6984.181476 Original Article https://doi.org/10.1590/1519-6984.181476 Brazilian Journal of Biology ISSN 1519-6984 (Print) ISSN 1678-4375 (Online) Brazilian Journal of Biology ISSN 1519-6984 (Print) ISSN 1678-4375 (Online) Volatilization of ammonia in systems of treatment of swine manure with aquatic macrophytes C. D. Pinaffia* and C. H. Santosa aUniversidade do Oeste Paulista – UNOESTE, Rodovia Raposo Tavares, Km 572, Limoeiro, CEP 19026-310, Presidente Prudente, SP, Brasil e-mail: camila_pinaffi@hotmal.com aUniversidade do Oeste Paulista – UNOESTE, Rodovia Raposo Tavares, Km 572, Limoeiro, CEP 19026-310, Presidente Prudente, SP, Brasil e-mail: camila_pinaffi@hotmal.com Received: June 16, 2017 – Accepted: January 11, 2018 – Distributed: August 8, 2019 (With 7 figures) 2. Material and Methods The experiment, carried out in May and June of 2016, was conducted in an open environment at Campus II/UNOESTE, whose geographic coordinates are: Latitude 22° 07’ S e Longitude 51° 09’ W, Presidente Prudente, São Paulo, Brazil. Located in a defined climate region, according to the climatic classification of Köppen, as Aw – tropical wet with dry winter, and average annual temperature of 29.2 °C. The average annual rainfall is 1254.9 mm (CEPAGRI, 2017). The systems made with aquatic macrophytes are used to reduce the nutrient concentration of pig slurry, minimize the impacts on aquatic ecosystems (Poach et al., 2003), and provide an operationally passive form of wastewater treatment (Hunt and Poach, 2001; Kadlec and Knight, 1996). Furthermore, they can effectively treat great amounts of animal excrements (Knight et al., 2000), especially in the nitrogen (N) removal (Poach et al., 2003). The treatment systems were composed of 16 polyvinyl chloride (PVC) containers with dimensions of 76.5 cm in height, 101.5 cm in superior diameter and 73.0 cm in inferior diameter, making a volume of 0.32 m3, in which only 0.25 m3 are used per container, margin adopted to avoid overflow. Pig slurry was used, coming from the zootechnical center of Campus II/UNOESTE, for the composition of the treatments, which are: T1 = Control sample – 50% pig slurry and 50% water; T2 = 50% pig slurry and 50% water + Eichhornia crassipes (Mart.) Solms (Figure 1); T3 = 50% According to Vesilind and Morgan (2011), the N is an important element in biological reactions, and may be connected to components that produce a lot of energy, such as amino acids and amines, also known as organic N. An intermediate element formed during the biological metabolism é the ammoniacal N. In accordance to Peng et al. (2005), the organic and ammoniacal N are the main forms presented in wastewater, and are considered indicators of recent pollution. The organic N is converted to ammoniacal under anaerobic and aerobic conditions, with the reduction of ammoniacal nitrogen concentrations contributing to the reduction of total nitrogen, and the temperature and pH have an impact on the bioactivity and volatilization processes. Figure 1. Eichhornia crassipes (Mart.) Solms (Lutzenberger, 1985). The wetlands, systems artificially designed for utilizing aquatic macrophytes, remove N through sedimentation, absorption, organic matter accumulation, microbial assimilation, nitrification/denitrification and volatilization of ammonia (Brix, 1993; Johnston, 1991; Poach et al., 2003). 1. Introduction scenario, where Brazil is the fourth largest producer and exporter of pork (ABCS, 2014). Pig farming represents a sector of great economic and social importance for Brazil (Gonzatto et al., 2013), responsible for increasing exports of meat and its industrialized products, and also for the generation of jobs. Brazilian pig farming occupies a prominent position in the world However, pig farming is characterized as an activity with great potential for pollution, due to the effluent generation normally in the liquid form, with high load of organic and 423 Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 Pinaffi, C.D. and Santos, C.H. Pinaffi, C.D. and Santos, C.H. Based on the previously said, this work aims at evaluating the losses of N to the atmosphere by volatilization of ammonia (NH3) in pig slurry treatment systems, with and without the presence of aquatic plants, as well as the concentration of ammonium (NH4 +) present in pig slurry, with the hypothesis that aquatic plants are able to reduce the concentration of NH4 + present in swine manures, as well as to reduce losses of N by NH3 volatilization. nutrient matter (mainly nitrogen and phosphorus, and also, potassium, calcium, sodium, magnesium, manganese, iron, zinc and copper) (Steinmetz et al., 2009). The pollutant load of liquid pig slurry may adversely affect environments as the water bodies, promoting the growth of microorganisms and the occurrence of eutrophication (Meade et al., 2011). Excess nutrients, applied to the soil at rates higher than those for capture, may get in surface and groundwater due to the flow and leaching (Stone et al., 1998). To reduce the nutrient load into the environment, alternative or additional ways of treating wastewater should be implemented. An option for the additional treating is phytoremediation, that is, the use of plants and associated microorganisms as an instrument for containment, isolation, removal or reduction of contaminant concentrations in solid, liquid or gaseous media (EPA, 2000), at safe levels compatible with the protection of human health, as well as preventing the spread of harmful substances to the environment (Andrade et al., 2007). 2. Material and Methods Sooknah and Wilkie (2004) verify the reduction of N because of the direct absorption by the aquatic macrophytes to the nitrification carried out by the nitrifying bacteria and the volatilization of non-ionized ammonia (NH3) which occurs predominantly in high pH environments. Ammonia emissions characterize a major threat to the environment, due to their implications, such as changes in the rainfall pH, contributions to the greenhouse effect, as well as effects on human and farmed animal health (Felix and Cardoso, 2004). Thus, monitoring these emissions is something fundamentally important for the control of atmospheric pollution. Figure 1. Eichhornia crassipes (Mart.) Solms (Lutzenberger, 1985). Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 424 Volatilization of ammonia in system with macrophytes Guarim Neto, 2005). Exotic in origin, it is distributed in all continents, in tropical semitropical latitudes and in some temperate countries (Metcalf and Tchobanoglous, 1991) and, in the absence of nutritional limitations, it rapidly develops in hot climate regions (Andrade et al., 2007). pig slurry and 50% water + Pistia stratiotes L. (Figure 2); T4 = 50% pig slurry and 50% water + Salvinia auriculata Aubl. (Figure 3), distributed in a randomized block design (RBD), in split plots, with four repetitions. The plots are represented by the treatments (with and without the presence of aquatic macrophytes) and the split plots, by the collection periods. The Pistia stratiotes L. is commonly known as water lettuce, belonging to the Araceae family (Kissmann and Groth, 1997), extensively distributed throughout the world. Its origin, which is attributed to Africa or South America (Lorenzi, 1982; Cardoso, et al., 2005), has not yet been fully defined. According to Pott and Pott (2000), P. stratiotes is considered a cosmopolitan tropical and subtropical species, being widely distributed throughout Brazil, occurring both in natural ecosystems and in aquatic environments impacted by anthropic activities (Henry-Silva and Camargo, 2000a). The Eichhornia crassipes (Mart.) Solms is characterized by being a free floating aquatic plant, native of South America, belonging to the monocotyledonous class, Pontederidaceae family, Pontederiales order (Esteves, 1998), and by being an angiosperm with perennial life cycle (Bortolotto and Figure 2. Pistia stratiotes L. (Pott and Pott, 2000). Belonging to the family Salviniaceae, the Salvinia auriculata Aubl. is a free floating aquatic plant, of annual or perennial occurrence (Oliveira, 1981). Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 2. Material and Methods NH3 collector scheme (adapted from Ros et al., 2005). Table 1. Values of F for pH, NH4 + and the volatilization of NH3 calculated based on the Tukey Test, at the level of 5% probability, in the treatments with aquatic species, in collection periods. Variation factors pH NH4 + NH3 (mg L-1) Treatment (a) 27.58* 12.69* 15.05* Collection periods (b) 239.36* 299.37* 28.67* Interaction (a) × (b) 3.62* 0.90ns 4.00* CV (a) (%) 1.49 10.47 31.25 DMS (a) 0.22 0.96 91.04 CV (b) (%) 1.47 20.81 30.33 DMS (b) 0.21 0.94 81.62 *significant at the level of 1% probability (p<0.01); ns = not significant. Figure 4. NH3 collector scheme (adapted from Ros et al., 2005). The pH values (Figure 5) decreased, throughout the collection periods, in all treatments. It was possible to notice that in the period of 7 days, the Eichhornia crassipes (Mart.) Solms provided a significant reduction in the pH compared to the others, and according to Granato (1995), Eichhornia crassipes (Mart.) Solms, in contact with alkaline solutions, has the ability of decreasing the pH of these solutions, due to the absorption of the potassium, calcium and magnesium nutrients, respectively, which are all present in the medium. According to Esteves (1998), the organic matter decomposition process reduces the pH, since there is an increase in the carbon dioxide (CO2) concentration in the medium and, at the same time, the consumption of dissolved oxygen. Lin et al. (2005) also observed a reduction in the pH of the effluent from de Litopenaeus vannamei nurseries treated with constructed wetland. According to Shah et al. (2015), a pH of 6.0-9.0 is the most adequate for the aquatic macrophytes performance and, in accordance with CONAMA (2011) Resolution no. 430/2011, the pH values of the effluent from any polluting source, to be directly released in the receiving body, must be between 5.0 and 9.0. Thus, the results found in the treatments with plant species are in agreement with the ones required by the Brazilian legislation. at the top with a polyethylene plate with 20.0 m in diameter, and a 2.0 cm opening between the plate and the collector base was kept, for establishing a steam pressure deficit and, thus, the NH3 volatilization could happen. 2. Material and Methods It presents a wide native distribution in the neotropics, extending from Mexico and the Galapagos Islands through Central America and the Antilles and most of South America to the south of Brazil (Sculthorpe, 1967). The aquatic macrophytes used were collected in lentic lakes from lands located in the West Paulista region, São Paulo, selected as young plants, with an established root system and aerial part of similar appearance. They were initially submitted to the environmental adaptation in boxes containing only water, at the experiment site, for a week. After this period, the treatment was initialized, and the amount of plants inserted in each experimental unit was determined to maintain an occupation of approximately 80% of the experimental units, as described by Henry-Silva and Camargo (2008). During the experimental period, 16 samples of the effluent to analyse the pH and ammoniacal nitrogen in the ionized form (NH4 +) were weekly collected, at the periods of 0 (day in which the experiment was implanted), 7, 14, 21 and 30 days. The pH measurements were obtained using the Micronal bench digital pH meter. The determination of the NH4 + content was obtained by the Kjeldahl method, consisted of three stages: sample digestion, distillation with the Kjeldahl nitrogen distiller and titration with sulfuric acid, through the method presented by Malavolta et al. (1997). The method is based on the decomposition of organic matter by digestion of the sample with concentrated sulfuric acid at 350 °C, in the presence of catalyst salts that accelerate the oxidation of organic matter. The digestion will be terminated after obtaining a colorless or slightly greenish liquid. The nitrogen present in the resulting acid solution is determined by steam distillation, collected by the boric acid 2% (m/v) solution and indicators, followed by titration with sulfuric acid (0.02 N).i Figure 2. Pistia stratiotes L. (Pott and Pott, 2000). Figure 3. Salvinia auriculata Aubl. (Gomes, 2011). For the quantification of NH3 volatilization a PVC collector base (Figure 4), with 15.0 cm in diameter and 14.0 cm in height, was placed in each treatment box, fixed to a 4.0 cm thick Styrofoam board, with a center opening beneath the surface of the effluent. Each base was protected Figure 3. Salvinia auriculata Aubl. (Gomes, 2011). Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 425 Pinaffi, C.D. and Santos, C.H. Figure 4. 2. Material and Methods Inside the PVC base a screen with an effluent height of 4.0 cm was place, fixed to a glass petri dish which housed in its interior a sponge with 2.0 cm thick and 7.0 cm width. The sponges, moistened with 30.0 mL of phosphoric acid (0.167 mol L-1) to capture the volatilized NH3 of the effluent, were replaced at the periods of 1, 4, 8, 12, 19 and 27 days after placing the collectors. In the collections, the sponges were gathered, stored in a plastic box with a lid and immediately taken to the laboratory of clinical analysis of vegetal tissues/UNOESTE to extract the ammonium phosphate solution ((NH4)3PO4) formed from the chemical reaction between NH3 and phosphoric acid (H3PO4). Their washings were performed with 500.0 mL deionized water, in five continuous washings of 100.0 mL each. Finally, an aliquot of 20.0 mL of the solution was submitted to distillation with the Kjheldad distiller by the method described by Cantarella and Trivelin (2001).fi The efficiencies of the treatments in the reduction of the content of NH4 + were calculated according to Equation 1: (%) [( ) / ] 100 E Ci Cf Ci = − × (1) (1) For Henry-Silva and Camargo (2000b), the highest pH found only in the effluent is probably related to the photosynthesis of the phytoplankton present in the site, which, by assimilating the CO2 available in the water, increases the pH values of the medium.fl In which: E = Removal efficiency (%); Ci = Initial concentration (mg L-1); Cf = Final concentration (mg L-1). The results were put under analysis of variance by the Test F and the comparison of the means by the Tukey Test, at the level of 5% probability, with software Assistat 7.7. Silva et al. (2014) found out that the effluent from water supply channels with tambaqui farming treated with Eichhornia crassipes (Mart.) Solms showed lower pH values when compared to the treatments without plants. According to the authors, it happens because of the removal of bases that enable the growth of Eichhornia crassipes (Mart.) Solms. Gentelini et al. (2008) observed that the organic pisciculture effluent, before passing through the system with aquatic macrophytes, was slightly alkaline, with a pH of 7.03, becoming somewhat Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 3. Results and Discussion The values of Test F, obtained by the contrast of means between the treatments (Table 1), show that there was a significant difference between the plant species used and the collection periods for the following variables: pH ammonium (NH4 +) and ammonia (NH3). Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 426 Volatilization of ammonia in system with macrophytes Volatilization of ammonia in system with macrophytes when analysing the refrigerated effluent treatment system with Eichhornia crassipes (Mart.) Solms, found that the industrial effluent presented average concentration of 79.85 mg L-1 of NH4 +, and the reductions of this element acid after crossing the system. This same pattern of pH reduction, after passing through the treatment system, was observed by Henry-Silva and Camargo (2006) with three floating macrophytes, Eichhornia crassipes (Mart.) Solms, Pistia stratiotes L. and Salvinia molesta D. S. Mitch., in the treatment of pisciculture effluent.fl Table 2. Initial and final values of the content of NH4 + in the treatments with aquatic species. Treatments NH4 + (mg L-1) Initial Final E (%) Control sample 9.87 2.38 75.9 Eichhornia crassipes (Mart.) Solms 8.89 0.58 93.4 Pistia stratiotes L. 9.52 3.15 66.9 Salvinia auriculata Aubl. 10.29 2.45 76.2 E(%): Removal efficiency. Table 2. Initial and final values of the content of NH4 + in the treatments with aquatic species. In relation to the NH4 + concentration in the effluent, it should be noted that there was no significant interaction between the treatments (without and with the plants) and between the collection periods. However, after evaluating the means obtained in each treatment (Figure 6), it was possible to notice that the Eichhornia crassipes (Mart.) Solms species statically differed from the others by presenting a lower NH4 + content in its effluent and, therefore, higher efficiency in the removal of this element (93.4%, according Table 2) in a 30 day detention time. Reidel et al. (2005), Figure 5. pH values obtained in the treatments with aquatic species, in five collection periods. Reference value according to CONAMA Resolution no 430/2011 = pH: 5-9. DMS for collection periods = 0.21; CV (%) = 1.47 classification on lowercase letters. DMS for aquatic species = 0.22; CV (%) = 1.49 classification on capital letters. Tukey Test was applied at the level of 5% probability. Figure 5. pH values obtained in the treatments with aquatic species, in five collection periods. 3. Results and Discussion Sezerino and Philippi (2000) highlight that, in such treatment systems with plants, approximately 74.0% from the NH4 + removal may be associated with the plant uptake. CONAMA Resolution no. 430/2011 puts a limit in the concentration of the NH4 + ion to values of 20 mg L-1 in effluents from any polluting source. Therefore, the values obtained in the treatments submitted to the aquatic plants (Figure 6) are in accordance with the Brazilian legislation. The NH3 volatilization (desorption to the atmosphere) is a physical process of disengaging this gas from the dissociation of the NH4 + ion in aquatic environments with high pH values (Assunção, 2009). The NH3 volatilization was observed in the interaction between treatments with aquatic macrophytes and the evaluation periods (Figure 7). Gonzatto et al. (2013), when evaluating, under field conditions, the NH3 volatilization and N2O emission after applying the pig slurry in maize, observed that approximately 80.0% of the N losses by volatilization happened in the first 22 hours after applying the slurry, quickly reducing with time. According to the authors, this kinetic observed in the NH3 emissions may be attributed to the high content of NH4 + of the pig slurry, which is one of the factors that influence the NH3 volatilization (Sommer and Hutchings, 2001). According to Figure 7, this volatilization was higher in the control sample treatment, stressing that the differentiation occurred, mainly, after the period of 12 days of collection. It is also stated that the systems containing aquatic species provided lower losses of NH3, with a noticeable decrease during the evaluation period. According to Körner and Vermaat (1998) and Sooknah and Wilkie (2004) in treatment systems with aquatic macrophytes, the removal of N happens by the direct absorption of the plant, by the action of microorganisms fixed in the roots and by the NH3 volatilization itself. Harper et al. (2004) state that the main factors that can be correlated to the NH3 volatilization in pig ponds are the wind speed, temperature, NH4 + concentration and the effluent pH. 3. Results and Discussion Reference value according to CONAMA Resolution no 430/2011 = pH: 5-9. DMS for collection periods = 0.21; CV (%) = 1.47 classification on lowercase letters. DMS for aquatic species = 0.22; CV (%) = 1.49 classification on capital letters. Tukey Test was applied at the level of 5% probability. Figure 6. Available NH4 + concentration, in mg L-1, in the treatments with aquatic species. Reference value according to CONAMA Resolution no 430/2011 = NH4 +: 20.0 mg L-1. DMS = 0.96; CV (%) = 10.47. The means followed by the same letter are not statistically different among themselves. Tukey Test was applied at the level of 5% probability. Figure 6. Available NH4 + concentration, in mg L-1, in the treatments with aquatic species. Reference value according to CONAMA Resolution no 430/2011 = NH4 +: 20.0 mg L-1. DMS = 0.96; CV (%) = 10.47. The means followed by the same letter are not statistically different among themselves. Tukey Test was applied at the level of 5% probability. Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 427 Pinaffi, C.D. and Santos, C.H. Pinaffi, C.D. and Santos, C.H. For Kiehl (1985), the NH3 losses increase when the compound reaction is in the alkalinity zone and when the pH reaches values above 8.0 and 9.0, a large part of N turned into NH3, and may be lost in the atmosphere. The balance represented by the reaction NH4 + ↔ NH3 + H+, at pH values around neutrality, is shifted to the left, and there is a predominance of NH3 only to pH values above 8.5 (SCHMIDELL et al., 2007). Probably, due to the fact that the pH of the effluent was higher (above 8.0) in the first days (0 to 7 days) (Figure 5), the release of NH3 was higher. Mkhabela et al. (2009) stated that the initial high pH of the slurry is a factor that contributes to such emissions, and Meade et al. (2011) observed that 95% of the NH3 emissions happened in the first 24 hours after applying the slurry.i were 58.9% for a five day detention time, 86.1% for seven days and 97.7% for ten days, thus increasing the removal efficiency with an increased hydraulic detention time. 3. Results and Discussion Thus, it is observed that the NH4 + concentrations found in the pig slurry (Figure 6) were higher in the beginning of the experiment; besides, in the aquatic macrophytes treatment, the NH3 volatilization may have been reduced from the 12 days of collection, compared to the control sample treatment (Figure 7), due to the high absorption of NH4 + by the plants, as can be observed in Figure 6. In the treatment with Eichhornia crassipes (Mart.) Solms, It is noteworthy that at the initial collection period (1 to 4 days), the NH3 concentration was higher in the four treatments proposed. According to Assunção (2009), when the pH is high, he balance between the free NH3 and NH4 + ion tends to shift toward the NH3 formation. For values above 9.26 there’s a predominance of NH3 (practically 100% at pH near 11.0) and for values close to neutrality (between 6.0 and 7.0), practically all NH3 is in the ionized form. Figure 7. Available NH3 concentration, in mg L-1, in the treatments with aquatic species, in six collection periods. DMS for collection periods = 81.62; CV (%) = 30.33 classification on lowercase letters. DMS for aquatic species = 91.04; CV(%) = 31.25 classification on capital letters. Tukey Test was applied at the level of 5% probability. Figure 7. Available NH3 concentration, in mg L-1, in the treatments with aquatic species, in six collection periods. DMS for collection periods = 81.62; CV (%) = 30.33 classification on lowercase letters. DMS for aquatic species = 91.04; CV(%) = 31.25 classification on capital letters. Tukey Test was applied at the level of 5% probability. Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 428 Volatilization of ammonia in system with macrophytes CANTARELLA, H. and TRIVELIN, P.C.O., 2001. Determinação de nitrogênio total em solos. In: B. RAIJ, J.C. ANDRADE, H. CANTARELLA and J.A. QUAGGIO, eds. Análise química para avaliação da fertilidade de solos tropicais. Campinas: Instituto Agronômico de Campinas, pp. 262-269. the NH4 + concentrations in the effluent were lower and statistically differ from the others.i Zimmo et al. (2003) verified that the NH3 volatilization rates in ponds with algae were higher than in ponds with aquatic plants (Lemna gibba). According to the authors, it can be explained by the lower NH3 values in ponds with plants due to the shadowing and to the lower pH values. 3. Results and Discussion In this case, the volatilization rate correlated to the free NH3 concentration in the water of the pond. CARDOSO, L.R., MARTINS, D., MORI, E.S. and TERRA, M.A., 2005. Variabilidade genética entre populações de Pistia Stratiotes. Planta Daninha, vol. 23, no. 2, pp. 181-185. http:// dx.doi.org/10.1590/S0100-83582005000200003. CENTRO DE PESQUISAS METEOROLÓGICAS E CLIMÁTICAS APLICADAS A AGRICULTURA – CEPAGRI, 2017 [viewed 24 April 2017]. Clima dos municípios paulistas [online]. Campinas: Unicamp. Available from: http://www.cpa.unicamp.br/outras- informacoes/clima_muni_467.html 3 When considering the percentages of reduction of the NH3 volatilization rates, it is possible to notice that the treatment systems with the species Eichhornia crassipes (Mart.) Solms and Salvinia auriculata Aubl. were more expressive, 77.8% and 76.2% in a 27 day retention time. This reduction in the NH3 volatilization shows the importance of the presence of aquatic plants in these treatment systems. Oron et al. (1988), when evaluating the treatment of domestic sewage under cultivation of floating aquatic plants of the family Lemnaceae, obtained NH3 removal efficiency of 90.0%, with a 10 day retention time and initial concentration of 520 mg L-1 of COD. Researches carried out in wetlands that treated pig slurry waters mention that the NH3 volatilization represented less than 20.0% of the N removed by the wetlands (Poach et al., 2002). CONSELHO NACIONAL DO MEIO AMBIENTE – CONAMA, 2011. Resolução Conama nº 430 de 13 de maio de 2011. Dispõe sobre as condições e padrões de lançamento de efluentes, complementa e altera a Resolução n. 357, de 17 de março de 2005, do Conselho Nacional do Meio Ambiente - CONAMA. Diário Oﬁcial da República Federativa do Brasil, Brasília, DF, n. 92, 16 maio, p. 89. ESTEVES, F.A., 1998. Fundamentos de limnologia. 2nd ed. Rio de Janeiro: Interciência. 602 p. FELIX, E.P. and CARDOSO, A.A., 2004. Amônia (NH3) atmosférica: fontes, transformação, sorvedouros e métodos de análise. Quimica Nova, vol. 27, no. 1, pp. 123-130. http://dx.doi. org/10.1590/S0100-40422004000100022. 4. Conclusions GENTELINI, A.L., GOMES, S.D., FEIDEN, A., ZENATTI, D., SAMPAIO, S.C. and COLDEBELLA, A., 2008. Produção de biomassa das macrófitas aquáticas Eichhornia crassipes (aguapé) e Egeria densa (egeria) em sistema de tratamento de efluente de piscicultura orgânica. Ciências Agrárias, vol. 29, no. 2, pp. 441-448. http://dx.doi.org/10.5433/1679-0359.2008v29n2p441. The presence of aquatic macrophytes promoted the reduction of NH4 + concentration and of the pH values of swine fluids, and this resulted in the reduction of NH3 volatilization rates to the environment, with emphasis on the system with Eichhornia crassipes (Mart.) Solms, which presented the lowest rate of volatilization. GOMES, M.A.C., 2011. Efeito da salinidade sobre a biomassa, morfologia e fisiologia de Salvinia auriculata Aubl. Campos dos Goytacazes: Universidade Estadual do Norte Fluminense Darcy Ribeiro, 90 p. Dissertação de Mestrado em Biociências e Biotecnologia. References ANDRADE, J.C.M., TAVARES, S.R.L. and MAHLER, C.F., 2007. Fitorremediação: o uso de plantas na melhoria da qualidade ambiental. São Paulo: Oficina de Textos. 176 p. GONZATTO, R., MIOLA, E.C.C., DONEDA, A., PUJOL, S.B., AITA, C. and GIACOMINI, S.J., 2013. Volatilização de amônia e emissão de óxido nitroso após aplicação de dejetos líquidos de suínos em solo cultivado com milho. Ciência Rural, vol. 43, no. 9, pp. 1590-1596. http://dx.doi.org/10.1590/S0103- 84782013000900009. ASSOCIAÇÃO BRASILEIRA DOS CRIADORES DE SUÍNOS – ABCS, 2014. Produção de suínos: teoria e prática. 1st ed. Brasília: ABCS. 905 p. GRANATO, M., 1995. Utilização do aguapé no tratamento de efluentes com cianetos. Rio de Janeiro: CETEM/CNPq. 39 p. Série Tecnologia Ambiental, vol. 5. ASSUNÇÃO, F.A.L., 2009. Estudo da remoção de nitrogênio, com ênfase na volatilização de amônia, em lagoas de polimento de efluentes de reatores UASB tratando esgotos urbanos de Belo Horizonte/MG. Belo Horizonte: Universidade Federal de Minas Gerais, 89 p. Dissertação de Mestrado em Saneamento, Meio Ambiente e Recursos Hídricos. HARPER, L.A., SHARPE, R.R., PARKIN, T.B., VISSCHER, A., VAN CLEEMPUT, O. and BYERS, F.M., 2004. Nitrogen cycling through swine production systems: Ammonia, dinitrogen, and nitrous oxide emissions. Journal of Environmental Quality, vol. 33, no. 4, pp. 1189-1201. http://dx.doi.org/10.2134/jeq2004.1189. PMid:15254100. BORTOLOTTO, I.M. and GUARIM NETO, G., 2005. O uso do camalote, Eichhornia crassipes (Mart.) Solms, Pontederiaceae, BORTOLOTTO, I.M. and GUARIM NETO, G., 2005. O uso do camalote, Eichhornia crassipes (Mart.) Solms, Pontederiaceae, para confecção de artesanato no Distrito de Albuquerque, Corumbá, MS, Brasil. Acta Botanica Brasílica, vol. 19, no. 2, pp. 331-337. http://dx.doi.org/10.1590/S0102-33062005000200016. para confecção de artesanato no Distrito de Albuquerque, Corumbá, MS, Brasil. Acta Botanica Brasílica, vol. 19, no. 2, pp. 331-337. http://dx.doi.org/10.1590/S0102-33062005000200016. HENRY-SILVA, G.G. and CAMARGO, A.F.M., 2000a. Composição química de quatro espécies de macrófitas aquáticas e possibilidade de uso de suas biomassas. Naturalia, vol. 26, no. 8, pp. 111-125. BRIX, H., 1993. Wastewater treatment in constructed wetlands: system design, removal processes, and treatment performance. In: G.A. MOSHIRI, ed. Constructed wetlands for water quality improvement. Boca Raton: CRC Press, pp. 9-22. HENRY-SILVA, G.G. and CAMARGO, A.F.M., 2000b. Impacto do lançamento de efluentes urbanos sobre alguns ecossistemas Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 429 Pinaffi, C.D. and Santos, C.H. aquáticos do município de Rio Claro (SP). Revista Ciências Biológicas e do Ambiente, vol. 2, pp. 317-330. MKHABELA, M.S., GORDON, R., BURTON, D., SMITH, E. References Plantas aquáticas do Pantanal. Brasília: Embrapa; Corumbá: Centro de Pesquisa Agropecuária do Pantanal. 404 p. REIDEL, A., DAMASCENO, S., ZENATTI, D.C., SAMPAIO, S.C., FEIDEN, A. and QUEIROZ, M.M.F., 2005. Utilização de efluente de frigorífico, tratado com macrófita aquática, no cultivo de tilápia do Nilo. Revista Brasileira de Engenharia Agrícola e Ambiental, vol. 9, suppl, pp. 181-185. KÖRNER, S. and VERMAAT, J.E., 1998. The relative importance of Lemna gibba, bacteria and algae for the nitrogen and phosphorus removal in duckweed – covered domestic wastewater. Water Research, vol. 32, no. 12, pp. 3651-366. http://dx.doi.org/10.1016/ S0043-1354(98)00166-3. ROS, C.O., AITA, C. and GIACOMINI, S.J., 2005. Volatilização de amônia com aplicação de uréia na superfície do solo, no sistema plantio direto. Ciência Rural, vol. 35, no. 4, pp. 799-805. http:// dx.doi.org/10.1590/S0103-84782005000400008. LIN, Y.F., JING, S.R., LEE, D.Y., CHANG, Y.F., CHEN, Y.M. and SHIH, K.C., 2005. Performance of a constructed wetland treating intensive shrimp aquaculture wastewater under high hydraulic loading rate. Environmental Pollution, vol. 134, no. 3, pp. 411-421. PMid:15620586. SCHMIDELL, W., SOARES, H.M., ETCHEBEHERE, C., MENES, R.J., BERTOLA, N.C. and CONTRERAS, E.M., 2007. Tratamento biológico de águas residuárias. Florianópolis: Tribo da Ilha. LORENZI, H. 1982. Plantas daninhas do Brasil: terrestres, aquáticas, parasitas, tóxicas e medicinais. São Paulo: Nova Odessa. 425 p. SCULTHORPE, C.D., 1967. The biology of aquatic vascular plants. London: Edward Arnold Ltd. 610 p. LUTZENBERGER, J., 1985. Ecologia: do jardim do poder. Porto Alegre: L&PM Editores, 102 p. SEZERINO, P.H. and PHILIPPI, L.S., 2000. Utilização de um sistema experimental por meio de “wetland” construído no tratamento de esgotos domésticos pós tanque séptico. In: Anais do Simpósio Luso-Brasileiro de Engenharia Sanitária e Ambiental; 2000; Porto Seguro. Rio de Janeiro: ABES, pp. 688-697. SEZERINO, P.H. and PHILIPPI, L.S., 2000. Utilização de um sistema experimental por meio de “wetland” construído no tratamento de esgotos domésticos pós tanque séptico. In: Anais do MALAVOLTA, E., VITTI, G.C. and OLIVEIRA, S.A., 1997. Avaliação do estado nutricional das plantas: princípios e aplicações. 2nd ed. Piracicaba: Potafós. 319 p. Simpósio Luso-Brasileiro de Engenharia Sanitária e Ambiental; 2000; Porto Seguro. Rio de Janeiro: ABES, pp. 688-697. MEADE, G., PIERCE, J.V., O’DOHERTY, C., MUELLER, G., LANIGAN, G. and MC CABE, T., 2011. Ammonia and nitrous oxide emissions following land application of high and low nitrogen pig manures to winter wheat at three growth stages. Agriculture, Ecosystems & Environment, vol. 140, no. 1-2, pp. 208-217. http://dx.doi.org/10.1016/j.agee.2010.12.007. SHAH, M., HASHMI, H.N., GHUMMAN, A.R. References and MADANI, A., 2009. The impact of management practices and meteorological conditions on ammonia and nitrous oxide emissions following application of hog slurry to forage grass in Nova Scotia. Agriculture, Ecosystems & Environment, vol. 130, no. 1-2, pp. 41-49. http://dx.doi.org/10.1016/j.agee.2008.11.012. HENRY-SILVA, G.G. and CAMARGO, A.F.M., 2006. Efficiency of aquatic macrophytes to treat Nile tilapia pond effluents. Scientia Agrícola, vol. 63, no. 5, pp. 433-438. http://dx.doi.org/10.1590/ S0103-90162006000500003. OLIVEIRA, P., 1981. Plantas de aquário. Lisboa: Editorial Presença. 238 p. HENRY-SILVA, G.G. and CAMARGO, A.F.M., 2008. Tratamento de efluentes de carcinicultura por macrófitas aquáticas flutuantes. HENRY-SILVA, G.G. and CAMARGO, A.F.M., 2008. Tratamento de efluentes de carcinicultura por macrófitas aquáticas flutuantes. Revista Brasileira de Zootecnia, vol. 37, no. 2, pp. 181-188. http:// dx.doi.org/10.1590/S1516-35982008000200002. ORON, G., DE-VEGT, A. and PORATH, D., 1988. Nitrogen removal and conversion by duckweed grown on wastewater. Water Research, vol. 22, no. 2, pp. 179-184. http://dx.doi. org/10.1016/0043-1354(88)90076-0. p q Revista Brasileira de Zootecnia, vol. 37, no. 2, pp. 181-188. http:// dx.doi.org/10.1590/S1516-35982008000200002. HUNT, P.G. and POACH, M.E., 2001. State of the art for animal wastewater treatment in constructed wetlands. Water Science and Technology, vol. 44, no. 11-12, pp. 19-25. http://dx.doi. org/10.2166/wst.2001.0805. PMid:11804093. PENG, J.F., WANG, B.Z. and WANG, L., 2005. Mult- stage ponds-wetlands ecosystem for effective wastewater treatment. Journal of Zhejiang University. Science, vol. 6B, no. 5, pp. 346- 352. http://dx.doi.org/10.1631/jzus.2005.B0346. JOHNSTON, C.A., 1991. Sediment and nutrient retention by freshwater wetlands: effects on surface water quality. Critical Reviews in Environmental Control, vol. 21, no. 5-6, pp. 491-565. http://dx.doi.org/10.1080/10643389109388425. POACH, M.E., HUNT, P.G., SADLER, E.J., MATHENY, T.A., JOHNSON, M.H., STONE, K.C., HUMINEK, F.J. and RICE, J.M., 2002. Ammonia volatilization from constructed wetlands that treat swine wastewater. Transactions of the ASAE. American Society of Agricultural Engineers, vol. 45, no. 3, pp. 619-627. http://dx.doi.org/10.13031/2013.8825. KADLEC, R.H. and KNIGHT, R.L., 1996. Treatment wetlands. Boca Raton: Lewis Publishers. 893 p. KIEHL, E.J., 1985. Fertilizantes orgânicos. Piracicaba: Agronômica Ceres. 492 p. POACH, M.E., HUNT, P.G., VANOTTI, M.B., STONE, K.C., MATHENY, T.A., JOHNSON, M.H. and SADLER, E.J., 2003. Improved nitrogen by constructed wetlands receiving partially nitrified swine manure. Ecological Engineering, vol. 20, no. 2, pp. 183-197. http://dx.doi.org/10.1016/S0925-8574(03)00024-7. KISSMANN, K.G. and GROTH, D., 1997. Plantas infestantes e nocivas. 2nd ed. São Paulo: BASF Brasileira. KNIGHT, R.L., PAYNE-JÚNIOR, V.W.E., BORER, R.E., CLARKE-JÚNIOR, R.A. and PRIES, J.H., 2000. Constructed wetlands for livestock wastewater management. Ecological Engineering, vol. 15, no. 1-2, pp. 41-55. http://dx.doi.org/10.1016/ S0925-8574(99)00034-8. POTT, V.J. and POTT, A., 2000. Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 References and ZEESHAN, M., 2015. Performance assessment of aquatic macrophytes for treatment of municipal wastewater. Journal of the South African Institution of Civil Engineering, vol. 57, no. 3, pp. 18-25. http:// dx.doi.org/10.17159/2309-8775/2015/V57N3A3. SILVA, A.D.R., SANTOS, R.B., BRUNO, A.M.S.S., GENTELINI, A.L., SILVA, A.H.G. and SOARES, E.C., 2014. Eficiência do aguapé sobre variáveis limnológicas em canais de abastecimento METCALF, E. and TCHOBANOGLOUS, G., 1991. Wastewater engineering: treatment, disposal and reuse. 3rd ed. New York: MacGraw Hill. Braz. J. Biol., 2019, vol. 79, no. 3, pp. 423-431 430 Volatilization of ammonia in system with macrophytes utilizados no cultivo de tambaqui. Acta Amazonica, vol. 44, no. 2, pp. 255-262. http://dx.doi.org/10.1590/S0044-59672014000200011. utilizados no cultivo de tambaqui. Acta Amazonica, vol. 44, no. 2, pp. 255-262. http://dx.doi.org/10.1590/S0044-59672014000200011. STONE, K.C., HUNT, P.G., HUMENIK, F.J. and JOHNSON, M.H., 1998. Impact of swine waste application on ground and stream water quality in an Eastern Coastal Plain watershed. American Society of Agricultural and Biological Engineer, vol. 41, no. 6, pp. 1665-1670. http://dx.doi.org/10.13031/2013.17342. SOMMER, S.G. and HUTCHINGS, N.J., 2001. Ammonia emission from field applied manure and its reduction: invited paper. European Journal of Agronomy, vol. 15, no. 1, pp. 1-15. http://dx.doi.org/10.1016/S1161-0301(01)00112-5. UNITED STATES ENVIRONMENTAL PROTECTION AGENCY – EPA, 2000 [viewed 03 November 2017]. Introduction to phytoremediation. Cincinnati, Ohio: EPA. EPA/600/R-99/107. Available from: http://nepis.epa.gov SOOKNAH, R.D. and WILKIE, A.C., 2004. Nutrient removal by floating aquatic macrophytes cultured in anaerobically disgested flushed dairy manure wastewater. Ecological Engineering, vol. 22, no. 1, pp. 27-42. http://dx.doi.org/10.1016/j.ecoleng.2004.01.004. VESILIND, P.A. and MORGAN, S.M., 2011. Introdução à engenharia ambiental. 2nd ed. São Paulo: Cengage Learning. 438 p. ZIMMO, O., VAN DER STEEN, N. and GIJZEN, H., 2003. Comparison of ammonia volatilization rates in algae and duckweed- based waste stabilization ponds treating domestic wastewater. Water Research, vol. 37, no. 19, pp. 4587-4594. http://dx.doi. org/10.1016/j.watres.2003.08.013. PMid:14568043. STEINMETZ, R.L.R., KUNZ, A., DRESSLER, V.L., FLORES, É.M.M. and MARTINS, A.F., 2009. Study of metal distribution in raw end screened swine manure. Clean Soil Air Water, vol. 37, no. 3, pp. 239-244. http://dx.doi.org/10.1002/clen.200800156. 431
https://openalex.org/W4389048099	https://bmchealthservres.biomedcentral.com/counter/pdf/10.1186/s12913-023-10270-8	English	null	Burnout, employee engagement, and changing organizational contexts in VA primary care during the early COVID-19 pandemic	BMC health services research	2,023	cc-by	9,852	This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permit- ted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecom- mons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Burnout, employee engagement, and changing organizational contexts in VA primary care during the early COVID‑19 pandemic Eric A. Apaydin1,2, Danielle E. Rose1, Michael R. McClean1, David C. Mohr3,4, Elizabeth M. Yano1,5,6, Paul G. Shekelle1,6, Karin M. Nelson7,8, Rong Guo1, Caroline K. Yoo1,5 and Susan E. Stockdale1,9 BMC Health Services Research BMC Health Services Research Apaydin et al. BMC Health Services Research (2023) 23:1306 https://doi.org/10.1186/s12913-023-10270-8 Open Access Abstract Apaydin eric.apaydin@va.gov Full list of author information is available at the end of the article Abstract Background The COVID-19 pandemic involved a rapid change to the working conditions of all healthcare work- ers (HCW), including those in primary care. Organizational responses to the pandemic, including a shift to virtual care, changes in staffing, and reassignments to testing-related work, may have shifted more burden to these HCWs, increasing their burnout and turnover intent, despite their engagement to their organization. Our objectives were (1) to examine changes in burnout and intent to leave rates in VA primary care from 2017–2020 (before and during the pandemic), and (2) to analyze how individual protective factors and organizational context affected burnout and turn- over intent among VA primary care HCWs during the early months of the pandemic. Methods We analyzed individual- and healthcare system-level data from 19,894 primary care HCWs in 139 healthcare systems in 2020. We modeled potential relationships between individual-level burnout and turnover intent as out- comes, and individual-level employee engagement, perceptions of workload, leadership, and workgroups. At health- care system-level, we assessed prior-year levels of burnout and turnover intent, COVID-19 burden (number of tests and deaths), and the extent of virtual care use as potential determinants. We conducted multivariable analyses using logistic regression with standard errors clustered by healthcare system controlled for individual-level demographics and healthcare system complexity. Results In 2020, 37% of primary care HCWs reported burnout, and 31% reported turnover intent. Highly engaged employees were less burned out (OR = 0.57; 95% CI 0.52–0.63) and had lower turnover intent (OR = 0.62; 95% CI employees were less burned out (OR = 0.57; 95% CI 0.52–0.63) and had lower turnover intent (OR = 0.62; 95% CI 0.57–0.68). Pre-pandemic healthcare system-level burnout was a major predictor of individual-level pandemic burn- out (p = 0.014). Perceptions of reasonable workload, trustworthy leadership, and strong workgroups were also related to lower burnout and turnover intent (p < 0.05 for all). COVID-19 burden, virtual care use, and prior year turnover were not associated with either outcome. Conclusions Employee engagement was associated with a lower likelihood of primary care HCW burnout and turn- over intent during the pandemic, suggesting it may have a protective effect during stressful times. COVID-19 burden Correspondence: Eric A. Apaydin eric.apaydin@va.gov Full list of author information is available at the end of the article Correspondence: Eric A. Apaydin eric.apaydin@va.gov Full list of author information is available at the end of the article p Eric A. Introduction Crises such as the COVID-19 pandemic can have wide- spread impacts on the healthcare system, resulting in organizational changes that can negatively impact health- care worker burnout and turnover. During previous crises, like the 2005 Kashmir earthquake [1], the 2011 Fukushima earthquake, tsunami, and nuclear incident [2], and 2017 Harvey and Maria hurricanes [3], burnout and other mental health issues increased among healthcare, disaster, and related workers. Healthcare worker (HCW; i.e., providers and staff) burnout, primarily characterized by emotional exhaustion and depersonalization, [4] has been widespread across many specialties, in the US [5, 6] and globally [7–9], during the COVID-19 pandemic. Turnover intent, or the intent to leave one’s job, among HCWs was also high in the US [10] and internationally [11, 12] during the pandemic. Prior to the pandemic, the burnout phenomenon [13] was associated with both individual and organizational drivers [14], and linked to numerous negative healthcare consequences [14–16], including increased medical errors, reduced patient sat- isfaction, poorer quality of care, and increased turnover. Similarly, previous research has identified turnover as a possible driver of primary care shortages, [17, 18] and found that it is very expensive, with the estimated costs of replacing a single HCW ranging from $14,000 for a medical assistant [19] to $1,000,000 for a physician [20].h Some characteristics of individual HCWs have been shown to be protective against burnout and turnover. Specifically, more engaged employees may be more resil- ient and less susceptible to burnout. Engagement is char- acterized by energy, involvement, and effectiveness on the job [24]. A systematic review and meta-analysis [25] of burnout and engagement in 37 studies across job fields showed that higher engagement and lower burnout were consistently correlated with fewer health complaints, more job satisfaction, and more organizational commit- ment. To date, employee engagement has not been well- studied in primary care. Protective factors, like employee engagement and positive working environments (e.g., a workplace with a sense of community), may attenuate increases in HCW burnout [5] and turnover intentions [26], but it is unknown whether engagement can buffer the impact of crisis-induced organizational changes on burnout and turnover. Good leadership and strong workgroups may have also been protective of burnout among individual HCWs dur- ing the pandemic. High quality leadership, characterized by values like trust, respect, mentorship, and inspira- tion, has been consistently associated [14, 27] with lower burnout among physicians and other HCWs. This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permit- ted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecom- mons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Apaydin et al. BMC Health Services Research (2023) 23:1306 Page 2 of 12 and virtual care use were not related to either outcome. Future research should focus on understanding the relation- ship between engagement and burnout and improving well-being in primary care. Keywords Burnout, Primary care, Healthcare workforce, Employee engagement, COVID-19, Virtual care and virtual care use were not related to either outcome. Future research should focus on understanding the relation- ship between engagement and burnout and improving well-being in primary care. Keywords Burnout, Primary care, Healthcare workforce, Employee engagement, COVID-19, Virtual care and social distancing restrictions, may have increased primary care HCW burnout via changes in work flows, workload, learning how to use the virtual care platforms, and/or educating patients on how to use the modality [10, 23]. Conversely, virtual care use may have reduced HCW stress and associated burnout by reducing the risk of COVID-19 infection. Data sources The outcomes included individual-level burnout and turnover intent. Burnout was assessed using a single-item measure for emotional exhaustion (“I feel burned out from my work.”) and a single-item measure for deperson- alization (“I worry that this job is hardening me emotion- ally.”). Both were measured on a 7-point frequency scale (never, a few times a year or less, once a month or less, a few times a month, once a week, a few times a week, or every day). For analysis, we created a dichotomous measure of burnout, with 1 = experiencing symptoms once a week or more versus 0 = less than once per week on either the emotional exhaustion or the depersonaliza- tion item, as in previous studies [34, 35]. Similarly, turno- ver intent was a dichotomous measure, with 1 = a “yes” response (“yes, but taking another job within VA”, “yes, to retire,” “yes, to take another job within the federal gov- ernment,” “yes, to take another job outside the federal government,” and “yes, other”) to the question of “Are you considering leaving your job in the next year, and if so why?” and 0 = “no”. Data sources included survey data from the 2017 to 2020 VA All Employee Surveys (AES) and 2020 administrative data from the VA COVID Shared Data Resource (COVID SDR), the VA Corporate Data Warehouse (CDW), and the VHA Service Support Center (VSSC).h The AES is a yearly survey of VA employee attitudes [31] and is typically administered in June (see Table 1 for yearly response rates), except in 2020 when COVID delayed administration until September. The AES is anonymous at the individual-level, with identifiers for an individual’s healthcare system. Each annual survey is cross-sectional, and responses are not linked across years. The CDW is a national repository of VA clinical and administrative information that includes clinical, administrative, financial, enrollment, and benefits data [32]. The VSSC contains information on healthcare sys- tem complexity, detailed below. The COVID SDR [33] was specifically created at the beginning of the pandemic by the VA to pool COVID-19 clinical data from all VA healthcare systems. Introduction Coworkers also matter – evidence suggests that in VA primary care, workgroups with good collaboration and communication and high competency [28–30] have members that are less likely to be burned out. Strong leadership and good relationships with coworkers were likely very important during the chaotic times of the early pandemic, and these characteristics may have protected against higher HCW burnout. The COVID-19 pandemic tested the capacity of health- care systems to respond to a crisis, requiring healthcare system administrators to adapt organizational structures and processes overnight and changing organizational contexts in ways that may have contributed to higher lev- els of burnout and turnover among primary care health- care workers (HCWs). Little is known, however, about how COVID crisis-induced changes may have impacted primary care HCW burnout. Crisis levels of acute and emergent care often required surge staffing, [21] which may have included temporarily reassigning primary care HCWs into COVID-related screening or inpatient care. These staffing changes may have contributed to increased burnout among the reassigned HCWs exposed to new stressors and among the remaining HCWs with increased workloads. Those HCWs with higher workloads may have experienced increased burnout, as has been documented in research prior to the pandemic [14]. In addition, the rapid shift towards virtual care (e.g., video- or telephone- based care), [22] due to COVID-related quarantining Our objectives in this study were to 1) examine changes in reported VA primary care HCW burnout and turno- ver intent prior to and during the early COVID-19 crisis and 2) examine how individual protective characteristics (engagement, and perceptions of workload, leadership, and workgroups) and COVID-related organizational contextual factors (via COVID-19 burden, virtual care use, and prior year burnout and turnover intent), were Page 3 of 12 Apaydin et al. BMC Health Services Research (2023) 23:1306 Apaydin et al. BMC Health Services Research (2023) 23:1306 (administrative or clerical employee working in a clinical area, or a general administrative employee). associated with burnout and turnover intent during the first six months of the pandemic. associated with burnout and turnover intent during the first six months of the pandemic. Abbreviations: AES All Employee Survey, d denominator who answered question, DP depersonalization, EE emotional exhaustion, n nume question affirmatively, VA Veterans Health Administration Sampleh The sample included primary care HCWs who completed the AES surveys from 2017–2020. Data from 2017–19 was also used to describe pre-pandemic trends and to create healthcare system-level predictors for burnout and turnover intent for the year prior to the beginning of the pandemic (2019). Eligible respondents indicated that they worked on a VA primary care team in one of the following professions: primary care provider (PCP: physician [MD/DO], nurse practitioner [NP], and physi- cian assistant [PA]), registered nurse (RN), clinical asso- ciate (licensed practical nurse [LPN], licensed vocational nurse [LVN], nursing assistant, intermediate care techni- cian, and health technician), and administrative associate We also include several individual-level factors that that may be protective against burnout and turnover intent.hi The first, employee engagement, was a factor that we hypothesize may be protective against crisis-induced organizational stressors, and that previous studies have suggested might be protective against burnout and turn- over intent. Individual-level employee engagement was assessed using a 5-point agreement scale ranging from “strongly disagree” to “strongly agree,” ranked from 1 to 5, for four statements: “I recommend my organization as a good place to work.”; “This organization really inspires the very best in me in the way of job performance.”; “I always do more than is actually required.”; and “My job Table 1 Average burnout and turnover intent for VA primary care HCWs from 2017–20 AES Abbreviations: AES All Employee Survey, d denominator who answered question, DP depersonalization, EE emotional exhaustion, n numerator who responded to question affirmatively, VA Veterans Health Administration Year Response rate (VA-wide) Burnout (% EE or DP > = once a week or more) n/d Turnover Intent (%, 1 year) n/d 2017 59.5 41.6 6170/14825 33.0 4821/14623 2018 61.6 35.7 5781/16201 34.5 5720/16649 2019 63.9 35.4 6191/17504 34.3 6177/18017 2020 69.4 37.2 7320/19680 30.6 6096/19894 ble 1 Average burnout and turnover intent for VA primary care HCWs from 2017–20 AES Abbreviations: AES All Employee Survey, d denominator who answered question, DP depersonalization, EE emotional exhaustion, n numerator who responded to question affirmatively, VA Veterans Health Administration Apaydin et al. BMC Health Services Research (2023) 23:1306 Page 4 of 12 Page 4 of 12 measure for healthcare system-level proportion of vir- tual care use was created by dividing total virtual primary care visits by all primary care visits (in-person + virtual visits) for each healthcare system. Other covariates d d l l l Individual-level demographic covariates included profes- sional role (provider, registered nurse, clinical associate, or administrative associate); gender (male or female); race (white, black/African American, Asian, American Indian or Alaskan Native, or Native Hawaiian or other Pacific Islander); ethnicity (non-Hispanic or Hispanic); age (29 and under, 30–49, or 50 +); VA tenure (less than 2 years, between 2 and 10 years, between 10 and 20 years, or more than 20 years); and supervisor status (yes or no). We also included a healthcare system-level complexity measure using the VHA Complexity Model [37] catego- ries available from VSSC: group 1 (combining groups 1a, 1b, and 1c; most complex), group 2, and group 3 (least complex). In general, group 1 healthcare systems have medium-to-high patient volumes, medium-to-high risk patients, more complex clinical programs, and have medium-to-large teaching and research programs [38]. Group 2 and 3 healthcare systems generally have low- to-medium patient volumes, lower risk patients, and less complex clinical programs. Both group 2 nor 3 healthcare systems typically have small or no teaching or research programs. Individual-level demographic covariates included profes- sional role (provider, registered nurse, clinical associate, or administrative associate); gender (male or female); race (white, black/African American, Asian, American Indian or Alaskan Native, or Native Hawaiian or other Pacific Islander); ethnicity (non-Hispanic or Hispanic); age (29 and under, 30–49, or 50 +); VA tenure (less than 2 years, between 2 and 10 years, between 10 and 20 years, or more than 20 years); and supervisor status (yes or no). y ); p (y ) We also included a healthcare system-level complexity measure using the VHA Complexity Model [37] catego- ries available from VSSC: group 1 (combining groups 1a, 1b, and 1c; most complex), group 2, and group 3 (least complex). In general, group 1 healthcare systems have medium-to-high patient volumes, medium-to-high risk patients, more complex clinical programs, and have medium-to-large teaching and research programs [38]. Group 2 and 3 healthcare systems generally have low- to-medium patient volumes, lower risk patients, and less complex clinical programs. Both group 2 nor 3 healthcare systems typically have small or no teaching or research programs. Healthcare system‑level predictors of burnout and turnover intenti Organizational contextual factors specific to the COVID crisis included healthcare system-level COVID-19 bur- den, virtual care use, and prior year average HCW burn- out and turnover intent. Using data from the COVID SDR, COVID-19 burden was operationalized as the number of COVID-19 tests administered and deaths per 1000 patients from 03/15/20 to 09/15/20 (i.e., the first six months of the pandemic and six months prior to the administration of the AES 2020 survey). These measures were averaged by healthcare system and grouped into quartiles (tests) or terciles (deaths) for analysis, to adjust for uneven distribution of data and ease interpretation of results. Testing rates were grouped into four catego- ries: 9.2 to 38.7 (lowest quartile), 38.9 to 47.6 (2nd quar- tile), 47.8 to 61.4 (3rd quartile), and 65.2 to 471.7 tests per 1000 unique patients (highest quartile). Death rates were grouped into three categories: 0.0 to 0.27 (lowest tercile), 0.27 to 0.69 (middle tercile), and 0.69 to 3.65 deaths per 1000 unique patients (highest tercile).h Sampleh is more than just a paycheck to me.” Respondents were rated as highly engaged if they agreed or strongly agreed with most items in this set (i.e., a score of 18 or above) [36]. All four items loaded onto one factor with an eigen- value greater than 1 (Cronbach’s alpha = 0.76). We used 2019 AES data to create a measure of the proportion of primary care HCWs with high emotional exhaustion or depersonalization in 2019 by healthcare system, using the same definition of dichotomous burn- out described above. A similar healthcare system-level turnover intent predictor variable was also constructed using 2019 AES data. We also included three other individual-level factors that may be associated with burnout and turnover intent: perceptions of workload, leadership, and workgroup. These individual factors and their corresponding survey items are as follows: reasonable workload (“My work- load is reasonable.”), supervisor listening (“My supervisor listens to what I have to say.”), supervisor respect (“My supervisor treats me with respect.”), supervisor trust (“I have trust and confidence in my supervisor.”), workgroup cooperation (“The people I work with cooperate to get the job done.”), workgroup competency (“My work unit has the job-relevant knowledge and skills necessary to accomplish organizational goals.”), and workgroup col- laboration (“Workgroups collaborate to accomplish shared objectives.”). All of these factors were assessed using 5-point agreement scales, ranging from “strongly disagree” to “strongly agree,” and ranked from 1 to 5. Respondents who “agreed” or “strongly agreed” with a factor’s survey item were considered to endorse that factor. Descriptive statistics for 2020 analytic sample Descriptive statistics for 2020 analytic sample In 2020, high emotional exhaustion and deperson- alization burnout were reported by 34.2% and 26.6% of respondents, respectively (Table 2). In the same year, almost 31% of the sample indicated their intent to leave VA within the next year. Nearly 40% of respondents reported high employee engagement, and over half (57.4%) stated that their workload was reasonable. Over 70% of respondents felt that their supervisors listened, were respectful and were trustworthy, and over 60% agreed that their workgroups cooperated, collaborated, and were competent. Respondents were mostly female (74.5%), white (56.5%), and non-Hispanic (78.8%). Most of the sample was 30 to 49 years of age (45.4%) or 50 years or older (47.9%), and nearly half had 2–10 years of VA experience (44.4%). Supervisors were the minority among respond- ents (27.4%), and registered nurses (33.2%) and clinical associates (29.3%) were the most common professions in the sample. Organizational context regarding COVID burden (Table 3) showed high variation across healthcare sys- tems for COVID-19 testing per 1000 patients (mean [M] 56.5, standard deviation [SD] 46.9; range [R] 9.2–471.7), COVID-specific mortality rates per 1000 patients (M 0.46; SD 0.52; R 0.0–3.65), and virtual primary care visits (M 29%, SD 21%, R 0–76%). Less variation was observed in prior year healthcare system-level average burnout (M 31%, SD 3%, R 24–41%) and intent to leave within one year (M 35%, SD 3%, R 26–44%). Statistical analyses W d d i i BMC Health Services Research (2023) 23:1306 Page 5 of 12 Table 2 Individual characteristics (n = 19,894) Characteristic n % Professional role Provider (MD/DO, NP, PA) 4,851 24.4 RN 6,635 33.4 Clinical associate 5,877 29.5 Administrative associate 2,531 12.7 Gender Male 4,544 22.8 Female 14,824 74.5 Unknown 526 2.6 Race White 11,241 56.5 Black or African American 4,199 21.1 Asian 2,216 11.1 American Indian or Alaskan Native 490 2.5 Native Hawaiian or other Pacific Island 381 1.9 Unknown 1,367 6.9 Ethnicity Non-Hispanic 15,680 78.8 Hispanic 1,796 9.0 Unknown 2,418 12.2 Age 29 and under 684 3.4 30–49 9,028 45.4 50 + 9,534 47.9 Unknown 648 3.3 VA tenure Less than 2 years 4,494 22.6 Between 2 and 10 years 8,841 44.4 Between 10 and 20 years 4,477 22.5 More than 20 years 1,624 8.2 Unknown 458 2.3 Supervisor status No 14,143 71.1 Yes 5,445 27.4 Unknown 306 1.5 High employee engagement (mostly agree or strongly agree on four items) No 12,005 60.3 Yes 7,889 39.7 High EE or DP burnout (once a week or more for either) No 12,360 62.1 Yes 7,320 36.8 Unknown 214 1.1 High EE burnout (once a week or more) No 12,801 64.4 Yes 6,798 34.2 Unknown 295 1.5 High DP burnout (once a week or more) No 14,288 71.8 Yes 5,285 26.6 Table 2 Individual characteristics (n = 19,894) complexity as control variables, with cluster adjustments at the healthcare system level. Statistical analyses W d d i i We used descriptive statistics to characterize study out- comes, predictors, and covariates. To compare burnout and turnover intent levels before and during the pan- demic crisis period, we examined the prevalence of burn- out and turnover intent among VA primary care HCWs from the 2017 to 2020 administrations of the AES. For multivariable analyses, individual HCW survey responses for 2020 were linked to 2020 and 2019 health- care system-level (e.g., VA hospital-based medical center and its affiliated clinics) data. We used logistic regression, modeling burnout and turnover separately, and includ- ing employee engagement; perceptions of workload, leadership, and workgroup; COVID-19 burden; virtual care use; and prior year burnout as predictors. Mod- els also included demographics and healthcare system The number of primary care visits delivered virtually by patient location (including video, phone and supple- mentary remote [e.g., store-and-forward video, audio, or image messages, or specialty visits like tele-smoking ces- sation]) [22] and in-person from 03/15/20 to 09/15/20 by healthcare system were extracted from the CDW. A Apaydin et al. Historical trends in VA primary care burnout and turnover intent VA primary care HCWs reported burnout rates of 35–42% from 2017–2020 (Table 1), with rates declining from 42% in 2017 to 35–36% in 2018–2019, and then increasingly slightly to 37% during the early COVID-19 pandemic. These HCWs also reported one-year turno- ver intention rates of 31–34% from 2017–2020, with a decrease in turnover intent during the early pandemic (2020). Response rates to the AES among all employees during this period ranged from 60–69%. Multivariable model results Turnover intent rates in late 2020 among non-VA physicians and advanced practitioners [39] were All data from 2020 Abbreviations: DO Doctor of Osteopathy, DP depersonalization, EE emotional exhaustion, MD Doctor of Medicine, RN registered nurse, VA Veterans Health Administration Table 2 (continued) Characteristic n % Unknown 321 1.6 Turnover intent in next year No 13,798 69.4 Yes 6,096 30.6 Reasonable workload (agree or strongly agree) No 8,467 42.6 Yes 11,427 57.4 Supervisor listening (agree or strongly agree) No 5,043 25.4 Yes 14,851 74.7 Supervisor respect (agree or strongly agree) No 3,818 19.2 Yes 16,076 80.8 Supervisor trust (agree or strongly agree) No 5,789 29.1 Yes 14,105 70.9 Workgroup cooperation (agree or strongly agree) No 5,443 27.4 Yes 14,451 72.6 Workgroup competency (agree or strongly agree) No 4,485 22.5 Yes 15,409 77.5 Workgroup collaboration (agree or strongly agree) No 7,642 38.4 Yes 12,252 61.6 All data from 2020 Abbreviations: DO Doctor of Osteopathy, DP depersonalization, EE emotional exhaustion, MD Doctor of Medicine, RN registered nurse, VA Veterans Health Administration Table 2 (continued) Characteristic n % Unknown 321 1.6 Turnover intent in next year No 13,798 69.4 Yes 6,096 30.6 Reasonable workload (agree or strongly agree) No 8,467 42.6 Yes 11,427 57.4 Supervisor listening (agree or strongly agree) No 5,043 25.4 Yes 14,851 74.7 Supervisor respect (agree or strongly agree) No 3,818 19.2 Yes 16,076 80.8 Supervisor trust (agree or strongly agree) No 5,789 29.1 Yes 14,105 70.9 Workgroup cooperation (agree or strongly agree) No 5,443 27.4 Yes 14,451 72.6 Workgroup competency (agree or strongly agree) No 4,485 22.5 Yes 15,409 77.5 Workgroup collaboration (agree or strongly agree) No 7,642 38.4 Yes 12,252 61.6 Healthcare system-level COVID-19 tests and deaths, proportion of virtual care visits, and healthcare system complexity were not associated with either burnout or turnover intent. Average prior-year healthcare system- level burnout was significantly associated with 2020 individual-level burnout (OR 6.75; 95% CI 1.47–30.94) but not turnover intent. Prior year average healthcare system-level turnover intent was also not associated with current year individual-level turnover intent. p y Thirty-seven percent of VA primary care HCWs reported high burnout in 2020, and nearly 31% reported an intent to leave their job. Burnout rates among physi- cians and other advanced practitioners outside of VA ranged from 38 to 57% in late 2020 according to two national surveys, [39, 40] indicating that VA experienced less burnout during the early pandemic than commu- nity physicians. Multivariable model results In multivariable models (n = 16,191 [burnout model] and n = 16,333 [intent to leave model] in 139 healthcare systems), highly engaged employees were significantly less likely to report burnout (odds ratio [OR] 0.57; 95% confidence interval [CI] 0.52–0.63) or an intention to Page 6 of 12 Page 6 of 12 Page 6 of 12 Apaydin et al. BMC Health Services Research (2023) 23:1306 Healthcare system-level COVID-19 tests and deaths, proportion of virtual care visits, and healthcare system complexity were not associated with either burnout or turnover intent. Average prior-year healthcare system- level burnout was significantly associated with 2020 individual-level burnout (OR 6.75; 95% CI 1.47–30.94) but not turnover intent. Prior year average healthcare system-level turnover intent was also not associated with current year individual-level turnover intent. Discussion Trends in VA primary care burnout between 2017–2020 varied but hovered around 37%, with a high rate of 42% in 2017, suggesting little impact due to the early COVID- 19 pandemic. Thirty-seven percent of VA primary care HCWs reported high burnout and nearly one-third reported turnover intent in September 2020. The lack of large increases in rates of burnout and turnover intent during 2020 may indicate the presence of local supports (e.g., strong leadership or good crisis response) that offset HCW burnout, or persistent non-COVID drivers of both outcomes in VA primary care. Thirty-seven percent of VA primary care HCWs reported high burnout in 2020, and nearly 31% reported an intent to leave their job. Burnout rates among physi- cians and other advanced practitioners outside of VA ranged from 38 to 57% in late 2020 according to two national surveys, [39, 40] indicating that VA experienced less burnout during the early pandemic than commu- nity physicians. Characteristic 1 (most complex) 3 (least complex) data as well. Higher workload during the COVID-19 pandemic was found to be related to more burnout in a wide range of studies of HCWs conducted across the world [51]. Workloads pushing HCWs beyond their training or interfering with their personal lives [52] and a desire to decrease one’s workload, [7] were found to be particularly related to high burnout. Leader- ship, overall, [53] and in terms of communication and support [54], has also been previously linked to lower burnout among HCWs during the pandemic. The rela- tionships between good workgroups, or teams, and burnout during the pandemic have not been extensively studied in the literature, but there is some evidence that team identification [55] and support [56] were related to lower pandemic-era burnout in primary care and emergency medicine. Finally, it is important to note that only workload had a stronger relationship with burnout than engagement in our findings, suggesting that engagement may encompass aspects of the work- ing environment that are more than just the sum of workload, leadership, and workgroups. Perceptions of good leadership and workgroups were less related to decreased burnout, implying that these constructs were less protective than overall engagement among primary care HCWs during the early pandemic. Model formalizes these differences between burnout and engagement, and proposes that burnout is driven by job demands (e.g., workload), whereas engagement is driven by job resources (e.g., high quality supervisors or lead- ers, social support from colleagues, etc.) [49]. Cultivating these resources in “normal” times can increase engage- ment, and potentially protect against a large increase in burnout that results from increases in demands dur- ing crises. Engagement has been previously associated with positive organizational culture in a small sample of nurses [50] (as conceived by the six areas of worklife: workload, control, reward, community, fairness, and val- ues [26]) and with lower burnout across professions and countries [5, 26]. In our sample, high engagement was linked to lower burnout during the pandemic, suggest- ing that engagement may be a bulwark against drivers of burnout external to an organization. Our analyses also showed that healthcare system-level average burnout in 2019 was associated with individual-level burnout and turnover intent in 2020, suggesting the persistent and longitudinal relationship between an organizational cul- ture of burnout and these individual outcomes. Multivariable model results BMC Health Services Research (2023) 23:1306 Page 7 of 12 Table 3 Healthcare system characteristics (n = 139) All data from 2020, except 2019 burnout and turnover intent data Abbreviations: DP depersonalization, EE emotional exhaustion, M mean, PC primary care, SD standard deviation Characteristic M SD Range Healthcare system PC visits conducted virtually 29% 21% 0–76% 2020 healthcare system-level high burnout (EE or DP) 31% 3% 24–40% 2020 healthcare system-level turnover intent 31% 3% 22–40% 2019 healthcare system-level high burnout (EE or DP) 31% 3% 24–41% 2019 healthcare system-level turnover intent 35% 3% 26–44% Healthcare system COVID-19 deaths per 1000 unique patients 0.46 0.52 0.0–3.65 Healthcare system COVID-19 tests per 1000 unique patients 56.1 46.9 9.2–471.7 N % Healthcare system COVID-19 death rate terciles Lowest tercile (0–0.27 deaths per 1000 unique patients) 59 42.6 Middle tercile (0.27–0.69 deaths per 1000 unique patients) 58 41.7 Highest tercile (0.69–3.65 deaths per 1000 unique patients) 22 15.8 Healthcare system COVID-19 test quartiles Lowest quartile (9.2–38.7 tests per 1000 unique patients) 47 33.8 2nd quartile (38.9–47.6 tests per 1000 unique patients) 30 21.6 3rd quartile (47.8–61.4 tests per 1000 unique patients) 27 19.4 Highest quartile (65.2–471.7 tests per 1000 unique patients) 35 25.2 Healthcare system complexity 1 (most complex) 92 61.2 2 20 14.4 3 (least complex) 27 19.4 Table 3 Healthcare system characteristics (n = 139) Multivariable model results Turnover intent rates in late 2020 among non-VA physicians and advanced practitioners [39] were similar to the rates we found in among VA HCWs in our analysis. National data for nurses has not been published, but a state-level survey from New Jersey indicated that 37% of nurses intended to leave their jobs in late 2020, [41] which is higher in our sample of VA primary care HCWs (31%). In another state-wide analysis in early 2022, intent to leave was even higher among nurses in Michigan at 39% [42]. These comparator studies contain non-primary care specialties, so this analysis may under- estimate the gap in burnout between VA and non-VA primary care HCWs, as burnout in both community and VA primary care is generally higher than in most other specialties [43, 44]. VA HCWs in primary care may have had even lower rates of burnout and turnover intent than equivalent HCWs in the community than what is sug- gested here. Abbreviations: DO Doctor of Osteopathy, DP depersonalization, EE emotional exhaustion, MD Doctor of Medicine, RN registered nurse, VA Veterans Health Administration leave their jobs (OR 0.62; 95% CI 0.57–0.68; Table 4), adjusting for professional role, gender, age, race, eth- nicity, VA tenure, and supervisor status. y p Reasonable workload was also associated with lower burnout (OR 0.26, 95% CI 0.24–0.28) and lower turn- over intent (OR 0.54, 95% CI 0.50–0.58). Supervi- sor trust (burnout model: OR 0.71, 95% CI 0.62–0.82; turnover intent model: OR 0.69, 95% CI 0.62–0.78), and workgroup cooperation (burnout model: OR 0.78, 95% CI 0.72–0.85; turnover intent model: OR 0.77, 95% CI 0.69–0.85), competency (burnout model: OR 0.82, 95% CI 0.74–0.91; turnover intent model: OR 0.78, 95% CI 0.69–0.87), and collaboration (burnout model: OR 0.74, 95% CI 0.67–0.80; turnover intent model: OR 0.77, 95% CI 0.70–0.85) were also associated with lower likeli- hoods of both outcomes. Supervisor listening (turnover intent model: OR 0.83, 95% CI 0.73–0.95), and respect (turnover intent model: OR 0.85, 95% CI 0.74–0.98) were related to lower odds of turnover intent but not burnout. We found that highly engaged HCWs were less likely to report either high burnout or their intent to leave practice in 2020. Engagement, commonly conceived as either the opposite of burnout [45] or an independent positive construct, [46] is a cultivated, rather than emer- gent, phenomenon [47, 48]. The Job Demands-Resources Apaydin et al. All data from 2020, except 2019 burnout and turnover intent data Characteristic Individual perceptions of reasonable workload, high quality leadership, and good workgroups were linked with lower rates of burnout and turnover intent in our Apaydin et al. BMC Health Services Research (2023) 23:1306 Page 8 of 12 Table 4 Odds of burnout and turnover intent by healthcare system and individual characteristics EE or DP burnout Turnover intent n = 16,191 in 139 healthcare systems n = 16,333 in 139 healthcare systems Characteristic OR 95% CI OR 95% CI High employee engagement No Ref Ref Yes 0.57 0.52–0.63 0.62* 0.57–0.68 Reasonable workload No Ref Ref Yes 0.26* 0.24–0.28 0.54* 0.50–0.58 Supervisor listening No Ref Ref Yes 0.91 0.78–1.06 0.83* 0.73–0.95 Supervisor respect No Ref Ref Yes 1.04 0.89–1.21 0.85* 0.74–0.98 Supervisor trust No Ref Ref Yes 0.71* 0.62–0.82 0.69* 0.62–0.78 Workgroup cooperation No Ref Ref Yes 0.78* 0.72–0.85 0.77* 0.69–0.85 Workgroup competency No Ref Ref Yes 0.82* 0.74–0.91 0.78* 0.69–0.87 Workgroup collaboration No Ref Ref Yes 0.74* 0.67–0.80 0.77* 0.70–0.85 2019 healthcare system-level high burnout (EE or DP) 6.75* 1.47–30.94 – – 2019 healthcare system-level high turnover intent – – 2.93 0.59–14.50 Healthcare system COVID-19 death terciles Lowest third (0–0.27 deaths per 1000 unique patients) Ref Ref Middle third (0.27–0.69 deaths per 1000 unique patients) 0.99 0.91–1.07 0.92 0.83–1.02 Highest third (0.69–3.65 deaths per 1000 unique patients) 0.91 0.81–1.03 0.84* 0.72–0.98 Healthcare system COVID-19 test quartiles Lowest quartile (9.2–38.7 tests per 1000 unique patients) Ref Ref 2nd quartile (38.9–47.6 tests per 1000 unique patients) 0.99 0.90–1.10 1.03 0.90–1.18 3rd quartile (47.8–61.4 tests per 1000 unique patients) 0.92 0.82–1.02 0.97 0.84–1.13 Highest quartile (65.2–471.7 tests per 1000 unique patients) 0.91 0.79–1.05 0.98 0.86–1.13 Healthcare system proportion of virtual PC visits per all visits 1.00 0.83–1.22 0.86 0.70–1.05 Healthcare system complexity 1 Ref Ref 2 0.99 0.85–1.14 0.92 0.76–1.11 3 1.01 0.89–1.16 0.97 0.83–1.12 Professional role Provider (MD/DO, NP, PA) Ref Ref RN 0.72* 0.64–0.79 1.27* 1.14–1.41 Clinical associate 0.71* 0.63–0.80 1.12 1.00–1.25 Administrative associate 0.87* 0.76–0.99 2.23* 1.93–2.57 Table 4 Odds of burnout and turnover intent by healthcare system and individual characteristics Clinical associate Apaydin et al. Characteristic BMC Health Services Research (2023) 23:1306 Page 9 of 12 Table 4 (continued) Table 4 (continued) EE or DP burnout Turnover intent n = 16,191 in 139 healthcare systems n = 16,333 in 139 healthcare systems Characteristic OR 95% CI OR 95% CI Gender Male Ref Ref Female 1.00 0.92–1.08 0.75* 0.69–0.81 Race White Ref Ref Black or African American 0.80* 0.73–0.88 1.11* 1.01–1.22 Asian 0.87* 0.76–0.99 0.80* 0.69–0.93 American Indian or Alaskan Native 0.87 0.68–1.11 0.90 0.70–1.15 Native Hawaiian or other Pacific Island 1.13 0.88–1.46 1.44* 1.12–1.86 Ethnicity Non-Hispanic Ref Ref Hispanic 1.10 0.94–1.29 1.04 0.91–1.18 Age 29 and under Ref Ref 30–49 0.72* 0.60–0.86 0.80* 0.67–0.96 50 + 0.52* 0.43–0.63 0.74* 0.61–0.89 VA tenure Less than 2 years Ref Ref Between 2 and 10 years 1.60* 1.46–1.77 1.29* 1.16–1.44 Between 10 and 20 years 1.77* 1.57–1.99 1.30* 1.14–1.47 More than 20 years 1.81* 1.52–2.15 1.68* 1.44–1.97 Supervisor status No Ref Yes 1.18* 1.08–1.29 1.12* 1.02–1.22 * = p < 0.05; All data from 2020, except 2019 burnout and turnover intent data Abbreviations: CI confidence interval, DO Doctor of Osteopathy, DP depersonalization, EE emotional exhaustion, MD Doctor of Medicine, OR odds ratios, Ref reference category, RN registered nurse, VA Veterans Health Administration terval, DO Doctor of Osteopathy, DP depersonalization, EE emotional exhaustion, MD Doctor of Medicine, OR odds ratios, Ref reference VA Veterans Health Administration Our results also show that healthcare system-level COVID-19 burden, shift to virtual care use, and com- plexity were not associated with either high burnout or turnover intent. While high burnout has been reported among HCWs across specialties and geographies during the pandemic, [5–9] the relationships between primary care burnout or turnover intent and COVID-19 burden has not been widely studied. We found only one study conducted in a Belgian intensive care unit with nurses that found little association between burnout and the proportions of COVID-19 patients or deaths (to total patients or deaths) [57]. Similarly, the effect of virtual care on healthcare outcomes during the pandemic has been widely studied, [58] but its relationship with burn- out or turnover intent has only been the subject of a few analyses. These previous studies suggest that greater telehealth self-efficacy (i.e., comfort with the use of tele- health or virtual care) [10] or fewer “difficulties with new technologies” [23] during the pandemic may be associ- ated with lower burnout or turnover intent. Availability of data and materials The data that support the findings of this study are available from the VA Office of Primary Care, but this data is not publicly available outside of the VA. Data are however available for VA employees from the authors upon reason- able request and with permission of the VA Office of Primary Care. Interested parties should contact the corresponding author, Eric Apaydin, PhD, MPP, MS, at eric.apaydin@va.gov with their request. Funding This project was supported in part by the Veterans Health Administration Office of Primary Care (Project #XVA 65–018). Dr. Apaydin was supported by the VA Office of Academic Affiliations through the Advanced Fellowship in Health Services Research & Development (HSR&D) and through a VA HSR&D Career Development Award (IK2HX003534). Dr. Yano was supported by a VA HSR&D Senior Research Career Scientist Award (IK6HX002836). Characteristic Our results suggest that the COVID-19 crisis and associated virtual care use had no relationship with burnout or turnover intent. It is possible that neither COVID-19 nor virtual care use were as impactful as longer-term organizational contextual factors associated with burnout and turnover, such as staffing shortages and instability [29, 59]. fi Our study had a few notable limitations, including the inability to identify individuals in the anonymous AES data and match those data with prior-year individ- ual-level AES data. In addition, our COVID-19 deaths and tests and virtual care data were healthcare system averages across the first six months of the pandemic (03/15/20 to 09/15/20) and may not have adequately captured variations due to local virus surges. Finally, our measures of COVID burden may not have reflected the true impact of the pandemic on primary care, Apaydin et al. BMC Health Services Research (2023) 23:1306 Page 10 of 12 Page 10 of 12 including the uncertainties associated with staff reas- signments, patient screening procedures, care manage- ment for vulnerable patients, and the sudden shift to virtual modalities. 1. Ehring T, Razik S, Emmelkamp PM. Prevalence and predictors of post- traumatic stress disorder, anxiety, depression, and burnout in Pakistani earthquake recovery workers. Psychiatry Res. 2011;185(1–2):161–6. https://doi.org/10.1016/j.psychres.2009.10.018. Disclosure h The views expressed are those of the authors and do not represent the views of the US Department of Veterans Affairs or the United States Government. The views expressed are those of the authors and do not represent the views of the US Department of Veterans Affairs or the United States Government. Author details 1 1 Center for the Study of Healthcare Innovation, Implementation & Policy, VA Greater Los Angeles Healthcare System, 11301 Wilshire Blvd. (151), Los Angeles, CA 90073, USA. 2 RAND Corporation, Santa Monica, CA, USA. 3 National Center for Organization Development, Veterans Health Administration, Cincin- nati, OH, USA. 4 Department of Health Law, Policy & Management, School of Public Health, Boston University, Boston, MA, USA. 5 Department of Health Policy and Management, Fielding School of Public Health, University of Cali- fornia, Los Angeles, Los Angeles, CA, USA. 6 Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA, USA. 7 Seattle-Denver Center of Innovation, VA Puget Sound Health Care System, Seattle, WA, USA. 8 Division of General Internal Medicine, Department of Medi- cine, University of Washington School of Medicine, University of Washington, Seattle, WA, USA. 9 Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA. Received: 15 June 2023 Accepted: 2 November 2023 Received: 15 June 2023 Accepted: 2 November 2023 Acknowledgements g The authors wish to thank the US Department of Veterans Affairs, the Veterans Health Administration, and the Organizational Assessment Sub-Committee of the Human Resources Committee for access to its survey data. We especially thank the Veterans Health Administration National Center for Organization Development (VHA NCOD) for their contribution to collecting, cleaning, and managing the VHA All Employee Survey data used in our study. Authors’ contributions EAA, DER, and SES conceived and designed the study, and interpreted the data. MRM, RG, and CKY conducted all data analyses. DCM provided access to and expertise on survey data. EAA drafted the manuscript. All authors (includ- ing EMY, PGS, and KMN) reviewed and substantially revised the manuscript. The authors read and approved the final manuscript. More research is needed to explore whether interven- tions to improve job resources and employee engage- ment can protect HCW from the negative impacts of crises affecting healthcare delivery and other system- level changes that may increase stress among HCW. There is some evidence that evidence-based quality improvement, an implementation strategy for clini- cal guidelines and care models, may have reduced VA PCP burnout during patient-centered medical home implementation [60]. This reduction in burnout may have been related to engaging primary care HCWs in participatory decision-making and the empowerment of frontline HCWs to address workplace quality issues. In another study of PCPs and staff outside the VA, par- ticipants who did not help design burnout reduction interventions actually experienced an increase in burn- out [61]. Edwards and colleagues examined 715 small- to-medium-size non-VA primary care practices, and found that those with zero burnout had better work- ing environments and used more quality improvement strategies, compared to high burnout practices [62]. These results suggest that interventions that empower and engage providers and staff may be key to improving organizational climate and reducing burnout. g The authors wish to thank the US Department of Veterans Affairs, the Veterans Health Administration, and the Organizational Assessment Sub-Committee of the Human Resources Committee for access to its survey data. We especially thank the Veterans Health Administration National Center for Organization Development (VHA NCOD) for their contribution to collecting, cleaning, and managing the VHA All Employee Survey data used in our study. Consent for publication Not applicable. Consent for publication Not applicable. Ethics approval and consent to participate This analysis was conducted as a non-research evaluation approved by the VA Office of Primary Care. Non-research evaluations designed to evaluate and improve public service programs are exempt from ethics approval and consent requirements under United States federal regulation 45 CFR 46.104(d) (5). All methods were carried out in accordance with relevant guidelines and regulations. In conclusion, burnout and turnover intent among VA primary care HCWs was high, but lower than pre- pandemic trends during the first six months of the COVID-19 pandemic, and COVID-specific organi- zational contextual factors were not associated with high burnout or turnover intent. We found that highly engaged providers and staff were less likely to be burned out or to intend to leave their jobs. Those with reasonable workloads, and positive perceptions of their leadership and workgroups were also less likely to be burned out. Interventions to improve organizational climate, and increase job resources and engagement, should be developed and evaluated, with the intent of buffering the impacts of external and organizational contextual factors on burnout. Improving primary care HCW working conditions now may be the key to pro- tecting against high burnout in the face of future chal- lenges, whether they be pandemics or other healthcare crises. References Page 11 of 12 Page 11 of 12 Apaydin et al. BMC Health Services Research (2023) 23:1306 Apaydin et al. BMC Health Services Research (2023) 23:1306 2. Fujitani K, Carroll M, Yanagisawa R, Katz C. Burnout and Psychiatric Distress in Local Caregivers Two Years After the 2011 Great East Japan Earthquake and Fukushima Nuclear Radiation Disaster. Community Ment Health J. 2016;52(1):39–45. https://doi.org/10.1007/s10597-015-9924-y. 21. Dichter JR, Devereaux AV, Sprung CL, et al. Mass Critical Care Surge Response During COVID-19: Implementation of Contingency Strategies - A Preliminary Report of Findings From the Task Force for Mass Critical Care. Chest. 2021. https://doi.org/10.1016/j.chest.2021.08.072. y ealth J. 2016;52(1):39–45. https://doi.org/10.1007/s10597-015-9924 3. Powell TM, Yuma PJ, Scott J, et al. In the aftermath: The effects of hur- ricanes Harvey and Maria on the well-being of health-care and social service providers. Traumatology. 2020;26(3):298–307. https://doi.org/10. 1037/trm0000228. 22. Ferguson JM, Jacobs J, Yefimova M, Greene L, Heyworth L, Zulman DM. Virtual care expansion in the Veterans Health Administration during the COVID-19 pandemic: clinical services and patient characteristics associ- ated with utilization. J Am Med Inform Assoc. 2021;28(3):453–62. https:// doi.org/10.1093/jamia/ocaa284. 4. Maslach C, Jackson SE, Leiter MP, Schaufeli WB, Schwab RL. Maslach Burnout Inventory Manual: Fourth Edition. Menlo Park: Mind Garden, Inc.; 2018. 4. Maslach C, Jackson SE, Leiter MP, Schaufeli WB, Schwab RL. Maslach Burnout Inventory Manual: Fourth Edition. Menlo Park: Mind Garden, Inc.; 2018. 23. Kelly EL, Cunningham A, Sifri R, Pando O, Smith K, Arenson C. Burnout and Commitment to Primary Care: Lessons From the Early Impacts of COVID- 19 on the Workplace Stress of Primary Care Practice Teams. Ann Fam Med Jan-Feb. 2022;20(1):57–62. https://doi.org/10.1370/afm.2775. 5. Apaydin EA, Rose DE, Yano EM, et al. Burnout Among Primary Care Healthcare Workers During the COVID-19 Pandemic. J Occup Environ Med. 2021; https://doi.org/10.1097/JOM.0000000000002263. 5. Apaydin EA, Rose DE, Yano EM, et al. Burnout Among Primary Care Healthcare Workers During the COVID-19 Pandemic. J Occup Environ Med. 2021; https://doi.org/10.1097/JOM.0000000000002263. Jan-Feb. 2022;20(1):57–62. https://doi.org/10.1370/afm.2775 24. Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev Psychol. 2001;52:397–422. https://doi.org/10.1146/annurev.psych.52.1.397. 6. Azoulay E, De Waele J, Ferrer R, et al. Symptoms of burnout in intensive care unit specialists facing the COVID-19 outbreak. Ann Intensive Care. 2020;10(1):110. https://doi.org/10.1186/s13613-020-00722-3. 25. Cole MS, Walter F, Bedeian AG, O’Boyle EH. Job Burnout and Employee Engagement. J Manag. 2011;38(5):1550–81. https://doi.org/10.1177/ 0149206311415252. 7. Matsuo T, Kobayashi D, Taki F, et al. References World Psychiatry. 2016;15(2):103–11. https://doi.org/10.1002/wps.20311. 32. Graham LA. A Primer to VA Data: Accessing, Requesting, and Analyzing. VA Health Economics Resource Center. Published 12 Oct 2020. https:// med.stanford.edu/content/dam/sm/s-spire/documents/VA-Data-Primer- 2020-_Laura-Graham_10.12.20.pdf. Accessed 12 Nov 2023. 14. Dyrbye LN, Shanafelt TD, Sinsky CA, et al. Burnout Among Health Care Professionals: A Call to Explore and Address This Underrecognized Threat to Safe, High-Quality Care. NAM Perspect. 2017;7(7). https://doi.org/10. 31478/201707b. 33. VA Informatics and Computing Infrastructure. VA COVID-19 Shared Data Resource. Published 2020. https://www.hsrd.research.va.gov/for_resea rchers/cyber_seminars/archives/3810-notes.pdf. Accessed 12 Nov 2023. 15. Abraham CM, Zheng K, Poghosyan L. Predictors and Outcomes of Burnout Among Primary Care Providers in the United States: A Systematic Review. Med Care Res Rev. 2020;77(5):387–401. https://doi.org/10.1177/ 1077558719888427. 34. West CP, Dyrbye LN, Satele DV, Sloan JA, Shanafelt TD. Concurrent validity of single-item measures of emotional exhaustion and depersonalization in burnout assessment. J Gen Intern Med. 2012;27(11):1445–52. https:// doi.org/10.1007/s11606-012-2015-7. 16. Willard-Grace R, Knox M, Huang B, Hammer H, Kivlahan C, Grum- bach K. Burnout and Health Care Workforce Turnover. Ann Fam Med. 2019;17(1):36–41. https://doi.org/10.1370/afm.2338. 35. West CP, Dyrbye LN, Sloan JA, Shanafelt TD. Single item measures of emo- tional exhaustion and depersonalization are useful for assessing burnout in medical professionals. J Gen Intern Med. 2009;24(12):1318–21. https:// doi.org/10.1007/s11606-009-1129-z. 17. Basu S, Berkowitz SA, Phillips RL, Bitton A, Landon BE, Phillips RS. Associa- tion of Primary Care Physician Supply With Population Mortality in the United States, 2005–2015. JAMA Intern Med. 2019;179(4):506–14. https:// doi.org/10.1001/jamainternmed.2018.7624. 36. VA National Center for Organization Development. All Employee Survey (AES) 2019 - 2020. Published 21 Mar 2023. https://www.datahub.va.gov/ dataset/All-Employee-Survey-AES-2022-2023/mqnc-hghc. Accessed 12 Nov 2023. 18. MacLean L, Hassmiller S, Shaffer F, Rohrbaugh K, Collier T, Fairman J. Scale, causes, and implications of the primary care nursing shortage. Annu Rev Public Health. 2014;35:443–57. https://doi.org/10.1146/annurev-publh ealth-032013-182508. 37. Veterans Health Administration. Productivity and Staffing Guidance for Specialty Provider Group Practice. Directive 1065. Washington, DC: Veter- ans Health Administration; 2020. 19. Friedman JL, Neutze D. The Financial Cost of Medical Assistant Turno- ver in an Academic Family Medicine Center. J Am Board Fam Med. 2020;33(3):426–30. https://doi.org/10.3122/jabfm.2020.03.190119. 38. National Academies of Sciences E, Medicine. Facilities Staffing Require- ments for the Veterans Health Administration–Resource Planning and Methodology for the Future. Washington: The National Academies Press; 2020; 136. 20. Noseworthy J, Madara J, Cosgrove D, et al. Physician Burnout Is A Public Health Crisis: A Message To Our Fellow Health Care CEOs. Health Affairs Blog. Published 28 Mar 2017. https://www.healthaffairs.org/do/10.1377/ hblog20170328.059397/full/. References Prevalence of Health Care Worker Burnout During the Coronavirus Disease 2019 (COVID-19) Pandemic in Japan. JAMA Netw Open. 2020;3(8):e2017271. https://doi.org/10.1001/ jamanetworkopen.2020.17271. 26. Leiter MP, Maslach C. Areas of Worklife: A Structured Approach to Organizational Predictors of Job Burnout. In: Perrewe PL, Ganster DC, eds Emotional and Physiological Processes and Positive Intervention Strate- gies: Research in Occupational Stress and Well Being. Bingley: Emerald Group Publishing Limited; 2003:91–134. 8. Barello S, Palamenghi L, Graffigna G. Burnout and somatic symptoms among frontline healthcare professionals at the peak of the Italian COVID-19 pandemic. Psychiatry Res. 2020;290: 113129. https://doi.org/10 1016/j.psychres.2020.113129. 27. Shanafelt TD, Gorringe G, Menaker R, et al. Impact of organizational leadership on physician burnout and satisfaction. Mayo Clin Proc. 2015;90(4):432–40. https://doi.org/10.1016/j.mayocp.2015.01.012. 9. Jalili M, Niroomand M, Hadavand F, Zeinali K, Fotouhi A. Burnout among healthcare professionals during COVID-19 pandemic: a cross-sectional study. Int Arch Occup Environ Health. 2021;94(6):1345–52. https://doi. org/10.1007/s00420-021-01695-x. 28. Apaydin EA, Rose DE, Yano EM, Shekelle PG, Stockdale SE, Mohr DC. Gender Differences in the Relationship Between Workplace Civility and Burnout Among VA Primary Care Providers. J Gen Intern Med. 2021. https://doi.org/10.1007/s11606-021-06818-1. 10. Sklar M, Ehrhart MG, Aarons GA. COVID-related work changes, burnout, and turnover intentions in mental health providers: A moderated media- tion analysis. Psychiatr Rehabil J. 2021;44(3):219–28. https://doi.org/10. 1037/prj0000480. 29. Helfrich CD, Dolan ED, Simonetti J, et al. Elements of team-based care in a patient-centered medical home are associated with lower burnout among VA primary care employees. J Gen Intern Med. 2014;29 Suppl 2(Suppl 2):S659–66. https://doi.org/10.1007/s11606-013-2702-z. 11. Falatah R. The Impact of the Coronavirus Disease (COVID-19) Pan- demic on Nurses’ Turnover Intention: An Integrative Review. Nurs Rep. 2021;11(4):787–810. https://doi.org/10.3390/nursrep11040075. 30. Kim LY, Rose DE, Soban LM, et al. Primary Care Tasks Associated with Provider Burnout: Findings from a Veterans Health Administration Survey. J Gen Intern Med. 2018;33(1):50–6. https://doi.org/10.1007/ s11606-017-4188-6. 12. Chen HM, Liu CC, Yang SY, Wang YR, Hsieh PL. Factors Related to Care Competence, Workplace Stress, and Intention to Stay among Novice Nurses during the Coronavirus Disease (COVID-19) Pandemic. Int J Environ Res Public Health. 2021;18(4). https://doi.org/10.3390/ijerph1804 2122. 31. Otasuke K, Draime J, Moore SC, et al. Organization development in the Department of Veterans Affairs. In: Miller T, ed. The Praeger handbook of Vet- erans Health: History, challenges, issues and developments, Volume IV: Future directions in Veterans healthcare. Santa Barbara: Praeger; 2012:21–76. 13. Maslach C, Leiter MP. Understanding the burnout experience: recent research and its implications for psychiatry. 39. Linzer M, Jin JO, Shah P, et al. Trends in Clinician Burnout With Associated Mitigating and Aggravating Factors During the COVID-19 Pandemic. References Accessed 12 Nov 2023. 39. Linzer M, Jin JO, Shah P, et al. Trends in Clinician Burnout With Associated Mitigating and Aggravating Factors During the COVID-19 Pandemic. Page 12 of 12 Page 12 of 12 Apaydin et al. BMC Health Services Research (2023) 23:1306 JAMA Health Forum. 2022;3(11):e224163. https://doi.org/10.1001/jamah ealthforum.2022.4163. JAMA Health Forum. 2022;3(11):e224163. https://doi.org/10.1001/jamah ealthforum.2022.4163. 59. Helfrich CD, Simonetti JA, Clinton WL, et al. The Association of Team- Specific Workload and Staffing with Odds of Burnout Among VA Primary Care Team Members. J Gen Intern Med. 2017;32(7):760–6. https://doi.org/ 10.1007/s11606-017-4011-4. 40. Shanafelt TD, West CP, Dyrbye LN, et al. Changes in Burnout and Satisfac- tion With Work-Life Integration in Physicians During the First 2 Years of the COVID-19 Pandemic. Mayo Clin Proc. 2022;97(12):2248–58. https:// doi.org/10.1016/j.mayocp.2022.09.002. 60. Meredith LS, Batorsky B, Cefalu M, et al. Long-term impact of evidence- based quality improvement for facilitating medical home implemen- tation on primary care health professional morale. BMC Fam Pract. 2018;19(1):149. https://doi.org/10.1186/s12875-018-0824-4. 41. de Cordova PB, Johansen ML, Grafova IB, Crincoli S, Prado J, Pogor- zelska-Maziarz M. Burnout and intent to leave during COVID-19: A cross-sectional study of New Jersey hospital nurses. J Nurs Manag. 2022;30(6):1913–21. https://doi.org/10.1111/jonm.13647. 61. Hung DY, Harrison MI, Truong Q, Du X. Experiences of primary care physi- cians and staff following lean workflow redesign. BMC Health Serv Res. 2018;18(1):274. https://doi.org/10.1186/s12913-018-3062-5. 42. Medvec BR, Marriott DJ, Khadr L, et al. Patterns and Correlates of Nurse Departures From the Health Care Workforce: Results From a Statewide Survey. Med Care. 2023;61(5):321–7. https://doi.org/10.1097/MLR.00000 00000001837. 62. Edwards ST, Marino M, Solberg LI, et al. Cultural And Structural Fea- tures Of Zero-Burnout Primary Care Practices. Health Aff (Millwood). 2021;40(6):928–36. https://doi.org/10.1377/hlthaff.2020.02391. 43. Shanafelt TD, Hasan O, Dyrbye LN, et al. Changes in Burnout and Satisfac- tion With Work-Life Balance in Physicians and the General US Working Population Between 2011 and 2014. Mayo Clin Proc. 2015;90(12):1600– 13. https://doi.org/10.1016/j.mayocp.2015.08.023. Publisher’s Note S i N i Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. y 44. Rinne ST, Mohr DC, Swamy L, Blok AC, Wong ES, Charns MP. National Burnout Trends Among Physicians Working in the Department of Veter- ans Affairs. J Gen Intern Med. 2020;35(5):1382–8. https://doi.org/10.1007/ s11606-019-05582-7. 45. Maslach C, Leiter MP. Early predictors of job burnout and engagement. J Appl Psychol. 2008;93(3):498–512. https://doi.org/10.1037/0021-9010. 93.3.498. 46. Schaufeli WB, Salanova M, González-romá V, Bakker AB. J Happiness Stud. 2002;3(1):71–92. https://doi.org/10.1023/a:1015630930326. 47. Maslach C. Engagement research: Some thoughts from a burnout per- spective. Eur J Work Organ Psy. 2011;20(1):47–52. https://doi.org/10.1080/ 1359432X.2010.537034. 48. Bakker AB, Albrecht SL, Leiter MP. Key questions regarding work engage- ment. Eur J Work Organ Psy. 2011;20(1):4–28. https://doi.org/10.1080/ 1359432X.2010.485352. 49. Bakker AB, Demerouti E, Sanz-Vergel AI. Burnout and Work Engagement: The JD–R Approach. Annu Rev Organ Psych Organ Behav. 2014;1(1):389– 411. https://doi.org/10.1146/annurev-orgpsych-031413-091235.l 50. Bamford M, Wong CA, Laschinger H. The influence of authentic leader- ship and areas of worklife on work engagement of registered nurses. J Nurs Manag. 2013;21(3):529–40. https://doi.org/10.1111/j.1365-2834. 2012.01399.x. 51. Sharifi M, Asadi-Pooya AA, Mousavi-Roknabadi RS. Burnout among Healthcare Providers of COVID-19; a Systematic Review of Epidemiology and Recommendations. Arch Acad Emerg Med. 2021;9(1): e7. https://doi. org/10.22037/aaem.v9i1.1004. 52. Morgantini LA, Naha U, Wang H, et al. Factors contributing to healthcare professional burnout during the COVID-19 pandemic: A rapid turnaround global survey. PLoS ONE. 2020;15(9): e0238217. https://doi.org/10.1371/ journal.pone.0238217. 53. Ma Y, Faraz NA, Ahmed F, et al. Curbing nurses’ burnout during COVID-19: The roles of servant leadership and psychological safety. J Nurs Manag. 2021;29(8):2383–91. https://doi.org/10.1111/jonm.13414. 54. Sangal RB, Bray A, Reid E, et al. Leadership communication, stress, and burnout among frontline emergency department staff amid the COVID- 19 pandemic: A mixed methods approach. Healthc (Amst). 2021;9(4): 100577. https://doi.org/10.1016/j.hjdsi.2021.100577. 55. Sangal RB, Wrzesniewski A, DiBenigno J, et al. Work team identification associated with less stress and burnout among front-line emergency department staff amid the COVID-19 pandemic. BMJ Leader. 2020. https://doi.org/10.1136/leader-2020-000331. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: Publisher’s Note S i N i Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 56. Cunningham AT, Felter J, Smith KR, et al. Burnout and Commitment After 18 Months of the COVID-19 Pandemic: A Follow-Up Qualitative Study with Primary Care Teams. J Am Board Fam Med. 2023. https://doi.org/10. 3122/jabfm.2022.220226R1. 57. Bruyneel A, Smith P, Tack J, Pirson M. Prevalence of burnout risk and fac- tors associated with burnout risk among ICU nurses during the COVID-19 outbreak in French speaking Belgium. Intensive Crit Care Nurs. 2021;65: 103059. https://doi.org/10.1016/j.iccn.2021.103059. 58. Monaghesh E, Hajizadeh A. The role of telehealth during COVID-19 out- break: a systematic review based on current evidence. BMC Public Health. 2020;20(1):1193. https://doi.org/10.1186/s12889-020-09301-4.
https://openalex.org/W2498319347	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0159475&type=printable	English	null	Compatibility of Injectable Anticoagulant Agents in Ethanol; In Vitro Antibiofilm Activity and Impact on Polyurethane Catheters of Enoxaparin 400 U/mL in 40% v/v Ethanol	PloS one	2,016	cc-by	11,014	RESEARCH ARTICLE OPEN ACCESS ☯These authors contributed equally to this work. * bsouweine@chu-clermontferrand.fr Citation: Balestrino D, Quintana M, Charbonnel N, Forestier C, Lartigue C, Souweine B (2016) Compatibility of Injectable Anticoagulant Agents in Ethanol; In Vitro Antibiofilm Activity and Impact on Polyurethane Catheters of Enoxaparin 400 U/mL in 40% v/v Ethanol. PLoS ONE 11(7): e0159475. doi:10.1371/journal.pone.0159475 Background and Objectives Editor: Partha Mukhopadhyay, National Institutes of Health, UNITED STATES Received: May 24, 2016 Accepted: June 8, 2016 Published: July 21, 2016 Editor: Partha Mukhopadhyay, National Institutes of Health, UNITED STATES Received: May 24, 2016 Accepted: June 8, 2016 Published: July 21, 2016 Interdialytic lock solutions should maintain catheter patency and prevent catheter infections. We aimed to determine in which conditions injectable anticoagulant agents (IAAs) com- bined with ethanol are compatible and to assess the antibiofilm activity of the selected com- bination and its effects on dialysis catheters (DC). Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was supported by a financial support from the Auvergne FEDER (Fonds Européen de Développement Économique et Régional) and a grant from Hemotech laboratories. DualCath1 used in this study were specifically designed and provided by Hemotech, Ramonville, France. Compatibility of Injectable Anticoagulant Agents in Ethanol; In Vitro Antibiofilm Activity and Impact on Polyurethane Catheters of Enoxaparin 400 U/mL in 40% v/v Ethanol a11111 Damien Balestrino1,2☯, Mercédès Quintana3,4☯, Nicolas Charbonnel2, Christiane Forestier1,2, Claire Lartigue4, Bertrand Souweine1,5* 1 UMR CNRS 6023, Laboratoire Microorganismes: Génome et Environnement, Clermont Université, Université d'Auvergne, Clermont Ferrand, 63000, France, 2 Université d’Auvergne, Faculté de Pharmacie, Laboratoire de Bactériologie, F- 63001, Clermont-Ferrand, France, 3 Université d’Auvergne, Faculté de Pharmacie, Laboratoire de Chimie analytique et spectrométrie de masse, F- 63001, Clermont-Ferrand, France, 4 Inserm, UMR 990, IMTV, F-63005, Clermont-Ferrand, France, 5 Réanimation Médicale, Hôpital Gabriel Montpied, CHU-Clermont-Ferrand, 63000, France Methods Copyright: © 2016 Balestrino et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The solubility and compatibility of unfractionated heparin (UFH), low molecular weight hepa- rins (LMWHs), heparinoids and fondaparinux (50 to 2,500 U/mL) in 30 to 70% ethanol were determined by visual observation. The stability of enoxaparin in ethanol and the ethanol content were assessed by high performance liquid chromatography (HPLC) and titrimetric control, respectively. The bactericidal effect was determined on 24h-old biofilms embedded in silicone-DC. The integrity of polyurethane-DC immersed in anticoagulant-ethanol was assessed by gas chromatography-mass spectrometry (GC-MS) and compared with previ- ously published results. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Introduction Dialysis catheters (DCs) are widely used for the provision of dialysis in end stage renal disease patients with a non-functioning arteriovenous fistula or synthetic graft [1–2]. Infection and thrombosis are two major complications associated with DC use [2–3]. DC infections represent one of the most common causes of death in chronic dialysis patients [4], with mortality rates estimated between 12 and 25% [5]. In addition, thrombosis-induced DC dysfunction is a fre- quent reason for DC removal or replacement [6]. At the end of each dialysis session, unfractio- nated heparin (UFH) is currently instilled into DC lumens and left in between the dialysis sessions to maintain catheter patency. UFH is the injectable anticoagulant agent (IAA) gener- ally used for this purpose but other IAAs such as low molecular weight heparins (LMWHs), pentasaccharide Factor Xa inhibitor (fondaparinux) and heparinoids could also be envisaged as lock solutions. In particular, LMWHs seem to generate less adverse reactions in the long term [7]. In vitro studies assessing the antibiofilm activity of UFH have yielded conflicting results [8–9]. UFH at low concentrations (< 1,000 U/mL) have been shown to stimulate Staph- ylococcus aureus biofilm formation [8,10]. Higher inhibitory effects are obtained when chelat- ing agents are used in combination with anti-microbial agents such as antibiotics [11]. Dialysis catheters (DCs) are widely used for the provision of dialysis in end stage renal disease patients with a non-functioning arteriovenous fistula or synthetic graft [1–2]. Infection and thrombosis are two major complications associated with DC use [2–3]. DC infections represent one of the most common causes of death in chronic dialysis patients [4], with mortality rates estimated between 12 and 25% [5]. In addition, thrombosis-induced DC dysfunction is a fre- quent reason for DC removal or replacement [6]. At the end of each dialysis session, unfractio- nated heparin (UFH) is currently instilled into DC lumens and left in between the dialysis sessions to maintain catheter patency. UFH is the injectable anticoagulant agent (IAA) gener- ally used for this purpose but other IAAs such as low molecular weight heparins (LMWHs), pentasaccharide Factor Xa inhibitor (fondaparinux) and heparinoids could also be envisaged as lock solutions. In particular, LMWHs seem to generate less adverse reactions in the long term [7]. In vitro studies assessing the antibiofilm activity of UFH have yielded conflicting results [8–9]. UFH at low concentrations (< 1,000 U/mL) have been shown to stimulate Staph- ylococcus aureus biofilm formation [8,10]. Conclusions A 40% ethanol solution can be combined with all IAAs but UFH. Enox/Eth was effective as an anti-biofilm agent with minor impacts on DC integrity and could be a useful interdialytic lock solution. Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol in 40% ethanol were 1350, 575, 307 and 207 U/ml, respectively, and up to 300 U/ml for danaparoid and 1 mg/mL for fondaparinux. Enoxaparin 400 U/mL in 40% ethanol (Enox/ Eth) eradicated biofilm after 4 hours of exposure for Staphylococcus epidermidis, Pseudo- monas aeruginosa and Candida albicans and after 24 hours for Klebsiella pneumoniae and S. aureus. Aliphatic carbonate and alcohol compounds were released by polyurethane-DC after Enox/Eth exposure, as after 40% ethanol or saline exposure. There was no significant difference between the amounts released after 30 minutes of exposure to Enox/Eth and 15 days to saline. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Results The compatibility of IAAs and ethanol varied according to IAA type and concentration, and ethanol content. UFH in 40% ethanol was not compatible, whatever the UFH concentration used. Established limits of compatibility of enoxaparin, nadroparin, dalteparin and tinzaparin Competing Interests: The authors have declared that no competing interest exists. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 1 / 21 Ethic Statement Whole blood (WB) samples were obtained from healthy blood donors at the local French blood agency (Etablissement Français du Sang [EFS], Saint-Etienne, France). In France, the use of blood samples from donors for research purpose is controlled by the State. The Law indi- cates that blood donation requires the systematic information of the volunteers (article R.1221- 5 of the Public Health Code, 01/12/2009 and 06/11/2006 decrees) and that written informed consents must be obtained by EFS from all donors whose samples are involved in research studies. The EFS controls the storage of the samples and their use in any study, in a legal and ethical framework, and there is no requirement for approval by a local Ethical Committee Introduction Higher inhibitory effects are obtained when chelat- ing agents are used in combination with anti-microbial agents such as antibiotics [11]. However, the widespread use of solutions containing a high concentration of antibiotics as lock solutions raises concerns in clinical practice because of side effects [12] and the risk for devel- opment of antimicrobial-resistant organisms [13–14]. Ethanol is an antiseptic that exerts bac- tericidal and fungicidal activity against a broad range of microorganisms. It acts by non- specific protein denaturation and thus is less likely to promote antimicrobial resistance. Several randomized control studies have assessed the efficacy of ethanol locks in preventing catheter infections and have yielded conflicting results [15–19]. These discrepancies may be related to differences in study design, case definition, case mix population, type of catheter, ethanol lock concentration and dwell time. Of these studies, only one focused on chronic hemodialysis patients with long-term tunneled catheters. The authors reported that locking catheter between dialysis sessions once a week with 70% (v/v) ethanol and twice with UFH was associated with a 67% decrease in catheter-related bloodstream infections as compared to standard interdialytic UFH lock thrice a week [17]. However, prolonged and repeated instillation with 70% (v/v) eth- anol concentration may be associated with catheter damage and obstruction requiring catheter PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 2 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol withdrawal [18]. Experimental data suggest that there is no need to use a 70% (v/v) ethanol solution to eradicate sessile microorganisms and that ethanol at lower concentrations, about 40% (v/v), exert antibiofilm effects [20–23]. Our preliminary studies on DCs reported that immersion in 40% ethanol has only a marginal impact on catheter integrity [24,25]. However, ethanol has no anticoagulant properties and should not be mixed with UFH because of precipi- tation [26]. UFH is a heterogeneous mixture of polysaccharides with variable molecular weight and whether ethanol causes precipitation of fractionated heparin, pentasaccharide Factor Xa inhibitor, and heparinoids remains unknown. These treatments are increasingly used for antic- oagulation during dialysis sessions [7,27,28] and could be an appropriate alternative for inter- dialytic lock solutions [29]. The aim of the study was to assess the compatibility of IAAs in eth- anol and to determine the anti-biofilm activity of the optimal combination against several microorganisms commonly involved in catheter-related infections and its effect on the integ- rity of polyurethane DCs. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol microbial study. Bacterial strains were grown in lysogeny broth and in minimal medium (M63B1) and the fungal species in 0.67% yeast nitrogen base (YNB, Difco) supplemented with 0.4% glucose. The organisms were maintained at −70°C in their respective medium with 15% glycerol, and on each occasion the biofilm was established from the original stock. microbial study. Bacterial strains were grown in lysogeny broth and in minimal medium (M63B1) and the fungal species in 0.67% yeast nitrogen base (YNB, Difco) supplemented with 0.4% glucose. The organisms were maintained at −70°C in their respective medium with 15% glycerol, and on each occasion the biofilm was established from the original stock. The catheters tested for microbial analyses were segments (each 1 cm long) of sterile silicone dialysis catheters (S-DCs) (DualCath1, Medcomp, Harleyville, PA, USA, and Hemotech, Ramonville, France), which are easier to cut and fix onto the glass slide of the microfermentors than polyurethane catheters. For chemical analysis, the catheters were 60 cm long unstuffed tunneled polyurethane catheters (PU-DCs) (Carbothane1, 85A, Medcomp, Harleyville, PA, USA) with no outside paint marks. The catheters tested for microbial analyses were segments (each 1 cm long) of sterile silicone dialysis catheters (S-DCs) (DualCath1, Medcomp, Harleyville, PA, USA, and Hemotech, Ramonville, France), which are easier to cut and fix onto the glass slide of the microfermentors than polyurethane catheters. For chemical analysis, the catheters were 60 cm long unstuffed tunneled polyurethane catheters (PU-DCs) (Carbothane1, 85A, Medcomp, Harleyville, PA, USA) with no outside paint marks. Visual Observations A single blinded observer assessed the IAA in ethanol for color and clarity under normal light- ing in a colorless glass vial held against white and black backgrounds. The appearance of any precipitate including cloudiness, film deposit and droplets was recorded after 1 min and 1, 24, 48 and 72 h of incubation at room temperature and at 37°C. Visual observation was used to determine the limits of solubility of IAA in ethanol at various IAA concentrations and in vari- ous ethanol contents. Observed solubility data of IAA in ethanol were plotted on a graph between IAA concentration (x) and ethanol content in % (y) for 72 h at room temperature and at 37°C. The relationship between the particular values of x corresponding to the IAA limit of solubility at a particular level of ethanol content y was subsequently established. Ethanol Assay Ethanol content was determined by a titrimetric assay used to measure blood alcohol levels [30] with samples of IAA in ethanol previously diluted at 1:100. Ethanol stability was assayed on Enox at 400 U/mL (Enox400) in ethanol. A control assay without enoxaparin was carried out at the same time to take into account possible errors resulting from oxidation of Enox. HPLC-ELSD Determination of Enoxaparin The HPLC-ELSD method (S1 Text) was used to measure enoxaparin concentrations in Enox400/ethanol mixtures at different levels of ethanol content (range 30 to 70%). Enoxaparin concentrations in Enox400 without ethanol were determined at the same time. The 95% confi- dence interval was determined by repetitive analysis (n = 6) of Enox400 samples in saline. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Chemicals, Reagents and Materials Mixing solutions were prepared using various commercially available injectable anticoagulant agents (IAAs): unfractionated heparin (UFH), heparin sodium (Choay, 5,000 U/mL, Sanofi- Aventis, France); low molecular weight heparins (LMWHs), enoxaparin sodium (Enox), (Love- nox1, 10,000 U/mL, Sanofi-Aventis, France), tinzaparin sodium (Innohep1, 20,000 U/mL, Leo Pharma, France), nadroparin calcium (Fraxiparine1, 9,500 U/mL, GlaxoSmithKline, France), dalteparin sodium (Fragmine1, 10,000 U/mL, Pfizer, France); and a heparinoid, nox1, 10,000 U/mL, Sanofi-Aventis, France), tinzaparin sodium (Innohep1, 20,000 U/mL, Leo Pharma, France), nadroparin calcium (Fraxiparine1, 9,500 U/mL, GlaxoSmithKline, France), dalteparin sodium (Fragmine1, 10,000 U/mL, Pfizer, France); and a heparinoid, danaparoid sodium (Orgaran1, 1,250 U/mL, MSD, France), pentasaccharide Factor Xa inhibi- tor, sodium fondaparinux (Arixtra1, 12.5 mg/mL, GlaxoSmithKline, France). In this study the wording IAA refers to UFH, LMWHs, fondaparinux and danaparoid sodium. Mixing solutions were prepared in 5 mL glass hemolysis tubes, combining either a IAA fixed concentration and increased concentrations of ethanol and prepared with absolute ethanol in 0.9% sodium chlo- ride solution (saline solution), or increased concentrations of IAA between 50 and 1,000 U/mL (and in some cases up to 2,500 U/mL) and a fixed ethanol content. All ethanol contents in the work are expressed in % (v/v). The ethanol content in the mixing solutions ranged between 30 and 70% in saline or in some experiments up to 95% ethanol. Absolute ethanol (purity more than 99.9%) for high performance liquid chromatography (HPLC) was obtained from Carlo Erba (Peypin, France), HPLC quality methanol was from Acros-Organics (Van Overbeek, Bel- gium) and 0.9% sodium chloride from Aguettant (Lyon, France). Ethyl acetate, used as solvent for extraction, was of analytical grade (Carlo Erba, Peypin, France), and cyclododecanol, used as GC-MS internal standard (IS), was from Sigma Aldrich (Saint Quentin Fallavier, France). Staphylococcus epidermidis CIP 68.21, S. aureus CIP 65.25 (methicillin resistant), P. aerugi- nosa ATCC 27853, Klebsiella pneumoniae LM21 and C. albicans SC5314 were selected for the PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 3 / 21 Silicone Catheter Biofilm Formation Biofilm was formed on an S-DC in 60-ml aerated microfermentors as described by Ghigo [32]. Sterile S-DCs were fixed onto the internal removable glass slide of the microfermentors. Strains from the frozen stocks were cultivated in M63B1-0.4% Glu or YNB-0.4% Glu medium over- night. An inoculum of 109 bacilli, 108 cocci or 107 C. albicans cells was used to inoculate micro- fermentors containing the silicone segments. Continuous flow of 100 mL/h of either M63B1- 0.4% Glu medium (bacterial strains) or YNB-0.4% Glu (yeast) and constant aeration with ster- ile pressed air (0.3 bar) were used to obtain continuous flow-through culture conditions. Our experimental model received a high input of fresh medium to avoid significant planktonic growth. After 24 h of incubation, the S-DC segments were removed from the incubator and separated from the device. The biofilms formed on the S-DC segments were resuspended in 5 ml M63B1 minimal or YNB medium by sonication and vortexing. Serial dilutions of the result- ing suspensions were performed and plated onto appropriate agar plates to determine the number of viable cells [Colony Forming Unit (CFU)] after overnight incubation at 37°C. The bacteria count was expressed as a decimal logarithm (log10). The limit of detection in our experimental conditions was 1.6 log10 (40 CFU) per KT segment. Definitions Solubility—IAA solubility in ethanol was defined at a given temperature as the absence in visual observations of any precipitate, including cloudiness, film deposit and droplets. Compatibility—The compatibility of IAA in ethanol was defined when IAA solubility in ethanol was observed after 1 min, 1, 24, 48 and 72 h of incubation at room temperature and at 37°C. Limit of solubility—The limit of solubility was defined at a given temperature as the highest IAA concentration soluble in ethanol at a particular level of ethanol content. Limit of compatibility—The limit of compatibility was defined as the highest concentration of IAA soluble in ethanol at a particular level of ethanol content at room temperature and at 37°C. Ethanol stability—The stability of ethanol in Enox/ethanol solutions was defined when the measured ethanol content was within 100.0 ± 2.0% of the theoretical value. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 4 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Enox stability—As requested in industry quality guidelines [31], a 95% two-sided confi- dence interval was used to verify enoxaparin stability. Enox400 stability in mixing solutions was defined when the measured concentration of Enox was within the range of the mean calcu- lated concentration ± 2 SD ([387.2–415.2]). Microbial Treatment Protocol After incubation, the S-DC segments harboring 24-hour biofilm were removed. Each segment was carefully rinsed in 1 mL of saline, and then placed in a tube containing 1 mL of the differ- ent lock solutions: (i) ethanol at 40%, (ii) sodium enoxaparin 400 U/mL, (iii) Enox/Eth mixing solution and (iv) 0.9% sodium chloride as control. For every organism, the experiments were repeated in triplicate or quadruplicate, and during each treatment assay S-DC segments were exposed to the different solutions for 4, 24 and 48 h at 37°C. Subsequently, the S-DC segments were removed, rinsed once with saline and the number of adherent viable microorganisms (CFU) was determined as described above. In addition, the biofilm biomass was determined in triplicate for each strain before treatment. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol recorded on the ethyl acetate extracts of Enox/Eth immersion solutions and compared with those obtained in saline as control and in 40% ethanol alone. The areas of all GC-MS peaks were summed and the mean ratios (sum of all peaks areas/internal standard area) were calcu- lated for comparative quantitative analysis. Protein Precipitation and Hemolysis Assessments Protein precipitation and hemolysis capacity of Enox/Eth was assessed using whole blood (WB) samples obtained from five healthy blood donors at the local French blood agency (Eta- blissement Français du Sang [EFS], Saint-Etienne). Blood donation requires the systematic information of the volunteers (article R.1221-5 of the Public Health Code, 01/12/2009 and 06/ 11/2006 decrees) and written informed consents were obtained by EFS from all donors involved in our study. WB samples (0.5 and 1 ml) were diluted with 4.5 or 4 mL, respectively, of the solution to be tested, i.e. enoxaparin 400, Enoxaparin 400 in 40% ethanol mixing solution or saline (control), thus the final suspensions contained 10% and 20% of WB. After 20 min of incubation at room temperature (20°C), the samples were centrifuged at 20°C and 4,000 r.p.m. for 10 min (Eppendorf centrifuge 5810R). If there were visible signs of precipitation, serum and precipitate were transferred into polypropylene tubes and centrifuged at 20°C and 10,800 r.p.m. for 10 min (Abbott Laboratories centrifuge 3530). The supernatant was removed and the precipitate was dissolved in 0.9% sodium chloride. Albumin was assayed with a Vista1 ana- lyzer (Siemens Healthcare Diagnostics, Saint-Denis, France). The tests were performed in triplicate. To assess hemolysis capacity, WB samples were diluted 1:200 in saline containing enoxa- parin 400 or Enox/Eth, or neither (control). The dilute WB samples were incubated at room temperature for 30 min and then centrifuged at 400g/5 min/25°C. An aliquot of supernatant from each centrifuged sample and absorbance (Abs) was read at 540 nm. Statistical Analysis The level of significance was set at P <0.05 for all the tests. The HPLC data (means ± SD) of enoxaparin determined in various ethanol contents and in saline for stability studies were com- pared using Statview software (SAS Institute, Cary, NC, USA) and the Wilcoxon and Mann- Whitney U tests. Concordance between the solubility and stability tests was assessed in Enox/ethanol solu- tions at a fixed Enox concentration of 400 U/mL (Enox400) and 40% ethanol using the Mac Nemar test. The microbial data expressed as means ± SD decimal logarithm (log10) of CFU were com- pared using the Mann-Whitney U test. The Enox/Eth effect on PU-DC was carried out using the mean ratios of released com- pounds determined by GC-MS on three catheters for each immersion condition (solvent and time of contact). The Wilcoxon test was used to compare repeated measurements over time. Differences between groups (Enox/Eth, 40% ethanol and saline) were compared using the Mann-Whitney U test. Data were analyzed using Statview 5.0 software (SAS Institute, Cary, NC, USA). Chemical Analysis of Polyurethane Catheters Each Carbothane1 PU-DC was entirely immersed in Enox/Eth and kept at 37°C for 30 min- utes, 4 hours or 15 days using the protocol described elsewhere [25]. Three catheters were used with each immersion condition and GC-MS analysis in electron impact mode was performed as previously described on the immersion media using dodecanol as internal standard and extraction with ethyl acetate [25]. GC-MS qualitative analysis was performed in full scan mode to determine the chro- matographic profile and establish structures of migrating components observed following immersion of Carbothane1 PU-DC in Enox/Eth for 15 days. Mass spectra of compounds sep- arated by gas chromatography were recorded and compared with those of previously identified compounds in immersion solutions (40% ethanol and control saline) of PU-DC [25]. All char- acterized structures were related with the aliphatic polycarbonate-polyurethane structure of the Carbothane1 polymer. Quantitative GC-MS analysis was carried out by selected ion mon- itoring mode (using characteristic ions in mass spectra at m/z 82, 83, 89, 101 and 139) on ethyl acetate extracts [25]. Reconstructed ion chromatograms (sum of characteristic ions) were 5 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol ethanol was observed using ethanol content at 30% and UFH concentrations from 50 to 2,500 U/ml, or ethanol at 35% and UFH concentration 100 U/ml. The results of the visual observa- tions of UFH in ethanol are given in S1 Table. In Enox/ethanol solutions for an ethanol content of 40%, compatibility was observed at all Enox concentrations from 50 to 1,200 U/ml. For a 45% ethanol content, compatibility was observed when Enox concentration was 100 U/ml. When ethanol content was 50%, no solubility was observed whatever the Enox concentration 50 U/ml. Precipitates were observed immediately (1 min) and were more extensive at room temperature than at 37°C. At longer contact times, decantation occurred and a film deposit or insoluble fine droplets appeared at the bottom of the tubes, after which the supernatant solution gradually became clear again. The results of visual observations of Enox in ethanol at room temperature and at 37°C are given in Table 1 and S2 Table. The linear trend curves of the limits of solubility of Enox in ethanol are shown in Fig 1A and 1B. Similarly, the limits of solubility of tinzaparin, nadroparin, dalteparin and UFH in ethanol based on visual observations at room temperature and at 37°C are shown in Fig 2A and 2B. Table 1. Testing grid for the visual determination of enoxaparin precipitation in ethanol. Influence of relative concentrations, time of contact and temperature. Ethanol Enoxaparin Room temperature 37°C (%, v/v) (U/mL) 1min 1h 24h 48h 72h 1min 1h 24h 48h 72h 40 50 0 0 0 0 0 0 0 0 0 0 100 0 0 0 0 0 0 0 0 0 0 200 0 0 0 0 0 0 0 0 0 0 300 0 0 0 0 0 0 0 0 0 0 400 0 0 0 0 0 0 0 0 0 0 500 0 0 0 0 0 0 0 0 0 0 600 0 0 0 0 0 0 0 0 0 0 800 0 0 0 0 0 0 0 0 0 0 1000 0 0 0 0 0 0 0 0 0 0 1200 0 0 0 0 0 0 0 0 0 0 1500 + + + + + 0 0 0 0 0 2500 + + + + + 0 0 0 0 0 45 50 0 0 0 0 0 0 0 0 0 0 100 0 0 0 0 0 0 0 0 0 0 200 + + + + + 0 0 0 0 0 300 + + + + + 0 0 0 0 0 400 + + + + + 0 0 0 0 0 500 + + + + + 0 0 0 0 0 600 + + + + + 0 0 0 0 0 800 + + + + + + + + + + 1000 + + + + + + + + + + 50 50 + + + + + + + + + + 100 + + + + + + + + + + 400 + + + + + + + + + + 500 + + + + + + + + + + 1000 + + + + + + + + + + 0 b f i i i l di l di ﬁl d i d d l nation of enoxaparin precipitation in ethanol. Influence of relative concentrations, time of contact and Table 1. Testing grid for the visual determination of enoxaparin precipitation in ethanol. Influence of relative concentrations, time of contact and temperature. ng grid for the visual determination of enoxaparin precipitation in ethanol. Influence of relative concentrations Table 1. Testing grid for the visual determination of enoxaparin precipitation in ethanol. Influence of relative temperature. 0, absence of precipitates including cloudiness, ﬁlm deposit and droplets +, presence of precipitates including cloudiness, ﬁlm deposit and droplets 0, absence of precipitates including cloudiness, ﬁlm deposit and droplets Solubility Assessed by Visual Observations Solubility Assessed by Visual Observations In UFH/ethanol solutions, for an ethanol content 40%, no compatibility was observed what- ever the UFH concentration between 50 and 2,500 U/ml used. The compatibility of UFH in In UFH/ethanol solutions, for an ethanol content 40%, no compatibility was observed what- ever the UFH concentration between 50 and 2,500 U/ml used. The compatibility of UFH in 6 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 0, absence of precipitates including cloudiness, ﬁlm deposit and droplets +, presence of precipitates including cloudiness, ﬁlm deposit and droplets 0, absence of precipitates including cloudiness, ﬁlm deposit and droplets + presence of precipitates including cloudiness ﬁlm deposit and droplets PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 7 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Fig 1. Visual determination of Enox solubility (in U/mL) according to ethanol content (in %) and linear trend curves of the limits of solubility following 72h of mixing a) at room temperature and b) at 37°C. doi:10.1371/journal.pone.0159475.g001 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Fig 1. Visual determination of Enox solubility (in U/mL) according to ethanol content (in %) and linear trend curves of the limits of solubility following 72h of mixing a) at room temperature and b) at 37°C. doi:10.1371/journal.pone.0159475.g001 Fig 1. Visual determination of Enox solubility (in U/mL) according to ethanol content (in %) and linear trend curves of the limits of solubility following 72h of mixing a) at room temperature and b) at 37°C. doi:10.1371/journal.pone.0159475.g001 Fig 1. Visual determination of Enox solubility (in U/mL) according to ethanol content (in %) and linear trend curves of the limits of solubility following 72h of mixing a) at room temperature and b) at 37°C. doi:10.1371/journal.pone.0159475.g001 doi:10.1371/journal.pone.0159475.g001 8 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Fig 2. Linear trend curves of the limits of solubility of UFH (heparin: H) and LMWHs (Enox: E, tinzaparin: T, nadroparin: N and dalteparin: D) in ethanol a) at room temperature and b) at 37°C. doi:10.1371/journal.pone.0159475.g002 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Fig 2. Linear trend curves of the limits of solubility of UFH (heparin: H) and LMWHs (Enox: E, tinzaparin: T, nadroparin: N and dalteparin: D) in ethanol a) at room temperature and b) at 37°C. doi:10.1371/journal.pone.0159475.g002 Fig 2. Linear trend curves of the limits of solubility of UFH (heparin: H) and LMWHs (Enox: E, tinzaparin: T, nadroparin: N and dalteparin: D) in ethanol a) at room temperature and b) at 37°C. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 9 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Table 2. Equations estimating the limit of compatibility of IAAs in ethanol. Stability of Enox400/Ethanol Solutions Measured ethanol contents in Enox400/ethanol solutions at ethanol contents of 40%, 45%, and 50%, were within 100.0 ± 2.0% of the theoretical values, showing that ethanol was stable in these combinations (S5 Table). In Enox400/ethanol solutions, Enox400 was stable when ethanol content was 40% (Table 3). There was a significant decrease between theoretical and calculated Enox concentra- tions when the ethanol content was 45%. The higher the ethanol content increased the greater the stability of Enox400 decreased (Table 3 and S5 Table). For an ethanol content of 70%, Enox concentration in the clear supernatant obtained after centrifugation was below the limit of detec- tion (LOD) of the method (< 25 U/mL in diluted samples) (S1 Fig). The HPLC analysis of drop- lets formed at the bottom of the tubes after centrifugation evidenced the presence of Enox in the droplets (data not shown). For Enox400/ethanol solutions there was a concordance between compatibility and stability. HPLC determination of enoxaparin concentration in Enox400/etha- nol solutions at various ethanol contents (30% to 70%) indicated that Enox concentrations were unchanged in Enox/ethanol solutions without precipitates, whereas Enox concentrations decreased in mixing solutions containing precipitates (P = 1). Injectable anticoagulant agents / ethanol lock solutions Formula Unfractionated heparin / ethanol y = -0.005x + 36.5 Enoxaparin / ethanol y = -0.004x + 45.4 Nadroparin / ethanol y = -0.008x + 44.6 Dalteparin / ethanol y = -0.014x + 44.3 Tinzaparin / ethanol y = -0.015x + 43.1 x, injectable anticoagulant agents; y, ethanol content in % (y) doi:10.1371/journal.pone.0159475.t002 Table 2. Equations estimating the limit of compatibility of IAAs in ethanol. Since all heparins (UFH and LMWHs) in ethanol without any precipitation at room tempera- ture were precipitate-free at 37°C, their limits of solubility in ethanol at room temperature were used to define the limits of compatibility. Equations used to determine the limits of compatibil- ity of heparins in ethanol are given in Table 2. As estimated by equations, the limits of compati- bility of heparins in 40% ethanol were 1,350 U/ml for Enox, 575 U/ml for nadroparin, 307 U/ ml for dalteparin, and 207 U/ml for tinzaparin. In danaparoid/ethanol solutions, for 40% ethanol, compatibility was observed when dana- paroid concentration was up to 300 U/ml. For 45% ethanol, compatibility was observed at danaparoid concentrations up to 200 U/ml. No compatibility was observed for danaparoid concentrations of 100 U/ml in 50% ethanol (S3 Table). In fondaparinux/ethanol solutions, for ethanol content up to 60%, compatibility was observed at all fondaparinux concentrations between 0.1 and 1 mg/mL (S4 Table). Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Table 3. HPLC-ELSD determination of enoxaparin concentrations of Enox400 in 30% to 70% ethanol at room temperature. % EtOH Calculated enoxaparin concentration Mean value Coefﬁcient of variation Recovery P (v/v) (U/mL) (U/mL) ± SD (%) (%) 30% 396.5 396.8 ± 2.1 0.5 99.2 0.4386 394.9 399.1 40% 403.8 405.5 ± 2.5 0.6 101.4 0.4386 408.4 404.4 45% 381.3 381.9 ± 0.8 0.2 95.5 0.0201 382.9 381.6 50% 333.5 332.1 ± 1.3 0.4 83.0 0.0201 331.1 331.6 60% 221.8 226.6 ± 4.4 1,9 56.6 0.0201 227.6 230.4 70% < LOD - - - - < LOD < LOD LOD: limit of detection P value in the 95% conﬁdence interval doi:10 1371/journal pone 0159475 t003 ermination of enoxaparin concentrations of Enox400 in 30% to 70% ethanol at room temperature. Table 3. HPLC-ELSD determination of enoxaparin concentrations of Enox4 decrease in biofilm mass, whatever the microorganism tested as compared to saline. Microor- ganisms were eradicated after a 4-hour exposure to both 40% ethanol and Enox/Eth for P. aer- uginosa, S. epidermidis, and C. albicans biofilms, and after 24 hours for S. aureus and K. pneumoniae biofilms. Whatever the sessile microorganisms tested, eradication was achieved with both 40% ethanol and Enox/Eth after 24, 48, and 72h of treatment. There were no differ- ences in the number of viable microorganisms between treatment with 40% ethanol and Enox/ Eth whatever the microorganisms and treatment duration. Anti-Biofilm Activities of Enoxaparin/Ethanol-Lock Solution Results of the anti-biofilm activity of the Enox400/Ethanol 40% solution are presented in Fig 3, 3A, 3B and 3C. The mean baseline counts of S. epidermidis, S. aureus, K. pneumoniae, P. aeru- ginosa and C. albicans adhered to catheters were 7.4× 106 cfu, 4 .6× 106 cfu, 7.2× 108 cfu, 3.2 × 109 cfu and 1.8× 106 cfu, respectively. Whatever the microorganism, there were no differ- ences in the number of viable cells after 4, 24 and 48 hours between enoxaparin 400U/mL and control saline. In contrast, after a 4-hour exposure to 40% ethanol or Enox/Eth, there was a PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 10 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Integrity of PUR Catheters Immersed in Enoxaparin/Ethanol-Lock Solution GC-MS profiles for qualitative analysis of the compounds, released following immersion of Carbothane1 PU-DC in Enox/Eth, were identical to those obtained following immersion in 40% ethanol and control saline solution (Fig 4A, 4B and 4C). Mass spectra analysis of chro- matographic peaks showed that the release compounds were the same as those previously observed following immersion in 40% ethanol alone, and no new compound was detected. All the characterized structures were related to the complex polycarbonate-polyurethane structure of the Carbothane1 polymer whose synthesis constituents are aliphatic polycarbonates and isocyanates. Major released compounds were aliphatic carbonate and alcohol structures (Fig 4A). Minor compounds (< 2% of the sum of all the compounds released) were related to the aliphatic 4,4’-dicyclohexylmethane diisocyanate isomer (H12MDI) structure of the polymer (Fig 4A). Table 4 shows the mean ratios and standard deviations (n = 3) of the sum of the compounds released observed following immersion of Carbothane1 PU-DC in Enox/Eth and those 11 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol y, g 371/journal.pone.0159475 July 21, 2016 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 12 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Fig 3. Effect of 40% Ethanol, Enoxaparin 400 U, Enoxaparin 400–40% Ethanol and saline control on microbial colony counts of different organisms in 24-hour old biofilms. Each bar represents mean CFU per centimeter of catheter at different time points along with respective standard errors of the means. The y axis is on a logarithmic scale. The limit of detection is 40 CFU and * indicates P<0.05 compared to the control at the same time. doi:10.1371/journal.pone.0159475.g003 previously measured in ethanol at 40% [25]. There was no significant difference in the amounts of compounds released between Enox/Eth and 40% ethanol after 30 min, 4 hours and 15 days (P = 0.8273, P = 0.2752, and P = 0.1266 respectively). The mean value of compounds released after 30 minutes of exposure to Enox/Eth was lower than this observed after 15 days of expo- sure to saline: 0.694 ± 0.049 vs 0.837 ± 0.127 (P = 0.0492), respectively. In contrast, as com- pared to 15 days exposure to saline, the mean values of compounds released after 4 hours and 15 days of exposure to Enox/Eth were higher: 1.920 ± 0.447 (P = 0.492), and 4.775 ± 0.619 (P = 0.492), respectively. Protein Precipitation and Hemolysis Assessments Protein precipitation was visually detected in all WB suspensions performed in enoxaparin 400 and in Enox/Eth, but not when WB was mixed with saline alone. The mean values of albumin in the precipitates of the suspensions at 10% and 20% WB ratios were 14.2 ± 7.1 and 11.0 ± 3.4 mg/L for Enoxaparin and 26.9 ± 13.8 and 103.7 ± 43.3 for Enox/Eth, respectively (Table 5). The hemolysis capacities of the different solutions are given in Table 6. Discussion UFH is the predominant anticoagulant lock solution used to prevent interdialytic catheter occlusion. The UFH lock concentrations used range between 1,000 and 10,000 U/mL [33]- [34]. Because of potential bleeding complications, the American Society for Diagnostic and Interventional Nephrology has recommended administering UFH at a concentration of 1,000 U/mL for locking dialysis catheters in patients without catheter occlusion [33]. UFH has no antibiofilm properties. Combining UFH with antibiotics reduces the rate of catheter infections [35]. However, the strategy is not recommended since it may promote bacterial resistance [13]. Ethanol is an antiinfectious agent with low risk of inducing bacterial resistance since it acts by protein denaturation. It could therefore be an attractive antimicrobial lock agent [36]. Its anti- biofilm efficacy increases with its concentration [23]. Although catheter exposure to high ethanol concentration was suspected of reducing the elasticity of elastomers, Crnich et al [37] concluded that exposure of silicone and polyurethane catheters to a 70% ethanol lock solution does not appreciably alter their mechanical properties. In the present study, we focused on 40% ethanol solutions since this ethanol content has anti- biofilm properties [23],[38] with only a marginal impact on catheter structural degradation [24], [25]. Ethanol has no anticoagulant properties and, therefore, using ethanol as a lock solu- tion without an anticoagulant could alter catheter patency [39]. There are concerns about com- bining ethanol with heparin since ethanol is classically used for fractional precipitation of polysaccharides [40], [41]. Our results indicate that UFH cannot be mixed with 40% ethanol owing to incompatibility. Although LMWHs have become the anticoagulation of choice for intermittent hemodialysis sessions in Europe [27], their use for interdialytic catheter locking has been scarcely reported [28], [29]. To the best of our knowledge, the compatibility of LMWHs/ethanol mixing solu- tions has not been previously documented. The optimal LMWH concentration for catheter lock remains unknown and probably differs depending on the kind of LMWH used, their 13 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Fig 4. GC-MS Total ion current profiles of ethyl acetate extracts obtained from immersion solutions in contact with Carbothane1 PU-DC at 37°C: (A) Enox/Eth immersion solution for 30 min. (B) 40% ethanol Fig 4. GC-MS Total ion current profiles of ethyl acetate extracts obtained from immersion solutions in contact with Carbothane1 PU-DC at 37°C: (A) Enox/Eth immersion solution for 30 min. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Discussion (B) 40% ethanol 14 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol solution for 30 min. (C) Reference saline solution for 15 days. Proposed structures of release compounds are given in a) 2 and 4 are H12MDI isomers, 3, 6, 7 and 8 are main aliphatic alcohol and carbonates structures. C1 to C5 are minor aliphatic alcohol and carbonate structures. solution for 30 min. (C) Reference saline solution for 15 days. Proposed structures of release compounds are given in a) 2 and 4 are H12MDI isomers, 3, 6, 7 and 8 are main aliphatic alcohol and carbonates structures. C1 to C5 are minor aliphatic alcohol and carbonate structures. doi:10.1371/journal.pone.0159475.g004 doi:10.1371/journal.pone.0159475.g004 pharmacodynamic properties being not clinically interchangeable [7]. A theoretical concentra- tion of IAA for catheter locking could be estimated by extrapolating the pharmacodynamics of UFH. Thus, the recommended concentration of UFH locks represents about 10% of the daily dose of UFH currently used for venous thrombosis prophylaxis (10,000 U/day). If this 10% amount was applied to IAAs on the basis of their recommended daily dose for venous throm- bosis prophylaxis [42] the resulting estimated concentrations of LMWHs for lock concentra- tions would approximate 200 U/mL for Enox, 280 U/mL for nadroparin and 250 U/ml for dalteparin and tinzaparin. We carried out the solubility tests on a wide range of LMWH con- centrations including these estimated theoretical concentrations. The study showed that differ- ent LMWHs had different solubilities at a given ethanol content and, as generally admitted for molecules having similar structures, solubility decreased with the increase in molecular weight. Our results are in agreement with this observation: among the IAAs tested, UFH, which has the highest molecular weight, had the lowest solubility in ethanol, whereas fondaparinux, which has the lowest molecular weight, had the highest solubility. At the estimated theoretical concentrations suitable for catheter locking in 40% ethanol, tinzaparin precipitated whereas compatibility was observed for nadroparin, dalteparin and Enox. Whether these estimated the- oretical LMWH concentrations could be used in a clinical setting to maintain catheter patency is unknown. Whatever the case, our study strongly suggests that if physicians decided to administer LMWH/ethanol lock solutions, they should prefer Enox, nadroparin and dalteparin over tinzaparin because the last is less soluble in ethanol. Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Table 5. Results of in vitro albumin precipitation test (mg/L). Test solutions (1) Saline (control) Enox. 400 Enox. 400–40% Ethanol 10% WB 20% WB 10% WB 20% WB 10% WB 20% WB Blood sample Donor sex (Age) #1 H (56) - - 21.7 6.7 33.2 82.8 #2 F (32) - - 11.7 15.1 17.8 46.6 #3 F (32) - - 21.4 13.8 46 146 #4 H (34) - - 11 8.61 27.7 148 #5 F (30) - - 5.2 11.0 10.2 95.3 mean - - 14.2 11.0 26.98 103.74 SD - - 7.1 3.4 13.85 43.36 (1) Test solutions consisted of 0.5 (10%) or 1 mL (20%) WB and 4 mL or 4.5 mL of either saline (control), Enoxaparin 400, or Enoxaparin 400–40% Ethanol solution. N i it ti b d Table 5. Results of in vitro albumin precipitation test (mg/L). (1) Test solutions consisted of 0.5 (10%) or 1 mL (20%) WB and 4 mL or 4.5 mL of either saline (control), Enoxaparin 400, or Enoxaparin 400–40% Ethanol solution. As the Enox/Eth mixing solution was stable, the anti-infectious activity of such a solution was assessed using five microorganisms embedded in 24-hour old monospecies biofilms that are commonly involved in catheter infections. After 4h of exposure, significant decreases in the numbers of viable microorganisms within the biofilms were observed with 40% ethanol alone or in combination with enoxaparin. The anti-microbial activities of ethanol were not modified by the addition of enoxaparin (400 U/ml). In addition, treatment with sodium enoxaparin at 400 U/mL alone (control) did not modify the biofilm biomass whatever the microorganisms, contrary to findings reported by Shanks et al with S. aureus [8]. Depending on the microorganism studied, complete eradication of the biofilm was observed after 4 to 24h of incubation, with K. pneumoniae and S. aureus being the most resistant. In all cases, the length of time required to achieve in vitro biofilm eradication was shorter than the 48 to 72-hour dwell time of interdialytic locks. In a previous study using 60% ethanol treatment, we showed that a shorter incubation time (30 min) completely eradicated established biofilms formed with the same microorganisms [36]. However, such high concentrations are not compati- ble with the addition of injectable anticoagulant agents (IAAs) and would result in precipitation, except for sodium fondaparinux. Discussion There are few reports of anticoagulation with danaparoid and fondaparinux for circuit patency during hemodialysis [43], [44] and no data are available on their use and dose as a catheter lock solution. As calculated for LMWHs, estimated theoretical lock concentrations would approximate 150 U/mL for danaparoid and 0.25 mg/L (i.e. 232.5 U/mL) for fondapari- nux. Our results indicate that danaparoid and fondaparinux at these concentrations are com- patible in 40% ethanol. Among the IAAs assessed in the study, fondaparinux had the highest compatibility in ethanol. According to the European Pharmacopeia [45], LMWH concentrations are usually deter- mined by measuring the anti-factor Xa and anti-factor II activities in TRIS buffer and the assay is performed by absorptiometry at 405 nm. Given the various hydroalcoholic medium condi- tions and the non-specific low absorbance wavelengths (190–210 nm) of IAA polysaccharides, an HPLC-ELSD assay was developed. Our study suggests that visual observation of IAAs in ethanol is an accurate surrogate of stability. Table 4. Mean ratios ± standard deviations (SD) of the sum of released products observed following immersion of Carbothane 1catheters (n = 3) in the mixing solutions of Enox/Eth, 40% ethanol and saline for storage times of 30 min, 4 hours and 15 days. Storage time Sum of released products Mean ratio ± SD (n = 3) Enox/Eth 40% ethanol Saline 30 min 0.694 ± 0.049 0.677 ± 0.068 0.038 ± 0.007 4 h 1.920 ± 0.447 2.443 ± 0.289 0.151 ± 0.048 15 days 4.775 ± 0.619 6.122 ± 0.730 0.837 ± 0.127 doi:10.1371/journal.pone.0159475.t004 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 15 / 21 Table 4. Mean ratios ± standard deviations (SD) of the sum of released products observed following immersion of Carbothane 1catheters (n = 3) in the mixing solutions of Enox/Eth, 40% ethanol and saline for storage times of 30 min, 4 hours and 15 days. Storage time Sum of released products Mean ratio ± SD (n = 3) Enox/Eth 40% ethanol Saline 30 min 0.694 ± 0.049 0.677 ± 0.068 0.038 ± 0.007 4 h 1.920 ± 0.447 2.443 ± 0.289 0.151 ± 0.048 15 days 4.775 ± 0.619 6.122 ± 0.730 0.837 ± 0.127 doi 10 1371/jo rnal pone 0159475 t004 15 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol The catheters used were made of polyurethane elastomer, whose precursors are aliphatic polycarbonates and an aliphatic isocyanate (H12MDI). Aliphatic polycarbonates are considered to be biocompatible and slightly toxic compounds [46]. H12MDI is only highly toxic when inhaled [47]. In our study, GC-MS qualitative and quantitative analyses of PU-DC migrating components in Enox/Eth immersion solution showed that the major compounds released were aliphatic polycarbonate structures with only a slight release of aliphatic H12MDI isomers, sug- gesting that PU-DC exposure to ethanol is safe. The safety of 40% ethanol used as a lock solution and its low impact on catheter integrity observed in our study are in agreement with a recent study reporting no adverse event when locking dialysis PU-DCs were repetitively used [37]. Since ethanol can cause plasma protein precipitation at concentrations above 28% [48], we assessed plasma albumin precipitation induced by Enox/Eth. About 20% of locking solution can spill out from catheters into the blood stream during locking [49] and is replaced by WB in the catheter. Hence, we investigated plasma albumin precipitation in suspensions with ratios of 10% and 20% of WB. Albumin precipitation was observed as previously reported [48]. When the current hematocrit value (45%) and the current albumin concentration in plasma (40 g/L) were taken into account, the maximum amount of albumin precipitated in our study was lower than 0.5% of the initial albumin content of the suspension, suggesting that this precipitation has only a marginal impact in a clinical setting. Enox/Eth has been shown to preserve anticoagulant properties [50]. This finding, in combi- nation with our observations, strongly suggests that it could be safely used as an interdialytic lock solution in a clinical setting and be effective in preventing catheter infections and main- taining catheter patency. We are aware that our study has some important limitations. First, the compatibility study was based on visual observations, which are less accurate than turbidimetry in measuring pre- cipitation [51]. However, the visual observation approach is classically used to assess the solu- bility of antibiotics/UFH lock solutions [52] and the precipitation of protein in ethanol [48]. Second, we cannot exclude the possibility that mixing solutions defined as compatible in vitro may precipitate in vivo. Indeed, the solution locked into the patient’s catheter partly leaks into the systemic circulation. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Concentrations of ethanol between 30 and 80% have also been shown to eradicate Candida biofilms in a dose-dependent manner, with optimal concentration— determined as the fastest eradication with lowest ethanol strength—being 40% [38]. Exposure of catheters to ethanol is still largely debated owing to the fear of release of com- pounds and impairment of catheter integrity [39]. We previously described the conditions of ultra-structural integrity and chemical release of S-DC and PU-DC materials following ethanol immersion [24], [25]. S-DC material maintained integrity in ethanol at 60% [24]. For this rea- son, we elected to perform experiments in this study on PU-DC material. Table 6. Assessment of in vitro hemolysis (OD measured at 540 nm). Blood sample Donor sex (age) Saline (Control) Enox. 400 Enox. 400–40% Ethanol #1 H (56) 0.003 0.009 0.437 #2 F (32) 0.004 0.008 0.450 #3 F (32) 0.005 0.005 0.431 #4 H (34) 0.004 0.008 0.436 #5 F (30) 0.005 0.049 0.443 d i 10 1371/j l 0159475 t006 Table 6. Assessment of in vitro hemolysis (OD measured at 540 nm). PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 16 / 21 This leakage is accompanied by a concomitant blood inflow allowing plasma proteins to enter the catheter lumen and precipitate in the ethanol [48]. Third, the sta- bility of IAAs in ethanol and the concordance between compatibility and stability were only determined with Enox400/ethanol solutions. Whether the results can be extrapolated to solu- tions with different Enox concentrations and to other IAAs in ethanol remains questionable. Fourth, the anti-thrombotic activity of all mixtures of Enox400 in ethanol was not assessed for all concentrations of ethanol. However, the stability of Enox400 in ethanol was shown by an HPLC-ELSD method and the absence of degradation of Enox was confirmed by the chro- matographic analysis of droplets recovered at the bottom of tubes following precipitation of enoxaparin in ethanol. In addition, we recently demonstrated that Enox400 in 40% ethanol has both anti-thrombotic and anti-Xa activities [50]. With regard to the antibiofilm capacities of the Enox/Eth solution, the possibility that the eradication of biofilm embedded in silicone cath- eters observed after a 4-hour exposure to Enox/Eth for S. epidermidis, P. aeruginosa and C. albi- cans and after a 24-hour for S. aureus and K. pneumoniae was not definite since we did not assess regrowth. Another limit of our study is that polyurethane belongs to the family of elasto- mers, which are characterized by the heterogeneity of their components and whether the results observed after Carbothane1 PU-DC exposure to Enox/Eth in our study can be extrap- olated to other polyurethane catheters remains questionable. Finally, the study did not assess the mechanical properties of catheters immersed in Enox/Eth. However it has been previously shown that prolonged exposure of polyurethane and silicone catheters to 70% ethanol lock PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 17 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol solution has a negligible impact on their mechanical properties [37]. In addition, in clinical set- tings, the adverse effects observed after exposure of polyurethane catheters to ethanol were only observed when ethanol content was 70% v/v or higher [39]. Conclusions Our study showed that, first, the compatibility of IAAs in ethanol varies depending on the kind of anticoagulant used. Second, satisfactory compatibility of enoxaparin, nadroparin, dalteparin, fondaparinux and danaparoid was observed in 40% ethanol. Of the LMWHs, enoxaparin exhibited the highest solubility and compatibility in 40% ethanol. In addition, we showed that 40% ethanol does not alter the silicone structure of catheters [24] and has only a marginal impact on polyurethane (Carbothane1) catheter structural degradation [25]. The selected Enox/Eth solution is an effective anti-biofilm solution with minor impacts on Carbothane1 polymer integrity. The compounds, released after Enox/Eth or 40% ethanol exposure, were similar to those observed after prolonged saline exposure and were non-harmful. Enox/Eth may be of value as an interdialytic lock solution in preventing catheter infection. It would be timely now to perform large trials to assess the efficacy of IAAs in ethanol as interdialytic lock solutions in preventing catheter infections. Supporting Information S1 Fig. HPLC-ELSD profiles of enoxaparin solutions. Enoxaparin at 400 U/l diluted (A) in H2O, (B) In 0.9% NaCl, (C) In 50% ethanol. Enlargement of the profile from 2.5 to 5.5 min, and (D) In 70% ethanol. Enlargement of the profile from 2.5 to 5.5 min. (DOCX) S1 Table. Testing grid for the visual determination of unfractionated heparin precipitation in ethanol. Influence of relative concentrations, time of contact and temperature on UFH solu- bility. (DOCX) S1 Table. Testing grid for the visual determination of unfractionated heparin precipitation in ethanol. Influence of relative concentrations, time of contact and temperature on UFH solu- bility. (DOCX) References 1. Collins AJ, Foley RN, Herzog C, Chavers B, Gilbertson D, Ishani A, et al. United States Renal Data Sys- tem 2008 Annual Data Report Abstract. Am J Kidney Dis 53(vi–vii): S8–S374, 2009 Bas du formulaire. 1. Collins AJ, Foley RN, Herzog C, Chavers B, Gilbertson D, Ishani A, et al. United States Renal Data Sys- tem 2008 Annual Data Report Abstract. Am J Kidney Dis 53(vi–vii): S8–S374, 2009 Bas du formulaire. 2. Rayner HC, Pisoni RL. The increasing use of hemodialysis catheters: evidence from the DOPPS on its significance and ways to reverse it. Semin Dial. 2010; 23: 6–10. doi: 10.1111/j.1525-139X.2009. 00675.x PMID: 20331810 2. Rayner HC, Pisoni RL. The increasing use of hemodialysis catheters: evidence from the DOPPS on its significance and ways to reverse it. Semin Dial. 2010; 23: 6–10. doi: 10.1111/j.1525-139X.2009. 00675.x PMID: 20331810 3. Williams AW, Dwyer AC, Eddy AA, Fink JC, Jaber BL, Linas SL, et al. American Society of Nephrology Quality, and Patient Safety Task Force. Critical and honest conversations: the evidence behind the "Choosing Wisely" campaign recommendations by the American Society of Nephrology. Clin J Am Soc Nephrol. 2012; 7:1664–1672. doi: 10.2215/CJN.04970512 PMID: 22977214 4. Renal Data System. USRDS 2003 annual data report: atlas of end-stage renal disease in the United States. Bethesda, MD: National Institute of Diabetes and Digestive and Kidney Diseases, 2003. 5. O’Grady NP, Alexander M, Dellinger EP, Gerberding JL, Heard SO, Maki DG, Masur H, et al. Guide- lines for the prevention of intravascular catheter-related infections. MMWR Recomm Rep. 2002; 51:1– 26. 6. Jain G, Allon M, Saddekni S, Barker JF, Maya ID. Does heparin coating improve patency or reduce infection of tunneled dialysis catheters? Clin J Am Soc Nephrol. 2009; 4:1787–1790. doi: 10.2215/ CJN.03920609 PMID: 19729425 7. Malo J, Jolicoeur C, Theriault F, Lachaine J, Senecal L. Comparison between standard heparin and tin- zaparin for haemodialysis catheter lock. 2010; ASAIO J 56: 42–7. doi: 10.1097/MAT. 0b013e3181c84e61 PMID: 20051836 8. Shanks RMQ, Donegan NP, Graber ML, Buckingham SE, Zegans ME, Cheung AL, et al. Heparin stim- ulates Staphylococcus aureus biofilm formation. Infect Immun. 2005; 73:4596–4606. PMID: 16040971 9. Miceli MH, Bernardo SM, Ku TS, Walraven C, Lee SA. In vitro analyses of the effects of heparin and parabens on Candida albicans biofilms and planktonic cells. Antimicrob Agents Chemother. 2012; 56:148–153. doi: 10.1128/AAC.05061-11 PMID: 21986822 10. Shanks RM, Sargent JL, Martinez RM, Graber ML, O'Toole GA. S2 Table. Testing grid for the visual determination of enoxaparin 400 U/mL precipitation in ethanol. Influence of relative ethanol contents, time of contact and temperature on enoxa- parin 400 U/mL solubility. (DOCX) S2 Table. Testing grid for the visual determination of enoxaparin 400 U/mL precipitation in ethanol. Influence of relative ethanol contents, time of contact and temperature on enoxa- parin 400 U/mL solubility. (DOCX) S3 Table. Testing grid for the visual determination of danaparoid precipitation in ethanol. Influence of relative concentrations, time of contact and temperature on danaparoid solubility. (DOCX) S4 Table. Testing grid for the visual determination of fondaparinux precipitation in etha- nol. Influence of relative concentrations, time of contact and temperature on fondaparinux sol- ubility. (DOCX) S5 Table. Ethanol content in enoxaparin 400U/mL/ethanol at different levels of ethanol content (40%, 45%, and 50%). (DOCX) S1 Text. Description of the HPLC-ELSD method used for enoxaparin concentration mea- surement. (DOCX) 18 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol Acknowledgments This work was supported by a financial support from the Auvergne FEDER and a grant from Hemotech laboratories. The authors would like to thank Dr Hélène Bargnoux for helpful dis- cussions, Elsa Aigouy for her technical assistance, Damien Bouvier for determination of albu- min concentrations and Jeffrey Watts for his help in preparing the manuscript. Author Contributions Conceived and designed the experiments: DB MQ CF CL BS. Performed the experiments: DB MQ NC. Analyzed the data: DB MQ CF CL BS. Contributed reagents/materials/analysis tools: CF CL BS. Wrote the paper: DB MQ CF CL BS. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 References Catheter lock solutions influence staphylococcal biofilm formation on abiotic surfaces. Nephrol Dial Transplant. 2006; 21: 2247–2255. PMID: 16627606 11. Chauhan A, Lebeaux D, Ghigo JM, Beloin C. Full and broad-spectrum in vivo eradication of catheter- associated biofilms using gentamicin-EDTA antibiotic lock therapy. Antimicrob Agents Chemother. 2012; 56: 6310–6318. doi: 10.1128/AAC.01606-12 PMID: 23027191 12. Saxena AK, Panhotra BR, Naguib M. Sudden irreversible sensory-neural hearing loss in a diabetic on hemodialysis, receiving amikacin as antibiotic-heparin lock. Pharmacotherapy. 2002; 22:105–108. PMID: 11794420 13. Landry DL, Braden GL, Gobeille SL, Haessler SD, Vaidya CK, Sweet SJ. Emergence of gentamicin- resistant bacteremia in hemodialysis patients receiving gentamicin lock catheter prophylaxis. Clin J Am Soc Nephrol. 2010; 5:1799–804. doi: 10.2215/CJN.01270210 PMID: 20595689 14. Dixon JJ, Steele M, Makanjuola AD. Anti-microbial locks increase the prevalence of Staphylococcus aureus and antibiotic-resistant Enterobacter: observational retrospective cohort study. Nephrol Dial Transplant. 2012; 27:3575–81. doi: 10.1093/ndt/gfs081 PMID: 22513704 15. Sanders J, Pithie A, Ganly P, Surgenor L, Wilson R, Merriman E, et al. A prospective double-blind ran- domized trial comparing intraluminal ethanol with heparinized saline for the prevention of catheter- associated bloodstream infection in immunosuppressed haematology patients. J Antimicrob Che- mother. 2008; 62:809–815. doi: 10.1093/jac/dkn284 PMID: 18621987 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 19 / 21 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol 16. Slobbe L, Doorduijn JK, Lugtenburg PJ, el Barzouhi A, Boersma E, van Leeuwen WB, et al. Prevention of Catheter-Related Bacteremia with a Daily Ethanol Lock in Patients with Tunneled Catheters: A Ran- domized, Placebo-Controlled Trial. Plos One. 2010; 5: e10840. doi: 10.1371/journal.pone.0010840 PMID: 20520776 16. Slobbe L, Doorduijn JK, Lugtenburg PJ, el Barzouhi A, Boersma E, van Leeuwen WB, et al. Prevention of Catheter-Related Bacteremia with a Daily Ethanol Lock in Patients with Tunneled Catheters: A Ran- domized, Placebo-Controlled Trial. Plos One. 2010; 5: e10840. doi: 10.1371/journal.pone.0010840 PMID: 20520776 17. Broom JK, Krishnasamy R, Hawley CM, Playford EG, Johnson DW. A randomised controlled trial of heparin versus ethanol lock therapy for the prevention of catheter associated infection in haemodialysis patients—the healthy-cath trial. BMC Nephrol. 2012; 13: 146–154. doi: 10.1186/1471-2369-13-146 PMID: 23121768 18. Perez-Granda MJ, Barrio JM, Munoz P, Hortal J, Rincon C, Rabadan PM, et al. Ethanol Lock Therapy (E-Lock) in the prevention of catheter-related bloodstream infections (CR-BSI) after major heart surgery (MHS): a randomized clinical trial. PloS One. 2014; 9: e918378. 19. References Souweine B, Lautrette A, Gruson D, Canet E, Klouche K, Argaud L, et al. Ethanol Lock and Risk of Hemodialysis Catheter Infection in Critically Ill Patients: A Randomized Controlled Trial. Am J Respir Crit Care Med. 2015; 191:1024–32. doi: 10.1164/rccm.201408-1431OC PMID: 25668557 20. Qu Y, Istivan TS, Daley AJ, Rouch DA, Deighton MA. Comparison of various antimicrobial agents as catheter lock solutions: preference for ethanol in eradication of coagulase-negative staphylococcal bio- films. J Med Microbio. 2009; 58: 442–450. 21. Rane HS, Bernardo SM, Walraven CJ, Lee SA. In Vitro analyses of ethanol activity against Candida albicans biofilms. Antimicrob Agents Chemother. 2012; 56: 4487–4489. doi: 10.1128/AAC.00263-12 PMID: 22615286 22. Peters BM, Ward RM, Rane HS, Lee SA, Noverr MC. Efficacy of ethanol against Candida albicans and Staphylococcus aureus polymicrobial biofilms. Antimicrob Agents Chemother. 2013; 57(1): 74–82. doi: 10.1128/AAC.01599-12 PMID: 23070170 23. Lesens O, Balestrino D, Charbonnel N, Aumeran C, Traoré O, Forestier C, et al. Effectiveness of etha- nol-based lock solutions on catheter biofilm microorganisms. Presented at the ICAAC Congress 2013 p 952. 24. Guenu S, Heng AE, Charbonne F, Galmier MJ, Charlès F, Deteix P, et al. Mass spectrometry and scan- ning electron microscopy study of silicone tunnelled dialysis catheter integrity after an exposure of 15 days to 60% ethanol solution. Rapid Commun Mass Spectrom. 2007; 21: 229–236. PMID: 17171773 25. Msakni N, Galmier MJ, Couret MJ, Szczepaniak C, Bouchon B, Souweine B, et al. Complementary mass spectrometric approaches and scanning electron microscopy for the study of structural stability of polyurethane tunneled dialysis catheters after exposure to ethanol solutions. Rapid Commun Mass Spectrom. 2013; 27: 2343–2354. doi: 10.1002/rcm.6691 PMID: 24097390 26. Volpi N. Purification of heparin, dermatan sulfate and chondroitin sulfate from mixtures by sequential precipitation with various organic solvents. J Chromatogr B. 1996; 685: 27–34. 27. Davenport A. The rationale for the use of low molecular weight heparin for haemodialysis treatments. Hemodial Int. 2013 Oct; 17 Suppl 1:S28–32. doi: 10.1111/hdi.12086 PMID: 24134328 28. Speeckaert MM, Devreese KM, Vanholder RC, Dhondt A. Fondaparinux as an alternative to vitamin K antagonists in haemodialysis patients. Nephrol Dial Transplant. 2013; 28: 3090–3095. doi: 10.1093/ ndt/gft293 PMID: 24021678 29. Canaud B, My H, Morena M, Lamy-Lacavalerie B, Leray-Moragues H, Bosc JY, et al. a new vascular access device for extracorporeal renal replacement therapy. Preliminary clinical results. Nephrol Dial Transplant. 1999; 14: 692–698. PMID: 10193821 30. Jaulmes P, Mestres R. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 References Dosage des alcools par oxidation (Determination of alcohols by oxidation). Chim Analytique. 1958; 40: 413. 31. Guidance for Industry Q1A (R2) Stability Testing of New Drug Substances and Products. U.S. Depart- ment of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research (CDER), Center for Biologics Evaluation and Research (CBER). November 2003; ICH Revi- sion 2. 32. Ghigo JM. Natural conjugative plasmids induce bacterial biofilm development. Nature 2001; 412:442– 445. PMID: 11473319 33. Maya ID, Smith T, Allon M. Does the heparin lock concentration affect hemodialysis catheter patency? Clin J Am Soc Nephrol. 2010; 5: 1458–1462. doi: 10.2215/CJN.01230210 PMID: 20498241 34. Moran JE, Ash SR. For the ASDIN Clinical Practice Committee: Locking solutions for hemodialysis catheters; heparin and citrate: A position paper by ASDIN. Semin Dial. 2008; 21: 490–492. doi: 10. 1111/j.1525-139X.2008.00466.x PMID: 18764795 35. Snaterse M, Ruger W, Scholte Op Reimer WJ, Lucas C. Antibiotic-based catheter lock solutions for pre- vention of catheter-related bloodstream infection: A systematic review of randomised controlled trials. J Hosp Infect. 2010; 75: 1–11. doi: 10.1016/j.jhin.2009.12.017 PMID: 20227787 20 / 21 PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 Stability, Effects on Biofilm and Catheter of Anticoagulants in Ethanol 36. Balestrino D, Souweine B, Charbonnel N, Lautrette A, Aumeran C, Traoré O, et al. Eradication of micro- organisms embedded in biofilm by an ethanol-based catheter lock solution. Nephrol Dial Transplant. 2009; 24: 3204–3209. doi: 10.1093/ndt/gfp187 PMID: 19383833 37. Crnich CJ, Halfmann JA, Crone WC, Maki DG. The effects of prolonged ethanol exposure on the mechanical properties of polyurethane and silicone catheters used for intravascular access. Infect Con- trol Hosp Epidemiol. 2005; 26: 708–714. PMID: 16156328 38. Öncü S. Optimal dosage and dwell time of ethanol lock therapy on catheters infected with Candida spe- cies. Clin Nutr. 2014; 33: 360–362. doi: 10.1016/j.clnu.2013.04.014 PMID: 23672805 39. Mermel LA, Alang N. Adverse effects associated with ethanol catheter lock solutions: a systematic review. J Antimicrob Chemother. 2014; 69: 2611–2619. doi: 10.1093/jac/dku182 PMID: 24891431 40. Kanwar YS, Farquhar MG. Isolation of glycosaminoglycans (heparin sulfate) from glomerular basement membranes. Proc Natl Acad Sci USA. 1979; 76: 4493–4497. PMID: 159457 41. Thornton DJ, Hunt S, Huckerby TN. The glycosaminoglycans of pig colonic wall connective tissue. Bio- chim Biophys Acta. 1983; 757: 219–225. PMID: 6849974 42. Guyatt GH, Norris SL, Schulman S, Hirsh J, Eckman M H, Akl EA, et al. PLOS ONE \| DOI:10.1371/journal.pone.0159475 July 21, 2016 References Methodology for the Develop- ment of Antithrombotic Therapy and Prevention of Thrombosis Guidelines: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012; 141: 53S–70S. doi: 10.1378/chest.11-2288 PMID: 22315256 43. Polkinghorne KR, McMahon LP, Becker GJ. Pharmacokinetic studies of dalteparin (Fragmin), enoxa- parin (Clexane), and danaparoid sodium (Orgaran) in stable chronic hemodialysis patients. Am J Kid- ney Dis. 2002; 40: 990–995. PMID: 12407644 44. Ho G, Leblanc K, Selby R, Richardson R, Hladunewich M, Battistella M. Use of fondaparinux for circuit patency in hemodialysis patients. Am J Kidney Dis. 2013; 61: 525–526. doi: 10.1053/j.ajkd.2012.09. 015 PMID: 23157936 45. Héparines de basse masse moléculaire. European Pharmacopeia 5.0. 2005:1862–1865. 46. Olsson JV, Hult D, Cai Y, Garcia-Gallego S, Malkoch M. Reactive imidazole intermediates: simplified synthetic approach to functional aliphatic cyclic carbonates. Polymer Chem. 2014; 5: 6651–6655. 47. Safety Data Sheets: Properties, Hazards and Safety Information for H12MDI. Available: www.alipa.org/ uploads/H12MDI.pdf. 2013. Accessed 8 April 2015. 48. Schilcher G, Schlagenhauf A, Schneditz D, Scharnagl H, Ribitsch W, Krause R, et al. Ethanol causes protein precipitation-new safety issues for catheter locking techniques. PloS One. 2013; 8: e84869. doi: 10.1371/journal.pone.0084869 PMID: 24391979 49. Polaschegg HD. Loss of catheter locking solution caused by fluid density. ASAIO J. 2005; 51:230–5. PMID: 15968952 50. Calvet L, Piot M, Lartigue C, Souweine B, Tardy-Poncet B. Anticoagulant properties of enoxaparin 400 IU/mL–40% ethanol catheter lock solution. Springer Plus. 2015; 4:746. doi: 10.1186/s40064-015- 1533-2 PMID: 26693105 51. Restrepo D, Laconi NS, Alcantar NA, West LA, Buttice AL, Patel S, et al. Inhibition of heparin precipita- tion, bacterial growth, and fungal growth with a combined isopropanol-ethanol locking solution for vas- cular access devices. J Pediatr Surg. 2015; 50: 472–477. doi: 10.1016/j.jpedsurg.2014.07.003 PMID: 25746710 52. Krishnasami Z, Carlton D, Bimbo L, Taylor ME, Balkovetz DF, Barker J, et al. Management of hemodial- ysis catheter-related bacteremia with an adjunctive antibiotic lock solution. Kidney Int. 2002; 61:1136– 1142. PMID: 11849468 21 / 21
https://openalex.org/W2127155215	https://www.scielo.br/j/csp/a/36vS8h6y6PXP4Y7nxtxJ5PJ/?lang=en&format=pdf	English	null	Likely transmission of hepatitis C virus through sharing of cutting and perforating instruments in blood donors in the State of Pará, Northern Brazil	Cadernos de Saúde Pública	2,010	cc-by	5,030	1 Instituto de Ciências Biológicas, Universidade Federal do Pará, Belém, Brasil. 2 Fundação Centro de Hematologia e Hemoterapia do Pará, Belém, Brasil. 3 Secretaria de Estado de Saúde Pública do Pará, Belém, Brasil. 4 Núcleo de Medicina Tropical, Universidade Federal do Pará, Belém, Brasil. 837 ARTIGO ARTICLE 837 ARTIGO ARTICLE 837 ARTIGO ARTICLE Correspondence J. A. R. Lemos Instituto de Ciências Biológicas, Universidade Federal do Pará. Av. Augusto Correa 1, Belém, PA 66075-110, Brasil. lemos@ufpa.br Likely transmission of hepatitis C virus through sharing of cutting and perforating instruments in blood donors in the State of Pará, Northern Brazil Possível transmissão do vírus da hepatite C por compartilhamento de materiais cortantes e perfurantes em doadores de sangue no Estado do Pará, Norte do Brasil Aldemir Branco de Oliveira-Filho 1 Adriana do Socorro Coelho Pimenta 2 Márcia de Fátima Maciel Rojas 2 Márcia Cristina Munhoz Chagas 2 Deborah Maia Crespo 3 José Ângelo Barletta Crescente 4 José Alexandre Rodrigues de Lemos 1 Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 Aldemir Branco de Oliveira-Filho 1 Adriana do Socorro Coelho Pimenta 2 Márcia de Fátima Maciel Rojas 2 Márcia Cristina Munhoz Chagas 2 Deborah Maia Crespo 3 José Ângelo Barletta Crescente 4 José Alexandre Rodrigues de Lemos 1 Introduction We determined the risk factors for HCV infection in blood donors in the State of Pará, Northern Brazil. We examined 256 blood donors seen at the Blood Bank of Pará State between 2004 and 2006. They were divided into two groups, depending on whether they were infected with HCV or not; 116 donors were infected with HCV, while the other 140 were free of infection. The HCV-RNA was de- tected by real-time PCR. All of the participants filled out a questionnaire about possible risk fac- tors. The data were evaluated using simple and multiple logistic regressions. The main risk fac- tors for HCV were found to be use of needles and syringes sterilized at home (OR = 4.55), invasive dental treatment (OR = 3.08), shared use of razors at home (OR = 1.99), sharing of disposable razors in barbershops, beauty salons, etc. (OR = 2.34), and sharing manicure and pedicure material (OR = 3.45). Local and regional health authorities should educate the public about sharing perfo- rating and cutting materials at home, in barber/ beauty shops, and in dental clinics as risk factors for HCV infection. Hepatitis C virus (HCV) is a major cause of acute and chronic hepatitis, cirrhosis, and hepatocel- lular carcinoma. It is estimated that 2% of the world population (approximately 123 million people) is infected with HCV; most cases are as- ymptomatic and the victims are unaware that they have a viral infection 1,2. HCV infection is a pandemic. Seroprevalence for HCV varies from less than 2% in some countries, such as Germa- ny, India, France, and the United States, to high rates in countries and regions in Africa, Asia, and Europe, such as the Chinese province Hu- bei (30.13%), Mongolia (31.86%), and in Egypt (28%) 1,3. In South America, Brazil is estimated to have the highest HCV seroprevalence (1.6%), which varies across different geographical re- gions 4. Northern Brazil has the highest HCV se- roprevalence (2.12%), with a high frequency of genotype 1 4,5,6. The various genotypes of HCV differ in their susceptibility to antiviral therapy, and evidence suggests a variation in clinical con- sequences, with patients with genotype 1 having the poorest prognosis 7,8. Epidemiological stud- ies in the population of Brazilian blood donors have found similar results, with reported preva- lences of anti-HCV from 0.38% to 2.6% and sig- nificant frequencies of genotype 1 6,9,10,11. Correspondence J. A. R. Risk factors In northern Brazil, there have been very few epidemiological studies of HCV infection 5. In the State of Pará, the prevalence of anti-HCV rang- es from 0.22% to 3.6%, and it is predominantly found in individuals over 50 years old 21,22. Based on the scarcity of epidemiological information, we decided to determine the risk factors associ- ated with transmission of HCV infection in blood donors in the State of Pará. The participants completed a questionnaire about possible risk factors for infection with HCV. This questionnaire included questions about age, use of home-sterilized needles and syringes, use of illicit drugs, blood transfusion before or after 1993, surgery before or after 1993, use of con- doms during sexual intercourse, more than five sex partners during the previous year, invasive dental treatment (root canal and surgical tartar removal), shared use of razors at home, shared use of disposable razors in public places (beauty salons, barber shops, etc.), and shared use of in- struments for manicures and pedicures. Screening for HCV infection Plasma samples from all of the blood donors were screened for anti-HCV antibodies, using the Murex anti-HCV version 4.0 (Murex Biotech S.A., Kyalami, South Africa). Confirmation tests were done using real-time PCR (ABI Prism 7000, Applied Biosystems, Foster City, U.S.A.). The viral RNA was extracted using the QIAmp Viral RNA Mini Kit (Qiagen). Molecular diagnosis was ac- complished by detecting the nucleotide fragment (67 bp) of 5’ UTR with the TaqMan EZ RT-PCR Core Reagents kit (Applied Biosystems), accord- ing to the manufacturer’s instructions, adding the primers (5’ﬁ3’: CGCTCAATGCCTGGAGATTT and TTTCGCGACCCAACACTACTC) and the probe (5’ﬁ3’:FAM-TGCCCCCGCAAGACTGCTAGC- TAMRA). The amplification conditions were: 1 cycle: 50oC/2 min, 60oC/30 min, and 95oC/5 min; 50 cycles: 94oC/20 s and 60oC/1 min. Nucle- ic acid tests (NAT) directly detect the presence of HCV RNA, using a combination of amplifica- tion and detection techniques. In general, NAT are quite sensitive and specific. HCV RNA can be detected one to three weeks after infection, ap- proximately one month before the appearance of antibodies 24,25. After HCV screening tests were added to the routine protocols of blood centers, and blood donors were screened clinically and epidemio- logically based on knowledge of HCV transmis- sion, the rate of post-transfusion hepatitis was significantly reduced 1,13,14. However, some cases of post-transfusion hepatitis C continue to be re- ported, because of the 4-20-week immunological window of HCV 20. Currently, the groups at risk of HCV infection are associated with frequent or in- adequate parenteral procedures, such as: sharing syringes among drug users, sharing hemodialysis tubing and filters among patients with chronic kidney disease, multiple blood/hemoderivative transfusion in patients with chronic blood dis- eases, etc. Even with current knowledge of the various forms of HCV transmission, about 30% of cases are of unknown origin 13,14. Introduction Lemos Instituto de Ciências Biológicas, Universidade Federal do Pará. Av. Augusto Correa 1, Belém, PA 66075-110, Brasil. lemos@ufpa.br Hepacivirus; Hepatitis C; Blood Donors Hepacivirus; Hepatitis C; Blood Donors HCV is primarily transmitted via parenter- al routes. Blood transfusions without previous testing for HCV, and the reuse and the sharing Oliveira-Filho AB et al. 838 of gloves, syringes, and other material that can become contaminated with blood in hospitals, clinics, emergency rooms, and doctors’ offices are the most commonly documented routes of parenteral transmission of HCV 12,13,14. There are also other forms of transmission associat- ed with parenteral procedures when contami- nated equipment is used, such as intrafamilial transmission through sharing of razor blades or other perforating/cutting instruments 12,15,16,17. Transmission of HCV through sexual and peri- natal routes has been reported, although both are relatively inefficient, except when HCV is as- sociated with other microorganisms such as HIV and HBV 18,19. ian law 23, all blood donations obtained at the HEMOPA were voluntary, anonymous, altruistic, and unremunerated, either directly or indirectly. Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 Statistical analysis This case-control epidemiological study was composed of HCV-infected and uninfected voluntary blood donors seen at the Fundação Centro de Hematologia e Hemoterapia do Pará (HEMOPA), northern Brazil. The blood donors were selected during routine blood collection at nine hematology and hemotherapy units and in blood-donation campaigns held in public and private institutions in Pará from January 1, 2004- December 31, 2006 22. As is mandated by Brazil- Each possible risk factor for HCV infection (as diagnosed by molecular analysis) was analyzed separately, by simple logistic regression. All the potential risk factors with probabilities of p ≤ 0.2 were examined and included in the final model of HCV transmission using backward stepwise multiple logistic regression. A multiple logistic regression was then run to determine the asso- Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 LIKELY TRANSMISSION OF HCV INFECTION IN NORTHERN BRAZIL 839 ciation of each risk factor with HCV infection, as diagnosed by molecular analysis. Various possi- ble types of interactions were evaluated in order to determine how they might improve the final model 26. The fit of the final model was assessed using the Hosmer-Lemeshow goodness-of-fit test 27. Finally, the risk factors listed by multiple logistic regression were quantified and evaluated by the chi-square test, with expected equal pro- portions as the criterion to distinguish between infected and uninfected groups. All statistical analyses were carried out in duplicate with the programs BioEstat version 5.0 (Sociedade Civil Mamirauá, Manaus, Brazil) 28 and PASW Statistic version 18.0 (SPSS Inc., Chicago, U.S.A.); except that the Hosmer-Lemeshow test was performed only by PASW Statistic. was 30 years in the uninfected blood donors. Blood donors over 35 years of age were sig- nificantly more likely to be infected than those under 35 (Table 1); this information should be considered in the construction of an epidemio- logical model for HCV transmission. Other factors associated with the molecular diagnosis of infection were “use of needles and/ or syringes at home”, “blood transfusion”, “inva- sive dental treatment (root canal or surgical tar- tar removal)”, “sharing of razors at home”, “us- ing disposable blades in public places (beauty salons, barber shops, or similar establishments)”, “shared use of manicure and pedicure instru- ments”, and “use of illicit drugs (injectable or non-injectable)” (Table 1). However, it was not possible to determine precisely which factors sig- nificantly contributed to HCV transmission, by means of univariate analysis (Table 2). Ethics This study was approved by the Research Eth- ics Committee of the Tropical Medicine Center at the Federal University of Pará, Brazil. All the blood donors were informed about the objectives of the research and gave their written informed consent to participate in this study. We then used multiple logistic regression to determine which risk factors significantly con- tributed to HCV transmission (Table 3). The im- portance (odds ratio) of these risk factors was 4.55 for “use of needles and/or syringes at home”, 3.08 for “invasive dental treatment (root canal or sur- gical tartar removal)”, 1.99 for “sharing of razors at home”, 2.34 for “shared use of disposable blades in public places (beauty salons, barber shops, or similar establishments)”, and 3.45 for “shared use of manicure and pedicure instruments” (Ta- ble 3). The Hosmer-Lemeshow goodness-of-fit test showed a good fit for the final model (χ2 = 2,734, p = 0.950). We see in Table 4 that 55.17% of the blood donors infected with HCV had 3-5 risk factors, whereas 87.15% of the uninfected blood donors had 0-2 risk factors, based on mul- tiple logistic regression. Most of the infected do- nors had three or more risk factors (χ2 = 69.246; p < 0.001). The epidemiological model for HCV transmission in blood donors in Pará included five parenteral risk factors, with sharing perfo- rating/cutting materials in the home and inva- sive dental treatment being the most important. Results From 2004-2006, the HEMOPA attended 242,726 voluntary blood donors in the State of Pará. HCV RNA was detected in 304 of these donors (anti- HCV antibody-positive or indeterminate). All of the 304 infected donors were invited to partici- pate in this study, and 116 agreed to participate 22. In addition, we invited 304 uninfected blood donors (anti-HCV antibody-negative and HCV RNA-negative), randomly selected from 241,614 uninfected donors, to participate as a control group 22; 140 of them visited the HEMOPA to complete the epidemiological questionnaire. In this way, we selected a total of 256 blood donors in Pará to participate in the research. These blood donors were divided into two groups, based on detection of anti-HCV anti- bodies and HCV RNA. The numerical relation- ship between the case and control groups was approximately 1:1 (116:140; no statistical differ- ence in the sample size; χ2 = 2.25; p = 0.134). The case group consisted of 116 blood donors infected with HCV (anti-HCV antibody-positive or indeterminate, and HCV RNA-positive). The control group consisted of 140 uninfected blood donors (anti-HCV antibody-negative and HCV RNA-negative). Age appeared to be a risk fac- tor for infection with HCV. In the infected donor group, the median age was 40 years, whereas it Statistical analysis We found that 85.39% of the blood donors infected by HCV had 1-5 risk factors, and 87.15% of the uninfected blood donors had 0-3 risk factors. Discussion In this study, the final model of HCV transmission based on multiple logistic regression showed that the use of home-sterilized needles and glass sy- ringes, invasive dental treatments, shared use of razors at home, shared use of disposable razors in beauty salons, barber shops, and similar es- tablishments, and shared use of manicure and pedicure material should be taken into account In this study, the final model of HCV transmission based on multiple logistic regression showed that Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 Oliveira-Filho AB et al. 840 840 Oliveira-Filho AB et al. Table 1 Initial discrimination of risk factors for HCV infection using simple logistic regression (molecular diagnosis versus risk factor). Discussion Risk factors for HCV infection n HCV RNA (%) Odds ratio 95%CI Gender Male 180 85 (47) 1.30 0.75-2.24 Female 76 39 (51) 1.00 Marital status Single 98 38 (39) 1.00 0.99-2.76 Married 158 79 (50) 1.65 * Age (years) > 35 132 77 (76) 3.05 1.83-5.10 ≤ 35 124 39 (31) 1.00 Use of needles and/or syringes sterilized at home Yes 70 53 (72) 6.09 3.26-11.37 No 186 63 (33) 1.00 Recipient of a blood transfusion Yes 49 30 (61) 2.22 1.17-4.20 No 207 86 (42) 1.00 Recipient of a blood transfusion prior to 1993 * Yes 20 13 (65) 1.18 0.35-3.94 No 29 17 (59) 1.00 Had surgery Yes 117 58 (50) 1.37 0.84-2.25 No 139 58 (48) 1.00 Had surgery prior to 1993 # Yes 53 31 (58) 1,69 0.79-3.65 No 64 27 (42) 1.00 Has a tattoo Yes 37 22 (59) 1.95 * 0.96-3.96 No 219 94 (43) 1.00 Use of condoms during the sexual act No 157 77 (49) 1.48 * 0.89-2.47 Yes 99 39 (39) 1.00 More than five sexual partners during the previous year Yes 50 27 (54) 1.54 * 0.83-2.87 No 206 89 (43) 1.00 Invasive dental treatment (canal work and tartar surgery) Yes 126 70 (56) 2.28 1.38-3.77 No 130 46 (35) 1.00 Sharing razors at home Yes 94 56 (60) 2.51 1.49-4.22 No 162 60 (37) 1.00 Shared use of disposable blades in public places (beauty salons, barber shops and similar) Yes 162 59 (36) 1.00 1.59-4.54 No 94 57 (61) 2.67 Use of shared instruments for manicure and pedicure Yes 134 41 (31) 1.00 2.16-6.07 No 122 75 (61) 3.62 Use of illicit drugs (injectable or not) Yes 28 18 (64) 2.39 ** 1.06-5.40 No 228 98 (43) 1.00 * p < 0.05; ** p ≤0.2; Initial discrimination of risk factors for HCV infection using simple logistic regression (molecular diagnosis versus risk factor). Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 LIKELY TRANSMISSION OF HCV INFECTION IN NORTHERN BRAZIL 84 841 through sharing of inadequately sterilized sy- ringes and needles in homes with individuals who were asymptomatic and unaware that they were infected. Administering injectable medica- tion without adequately sterilizing syringes or needles has been the main cause of HCV trans- mission worldwide, especially in developing countries 2,15,33. Discussion Probably the risk factor “use of home-sterilized syringes and needles” was partly responsible for the significantly higher infection rate in the group older than 35 years. This may also indicate that longer exposure to risk factors increases the probability of infection with HCV. Table 2 Table 2 Division of blood donors into groups according to the number of signiﬁ cant risk factors for HCV infection, based on univariate analysis. Number of risk factors Infected Uninfected n % n % 0 1 0.86 14 10.00 1 7 6.03 43 30.72 2 20 17.24 38 27.14 3 22 18.97 27 19.29 4 29 25.00 12 8.57 5 21 18.11 5 3.57 6 11 9.48 1 0.71 7 4 3.45 0 0.00 8 1 0.86 0 0.00 The use of illicit injectable or non-injectable drugs is currently considered as the most impor- tant risk factor for HCV 1,2,32. However, we did not find it to be a significant factor in our study. Probably this was because of the clinical/epide- miological preselection of blood donors by the HEMOPA, which considers the use of illicit drugs as an exclusion factor for blood donation. How- ever, we still detected 28 blood donors who were illicit drug users; they had withheld information during the selection process in order to obtain free blood tests. by local and regional health authorities and by those of other countries with cultures similar to that of northern Brazil, in order to inform politi- cal and public strategies to control HCV trans- mission. The detection and treatment of HCV-in- fected patients are essential public-health mea- sures for containing viral transmission. However, public awareness about the risks of specific ac- tions also significantly contributes to reducing prevalence and to the prevention of new infec- tions 29,30. The risk factors that we examined are parenteral procedures that can be avoided if the population is aware of the risks of transmitting microorganisms. Shaving or trimming nails can generate trauma or microtrauma on the skin surface, resulting in exposure to HCV on blades or nail scissors contaminated by asymptomatic infected family members or by other people who are unknowingly infected and who frequent bar- bershops and beauty salons. In Italy, Pakistan, and Nigeria, HCV transmission by sharing razor blades and other cutting instruments, as well as other forms of viral dissemination due to a lack of public awareness, have been reported as risk factors for infection 12,31,32. Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 Discussion We found invasive dental treatment (root ca- nals and surgical removal of tartar) to be a risk factor for HCV infection. Various studies have detected HCV RNA in the saliva of HCV-infected patients 34,35. Consequently, HCV in the saliva, combined with inadequate sterilization of od- ontological instruments, could be a means of HCV transmission. Nosocomial transmission could be investigated in greater detail through phylogenetic analysis of the viral strains circu- lating among those patients in our study who apparently had been infected during dental treatment. In this study, the high odds ratio (3.08) and the narrow 95% confidence interval (1.68-5.65) indicate the high probability of this type of event. Various studies have recognized blood trans- fusion as a significant risk factor for HCV infec- tion 13,30. However, this was not found in our study. The univariate analysis indicated a sig- nificant risk from blood transfusion. However, it was not a significant factor in the model con- structed based on multiple regression analysis. Probably this was due to the influence of other factors that increased the apparent significance of receiving blood transfusions. Transmission of HCV infection by blood transfusion was signifi- cantly reduced by the introduction of HCV tests for screening blood donors 14, which is the case in the Pará state blood bank. The rate of post-trans- fusion hepatitis C infection remains very high in countries in which HCV screening is deficient or nonexistent 13,14,36. The final HCV transmission model indicates that the use of home-sterilized needles and glass syringes is the greatest risk factor for infection among blood donors in Pará. In Brazil, dispos- able perforating and cutting materials for health procedures began to be used on a large scale during the second half of the 1980s. This situ- ation, together with a lack of knowledge about HCV transmission, likely accounts for infection being more common among blood donators over 35, since transmission could have occurred Oliveira-Filho AB et al. 842 Oliveira-Filho AB et al. Table 3 Risk factors for HCV infection in blood donors in the State of Pará based on multiple logistic regression. Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 Resumo Nós determinamos os fatores de risco à infecção pelo HCV em doadores de sangue no Estado do Pará, Brasil. Foram analisados 256 doadores de sangue atendidos na Fundação HEMOPA de 2004 a 2006, sendo divi- didos em dois grupos: infectados e não-infectados. O diagnóstico foi realizado por PCR em tempo real. To- dos os participantes responderam a questionário sobre possíveis fatores de risco, sendo a modelagem estatísti- ca feita por regressão logística simples e múltipla. Os fatores de risco à infecção foram: uso de agulhas e se- ringas de vidros esterilizadas em casa (OR = 4,55), re- alização de tratamento dentário invasivo (OR = 3,08), compartilhamento de lâminas em domicílio (OR = 1,99), compartilhamento de lâminas descartáveis em barbearias, salões de beleza (OR = 2,34), e comparti- lhamento de material de manicure e pedicure (OR = 3,45). As autoridades de saúde devem conscientizar a população sobre o compartilhamento de materiais perfuro-cortantes em domicílio, salões de beleza e con- sultórios dentários como fatores de risco à infecção. J. A. R. Lemos contributed to the study design, data interpretation, and the corrections to the manuscript. A. B. Oliveira-Filho contributed to the study design, molecular data acquisition, data analysis and interpre- tation, and manuscript preparation. A. S. C. Pimenta contributed to the study design, molecular data ac- quisition and manuscript preparation. M. F. M. Rojas, M. C. M. Chagas, D. M. Crespo and J. Â. B. Crescente were responsible for data collection, data interpreta- tion and corrections to the manuscript. Acknowledgments We wish to thank Dr. M. Sidia Callegari-Jacques (Bios- ciences Institute, Federal University of Rio Grande do Sul, Brazil) and Dr. Manuel Ayres (Institute of Biological Sciences, Federal University of Pará, Brazil) for support and suggestions concerning the statistical analyses of this study. This work was supported by grants from the Progra- ma Nacional de DST/AIDS/Hepatites (CSV298/2007), Ministry of Health, co-supported by former Secretaria de Estado de Ciência, Tecnologia e Meio-Ambiente do Pará (current Secretaria de Estado de Desenvolvimento, Ciência e Tecnologia do Pará), and scholarships provi- ded by CNPq, Brazil. Hepacivirus; Hepatite C; Doadores de Sangue Discussion Risk factors for HCV infection n HCV RNA (%) Odds ratio * 95%CI Use of needles and syringes sterilized at home Yes 70 53 (72) 4.55 2.28-9.10 No 186 63 (33) 1.00 Invasive dental treatment (canal work and tartar removal) Yes 126 70 (56) 3.08 1.68-5.65 No 130 46 (35) 1.00 Shared use of razors at home Yes 94 56 (60) 1.99 1.08-3.66 No 162 60 (37) 1.00 Shared use of disposable razors in a public place (beauty salons and barber shops) Yes 162 59 (36) 1.00 1.26-4.32 No 94 57 (61) 2.34 Shared use of manicure and pedicure instruments Yes 134 41 (31) 1.00 1.88-6.33 No 122 75 (61) 3.45 * p < 0.05. Risk factors for HCV infection in blood donors in the State of Pará based on multiple logistic regression. We propose here an HCV transmission mod- el for blood donors in Pará, based on the use of home-sterilized needles and/or syringes, inva- sive dental treatment (root canal or surgical tar- tar removal), sharing of razors at home, shared use of disposable razors in public places (beauty salons, barber shops, or similar establishments), and shared use of manicure and pedicure instru- ments. This information should help to inform political and public strategies of local and region- al heath authorities to help control the transmis- sion of HCV infection in the state of Pará, through public awareness of the risk of HCV infection due to sharing perforating and cutting instruments at home, in beauty salons and barber shops, and in dental clinics. Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 LIKELY TRANSMISSION OF HCV INFECTION IN NORTHERN BRAZIL 843 Hepacivirus; Hepatite C; Doadores de Sangue Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 References 1. Shepard CW, Finelli L, Alter MJ. Global epidemiol- ogy of hepatitis C virus infection. Lancet Infect Dis 2005; 5:558-67. 7. Hadziyannis SJ, Koshina, JS. Differences in epidemi- ology, liver disease and treatment response among HCV genotypes. Hepatol Res 2004; 29:129-35. 2. Alter MJ. Epidemiology of hepatitis C virus infec- tion. World J Gastroenterol 2007; 13:2436-41. 8. Zeuzem S. Heterogeneous virologic response rates to interferon-based therapy in patients with chronic hepatitis C: who responds less well. Ann Intern Med 2004; 140:370-81. 3. Lo Re 3rd V, Kostman JR. Management of chronic hepatitis C. Postgrad Med J 2005; 81:376-82. 9. Gonçalves FL, Boccato RSBS, Pedro RJ, Papaiorda- nou PMO, Souza CA, Gonçalves NSL, et al. Preva- lence of HBsAg, anti-HBc and anti-HCV in blood donors of “Hemocentro-Campinas”. Rev Inst Med Trop São Paulo 1993; 35:45-51. 4. Grupo de Estudo da Sociedade Brasileira de He- patologia. Epidemiologia da infecção pelo vírus da hepatite C no Brasil. Gastroenterol End Digest 1999; 18:S3-8. 5. Da Fonseca JCF, Brasil LM. Hepatitis C virus infec- tion in the Amazon Brazilian region. Rev Soc Bras Med Trop 2004; 37:1-8. 10. Narciso-Schiavon JL, Schiavon LL, Carvalho-Filho RJ, Freire FCF, Cardoso JR, Bordin JO, et al. Anti- hepatitis C virus-positive blood donors: are wom- en any different? Transfus Med 2008; 18:175-83. 6. Campiotto S, Pinho JR, Carrilho FJ, Da Silva LC, Souto FJ, Spinelli V, et al. Geographic distribution of hepatitis C virus genotypes C in Brazil. Braz J Med Biol Res 2005; 38:41-9. Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 Oliveira-Filho AB et al. 844 Oliveira-Filho AB et al. 11. Nascimento MC, Mayaud P, Sabino EC, Torres KL, Franceschi S. Prevalence of hepatitis B and C sero- logical markers among first-time blood donors in Brazil: a multi-center serosurvey. J Med Virol 2008; 80:53-7. 24. Scott JD, Gretch DR. Molecular diagnostics of hep- atitis C virus infection. A systematic review. JAMA 2007; 297:724-32. 25. Garcia FB, Gomide GPM, Pereira GA, Moraes- Souza H. Importance of screening and confirma- tory tests to detect blood donors infected by the hepatitis C virus. Rev Bras Hematol Hemoter 2008; 30:218-22. 12. Bari A, Akhtar S, Rahbar MH, Luby SP. Risk fac- tors for hepatitis C virus infection in male adults in Rawalpindi–Islamabad, Pakistan. Trop Med Int Health 2001; 6:732-8. 26. Bruzzi P, Green SB, Byar DP, Brinton LA, Schairer C. References Estimating the population attributable risk for multiple risk factors using case–control data. Am J Epidemiol 1985; 122:904-14. 13. Memon MI, Memon MA. Hepatitis C: an epidemi- ological review. J Viral Hepat 2002; 9:84-100. 14. Prati, D. Transmission of hepatitis C virus by blood transfusions and other medical procedures: a global review. J Hepatol 2006; 45:607-16. 27. Hosmer DW, Lemeshow S. Applied logistic regres- sion. 2nd Ed. New York: John Wiley & Sons; 2000. 15. Kane M. Unsafe injections. Bull World Health Or- gan 1998; 76:99-100. 28. Ayres M, Ayres Jr. M, Ayres DL, Santos ASS. BioEstat 5.0: aplicações estatísticas nas áreas das ciências biológicas e médicas [CD-ROM]. Belém: Sociedade Civil Mamirauá/Brasília: Conselho Nacional de Desenvolvimento Científico e Tecnológico; 2005. 16. Marx MA, Murugavel KG, Sivaram S, Balakrishnan P, Steinhoff M, Anand S, et al. The association of health care use and hepatitis C virus infection in a random sample of urban slum community resi- dents in southern India. Am J Trop Med Hyg 2003; 68:258-62. 29. Lavanchy D. The global burden of hepatitis C. Liver Int 2009; 29:S74-81. 30. Wasley A, Alter MA. Epidemiology of hepatitis C: geographic differences and temporal trends. Semin Liver Dis 2001; 20:1-16. 17. Hauri AM, Armstrong GL, Hutin YJF. The global burden of disease attributable to contaminated injections given in health care settings. Int J STD AIDS 2004; 15:7-16. 31. Mele A, Corona R, Tosti ME, Palumbo F, Moiraghi A, Novaco F, et al. Beauty treatments and risk of par- enterally transmitted hepatitis. Results from the hepatitis surveillance system in Italy. Scand J In- fect Dis 1995; 27:441-4. 18. Conte D, Fraquelli M, Prati D, Colucci A, Minola E. Prevalence and clinical course of chronic hepatitis C virus (HCV) infection and rate of HCV vertical transmission in a cohort of 15.250 pregnant wom- en. Hepatology 2000; 31:751-5. 32. Koate BBD, Buseri FI, Jeremiah ZA. Seroprevalence of hepatitis C virus among blood donors in Rivers State, Nigeria. Transfus Med 2005; 15:449-51. 19. Gibb DM, Goodall RL, Dunn DT, Healy M, Neave P, Cafferkey M, et al. Mother-to-child transmission of hepatitis C virus: evidence for preventable peri- partum transmission. Lancet 2000; 356:904-7. 33. Reeler AV. Injections: a fatal attraction. Soc Sci Med 1990; 31:1119-25. 34. Arrieta JJ, Rodriguez-Inigo E, Casqueiro M, Barto- lomé J, Manzarbeitia F, Herrero M, et al. Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010 References Detection of hepatitis C virus replication by in situ hybridiza- tion in epithelial cells of anti-hepatitis C virus-pos- itive patients with and without oral lichen planus. Hepatology 2000; 32:97-103. 20. Strauss E. Hepatitis C. Rev Soc Bras Med Trop 2001; 34:69-82. 21. Aquino JA, Pegado KA, Barros LP, Machado LFA. Seroprevalence of hepatitis B virus and hepatitis C virus infections among individuals in the State of Pará. Rev Soc Bras Med Trop 2008; 41:334-7. 35. Hermida M, Ferreiro MC, Barral S, Laredo R, Cas- tro A, Diz Dios P. Detection of HCV RNA in saliva of patients with hepatitis C virus infection by us- ing a highly sensitive test. J Virol Methods 2002; 101:29-35. 22. Oliveira-Filho AB, Pimenta ASC, Rojas MFM, Cha- gas MCM, Crescente JAB, Crespo DM, et al. Preva- lence and genotyping of hepatitis C virus in blood donors in the state of Pará, Northern Brazil. Mem Inst Oswaldo Cruz 2010; 105:103-6. 36. Madhava V, Burgess C, Drucker E. Epidemiology of chronic hepatitis C virus infection in sub-Saharan Africa. Lancet Infect Dis 2002; 2:293-302. 23. Agência Nacional de Vigilância Sanitária/Ministé- rio da Saúde. Resolução RDC nº. 153. http://e-legis. bvs.br/leisref/public/showAct.php?mode=PRINT_ VERSION&id=11662 (accessed on 07/Apr/2008). Submitted on 18/Apr/2009 Final version resubmitted on 04/Feb/2010 Approved on 01/Mar/2010 Submitted on 18/Apr/2009 Final version resubmitted on 04/Feb/2010 Approved on 01/Mar/2010 Cad. Saúde Pública, Rio de Janeiro, 26(4):837-844, abr, 2010
https://openalex.org/W4237621445	https://www.researchsquare.com/article/rs-845733/latest.pdf	English	null	Non-invasive imaging of tau-targeted probe uptake by whole brain multi-spectral optoacoustic tomography	bioRxiv (Cold Spring Harbor Laboratory)	2,021	cc-by	11,171	Non-invasive imaging of tau-targeted probe uptake by whole brain multi-spectral optoacoustic tomography Patrick Vagenknecht ETH Zürich: Eidgenossische Technische Hochschule Zurich Artur Luzgin ETH Zurich: Eidgenossische Technische Hochschule Zurich Maiko Ono National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo BIN JI Fudan University Makoto Higuchi National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo Daniela Noain University Hospital Zurich: UniversitatsSpital Zurich Cinzia A Maschio University of Zurich: Universitat Zurich Jens Sobek ETH Zurich: Eidgenossische Technische Hochschule Zurich Zhenyue Chen University of Zurich: Universitat Zurich Uwe Konietzko University of Zurich: Universitat Zurich Juan Gerez ETH Zurich: Eidgenossische Technische Hochschule Zurich Roland Riek ETH Zurich: Eidgenossische Technische Hochschule Zurich Daniel Razansky ETH Zurich: Eidgenossische Technische Hochschule Zurich Jan Klohs University of Zurich: Universitat Zurich Roger Nitsch Non-invasive imaging of tau-targeted probe uptake by whole brain multi-spectral optoacoustic tomography Patrick Vagenknecht ETH Zürich: Eidgenossische Technische Hochschule Zurich Artur Luzgin ETH Zurich: Eidgenossische Technische Hochschule Zurich Maiko Ono National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo BIN JI Fudan University Makoto Higuchi National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo Daniela Noain University Hospital Zurich: UniversitatsSpital Zurich Cinzia A Maschio University of Zurich: Universitat Zurich Jens Sobek ETH Zurich: Eidgenossische Technische Hochschule Zurich Zhenyue Chen University of Zurich: Universitat Zurich Uwe Konietzko University of Zurich: Universitat Zurich Juan Gerez ETH Zurich: Eidgenossische Technische Hochschule Zurich Roland Riek ETH Zurich: Eidgenossische Technische Hochschule Zurich Daniel Razansky ETH Zurich: Eidgenossische Technische Hochschule Zurich Jan Klohs University of Zurich: Universitat Zurich Roger Nitsch Non-invasive imaging of tau-targeted probe uptake by whole brain multi-spectral optoacoustic tomography by whole brain multi-spectral optoacoustic tomography Patrick Vagenknecht ETH Zürich: Eidgenossische Technische Hochschule Zurich Artur Luzgin ETH Zurich: Eidgenossische Technische Hochschule Zurich Maiko Ono National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo BIN JI Fudan University Makoto Higuchi National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihatsu Hojin Ryoshi Kagaku Gijutsu Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo Daniela Noain University Hospital Zurich: UniversitatsSpital Zurich Cinzia A Maschio University of Zurich: Universitat Zurich Jens Sobek ETH Zurich: Eidgenossische Technische Hochschule Zurich Zhenyue Chen University of Zurich: Universitat Zurich Uwe Konietzko University of Zurich: Universitat Zurich Juan Gerez ETH Zurich: Eidgenossische Technische Hochschule Zurich Roland Riek ETH Zurich: Eidgenossische Technische Hochschule Zurich Daniel Razansky ETH Zurich: Eidgenossische Technische Hochschule Zurich Jan Klohs U i i f Z i h U i i Z i h by whole brain multi-spectral optoa tomography Patrick Vagenknecht ETH Zürich: Eidgenossische Technische Hochschule Zurich Artur Luzgin ETH Zurich: Eidgenossische Technische Hochschule Zurich Maiko Ono National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihat Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo BIN JI Fudan University Makoto Higuchi National Institute of Radiological Sciences: Kokuritsu Kenkyu Kaihat Kenkyu Kaihatsu Kiko Hoshasen Igaku Sogo Kenkyujo Daniela Noain University Hospital Zurich: UniversitatsSpital Zurich Cinzia A Maschio University of Zurich: Universitat Zurich Jens Sobek ETH Zurich: Eidgenossische Technische Hochschule Zurich Zhenyue Chen University of Zurich: Universitat Zurich Uwe Konietzko University of Zurich: Universitat Zurich Juan Gerez ETH Zurich: Eidgenossische Technische Hochschule Zurich Roland Riek ETH Zurich: Eidgenossische Technische Hochschule Zurich Daniel Razansky ETH Zurich: Eidgenossische Technische Hochschule Zurich Jan Klohs University of Zurich: Universitat Zurich Roger Nitsch University of Zurich: Universitat Zurich Xose Luis Dean-Ben University of Zurich: Universitat Zurich Ruiqing Ni (  ruiqing.ni@uzh.ch ) University of Zurich: Universitat Zurich https://orcid.org/0000-0002-0793-2113 University of Zurich: Universitat Zurich Xose Luis Dean-Ben University of Zurich: Universitat Zurich Ruiqing Ni (  ruiqing.ni@uzh.ch ) University of Zurich: Universitat Zurich https://orcid.org/0000-0002-0793-2113 Makoto Higuchi Page 1/26 Research Article Posted Date: August 27th, 2021 Results PBB5 showed specific binding to recombinant K18 tau fibrils by fluorescence assay, to post-mortem Alzheimer’s disease brain tissue homogenate by competitive binding against [11C]PBB3, and to tau deposits (AT-8 positive) in post-mortem corticobasal degeneration and progressive supranuclear palsy brains. Concurrent vMSOT and epi-fluorescence imaging of in vivo PBB5 targeting (i.v.) was performed in P301L and non-transgenic littermate mice. A dose dependent optoacoustic and fluorescence signal intensity was observed in the mouse brains with i.v. administration of different concentrations of PBB5. i.v. administration of PBB5 in P301L mice showed higher retention in tau-laden cortex and hippocampus compared to wild-type, confirmed by ex vivo vMSOT, epi-fluorescence, multiphoton microscopy, immunofluorescence staining using AT-8 antibody for phosphorylated tau. DOI: https://doi.org/10.21203/rs.3.rs-845733/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Version of Record: A version of this preprint was published at European Journal of Nuclear Medicine and Molecular Imaging on February 7th, 2022. See the published version at https://doi.org/10.1007/s00259- 022-05708-w. Page 2/26 Background Abnormal tau accumulation within the brain plays an important role in tauopathies such as Alzheimer’s disease and Frontotemporal dementia. High-resolution imaging of tau deposits at the whole-brain scale in animal disease models are highly desired. Herein, we approach this challenge by non-invasively imaging the brain of P301L mice of 4-repeat tau with concurrent volumetric multi-spectral optoacoustic tomography (vMSOT) at ~ 115 µm spatial resolution using tau-targeted pyridinyl-butadienyl- benzothiazole derivative PBB5 (i.v.). Conclusions We demonstrated non-invasive 3D whole-brain imaging of tau in P301L mice with a vMSOT system using PBB5 at a previously unachieved ~ 115 µm spatial resolution. This platform provides new tool to study tau spreading and clearance in tauopathy mouse model, foreseeable in monitoring of tau targeting putative therapeutics. Background The abnormal cerebral deposition of pathological tau fibrils is a characteristic feature of tauopathy- related neurodegenerative diseases including Alzheimer’s disease (AD), corticobasal degeneration (CBD), progressive supranuclear palsy (PSP) and parkinsonism linked to chromosome 17[1]. The microtubule- associated protein tau (MAPT) is located intracellularly and is composed of six isoforms classified into 4- repeat (4R) and 3-repeat (3R) species [2]. Several tau positron emission tomography (PET) tracers have been developed, including the first generation [18F]flortaucipir, [11C]PBB3, and [11C]THK5351, [18F]THK5117 [3-7]; second generation [18F]MK-6240, [18F]PM-PBB3 (APN1607), [18F]JNJ-64326067, [18F]RO948, [18F]PI-2620, and [18F]GTP1 [8-13]. PET showed the spreading of tau in patients with AD, Page 3/26 which correlates with axonal damage, neurodegeneration, functional network alterations, and cognitive impairment. Thereby, the tau bio-distribution represents a powerful bio-marker with great potential in disease staging [14-23]. In addition, the tau tracer [18F]PM-PBB3 has been shown to facilitate detecting distinct patterns in patients with PSP and CBD compared to AD, indicating its capability for differential diagnosis [9]. Transgenic mouse models (mutations in the MAPT gene) recapitulate pathological features of tauopathy and have greatly advanced our understanding of disease mechanisms [24-28]. Ex vivo high-resolution light-sheet microscopy with anti-tau antibodies or luminescent-conjugated-oligothiophenes enabled whole-brain mapping of tau bio-distribution and spread [29-31]. However, capturing early tau deposits in vivo is needed for a better understanding of the link with other pathological alterations in the deep brain regions. In vivo positron emission tomography (PET) imaging of the cerebral tau accumulation in the transgenic tauopathy mouse has been achieved using [18F]PM-PBB3, [11C]PBB3, [11C]mPBB5, [18F]THK5117, [18F]JNJ-64349311, and 4R-tau specific tracers [18F]CBD-2115[9, 32-39]. PET provides excellent accuracy to map the bio-distribution of tau in human subjects. However, microPET has a limited spatial resolution (0.7-1.5mm) relative to the small mouse brain, which hinders accurate detection of tau, especially in small subcortical brain regions[40]. Fluorescence tau imaging studies using PBB5, luminescent oligothiophene conjugated probes, BF-158, Q-tau 4, pTP-TFE, BODIPY derivative [36, 41-46] and fluorescent-labelled antibodies[47] have been reported. However, fluorescence imaging provides a planar view and limited detection depth. Two-photon imaging of mice with a cranial window using HS-84, methoxy-X04, fluorescent-labelled antibodies[48-50] can follow the development of tau at cellular resolution, but with sub-millimeter field-of-view (FOV) and low penetration depth. Overall, existing imaging approaches are either limited by penetration depth or spatial resolution, which demands for non-invasive imaging tools providing high-resolution performance at whole-brain scales. Background Recently, volumetric multi-spectral optoacoustic tomography (vMSOT) imaging has been shown to provide previously unavailable capabilities to visualize the bio-distribution of amyloid-b (Ab) deposits in mouse models of AD amyloidosis [51-53]. vMSOT capitalizes on the high sensitivity of optical contrast and the high resolution provided by ultrasound [54, 55], and can attain a sufficient penetration depth to cover the whole mouse brain. State-of-the-art vMSOT embodiments enable whole-brain non-invasive imaging with ~115 µm spatial resolution [51, 52, 56-61], i.e., almost an order of magnitude finer compared to modern small-animal microPET scanners. In this study, we investigate on the capabilities of vMSOT assisted with the pyridinyl-butadienyl-benzothiazole derivative PBB5 probe to enable in vivo high- resolution 3D transcranial mapping of tau across the entire mouse brain in 4R-tau P301L mouse models [26]. The targeting performance of the PBB5 probe is further evaluated using post-mortem human brain tissues from patients with AD, PSP and CBD. Staining in human brain In order to investigate whether PBB5 binds tau aggregates in the mammalian brain, we stained caudate/putamen from patients with CBD and motor cortex from PSP with both PBB5 and anti- phosphorylated tau antibody (AT-8), the later used as positive control as it was shown to bind specifically tau inclusions in the brain [62]. Staining using PBB5 and anti-phosphorylated tau antibody (AT-8) in the caudate/putamen from patients with CBD and motor cortex from PSP showed an overlapping signal, which indicates that PBB5 is capable of recognizing AT-8 positive coiled body (Figs. 1h,i) and argyrophilic threads in oligodendrocytes (SFigs. 3a,b), and tufted astrocyte (SFig. 3c). Results vMSOT resolution characterization vMSOT resolution characterization Page 4/26 Page 4/26 Here used a recently developed concurrent optoacoustic and fluorescence imaging set-up and data analysis pipeline for non-invasive transcranial 3D mouse brain imaging (Fig. 1b-d). First we performed a characterization of the reconstructed size of cerebral vessels was assessed by considering a vMSOT image of a 5-month nude mouse brain recorded in vivo (SFig. 1). Specifically, the unmixed signal corresponding to oxygenated hemoglobin was considered as it was shown to provide the best vascular contrast. The vessel size was estimated as the full width at half maximum of the fitted Gaussian curve. Binding assays on human brain tissue We further characterized the binding properties of PBB5 using brain tissues from patients with different tauopathies including AD brain tissue with mixed 3R, 4R-tau as well as CBD and PSP brain tissues with 4R-tau. Competitive binding assay in AD brain homogenates using different concentrations of unlabeled PBB5 and PBB3 against [11C]PBB3 (concentration: 5 nM, specific activity: 86.9 GBq/mmol, radiochemical purity: 96.7 %) indicated an inhibition constant (Ki) = 181.5 nM, and partial replacement for PBB5 (R2 = 0.9889, n = 4), compared to Ki = 2.5 nM for PBB3 (R2 = 0.9669, n = 4) (SFig. 3c). In-vitro fluorescence binding assays in recombinant fibrils We produced tau fibrils using bacterially-produced recombinant monomers of the 4R-tau isoform called K18. The K18 tau fibrils were validated using ThT assay (SFig. 2a), transmission electron microscopy (Fig. 1f) and western blot (SFig. 2b). In order to characterize the binding properties of PBB5 to tau fibrils and aggregates in vivo, we first studied the absorbance spectrum, affinity, binding kinetics and specificity of PBB5 towards recombinant tau K18 fibrils (Fig. 1g). Dosage-dependent performance The optimal dosage of PBB5 to allow clearly optoacoustic (photoacoustic, OA) signal detection in the vMSOT images was established by testing different concentrations of PBB5 (5, 25, 50 mg/kg weight) in P301L and wild-type mice (n = 2-3 each group at each concentration). A dependence on the unmixed PBB5 signal in the vMSOT images with the concentration of the probe was clearly observed at 20-60 minute post-injection (Figs. 2a-c, f). Due to the abundant endogenous hemoglobin signal in the mouse brain, negligible signal increase was detected using 5 mg/kg PBB5. 25 mg/kg PBB5 (i.v.) provided sufficient vMSOT signal increase to be detected in the unmixed images. Fluorescence imaging results indicate a similar dose-dependent signal with PBB5: very intense signal at 25 mg/kg of PBB5, while sufficient fluorescence signal increase was also observed using 5 mg/kg PBB5 (Figs. 2d, e, g). Spectral unmixing of the vMSOT data Spectral unmixing can generally isolate the bio-distribution of any spectrally-distinctive probe from endogenous absorbers in biological tissues. However, spectral coloring effects associated to wavelength- dependent attenuation of light lead to cross-talk artefacts when considering the theoretical spectra of the absorbing substances present in the sample [64, 65]. This is particularly important for spectral windows exhibiting sharp variations of the hemoglobin absorption, e.g. around the 600-630 nm wavelengths (Fig. 1a) [66]. Specifically, the wavelengths and absorbing components were optimized so that the unmixed bio-distribution of PBB5 matches that obtained by subtracting a reference image taken before injection for the sequence vMSOT images taken at 640 nm wavelength. We found that the unmixing performance was optimal when considering five wavelengths (600, 610, 620, 630 and 640 nm) and only HbO and PBB5 as absorbing components. The unmixed bio-distribution of PBB5 is shown to match the differential (baseline-subtracted) vMSOT image at 640 nm (Figs. 1k-o). This corroborates the validity of multispectral unmixing with the selected wavelengths and components as a method to isolate the bio- distribution of PBB5. Non-invasive in vivo vMSOT of PBB5 uptake in the mouse brain The absorption spectrum of PBB5 expands within the far-red range (~590-690 nm, Fig. 1a), where light penetration is significantly enhanced with respect to shorter wavelengths. This facilitates distinguishing the bio-distribution of PBB5 from endogenous chromophores such as deoxyhemoglobin (Hb) and oxyhemoglobin (HbO) via spectral unmixing of vMSOT images acquired in vivo. The surface-weighted PBB5 bio-distribution was also measured in the epi-fluorescence mode in both P301L and wild-type mice by means of a custom-build concurrent planar fluorescence-vMSOT system (Fig. 1b) as described in Page 5/26 Page 5/26 detail elsewhere [51, 52]. The vMSOT imaging data analysis pipeline consisted on the following steps. First, 3D vMSOT images were reconstructed for multiple excitation wavelengths (Fig. 1c). Then, spectral unmixing was performed to isolate the bio-distributions of HbO and PBB5. Finally, co-registration with a magnetic resonance imaging (MRI) mouse brain atlas [63] was performed for volume-of-interest (VOI) analysis (Fig. 1d). After i.v. bolus injection of PBB5 in mice through the mouse tail vein (n = 20 in total), an increase in the fluorescence and/or spectrally unmixed PBB5 signal was observed in the mouse brain parenchyma, arguably indicating that the probe passed the blood-brain barrier. Epi-fluorescence images of the brain corroborated the increase in signal associated to PBB5, albeit providing no depth information and significantly inferior resolution compared to vMSOT (Fig. 1e). PBB5 bio-distribution in P301L and wild-type mice P301L (n = 3) and wild-type mice (n = 3) were imaged at different time points before, during and after injection of PBB5 (25 mg/kg weight i.v.) using the vMSOT system. The unmixed images for the PBB5 Page 6/26 Page 6/26 channel were superimposed onto the MRI atlas for VOI analysis (Fig. 3a, SFig. 4). The time courses of PBB5 (absolute OA (a.u.)) in different brain regions of P301L and wild-type mice were assessed (Fig. 3c). A significantly higher PBB5 OA at 60 minutes post-injection was observed in the cortex, hippocampus and thalamus of P301L mice compared to wild-type mice (Fig. 3e, SVideo 1,2). Similar temporal profiles of vMSOT and planar fluorescence signals were observed throughout the cortical region (Fig. 3d, SFig. 5). Robust correlation was observed between fluorescence and unmixed vMSOT PBB5 absorbance signal (p<0.0001, Pearson’s rank correlation analysis (Fig. 3g, Fig. 4a-c). The test-retest correlation analysis between independent analyses was shown in Fig. 4 indicating the repeatability of the VOI analysis (interrater and intra-rater reliability). Ex vivo validation To validate the in vivo imaging results, the mouse brains were dissected after in vivo imaging and imaged ex vivo using the same vMSOT set-up. The accumulation of PBB5 signal in the cortex and the hippocampus of P301L mouse suggests specific binding of the probe to these regions known to express high tau load. Ex vivo PBB5 epifluorescence images corroborated the tau accumulation in vMSOT, although it was not possible to resolve different regions (Figs. 5a, b, d). Imaging on coronal brain slices (~2 mm thickness, coronal slices cut using a brain matrix at Bregma -2 - 0 mm) indicate retention of signal in the brain of P301L mouse (Figs. 5c, e). To further validate the in vivo PBB5 signal distribution imaged with vMSOT with higher resolution, we imaged fixed brains from P301L and wild-type mice by multiphoton microscopy. In congruence with the in vivo imaging findings, tau deposits morphology was clearly observed in tissue slices with stronger PBB5 signal found in the cortex and hippocampus of P301L mice (Fig. 5f). Immunofluorescence staining performed on horizontal brain tissue sections from P301L and wild-type mice co-staining with anti-phosphorylated tau AT-8 antibody, further confirmed the detection of PBB5 in tau (Fig. 1j, Figs. 5g-i). Discussion New tools for non-invasive mapping of tau deposits with high-resolution in animal models of tauopathy are imperative for understanding the accumulation and spreading of tau deposits [67] and for translational development of tau-targeted therapeutic and diagnostic tools [68, 69]. Herein, we identified PBB5 as a suitable tau imaging probe for vMSOT that binds with high sensitivity and specificity to tau aggregates in-vitro as well as in vivo. This was used to establish a novel in vivo transcranial vMSOT imaging approach to map whole brain tau deposits at ~ 115 µm resolution in a P301L mouse model. The criteria for selecting an appropriate tau-specific probe for vMSOT imaging include suitable absorption spectrum to allow unambiguous unmixing from the endogenous signal of blood (preferably with peak absorption at > 600 nm optical wavelength), high-affinity, low toxicity, low non-specific binding, photostability, low toxicity as well as low molecular weight and suitable lipophilicity to allow sufficient blood-brain barrier passage, and biocompatibility [70]. Herein, we chose PBB5 for its peak absorption at 630–640 nm (where the absorption of hemoglobin decays), which facilitates distinguishing it from blood. Page 7/26 Page 7/26 A competitive binding assay against [11C]PBB3, PBB5 was further shown to have an affinity Ki of 181 nM in post-mortem brain tissue from patients with AD cortex. The binding affinity is in line with the previously reported affinity of PBB5 [36]. Although the specificity and brain penetration of PBB5 is lower than that of A competitive binding assay against [11C]PBB3, PBB5 was further shown to have an affinity Ki of 181 nM in post-mortem brain tissue from patients with AD cortex. The binding affinity is in line with the previously reported affinity of PBB5 [36]. Although the specificity and brain penetration of PBB5 is lower than that of PBB3 (with peak absorption at 405 nm)[36] or PM-PBB3 (emission at 525 nm)[9], its near-infrared (NIR) absorption spectrum allows for epi-fluorescence and vMSOT imaging of deep brain regions. Staining with PBB5 and AT-8 of brain tissues from caudate/putamen patients with CBD and motor cortex from PSP showed an overlapping signal demonstrating that PBB5 is capable of recognizing tau accumulation in coiled body and argyrophilic threads inside oligodendrocytes in brain from CBD and PSP, as well as tufted astrocytes in brain from PSP. Tau plays an important role in the pathogenesis of AD and other primary tauopathy diseases such as CBD and PSP [29, 71, 72]. Discussion Ongoing clinical trials targeting at reducing tau have shown promising results. These include antibodies gosuranemab BIIB092 or non-pharmacological treatments [73–77]. Tau imaging has however been challenging due to the structural diversity of tau isoforms, the difference between 4R and 3R-tau, its intracellular location, as well as the specificity and off-target binding of tau imaging probes [78, 79]. PET assisted with the tau tracer [18F]PM-PBB3 has been shown to detect different patterns in patients with PSP and CBD compared to AD, indicating a role in differential diagnosis [9]. Recent cryogenic electron microscopy has shown that PM-PBB3 binds to tau fibrils in AD brain [80]. An in silico study reported THK5351 probes, T807 binding to different sites on tau fibrils [3, 81] as well as off-target binding sites [79]. Previous autoradiography and PET studies indicated that PBB analogs, THK5351 or THK5117 and JNJ-64349311 but not T807 can detect tauopathy in tau mouse models (P301L, PS19 line) [32, 34–37, 82, 83]. In P301L (CaMKII) mice, tau deposits start at 5 months-of-age, first in the limbic system (entorhinal cortex and hippocampus) and subsequently spreading to the neocortex [26, 84]. Tauopathy deposits in P301L (Thy1.2) mice [26] are most pronounced in the cortex, amygdala and hippocampus, moderate in the brain stem and striatum, and negligible in the cerebellum. Thus we chose cerebellum as reference brain region. Similar to PBB3 and PM-PBB3, PBB5 detects the AT-8 stained neurofibrillary tangle, ghost tangles, tau deposits in astrocytes and oligodendrocytes in the brain from PSP, CBD [62]. In P301L as well as in other tauopathy mouse models, the neurofibrillary tangle is rear and less fibrillar structure is present in the mouse brain [26, 37, 84]. The cortical and hippocampal signals detected by vMSOT in vivo and ex vivo using PBB5 are in accordance with immunofluorescence staining results, and with the known tau distribution in the P301L mouse brain [84, 85]. NIR fluorescence imaging using PBB5 and PET using [11C]mPBB5, respectively, have been previously reported for mapping tau deposition in the brain stem and spinal cord of P301S mice [36]. However, NIR fluorescence imaging detection in deep brain regions was hindered by strong absorption and scattering of the excitation light and emitted fluorescence. Sub- millimeter scale intravital microscopy enables the visualization of tau deposits, but is highly invasive and can only cover a very limited FOV [49]. Immunohistochemical staining on post-mortem brain tissues from patients with CBD and PSP For fluorescence labeling with PBB5, deparaffinized sections were incubated in 50 % ethanol containing 2 µM of PBB5 at room temperature for 30 minutes. The samples were rinsed with 50 % ethanol for 5 minutes, dipped into distilled water twice for 3 minutes, and mounted in non-fluorescent mounting media (VECTASHIELD; Vector Laboratories). Fluorescence images were captured using an FV-1000 confocal laser scanning microscope (Olympus, excitation at 635 nm and emission at 645-720 nm). Following fluorescence microscopy, all sections were autoclaved for antigen retrieval and immunohistochemical stained with anti-phosphorylated tau antibodies AT-8 (pSer202/pThr205, MN1020, Invitrogen, 1:250). Immunolabeling was then examined using a DM4000 microscope (Leica, Germany). Discussion We recently reported on large FOV fluorescence microscopy imaging of tau in P301L mice with 6 micron resolution, which however only provided a planar view[41]. As the spatial resolution of vMSOT is not altered by photon scattering but rather governed by ultrasound Page 8/26 Page 8/26 diffraction, it enables high-resolution mapping and quantification of endogenous tissue chromophores or spectrally distinctive exogenous probes at millimeter to centimeter scale depths [54, 86, 87]. There are several limitations in the current study that need to be highlighted. We did not take into account the spectral colouring effect associated to wavelength-dependent optical attenuation, which may cause distortion in the vMSOT spectra rendered from deep locations [65, 88]. These factors may lead to cross- talk artefacts in the unmixed images corresponding to the contrast agent. Advanced algorithms are required for attaining more accurate performance [88]. Reference tissue model for kinetic model will be potentially useful for improved quantification. In addition, future longitudinal studies are required to determine the sensitivity and specificity of the proposed methodology, how early PBB5 positive tau can be detected, and whether it can follow the spreading of tau in the brain [89]. Conclusions We demonstrated non-invasive whole-brain imaging of tau in P301L mice with a state-of-the-art vMSOT system at ~ 115 µm spatial resolution, which is not feasible with other imaging modalities. This platform provides new tool to study tau spreading and clearance in tauopathy mouse model, foreseeable in monitoring of tau targeting therapeutics. Transmission electron microscopy 4 mL of the fibril samples (~50 mM) in PBS were applied directly to the negatively glow-discharged carbon-coated copper grids, followed by incubation for 1 minute at room temperature. The excess of the solution was gently removed using Whatman filter paper. This step was followed by staining the samples with 10 mL of an aqueous phosphotungstic acid solution (1 %, pH 7.2) for 1 minute. The excess of the stain on the grid was then wiped off with filter paper, and the grid was washed with double-distilled water and air-dried. Finally, the images were recorded at ScopeM (ETH core facility) on an FEI Morgagni 268 electron microscope. In vitro fluorescence assay for the binding of probes to recombinant K18 tau fibrils ed information of the probes and chemical compounds are listed in Suppl. T of the probes and chemical compounds are listed in Suppl. Table 1 [90, 91]. Recombinant K18 4R tau, were expressed and produced by E.coli as described previously [92, 93] (Supplementary material). Details on the recombinant K18 tau fibrils production and characterization in SFig. 2. and supplementary methods. The absorbance of the compounds were measured with a spectrofluometer. Thioflavin T assays against K18 tau fibrils using fluorometer (Fluoromax 4, Horiba scientific, Japan) were performed as described previously [92], with two independent experiments and three technical replicates. PBB5 (excitation peak 630 nm, concentration 1.6 mM) was dissolved in MilliQ H2O or dimethyl sulfoxide and further diluted in 1×PBS (Gibco). PBB5 were then mixed with 5 μL of tau K18 fibril solution in a 45 µL quartz cuvette (quartz SUPRASIL Ultra Micro Cell, Hellma). The solution was incubated for 1 minute at room temperature, resuspended, and fluorescence was measured with a spectrofluorometer using the corresponding excitation wavelength. excitation. Animal models Mice transgenic for MAPT P301L, overexpressing the human 2N/4R tau under neuron-specific Thy1.2 promoter (pR5 line, C57B6.Dg background)[26, 41, 85, 94, 95], and wild-type littermates were used (18 months-old, n = 10 each group, both genders). For resolution characterization, one female athymic nude mice (5-weeks-old, JanvierLab, France) was used. Animals were housed in individually-ventilated cages inside a temperature-controlled room, under a 12-hour dark/light cycle. Pelleted food (3437PXL15, CARGILL) and water were provided ad-libitum. All experiments were performed in accordance with the Swiss Federal Act on Animal Protection. CARGILL) and water were provided ad-libitum. All experiments were performed in accordance with the Swiss Federal Act on Animal Protection. In vitro [11C]PBB3 radiosynthesis and binding assay Frozen tissues derived from the frontal cortex of an AD patient were homogenized in 50 mM Tris-HCl buffer, pH 7.4, containing protease inhibitor cocktail (cOmpleteTM, EDTA-free; Roche), and stored at -80°C until analyses. [11C]PBB3 was synthesized as described previously [62]. To assay radioligand binding with homologous or heterologous blockade, these homogenates (100 µg tissue) were incubated with 5 nM [11C]PBB3 (specific radioactivity: 86.9 GBq/µmol) in the absence or presence of non-radiolabeled PBB3 or PBB5 at varying concentrations ranging from 1×10-11 to 5×10-7 M in Tris-HCl buffer containing 10 % ethanol, pH 7.4, for 30 minute at room temperature. Non-specific binding of [11C]PBB3 was Page 9/26 Page 9/26 Page 9/26 determined in the presence of 5×10-7 M PBB3. Samples were run in quadruplicate. Inhibition constant Ki was determined by using non-linear regression to fit a concentration-binding plot to one-site and two-site binding models derived from the Cheng-Prusoff equation with GraphPad Prism version 5.0 (GraphPad Software), followed by F-test for model selection. In vivo imaging with the hybrid fluorescence and vMSOT system and resolution characterization Simultaneous vMSOT and planar fluorescence imaging at pre-, during, and post i.v. bolus injection of PBB5 was performed using a previously established hybrid system, consisting of an epi-fluorescence fiberscope and a vMSOT system capable of covering the entire mouse brain. The FOV is 10×10mm2 for epi-fluorescence imaging and 15×15×15mm3 for vMSOT, while the spatial resolution is approximately 40 mm and 115 mm for epi-fluorescence and vMSOT, respectively (SFig. 1)[51, 52, 58, 87, 96-99]. Mice were first anesthetized with an initial dose of 4 % isoflurane (Abbott, Cham, Switzerland) in an oxygen/air mixture (200/800 mL/minute), and subsequently maintained at 1.5 % isoflurane in oxygen/air (100/400 mL/minute) throughout the measurement. The fur and the scalps over the head of the mice were then removed. The mice were placed in prone position on a heating pad with feedback control to maintain a constant body temperature. The mice were subsequently injected with a 100 ml bolus containing PBB5 (Fig. 2, dissolved in dimethyl sulfoxide, 0.1 M PBS pH 7.4) through the tail vein. To establish the optimal dosage four P301L and four wild-type mice were used for dose response experiment (5, 25, 50 mg/kg weight). In the subsequent experiment the dose of 25 mg/kg body weight is chosen and used in the following experiment. For vMSOT, the pulse repetition frequency of the laser was set to 25 Hz and the laser wavelength tuned between 550 and 660 nm (5 nm step) on a per pulse basis. Epi-fluorescence imaging was performed by coupling the same beam from the pulsed OPO laser into the excitation fiber bundle. The excited fluorescence field was collected by an imaging fiber bundle comprised of 100,000 fibers and then projected onto an EMCCD camera (Andor iXon life 888, Oxford Instruments, UK). vMSOT and epi-fluorescence signals were recorded simultaneously before injection (108 s duration), during injection (432 s duration with i.v. injection starting at 30 s after the beginning of acquisition) and 20, 40, 60, 90 and 120 minute post-injection (108 s duration each). For the resolution characterization, one female athymic nude mice (n = 1, 5 weeks old, Janvier Lab, France) were used for in vivo experiments. Post-mortem human brain tissues Page 10/26 Post-mortem human brains were obtained from autopsies carried out at the Center for Neurodegenerative Disease Research of the University of Pennsylvania Perelman School of Medicine on patients with AD, CBD and PSP. Tissues for homogenate binding assays were frozen, and tissues for histochemical and immunohistochemical labeling were fixed in 10% neutral buffered formalin followed by embedding in paraffin blocks. All procedures involving the use of human materials were performed in accordance with the ethical guidelines of the Institutional Review Boards of the University of Pennsylvania, and the National Institutes for Quantum and Radiological Science and Technology. Ex vivo hybrid vMSOT and fluorescence imaging To validate the in- and ex vivo signal, one P301L mice were perfused unde ketamine/xylazine/acepromazine maleate anesthesia (75/10/2 mg/kg bo with ice-cold 0.1 M PBS (pH 7.4) and in 4 % paraformaldehyde in 0.1 M P 4 % paraformaldehyde (pH 7.4) and then stored in 0.1 M PBS (pH 7.4) at 4 imaged using vMSOT and hybrid epifluorescence imaging. The brain was brain matrix (World precision medicine, US) into 2 mm thickness at appro imaged again using the same set-up. For this, the spherical array was pos filled with agar gel to guarantee acoustic coupling, which served as a soli brain and brain slice. Uniform illumination of the brain surface was achie the fiber bundle in the lateral apertures of the array and a fourth one provi All recorded OA signals were normalized with the calibrated wavelength-d pulse. The bio-distribution of the probe was estimated via multi-spectral u component algorithm (VCA) considering optical wavelengths from 600 to To validate the in- and ex vivo signal, one P301L mice were perfused under ketamine/xylazine/acepromazine maleate anesthesia (75/10/2 mg/kg body weight, i.p. bolus injection) with ice-cold 0.1 M PBS (pH 7.4) and in 4 % paraformaldehyde in 0.1 M PBS (pH 7.4), and fixed for 4 h in 4 % paraformaldehyde (pH 7.4) and then stored in 0.1 M PBS (pH 7.4) at 4°C. The dissected brain was imaged using vMSOT and hybrid epifluorescence imaging. The brain was cut coronally using a mouse brain matrix (World precision medicine, US) into 2 mm thickness at approximately -2 mm - 0 mm, and imaged again using the same set-up. For this, the spherical array was positioned pointing upwards and filled with agar gel to guarantee acoustic coupling, which served as a solid platform to place the excised brain and brain slice. Uniform illumination of the brain surface was achieved by inserting three arms of the fiber bundle in the lateral apertures of the array and a fourth one providing light delivery from the top. All recorded OA signals were normalized with the calibrated wavelength-dependent energy of the laser pulse. The bio-distribution of the probe was estimated via multi-spectral unmixing considering the vortex component algorithm (VCA) considering optical wavelengths from 600 to 655 nm (5 nm step) [103, 104]. Co-registration with MRI atlas and VOI analysis of the vMSOT data Registration between vMSOT and MRI/atlas provides anatomical reference for regional analysis [52, 101, 102]. These images were co-registered with T2-weighted structural MRI images (Ma-Benveniste-Mirrione- T2 [63]) in PMOD 4.2 (Bruker, Germany) by two readers independently. VOI analysis of 15 brain regions was performed using the embedded Mouse VOI atlas (Ma-Benveniste-Mirrione) in PMOD [52]. Specifically, dynamic time course and retention (60 min) of regional PBB5 absorbance intensity (a.u.) were calculated. Extra-cranial background signal was removed with a mask from the VOI atlas. vMSOT image reconstruction and multi-spectral analysis During the experiments, vMSOT images were reconstructed in real-time by using a graphics processing unit-based implementation of a back-projection formula [51, 52, 100]. The reconstructed images were further processed off-line to unmix the bio-distribution of PBB5 [52]. Specifically, per-voxel least square fitting of the spectral signal profiles to a linear combination of the absorption spectra of oxygenated HbO and PBB5 was performed. Wavelengths between 600 and 640 nm (10 nm step) were considered. The optimum wavelengths and unmixing components were determined by comparing the unmixed bio- Page 11/26 distribution of the probe with that obtained by the pre-injection image during injection of the probe. It was found that including deoxygenated hemoglobin led to larger errors in the bio-distribution of the probe. The probe absorption spectra was experimentally determined as the average spectra of the differential (baseline-subtracted) vMSOT image during bolus perfusion at several major vessels in the brain. The vMSOT spectrum of PBB5 approximately matched the absorption spectrum measured with a spectrophotometer (Avantes BV, Apeldoorn, The Netherlands). The absorption spectrum of HbO was taken from an online database [66]. The effective attenuation coefficient was estimated by considering a constant reduced scattering coefficient of 10 cm-1 for all mice and an optical absorption coefficient corresponding to the unmixed bio-distribution of blood and PBB5. Ex vivo multiphoton microscopy Fixed brains from one P301L and one wild-type mice were imaged at × 20 magnification using Leica TCS SP8 Multiphoton microscopy and analyzed using ImageJ (NIH, United States). Lambda scan 3D rendering Identical settings resolution with Z stack and gain were used. Ex vivo immunofluorescence and confocal imaging Statistics Group comparison of PBB5 absorbance in multiple brain regions at different time points was performed by using two-way analysis of variance with Bonferroni post-hoc analysis (Graphpad Prism, Switzerland). The difference in the fluorescence at 60 minute was compared using two-tail student t test. All data are presented as mean ± standard deviation. Pearson’s rank correlation analysis was used for comparing vMSOT and epi-fluorescence imaging data; and reliability analysis. Significance was set at ∗p < 0.05. Ex vivo immunofluorescence and confocal imaging Page 12/26 Page 12/26 Horizontal brain sections (40 mm) were cut and co-stained with PBB5 and anti-phosphorylated tau (pSer202/pThr205) antibody AT-8 (details in Suppl. Table 1). Sections were counterstained using 4’,6- diamidino-2-phenylindole DAPI [85]. The brain sections were imaged at × 20 magnification using Axio Oberver Z1 and at × 63 magnification using a Leica SP8 confocal microscope (Leica, Germany) for co- localization of PBB5 with AT-8. The images were analyzed using ImageJ (NIH, U.S.A). Competing interests The authors declare no conflicts of interest. Funding Authors' contributions Availability of data and materials Availability of data and materials The datasets generated and/or analyzed during the current study are available in the repository zenodo 10.5281/zenodo.4699067. The datasets generated and/or analyzed during the current study are available in the repository zenodo 10.5281/zenodo.4699067. Competing interests Declarations Ethics approval and consent to participate All experiments were approved by the Cantonal Veterinary Office Zurich (ZH082/18, ZH162/20, ZH161/18). Consent for publication Not applicable. Funding JK received funding from the Swiss National Science Foundation (320030_179277), in the framework of ERA-NET NEURON (32NE30_173678/1), the Synapsis foundation and the Vontobel foundation. RN received funding from Synapsis foundation career development award (2017 CDA-03), Helmut Horten Stiftung, Jubiläumsstiftung von SwissLife, Vontobel Stiftung and UZH Entrepreneur Fellowship, reference no. [MEDEF-20-021] Page 13/26 Page 13/26 The study was designed by RN. MO, BJ, MH performed radiosynthesis, histology and provided binding assay on postmortem human brain. PV, JG were performed fibril production and ThT binding studies. JS performed the SPR assay. ZC, DR designed and built the hybrid fluorescence and optoacoustic tomography system. XLDB, and RN performed in vivo imaging. DN, CM, AL and UK performed histology, confocal and multiphoton microscopy. PV, AL, XLDB, RN performed data analysis. RN wrote the first draft. All authors contributed to the revising of the manuscript. All authors read and approved the final manuscript. Acknowledgements The authors acknowledge Prof. John Robinson, Prof. John Q. Trojanowski, and Prof. Virginia M.-Y. Lee at the University of Pennsylvania for case selection and kindly sharing postmortem brain tissues; Dr Mark Aurel Augath, Michael Reiss at the Institute for Biomedical Engineering, ETH Zurich/University of Zurich; Larissa Kägi, Nadja Straumann, Daniel Schuppli at the Institute for Regenerative Medicine, and ZMB, University of Zurich, Saroj Kumar Rout for technical assistance. Abbreviations Aβ amyloid-beta AD Alzheimer’s disease CBD corticobasal degeneration FOV field-of-view Hb deoxyhemoglobin HbO oxyhemoglobin Ki inhibition constant MAPT microtubule-associated protein tau MRI magnetic resonance imaging NIR near-infrared OA optoacoustic, photoacoustic PBB Page 14/26 pyridinyl-butadienyl-benzothiazole PBS Phosphate-buffered saline PET positron emission tomography PSP progressive supranuclear palsy vMSOT volumetric multi-spectral optoacoustic tomography VOI volume-of-interest 3R 3-repeat 4R 4-repeat pyridinyl-butadienyl-benzothiazole PBS Phosphate-buffered saline PET positron emission tomography PSP progressive supranuclear palsy vMSOT volumetric multi-spectral optoacoustic tomography VOI volume-of-interest 3R 3-repeat 4R 4-repeat References 1. Spillantini MG, Goedert M: Tau pathology and neurodegeneration. The Lancet Neurology 2013, 12(6):609-622. 2. Lee VM, Goedert M, Trojanowski JQ: Neurodegenerative tauopathies. Annu Rev Neurosci 2001, 24:1121-1159. 3. Ono M, Sahara N, Kumata K, Ji B, Ni R, Koga S, Dickson DW, Trojanowski JQ, Lee VM-Y, Yoshida M et al: Distinct binding of PET ligands PBB3 and AV-1451 to tau fibril strains in neurodegenerative tauopathies. Brain : a journal of neurology 2017, 140:764-780. 4. Fleisher AS, Pontecorvo MJ, Devous MD, Sr., Lu M, Arora AK, Truocchio SP, Aldea P, Flitter M, Locascio T, Devine M et al: Positron Emission Tomography Imaging With [18F]flortaucipir and Postmortem Assessment of Alzheimer Disease Neuropathologic Changes. JAMA neurology 2020, 77(7):829-839. 5. Fodero-Tavoletti MT, Okamura N, Furumoto S, Mulligan RS, Connor AR, McLean CA, Cao D, Rigopoulos A, Cartwright GA, O'Keefe G et al: 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer's disease. Brain : a journal of neurology 2011, 134(Pt 4):1089-1100. 6. Okamura N, Furumoto S, Fodero-Tavoletti MT, Mulligan RS, Harada R, Yates P, Pejoska S, Kudo Y, Masters CL, Yanai K et al: Non-invasive assessment of Alzheimer's disease neurofibrillary pathology using 18F-THK5105 PET. Brain : a journal of neurology 2014, 137(Pt 6):1762-1771. Page 15/26 7. Harada R, Okamura N, Furumoto S, Furukawa K, Ishiki A, Tomita N, Tago T, Hiraoka K, Watanuki S, Shidahara M et al: 18F-THK5351: a novel PET radiotracer for imaging neurofibrillary pathology in Alzheimer disease. Journal of nuclear medicine : official publication, Society of Nuclear Medicine 2016, 57(2):208-214. 8. Pascoal TA, Therriault J, Benedet AL, Savard M, Lussier FZ, Chamoun M, Tissot C, Qureshi MNI, Kang MS, Mathotaarachchi S et al: 18F-MK-6240 PET for early and late detection of neurofibrillary tangles. Brain : a journal of neurology 2020, 143(9):2818-2830. 9. Tagai K, Ono M, Kubota M, Kitamura S, Takahata K, Seki C, Takado Y, Shinotoh H, Sano Y, Yamamoto Y et al: High-Contrast In Vivo Imaging of Tau Pathologies in Alzheimer’s and Non-Alzheimer’s Disease Tauopathies. Neuron 2020. 10. Schmidt ME, Janssens L, Moechars D, Rombouts FJR, Timmers M, Barret O, Constantinescu CC, Madonia J, Russell DS, Sandiego CM et al: Clinical evaluation of [(18)F] JNJ-64326067, a novel candidate PET tracer for the detection of tau pathology in Alzheimer's disease. European journal of nuclear medicine and molecular imaging 2020, 47(13):3176-3185. 11. References Franzmeier N, Rubinski A, Neitzel J, Kim Y, Damm A, Na DL, Kim HJ, Lyoo CH, Cho H, Finsterwalder S et al: Functional connectivity associated with tau levels in ageing, Alzheimer's, and small vessel disease. Brain : a journal of neurology 2019, 142(4):1093-1107. S et al: Functional connectivity associated with tau levels in ageing, Alzheimer's, and small vessel disease. Brain : a journal of neurology 2019, 142(4):1093-1107. 19. La Joie R, Visani AV, Baker SL, Brown JA, Bourakova V, Cha J, Chaudhary K, Edwards L, Iaccarino L, Janabi M et al: Prospective longitudinal atrophy in Alzheimer’s disease correlates with the intensity and topography of baseline tau-PET. Science Translational Medicine 2020, 12(524):eaau5732. 20. Johnson KA, Schultz A, Betensky RA, Becker JA, Sepulcre J, Rentz D, Mormino E, Chhatwal J, Amariglio R, Papp K et al: Tau positron emission tomographic imaging in aging and early Alzheimer disease. Annals of neurology 2016, 79(1):110-119. 21. Schöll M, Lockhart SN, Schonhaut DR, O'Neil JP, Janabi M, Ossenkoppele R, Baker SL, Vogel JW, Faria J, Schwimmer HD et al: PET Imaging of Tau Deposition in the Aging Human Brain. Neuron 2016, 89(5):971-982. 22. Wang L, Benzinger TL, Su Y, Christensen J, Friedrichsen K, Aldea P, McConathy J, Cairns NJ, Fagan AM, Morris JC et al: Evaluation of Tau Imaging in Staging Alzheimer Disease and Revealing Interactions Between β-Amyloid and Tauopathy. JAMA Neurology 2016, 73(9):1070-1077. 23. Jacobs HIL, Hedden T, Schultz AP, Sepulcre J, Perea RD, Amariglio RE, Papp KV, Rentz DM, Sperling RA, Johnson KA: Structural tract alterations predict downstream tau accumulation in amyloid-positive older individuals. Nature Neuroscience 2018, 21(3):424-431. 24. SantaCruz K, Lewis J, Spires T, Paulson J, Kotilinek L, Ingelsson M, Guimaraes A, DeTure M, Ramsden M, McGowan E et al: Tau Suppression in a Neurodegenerative Mouse Model Improves Memory Function. Science 2005, 309(5733):476-481. 25. Lewis J, McGowan E, Rockwood J, Melrose H, Nacharaju P, Van Slegtenhorst M, Gwinn-Hardy K, Paul Murphy M, Baker M, Yu X et al: Neurofibrillary tangles, amyotrophy and progressive motor disturbance in mice expressing mutant (P301L) tau protein. Nat Genet 2000, 25(4):402-405. 26. Gotz J, Chen F, van Dorpe J, Nitsch RM: Formation of neurofibrillary tangles in P301l tau transgenic mice induced by Abeta 42 fibrils. Science 2001, 293(5534):1491-1495. P 17/26 27. References Mueller A, Bullich S, Barret O, Madonia J, Berndt M, Papin C, Perrotin A, Koglin N, Kroth H, Pfeifer A et al: Tau PET imaging with (18)F-PI-2620 in patients with Alzheimer's disease and healthy controls: a first-in-human study. Journal of nuclear medicine : official publication, Society of Nuclear Medicine 2019. 12. Leuzy A, Smith R, Ossenkoppele R, Santillo A, Borroni E, Klein G, Ohlsson T, Jögi J, Palmqvist S, Mattsson-Carlgren N et al: Diagnostic Performance of RO948 F 18 Tau Positron Emission Tomography in the Differentiation of Alzheimer Disease From Other Neurodegenerative Disorders. JAMA Neurol 2020, 77(8):955-965. 13. Sanabria Bohórquez S, Marik J, Ogasawara A, Tinianow JN, Gill HS, Barret O, Tamagnan G, Alagille D, Ayalon G, Manser P et al: [(18)F]GTP1 (Genentech Tau Probe 1), a radioligand for detecting neurofibrillary tangle tau pathology in Alzheimer's disease. European journal of nuclear medicine and molecular imaging 2019, 46(10):2077-2089. 14. Ahmed Z, Cooper J, Murray TK, Garn K, McNaughton E, Clarke H, Parhizkar S, Ward MA, Cavallini A, Jackson S et al: A novel in vivo model of tau propagation with rapid and progressive neurofibrillary tangle pathology: the pattern of spread is determined by connectivity, not proximity. Acta Neuropathol 2014, 127(5):667-683. 15. Hanseeuw BJ, Betensky RA, Jacobs HIL, Schultz AP, Sepulcre J, Becker JA, Cosio DMO, Farrell M, Quiroz YT, Mormino EC et al: Association of Amyloid and Tau With Cognition in Preclinical Alzheimer Disease: A Longitudinal Study. JAMA Neurology 2019, 76(8):915-924. Page 16/26 16. Vogel JW, Iturria-Medina Y, Strandberg OT, Smith R, Levitis E, Evans AC, Hansson O: Spread of pathological tau proteins through communicating neurons in human Alzheimer's disease. Nature Page 16/26 Page 16/26 communications 2020, 11(1):2612. 17. Hoenig MC, Bischof GN, Seemiller J, Hammes J, Kukolja J, Onur Ö A, Jessen F, Fliessbach K, Neumaier B, Fink GR et al: Networks of tau distribution in Alzheimer's disease. Brain : a journal of neurology 2018, 141(2):568-581. 18. Franzmeier N, Rubinski A, Neitzel J, Kim Y, Damm A, Na DL, Kim HJ, Lyoo CH, Cho H, Finsterwalder S et al: Functional connectivity associated with tau levels in ageing, Alzheimer's, and small vessel 18. Franzmeier N, Rubinski A, Neitzel J, Kim Y, Damm A, Na DL, Kim HJ, Lyoo CH, Cho H, Finsterwalder 18. References Saito T, Mihira N, Matsuba Y, Sasaguri H, Hashimoto S, Narasimhan S, Zhang B, Murayama S, Higuchi M, Lee VMY et al: Humanization of the entire murine Mapt gene provides a murine model of pathological human tau propagation. J Biol Chem 2019, 294(34):12754-12765. Page 17/26 28. de Calignon A, Polydoro M, Suárez-Calvet M, William C, Adamowicz DH, Kopeikina KJ, Pitstick R, Sahara N, Ashe KH, Carlson GA et al: Propagation of tau pathology in a model of early Alzheimer's disease. Neuron 2012, 73(4):685-697. 29. Fung CW, Guo J, Fu H, Figueroa HY, Konofagou EE, Duff KE: Atrophy associated with tau pathology precedes overt cell death in a mouse model of progressive tauopathy. Science Advances 2020, 6(42):eabc8098. 30. Detrez JR, Maurin H, Van Kolen K, Willems R, Colombelli J, Lechat B, Roucourt B, Van Leuven F, Baatout S, Larsen P et al: Regional vulnerability and spreading of hyperphosphorylated tau in seeded mouse brain. Neurobiol Dis 2019, 127:398-409. 31. Liebmann T, Renier N, Bettayeb K, Greengard P, Tessier-Lavigne M, Flajolet M: Three-Dimensional Study of Alzheimer's Disease Hallmarks Using the iDISCO Clearing Method. Cell reports 2016, 16(4):1138- 1152. 32. Brendel M, Jaworska A, Probst F, Overhoff F, Korzhova V, Lindner S, Carlsen J, Bartenstein P, Harada R, Kudo Y et al: Small-Animal PET Imaging of Tau Pathology with 18F-THK5117 in 2 Transgenic Mouse Models. Journal of nuclear medicine : official publication, Society of Nuclear Medicine 2016, 57(5):792-798. 33. Chaney AM, Lopez-Picon FR, Serrière S, Wang R, Bochicchio D, Webb SD, Vandesquille M, Harte MK, Georgiadou C, Lawrence C et al: Prodromal neuroinflammatory, cholinergic and metabolite dysfunction detected by PET and MRS in the TgF344-AD transgenic rat model of AD: a collaborative multi-modal study. Theranostics 2021, 11(14):6644-6667. 34. Declercq L, Rombouts F, Koole M, Fierens K, Mariën J, Langlois X, Andrés J, Schmidt M, Macdonald G, Moechars D et al: Preclinical evaluation of 18F-JNJ64349311, a novel PET tracer for tau imaging. Journal of Nuclear Medicine 2017, 58:jnumed.116.185199. 35. Ishikawa A, Tokunaga M, Maeda J, Minamihisamatsu T, Shimojo M, Takuwa H, Ono M, Ni R, Hirano S, Kuwabara S et al: In Vivo Visualization of Tau Accumulation, Microglial Activation, and Brain Atrophy in a Mouse Model of Tauopathy rTg4510. Journal of Alzheimer's disease : JAD 2018, 61(3):1037- 1052. 36. References Maruyama M, Shimada H, Suhara T, Shinotoh H, Ji B, Maeda J, Zhang MR, Trojanowski JQ, Lee VM, Ono M et al: Imaging of tau pathology in a tauopathy mouse model and in Alzheimer patients compared to normal controls. Neuron 2013, 79(6):1094-1108. 37. Ni R, Ji B, Ono M, Sahara N, Zhang MR, Aoki I, Nordberg A, Suhara T, Higuchi M: Comparative in- vitro and in-vivo quantifications of pathological tau deposits and their association with neurodegeneration in tauopathy mouse models. Journal of nuclear medicine : official publication, Society of Nuclear Medicine 2018, 59(6):960-966. Page 18/26 Page 18/26 38. Barron AM, Ji B, Fujinaga M, Zhang MR, Suhara T, Sahara N, Aoki I, Tsukada H, Higuchi M: In vivo positron emission tomography imaging of mitochondrial abnormalities in a mouse model of tauopathy. Neurobiol Aging 2020, 94:140-148. 39. Lindberg A, Knight AC, Sohn D, Rakos L, Tong J, Radelet A, Mason NS, Stehouwer JS, Lopresti BJ, Klunk WE et al: Radiosynthesis, In Vitro and In Vivo Evaluation of [(18)F]CBD-2115 as a First-in-Class Radiotracer for Imaging 4R-Tauopathies. ACS Chem Neurosci 2021, 12(4):596-602. 40. Mannheim JG, Schmid AM, Schwenck J, Katiyar P, Herfert K, Pichler BJ, Disselhorst JA: PET/MRI Hybrid Systems. Semin Nucl Med 2018, 48(4):332-347. 41. Ni R, Chen Z, GEREZ JA, Shi G, Zhou Q, Riek R, Nilsson PR, Razansky D, Klohs J: Detection of cerebral tauopathy in P301L mice using high-resolution large-field multifocal illumination fluorescence microscopy. Biomed Opt Express 2020. 42. Elbatrawy AA, Hyeon SJ, Yue N, Osman EEA, Choi SH, Lim S, Kim YK, Ryu H, Cui M, Nam G: "Turn- On" Quinoline-Based Fluorescent Probe for Selective Imaging of Tau Aggregates in Alzheimer's Disease: Rational Design, Synthesis, and Molecular Docking. ACS Sens 2021. 43. Zhao Y, Tietz O, Kuan WL, Haji-Dheere AK, Thompson S, Vallin B, Ronchi E, Tóth G, Klenerman D, Aigbirhio FI: A fluorescent molecular imaging probe with selectivity for soluble tau aggregated protein. Chem Sci 2020, 11(18):4773-4778. 44. Okamura N, Suemoto T, Furumoto S, Suzuki M, Shimadzu H, Akatsu H, Yamamoto T, Fujiwara H, Nemoto M, Maruyama M et al: Quinoline and benzimidazole derivatives: candidate probes for in vivo imaging of tau pathology in Alzheimer's disease. The Journal of neuroscience : the official journal of the Society for Neuroscience 2005, 25(47):10857-10862. 45. References Verwilst P, Kim H-R, Seo J, Sohn N-W, Cha S-Y, Kim Y, Maeng S, Shin J-W, Kwak JH, Kang C et al: Rational Design of in Vivo Tau Tangle-Selective Near-Infrared Fluorophores: Expanding the BODIPY Universe. Journal of the American Chemical Society 2017, 139(38):13393-13403. Page 19/26 nal Design of in Vivo Tau Tangle-Selective Near-Infrared Fluorophores: Expanding the BODIPY rse. Journal of the American Chemical Society 2017, 139(38):13393-13403. Zhou K, Yuan C, Dai B, Wang K, Chen Y, Ma D, Dai J, Liang Y, Tan H, Cui M: Environment-Sensitive nfrared Probe for Fluorescent Discrimination of Aβ and Tau Fibrils in AD Brain. Journal of Medicinal mistry 2019, 62(14):6694-6704. Krishnaswamy S, Lin Y, Rajamohamedsait WJ, Rajamohamedsait HB, Krishnamurthy P, dsson EM: Antibody-derived in vivo imaging of tau pathology. The Journal of neuroscience : the al journal of the Society for Neuroscience 2014, 34(50):16835-16850. Kuchibhotla KV, Wegmann S, Kopeikina KJ, Hawkes J, Rudinskiy N, Andermann ML, Spires-Jones acskai BJ, Hyman BT: Neurofibrillary tangle-bearing neurons are functionally integrated in cortical ts in vivo. Proceedings of the National Academy of Sciences 2014, 111(1):510-514. 46. Zhou K, Yuan C, Dai B, Wang K, Chen Y, Ma D, Dai J, Liang Y, Tan H, Cui M: Environment-Sensitive Near-Infrared Probe for Fluorescent Discrimination of Aβ and Tau Fibrils in AD Brain. Journal of Medicinal Chemistry 2019, 62(14):6694-6704. 47. Krishnaswamy S, Lin Y, Rajamohamedsait WJ, Rajamohamedsait HB, Krishnamurthy P, Sigurdsson EM: Antibody-derived in vivo imaging of tau pathology. The Journal of neuroscience : the official journal of the Society for Neuroscience 2014, 34(50):16835-16850. 48. Kuchibhotla KV, Wegmann S, Kopeikina KJ, Hawkes J, Rudinskiy N, Andermann ML, Spires-Jones TL, Bacskai BJ, Hyman BT: Neurofibrillary tangle-bearing neurons are functionally integrated in cortical circuits in vivo. Proceedings of the National Academy of Sciences 2014, 111(1):510-514. Page 19/26 Page 19/26 49. Calvo-Rodriguez M, Hou SS, Snyder AC, Dujardin S, Shirani H, Nilsson KPR, Bacskai BJ: In vivo detection of tau fibrils and amyloid beta aggregates with luminescent conjugated oligothiophenes and multiphoton microscopy. Acta Neuropathol Commun 2019, 7(1):171. 50. Wu Q, Lin Y, Gu J, Sigurdsson E: Dynamic assessment of tau immunotherapies in the brains of live animals by two-photon imaging. EBioMedicine 2018, 35. 51. References Ni R, Villois A, Dean-Ben XL, Chen Z, Vaas M, Stavrakis S, Shi G, deMello A, Ran C, Razansky D et al: In-vitro and in-vivo characterization of CRANAD-2 for multi-spectral optoacoustic tomography and fluorescence imaging of amyloid-beta deposits in Alzheimer mice. bioRxiv 2020:2020.2010.2027.353862. 52. Ni R, Dean-Ben XL, Kirschenbaum D, Rudin M, Chen Z, Crimi A, Voigt F, Nilsson PR, Helmchen F, Nitsch RM et al: Whole brain optoacoustic tomography reveals strain-specific regional beta-amyloid densities in Alzheimer`s disease amyloidosis models. bioRxiv 2020:DOI: 10.1101/2020.1102.1125.964064. 53. Ni R, Chen Z, Shi G, Villois A, Zhou Q, Arosio P, Nitsch RM, Nilsson KPR, Klohs J, Razansky D: Transcranial <em>in vivo</em> detection of amyloid-beta at single plaque resolution with large-field multifocal illumination fluorescence microscopy. bioRxiv 2020:2020.2002.2001.929844. 54. Wang LV, Hu S: Photoacoustic tomography: in vivo imaging from organelles to organs. Science 2012, 335(6075):1458-1462. 54. Wang LV, Hu S: Photoacoustic tomography: in vivo imaging from organelles to organs. Science 2012, 335(6075):1458-1462. 55. Razansky D, Distel M, Vinegoni C, Ma R, Perrimon N, Köster RW, Ntziachristos V: Multispectral opto-acoustic tomography of deep-seated fluorescent proteins in vivo. Nature Photonics 2009, 3(7):412- 417. 56. Razansky D, Distel M, Vinegoni C, Ma R, Perrimon N, Köster RW, Ntziachristos V: Multispectral opto-acoustic tomography of deep-seated fluorescent proteins in vivo. Nature Photonics 2009, 3:412. 57. Wang LV, Yao J: A practical guide to photoacoustic tomography in the life sciences. Nature methods 2016, 13(8):627. 58. Deán-Ben XL, Sela G, Lauri A, Kneipp M, Ntziachristos V, Westmeyer GG, Shoham S, Razansky D: Functional optoacoustic neuro-tomography for scalable whole-brain monitoring of calcium indicators. Light Sci Appl 2016, 5(12):e16201. 59. Gottschalk S, Degtyaruk O, Mc Larney B, Rebling J, Hutter MA, Deán-Ben XL, Shoham S, Razansky D: Rapid volumetric optoacoustic imaging of neural dynamics across the mouse brain. Nature Biomedical Engineering 2019, 3(5):392-401. 60. Vaas M, Ni R, Rudin M, Kipar A, Klohs J: Extracerebral Tissue Damage in the Intraluminal Filament Mouse Model of Middle Cerebral Artery Occlusion. Front Neurol 2017, 8:85. Page 20/26 61. Razansky D, Klohs J, Ni R: Multi-scale optoacoustic molecular imaging of brain diseases. European journal of nuclear medicine and molecular imaging 2021. 62. Ono M, Sahara N, Kumata K, Ji B, Ni RQ, Koga S, Dickson DW, Trojanowski JQ, Lee VMY, Yoshida M et al: Distinct binding of PET ligands PBB3 and AV-1451 to tau fibril strains in neurodegenerative tauopathies. Brain 2017, 140:764-780. 63. Ma Y, Hof PR, Grant SC, Blackband SJ, Bennett R, Slatest L, McGuigan MD, Benveniste H: A three- dimensional digital atlas database of the adult C57BL/6J mouse brain by magnetic resonance microscopy. Neuroscience 2005, 135(4):1203-1215. 64. Cox B, Laufer JG, Arridge SR, Beard PC: Quantitative spectroscopic photoacoustic imaging: a review. J Biomed Opt 2012, 17(6):061202. 65. Tzoumas S, Deliolanis N, Morscher S, Ntziachristos V: Unmixing Molecular Agents From Absorbing Tissue in Multispectral Optoacoustic Tomography. IEEE Trans Med Imaging 2014, 33(1):48-60. 66. Prahl S: https://omlc.org/spectra/hemoglobin/. 65. Tzoumas S, Deliolanis N, Morscher S, Ntziachristos V: Unmixing Molecular Agents From Absorbing Tissue in Multispectral Optoacoustic Tomography. IEEE Trans Med Imaging 2014, 33(1):48-60. 66. Prahl S: https://omlc.org/spectra/hemoglobin/. 66. Prahl S: https://omlc.org/spectra/hemoglobin/. 67. He Z, McBride JD, Xu H, Changolkar L, Kim SJ, Zhang B, Narasimhan S, Gibbons GS, Guo JL, Kozak M et al: Transmission of tauopathy strains is independent of their isoform composition. Nature communications 2020, 11(1):7. 68. 54. Wang LV, Hu S: Photoacoustic tomography: in vivo imaging from organelles to organs. Science 2012, 335(6075):1458-1462. Congdon EE, Sigurdsson EM: Tau-targeting therapies for Alzheimer disease. Nat Rev Neurol 2018, 14(7):399-415. 69. Lo CH, Lim CK, Ding Z, Wickramasinghe SP, Braun AR, Ashe KH, Rhoades E, Thomas DD, Sachs JN: Targeting the ensemble of heterogeneous tau oligomers in cells: A novel small molecule screening platform for tauopathies. Alzheimer's & dementia : the journal of the Alzheimer's Association 2019, 15(11):1489-1502. 70. Zeng L, Ma G, Lin J, Huang P: Photoacoustic Probes for Molecular Detection: Recent Advances and Perspectives. Small 2018, 14(30):e1800782. 71. Ittner A, Chua SW, Bertz J, Volkerling A, van der Hoven J, Gladbach A, Przybyla M, Bi M, van Hummel A, Stevens CH et al: Site-specific phosphorylation of tau inhibits amyloid-β toxicity in Alzheimer's mice. Science 2016, 354(6314):904-908. 72. Busche MA, Wegmann S, Dujardin S, Commins C, Schiantarelli J, Klickstein N, Kamath TV, Carlson GA, Nelken I, Hyman BT: Tau impairs neural circuits, dominating amyloid-β effects, in Alzheimer models in vivo. Nat Neurosci 2019, 22(1):57-64. 73. Leinenga G, Gotz J: Scanning ultrasound removes amyloid-beta and restores memory in an Alzheimer's disease mouse model. Sci Transl Med 2015, 7(278):278ra233. Page 21/26 Page 21/26 74. Gibbons GS, Kim SJ, Wu Q, Riddle DM, Leight SN, Changolkar L, Xu H, Meymand ES, O'Reilly M, Zhang B et al: Conformation-selective tau monoclonal antibodies inhibit tau pathology in primary neurons and a mouse model of Alzheimer's disease. Mol Neurodegener 2020, 15(1):64. 75. Boxer AL, Qureshi I, Ahlijanian M, Grundman M, Golbe LI, Litvan I, Honig LS, Tuite P, McFarland NR, O'Suilleabhain P et al: Safety of the tau-directed monoclonal antibody BIIB092 in progressive supranuclear palsy: a randomised, placebo-controlled, multiple ascending dose phase 1b trial. The Lancet Neurology 2019, 18(6):549-558. 76. Sopko R, Golonzhka O, Arndt J, Quan C, Czerkowicz J, Cameron A, Smith B, Murugesan Y, Gibbons G, Kim SJ et al: Characterization of tau binding by gosuranemab. Neurobiol Dis 2020, 146:105120. 77. Brendel M, Deussing M, Blume T, Kaiser L, Probst F, Overhoff F, Peters F, von Ungern-Sternberg B, Ryazanov S, Leonov A et al: Late-stage Anle138b treatment ameliorates tau pathology and metabolic decline in a mouse model of human Alzheimer's disease tau. Alzheimer's research & therapy 2019, 11(1):67. 78. Villemagne VL, Dore V, Burnham SC, Masters CL, Rowe CC: Imaging tau and amyloid-beta proteinopathies in Alzheimer disease and other conditions. Nat Rev Neurol 2018, 14(4):225-236. 79. 54. Wang LV, Hu S: Photoacoustic tomography: in vivo imaging from organelles to organs. Science 2012, 335(6075):1458-1462. Murugan NA, Chiotis K, Rodriguez-Vieitez E, Lemoine L, Agren H, Nordberg A: Cross-interaction of tau PET tracers with monoamine oxidase B: evidence from in silico modelling and in vivo imaging. European journal of nuclear medicine and molecular imaging 2019, 46(6):1369-1382. 80. Shi Y, Murzin A, Falcon B, Epstein A, Machin J, Tempest P, Newell K, Vidal R, Garringer H, Sahara N et al: Cryo-EM structures of tau filaments from Alzheimer’s disease with PET ligand APN-1607. Acta Neuropathol 2021:1-12. 81. Murugan NA, Nordberg A, Ågren H: Cryptic Sites in Tau Fibrils Explain the Preferential Binding of the AV-1451 PET Tracer toward Alzheimer’s Tauopathy. ACS Chemical Neuroscience 2021. 82. Brendel M, Yousefi BH, Blume T, Herz M, Focke C, Deussing M, Peters F, Lindner S, von Ungern- Sternberg B, Drzezga A et al: Comparison of 18F-T807 and 18F-THK5117 PET in a Mouse Model of Tau Pathology. Frontiers in Aging Neuroscience 2018, 10(174). 83. Lindberg A, Knight AC, Sohn D, Rakos L, Tong J, Radelet A, Mason NS, Stehouwer JS, Lopresti BJ, Klunk WE et al: Radiosynthesis, In Vitro and In Vivo Evaluation of [18F]CBD-2115 as a First-in-Class Radiotracer for Imaging 4R-Tauopathies. ACS Chem Neurosci 2021, 12(4):596-602. 84. Gotz J, Chen F, Barmettler R, Nitsch RM: Tau filament formation in transgenic mice expressing P301L tau. J Biol Chem 2001, 276(1):529-534. 84. Gotz J, Chen F, Barmettler R, Nitsch RM: Tau filament formation in transgenic mice expressing P301L tau. J Biol Chem 2001, 276(1):529-534. Page 22/26 85. Ni R, Zarb Y, Kuhn GA, Müller R, Yundung Y, Nitsch RM, Kulic L, Keller A, Klohs J: SWI and phase imaging reveal intracranial calcifications in the P301L mouse model of human tauopathy. Magnetic Resonance Materials in Physics, Biology and Medicine 2020. Resonance Materials in Physics, Biology and Medicine 2020. Resonance Materials in Physics, Biology and Medicine 2020. 86. Taruttis A, Ntziachristos V: Advances in real-time multispectral optoacoustic imaging and its applications. Nature Photonics 2015, 9(4):219-227. 87. Deán-Ben XL, Razansky D: Adding fifth dimension to optoacoustic imaging: volumetric time- resolved spectrally enriched tomography. Light: Science &Amp; Applications 2014, 3:e137. 88. Tzoumas S, Nunes A, Olefir I, Stangl S, Symvoulidis P, Glasl S, Bayer C, Multhoff G, Ntziachristos V: Eigenspectra optoacoustic tomography achieves quantitative blood oxygenation imaging deep in tissues. Nature communications 2016, 7:12121. 89. Colin M, Dujardin S, Schraen-Maschke S, Meno-Tetang G, Duyckaerts C, Courade JP, Buee L: From the prion-like propagation hypothesis to therapeutic strategies of anti-tau immunotherapy. Acta neuropathologica 2019. 90. Shirani H, Linares M, Sigurdson CJ, Lindgren M, Norman P, Nilsson KP: A Palette of Fluorescent Thiophene-Based Ligands for the Identification of Protein Aggregates. Chemistry 2015, 21(43):15133- 15137. 91. Klingstedt T, Aslund A, Simon RA, Johansson LB, Mason JJ, Nystrom S, Hammarstrom P, Nilsson KP: Synthesis of a library of oligothiophenes and their utilization as fluorescent ligands for spectral assignment of protein aggregates. Organic & biomolecular chemistry 2011, 9(24):8356-8370. 92. Burmann BM, Gerez JA, Matecko-Burmann I, Campioni S, Kumari P, Ghosh D, Mazur A, Aspholm EE, Sulskis D, Wawrzyniuk M et al: Regulation of alpha-synuclein by chaperones in mammalian cells. Nature 2019. 93. Gerez JA, Prymaczok NC, Rockenstein E, Herrmann US, Schwarz P, Adame A, Enchev RI, Courtheoux T, Boersema PJ, Riek R et al: A cullin-RING ubiquitin ligase targets exogenous alpha-synuclein and inhibits Lewy body-like pathology. Sci Transl Med 2019, 11(495). 94. Massalimova A, Ni R, Nitsch RM, Reisert M, von Elverfeldt D, Klohs J: DTI reveals whole-brain microstructural changes in the P301L mouse model of tauopathy. Neurodegener Dis 2021:2020.2010.2028.358465. 95. Vagenknecht P, Dean-Ben XL, Gerez JA, Chen Z, Nitsch RM, Klohs J, Razansky D, Ni R: Non- invasive optoacoustic imaging of tau in P301L mice. In: Biophotonics Congress: Optics in the Life Sciences 2021: 2021; 2021: OTu1E.4. 96. Dean-Ben XL, Robin J, Ni R, Razansky D: Noninvasive three-dimensional optoacoustic localization microangiography of deep tissues. 2020. Page 23/26 97. Chen Z, Dean-Ben XL, Gottschalk S, Razansky D: Hybrid system for in vivo epifluorescence and 4D optoacoustic imaging. Opt Lett 2017, 42(22):4577-4580. 98. Chen Z, Deán-Ben XL, Liu N, Gujrati V, Gottschalk S, Ntziachristos V, Razansky D: Concurrent fluorescence and volumetric optoacoustic tomography of nanoagent perfusion and bio-distribution in solid tumors. Biomedical optics express 2019, 10(10):5093-5102. 99. Resonance Materials in Physics, Biology and Medicine 2020. Chen Z, Deán-Ben XL, Gottschalk S, Razansky D: Performance of optoacoustic and fluorescence imaging in detecting deep-seated fluorescent agents. Biomed Opt Express 2018, 9(5):2229-2239. 100. Ali O, Deán-Ben XL, Daniel R: Realtime parallel back-projection algorithm for three-dimensional optoacoustic imaging devices. In: ProcSPIE: 2013; 2013. 101. Ni R, Vaas M, Ren W, Klohs J: Non-invasive detection of acute cerebral hypoxia and subsequent matrix-metalloproteinase activity in a mouse model of cerebral ischemia using multispectral- optoacoustic-tomography. Neurophotonics 2018, 5(1):015005. 102. Ren W, Skulason H, Schlegel F, Rudin M, Klohs J, Ni R: Automated registration of magnetic resonance imaging and optoacoustic tomography data for experimental studies. Neurophotonics 2019, 6(2):1-10, 10. 103. Ding L, Dean-Ben XL, Burton NC, Sobol RW, Ntziachristos V, Razansky D: Constrained Inversion and Spectral Unmixing in Multispectral Optoacoustic Tomography. IEEE transactions on medical imaging 2017, 36(8):1676-1685. 104. Nascimento JMP, Dias JMB: Vertex component analysis: a fast algorithm to unmix hyperspectral data. IEEE Transactions on Geoscience and Remote Sensing 2005, 43(4):898-910. 104. Nascimento JMP, Dias JMB: Vertex component analysis: a fast algorithm to unmix hyperspectral data. IEEE Transactions on Geoscience and Remote Sensing 2005, 43(4):898-910. 104. Nascimento JMP, Dias JMB: Vertex component analysis: a fast algorithm to unmix hyperspectral data. IEEE Transactions on Geoscience and Remote Sensing 2005, 43(4):898-910. Figure 1 P 24/26 Non-invasive tau epifluorescence-vMSOT imaging pipeline (a) Chemical structure of the probe PBB5 and extinction spectrum of HbO and Hb along with the spectrum of PBB5 measured by volumetric multi- spectral optoacoustic tomography (vMSOT); (b) Set-up of the hybrid epifluorescence-vMSOT system for tau mapping across entire mouse brain; (c) Volumetric reconstructions of the in vivo vMSOT data for five distinct excitation wavelengths (600, 610, 620, 630, 640 nm) used for spectral unmixing; (d) coronal, horizontal and sagittal view of PBB5 and HbO; Absorbance intensity scale: 0-1. (e) Simultaneous epi- fluorescence imaging in one P301L moues brain after i.v. injection of PBB5. (f-j) PBB5 characterization on recombinant fibrils and staining on human brain. (f) Transmission electron microscopes image of K18 tau fibril. scalebar=200 nm. (g) Fluorescence binding assay using PBB5 on K18 4R tau fibrils and blank Page 24/26 Page 24/26 (dd. water) with PBB5, CPS, counts per sounds; (h, i, j) PBB5-positive and AT-8-positive inclusions indicated coiled bodies in the caudate/putamen from patients with corticobasal degeneration (CBD) and motor cortex from progressive supranuclear palsy (PSP) and hippocampus sections from P301L mice; scalebar=10 μm. AT-8: an anti-phosphorylated tau antibody. (k-l) Comparison of vMSOT processing methods. The baseline-subtracted single wavelength vMSOT 3D rendering image acquired at 640 nm (k) matches the multispectrally unmixed bio-distribution of PBB5 (n). (l, o) Time-lapse curves of multispectrally unmixed PBB5 signals and baseline-subtracted 640 nm signals corresponding to selected points, cortex (red), superior sagittal sinus (dark blue), hippocampus (green), vessel (light blue), indicated in (k) and (n). The multispectrally unmixed bio-distribution of HbO reveals major cerebral vessels (m). Figure 2 Dose determination for in vivo tau imaging with vMSOT. (a-e) vMSOT images of three different concentration of PBB5, 5 mg/kg weight (blue square), 25 mg/kg weight (red square); b), and 50 mg/kg weight (green square, a) and epifluorescence images of 5 mg/kg weight (blue square, e), 25 mg/kg weight (red square, d); (f) Time curve of unmixed PBB5 absorbance profile during the first 300 seconds (within 7 minute dynamic i.v. injection using three different concentration of PBB5, 5 mg/kg weight (blue line), 25 mg/kg weight (red line), 50 mg/kg weight (green line); No clear signal increase was detected using 5 mg/kg weight dose; (g) Fluorescence intensity curve of PBB5 using 5 mg/kg weight (light blue) and 25 mg/kg weight (dark blue). PBB5 was injected i.v. at 30 s. Figure 3 Regional tau distribution revealed by in vivo vMSOT imaging using PBB5 probe in P301L and wild-type mice, (a) Wild-type (WT) and transgenic P301L mice; at pre-injection, 20, 40, 60 min following dye administration showing coronal, sagittal and horizontal views overlaid over the masked magnetic resonance imaging-based brain atlas. PBB5 absorbance signal strength is indicated by rainbow color- map; (b) Example of epi-fluorescence images from one P301L mouse at 20, 40, 60 min following dye administration; (c, d) Time course of cortical, hippocampal, thalamic volume-of-interest PBB5 signal (absorbance signal) and cortical region-of-interest fluorescence intensity; (e, f) regional comparison of probe absorbance signal retention and fluorescence intensity at 60 min post-injection, Data are presented as mean±SD; P301L (n=3), and NTL (n=3); p<0.05, p<0.01, **p<0.001 comparison between WT and P301L mice. Cortex: Ctx; Hippocampus: Hip; Thalamus: TH; (h) Correlation between optoacoustic and Fluorescence imaging across different mice using Pearson rank analysis. Figure 5 Ex vivo validation using vMSOT, epi-fluorescence imaging, multiphoton microscopy and immunofluorescence staining, (a-c) Ex vivo vMSOT of whole brain, and brain slice at 90 minutes after PBB5 i.v. injection; (a) 3D rendering of ex vivo vMSOT data unmixed for PBB5 distribution in P301L mouse brain; (b) Overlay of (a) on MRI structural data; (c) ex vivo vMSOT of 1 mm mouse brain slice data unmixed for PBB5 distribution in P301L mouse brain. PBB5 absorbance signal strength is indicated by blue-green color-map; (d, e) Epi-fluorescence of (a, c); (f) Multiphoton microscopy (MPM) regional quantification multiphoton. Scale bar = 20 μm; (g-i) Confocal microscopic images of hippocampus sections from P301L mice. PBB5 (white), Alexa488-AT-8 (green) in the hippocampus areas. Scale bar = 5 μm; Figure 4 Figure 4 Page 25/26 Reliability of volume-of-interest (VOI) analysis. (a) VOI labeling of the segmented brain areas - Cortex: red; Thalamus: green; Hippocampus, blue. (b) Intra-rater reliability. (c) Inter-rater reliability. Analysis and reanalysis using PMOD volume-of-interest analysis process. Pearson rank analysis indicate robust correlation between two independent analysis for the cortical PBB5 absorbance intensity (a.u.). Reliability of volume-of-interest (VOI) analysis. (a) VOI labeling of the segmented brain areas - Cortex: red; Thalamus: green; Hippocampus, blue. (b) Intra-rater reliability. (c) Inter-rater reliability. Analysis and reanalysis using PMOD volume-of-interest analysis process. Pearson rank analysis indicate robust correlation between two independent analysis for the cortical PBB5 absorbance intensity (a.u.). Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. suppl.material.docx Suppl.vieo1.avi Suppl.video2.avi suppl.material.docx Suppl.vieo1.avi Suppl.video2.avi Page 26/26
https://openalex.org/W2902663058	https://iris.unitn.it/bitstream/11572/220569/1/environments-05-00132.pdf	English	null	Demand-Side Management of Air-Source Heat Pump and Photovoltaic Systems for Heating Applications in the Italian Context	Environments	2,018	cc-by	9,058	Received: 17 October 2018; Accepted: 29 November 2018; Published: 6 December 2018 Abstract: Matching demand proﬁle and solar irradiance availability is necessary to meet space heating and domestic hot water needs by means of an air-source heat pump and photovoltaic system in a single-family house. Demand-side management, with smart control of the water storage set-point, is a simple but effective technique. Several studies in the literature pursue demand-side matching and self-consumption goals through system adjustments based on the model predictive control. This study proposes a rule-based control strategy, based on instantaneous photovoltaic (PV) power production, with the purpose of enhancing the self-consumption. This strategy exploits the building’s thermal capacitance as a virtual battery, and the thermal storage capacity of the system by running the heat pump to its limit when PV surplus power is available, and by eventually using an electric heater in order to reach higher temperatures. Results of annual dynamic simulations of a building and its heating system show that the proposed rule-based control strategy is able to reduce signiﬁcantly the energy exchanges between the system and the grid. Despite the enlarged renewable energy share, economic analysis points out the pursuit of the self-consumption goal may lead to a diminution of the economic advantage in the Italian context (Italian weather data and the electric power pricing scheme). Keywords: demand-side management; photovoltaic; air-source heat pump; self-consumption; net-metering environments Demand-Side Management of Air-Source Heat Pump and Photovoltaic Systems for Heating Applications in the Italian Context Elena Bee , Alessandro Prada and Paolo Baggio Elena Bee , Alessandro Prada and Paolo Baggio Department of Civil, Environmental and Mechanical Engineering, University of Trento, 38123 Trento, Italy; alessandro.prada@unitn.it (A.P.); paolo.baggio@unitn.it (P.B.) * Correspondence: elena.bee@unitn.it Elena Bee , Alessandro Prada and Paolo Baggio Department of Civil, Environmental and Mechanical Engineering, University of Trento, 38123 Trento, Italy; alessandro.prada@unitn.it (A.P.); paolo.baggio@unitn.it (P.B.) Correspondence: elena.bee@unitn.it Elena Bee , Alessandro Prada and Paolo Baggio Department of Civil, Environmental and Mechanical Engineering, University of Trento, 38123 Trento, Italy; alessandro.prada@unitn.it (A.P.); paolo.baggio@unitn.it (P.B.) Correspondence: elena.bee@unitn.it Environments 2018, 5, 132; doi:10.3390/environments5120132 1. Introduction In Europe, there is a clear long-term objective to decarbonise the energy system. The residential sector is responsible for 25.4% of the ﬁnal energy consumption in the EU28 [1]; however it also represents 63% of the potential building energy savings in 2050 [2]. Currently, the average ﬁnal energy demand in the EU28 residential stock is 160 kWh m−2 y−1 for heating, 31 kWh m−2 y−1 for domestic hot water, and 12 kWh m−2 y−1 for space cooling [3]. The share of energy consumption for space cooling is quite low, since less than 20% of the residential buildings in Europe actually meet their cooling needs, with many buildings opting to live with the discomfort of overheating rather than pay for the cost of cooling to a comfortable level [4]. An increase in the share of energy from renewable energy sources, together with more efﬁcient energy use, are therefore required to meet the transition target toward a low-carbon society. An air-to-water heat pump (AWHP) coupled with photovoltaic (PV) panels can play an important role in meeting the European targets, as evidenced by their increasing share in the European market. The AWHP represents the fastest growing heat pump segment across Europe, according to the European Heat Pump Market and Statistics Report 2015 [5]. The coefﬁcient of performance (COP) of Environments 2018, 5, 132; doi:10.3390/environments5120132 www.mdpi.com/journal/environments 2 of 12 Environments 2018, 5, 132 heat pump thermodynamic cycle depends on the operating conditions (especially on the source and the sink temperatures) and on how the partial load operation is performed. For AWHPs, the large variation of the source temperature (i.e., outdoor air) signiﬁcantly affects the COP, and consequently, the energy consumption. Nonetheless, inverter-driven (i.e., variable speed) compressors and other thermodynamic cycle enhancements allowed for performance improvements, both at partial and full load conditions. Therefore, the performance of the average commercial products has consistently improved in recent years [6]. Besides, the new optimized management of defrosting cycles allows the AWHP to operate even at low air temperatures, although with the drawback of a reduced COP. However, the increased nominal and partial load efﬁciencies of the AWHP are not enough to ensure optimal behavior of the building and its heating, ventilation, and air conditioning (HVAC) system. 1. Introduction For instance, some issues arise with regard to the control of the HVAC systems due to fast changes in energy demand of high performance buildings, and consequently, the building might be easily subject to poor comfort conditions [7]. Hence, an optimal design and control of the HVAC systems is essential to ensure the reduction of energy consumption and the achievement of thermal comfort for the entire heating season [8]. The challenge of minimizing energy use and cost while maintaining thermal comfort is always open. The complexity of the systems and their control strategies is increasing, since it also includes the efﬁcient use of renewable energy sources [9]. Moreover, the AWHPs have the potential to provide ﬂexibility to the power system, and they can reduce the grid imbalance problems due to the peaks in the PV production [10]. This is relevant in a scenario where the peaks in electricity demand are growing or where the non-dispatchable generation share is increasing [11]. Hence, the pursue of optimal self-consumption is one of the main objectives of designing the AWHP coupled with PV panels for space heating and domestic hot water preparation. In the literature, several authors have coped with the enhancement of the self-consumption rate of locally generated power [12], and storage is becoming the key aspect [13]. The water storage tank is, to a large extent, the most common tool for reducing the discrepancy between the renewable source availability and the energy demand. Nonetheless, some authors [13,14] have demonstrated that an increase in thermal storage has very limited beneﬁts during winter, and the self-consumption target can be achieved with standard tank sizes. Other authors [15] have found a negligible effect of storage size, except for large PV sizes or highly ﬂuctuating electricity prices. Schibuola et al. [16] applied three control strategies to different combinations of water storage and AWHP capacities, but the results do not show any relevant variation in the energy consumption related to water storage sizes. On the contrary, large tank sizes can reduce the system efﬁciency unless a proper control strategy is used [17]. Prada et al. [18] studied the cost-optimal mix between the water storage tank, electrical batteries, and storage capacity of the envelope [19] in a simpliﬁed residential building in three Italian cities. 1. Introduction The results of a multi-objective optimization process point out that about 60% of the optimal solutions have a volume lower than 150 L in Trento and Rome, and lower than 100 L in Palermo. The necessity of a more advanced control system to maximize PV self-consumption emerges from these studies, and some attempts have been made in the literature. Psimopoulos et al. [20] used weather forecasts to take rule-based decisions on the system operation. Similarly, Thygesen and Karlsson [14] proposed a controller of the tank set point based on the radiation forecast for a ground source heat pump in the Swedish climate. However, such a controller proved to be economically unproﬁtable. Pospisil et al. [21] proposed a predictive control, to operate the AWHP during periods with the highest outdoor temperature. Henze et al. [22] used the time-of-use price signal to shift the electrical loads to off-peak periods at night and weekends in cooling applications. An extensive review by Péan et al. [23] clearly distinguishes between rule-based controls and model predictive controls. Almost all of the cases analyzed in the literature belong to one of these two categories. Most of these strategies, however, require advanced control systems that make them more suitable for new installations. On the contrary, this study proposed a simple rule-based strategy that can be implemented through a low-cost controller. The simple, rule-based (i.e., if/then) control logic is based on instantaneous measurements of PV production, in order to exploit the storage tank and building 3 of 12 Environments 2018, 5, 132 capacity with a few changes to the HVAC management system. Nevertheless, the simplicity of the rule-based control algorithm could undermine its ability to produce near-optimal control strategies. For this reason, the beneﬁts of the proposed strategy for increasing PV self-consumption and at the same time reducing the energy drawn from the grid are analyzed. A coupled simulation of a single-family house and its energy system was set up in the TRNSYS simulation suite. Standard and TESS libraries [24] were used to model the building and the HVAC components (e.g., storage tank, PV modules, AWHP, and the control system). A TRNSYS type able to simulate variable-speed units was developed by the authors, in order to correctly model the partial load operation of the AWHP. Finally, the study focuses on the Italian context and on the analysis of the HVAC running costs. 1. Introduction For this reason, the Italian net metering tariffs for PV systems were implemented in the simulation code. 2. Methods This study proposes and tests two control strategies for demand-side management, based on instantaneous PV power production by means of dynamic simulations. A coupled simulation model of a single-family house, and the energy systems for heating (SH) and domestic hot water production (DHW), was set up by means of the TRNSYS software, using standard and TESS libraries. Annual simulations were run with a time-step of 1 min. The Italian electricity tariffs for the year 2017 and the typical reference year for Milan (a city having a 4A climate, according to ASHRAE 90.1 classiﬁcation [25]) were used as boundary conditions. Weather data included hourly proﬁles of dry bulb air temperature, relative humidity, global radiation, and wind speed. However, the climate can affect signiﬁcantly the ASWHP performance [26]. This is an initial study focused on Northern Italy, and for this reason, does not investigate the extent to which the weather data can affect the control strategy, even if it signiﬁcantly inﬂuences the PV production and the energy performance of the AWHP [26]. 2.1. Building Model The building is a small, bi-level, single-family house, having a volume of 275 m3 and façades oriented towards the main cardinal directions. The windows are exposed to south, east, and west. The choice of this building is due to a previous modelling work of a real high-performance prototype building [26]. The walls and roof have a thermal transmittance equal to 0.25 W m−2 K−1, and windows have a thermal transmittance equal to 0.9 W m−2 K−1. The ventilation rate of the building is 0.5 ACH (air change per hour), according to Italian technical speciﬁcation UNI/TS 11300-1 [27], and ventilation is performed by a mechanical ventilation system with a heat recovery exchanger. The building is representative of a European family house, since the ﬂoor area (about 80 m2) is quite close to the average European ﬂoor area of a residential dwelling (i.e., 84 m2) [28]. The building modeled is divided into four thermal zones, three of which are equipped with radiant ﬂoor panels. A small mechanical room is without a heating terminal, but has internal heat gains due to the tank thermal losses. A weekly pattern represents the total internal gain due to people and appliances, with two different proﬁles for the living room/kitchen and for the bedrooms. The proﬁles are taken from the Italian technical speciﬁcation UNI TS 11300-1 [27]. The building model and the input/output management is performed by means of Type 56 (standard TRNSYS library). 2.2. Heating, Ventilation, and Air Conditioning System and the Domestic Hot Water Production Model The heating system was based on a variable-speed AWHP, coupled with radiant ﬂoor panels (Figure 1). The AWHP simulation model was based on the performance map and on the partial-load performance function provided by the manufacturer. A high-performance unit with a COP value of 4.5 at 7–35 ◦C and a nominal heating capacity of 5 kW is considered in this study. As usual with high-performance units, the partial-load operation performance curve is such that the optimal AWHP performance happens when it is operating at around 50% of the load, and the maximum COP is up to 30% higher than the COP at the nominal capacity. The compressor speed depends on the control 4 of 12 AWHP P is up Environments 2018, 5, 132 performance units, t performance happen strategy, and it takes different variables as input. In the reference case, for the SH mode the speed linearly depended on the distance of the temperature from the set-point in the SH tank, while for the DHW mode the AWHP was always run at the maximum speed. Cross-linked polyethylene (PEX) pipes were used for the radiant panels, with a diameter of 0.016 m, pipe spacing of 0.12, and thermal conductivity of 0.44 W m−1 K−1. An on-off temperature controller in each room controlled the ambient temperature, with a dead-band of ±0.5 ◦C. Two separate water storage tanks were used for SH and DHW, and a temperature controller (with dead-band of 3 ◦C around the set-point) controls their temperatures. The set-point temperature of the SH tank is reset based on the outdoor temperature, and it varies linearly between 38 ◦C and 28 ◦C as the outdoor air temperature varies between −5 ◦C and 15 ◦C. The set-point temperature of the DHW tank was set to 45 ◦C in the reference case. The water storage tanks were modeled as stratiﬁed vertical cylindrical tanks, with a volume of 150 L for space heating and 250 L for DHW production. Priority is given to DHW demand, which is modeled through the M standard tapping proﬁle according to the EN 16147 [29]. A mechanical ventilation system with heat recovery (with a nominal efﬁciency of 75%) guarantees a ventilation rate of 0.5 ACH in each room. A typical commercial PV module was used in the simulation (Table 1). Since the PV production greatly affects the control efﬁciency, the PV area is a parametric value in this study. 2.2. Heating, Ventilation, and Air Conditioning System and the Domestic Hot Water Production Model Hence, three different PV areas are considered, i.e., 15 m2 (PV15), 20 m2 (PV20), or 30 m2 (PV30), corresponding, respectively, to the peak powers of 2.1 kW, 2.8 kW, and 4.2 kW. The PV electric power production, when available, is used by the AWHP, and if necessary, additional electric power is taken from the grid. PV surplus power is delivered to the grid, and the model assumes a conversion efﬁciency (DC/AC inverter) of 90%. strategy, and it takes different variables as input. In the reference case, for the SH mode the speed linearly depended on the distance of the temperature from the set-point in the SH tank, while for the DHW mode the AWHP was always run at the maximum speed. Cross-linked polyethylene (PEX) pipes were used for the radiant panels, with a diameter of 0.016 m, pipe spacing of 0.12, and thermal conductivity of 0.44 W m−1 K−1. An on-off temperature controller in each room controlled the ambient temperature, with a dead-band of ±0.5 °C. Two separate water storage tanks were used for SH and DHW, and a temperature controller (with dead-band of 3 °C around the set-point) controls their temperatures. The set-point temperature of the SH tank is reset based on the outdoor temperature, and it varies linearly between 38 °C and 28 °C as the outdoor air temperature varies between −5 °C and 15 °C. The set-point temperature of the DHW tank was set to 45 °C in the reference case. The water storage tanks were modeled as stratified vertical cylindrical tanks, with a volume of 150 liters for space heating and 250 liters for DHW production. Priority is given to DHW demand, which is modeled through the M standard tapping profile according to the EN 16147 [29]. A mechanical ventilation system with heat recovery (with a nominal efficiency of 75%) guarantees a ventilation rate of 0.5 ACH in each room. A typical commercial PV module was used in the simulation (Table 1). Since the PV production greatly affects the control efficiency, the PV area is a parametric value in this study. Hence, three different PV areas are considered, i.e., 15 m2 (PV15), 20 m2 (PV20), or 30 m2 (PV30), corresponding, respectively, to the peak powers of 2.1 kW, 2.8 kW, and 4.2 kW. The PV electric power production, when available, is used by the AWHP, and if necessary, additional electric power is taken from the grid. 2.2. Heating, Ventilation, and Air Conditioning System and the Domestic Hot Water Production Model PV surplus power is delivered to the grid, and the model assumes a conversion efficiency (DC/AC inverter) of 90%. Figure 1. Heating system layout. Figure 1. Heating system layout. Figure 1. Heating system layout. Figure 1. Heating system layout. Table 1. Technical specifics of the photovoltaic (PV) modules. Table 1. Technical speciﬁcs of the photovoltaic (PV) modules. p p ( ) Module Specifics Value Units Area 1.6 m2 Power Rating at Standard Test Conditions 230 W Efficiency 14.1% - Number of cells 60 - Maximum power point current 7.8 A Maximum power point voltage 29.5 V Short circuit current 8.4 A Open circuit voltage 37 V Tilt angle 10 deg Module Speciﬁcs Value Units Area 1.6 m2 Power Rating at Standard Test Conditions 230 W Efﬁciency 14.1% - Number of cells 60 - Maximum power point current 7.8 A Maximum power point voltage 29.5 V Short circuit current 8.4 A Open circuit voltage 37 V Tilt angle 10 deg 2.3. Economics The electricity market is set up to match energy supply and demand in Italy. For this reason, there is trading between generators and suppliers in the Italian Power Exchange. The increasingly 5 of 12 eason Environments 2018, 5, 132 2.3. Economics Th l i i widespread diffuse electricity production and the growing number of heat pump installations are altering the grid balance and in order to face those changes, the electricity marked is evolving. Italian energy providers have recently introduced a time-of-use tariff for the domestic users that does not penalize consumers with large annual electricity needs. This rate will promote the adoption of AWHPs for space heating or cooling. The time-of-use tariff has two electricity prices that change every three months (Figure 2a). During the weekdays, the F1 price is used from 8:00 AM to 7:00 PM, while the F23 price is available from 7:00 PM to 8:00 AM. The F23 is also applied for the weekends. The annual electricity bill also includes ﬁxed costs (about 159 €/y for a committed power of 4.5 kW) and 13% of fees (in 2017) [30]. The daily proﬁles of the time-of-use tariff, including transport/management and system charges, are shown in details in Figure 2a. Figure 2b represents instead a typical proﬁle of the real price on the market (unique national price; PUN) that reﬂects the unbalance of electricity exchanges between the grid and the consumers/producers. there is trading between generators and suppliers in the Italian Power Exchange. The increasingly widespread diffuse electricity production and the growing number of heat pump installations are altering the grid balance and in order to face those changes, the electricity marked is evolving. Italian energy providers have recently introduced a time-of-use tariff for the domestic users that does not penalize consumers with large annual electricity needs. This rate will promote the adoption of AWHPs for space heating or cooling. The time-of-use tariff has two electricity prices that change every three months (Figure 2a). During the weekdays, the F1 price is used from 8:00 AM to 7:00 PM, while the F23 price is available from 7:00 PM to 8:00 AM. The F23 is also applied for the weekends. The annual electricity bill also includes fixed costs (about 159 €/y for a committed power of 4.5 kW) and 13% of fees (in 2017) [30]. 2.3. Economics The daily profiles of the time-of-use tariff, including transport/management and system charges, are shown in details in Figure 2a. Figure 2b represents instead a typical profile of the real price on the market (unique national price; PUN) that reflects the unbalance of electricity exchanges between the grid and the consumers/producers. (a) (b) Figure 2. (a) Italian electricity time-of-use tariffs for domestic users in 2017. Prices include transport/management and system charges and annual fixed costs, but fees are not included. (b) Unique national price (PUN) on a weekday for the year 2017 with the typical double-peak profile. Figure 2. (a) Italian electricity time-of-use tariffs for domestic users in 2017. Prices include transport/management and system charges and annual ﬁxed costs, but fees are not included. (b) Unique national price (PUN) on a weekday for the year 2017 with the typical double-peak proﬁle. (b) (a) (a) (b) Figure 2. (a) Italian electricity time-of-use tariffs for domestic users in 2017. Prices include transport/management and system charges and annual fixed costs, but fees are not included. (b) Unique national price (PUN) on a weekday for the year 2017 with the typical double-peak profile. Figure 2. (a) Italian electricity time-of-use tariffs for domestic users in 2017. Prices include transport/management and system charges and annual ﬁxed costs, but fees are not included. (b) Unique national price (PUN) on a weekday for the year 2017 with the typical double-peak proﬁle. A separate national net-metering scheme rewards and incentivizes power production from renewable energy sources. In the Italian net-metering scheme, the power generated by eligible on- site renewable plants can be delivered to the grid and used to offset the electricity withdrawn from the grid. The grid authority pays a contribution based on the balance and the economic value of power injections and withdrawals in a given calendar year. The contribution amount, calculated with Equation (2), is based on the energy exchanges with the grid (𝐸௚௥௜ௗ, 𝐸௚௥௜ௗ) and some economic indices of the national market, such as the day ahead of auction market (MGP), the unique national price (PUN), and the price of the flat-rate exchange (𝐶𝑈௦௙) [31,32]. A separate national net-metering scheme rewards and incentivizes power production from renewable energy sources. In the Italian net-metering scheme, the power generated by eligible on-site renewable plants can be delivered to the grid and used to offset the electricity withdrawn from the grid. 2.3. Economics The grid authority pays a contribution based on the balance and the economic value of power injections and withdrawals in a given calendar year. The contribution amount, calculated with Equation (2), is based on the energy exchanges with the grid (Egrid, Egrid) and some economic indices of the national market, such as the day ahead of auction market (MGP), the unique national price (PUN), and the price of the ﬂat-rate exchange (CUs f ) [31,32]. ( U ), a e p i e o e a a e e a ge ( ௦௙) [ , ] The annual economic balance is defined in Equation (1), where 𝐵𝑖𝑙𝑙 is the cost of purchased electricity, 𝐶𝑠 and 𝐸𝑥𝑐 are the contributions, defined respectively with Equations (2) and (3), and the 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 is the amount actually paid at the end of the year by the householder. The economic credit (𝐸𝑥𝑐) is granted only if the economic value of the electricity fed to the grid is higher than the economic value of the withdrawn power, according to Equation (3). f The annual economic balance is deﬁned in Equation (1), where Bill is the cost of purchased electricity, Cs and Exc are the contributions, deﬁned respectively with Equations (2) and (3), and the Netbill is the amount actually paid at the end of the year by the householder. The economic credit (Exc) is granted only if the economic value of the electricity fed to the grid is higher than the economic value of the withdrawn power, according to Equation (3). 𝑖𝑙𝑙−𝐶௦−𝐸𝑥𝑐 (1) ௥௢௠ ௚௥௜ௗ∙𝑃𝑈𝑁; 𝐸௧௢ ௚௥௜ௗ∙𝑀𝐺𝑃) + 𝐶𝑈௦௙∙min (𝐸௙௥௢௠ ௚௥௜ௗ, 𝐸௧௢ ௚௥௜ௗ) (2) 0; 𝐸 𝑀𝐺𝑃 𝐸 𝑃𝑈𝑁) (3) Net bill = Bill −Cs −Exc (1) Cs = min(Ef rom grid·PUN; Eto grid·MGP) + CUs f ·min Ef rom grid, Eto grid (2) Exc = max 0; Eto grid·MGP −Ef rom grid·PUN (3) (1) (2) ௧௢ ௚௥௜ௗ ) ௦௙ ( ௙௥௢௠ ௚௥௜ௗ ௧௢ ௚௥௜ௗ) 𝐺𝑃 𝐸 𝑃𝑈𝑁) (3) Exc = max 0; Eto grid·MGP −Ef rom grid·PUN (3) (3) 𝐸𝑥𝑐 max (0; 𝐸௧௢ ௚௥௜ௗ𝑀𝐺𝑃 2.4. Control System Strategy 𝐸𝑥𝑐 max (0; 𝐸௧௢ ௚௥௜ௗ𝑀𝐺𝑃 2.4. Control System Strategy The proposed control strategy aims to increase self-consumption by changing the set-point temperatures as a function of the actual PV generation. In a previous work [30], different algorithms, 6 of 12 Environments 2018, 5, 132 based on either the PV production or the outdoor temperature, were compared to each other. The authors found that the temperature has a limited impact on the control reliability with respect to the control, based on the PV production. In the previous study, the algorithm checks whether PV power is still available after standard set-points are met. If this is the case, the algorithm increases the set-points of the DHW tank, the SH tank, and ambient thermostats and, consequently, it runs the AWHP to satisfy the new set-points. In addition, the set-points are slightly reduced in case of low or absent PV power. p In this study, we propose a new control strategy (HP+) as a further development of the algorithm, based on PV production [33]. The new algorithm ﬁrst checks the AWHP capacity, and in this respect, the available PV power is compared with the minimum and the maximum power absorption of the AWHP. The increase in set-points is therefore calculated as a function of this difference. The maximum set-point variation is applied in the case when PV production is greater than the power absorption at the maximum AWHP speed. Otherwise, only a fraction of the maximum set-point variation is applied. The set-points are only increased and not reduced, with the purpose of avoiding any possible discomfort (even if acceptable). Besides, the algorithm comparability with the reference case is ensured since the same indoor conditions are pursued. The maximum set-points are 55 ◦C for the water tanks (the maximum temperature supply of the AWHP) and 22 ◦C for the air ambient temperature. Moreover, another rule-based control scheme is implemented. This new function checks whether PV power is still available after the maximum set-point temperature is reached in the tanks. In this case, the excess power is used to heat the DHW storage to a higher temperature (up to 90 ◦C for safety reasons) by means of an electric heater (two subsequent stages of 500 W). This strategy maximizes the self-consumption of PV production, although with the drawback of a high exergetic cost of the electric resistance, which affects the overall efﬁciency of the system. The function is called electric heater plus (EH+). 3. Results 3. Results The results are presented as a comparison between the reference case, in which the system is controlled using the standard approach, and the cases in which the new control strategies (Section 2.4) are applied. The results are presented as a comparison between the reference case, in which the system is controlled using the standard approach, and the cases in which the new control strategies (section 2.4) are applied. Energy use, self-consumption (direct use of the energy generated by PV panels), and energy exchanged with the grid, are presented on a monthly or annual basis. Bills and contributions are evaluated on an annual basis. Energy use, self-consumption (direct use of the energy generated by PV panels), and energy exchanged with the grid, are presented on a monthly or annual basis. Bills and contributions are evaluated on an annual basis. In the reference case, the annual SCratio is 7%, and the remaining energy from the PV system is delivered to the grid. Such a low value is due to the conﬁguration of the model, where no cooling applications or domestic appliances are taken into account, as explained in the introduction. Only the DHW needs are covered in the summer months, and consequently, a large amount of surplus energy is produced. Figure 3 includes some signiﬁcant results about the simulation of the reference case. On the left, the monthly energy exchanged with the grid is represented, and it is evident that the energy to the grid exceeds the energy from the grid, especially in summer. On the right, annual costs/incomes for the householder on the base of the net-metering scheme indicate that the contribution considerably reduces the Netbill. In the reference case, the annual 𝑆𝐶௥௔௧௜௢ is 7%, and the remaining energy from the PV system is delivered to the grid. Such a low value is due to the configuration of the model, where no cooling applications or domestic appliances are taken into account, as explained in the introduction. Only the DHW needs are covered in the summer months, and consequently, a large amount of surplus energy is produced. Figure 3 includes some significant results about the simulation of the reference case. On the left, the monthly energy exchanged with the grid is represented, and it is evident that the energy to the grid exceeds the energy from the grid, especially in summer. 𝐸𝑥𝑐 max (0; 𝐸௧௢ ௚௥௜ௗ𝑀𝐺𝑃 2.4. Control System Strategy The two functions are applied either independently or in combination (HP+ and EH+), but with HP+ priority. Tset,HP+ = Tset,re f + Tset,max −Tset,re f Pabs,max −Pabs,min ·max 0; PPV,surplus −Pabs,min (4) (4) Tset,EH+ = Tmax,DHW·max 0; PPV,surplus −Pabs,max ·max 0; Ttank −Tset,re f (5) (5) Notice that a temperature-controlled mixing valve is installed, both on the DHW and the SH supply system. This secondary control, connected to a three-way valve, mixes the water exiting from the tanks with cold water in order to meet the supply water temperature (i.e., 45 ◦C for DHW and depending on the outdoor reset control for the SH). The different level of self-consumption is quantiﬁed by two indices. These two key ﬁgures are the self-consumption ratio (SCratio) and the self-sufﬁciency ratio (SSratio), which are detailed in Equations (6) and (7), respectively. SCratio = Sel f consumption[kWh] PVproduction[kWh] (6) SSratio = Sel f consumption[kWh] Totalconsumption[kWh] (7) (6) (6) (7) The two indices quantify the optimality of the control in a different way. The self-consumption ratio evaluates the portion of energy produced by the PV that is directly consumed, while the self-sufﬁcient ratio quantiﬁes the share of total energy consumption for SH and DHW produced by using the PV source. 7 of 12 7 of 12 7 of 12 7 of 12 Environments 2018, 5, 132 Environments 2018 5 x FO 3. Results 3. Results On the right, annual costs/incomes for the householder on the base of the net-metering scheme indicate that the contribution considerably reduces the 𝑁𝑒𝑡𝑏𝑖𝑙𝑙. Monthly energy exchanges with the grid for the reference case, and for the cases with the new control algorithms (PV area of 20 m2). Figure 4. Monthly energy exchanges with the grid for the reference case, and for the cases with the new control algorithms (PV area of 20 m2). Figure 4. Monthly energy exchanges with the grid for the reference case, and for the cases with the new control algorithms (PV area of 20 m2). The reduction of the energy exchanges is also evident from Figure 5, in which the increase of self-consumption is also shown on the annual balance. This increase is quantified as 167% for HP+, 660% for EH+, and 729% for HP+ and EH+, with respect to the reference case. Note that the self- consumption exceeds the energy drawn from the grid when the EH+ or the HP+ combined with EH+ algorithms are adopted. The reduction of the energy exchanges is also evident from Figure 5, in which the increase of self-consumption is also shown on the annual balance. This increase is quantiﬁed as 167% for HP+, 660% for EH+, and 729% for HP+ and EH+, with respect to the reference case. Note that the self-consumption exceeds the energy drawn from the grid when the EH+ or the HP+ combined with EH+ algorithms are adopted. The reduction of the energy exchanges is also evident from Figure 5, in which the increase of self-consumption is also shown on the annual balance. This increase is quantified as 167% for HP+, 660% for EH+, and 729% for HP+ and EH+, with respect to the reference case. Note that the self- consumption exceeds the energy drawn from the grid when the EH+ or the HP+ combined with EH+ algorithms are adopted. Figure 5. Annual amounts of energy exchanged with grid and self-consumed for the reference case and for the cases with the new control strategies (PV area of 20 m2). Figure 5. Annual amounts of energy exchanged with grid and self-consumed for the reference case and for the cases with the new control strategies (PV area of 20 m2). Figure 5. Annual amounts of energy exchanged with grid and self-consumed for the reference case and for the cases with the new control strategies (PV area of 20 m2). Figure 5. Figure 3. Reference case: monthly energy exchanges with the grid and annual bills/takings for the householder with net-metering (Equations (1)–(3)). Figure 3. Reference case: monthly energy exchanges with the grid and annual bills/takings for the householder with net-metering (Equations (1)–(3)). Figure 3. Reference case: monthly energy exchanges with the grid and annual bills/takings for the householder with net-metering (Equations (1)–(3)). Figure 3. Reference case: monthly energy exchanges with the grid and annual bills/takings for the householder with net-metering (Equations (1)–(3)). Figure 4 compares the three simulations with the new control strategies to each other and to the reference case, in terms of monthly energy fluxes. All of the algorithms applied have the effect of reducing the exchanges with the grid (energy drawn from the grid and energy delivered to the grid). In other words, the algorithms improve the self-consumption and the self-sufficiency, especially when the HP+ and EH+ are combined together. In this case, the 𝑆𝐶௥௔௧௜௢ increases from 7% to 60%, and the 𝑆𝑆௥௔௧௜௢ from 12% to 65%. This result therefore highlights the potential benefit for the grid manager if this algorithm were extensively applied in buildings with PV systems. Figure 4 compares the three simulations with the new control strategies to each other and to the reference case, in terms of monthly energy ﬂuxes. All of the algorithms applied have the effect of reducing the exchanges with the grid (energy drawn from the grid and energy delivered to the grid). In other words, the algorithms improve the self-consumption and the self-sufﬁciency, especially when the HP+ and EH+ are combined together. In this case, the SCratio increases from 7% to 60%, and the SSratio from 12% to 65%. This result therefore highlights the potential beneﬁt for the grid manager if this algorithm were extensively applied in buildings with PV systems. 8 of 12 Environments 2018, 5, 132 Figure 4. Monthly energy exchanges with the grid for the reference case, and for the cases with the new control algorithms (PV area of 20 m2). Figure 4. Monthly energy exchanges with the grid for the reference case, and for the cases with the new control algorithms (PV area of 20 m2). Figure 4. Monthly energy exchanges with the grid for the reference case, and for the cases with the new control algorithms (PV area of 20 m2). Figure 4. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). Figure 6. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). Figure 6. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2) Figure 6. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). Figure 6. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). Figure 6. Annual bills/takings for the householder on the base of the net-metering scheme (Equa (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2) g g g q (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). Finally, Figures 7 and 8 show the influence of the PV array area (or aperture) on the energy exchanges and costs. Figure 7 clearly shows how, passing from 15 m2 to 30 m2, the energy withdrawn from the grid obviously decreases, and the energy fed into the grid increases. With a PV area of 15 m2, the energy withdrawn exceeds the energy delivered to the grid. The 𝑆𝑆௥௔௧௜௢ increases from 65% (PV15) to 58% (PV20) and 72% (PV30) by oversizing the area, while the 𝑆𝐶௥௔௧௜௢ increases from 60% (PV15) to 65% (PV20) but then decreases to 48% (PV30) Finally, Figures 7 and 8 show the inﬂuence of the PV array area (or aperture) on the energy exchanges and costs. Figure 7 clearly shows how, passing from 15 m2 to 30 m2, the energy withdrawn from the grid obviously decreases, and the energy fed into the grid increases. With a PV area of 15 m2, the energy withdrawn exceeds the energy delivered to the grid. The SSratio increases from 65% (PV15) to 58% (PV20) and 72% (PV30) by oversizing the area, while the SCratio increases from 60% (PV15) to 65% (PV20), but then decreases to 48% (PV30). Finally, Figures 7 and 8 show the influence of the PV array area (or aperture) on the energy exchanges and costs. Figure 7 clearly shows how, passing from 15 m2 to 30 m2, the energy withdrawn from the grid obviously decreases, and the energy fed into the grid increases. With a PV area of 15 m2, the energy withdrawn exceeds the energy delivered to the grid. The 𝑆𝑆௥௔௧௜௢ increases from 65% (PV15) to 58% (PV20) and 72% (PV30) by oversizing the area, while the 𝑆𝐶௥௔௧௜௢ increases from 60% (PV15) to 65% (PV20), but then decreases to 48% (PV30). Figure 7. Monthly energy exchanges with the grid for the case with control strategy HP+ and EH+, d f PV f 15 20 d 30 2 Figure 7. Monthly energy exchanges with the grid for the case with control strategy HP+ and EH+, and for a PV area of 15, 20, and 30 m2. Figure 7. Monthly energy exchanges with the grid for the case with control strategy HP+ and EH+, and for a PV area of 15, 20, and 30 m2. Figure 7. Monthly energy exchanges with the grid for the case with control strategy HP+ and EH+, Figure 7. Monthly energy exchanges with the grid for the case with control strategy HP+ and EH+, and for a PV area of 15, 20, and 30 m2. Figure 7. Monthly energy exchanges with the grid for the case with control strategy HP+ and EH+, and for a PV area of 15, 20, and 30 m2. , , Figure 8 shows the different effects of the contribution to the 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 from an economic point of view. In both the case with HP+ only and with HP+ and EH+, the 𝐸𝑐𝑥 term is equal to zero, and the 𝐶௦ term is lower for the smaller area size; the resulting 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 penalizes this situation. Therefore, an oversized system is more profitable, compared to one that is more suitable to the actual energy needs of the building by considering the running costs. Nonetheless, the different installation costs and the Figure 8 shows the different effects of the contribution to the 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 from an economic point of view. Annual amounts of energy exchanged with grid and self-consumed for the reference case and for the cases with the new control strategies (PV area of 20 m2). Figure 5. Annual amounts of energy exchanged with grid and self-consumed for the reference case and for the cases with the new control strategies (PV area of 20 m2). Figure 5. Annual amounts of energy exchanged with grid and self-consumed for the reference case and for the cases with the new control strategies (PV area of 20 m2). Figure 6 compares the three simulations with the new control strategies to one another and to the reference case, in economic terms. The new algorithms are not advantageous in economic terms, although the energy taken from the network is greatly reduced. The resulting 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 increases because the contribution (𝐶௦) decreases for higher self-consumption levels. That means that these control strategies aimed at increasing the self-consumption are not favored by the Italian net- metering scheme, which does not adequately reward such behavior. Figure 6 compares the three simulations with the new control strategies to one another and to the reference case, in economic terms. The new algorithms are not advantageous in economic terms, although the energy taken from the network is greatly reduced. The resulting 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 increases because the contribution (𝐶௦) decreases for higher self-consumption levels. That means that these control strategies aimed at increasing the self-consumption are not favored by the Italian net- metering scheme, which does not adequately reward such behavior. Figure 6 compares the three simulations with the new control strategies to one another and to the reference case, in economic terms. The new algorithms are not advantageous in economic terms, although the energy taken from the network is greatly reduced. The resulting Netbill increases because the contribution (Cs) decreases for higher self-consumption levels. That means that these control strategies aimed at increasing the self-consumption are not favored by the Italian net-metering scheme, which does not adequately reward such behavior. 9 of 12 9 of 12 Environments 2018, 5, 132 Environments 2018 5 x FO vironments 2018, 5, x FOR PEER REVIEW 9 of 1 Figure 6. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the reference case and the cases with control strategies applied (PV area of 20 m2). Figure 6. 4 Conclusions 4. Conclusions Householders who take part in the Italian net-metering scheme receive a contribution that significantly reduces their net annual spending for electricity. The contribution boosts the installation of PV plants helping to meet the goal of the progressive de carbonization of the dwelling sector Householders who take part in the Italian net-metering scheme receive a contribution that signiﬁcantly reduces their net annual spending for electricity. The contribution boosts the installation of PV plants, helping to meet the goal of the progressive de-carbonization of the dwelling sector. of PV plants, helping to meet the goal of the progressive de-carbonization of the dwelling sector. The regular bill (time-of-use tariff) linearly depends on the purchased electricity, and decreases when a control strategy oriented to self-consumption is applied. Despite the benefits of those control strategies, in terms of self-consumption and energy exchanges, the Italian net-metering contribution does not boost the optimal use of the self-produced electricity in terms of net cost, resulting in larger net money savings for management of the system with larger energy exchanges with the national grid. In particular, if the HP+ and EH+ strategies are applied together, the energy delivered to the g id a d d a f o the g id dec ease by 56% a d 36% especti ely but the et bill i c eases by The regular bill (time-of-use tariff) linearly depends on the purchased electricity, and decreases when a control strategy oriented to self-consumption is applied. Despite the beneﬁts of those control strategies, in terms of self-consumption and energy exchanges, the Italian net-metering contribution does not boost the optimal use of the self-produced electricity in terms of net cost, resulting in larger net money savings for management of the system with larger energy exchanges with the national grid. In particular, if the HP+ and EH+ strategies are applied together, the energy delivered to the grid and drawn from the grid decrease by 56% and 36%, respectively, but the net bill increases by 34%. grid and drawn from the grid decrease by 56% and 36%, respectively, but the net bill increases by 34%. A threshold for the PV array area, below which the energy withdrawn exceeds the energy delivered to the grid, can be identified for a particular system configuration and energy use. The economic advantages are obviously lower for smaller areas, although more suitable for the building energy demand. In both the case with HP+ only and with HP+ and EH+, the 𝐸𝑐𝑥 term is equal to zero, and the 𝐶௦ term is lower for the smaller area size; the resulting 𝑁𝑒𝑡𝑏𝑖𝑙𝑙 penalizes this situation. Therefore, an oversized system is more profitable, compared to one that is more suitable to the actual energy needs of the building by considering the running costs. Nonetheless, the different installation costs and the payback times should also be considered. Figure 8 shows the different effects of the contribution to the Netbill from an economic point of view. In both the case with HP+ only and with HP+ and EH+, the Ecx term is equal to zero, and the Cs term is lower for the smaller area size; the resulting Netbill penalizes this situation. Therefore, an oversized system is more proﬁtable, compared to one that is more suitable to the actual energy needs of the building by considering the running costs. Nonetheless, the different installation costs and the payback times should also be considered. 10 of 12 Environments 2018, 5, 132 nvironments 2018, 5, x FOR PEER REVIEW 10 of Figure 8. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the case with control strategy HP+ only (above) and HP+ EH+ (below), as well as for a PV area of 15, 20, and 30 m2. Figure 8. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the case with control strategy HP+ only (above) and HP+ EH+ (below), as well as for a PV area of 15, 20, and 30 m2. , , Figure 8. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the case with control strategy HP+ only (above) and HP+ EH+ (below), as well as for a PV area of 15, 20, and 30 m2. Figure 8. Annual bills/takings for the householder on the base of the net-metering scheme (Equations (1)–(3)) for the case with control strategy HP+ only (above) and HP+ EH+ (below), as well as for a PV area of 15, 20, and 30 m2. Acknowledgments: The authors want to thank INNO collaboration in analyzing the heat pump operation Conﬂicts of Interest: The authors declare no conﬂict of interest. 4 Conclusions 4. Conclusions It worth noting, however, that a correct sizing procedure would require taking into account other electrical uses that are not included in this study, and a comprehensive economic analysis should include the initial investment maintenance and replacement costs of the PV system A threshold for the PV array area, below which the energy withdrawn exceeds the energy delivered to the grid, can be identiﬁed for a particular system conﬁguration and energy use. The economic advantages are obviously lower for smaller areas, although more suitable for the building energy demand. It worth noting, however, that a correct sizing procedure would require taking into account other electrical uses that are not included in this study, and a comprehensive economic analysis should include the initial investment, maintenance, and replacement costs of the PV system. Author Contributions: Elena Bee and Alessandro Prada defined the methods, ran simulation and analyzed the results. Paolo Baggio identified the aims of the research and supervised the study. All Author Contributions: E.B. and A.P. deﬁned the methods, ran simulation and analyzed the results. P.B. identiﬁed the aims of the research and supervised the study. All authors contributed in writing, editing, and structuring the paper. y gg p y authors contributed in writing, editing, and structuring the paper. Acknowledgments: The authors want to thank INNOVA Renewing Energies for the fruitful collaboration in analyzing the heat pump operation. Acknowledgments: The authors want to thank INNO collaboration in analyzing the heat pump operation Conﬂicts of Interest: The authors declare no conﬂict of interest. 11 of 12 Environments 2018, 5, 132 References 1. Eurostat. Complete Energy Balances—Annual Data 2016. Available online: https://ec.europa.eu/eurostat/ statistics-explained/index.php (accessed on 1 October 2018). 1. Eurostat. Complete Energy Balances—Annual Data 2016. Available online: https://ec.europa.eu/eurostat/ statistics-explained/index.php (accessed on 1 October 2018). 2. IEA. Energy Technology Perspectives 2010—Scenarios & Strategies to 2050; IEA: Paris, France, 2010. 3. HotMaps. Hotmaps Report D2.3 WP2 Open Dataset for the EU28; HotMaps: Vienna, Austria, 2018. [CrossR 3. HotMaps. Hotmaps Report D2.3 WP2 Open Dataset for the EU28; HotMaps: Vienna, Austria, 2018. [CrossRef] 4. Stratego. Low Carbon Heating and Cooling Strategies for Europe. Final Report of the EU Project “Stratego Enhanced Heating & Cooling Plans”. Available online: http://stratego-project.eu/reports/ (accessed on 1 October 2018). 5. Nowak, T.; Westring, P. European Heat Pump Market and Statistics Report 2016; Technical report; The European Heat Pump Association AISBL (EHPA): Brussels, Belgium, 2016. 6. Chua, K.J.; Chou, S.K.; Yang, W.M. Advances in heat pump systems: A review. Appl. Energy 2010, 87, 3611–3624. [CrossRef] 7. Penna, P.; Prada, A.; Cappelletti, F.; Gasparella, A. Multi-objectives optimization of energy efﬁciency measures in existing buildings. Energy Build. 2015, 95, 57–69. [CrossRef] 8. Carlon, E.; Schwarz, M.; Prada, A.; Golicza, L.; Verma, V.K.; Baratieri, M.; Gasparella, A.; Haslinger, W.; Schmidl, C. On-site monitoring and dynamic simulation of a low energy house heated by a pellet boiler. Energy Build. 2016, 116, 296–306. [CrossRef] 9. Fabrizio, E.; Seguro, F.; Filippi, M. Integrated HVAC and DHW production systems for Zero Energy Buildings. Renew. Sustain. Energy Rev. 2014, 40, 515–541. [CrossRef] 10. Fischer, D.; Madani, H. On heat pumps in smart grids: A review. Renew. Sustain. Energy Rev. 2017, 70, 342–357. [CrossRef] 11. Arteconi, A.; Hewitt, N.J.; Polonara, F. State of the art of thermal storage for demand-side management. Appl. Energy 2012, 93, 371–389. [CrossRef] 12. Luthander, R.; Widén, J.; Nilsson, D.; Palm, J. Photovoltaic self-consumption in buildings: A review. Appl. Energy 2015, 142, 80–94. [CrossRef] 13. Fischer, D.; Rautenberg, F.; Wirtz, T.; Wille-Haussmann, B.; Madani, H. Smart meter enabled control for variable speed heat pumps to increase PV self-consumption. In Proceedings of the 24th IIR, Yokohama, Japan, 16–22 August 2015. 14. Thygesen, R.; Karlsson, B. Simulation of a proposed novel weather forecast control for ground source heat pumps as a mean to evaluate the feasibility of forecast controls’ inﬂuence on the photovoltaic electricity self-consumption. Appl. Energy 2016, 164, 579–589. [CrossRef] 15. Arteconi, A.; Hewitt, N.J.; Polonara, F. References Domestic demand-side management (DSM): Role of heat pumps and thermal energy storage (TES) systems. Appl. Therm. Eng. 2013, 51, 155–165. [CrossRef] 6. Schibuola, L.; Scarpa, M.; Tambani, C. Demand response management by means of heat pumps contro via real time pricing. Energy Build. 2015, 90, 15–28. [CrossRef] 17. De Coninck, R.; Baetens, R.; Verbruggen, B.; Driesen, J.; Saelens, D.; Helsen, L. Modelling and simulation of a grid connected photovoltaic heat pump system with thermal energy storage using Modelica. In Proceedings of the 8th International Conference on System Simulation, Paris, France, 8–10 November 2010. 18. Prada, A.; Bee, E.; Grigiante, M.; Baggio, P. On the optimal mix between lead-acid battery and thermal storage tank for PV and heat pump systems in high performance buildings. Energy Procedia 2017, 140, 423–433. [CrossRef] 19. Kats, G.; Seal, A. Buildings as Batteries: The Rise of ‘Virtual Storage’. Electr. J. 2012, 25, 59–70. [CrossR 19. Kats, G.; Seal, A. Buildings as Batteries: The Rise of ‘Virtual Storage’. Electr. J. 2012, 25, 59–70. [CrossRef] 20 Psimopoulos E ; Bee E ; Luthander R ; Bales C Smart control strategy for PV and heat pump system 20. Psimopoulos, E.; Bee, E.; Luthander, R.; Bales, C. Smart control strategy for PV and heat pump system utilizing thermal and electrical storage and forecast services. In Proceedings of the Solar World Congress 2017, Abu Dhabi, UAE, 29 October–2 November 2017. 21. Pospísil, J.; Spilacek, M.; Kudela, L. Potential of predictive control for improvement of seasonal coefﬁcient of performance of air source heat pump in Central European Climate Zone. Energy 2018, 154, 415–423. [CrossRef] 22. Henze, G.P.; Felsmann, C.; Knabe, G. Evaluation of optimal control for active and passive building thermal storage. Int. J. Therm. Sci. 2004, 43, 173–183. [CrossRef] 23. Péan, T.Q.; Saloma, J.; Costa-Castelló, R. Review of control strategies for improving the energy ﬂexibility provided by heat pump systems in buildings. J. Process Control 2018, in press. 12 of 12 Environments 2018, 5, 132 24. TESS Component Libraries. Available online: http://www.trnsys.com/tess-libraries/ (accessed on 10 September 2018). 25. ASHRAE. International Weather for Energy Calculations (IWEC Weather Files); ASHRAE: Atlanta, GA, USA, 2001. 26. Bee, E.; Prada, A.; Baggio, P. Air-source heat pump and photovoltaic systems for residential heating and cooling: Potential of self-consumption in different European climates. Build. Simul. Int. J. 2019, in press. 27. UNI. UNI/TS 11300. References Prestazioni energetiche degli ediﬁce—Parte 1: Determinazione del fabbisogno di energia termica dell’ediﬁcio per la climatizzazione estiva ed invernale; Ente Nazionale Italiano di Normazione: Milano, Italy, 2016. 28. ENTRA NZE Project, Co-Founded by Intelligent Energy Europe Programme. Available online: www.entranze.enerdata.eu (accessed on 1 October 2018). 29. CEN. EN 16147:2017 Heat Pumps with Electrically Driven Compressors—Testing, Performance Rating and Requirements for Marking of Domestic Hot Water Units; European Committee for Standardization: Brussels, Belgium, 2017. 30. ARERA. Available online: www.arera.it/it/prezzi.htm (accessed on 1 October 2018). 31. ARERA. Available online: www.arera.it/it/comunicati/18/180403.htm (accessed on 1 October 201 32. GSE. Servizio di Scambio sul Posto Disposizioni Tecniche di Funzionamento. 2016. Available online: www.gse.it/servizi-per-te/fotovoltaico/scambio-sul-posto/documenti (accessed on 1 October 2018). 33. Bee, E.; Prada, A.; Baggio, P. Rule Based Control Strategies of Thermal Storage in Residential Heating Systems with Air-Source Heat Pump and Photovoltaic Panels. In Proceedings of the 5th International High Performance Buildings Conference, West Lafayette, IN, USA, 9–12 July 2018. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
W4255245316.txt	https://joghr.scholasticahq.com/article/11930.pdf	en		Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island population	Journal of global health reports	2,018	cc-by	6,521	Hernandez BY, Tareg AC, Reichhardt M, et al. Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island population. Journal of Global Health Reports. 2018;2:e2018016. doi:10.29392/joghr.2.e2018016 Research Article Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island population Brenda Y. Hernandez 1, Aileen C Tareg 2, Martina Reichhardt 3, Angelica Agapito 4, Xuemei Zhu 1, Angela Sy 5, Arnice Yuji 6, Jeffrey Killeen 7, Owen Chan 1, Lee E Buenconsejo-Lum 5 1 University of Hawaii Cancer Center, Honolulu, Hawaii, USA, 2 Public Health, Colonia, Yap, Micronesia (the Federated States of), 3 Yap State Department of Health Services, Colonia, Yap, Micronesia (Federated States of), 4 Waab Community Health Center, Colonia, Yap, Micronesia (the Federated States of), 5 John A. Burns School of Medicine, Honolulu, Hawaii, USA, 6 John A. Burns School of Medicine, Pacific Programs, Honolulu, Hawaii, USA, 7 Kapiolani Medical Center for Women and Children, Honolulu, Hawaii, USA Keywords: global health https://doi.org/10.29392/joghr.2.e2018016 Journal of Global Health Reports Vol. 2, 2018 Background Non-invasive, self-collection sampling methods for human papillomavirus (HPV) DNA detection have the potential to address logistical and cultural barriers to Pap screening, particularly in under resourced settings such as Yap state in the Federated States of Micronesia – a population with low levels of screening and high incidence of cervical cancer. Methods A randomized controlled trial was conducted among adult women in Yap to compare cervical HPV DNA in self-collected urine and clinician-collected liquid cytology. Adult women aged 21-65 (n=217) were randomized by the order of sample collection. Concordance of HPV DNA, evaluated by the Roche Linear Array, was compared in paired self-collected urine and clinician-collected liquid cytology samples. The sensitivity and specificity of urine HPV DNA for prediction of cervical HPV and abnormal cytology was also evaluated. p16 in urine cytology samples was additionally assessed. Results Overall, HPV DNA detection was significantly lower in urine than cervical samples for any HPV (27.8% and 38.3%, respectively) and high-risk HPV (15.1% and 23.8%, respectively). For paired samples, there was moderate agreement for the overall study population (Kappa=0.54, 95% confidence interval CI=0.40-0.68) and substantial agreement for women ≥40 years (Kappa=0.65, 95% CI=0.46-0.85). The sensitivity and specificity of urine for the detection of cervical high-risk HPV was 51.0% and 96.2%, respectively. The sensitivities of HPV DNA in urine and liquid cytology for prediction of abnormal cytology (ASCUS/LSIL/HSIL) were 47.4% (95% CI=31.0-64.2) and 57.9% (95% CI=40.8-73.7), respectively; specificities were 92.0% (95% CI=86.9%-95.5%) and 83.5% (95% CI=77.2-88.7). Urine p16 was poorly correlated with urine HPV DNA positivity. Conclusions Urine is less sensitive but more specific than directed cervical sampling for detection of cytologic abnormalities and may have utility for screening in older populations within low-resource communities when clinically-collected samples cannot be obtained. Globally, cervical cancer is the third most common cancer in women and the second most frequent cause of cancer death with the highest burden found in developing areas of the world.1 Human papillomavirus (HPV) infection, primarily oncogenic types HPV 16 and 18, is the principal cause of nearly all cervical cancers.2 Even with the availability of highly efficacious prophylactic HPV vaccines, screening remains an important component of cervical cancer prevention. In many developing countries, however, screening is underutilized and cervical cancer remains a major public health challenge.3,4 The Federated States of Micronesia (FSM) is comprised of 607 volcanic islands and atolls scattered over 1 million square miles of the Northwestern Pacific Ocean (WHO, 2011). FSM is one of the most resourcelimited US Affiliated Pacific Island (USAPI) jurisdictions. Yap State, FSM has a population of approximately 12,000 people living on 22 inhabited small islands and atolls spread across 500 square miles of Western Pacific ocean (Figure 1). Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... Figure 1. The Federated States of Micronesia (FSM) is comprised of 607 islands and atolls scattered over 1 million square miles of the Northwestern Pacific Ocean Yap State, FSM has a population of approximately 12,000 people living on 22 inhabited small islands and atolls spread across 500 square miles of Western Pacific ocean. Map source: http://legacy.lib.utexas.edu/maps/islands_oceans_poles/micronesia_pol99.jpg. Micronesian women throughout the Pacific have among the highest rates of cervical cancer in the world and often present with late stage disease. The incidence of cervical cancer in Yap is over twice that of the U.S. and most cases are diagnosed at advanced stages.4 The high burden of cervical cancer in Yap is consistent with low levels of screening, which remain at less than 40% throughout the FSM.5 Major barriers to cervical cancer screening in Yap include geography, lack of trained personnel, limited clinical resources, as well as issues of cultural and personal acceptability.5 Primary health care is provided through a hospital and public health clinics on the main Yap island and, for the outer islands, through small health dispensaries run by health assistants and equipped with variable electricity and limited supplies and medication. Cervical cancer screening, largely comprised of cytology (Pap testing) and visual inspection with acetic acid (VIA), is available on the main island and, sporadically, on the outer islands by traveling public health teams.5 Follow-up colposcopy and biopsy as well as treatment for precancerous and early stage cervical cancer are also available on the main island with more advanced stage cancers referred to medical facilities off-island.5 For low-resource communities like Yap, the need for more culturally-, resource-, and health workforce-appropri- ate methods of cervical cancer screening has been recognized.6 The need for alternatives screening approaches has also been recognized in high resource settings such as the U.S. where over half of cervical cancers are diagnosed in women who are unscreened despite having access to health care.7 HPV DNA testing has been shown to be effective for cervical cancer screening when used as an adjunct to cytology or as a primary test with similar or better sensitivity for the detection of precancerous lesions compared to cytology alone.8–10 Nonetheless, the improved sensitivity offered by HPV DNA testing to supplement or to replace cervical cytology does not address current barriers to cervical cancer screening. Similar to Pap smear collection, current methods for the collection of samples for HPV DNA testing require a trained clinician to directly sample the cervix. Non-invasive, self-collection sampling strategies which are reliable, efficient, and acceptable have the potential to address current barriers to cervical cancer screening in underserved communities. Self-sampling methods for HPV DNA testing are generally more acceptable and preferable to women compared to collection methods performed by a clinician.11 Evaluation of HPV self-sampling methods has largely focused on the collection of cervical/vaginal or vagi- Journal of Global Health Reports 2 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... nal samples using swabs, brushes, tampons, or lavage.11 In a study of HPV transmission between male-female partners, we found urine to be a good proxy for cervical HPV infection.12 In a meta-analysis, urine was found to be generally accurate for the detection of cervical HPV DNA.13 However, few studies have also compared urine and cervical samples for the prediction of cervical cytologic outcomes.14 METHODS OBJECTIVES A randomized controlled trial was conducted in the state of Yap in the FSM. The purpose of the project was to evaluate the detection of HPV DNA in self-collected urine compared to clinician-collected cervical cell samples. STUDY SETTINGS AND STUDY SUBJECTS The study was approved by the Western Institutional Review Board. Study participants were enrolled through six Wa’ab community clinics located throughout the region. Written informed consent was obtained from all participants who were enrolled between March-May 2016. Eligible subjects included women aged 21-65 who had not had a hysterectomy, were not currently pregnant, and either who had not been screened within the past 3 years or had abnormal screening results within the past 3 years. The latter criteria based on screening history was intended to target high-risk populations with a high prevalence of HPV. Following completion of the study visit, participants were each provided with a tote bag as a token of appreciation. TRIAL DESIGN Study subjects were randomized into one of two groups to account for the potential influence of the order of sampling procedures on HPV DNA detection: 1) Cervical sampling by a trained clinician followed by self-collection of urine; 2) Self-collection of urine followed by clinician-collected cervical sampling. Randomization was based on sequential enrollment into the study at each of the six clinics. SPECIMEN COLLECTION At each study site, cervical cell specimens were collected by trained clinicians in private examination rooms. A sterile cytobrush was used to sample the endocervical canal and transformation zone then placed into liquid cytology collection media (ThinPrep, Hologic, Inc, Marlborough, Massachusetts, USA). Urine specimens were collected by the participant in a private restroom using a labeled sterile collection cup. Individuals were instructed to collect up to 30 mL of first-void urine. Cervical cell and urine samples were stored at 4 degrees Celsius until they were transported to a central facility on the main Yap island where they were packed and shipped on ice to Honolulu, Hawaii, U.S.A. INTERVIEWER-ADMINISTERED SURVEY Interviews were conducted and medical records reviewed to collect demographic characteristics, cervical screening and HPV vaccine history, sexual history, and medical conditions and heath behaviors. The survey also addressed the acceptability of the cervical sampling and urine collection procedures; these results were recently reported.15 CERVICAL CYTOLOGIC EVALUATION Cervical specimens in the liquid cytology media were processed for cytologic evaluation at a College of American Pathologists (CAP)-certified pathology laboratory in Honolulu, Hawaii. Papanicolaou (Pap) smear stained slides were read by board-certified cytotechnologists using standard cervical cytology criteria based on the Bethesda system.16 Abnormal results were confirmed by a board-certified pathologist. Diagnostic follow-up for abnormal cytology results, including follow-up colposcopy and biopsy, were in accordance with the recommendations of the American Society for Colposcopy and Cervical Pathology.17 For a subset of women with abnormal cytology, liquid cytology specimens underwent reflex HPV testing through the Honolulu pathology laboratory. The Roche Cobas 4800 system (Roche Molecular Systems, Inc.) was used for real-time PCR using group probes for high-risk HPV genotypes 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68 and individual probes for HPV 16 and HPV 18 (for specimens positive by group probe). HPV DNA TESTING HPV DNA testing of cervical and urine specimens was conducted the University of Hawaii Cancer Center in Honolulu, Hawaii where liquid cytology specimens were sent following cytologic evaluation and reflex testing. Following DNA isolation, specimens were evaluated for HPV DNA using a PCRbased assay to target a consensus region of the HPV L1 gene. Amplicons were genotyped with the Linear Array HPV Genotyping Test (LA, Roche Diagnostics, Indianapolis, IN) which distinguishes 37 HPV genotypes (6, 11, 16, 18, 26, 31, 33, 35, 39, 40, 42, 45, 51, 52 (XR), 53, 54, 55, 56, 58, 59, 61, 62, 64, 66, 67, 68, 69, 70, 71, 72, 73, 81, 82, 83, 84, 89, IS39). Human beta-globin PCR was included as a measure of sample sufficiency. Samples negative for beta-globin were considered inadequate and were excluded from the statistical analyses. P16 IN URINE CYTOLOGY Aliquots of urine specimens were concentrated on glass slides using a cytospin. Slides were stained with a p16 mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) (dilution 1:400) according to the manufacturer’s specifications. Slides were read by a study pathologist who was blinded to the HPV status of cases. p16 was classified as positive or negative based on any nuclear and/or cytoplasmic staining. STATISTICAL ANALYSIS SAS software version 9.4 (SAS Institute, Inc., Cary, North Carolina, USA) was used for analyses of data. HPV genotypes were grouped as any HPV and high-risk HPV. HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68 were clas- Journal of Global Health Reports 3 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... sified as high-risk (oncogenic).18 (Non-oncogenic types and HPV types of undetermined risk status included HPV 6, 11, 26, 34, 40, 42, 44, 53, 54, 55, 61, 62, 64, 66, 67, 70, 71, 72, 73, 81, 82, 83, 84, and 89). Samples positive for 1 or more high-risk genotypes with or without other genotypes were classified as oncogenic, or high-risk. Agreement of HPV status between self-collected urine and clinician-collected cervical samples was measured by percent agreement and Cohen’s Kappa and McNemar statistics.19 Kappa values were defined as ≤0 (no agreement); 0.01-0.20 (slight agreement); 0.21-0.40 (fair agreement); 0.41–0.60 (moderate agreement); 0.61-0.80 (substantial agreement); (0.81-1.00) excellent agreement.19 The sensitivity and specificity of urine for the prediction of cervical high-risk HPV were evaluated. The sensitivities and specificities of high-risk HPV in both urine and cervical samples for the prediction of cervical cytology were also evaluated. Comparisons between categorical variables utilized the χ2 statistic. All tests were two-sided, and P<0.05 was considered statistically significant. RESULTS STUDY POPULATION A total of 217 women aged 21-65 years were enrolled (Table 1). Five percent of women had prior HPV vaccination (at least one dose) and 59.5% had prior cervical cancer screening via Pap smear and/or visual inspection with acetic acid (VIA). Among those previously screened, 17.1% had abnormal cytology; 3 women had a history of biopsy-confirmed CIN. Sexually transmitted disease history was positive for 17.5% of participants. Family history of cancer was reported by 45.2% of subjects; cervical cancer was the 4th most frequent (data not shown). Two-thirds of women reported 2-9 male sexual partners in their lifetime and over two-thirds reported never using a condom during vaginal intercourse. CERVICAL CYTOLOGY AND HISTOPATHOLOGY Cervical cytology was normal for 83% of women; abnormal cytology included atypical cells of unknown significance (ASCUS) (14%), low-grade squamous intraepithelial lesions (LSIL) (1.4%) and high-grade squamous intraepithelial lesions (HSIL) (1.8%). Biopsy for follow-up of abnormal cytology was completed for 13 women. Histologically-confirmed invasive cervical cancer was diagnosed in two women, carcinoma in situ in two, CIN II/III in 2, and CIN III in five; two had normal (negative) biopsies. SAMPLE SUFFICIENCY Overall, 97.7% of self-collected urine specimens were sufficient compared to 98.6% of cervical specimens (P=0.0003) (Table 2). The sufficiency of urine and cervical samples did not significantly vary (P≥0.05) by the order of sample collection, age, clinic site, urine pH, or time since last urination. HPV DNA DETECTION HPV DNA (any genotype) was detected in 27.8% of urine samples and 38.3% of cervical samples (P<0.0001) (Table 2). High-risk HPV was detected in 15.1% of urine and 23.8% of cervical samples (P<0.0001). A total of 24 distinct HPV genotypes were detected in urine and 29 genotypes in cervical samples (Figure 2). The most frequently detected types in urine were oncogenic HPV 51, 58, and 68 and other HPV 54, 62, and 72. For cervical samples, the most frequently observed types were oncogenic HPV 16, 31, 51, 52, 58 and other HPV 62. Multiple genotypes were detected in 36% and 31% of HPV positive urine and cervical samples, respectively. HPV DNA detection was compared by study and participant variables (data not shown). Detection of HPV DNA (any genotype) in both urine and cervical samples did not vary by the order of collection, clinic site, urine pH, time since last urination, number of sexual partners, or condom use (P≥0.05 for all). Urine HPV DNA detection by age group did not vary by age: 35.1% (20-29 yrs.), 30.8% (30-39), 21.7% (40-49), and 26% (50 and over) (P=0.48). In contrast, cervical HPV DNA detection decreased with age: 58.3% (20-29), 40% (30-39), 32.3% (40-49), and 29.4% (50 and over) (P=0.03). Cervical HPV DNA detection also varied by alcohol and betel nut use. HPV DNA was detected in 50.5% of current alcohol drinkers compared to 28.6% of nondrinkers (P=0.001). HPV DNA was detected in cervical samples of 40.4% of betel nut chewers compared to 15.8% of non-chewers (P=0.04). Urine HPV detection did not vary by alcohol or betel nut use. Although there was no variation of HPV DNA by hypertension status, there was some variation of HPV DNA detection by use of hypertension medication. Urine HPV DNA was not detected among any women who used hypertension medication compared to 29% of those non-users although the difference was not statistically significant (P=0.06). Of the 11 women who had a history of HPV vaccination, 5 were HPV positive in urine and cervical samples for genotypes other than quadrivalent vaccine-covered types (HPV 6, 11, 16, and 18). (Normal cervical cytology was observed in 10 of 11 HPV vaccinated women; 1 was ASCUS). AGREEMENT OF HPV DNA IN PAIRED URINE AND CERVICAL SAMPLES Overall, HPV DNA detection in paired urine and cervical samples showed moderate agreement (Kappa=0.55, 95% CI=0.43-0.66) (Table 3). Genotype concordance (partial or complete) was 81.6% for HPV-positive urine-cervical pairs. Agreement was similar for samples positive for high-risk genotypes (with or without concurrent presence of other types) (Kappa=0.54, 95% CI=0.40-0.68). High-risk HPV agreement was moderate when urine samples were collected first (Kappa=0.57, 95% CI=0.39-0.76) as well as when cervical samples were collected first (Kappa=0.51, 95% CI=0.31 - 0.71). Agreement between paired urine and cervical samples substantially varied by age. For high-risk HPV, agreement was moderate for women age 20-39 (Kappa=0.45 95% CI=0.25-0.64). Among women ages 40 and older, highrisk HPV agreement was substantial (Kappa=0.65, 95% CI=0.46-0.85). Agreement varied across the six Yap study sites ranging from fair levels of agreement (Kappa=0.29, 95% CI=-0.11-0.70) to substantial levels of agreement Journal of Global Health Reports 4 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... Table 1. Study population: Yap, Federated States of Micronesia (n=217) No. % 117 53.9 100 46.1 Collection order: Cervical cytology followed by urine Urine followed by cervical cytology Age: range (21-65) 20-29 37 17.1 30-39 66 30.4 40-49 63 29.0 ≥60 51 23.5 Medical history: HPV vaccination 11 5.1 Cervical cancer screening (Pap and/or VIA) 129 59.5 Abnormal cervical cancer screening (n=129)* 22 17.1 Sexually transmitted infection 38 17.5 Betel nut chewing† 186 85.7 Alcohol use† 96 44.2 Cigarette smoking† 24 11.1 Diabetes mellitus 15 6.9 High cholesterol 36 16.6 Hypertension‡ 40 18.4 Overweight or obese 100 46.1 Diabetes medication 6 2.8 Family history of cancer 98 45.2 1 partner 39 18.7 2-9 partners 141 67.5 10+ partners 29 13.9 Never 147 67.7 Rarely 21 9.8 Sometimes 45 20.7 Most of the time 4 1.8 Lifetime no. partners (n=209): Frequency of condom use during vaginal intercourse: Two women had a history of biopsy-confirmed CIN 2-3 and one biopsy-confirmed CIN 1. †Current use. ‡Includes 9 women using hypertension medication. (Kappa=0.78, 95% CI=0.50-1.00). Age is unlikely to have influenced HPV agreement by study site as the age distribution of study subjects across study sites did not significantly vary (P=0.11). Given the variation in HPV DNA detection in cervical samples by alcohol and betel nut use, agreement between paired urine and cervical samples was compared by use of these substances. Agreement did not vary by betel nut use but significantly varied by alcohol use. Among current drinkers, agreement between paired urine and cervical samples was moderate (Kappa=0.43 95% CI=0.26-0.60) while among non-drinkers, agreement was substantial (Kappa=0.65, 95% CI=0.49-0.81). Agreement could not be compared by hypertension medication use as urine HPV DNA was not detected among any users. REFLEX HPV TESTING Reflex HPV testing of liquid cytology cervical specimens was conducted for 29 individuals with abnormal cytology. Reflex testing yielded 14 of 29 positive for HPV 16/18/31/33/ 35/39/ 45/51/52/56/58/59/68; 1 result was indeterminate. One of the 14 cases were positive for HPV 18 upon additional testing for HPV 16 and HPV 18. For the 29 cases undergoing reflex HPV testing, compared to with samples tested by the Roche linear array, agreement in high-risk HPV DNA detection was excellent for cervical samples (Kappa=0.85, 95% CI=0.65-1.00) and substantial for urine (Kappa=0.64, 95% CI=0.37-0.92). Journal of Global Health Reports 5 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... Table 2. Urine and cervical cytology samples: comparison of sample sufficiency and HPV DNA Urine (n=217) Cervical cytology (n=217) No. No. P-value 0.0003 Human beta-globin (sample sufficiency): Negative 5 (2.3%) 3 (1.4%) Positive 212 (97.7%) 214 (98.6%) HPV DNA negative 153 (72.2%) 132 (61.7%) HPV DNA positive 59 (27.8%) 82 (38.3%) Negative 180 (84.9%) 163 (76.2%) Positive Any HPV DNA: High-risk HPV DNA: <0.0001 <0.0001 32 (15.1%) 51 (23.8%) Total number HPV genotypes 24 29 % samples with multiple types 36.0% 31.0% Figure 2. High-risk HPV genotype detection in urine and cervical cytology samples Includes 32 urine samples and 51 cervical samples positive for high-risk HPV. P16 IN URINE p16 was expressed in 45% of urine cytology samples including strong expression in koilocytes of a sub-set of cases (Figure 3). However, p16 was poorly correlated with urine HPV positivity (percent agreement 57.2% any HPV; 55.3% high-risk HPV). p16 expression also did not correlate with cervical cytologic status (data not shown). URINE SENSITIVITY AND SPECIFICITY FOR CERVICAL HPV DNA AND ABNORMAL CERVICAL CYTOLOGY CUS, LSIL, and HSIL were combined as the numbers were too few for separate evaluation.) For the prediction of cervical high-risk HPV DNA, the sensitivity of urine was 51.0% (95% CI=37%-65%) and specificity was 96.2% (95% CI=92.0%-99.0%). For ASCUS/LSIL/HSIL, the sensitivity of high-risk HPV in urine (47.4%, 31.0%-64.2%) was less than that of cervical HPV DNA (57.9%, 95% CI=40.8%-73.7%). In contrast, the specificity of high-risk HPV in urine (92.0%, 95% CI=86.9%-95.5%) was greater than that of cervical HPV (83.5, 95% CI=77.2%-88.7%). The sensitivity and specificity of urine high-risk HPV for prediction of cervical high-risk HPV and abnormal cytology were evaluated (Table 4). (Abnormal cytology including AS- Journal of Global Health Reports 6 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... Table 3. Agreement of HPV DNA detection in paired urine and cervical samples Concordant urine/cervix Discordant urine/cervix Total no.* Pos/ pos Neg/ neg Pos/ neg Neg/ pos % agreement Kappa 95% CI P-value† 210 49 118 10 33 79.5% 0.55 0.43-0.66 0.0005 210 26 153 6 25 85.2% 0.54 0.40-0.68 0.0006 Urine sample first 98 13 71 0 14 85.7% 0.57 0.39-0.76 0.0002 Cervical sample first 112 13 82 6 11 84.8% 0.51 0.31-0.71 0.2300 20-39 100 14 65 4 17 79.0% 0.45 0.26-0.64 0.0046 ≥40 110 12 88 2 8 90.9% 0.65 0.46-0.85 0.0600 Any HPV: All High-risk HPV: All Collection order: Age (years): CI – confidence interval *Excludes pairs with insufficient samples. †McNemar P-value. Table 4. Sensitivity and specificity of urine high-risk HPV for prediction of cervical HPV and abnormal cytology Sensitivity Urine high-risk HPV Clinical endpoint Specificity Cervical high-risk HPV % 95% CI Cervical HPV (n=51) 51.0 37.0 65.0 % N/A ASCUS/LSIL/HSIL (n=38) 47.4 31.0 64.2 57.9 Urine high-risk HPV 95% CI 40.8 73.7 Cervical high-risk HPV % 95% CI % 96.2 92.0 99.0 N/A 92.0 86.9 95.5 83.5 95% CI 77.2 88.7 HPV – Human papilloma virus, CI – confidence interval, ASCUS – atypical squamous cells of undetermined significance, LSIL – low-grade squamous intraepithelial Lesion, HSIL – high-grade squamous intraepithelial lesion, N/A – not applicable DISCUSSION Figure 3. p16 expression in urine cytology p16 was expressed in 45% of urine cytology samples including strong expression in koilocytes of a subset of cases. However, p16 was poorly correlated with urine HPV positivity. In this randomized controlled trial, self-collected urine was generally inferior to clinician-collected cervical samples for the detection of cervical HPV DNA. HPV DNA detection was lower in urine compared to cervical samples and agreement was moderate between paired samples. The observed sensitivity of urine for the detection of cervical HPV (59.8%) in this study was on the lower end of sensitivity demonstrated in other studies evaluating urine which ranged from 53% to 99%.13,20 In contrast, the specificity (92.2%) was on the higher end of the range of specificities (38% to 99%). Nonetheless, comparisons across studies are limited by the variation in study populations, age distribution, collection methods, and laboratory assays.13 For the prediction of abnormal cytology, HPV measured in urine was less sensitive but more specific than cliniciancollected cervical samples. Our findings support that urine HPV detection may be most clinically useful in older women. Agreement in HPV DNA detection between paired urine and cervical samples was substantial among older women. This is consistent with evidence that HPV DNA testing as primary screening tool or as co-test with cytology is most suitable for women aged 30 Journal of Global Health Reports 7 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... years and older.8–10 Sample size limitations did not permit separate age group comparisons of urine and cervical HPV DNA for the prediction of cervical lesions. Agreement between paired urine and cervical samples and adequacy of urine specimens did not vary by the order of collection indicating that sufficient cervical cells remained to be shed into urine following directed sampling. Moreover, urine HPV was not influenced by the time since last urination suggesting the continuous shedding of HPVinfected cervical cells. The wide variation across clinical sites underscores the potential influence of logistical factors that may have influenced the integrity of samples. Clinical sites across Yap included more resourced facilities on the main island as well as the less resourced, smaller dispensaries. Consequently, there may have been variation in the quality of clinicianand self-collected samples. Although only a fraction of all samples were insufficient, they included significantly more urine than cervical specimens. There was no way to verify that the participants fully complied with the instructions for self-collection including the collection of first-void rather than mid-stream urine, the former of which is superior for the detection of HPV.13,20 Sample integrity may also have been influenced by other factors related to specimen collection, storage, and processing. In contrast to cervical samples collected into liquid cytology media, urine samples were collected without the use of a DNA-preserving media due to concerns of the spillage of media during the self-collection process. Degradation of viral DNA in urine samples in the absence of a stabilizing media has been reported.21 Urine and cytology specimens collected at the six sites were transported to a central facility on the main Yap island prior to shipment to the testing laboratory in Honolulu. It is possible that variable specimen handing practices across clinical sites and transport delays may also have affected sample integrity and subsequent measures of specimen sufficiency and HPV DNA. Concordance of genotypes was observed among the majority of samples positive in both urine and cervical paired samples. This underscores that urine and cervical samples were largely measuring the same cervical infection. Conversely, our findings provide some evidence that HPV detected in urine and cervical samples to some extent represented different anatomical sources. Collectively, the most frequently detected genotypes were somewhat different in urine and cervical samples. HPV in urine may represent viral infection from cells shed by the vagina and vulva in addition to the cervix. There is some evidence that genotypes trophic to the lower genital tract are not entirely consistent with cervical HPV genotypes.22 That urine HPV is not specific to the cervical infection is also supported by observed differences in age distributions. Cervical HPV DNA significantly varied with age with the highest prevalence in young women while urine HPV prevalence did not vary with age. As we and others have observed in U.S. and other populations, cervical HPV prevalence is strongly correlated with age with the highest prevalence in young women and declining with increasing age.23,24 Cervical HPV has also been shown to be correlated with number of sexual partners across population.25 Interestingly, we observed that HPV detection in both urine and cervical samples did not vary by number of sexual partners. This might reflect a limited exposure range in this generally high-risk study population. In the cervix, elevated expression of p16(INK4A), or p16, a cyclin-dependent kinase-4 inhibitor, is strongly correlated with HPV positivity and the presence of high-grade lesions.26 p16 expression in urine did not correlate with urine HPV or cervical cytology. Nonetheless, p16 was strongly expressed in koilocytes, which were found in a limited number of urine specimens. The detection of koilocytes in urine does lend support to the notion that HPV-infected cells of gynecologic origin are shed into the urine. The presence of urinary koilocytes has been reported in an immunosuppressed patient with cervical dysplasia and condylomatous lesions in the vulva, vagina, and cervix.27 Our results provide some evidence that the detection of HPV in urine and cervical samples and agreement between the two media is influenced by substance use. Agreement of HPV DNA detection in urine and cervical samples was substantial among non-drinkers but moderate among drinkers. HPV prevalence in cervical samples was significantly higher in drinkers compared to non-drinkers suggesting that this variation accounted for the better agreement of the two media among non-drinkers. Interestingly, HPV DNA detection in cervical samples also significantly varied by betel nut use with higher prevalence among daily users than non-users. Urine HPV did not vary by alcohol or betel nut use. Interestingly, none of the women taking medication for hypertension were positive for urine HPV. It is possible albeit speculative that the detection of urine HPV is impeded by the diuretic effects of such medications which result in increased urine volume. It should be noted that the Roche Linear Array assay utilized for the present study is not among the U.S. FDA-approved HPV assays that have been validated as a primary screening tool. Unlike other assays which utilize group probes, the Linear Array allows for discrimination of 37 individual HPV genotypes and has been extensively used for research purposes in the U.S. and worldwide. The Linear Array has been shown to be comparable to other HPV assays including the FDA-approved Cobas 4800 test. We observed excellent agreement between cervical samples tested in the linear array and the subset which underwent reflex testing with the Cobas 4800 test. In fact, high-risk HPV agreement was substantial between reflex tested cervical samples and urine samples although this was based on a very small sample. Other evaluations have also shown the linear array assay to correlate well with the Cobas 4800 test28 as well as with the FDA-approved Hybrid Capture 2 assay.29 In a number of developing countries where screening is available, screening rates remain low and a high burden cervical cancer persists.3 For low-resource settings such as Yap, there is a need for alternative screening strategies. The development of urine-based or other self-collection strategies for cervical cancer screening has the potential to transform prevention worldwide including in low-resource populations as well as underserved communities within developed areas of the world. Such strategies may include primary screening in accordance with current age-based clinical guidelines and would be particularly useful in populations where clinically-collected cervical samples cannot Journal of Global Health Reports 8 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... be obtained. Urine or other self-collected sampling could also be the basis of novel strategies such as the identification of high-risk women through mass HPV testing of selfcollected samples followed by targeted cytologic and HPV screening of high-risk HPV-positive older females. Self-collected samples may also be useful for follow-up of patients with abnormal cytology incorporating periodic HPV testing in order to identify those with persistent high-risk infection as these individuals bear the greatest risk for neoplastic progression.18 Such non-invasive follow up could reduce unnecessary colposcopy and biopsy procedures along with their associated medical and psychosocial sequelae, costs, and resources. Urine-based or other self-collected HPV testing may also be useful for monitoring the uptake and effectiveness of prophylactic HPV vaccination across populations.30 This is particularly relevant in adolescent female populations for which invasive cervical sampling is not appropriate. CONCLUSIONS Urine is less sensitive but more specific than directed cervical sampling for detection of cytologic abnormalities. Although the limited study population restricts our findings, our study provides evidence that urine may have utility for cervical screening, particularly in older populations of women when clinician-collected samples cannot be obtained. Confirmation of our study results in larger studies of low- and high-risk populations is needed. ACKNOWLEDGEMENTS We would like to acknowledge the project steering committee in Yap, comprised of public health and medical leaders, including key staff from the Yap Hospital Lab, MCH and Family Planning programs, Wa`ab Community Health Centers, Yap Cancer Program, Fais Outer Island Dispensary staff and local health board. Thank you to the women of Yap who participated in this first randomized controlled trial in the FSM. DISCLAIMER The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Centers for Disease Control and Prevention. FUNDING National Cancer Institute of the National Institutes of Health under award number NCI=3P30CA071789-16S3 (BYH, ACT, MR AG, XZ, AS, AY, LEBL), as well as the Centers for Disease Control and Prevention awards: CDC U58 DP000976 and U58 DP003906 Pacific Regional Central Cancer Registry (LEBL) and CDC U58 DP000779 FSM National Comprehensive Cancer Control Program (ACT, MR, AA). COMPETING INTERESTS The authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare no conflict of interest. CORRESPONDENCE TO: Brenda Y. Hernandez University of Hawaii Cancer Center 701 Ilalo Street Honolulu, Hawaii, United States of America brenda@cc.hawaii.edu This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CCBY-4.0). View this license’s legal deed at http://creativecommons.org/licenses/by/4.0 and legal code at http://creativecommons.org/licenses/by/4.0/legalcode for more information. Journal of Global Health Reports 9 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... REFERENCES 1. Torre LA, Siegel RL, Ward EM, Jemal A. Global cancer incidence and mortality rates and trends-an update. Cancer Epidemiol Biomarkers Prev. 2016;25:16-27. doi:10.1158/1055-9965.epi-15-0578 2. Schiffman M, Castle PE, Jeronimo J, Rodriguez AC, Wacholder S. Human papillomavirus and cervical cancer. Lancet. 2007;370(9590):890-907. doi:10.1016/s 0140-6736(07)61416-0 3. Steben M, Jeronimo J, Wittet S, et al. Upgrading public health programs for human papillomavirus prevention and control is possible in low- and middle-income countries. Vaccine. 2012;30(Suppl 5):F183-F191. doi:10.1016/j.vaccine.2012.06.031 4. Buenconsejo-Lum LE, Navasca D, Jeong Y, Wong E, Torris P, Palafox NA. Cancer in the US affiliated Pacific Islands 2007-2012. Published June 1, 2015. Accessed April 9, 2018. http://www.pacificcancer.org/ site-media/docpdfonwebpage/2015/PIJ%20Cancer%2 0Facts%26%20Figures%202007-2012%20060115.pdf 10. Pileggi C, Flotta D, Bianco A, Nobile CGA, Pavia M. Is HPV DNA testing specificity comparable to that of cytological testing in primary cervical cancer screening? Results of a meta-analysis of randomized controlled trials. Int J Cancer. 2014;135:166-177. doi:1 0.1002/ijc.28640 11. Nelson EJ, Maynard BR, Loux T, Fatla J, Gordon R, Arnold LD. The acceptability of self-sampled screening for HPV DNA: a systematic review and meta-analysis. Sex Transm Infect. 2017;93:56-61. do i:10.1136/sextrans-2016-052609 12. Hernandez BY, Wilkens LR, Zhu X, et al. Transmission of human papillomavirus in heterosexual couples. Emerg Infect Dis. 2008;14(6):888-894. doi:10.3201/eid1406.0706162 13. Pathak N, Dodds J, Zamora J, Khan K. Accuracy of urinary human papillomavirus testing for presence of cervical HPV: systematic review and meta-analysis. BMJ. 2014;349(sep16 12):g5264. doi:10.1136/bmj.g52 64 5. Obel J, McKenzie J, Buenconsejo-Lum L, et al. Mapping HPV vaccination and cervical cancer screening practice in the Pacific Regional – strengthening national and regional cervical cancer prevention. Asian Pac J Cancer Prev. 2015;16(8):3435-3442. doi:10.7314/apjcp.2015.16.8.3 435 14. Sahasrabuddhe VV, Gravitt PE, Dunn ST, et al. Comparison of human papillomavirus detections in urine, vulvar, and cervical samples from women attending a colposcopy clinic. J Clin Microbiol. 2014;52(1):187-192. doi:10.1128/jcm.01623-13 6. American College of Obstetricians and Gynecologists. Committee Opinion No. 624: Cervical cancer screening in low-resource settings. Obstet Gynecol. 2015;125(624):526-528. doi:10.1097/01.AO G.0000460763.59152.9e25611643 15. Sy AU, Hernandez BY, Tareg A, Reichhardt M, Buenconsejo-Lum L. Acceptability and feasibility of a community based participatory research project comparing cytology and urine HPV DNA testing for cervical cancer screening in Yap, Federated States of Micronesia. Cancer Epidemiol. 2017;50:283-288. doi:1 0.1016/j.canep.2017.07.008 7. Leyden WA, Manos MM, Geiger AM, et al. Cervical cancer in women with comprehensive health care access: attributable factors in the screening process. J Natl Cancer Inst. 2005;97(9):675-683. doi:10.1093/jnc i/dji115 16. Nayar R. The Bethesda System for Reporting Cervical Cytology. Definitions, Criteria, and Explanatory Notes. Springer; 2015. 8. Dijkstra MG, Snijders PJF, Arbyn M, Rijkaart DC, Berkhof J, Meijer CJLM. Cervical cancer screening: on the way to a shift from cytology to full molecular screening. Ann Oncol. 2014;25(5):927-935. doi:10.109 3/annonc/mdt538 9. Arbyn M, Ronco G, Anttila A, et al. Evidence regarding human papillomavirus testing in secondary prevention of cervical cancer. Vaccine. 2012;30(Suppl 5):F88-F99. doi:10.1016/j.vaccine.2012.06.095 17. Massad LS, Einstein MH, Huh WK, et al. 2012 updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. J Low Genit Tract Dis. 2013;17(5):S1-S27. d oi:10.1097/lgt.0b013e318287d329 18. Muñoz N, Castellsagué X, de González AB, Gissmann L. Chapter 1: HPV in the etiology of human cancer. Vaccine. 2006;24(S3):S1-S10. doi:10.1016/j.vac cine.2006.05.115 19. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas. 1960;20:37-46. doi:10.117 7/001316446002000104 Journal of Global Health Reports 10 Randomized controlled trial evaluating the utility of urine HPV DNA for cervical cancer screening in a Pacific Island... 20. Hagihara M, Yamagishi Y, Izumi K, et al. Comparison of initial stream urine samples and cervical samples for detection of human papillomavirus. J Infect Chemother. 2016;22(8):559-562. doi:10.1016/j.jiac.2016.05.009 26. Bergeron C, Ordi J, Schmidt D, Trunk MJ, Keller T, Ridder R. Conjunctive p16INK4a testing significantly increases accuracy in diagnosing high-grade cervical intraepithelial neoplasia. Am J Clin Pathol. 2010;133(3):395-406. doi:10.1309/ajcpxsvcdz3d5mzm 21. Vorsters A, Van den Bergh J, Micalessi I, et al. Optimization of HPV DNA detection in urine by improving collection, storage, and extraction. Eur J Clin Microbiol Infect Dis. 2014;33(11):2005-2014. doi:1 0.1007/s10096-014-2147-2 27. Altamirano E, Drut R. Koilocytes in urinary cytology in a patient with kidney transplant. Diagn Cytopathol. 2008;36(5):338-340. doi:10.1002/dc.20805 22. Castle PE, Rodriguez AC, Porras C, et al. A comparison of cervical and vaginal human papillomavirus. Sex Transm Dis. 2007;34(11):849-855. doi:10.1097/olq.0b013e318064c8c5 23. Hernandez BY, McDuffie K, Zhu X, Wilkens LR, Killeen J, Kessel B. Anal human papillomavirus infection in women and its relationship with cervical infection. Cancer Epidemiol Biomarkers Prev. 2005;14(11 Pt 1):2550-2556. doi:10.1158/1055-9965.e pi-05-0460 24. Dunne EF, Unger ER, Sternberg M, et al. Prevalence of HPV infection among females in the United States. JAMA. 2007;297(8):813-819. doi:10.100 1/jama.297.8.813 28. Castle PE, Sadorra M, Lau T, Aldrich C, Garcia FAR, Kornegay J. Evaluation of a prototype real-time PCR assay for carcinogenic human papillomavirus (HPV) detection and simultaneous HPV genotype 16 (HPV16) and HPV18 genotyping. J Clin Microbiol. 2009;47(10):3344-3347. doi:10.1128/jcm.00725-09 29. Gravitt PE, Schiffman M, Solomon D, Wheeler CM, Castle PE. A comparison of linear array and hybrid capture 2 for detection of carcinogenic human papillomavirus and cervical precancer in ASCUS-LSIL triage study. Cancer Epidemiol Biomarkers Prev. 2008;17(5):1248-1254. doi:10.1158/1055-9965.epi-0 7-2904 30. Franceschi S, Chantal Umulisa M, Tshomo U, et al. Urine testing to monitor the impact of HPV vaccination in Bhutan and Rwanda. Int J Cancer. 2016;139(3):518-526. doi:10.1002/ijc.30092 25. Vaccarella S, Franceschi S, Herrero R, Munoz N, Snijders PJ, Clifford GM. Sexual behavior, condom use, and human papillomavirus: pooled analysis of the IARC human papillomavirus prevalence surveys. Cancer Epidemiol Biomarkers Prev. 2006;15(2):326-333. doi:10.1158/1055-9965.epi-05-0 577 Journal of Global Health Reports 11
https://openalex.org/W4225647609	https://www.scielo.br/j/rbso/a/8Zr7FvkgbyNpbhNqJXCNvcm/?lang=pt&format=pdf	Portuguese	null	Estresse no trabalho e níveis de hemoglobina glicada: o papel da escolaridade. Dados da linha de base do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil)	Revista Brasileira de Saúde Ocupacional	2,022	cc-by	8,513	Contato: Raíla de Souza Santos E-mail: raila.santos@uerj.br Palavras-chave: hemoglobina A glicada; estresse ocupacional; escolaridade; estudos transversais; saúde do trabalhador. Os autores informam em Agradecimentos sobre apoios recebidos pelo estudo e declaram que não há conflitos de interesses. 317-6369/24319PT2022v47e5 Estresse no trabalho e níveis de hemoglobina glicada: o papel da escolaridade. Dados da linha de base do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil) Job stress and glycated hemoglobin levels: the role of educational attainment. Baseline data from the Longitudinal Study of Adult Health (ELSA-Brasil) 1 Raíla de Souza Santosa,b https://orcid.org/0000-0002-7983-6462 Rosane Harter Griepc https://orcid.org/0000-0002-6250-2036 Maria de Jesus Mendes da Fonsecaa https://orcid.org/0000-0002-5319-5513 Dóra Chora https://orcid.org/0000-0002-3941-5786 Itamar Souza Santosd https://orcid.org/0000-0003-3212-8466 Enirtes Caetano Prates Meloa https://orcid.org/0000-0003-4240-8365 Raíla de Souza Santosa,b https://orcid.org/0000-0002-7983-6462 Rosane Harter Griepc https://orcid.org/0000-0002-6250-2036 Maria de Jesus Mendes da Fonsecaa https://orcid.org/0000-0002-5319-5513 Dóra Chora https://orcid.org/0000-0002-3941-5786 Itamar Souza Santosd https://orcid.org/0000-0003-3212-8466 Enirtes Caetano Prates Meloa https://orcid.org/0000-0003-4240-8365 Job stress and glycated hemoglobin levels: the role of educational attainment. Baseline data from the Longitudinal Study of Adult Health (ELSA-Brasil) Enirtes Caetano Prates Meloa https://orcid.org/0000-0003-4240-8365 Enirtes Caetano Prates Meloa https://orcid.org/0000-0003-4240-8365 Revista Brasileira de Saúde Ocupacional ISSN: 2317-6369 (online) http://dx.doi.org/10.1590/2317-6369/24319PT2022v47e5 Artigo de pesquisa Resumo Introdução: as condições estressantes do trabalho estão associadas ao aumento dos níveis glicêmicos, mas pouco se conhece sobre o papel da escolaridade neste contexto. Objetivos: analisar a associação entre o estresse psicossocial no trabalho e os níveis de hemoglobina glicada (HbA1c) e a influência da escolaridade como modificador de efeito. Métodos: estudo transversal com dados de 11.922 trabalhadores ativos da linha de base do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil). O estresse psicossocial no trabalho foi avaliado pelo modelo demanda-controle. Foram empregadas a regressão logística multinomial e interações multiplicativas. Resultados: em trabalhadoras do sexo feminino com baixa escolaridade, observou-se associação entre baixo uso de habilidades no trabalho (OR 1,56; IC95% 1,09-2,24) e HbA1c elevada. A baixa autonomia no trabalho foi relacionada à HbA1c limítrofe (OR 1,21; IC95% 1,01-1,45) e elevada (OR 1,73; IC95% 1,19-2,51). Entre trabalhadores do sexo masculino com baixa escolaridade, o trabalho de alto desgaste (OR 1,94; IC95% 1,18-3,21), o baixo uso de habilidades (OR 2,00; IC95% 1,41-2,83) e a baixa autonomia no trabalho (OR 1,58; IC95% 1,13-2,21) foram associados à HbA1c elevada. Conclusão: o estresse psicossocial no trabalho foi associado a níveis limítrofes e elevados de HbAlc para trabalhadores com baixa escolaridade de ambos os sexos. Assim, ações para modificar as relações de trabalho e prevenir doenças crônicas devem ser priorizadas. a Fundação Oswaldo Cruz (Fiocruz), Escola Nacional de Saúde Pública, Departamento de Epidemiologia e Métodos Quantitativos. Rio de Janeiro, RJ, Brasil. b Universidade do Estado do Rio de Janeiro (UERJ), Policlínica Piquet Carneiro. Rio de Janeiro, RJ, Brasil. c Fundação Oswaldo Cruz (Fiocruz), Instituto Oswaldo Cruz, Laboratório de Educação em Ambiente e Saúde. Rio de Janeiro, RJ, Brasil. d Universidade de São Paulo (USP), Hospital Universitário, Centro de Pesquisa Clínica e Epidemiológica. São Paulo, SP, Brasil. d Universidade de São Paulo (USP), Hospital Universitário, Centro de Pesquisa Clínica e Epidemiológica. São Paulo, SP, Brasil. Introdução e maiores concentrações de HbA1c16. O alto desgaste e o baixo apoio social no trabalho também têm sido associados a maiores concentrações de HbA1c6. O estresse psicossocial tem sido identificado como um importante fator de risco para doenças crônicas como diabetes1,2. A hipótese de que o estresse psicos- social afeta diretamente os níveis glicêmicos encon- tra plausibilidade biológica na neuroendocrinologia (catecolaminas, glicocorticoides e biomarcadores de inflamação), resultando em alterações na produção de glicose hepática e na secreção e sensibilidade à insu- lina1. O estresse psicossocial também pode desenca- dear comportamentos considerados fatores de risco para o aumento dos níveis glicêmicos1,2. Para diabetes há mais evidências. Estudos seccio- nais17,18; controles de caso19; e coortes longitudinais na Europa20, Suécia21, Inglaterra7,22, Alemanha23 e Canadá24 mostraram que o estresse no traba- lho tem sido positivamente associado ao diabetes. No entanto, existem estudos que não confirmaram essa associação, por exemplo, os dos EUA8,25, Japão6, e Israel4; um estudo com três coortes [francês (Estudo GAZEL), sueco (Slosh Study) e britânico (British Whitehall II Study)]9 e duas meta-análises26,27. No entanto, pouca atenção tem sido dada na lite- ratura internacional à identificação de fatores de risco psicossociais que podem aumentar os níveis glicêmi- cos3. A etiologia múltipla do estresse e o tempo gasto na vida adulta com atividades laborais, exigem a iden- tificação de mecanismos pelos quais o ambiente de trabalho afeta a saúde dos trabalhadores3. Como exposto, a associação do estresse psicosso- cial no trabalho e o aumento dos níveis glicêmicos medidos pela HbA1c ainda são pouco explorados. Para o diabetes, há um maior número de estudos realizados em países desenvolvidos, especialmente na Europa e nos EUA. No entanto, há divergências entre esses achados do estudo. Além disso, não foram identificados estudos explorando a escola- ridade como potencial modificador de efeito nessa relação. Assim, este estudo tem como objetivo ava- liar a associação do estresse psicossocial e dos níveis glicêmicos, utilizando valores de hemoglobina gli- cada (HbA1c) como marcador, e analisar a influên- cia da escolaridade como modificador de efeito dessa associação em ambos os sexos. Vários estudos apontaram que as características do trabalho influenciam direta ou indiretamente os efei- tos do estresse na glicemia, como os turnos de traba- lho, a carga horária semanal, as relações interpessoais e o tipo de posição ou função4-9. Da mesma forma, a escolaridade pode melhorar ou reduzir esse tipo de efeito em mudanças glicêmicas5,7. Introdução Apesar de pouco explorada, a realização educacional, além de determi- nar o tipo de ocupação, pode interferir em estratégias de enfrentamento ao estresse e até mesmo modificar os efeitos das condições de trabalho na saúde7. Abstract Os autores informam que este estudo não foi apresentado em evento cientifico. Introduction: stressful work conditions are associated to increased glycemic levels, but little is known about the role of educational attainment in this association. Objectives: to analyze the association between psychosocial stress at work, levels of glycated hemoglobin (HbA1c), and the role of educational attainment as an effect modifier. Methods: a cross-sectional study with baseline data from 11,922 active workers who participated in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Psychosocial stress at work was measured via the Demand- Control model. Multinomial logistic regression and multiplicative interactions were performed. Results: among female workers with low educational attainment, low skill discretion was associated to elevated HbA1c (OR 1.56; 95% CI 1.09- 2.24). Low decision authority was associated to borderline (OR 1.21; 95% CI 1.01-1.45) and high (OR 1.73; 95% CI 1.19-2.51) HbA1c. Among male workers with low educational attainment, high strain (OR 1.94; 95% CI 1.18-3.21), low skill discretion (OR 2.0; 95% CI 1.41-2.83), and low decision authority (OR 1.58; 95% CI 1.13-2.21) were associated to high HbA1c. Conclusion: stress at work was associated to high and borderline levels of HbAlc in workers from both genders with low educational attainment. Actions to modify work relations and to prevent chronic diseases should be prioritized for this group. Os autores informam que o trabalho é baseado na dissertação de mestrado “Associação entre estresse psicossocial no trabalho e alteração dos níveis glicêmicos em servidores públicos: resultados do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil)”, por Raíla de Souza Santos, defendida em 2018 junto ao Programa de Pós-Graduação em Epidemiologia em Saúde Pública da Escola Nacional de Saúde Pública Sergio Arouca, Fundação Oswaldo Cruz. Recebido: 17/10/2019 Revisado: 08/05/2020 Aprovado: 28/07/2020 Keywords: glycated hemoglobin A; occupational stress; educational status; cross-sectional studies; occupational health. 1/12 Rev Bras Saude Ocup 2022;47:e5 Métodos Um dos modelos teóricos mais utilizados para avaliar os efeitos deletérios do estresse relacio- nado ao trabalho na saúde é o modelo demanda- controle (DC), desenvolvido por Karasek-Theorell, que se baseia em questões situacionais do ambiente de trabalho psicossocial, mais especificamente na forma como o trabalho é organizado e nas caracterís- ticas das tarefas realizadas10. Os efeitos do estresse ocupacional medidos por este modelo são bem esta- belecidos para doenças cardiovasculares11-13. É pos- sível que essa associação envolva alterações nos níveis glicêmicos, um dos principais fatores de risco para doenças cardiovasculares14. Entretanto, a evi- dência dos efeitos do estresse ocupacional nos níveis glicêmicos e do desenvolvimento de alterações glicê- micas e diabetes ainda é contraditória. Covariáveis Foram incluídas as seguintes características socio- demográficas: sexo (masculino ou feminino), idade (contínua), escolaridade (até ensino médio completo e superior completo). As variáveis relacionadas ao trabalho englobavam: carga horária semanal (até 40 horas semanais ou mais de 40 horas semanais) e turno de trabalho (diurno, noturno e ex-noturno, para trabalhadores que, em algum momento, traba- lharam em turno noturno). Os escores obtidos para os domínios de demanda psicológica (5-20 pontos) e latitude de decisão (6-24 pontos) foram dicotomizados em baixo e alto, através de um ponto de corte mediano. A latitude de decisão foi analisada por duas subdimensões não agrupadas, como proposto em outros estudos que indicam melhores ajustes30,31. O estresse psi- cossocial do trabalho foi categorizado em quatro quadrantes: “alto desgaste no trabalho” (caracteri- zada por trabalhadores com alta demanda psicoló- gica e baixa latitude de decisão; o subgrupo mais propenso ao estresse), “baixo desgaste no traba- lho” (baixa demanda e alta latitude de decisão no processo de trabalho), “trabalho passivo” (baixa demanda e baixa latitude de decisão; circunstância em que há limitações de habilidades e desânimo) e “trabalho ativo” (que associa altas demandas e alta latitude de decisão e consiste em circunstâncias menos prejudiciais aos trabalhadores mesmo na presença de altas demandas)10. Também foram avaliadas variáveis relaciona- das a hábitos de saúde: tabagismo (não fumante, ex-fumante e fumante) e atividades físicas, avalia- das pelo Questionário Internacional de Atividade Física (IPAQ), traduzidas para o português e vali- dadas, que contempla o tipo de atividade e sua intensidade, classificadas posteriormente como fortes, moderadas e fracas. O índice de massa cor- poral (IMC), representando a adiposidade, foi esti- mado como variável contínua, a partir da razão entre peso (kg) e altura quadrada (kg/m2), e categorizado para análise descritiva como “abaixo do peso ou nor- mal” (IMC abaixo de 24,9), “sobrepeso” (IMC entre 25 e 29,9) e “obeso” (IMC igual ou superior a 30). Para construir os indicadores para cada compo- nente do modelo, os escores gerados somando as respostas de demandas psicológicas (mediana = 14), uso de habilidade (mediana = 12) e autoridade de decisão (mediana = 6) foram dicotomizados. Para demandas psicológicas, a categoria de referência foi “baixa”, e para todas as subdimensões de latitude de decisão, a categoria de referência foi “alta”. Variável de exposição: estresse psicossocial no trabalho três a quatro meses anteriores à sua medição14,32. Correlaciona-se adequadamente com o risco de longo prazo de complicações do diabetes e apresenta van- tagens técnicas em avaliações pré-analíticas (não necessariamente rápidas) e analíticas (menos dis- túrbios diários durante o estresse e doença), quando comparadas às medições laboratoriais de glicemia utilizadas hoje14,32. A variável explicativa de interesse foi o estresse no trabalho, medido por meio da versão brasileira29 do questionário sueco de demanda-controle, desen- volvido por Theorell13 com base no Job Content Questionnaire (Questionário do Conteúdo do Trabalho)10. Este questionário abrange duas dimen- sões: demandas psicológicas, que envolvem carga de trabalho e as demandas psicológicas da realização das tarefas, e latitude de decisão, composta por duas subdimensões: autoridade de decisão, ou seja, auto- nomia para decidir como realizar o trabalho, e uso de habilidade, ou seja, as habilidades intelectuais adequadas para o trabalho10. Análise de dados Todas as análises foram estratificadas por sexo, uma vez que tanto o estresse ocupacional quanto a ocorrência de alterações glicêmicas diferiram para trabalhadores do sexo masculino e feminino17,21,23,25. O teste qui-quadrado de Pearson, com correção de Yates, foi utilizado para variáveis com apenas duas categorias. O nível de significância utilizado nos tes- tes foi de 5%. Projeto do estudo e participantes Este estudo transversal utiliza dados bási- cos do Estudo Longitudinal de Saúde do Adulto (ELSA-Brasil), um estudo multicêntrico que visa investigar a ocorrência e progressão de doenças crô- nicas, particularmente cardiovasculares e diabetes. A população do estudo do ELSA foi composta por 15.105 servidores públicos, com idades entre 35 e 74 anos, de cinco universidades e um instituto de pesquisa em seis capitais brasileiras28. Uma des- crição detalhada dos aspectos metodológicos desse estudo, como coleta de dados, medições clínicas e laboratoriais e medidas de controle de qualidade, são encontradas em outras publicações28. Nos últimos anos, apesar das evidências de como o estresse ocupacional afeta o desenvolvimento de diferentes padrões de alterações glicêmicas, medi- dos pela hemoglobina glicada (HbA1c), Cesana et al. relataram maiores concentrações de HbA1c em traba- lhadores expostos a um ambiente de trabalho estres- sante15. O estudo transversal de Netterstrom e Sjol observou a associação entre alto desgaste no trabalho Apenas os participantes ativos da linha de base foram selecionados para este estudo. Foram excluídos os trabalhadores aposentados, aqueles com níveis de HbA1c não testada e aqueles que não responderam a todas as perguntas relacionadas ao estresse ocu- pacional ou apresentaram dados faltantes sobre as covariáveis utilizados neste estudo. 2/12 Rev Bras Saude Ocup 2022;47:e5 Variável de exposição: estresse psicossocial no trabalho Resultados Da população do estudo ELSA, composta por 15.105, este estudo excluiu 3.059 trabalhadores apo- sentados. Entre os 12.046 participantes ativos da linha de base selecionados, 124 foram excluídos por não terem sua hemoglobina glicada (HbA1c) testada, deixarem perguntas relacionadas ao estresse ocu- pacional não respondidas ou apresentarem dados faltantes sobre as covariáveis utilizadas. Assim, nossa amostra final continha 11.922 trabalhadores (6.229 mulheres e 5.693 homens). Entre as mulheres, o baixo uso de habilidade esteve associada às alterações na HbA1c nos níveis limítrofe e elevado (OR 1,18; IC 95% 1,04-1,33 e OR 1,61; IC 95% 1,30-2,00, respectivamente). Entre os homens, foi associado à HbA1c elevada (OR 1,62; IC 95% 1,34-1,95). Da mesma forma, trabalhadores do sexo feminino e masculino com baixa autoridade de decisão mostraram cerca de 30% mais chances de apresentar HbA1c elevada em relação àqueles com alta autoridade de decisão (Tabela 3). A proporção de HbA1c elevada e limítrofe foi, respectivamente, de 6% e 21% entre as mulheres, e 8% e 19% entre os homens. A média de idade da população do estudo foi de 48,8 (DP = 7,0) anos para as mulheres e 49,5 (DP = 7,0) anos para os homens e cerca de 50% dos participantes apresen- taram alta escolaridade. Quanto aos fatores relacio- nados ao trabalho, as mulheres trabalhavam mais em turnos noturnos e os homens relataram maior carga horária semanal. Proporções semelhantes de homens (14,8%) e mulheres (12,6%) declararam-se fumantes; os homens trabalhavam em atividades físi- cas mais intensas e a obesidade foi mais frequente entre as mulheres (Tabelas 1 e 2). Após o ajuste, a associação de interesse tanto para os domínios do desgaste no trabalho quanto do estresse isolado foi fortemente afetada pela idade, o que aumentou a magnitude das associações, mas o mesmo padrão não ocorreu para a escolaridade; alguns estratos deixaram de ser significativos, e mag- nitudes diminuíram (Tabela 3). Observou-se interação entre escolaridade, des- gaste no trabalho, baixo uso de habilidade e baixa autoridade de decisão em homens (valores de p = 0,023, < 0,001 e 0,004, respectivamente). Entre as mulheres, encontramos apenas interação entre a escolaridade e as subdimensionais da latitude de decisão (valor de p = 0,019) (Tabela 3). Em geral, para todos os subgrupos HbA1c, a pre- valência de HbA1c limítrofe e elevada aumentou com a idade, um padrão inversamente proporcional ao aumento da escolaridade. Variável dependente: níveis glicêmicos A força da associação entre o estresse no traba- lho e os níveis glicêmicos foi avaliada por razão de probabilidade, e seus respectivos intervalos de con- fiança de 95% (IC 95%) por uma análise de regressão logística multinomial. As razões de chances (OR, do inglês, odds ratio) foram estimadas para o modelo bruto (modelo 1) e os modelos subsequentes foram progressivamente ajustados a um conjunto de variá- veis para controle por fatores sociodemográficos de confusão: idade (modelo 2); escolaridade (modelo 3); e características relacionadas ao trabalho, hábitos de A hemoglobina glicada (HbA1c) foi calibrada por cromatografia líquida de alto desempenho. As aná- lises foram realizadas em laboratório central para garantir a uniformidade nas análises dos exames30. A HbA1c foi classificada em três categorias: HbA1c “normal” < 5,7% (< 39 mmol/mol), HbA1c “limí- trofe” 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c “elevada” ≥ 6,5% (≥ 48 mmol/mol)14. A HbA1c é um marcador de nível glicêmico confiável, refletindo esses níveis nos últimos 3/12 Rev Bras Saude Ocup 2022;47:e5 vida e adiposidade (modelo 4). Apenas as variáveis significativas (p < 0,05) no teste ANOVA permane- ceram no modelo final. Para avaliar o efeito modifi- cador da escolaridade sobre a associação de interesse na escala multiplicativa, foi estimada a medida de seu efeito e seus respectivos intervalos de confiança de 95%. A presença de interação multiplicativa entre cada componente do modelo demanda-controle e a escolaridade foi testada nos modelos finais. As aná- lises foram realizadas em R, versão 3.3.1. semanais e entre trabalhadores noturnos e ex-notur- nos. No que se refere ao comportamento de saúde, concentraram-se entre os participantes obesos, aque- les que praticaram atividades físicas de baixa inten- sidade (moderada e baixa), fumantes e ex-fumantes (Tabelas 1 e 2). Quanto ao desgaste no trabalho, os participantes de ambos os sexos apresentaram maior frequência de trabalho passivo (Tabelas 1 e 2). Em geral, obser- vamos valores limítrofes e elevados de HbA1c entre mulheres com trabalho passivo, e baixas deman- das psicológicas, uso de habilidade e autoridade de decisão (Tabela 1). Entre os homens, ocorreu maior frequência de valores alterados de HbA1c naqueles com alto desgaste no trabalho e trabalho passivo, e entre aqueles classificados com baixa demanda psi- cológica, uso de habilidade e autoridade de decisão (Tabela 2). Aspectos éticos O estudo do ELSA foi aprovado pelos Comitês de Ética de cada instituição envolvida e pelo Conselho Nacional de Ética em Pesquisa (CONEP). Este estudo foi aprovado em 10 de abril de 2017, pelo Comitê de Ética em Pesquisa da Fundação Oswaldo Cruz (Fiocruz) e pela Escola Nacional de Saúde Pública (CAAE 656716.0.0000.5240). Todos os par- ticipantes assinaram um termo de consentimento livre e esclarecido. Comparando modelos brutos entre os sexos, as mulheres expostas ao trabalho passivo (baixo con- trole e baixa demanda) apresentam mais chances de HbA1c elevada (OR 1,79; IC 95% 1,35-2,38) compa- rado com mulheres expostas a trabalho de baixo des- gaste (Tabela 3). Entre os homens, as chances foram maiores entre aqueles expostos ao trabalho passivo (OR 1,56; IC 95% 1,24-1,97) ou de alto desgaste (OR 1,56; IC 95% 1,60-2,08), ambos com baixa autoridade de decisão (Tabela 3). Mulheres com altas deman- das psicológicas no trabalho apresentaram menores chances de apresentarem HbA1c elevada (OR 0,73; IC 95% 0,59-0,91) (Tabela 3). Resultados Para ambos os sexos, os valores limítrofes e elevados de HbA1c foram mais frequentes entre aqueles que trabalham até 40 horas 4/12 Rev Bras Saude Ocup 2022;47:e5 Tabela 1 Caracterização dos participantes do sexo feminino segundo níveis glicêmicos, trabalhadores ativos da linha de base do ELSA-Brasil, 2008-2010 MULHERES HbA1c§ Total Normal Limítrofe Elevada n = 6.229 n = 4.546 n = 1.297 n = 386 Idade Média (DP) Média (DP) Média (DP) Média (DP) 48,8 (7,0) 48,0 (6,9) 50,7 (7,0) 52,9 (6,5) n (%) n (%) n (%) n (%) Escolaridade Ensino médio completo 2.762 (44,3) 1.834 (66,4) 674 (24,4) 254 (9,2) Superior completo 3.467 (55,7) 2.711 (78,2) 624 (18,0) 132 (3,8) Carga horária semanal Até 40 horas por semana 4.477 (71,9) 3.192 (71,3) 976 (21,8) 309 (6,9) Mais de 40 horas por semana 1.752 (28,1) 1.354 (77,3) 323 (18,4) 75 (4,3) Turno de trabalho Diurno 3.911 (62,8) 2.945 (75,3) 763 (19,5) 203 (5,2) Noturno 1.117 (17,9) 765 (68,5) 266 (23,8) 86 (7,7) Ex-noturno 1.201 (19,3) 836 (69,6) 267 (22,2) 98 (8,2) Tabagismo Não fumante 3.926 (63,0) 2.964 (75,5) 754 (19,2) 204 (5,2) Ex-fumante 1.520 (24,4) 1.076 (70,8) 336 (22,1) 108 (7,1) Fumante 783 (12,6) 504 (64,4) 207 (26,4) 72 (9,2) Atividade física Forte 322 (5,3) 262 (81,4) 52 (16,1) 8 (2,5) Moderada 805 (13,1) 593 (73,7) 162 (20,1) 50 (6,2) Fraca 5.003 (81,6) 3.612 (72,2) 1.066 (21,3) 325 (6,5) Índice de massa corporal Abaixo do peso/normal 2.553 (41,0) 2.064 (80,8) 417 (16,3) 72 (2,8) Sobrepeso 2.183 (35,0) 1.589 (72,8) 463 (21,2) 131 (6,0) Obeso 1.493 (24,0) 891 (59,7) 418 (28,0) 184 (12,3) Desgaste no trabalho † Baixo desgaste ‡ 1.453 (23,3) 1.090 (75,0) 292 (20,1) 71 (4,9)** Ativo 1.204 (19,3) 926 (76,9) 226 (18,8) 52 (4,3) Passivo 2.252 (36,2) 1.576 (70,1) 491 (21,8) 185 (8,2) Alto desgaste 1.320 (21,2) 952 (72,1) 289 (21,9) 79 (6,0) Demandas Psicológicas Baixa ‡ 3.705 (59,5) 2.667 (72,0)* 782 (21,1)* 256 (6,9)* Alta 2.524 (40,5) 1.878 (74,4) 515 (20,4) 131 (5,2) Uso de habilidade Alto ‡ 2.846 (45,7) 2.148 (75,5) 561 (19,7) 137 (4,8)** Baixo 3.383 (54,3) 2.399 (70,9) 737 (21,8) 247 (7,3) Autonomia Alta ‡ 2.142 (34,4) 1.598 (74,6)* 433 (20,2)* 111 (5,2)* Baixa 4.087 (65,6) 2.947 (72,1) 866 (21,2) 274 (6,7) §HbA1c: HbA1c normal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡Categorias de referência; p < 0,05; p < 0,01 no teste qui- quadrado de Pearson com correção de Yates para nível glicêmico. §HbA1c: HbA1c normal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡Categorias de referência; p < 0,05; *p < 0,01 no teste qui- quadrado de Pearson com correção de Yates para nível glicêmico. DP: Desvio Padrão. ormal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); lh b lh d b i d (b i d d l l ) b lh i ( l d d l l ) b lh i (b Resultados DP: Desvio Padrão. §HbA1c: HbA1c normal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡Categorias de referência; p < 0,05; p < 0,01 no teste qui- quadrado de Pearson com correção de Yates para nível glicêmico. DP: Desvio Padrão. Resultados 5/12 Rev Bras Saude Ocup 2022;47:e5 Tabela 2 Caracterização dos participantes do sexo masculino segundo níveis glicêmicos, trabalhadores ativos da linha de base do ELSA-Brasil, 2008-2010 HOMENS HbA1c§ Total Normal Limítrofe Elevada n = 5.693 n = 4.051 n = 1.138 n = 504 Idade Média (DP) Média (DP) Média (DP) Média (DP) 49,5 (7,4) 49,1 (7,5) 50,1 (7,2) 52,1 (6,7) n (%) n (%) n (%) n (%) Escolaridade Ensino médio completo 2.872 (50,4) 1.884 (65,6) 643 (22,4) 345 (12,0) Superior completo 2.821 (49,6) 2.167 (76,8) 496 (17,6) 158 (5,6) Carga horária semanal Até 40 horas por semana 3.529 (62,0) 2.446 (69,3) 731 (20,7) 352 (10,0) Mais de 40 horas por semana 2.164 (38,0) 1.604 (74,1) 409 (18,9) 151 (7,0) Turno de trabalho Diurno 3.652 (64,1) 2.650 (72,6) 714 (19,6) 288 (7,9) Noturno 714 (12,5) 504 (70,6) 131 (18,3) 79 (11,1) Ex-noturno 1.327 (23,3) 897 (67,6) 292 (22,0) 138 (10,4) Tabagismo Não fumante 2.981 (52,4) 2.250 (75,5) 540 (18,1) 191 (6,4)** Ex-fumante 1.868 (32,8) 1.262 (67,6) 387 (20,7) 219 (11,7) Fumante 844 (14,8) 536 (63,5) 213 (25,2) 95 (11,3) Atividade física Forte 521 (9,3) 401 (77,0)* 90 (17,3)* 30 (5,8)* Moderada 887 (15,8) 633 (71,4) 179 (20,2) 75 (8,5) Fraca 4.197 (74,9) 2.946 (70,2) 852 (20,3) 399 (9,5) Índice de massa corporal Abaixo do peso/normal 1.952 (34,3) 1.495 (76,6) 359 (18,4) 98 (5,0) Sobrepeso 2.567 (45,1) 1.843 (71,8) 501 (19,5) 223 (8,7) Obeso 1.174 (20,6) 711 (60,6) 279 (23,8) 183 (15,6) Desgaste no trabalho † Baixo desgaste ‡ 1.736 (30,5) 1.257 (72,4) 356 (20,5) 123 (7,1) Ativo 1.047 (18,4) 792 (75,6) 183 (17,5) 72 (6,9) Passivo 2.081 (36,6) 1.442 (69,3) 416 (20,0) 223 (10,7) Alto desgaste 829 (14,6) 560 (67,6) 183 (22,1) 86 (10,4) Demandas Psicológicas Baixa ‡ 3.817 (67,0) 2.699 (70,7) 771 (20,2) 347 (9,1) Alta 1.876 (33,0) 1.352 (72,1) 366 (19,5) 158 (8,4) Uso de habilidade Alto ‡ 3.030 (53,2) 2.206 (72,8) 609 (20,1) 215 (7,1) Baixo 2.663 (46,8) 1.845 (69,3) 527 (19,8) 291 (10,9) Autonomia Alta ‡ 2.124 (37,3) 1.559 (73,4) 399 (18,8) 166 (7,7) Baixa 3.569 (62,7) 2.491 (69,8) 739 (20,7) 339 (9,5) §HbA1c: HbA1c normal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡Categorias de referência; p < 0,05; p < 0,01 no teste qui- quadrado de Pearson com correção de Yates para nível glicêmico. §HbA1c: HbA1c normal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡Categorias de referência; p < 0,05; *p < 0,01 no teste qui- quadrado de Pearson com correção de Yates para nível glicêmico. DP: Desvio Padrão. Resultados DP: Desvio Padrão. §HbA1c: HbA1c normal < 5,7% (< 39 mmol/mol), HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-47 mmol/mol) e HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡Categorias de referência; p < 0,05; *p < 0,01 no teste qui- quadrado de Pearson com correção de Yates para nível glicêmico. DP: Desvio Padrão. HbA1c: HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-45 mmol/mol); HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); IC 95%: Intervalo de confiança de 95%; OR: razão de chances; †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle). a Modelo bruto 1; b Modelo bruto 1 + ajuste por idade; c Modelo 2 + ajuste por escolaridade; d Modelo 3 + ajuste por turno de trabalho, hábito de fumar e índice de massa corporal. Resultados Rev Bras Saude Ocup 2022;47:e5 6/12 Tabela 3 Regressão logística multinomial com razão de chances (OR) e intervalos de confiança de 95% da associação entre estresse psicossocial no trabalho (quadrantes e dimensões isoladas) e níveis glicêmicos, ajustados por variáveis selecionadas, em trabalhadores ativos da linha de base do ELSA-Brasil, 2008-2010 associação entre estresse psicossocial no trabalho (quadrantes e dimensões isoladas) e níveis glicêmicos, ajustados por variáveis selecionadas, em trabalhadores ativos da linha de base do ELSA-Brasil, 2008-2010 Modelos Mulheres (n = 6.229) Homens (n = 5.693) HbA1c* HbA1c* Limítrofe Elevada Limítrofe Elevada Desgaste no trabalho † OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) Modelo bruto 1a Baixo desgaste 1,00 1,00 1,00 1,00 Ativo 0,91 (0,75-1,11) 0,86 (0,60-1,25) 0,82 (0,67-0,99) 0,92 (0,68-1,25) Passivo 1,16 (0,98-1,37) 1,79 (1,35-2,38) 1,02 (0,87-1,19) 1,56 (1,24-1,97) Alto desgaste 1,13 (0,94-1,36) 1,27 (0,91-1,77) 1,15 (0,94-1,41) 1,56 (1,16-2,08) Modelo 2b Baixo desgaste 1,00 1,00 1,00 1,00 Ativo 0,91 (0,75-1,11) 0,87 (0,60-1,26) 0,83 (0,68-1,01) 0,98 (0,72-1,33) Passivo 1,17 (0,99-1,38) 1,83 (1,37-2,45) 1,04 (0,88-1,22) 1,67 (1,32-2,11) Alto desgaste 1,22 (1,02-1,48) 1,50 (1,07-2,11) 1,19 (0,97-1,46) 1,72 (1,28-2,31) Modelo 3c Baixo desgaste 1,00 1,00 1,00 1,00 Ativo 0,93 (0,77-1,14) 0,92 (0,64-1,34) 0,85 (0,70-1,04) 1,04 (0,76-1,41) Passivo 1,00 (0,84-1,19) 1,28 (0,95-1,74) 0,87 (0,73-1,03) 1,14 (0,88-1,46) Alto desgaste 1,07 (0,88-1,30) 1,12 (0,79-1,59) 1,02 (0,82-1,26) 1,23 (0,90-1,67) Modelo 4d Baixo desgaste 1,00 1,00 1,00 1,00 Ativo 0,91 (0,74-1,11) 0,85 (0,58-1,24) 0,86 (0,70-1,05) 1,05 (0,77-1,44) Passivo 0,99 (0,83-1,18) 1,27 (0,94-1,73) 0,88 (0,74-1,05) 1,15 (0,89-1,49) Alto desgaste 1,02 (0,84-1,24) 1,05 (0,73-1,49) 1,01 (0,82-1,26) 1,19 (0,87-1,62) interações com escolaridade p = 0,4273 p = 0,0231 Dimensões Altas Demandas Psicológicas Modelo bruto 1a 0,94 (0,82-1,06) 0,73 (0,59-0,91) 0,95 (0,82-1,09) 0,91 (0,75-1,11) Modelo 2b 0,97 (0,85-1,10) 0,79 (0,63-0,98) 0,96 (0,83-1,10) 0,95 (0,77-1,16) Modelo 3c 1,01 (0,89-1,15) 0,88 (0,70-1,10) 1,00 (0,87-1,15) 1,05 (0,85-1,28) Modelo 4d 0,97 (0,85-1,11) 0,81 (0,65-1,03) 0,99 (0,86-1,15) 1,03 (0,84-1,26) interações com escolaridade p = 0,745 p = 0,579 Baixo uso de habilidade Modelo bruto 1a 1,18 (1,04-1,33) 1,61 (1,30-2,00) 1,03 (0,91-1,18) 1,62 (1,34-1,95) Modelo 2b 1,19 (1,05-1,35) 1,65 (1,32-2,05) 1,05 (0,92-1,19) 1,69 (1,40-2,04) Modelo 3c 1,01 (0,89-1,16) 1,15 (0,90-1,46) 0,86 (0,74-0,99) 1,15 (0,93-1,42) Modelo 4d 1,01 (0,88-1,16) 1,15 (0,90-1,46) 0,86 (0,74-1,00) 1,16 (0,93-1,43) interações com escolaridade p = 0,058 p < 0,00001 Baixa autonomia Modelo bruto 1a 1,09 (0,95-1,24) 1,33 (1,06-1,67) 1,16 (1,01-1,33) 1,30 (1,07-1,58) Modelo 2b 1,18 (1,04-1,35) 1,56 (1,23-1,97) 1,18 (1,03-1,36) 1,40 (1,15-1,71) Modelo 3c 1,08 (0,94-1,24) 1,25 (0,99-1,59) 1,06 (0,92-1,22) 1,07 (0,87-1,32) Modelo 4d 1,07 (0,93-1,23) 1,27 (1,00-1,63) 1,06 (0,92-1,22) 1,04 (0,84-1,28) interações com escolaridade p = 0,019 p = 0,004 HbA1c: HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-45 mmol/mol); HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); IC 95%: Intervalo de confiança de 95%; OR: razão de chances; †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle). Resultados HbA1c: HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-45 mmol/mol); HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); IC 95%: Intervalo de confiança de 95%; OR: razão de chances; †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle). 7/12 Rev Bras Saude Ocup 2022;47:e5 HbA1c limítrofe (OR 1,21; IC 95% 1,01-1,45) quanto à elevada (OR 1,73; IC 95% 1,19-2,51) (Tabela 4). HbA1c limítrofe (OR 1,21; IC 95% 1,01-1,45) quanto à elevada (OR 1,73; IC 95% 1,19-2,51) (Tabela 4). A escolaridade apresentou alteração na associa- ção de interesse com significância estatística con- firmada na escala multiplicativa (Tabela 4) quando comparadas as categorias de baixa (até ensino médio completo) versus alta (ensino médio completo e superior completo) escolaridade. Entre os homens com baixa escolaridade e traba- lho de alto desgaste que combina altas demandas e baixa latitude de decisão, as chances foram maiores para a HbA1c elevada (OR 1,94; IC 95% 1,18-3,21), em comparação com homens expostos a trabalho de baixo desgaste. Da mesma forma, as seguintes subdi- mensionais de latitude de decisão estão associadas à HbA1c elevada: baixo uso de habilidade (OR 2,00; IC 95% 1,41-2,83) e baixa autoridade de decisão (OR 1,58; IC 95% 1,13-2,21) para homens com baixa escolaridade (Tabela 4). Além disso, para homens com alta escolaridade, o efeito observado foi o oposto, o trabalho passivo e o uso de habilidade têm associa- ção inversa à ocorrência de HbA1c limítrofe (OR 0,77; IC 95% 0,62-0,98 e OR 0,78; IC 95% 0,54-0,95). No modelo final ajustado, após o controle de potenciais fatores de confusão, a associação de interesse permaneceu apenas entre as mulheres com baixa escolaridade (Tabela 4). Para os homens, encontramos associação tanto entre aqueles com alta quanto com baixa escolaridade, embora ao contrá- rio. As chances de HbA1c elevada entre mulheres com baixa escolaridade, submetidas abaixo uso de habilidade, são maiores (OR 1,56; IC 95% 1,09-2,24). Resultados Observou-se um padrão equivalente para a baixa autoridade de decisão, que está associada tanto à Tabela 4 Regressão logística multinomial com razão de chances (OR) e intervalos de confiança de 95% da associação do estresse psicossocial no trabalho (quadrantes e dimensões isoladas) e níveis glicêmicos, ajustados por variáveis selecionadas e pela interação com escolaridade de trabalhadores ativos do sexo feminino e masculino da linha de base do ELSA-Brasil 2008-2010 Modelos Interação Escolaridade Mulheres (n = 3.467) Homens (n = 2.821) Mulheres (n = 2.762) Homens (n = 2.872) HbA1c* HbA1c* Limítrofe Elevada Limítrofe Elevada Limítrofe Elevada Limítrofe Elevada OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) BAIXA ESCOLARIDADE ALTA ESCOLARIDADE †Desgaste no trabalho ‡ Baixo desgaste 1,00 1,00 1,00 1,00 Ativo 0,87 (0,68-1,11) 0,83 (0,54-1,28) 0,87 (0,60-1,26) 1,45 (0,92-2,31) Passivo 1,04 (0,80-1,36) 1,35 (0,87-2,10) 0,77 (0,61-0,98) 1,06 (0,76-1,47) Alto desgaste 1,14 (0,80-1,61) 1,94 (1,18-3,21) 0,90 (0,68-1,20) 0,97 (0,65-1,44) Dimensões de controle ‡ Alto uso de habilidade 1,00 1,00 1,00 1,00 1,00 1,00 1,00 1,00 Baixo uso de habilidade 0,97 (0,81-1,17) 1,56 (1,09-2,24) 0,98 (0,77-1,23) 2,00 (1,41-2,83) 1,06 (0,86-1,30) 0,92 (0,67-1,25) 0,78 (0,65-0,95) 0,87 (0,68-1,12) Alta autonomia 1,00 1,00 1,00 1,00 1,00 1,00 1,00 1,00 Baixa autonomia 1,21 (1,01-1,45) 1,73 (1,19-2,51) 1,13 (0,92-1,37) 1,58 (1,13-2,21) 0,91 (0,73-1,12) 0,98 (0,71-1,35) 0,97 (0,79-1,20) 0,78 (0,60-1,02) HbA1c: HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-45 mmol/mol); HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); IC 95%: intervalo de confiança 95%; OR: razão de chances; †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡modelo ajustado por idade, escolaridade, turno de trabalho, hábito de fumar, índice de massa corporal + interação com a realização educacional. Discussão baixa escolaridade têm recursos limitados para lidar com cargas de trabalho estressantes, em parte como resultado de múltiplos fatores de risco concorrentes que podem sobrecarregar seus esforços e resultar em habilidades de enfrentamento menos eficazes36,38,39. Nossos resultados mostraram que as chances de níveis de HbA1c elevados e limítrofes aumentam na presença de estresse psicossocial no trabalho para pessoas com baixa escolaridade. Essa covariável foi um modificador de efeito na associação investigada. O efeito do estresse no trabalho sobre os valores da hemoglobina glicada é reduzido com o aumento da escolaridade. Mulheres com baixa escolaridade, submetidas ao trabalho passivo, de baixa autoridade de decisão ou com baixo uso de habilidade apresen- taram maiores chances de apresentarem valores de hemoglobina glicadas elevados e limítrofes. Homens com baixa escolaridade, trabalho de alto desgaste, baixo uso de habilidade e baixa autoridade de deci- são mostraram associação com HbA1c elevada. Nossos resultados mostraram que as chances de níveis de HbA1c elevados e limítrofes aumentam na presença de estresse psicossocial no trabalho para pessoas com baixa escolaridade. Essa covariável foi um modificador de efeito na associação investigada. Embora a associação entre estresse ocupacio- nal e diabetes seja mais frequente entre as mulhe- res17,19,21-23,25, este estudo encontrou associações relevantes de estresse psicossocial no trabalho e variações nos valores de hemoglobina glicadas em ambos os sexos. Achados semelhantes foram vistos em estudos longitudinais europeus20 e no estudo sec- cional de Leynen18. Mesmo assim, há diferenças no tipo de trabalho desenvolvido e variação glicêmica para ambos os grupos. O efeito do estresse no trabalho sobre os valores da hemoglobina glicada é reduzido com o aumento da escolaridade. Mulheres com baixa escolaridade, submetidas ao trabalho passivo, de baixa autoridade de decisão ou com baixo uso de habilidade apresen- taram maiores chances de apresentarem valores de hemoglobina glicadas elevados e limítrofes. Homens com baixa escolaridade, trabalho de alto desgaste, baixo uso de habilidade e baixa autoridade de deci- são mostraram associação com HbA1c elevada. O trabalho passivo, o baixo usos de habilidade e, principalmente, a baixa autoridade de decisão no trabalho estão mais associadas às variações glicê- micas entre as mulheres com baixa escolaridade. Nossos achados confirmam os resultados disponí- veis relacionados ao impacto isolado do baixo con- trole no trabalho na ocorrência de HbA1c elevada entre as mulheres, que não foram observadas para altas demandas psicológicas17-19,21,24. Discussão Da mesma forma, observamos o trabalho de alto desgaste asso- ciado a altos valores de hemoglobina glicada apenas entre os homens, contradizendo estudos que mos- tram associação para esse tipo de trabalho – altas demandas combinadas com baixo controle – e dia- betes entre as mulheres18-21,23. O estresse é um dos fatores de risco psicossocial mais relevantes no desenvolvimento do diabetes. Diferentes mecanismos neuroendócrinos podem afe- tar diretamente a glicemia através de alterações na produção de glicose hepática e sensibilidade e secre- ção de insulina1,22,23. Além disso, o estresse mantém uma ação indireta relacionada ao enfrentamento negativo por meio de comportamentos de risco para doença1,23. Na perspectiva de que o estresse tem múltiplas etiologias, foi colocada ênfase no ambiente de trabalho, muitas vezes considerado estressante. Diversos aspectos do trabalho têm sido desta- cados como responsáveis por ampliar o risco de diabetes, como turnos noturnos33, longas horas e alta carga de trabalho5,6,34, qualidade das relações interpessoais no trabalho4,9 e o tipo de posição ou função7,8. Além disso, o risco de diabetes pode ser modificado pela escolaridade dos trabalhadores5,8 e sexo, que desempenham um papel determinante em sua prevalência18,19,21-23,25,27,35. Segundo Karasek et al.10, o estresse é gerado por restrições ambientais de longo prazo. Portanto, em alguns casos, os efeitos do estresse no trabalho só poderiam ser explicados pelo baixo controle. Tal hipótese explicaria a baixa magnitude de controle, independentemente das demandas psicológicas, nas mulheres, que apresentaram menor autoridade de decisão no trabalho em comparação com os homens10. Além disso, estudos que utilizaram com- ponentes da escala isoladamente encontraram resul- tados semelhantes para diabetes; Agardh et al.17 e Smith et al.24 apontam que as altas demandas no trabalho não têm influência na ocorrência de diabe- tes. Eriksson et al.21 também reforçam a necessidade de análises separadas para demanda e controle, pois não encontram associação com altas demandas psi- cológicas no trabalho isoladamente. Estudos com diferentes populações de trabalha- dores mostram o efeito da escolaridade no estresse do trabalho em relação a outros problemas de saúde, como doenças cardiovasculares, depressão e autoa- valiação de saúde ruim36-38. Estudos mostram que a escolaridade determina o tipo de ocupação; assim, os trabalhadores com cargos de alta escolaridade estão mais protegidos contra os efeitos nocivos do estresse36,39. Para a HbA1c elevada, mostrou-se que, mesmo na presença de altas demandas de trabalho e alta carga horária, o risco foi reduzido entre aqueles com alta escolaridade. Resultados Modelos Interação Escolaridade Mulheres (n = 3.467) Homens (n = 2.821) Mulheres (n = 2.762) Homens (n = 2.872) HbA1c HbA1c* Limítrofe Elevada Limítrofe Elevada Limítrofe Elevada Limítrofe Elevada OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) OR (IC 95%) BAIXA ESCOLARIDADE ALTA ESCOLARIDADE *HbA1c: HbA1c limítrofe 5,7%-6,4% (39 mmol/mol-45 mmol/mol); HbA1c elevada ≥ 6,5% (≥ 48 mmol/mol); IC 95%: intervalo de confiança 95%; OR: razão de chances; †Desgaste no trabalho: trabalho de baixo desgaste (baixa demanda e alto controle), trabalho ativo (alta demanda e alto controle), trabalho passivo (baixa demanda e baixo controle) e trabalho de alto desgaste (alta demanda e baixo controle); ‡modelo ajustado por idade, escolaridade, turno de trabalho, hábito de fumar, índice de massa corporal + interação com a realização educacional. Rev Bras Saude Ocup 2022;47:e5 8/12 Rev Bras Saude Ocup 2022;47:e5 Conclusão Em conclusão, nosso estudo aponta que a baixa escolaridade potencializa o efeito observado na asso- ciação entre o estresse no trabalho e os valores da hemoglobina glicada. O controle no trabalho (latitude de decisão) foi um fator determinante do estresse ocu- pacional associado aos valores de hemoglobina gli- cada entre os trabalhadores com baixa escolaridade em ambos os sexos. Assim, mudanças nas relações de trabalho que promovam maior uso de habilidades pessoais e maior autonomia para a tomada de deci- são para reduzir o estresse ocupacional podem ter impacto sobre esse marcador. Embora a realização educacional explique parcialmente a associação de interesse, a promoção de estratégias voltadas à melho- ria das condições de trabalho oferece um efeito posi- tivo, em princípio, mais viável do que mudar o nível de escolaridade na idade adulta. Portanto, ações que reduzam o estresse ocupacional podem representar o objetivo preferencial de intervenção para o desen- volvimento de estratégias de prevenção, incluindo doenças crônicas, como o diabetes. Deve-se considerar que os resultados relata- dos podem não representar a real magnitude do problema no país, uma vez que a distribuição da escolaridade na amostra pode não representar ade- quadamente o padrão geral de escolaridade dos trabalhadores brasileiros. Da mesma forma, as esti- mativas de mudança glicêmica são provavelmente subestimadas como resultado do uso de um único marcador (hemoglobina glicada) para a avaliação do metabolismo da glicose. No entanto, estudos mos- tram que a hemoglobina glicada tem sido o marcador mais adequado para avaliar o estresse psicossocial no trabalho6,15,16. É necessário ressaltar que modelos Agradecimentos O estudo de linha de base do ELSA-Brasil recebeu apoio do Ministério da Saúde e do Ministério da Ciência e Tecnologia (Agência Brasileira de Inovação – FINEP e Conselho Nacional de Pesquisa – CNPq) (bolsas 01 06 0010,00 RS, 01 06 0212,00 BA, 01 06 0300,00 ES, 01 06 0278,00 MG, 01 06 0115,00 SP, 01 06 0071,00 RJ). R.H.G é premiado pelo Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (número 301807/2016-7) e pelo Cientistas do Estado (Faperj). Os financiadores não desempenharam nenhum papel na concepção, coleta, análise e interpretação dos dados, redação do relatório e decisão de submeter o artigo para publicação. Discussão Atualmente, outros estudos30,31,40 analisa- ram as subdimensionais da latitude da decisão. Hökerberg et al.30 indicaram que o melhor modelo de ajuste para o contexto brasileiro foi alcançado utilizando-se as subdimensionais de controle de forma desagrupada. Tal método se justifica, pois os distintos aspectos30 do controle mensuram as sub- dimensões. Neste estudo, o baixo uso de habilidade e Outros achados confirmaram que indivíduos com baixa escolaridade têm menor controle sobre o trabalho e, consequentemente, menor uso de habi- lidade e autoridade. Como resultado, muitas vezes são privados de experiências satisfatórias no tra- balho38. É importante considerar que pessoas com 9/12 Rev Bras Saude Ocup 2022;47:e5 multinomiais não permitem a análise de outros tipos de interações, como, por exemplo, a interação adi- tiva que provavelmente estaria presente nas análises. Por fim, a natureza seccional das análises limita as interpretações sobre a direcionalidade das associa- ções observadas, e a causalidade reversa não pode ser descartada. Da mesma forma, a percepção dos trabalhadores sobre o estresse é provavelmente dinâ- mica e, portanto, varia em longos períodos. a autoridade de decisão estiveram associadas a varia- ções nos níveis de HbA1c para homens e mulheres. No entanto, a magnitude das associações relaciona- das às variações nos valores de hemoglobina glicadas foram diferentes. Em mulheres com baixa escolari- dade, houve uma associação mais forte para baixa autoridade de decisão no trabalho. Por outro lado, em homens com menor escolaridade, houve maior impacto para baixo uso de habilidade no trabalho. Os pontos relevantes deste estudo incluem o rigor metodológico em todas as etapas da coleta de dados e o fato de ser o primeiro estudo nacional brasileiro a testar a hipótese de interação do estresse psicosso- cial no trabalho e a escolaridade, importante medida de contexto social para mudanças nos níveis glicêmi- cos. Assim, a escolaridade foi mais do que um fator de confusão nessa relação e mostrou-se um modifi- cador de efeito para grupos com baixos níveis dela, que devem ser grupos prioritários para ações de pre- venção de doenças desencadeadas pelo estresse no ambiente de trabalho. Rev Bras Saude Ocup 2022;47:e5 Referências 13. Theorell T, Karasek RA. Current issues relating to psychosocial job strain and cardiovascular disease research. J Occup Health Psychol. 1996;1(1):9-26. 1. Lloyd C, Smith J, Weinger K. Stress and diabetes: a review of the links. Diabetes Spectr. 2005;18(2):121-7. 2. Surwit RS, van Tilburg MAL, Zucker N, McCaskill CC, Parekh P, Feinglos MN, et al. Stress management improves long-term glycemic control in type 2 diabetes. Diabetes Care. 2002;25(1):30-4. 14. Diabetes Care: the Journal of Clinical and Applied Research and Education. New York: American Diabetes Association. Vol. 41, Suppl. 1, 2018. 15. Cesana G, Panza G, Ferrario M, Zanettini R, Arnoldi M, Grieco A. Can glycosylated hemoglobin be a job stress parameter? J Occup Med. 1985;27(5):357-60. 3. Li J, Jarczok MN, Loerbroks A, Schöllgen I, Siegrist J, Bosch JA, et al. Work stress is associated with diabetes and prediabetes: cross-sectional results from the MIPH Industrial Cohort Studies. Int J Behav Med. 2013;20(4):495-503. 16. Netterstrøm B, Sjøl A. Glycated haemoglobin (HbA1C) as an indicator of job strain. Stress Med. 1991;7(2):113-8. 4. Toker S, Shirom A, Melamed S, Armon G. Work characteristics as predictors of diabetes incidence among apparently healthy employees. J Occup Health Psychol. 2012;17(3):259-67. 17. Agardh EE, Ahlbom A, Andersson T, Efendic S, Grill V, Hallqvist J, et al. Work stress and low sense of coherence is associated with type 2 diabetes in middle-aged Swedish women. Diabetes Care. 2003;26(3):719-24. 5. Nakanishi N, Nishina K, Yoshida H, Matsuo Y, Nagano K, Nakamura K, et al. Hours of work and the risk of developing impaired fasting glucose or type 2 diabetes mellitus in Japanese male office workers. Occup Environ Med. 2001;58(9):569-74. 18. Leynen F, Moreau M, Pelfrene E, Clays E, De Backer G, Kornitzer M. Job stress and prevalence of diabetes: results from the Belstress study. Arch Public Health. 2003;61:75-90. 6. Kawakami N, Haratani T. Epidemiology of job stress and health in Japan: review of current evidence and future direction. Ind Health. 1999;37(2):174-86. 19. Norberg M, Stenlund H, Lindahl B, Andersson C, Eriksson JW, Weinehall L. Work stress and low emotional support is associated with increased risk of future type 2 diabetes in women. Diabetes Res Clin Pract. 2007;76(3):368-77. 7. Kumari M, Head J, Marmot M. Prospective study of social and other risk factors for incidence of type 2 diabetes in the Whitehall II study. Arch Intern Med. 2004;164(17):1873-80. 20. Contribuições dos autores Santos RS, Griep RH, Fonseca MJM, Chor D, Santos IS e Melo ECP contribuíram substancialmente para a concepção e elaboração do estudo, para a coleta, análise e interpretação dos dados, para a elaboração e revisões críticas do manuscrito e para a aprovação da versão final publicada. Todos os autores assumem total responsabilidade pelo estudo e pelo conteúdo publicado. 10/12 Rev Bras Saude Ocup 2022;47:e5 Referências Sui H, Sun N, Zhan L, Lu X, Chen T, Mao X. Association between work-related stress and risk for type 2 diabetes: a systematic review and meta- analysis of prospective cohort studies. PLoS One. 2016;11(8):e0159978. 35. Griep RH, Toivanen S, van Diepen C, Guimarães JMN, Camelo LV, Juvanhol LL, et al. Work–family conflict and self-rated health: the role of gender and educational level. Baseline Data from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Int J Behav Med. 2016;23(3):372-82. 27. Cosgrove MP, Sargeant LA, Caleyachetty R, Griffin SJ. Work-related stress and type 2 diabetes: systematic review and meta-analysis. Occup Med (Lond). 2012;62(3):167-73. 28. Aquino EML, Barreto SM, Bensenor IM, Carvalho MS, Chor D, Duncan BB, et al. Brazilian Longitudinal Study of Adult Health (ELSA- Brasil): objectives and design. Am J Epidemiol. 2012;175(4):315-24. 36. Landsbergis PA, Schnall PL, Pickering TG, Warren K, Schwartz JE. Lower socioeconomic status among men in relation to the association between job strain and blood pressure. Scand J Work Environ Health. 2003;29(3):206-15. 29. Alves MGM, Chor D, Faerstein E, Lopes CS, Werneck GL. Versão resumida da “job stress scale”: adaptação para o português. Rev Saude Publica. 2004;38(2):164-71. 37. Tang M, Chen Y, Krewski D. Gender-related differences in the association between socioeconomic status and self-reported diabetes. Int J Epidemiol. 2003;32(3):381-5. 30. Hökerberg YHM, Aguiar OB, Reichenheim M, Faerstein E, Valente JG, Fonseca MJ, et al. Dimensional structure of the demand control support questionnaire: a Brazilian context. Int Arch Occup Environ Health. 2010;83(4):407-16. 38. Wege N, Dragano N, Erbel R, Jockel KH, Moebus S, Stang A, et al. When does work stress hurt? Testing the interaction with socioeconomic position in the Heinz Nixdorf Recall Study. J Epidemiol Community Health. 2008;62(4):338-41. 31. Fransson EI, Nyberg ST, Heikkilä K, Alfredsson L, Bacquer DD, Batty GD, et al. Comparison of alternative versions of the job demand-control scales in 17 European cohort studies: the IPD-Work consortium. BMC Public Health. 2012;12(1):62. 39. Griep RH, Nobre AA, Alves MGM, Fonseca MJM, Cardoso LO, Giatti L, et al. Job strain and unhealthy lifestyle: results from the baseline cohort study, Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). BMC Public Health. 2015;15(1):309. 32. Camargo JL, Gross JL. Glico-hemoglobina (HbA1c): aspectos clínicos e analíticos. Arq Bras Endocrinol Metabol. 2004;48(4):451-63. 33. Silva-Costa A, Rotenberg L, Nobre AA, Schmidt MI, Chor D, Griep RH. Gender-specific association between night-work exposure and type-2 diabetes: results from longitudinal study of adult health, ELSA-Brasil. Referências Nyberg ST, Fransson EI, Heikkilä K, Ahola K, Alfredsson L, Bjorner JB, et al. Job strain as a risk factor for type 2 diabetes: a pooled analysis of 124,808 men and women. Diabetes Care. 2014;37(8):2268-75. 8. Kroenke CH, Spiegelman D, Manson J, Schernhammer ES, Colditz GA, Kawachi I. Work characteristics and incidence of type 2 diabetes in women. Am J Epidemiol. 2006;165(2):175-83. 21. Eriksson AK, van den Donk M, Hilding A, Östenson CG. Work stress, sense of coherence, and risk of type 2 diabetes in a prospective study of middle-aged Swedish men and women. Diabetes Care. 2013;36(9):2683-9. 9. Mortensen J, Clark AJ, Lange T, Andersen GS, Goldberg M, Ramlau-Hansen CH, et al. Informal caregiving as a risk factor for type 2 diabetes in individuals with favourable and unfavourable psychosocial work environments: a longitudinal multi-cohort study. Diabetes Metab. 2018;44(1):38-44. 22. Heraclides AM, Chandola T, Witte DR, Brunner EJ. Work stress, obesity and the risk of type 2 diabetes: gender-specific bidirectional effect in the Whitehall II study. Obesity (Silver Spring). 2012;20(2):428-33. 10. Karasek R, Brisson C, Kawakami N, Houtman I, Bongers P, Amick B. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol. 1998;3(4):322-55. 23. Huth C, Thorand B, Baumert J, Kruse J, Emeny RT, Schneider A, et al. Job strain as a risk factor for the onset of type 2 diabetes mellitus: findings from the MONICA/KORA Augsburg cohort study. Psychosom Med. 2014;76(7):562-8. 11. Pimenta AM, Assunção AA. Estresse no trabalho e hipertensão arterial em profissionais de enfermagem da rede municipal de saúde de Belo Horizonte, Minas Gerais, Brasil. Rev Bras Saude Ocup. 2016;41:e6. 24. Smith PM, Glazier RH, Lu H, Mustard CA. The psychosocial work environment and incident diabetes in Ontario, Canada. Occup Med (Lond). 2012;62(6):413-9. 12. Backé EM, Seidler A, Latza U, Rossnagel K, Schumann B. The role of psychosocial stress at work for the development of cardiovascular diseases: a systematic review. Int Arch Occup Environ Health. 2012;85(1):67-79. 25. Annor FB, Masyn KE, Okosun IS, Roblin DW, Goodman M. Psychosocial stress and changes in estimated glomerular filtration rate among adults with diabetes mellitus. Kidney Res Clin Pract. 2015;34(3):146-53. 11/12 Rev Bras Saude Ocup 2022;47:e5 34. Tayama J, Li J, Munakata M. Working long hours is associated with higher prevalence of diabetes in urban male Chinese workers: the Rosai Karoshi study. Stress Health. 2016;32(1):84-7. 26. Rev Bras Saude Ocup 2022;47:e5 Referências Scand J Work, Environ Health. 2015;41(6): 569-78. 40. Joensuu M, Kivimäki M, Koskinen A, Kouvonen A, Pulkki-Råback L, Vahtera J, et al. Differential associations of job control components with mortality: a cohort study, 1986-2005. Am J Epidemiol. 2012;175(7):609-19. 12/12 Rev Bras Saude Ocup 2022;47:e5
https://openalex.org/W4383175835	https://zenodo.org/records/8108064/files/VOLODIN_Andrey_Sibiriana__designing_a_platform_for_aggregati.pdf	English	null	Sibiriana: designing a platform for aggregation of the historical and cultural heritage of the Angara-Yenisei macroregion	Zenodo (CERN European Organization for Nuclear Research)	2,023	cc-by	1,488	Rumyantzev, Maksim mrumyantsev@sfu-kras.ru Siberian Federal University, Russian Federation Rumyantzev, Maksim mrumyantsev@sfu-kras.ru Siberian Federal University, Russian Federation Pikov, Nikita npikov@sfu-kras.ru Siberian Federal University, Russian Federation Pikov, Nikita npikov@sfu-kras.ru Siberian Federal University, Russian Federation For the prototype stage of Siberiana, several types of collections were selected, guaranteed to be complete, amenable to a complete description, and in demand in research, and at the same time, they are quite diverse, so it would be possible to find out to what extent different essences of the historical and cultural heritage can be brought to a common denominator. They are the following: Digital Humanities 2023 Digital Humanities 2023 (pro.europeana.eu) is the benchmark for metadata organization, Wikidata (wikidata.org) gives good examples for data identifica- tion and linking open data, Global Digital Heritage Sibiriana: designing a platform for aggregation of the historical and cultural heritage of the Angara- Yenisei macroregion Sibiriana: designing a platform for aggregation of the historical and cultural heritage of the Angara- Yenisei macroregion (GlobalDigitalHeritage.org) is the reference point for cultural and historical objects presentation, and PhotoGrammar (Photo- Grammar.org) is the example we’re focusing on successfully com- bination of retrieval, analysis, and mapping of the artifacts in dif- ferent collections. The main aim of Siberiana (or Sibiriana, or Сибириана) inclu- des three relatively independent tasks. The first task is to create the research aggregator of historical and cultural heritage of the huge Krasnoyarsk region. This task needs to pose many questions concerning semantic meta descriptions, open licenses, facet classification, fuzzy sets, ontologies, up to ma- chine learning. The other task is to open the project for crowdsour- cing, to invite students, volunteers, independent researchers for digitalization, moderation, editing, quality control, and data cura- tion. The third task is to present a "business card" of the region, to lead the positioning of the region through cultural heritage: it starts from the 200 th anniversary of the Yenisei province of Rus- sian Empire (in 2022) to the 400 th anniversary of Krasnoyarsk city (in 2028). This task gives an extra opportunity for interaction with holders of historical and cultural heritage, for creation of thematic research collections, and it gives challenges for formulating auto- matic selection goals (Kizhner et al, 2021). Volodin, Andrey volodin@hist.msu.ru Siberian Federal University & Moscow State Unversity, Russian Federation Senotrusova, Polina psenotrusova@sfu-kras.ru Siberian Federal University, Russian Federation Antamoshkin, Oleslav oantamoskin@sfu-kras.ru Siberian Federal University, Russian Federation The consortium of Russian Digital Humanities initiatives was created to unite museums, libraries, archives with researchers and data scientists and engineers in developing such complex tasks of Siberiana. Kizhner, Inna inna.kizhner@gmail.com Siberian Federal University, Russian Federation & Haifa University, Israel Now, tools have been developed for aggregation and represen- tation of already digitized data (annotated images, electronic co- pies of historical documents, 3D/VR/AR/MR models). Various methods and standards for digitizing objects of historical, cultu- ral, and natural heritage are described for the platform needs. Ex- pert support and usability analysis is an important component of each of the stages of project creation. Intelligent technologies for processing large collections of data, texts, images, 3D-models are under development. Examples of best practices are collected that can be included and scaled for the tasks of Siberiana. Gruzdev, Andrey agruzdev@sfu-kras.ru Siberian Federal University, Russian Federation Gruzdev, Andrey agruzdev@sfu-kras.ru Siberian Federal University, Russian Federation Gruzdev, Andrey Bibliography Kizhner, Inna / Terras, Melissa / Manovich, Lev / Orekhov, Boris / Kim, Igor / Rumyantsev, Maxim / Bonch-Osmolovs- kaya, Anastasia (2022): “The history and context of the Digital Humanities in Russia”, in: Global Debates in the Digital Huma- nities (Eds Domenico Fiormonte, Sukanta Chaudhuri). University of Minnesota Press: 55-70. Kizhner, Inna / Terras, Melissa / Rumyantsev, Maxim / Khokhlova Valentina / Demeshkova Elisaveta / Rudov Ivan Collection types (7): minerals Collection types (7): minerals Collection types (2): text corpus of “Minusinsk Library” (1901) Collection types (3): aggregation of open digital copies yp ( ) p y ( Collection types (3): aggregation of open digital copies Collection types (4): postcards Collection types (5): cartographic (with the possibility of GIS implementation) Collection types (6): herbaria Collection types (7): minerals Collection types (4): postcards Collection types (4): postcards Collection types (5): cartographic (with the possibility of GIS implementation) implementation) Collection types (6): herbaria For the prototype reasons of Siberiana, the following database fields for faceted classification were the obligatory: what (object type), when (time period), where (location), where it is stored (in- stitution), as well as copyright or open license information. For the prototype reasons of Siberiana, the following database fields for faceted classification were the obligatory: what (object type), when (time period), where (location), where it is stored (in- stitution), as well as copyright or open license information. Siberiana is plotting as a digital model of the historical, cultu- ral, and natural heritage of the Angara-Yenisei macro-region. Our goal is to unlock the potential of this heritage resources in two di- mensions. For all types of users: it's to make such content easily accessible, visually well-presented, conveniently organized, and for researchers: it’s to provide a wide coverage of collections, va- luable services for exploratory data analysis, to give a platform for collecting specific research data for a significant agenda (Kizhner et al, 2022). Thus, historical, cultural, and natural heritage should be unders- tood as a resource that allows people to make life interesting, rich, valuable, and in some respects socially significant. Gruzdev, Andrey Collection types (1): archaeological artifacts In 2023, Digital Humanities Research Institute at Siberian Fe- deral University starts a working prototype of a research digital infrastructure for the aggregation, preservation, dissemination of Siberian historical and cultural heritage for historical, literary, eth- nographic, art history and other kinds of research at the intersec- tion of the humanities and computer sciences — Siberiana.online. The aim of the project is to launch a long-term initiative for digi- tization, analysis, and curation of the different collections of cultu- ral heritage of the Central Siberia (so-called Angara-Yenisei ma- croregion). The project is designed for research and education needs of the digital humanists at Siberian Federal University and world over, because judging by the current literature Siberian artifacts and collections evoke a steady interest (Kizhner et al, 2021). We took several online resources for benchmarks. Europeana 1 1 Digital Humanities 2023 Collection types (2): text corpus of “Minusinsk Library” (1901) Collection types (3): aggregation of open digital copies Collection types (4): postcards Collection types (5): cartographic (with the possibility of GIS implementation) Collection types (7): minerals Collection types (2): text corpus of “Minusinsk Library” (1901) Collection types (3): aggregation of open digital copies Collection types (4): postcards Collection types (5): cartographic (with the possibility of GIS implementation) Collection types (6): herbaria Collection types (7): minerals For the prototype reasons of Siberiana, the following database fields for faceted classification were the obligatory: what (object type), when (time period), where (location), where it is stored (in- stitution), as well as copyright or open license information. Siberiana is plotting as a digital model of the historical, cultu- ral, and natural heritage of the Angara-Yenisei macro-region. Our goal is to unlock the potential of this heritage resources in two di- mensions. For all types of users: it's to make such content easily accessible, visually well-presented, conveniently organized, and for researchers: it’s to provide a wide coverage of collections, va- luable services for exploratory data analysis, to give a platform for collecting specific research data for a significant agenda (Kizhner et al, 2022). Thus, historical, cultural, and natural heritage should be unders- tood as a resource that allows people to make life interesting, rich, valuable, and in some respects socially significant. “Siberiana”. Aggregator of historical and cultural heritage of the Yenisei Siberia. URL: https://siberiana.online/ [30.04.2023] ring bias in aggregated digitized content held in Google Arts and Culture”, in: Digital Scholarship in the Humanities, V. 36, Is. 3: 607–640. Bibliography Collection types (6): herbaria Collection types (6): herbaria Collection types (6): herbaria Kizhner, Inna / Terras, Melissa / Manovich, Lev / Orekhov, Boris / Kim, Igor / Rumyantsev, Maxim / Bonch-Osmolovs- kaya, Anastasia (2022): “The history and context of the Digital Humanities in Russia”, in: Global Debates in the Digital Huma- nities (Eds Domenico Fiormonte, Sukanta Chaudhuri). University of Minnesota Press: 55-70. Kizhner, Inna / Terras, Melissa / Rumyantsev, Maxim / Khokhlova, Valentina / Demeshkova, Elisaveta / Rudov, Ivan / Afanasieva, Julia (2021): “Digital cultural colonialism: measu- Kizhner, Inna / Terras, Melissa / Rumyantsev, Maxim / Khokhlova, Valentina / Demeshkova, Elisaveta / Rudov, Ivan / Afanasieva, Julia (2021): “Digital cultural colonialism: measu- 2 Digital Humanities 2023 ring bias in aggregated digitized content held in Google Arts and Culture”, in: Digital Scholarship in the Humanities, V. 36, Is. 3: 607–640. Kizhner, Inna / Terras, Melissa / Rumyantsev, Maxim / Sy- cheva, Kristina / Rudov, Ivan (2019): “Accessing Russian cul- ture online: The scope of digitization in museums across Russia”, in: Digital Scholarship in the Humanities, V. 34, Is. 2: 350–367. “Siberiana”. Aggregator of historical and cultural heritage of the Yenisei Siberia. URL: https://siberiana.online/ [30.04.2023] 3 3
https://openalex.org/W3204662873	http://www.scielo.br/pdf/rn/v19n5/a08v19n5.pdf	English	null	Idea Factory Chou Jigen Game Neptune Mk-IIand alcoholic liver disease	null	2,006	cc-by	6,339	1 Nutricionista, Hospital Universitário Clementino Fraga Filho, Universidade Federal do Rio de Janeiro. Av. Prof. Rodolpho P. Rocco, s/n, Ilha do Fundão, Rio de Janeiro, RJ, Brasil. Correspondência para/Correspondence to: E-mail: <tatidepaula@yahoo.com.br>. 2 Departamento de Nutrição e Dietética, Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro. Rio de Janeiro, RJ, Brasil. 3 Departamento de Nutrição Social e Aplicada, Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro. Rio de Janeiro, RJ, Brasil. 4 Departamento de Clínica Médica, Setor Hepatologia, Hospital Universitário Clementino Fraga Filho, Universidade Federal do Rio de Janeiro. Rio de Janeiro, RJ, Brasil. Aspectos metabólicos da vitamina A e doença hepática alcoólica Tatiana Pereira de PAULA1 Wilza Arantes Ferreira PERES2 Rejane Andréa RAMALHO3 Henrique Sérgio Moraes COELHO4 Tatiana Pereira de PAULA1 Wilza Arantes Ferreira PERES2 Rejane Andréa RAMALHO3 Henrique Sérgio Moraes COELHO4 VITAMINA A E DOENÇA HEPÁTICA ALCOÓLICA \| 601 REVISÃO \| REVIEW Vitamin A metabolic aspects and alcoholic liver disease Aspectos metabólicos da vitamina A e doença hepática alcoólica VITAMINA A E DOENÇA HEPÁTICA ALCOÓLICA \| 601 REVISÃO \| REVIEW Vitamin A metabolic aspects and alcoholic liver disease Aspectos metabólicos da vitamina A e doença hepática alcoólica Vitamin A metabolic aspects and alcoholic liver disease Vitamin A metabolic aspects and alcoholic liver disease 4 Departamento de Clínica Médica, Setor Hepatologia, Hospital Universitário Clementino Fraga Filho, Universidade Federal do Rio de Janeiro. Rio de Janeiro, RJ, Brasil. I N T R O D U C T I O N alteration of the metabolism of this vitamin in the liver disease. This approach aims to increase awareness of this nutritional condition in different stages of the liver disease. Alcoholic liver disease is one of the most common causes of liver cirrhosis in Occident, and it is among the ten most common causes of death by liver disease in the world1. Alcoholic liver disease has a wide clinical spectrum; it may progress to cirrhosis and to end-stage liver disease2. It’s a review work, based in the available articles of periodicals enlisted by Medline. To make the search of the bibliographic references, it was considered the ones published within 1979 to 2005, which contained the key-words: liver, vitamin A metabolism, alcoholic liver disease, vitamin A deficiency, retinoic acid and ethanol. Considering the amount of scientific evidence available about the proposed subject and aiming the better application, the information obtained was assembled, organized and critically evaluated. The selection of the works was guided by experimental studies and studies with human beings, and the inconclusive studies were excluded of this review. Vitamin A participates in several biological mechanisms, as epithelial differentiation, gene expression, reproduction, normal vision and integrity of the immune system3. The liver plays a central role in the uptake, storage and is also the oxidation site of vitamin A catabolism and responsible for the regulated release of this vitamin to other tissues4. By this way, direct effects of ethanol on liver may interfere on vitamin A metabolism and hepatic levels. Patients with alcoholic liver disease have been found to have low hepatic levels of retinol at all the stages of their disease5. Furthermore, the interaction between vitamin A and ethanol may result in structural hepatic alterations. R E S U M O O fígado é um órgão estratégico no metabolismo de macro e de micronutrientes e, portanto, é de esperar que o comprometimento de sua função seja acompanhado de alterações no estado nutricional de vitamina A. O objetivo deste artigo é revisar na literatura evidências científicas sobre o metabolismo hepático da vitamina A, o efeito das interações entre a vitamina A e o etanol sobre a morfologia hepática, além das alterações do metabolismo dessa vitamina na doença hepática alcoólica. Os dados foram selecionados na base de dados Medline no período de 1979 a 2005. O fígado é o principal órgão responsável pelo armazenamento, metabolismo e distribuição da vitamina A para os tecidos periféricos. Esse órgão utiliza retinol para seu funcionamento normal como proliferação e diferenciação celular. Dessa forma, a deficiência dessa vitamina parece alterar a morfologia hepática. Baixos níveis de retinol hepático têm sido encontrados em todos os estágios da doença hepática alcoólica. A deficiência de vitamina A na doença hepática alcoólica pode resultar da diminuição da sua ingestão ou absorção, na redução da síntese de ácido retinóico ou na diminuição da sua degradação. A ingestão crônica de álcool resulta em níveis reduzidos de ácido retinóico, o que favorece a formação de tumor hepático. Logo, em etilistas crônicos o estado nutricional de vitamina A deve ser monitorado, para evitar sua deficiência e seus sintomas clínicos, embora a suplementação deva ser feita com cautela, pois doses comumente usadas podem ser tóxicas para aqueles que consomem etanol. Termos de indexação: deficiência de vitamina A; etanol; fígado; doença hepática alcoólica; tretinoína; vitamina A. A B S T R A C T The liver is a strategic organ in the metabolism of macro and micronutrients; when its functioning is compromised, it may cause some change in the nutritional status of vitamin A. The purpose of this article is to review scientific evidence in literature on the liver metabolism of vitamin A, the role of ethanol and retinol interactions on hepatic morphology, besides the alterations in the metabolism of this vitamin in alcoholic liver disease. Data were collected from Medline database. The liver is the main organ responsible for the storage, metabolism and distribution of vitamin A to peripheral tissues. This organ uses retinol for its normal functioning such as cell proliferation and differentiation. This way, vitamin A deficiency seems to alter liver morphology. Patients with alcoholic liver disease have been found to have low hepatic levels of retinol in all stages of their disease. In alcoholic liver disease, vitamin A deficiency may result from decreased ingestion or absorption, reduction in retinoic acid synthesis or increased degradation. Long-term alcohol intake results in reduced levels of retinoic acid, which may promote the development of liver tumor. So, in chronic alcoholic subjects, vitamin A status needs to be closely monitored to avoid its deficiency and clinical effects, however its supplementation must be done with caution since the usual dose may be toxic for those who consume ethanol. Indexing terms: vitamin A deficiency; ethanol; liver; alcoholic liver disease; tretinoin; vitamin A. Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição 602 \| T.P. PAULA et al. Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Vitamin A metabolism in the liver The liver consists of several cell types, of which two (hepatocytes or parenchymal cells and stellate cells) are directly involved in the metabolism of vitamin A6. The current work intents to gather information about the liver metabolism of vitamin A, the role of the interaction between retinol and ethanol in the liver morphology, besides the Around 50%-90% of retinol ingested is absorbed via the lymphatics and carried by chylomicrons and chylomicron remnants, as retinyl Revista de Nutrição Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 VITAMINA A E DOENÇA HEPÁTICA ALCOÓLICA \| 603 it can be explained by the reduced hepatic vitamin storage, synthesis and/or diminished release of binding proteins by the liver13. Another factor could be influenced by the deficient enzymatic conversion of β-carotene into retinol, which also occurs in liver14. Other factors such as chronic inflammation and infection, which are part of clinical manifestations of liver cirrhosis, are also responsible for the reduction of the serum levels of retinol. This can be attributed to the reduction in the synthesis and release of retinol binding protein, during acute phase response15,16. In this case, decrease in serum levels can be transient and not related to the lower hepatic vitamin A storage16. esters, to the liver and taken up by the parenchymal cells7. In these cells, the retinyl esters are hydrolyzed and the free retinol is bound to cellular retinol binding protein I (CRPB1) and then transferred to the stellate cells, also called lipocytes, fat-storing cells or Ito cells8. The stellate cells, which in normal conditions contain about 90% of hepatic retinol, are responsible for uptake, storage and release of the retinol. In these cells, retinol bound to CRPB1 is esterified by the lecithin: retinol acyltransferase and retinyl esters are stored in lipid droplets of these cells. When retinol is present at high levels and the CRBP1 becomes saturated, acyl CoA: retinol acyltransferase may esterify the excess6. Thus, the hepatic retinol esterification depends on the amount of retinol present and whether retinol is bound to cellular retinol binding protein8,9. About 98% percent of the stellate cell vitamin A is in the form of retinyl esters6. The mobilization of retinol from stellate cells, occurs after the hydrolysis of retinyl esters and retinol is bound to retinol binding protein, and the holo-retinol binding protein complex formed is released. Vitamin A metabolism in the liver However, both parenchymal cells and stellate cells can secrete this complex to the bloodstream, but the release from stellate cells seems to be the predominant pathway6,10. In the blood, the holo-retinol binding protein complex is associated with the transthyretin, a protein also synthesized in the liver. Retinol then is removed from bloodstream and used by the target cells through specific receptors for active retinoids or its metabolites, especially retinoic acid, existing in the surface of the cells and nucleus11,12. The decrease in serum levels of vitamin A can be aggravated by the degree of liver disease. Rocchi et al.17, in a study developed with cirrhotic patients, found a narrow correlation between the serum reduction of retinol and the severity of the disease, according to the classification of Child & Pugh18, that it is based on clinical and biochemical data and may be considered as a survival predictor for these patients. The inadequate consumption of vitamin A leads to the depletion of its organic reserves. In patients with cirrhosis, the inadequate intake of this vitamin, mainly the vitamin A from animal sources, which is highly available, can exacerbate an expected reduction in the serum levels of vitamin A. In addition, the reduction in protein ingestion, by unfounded patient’s belief or inadequate recommendations in clinical practice, as prophylaxis for the hepatic encephalopathy contributes for the reduction of vitamin A from animal sources daily ingestion and the decreasing of bioconversion of provitamin A carotenoids. In agreement, the administration of low protein diets in rats showed to reduce the activity of β-carotene 15-15’monooxygenase, besides limiting the retinol binding protein synthesis19. Retinol, the major circulating form of vitamin A, is not biologically active and serves as a metabolic precursor of active retinoids. These are generated intracellularly by two oxidative enzymatic reactions in which retinol is first converted into retinaldehyde and then to retinoic acid8. Ethanol and retinol interactions associated with hepatic structural alterations patients with low vitamin A concentration and liver disease induced by drugs did not have any Mallory bodies27. Leo et al.5,23 did not verify relationship between the decrease of hepatic vitamin A concentration and the appearance of Mallory bodies in livers biopsied from alcoholic subjects and in rats fed with ethanol. Although this relationship has not been observed, one cannot exclude that vitamin A deficiency may potentiate the role of ethanol in the appearance of Mallory bodies27. It is suggested that the liver also works as a vitamin A storage site and it can use retinol for a normal functioning like cell proliferation and differentiation21,22. This way, vitamin A deficiency seems to alter the hepatic morphology and may be associated with hepatic structural alterations . Leo et al.23 evaluated the serum and hepatic levels of vitamin A in 41 patients with liver disease from several etiologies and compared these results to some alterations in hepatic morphology. These authors observed the relationship between hepatic vitamin A deficiency and the appearance of multivesicular lysosomes. These organelles seem to be filled with numerous particles like lipoprotein that suggest an abnormality in lipoprotein secretion. Searching for differentiating the ethanol toxic action from vitamin A action in lesion etiology, these researchers performed an experimental protocol in rats by administrating vitamin A-replete or deplete diets, with or without ethanol. As result, these authors suggested that ethanol can potentiate the effect of vitamin A deficiency, which was indicated by high frequency of multivesicular lysosomes in animals that received ethanol and deficient diet in vitamin A. However, these lesions were observed in animals fed with vitamin A deficient diet without ethanol as well as in patients with vitamin A deficiency and without alcoholic liver disease, while normal vitamin A diet, with or without ethanol, did not result in the appearance of multivesicular lysosomes in rats. Thus, multivesicular lysosomes appearance may be associated with low hepatic vitamin A levels, although this type of lesion is not considered a histopathological sign of vitamin A deficiency. Ethanol can activate stellate cells by several mechanisms, associated with ethanol metabolism, including direct fibrogenic action of acethaldehyde (a highly toxic metabolite), induction of oxidative stress, and increased lactate28-32. Stellate cells activation into myofibroblasts-like cells, responsible for the synthesis of collagen, results in hepatic fibrosis and is associated with a decrease in vitamin A storage in these cells33. Vitamin A deficiency and liver disease Moreover, the liver disease can course with intestinal alterations20 that can compromise the bioavailability and the bioconversion of carotenoids in vitamin A. The decrease in serum levels of retinol is frequently found in patients with liver cirrhosis and Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição 604 \| T.P. PAULA et al. Ethanol and retinol interactions associated with hepatic structural alterations This reduction may have a casual role in this cellular transformation, since in a culture of stellate cells was demonstrated that retinol and especially the retinoic acid reduced the proliferation of these cells, the collagen synthesis and the transformation into myofibroblasts-like cells. Thus, vitamin A deficiency may lead to hepatic fibrosis34. In vitro, rats’ stellate cell transformation into fibroblasts was inhibited by the cell exposition to retinol33. Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição Vitamin A and alcoholic liver disease Alcoholic liver disease normally courses with alteration in nutritional status of vitamin A35,36. Reduced levels of retinol have been found in serum of chronic alcoholic subjects, with or without hepatic disease, and in liver biopsies from alcoholic subjects5. In patients with alcoholic liver disease, low levels of hepatic vitamin A has been associated with the presence of Mallory bodies which seems a pathological keratinization of the hepatocytes24-26. However, Mallory bodies were also observed in patients with alcoholic liver disease and normal concentration of vitamin A while Calamita et al.37, in a study with patients with alcoholic cirrhosis, observed that all of them presented decreased serum retinol concentrations. Several factors may contribute for this alteration, such as: decrease in micro and macronutrient intake, decreased retinol absorption, Revista de Nutrição Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 VITAMINA A E DOENÇA HEPÁTICA ALCOÓLICA \| 605 consequently, impairs vitamin A mobilization27. Protein-calorie malnutrition usually present in these patients42, with prevalence between 34% to 87%, reaching about 100% in hospitalized patients43, also contributes to reduced transport of vitamin A since one of the main causes is decreased dietary intake of both micro and macronutrients, mainly reduced protein intake, which also limits RBP synthesis44. enhanced degradation in the liver and an increased mobilization of retinol from the liver to other organs27,35. Moreover, oxidative stress, induced by chronic ethanol ingestion, may increase demands of several antioxidants including vitamin A38. Like ethanol, retinol is an alcohol and, in vitro, both can be converted to corresponding aldehydes in reactions catalyzed by several isoenzymes of cytosolic alcohol dehydrogenase. It is expected that in vitro and, possibly, in vivo these two alcohols may compete for the same or similar enzymatic pathways. By this way, ethanol consumption affects retinol metabolism and function27. Vitamin A status may be worse in late stages of liver disease. Bell et al.45, found relation between reduced levels of hepatic vitamin A and decreased activity of prothrombin in patients with alcoholic liver disease and suggested that the hepatic vitamin A level is related to the decrease of the hepatic function Ethanol, both in acute and chronic intakes, induces an important depletion in hepatic vitamin A and alters the distribution of this vitamin to others tissues23,39. This reduction may be caused by decreased of hepatic uptake40 or by the increased metabolism of this vitamin39. Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Vitamin A and alcoholic liver disease was a significant increase of hepatic vitamin A when compared to the group that received normal vitamin A diet (5800UI/day for 8 weeks). A slighter increase was observed in the group that received the combination high vitamin A diet and ethanol (36% of total colories) possibly because of an increased vitamin A catabolism in the liver after chronic ethanol consumption. These researchers observed that even the group that received ethanol and normal vitamin A diet presented mild mitochondrial alterations and proliferation of the smooth endoplasmic reticulum, just as the control group that received only high vitamin A diet. By contrast, the high vitamin A-ethanol group showed much more striking lesions with enlarged mitochondria, confirming the hepatotoxicity of this vitamin and potentiation of its effect by ethanol. baboons and 400UI/day in rats). Researchers concluded that the vitamin A hepatic depletion induced by ethanol intake occurs in initial stages of liver disease. In baboons with hepatic steatosis, increased vitamin A plasma levels and retinol binding protein were not explained. The possible reasons found by the authors to explain the vitamin A hepatic depletion, in these two experimental groups, were the increase of mobilization of vitamin A evidenced by the increase of concentration of vitamin A in the kidneys and the testis, and the increase of catabolism of vitamin A through the enzyme cytochrome P450 (CYP) induced by ethanol consumption. The main CYP induced by ethanol is CYP2E1, which possesses a pronounced oxidative activity in metabolism of several compounds, resulting in increased production of acetaldehyde and free radical release48. CYP2E1 also has a high capacity to activate some commonly used drugs to their toxic metabolites and to promote carcinogenesis49. Liu et al.50 observed in vitro that the incubation of retinoic acid with microsomal fractions of hepatic tissue containing CYP2E1 from rats exposed to ethanol resulted in decrease of retinoic acid and the appearance of polar metabolites, which can be toxic to cellular membrane. Moreover, the treatment with chlormethiazole, an inhibitor of cytochrome system, in ethanol-fed rats restored the serum and hepatic concentrations of retinoic acid to normal levels. Vitamin A and alcoholic liver disease It has been observed that alcoholic liver disease is associated with the reduction of the concentration of vitamin A in the liver, even when the hepatic injury is mild, as in steatosis, and when serum concentration of this vitamin, retinol binding protein and transthyretin are normal27,28,36. This way, the serum levels of vitamin A may not reflect the vitamin A status in chronic alcoholic subjects with or without hepatic disease14, as in healthy subjects, since serum retinol level is kept constant until vitamin A liver reserve is nearly exhausted. Vitamin A deficiency may result in hipogonadism as consequence of testicular atrophy and decreased spermatogenesis, and night blindness46 due to reduced rate of regeneration of visual pigment rhodopsin (opsin conjugated to retinal), compromising visual adaptation to dim light11. In a study using two animal models, Sato & Lieber47 observed in baboons the decreasing in vitamin A hepatic concentration equivalent to 59%, 67% and 95% after a mean of 4, 12 and 60 months of ethanol feeding, respectively. In the first two groups was observed the appearance of hepatic steatosis while the last one developed fibrosis or cirrhosis. In rats, was demonstrated the decrease of vitamin A hepatic concentration after three weeks of ethanol administration with progressive reduction during the nine weeks of study. In relation to plasma retinol levels and retinol binding protein, they were found significantly increased in baboons with hepatic steatosis when compared to the control group, but they did not present any significant alteration in animals with fibrosis and cirrhosis. In rats, plasma retinol levels and RPB were significantly different between ethanol fed rats and control group. In both experiments the diets contained vitamin A as retynil acetate in usual amounts (5600UI/day in Ukleja et al.41 in a prospective study with liver transplantation candidates with several etiologies, found that patients with alcohol-related cirrhosis had the lowest median for total hepatic vitamin A, as well as the lowest median for serum retinol concentration, when compared to patients with other etiologies. Low levels of plasma vitamin A may even be attributed to decreased synthesis of retinol binding protein by the liver. Moreover, alcoholic cirrhosis is normally associated with zinc deficiency, which may contribute to the decrease of retinol binding protein synthesis, and Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição Revista de Nutrição 606 \| T.P. PAULA et al. Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Vitamin A and alcoholic liver disease These authors demonstrated that this toxicity was due to induction of apoptosis by polar retinol metabolites. These findings explain the mechanism by which ethanol potentiate vitamin A hepatotoxicity. The reduction of serum levels of retinoic acid can, in part, be justified by the increase of its degradation as consequence of chronic ethanol consumption. The decrease of biosynthesis of retinoic acid can also be involved, since ethanol inhibits the oxidation of retinol by alcohol dehydrogenase58. Wang et al.59 suggested that decreased liver and plasma retinoic acid induced by chronic alcohol feeding (36% of total calorie intake for 1 month) in rats could be caused by inhibition of retinol oxidation, increased catabolism of retinoids via ethanol-induced cytochrome P450 enzymes, lower amounts of retinol and retinyl palmitate, precursors of retinoic acid, in liver of ethanol-fed rats, which could result from reduced intestinal absorption or hepatic uptake, and increased mobilization of vitamin A from liver to other organs. Based on these facts one should ask how supplement patients with alcoholic liver disease, mainly those who are still alcoholics, who demonstrate clinical signs of vitamin A deficiency. As previously shown, ethanol and vitamin A interactions may result in hepatotoxicity. Majumdar et al.46 supplemented 25 chronic alcoholic patients with 2500UI/day of vitamin A orally for 5 days. Three out of these patients were found to be vitamin A deficient (low serum levels) and after treatment was observed improvement of the blood levels in all, except two elderly males. However, these authors do not describe which type of liver injury these patients presented, although biochemical evidence of ethanol-induced liver damage were found in many of the patients. By this way, treated patients could be in different stages of liver disease and consequently may present different response to supplementation. Molotkov & Duester60 observed in retinoid- treated (10mg/kg of retinoic acid) mice exposed to acute ethanol intoxication (3,5g/kg) that the reduced levels of serum retinoic acid are due primarily to a decrease in retinoic acid synthesis, and secondarily to an increase in retinoic acid degradation, but both are significant. Since zinc is essential in vitamin A metabolism, concomitant administration of zinc and vitamin A supplementation must be considered. Mobarhan et al.62, evaluated dark adaptation, zinc and retinol levels in a study carried out on 21 hospitalized male patients. Ten patients were found vitamin A deficient and with abnormal dark adaptation and two of these patients, were also zinc deficient. Vitamin A and alcoholic liver disease In another study, Liu et al.51, evaluating the effect of two different doses of chlormethiazole, in rats exposed or not to ethanol, observed that this inhibitor can restore the concentrations of retinol and hepatic retinyl esters in ethanol-fed rats by inhibiting the degradation of vitamin A and the mobilization of vitamin A from the liver to the bloodstream. In another analysis, Leo & Lieber53 observed an increased number of Ito cells after the administration of diet with high concentration of vitamin A (29000UI/day for 2 or 9 months) in rats, but in the groups that received diet containing ethanol and high concentration of vitamin A was observed a decrease in the number of these cells, which matches with the decrease of hepatic vitamin A storage and the appearance of myofibroblasts and collagen fibers. This way, retinoids may play a role in the development of cirrhosis induced by ethanol. These researchers also observed that retinol is not directly responsible for these hepatic alterations since after a chronic ethanol intake, hepatic levels of vitamin A were smaller than when compared to control group, yet signs of toxicity were more prominent. It can be explained by the induction of cytochrome P450 by ethanol, which promotes the hydrolysis of retinoic acid, a physiologically active derivate from retinol, in polar metabolites, which might participate in toxicity23,54-56. Thus, ethanol can potentiate the toxicity of retinol. In order to evaluate the influence of ethanol on vitamin A supplementation and its hepatic effects, Leo et al.52 observed in an experimental study that in rats fed with a high dose of vitamin A (29000UI/day for 8 weeks) there Recently, Dan et al.57 showed that polar retinol metabolites caused marked cytotoxicity in Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição VITAMINA A E DOENÇA HEPÁTICA ALCOÓLICA \| 607 in liver disease patients by a hepatic function impaired. In addition, animal models are useful in exploring basic science questions related to molecular biology, biochemical pathways and genetics that cannot be explored in human beings. Although these experimental studies have shown an adverse interaction between ethanol and retinol, they must be carefully interpreted. However, according to its results, in vitamin A deficiency the administration of pharmacological supplements needs cautions in relation to dose and adverse effects. a concentration and time-dependent manner in HepG2 cells (human hepatoma cell line) and primary hepatocytes from rats fed with ethanol. Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Vitamin A and alcoholic liver disease Those patients were treated with oral vitamin A supplementation (30000UI/day for 4 weeks). Only three patients experienced a rise in serum retinol levels, suggesting that retinol levels may not be sensitive to monitor changing in vitamin A status than dark Decreasing in hepatic retinoic acid may be a mechanism by which alcohol causes hepatocellular carcinoma, since retinoic acid is necessary for the cellular proliferation and differentiation60,61. Only few studies have been carried out investigating the mechanisms leading to low circulating vitamin A levels in alcoholics or chronic liver disease patients, since methodological and ethical aspects limit studies in human beings. Besides, the utilization of hepatic biopsies in order to evaluate hepatic structural changes is limited Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição Revista de Nutrição 608 \| T.P. PAULA et al. 2. Mendez Sanchez N. Alcoholic liver disease. An update. Ann Hepatol. 2005; 4(1):32-42. adaptation. After this period the two zinc-deficient patients remained with abnormal dark adaptation. These two patients were then additionally treated with 220mg of ZnSO4/day for 2 weeks, after which their dark-adapted thresholds returned to normal. By this way, vitamin A and zinc supplementation together may be more efficient than vitamin A alone. 3. Goodman DS. Vitamin A and retinoids in health and disease. N Engl J Med. 1984; 310(16): 1023-31. 4. Hendriks HF, Bosma A, Brouwer A. Fat-storing cells: hyper- and hypovitaminosis A and the relationships with liver fibrosis. Semin Liver Dis. 1993; 13(1): 72-80. 5. Leo MA, Lieber CS. Hepatic vitamin A depletion in alcoholic liver injury. N Engl J Med. 1982; 307(10): 597-601. This review shows that ethanol can either potentiate vitamin A hepatotoxicity or its deficiency. So, other studies are necessary to investigate adequate dose and time of vitamin A supplementation in chronic alcoholism with or without liver disease, when and how to associate it with zinc supplementation. 6. Blomhoff R. Transport and metabolism of vitamin A. Nutr Rev. 1994; 52(2):Pt 2:S13-23. 7. Blomhoff R, Wake K. Perisunusoid stellate cells of the liver: important roles in retinol metabolism and fibrosis. Faseb J. 1991; 5(3):271-7. 8. Quadro L, Hamberger L, Colantuoni V, Gottesman ME, Blaner WS. Understanding the physiological role of retinol-binding protein in vitamin A metabolism using transgenic and knockout mouse models. Mol Aspects Med. 2003; 24(6):421-30. 1. Stewart SF, Day CP. The management of alcoholic liver disease. J Hepatol. 2003; 38(Suppl 1):S2-13. C O N C L U S I O N Ethanol toxicity on liver is a function of duration of alcoholism, amount of daily intake of alcohol and patient’s nutrition. Long-term alcohol intake results in reduced vitamin A ingestion and absorption, enhanced degradation in the liver, increased mobilization of retinol from the liver to other organs and in oxidative stress, increasing vitamin A demands. Moreover, chronic alcoholism results in impaired nutritional status of retinoic acid, the most active derivate of vitamin A, which may promote an environment for tumor formation, since retinoic acid is necessary for the cellular proliferation and differentiation. Indeed, reduced hepatic levels of vitamin A induced by ethanol may also result in hepatic structural alterations. So, in chronic alcoholic subjects, vitamin A serum levels and hepatic reserve need to be closely monitored to avoid deficiency and its adverse effects, although it is important to emphasize that vitamin A supplementation must be done carefully in vitamin A deficient patients, because dose commonly administrated in diseases of others etiology may be toxic for chronic ethanol consumer. 9. Ghyselinck NB, Bavik C, Sapin V, Mark M, Bonnier D, Hindelang C, et al. Cellular retinol-binding protein I is essential for vitamin A homeostasis. EMBO J. 1999; 18(18):4903-14. 10.Dawson HD, Yamamoto Y, Zolfaghari R, Rosales FJ, Dietz J, Shimada T, et al. Regulation of hepatic vitamin A storage in a rat model of controlled vitamin A status during aging. J Nutr. 2000; 130(5): 1280-6. 11.Sommer A. Vitamin A deficiency and its consequences: a field guide to detection and control - epidemiology. 3rd ed. Geneva: World Health Organization; 1995. 65p. 12.Blomhoff R, Green MH, Green JB, Berg T, Norum KR. Vitamin A metabolism: new perspectives on absorption, transport and storage. Physiol Rev. 1991; 171(4):951-90. 13.Janczewska I, Ericzon BG, Eriksson LS. Influence of orthotopic liver transplantation on serum vitamin A levels in patients with chronic liver disease. Scand J Gastroenterol. 1995; 30(1):68-71. 14.Institute of Medicine, Food and Nutrition Board. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, vanadium, manganese, molybdenum, nickel, silicon, vanadium and zinc. Washington (DC): National Academy Press; 2002. R E F E R E N C E S 15.Shiota G, Umeki K, Okano J, Kawasaki H. Hepatocyte growth factor and acute phase proteins in patients with chronic liver diseases. J Med. 1995; 26(5-6):295-308. 1. Stewart SF, Day CP. The management of alcoholic liver disease. J Hepatol. 2003; 38(Suppl 1):S2-13. Revista de Nutrição Revista de Nutrição Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 VITAMINA A E DOENÇA HEPÁTICA ALCOÓLICA \| 609 16.Stephensen CB, Gildengorin G. Serum retinol, the acute phase response, and the apparent misclassification of vitamin A status in the Third National Health and Nutrition Examination Survey. Am J Clin Nutr. 2000; 72(5):1170-8. signaling pathways involved in acethaldeyde induced collagen and gene expression in human hepatic stellate cells. Hepatology. 2001; 33(5):1130-40. 30.Savolainen ER, Leo MA, Timpl R, Lieber CS. Acetaldehyde and lactate stimulate collagen synthesis of cultured baboon liver myofibroblasts. Gastroenterology. 1984; 87(4):777-87. 17.Rocchi E, Borghi A, Paolillo F, Pradelli M, Casalgrandi G. Carotenoids and liposoluble vitamins in liver cirrhosis. J Lab Clin Med. 1991; 118(2):176-85. 31.Moshage H, Casini A, Lieber CS. Acetaldehyde selectively stimulates collagen production in cultured rat liver-storing cells but not in hepatocytes. Hepatology. 1990; 12(3)Pt 1:511-8. 18.Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Willians R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg. 1973; 60(8):646-9. 32.Tanaka Y, Funaki N, Mak IM, Kim C, Lieber CS. Effects of ethanol and hepatic vitamin A on proliferation of lipocytes in regenerating rat liver. J Hepatol. 1991; 12(3) 344-50. 19.Sklan D, Yosefov T, Friedman A. The effects of vitamin A, beta-carotene and canthaxanthin on vitamin A metabolism and immune responses in the chick. Int J Vit Nutr Res. 1989; 59(3):245-50. 33.Davis BH, Vucic A. The effect of retinol on Ito cell proliferation in vitro. Hepatology. 1988; 8(4):788- 93. 20.Albillos A, Hera A. Multifactorial gut barrier failure in cirrhosis and bacterial translocation: working out the role of probiotics and antioxidants. J Hepatol. 2002; 37(4):523-6. 34.Friedman SL, Wei S, Blaner WS. Retinol release by activated rat hepatic lipocytes: regulation by Kupffer cell- conditioned medium and PDGF. Am J Physiol. 1993; 264(5)Pt 1:G947-52. 21.Roenigk HH. Liver toxicity of retinoid therapy. Pharmacol Therap. 1989; 40(1):145-55. 35.Seitz HK. Alcohol and retinoid metabolism. Gut. 2000; 47(6):748-50. 22.Davis BH, Kramer RT, Davidson NO. Retinoic acid modulates rat Ito cell proliferation, collagen, and transforming growth factor beta production. J Clin Invest. 1990; 86(6):2062-70. 36.Stickel F, Hoehn B, Schuppan D, Seitz HK. Nutritional therapy in alcoholic liver disease. Alim Pharmacol Therap. 2003; 18(4):357-73. R E F E R E N C E S The vitamin A spectrum: from deficiency to toxicity. Am J Clin Nutr. 2000; 71(4):878-84. 44.Maio R, Dichi JB, Burini RC. Nutritional consequences of metabolic impairment of macronutrients in chronic liver disease. Arq Gastroenterol. 2000; 37(1):52-7. 56.Russel RM. The impact of disease states as a modifying factor for nutrition toxicity. Nutr Rev. 1997; 55(2):50-3. 45.Bell H, Nilsson A, Norum KR, Pedersen LB, Racknerud N, Rasmussen M. Retinol and retinyl esters in patients with alcoholic liver disease. J Hepatol. 1989; 8(1):26-31. 57.Dan Z, Popov Y, Patsenker E, Preimel D, Liu C, Wang XD, et al. Hepatotoxicity of alcohol-induced polar retinol metabolites involves apoptosis via loss of mitochondrial membrane potential. FASEB J. 2005; 19(7):845-7. 46.Majumdar SK, Shaw GK, Thomson AD. Vitamin A utilization status in chronic alcoholic patients. Int J Vitam Nutr Res. 1983; 53(3):273-9. 58.Deltour L, Ang HL, Duester G. Ethanol inhibition of retinoic acid synthesis as a potential mechanism for fetal alcohol syndrome. FASEB J. 1996; 10(9): 1050-7. 47.Sato M, Lieber CS. Hepatic vitamin A depletion after chronic ethanol consumption in baboons and rats. J Nutr. 1981; 111(11):2015-23. 48.Lieber CS. Cytochrome P-4502E1: its physiological and pathological role. Physiol Rev. 1997; 77(2): 517-44. 59.Wang XD, Liu C, Chung J, Stickel F, Seitz HK, Russel RM. Chronic alcohol intake reduces retinoic acid concentration and enhances AP1 (c-Jun and c-Fos) expression in rat liver. Hepatology. 1998; 28(3): 744-50. 49.Lieber CS. Microsomal ethanol oxidizing system (MEOS): the first 30 years (1968-1998): a review. Alcohol Clin Exp Res. 1999; 23(6):991-1007. 60.Molotkov A, Duester G. Retinol/ethanol drug interaction during acute alcohol intoxication in mice involves inhibition of retinol metabolism to retinoic acid by alcohol dehydrogenase. J Biol Chem. 2002; 277(25):22553-7. 50.Liu C, Russel RM, Seitz HK, Wang XD. Ethanol enhances retinoic acid metabolism into polar metabolites in rat liver via induction of cytochrome P4502E1. Gastroenterology. 2001; 120(1):179-89. 61.Morgan TR, Mandayam S, Jamal MM. Alcohol and hepatocellular carcinoma. Gastroenterology. 2004; 127( 5 Suppl 1):S87-96. 51.Liu C, Chung J, Seitz HK, Russel RM, Wang XD. Chlormethiazole treatment prevents reduced hepatic vitamin A levels in ethanol-fed rats. Alcohol Clin Exp Res. 2002; 26(1):1073-9. 62.Mobarhan S, Russel RM, Underwood BA, Wallingford J, Mathieson RD, Al Midani H. Evaluation of the relative dose response test for vitamin A nutriture in cirrhotics. Am J Clin Nutr. 1981; 34(10):2264-70. 52.Leo MA, Arai M, Sato M, Lieber CS. Hepatotoxicity of vitamin A and ethanol in the rat. Gastroenterology. 1982b; 82(2):194-205. 53.Leo MA, Lieber CS. R E F E R E N C E S 23.Leo MA, Sato M, Lieber CS. Effect of hepatic vitamin A depletion on the liver in humans and rats. Gastroenterology. 1983; 84(3):562-72. 37.Calamita A, Dichi I, Papini-Berto SJ, Dichi JB, Angeleli AV, Vannucchi H. Plasma levels of transthyretin and retinol-binding protein in Child- A cirrhotic patients in relation to protein-calorie status and plasma amino acids, zinc, vitamin A and plasma thyroid hormones. Arq Gastroenterol. 1997; 34(3):139-47. 24.Denk H, Franke WW, Kerjaschki D, Eckerstorfer R. Mallory bodies in experimental animals and man. Int Rev Exp Pathol. 1979; 20:77-121. 25.Franke WW, Schmid E, Kartenbeck, Mayer D, Hacker H, Bannash P, et al. Characterization of the intermediate-sized filaments in liver cells by immunofluorescence and electron microscopy. Biol Cellulaire. 1979; 34:99-110. 38.Masalkar PD, Abhang SA. Oxidative stress and antioxidant status in patients with alcoholic liver disease. Clin Chim Acta. 2005; 355(1-2):61-5. 26.Borenfreund E, Schmid E, Bendich A, Franke WW. Constitutive aggregates of intermediate-sized filaments of the vimentin and cytokeratin type in cultured hepatoma cells and their dispersal by butyrate. Exp Cell Res. 1980; 127(1):215-35. 39.Mobarhan S, Seitz HK, Russel RM, Mehta R, Hupert J, Friedman H, et al. Age-related effects of chronic ethanol intake on vitamin A status in Fisher 344 rats. J Nutr. 1991; 121(4):510-7. 40.Sato M, Lieber CS. Changes in vitamin A status after acute ethanol administration in the rat. J Nutr. 1982; 112(6):1188-96. 27.Leo MA, Lieber CS. Alcohol, vitamin A and beta carotene: adverse interactions, including hepatotoxicity and carcinogenicity. Am J Clin Nutr. 1999; 69(6):1071-85. 41.Ukleja A, Scolapio JS, McConnel JP, Spivey JR, Dickson RC, Ngayen JH, et al. Nutritional assessment of serum and hepatic vitamin A levels in patients with cirrhosis. JPEN J Parenter Enteral Nutr. 2002; 26(3):184-8. 28.Reeves HL, Burt AD, Wood S, Day CP. Hepatic stellate cell activation in the absence of hepatitis in alcoholic liver disease and correlates with the severity of steatosis. J Hepatol. 1996; 25(5): 677-83. 42.McCullough AJ, Bugianesi E. Protein-calorie malnutrition and the etiology of cirrhosis. Am J Gastroenterol. 1997; 92(5):734-8. 29.Svegliati-Baroni G, Ridolfi F, Sario A, Saccomanno S, Bendia E, Benedetti A, et al. Intracellular Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 Revista de Nutrição Revista de Nutrição 610 \| T.P. PAULA et al. content in hepatocellular carcinoma with special reference to alcohol consumption. Hepatology. 1991; 14(5):776-80. 43.Zaína FE, Parolin MB, Lopes RW, Coelho JCU. Prevalence of Malnutrition in Liver Transplant Candidates. Transplant Proc. 2004; 36(4):923-5. 55.Russel RM. Submitted on: 28/2/2005 Final version resubmitted on: 21/3/2006 Approved on: 15/8/2006 Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006 R E F E R E N C E S Hepatic fibrosis after long- term administration of ethanol and moderate vitamin A supplementation in the rat. Hepatology. 1983; 3(1):1-11. Submitted on: 28/2/2005 Final version resubmitted on: 21/3/2006 Approved on: 15/8/2006 Submitted on: 28/2/2005 Final version resubmitted on: 21/3/2006 Approved on: 15/8/2006 54.Adachi S, Moriwaki H, Muto Y, Yamada Y, Fukutomi Y, Shimazaki M, et al. Reduced retinoid Revista de Nutrição Rev. Nutr., Campinas, 19(5):601-610, set./out., 2006
https://openalex.org/W4390479654	https://berkeleypharmatechjournalofmedicine.com/index.php/bptjm/article/download/46/57	English	null	Effective Reprogramming Strategies for Treating Diabetes-Related Cancer: A Focus on Fasentin, Metformin, and Panitumumab Therapies	Berkeley Pharma Tech journal of medicine	2,023	cc-by	6,957	Berkeley Pharma Tech Journal of Medicine By: Heyi Fang, Hailey Guinto, and Harshita Pinneboina Correspondence: Robertfang17@gmail.com Correspondence: Robertfang17@gmail.com Correspondence: Robertfang17@gmail.com Correspondence: Robertfang17@gmail.com Keywords: Metabolic-reprogramming Obesity Diabetes Metformin Fasentin Panitumumab Submitted May 5, 2023 Accepted July 7, 2023 Published December 30, 2023 Submitted May 5, 2023 Accepted July 7, 2023 Published December 30, 2023 Attribution License 4.0 Submitted May 5, 2023 Accepted July 7, 2023 Published December 30, 2023 Abstract Cancer is a condition characterized by the uncontrolled growth and spreading of certain cells within the body. The formation of malignant tumors necessitates a substantial amount of energy to sustain the abnormal rate of cell division. This process leads to a significant alteration in the primary metabolic pathway, transitioning from mitochondrial respiration to aerobic glycolysis, particularly in cancers associated with diabetes. This shift creates an opportunity for less invasive treatment options that can limit cancer growth by targeting specific transporters and enzymes crucial for energy production. This article focuses on the biological functions of Fasentin and Metformin, exploring their effectiveness in constraining cancer development. The discussion delves into their roles in regulating metabolism and highlights how these drugs can be instrumental in impeding the progression of cancer. Full Open Access Creative Commons Full Open Access Creative Commons Attribution License 4.0 Attribution License 4.0 Berkeley Pharma Tech Journal of Medicine \| 102 1. Introduction Since its initial discovery, cancer has been one of the most formidable diseases. Though its early-stage symptoms are less virulent and the rate of recovery can be as high as 90%, advanced stages are faced with a high mortality rate. During intermediate and advanced stages, common treatments are invasive and seek to eradicate all cancer cells -- though recurrence is possible. If the cancer progresses to advanced stages, however, the mortality rate will quickly rise to 70% over the span of Ξve years.1 There are various causes and risk factors for diΛerent types of cancer, but chronic diseases2, high body mass index, (BMI) and unhealthy lifestyles appear as general components in all types. Diabetes mellitus, one of the most common chronic diseases, has proved to be positively correlated with the progression of cancer. Diabetes Mellitus refers to a group of diseases that aΛect how the body uses blood glucose with regulations by Insulin, adiponectin, leptin, and glucose transporter groups (GLUTs) that manipulates the metabolic Οux of metabolites between bloodstream and peripheral tissues with energy needs. Type I Diabetes will disable the pancreas from producing insulin and type II Diabetes is characterized by insulin resistance, which heavily impairs the patient's ability to utilize insulin for the regulation of sugar. Insulin is a hormone secreted by the pancreas and is used extensively for the metabolism of glucose and adipose tissue after food intake, especially for the synthesis of glycogen. Glucose is the major energy source and upon the damage of insulin's function, the homeostasis for glucose and glycogen is broken. Insulin also serves as a negative feedback signal for glycogenolysis, which transforms stored glycogen to free glucose. Moreover, more insulin circulating in the bloodstream will inhibit the synthesis of various proteins in diΛerent tissues, interfering with normal physiological functions throughout the body, especially in the digestive system. In discussion of diabetes and the risk of faster cancer progression, diabetes provides extra energy supply to the uncontrolled dividing cancer cells, Ξlling the gap with an abnormally high consumption of energy. Heavily depending on the intake of energy, the shifted metabolic mechanism is a novel interest for the treatment of diabetes-related cancer. 1. Introduction The complex energy transfer process including the Berkeley Pharma Tech Journal of Medicine \| 102 Berkeley Pharma Tech Journal of Medicine \| 102 use of insulin, glucose transporter, and the change-of-function of many metabolism-related cells in mitochondria are hypothesized as new aims with a non-invasive treatment: treatment with no need of physical insertion of instruments into the patient’s body. In this article, the relationship among the chain of obesity, diabetes mellitus, and cancer development researched by different articles regarding to distinct diabetes-associated cancer types is reviewed, the potential metabolic reprogramming pathways of glucose uptake and digestion by cancer cells are studied, and the mechanisms of inhibitory functions on process of reprogramming by Fasentin, Panitumumab, and Metformin are possessed for the possibility as drug treatments of diabetes-associated cancer.ff 2. Obesity and its Correlation to Diabetes Development Obesity is considered a common risk factor for many diseases3. Regarding the accumulation of excess adipose tissue, a Body Mass Index (BMI) of over 30 kg/m2 is utilized as an indicator. Obesity leads to metabolic disturbance, the severe impairments of the ability to regulate the synthesis of adipocytokines and the conversion between glucose and glycogen. Researchers have revealed that common metabolic disturbances like dyslipidemia and hyperinsulinemia are the causal factors of cancer development. Many adipose cytokines are critical regulators for the maintenance of homeostasis between glucose and energy supply with some of them impacting a range of metabolic pathways to other organs. Two of the most intensively studied adipocytokines are adiponectin and leptin. While both are heavily impacted by the change in adiposity, the change in concentration and distribution of adipose tissue4 impacts the body’s response to metabolites. Adiponectin is secreted by adipose tissue into the bloodstream, with a function of mediating glucose metabolism and fatty acid oxidation process5. Study conducted by Reneau, James et al reveals that the linkage between the accumulation of adipose tissue and secretion of adiponectin shows a negative correlation6. Therefore, the increased adiposity inhibits the production of adiponectin, disabling its function to regulate gluconeogenesis6. Leptin, another cytokine secreted by adipose tissue, has an opposite eΛect to adiponectin. Acting on the central nervous system, namely the hypothalamus7,8, leptin serves to decrease the sense of hunger while inducing Berkeley Pharma Tech Journal of Medicine \| 103 Berkeley Pharma Tech Journal of Medicine \| 103 the activity of beta pancreatic cells that produce insulin. It is shown that leptin production is positively correlated with the increased adiposity, with a positive feedback loop involving increased circulating insulin and the resulting increase of adipose tissue. Both researched adipokines promote the risk of getting diabetes. Furthermore, insulin also plays a critical role in the hydrolyzation of fatty acid, which serves to increase the rate of glucose uptake and fatty acid synthesis, as well as decrease the rate of fatty acid decomposition9. The tumor, upon receiving beneficial signals from the adipose tissue, also feeds back with inhibitory signals on the decomposition of fats. Therefore, a positive feedback loop is observed among obesity-high blood insulin-tumor development, as shown here in figure 1. 2. Obesity and its Correlation to Diabetes Development Furthermore, insulin also plays a critical role in the hydrolyzation of fatty acid, which serves to increase the rate of glucose uptake and fatty acid synthesis, as well as decrease the rate of fatty acid decomposition9. The tumor, upon receiving beneficial signals from the adipose tissue, also feeds back with inhibitory signals on the decomposition of fats. Therefore, a positive feedback loop is observed among obesity-high blood insulin-tumor development, as shown here in figure 1. Figure 1. The positive feedback of adipose tissue accumulation and cancer development.fl Figure 1. The positive feedback of adipose tissue accumulation and cancer development.fl For summary, the abnormally high BMI level indicates the obesity level of an individual, resulting in the Οuctuation of production for critical metabolic regulators, which leads to the increasing level of insulin circulating in the body and subsequently, will cause Type II Diabetes Mellitus by inducing insulin resistance. Over time, the disturbance of homeostasis can generate enough energy to promote the growth of cancer, particularly in organs with high concentrations of adipose tissue. 3. Type II Diabetes Mellitus (T2DM) and its Correlation to Cancer Progression 3. Type II Diabetes Mellitus (T2DM) and its Correlation to Cancer Progression Berkeley Pharma Tech Journal of Medicine \| 104 The compilation of the mentioned negative impacts from the previous section will lead to the development of T2DM. Establishment of insulin resistance (IR) by tissues will be disabled to respond normally to the hormone insulin or downregulate insulin receptors in response to hyperinsulinemia10. The detailed development of IR remains unclear, but the hypothesis is that decrease in insulin sensitivity with adjustments in the PI3K/Akt/mTOR signaling pathway11. Hyperinsulinemia boosts the advancement of cancer by providing access to excess glucose from the bloodstream12. Normal functioning cells, upon receiving glucose, will degrade it for energy with the assistance of oxygen in the process of glycolysis, Citric Acid Cycle, and Oxidative Phosphorylation, producing approximately 34 ATP per glucose used13. However, during the pre-malignant expansion stage of tumor, the development of peripheral Tumor Microenvironment (TME) will separate the interior of tumor cells farther away from the local bloodstream14, which is the carrier of oxygen-rich hemoglobin. The result of this isolation is the leveled average partial pressure of oxygen in TME around 5 mmHg, which is only 12.5% in average of the venous oxygen pressure15. This shift of oxygen level disallows cancer tissue to utilize oxygen for the mitochondrial respiration process and instead, it switches its dependence for energy to the Warburg Effect, the metabolic reprogramming effect done by tumor cells to rely on the energy provided by aerobic glycolysis. This is only effective when there is an excessive supply of glucose and an anaerobic environment, for the production of energy is only 4 ATP per glucose consumed, rather inefficient compared to the normal process16. Therefore, the T2DM with its influence on tissues incapable of utilizing glucose efficiently, the extra circulating glucose in the bloodstream will be recruited and made use of by the tumor complex17. On the other hand, T2DM also provides conveniences for the progression of cancer by the hyper-expressed insulin. Insulin serves as a mitogen in the human body18. With high levels of insulin expressed, cells are excited by the signal from mitogens to proceed with more mitosis. Besides insulin, Insulin-like Growth Factor 1 (IGF1) also serves as an important mitogen. However, IGF1 is not induced by the increased level of blood glucose, but the excess circulating insulin serves to compete with IGF for the constant amount of IGF Binding Berkeley Pharma Tech Journal of Medicine \| 105 Protein 3 (IGFBP3). 3. Type II Diabetes Mellitus (T2DM) and its Correlation to Cancer Progression As IGF1 is a mitogen as well, the imbalance between IGF1-IGFBP3 ratio leads to excess proliferation of cancer cells. On top of that, IGFBP3 is considered a negative regulator of cancer as a low-penetrance tumor suppressor gene19. 4. The Role of Glucose Transporter (GLUT) and Treatment with Fasentin and Panitumumab The two mentioned changes negatively impact the in vivo microbiological environment for all cancer types, because energy is a common restriction for the development of all tumors. This has led the discussion to the molecular metabolic pathway of energy generation, especially the relationship of blood glucose and insulin level, as well as the mechanism of how glucose enters into tissues. In order for the intake of glucose to peripheral tissues from the bloodstream for further mitochondrial oxidation and ATP production, mammalian cells have developed a family of glucose transporter proteins for the transportation of sugar through the plasma membrane20. The abnormal blood glucose level within the T2DM patients will induce a stronger expression of GLUT families, especially in the energy-demanding tissue like adipose tissue and breast cells, the emphasis of our paper. There are 3 classes of GLUT, and the responsible group here is class I GLUT, which contains GLUT1-4 and GLUT1421. DiΛerent GLUT types are responsible for the intake by diΛerent tissues, and our emphasis will be on GLUT4, the one that controls uptake of glucose by skeletal muscle, cardiac muscle, and most importantly for cancer development, adipose tissue22. The expression of GLUT4 is heavily regulated by the level of presenting insulin. Stored in intracellular vesicles, GLUT4 is released vesicular fusion once insulin binds with the insulin receptors expressed on the plasma membrane23. Increased GLUT4 availability to glucose in the bloodstream causes more glucose uptake by fat. This is thought to be the eΛort against the development of Insulin Resistance, although it also promotes the creation of a positive feedback loop for fat accumulation and development of diabetes22. This has led Berkeley Pharma Tech Journal of Medicine \| 106 Berkeley Pharma Tech Journal of Medicine \| 106 the research interest to a potential drug that restricts metabolic efficiency of cancer cells by inhibiting the process of glucose intake. There are two potential targets in regard to this possible map. First, the insulin receptors are Receptor Tyrosine Kinase (RTK), which requires the process of dimerization and phosphorylation to the transduction of the chemical substances from ligands to intracellular electric signals24. This is not unique because many signal regulation pathways utilize RTK as well, and they can also be seen as targets. Therefore, a potential therapy is to block or terminate the phosphorylation process of RTKs with pharmaceutical interventions. 4. The Role of Glucose Transporter (GLUT) and Treatment with Fasentin and Panitumumab There are already drugs developed for cancer treatments via this pathway, namely the Panitumumab25. Panitumumab is an agent serving to restrict progression of colorectal cancer on epithelial level26. Epithelial cancer is a critical cancer type that is affected by T2DM, for the transformation of normal epithelial to cancerous cells requires enormous amounts of energy consumption. Panitumumab inhibits the function of Epithelial-Growth Factor Receptor (EGFR), which after cancerous lesion, starts the process of uncontrolled replication27. Panitumumab, by binding to the extracellular receptor of EGFR and outcompeting the essential nutrients for cell proliferation, effectively inhibits the progression. Recently, clinical trials were made with Panitumumab on its pharmaceutical effect on cancer types other than colorectal cancer. Records show that it can also repress the development of head cancer and neck cancer28. Secondly, the drug can also target the intake of glucose by the direct blockage of GLUT4. Fasentin serves 2 functions for the suppression of tumor establishment. Firstly, Fasentin is a direct inhibitor of GLUT4. By outcompeting the glucose on binding affinity represented by a higher IC50 value29, Fasentin effectively reduces the glucose uptake by cancer-surrounding tissue and therefore, an inhibitory effort is made to reduce energy production. Secondly, Fasentin presence serves as a stimulatory signal for the activation of Fas-directed apoptosis process of cancer30. Therefore, intake of Fasentin can both directly, by inducing cancer cell death by apoptosis and indirectly, by restricting the amount of glucose inflow. Berkeley Pharma Tech Journal of Medicine \| 107 Berkeley Pharma Tech Journal of Medicine \| 107 5. Treatment Incorporating Monocarboxylate Transporter 4 and Cannabinoid Receptor 2 The distorted glucose in cancer cells results in the upregulation of glycolysis in cancer cells. This induces a high amount of lactate production, and consequently, its accumulation in these cells. Blocking upregulation of aerobic glycolysis has been ineΛective, such as using 2-DG as an anti-cancer agent. Mere inhibition of glycolysis is insuΜcient for the eradication of cancer cells due to the reason that cancer cells have the potential to adapt their metabolism to their environmental conditions. Upon glycolytic suppression in multiple types of tumor cells, intracellular energy metabolism is reprogrammed in an autophagy-dependent manner to ensure cellular survival. Rather, a possible treatment method for these cancer cells is aggravation into a hyper-glycemic condition, followed by blocking products from TCA cycle, resulting in high amounts of lactate production. This lactate export is then blocked, causing intracellular acidiΞcation and consequently cell death. The high amount of lactate can cause a strong acidiΞcation process, in which most of the normal cell functions are inhibited, including division. The high amount of lactate can cause a strong acidiΞcation process, in which most of the normal cell functions are inhibited, including division. In addition, lactate released from tumor cells through Monocarboxylate Transporter 4 (MCT4) is enough to stimulate angiogenesis and tumor growth. Increased lactic acid can in turn enhance glycolysis in cancer cells, causing a vicious cycle. However, this high lactate content causes normal cell functions to be inhibited, including the process of replication and division. Using this manner, a treatment of poisoning the cancer cells is feasible by inducing over-production of lactate to decrease pH level. However, research shows that a highly acidic environment surrounding the tumor is responsible for the development of chemotherapy resistance. Thus, two needs that must be met are to increase lactate formation and block the lactate from exiting the cancer cell. A potential pathway of this treatment is the utilization of Monocarboxylate Transporter 4 (MCT4) along with the cannabinoid receptor 2 (CB2).31 Berkeley Pharma Tech Journal of Medicine \| 108 Monocarboxylate transporter 4 (MCT4) is highly expressed in metastatic tumors and at inflammatory sites, referentially in glycolytic muscle fibers and facilitating the lactate efflux. MCT4 is responsible for the bidirectional transport of lactate across the plasma membrane. The CB2 receptor modulates immune cell functions. Cannabinoid receptors (CB1 and CB2)– G-protein coupled receptors, inhibit adenylate cyclase activity in response to psychoactive cannabinoids32. 5. Treatment Incorporating Monocarboxylate Transporter 4 and Cannabinoid Receptor 2 The activation of CB2 receptors does not appear to produce psychotropic effects, and therefore, it may also be helpful in treating diseases that have a neuroinflammatory or neurodegenerative component, such as multiple sclerosis. MAPKs are enzymes involved in a wide variety of important signaling cascades in many cellular responses– cell proliferation, migration, transformation, and cell death. MAPK activation by a nonselective CB2 receptor agonist (D9-THC) was found to have a proapoptotic effect in the Jurkat human leukemia cell line (Herrera et al., 2005) and cytotoxicity in J774-1 macrophages; In the same cells, there was also a c-Jun N-terminal kinase–mediated cytoprotective effect mediated by Δ9-THC activation of CB2 receptors, displaying the same CB2 receptor ligand can activate multiple MAPKs, each with different outcomes33. The activation of CB2 receptors by natural or synthetic ligands favors a range of receptor conformations that can variably affect different signaling pathways as the following procedure– inhibition of adenylyl cyclase, decreased cAMP (production, and less activation of cAMP-dependent protein kinase (PKA), inhibiting A-type potassium channels as well as specific gene expression. This is followed by activation of Akt/protein kinase B– stimulating cell survival, migration, and growth. Proceeding activation of the mitogen-activated protein kinase (MAPK) cascade favors cell survival and modulates gene expression. In addition, there is inhibition of specific calcium channels and enhanced opening of G protein–gated inwardly rectifying potassium (GIRK) channels. Lastly, stimulation of de novo synthesis of ceramide and inhibition of the MAPK cascade promotes apoptosis34,35. Recruitment of b-arrestin to the activated CB2 receptor results in desensitization and/or internalization of the receptor and potential activation of arrestin-specific signaling. Decreased PKA activity increases Raf-1 to stimulate the MAPK cascade, positively regulating the Berkeley Pharma Tech Journal of Medicine \| 109 expression of many genes and indicating activation of a pathway by CB2 receptor agonists. Functional selectivity of CB2 receptor agonists must be considered during the therapeutic development of CB2 agonists, which increases the possibilities for developing drugs targeting CB2 receptors. By forcing glycolysis with metformin and a NF-kB inhibitor lowers the pH of a cell, our treatment has a minor effect on normal cells and most effective in cancer patients with any hyperglycemic stages, including. As an example, NK-kB inhibition causes increased lactate secretion from breast cancer cell line MCF-736,37. The effect of this metabolic reprogramming strategy was observed by checking oxygen consumption and extracellular acidification rates. 5. Treatment Incorporating Monocarboxylate Transporter 4 and Cannabinoid Receptor 2 This resulted in reduced OCR and blocked ECAR, essentially leading to an accumulation of lactate. Fluorescent BCFL-AM was used as a probe for detecting pHi. Therefore, the reprogramming did not decrease viability in (normal) MCF-12A cells. It did, however, affect MCF-7, T47D, and MDA-MB-231 (breast cancer cells), demonstrating significantly inhibited migration and invasion ability38. Consequently, in high glucose cancer cells, a higher glycolysis rate is present, allowing the reprogramming strategy to further promote the process and raise intracellular lactate, permitting cancer cells to poison themselves. Essentially, there is limited cytotoxic effect on non-subject cells, reducing the chance of untargeted damage. CB-2 might bind to a pocket of MCT4 composed of Ser156, Phe243, Tyr332, Gln339 and Glu363. The combination of Metformin and CB-2 exerts a deleterious effect on breast cancer cell viability and exhibits synergistic antitumor effects. Results of the combination treatment showed a 63% inhibition of cell viability in MDA-MB-231 breast cancer cells39. In addition, moderate effect on cell viability was observed in normal MCF-10A human mammary epithelial cells. Disrupting MCT4 function leads to an accumulation of intracellular lactate and a decrease in intracellular pH which may rapidly damage a cell thereby inducing necrosis, apoptosis or growth arrest. Metformin increases glycolysis thereby increasing the buildup of intracellular lactate, thereby accelerating the effects of the MCT4 inhibitor. This can be used to treat high glycolytic rate/MCT4-expressing malignancies. Another possible combination is the MCT4 inhibitor CB-2 and a GLS1 inhibitor CB-839 to reengineer cancer metabolism40. This combination solves Berkeley Pharma Tech Journal of Medicine \| 110 the problem of increasing ammonia production to neutralize lactate via restricting the metabolic Οexibility of these cancer cells. and mTOR Pathway The collection of genetic alterations in cancer cells causes interference with the regular cellular signaling pathways, this then leads to cell growth for cancer. Despite the diΛerent current treatment plans that are oΛered, many recoveries fail due to the drug resistance and its adverse side eΛects. Though in recent studies, metabolic reprogramming has served as a possible cancer therapy. With this, we want to identify the glucose metabolism of cancer cells to alter so it causes glucose-lowering agents like metformin to be a possible treatment in cancer cells. One change we can see in cancer metabolism is known as the Warburg eΛect. The metabolic adaptation shifts their energy production from oxidative phosphorylation in the process of aerobic glycolysis. Metformin is a commonly prescribed drug used for type 2 diabetes, which also displays anticancer properties in inhibiting mitochondrial complex I,41,42 activating AMPK, a regulator for energy metabolism, and reducing insulin and insulin-like growth factor 1 (IGF-1), which performs anti-tumor functions. In addition, Metformin has shown ability to inhibit the mTOR pathway, which is involved in the protein synthesis and cell growth process. There are two types of forms of mTOR: complex 1 (mTORC1) used in growth factors, glucose, and helps with protein synthesis; and complex 2 (mTORC1) used for regulating cell survival and metabolism43. Metformin inhibits mTORC1 which then activates AMPK and phosphorylates the TSC2 protein, the negative regulator of mTOC1. This inhibition can occur both dependent and independent of AMP-activated protein kinase (AMPK) activation, leading to the decrease in protein synthesis and cell growth. Metformin is able to inhibit mTORC1 independently when AMPK is activated as it is binding to the complex.43 This dual mechanism causes a positive treatment for the host, since metformin has multiple strategies to repress pathways critical for cancer cell growth. Berkeley Pharma Tech Journal of Medicine \| 111 Berkeley Pharma Tech Journal of Medicine \| 111 Recent studies with mice given metformin after being exposed to carcinogen show a reduction of lung tumor burden by up to 53%. However, only modest effects presented as mTOR was inhibited in lung tumors 45. The researchers then inject mice with metformin to assess whether this method would improve mTOR inhibition. The result shows that plasma levels of metformin were higher after injection than oral administration. and mTOR Pathway On the other hand, Metformin also activates AMPK and inhibits mTOR in liver tissue, but it only inhibits phosphorylation process of IGF-IR/IR, Akt, ERK, and mTOR in lung tissue. This suggests that Metformin indirectly inhibited mTOR in lung tissue by decreasing activation of IGF-1R/IR and Akt upstream of mTOR45. A follow-up study showed that intraperitoneal administration of metformin decreased tumor burden by 72%, which correlates with decreased cellular proliferation and marked inhibition of mTOR in tumors, as shown in figure 2. Figure 2. As the intake of Metformin treatment inhibits plasma IGF-1 level in all tissues. As mentioned, the Metformin inhibition also has specificity in lung and liver tissue. Figure 2. As the intake of Metformin treatment inhibits plasma IGF-1 level in all tissues. As mentioned, the Metformin inhibition also has specificity in lung and liver tissue. Metformin swords decrease with the phosphorylation of IGF-1 and the insulin receptors in lung tissues. The drug is known for reducing the levels of hormones presentation in patients. With oral administration of Metformin, study shows a decrease of 1 or 5 mg/ml of circulating IGF-1 by approximately 20% and the insulin by 20% and 35%44. Berkeley Pharma Tech Journal of Medicine \| 112 Berkeley Pharma Tech Journal of Medicine \| 112 There was an injection of metformin intraperitoneally to assess its inhibitory effect on mTOR pathway. However, the intraperitoneal injection did not decrease the IGF-I levels significantly. Instead, the levels of IGF-1 inside the mice shows a significant reduction from the wild type comparison group. This can indicate that stress in daily injection can alter the masked inhibitory effect of metformin in circulating the levels of IGF-1. This supports the hypothesis that metformin can be used in mTOR pathways with cancer patients, since it will decrease the levels of circulating IGF-1 (insulin) in preventing the NNK-induced lung tumorigenesis45. In addition to its ability to inhibit the mTOR pathway, Metformin triggers a decrease in cap-dependent translation. A study utilizes MCF-7 cells, Metformin treatment and led to a maximal inhibition of 40% in cap-dependent translation45. The polysome profile analysis shows how the metformin treatment of MCF-7 cells leads to a shift of mRNAs from heavy to light polysomes and how concomitant increased 80s ribosomes46. This suggests that metformin can be a treat and cause significant impact on the translation of specific mRNAs and leads to alteration of protein expression in cancer cells. and mTOR Pathway The change in polysome profiles towards lighter polysomes indicates a reduction in the translation efficiency of specific mRNAs. This can help with further implication for cancer cells that rely on increased protein synthesis for their survival and proliferation47. The increase of 80s ribosomes indicates that Metformin treatment may also affect the biogenesis of ribosomes, which is essential for protein synthesis. This decreases the rate of protein synthesis within cancer hosts, which can then be added to the growing body of evidence supporting the potential of Metformin as a therapeutic agent for cancer. In MCF-7 breast cancer cells, Metformin has been shown to inhibit translation initiation by activating SMP-activated protein kinase (AMPK) through its upstream kinase, liver kinase b1 (LKB1)48. This results in the inhibition of the mammalian target of mTORC1. Although in contrast, in MDA-MB-231 breast cancer cells, didn’t express LKB1 mRNA, metformin had no effect on its protein synthesis, confirming the requirement of LKB1 for inhibition of translation by metformin in MCR-7 breast cancer cells49. This confirms the requirement of LKB1 for inhibition of translation by metformin in MCF-7 Berkeley Pharma Tech Journal of Medicine \| 113 Berkeley Pharma Tech Journal of Medicine \| 113 breast cancer cells. It is an important biomarker for predicting the response to metformin treatment. Berkeley Pharma Tech Journal of Medicine \| 114 7. Conclusion As discussed, the development of diabetes is a result of imbalance between glucose metabolism and anabolism. This distorted metabolic system can be caused by abnormally increased glucose intake and the subsequent development of insulin resistance, causing T2DM. Furthermore, the establishment of T2DM, along with an anaerobic environment that is beneΞcial for the cancer progression, will reshape the metabolic dependence of tumor from mitochondrial respiration to aerobic glycolysis. Extensive supply of glucose can be advantageous for cancers with high energy demand to fulΞll its need of rapid division and proliferation. Therefore, a new target for non-invasive cancer treatment emerges. By limiting the energy inΟux to tumor tissue with either direct blockage on membrane receptors or indirect pathways that induce speciΞc cell killing, three of the potential candidates are available for consideration, which are Panitumumab, Fasentin, and Metformin. Even though all the mentioned treatments are proven to be eΛective to some extent, assistance from corresponding drugs and therapies is highly recommended since some of the directed pathways are not exclusive to cancer, which leaves potential risk of aΛecting normal cell functioning. Moreover, many of the mentioned pathways are distinct from each other, which gives the possibility of combinational use, similar to a cocktail therapy for HIV patients50. References 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer 2. Tu H, Wen CP, Tsai SP, et al. Cancer risk associat with chronic diseases and disease markers: prospecti cohort study. BMJ. 2018;360:k134. Published 2018 31. doi:10.1136/bmj.k134 3. Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crises in adult patients with diabetes. Diabetes Care. 2009;32(7):1335-1343. doi:10.2337/dc09-9032 4. Ghoshal, K., Chatterjee, T., Chowdhury, S. et al. Adiponectin Genetic Variant and Expression Coupled with Lipid Peroxidation Reveal New Signatures in Diabetic Dyslipidemia. Biochem Genet 59, 781–798 (2021). https://doi.org/10.1007/s10528-021-10030-5 5. Reneau, James et al. “Effect of adiposity on tissue-specific adiponectin secretion.” PloS one vol. 13,6 e0198889. 20 Jun. 2018, doi:10.1371/journal.pone.0198889 6. Margetic, S et al. “Leptin: a review of its peripheral actions and interactions.” International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity vol. 26,11 (2002): 1407-33. doi:10.1038/sj.ijo.0802142 7. Dimitriadis, George et al. “Insulin effects in muscle and adipose tissue.” Diabetes research and clinical practice vol. 93 Suppl 1 (2011): S52-9. doi:10.1016/S0168-8227(11)70014-6 8. Forny-Germano L, De Felice FG, Vieira References MNDN. The Role of Leptin and Adiponectin in Obesity-Associated Cognitive Decline and Alzheimer's Disease. Front Neurosci. 2019;12:1027. Published 2019 Jan 14. doi:10.3389/fnins.2018.01027 8. Forny-Germano L, De Felice FG, Vieira 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer FDA Fact Sheet, Panitumumab use. https://www.accessdata.fda.gov/drugsatfda_docs/ label/2009/125147s080lbl.pdf 25. Zhen, Y et al. “Knockdown of EGFR inhibits growth and invasion of gastric cancer cells.” Cancer gene therapy vol. 21,11 (2014): 491-7. doi:10.1038/cgt.2014.55 26. Clinicaltrial.gov, searching criteria Panitumumab. https://www.clinicaltrials.gov/ct2/results?cond=P anitumumab&term=&cntry=US&state=&city= &dist=&Search=Search 27. Ocaña, Mª Carmen et al. “Fasentin diminishes endothelial cell proliferation, differentiation and invasion in a glucose metabolism-independent manner.” Scientific reports vol. 10,1 6132. 9 Apr. 2020, doi:10.1038/s41598-020-63232-z 28. Chou, Ping‐Chieh, et al. “Impact of Diabetes on Promoting the Growth of Breast Cancer.” C C i i l 41 5 2021 Opportunities. Cancers (Basel). 2021;13(9):2053. Published 2021 Apr 23. doi:10.3390/cancers13092053 alterations in Alms1GT/GT, a mouse model for obesity and insulin resistance.” PloS one vol. 9,10 e109540. 9 Oct. 2014, doi:10.1371/journal.pone.0109540 15. Vander Heiden, Matthew G et al. “Understanding the Warburg effect: the metabolic requirements of cell proliferation.” Science (New York, N.Y.) vol. 324,5930 (2009): 1029-33. doi:10.1126/science.1160809 15. Vander Heiden, Matthew G et al. “Understanding the Warburg effect: the metabolic requirements of cell proliferation.” Science (New York, N.Y.) vol. 324,5930 (2009): 1029-33. doi:10.1126/science.1160809 22. Hubbard, S R, and J H Till. “Protein tyrosine kinase structure and function.” Annual review of biochemistry vol. 69 (2000): 373-98. doi:10.1146/annurev.biochem.69.1.373 16. Jogie-Brahim, Sherryline et al. “Unraveling insulin-like growth factor binding protein-3 actions in human disease.” Endocrine reviews vol. 30,5 (2009): 417-37. doi:10.1210/er.2008-0028 23. Lemmon, Mark A, and Joseph Schlessinger. “Cell signaling by receptor tyrosine kinases.” Cell vol. 141,7 (2010): 1117-34. doi:10.1016/j.cell.2010.06.011 16. Jogie-Brahim, Sherryline et al. “Unraveling insulin-like growth factor binding protein-3 actions in human disease.” Endocrine reviews vol. 30,5 (2009): 417-37. doi:10.1210/er.2008-0028 17. Olson, A L, and J E Pessin. “Structure, function, and regulation of the mammalian facilitative glucose transporter gene family.” Annual review of nutrition vol. 16 (1996): 235-56. doi:10.1146/annurev.nu.16.070196.001315 18. Call R, Grimsley M, Cadwallader L, et al. Insulin--carcinogen or mitogen? Preclinical and clinical evidence from prostate, breast, pancreatic, and colorectal cancer research. Postgrad Med. 2010;122(3):158-165. 19. Bell, G I et al. “Molecular biology of mammalian glucose transporters.” Diabetes care vol. 13,3 (1990): 198-208. doi:10.2337/diacare.13.3.198 19. Bell, G I et al. “Molecular biology of mammalian glucose transporters.” Diabetes care vol. 13,3 (1990): 198-208. doi:10.2337/diacare.13.3.198 27. Ocaña, Mª Carmen et al. “Fasentin diminishes endothelial cell proliferation, differentiation and invasion in a glucose metabolism-independent manner.” Scientific reports vol. 10,1 6132. 9 Apr. 2020, doi:10.1038/s41598-020-63232-z 20. Shepherd, P R et al. 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer 9. Kennedy A, Gettys TW, Watson P, et al. The metabolic significance of leptin in humans: gender-based differences in relationship to adiposity, insulin sensitivity, and energy expenditure. J Clin Endocrinol Metab. 1997;82(4):1293-1300. doi:10.1210/jcem.82.4.3859 9. Kennedy A, Gettys TW, Watson P, et al. The metabolic significance of leptin in humans: gender-based differences in relationship to adiposity, insulin sensitivity, and energy expenditure. J Clin Endocrinol Metab. 1997;82(4):1293-1300. doi:10.1210/jcem.82.4.3859 2. Tu H, Wen CP, Tsai SP, et al. Cancer risk associated with chronic diseases and disease markers: prospective cohort study. BMJ. 2018;360:k134. Published 2018 Jan 31. doi:10.1136/bmj.k134 3. Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crises in adult patients with diabetes. Diabetes Care. 2009;32(7):1335-1343. doi:10.2337/dc09-9032 10. Wang, Guanyu. “Singularity analysis of the AKT signaling pathway reveals connections between cancer and metabolic diseases.” Physical biology vol. 7,4 046015. 22 Dec. 2010, doi:10.1088/1478-3975/7/4/046015 4. Ghoshal, K., Chatterjee, T., Chowdhury, S. et al. Adiponectin Genetic Variant and Expression Coupled with Lipid Peroxidation Reveal New Signatures in Diabetic Dyslipidemia. Biochem Genet 59, 781–798 (2021). 11. Gatenby, R., Gillies, R. Why do cancers have high aerobic glycolysis?. Nat Rev Cancer 4, 891–899 (2004). 5. Reneau, James et al. “Effect of adiposity on tissue-specific adiponectin secretion.” PloS one vol. 13,6 e0198889. 20 Jun. 2018, doi:10.1371/journal.pone.0198889 12. Del Puerto-Nevado L, Santiago-Hernandez A, Solanes-Casado S, et al. Diabetes-mediated promotion of colon mucosa carcinogenesis is associated with mitochondrial dysfunction. Mol Oncol. 2019;13(9):1887-1897. 6. Margetic, S et al. “Leptin: a review of its peripheral actions and interactions.” International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity vol. 26,11 (2002): 1407-33. doi:10.1038/sj.ijo.0802142 13. Zhang, Daoxiang et al. “Metabolic reprogramming of cancer-associated fibroblasts by IDH3α downregulation.” Cell reports vol. 10,8 (2015): 1335-48. 7. Dimitriadis, George et al. “Insulin effects in muscle and adipose tissue.” Diabetes research and clinical practice vol. 93 Suppl 1 (2011): S52-9. doi:10.1016/S0168-8227(11)70014-6 8. Forny-Germano L, De Felice FG, Vieira Berkeley Pharma Tech Journal of Medicine \| 115 alterations in Alms1GT/GT, a mouse model for obesity and insulin resistance.” PloS one vol. 9,10 e109540. 9 Oct. 2014, doi:10.1371/journal.pone.0109540 22. Hubbard, S R, and J H Till. “Protein tyrosine kinase structure and function.” Annual review of biochemistry vol. 69 (2000): 373-98. doi:10.1146/annurev.biochem.69.1.373 23. Lemmon, Mark A, and Joseph Schlessinger. “Cell signaling by receptor tyrosine kinases.” Cell vol. 141,7 (2010): 1117-34. doi:10.1016/j.cell.2010.06.011 24. 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer “Adipose cell hyperplasia and enhanced glucose disposal in transgenic mice overexpressing GLUT4 selectively in adipose tissue.” The Journal of biological chemistry vol. 268,30 (1993): 22243-6. 28. Chou, Ping‐Chieh, et al. “Impact of Diabetes on Promoting the Growth of Breast Cancer.” Cancer Communications, vol. 41, no. 5, 2021, doi:10.1002/cac2.12147. doi:https://doi.org/10.1016/j.ejphar.2018.11.028 28. Chou, Ping‐Chieh, et al. “Impact of Diabetes on Promoting the Growth of Breast Cancer.” Cancer Communications, vol. 41, no. 5, 2021, doi:10.1002/cac2.12147. 21. Favaretto, Francesca et al. “GLUT4 defects in adipose tissue are early signs of metabolic Berkeley Pharma Tech Journal of Medicine \| 116 Berkeley Pharma Tech Journal of Medicine \| 116 29. Kraus D, Reckenbeil J, Veit N, et al. Targeting glucose transport and the NAD pathway in tumor cells with STF-31: a re-evaluation. Cell Oncol (Dordr). 2018;41(5):485-494. doi:10.1007/s13402-018-0385-5 cancer cells. Cancer Res. 2010;70:2465–2475. – PubMed 29. Kraus D, Reckenbeil J, Veit N, et al. Targeting glucose transport and the NAD pathway in tumor cells with STF-31: a re-evaluation. Cell Oncol (Dordr). 2018;41(5):485-494. doi:10.1007/s13402-018-0385-5 cancer cells. Cancer Res. 2010;70:2465–2475. – PubMed 36. Chou PC, Choi HH, Huang Y, et al. Impact of diabetes on promoting the growth of breast cancer. Cancer Commun (Lond). 2021;41(5):414-431. doi:10.1002/cac2.12147 36. Chou PC, Choi HH, Huang Y, et al. Impact of diabetes on promoting the growth of breast cancer. Cancer Commun (Lond). 2021;41(5):414-431. doi:10.1002/cac2.12147 30. Wood TE, Dalili S, Simpson CD, et al. A novel inhibitor of glucose uptake sensitizes cells to FAS-induced cell death. Mol Cancer Ther. 2008;7(11):3546-3555. 30. Wood TE, Dalili S, Simpson CD, et al. A novel inhibitor of glucose uptake sensitizes cells to FAS-induced cell death. Mol Cancer Ther. 2008;7(11):3546-3555. 37. Goodwin PJ, Chen BE, Gelmon KA, et al. Effect of Metformin vs Placebo on Invasive Disease-Free Survival in Patients With Breast Cancer: The MA.32 Randomized Clinical Trial. JAMA. 2022;327(20):1963-1973. doi:10.1001/jama.2022.6147 doi:10.1158/1535-7163.MCT-08-0569 31. Zakikhani, Mahvash, et al. “Metformin Is an AMP Kinase–Dependent Growth Inhibitor for Breast Cancer Cells.” Cancer Research, vol. 66, no. 21, 2006, pp. 10269–10273., 38. Au Yeung SL, Luo S, Schooling CM. The impact of GDF-15, a biomarker for metformin, on the risk of coronary artery disease, breast and colorectal cancer, and type 2 diabetes and metabolic traits: a Mendelian randomisation study. Diabetologia. 2019;62(9):1638-1646. doi:10.1007/s00125-019-4913-2 doi:10.1158/0008-5472.can-06-1500. 32. Dhopeshwarkar A, Mackie K. CB2 Cannabinoid receptors as a therapeutic target-what does the future hold?. Mol Pharmacol. 2014;86(4):430-437. doi:10.1124/mol.114.094649 doi:10.1124/mol.114.094649 39. Lee SH, Hwang HK, Lee WJ, Kang CM. 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer MCT4 as a potential therapeutic target to augment gemcitabine chemosensitivity in resected pancreatic cancer. Cell Oncol (Dordr). 2021;44(6):1363-1371. doi:10.1007/s13402-021-00643-8Chou PC, Choi HH, Huang Y, et al. Impact of diabetes on promoting the growth of breast cancer. Cancer Commun (Lond). 2021;41(5):414-431. doi:10.1002/cac2.12147 33. Dowling, Ryan J.O., et al. “Metformin Inhibits Mammalian Target of Rapamycin–Dependent Translation Initiation in Breast Cancer Cells.” Cancer Research, vol. 67, no. 22, 2007, pp. 10804–10812., doi:10.1158/0008-5472.can-07-2310. doi:10.1158/0008-5472.can-07-2310. doi:10.1158/0008-5472.can-07-2310. 34. Apr 25, 2019. (n.d.). Novel MCT4 inhibitors and uses thereof. Justia. Retrieved April 8, 2023, from https://patents.justia.com/patent/20190352282 35. Ben Sahra I., Laurent K., Giuliano S., Larbret F., Ponzio G., Gounon P., Le Marchand-Brustel Y., Giorgetti-Peraldi S., Cormont M., Bertolotto C. Targeting cancer cell metabolism: the combination of metformin and 2-deoxyglucose induces p53-dependent apoptosis in prostate 34. Apr 25, 2019. (n.d.). Novel MCT4 inhibitors and uses thereof. Justia. Retrieved April 8, 2023, from 34. Apr 25, 2019. (n.d.). Novel MCT4 inhibitors and uses thereof. Justia. Retrieved April 8, 2023, from https://patents.justia.com/patent/20190352282 40. Goldberg FW, Kettle JG, Lamont GM, et al. Discovery of Clinical Candidate AZD0095, a Selective Inhibitor of Monocarboxylate Transporter 4 (MCT4) for Oncology. J Med Chem. 2023;66(1):384-397. doi:10.1021/acs.jmedchem.2c01342 https://patents.justia.com/patent/20190352282 35. Ben Sahra I., Laurent K., Giuliano S., Larbret F., Ponzio G., Gounon P., Le Marchand-Brustel Y., Giorgetti-Peraldi S., Cormont M., Bertolotto C. Targeting cancer cell metabolism: the combination of metformin and 2-deoxyglucose induces p53-dependent apoptosis in prostate 35. Ben Sahra I., Laurent K., Giuliano S., Larbret F., Ponzio G., Gounon P., Le Marchand-Brustel Y., Giorgetti-Peraldi S., Cormont M., Bertolotto C. Targeting cancer cell metabolism: the combination of metformin and 2-deoxyglucose induces p53-dependent apoptosis in prostate 41. Lord SR, Cheng WC, Liu D, et al. Integrated Pharmacodynamic Analysis Identifies Two Berkeley Pharma Tech Journal of Medicine \| 117 Metabolic Adaption Pathways to Metformin in Breast Cancer. Cell Metab. 2018;28(5):679-688.e4. Metabolic Adaption Pathways to Metformin in Breast Cancer. Cell Metab. 2018;28(5):679-688.e4. doi:10.1158/0008-5472.CAN-07-2310 48. Vernieri C, Signorelli D, Galli G, et al. Exploiting FAsting-mimicking Diet and MEtformin to Improve the Efficacy of Platinum-pemetrexed Chemotherapy in Advanced LKB1-inactivated Lung Adenocarcinoma: The FAME Trial. Clin Lung Cancer. 2019;20(3):e413-e417. doi:10.1016/j.cllc.2018.12.011 doi:10.1016/j.cmet.2018.08.021 doi:10.1016/j.cmet.2018.08.021 42. Birsoy K., Wang T., Chen W.W., Freinkman E., Abu-Remaileh M., Sabatini D.M. An essential role of the mitochondrial electron transport chain in cell proliferation is to enable aspartate synthesis. Cell. 2015;162:540–551. - PMC - PubMed 42. Birsoy K., Wang T., Chen W.W., Freinkman E., Abu-Remaileh M., Sabatini D.M. 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer An essential role of the mitochondrial electron transport chain in cell proliferation is to enable aspartate synthesis. Cell. 2015;162:540–551. - PMC - PubMed 49. Marinello PC, da Silva TN, Panis C, et al. Mechanism of metformin action in MCF-7 and MDA-MB-231 human breast cancer cells involves oxidative stress generation, DNA damage, and transforming growth factor β1 induction. Tumour Biol. 2016;37(4):5337-5346. doi:10.1007/s13277-015-4395-x 43. U.S. National Library of Medicine. (n.d.). Integrated pharmacodynamic analysis identifies two metabolic adaption pathways to metformin in breast cancer. Cell metabolism. Retrieved April 8, 2023, from p 2023, from https://pubmed.ncbi.nlm.nih.gov/30244975/ 44. 44. Anders S., Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010;11:R106. - PMC - PubMed 45. Memmott, Regan M., et al. “Metformin Prevents Tobacco Carcinogen–Induced Lung Tumorigenesis.” Cancer Prevention Research, vol. 3, no. 9, 2010, pp. 1066–1076., doi:10.1158/1940-6207.capr-10-0055. 46. Wilson PM, Yang D, Azuma M, et al. Intratumoral expression profiling of genes involved in angiogenesis in colorectal cancer patients treated with chemotherapy plus the VEGFR inhibitor PTK787/ZK 222584 (vatalanib). Pharmacogenomics J. 2013;13(5):410-416. doi:10.1038/tpj.2012.23 47. Dowling RJ, Zakikhani M, Fantus IG, Pollak M, Sonenberg N. Metformin inhibits mammalian target of rapamycin-dependent translation initiation in breast cancer cells. Cancer Res. 2007;67(22):10804-10812. https://pubmed.ncbi.nlm.nih.gov/30244975/ 50. Gilad Y, Gellerman G, Lonard DM, O'Malley BW. Drug Combination in Cancer Treatment-From Cocktails to Conjugated Combinations. Cancers (Basel). 2021;13(4):669. Published 2021 Feb 7. d i 10 3390/ 13040669 44. 44. Anders S., Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010;11:R106. - PMC - PubMed 44. 44. Anders S., Huber W. Differential expression analysis for sequence count data. Genome Biol. 2010;11:R106. - PMC - PubMed 45. Memmott, Regan M., et al. “Metformin Prevents Tobacco Carcinogen–Induced Lung Tumorigenesis.” Cancer Prevention Research, vol. 3, no. 9, 2010, pp. 1066–1076., doi:10.1158/1940-6207.capr-10-0055. 46. Wilson PM, Yang D, Azuma M, et al. Intratumoral expression profiling of genes involved in angiogenesis in colorectal cancer patients treated with chemotherapy plus the VEGFR inhibitor PTK787/ZK 222584 (vatalanib). Pharmacogenomics J. 2013;13(5):410-416. doi:10.1038/tpj.2012.23 46. Wilson PM, Yang D, Azuma M, et al. Intratumoral expression profiling of genes involved in angiogenesis in colorectal cancer patients treated with chemotherapy plus the VEGFR inhibitor PTK787/ZK 222584 (vatalanib). Pharmacogenomics J. 2013;13(5):410-416. doi:10.1038/tpj.2012.23 47. Dowling RJ, Zakikhani M, Fantus IG, Pollak M, Sonenberg N. Metformin inhibits mammalian target of rapamycin-dependent translation initiation in breast cancer cells. Cancer Res. 2007;67(22):10804-10812. 47. Dowling RJ, Zakikhani M, Fantus IG, Pollak M, Sonenberg N. Berkeley Pharma Tech Journal of Medicine \| 118 1. World Health Organization, Cancer Overview. https://www.who.int/news-room /fact-sheets/detail/cancer Metformin inhibits mammalian target of rapamycin-dependent translation initiation in breast cancer cells. Cancer Res. 2007;67(22):10804-10812. Berkeley Pharma Tech Journal of Medicine \| 118
https://openalex.org/W2339877969	https://academicjournals.org/journal/JGRP/article-full-text-pdf/A75C2FB57229	English	null	Drainage basin morphometry of the Encadenadas del Oeste lakes, Argentina	Journal of geography and regional planning	2,016	cc-by	9,695	Full Length Research Paper Drainage basin morphometry of the Encadenadas del Oeste lakes, Argentina Alejandra Mabel Geraldi1,2, María Cintia Piccolo1,2 and Gerardo Miguel Eduardo Perillo1,2 1Argentine Institute of Oceanography. Camino de La Carrindanga km 7 Bahía Blanca, Buenos Aires, Argentina. 2Universidad Nacional del Sur. 12 de Octubre y San Juan. Bahía Blanca. 8000. Argentina. 1Argentine Institute of Oceanography. Camino de La Carrindanga km 7 Bahía Blanca, Buenos Aires, Argentina. 2Universidad Nacional del Sur. 12 de Octubre y San Juan. Bahía Blanca. 8000. Argentina. Received 14 June 2009: Accepted 25 February, 2011 The Las Encadenadas basin can be defined as an endorheic fluviolacustrine system. The aim of this study is to identify hydrographic sectors and subbasins within the Encadenada’s drainage basin and analyze the former’s morphometric properties including hypsometry. The morphometric analysis allowed for quantification of variables and indices for example area, perimeter, total length of streams, etc. Hypsometric curves were also plotted for each subbasin and finally, principal components analysis was used to sort basins based on results from individually calculated parameters and indices. This study’s aim was to define for the first time the various drainage subbasins that comprise the Encadenadas del Oeste’s basin. The characterization of these units shows that the basin is morphologically diverse due to the dynamic fluvial activity that prevails within its limits. One of the above mentioned morphological units are the alluvial cones which form at the mouths of the mainstreams and delineate the bases of the different subbasins. The drainage network exhibits overall a low level of ramification and hierarchy which is likely due to the sedimentary nature and high permeability of the sub-surface soil. Keywords: Basin, morphometric, drainage subbasins, hypsometry. Vol. 9(2), pp. 12-27, February, 2016 DOI: 10.5897/JGRP09.050 Article Number: A75C2FB57229 ISSN 2070-1845 Copyright © 2016 Author(s) retain the copyright of this article http://www.academicjournals.org/JGRP Journal of Geography and Regional Planning Vol. 9(2), pp. 12-27, February, 2016 DOI: 10.5897/JGRP09.050 Article Number: A75C2FB57229 ISSN 2070-1845 Copyright © 2016 Author(s) retain the copyright of this article http://www.academicjournals.org/JGRP Journal of Geography and Regional Planning Journal of Geography and Regional Planning INTRODUCTION C ó Silva et al. (2006) analyzed and compared the morphometric attributes of some drainage basins located in Brazil. Their characterization is based on various parameters and indices such as area, perimeter, mainstream length, compactness index, drainage density, etc. They linked these parameters to the geological characteristics of the area and demonstrated that morphology is controlled by the main regional rock type. Basins overlying sedimentary bedrock and basins with a dissected relief exhibit lower current density and compactness index values than those with a smooth topography. The regional morphostructure underlying the basin is divided into two units: the southward Norventánica prairie (Fidalgo et al., 1975) which includes the Positivo Bonaerense (Ventania mountain range) (Yrigoyen, 1975) and the northward Western sandy prairie (Figure 1). The former comprises sub-parallel ridges up to 180 km in length and 50 km in width ranging in age from the pre- Cenozoic to the late Holocene (Zarate et al., 2005). The Ventania is sigmoid in shape and likely arised from block faulting (Martinez, 2001). The Puán and Pigüé mountains are located in its northeastern sector and border an arc- shaped valley flanking the central section of the range, across which lie the western Curamalal and eastern Bravard mountains (Martinez, 2001). In Argentina, Doffo and Bonorino (2005) produced a morphometric characterization of Las Lajas drainage basin in Cordoba using statistical analysis to look specifically at the control of structure and lithology over drainage network patterns. Morphometric data were analyzed using techniques like correlation, dispersion, factorial analysis, etc. and ultimately revealed no lithological control over the drainage network. Subbasin analysis indicated they belonged to two different groups, neither of which was found to be controlled by structure or substrate lithology. The Western sandy prairie consists exclusively of dunes arranged parallelly to the Encadenadas lakes. The area‟s highest elevation point occurs 130 meters above mean sea level (amsl). The concave dune slacks are filled by intermittent lakes. These landforms extend towards the north of the province, their crescent shapes clearly indicating an eolian origin. Their formation has been cyclic, alternating between periods of pedogenesis and erosion (Uriarte and Orioli, 1998). The Claromecó stream catchment was studied by Carbone and Piccolo (2002) whom described the drainage network hierarchy and characterized the basin based on edaphic and geomorphological parameters. Munguía and Campo de Ferreras (2003) studied geomorphological features of the Pescado Castigado creek basin associated with runoff. INTRODUCTION They also carried out a description of slope morphology, hypsometric curves and topographic profiles. ( ) The region‟s oldest geological units are 500 to 200 million years old (Harrington, 1947). It is characterized by linear structures of younger age which formed as a result of tectonic faulting that caused the uplifting of the Positivo Bonaerense (also known as Ventania system) and the downlifting of other areas such as those extending north of the Encadenadas depression (Uriarte and Orioli, 1998). These events were determinant in the pattern and distribution of the latter‟s drainage basins. Surface runoff is controlled by a five million year old deposit of eolian sediments covering the area whose top layer is as much as 100 m thick (Uriarte and Orioli, 1998). The layers at the base of this deposit were named Pampean (Pampeanos) by Ameghino (1880) while the top ones, Postpampean (Postpampeano) (Uriarte and Orioli, 1998). Knowledge of these sediments is important since they exert control over the present landscape and materials over which soils form. The region‟s oldest geological units are 500 to 200 million years old (Harrington, 1947). It is characterized by linear structures of younger age which formed as a result of tectonic faulting that caused the uplifting of the Positivo Bonaerense (also known as Ventania system) and the downlifting of other areas such as those extending north of the Encadenadas depression (Uriarte and Orioli, 1998). These events were determinant in the pattern and distribution of the latter‟s drainage basins. One study which stands out is Uriarte and Orioli‟s (1998) geo-environmental map of Guaminí county (Province of Buenos Aires). This study characterizes the potential and behaviour of water resources through pluviometric and hydrological budget analysis and also does a detailed study of groundwater resources. Navarro and González (1998) used a mathematical model running least squares and regression analysis to estimate the soil storage capacity for excess water based on a geomor- phological model. They also evaluated the area‟s geomorphological characteristics and quantified areas in- between contour lines. Lastly, calibration of their mathematical model confirmed realistic peakflow estimates. Surface runoff is controlled by a five million year old deposit of eolian sediments covering the area whose top layer is as much as 100 m thick (Uriarte and Orioli, 1998). The layers at the base of this deposit were named Pampean (Pampeanos) by Ameghino (1880) while the top ones, Postpampean (Postpampeano) (Uriarte and Orioli, 1998). INTRODUCTION Drainage basins and morphometric analysis have been the topic of numerous studies worldwide after interest in the subject picked up in the mid-twentieth century in the United States (Horton, 1945, Miller, 1953; Strahler, 1952, Schumm, 1956). These studies set groundwork for current research such as van der Linden and Woo‟s (2002) which modeled runoff and hypsometric curves in the Liard basin, Canada. Luo and Howard (2005) used a “circularity” function to quantify drainage basin morphology, infer erosion processes, characterize geomorphology and link them to climatic conditions. It is also used to derive drainage basin elongation and hypsometric curves at various elevation levels. The authors found that basins which are morphologically similar to their terrestrial counterparts are clustered around mainstreams at lower elevations, while cliffed basins whose origin likely is groundwater sapping develop around tributaries at higher elevations. This classification fits dry climate conditions Corresponding author. E-mail: ageraldi@criba.edu.ar. Tel: 054-0291-4861112. Corresponding author. E mail: ageraldi@criba.edu.ar. Tel: 054 0291 4861112. Authors agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License Authors agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License Geraldi et al. 13 punctuated by wet periods. piedmont area is smooth and dissected by a well-defined dendritic drainage network formed by the sloping terrain (De Francesco, 1971). pu ctuated by et pe ods Silva et al. (2006) analyzed and compared the morphometric attributes of some drainage basins located in Brazil. Their characterization is based on various parameters and indices such as area, perimeter, mainstream length, compactness index, drainage density, etc. They linked these parameters to the geological characteristics of the area and demonstrated that morphology is controlled by the main regional rock type. Basins overlying sedimentary bedrock and basins with a dissected relief exhibit lower current density and compactness index values than those with a smooth topography. In Argentina, Doffo and Bonorino (2005) produced a morphometric characterization of Las Lajas drainage basin in Cordoba using statistical analysis to look specifically at the control of structure and lithology over drainage network patterns. Morphometric data were analyzed using techniques like correlation, dispersion, factorial analysis, etc. and ultimately revealed no lithological control over the drainage network. Subbasin analysis indicated they belonged to two different groups, neither of which was found to be controlled by structure or substrate lithology. INTRODUCTION Knowledge of these sediments is important since they exert control over the present landscape and materials over which soils form. The area in which the lakes are located (Epecúen, Venado, Cochicó, Del Monte y Alsina), between the Norventánica plain and the Western sandy plain is refered to as a collector belt (Uriarte and Orioli, 1998). The development of this belt is a topic of scientific debate which has not been solved yet although it is believed that this break in topography formed from a regional fracture, a fault trending SW-NE which would exert structural control over the lakes‟ striking alignment. A dynamic fluvio-morphological regime is also evidenced by the occurence of alluvial cones in-between the lakes. The Province of Buenos Aires mostly comprises planation surfaces of low morphogenic energy. These overlie unconsolidated quaternary sediments that were reworked and redistributed by eolian - and to a lesser degree - hydrological processes. The apparent monotony of these plains is interrupted by mountain ranges such as the Tandil and Ventana (Figure 1). The latter are charaterized by rugged relief in the highest sectors where rock outcrops are overlaid by thin loess layers. Their The depression formed by the Encadenadas del Oeste J. Geogr. Reg. Plann. 14 Sandy plain Political limit Basin margin Mountain ranges Lagoons References Figure 1. Location of the study area. References Figure 1. Location of the study area. attempt will also be made to analyze behaviour of the drainage network in light of basin dynamics and identify those factors that control fluvial morphology. Moreover, since few regional studies have focused on this topic, the present study has a greater significance. drainage basin gives way southward to mountains where elevations reach a maximum of 1015 meters amsl. They decrease to a minimum of 85 m amsl northward in Epecuén lake, with most of the area lying between 150 and 300 m amsl. The basin can be defined as an endorheic fluvio- lacustrine system. The southern part of its catchment area is delineated by mountains whose streams supply Epecuén Lake. From West to East, the lakes and their mainstreams are Epecuén lake, Pigüé and Pull streams; Venado lake and Venado stream; Monte lake, Guaminí and Malleo Leufú streams; Alsina lake and Cura Malal, Pescado and Corto streams. METHODS Environmental characterization of a drainage basin first involves looking at its morphometric properties. These have a great impact on aquatic ecology given that many indicators, physico-chemical properties, sedimentation, etc. are controled by basin shape. Morphometric studies typically seek parameters which on one hand properly describe a geomorphological environment and on the other, allow for their statistical analysis thereby limiting the subjectivity of their conclusions (Doffo and Bonorino, 2005). This study relied on four different methods: cartographic analysis, field work, use of morphometric indices and statistical techniques. The physical setting of this drainage basin is a low sloping plain with steepness increasing in the mountain area and with surface runoff trending South-North. The most peculiar characteristic of this basin is its climate which alternates between very humid periods and intense droughts that occur in 8 to 12 year cycles. The first step involved quantifying the relief. Contour lines with a 2.5 m interval were digitized from the Military Geographic Institute‟s 1:50 000 and 1:100 000-scale topographic maps. They were then georeferenced and mosaiced into one set covering the entire basin. The drainage network was also digitized from topographic maps and updated with Landsat 5 TM satellite imagery provided by the Argentine National Commission for Spatial Activities (CONAE). Consequently and because it is an endorheic basin, characterization of its morphometry is of great importance for land-use planning. The aim of this study is to identify hydrographic sectors and subbasins within the Encadenada‟s drainage basin and analyze the former‟s morphometric properties including hypsometry. An Given that the area is highly impacted by anthropogenic activities, a systemic analysis (Capitanelli, 1998) was performed to Geraldi et al. 15 1. Corto stream subbasin (CAC) 2. Pescado stream subbasin (CP) 3. Cura Matal stream subbasin (CACM) 4. Guaminí stream subbasin (CAG) 5. Pigüé stream subbasin (CAP) 1. Corto stream subbasin (CAC) 2. Pescado stream subbasin (CP) 3. Cura Matal stream subbasin (CACM) 4. Guaminí stream subbasin (CAG) 5. Pigüé stream subbasin (CAP) 1. Corto stream subbasin (CAC) quantify regional space and embedded cultural elements such as channels, railways, sluice-gates, etc. Field trips were carried out to determine stream junctions and ground truth elevations as well as geographic coordinates with a GPS (global positioning system). Data thereby acquired were then used to delineate each subbasin by choosing points of highest elevation on either sides of the mainstream and its outlet. METHODS ArcGIS 9 and relevant extensions were used for this task and enabled instantaneous acquisition of parameters related to topography, basin length, etc. Once delineation of the subbasins was completed, morphometric analysis of their properties was carried out. The latter is of great importance if only to estimate the delay required for water to flow from a basin‟s edge to its outlet. The morphometric analysis allowed for quantification of variables and indices originally developped by Horton (1945), Strahler (1952) and Schumm (1956). 2. Pescado stream subbasin (CP) 4. Guaminí stream subbasin (CAG) 5. Pigüé stream subbasin (CAP) b. Intermediate subbasins: Located in the central part of the Encadenadas basin which is characterized by prairie (Fig. 2). They fill the areas in-between the larger subbasins and are catchments for smaller typically intermittent streams whose sources are found in gullies rather than the mountains: The drainage area (A) and perimeter (P) are two parameters which are useful for determining preliminary territorial dimensions of a drainage basin. These parameters are also useful for establishing general characteristics of the latter such as the climate zone it is part of, topography of the territory it encompasses, urbain and rural areas and others which can be infered from a global cartographical interpretation. 1. Pull stream subbasin 2. Venado stream subbasin (CAV) 3. Malleo Leufú subbasin (CAML) 4. Cochicó stream subbasin (CaC) c. Alluvial cones: Sedimentary formations of triangular shape whose wide bases are contiguous to the lakes and apexes intersect with the subbasins. Alluvial fans form at the outlet of streams due to the decreasing slope, slower currents and widening valleys where material from higher grounds typically tend to acrete (Strahler, 1952). They are flat and are formed by alluvial deposits. These units lie in-between the lakes and subbasins and are named after the latter: p In order to determine linear properties of the basin, a number of parameters were quantified including basin length (L), total length of streams (Lt), mainstream length (Lc), number of perennial and intermittent tributaries (NS) and number of streams of order 1 and 2 (N1 and N2). Area, fluvial streams, slopes, etc. were also analyzed in a GIS. Shape was determined using a variety of indices presented in Table 1. p With regards to basin shape, it is assumed that as values trend away from zero, peakflow levels and erosive action increase. If values trend towards zero, this potential is reversed. METHODS Basin shape was established on the basis of a morphological classification elaborated by López in 1988 (Fuentes Junco, 2004) where values ranging between 1.00 and 1.25 are deemed round; 1.25 to 1.50, oval; and 1.50 to 1.75, oblong. Hierarchical categorization was based on Strahler‟s version (1952) of Horton‟s system (1945) which Gardiner (1974) qualified as adequate for objective and hierarchical applications. Analysis of surfaces and visualisation of data in three dimensions was carried out using a DEM (digital elevation model) derived from contour lines data and plotted as a TIN (triangular interpolation network). A TIN is essentially a vectorial data structure that stores and displays a surface model. It partitions a topographic surface into a group of contiguous and separate triangles. Sampled points are converted into triangular vertices or nodes whose elevation values are interpolated into a continuous surface. Triangles are calculated based on the Delaunay triangulation model which states that a circle drawn around the node of any of these will contain that point and no other ones (Sione et al., 2004). Hypsometric curves were also plotted for each subbasin and finally, principal components analysis (PCA) was used to sort basins based on results from individually calculated parameters and indices. a . Epecuén Lake alluvial cone b .Epecuén - Venado alluvial cone c. Venado Del Monte alluvial cone d. Del Monte - Cochicó alluvial cone e. Cochicó - Alsina alluvial cone f. Cura Malal - Pescado alluvial cone g. Pescado - Corto alluvial cone a . Epecuén Lake alluvial cone b .Epecuén - Venado alluvial cone c. Venado Del Monte alluvial cone d. Del Monte - Cochicó alluvial cone e. Cochicó - Alsina alluvial cone f. Cura Malal - Pescado alluvial cone g. Pescado - Corto alluvial cone Sand dune subbasin: Located north of the lakes and populated by sand-dunes. No stream drains into it. Notable morphological units in the area include sand dunes, sand veneers, dune slacks and blow outs. Four litho-stratigraphic units and paleosoils have been described from the area (Dillon et al., 1985) as have compact alluvia at various depths of the Epecuén formation (Dillon et al., 1985). Fine to silty alluvial sands from the Hereford Formation are thought to date back to the Upper Pleistocene. These are overlain by the clayey to sandy and clayey to silty alluvial deposits of the Carlos Tejedor formation. METHODS These may be related to the extensive and typical fluvial deposits known as lujanenses (from Lujan) that are characteristic of the Late Pleistocene in the Province of Buenos Aires. RESULTS AND DISCUSSION The topographic analysis of the Encadenadas del Oeste basin resulted in the identification of each subbasin it is comprised of. This typology, the first for the area, will be a useful tool for future regional studies. The basins were classified as follows: The topmost layer of this sequence is made of an eolian deposit of fine sands known as Las Lilas formation (Holocene) (Isla et al., 2003). A drainage basin‟s size is of great importance when it comes to its hydrological behaviour. Typically, the larger it is in area, the more a. Main subbasins: Include those that flank the mountain range (Figure 2) and are bisected by mainstreams: 16 J. Geogr. Reg. Plann. J. Geogr. Reg. Plann. Table 1. Analyzed morphometry parameters, Encadenadas del Oeste lakes. Variable Parameter Equation Symbol Unit Form Compactness coefficient IK= 0.28* P IK A  IK A dimensional Circularity ratio 2 4 ci A R P   Rci A dimensional Shape factor  2 A Ff Lc Ff A dimensional Elongation ratio 1.128 e  A R L Re A dimensional Drainage System Drainage density Lt Dd A Dd km-1 Stream density Ns Dc A (a) Dc km-2 Average extent of drainage 4 A Es = Lt Es km Bifurcation ratio 1 2 N Rb N  (b) Rb A dimensional Concentration time       0.385 3 0.87* Lc Tc D (c) Tc min Relief Mainstream slope max min100 H H P L   (d) P % Mean basin elevation    hi Si H A (e) H m (amsl) Massivity coefficient H Cm A Cm m/km2 Sources: Lk: Gravelius (1914); Rc and Ff: Llamas (1993); Re: Schumm (1956); Dd: Gregory and Walling (1973); Dc, Es, Rb: Horton (1945); Tc: Junco (2004); P, H and Cm: Strahler (1968) a) Ns: Number of perennial and intermittent streams b) N1: First order channel; N2: Second order channel 2 c) D: Altitude difference in meters between the exit of the drainage basin and the moved away point more. d) Hmax: Maximum elevation; Hmin: Minimum elevation e) Hi: Length of the basin‟s contour lines and Si: Equidistance between this. RESULTS AND DISCUSSION Variables IK Re Rc Ff Es Rb Dd Dc TC CAG 1.93 0.36 0.26 0.1 0.96 2.6 0.25 0.03 19 CP 2.14 0.3 0.21 0.072 1.09 3 0.22 0.02 15 CACM 2.08 0.3 0.22 0.07 1.22 3.6 0.2 0.024 17 CAP 1.48 0.54 2.9 0.42 1.32 3.25 0.15 0.013 35 CAC 1.7 0.37 0.33 0.11 1.35 2.8 0.18 0.001 26 CAV 1.4 0.52 0.5 2.21 2.02 2 0.12 0.01 36 CAML 1.61 0.47 0.37 0.18 2.1 2 0.11 0.01 33 CaC 1.6 0.9 2.55 0.63 4.14 0.5 0.06 0.002 40 Table 3. Sub-basin morphometry index. elongation and low probability of peakflow events. Based on this index, it is unlikely that one of these subbasins would experience a rainstorm event over its entire area. This remains a general assumption though as the area affected by a storm depends on the latter‟s size, extent, duration and intensity, as well as characteristics of the event. The Pescado, Corto and Cura Malal subbasins display the lowest values by being the most elongated and thus the least likely to have their entire area affected by a single storm. Malleo Leufú and Corto‟s subbasins can be designated as oblong, the Venado‟s oval and the rest, elongated as they exceed the maximum value (Table 2). y ( ) The elongation ratio index (Re) correlates best with basin hydrology (Lopez Bermudez and Romero Diaz, 1988). Values lower than 1 correspond to elongated basins and the smaller they are, the more elongated their shape is. Re emphasizes the relationship between area and mainstream. All subbasins within the study area were found to be elongated as their Re values are all less than 1. Differences can still be noted between them, however. The Guamini, Pescado, Cura Malal and Corto basin display the lowest values which range between 0.3 and 0.37. These subbasins are located in the eastern part of the drainage basin and are also the widest in area. It should also be noted that their channel banks are free of cliffs and their courses are meandering. These values corroborate Senciales‟ (1999) findings that lower index values are typical of areas where meandering is well developped and slope gradients low. The Cochicó basin sets itself apart by displaying a value of 0.9 which is evidence of an almost circular shape. RESULTS AND DISCUSSION Sources: Lk: Gravelius (1914); Rc and Ff: Llamas (1993); Re: Schumm (1956); Dd: Gregory and Walling (1973); Dc, Es, Rb: Horton (1945); Tc: Junco (2004); P, H and Cm: Strahler (1968) a) Ns: Number of perennial and intermittent streams b) N1: First order channel; N2: Second order channel 2 c) D: Altitude difference in meters between the exit of the drainage basin and the moved away point more. d) Hmax: Maximum elevation; Hmin: Minimum elevation e) Hi: Length of the basin‟s contour lines and Si: Equidistance between this. for which indices of compactness (IK), elongation ratio (Re), circularity ratio (Rc) and shape factor (Ff) were subsequently calculated. The compactness index varies between zero and one. The closer a value is to one, the more compact is the basin. Within the subbasin set, the Pigüe and Venado are more compact than the CAG, CP and CACM subbasins which display the lowest lK. The resulting value was compared with a morphology-based classification of basins designed by López (Fuentes Junco, 2004). Based on that scheme, the Cochicó, rainwater it collects and the greater is its peakflow level. Nonetheless, there is a delay between rainstorm events and timing of the peakflow and associated discharge. Of the subbasin set, the Corto is the widest in area, followed by the Pigüe (Table 2). Their mainstreams, along with that of the Cura Malal are also the system‟s longest ones. The basin analysis was carried out using various morphometric parameters whose results are presented in Table 3. Shape is a complex morphometric concept. In this study, basin shape was compared with ideal shapes Geraldi et al. 17 Table 2. Sub-basin area (km2), mainstream length (Lc) and total channel length (Lt). Table 2. Sub-basin area (km2), mainstream length (Lc) and total channel length (Lt). Sub-basin Area(km2) Lc (km) Lt (km) Shape Malleo Leufú 270.00 16.60 32.06 Oblong Venado 293.03 23.04 23.00 Oval Cochicó 333.97 20.14 20.14 Oblong Guaminí 605.12 50.70 156.21 Elongated Pescado 791.91 70.22 180.40 Elongated Cura Malal 829.63 106.38 168.77 Elongated Pigué 998.56 80.18 80.18 Elongated Corto 2893.62 160.37 533.00 Oblong Table 3. Sub-basin morphometry index. RESULTS AND DISCUSSION Values for the Arroyo Venado and Pigüe basins lie in the medium range which might be related to a rounder shape upstream and elongated in the middle and lower sections. y g The circularity ratio (Rc) is defined as the relationship between basin area and that of a circle of equal perimeter. The lowest values are found in the eastern section of the basin and belong to Pescado, Corto, Cura Malal and Guaminí subbasins. These units display the lowest circularity, a morphometric characteristic which is unlikely to intensify peakflow events. It can also be inferred that torrentiality is greater within the tributaries rather than in the main channel at their points of confluence. As for mainstream length, the longest is the Corto (Table 2) whose tributary network is also the longest (533 km). The shortest mainstreams are Malleo Leufú (16.6 km) and Cochicó (20.14 km). The shortest network is the Cochicó„s to which no tributaries are connected. The shape factor (Ff) is very low, ranging between 0.07 and 0.6 for all cases except the Arroyo Venado basin (Table 3). This implies an overall trend towards A drainage network is a hierarchized system of streams that takes into account all orders and junctions of tributaries connected to a mainstream. It plays an J. Geogr. Reg. Plann. 18 the Encadenadas are long in the rounder basins (Pigüé, Venado and Cochicó) with values of 35, 36 and 40 min respectively. important role in the transport of materials and energy. The subbasins‟ drainage network varies from third to fourth order which indicates a low level hierarchy. The basin with poorest drainage is the Cochicó, of first order. Those of highest order (fourth) are the Guaminí and Corto basins. The Pigüé, Malleo Leufú, Pescado and Cura Malal basins are of third order and the Venado of second order. The network pattern or ramification shape is dendritic to sub-dendritic, however parallel to sub-parallel is also developed locally. In the widest basins, this pattern tends to occur above 600 m amsl. p y Basin topography was analyzed using the digital elevation model (Fig. 3). Elevations within the study area are comprised between 82 m amsl (lowest elevation) of the western Epecuén lake and 1015 m amsl (highest elevation) where the mainstreams that drain the Ventania‟s main basins take their source. RESULTS AND DISCUSSION The difference in altitude between these two extremes characterizes a relief whose elevations and slopes are likely to be significant and thus favour the development and activation of erosive processes and transport of materials. It should also be mentioned that the highest elevations lie southward of the basin and the lowest ones, northward. The remaining landscape lies below 400 m amsl. The mean slope of the mainstream is an important parameter for determining hydrological behaviour in a basin. Basins with a steeper slope are characterized by faster runoff and respond more rapidly to rainfalls, therby increasing discharge at any given point (Horton, 1945). In basins of greater elongation and lower gradients, water circulates more slowly along longer channels in-between sources and outlets. p Drainage density (Table 3) is defined as the relationship between stream length and total area. This parameter is used to describe basin regime and morphology (Sanchez, 1991). In general, magnitude is indirectly related to soil cover, erosion, lithology and infiltration. Areas of permeable rocks typically display low drainage density. The study area‟s basins posess a very low drainage density, including some extreme cases like the Cochicó which exhibits a value of 0.06 km/km2. This may be caused in part, first, by the coarse texture of its silty and sandy alluvia (INTA, 1994) which are rich in organic materials and are characterized by large interstitial pores, thereby increasing water infiltration. Sands are also permeable enough that input to nearby streams is minimized. Land-use being mostly agricultural, the vegetation cover also promotes water infiltration and reduces runoff. Once a rainfall is over, roots absorb soil humidity and increase the capacity for infiltration in- between rain events. The low drainage density indicates that basin response to a rainstorm is likely to be slower for evacuating excess water than basins with a higher density. However, when water reaches streams of low ramification and hierarchization, this may cause a rapid rise in waters and waves of peakflow events all the way to the stream outlet. Due to the higher declivity between stream sources and outlets in the western part of the basin, cliffs have developped along the channel banks of streams such as Pigüé et al. (Figure 3) whose mean slopes are steepest (Table 4). The Cura Malal, Pescado and Corto streams exhibit slopes below 0.18, 0.02 and 0.01 respectively. RESULTS AND DISCUSSION Consequently, streams located to the east of the basin respond faster to rainfalls than those to the west. The headwaters sector is influenced by the presence of mountains which cause mean slope values to reverse. The Corto, Pescado and Cura Malal streams exhibit the steepest mean slopes, the highest levels of ramification and the most cliffs. This is explained by the range‟s curved shape (Figures 2 and 3) whose elevation decreases towards the northwest where the sources of Pigüé and Guaminí streams lie. In terms of bifurcation ratio, subbasins displaying the lowest values are Venado, Cochicó and Malleo Leufú (Table 3) while the remaining are all higher. With regards to the Encadenadas, the rounder the drainage basin, the higher the bifurcation ratio and the more elongated the basin, the lower it is. This can be explained the following way: elongated basins tend to delay the concentration of water due to the longer length of their mainstream. However, when that threshold is reached, waters may start to rise suddenly and rapidly (Gonzáles Matauco, 2004). Thus, the fact that elongated basins display the lowest bifurcation ratio results from the length of their mainstreams which delays the concentration of waters, further enhanced by low gradients (1%) in the basins‟ middle and lower sections (Senciales, 1999). Altitudinal difference is the difference between the highest and the lowest elevation point of a basin and influences climatic and ecological variability. A basin with a greater number of altitudinal surfaces is likely to harbour more ecosystems as a result of important precipitation and temperature variations (Junto, 2004). Table 4 sums up altitudinal differences including and excluding the nearby mountain topography. Based on this data, the Cura Malal, Pescado and Corto subbasins exhibit the greatest climatic and ecological variations. The massivity index (Cm) indicates that for each square kilometre within the basin, there is a mean increment in elevation of 0.35 m, a typical value for basins whose relief are for the most part flat. This parameter is thus an indicator of increments in slope as elevation increases. The Cm index may also reflect Concentration time is defined as the time required for a water particle to travel from the furthest point to a specific point. It corresponds to the time elapsed between the end of a rainfall event and the time when surficial runoff ceases (Fuentes Junco, 2004). Concentration times in Geraldi et al. RESULTS AND DISCUSSION 19 p g g g g Subbasin Hmax cs Hmaxss Hmin Dss Dcs Pss Pcs Cm H Malleo Leufu 148 148 105 43 43 0.25 0.25 0.56 153 Venado 163.94 163.94 105.72 58.22 58.22 0.25 0.25 0.53 158 Cochicó 147.45 147.45 107.04 40.41 40.41 0.2 0.2 0.46 155 Guamini 376.38 376.38 105 271.38 271.38 0.38 0.53 0.38 235 Pescado 600.51 300 110 190 490.51 0.02 0.69 0.28 222 Cura Malal 824.84 300 108.61 191.39 716.23 0.01 0.67 0.31 259 Pigue 375 300 100 200 275 0.24 0.34 0.24 245 Corto 853 300 110 190 743 0.01 0.46 0.1 317 Subbasin Hmax cs Hmaxss Hmin Dss Dcs Pss Pcs Cm H Malleo Leufu 148 148 105 43 43 0.25 0.25 0.56 153 Venado 163.94 163.94 105.72 58.22 58.22 0.25 0.25 0.53 158 Cochicó 147.45 147.45 107.04 40.41 40.41 0.2 0.2 0.46 155 Guamini 376.38 376.38 105 271.38 271.38 0.38 0.53 0.38 235 Pescado 600.51 300 110 190 490.51 0.02 0.69 0.28 222 Cura Malal 824.84 300 108.61 191.39 716.23 0.01 0.67 0.31 259 Pigue 375 300 100 200 275 0.24 0.34 0.24 245 Corto 853 300 110 190 743 0.01 0.46 0.1 317 Figure 2. Encadenadas del Oeste lakes: map of sub-basins. Figure 2. Encadenadas del Oeste lakes: map of sub-basins. Figure 2. Encadenadas del Oeste lakes: map of sub-basins. more important and render the implementation of appropriate agricultural practices essential. It can be noted that the basin‟s hypsometric curve, excluding the mountain range, comprises a gentle slope below 300 m amsl. The altimetric frequency histogram reveals that only 5% of the basin lies above 350 meters which corroborates the DEM analysis. As part of the targeted morphometric analysis, the following presents results from each subbasin‟s hypsometric curve and histogram analysis. erosion prone basins, the most vulnerable ones being the smaller ones who display higher coefficients versus the larger ones whose values tend to be smaller. According to Llamas (1993), the hypsometric curve is the most precise rendition of a basin‟s elevation. Hypsometry analysis (Figures 4 and 5) demonstrates that the highest section is that of steepest slope and also the most restricted in area. During wet periods, water precipitations in the mountains are thus likely to result in intense runoffs above 400 m amsl. Erosion processes in that area are J. Geogr. Reg. Plann. RESULTS AND DISCUSSION It does not fit well any of Strahler‟s models (1974) though as it is located between two active subbasins both in evolutionary states and has two alluvial cones resting on its margins, all of which suggest a subbasin in a state of disequilibrium evolving towards “youth”. This illustrates the prevalence of low lying areas in the basin over higher ones. Its hypsometric curve is overall convex which suggests an important potential for erosion and according to Strahler‟s (1974) model corresponds to the disequilibrium or “youth” stage indicating the area is primarily a source of sediment and water. This illustrates the prevalence of low lying areas in the basin over higher ones. Its hypsometric curve is overall convex which suggests an important potential for erosion and according to Strahler‟s (1974) model corresponds to the disequilibrium or “youth” stage indicating the area is primarily a source of sediment and water. The Cochicó stream which drains the equally named basin is the second shortest stream of the Encadenadas, being 20 km in length from source to outlet. Its channel is fairly straight for that of a plain stream. It‟s discharge decreases during droughts to the point where it may completely dry out, defining it as intermittent. The basin is 333 km2 in area, 1% in slope and has a mean elevation of 155 m amsl. The frequency histogram (Figure 6e) shows that elevations ranging from 95 to 150 m are prevalent and that low lying areas are more extensive than those at higher altitudes (200 - 250). Its convex shaped curve displays a negative asymmetry (Figure 7) whose mean equals to 175 and its median 155 m amsl. It has reached the youth phase of the erosion cycle indicating a high morphodynamic potential for erosion processes to take place. The Guaminí sub-basin exhibits a regular slope which is evidenced by its fairly straight curve (Figure 7c). Within the erosion cycle, it has reached the equilibrium state or “maturity” which is characterised by sediment and water transport. Its altimetric frequency histogram (Figure 6c) denotes the predominance of elevations between 200 and 250 m amsl which encompass 67% of its total area with another 3% located above 400 m. The stream that drains the Malleo Leufú basin is of intermittent character and its channel the shortest in length (16 km). RESULTS AND DISCUSSION 20 ee ee Figure 3 Digital elevation model of the Encadenadas del Oeste lakes ee Figure 3. Digital elevation model of the Encadenadas del Oeste lakes. 0.000 5.000 10.000 15.000 20.000 25.000 30.000 35.000 40.000 45.000 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 % Area Área km2 Figure 4. Altimetric histogram of the Encadenadas del Oeste lakes basin. 0.000 5.000 10.000 15.000 20.000 25.000 30.000 35.000 40.000 45.000 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 % Area Área km2 Figure 4. Altimetric histogram of the Encadenadas del Oeste lakes basin. Figure 4. Altimetric histogram of the Encadenadas del Oeste lakes basin. Figure 4. Altimetric histogram of the Encadenadas del Oeste lakes basin. he Encadenadas del Oeste reached an intermediate phase, between relative equilibrium or “maturity”, and “old age”, evolving towards the latter. At that stage, its morphodynamic potential for undergoing erosion and mass slumping of its slopes is low. Figure 4: Altimetric histogram of lakes basin. The Pigüé Basin mostly comprises (78%) of elevations ranging between 200 and 300 m. The highest lands make up only 1% of the basin area and are where the Pigüé stream source lies (Figure 6a). Its topography is smooth which corroborates well with its almost straight hypsometric curve (Figure 7a). This is also evidenced by the agreement between the mean elevation and the elevation frequency mean. The Venado sub-basin covers 293 km2, has a perimeter of 85 km and a mean slope of 2%. Of the lesser subbasins, this one has the greatest mean elevation (Table 5) which is worth considering since it has an influence on runoff and erosion processes. Predominant elevations range between 150 to 200 metres followed by Figure 8 illustrates the typical phases of basin erosion cycle and potential evolutionary scenarios as put forth by Strahler (1974). Based on his model, the Pigüé basin has 21 Geraldi et al. 0 200 400 600 800 1000 1200 1400 0.00 10.00 20.00 30.00 40.00 50.00 Altitud en m/s/n/m Area km2 Area km2 Figure 5. Hypsometric curve (km2) of the Encadenadas del Oeste lakes basin. Figure 5. Hypsometric curve (km2) of the Encadenadas del Oeste lakes basin. 95 to 150 metres amsl (Figure 6b). Its hypsometric curve (Figure 7b) increases as it reaches the basin‟s predominant elevation values. RESULTS AND DISCUSSION This is a typical plain stream, winding, its course interspersed with meanders where it migrated to connect to the outlet. Its banks are free of cliffs and allow this shallow channel to easily overflow its banks in response to increases in discharge. Its basin is 270 km2 in area with a perimeter of 94 km and a slope of 2%. The altimetric frequency histogram (Figure 6d) indicates that 52% of elevations lie between 95 and 150 m. Its hypsometric curve is asymetric (Figure 7d) given that the mean elevation is lower than the 175 m amsl median. The 106 km long Cura Malal stream runs through a diverse landscape of mountains and plains and drains into Alsina Lake. Its middle section which lies over a plain is characterized by meandering. This basin, 829 km2 in area, is comprised of a diverse range of elevations with an almost symmetrical frequency as shown in Figure 7f. J. Geogr. Reg. Plann. 22 0 50 100 150 200 250 300 350 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 Área km2 f 0 20 40 60 80 100 120 140 160 95-150 150-200 200-250 Área km 2 0 20 40 60 80 100 120 140 160 95-150 150-200 200-250 Área km2 b 0 50 100 150 200 250 300 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 Área km2 g 0 100 200 300 400 500 600 700 800 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 Área km2 h 0 50 100 150 200 250 300 95-100 100-200 200-250 250-300 300-350 350-400 400-1015 Área km2 c 0 20 40 60 80 100 120 140 160 180 200 95-150 150-200 200-250 Área km2 e d a Area km2 gure 6. RESULTS AND DISCUSSION Altimetric histogram of sub-basins a- Pigüé b-Venado c- Guaminí d- Malleo Leufú e- Cochicó f- Cura Malal g- Pescado h- Corto a Area km2 0 20 40 60 80 100 120 140 160 95-150 150-200 200-250 Área km 2 b 0 50 100 150 200 250 300 350 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 Área km2 c 0 20 40 60 80 100 120 140 160 95-150 150-200 200-250 Área km2 d d f 0 50 100 150 200 250 300 95-100 100-200 200-250 250-300 300-350 350-400 400-1015 Área km2 0 20 40 60 80 100 120 140 160 180 200 95-150 150-200 200-250 Área km2 e f 0 50 100 150 200 250 300 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 Área km2 g 0 100 200 300 400 500 600 700 800 95-150 150-200 200-250 250-300 300-350 350-400 400-1015 Área km2 h g h Figure 6. Altimetric histogram of sub-basins a- Pigüé b-Venado c- Guaminí d- Malleo Leufú e- Cochicó f- Cura Malal g- Pescado h- Corto igure 6. Altimetric histogram of sub-basins a- Pigüé b-Venado c- Guaminí d- Malleo Leufú e- Cochicó f- Cura Ma 23 Geraldi et al. 0 200 400 600 800 1000 1200 0 200 400 600 Altura m/s/n/m Area km 2 C. Pigue 0 50 100 150 200 250 300 100 120 140 160 Altura m/s/n/m Area km 2 Venado 0 50 100 150 200 250 300 0 50 100 150 Altura m/s/n/m Area km2 Mallao Leufu 0 100 200 300 400 500 0 100 200 300 400 500 Altura m/s/n/m Area km 2 Guaminí 0 50 100 150 200 250 300 0 50 100 150 200 Altura m/s/n/m Area. km2 Cochicó 0 200 400 600 800 1000 1200 0 200 400 600 Altura m/s/n/m Area km 2 Cura Malal a b 0 200 400 600 800 1000 1200 0 200 400 600 800 Altura m/s/n/m Area km 2 Pescado 0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200 Altura m/s/n/m Area km 2 Corto g c e d f h Figure 7a-h. Hypsometric curves of subbasins a)Pigüé b)Venado c)Guaminí d)Malleo Leufú e)Cochicó f)Cura Malal g)Pescado h)Corto. 0 50 100 150 200 250 300 100 120 140 160 Altura m/s/n/m Area km 2 Venado b 0 200 400 600 800 1000 1200 0 200 400 600 Altura m/s/n/m Area km 2 C. RESULTS AND DISCUSSION Pigue a Venado Venado Altura m/s/n/m 0 100 200 300 400 500 0 100 200 300 400 500 Altura m/s/n/m Area km 2 Guaminí d 0 50 100 150 200 250 300 0 50 100 150 Altura m/s/n/m Area km2 Mallao Leufu c d c c 0 200 400 600 800 1000 1200 0 200 400 600 Altura m/s/n/m Area km 2 Cura Malal f 0 50 100 150 200 250 300 0 50 100 150 200 Altura m/s/n/m Area. km2 Cochicó e Cura Malal f e 0 200 400 600 800 1000 1200 0 200 400 600 800 1000 1200 Altura m/s/n/m Area km 2 Corto h 0 200 400 600 800 1000 1200 0 200 400 600 800 Altura m/s/n/m Area km 2 Pescado g Figure 7a-h. Hypsometric curves of subbasins a)Pigüé b)Venado c)Guaminí d)Malleo Leufú e)Cochicó f)Cura Malal g)Pescado h)Corto. J. Geogr. Reg. Plann. 24 Figure 8. Phases of the erosion cycle based on basin hypsometric curves (adapted from Strahler, 1974). Figure 8. Phases of the erosion cycle based on basin hypsometric curves (adapted from Strahler, 1974). Figure 8. Phases of the erosion cycle based on basin hypsometric curves (adapted from Strahler, 1974). Table 5. Correlation matrix between the variables and the first three components. from 150 m to 250 m amsl prevail. Mean elevation in the basin is 222 m. Its properties are similar to the Cura Malal‟s nevertheless, its hypsometric curve (Figure 7g) is slightly more pronounced thereby reflecting the steeper terrain. The basin has reached the “maturity” stage characterized by sediment deposition (Strahler, 1974). 2 IK - 0.79064 - 0.17754 - 0.40242 Re 0.92922 - 0.10657 - 0.29317 RC 0.72409 0.07818 - 0.31155 Ff 0.53774 - 0.31434 0.71700 Es 0.84875 - 0.14464 - 0.36142 Dd - 0.94937 0.02185 0.05111 Dc - 0.74746 - 0.56683 - 0.04081 Tc 0.97334 0.04702 0.14118 Área - 0.27066 0.94159 0.06785 y p ( , ) The Corto Basin extends over 2 893 km2 making it the Encadenada‟s widest subbasin. Its mainstream, 160 km in length, is also the longest and its relief, the system‟s highest as it has the widest surface covered by mountains even though it amounts to only 8% of the total. This is the Encadenada‟s easternmost subbasin displaying a difference of 72 m in elevation with the Pigüé subbasin (westernmost). The altimetric frequencies histogram (Figure 6 h) shows a uniform distribution and variation in elevation. RESULTS AND DISCUSSION Its hypsometric curve (Figure 7 h) is much steeper than that of the smaller basins (Venado, Malleo Leufú and Cochicó) and displays a positive asymmetry due to its mean elevation, lower than the 275 m median. This basin is in its maturity phase during which sediment deposition prevails (Strahler, 1974). Areas lying above 400 m make-up only 5% of the total. The altimetric frequency histogram (Figure 6 f) highlights the basin‟s symmetry and shows most area falling in the 200 to 300 m range. It has reached an equilibrium state, evolving towards maturity. 2 The Pescado Basin covers a 791 km2 wide area. Its histogram (Figure 6 g) shows a less symmetrical distribution than the previous one. Higher grounds only represent 1% of the total area where altitudes ranging A principal components analysis (PCA) was carried out on the basins‟ calculated morphometric indices so as to order them. The graphic in Figure 9 corresponds to a 25 Geraldi et al. Table 6. Total variance accumulated in the first six components. % Total Associated variance with c/Component 0.61145 0.76470 0.87787 0.96360 0.99021 0.99814 Figure 9. PCA as calculated from matrix correlation. Both graphics correspond to the defined plane by the two first main components, with 87% of the total variety. Figure 9. PCA as calculated from matrix correlation. Both graphics correspond to the defined plane by the two first main components, with 87% of the total variety. Encadenadas del Oeste‟s basin. The characterization of these units shows that the basin is morphologically diverse due to the dynamic fluvial activity that prevails within its limits. One of the above mentioned morphological units are the alluvial cones which form at the mouths of the mainstreams and delineate the bases of the different subbasins. Units designated “lesser” subbasins by this study are in a phase of disequilibrium or “youth” and therefore are actively being eroded. These basins are smaller in terms of area and their mainstreams shorter. Their hypsometric curves are negative and asymetric due to the mean elevation being lower than its median. The subbasins whose sources are located in the mountains have reached a state of equilibrium refered to as “maturity” or “old age” by Strahler, indicating that transport of water and sediment is the predominant process. These subbasins also display a positive asimetric curve. RESULTS AND DISCUSSION Corto and Pescado set themselves apart by being the only two basins to have reached the “old age” phase. projection of the initial variables on a two-dimensional plane delineated by two axes. Axis one correlates with shape indices and axis two, with drainage and area. The correlation matrix (Table 6) shows that the first three components explain 87 % of the variance. Pigüé, Venado, Malleo Leufú and Cochicó subbasins are grouped together as they all display a strong positive correlation with the first principal component. Similarity between them is due to their rounder shape and longer concentration times. projection of the initial variables on a two-dimensional plane delineated by two axes. Axis one correlates with shape indices and axis two, with drainage and area. The correlation matrix (Table 6) shows that the first three components explain 87 % of the variance. Pigüé, Venado, Malleo Leufú and Cochicó subbasins are grouped together as they all display a strong positive correlation with the first principal component. Similarity between them is due to their rounder shape and longer concentration times. As can be observed, the second principal component correlates positively with area, drainage density and concentration times. The group formed by Pescado, Cura Malal and Guaminí is characterized by larger areas, greater drainage densities and shorter concentration times. The Corto basin sets itself apart by exhibiting a strong positive correlation with the second principal component. This is to be expected as the Corto basin is the widest in terms of area and its mainstream, the longest. With regards to the shape index, the Venado subbasin is characterized by an oval outline, the Cochicó and Malleo Leufú an oblong one, while the rest are significantly elongated. The Pigüé and Venado basins are the most compact and circular indicating that rises in water are also likely to result in greater discharges at REFERENCES their mouths. Their steeper topography favours the rapid concentration of waters which is corroborated by the occurrence of cliffed banks along their mainstreams‟ channels. Capitanelli R (1998). Geografía Física y Medio Ambiente. Vol. 1-Ed. Mendoza. Ecogeo. Carbone ME, Piccolo MC (2002). Morfometría de la cuenca del arroyo Claromecó. Provincia de Buenos Aires, Argentina. Revista Geofísica. Instituto Panamericano de Geografía e Historia 56:51-66 The drainage network exhibits overall a low level of ramification and hierarchy which is likely due to the sedimentary nature and high permeability of the soil below. The vegetation cover (crops are cultivated year- round) may exacerbate the restricted ramification which could provoke a rapid swelling of its waters downriver. De Francesco FD (1971). Geología superficial del área pedemontana del flanco Sudoccidental de la Sierra de la Ventana. Prov. Buenos Aires. Com. Invest. Cient., Reunión Científica. La Plata. Dillon AA, Hurtado MA, Jiménez JE, Castillo RJ (1985). Consideraciones geomorfológicas y estratigráficas como base del carteo de suelos en un sector de la Pampa Arenosa (Pcia. De Buenos Aires). 1as. Jornadas Geológicas Bonaerenses, Actas, Tandil, CIC, pp. 737-749. Concentration times are short in all of the basins. The Cochicó and Venado exhibit the highest values at 40 and 36 min respectively. These basins‟ roundness considerably lengthens the time needed for water to flow from their edges to their mainstream and may end up concentrating the discharge downriver. The Corto, Pescado and Cura Malal display the shortest concentration times given their significant elongation which has water flowing through a shorter area before reaching the subbasins‟ mainstreams. pp Doffo N, Bonorino GG (2005). Caracterización morfométrica de la cuenca alta del arroyo Las Lajas, Córdoba: Un análisis estadístico. Revista de la Asociación Geológica Argentina 60(1):16-22 g g ( ) Fidalgo F, De Francesco F and Pascual R (1975). Geología superficial de la llanura Bonaerense. En: VI Congreso Geológico Argentino. g g g Fuentes JJJA (2004). Análisis Morfométrico De Cuencas: Caso De Estudio Del Parque Nacional Pico De Tancítaro. Dirección General de Investigación de Ordenamiento Ecológico y Conservación de Ecosistemas Instituto Nacional De Ecología. P. 47. Gardiner V (1974). “Drainage basin morphometry”. Bristish Geomorphological Research Group. Technical Bulletin Nº 14. University of East Anglia. England 44 p. g Principal components analysis ordered the subbasins into two groups. The first one includes Pigüé, Venado and Malleo Leufú subbasins. Rounder rather than elongated, they have longer concentration times. REFERENCES This means that there may be some delay for stormwaters to reach their mainstreams but that once they are channelized, their discharge may swell considerably downriver, more so because of these basins‟ steeper topography. This fact is extremely important as Carhué, a resort town, is located nearby. The second group includes Pescado, Cura Malal and Guaminí basins to which the Corto basin can be added as only its larger area sets it apart. These basins‟ shape is elongated, their streams long and straight thus lengthening the distance from source to outlet. Low declivity and the absence of cliffs allow these streams to easily overflow their banks during rainstorm events. Rapid swelling of stream waters is uncommon in these subbasins and easily forecasted. González de MAI (2004). Análisis Morfométrico De La Cuenca y de la Red De Drenaje del Río Zadorra y sus Afluentes Aplicado a la Peligrosidad de Crecidas. Dpto. Geografía, Prehistoria y Arqueología Universidad del País Vasco. Boletín de la A.G.E. 38:311-329 Hernández E (2002). La cuenca hidrográfica. FCFA. Universidad de los Andes. Mérida – Venezuela. Horton R (1945). Erosional development of streams and their drainage basins: Hydrophysical approach to quantitative morphology. Geolog. Amer. Bull. 56:275-370. Instituto Nacional de Tecnología Agropecuaria (INTA) Secretaría de Agricultura, Ganadería y Pesca. (1994). Atlas del Suelo de la República Argentina, Tomo I Buenos Aires 525 p. p g p Isla FI, Ruiz Barlett E, Marquez J, Urrutia A (2003). Efectos Enso En La Transición Entre El Espinal Y La Pradera Cultivada En La Diagonal Isla FI, Ruiz Barlett E, Marquez J, Urrutia A (2003). Efectos Enso En La Transición Entre El Espinal Y La Pradera Cultivada En La Diagonal Sudamericana, Argentina Central. Rev. C&G 17(1-2):63-74 Lexow C (2002). Hidrodinámica de la zona no saturada aplicada al estudio del Balance Hidrológico y de contaminantes de los recursos hídricos subterráneos (Cuenca del Arroyo del Aguila). Tesis Doctoral en Geología. Universidad Nacional del Sur. Bahía Blanca. Argentina 175 p. Linden SVD and Woo MK (2002). Transferability of hydrological model parameters between basins in data-sparse areas, subarctic Canada- Elsevier - J. Hydrol. 270 (2003):182-194. Two facts need to be taken into account as part of any land-use plan or major intervention in the area. Not only do both mainstreams from the Malleo Leufú and Guaminí basins drain into the same lake, but they also do so at a short distance from one another. REFERENCES In the event of a rainstorm occuring over both subbasins, fairly similar concentration times (19 and 33 min) will have the increased discharge of waters from both subbasins reaching the lake at a short distance from the town of Guaminí. The same could occur at the mouths of Cura Malal and Pescado streams which also both drain into Alsina Lake. Although no towns are located nearby, a number of farms are. y ( ) López Bermúdez F, Romero Díaz MA (1988). Relación entre escorrentías superficiales y características físicas y ambientales (Alto Guadalquivir). En Avenidas fluviales e inundaciones cuenca del Mediterráneo. Alicante, Instituto Universitario de Geografía de la Universidad de Alicante. Caja de Ahorros del Mediterráneo, pp. 171- 183. Luo W, Howard A (2005). Morphometric analysis of Martian Valley Network basins using a circularity function. J. Geophysical Res. 110: E12S13. Llamas J (1993). Hidrología general. Universidad del Estado de México. Toluca. México 627 p. p Martinez JS (2001). The geology of Ventania (Buenos Aires Province, Argentina). J. liberian Geol. 27:43-69. Miller VC (1953). A quantitative geomorphic study of drainage basin characteristics in the Clinch Mountain area, Virginia and Tennessee. Office of Naval Research, Geography Branch, Project NR. Technical Report, 3, Columbia University. pp. 389-042 Report, 3, Columbia University. pp. 389-042 Conclusion This study‟s aim was to define for the first time the various drainage subbasins that comprise the J. Geogr. Reg. Plann. 26 REFERENCES Conflict of Interests Munguía S, Campo de Ferreras AM (2003). Características Hidro- Geomorfológicas De La Cuenca Del Arroyo Pescado Castigado. Buenos Aires. Argentina. Papeles de Geografía 38:137-150. The authors have not declared any conflict of interests. Geraldi et al. 27 Strahler A (1974). Geografía física. Ediciones Omega, S.A. Barcelona, España. 765 p. Navarro E, González UM (1998). Geomorfometria aplicada a la gestión hídrica del ecosistema de las lagunas Las Encadenadas del Oeste. Revista del Instituto de Geología y Minería 12(1):35-49 Strahler A (1985). Geografía Física (7a.ed.). Editorial Omega, S. A. Barcelona – España. 567 p. g y ( ) Sánchez T (1991). Estudio morfoclimático del CabeVó D´OR. Universidad de Alicante. España. 69 p. Uriarte González M and Orioli G (1998). Carta Geoambiental del Partido de Guaminí. Editorial Universidad Nacional del Sur. Bahía Blanca. Argentina. p p Schumm SA (1956). Evolution of Drainage Systems and Slopes in Badlands at Perth Ambos, New Jersey. Bull. Geol. Soc. Am. 67:646. Senciales JM (1999). Redes fluviales. Metodología de Análisis. Estudios y Ensayos, 34. Universidad de Málaga, 337 p. Yrigoyen MR (1975). Geología del subsuelo y plataforma continental. Relatorio 6º Congreso Geológico Argentino: 139-168. Bahía Blanca. Silva A, Herpin U, Martinelli L (2006). Morfhometric Characteristics of seven messo-scale river basins in state of Sao Paulo (Southestern Brazil). Caminhos de Geografia. 3(17):20-30. Zarate M, Rabassa J, Partridge T, Maud R (1995). La brecha Cerro Colorado (Abra de la Ventana, Buenos Aires): ¿Es miocena? Actas de las Cuartas Jornadas Geológicas y Geofísicas Bonaerenses, Junin, 1:159-167. Sione W, Cuello A, Angelini M (2004). Análisis Espacial, Técnicas cuantitativas aplicadas. Curso 9. Especialización en Teledetección y Sig. Universidad Nacional de Lujan. Strahler A (1952). Hypsometric (area-altitud) analysis of erosional relief. Bull. Geol. Soc. Am. 63:1117-1142. Strahler A (1968). Quantitative Geomorphology. En: R. Fairbridge (ed.). The Encyclopedia of Geomorphology. Encyclopedia of Earth Sciences Series, T. 3. Dowden, Hutchinson & Ross. Pennsylvania- USA. 912 p.
https://openalex.org/W4239329404	https://www.qeios.com/read/LCD7OP/pdf	English	null	Collaborator	Definitions	2,020	cc-by	68	Qeios · Definition, February 7, 2020 Open Peer Review on Qeios Open Peer Review on Qeios Open Peer Review on Qeios Collaborator National Cancer Institute National Cancer Institute Qeios ID: LCD7OP · https://doi.org/10.32388/LCD7OP Source National Cancer Institute. Collaborator. NCI Thesaurus. Code C84336. National Cancer Institute. Collaborator. NCI Thesaurus. Code C84336. Any of two or more parties working jointly towards a common goal. Qeios ID: LCD7OP · https://doi.org/10.32388/LCD7OP 1/1
https://openalex.org/W2987202385	http://ejournal.iaiibrahimy.ac.id/index.php/tarbiyatuna/article/download/198/216	Indonesian	null	Implementasi Media Pembelajaran Berbasis Teknologi Informasi dan Komunikasi dalam Distance Learning	Jurnal Tarbiyatuna	2,019	cc-by-sa	3,728	TARBIYATUNA: Kajian Pendidikan Islam Volume 3 Nomor 1 Tahun 2019 Print ISSN : 2597-4807 Online ISSN : 2622-1942 TARBIYATUNA: Kajian Pendidikan Islam Volume 3 Nomor 1 Tahun 2019 Print ISSN : 2597-4807 Online ISSN : 2622-1942 TARBIYATUNA: Kajian Pendidikan Islam Volume 3 Nomor 1 Tahun 2019 Print ISSN : 2597-4807 Online ISSN : 2622-1942 Abstract The rapid development of science and advances in advanced technology have implications for the rhythms of human communication. A very striking implication is that people are increasingly free and easy (easy access) in obtaining information and knowledge, including in the field of education. The global education environment is seen in several ways to provide solutions to other gaps and problems through the soaring distributed learning opportunities. Learning process activities with the help of information and communication technology take place not only applied limited by space, location or level of education such as at school or college, but can be done in many different places and involve a lot of people. Learning methods from learners that are not limited to time and place are called information and communication technology based learning. Learning media that can be used in distance learning are: computers, television, radio, internet, voice recordings, and VCD tapes. Keywords: technology information & communication, distance Learning Accepted: Desember 24 2018 Reviewed: Januari 10 2019 Publised: Februari 28 2019 Keywords: technology information & communication, distance Learning This work is licensed under Creative Commons Attribution Non Commercial 4.0 International License Available iaiibrahimy.ac.id PENDAHULUAN Teknologi informasi dan komunikasi memiliki peran penting dalam denyut aktivitas saat ini dan periode yang akan datang, salah satunya dalam elemen pendidikan. The development of the information and communication technology sector as one of the changing products of the times offers new things for education(Islam, Baharun, Muali, Ghufron, & Bali, 2018). Dapat dipahami bahwa perkembangan sektor teknologi informasi dan komunikasi sebagai salah satu produk yang berubah sesuai zaman sekarang untuk menawarkan hal-hal baru bagi pendidikan. Salah satu kelebihan yang diunggulkan oleh teknologi informasi modern adalah dapat melaksanakan kegiatan belajar-mengajar tanpa kehadiran This work is licensed under Creative Commons Attribution Non Commercial 4.0 International License Available iaiibrahimy.ac.id Muhammad Mushfi El Iq Bali pembelajar. Artinya, pebelajar bisa belajar di periode dan lokasi yang berbeda, serta melakukan interaksi jarak-jauh dengan pembelajar. Sebelum pembelajaran jarak jauh dengan memanfaatkan media teknologi informasi sebagai penghubung antara pembelajar dan pebelajar, selama hampir satu abad manusia dibeberapa bagian dunia memiliki independensi untuk menentukan sistem belajar. Sistem belajar yang umum digunakan adalah kegiatan surat-menyurat melalui sistem surat tradisional. Pebelajar menerima pelajaran dalam bentuk panduan tertulis, dan membuat tugas dalam bentuk surat kemudian mendapatkan umpan balik dari pembelajar (Bali, 2018). Tidak dapat dipungkiri pesatnya kemajuan teknologi informasi dan komunikasi dalam dunia pendidikan menjadi pemicu tendensi transisi dari pembelajaran konvensional secara tatap muka ke arah pembelajaran kontemporer yang berbasis e-learning atau pembelajaran jarak jauh yang mampu diakses dengan memanfaatkan media, seperti perangkat hardware dan software, multimedia interaktif dan jaringan internet tanpa dibatasi jarak, ruang, dan waktu oleh siapapun yang memerlukannya. Transformasi sistem pembelajaran tersebut sangat membantu para pebelajar, termasuk di kalangan perguruan tinggi. Pola pembelajaran yang didukung dengan paradigma berpikir tingkat tinggi sangat membutuhkan peranan dan kecanggihan teknologi dalam transfer ilmu tanpa mengesampingkan mobilitas individu. Perguruan tinggi identik dengan kemodernan dan lebih menekankan pendekatan-pendekatan yang bersifat liberal (Bali, 2017b). Dalam artikel ini akan mengulas tentang distance learning atau pembelajaran jarak jauh serta perangkat media yang dapat digunakan dalam pembelajaran jarak jauhdan jenjang pendidikan yang dapat melaksanakan sistem pendidikan jarak jauh. Penggunaan media dalam sistem pendidikan jarak jauh digunakan dengan tujuan dapat mentransfer materi pelajaran dan dapat menjangkau keberadaan pebelajar dimanapun berada. Adapun media yang dimaksud seperti; komputer, televisi, radio, internet, rekaman suara, dan kaset VCD. Tarbiyatuna: Volume 3 Nomor 1, 2019 A. Distance Learning g Distance learning (pembelajaran jarak jauh) sebagai model dari distance education (pendidikan jarak jauh) bukanlah model pendidikan yang baru. Cikal bakal sistem pembelajaran jarak jauh dimulai dengan kursus tertulis, kemudian berkembang sistemnya menjadi pendidikan tinggi formal. Alasan utama diselenggarakannyasistem pembelajaran jarak jauh diperuntukkan bagi orang atau karyawan yang setiap harinya bekerja dengan memiliki waktu kerja yang padat, Tarbiyatuna: Volume 3 Nomor 1, 2019 30 Muhammad Mushfi El Iq Bali berdomisili dan berdinas di lokasi yang berjauhan dari lembaga pendidikan. Pebelajar dengan background karyawan merasa banyak kehilanganwaktu dan berbenturan dengan jam kerja jika harus mengikuti pembelajaran atau perkuliahan secara konvensional pada lembaga pendidikan tersebut. Sedangkan untuk mengikuti jadwal perkuliahan, praktikum, dan semua kegiatan lainnya membutuhkan waktu berjam-jam setiap harinya untuk duduk di kelas. Dengan demikian, pembelajaran jarak jauh dapat mengatasi jarak, tempat, dan waktu dalam menyampaikan materi pembelajaran. Oleh karena itu, pembelajaran jarak jauh memiliki keistimewaan atau distingtif dengan sistem pendidikan yang diselenggarakan secara stereotipyaitu tatap muka. Terpisahnya jasmani antara aktivitas pembelajar dan pebelajartanpa disertai proses tatap muka secara langsung (face to face), sehingga terjadi learning limitednessyang dilakukan dalam bentuk tatap muka. Untuk menanganilearning limitednesspembelajaran jarak jauh yang tanpa disertai dengan tatap mukanya, maka pembelajaran dapat dilengkapi dengan pemanfaatan media yang dapat mengakomodir dan memfasilitasi terjadinya interaksi antara pembelajar dan pebelajar sehingga menjadi lebih efektif dan efisien. Hal tersebut beranjak dari paradigma bahwa individu tidak mungkin bisa membebaskan dirinya dari interaksi dengan orang lain (Bali, 2017a). Pembelajar dapat menetapkanalokasi waktu studinyasesuai keinginan serta luwes dalam menetukan ruang belajarnya, yang disesuaikan dengan akselerasi dan teknik belajarnya. Media pembelajaran yang esensial dalam pembelajaran jarak jauh pada awalnya cukup menggunakan modul, namun selaras dengan kemajuan teknologi informasi dan komunikasi (Munir, 2009). Maka muncullah media pembelajaran berbantuan komputer, audio, vidio, media cetak, multimedia, internet, dan lain- lain. Program pendidikan dengan menerapkan sistem pembelajaran jarak jauh umumnya ada dua kategori yaitu program reguler dan non reguler. Program non reguler merupakan program yang diselenggarakan secara khusus, misalnya khusus untuk pegawai atau karyawan suatu instansi yang ingin meningkatkan sumber daya manusianya. Hal tersebut menjadi alasan ilmu ekonomi memberikan pelatihan yang efektif yang didistribusikan untuk sejumlah orang di berbagai situs. Dalam banyak kasus, misalnya perusahaan multinasional sering menegaskan bahwa pelatihan penting untuk diberikan. Sebagai contoh, strategi pemasaran untuk jaringan traktor kebun atau petugas asuransi dituntut untuk menekan jumlah barang dengan presentasi penjualan mereka. Tarbiyatuna: Volume 3 Nomor 1, 2019 A. Distance Learning Kursus pelatihan dikemas dengan menggunakan media telekomunikasi berupa program televisi sehingga Tarbiyatuna: Volume 3 Nomor 1, 2019 31 Muhammad Mushfi El Iq Bali dengan cepat dapat memberikan informasi. dengan cepat dapat memberikan informasi. Seperti sistem pendidikan lainnya, sistem pembelajaran jarak jauh juga memiliki jenjang/tingkat pendidikan yang dapat disinkronkan dengan kebutuhan dan segmen di kalangan masyarakat. Pembelajaran jarak jauh untuk jenjang sekolah dasar dan tingkat menengah, pemanfaatan media pembelajaran oleh pebelajarlebih condong pada kaset video rekaman (cassette recorder) daripada program-program siaran televisi langsung. Beberapa serialprogram siaran televisi yang sering digunakan di tingkat awal sekolah dasar adalah Edukasi TV. Program ini sebagai pengayaan bukan sebagai esensi dari pembelajaran. Sementara itu sistem pembelajaran jarak jauh di tingkat tinggi, cenderung menggunakan sistem telekomunikasi. Sistem telekomunikasi tersebut sering diterapkanpada perkuliahan di kampus maupun di luar kampus. Beberapa perguruan tinggi menghadirkan telekomunikasi sebagai rutinitas pokok dari program distingsi kampus tertentu. Tujuan lazimnya adalah berupaya untuk meningkatkan kompetensi mahasiswa yang dapat direngkuh oleh pembelajar. Contohnya ketika seorang dosen atau Guru Besar dalam proses pembelajarannya dapat menyampaikan materi perkuliahan dengan berbicara melalui sebuah studio (microteaching)atau ruang kelas yang difasilitasidengan kamera dan mikrofon. B. Fungsi Komunikasi Instruksional g Terlepas dari teknologi yang digunakan, hal yang disajikan pembelajar melalui sistem telekomunikasi instruksional harus memenuhi syarat tertentu untuk menjadi efektif dalam penyajian informasi. Teacher need the skills to construct and manage class- room activities efficiently, communicate well, use technology, and reflect on their practice to learn from and improve it continually (Darling-Hammond, 2006). Dapat disimpulkan bahwa pembelajar membutuhkan keterampilan untuk membangun dan mengelola kegiatan ruang kelas secara efisien, berkomunikasi dengan baik, menggunakan teknologi, dan merefleksikan praktik pembelajaran dan meningkatkannya secara terus menerus. Barometer dalam materi pelajaran apapun untuk penyajian informasi dapatberbentuk presentasi yang disampaikan pada pebelajar. Berikut ini contoh umum yang sering ditemui yaitu: 1) Guru ceramah dan demonstrasi; 2) Teks tercetak dan ilustrasi (misalnya, buku teks, handout, bahan studi korespondensi); 3) Perekam suara, musik, dan suara lainnya; 4) Gambar full-motion (video, CD-ROM); dan 5) Kegiatan mahasiswa, karena sebagian besar pembelajaran terjadi ketika pembelajar memproses materi dengan mental yang berpartisipasi aktif (Munadi, 2013). Tarbiyatuna: Volume 3 Nomor 1, 2019 32 Muhammad Mushfi El Iq Bali Tabel 1. Sistem Komunikasi Sistem Presentasi Interaksi Siaran radio Rekaman audio Tes melalui pos Rekaman suara Tanya jawab melalui rekaman dengan memberikan PR dan umpan balik. Uji melalui surat Rekaman suara bergambar Rekaman dan melakukan komunikasi interaktif tentang materi yang didengarkan. Grafik gambar diam, uji melalui surat/fax Konferensi via komputer Teks elektronik, data, grafik Tugas yang di tulis ditukar dengan siswa lain dan guru Televisi Audiovisual Suara, musik, gambar diam, grafik, gambar bergerak Uji melalui surat Televisi Audio dua arah dan video Suara, musik langsung, gambar gerak Adanya interaksi tanya jawab dengan siswa lain melalui gambar dan suara. Televisi Vidio dan audio dua arah Suara, musik langsung, gambar diam, grafik, gambar bergerak Tugas dikirim dengan surat. Tabel 1. Sistem Komunikasi Bagi tujuan pendidikan, interaksi antar pebelajardi dalam kelompok atau kelompok-kelompok kecildapat sangat efektif. Beberapa kaidahlazim interaksi pebelajar dalam kelompok sebagai berikut: 1) Diskusi kelompok (dalam atau di luar kelas); 2) Kegiatan kelompok terstruktur (misalnya, bermain peran atau permainan); 3) Proyek kelompok; 4) Tutor sebaya; dan 5) Akses ke sumber belajar. Pelajaran dan materi pelajaran umumnyaintegral dengan proposisi bahwa pebelajar akan menamatkan waktu di luar kelas dengan bekerja secara individual Tarbiyatuna: Volume 3 Nomor 1, 2019 33 Muhammad Mushfi El Iq Bali dalam mempelajari materi kuliah, mengerjakan PR, proyek, kertas, dan sejenisnya. B. Fungsi Komunikasi Instruksional Sumber daya eksternal belajar memiliki bentuk yang bervariasi, diantaranya: 1) Bahan ajar cetak (misalnya, buku teks, bacaan, lembar kerja); 2) Materi audiovisual (misalnya, audio atau kaset video, multimedia, CD-ROM); 3) Database komputer (misalnya, untuk pencarian online); dan 4) Bahan pustaka (misalnya, dokumen sumber asli). Masing-masing dari sistem telekomunikasi yang difungsikan sebagai media mempunyainilai positif dan negatif di daerah-daerah tertentu. Oleh karena itu, pemanfaatan media telekomunikasi harus relevan, ajeg dan konsisten. Karakteristik dari sistem media telekomunikasi tersebut dibahas secara rinci berikut ini. Tarbiyatuna: Volume 3 Nomor 1, 2019 1. Telekonferensi audio Sistem yang berbeda dengan melibatkan perangkat media telekomunikasi yang bisa diterapkan dalam pembelajaran jarak jauh yaitu telekonferensi audio. Telekonferensi audio atau lebih familiar disebut audio teleconference hakikatnya merupakan perkembangan dari pemanfaatan pesawat telepon. Kecanggihan teknologi komunikasi saat ini memberikan peluang untuk melakukan komunikasi dua arah melalui percakapan langsung menggunakan audio teleconference yang terkoneksi dengan jaringan telepon maupun satelit. Telekonferens dapat dijadikan sebagai mediasi komunikasi, dialog dan diskusi antar individu maupun kelompok dengan lokasi yang berbeda, pada satuan waktu yang bersamaan. a. Keunggulan telekonferensi audio Adapun kelebihan yang dimiliki audio teleconference meliputi: 1) Penggunaan telepon bahkan telepon seluler sangat familiar di masyarakat, sehingga penggunaan telekonferens pun sangat potensial; 2) Terjangkau, efektif dan efisien untuk dioperasikan; 3) Jangkauan yang luas, sehingga mendorong mahasiswa berpartisipasi aktif dimanapun berada; dan 4) Tingkat interaktivitas yang tinggi, sehingga memungkinkan audience dan narasumber dapat saling berbicara antara yang satu dengan yang lain. b. Keterbatasan telekonferensi audio 2. Internet Teknologi online yang bergantung pada sistem berbasis komputer telah membuka berbagai peluang baru pembelajaran jarak jauh. Program perangkat lunak, kamera, dan berbagai sumber daya dapat memberikan pebelajar pengalaman pembelajaran jarak jauh sama dengan yang digunakan sistem konvensional. Pebelajar dapat menghadiri kelas pada waktu yang ditentukan, dapat berpartisipasi dalam audio, video, atau berbasis teks chat, bisa membalas diskusi papan buletin, dan dapat mengakses informasi yang berkaitan dengan kursus langsung dari komputer. Konferensi melalui internet memiliki keunggulan antara lain: 1) Menjangkau pebelajar yang jumlahnya tidak terbatas pada waktu yang bersamaan; 2) Tidak dibatasi oleh ruang dan waktu, bahkan teritorial Negara; dan 3) Mampu melakukan hubungan ke lokasi (site) di negara lain. Konferensi melalui internet memiliki keunggulan antara lain: 1) Menjangkau pebelajar yang jumlahnya tidak terbatas pada waktu yang bersamaan; 2) Tidak dibatasi oleh ruang dan waktu, bahkan teritorial Negara; dan 3) Mampu l k k h b k l k i ( i ) di l i j g p j y g j y p y g 2) Tidak dibatasi oleh ruang dan waktu, bahkan teritorial Negara; dan 3) Mampu melakukan hubungan ke lokasi (site) di negara lain. Adapun kelemahan internet dalam pembelajaran jarak jauh, yaitu: 1) Membutuhkan keterampilan menggunakan computer; dan 2) Jaringan internet yang fluktuatif sehingga berpengaruh pada arus koneksi internet. c. Integrasi Penggunaan Internet dalam Pembelajaran Jarak Jauh Denganinternet pebelajar tidak harus berkendara jarak jauh untuk menghadiri kelas untuk menerima materi pelajaran, melalui media yang disampaikan oleh koneksi video atau audio, sistem berbasis komputer membantu mereka tetap belajar dan mendapatkan materi pelajaran walaupun mereka berada di rumah b. Keterbatasan telekonferensi audio Sedangkan keterbatasan audio teleconference antara lain: 1) Tidak mampu menyuguhkan materi yang visualis. Solusinya, mempersiapkan terlebih dahulu materi yang bersifat visual di lokasi konferensi sebelum kegiatan dimulai; 2) Kualitas suara yang diterima kurang baik. Hal yang harus dilakukan adalah memperhatikan peralatan microphone-amplifier khusus di setiap lokasi; dan 3) Terbatasnya tenaga ahli yang berpengalaman sehingga mempengaruhi minat kegiatan telekonferensi audio. Tarbiyatuna: Volume 3 Nomor 1, 2019 34 Muhammad Mushfi El Iq Bali c. Integrasi penggunaan telekonferensi dalam pembelajaran jarak jauh c. Integrasi penggunaan telekonferensi dalam pembelajaran jarak jauh Sistem ini sering digunakan di bagian tingkat sekolah menengah untuk menghubungkan pebelajar dalam dua atau lebih situs, dengan seorang penulis untuk membahas tulisannya atau dengan pejabat publik untuk membahas undang- undang. Audio telekonferensi populer dalam pendidikan perusahaan dan profesional untuk pelatihan, misalnya membahas fitur layanan baru, mengajar perwakilan penjualan teknik terbaru, untuk memperbarui akuntan pada perubahan dalam hukum pajak, dan sebagainya. Tarbiyatuna: Volume 3 Nomor 1, 2019 1. Peran pebelajar Pebelajar perlu mengetahui peranannya dalam pengalaman pembelajaran jarak jauh. Upaya awal pada proses pembelajaran cenderung melibatkan pebelajar berbicara dengan pebelajar yang pasif dan hanya duduk menjelajahi situs jarak Tarbiyatuna: Volume 3 Nomor 1, 2019 35 Muhammad Mushfi El Iq Bali jauh yang sering menunjukkan ketidakhadiran pebelajar (Marisa, 2014). Dengan kemajuan teknologi terbaru, ada interaksi antar pebelajar melalui situs layanan dalam aplikasi di internet. Pebelajar dapat menjadi lebih terlibat dalam pembelajarannya. Tanggung jawab pembelajar dalam pembelajaran jarak jauh adalah membimbing pebelajar untuk berinteraksi dengan tepat. The students may use imprecise language when communicating their views,questions, and criticism to their peers (Kumpulainen, K and Wray, 2002). Artinya pebelajar dapat menggunakan bahasa yang tidak tepat saat mengkomunikasikan pandangan, pertanyaan, dan kritikan pebelajar kepada rekan-rekannya. Pebelajar perlu mengetahui cara menggunakan teknologi untuk berkomunikasi dengan pembelajar dan dengan pebelajaryang lain. Tarbiyatuna: Volume 3 Nomor 1, 2019 2. Peran pembelajar Ketika mulai berbicara tentang pembelajar di kelas pembelajaran jarak jauh, maka perlu untuk berpikir tentang pengaturan dalam kelas (Karawati, 2015). Kelas yang digunakan ini adalah serangkaian kelas yang terhubung secara elektronik. Lokasi ruang kelas yang jauh akan terhubung oleh sistem telekomunikasi. Di lokasi jarak jauh, mungkin ada satu atau dua pebelajar yang dapat dijadikan relasi pembelajar sebagai tutor di kelas. Selain itu, mungkin ada fasilitator yang bertanggung jawab untuk bekerja sama dengan pembelajar. Fasilitator merupakan pembelajar lain atau asisten di kelas. Tugas fasilitator bervariasi tergantung pada isi kegiatan dan kebutuhan pembelajar di kelas. Beberapa hal yang harus dilakukanpembelajardalam pembelajaran jarak jauh yaitu: 1) Mengawasi dan berpartisipasi aktif dalam semua program dengan pebelajar; 2) Mendorong interaksi dengan pembelajar dan pebelajar lainnya; 3) Menjawab pertanyaan di situs tersebut; 4) Memecahkan masalah dengan segera; 5) Disediakan tambahan kuis dan lembar kegiatan/kerja; dan 6) Mengambil tanggung jawab untuk operasi dan troubleshooting peralatan. Dalam lingkup belajar online, pembelajar memikul tanggung jawab lebih baik dalam hal perencanaan. Teacher planning is a multifaceted and ongoing process that covers almost everything teachers do (Arends, 2012). Dapat diartikan bahwa perencanaan pembelajar adalah proses beragam dan berkelanjutan yang mencakup hampir semua yang dilakukan pembelajar. Bahan ajar harus dipersiapkan sebelumnya seperti persiapan pembelajaran di dalam kelas yang sebenarnya. Pembelajar juga harus memahami prihal yang diharapkan dari mereka dalam hal pola responsif pebelajar. Misalnya, jika pembelajar menulis tanggapan terhadap pertanyaan di papan buletin, mereka tidak hanya perlu tahu di mana untuk mengirim, tetapi harapan atau target pencapaian yang harus dikuasai Tarbiyatuna: Volume 3 Nomor 1, 2019 36 Muhammad Mushfi El Iq Bali pebelajar dari materi yang akan diposting. Tarbiyatuna: Volume 3 Nomor 1, 2019 D. Mengakses Sumber Daya Jarak Jauh g y Salah satu elemen yang sering diabaikan dalam situasi pembelajaran jarak jauh adalah akses pebelajar untuk mendapatkan sumber daya informasi. Jumlah sumber daya informasi yang tersedia terus bertambah pada tingkat yang eksponensial. Jika seorang pembelajar ingin pebelajar terlibat dalam penelitian atau jenis kegiatan tertentu, sangat penting bahwa mereka memiliki akses ke bahan-bahan terkait. Seorang pembelajar mungkin perlu mengubah jenis kegiatan tertentu atau untuk membuat pengaturan khusus untuk bahan yang akan dikirim ke kelas jarak jauh. Pebelajar di situs on-line seharusnya tidak merasakan kerugian belajar karena sumber daya yang terbatas. Ini adalah tanggung jawab pembelajar yang harus berkoordinasi dan berkolaborasi dengan spesialis media atau ahli teknologi informatika (teknisi), untuk memastikan bahwa semua pebelajar memiliki akses yang simetris terhadap bahan penting untuk belajar. Sementara internet dan World Wide Web telah sedikit mengurangi kekhawatiran ini. Web- based instruction pro- vides students with access to more resources in less time than is possible in traditional ways(Schunk, 1996), Instruksi berbasis web memberi siswa akses ke lebih banyak sumber daya dalam waktu yang lebih singkat daripada yang dimungkinkan dengan cara tradisional. Adanya beberapa program sumber daya bagi pebelajar yang tidak tersedia di Web, atau masalah hak cipta tidak memungkinkan menggunakan Web untuk menyediakan sumber daya tersebut. Adapun keuntungan yang diperoleh dalam mengakses sumber daya jarak jauh, antara lain: 1) Efisiensi biaya, Segala bentuk saham penyiaran atribut mencapai geografis audience dengan cara yang hemat biaya; 2) Kapasitas audiovisual, Semua sistem televisi memungkinkan transmisi gambar gerak dan suara melalui jarak jauh; 3) Kemungkinan dua arah, Ketika pebelajar dapat berkomunikasi dengan instruktur dan pebelajar lainnya melalui telepon atau video dua-arah, pembelajar dapat memungkinkan interaksi kelas hidup; 4) Kemungkinan Online, Jika tersedia, teknologi online dapat memberikan akses ke sumber daya dan instruksi yang mungkin tidak dinyatakan mungkin. Penilaian portofolio Pembelajar memberikan evaluasi ke pebelajar berupa tugas Portofolio dengan CD-ROM dan Companion Website sebagai sumber daya informasi dalam menyelesaikan tugas portofolio, meliputi; 1) Perencanaanpengajaran jarak jauh. Mencari materi pelajaran pada Portofolio CD-ROM atau internet pada topik atau standar pilihan pebelajar. Pertama, perhatikan kegiatan dan media yang diaplikasikan ke dalam pelajaran. Kedua, perhatikan perubahan yang diperlukan untuk menyediakan bahwa pengalaman belajar di kejauhan. Ketiga, mengidentifikasi alasan untuk perubahan. Kemudian, menggambarkan strategi penilaian yang akan digunakan; dan 2) Menulis refleksi dan merenungkan pengalaman belajar sebelumnya. Untuk melengkapi kegiatan tertentu secara online, pebelajar diminta untuk mengacu pada Penilaian Integrasi, meliputi; 1) Mengawasi penggunaan radio untuk tujuan instruksional di sekolah maupun perguruan tinggi. Pemeriksaandengan stasiun radio publik lokal untuk melihat jika mendukung kegiatan pembelajaran khusus; 2) Wawancara seorang pembelajar secara teratur. Siapkan dalam bentuk tertulis atau direkam laporan singkat menangani tujuan pembelajaran, teknik yang digunakan, dan masalah yang dihadapi saat menggunakan media on-line; 3) Menampilkan daftar siswa yang menggunakan aplikasi telekomunikasi untuk proses belajar; 4) Menyiapkan abstrak laporan sebuah proyek penelitian atau demonstrasi terkait dengan telekomunikasi instruksional (misalnya, dua sekolah berbagi satu pembelajar lewat telekonferensi interaktif). D. Mengakses Sumber Daya Jarak Jauh Sedangkan keterbatasan dari mengakses sumber daya jarak jauh, meliputi; 1) Biaya untuk dua arah, Menambahkan kapasitas untuk komunikasi dua arah mungkin memerlukan instalasi hardware yang mahal, termasuk sebuah connector jika beberapa situs yang terhubung sekaligus; 2) Fasilitas untuk dua arah, Para setup khusus yang diperlukan untuk video dua-arah mengharuskan kelas didedikasikan untuk penggunaan e-learning, sehingga membuatnya sulit untuk menggunakan ruang kelas untuk tujuan lain; 3) Isolasi, Pebelajar yang Tarbiyatuna: Volume 3 Nomor 1, 2019 37 Muhammad Mushfi El Iq Bali berpartisipasi dalam pelajaran siaran tanpa kemampuan talkbackdapat merasa seperti warga kelas dua memiliki sedikit hubungan dengan anggota kelompok lainnya; 4) Masalah teknis, Masalah teknis dapat mengganggu instruksi dan dapat menciptakan kebingungan dan frustrasi bagi instruktur dan pebelajar; 5) Pengalaman, Instruktur mungkin tidak merasa nyaman mengajar dalam jenis pengaturan; dan 6) Keengganan,Pebelajar mungkin enggan untuk memikul tanggung jawab yang lebih besar untuk pembelajaran mereka sendiri. Tarbiyatuna: Volume 3 Nomor 1, 2019 SIMPULAN Pembelajaran jarak jauh (distance learning) merupakan salah satu metode pembelajaran modern yang memberikan peluang bagi pebelajar yang telah berkarir di dunia kerja. Metode ini menjawab harapan pebelajar karir untuk Tarbiyatuna: Volume 3 Nomor 1, 2019 38 Muhammad Mushfi El Iq Bali meningkatkan pengetahuan, pengalaman, dan aktualisasi diri dalam menyelaraskan skill individu dengan perkembangan dan tuntutan kerja. Pemanfaatan teknologi informasi dan komunikasi dalam pembelajaran jarak jauh sangat membantu keterlaksanaan proses pembelajaran tanpa harus melakukan tatap muka. Media komunikasi yang digunakan meliputi; audio atau kaset video, multimedia, CD-ROM, jaringan internet, dsb. Kecanggihan teknologi komunikasi tersebut menjadi media yang optimal dalam penyampaian materi/konsep, tugas berkala, ujian semester, bahkan dalam penyelesaian tugas akhir. Penilaian yang dilakukan dalam pembelajaran jarak jauh (distance learning) menggunakan dua cara, yaitu penilaian portofolio dan penilaian terintegrasi. DAFTAR RUJUKAN . Arends, R. I. (2012). Learning to Teach (9th ed.). New York: McGraw-Hill Bali, M. M. E. I. (2017)a. Model Interaksi Sosial dalam Mengelaborasi Keterampilan Sosial. Pedagogik, 04(02), 211–227 Bali, M. M. E. I. (2017)b. Perguruan Tinggi Islam Berbasis Pondok Pesantren. Al- tanzim, 1(2), 1–14 Bali, M. M. E. I. (2018). Strategi Pembelajaran Pendidikan Agama Islam. Probolinggo: Pustaka Nurja Darling-Hammond, L. (2006). Constructing 21st-century Teacher Education. Journal of Teacher Education, 57(3), 300–314. https://doi.org/10.1177/0022487105285962 Heinich, Robert dkk. (2002). Intructional Media and Technologies for Learning. Seventh Edition. New Jersey: Pearson Education inc Islam, S., Baharun, H., Muali, C., Ghufron, M. I., & Bali, M. M. E. I. (2018). To Boost Students ’Motivation and Achievement through Blended Learning. Journal of Physics: Conference Series, 1–11 Karawati, Euis & Juni P, D. (2015). Manajemen Kelas. Bandung: Alfabeta Kumpulainen, K & Wray, D. (2002). Classroom Interaction and Social Learning. New York: Taylor & Francis Group Marisa, dkk. (2014). Komputer dan Media Pembelajaran. Banten: UT-Kementerian Tarbiyatuna: Volume 3 Nomor 1, 2019 39 Muhammad Mushfi El Iq Bali Pendidikan dan Kebudayaan Munadi, Yudhi. (2013). Media Pembelajaran: Sebuah Pendekatan Baru. Jakarta: Referensi Munir. (2009). Pembelajaran Jarak Jauh Berbasis Teknologi Informasi dan Komunikasi. Bandung: Alfabeta Schunk, D. H. (1996). Learning Theories an Educational Perspective. Printice Hall Inc., New Jersey (Sixth Edit, Vol. 53). Boston: Allyn & Bacon. https://doi.org/10.1017/CBO9781107415324.004 Schunk, D. H. (1996). Learning Theories an Educational Perspective. Printice Hall Inc., New Jersey (Sixth Edit, Vol. 53). Boston: Allyn & Bacon. https://doi.org/10.1017/CBO9781107415324.004 Pendidikan dan Kebudayaan unadi, Yudhi. (2013). Media Pembelajaran: Sebuah Pendekatan Baru. Jakarta: Referensi Munadi, Yudhi. (2013). Media Pembelajaran: Sebuah Pendekatan Baru. Jakarta: Referensi Munir. (2009). Pembelajaran Jarak Jauh Berbasis Teknologi Informasi dan Komunikasi. Bandung: Alfabeta Schunk, D. H. (1996). Learning Theories an Educational Perspective. Printice Hall Inc., New Jersey (Sixth Edit, Vol. 53). Boston: Allyn & Bacon. https://doi.org/10.1017/CBO9781107415324.004 Tarbiyatuna: Volume 3 Nomor 1, 2019 40
https://openalex.org/W2154700639	http://ibimapublishing.com/articles/JMED/2014/539605/539605.pdf	English	null	Evaluation of the Anti-Diabetic and Anti-Ulcer Properties of Some Jordanian and Iraqi Medicinal Plants; a Screening Study	JMED research	2,014	cc-by	4,273	Cite this Article as: Husni A Twaij and Emad A S Al-Dujaili (2014), “Evaluation of the Anti-Diabetic and Anti- Ulcer Properties of Some Jordanian and Iraqi Medicinal Plants; a Screening Study,” JMED Research, Vol. 2014 (2014), Article ID 539605, DOI: 10.5171/2014.539605 Abstract A large number of Jordanian and Iraqi medicinal plants have been used in traditional medicine for hypoglycemic and anti-ulcer activities. We have started a screening program to evaluate the activity of some of these plants and their extracts taking in consideration their use as an alternative therapy. Various parts of several plants were extracted and the hypoglycemic effects of the aqueous extracts have been examined in normal, glucose loaded and alloxan diabetic experimental animals. Teucrium polium aqueous extract was studied for its anti-ulcer activity of ulcers induced in rats by starvation and reserpine and/or stress. The aqueous extract of the aerial parts of Artemisia herba alba was administered orally to normoglycemic and alloxan diabetic rabbits and rats and found to produce a significant hypoglycemic activity (normoglycaemic: plasma glucose levels reduced from 116±4 to 90±3mg/dL after 90minutes, p<0.001; diabetic animals from 471±7 to 399±9mg/dL, p<0.001). The aerial aqueous extract of Phangnalon rupestre, Sinapis arvensis and Eryngium creticum also exhibited significant hypoglycemic effects in euglycaemic and glucose loaded rats. The other plants examined (Ditrichia graveolens, Alkanna strigosa, , Centaurea iberica, Asphodelin lutea, Paronychia argentea, rosmarinus officinalis, Achillea santolina, Achillea micrantha, Allium sativum, Centaurea phyllacephala, Centaurea behen, Crataegus azarolus, Prosopis farcta and Myrtus communis) failed to produce a significant change in the plasma glucose levels. One herb was comprehensively tested in detail for its anti-ulcer activity; Teucrium polium aqueous extract (intra- peritoneally;150mg/kg) produced an average of 50% healing of ulcers where as the orally administered extract produced 85% healing activity of ulcers when compared with the controls. Keywords: Medicinal plants, Anti-diabetic, Anti-ulcer. offer a useful support to treat the disease. The present investigative project represents the up to date results obtained through a continuing programme related to the Research Article Evaluation of the Anti-Diabetic and Anti-Ulcer Properties of Some Jordanian and Iraqi Medicinal Plants; a Screening Study Husni A Twaij1 and Emad A S Al-Dujaili2 1Faculty of Pharmacy, Philadelphia University, Jordan Academic Editor: Ozra Tabatabaei-Malazy Copyright © 2014 Husni A Twaij and Emad A S Al-Dujaili. Distributed under Creative Commons CC-BY 3.0 IBIMA Publishing JMED Research http://www.ibimapublishing.com/journals/JMED/jmed.html Vol. 2014 (2014), Article ID 539605, 10 pages DOI: 10.5171/2014.539605 IBIMA Publishing JMED Research http://www.ibimapublishing.com/journals/JMED/jmed.html Vol. 2014 (2014), Article ID 539605, 10 pages DOI: 10.5171/2014.539605 Animals Ethical approval for all the experiments performed was granted by the Faculty of Pharmacy, Philadelphia University Ethics Committee. Male adult Wistar rats (250-300 g), male adult Swiss albino mice (20-25 g) and male white New Zealand rabbits (1-2 kg) were used in the present studies. Animals described as fasted were deprived of food for at least 16 hours but allowed free access to tap water. The rats were anesthetized with sodium pentobarbitone, administered intraperitoneally at a dose of 40mg/kg. Collection of blood samples and administration of the materials to the rats were facilitated by the cannulated femoral arteries and veins respectively. Regarding the rabbits the marginal ear vein is used for administration of the materials and collection of blood samples. Alloxan diabetic animals were prepared by giving a dose of 150mg /kg of freshly prepared alloxan monohydrate (BDH Chemicals). Glucose loads (50% w/v, 4mL /kg) were given to all animals 30 min following the plant extract or control vehicle administration. Ethical approval for all the experiments performed was granted by the Faculty of Pharmacy, Philadelphia University Ethics Committee. Male adult Wistar rats (250-300 g), male adult Swiss albino mice (20-25 g) and male white New Zealand rabbits (1-2 kg) were used in the present studies. Animals described as fasted were deprived of food for at least 16 hours but allowed free access to tap water. The rats were anesthetized with sodium pentobarbitone, administered intraperitoneally at a dose of 40mg/kg. Collection of blood samples and administration of the materials to the rats were facilitated by the cannulated femoral arteries and veins respectively. Regarding the rabbits the marginal ear vein is used for administration of the materials and collection of blood samples. Alloxan diabetic animals were prepared by giving a dose of 150mg /kg of freshly prepared alloxan monohydrate (BDH Chemicals). Glucose loads (50% w/v, 4mL /kg) were given to all animals 30 min following the plant extract or control vehicle administration. Effect on Plasma Glucose The animals were divided into 3 groups of eight animals each and were then given the following to each group respectively. 1. The aqueous extract of the plants under study. Introduction Diabetes mellitus rely primarily on a life style changes and natural remedies may definitely JMED Research JMED Research 2 biological and biochemical screening of the local plants used in folk medicine. Throughout the ages, Arabian physicians have attempted the treatment of diabetes mellitus patients with indigenous plants (Said, 1969). Farnsworth and Segelman (1971) have described the activity of many herbs and plants that have also been pharmacologically proven to be of some value in diabetes mellitus (Lewis and Elvin- Lewis, 1977). Recently, primary prevention of diabetes mellitus has become necessary to control the huge spread of the disease to people of all ages as evidenced by several studies reported by Saravanamuttu (2012). It provides effective management and may prevent the onset of disease particularly in high-risk subjects. All of the plants tested under this study are commonly known by Arabic names, are popular folk remedy and claimed to be effective for the treatment of diabetes mellitus and some plants as anti- ulcer in Iraq and Jordan. For this purpose, the native populations have usually used a hot water decoction made from the fresh leaves and branches. The aim of this investigative study was to use the aqueous extracts of various plants in our experiments to mimic the procedure used in the traditional practices and report if there is really any effect on glucose levels. One plant extract was chosen to assess its anti-ulcer characteristics due to time and finance constraints. Behavioral Effects and Toxicity Gross behavioral studies were carried out in mice together with the LD50 studies including spontaneous motor activity and rectal temperature measurements. The animals were observed continuously for one hour, intermittently every 3 hours and then after 24 hours. Determination of the LD 50 Aqueous extract of T. polium (150mg/kg i.p.), saline (1ml) and Cimitidine 20mg/kg subcutaneously were given three times as follow; 120 min before, 15 min and 120 min after reserpine administration. The twelve hours period starvation was then continued as above. Afterwards the animals were allowed for normal housing with food ad lib for 48 hours during which T.polium aqueous extract (150 mg/kg) or controls were given orally twice daily (Twaij and Al-Badr, 1988). The ulcers were assessed as above and the degree of healing was determined accordingly and reported as percentage in comparison with the untreated animals. Increasing doses of the plant extracts were given to two groups of twelve mice. They were kept in transparent cages at 24 C. Mortality was recorded within 24 hours period (Litchfield and Wilcoxon, 1949). Materials and Methods 2. The negative control (the vehicle solvent of the extracts). The plants materials used in this study were collected in spring from the region of Iraq and Jordan. The samples were identified and authenticated by a specialist at the National Herbarium Department, Amman, Jordan. The aerial parts of the plants (100g) were shade dried, pulverized and extracted with 500 mL of distilled water at room temperature overnight, then filtered and the filtrate evaporated in vacuum at 40oC. The extracts obtained were dissolved in distilled water and administered to the animals following the proper dilutions to achieve the dose required. 3. The positive control (the reference drug Glyburide). Blood samples were then taken just before the administration of the extracts and the controls and after 15, 30, 60 and 90 min post treatment. For the detailed Glucose load experiments, basal blood samples were taken at -30 min, just before the administration of the extracts and the controls (zero time) and after 30, 60, 90, 120, and 150 min there after. Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 3 Anti- Ulcer Activity Induction of ulcer: Normal white Wistar rats (300-350 g) of both sexes were used. The animal were subdivided into 2 groups, each animal was isolated in a separate cage with raised bottom and a wide wire mesh to prevent caprophagia. The animals were starved for 24 hours but water was permitted ad lib, then reserpine was administered (20mg/kg, ip). The animals were then continued on starvation in the same cages for another 12 hours before sacrificing them after reserpine or vehicle treatments. The stomachs were removed, filled with 10mL of 1% formalin in saline and then immersed in 10% formalin in saline, then each stomach was opened along the greater curvature and the mucosa was examined by the naked eye and under dissecting microscope. Ulcers were counted and measured to generate the ulcer index, which represent the total surface area of ulcer in mm² per stomach. Representative ulcers were sampled for histopathological examination, processed by conventional technique using the Shandon automatic processor, and stained by hematoxylin and eosin. Sections were used to assess the depth of ulcers. All data collected was first stored using Microsoft Excel. A two tailed paired t-test was used first to determine any significant difference between baseline and intervention. Statistical analysis was then performed using One-Way ANOVA in SPSS software for Windows version 17.0 and using Student’s T-tests in Microsoft Excel 2010. All data is expressed as Mean ± Standard deviation or Error of the Mean (±SD or SEM) unless otherwise stated. P ≤ 0.05 was considered statistically significant for all tests. Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 Behavioral Effects and Acute Toxicity LD50 and extract’s yield for the plants investigated are shown in Table 1. All of the extracts under study had no significant changes in the spontaneous motor activity and rectal temperature. They did not produce any other changes in behaviour, food and water intake and the morphology of viscera (Twaij et al, 1987). Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 4 4 Table 1: LD50 and Extract’s Yield for the Plants Investigated. Name in Brackets is the Plant Family Scientific name Local name % Yield LD50 Asphodelin Lutea (Liliaceae) Abu Swai 12.37 6800mg/kg equivalent to 54.97g crude Eryngium Creticum Lam (Umbelliferaceae) Shuk al-Akrabati 10.08 2400mg/kg equivalent to 23.8g crude Paronychia Argentea (Caryophyllaceae) Regel al-Hamama 12.5 420mg/kg equivalent to 3.36g crude Rosmarinus Officinalis Ekleel-elbahar 14.0 55.86g/kg of the crude powder Centaurea phyllacephala 12.56 2050mg/kg equivalent to 33.8g crude Centaurea behen 11.44 2010mg/kg equivalent to 25.8g crude Crataegus EL-Nabeq 10.82 3200mg/kg equivalent to 43.8g crude Teucrium polium 13.6 380mg/kg equivalent to 3.46g crude Table 1: LD50 and Extract’s Yield for the Plants Investigated. Name in Brackets is the Plant Family Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 Anti-diabetic and Hypoglycemic Study Asphodelin lutea, Paronychia argentea, rosmarinus officinalis, Achillea santolina, Achillea micrantha, Allium sativum, Centaurea phyllacephala, Centaurea behen, Crataegus azarolus failed to produce a significant hypoglycemic or hyperglycemic effects. Some produced slight hyperglycemic effects or mild elevation in the plasma glucose levels (AI-Khazarji et al, 1993; Twaij et al, 1989; Al-Shamaony et al, 1994; Twaij et al, 1983; Twaij et al, 1987b; Twaij et al, 1988a & b), in comparison with effects obtained from the controls. On the other hand, Myrtus communis and Prosopis farcta produced a significant hyperglycemic activity at 90 min compared to zero time. See Tables 2 and 3. The results obtained from the present studies showed that the oral administration of the aerial parts of Artemisea herba alba to glucose-fed, and to alloxan-induced diabetic rats and rabbits produced significant hypoglycemic activity (Twaij et al, 1987a). The aerial aqueous extract of both Phangnalon rupestre and Eryngium creticum also exhibited a potent and significant hypoglycemic effects in normal-fed and glucose loaded rats (Twaij and A-Jaleel, 2002; A-Jaleel and Twaij, 2002). In contrast to the above mentioned positive effects, all the following plant extracts (Ditrichia graveolens, Alkanna strigosa, Sinapis arvensis, Centaurea iberica, The results obtained from the present studies showed that the oral administration of the aerial parts of Artemisea herba alba to glucose-fed, and to alloxan-induced diabetic rats and rabbits produced significant hypoglycemic activity (Twaij et al, 1987a). The aerial aqueous extract of both Phangnalon rupestre and Eryngium creticum also exhibited a potent and significant hypoglycemic effects in normal-fed and glucose loaded rats (Twaij and A-Jaleel, 2002; A-Jaleel and Twaij, 2002). In contrast to the above mentioned positive effects, all the following plant extracts (Ditrichia graveolens, Alkanna strigosa, Sinapis arvensis, Centaurea iberica, Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 5 Table 2: Effect of the Aqueous Extract of Various Types of Plant on the Plasma Level of glucose (mg/dL) in Normal, Glucose Load and Alloxan Diabetic Rats in Comparison with their Vehicle Solvent (Saline) in Control Rats (Values are Mean±sem) Name of the plants and their Controls Pretreatment Values Post treatment values (min.) 0 15 30 60 90 1. Artemisea herba alba 116±4 113±4 105±4 97±3* 90±3** Saline 125 ±1 128.5±1 129.5±4 128±3 128±3 Artemisea herba.alba Alloxan(Solvent) 452.5±4 459±3 460.5±3 462±3 457±2 Artemisea herba alba Alloxan(diabetic) 471±7 456±7 437±8 416±10* 399±9* 2. Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 Statistically different between pretreatment and post treatment values: p<0.01; p< 0.001 Anti-diabetic and Hypoglycemic Study Crataegus azarolus 104±4 109±4 108±5 111±4 110±5 Solvet 104±2 108±3 113±2 106±4 110±4 3. Allium sativum 93±5 99±9 99±7 96 ± 5 95±4 Solvent 84±4 88±4 82±4 90± 6 89±3 4. Myrtus communis 100±3 107±3 115±2 98±3 98±2 Solvent 95±2 105±3 99±2 90±3 92±2 Myrtus communis 82±2 132±2 152±6 150±2 145±5 On glucose load Solvent 81±2 131±2 117±2 91±3 95± 4 5. Prosopis farcta 121±7 217±19 221±14 233±25 244±24* Solvent 116±4 127 ±7 120±8 120±8 121±8 6. Centaurea phylloceph 95±6 115±4 125±5 99±5 100±6 -ala Solvent 100±8 101±6 108±8 107±5 105±6 Centaurea Phylloceph -ala + glucose load 111±5 121±4 117±4 109±5 110±6 solvent + glucose load 116 ±4 120±7 119±6 120±6 121±4 7. Centaurea behen 105 ±6 139±10 123±12 106±5 115±7 Solvent 107 ±6 106±4 108±5 109±6 110±5 8. Achillea santolina 115 ±5 160±3 159±5 112±5 115±4 Solvent 107 ±3 104±6 102±6 107±9 109±6 9. Achillea micrantha 106 ±2 121±4 131±3 116±4 111±5 Statistically different between pretreatment and post treatment values: p<0.01; Table 2: Effect of the Aqueous Extract of Various Types of Plant on the Plasma Level of glucose (mg/dL) in Normal, Glucose Load and Alloxan Diabetic Rats in Comparison with their Vehicle Solvent (Saline) in Control Rats (Values are Mean±sem) Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 6 6 Table 3: Effect of the Orally Given Aqueous Extracts of Tested Herbs on Blood Glucose Concentration, before and after an Oral Glucose Load (4ml/kg of 50% w/v) in Comparison with their Negative (DW. Water) and Positive (Glyburide 5mg/kg) Controls when Given to Fasted Rats. Values are Mean ±SEM, N=9. (A) Ditrichia Graveolens, Alkanna Strigosa, Phagnalon Rupestre, Sinapis Arvensis, Centaurea Iberica Name of plants And their controls Blood glucose concentration mg/dL Pre-Treatment and Time of drug administration After treatment and time of glucose load administration After glucose administration - 30 min 0 time 30 min 60 min 90 min 120 min 150 min I. a. Ditrichia graveolens 90.13±1.47 101.75 ±1.51 163.5 ±1.96 138.75 ±1.86 107.63 ±0.96 113.0 ±1.09 105.25 ±0.98 b .D. Water 89.38±1.53 99.88 ±1.86 163.95 ±1.55 129.63 ±2.47 112.13 ±1.62 104.38 ±1.43 91.5 ±0.73 c. Glibenclamide 91.25±1.75 101.38 ±2.39 155.63 ±2.68 126.13 ±2.59 94.75 ±1.95 73.13 ±1.00 65.38 ±1.89 II .a.Alkanna Strigosa 91.5±1.43 100.38 ±1.15 164.38 ±2.01 139.75 ±0.80 119.63 ±0.63 107.0 ±1.16 99.0 ±0.94 b.D Water 94.63±1.56 101.25±1.0 155.88 ±1.16 130.25 ±1.03 113.63±1.6 104.5±0.87 92.38±0.86 c. Refer to p <0.05, * refer to p<0.01, * refer to p<0.001 Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 Anti-diabetic and Hypoglycemic Study Glibenclamide 100.0±1.85 154.88 ±1.63 121.5±1.4 97.0±2.16 74.38±2.00 64.00 ±1.43 III .a .Phagnalon rupestre 93.75±2.12 99.38±2.05 163.5±5.22 116.25±3.63* 94.63±2.12* 84.88±1.91 80.00±1.44* b. D. Water 90.38±1.75 97.38±1.59 158.75 ±4.42 134.13 ±3.80 116.38±2.4 103.38 ±2.39 94.38±1.68 c. Glibenclamide 91.50±1.93 102.78 ±2.47 157.75 ±3.82 118.0±2.69 91.63±1.84 68.75±1.15 61.50 ±1.94 IV a. Sinapis arvensis 91.38±1.31 104.88±1.9 178.63 ±2.28 145.±5.06 116.63 ±2.23 104.88 ±0.83 88.5±1.96* b. D. Water 92.5±0.98 100.5±2.15 170.88 ±3.26 137.5 ±1.88 123.25 ±1.93 106.75±1.3 94.38±1.79 c. Glibenclamide 91.88±1.26 103.88 ±1.77 154.25 ±3.04 107.88 ±2.54 84.63 ±2.53 68.75±1.15 59.38±2.47 V.a. Centaurea Iberica 91.75 ±1.37 103.00 ±3.04 172.75 ±4.75 149.25 ±2.52 138.38 ±3.42 124.25 ±2.48 111.00±2.5 b. D. Water 91.88±1.49 97.13±1.49 160.5 ±5.19 138.5 ±3.65 121.75 ±2.55 108.75 ±3.15 98.13±2.08 c. Glibenclamide 93.13±1.77 102.63 ±2.23 155.25 ±5.20 117.38 ±5.71 89.63±3.00 65.75±1.74 58.75±1.69 Refer to p <0.05, * refer to p<0.01, * refer to p<0.001 After glucose administration Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 7 JMED Research (B) Asphodeline Lutea, Eryngium Creticum, Paronychia Argentea and Rosmarinus Officinalis Name of plants And their controls Blood glucose concentration mg/dL Pre-Treatment and Time of drug administration After treatment and time of glucose load administration After glucose administration - 30 min O time 30 min 60 min 90 min 120 min 150 min VI. a. Asphodelin Lutea 92±2.71 102.13±1.57 160.13±5.3 147.5±4.07 124.75±3.51 114.75 ±2.44 103.38 ±2.97 b .D. Water 88.13±1.16 103.25±1.59 147.13 ±2.85 147.13 ±3.99 123.38±2.38 106.25 ±1.58 94.75±0.84 c. Glibenclamide 91.88±1.30 100.25±1.58 158.85 ±4.42 117.75 ±1.55 96.00±2.83 64.88±3.65 ND VII .a.Eryngium certicum 92.0±1.25 106.5±2.20 152.25±3.37 115.88 ±1.58 * 101.13 ±2.48 * 91.00 ±2.15 * 86.13 ±1.30 *** b.D Water 91.38±1.58 102.63±2.02 165.88 ±4.87 143.38 ±4.21 121.63±3.05 106.5±1.91 95.75±1.25 c. Glibenclamide 90.75±1.71 99.50±2.35 157.25 ±4.92 122.63 ±2.79 95.75±1.68 71.00±1.95 ND VIII .a Paronychia argentea 91.25±1.35 105.75±1.78 171.38 ±4.86 137.75±3.41 133.63±2.43 118.63 ±4.06 114.88 ±3.25 b. D. Water 88.13±1.56 103.25±1.28 172.5±3.09 140.13 ±2.64 125.63±0.94 101.0±1.07 91.25±0.90 c. Glibenclamide 88.13±1.53 99.63±2.27 152±3.15 115.0±2.01 91.13±1.79 74.25±1.75 65.38±2.23 IX a. Rosmarinus officinslis 94.5±1.08 103.25±1.08 162.25 ±3.08 128.75±2.3 103.13±2.17 100.88 ±2.57 100.88 ±2.57 b. D. Water 92.38±0.91 102.±1.21 185.86 ±3.56 147.38 ±2.96 111.5±2.39 101.75±1.8 111.13±2.4 0 c. Glibenclamide 92.25±1.10 101.25±1.16 156.38±1.92 122.5±1.65 97.88±1.29 67.38±1.64 ND Significance levels: * p <0.05, refer to p<0.01, * refer to p<0.001. ND= not done (B) Asphodeline Lutea, Eryngium Creticum, Paronychia Argentea and Rosmarinus Officinalis (B) extract produced 85% healing activity of the ulcers. Anti-diabetic and Hypoglycemic Study These results were statistically significant when compared with the saline and cimetidine in control animals (Twaij and Al-Badr, 1988). See figure 1. Anti-Ulcer Study The aqueous extract of Teucrium polium (150mg/kg IP) produced 50% healing of ulcers while the oral administration of the Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 8 8 Figure 1. Effect of Teucrium Polium Aqueous Extract Given Orally at 150mg/kg (TPO), and Intraperitonealy (TIP at 150mg/kg) in Comparison with the Effect of Proglumide (135mg/kg Subcutaneously: PRG). Control Animals Received Saline Orally (SPO) or Saline Intraperitonealy (SIP). Values in Brackets Represent Number of Animals and Data Represents mean±sem. *p<0.001 TPO versus TIP or PRG. p=0.015 TIP versus PRG. 0 10 20 30 40 50 60 70 80 90 100 TPO (n=24) TIP (n=24) PRG (n=12) SPO (n=12) SIP (n=12) Anti-ulcer activity % Healing * 0 10 20 30 40 50 60 70 80 90 100 TPO (n=24) TIP (n=24) PRG (n=12) SPO (n=12) SIP (n=12) Anti-ulcer activity % Healing * Anti-ulcer activity Figure 1. Effect of Teucrium Polium Aqueous Extract Given Orally at 150mg/kg (TPO), and Intraperitonealy (TIP at 150mg/kg) in Comparison with the Effect of Proglumide (135mg/kg Subcutaneously: PRG). Control Animals Received Saline Orally (SPO) or Saline Intraperitonealy (SIP). Values in Brackets Represent Number of Animals and Data Represents mean±sem. *p<0.001 TPO versus TIP or PRG. p=0.015 TIP versus PRG. Discussion plants from different sources including some molecular docking studies on three plants to authenticate their affinity and therapeutic efficacy. Our laboratory conduct studies to explore the anti-diabetic efficacy of herbs and plant parts mainly due to 3 reasons: 1) to investigate claims for several herbal and plant extracts that exist in the middle east and used by traditional healers to treat diabetes; 2) to identify the active ingredients that are responsible for the anti-diabetic effect; 3) to compare the plants in the area with those used in other parts of the world. In the last few decades, there was a renewed interest in herbal medicines and several potential anti- diabetic plants have been studied to identify a wide array of chemically derived plant compounds for their possible diabetes therapy. Usually natural extracts can provide marked pharmacological actions with no or rare side effects if they were taken in the specified doses. Recently, Saravanamuttu and Sudarsanam (2012) reviewed anti-diabetic The present investigative study showed that the aqueous extracts of the aerial parts of Artemisia herba alba, phangnalon rupestre and Eryngium creticum produced significant hypoglycemic as well as antidiabetic activities in experimental animals while all the other plant extracts investigated failed to produce any hypoglycemic effect. In fact, some of the plant extracts produced significantly hyperglycemic effects following a glucose load such as myrtus communis and prosopis farcta (Twaij et al, 1987b; Twaij et al, 1989). Therefore, the later plant extracts results contrasted with the claimed hypoglycemic benefit of these as recommended by the local herbalists in folk medicine (Twaij et al, 1988a). Aqueous Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 9 Al-Shamaony, L., Al-Khazraji, S. M. & Twaij, H. A. A. (1994). “Hypoglycaemic Effect of Artemisia Herba Alba. II. Effect of a Valuable Extract on Some Blood Parameters in Diabetic Animals,” Journal of Ethnopharmacology 43: 167-171 extracts and not organic solvent or other extracts are used in all of our studies. We think that this could be the cause behind their claimed beneficial application in herbal medicine and in addition, it seemed that the aqueous extracts had minimum or undiscovered adverse effects. Farnsworth, R. & Segelman, A. B. (1971). 'Hypoglycemic Plants,' Volume 57, Tile & Till: 52-56. We have also demonstrated in this paper that the T. Polium aqueous extract had a healing effect on ulcers induced by reserpine in rats. Discussion This healing effect was more obvious after oral administration of the extract. The mechanism underlying this effect is not evident from the data available. Further investigations are needed, such as studying its effect on serum calcium, gastric acid secretion and intestinal motility. The extracts of other plants were briefly tested for anti- ulcer activity but none in our laboratory produced a significant effect. Future studies would now be justified trying to identify and isolate the active constituents responsible for the anti-diabetic and anti-ulcer activities. Lewis, H. W. & Elvin-Lewis, M. P. H. (1977). 'Plant Botany: Plants Affecting Man’s Health,' john Wiley and Sons, New York, pp.1, 36, 98, 218, 515. Litchfield, Jr., J. T. & Wilcoxon, F. (1949). “A Simplified Method of Evaluating Dose-Effect Experiments,” Journal of. Pharmacology and Experimental Therapeutics 96; 99-113. Said, M. (1969). 'Hamdard Pharmacopoeia of Eastern Medicine,' Hamadard National Foundation, Time press, Karachi, Pakistan, page 42. Acknowledgements Saravanamuttu, S. & Sudarsanam, D. (2012). “Antidiabetic Plants and their Active Ingredients: A Review,” International Journal of Pharmaceutical Sciences & Research 3(10); 3639-3650. The authors would like to thanks the staff of the Pharmacology Department and animal house at the Faculty of Pharmacy, Philadelphia University for their help and support in providing the lab space to conduct the experiments. The authors wish to declare no conflict of interest. Twaij, H. A. A. & Al-badr, A. A. (1988). “Hypoglycemic Activity of Artemisia Herba Alba,” Elsevier Scientific Publishers Ireland. Journal of Ethnopharmacology 24: 123-126. Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 References A-Jaleel, H. & Twaij, H. (2002). 'Screening for a Possible Hypoglycemic Activity of Selected Jordanian Medicinal Plants,' Part I. Jordanian Journal for Applied Science,4: 1-7 Twaij, H. A. A., Al-badr, A. A. & Abul-Khail, A. (1987a). “Anti-Ulcer Activity of Teucrium Polium,” International Journal of Crude Drug Research 25: 125-128. Twaij, H. A. A. & A-jaleel, H. (2002). 'Screening for a Possible Hypoglycemic Activity of Selected Jordanian Medicinal Plants,' Part II. Jordanian Journal for Applied Science, 4: 16-21. AI-Khazarji, S. M., AI-Shamony, L. A. & Twaij, H. A. A. (1993). “Hypoglycaemic Effect of. Artemisia Herba-Alba. Effect of Different Parts and Influence of the Solvent on Hypoglycaemic Activity,” Journal of Ethnopharmacology. 40: 163–166. Husni A Twaij and Emad A S Al-Dujaili (2014), JMED Research, DOI: 10.5171/2014.539605 JMED Research 10 Twaij, H. A. A., Kerry, A., Al-jebory, A. A. & Hammad, M. N. (1987b). 'Crataegus Azarolus Linn.Pharmacology and Phytochemistry,' Part I. Pharmacology J. Bio. Sci. Res, 18,105- 125. Twaij, H. A. A., Kerry, A. & Al-Khazraji, S. M. (1983). 'Some Pharmacological, Toxicological and Phytochemical Investigations on Centaurea Phylloceph,' Journal of Ethnopharmacology, 9 (1983) 29914 299 Twaij, H. A. A., Mahmoud, S. N., & Khalid, R. M. (1989). “Screening of Some Iraqi Medicinal Plants for their Molluscicidal Activities,” Fitoterapia 1989 Vol. 60 No. 3 pp. 267-268 Twaij, H. A. A., Sayed-Ali, H. M. & AlZohry, A. M. (1988a). “Pharmacological Phytochemical and Antimicrobial Studies on Myrtus Communis Part I. Cardiovascular and Phytochemical Studies,” Journal of Biological Sciences Research, 19(1): 29-40. Twaij, H. A. A., Sayed-Ali, H. M. & AlZohry, A. M. (1988b). "Pharmacological Phytochemical and Antimicrobial Studies on Myrtus Communis Part 2. Glycemic and Antimicrobial Studies," Journal of Biological Sciences Research, 19(1): 41-52.
https://openalex.org/W3100909419	https://link.springer.com/content/pdf/10.1007%2FJHEP02%282016%29165.pdf	English	null	Three-point functions in N = 4 $$ \mathcal{N}=4 $$ SYM: the hexagon proposal at three loops	The Journal of high energy physics/The journal of high energy physics	2,016	cc-by	9,110	Open Access, c⃝The Authors. Article funded by SCOAP3. Published for SISSA by Springer Received: December 21, 2015 Accepted: February 3, 2016 Published: February 24, 2016 Received: December 21, 2015 Accepted: February 3, 2016 Published: February 24, 2016 1See refs. [6–10] for reviews and a list of references. Three-point functions in N = 4 SYM: the hexagon proposal at three loops JHEP02(2016)165 Burkhard Edena and Alessandro Sfondrinib aInstitut f¨ur Mathematik & Institut f¨ur Physik, Humboldt-Universit¨at zu Berlin, Zum großen Windkanal 6, D-12489 Berlin, Germany bInstitut f¨ur Theoretische Physik, ETH Z¨urich, Wolfgang-Pauli-Str. 27, CH-8093 Z¨urich, Switzerland E-mail: eden@math.hu-berlin.de, sfondria@itp.phys.ethz.ch Abstract: Basso, Komatsu and Vieira recently proposed an all-loop framework for the computation of three-point functions of single-trace operators of N = 4 super-Yang-Mills, the “hexagon program”. This proposal results in several remarkable predictions, including the three-point function of two protected operators with an unprotected one in the SU(2) and SL(2) sectors. Such predictions consist of an “asymptotic” part — similar in spirit to the asymptotic Bethe Ansatz of Beisert and Staudacher for two-point functions — as well as additional ﬁnite-size “wrapping” L¨uscher-like corrections. The focus of this paper is on such wrapping corrections, which we compute at three-loops in the SL(2) sector. The resulting structure constants perfectly match the ones obtained in the literature from four-point correlators of protected operators. Keywords: AdS-CFT Correspondence, Integrable Field Theories, Supersymmetric gauge theory ArXiv ePrint: 1510.01242 ArXiv ePrint: 1510.01242 ArXiv ePrint: 1510.01242 Open Access, c⃝The Authors. Article funded by SCOAP3. doi:10.1007/JHEP02(2016)165 doi:10.1007/JHEP02(2016)165 Contents 1 Introduction 1 2 The hexagon proposal 2 2.1 Asymptotic three-point function 3 2.2 Wrapping eﬀects 4 3 Three-point functions at three loops 5 3.1 Wrapping in the opposite channel 6 3.2 Wrapping in the adjacent channels 7 4 Results, conclusions and outlook 8 A Telescoping the transfer matrix 10 B Evaluation of the opposite-channel wrapping 10 B.1 Corrections to the measure 11 B.2 Corrections to the transfer matrix 11 B.3 Measure and residues from the ﬁrst order mirror dressing phase 12 C Evaluation of the adjacent-channel wrapping 13 Contents 1 Introduction 1 2 The hexagon proposal 2 2.1 Asymptotic three-point function 3 2.2 Wrapping eﬀects 4 3 Three-point functions at three loops 5 3.1 Wrapping in the opposite channel 6 3.2 Wrapping in the adjacent channels 7 4 Results, conclusions and outlook 8 A Telescoping the transfer matrix 10 B Evaluation of the opposite-channel wrapping 10 B.1 Corrections to the measure 11 B.2 Corrections to the transfer matrix 11 B.3 Measure and residues from the ﬁrst order mirror dressing phase 12 C Evaluation of the adjacent-channel wrapping 13 1 2 3 4 5 6 7 8 10 10 11 11 12 13 JHEP02(2016)165 C Evaluation of the adjacent-channel wrapping 1 Introduction A remarkable advance in the study of holographic [1], or gauge/string correspondence [2–4] was the discovery of integrability in the planar limit [5] of the correspondence: the duality’s dynamics is severely constrained by inﬁnitely many hidden symmetries. These powerful symmetries can be used to compute the energies of string states, or equivalently the two- point functions of operators of the dual CFT. This was ﬁrstly done for the most supersym- metric case of AdS5/CFT4, and more recently for other less supersymmetric dual pairs.1 The missing ingredient to fully describe generic (non-protected) states in the planar limit through integrability is to exploit symmetries to compute three-point functions. A great deal of eﬀort has been devoted to this problem in AdS5/CFT4, both from the point of view of string theory and of gauge theory [11–39]. Recently, a crucial development was the proposal from Basso, Komatsu and Vieira (BKV) of an all-loop framework for the computation of three-point functions using in- tegrability: the hexagon program [34]. This framework automatically incorporates the weakly-coupled “tailoring” procedure [16–18, 21], but it is “all-loop” in nature. It is in fact – 1 – a generalisation of the integrable bootstrap to three-point functions, with a new fundamen- tal object — the hexagon amplitude — playing the role of the scattering matrix. This can be quite easily used to construct the asymptotic part of the structure constant, in the sense of the asymptotic Bethe Ansatz of Beisert and Staudacher for two-point functions [40]. For short operators, corrections due to wrapping eﬀects similar to those familiar from the spectral problem [41] should be added. BKV propose an explicit recipe for doing so in a manner reminiscent of L¨uscher corrections [42, 43]. In ref. [34] BKV put their proposal to several tests, including the direct comparison of certain structure constants against known weak-coupling [44, 45] and strong-coupling [31] results. At weak-coupling, BKV explicitly compute the two-loop, three-point function of one non-BPS operator with two BPS ones, in the SU(2) and SL(2) sectors. This is matched to the ﬁeld theory results, which have been independently obtained from computing the four-point correlators of BPS operators [44–51]. Since the operators appearing in the three- point function are very short, this two-loop calculation already probes the ﬁrst “wrapping” correction to the BKV asymptotic formula — the ones coming from the edge of the hexagon opposite to the non-BPS state, see also ﬁgure 1 below. 1 Introduction JHEP02(2016)165 The aim of this paper is to apply the hexagon approach to compute three-loop three- point functions in the SL(2) sector. This calculation will for the ﬁrst time simultaneously probe the wrapping corrections on all the hexagon edges that have a “mirror” kinemat- ics. What is more, a prediction for the structure constant of these operator is known in the literature [45]. This yields a further quite non-trivial (and successful!) check of the hexagon proposal. This paper is structured as follows. In section 2 we brieﬂy summarise the BKV pro- posal, working out explicitly some formulae which we will need later — namely, the ones for the wrapping corrections in the two “adjacent” channels. In section 3 we specialise these formulae to the aforementioned three-loop computation and discuss how to evaluate them. In section 4 we present our results and conclusions. We relegate the technical details concerning the evaluation of the wrapping corrections to the appendices. Note added. Shortly after the submission of this pre-print we became aware of an up- coming work by Basso, Goncalves, Komatsu and Vieira where, among other things, these three-loop structure constants are computed using the hexagon approach and successfully matched to gauge theory [52]. 2 The hexagon proposal Let us brieﬂy review the BKV proposal [34]. One starts by cutting a three-string interaction (which has the topology of a pair of pants at leading order) into two pieces “along the pants’ seams”. This results into two patches, each having six distinguished edges — three corresponding to the cuts, and three to half of a “cuﬀ” each, see ﬁgure 1. These hexagons are the central objects of the proposal. The hexagon can be decorated with six sets of particles. Along the edges corresponding to “cuﬀs”, we will have closed-string excitations. Along the edges which will be glued back – 2 – u1 u2 v u3 v Figure 1. A stringy interaction with the topology of a pair of pants can be cut into two hexagons by cutting along the pants’ seams. In the ﬁgure the dashed edges (seams) of the two hexagons are glued, while the solid lines represent the cuﬀs of the pants. Closed-string excitations live on the “cuﬀs”: here we depicted excitations with rapidities u1, u2, u3 on the upper cuﬀ. Mirror excitations live on the dashed edges, and are necessary to properly account for wrapping processes. Here we depict one mirror excitation v on the mirror edge opposite to the excited string edge. Generically, the rapidities of each closed string will be the Bethe roots of the corresponding state, while one should integrate over all possible mirror rapidities. u1 u2 v u3 v Figure 1. A stringy interaction with the topology of a pair of pants can be cut into two hexagons by cutting along the pants’ seams. In the ﬁgure the dashed edges (seams) of the two hexagons are glued, while the solid lines represent the cuﬀs of the pants. Closed-string excitations live on the “cuﬀs”: here we depicted excitations with rapidities u1, u2, u3 on the upper cuﬀ. Mirror excitations live on the dashed edges, and are necessary to properly account for wrapping processes. Here we depict one mirror excitation v on the mirror edge opposite to the excited string edge. Generically, the rapidities of each closed string will be the Bethe roots of the corresponding state, while one should integrate over all possible mirror rapidities. JHEP02(2016)165 to yield the three-point functions, we will have excitations in the “mirror” kinematics, similarly to what happens when computing ﬁnite-size corrections to the energy of string states. 2 The hexagon proposal Given a three-point function by specifying the physical closed-string excitations at each of its cuﬀs, we can compute the relative structure constant by summing hexagon amplitudes over all possible ways of distributing the physical excitations over the two pieces of the cuﬀs, as well as summing over all possible mirror states and integrating over the mirror rapidities. Cleverly using the SU(2\|2) super-symmetry of the hexagon, as well the crossing trans- formation and imposing scattering factorisation `a la Zamolodchikov [53], the hexagon am- plitude h was ﬁxed exactly in [34], at least up to a scalar factor h(x, y). This is constrained by crossing symmetry to satisfy h(1/x, y) h(x, y) = c(x, y) , c(x, y) = x−−y− x−−y+ 1 −1/x+y− 1 −1/x+y+ , (2.1) (2.1) where we crossed the Zhukovski variables as x± →1/x±, corresponding to crossing the rapidity u →u2γ [34]. The last ingredient of the BKV proposal is then to set h(x, y) = x−−y− x−−y+ 1 −1/x−y+ 1 −1/x+y+ 1 σ(x, y) , (2.2) (2.2) where σ is the dressing factor of Beisert, Eden and Staudacher [54]. where σ is the dressing factor of Beisert, Eden and Staudacher [54]. 2.1 Asymptotic three-point function Using these ingredients, BKV predict the asymptotic part of the three-point function of two protected and one non-protected operator in the SL(2) sector to be C•◦◦ 123 C◦◦◦ 123 2 = QS k=1 µ(uk) det∂ujφk Q j<k Sjk   X α∪¯α={u} A(α,¯α)   2 . (2.3) (2.3) – 3 – – 3 – Here Sjk is the SL(2) diagonal scattering element of the S matrix by Beisert [55], µ is a measure deﬁned by the residue of the pole in the transition hD\|D of an excitation from one physical edge to another [34], Here Sjk is the SL(2) diagonal scattering element of the S matrix by Beisert [55], µ is a measure deﬁned by the residue of the pole in the transition hD\|D of an excitation from one physical edge to another [34], µ(u) = i resv=u hD\|D(u\|v) −1 , (2.4) (2.4) and the determinant is the Gaudin norm deﬁned in terms of φj which satisﬁes and the determinant is the Gaudin norm deﬁned in terms of φj which satisﬁes eiφj = eipjL Y k̸=j Sjk . (2.5) (2.5) JHEP02(2016)165 Finally, we should sum over the partitions α and ¯α the expression A(α,¯α) = (−1)\|¯α\| Y j<k j,k∈α∪¯α hjk Y k∈¯α eipkℓY j∈α k∈¯α 1 hjk . (2.6) (2.6) Here ℓis the separation between the non-protected operator and the others. Denoting the length of the non-protected operator as L = L1, we have ℓ= ℓ12 = ℓ31 with ℓij = 1 2(Li + Lj −Lk), all indices being distinct. 2.2 Wrapping eﬀects The contributions of the two adjacent channels (the two lateral dashed edges in ﬁgure 1)can be easily found and are given by JHEP02(2016)165 intγ a(u, {ui}) = Aα,¯α(−1)aTa(u−γ) Y j∈α ha(uγ, uj) ca(u−γ, uj) Y j∈¯α ha(uγ, uj) ca(u+γ, uj), (2.12) (2.12) and and int5γ a (u, {ui}) = Aα,¯α(−1)aTa(u−γ) Y j∈α ha(uγ, uj) ca(u+γ, uj) Y j∈¯α ha(uγ, uj) ca(u−γ, uj), (2.13) (2.13) where ca(u, v) can be found from eq. (2.1) by fusion. There are two important diﬀerences between eq. (2.11) and (2.12)–(2.13). Firstly, the contributions of the adjacent channels are sub-leading since Ta(uγ) = O(1) while Ta(u−γ) = O(g2); for this reason, (2.12)–(2.13) did not contribute in the evaluation of the two-loop SL(2) three-point functions in ref. [34]. Secondly, in (2.12)–(2.13) the sum over partitions and the integration over u do not factor, making the evaluation of these contributions somewhat more involved. It is worth noting that at higher-loop level there are additional wrapping eﬀects con- tributing, namely the ones described by ordinary L¨uscher corrections for each of the single- trace operators in the three-point function. Of course in this set-up such corrections only appear for the non-protected operator. In the SL(2) sector, as it is well known [59–61], such wrapping corrections ﬁrst appear at four loops, and therefore go beyond the scope of this work. 2.2 Wrapping eﬀects So far we have not accounted for the presence of mirror particles on the edges of the hexagon to be glued. This can be done in a L¨uscher-like approach, where the leading ﬁnite-volume contribution is given by allowing at most a single mirror particle per edge. Then one has to correct the asymptotic expression by A(α,¯α) →A(α,¯α) + δA(α,¯α) 12 + δA(α,¯α) 23 + δA(α,¯α) 31 . (2.7) (2.7) Each of the δA is related to one of the mirror channels. They are given by δA(α,¯α) jk = X a>0 Z du 2π µγ a(u) 1 x[+a]x[−a] ℓjkint(2j−1)γ a (u\|{ui}). (2.8) (2.8) Here and in what follows a denotes the bound-state number, and x[±a] are the bound-state Zhukovski variables, which depend on the shifted rapidities u ± a 2i. Following ref. [34], we indicate mirror-like shifts of the rapidity, corresponding to shifting an excitation to a neighbouring edge of the hexagon n times, by nγ. The mirror measure is the same in all channels and reads µγ a(u) = a (x[+a]x[−a])2 g2(x[+a]x[−a] −1)2(x[+a]2 −1)(x[−a]2 −1). (2.9) (2.9) The integrand depends on which mirror channel we consider. Schematically intnγ a (u, {ui}) = (−1)\|¯α\| Y j∈¯α eipjℓ Y k>j j∈¯α,k∈α Sjk X Xa (−1)fXahXaD...D(unγ, α)hD...D ¯ Xa(¯α, u−nγ), (2.10) intnγ a (u, {ui}) = (−1)\|¯α\| Y j∈¯α eipjℓ Y k>j j∈¯α,k∈α Sjk (2.10) (2.10) X Xa (−1)fXahXaD...D(unγ, α)hD...D ¯ Xa(¯α, u−nγ), ( ) – 4 – which involves the scattering of the mirror (bound-state) particles Xa with all the physical particles in the α partitions, and similarly for their conjugates ¯ Xa with the ¯α partition. The channel opposite to the non-protected state (the bottom edge of the hexagon in ﬁgure 1), corresponding to a shift of 3γ, has been computed in ref. [34] int3γ a (u, {ui}) = Aα,¯α (−1)aTa(uγ) Q j∈α∪¯α ha(uγ, uj), (2.11) (2.11) where Ta is the transfer matrix in the anti-symmetric representation (see appendix H in ref. [34]), and ha is bound-state scalar factor, which can be found by fusion [56–58]. 3.1 Wrapping in the opposite channel The computation of the opposite (3γ) channel follows the one of ref. [34], but must include next-to-leading order corrections. Firstly, it is useful to massage a bit the transfer matrix Ta(uγ) from ref. [34] — see appendix A for details. Then, we ﬁnd it convenient to strip a denominator out of Ta(uγ), introducing eTa(uγ) = (−1)a Ta(uγ) Qs j=1 dena(uγ, uj) , dena(uγ, uj) = (x[−a] −x+ j )(1 −1/x[−a]x− j ). (3.1) (3.1) The reason for doing so is that, when plugging Ta into (2.11), this denominator will simplify against the rational part of the mirror-string dressing factor as we will see. We need to expand eTa up to O(g2). On top of the tree-level term computed in [34], we have a correction coming from the one-loop rapidities and one coming from the explicit g-dependence of eTa, JHEP02(2016)165 eTa(uγ) = 1 + g2 s X j=1 u(1) i ∂ui eT (0) a (uγ) + g2E(1) eT (1) a (uγ) + O(g4). (3.2) (3.2) In terms of the Baxter polynomials Q(u) = Qs j=1(u −uj) we then have2 s of the Baxter polynomials Q(u) = Q j=1(u −uj) we then have eT (0) a (uγ) = Q(u[a+1]) + Q(u[−a−1]) −Q(u[a−1]) −Q(u[−a+1]), eT (1) a (uγ) = Q(u[−a−1]) u[−a] −Q(u[a+1]) u[+a] + a−1 X k=1 Q(u[2k−1−a]) −Q(u[2k+1−a]) u[2k−a] . (3.3) (3.3) Note that the one-loop transfer matrix is multiplied by the one-loop energy E(1) = s X j=1 −i v2 j + 1/4 = −2iS1(s) , (3.4) (3.4) where the last expression is the well-known representation of the one-loop, spin-s energy as a harmonic sum [59]. In (3.3) one might worry that eT (1) a seems to have a pole on the real-line when 2k = a. However, it is easy to see that the summand is regular at u = 0 when one imposes the zero-momentum or level-matching condition Q(+i/2) = Q(−i/2). As we mentioned, the denominator dena simpliﬁes drastically against the rational part of the dressing factor ha. In fact, for the purpose of our calculation we can write dena(uγ, v)ha(uγ, v) ≈1 x+ v 1 + ig2E(1)Ψ(u) , Ψa(u) = 4γ + ψ 1 + iu[−a] + ψ 1 −iu[−a] + ψ 1 + iu[+a] + ψ 1 −iu[+a] , (3.5) (3.5) where the digamma functions ψ come from the expansion of the mirror-string dressing factor [58, 62]. 3 Three-point functions at three loops Using these ingredients, we can now compute three-loop three-point functions involving twist-two, spin-s operators. Speciﬁcally, we pick one twist-two (non-protected) operator O1 = tr(DsZ2), and two protected operators O2 = tr( ¯ZY ) and O3 = tr( ¯Z ¯Y ). This will probe the wrapping contributions to the adjacent channels at γ, 5γ, as well as the next- to-leading-order contribution of the opposite (3γ) channel. Of course the ﬁnal result will also depend on the three-loop expansion of the asymptotic term described in section 2.1, as well as on the loop corrections to the rapidities from the Bethe Ansatz. Accounting for these corrections is straightforward, and we will therefore focus our attention on the wrapping eﬀects. – 5 – 2The Zhukovski variables x± are particularly convenient when discussing crossing transformations, but for our perturbative computations it is convenient to expand them as customary x± → √ 2/g u± + . . .. 3.1 Wrapping in the opposite channel The last ingredient for the evaluation of the opposite channel contribution is the prod- uct of the mirror measure µ(uγ) and of the mirror energy 1/x[+a]x[−a] from eq. (2.8). Since in this case ℓ= 1, we ﬁnd that this is g4 a (u[+a]u[−a])3 1 + g2 2 (u[+a])2 + 1 u[+a]u[−a] + 2 (u[−a])2 + . . . . (3.6) (3.6) 2The Zhukovski variables x± are particularly convenient when discussing crossing transformations, but for our perturbative computations it is convenient to expand them as customary x± → √ 2/g u± + . . .. – 6 – In order to evaluate this expression we have to perform the integration over u, sum over the bound-state number a and an additional sum from eq. (3.3). To this end, it is necessary to massage the expressions we have found, in a way reminiscent of [34]. We comment on the necessary manipulations in appendix B. Eventually, we obtain an explicit expression for the integrand to be integrated over the real line, as well as some residues due to manipulations of the integration contours. However, just as it was the case in [34], the integral cannot be evaluated analytically — to our knowledge. Numeric integration is however possible to good accuracy, even if some care is necessary in the estimate of the numerical errors.3 Nonetheless we could always reduce the result to rational numbers and ζ3, ζ5 by the pslq algorithm [63]. It is interesting to note that the residues subtract all terms involving even ζ values, while sometimes not contributing to the pure ζ3 or ζ5. Remarkably, the integral arising from corrections to mirror dressing phase (see appendix B.3) is apparently always purely rational once the residues are extracted. JHEP02(2016)165 3In particular, we have encountered some issues in this sense when using the NIntegrate algorithms of Wolfram Mathematica. 3.2 Wrapping in the adjacent channels Let us proceed as in the previous subsection. We ﬁrstly deﬁne eTa(u−γ) = (−1)a Ta(u−γ) Qs j=1 dena(u−γ, uj) , dena(u−γ, uj) = (x[−a] −x− j )(1 −1/x[−a]x+ j ). (3.7) (3.7) xpanding eTa(u−γ), we ﬁnd that the ﬁrst non-vanishing contribution appears at order g2: eTa(u−γ) = g2E(1) " −Q(u[−a+1]) u[−a] + Q(u[a−1]) u[+a] + a−1 X k=1 Q(u[2k−1−a]) −Q(u[2k+1−a]) u[2k−a] # . (3.8) As we have discussed, the remaining expressions in eqs. (2.12)–(2.13) contain terms which are both u-dependent and partition dependent. Let us focus on one channel, say 1γ (2.12), and introduce the incomplete Baxter polynomial Qα(u) = Q j∈α(u −uj). Using that Q(u) = Qα(u)Q¯α(u) we can e.g. eliminate the products over the ¯α partition. The price is to introduce a product over the α partition As we have discussed, the remaining expressions in eqs. (2.12)–(2.13) contain terms which are both u-dependent and partition dependent. Let us focus on one channel, say 1γ (2.12), and introduce the incomplete Baxter polynomial Qα(u) = Q j∈α(u −uj). Using that Q(u) = Qα(u)Q¯α(u) we can e.g. eliminate the products over the ¯α partition. The price is to introduce a product over the α partition Mα =  Y j∈α e−2ipj  Qα(u[a−1])Qα(u[−a−1]) Qα(u[a+1])Qα(u[−a+1]), (3.9) (3.9) which modiﬁes the partition dependent term so that it diﬀers from the asymptotic one A = P α∪¯α Aα,¯α. Instead we now have e A = X α∪¯α Aα,¯αMα, (3.10) (3.10) which depends on a and u. Clearly the 5γ channel can be found by exchanging α ↔¯α. As it turns out, e A is identical in the two cases. which depends on a and u. Clearly the 5γ channel can be found by exchanging α ↔¯α. As it turns out, e A is identical in the two cases. 3In particular, we have encountered some issues in this sense when using the NIntegrate algorithms of Wolfram Mathematica. – 7 – Given that we are interested in integrating this expression, it is convenient to make all possible poles manifest. To this end, we introduce a function eQ(u, a, s) that satisﬁes e A Q(u[a+1]) Q(u[−a+1]) = A eQ(u, a, s). (3.11) (3.11) By construction, eQ(u, a, s) is a polynomial in u, i.e. it has no poles. Such a polynomial can be explicitly evaluated for every given spin s, see appendix C. 4 Results, conclusions and outlook Summing up the various contributions described in the last two sections we ﬁnally obtain the structure constants of two half-BPS operators of length-2 going into a twist-2 operator: s C•◦◦ C◦◦◦ 2 for twist L = 2, bridge ℓ= 1 and spin s 2 1 6 −g2 + (7 + 3ζ3)g4 −(48 + 8ζ3 + 25ζ5)g6 + . . . 4 1 70 −205 1764g2 + 76393 74088 + 5 14ζ3 g4 − 242613655 28005264 + 1315 1323ζ3 + 125 42 ζ5 g6 + . . . 6 1 924 − 553 54450g2 + 880821373 8624880000 + 7 220ζ3 g4 − 1364275757197 1423105200000 + 520093 6534000ζ3 + 35 132ζ5 g6 + . . . 8 1 12870 − 14380057 18036018000g2 + 5944825782678337 682443241880400000 + 761 300300ζ3 g4 − 758072803634287465765957 8607383632540733040000000 + 15248925343 2840672835000ζ3 + 761 36036ζ5 g6 + . . . 10 1 184756 − 3313402433 55983859495200g2 + 171050793565932326659 248804677619932936320000 + 671 3527160ζ3 g4 − 9135036882706194334305789554347 1243961012766985364412864576000000 + 11482697774339 35269831481976000ζ3 + 3355 2116296ζ5 g6 + . . . s C•◦◦ C◦◦◦ 2 for twist L = 2, bridge ℓ= 1 and spin s 2 1 6 −g2 + (7 + 3ζ3)g4 −(48 + 8ζ3 + 25ζ5)g6 + . . . 4 1 70 −205 1764g2 + 76393 74088 + 5 14ζ3 g4 − 242613655 28005264 + 1315 1323ζ3 + 125 42 ζ5 g6 + . . . 6 1 924 − 553 54450g2 + 880821373 8624880000 + 7 220ζ3 g4 − 1364275757197 1423105200000 + 520093 6534000ζ3 + 35 132ζ5 g6 + . . . 8 1 12870 − 14380057 18036018000g2 + 5944825782678337 682443241880400000 + 761 300300ζ3 g4 − 758072803634287465765957 8607383632540733040000000 + 15248925343 2840672835000ζ3 + 761 36036ζ5 g6 + . . . 10 1 184756 − 3313402433 55983859495200g2 + 171050793565932326659 248804677619932936320000 + 671 3527160ζ3 g4 − 9135036882706194334305789554347 1243961012766985364412864576000000 + 11482697774339 35269831481976000ζ3 + 3355 2116296ζ5 g6 + . . . The O(g6) values in the table are in exact agreement with the conformal partial-wave analysis of the four-point function of stress energy tensor multiplets [44, 51]. This test of the hexagon conjecture [34] was the main motivation for our work. 3.2 Wrapping in the adjacent channels Using the expansion of the remaining terms in (2.12) we ﬁnd that the integrand takes the form A Q(i/2) eQ(u, a, s) Q(u[a+1])Q(u[−a−1])Q(u[a−1])Q(u[−a+1]) Ta(u−γ). (3.12) (3.12) JHEP02(2016)165 Even if eQ(u, a, s) in general is not a Baxter polynomial, it has degree s. It follows that the integrand decays fast for large \|u\|, and can be evaluated by residues. We further comment on the related technicalities in appendix C. Even if eQ(u, a, s) in general is not a Baxter polynomial, it has degree s. It follows that the integrand decays fast for large \|u\|, and can be evaluated by residues. We further comment on the related technicalities in appendix C. 4 Results, conclusions and outlook s C•◦◦ C◦◦◦ 2 for twist L = 2, bridge ℓ12 = ℓ31 = 1, ℓ23 > 1 and spin s 2 1 6 −g2 + 7g4 + (10ζ5η −10ζ5 + 7ζ3 −48)g6 + . . . 4 1 70 −205 1764g2 + 36653 37044g4 + 1 6ζ3 + 25 21ζ5 η −25 21ζ5 + 193 216ζ3 −442765625 56010528 g6 + . . . 6 1 924 − 553 54450g2 + 826643623 8624880000g4 + − 1 1440 + 7 264ζ3 + 7 66ζ5 η −7 66ζ5 + 24143 297000ζ3 −1183056555847 1423105200000 g6 + . . . 8 1 12870 − 14380057 18036018000g2 + 2748342985341731 341221620940200000g4 + − 79 604800 + 3 1040ζ3 + 761 90090ζ5 η −761 90090ζ5 + 1039202363 158918760000ζ3 −1270649655622342732745039 17214767265081466080000000 g6 + . . . 10 1 184756 − 3313402433 55983859495200g2 + 156422034186391633909 248804677619932936320000g4 + − 45071 2813045760 + 781 2930256ζ3+ 671 1058148ζ5 η − 671 1058148ζ5 + 8295615163 16799157648000ζ3 − 7465848687069712820911408164847 1243961012766985364412864576000000 g6 + . . . s C•◦◦ C◦◦◦ 2 for twist L = 2, bridge ℓ12 = ℓ31 = 1, ℓ23 > 1 and spin s 2 1 6 −g2 + 7g4 + (10ζ5η −10ζ5 + 7ζ3 −48)g6 + . . . 4 1 70 −205 1764g2 + 36653 37044g4 + 1 6ζ3 + 25 21ζ5 η −25 21ζ5 + 193 216ζ3 −442765625 56010528 g6 + . . . 6 1 924 − 553 54450g2 + 826643623 8624880000g4 + − 1 1440 + 7 264ζ3 + 7 66ζ5 η −7 66ζ5 + 24143 297000ζ3 −1183056555847 1423105200000 g6 + . . . 8 1 12870 − 14380057 18036018000g2 + 2748342985341731 341221620940200000g4 + − 79 604800 + 3 1040ζ3 + 761 90090ζ5 η −761 90090ζ5 + 1039202363 158918760000ζ3 −1270649655622342732745039 17214767265081466080000000 g6 + . . . 10 1 184756 − 3313402433 55983859495200g2 + 156422034186391633909 248804677619932936320000g4 + − 45071 2813045760 + 781 2930256ζ3+ 671 1058148ζ5 η − 671 1058148ζ5 + 8295615163 16799157648000ζ3 − 7465848687069712820911408164847 1243961012766985364412864576000000 g6 + . . . JHEP02(2016)165 It would be very interesting to extend this analysis to higher-order corrections. A ﬁrst question is how one can make sense of wrapping corrections to the operators of the three- point functions. While this is a familiar problem in the context of two-point functions, this issue still to be explored in the hexagon program. 4 Results, conclusions and outlook As we mentioned, these eﬀects would ﬁrst appear at order O(g8), and therefore are accessible to gauge-theoretical computations, which would provide another crucial check of the hexagon approach. In the long run one could try to constrain a putative octagon operator [34] for four- point functions by perturbative data. Ideally we obtain a machinery that will directly furnish non-trivial kinematics, so which will allow us to reach out beyond the computation of sets of constants. Another diﬃcult but very interesting question is whether the hexagon approach can be promoted to a truly non-perturbative formalism. While as we saw L¨uscher-like cor- rections work remarkably well, one should also account account for more than one virtual particle at a time4 within the hexagon approach [34]. Ideally one would hope that a sort of “Thermodynamic Bethe Ansatz” formulation could be constructed for the hexagon. We are conﬁdent to witness remarkable developments in this direction in the near future. 4For advances in this direction in the context of two-point functions see ref. [64]. 4 Results, conclusions and outlook It probes the correctness of the new approach to ﬁnite size corrections already rather deeply, as is well illustrated by the multitude of eﬀects we had to take into account. The O(g6) values in the table are in exact agreement with the conformal partial-wave analysis of the four-point function of stress energy tensor multiplets [44, 51]. This test of the hexagon conjecture [34] was the main motivation for our work. It probes the correctness of the new approach to ﬁnite size corrections already rather deeply, as is well illustrated by the multitude of eﬀects we had to take into account. An obvious extension to this work is to predict structure constants for the fusion of two higher-length single-trace half-BPS operators into twist 2 operators. In that case the bridge length for the adjacent channels would stay put at ℓ12 = ℓ31 = 1, while ℓ23 for the opposite channel would rise. For length-3 operators we have ℓ23 = 2, so that the leading order analysis of [34] now applies to the three-loop correction. As the mirror measure now starts on 1/(u+u−)4 we obtain a modiﬁed eﬀective integration measure ˆµ = 16π3 3(1 + 4u2)4 cosh2(πu) π(−1 + 8u2 + 48u4) cosh2(πu) (4.1) + 48u(−1 + 4u2) tanh(πu) −2π(−1 + 8u2 + 48u4) tanh2(πu) . ˆµ = 16π3 3(1 + 4u2)4 cosh2(πu) π(−1 + 8u2 + 48u4) cosh2(πu) (4.1) (4.1) + 48u(−1 + 4u2) tanh(πu) −2π(−1 + 8u2 + 48u4) tanh2(πu) . + 48u(−1 + 4u2) tanh(πu) −2π(−1 + 8u2 + 48u4) tanh2(πu) . – 8 – – 8 – with the subtraction of residues following the by now standard path. The contribution of the opposite channel for length 3 is marked in the table by the coeﬃcient η, which ought to be put to 1/2 in this case. If the length of the BPS is greater or equal 4, we have bridge length ℓ23 > 2 and the leading contribution in the opposite channel moves out to O(g8) or higher. The approach of [34] then predicts the result in the table below at η = 0, so notably perfect universality of the structure constants up to three loops. We will check what constraints these results can impose on Ans¨atze for higher-charge planar correlation functions [44]. A Telescoping the transfer matrix The transfer matrix in the antisymmetric bound state representation as deﬁned in formula (H1) in ref. [34] can be substantially simpliﬁed when the level-matching condition, or, in gauge theory parlance, the zero momentum condition Q j x+ j /x− j = 1 is satisﬁed. Using JHEP02(2016)165 Y j (x[a] −x∓ j ) 1 − 1 x[a]x∓ j ! = Q(u[a±1]) (A.1) (A.1) it follows that R+(u[a]) B+(u[a]) R−(u[a]) B−(u[a]) = Q(u[a+1]) Q(u[a−1]), (A.2) (A.2) where we followed the notation of [34] R±(u) = Y j (x(u) −x∓ j ) , B±(u) = Y j 1 x(u) −x∓ j . (A.3) (A.3) In every summand in Ta(uγ) and Ta(u−γ) the Q factors “telescope”, so that all terms but the ﬁrst in the denominator and the last in the numerator cancel. In terms of the function eTa deﬁned in (3.1) eTa(uγ) = R−(u[−a])B+(u[−a]) + R+(u[a])B−(u[a]) −2R−(u[a])B−(u[a]) + a−1 X k=1 R+(u[2k−a])B−(u[2k−a]) + R−(u[2k−a])B+(u[2k−a]) −2R−(u[2k−a])B−(u[2k−a]) . (A.4) (A.4) A similar expression for eTa(uγ) can be found immediately by crossing x[±a] →1/x[±a], i.e. by swapping B± ↔R±. A similar expression for eTa(uγ) can be found immediately by crossing x[±a] →1/x[±a], i.e. by swapping B± ↔R±. Acknowledgments We would like to thank G. Arutyunov and S. van Tongeren for discussions. B.E. is sup- ported by the DFG, “eigene Stelle” ED 78/4-2 and acknowledges partial support by the 4For advances in this direction in the context of two-point functions see ref. [64]. We would like to thank G. Arutyunov and S. van Tongeren for discussions. B.E. is sup- ported by the DFG, “eigene Stelle” ED 78/4-2 and acknowledges partial support by the – 9 – Marie Curie network GATIS under REA Grant Agreement No 317089. A.S. would like to thank the group for Mathematical Physics of Space, Time and Matter at Humboldt Uni- versity where part of this research was carried out. A.S.’s research was partially supported by the NCCR SwissMAP, funded by the Swiss National Science Foundation. B Evaluation of the opposite-channel wrapping With respect to the computation detailed in ref. [34] we now have to include sub-leading order O(g6) contributions. These may come from corrections to diﬀerent bits of the leading- order expansion of eq. (2.8): 1. From corrections to the rapidities, through Ps j=1 u(1) j ∂uj. This does not substantially alter the analytic form of leading-order expression, and can be easily evaluated by shifting the integration variable to reduce the integral to take values over a single Q-function [34]. 1. From corrections to the rapidities, through Ps j=1 u(1) j ∂uj. This does not substantially alter the analytic form of leading-order expression, and can be easily evaluated by shifting the integration variable to reduce the integral to take values over a single Q-function [34]. – 10 – 2. From corrections to the integration measure µ(uγ), cf. eq. (3.6). 3. From corrections eT (1) a (uγ) to the transfer matrix eTa(uγ), cf. eq. (3.3). 4. From corrections coming from the dressing factor, cf. eq. (3.5). Below we will discuss in more detail these last three contributions. Below we will discuss in more detail these last three contributions. B.1 Corrections to the measure The strategy here is once again to perform shifts in the integration variable u in such a way as to end up with a a single Q-function Q(u). These shifts produce the total eﬀective measure 2 JHEP02(2016)165 µc = 16π2 (3(1 + 4u2)5) cosh2(πu) h 24(1 −40u2 + 80u4) + 48πu(1 −16u4) tanh(πu) −4π3u(1 + 4u2)3 tanh3(πu) + π2(1 + 4u2)2 cosh−2(πu) (−1 + 12u2)(−2 + cosh(2πu)) + 8πu(1 + 4u2) tanh(πu) i . (B.1) (B.1) Like the leading measure this falls oﬀexponentially for large \|u\| so that integration against the Baxter polynomials Q(u) is possible for any spin. The shifts u →u ± a+1 2 i must again be accompanied by the subtraction of residues from crossing the poles at ∓a 2i. B.3 Measure and residues from the ﬁrst order mirror dressing phase B.3 Measure and residues from the ﬁrst order mirror dressing phase B.3 Measure and residues from the ﬁrst order mirror dressing phase Beyond a rational factor the mirror dressing phase contains the very special combination of digamma functions Ψa(u) of eq. (3.5). This results in an integrand of the form JHEP02(2016)165 yond a rational factor the mirror dressing phase contains the very special combination digamma functions Ψa(u) of eq. (3.5). This results in an integrand of the form S1(s) X a>0 a Ψa(u) (u[+a] u[−a])3 Q(u[a+1]) + Q(u[−a−1]) −Q(u[a−1]) −Q(u[−a+1]) . (B.5) (B.5) Once again the strategy is to shift each term in the integrand in such a way as to extract an overall factor of Q(u). The presence of the digamma functions mandates some extra care. For each shift, two of the four digamma functions ψ in Ψa become independent of the bound state number a. The sum over a then simply yields a second polygamma factor. As for the digamma functions that do depend on the bound state number even after shifting, we recall that ψ(1 + x) = ∞ X k=1 1 k − 1 k + x −γ, (B.6) (B.6) which leads to a second summation over terms of the type 1/(a−k +y) or 1/(a+k +y), as well as formally some ζ(1) terms. One of the factors (u[+]a)−3 or (u[−a])−3 from the bound state measure yields 1/(a ± 1/2 ± iu)3, too. The idea is now to use partial fractions w.r.t. a to decompose products of the two factors involving a into single (if higher order) poles. We then swap the sums as before and shift to a′ = a −k and a′′ = a + k. The sums from 1 −k, k + 1 to inﬁnity, respectively, are conveniently split into an inﬁnite part with bounds 1 . . . ∞and a ﬁnite bit −a′ ∈{0 . . . k −1} and a′′ ∈{1 . . . k}. After some algebra it is found that all ζ(1) terms cancel and what is more, the ﬁnite sums “telescope” whereby the nested double sum totally disappears. The net result of this complicated looking exercise is surprisingly concise γ (u−)3 h 2iψ′(1−iu−)+2iψ′(1+iu−)+u−ψ′′(1−iu−)−u−ψ′′(1+iu−) i −(u−↔u+) . (B. B.2 Corrections to the transfer matrix Let us consider eT (1)) a (uγ) (3.3). Once again, we want to shift u in such a way as to integrate only on Q(u). For the boundary terms with Q(u[−a−1]), Q(u[a+1]) this is quite as before. The sum over k can be concisely rewritten: originally one has Q(u) ∞ X a=1 a a−1 X k=1 1 u+ ((a −k) −i u+)3 (k + i u+)3 −(u+ ↔u−). (B.2) (B.2) We can swap the order of summations and rewrite the sum in terms of a′ = a −k. In this way, the sums decouple and we can evaluate in terms of polygamma functions. The “eﬀective” integration measure emerging form these shifts is µp = 32iπ2 (1 + 4u2)4 cosh2(πu) 1 −24u2 + 16u4 + πu(3 + 8u2 −16u4) tanh(πu) + 2i 1 + 4u2 ψ′ iu− ψ′′ −iu+ + ψ′ −iu+ ψ′′ iu− . (B.3) (B.3) The terms in the second line cannot straightforwardly be rewritten in terms of trigonometric functions; nonetheless this part also has the desired asymptotic behaviour e−2π\|u\| for large values of the argument. Residues from the boundary terms in the second line of (3.3) are computed as usual. A further remark concerns the subtraction for the sum part of (3.3): to begin with let a be odd. Then the shift of Q(u−a+2k−1) →Q(u) crosses the denominator pole only if −a + 2k < 0, likewise for −a + 2k > 0 only the other term picks up a residue. For even – 11 – a, k = a/2 we look at (Q(u−)−Q(u+))/u. In either term the shift approaches the pole but does not cross it. As in a principle value prescription both terms contribute half a residue. Therefore in both cases — even and odd bound state number — we have to subtract a, k = a/2 we look at (Q(u−)−Q(u+))/u. In either term the shift approaches the pole but does not cross it. As in a principle value prescription both terms contribute half a residue. Therefore in both cases — even and odd bound state number — we have to subtract i Q (i/2) ∞ X a=1 a−1 X k=1 a (a −k)3 k3 = 2i ζ2ζ3 Q (i/2) , (B.4) (B.4) where the right-hand-side can be found by the same manipulations on the double sum as above. B.3 Measure and residues from the ﬁrst order mirror dressing phase B.3 Measure and residues from the ﬁrst order mirror dressing phase (B.7) Adding it to the contribution from the a independent digamma functions we obtain the complete eﬀective measure µm = 32π2 (1 + 4u2)4 cosh2(πu) h −1 + 24u2 −16u4 + πu(−3 −8u2 + 16u4) tanh(πu) i − 16π2 (1 + 4u2)3 cosh2(πu) h 1 −12u2 + 2πu(1 + 4u2) tanh(πu) i × h 2γ + ψ(−iu+) + ψ(iu−) i , (B.8) µm = 32π2 (1 + 4u2)4 cosh2(πu) h −1 + 24u2 −16u4 + πu(−3 −8u2 + 16u4) tanh(πu) i (B.8) × h 2γ + ψ(−iu+) + ψ(iu−) i , × h 2γ + ψ(−iu+) + ψ(iu−) i , induced by the mirror dressing phase. – 12 – Here the subtraction of residues deserves further attention. To begin with note that Ψa has simple poles at ±i(a/2+k). Shifting thus needs to be complemented by the subtraction of residues only at ±ia/2. Let us focus on the Q(u[a+1]) polynomial, i.e. on the residue at u = +ia/2. We decompose Ψa(u) = Ψreg a (u) + i u −i 2a (B.9) (B.9) by adding and subtracting the pole. Now, the pole part of Ψ(u) will combine with the (u[−a])3 factor from the bound-state mirror measure to form a fourth-order pole. This residue is then given by a third derivative acting on Q(u[a+1])/(u[+a])3. by adding and subtracting the pole. Now, the pole part of Ψ(u) will combine with the (u[−a])3 factor from the bound-state mirror measure to form a fourth-order pole. This residue is then given by a third derivative acting on Q(u[a+1])/(u[+a])3. Secondly, the bound-state measure itself has the familiar third-order pole, with now a residue Ψreg a (u) Q(u[a+1])/(u[a])3. It is not hard to work out that JHEP02(2016)165 Ψreg a \| i 2 a = 2S1(a) −1 a , Ψreg a ′\| i 2 a = i 2S2(a) −1 a2 , Ψreg a ′′\| i 2 a = −2 2S3(a) −1 a3 −4ζ(3) . (B.10) Ψreg a ′\| i 2 a = i 2S2(a) −1 a2 , Ψreg a ′′\| i 2 a = −2 2S3(a) −1 a3 −4ζ(3) . (B.10) The bound state sum therefore creates all the double ζ-values {ζ21, ζ22, ζ23, ζ31, ζ32, ζ41} up to transcendentality weight 5. All of these can be recast in terms of ordinary zeta-values. B.3 Measure and residues from the ﬁrst order mirror dressing phase The bound state sum therefore creates all the double ζ-values {ζ21, ζ22, ζ23, ζ31, ζ32, ζ41} up to transcendentality weight 5. All of these can be recast in terms of ordinary zeta-values. C Evaluation of the adjacent-channel wrapping The crucial ingredient in the evaluation of the adjacent-channel wrapping is the polyno- mial eQ(u, a, s) introduced in (3.11). This is a degree-s polynomial that can be explicitly evaluated for each given s. Interestingly, if we extract a normalisation pre-factor eQ = n eQ′ so that the highest-degree monomial in eQ′ has unit coeﬃcient, we ﬁnd n = Q(i/2). We collect the ﬁrst few expression for eQ in this normalisation in the table below. Curiously, when a = 1, the ˜Qs are in fact Baxter polynomials. spin eQ(u, a, s)/Q(i/2) 2 −1 3 + 1 4a2 + u2 4 12 35 −23 84a2 −1 48a4 −23 21u2 + 1 6a2u2 + u4 6 −60 77 + 139 220a2 + 3 44a4 + 1 320a6 + 139 55 u2 −6 11a2u2 −9 80a4u2 −36 11u4 −1 4a2u4 + u6 Let us now see how to integrate (3.12). As had been mentioned above, in the sum part of (3.8) the poles 1/u−a+2k are absent because the diﬀerence in the respective numerator factors out a power of u−a+2k. Since we are not interested in shifting u it is best to cancel these factors and to work with the remaining total numerator polynomials. In these it is of course trivial to execute the sum over k. Note that eQ times this polynomial or Q(u[a−1]), Q(u[−a+1]) contains powers up to a2s, to be augmented by the explicit factor a in the bound state measure. – 13 – Any residue has various denominator terms (a + x)n where x may contain ±1/2, ±1 and/or one or two Bethe roots. We may now use partial fractions w.r.t. to a to reduce to single (generically higher order) poles which can be summed over the bound state counter to yield polygamma functions. The procedure is well-behaved in that 1. all potentially divergent sums ζ0, ζ−1 . . . ζ−2s−1 cancel, 1. all potentially divergent sums ζ0, ζ−1 . . . ζ−2s−1 cancel, 2. ζ1 also cancels when the sums 1/(a + x) are expressed as ψ(1 + x) by eq. (B.6), 2. ζ1 also cancels when the sums 1/(a + x) are expressed as ψ(1 + x) by eq. (B.6), 2. ζ1 also cancels when the sums 1/(a + x) are expressed as ψ(1 + x) by eq. (B.6) 3. and the transcendentality level does not increase with the spin. C Evaluation of the adjacent-channel wrapping One then ﬁnds some ζ-values with rational coeﬃcients and a sum over polygamma functions containing the Bethe roots in their arguments. For spin 2 this can immediately be simpliﬁed to rational numbers and ζ3, ζ5 due to the property ψ(1+x) = ψ(x)+1/x and its derivatives. In general, we have not tried to analytically solve this — which may well be possible — but rather numerically evaluated to high precision and reconverted by the pslq algorithm. The adjacent channel calculation has a deﬁnite advantage on the opposite channel case because no integration is needed to arrive at the ﬁnal result whereby the numerical precision can be much better. JHEP02(2016)165 Open Access. This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited. References [1] G. ’t Hooft, Dimensional reduction in quantum gravity, in Salamfest (1993) 0284 [gr-qc/9310026] [INSPIRE]. [2] J.M. Maldacena, The large-N limit of superconformal ﬁeld theories and supergravity, Int. J. Theor. Phys. 38 (1999) 1113 [hep-th/9711200] [INSPIRE]. [3] E. Witten, Anti-de Sitter space and holography, Adv. Theor. Math. Phys. 2 (1998) 253 [hep-th/9802150] [INSPIRE]. [4] S.S. Gubser, I.R. Klebanov and A.M. Polyakov, Gauge theory correlators from noncritical string theory, Phys. Lett. B 428 (1998) 105 [hep-th/9802109] [INSPIRE]. [5] G. ’t Hooft, A Planar Diagram Theory for Strong Interactions, Nucl. Phys. B 72 (1974) 461 [INSPIRE]. [5] G. ’t Hooft, A Planar Diagram Theory for Strong Interactions, Nucl. Phys. B 72 (1974) 461 [INSPIRE]. [6] G. Arutyunov and S. Frolov, Foundations of the AdS5 × S5 Superstring. Part I, J. Phys. A 42 (2009) 254003 [arXiv:0901.4937] [INSPIRE]. [7] N. Beisert et al., Review of AdS/CFT Integrability: An Overview, Lett. Math. Phys. 99 (2012) 3 [arXiv:1012.3982] [INSPIRE]. [8] T. Klose, Review of AdS/CFT Integrability, Chapter IV.3: N = 6 Chern-Simons and Strings on AdS4 × CP 3, Lett. Math. Phys. 99 (2012) 401 [arXiv:1012.3999] [INSPIRE]. [9] A. Sfondrini, Towards integrability for AdS3/CFT2, J. Phys. A 48 (2015) 023001 [arXiv:1406.2971] [INSPIRE]. – 14 – [10] S.J. van Tongeren, Integrability of the AdS5 × S5 superstring and its deformations, J. Phys. A 47 (2014) 433001 [arXiv:1310.4854] [INSPIRE]. [11] K. Okuyama and L.-S. Tseng, Three-point functions in N = 4 SYM theory at one-loop, JHEP 08 (2004) 055 [hep-th/0404190] [INSPIRE]. [12] L.F. Alday, J.R. David, E. Gava and K.S. Narain, Structure constants of planar N = 4 Yang-Mills at one loop, JHEP 09 (2005) 070 [hep-th/0502186] [INSPIRE]. [13] R. Roiban and A. Volovich, Yang-Mills correlation functions from integrable spin chains, JHEP 09 (2004) 032 [hep-th/0407140] [INSPIRE]. [14] M.S. Costa, R. Monteiro, J.E. Santos and D. Zoakos, On three-point correlation functions in the gauge/gravity duality, JHEP 11 (2010) 141 [arXiv:1008.1070] [INSPIRE]. JHEP02(2016)165 [15] K. Zarembo, Holographic three-point functions of semiclassical states, JHEP 09 (2010) 030 [arXiv:1008.1059] [INSPIRE]. [16] J. Escobedo, N. Gromov, A. Sever and P. Vieira, Tailoring Three-Point Functions and Integrability, JHEP 09 (2011) 028 [arXiv:1012.2475] [INSPIRE]. [17] J. Escobedo, N. Gromov, A. Sever and P. Vieira, Tailoring Three-Point Functions and Integrability II. Weak/strong coupling match, JHEP 09 (2011) 029 [arXiv:1104.5501] [INSPIRE]. [18] N. Gromov, A. Sever and P. Vieira, Tailoring Three-Point Functions and Integrability III. Classical Tunneling, JHEP 07 (2012) 044 [arXiv:1111.2349] [INSPIRE]. [19] R.A. References Janik and A. Wereszczynski, Correlation functions of three heavy operators: The AdS contribution, JHEP 12 (2011) 095 [arXiv:1109.6262] [INSPIRE]. [20] Y. Kazama and S. Komatsu, On holographic three point functions for GKP strings from integrability, JHEP 01 (2012) 110 [Erratum ibid. 1206 (2012) 150] [arXiv:1110.3949] [INSPIRE]. [21] N. Gromov and P. Vieira, Tailoring Three-Point Functions and Integrability IV. Theta-morphism, JHEP 04 (2014) 068 [arXiv:1205.5288] [INSPIRE]. [22] I. Kostov, Classical Limit of the Three-Point Function of N = 4 Supersymmetric Yang-Mills Theory from Integrability, Phys. Rev. Lett. 108 (2012) 261604 [arXiv:1203.6180] [INSPIRE]. [23] I. Kostov, Three-point function of semiclassical states at weak coupling, J. Phys. A 45 (2012) 494018 [arXiv:1205.4412] [INSPIRE]. [24] I. Kostov and Y. Matsuo, Inner products of Bethe states as partial domain wall partition functions, JHEP 10 (2012) 168 [arXiv:1207.2562] [INSPIRE]. [25] O. Foda, Y. Jiang, I. Kostov and D. Serban, A tree-level 3-point function in the SU(3)-sector of planar N = 4 SYM, JHEP 10 (2013) 138 [arXiv:1302.3539] [INSPIRE]. [26] Y. Kazama and S. Komatsu, Wave functions and correlation functions for GKP strings from integrability, JHEP 09 (2012) 022 [arXiv:1205.6060] [INSPIRE]. [27] Y. Kazama and S. Komatsu, Three-point functions in the SU(2) sector at strong coupling, JHEP 03 (2014) 052 [arXiv:1312.3727] [INSPIRE]. [28] T. Klose and T. McLoughlin, Worldsheet Form Factors in AdS/CFT, Phys. Rev. D 87 (2013) 026004 [arXiv:1208.2020] [INSPIRE]. – 15 – [29] Y. Jiang, I. Kostov, F. Loebbert and D. Serban, Fixing the Quantum Three-Point Function, JHEP 04 (2014) 019 [arXiv:1401.0384] [INSPIRE]. [30] Z. Bajnok, R.A. Janik and A. Wereszczy´nski, HHL correlators, orbit averaging and form factors, JHEP 09 (2014) 050 [arXiv:1404.4556] [INSPIRE]. [31] Y. Kazama, S. Komatsu and T. Nishimura, Novel construction and the monodromy relation for three-point functions at weak coupling, JHEP 01 (2015) 095 [Erratum ibid. 1508 (2015) 145] [arXiv:1410.8533] [INSPIRE]. [32] Y. Jiang, I. Kostov, A. Petrovskii and D. Serban, String Bits and the Spin Vertex, Nucl. Phys. B 897 (2015) 374 [arXiv:1410.8860] [INSPIRE]. JHEP02(2016)165 [33] Z. Bajnok and R.A. Janik, String ﬁeld theory vertex from integrability, JHEP 04 (2015) 042 [arXiv:1501.04533] [INSPIRE]. [34] B. Basso, S. Komatsu and P. Vieira, Structure Constants and Integrable Bootstrap in Planar N = 4 SYM Theory, arXiv:1505.06745 [INSPIRE]. [35] L. Hollo, Y. Jiang and A. Petrovskii, Diagonal Form Factors and Heavy-Heavy-Light Three-Point Functions at Weak Coupling, JHEP 09 (2015) 125 [arXiv:1504.07133] [INSPIRE]. [36] I. Balitsky, V. Kazakov and E. References Sobko, Three-point correlator of twist-2 operators in BFKL limit, arXiv:1506.02038 [INSPIRE]. [37] Y. Jiang, S. Komatsu, I. Kostov and D. Serban, The hexagon in the mirror: the three-point function in the SoV representation, arXiv:1506.09088 [INSPIRE]. [38] C. Kristjansen, S. Mori and D. Young, On the Regularization of Extremal Three-point Functions Involving Giant Gravitons, Phys. Lett. B 750 (2015) 379 [arXiv:1507.03965] [INSPIRE]. [39] D. Arnaudov and R.C. Rashkov, Three-point correlation functions from pulsating strings in AdS5 × S5, Phys. Rev. D 92 (2015) 126001 [arXiv:1509.02831] [INSPIRE]. [40] N. Beisert and M. Staudacher, Long-range PSU(2, 2\|4) Bethe Ansatze for gauge theory and strings, Nucl. Phys. B 727 (2005) 1 [hep-th/0504190] [INSPIRE]. [40] N. Beisert and M. Staudacher, Long-range PSU(2, 2\|4) Bethe Ansatze for gauge theory and strings, Nucl. Phys. B 727 (2005) 1 [hep-th/0504190] [INSPIRE]. [41] J. Ambjørn, R.A. Janik and C. Kristjansen, Wrapping interactions and a new source of corrections to the spin-chain/string duality, Nucl. Phys. B 736 (2006) 288 [hep-th/0510171] [INSPIRE]. [42] M. L¨uscher, Volume Dependence of the Energy Spectrum in Massive Quantum Field Theories. 1. Stable Particle States, Commun. Math. Phys. 104 (1986) 177 [INSPIRE]. [43] M. L¨uscher, Volume Dependence of the Energy Spectrum in Massive Quantum Field Theories. 2. Scattering States, Commun. Math. Phys. 105 (1986) 153 [INSPIRE]. [44] B. Eden, P. Heslop, G.P. Korchemsky and E. Sokatchev, Hidden symmetry of four-point correlation functions and amplitudes in N = 4 SYM, Nucl. Phys. B 862 (2012) 193 [arXiv:1108.3557] [INSPIRE]. [45] B. Eden, P. Heslop, G.P. Korchemsky and E. Sokatchev, Constructing the correlation function of four stress-tensor multiplets and the four-particle amplitude in N = 4 SYM, Nucl. Phys. B 862 (2012) 450 [arXiv:1201.5329] [INSPIRE]. [46] G. Arutyunov, B. Eden, A.C. Petkou and E. Sokatchev, Exceptional nonrenormalization properties and OPE analysis of chiral four point functions in N = 4 SYM(4), Nucl. Phys. B 620 (2002) 380 [hep-th/0103230] [INSPIRE]. – 16 – [47] G. Arutyunov, F.A. Dolan, H. Osborn and E. Sokatchev, Correlation functions and massive Kaluza-Klein modes in the AdS/CFT correspondence, Nucl. Phys. B 665 (2003) 273 [hep-th/0212116] [INSPIRE]. [48] G. Arutyunov, S. Penati, A. Santambrogio and E. Sokatchev, Four point correlators of BPS operators in N = 4 SYM at order g4, Nucl. Phys. B 670 (2003) 103 [hep-th/0305060] [INSPIRE]. [49] G. Arutyunov and E. Sokatchev, On a large-N degeneracy in N = 4 SYM and the AdS/CFT correspondence, Nucl. Phys. B 663 (2003) 163 [hep-th/0301058] [INSPIRE]. References [50] F.A. Dolan and H. Osborn, Conformal partial wave expansions for N = 4 chiral four point functions, Annals Phys. 321 (2006) 581 [hep-th/0412335] [INSPIRE]. JHEP02(2016)165 JHEP02(2016)165 [51] B. Eden, Three-loop universal structure constants in N = 4 SUSY Yang-Mills theory, arXiv:1207.3112 [INSPIRE]. [52] B. Basso, V. Goncalves, S. Komatsu and P. Vieira, Gluing Hexagons at Three Loops, arXiv:1510.01683 [INSPIRE]. [53] A.B. Zamolodchikov and A.B. Zamolodchikov, Factorized S-matrices in two-dimensions as the exact solutions of certain relativistic quantum ﬁeld models, Annals Phys. 120 (1979) 253 [INSPIRE]. [54] N. Beisert, B. Eden and M. Staudacher, Transcendentality and Crossing, J. Stat. Mech. (2007) P01021 [hep-th/0610251] [INSPIRE]. [55] N. Beisert, The SU(2\|2) dynamic S-matrix, Adv. Theor. Math. Phys. 12 (2008) 948 [hep-th/0511082] [INSPIRE]. [56] H.-Y. Chen, N. Dorey and K. Okamura, On the scattering of magnon boundstates, JHEP 11 (2006) 035 [hep-th/0608047] [INSPIRE]. [57] R. Roiban, Magnon Bound-state Scattering in Gauge and String Theory, JHEP 04 (2007) 048 [hep-th/0608049] [INSPIRE]. [58] G. Arutyunov and S. Frolov, The Dressing Factor and Crossing Equations, J. Phys. A 42 (2009) 425401 [arXiv:0904.4575] [INSPIRE]. [59] A.V. Kotikov, L.N. Lipatov, A. Rej, M. Staudacher and V.N. Velizhanin, Dressing and wrapping, J. Stat. Mech. (2007) P10003 [arXiv:0704.3586] [INSPIRE]. [60] Z. Bajnok and R.A. Janik, Four-loop perturbative Konishi from strings and ﬁnite size eﬀects for multiparticle states, Nucl. Phys. B 807 (2009) 625 [arXiv:0807.0399] [INSPIRE]. [61] Z. Bajnok, R.A. Janik and T. Lukowski, Four loop twist two, BFKL, wrapping and strings, Nucl. Phys. B 816 (2009) 376 [arXiv:0811.4448] [INSPIRE]. [62] Z. Bajnok, A. Hegedus, R.A. Janik and T. Lukowski, Five loop Konishi from AdS/CFT, Nucl. Phys. B 827 (2010) 426 [arXiv:0906.4062] [INSPIRE]. [63] H.R.P. Ferguson, D.H. Bailey and S. Arno, Analysis of pslq, an integer relation ﬁnding algorithm, Math. Comput. 68 (1999) 351. [64] D. Bombardelli, A next-to-leading Luescher formula, JHEP 01 (2014) 037 [arXiv:1309.4083] [INSPIRE]. – 17 –
https://openalex.org/W2069915922	https://zenodo.org/records/1701242/files/article.pdf	English	null	XLV. <i>The photoelectric effect</i>.—II	The London, Edinburgh and Dublin philosophical magazine and journal of science	1,913	public-domain	8,275	Philosophical Magazine Series 6 ISSN: 1941-5982 (Print) 1941-5990 (Online) Journal homepage: http://www.tandfonline.com/loi/tphm17 Date: 04 June 2016, At: 22:31 Download by: [UQ Library] , Communicated by the Authors. t K. T. Compton, Phil. Mag. vol. xxiii, p. 579 (]912). Phil. Mag. S. 6. V ol. 26. No. 154. Oct. 1913. 2 P Karl T. Compton & O.W. Richardson To cite this article: Karl T. Compton & O.W. Richardson (1913) XLV. The photoelectric effect.—II, Philosophical Magazine Series 6, 26:154, 549-567, DOI: 10.1080/14786441308635002 To link to this article: http://dx.doi.org/10.1080/14786441308635002 Published online: 08 Apr 2009. Submit your article to this journal Article views: 19 View related articles Citing articles: 12 View citing articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tphm17 Download by: [UQ Library] [SIXTH SERIES.] Downloaded by [UQ Library] at 22:31 04 June 2016 0 C TOBE_R 1913. XLV. The Photoelectric Effect.--II. B~j KARL T. COMPTOS and O. W. RICHARDSON, Princeton University . INTRODUCTION. r\|'~HERE are two lines of experimental investigation which 9 have been followed in the search for the ultimate explanation of the photoelectric effect from metals. Most of the work in this subject has had to do with the velocities of the emitted electrons and the relation of these velocities to the nature of the emitting metal and the frequency and intensity of the incident light. Numerous experimenters have shown that the velocities of the electrons are independent of the in~ensir, y of the incident light, and are nearly, if not entirely, independent of the photoelectric fatigue, except in so far as this fatigue alters the intrinsic potential of the metal t, ~md therefore alters the amount of energy lost by the electrons in escaping. It has therefore been possible to investigate the relation between velocity and frequency without having to take special means of allowing for these two factors. To this extent the investigation of photoelectron velocities is free from complications which are troublesome when the nmnber of emitted electrons instead of their velocities is being studied. This line of investigation has therefore received the most attention. r\|'~ 9 ) 2 P 550 50 Mr. K. T. Compton and Prof. O. W. Richardson on Recent work by the writers* and by Hughes t has well established the fact that there is a linear relation between the maximum initial kinetic energy of the emitted electrons and the frequency of the incident light. This relation has been suggested before, but secondary effects due to reflected light and reflected electrons, to impurity of the spectrum, and several other factors have rendered the experimental proofs of this relation uncertain and contradictory. The relation may be expressed by the formula V = kv-- w0, where V is the maximum initial kinetic energy of the liberated electrons, expressed in equivalent volts, k and we are constants, and v is the frequency of the incident light. The average value of ]c for eight different metals we found to be Downloaded by [UQ Library] at 22:31 04 June 2016 k= 3005"3 10 -27 erg see., e while Hughes ibund k= 30-0-5"6 10 -:7 erg see. e as the average for a number of metals, w0 is the work, expressed in equivalent volts, done by an electron in escaping from the metal. where h is Planck's constant, 6"55 10 -27 erg sec. The 9 Phil. Mug'. vol. xxiv. p. 575 (1912). ~- :Phil. Trans. Roy. Soc. London, A. vol. ccxii, p. 205 (]912). $ O. W. Richardson, Phys. Rev. vol. xxxiv, p. 119 (19] 2) ; Phi]. Mug. vol. xxiii, p. 615 (1912) ; ~-ol. xxiv. p. 570 (1912} ; Science, vcl..~xxvi. p. w176 300_e__ appears throughout this paper because the practical system of electrical units is employed. g p ( ) ~- :Phil. Trans. Roy. Soc. London, A. vol. ccxii, p. 205 (]912). $ O W Ri h d Ph R l i 119(19]2) P INTRODUCTION. Our work showed that, within the limits of experimental error, w o for any metal has a value which may be calculated independently from a direct knowledge of we for platinum and the contact difference of potential between platinum and the given metal. p g These experimental results confirm the theory of photo- electric action which was recently developed by one of the writers $. By thermodynamic and statistical methods the conclusion was reached that the relation between velocity and frequency should be given by the equation w V = 30O I~--We, 9 . . . . . (1) e (1) where h is Planck's constant, 6"55 10 -27 erg sec. The 9 Phil. Mug'. vol. xxiv. p. 575 (1912). Phil T S L d A l ii 205 (]912) where h is Planck's constant, 6"55 10 -27 erg sec. The 9 Phil. Mug'. vol. xxiv. p. 575 (1912). 9 Phil. Mug. vol. xxiv. p. 575 (1912). ~- :Phil. Trans. Roy. Soc. London, A. vol. ccxii, p. 205 (]912). $ i h d h l i h y , , p (] ) $ O. W. Richardson, Phys. Rev. vol. xxxiv, p. 119 (19] 2) ; Phi]. Mug. vol. xxiii, p. 615 (1912) ; ~-ol. xxiv. p. 570 (1912} ; Science, vcl..~xxvi. p 176 300eappears thro gho t this paper beca sethe practical p. w176 300_e__ appears throughout this paper because the practical system of electrical units is employed. the _Photoelectric Effect. 551 the _Photoelectric Effect. 551 551 theory also led to an expression connecting w 0 for two metals s and p of the form* ~w o-~,wo = V~-- V~-- O ~O (V,-- V~), . (2) (2) where Vp--V, is the contact difference of potential between the metals, expressed in volts. The term involving the absolute temperature t? expresses the Peltier effect, and may be neglected in comparison with Vp--Vs. The theoretical and experimental values of w0 agree fairly well. The fact that ]r is considerably less than the theoretical value h is probably not due to manipulative error, but to experimental error of a more insiduous nature. On the whole, however, the support of the theory is consistent and satisfactory. * O. W. Richardson, Phil. Mag. vol. xxiii, p. 264 (1912) ; vol. xxiii. p. 615 (1912). t .Ann. d..phys, vol. viii. p. 149 (1902). j Hallwachs, .Ann. d..Phys, vol. xxx. p. 593 (1909). 2P2 INTRODUCTION. Downloaded by [UQ Library] at 22:31 04 June 2016 pp y y The second line of attack on the photoelectric problem is the investigation of the effect of the intensity of the light, the frequency of the light, and the nature of the metal on the number of electrons emitted in unit time. Lenard t and others have shown that the number of emitted electrons is simply proportional to the intensity of the incident light. As to the influence of the nature of the metal, it is known in a general way that the more eleetropositive metals give larger photoelectric currents. It is also known that the slightest oxidation or other chemical action on the illuminated surfhce of the metal greatly decreases the size of tlle photoelectric currents. This "fatigue " is prevented in proportion as a perfect vacuum is approached. Except for the work of Pohl and Pringsheiin, which will be discussed later, little is known concerning the relation between the frequency and the rate of emission of electrons except the fact that photoelectric currents apparently increase as shorter wave-lengths of light are employed $. p y The theory of photoelectric action developed by one of the writers w yields an expression connecting the frequency, nature of the metal, and number of electrons emitted in unit time, as well as the expression for the initial kinetic energy tdready referred to. The complete theory leads to the following equations: N=0 when 0< 300by<we . . . . . (3) e (3) J~{r. K. T. Compton and Prof. 0. W. Richardson o~ 552 552 and J~{r. K. T. Compton and Prof. 0. W. Richardson o~ and N= Ah/" ewo "~ 300 ~OOh-v] and V=--e by--w~ when Wo< 300-hv<O. e (4) (4) In these equations N is the number of electrons emitted per unit area per unit time by light of unit intensity, A is a constant characteristic of the metal, and R is the absolute gas constant reckoned for one molecule. The other symbols have the meanings assigned to them previously. g g p y An examination of the expressions for N shows several facts in support of the theory. There is apparently a critical frequency, given by Downloaded by [UQ Library] at 22:31 04 June 2016 y Wo e 300 h or X 0- c, . (5) v~= h 300' e w0 (5) below which there is no photoelectric emission. * H. L. Cooke and O. W. Richardson, Phys. Review E2] vol. i. p. 71 (1913); 1%il. Mag. vol. xxv. p. 624 (1913). INTRODUCTION. e e N 0, when O 300hv wo e and V= 300hv--Wo, when w0< 300by < 0. e e Downloaded by [UQ Library] at 22:31 04 June 2016 There still remains to be tested the equation Ah[l_ e Wo~ 300 N= R~v~ \ ~ ITs,/' when Wo<---e hv< O. (6) (6) The purpose of this paper is the experimental investigation of the relations involved in this equation. INTRODUCTION. p This we found to be true. The agreement on this point between theory and experiment is illustrated by Table I., TABLE I. Metal. Xo observed, ~o calculated. l~a ............ A1 ............... Mg ............ Zn ............ I Sll ............ ]3i ............... Cll . . . . . . . . . . . . Pt ............ 577 tqz 360 375 357 337 337 309 280 550 y/L 360 390 340 305 290 275 270 which shows the values of the critical wave-lengths for eight metals. Tile calculated values of ko are obtained from equations (5) and (2), taking w0 for platinum to be 4"7 volts, and using values of the contact difference of potential of each metal with respect to platinum as taken from standard tables. The observed values are a little uncertain, since the photoelectric currents are small in this region, and since we used a quartz-mercury arc, which gives a discontinuous spectrum. However, we determined X o as best we could under the circumstances. Some of these values were deter- mined during the course of this investigation. The rest are which shows the values of the critical wave-lengths for eight metals. Tile calculated values of ko are obtained from equations (5) and (2), taking w0 for platinum to be 4"7 volts, and using values of the contact difference of potential of each metal with respect to platinum as taken from standard tables. The observed values are a little uncertain, since the photoelectric currents are small in this region, and since we used a quartz-mercury arc, which gives a discontinuous spectrum. However, we determined X o as best we could under the circumstances. Some of these values were deter- mined during the course of this investigation. The rest are the Photoelectric Effect. 553 taken from our last paper. Considering the ihct that the intrinsic potentials of the investigated surfaces may have been somewhat different from the values given in the tables, and since the finite size of the spectrometer-slits caused a slight overlapping in the spectrum, these results are in very fair support of the theory. pp y We have, therefore, good experimental support of those results of the theory expressed by y p y N=0, when O<300hv<wo e y p y N=0, when O<300hv<wo e V= 300hv--Wo, when w0< 300by < 0. Preliminar!/ Considerations. At the outset it should be remarked that equation (6) is not necessarily the only or the complete expression of the theory. The thermodynamic and statistical treatment yielded two integral equations which were not satisfied by any analytic function continuous throughout the entire range o~ variation. It was found, however, that equations (3) and (4) are a solution of these integral equations. The complete solution may therefore consist of the sum of several terms, of which equation (6) is only one part. This suggestion will be considered again later in the paper. It is also possible that the approxinmtions used lead to an error in this part of the solution. One evident requirement of the theory is that the rate of photoelectric emission must reach a maximmn at some par- ticular frequency, instead of increasing indefinitely with the frequency as was at first supposed. For from equation (6) dN Ah { e 3we ) dv -- -ff~3 \3()0 hv '2, = O, whence 3 e we 3 v .... = 2 300 h -- ~vo, 9 . . . . . (7) and Vmi=. = cr . . . . . . . . (8) d2N . ~ negative when v=~v o d~ve is (positive when v= dN Ah { e 3we ) dv -- -ff~3 \3()0 hv '2, = O, whence 3 e we 3 v .... = 2 300 h -- ~vo, 9 . . . . . (7) and Vmi=. = cr . . . . . . . . (8) d2N . ~ negative when v=~v o d~ve is (positive when v= (7) (8) and (8) 554 Mr. K. T. Compton and Prof. O. W. Richardson on Therefore equation (6) requires the photoelectric sensitiveness of any metal to be a maximum at :~ wave-length equal to -~X0, where X0 is the longest wave-length at which electrons are emitted, and it requires that the sensitiveness approach zero when the wave-length is exceedingly small. g g y The existence of a maximum in the curve representing the relation between frequency and rate of emission is proved by the work done on a large number of metals by Pohl and Pringsheim$. The results of their experiments may bo generalized in the statement that the wave-length-rate of emission curve for every metal is of the type shown in fig. * Verb. d. 1).-Phys. Ges. voI. xiii. p. 474 (1911); vol. xii. p. 349 (1910) ; vol. xii. p. 682 (1910) ; vol. xiv. p. 46 (1912), &c. Preliminar!/ Considerations. 1, a b c d e, and that for each metal the starting-point and the position of the maximum are at wave-lengths peculiar to the given metal. N is the nmnber of electrons emitted Downloaded by [UQ Library] at 22:31 04 June 2016 Fig. 1. C (1 per second by a unit amount of light-energy. They find that the part b c d of the curve is present only when there is u component of the electric intonslty in the light-wave which is perpendicular to the mirror surface of the metal. When the light is incident perpendicularly the part b c d degenerates into the dotted line b d, which forms part of a continuous curve a bdg. This curve is not in good agreement with equation (6), which is represented by the curve afcg. However, the theoretical curve afcg was obtained on the supposition that the radiation incident on the metal is isotropic, while the experimental curve of Pohl and Pringsheim was obtained with the light incident at a certain definite angle. They found that the relative importance of the "hmnp " b c d depends greatly on the angle of incidence. Therefore these curves are not strictly comparable. per second by a unit amount of light-energy. They find that the part b c d of the curve is present only when there is u component of the electric intonslty in the light-wave which is perpendicular to the mirror surface of the metal. When the light is incident perpendicularly the part b c d degenerates into the dotted line b d, which forms part of a continuous curve a bdg. This curve is not in good agreement with equation (6), which is represented by the curve afcg. However, the theoretical curve afcg was obtained on the supposition that the radiation incident on the metal is isotropic, while the experimental curve of Pohl and Pringsheim was obtained with the light incident at a certain definite angle. They found that the relative importance of the "hmnp " b c d depends greatly on the angle of incidence. Therefore these curves are not strictly comparable. y p However, Pohl and Pringshcim found that the position of the maximmn is nearly or entirely independent of the angle 555 the .Photoelectric J';fl'ect. of incidence. Preliminar!/ Considerations. 556 M K T C t dP f O W Ri h d 556 Mr. K. T. Compton The Photoe The apparatus which making the photoelectri fig. 2. It is similar to tha (a i 5 Downloaded by [UQ Library] at 22:31 04 June 2016 556 Mr. K. The appar making the fig. 2. It is (a Downloaded by [UQ Library] at 22:31 04 June 2016 556 Mr. K. T. Compton andProf. O. W. Richardson on Preliminar!/ Considerations. Therefore we may use their data on the wave- length at which the photoelectri~c emission is a maximum for various metals, and compare these results with the require- ments of the theory expressed by equation (6), or rather, by equation (7). This comparison is shown in Table II. The calculated values of X .... are obtained from equation (7), taking w0 for each metal as given by equation (2), taking values of the contact difference of potential with respect to platinum from standard tables, and taking for platinum w 0--4"7 volts. The observed values of ~ .... are taken directly from Pohl and Pringsheim's papers. The blank spaces re- present cases in which the data are not available and are put in because they may be of interest later. Downloaded by [UQ Library] at 22:31 04 June 2016 TABLE II. Metal. kmax. calculated. ~max. observed. I~b ......... K ............ ~a ......... Ca ......... ~a ........ Li ........... Mg ......... 3_1 ............ Zn ......... Sn ......... ~u ......... Pt ......... 4~ 367 .oo ,,~ 2~5 240 223 200 184 180 475/~ 440 335 360 280 280 250 247.5 ... TABLE II. Metal. kmax. calculated. ~max. observed. I~b ......... K ............ ~a ......... Ca ......... ~a ........ Li ........... Mg ......... 3_1 ............ Zn ......... Sn ......... ~u ......... Pt ......... 4~ 367 .oo ,,~ 2~5 240 223 200 184 180 475/~ 440 335 360 280 280 250 247.5 ... Wherever comparison is possible the agreement with the theory is quite good. Pohl and Pringsheim did not deter- mine points close together on their curves, and the exact positions of the maxima are doubtful within limits large enough to account for most of the discrepancies in the table. The most obvious discrepancy between the theoretical equation and the curves of Pohl and Pringsheim is that the .sensitiveness in the region of very short wave-lengths is increasing as the wave-length decreases, while the theory requires that it should finally fall to zero. The further ex- perimental test of the theory, as expressed by equation (6), which we shall now proceed to describe, differs from the work of Pohl and Pringsheim principally in the fact that we secure approximate isotropic monochromatic radiation falling on the strip, and that we extend the measurements farther out into the ultra-violet than has hitherto been done. The Photoelectric Measurements. The apparatus which we found most satisfactory for making the photoelectric measurements is illustrated in fig. 2. It is similar to that used in our previous experiments fig. 2. It is similar to that used in our previous experiments Fig. 2. (a i 5 3 ' e-.., ?m air pare on electron velocities, with the addition of arrangements designed to make possible a more perfect vacuum and to give an absolutely clean, untarnished meial strip upon which to focus the light from the spectrometer-slit. Light from the Downloaded by [UQ Library] at 22:31 04 June 2016 Fig. 2. on electron velocities, with the addition of arrangements designed to make possible a more perfect vacuum and to give an absolutely clean, untarnished meial strip upon which to focus the light from the spectrometer-slit. Light from the on electron velocities, with the addition of arrangements designed to make possible a more perfect vacuum and to give an absolutely clean, untarnished meial strip upon which to focus the light from the spectrometer-slit. Light from the 557 t]~e lO /totoelectric E~ect. Heraeus quartz mercury arc-lamp Q entered the collimntor- slit of a I-Iilger monochromatic ultra-violet illuminator and entered the photoelectric cell through a quartz window P. The apparatus was adjusted so that the light from the slit was exactly focussed on the strip S of metal to be tested~ so that the strip and the image of the slit exactly coincided for each wave-length of light. The wave-length of light falling on the strip was shown directly by the graduated drmn W. The degree of adjustment could be examined by looking" into the bulb at G. The requisite vacuum was obtained by first using a Gaede pump~ then sealing off the pump connexion at F and immersing the tube of coconut charcoal R1 in liquid air. Under the influence of the light, electrons escaped from the strip S to the inside silvered surface of the bulb B. The rate of emission of electrons was measured by observing the rate of deflexion of the electrometer E, whic~ was connected with the strip S. The bulb B was maintained at a positive potential of 5 volts (contact difference of potential being allowed for). This gave complete saturation and prevented any reverse current from B to S due to light reflected from the strip S to the walls of the bulb B. The Photoelectric Measurements. Downloaded by [UQ Library] at 22:31 04 June 2016 p It has been mentioned that the theory is based on the assumption that isotropic radiation falls" upon the metal strip. Since the light falling upon the strip in our apparatus was practically a parallel beam, we secured the equivalent of the incidence of isotropic light by bending the strip around a lengthwise axis into such a shape that, with a parallel beam of light striking it, there would be the same proportion of light incident at each angle as if isotropie radiation were incident on a plane surface. We determined graphically the shape of cross-section of the strip necessary to fulfil these conditions, and found that it is approximately elliptical, with the major axis twice the minor axis, and pointing in the direction of the light. g In order to obtain absolutely fresh metal surf~lce for the photoelectric tests we adopted the following device. The metal strip S, of elliptical cross-section and just the size of the image of the slit, was fixed to the end of a light aluminium rod R which was capable of a motion up and down in a closely fitting brass tube. Stops D were arranged so that the upper limit of nmtion brought the strip exactly in the path ot the light, while the lower limit left the strip inside of a small bulb T directly below the main bulb B. The brass tube was firmly waxed into the outer glass tube at A. A soft iron block I was attached to the upper end of the aluminimn rod, so that the strip could be moved up or down 558 Mr. K. T. Compton and Prof. O. W. Richardson on at will by sliding an electromagnet up or down outside of the apparatus. A suspending spring C was just strong enough to keep the strip in its regular position when the electro-, magnet was removed. A bit of the metal to be investigated was placed in the quartz furnace ld, which was wound with a platinum heating-coil. After the requisite vacumn had been obtained, the strip S was drawn into the bulb T and the metal distilled from the furnace over on to the strip, where a thin film of the metal to be investigated was deposited. The Photoelectric Measurements. Then the strip was returned to its position in the bulb B, and photoelectric measurements were made as soon as pos- sible. This method worked very satisfactorily, and can be used for most metals which melt below about 1000 ~ C. Downloaded by [UQ Library] at 22:31 04 June 2016 This was the apparatus used in testing sodium. The low melting-point of caesium made a variation of this process more convenient. The caesium, after being washed in an- hydrous ether, was placed in the small bulb i~, which was at once sealed on to the tube ML. After a liquid-air vacuum had been maintained for several hours the caesium was melted so that it flowed successively through the constrictions at ]~[ and L. The csesium oxide and other impurities were leit behind at L, and the tube was sealed offat 3{ wi~h a blowpipe. Then the caesium was distilled on to the strip by gently heating the bulb T. This method proved very satisfactory, for the c~esium gave practically constant photoelectric currents during the time necessary to complete a test. g y p These precautions were not taken when aluminium and platinum were tested, although in the ease of aluminimn we should probably have obtained better results had we taken them. in these eases the metal strip was scraped with a clean knife-blade in air, quickly placed in position, and the vacuum pmnp started. Certa{nly some oxidation of the aluminium surface occurred before the vacuum was obtained. As soon as the strip S was ready for the test the drmn W was turned so as to cause light of different wave-lengths to fall successively on the strip. For each wave-length we measured the deflexion of the electrometer in some definite interval of time, say ten seconds. Thus data were obtained giving, in arbitrary units, the number of electrons liberated in unit time front the metal by the light of each wave-length represented in the spectrum of the mercury are. But in order to apply these results ~o test equation (6) it was neces- sary first to make a correction for the fact that the various lines in the arc spectrum were not all of equal intensity. This necessitated a calibration in order to find the relative energies in different parts of the spectrum. The Photoelectric Measurements. Then, when the g y p These precautions were not taken when aluminium and platinum were tested, although in the ease of aluminimn we should probably have obtained better results had we taken them. in these eases the metal strip was scraped with a clean knife-blade in air, quickly placed in position, and the vacuum pmnp started. Certa{nly some oxidation of the aluminium surface occurred before the vacuum was obtained. As soon as the strip S was ready for the test the drmn W was turned so as to cause light of different wave-lengths to fall successively on the strip. For each wave-length we measured the deflexion of the electrometer in some definite interval of time, say ten seconds. Thus data were obtained giving, in arbitrary units, the number of electrons liberated in unit time front the metal by the light of each wave-length represented in the spectrum of the mercury are. But in order to apply these results ~o test equation (6) it was neces- sary first to make a correction for the fact that the various lines in the arc spectrum were not all of equal intensity. This necessitated a calibration in order to find the relative energies in different parts of the spectrum. Then, when the the Photoelectric Effect. 559 number of electrons liberated per unit time by light of each wave-length was divided by the energy of the light of the corresponding wave-length, the quotient represented the relative number of electrons liberated in unit time by a nnit 9 amount of light-energy for each wave-length. These values. may be used directly to test equation (6). They are called N in the curves and are plotted as ordinates. * Franklin Inst. Journ. clxxv, p. 151 (1913). The Energy Calibration. The relative energies in different parts of the spectrum were measured by means of a bismuth-silver linear thermo- pile made by W. W. Coblentz % This thermopile was adapted to the monochromatic illuminator and was placed directly back of the slit in the telescope arm. This slit was adjusted to the exact width and position of the strip which had been tested photoelectrically. The energy was determined by a compensation method illustrated in fig. 3. The thermopile Downloaded by [UQ Library] at 22:31 04 June 2016 Fig. 3. T Fig. 3. T Fig. 3. T was connected in series with a shunt S and a Thomson astatic galvanometer G. A second circuit contained, in series, the shunt S, a battery B, and a variable resistance R. The battery was connected so that its electromotive force opposed that generated by the thermopile. The resistance tt was adjusted so that no current flowed through the galva- nometer. Under these conditions the radiation energy falling on the ~hermopile was inversely proportional to the resistance R. For the current through the battery is E R+b+ 1 - 1 ~+ E R+b+ 1 - 1 ~+ E where E is the electromotive force of the battery, and b, s, t, and g are the resistances of the battery, shunt, thermopile, and galvanometer respectively. R varied from 20,000 to 1,200,000 ohms in different parts of the spectrum, b and s were each 0"01 ohm, t was 8 ohms and g was 10"5 ohms. 560 ?cir. K. T. Compton and Prof. O. W. Richardson on Therefore we may neglect all of the resistance except R, and write 560 p Therefore we may neglect all of the resistance except R, and write E I=~. E I=~. The drop across S was as Es K d=Is-- R --R' where K is a constant. When the resistance R was ad- justed so that no current passed through the galvanometer, obviously K K Downloaded by [UQ Library] at 22:31 04 June 2016 where e is the electromotive force generated by the thermo- pile. Thus the electromotive force was inversely propor- tional to the resistance R. But the electromotive ~brcs is proportional to the rise in temperature, which in turn is proportional to the energy of the incident light. Therefore the reciprocal of the resistance required for compensation was a measure of" the energy of the light incident on the strip. gy g p By this method the galvanometer was a detector only, and ~he measurements were inde~endent of any changes in its sensitiveness. For measurements in the extreme ultra-violet between wave-lengths 200/z/~ and 250/~/z the sensitiveness of the galvanometer was about 5 x 10 -11 ampere per division. In the rest of the spectrum it was unnecessary to maintain such high sensitiveness. g The calibration of the spectrum was repeated at intervals during the course of the investigation. It was found that, for a given number of amperes running through the arc- lamp, the energy distribution in the spectrum remained practically constant during the period of the three months' 9 9 ~. 1~ 9 9 mvestlgatmn. The slight varmtmns that occurred were allowed for. There is one peculiarity of the apparatus which should be mentioned before taking up the experimental results. The collimator slit, the metal strip~ and the slit in front of the thermopile were each ~g inch wide. This large width re- suited in considerable overlapping and impurity in the visible part of the spectrum from wave-lengths 4=00/~/~ to 700/z/~. E But out in the n]tra-vlolet the dispersive power of the instrument was about twenty times as great as in the visible spectrum, so that in this region, which was the most important, the error due to the finite sizes of the slits was negligible. The reason for making the slits so wide was that in the visible spectrum the photoelectric currents were the Photoelectric E]f ect. 561 small and in the ultra-violet the energy was small, so that in both parts of the spectrum it was found best to use fairly wide slits and strips. The effect of this width of slits will he considered when the experimental results are discussed. Ex29erlmental Results. 150 (10) t3 202 148"5 204 147 208 144 212 141"5 214 140 220 136 224 133"7 226 132"5 228 131"5 230 130 232 129"2 234 128"2 236 127"1 240 125 242 124 254 118 265 113'2 275 109 290 103"5 313 95"5 339 88'5 366 82 d I ~ 103 9O 85 99 149 172 249 319 343 368 401 423 439 500 50O 53l 1300 518 294 141 82 17 7 d 2 9 87 76 70 92 t 130 150 223 294 322 343 385 406 t30 t81 493 518 1 200 512 '287 140 78 17 7 d, 95 82 78 96 140 161 236 306 333 353 393 415 435 490 497 525 25O 515 29O 140 8O 17 7 1"1 0"9 0"8; 0"9 1'2~ 1'4; 2"01 2"8~ 2"9~ 3"41 4"2( 4"7i 5'3i 5"9( 7'1 7"7 285 18"5 14"5 21"2 55"5 295 775 d 85 90 94 105 112 113 118 113 113 102 93 87 82 83 70 68 44 28 2O 6"5 1"5 0'6 0'09 Fig. 5. A] 120 ~ / 70 80 90 100 lt0 120 130 14"0 1~0 of N for the different metals canno~ be compared w ach other, since the experimental conditions were usua Aluminlum.--Table IV. contains the results of the first set of measurements with aluminium. The absolute values Aluminlum.--Table IV. contains the results of the fi set of measurements with aluminium. The absolute valu TABLE IV. Alumlnium. Ex29erlmental Results. Platinum.--The experimental results in the case of platinum are given in Table III. ~, and v are tile wave-length and frequency of the incident light, dl and d2 are deflexions of the electrometer per half minute, d is the average of dl and d~, E is the energy of the incident light in arbitrary units, and N is proportional to the number of electrons enfitted per unit time by light of unit intensity. The readings were taken in order down the colmnn dl and then up the column d~. The averages should be relatively independent of any fatigue Downloaded by [UQ Library] at 22:31 04 June 2016 TABLE lIT. Platinum. TABLE lIT. Platinum. )k~ 04 ',14 20 :30 ',40 ',54 60 ~70 :80 P, 150(10)r~ 148'5 147 140 136 130 125 118 115"5 111 107 103"5 100 d 1 9 2 4 5 7 5 3 1 3 4"5 6'5 2'5 2'5 5 05 0 E, ,N 1 '0( 1 "0{ 0'8; 1"7 2'0 4"1 6'3 24"3 22:7 16"6 14"5 21 "2 43"5 -:E 13 l0 ~7 ,~5 ),7 .6 8"3 1"5 0"7 0'3 0'2 0'02 0 40 Fig. 4. Pt A 5O 10 0 90 100 llO l~O 150 140 1~0 effect. The relation between X, v, and N is shown graphically in fig. 4, curve 1. Curve 2 is the graph of equation (6), the ordinates being in arbitrary units. effect. The relation between X, v, and N is shown graphically in fig. 4, curve 1. Curve 2 is the graph of equation (6), the ordinates being in arbitrary units. 562 562 Mr. K. T. Compton and Prof. O. W. Richardson on 562 Mr. K. T. Compton and Prof. O. W. Richardson on p Aluminlum.--Table IV. contains the results of the fi et of measurements with aluminium. The absolute valu TABLE IV. Alumlnium. Ex29erlmental Results. 150 (10) t3 202 148"5 204 147 208 144 212 141"5 214 140 220 136 224 133"7 226 132"5 228 131"5 230 130 232 129"2 234 128"2 236 127"1 240 125 242 124 254 118 265 113'2 275 109 290 103"5 313 95"5 339 88'5 366 82 d I ~ 103 9O 85 99 149 172 249 319 343 368 401 423 439 500 50O 53l 1300 518 294 141 82 17 7 d 2 9 87 76 70 92 t 130 150 223 294 322 343 385 406 t30 t81 493 518 1 200 512 '287 140 78 17 7 d, 95 82 78 96 140 161 236 306 333 353 393 415 435 490 497 525 25O 515 29O 140 8O 17 7 1"1 0"9 0"8; 0"9 1'2~ 1'4; 2"01 2"8~ 2"9~ 3"41 4"2( 4"7i 5'3i 5"9( 7'1 7"7 285 18"5 14"5 21"2 55"5 295 775 d 85 90 94 105 112 113 118 113 113 102 93 87 82 83 70 68 44 28 2O 6"5 1"5 0'6 0'09 Fig. 5. A] 120 ~ / 70 80 90 100 lt0 120 130 14"0 1~0 of N for the different metals canno~ be compared w each other, since the experimental conditions were usua altered between each tesL The curve is shown in fig. of N for the different metals canno~ be compared with each other, since the experimental conditions were usually altered between each tesL The curve is shown in fig. 5 563 the Photoelectric .Effect. (curve 1). Curve 2 is the graph of equation (6), reckoned for aluminium, • being expressed in arbitrary units. The readings for curve 3 were taken 72 hours after those ibr curve 1. The shrinkage represents the photoelectric fatigue during this interval. Curves 4 and 5 were taken under the same conditions as curve 1, except that the strip S was bent so as to receive the light at perpendicular and acute incidence in the two cases. Apparently the angle of inci- dence wa~ of' little or no influence. It mus~ be remem- bered, however, that the aluminium strip was scraped with a knife-blade and was far from having a polished plane surface. Therefore this test may indicate little except that there can be no error in curve 1 due to any slight departure of the shape of the strip from its supposed elliptical cross-section. Ex29erlmental Results. / 20 i / z0 ..// 50 60 /o % I 70 80 90 100 110 120 150 140 150 Fig'. 7. Downloaded by [UQ Library] at 22:31 04 June 2016 Evidently the method employed in purifying and distilling the emsium made possible a better vacumn, and consequently there was less fatigue than in the preceding cases. Evidently the method employed in purifying and distilling the emsium made possible a better vacumn, and consequently there was less fatigue than in the preceding cases. Ex29erlmental Results. Downloaded by [UQ Library] at 22:31 04 June 2016 Downloaded by [UQ Library] at 22:31 04 June 2016 p pp p Sodium.--The results in the case of sodium are shown in fig. 67 curves 1, 2~ 3. Experimental difficulties prevented Fig. 6. 80 7O " // 50 40 N 1[] /" 50 60 70 80 90 I00 II0 I~0 150 l~tO 150 0o) the photoelectric measurements being taken within less than one hour after the sodium had been distilled on to the strip S. Curves 1, 2, and 3 represent the variation of ~T with v, 1, 289 and 18 hours respectively after distillation. Curve ~t repre- sents (6) plotted for sodium, the ordinates being in arbitrary units. The photoelectric currents were about fifty times as large as in the case of aluminium. The experimental results were o[ exactly the same type as those shown in Tables IIL and IV. Fig. 6. Fig. 6. Fig. 6. 80 7O " // 50 40 N 1[] /" 50 60 70 80 90 I00 II0 I~0 150 l~tO 150 0o) 0o) the photoelectric measurements being taken within less than one hour after the sodium had been distilled on to the strip S. Curves 1, 2, and 3 represent the variation of ~T with v, 1, 289 and 18 hours respectively after distillation. Curve ~t repre- sents (6) plotted for sodium, the ordinates being in arbitrary units. The photoelectric currents were about fifty times as large as in the case of aluminium. The experimental results were o[ exactly the same type as those shown in Tables IIL and IV. 56t: C(esium.~The resnlts in the ease of cmsimn are illustrated by fig. 7, curare 1. The energy calibration in this ease was less satisfactory than in the preeedin~ eases, since the galvanometer was rather unsteady. On the other hand, the photoelectric measurements were exceedingly satisfactory. p g y y Fig'. 7. 8O Cs 70 N r 50 ,./ 4-0 / 1 SO! / 20 i / z0 ..// 50 60 /o % I 70 80 90 100 110 120 150 140 150 Evidently the method employed in purifying and distilling the emsium made possible a better vacumn, and consequently there was less fatigue than in the preceding cases. p g y y Fig'. 7. 8O Cs 70 N r 50 ,./ 4-0 / 1 SO! Discussion of Restdts. The most striking conclusion to be drawn from these results is that there are two maxima in the sensitiveness- frequency curve, instead of but one maximum, as indicated bv tile theory and by tile experiments of Pohl and Prino4sheim. In the Pohl and Pringsheim curve in fig. 1 we see that the "normal " photoelectric effec L represented by a b c d, reaches a maximmn similar to that reached by the "selective" effect b c d, but farther out in the ultra-violet. Experiments with RSntgen rays indicate that the "normal" effect does net increase indefinitely with the frequency. The fact that we were able to discover this second maximum was due to the use of very electropositive metals and to the sensitiveness of onr energy calibration system, which enabled the investigation to be carried further into the ultra-violet than ever before, 565 the Photoelectric E]ect. An examination of the curves for sodium shows that the first maximum (a) decreases more rapidly as "fatigue" proceeds than does the second maximum (b). This is what we should expect, since the electrons liberated by light in the region of the second maximum possess much greater initial velocities, and are therefore better able to penetrate the surface layer of oxide. Evidently, if measurements could have been made immediately after the sodium had been distilled, the sensitiveness at (a) would have been con- siderably greater than at (b). y g ( ) In the case of platinum, fig. 4~ both maxima lie beyond the region in which measurements are possible. The value of Xo, about 280/~/z, agrees well with the value 270 ~tk calculated theoretically. Downloaded by [UQ Library] at 22:31 04 June 2016 y In fig. 5 for aluminimn the first maximum only appears. It is at a wave-length Xm~x.----220/t/t, whereas the theory would put it at 240/~/z. A possible explanation of this dis- crepancy is suggested by the fact that the maximum shifts to the right as photoelectric fatigue proceeds, as is seen by comparing curves 1 and 3. This is consistent with the fact, mentioned previously, thai metal surfaces become more electronegative and assume properties of more electronegative metals as fatigue takes place. Since the aluminium was in contact with the air for a short while before making the test, it is certain that the true maximum for aluminimn is at a wave-length longer than 220 I~tL. p Phil. Mag. S. 6. u 26. No. 154. Oct. 1913. Discussion of Restdts. If it should lie at a wave-length longer than about 550 tq, it is also possible that it would be masked by the low dispersive power of the instrument in this region. Thus the left end of the curve is unsatisfactory. The second maximum (b), however, is shown very distinctly and is at a wave-length of about 250/q,. 7o 65 5O ,33 ~ 40 30 o-,.o c_9 ~._ rio r'j5 0 ____, 700 I i i t ! t 1 I I I ] -'--- l :Fig. 8. l 4oo 200 A Downloaded by [UQ Library] at 22:31 04 June 2016 that the maximum lies beyond the range of the illuminator and the mercury arc. If it should lie at a wave-length longer than about 550 tq, it is also possible that it would be masked by the low dispersive power of the instrument in this region. Thus the left end of the curve is unsatisfactory. The second maximum (b), however, is shown very distinctly and is at a wave-length of about 250/q,. Discussion of Restdts. Whether the entire discrepancy may be explained in this way is no~ known. The value of X0, 360/z/+, agrees exactly with ~hat calculated from the theory. y The fatigue in the case of sodimn, fig. 5, was less than in the case of almninimn, because much greater pains were taken to secure a good vacuum. There is no certain evidence of any shift of the maxima with time. The first maximmn (a) is at wave-length 360/zt~, which agrees almost exactly with the theory. The second maximum (b) is at wave-length 227/~/~. Because of the wide slit and strip, and consequent overlapping in the visible spectrum, Xo cannot be accurately determined from these curves, and it appears larger than it really is. It is probably not far from the calculated value X0=550 t+/~. The large overlapping in this spectral region is shown in fig. 8 (p. 566), which represents the dispersive power of the instrument at different wave-lengths. There is very little overlapping of spectral lines for wave-lengths less than 300/~/z. But the overlapping is very considerable in the visible spectrmn. 2 Q 566 On the Plwtoeleetric Effect 566 On the Plwtoele The most serious effect of thi case of cmsium, fig. 7. Here t of the first maximum, which the in the red or infra-red part of 7o 65 5O ,33 ~ 40 30 o-,.o c_9 ~._ rio r'j5 0 ____, 700 I i i t ! t 1 I I I ] -'--- l :Fig. l A that the maximum lies beyond t and the mercury arc. If it longer than about 550 tq, it is a masked by the low dispersive this region. Thus the left end The second maximum (b), howe and is at a wave-length of abou Conclus The experimental results ind y y 566 The most serious effect of this overlapping appears in the case of cmsium, fig. 7. Here there is little or no evidence of the first maximum, which theory and analogy would place in the red or infra-red part of the spectrum It is possible 7o 65 5O ,33 ~ 40 30 o-,.o c_9 ~._ rio r'j5 0 ____, 700 I i i t ! t 1 I I I ] -'--- l :Fig. 8. l 4oo 200 A that the maximum lies beyond the range of the illuminator and the mercury arc. Conclusion. The experimental results indicate that the photoelectric sensitiveness is the same function of the frequency for all metals, the difference being that the curve is shift6d bodily out to the region of short wave-lengths for the electronegative metals. This function is evidently not that expressed by equation (6). Equation (6), however, accurately predicts the values of Xo and )~m~, and seems to be at least a part of the truth. It will be remembered that earlier in the paper reference was made to the fact that equation (6) is only one solution of the theoretical equations, and that it is probably not the only or the complete solution. The results of this investigation point 567 On Ionization and Wireless Telegraphj. to the existence of another term in the solution. If this is true, the real expression of the theory should not be equation (6), but an equation in which N equals the sum of two terms. The first term would be fhe right member of equation (6), or something very similar to it, and would provide the first maximum and account for the "selective" effect. The second term would provide the second maximmn and account for the "normal" effect. The discovery of the equation of the relation between frequency and sensitiveness would be of practical as well as theoretical importance, since it would render possible the use of photoelectric cells as the most sensitive of spectrophotometers. We have not succeeded as yet in discovering such a solution of the theoretical equations. Downloaded by [UQ Library] at 22:31 04 June 2016 q We are glad to take this opportunity to express our thanks to Professor Augustus Trowbridge for many valuable suggestions with regard to the adjustment and use of the galvanometer. Palmer Physical Laboratory, Princeton, N.J. Palmer Physical Laboratory, Princeton, N.J. x LVI. The L:~hct (lionization of Air on Electrical Oscillation., and its bearing on Long-Distance Wireless Telegraphy. B# EDwlx 1-I. BAR'ro~, }).Sc., t~:R.S.E., t)rofessor of Experimental Physics, and WALTER B. KILnY~ B.Sc., " 1851 E%hibltion " Research Bursar, University College~ ,cVottingham . Communicated by the Authors. t Prec. Roy. 8oc. A. vol. ixxxvii, pp. 79299 (1912). 2Q2 * Communicated by the Authors. t Prec. Roy. 8oc. A. vol. ixxxvii, pp. 79299 (1912). * Communicated by the Authors. P R 8 A l i ii * Communicated by the Authors. P R 8 A l i ii 79299(1912) [Plate XI.] I I N his stimulating address before the physical and engi- neering sections of the British Association at Dundee on Sept. 6, 1912, Dr. Fleming dealt with the various unsolved problems of long-distance wireless telegraphy. p g g p y One of the poims of outstanding difficulty then commented upon was the mechanism of the propagation of the eether radiatiou round the curvature of the earth in those cases where about a quarter of its circumference is passed over. q p In the subsequent discussion Dr. Eecles brought forward his theory t, which attributes the bending of the waves round the earth to a higher velocity of propagation in the upper regions of the atmosphere, owing to the ionization there present. This increased speed of propagation was
https://openalex.org/W2745416706	https://hal.sorbonne-universite.fr/hal-01585564/file/s41598-017-09721-0.pdf	English	null	Neuroprotective effect of kinin B1 receptor activation in acute cerebral ischemia in diabetic mice	Scientific reports	2,017	cc-by	9,023	To cite this version: Dorinne Desposito, Georges Zadigue, Christopher Taveau, Clovis Adam, François Alhenc-Gelas, et al.. Neuroprotective effect of kinin B1 receptor activation in acute cerebral ischemia in diabetic mice. Scientific Reports, 2017, 7, pp.9410. 10.1038/s41598-017-09721-0. hal-01585564 Distributed under a Creative Commons Attribution 4.0 International License Neuroprotective effect of kinin B1 receptor activation in acute cerebral ischemia in diabetic mice Dorinne Desposito1,2,3, Georges Zadigue1, Christopher Taveau1,2,3, Clovis Adam4, François Alhenc-Gelas1,2,3, Nadine Bouby1,2,3 & Ronan Roussel1,5,6 Received: 2 February 2017 Accepted: 10 July 2017 Published: xx xx xxxx Received: 2 February 2017 Accepted: 10 July 2017 Published: xx xx xxxx Activation of the kallikrein-kinin system enhances cardiac and renal tolerance to ischemia. Here we investigated the effects of selective agonists of kinin B1 or B2 receptor (R) in brain ischemia- reperfusion in diabetic and non-diabetic mice. The role of endogenous kinins was assessed in tissue kallikrein deficient mice (TK−/−). Mice underwent 60min-middle cerebral artery occlusion (MCAO), eight weeks after type 1-diabetes induction. Treatment with B1R-, B2R-agonist or saline was started at reperfusion. Neurological deficit (ND), infarct size (IS), brain water content (BWC) were measured at day 0, 1 and 2 after injury. MCAO induced exaggerated ND, mortality and IS in diabetic mice. B2R- agonist increased ND and mortality to 60% and 80% in non-diabetic and diabetic mice respectively, by mechanisms involving hemodynamic failure and renal insufficiency. TK−/− mice displayed reduced ND and IS compared to wild-type littermate, consistent with suppression of B2R activity. B1R mRNA level increased in ischemic brain but B1R-agonist had no effect on ND, mortality or IS in non-diabetic mice. In contrast, in diabetic mice, B1R-agonist tested at two doses significantly reduced ND by 42–52% and IS by 66–71%, without effect on BWC or renal function. This suggests potential therapeutic interest of B1R agonism for cerebral protection in diabetes. Acute brain ischemia secondary to cerebral artery occlusion is a major cause of mortality or permanent disability. Risk of ischemic stroke is increased in diabetic patients and prognosis is poorer1, 2. Cerebral artery occlusion causes acute (minutes to hours) and delayed (hours to days or weeks) injury cascades, both implicating mul- tiple pathogenic factors like thrombosis, neuron stunning or necrosis, brain oedema and inflammation3, 4. The complexity of mechanisms involved in brain damage explains in part that there is still no clinically effective neuroprotective treatment besides revascularization. The kallikrein-kinin system (KKS) is implicated in physio- logical vasodilatation, exerts antithrombotic and profibrinolytic actions and reduces oxidative stress in different organs5–8. KKS protects against cardiac and renal damage in the setting of acute ischemia secondary to arterial occlusion. Inhibition of KKS aggravates cardiac and renal ischemic lesions while activation of kinin receptors enhances cardiac tolerance to ischemia and reperfusion9–12. HAL Id: hal-01585564 https://hal.sorbonne-universite.fr/hal-01585564v1 Submitted on 11 Sep 2017 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License www.nature.com/scientificreports www.nature.com/scientificreports Received: 2 February 2017 Accepted: 10 July 2017 Published: xx xx xxxx Neuroprotective effect of kinin B1 receptor activation in acute cerebral ischemia in diabetic mice Kinins are generated by proteolytic cleavage of pro- tein precursors, kininogens, by tissue kallikrein (TK) and are mainly inactivated in the circulation by the angio- tensin I-converting enzyme (ACE/kininase II)5. Kinins, activate two receptor subtypes: B1 (B1R) and B2 (B2R). All components of KKS have been identified in brain tissue from rodents and humans3, 13–18. B1R gene expression is low in the brain under normal condition, but it is upregulated by inflammation and ischemia19, 20. By contrast, B2R is constitutively present in different brain structures and in cerebral arteries and microvessels18, 21. yf Role of kinins in brain ischemia has been addressed so far by performing pharmacological blockade of B1R or B2R in rodents or studying mice genetically deficient in either B1 or B2 receptor. However, these studies have pro- duced conflicting results3. Some studies have shown that B2R blockade reduced infarct size and neuronal damage after transient middle cerebral artery occlusion (MCAO)19, 22–24 but other suggested that inactivation of this recep- tor has no effect or even aggravates ischemic brain damage20, 25–27. It has also been reported that pharmacological blockade or genetic inactivation of B1R confer neuroprotection in mice20. Single receptor inactivation however is well known to result in induction, coupling and activation of the remaining alternate receptor that can explain, 1INSERM U 1138, Cordeliers Research Center, Paris, France. 2Paris Descartes University, Paris, France. 3Pierre et Marie Curie University, Paris, France. 4Anatomopathology Department, Kremlin-Bicêtre Hospital, Paris, France. 5Denis Diderot University, Paris, France. 6Diabetology, Endocrinology and Nutrition Department, DHU FIRE, Bichat Hospital, AP-HP, Paris, France. Nadine Bouby and Ronan Roussel contributed equally to this work. Correspondence and requests for materials should be addressed to N.B. (email: nadine.bouby@crc.jussieu.fr) Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 1 www.nature.com/scientificreports/ Figure 1. B1R mRNA level in brain increased 24 h after transient MCAO. Kinin receptor mRNA levels in NonDiab mice brain measured at day 0 (D0), 1 (D1), 3 (D3) and 7 (D7) after transient MCAO, by RT-qPCR. Data were normalized to 18 S rRNA. Values are mean ± SEM, n = 5/group. Figure 1. B1R mRNA level in brain increased 24 h after transient MCAO. Kinin receptor mRNA levels in NonDiab mice brain measured at day 0 (D0), 1 (D1), 3 (D3) and 7 (D7) after transient MCAO, by RT-qPCR. Data were normalized to 18 S rRNA. Values are mean ± SEM, n = 5/group. Neuroprotective effect of kinin B1 receptor activation in acute cerebral ischemia in diabetic mice at least in part, the effects observed making data interpretation ambiguous8, 9, 28, 29. Moreover, the effect of kinins and their receptors may depend in part on the stage of infarct development27. Therefore, the role of kinins, B1R and B2R in brain ischemia and the potential therapeutic interest of pharmacological manipulation of KKS need to be further documented by using new experimental approaches. Involvement of KKS in cerebral ischemia in the setting of diabetes has not been studied, except for a very recent report in the rat based on receptor inhibition30.h at least in part, the effects observed making data interpretation ambiguous8, 9, 28, 29. Moreover, the effect of kinins and their receptors may depend in part on the stage of infarct development27. Therefore, the role of kinins, B1R and B2R in brain ischemia and the potential therapeutic interest of pharmacological manipulation of KKS need to be further documented by using new experimental approaches. Involvement of KKS in cerebral ischemia in the setting of diabetes has not been studied, except for a very recent report in the rat based on receptor inhibition30.h The aim of the present study was to address the role of KKS in cerebral ischemia in non-diabetic (NonDiab) and diabetic (Diab) mice, by using gain and loss of function approaches. We firstly probed the role of each recep- tor by performing pharmacological activation using potent, peptidase resistant synthetic agonists, selective B1R (B1R-ag) or B2R (B2R-ag) in a model of transient MCAO. We then addressed the role of endogenously produced kinins acting through both receptors by studying a genetic mouse model of TK and kinin deficiency.hi i The study shows that a) MCAO induced bradycardia, mild hypotension, neurological deficit, and resulted in partial brain infarction. Neurological deficit, mortality and infarct size were all increased in diabetic mice compared to non-diabetic mice; b) B2R activation increased neurological deficit and mortality; c) B1R activation had no effect in non-diabetic mice but in diabetic mice a B1R agonist, tested at two different dosages, reduced neurological deficit and infarct size; d) TK deficiency reduced neurological deficit and infarct size in non-diabetic mice but had no effect in diabetic mice. The data are consistent with a deleterious role of kinins, through B2R activation in cerebral ischemia. They however show that in diabetic condition B1R signalling is neuroprotective. Resultsf Effects of B1R or B2R agonists (ag) on cerebral ischemia were investigated, in both non-diabetic (NonDiab) and diabetic (Diab) mice. Diabetic mice were studied 8 weeks after streptozotocin injections. Focal cerebral ischemia (Isch) was induced via a transient intraluminal filament middle cerebral artery occlusion method. Non-ischemic (NonIsch) mice underwent sham operation. Chronic treatment with B1R-, B2R-agonist or saline was started at reperfusion, using osmotic minipumps implanted s.c. and lasted two days. Effect of ischemia on B1R and B2R mRNA levels. B1R and B2R mRNAs were both detected in affected brain tissue. B2R mRNA level was not influenced by transient MCAO and did not change at day 1, 3 and 7 after ischemia in our model (Fig. 1). By contrast B1R mRNA level increased 2.35 fold (p < 0.05) 24 h after transient MCAO before returning to basal values (Fig. 1). B1R and B2R mRNA levels were not influenced by diabetes (data not shown). Effect of transient cerebral ischemia on neurological score and infarct size. Transient MCAO did not affect body weight, glycaemia and plasma creatinine in either NonDiab or Diab mice (data not shown). In NonDiab mice, transient MCAO induced bradycardia (Isch: 382 ± 15 bpm vs NonIsch: 655 ± 14 bpm, p < 0.01) and a tendency to hypotension (Isch: 101 ± 2 mmHg vs NonIsch: 113 ± 2 mmHg, p = 0.055), measured at 24 h. At 24 and 48 h after transient MCAO, the neurological score reflected severe impairment of sensorimotor function in Isch mice versus NonIsch mice (p < 0.01) (Fig. 2a). Infarction was observed in cerebral cortex and striatum (TTC staining: 24.1 ± 2.0%, p < 0.01 vs NonIsch). Mortality remained low (Fig. 2b).i g p ) y ( g ) Diabetes increased neurological deficits (p < 0.01) at 48 h when compared to NonDiab mice (Fig. 2a,b). Infarct size was increased by 55% in Diab mice compared to NonDiab mice (TTC staining: 37.2 ± 2.7 vs 24.0 ± 2.8%, p < 0.01, Fig. 2c). Effect of B2R agonist treatment after transient cerebral ischemia. B2R-ag treatment significantly increased mortality to 60% in NonDiab mice and 80% in Diab mice 48 h after ischemia (both p < 0.05 compared to saline, Fig. 3). Mortality occurred mainly after 24 h. This did not allow gathering enough data concerning neu- rological deficits and histological lesions at two days for these groups. Resultsf i In NonDiab ischemic mice, B2R-ag treatment did not influence bradychardia (Isch: 382 ± 15 bpm, Isch + B2R-ag 720 nmol/kg.day−1: 460 ± 17 bpm, Isch + B2R-ag 240 nmol/kg.day−1: 404 ± 20 bpm, both NS) and aggravated hypotension at the two different dosages used (Isch: 101 ± 2 mmHg, Isch + B2R-ag 720 nmol/ Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 2 www.nature.com/scientificreports/ k d 1 8 ±2 H I h B2R 240 l/k d 1 82±6 H b h 0 01 N I h) Figure 2. Diabetes increased neurological impairment, mortality and infarct size 48 h after transient MCAO. (a) Neurological score (0–30) measured in NonDiab and Diab mice at day 0, 1 and 2 after transient MCAO or sham-operation. (b) Survival curve of NonDiab and Diab mice after transient MCAO. Numbers in parenthese refer to surviving/operated animals. (c) Ischemic area measured at 48 h after transient MCAO or sham- operation using TTC staining. Values are mean ± SEM, n = 8–14/group. p < 0.01 vs corresponding non ischemic group, other statistics shown on figure. k d 1 h l k d 1 Figure 2. Diabetes increased neurological impairment, mortali (a) Neurological score (0–30) measured in NonDiab and Diab m sham-operation. (b) Survival curve of NonDiab and Diab mice refer to surviving/operated animals. (c) Ischemic area measured operation using TTC staining. Values are mean ± SEM, n = 8–1 ischemic group, other statistics shown on figure. Figure 2. Diabetes increased neurological impairment, mortality and infarct size 48 h after transient MCAO. (a) Neurological score (0–30) measured in NonDiab and Diab mice at day 0, 1 and 2 after transient MCAO or sham-operation. (b) Survival curve of NonDiab and Diab mice after transient MCAO. Numbers in parentheses refer to surviving/operated animals. (c) Ischemic area measured at 48 h after transient MCAO or sham- operation using TTC staining. Values are mean ± SEM, n = 8–14/group. p < 0.01 vs corresponding non ischemic group, other statistics shown on figure. kg.day−1: 87 ± 2 mmHg, Isch + B2R-ag 240 nmol/kg.day−1: 82 ± 6 mmHg, both p < 0.01 vs NonIsch), measured at day 1 after MCAO. The treatment severely increased plasma creatinine, in surviving animals at day 2 (Isch: 17.5 ± 4.2 μmol/l vs Isch + B2R-ag 720 nmol/kg.day−1: 72.8 ± 19.1μmol/l, p < 0.05). B2R-ag treatment did not affect body weight and glycaemia at the two dosages (data not shown). Resultsf In NonIsch mice, B2R-ag treatment did not influence blood pressure (NonIsch + B2R-ag: 102 ± 2 mmHg, NS), heart rate (667 ± 23 bpm, NS) or plasma creatinine (NonIsch + B2R-ag: 15.4 ± 0.5 μmol/l, NS). Effect of B1R agonist treatment in mice after transient cerebral ischemia. Neurological defi- cit and mortality was not influenced by B1R-ag treatment (720 nmol/kg.day−1) in NonDiab mice (Fig. 4a,b). Treatment had no effect on body weight, glycaemia, plasma creatinine, blood pressure and heart rate when Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 3 www.nature.com/scientificreports/ Figure 3. B2R-ag treatment was associated with increased mortality after transient MCAO. (a) Survival curve of NonDiab mice treated with B2R-ag (720 or 240 nmol/kg.day−1) or non-treated (Isch). (b) Survival curve of Diab mice treated with B2R-ag (720 nmol/kg.day−1) or non-treated (Diab-Isch). Numbers in parentheses refer to surviving/operated animals. Values are mean ± SEM, n = 8–14/group. p < 0.05 refers to corresponding ischemic group. Figure 3. B2R-ag treatment was associated with increased mortality after transient MCAO. (a) Survival curve of NonDiab mice treated with B2R-ag (720 or 240 nmol/kg.day−1) or non-treated (Isch). (b) Survival curve of Diab mice treated with B2R-ag (720 nmol/kg.day−1) or non-treated (Diab-Isch). Numbers in parentheses refer to surviving/operated animals. Values are mean ± SEM, n = 8–14/group. p < 0.05 refers to corresponding ischemic group. compared to saline (data not shown). B1R-ag treatment had no effect on infarct volume and histological score at day 2 after transient MCAO (Fig. 4c).f yt g By contrast, in Diab mice, B1R-ag tested at two different dosages (240 or 720 nmol/kg.day−1) improved neu- rological score compared to saline treated Diab-Isch group (Fig. 5a,b). The clinical beneficial effect of B1R-ag was associated with a decrease of infarct size by 71 and 66% at the two dosages respectively (both p < 0.01) (Fig. 5c). These results were confirmed by the histological score (Fig. 5d). B1R-ag did not induce mortality (Fig. 5b) and did not increase plasma creatinine (Diab-Isch: 17.3 ± 0.3 μmol/l vs Diab-Isch + B1R-ag: 16.7 ± 1.1 μmol/l, NS). Moreover, B1R-ag at 720 nmol/kg.day−1 did not affect BWC 24 h after cerebral ischemia-reperfusion compared to saline-treated Diab-Isch group (Diab-Isch: 83.2 ± 0.82% vs Diab-Isch + B1R-ag: 82.3 ± 0.27%, NS). Effect of TK deficiency on transient cerebral ischemia. In NonDiab condition, TK−/− mice displayed better post MCAO outcome compared to TK+/+ and TK+/− mice. Resultsf Indeed, TK deficiency improved neurological score and significantly decreased infarct size at 48 hours after transient MCAO (Fig. 6a–c). In Diab mice however, TK deficiency had no effect on cerebral ischemia-reperfusion outcome (Fig. 6d–f). Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 Discussion In the present study, we evaluated effect of kinin signalling in mice submitted to transient focal cerebral ischemia. We studied both diabetic and non-diabetic mice because of the deleterious effect of diabetes on outcome of cerebral ischemia in man and also of diabetes effect on cellular signalling pathways2, 11. We considered the two kinin receptor subtypes, B1 or B2 and probed their effects by selectively activating these receptors, pharmacolog- ically. We then addressed the role of endogenously produced kinins by studying mice deficient in TK, the main kinin-forming enzyme. We show that B2R activation increases mortality after transient MCAO in non-diabetic or diabetic mice, by mechanisms that may involve peripheral hemodynamic failure. B1R signalling on the other hand is not detrimental and its effect is strongly influenced by diabetes. While B1R activation has no effect in non-diabetic mice it reduces brain infarction and improves MCAO outcome in diabetic mice. Data obtained in TK deficient mice are consistent with disappearance of B1 and B2 receptors activity and suggest a role of endog- enously produced kinins in cerebral tolerance to ischemia. Finding that treatment with a selective B1R agonist, at different dosages, reduces brain infarct volume and improve neurological deficit in diabetic mice may have therapeutic implication. p p Previous studies indicated that B1R and B2R are present in the brain from various species including man3, 13, 14, 16–18. We show that genes for both receptors are expressed in murine brain and their expression level is not altered by diabetes. Whereas B2R mRNA was not influenced by ischemia, B1R mRNA level increased in the ischemic Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 4 www.nature.com/scientificreports/ Figure 4. B1R-ag had no effect on neurological impairment, mortality and infarct size 48 h after transient MCAO in NonDiab mice. (a) Neurological score (0–30) measured in NonDiab mice, treated with B1R-ag (720 nmol/kg.day−1) (grey bars) or saline (black bars), at day 0, 1 and 2 after transient MCAO or sham-operation. (b) Survival curve of NonDiab mice treated with B1R-ag or non-treated after transient MCAO. Numbers in parentheses refer to surviving/operated animals. (c) Ischemic area measured at 48 h after transient MCAO using TTC staining. Values are mean ± SEM, n = 9–14/group. p < 0.01 vs non ischemic group. Figure 4. B1R-ag had no effect on neurological impairment, mortality and infarct size 48 h after transient MCAO in NonDiab mice. Discussion (a) Neurological score (0–30) measured in NonDiab mice, treated with B1R-ag (720 nmol/kg.day−1) (grey bars) or saline (black bars), at day 0, 1 and 2 after transient MCAO or sham-operation. (b) Survival curve of NonDiab mice treated with B1R-ag or non-treated after transient MCAO. Numbers in parentheses refer to surviving/operated animals. (c) Ischemic area measured at 48 h after transient MCAO using TTC staining. Values are mean ± SEM, n = 9–14/group. p < 0.01 vs non ischemic group. hemisphere after transient MCAO. This is consistent with previous studies19, 20 and extends to the brain observa- tion of induction of B1R gene expression by ischemia made in the heart and kidney9, 10, 31. Increase in B1R mRNA in the ischemic brain was transient, peaking at 24 h and then subsiding, despite brain infarction. This is similar to what has been observed in the ischemic heart suggesting that acute phase secretion of proinflammatory cytokines acting through MAP-kinases and NFkappaB activation but not post-necrosis tissue remodelling and fibrosis development is involved in B1R induction in the ischemic brain31, 32. hemisphere after transient MCAO. This is consistent with previous studies19, 20 and extends to the brain observa- tion of induction of B1R gene expression by ischemia made in the heart and kidney9, 10, 31. Increase in B1R mRNA in the ischemic brain was transient, peaking at 24 h and then subsiding, despite brain infarction. This is similar to what has been observed in the ischemic heart suggesting that acute phase secretion of proinflammatory cytokines acting through MAP-kinases and NFkappaB activation but not post-necrosis tissue remodelling and fibrosis development is involved in B1R induction in the ischemic brain31, 32. p Transient MCAO resulted in downstream brain infarction and, clinically, severe neurological impairment. Bradychardia is believed to result from reflex activation of baroreflex loop33 and may contribute to hemody- namic instability with mild decrease in blood pressure. Activation of B2R signalling pathways during reperfusion Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 5 www.nature.com/scientificreports/ Figure 5. B1R-ag, at two different dosages (720 or 240 nmol/kg.day−1), reduced neurological score and infarct size at 48 h in Diab mice. (a) Neurological score (0–30) measured in Diab mice treated with B1R-ag, 720 nmol/ kg.day (grey bars) or 240 nmol/kg.day−1 (hatched bars) or with saline (white bars), at day 0, 1 and 2 after transient MCAO. (b) Survival curve of Diab mice after transient MCAO. Discussion Thus the effect of B2R activation on blood pressure, renal function and mortality in cerebral ischemia-reperfusion is peculiar to this experimental setting. One can speculate that in presence of inappropriate baroreflex activation triggered by brain ischemia B2R agonist administration induces hypotension that may, if becoming severe, even transiently, result in renal failure and death. This phenomenon would likely not occur in man where sympathetic activation raising blood pressure is a major effect of acute brain ischemia36.fi induced mortality. While transient MCAO did not result in significant mortality, more than 50% of B2R agonist treated mice died within two days, mainly after one day. Analysis of the brain in surviving animals did not show larger brain infarction or evidence for aggravated cerebral oedema, despite known effects of kinins on cerebral oedema19, 34, compared to saline treated animals (data not shown). However, these observations are difficult inter- preting in term of causality, or lack of it, between B2R activity and brain damage and no conclusion can be made in absence of data concerning brains of deceased animals. Data obtained in TK deficient mice however indirectly suggest that B2R activity may aggravate brain infarction (see below). The cause of death of B2R agonist treated animals remains undocumented but may be related, at least in part to peripheral hemodynamic failure. These animals indeed displayed aggravated hypotension when compared to saline treated animals and had severe renal insufficiency. These two effects were not observed during B2R agonist treatment at the same dosage and by the same route in mice non-submitted to cerebral ischemia or in other experimental settings, including in diabetes, ruling out renal toxicity of the B2R agonist12, 35 (unreported data). Thus the effect of B2R activation on blood pressure, renal function and mortality in cerebral ischemia-reperfusion is peculiar to this experimental setting. One can speculate that in presence of inappropriate baroreflex activation triggered by brain ischemia B2R agonist administration induces hypotension that may, if becoming severe, even transiently, result in renal failure and death. This phenomenon would likely not occur in man where sympathetic activation raising blood pressure is a major effect of acute brain ischemia36.fi jf B1R activation had no effect on mortality and brain infarction in non-diabetic animals. Interestingly, TK defi- ciency reduced infarct size and improved neurological defects in the non-diabetic animals. This result suggests that endogenously produced kinins are involved in brain damage during ischemia-reperfusion. Discussion Numbers in parentheses refer to surviving/operated animals. (c) Ischemic area measured at 48 h after transient MCAO using TTC staining. (d) Histological score (0–3) measured at 48 h after transient MCAO using haematoxylin and eosin (H&E) staining. Values are mean ± SEM, n = 6–13/group. p < 0.05, p < 0.01 vs non ischemic group. Other statistics shown on figure. Figure 5. B1R-ag, at two different dosages (720 or 240 nmol/kg.day−1), reduced neurological score and infarct size at 48 h in Diab mice. (a) Neurological score (0–30) measured in Diab mice treated with B1R-ag, 720 nmol/ kg.day (grey bars) or 240 nmol/kg.day−1 (hatched bars) or with saline (white bars), at day 0, 1 and 2 after transient MCAO. (b) Survival curve of Diab mice after transient MCAO. Numbers in parentheses refer to surviving/operated animals. (c) Ischemic area measured at 48 h after transient MCAO using TTC staining. (d) Histological score (0–3) measured at 48 h after transient MCAO using haematoxylin and eosin (H&E) staining. Values are mean ± SEM, n = 6–13/group. p < 0.05, p < 0.01 vs non ischemic group. Other statistics shown on figure. induced mortality. While transient MCAO did not result in significant mortality, more than 50% of B2R agonist treated mice died within two days, mainly after one day. Analysis of the brain in surviving animals did not show larger brain infarction or evidence for aggravated cerebral oedema, despite known effects of kinins on cerebral oedema19, 34, compared to saline treated animals (data not shown). However, these observations are difficult inter- preting in term of causality, or lack of it, between B2R activity and brain damage and no conclusion can be made in absence of data concerning brains of deceased animals. Data obtained in TK deficient mice however indirectly suggest that B2R activity may aggravate brain infarction (see below). The cause of death of B2R agonist treated animals remains undocumented but may be related, at least in part to peripheral hemodynamic failure. These animals indeed displayed aggravated hypotension when compared to saline treated animals and had severe renal insufficiency. These two effects were not observed during B2R agonist treatment at the same dosage and by the same route in mice non-submitted to cerebral ischemia or in other experimental settings, including in diabetes, ruling out renal toxicity of the B2R agonist12, 35 (unreported data). Discussion Data obtained with subtype selective pharmacological receptor agonists suggest that effect of endogenous kinins can be ascer- tained to B2R but not B1R activation.h Diabetes increased infarct size and enhanced neurological impairment, consistent with clinical studies. The effect of the kallikrein-kinins system in brain ischemia in the setting of diabetes has not been studied until now. Like in non-diabetic animals B2R activation enhances mortality. But interestingly, a B1R agonist, when admin- istered at time of reperfusion, improves neurological deficit and decreases brain infarct size by more than 60% in diabetic mice submitted to transient focal cerebral ischemia. No adverse effect on mortality or renal function Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 6 www.nature.com/scientificreports/ ntificreports/ Figure 6. TK deficiency decreased neurological impairment and infarct size 48 h after transient MCAO in NonDiab mice but had no effect in Diab mice. (a,d) Neurological score (0–30) measured in TK-deficient mice at day 0, 1 and 2 after transient MCAO. (b,e) Survival curve of TK-deficient mice after transient MCAO. Numbers in parentheses refer to surviving/operated animals. (c,f) Ischemic area measured at 48 h after transient MCAO using TTC staining. a, b and c: NonDiab mice; d, e and f: Diab mice. TK+/+: white bars, TK+/−: grey bars, TK−/−: black bars. Data are mean ± SEM, n = 8–10/group. p < 0.01 vs non ischemic group. Other statistics shown on figure. Figure 6. TK deficiency decreased neurological impairment and infarct size 48 h after transient MCAO in fi Figure 6. TK deficiency decreased neurological impairment and infarct size 48 h after transient MCAO in NonDiab mice but had no effect in Diab mice. (a,d) Neurological score (0–30) measured in TK-deficient mice at day 0, 1 and 2 after transient MCAO. (b,e) Survival curve of TK-deficient mice after transient MCAO. Numbers in parentheses refer to surviving/operated animals. (c,f) Ischemic area measured at 48 h after transient MCAO using TTC staining. a, b and c: NonDiab mice; d, e and f: Diab mice. TK+/+: white bars, TK+/−: grey bars, TK−/−: black bars. Data are mean ± SEM, n = 8–10/group. p < 0.01 vs non ischemic group. Other statistics shown on figure. gi y g pt NonDiab mice but had no effect in Diab mice. (a,d) Neurological score (0–30) measured in TK-deficient mice at day 0, 1 and 2 after transient MCAO. (b,e) Survival curve of TK-deficient mice after transient MCAO. Materials and Methods Animals. Experiments were performed in male C57/BL6 mice (JanvierLabs, France or in house strain). TK deficient mice were generated in our laboratory by disruption of the TK gene as previously described46. Littermate wild type, homozygous and heterozygous TK deficient mice were obtained by heterozygous crossing46, 47. All mice were housed with a 12 h light/dark cycle and had free access to standard mice chow and water. All experimental procedures were performed in accordance with the Directive 2010/63/eu of the European Union. The study has received approval from the Ethical Committee Charles Darwin (CEEACD/N°5). Reporting of this work complies with ARRIVE guidelines. Murine model of type 1 diabetes. Diabetes (Diab) was induced in ten weeks-old mice by 5 daily i.p. injec- tions of streptozotocin (STZ) (Sigma-Aldrich, France) (50 mg/kg body weight in 0.05 mol/L sodium citrate, pH 4.5)12. After 8 weeks of established diabetes (fasting blood glucose >250 mg/dl), transient focal cerebral ischemia was induced as described below. Transient focal cerebral ischemia. Transient focal cerebral ischemia was induced by MCAO using the intraluminal filament technique previously described48. Briefly, mice were anesthetised with 3.5% isoflurane in an anaesthetic chamber and maintained during surgery at 2% isoflurane using a rodent mask. Body temperature was maintained at 37 ± 0.5 °C with a heating blanket throughout the entire experimental procedure. MCAO was car- ried out for 60 min by inserting a calibrated monofilament (Doccol Corporation, USA) according to body weight via the right external carotid artery into the internal carotid artery to block the origin of the MCA. Sham-operated controls (NonIsch) were treated similarly to the ischemic (Isch) mice, but the filament was not inserted. After sur- gery and before being returned to cages, animals were placed for 4 hours in a heating incubator at 37 °C. B1R or B2R agonist treatments. Chronic treatment with the selective B1R agonist SarLys[Hyp3, Igl5, DPhe8]desArg9-bradykinin (B1R-ag) or the selective B2R agonist [Hyp(3),Thi(5),(N)Chg(7),Thi(8)]-bradykinin (B2R-ag)49, 50 was started at reperfusion, using osmotic minipumps implanted s.c. (Alzet 1007D, Charles River Laboratories, France). These compounds are resistant to kininase hydrolysis. Two different dosages chosen from previous studies based on therapeutic efficiency and lack of hypotensive effect, 720 nmol/kg.day and 240 nmol/ kg.day were used12, 35. Control mice received saline infusion. Experimental groups. Several sets of experiments were performed to analyse the effects of kinin receptor agonist treatments on mortality, neurological deficit and infarction volume (n = 8–14/group). www.nature.com/scientificreports/ and B1R coupling is activated11. Effect of B2R on mortality in diabetic animals may be due to peripheral rather than cerebral action, as discussed above. A recent report by Sang et al.30 suggests that acute administration of a B1R antagonist in type 2 diabetic rats submitted to MCAO reduces brain infarction. Our data may not appear consistent with this observation. However, result of the Sang et al. study is difficult interpreting given that the compound used as B1R antagonist is an analog of human rather than rat kinins42–45 and can behave as a partial agonist instead, depending on kinin level.il g p g TK deficiency in diabetic mice did not influence MCAO outcome suggesting that TK is not involved in kinin production in brain of diabetic animals. An alternate explanation is that the lack of effect of TK deficiency results from disappearance of both the beneficial effect of B1R activation and the deleterious effect of B2R activation.i ppiff Our study clarifies to some extent the controversial issue of role of kinin and their receptors in cerebral ischemia-reperfusion. The study show that B2R activation is detrimental in this experimental setting but B1R activation can be beneficial. Documentation of neuroprotective effect of a pharmacological B1R agonist in brain ischemia in diabetic mice can have therapeutic implication. Together with previous studies documenting cardio- protective effect of the B1R agonist in the diabetic and ischemic mouse heart11 and proangiogenic effect in periph- eral ischemia in diabetic mice12, the present observations argue for clinical development of kinin B1R agonist for cardiovascular and cerebral protection in diabetes. Discussion Numbers in parentheses refer to surviving/operated animals. (c,f) Ischemic area measured at 48 h after transient MCAO using TTC staining. a, b and c: NonDiab mice; d, e and f: Diab mice. TK+/+: white bars, TK+/−: grey bars, TK−/−: black bars. Data are mean ± SEM, n = 8–10/group. p < 0.01 vs non ischemic group. Other statistics shown on figure. was observed during B1R agonist treatment. Thus, in diabetic mice B1R activation has neuroprotective effect in cerebral ischemia. Mechanisms remain unclear but may be related to endothelial activation with release of anticlotting, profibrinolytic and vasodilatory agents37–39. Vasodilatation of collateral arteries could improve cer- ebral blood supply. Also, kinins are known to modulate mitochondrial permeability transition pore opening and trigger production of reactive oxygen species, which afford organ protection7, 40, 41. These mechanisms are believed to be operative in the ischemic heart or kidney10, 11. The lack of effect of B1R agonist on brain infarc- tion in non-diabetic animals may appear surprising but the situation is similar to the ischemic heart where B1R activation had no effect in non-diabetic mice while it dramatically reduced infract size in diabetic animals11. Hypothesis put forward for the heart and kidney and based on the well documented balance between B1R and B2R activity8, 9, 28 may also be valid for the brain: when B2R is functional in non-diabetic animals B1R remains uncoupled. However, in the ischemic heart and tentatively brain of diabetic animals, B2R signalling is impaired Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 7 www.nature.com/scientificreports/ Materials and Methods The score was obtained by using a series of behavioural tests including grip test, scotch test, tail suspension test, beam test, wire hang test, circles tests and by assessing comportment into home cage as previously described52–54. Higher score indicate greater functional impairment. Mice were studied in random order in each series and blindly with regard to treatment. Determination of infarct volume. Two days after reperfusion and after neurological score evaluation mice were sacrificed. Brains were rapidly removed and sectioned into six coronal sections, 2 mm thick, using a mice brain matrix. Coronal brain sections were stained by incubation in a 0.5% 2,3,5-triphenyltetrazolium chloride (TTC, Sigma-Aldrich, France) solution for 30 min at 37 °C in the dark55 and fixed in 10% formalin (Sigma-Aldrich, France) for two hours prior to analysis. Photographs of the sections were obtained using digital camera attached to microscope (Nikon SMZ800, Italy). The infarction area, outlined in white, and the entire section area were measured under microscope (Nikon SMZ800, Italy) on the anterior surface of each section in a blinded manner with regard to protocol and treatment, using Image Analyzer Software (ImageJ, NIH). For each section, infarction area was normalized to the whole section area. For each animal, results for six sections were averaged. g Cerebral infarction was also evaluated by using Haematoxylin and eosin staining on the same sections of the brain for confirmation56. Brain sections were fixed in 10% formalin during 24 h, embedded in paraffin, cut into 6-μm section and stained with haematoxylin and eosin according to the manufacturer’s instructions (Sigma-Aldrich, France). Photomicrographs were obtained using digital camera attached to light microscope (Leica DM 4000B and LAS v3.8 software). Histological lesions were assessed in a blinded manner regarding protocol and treatment using a numerical scale of 0–3 for each animal57. Higher score indicate more severe his- tological lesions. Evaluation of brain oedema by measurement of brain water content (BWC). Procedure was per- formed as previously described58, 59. Briefly, mice were killed by decapitation and brains were removed. Ischemic hemisphere was weighed before (wet weight) and after being dried at 110 °C for 24 h (dry weight). BWC was calculated and expressed as follows: BWC (%) = (wet weight − dry weight)/wet weight ×100. Quantification of B1R and B2R mRNA by real-time PCR. Materials and Methods All mice were ≈18 week-old at the time of MCAO. Animals were sacrificed after two days, unless otherwise indicated. Series 1 was dedicated at comparing effect of ischemia-reperfusion in NonDiab and Diab mice. Series 2 and 3 were dedicated at testing effect of B2R agonist at two dosages in NonDiab and Diab mice, respectively. Series 4 and 5 were dedi- cated at testing effect of B1R agonist at two dosages in NonDiab and Diab mice, respectively. The effect of B1R-ag (720 nmol/kg.day−1) on cerebral oedema in Diab mice 24 h after transient MCAO was tested in a separate series (n = 5–6/group).f g p Additional groups of NonDiab mice were dedicated at studying effect of B2R-ag (240 nmol/kg.day−1 or 720 nmol/kg.day−1) on blood pressure. Mice were treated for 24 h after transient MCAO. Neurological score was determined, blood pressure and heart rate was recorded, and animals were sacrificed (n = 5–6/group). i Other groups of mice rendered or not diabetic and submitted to MCAO occlusion or sham operation were used for studying time related effect of brain ischemia-reperfusion on receptor gene expression. Animals were sacrificed at 1, 3 or 7 days after surgery, brain was sampled and kept at −80% until processed for measurement of receptor mRNAs by RT-PCR (n = 5 per group and time point).hfii p y p g p p The effect of TK deficiency on neurological deficit and ischemic volume was tested in NonDiab mice at 48 h after transient MCAO, in TK+/+, TK+/− and TK−/− mice (n = 8–10/group). Same protocol was performed in series dedicated at testing effect of TK deficiency in Diab mice (Diab-TK+/+, Diab- TK−/−, n = 8–9/group). Measurement of blood pressure. Blood pressure was measured by tail-cuff plethysmography (BP-2000 eries II, BIOSEB Instruments, France) in trained animals as previously described51. 8 Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 www.nature.com/scientificreports/ Measurement of plasma creatinine. Plasma creatinine was assessed in blood samples taken at sacrifice using a colorimetric enzymatic assay (automatic analyser Konelab 201, France) (n = 4–5/group). Measurement of plasma creatinine. Plasma creatinine was assessed in blood samples taken at sacrifice using a colorimetric enzymatic assay (automatic analyser Konelab 201, France) (n = 4–5/group). Evaluation of neurological deficits. Neurological deficit was assessed in each animal on a numerical scale of 0–30 before ischemia and at day 1 and/or 2 after ischemia, depending on protocol. Materials and Methods Total RNA was isolated from the whole ipsilateral hemisphere, in occluded or sham operated mice (day 0, 1, 3 and 7), using TRIzol (Invitrogen, France) and reverse transcribed with superscript II reverse transcriptase. The cDNAs were amplified and quantified using TaqMan Universal Master Mix and Assays-on-Demand Gene Expression Probes for gene of B1R and B2R (Applied Biosystems, France) in an ABI PRISM-7000 Sequence Detection System (Applied Biosystems, France), as previously described60. Each sample was tested in triplicate. Data were normalized to 18 S rRNA. Changes in the target gene were calculated by the 2−∆∆CT comparative method for each sample61. Data expression and statistical analysis. Data are expressed as mean ± SEM. Effects of surgical pro- cedure, diabetes and treatments on mortality and infarct size were evaluated by chi2 test or one-way ANOVA. For repeated measurements of neurological score two-way ANOVA was used. ANOVA was followed by Tukey multiple comparison tests. For comparison of mRNA level, data were analysed by Mann-Whitney test. Statistical significance was accepted at p-value less than 0.05. Data availability statement. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Data availability statement. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 References 1. Allen, C. L. & Bayraktutan, U. Risk factors for ischaemic stroke. Int J Stroke 3, 105–116 (2008). 1. Allen, C. L. & Bayraktutan, U. Risk factors for ischaemic stroke. Int J Stroke 3, 105–116 (2008). 1. Allen, C. L. & Bayraktutan, U. Risk factors for ischaemic stroke. Int J Stroke 3, 105–116 (2008). l ll b h l d y 2. Luitse, M. J., Biessels, G. J., Rutten, G. E. & Kappelle, L. J. Diabetes, hyperglycaemia, and acute ischaemic stroke. Lancet Neurol 11 261–271 (2012). 3. Albert-Weissenberger, C., Siren, A. L. & Kleinschnitz, C. Ischemic stroke and traumatic brain injury: the role of the kallikrein-kinin system. Prog Neurobiol 101–102, 65–82 (2013). 3. Albert-Weissenberger, C., Siren, A. L. & Kleinschnitz, C. Ischemic stroke and traumatic brain injury: the role of the kallikrein-kinin system. Prog Neurobiol 101–102, 65–82 (2013). y g 4. Doyle, K. P., Simon, R. P. & Stenzel-Poore, M. P. Mechanisms of ischemic brain damage. Neuropharmacology 55, 310–318 (2008).h y g p gy 5. Rhaleb, N. E., Yang, X. P. & Carretero, O. A. The kallikrein-kinin system as a regulator of cardiovascular and renal function. Compr Physiol 1, 971–993 (2011). y 6. Leeb-Lundberg, L. M., Marceau, F., Muller-Esterl, W., Pettibone, D. J. & Zuraw, B. L. International union of pharmacology. XLV. Classification of the kinin receptor family: from molecular mechanisms to pathophysiological consequences. Pharmacol Rev 57, 27–77 (2005). 7. Kayashima, Y., Smithies, O. & Kakoki, M. The kallikrein-kinin system and oxidative stress. Curr Opin Nephrol Hypertens 21, 92–96 (2012). ( ) 8. Girolami, J. P., Blaes, N., Bouby, N. & Alhenc-Gelas, F. Genetic manipulation and genetic variation of the kallikrein-kinin system impact on cardiovascular and renal diseases. Prog Drug Res 69, 145–196 (2014).f p g g 9. Griol-Charhbili, V. et al. Role of tissue kallikrein in the cardioprotective effects of ischemic and pharmacological preconditioning in myocardial ischemia. FASEB J 19, 1172–1174 (2005). 9. Griol-Charhbili, V. et al. Role of tissue kallikrein in the cardioprotective effects of ischemic and pharmacological preconditioning in myocardial ischemia. FASEB J 19, 1172–1174 (2005). y 10. Kakoki, M., McGarrah, R. W., Kim, H. S. & Smithies, O. Bradykinin B1 and B2 receptors both have protective roles in renal ischemia/reperfusion injury. Proc Natl Acad Sci USA 104, 7576–7581 (2007). 11. Potier, L. et al. Selective kinin receptor agonists as cardioprotective agents in myocardial ischemia and diabetes. J Pharmacol Exp h y 10. Kakoki, M., McGarrah, R. www.nature.com/scientificreports/ www.nature.com/scientificreports/ 2. Desposito, D. et al. Kinin receptor agonism restores hindlimb postischemic neovascularization capacity in diabetic mice. Pharmacol Exp Ther 352, 218–226 (2015). ph 3. Kizuki, K., Suzuki, T., Kudo, M. & Noguchi, T. Immunohistochemical demonstration of tissue kallikrein in the neurons of rat brain Brain Res 634, 305–309 (1994). 4. Walker, K., Perkins, M. & Dray, A. Kinins and kinin receptors in the nervous system. Neurochem Int 26, 1–16; discussion 17–26 (1995). M. et al. Visualisation of bradykinin B2 receptors on human brain neurons. Immunopharmacology 33, 104–107 (1996). M. & Bhoola, K. D. Kinin receptors on human neurones. J Neuroimmunol 77, 39–44 (1997). et al. Visualisation of bradykinin B2 receptors on human brain neu 5. Raidoo, D. M. et al. Visualisation of bradykinin B2 receptors on human brain neurons. Immunopharmacology 33, 104–107 (1996). 6. Raidoo, D. M. & Bhoola, K. D. Kinin receptors on human neurones. J Neuroimmunol 77, 39–44 (1997). y p gy 16. Raidoo, D. M. & Bhoola, K. D. Kinin receptors on human neurones. J Neuroimmunol 77, 39–44 (1997). 7. Ongali, B. et al. Autoradiographic analysis of rat brain kinin B1 and B2 receptors: normal distribution and alterations induced by epilepsy. J Comp Neurol 461, 506–519 (2003).ht p p y p 8. Trabold, R. et al. The role of bradykinin B(1) and B(2) receptors for secondary brain damage after traumatic brain injury in mice. Cereb Blood Flow Metab 30, 130–139 (2010). ( ) 9. Groger, M. et al. Release of bradykinin and expression of kinin B2 receptors in the brain: role for cell death and brain edema formation after focal cerebral ischemia in mice. J Cereb Blood Flow Metab 25, 978–989 (2005). t 0. Austinat, M. et al. Blockade of bradykinin receptor B1 but not bradykinin receptor B2 provides protection from cerebral infarction and brain edema. Stroke 40, 285–293 (2009). 1. Bovenzi, V. et al. Bradykinin protects against brain microvascular endothelial cell death induced by pathophysiological stimuli. J Cel Physiol 222, 168–176 (2010). y elton, J. K., Beckey, V. E., Hanson, W. L. & Whalley, E. T. CP-0597, a selective bradykinin B2 receptor antagonist, inhibits brain jury in a rat model of reversible middle cerebral artery occlusion Stroke 28 1430 1436 (1997) y 2. Relton, J. K., Beckey, V. E., Hanson, W. L. & Whalley, E. T. www.nature.com/scientificreports/ CP-0597, a selective bradykinin B2 receptor antagonist, inhibits brain injury in a rat model of reversible middle cerebral artery occlusion. Stroke 28, 1430–1436 (1997). 23. Ding-Zhou, L. et al. LF 16-0687 Ms, a bradykinin B2 receptor antagonist, reduces ischemic brain injury in a murine model of transient focal cerebral ischemia. Br J Pharmacol 139, 1539–1547 (2003).ht 24. Klasner, B., Lumenta, D. B., Pruneau, D., Zausinger, S. & Plesnila, N. Therapeutic window of bradykinin B2 receptor inhibition f l b l i h i i t N h I t 49 442 447 (2006) 24. Klasner, B., Lumenta, D. B., Pruneau, D., Zausinger, S. & Plesnila, N. Therapeutic window of bradykinin B2 receptor inhibition after focal cerebral ischemia in rats. Neurochem Int 49, 442–447 (2006). . Klasner, B., Lumenta, D. B., Pruneau, D., Zausinger, S. & Plesnila, g focal cerebral ischemia in rats. Neurochem Int 49, 442–447 (2006) ( ) 25. Chao, J. & Chao, L. Kallikrein-kinin in stroke, cardiovascular and renal disease. Exp Physiol 90, 291–298 (2005). 26. Xia, C. F. et al. Postischemic brain injury is exacerbated in mice lacking the kinin B2 receptor. Hypertension 47, 752–761 (2006 27. Danielisova, V., Gottlieb, M., Bonova, P., Nemethova, M. & Burda, J. Bradykinin postconditioning ameliorates focal cerebral ischemia in the rat. Neurochem Int 72, 22–29 (2014). 8. Duka, I. et al. Vasoactive potential of the b(1) bradykinin receptor in normotension and hypertension. Circ Res 88, 275–281 (2001) p y p yp 29. Marcon, R. et al. Exacerbation of DSS-induced colitis in mice lacking kinin B(1) receptors through compensatory up-regulation of kinin B(2) receptors: the role of tight junctions and intestinal homeostasis. Br J Pharmacol 168, 389–402 (2013). 29. Marcon, R. et al. Exacerbation of DSS-induced colitis in mice lacking kinin B(1) receptors through compensator kinin B(2) receptors: the role of tight junctions and intestinal homeostasis. Br J Pharmacol 168, 389–402 (2013). 0. Sang, H. et al. Opposite roles of bradykinin B1 and B2 receptors during cerebral ischaemia-reperfusion injury in experimenta diabetic rats. Eur J Neurosci 43, 53–65 (2016).t 1. Tschope, C. et al. Upregulation of bradykinin B1-receptor expression after myocardial infarction. Br J Pharmacol 129, 1537–1538 (2000). 2. Campos, M. M., Souza, G. E. & Calixto, J. B. In vivo B1 kinin-receptor upregulation. Evidence for involvement of protein kinases and nuclear factor kappaB pathways. Br J Pharmacol 127, 1851–1859 (1999).hl y 33. Matsumoto, S., Yamasaki, M., Kanno, T., Nagayama, T. References W., Kim, H. S. & Smithies, O. Bradykinin B1 and B2 receptors both have protective roles in renal ischemia/reperfusion injury. Proc Natl Acad Sci USA 104, 7576–7581 (2007). y 10. Kakoki, M., McGarrah, R. W., Kim, H. S. & Smithies, O. Bradykinin B1 and B2 receptors both have protective roles in renal ischemia/reperfusion injury. Proc Natl Acad Sci USA 104, 7576–7581 (2007). p j y , ( ) 1. Potier, L. et al. Selective kinin receptor agonists as cardioprotective agents in myocardial ischemia and diabetes. J Pharmacol Exp Ther 346, 23–30 (2013). Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 9 Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 www.nature.com/scientificreports/ & Shimizu, T. The reflex bradycardia during brain ischemia in the rabbit. Jpn J Physiol 41, 703–715 (1991).tl y 4. Dobrivojevic, M., Spiranec, K. & Sindic, A. Involvement of bradykinin in brain edema development after ischemic stroke. Pfluger Arch 467, 201–212 (2015). 35. Desposito, D. et al. Improvement of skin wound healing in diabetic mice by kinin B2 receptor blockade. Clin Sci (Lond) 130, 45–56 (2016).h muels, M. A. The brain-heart connection. Circulation 116, 77–84 (2 h 37. Alhenc-Gelas, F., Tsai, S. J., Callahan, K. S., Campbell, W. B. & Johnson, A. R. Stimulation of prostaglandin formation mediators in cultured human endothelial cells Prostaglandins 24 723–742 (1982) h 37. Alhenc-Gelas, F., Tsai, S. J., Callahan, K. S., Campbell, W. B. & Johnson, A. R. Stimulation of prostaglandin formation by vasoa h 37. Alhenc-Gelas, F., Tsai, S. J., Callahan, K. S., Campbell, W. B. & Johnson, A. R. Stimulation of pr mediators in cultured human endothelial cells. Prostaglandins 24, 723–742 (1982). mediators in cultured human endothelial cells. Prostaglandins 24, 723–742 (1982). Furchgott, R. F. & Vanhoutte, P. M. Endothelium-derived relaxing a F. & Vanhoutte, P. M. Endothelium-derived relaxing and contractin 39. Brown, N. J., Gainer, J. V., Murphey, L. J. & Vaughan, D. E. Bradykinin stimulates tissue plasminogen activator release from human forearm vasculature through B(2) receptor-dependent, NO synthase-independent, and cyclooxygenase-independent pathway. Circulation 102, 2190–2196 (2000). 0. Park, S. S., Zhao, H., Mueller, R. A. & Xu, Z. Bradykinin prevents reperfusion injury by targeting mitochondrial permeability transition pore through glycogen synthase kinase 3beta. J Mol Cell Cardiol 40, 708–716 (2006).f 1. Youcef, G. et al. Tissue kallikrein is required for the cardioprotective effect of cyclosporin A in myocardial ischemia in the mouse Biochem Pharmacol 94, 22–29 (2015).f 42. Alhenc-Gelas, F., Marchetti, J., Allegrini, J., Corvol, P. & Menard, J. Measurement of urinary kallikrein activity. Species differences in kinin production. Biochim Biophys Acta 677, 477–488 (1981). p p y 3. Paiva, A. C., Paiva, T. B., Pereira, C. C. & Shimuta, S. I. Selectivity of bradykinin analogues for receptors mediating contraction and relaxation of the rat duodenum. Br J Pharmacol 98, 206–210 (1989).h elaxation of the rat duodenum. Br J Pharmacol 98, 206–210 (1989) 4. Meini, S., Lecci, A. & Maggi, C. A. The longitudinal muscle of rat ileum as a sensitive monoreceptor assay for bradykinin B1 receptors. Br J Pharmacol 117, 1619–1624 (1996). 44. Meini, S., Lecci, A. & Maggi, C. www.nature.com/scientificreports/ Hatashita, S., Hoff, J. T. & Salamat, S. M. Ischemic brain edema and the osmotic gradient between blood and brain. J Cereb Blood Flow Metab 8, 552–559 (1988).ft Flow Metab 8, 552 559 (1988). 59. Zweckberger, K. et al. Effect of early and delayed decompressive craniectomy on secondary brain damage after controlled cortical impact in mice. J Neurotrauma 23, 1083–1093 (2006). , ( ) 9. Zweckberger, K. et al. Effect of early and delayed decompressive craniectomy on secondary brain damage after controlled cortica impact in mice. J Neurotrauma 23, 1083–1093 (2006). B d S l K ll k lb l I d b Kid I 6 ( ) 60. Bodin, S. et al. Kallikrein protects against microalbuminuria in experimental type I diabetes. Kidney Int 76, 395–403 (2009). 61. Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408 (2001). p g p yp y 1. Livak, K. J. & Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408 (2001). Author Contributions D.D., F.A.-G., R.R., N.B. made substantial contribution to the design and have drafted the manuscript script. D.D., G.Z., C.T., C.A. made substantial contribution to acquisition and analysis of data. All authors have reviewed and approved the final version. Acknowledgements g The authors thank the crew of the Functional Exploration Center (Cordeliers Research Center, Paris) for technical assistance and support in mice studies. This work was supported by Institut National de la Santé Et de la Recherche Medicale, Paris-Descartes University, Pierre and Marie Curie University. www.nature.com/scientificreports/ A. The longitudinal muscle of rat ileum as a sensitive mon receptors. Br J Pharmacol 117, 1619–1624 (1996). 45. Ni, A. et al. Overexpression of kinin B1 receptors induces hypertensive response to des-Arg9-bradykinin and susceptibility to inflammation. J Biol Chem 278, 219–225 (2003).i l 6. Meneton, P. et al. Cardiovascular abnormalities with normal blood pressure in tissue kallikrein-deficient mice. Proc Natl Acad Sc USA 98, 2634–2639 (2001).i 7. Trabold, F. et al. Cardiovascular phenotypes of kinin B2 receptor- and tissue kallikrein-deficient mice. Hypertension 40, 90–95 (2002). 48. Rousselet, E., Kriz, J. & Seidah, N. G. Mouse model of intraluminal MCAO: cerebral infarct evaluation by cresyl violet staining. J Vis Exp (2012).i p 49. Cote, J. et al. Novel kinin B1 receptor agonists with improved pharmacological profiles. Peptides 30, 788–795 (2009). 50. Belanger, S. et al. Structure-activity relationships of novel peptide agonists of the human bradykinin B2 receptor. Peptides 30, 777–787 (2009). 51. Waeckel, L. et al. Antihypertensive role of tissue kallikrein in hyperaldosteronism in the mouse. Endocrinology 153, 3886–3896 (2012).t 52. Hunter, A. J. et al. Functional assessments in mice and rats after focal stroke. Neuropharmacology 39, 806–816 (2000). t 53. Schallert, T. Behavioral tests for preclinical intervention assessment. NeuroRx 3, 497–504 (2006). 54. Balkaya, M., Krober, J. M., Rex, A. & Endres, M. Assessing post-stroke behavior in mouse models of focal ischemia. J Cereb Blood Flow Metab 33, 330–338 (2013).i 55. Joshi, C. N., Jain, S. K. & Murthy, P. S. An optimized triphenyltetrazolium chloride method for identification of cerebral infarcts. Brain Res Brain Res Protoc 13, 11–17 (2004). 6. Okuno, S., Nakase, H. & Sakaki, T. Comparative study of 2,3,5-triphenyltetrazolium chloride (TTC) and hematoxylin-eosin staining for quantification of early brain ischemic injury in cats. Neurol Res 23, 657–661 (2001). Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 10 www.nature.com/scientificreports/ 7. Baker, A. J. et al. Hypothermia prevents ischemia-induced increases in hippocampal glycine concentrations in rabbits. Stroke 22 666–673 (1991).f , J yp p pp p g y , 666–673 (1991). 58. Hatashita, S., Hoff, J. T. & Salamat, S. M. Ischemic brain edema and the osmotic gradient between blood and brain. J Cereb Blood Fl M b 8 552 559 (1988) 666 673 ( 99 ). 58. Hatashita, S., Hoff, J. T. & Salamat, S. M. Ischemic brain edema and the osmotic gradient between blood and brain. J Cereb Blood Flow Metab 8, 552–559 (1988). 8. Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 Additional Informationh Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Scientific Reports \| 7: 9410 \| DOI:10.1038/s41598-017-09721-0 11
https://openalex.org/W2588445271	https://europepmc.org/articles/pmc5367300?pdf=render	English	null	A novel locus on chromosome 1 underlies the evolution of a melanic plumage polymorphism in a wild songbird	Royal Society open science	2,017	cc-by	14,782	Research Cite this article: Bourgeois YXC etal. 2017 A novel locus on chromosome 1 underlies the evolutionofamelanicplumagepolymorphism in a wild songbird. R.Soc.opensci. 4: 160805. http://dx.doi.org/10.1098/rsos.160805 Received: 11 October 2016 Accepted: 12 January 2017 Subject Category: Biology (whole organism) Subject Areas: ecology/evolution/genomics Keywords: melanism, chromosome 1, polymorphism, selective sweep, Zosterops Authors for correspondence: Yann X. C. Bourgeois e-mail: yann.x.c.bourgeois@gmail.com Borja Milá e-mail: b.mila@csic.es Christophe Thébaud e-mail: christophe.thebaud@univ-tlse3.fr †Present address: New York University Abu Dhabi, PO Box 129188, United Arab Emirates. ‡These authors contributed equally to this study. Electronic supplementary material is available online at https://dx.doi.org/10.6084/m9. figshare.c.3679084. Cite this article: Bourgeois YXC etal. 2017 A novel locus on chromosome 1 underlies the evolutionofamelanicplumagepolymorphism in a wild songbird. R.Soc.opensci. 4: 160805. http://dx.doi.org/10.1098/rsos.160805 Yann X. C. Bourgeois1,†, Boris Delahaie1, Mathieu Gautier2, Emeline Lhuillier3,4, Pierre-Jean G. Malé1, Joris A. M. Bertrand1, Josselin Cornuault1, Kazumasa Wakamatsu5, Olivier Bouchez3,6, Claire Mould1, Jade Bruxaux1, Hélène Holota1, Borja Milá7,‡ and Christophe Thébaud1,‡ 1Laboratoire Évolution et Diversité Biologique, UMR5174 CNRS, Université Paul Sabatier – ENFA, 31062 Toulouse Cedex 9, France 2INRA, UMR 1062 CBGP (INRA, IRD, Cirad, Montpellier SupAgro), Campus de Baillarguet, 34988 Montferrier-sur-Lez, France 3INRA, GeT-PlaGe, Genotoul, 24 chemin de Borde Rouge, Auzeville, CS 52627, 31326 Castanet-Tolosan, France 4INRA, UAR1209, 24 chemin de Borde Rouge, Auzeville, CS 52627, 31326 Castanet-Tolosan, France 5Department of Chemistry, Fujita Health University, School of Health Sciences, Toyoake Aichi 470-1192, Japan 6GenPhySE, Université de Toulouse, INRA, INPT, INP-ENVT, 24 chemin de Borde Rouge, Auzeville, CS 52627, 31326 Castanet-Tolosan, France 7National Museum of Natural Sciences, Spanish National Research Council (CSIC), 28006 Madrid, Spain Authors for correspondence: Yann X. C. Bourgeois e-mail: yann.x.c.bourgeois@gmail.com Borja Milá e-mail: b.mila@csic.es Christophe Thébaud e-mail: christophe.thebaud@univ-tlse3.fr YXCB, 0000-0002-1809-387X; EL, 0000-0003-2629-7773 †Present address: New York University Abu Dhabi, PO Box 129188, United Arab Emirates. ‡These authors contributed equally to this study. Understanding the mechanisms responsible for phenotypic diversiﬁcation within and among species ultimately rests with linking naturally occurring mutations to functionally and ecologically signiﬁcant traits. Colour polymorphisms are of great interest in this context because discrete colour patterns within a population are often controlled by just a few genes in a common environment. We investigated how and why phenotypic diversity arose and persists in Electronic supplementary material is available online at https://dx.doi.org/10.6084/m9. figshare.c.3679084. 2017 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original author and source are credited. 1. Introduction Identifying the speciﬁc genes underlying inter-individual phenotypic variation and reconstructing their evolutionary history is a key issue to link naturally occurring mutations to ecologically signiﬁcant traits, and help resolve questions relative to the origin and maintenance of genetic and phenotypic variability in natural populations [1]. In this context, colour polymorphisms have played an essential role by enhancing our understanding of how selection and demography can impact phenotypes [2–5], and by identifying proximate causes of phenotypic variation [6,7]. Melanins, in contrast to other pigments such as carotenoids, allow the production of colour traits that in most cases are independent of the environment and the individuals’ phenotypic condition [8,9] (but see also [10,11]) and show heritable segregation among colour morphs [12], a feature that appears to be critical for explaining the origin and maintenance of distinct colour morphs in natural populations [13]. Moreover, the observed covariation between melanin-based pigmentation and life-history or social strategies [14,15] suggests a major role for intergenic interactions and pleiotropic effects in the evolution of colour polymorphisms [16]. A great deal of research over the last decade has been devoted to elucidating the molecular basis of melanin-based colour polymorphism in natural populations. Early studies have assessed the role of the melanocortin-1-receptor (MC1R) gene, currently one of the most studied ‘colour genes’. This gene displays relatively low pleiotropic effect (but see [17]) and has been shown to explain colour variation in a broad variety of independent lineages, such as lizards [18], mice [19], humans [20], mammoths [21] and birds [22–25]. Although this suggested at ﬁrst a relatively simple and general genetic mechanism underlying melanin-based polymorphisms, there is mounting evidence suggesting that mechanisms of hair/feather colour evolution may involve many other genes, and this seems particularly likely in species or populations in which the polymorphism arises from complex patterns of eumelanin/phaeomelanin deposition [26,27]. Studies in model species such as laboratory mice have shown that several genes other than MC1R could play a role in explaining differences in the patterning of melanin pigments across the body [28], and alternative candidate loci for melanic colour polymorphisms have indeed been identiﬁed in wild non-model species [29–32]. The recent development of next-generation sequencing has allowed research to transcend candidate- gene approaches in wild non-model species, leading to the identiﬁcation of genomic regions associated with ﬁtness-related traits, including colour traits [33,34]. Research 2 the Zosterops borbonicus white-eye of Reunion (Mascarene archipelago), a colour polymorphic songbird in which all highland populations contain individuals belonging to either a brown or a grey plumage morph. Using extensive phenotypic and genomic data, we demonstrate that this melanin- based colour polymorphism is controlled by a single locus on chromosome 1 with two large-effect alleles, which was not previously described as affecting hair or feather colour. Differences between colour morphs appear to rely upon complex cis-regulatory variation that either prevents the synthesis of pheomelanin in grey feathers, or increases its production in brown ones. We used coalescent analyses to show that, from a ‘brown’ ancestral population, the dominant ‘grey’ allele spread quickly once it arose from a new mutation. Since colour morphs are always found in mixture, this implies that the selected allele does not go to ﬁxation, but instead reaches an intermediate frequency, as would be expected under balancing selection. rg R.Soc. opensci. 4: 160805 1. Introduction Phenotypic variation in highland populations of the Reunion grey white-eye and the population sampling scheme. (a) Photographs of brown and grey morphs (i), microscopic view of grey and brown feathers (ii) and melanin content per morph (iii) for 12 brown and 12 grey individuals. (b) PCA on spectrophotometry measures for feathers sampled from the back of brown and grey birds. (c) Morph frequencies across different localities. ‘P’ stands for localities included in pedigree analyses. Localities used for GBS and RAD-seq analyses are labelled. The size of the pie charts is proportional to sample size (ranging from 11 to 111 birds). (b) 3 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BBH GRY 2 1 –1 0 0 1 2 3 –1 –2 –3 PC1 (69%) PC2 (16%) (b) 3 (a) brown grey 1800 1900 2000 2100 2200 morph PTCA (ng/mg) − eumelanin marker brown grey 0 100 200 300 400 500 600 morph 4−AHP (ng/mg) − pheomelanin marker ing.org R.Soc. opensci. 4: 160805 PC1 (69%) (c) (c) P Tévelave Pas de Bellecombe P Bois Ozoux P scale approx 1:220,000 0 15 km 10 5 brown grey 0 100 200 300 400 500 600 morph 4−AHP (ng/mg) − pheomelanin marker brown grey 1800 1900 2000 2100 2200 morph PTCA (ng/mg) − eumelanin marker Pas de Bellecombe P Figure 1. Phenotypic variation in highland populations of the Reunion grey white-eye and the population sampling scheme. (a) Photographs of brown and grey morphs (i), microscopic view of grey and brown feathers (ii) and melanin content per morph (iii) for 12 brown and 12 grey individuals. (b) PCA on spectrophotometry measures for feathers sampled from the back of brown and grey birds. (c) Morph frequencies across different localities. ‘P’ stands for localities included in pedigree analyses. Localities used for GBS and RAD-seq analyses are labelled. The size of the pie charts is proportional to sample size (ranging from 11 to 111 birds). Table 1. Summary of datasets used in this study. 1. Introduction y y method used goal notes sample size references microsatellites pedigree analysis to characterize inheritance patterns of plumage colour obtained from two populations monitored since 2008 (Pas de Bellecombe and Bois Ozoux) and one population sampled in 2007 and 2012 (figure 1) 261 [39] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pooled RAD-sequencing mapping loci associated with plumage colour high density of markers (more than 600 000) 137 (6 pools) [40] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GBS validating associations obtained from pooled data with individual genotyping lower density of markers than RAD-seq (25 000) 42 this study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . whole-genome resequencing annotating coding/non-coding mutations; characterizing selection in the genomic region associated with plumage colour near-exhaustive coverage of the region associated with plumage colour 12 (6 parents and 6 offspring) this study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction Here, we took advantage of this technology and used a population genomic approach to investigate the genetic architecture underpinning a melanic plumage polymorphism in natural populations of the Reunion grey white-eye, Zosterops borbonicus, a textbook example of intraspeciﬁc variation in plumage coloration in birds [35]. This species displays four geographically structured plumage forms distributed across the small oceanic island of Reunion (2512 km2) [36]. One of these forms, restricted to the highlands of Reunion, comprises two distinct and sympatric colour morphs, with birds showing predominantly grey or brown plumage, respectively (ﬁgure 1). Melanic plumage polymorphism is widely maintained across the range of this highland form, with both grey and brown birds always present at any given locality, although populations vary in morph frequencies [30]. Field observations suggesting that mating between grey individuals could produce both grey and brown offspring, and that mating between brown individuals always produced brown offspring, led to the proposition that differences between grey and brown morphs could be due to genetic changes at a few loci of major effect [30]. Previously, we used a candidate-gene approach to search for associations between plumage colour and genetic variants in the Reunion grey white-eye, and did not detect any for a series of genes 3 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BBH GRY 2 1 –1 0 0 1 2 3 –1 –2 –3 PC1 (69%) PC2 (16%) (c) (b) P Tévelave Pas de Bellecombe P Bois Ozoux P scale approx 1:220,000 0 15 km 10 5 he Reunion grey white-eye and the population sampling scheme. w of grey and brown feathers (ii) and melanin content per morph (iii) etry measures for feathers sampled from the back of brown and grey BBH GRY brown grey 1800 1900 2000 2100 2200 morph PTCA (ng/mg) − eumelanin marker brown grey 0 100 200 300 400 500 600 morph 4−AHP (ng/mg) − pheomelanin marker 2 1 –1 0 0 1 2 3 –1 –2 –3 PC1 (69%) PC2 (16%) (c) (a) (b) P Tévelave Pas de Bellecombe P Bois Ozoux P scale approx 1:220,000 0 15 km 10 5 Figure 1. 2.3. Melanin content We used microanalytical methods to quantify the eumelanin and pheomelanin content of the feathers from brown and grey morphs (see [49] for a detailed description of the method). Feather samples (ca 7 mg) from the back of 12 brown and 12 grey individuals were homogenized with a Ten-Broeck homogenizer at a concentration of 10 mg ml−1 and 100 µl aliquots were subjected to Soluene-350 solubilization [50], alkaline hydrogen peroxide oxidation [51] and hydriodic acid hydrolysis [52]. Values are from single determinations. A500/mg and A650/mg values were substracted by background values of 0.019 and 0.001. 2.1. Field sampling Birds were captured in the ﬁeld using mist nets between 2007 and 2012 on Reunion (55°39′ E; 21°00′ S), were weighed and marked with a uniquely numbered aluminium ring, and approximately 10 µl of blood was collected from each bird. Blood was conserved in Queen’s lysis buffer [43] and stored at −20°C for long-term preservation. Individuals were sexed by PCR [44] in order to infer the number of distinct Z chromosomes included in each pool. On each bird, 10 feathers from four different body parts (head, back, ﬂank and belly) were collected for spectrophotometric analyses. We also measured ﬁve morphological traits with a dial calliper (to the nearest 0.1 mm): tail length (from the uropygial gland to the tip of the longest rectrix), tarsus length (from the intertarsal joint to the most distal undivided scute on the tarsometatarsus), bill length (from the anterior end of the nares to the tip of the upper mandible), bill width and depth (both measured at the anterior end of the nares). 2.2. Measurement of the reflectance spectra of plumage patches Ninety-six individuals from the 137 used for pooled RAD-sequencing were used for characterizing phenotypes by spectrophotometry analyses. We used reﬂectance spectrophotometry to characterize variation in plumage colour, because it provides an objective quantiﬁcation of colour [45,46]. We summarized colour variation by conducting a principal components analysis (PCA) on a set of ﬁve variables summarizing the spectrum for back feathers that were then used for estimating melanin content. We determined whether birds were actually able to discriminate morphs by using Vorobyev & Osorio’s model of colour perception [47]. This model allows chromatic contrasts between two colours to be compared with a threshold value of colour discrimination by birds (for further details of how this was perfomed, see [48]). 4 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . that play a critical role in melanin-based pigmentation patterns in other species, including MC1R, Agouti, Corin, Pro-opiomelanocortin and Tyrosinase-related protein 1 [37,38]. In this study, we ﬁrst conﬁrmed that segregating phenotypes conform to a Mendelian model using genetic marker-based pedigree reconstruction (table 1). Then, using restriction site-associated DNA (RAD)-sequencing [41] on pools of individuals from three geographically separate populations [40], we performed a genome-wide scan relative to colour morphs while accounting for population structure. This approach provides a cost- effective way to detect associations by allowing large numbers of individuals to be included into a sample and by providing a sufﬁcient density of markers for mapping (see table 1 for more details on the approaches used in this study). This led us to discover a single narrow region on chromosome 1 showing a strong association with melanin-based plumage colour phenotypes. We conﬁrmed the existence of this association by using individual genotyping-by-sequencing (GBS) [42] and whole- genome sequencing of individuals with known pedigrees, and assessed the functional consequences of the molecular variants underlying plumage colour variation. Finally, we addressed the origin of the polymorphism by estimating the relative age of the genetic variants and tested the role of selection in explaining how the polymorphism could have spread and been widely maintained across populations. 2.6. Genotyping by sequencing using individual DNA samples We further validated the results obtained by the pooled RAD-seq approach by performing a genotyping by sequencing analysis [42] on 42 individuals from the same populations that were used to build the pooled DNA samples (seven individuals per morph per population for a total of 14 individuals in each population). Approximately 1 µg of DNA was extracted with a QIAGEN Blood and Tissue kit following the manufacturer’s instructions and sent to the BRC Genomic Diversity Facility at Cornell University [42]. Reads were trimmed with Trimmomatic (v0.33) and mapped on Z. lateralis genome using BWA. SNPs were called using freebayes (v0.9.15-1) and ﬁltered with VCFTOOLS (0.1.12b) using the following criteria: (i) a mean sequencing depth between 6 and 20×; (ii) a minimal genotype quality of 20; (iii) a minor allele count of 3, which resulted in around 25 000 unambiguous SNPs. 2.5. RAD-sequencing using pooled DNA samples To identify loci associated with colour, we used a paired-end RAD-sequencing protocol, using a dataset described previously in which six pools of 18–25 individuals representing the two colour morphs in each of three separate populations (named ‘Bois Ozoux’, ‘Pas de Bellecombe’ and ‘Tévelave’) were sequenced [40]. We took advantage of the recent sequencing of the Z. lateralis genome [56] to map reads back onto this reference with BWA MEM (v. 0.7.12) [57], instead of creating consensuses directly from data as in [40]. We aligned contigs and scaffolds from Z. lateralis on the zebra ﬁnch genome (version July 2008, assembly WUGSC v. 3.2.4) using LASTZ [58]. We removed PCR duplicates using samtools [59]. SNPs were called using Popoolation2 (v. 1.201) [60]. 2.4. Pedigree analysis Performing controlled matings in wild populations is challenging, and hence to study inheritance patterns, we inferred pedigrees in three natural populations monitored over multiple years at three localities. Given the reduced dispersal in Z. borbonicus [53,54], samples from these populations were more likely to include parents and their offspring. Pedigree reconstruction was performed using the Bayesian parentage assignment algorithm implemented in the R package MASTERBAYES v. 2.50 [55] and genotype data from an informative panel of 11 neutral microsatellites [39]. A total of 261 birds of known sex were genotyped and included in this analysis. When known, a birth cohort identiﬁer was speciﬁed to minimize the possibility that individuals appear as potential parents of offspring from the same or a previous cohort. Since MASTERBAYES allows the simultaneous use of genetic and phenotypic data that may inform on parentage, the reconstruction was performed twice: a ﬁrst time using genotypic information only and a second time including both genetic and morphometric data (including body mass). The genotyping error rates for each locus, the number of unsampled sires and the number of unsampled females were all estimated jointly from the pedigree. Markov chains were run for 1.1 million iterations, with a burn-in of 100 000 iterations and a thinning interval of 1000. Parents (or father or mother) were assigned to an offspring with a 95% joint posterior probability threshold. 5 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Putative pedigrees obtained from microsatellite data were conﬁrmed by checking the coefﬁcients of relatedness between individuals in KING (v. 1.4) [66] using 10 000 biallelic SNPs randomly sampled across autosomal scaffolds. An association analysis (including indels) taking into account family structure was performed in LAMP (v. 0.0.12) [67]. We present here results taking into account the putative heredity of the colour trait (brown recessive, grey dominant) as they were similar to those obtained when not taking the transmission mode into account. To assess whether any associated SNP might lead to changes in protein sequence, we performed an annotation using the software SNPdat (v. 1.0.5) [68] and using gene coordinates from [56]. 2.7. Genetic structure and association analysis To validate the lack of genetic structure according to morphs, we performed a PCA [61], using custom scripts in R (v3.2.2) for allele frequencies obtained from pooled data, and the Bioconductor package SeqVarTools for GBS data [62]. Given our sampling strategy, we expected morphs to group together in the PCA and population structure, if any, to correlate with geography. To further rule out the possibility of any neutral structure according to colour morph, we also conducted a locus by locus analysis of molecular variance (AMOVA) in ARLEQUIN v. 3.5 [63] on the 10 115 autosomal SNPs with less than 95% of missing data. g To detect loci displaying a signiﬁcant association with coloration, we performed an association analysis on the pooled RAD-seq data using the software BAYPASS (v. 2.1) [64]. We computed the empirical Bayesian p-value (eBPis) and Bayes factors (BF) expressed in deciban units (dB) to determine the level of association of each SNP with the grey/brown trait. Decibans are a commonly used statistic that describes here the probability that data are produced under a given model. They are often used as a unit for BF. Here, it describes the odds of association to morphs versus the null hypothesis of non-association. BAYPASS was run using default parameters under the core model. Empirical Bayesian p-values and BF were then computed for a pseudo-observed dataset containing 1 000 000 SNPs simulated from the actual data. We then compared the observed values to this distribution in order to calibrate the statistics. For GBS data, we performed a single analysis in PLINK [65] by correcting for population structure with a Cochran–Mantel–Haenszel test (option –mh in PLINK). 2.9. Detecting selective sweeps and characterizing derived and ancestral states To determine when and on which speciﬁc allele selection occurred, we used the coalescent framework implemented in ARGWeaver [69]. ARGWeaver models the coalescent process across non-recombining blocks of sequences and thus provides access to the evolutionary history of DNA sequences. It allows recovering several statistics that describe local genealogies, like coalescence times and local effective population sizes. In the case of a recent and partial selective sweep, selected lineages should display shorter coalescence times than the ancestral ones, i.e. alleles under selection will tend to be younger than neutral alleles. We ran the analyses using whole-genome sequences from the three parental pairs, as these individuals displayed low levels of relatedness in the whole-genome association analysis and were predicted to display an equal number of brown and grey alleles. To reduce computational burden, we performed the analysis on the scaffold covering the candidate region (scaffold 40) and one on chromosome 2 (scaffold 30) with no SNPs associated with colour. The algorithm was run for 1000 iterations. We then extracted trees and half-time TMRCAs from the output. 6 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 To validate the hereditary transmission of plumage colour and conﬁrm the validity of pedigrees, 12 individuals from three distinct families were selected for 150 bp paired-end Illumina sequencing. Individuals included 4 brown and 8 grey birds (for reconstructed pedigrees see the electronic supplementary material, table S4). Three types of matings were considered: two grey parents with only grey offspring, two grey parents with both brown and grey offspring, and one parent of each morph with both types of offspring. DNA sequencing libraries were prepared according to Illumina’s protocols using the Illumina TruSeq Nano DNA LT Library Prep Kit. Brieﬂy, DNA was fragmented by sonication (Covaris M220) and adaptators were ligated to be sequenced. Eight cycles of PCR were then applied to amplify the libraries. Library quality was assessed using an Agilent Bioanalyzer and libraries were quantiﬁed by QPCR using the Kapa Library Quantiﬁcation Kit. Sequencing was performed in paired-end (2 × 125 pb) on an Illumina HiSeq2500 sequencer at the GeT-PlaGe core facility (INRA, Toulouse). The mean sequencing depth ranged from 6.9× to 11.6× after removing duplicates (9.3× in mean, electronic supplementary material, table S6). After calling with freebayes, SNPs were ﬁltered with VCFTOOLS using the following criteria: (i) a mean sequencing depth between 6 and 20×; (ii) a minimal Phred score of 20 and a genotype quality above 20; (iii) at least eight individuals genotyped. 3.1. Differences in melanin-based pigmentation patterns between morphs are functionally significant Brown and grey birds were different in coloration when examined in an avian-appropriate colour space, clustering into two distinct groups along the ﬁrst component axis in a PCA (ﬁgure 1; electronic supplementary material, table S1). All colour variables differed signiﬁcantly between grey and brown morphs (electronic supplementary material, table S1). In addition, we found that, for all patches, the differences between morphs are greater than the discrimination threshold in the avian visual space (electronic supplementary material, table S1). Taken together, these results highlight the functional signiﬁcance in relation to colour perception of the differences in pigmentation patterns that distinguish grey and brown birds. We analysed the melanic content of feathers by determining eumelanin and pheomelanin concentrations (ﬁgure 1; electronic supplementary material, table S2). Although levels of eumelanin were 7 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.4 –0.2 0 0.2 0.4 –0.4 0 0.2 0.4 PC1 (19.72%) PC2 (18.51%) Pas de Bellecombe Bois ozoux Tévelave –0.4 –0.3 –0.2 –0.1 0 0.1 0.2 –0.2 0 0.2 0.4 PC1 (5.62%) PC2 (5.55%) Pas de Bellecombe Bois Ozoux Tévelave 1 2 0 chr1 chr10 chr11 chr12 chr13 chr14 chr15 chr16 chr17 chr18 chr19 chr1A chr1B chr2 chr20 chr21 chromosome chr22 chr23 chr24 chr25 chr26 chr27 chr28 chr3 chr4 chr4A chr5 chr6 chr7 chr8 chr9 chrLG2 chrLG5 chrLGE22 chrM chrZ eBPis position on Z. lateralis scaffold 40 frequency of SNPs highly associated (eBPis < 1.176) 0 1 × 106 2 × 106 3 × 106 5 × 106 4 × 106 0 1 × 106 1.5 × 107 1.7 × 107 1.9 × 107 2.1 × 107 2 × 106 3 × 106 5 × 106 4 × 106 0 2 4 6 8 * * * * * * * * * * * * * * position on zebra finch chromosome 1 Z. lateralis scaffold 40 0.6 (a) (b) (c) Figure2. Populationstructureandassociationstudy.(a)PCAonallelefrequenciesforpooledRAD-seq(top)andindividualGBS(bottom) data. (b) Genome-wide association analysis performed with BAYPASS for 627 795 RAD-seq loci. Horizontal dashed lines correspond to 0.001% (grey) and 0.01 (black) significance thresholds. (c) Density of highly associated SNPs on scaffold 40 from the Zosterops lateralis referencegenome.Upperstars:SNPsdisplayingbothhighlysignificantBFandempiricalBayesianp-values(abovethe0.001%threshold). 3.1. Differences in melanin-based pigmentation patterns between morphs are functionally significant The region including all 100 kb intervals with at least three highly associatedSNPsisshowninorange,andthethreeintervalswiththehighestdensityofassociatedSNPsareshowninred.LASTZalignment between scaffold 40 and the zebra finch chromosome 1 is also illustrated. org R.Soc. opensci. 4: 160805 –0.4 –0.3 –0.2 –0.1 0 0.1 0.2 –0.2 0 0.2 0.4 PC1 (5.62%) Pas de Bellecombe Bois Ozoux Tévelave chromosome 2 position on Z. lateralis scaffold 40 frequency of SNPs highly associated (eBPis < 1.176) 0 1 × 106 2 × 106 3 × 106 5 × 106 4 × 106 0 1 × 106 1.5 × 107 1.7 × 107 1.9 × 107 2.1 × 107 2 × 106 3 × 106 5 × 106 4 × 106 0 2 4 6 8 * * * * * * * * * * * * * * position on zebra finch chromosome 1 Z. lateralis scaffold 40 (c) –0.4 –0.3 –0.2 –0.1 0 0.1 0.2 –0.2 0 0.2 0.4 PC1 (5.62%) PC2 (5.55%) Pas de Bellecombe Bois Ozoux Tévelave Figure2. Populationstructureandassociationstudy.(a)PCAonallelefrequenciesforpooledRAD-seq(top)andindividualGBS(bottom) data. (b) Genome-wide association analysis performed with BAYPASS for 627 795 RAD-seq loci. Horizontal dashed lines correspond to 0.001% (grey) and 0.01 (black) significance thresholds. (c) Density of highly associated SNPs on scaffold 40 from the Zosterops lateralis referencegenome.Upperstars:SNPsdisplayingbothhighlysignificantBFandempiricalBayesianp-values(abovethe0.001%threshold). Lower stars: 7 GBS SNPs associated with a p-value < 1 × 10−4. The region including all 100 kb intervals with at least three highly associatedSNPsisshowninorange,andthethreeintervalswiththehighestdensityofassociatedSNPsareshowninred.LASTZalignment between scaffold 40 and the zebra finch chromosome 1 is also illustrated. opulationstructureandassociationstudy.(a)PCAonallelefrequenciesforpooledRAD-seq(top)andindividualGBS(bottom) similar in both morphs, levels of pheomelanin were clearly lower in the grey morph when compared with the brown morph (ANOVA, F1,20 = 132.817, p = 2.77 × 10−10). Pheomelanin was mostly concentrated in feather barbs and rachis (ﬁgure 1), suggesting that it is produced and deposited in a timely fashion during feather growth [70]. We also found a highly signiﬁcant relationship between PC1 scores and the pheomelanin marker (Spearman’s ρ = 0.834, p-value = 7.6 × 10−6), consistent with the fact that most of the variation in reﬂectance spectra between morphs is related to variation in pheomelanin content. 3.2. Lack of population structure and Mendelian inheritance reveal a true genetic polymorphism with single-locus control 3.1. Differences in melanin-based pigmentation patterns between morphs are functionally significant Lower stars: 7 GBS SNPs associated with a p-value < 1 × 10−4. The region including all 100 kb intervals with at least three highly associatedSNPsisshowninorange,andthethreeintervalswiththehighestdensityofassociatedSNPsareshowninred.LASTZalignment between scaffold 40 and the zebra finch chromosome 1 is also illustrated. 7 rsos.royalsocietypublishing.org . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.4 –0.2 0 0.2 0.4 –0.4 0 0.2 0.4 PC1 (19.72%) PC2 (18.51%) Pas de Bellecombe Bois ozoux Tévelave 1 2 0 chr1 chr10 chr11 chr12 chr13 chr14 chr15 chr16 chr17 chr18 chr19 chr1A chr1B chr2 chr20 chr21 chromosome chr22 chr23 chr24 chr25 chr26 chr27 chr28 chr3 chr4 chr4A chr5 chr6 chr7 chr8 chr9 chrLG2 chrLG5 chrLGE22 chrM chrZ eBPis 0.6 (a) (b) 7 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (a) 7 –0.4 –0.2 0 0.2 0.4 –0.4 PC1 (19.72%) Pas de Bellecombe –0.4 –0.3 –0.2 –0.1 0 0.1 0.2 –0.2 0 0.2 0.4 PC1 (5.62%) PC2 (5.55%) Pas de Bellecombe Bois Ozoux Tévelave 0 chr1 chr10 chr11 chr12 chr13 chr14 chr15 chr16 chr17 chr18 chr19 chr1A chr1B chr2 chr20 chr21 chromosome chr22 chr23 chr24 chr25 chr26 chr27 chr28 chr3 chr4 chr4A chr5 chr6 chr7 chr8 chr9 chrLG2 chrLG5 chrLGE22 chrM chrZ position on Z. lateralis scaffold 40 frequency of SNPs highly associated (eBPis < 1.176) 0 1 × 106 2 × 106 3 × 106 5 × 106 4 × 106 0 1 × 106 1.5 × 107 1.7 × 107 1.9 × 107 2.1 × 107 2 × 106 3 × 106 5 × 106 4 × 106 0 2 4 6 8 * * * * * * * * * * * * * * position on zebra finch chromosome 1 Z. lateralis scaffold 40 (c) Figure2. Populationstructureandassociationstudy.(a)PCAonallelefrequenciesforpooledRAD-seq(top)andindividualGBS(bottom) data. (b) Genome-wide association analysis performed with BAYPASS for 627 795 RAD-seq loci. Horizontal dashed lines correspond to 0.001% (grey) and 0.01 (black) significance thresholds. (c) Density of highly associated SNPs on scaffold 40 from the Zosterops lateralis referencegenome.Upperstars:SNPsdisplayingbothhighlysignificantBFandempiricalBayesianp-values(abovethe0.001%threshold). Lower stars: 7 GBS SNPs associated with a p-value < 1 × 10−4. 3.4. Functionally important changes lie outside candidate gene coding regions Having pinpointed the locus responsible for plumage colour variation, we searched for functional polymorphism at non-synonymous sites in the candidate region by obtaining whole-genome sequences from 12 individuals sampled across three distinct families from different populations (electronic supplementary material, table S4). Coefﬁcients of kinship estimated from whole-genome data for these individuals were consistent with the pedigrees estimated from microsatellite data (electronic supplementary material, table S5). The highest density of associated SNPs was found between 17.7 and 18.7 Mb on the zebra ﬁnch chromosome 1 (ﬁgure 3a). Only one non-synonymous substitution at a level of signiﬁcance less than 1 × 10−3 was found in ACE2, an angiotensin (ﬁgure 3a). The induced change led to the substitution of methionin to isoleucin, two hydrophobic amino acids which are extremely similar. Therefore, there is no non-synonymous change that would make a suitable candidate for the observed phenotypic variation. This suggests a role for regulatory mutations in determining the observed phenotype. 3.3. Genome-wide association analysis identifies a single genomic region associated with colour phenotype 3.3. Genome-wide association analysis identifies a single genomic region associated with colour phenotype To identify the genomic regions responsible for the observed colour polymorphism, we performed a genome-wide association analysis on a total of 627 795 SNPs obtained from the pooled RAD-sequencing data, and mapped it onto the Z. lateralis genome (ﬁgure 2b). Thirty-three SNPs had an empirical Bayesian p-value (eBPis) higher than 1.48 (at the 0.001% threshold). Among these SNPs, 28 were found on scaffold 40 on the reference genome for Z. lateralis which covers approximately positions 15–21 Mb on chromosome 1 of the better assembled zebra ﬁnch reference genome (ﬁgure 2c). Density of strongly associated SNPs (with an eBPis above the 0.01% threshold) was particularly high around 3 Mb of this scaffold, which corresponds approximately to position 18 Mb on the zebra ﬁnch chromosome 1. This pattern was further conﬁrmed by individual GBS data, with which seven variants were found to be associated to coloration with an uncorrected p-value between 9.1 × 10−5 and 8.6 × 10−7 on scaffold 40, while only two other variants were found to be associated with similar levels of signiﬁcance on scaffold 172, which is directly upstream of scaffold 40 according to the zebra ﬁnch reference genome. In sum, we found a single genomic region displaying a clear signal of association with phenotype. Within this region, we identiﬁed seven candidate genes based on their known role or on the role of homologues in melanic pigmentation (table 3). Among those genes, four—AP1S2, GPM6B, RAB9A and TRAPPC2—lie within regions where associated SNP density is highest (ﬁgure 2). rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . parental phenotype brown offspring grey offspring brown × brown 16 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . grey × grey 1 8 (+1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . grey × brown 6 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Matings involving two brown parents systematically led to brown offspring, while grey offspring always had at least one grey parent (table 2), suggesting a single-locus control with grey dominant to brown. Matings involving two brown parents systematically led to brown offspring, while grey offspring always had at least one grey parent (table 2), suggesting a single-locus control with grey dominant to brown. rg R.Soc. opensci. 4: 160805 3.2. Lack of population structure and Mendelian inheritance reveal a true genetic polymorphism with single-locus control Previous ﬁeld observations suggested a lack of assortative mating relative to plumage colour in white- eye populations, which should lead to a lack of genetic structure according to morph [30]. Pooled RAD-sequencing and GBS data on 137 and 42 individuals, respectively, from three geographically separate populations revealed that genetic structure was consistent with geography but not with colour morphs. PCA on allele frequencies from pooled RAD-seq data highlighted a clear grouping of brown and grey pools by locality (ﬁgure 2a). This lack of structure was also found for individual GBS data (ﬁgure 2b) and was further conﬁrmed by an AMOVA (electronic supplementary material, table S3) which revealed a signiﬁcant effect of population structure in shaping neutral genetic diversity (Fst = 0.022, p-value < 1 × 10−5), yet no effect of colour morph (Fct = −0.011, p-value = 1). To study the pattern of inheritance of the two colour morphs, we generated pedigrees for 36 putative parent–offspring triads based on the genotyping of 11 microsatellite markers in a total of 260 individuals. 8 rs. Table 2. Link between parent and offspring Zosteropsborbonicus phenotypes. Results were obtained from a pedigree analysis based on 11microsatellitesandmorphometricmeasurementsotherthancolour.Valuescorrespondtopedigreereconstructionscongruentbetween analyses including and excluding morphometric measurements. +1: found in analysis with genetic data only. 8 rs. . Table 2. Link between parent and offspring Zosteropsborbonicus phenotypes. Results were obtained from a pedigree analysis based on 11microsatellitesandmorphometricmeasurementsotherthancolour.Valuescorrespondtopedigreereconstructionscongruentbetween analyses including and excluding morphometric measurements. +1: found in analysis with genetic data only. 8 3.5. Evidence for a selective sweep suggests recent selection on the grey allele To assess whether selection had a role in the emergence of the observed polymorphism and to identify which of the two allele categories (brown or grey) is ancestral, we sampled ancestral recombination graphs for the entire candidate scaffold. Here, we took advantage of the fact that the individuals used in this test displayed six ‘brown’ and six ‘grey’ alleles at the most strongly associated markers. We found that the candidate region displayed a clear signature of a recent selective sweep, with an extremely short half-TMRCA between positions 2 700 000 and 3 100 000 on scaffold 40 (ﬁgure 3b). An examination of the local trees for regions containing SNPs associated with phenotype clearly showed that the most recent . Table3. SummaryofsevencandidategenesforcolourvariationinZosteropsborbonicus.InformationbasedonOMIM(OnlineMendelian Inheritance in Man), a database reporting large-scale genotype–phenotype associations in humans and laboratory mice. Positions correspond to coordinates on zebra finch reference genome. 9 sos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . candidate gene complete name position (from first to last exon) role in melanocyte impact on phenotype references RS1 retinoschisin 15 288 297– 15 291 738 retinal cells adhesion, cell–cell interaction depigmentation in retinal pigment epithelium [71] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Evidence for a selective sweep suggests recent selection on the grey allele . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AP1S2 adaptor-relatedprotein complex1,sigma2 subunit 16 724 833– 16 754 005 AP-1 complex is involved in melanosome genesis and is necessary for TYRP1 to reach the melanosome and produce eumelanin mutations on another complex, AP-3, lead to the pearl (Ap3b1; coat hypopigmentation) and mocha (Ap3d1; coat colour dilution) phenotypes in mice [72–74] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GPM6B glycoproteinM6B 17 812 158– 17 843 054 membrane protein involved in neuronal tissues. Up-regulated by MITF — [75] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Evidence for a selective sweep suggests recent selection on the grey allele . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TRAPPC2 traffickingprotein particlecomplex, subunit2 17 877 165– 17 880 771 TRAPP complex is involved in vesicle transport and tethering in mice, mutations on the subunit TRAPPC6A lead to hypopigmented patches in the coat and retinal epithelium [76] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Evidence for a selective sweep suggests recent selection on the grey allele . . . . . . . . RAB9A Ras-associatedprotein 9A 17 886 048– 17 886 836 regulation of vesicular trafficking. Interacts with BLOC-3, involved in Hermansky-Pudlak syndrome Rab38: chocolate phenotype in mice. Rab27a: Griscelli syndrome (hypomelanosis and neurological defects) [28,77,78] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . APXL apicalproteinof Xenopus-like 19 986 240– 20 035 565 melanosome biogenesis and transport. Activity requires the G-protein RAB27A ocular albinism [79] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Evidence for a selective sweep suggests recent selection on the grey allele . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OA1 ocularalbinism1 20 136 404– 20 149 393 melanosome transport and interactions with cytoskeleton ocular albinism [80,81] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . haplotypes were found in grey individuals only (ﬁgure 3c). These haplotypes also harboured the allele associated with grey coloration in the association analysis. Median half-TMRCA was estimated at 1733 generations in the region, signiﬁcantly shorter than the estimate for the remaining regions included in the analysis (median: 54 460 generations; Wilcoxon rank test, p < 2.2 × 10−16). 3.5. Evidence for a selective sweep suggests recent selection on the grey allele This indicates that grey coloration has been positively selected and that brown and grey phenotypes represent ancestral and derived conditions, respectively. 4. Discussion We conﬁrmed that highland populations of the Reunion grey white-eye present a genetic colour polymorphism that results from the differential deposition of pheomelanin on feathers, leading to predominantly grey or brown plumage, and is controlled by a simple genetic mechanism with a recessive ‘brown’ allele and a dominant ‘grey’ allele. A genome-wide association analysis designed to search for loci involved in the polymorphism led to the identiﬁcation of a single genomic region on chromosome 1 characterized by SNPs strongly associated with colour phenotype. A coalescent analysis of selection pinpoints the same region, and suggests recent selection on the grey allele. Annotation of these SNPs 10 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 2 × 106 4 × 106 6 × 106 1 × 105 1 × 105 3 × 106 1 × 106 5 × 106 5 × 106 4 × 106 3 × 106 2 × 106 1 × 106 0 0 position on scaffold 40 position on scaffold 40 position on scaffold 40 1 2 3 4 –log10 (p-value) * * * frequency of mutations with a p-value <10−3 0 10 20 30 40 50 60 * * * 0 50 000 100 000 150 000 200 000 mean half TMRCA * * * locus 2955711 1588 (grey) 314 (brown) 1430 (grey) 1337 (grey) 317 (grey) 314 (brown) 1588 (grey) 317 (grey) 1434 (grey) 1430 (grey) 1337 (grey) 1434 (grey) locus 2994089 314 (brown) 1588 (grey) 314 (brown) 1430 (grey) 317 (grey) 1337(grey) 1337 (grey) 1588 (grey) 317 (grey) 1434 (grey) 1434 (grey) 1430 (grey) locus 3068841 1588 (grey) 314 (brown) 317 (grey) 314 (brown) 1337 (grey) 1430 (grey) 1588 (grey) 1430 (grey) 1434 (grey) 1337 (grey) 1434 (grey) 317 (grey) 50 000 grey haplotypes brown haplotypes grey haplotypes brown haplotypes grey haplotypes brown haplotypes (a) (b) (c) Figure 3. Test for selection using whole-genome sequencing. (a) Association analysis of whole-genome sequencing data, including four brown and eight grey individuals from three different families. Synonymous and non-synonymous SNPs are highlighted in blue and red, respectively. Stars indicate the positions of the seven candidate genes listed in table 3. 4. Discussion Density of SNPs associated with a p-value < 1 × 10−3 isalsoprovided.(b)Plotofhalf-TMRCA(innumberofgenerations)forscaffold40.Starsindicatethepositionofthe threenon-recombiningblocksexaminedin(c).(c)Genealogiesobservedatthreedistinctpointsfromthecolourlocus.Allthreeblocksof sequences included SNPs associated with colour with a p-value < 1 × 10−3 in the whole-genome association analysis. Branch lengths represent time in generations. Individual phenotype is indicated in brackets. Individuals 1434 (grey morph) and 314 (brown morph) are found to be homozygous at SNPs strongly associated with colour. 6 5 × 106 4 × 106 3 × 106 2 × 106 1 × 106 0 position on scaffold 40 0 50 000 100 000 150 000 200 000 mean half TMRCA * * * (b) 10 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (b) (a) 10 Figure 3. Test for selection using whole-genome sequencing. (a) Association analysis of whole-genome sequencing data, including four brown and eight grey individuals from three different families. Synonymous and non-synonymous SNPs are highlighted in blue and red, respectively. Stars indicate the positions of the seven candidate genes listed in table 3. Density of SNPs associated with a p-value < 1 × 10−3 isalsoprovided.(b)Plotofhalf-TMRCA(innumberofgenerations)forscaffold40.Starsindicatethepositionofthe threenon-recombiningblocksexaminedin(c).(c)Genealogiesobservedatthreedistinctpointsfromthecolourlocus.Allthreeblocksof sequences included SNPs associated with colour with a p-value < 1 × 10−3 in the whole-genome association analysis. Branch lengths represent time in generations. Individual phenotype is indicated in brackets. Individuals 1434 (grey morph) and 314 (brown morph) are found to be homozygous at SNPs strongly associated with colour. using the Z. lateralis reference genome [56] did not reveal any non-synonymous mutations or deletions in coding parts of the genomic region associated with variation between colour morphs, indicating that functionally important nucleotide changes in the Reunion grey white-eye must lie outside coding regions. This suggests a role for cis-regulatory mutations that either prevent the synthesis of pheomelanin in grey feathers, or increase its production in brown ones. The ﬁrst mechanism seems more likely in the present case since the grey phenotype seems to represent the derived condition. The fact that eumelanin is found in both brown and grey feathers, while pheomelanin is only present in brown feathers suggests that the underlying genes may be involved in melanogenesis and melanosome transport. rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A large-effect locus can break up into many small- to moderate-effect mutations [32,84]. Thus, although the pattern of inheritance of plumage colour and the existence of a single strongly associated genomic region points towards a relatively simple genetic mechanism, it remains possible that several of the candidate genes we identiﬁed are simultaneously involved in shaping the observed colour variation in this study. If mutations arose independently at several genes involved in melanogenesis, their physical proximity would favour linkage between the best combinations of mutations by sieve. Structural rearrangements such as large inversions covering the genomic region associated to colour, if any, would also stabilize associations between several alleles [85]. In that case, the colour locus that we identiﬁed in this study may be a further example of the so-called ‘largesse of the genome’ [86,87]—i.e. the predisposition of certain genomic regions to mediate integrated phenotypic shifts—in vertebrates. hing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genes like RS1, APXL or OA1 have been studied in the context of ocular diseases or melanomas [88], in a way that suggests important functions in melanocyte development (table 3). Therefore, they probably play a role in ocular development and in colour recognition, which might have an impact on how individuals recognize and favour conspeciﬁcs. In addition, GPM6B is known to be involved in neural development, serotonin uptake and bone formation [89,90]. Thus, this gene could inﬂuence feather colour, brain development and behaviour. This feature is particularly interesting, since changes in behavioural strategies associated with visual signalling can explain the persistence of morphs over time [91]. Our work reveals a new genomic region not previously associated with melanic coloration in vertebrates, and underscores the importance of genome-based research on non-model species to understand the genetic basis of ecologically signiﬁcant traits and their role in phenotypic diversiﬁcation [92]. Our results suggest a strong selective advantage for the dominant ‘grey’ allele once it arose from a new mutation, leading to its fast spread across all highland populations of the Reunion grey white-eye [93]. Similar patterns of selection on de novo mutations at colour genes have been recently described in deer mice (Peromyscus maniculatus), where multiple independent mutations were selected for cryptic coloration after the colonization of a novel selective environment [31,32]. rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . In white-eyes, both colour morphs are found at an appreciable frequency in all populations, implying that the selected allele does not go to ﬁxation, but instead reaches an intermediate frequency, as would be expected under balancing selection. This may happen as the result of, for example, heterozygote advantage or negative frequency-dependent selection [94]. Distinguishing between these possibilities is notoriously difﬁcult, and future studies should estimate the relative survival and fertility rates of the different ‘colour’ genotypes in order to determine the nature of selection acting on this colour polymorphism and to come closer to an explanation about its persistence. Ethics. Measurements and manipulations were performed by T.D. and B.M. Manipulation of live birds for ringing, measurements and blood sampling was done under a ringing permit (number 14 631) issued by CRBPO Museum National d’Histoire Naturelle (Paris, France), and in compliance with ethical guidelines at Université Paul Sabatier. A special authorization to work on protected species was issued by Direction Régionale de l’Environnement (DIREN Réunion). Data accessibility. All data are deposited on Dryad (phenotypic data, GBS, http://dx.doi.org/10.5061/dryad.1q9g7) [95] and European Nucleotide Archive repositories (raw reads for whole genomes and pooled RAD-seq, accession nos. ERP020509 and ERP002555). Methods and options are further detailed in the electronic supplementary material. Scripts used in this study are freely available on GitHub at https://github.com/YannBourgeois/Scripts_Pool_ RADseq. Authors’ contributions. B.M. and C.T. initiated, coordinated and supervised the project, and coordinated the collection of ﬁeld samples; Y.B., B.M. and C.T. conceived the study and designed the experiments. Molecular data were obtained by Y.B. with contributions from E.L., J.A.M.B., B.D., J.B., C.M., H.H. and O.B.; pigment data were obtained by K.W.; Y.B. and M.G. analysed the data, with contributions from P.J.M., B.D. and J.C.; and Y.B., B.M. and C.T. wrote the paper with comments from other authors. All authors gave their ﬁnal approval for publication. Competing interests. We declare we have no competing interests. ompeting interests. We declare we have no competing interests. ng interests. We declare we have no competing interests. Funding. This work was supported by Fondation pour la Recherche sur la Biodiversité (FRB), Agence Française pour le Développement (AFD), Agence Nationale de la Recherche (ANR-2006-BDIV002), Centre National de la Recherche Scientiﬁque (CNRS) through a PEPS grant, the National Geographic Society and the ‘Laboratoire d’Excellence’ TULIP (ANR-10-LABX-41). The ﬁrst author was supported by a MESR (Ministère de l’Enseignement Supérieur et de la Recherche) PhD scholarship during this study. 4. Discussion Importantly, the genomic scaffold associated with plumage colour does not include, to the best of our knowledge, genes previously known to be involved in hair or feather coloration in other species, even in model species such as laboratory mice. Three genes found in this region, RS1, APXL and OA1, are known to affect melanogenesis and pigmentation in retinal pigment cells. In addition to these genes, the colour locus includes genes (GPM6B, RAB9A, TRAPPC2, AP1S2) known to be involved in intracellular trafﬁcking and in shaping melanosomes in melanocytes (table 3). Among them, GPM6B is of particular interest as it is regulated by MITF, a key transcription factor involved in melanocyte speciﬁcation [75]. However, while changes in GPM6B expression were found in a comparison of black and hooded crows [30], there is currently no direct evidence for a role of this gene in feather coloration. Our results also conﬁrm previous ﬁndings suggesting that variation at genes often studied in natural populations, such as MC1R, Agouti or POMC [32,82,83], is not directly related to colour variation in the Reunion grey white-eye [38] and provide an impressive illustration of the diversity of the mechanisms underlying melanin-based plumage colour evolution in birds and perhaps other vertebrates. Future work should focus on gene expression differentiation to deﬁne the developmental basis of this colour polymorphism and to determine, for example, if the candidate loci found on chromosome 1 are epistatic over other genes classically found associated with melanin-based colour variation. work should focus on gene expression differentiation to deﬁne the developmental basis of this colour polymorphism and to determine, for example, if the candidate loci found on chromosome 1 are epistatic over other genes classically found associated with melanin-based colour variation. 11 References 1. Hoekstra HE. 2006 Genetics, development and evolution of adaptive pigmentation in vertebrates. Heredity 97, 222–234. (doi:10.1038/sj.hdy.6800861) 1. Hoekstra HE. 2006 Genetics, development and evolution of adaptive pigmentation in vertebrates. Heredity 97, 222–234. (doi:10.1038/sj.hdy.6800861) 17. Gangoso L, Roulin A, Ducrest AL, Grande JM, Figuerola J. 2015 Morph-specific genetic and environmental variation in innate and acquired immune response in a color polymorphic raptor. Oecologia 178, 1113–1123. (doi:10.1007/s00442- 015-3306-6) . . . . . . . . . . 28. Bennett DC, Lamoreux ML. 2003 Pigment gene focus the color loci of mice—a genetic century. PigmentCellRes. 16, 333–344. (doi:10.1034/j.1600- 0749.2003.00067.x) 2. Huxley J. 1955 Morphism in Birds. In ActaXI InternationalCongressofOrnithology (eds A Portmann, E Sutter), pp. 309–328. Basel, Switzerland: Birkhäuser Verlag. 2. Huxley J. 1955 Morphism in Birds. In ActaXI InternationalCongressofOrnithology (eds A Portmann, E Sutter), pp. 309–328. Basel, Switzerland: Birkhäuser Verlag. 29. Kingsley EP, Manceau M, Wiley CD, Hoekstra HE, Borevitz JO. 2009 Melanism in peromyscus is caused by independent mutations in agouti. PLoSONE 4, e6435. (doi:10.1371/journal.pone.0006435) 18. Rosenblum EB, Hoekstra HE, Nachman MW. 2004 Adaptive reptile color variation and the evolution of the Mc1r gene. Evolution 58, 1794–1808. (doi:10.1554/03-741) 3. Ford EB. 1975 Ecologicalgenetics, 4th edn. London, UK: Chapman and Hall. 3. Ford EB. 1975 Ecologicalgenetics, 4th edn. London, UK: Chapman and Hall. 30. Poelstra JW, Vijay N, Hoeppner MP, Wolf JBW. 2015 Transcriptomics of colour patterning and coloration shifts in crows. Mol.Ecol. 24, 4617–4628. (doi:10.1111/mec.13353) 4. Wright S. 1943 An analysis of local variability of flower color in Linanthus parryae. Genetics 28, 139–156. 4. Wright S. 1943 An analysis of local variability of flower color in Linanthus parryae. Genetics 28, 139–156. 19. Mullen LM, Hoekstra HE. 2008 Natural selection along an environmental gradient: a classic cline in mouse pigmentation. Evolution 62, 1555–1570. (doi:10.1111/j.1558-5646.2008.00425.x) 5. Gray SM, McKinnon JS. 2007 Linking color polymorphism maintenance and speciation. Trends Ecol.Evol. 22, 71–79. (doi:10.1016/j.tree.2006.10.005) 31. Linnen CR, Kingsley EP, Jensen JD, Hoekstra HE. 2009 On the origin and spread of an adaptive allele in deer mice. Science 325, 1095–1098. (doi:10.1126/ science.1175826) 20. Rees JL. 2003 Genetics of hair and skin color. Annu. Rev.Genet. 37, 67–90. (doi:10.1146/annurev.genet. 37.110801.143233) 6. Lewontin R. 1974 Thegeneticbasisofevolutionary change. New York, NY: Columbia University Press. 32. Linnen CR, Poh Y-P, Peterson BK, Barrett RDH, Larson JG, Jensen JD, Hoekstra HE. 2013 Adaptive evolution of multiple traits through multiple mutations at a single gene. Science 339, 1312–1316. (doi:10.1126/science.1233213) 21. References Römpler H, Rohland N, Lalueza-Fox C, Willerslev E, Kuznetsova T, Rabeder G, Bertranpetit J, Schöneberg T, Hofreiter M. 2006 Nuclear gene indicates coat-color polymorphism in mammoths. Science 313, 62. (doi:10.1126/science.1128994) 7. Manceau M, Domingues VS, Linnen CR, Rosenblum EB, Hoekstra HE. 2010 Convergence in pigmentation at multiple levels: mutations, genes and function. Phil.Trans.R.Soc.B 365, 2439–2450. (doi:10.1098/ rstb.2010.0104) 33. Poelstra JW etal. 2014 The genomic landscape underlying phenotypic integrity in the face of gene flow in crows. Science 344, 1410–1414. (doi:10.1126/ science.1253226) 8. McGraw KJ, Hill GE. 2000 Differential effects of endoparasitism on the expression of carotenoid- and melanin-based ornamental coloration. Proc.R. Soc.Lond.B 267, 1525–1531. (doi:10.1098/rspb. 2000.1174) 22. Theron E, Hawkins K, Bermingham E, Ricklefs RE, Mundy NI. 2001 The molecular basis of an avian plumage polymorphism in the wild: a melanocortin-1-receptor point mutation is perfectly associated with the melanic plumage morph of the bananaquit, Coerebaflaveola. Curr.Biol. 11, 550–557. (doi:10.1016/S0960-9822(01)00158-0) 34. Ellegren H. etal. 2012 The genomic landscape of species divergence in Ficedula flycatchers. Nature 491, 756–760. (doi:10.1038/nature11584) 9. McGraw KJ, Mackillop EA, Dale J, Hauber ME. 2002 Different colors reveal different information: how nutritional stress affects the expression of melanin- and structurally based ornamental plumage. J.Exp. Biol. 205, 3747–3755. (doi:10.1007/s00265-010- 1135-5) 35. Gill F. 2007 Ornithology, 3rd edn. New York, NY: Freeman & Co. 23. Mundy NI, Badcock NS, Hart T, Scribner K, Janssen K, Nadeau NJ. 2004 Conserved genetic basis of a quantitative plumage trait involved in mate choice. Science 303, 1870–1873. (doi:10.1126/science. 1093834) 36. Gill FB. 1973 Intra-island variation in the Mascarene white-eye Zosteropsborbonica. Ornithol.Monogr. 12, 1–66. (doi:10.2307/40166689) 37. Bourgeois Y, Bertrand J, Delahaie B, Cornuault J, Duval T, Milá B, Thébaud C. 2016 Candidate gene analysis suggests untapped genetic complexity in melanin-based pigmentation in birds. J.Hered. 107, 327–335. (doi:10.1093/jhered/esw017) 10. McGraw KJ. 2008 An update on the honesty of melanin-based color signals in birds. PigmentCell MelanomaRes. 21, 133–138. (doi:10.1111/j.1755- 148X.2008.00454.x) 24. Haas F, Pointer MA, Saino N, Brodin A, Mundy NI, Hansson B. 2009 An analysis of population genetic differentiation and genotype-phenotype association across the hybrid zone of carrion and hooded crows using microsatellites and MC1R. Mol.Ecol. 18, 294–305. (doi:10.1111/j.1365-294X. 2008.04017.x) 11. Roulin A. 2016 Condition-dependence, pleiotropy and the handicap principle of sexual selection in melanin-based colouration. Biol.Rev. 91, 328–348. (doi:10.1111/brv.12171) 38. Bourgeois YXC, Bertrand JAM, Thébaud C, Milá B. rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements. We thank Alexandre Roulin, Dieter Ebert, Anne Roulin, Marie Manceau and one anonymous reviewer for insightful comments and assistance that greatly improved the manuscript. We want to thank the . . Reunion National Park for granting us permission to conduct ﬁeldwork and to collect blood samples. Thomas Duval, Guillaume Gélinaud, Philipp Heeb, Dominique Strasberg, Ben Warren and Juli Broggi assisted with ﬁeldwork. Isabelle Bourgeois provided assistance with drawing ﬁgures. 12 References 2012 Investigating the role of the melanocortin-1 receptor gene in an extreme case of microgeographical variation in the pattern of melanin-based plumage pigmentation. PLoSONE 7, e50906. (doi:10.1371/journal.pone.0050906) 12. Roulin A, Ducrest A-L. 2013 Genetics of colouration in birds. Semin.CellDev.Biol. 24, 594–608. (doi:10.1016/j.semcdb.2013.05.005) 25. Uy JAC, Moyle RG, Filardi CE, Cheviron ZA. 2009 Difference in plumage color used in species recognition between incipient species is linked to a single amino acid substitution in the melanocortin-1 receptor. Am.Nat. 174, 244–254. (doi:10.1086/600084) 39. Bertrand JA, García-Jiménez R, Bourgeois Y, Duval T, Heeb P, Thébaud C, Milá B. 2012 Isolation and characterization of twelve polymorphic microsatellite loci for investigating an extreme case of microgeographical variation in an island bird (Zosteropsborbonicus). Conserv.Genet.Resour. 4, 323–326. (doi:10.1007/s12686-011-9538-9) 13. Roulin A. 2004 The evolution, maintenance and adaptive function of genetic colour polymorphism in birds. Biol.Rev.Camb.Philos.Soc. 79, 815–848. (doi:10.1017/S1464793104006487) 26. Cheviron Z, Hackett SJ, Brumfield RT. 2006 Sequence variation in the coding region of the melanocortin-1 receptor gene (MC1R) is not associated with plumage variation in the blue-crowned manakin (Lepidothrixcoronata). Proc.R.Soc.B 273, 1613–1618. (doi:10.1098/ rspb.2006.3499) 14. Ducrest A-L, Keller L, Roulin A. 2008 Pleiotropy in the melanocortin system, coloration and behavioural syndromes. TrendsEcol.Evol. 23, 502–510. (doi:10.1016/j.tree.2008.06.001) 40. Bourgeois YXC etal. 2013 Mass production of SNP markers in a nonmodel passerine bird through RAD sequencing and contig mapping to the zebra finch genome. Mol.Ecol.Resour. 13, 899–907. (doi:10.1111/1755-0998.12137) 15. Galeotti P, Rubolini D, Dunn PO, Fasola M. 2003 Colour polymorphism in birds: causes and functions. J.Evol.Biol. 16, 635–646. (doi:10.1046/ j.1420-9101.2003.00569.x) 15. Galeotti P, Rubolini D, Dunn PO, Fasola M. 2003 Colour polymorphism in birds: causes and functions. J.Evol.Biol. 16, 635–646. (doi:10.1046/ j.1420-9101.2003.00569.x) 27. MacDougall-Shackleton EA, Blanchard L, Igdoura SA, Gibbs HL. 2003 Unmelanized plumage patterns in Old World leaf warblers do not correspond to sequence variation at the melanocortin-1 receptor locus (MC1R). Mol.Biol.Evol. 20, 1675–1681. (doi:10.1093/molbev/msg186) 16. McKinnon JS, Pierotti MER. 2010 Colour polymorphism and correlated characters: genetic mechanisms and evolution. Mol.Ecol. 19, 5101–5125. (doi:10.1111/j.1365-294X.2010.04846.x) 41. Davey JW, Davey JL, Blaxter ML, Blaxter MW. 2010 RADSeq: next-generation population genetics. Brief.Funct.Genomics 9, 416–423. (doi:10.1093/ bfgp/elq031) 73. Theos A, Tenza D, Martina J. 2005 Functions of adaptor protein (AP)-3 and AP-1 in tyrosinase sorting from endosomes to melanosomes. Mol.Biol. Cell 16, 5356–5372. (doi:10.1091/mbc.E05) 57. Li H, Durbin R. 2009 Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760. (doi:10.1093/ bioinformatics/btp324) 42. 13 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58. Schwartz S, Kent W, Smit A. 2003 Human–mouse alignments with BLASTZ. GenomeRes. 13, 103–107. (doi:10.1101/gr.809403) 74. Feng L etal. 1999 The beta3A subunit gene (Ap3b1) of the AP-3 adaptor complex is altered in the mouse hypopigmentation mutant pearl, a model for Hermansky-Pudlak syndrome and night blindness. Hum.Mol.Genet. 8, 323–330. (doi:10.1093/hmg/ 8.2.323) 43. Seutin G, White BN, Boag PT. 1991 Preservation of avian blood and tissue samples for DNA analyses. Can.J.Zool. 69, 82–90. (doi:10.1139/z91-013) 59. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R. 2009 The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079. (doi:10.1093/bioinfor matics/btp352) 44. Griffiths R, Double M, Orr K, Dawson RJG. 1998 A DNA test to sex most birds. Mol.Ecol. 7, 1071–1075. (doi:10.1046/j.1365-294x.1998.00389.x) 75. Hoek KS etal. 2008 Novel MITF targets identified using a two-step DNA microarray strategy. Pigment CellMelanomaRes. 21, 665–676. (doi:10.1111/ j.1755-148X.2008.00505.x) 45. Cuthill IC, Bennett ATD, Partridge JC, Maier EJ. 1999 Plumage reflectance and the objective assessment of avian sexual dichromatism. Am.Nat. 153, 183–200. (doi:10.1086/303160) 60. Kofler R, Pandey RV, Schlötterer C. 2011 PoPoolation2: identifying differentiation between populations using sequencing of pooled DNA samples (Pool-Seq). Bioinformatics 27, 3435–3436. (doi:10.1093/bioinformatics/btr589) 76. Gwynn B, Smith RS, Rowe LB, Taylor BA, Peters LL. 2006 A mouse TRAPP-related protein is involved in pigmentation. Genomics 88, 196–203. (doi:10.1016/ j.ygeno.2006.04.002) 46. Endler JA. 1990 On the measurement and classification of colour in studies of animal colour patterns. Biol.J.Linn.Soc. 41, 315–352. (doi:10.1111/ j.1095-8312.1990.tb00839.x) 61. Patterson N, Price AL, Reich D. 2006 Population structure and eigenanalysis. PLoSGenet. 2, e190. (doi:10.1371/journal.pgen.0020190) 77. Ohbayashi N, Fukuda M. 2012 Role of Rab family GTPases and their effectors in melanosomal logistics. J.Biochem. 151, 343–351. (doi:10.1093/ jb/mvs009) 47. Vorobyev M, Osorio D. 1998 Receptor noise as a determinant of colour thresholds. Proc.R.Soc.Lond. B 265, 351–358. (doi:10.1098/rspb.1998.0302) 62. Huber W etal. 2015 Orchestrating high-throughput genomic analysis with Bioconductor. Nat.Methods 12, 115–121. (doi:10.1038/nmeth.3252) 48. Cornuault J, Delahaie B, Bertrand JAM, Bourgeois YXC, Milá B, Heeb P, Thébaud C. 2015 Morphological and plumage colour variation in the Réunion grey white-eye (Aves: Zosteropsborbonicus): assessing the role of selection. Biol.J.Linn.Soc. 114, 459–473. (doi:10.1111/bij.12428) 78. Kloer DP etal. 2010 Assembly of the biogenesis of lysosome-related organelles complex-3 (BLOC-3) and its interaction with Rab9. J.Biol. Chem. 285, 7794–7804. (doi:10.1074/jbc.M109. 069088) 63. Excoffier L, Lischer HEL. 2010 Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. 13 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mol. Ecol.Resour. 10, 564–567. (doi:10.1111/j.1755-0998. 2010.02847.x) 79. Fairbank PD, Lee C, Ellis A, Hildebrand JD, Gross JM, Wallingford JB. 2006 Shroom2 (APXL) regulates melanosome biogenesis and localization in the retinal pigment epithelium. Development 133, 4109–4118. (doi:10.1242/dev. 02563) 64. Gautier M. 2015 Genome-wide scan for adaptive divergence and association with population-specific covariates. Genetics 201, 1555–1579. (doi:10.1534/ genetics.115.181453) 49. Saino N, Romano M, Rubolini D, Teplitsky C, Ambrosini R, Caprioli M, Canova L, Wakamatsu K. 2013 Sexual dimorphism in melanin pigmentation, feather coloration and its heritability in the barn swallow (Hirundorustica). PLoSONE 8, e58024. (doi:10.1371/journal.pone.0058024) 65. Purcell S etal. 2007 PLINK: a tool set for whole-genome association and population-based linkage analyses. Am.J.Hum.Genet. 81, 559–575. (doi:10.1086/519795) 80. Palmisano I etal. 2008 The ocular albinism type 1 protein, an intracellular G protein-coupled receptor, regulates melanosome transport in pigment cells. Hum.Mol.Genet. 17, 3487–3501. (doi:10.1093/hmg/ 50. Ozeki H, Ito S, Wakamatsu K, Thody AJ. 1996 Spectrophotometric characterization of eumelanin and pheomelanin in hair. PigmentCellRes. 9, 265–270. (doi:10.1111/j.1600-0749.1996.tb00116.x) 66. Manichaikul A, Mychaleckyj JC, Rich SS, Daly K, Sale M, Chen W-M. 2010 Robust relationship inference in genome-wide association studies. Bioinformatics 26, 2867–2873. (doi:10.1093/bioinformatics/ btq559) Hum.Mol.Genet. 17, 3487–3501. (doi:10.1093/hmg/ ddn241) 51. Ito S, Nakanishi Y, Valenzuela RK, Brilliant MH, Kolbe L, Wakamatsu K. 2011 Usefulness of alkaline hydrogen peroxide oxidation to analyze eumelanin and pheomelanin in various tissue samples: application to chemical analysis of human hair melanins. PigmentCellMelanomaRes. 24, 605–613. (doi:10.1111/j.1755-148X.2011.00864.x) 81. Incerti B etal. 2000 Oa1 knock-out: new insights on the pathogenesis of ocular albinism type 1. Hum. Mol.Genet. 9, 2781–2788. (doi:10.1093/hmg/ 9.19.2781) 67. Li M, Boehnke M, Abecasis GR. 2006 Efficient study designs for test of genetic association using sibship data and unrelated cases and controls. Am.J.Hum.Genet. 78, 778–792. (doi:10.1086/ 503711) 82. Uy JAC etal. 2016 Mutations in different pigmentation genes are associated with parallel melanism in island flycatchers. Proc.R.Soc.B 283, 2115–2118. (doi:10.1098/rspb.2016.0731) 52. Wakamatsu K, Ito S, Rees JL. 2002 The usefulness of 4-amino-3-hydroxyphenylalanine as a specific marker of pheomelanin. PigmentCellRes. 15, 225–232. (doi:10.1034/j.1600-0749.2002. 02009.x) 68. Doran AG, Creevey CJ. 2013 Snpdat: easy and rapid annotation of results from de novo snp discovery projects for model and non-model organisms. BMCBioinform. 14, 45. (doi:10.1186/1471-2105- 14-45) 83. Roulin A, Emaresi G, Bize P, Gasparini J, Piault R, Ducrest A-L. 2011 Pale and dark reddish melanic tawny owls differentially regulate the level of blood circulating POMC prohormone in relation to environmental conditions. Oecologia 166, 913–921. (doi:10.1007/s00442-011-1955-7) 53. References Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, Mitchell SE, Orban L. 2011 A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoSONE 6, e19379. (doi:10.1371/journal.pone.0019379) 13 90. Fjorback AW, Müller HK, Wiborg O. 2009 Membrane glycoprotein M6B interacts with the human serotonin transporter. J.Mol.Neurosci. 37, 191–200. (doi:10.1007/s12031-008-9092-4) 89. Drabek K, van de Peppel J, Eijken M, van Leeuwen JPTM. 2011 GPM6B regulates osteoblast function and induction of mineralization by controlling cytoskeleton and matrix vesicle release. J.Bone Miner.Res. 26, 2045–2051. (doi:10.1002/jbmr.435) 13 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bertrand JAM etal. 2014 Extremely reduced dispersal and gene flow in an island bird. Heredity 112, 190–196. (doi:10.1038/hdy.2013.91) 69. Rasmussen MD, Hubisz MJ, Gronau I, Siepel A, Coop G. 2014 Genome-wide inference of ancestral recombination graphs. PLoSGenet. 10, e1004342. (doi:10.1371/journal.pgen.1004342) 84. Flint J, Mackay TFC. 2009 Genetic architecture of quantitative traits in mice, flies, and humans. GenomeRes. 19, 723–733. (doi:10.1101/gr.086660. 108.19) 54. Bertrand JAM, Delahaie B, Bourgeois YXC, Duval T, García-Jiménez R, Cornuault J, Pujol B. 2016 The role of selection and historical factors in driving population differentiation along an elevational gradient in an island bird. J.Evol.Biol.29, 824–836. (doi:10.1111/jeb.12829) 70. Prum RO, Dyck J. 2003 A hierarchical model of plumage: morphology, development, and evolution. J.Exp.Zool.B.Mol.Dev.Evol. 298B, 73–90. (doi:10.1002/jez.b.27) 85. Joron M etal. 2011 Chromosomal rearrangements maintain a polymorphic supergene controlling butterfly mimicry. Nature 477, 203–206. (doi:10.1038/nature10341) 55. Hadfield JD, Richardson DS, Burke T. 2006 Towards unbiased parentage assignment: combining genetic, behavioural and spatial data in a Bayesian framework. Mol.Ecol. 15, 3715–3730. (doi:10.1111/ j.1365-294X.2006.03050.x) 71. Weber BHF etal. 2002 Inactivation of the murine X-linked juvenile retinoschisis gene, Rs1 h, suggests a role of retinoschisin in retinal cell layer organization and synaptic structure. Proc.Natl Acad.Sci.USA 99, 6222–6227. (doi:10.1073/pnas. 092528599) 86. Mallet J, Barton N. 1989 Strong natural selection in a warning color hybrid zone. Evolution 43, 421–431. (doi:10.2307/2409217) 87. Turner JRG. 1977 Butterfly mimicry: the genetical evolution of an adaptation. Evol.Biol. 11, 163–206. (doi:10.1007/978-1-4615-6953-4_3) 56. Cornetti L, Valente LM, Dunning LT, Quan X, Black RA, Hébert OH, Savolainen V. 2015 The genome of the ‘great speciator’ provides insights into bird diversification. GenomeBiol.Evol. 7, 2680–2691. (doi:10.1093/gbe/evv168) 72. Delevoye C etal. 2009 AP-1 and KIF13A coordinate endosomal sorting and positioning during melanosome biogenesis. J.CellBiol. 187, 247–264. (doi:10.1083/jcb.200907122) 72. Delevoye C etal. 2009 AP-1 and KIF13A coordinate endosomal sorting and positioning during melanosome biogenesis. J.CellBiol. 187, 247–264. (doi:10.1083/jcb.200907122) 72. Delevoye C etal. 2009 AP-1 and KIF13A coordinate endosomal sorting and positioning during 88. Bai J, Xie X, Lei Y, An G, He L, Lv X. 2014 Ocular albinism type 1 induced melanoma cell migration is mediated through the RAS/RAF/MEK/ERK signaling 88. Bai J, Xie X, Lei Y, An G, He L, Lv X. 2014 Ocular albinism type 1 induced melanoma cell migration is mediated through the RAS/RAF/MEK/ERK signaling endosomal sorting and positioning during melanosome biogenesis. J.CellBiol. 187, 247–264. (doi:10.1083/jcb.200907122) melanosome biogenesis. J.CellBiol. 187, 247–264. (doi:10.1083/jcb.200907122) 14 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . 92. Abzhanov A, Extavour CG, Groover A, Hodges SA, Hoekstra HE, Kramer EM, Monteiro A. 2008 Are we there yet? Tracking the development of new model systems. TrendsGenet. 24, 353–360. (doi:10.1016/ j.tig.2008.04.002) 13 rsos.royalsocietypublishing.org R.Soc. opensci. 4: 160805 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pathway. Mol.Med.Rep. 10, 491–495. (doi:10.3892/ mmr.2014.2154) 89. Drabek K, van de Peppel J, Eijken M, van Leeuwen JPTM. 2011 GPM6B regulates osteoblast function and induction of mineralization by controlling cytoskeleton and matrix vesicle release. J.Bone Miner.Res. 26, 2045–2051. (doi:10.1002/jbmr.435) 90. Fjorback AW, Müller HK, Wiborg O. 2009 Membrane glycoprotein M6B interacts with the human serotonin transporter. J.Mol.Neurosci. 37, 191–200. (doi:10.1007/s12031-008-9092-4) 91. Sinervo B, Svensson E. 2002 Correlational selection and the evolution of genomic architecture. Heredity 89, 329–338. (doi:10.1038/sj.hdy. 6800148) 92. Abzhanov A, Extavour CG, Groover A, Hodges SA, Hoekstra HE, Kramer EM, Monteiro A. 2008 Are we there yet? Tracking the development of new model systems. TrendsGenet. 24, 353–360. (doi:10.1016/ j.tig.2008.04.002) 93. Teshima KM, Przeworski M. 2006 Directional positive selection on an allele of arbitrary dominance. Genetics 172, 713–718. (doi:10.1534/genetics.105. 044065) 94. Charlesworth B, Charlesworth D. 2010 Elements of evolutionary genetics. Bioscience 61, 409–411. (doi:10.1525/bio.2011.61.5.12) 95. Bourgeois YXC etal. 2017 Data from: A novel locus on chromosome 1 underlies the evolution of a melanic plumage polymorphism in a wild songbird. Dryad Digital Repository. (http://dx.doi.org/10. 5061/dryad.1q9g7) 91. Sinervo B, Svensson E. 2002 Correlational selection and the evolution of genomic architecture. Heredity 89, 329–338. (doi:10.1038/sj.hdy. 6800148) pathway. Mol.Med.Rep. 10, 491–495. (doi:10.3892/ mmr.2014.2154) 91. Sinervo B, Svensson E. 2002 Correlational selection and the evolution of genomic architecture. Heredity 89, 329–338. (doi:10.1038/sj.hdy. 6800148) pathway. Mol.Med.Rep. 10, 491–495. (doi:10.3892/ mmr.2014.2154) 14 89. Drabek K, van de Peppel J, Eijken M, van Leeuwen JPTM. 2011 GPM6B regulates osteoblast function and induction of mineralization by controlling cytoskeleton and matrix vesicle release. J.Bone Miner.Res. 26, 2045–2051. (doi:10.1002/jbmr.435) 92. Abzhanov A, Extavour CG, Groover A, Hodges SA, Hoekstra HE, Kramer EM, Monteiro A. 2008 Are we there yet? Tracking the development of new model systems. TrendsGenet. 24, 353–360. (doi:10.1016/ j.tig.2008.04.002) 90. Fjorback AW, Müller HK, Wiborg O. 2009 Membrane glycoprotein M6B interacts with the human serotonin transporter. J.Mol.Neurosci. 37, 191–200. (doi:10.1007/s12031-008-9092-4) 93. Teshima KM, Przeworski M. 2006 Directional positive selection on an allele
W4317868556.txt	null	en		SARS-CoV-2 evolution influences GBP and IFITM sensitivity	Proceedings of the National Academy of Sciences of the United States of America	2,023	cc-by	10,004	RESEARCH ARTICLE \| OPEN ACCESS MICROBIOLOGY SARS-CoV-2 evolution influences GBP and IFITM sensitivity Dejan Mesnera , Ann-Kathrin Reuschla, Matthew V. X. Whelana, Taylor Bronzovicha, Tafhima Haidera,b Paola Bonfantia,c , Greg J. Towersa , and Clare Jollya,1 , Lucy G. Thornea, Roberta Ragazzinia,c , Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. Edited by Stephen Goff, Columbia University Irving Medical Center, New York, NY; received July 22, 2022; accepted December 6, 2022 SARS-CoV-2 spike requires proteolytic processing for viral entry. A polybasic furin-cleavage site (FCS) in spike, and evolution toward an optimized FCS by dominant variants of concern (VOCs), are linked to enhanced infectivity and transmission. Here we show interferon-inducible restriction factors Guanylate-binding proteins (GBP) 2 and 5 interfere with furin-mediated spike cleavage and inhibit the infectivity of early-lineage isolates Wuhan-Hu-1 and VIC. By contrast, VOCs Alpha and Delta escape restriction by GBP2/5 that we map to the spike substitution D614G present in these VOCs. Despite inhibition of spike cleavage, these viruses remained sensitive to plasma membrane IFITM1, but not endosomal IFITM2 and 3, consistent with a preference for TMPRSS2-dependent plasma membrane entry. Strikingly, we find that Omicron is unique among VOCs, being sensitive to restriction factors GBP2/5, and also IFITM1, 2, and 3. Using chimeric spike mutants, we map the Omicron phenotype and show that the S1 domain determines Omicron’s sensitivity to GBP2/5, whereas the S2’ domain determines its sensitivity to endosomal IFITM2/3 and preferential use of TMPRSS2independent entry. We propose that evolution of SARS-CoV-2 for the D614G substitution has allowed for escape from GBP restriction factors, but the selective pressures on Omicron for spike changes that mediate antibody escape, and altered tropism, have come at the expense of increased sensitivity to innate immune restriction factors that target virus entry. Significance GBPs and IFITMs are potent innate immune restriction factors that can inhibit viral infectivity. Specifically, GBPs perturb furin-mediated processing of viral envelope proteins, targeting viruses that rely on proteolytic processing for optimal infectivity. Here, we report that GBP2 and GBP5 inhibit the cleavage of SARS-CoV-2 spike and reduce viral infection. Notably, while the infectivity of early-lineage SARS-CoV-2 isolates is restricted by GBP2/5, VOCs Alpha and Delta have evolved to escape this inhibition. By contrast, Omicron is sensitive to inhibition by GBP2/5, as well as inhibition by endosomal IFITM2 and IFITM3 consistent with Omicron’s use of alternative cell entry pathways. Our data show how VOC evolution under different selective pressures has influenced sensitivity to antiviral restriction factors, and thus, innate immunity. SARS-CoV-2 \| GBP \| IFITM \| restriction \| spike SARS-CoV-2 infects cells by binding of the viral spike (S) protein to the angiotensin-converting enzyme 2 (ACE2) receptor on host cells (1). For fusion to proceed after ACE2 binding, spike must be cleaved by host cell proteases to become activated and fusion-competent. In this step-wise process, spike is preprocessed by furin-like proteases in virus-producing cells at the S1/S2 junction (2, 3), followed by a second cleavage event at the S2’ site mediated by TMPRSS2 protease at the target cell surface (1), releasing the fusion peptide and allowing for viral fusion at the plasma membrane. Alternatively, TMPRSS2independent endocytic uptake can occur in some cell types resulting in spike being cleaved and activated for fusion by endosomal cathepsin proteases (2). The polybasic furin cleavage site (FCS) Arg–Arg–Ala–Arg (or RRAR) motif that is targeted by furin is absent in closely related coronaviruses, including the closest relatives of SARS-CoV-2, bat RaTG13 and pangolin CoV (2, 4, 5). This has led to the notion that presence of an FCS in the SARSCoV-2 ancestor was associated with successful zoonosis and pandemic transmission between humans. In support, the FCS is required for efficient proteolytic cleavage of SARS-CoV-2 spike (2), virus infection of human airway cells (1, 2, 6, 7), cell–cell fusion and syncytia formation (2, 8, 9), and transmission (6, 7). Following the identification of the first SARS-CoV-2 strain circulating in humans (Wuhan-Hu-1), several variants of concern (VOCs) have emerged, each containing a constellation of mutations and further adaptations to host that have been associated with increased transmission. These major previous and current VOCs are designated Alpha (PANGO lineage B.1.1.7), Beta (B.1.351), Gamma (P1), Delta (B.1.617.2), and latterly Omicron (B.1.1.529). Of these, Alpha, Delta, and Omicron have been the most successful globally, each rapidly replacing the previous dominating VOC over time (Omicron > Delta > Alpha). These VOCs have an increasing number of nonsynonymous mutations in spike, which alter entry efficiency and kinetics (10) and enhance immune escape, including innate immunity (11–13). Much focus has been on spike mutations arising from selective pressure for antibody escape; however, SARS-CoV-2 spike continuously adapts in other ways to the human host. For example, dominant VOCs (Alpha, Delta, and Omicron) harbor mutations near and within the FCS, which enhances spike cleavage, indicative of evolution toward an optimized FCS (14, 15). Successful viral replication and transmission requires evasion or antagonism of host defensive processes, notably innate immunity, and is particularly important for zoonotic viruses that must adapt quickly or be suitably preadapted to the new host. Innate immune PNAS 2023 Vol. 120 No. 5 e2212577120 Author contributions: D.M., A.-K.R., M.V.X.W., and C.J. designed research; D.M., A.-K.R., M.V.X.W., T.B., and T.H. performed research; L.G.T., R.R., P.B., and G.J.T. contributed new reagents/analytic tools; D.M., A.-K.R., M.V.X.W., T.B., T.H., and C.J. analyzed data; and D.M., A.K.R., M.V.X.W., T.B., T.H., L.G.T., G.J.T., and C.J. wrote the paper. The authors declare no competing interest. This article is a PNAS Direct Submission. Copyright © 2023 the Author(s). Published by PNAS. This open access article is distributed under Creative Commons Attribution License 4.0 (CC BY). 1 To whom correspondence may be addressed. Email: c.jolly@ucl.ac.uk. This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas. 2212577120/-/DCSupplemental. Published January 24, 2023. https://doi.org/10.1073/pnas.2212577120 1 of 10 Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. activation up-regulates host cell proteins termed restriction factors that target key steps in viral replication to limit and control infection. Guanylate-binding proteins (GBP) are type-I and type-II interferon-stimulated genes (ISGs) and a subfamily of guanosine triphosphatases (GTPases) that can act as intracellular antiviral restriction factors (16). GBP2 and 5 potently inhibit furin-mediated processing of viral envelope proteins, inhibiting infection of HIV-1, Influenza A Virus, Zika, and measles viruses, all of which require furin cleavage for optimal infectivity (16–19). Notably, GBPs are up-regulated in airway epithelial cells and during SARSCoV-2 infection (20, 21). Thus, GBPs comprise a key effector of the antiviral innate immune response that can act to limit infectious virus production during replication. Likewise, the interferon-induced transmembrane (IFITM) protein family also acts broadly to block viral entry, inhibiting viral fusion with cellular membranes, including SARS-CoV-2 (7, 12, 21–24). Here, we investigated the capacity of GBP2 and 5 to inhibit SARS-CoV-2 spike cleavage and virus infectivity, and tested whether evolution of VOCs has led to escape from GBPs. We find differential sensitivity of SARS-CoV-2 spikes to GBP indicative of independent adaptation to host. We also test the consequence of GBP inhibition of furin-cleavage on SARS-CoV-2 sensitivity to another family of entry-targeting restriction factors, IFITMs. Notably, while Alpha and Delta have evolved to escape restriction by GBPs through the D614G substitution in spike, we find that Omicron is uniquely sensitive to inhibition GBP2/5 and also endosomal IFITM2 and 3, consistent with Omicron evolving under different selective pressures, driving increased spike mutations that alter spike activity, cell entry, and tropism. Results GBP2 and 5 Inhibit Wuhan-Hu-1 and Omicron, but not Alpha and Delta Spike-Mediated Infectivity. To determine whether the antiviral restriction factors GBP2 and 5 have activity against SARS-CoV-2, a pseudovirus (PV) assay was used in which SARSCoV-2 spike is incorporated into lentiviral particles (herein termed PV) (SI Appendix, Fig. S1A). This allows direct comparison of how evolution of amino acid changes in spike alone (SI Appendix, Fig. S1B) has influenced GBP sensitivity, without confounding contributions of other SARS-CoV-2 variant proteins on infectivity. 293T cells were cotransfected with plasmids encoding SARS-CoV-2 spike, lentiviral genome, and increasing doses of GBP-expressing plasmid. GBP expression was confirmed by flow cytometry staining for the HA-tag and immunoblotting (Fig. 1A and SI Appendix, Fig S2 A and B). Fig. 1B shows that WuhanHu-1 PV made in the presence of GBP2 or 5 was significantly less infectious (50%) when titrated onto Caco2 target cells, with both GBP2 and 5 inhibiting PV infectivity in a dose-dependent manner (Fig. 1B and SI Appendix, Fig. S2C). We selected naturally permissive intestinal epithelial Caco2 cells as targets for PV infection for their endogenous expression of both ACE2 and TMPRSS2 (25, 26). As expected, the isoprenylation-deficient mutants of GBP2 (GBP2 C588A) and GBP5 (GBP5 C583A), which are mislocalized in the cell and lose antiviral activity against other viruses such as HIV-1 (17, 18) (SI Appendix, Fig. S2 G and H), showed no inhibitory activity against Wuhan-Hu-1 (Fig. 1B). Inhibition of Wuhan-Hu-1 PV infectivity was not due to lack of spike expression on 293T cells since expressing GBPs did not alter plasma membrane levels of spike measured by flow cytometry (SI Appendix, Fig. S3B). Immunofluorescence imaging of spike in 293T cells revealed a similar pattern of diffuse and punctate spike staining in the presence and absence of GBP5, and a small but significant increase in spike colocalization with the ER marker 2 of 10 https://doi.org/10.1073/pnas.2212577120 calnexin in the presence of GBP5 (SI Appendix, Fig. S3 C–E). Strikingly, and in contrast to Wuhan-Hu-1, PV particles containing the Alpha and Delta spikes were completely resistant to GBP2 and 5 restriction and showed no loss of PV infectivity (Fig. 1 C and D and SI Appendix, Fig. S2 D and E). By contrast, PVs containing the Omicron BA.1 (Omicron) spike behaved like Wuhan-Hu-1 and were sensitive to restriction by GBP2 and 5, evidenced by a significant 60% loss of infectivity (Fig. 1E and SI Appendix, Fig. S2F). To measure the contribution of the SARS-CoV-2 furin-cleavage site to PV infectivity, and thus determine the maximum loss of infectivity that be expected if spike processing is completely prevented (comparing to the effects of GBP2/5), we used a Wuhan-Hu-1 spike in which the FCS has been deleted (ΔFCS) to prevent spike cleavage (7). Deletion of the FCS resulted in a significant 75% reduction in PV infectivity, when compared with wild-type Wuhan-Hu-1 spike (Fig. 1F). This was not further reduced by GBP5 expression (Fig. 1F). Consistent with the loss of Wuhan-Hu-1 PV infectivity being mediated by GBP2/5 inhibition of furin cleavage (and not off-target effects), no difference in infectivity was seen when PV were titrated onto Vero.E6 cells (SI Appendix, Fig. S2I) that do not require furin-processing for SARS-CoV-2 infection, as spike is processed by endosomal cathepsins (2, 27). Furthermore, GBP2/5 expression also significantly reduced infectivity mediated by MERS-CoV spike, which contains a furin-cleavage site (28), but not by SARS-CoV-1 spike that lacks a furin-cleavage site (SI Appendix, Fig. S2 J and K). Next, we tested whether GBP2/5 also inhibited the infectivity of live, replication competent SARS-CoV-2 isolates. To do this, Calu-3 cells were pretreated with IFNγ to up-regulate expression of GBP2/5 (Fig. 1G) (16, 18) and then infected with SARSCoV-2 isolates: VIC (an early-lineage, Wuhan-like isolate) and VOCs Alpha, Delta, and Omicron (BA.1). Virus-containing supernatants were harvested from Calu-3 infections and used to infect Caco2 target cells. Input doses were equalized by viral E gene copies and nucleocapsid-positive (N+) Caco2 cells were quantified by flow cytometry. We observed a significant reduction in the infectivity of VIC and Omicron viruses produced by IFNγtreated Calu-3 cells (that was dose dependent) when compared with virus produced by Calu-3 cells not treated with IFNγ (Fig. 1H and SI Appendix, Fig. S4A). By contrast, Alpha and Delta viruses showed no loss of infectivity (Fig. 1H), a finding consistent with our PV assay data. Importantly, depleting GBP2 and GBP5 with siRNA in IFNγ-treated Calu-3 cells (Fig. 1I) completely rescued the infectivity of GBP-sensitive isolates VIC and Omicron (Fig. 1J and SI Appendix, Fig. S4B) confirming that the antiviral effects of IFNγ on reducing virus infectivity in producer cells is mediated by GBP2/5. We note the presence of an additional band detected by the GBP5, but not GBP2 antibody in IFNγ-treated Calu-3 cells, whether this represents a posttranslational modification of GBP5 or a GBP5 splice variant for example remains unclear. Concordant with our Calu-3 results, virus recovered from infections of primary human airway epithelial (HAE) cells, treated with IFNγ to up-regulate GBP2/5, showed a significant loss of infectivity of VIC and Omicron, but not Alpha and Delta viruses, despite all VOCs showing similar infection levels in HAE (Fig. 1 K–M and SI Appendix, Fig. S4C). Collectively these data reveal that VOCs Alpha and Delta have evolved spikes that escape restriction factors GBP2 and 5 that inhibit early-lineage SARS-CoV-2 isolates Wuhan-Hu-1 and VIC, whereas Omicron has not. GBP2 and 5 Inhibit SARS-CoV-2 Spike Cleavage. Given spike is cleaved by furin, and that furin is inhibited by GBPs, we next measured the effects of GBP2 and 5 on spike S1/S2 pnas.org B A Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. F D C G I E H J K L M GBP2 GBP5 Fig. 1. GBP2 and 5 inhibit Wuhan-Hu-1 and Omicron, but not Alpha and Delta spike-mediated infectivity. (A) Expression of HA-tagged GBPs in pseudovirus (PV)-producing 293T cells measured by flow cytometry. (B–E) Infectivity of PVs produced by 293T cells in the presence of increasing amounts of plasmid encoding GBP2, GBP2 C588A, GBP5, or GBP5 C583A measured by luciferase assay (RLU) on Caco2 cells. Shown are percentage infectivity of PV made in the presence of GBPs normalized to empty vector (EV) control (no GBP, set to 100%). Percent infectivity of (B) Wuhan-Hu-1, (C) Alpha, (D) Delta, and (E) Omicron spike PV infection is shown. Shown are the mean ± SEM from three independent experiments. (F) Infectivity of WT and ΔFCS Wuhan-Hu-1 spike PV made in the presence of 120 ng GBP5 or EV control and titrated on Caco2 cells. Shown is percent infectivity (±SEM from three independent experiments) normalized to WT spike EV control. (G, H) Calu-3 cells were treated with indicated doses of IFNγ for 8 h prior to infection with indicated SARS-CoV-2 variants for 36 h. (G) Cell lysates were immunoblotted for GBP2, GBP5, and tubulin. Quantification shows relative expression of GBP2/5 over tubulin and normalized to untreated control. (H) Equal doses (E copies/ cells) of SARS-CoV-2 virus produced in cells from (G) were used to infect Caco2 cells, and infection levels were determined at 24 hpi by intracellular staining for nucleocapsid (N) protein. Percentage positive cells are shown (% N+ cells). Bars show the mean and individual values from two independent experiments. (I, J) Calu-3 cells were pretreated with combined GBP2 and GBP5 siRNA (KD) or nontargeting control (Ctrl, 120 and 32 h preinfection) and indicated doses of IFNγ (8 h preinfection) before infection with indicated SARS-CoV-2 variants for 36 h. (I) Cell lysate immunoblots for GBP2, GBP5, and tubulin. Quantification shows relative expression of GBP2/5 over tubulin and normalized to Ctrl siRNA IFNγ-untreated control. (J) Equal doses (E copies/cells) of SARS-CoV-2 virus produced in cells from (I) were used to infect Caco2 cells for 24 h and infection was quantified by N protein staining (% N+ cells). Bars show the mean and replicate values from one experiment. (K–M) Primary human airway epithelial cells (HAEs) were treated with indicated doses of IFNγ for 12 h prior to infection with equal doses (1,500 E copies/cell) of indicated SARS-CoV-2 variants for 72 h. (K) GBP2 and GBP5 gene expression (fold increase over 0ng/mL IFNγ) at 72 hpi. (L) Virus release from infected HAE was collected in apical washes at 72 hpi, equal volumes were used to infect Caco2 cells for 24 h, and infection was quantified by N protein staining (% N+ cells). (M) Intracellular replication in HAE measured by E copies at 72 hpi. Two-way ANOVA (B–F, H, J–M) with Dunnett’s posttest was used. (B–E) Statistical significance for GBPs (120 ng) compared EV control is indicated. FCS, furin cleavage site. ns, not significant; P < 0.05; P < 0.01; P < 0.001; *P < 0.0001. cleavage. Visualizing uncleaved (S) and cleaved spikes (S2) by immunoblotting of purified PV particles (Fig. 2A and SI Appendix, Fig. S5A) and cell lysates (Fig. 2B and SI Appendix, Fig. S5B) revealed clear differences in the processing of Wuhan-Hu-1 spike in the presence of GBP2 and 5, compared with either no GBPs, or the inactive mutants GBP2 C588A and GBP5 C583A (SI Appendix, Fig. S5 A and B). Quantifying this across all spike variants from four independent experiments showed that GBP5 PNAS 2023 Vol. 120 No. 5 e2212577120 significantly reduced spike cleavage (lower S2/S ratio) and incorporation (S+S2/p24) into Wuhan-Hu-1, Alpha, Delta, and Omicron PV particles (Fig. 2A). A significant spike cleavage defect was also apparent when immunoblotting cell lysates (Fig. 2B), consistent with GBP perturbation of intracellular spike processing during PV production. Having observed that the S2 spike subunit migrated at a lower molecular weight in the presence of GBP, we sought to test whether this may reflect a change in https://doi.org/10.1073/pnas.2212577120 3 of 10 Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. A C B D Fig. 2. GBP2 and 5 inhibit SARS-CoV-2 spike cleavage. (A, B) Spike PVs were produced in 293T cells in the presence of 80 ng GBP plasmid or empty vector (EV) control. PVs and producer cell lysates were immunoblotted for spike, lentiviral Gag (p24 and p55), GBP (HA-tag) and tubulin. (A) Immunoblot of VOC spike PV produced in the presence of GBP5 or EV. A representative immunoblot is shown. Graphs show quantification pooled from four independent experiments, measuring the proportion of cleaved spike in PV (S2/S) (Top) and total spike incorporation (S+S2/p24) (Bottom). Mean and individual values are shown. (B) Producer 293T cell lysate from (A). (C) Calu-3 cells were pretreated with GBP2 and GBP5 siRNA (KD) or nontargeting control (Ctrl) and indicated doses of IFNγ before infection with indicated SARS-CoV-2 variants for 36 h as described in Fig. 1. Equal doses (E copies) of clarified viruses were sucrose-purified and immunoblotted for spike (S) and nucleocapsid (N) protein. Quantification shows proportion of spike cleavage (S2/S) and cleaved spike incorporation into virions (S2/N). (D) Indicated SARS-CoV-2 variants were produced in Caco2 cells, and equal doses of virus (E copies) were immunoblotted for S and N protein. Quantification shows proportion of cleaved spike (S2/S) and cleaved spike incorporation into virions (S2/N). Two independent Omicron BA.1 isolates (A and B) are shown. Two-way ANOVA with Dunnett’s posttest (A, B) was used. ns, not significant; P < 0.05; P < 0.01; P < 0.001; **P < 0.0001. spike glycosylation. A similar alteration in HIV-1 envelope (Env) glycosylation has been reported when furin-cleavage is inhibited by GBPs, presumably due to differences in the intracellular trafficking of uncleaved versus cleaved Env influencing protein glycosylation (17, 18). SI Appendix, Fig. S5C shows that PNGase treating 293T cell lysates abolished the spike band shift, such that the S2 subunit migrated at the same molecular weight +/− GBP. Taken together, these data indicate that GBP expression inhibits furin cleavage and also influences spike N-linked glycosylation (29, 30) resulting in reduced particle infectivity. Quantifying spike cleavage (S2/S) in virus produced by IFNγtreated Calu-3cells was more challenging due to low levels of uncleaved spike detected in VOC particles (Fig. 2C) and here immunoblotting cell lysates did not show an obvious cleavage defect (SI Appendix, Fig. S5D), unlike PV assays. However, quantifying spike cleavage by immunoblotting purified VIC viral particles (from the experiment shown in Fig. 1 I and J) revealed that 4 of 10 https://doi.org/10.1073/pnas.2212577120 RNAi depletion of GBP2/5 in IFNγ-treated Calu-3 increased the amount of cleaved spike product in virions (S2/S) (Fig. 2C), although this effect was not as striking as what was seen in PV assays. The observation that GBP2/5 knockdown also increased VIC spike cleavage in Calu-3 cells that were not treated with IFNγ (Fig. 2C, 0ng/mL IFNγ condition) is explained by the presence of basal levels of inhibitory GBP2/5 expression in these cells (Fig. 1I and SI Appendix, Fig. S5D) that was also depleted using RNAi. Alpha, Delta, and Omicron spikes contain optimizing mutations in the FCS that have been reported to enhance S1/S2 cleavage (14, 15) (P681H, P681R and P681H/N679K/H655Y respectively). Consistent with this, immunoblotting of purified viral isolates (produced in the absence of GBPs) showed a spike cleavage hierarchy of Omicron > Delta > Alpha > IC19/VIC (Fig. 2D); however, Omicron’s optimized FCS does not protect from GBP restriction (Fig. 1E). These data, as well as our pnas.org observation that both Alpha and Delta are resistant to GBPmediated restriction of viral infectivity despite GBPs inhibiting spike cleavage, suggests that evolution of VOCs Alpha and Delta for escape from GBPs cannot simply be attributed to the presence of optimized FCS. This implicates other spike adaptations, beyond the FCS, in allowing Alpha and Delta to overcome GBP restriction of infectivity. the FCS mutations in Alpha (H681) and Delta (R681) spike back to Wuhan-Hu-1 like sequences (creating Alpha H681P and Delta R681P, respectively) had no effect on the restriction phenotype, with these mutants remaining fully resistant to GBP-mediated inhibition of infectivity (Fig. 3C). Importantly, increasing the amount of spike incorporated into Wuhan-Hu-1 PV using a construct in which the 19 residues of the spike cytoplasmic tail are deleted (ΔCT) boosted spike incorporation and particle infectivity as described previously (34, 35), but did not rescue from GBP restriction (SI Appendix, Fig. S6 A–C). Similarly, titrating increasing amounts of full-length Wuhan-Hu-1 spike plasmid into 293T cells during PV production in the presence of GBP also failed to rescue from GBP restriction (SI Appendix, Fig. S6G). Extending this, reducing the amount of Alpha and Delta spikes transfected into 293T cells failed to sensitize these PV to GBP restriction (SI Appendix, Fig. S6 J and K). Taken together, these data identify the D614G substitutions as the evolutionary change allowing Alpha and Delta to evade GBP-mediated restriction. Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. D614G Substitution in Alpha and Delta Spikes Confers Resistance to GBP Restriction. The D614G mutation in spike manifested as an early host adaptation in Wuhan-like isolates that enhances infectivity and is now ubiquitous in circulating SARS-CoV-2 variants (31–33). Alpha and Delta contain the infectivity-enhancing D614G mutation in spike (SI Appendix, Fig. S1B), whereas Wuhan-Hu-1 and VIC do not. Having observed that the most infectious PV (Alpha and Delta) were resistant to GBP restriction, whereas the less infectious Wuhan-Hu-1 PV was sensitive (Fig. 3A), we sought to test whether the D614G substitution may explain the resistance of Alpha and Delta VOCs to GBP2/5 inhibition of infectivity. Strikingly, introducing the D614G mutation into Wuhan-Hu-1 spike (Wuhan D614G) completely rescued PV from GBP inhibition (Fig. 3C), although spike cleavage was still inhibited (Fig. 2 A and B). Consistent with previous reports, D614G also rendered Wuhan-Hu-1 PV more infectious on a per particle basis, but it remained less infectious than Alpha and Delta (Fig. 3B) (31–33). Importantly, reverting Alpha and Delta spikes back to the ancestral Wuhan-Hu-1 by introducing a G614D substitution resulted in these PV becoming sensitive to GBP restriction, evidenced by a significant 50% reduction in infectivity (Fig. 3C). Consistent with the optimized FCS not explaining Alpha and Delta resistance to GBPs, reverting B A as Alpha and Delta but remained sensitive to GBPs, behaving like Wuhan-Hu-1 (Fig. 3A). However, in addition to D614G, Omicron contains a large number of antibody escape mutations in its spike compared with other SARS-CoV-2 isolates (SI Appendix, Fig. S1B) (36–39), suggesting that this constellation of changes may have negatively influenced spike activity and its capacity to escape GBP restriction. Concordantly, Omicron PV infectivity was significantly lower than other variants tested (Fig. 3A), consistent with reports that Omicron displays reduced infectivity in cell lines commonly used to study SARS-CoV-2 entry (36–39) and C E D The S1 Domain of Spike Confers Sensitivity of Omicron to GBP Restriction. Omicron also contains the same D614G substitution F Fig. 3. Mapping spike determinants mediating the differential sensitivity of SARS-CoV-2 isolates to GBP5 inhibition. (A) Comparison of particle infectivity (RLU/ RT) of spike PVs on Caco2 cells in the absence of GBPs. (B and C) Wuhan D614G, Alpha G614D, and H681P, and Delta G614D and R681P spike mutants were generated and tested for PV infectivity and GBP5 sensitivity. (B) Comparison of spike mutants PV particle infectivity (RLU/RT) on Caco2 cells in the absence of GBP expression. (C) Percentage infectivity of spike mutants PV made in the presence of GBP5 normalized to empty vector (EV) control for each spike (set to 100%). (D–F) Chimeric spikes were generated with S2’ domains swapped between Delta and Omicron spikes to produce S1-Omicron-S2’-Delta and S1-Delta-S2’-Omicron chimeras (schematic). Indicated spike PV were produced in the presence of GBP5 or EV control and titrated on Caco2 cells. (D) Raw infectivity (RLU/RT) values in the absence of GBP5 expression (EV control) and (E) percent infectivity normalized to EV control are shown. (F) S1-Omicron-S2’-Delta and S1-Delta-S2’-Omicron spike PVs made in the presence of GBP5 or EV control were immunoblotted for spike and p24. Bars show mean ± SEM from three independent experiments. One-way ANOVA (A, D) or two-way ANOVA (B, C, E) with Dunnett’s posttest were used. ns, not significant; P < 0.05; P < 0.01; P < 0.001; **P < 0.0001. PNAS 2023 Vol. 120 No. 5 e2212577120 https://doi.org/10.1073/pnas.2212577120 5 of 10 Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. in animal models (38). These spike changes are present both in the receptor-binding S1 subunit, and in the S2’ domain that harbors the viral fusion peptide (SI Appendix, Fig. S1B). Therefore, we generated chimeric Omicron–Delta spikes to determine whether S1 or S2’ determines Omicron sensitivity to GBPs (Fig. 3D). We used Delta because its spike has the highest fusogenicity (40) and is resistant to GBPs. Fig. 3 D–F shows that replacing the S1 domain of Omicron with that of Delta (creating S1-Delta-S2’Omicron spike) conferred complete resistance to GBP inhibition of infectivity. By contrast the Delta-S2’ domain did not rescue, and this spike (S1-Omicron-S2’-Delta) remained GBP sensitive, phenocopying the native Omicron spike. Similar to both the fulllength Delta and Omicron spikes, these chimeric spikes remained sensitive to GBP inhibition of furin-cleavage (Fig. 3F). Finally, increasing Omicron spike incorporation into PV using the ΔCT spike mutant, or titrating increasing amounts of full-length Omicron spike plasmid into 293T cells during PV production failed to rescue Omicron from GBP restriction (SI Appendix, Fig. S6 D–F and H), similar to Wuhan-Hu-1. These data suggest that restriction of Wuhan-Hu-1 and Omicron by GBPs cannot be simply explained by reduced spike incorporation into virions. We conclude that the S1 domain contains the determinant for Omicron’s sensitivity to GBP-mediated inhibition of infection, and that for Omicron the changes within the S1 domain of spike have compromised the ability of the D614G mutation to confer resistance to GBPs. GBPs Do Not Sensitize to Restriction by Endosomal IFITM2 and 3. It has been reported that mutating the FCS in Alpha to interfere with furin-cleavage can modulate sensitivity to inhibition by endosomal IFITM2 (12, 22). Having shown that GBPs interfere with SARS-CoV-2 spike cleavage, we sought to test whether this may sensitize SARS-CoV-2 to IFITM-mediated inhibition. IFITMs are another family of spike-targeting interferon-inducible restriction factors that can inhibit infection of a range of viruses, including SARS-CoV-2 (7, 12, 21–24, 41), by perturbing viral fusion with host cell membranes and thus inhibiting infection of target cells. IFITMs are differentially localized in cells with IFITM1 being found mostly at the plasma membrane and IFITM2/3 predominantly endosomal. To explore the effects of GBPs on IFITM sensitivity, we used Caco2 cells stably overexpressing either IFITM1, 2, or 3 and confirmed the expected IFITM localization by immunofluorescence microscopy (Fig. 4A). We first established the IFITM restriction phenotype in these cells. Consistent with TMPRSS2-dependent plasma membrane fusion, IFITM1 was found to potently and significantly inhibit infection of early-lineage viruses (Wuhan-Hu-1 and VIC) and VOCs Alpha and Delta in both PV (Fig. 4 B–D) and live virus infections (Fig. 4 F–H). By contrast, IFITM2 and IFITM3 did not inhibit but rather enhanced infection in agreement with previous studies (12, 22, 24, 41, 42). Notably, Omicron was unique among SARS-CoV-2 viruses in being sensitive to inhibition by endosomal IFITM2 and 3 in Caco2 cells (Fig. 4 E and I). This is consistent with Omicron favoring TMPRSS2-independent endosomal entry pathways, evidenced by increased sensitivity to the cathepsin inhibitor E64d and reduced sensitivity to the TMPRSS2 inhibitor Camostat (SI Appendix, Fig. S7), in agreement with others (36– 39, 43). Infecting IFITM-expressing Caco2 cells with VIC or Omicron (sensitive to GBP restriction) produced from IFNγtreated Calu-3 (from Fig. 1) showed no difference in the sensitivity of these viruses to IFITMs when compared with virus produced from untreated Calu-3 cells (Fig. 4 J and K). Similar results were obtained using Wuhan-Hu-1 and Omicron PV made in the presence of overexpressed GBP5, where we found no change in 6 of 10 https://doi.org/10.1073/pnas.2212577120 IFITM restriction and no increase in inhibition by endosomal IFITM2 or 3 (SI Appendix, Fig. S8). Together, these data show that inhibiting furin cleavage by GBPs does not make a virus that is resistant to IFITMs become sensitive to restriction. Finally, to map the domains in Omicron that dictate its sensitivity to endosomal IFITM2 and 3, we used our chimeric Omicron–Delta spike mutants (Fig. 3D). Fig. 4 L–O show that replacing the S2’ domain of Omicron with that of Delta rendered this PV resistant to IFITM2 and 3 inhibition and sensitive to IFITM1, thus behaving like Delta. This S1-Omicron-S2’-Delta spike PV also became sensitive to the TMPRSS2 inhibitor Camostat and resistant to the endosomal E64d inhibitor, thus behaving like Delta (Fig. 4P), a finding supported by recent studies (37, 39). These results identify the S2’ domain of spike as the determinant for IFITM sensitivity, by contrast to GBP-sensitivity that is mediated by the S1 domain, and reveal that Omicron’s unique sensitivity to IFITM2 and 3 (when compared with other SARS-CoV-2 variants) is dictated by its altered entry route. Discussion Innate immunity is a potent first-line host cell defense against viruses, up-regulating a group of ISGs which can act directly as restriction factors, targeting key steps in viral replication and collectively inducing an antiviral state. SARS-CoV-2 triggers innate immune sensing and induces an interferon-response (11, 25, 44, 45), upregulating canonical ISGs including GBP2 and 5 in primary human airway epithelial cells (20). Evolution of mutations outside of spike allow for SARS-CoV-2 evasion/antagonism of innate immune sensing (11); however, spike itself is a target of the innate immune response. Here we report that the interferon-inducible restriction factors GBP2 and 5 interfere with SARS-CoV-2 spike cleavage, and significantly inhibit infection by the early-lineage SARS-CoV-2 strains Wuhan-Hu-1 and VIC, but that previously dominant VOCs Alpha and Delta have evolved to evade GBP-mediated inhibition of infectivity. Notably, Alpha and Delta have both evolved an optimized FCS by acquiring the P681H and P681R substitutions, respectively, resulting in enhanced spike processing (12, 14, 15); however, we find that evolution of these VOCs to escape GBP restriction is not due to this optimized FCS. Instead, we find that it is the presence of the now ubiquitous D614G substitution in spike that mediates Alpha and Delta resistance to GBPs. Specifically, we show that reverting Alpha and Delta to 614D led these PV becoming sensitive to GBP-mediated inhibition of infection, behaving like Wuhan-Hu-1 and VIC. By contrast, introducing the 614G mutation into Wuhan-Hu-1 spike rescued from GBP inhibition and increased spike infectivity. The D614G variant has been reported to enhance SARS-CoV-2 spike infectivity (31, 33, 46, 47) by stabilizing the S1/S2 subunit noncovalent association and altering spike conformation, shifting it toward a more open, fusion competent state, without increasing ACE2-binding affinity (33, 46, 48). Therefore, by allowing spike to adopt the more open conformation that is on pathway for fusion, the 614G change essentially makes spike better primed for function. Consistent with this, we saw a correlation between the presence of 614G in spike and increased PV infectivity. By contrast, we found that early lineage isolates that do not contain D614G cannot overcome the restriction by GBP. The fact that we could not rescue Wuhan-Hu-1 or Omicron from GBP restriction by increasing spike incorporation argues against the effects of GBPs being mediated simply by reducing overall spike content in virions. Instead, our data suggest that GBPs restrict SARS-CoV-2 infectivity by impairing furin-cleavage of spike and altering spike pnas.org Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. A B C F G J D H E I K L M N O P Fig. 4. Omicron is uniquely restricted by IFITM2/3, which maps to the spike S2’ domain. (A) Localization of HA-tagged IFITMs 1-3 in Caco2 cells was assessed by immunofluorescence imaging. HA-tag IFITM (green), CD63 (red), and nuclei (blue). (Scale bar is 50 μm.) (B–E) Infection of IFITM-expressing Caco2 cells with (B) Wuhan-Hu-1, (C) Alpha, (D) Delta, and (E) Omicron spike PV. Percent infectivity normalized to WT Caco2 cells (no IFITM over-expression) are shown. (F–I) IFITMexpressing Caco2 cells were infected with SARS-CoV-2. Shown is percent N+ cells at 24 hpi of (F) VIC, (G) Alpha, (H) Delta, and (I) Omicron isolates. Mean ± SEM from three independent experiments are shown (J, K) Calu-3 cells were treated with indicated doses of IFNγ for 8 h to induce expression of GBP2/5 and infected with indicated SARS-CoV-2 variants as shown in Fig. 1. At 36 hpi, virus-containing supernatant of (J) VIC and (K) Omicron isolates were harvested, and equal doses of virus (E copies/cell) from the supernatants were used to infect WT or IFITM expressing Caco2 cells for 24 h. Infection was quantified as the percentage N+ cells. Mean and individual replicates from two independent experiments are shown. (L–O) IFITM-expressing Caco2 cells were infected with (L) Omicron, (M) S1-Omicron-S2’-Delta, (N) S1-Delta-S2’-Omicron, and (O) Delta spike PV. Percent infectivity normalized to WT Caco2 cells (no IFITM overexpression) are shown. (P) Indicated spike PV were used to infect Caco2 cells pretreated with 25 μM Camostat or E64d. Shown is percent infection normalized to DMSO control. One-way ANOVA (B–I, L–O) or two-way ANOVA (J, K, P) with Dunnett’s posttest was used. ns, not significant; P < 0.05; P < 0.01; P < 0.001; ***P < 0.0001. PNAS 2023 Vol. 120 No. 5 e2212577120 https://doi.org/10.1073/pnas.2212577120 7 of 10 Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. processing (glycosylation), leading to reduced spike function that consequently reduces particle infectivity and entry into target cells. However, evolution for the D614G substitution has allowed Alpha and Delta to overcome the loss of spike function caused by GBPs by improving spike activity and particle infectivity, allowing these VOCs to evade restriction. It is therefore tempting to speculate that the selection for and subsequent dominance of D614G in SARS-CoV-2 isolates was due, in part, to adaptation to host in order to escape from interferon-induced innate immunity, as well as inherent effects of D614G on improving spike function to increase transmissibility. It has been shown that interfering with spike cleavage by mutating the FCS in Alpha sensitizes virus to inhibition by another spike targeting restriction factor, namely endosomal IFITM2 (12, 22). However, despite GBPs perturbing spike cleavage and reducing Wuhan-Hu-1 infectivity, we did not find that GBPs sensitized SARS-CoV-2 to inhibition by endosomal IFITM2 or 3. We explain this by GBPs reducing, but not completely preventing, spike S1/S2 cleavage, thus allowing Wuhan-Hu-1 to retain a preference for plasma membrane entry, and therefore retaining sensitivity to plasma membrane localized IFITM1. These results are consistent with reports that other cellular proteases can mediate some spike processing at the polybasic cleavage site in the absence of furin (8). It is notable however, that although GBPs did not sensitize to IFITMs, combining GBP5 in producer cells with IFITM1 in target cells led to an almost complete inhibition of Wuhan-Hu-1 infection when compared with infection in the absence of these restriction factors. Thus, during an innate immune response to viral infection where multiple ISGs including GBP2/5 and IFITM1/2/3 are induced, we might expect to see stronger inhibitory effects. We show that Omicron is unique among VOCs we tested in being sensitive to inhibition by GBP2/5, despite containing the same D614G substitution as Alpha and Delta. Using chimeric Omicron–Delta spike mutants, we mapped the determinant of Omicron’s sensitivity to GBPs to the S1 domain. We propose that the constellation of mutations present in Omicron spike compared with other VOCs has compromised the ability of D614G to overcome the inhibitory effects of GBPs. Consistent with this notion, Omicron was significantly less infectious than other SARS-CoV-2 isolates we tested, indicative of the multiple mutations impacting spike function. Given the significant number of substitutions in Omicron, this is not surprising. Further work will be needed to define precisely which combination of changes in Omicron S1 mediates its sensitivity to GBP2/5 and compromise the effects of D614G in driving escape. Omicron was also unique in being sensitive to inhibition by IFITM1, 2, and 3, unlike other isolates we tested that were not inhibited by IFITM2 and 3. This is explained by Omicron having evolved toward an altered entry route of TMPRSS2-independence and endosomal-dependent fusion (36, 37, 39), thus exposing it to endosomal IFITMs. The domains that mediate Omicron’s sensitivity to IFITM2 and 3 are distinct from those that mediate GBP-sensitivity, and we show that it is the S2’ domain of spike that dictates its entry phenotype and sensitivity to endosomal IFITM2 and 3. Specifically, replacing the S2’ domain in Omicron spike with that of Delta (S1-OmicronS2’-Delta), converted the phenotype rendering this PV “Deltalike” and resistant to IFITM2/3 by switching Omicron PV entry back to TMPRSS2-dependence and away from endosomal entry. While this manuscript was in preparation, Omicron subvariants BA.4 and BA.5 began dominating infections globally. We also confirmed that like BA.1, BA.2 and BA.4/5 spikes are sensitive to both GBP and IFITM restriction and retain the same preference for TMPRSS2-independent entry (SI Appendix, Fig. S9). The 8 of 10 https://doi.org/10.1073/pnas.2212577120 in vivo interplay between Omicron and innate immunity, restriction factors, and tropism remains ill-defined. It is possible that the expression of different innate immune restriction factors varies across cells/tissues, such that Omicron has evolved into a niche that allows it to avoid or tolerate GBPs and IFITMs. Moreover, IFITM2 and 3 are suggested to act as cofactors for SARS-CoV-2 in some cases (41, 42), and we and others also see enhancement of early-lineage isolates and Alpha and Delta infection by IFITM2/3 (Fig. 4) (12, 22, 24, 41, 42), therefore it cannot be excluded that Omicron may similarly exploit IFITMs in some settings. It is clear that SARS-CoV-2 needs to balance efficient cell entry with evasion of compartmentalized restriction factors, and it will be intriguing to see how Omicron does this to successfully infect target cells in vivo. Omicron has evidently evolved to do things differently, but effectively, and exploit a different cellular niche, one to which it is clearly well adapted. SARS-CoV-2 VOCs have evolved separately from early lineage strains and not from each other. It is therefore not surprising that these VOCs have explored different evolutionary solutions to the problems they faced, and that the selective pressures encountered by each VOC are not identical. For example, Alpha and Delta evolved prior to significant levels of adaptive immunity in the population. By contrast Omicron, the first real antibody escape variant, evolved at a time of much greater population level humoral immunity, therefore Omicron has been exposed to different selective pressures, requiring different evolutionary solutions. Our data showing differential sensitivity of Alpha/Delta vs Omicron to GBP and IFITM restriction, as well as others showing significant Omicron antibody escape (36, 39), are consistent with this, reinforcing the notion that Omicron has taken a different evolutionary path to preceding VOCs. We propose a scenario in which evolution of Omicron spike for neutralizing antibody escape has influenced the ability to evade innate immunity. The critical balance between viral evasion of innate and adaptive immunity has precedent. This is borne out of studies of HIV-1 evolution in a host, where HIV-1 isolates from early in infection (so called transmitter/ founder viruses) are completely resistant to IFITM restriction, but overtime, the selective pressure from adaptive immunity, and the resulting neutralizing antibody escape mutations in HIV-1 Env, leads to viral isolates having increased sensitivity to IFITMs and interferons (49). We propose that similar processes have occurred during SARS-CoV-2 evolution to host, in which the need to escape from neutralizing antibody became the dominant selective pressure on Omicron, resulting in a compensatory, but tolerable, increase in sensitivity to innate immunity, while also impacting on spike activity and cell tropism. We predict that this interplay between evasion of innate and adaptive immunity, and the consequences for transmission and tropism, will be features of future SARS-CoV-2 evolution, and emergence of new VOCs, and that linking this evolution to phenotype will become important aspects for understanding and predicting SARS-CoV-2 biology, and ultimately pathogenesis. Materials and Methods Cells. HEK293T/17 cells (abbreviated herein as 293T cells) were obtained from American Type Culture Collection (ATCC, CRL-11268). Caco2 cells were a gift from Dalan Bailey (Pirbright Institute) and originally obtained from ATCC. Calu-3 cells were purchased from AddexBio (C0016001). Vero.E6 cells were obtained from the National Institute for Biological Standards and Control. HeLa-TZM-bl cells (expressing luciferase and beta-galactosidase under the control of HIV-1 LTR) were obtained from the Centre for AIDS Reagents (CFAR). All cell lines were grown in Dulbecco’s modified Eagle’s medium (DMEM, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (Labtech) and 1% Pen Strep pnas.org Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. (penicillin-streptomycin, Thermo Fisher Scientific), and maintained in humidified 5% CO2 incubators at 37 °C. Cells were passaged every 2 to 4 d when they reached 80 to 90% confluency. Caco2 cells were transduced with IFITM1/2/3 lentivectors as described previously (50) and selected with 10 μg/mL puromycin (Merck) to produce stable cell lines expressing individual HA-tagged IFITM proteins. IFITM expression was confirmed by flow cytometry. Primary normal (healthy) bronchial epithelial (NHBE-A) cells were cultured for 5 to 7 passages and differentiated at an air-liquid interface as previously described (11). After 21 to 24 d of differentiation, cells were used in infection experiments. Plasmids. SARS-CoV-2 spike expression vectors were originally synthesized by Genewiz and subcloned into pcDNA3.1+ vector. All spike sequences are fulllength, unless otherwise stated. Wuhan-Hu-1 WT (51), Wuhan-Hu-1 D614G, and Alpha (52) spike expression vectors were gifts from Laura McCoy (UCL). Delta and Omicron BA.1 spike expression vectors were a gift from Katie Doores (King’s College London). Wuhan-Hu-1 WT, Wuhan-Hu-1 ΔFCS, Omicron BA.1, BA.2, and BA.4/5 spike ΔCT expression vectors were a gift from Wendy Barclay (Imperial College London) (7, 37). SARS-CoV-1 spike and MERS-CoV spike expression vectors were a gift from Joe Grove (Centre for Virus Research, Glasgow). Plasmid encoding HIV-1 Env pSVIII_JRFL was a gift from Laura McCoy (UCL). Plasmid encoding full-length HIV-1 pNL4.3 was donated by Dr M Martin and obtained from CFAR. Lentiviral backbone packaging plasmid (expressing HIV Gag, Pol, Tat and Rev) p8.91 and the reporter plasmid-encoding luciferase gene pCSLW were a gift from Greg Towers (UCL). GBP expression vectors encoding HA-tagged GBPs (GBP2 WT, GBP2 C588A, GBP5 WT, GBP5 C583A) and BFP reporter expressed from an IRES (17, 18) were a gift from Daniel Sauter (University Hospital Tubingen). Mutagenesis. Mutagenesis was performed as described in SI Appendix, Methods. SARS-CoV-2 Viruses. SARS-CoV-2 isolates VIC (BetaCoV/Australia/VIC01/2020, lineage B), IC19 (hCoV-19/England/IC19/2020, lineage B.1.13) and Alpha (hCoV19/England/204690005/2020, lineage B.1.1.7) have been described previously (11). SARS-CoV-2 Delta (lineage B.1.617.2) and Omicron (lineage B.1.1.529/BA.1) isolates were a kind gift from Wendy Barclay (Imperial College London, UK) (14, 37). Viruses were propagated by infecting Caco2 cells at MOI 0.01 TCID50 per cell, in DMEM culture medium supplemented with 1% FBS and 1% penicillin/ streptomycin, at 37 °C. Virus was collected at 72 hpi and clarified by centrifugation at 2,100 × g for 15 min at 4 °C to remove any cellular debris. Virus stocks were aliquoted and stored at −80 °C. Virus stocks were quantified by extracting RNA from 100 µL supernatant with 1 µg carrier RNA using Qiagen RNeasy clean-up RNA protocol, before measuring viral E RNA copies per mL by RT-qPCR as described previously (25). Live Virus Infections. Caco2 (1 × 10 5 cells/well) were seeded in 24-well plates one day before infection. Cells were infected with 1000 E RNA copies per cell in 200μL culture medium. After 2h incubation at 37 °C, cells were carefully washed with PBS to remove excess virus, and a fresh culture medium was added. For inhibition assays, cells were pretreated with inhibitors at the indicated concentrations for 2 h prior to infections and maintained throughout the experiment. At the indicated time points, cells were collected for analysis. For Calu-3 cell infections, 2 × 105 cells/well were seeded into 12-well plates and grown until confluent. Where indicated, cells were pretreated with indicated concentrations of recombinant human IFNγ (Peprotech) for 8 h before cells were infected with 1000 E RNA copies SARS-CoV-2 per cell in 400 μL culture medium. The inoculum was thoroughly washed off with PBS after 2 h, and a fresh culture medium added. At 36 hpi, cells were harvested for protein lysates and flow cytometry. Culture supernatants from infected cells were clarified by centrifugation at 2,100 × g for 15 min at 4 °C and viral E RNA copies measured by RT-qPCR. To determine infectivity of these viral supernatants, Caco2 cells were pretreated with 5 μM Ruxolitinib (Bio-Techne) for 1 h to inhibit JAK signaling (from carryover interferon) and then infected with 1000 E copies/cell of virus as described above and harvested at 24 hpi for flow cytometry analysis. Ruxolitinib was maintained throughout. Primary human airway epithelial cells (HAEs) were infected by adding 1500 E copies/cell to the apical side for 3 h at 37 °C. Supernatant was then removed and cells were gently washed twice with PBS. All liquid was removed from the apical side, and the basal medium was replaced with fresh Pneumacult ALI medium for the duration of the experiment. For IFNγ-stimulation, 10 ng/mL recombinant IFNγ (Peprotech) was added basally 12 h before HAE infection and maintained throughout the PNAS 2023 Vol. 120 No. 5 e2212577120 experiment. Intracellular replication was determined at 72 hpi, and viral release was measured at 24 and 72 h by extracting viral RNA from apical PBS washes as described previously (11). 5 GBP2/5 RNAi Depletion. Calu-3 cells (2 × 10 cells/well) were seeded into 12-well plates and transfected with 20 pmol siRNA SMART pool against GBP2 (L-011867-00-0005) and GBP5 (L-018178-00-0005) or nontargeting control (D-001810-10-05) (Dharmacon) using Lipofectamine RNAiMAX Transfection Reagent (Invitrogen). Cells were treated with siRNA transfection reagents at both 120 h and 36 h before infection. Cells were then treated with indicated concentrations of recombinant human IFNγ (Peprotech) for 8 h before infection as described above. GBP2/5 depletion was confirmed by immunoblotting (described below). Pseudovirus Production. Spike pseudoviruses (PVs) were made by cotransfec- tion of spike, p8.91 and pCSLW plasmids as described previously (53). To determine GBP inhibition, PV plasmids were transfected together with GBP vector or empty vector (EV) control expressing only BFP reporter. Briefly, 5 × 104 293T cells were seeded onto 24-well plates for 24 h and then transfected with 260 ng p8.91, 260 ng pCSLW, 40 ng spike, and 20 to 120 ng GBP vectors or EV control using Fugene6 (Promega). For larger scale production cells were seeded in 6-well plates with cell numbers and transfection reagents scaled up fivefold. For HIV-1 Env PV, 293T cells were seeded as described above and transfected with 240 ng p8.91, 240 ng pCSLW,120 ng pSVIII-JRFL Env, and indicated doses of GBP or empty vector control. PV supernatants were collected at 48 and 72 h posttransfection and purified through 0.45 μm centrifuge tube filters (Corning) or 0.45 μm syringe-filters (Starlab) and used within 24 h without freeze–thawing. The amount of PV in the supernatant was determined by measuring the supernatant RT activity using SYBR-green-based product enhanced reverse transcription assay (SG-PERT) by qPCR, performed as described previously (54). Pseudovirus Infection. Target cells were seeded into white 96-well plates 24 h before infection (Caco2 and Vero.E6 cells seeded at 1.5 × 104 cells/well, and HeLaTZMbl cells seeded at 1 × 104 cells/well). Cells were infected with increasing doses (2 to 15 mU RT/well) of PV supernatant (to confirm linear increase in infection at increased PV doses) and incubated at 37 °C for 48 h without changing the medium. Luciferase expression (RLU) was measured at 48 h postinfection using BrightGlo substrate (Promega) according to the manufacturer’s instruction on the Glomax luminometer (Promega). For inhibitor studies, cells were pretreated before infections for 2 h with Camostat mesylate (Apexbio, 0.2 to 100 μM) or E64d (Focus Biomolecules, 0.2 to 25 μM). To obtain infectivity (RLU/RT) values, RLU values were normalized to input supernatant RT activity (measured by SG-PERT assay, described above). GBP Inhibition Assay. Spike PV, HIV-1 Env PV, or HIV-1 virus were made in the presence of increasing doses of GBP2/5 or their mutants or empty vector control as described above. Supernatant RT activity (RT units) was measured by SG–PERT assay to determine PV or virus content in supernatants. As indicated, different cell lines were infected with equal doses of PV/virus supernatants for 48 h and luciferase expression (RLU) was measured as describe above. Infectivity of the supernatant was determined as a ratio of RLU to RT units (RLU/RT) and normalized to empty vector control (no GBP), set at 100% for each PV or virus. IFITM Inhibition Assay. Caco2 cells, WT or stably expressing IFITM1/2/3 (described above), were infected with indicated spike PVs, and luciferase expression was measured 48 h postinfection as described above. Infectivity (RLU/RT) was normalized to Caco2 WT control (100%). Alternatively, spike PVs were made in the presence of GBP5 or EV control and used to infect Caco2 WT- and IFITMexpressing cells. Infection of Caco2-IFITM1/2/3 cells with live SARS-CoV-2 virus was done as described above. HIV-1 Infection. HIV-1 infections were performed as described in SI Appendix, Methods. Quantitative PCR. Quantitative PCR was performed as described in SI Appendix, Methods. Immunoblotting. Immunoblotting was performed as described in SI Appendix, Methods. https://doi.org/10.1073/pnas.2212577120 9 of 10 Flow Cytometry. Flow cytometry was performed as described in SI Appendix, Methods. Immunofluorescence Microscopy. Immunofluorescence microscopy was per- formed as described in SI Appendix, Methods. Statistical Analysis. Statistical significance was calculated using Prism 9 (GraphPad Prism) using indicated statistical tests and significance was assumed when P < 0.05. Data, Materials, and Software Availability. All study data are included in the article and/or SI Appendix. This work was funded by Wellcome Investigator Award 108079 followed by 223065 to C.J. G.J.T. is funded by Wellcome Investigator Award 220863. C.J. and G.J.T. were also funded by MRC/UKRI G2P-UK National Virology consortium (MR/W005611/1) and the UCL COVID19 fund. M.V.X.W. is part-supported by the NIHR Biomedical Research Centre at UCLH and IDEA Bio-Medical. R.R. is supported by a Marie Skłodowska-Curie Individual Fellowships no. 896014. P.B. received funding from the European Downloaded from https://www.pnas.org by UNIVERSITY COLLEGE LONDON LIBRARY-PERIODICALS DEPT on February 1, 2023 from IP address 193.60.238.99. ACKNOWLEDGMENTS. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 10 of 10 M. Hoffmann et al., SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell 181, 271–280.e8 (2020). M. Hoffmann, H. Kleine-Weber, S. Pöhlmann, A multibasic cleavage site in the spike protein of SARS-CoV-2 is essential for infection of human lung cells. Mol. Cell 78, 779–784.e5 (2020). A. C. Walls et al., Structure, function, and antigenicity of the SARS-CoV-2 spike glycoprotein. Cell 181, 281–292.e6 (2020). S. Temmam et al., Bat coronaviruses related to SARS-CoV-2 and infectious for human cells. Nature 604, 330–336 (2022), 10.1038/s41586-022-04532-4. M. F. Boni et al., Evolutionary origins of the SARS-CoV-2 sarbecovirus lineage responsible for the COVID-19 pandemic. Nat. Microbiol. 5, 1408–1417 (2020). B. A. Johnson et al., Loss of furin cleavage site attenuates SARS-CoV-2 pathogenesis. Nature 591, 293–299 (2021). T. P. Peacock et al., The furin cleavage site in the SARS-CoV-2 spike protein is required for transmission in ferrets. Nat. Microbiol. 6, 899–909 (2021). G. Papa et al., Furin cleavage of SARS-CoV-2 spike promotes but is not essential for infection and cell-cell fusion. PLoS Pathog. 17, e1009246 (2021). M. M. Rajah et al., SARS-CoV-2 alpha, beta, and delta variants display enhanced spike-mediated syncytia formation. EMBO J. 40, e108944 (2021). C. B. Jackson, M. Farzan, B. Chen, H. Choe, Mechanisms of SARS-CoV-2 entry into cells. Nat. Rev. Mol. Cell Biol. 23, 3–20 (2022). L. G. Thorne et al., Evolution of enhanced innate immune evasion by SARS-CoV-2. Nature 602, 487–495 (2022). M. J. Lista et al., The P681H mutation in the spike glycoprotein of the alpha variant of SARS-CoV-2 escapes IFITM restriction and is necessary for type I interferon resistance. J. Virol. 96, e0125022 (2022), 10.1128/jvi.01250-22. K. Guo et al., Interferon resistance of emerging SARS-CoV-2 variants. Proc. Natl. Acad. Sci. U.S.A. 119, e2203760119 (2022). T. P. Peacock, The SARS-CoV-2 variants associated with infections in India, B.1.617, show enhanced spike cleavage by furin. bioXriv [Preprint] (2021), https://doi. org/10.1101/2021.05.28.446163. (Accessed 22 November 2022). A. Escalera et al., Mutations in SARS-CoV-2 variants of concern link to increased spike cleavage and virus transmission. Cell Host Microbe 30, 373–387.e7 (2022). S. Martens, J. Howard, The interferon-inducible GTPases. Annu. Rev. Cell Dev. Biol. 22, 559–589 (2006). E. Braun et al., Guanylate-binding proteins 2 and 5 exert broad antiviral activity by inhibiting furinmediated processing of viral envelope proteins. Cell Rep. 27, 2092–2104.e10 (2019). C. Krapp et al., Guanylate binding protein (GBP) 5 is an interferon-inducible inhibitor of HIV-1 infectivity. Cell Host Microbe 19, 504–514 (2016). D. Hotter, D. Sauter, F. Kirchhoff, Guanylate binding protein 5: Impairing virion infectivity by targeting retroviral envelope glycoproteins. Small GTPases 8, 31–37 (2017). C. G. K. Ziegler et al., SARS-CoV-2 receptor ACE2 is an interferon-stimulated gene in human airway epithelial cells and is detected in specific cell subsets across tissues. Cell 181, 1016–1035.e19 (2020). L. Martin-Sancho et al., Functional landscape of SARS-CoV-2 cellular restriction. Mol. Cell 81, 2656–2668.e8 (2021). H. Winstone et al., The polybasic cleavage site in SARS-CoV-2 spike modulates viral sensitivity to type I interferon and IFITM2. J. Virol. 95, e02422-20 (2021). G. Shi et al., Opposing activities of IFITM proteins in SARS-CoV-2 infection. EMBO J 40, e106501. (2021). J. Buchrieser et al., Syncytia formation by SARS-CoV-2-infected cells. EMBO J. 39, e106267 (2020). L. G. Thorne et al., SARS-CoV-2 sensing by RIG-I and MDA5 links epithelial infection to macrophage inflammation. EMBO J. 40, e107826 (2021). J. Koch et al., TMPRSS2 expression dictates the entry route used by SARS-CoV-2 to infect host cells. EMBO J. 40, e107821 (2021). M. Hoffmann et al., SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell 181, 271–280.e8 (2020). https://doi.org/10.1073/pnas.2212577120 Research Council (ERC-Stg no. 639429), the Rosetrees Trust (M362-F1; M553), the NIHR GOSH BRC and the CF Trust (SRC006; SRC020). We thank the G2P-UK National Virology Consortium; Wendy Barclay and Tom Peacock at Imperial College London, UK; Gavin Screaton Oxford University, UK; Dalan Bailey at Pirbright Institute UK; Katie Doores at King’s College London UK, as well as Daniel Sauter at University Hospital Tubingen, Germany, Laura McCoy at University College London UK, J.E. Voss and D. Huang at Scripps Research Institute USA and Joe Grove at the Centre for Virus Research, Glasgow, UK, for provision of reagents and variant isolates. We also acknowledge members of the G2P-UK Consortium, as well as the Jolly lab and Towers lab for helpful discussions. For the purpose of Open Access, we have applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. Author affiliations: aDivision of Infection and Immunity, University College London, WC1E 6BT London, UK; bCentre for Genomics and Child Health, Blizard Institute, Queen Mary University of London, E1 2AT London, UK; and cEpithelial Stem Cell Biology and Regenerative Medicine Laboratory, The Francis Crick Institute, London NW1 1AT, UK 28. J. K. Millet, G. R. Whittaker, Host cell entry of Middle East respiratory syndrome coronavirus after two-step, furin-mediated activation of the spike protein. Proc. Natl. Acad. Sci. U.S.A. 111, 15214–15219 (2014). 29. Q. Yang et al., Inhibition of SARS-CoV-2 viral entry upon blocking N- and O-glycan elaboration. Elife 9, e61552 (2020). 30. Q. Yang et al., Role for N-glycans and calnexin-calreticulin chaperones in SARS-CoV-2 Spike maturation and viral infectivity. Sci. Adv. 8, eabq8678 (2022). 31. B. Korber et al., Tracking changes in SARS-CoV-2 spike: Evidence that D614G increases infectivity of the COVID-19 virus. Cell 182, 812–827.e19 (2020). 32. J. A. Plante et al., Spike mutation D614G alters SARS-CoV-2 fitness. Nature 592, 116–121 (2021). 33. L. Yurkovetskiy et al., Structural and functional analysis of the D614G SARS-CoV-2 spike protein variant. Cell 183, 739–751.e738 (2020). 34. F. Schmidt et al., Measuring SARS-CoV-2 neutralizing antibody activity using pseudotyped and chimeric viruses. J. Exp. Med. 217, e20201181 (2020). 35. X. Ou et al., Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune cross-reactivity with SARS-CoV. Nat. Commun. 11, 1620 (2020). 36. B. Meng et al., Altered TMPRSS2 usage by SARS-CoV-2 omicron impacts infectivity and fusogenicity. Nature 603, 706–714 (2022). 37. T. P. Peacock, The altered entry pathway and antigenic distance of the SARS-CoV-2 omicron variant map to separate domains of spike protein. bioXriv [Preprint] (2022), https://doi. org/10.1101/2021.12.31.474653. (Accessed 22 November 2022). 38. H. Shuai et al., Attenuated replication and pathogenicity of SARS-CoV-2 B.1.1.529 omicron. Nature 603, 693–699 (2022). 39. B. J. Willett et al., SARS-CoV-2 omicron is an immune escape variant with an altered cell entry pathway. Nat. Microbiol. 7, 1161–1179 (2022), 10.1038/s41564-022-01143-7. 40. P. Mlcochova et al., SARS-CoV-2 B.1.617.2 delta variant replication and immune evasion. Nature 599, 114–119 (2021). 41. R. Nchioua et al., SARS-CoV-2 variants of concern hijack IFITM2 for efficient replication in human lung cells. J. Virol. 96, e0059422 (2022). 42. C. Prelli Bozzo et al., IFITM proteins promote SARS-CoV-2 infection and are targets for virus inhibition in vitro. Nat. Commun. 12, 4584 (2021). 43. X. Du et al., Omicron adopts a different strategy from delta and other variants to adapt to host. Signal Transduct. Target. Ther. 7, 45 (2022). 44. X. Lei et al., Activation and evasion of type I interferon responses by SARS-CoV-2. Nat. Commun. 11, 3810 (2020). 45. M. S. Diamond, T.-D. Kanneganti, Innate immunity: The first line of defense against SARS-CoV-2. Nat. Immunol. 23, 165–176 (2022). 46. S. M. Gobeil et al., D614G mutation alters SARS-CoV-2 spike conformation and enhances protease cleavage at the S1/S2 junction. Cell Rep. 34, 108630 (2021). 47. E. Volz et al., Evaluating the effects of SARS-CoV-2 spike mutation D614G on transmissibility and pathogenicity. Cell 184, 64–75.e11 (2021). 48. J. Zhang et al., Structural impact on SARS-CoV-2 spike protein by D614G substitution. Science 372, 525–530 (2021). 49. T. L. Foster et al., Resistance of transmitted founder HIV-1 to IFITM-mediated restriction. Cell Host Microbe 20, 429–442 (2016). 50. C. Petrillo et al., Cyclosporine H overcomes innate immune restrictions to improve lentiviral transduction and gene editing in human hematopoietic stem cells. Cell Stem Cell 23, 820–832.e9 (2018). 51. J. Seow et al., Longitudinal observation and decline of neutralizing antibody responses in the three months following SARS-CoV-2 infection in humans. Nat. Microbiol. 5, 1598–1607 (2020). 52. C. Rees-Spear et al., The effect of spike mutations on SARS-CoV-2 neutralization. Cell Rep. 34, 108890 (2021). 53. S. J. Dicken, Characterisation of B.1.1.7 and pangolin coronavirus spike provides insights on the evolutionary trajectory of SARS-CoV-2. bioXriv [Preprint] (2021), https://doi. org/10.1101/2021.03.22.436468. (Accessed 22 November 2022). 54. M. Pizzato et al., A one-step SYBR Green I-based product-enhanced reverse transcriptase assay for the quantitation of retroviruses in cell culture supernatants. J. Virol. Methods 156, 1–7 (2009). pnas.org
https://openalex.org/W2131898777	https://europepmc.org/articles/pmc4864488?pdf=render	English	null	Divergent modulation of Rho‐kinase and Ca<sup>2+</sup> influx pathways by Src family kinases and focal adhesion kinase in airway smooth muscle	British journal of pharmacology	2,015	cc-by	12,975	Divergent modulation of Rho- kinase and Ca2+ inﬂux pathways by Src family kinases and focal adhesion kinase in airway smooth muscle Received 12 May 2015 Revised 2 August 2015 Accepted 19 August 2015 Received 12 May 2015 Revised 2 August 2015 Accepted 19 August 2015 Yasin Shaifta, Nneka Irechukwu, Jesus Prieto-Lloret, Charles E MacKay, Keisha A Marchon, Jeremy P T Ward and Greg A Knock Yasin Shaifta, Nneka Irechukwu, Jesus Prieto-Lloret, Charles E MacKay, Keisha A Marchon, Jeremy P T Ward and Greg A Knock Division of Asthma, Allergy and Lung Biology, Faculty of Life Sciences and Medicine, King's College London, London, UK London, London, UK KEY RESULTS Contractile responses in intact bronchioles were inhibited by antagonists of SrcFK, FAK and Rho-kinase, while after α-toxin permeabilization, they were sensitive to inhibition of SrcFK and Rho-kinase, but not FAK. CCh and BK increased phosphorylation of MYPT-1 and MLC20 and auto-phosphorylation of SrcFK and FAK. MYPT-1 phosphorylation was sensitive to inhibition of Rho-kinase and SrcFK, but not FAK. Contraction induced by SR Ca2+ depletion and equivalent [Ca2+]i responses in hASMC were sensitive to inhibition of both SrcFK and FAK, while depolarization-induced contraction was sensitive to FAK inhibition only. SrcFK auto-phosphorylation was partially FAK-dependent, while FAK auto-phosphorylation was SrcFK-independent. EXPERIMENTAL APPROACH EXPERIMENTAL APPROACH Contraction was recorded in intact or α-toxin permeabilized rat bronchioles. Phosphorylation of SrcFK, FAK, myosin light-chain- 20 (MLC20) and myosin phosphatase targeting subunit-1 (MYPT-1) was evaluated in cultured human ASM cells (hASMC). [Ca2+]i was evaluated in Fura-2 loaded hASMC. Responses to carbachol (CCh) and bradykinin (BK) and the contribution of SrcFK and FAK to these responses were determined. CONCLUSIONS AND IMPLICATIONS SrcFK mediates Ca2+-sensitization in ASM, while SrcFK and FAK together and individually inﬂuence multiple Ca2+ inﬂux pathways. Tyrosine phosphorylation is therefore a key upstream signalling event in ASM contraction and may be a viable target for modulating ASM tone in respiratory disease. BACKGROUND AND PURPOSE The importance of tyrosine kinases in airway smooth muscle (ASM) contraction is not fully understood. The aim of this study was to investigate the role of Src-family kinases (SrcFK) and focal adhesion kinase (FAK) in GPCR-mediated ASM contraction and associated signalling events. RESEARCH PAPER Correspondence Dr Greg Knock, 1.20 Henriette Raphael House, Guy’s Campus, King’s College London, London, SE1 1UL, UK. E-mail: greg.knock@kcl.ac.uk Divergent modulation of Rho- kinase and Ca2+ inﬂux pathways by Src family kinases and focal adhesion kinase in airway smooth muscle BJP British Journal of Pharmacology BJP British Journal of Pharmacology BJP British Journal of Pharmacology BJP British Journal of Pharmacology DOI:10.1111/bph.13313 www.brjpharmacol.org BJP British Journal of Pharmacology BJP RESEARCH PAPER Introduction was ﬁrst suggested by the relaxant effect of non-selective tyro- sine kinase inhibitors on rat isolated bronchioles (Chopra et al., 1997). Subsequently, selective inhibition of SrcFK and FAK was shown to depress GPCR-induced contraction in human, rodent or canine upper airways (Tang and Gunst, 2001; Katsumoto et al., 2013). FAK was linked to elevated [Ca2+]i in response to various stimuli in trachea, but the relative inﬂuence of the kinase on VOCE, ROCE or SOCE or on Rho-kinase was not determined (Tang et al., 1999; Tang and Gunst, 2001). SrcFKs have been identiﬁed as upstream mediators of Rho-kinase in vascular smooth muscle (Nakao et al., 2002; Knock et al., 2008), but neither this relationship nor the inﬂuence of SrcFK on GPCR [Ca2+]i responses has yet been examined in ASM. To our knowledge, only one previous study has examined the involvement of SrcFK or FAK speciﬁcally in the contraction of intralobar bronchioles, and this was limited to the role of SrcFK in mediating sensitization of rat bronchioles to muscarinic agonists (Sakai et al., 2010). Airway smooth muscle (ASM) tone is subject to regulation by cholinergic, catecholamine and NANC neurotransmitters as well as local inﬂammatory mediators. In healthy airways, muscle tone is normally low, providing a low resistance path for airﬂow. However, contraction may be enhanced in re- sponse to chemical irritants or allergens, particularly in lower respiratory tract bronchioles (Gilbert and Auchincloss, 1989; Pinelli et al., 2009). In asthma, airway resistance is increased, partly due to increased basal tone and hypersensitivity to constrictor stimuli in these bronchioles (Doeing and Solway, 2013; Meurs et al., 2008). Smooth muscle contractile force depends on the degree of myosin light-chain-20 (MLC20) phosphorylation, which is in turn determined by the balance between Ca2+-dependent ac- tivation of myosin light-chain kinase (MLCK) and Ca2+-inde- pendent inhibition of myosin light-chain phosphatase (MLCP), as well as the formation and recruitment of myoﬁla- ments (Gunst et al., 2003; Somlyo and Somlyo, 2003). Increases in [Ca2+]i result from Ca2+ release from the sarco- endoplasmic reticulum (SR) and a combination of Ca2+ entry through receptor-operated, store-operated and voltage- operated Ca2+ channels (ROCE, SOCE and VOCE, respec- tively). Inhibition of MLCP occurs via phosphorylation of myosin phosphatase targeting subunit-1 (MYPT-1), primarily by Rho-kinase (Feng et al., 1999), resulting in a further increase in MLC20 phosphorylation and contraction without the need for a further increase in [Ca2+]i (Somlyo and Somlyo, 2003). Methods Rats and tension measurement by wire myography Tables of Links LIGANDS Bradykinin (BK) GTP Carbachol (CCh) Nifedipine Cyclopiazonic acid (CPA) Y27632 These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2013/14 (a,bAlexander et al., 2013a,b). Introduction Although it is likely that bronchoconstrictors act via a combination of the above pathways, the precise mechanisms through which Ca2+ inﬂux and Rho-kinase activity are medi- ated by GPCRs are not fully understood. In this study, we hypothesized that SrcFK and FAK mediate GPCR-induced ASM contraction via multiple signalling path- ways and examined their inﬂuence on Rho-kinase-dependent MLCP inhibition/Ca2+-sensitization and on SOCE/ROCE and VOCE Ca2+ entry pathways, in intra-lobar bronchioles of rat and cultured human ASM cells (hASMC). We found that SrcFK, most likely c-Src itself, modulate Rho-kinase dependent Ca2+-sensitization, but FAK does not, and that the two tyrosine kinases differentially regulate SOCE/ROCE and VOCE. We also suggest the existence of two subpopulations of GPCR- activated SrcFK, one being FAK-dependent and the other FAK-independent. Src family kinases (SrcFK) and focal adhesion kinase (FAK) are widely expressed non-receptor tyrosine kinases (TKs) im- portant in many aspects of cellular function, being activated in response to various stimuli including growth factors, GPCRs, reactive oxygen species and adhesion. SrcFK and FAK are often described as being mutually dependent or recip- rocally activated, especially when associated with integrin engagement and/or growth factor receptor activation (Owen et al., 1999; Ishigaki et al., 2011). An effect of TKs on ASM tone Abbreviations ASM, airway smooth muscle; hASMC, cultured human airway smooth muscle cells; KPSS, PSS with 80 mM equimolar substitution of Na+ for K+; MLC20, myosin light-chain 20 KDa subunit; MLCP, myosin light-chain phosphatase; MYPT-1, myosin phosphatase targeting subunit-1; ROCE, receptor-operated Ca2+ entry; SOCE, store-operated Ca2+ entry; VOCE, voltage-operated Ca2+ entry British Journal of Pharmacology (2015) 172 5265–5280 5265 © 2015 The Authors. British Journal of Pharmacology published by John Wiley & Sons Ltd on behalf of The British Pharmacological Society. on behalf of The British Pharmacological Society. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. g y der the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medi perly cited. g y This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. BJP Tables of Links TARGETS Ion channelsa Enzymesb Store-operated Ca2+ channels FAK Voltage-gated Ca2+ channels MLCK PYK2 Src family kinases Rho-kinase (ROCK) These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2013/14 (a,bAlexander et al., 2013a,b). LIGANDS Bradykinin (BK) GTP Carbachol (CCh) Nifedipine Cyclopiazonic acid (CPA) Y27632 BJP S a ta et a . 5266 British Journal of Pharmacology (2015) 172 5265–5280 Protein lysate preparation and western blot Protein lysate preparation and western blot Cultured hASMCs were treated in serum-free DMEM at 37°C. Preliminary studies showed that phosphorylation responses, although sustained for at least 5 min, peaked at ~30 s, so all subsequent acute treatments were for 30 s. Cells were imme- diately washed twice with ice-cold PBS, followed immediately by application of cell lysis buffer (NEB) containing 1% phos- phatase inhibitor cocktails 2 and 3 and 1% protease inhibitor cocktail (all Sigma). Cells were scraped into a tube and agitated before being placed on ice. Rat trachealis muscle was dissected free of adjoining cartilage, and epithelium was removed by scraping. Acute treatments were conducted in PSS/5% CO2, at 37°C, before the tissue was snap frozen in liquid nitrogen, pulverized and lysed in cell lysis buffer. All lysates were centrifuged at 9.2x g, and the supernatants were stored at 80°C. Tyrosine kinases in airway smooth muscle deemed to be as humane as possible. All results involving an- imals are reported in accordance with the ARRIVE guidelines for reporting experiments involving animals (McGrath et al., 2010). A total of 98 rats were used. Male Wistar rats (~250 g) had free access to food and water and were maintained on a 12:12 h light/dark schedule. The rats were killed by an i.p. injection of sodium pentobarbital and the lungs and trachea were immediately removed. First or second-order intralobar bronchioles (~2 mm length) were dissected free of surround- ing parenchyma and mounted on a wire myograph (DMT. dk), bathed in PSS (in mM: 118 NaCl; 24 NaHCO3; 1 MgSO4; 4 KCl; 5.56 glucose; 0.435 NaH2PO4; 1.8 CaCl2, pH 7.4), gassed with 95% air, 5% CO2 at 37°C. Bronchioles were incrementally stretched and alternately exposed to PSS con- taining 80 mM [K+] (equimolar substitution for Na+, KPSS) until the point on the length tension curve at which muscle length was optimum for active tension development was achieved, as described previously (Moir et al., 2003). Viability for contraction experiments was conﬁrmed by a response of at least 3 mN to the last challenge with KPSS. Bronchiole internal diameter after stretch was typically in the range 300–800 μm. anti-smooth muscle α-actin, anti-desmin and anti-calponin, with Alexa Fluor®488 labelled secondary antibody (Lifetechnologies) and with TRITC-labelled phalloidin to conﬁrm the presence of stress ﬁbres in resting cells (Supporting Information Fig. S1). Cells were used for experi- ments at passages 4–9, grown to conﬂuence and serum starved for 7 days in DMEM plus supplements, and the addition of 1% BSA, 5 μg ml-1 transferrin, 1 μM insulin and 100 μM ascorbate. Rats and tension measurement by wire myography y g p y All animal care and experimental procedures complied with UK legislation under the Animals (Scientiﬁc Procedures) Act 1986 Amendment Regulations (SI 2012/3039) and were 5266 British Journal of Pharmacology (2015) 172 5265–5280 British Journal of Pharmacology (2015) 172 5265–5280 5267 siRNA design and transfection Two siRNAs against human SRC (GenBank accession no. NM_005417) were designed as described previously (Reynolds et al., 2004; Ui-Tei et al., 2004). The 19 nucleotide target sequences (SRC-siRNA1: position 1489–1507 and SRC-siRNA2: position 1684–1702) were synthesized into 64–65 mer oligonu- cleotides with BamHI/HindIII overhangs (Sigma Aldrich) and cloned into the expression vector pSilencer 3.0-H1, containing pmaxGFP (Ambion Inc.). All clones were puriﬁed using an EndoFree Plasmid Maxi Kit (Qiagen Ltd) and sequenced (Geneservice Ltd). hASMC were transfected using the Basic Nucleofector® Kit and nucleofector device (Amaxa Biosystems). After 72 h, the transfection efﬁciency was >90%, conﬁrmed by ﬂuorescence microscopy. Speciﬁc examination of the Rho-kinase dependent Ca2+- sensitization component of contraction was achieved by permeabilizing myograph-mounted bronchioles with α-haemolysin (α-toxin). PSS was ﬁrst exchanged for relaxing solution (pCa = 10, in mM: 200 PIPES; 100 Mg(Ms)2; 1000 KMS; 100 K2EGTA; 5 Na2ATP; 10 Na2creatine phosphate, pH 7.1), gassed with air at 26°C. α-toxin (60 μg ml-1) was then applied in relaxing solution with pCa raised to 6.7 for 30 min, permeabilization being conﬁrmed by the development of active tension. pCa was adjusted via proportionate substitu- tion of K2EGTA for CaEGTA, with 100 CaEGTA, 0 K2EGTA being equivalent to pCa4.5. Contractile responses to bronchoconstrictors were conducted at pCa 6.5 (~300 nM [Ca2+]), which induced a contraction equivalent to 10-20% of that achieved by pCa 4.5. GTP 1 μM and 10 μM cyclopiazonic acid (CPA) were included to support G-protein signalling and to prevent the inﬂuence of SR Ca2+ release on contraction respectively. Y Shaifta et al. BJP BJP component (PD2-2 and Max-2). Statistical analysis of data was by Student’s paired or un-paired t-test (two groups of data, single factor), one-way ANOVA (more than two groups of data, single factor) or two-way ANOVA (more than two groups of data, two factors), with Holm–Sidak post tests where appropriate, and as indicated in ﬁgure or table legends, using SigmaPlot 10. Differences were consi- dered signiﬁcant if P < 0.05. component (PD2-2 and Max-2). Statistical analysis of data was by Student’s paired or un-paired t-test (two groups of data, single factor), one-way ANOVA (more than two groups of data, single factor) or two-way ANOVA (more than two groups of data, two factors), with Holm–Sidak post tests where appropriate, and as indicated in ﬁgure or table legends, using SigmaPlot 10. Differences were consi- dered signiﬁcant if P < 0.05. Femto chemi-luminescent Substrate (Thermo scientiﬁc). Membranes were then stripped in Restore western blot strip- ping buffer (Thermo Scientiﬁc), re-blocked and re-incubated with corresponding ‘total’ antibody and appropriate secondary antibodies, as above. ‘Total’ proteins were visualized with either ECL plus or ECL prime (Amersham, GE healthcare). Images were captured and quantiﬁed using the ChemiDoc XRS+ gel-imaging system (Biorad). An estimate of the propor- tion of target protein that was phosphorylated was calculated as a ratio of ‘phospho’ over ‘total’ signal for each protein band from each gel, and the effects of acute treatments on these ratios was expressed as a percentage of control (untreated samples run on the same gel). [Ca2+]i measurement C l d hAS C [ ]i Cultured hASMCs were grown on glass cover-slips until 70% conﬂuent, followed by 7 days of serum starvation. Cells were loaded with 1 μM Fura PE-3/AM in HBSS (containing in mM: 0.49 MgCl, 0.41 MgSO4, 4 KCl, 0.44 KH2PO4, 4.2 NaHCO3, 120 NaCl, 0.34 Na2HPO4, 20 HEPES and 2 CaCl) at room temperature for 40 min. Coverslips were mounted on an up- right microscope and cells perfused with HBSS, containing test reagents as required. Changes in [Ca2+]i were measured as a ratio of 340 nm over 380 nm emission intensities with a ×20 oil immersion UV objective and a microspectro- ﬂuorimeter (CairnResearch Ltd., U.K.). For each coverslip, ratios obtained in zero [Ca2+]o and the absence of drug were taken as background ﬂuorescence (auto-ﬂuorescence + residual basal [Ca2+]i) and subtracted from all subsequent measurements. Materials and reagents g Antibodies were obtained from cell signalling (anti-phospho- Src (tyr416); anti-Src; anti-phospho-FAK (Y397); anti-phospho- FAK (Y576/577); anti-FAK; anti-phospho-MLC (S19); anti-MYPT1; anti-MLC), Millipore (anti-phospho-MYPT1 (T696)), Sigma (anti-rabbit IgG; anti-mouse IgG). Kinase inhibitors were obtai- ned from Sigma ((1R, 4r)-4((R)-1-aminoethyl)-N-(pyridine-4-yl) cyclohexane carboxamide (Y27632); 6-[4-(3-methanesulfonyl- benzylamino)-5-triﬂuoromethyl-pyrimidin-2-ylamino]-3,4-dihy- dro-1H-quinolin-2-one (PF-573228); N-[2-[[[2-[(2,3-dihydro-2- oxo-1H-indol-5-yl)amino]-5-(triﬂuoromethyl)-4-pyrimidinyl] amino]methyl]phenyl]-N-methyl-methanesulfonamide hydrate (PF-431396) or Calbiochem: (4-amino-5-(4-chlorophenyl)-7- (dimethylethyl) pyrazolo[3,4-d]pyrimidine (PP2); 4-amino- 7-phenylpyrazol[3,4-d]pyrimidine (PP3). Cell culture and western blot materials were obtained from Cell Signalling, Invitrogen, GE Healthcare or Thermo Scientiﬁc. Nifedipine, YM58483, and cyclopiazonic acid and α-haemolysin were from Sigma. Bradykinin caused a prominent transient contraction and a smaller sustained component at each dose (Figure 1E). The concentration-dependence of these responses appeared biphasic. Contraction at all concentrations of BK was signiﬁcantly inhibited by PP2 (Figure 1F), with maximum amplitudes of both high and low afﬁnity components being reduced (Max-1 = 1.66 ± 0.5% vs. control 12.1 ± 4.1%; Max- 2 = 6.81 ± 2.0% vs. control 24.4 ± 6.0%, P < 0.05, n = 11), while PD2-1 and PD2-2 were both unchanged (PD2-1 = 6.7 ± 0.21 vs. control 7.08 ± 0.11; PD2-2 = 4.85 ± 0.27 vs. con- trol 4.67 ± 0.22, n = 11) (see Supporting Information Fig. S4 5268 British Journal of Pharmacology (2015) 172 5265–5280 Results GPCR-mediated contraction of rat bronchioles is dependent on SrcFK, Rho-kinase and FAK. We examined the contractile responses to CCh and bradykinin (BK) in rat bronchioles, whereby the bronchoconstrictors were applied cumulatively at 5 min intervals. Two concentration-response curves were performed in each bronchiole (0.01–100μM), the ﬁrst acting as a control and the second after pre-incubation with either the SrcFK inhibitor PP2 (30 μM), the Rho-kinase inhibitor Y27632 (10 μM), the FAK inhibitor PF-573228 (10 μM) or no inhibitor (control). In addition, to account for possible off-target effects of PP2 and PF-573228, key contractile responses were also repeated with PP3 (30 μM), the negative control for PP2 and a dual FAK/PYK2 inhibitor, PF-431396 (10 μM). CCh caused a sustained contraction at each dose (Figure 1A). The maximum response to CCh was signiﬁcantly reduced by PP2 (P < 0.01, paired t-test, n = 8), Y27632 (P < 0.01, paired t-test, n = 6) and PF-573228 (P < 0.05, paired t-test, n = 8), and the PD2 was signiﬁcantly increased by PP2 (5.55 ± 0.09 vs. control 5.8 ± 0.14, P < 0.05, paired t-test, n = 8), Y27632 (5.4 ± 0.07 vs. control 5.82 ± 0.07, P < 0.01, paired t-test, n = 6) and PF-573228 (5.21 ± 0.08 vs. control 5.69 ± 0.07, P < 0.001, paired t-test, n = 8) (Figure 1A–D). PP3 had no signiﬁcant effect on either PD2 (5.6 ± 0.05 vs. control 5.72 ± 0.08, n = 7) or maximum contraction (144 ± 5% vs. control 149 ± 3%, n = 7). Conversely, PF-431396 had similar effects as those of PF-573228, causing a similar increase in PD2 (5.20 ± 0.05 vs. control 5.90 ± 0.07, P < 0.001, paired t-test, n = 7) and a similar reduction in maximum contraction (139 ± 6.3% vs. control 176 ± 9.3%, P < 0.001, paired t-test, n = 7) (Supporting Information Figs. S2A and S3A). In time- matched control responses, repeated in the absence of inhibitor, the maximum contraction of the second response was slightly increased (ﬁrst repeat: 203 ± 22% vs. second repeat 228 ± 25%, P < 0.01, paired t-test, n=10), but there was no signiﬁcant change in PD2 (ﬁrst repeat: 5.67 ± 0.11 vs. second repeat 5.63 ± 0.09, n = 10). Human tissue and cell culture Samples were boiled in NuPAGE LDS Sample Buffer (Invitrogen) at 95°C for 5 min before being loaded onto 4–12% NuPAGE Bis-Tris gels (Invitrogen) for SDS-PAGE. Sam- ple protein content was determined using the bicinchoninic acid assay, calibrated against BSA protein standards, to enable loading of ~20 μg of protein per lane. Gels were run at 180 V for 1 h using an Xcell SureLock Mini-Cell (Invitrogen) and MOPS running buffer (Invitrogen). Protein was transferred to a nitrocellulose membrane (Amersham) in 25 mM Tris, 192 mM glycine and 20% methanol, at 35 V for 1 h. Donations of human tissue were obtained following written informed consent and with the approval of the South East London Research Ethics Committee, REC reference number 10/H0804/66. All clinical procedures conformed to the standards set by the latest Declaration of Helsinki. hASMC were obtained from healthy volunteers (n = 11; 7 women, 4 males; age range 22–53 years; life-long absence of respiratory symptoms; lung functions within normal limits) by deep endobronchial biopsy. ASM bundles were bathed in DMEM containing 10% FBS, L-glutamine (2 mM), sodium pyruvate (1 mM), non-essential amino acids and amphotericin B (2 μg ml-1), and subjected to enzymatic digestion in nomi- nally Ca2+-free HEPES buffer containing: 5.56 mM glucose, 2 mg ml-1 collagenase Type XI, 1 mg ml-1 papaine, 1 mg ml-1 trypsin inhibitor and 1 mM DTT, for 30 min at 37°C. Cells were then dispersed into culture ﬂasks containing DMEM (plus supplements) and incubated at 37°C, pH 7.4. Smooth muscle phenotype was conﬁrmed by positive staining with Membranes were blocked with 5% skimmed milk in Tris buffered saline (TBS) for 1 h at room temperature, followed by incubation with speciﬁc anti-phospho-protein primary antibody (typically 1:1000 dilution) in TBS with 1% skimmed milk and 0.1% Tween-20 (TBS-T), overnight at 4°C. Following washes in TBS-T, HRP-conjugated secondary antibody (typically 1:5000 dilution) was applied for 1 h at room temperature, followed by a ﬁnal wash in TBS-T. ‘Phospho’ proteins were visualized with Super-Signal West British Journal of Pharmacology (2015) 172 5265–5280 5267 Data analysis and statistics y All values are expressed as mean ± SEM. Non-linear regres- sion curve ﬁtting was performed with SigmaPlot 10. Carba- chol (CCh) concentration-response curves were ﬁtted using the Hill equation for the calculation of PD2 (-LogM EC50) and maximum response (Max). Bradykinin concentration responses were biphasic, and were best ﬁtted using a two- site saturation model, for the characterization of a high afﬁnity component (PD2-1 and Max-1) and a low afﬁnity 5268 British Journal of Pharmacology (2015) 172 5265–5280 Tyrosine kinases in airway smooth muscle BJP Figure 1 Effects of kinase inhibitors on carbachol and bradykinin-induced contraction in rat bronchioles. Measurement of isometric te isolated rat bronchioles. CCh (A–D) or BK (E–H) was applied cumulatively (0.01–100μM) at 5 min intervals. Representative trac cumulative contractile responses to CCh (A) and BK (E). Arrows indicate the points where the ﬁrst concentration was applied. Two performed in each bronchiole, the second after application of the Src inhibitor PP2 (30 μM, 10 min, B: CCh, n = 8 or F: BK, n = 11), inhibitor Y27632 (10 μM, 10 min, C: CCh, n = 6 or G: BK, n = 4), the FAK inhibitor PF-573228 (10 μM, 10 min, D: CCh, n = 8 or H: inhibitor (not shown). Measurements were taken at the end of each 5 min exposure and data ﬁtted by nonlinear regression. Data ex of that induced by 80 mM KPSS (mean ± SEM); see main text or Supporting Information Fig. S4 for effects on PD2 and max valu g Effects of kinase inhibitors on carbachol and bradykinin-induced contraction in rat bronchioles. Measurement of isometric tension in freshly isolated rat bronchioles. CCh (A–D) or BK (E–H) was applied cumulatively (0.01–100μM) at 5 min intervals. Representative traces show typical cumulative contractile responses to CCh (A) and BK (E). Arrows indicate the points where the ﬁrst concentration was applied. Two responses were performed in each bronchiole, the second after application of the Src inhibitor PP2 (30 μM, 10 min, B: CCh, n = 8 or F: BK, n = 11), the Rho-kinase inhibitor Y27632 (10 μM, 10 min, C: CCh, n = 6 or G: BK, n = 4), the FAK inhibitor PF-573228 (10 μM, 10 min, D: CCh, n = 8 or H: BK, n = 5) or no inhibitor (not shown). Measurements were taken at the end of each 5 min exposure and data ﬁtted by nonlinear regression. Y Shaifta et al. 30.2 ± 6.3% inhibition of 40 mM KPSS, n = 7; P < 0.05 by unpaired t-test). We then examined the effects of PP2 and PF-573228 on SOCE- mediated contraction. SR Ca2+ was ﬁrst depleted with cyclopiazonic acid (CPA, 10 μM) in the absence of extracellular Ca2+ and presence of 200 μM EGTA, then 2 mM Ca2+ was re-applied. CPA was used here instead of a GPCR agonist because it would have been difﬁcult to separate effects of the agonist on Ca2+ entry from those on Rho- kinase-mediated Ca2+ sensitization. In control experiments, re-application of 2 mM Ca2+ induced a biphasic contraction, which peaked at ~2 min and slowly decayed to ~20% of 80 mM KPSS after 30 min. In the presence of either PP2 or PF-573228, the sustained component was signiﬁcantly smaller, decaying to <10% of 80 mM KPSS after 30 min (Figure 4C and D). PP3 was without signiﬁcant effect on this response, ruling out the possibility of a non-speciﬁc SOCE blocking effect of PP2 (Supporting Information Fig. S2B), while PF-431396 inhibited the response in a near-identical way to that of PF-573228, supporting a speciﬁc role for FAK in this response (Supporting Information Fig. S3B). To see whether these effects of PP2 and PF-573228 were due to an effect of SrcFK or FAK on SOCE itself or on secondary activation of VOCE, the effect of the SOCE blocker YM58483 (10 μM) and the Ca2+ channel antagonist nifedipine (2 μM) on the SOCE-mediated contraction was also determined. Apart from a small residual transient contraction, the response was abolished by YM58483, while nifedipine was without effect, apart from a small reduction in the peak response (Figure 4D). To conﬁrm that Rho-kinase is being activated by 30 s exposure to BK (1 μM) or CCh (100 μM), and whether this activation relates to subsequent activation of MLCK, we measured phosphorylation of MYPT-1 at T696, a known phosphorylation target of Rho-kinase (Feng et al., 1999), and of MLC20 at S19, the main target of MLCK. Furthermore, in order to reveal a possible interaction between Rho-kinase and SrcFK or FAK with relation to MLC20 phosphorylation, we examined the effects of inhibitors of Rho-kinase, SrcFK or FAK on both phosphorylation responses in hASMC. Phosphorylation of MLC20 and MYPT-1 were both signiﬁ- cantly enhanced by BK. Y Shaifta et al. BJP kinase inhibitors on bronchoconstrictor-induced contraction is mediated through a Rho-kinase-dependent Ca2+- sensitization pathway, CCh or BK concentration-response curves were repeated in α-toxin-permeabilized rat bronchioles, with [Ca2+]i ﬁxed at pCa 6.5, in the absence or presence of PP2, Y27632 or PF-573228. After α-toxin- permeabilization, bronchoconstictor concentration- response curves were not repeatable (not shown), so controls and effects of antagonists were compared in separate bronchioles. CCh-induced contraction was almost absent in the presence of Y27632 and was signiﬁcantly smaller in the presence of PP2 (P < 0.05, unpaired t-test, n = 9), but in the presence of PF-573228 was not different from controls (Figure 2B). The PD2 was signiﬁcantly greater after PP2 (4.64 ± 0.13, vs. control -5.08 ± 0.12, P < 0.05, unpaired t-test, n = 9), but was no different in PF-573228. The underlying pCa 6.5 contraction was unaffected by either PP2 or PF-573228, but was partially inhibited by Y27632 (61 ± 8% block, P < 0.05 vs. absence of Y27632, paired t-test, n = 6). BK also produced a modest (relative to CCh) concentration- dependent contraction in permeabilized bronchioles (Figure 2A and C). The concentration-dependence of these responses again appeared biphasic, but their small amplitude and poor sustainability rendered curve-ﬁtting impossible. Nevertheless, peak responses were signiﬁcantly smaller or absent in the presence of PP2 or Y27632, respectively, and no different in PF-573228 (Figure 2C). validating the choice of kinase inhibitor concentrations used. In rat trachealis muscle, SrcFK and FAK auto-phosphorylation were also enhanced by BK and CCh, as was phosphorylation of MLC20 (S19) and MYPT-1 (Y397), (Figure 3E, F). Bronchoconstrictor-induced FAK Y397 phosphorylation was noticeably weaker in rat trachealis than in hASMC. FAK and SrcFK influence SOCE/ROCE and VOCE. We examined the effects of PP2 and PF-753228 on VOCE- mediated contraction by sub-maximal depolarization with 40 mM KPSS. In control experiments, contraction amplitude induced by two consecutive KPSS exposures was not signiﬁcantly different, while when PP2 (30 μM) or PF- 573228 (10 μM) was applied between the ﬁrst and second exposures, the contractile response was modestly but signiﬁcantly reduced by PF-573228, but not by PP2 (Figure 4A and B). To rule out a direct Ca2+-channel antagonist effect of PF-573228, we also examined its effect on contraction induced by maximal depolarization with 80 mM KPSS. This contraction was signiﬁcantly less sensitive to PF-573228 than the 40 mM KPSS contraction (11.5 ± 2.7% inhibition, n = 6, vs. Data analysis and statistics Data expressed as a % of that induced by 80 mM KPSS (mean ± SEM); see main text or Supporting Information Fig. S4 for effects on PD2 and max values. values for either the high or the low afﬁnity component (not shown). for all BK dose responses curve ﬁt data). BK-induced responses were nearly abolished by Y27632 (Figure 1G) and abolished by PF-573228 (Figure 1H), rendering curve ﬁtting impossible. In time-matched control responses, repeated in the absence of inhibitor, there were no changes in either Max or PD2 SrcFKs mediate GPCR-induced Ca2+-sensitization and Rho-kinase activation, but FAK does not. To clarify whether the effects of 5270 British Journal of Pharmacology (2015) 172 5265–5280 Y Shaifta et al. (B) CCh data were ﬁtte ntrol, n = 10; PP2, n = 9; Y27632, n = 4; PF-573228, n = 11); see main text for effects on CCh PD2. (C) BK data could n ession so were compared by two-way RM ANOVA (control, n = 6; PP2, n = 6; Y27632, n = 4; PF-573228, n = 6; P < 0. –G) Measurement of phosphorylation of MLC20 at S19 (P-MLC20, D, E) and MYPT-1 at T696 (P-MYPT-1, F, G), in hASMC. ects of treatment on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on on (mean ± SEM), expressed as a % of values from untreated (control) samples run on the same gels. (D) Effects of BK (1 μM on in the absence of inhibitor (n = 16) or after pretreatment with either PP2 (30 μM, 10 min, n = 11), Y27632 (10 μM, (PF, 10 μM, 10 min, n = 11). (E) Effects of inhibitors on basal MLC20 phosphorylation (n = 4–11). (F) Effects of BK ( phorylation in the absence of inhibitor (n = 13), or after pre-application of PP2 (30 μM, 10 min, n = 13), Y27632 (1 Tyrosine kinases in airway smooth mu Tyrosine kinases in airway smooth muscle BJP Tyrosine kinases in airway smooth muscle B gure 2 cts of kinase inhibitors on contraction in α-toxin permeabilized rat bronchioles and MLC20/MYPT-1 phosphorylation in hASMC. (A–C) Measurement sometric tension in α-toxin permeabilized rat bronchioles. All responses were performed with pCa ﬁxed at 6.5 and in the presence of 10 μM CPA and M GTP. (A) Representative traces showing CCh (upper panel) or BK (lower panel) being applied cumulatively (0.01–100 μM) at 5 min intervals, with ows indicating where the ﬁrst dose was applied. Responses were performed in the absence of inhibitor (control) or after pre-incubation with either the inhibitor PP2 (30 μM, 10 min), the Rho-kinase inhibitor Y27632 (10 μM, 10 min) or the FAK inhibitor PF-573228 (10 μM, 10 min). Measurements re taken at the end of each 5 min exposure. Data are expressed as a % of that induced by pCa 4.5 (mean ± SEM). (B) CCh data were ﬁtted by nonlinear ression (control, n = 10; PP2, n = 9; Y27632, n = 4; PF-573228, n = 11); see main text for effects on CCh PD2. Figure 2 Figure 2 Effects of kinase inhibitors on contraction in α-toxin permeabilized rat bronchioles and MLC20/MYPT-1 phosphorylation in hASMC. (A–C) Measurement of isometric tension in α-toxin permeabilized rat bronchioles. All responses were performed with pCa ﬁxed at 6.5 and in the presence of 10 μM CPA and 1 μM GTP. (A) Representative traces showing CCh (upper panel) or BK (lower panel) being applied cumulatively (0.01–100 μM) at 5 min intervals, with arrows indicating where the ﬁrst dose was applied. Responses were performed in the absence of inhibitor (control) or after pre-incubation with either the Src inhibitor PP2 (30 μM, 10 min), the Rho-kinase inhibitor Y27632 (10 μM, 10 min) or the FAK inhibitor PF-573228 (10 μM, 10 min). Measurements were taken at the end of each 5 min exposure. Data are expressed as a % of that induced by pCa 4.5 (mean ± SEM). (B) CCh data were ﬁtted by nonlinear regression (control, n = 10; PP2, n = 9; Y27632, n = 4; PF-573228, n = 11); see main text for effects on CCh PD2. (C) BK data could not be ﬁtted by nonlinear regression so were compared by two-way RM ANOVA (control, n = 6; PP2, n = 6; Y27632, n = 4; PF-573228, n = 6; P < 0.05, *P < 0.01 vs. control). (D–G) Measurement of phosphorylation of MLC20 at S19 (P-MLC20, D, E) and MYPT-1 at T696 (P-MYPT-1, F, G), in hASMC. Representative blots show effects of treatment on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on the degree of phosphorylation (mean ± SEM), expressed as a % of values from untreated (control) samples run on the same gels. (D) Effects of BK (1 μM 30 s) on MLC20 phosphorylation in the absence of inhibitor (n = 16) or after pretreatment with either PP2 (30 μM, 10 min, n = 11), Y27632 (10 μM, 10 min, n = 9) or PF-573228 (PF, 10 μM, 10 min, n = 11). (E) Effects of inhibitors on basal MLC20 phosphorylation (n = 4–11). (F) Effects of BK (1 μM, 30 s) on MYPT-1 phosphorylation in the absence of inhibitor (n = 13), or after pre-application of PP2 (30 μM, 10 min, n = 13), Y27632 (10 μM, 10 min, n = 13), or PF-573228 (10 μM, 10 min, n = 8). (G) Effects of inhibitors on basal MYPT-1 phosphorylation (n = 4–11). Y Shaifta et al. (C) BK data could not be ﬁtted by nlinear regression so were compared by two-way RM ANOVA (control, n = 6; PP2, n = 6; Y27632, n = 4; PF-573228, n = 6; P < 0.05, *P < 0.01 control). (D–G) Measurement of phosphorylation of MLC20 at S19 (P-MLC20, D, E) and MYPT-1 at T696 (P-MYPT-1, F, G), in hASMC. Representative ts show effects of treatment on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on the degree of osphorylation (mean ± SEM), expressed as a % of values from untreated (control) samples run on the same gels. (D) Effects of BK (1 μM 30 s) on MLC20 osphorylation in the absence of inhibitor (n = 16) or after pretreatment with either PP2 (30 μM, 10 min, n = 11), Y27632 (10 μM, 10 min, n = 9) PF-573228 (PF, 10 μM, 10 min, n = 11). (E) Effects of inhibitors on basal MLC20 phosphorylation (n = 4–11). (F) Effects of BK (1 μM, 30 s) on PT-1 phosphorylation in the absence of inhibitor (n = 13), or after pre-application of PP2 (30 μM, 10 min, n = 13), Y27632 (10 μM, 10 min, 13), or PF-573228 (10 μM, 10 min, n = 8). (G) Effects of inhibitors on basal MYPT-1 phosphorylation (n = 4–11). Comparisons were by one- y ANOVA with Holm–Sidak post tests: P < 0.05 and *P < 0.01 versus control; #P < 0.5 and ##P <0.01 versus BK alone. Y Shaifta et al. This enhancement was signiﬁcantly reduced by PP2 and abolished by Y27632 at both sites, while PF-573228 signiﬁcantly reduced the enhancement of MLC20, but not MYPT-1 phosphorylation (Figure 2D and F). CCh also enhanced phosphorylation of both proteins, but to a lesser extent than BK. Basal MLC20 phosphorylation was insensi- tive to all three inhibitors (Figure 2E), while basal MYPT-1 phosphorylation was partially sensitive to Y27632, but insensitive to PP2 or PF-573228 (Figure 2G). Bronchoconstrictors enhance SrcFK and FAK auto- phosphorylation. In order to conﬁrm that the inﬂuence of SrcFK and FAK on contraction and Rho-kinase activity occurs in direct response to bronchoconstrictor stimulation, we also examined the effects of BK or CCh on SrcFK auto- phosphorylation at Y416 and FAK auto-phosphorylation at Y397, as a reﬂection of respective changes in kinase activity (Calalb et al., 1995; Xu et al., 1999). In hASMC, auto- phosphorylation of both kinases was signiﬁcantly enhanced by both agents (Figure 3 A–D). As expected, SrcFK phosphorylation was almost abolished by PP2, and FAK phosphorylation was almost abolished by PF-573228, both conﬁrming the selectivity of the phospho-antibodies and To support contraction data and to further eliminate the possibility that SrcFK and FAK were indirectly inﬂuencing SOCE via an action on SR Ca2+ release, we also examine the effects of PP2 and PF-753228 on SOCE [Ca2+]i responses in Fura-2 loaded hASMC, using BK as the initial SR-emptying stimulus. After the recording of an initial baseline in 2 mM Ca2+, and then for 5 min in nominally Ca2+-free buffer, the addition of BK (1 μM) caused a near instantaneous increase in [Ca2+]i. This decayed back to the baseline within ~2 min; a response 5270 British Journal of Pharmacology (2015) 172 5265–5280 e inhibitors on contraction in α-toxin permeabilized rat bronchioles and MLC20/MYPT-1 phosphorylation in hASMC. (A–C nsion in α-toxin permeabilized rat bronchioles. All responses were performed with pCa ﬁxed at 6.5 and in the presence of 1 Representative traces showing CCh (upper panel) or BK (lower panel) being applied cumulatively (0.01–100 μM) at 5 min ng where the ﬁrst dose was applied. Responses were performed in the absence of inhibitor (control) or after pre-incubation P2 (30 μM, 10 min), the Rho-kinase inhibitor Y27632 (10 μM, 10 min) or the FAK inhibitor PF-573228 (10 μM, 10 min). he end of each 5 min exposure. Data are expressed as a % of that induced by pCa 4.5 (mean ± SEM). British Journal of Pharmacology (2015) 172 5265–5280 5271 Figure 2 Comparisons were by one- way ANOVA with Holm–Sidak post tests: P < 0.05 and *P < 0.01 versus control; #P < 0.5 and ##P <0.01 versus BK alone. British Journal of Pharmacology (2015) 172 5265–5280 5271 Y Shaifta et al. BJP Figure 3 Effects of broncoconstrictors on SrcFK and FAK auto-phosphorylation in hASMC and rat trachealis. (A–D) Measurements of aut of SrcFK at Y416 (P-SrcFK, A, B) and auto-phosphorylation of FAK at Y397 (P-FAK, C, D) in hASMC. Representative blots show ef on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on the degree of phosp SEM), expressed as a % of values from untreated (control) samples run on the same gels. (A) Effect of BK (1 μM, 30 s) in the a presence of PP2 (30 μM, 10 min, n = 7) on P-SrcFK (Y416). (B) Effect of CCh (100 μM, 30 s) in the absence (n = 13) or presenc P-SrcFK (Y416). (C) Effect of BK in the absence (n = 15) or presence of PF-573228 (PF, 10 μM, 10 min, n = 13) on P-FAK (Y397). the absence (n = 16) or presence of PF-573228 (n = 11) on P-FAK (Y397). Comparisons by one-way ANOVA with Holm–Sidak po versus control; ##P < 0.01 versus BK or CCh alone. (E, F) Measurements of phosphorylation of MLC20 (S19), MYPT-1 (T696), Src (Y397) in rat trachealis muscle. (E) Effects of BK (1 μM, 30 s, n = 8). (F) Effects of CCh (100 μM, 30 s, n = 8). Comparisons P<0.05, P<0.01 versus control. BJP 5272 British Journal of Pharmacology (2015) 172 5265–5280 Figure 3 Representative traces (A) and mean measurements of peak amplitude (B, ± SEM), showing effects of two contractions in the absence of inhibitor (left panels, n = 8), the effect of PF-573228 on the second contraction (PF, middle panels, 10 μM, 5 min pre-incubation, n = 7) or the effect of PP2 on the second contraction (30 μM, right panels, 5 min pre-incubation, n = 9). Comparisons by paired t-test: P < 0.01 versus control. (C, D) SOCE-associated contraction induced by 10 μM CPA/200 μM EGTA/zero Ca2+, followed by re-application of 2 mM Ca2+. Representative traces (C) and mean measurements of amplitude of contractile responses at 2 min intervals after re-application of Ca2+ (D, ± SEM), showing control response (n = 14) and effects of pre-incubation with PF-573228 (n = 9), PP2 (n = 9), nifedipine (n = 10) or YM58483 (n = 5). Comparisons by two-way ANOVA with Holm–Sidak post tests: P < 0.05 for control versus PF-573228 or PP2. #P<0.01 for control versus YM58483. ^P < 0.05 for control versus nifedipine at 2 min only. Figure 3 g Effects of broncoconstrictors on SrcFK and FAK auto-phosphorylation in hASMC and rat trachealis. (A–D) Measurements of auto-phosphorylation of SrcFK at Y416 (P-SrcFK, A, B) and auto-phosphorylation of FAK at Y397 (P-FAK, C, D) in hASMC. Representative blots show effects of treatments on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on the degree of phosphorylation (mean ± SEM), expressed as a % of values from untreated (control) samples run on the same gels. (A) Effect of BK (1 μM, 30 s) in the absence (n = 12) or presence of PP2 (30 μM, 10 min, n = 7) on P-SrcFK (Y416). (B) Effect of CCh (100 μM, 30 s) in the absence (n = 13) or presence of PP2 (n = 8) on P-SrcFK (Y416). (C) Effect of BK in the absence (n = 15) or presence of PF-573228 (PF, 10 μM, 10 min, n = 13) on P-FAK (Y397). (D) Effect of CCh in the absence (n = 16) or presence of PF-573228 (n = 11) on P-FAK (Y397). Comparisons by one-way ANOVA with Holm–Sidak post tests: P < 0.01 versus control; ##P < 0.01 versus BK or CCh alone. (E, F) Measurements of phosphorylation of MLC20 (S19), MYPT-1 (T696), SrcFK (Y416) and FAK (Y397) in rat trachealis muscle. (E) Effects of BK (1 μM, 30 s, n = 8). (F) Effects of CCh (100 μM, 30 s, n = 8). Comparisons by unpaired t-test: P<0.05, P<0.01 versus control. typical of GPCR-induced SR Ca2+-release. 2 mM Ca2+ was then re-applied for 20 min, with simultaneous washout of BK. This induced a biphasic rise in [Ca2+]i with a similar time course to the SOCE contractile responses. In the presence of either PP2 or PF-573228, the sustained com- ponent was reduced by ~50%, while the initial transient component and the initial BK-induced SR release were both unaffected (Figure 5). Interaction between SrcFK and FAK. Because SrcFK and FAK appear to be sharing some but not all of the contraction signalling pathways investigated in this study, and in the 5272 British Journal of Pharmacology (2015) 172 5265–5280 Tyrosine kinases in airway smooth muscle BJP Tyrosine kinases in airway smooth muscle BJP Figure 4 Effects of SrcFK and FAK inhibitors on VOCE- or SOCE-associated contraction in rat bronchioles. (A, B) VOCE-associated contractions induced by 40 mM KPSS. British Journal of Pharmacology (2015) 172 5265–5280 5273 Figure 4 g Effects of SrcFK and FAK inhibitors on VOCE- or SOCE-associated contraction in rat bronchioles. (A, B) VOCE-associated contractions induced by 40 mM KPSS. Representative traces (A) and mean measurements of peak amplitude (B, ± SEM), showing effects of two contractions in the absence of inhibitor (left panels, n = 8), the effect of PF-573228 on the second contraction (PF, middle panels, 10 μM, 5 min pre-incubation, n = 7) or the effect of PP2 on the second contraction (30 μM, right panels, 5 min pre-incubation, n = 9). Comparisons by paired t-test: P < 0.01 versus control. (C, D) SOCE-associated contraction induced by 10 μM CPA/200 μM EGTA/zero Ca2+, followed by re-application of 2 mM Ca2+. Representative traces (C) and mean measurements of amplitude of contractile responses at 2 min intervals after re-application of Ca2+ (D, ± SEM), showing control response (n = 14) and effects of pre-incubation with PF-573228 (n = 9), PP2 (n = 9), nifedipine (n = 10) or YM58483 (n = 5). Comparisons by two-way ANOVA with Holm–Sidak post tests: P < 0.05 for control versus PF-573228 or PP2. #P<0.01 for control versus YM58483. ^P < 0.05 for control versus nifedipine at 2 min only. inﬂuenced by Src-dependent phosphorylation on FAK Y576/577 (Calalb et al., 1995). Phosphorylation at this dual site was enhanced by BK, and this enhancement was nearly abolished by PP2 and partially inhibited by PF-573228 (Figure 6C), while basally, this phosphorylation was inhibited by PP2, but not PF-573228 (Figure 6D). However, BK-induced enhancement of FAK (Y397) auto-phosphorylation was light of previous evidence suggesting cooperation between the two kinases, we investigated the inﬂuence of FAK on SrcFK auto-phosphorylation and vice versa. Enhancement of SrcFK (Y416) auto-phosphorylation by BK was inhibited by PF-573228 by ~50% (Figure 6A), while basally, this phosphorylation was insensitive to PF-573228 but inhibited by PP2 (Figure 6B). FAK kinase activity is also reportedly British Journal of Pharmacology (2015) 172 5265–5280 5273 Y Shaifta et al. BJP gure 5 ects of SrcFK and FAK inhibitors on SOCE/ROCE-associated [Ca2+]i responses in hASMC. A: Representative control trace of [Ca2+]i in Fura- ded hASMC, as determined by the ratio of ﬂuorescence at 340 nm/380 nm. Arrow indicates the point at which pre-stimulus basal [Ca2+ s recorded in 2 mM [Ca2+]o (and extrapolated by dashed line). Figure 4 The buffer was then switched to zero [Ca2+]o until a new baseline wa ablished, and 1 μM BK added for 5 min. Finally, 2 mM [Ca2+]o was reapplied for 20 min. Responses were performed either in the absence o sence of FAK inhibitor PF-573228 (B, C: PF, 10 μM, added 5 min prior to BK, n = 15 vs. 15 matched controls) or SrcFK inhibitor PP2 (D, E μM, added 5 min prior to BK, n = 10 vs. eight matched controls). Measurements were made of the peak BK-induced transient (B, D itrary units, mean ± SEM) and of the response to reapplication of 2 mM [Ca2+]o (C, E: arbitrary units, measured at 1 min intervals, mean ± SEM) mpared with the pre-stimulus basal [Ca2+]i in 2 mM [Ca2+]o (indicated by arrows). Background ﬂuorescence (in zero [Ca2+]o, prior to the applica n of BK) was subtracted from all other measurements. Comparisons by un-paired t-test (B, D) or two-way ANOVA with Holm–Sidak post tests (C, E < 0 05 vs matched controls) Figure 5 g Effects of SrcFK and FAK inhibitors on SOCE/ROCE-associated [Ca2+]i responses in hASMC. A: Representative control trace of [Ca2+]i in Fura-2 loaded hASMC, as determined by the ratio of ﬂuorescence at 340 nm/380 nm. Arrow indicates the point at which pre-stimulus basal [Ca2+]i was recorded in 2 mM [Ca2+]o (and extrapolated by dashed line). The buffer was then switched to zero [Ca2+]o until a new baseline was established, and 1 μM BK added for 5 min. Finally, 2 mM [Ca2+]o was reapplied for 20 min. Responses were performed either in the absence or presence of FAK inhibitor PF-573228 (B, C: PF, 10 μM, added 5 min prior to BK, n = 15 vs. 15 matched controls) or SrcFK inhibitor PP2 (D, E: 30 μM, added 5 min prior to BK, n = 10 vs. eight matched controls). Measurements were made of the peak BK-induced transient (B, D: arbitrary units, mean ± SEM) and of the response to reapplication of 2 mM [Ca2+]o (C, E: arbitrary units, measured at 1 min intervals, mean ± SEM), compared with the pre-stimulus basal [Ca2+]i in 2 mM [Ca2+]o (indicated by arrows). Background ﬂuorescence (in zero [Ca2+]o, prior to the applica- tion of BK) was subtracted from all other measurements. Comparisons by un-paired t-test (B, D) or two-way ANOVA with Holm–Sidak post tests (C, E: P < 0.05 vs. matched controls). unaffected by PP2 (Figure 6E). Basal phosphorylation of FAK (Y397) was also insensitive to PP2, but was inhibited by PF-53228 (Figure 6F). Yes and Lyn (Sakai et al., 2010). For this reason, and the fact that the phospho-SrcFK antibody cannot distinguish between Src family members, we re-examined the effects of acute BK treatment on MLC20 (S19), MYPT-1 (T696), SrcFK (Y416) and FAK (Y576/577) phosphorylation in hASMC after transfection with c-Src siRNA or scrambled siRNA. Speciﬁcity of c-Src siRNA was veriﬁed by a ~70% reduction in c-Src c-Src is the principle SrcFK mediating bronchoconstrictor-induced phosphorylation responses. Multiple members of the Src- family of kinases are expressed in ASM, including c-Src, Fyn, 5274 British Journal of Pharmacology (2015) 172 5265–5280 Tyrosine kinases in airway smooth muscle Tyrosine kinases in airway smooth muscle BJP i i hil i f MLC MYPT 1 d Di i ure 6 action between SrcFK and FAK in response to BK in hASMCs. Figure 6 g Interaction between SrcFK and FAK in response to BK in hASMCs. Measurements of auto-phosphorylation of SrcFK at Y416 (P-SrcFK, A, B), phosphorylation of FAK at the Y576/577 dual site (C, D), and auto-phosphorylation of FAK at Y397 (P-FAK, E, F) in hASMC. Representative blots show effects of treatments on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on the degree of phosphorylation (mean ± SEM), expressed as a % of values from untreated (control) samples run on the same gels. (A) Effects of BK (1μM, 30 s) on P-SrcFK (Y416) in the absence (n = 14) or presence of the FAK inhibitor PF-573228 (PF, 10μM 10 min, n = 7). (B) Effects of PF (n = 4) or the SrcFK inhibitor PP2 (30μM, 10 min, n = 4) on basal P-SrcFK (Y416). (C) Effects of BK on P-FAK (Y576/577) in the absence (n = 11) or presence of PP2 (n = 6) or PF (n = 7). (D) Effects of PF (n = 4) or PP2 (n = 5) on basal P-FAK (Y576/577). (E) Effect of BK on P-FAK (Y397) in the absence (n = 14) or presence of PP2 (n = 4). (F) Effects of PF (n = 4) or PP2 (n = 4) on basal P-FAK (Y397). Comparisons by one-way ANOVA with Holm–Sidak post tests: P < 0.01 versus control or ##P < 0.01 versus BK alone. Figure 5 Measurements of auto-phosphorylation of SrcFK at Y416 (P-SrcFK, A, B), phorylation of FAK at the Y576/577 dual site (C, D), and auto-phosphorylation of FAK at Y397 (P-FAK, E, F) in hASMC. Representative blots effects of treatments on ‘phospho’ and ‘total’ immunoreactivity for each protein. Bar charts show the effects of treatments on the degree of phorylation (mean ± SEM), expressed as a % of values from untreated (control) samples run on the same gels. (A) Effects of BK (1μM, 30 s) on FK (Y416) in the absence (n = 14) or presence of the FAK inhibitor PF-573228 (PF, 10μM 10 min, n = 7). (B) Effects of PF (n = 4) or the SrcFK itor PP2 (30μM, 10 min, n = 4) on basal P-SrcFK (Y416). (C) Effects of BK on P-FAK (Y576/577) in the absence (n = 11) or presence of PP2 6) or PF (n = 7). (D) Effects of PF (n = 4) or PP2 (n = 5) on basal P-FAK (Y576/577). (E) Effect of BK on P-FAK (Y397) in the absence (n = 14) esence of PP2 (n = 4). (F) Effects of PF (n = 4) or PP2 (n = 4) on basal P-FAK (Y397). Comparisons by one-way ANOVA with Holm–Sidak post P < 0.01 versus control or ##P < 0.01 versus BK alone. BJP Discussion protein expression, while expressions of MLC20, MYPT-1 and FAK were all unaffected (Figure 7A). The scrambled siRNA had no effect on any of the proteins examined. c-Src siRNA inhibited BK-induced phosphorylation of MLC20 by ~80%, while responses of MYPT-1 (T696), SrcFK (Y416) and FAK (Y576/577) to BK were all inhibited by ~60%, compared with matched scrambled siRNA transfected cells (Figure 7 B–D). We examined the role of SrcFK and FAK kinase activity in bronchoconstrictor-induced contraction of rat-isolated bronchioles and in [Ca2+]i and phosphorylation responses in hASMC. Contraction was induced and MLC20 phosphor- ylation was enhanced by the bronchoconstrictors BK and British Journal of Pharmacology (2015) 172 5265–5280 5275 Y Shaifta et al. BJP Figure 7 Effects of c-Src siRNA on bronchoconstrictor-induced MLC20, MYPT-1, SrcFK and FAK phosphorylation in hASMCs. Meas expression and phosphorylation responses to BK (1 μM, 30 s) in hASMCs after transfection with c-Src siRNA or scrambled siRN of c-Src siRNA on c-Src protein expression and (lack of) effect on MLC20, MYPT-1 and FAK expression. Data normalized to GAPD same samples (arbitrary units, n = 8–12). Comparisons by unpaired t-test: P < 0.01 versus scram siRNA. (B) Effect of BK on lation at S19 (P-MLC20) after transfection with scram siRNA (n =16) or c-Src siRNA (n = 16). (C) Effect of BK on MYPT-1 phosp (P-MYPT-1) after transfection with scram siRNA (n = 14) or c-Src siRNA (n = 14). (D) Effect of BK on SrcFK phosphorylation at Y transfection with scram siRNA (n = 16) or c-Src siRNA (n = 16). (E) Effect of BK on FAK phosphorylation at Y576/577 (P-FAK) aft scram siRNA (n = 10) or c-Src siRNA (n = 10). Comparisons by two-way ANOVA: P < 0.01 versus control; ##P < 0.01 versu BJP 5276 British Journal of Pharmacology (2015) 172 5265–5280 Tyrosine kinases in airway smooth muscle Some GPCR agonists, notably angiotensin II, also mediate SR Ca2+ release via SrcFK-dependent tyrosine phosphorylation of PLC-γ (Schmitz et al., 1997). However, there is no indi- cation that this is occurring in our study, because we found that neither PP2 nor PF-573228 altered the BK-induced Ca2+ transients indicative of SR release. Our ﬁnding that the CPA-induced SOCE contraction was minimally affected by the Ca2+ channel antagonist nifedipine suggests that VOCE secondary to SOCE-induced depolarization was not contributing substantially to this response. However, VOCE may also be activated more directly via a number of signalling pathways including stretch-activated phosphatidylcholine-speciﬁc PLC-derived DAG (Mauban et al., 2015) or integrin-directed tyrosine phosphorylation. Regarding the latter, engagement of integrin α5β1 induces SrcFK and FAK-dependent phosphory- lation of the α1c subunit of the L-type Ca2+ channel in vascular smooth muscle, with the likeliest sequence of events being integrin-induced FAK auto-phosphorylation followed by SrcFK recruitment by FAK (Owen et al., 1999; Salazar and Rozengurt, 2001) and subsequent direct phos- phorylation of the channel by SrcFK (Wu et al., 2001; Gui et al., 2006). Similarly in ASM, VOCE may also be enhanced via stretch or adhesion-induced FAK activity (Smith et al., 1998; Tang et al., 1999). In bronchioles, we found that contraction induced by depolarization with sub-maximal (40 mM) K+ was sensitive to PF-573228 but not PP2, which suggests that FAK may be activating VOCE independently of SrcFK. However, contraction induced by maximum depo- larization with 80 mM K+ was considerably less sensitive to PF-573228, ruling out a non-speciﬁc effect of PF-573228 on Ca2+ channel opening per se. We did not systematically examine the effects of stretch on contractile responses, as carried out previously in trachea (Tang et al., 1999), but applied a standard degree of stretch to maximize active tension responses to bronchoconstrictors or KPSS. Similarly, adherent hASMC are assumed to be under a degree of self-induced basal tension (Deguchi et al., 2005). In light of this, it is worth noting that the relatively weak FAK auto- phosphorylation observed in trachealis samples treated with BK or CCh, compared with similarly treated hASMC, may have been because no stretch was applied to trachealis samples during BK or CCh treatment prior to snap freezing for protein extraction. Rho-kinase is directly activated by the small G-protein RhoA, which is itself activated by guanine nucleotide exchange factors (RhoGEFs). Tyrosine kinases in airway smooth muscle To further characterize the signalling pathways through which SrcFK and FAK mediate ASM contraction, we ﬁrst fo- cussed on their role in Rho-kinase dependent Ca2+-sensitiza- tion, a process whereby inhibition of MLCP results in enhanced MLC20 phosphorylation, and hence force genera- tion, without the requirement for an increase in [Ca2+]i (Somlyo and Somlyo, 2003). Both BK and CCh-induced con- traction were highly sensitive to Rho-kinase inhibition with Y27632. Furthermore, a component of the contractile re- sponse to both BK and CCh persisted when bronchioles were permeabilized with α-toxin to prevent changes in intracellu- lar Ca2+. We found that these contractile responses were dependent on Rho-kinase and SrcFK, but not FAK. Further- more, we found that MYPT-1 phosphorylation on T696, an indicator of Rho-kinase-mediated MLCP inhibition (Feng et al., 1999), is also enhanced by BK and CCh and that this enhancement is sensitive to inhibition of Rho-kinase and SrcFK, but not of FAK. This inﬂuence of SrcFK on Rho-kinase activity occurred speciﬁcally in response to agonist stimula- tion, because baseline phosphorylation of MYPT-1 and MLC20 and baseline pCa6.5 contraction in permeabilized bronchioles were not affected by SrcFK inhibition. Clearly, these results indicate that SrcFK mediates GPCR-induced smooth muscle contraction in part via activation of Rho- kinase. Importantly, this is the ﬁrst direct demonstration of an interaction between SrcFK and Rho-kinase in ASM and in vessels of a size relevant to the control of airway resistance, consistent with the implied importance of Rho-kinase in airway hyper-responsiveness from studies in whole animal or isolated upper airways (Yoshii et al., 1999; Schaafsma et al., 2006). Interestingly, receptor TK stimulation also induces Rho-kinase-dependent ASM contraction (Gosens et al., 2004), and other responses of ASM to growth factor stimulation are also SrcFK-dependent (Krymskaya et al., 2005). Thus SrcFK may be a point of convergence for the activation of Rho-kinase in response to either GPCR or growth factors. inﬂux, is partially dependent on SrcFK-mediated phosphory- lation of STIM1 (Lopez et al., 2012). In accord with these pre- vious studies, we found that CPA-induced SOCE-dependent contraction in rat bronchioles and BK-induced Ca2+ inﬂux in hASMC were both similarly sensitive to SrcFK inhibition. Interestingly, we found that these responses were also similarly sensitive to FAK inhibition. This, to our knowledge, is the ﬁrst indication that FAK may be contributing to GPCR- induced Ca2+ responses and contraction in human and rodent ASM, via upstream modulation of SOCE and/or ROCE. Figure 7 g Effects of c-Src siRNA on bronchoconstrictor-induced MLC20, MYPT-1, SrcFK and FAK phosphorylation in hASMCs. Measurement of protein expression and phosphorylation responses to BK (1 μM, 30 s) in hASMCs after transfection with c-Src siRNA or scrambled siRNA (scram). (A) Effect of c-Src siRNA on c-Src protein expression and (lack of) effect on MLC20, MYPT-1 and FAK expression. Data normalized to GAPDH expression in the same samples (arbitrary units, n = 8–12). Comparisons by unpaired t-test: P < 0.01 versus scram siRNA. (B) Effect of BK on MLC20 phosphory- lation at S19 (P-MLC20) after transfection with scram siRNA (n =16) or c-Src siRNA (n = 16). (C) Effect of BK on MYPT-1 phosphorylation at T696 (P-MYPT-1) after transfection with scram siRNA (n = 14) or c-Src siRNA (n = 14). (D) Effect of BK on SrcFK phosphorylation at Y416 (P-SrcFK) after transfection with scram siRNA (n = 16) or c-Src siRNA (n = 16). (E) Effect of BK on FAK phosphorylation at Y576/577 (P-FAK) after transfection with scram siRNA (n = 10) or c-Src siRNA (n = 10). Comparisons by two-way ANOVA: P < 0.01 versus control; ##P < 0.01 versus scram siRNA. CCh, and these responses were sensitive to inhibition of both SrcFK and FAK. Using auto-phosphorylation as an in- dication of kinase activity, both SrcFK and FAK were acti- vated by both agents in hASMC and rat trachealis, suggesting an important role for these kinases in GPCR- induced ASM contraction in both humans and rodents. PP3, the negative control for PP2, was without effect on contraction, while PF-431396, another inhibitor of FAK, had similar effects as PF-573228, reducing the likelihood of our results being inﬂuenced by non-speciﬁc effects of PP2 or PF-573228. Amongst the several Src family members, both c-Src and Fyn have been implicated in vascular smooth mus- cle contraction (Nakao et al., 2002; Knock et al., 2008), and Lyn is activated by muscarinic agonists in ASM (Pertel et al., 2006). Here however, the effects of c-Src siRNA suggest that 60–80% of all the bronchoconstrictor-induced phosphoryla- tion responses investigated herein were speciﬁcally mediated by c-Src. 5276 British Journal of Pharmacology (2015) 172 5265–5280 British Journal of Pharmacology (2015) 172 5265–5280 5277 Tyrosine kinases in airway smooth muscle RhoGEFs are known to be acti- vated or modulated by various non-receptor TKs (Chikumi et al., 2002; Ying et al., 2009; Guilluy et al., 2010), in addition to G12/13 binding. It is therefore conceivable that SrcFK may be activating RhoA/Rho-kinase via direct phosphorylation of a RhoGEF. Alternatively, they may do so via the prior acti- vation of another kinase, such as FAK, PYK2 or JAK2 (Calalb et al., 1995; Andreev et al., 2001; Singh et al., 2011). Our results are not inconsistent with this hypothesis, but exclude FAK as the intermediary kinase in this instance. We next focussed on the role of SrcFK and FAK in Ca2+ sig- nalling. Gq/PLC-β-coupled GPCRs induce Ca2+ entry through three main pathways: DAG-sensitive TRP channel opening (ROCE), IP3-dependent depletion of SR Ca2+ and subsequent STIM1/Orai1/TRP-dependent inﬂux (SOCE), and subsequent depolarization-induced opening of L-type Ca2+ channels (VOCE) (Kawasaki et al., 2006; Wang et al., 2008). Several members of the TRPC family of channels, in addition to being modulated by DAG, are subject to modulation by phosphory- lation, and tyrosine phosphorylation of TRPC channels is SrcFK-dependent, contributing to either ROCE or SOCE (Kawasaki et al., 2006). Moreover, association of STIM1 with Orai1 in response to SR depletion, and subsequent Ca2+ In smooth muscle contraction, it has been assumed that the mutual dependence between SrcFK and FAK relates pri- marily to the recruitment of contractile ﬁbres through actin polymerization and focal attachment formation (Gerthoffer and Gunst, 2001; Tang et al., 1999; Gunst et al., 2003). It is therefore of note that we only see possible evidence of such mutuality with regard to SOCE/ROCE activity, but not in British Journal of Pharmacology (2015) 172 5265–5280 5277 ournal of Pharmacology (2015) 172 5265–5280 5277 Y Shaifta et al. BJP Figure 8 Proposed role for SrcFK and FAK in bronchoconstrictor or depolarization-induced ASM contraction, based on the results of the current study and existing literature. (A) GPCR activate both SrcFK and FAK, presumably via interaction with heterotrimeric G-protein sub-units (e.g. Gαq, Gα12/13 or Gβγ) or downstream signalling molecules. Activated SrcFK forms two distinct sub-populations: one FAK-independent and one FAK-dependent. (B) FAK-independent SrcFK induces Rho-kinase to phosphorylate MYPT-1, thus enhancing myosin light-chain phosphorylation (P-MLC) through inhibition of myosin light-chain phosphatase (MLCP). SrcFK are perhaps activating Rho-kinase via the tyrosine phosphorylation of RhoA- associated proteins such as RhoGEFs. Figure 8 The selective sensitivity of basal SrcFK auto-phosphorylation to PP2 and of basal FAK auto- phosphorylation to PF-537228 conﬁrms the speciﬁcity of these two antagonists for their intended targets at the concentrations used in this study. Therefore, our ﬁnding that BK-induced SrcFK auto-phosphorylation is partially PF- 537228-sensitive suggests a partial dependence of GPCR- induced SrcFK activity on FAK. This is probably because SrcFK can be activated via disruption of intramolecular auto- inhibition by association of the SH2 domain with the phos- phorylated Y397 of FAK (Xing et al., 1994). Conversely, we show that BK-induced FAK auto-phosphorylation is wholly independent of SrcFK activity. In response to adhesion, in non-muscle cells, FAK auto-phosphorylation is enhanced by SrcFK-mediated phosphorylation on Y576/577 (Calalb et al., 1995; Salazar and Rozengurt, 2001). However, despite the fact that we show that FAK Y576/577 phosphorylation is also induced by BK and almost entirely SrcFK-mediated, this phosphorylation does not correlate with enhanced FAK auto- phosphorylation. A similar discrepancy was observed in ﬁbroblasts where adhesion-induced FAK auto-phosphorylation was SrcFK-dependent but GPCR-induced FAK auto- phosphorylation was not, despite both stimuli inducing SrcFK-dependent Y576/577 phosphorylation (Salazar and Rozengurt, 2001). Taken together, our results imply two things about these interactions in ASM: ﬁrstly, that GPCR can induce FAK activation without prior activation of SrcFK, and secondly, that there may be two sub-populations of BK/CCh-activated SrcFK; one FAK-dependent, resulting in modulation of SOCE/ROCE, and the other FAK-independent, resulting in activation of Rho-kinase (as summarized in Figure 8). In conclusion, our data suggest an important role for SrcFK and FAK in bronchoconstrictor-mediated contraction in ASM, with the two kinases acting together to induce SOCE/ROCE, and independently to mediate Rho-kinase- dependent Ca2+ sensitization and VOCE respectively. These ﬁndings may inform the search for new drug targets for the treatment of obstructive lung diseases such as asthma, and in particular, support the suggested key role for SrcFK in experimental airway hyper-responsiveness (Sakai et al., 2010; Katsumoto et al., 2013). Tyrosine kinases in airway smooth muscle (C) A FAK/Src complex may be mediating store-operated and/or receptor-operated Ca2+ entry (SOCE/ ROCE) via the tyrosine phosphorylation of TRP channels, or associated signalling proteins, such as STIM1, in association with DAG. (D) FAK is also independently enhancing voltage-operated Ca2+ entry (VOCE), perhaps via direct phosphorylation of voltage-dependent Ca2+ channels, in response to the additional stimulus of stretch or cellular adhesion, via integrin engagement. BJP S a ta et a . Figure 8 Proposed role for SrcFK and FAK in bronchoconstrictor or depolarization-induced ASM contraction, based on the results of the current study and existing literature. (A) GPCR activate both SrcFK and FAK, presumably via interaction with heterotrimeric G-protein sub-units (e.g. Gαq, Gα12/13 or Gβγ) or downstream signalling molecules. Activated SrcFK forms two distinct sub-populations: one FAK-independent and one FAK-dependent. (B) FAK-independent SrcFK induces Rho-kinase to phosphorylate MYPT-1, thus enhancing myosin light-chain phosphorylation (P-MLC) through inhibition of myosin light-chain phosphatase (MLCP). SrcFK are perhaps activating Rho-kinase via the tyrosine phosphorylation of RhoA- associated proteins such as RhoGEFs. (C) A FAK/Src complex may be mediating store-operated and/or receptor-operated Ca2+ entry (SOCE/ ROCE) via the tyrosine phosphorylation of TRP channels, or associated signalling proteins, such as STIM1, in association with DAG. (D) FAK is also independently enhancing voltage-operated Ca2+ entry (VOCE), perhaps via direct phosphorylation of voltage-dependent Ca2+ channels, in response to the additional stimulus of stretch or cellular adhesion, via integrin engagement. relation to Rho-kinase activation (SrcFK only) or VOCE activity (FAK only). The selective sensitivity of basal SrcFK auto-phosphorylation to PP2 and of basal FAK auto- phosphorylation to PF-537228 conﬁrms the speciﬁcity of these two antagonists for their intended targets at the concentrations used in this study. Therefore, our ﬁnding that BK-induced SrcFK auto-phosphorylation is partially PF- 537228-sensitive suggests a partial dependence of GPCR- induced SrcFK activity on FAK. This is probably because SrcFK can be activated via disruption of intramolecular auto- inhibition by association of the SH2 domain with the phos- phorylated Y397 of FAK (Xing et al., 1994). Conversely, we show that BK-induced FAK auto-phosphorylation is wholly independent of SrcFK activity. In response to adhesion, in non-muscle cells, FAK auto-phosphorylation is enhanced by SrcFK-mediated phosphorylation on Y576/577 (Calalb et al., 1995; Salazar and Rozengurt, 2001). However, despite the fact that we show that FAK Y576/577 phosphorylation is also induced by BK and almost entirely SrcFK-mediated, this phosphorylation does not correlate with enhanced FAK auto- phosphorylation. A similar discrepancy was observed in ﬁbroblasts where adhesion-induced FAK auto-phosphorylation was SrcFK-dependent but GPCR-induced FAK auto- phosphorylation was not, despite both stimuli inducing SrcFK-dependent Y576/577 phosphorylation (Salazar and Rozengurt, 2001). Taken together, our results imply two things about these interactions in ASM: ﬁrstly, that GPCR can induce FAK activation without prior activation of SrcFK, relation to Rho-kinase activation (SrcFK only) or VOCE activity (FAK only). 5278 British Journal of Pharmacology (2015) 172 5265–5280 References Kawasaki BT, Liao Y, Birnbaumer L (2006). Role of Src in C3 transient receptor potential channel function and evidence for a heterogeneous makeup of receptor- and store-operated Ca2+ entry channels. Proc Natl Acad Sci U S A 103: 335–340. Alexander SP, Benson HE, Faccenda E, Pawson AJ, Sharman JL, Catterall WA et al. (2013a). The concise guide to pharmacology 2013/14: ion channels. Br J Pharmacol 170: 1607–1651. Alexander SP, Benson HE, Faccenda E, Pawson AJ, Sharman JL, Spedding M et al. (2013b). The concise guide to pharmacology 2013/14: enzymes. Br J Pharmacol 170: 1797–1867. Knock GA, Shaifta Y, Snetkov VA, Vowles B, Drndarski S, Ward JP et al. (2008). Interaction between Src family kinases and Rho-kinase in agonist-induced Ca2+-sensitization of rat pulmonary artery. Cardiovasc Res 77: 570–579. Andreev J, Galisteo ML, Kranenburg O, Logan SK, Chiu ES, Okigaki M et al. (2001). Src and Pyk2 mediate G-protein-coupled receptor activation of epidermal growth factor receptor (EGFR) but are not required for coupling to the mitogen-activated protein (MAP) kinase signaling cascade. J Biol Chem 276: 20130–20135. Krymskaya VP, Goncharova EA, Ammit AJ, Lim PN, Goncharov DA, Eszterhas A et al. (2005). Src is necessary and sufﬁcient for human airway smooth muscle cell proliferation and migration. FASEB J 19: 428–430. Calalb MB, Polte TR, Hanks SK (1995). Tyrosine phosphorylation of focal adhesion kinase at sites in the catalytic domain regulates kinase activity: a role for Src family kinases. Mol Cell Biol 15: 954–963. Lopez E, Jardin I, Berna-Erro A, Bermejo N, Salido GM, Sage SO et al. (2012). STIM1 tyrosine-phosphorylation is required for STIM1-Orai1 association in human platelets. Cell Signal 24: 1315–1322. Chikumi H, Fukuhara S, Gutkind JS (2002). Regulation of G protein- linked guanine nucleotide exchange factors for Rho, PDZ-RhoGEF, and LARG by tyrosine phosphorylation: evidence of a role for focal adhesion kinase. J Biol Chem 277: 12463–12473. Mauban JR, Zacharia J, Fairfax S, Wier WG (2015). PC-PLC/ sphingomyelin synthase activity plays a central role in the development of myogenic tone in murine resistance arteries. Am J Physiol Heart Circ Physiol 308: H1517–H1524. Chopra LC, Hucks D, Twort CH, Ward JP (1997). Effects of protein tyrosine kinase inhibitors on contractility of isolated bronchioles of the rat. Am J Respir Cell Mol Biol 16: 372–378. McGrath JC, Drummond GB, McLachlan EM, Kilkenny C, Wainwright CL (2010). Guidelines for reporting experiments involving animals: the ARRIVE guidelines. Br J Pharmacol 160: 1573–1576. Author contributions Guilluy C, Bregeon J, Toumaniantz G, Rolli-Derkinderen M, Retailleau K, Loufrani L et al. (2010). The Rho exchange factor Arhgef1 mediates the effects of angiotensin II on vascular tone and blood pressure. Nat Med 16: 183–190. All other authors reviewed the manuscript critically for important intellectual content. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Arhgef1 mediates the effects of angiotensin II on vascular tone and blood pressure. Nat Med 16: 183–190. Gunst SJ,Tang DD,Opazo SA(2003).Cytoskeletal remodeling oftheairway smoothmusclecell:amechanismforadaptationtomechanicalforcesinthe lung. Respir Physiol Neurobiol 137: 151–168. Ishigaki T, Imanaka-Yoshida K, Shimojo N, Matsushima S, Taki W, Yoshida T (2011). Tenascin–C enhances crosstalk signaling of integrin alphavbeta3/PDGFR-beta complex by SRC recruitment promoting PDGF-induced proliferation and migration in smooth muscle cells. J Cell Physiol 226: 2617–2624. Conﬂict of interest Authors declare that they have not any conﬂict of interest. Katsumoto TR, Kudo M, Chen C, Sundaram A, Callahan EC, Zhu JW et al. (2013). The phosphatase CD148 promotes airway hyperresponsiveness through Src family kinases. J Clin Invest 123: 2037–2048. Funding Gosens R, Schaafsma D, Grootte Bromhaar MM, Vrugt B, Zaagsma J, Meurs H et al. (2004). Growth factor-induced contraction of human bronchial smooth muscle is Rho-kinase-dependent. Eur J Pharmacol 494: 73–76. Gosens R, Schaafsma D, Grootte Bromhaar MM, Vrugt B, Zaagsma J, Meurs H et al. (2004). Growth factor-induced contraction of human bronchial smooth muscle is Rho-kinase-dependent. Eur J Pharmacol 494: 73–76. Wellcome Trust: #087776. British Heart Foundation: FS/12/ 43/29608. Gui P, Wu X, Ling S, Stotz SC, Winkfein RJ, Wilson E et al. (2006). Integrin receptor activation triggers converging regulation of Cav1.2 calcium channels by c-Src and protein kinase A pathways. J Biol Chem 281: 14015–14025. Acknowledgements Thanks to Mrs Kheem Jones and other NHS staff for patient recruitment and provision of bronchoscopy biopsies. This work was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London. The views expressed are those of the author(s) and not necessarily those of the NHS, NIHR or the Department of Health. 5278 British Journal of Pharmacology (2015) 172 5265–5280 BJP Y Shaifta et al. Y Shaifta et al. BJP BJP Tang DD, Gunst SJ (2001). Depletion of focal adhesion kinase by antisense depresses contractile activation of smooth muscle. Am J Physiol Cell Physiol 280: C874–C883. migration requiring both auto- and activation loop phosphorylation sites and inhibits adhesion-dependent tyrosine phosphorylation of Pyk2. Mol Cell Biol 19: 4806–4818. migration requiring both auto- and activation loop phosphorylation sites and inhibits adhesion-dependent tyrosine phosphorylation of Pyk2. Mol Cell Biol 19: 4806–4818. Ui-Tei K, Naito Y, Takahashi F, Haraguchi T, Ohki-Hamazaki H, Juni A et al. (2004). Guidelines for the selection of highly effective siRNA sequences for mammalian and chick RNA interference. Nucleic Acids Res 32: 936–948. Pawson AJ, Sharman JL, Benson HE, Faccenda E, Alexander SP, Buneman OP, Davenport AP, McGrath JC, Peters JA, Southan C, Spedding M, Yu W, Harmar AJ; NC-IUPHAR. (2014) The IUPHAR/BPS Guide to PHARMACOLOGY: an expert-driven knowledgebase of drug targets and their ligands. Nucl. Acids Res. 42 (Database Issue): D1098-106. Spedding M, Yu W, Harmar AJ; NC-IUPHAR. (2014) The IUPHAR/BPS Guide to PHARMACOLOGY: an expert-driven knowledgebase of drug targets and their ligands. Nucl. Acids Res. 42 (Database Issue): D1098-106. Wang Y, Deng X, Hewavitharana T, Soboloff J, Gill DL (2008). Stim, ORAI and TRPC channels in the control of calcium entry signals in smooth muscle. Clin Exp Pharmacol Physiol 35: 1127–1133. Pertel T, Zhu D, Panettieri RA, Yamaguchi N, Emala CW, Hirshman CA (2006). Expression and muscarinic receptor coupling of Lyn kinase in cultured human airway smooth muscle cells. Am J Physiol Lung Cell Mol Physiol 290: L492–L500. Wu X, Davis GE, Meininger GA, Wilson E, Davis MJ (2001). Regulation of the L-type calcium channel by alpha 5beta 1 integrin requires signaling between focal adhesion proteins. J Biol Chem 276: 30285–30292. Pinelli V, Marchica CL, Ludwig MS (2009). Allergen-induced asthma in C57Bl/6 mice: hyper-responsiveness, inﬂammation and remodelling. Respir Physiol Neurobiol 169: 36–43. Xing Z, Chen HC, Nowlen JK, Taylor SJ, Shalloway D, Guan JL (1994). Direct interaction of v-Src with the focal adhesion kinase mediated by the Src SH2 domain. Mol Biol Cell 5: 413–421. Reynolds A, Leake D, Boese Q, Scaringe S, Marshall WS, Khvorova A (2004). Rational siRNA design for RNA interference. Nat Biotechnol 22: 326–330. Xu W, Doshi A, Lei M, Eck MJ, Harrison SC (1999). Crystal structures of c-Src reveal features of its autoinhibitory mechanism. Mol Cell 3: 629–638. Y Shaifta et al. Sakai H, Nishimura A, Watanabe Y, Nishizawa Y, Hashimoto Y, Chiba Y et al. (2010). Involvement of Src family kinase activation in angiotensin II-induced hyperresponsiveness of rat bronchial smooth muscle. Peptides 31: 2216–2221. Ying Z, Giachini FR, Tostes RC, Webb RC (2009). PYK2/PDZ- RhoGEF links Ca2+ signaling to RhoA. Arterioscler Thromb Vasc Biol 29: 1657–1663. Yoshii A, Iizuka K, Dobashi K, Horie T, Harada T, Nakazawa T et al. (1999). Relaxation of contracted rabbit tracheal and human bronchial smooth muscle by Y-27632 through inhibition of Ca2+ sensitization. Am J Respir Cell Mol Biol 20: 1190–1200. Salazar EP, Rozengurt E (2001). Src family kinases are required for integrin-mediated but not for G protein-coupled receptor stimulation of focal adhesion kinase autophosphorylation at Tyr-397. J Biol Chem 276: 17788–17795. Schaafsma D, Bos IS, Zuidhof AB, Zaagsma J, Meurs H (2006). Inhalation of the Rho-kinase inhibitor Y-27632 reverses allergen- induced airway hyperresponsiveness after the early and late asthmatic reaction. Respir Res 7(121): 1–7. References Deguchi S, Ohashi T, Sato M (2005). Intracellular stress transmission through actin stress ﬁber network in adherent vascular cells. Mol Cell Biomech 2: 205–216. Meurs H, Gosens R, Zaagsma J (2008). Airway hyperresponsiveness in asthma: lessons from in vitro model systems and animal models. Eur Respir J 32: 487–502. Doeing DC, Solway J (2013). Airway smooth muscle in the pathophysiology and treatment of asthma. J Appl Physiol 114: 834–843. Moir LM, Leung SY, Eynott PR, McVicker CG, Ward JP, Chung KF et al. (2003). Repeated allergen inhalation induces phenotypic modulation of smooth muscle in bronchioles of sensitized rats. Am J Physiol Lung Cell Mol Physiol 284: L148–L159. Feng J, Ito M, Ichikawa K, Isaka N, Nishikawa M, Hartshorne DJ et al. (1999). Inhibitory phosphorylation site for Rho-associated kinase on smooth muscle myosin phosphatase. J Biol Chem 274: 37385–37390. NakaoF,KobayashiS,MogamiK,MizukamiY,ShiraoS,MiwaSetal.(2002). InvolvementofSrcfamilyproteintyrosinekinasesinCa2+sensitizationof coronaryarterycontractionmediatedbyasphingosylphosphorylcholine- Rho-kinase pathway. Circ Res 91: 953–960. Gerthoffer WT, Gunst SJ (2001). Invited review: focal adhesion and small heat shock proteins in the regulation of actin remodeling and contractility in smooth muscle. J Appl Physiol 91: 963–972. Gilbert R, Auchincloss JH Jr (1989). Reactive airways dysfunction syndrome presenting as a reversible restrictive defect. Lung 167: 55–61. Owen JD, Ruest PJ, Fry DW, Hanks SK (1999). Induced focal adhesion kinase (FAK) expression in FAK-null cells enhances cell spreading and British Journal of Pharmacology (2015) 172 5265–5280 5279 Supporting Information Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Schmitz U, Ishida M, Berk BC (1997). Angiotensin II stimulates tyrosine phosphorylation of phospholipase C-gamma-associated proteins. Characterization of a c-Src-dependent 97-kD protein in vascular smooth muscle cells. Circ Res 81: 550–557. http://dx.doi.org/10.1111/bph.13313 Singh NK, Wang D, Kundumani-Sridharan V, Van QD, Niu J, Rao GN (2011). 15-Lipoxygenase-1-enhanced Src-Janus kinase 2-signal transducer and activator of transcription 3 stimulation and monocyte chemoattractant protein-1 expression require redox- sensitive activation of epidermal growth factor receptor in vascular wall remodeling. J Biol Chem 286: 22478–22488. Figure S1 Identiﬁcation of hASMC as smooth muscle by positive staining with anti-smooth muscle-actin (panel A), anti-desmin (panel B) and anti-calponin (panel C), visualised with Alexa Fluor®488 labelled secondary antibody (Lifetechnologies) and ﬂuorescent microscopy. Cells were also stained with TRITC-labelled phalloidin to conﬁrm the presence of stress ﬁbres in resting cells (Panel D). In Panel D, nuclei are visualised by staining with Hoechst. Scale bar =20μm. Smith PG, Garcia R, Kogerman L (1998). Mechanical strain increases protein tyrosine phosphorylation in airway smooth muscle cells. Exp Cell Res 239: 353–360. Somlyo AP, Somlyo AV (2003). Ca2+ sensitivity of smooth muscle and nonmuscle myosin II: modulated by G proteins, kinases, and myosin phosphatase. Physiol Rev 83: 1325–1358. Figure S2 Effect of PP3 on contractile responses in rat bronchioles. Figure S2 Effect of PP3 on contractile responses in rat bronchioles. Figure S3 Effect of PF-431396 on contractile responses in rat bronchioles. Figure S3 Effect of PF-431396 on contractile responses in rat bronchioles. Tang D, Mehta D, Gunst SJ (1999). Mechanosensitive tyrosine phosphorylation of paxillin and focal adhesion kinase in tracheal smooth muscle. Am J Physiol 276: C250–C258. Figure S4 Effects of SrcFK, Rho-kinase and FAK inhibition on bradykinin-induced contractile responses in rat bronchioles. Figure S4 Effects of SrcFK, Rho-kinase and FAK inhibition on bradykinin-induced contractile responses in rat bronchioles. 5280 British Journal of Pharmacology (2015) 172 5265–5280
https://openalex.org/W4225001173	https://discovery.dundee.ac.uk/files/74583304/s12909_022_03381_z.pdf	English	null	The educational value of an audience response system use in an Iraqi medical school	BMC medical education	2,022	cc-by	7,709	The educational value of an audience response system use in an Iraqi medical school Tuma, Faiz; Majeed, Husam; Blebea, John; Nassar, Aussama; Durchholz, William C.; Schofield Susie The educational value of an audience response system use in an Iraqi medical school Tuma, Faiz; Majeed, Husam; Blebea, John; Nassar, Aussama; Durchholz, William C.; Schofield Susie The educational value of an audience response system use in an Iraqi medical school Citation for published version (APA): Tuma, F., Majeed, H., Blebea, J., Nassar, A., Durchholz, W. C., & Schofield, S. (2022). The educational value of an audience response system use in an Iraqi medical school. BMC Medical Education, 22, Article 319. https://doi.org/10.1186/s12909-022-03381-z General rights Copyright and moral rights for the publications made accessible in Discovery Research Portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Citation for published version (APA): Tuma, F., Majeed, H., Blebea, J., Nassar, A., Durchholz, W. C., & Schofield, S. (2022). The educational value of an audience response system use in an Iraqi medical school. BMC Medical Education, 22, Article 319. https://doi.org/10.1186/s12909-022-03381-z University of Dundee The educational value of an audience response system use in an Iraqi medical school Tuma, Faiz; Majeed, Husam; Blebea, John; Nassar, Aussama; Durchholz, William C.; Schofield, Susie Published in: University of Dundee The educational value of an audience response system use in an Iraqi medical school Tuma, Faiz; Majeed, Husam; Blebea, John; Nassar, Aussama; Durchholz, William C.; Schofield, Susie Published in: Abstract Background: Medical education is continually evolving particularly through the modern implementation of edu- cational technology. Enhancing interactive learning in the classroom or lecture settings is one of the growing uses of educational technology. The role and potential benefits of such technology may not be as evident in developing educational systems like the one in Iraq. The purpose of this study was to examine the effect and perception of the use of an audience response system (ARS) on interactive medical education in Iraq. A mixed quantitative and qualita- tive research methodology approach was used to study the effects and users’ perceptions (both student and tutor) of the ARS. Method: The study was conducted in an Iraqi medical school in the Head and Neck course during the spring semes- ter for third-year medical students. The course involved fifteen one-hour lectures over fifteen weeks. Users’ percep- tions were evaluated by survey and focus group discussions (FGD). Descriptive statistics were used for quantitative measures and thematic analysis for the qualitative data. An ARS system was installed and integrated into the course lectures throughout the course period of three months to enhance interactive learning. Three to five interactive questions were used in each lecture. Anonymous participa- tion and answers were maintained. The appropriate discussion was initiated when pertinent depending on students’ answers. Result: Most students (77% of survey, 85% of FGD) perceived the use of ARS as impactful on their learning. They found the ARS engaging (70%), motivating (76%), promoting interactions (73%), and augment learning through better understanding and remembering (81%). Through the FGD, students expressed improved focus, enhanced thinking and reflection, and joyful learning. The educator perceived the ARS use as practical, interactive, thinking- stimulator, and reflective of student’s understanding. The required technology skills were reasonable; however, it demanded extra non-insignificant time to learn the use. Conclusion: The perception of the ARS in this study was overall positive, providing encouragement for wide appli- cation of this technology in medical education in the developing world. Further studies are needed to validate and prioritize ARS usage in medical education in Iraq. Keywords: Educational Technology, Education, Medical, Students, Medical, Iraq, Surveys and Questionnaires, Educational Measurement Introduction Medical education is continually evolving with very sig- nificant changes having taken place during the last two decades [1]. There has been a gradual shift in medi- cal trainees’ education in various aspects including the *Correspondence: faiz.tuma@gmail.com p g 1 Central Michigan University College of Medicine, PO Box 4181, Saginaw, MI 48606, USA Full list of author information is available at the end of the article General rights i h d Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Download date: 24. Oct. 2024 Tuma et al. BMC Medical Education (2022) 22:319 https://doi.org/10.1186/s12909-022-03381-z Open Access Methodsh The study was conducted in an Iraqi medical school (Wasit University) during the 2018-2019 academic year as part of introduction of new simple educational tech- nology. The medical school follows the same national six-year curriculum (3 basic and 3 clinical years). The ARS has been considered for use as an introduction of a simple technology enhanced interactive education. Upon the completion of the course, the following queries were addressed: 1) students’ perception of the ARS use experi- ence as assessed by a quantitative survey and a qualitative method using Focus Group Discussion (FGD) [17]; 2) and an evaluation of the instructor’s perception of the experi- ence using a survey. Approval from the Dean’s office of Wasit University College of Medicine, the official licens- ing authority to approve all experiments in the university medical school, was obtained. There is no institutional or licensing committee at this university. Participation was voluntary and anonymized for the quantitative part (the survey questionnaire), and informed consent was obtained for the qualitative part (FGD). Another change has been the utilization of educational technology on medical curricula. This has been partly driven by improved availability and power of hardware, software and Wi-Fi, and also by growing class sizes [5]. Many educators have begun to use educational technol- ogy to allow learners to respond to and interact with materials, both within the face-to-face context and online [6, 7]. Enhancing involvement, participation, and maxi- mum interactivity for both students and faculty is an area for potential improvement in medical education [8]. There is an increasing trend toward shifting from tradi- tional teaching to student-centered teaching that actively engages students [9]. Creating interactivity within the classroom is becom- ing easier than ever with available educational technol- ogy tools [10]. One way of increasing interaction between educators and learners during a learning activity is via an audience response system (ARS). ARS is a relatively simple technology tool that allows educators to poll the audience and collect instant responses which can then be shared with all participants instantaneously [11]. Polling the audience instantly promotes further exploration and discussion of point of special importance. Furthermore, quizzing all students encourages individual engage- ment while also informing the instructor immediately of the students’ levels of understanding. Various meth- ods have been used over the years [12]. © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visithttp://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Tuma et al. BMC Medical Education (2022) 22:319 Tuma et al. BMC Medical Education (2022) 22:319 Page 2 of 10 Page 2 of 10 traditional lectures where delivery of information is the focus of the activity to a more engaging and participatory style of teaching. This is, in part, due to the increasing evidence that lectures alone (as a method of delivering information) are not effective in solidifying long-term knowledge acquisition nor in promoting its application to the clinical setting [2, 3]. There is an increasing trend toward self-directed learning that actively engages stu- dents in enquiry-based learning [4]. The utilization and extent of this trend in developing educational systems such as Iraq are not well known. been in the form of sporadic projects of limited appli- cation initiated by individuals or small groups and with little subsequent formal evaluation. However, any inter- vention must be acceptable to all stakeholders to have lasting utility [16]. Therefore, the aim of this study was to provide such a formal evaluation of using a simple introductory technology to enhance interactivity in large group teaching and improve our understanding about medical education in the context of the ARS in Iraq. Methodsh However, rather than a "magic bullet" to educational woes, these systems are merely tools which can be used in a number of ways; therefore, defining what the ARS can add to a learning environment is required [13]. Examining the exact role in a particular learning environment provides further guid- ance of how to apply and better use such a tool. The ARS’ potential role and best practice in medical education was explored in this study. ARSs are generally used in multiple ways: as a learn- ing strategy to facilitate improving attention, interac- tion, instruction, student preparation and discussion, and formative and summative knowledge assessments. The particular use and setting of the ARS might determine the extent of its usefulness. It is therefore important to study and choose the appropriate setting before imple- menting ARS use, especially with changing educational styles and evolving technology.h The Head and Neck course was chosen due to the motivation and technology competency of the instruc- tor who was willing to implement the new change. This course is taken in the third year of the curriculum. The course involved fifteen one-hour lectures over fifteen weeks. The ARS system with the associated software was installed and tested prior to the beginning of the course. This particular ARS system used a small, dedicated hand- held keypad to respond to questions posed by the edu- cator with various different types of questions. At the beginning of the course, the instructor gave the students a short introduction in using ARS. He then used ARS- based questions throughout the lecture presentation. Interactive education has been used to a limited extent in Iraq [14]. The use of educational technology is a rela- tively new experience to the medical education there [15]. The country has been through a prolonged period of political conflict and civil war. Only a very limited amount of educational technology has been introduced in a few Iraqi educational institutions [14]. These have Page 3 of 10 Tuma et al. BMC Medical Education (2022) 22:319 Tuma et al. BMC Medical Education (2022) 22:319 Three to five questions were incorporated in each lecture based on the instructor selection where deemed neces- sary to enhance interactivity. Questions were designed to evaluate understanding and enhance critical thinking. analysis of the two data sets combining and comparing the results. Methodsh In this convergent use, the two types of data can create a stronger foundation for conclusions and vali- dation of the results [24]. Students’ participation in answering the lecturer’s questions using the ARS was voluntary; however, they were encouraged to participate to learn. At the end of the course, students were invited to participate anonymously in the study survey. Voluntary and anonymous participa- tion were maintained in the survey. This was important for the validity of the study results as anonymity enhances the rate of response, accuracy of information, and validity [18]. There were no positive reward or gifts for participat- ing in the study. Neither was there a punishment or nega- tive reward for not participating in the study. After the survey, focus group discussion (FGD) was conducted in a sequential explanatory design to expand and strengthen the study conclusions [19–21]. Educator’s perception of using the ARS As part of the educational process and experience, the educator’s perception is of great importance in estimating how ARS could become part of their educational curricu- lum in the future. A survey was therefore developed to assess the educational value and technical aspects from the instructor’s perspective. The survey was conducted at the end of the course and was divided into the prepara- tion and lecture delivery phases. Similar to the student questionnaire, a 5-point Likert scale from “Strongly Disa- gree” to “Strongly Agree” was used to measure responses to each question (Appendix B). Focus Group Discussion (FGD) Focus Group Discussion (FGD) FGD aims to gain data about specific study parts from purposely selected participants. Participants were selected and a comfortable environment for the FGD was prepared. Open discussion was facilitated by an experienced moderator (FT) and systematic theme con- tent analysis was subsequently performed. Recurrent thematic statements that were related to the educa- tional experience and perception were used. The mixed method was used to verify and validate findings from the quantitative and qualitative data. This involved separate Study validity and reliability A comprehensive survey was designed and structured for this study using the principles of evaluation. The main four dimensions of the evaluation of the educa- tional activities (structure, process, instructor, and out- come) as discussed by Schiekirka et al. were considered. [22] Therefore, the survey questions covered the essen- tial components of the educational activity, measuring the intended outcomes of students learning, focusing on the adult style of learning that uses learner-guided activi- ties and goals for learning, and using as valid and reliable questions as possible. The survey questions were used with a small group as a pilot test before the main study. Modifications of some of the questions were accordingly applied. A 5-point Likert scale from “Strongly Disagree” to Strongly Agree” was used to measure responses to each question. Percentage calculations and interpretation of the answers on the fifteen survey questions (Table 1 and Fig. 1) were collected and described as both the mean and mode. The mode was used to define the most common response while the mean was employed to as a measure of central tendency [23]. Validity and reliability were considered and enforced at different levels of the study. Content validity that indi- cates the adequacy and effectiveness of measuring a vari- able was optimized by changing and adjusting some of the survey questions according to the pilot study. Inter- nal validity was enhanced by using different ways of col- lecting data (survey and FGD). External validity in this study is challenging. The applicability of this study in other setting is difficult to determine since it is a single- site single-course study. However, this time-limited study will be considered a first step to promote further studies in related or different contexts before implementing the results. External reliability that measures the replicability of the results is one of the known weaknesses of mixed method research. The different modalities of the data collection instrument and the availability of the participants are dif- ficult to replicate. The internal reliability that reflects the consistency of data collection, analysis and interpreta- tion has one weakness that is related to the analyzing of the FGD. This is an inherent interviewer variability with FGD. However, inter-rater, or inter-observer reliability was high considering the educational level and the scope of the authors’ practice. Results Survey Sixty-three responses from all the participants in the course were obtained for all the fifteen questions (100% participation rate) (Table 1). The results strongly indicate that the students perceived several advantages of using the ARS. Within each domain, they demonstrate: Tuma et al. Attendance, engagement and memoryh questions about a lecture topic had their questions answered using the ARS. Students strongly agreed that the use of the ARS encouraged the use of technology in their education (80%) and it was not difficult to use the ARS (77%). Participants (84%) wished to use this ARS in other courses (Table 1). The majority of students agreed the ARS helped to improve attention and focus during lectures. Eight three per cent of students agree or strongly agree that the ARS helped with memorization of information, and 70% felt it helped answer questions without embarrass- ment. Most students (81%) strongly agreed the use of the ARS motivated them to attend lectures (Table 1). Adult and active learning ( Most of the students (76%) felt that the ARS motivated them to prepare for lectures in advance. Sixty-eight per cent of the students agree or strongly agreed that the ARS stimulated them to study and review further after lecture, and 73% felt that it encouraged discussion with colleagues and instructors (Table 1). Results Survey BMC Medical Education (2022) 22:319 Page 4 of 10 Table 1 All answers by students’ survey questions with mean and mode Table 1 All answers by students’ survey questions with mean and mode a Q 10 & 14 were reverse scored for the mean b Answers by attendance, engagement and memory domain c Answers by learning preferences and use of technology domain d Answers by adult and active learning domain e Answers by learning efficiency/quality and the effect of educational technology domain SD D N A SA mean mode # Number assigned for calculating mean 1 2 3 4 5 1 b The ARS helped to improve my attention and focus during the lecture 2 3% 0 0% 7 11% 25 40% 29 46% 4.25 SA 2 e The ARS helped me to understand the topics of the lecture better 2 3% 2 3% 8 13% 30 48% 21 33% 4.05 A 3 d The ARS stimulated me to discuss the topic with my colleagues and teacher 2 3% 2 3% 13 21% 22 35% 24 38% 4.02 SA 4 d The ARS stimulated me to prepare for the topic in advance 2 3% 3 5% 10 16% 20 32% 28 44% 4.10 SA 5 d The ARS stimulated me to study and review the topic more after the lecture 2 3% 3 5% 15 24% 19 30% 24 38% 3.95 SA 6 b The ARS helped me to memorize information more 3 3% 1 2% 7 11% 29 46% 23 36% 4.08 A 7 b The ARS motivated me to attend lectures 4 6% 1 2% 7 11% 20 32% 31 49% 4.16 SA 8 b The ARS helps me answer questions and participate with no embarrassment 3 5% 0 0% 16 25% 23 37% 21 33% 3.94 A 9 c Using ARS provided answers to some of the questions that I have about the topic 1 2% 2 3% 17 27% 31 49% 12 19% 3.81 A a 10 e Using ARS was a waste of timea 28 44% 22 35% 0 0% 10 16% 3 5% 3.98 SD 11 c I wish ARS used in all other subjects 2 3% 0 0% 8 13% 19 30% 34 54% 4.32 SA 12 e Using ARS made me like the topic more than other topics 3 5% 3 5% 9 14% 20 32% 28 44% 4.06 SA 13 c Using ARS encouraged me to use technology in learning 2 3% 1 2% 9 14% 21 33% 30 47% 4.21 SA a14 c It was difficult to use the ARSa 28 44% 21 33% 6 10% 3 5% 5 8% 4.02 SD 15 c Overall, I find the ARS is an efficient tool of teaching 2 3% 3 5% 6 10% 22 35% 30 48% 4.19 SA Total 38 34 138 344 391 4.08 Learning efficiency, quality, and technology effecth The majority of the students liked the ARS use and felt it improved their understanding (Table 1). Eighty-one per- cent (81%) of students agreed or strongly agreed the ARS augmented their learning, helping them to understand the topics of the lecture better. Most students (79%) disa- greed or strongly disagreed that the ARS was a waste of time. Students also report they liked the topic more than Learning preferences and use of technologyh There is predominant agreement on the overall use- fulness of the ARS and their wish to use it in other courses. Sixty-eight percent (68%) of students with Tuma et al. BMC Medical Education (2022) 22:319 Page 5 of 10 Fig. 1 All questions from students’ survey (1-15) responses Learning efficiency, quality, and technology effecth other topics they have covered because of the use of the ARS (Table 1). Learning efficiency, quality, and technology effecth Discussionh 1- Time consuming. Students expressed concerns about consuming lecture time on the ARS use - “We spend a lot of time and we wait for the answers” (S1) The use of ARS in medical education is a new experi- ence for many medical schools, including those in Iraq. The introduction of the ARS use has promoted the use of technology in education and enhanced the concept of 2- Repetition. Some students felt that there is excessive repetition of steps in the ARS use - “Every time it is Focus Group Discussion (FGD) In reviewing the FGD content, the following themes emerged: disadvantages and advantages. Within each theme, a few subthemes became evident (Table 2). These are labelled Sn where n is consecutive number, to Table 2 FGD Themes and frequencies Caterory 1: Advantages Subcategory Theme Frequency Students Example Qoute A Summarizing the topic 2 S2, S4 “Summary of topics is good” B Improving focus 5 S2, S3, S4, S7, S8 “I like it and I can pay attention” C Thinking simulation 6 S1, S4, S5, S6, S7, S8 “we understand and ask questions”. D Helps preparing for the exam 1 S4 “It helps study for the exam” E Special way of learning. 2 S6, S7 “It is different from other subjects; I learn more here” F Entertaining 6 S2, S3, S4, S5, S6, S8 “We like it and it’s like playing game” Total 22 Caterory 2: Disadvantages Subcategory Theme Frequency Students Example Qoute A Time consuming 3 S1, S3, S8 We spend a lot of time and we wait for the answers B Repetition 1 S5 Every time it is the same things we answer ques- tions and wait until we see the answers Total 4 Table 2 FGD Themes and frequencies Tuma et al. BMC Medical Education (2022) 22:319 Page 6 of 10 anonymize the respondents and to show a wide spread of contributors to the discussions. the same thing, we answer questions and wait until we see the answers” (S5). Educator’s perception of ARSh B- Improving focus. Several students commented on how the ARS focused their attention - ‘I like it and I can pay attention’ (S4).h B- Improving focus. Several students commented on how the ARS focused their attention - ‘I like it and I can pay attention’ (S4).h The answers and results obtained from the educator’s survey are summarized in Table 3. The educator strongly agreed that the level of difficulty in learning and using the ARS was within his skill level. The educator felt the time consumed in preparing ARS questions for each lec- ture was reasonable and that it is practical and conveni- ent for lecture preparation. He also agreed that the use of the ARS did not affect the amount of lecture content delivered per session. He strongly agreed there was more interaction, deeper thinking, and broader involvement from students while using the ARS. C- Thinking stimulation. Several students also expressed better understanding and critical thinking with using the ARS - “we understand and ask questions” (S8). D- Helps preparing for the exam. Students felt that the use of ARS helped them to prepare for the exam - “It helps study for the exam” (S4). E- Special way of learning. Some students found learn- ing with the ARS is a special way of learning - “It is different from other subjects; I learn more here” (S6). The educator felt more enthusiastic about teaching with better-quality lectures while using the ARS. The ARS helped to tailor the lecture according to the stu- dents’ understanding and needs, while also helping to evaluate students’ knowledge and performance. Overall, the educator enjoyed using the ARS and felt that the ARS is an efficient tool for teaching. f F- Entertaining. Several students felt it was entertaining to learn with the ARS use - “We like it and it’s like playing game” (S8). II- Students’ perceived disadvantages of the ARS use by subthemes with quotes examples are presented below: I‑ Advantages: The frequencies associated with categories and subcat- egories were used to adjudicate the weight of different responses [25, 26]. A- Use to summarize a topic. Students felt the use of ARS to summarize topics was useful - “summary of topics is good’ (S2). Table 3 Educator’s perception of the ARS use A Preparation phase: 1 Time consumed in learning how to use the ARS is reasonable. Disagree 2 Time consumed in preparing ARS questions and slides for the lecture is reasonable. Agree. 3 Level of difficulty in learning and using ARS is within my skills level. Strongly agree. 4 Using ARS is practical and convenient in terms of lecture preparation. Agree. 5 Preparing and incorporating ARS questions for each lecture requires reasonable efforts. Neutral B Lecture delivery phase: 1 Using ARS does not affect the amount of lecture content I can deliver. Agree 2 I observe more interaction from students using ARS in the lecture. Strongly agree. 3 I noticed more thinking and deeper involvement from students using ARS. Strongly agree. 4 I feel more enthusiastic to teach more and better using ARS. Strongly agree 5 I enjoy using ARS. Strongly agree. 6 I feel I deliver a better-quality lecture using ARS. Strongly agree. 7 I have a better sense of students understanding of the topic using ARS. Strongly agree 8 ARS helps me tailor the lecture according to the students’ understanding and needs. Agree 9 ARS helps me to evaluate students’ overall knowledge and performance. Agree 10 Overall, I find the ARS is an efficient tool of teaching. Strongly agree Tuma et al. BMC Medical Education (2022) 22:319 Page 7 of 10 interactive learning. Although there is an abundancy of studies evaluating the use of ARS in Western culture, it is important that contextualized studies are done. It is imperative to understand both student and tutor percep- tions, and the impact on using such a technology, both from the view of the dynamics of the group and from stu- dent results. This study provides an early evaluation of this novel experience. and attendance [33]. Attention and interaction of learn- ers may have long term consequences on memory [34]. Focus and attention are indicators of engagement in the educational activity which is an essential component of adult interactive learning. The use of technology and learning preferences domain was addressed by both the survey and FGD. The poten- tial issue with the use of new technology difficulties did not seem to be an obstacle in the use of the ARS. On the contrary, most of the students found it stimulating to use technology in learning. Students’ perceptionh The results of the survey collectively demonstrate a posi- tive preference and advantages of ARS use in all the four domains. ARS use was perceived as liked, preferred, wanted, entertained, and advantageous. Furthermore, students’ responses revealed that the ARS helped them to better understand the topic and stimulated them to fur- ther discussion and thinking. FGD analysis revealed that students liked the ARS mostly because it stimulated them to think more deeply. Questions in education are often used by teachers to stimulate building knowledge and critical thinking skills [27, 28]. When the questions are embedded within a lecture, this becomes an invitation to think more deeply about the content of the lecture. This is a first step of the thinking process followed by addi- tional stimulation from the shared responses of the audi- ence and comparing answers. Such answer comparisons are a natural form of learning [29, 30]. Students like to compare themselves and their responses to the classroom responses. This generates a second round of thinking, reviewing and comparing, especially when the responses are diverse. Depending on the setting of the group activ- ity, this could be led by the course instructor for further interactive discussion and shared learning. Question 5 in the students’ survey, “The ARS stimulated me to study and review the topic more after the lecture”, addresses this point specifically. Two thirds of the students agreed or strongly agreed on this effect. This reflects the extended positive effect of the ARS use on continuing learning beyond the session. Motivation increases persis- tence in achieving learning goals [31]. The use of ARS technology provides excitement to learners [36]. This can lead to the possibility that the other features of the ARS use were liked because of other factors e.g., entertainment rather than real positive per- ception of the features. The Technology Acceptance Model (TAM) provides some explanation of the tech- nology use behavior and intention by associating it with the attitude toward technology and ease of use [37]. This uncertainty needs further exploration.h There is clear evidence from the study of the multiple advantages of the ARS use as perceived by the students. As a new electronic tool, the ARS was attractive to the students. Entertaining tools and activities may influence the entire activity to the positive side [38]. Table 3 Educator’s perception of the ARS use The use of technology by student continues to grow worldwide, with students reporting they use desktop computers, interactive whiteboards, smart phones, and tablets [35]. The positive percep- tion and openness to use technology in education is an important factor in the introduction of more technology in education with expanding the use and application. Students’ perceptionh Another factor that might have contributed is the novelty of the expe- rience to the students and the medical school. Students expressed excitement to use this technology as the first medical school in the country of Iraq. The course organ- izer and moderator expressed similar excitement.h The disadvantages of using the ARS reported from the FGD were much fewer than the advantages. The two main disadvantages that were reported are “time con- sumed” and “repetition” of the question posting process. The extra time needed for the ARS use and its questions is a known issue and disadvantage [39]. Interestingly, the instructor did not believe that using ARS affected the amount of lecture content that was delivered. The advantages in the domains of attendance, engage- ment, and memory were evident. Survey results showed perception of improving focus and attracting to attend the educational activity (lecture). Most students found the ARS helpful in participating and interacting in the lecture without embarrassment. Nelson [32] reported five out of six studies he reviewed favored learner inter- actions in ARS lectures. Facilitating sharing input helps increase the interaction. The students also found the sys- tem stimulating to prepare for the lecture and discuss the topic with their instructor and colleagues. Hassanin et al. reported that the ARS use encouraged students to discuss the topic with peer, in addition to improving engagement Educator’s perceptionh The instructor’s perception was generally positive to all the questions of the survey except the two technical questions. Educators using technology often complain about the extra time needed to use technology. But, this extra time to learn the new technology can be evaluated against the length or term of the technology use as well as with the importance of the technology. Technology Tuma et al. BMC Medical Education (2022) 22:319 Page 8 of 10 acceptance and use by teachers has been the focus of prior research, e.g. the Technology Acceptance Model (TAM) was developed to explain the influential factors and mechanisms of technology use, including in class- rooms [40]. It is, therefore, important to consider all factors that influence the adoption of the technology con- sidered for use. part of the medical educational system. However, it is important to choose the appropriate technology for the specific educational purpose. This ARS study demon- strated the smooth implementation of technology with acceptable and manageable challenges. Further studies and information are needed to select, prioritize, and design the optimal ARS use in the vari- ous educational environments. Caution should be exercised to avoid generalizations too quickly as there might be other factors that could determine the utility and advantages of using ARS in other types of courses and different educational activities. Ours was a rela- tively short course with immediate outcomes measure- ment that might be different than long term outcomes and for longer use. The course had one instructor which limits the data available for evaluation. Alternative way of evaluating the perception was to conduct personal interview that may provide in-depth perception input. However, the course was expected to be delivered by more than one instruc- tor but ended up completed by one instructor for reasons out of the study control.h The instructor was very motivated and enthusiastic which may not reflect the average medical educator’s attitude and availability. According to Sharma and Sriv- astava, teachers who are willing to adopt new tools are motivated in adopting new teaching approaches [41]. Teachers often resist using new technologies in their classrooms because of the challenges of the new expe- rience [42]. Instructors of courses that are planning to introduce the use of ARS should be prepared to spend initial extra time to learn how to use the technology. Abbreviations d Abbreviations ARS: Audience response system; FGS: Focus group discussions; TAM: Technol- ogy Acceptance Model. Ethics approval and consent to participatefi The Dean’s office of Wasit University College of Medicine approved the study. There is no ethics committee/institutional review board at this university. The Dean’s office is the official licensing authority to approve all experiments in the university medical school. This study was performed in accordance with the Declaration of Helsinki. Participation was voluntary, anonymized (for the quantitative part), and was completed with informed consent. Funding Another limitation is the short-term use of the ARS and the lack of long term follow up. Long-term out- comes and knowledge retention were not embedded into this study due to time limitations but would be useful to study in the future. The scope of this study allows to focus on learning, understanding and short-term applica- tion of information. Larger and long-term studies can be designed to evaluate higher level of learning and knowl- edge retention. No funding was received. Availability of data and materials The datasets generated and/or analyzed during the current study are available in the appendices. Consent for publication Consent for publication Not applicable. Acknowledgements Acknowledgements The assistance of Salih Salman, Haider, and Mohammad Tuma in preparing the manuscript and references styling is greatly valued. Authors’ contributions FT: study conceptualization and design, data collection and analysis, manu- script writing, editing, revising, and project supervision. HM: study design and moderation, data collection and analysis, and supervision. JB: manuscript writ- ing, editing, revising, and supervising. AN: manuscript editing, revising, discuss- ing, and submitting. WD: manuscript editing, revising, and submitting. SS: study conceptualizing and design, data analysis, manuscript writing, editing, revising, and project supervision. The author(s) read and approved the final manuscript. Study limitationsh The study evaluated one instructor’s perception through a written survey. More instructors’ perception, if avail- able, would be more accurate. In addition, qualitative personal interview might provide more detailed and accurate input from the instructor. Conclusions and recommendationsh This study showed a strong agreement on multiple per- ceived advantages of the ARS use in lectures. The ARS induced interactivity and improved the learning process during lectures. The instructor’s role is crucial in the introduction of such technology successfully. Educa- tional technology will play an increasing and important Educator’s perceptionh With the current level of technology use in all aspect of life and the wide exposure of people to it, learning ARS technology is not lengthy nor difficult. Easy students’ learning and adjustment to technology and the use of the ARS was observed in other studies [39, 43]. Supplementary Information The online version contains supplementary material available at https://doi. org/10.1186/s12909-022-03381-z. Additional file 1. References 2013;25(Suppl 1):S39–43. https://doi.org/10.1080/ 10401334.2013.842914. 29. Colt HG, Davoudi M, Murgu S, Zamanian Rohani N. Measuring learning gain during a one-day introductory bronchoscopy course. Surg Endosc. 2011;25(1):207–16. https://doi.org/10.1007/s00464-010-1161-4. 8. Faghihi SA, Khankeh HR, Hosseini SJ, et al. Improving continuing medical education by enhancing interactivity: lessons from Iran. J Adv Med Educ Prof. 2016;4(2):54–63. 30. Silver HF. Compare & Contrast: Teaching Comparative Thinking to Strengthen Student Learning. Alexandria, VA: ASCD; 2010. 31. Higashi RM, Schunn CD, Flot JB. Different underlying motivations and abilities predict student versus teacher persistence in an online course. Educ Tech Res Dev. 2017;65(6):1471–93. https://doi.org/10.1007/ s11423-017-9528-z. 9. Abdel Meguid E, Collins M. Students’ perceptions of lecturing approaches: traditional versus interactive teaching. Adv Med Educ Pract. 2017;8:229-241. Published 2017 Mar 17. doi:https://doi.org/10.2147/ AMEP.S131851 32. Nelson C, Hartling L, Campbell S, Oswald AE. The effects of audi- ence response systems on learning outcomes in health professions education. A BEME systematic review: BEME Guide No. 21. Med Teach. 2012;34(6):e386–405. https://doi.org/10.3109/0142159X.2012.680938. 10. Bartrum P. The Impact of Interactive Technology on the Future of School Education. MDR. https://mdreducation.com/2018/05/22/interactive- technology-education/. Published December 31, 2018. Accessed Febru- ary 14, 2021. 33. Hassanin H, Essa K, Ahmed El-Sayed M, Attallah MM. Enhancement of student learning and feedback of large group engineering lectures using audience response systems. J Mater Educ. 2016;3856:175–90. 11. Toscos T, Drouin M, Flanagan M, et al. Audience Response Systems and Missingness Trends: Using Interactive Polling Systems to Gather Sensitive Health Information From Youth. JMIR Form Res. 2019;3(3):13798. https:// doi.org/10.2196/13798. 34. Dolcos F, Katsumi Y, Moore M, et al. Neural correlates of emotion-atten- tion interactions: From perception, learning, and memory to social cogni- tion, individual differences, and training interventions. Neurosci Biobehav Rev. 2020;108:559–601. https://doi.org/10.1016/j.neubiorev.2019.08.017. 12. Knapen R. 20 interactive teaching activities for in the interactive class- room. BookWidgets Blog. https://www.bookwidgets.com/blog/2018/ 06/20-interactive-teaching-activities-for-in-the-interactive-classroom. Published June 13, 2018. Accessed February 21, 2021. 35. Bernstein L. New Global Survey Offers Snapshot of Technology in the Classroom. Technology Solutions That Drive Education. https://edtec hmagazine.com/k12/article/2019/02/new-global-survey-offers-snapshot- technology-classroom-2019. Published September 15, 2020. Accessed February 14, 2021. Published June 13, 2018. Accessed February 21, 2021. 13. Roschelle J, Penuel W, Abrahamson L. Classroom Response and Com- munication Systems: Research Review and Theory. 2004. 14. Weber AS, Hamlaoui S. E-Learning in the Middle East and North Africa (MENA) Region. Springer, 2018. 36. Gousseau M, Sommerfeld C, Gooi A. Tips for using mobile audience response systems in medical education. Adv Med Educ Pract. 2016;7:647– 52. https://doi.org/10.2147/AMEP.S96320. Competing interests h There are no competing interests to declare. There are no competing interests to declare. Page 9 of 10 Tuma et al. BMC Medical Education (2022) 22:319 Tuma et al. BMC Medical Education (2022) 22:319 Received: 12 November 2021 Accepted: 4 April 2022 21. Greene JC, Caracelli VJ, Graham WF. Toward a conceptual framework for mixed-method evaluation designs. Educ Eval Policy Anal. 1989;11(3):255– 74. https://doi.org/10.3102/01623737011003255. 22. Schiekirka S, Feufel MA, Herrmann-Lingen C, Raupach T. Evaluation in medical education: a topical review of target parameters, data collection tools and confounding factors. Ger Med Sci. 2015;13:Doc15. https://doi. org/10.3205/000219. References 1. Collado-Yurrita L, Ciudad-Cabañas MJ, Cuadrado-Cenzual MA. Evolution of medical education in the Department of Medicine of the Complutense University of Madrid in the last decade. Med Teach. 2018;40(5):449–52. https://doi.org/10.1080/0142159X.2018.1441987. 23. Sauro J. Can You Take the Mean of Ordinal Data? MeasuringU. https:// measuringu.com/mean-ordinal/. Published May 14, 2016. Accessed February 21, 2021. g 2. Alexander CJ, Crescini WM, Juskewitch JE, Lachman N, Pawlina W. Assess- ing the integration of audience response system technology in teaching of anatomical sciences. Anat Sci Educ. 2009;2(4):160–6. https://doi.org/10. 1002/ase.99. 2. Alexander CJ, Crescini WM, Juskewitch JE, Lachman N, Pawlina W. Assess- ing the integration of audience response system technology in teaching of anatomical sciences. Anat Sci Educ. 2009;2(4):160–6. https://doi.org/10. 1002/ase.99. 24. Wisdom J, Creswell JW. Mixed Methods: Integrating Quantitative and Qualitative Data Collection and Analysis While Studying Patient-Centered Medical Home Models. AHRQ. https://pcmh.ahrq.gov/page/mixed- methods-integrating-quantitative-and-qualitative-data-collection-and- analysis-while. Published March 2013. Accessed February 21, 2021. 3. Tuma F. The use of educational technology for interactive teaching in lectures. Ann Med Surg (Lond). 2021;62:231-235. Published 2021 Jan 21. doi:https://doi.org/10.1016/j.amsu.2021.01.051 3. Tuma F. The use of educational technology for interactive teaching in lectures. Ann Med Surg (Lond). 2021;62:231-235. Published 2021 Jan 21. doi:https://doi.org/10.1016/j.amsu.2021.01.051 25. Doody O, Slevin E, Taggart L. Focus group interviews part 3: analysis. Br J Nurs. 2013;22(5):266–9. https://doi.org/10.12968/bjon.2013.22.5.266. p g j 4. Hutchings B. Principles of Enquiry-Based Learning. Centre for Excellence in Enquiry-Based Learning. 2006. http://www.ceebl.manchester.ac.uk/ resources/papers/ceeblgr002.pdf. 4. Hutchings B. Principles of Enquiry-Based Learning. Centre for Excellence in Enquiry-Based Learning. 2006. http://www.ceebl.manchester.ac.uk/ resources/papers/ceeblgr002.pdf. 26. Cyr J. Focus Groups for the Social Science Researcher. Cambridge Univer- sity Press; 2019. 27. Tregonning AM, Doherty DA, Hornbuckle J, Dickinson JE. The audience response system and knowledge gain: a prospective study. Med Teach. 2012;34(4):e269–74. https://doi.org/10.3109/0142159X.2012.660218. 5. US Department of Education. Reimagining the Role of Technology in Education. Tech Ed. 2017. https://tech.ed.gov/files/2017/01/NETP17.pdf. 5. US Department of Education. Reimagining the Role of Technology in Education. Tech Ed. 2017. https://tech.ed.gov/files/2017/01/NETP17.pdf. 6. Caldwell JE. Clickers in the large classroom: current research and best- practice tips. CBE Life Sci Educ. 2007;6(1):9–20. https://doi.org/10.1187/ cbe.06-12-0205. 6. Caldwell JE. Clickers in the large classroom: current research and best- practice tips. CBE Life Sci Educ. 2007;6(1):9–20. https://doi.org/10.1187/ cbe.06-12-0205. 28. Tofade T, Elsner J, Haines ST. Best practice strategies for effective use of questions as a teaching tool. Am J Pharm Educ. 2013;77(7):155. https:// doi.org/10.5688/ajpe777155. 7. Han H, Resch DS, Kovach RA. Educational technology in medical educa- tion. Teach Learn Med. Author details 1 representativeness, and disclosure of sensitive information: a randomized controlled trial. BMC Med Res Methodol. 2014;14(1):1–11. https://doi.org/ 10.1186/1471-2288-14-90. 1 Central Michigan University College of Medicine, PO Box 4181, Saginaw, MI 48606, USA. 2 Western Michigan University Homer Stryker MD School of Medi- cine, Kalamazoo, USA. 3 Wasit University College of Medicine, Kut, Iraq. 4 School of Medicine, Stanford University, Stanford, USA. 5 School of Medicine, University of Dundee, Dundee, United Kingdom. 19. Schoonenboom J, Johnson RB. How to Construct a Mixed Methods Research Design. Kolner Z Soz Sozpsychol. 2017;69(Suppl 2):107–31. https://doi.org/10.1007/s11577-017-0454-1. 20. Creswell, J (2013). What is mixed method research [video]. Retrieved from: https://www.youtube.com/watch?v=1OaNiTlpyX8 Received: 12 November 2021 Accepted: 4 April 2022 Received: 12 November 2021 Accepted: 4 April 2022 Received: 12 November 2021 Accepted: 4 April 2022 References 15. Al-Radhi AA-D. Information professionals in a GLOBALIZED WORLD: dis- tance learning/E-learning for IRAQ: concept and road map. Bull Am Soc Inform Sci Tech. 2008;34(3):34–7. https://doi.org/10.1002/bult.2008.17203 40311. 37. Scherer R, Siddiq F, Tondeur J. The technology acceptance model (TAM): A meta-analytic structural equation modeling approach to explaining teachers’ adoption of digital technology in education. Comput Educ. 2019;128:13–35. https://doi.org/10.1016/j.compedu.2018.09.009. 16. van der Vleuten C, Lindemann I, Schmidt L. Programmatic assessment: the process, rationale and evidence for modern evaluation approaches in medical education. Med J Aust. 2018;209(9):386–8. https://doi.org/10. 5694/mja17.00926. 38. Brown WJ. Entertainment-Education.Oxford Bibliographies. 2017 https:// doi.org/10.1093/OBO/9780199756841-0180 17. Ravitch SM, Carl NM. Qualitative Research: Bridging the Conceptual, Theoretical, and Methodological. Thousand Oaks, CA: SAGE Publications, Inc.; 2021. 39. Medina MS, Medina PJ, Wanzer DS, Wilson JE, Er N, Britton ML. Use of an audience response system (ARS) in a dual-campus classroom environ- ment. Am J Pharm Educ. 2008;72(2):38. https://doi.org/10.5688/aj720238. 18. Murdoch M, Simon AB, Polusny MA, et al. Impact of different pri- vacy conditions and incentives on survey response rate, participant Page 10 of 10 Tuma et al. BMC Medical Education (2022) 22:319 Tuma et al. BMC Medical Education (2022) 22:319 40. Scherer R, Siddiq F, Tondeur J. The technology acceptance model (TAM): a meta-analytic structural equation modeling approach to explaining teachers’ adoption of digital technology in education. Comput Educ. 2019;128:13–35. https://doi.org/10.1016/j.compedu.2018.09.009. 40. Scherer R, Siddiq F, Tondeur J. The technology acceptance model (TAM): a meta-analytic structural equation modeling approach to explaining teachers’ adoption of digital technology in education. Comput Educ. 2019;128:13–35. https://doi.org/10.1016/j.compedu.2018.09.009. 41. Sharma L, Srivastava M. Teachers’ motivation to adopt technology in higher education. J Appl Res High Educ. 2019;12(4):673–92. https://doi. org/10.1108/jarhe-07-2018-0156. 41. Sharma L, Srivastava M. Teachers’ motivation to adopt technology in higher education. J Appl Res High Educ. 2019;12(4):673–92. https://doi. org/10.1108/jarhe-07-2018-0156. 42. Gläsel A. 6 reasons why teachers don’t use tech in the classroom - what can EdTech companies learn? Medium. https://medium.com/the-edtech- world/resistance-tech-classrooms-981b86d862fc. Published July 20, 2018. Accessed February 14, 2021. 42. Gläsel A. 6 reasons why teachers don’t use tech in the classroom - what can EdTech companies learn? Medium. https://medium.com/the-edtech- world/resistance-tech-classrooms-981b86d862fc. Published July 20, 2018. Accessed February 14, 2021. y 43. Spickard A 3rd, Ahmed T, Lomis K, Johnson K, Miller B. Changing medical school it to support medical education transformation. Teach Learn Med. 2016;28(1):80–7. https://doi.org/10.1080/10401334.2015.1107488. Publisher’s Note S i N i Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from:
https://openalex.org/W4389214206	https://ojs.unm.ac.id/UJTV/article/download/34309/pdf	Indonesian	null	PENGARUH KINERJA GURU PRODUKTIF TERHADAP MINAT BELAJAR DAN KREATIVITAS SISWA PROGRAM KEAHLIAN TATA BUSANA DI SMK NEGERI 2 GOWA	UNM Journal of Technology and Vocational	2,022	cc-by-sa	2,582	UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 PENGARUH KINERJA GURU PRODUKTIF TERHADAP MINAT BELAJAR DAN KREATIVITAS SISWA PROGRAM KEAHLIAN TATA BUSANA DI SMK NEGERI 2 GOWA UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 PENDAHULUAN Perkembangan ilmu pengetahuan dan teknologi dalam era globalisasi seperti sekarang ini sangat mempengaruhi berbagai dimensi kehidupan manusia baik itu kehidupan sosial, ekonomi, pendidikan dan kebudayaan. Belajar merupakan proses dari pendidikan untuk meningkatkan sumber daya manusia, proses belajar pada umumnya dilaksanakan di sekolah oleh dua pihak yaitu guru dan siswa. hal ini berarti berhasil tidaknya pencapaian tujuan belajar bergantung pada bagaimana proses belajar mengajar yang dijalani oleh siswa sebagai peserta didik yang menjadi tanggung jawab guru sebagai pendidik. Menurut Srinalia (2015) Guru adalah unsur manusiawi dalam pendidikan. Dalam pelaksanaan fungsi dan tugasnya, guru menyandang persyaratan tertentu sebagaimana tertuang dalam Undang- Undang Republik Indonesia Nomor 20 tahun 2003 tentang Sistem Pendidikan Nasional. Menurut Supardi (2016) kinerja merupakan ungkapan dari kata performance dalam bahasa inggris yang memiliki tiga makna yaitu: prestasi, pertunjukan, dan pelaksanaan tugas. Keberhasilan pendidikan di sekolah sangat ditentukan oleh kinerja gurunya, seorang guru yang mempunyai kinerja tinggi seharusnya memiliki sikap positif terhadap pekerjaan yang menjadi tanggung jawabnya. Menurut Manullang (2017) kinerja guru adalah kemampuan yang ditunjukan oleh guru dalam melaksanakan tugas dan pekerjaannya, kinerja dikatakan baik atau memuaskan apabila tujuan yang dicapai sesuai dengan standar yang telah ditetapkan. Kinerja guru menyangkut seluruh aktivitas yang dilakukan oleh seorang guru dalam mengembangkan amanat dan tanggung jawabnya dalam mendidik, mengajar, membimbing, mengarahkan, dan memandu siswa untuk mencapai tingkat kedewasaan dan kematangan (Kompri, 2015). Kinerja g r memang mer pakan ses at Seorang guru diharapkan memiliki seperangkat pengetahuan dan keterampilan teknik mengajar serta menguasai bahan pelajaran yang akan disajikan kepada siswa. kemampuan mengajar menjadi sangat penting karena tanpa kemampuan mengajar yang baik guru tidak mampu melakukan inovasi atau kreasi dalam proses pembelajaran sehingga dapat mengurangi minat belajar bagi siswa, apabila guru tidak memiliki sikap profesional maka berakibat menurunkan tingkat kinerja guru itu sendiri. Kinerja guru yang kurang optimal dapat mempengaruhi minat belajar dari siswa. Menurut Widyaningsih (2018) mendefinisikan bahwa minat memiliki peran yang sangat penting dalam proses belajar karena minat menjadi salah satu faktor penentu dalam berhasil atau tidaknya tujuan pengajar yang akan dicapai. Minat besar pengaruhnya terhadap belajar, karena bila bahan pelajaran yang dipelajari tidak sesuai dengan minat siswa, siswa tidak akan belajar dengan sebaik-baiknya. Bahan pelajaran yang menarik minat siswa akan lebih mudah dipelajari oleh siswa. Menurut Slameto (2015) minat belajar merupakan hal yang sangat penting untuk diperhatikan dalam proses pembelajaran karena tanpa adanya minat belajar dari siswa maka proses pembelajaran tidak akan berlangsung secara maksimal. PENGARUH KINERJA GURU PRODUKTIF TERHADAP MINAT BELAJAR DAN KREATIVITAS SISWA PROGRAM KEAHLIAN TATA BUSANA DI SMK NEGERI 2 GOWA Devi Darnita1, Muhammad Andi Idkhan2, Aisyah3, 1,2,3, Universitas Negeri Makassar Abstrak: Penelitian ini bertujuan untuk mengetahui tingkat kinerja guru produktif, tingkat minat belajar, dan tingkat kreativitas serta pengaruh kinerja guru produktif terhadap minat belajar, dan pengaruh kinerja guru produktif terhadap kreativitas siswa. pada Program Studi Tata Busana SMKN 2 Gowa tahun ajaran 2020/2021. Jenis penelitian ini adalah penelitian kuantitatif, dan teknik pengambilan sampel yang digunakan adalah teknik proportional cluster random sampling dengan jumlah sampel 64 responden. Metode pengumpulan data menggunakan observasi, dokumentasi, dan angket. Hasil penelitian menunjukkan bahwa 56,3% kinerja guru dalam kategori tinggi, 68,8% siswa dalam minat belajar sedang, dan 57,8% siswa berada pada tingkat kreativitas sedang. Berdasarkan hasil penelitian, secara parsial kinerja guru berpengaruh signifikan terhadap minat belajar dengan persentase 28,2%, sedangkan kinerja guru tidak berpengaruh terhadap kreativitas siswa. Kata Kunci: Kinerja Guru, Minat Belajar, Kreativitas Abstract: The purpose of this study was to discover the level of productive teacher’s performance, level of learning interest, and level of creativity as well as the influence of productive teacher’s performance on learning interest, and the influence of productive teacher’s performance on students’ creativity in Fashion Design Program at SMKN 2 Gowa in academic year of 2020/2021. This type of research is quantitative research, and the sampling technique used is the proportional cluster random sampling technique with a sample of 64 respondents. Methods of data collection employed observation, documentation, and questionnaires. The results show that 56.3% of teacher's performance is in high category, 68,8% of students is in medium level of learning interest, and 57,8% of students is in medium level of creativity. Based on the results of the study, partially the teacher's performance has a significant influence on learning interest with a percentage of 28.2%, while the teacher's performance has no influence on students' creativity. Keywords: Teacher’s Performance, Learning Interest, Creativity 99 UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 kinerja dan individu, termasuk kinerja guru antara lain: 1) kemampuan, 2) motivasi, 3) dukungan yang diterima, 4) keberadaan pekerjaan yang mereka lakukan dan 5) hubungan mereka dengan organisasi. Berbagai upaya telah dilakukan dalam meningkatkan kinerja guru antara lain dengan peningkatan profesionalitas guru melalui pelatihan, seminar, kursus atau pendidikan formal yang tinggi serta pembinaan dan pengembangan untuk mendukung pembelajaran yang efektif. PENGARUH KINERJA GURU PRODUKTIF TERHADAP MINAT BELAJAR DAN KREATIVITAS SISWA PROGRAM KEAHLIAN TATA BUSANA DI SMK NEGERI 2 GOWA Dalam pelaksanaannya, kita tidak hanya menuntut keahlian dari para ahli pengembang kompetensi guru saja melainkan juga harus memperhatikan berbagai faktor yang mempengaruhi kinerja seorang guru. PENDAHULUAN Untuk membangkitkan minat belajar sisw banyak cara yang bisa digunakan antara lain dengan Kinerja guru memang merupakan sesuatu yang kompleks dan melibatkan berbagai faktor dengan itu, Mathis dan Jackson (2010) mengatakan banyak faktor yang mempengaruhi 100 UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 efisien. Berdasarkan hal tersebut penelitian ini bertujuan untuk mengetahui pengaruh kinerja guru produktif terhadap minat belajar dan kreativitas siswa program keahlian tata busana di SMK Negeri 2 Gowa. membuat materi yang akan dipelajari semenarik mungkin dan tidak membosankan, baik dari bentuk buku materi, desain pembelajaran yang membebaskan siswa untuk mengesplor apa yang dipelajari, melibatkan seluruh dominan belajar siswa (kognitif, afektif, psikomotor) sehingga siswa menjadi aktif maupun performansi guru yang menarik saat mengajar (Baharuddin, 2015). Dengan demikian disimpulkan minat belajar adalah dorongan dalam diri sendiri untuk melakukan sesuatu yang dapat membuatnya tertarik dan senang. METODE PENELITIAN Penelitian dilakukan menggunakan pendekatan kuantitatif selanjutnya data dipaparkan secara statistik deskriptif dan uji hipotesis menggunakan analisis regresi linear sederhana ini akan di ketahui pengaruh dari variabel independent terhadap variabel dependen menggunakan aplikasi Stastical Package for The Social Sciences (SPSS). Penelitian dilaksanakan di SMK Negeri 2 Gowa. Data dalam penelitian ini diambil melalui kuesioner penelitian. Selanjutnya, populasi penelitian adalah seluruh siswa kelas XI program keahlian tata busana di SMK Negeri 2 Gowa sejumlah 64 orang, sementara sampel ditentukan dengan teknik random sampling Selain itu, kinerja guru yang kurang optimal juga dapat mempengaruhi kreativitas siswa. Iklim pendidikan di negara kita saat ini belum banyak memberikan perhatian bagi pertumbuhan kreativitas anak bangsa. Padahal seiring perkembangan jaman, persaingan akan semakin meningkat. Ketatnya persaingan menuntut anak untuk lebih kreatif dan inovatif. Menurut Kristin (2016) kreativitas adalah kemampuan yang dimiliki seseorang untuk menghasilkan sesuatu yang baru dari ide, dan gagasan yang dimikinya sehingga yang dihasilkan tersebut memiliki daya guna. Menurut Slameto (2015) mengatakan bahwa kreativitas merupakan hasil belajar dalam kecakapan kognitif, sehingga untuk menjadi kreatif dapat dipelajari melalui proses belajar mengajar. Untuk itu para siswa harus dibimbing agar memiliki kemampuan kreativitas, mampu berpikir kritis, dan mampu memecahkan masalah. Karena melalui proses belajar tertentu diupayakan tercapainya tujuan-tujuan tersebut. Guru perlu menyediakan kondisi-kondisi belajar yang memungkinkan terjadinya penambahan aspek keluwesan, keaslian, dan kuantitas dari ablite kreativitas yang dimiliki oleh para siswa, dengan cara mengklasifikasikan jenis masalah yang akan disajikan kepada siswa, mengembangkan dan menggunakan keterampilan-keterampilan pemecahan masalah, ganjaran bagi prestasi belajar kreatif. Maka dari itu kinerja guru harus dikelola dengan baik dan dijaga agar tidak mengalami penurunan dan juga agar visi dan misi sekolah dapat tercapai dengan efetif serta Sumber: olah data 2021 minat belajar H2. Terdapat pengaruh kinerja guru terhadap kreativitas 0,05 0,192 0,027 Ho diterima H2 ditolak Tabel 1 menunjukkan hasil uji normalitas dengan nilai deviation from normality pada seluruh variabel lebih besar dari 0,05. Dengan demikian dapat disimpulkan bahwa seluruh data pada variabel berdistribusi normal. Selanjutnya hasil uji linearitas dapat dilihat pada tabel berikut: Tabel 2 Hasil Uji Linearitas Variabel α Sig. Deviatiom from linearity Keterangan Kinerja Guru + Minat Belajar 0,05 0,268 Linear Kinerja Guru + Kreativitas 0,05 0,159 Linear Sumber: olah data 2021 Tabel 2 Hasil Uji Linearitas Tabel 2 Hasil Uji Linearitas Tabel 2 Hasil Uji Linearitas Hipotesis pertama (H1) dalam penelitian ini adalah “terdapat pengaruh signifikan antara kinerja guru terhadap minat belajar siswa di SMK Negeri 2 Gowa”. Pada tabel 3 dapat diketahui nilai Sig. untuk variabel minat belajar adalah 0,000 > 0,05. Dengan demikian dapat disimpulkan bahwa H0 ditolak dan H1 dinyatakan diterima, artinya bahwa terdapat pengaruh signifikan kinerja guru terhadap minat belajar siswa di SMK Negeri 2 Gowa dengan nilai pengaruh sebesar 28,2%. Dengan kata lain kinerja mengajar guru yang baik dapat mempengaruhi minat belajar siswa lebih giat dan rajin. Begitu pula sebaliknya guru produktif yang memiliki kinerja mengajar yang kurang baik maka minat belajar siswa akan jadi menurun. Sumber: olah data 2021 Tabel 2 Hasil uji linearitas menunjukkan seluruh nilai Sig. Dev. From Linearity lebih besar dari nilai α (0,05). Dengan demikian dapat disimpulkan bahwa terdapat hubungan yang linear antara variabel kinerja guru terhadap minat belajar, dan terdapat hubungan yang linear antara variabel kinerja guru terhadap kreativitas. Setelah seluruh uji prasyarat analisis telah memenuhi syarat maka selanjutnya dilakukan uji hipotesis, menggunakan analisis regresi linear sederhana. Berikut adalah hasil uji regresi linear sederhana: Hasil penelitian ini mendukung dari hasil penelitian terdahulu yang dilakukan oleh Harnipa (2016) terkait kinerja guru menunjukan adanya pengaruh secara signifikan terhadap minat belajar siswa. Untuk meningkatkan kinerja guru yang baik dalam menumbuhkan minat belajar siswa diperlukan kerjasama dari semua komponen. Baik dan buruknya suatu kinerja mengajar guru terhadap siswa tentunya dipengaruhi oleh semua komponen. Setiap komponen harus memperbaiki kualitasnya masing-masing agar interaksi antara kinerja guru dan minat belajar siswa di dalam kelas menjadi meningkat. Tabel 3 Hasil Uji Regresi Linear Sederhana Hipotesis α Sig. R Square Ket H1.Terdapat pengaruh kinerja guru terhadap 0,05 0,000 0,282 Ho ditolak H1 diterima Tabel 3 Hasil Uji Regresi Linear Sederhana Minat belajar merupakan salah satu faktor yang mempengaruhi siswa dalam proses belajar mengajar. UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 minat belajar H2. Terdapat pengaruh kinerja guru terhadap kreativitas 0,05 0,192 0,027 Ho diterima H2 ditolak HASIL DAN PEMBAHASAN Hasil Penelitian Data penelitian diolah menggunakan SPSS untuk memperoleh jawaban rumusan masalah penelitian. Analisis data untuk uji hipotesis dilakukan menggunakan uji regresi linear sederhana. Sebelum uji analisis regresi linear sederhana, terlebih dahulu dilakukan uji prasyarat analisis yakni uji normalitas, dan uji linearitas. Berikut adalah hasil uji prasyarat analisis. Tabel 1 Hasil Uji Normalitas Variabel Variabel α Sig. Deviatio m from linearity Keteranga n Kinerja Guru (X) 0,0 5 0,696 Normal Minat Belajar (Y1) 0,0 5 0,148 Normal Kreativita s (Y2) 0,05 0,097 Normal Tabel 1 Hasil Uji Normalitas Variabel Tabel 1 Hasil Uji Normalitas Variabel 101 UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 KESIMPULAN Berdasarkan hasil analisis dan pembahasan, maka dapat ditarik kesimpulan sebagai berikut: 1) Terdapat pengaruh signifikan kinerja guru terhadap minat belajar siswa program keahlian tata busana di SMK Negeri 2 Gowa dengan nilai pengaruh sebesar 28,2%. 2) tidak terdapat pengaruh signifikan kinerja guru terhadap kreativitas siswa program keahlian tata busana di SMK Negeri 2 Gowa. Hal ini disebabkan oleh faktor lain yakni fasilitas sekolah yang kurang memadai, sehingga pembelajaran siswa menjadi kurang efektif. Dengan demikian diharapkan sekolah dapat memperhatikan ketersediaan fasilitas siswa dalam belajar. Hipotesis kedua (H2) dalam penelitian ini adalah “terdapat pengaruh signifikan antara kinerja guru terhadap kreativitas siswa di SMK Negeri 2 Gowa”. Pada tabel 4.12 dapat diketahui nilai Sig. untuk variabel minat belajar adalah 0,192 > 0,05. Dengan demikian dapat disimpulkan bahwa H0 diterima dan H1 dinyatakan ditolak, artinya bahwa tidak terdapat pengaruh signifikan kinerja guru terhadap kreativitas siswa di SMK Negeri 2 Gowa. Hasil penelitian ini memiliki perbedaan dengan penelitian sebelumnya oleh Moh. Rizal (2016) hasil menunjukan bahwa kinerja guru berpengaruh signifikan terhadap kreativitas siswa sebesar 35,60%. 3. Profil belajar siswa Sebagai guru, kita semua tentu tahu bahwa siswa akan menunjukkan kinerja yang lebih baik jika tugas-tugas yang diberikan sesuai dengan keterampilan dan pemahaman yang mereka miliki sebelumnya (kesiapan belajar). Lalu jika tugas-tugas tersebut memicu keingintahuan atau hasrat dalam diri seorang siswa (minat), dan jika tugas itu memberikan kesempatan bagi mereka untuk bekerja dengan cara yang mereka sukai (profil belajar). Sumber: olah data 2021 Menumbuhkan minat belajar pada siswa dengan tujuan agar siswa lebih cepat dan 102 UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 ketersediaan fasilitas sekolah. Kurangnya fasilitas yang tersedia disekolah dapat mempengaruhi kreativitas siswa dalam belajar, fasilitas menjadi salah satu penunjang dalam proses pembelajaran karena adanya fasilitas yang mendukung dapat memberikan kemudahan bagi siswa. Berdasarkan hasil wawancara dengan guru produktif tata busana di SMK Negeri 2 Gowa fasilitas mesin bordir disekolah tidak memadai dikarenakan jumlahnya hanya 1 sedangkan siswa berjumlah 35 orang sehingga pembelajaran menjadi kurang efektif. Meskipun tingkat kreativitas siswa tinggi jika fasilitas sekolah yang kurang memadai sehingga akan memberikan pengaruh signifikan terhadap kreativitas siswa. Dengan demikian, diharapkan sekolah dapat memperhatikan ketersedian fasilitas siswa dalam belajar. mudah dalam menerima pelajaran dengan cara memberikan motivasi yang baik pada diri anak didik, sehingga rela belajar tanpa adanya keterpaksaan, dan memberikan kesempatan pada anak didik untuk belajar dengan lingkungan belajar yang kondusif dan kreatif. Berkaitan dengan itu, salahsatu hal yang penting dilakukan oleh guru adalah melakukan pemetaan kebutuhan belajar siswa. Seiring dengan itu, Tomlinson (dalam Kusuma dan Luthfah, 2020) menyampaikan bahwa kita dapat mengkategorikan kebutuhan belajar murid, paling tidak berdasarkan tiga aspek. Ketiga aspek tersebut adalah: 1. Kesiapan belajar (readiness) siswa 2. Minat siswa 3. Profil belajar siswa DAFTAR PUSTAKA Baharuddin. (2015). Teori Belajar dan Pembelajaran. Yogyakarta: Ar-Rumzz Media. Depdiknas. (2003). Undang-undang RI No.20 Tahun 2003 Tentang Sistem Pendidikan Nasional. Tidak terdapat pengaruh signifikan kinerja guru terhadap kreativitas siswa. Hal ini menunjukan kreativitas dapat diipengaruhi oleh faktor lain. Menurut Munandar (dalam Ali & Asrori 2016) terdapat sejumlah faktor yang dapat mempengaruhi kreativitas yakni Kompri. (2015). Manajemen Pendidikan. In Komponen-komponen elementer 103 UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 kemajuan sekolah. Yogyakarta: Ar- Ruzz Media. UNM of Journal Technologycal and Vocational Vol. 6 No. 1, Februari 2022 ISSN e: 2580-0434; p: 2580-0418 Kristin, Kristin, F. (2018). Penerapan Model Pembelajaran Active Learning Untuk Meningkatkan Kreativitas Belajar Mahasiswa Pada Mata Kuliah Konsep Dasar I IPS. Jurnal Pendidikan Edutama, 3(2), 9-19. Manullang. (2017). Pengaruh Kinerja Guru Terhadap Hasil Belajar Siswa SMP Negeri 4 Tanjung Jambung Timur Kota Jambi. Jambi: Jurnal Ilmiah Universitas Batanghari Jambi. Volume 17 Nomor 3 Tahun 2017 Mathis dan Jackson, RL. (2010). Manajemen Sumber Daya Manusia. Penterjemah Jimmy Sadeli dan Bayu Prawira Hie. Jakarta: Penerbit PT. Salemba Empat. kemajuan sekolah. Yogyakarta: Ar- Ruzz Media. kemajuan sekolah. Yogyakarta: Ar- Ruzz Media. Slameto. (2015). Belajar dan faktor-faktor yang mempengaruhi. Jakarta: Rineka Cipta. Slameto. (2015). Belajar dan faktor-faktor yang mempengaruhi. Jakarta: Rineka Cipta. Srinalia. (2015). Faktor-faktor Penyebab Rendahnya Kinerja Guru dan Korelasinya terhadap Pembinaan Siswa: Studi kasus di SMAN 1 darul Imarah Aceh Besar. Jurnal Ilmiah Didaktita. Vol. 15 No. 2 Supardi. (2016). Kinerja Guru. Jakarta: Rajawali Press. Widyaningsih, O. (2018). Survey Minat Mahasiswa Mengajar Di Daerah Terpencil Di STKIP Kusuma Negara Jakarta Tahun 2017. Jurnal Edukasi Gemilang, 15-19. 104
https://openalex.org/W3159398451	https://figshare.com/articles/journal_contribution/Synthesis_Characterization_and_First_Application_of_Graphene_Oxide_Functionalized_Cu_II_Complex_for_the_Synthesis_of_1_2_3-Triazole_Derivatives/14497889/1/files/27771122.pdf	English	null	Synthesis, Characterization and First Application of Graphene Oxide Functionalized Cu(II) Complex for the Synthesis of 1,2,3-Triazole Derivatives	Polycyclic aromatic compounds	2,021	cc-by	445	Supplementary data Synthesis, characterization and first application of graphene oxide functionalized Cu(II) complex for the synthesis of 1,2,3-triazole derivatives Melika Eftekhar and Farveh Raoufi Table of Content Page Nitrogen adsorption–desorption isotherms Fig. S1 and Fig. S2 2-3 1H-NMR, 13C-NMR and FT-IR Characterization 4-18 Supplementary data Synthesis, characterization and first application of graphene oxide functionalized Cu(II) complex for the synthesis of 1,2,3-triazole derivatives Melika Eftekhar and Farveh Raoufi Table of Content Page Nitrogen adsorption–desorption isotherms Fig. S1 and Fig. S2 2-3 1H-NMR, 13C-NMR and FT-IR Characterization 4-18 Table of Content Fig. S1 Nitrogen adsorption–desorption isotherms of GO Fig. S1 Nitrogen adsorption–desorption isotherms of GO Fig. S1 Nitrogen adsorption–desorption isotherms of GO Fig. S2 Nitrogen adsorption–desorption isotherms of GO-M-Cu Fig. S2 Nitrogen adsorption–desorption isotherms of GO-M-Cu Fig. S3 FT-IR of 1-Benzyl-4-phenyl-1H-1,2,3-triazole (5a, 5b and 5c, Table 3, Entries 1-3) Fig. S3 FT-IR of 1-Benzyl-4-phenyl-1H-1,2,3-triazole (5a, 5b and 5c, Table 3, Entries 1-3) Fig. S4 1H-NMR of 1-Benzyl-4-phenyl-1H-1,2,3-triazole (5a, 5b and 5c, Table 3, Entries 1-3) Fig. S4 1H-NMR of 1-Benzyl-4-phenyl-1H-1,2,3-triazole (5a, 5b and 5c, Table 3, Entries 1-3) Fig. S5 13C-NMR of 1-Benzyl-4-phenyl-1H-1,2,3-triazole (5a, 5b and 5c, Table 3, Entries 1-3) Fig. S5 13C-NMR of 1-Benzyl-4-phenyl-1H-1,2,3-triazole (5a, 5b and 5c, Table 3, Entries 1-3) Fig. S6 FT-IR of 1-(4-chlorobenzyl)-4-phenyl-1H-1,2,3-triazole (5d, 5e, Table 3, Entries 4,5) Fig. S6 FT-IR of 1-(4-chlorobenzyl)-4-phenyl-1H-1,2,3-triazole (5d, 5e, Table 3, Entries 4,5) Fig. S7 1H-NMR of 1-(4-chlorobenzyl)-4-phenyl-1H-1,2,3-triazole (5d, 5e, Table 3, Entries 4,5) Fig. S7 1H-NMR of 1-(4-chlorobenzyl)-4-phenyl-1H-1,2,3-triazole (5d, 5e, Table 3, Entries 4,5) Fig. S8 FT-IR of 1-(4-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5f, Table 3, Entry 6) Fig. S8 FT-IR of 1-(4-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5f, Table 3, Entry 6) Fig. S9 1H-NMR of 1-(4-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5f, Table 3, Entry 6) Fig. S9 1H-NMR of 1-(4-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5f, Table 3, Entry 6) Fig. S10 13C-NMR of 1-(4-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5f, Table 3, Entry 6) Fig. S10 13C-NMR of 1-(4-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5f, Table 3, Entry 6) Fig. S11 1H-NMR of 1-(4-bromobenzyl)-4-phenyl-1H-1,2,3-triazole (5g, Table 3, Entry 7) Fig. S11 1H-NMR of 1-(4-bromobenzyl)-4-phenyl-1H-1,2,3-triazole (5g, Table 3, Entry 7) Fig. S12 1H-NMR of 1-(3-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5h, Table 3, Entry 8) 2 1H-NMR of 1-(3-nitrobenzyl)-4-phenyl-1H-1,2,3-triazole (5h, Table 3, Entry 8) Fig. S13 FT-IR of 1,4-diphenyl-1H-1,2,3-triazole (6a, Table 3, Entry 9) Fig. S13 FT-IR of 1,4-diphenyl-1H-1,2,3-triazole (6a, Table 3, Entry 9) Fig. S14 1H-NMR of 1,4-diphenyl-1H-1,2,3-triazole (6a, Table 3, Entry 9) Fig. S14 1H-NMR of 1,4-diphenyl-1H-1,2,3-triazole (6a, Table 3, Entry 9) Fig. S15 FT-IR of 1-(4-bromophenyl)-4-phenyl-1H-1,2,3-triazole (6c, Table 3, Entry 11) Fig. S15 FT-IR of 1-(4-bromophenyl)-4-phenyl-1H-1,2,3-triazole (6c, Table 3, Entry 11) Fig. S16 FT-IR of (4-(4-nitrophenyl)-1H-1,2,3-triazol-1-yl)methanol (6g, Table 3, Entry 15) Fig. S16 FT-IR of (4-(4-nitrophenyl)-1H-1,2,3-triazol-1-yl)methanol (6g, Table 3, Entry 15) Fig. S17 FT-IR of (4-(4-bromophenyl)-1H-1,2,3-triazol-1-yl)methanol (6h, Table 3, Entry 16) Fig. S17 FT-IR of (4-(4-bromophenyl)-1H-1,2,3-triazol-1-yl)methanol (6h, Table 3, Entry 16)
https://openalex.org/W2935252922	https://www.nature.com/articles/s41598-019-42079-z.pdf	English	null	Serum of patients with acute myocardial infarction prevents inflammation in iPSC-cardiomyocytes	Scientific reports	2,019	cc-by	9,544	Serum of patients with acute myocardial infarction prevents inflammation in iPSC- cardiomyocytes Received: 11 October 2017 Accepted: 20 March 2019 Published: xx xx xxxx Received: 11 October 2017 Accepted: 20 March 2019 Published: xx xx xxxx Katherine Sattler1, Ibrahim El-Battrawy1,2, Zhihan Zhao1,2, Christoph Schrottenberg1, Gökhan Yücel1,2, Huan Lan1,2, Xin Li1, Siegfried Lang1,2, Wolfram-Hubertus Zimmermann3,2, Lukas Cyganek2,4, Jochen Utikal 5, Thomas Wieland 2,6, Ursula Ravens7, Karen Bieback8, Martin Borggrefe1,2, Xiaobo Zhou1,2,9 & Ibrahim Akin1,2 Acute myocardial infarction (MI) evokes a systemic inflammatory response and locally the degradation of the necrotic tissue, followed by scar formation. The mechanisms for containment of the infarct zone are not studied well. The study aimed to examine the response of healthy cardiomyocytes to serum of patients with myocardial infarction. Human iPSC-cardiomyocytes (iPSC-CM) generated from two healthy donors were incubated with serum of patients with MI with and without ventricular fibrillation (VF) or of healthy controls. Different cell adhesion molecules were studied by flow cytometry and immunostaining. Cellular electrophysiology was studied by patch clamp. The cell adhesion molecules CD54/ICAM-1, CD58/LFA-3 and CD321/JAM-A were expressed on iPSC-CM within the plasma membrane. Incubation with serum of MI patients reduced the levels of expression of CD54/ICAM-1 and CD321/JAM-A by 15–20%. VF serum was less effective than serum of MI patients without VF. MI serum or VF serum did not affect resting potential, action potential duration or maximum depolarization velocity. Myocardial infarction serum exerts anti-inflammatory effects on healthy cardiomyocytes without affecting their electrical activity, thus helping to contain the infarct zone and to protect healthy tissue. Ventricular fibrillation during MI drives healthy cardiomyocytes towards a pro-inflammatory phenotype. Myocardial infarction (MI) is a primarily local event which leads to the activation of an acute systemic inflam- matory response. This is mirrored by a systemic increase of acute phase proteins, pro-inflammatory mediators, recruitment of inflammatory cells towards the myocardium and stem cell mobilization1,2, which ultimately ini- tiates the stabilization of the infarcted area3. One of the first steps of this process is the adhesion of myeloid cells and macrophages to the infarct area to clear away the necrotic tissue4, followed by initiation of scar formation3. 1First Department of Medicine, Faculty of Medicine, University Medical Centre Mannheim (UMM), University of Heidelberg, Mannheim, Germany. 2DZHK (German Center for Cardiovascular Research), Partner Site Heidelberg- Mannheim and Göttingen, Göttingen, Germany. 3Institute of Pharmacology and Toxicology, University of Göttingen, Göttingen, Germany. 4Stem Cell Unit, Clinic for Cardiology and Pneumology, University Medical Center Göttingen, Göttingen, Germany. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports Received: 11 October 2017 Accepted: 20 March 2019 Published: xx xx xxxx www.nature.com/scientificreports/ www.nature.com/scientificreports/ p p Age [years] 62 (38–75) Male 7 (87.5) Clinical course Cardiopulmonary resuscitation 4 (50.0) Ventricular fibrillation requiring defibrillation 3 (37.5) Cardiogenic shock 2 (25.0) Impella 2 (25.0) ECMO 2 (25.0) 1-vessel disease 0 (0.0) 2-vessel disease 1 (12.5) 3-vessel disease 7 (87.5) Status post CABG 1 (12.5) Primary PCI, target vessel LAD 6 (75.0) RCX 4 (50.0) RCA 3 (37.5) Bypass 1 (12.5) Number of implanted stents 3 (1–7) abciximab 2 (25.0) Blood chemistry Creatinine [mg/dl] 1.17 (0.5–3.4) Creatinephosphokinase [U/l] 987.0 (116.0– 2816.0) Troponin I [µg/l] 9.53 (0.5–144.9) Cholesterol [mg/dl] 180.5 (75.0– 206.0) Triglycerides [mg/dl] 155.0 (77.0– 262.0) Leucocyte count [10E9/l] 12.5 (4.1–23.6) C-reactive protein [mg/dl] 5.8 (2.9–26.0) Medication prior to event Betablockers 3 (37.5) ACE-inhibitors/ARBs 2 (25.0) Calcium antagonist 1 (12.5) CSE-inhibitors 2 (25.0) Acetylsalicylate 1 (12.5) Table 1. Demographic and clinical characteristics of the patient group. Continuous variables are shown as median (min-max). Quantitative variables are shown as number [%]. In several patients, more than 1 vessel was treated during PCI. Information about medication prior to event is available for 6 of 8 patients. Creatinphosphokinase levels are based on information from 7 of 8 patients. ACE, angiotensin converting enzyme; ARB, aldosterone receptor blocker; CABG, coronary artery bypass graft; CSE, cholesterol synthesis enzyme; ECMO, extracorporeal membrane oxygenation; LAD, left anterior descending; NSTEMI, non-ST- segment elevation myocardial infarction; PCI, percutaneous coronary intervention; RCA, right coronary artery; RCX, ramus circumflexus; STEMI, ST-segment elevation myocardial infarction. Age [years] 62 (38–75) Male 7 (87.5) Clinical course Cardiopulmonary resuscitation 4 (50.0) Ventricular fibrillation requiring defibrillation 3 (37.5) Cardiogenic shock 2 (25.0) Impella 2 (25.0) ECMO 2 (25.0) 1-vessel disease 0 (0.0) 2-vessel disease 1 (12.5) 3-vessel disease 7 (87.5) Status post CABG 1 (12.5) Primary PCI, target vessel LAD 6 (75.0) RCX 4 (50.0) RCA 3 (37.5) Bypass 1 (12.5) Number of implanted stents 3 (1–7) abciximab 2 (25.0) Blood chemistry Creatinine [mg/dl] 1.17 (0.5–3.4) Creatinephosphokinase [U/l] 987.0 (116.0– 2816.0) Troponin I [µg/l] 9.53 (0.5–144.9) Cholesterol [mg/dl] 180.5 (75.0– 206.0) Triglycerides [mg/dl] 155.0 (77.0– 262.0) Leucocyte count [10E9/l] 12.5 (4.1–23.6) C-reactive protein [mg/dl] 5.8 (2.9–26.0) Medication prior to event Betablockers 3 (37.5) ACE-inhibitors/ARBs 2 (25.0) Calcium antagonist 1 (12.5) CSE-inhibitors 2 (25.0) Acetylsalicylate 1 (12.5) Table 1. Demographic and clinical characteristics of the patient group. Continuous variables are shown as median (min-max). Quantitative variables are shown as number [%]. In several patients, more than 1 vessel was treated during PCI. www.nature.com/scientificreports/ Information about medication prior to event is available for 6 of 8 patients. Creatinphosphokinase levels are based on information from 7 of 8 patients. ACE, angiotensin converting enzyme; ARB, aldosterone receptor blocker; CABG, coronary artery bypass graft; CSE, cholesterol synthesis enzyme; ECMO, extracorporeal membrane oxygenation; LAD, left anterior descending; NSTEMI, non-ST- segment elevation myocardial infarction; PCI, percutaneous coronary intervention; RCA, right coronary artery; RCX, ramus circumflexus; STEMI, ST-segment elevation myocardial infarction. For inflammatory response, distant organs producing inflammatory cells such as bone marrow or spleen are “activated” by myocardial infarction5. However, the systemic response can spread the inflammation to organs that are not primarily involved in the inflammatory defense such as carotid and aortic tissue, as demonstrated recently2. Transfer of inflammation to different organs is carried out by cytokines, a mechanism demonstrated for other disease conditions as well6. Although there is extensive knowledge of the inflammatory systemic and local responses after myocardial infarction, the mechanisms for containment of the infarct zone are not clear yet3. Therefore, in the current study we examined the response of healthy cardiomyocytes to serum of patients with myocardial infarction. Due to the limited availability of human adult ventricular cardiomyocytes, we chose induced pluripotent stem cell (iPSC)-derived cardiomyocytes of healthy human donors. These cells show proper inflammatory response upon pro-inflammatory stimuli, as demonstrated by our group recently7. Serum of patients with acute myocardial infarction prevents inflammation in iPSC- cardiomyocytes 5Skin Cancer Unit, German Cancer Research Center (DKFZ), Heidelberg and Department of Dermatology, Venereology and Allergology, University Medical Center Mannheim, University of Heidelberg, Mannheim, Germany. 6Institute of Experimental and Clinical Pharmacology and Toxicology, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany. 7Institue of Experimental Cardiovascular Medicine, University Heart Centre Freiburg, Bad Krozingen, Freiburg, Germany. 8Institute for Transfusion Medicine and Immunology, Mannheim, Germany. 9Key Laboratory of Medical Electrophysiology of the Ministry of Education, Collaborative Innovation Center for Prevention and Treatment of Cardiovascular Disease, Institute of Cardiovascular Research, Southwest Medical University, Luzhou, Sichuan, China. Katherine Sattler and Ibrahim El-Battrawy contributed equally. Correspondence and requests for materials should be addressed to I.E.-B. (email: Ibrahim.El- Battrawy@umm.de) 1 xpression of cell adhesion molecules on human iPSC-cardiomyocytes. For monitoring the dif- Expression of cell adhesion molecules on human iPSC-cardiomyocytes. For monitoring the dif- ferentiation process of hiPSC to cardiomyocytes, we performed quantitative real-time PCR of the mRNA expres- sion of the pluripotency gene POU5F1 and the cardiomyocyte marker troponin T (TNNT2). Indeed, while the change of the expression of POU5F1 was decreased tremendously to a minimal amount at day 20 of differentia- tion relative to the house keeping gene GAPDH, the expression of mRNA of TNNT2 increased during the first 20 days of differentiation (Fig. 1A).f yf g We then tested whether hiPSC-cardiomyocytes expressed adhesion molecules during their differentiation process. At day 25, CD 31, CD 162, CD 62 and CD 102 were expressed by <2% of the cells and were thus defined as “negative” (Fig. 1B,C), while the expression of CD54/ICAM-1, CD58/LFA-3 and CD321/JAM-A was found on >2% of hiPSC-cardiomyocytes and was thus termed positive (Fig. 1B). For the three molecules CD54/ICAM-1, CD58/LFA-3 and CD321/JAM-A, a dynamic expression over time was found with stable expression at day 25 (Fig. 2A,B). At day 25, the percentage of cardiomyocyte positive for the adhesion molecules relative to all cardi- omyocytes (as defined by positive expression of troponin I [TNNT2]) was 25.5%, 26.8% and 69.6% for CD54/ ICAM-1, CD58/LFA-3 and CD321/JAM-A, respectively (Fig. 2C). Immunofluorescence showed the expression of all three adhesion molecules in iPSC-cardiomyocytes as well. Apart from cardiomyocytes, other cells of the cell culture expressed the markers as well as is demonstrated in the staining of CD58, leading to the impression of heavy background artefacts. However, localization in cardiomyocytes was demonstrated by positive doublestain- ing for the cardio-specific marker troponin I (TNNT2, Fig. 3). Serum of myocardial infarction alters number of stimulated cardiomyocytes. Stimulation with serum reduced the percentage of cardiomyocytes expressing the respective cell adhesion molecule when com- pared to unstimulated cells. However, different changes were observed regarding effect of myocardial infarction vs. control serum. Serum of patients with myocardial infarction (MI) exerted the strongest effect on cells expressing CD58/LFA- 3. When compared to the incubation with control serum, 5% (serum/medium, v/v) MI serum reduced positive cells by 40% (p = 0.02), and incubation with 20% MI serum reduced cells by 49% (p < 0.01). For CD54/ICAM-1, an increased number of cells was observed when stimulation was performed with 20% MI serum (p < 0.01 vs. control serum), while 5% MI serum did not change cell numbers. Results P i Patients’ characteristics. In the current study, effects of serum of eight patients presenting with acute myocardial infarction were compared to the effects of serum of three healthy volunteers. The demographic and clinical characteristics of the patient group are shown in Tables 1 and 2. Due to the acuteness of the disease, blood drawings were undertaken after restoration of blood flow by primary percutaneous coronary intervention, thus, Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 2 www.nature.com/scientificreports/ Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8 Age 67 76 39 65 70 49 51 61 Sex Male Male Male Female Male Male Male Male Immediate diagnosis STEMI NSTEMI STEMI STEMI NSTEMI STEMI STEMI STEMI Ventricular fibrillation Yes Yes Yes Yes No No No No Troponin I [µg/l] 5.48 5.64 144.9 1.3 0.5 13.4 76.3 87.2 CK [U/l] 455 na 2321 182 116 987 945 2816 CPR during hospital stay Yes Yes Yes Yes No No No No Diabetes mellitus type II Yes Yes No No Yes No No No Arterial hypertension Yes Yes Yes Yes Yes Yes No Yes CAD No Yes No No Yes Yes Yes Yes Table 2. Case-wise presentation of patients’ characteristics. CAD, previously known coronary artery disease; CK, Creatinphosphokinase; CPR, cardiopulmonary resuscitation; NSTEMI, non-ST-segment-elevation myocardial infarction; STEMI, ST-segment-elevation myocardial infarction. Table 2. Case-wise presentation of patients’ characteristics. CAD, previously known coronary artery disease; CK, Creatinphosphokinase; CPR, cardiopulmonary resuscitation; NSTEMI, non-ST-segment-elevation myocardial infarction; STEMI, ST-segment-elevation myocardial infarction. at the time of blood collection all patients had received acetylsalicylate and weight-adjusted heparin, as well as a second platelet aggregation inhibitor. Serum of patients with myocardial infarction is pro-inflammatory. Interleukin-6 (Il-6) was meas- ured in seven serum samples of the patient group (due to lack of material of one patient) and in the serum sam- ples of all three healthy controls. As expected, serum of patients contained higher levels of Il-6 (median 50.71 pg/ ml, range 12.28–1339.65 pg/ml) than serum of healthy controls (median 4.38 pg/ml, range 2.08–16.15 pg/ml, p = 0.03). xpression of cell adhesion molecules on human iPSC-cardiomyocytes. For monitoring the dif- For CD321/JAM-A, cell number did not change upon incubation with MI serum compared to control serum. Table 3 shows the results in detail. Serum of myocardial infarction affects levels of cell adhesion molecule expression. In addition to the effects on cell numbers, incubation of cardiomyocytes with serum of patients with myocardial infarction (MI) for 48 hours resulted in changes of the expression levels of the different cell adhesion molecules, as expressed by the median fluorescence intensity (MIF) of each marker. Expression of CD58/LFA-3 was significantly reduced by approximately 15% by incubation with MI serum, irrespective of serum concentration. 5% MI serum reduced the levels of CD54/ICAM-1 by 20%, while 20% MI serum reduced the levels of CD321/JAM-A to 85%. Table 4 gives the exact data of the values of the median fluorescence intensity of each group. l To account for changes in cells numbers, the MIF values were normalized to the numbers of cardiomyocytes (defined as cells positive for TNNT2) expressing the respective molecule. CD58/LFA-3 expression levels normal- ized to cell numbers were not changed by addition to MI serum compared to control serum (p > 0.05, Fig. 4A). In contrast, 20% MI serum reduced per cell-expression of CD54/ICAM-1 when compared to control serum Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 3 www.nature.com/scientificreports/ Figure 1. (A). mRNA expression of the pluripotency gene POU5F1 and of the cardiac marker troponin I (TNNT2) during differentiation of hiPSC into cardiomyocytes. The fold change over time relative to the expression of the housekeeping gene GAPDH is presented. Mean of 3 technical replicates of 3–6 biological replicates of 3 different differentiation rounds is shown. Whiskers display standard error. (B). Percentage of marker-expressing iPSC-cardiomyocytes (% of TNNT2-positive cells) at day 25 of differentiation. Mean of n = 5–12 measurements per marker. Whiskers display standard deviation. (C). CD162, CD31, CD62, and CD102 (blue) were not found on iPSC-cardiomyocytes. Figure 1. (A). mRNA expression of the pluripotency gene POU5F1 and of the cardiac marker tropo Figure 1. (A). mRNA expression of the pluripotency gene POU5F1 and of the cardiac marker troponin I (TNNT2) during differentiation of hiPSC into cardiomyocytes. The fold change over time relative to the expression of the housekeeping gene GAPDH is presented. Mean of 3 technical replicates of 3–6 biological replicates of 3 different differentiation rounds is shown. Whiskers display standard error. (B). (p < 0.01, Fig. 4B). Similarly, CD321/JAM-A expression per cell was reduced by MI serum of the same concen- tration (p = 0.03, Fig. 4C). (p < 0.01, Fig. 4B). Similarly, CD321/JAM-A expression per cell was reduced by MI serum of the same concen- tration (p = 0.03, Fig. 4C). Serum of patients with ventricular fibrillation induces higher cell adhesion molecule expres- sion. Ventricular fibrillation (VF) is a complication which can emerge during myocardial infarction. We thus tested whether serum of patients who had developed VF had different effects than serum of patients with myo- cardial infarction without ventricular fibrillation. VF serum contained higher levels of Il-6 (n = 3; 622 pg/ml [38.4–1339.6] than MI serum without VF (n = 4; 43.2 pg/ml [12.3–71.3]). As expected, if VF serum was used, significantly higher expression of all three cell adhesion molecules was observed compared to serum of patients without VF (Fig. 5). Serum of patients with MI does not affect cellular electrophysiology. To evaluate the effect of serum on cellular electrophysiology of cardiomyocytes, hiPSC-CMs were incubated with serum of MI patients with or without VF or of controls for 48 hours. Treating cardiomyocytes with either serum had no significant effects on resting potential (RP), action potential amplitude (APA) or duration (APD), or maximum depolariza- tion velocity (Vmax), as shown in Figs 6 and 7. xpression of cell adhesion molecules on human iPSC-cardiomyocytes. For monitoring the dif- Percentage of marker-expressing iPSC-cardiomyocytes (% of TNNT2-positive cells) at day 25 of differentiation. Mean of n = 5–12 measurements per marker. Whiskers display standard deviation. (C). CD162, CD31, CD62, and CD102 (blue) were not found on iPSC-cardiomyocytes. Figure 1. (A). mRNA expression of the pluripotency gene POU5F1 and of the cardiac marker troponin I (TNNT2) during differentiation of hiPSC into cardiomyocytes. The fold change over time relative to the expression of the housekeeping gene GAPDH is presented. Mean of 3 technical replicates of 3–6 biological replicates of 3 different differentiation rounds is shown. Whiskers display standard error. (B). Percentage of marker-expressing iPSC-cardiomyocytes (% of TNNT2-positive cells) at day 25 of differentiation. Mean of n = 5–12 measurements per marker. Whiskers display standard deviation. (C). CD162, CD31, CD62, and CD102 (blue) were not found on iPSC-cardiomyocytes. Discussion Representative plots of flow cytometry for CD58/LFA-3, CD54/ICAM-1 and CD321/JAM-A (blue) showing the marker-positive sub-populations at differentiation day 25. The number of each marker subset gives the percentage of positive cells of all cells. (C). Percentage of iPSC-cardiomyocytes (% of TNNT2-positive cells) expressing CD58/LFA-3, CD54/ICAM-1 and CD321/JAM-A at different times of differentiation. N = 5–10 measurements per marker. The mean ± standard deviation is shown. Acute myocardial infarction stimulates a massive inflammatory response. The necrotic tissue releases proteins summarized by the term danger-associated molecular patterns (DAMP), which bind to Toll-like receptors and start the inflammatory reaction3,9. Chemokines and cytokines are secreted and start invasion of the infarcted area by neutrophils, monocytes and lymphocytes9. In the past, several animal studies have shown that non-infarcted myocardium, contrary to the infarcted tissue, does not express pro-inflammatory cytokines10–12. One reason might be the fact that in these models the perfusion of the non-infarcted myocardium is intact. However, it is unclear so far how healthy myocardium responds to the pro-inflammatory environment presenting after myo- cardial infarction. For sepsis, spreading of inflammation and cell damage by contact with septic serum has been shown recently in an ex-vivo model6, and myocardial cells showed dysfunction after incubation with plasma of patients in septic shock13. Interestingly, Youker et al. were able to transfer myocardial infarction-associated inflammation to isolated canine cardiomyocytes by incubating the cells with lymph collected directly from the area of ischemic canine myocardial tissue, thereby stimulating ICAM-1 expression as monitored by neutrophil adhesion, an observation attributed to Il-6 content of the lymph as it was deleted by Il-6-antibody14. Differences of this study to ours might be caused by the fact that Youker et al. used lymph fluid drained directly from the inflamed myocardium which might contain higher levels of pro-inflammatory mediators than serum does, and which might contain especially pro-inflammatory lipoid structures. In contrast, our model of using serum is more comparable to humans with myocardial infarction.ffl p y We decided to study three different adhesion molecules having different functions during inflammatory response. By immunostaining, we demonstrated the localization of the molecules in question in the plasma mem- brane of cardiomyocytes, as proven by co-staining for troponin T15. LFA-3 (CD58) is a ligand of the T lymphocyte CD2 protein and mediates adhesion and activation of T lymphocytes. The protein is associated with autoimmune diseases such as multiple sclerosis16, and with tumor growth17. Discussion In the current study, we found for the first time the stable expression of the cell adhesion molecules CD54/ICAM- 1, CD58/LFA-3 and CD321/JAM-A on iPSC-cardiomyocytes, and a reduction of the expression levels of these cell adhesion molecules in hiPSC-CMs induced by serum from patients with myocardial infarction. For studies on cardiac functions, hiPSC-CMs have important advantages over heterologous expression sys- tems such as Xenopus oocytes, human embryonic kidney (HEK) cells and Chinese Hamster Ovary (CHO) cells, which lack important constituents of cardiac ion channel macromolecular complexes that might be necessary for normal electrophysiological characteristics. Animals possess cardiac electrophysiological properties crucially different from that in humans, rendering animal cardiomyocytes not ideal for studies on human physiology and diseases. Therefore, hiPSC-CMs are a good alternative for modeling cardiac diseases and studying drug effects or mechanisms8. Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 4 www.nature.com/scientificreports/ p p Figure 2. (A). Expression levels (median fluorescence intensity) of CD58/LFA-3, CD54/ICAM-1 and CD321/ JAM-A on iPSC-cardiomyocytes (TNNT2-positive cells) at different times of differentiation. N = 5–10 measurements per marker. The mean ± standard deviation is shown. (B). Representative plots of flow cytometry for CD58/LFA-3, CD54/ICAM-1 and CD321/JAM-A (blue) showing the marker-positive sub-populations at differentiation day 25. The number of each marker subset gives the percentage of positive cells of all cells. (C). Percentage of iPSC-cardiomyocytes (% of TNNT2-positive cells) expressing CD58/LFA-3, CD54/ICAM-1 and CD321/JAM-A at different times of differentiation. N = 5–10 measurements per marker. The mean ± standard deviation is shown. igure 2. (A). Expression levels (median fluorescence intensity) of CD58/LFA-3, CD54/ICAM-1 and CD321/ff Figure 2. (A). Expression levels (median fluorescence intensity) of CD58/LFA-3, CD54/ICAM-1 and CD321/ JAM-A on iPSC-cardiomyocytes (TNNT2-positive cells) at different times of differentiation. N = 5–10 measurements per marker. The mean ± standard deviation is shown. (B). Representative plots of flow cytometry for CD58/LFA-3, CD54/ICAM-1 and CD321/JAM-A (blue) showing the marker-positive sub-populations at differentiation day 25. The number of each marker subset gives the percentage of positive cells of all cells. (C). Percentage of iPSC-cardiomyocytes (% of TNNT2-positive cells) expressing CD58/LFA-3, CD54/ICAM-1 and CD321/JAM-A at different times of differentiation. N = 5–10 measurements per marker. The mean ± standard deviation is shown. Figure 2. (A). Expression levels (median fluorescence intensity) of CD58/LFA-3, CD54/ICAM-1 and CD321/ JAM-A on iPSC-cardiomyocytes (TNNT2-positive cells) at different times of differentiation. N = 5–10 measurements per marker. The mean ± standard deviation is shown. (B). Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z Discussion 5% serum controls MI p value CD58/LFA-3 14.1 (11.7–16.5) 8.5 (2.6–19.2) 0.02 CD54/ICAM-1 8.1 (4.0–11.8) 5.8 (4.2–6.6) >0.05 CD321/JAM-A 44.4 (38.6–50.0) 45.2 (35.5–66.6) >0.05 20% serum controls MI p value CD58/LFA-3 10.2 (6.0–13.9) 5.2 (0.4–15.5) <0.01 CD54/ICAM-1 5.4 (4.5–5.7) 7.3 (4.8–24.5) <0.01 CD321/JAM-A 45.1 (36.5–58.9) 39.8 (33.1–100.0) >0.05 Table 3. Percentage of cardiomyocytes positive for cell adhesion molecules. Data are shown as median (minimum–maximum). Numbers present the percentage of cells positive for the respective marker relative to the number of all cardiomyocytes, as defined by positive TNNT2 expression. Results are based on n = 13–28 measurements per marker. 5% serum controls MI p value CD58/LFA-3 14.1 (11.7–16.5) 8.5 (2.6–19.2) 0.02 CD54/ICAM-1 8.1 (4.0–11.8) 5.8 (4.2–6.6) >0.05 CD321/JAM-A 44.4 (38.6–50.0) 45.2 (35.5–66.6) >0.05 20% serum controls MI p value CD58/LFA-3 10.2 (6.0–13.9) 5.2 (0.4–15.5) <0.01 CD54/ICAM-1 5.4 (4.5–5.7) 7.3 (4.8–24.5) <0.01 CD321/JAM-A 45.1 (36.5–58.9) 39.8 (33.1–100.0) >0.05 Table 3. Percentage of cardiomyocytes positive for cell adhesion molecules. Data are shown as median (minimum–maximum). Numbers present the percentage of cells positive for the respective marker relative to the number of all cardiomyocytes, as defined by positive TNNT2 expression. Results are based on n = 13–28 measurements per marker. of DAMP proteins3. CD321/JAM-A is an endothelial junction protein which mediates leukocyte diapedesis in ischemic tissues25,26. Interestingly, we found that the protein is expressed in 69.6% of cardiomyocytes at day 25 of differentiation. Previously, secretion of the peptide by cardiac progenitor cells was shown27. As our cell cul- ture contains a mixture of cells, a “transfer” from other cell types such as endothelial cells cannot be ruled out. However, cardiomyocytes were defined as TNNT2-positive cells in the flow cytometry analysis and immunos- taining. To our knowledge, this is the first time that its expression has been described in this cell type.f of DAMP proteins3. CD321/JAM-A is an endothelial junction protein which mediates leukocyte diapedesis in ischemic tissues25,26. Interestingly, we found that the protein is expressed in 69.6% of cardiomyocytes at day 25 of differentiation. Previously, secretion of the peptide by cardiac progenitor cells was shown27. As our cell cul- ture contains a mixture of cells, a “transfer” from other cell types such as endothelial cells cannot be ruled out. However, cardiomyocytes were defined as TNNT2-positive cells in the flow cytometry analysis and immunos- taining. Discussion CD58/LFA-3 was found on intercalating discs of the myocardial syncytium18. However, its precise function within the myocardium is not known yet. For CD54/ ICAM-1 many studies exist which have underlined the importance of this cytokine for the function of immune response. The primary function of this cytokine is the mediation of neutrophil adhesion to endothelium, rolling and diapedesis19. In the heart, CD54/ICAM-1 is released by cardiomyocytes under different conditions of cellular stress, such as hyperoxia20, ischemia/reperfusion21, hyperglycemia22, or viral inflammation23. Recently, a positive correlation was noted between expression levels of troponin and CD54/ICAM-1 in biopsies taken from human ischemic myocardium24. The expression of CD54/ICAM-1 in ischemic myocardial tissue follows the secretion Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 5 www.nature.com/scientificreports/ Figure 3. Immunofluorescence staining demonstrates the localization of CD58/LFA-3 (A, green), CD54/ ICAM-1 (B, green) and CD321/JAM-A (C, yellow) in iPSC-cardiomyocytes (A–C: troponin T, red; nuclei, blue). Each marker is also expressed by cells other than cardiomyocytes, which can best be seen with CD58. Images were taken at 40x magnification. Scale bars represent 50 µm with the exception of CD58 (200 µm for better overview of marker expression by different cells). Figure 3. Immunofluorescence staining demonstrates the localization of CD58/LFA-3 (A, green), CD54/ ICAM-1 (B, green) and CD321/JAM-A (C, yellow) in iPSC-cardiomyocytes (A–C: troponin T, red; nuclei, blue). Each marker is also expressed by cells other than cardiomyocytes, which can best be seen with CD58. Images were taken at 40x magnification. Scale bars represent 50 µm with the exception of CD58 (200 µm for better overview of marker expression by different cells). 5% serum controls MI p value CD58/LFA-3 14.1 (11.7–16.5) 8.5 (2.6–19.2) 0.02 CD54/ICAM-1 8.1 (4.0–11.8) 5.8 (4.2–6.6) >0.05 CD321/JAM-A 44.4 (38.6–50.0) 45.2 (35.5–66.6) >0.05 20% serum controls MI p value CD58/LFA-3 10.2 (6.0–13.9) 5.2 (0.4–15.5) <0.01 CD54/ICAM-1 5.4 (4.5–5.7) 7.3 (4.8–24.5) <0.01 CD321/JAM-A 45.1 (36.5–58.9) 39.8 (33.1–100.0) >0.05 Table 3. Percentage of cardiomyocytes positive for cell adhesion molecules. Data are shown as median (minimum–maximum). Numbers present the percentage of cells positive for the respective marker relative to the number of all cardiomyocytes, as defined by positive TNNT2 expression. Results are based on n = 13–28 measurements per marker. Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z www.nature.com/scientificreports/ www.nature.com/scientificreports/ 5% serum controls MI p value CD58/LFA-3 1272.5 (1210.0–1368.0) 1111.0 (889.0–1368.0) 0.03 CD54/ICAM-1 577.0 (459.0–927.0) 466.0 (388.0–530.0) <0.01 CD321/JAM-A 1963.0 (1786.0–2218.0) 1965.0 (1546.0–3121.0) >0.05 20% serum controls MI p value CD58/LFA-3 1084.5 (959.0–1174.0) 925.0 (580.0–1221.0) 0.02 CD54/ICAM-1 467.0 (443.0–488.0) 450.5 (325.0–805.0) >0.05 CD321/JAM-A 1965.5 (1634.0–2679.0) 1666.0 (1446.0–2377.0) <0.01 Table 4. Median of the fluorescence intensity of cell adhesion molecules. Data are shown as median (minimum–maximum). Fluorescence intensity [arbitrary units] was measured in cardiomyocytes defined as cells with positive TNNT2 expression. Results are based on n = 13–28 measurements per marker. Table 4. Median of the fluorescence intensity of cell adhesion molecules. Data are shown as median (minimum–maximum). Fluorescence intensity [arbitrary units] was measured in cardiomyocytes defined as cells with positive TNNT2 expression. Results are based on n = 13–28 measurements per marker. Table 4. Median of the fluorescence intensity of cell adhesion molecules. Data are shown as median (minimum–maximum). Fluorescence intensity [arbitrary units] was measured in cardiomyocytes defined as cells with positive TNNT2 expression. Results are based on n = 13–28 measurements per marker. Figure 4. Expression levels (median fluorescence intensity relative to cell number) on iPSC-cardiomyocytes after incubation with myocardial infarction serum (MI) or control serum (5% or 20% v/v). (A) CD58/LFA-3, (B) CD54/ICAM-1, (C) CD321/JAM-A. N = 13–26 measurements per marker. The line within box represents the median, additionally the 10th, 25th, 75th, and 90th percentile is shown; whiskers represent error bars. Figure 4. Expression levels (median fluorescence intensity relative to cell number) on iPSC-cardiomyocytes after incubation with myocardial infarction serum (MI) or control serum (5% or 20% v/v). (A) CD58/LFA-3, (B) CD /ICAM 1 (C) CD321/JAM A N 13 26 k Th li i hi b Figure 4. Expression levels (median fluorescence intensity relative to cell number) on iPSC-cardiomyocytes after incubation with myocardial infarction serum (MI) or control serum (5% or 20% v/v). (A) CD58/LFA-3, (B) CD54/ICAM-1, (C) CD321/JAM-A. N = 13–26 measurements per marker. The line within box represents the median, additionally the 10th, 25th, 75th, and 90th percentile is shown; whiskers represent error bars. after incubation with myocardial infarction serum, the response of CD58/LFA-3 was completely abolished in up to 50% of cells. When translated to myocardial tissue placed in a heart subjected to infarction at a remote area, this would imply a reduction of the “appeal” of healthy myocardium to T lymphocytes, especially in contrast to the highly “attractive” infarcted area. Discussion To our knowledge, this is the first time that its expression has been described in this cell type.f i Interestingly, in our study serum of patients with myocardial infarction affected both the expression levels of the cell adhesion molecules and the number of positive cardiomyocytes. The reduction of expression levels is sur- prising, as myocardial infarction serum is supposed to contain different pro-inflammatory mediators. Indeed, Il-6 was elevated in samples of patients with myocardial infarction compared to control samples. Thus, healthy car- diomyocytes do not react by changing to an inflammatory phenotype upon stimulation with pro-inflammatory serum, but rather suppress inflammation. Recently, we showed that several cell adhesion molecules were induced by incubating iPSC-derived cardiomyocytes of healthy donors with lipopolysaccharides7. Obviously, pathways are present in iPSC-cardiomyocytes which are specialized to cell responses depending on stimulus.if Even more interesting is the finding of an effect of myocardial infarction serum on the numbers of cell num- bers expressing the different cell adhesion molecules. While some cells downregulated the respective pathways Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 6 www.nature.com/scientificreports/ In accord, the stimulation of neutrophils takes place on a low level in healthy myocardium, as the CD54/ICAM-1 expression per cell was reduced by MI serum.i y y Ventricular fibrillation (VF) is a complication of myocardial infarction occurring in up to 12% of cases28, although many cases of sudden cardiac death might be based primarily on death due to arrhythmia, giving the assumption of even higher a number. The genesis of VF during myocardial infarction is probably multifacto- rial, as local necrosis29, cardiac sympathetic nerve stimulation30,31 and also several clinical and demographic pre-disposing factors, such as alcohol intake or pre-existing atrial fibrillation32, are discussed. Recently, an asso- ciation of previous episodes of ventricular tachyarrhythmia and elevated markers of inflammation was described in patients with stable coronary artery disease33. In an animal model, ischemia-triggered VF was associated with higher plasma levels of TNF-α than electrically induced VF, ascribed to the ischemia of the myocardium and the prolonged resuscitation phase34. In our study, serum of patients with VF was less able to reduce cell adhesion molecule expression. Indeed, VF serum contained higher levels of Il-6 than MI serum without VF, although only small numbers are available from our patients group. As we do not have the blood levels of Il-6 prior to the event, Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 7 www.nature.com/scientificreports/ Figure 5. Expression levels (median fluorescence intensity relative to cell number) on iPSC-cardiomyocytes after incubation with 5% or 20% (v/v) myocardial infarction serum with vs. without ventricular fibrillation (VF). (A) CD58/LFA-3, (B) CD54/ICAM-1, (C) CD321/JAM-A. N = 7–15 measurements per marker. The line within box represents the median, additionally the 10th, 25th, 75th, and 90th percentile is shown; whiskers represent error bars. Figure 5. Expression levels (median fluorescence intensity relative to cell number) on iPSC-cardiomyocytes after incubation with 5% or 20% (v/v) myocardial infarction serum with vs. without ventricular fibrillation (VF). (A) CD58/LFA-3, (B) CD54/ICAM-1, (C) CD321/JAM-A. N = 7–15 measurements per marker. The Figure 5. Expression levels (median fluorescence intensity relative to cell number) on iPSC-cardiomyocytes after incubation with 5% or 20% (v/v) myocardial infarction serum with vs. without ventricular fibrillation (VF). (A) CD58/LFA-3, (B) CD54/ICAM-1, (C) CD321/JAM-A. N = 7–15 measurements per marker. The line within box represents the median, additionally the 10th, 25th, 75th, and 90th percentile is shown; whiskers represent error bars. www.nature.com/scientificreports/ it is impossible to discern whether this serum property is a cause or a consequence of VF during myocardial infarction. Whether there is a serum component pre-disposing to VF after myocardial infarction or whether the VF event changes the serum towards a more inflammatory phenotype remains to be clarified.l gl y p ypi Based on our results, one can speculate that in “healthy parts” of the myocardial tissue pro-inflammatory signals are actively suppressed to protect against spreading of the tissue inflammation. Serum of patients with VF seems to be more pro-inflammatory, translating into a higher risk of inadequate infarct containment for the affected myocardium. Although adhesion molecule expression was affected by serum of myocardial infarc- tion, the electrophysiological properties of the cardiomyocytes involved in the generation of arrhythmias were not influenced by serum incubation. This is interesting, as several soluble mediators accumulating in ischemic conditions participate in the generation of arrhythmias occurring during reperfusion injury35,36. Although we determined only interleukin-6 levels in our samples, one can assume based on the blood chemistry values of the patients that the MI samples consisted the “typical” molecules found after reperfusion. Our results clearly show that serum of MI patients has no effects on action potentials and thus probably no proarrhythmic effect on healthy cardiomyocytes. The reason for the arrhythmogenesis after MI needs to be further investigated. Of further interest is the reaction of cardiomyocytes generated from stem cells of patients carrying a structural alteration, such as hypertrophic or dilated cardiomyopathy or arrhythmogenic rightventricular cardiomyopathy. In these cells, adhesion molecule expression and response might be different from healthy cardiomyocytes. The study of these cells is currently under way in our laboratory. Further studies are also needed to elucidate the signal trans- duction of inflammation relating to the cell adhesion molecules in cardiomyocytes. Summary. Myocardial infarction serum exerts anti-inflammatory effects on healthy iPSC-cardiomyocytes, while serum of patients with myocardial infarction and VF is less effective. MI serum or VF serum does not change electrophysiological properties of the cells. Further studies of the innate immunofunction of cardiomyo- cytes are needed to define the underlying pathways. Limitations. The blood samples were gathered after restoration of flow, thus, after reperfusion, not during the ischemic phase of the myocardial infarction. In a strict sense, the results might therefore represent the effects of an ischemia-reperfusion injury rather than “pure” myocardial ischemia. Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z www.nature.com/scientificreports/ However, due to the necessity of immediate treatment, sample collection had to be postponed until stabilization of the patients. We tested the effect on healthy cells living in a mixed cell culture setting. This is of course a different setting than cardiomyocytes contained in myocardium of an individual surviving an ischemic event of a distant area. We measured whole-cell-expression of Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 8 www.nature.com/scientificreports/ Figure 6. Incubation with myocardial infarction serum (MI, black) does not change electrophysiological properties (control serum – Ctr, white). RP – Resting potential, APA – action potential amplitude, Vmax – maximum depolarization velocity, APD – action potential duration. The mean of the indicated number of measurements is shown; whiskers display standard deviation. Figure 6. Incubation with myocardial infarction serum (MI, black) does not change electrophysiological properties (control serum – Ctr, white). RP – Resting potential, APA – action potential amplitude, Vmax – maximum depolarization velocity, APD – action potential duration. The mean of the indicated number of measurements is shown; whiskers display standard deviation. he adhesion molecules, not only surface expression, which does not represent the presentation of the molecules o cells of the immune system. Materials and Methodsh Ethics statement. The skin biopsies from the healthy donors were taken after written informed consent had been obtained. The study was approved by the Ethics Committee of the Medical Faculty Mannheim, University of Heidelberg (approval number: 2009–350N-MA) and by the Ethics Committee of University Medical Center Göttingen (approval number: 10/9/15), and carried out in accordance with the approved guidelines. Serum generation. Venous blood of patients presenting with acute myocardial infarction (ST-segment elevation infarction, STEMI, or non-ST-segment elevation infarction, NSTEMI) was collected within the first 12 hours after the event in serum monovettes® and centrifuged. The serum was stored at −80 °C until further use. Myocardial infarction (STEMI or NSTEMI) were defined according to the guidelines of the European Society of Cardiology37,38. For control, serum of three healthy volunteers was collected using the identical centrifugation protocol. All volunteers presented no clinical manifestation of CAD or of atherosclerosis at other sites, and were taking no medication. The study was approved by the Ethics Committee of University Medical Centre Mannheim and was conducted in accord with the Declaration of Helsinki. Enzyme-Linked Immunosorbent Assay. Interleukin-6 (Il-6) was measured according to manufacturer’s instructions in serum of patients and controls with an enzyme-linked immunosorbent assay (ELISA) kit (RayBio, USA). All measurements were done in duplicate. 9 9 Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z www.nature.com/scientificreports/ Figure 7. Incubation with myocardial infarction serum with ventricular fibrillation (VF, black) does not change electrophysiological properties (myocardial infarction serum without ventricular fibrillation – MI, white). RP – Resting potential, APA – action potential amplitude, Vmax – maximum depolarization velocity, APD – action potential duration. The mean of the indicated number of measurements is shown; whiskers display standard deviation. Figure 7. Incubation with myocardial infarction serum with ventricular fibrillation (VF, black) does not change electrophysiological properties (myocardial infarction serum without ventricular fibrillation – MI, white). RP – Resting potential, APA – action potential amplitude, Vmax – maximum depolarization velocity, APD – action potential duration. The mean of the indicated number of measurements is shown; whiskers display standard deviation Human iPS cells. Human iPS cells (hiPSCs) were generated from primary human fibroblasts derived from skin biopsies of two different healthy donors. For hiPSC line Donor 1 (D1), the reprogramming factors OCT4, SOX2, KLF4 and c-MYC were transfected using lentivirus particles carrying the transactivator rtTA and an inducible polycistronic cassette, as previously described39,40. Materials and Methodsh Donor 2 (D2)-iPSC were generated using the episo- mal 4-in-1 CoMiP reprogramming plasmid (OCT4, KLF4, SOX2, c-MYC and short hairpin RNA against p53) to reprogram the primary cells into iPSCs in feeder free culture conditions. Generation of hiPSC-cardiomyocytes. HiPSCs were cultured under feeder free conditions and dif- ferentiated into iPSC-cardiomyocytes as described previously8. In brief, culture flasks and dishes were coated with Matrigel (Corning). hiPSCs were cultured in TeSR-E8 media (Stemcell Technologies) and changed to RPMI 1640-Glutamax (Life Technologies) containing sodium pyruvate, penicillin/streptomycin, B27 (Life Technologies) and ascorbic acid (Sigma Aldrich) after start of the differentiation process. Differentiation into CMs was induced by addition of CHIR99021 (Stemgent), BMP-4 (R&D Systems), Activin A (R&D Systems), FGF-2 (Miltenyi Biotec) and IWP-4 (Stemgent) at different time points to the medium. Selection of cardiomyo- cytes occurred during the third week of cultivation by changing the medium to a lactate-supplemented (Sigma, Germany) RPMI-medium free of glucose and glutamine (WKS, Germany). Afterwards, cells were fed with RPMI 1640-Glutamax containing sodium pyruvate, penicillin/streptomycin, B27 and ascorbic acid until termination of the experiments. Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 10 www.nature.com/scientificreports/ RNA extraction, cDNA synthesis and quantitative real-time PCR. RNA was extracted from hiPSC-CMs by lysis with RLT lysis buffer, followed by the application of the RNeasy MiniKit (Qiagen) accord- ing to manufacturer’s instructions. cDNA synthesis was performed with oligo (dT) primers using AMV reverse transcriptase (Roche). For quantitative real time PCR, hot start Taq DNA-polymerase and SYBR-Green were used, together with commercially available primers (GAPDH, #PPH00150F; TNNT2, #PPH025619A; POUF51, #PPH02394E, Qiagen). The mean CT value of 3 to 6 biological replicates of three different differentiations was calculated from three technical replicates. Normalized mRNA expression was calculated by using ΔCT = (CTgene of interst − CThousekeeping gene). Incubation studies and flow cytometry analysis. Cardiomyocytes at differentiation day 25 and older were incubated with cell medium supplemented with 5% or 20% (v/v) of serum of patients or controls for 48 hours at 37 °C, 5% CO2 41. Cells were detached from wells by incubation with collagenase I (CLS 1, Worthington, Cat Nr. LS004196, 250 U/mg) for 40 minutes at 37 °C. Cells were washed twice with PBS and 0.05% trypsin. Afterwards, cells were fixated in 4% formaldehyde (10 minutes, 20 °C) and permeabilized (Perm/Wash Buffer, BD). Materials and Methodsh After another wash, antibodies were added and incubated for 30 minutes at 4 °C in the dark (BD Pharmingen: anti-CD62-PE,# 555524, anti-CD58-FITC #555920, anti-CD321-PE, #552556; Antibodies online: anti-CD102-FITC, #ABIN1383720). After washing, measurements of fluorescence were performed on a FACS Canto II, BD. Analysis of measurements was performed with the software FlowJo 10.1. Flow cytometry measure- ments were done by a technician blinded for the study protocol. Expression was defined “positive” if more than 2% of cardiomyocytes were stained with the respective marker. Immunofluorescence staining. hiPSC-CMs were grown on culture slides (Falcon) and allowed to rest for at least 2 days. Cells were fixed with 4% formaldehyde and permeabilized with 0.5% triton-X, followed by incu- bation with antibodies against cell adhesion molecules (anti-CD54-FITC, ABIN #2144636; anti-CD58-FITC, BD #555920; anti-CD321-PE, BD #552556, 1 hour, room temperature). Afterwards, staining of cardiac troponin T was performed with an AlexaFluor647-conjugated antibody (BD #565744, 1 hour, room temperature), followed by nuclear staining with DAPI (Vector, #H-1200). Photographs were taken with a Leica DMRE microscope (Leica Application Suite V4.4.0, Microsystems CMS GmbH, Switzerland). Patch clamp. Action potential characteristics were measured by standard patch-clamp recording techniques in the whole-cell configuration at room temperature according to a recently published protocol42. For measure- ments, 3 randomly chosen serum samples of each group were taken, and measurements were performed in at least 13 single hiPSC-CMs per group. Statistics. Data are presented as number (percent) for ordinal data or mean + −SD or as median (minimum-maximum) for continuous data, depending on data distribution. Group comparison was done with student’s t-test or Mann-Whitney-U-rank sum test. Multiple comparisons were done with ANOVA on ranks. P values are understood to be strictly descriptive. Statistical significance was assumed for p < 0.05. All analyses were done with SigmaPlot Version 13.0. Ethics approval and consent to participate. The study was approved by the Ethics Committee of the Medical Faculty Mannheim, University of Heidelberg (approval number: 2009-350N-MA) and by the Ethics Committee of University Medical Center Göttingen (approval number: 10/9/15), and carried out in accord- ance with the approved guidelines. Written informed consent was obtained from all participants or their legal representatives. Data Availabilityh y The datasets used and/or analyzed during the current study are available from the corresponding author on rea- onable request. References References 1. Libby, P., Nahrendorf, M. & Swirski, F. K. Leukocytes Link Local and Systemic Inflammation in Ischemic Cardiovascular Disease: An Expanded “Cardiovascular Continuum”. J Am Coll Cardiol. 67, 1091–1103 (2016).l p ( ) 2. Fang, L., Moore, X. L., Dart, A. M. & Wang, L. M. Systemic inflammatory response following acute myocardial infarction. J Geriat Cardiol. 12, 305–312 (2015).htl 3. Prabhu, S. D. & Frangogiannis, N. G. The Biological Basis for Cardiac Repair After Myocardial Infarction: From Inflammation to Fibrosis. Circ Res. 119, 91–112 (2016). Th fl d l d d ll d l ( ) 4. Frangogiannis, N. G. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol. 11, 255–265 (2014) 4. Frangogiannis, N. G. The inflammatory response in myocardial injury, repair, and remodelling. Nat Rev Cardiol. 11, 255–265 (2014). 5. Kim, E. J., Kim, S., Kang, D. O. & Seo, H. S. Metabolic activity of the spleen and bone marrow in patients with acute myocardial infarction evaluated by 18f-fluorodeoxyglucose positron emission tomograpic imaging. Circ Cardiovasc Imaging. 7, 454–460 (2014). 6. David, S., Thamm, K., Schmidt, B. M., Falk, C. S. & Kielstein, J. T. Effect of extracorporeal cytokine removal on vascular barrier function in a septic shock patient. Journal of intensive care. 5, 12 (2017). g ghl y y j y g 5. Kim, E. J., Kim, S., Kang, D. O. & Seo, H. S. Metabolic activity of the spleen and bone marrow in patients with acute myocardial infarction evaluated by 18f-fluorodeoxyglucose positron emission tomograpic imaging. Circ Cardiovasc Imaging. 7, 454–460 (2014). hl 5. Kim, E. J., Kim, S., Kang, D. O. & Seo, H. S. Metabolic activity of the spleen and bone marrow in patients with acute myocardial infarction evaluated by 18f-fluorodeoxyglucose positron emission tomograpic imaging. Circ Cardiovasc Imaging. 7, 454–460 (2014). 6. David, S., Thamm, K., Schmidt, B. M., Falk, C. S. & Kielstein, J. T. Effect of extracorporeal cytokine removal on vascular barrier infarction evaluated by 18f-fluorodeoxyglucose positron emission tomograpic imaging. Circ Cardiovasc Imaging. 7, 454–460 (2014). 6. David, S., Thamm, K., Schmidt, B. M., Falk, C. S. & Kielstein, J. T. Effect of extracorporeal cytokine removal on vascular barrier function in a septic shock patient. Journal of intensive care. 5, 12 (2017). yl yg p g p g g g g , ( ) 6. David, S., Thamm, K., Schmidt, B. M., Falk, C. S. & Kielstein, J. T. Acknowledgements g We thank Ms. Y. Wiegräfe and Ms. L. Rogge for excellent technical assistance in iPSC generation and characterization, and Ms. Stefanie Uhlig and Ms. Claudia Liebetrau for the excellent technical assistance of flow cytometry measurements. This study was funded by DZHK. g We thank Ms. Y. Wiegräfe and Ms. L. Rogge for excellent technical assistance in iPSC generation and characterization, and Ms. Stefanie Uhlig and Ms. Claudia Liebetrau for the excellent technical assistance of flow cytometry measurements. This study was funded by DZHK. www.nature.com/scientificreports/ J Cell Mol Med. 18, 1509–1518 (2014).l y y 42. Yucel, G. et al. Lipopolysaccharides induced inflammatory responses and electrophysiological dysfunctions in human-induced pluripotent stem cell derived cardiomyocytes. Sci Rep. 7, 2935, https://doi.org/10.1038/s41598-017-03147-4 (2017). www.nature.com/scientificreports/ Inflammation markers are associated with metabolic syndrome and ventricular arrhythmia in patients with di P t Hi M d D (O li ) 70 56 66 h //d i /10 5604/17322693 1194612 (2016) 3. Safranow, K. et al. Inflammation markers are associated with metabolic syndrome and ventricular arrhythmia in patients with coronary artery disease. Postepy Hig Med Dosw (Online) 70, 56–66, https://doi.org/10.5604/17322693.1194612 (2016).hl 4. Niemann, J. T. et al. The proinflammatory cytokine response following resuscitation in the swine model depends on the method o ventricular fibrillation induction. Acad Emerg Med. 15, 939–944 (2008). i g 5. Curtis, M. J., Pugsley, M. K. & Walker, M. J. Endogenous chemical mediators of ventricular arrhythmias in ischaemic heart disease Cardiovascular research. 27, 703–719 (1993). 6. Carmeliet, E. Cardiac ionic currents and acute ischemia: from channels to arrhythmias. Physiological reviews. 79, 917–1017 (1999) 7. Task Force on the management of STEMI. ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation. Eur Heart J. 33, 2569–2619 (2012).fi g ( ) offi, M. et al. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST- t l ti T k F f th M t f A t C S d i P ti t P ti ith t P i t t ST 38. Roffi, M. et al. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST- segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST- Segment Elevation of the European Society of Cardiology (ESC) Eur Heart J 37 267 315 (2016) 8. Roffi, M. et al. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST Segment Elevation of the European Society of Cardiology (ESC). Eur Heart J. 37, 267–315 (2016).f 39. Larribere, L. et al. NF1 loss induces senescence during human melanocyte differentiation in an iPSC-based model. Pigment cell & melanoma research. 28, 407–416 (2015).fi ( ) 40. Maherali, N. et al. A high-efficiency system for the generation and study of human induced pluripotent stem cells. Cell stem cell. 3, 340–345 (2008). 1. Dambrot, C. et al. Serum supplemented culture medium masks hypertrophic phenotypes in human pluripotent stem cell derived cardiomyocytes. www.nature.com/scientificreports/ www.nature.com/scientificreports/ 2. Trouw, L. A. et al. C4b-binding protein is present in affected areas of myocardial infarction during the acute inflammatory phase and covers a larger area than C3. PloS one 3, e2886, https://doi.org/10.1371/journal.pone.0002886 (2008). g p g j p 3. Carmona, R. H., Tsao, T., Dae, M. & Trunkey, D. D. Myocardial dysfunction in septic shock. Archives of surgery. 120, 30–35 (1985) k l h l dh l d d l d d b d l h ll d d h d y y y p f g y 4. Youker, K. et al. Neutrophil adherence to isolated adult cardiac myocytes. Induction by cardiac lymph collected during ischemia and reperfusion. J Clin Inves.t 89, 602–609 (1992). p ( ) 15. El-Battrawy, I. et al. Ion Channel Dysfunctions in Dilated Cardiomyopathy in Limb-Girdle Muscular Dystrophy. Circula Genomic and precision medicine. 11, e001893, https://doi.org/10.1161/CIRCGEN.117.001893 (2018).i p p g 6. Hoppenbrouwers, I. A. et al. Replication of CD58 and CLEC16A as genome-wide significant risk genes for multiple sclerosis. J Hum Genet. 54, 676–680 (2009). 7. Xu, S. et al. CD58, a novel surface marker, promotes self-renewal of tumor-initiating cells in colorectal cancer. Oncogene. 34 1520–1531 (2015). 18. Smith, M. E. & Thomas, J. A. Cellular expression of lymphocyte function associated antigens and the intercellular adhesion molecule-1 in normal tissue. J Clin Pathol. 43, 893–900 (1990).h J , ( ) 19. Lyck, R. & Enzmann, G. The physiological roles of ICAM-1 and ICAM-2 in neutrophil migration into tissues. Current opinion in hematology. 22, 53–59 (2015).l gy , ( ) 0. Hafner, C. et al. Hyperoxia Induces Inflammation and Cytotoxicity in Human Adult Cardiac Myocytes. Shock. 47, 436–444 (2017) 20. Hafner, C. et al. Hyperoxia Induces Inflammation and Cytotoxicity in Human Adult Cardiac Myocytes. Shock. 47, 436–444 (2017). 21. Youker, K. A. et al. Molecular evidence for induction of intracellular adhesion molecule-1 in the viable border zone associated with ischemia-reperfusion injury of the dog heart. Circulation. 89, 2736–2746 (1994).l l 21. Youker, K. A. et al. Molecular evidence for induction of intracellular adhesion molecule-1 in the viable border zone associated ischemia-reperfusion injury of the dog heart. Circulation. 89, 2736–2746 (1994).l p j y g 2. Qin, W. D. et al. Poly(ADP-ribose) polymerase 1 inhibition protects cardiomyocytes from inflammation and apoptosis in diabetic cardiomyopathy. Oncotarget. 7, 35618–35631 (2016). 22. Qin, W. D. et al. Poly(ADP-ribose) polymerase 1 inhibit cardiomyopathy. Oncotarget. 7, 35618–35631 (2016). Author Contributions K.S. and I.E.-B. analyzed and interpreted the results and were major contributors in writing the manuscript. I.E.-B., K.S., Z.Z., GY, H.L., X.L., S.L., X.Z., T.W., U.R., K.B., J.U., W.H.Z. and L.C. validated and visualized the data. C.S. performed FACS analysis and contributed to writing the manuscript. M.B., I.A., X.Z. and S.L. supervised the experiments. All authors read and approved the final manuscript. K.S. and I.E.-B. analyzed and interpreted the results and were major contributors in writing the manuscript. I.E.-B., K.S., Z.Z., GY, H.L., X.L., S.L., X.Z., T.W., U.R., K.B., J.U., W.H.Z. and L.C. validated and visualized the data. C.S. performed FACS analysis and contributed to writing the manuscript. M.B., I.A., X.Z. and S.L. supervised the experiments. All authors read and approved the final manuscript. References Effect of extracorporeal cytokine removal on vascular barrier function in a septic shock patient. Journal of intensive care. 5, 12 (2017). p p f ( ) 7. El-Battrawy, I. et al. Expression of Inflammation-related Intercellular Adhesion Molecules in Cardiomyocytes In Vitro and Modulation by Pro-inflammatory Agents. In vivo. 30, 213–217 (2016).lf f 7. El-Battrawy, I. et al. Expression of Inflammation-related Intercellular Adhesion Mo Modulation by Pro-inflammatory Agents. In vivo. 30, 213–217 (2016).lf yl y g ( ) 8. El-Battrawy, I. et al. Hyperthermia Influences the Effects of Sodium Channel Blocking Drugs in Human-Induced Pluripotent Stem Cell-Derived Cardiomyocytes. PloS one. 11, e0166143, https://doi.org/10.1371/journal.pone.0166143 (2016). yl y g ( ) 8. El-Battrawy, I. et al. Hyperthermia Influences the Effects of Sodium Channel Blocking Drugs in Human-Induced Pluripotent Stem Cell-Derived Cardiomyocytes. PloS one. 11, e0166143, https://doi.org/10.1371/journal.pone.0166143 (2016). y y p g j p ( ) 9. Christia, P. & Frangogiannis, N. G. Targeting inflammatory pathways in myocardial infarction. European journal of clinical investigation. 43, 986–995, https://doi.org/10.1111/eci.12118 (2013).t y y p g j p 9. Christia, P. & Frangogiannis, N. G. Targeting inflammatory pathways in myocardial infarction. European journal of clinical investigation. 43, 986–995, https://doi.org/10.1111/eci.12118 (2013).t g g 10. Sun, M. et al. Temporal response and localization of integrins beta1 and beta3 in the heart after myocardial infarction: regulatio cytokines. Circulation. 107, 1046–1052 (2003). y , ( ) 11. Irwin, M. W. et al. Tissue expression and immunolocalization of tumor necrosis factor-alpha in postinfarction dysfunctional myocardium. Circulation. 99, 1492–1498 (1999). y ( ) 11. Irwin, M. W. et al. Tissue expression and immunolocalization of tumor necrosis factor-alpha in postinfarction dysfunctional myocardium. Circulation. 99, 1492–1498 (1999). Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 11 www.nature.com/scientificreports/ y p y g 23. Seko, Y. et al. Expression of intercellular adhesion molecule-1 in murine hearts with acute myocarditis caused by coxsackievirus B3. J Clin Invest. 91, 1327–1336 (1993). 24. Ceausu, M., Curca, C., Dermengiu, D. & Ardeleanu, C. Simultaneous immunophenotypical assessment of troponin and extracellular matrix molecules in myocardium of patients with sudden cardiac death. Romanian journal of morphology and embryology=Revue roumaine de morphologie et embryologie. 50, 103–106 (2009).hil p g y g 25. Nourshargh, S., Krombach, F. & Dejana, E. The role of JAM-A and PECAM-1 in modulating leukocyte infiltration in inflamed and ischemic tissues. Journal of leukocyte biology. 80, 714–718 (2006).h f y gy 6. Mueller, M. et al. The receptor for activated complement factor 5 (C5aR) conveys myocardial ischemic damage by mediating neutrophil transmigration. Immunobiology. 218, 1131–1138 (2013).lt h neutrophil transmigration. Immunobiology. 218, 1131–1138 (2 27. Liu, M. L. et al. Anti-inflammatory peptides from cardiac progenitors ameliorate Heart Assoc. 3, e001101, https://doi.org/10.1161/JAHA.114.001101 (2014). 27. Liu, M. L. et al. Anti-inflammatory peptides from cardiac progenitors ameliorate dysfunction after myocardial infarction. J Am Heart Assoc. 3, e001101, https://doi.org/10.1161/JAHA.114.001101 (2014).i p g 8. Glinge, C., Sattler, S., Jabbari, R. & Tfelt-Hansen, J. Epidemiology and genetics of ventricular fibrillation during acute myocardia infarction. J Geriatr Cardiol. 13, 789–797 (2016).t 9. Kolettis, T. M. et al. Arrhythmogenesis after acute myocardial necrosis with and without preceding ischemia in rats. J Basic Clin Physiol Pharmacol. 25, 143–153 (2014).t y 0. Li, C. Y. & Li, Y. G. Cardiac Sympathetic Nerve Sprouting and Susceptibility to Ventricular Arrhythmias after Myocardial Infarction Cardiol Res Pract. 2015, 698368, https://doi.org/10.1155/2015/698368 (2015). g 31. Kolettis, T. M. Ventricular tachyarrhythmias during acute myocardial infarction: the role of endothelin-1. Life Sci. 118, 136 (2014).ii 32. Jabbari, R. et al. Incidence and risk factors of ventricular fibrillation before primary angioplasty in patients with first ST-elev myocardial infarction: a nationwide study in Denmark. J Am Heart Assoc 4, e001399, https://doi.org/10.1161/JAHA.114.00 (2015). J , R ki p y g p y pi myocardial infarction: a nationwide study in Denmark. J Am Heart Assoc 4, e001399, https://doi.org/10.1161/JAHA.114.001399 (2015). 33 Safranow K et al Inflammation markers are associated with metabolic syndrome and ventricular arrhythmia in patients with (2015). 33. Safranow, K. et al. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Additional Informationh Additional Information Competing Interests: The authors declare no competing interests. Competing Interests: The authors declare no competing interests. Competing Interests: The authors declare no competing interests. Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 12 www.nature.com/scientificreports/ Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. fi Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre- ative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not per- mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2019 Scientific Reports \| (2019) 9:5651 \| https://doi.org/10.1038/s41598-019-42079-z 13
W4249878956.txt	https://www.scielo.br/j/csp/a/dHjX7HXtccVNfvjgQVNvGmL/?lang=pt&format=pdf	fr		[NO TITLE AVAILABLE]	Cadernos de Saúde Pública	2,003	cc-by	4,558	RESENHAS REVIEWS NU & VESTIDO: DEZ ANTROPÓLOGOS REVELAM A CULTURA DO CORPO CARIOCA. Miriam Goldenberg (organizadora). Rio de Janeiro: Editora Record, 2002. 414 pp. ISBN: 85-0106-260-X O título Nu & Vestido, como observa a organizadora, foi inspirado na obra O Cru e o Cozido de LéviStrauss. Tanto numa quanto noutra obra, as díades (nu/vestido e cru/cozido) não são pensadas apenas como estados de vestimenta e alimento, respectivamente, mas servem também como referências facilitadoras para classificar pessoas, coisas, estilos de vida, sentimentos, rituais, crenças e valores. Essas díades são espaços privilegiados para se refletir acerca da oposição entre natureza e cultura. Os corpos das pessoas são a temática central do livro. Corpos que se cobrem, se descobrem e encobrem, dentre outros aspectos, traços identitários pessoais e grupais, construídos socialmente. Como o próprio subtítulo indica, dez antropólogos, brasileiros e estrangeiros, distribuem-se em nove estudos em que se propõem a desvendar a cultura do corpo carioca. Apesar de os autores focalizarem sua análise na Cidade do Rio de Janeiro, no final do século XX e início do século XXI, suas interpretações se deslocam dos casos e ganham tal densidade que conseguem ultrapassar os limites espaciais e temporais, promovendo uma polissemia corporal. Os estudos, em ordem de capítulos, com os seus focos e respectivos autores se configuram da seguinte forma: (1) a valorização e a transformação do corpo (Goldenberg & Ramos); (2) imagens identitárias cariocas (Gontijo); (3) a corpolatria com busca dos (h) alteres-ego (Malysse); (4) os corpos sob o efeito dos anabolizantes (Sabino); (5) a cirurgia plástica no universo de beleza do Rio de Janeiro (Edmonds); (6) o corpo e a classificação da cor numa praia carioca (Farias); (7) a relação entre estética e política (Fry); (8) o corpo da bruxa num leitura moderna (Osório) e (9) a construção de masculinidade a partir da vestimenta do homem (Dutra). São nove estudos que procuram: descrever etnograficamente especificidades dos corpos cariocas, interpretar etnologicamente tais singularidades e ensaiar uma reflexão que, juntos ou separados, podem servir para uma teoria antropológica do corpo em geral. Dentre os estudos, a título de ilustração, podem ser destacados o primeiro e o terceiro. No primeiro estudo, Goldenberg & Ramos, têm como foco a valorização e a transformação do corpo, tomando como ponto de partida a referência da ideologia do body building expressa na “cultura da malhação”. Seguem sua análise passando pelo corpo (des) coberto, observando que nem sempre o nu se associa a ser indecente e vestido a ser decente. Muitas vezes, o nu pode ser menos indecente do que certas vestimentas ou formas físicas. Continuando essa reflexão, assinalam que, dentro da moral da “boa forma”, o corpo sem marcas indesejáveis, a exemplo das rugas, e sem excessos, como a gordura, mesmo nu, pode ser considerado decentemente vestido. Os autores, na última parte do seu estudo, analisam a valorização do corpo nas camadas médias do Rio de Janeiro, observando que o corpo pode se tornar um valor que tanto identifica um indivíduo como parte de um grupo quanto o diferencia de outros. Concluem que o corpo pode sintetizar três idéias que se articulam entre si: insígnia ou emblemas, grife ou moda e prêmio ou medalha. O terceiro estudo, de autoria de Malysse, antropólogo francês, se destaca pela exposição metodológica clara e densa, servindo como referência para se desenhar estudos etnográficos. Colocando-se, primeiramente, na situação de turista, esse autor analisa as interações sociais como encontros entre os corpos, demarcados por contatos com o outro e com si próprio. Segue textualizando a construção corporal nas academias, a moda corporal como expressão de si, a relação corpolatria-cordialidade e o corpo como expressão de arte. Por último, Malysse, retomando o título de seu estudo, acentua que no universo da corpolatria, o corpo se traduz em alter ego e o ego se mede pelo peso dos halteres erguidos no cotidiano. Esses dois estudos e os demais que compõem a obra vêm ao encontro de um empreendimento, que continua atual, de se pensar a relação entre corpo e sociedade. Castro (2003) aponta uma das implicações dessa relação que é a do redesenhamento do corpo em busca da definição de identidades e do desenvolvimento de projetos do self. Para a autora, nesse redesenhamento ora o indivíduo se sobrepõe à sociedade, ora o inverso ocorre. Essa reflexão pode ser ampliada se for tomada como referência a modernidade tão bem descrita por Giddens (2002). Nesse cenário, como assinala esse conhecido sociólogo inglês, o corpo, cada vez menos visto como uma “entidade”, se constrói reflexivamente pela experiência de ser. Segundo ele, em meio às esferas da reprodução biológica, da engenharia genética e da intervenção médica, o corpo pode ser visto como uma questão de escolhas e opções. Ancorada ainda nesse autor, a discussão pode ganhar maior densidade de análise na medida em que aspectos do corpo são pensados em sua relevância para o eu e para a auto-identidade. Assim, a aparência, a postura corporal, a sensualidade e os regimes a que o corpo se submete podem ser categorias analíticas para se interpretar a inserção dos corpos nas interações da vida cotidiana. Outro desdobramento que a leitura da obra em tela pode proporcionar, articulada com a análise de outros autores, é uma maior compreensão do corpo como metáfora. Entendendo esse conceito à luz de (Bourdieu, 2002), como transferências analógicas de esquemas, pode-se considerar que o corpo tanto ser- Cad. Saúde Pública, Rio de Janeiro, 19(6):1893-1897, nov-dez, 2003 1893 1894 RESENHAS REVIEWS ve para falar da sociedade como esta pode ser utilizada para dele tratar. O livro Nu & Vestido: Dez Antropólogos Revelam a Cultura do Corpo Carioca, organizado por Mirian Goldenberg, devido à sua riqueza, pode suscitar outros diálogos com outros textos das ciências sociais. Com ou sem esses diálogos, sua leitura em si pode ser importante para todos os profissionais de saúde que buscam múltiplos olhares sobre o corpo humano. Com base em um olhar ampliado, tanto no ato de vestir como no de despir, é possível fugir de uma prática em que os corpos são considerados apenas como uma matéria inerte, excluídos de sua subjetividade e de seus significados (Silva, 2001). Esse não reducionismo se torna viável à medida em que se articule a visão anatomofisiológica com a dimensão simbólica que o corpo pode assumir na doença e na saúde, prestando atenção às narrativas corporais que se constroem reflexivamente entre personagens, enredos e cenários. Romeu Gomes Instituto Fernandes Figueira, Fundação Oswaldo Cruz, Rio de Janeiro, Brasil. romeu@iff.fiocruz.br BOURDIEU, P., 2002. Esboço de Uma Teoria da Prática, Precedido de Três Estudos de Etnologia Cabila. Oeiras: Editora Celta. CASTRO, A. L., 2003. Culto ao Corpo e Sociedade: Mídia, Estilo de Vida e Cultura de Consumo. São Paulo: Annublume/Fundação de Amparo à Pesquisa do Estado de São Paulo. GIDDENS, A., 2002. Modernidade e Identidade. Rio de Janeiro: Jorge Zahar. SILVA, A. M., 2001. O corpo no mundo: Algumas reflexões acerca da expectativa de corpo atual. In: A (Des)Construção do Corpo (J. C. Granado, org.), pp. 11-33, Blumenau: Edifurb. EPIDEMIOLOGIA E SAÚDE DOS POVOS INDÍGENAS NO BRASIL. Carlos E. A. Coimbra Jr., Ricardo Ventura Santos & Ana Lúcia Escobar (organizadores). Rio de Janeiro: Editora Fiocruz/ABRASCO, 2003. 257 pp. ISBN: 85-7541-022-9 O presente livro divulga contribuições apresentadas em uma oficina de trabalho denominada Saúde e Epidemiologia das Populações Indígenas no Brasil, realizada durante o V Congresso Brasileiro de Epidemiologia, promovido pela Associação Brasileira de PósGraduação em Saúde Coletiva (ABRASCO) em 2002. Ele pode ser considerado como uma seqüência e atualização de coletâneas anteriores realizadas pelos organizadores (Coimbra, 1991; Santos & Coimbra, 1994; Santos & Escobar, 2001). Um exame a esses trabalhos anteriores revela que embora muitos autores apareçam em todos, há também uma sucessão de outros especialistas, o que demonstra a vitalidade dos estudos e pesquisas que estão sendo desenvolvidos na área. Na verdade, essas investigações podem ser separadas em dois períodos bem marcados, antes e depois que Carlos E. A. Coimbra Jr. e Ricardo Ventura Santos iniciaram e desenvolveram seu programa de estudos entre esses povos. Cad. Saúde Pública, Rio de Janeiro, 19(6):1893-1897, nov-dez, 2003 As contribuições ao livro foram preparadas por 28 autores, que trabalham em número equivalente de instituições tanto universitárias como de assistência à saúde, distribuídas por 12 cidades de nove unidades da federação, localizadas entre aproximadamente 30º de latitude sul (Porto Alegre) até próximo da Linha do Equador (Manaus). Isto possibilitou ampla representatividade tanto entre estudiosos como com relação aos grupos indígenas estudados. Os assuntos cobertos foram também bastante abrangentes, envolvendo aspectos mais específicos, como a saúde bucal, mas também nutrição em diferentes faixas etárias, inclusive em idosos; detalhes epidemiológicos relacionados a doenças infecciosas e crônico-degenerativas (por exemplo, diabetes mellitus, alcoolismo), bem como estatísticas sobre internações hospitalares, e programas de saúde individualizados e sua logística, como a questão da formação de agentes de saúde, o financiamento das atividades e a distritalização das decisões. O desenvolvimento da pesquisa biomédica tornou claro que não há mais lugar para posições ingênuas (expressadas por muitos ativistas políticos), de que basta a introdução de uma série de medidas intervencionistas em grupos etnicamente diferenciados para a solução de seus problemas de saúde. Por outro lado, conservacionistas extremos chegam até a preconizar a abolição completa da chamada “medicina científica”, com o isolamento desses grupos e a manutenção estrita de suas práticas curativas tradicionais. Esta posição é claramente antiética, pois resulta na negação, a esses povos, de todos os benefícios que a medicina desenvolveu ao longo dos séculos para o bem-estar da humanidade como um todo. O que é necessário é o exame caso a caso de cada comunidade, tendo em vista as peculiaridades de seu ambiente físico, biótico e sociocultural, e por meio de uma interação harmônica com a comunidade montar programas de bem-estar físico e psíquico. A enunciação de uma política como a indicada é, naturalmente, bem mais fácil do que a sua implementação. Testemunha disto é a história do contato e convivência, nem sempre pacífica, dos nativos americanos com os colonizadores que por aqui aportaram a partir do século dezesseis. Esta história é monotonamente similar ao longo de todo o continente. Após o contato, esses grupos nativos, muitas vezes pequenos e isolados, são expostos a agentes patogênicos aos quais as suas histórias de vida não os havia preparado. O resultado é uma mortalidade espantosa, que pode inclusive levar ao seu extermínio. Em casos mais favoráveis há, no entanto, após a catástrofe, um período de lenta e posteriormente mais acelerada recuperação. Essa recuperação demográfica, no entanto, é apenas um dos aspectos do problema. Grupos de caçadores-coletores estão geralmente muito bem adaptados ao seu ambiente. A mudança brusca para uma prática agriculturalista e uma atualização intempestiva com a sociedade capitalista e industrial geralmente os condena a uma vida de semipárias, de pobreza marcante, na interface entre o rural e o urbano. Um grande número de pessoas bem-intencionadas tem tentado reverter este quadro. No começo do século 20, foi criado o Serviço de Proteção aos Índios (SPI) substituído posteriormente pela Fundação Nacional do Índio (FUNAI), que tinha como um dos RESENHAS REVIEWS seus objetivos o tratamento adequado da saúde indígena. A partir de 1999, foram implantados Distritos Sanitários Especiais Indígenas (DSEIs), pela Fundação Nacional de Saúde (FUNASA), com o reconhecimento explícito de que havia a necessidade de tratamento diferenciado para esses grupos. Apesar de todas as siglas no entanto, o quadro que é revelado pelo livro sob exame é desalentador. A maioria dessas comunidades apresenta condições de saúde equivalentes ou piores do que as das camadas menos favorecidas da sociedade brasileira. Enquanto as doenças infecto-contagiosas não estão sob controle, sobrepõe-se às mesmas o espectro de novas ameaças, como a obesidade e entidades clínicas como a diabetes mellitus, que já são problemas agudos em nativos da América do Norte. Para a superação deste quadro negativo há a necessidade de esforços conjugados das entidades assistenciais, governamentais e não-governamentais, com a comunidade acadêmica universitária. Mas é óbvio que a questão da saúde indígena está intimamente vinculada a todos os terríveis problemas econômicos e sociais que atualmente afligem a população brasileira como um todo. Financiamento adequado, mobilização e organização de quadros funcionais competentes, e colaboração íntima com os líderes comunitários são indicações óbvias para a solução do impasse. O livro em referência e as outras obras mencionadas, assim como os programas de apoio que elas descrevem, mostram os caminhos que devem ser trilhados. Cabe aos governantes e a outras entidades de apoio aos movimentos indígenas levar em consideração todo este conjunto de doutrinas e ações, para a concretização de uma qualidade de vida apropriada para os primeiros colonizadores de nosso continente. Francisco M. Salzano Departamento de Genética, Instituto de Biociências, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brasil. francisco.salzano@ufrgs.br COIMBRA Jr., C. E. A., 1991. Saúde de populações indígenas. Cadernos de Saúde Pública, 7:449-605. SANTOS, R. V. & COIMBRA Jr., C. E. A., 1994. Saúde e Povos Indígenas. Rio de Janeiro: Editora Fiocruz. SANTOS, R. V. & ESCOBAR, A. L., 2001. Saúde dos povos indígenas no Brasil: Perspectivas atuais. Cadernos de Saúde Pública, 17:258-445. PROGRAMAS DE SALUD REPRODUCTIVA PARA ADOLESCENTES. LOS CASOS DE BUENOS AIRES, MEXICO D.F. Y SAN PABLO. Mónica Cogna (coord.). Buenos Aires: CEDES, 2001. 289 pp. ISBN: 950-9572-19-5 Como conseqüência do aumento do número de gravidez na adolescência, assim como do maior número de adolescentes infectados pelo HIV nas duas últimas décadas, é preciso que se busquem iniciativas em diversas áreas do conhecimento, que possam contribuir para a diminuição desta prevalência. Vários são os fatores envolvidos nestes agravos, e o setor saúde assume um importante papel na sua prevenção, que pode ser obtida por meio da discussão sobre saúde sexual e reprodutiva. A publicação organizada por Mónica Cogna traz contribuições a este debate, atualizando-o e contextualizando-o. Ela apresenta as características e a análise de diferentes serviços que atendem adolescentes, com o enfoque na promoção de saúde e na prevenção de gravidez não desejada e de doenças de transmissão sexual. Foram selecionadas para estudo de caso, serviços de três cidades das Américas: Buenos Aires, Cidade do México e São Paulo. Em cada uma delas, escolhidos dois, considerados pelas pesquisadoras os mais representativos e/ou interessantes para o trabalho, sendo que, muitos, estavam vinculados a universidades. Na metodologia da pesquisa foram analisados os documentos oficiais dos países e das cidades sobre a política de saúde para adolescentes, assim como a proposta de atenção dos serviços destinados a este grupo. Cada país contou com uma pesquisadora específica. Elas apresentam uma descrição do espaço físico, a dinâmica da atenção, o material educativo disponível e as categorias profissionais envolvidas no atendimento. Esses profissionais foram entrevistados, assim como os usuários dos serviços. A apresentação dos serviços de cada cidade é feita em capítulos separados, com a utilização da mesma metodologia. Na última parte do livro a organizadora faz uma análise comparativa tanto entre os serviços como entre as cidades. Nesta pesquisa, com cunho avaliativo, buscaramse indicadores de qualidade que contemplassem a saúde sexual e reprodutiva dos adolescentes, como por exemplo, a realização de práticas educativas para os adolescentes que chegassem para atendimento nos serviços avaliados e a distribuição de material educativo. Os vários tópicos que as pesquisadoras abordam convergem para avaliar se existe uma adequação das políticas de saúde nacional e regional, e dos serviços pesquisados, aos compromissos assumidos nas Conferências do Cairo e de Beijing. Tanto a Plataforma de Ação da Conferência Internacional Sobre População e Desenvolvimento (CIPD) realizada no Cairo em 1984, quanto a Plataforma de Ação da Quarta Conferência Mundial sobre a Mulher, em Beijing, no ano de 1995, incluem referências sobre a saúde reprodutiva dos adolescentes. Nestes encontros foi feito um alerta quanto ao descuido dos serviços de saúde reprodutivo em relação a esta população, orientando que deveriam ser facilitadas informações e acesso aos serviços. Nestes, se discutiria a sexualidade dos adolescentes, seja individualmente numa consulta ou em grupos educativos, de modo que obtivessem informações que possibilitassem se proteger contra gravidez não desejada e de doenças de transmissão sexual. Na análise do sistema público de saúde, nos três locais pesquisados, ficam evidentes os problemas decorrentes da globalização, considerada por Cogna como geradora do processo de exclusão social. Em todos existiram cortes financeiros, com impacto para o setor saúde, manifestado, principalmente, pela progressiva escassez de recursos humanos e materiais. Foram detectados eixos comuns entre os serviços dos diferentes países, tais como o reconhecimento da importância da integralidade da atenção ao adolescente, que não deve ser atendido somente na sua queixa. Consideram que, independente do motivo que gerou a consulta, este momento deve ser apro- Cad. Saúde Pública, Rio de Janeiro, 19(6):1893-1897, nov-dez, 2003 1895 1896 RESENHAS REVIEWS veitado para abordagem de questões prevalentes neste grupo. A maior parte dos serviços trabalha a sexualidade somente com o enfoque de risco, mas um deles critica esta abordagem, sugerindo que o deve se trabalhar com vulnerabilidades. As autoras apresentam críticas aos serviços pesquisados, apontando algumas estratégias para otimizar a atenção integral e, efetivamente, se conseguir alcançar a saúde sexual e reprodutiva. Dentre suas sugestões encontram-se a valorização do adolescente como sujeito de direitos, o reforço do protagonismo juvenil, a concepção da integralidade como uma prática, e não somente uma filosofia. Também recomendam que os homens adolescentes sejam mais envolvidos nesta discussão e que se faça uma referência mais clara à perspectiva de gênero. O trabalho organizado por Mónica Cogna pode incentivar novas pesquisas, de tal forma que contemplem um maior número de serviços. Mas, desde já podem servir de norte aos profissionais que atendem adolescentes. O reconhecimento das críticas feitas à forma como a atenção é prestada, assim como das sugestões apresentadas pelas autoras, poderão repercutir na otimização da atenção a este grupo da população. Uma atenção que contemple de modo eficaz a saúde sexual e reprodutiva dos adolescentes, contribuirá para a diminuição do índice de agravos decorrentes de práticas sexuais não seguras. Olga Maria Bastos Instituto Fernandes Figueira, Fundação Oswaldo Cruz, Rio de Janeiro, Brasil. PROMOÇÃO DA SAÚDE: CONCEITOS, REFLEXÕES E TENDÊNCIAS. D. Czeresnia & C. M. Freitas (organizadores). Rio de Janeiro: Editora Fiocruz, 2003. 176 pp. ISBN: 85-7541-024-5 A promoção da saúde pode ser considerada como a síntese de desenvolvimentos teóricos nucleares e de movimentos práticos estratégicos do campo da saúde coletiva. Daí a dupla relevância do conjunto de ensaios reunidos por Dina Czeresnia & Carlos Machado Freitas. O livro introduz o leitor a questões centrais relacionadas com o tema que convidam ao debate. O primeiro deles, relativo à natureza do conceito de saúde, enquanto conceito vulgar, questão filosófica ou conceito científico traz novos insights e provoca algumas reflexões. Com perspectivas distintas, Dina Czeresnia, Naomar Almeida Filho e Sandra Caponi retomam as análises de Canguilhem sobre a saúde, particularmente aquelas desenvolvidas na conferência ministrada por este autor em 1988 na Universidade de Strasbourg, onde são discutidos o conceito de saúde como conceito vulgar e a questão filosófica aparentemente diferente de um conceito científico. Canguilhem, naquela conferência, mais que defender sistematicamente uma tese como em O Normal e o Patológico faz, segundo ele próprio, algumas reflexões sobre a saúde com base nas contribuições de diversos outros filósofos como Leibniz, Diderot, Kant, Descartes e Nietzsche. A partir de uma carta de Descartes a Chanut ele problematiza a possibilidade da saúde ser vista como a verdade do corpo invertendo a formulação cartesiana. Além dessa idéia, Can- Cad. Saúde Pública, Rio de Janeiro, 19(6):1893-1897, nov-dez, 2003 guilhem retoma alguns eixos do seu pensamento desenvolvido no livro O Normal e o Patológico, acerca da saúde como capacidade dos indivíduos serem normativos diante da vida. Ele próprio faz um alerta se, na sua tentativa de elucidar um conceito, não correria o risco de ser tomado por uma elocubração? Se interroga se, em buscando na filosofia ajuda para fortificar sua proposição de tomar a saúde por um conceito ao qual a experiência vulgar confere sentido de uma permissão de viver para o bom proveito do corpo, não estaria ele desprezando a medicina enquanto disciplina mais adequada para tratar desse objeto? Lembra que o corpo subjetivo, sentido como um poder e também como um entrave tem tido algumas relações com o corpo tal como o saber médico o representa e trata. Mesmo nos dias de hoje, o corpo, segundo o povo, incorpora a ideologia da profissão médica das especialidades e é, por esse motivo, um corpo dividido. Ao admitir, como recupera Sandra Caponi do pensamento de Canguilhem, que, ao tomar como objeto o corpo subjetivo, uma coisa é tentar livrar-se da tutela da medicina e a outra é que o reconhecimento da saúde como verdade do corpo no sentido ontológico implica pensá-la do ponto de vista da ciência, ou seja, pensar a verdade do ponto de vista lógico. O fato de a saúde ser uma noção vulgar e questão filosófica, não impede que a mesma possa ser objeto de construção científica. Na perspectiva epistemológica de Bachelard e Bourdieu, entre outros, os conceitos científicos são construídos contra o senso comum, embora tendo de levar em conta essas “primeiras noções” Dentre os modos de ruptura com o senso comum está a análise lógica dos conceitos com o que a filosofia pode e deve contribuir. Como bem aponta Dina Czeresnia, sem a questão filosófica “não há como lidar com pontos obscuros que se apresentam quando se procura dialogar e fluir entre as diferentes dimensões que caracterizam a complexidade da saúde”. A crítica de Naomar, no capítulo sobre a “Holopatogênese”, à limitação da teoria de Canguilhem diante dos desenvolvimentos paradigmáticos da ciência contemporânea é bastante pertinente. Eu acrescentaria que a dificuldade de Canguilhem em encontrar uma disciplina científica que tomasse por objeto a saúde, estava no fato de que suas análises se baseavam na medicina experimental de Claude Bernard e na fisiologia de Starling do início do século XX. O filósofo-médico Canguilhem, tinha uma percepção da higiene como ela se constituiu no século XVIII, ou seja, como disciplina médica com conotações morais e que pretendia regular a vida dos indivíduos. Analisar tanto a “higiene” como a “saúde pública” como movimentos de recomposição das práticas sanitárias decorrentes de diferentes articulações entre sociedade e Estado no campo da saúde, como o fazem Paim & Almeida-Filho no livro A Crise da Saúde Pública e a Utopia da Saúde Coletiva, ajuda a entender os sentidos históricos tanto da higiene do século XVIII como da Saúde Pública do século XIX. Nessa perspectiva é que o uso do termo “saúde pública” como um conceito que se refere ao campo geral da saúde no âmbito coletivo não é adequado. A saúde passa a ser objeto da ciência exatamente com a constituição do campo da saúde coletiva enquanto campo científico e campo de práticas a partir RESENHAS REVIEWS da década de 70, na América Latina. Esse movimento é contemporâneo dos processos e fatos ocorridos em âmbito internacional, relacionados com a produção de conhecimento e práticas voltadas para a promoção da saúde, muito bem analisados por Paulo Buss no capítulo Uma Introdução ao Conceito de Saúde Pública. Trata-se de fato de um leitura obrigatória para quem desejar situar-se em relação à constituição histórica dos movimentos pela promoção da saúde. O enfoque político e técnico operado pelo movimento da promoção da saúde e ocorrido no plano internacional, tem muitas interseções com o movimento latino-americano de constituição da saúde coletiva, bem como com a Reforma Sanitária Brasileira, como Paim e Freitas revelam. As diversas conferências internacionais articulam o mundo acadêmico com aquele dos serviços de saúde e com os movimentos sociais. A publicação da coletânea Porque Algumas Pessoas são Saudáveis e Outras Não, coordenada por Evans em 1997, no Canadá, reflete esse esforço de pensar de forma multidisciplinar a saúde visando a ação social. Um outro tema visitado de formas diferentes pelos autores diz respeito ao conceito de risco. O risco aparece como “perigo”, como “probabilidade”, no sentido estatístico dado pela epidemiologia e como conceito “construcionista” em tipologia usada por Castiel. Nesse sentido, o alerta de Ayres é adequado na medida em que freqüentemente o conceito de risco é substituído acriticamente pelo de vulnerabilidade. Esse autor analisa o processo histórico de constituição do conceito de vulnerabilidade procurando evidenciar seu potencial para a redefinição das práticas de saúde. Associa o conceito de risco epidemiológico à definição do problema de saúde e à intervenção sobre o mesmo. Já Caponi discute, o que, na sua opinião, seria a “prom oção da saúde como abertura ao risco” na perspectiva de que a promoção da saúde deve ir para além do controle sanitário sobre os riscos individuais e mesmo aceitar a existência dos mesmos. Paim, por sua vez, considera como uma das características da proposta da Vigilância da Saúde a busca da operacionalização do conceito epidemiológico de risco. Esse movimento não seria contraditório com a reorganização das práticas sanitárias a partir da recomposição do processo de trabalho, das suas relações sociais e técnicas e das relações interpessoais entre agentes e usuários, de forma a captar as especificidades culturais e as dimensões subjetiva e singular do processo saúde-doença nos sujeitos concretos. As relações entre os indivíduos e a sociedade também são retomadas sob diversos ângulos. Buss considera como conservadora a perspectiva da promoção da saúde que remete para os indivíduos a responsabilidade de tomarem conta de si próprios, em contraposição à perspectiva progressista que enfatiza a responsabilidade do Estado na implementação de políticas públicas saudáveis. Já Czeresnia, considera que um dos aspectos fundamentais da idéia da promoção é o estimulo à autonomia e problematiza este conceito nas suas relações com as representações científicas e culturais do conceito de risco. Há ainda a preocupação expressada por Caponi de que os programas de promoção da saúde reproduzem os programas higienistas e sua ambição de controle sobre a vida dos indivíduos. Para Castiel, a noção individualista de “identidade” ancora-se na tradição filosófica ocidental, ao contrário daquela dominante nos povos e culturas não-ocidentais onde a singularidade dos indivíduos é vista como envolvida em redes sociais. Esse autor, recorrendo ao mito e à metáfora como estratégia de análise, problematiza a dimensão individual das propostas da promoção à saúde ao interior das complexas redes sociais existentes no terreno da comunicação nas diferentes culturas e espaços sociais. Nesse sentido, estabelece um diálogo com a formulação de Norbert Elias, desenvolvida no seu livro A Sociedade dos Indivíduos, segundo a qual, a história é sempre a história de uma sociedade, mas uma sociedade de indivíduos. Essa formulação nos ajuda a refletir sobre a aparente contradição entre o indivíduo e a sociedade. Ou seja, que as possibilidades de manobra dos indivíduos dependem da estrutura e das características históricas da sociedade na qual eles vivem. Esse debate remete às teorias sobre as práticas sociais: se movidas pela ação racional, se determinadas pelas estruturas ou se influenciadas por disposições, produto da história incorporada nos inconscientes – o hab itus no sentido de Bourdieu? Há que se buscar integrar diferentes perspectivas na compreensão de objeto tão complexo. Assim, indivíduo e sociedade se articulam e se inter-relacionam numa sociedade dos indivíduos. Da mesma forma ocorre com o processo saúde-doença nas suas manifestações singular e coletiva, e com as estratégias voltadas para a promoção da saúde. Essa dupla dimensão do fenômeno em tela impõe desafios teóricos e práticos para a saúde coletiva – campo onde a tensão e complementaridade entre a teoria e a prática, entre a ciência e a política estão sempre presentes e para a qual o livro Promoção da Saúde: Conceitos, Reflexões e Tendência, aporta significativa contribuição. Ligia Maria Vieira-da-Silva Instituto de Saúde Coletiva, Universidade Federal da Bahia, Salvador, Brasil. Cad. Saúde Pública, Rio de Janeiro, 19(6):1893-1897, nov-dez, 2003 1897
https://openalex.org/W4285362224	https://www.vestnik-grekova.ru/jour/article/download/1878/1310	Russian	null	Locoregional methods of treatment of patients with extrahepatic bile duct cholangiocarcinoma (review of literature)	Vestnik hirurgii im. I.I. Grekova/Vestnik hirurgii imeni I.I. Grekova	2,021	cc-by	7,348	ОБЗОРЫ / REVIES ОБЗОРЫ / REVIES ОБЗОРЫ / REVIES © СС Коллектив авторов, 2021 УДК 616.367-006.6-08 (047) DOI: 10.24884/0042-4625-2021-180-5-111-117 * Автор для связи: Дарья Юрьевна Евстратьева, Военно-медицинская академия им. С. М. Кирова, 194044, Россия, Санкт-Петербург, ул. Академика Лебедева, д. 6. E-mail: plotnikowadarja@yandex.ru. для связи: Дарья Юрьевна Евстратьева, Военно-медицинская академия им. С. М. Кирова, 194044 Санкт-Петербург, ул. Академика Лебедева, д. 6. E-mail: plotnikowadarja@yandex.ru. Locoregional methods of treatment of patients with extrahepatic bile duct cholangiocarcinoma (review of literature) Sergey Ya. Ivanusa, Sergey A. Alentev, Daria Yu. Evstrateva, Alexander A. Molchanov Sergey Ya. Ivanusa, Sergey A. Alentev, Daria Yu. Evstrateva, Alexander A. Ivanusa, Sergey A. Alentev, Daria Yu. Evstrateva, Alexander A. Molchanov Military Medical Academy, Saint Petersburg, Russia Military Medical Academy, Saint Petersburg, Russia Military Medical Academy, Saint Petersburg, Russia Локорегионарные методы лечения больных с холангиокарциномой внепеченочных желчных протоков (обзор литературы) С. Я. Ивануса, С. А. Алентьев, Д. Ю. Евстратьева, А. А. Молчанов Федеральное государственное бюджетное военное образовательное учреждение высшего образования «Военно-медицинская академия имени С. М. Кирова» Министерства обороны Российской Федерации, Санкт-Петербург, Россия Поступила в редакцию 04.06.2021 г.; принята к печати 01.12.2021 г. Поступила в редакцию 04.06.2021 г.; принята к печати 01.12.2021 г. Представлен литературный обзор исследований, посвященных применению методов локальной деструкции (фото- динамической терапии и радиочастотной абляции) в лечении больных с нерезектабельной внепеченочной холангиокарциномой. В настоящее время опубликовано множество исследований, посвященных оценке фото- динамической терапии в лечении пациентов данной категории. Радиочастотная абляция при опухолях внепече- ночных желчных протоков применяется относительно недавно, но, по имеющимся данным, может представлять альтернативу фотодинамической терапии. В статье рассмотрены механизм действия, методика фотодинамической терапии и радиочастотной абляции, приведены результаты наиболее крупных исследований. Сообщается, что по результатам выживаемости фотодинамическая терапия сопоставима с R1/R2-резекцией. Имеются данные об успешном использовании ее в качестве адъювантного и неоадъювантного лечения, о возможности применения фотодинамической терапии в качестве метода локального контроля опухоли у пациентов с нерезектабельной холангиокарциномой, ожидающих трансплантации печени. Применение радиочастотной абляции на первом этапе, перед стентированием желчевыводящих путей, позволяет увеличить диаметр желчных протоков, способствует увеличению времени проходимости стента, а также выживаемости больных. Кроме того, радиочастотная абляция и фотодинамическая терапия могут успешно применяться при окклюзии ранее установленных стентов. Преиму- ществами фотодинамической терапии является возможность проведения ее у ослабленных больных, с повышен- ным уровнем билирубина сыворотки крови. К преимуществам радиочастотной абляции можно отнести меньшую стоимость процедуры, а также отсутствие необходимости соблюдения светового режима. Изучаются перспективы использования методов локорегионарной деструкции в составе комбинированных схем лечения (в сочетании с системной или регионарной химиотерапией), что позволяет достичь наилучших результатов. Ключевые слова: холангиоцеллюлярный рак гилюсная холангиокарцинома эндобилиарные вмешательства фото Ключевые слова: холангиоцеллюлярный рак, гилюсная холангиокарцинома, эндобилиарные вмешательства, фото- динамическая терапия, радиочастотная абляция Для цитирования: Ивануса С. Я., Алентьев С. А., Евстратьева Д. Ю., Молчанов А. А. Локорегионарные методы лечения больных с холангиокарциномой внепеченочных желчных протоков (обзор литературы). Вестник хирургии имени И. И. Грекова. 2021;180(5):111–117. DOI: 10.24884/0042-4625-2021-180-5-111-117. * Автор для связи: Дарья Юрьевна Евстратьева, Военно-медицинская академия им. С. М. Кирова, 194044, Россия, Санкт-Петербург, ул. Академика Лебедева, д. 6. E-mail: plotnikowadarja@yandex.ru. Received 04.06.2021; accepted 01.12.2021 Дозу световой энергии устанавливают в зави- симости от объема поражения протоков, обычно она состав- ляет 180–250 Дж/см2, длительность облучения рассчитывают исходя из заданной величины световой дозы и плотности мощ- ности излучения. Однако ФТ первого поколения свойственен ряд недостатков: неоптимальный спектр поглощения (вблизи λ=630 нм) и, как следствие, малая глубина проникновения света (3–5 мм), недостаточная избирательность накопления, длительное удерживание препарата в здоровых тканях и продолжительный период выведения, что обуславливает их высокую фототоксичность и требует длительного соблюдения светового режима. Перечисленных недостатков лишены ФТ второго поколения: Радахлорин (Россия), Фотолон (Беларусь), Фотодитазин (Россия), Темопорфирин/Фоскан (Германия). Препараты активируются светом в длинноволновой области спектра (λ=650–680 нм), который глубже проникает в ткани (7–8 мм). Более быстрое выведение препарата из организма (в течение 48 ч) обуславливает небольшой срок соблюдения светового режима пациентами [10]. Фотоактивацию производят светом с длиной волны 630 нм (красный свет). Дозу световой энергии устанавливают в зави- симости от объема поражения протоков, обычно она состав- ляет 180–250 Дж/см2, длительность облучения рассчитывают исходя из заданной величины световой дозы и плотности мощ- ности излучения. Однако ФТ первого поколения свойственен ряд недостатков: неоптимальный спектр поглощения (вблизи λ=630 нм) и, как следствие, малая глубина проникновения света (3–5 мм), недостаточная избирательность накопления, длительное удерживание препарата в здоровых тканях и продолжительный период выведения, что обуславливает их высокую фототоксичность и требует длительного соблюдения светового режима. Перечисленных недостатков лишены ФТ второго поколения: Радахлорин (Россия), Фотолон (Беларусь), Фотодитазин (Россия), Темопорфирин/Фоскан (Германия). Препараты активируются светом в длинноволновой области спектра (λ=650–680 нм), который глубже проникает в ткани (7–8 мм). Более быстрое выведение препарата из организма (в течение 48 ч) обуславливает небольшой срок соблюдения светового режима пациентами [10]. Лечение протоковой холангиокарциномы является акту- альной и сложной проблемой современной гепатопанкреато- билиарной хирургии. Среди всех злокачественных новооб- разований человека рак внепеченочных желчных протоков составляет от 2 до 4,5 %, при этом до 60–80 % из этого числа составляет опухоль Клацкина (опухоль проксимальных внепе- ченочных желчных протоков, воротная, гилюсная или хиляр- ная холангиокарцинома). За последние годы отмечается рост заболеваемости холангиоцеллюлярным раком. Ежегодно диаг ностируется приблизительно 7500 новых случаев протоковой холангиокарциномы [1, 2]. Единственным методом, позволяющим существенно уве- личить отдаленную выживаемость больных холангиоцеллю- лярным раком, является радикальная резекция [2–4]. Однако на момент постановки диагноза выполнение операции возможно не более чем у 30 % больных [3–5]. Кроме того, рецидив опу- холи после хирургического вмешательства достигает 50 % [3]. Таким образом, бóльшая часть пациентов нуждаются в палли- ативном лечении. Received 04.06.2021; accepted 01.12.2021 Стандартным вариантом терапии больных с опухолями внепеченочных желчных протоков являются раз- личные виды дренирования желчевыводящих путей с после- дующей имплантацией билиарных стентов (чрескожно или эндоскопически) [5]. Однако данные вмешательства не влияют на опухоль и часто сопровождаются рецидивом механической желтухи. Медиана выживаемости при этом составляет от 3 до 10 месяцев [6]. Учитывая локализацию опухоли, ее близ- кое расположение к сосудистым структурам ворот печени, а также относительно редкое метастазирование, перспективным представляется использование локорегионарных методов лече- ния. Наиболее изученными, показавшими обнадеживающие результаты, а также сопоставимыми по числу осложнений и выживаемости методами лечения являются фотодинамическая терапия (ФДТ) и радиочастотная абляция (РЧА). Для облучения опухоли используются гибкие оптические волокна (световоды), которые подсоединяются к источнику лазерного излучения. Световод, имеющий на конце цилиндри- ческий диффузор (концевая часть, обеспечивающая засветку пораженного участка), помещается внутрь катетера, содержа- щего рентгеноконтрастные метки на конце, и под рентгеноско- пическим контролем подводится к месту опухолевой стрик- туры. При облучении опухоли следует избегать повторной обработки полей. Источник света к опухоли может подводиться как с исполь- зованием чрескожного чреспеченочного доступа, так и эндо- скопически. При необходимости возможно проведение бал- лонной дилатации опухолевой стриктуры. В обоих случаях может быть использована холангиоскопия. На завершающем этапе стандартным считается выполнение дренирования или стентирования желчных протоков. Рекомендуется размещать пластиковые стенты, во время замены которых возможно выполнение повторных сеансов ФДТ. Однако имеются также сообщения о том, что ФДТ может безопасно и эффективно проводиться и через установленный ранее металлический саморасширяющийся стент [11]. ФДТ предполагает избирательное накопление фотосенси- билизатора (ФТ) в опухолевой ткани с последующей его акти- вацией путем локального облучения светом с определенной длиной волны. При этом в тканях происходят фотохимические реакции между ФТ и кислородом, приводящие к образованию в опухолевой ткани и (или) в сосудах опухоли цитотоксических агентов (прежде всего, активных форм кислорода), разруша- ющих опухолевые клетки [8]. Отчет о первом успешном применении ФДТ был опубли- кован в 1991 г. J. S. Jr McCaughan et al. [12]. Пациентке с нерезектабельной аденокарциномой общего желчного протока было выполнено 7 сеансов эндоскопической ФДТ, выживае- мость составила 4,5 года. Наиболее часто используемыми ФТ в лечении холангио карциномы являются фотодинамические агенты первого поколения (Фотофрин (США), Фотосан (Германия), Фотогем (Россия)), представляющие собой смесь мономерных и оли- гомерных производных гематопорфирина. Препарат вводят внутривенно в стандартной дозировке – 2 мг/кг массы тела, за 24–48 ч до оперативного вмешательства, за это время про- исходит максимальное накопление ФТ в опухолевой ткани [9]. Первое рандомизированное исследование было проведено в Германии в 2003 г. M. Ortner et al. [13]. Received 04.06.2021; accepted 01.12.2021 The article presents a literature review of studies on the use of local destruction methods (photodynamic therapy and radiofrequency ablation) in the treatment of patients with unresectable extrahepatic cholangiocarcinoma. Currently, many studies have been published on the assessment of photodynamic therapy in the treatment of patients of this category. Radiofrequency ablation in tumors of extrahepatic bile ducts has been used relatively recently, but, according to avail- able data, it may represent an alternative to photodynamic therapy. The article discusses the mechanism of action, the method of photodynamic therapy and radiofrequency ablation, and presents the results of the largest studies. It is reported that photodynamic therapy is comparable to R1/R2 resection according to the survival results. There is 111 Ивануса С. Я. и др. «Вестник хирургии» • 2021 • Том 180 • № 5 • С. 111–117 evidence of its successful use as adjuvant and neoadjuvant treatment, the possibility of using photodynamic therapy as a method of local tumor control in patients with unresectable cholangiocarcinoma awaiting liver transplantation. The use of radiofrequency ablation at the first stage before stenting the biliary tract allows to increase the diameter of the bile ducts, increases the stent patency time, as well as the survival orate f patients. In addition, radiofrequency ablation and photodynamic therapy can be successfully applied to occlude previously inserted stents. The advantage of photodynamic therapy is the possibility of using it in weakened patients with hyper bilirubinemia. The advantages of radiofrequency ablation include the lower cost of the procedure, as well as the absence of the need to keep the light regimen. The prospects of using methods of locoregional destruction as part of combined treatment regimens (in combination with systemic or regional chemotherapy) are being studied, which allows achieving the best results. Keywords: cholangiocellular carcinoma, hilus cholangiocarcinoma, endobiliary intervention, photodynamic therapy, ra- diofrequency ablation For citation: Ivanusa S. Ya., Alentev S. A., Evstrateva D. Yu., Molchanov A. A. Locoregional methods of treat- ment of patients with extrahepatic bile duct cholangiocarcinoma (review of literature). Grekov’s Bulletin of Surgery. 2021;180(5):111–117. (In Russ.). DOI: 10.24884/0042-4625-2021-180-5-111-117. * Corresponding author: Daria Yu. Evstrateva, Military Medical Academy, 6, Academika Lebedeva str., Saint Peters- burg, 194044, Russia. E-mail: plotnikowadarja@yandex.ru. Фотоактивацию производят светом с длиной волны 630 нм (красный свет). Received 04.06.2021; accepted 01.12.2021 Кроме того, при использовании Фоскана отмечена максимальная глубина поражения опухоли (>7,5 мм). Частота осложнений не отличалась. Имеются исследования, в которых авторы оценивали результаты ФДТ в зависимости от доступа, а также с или без использования холангиоскопии. Сообщается, что холангио- скопия способствует более целенаправленному, однородному облучению опухоли, но не влияет на результаты лечения [20]. В 2005 г. T. Zoepf et al. [14] было проведено еще одно рандо- мизированное исследование. Тридцать два пациента с нерезек- табельной холангиокарциномой внепеченочных желчных про- токов были разделены на две группы: 16 больных – ФДТ+стент и 16 – только стентирование. Девяти пациентам выполняли повторные сеансы ФДТ спустя 3–9 месяцев, 1 больному про- вели три сеанса ФДТ (третий – через 6 месяцев после второго). Спустя 4 месяца при контрольной холангиографии выявлялась реканализация желчных протоков в обработанной области. В 1-й группе больных билирубин сыворотки крови снизился с 2,75 до 1,3 мг/дл. Медиана выживаемости была значительно больше в группе ФДT (21 месяц против 7 месяцев, p=0,01). Случаев холангита отмечено больше в 1-й группе (4 против 1), но отличие не было статистически значимым (p=0,166). Таким образом, авторы делают предположение о том, что ФДТ имеет потенциал изменить все настоящие стандарты паллиативного лечения нерезектабельной холангиокарциномы. Наибольшая эффективность ФДТ наблюдается при исполь- зовании ФС последних поколений. Так, A. Wagner et al. [21] сравнили ФДТ с использованием Темопорфина (Фоскана) и Фотофрина в лечении больных с нерезектабельной холан- гиокарциномой. Медиана выживаемости составила 15,4 и 9,3 месяца, 6-месячная выживаемость – 83 и 70 % соответ- ственно. Кроме того, при использовании Фоскана отмечена максимальная глубина поражения опухоли (>7,5 мм). Частота осложнений не отличалась. Перспективным направлением, позволяющим достичь наи- лучших результатов ФДТ, является комбинация ее с другими методами. По результатам ретроспективного исследования M. J. Hong et al. [22], медиана выживаемости при сочетании ФДТ с системной ХТ составила 17,9 против 11,1 месяца при выполне- нии только ФДТ; 1-, 2- и 3-летняя выживаемость составила 40, 17 и 3 % в группе ФДТ и 93, 16 и 0 % соответственно при ФДТ с системной ХТ (Гемцитабин или Гемцитабин + Цисплатин). Еще в одном более позднем исследовании [23] со сходным дизайном пациенты, которым выполнялась ФДТ с системной ХТ (n=35), имели преимущество в выживаемости почти в 5 месяцев по сравнению с больными, получавшими только ФДТ (n=33), – 374 дня против 520 дней, 1-летняя выживаемость – 88 против 58 % (p=0,001). Кроме того, авторы провели анализ выживаемости больных в подгруппах в зависимости от схемы ХТ. Received 04.06.2021; accepted 01.12.2021 Всего в исследова- нии приняли участие 39 больных. Первую группу составили 112 Ivanusa S. Ya. et al. «Grekov’s Bulletin of Surgery» • 2021 • Vol. 180 • № 5 • P. 111–117 вании A. Höblinger et al. [17] были проанализированы 10 паци- ентов с нерезектабельной гилюсной холангиокарциномой, которым проводились множественные сеансы ФДТ (в среднем 8, диапазон 4–14), при этом 3 больных имели метастатиче- ское поражение. Повторные сеансы ФДТ выполняли через регулярные интервалы в 1–2 месяца (n=6) либо досрочно при прогрессировании заболевания (n=4). Процедура проводилась эндоскопически у всех пациентов. Из осложнений отмечены холангит (n=2, 20 %) и реакция фототоксичности у 1 больного после седьмого сеанса ФДТ. Медиана общей выживаемости не была достигнута, расчетная выживаемость всех пациентов составила 47,6 месяца. пациенты, которым выполнялась ФДТ со стентированием (n=19), вторую – больные, получающие только стентирова- ние (n=20). Для ФДТ применяли Фотофрин в стандартной дозировке. Медиана выживаемости составила 493 дня против 98 дней (P<0,0001). Уровень билирубина и качество жизни при выполнении ФДТ также улучшались, в отличие от группы стен- тирования. Частота осложнений была сопоставимой, кроме того, холангит чаще возникал во второй группе (7 против 3). Приводятся также результаты лечения пациентов, которые не вошли ни в одну из групп по причине несоответствия крите- риям включения (индекс Карновского <30 %, n=9; отказ от рандомизации, n=7; предыдущая химио- или лучевая терапия, n=5; частичная резекция, n=5). Все пациенты в качестве лече- ния выбрали стентирование с ФДТ. Медиана выживаемости у этих больных составила 426 дней, что выше, чем у пациен- тов, получающих стентирование. Авторы делают вывод о том, что ФДТ эффективна, в том числе и у ослабленных больных. Исследование было прекращено преждевременно, так как дальнейшая рандомизация считалась неэтичной. Помимо этого, прогностическими факторами, связанными с длительной выживаемостью, являются более короткий пери- од времени от момента постановки диагноза до выполнения процедуры, а также невысокий уровень билирубина перед ФДТ. Наоборот, низкий уровень альбумина сыворотки крови связан с меньшей выживаемостью пациентов [18, 19]. Имеются исследования, в которых авторы оценивали результаты ФДТ в зависимости от доступа, а также с или без использования холангиоскопии. Сообщается, что холангио- скопия способствует более целенаправленному, однородному облучению опухоли, но не влияет на результаты лечения [20]. Наибольшая эффективность ФДТ наблюдается при исполь- зовании ФС последних поколений. Так, A. Wagner et al. [21] сравнили ФДТ с использованием Темопорфина (Фоскана) и Фотофрина в лечении больных с нерезектабельной холан- гиокарциномой. Медиана выживаемости составила 15,4 и 9,3 месяца, 6-месячная выживаемость – 83 и 70 % соответ- ственно. Ивануса С. Я. и др. Данная конструкция обеспечивает цилиндрическую абляцию на рас- стоянии 24 мм между дистальным и проксимальным полями двух электродов. Катетер совместим со многими общедоступ- ными электрохирургическими генераторами и эндоскопами с рабочим каналом 3,2 мм или более. Настройка генератора включает мощность (оптимальной является от 7 до 10 Вт). Рекомендуемое время воздействия на всех участках стрикту- ры – не более 2 мин [30]. Коллектив авторов во главе с N. D. Cosgrove [27] пишут о возможности применения ФДТ в качестве метода локально- го контроля опухоли у пациентов с нерезектабельной холан- гиокарциномой, ожидающих трансплантации печени (ТП). Четырем больным после проведения ХТ выполнялась ФДТ. Среднее время от ФДТ до ТП составило 4 месяца. Ни у одного пациента не было отмечено прогрессирования опухоли за этот период. Выживаемость без прогрессирования у 3 пациентов после ТП составила 28,1 месяца (диапазон – 9,3–82,7 месяца). H. Witzigmann et al. [4] проанализировали проспективные данные лечения и исхода 184 пациентов с гилюсной холан- гиокарциномой, которым выполнялась резекция печени и желчных протоков, либо стентирование с ФДТ, либо только стентирование. Как и ожидалось, наилучшую выживаемость имели больные, у которых была достигнута R0-резекция (медиана – 33,1 месяца). Однако достичь отрицательных краев резекции удалось лишь в трети случаев. Пациенты с R1/R2 показали выживаемость, аналогичную тем, которые получали ФДТ со стентированием (12,2 и 12,0 месяца соответственно). При этом в группе ФДТ, по сравнению с 1-й группой, отмеча- лась значительно более низкая частота осложнений. В группе стентирования медиана выживаемости составила 6,4 месяца. RF-катетер ELRA позволяет контролировать температуру на границе раздела «ткань – электрод», состоит из 7 Fr (2,3 мм) катетера длиной 175 см, способного выдержать острый угол при отклонении в нижней части рабочего канала дуодено- скопа. Он содержит четыре биполярных электрода, которые обеспечивают линейную абляцию, в зависимости от длины которой выделяют два размера катетеров (18 и 33 мм). В каче- стве РЧ-источника применяются генераторы VIVA (Taewoong Medical, Южная Корея), настроенные в основном на двухми- нутный интервал, максимальную температуру 80 °C и мощ- ность 10 Вт. Сравнительные исследования между катетерами отсутствуют. В 2016 г. A. Schmidt et al. [28] провели сравнительное исследование ФДТ и РЧА при гилюсной холангиокарциноме. Первую группу составили 14 пациентов, которым проводилась РЧА, всего 31 сеанс (до 7 процедур на 1 больного, с интервалом от 2 до 3 месяцев). Второй группе (20 пациентов) выполнялась внутрипротоковая ФДТ, общее число сеансов ФДТ – 36 (мак- симально 5 на 1 пациента). Ивануса С. Я. и др. «Вестник хирургии» • 2021 • Том 180 • № 5 • С. 111–117 Ряд авторов предлагают использовать ФДТ в адъювант- ном и неоадъювантном режимах. В пилотном исследовании II фазы [3] 7 пациентам с нерезектабельной воротной холан- гиокарциномой выполнялся 1 либо 2 сеанса ФДТ. При гисто- логическом исследовании наблюдался положительный ответ опухоли в виде поверхностного некроза новообразования и отсутствия опухолевых клеток на глубину до 4 мм. В после- дующем проведено оперативное лечение, в 100 % случа- ев достигнута R0-резекция. Медиана времени до операции составила 6 недель. Опухоль рецидивировала у 2 пациентов через 6 и 19 месяцев. Показатель 1-летней выживаемости без прогрессирования составил 83 %. A. Nanashima et al. [26] сооб- щают об успешном использовании ФДТ у пациентов после выполнения нерадикальной резекции. Выявлялось заметное уменьшение опухоли, реканализация протоков после ФДТ. Прогрессирование заболевания отмечено через 6–31 месяц. токов применяется относительно недавно. Проведение ее стало возможным в связи с появлением специального оборудования. При РЧА происходит коагуляционный некроз опухоли посредством теплового повреждения тканей, вызванного переменным электрическим током, подводимым к опухоли при помощи специальных электродов. РЧА выполняется как эндоскопически, так и чрескожно. В отличие от ФДТ, при РЧА необходим непосредственный контакт электрода с опухолевой тканью. После выполнения абляции производится удаление некротически измененных тканей при помощи баллонного катетера, в обработанную зону устанавливается стент [30]. В настоящее время двумя коммерчески доступными устройствами для внутрипротоковой РЧА являются катетер Habib™ EndoHPB (Boston Scientific Corp., Marlborough MA, США) и катетер ELRA™ RF (Taewoong Medical, Южная Корея). Устройство Habib EndoHPB состоит из 8 Fr катетера (2,6 мм) длиной 180 см. На дистальном конце катетера находят- ся два циркулярных электрода из нержавеющей стали шириной 8 мм, разделенных неактивным расстоянием в 8 мм. Данная конструкция обеспечивает цилиндрическую абляцию на рас- стоянии 24 мм между дистальным и проксимальным полями двух электродов. Катетер совместим со многими общедоступ- ными электрохирургическими генераторами и эндоскопами с рабочим каналом 3,2 мм или более. Настройка генератора включает мощность (оптимальной является от 7 до 10 Вт). Рекомендуемое время воздействия на всех участках стрикту- ры – не более 2 мин [30]. В настоящее время двумя коммерчески доступными устройствами для внутрипротоковой РЧА являются катетер Habib™ EndoHPB (Boston Scientific Corp., Marlborough MA, США) и катетер ELRA™ RF (Taewoong Medical, Южная Корея). Устройство Habib EndoHPB состоит из 8 Fr катетера (2,6 мм) длиной 180 см. На дистальном конце катетера находят- ся два циркулярных электрода из нержавеющей стали шириной 8 мм, разделенных неактивным расстоянием в 8 мм. Received 04.06.2021; accepted 01.12.2021 Более эффек- тивной являлась комбинированная ХТ на основе Гемцитабина (Гемцитабин + Цисплатин или Оксалиплатин, Гемцитабин + Капецитабин) по сравнению с монотерапией Гемцитабином. В 2016 г. было проведено рандомизированное исследование [24] с участием двух групп пациентов: группа А – ФДТ + стентиро- вание + системная ХТ (GemCap), группа В – стентирование + системная ХТ (GemCap). Медиана выживаемости без прогрес- сирования составила 175 дней в группе А и 96 дней в группе В. Различий по числу осложнений между группами не было. р р В 2018 г. были опубликованы результаты рандоми- зированного многоцентрового исследования III фазы, PHOTOSTENT-02, проведенного в Великобритании [15]. Оценивалась эффективность и безопасность ФДТ со стенти- рованием (группа 1) по сравнению со стентированием желче- выводящих путей (группа 2) при местно-распространенном или метастатическом раке желчных протоков (как внутрипе- ченочных, так и внепеченочных холангиокарциномах). Всего в исследование были включены 92 пациента из восьми центров. Были получены неожиданные результаты, противоречащие всем предыдущим исследованиям. Авторы сообщают, что общая выживаемость была хуже в группе ФДТ по сравнению с группой стентирования (медиана выживаемости: 6,2 про- тив 9,8 месяца). Среди пациентов с ФДТ наблюдалась также более низкая выживаемость без прогрессирования (медиана 3,4 против 4,3 месяца). Частота и степень осложнений были сопоставимы. Полученные результаты авторы объясняют раз- личиями в последующем лечении. Во 2-й группе химиотера- пию (ХТ) получили 24 пациента (52 %) против 13 пациентов (28 %) из группы ФДТ. При этом больные из группы стен- тирования начинали получать ее раньше. ХТ проводили по схеме CisGem, которая является международным стандартом для лечения этого заболевания. Имеются сообщения об эффективности совместного при- менения методов локорегионарного лечения – внутрипрото- ковой ФДТ и регионарной химиотерапии (РХТ). Д. А. Гранов и др. [25] приводят данные лечения 55 пациентов с первич- но нерезектабельной опухолью Клацкина. Всего выполнено 318 сеансов внутрипротоковой ФДТ и 243 цикла РХТ. В 4 слу- чаях стало возможным выполнение радикального оперативно- го вмешательства. Медиана общей выживаемости пациентов составила 20,3 месяца. Однако на данный момент все остальные исследования, а также метаанализы [16] демонстрируют значимое увеличение выживаемости у пациентов, получающих ФДТ, по сравнению со стандартными методами декомпрессии желчных протоков. При этом наибольшую выживаемость имеют больные, полу- чающие повторные сеансы ФДТ. В ретроспективном исследо- 113 1. Макаров Е. С. и др. Рак проксимальных желчных протоков. М. : Печатка, 2018. 100 с. 2. Майстренко Н. А. и др. Холангиоцеллюлярный рак (особенности диагностики и лечения) // Практ. онкология. 2008. № 4. С. 229–236. 3. Wiedmann M. et al. Neoadjuvant photodynamic therapy as a new ap- proach to treating hilar cholangiocarcinoma : a phase II pilot study // Cancer. 2003. Vol. 97, № 11. P. 2783–2790. 3. Wiedmann M. et al. Neoadjuvant photodynamic therapy as a new ap- proach to treating hilar cholangiocarcinoma : a phase II pilot study // Cancer. 2003. Vol. 97, № 11. P. 2783–2790. 4. Witzigmann H. et al. Surgical and palliative management and outcome in 184 patients with hilar cholangiocarcinoma : palliative photodynamic therapy plus stenting is comparable to r1/r2 resection // Ann. Surg. 2006. Vol. 244, № 2. P. 230–239. 4. Witzigmann H. et al. Surgical and palliative management and outcome in 184 patients with hilar cholangiocarcinoma : palliative photodynamic therapy plus stenting is comparable to r1/r2 resection // Ann. Surg. 2006. Vol. 244, № 2. P. 230–239. РЧА может являться альтернативой традиционной техники «стент в стент» при прогрессировании заболевания. Первое сообщение об успешном использовании эндобилиарной РЧА у пациентов с окклюзированными саморасширяющимися метал- лическими стентами принадлежит J. Pozsa´r et al. [37]. Всего были пролечены 5 больных, 2 имели полностью покрытые стенты, 3 – непокрытые, со средним временем функциониро- вания 105 (17–240) дней. Внутрипротоковую РЧА проводили во всех случаях эндоскопически. Диаметр желчного протока в месте стриктуры увеличился в среднем с 2 до 4,7 мм, длина стриктуры уменьшалась с 15 до 10,6 мм. Проходимость стента после РЧА составила 62 (9–236) дня. 5. Goenka M. K., Goenka U. Palliation : Hilar cholangiocarcinoma // World J. Hepatol. 2014. Vol. 39, № 6. P. 559–569. 6. Farley D. R., Weaver A. L., Nagorney D. M. «Natural history» of unre- sected cholangiocarcinoma : patient outcome after noncurative interven- tion // Mayo Clin Proc. 1995. Vol. 70, № 5. P. 425–429. 7. Jarnagin W. R. et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma // Ann Surg. 2001. Vol. 234, № 4. P. 507–517. 8. Лапцевич Т. П. и др. Фотодинамическая терапия злокачественных опухолей : основы, история развития, перспективы (обзор) // Онколог. журн. 2008. № 1. С. 117–138. 9. Pahernik S. A. et al. Distribution and pharmacokinetics of Photofrin in human bile duct cancer // J. Photochem. Photobiol. 1998. Vol. 47, № 1. P. 58–62. A. Kadayifci et al. Compliance with ethical principles The authors confirm that they respect the rights of the people participated in the study, including obtaining informed consent when it is necessary, and the rules of treatment of animals when they are used in the study. Author Guidelines contains the detailed information. Ивануса С. Я. и др. Через 14 дней после проведения лечения в группе РЧА отмечено значимое снижение уровня билирубина (в среднем с 3,3 до 2,3 мг/дл), в отличие от группы ФДТ (в среднем с 4,1 до 3,5 мг/дл). Кроме того, в группе ФДТ наблюдалось бóльшее число преждевременных замен стента после первого вмешательства (13 из 20, 65 %) по сравнению с группой РЧА (4 из 14, 29 %). Также у пациентов, получающих ФДТ, отмечено несколько бóльшее число осложнений (у 40 %), чем у больных, которым проводилась РЧА (21 %), при этом серьезных осложнений не было ни в одной из групп. Сделан вывод о том, что РЧА может представлять альтернативу ФДТ. Ввиду малого числа больных, исследования по примене- нию РЧА при механической желтухе опухолевой этиологии чаще всего включают в себя опухоли различных локализаций – внутри- и внепеченочные холангиокарциномы, рак поджелу- дочной железы, рак желчного пузыря. При этом большинство исследователей отмечают увеличение диаметра желчного про- тока в обработанной зоне после РЧА, увеличение времени функционирования стента и общей выживаемости. Первое сообщение о применении РЧА для лечения нерезек- табельных опухолей желчных протоков проведено A. W. Steel et al. [31] в 2011 г. с включением 22 пациентов (16 – рак под- желудочной железы и 6 – холангиокарцинома внепеченочных желчных протоков). РЧА выполняли эндоскопически с после- дующей установкой металлического стента. Процедура была успешно проведена 21 больному. Отмечены нормализация уровня билирубина, восстановление проходимости желчных протоков, сохраняющееся до 90 дней. Долгосрочные результаты ФДТ и РЧА были оценены D. S. Strand et al. [29]. Выживаемость после эндоскопической РЧА (n=16) и ФДТ (n=32) статистически не различалась. Медиана выживаемости составила 9,6 и 7,5 месяца соответственно. В 2018 г. было опубликовано рандомизированное контро- лируемое исследование [32], включающее в себя 65 пациен- тов с опухолью Клацкина I и II типа (n=19) или дистальной холангиокарциномой (n=46). Больные распределялись между двумя группами: РЧА + пластиковый стент (n=32) и только пластиковый стент (n=33). В обеих группах наблюдались схо- жие осложнения, в то время как проходимость стента была в 2 раза больше в группе РЧА (6,8 против 3,4 месяца). Медиана выживаемости также была значительно выше в группе РЧА Метод РЧА на протяжении многих лет широко применя- ется в лечении больных с первичным или метастатическим раком печени. Технология хорошо изучена, накопленный опыт свидетельствует об ее высокой эффективности и безопасности у таких больных. Однако внутрипротоковая РЧА при нерезек- табельной холангиокарциноме внепеченочных желчных про- 114 Ivanusa S. Ya. et al. «Grekov’s Bulletin of Surgery» • 2021 • Vol. 180 • № 5 • P. Соответствие нормам этики Авторы подтверждают, что соблюдены права людей, принимав- ших участие в исследовании, включая получение информированного согласия в тех случаях, когда оно необходимо, и правила обращения с животными в случаях их использования в работе. Подробная инфор- мация содержится в Правилах для авторов. Y. Wang et al. [35] провели ретроспективный анализ лече- ния больных с нерезектабельной гилюсной холангиокарцино- мой III и IV типа по Bismuth. РЧА была успешно выполнена с использованием чрескожного чреспеченочного доступа. Технический успех составил 100 %. Одному пациенту, в связи с окклюзией стента, через 252 дня после процедуры была выполнена повторная абляция. Медиана проходимости стента с момента первой РЧА и выживаемость с момента постановки диагноза составили 100 дней и 5,3 месяца соответственно. T. T. Wu et al. [36] также сообщают о своем опыте лечения 47 пациентов, которым было проведено 65 процедур чрескож- ной РЧА с размещением саморасширяющихся металлических стентов. Значимо сниженные уровня билирубина наблюдали на 7-й день после операции. Медиана проходимости стента составила 149 дней (15–281). Медиана выживаемости – 181 (15–495) день. Y. Wang et al. [35] провели ретроспективный анализ лече- ния больных с нерезектабельной гилюсной холангиокарцино- мой III и IV типа по Bismuth. РЧА была успешно выполнена с использованием чрескожного чреспеченочного доступа. Технический успех составил 100 %. Одному пациенту, в связи с окклюзией стента, через 252 дня после процедуры была выполнена повторная абляция. Медиана проходимости стента с момента первой РЧА и выживаемость с момента постановки диагноза составили 100 дней и 5,3 месяца соответственно. T. T. Wu et al. [36] также сообщают о своем опыте лечения 47 пациентов, которым было проведено 65 процедур чрескож- ной РЧА с размещением саморасширяющихся металлических стентов. Значимо сниженные уровня билирубина наблюдали на 7-й день после операции. Медиана проходимости стента составила 149 дней (15–281). Медиана выживаемости – 181 (15–495) день. Conflict of interest The authors declare no conflict of interest. The authors declare no conflict of interest. Конфликт интересов Авторы заявили об отсутствии конфликта интересов. Литература 1. Макаров Е. С. и др. Рак проксимальных желчных протоков. М. : Печатка, 2018. 100 с. 1. Макаров Е. С. и др. Рак проксимальных желчных протоков. М. : Печатка, 2018. 100 с. Ивануса С. Я. и др. 111–117 ходимости стента через 90 дней составил 56 и 24 % в основной и контрольной группе соответственно. Среднее время про- ходимости стента было значительно больше при выполнении РЧА (119,5 против 65,3 дня). N. Xia et al. [39] опубликовали результаты применения РЧА при окклюзии стента с исполь- зованием чрескожного доступа. Среднее время проходимости стента составило 107 дней. Осложнений, связанных с про- ведением РЧА, не было. (13,2 против 8,3 месяца). Многовариантный регрессионный анализ Кокса показал, что РЧА является единственным фак- тором, влияющим на выживаемость пациентов. (13,2 против 8,3 месяца). Многовариантный регрессионный анализ Кокса показал, что РЧА является единственным фак- тором, влияющим на выживаемость пациентов. (13,2 против 8,3 месяца). Многовариантный регрессионный анализ Кокса показал, что РЧА является единственным фак- тором, влияющим на выживаемость пациентов. X. Zheng et al. [33] в 2016 г. провели первый метаанализ, включающий в себя 9 исследований, суммарно 263 пациента: холангиоцеллюлярный рак (65,8 %), рак поджелудочной желе- зы (29,3 %), метастазы других опухолей (1,5 %), другие типы рака (4,9 %). Всем больным выполнялась эндоскопическая РЧА. Описано увеличение диаметра желчных протоков в месте стриктуры – в среднем с 1,189 до 4,635 мм. Средняя продол- жительность функционирования стента составила 7,6 месяца. Смертность 30-, 90-дневная и 2-летняя составила 2, 21 и 48 % соответственно. Суммарная частота осложнений составила 17 %, при этом бóльшая часть лечилась консервативно. Таким образом, методами выбора в лечении пациентов с нерезектабельной холангиокарциномой внепеченочных желч- ных протоков могут являться ФДТ или РЧА. Однако необходи- мы дополнительные исследования с участием бóльшего числа пациентов, а также исследования сочетанного применения раз- личных методов локорегионарного воздействия. Необходима разработка новых фотосенсибилизаторов, которые бы позво- лили достичь максимальной эффективности ФДТ. р р В 2018 г. A. A. Sofi et al. [34] также представили систе- матический обзор и метаанализ, включающий в себя сум- марно 505 пациентов (9 исследований) со злокачественными стриктурами желчных протоков, которым выполнялось стен- тирование металлическими саморасширяющимися стента- ми после РЧА или самостоятельно. Пациенты, получившие РЧА (n=239), имели более длительную проходимость стента, а также выживаемость (285 против 248 дней). При этом раз- ница в выживаемости наблюдалась даже несмотря на то, что число пациентов, получавших системную ХТ, было значитель- но больше в группе контроля. Имеется предположение, что, помимо термического устранения новообразования, при РЧА, за счет некроза опухоли, происходит активизация противо- опухолевого иммунного ответа. References Комбинация методов регионарной терапии в ле- чении неоперабельной опухоли Клатскина // Высокотехнолог. мед. 2020. Т. 7, № 4. С. 8–16. 9. Pahernik S. A. et al. Distribution and pharmacokinetics of Photofrin in human bile duct cancer // J Photochem Photobiol. 1998;47(1): 58–62. 26. Nanashima A. et al. Adjuvant photodynamic therapy for bile duct car- cinoma after surgery : a preliminary study // J. Gastroenterol. 2004. Vol. 39, № 11. P. 1095–1101. 10. Chan Thi Hai Yen et al. Chlorine-type photosensitizers in tumor PDT // Russian Biotherapeutic Journal. 2009;8(4):99–104. 11. Gerhardt T. et al. Combination of bilateral metal stenting and transstent photodynamic therapy for palliative treatment of hilar cholangiocarci- noma // J Gastroenterol. 2010;48(1):28–32. 27. Cosgrove N. D. et al. Photodynamic therapy provides local control of cholangiocarcinoma in patients awaiting liver transplantation // Am. J. Transplant. 2014. Vol. 14, № 2. P. 466– 71. 12. McCaughan J. S. et al. Photodynamic therapy to treat tumors of the ex- trahepatic biliary ducts. A case report // Arch Surg. 1991;126(1):111–113. 28. Schmidt A. et al. Short-term effects and adverse events of endoscopi- cally applied radiofrequency ablation appear to be comparable with photodynamic therapy in hilar cholangiocarcinoma // United European Gastroenterol. J. 2016. Vol. 4, № 4. P. 570–579. 13. Ortner M. E. et al. Successful photodynamic therapy for nonresectable cholangiocarcinoma: a randomized prospective study // Gastroenterology. 2003;125(5):1355–1363. 29. Strand D. S. et al. ERCP-directed radiofrequency ablation and photo- dynamic therapy are associated with comparable survival in the treat- ment of unresectable cholangiocarcinoma // Gastrointest Endosc. 2014. Vol. 80, № 5. P. 794–804. 14. Zoepf T. et al. Palliation of nonresectable bile duct cancer: impro ved survival after photodynamic therapy // Am J Gastroenterol. 2005; 100(11):2426–2430. 15. Pereira S. et al. P. PHOTOSTENT-02: porfimer sodium photodynamic therapy plus stentingversus stenting alone in patients with locally ad- vanced or metastatic biliary tract cancer // ESMO Open. 2018;3(5). 30. Alvarez-Sánchez M. V., Napoléon B. Review of endoscopic radiofre- quency in biliopancreatic tumours with emphasis on clinical benefits, controversies and safety // World J. Gastroenterol. 2016. Vol. 22, № 37. P. 8257–8270. 16. Moole H. et al. Success of photodynamic therapy in palliating patients with nonresectable cholangiocarcinoma: A systematic review and meta- analysis // World J Gastroenterol. 2017;32(6):1278–1288. 31. Steel A. W. et al. Endoscopically applied radiofrequency ablation appears to be safe in the treatment of malignant biliary obstruction // Gastrointest Endosc. 2011. Vol. 73. P. 149–153. 1. Макаров Е. С. и др. Рак проксимальных желчных протоков. М. : Печатка, 2018. 100 с. [38] сравнили две группы больных c металлическими саморасширяющимися стентами, установ- ленными по поводу злокачественных новообразований желч- ных протоков. В основной группе (n=25) больным, в связи с окклюзией стента, выполнялась эндоскопическая внутри- протоковая РЧА. Контрольной группе (n=25) устанавливали дополнительный пластиковый стент. В первой группе РЧА была успешно проведена у 14 (56 %) пациентов, остальным 11 больным были установлены пластиковые стенты. Процент про- 10. Чан Тхи Хай Иен и др. Фотосенсибилизаторы хлоринового ряда в ФДТ опухолей // Рос. биотерапевт. журн. 2009. Т. 8, № 4. С. 99–104. 11. Gerhardt T. et al. Combination of bilateral metal stenting and transstent photodynamic therapy for palliative treatment of hilar cholangiocarci- noma // J. Gastroenterol. 2010. Vol. 48, № 1. P. 28–32. 12. McCaughan J. S. et al. Photodynamic therapy to treat tumors of the extrahepatic biliary ducts. A case report // Arch. Surg. 1991. Vol. 126, № 1. P. 111–113. 115 Ивануса С. Я. и др. «Вестник хирургии» • 2021 • Том 180 • № 5 • С. 111–117 35. Wang Y. et al. Percutaneous intraductal radiofrequency ablation in the management of unresectable Bismuth types III and IV hilar cholangio- carcinoma // Oncotarget. 2016. Vol. 7, № 33. P. 53911–53920. 13. Ortner M. E. et al. Successful photodynamic therapy for nonresecta ble cholangiocarcinoma : a randomized prospective study // Gastroen terology. 2003. Vol. 125, № 5. P. 1355–1363. 35. Wang Y. et al. Percutaneous intraductal radiofrequency ablation in the management of unresectable Bismuth types III and IV hilar cholangio- carcinoma // Oncotarget. 2016. Vol. 7, № 33. P. 53911–53920. 14. Zoepf T. et al. Palliation of nonresectable bile duct cancer : improved sur- vival after photodynamic therapy // Am. J. Gastroenterol. 2005. Vol. 100, № 11. P. 2426–2430. 36. Wu T. et al. Percutaneous Intraluminal Radiofrequency Ablation for Malig nant Extrahepatic Biliary Obstruction: A Safe and Feasible Method // Dig Dis Sci. 2015. Vol. 60, № 7. P. 2158–2163. 15. Pereira S. P. et al. PHOTOSTENT-02: porfimer sodium photodynamic therapy plus stentingversus stenting alone in patients with locally ad- vanced or metastatic biliary tract cancer // ESMO Open. 2018. Vol. 3, № 5. 37. Pozsa´r J. et al. Intraductal radiofrequency ablation can restore patency of occluded biliary selfexpanding metal stents // Z. Gastroenterol. 2011. Vol. 49. Р. 70. 16. Moole H. et al. Success of photodynamic therapy in palliating patients with nonresectable cholangiocarcinoma : A systematic review and meta- analysis // World J. Gastroenterol. 2017. References 18. Cheon Y. K. et al. Longterm outcome of photodynamic therapy compared with biliary stenting alone in patients with advanced hilar cholangiocar- cinoma // HPB (Oxford). 2012. Vol. 14, № 3. P. 185–193. 1. Makarov E. S. i dr. Cancer of the proximal bile ducts. Moscow, Pechatka, 2018:100. (In Russ.). 19. Prasad G. A. et al. Factors associated with increased survival after photodynamic therapy for cholangiocarcinoma // Clin. Gastroenterol. Hepatol. 2007. Vol. 5, № 6. P. 743–748. 2. Maystrenko N. A. i dr. Cholangiocellular cancer (features of diagnosis and treatment) // Practical Oncology. 2008;(4):229–236. (In Russ.). 3. Wiedmann M. et al. Neoadjuvant photodynamic therapy as a new ap- proach to treating hilar cholangiocarcinoma: a phase II pilot study // Cancer. 2003;97(11):2783–2790. 20. Talreja J. P. et al. Photodynamic therapy for unresectable cholangio- carcinoma : contribution of single operator cholangioscopy for targeted treatment // Photochem Photobiol Sci. 2011. Vol. 10, № 7. P. 1233–1238. 4. Witzigmann H. et al. Surgical and palliative management and outcome in 184 patients with hilar cholangiocarcinoma: palliative photodynamic therapy plus stenting is comparable to r1/r2 resection /// Ann Surg. 2006;244(2):230–239. 21. Wagner A. et al. Temoporfin improves efficacy of photodynamic therapy in advanced biliary tract carcinoma : A multicenter prospective phase II study // Hepatology. 2015. Vol. 62, № 5. P. 1456–1465. 22. Hong M. J. et al. Long-term outcome of photodynamic therapy with systemic chemotherapy compared to photodynamic therapy alone in patients with advanced hilar cholangiocarcinoma // Gut. Liver. 2014. Vol. 8, № 3. P. 318–323. 5. Goenka M. K., Goenka U. Palliation: Hilar cholangiocarcinoma // World J. Hepatol. 2014;39(6):559–569. 6. Farley D. R., Weaver A. L., Nagorney D. M. Natural history of unresected cholangiocarcinoma: patient outcome after noncurative intervention // Mayo Clin Proc. 1995;70(5):425–429. 23. Wentrup R. et al. Photodynamic Therapy Plus Chemotherapy Com pared with Photodynamic Therapy Alone in Hilar Nonresectable Cho langiocarcinoma // Gut. Liver. 2016. Vol. 10, № 3. P. 470–475. 7. Jarnagin W. R. et al. Staging, resectability, and outcome in 225 patients with hilar cholangiocarcinoma // Ann Surg. 2001;234(4):507–517. 24. Hauge T. et al. Randomised controlled trial of temoporfin photodynamic therapy plus chemotherapy in nonresectable biliary carcinoma – PCS Nordic study // Photodiagnosis Photodyn Ther. 2016. Vol. 13. P. 330–333. 8. Laptsevich T. P. et al. Photodynamic therapy of malignant tumors: fun damentals, history of development, prospects (review) // Journal of Oncology. 2008;(1):117–138. 25. Гранов Д. А. и др. 1. Макаров Е. С. и др. Рак проксимальных желчных протоков. М. : Печатка, 2018. 100 с. Vol. 32, № 6. P. 1278–1288. 38. Kadayifci A. et al. Radiofrequency ablation for the management of oc- cluded biliary metal stents // Endoscopy. 2016. Vol. 48. P. 1096. 39. Xia N. et al. Percutaneous intraductal radiofrequency ablation for treatment of biliary stent occlusion : A preliminary result // World J. Gastroenterol. 2017. Vol. 23, № 10. P. 1851–1856. 17. Höblinger A. et al. Feasibility and safety of long-term photodynamic therapy (PDT) in the palliative treatment of patients with hilar cholangio- carcinoma// Eur. J. Med Res. 2011. Vol. 16, № 9. P. 391–395. References 17. Höblinger A. et al. Feasibility and safety of long-term photodynamic therapy (PDT) in the palliative treatment of patients with hilar cholan- giocarcinoma // Eur J Med Res. 2011;16(9):391–395. 32. Yang J. et al. Efficacy and safety of endoscopic radiofrequency ablation for unresectable extrahepatic cholangiocarcinoma: a randomized trial // Endoscopy. 2018. Vol. 50. P. 751–760. 18. Cheon Y. K. et al. Longterm outcome of photodynamic therapy compared with biliary stenting alone in patients with advanced hilar cholangiocar- cinoma // HPB (Oxford). 2012;14(3):185–193. 33. Zheng X. et al. Endoscopic radiofrequency ablation may be preferable in the management of malignant biliary obstruction : A systematic re- view and meta-analysis // Journal of Digestive Diseases. 2016. Vol. 17. P. 716– 724. 19. Prasad G. A. et al. Factors associated with increased survival after pho todynamic therapy for cholangiocarcinoma // Clin Gastroenterol Hepatol. 2007;5(6):743–748. 34. Sofi A. A. et al. Radiofrequency ablation combined with biliary stent placement versus stent placement alone for malignant biliary strictures : a systematic review and meta-analysis // Gastrointest. Endosc. 2018. Vol. 87. P. 944–951. 20. Talreja J. P. et al. Photodynamic therapy for unresectable cholangiocar cinoma: contribution of single operator cholangioscopy for targeted treat- ment // Photochem Photobiol Sci. 2011;10(7):1233–1238. 116 Ivanusa S. Ya. et al. «Grekov’s Bulletin of Surgery» • 2021 • Vol. 180 • № 5 • P. 111–117 30. Alvarez-Sánchez M. V., Napoléon B. Review of endoscopic radiofre- quency in biliopancreatic tumours with emphasis on clinical benefits, con- troversies and safety // World J Gastroenterol. 2016;22(37) :8257–8270. 21. Wagner A. et al. Temoporfin improves efficacy of photodynamic therapy in advanced biliary tract carcinoma: A multicenter prospective phase II study// Hepatology. 2015;62(5):1456–1465. 22. Hong M. J. et al. Long-term outcome of photodynamic therapy with sys- temic chemotherapy compared to photodynamic therapy alone in patients with advanced hilar cholangiocarcinoma // Gut Liver. 2014;8(3):318–323. 31. Steel A. W. et al. Endoscopically applied radiofrequency ablation appears to be safe in the treatment of malignant biliary obstruction // Gastrointest Endosc. 2011;73:149–153. 32. Yang J. et al. Efficacy and safety of endoscopic radiofrequency ablation for unresectable extrahepatic cholangiocarcinoma: a randomized trial // Endoscopy. 2018;50:751–760. 23. Wentrup R. et al. Photodynamic Therapy Plus Chemotherapy Compa red with Photodynamic Therapy Alone in Hilar Nonresectable Cholan giocarcinoma // Gut Liver. 2016;10(3):470–475. 24. Hauge T. et al. References Randomised controlled trial of temoporfin photodynamic therapy plus chemotherapy in nonresectable biliary carcinoma – PCS Nordic study // Photodiagnosis Photodyn Ther. 2016;13:330–333. 33. Zheng X. et al. Endoscopic radiofrequency ablation may be preferable in the management of malignant biliary obstruction: A systematic review and meta-analysis // Journal of Digestive Diseases. 2016;17:716– 724. 34. Sofi A. A. et al. Radiofrequency ablation combined with biliary stent place- ment versus stent placement alone for malignant biliary strictures: a sys- tematic review and meta-analysis // Gastrointest Endosc. 2018;87:944–951. 25. Granov D. A. i dr. Combination of methods of regional therapy in the treat- ment of inoperable Klatskin tumor // High-tech medicine. 2020;7(4):8–16. (In Russ.). 35. Wang Y. et al. Percutaneous intraductal radiofrequency ablation in the management of unresectable Bismuth types III and IV hilar cholangio- carcinoma // Oncotarget. 2016;7(33):53911–53920. 26. Nanashima A. et al. Adjuvant photodynamic therapy for bile duct car cinoma after surgery: a preliminary study // J Gastroenterol. 2004; 39(11):1095–1101. 36. Wu T. et al. Percutaneous Intraluminal Radiofrequency Ablation for Malignant Extrahepatic Biliary Obstruction: A Safe and Feasible Method // Dig Dis Sci. 2015;60(7):2158–2163. 27. Cosgrove N. D. et al. Photodynamic therapy provides local control of cholangiocarcinoma in patients awaiting liver transplantation // Am J Transplant. 2014;14( 2):466– 71. 37. Pozsa´r J. et al. Intraductal radiofrequency ablation can restore patency of occluded biliary selfexpanding metal stents // Z Gastroenterol. 2011;49:70. 28. Schmidt A. et al. Short-term effects and adverse events of endoscopi- cally applied radiofrequency ablation appear to be comparable with photodynamic therapy in hilar cholangiocarcinoma // United European Gastroenterol J. 2016.;4(4):570–579. 38. Kadayifci A. et al. Radiofrequency ablation for the management of oc- cluded biliary metal stents // Endoscopy. 2016;48:1096. 29. Strand D. S. et al. ERCP-directed radiofrequency ablation and photodynamic therapy are associated with comparable survival in the treatment of unre- sectable cholangiocarcinoma // Gastrointest Endosc. 2014;80(5):794–804. 39. Xia N. et al. Percutaneous intraductal radiofrequency ablation for treatment of biliary stent occlusion: A preliminary result // World J Gastroenterol. 2017;23(10):1851–1856. Информация об авторах: Ивануса Сергей Ярославович, доктор медицинских наук, профессор, заслуженный врач РФ, начальник кафедры общей хирургии, Военно-медицинская акаде- мия им. С. М. Кирова (Санкт-Петербург, Россия), ORCID: 0000-0003-3948-6928; Алентьев Сергей Александрович, доктор медицинских наук, доцент кафедры общей хирургии, Военно-медицинская академия им. С. М. Кирова (Санкт-Петербург, Россия), ORCID: 0000-0002-4562-113X; Евстратьева Дарья Юрьевна, слушатель ординатуры, Военно-медицинская академия им. С. М. Кирова (Санкт-Петербург, Россия), ORCID: 0000-0002-2201-3959; Молчанов Александр Алек- сеевич, зав. отделением клиники общей хирургии, Военно-медицинская академия им. С. М. Кирова (Санкт-Петербург, Россия), ORCID: 0000-0001-5150-1700. Information about authors: Ivanusa Sergey Ya., Dr. of Sci. (Med.), Professor, Honored Doctor of the Russian Federation, Head of the Department of General Surgery, Military Medical Academy (Saint Petersburg, Russia), ORCID: 0000-0003-3948-6928; Alentev Sergey A., Dr. of Sci. (Med.), Associate Professor of the Department of General Surgery, S Military Medical Academy (Saint Petersburg, Russia), ORCID: 0000-0002-4562-113X; Evstrateva Daria Yu., Resident, Military Medical Academy (Saint Petersburg, Russia), ORCID: 0000-0002-2201-3959; Molchanov Alexander A., Head of the Department of the General Surgery Clinic, Military Medical Academy (Saint Petersburg, Russia), ORCID: 0000-0001-5150-1700. 117
https://openalex.org/W4304128155	https://hal.science/hal-03818965/document	English	null	Electro-Acoustic Properties of Scandium-Doped Aluminum Nitride (ScxAl1-xN) Material and its Application to Phononic Crystal-Coupled SAW Devices	Crystals	2,022	cc-by	17,528	Electro-Acoustic Properties of Scandium-Doped Aluminum Nitride (ScxAl1-xN) Material and its Application to Phononic Crystal-Coupled SAW Devices Fahima Arab, Fares Kanouni, Rafik Serhane, Yan Pennec, Zafer Özer, Khaled Bouamama To cite this version: Fahima Arab, Fares Kanouni, Rafik Serhane, Yan Pennec, Zafer Özer, et al.. Electro-Acoustic Prop- erties of Scandium-Doped Aluminum Nitride (ScxAl1-xN) Material and its Application to Phononic Crystal-Coupled SAW Devices. Crystals, 2022, 12 (10), pp.1431. 10.3390/cryst12101431. hal- 03818965 Distributed under a Creative Commons Attribution 4.0 International License Citation: Arab, F.; Kanouni, F.; Serhane, R.; Pennec, Y.; Özer, Z.; Bouamama, K. Electro-Acoustic Properties of Scandium-Doped Aluminum Nitride (ScxAl1-xN) Material and its Application to Phononic Crystal-Coupled SAW Devices. Crystals 2022, 12, 1431. https://doi.org/10.3390/ cryst12101431 Keywords: w-ScxAl1-xN compound; electro-acoustic properties; Density Functional Theory; ﬁnite element analysis; PnC-based SAW delay line Academic Editor: Abdolhamid Akbarzadeh Shafaroudi Received: 10 September 2022 Accepted: 3 October 2022 Published: 10 October 2022 Fahima Arab 1,, Fares Kanouni 1, Raﬁk Serhane 2 , Yan Pennec 3,, Zafer Özer 4 and Khal Fahima Arab 1,, Fares Kanouni 1, Raﬁk Serhane 2 1 Photonic Crystals Team, Research Unit in Optics and Photonics–Center for Development of Advanc Technologies (UROP-CDTA), El-Bez, Setif 19000, Algeria 1 Photonic Crystals Team, Research Unit in Optics and Photonics–Center for Development of Advanced Technologies (UROP-CDTA), El-Bez, Setif 19000, Algeria g g 2 MEMS and Sensors Team, Microelectronic and Nanotechnology Division, Center for Development o Ad d T h l i Cité 20 A ût 1956 B b H 16303 Al i 2 MEMS and Sensors Team, Microelectronic and Nanotechnology Division, Center for Developme Advanced Technologies, Cité 20 Août 1956, Baba Hassen 16303, Algeria Advanced Technologies, Cité 20 Août 1956, Baba Hassen 16303, Algeria 3 IEMN, UMR CNRS 8520, University of Lille, 59652 Villeneuve d’Ascq, France 4 Mersin Vocational High School, Electronic and Automation Department, Mersin University, Mersin 33110, Turkey 5 Laboratoire d’Optoélectronique et Composants, Département de Physique, Université Sétif 1, Sétif 19000, Algeria g Correspondence: farab@cdta.dz (F.A.); yan.pennec@univ-lille.fr (Y.P.) Abstract: Within the framework of the Density Functional Theory, the elastic, dielectric, and piezo- electric coefﬁcients of w-ScxAl1−xN material were investigated for scandium (Sc) concentrations x = 0 to 0.375. The electro-acoustic properties are used to investigate the frequency response of the SAW delay line, based on the tilt θ◦of the normal c-axis of the w-ScxAl1−xN piezoelectric thin ﬁlm. We found that the piezoelectric response is improved as the Sc concentration increases, which is consistent with existing works in the literature. A 2D-phononic crystal pillars was then grafted on top of the surface, and the dependence of the acoustic band gaps is investigated with the help of the ﬁnite element method as a function of the Sc concentration and the tilted angle of w-Sc0.375Al0.625N. It was found that the two ﬁrst band gaps exhibit a shift toward low frequencies with increasing Sc concentration. Moreover, the second acoustic bandgap is more sensitive to the inclination angle than the ﬁrst. Furthermore, the insertion loss (S21) of w-Sc0.375Al0.625N is improved by 22 dB at θ◦= 60◦. The c-axis tilted Sc0.375Al0.625N-SAW delay line coupled with 2D-phononic crystals is a promising structure for low-loss and high-frequency SAW devices. HAL Id: hal-03818965 https://hal.science/hal-03818965v1 Submitted on 18 Oct 2022 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License crystals Electro-Acoustic Properties of Scandium-Doped Aluminum Nitride (ScxAl1-xN) Material and its Application to Phononic Crystal-Coupled SAW Devices , Fares Kanouni 1, Raﬁk Serhane 2 , Yan Pennec 3,, Zafer Özer 4 and Khaled Bouamama 5 1. Introduction This can be explained by the fact that, in the growth process of the AlScN thin ﬁlm, a competition between Al3+ and Sc3+ for coordination with nitrogen is introduced, which weakens the resistance to nitrogen displacement in the crystal structure and increases the volume of the unit cell. Due to this remarkable piezoelectricity property, scandium (Sc)-doped AlN (ScAlN) is extensively used in high-frequency ﬁlters, sensors, and micro- electromechanical devices [15–18]. Basically, the high piezoelectric constant of ScAlN leads to a signiﬁcant increase in the electro-mechanical coupling factor, thus resulting in a signiﬁcant improvement in the performance of SAW devices. ScxAl1-xN alloy thin ﬁlms may represent an alternative material to replace AlN. y p p M. Akiyama et al. [19,20] found that the piezoelectric coefﬁcient d33 increases from 6 pC/N for pure AlN to 27.6 pC/N for Sc0.43Al0.57N, i.e., at least 500% larger than for AlN. They showed that the piezoelectric response is strongly dependent on the growth tempera- ture. Furthermore, W. Gunilla et al. [21] showed that the electromechanical coupling (kt2), inﬂuenced by the Sc concentration, increases from 7% in AlN to 15% in Al0.7Sc0.3N. As an example, Konno Akira et al. [22] reported that, when using a 40% Sc-doped AlN ﬁlm, the piezoelectric characteristic of Lamb wave resonators is approximately ﬁve times higher than in pure AlN. p From the perspective of different applications, ScxAl1-xN thin ﬁlms have been de- posited on several substrates such as Si [23], 6H-SiC [24], diamond [25], and sapphire [26]. In each case, the structures enhance the SAW velocity and the electro-mechanical coupling factor k2. Some examples of SAW devices properties based on ScxAl1-xN/sapphire structures have been investigated using experiments [27] and theories [13,14,28], showing relatively high insertion losses. By comparison, ScxAl1-xN/sapphire structures are promising SAW devices for high-temperature sensor applications [16,29,30]. Additionally, highly c-axis-oriented thin ﬁlms are useful for surface acoustic device fabrication, as evidenced by the rising electromechanical coupling coefﬁcient of c-axis tilted ScAlN thin ﬁlms. Many studies witness that the direction of wave propagation and the c-axis tilt angle of scandium-doped aluminum nitride piezoelectric thin ﬁlms have a major impact on the electromechanical properties of SAW devices [25,31]. Theoretical analysis of the Rayleigh SAW mode showed that (k2), in a tilted ScAlN ﬁlm on the R-sapphire, is 3.9% at a tilt angle of 90◦and 3.7% at a tilt angle of 54◦[32]. 1. Introduction Surface acoustic wave (SAW) devices are among the most attractive candidates for modern electronic devices because of their low cost, low losses, and great sensitivity and integrability. These devices have become crucial for mobile telecommunication systems that require high-frequency and low-loss sensing components. In SAW devices, piezoelectric materials are required to convert the incoming electric signal to an acoustic one and then back to electric. One of the most challenging features in producing high-performance SAW devices is to select an appropriate piezoelectric material in combination with a suitable substrate [1,2]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). The interest in piezoelectric thin ﬁlms for electromechanical systems increased after the discovery of aluminum nitride (AlN) with, in particular, the commercial success of the SAW sensors. Most conventional SAW devices made of silicon have been reproduced with enhanced performance using AlN thin-ﬁlm piezoelectric technology [3]. p g p gy Unlike other widely used piezoelectric materials such as zinc oxide (ZnO) [4] and lead zirconate titanate (PZT) [5], AlN has good chemical properties, thermal stability, high https://www.mdpi.com/journal/crystals Crystals 2022, 12, 1431. https://doi.org/10.3390/cryst12101431 Crystals 2022, 12, 1431 2 of 22 hardness, and high surface wave velocity, and is compatible with conventional CMOS technology on silicon [6–8]. Conversely, AlN piezoelectric thin ﬁlm has a lower effective piezoelectric constant, which can be improved using speciﬁc transition metals as dopants. p p g p p Recently, a number of works have been devoted to the AlN doping of scandium (Sc), yttrium (Y), ytterbium (Yb), and other elements to increase the piezoelectric properties and improve the efﬁciency of SAW devices [9–14]. Compared to other transition metals, scandium as a dopant exhibits a structural transition from the wurtzite phase to the rocky phase, which occurs at low alloying concentrations, leading to easily synthesized and more stable alloys. Second, the piezoelectric coefﬁcients are much higher in AlScN than in pure AlN. 1. Introduction By comparison, the insertion loss of fabricated IDT/ScAlN tilted at 33◦/R-sapphire structure is 34.4 dB, demonstrating that higher electromechanical coupling factor k2 improves the crystal orientation of ScAlN ﬁlms. The c-axis-tilted ScAlN ﬁlms with a Sc concentration of 40%, prepared on a silicon substrate via RF magnetron sputtering based on the self-shadowing effect, reach a maximum c-axis tilt angle of 57.4◦. In this structure, the electromechanical coupling coefﬁcient (k2) has been increased because of the c-axis tilt angle [33,34]. Furthermore, recent studies have shown that surface acoustic waves (SAWs) have a signiﬁcant interest in phononic crystals (PnCs) [35–38], in particular to generate and detect SAW using interdigital transducers [39–41]. In such structures, the well-known band gap, in which acoustic waves cannot propagate in any direction of the piezoelectric phononic crystals [42], plays an important role in the design of applications such as de- multiplexers [43], ﬁlters, and waveguides [44–46]. The insertion loss can be improved by −7 dB with PnCs used as reﬂected gratings [47]. Similarly, as grating sizes are reduced, a Crystals 2022, 12, 1431 3 of 22 novel proposed structure may provide a new idea for resonator, ﬁlter, or other possible application designs [48]. novel proposed structure may provide a new idea for resonator, ﬁlter, or other possible application designs [48]. We also mention the development of highly sensitive SAW sensors, using nano- or micro-pillar structures, attached like a cantilever between the ﬁngers of the IDT and used as a sensing medium [49]. The interaction of surface elastic waves with 2D phononic crystal (PnCs) on top of piezoelectric substrate has been studied to control the central frequency and the opening of acoustic band gaps [50,51]. Phononic crystals have been recently proposed as a platform for the investigation of the acoustic velocity and/or the density of the liquid ﬁlling parts of the structure. The sensing phenomena are based on the high sensitivity to external stimuli (temperature, pressure, nature of the liquid, etc.) of localized modes, associated with defects, appearing inside absolute phononic band gaps. Shaped and pillar-based phononic crystal anchors have recently been used in the ﬁeld of micro-electromechanical system (MEMS) resonators to reduce anchor losses and improve the quality factor (Q) [52–55]. It is demonstrated that using pillar-based PnC anchors in AlN lamb wave resonators can reduce the anchor loss and boost the Q factor by 24 times compared to the conventional resonators [55]. 1. Introduction Recently, SAW-PnC-based devices with non-metallic pillars have been designed on Si substrate for wide ultrahigh frequency range applications, such as wirelesses communications, ﬁlters, and duplexers [56–58]. pp p Despite the large number of studies on SAW-PnCs, none of them have examined the effect of ScxAl1-xN or c-tilted ScxAl1-xN on PnCs band gaps. For this purpose, because of the dependence of the elastic properties due to scandium concentration and c-axis tilted angle, we ﬁrst studied the inﬂuence of Sc concentration on elastic, piezoelectric, and dielectric properties using the Density Functional Theory (DFT). Then, we used the calculated values of the elastic parameters to investigate the inﬂuence of Sc concentration and c-axis tilt angle of ScxAl1-xN thin ﬁlms on both the acoustic band gap generated by the pillars structuration, and the insertion loss of the SAW device, i.e., the S21(dB)-scattering parameter. 2. Computational Methods Density Functional Theory (DFT), in combination with generalized gradient approxi- mation GGA-PBE [59], is used to predict the structural characteristics and elastic coefﬁcients of w-ScxAl1-xN Wurtzite crystals. A Monkhorst–Pack mesh of 5 × 5 × 5 k-points in the Brillouin-zone integral was used with a cutoff energy of 500 eV and total energy conver- gence threshold of 10−6 eV [60]. The Density Functional Theory Perturbation (DFTP) [61] was used to determine both piezoelectric and dielectric constants of w-ScxAl1-xN in the range of x = 0, 0.125, 0.25, and 0.375, and then, the elastic constants (Cij), the piezoelectric coefﬁcients (eij), and the permittivity tensor (εij) of w-ScxAl1-xN. The physical parameters were introduced in the Finite Element model (COMSOL) to obtain the acoustic band gaps of the 2D ScxAl1-xN pillar-based phononic crystals (PnCs), the mechanical transmission, and the transmission loss S21(dB) of the SAW delay line. The results were compared with the non-structured surface. The same methodology was used in the case of the c-axis tilt angle of w-ScxAl1-xN [62] (see diagram in Figure 1). 4 of 22 Crystals 2022, 12, 1431 rystals 2022, 12, x FOR Figure 1. Diagram of the calculation methodology. Figure 1. Diagram of the calculation methodology. Figure 1. Diagram of the calculation methodol Figure 1. Diagram of the calculation methodology. 3 Results and Discussion 3. Results and Discussion 3 Results and Discussion 3. Results and Discussion 3. Results and Discussion In the first step, we determined the equilibrium geometries, including the latt rameters (a, c) of w-AlN for a structural optimization. The equilibrium structural p ters were obtained at zero pressure and under the temperatures gathered in Table results are in good agreement with both experimental data [63], with an error of In the ﬁrst step, we determined the equilibrium geometries, including the lattice pa- rameters (a, c) of w-AlN for a structural optimization. The equilibrium structural parameters were obtained at zero pressure and under the temperatures gathered in Table 1. Our results are in good agreement with both experimental data [63], with an error of about 0.70%, and theoretical values [9,64] (see Table 1). Crystals 2022, 12, 1431 5 of 22 Table 1. Comparison of equilibrium lattice parameters (a, c) (Å) and elastic constants Cij (GPa) for w-ScxAl1-xN compound (x = 0%, 12.5%, 25% and 37.5%) between our results () and the literature. Material Constants w-AlN w-ScxAl1-xN Th. Exp. x 12.5% 25% 37.5% a (Å) 3.128 * 3.1163 3.1842* 3.2426 * 3.3136 * 3.13164 c (Å) 5.015 * 4.9863 5.0489* 5.0696 * 5.065* 5.01864 C11 (GPa) 376 * 402.5 ± 0.569 332.91 * 302.56 282.64 * 43264 37465 412.6 ± 0.0570 336.3765 305.6865 282.03665 378.867 34568 394 71 C12 (GPa) 123 * 135.6 ± 0.569 126.86 * 130.82 * 121.45 * 17064 12965 126.6 ± 0.570 128.967 12568 134 71 121.9565 116.0365 110.4465 C13 (GPa) 91* 101 ± 269 104.84 104.86 * 112.06 * 14764 10165 118.8 ± 0.970 91.1165 83.4365 77.2565 96.167 12068 9571 C33 (GPa) 354* 387.6 ± 169 293.99 * 251.91* 220.94 * 39064 35165 386.1 ± 4.570 302.1165 255.3365 211.3765 357.567 39568 40271 C44 (GPa) 116* 122.9 ± 0.569 103.47 * 96.09* 87.29 * 15564 11265 127.4 ± 0.970 102.1565 97.48965 97.5965 11267 11868 12171 : Our calculations. Table 1. Comparison of equilibrium lattice parameters (a, c) (Å) and elastic constants Cij (GPa) for w-ScxAl1-xN compound (x = 0%, 12.5%, 25% and 37.5%) between our results () and the literature. Table 1. Comparison of equilibrium lattice parameters (a, c) (Å) and elastic constants Cij (GPa) for w-ScxAl1-xN compound (x = 0%, 12.5%, 25% and 37.5%) between our results () and the literature. We then examined the effect of the proportion of Sc on the properties of aluminum scandium nitrite (w-ScxAl1−xN). 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants 6 (2) (2)(2) 𝑗 𝑗 For the hexagonal material, the tensor eij is given by: For the hexagonal material, the tensor eij is given by: For the hexagonal material, the tensor eij is given by: For the hexagonal material, the tensor eij is given by: 𝑒𝑖𝑗= ( 0 0 0 0 𝑒15 0 0 0 0 𝑒15 0 0 𝑒31 𝑒31 𝑒33 0 0 0 ) (3) eij =   0 0 0 0 e15 0 0 0 0 e15 0 0 e31 e31 e33 0 0 0   (3) (3) (3) 𝑑𝑖𝑘= ( 0 0 0 0 𝑑15 0 0 0 0 𝑑15 0 0 𝑑31 𝑑31 𝑑33 0 0 0 ) (4) 𝐶𝑖𝑗= ( 𝐶11 𝐶12 𝐶13 0 0 0 𝐶12 𝐶11 𝐶13 0 0 0 𝐶13 𝐶13 𝐶33 0 0 0 0 0 0 𝐶44 0 0 0 0 0 0 𝐶44 0 0 0 0 0 0 1 2 (𝐶11 −𝐶12)) (5) dik =   0 0 0 0 d15 0 0 0 0 d15 0 0 d31 d31 d33 0 0 0   (4) Cij =         C11 C12 C13 0 0 0 C12 C11 C13 0 0 0 C13 C13 C33 0 0 0 0 0 0 C44 0 0 0 0 0 0 C44 0 0 0 0 0 0 1 2(C11 −C12)         (5) 𝑑𝑖𝑘= ( 0 0 0 0 𝑑15 0 0 0 0 𝑑15 0 0 𝑑31 𝑑31 𝑑33 0 0 0 ) (4) dik =   0 0 0 0 d15 0 0 0 0 d15 0 0 d31 d31 d33 0 0 0   (4) (4) (4) 𝐶𝑖𝑗= ( 𝐶11 𝐶12 𝐶13 0 0 0 𝐶12 𝐶11 𝐶13 0 0 0 𝐶13 𝐶13 𝐶33 0 0 0 0 0 0 𝐶44 0 0 0 0 0 0 𝐶44 0 0 0 0 0 0 1 2 (𝐶11 −𝐶12)) (5) Cij =         C11 C12 C13 0 0 0 C12 C11 C13 0 0 0 C13 C13 C33 0 0 0 0 0 0 C44 0 0 0 0 0 0 C44 0 0 0 0 0 0 1 2(C11 −C12)         (5) 5) (5) The dielectric tensor is symmetric and respects all the symmetry operations of the corresponding point group. 3 Results and Discussion 3. Results and Discussion This behavior is similar with the observed theoretical [66,73,74] and experimental [9] works. We can therefore conclude that, as the concentration of Sc increases, the material softens along the c-axis, and hardens in the basal plane. EW 6 of 24 similar with the observed theoretical [66,73,74] and experimental [9] works. We can there- fore conclude that, as the concentration of Sc increases, the material softens along the c- axis, and hardens in the basal plane. Figure 2 Comparison between (a) our calculated elastic constants Cij (GPa), and (b) piezoelectric coefficients eij (C/m2), with data available from the literature. Figure 2. Comparison between (a) our calculated elastic constants Cij (GPa), and (b) piezoelectric coefﬁcients eij (C/m2), with data available from the literature. Figure 2 Comparison between (a) our calculated elastic constants Cij (GPa), and (b) piezoelectric coefficients eij (C/m2), with data available from the literature. Figure 2. Comparison between (a) our calculated elastic constants Cij (GPa), and (b) piezoelectric coefﬁcients eij (C/m2), with data available from the literature. 3 Results and Discussion 3. Results and Discussion For this study, we substituted the concentrations of two, four, and six Al by the Sc atoms in a supercell of (2 × 2 × 1) constructed from a w-AlN Wurtzite structure containing 16 atoms, then representing 12.5%, 25%, and 37.5%. In all concentrations (x = 0, 0.125, 0.25, and 0.375), the predicted lattice parameters of w-ScxAl1-xN agree with the experimental [9] and theoretical results [65] (see Table 1). Furthermore, when the Sc concentration increased, the lattice constants (a and c) increased respectively by 5.6% and 1%, while the c/a ratio decreased by 4.6%. This effect may be due to changes in the bond lengths and angles of the AlN4 tetrahedron, which deforms when Al atoms are replaced by Sc atoms [19,65]. This effect can then be estimated geometrically from the determination of the lattice constants a and c [9]. Theoretical and experimental details of the impact of scandium concentration on bond length can be found in several papers [9,19,65,66]. Again, these results agree with the experimental [19] and theoretical data [65] obtained from the Quantum Expresso (QE) software package with a (3 × 3 × 1) super cell. p Table 1 also compares the single-crystal elastic constants for w-ScxAl1−xN obtained from the proposed functional GGA-PBE for various concentrations x = 0, 0.125, 0.25, 0.375, theoretically [64,65,67,68] and experimentally [69–71]. Our result agrees with most of the accessible data, indicating the accuracy of our calculation. The elastic stiffness constants Cij should respect the mechanical stability criteria for hexagonal symmetry, as described below [72]: (1) (C11−C12) > 0 and C44 > 0 (C11 + C12) C33−2C213 > 0 (1) The elastic constants of w-ScxAl1-xN satisfy the above criteria in all compositions over the considered range, indicating that the compounds are mechanically stable. From The elastic constants of w-ScxAl1-xN satisfy the above criteria in all compositions over the considered range, indicating that the compounds are mechanically stable. From Crystals 2022, 12, 1431 6 of 22 Figure 2a, we observe that, when the Sc concentration increases, C33 and C11 present an almost linear variation, decreasing from 354 to 220 GPa and from 376 to 282 GPa, respectively, corresponding to a relative decrease of 38 and 25%. Similarly, C44 decreases as a function of the Sc concentration with a relative reduction of 25%. By comparison, doping AlN with Sc increases the mixed compression/shear C12 and C13 elastic constants. 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants 3.1. Piezoelectric and Dielectric Constants The piezoelectric coefficients eij explain the piezoelectric properties of the material, through the piezoelectric moduli d k and the elastic constants Cij as follows [75]: The piezoelectric coefﬁcients eij explain the piezoelectric properties of the material, through the piezoelectric moduli dik and the elastic constants Cij as follows [75]: piezoelectric moduli dik and the elastic constants Cij as follows [75]: 𝑒𝑖𝑗 = ∑𝑑𝑖𝑘 k 𝐶𝑘𝑗, where i = 1, 2, 3 j = 1,… 6 and k = 1,…6 (2) eij = ∑k dikCkj, where = 1, 2, 3 j = 1, . . . 6 and k = 1, . . . 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants This limits the number of independent elements in the tensor to a minimum of 1 and a maximum of 6 depending on the crystal symmetry. For a hexag- onal crystal it only has two components (𝜀11, 𝜀33) different from zero, expressed as: The dielectric tensor is symmetric and respects all the symmetry operations of the corresponding point group. This limits the number of independent elements in the tensor to a minimum of 1 and a maximum of 6 depending on the crystal symmetry. For a hexagonal crystal it only has two components (ε11, ε33) different from zero, expressed as: p ( 11 33) p 𝜀𝑖𝑗= ( 𝜀11 0 0 0 𝜀11 0 0 0 𝜀33 ) (6) εij=   ε11 0 0 0 ε11 0 0 0 ε33   (6) (6) (6) Crystals 2022, 12, 1431 7 of 22 7 of 22 DFTP is used to calculate the piezoelectric and dielectric tensors of w-ScxAl1-xN, which are the sum of the ionic and electronic contributions. Table 2 reports our calculated values, which are in good agreement with other theoretical data and measured values. Table 2. Comparison of calculated piezoelectric (e15, e31, e33) and dielectric constants (ε11, ε33) for w-ScxAl1-xN compound between our results () and the literature. Table 2. Comparison of calculated piezoelectric (e15, e31, e33) and dielectric constants (ε11, ε33) for w-ScxAl1-xN compound between our results () and the literature. w-ScxAl1-xN compound between our results () and the literature. Material Coefﬁcient w-AlN w-ScxAl1-xN Th. Exp. x 12.5% 25% 37.5% e31 (C/m2) −0.58 * −0.54 ± 0.059 −0.624 * −0.660 * −0.686 * −0.5564–0.59365 −0.6 ± 0.269 −0.62565 −0.67565 −0.74365 –0.42467–0.5868 −0.47 ± 0.270 e33 (C/m2) 1.45 * 1.52 ± 0.439 1.705 * 2.026 * 2.421 * 1.3964 1.47165 1.34 ± 0.169 1.7065 2.14265 2.78865 1.44967 1.5568 2.09 ± 0.470 e15 (C/m2) −0.29 * −0.30 ± 0.229 −0.311* −0.306 * −0.282 * −0.3064–0.31365 −0.32 ± 0.0569 −0.29365 −0.25665 −0.20465 –0.36767−0.4868 −0.24 ± 0.0570 ε11 (10−11 F/m) 8.3 * 868 9.8 ± 07 9 9.03 * 9.75 * 10.47 * 9 ± 0.0169 8.44 ± 0.170 ε33 (10−11 F/m) 9.75 * 9.568 9.1 ± 0.3 9 10.72 * 11.96 * 13.50 * 9.5 ± 0.0169 10.51 ± 0.170 *: Our calculations. Figure 2b depicts the evolution of the piezoelectric coefﬁcients (e15, e31, e33) as a function of the Sc concentration. 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants Dependence on Electro-Acoustic Parameters of w-Sc0.375Al0.625N with tilted c-A Orientation Orientation The description of the dependence of the tilted c-axis orientation angle on the acoustic properties is a very important issue for the growth of films with oriented ture. In order to investigate the inclination angle on the electro-acoustic prope Sc0.375Al0.625N, a rotation following a clockwise angle θ against the y-axis was cons The original coordinates (x, y, z) were changed to a set of new coordinates (x′, y′, z′ found that the elastic stiffness Cij, the piezoelectric stress eij coefficient, and the di permittivity εij can be computed through properties in original coordinate system The description of the dependence of the tilted c-axis orientation angle on the electro- acoustic properties is a very important issue for the growth of ﬁlms with oriented struc- ture. In order to investigate the inclination angle on the electro-acoustic properties of Sc0.375Al0.625N, a rotation following a clockwise angle θ against the y-axis was considered. The original coordinates (x, y, z) were changed to a set of new coordinates (x ′, y ′, z ′). It was found that the elastic stiffness Cij, the piezoelectric stress eij coefﬁcient, and the dielectric permittivity εij can be computed through properties in original coordinate system (x ′, y ′, z ′) with the help of matrix algebra [76]. The description of the dependence of the tilted c-axis orientation angle on the electro- acoustic properties is a very important issue for the growth of films with oriented struc- ture. In order to investigate the inclination angle on the electro-acoustic properties of Sc0.375Al0.625N, a rotation following a clockwise angle θ against the y-axis was considered. The original coordinates (x, y, z) were changed to a set of new coordinates (x′, y′, z′). It was found that the elastic stiffness Cij, the piezoelectric stress eij coefficient, and the dielectric permittivity εij can be computed through properties in original coordinate system (x′, y′, z′) with the help of matrix algebra [76]. A di t E l ’ l th t ti ll h i l t i i i t ti permittivity εij can be computed through properties in original coordinate system z′) with the help of matrix algebra [76]. According to Euler’s laws on the rotation, all physical matrices in a new orie are obtained by a transformation matrices methodology [77]. 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants The most striking feature of this graph is the quick evolution of e33, which increases by 67%. By comparison, the two coefﬁcients e15 and e31 show a relative increase of 18 and 2.75%, respectively. This result indicates that the addition of a trivalent dopant (Sc3+) constrains the displacement along the c-axis. Moreover, because the electronegativity of the Sc atom is lower than that of the Al atom, the w-ScxAl1-xN material is more electrovalent, which improves the piezoelectric properties [19]. The dependence of the piezoelectric coefﬁcients e15, e31, and e33 with increasing Sc content is consistent and follows the general trend obtained from previous theoretical results [64,65,67,68] and experimental measurements [9,69,70]. Table 2 shows that our calculated dielectric constants ε11 and ε33 at x = 0% are in good agreement with previous data [68–70]. Then, as seen in Figure 3, the two components ε33 and ε11 increase almost linearly with the concentration of Sc [21]. Obviously, ε33 is more sensitive than ε11, increasing by 38% compared to 26%. This is mainly due to the strong lattice polarization induced by the out-of-plane (c-axis) Sc atom. These results indicate an increasing charge and polarization binding capacity of the compound w-ScxAl1-xN. All these results show an enhanced electromechanical coupling (k2), which justify the selection of w-Sc0.375Al0.625N material for SAW devices. Crystals 2022, 12, 1431 Crystals 2022, 12, x FOR C l 2022 12 FOR 8 of 22 8 f 24 Figure 3. Calculated dielectric constants ε11 and ε33 as a function of Sc concentration for w-S 3.2. Dependence on Electro-Acoustic Parameters of w-Sc0.375Al0.625N with tilted c-Axis Figure 3. Calculated dielectric constants ε11 and ε33 as a function of Sc concentration for w-ScxAl1-xN. 3.2. Dependence on Electro-Acoustic Parameters of w-Sc0.375Al0.625N with tilted c-Axis Orientation Figure 3. Calculated dielectric constants ε11 and ε33 as a function of Sc concentration for w-ScxAl1-xN. 3.2. Dependence on Electro-Acoustic Parameters of w-Sc0.375Al0.625N with tilted c-Axis Orientation Figure 3. Calculated dielectric constants ε11 and ε33 as a function of Sc concentration for w-S Figure 3. Calculated dielectric constants ε11 and ε33 as a function of Sc concentration for w-ScxAl1-xN. Figure 3. Calculated dielectric constants ε11 and ε33 as a function of Sc concentration for w-ScxAl1-xN. 3.2. Dependence on Electro-Acoustic Parameters of w-Sc0.375Al0.625N with O i t ti 3.2. Dependence on Electro-Acoustic Parameters of w-Sc0.375Al0.625N with tilted c-Axis Orientation 3.2. 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants (a) Elastic constants, (b) piezoelectric coefficient, and (c) dielectric constants in (10−11 F/m) of tilted w-Sc0.375Al0.625N. Figure 4. (a) Elastic constants, (b) piezoelectric coefﬁcient, and (c) dielectric constants in (10−11 F/m) of tilted w-Sc0.375Al0.625N. (c) (c) Figure 4. (a) Elastic constants, (b) piezoelectric coefficient, and (c) dielectric constants in (10−11 F/m) of tilted w-Sc0.375Al0.625N. Figure 4. (a) Elastic constants, (b) piezoelectric coefﬁcient, and (c) dielectric constants in (10−11 F/m) of tilted w-Sc0.375Al0.625N. Figure 4b shows the variation of direct piezoelectric constants eij with the c-axis tilted angle. We observed that e15 and e31 are negative at 0° and behave as a sinus. With the increase of the inclination angle θ, e15 and e31 reach a maximum at 56° and 58° respectively. This dependence was also observed in c-axis-tilted AlN with the same trends [78]. In addition, e33 reaches a maximum at θ = 0°, a minimum at θ = 70°, then becomes zero at 90°, in addition to e15 and e33. Figure 4b shows the variation of direct piezoelectric constants eij with the c-axis tilted angle. We observed that e15 and e31 are negative at 0◦and behave as a sinus. With the increase of the inclination angle θ, e15 and e31 reach a maximum at 56◦and 58◦respectively. This dependence was also observed in c-axis-tilted AlN with the same trends [78]. In addition, e33 reaches a maximum at θ = 0◦, a minimum at θ = 70◦, then becomes zero at 90◦, in addition to e15 and e33. In Figure 4c, we plot the dielectric constants ε11 and ε33 as a function of the c-axis tilted angle θ°. The dielectric constants ε11 and ε33 present an opposite trend with equal values at (θ = 45°). Indeed, the absolute value of ε11 is maximum at θ = 90°, while ε33 component has a maximum at θ = 0°, equivalent to the behavior of undoped AlN [78]. This is attributed to the fact that the x1′ axis becomes x3, and x3′ coincides with negative x1 when θ is equal to 90°, and is a direct consequence of the Wurtzite crystal structure of group III-Nitrides [79,80]. In Figure 4c, we plot the dielectric constants ε11 and ε33 as a function of the c-axis tilted angle θ ◦. The dielectric constants ε11 and ε33 present an opposite trend with equal values at (θ = 45◦). 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants Figure 4a illustrates pendence of Cij against the inclination angle θ. It is clearly seen that, when cha from −90° to 90°, a symmetric behavior for all Cij is observed at θ = 0°. The incre leads to decreases in C11, C12, and C13. An opposite dependence is observed betw and C11 due to the c-axis inclination angle, as these two elements are related to the plane and in-plane component. It should be mentioned that C44 does not depend According to Euler’s laws on the rotation, all physical matrices in a new orientation are obtained by a transformation matrices methodology [77]. Figure 4a illustrates the dependence of Cij against the inclination angle θ. It is clearly seen that, when changing θ from −90◦to 90 ◦, a symmetric behavior for all Cij is observed at θ = 0◦. The increase in θ leads to decreases in C11, C12, and C13. An opposite dependence is observed between C33 and C11 due to the c-axis inclination angle, as these two elements are related to the out-of-plane and in-plane component. It should be mentioned that C44 does not depend on the inclination angle, which can be explained by the fact that x2 and x2 ′ present the same axis. According to Euler’s laws on the rotation, all physical matrices in a new orientation are obtained by a transformation matrices methodology [77]. Figure 4a illustrates the de- pendence of Cij against the inclination angle θ. It is clearly seen that, when changing θ from −90° to 90°, a symmetric behavior for all Cij is observed at θ = 0°. The increase in θ leads to decreases in C11, C12, and C13. An opposite dependence is observed between C33 and C11 due to the c-axis inclination angle, as these two elements are related to the out-of- plane and in-plane component. It should be mentioned that C44 does not depend on the inclination angle, which can be explained by the fact that x2 and x2′ present the same axis. p p p inclination angle, which can be explained by the fact that x2 and x2′ prese (a) (b) Figure 4. Cont. an be explained by the fact that x2 and x2′ p (b) inclination angle, which (b) (a) Figure 4. Cont. 9 of 22 10 of 25 Crystals 2022, 12, 1431 Crystals 2022, 12, x FO (c) Figure 4. 3 3 FEM Simulation of PnCs Unit Cell Dispersion Modes 3.3. FEM Simulation of PnCs Unit Cell Dispersion Modes EM Simulation of PnCs Unit Cell Dispersion Modes M Simulation of PnCs Unit Cell Dispersion Modes 3.3. FEM Simulation of PnCs Unit Cell Dispersion Modes In this study, we define a phononic crystal (PnC) made of AlN cylindrical pillars, deposited on a semi-infinite AlN/Al2O3 substrate, arranged in a square lattice. The elementary unit cell, shown in Figure 5a, is repeated periodically in the (x, y) plane with the pillar axis oriented along the z-axis. The filling factor is defined by f = π.r2/a2, where a is the lattice parameter of the phononic crystal and r the radius of the cylindrical pillar. The pillar height is h1, h2 is the thickness of the AlN layer, and h3 is the thickness of the bulk material (Al2O3) chosen to be five times the lattice constant of the PnCs unit cell (h3= 5 × a) In this study, we deﬁne a phononic crystal (PnC) made of AlN cylindrical pillars, de- posited on a semi-inﬁnite AlN/Al2O3 substrate, arranged in a square lattice. The elementary unit cell, shown in Figure 5a, is repeated periodically in the (x, y) plane with the pillar axis oriented along the z-axis. The ﬁlling factor is deﬁned by f = π.r2/a2, where a is the lattice parameter of the phononic crystal and r the radius of the cylindrical pillar. The pillar height is h1, h2 is the thickness of the AlN layer, and h3 is the thickness of the bulk material (Al2O3), chosen to be ﬁve times the lattice constant of the PnCs unit cell (h3 = 5 × a) [81]. IEW 10 of 24 [81]. In the elementary unit cell, we applied, on the upper surface, free mechanical displacement as boundary conditions, whereas the bottom surface of the substrate has fixed boundary conditions (i.e., zero mechanical displacement). On the lateral sides of the unit cell, we applied Bloch–Floquet periodic boundary conditions (PBCs). The Bloch wave vector components (kB) (in the x and y directions) are swept between the high symmetry points of the first irreducible Brillouin zone (Γ-X-M-Γ) represented Figure 5b. Figure 5. (a) PnC unit cell used for the dispersion curve calculation. (b) First irreducible Brillouin zone. Figure 5. (a) PnC unit cell used for the dispersion curve calculation. (b) First irreducible Brillouin zone. Figure 5. (a) PnC unit cell used for the dispersion curve calculation. (b) First irreducible Brillouin o e Figure 5. 3 1 Pi l t i d Di l t i C t t 3.1. Piezoelectric and Dielectric Constants Indeed, the absolute value of ε11 is maximum at θ = 90◦, while ε33 component has a maximum at θ = 0◦, equivalent to the behavior of undoped AlN [78]. This is attributed to the fact that the x1 ′ axis becomes x3, and x3 ′ coincides with negative x1 when θ is equal to 90◦, and is a direct consequence of the Wurtzite crystal structure of group III-Nitrides [79,80]. 3 3 FEM Simulation of PnCs Unit Cell Dispersion Modes 3.3. FEM Simulation of PnCs Unit Cell Dispersion Modes (a) PnC unit cell used for the dispersion curve calculation. (b) First irreducible Brillouin zone. Crystals 2022, 12, 1431 10 of 22 10 of 22 In the elementary unit cell, we applied, on the upper surface, free mechanical dis- placement as boundary conditions, whereas the bottom surface of the substrate has ﬁxed boundary conditions (i.e., zero mechanical displacement). On the lateral sides of the unit cell, we applied Bloch–Floquet periodic boundary conditions (PBCs). The Bloch wave vector components (kB) (in the x and y directions) are swept between the high symmetry points of the ﬁrst irreducible Brillouin zone (Γ-X-M-Γ) represented Figure 5b. p p g The dispersion curves were computed by plotting the calculated eigenfrequency values versus the wave vector magnitude, swept in the ﬁrst irreducible Brillouin zone (Figure 5b). The eigenfrequencies and the corresponding vibration modes of the PnC unit cell structure were computed by solving the dispersion relation using the Finite Element Method (FEM), in the COMSOL Multiphysics software. The FEM has been proven to be an efﬁcient tool to obtain dispersion curves and mechanical displacement ﬁelds in phononic structures [82]. All calculations were performed with the physical parameters reported in Table 3. Table 3. Physical parameters used in the calculations. Table 3. Physical parameters used in the calculations. Material Constants w-AlN Al2O3 C11 (GPa) 376 452 C12 (GPa) 123 150 C13 (GPa) 91 107 C33 (GPa) 354 454 C44 (GPa) 116 132 e15 (C/m2) 0.29 e31 (C/m2) −0.58 e33 (C/m2) 1.45 ε11/ε0 (C/m2) 8.31 11.07 ε33/ε0 9.75 9.48 Young Modulus 109 [Pa] 364.05 Poisson’s ratio 0.24 ρ (kg/m3) 3214.21 3870 Figure 6 depicts the computed acoustic band structure in the ﬁrst irreducible Brillouin zone, along the high symmetry directions, for pillars with a relative height h/a, and radius r/a of 30% and 37.5% respectively. The sound line delimits the bulk modes (gray area) from the surface coupled modes, below the line. The latter correspond to SAW modes with an acoustic energy localized at the surface and/or in the pillars. Under those conditions, the blue area represents an acoustic bandgap, extending from 313 to 350 MHz, with an 11% relative bandwidth, generated by local resonances of pillars interacting with the SAW [58]. EW Table 3. Physical parameters used in the calculations. 3.3.1. Effect of the Structural Parameters quency fc and the bandgap w 3.3.1. Effect of the Structural Parameters quency fc and the bandgap width Bw. These two indicators determine th 3.3.1. Effect of the Structural Parameters quency fc and the bandgap width Bw. These two indicators determine t The band gaps can be characterized by two key indicators, namely the center frequency f c and the bandgap width Bw. These two indicators determine the operating parameters of the PnCs and are deﬁned as [55]: rameters of the PnCs and are defined as [55]: 𝑓c = (𝑓𝑢+ 𝑓𝑙) 2 ⁄ , 𝐵𝑤= 𝑓𝑢−𝑓𝑙 (7) cy lim a fun fc = ( fu + fl)/2, Bw = fu −fl (7) represent respectively the upper and lower frequency lim aps operating parameters have been investigated as a fun where f u and f l represent respectively the upper and lower frequency limits of the band gap. The band gaps operating parameters have been investigated as a function of the geometrical parameters, namely the pillar’s height (h) and radius (r). Figure 7a–c show the phononic crystal band structure for three different characteristic values of (h), i.e., 4, 6, and 8 µm, with a ﬁxed radius value of r = 3 µm and a = 8 µm. The dependence of the forbidden band according to the height of the pillars is highlighted. The absolute band gap of the SAWs is limited to the domain below the sound line. When h = 4 µm, only one absolute band gap is observed (Figure 7a), whereas two band gaps occur for 6 µm, then three for 8 µm. Additionally, when the height of the pillars h increases, the band gaps downshift toward low frequencies and new band gaps appear. According to [58,83], they can be attributed to local resonance bandgaps. ometrical parameters, namely the pillar’s height (h) and radius (r). Figur phononic crystal band structure for three different characteristic values of 8 µm, with a fixed radius value of r = 3 µm and a = 8 µm. The dependence band according to the height of the pillars is highlighted. The absolute SAWs is limited to the domain below the sound line. When h = 4 µm, on band gap is observed (Figure 7a), whereas two band gaps occur for 6 µm, µm. Additionally, when the height of the pillars h increases, the band gap ward low frequencies and new band gaps appear. 3 3 FEM Simulation of PnCs Unit Cell Dispersion Modes 3.3. FEM Simulation of PnCs Unit Cell Dispersion Modes Material Constants w-AlN Al2O3 C11 (GPa) 376 452 C12 (GPa) 123 150 C13 (GPa) 91 107 C33 (GPa) 354 454 C44 (GPa) 116 132 e15 (C/m2) 0.29 e31 (C/m2) −0.58 e33 (C/m2) 1.45 ε11/ε0 (C/m2) 8.31 11.07 ε33/ε0 9.75 9.48 Young Modulus 109 [Pa] 364.05 Poisson’s ratio 0.24 ρ (kg/m3) 3214.21 3870 Table 3. Physical parameters used in the calculations. Table 3. Physical parameters used in the calculations. Figure 6 depicts the computed acoustic band structure in the ﬁrst irreducible Brillouin zone, along the high symmetry directions, for pillars with a relative height h/a, and radius r/a of 30% and 37.5% respectively. The sound line delimits the bulk modes (gray area) from the surface coupled modes, below the line. The latter correspond to SAW modes with an acoustic energy localized at the surface and/or in the pillars. Under those conditions, the blue area represents an acoustic bandgap, extending from 313 to 350 MHz, with an 11% relative bandwidth, generated by local resonances of pillars interacting with the SAW [58]. EW Figure 6. Acoustic band structure of PnC AlN pillars with r = 3 µm, and h = 2.4 µm. The b hatched area corresponds to the position of the acoustic band gap. Figure 6. Acoustic band structure of PnC AlN pillars with r = 3 µm, and h = 2.4 µm. The blue hatched area corresponds to the position of the acoustic band gap. Figure 6. Acoustic band structure of PnC AlN pillars with r = 3 µm, and h = 2.4 µm. The hatched area corresponds to the position of the acoustic band gap. Figure 6. Acoustic band structure of PnC AlN pillars with r = 3 µm, and h = 2.4 µm. The blue hatched area corresponds to the position of the acoustic band gap. Crystals 2022, 12, 1431 11 of 22 11 of 22 3.3.1. Effect of the Structural Parameters quency fc and the bandgap w According to [58,83], tributed to local resonance bandgaps. 12 of 24 PnC band structures of lattice parameter a = 8 µm for AlN pillars of radius r = 3 µm and alues of h: (a) h = 4 µm, (b) h = 6 µm and (c) h = 8 µm. re 8a outlines the evolution of the first and second acoustic band gaps with the tio h/a in the range [0.3, 1]. We find that the height of the pillars plays an im- ffect on both the width and the frequency position of the gaps. Increasing h shifts d t d l f i B i th d ti Figure 7. PnC band structures of lattice parameter a = 8 µm for AlN pillars of radius r = 3 µm and different values of h: (a) h = 4 µm, (b) h = 6 µm and (c) h = 8 µm. Figure 8a outlines the evolution of the ﬁrst and second acoustic band gaps with the aspect ratio h/a in the range [0.3, 1]. We ﬁnd that the height of the pillars plays an important effect on both the width and the frequency position of the gaps. Increasing h shifts the band gaps down toward low frequencies. By comparison, the second acoustic bandgap does not 12 of 24 nC band structures of lattice parameter a = 8 µm for AlN pillars of radius r = 3 µm and Figure 7. PnC band structures of lattice parameter a = 8 µm for AlN pillars of radius r = 3 µm and different values of h: (a) h = 4 µm, (b) h = 6 µm and (c) h = 8 µm. C band structures of lattice parameter a = 8 µm for AlN pillars of radius r = 3 µm and Figure 7. PnC band structures of lattice parameter a = 8 µm for AlN pillars of radius r = 3 µm and different values of h: (a) h = 4 µm, (b) h = 6 µm and (c) h = 8 µm. nC band structures of lattice parameter a 8 µm for AlN pillars of radius r 3 µm and alues of h: (a) h = 4 µm, (b) h = 6 µm and (c) h = 8 µm. 3.3.1. Effect of the Structural Parameters quency fc and the bandgap w e 8a outlines the evolution of the first and second acoustic band gaps with the io h/a in the range [0.3, 1]. We find that the height of the pillars plays an im- Figure 8a outlines the evolution of the ﬁrst and second acoustic band gaps with the aspect ratio h/a in the range [0.3, 1]. We ﬁnd that the height of the pillars plays an important effect on both the width and the frequency position of the gaps. Increasing h shifts the band gaps down toward low frequencies. By comparison, the second acoustic bandgap does not p µ p µ lues of h: (a) h = 4 µm, (b) h = 6 µm and (c) h = 8 µm. e 8a outlines the evolution of the first and second acoustic band gaps with the o h/a in the range [0.3, 1]. We find that the height of the pillars plays an im- Figure 8a outlines the evolution of the ﬁrst and second acoustic band gaps with the aspect ratio h/a in the range [0.3, 1]. We ﬁnd that the height of the pillars plays an important effect on both the width and the frequency position of the gaps. Increasing h shifts the band gaps down toward low frequencies. By comparison, the second acoustic bandgap does not Crystals 2022, 12, 1431 12 of 22 d acou c bandga 12 of 22 d acou c bandga exist until h/a = 0.7. For h/a = 0.7, the ﬁrst and second acoustic bandgaps now extend from 136 to 225 MHz and from 309 to 352 MHz, respectively. Figure 8b shows, as an example, the variation in fc and Bw of the ﬁrst and second bandgap as a function of h, keeping constant a = 8 µm and r = 2.6 µm. For the ﬁrst (resp. second) band gap, the width Bw decreases from 98 MHz (resp. 45 MHz) to 78 MHz (resp. 24 MHz) when h increases from 4 µm (resp. 6µm) to h = 8 µm. Furthermore, as shown in Figure 8b, the center frequency fc1 decreases almost linearly from 339 to 121 MHz when h changes from 2.4 to 8 µm, and fc2 decreases from 331 to 252 MHz when h changes from 5.6 to 8 µm. 3.3.1. Effect of the Structural Parameters quency fc and the bandgap w an example, the variation in fc and Bw of the first and second bandgap as a function keeping constant a = 8 µm and r = 2.6 µm. For the first (resp. second) band gap, dth Bw decreases from 98 MHz (resp. 45 MHz) to 78 MHz (resp. 24 MHz) when h eases from 4 µm (resp. 6µm) to h = 8 µm. Furthermore, as shown in Figure 8b, the cen quency fc1 decreases almost linearly from 339 to 121 MHz when h changes from 2.4 µm, and fc2 decreases from 331 to 252 MHz when h changes from 5.6 to 8 µm. (a) W 13 of 2 (b) gure 8. (a) Evolution of the acoustic bandgaps (gaps map) as a function of the aspect ratio (h/a ) Evolution of the center frequency fc and the bandgap width Bw of the first and second bandgap a function of the height of the pillars of radius r = 2.6 µm. By comparison the influence of the radius r even if optimized on the two bandgap Figure 8. (a) Evolution of the acoustic bandgaps (gaps map) as a function of the aspect ratio (h/a). (b) Evolution of the center frequency fc and the bandgap width Bw of the ﬁrst and second bandgaps as a function of the height of the pillars of radius r = 2.6 µm. By comparison, the inﬂuence of the radius r, even if optimized, on the two bandgaps (a) (b) (b) re 8. (a) Evolution of the acoustic bandgaps (gaps map) as a function of the aspect ratio (h/ volution of the center frequency fc and the bandgap width Bw of the first and second bandga function of the height of the pillars of radius r = 2.6 µm. Figure 8. (a) Evolution of the acoustic bandgaps (gaps map) as a function of the aspect ratio (h/a). (b) Evolution of the center frequency fc and the bandgap width Bw of the ﬁrst and second bandgaps as a function of the height of the pillars of radius r = 2.6 µm. By comparison, the influence of the radius r, even if optimized, on the two bandgap icators, is low (see Figures S1 and S2, Supplementary Materials). Therefore, h = 6 µ r = 2.6 µm represent an appropriate choice to obtain the largest acoustic gap openin SAW-coupled PnCs applications. 3.3.1. Effect of the Structural Parameters quency fc and the bandgap w It is worth noting that modifying the geometric racteristics of the pillars can modulate the acoustic band gaps of pillar-based PnC By comparison, the inﬂuence of the radius r, even if optimized, on the two bandgaps indicators, is low (see Figures S1 and S2, Supplementary Materials). Therefore, h = 6 µm and r = 2.6 µm represent an appropriate choice to obtain the largest acoustic gap openings for SAW-coupled PnCs applications. It is worth noting that modifying the geometrical characteristics of the pillars can modulate the acoustic band gaps of pillar-based PnCs over a large frequency range. By comparison, the influence of the radius r, even if optimized, on the two bandga icators, is low (see Figures S1 and S2, Supplementary Materials). Therefore, h = 6 µ r = 2.6 µm represent an appropriate choice to obtain the largest acoustic gap openin SAW-coupled PnCs applications. It is worth noting that modifying the geometric racteristics of the pillars can modulate the acoustic band gaps of pillar-based Pn By comparison, the inﬂuence of the radius r, even if optimized, on the two bandgaps indicators, is low (see Figures S1 and S2, Supplementary Materials). Therefore, h = 6 µm and r = 2.6 µm represent an appropriate choice to obtain the largest acoustic gap openings for SAW-coupled PnCs applications. It is worth noting that modifying the geometrical characteristics of the pillars can modulate the acoustic band gaps of pillar-based PnCs over a large frequency range. Crystals 2022, 12, 1431 13 of 22 13 of 22 3.3.2. Effect of AlN Doping on the Acoustic Band Gaps 3.3.2. Effect of AlN Doping on the Acoustic Band Gaps As mentioned in Section 1, the elastic, dielectric, and piezoelectric coefﬁcients of AlN are affected by the doping concentration in Sc which, in consequence, has the potential to affect the acoustic bandgap characteristics, i.e., the frequency position fc and the bandwidth Bw. Indeed, as seen Figure 9a, when the Sc concentration increases, both ﬁrst and second band gaps exhibit a consequent shift down toward low frequencies, while their widths are slightly modiﬁed. This is conﬁrmed by a representation of fc and Bw as a function of the Sc concentration (Figure 9b). From this ﬁgure, we notice that the center frequencies vary linearly from 161 to 131 MHz and from 321 to 259 MHz when the Sc concentration increases from 0 to 37.5% for the ﬁrst and second acoustic band gaps, respectively. Such a variation is due to the acoustic velocity which is modiﬁed by the elastic constant Cij and density ρ (g/cm3) of w-ScxAl1-xN. For the two band gaps, the bandgap width (Bw) has been slightly modiﬁed, with a variation of −10% for the ﬁrst gap and +7% for the second. 14 of 24 (a) (b) igure 9. Evolution of (a) the bandgaps map, and (b) the center frequency fc and bandgap width Bw f the first and second bandgaps with Sc concentration (x). Figure 9. Evolution of (a) the bandgaps map, and (b) the center frequency fc and bandgap width Bw of the ﬁrst and second bandgaps with Sc concentration (x). ll d d h d d f h b d l d (a) (a) (b) (b) gure 9. Evolution of (a) the bandgaps map, and (b) the center frequency fc and bandgap width Bw the first and second bandgaps with Sc concentration (x). Figure 9. Evolution of (a) the bandgaps map, and (b) the center frequency fc and bandgap width Bw of the ﬁrst and second bandgaps with Sc concentration (x). Finally, we studied the dependence of the acoustic bandgaps on a tilted c-axis orien- tion angle of a w-Sc0.375Al0.625N PnC-based structure. From the results shown in Figure a,b, it is clear that when θ changes from −90° to 90°, a symmetric behavior with respect θ = 0° is observed for the two acoustic bandgaps, due to the dependence of Cij on the gle of inclination. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic To gains more insight into the involved phononic modes, we report in Figure 11a the dispersion curves of w-Sc0.375Al0.625N pillars with a = 8 µm, h = 6 µm, and r = 2.6 µm, and in Figure 11b the corresponding mechanical displacement field of the first modes, desig- nated by points A, B, C, D, and E. The acoustic branches, passing by A and D, are degen- erated because of the x and y polarization modes appearing at the same frequency. To gains more insight into the involved phononic modes, we report in Figure 11a the dispersion curves of w-Sc0.375Al0.625N pillars with a = 8 µm, h = 6 µm, and r = 2.6 µm, and in Figure 11b the corresponding mechanical displacement ﬁeld of the ﬁrst modes, designated by points A, B, C, D, and E. The acoustic branches, passing by A and D, are degenerated because of the x and y polarization modes appearing at the same frequency. To gains more insight into the involved phononic modes, we report in Figure 11 dispersion curves of w-Sc0.375Al0.625N pillars with a = 8 µm, h = 6 µm, and r = 2.6 µm, a Figure 11b the corresponding mechanical displacement field of the first modes, d nated by points A, B, C, D, and E. The acoustic branches, passing by A and D, are d erated because of the x and y polarization modes appearing at the same frequency. Figure 11. (a) Acoustic band structures of PnC w-Sc0.375Al0.625N pillars on half infinite substrate. (b) Transmission through a finite PnC constituted of an 11-unit cell, with h = 6 µm, r = 2.6 µm, and a = Figure 11. (a) Acoustic band structures of PnC w-Sc0.375Al0.625N pillars on half infinite substra Transmission through a finite PnC constituted of an 11-unit cell, with h = 6 µm, r = 2.6 µm, an Figure 11. (a) Acoustic band structures of PnC w-Sc0.375Al0.625N pillars on half inﬁnite substrate. 3.3.2. Effect of AlN Doping on the Acoustic Band Gaps Additionally, we can see that the second gap is more sensitive to c- Finally, we studied the dependence of the acoustic bandgaps on a tilted c-axis ori- entation angle of a w-Sc0.375Al0.625N PnC-based structure. From the results shown in Figure 10a,b, it is clear that when θ changes from −90◦to 90◦, a symmetric behavior with respect to θ = 0◦is observed for the two acoustic bandgaps, due to the dependence of Cij on the angle of inclination. Additionally, we can see that the second gap is more sensitive to Finally, we studied the dependence of the acoustic bandgaps on a tilted c-axis orien- tion angle of a w-Sc0.375Al0.625N PnC-based structure. From the results shown in Figure 0a,b, it is clear that when θ changes from −90° to 90°, a symmetric behavior with respect θ = 0° is observed for the two acoustic bandgaps, due to the dependence of Cij on the ngle of inclination. Additionally, we can see that the second gap is more sensitive to c- Finally, we studied the dependence of the acoustic bandgaps on a tilted c-axis ori- entation angle of a w-Sc0.375Al0.625N PnC-based structure. From the results shown in Figure 10a,b, it is clear that when θ changes from −90◦to 90◦, a symmetric behavior with respect to θ = 0◦is observed for the two acoustic bandgaps, due to the dependence of Cij on the angle of inclination. Additionally, we can see that the second gap is more sensitive to Crystals 2022, 12, 1431 Crystals 14 of 2215 14 of 2215 c-axis inclination angle (see Figure 10b). All the modiﬁcation of the physical parameters can provide a new perspective for controlling SAW-coupled PnC devices and applications. c-axis inclination angle (see Figure 10b). All the modiﬁcation of the physical parameters can provide a new perspective for controlling SAW-coupled PnC devices and applications. (a) (a) Fi 1 E l i (a) (b) Figure 10. Evolution of (a) the bandgaps map, and (b) the center frequencies fc and bandgap widths Bw of the first and second bandgaps with the angle θ of tilted w-Sc0.375Al0.625Nbased PnCs structure. Figure 10. Evolution of (a) the bandgaps map, and (b) the center frequencies fc and bandgap widths Bw of the ﬁrst and second bandgaps with the angle θ of tilted w-Sc0.375Al0.625N based PnCs structure. (a) (b) Figure 10. 3.3.2. Effect of AlN Doping on the Acoustic Band Gaps Evolution of (a) the bandgaps map, and (b) the center frequencies fc and bandgap w Bw of the first and second bandgaps with the angle θ of tilted w-Sc0.375Al0.625N based PnCs stru 3 3 3 Analysis of Surface Phononic Modes in the SAW-PnCs (b) (b) f (a) the bandgaps map and (b) the center frequenci (b) p and (a) Figure 10. Evolution of (a) the bandgaps map, and (b) the center frequencies fc and bandgap widths Bw of the first and second bandgaps with the angle θ of tilted w-Sc0.375Al0.625Nbased PnCs structure. Figure 10. Evolution of (a) the bandgaps map, and (b) the center frequencies fc and bandgap widths Bw of the ﬁrst and second bandgaps with the angle θ of tilted w-Sc0.375Al0.625N based PnCs structure. Figure 10. Evolution of (a) the bandgaps map, and (b) the center frequencies fc and bandgap w Bw of the first and second bandgaps with the angle θ of tilted w-Sc0.375Al0.625N based PnCs stru 3 3 3 A l i f S f Ph i M d i h SAW P C 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic To achieve this objective, we built a super unit cell containing 11 pillars distributed along the direction of propagation x, bounded with PML, and inﬁnite along y by applying periodic conditions. In this ﬁrst calculation, the incident plane wave was generated roughly by considering a displacement perpendicular to the surface along a line source in front of the phononic crystal. The transmission coefﬁcient, given by (uout/uin), was then recorded after the phononic crystal (Figure 11b). The transmission spectrum presents zero transmissions at frequencies 88, 205, and 282 MHz, which correspond to the frequencies of modes A, C, and D, respectively. One can note that mode C is at the origin of a large band gap compared to modes A and D, while mode B does not lead to any effect in the transmission curve. This is due to the symmetry of mode B, which cannot be excited under the SAW. We now turn to the study of the interaction of the elastic waves generated by the interdigital transducers (IDTs). The Sc0.625Al0.375N phononic crystal was set between the IDTs transmitter and receiver (see Figure 12a). To avoid time-consuming calculations and to focus on the physical effects induced by the Sc concentration, we performed all the following calculations based on a 2D model. In that case, the pillared structure will be transformed in a phononic crystal made of ridges, oriented inﬁnitely along y. Compared to the 3D pillared crystal, it results eigenmodes A′, B′, and D′ operating now at the corresponding frequencies 89, 190, and 277 MHz, obtained by the calculation of dispersion curves and transmission curves performed on the 2D model (see Supplementary Materials). p pp y The input and output IDTs consist of 20 pairs of ﬁngered aluminum (Al) electrodes of thickness 0.2 µm and width λ/4. In the IDTs, the even electrodes are connected with Vin = 1 V (input), while the odd ones are connected to the ground. In the output, Vout is connected between the even electrodes and the odd ones as ground. We ﬁrst generated a SAW of wavelength λ equal to 16 µm through the ScxAl1-xN piezoelectric layer of 2 µm thickness in the x direction. Figure 12b shows the insertion loss (S21) for different values of Sc concentration, i.e., x = 0, 12.5, 25, and 37.5%. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic (b) Transmission through a ﬁnite PnC constituted of an 11-unit cell, with h = 6 µm, r = 2.6 µm, and a = 8 µm (red solid line) compared to the SAW through a non-structured surface (dashed lines). (c) Map of the mechanical displacement ﬁelds of modes (A, B, C, D, and E). Figure 11. (a) Acoustic band structures of PnC w-Sc0.375Al0.625N pillars on half infinite substrate. (b) Transmission through a finite PnC constituted of an 11-unit cell, with h = 6 µm, r = 2.6 µm, and a = Figure 11. (a) Acoustic band structures of PnC w-Sc0.375Al0.625N pillars on half infinite substra Transmission through a finite PnC constituted of an 11-unit cell, with h = 6 µm, r = 2.6 µm, an Figure 11. (a) Acoustic band structures of PnC w-Sc0.375Al0.625N pillars on half inﬁnite substrate. (b) Transmission through a ﬁnite PnC constituted of an 11-unit cell, with h = 6 µm, r = 2.6 µm, and a = 8 µm (red solid line) compared to the SAW through a non-structured surface (dashed lines). (c) Map of the mechanical displacement ﬁelds of modes (A, B, C, D, and E). Crystals 2022, 12, 1431 15 of 22 15 of 22 The calculations of the displacement ﬁelds of modes A, B, C, and D show that the elastic ﬁelds are mostly localized inside the pillars (see Figure 11c). Mode A, occurring at 88 MHz, corresponds to a ﬁrst-order ﬂexural oscillation of the pillar, either in the sagittal plane (x, z) or in the (y, z) plane. Mode B, at 182 MHz, shows a radial oscillation. Mode C, at f = 205 MHz, is a compressional mode, exhibiting an axial elongation of the pillar in the direction z. Finally, mode D at f = 282 MHz is associated with a second-order ﬂexural mode of the pillar. All these modes interact with the Rayleigh wave and generate dispersion branches of more or less slow group velocity [84–87]. Mode E at 413 MHz is located over the sound cone, and corresponds to a radiative mode in the substrate, for which the mechanical displacement ﬁeld penetrates into the bulk of the substrate (w-Sc0.375Al0.625N/Al2O3 structure). p To go further, we calculated the transmission spectra of the ﬁnite 3D PnC with h = 6 µm, r = 2.6 µm, and a = 8 µm to understand the effect of the modes on the SAW propagation. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic (b) Insertion loss (S21) for different values of Sc concentration x of the non-structured surface. gure 12. (a) 2D model used for the calculation of the transmission using the IDT. The PnC w tween the input and output IDTs. (b) Insertion loss (S21) for different values of Sc concentra the non-structured surface. Figure 12. (a) 2D model used for the calculation of the transmission using the IDT. The PnC was set between the input and output IDTs. (b) Insertion loss (S21) for different values of Sc concentration x of the non-structured surface. We first generated a SAW of wavelength 𝝀 equal to 16 µm through the ScxAl1-xN zoelectric layer of 2 µm thickness in the x direction. Figure 12b shows the insertion S21) for different values of Sc concentration, i.e., x = 0, 12.5, 25, and 37.5%. As see igure 12b, we were able to generate an acoustic surface pulse in the frequency ra 355 MHz, 385 MHz) for pure AlN (x = 0%). When now increasing the Sc concentra he maximum of the transmitted amplitude increases and shifts toward the low frequ ies, from 371 MHz (x = 0%) to 360 MHz (x = 37.5%). Under the sameconditions of ex We now turn to the interaction of the SAW generated by the interdigital transducers (IDTs) with the phononic crystal made of parallel ridges. Figure 13 shows the scattering parameters S21 of the Al/ScxAl1-xN/Sapphire for x = 37.5%, obtained with and without the phononic crystal at different wavelengths. The wavelengths were chosen in order to track the frequencies’ eigenmodes of the ridges, namely A′, C′, and D′. To deﬁne the appropriated wavelengths, we used the expression f = vsaw/λ, where the surface wave velocity results from the calculation at x = 37.5%. We found that the wavelengths λ = 64 µm, λ = 30 µm, and λ = 20 µm cover respectively the eigenfrequencies A′, C′, and D′. es, from 371 MHz (x = 0%) to 360 MHz (x = 37.5%). Under the same conditions of ex on (𝜆= 16 µ𝑚) and for pure AlN (x = 0%), the deviation in the center frequency o AW device from the experimental value of 355 MHz obtained by Ginlinger et al. [2 5%. Moreover, their experimental work shows that doping the AlN (27% Sc) enha he performance of ScxAl1-xN-based SAW devices. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic As seen in Figure 12b, we were able to generate an acoustic surface pulse in the frequency range (355 MHz, 385 MHz) for pure AlN (x = 0%). When now increasing the Sc concentration, the maximum of the transmitted amplitude increases and shifts toward the low frequencies, from 371 MHz (x = 0%) to 360 MHz (x = 37.5%). Under the same conditions of excitation (λ = 16 µm) and for pure AlN (x = 0%), the deviation in the center frequency of the SAW device from the experimental value of 355 MHz obtained by Ginlinger et al. [27] is 4.5%. Moreover, their experimental work shows that doping the AlN (27% Sc) enhanced the performance of ScxAl1-xN-based SAW devices. The shift in the passing band toward the low frequencies when x increases comes from the decrease of the phase velocity, given by vsaw = λ × f, where λ is the acoustic wavelength equal to the spatial period of the IDTs and f is the center frequency. Indeed, the phase velocities corresponding to x = 0, 12.5, 25, and 37.5% are respectively 5936, 5856, 5808, and 5760 m/s. The insertion loss (S21) of ScxAl1-xN-based delay line has been increased from −34.52, −32.56, −30.56, and −28.81 dB for the respective concentra- tions x = 0, 12.5, 25, and 37.5%. This behavior agrees the conclusion that the insertion loss (S21) of pure AlN can be improved by a Sc concentration of 37.5%. 16 of 22 16 of 22 Crystals 2022, 12, 1431 Figure 12. (a) 2D model used for the calculation of the transmission using the IDT. The PnC was between the input and output IDTs. (b) Insertion loss (S21) for different values of Sc concentratio of the non-structured surface. Figure 12. (a) 2D model used for the calculation of the transmission using the IDT. The PnC was set between the input and output IDTs. (b) Insertion loss (S21) for different values of Sc concentration x of the non-structured surface. igure 12. (a) 2D model used for the calculation of the transmission using the IDT. The PnC wa etween the input and output IDTs. (b) Insertion loss (S21) for different values of Sc concentrat f the non-structured surface. Figure 12. (a) 2D model used for the calculation of the transmission using the IDT. The PnC was set between the input and output IDTs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic The shift in the passing band toward the low frequencies when x increases co om the decrease of the phase velocity, given by 𝑣𝑠𝑎𝑤= 𝜆× 𝑓, where λ is the aco µ µ p y g q In all cases, it is seen in Figure 13 that the electrical performances of S21 are affected in the vicinity of the SAW central frequency. In other words, the propagation of the SAW is disturbed by the presence of the ridged structure, resulting in a decrease in the mechanical energy of the IDT receiver. S21 also presents a slight frequency shift toward low frequencies because of the mass loading effect caused by the presence of the PnC [87,88]. S21 in the delay line also shows small oscillations, which are induced by Fabry–Perot reﬂection from the metallic IDT ﬁngers [89]. 17 of 22 PnC [87,88 y–Perot re 17 of 22 PnC [87,88 y–Perot re Crystals 2022, 12, 1431 (a) (b) (c) Figure 13. Insertion loss (S21) with (red solid lines) and without (black solid lines) PnCs for dif SAW device wavelengths, (a) λ = 64 µm, (b) λ = 30 µm, and (c) λ = 20 µm closed to the eigenm A, C, and D. Figure 13. Insertion loss (S21) with (red solid lines) and without (black solid lines) PnCs for different SAW device wavelengths, (a) λ = 64 µm, (b) λ = 30 µm, and (c) λ = 20 µm closed to the eigenmodes A, C, and D. x FOR PEER REVIEW 19 (a) (b) (a) (c) (c) Figure 13. Insertion loss (S21) with (red solid lines) and without (black solid lines) PnCs for dif SAW device wavelengths, (a) λ = 64 µm, (b) λ = 30 µm, and (c) λ = 20 µm closed to the eigenm A, C, and D. Figure 13. Insertion loss (S21) with (red solid lines) and without (black solid lines) PnCs for different SAW device wavelengths, (a) λ = 64 µm, (b) λ = 30 µm, and (c) λ = 20 µm closed to the eigenmodes A, C, and D. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic OR PEER REVIEW 19 Finally, using the values of the elastic constants Cij (Figure 4a), and the piezoelectric eij (Figure 4b) and dielectric values εij (Figure 4c), the dependence of the insertion loss (S21) of the SAW delay line without PnCs versus the c-axis tilted angle (θ◦) of Sc0.375Al0.625N was examined for θ = 20◦, 40◦, 60◦and 80◦, and compared to that when the c-axis is normally oriented (θ = 0◦). The result is reported in Figure 14a, which shows that the insertion loss (S21) is affected by the c-axis tilting. The variation in the insertion loss is directly linked to the electromechanical coupling coefﬁcient, which is very sensitive to the tilted angle [90]. Finally, using the values of the elastic constants Cij (Figure 4a), and the piezoele eij (Figure 4b) and dielectric values ɛij (Figure 4c), the dependence of the insertion loss of the SAW delay line without PnCs versus the c-axis tilted angle (θ°) of Sc0.375Al0.625N examined for θ = 20°, 40°, 60° and 80°, and compared to that when the c-axis is norm oriented (θ = 0°). The result is reported in Figure 14a, which shows that the insertion (S21) is affected by the c-axis tilting. The variation in the insertion loss is directly linke the electromechanical coupling coefficient, which is very sensitive to the tilted angle Figure 14. (a) Insertion loss (S21) of the SAW delay line without PnCs of w-Sc0.375Al0.625N. (b) Inse loss (S21) and frequency of SAW delay line without PnCs versus the c-axis tilted angle (θ°) o Sc0.375Al0.625N. Figure 14. (a) Insertion loss (S21) of the SAW delay line without PnCs of w-Sc0.375Al0.625N. (b) Insertion loss (S21) and frequency of SAW delay line without PnCs versus the c-axis tilted angle (θ◦) of w-Sc0.375Al0.625N. Figure 14. (a) Insertion loss (S21) of the SAW delay line without PnCs of w-Sc0.375Al0.625N. (b) Inse loss (S21) and frequency of SAW delay line without PnCs versus the c-axis tilted angle (θ°) Sc0.375Al0.625N. Figure 14. (a) Insertion loss (S21) of the SAW delay line without PnCs of w-Sc0.375Al0.625N. (b) Insertion loss (S21) and frequency of SAW delay line without PnCs versus the c-axis tilted angle (θ◦) of w-Sc0.375Al0.625N. Crystals 2022, 12, 1431 18 of 22 18 of 22 Figure 14b shows that the minimum value of insertion loss is found at the tilting angle θ = 60◦. 4. Conclusions In this work, the effects of Sc concentrations on electro-acoustic material properties were theoretically investigated for w-ScxAl1-xN in the x range from 0 to 37.5% by means of Density Functional Theory. The calculated elastic, piezoelectric, and dielectric properties using the GGA-PBE function show very good agreement with experiments and theoretical works. For all considered Sc contents, w-ScxAl1-xN material exhibits good mechanical stability criteria. By increasing the Sc concentrations for pure AlN, the elastic constants C11 and C33 decrease, whereas the piezoelectric (e33) and dielectric (ε33) constants increase, enhancing the performance of the SAW devices based on w-Sc0.375Al0.625N. Furthermore, the dependences of the electro-acoustic properties (Cij, eij, and εij) with the tilted angle of w-Sc0.375Al0.625N was investigated. It is observed that all these properties exhibit a symmetric behavior at a 0◦tilted angle, whereas an opposite trend for both C11, C33, ε11, and ε33 was found at the tilted angle of 45◦. The material becomes non-piezoelectric at the tilted angle of 90◦. The electro-acoustic properties of w-Sc0.375Al0.625N calculated by DFT were used to calculate the dispersion curves of a pillared phononic crystal deposited on top of the substrate. The effects of scandium (Sc) concentration and tilted angle θ ◦of w-Sc0.375Al0.625N on acoustic band gaps and S21 scattering parameters were studied for the ﬁrst time. The geometrical, (h = 6 µm and r = 2.6 µm), and physical (x = 37.5%) parameters were found to be the appropriate choice to obtain a maximum bandwidth of 44.83 MHz. By comparison, a symmetric behavior at a 0◦tilted angle was also revealed for larger acoustic band gaps. An improvement in the S21 intensity (in dB) of SAW delay lines is demonstrated when x = 37.5% for a normally oriented c-axis. We found that 60 ◦is an optimal tilted angle to improve the insertion loss (S21) from −19 dB for AlN to −12.8 dB for w-Sc0.375Al0.625N. Ongoing work is dealing with AlScN-SAW devices coupled with 2D phononic crystal as a highly sensitive micro-sensor for liquid property determination. Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/cryst12101431/s1. The figure S1 describes the evolution of the acoustic bandgap (gap map) as a function of the aspect ratio (r/a). The figure S2 represents the evolution of the center’s frequency fc and the widths Bw of the first and second bandgap as a function of r. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs. 3.3.3. Analysis of Surface Phononic Modes in the SAW-PnCs 3.3.3. Analysis of Surface Phononic Modes in the SAW To gains more insightinto the involved phononic This is due to the coupling factor, whose value is maximum at this angle, which also in good concordance with experimental data [33]. This maximum coupling factor is due to the large piezoelectric constant e15 and e31 of Sc0.375Al0.625N thin ﬁlms (see Figure 4b). In addition, the variation om the resonances frequencies as a function of x behaves as the calculated SAW velocities at λ= 20 µm (5531.4, 5473.2, 5495.6, 5466.4, and 5405 m/s). In summary, we observed a signiﬁcant improvement in the insertion loss at the 60◦ inclined Sc0.375Al0.625N SAW delay line compared to the pure AlN SAW delay line. The same improvement was observed experimentally by A. Kochar et al. [26,30]. Moreover, the passing frequency band of the SAW delay line based on 60◦inclined pure AlN is shifted down to the low frequencies compared to Sc0.375Al0.625N at the same angle. This is due to the inﬂuence of the Sc concentration on the acoustic velocity [91]. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References ScAlN polarization inverted resonators and enhancement of kt 2 in new YbAlN materials for BAW devices. In Proceedings of the IEEE International Ultrasonics Symposium (IUS); 2019; pp. 894–899. [CrossRef] 1 A l A L d d F S h R K F M l O H hl E h d El P f YAlN/S h B d 11. Yanagitani, T.; Jia, J. ScAlN polarization inverted resonators and enhancement of kt 2 in new YbAlN In Proceedings of the IEEE International Ultrasonics Symposium (IUS); 2019; pp. 894–899. [CrossR 12. Assali, A.; Laidoudi, F.; Serhane, R.; Kanouni, F.; Mezilet, O. Highly Enhanced Electro-acoustic Properties of YAlN/Sapphire Based Surface Acoustic Wave Devices for Next Generation of Microelectromechanical Systems. Mater. Today Commun. 2021, 26, 102067. [CrossRef] 13. Laidoudi, F.; Amara, S.; Caliendo, C.; Boubenider, F.; Kanouni, F.; Assali, A. High quality and low los resonator based on chromium-doped AlN on sapphire. Appl. Phys. A 2021, 127, 1–11. [CrossRef] 14. Amara, S.; Kanouni, F.; Laidoudi, F.; Bouamama, K. Low loss surface acoustic wave SAW devices based on Al1-xMxN (M=Cr, Y, Sc) thin ﬁlms. Phys. B: Condens. Matter 2021, 615, 412990. [CrossRef] 15. Fei, C.; Liu, X.; Zhu, B.; Li, D.; Yang, X.; Yang, Y.; Zhou, Q. AlN piezoelectric thin ﬁlms for energy harvesting and acoustic devices. Nano Energy 2018, 51, 146–161. [CrossRef] 16. Aubert, T.; Naumenko, N.; Bartoli, F.; Pigeat, P.; Streque, J.; Ghanbaja, J.; Elmazria, O. Non-leaky longitudinal acoustic modes in ScxAl1-xN/sapphire structure for high-temperature sensor applications. Appl. Phys. Lett. 2019, 115, 83502. [CrossRef] 17. Qi, W.; Lu, Y.; Fung, S.; Jiang, X.; Horsley, D. Scandium Doped Aluminum Nitride Based Piezoelectric Micromachined Ultrasound Transducers. In Proceedings of the Hilton Head Workshop 2016: A Solid-State Sensors, Actuators and Microsystems Workshop, Hilton Head, SC, USA, 5–9 June 2016. 18. Casamento, J.; Chang, C.S.; Shao, Y.-T.; Wright, J.; Muller, D.A.; Xing, H.; Jena, D. Structural and piezoelectric properties of ultra-thin ScxAl1−xN ﬁlms grown on GaN by molecular beam epitaxy. Appl. Phys. Lett. 2020, 117, 112101. [CrossRef] 18. Casamento, J.; Chang, C.S.; Shao, Y.-T.; Wright, J.; Muller, D.A.; Xing, H.; Jena, D. Structural and piezoelectric properties of ultra-thin ScxAl1−xN ﬁlms grown on GaN by molecular beam epitaxy. Appl. Phys. Lett. 2020, 117, 112101. [CrossRef] 19. Akiyama, M.; Kamohara, T.; Kano, K.; Teshigahara, A.; Takeuchi, Y.; Kawahara, N. Enhancement of Piezoelectric Response in ultra-thin ScxAl1−xN ﬁlms grown on GaN by molecular beam epitaxy. Appl. Phys. Lett. 2020, 117, 112101. References 1. Panneerselvam, G.; Thirumal, V.; Pandya, H.M. Review of surface acoustic wave sensors for the detection and identiﬁcation of toxic environmental Gases/Vapours. Arch. Acoust. 2019, 44, 117–127. p 2. Caliendo, C.; Imperatori, P. High-frequency, high-sensitivity acoustic sensor implemented on ALN/Si substrate. Appl. Phys. Lett. 2003, 83, 1641–1643. [CrossRef] 3. Korovin, A.V.; Pennec, Y.; Stocchi, M.; Mencarelli, D.; Pierantoni, L.; Makkonen, T.; Ahopelto, J.; Rouhani, B.D. Conversion between surface acoustic waves and guided modes of a quasi-periodic structured nanobeam. J. Phys. D Appl. Phys. 2019, 52, 32LT01. [CrossRef] 4. Water, W.; Yan, Y.-S.; Meen, T.-H. Effect of magnesium doping on thin ﬁlm. Sens. Actuators A: Phys. 2008, 144, 105–108. [CrossRef] 4. Water, W.; Yan, Y.-S.; Meen, T.-H. Effect of magnesium doping on the structural and piezoelectric properties of sputtered ZnO thin ﬁlm. Sens. Actuators A: Phys. 2008, 144, 105–108. [CrossRef] 4. Water, W.; Yan, Y.-S.; Meen, T.-H. Effect of magnesium doping on the structural and piezoelec thin ﬁlm. Sens. Actuators A: Phys. 2008, 144, 105–108. [CrossRef] R.; Takata, E.; Tanaka, K.; Ohwada, K. Characterization of piezoelectric properties of PZT thin ﬁlms deposited ring. Sens. Actuators A: Phys. 1994, 45, 125–129. [CrossRef] 5. Toyama, M.; Kubo, R.; Takata, E.; Tanaka, K.; Ohwada, K. Characterization of piezoelectric propertie on Si by ECR sputtering. Sens. Actuators A: Phys. 1994, 45, 125–129. [CrossRef] 6. Lin, C.-M.; Chen, Y.-Y.; Felmetsger, V.V.; Lien, W.-C.; Riekkinen, T.; Senesky, D.; Pisano, A.P. Surface acoustic wave devices on AlN/3C–SiC/Si multilayer structures. J. Micromech. Microeng. 2013, 23, 25019. [CrossRef] y g 7. Deger, C.; Born, E.; Angerer, H.; Ambacher, O.; Stutzmann, M.; Hornsteiner, J.; Riha, E.; Fischerauer, G. Sound velocity of AlxGa1−xN thin ﬁlms obtained by surface acoustic-wave measurements. Appl. Phys. Lett. 1998, 72, 2400–2402. [CrossRef] y pp y 8. Doll, J.C.; Petzold, B.C.; Ninan, B.; Mullapudi, R.; Pruitt, B.L. Aluminum nitride on titanium for CMOS compatible piezoelectric transducers. J. Micromechanics Microengineering 2009, 20, 25008. [CrossRef] g g 9. Kurz, N.; Ding, A.; Urban, D.F.; Lu, Y.; Kirste, L.; Feil, N.M.; Žukauskait˙e, A.; Ambacher, O. Experimental determination of the electro-acoustic properties of thin ﬁlm AlScN using surface acoustic wave resonators. J. Appl. Phys. 2019, 126, 075106. [CrossRef] 10. Mayrhofer, P.; Riedl, H.; Euchner, H.; Stöger-Pollach, M.; Bittner, A.; Schmid, U. Microstructure and piezoelectric response of Y Al1−N thin ﬁlms. Acta Mater. 2015, 100, 81–89. [CrossRef] 11. Yanagitani, T.; Jia, J. Acknowledgments: This work was supported by the General Directorate of Scientiﬁc Research and Technological Development DGRSDT of Algeria. The High-Performance Computing Network of the University of Setif 1 is acknowledged for computer time. 4. Conclusions The figure S3 shows the dispersion curves, the transmission spectrum and displacement field of modes A’, B’, and C’. Author Contributions: Conceptualization, F.A., F.K. and R.S.; methodology, F.A., F.K. and Y.P.; formal analysis, F.A., Z.Ö. and K.B.; writing—original draft preparation, F.A.; supervision, Y.P. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data are available upon request to the corresponding authors. Crystals 2022, 12, 1431 19 of 22 19 of 22 Acknowledgments: This work was supported by the General Directorate of Scientiﬁc Research and Technological Development DGRSDT of Algeria. The High-Performance Computing Network of the University of Setif 1 is acknowledged for computer time. Acknowledgments: This work was supported by the General Directorate of Scientiﬁc Research and Technological Development DGRSDT of Algeria. The High-Performance Computing Network of the University of Setif 1 is acknowledged for computer time. References [CrossRef] 19. Akiyama, M.; Kamohara, T.; Kano, K.; Teshigahara, A.; Takeuchi, Y.; Kawahara, N. Enhancement of Piezoelectric Response in Scandium Aluminum Nitride Alloy Thin Films Prepared by Dual Reactive Cosputtering. Adv. Mater. 2008, 21, 593–596. [CrossRef] 20. Akiyama, M.; Kano, K.; Teshigahara, A. Inﬂuence of growth temperature and scandium concentration on piezoelectric response of scandium aluminum nitride alloy thin ﬁlms. Appl. Phys. Lett. 2009, 95, 162107. [CrossRef] 19. Akiyama, M.; Kamohara, T.; Kano, K.; Teshigahara, A.; Takeuchi, Y.; Kawahara, N. Enhancement of Piezoelectric Response in Scandium Aluminum Nitride Alloy Thin Films Prepared by Dual Reactive Cosputtering. Adv. Mater. 2008, 21, 593–596. [CrossRef] 20 Aki M K K T hi h A I ﬂ f th t t d di t ti i l t i 19. Akiyama, M.; Kamohara, T.; Kano, K.; Teshigahara, A.; Takeuchi, Y.; Kawahara, N. Enhancement of Piezoelectric Response in Scandium Aluminum Nitride Alloy Thin Films Prepared by Dual Reactive Cosputtering. Adv. Mater. 2008, 21, 593–596. [CrossRef] 20. Akiyama, M.; Kano, K.; Teshigahara, A. Inﬂuence of growth temperature and scandium concentration on piezoelectric response of scandium aluminum nitride alloy thin ﬁlms. Appl. Phys. Lett. 2009, 95, 162107. [CrossRef] 20. Akiyama, M.; Kano, K.; Teshigahara, A. Inﬂuence of growth temperature and scandium concentrat of scandium aluminum nitride alloy thin ﬁlms. Appl. Phys. Lett. 2009, 95, 162107. [CrossRef] o sca d u a u u t de a oy t s pp y 009, , 6 0 [C oss e ] 21. Wingqvist, G.; Tasnadi, F.; Zukauskaite, A.; Birch, J.; Arwin, H.; Hultman, L. Increased electromechanical coupling in w−ScxAl1−xN. Appl. Phys. Lett. 2010, 97, 112902. [CrossRef] 21. Wingqvist, G.; Tasnadi, F.; Zukauskaite, A.; Birch, J.; Arwin, H.; Hultman, L. Increased ele w−ScxAl1−xN. Appl. Phys. Lett. 2010, 97, 112902. [CrossRef] pp y 22. Konno, A.; Sumisaka, M.; Teshigahara, A.; Kano, K.; Hashimo, K.-Y.; Hirano, H.; Esashi, M.; Kadota, M.; Tanaka, S.; Konno, A.; et al. ScAlN Lamb wave resonator in GHz range released by XeF2 etching. In Proceedings of the 2013 IEEE International Ultrasonics Symposium (IUS), Prague, Czech Republic, 21–25 July 2013; 2013; pp. 1378–1381. [CrossRef] 23. Wang, W.; Mayrhofer, P.M.; He, X.; Gillinger, M.; Ye, Z.; Wang, X.; Bittner, A.; Schmid, U.; Luo, J. High performance AlScN thin ﬁlm based surface acoustic wave devices with large electromechanical coupling coefﬁcient. Appl. Phys. Lett. 2014, 105, 133502. [CrossRef] 20 of 22 20 of 22 Crystals 2022, 12, 1431 24. References Bartoli, F.; Moutaouekkil, M.; Streque, J.; Pigeat, P.; Hage-Ali, S.; Boulet, P.; M’Jahed, H.; Elmazria, O Theoretical and experimental study of ScAlN/Sapphire structure based SAW sensor. IEEE Sens. 20 p y / pp , [ ] 31. Kochhar, A.; Yamamoto, Y.; Teshigahara, A.; Hashimoto, K.-Y.; Tanaka, S.; Esashi, M. NSPUDT using c-axis tilted ScAlN thin ﬁlm. In Proceedings of the Joint Conference of the IEEE International Frequency Control Symposium & the European Frequency and Time Forum, Orlando, FL, USA, 14–18 April 2015; pp. 633–636. [CrossRef] p pp 32. Tokuda, S.; Takayanagi, S.; Matsukawa, M.; Yanagitani, T. Film growth of c-axis tilted ScAlN on the sapphire substrate for SAW devices. In Proceedings of the 2017 IEEE International Ultrasonics Symposium (IUS), Washington, DC, USA, 6–9 September 2017; pp. 1–4. [CrossRef] pp [ ] 33. Tominaga, T.; Takayanagi, S.; Yanagitani, T. c-axis-tilted ScAlN ﬁlm on silicon substrate for surface acoustic wave device. In Proceedings of the 2021 IEEE International Ultrasonics Symposium (IUS), Xi’an, China, 11–16 September 2021; pp. 1–4. [CrossRef] 34. Tominaga, T.; Takayanagi, S.; Yanagitani, T. c-Axis-tilted ScAlN ﬁlms grown on silicon substrates for surface acoustic wave devices. Jpn. J. Appl. Phys. 2022, 61, SG1054. [CrossRef] pp 33. Tominaga, T.; Takayanagi, S.; Yanagitani, T. c-axis-tilted ScAlN ﬁlm on silicon substrate for surface acoustic wave device. In Proceedings of the 2021 IEEE International Ultrasonics Symposium (IUS), Xi’an, China, 11–16 September 2021; pp. 1–4. [CrossRef] 34 Tominaga T ; Takayanagi S ; Yanagitani T c-Axis-tilted ScAlN ﬁlms grown on silicon substrates for surface acoustic wave g y p ( ), , , p 34. Tominaga, T.; Takayanagi, S.; Yanagitani, T. c-Axis-tilted ScAlN ﬁlms grown on silicon substrates fo devices. Jpn. J. Appl. Phys. 2022, 61, SG1054. [CrossRef] Jp J pp y [ ] 35. Djafari-Rouhani, B.; Maradudin, A.A.; Wallis, R.F. Rayleigh waves on a superlattice stratiﬁed normal to the surface. Phys. Rev. B 1984, 29, 6454–6462. [CrossRef] 36. Tanaka, Y.; Tamura, S.-I. Surface acoustic waves in two-dimensional periodic elastic structures. Phys. Rev. B 1998, 58, 7958–7965. [CrossRef] 37. Wu, T.-T.; Wu, L.-C.; Huang, Z.-G. Frequency band-gap measurement of two-dimensional air/silicon phononic crystals using layered slanted ﬁnger interdigital transducers. J. Appl. Phys. 2005, 97, 094916. [CrossRef] 38. Hsu, J.-C.; Lin, Y.-D. Microparticle concentration and separation inside a droplet using phononic-crystal scattered standing surface acoustic waves. Sens. Actuators A Phys. 2019, 300, 111651. [CrossRef] 39. Benchabane, S.; Khelif, A.; Rauch, J.-Y.; Robert, L.; Laude, V. References Teshigahara, A.; Hashimoto, K.-Y.; Akiyama, M. Scandium aluminum nitride: Highly piezoelectric thin film for RF SAW devices in multi GHz range. In Proceedings of the IEEE International Ultrasonics Symposium, Dresden, Germany, 7–10 October 2012; pp. 1–5. [CrossRef] M.Y.; Huang, J.-L.; Lii, D.-F. Characterization and Piezoelectric Properties of Reactively Sputtered (Sc, Al) mond Structure. Int. J. Appl. Ceram. Technol. 2013, 11, 894–900. [CrossRef] [ ] 25. Wu, S.; Wu, M.Y.; Huang, J.-L.; Lii, D.-F. Characterization and Piezoelectric Properties of Reactive Films on Diamond Structure. Int. J. Appl. Ceram. Technol. 2013, 11, 894–900. [CrossRef] 26. Kochhar, A.; Yamamoto, Y.; Teshigahara, A.; Hashimoto, K.-Y.; Tanaka, S.; Esashi, M. Wave Propagation Direction and c-Axis Tilt Angle Inﬂuence on the Performance of ScAlN/Sapphire-Based SAW Devices. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2016, 63, 953–960. [CrossRef] 27. Gillinger, M.; Shaposhnikov, K.; Knobloch, T.; Schneider, M.; Kaltenbacher, M.; Schmid, U. Impact of layer and substrate properties on the surface acoustic wave velocity in scandium doped aluminum nitride based SAW devices on sapphire. Appl. Phys. Lett. 2016, 108, 231601. [CrossRef] 28. Kanouni, F.; Amara, S.; Assali, A.; Arab, F.; Qin, Z. A P-matrix-based model for the frequency analysis of IDT/AlScN/Sapphire SAW-delay line. Sens. Actuators A Phys. 2020, 307, 111980. [CrossRef] y y 29. Gillinger, M.; Knobloch, T.; Schneider, M.; Schmid, U. Harsh Environmental Surface Acoustic Wave T 29. Gillinger, M.; Knobloch, T.; Schneider, M.; Schmid, U. Harsh Environmental Surface Acoustic Wave Temperature Sensor Based on Pure and Scandium doped Aluminum Nitride on Sapphire. Multidiscip. Digit. Publ. Inst. Proc. 2017, 1, 341. [CrossRef] 30. Bartoli, F.; Moutaouekkil, M.; Streque, J.; Pigeat, P.; Hage-Ali, S.; Boulet, P.; M’Jahed, H.; Elmazria, O.; Zhgoon, S.; Aubert, T.; et al. 29. Gillinger, M.; Knobloch, T.; Schneider, M.; Schmid, U. Harsh Environmental Surface Acoustic Wave T Pure and Scandium doped Aluminum Nitride on Sapphire. Multidiscip. Digit. Publ. Inst. Proc. 2017 29. Gillinger, M.; Knobloch, T.; Schneider, M.; Schmid, U. Harsh Environmental Surface Acoustic Wave Temperature Sensor Based on Pure and Scandium doped Aluminum Nitride on Sapphire. Multidiscip. Digit. Publ. Inst. Proc. 2017, 1, 341. [CrossRef] 30. Bartoli, F.; Moutaouekkil, M.; Streque, J.; Pigeat, P.; Hage-Ali, S.; Boulet, P.; M’Jahed, H.; Elmazria, O.; Zhgoon, S.; Aubert, T.; et al. h l d l d f S l /S h b d S [C f] p pp p g 30. References Evidence for complete surface wave band gap in a piezoelectric phononic crystal. Phys. Rev. E 2006, 73, 65601. [CrossRef] 40. Yudistira, D.; Pennec, Y.; Rouhani, B.D.; Dupont, S.; Laude, V. Non-radiative complete surface aco ﬁnite-depth holey phononic crystal in lithium niobate. Appl. Phys. Lett. 2012, 100, 61912. [CrossRef] ec, Y.; Rouhani, B.D.; Dupont, S.; Laude, V. Non-radiative complete surface acoustic wave bandgap for hononic crystal in lithium niobate. Appl. Phys. Lett. 2012, 100, 61912. [CrossRef] 41. White, R.M.; Voltmer, F.W. Direct piezoelectric Coupling to surface elastic waves. Appl. Phys. Lett. 1965, 7, 314–316. [CrossRef] 42. Sigalas, M.; Economou, E. Elastic and acoustic wave band structure. J. Sound Vib. 1992, 158, 377–382. [CrossRef] g 43. Li, G.; Ma, F.; Guo, J.; Zhao, H. Case Study of Roadway Deformation Failure Mechanisms: Field Investigation and Numerical Simulation. Energies 2021, 14, 1032. [CrossRef] 44. Mohammadi, S.; Adibi, A. On chip complex signal processing devices using coupled phononic crystal slab resonators and waveguides. AIP Adv. 2011, 1, 41903. [CrossRef] 45. Wu, T.; Sun, J. 4G-3 guided surface acoustic waves in phononic crystal waveguides. In Proceedings of the 2006 IEEE Ultrasonics Symposium 2006, Vancouver, BC, Canada, 3–6 October 2006; pp. 673–676. 46. Salman, A.; Kaya, O.A.; Cicek, A. Determination of concentration of ethanol in water by a linear waveguide in a 2-dimensional phononic crystal slab. Sens. Actuators A Phys. 2014, 208, 50–55. [CrossRef] 47. Wu, T.; Wang, W.; Sun, J. A layered SAW device using phononic-crystal reﬂective gratings. In Proce Ultrasonics Symposium, Beijing, China, 2–5 November 2008; pp. 709–712. 48. Imanian, H.; Noori, M.; Abbasiyan, A. Highly efﬁcient gas sensor based on quasi-periodic phononic crystals. Sens. Actuators B Chem. 2021, 345, 130418. [CrossRef] 49. Ramakrishnan, N.; Palathinkal, R.P.; Nemade, H.B. Mass Loading Effect of High Aspect Ratio Structures Grown Over Surface Acoustic Wave Resonators. Sens. Lett. 2010, 8, 253–257. [CrossRef] 50. Benchabane, S.; Gaiffe, O.; Ulliac, G.; Salut, R.; Achaoui, Y.; Laude, V. Observation of surface-guide phononic crystal. Appl. Phys. Lett. 2011, 98, 171908. [CrossRef] 50. Benchabane, S.; Gaiffe, O.; Ulliac, G.; Salut, R.; Achaoui, Y.; Laude, V. Observation of surface-guided waves in holey hypersonic phononic crystal. Appl. Phys. Lett. 2011, 98, 171908. [CrossRef] 21 of 22 Crystals 2022, 12, 1431 51. Yankin, S.; Talbi, A.; Du, Y.; Gerbedoen, J.-C.; Preobrazhensky, V.; Pernod, P.; Matar, O.B. Finite element analysis and experimental study of surface acoustic wave propagation through two-dimensional pillar-based surface phononic crystal. J. Appl. Phys. References Crystal structure reﬁnement of AlN and GaN. Solid State Commun. 1977, 23, 8 64. Momida, H.; Teshigahara, A.; Oguchi, T. Strong enhancement of piezoelectric constants in ScxAl1−xN: First-principles calculations. AIP Adv. 2016, 6, 65006. [CrossRef] 65. Urban, D.F.; Ambacher, O.; Elsässer, C. First-principles calculation of electroacoustic properties of wurtzite (Al,Sc)N. Phys. Rev. B 2021, 103, 115204. [CrossRef] 66. Ambacher, O.; Christian, B.; Feil, N.; Urban, D.F.; Elsässer, C.; Prescher, M.; Kirste, L. Wurtzite ScAlN, InAlN, and GaAlN crystals, a comparison of structural, elastic, dielectric, and piezoelectric properties. J. Appl. Phys. 2021, 130, 45102. [CrossRef] , ; , ; , ; , ; , ; , ; , , , parison of structural, elastic, dielectric, and piezoelectric properties. J. Appl. Phys. 2021, 130, 45102. [CrossR 67. Caro, M.A.; Zhang, S.; Riekkinen, T.; Ylilammi, M.; Moram, M.A.; Lopez-Acevedo, O.; Molarius, J.; coefﬁcients and spontaneous polarization of ScAlN. J. Physics Condens. Matter 2015, 27, 245901. [CrossR Caro, M.A.; Zhang, S.; Riekkinen, T.; Ylilammi, M.; Moram, M.A.; Lopez-Acevedo, O.; Molarius, J.; La fﬁi t d t l i ti f S AlN J Ph i C d M tt 2015 27 245901 [C R f] M.A.; Zhang, S.; Riekkinen, T.; Ylilammi, M.; Moram, M.A.; Lopez-Acevedo, O.; Molarius, J.; Laurila, T cients and spontaneous polarization of ScAlN. J. Physics Condens. Matter 2015, 27, 245901. [CrossRef] p p y 68. Tsubouchi, K.; Sugai, K.; Mikoshiba, N. AlN material constants evaluation and SAW properties on AlN/Al2O3 and AlN/Si. In Proceedings of the 1981 Ultrasonics Symposium, Chicago, IL, USA, 14–16 October 1981; pp. 375–380. 69. Sotnikov, A.V.; Schmidt, H.; Weihnacht, M.; Smirnova, E.P.; Chemekova, T.Y.; Makarov, Y.N. Elastic and piezoelectric properties of AlN and LiAlO2 single crystals. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2010, 57, 808–811. [CrossRef] T.; Kim, J.; Dalmau, R.; Schlesser, R.; Preble, E.; Jiang, X. High-temperature electromechanical characterization Kim, T.; Kim, J.; Dalmau, R.; Schlesser, R.; Preble, E.; Jiang, X. High-temperature electromechanical characte crystals IEEE Trans Ultrason Ferroelectr Freq Control 2015 62 1880 1887 [CrossRef] 70. Kim, T.; Kim, J.; Dalmau, R.; Schlesser, R.; Preble, E.; Jiang, X. High-temperature electromechanical char crystals. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2015, 62, 1880–1887. [CrossRef] 70. Kim, T.; Kim, J.; Dalmau, R.; Schlesser, R.; Preble, E.; Jiang, X. High-temperature electromechanical characterization of AlN single crystals. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2015, 62, 1880–1887. [CrossRef] ls. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2015, 62, 1880–1887. [CrossRef] 72. References Watt, J.P.; Peselnick, L. Clariﬁcation of the Hashin-Shtrikman bounds on the effective elastic moduli of polycrystals with hexagonal, trigonal, and tetragonal symmetries. J. Appl. Phys. 1980, 51, 1525–1531. [CrossRef] 73. Mayrhofer, P.; Euchner, H.; Bittner, A.; Schmid, U. Circular test structure for the determination of piezoelectric constants of Scx Al1−xN thin ﬁlms applying Laser Doppler Vibrometry and FEM simulations. Sens. Actuators A Phys. 2014, 222, 301–308. [CrossRef] [ ] 74. Manna, S.; Talley, K.R.; Gorai, P.; Mangum, J.; Zakutayev, A.; Brennecka, G.L.; Stevanovi´c, V.; Ciobanu, C.V. Enhanced Piezoelectric Response of AlN via CrN Alloying. Phys. Rev. Appl. 2018, 9, 034026. [CrossRef] p y g y pp [ ] 75. Nye, J.F. Physical Properties of Crystals: Their Representation by Tensors and Matrices; Oxford University Press: Oxford, UK, 1985. y g y pp Properties of Crystals: Their Representation by Tensors and Matrices; Oxford University Press: Oxford, UK, 198 p y g y pp 75. Nye, J.F. Physical Properties of Crystals: Their Representation by Tensors and Matrices; Oxford Universit 75. Nye, J.F. Physical Properties of Crystals: Their Representation by Tensors and Matrices; Oxford University Press: Oxford, UK, 1985. 76 Qin L ; Chen Q ; Cheng H ; Wang Q M Analytical study of dual mode thin ﬁlm bulk acoustic resonators (FBARs) based on 76. Qin, L.; Chen, Q.; Cheng, H.; Wang, Q.-M. Analytical study of dual-mode thin ﬁlm bulk acoustic reso ZnO and AlN ﬁlms with tilted c-axis orientation. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2010, 57 77. Newnham, R.E. Properties of Materials: Anisotropy, Symmetry, Structure; Oxford University Press: Oxford, UK, 2005. , p f py, y y, ; y , , 78. Kong, L.; Zhang, J.; Wang, H.; Ma, S.; Li, F.; Wang, Q.-M.; Qin, L. Simulation study of MEMS piezoelectric vibration energy harvester based on c-axis tilted AlN thin ﬁlm for performance improvement. AIP Adv. 2016, 6, 125128. [CrossRef] 79. Qin, L.; Wang, Q.-M. Analysis of dual-mode thin ﬁlm bulk acoustic resonators based on polar c-axis tilted wurtzite gallium 78. Kong, L.; Zhang, J.; Wang, H.; Ma, S.; Li, F.; Wang, Q.-M.; Qin, L. Simulation study of MEMS piezoelectric vibration energy harvester based on c-axis tilted AlN thin ﬁlm for performance improvement. AIP Adv. 2016, 6, 125128. [CrossRef] 80. Xie, M.-Y.; Tasnádi, F.; Abrikosov, I.A.; Hultman, L.; Darakchieva, V. Elastic constants, composition, and piezolectric polarization in InxAl1−xN: From ab initio calculations to experimental implications for the applicability of Vegard’s rule. References 2014, 115, 244508. [CrossRef] , [ ] 52. Binci, L.; Tu, C.; Zhu, H.; Lee, J.E.-Y. Planar ring-shaped phononic crystal anchoring boundaries for enh of Lamb mode resonators. Appl. Phys. Lett. 2016, 109, 203501. [CrossRef] 53. Ardito, R.; Cremonesi, M.; D’Alessandro, L.; Frangi, A. Application of optimally-shaped phononic crystals to reduce anchor losses of MEMS resonators. In Proceedings of the IEEE International Ultrasonics Symposium (IUS), Montreal, QC, Canada, 5–8 September 2016; pp. 1–3. [CrossRef] 54. Siddiqi, M.W.U.; Lee, J.E.-Y. Quality factor enhancement of AlN-on-Si lamb wave resonators using a hybrid of phononic crystal shapes in anchoring boundaries. In Proceeding of the 20th International Conference on Solid-State Sensors, Actuators and Microsystems &Eurosensors XXXIII (Transducers & Eurosensors XXXIII), Berlin, Germany, 23–27 June 2019; pp. 913–916. [CrossRef] 55. Tong, Y.; Han, T. Anchor Loss Reduction of Lamb Wave Resonator by Pillar-Based Phononic Crystal. Micromachines 2021, 12, 62. [CrossRef] [ ] 56. Pourabolghasem, R.; Dehghannasiri, R.; Eftekhar, A.A.; Adibi, A. Waveguiding Effect in the Gigahertz Frequency Range in Pillar-based Phononic-Crystal Slabs. Phys. Rev. Appl. 2018, 9, 14013. [CrossRef] 57. Dehghannasiri, R.; Eftekhar, A.A.; Adibi, A. Hypersonic Surface Phononic Bandgap Demonstration in a CMOS-Compatible Pillar-Based Piezoelectric Structure on Silicon. Phys. Rev. Appl. 2018, 10, 64019. [CrossRef] y pp 58. Taleb, F.; Darbari, S. Tunable Locally Resonant Surface-Acoustic-Waveguiding Behavior by Acoustoelectric Interaction in ZnO -Based Phononic Crystal. Phys. Rev. Appl. 2019, 11, 24030. [CrossRef] y y pp .P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865. st, H.J.; Pack, J.D. Special points for Brillouin-zone integrations. Phys. Rev. B 1976, 13, 5188. [CrossRef] y y pp 59. Perdew, J.P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. L 59. Perdew, J.P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 60. Monkhorst, H.J.; Pack, J.D. Special points for Brillouin-zone integrations. Phys. Rev. B 1976, 13, 5188. [C , J ; , ; , g pp p y 60. Monkhorst, H.J.; Pack, J.D. Special points for Brillouin-zone integrations. Phys. Rev. B 1976, 13, 518 p p g y 61. Baroni, S.; Giannozzi, P.; Testa, A. Green’s-function approach to linear response in solids. Phys. Rev. Lett. 1987, 58, 1861–1864. [CrossRef] 61. Baroni, S.; Giannozzi, P.; Testa, A. Green’s-function approach to linear response in solids. Phys. R [CrossRef] [ ] 62. COMSOL Multiphysics. Available online: https://www.comsol.com/ (accessed on 10 September 2022). 63. Schulz, H.; Thiemann, K. References Phys. Rev. B 2012, 86, 155310. [CrossRef] 22 of 22 Crystals 2022, 12, 1431 22 of 22 81. Benchabane, S.; Gaiffe, O.; Salut, R.; Ulliac, G.; Laude, V.; Kokkonen, K. Guidance of surface waves in a micron-scale phononic crystal line-defect waveguide. Appl. Phys. Lett. 2015, 106, 81903. [CrossRef] y g pp y 82. Wu, T.-C.; Wu, T.-T.; Hsu, J.-C. Waveguiding and frequency selection of Lamb waves in a plate with a periodic stubbed surface. Phys. Rev. B 2009, 79, 104306. [CrossRef] y 83. Pennec, Y.; Laude, V.; Papanikolaou, N.; Djafari-Rouhani, B.; Oudich, M.; El Jallal, S.; Beugnot, J.-C.; Escalante, J.M.; Martínez, A. Modeling light-sound interaction in nanoscale cavities and waveguides. Nanophotonics 2014, 3, 413–440. [CrossRef] 84. Addouche, M.; Al-Lethawe, M.A.; Choujaa, A.; Khelif, A. Superlensing effect for surface acoustic waves in a pillar-based phononic crystal with negative refractive index. Appl. Phys. Lett. 2014, 105, 023501. [CrossRef] 85. Oudich, M.; Li, Y.; Assouar, B.; Hou, Z. A sonic band gap based on the locally resonant phononic plates with stubs. New J. Phys. 2010, 12. [CrossRef] A.; Benchabane, S.; Robert, L.; Laude, V. Experimental observation of locally-resonant and Bragg band gaps waves in a phononic crystal of pillars. Phys. Rev. B 2011, 83. [CrossRef] 86. Achaoui, Y.; Khelif, A.; Benchabane, S.; Robert, L.; Laude, V. Experimental observation of locally-reson for surface guided waves in a phononic crystal of pillars. Phys. Rev. B 2011, 83. [CrossRef] 87. Tian, Y.; Li, H.; Chen, W.; Lu, Z.; Luo, W.; Mu, X.; Wang, L. A Novel Love Wave Mode Sensor Waveguide Layer with Microphononic Crystals. Appl. Sci. 2021, 11, 8123. [CrossRef] y pp 88. Serhane, R.; Belkhelfa, N.; Hadj-Larbi, F.; Merah, S.; Bakha, Y. Electrical Performances of a Surface Acoustic Wave Device With Inter Digital Transducers Electrodes in Local Resonances. J. Vib. Acoust. AMSE 2021, 143, 011009 [CrossRef]. [CrossRef] 89. Shao, L.; Maity, S.; Zheng, L.; Wu, L.; Shams-Ansari, A.; Sohn, Y.-I.; Puma, E.; Gadalla, M.; Zhang, M.; Wang, C.; et al. Phononic Band Structure Engineering for High- Q Gigahertz Surface Acoustic Wave Resonators on Lithium Niobate. Phys. Rev. Appl. 2019, 12, 14022. [CrossRef] 90. Laidoudi, F.; Boubenider, F.; Caliendo, C.; Hamidullah, M. Numerical Investigation of Rayleigh, Sezawa and Love Modes in C-Axis Tilted ZNO/SI for Gas and Liquid Multimode Sensor. J. Mech. 2019, 36, 7–18. [CrossRef] 91. Caliendo, C.; Hamidullah, M.; Mattioli, F. References Finite Element Modeling and Synthesis of c-axis Tilted AlN TFBAR for Liquid Sensing Applications. Procedia Eng. 2016, 168, 1032–1035. [CrossRef]
https://openalex.org/W4313590179	https://journals.iaepan.pl/apol/article/download/3348/3137	Polish	null	Podlaskie cmentarzyska z grobami w obstawach kamiennych w świetle wyników badań interdyscyplinarnych	Archeologia Polski	2,023	cc-by	23,218	Archeologia Polski, LXVII: 2022 PL ISSN 0003-8180 DOI: 10.23858/APol67.2022.005 Archeologia Polski, LXVII: 2022 PL ISSN 0003-8180 DOI: 10.23858/APol67.2022.005 KATARZYNA SKRZYŃSKAa a Dr Katarzyna Skrzyńska, Ośrodek Interdyscyplinarnych Badań Archeologicznych, Instytut Archeologii i Etnologii PAN, al. Solidarności 105, 00-140 Warszawa, kasiaskrzynska@tlen.pl, ORCID iD: https://orcid.org/0000-0001-5606-1177. 1 Dane dla Rusi wg: Kvâtkovskaâ 1998, ryc. 1; Sedov 2000, ryc. 1 – uwaga, przybliżone usytuo- wanie stanowisk nietworzących spójnego horyzontu chronologicznego; dane dla Polski: por. ryc. 2. PODLASKIE CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH W ŚWIETLE WYNIKÓW BADAŃ INTERDYSCYPLINARNYCH Abstrakt: W artykule prezentowane są wybrane wyniki badań projektu pt. „Początki chrześ- cijaństwa na pograniczu mazowiecko-ruskim”, realizowanego w latach 2013–2016 w Instytucie Archeologii i Etnologii PAN pod kierownictwem prof. dr. hab. Andrzeja Buko. Przedmiotem tych badań były wczesnośredniowieczne cmentarzyska szkieletowe na Podlasiu. Analizy aDNA wykazały, że ludność pochowana w kurhanach genetycznie różniła się od użytkującej tamtejsze cmentarze płaskie – zarówno te z konstrukcjami kamiennymi, jak i bez nich. Antropologiczna analiza porównawcza populacji z cmentarzysk w Surażu i Czarnej Wielkiej wskazuje także na róż- nice biostrukturalne pomiędzy użytkownikami obydwu typów nekropoli płaskich. Wyniki badań wspierają koncepcję o obcym pochodzeniu ludności pochowanej na podlaskich cmentarzyskach z grobami w obstawach kamiennych oraz co najmniej częściowo tej użytkującej nekropole płaskie założone przy głównych tamtejszych grodach. Słowa kluczowe: pogranicze polsko-ruskie, obrządek pogrzebowy, kurhany, groby w obstawach kamiennych, badania paleogenetyczne, badania antropologiczne, wczesne średniowiecze Abstract: The article presents selected research results of a NPRH (National Program for the Development of the Humanities) project “The origins of Christianity in the Masovian-Rus’ bor- derland”, carried out under the direction of Prof. Andrzej Buko in 2013–2016 at the Institute of Archaeology and Ethnology of the Polish Academy of Sciences. The research focused on early medieval inhumation cemeteries in Podlasie. Ancient DNA analyses demonstrated that individu- als buried in the barrow graves were genetically different from the population interred in the flat cemeteries, whether with a stone casing around the grave or not. An anthropological compara- tive analysis of the populations from the cemeteries at Suraż and Czarna Wielka also indicated biostructural differences between the users of the two types of flat cemeteries. The results support the thesis about the foreign origins of the people buried in the cemeteries of graves with stone casing in Podlasie, and the at least partly foreign provenance of those using the burial grounds established around the major strongholds of the region. Keywords: Polish-Rus’ borderland, funerary rites, barrow graves, graves with stone casing, palaeo genetic studies, anthropological research, early medieval times a Dr Katarzyna Skrzyńska, Ośrodek Interdyscyplinarnych Badań Archeologicznych, Instytut Archeologii i Etnologii PAN, al. Solidarności 105, 00-140 Warszawa, kasiaskrzynska@tlen.pl, ORCID iD: https://orcid.org/0000-0001-5606-1177. 118 KATARZYNA SKRZYŃSKA 1. WSTĘP Cmentarzyska z grobami w obstawach kamiennych są wyjątkowym zjawiskiem w średniowiecznej kulturze funeralnej Europy środkowo-wschodniej. Pod wzglę- dem konstrukcji i form mogił oraz organizacji przestrzennej stanowią one fenomen charakterystyczny dla styku etnokulturowego, jaki wykrystalizował się we wczesnym średniowieczu na pograniczu Słowiańszczyzny i ziem Bałtów, pomiędzy Ziemią Nowogrodzką na północnym wschodzie a Mazowszem na południowym zachodzie (ryc. 11; Kordala 2006; Sobolev 2015). 0 200 400 800 800 km Ryc. 1. Zasięg występowania cmentarzysk z grobami w obstawach kamiennych w Europie środkowo- -wschodniej. Opracowała K. Skrzyńska Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Ryc. 1. Zasięg występowania cmentarzysk z grobami w obstawach kamiennych w Europie środkowo- -wschodniej. O ł K Sk ń k Ryc. 1. Zasięg występowania cmentarzysk z grobami w obstawach kamiennych w Europie środkowo- -wschodniej. Opracowała K. Skrzyńska Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K Skrzyńska g. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K Skrzyńska Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Nekropole te wyróżnia kilka głównych cech. Przede wszystkim są one niewąt- pliwie związane ze szkieletowym obrządkiem pogrzebowym i na tej podstawie interpretowane jako miejsca pochówku ludności schrystianizowanej. O tym, że w najstarszym horyzoncie mogło to mieć wymiar wyłącznie formalny, świadczy obecność pochówków ciałopalnych, które reprezentują regresywny nurt obycza- jowości przedchrześcijańskiej, sukcesywnie zanikający już w kontekście nowych realiów kulturowych. Czytelne tam są również inne wątki charakterystyczne dla 119 CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 0 5 m Ryc. 2. Cmentarzysko z grobami w obstawach kamiennych w Łączynie Starym, pow. mławski. Wydzie- lone sektory cmentarne a – mężczyzna; b – kobieta; c – dziecko; d – płeć biologiczna odmienna od kulturowej lub nieustalona. Wg Rauhuta i Długopolskiej 1972, ryc. 119; opracowała K. Skrzyńska Fig. 2. Cemetery of graves with stone casing in Łączyno Stare, Mława district. Separate cemetery sectors a – male; b – female; c – child; d – biological sex different from cultural sex or undetermined 0 5 m Cmentarzysko z grobami w obstawach kamiennych w Łączynie Starym, pow. mławski. Wydzie- lone sektory cmentarne b k b d k d ł ć b l d d k l l b l Ryc. 2. Cmentarzysko z grobami w obstawach kamiennych w Łączynie Starym, pow. mławski. Wydzie- lone sektory cmentarne a – mężczyzna; b – kobieta; c – dziecko; d – płeć biologiczna odmienna od kulturowej lub nieustalona. Wg Rauhuta i Długopolskiej 1972, ryc. 119; opracowała K. Skrzyńska y mężczyzna; b – kobieta; c – dziecko; d – płeć biologiczna odmienna od kulturowej lub nieustalona. Wg Rauhuta i Długopolskiej 1972, ryc. 119; opracowała K Skrzyńska Fig. 2. Cemetery of graves with stone casing in Łączyno Stare, Mława district. Separate cemetery sectors a – male; b – female; c – child; d – biological sex different from cultural sex or undetermined. After Rauhut and Długopolska 1972, Fig. 119; processing K. Skrzyńska Fig. 2. Cemetery of graves with stone casing in Łączyno Stare, Mława district. Separate cemetery sectors a – male; b – female; c – child; d – biological sex different from cultural sex or undetermined. After Rauhut and Długopolska 1972, Fig. 119; processing K. Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Skrzyńska etapu konwersji wyznaniowej i związanej z nim akulturacji/adaptacji nowych prak- tyk, tj.: usytuowanie nekropoli poza strefą bezpośredniej aktywności osadniczej – najczęściej w miejscach naturalnie eksponowanych, brak ujednoliconej orientacji zwłok na osi W–E głową ku zachodowi, obecność wyposażenia zarówno grobów, jak i samych pochówków oraz różnorodnych pobocznych motywów obyczajowości, które można interpretować jako nawiązujące do kształtującej się wówczas funeralnej symboliki apotropeicznej i rezurekcyjnej. Głównym wyróżnikiem tych cmentarzysk, decydującym o ich unikatowym charakterze, pozostają jednak kamienne konstruk- etapu konwersji wyznaniowej i związanej z nim akulturacji/adaptacji nowych prak- tyk, tj.: usytuowanie nekropoli poza strefą bezpośredniej aktywności osadniczej – najczęściej w miejscach naturalnie eksponowanych, brak ujednoliconej orientacji zwłok na osi W–E głową ku zachodowi, obecność wyposażenia zarówno grobów, jak i samych pochówków oraz różnorodnych pobocznych motywów obyczajowości, które można interpretować jako nawiązujące do kształtującej się wówczas funeralnej symboliki apotropeicznej i rezurekcyjnej. Głównym wyróżnikiem tych cmentarzysk, decydującym o ich unikatowym charakterze, pozostają jednak kamienne konstruk- 120 KATARZYNA SKRZYŃSKA Ryc. 3. Rozmieszczenie cmentarzysk w centralnej części dorzecza środkowej Wisły pewnie lub hipotetycznie datowanych na XI–XIII w. –c – cmentarzyska z grobami w obstawach kamiennych (a – badane wykopaliskowo; b – rozpoznane przez amatorów; c – niebadane); d, e – cmentarzyska z grobami pła- kimi oraz w tzw. typie mieszanym, tj. z pojedynczymi mogiłami z brukowanym stropem (d – badane wykopaliskowo; e – niebadane); f – cmentarzyska kurhanowe badane wykopaliskowo. Wg Rauhuta 1971, s. 486–647; Kordali 2006, s. 36–87; Skrzyńskiej-Jankowskiej 2014, załącznik II; Dzika 2015a; opracowała K. Skrzyńska Fig. 3. Distribution of cemeteries in the central part of the middle Vistula basin dated either certainly or tentatively to the 11th–13th centuries –c – cemeteries of graves with stone casing (a – excavated; b – recognized by amateurs; c – not investigated); d, e – cemeteries with flat graves and graves of mixed type, that is, singular graves with a pavement superstructure (d – excavated; e – not investigated); f – cemetery of excavated barrow graves. After Rauhut 1971, pp. 486–647; Kordala 2006, pp. 36–87; Skrzyńska-Jankowska 2014, appendix II; Dzik 2015a; processing K. Skrzyńska 0 – a – b – c – d – e – f 20 30 40 50 km Ryc. 3. Rozmieszczenie cmentarzysk w centralnej części dorzecza środkowej Wisły pewnie lub hipotetycznie datowanych na XI–XIII w. Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska a–c – cmentarzyska z grobami w obstawach kamiennych (a – badane wykopaliskowo; b – rozpoznane przez amatorów; c – niebadane); d, e – cmentarzyska z grobami pła- skimi oraz w tzw. typie mieszanym, tj. z pojedynczymi mogiłami z brukowanym stropem (d – badane wykopaliskowo; e – niebadane); f – cmentarzyska kurhanowe badane wykopaliskowo. Wg Rauhuta 1971, s. 486–647; Kordali 2006, s. 36–87; Skrzyńskiej-Jankowskiej 2014, załącznik II; Dzika 2015a; opracowała K. Skrzyńska Fig. 3. Distribution of cemeteries in the central part of the middle Vistula basin dated either certainly or tentatively to the 11th–13th centuries a–c – cemeteries of graves with stone casing (a – excavated; b – recognized by amateurs; c – not investigated); d, e – cemeteries with flat graves and graves of mixed type, that is, singular graves with a pavement superstructure (d – excavated; e – not investigated); f – cemetery of excavated barrow graves. After Rauhut 1971, pp. 486–647; Kordala 2006, pp. 36–87; Skrzyńska-Jankowska 2014, appendix II; Dzik 2015a; processing K. Skrzyńska CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 121 cje grobów (Rauhut 1971, s. 454–458, 460–461). Są to czworokątne lub owalne w planie obstawy zbudowane z głazów, współtworzące rzędy jednostkowych lub połączonych ze sobą kwater. Do konwencji należą też układane wewnątrz grobów płaszcze z drobniejszych kamieni, o kształcie, zasięgu i liczbie warstw wykazujących znaczne zróżnicowanie między- i wewnątrzregionalne. Różnice takie notowane są także w odniesieniu do form samych obstaw – inaczej konstruowanych na Mazow- szu i na Podlasiu (Miśkiewicz 1981, s. 94; Kordala 2005a, s. 231–235). Rozbieżności te, podkreślane najczęściej przy porównywaniu cmentarzysk mazowieckich i podlaskich, świadczyć mogą o polaryzowaniu się bazowego wzorca sepulkralnego. Mogło to być przestrzennie warunkowane zarówno przez odległość dzielącą poszczególne skupiska ludności kultywującej tę obrzędowość, jak też przez odmienne realia kulturowe, w jakich funkcjonowały poszczególne grupy. Odmien- ności konstrukcyjne świadczą też zapewne o chronologicznym zróżnicowaniu tej tradycji sepulkralnej. Jest to widoczne zarówno przy porównywaniu całych cmen- tarzysk względem siebie, jak też odrębnych grup pochówków, współtworzących typologicznie wydzielające się sektory pojedynczych nekropoli (ryc. 2). Względne zróżnicowanie chronologiczne jest też często sygnalizowane dzięki usytuowaniu pojedynczych mogił identyfikowanych jako skonstruowane według „młodszej” konwencji pomiędzy grobami o typologicznie „starszych” formach (Dzik 2009; 2014b, ryc. 1). i y Przykłady dywersyfikacji przestrzenno-chronologicznej na poziomie makro mogą być analizowane na podstawie regionalnych koncentracji takich cmentarzysk (ryc. 3). Udokumentowano je na terenie Mazowsza: na północ i północny wschód od Płocka oraz na północ od Ciechanowa – w dolinie górnego Orzyca. Fig. 4. Spatial distribution of cemeteries of graves with stone casing a – Tańsk Przedbory, Mława district; b – Skiwy Małe, Siemiatycze district; c – Starogród, Mińsk Mazowiecki district. After Rauhut and Długopolska 1973, Fig. 4 – fragment (a); A. Bieńkow- ska 2010, Fig. 2 – fragment (b); Rauhut 1971, Fig. 44 (c); processing K. Skrzyńska 2 Wyniki datowań 14C wykonanych w ramach projektu badawczego pt. „Populus Masoviae Medi Aevi i zagadka stulecia: kogo chowano w grobach z obstawami kamiennymi na Mazowszu wczes- nośredniowiecznym?”, nr NCN 2019/33/B/HS3/02453 (dalej: PMMA), realizowanego od 2020 r. w Instytucie Archeologii i Etnologii PAN, pod kierunkiem prof. dr. hab. Andrzeja Buko (D. Błasz- czyk – prezentacja wstępnych wyników badań na konferencji „Przeszłość ma przyszłość” w Instytucie Archeologii UW w marcu 2022 r.). Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Kolejne skupiska znane są ze średniowiecznego pogranicza mazowiecko-ruskiego: najwięk- sze jest w okolicach Drohiczyna, a mniejsze – na północny wschód od Łomży i na wschód od Bielska Podlaskiego. To ostatnie być może wyznacza północne peryferie kolejnej takiej koncentracji, zarejestrowanej już na terenie współczesnej Białorusi, na północ od Brześcia – w dorzeczu Leśnej. Wszystkie te skupiska łączy jedna pra- widłowość: pozostają one w konkretnej relacji przestrzennej do ośrodków władzy wyższej rangi – Płocka, Drohiczyna i Brześcia oraz do tych o znaczeniu drugopla- nowym, jak Ciechanów, Łomża, Tykocin i Bielsk. Współtworzące je cmentarze były bowiem zakładane z dala od nich, jednak ciągle jeszcze w zasięgu ich głębokiego zaplecza osadniczego. p g Każdą z tych koncentracji charakteryzuje odrębny obraz kulturowy, współtwo- rzony przez definiowalny archeologicznie zespół konkretnych cech regionalnych: drugorzędnych w odniesieniu do zgeneralizowanego wzorca, nakreślonego przez konstruowanie kamiennych kwater (np. rozmiary, kształt i sposób konstruowania obstaw, liczba warstw kamieni płaszcza wewnętrznego, obecność/brak jamy gro- bowej oraz jej głębokość, asortyment wyposażenia grobowego itp.). Różnice te są czytelne już na etapie wstępnej analizy rozplanowania przestrzennego zbadanych sektorów cmentarnych (ryc. 4) i w widoczny sposób pogłębiają się w miarę posze- rzania i uszczegóławiania zakresu badań. Cechuje je także brak międzyregionalnej synchroniczności zarówno w odniesieniu do początków tradycji pogrzebowej, jak też KATARZYNA SKRZYŃSKA 122 0 0 10 m a c b 0 10 m 10 m Ryc. 4. Rozplanowanie przestrzenne cmentarzysk z grobami w obstawach kamiennych a – Tańsk Przedbory, pow. mławski; b – Skiwy Małe, pow. siemiatycki; c – Starogród, pow. miński. Wg Rauhuta i Długopolskiej 1973, ryc. 4 – fragment (a); A. Bieńkow- skiej 2010, ryc. 2 – fragment (b); Rauhuta 1971, ryc. 44 (c); opracowała K. Skrzyńska 0 10 m a b 0 10 m b 0 c 10 m Ryc. 4. Rozplanowanie przestrzenne cmentarzysk z grobami w obstawach kamiennych a – Tańsk Przedbory, pow. mławski; b – Skiwy Małe, pow. siemiatycki; c – Starogród, pow. miński. Wg Rauhuta i Długopolskiej 1973, ryc. 4 – fragment (a); A. Bieńkow- skiej 2010, ryc. 2 – fragment (b); Rauhuta 1971, ryc. 44 (c); opracowała K Skrzyńska 0 c 10 m c Ryc. 4. Rozplanowanie przestrzenne cmentarzysk z grobami w obstawach kamiennych a – Tańsk Przedbory, pow. mławski; b – Skiwy Małe, pow. siemiatycki; c – Starogród, pow. miński. Wg Rauhuta i Długopolskiej 1973, ryc. 4 – fragment (a); A. Bieńkow- skiej 2010, ryc. 2 – fragment (b); Rauhuta 1971, ryc. 44 (c); opracowała K. 4 Zweryfikowanie wniosków sformułowanych na podstawie analizy znalezisk i zapisków Ludwika E. de Fleury’ego, który rozkopał kilka takich cmentarzysk na Wysoczyźnie Kolneńskiej w latach osiem- dziesiątych i dziewięćdziesiątych XIX w., zapewne nie będzie możliwe z uwagi na prawie całkowite zniszczenie tamtejszych stanowisk (Dzik 2021). 3 Np. cmentarzysko w Łączynie Starym, pow. mławski. Najmłodsze groby, płaskie i pozbawione obstaw, datowane są tam na XVI–XVII w. (D. Błaszczyk – prezentacja wstępnych wyników badań na konferencji „Przeszłość ma przyszłość” w Instytucie Archeologii UW w marcu 2022 r.).i Aevi i zagadka stulecia: kogo chowano w grobach z obstawami kamiennymi na Mazowszu wczes- nośredniowiecznym?”, nr NCN 2019/33/B/HS3/02453 (dalej: PMMA), realizowanego od 2020 r. w Instytucie Archeologii i Etnologii PAN, pod kierunkiem prof. dr. hab. Andrzeja Buko (D. Błasz- czyk – prezentacja wstępnych wyników badań na konferencji „Przeszłość ma przyszłość” w Instytucie Archeologii UW w marcu 2022 r.). 3 Np. cmentarzysko w Łączynie Starym, pow. mławski. Najmłodsze groby, płaskie i pozbawione obstaw, datowane są tam na XVI–XVII w. (D. Błaszczyk – prezentacja wstępnych wyników badań na konferencji „Przeszłość ma przyszłość” w Instytucie Archeologii UW w marcu 2022 r.). 4 Zweryfikowanie wniosków sformułowanych na podstawie analizy znalezisk i zapisków Ludwika E. de Fleury’ego, który rozkopał kilka takich cmentarzysk na Wysoczyźnie Kolneńskiej w latach osiem- dziesiątych i dziewięćdziesiątych XIX w , zapewne nie będzie możliwe z uwagi na prawie całkowite Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Skrzyńska CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 123 do czasu jej dezintegracji, wiążącej się bądź z porzuceniem cmentarzy, bądź z kon- tynuacją ich użytkowania, ale bez konstruowania obstaw wokół młodszych grobów. y ją y y g Najstarsze z tych cmentarzysk reprezentują czytelny horyzont, datowany na Mazowszu północnym od około połowy XI w. (Dulinicz 2005, s. 197; Kordala 2005a, s. 227), a według najnowszych badań być może nawet od początku tego stulecia2. W okolicach Drohiczyna są one wyraźnie młodsze, bo pochodzą dopiero z 2 poł. lub z 4 ćw. tego samego stulecia (Dzik 2012, s. 428–448, 697; Skrzyńska 2019a, s. 283). W tym samym czasie, chociaż najpewniej relatywnie później, pojawiły się one na Wysoczyźnie Kolneńskiej, gdzie interpretowane są jako świadczące o migracji lud- ności spod Drohiczyna (Dzik 2021, s. 154–156, 158–161). Określenie schyłkowej fazy tej tradycji funeralnej jest już trudniejsze. Na Mazowszu północnym iden- tyfikowana jest ona z połową XII w. (Kordala 2006, s. 233), a szerzej z 2 poł. tego samego stulecia i początkiem XIII w. (Kordala 2003, s. 308–309). Istnieją jednak przesłanki ku temu, że niektóre cmentarze użytkowane były nadal, na przykład we wczesnej nowożytności3. Hipotetycznie na koniec XII w. datowany jest też schyłek istnienia tych cmentarzy w okolicach Łomży (Dzik 2021, s. 156)4. Pod Drohiczy- nem korzystano z nich znacznie dłużej, co najmniej do XIV–XV w. (Dzik 2012, s. 733–739), a prawdopodobnie – przy zredukowanej formie konstrukcji kamien- nych – rytuał ten mógł być wygasającym wątkiem praktyk pogrzebowych nawet do XVI–XVII w. (por. Dzik 2010). Jako relikt starych tradycji mógł on utrzymywać się w obyczajowości wspólnot długo będących w kulturowej izolacji, charakteryzującej zapewne peryferie regionu w późnym średniowieczu i wczesnej nowożytności. Być może też stanowił inspirację lub na niektórych obszarach bezpośrednio ewoluował w obyczajowość polegającą na umieszczaniu na grobach kamiennych stel (o nowych badaniach tych cmentarzysk: zob. Lepionka 2020). Przypuszczalnie w tym samym czasie cmentarze te istniały również w dorzeczu Leśnej (zachodnia Białoruś), gdzie, podobnie jak nad środkową i górną Dźwiną, obyczaj ten mógł przetrwać do wczes- nej nowożytności, przy nieznacznej tylko modyfikacji kanonu wykształconego we wczesnym średniowieczu (Kvâtkovskaâ 1998, s. 159–167; Čaraŭko 2018). C k b b h k h d y Cmentarzyska z grobami w obstawach kamiennych są przedmiotem zaintere- sowania archeologów od około 150 lat (Rauhut 1971, s. 438–442; Kordala 2000; Krasnodębski 2003), a mimo to główne kwestie związane z tą formą obrzędowo- 124 KATARZYNA SKRZYŃSKA ści pogrzebowej pozostają nierozstrzygnięte. Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska Przedmiotem polemiki jest przede wszystkim geneza tych cmentarzysk i problem przynależności etnicznej i kulturowej pochowanej tam ludności. Wieloaspektową dyskusję w zakresie tych zagadnień można zaprezentować jako oscylującą w tematycznych granicach dwóch koncep- cji. Pierwsza, roboczo nazwana „autochtoniczną”, zakłada miejscowe pochodzenie zmarłych, więc generuje pogląd o rodzimej proweniencji pochowanych i samej tra- dycji sepulkralnej; druga natomiast, „allochtoniczna”, zakłada obce pochodzenie ludności użytkującej te cmentarze – przynajmniej w pierwszym pokoleniu ich zało- życieli (szczegółowe omówienie poglądów w obu opcjach – Dzik 2014a, s. 87–94). W myśl idei autochtonicznej (Rauhut 1971, s. 460) płaskie groby z konstrukcjami kamiennymi są efektem zmian w obrządku pogrzebowym, związanych z recepcją chrześcijaństwa. Teza ta nie spotkała się jednak z akceptacją badaczy z uwagi na regionalny zasięg tej tradycji funeralnej i brak dowodów powszechnego kultywo- wania jej w innych regionach Polski (Dulinicz 1998, s. 105 – tam literatura). Wątek ten pojawia się jednak w tle innych poglądów uznających miejscową genezę oma- wianych praktyk pogrzebowych (por. niżej). Strukturom kamiennym przypisywano także funkcje symboliczne, których genezy upatrywano albo w praktykach pogańskich (Miśkiewicz 1998, s. 116; Kvâtkovskaâ 1998, s. 49–51), albo w zróżnicowaniu ekonomiczno-społecznym (Sikora 2013 – tam literatura). Według tej ostatniej koncepcji najstarsze groby z konstrukcjami kamien- nymi miałyby należeć do osób z kształtującej się elity. W „autochtonicznym” podejściu mieszczą się też hipotezy zakładające, że budowanie obstaw stanowi reminiscencję kurhanowej obrzędowości pogrzebowej na etapie poprzedzającym chrystianizację (Gierlach 1975, s. 91). W takim ujęciu korzeni kulturowych omawianych praktyk poszukiwano nawet w okresie wpływów rzymskich (Kostrzewski 1939–1948, s. 325; Okuliczowie 1963; Dzik 2012, s. 714). Na terenie północnej Białorusi prowadzono badania mające na celu wykazanie bezpośrednich takich powiązań. Alla V. Kvât- kovskaâ (1998, s. 190; tam literatura), opracowując materiały z tamtejszych cmenta- rzysk, zwróciła uwagę, że często zakładano je w bezpośrednim sąsiedztwie kurhanów zaopatrzonych w konstrukcje kamienne: obwodowe, nasypowe lub podnasypowe ze szkieletami składanymi do grobów według analogicznych zasad. Analiza przedmiotów wchodzących w skład wyposażenia pozwoliła jej na sformułowanie wniosku, że groby z obstawami stanowią formę pośrednią pomiędzy kurhanem a grobem płaskim, w któ- rej kulturową łączność pomiędzy obydwiema tradycjami potwierdza obecność kon- strukcji kamiennych. Wsparła w ten sposób wcześniejsze koncepcje Fridy D. Gurevič, Valentina V. Sedova (starszą)5 i Âroslava G. Zvâruga, pośrednio wskazując, że cmen- tarzyska te łączyć należy z lokalnym lub napływowym, schrystianizowanym i zasy- milowanym osadnictwem jaćwieskim (Kvâtkovskaâ 1998, s. 168–187; tam literatura). y j ( ) Podobnie wypowiadał się Michał Dzik ([2010] 2015, s. 5 Ostatecznie badacz ten opowiedział się za migracją osadników z Mazowsza (Sedov 2000). Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska 132–135), zwracając uwagę na zbieżności kształtów podkurhanowych struktur kamiennych odkry- wanych na Podlasiu z konstrukcjami grobów z obstawami, które, jak zauważył, bardzo często występują w bezpośrednim sąsiedztwie. Nie łączył ich jednak z bał- CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 125 tyckimi wpływami kulturowymi, ale z rozpowszechnioną w kulturze słowiańskiej ideą „domów zmarłych” (Dzik 2012, s. 643–650; tenże 2014a; 2015b, s. 137–151), materializowaną regionalnie przy użyciu kamiennego budulca. Ostatecznie uznał, że na terenie dorzecza środkowego Bugu groby z obstawami kamiennymi są sło- wiańską formą rodzimą, wywodzącą się bezpośrednio od kurhanów z konstruk- cjami kamiennymi, a ich wznoszenia zaprzestano w dobie chrystianizacji w związku z zakazem kultywowania pogańskich praktyk pogrzebowych, rygorystycznie egze- kwowanym przez kościół katolicki, który w ostatnich dziesięcioleciach XI w. miał tam prowadzić intensywną działalność misyjną (Dzik 2012, s. 712–720; tenże 2015b, s. 187–192; tenże 2021, s. 165). ) Koncepcja allochtoniczna zakłada obce pochodzenie ludności użytkującej te cmentarze. Archeologiczna argumentacja tej tezy opiera się m.in. na spostrzeże- niach odnoszących się do Mazowsza i braku „zakorzenienia” osadniczego użyt- kujących je tam społeczności (brak lokalnych śladów osadnictwa sprzed XI w.), pojawienia się tej konstrukcji grobów jako obyczaju w pełni ukształtowanego, nie- mającego miejscowych pierwowzorów, a także na stosunkowo szybkiej jego tamtej- szej dezintegracji (Kordala 2005a, s. 237–238). Wysunięto przy tym tezę o wareskim pochodzeniu osadników grzebiących zmarłych w tej tradycji (Kiersnowska 1992, s. 69–72; Kordala 2006 – tam literatura). Ustalenia te korespondują z tezą o traktacie datowanym na lata trzydzieste XI w., zawartym przez polskiego księcia Kazimierza Odnowiciela i ruskiego – Jarosława Mądrego, którego efektem było m.in. ustano- wienie granicy mazowiecko-ruskiej (Kuczyński [1954–1956] 1956). Przypuszcza się, że mogło wtedy dojść do osadzenia na Mazowszu wojowników ruskich, gdzie ich zadaniem było zagospodarowanie pustek osadniczych i militarna ochrona pogra- niczy: mazowiecko-pruskiego i mazowiecko-ruskiego (Kiersnowska 1992, s. 69). Pomimo licznych zastrzeżeń i głosów krytycznych (m.in. Kowalczyk 2003, s. 124; Piotrowski 2003, s. 184–185; Dzik 2008, s. 261; tenże 2015b, s. 185–186), pogląd ten nie został ostatecznie podważony i nadal funkcjonuje w literaturze, będąc czasami odtwarzanym bez głębszej refleksji (Grzegorczyk 2009). y g ę jl j g y Pomimo bogatej literatury przedmiotu nie udało się więc ustalić, czy opisane wzorce konstrukcyjne grobów i cmentarzy wykształciły się lokalnie na terenach Mazowsza lub Podlasia, czy też zostały skopiowane lub przeniesione z innych części Europy. Zwolennicy teorii allochtonicznej jako teren macierzysty wykrystalizowania się tej tradycji wskazują m.in. Szwecję, a w szczególności cmentarzyska w Birce i Ihre (Kordala 2005b, s. 2. WYNIKI BADAŃ INTERDYSCYPLINARNYCH Pierwszych danych w tym zakresie dostarczyły badania cmentarzysk podlaskich, przeprowadzone w ramach projektu Narodowego Programu Rozwoju Humanistyki, pt. „Początki chrześcijaństwa na pograniczu mazowiecko-ruskim”, realizowanego w latach 2013–2016 w Instytucie Archeologii i Etnologii PAN pod kierunkiem prof. dr. hab. Andrzeja Buko (Buko red. 2019)6, i finansowanego ze środków Narodowego Programu Rozwoju Humanistyki. Poza opracowaniami: historycznym (Pacuski 2019) i wieloma archeologicznymi (por. niżej), w jego ramach wykonano serię prób aDNA pobranych ze szkieletów pochodzących z kurhanów, cmentarzysk z grobami w obstawach kamien- nych i z cmentarzysk z grobami płaskimi (Molak i in. 2019). Zaprezentowano też wyniki badań antropologicznych – zarówno szkieletów z rożnych cmentarzysk, wytypowa- nych do rozpoznania paleogenetycznego (Stanaszek 2019), jak też dotyczących całej pojedynczej nekropoli przygrodowej w Surażu (Gładykowska-Reczycka 2019) z pła- skimi grobami szkieletowymi (Olczak i in. 2019, s. 83–87). Wśród najważniejszych z rozpatrywanych aspektów badawczych znalazła się kwestia pojawienia się, rozkwitu oraz schyłku tradycji użytkowania cmentarzy z grobami w obstawach kamiennych, jako związanych z fazą wprowadzenia inhumacji, łączonej z upowszechnianiem się głównych wymogów chrześcijańskiej liturgii pogrzebu (Skrzyńska 2019a, s. 283–287; Buko i in. 2020, s. 583–586). Najważniejsze wnioski w tym zakresie prezentuję poniżej. Fig. 1. Range of cemeteries with graves with stone casing in central and eastern Europe. Processing K. Skrzyńska 152–153; o innych opcjach: Dzik 2014a, s. 93–94). Z kolei w teorii autochtonicznej za terytoria macierzyste dla obstaw uznawane są północne Mazowsze lub międzyrzecze Bugu i Nurca. W pierwszej alternatywie należy brać pod uwagę szczególnie rejon pogranicza mazowiecko-pruskiego, gdzie znajdowały się nekropole nad górnym Orzycem – najpewniej najstarsze z należących do oma- wianego typu. Niestety, nie może to być jednak uznawane za równoznaczne ze wskazaniem tego regionu jako miejsca, gdzie obyczaj ten kulturowo wyewoluował, chociaż taka próba została dawno temu podjęta (Okuliczowie 1963, s. 113–114). W drugim wariancie za obszar ukształtowania się tego wzorca pogrzebowego uzna- wana jest Wysoczyzna Drohiczyńska (międzyrzecze Bugu i Nurca), gdzie znajdować 126 KATARZYNA SKRZYŃSKA się mają najstarsze tego typu cmentarzyska, wywodzące się bezpośrednio z kurha- nowej formy pochówku (por. Dzik [2010] 2015, s. 132–135).i Pomimo prób weryfikacji dotychczasowych ustaleń poprzez ponawianie analizy źródeł za pomocą doskonalonych metod archeologicznych, dyskusja o genezie cmen- tarzysk z grobami w obstawach kamiennych, a pośrednio o pochodzeniu użytkującej je ludności, od długiego czasu pozostaje więc w stanie wyraźnego impasu. Odczu- walny jest niedostateczny stan rozpoznania cmentarzysk (szczególnie na północnym Mazowszu), a wnioski formułowane na podstawie sporządzania kolejnych zestawień typologicznych w zakresie form grobów i towarzyszących im konstrukcji z kamieni, układu szkieletów oraz analiz przedmiotów wchodzących w skład wystroju i wyposa- żenia zmarłych nie generują już nowych wartości poznawczych. Szansą na przełama- nie tej stagnacji stały się badania interdyscyplinarne, w których wnioskowanie arche- ologiczne weryfikowane jest za pomocą analiz biologicznych i fizykochemicznych. 6 Badania te są kontynuowane w skali poszerzonej o analizy stabilnych izotopów strontu, tlenu, węgla i siarki w projekcie PMMA (zob. przyp. 2). 2.1. DANE ARCHEOLOGICZNE Według wyników nowych badań, szkieletowy obrządek pogrzebowy w mię- dzyrzeczu Bugu i górnej Narwi pojawił się być może jeszcze przed aneksją tego CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 127 obszaru przez Polskę i Ruś, co nastąpić mogło około połowy XI w. (np. Bieniak 1963, s. 172–181; Pacuski 2019, s. 31; Sianko 2019, s. 153–155). Bezpośrednich danych dostarczyły tu badania cmentarzyska w Daniłowie Małym, pow. białostocki, którego najstarsza faza sytuowana jest w 2 ćw. XI w., z dopuszczeniem możliwości przesu- nięcia początkowej fazy nekropoli do przełomu X i XI w. (Olczak i in. 2019, s. 87). Zadokumentowano tam pochówki szkieletowe, którym towarzyszyło kilka ciałopal- nych. Zarejestrowane groby były pozbawione zewnętrznych konstrukcji kamiennych, z wyjątkiem nielicznych, gdzie stwierdzono bruk z drobnych otoczaków. W zestawie- niu z innymi przykładami, cmentarzysko to może być interpretowane jako miejsce pochówku formalnie lub częściowo schrystianizowanej społeczności, która osiedliła się/została osadzona (?) w strategicznym miejscu, związanym z funkcjonowaniem przeprawy rzecznej (Skrzyńska 2019a, s. 288). W młodszym horyzoncie ostatniego dziesięciolecia XI w. i całego XII stulecia sytuować należy płaskie szkieletowe cmen- tarzyska przygrodowe, różniące się od daniłowskiego brakiem grobów ciałopalnych oraz bardziej zaawansowanym wdrożeniem chrześcijańskich zasad organizacji pochówku. Do grupy najstarszych w regionie należą nekropole płaskie w Drohiczy- nie, pow. siemiatycki (Pawlata 1991; Andrzejewski, Sikora 2009, s. 164–165; ci sami 2011, s. 85 i ryc. 4), Surażu, pow. białostocki (A. Bieńkowska 2005) oraz w Święcku- -Strumianach, pow. wysokomazowiecki (Jaskanis 2008, s. 147–256). p y Szkieletowa forma pochówku, jako innowacyjna, wprowadzona została także do tradycyjnego na tych terenach, kurhanowego obrządku pogrzebowego. Jest ona notowana zarówno w kurhanach z wewnętrznymi i zewnętrznymi konstrukcjami kamiennymi, jak też w kopcach ich pozbawionych – usypanych tylko z ziemi. Najstarsze takie kurhany odnotowano w dorzeczu Leśnej, na północ od Brześcia. Początek tego zjawiska odnoszony jest tam do przełomu X i XI w. (np. Korobuškina 1993, s. 45–46). Cezura ta jest jednak niepewna z uwagi na to, że wyznaczono ją arbitralnie, odnosząc do czasu, kiedy obszar ten znalazł się w strefie wpływów już wówczas chrześcijańskiego państwa kijowskiego (Skrzyńska 2019a, s. 278– 283), a być może też do czasu powstania grodu w Brześciu, co także datowane jest na przełom X i XI w. (Lysenko 2007). Niezależnie jednak od tego, dla fazy XI–XII w. powszechna obecność takich kurhanów jest tam dobrze udokumento- wana archeologicznie (Lysenko 1991, s. 43–46). 2.1. DANE ARCHEOLOGICZNE Dotychczasowe badania Dariusza Krasnodębskiego i Hanny Olczak dostarczają dowodów na obecność pochówków szkieletowych także na północny zachód od wskazanego obszaru, tj. na terenie Równiny Bielskiej, gdzie najstarsze szkieletowe groby kurhanowe, w świetle dotych- czasowych wyników badań, datowane są na XII w. (Krasnodębski, Olczak 2019, s. 58–61; Olczak, Krasnodębski 2019a, s. 67–72; ci sami 2019b, s. 74–77). Analo- giczną chronologię mają nieliczne, najstarsze szkieletowe groby kurhanowe zbadane w okolicach Drohiczyna. Dotychczasowe odkrycia dowodzą jednak, że inhumacja w połączeniu z kurhanami nigdy nie rozpowszechniła się tam w skali porównywal- nej z pobliskim dorzeczem Leśnej. Uzasadnienie tej sytuacji nadal pozostaje kwestią dalszych badań (Skrzyńska 2019a, s. 279–280). p y Szkieletowa forma pochówku, jako innowacyjna, wprowadzona została także do tradycyjnego na tych terenach, kurhanowego obrządku pogrzebowego. Jest ona notowana zarówno w kurhanach z wewnętrznymi i zewnętrznymi konstrukcjami kamiennymi, jak też w kopcach ich pozbawionych – usypanych tylko z ziemi. Najstarsze takie kurhany odnotowano w dorzeczu Leśnej, na północ od Brześcia. y y Z fazą organizowania pograniczy w ramach wczesnopaństwowej struktury terytorialnej Polski i Rusi związane są także cmentarzyska z płaskimi grobami 128 KATARZYNA SKRZYŃSKA w obstawach kamiennych. Wnioskując na podstawie dużej ich liczby w okolicach Drohiczyna oraz spójnego datowania ich faz inicjalnych (2 poł. lub 4 ćw. XI w.), przypuszczać można, że wzorzec ten nie został upowszechniony w drodze długo- falowego procesu, lecz w efekcie krótkotrwałej akcji, być może osiedleńczej (tu: koncepcja allochtoniczna?) albo chrystianizacyjnej, przeprowadzonej jednocześnie w wielu punktach (tu teoria autochtoniczna?) i zakończonej symbolicznym sukce- sem polegającym na wprowadzeniu inhumacji. Jednocześnie w kulturze funeralnej społeczności użytkujących te cmentarze początkowo obecne były elementy nie- mieszczące się w chrześcijańskim kanonie praktyk pogrzebowych, np. konstrukcje kamienne, pochówki ciałopalne, brak ujednoliconej orientacji, obecność wyposa- żenia, a w przypadku grobów kobiet – bogatego wystroju. Odejście od obyczajów odbiegających od chrześcijańskiej konwencji postępowało stopniowo (Dzik 2015b, s. 24–26). Składały się na to m.in. takie zjawiska jak: zanik pochówków ciałopal- nych i ujednolicenie orientacji grobów, rezygnacja z ich wyposażania, odejście od przypowierzchniowego chowania zmarłych przysypywanych drobnymi kamie- niami. Obserwowana jest także tendencja do sukcesywnej redukcji form konstrukcji kamiennych i widoczne dążenie do dostosowania wyglądu grobu do chrześcijańskiej konwencji (szeroko na ten temat w: Dzik 2015b, s. 179–203). Efekt ten nie miał więc charakteru radykalnej przemiany, ale był konsekwencją długotrwałego procesu dekompozycji wzorców, które w przypadku wczesnego średniowiecza postrzegane są przez archeologów jako decydujące o odrębności kulturowej użytkujących je społeczności. 2.1. DANE ARCHEOLOGICZNE p Kwestia „obcości” tego obyczaju pogrzebowego zyskuje szczególny wymiar poznawczy w międzyrzeczu górnej Narwi i Bugu oraz Bugu i Krzny, gdzie spo- łeczności konstruujące groby z kamiennymi obstawami w 2 poł. XI i w XII w. mogły współfunkcjonować z tymi, które kultywowały regionalną tradycję, pole- gającą na budowaniu kurhanów. Jak wykazały badania w zachodniej części Przed- polesia (dorzecze Leśnej) i na Wysoczyźnie Siedleckiej, w efekcie chrystianizacji ani forma tych ostatnich, ani towarzyszące jej praktyki pogrzebowe nie uległy zasadniczej zmianie (charakterystyka obrządku ciałopalnego – por. Kalaga 2006; charakterystyka obrządku szkieletowego – por. Korobuškina 1993; Karobuškina 1999). W miejsce rytuału ciałopalnego wprowadzano jednak szkieletowy, według którego zmarłych grzebano pod kurhanami zgodnie z zasadami podobnymi do tych rekonstruowanych dla cmentarzysk z grobami w obstawach kamiennych. Niemniej jednak, pozostałe wątki obrzędowości pogańskiej były tam nadal obecne, m.in. najstarsze kurhanowe pochówki szkieletowe układane były na warstwie popiołu, w moim przekonaniu imitującej pozostałości stosu ciałopalnego. Tradycyjnie popioły te grzebano bowiem wraz ze skremowanymi zwłokami, a obyczaj ten był jedną z cech definiujących obrządek pogrzebowy regionu. Jest to dobrze czytelne na południe od Bugu (por. Kalaga 2006, s. 108–110), a prawdopodobnie również nad Leśną i Narewką, gdzie jak się zdaje, starsze kurhany ciałopalne niejednokrotnie inicjowały też powstawanie cmentarzy, w XI–XII w. już szkieletowych (Korobuškina 1993, s. 115–117, 120–121, 126–127, 130; Krasnodębski, Olczak 2019, s. 57–61). Wyraźna różnica, „wymuszona” inhumacyjną formą pochówku, dotyczyła nato- CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 129 miast liczebności kurhanów współtworzących cmentarze. Badania Joanny Kalagi (2006, s. 131) wskazują, że w wydaniu ciałopalnym były to groby zbiorowe. Oznacza to, że jedna społeczność, w zależności od liczebności jej członków oraz długości czasu zasiedlenia danego miejsca, mogła ich budować o wiele mniej niż byłoby to konieczne dla pochowania pojedynczych osób. Po wdrożeniu inhumacji groby kurhanowe stały się mogiłami głównie indywidualnymi (por. Dzik 2015b, s. 63–64), co sprawiło, że współtworzone przez nie nekropole liczyły często od kilkunastu do nawet ponad stu obiektów (Skrzyńska 2019a, s. 280–281). Pod względem liczebno- ści kwater upodabnia je to do cmentarzysk z grobami w obstawach kamiennych, w których również chowano głównie pojedynczych zmarłych (por. Dzik 2015b, s. 66–68). Niemniej jednak zdaje się, że obydwie formy grobów: kurhanową i obsta- wową stosowano równolegle, co najmniej do czasu zaniku pierwszej z nich w XII i XIII w. Współcześnie teza ta ma jedynie wstępne wsparcie w materiałach archeolo- gicznych z Wysoczyzny Drohiczyńskiej oraz z Równiny Bielskiej (Zoll-Adamikowa 1975, s. 53–56; Olczak i in. 2019, s. 7 Przykłady zestawienia danych: dla kurhanów por. Štyhov 1971, s. 20: nry 110, 112, s. 21: nr 123, s. 23: nr 156 oraz Korobuškina 1993, s. 76, 112, 121–123, 128; dla cmentarzysk z grobami w obstawach por. Kvâtkovskaâ 1998, s. 28, ryc. 1: nry 123, 128, 132, 139, 143. p y y 8 Obydwaj badacze reprezentują przy tym odmienne poglądy na temat genezy tych cmentarzysk na Podlasiu: T. Kordala (2005a, s. 236–237) uznaje je za efekt osadzenia tam ludności po przejęciu 2.1. DANE ARCHEOLOGICZNE 72–77), z uwagi na ograniczone możliwości uściślenia datowania znalezisk do przedziału węższego niż dwa stulecia. Olbrzymi potencjał poznawczy mają w tym aspekcie także cmentarzyska w dorzeczu Leśnej, gdzie groby kurhanowe i płaskie w obstawach często tworzą odrębne sektory tych samych nekropoli lub osobne zespoły, położone jednak w niewielkiej odległości7. Wykopaliskowo przebadano tam tylko kurhany, trudno więc powiedzieć, w jakiej relacji chronologicznej pozostają one z niebadanymi jeszcze grobami w obstawach. j g j p ją y j g Koncepcja zakładająca, że społeczności użytkujące cmentarze z płaskimi gro- bami w obstawach kamiennych nie stanowią rodzimego substratu osadniczego na Podlasiu, a ich tamtejsza obecność jest efektem migracji, nie jest nowa (szczegółowo na ten temat: Dzik 2019, s. 80–88). Badacze radzieccy a potem białoruscy długo uznawali je za miejsca pochówku zeslawizowanych Bałtów, a nawet bezpośrednio Jaćwingów, chociaż na obszarze uznawanym za ich wczesnośredniowieczne teryto- rium, wskazywanym jako macierzysty dla wykrystalizowania się tradycji związanej z konstruowaniem „kamiennych mogił”, brakuje archeologicznego potwierdzenia obecności takich obyczajów zarówno w starszych fazach wczesnego średniowiecza, jak też w X–XIII w. (Nowakowski 2011). W Polsce na długie lata dyskurs o pocho- dzeniu ludności pochowanej na tych cmentarzyskach zdominowała teza o migracji osadników z Mazowsza (m.in. Musianowicz 1955; Rauhut 1971, s. 484–485; Miś- kiewicz 1981, s. 92–104; 112–113; Kiersnowska 1992, s. 71). W opozycji do niej pozostają opinie Tomasza Kordali (2005c; 2006, s. 244–246) oraz Michała Dzika (2019, s. 96–99), wskazujących, że to właśnie na Podlasiu cmen- tarze z konstrukcjami kamiennymi mogły pojawić się jako pierwsze, a dopiero potem (lub równocześnie – Kordala 2005c, s. 54) idea ta została przeniesiona na Mazowsze8. Koncepcja ta jest nie do utrzymania w świetle nowych datowań 130 KATARZYNA SKRZYŃSKA radiowęglowych, wskazujących na to, że mazowieckie cmentarze z konstrukcjami kamiennymi są dużo starsze od podlaskich9. y ą p W starszej literaturze przedmiotu (Musianowicz 1960; Rauhut 1971; Miśkie- wicz 1981) dominuje teza o koegzystencji ludności autochtonicznej, kultywują- cej kurhanowy obrządek pogrzebowy, z napływową z Mazowsza, grzebiącą swych zmarłych w grobach w obstawach kamiennych. Opinia ta, sformułowana w latach sześćdziesiątych i siedemdziesiątych ubiegłego wieku, z uwagi na niedostateczne rozpoznanie archeologiczne długo argumentowana była jedynie faktem wielo- krotnie potwierdzonego, bezpośredniego sąsiedztwa obydwu typów cmentarzysk. Brakowało natomiast weryfikacji wykopaliskowej, pozwalającej na rozpoznanie względnej chronologii takich stanowisk. W zaobserwowanej sytuacji dopusz- czalna jest bowiem zarówno opcja równoległego funkcjonowania obydwu rytów pogrzebowych, jak też ta zakładająca kontynuację użytkowania cmentarzy przy jednoczesnej zmianie formy grobu. kontroli politycznej na tym obszarze przez Jarosława Mądrego, a M. Dzik (2019, s. 102–103, 110–112) wiąże z lokalnym wykrystalizowaniem się nowego wzorca pogrzebu w następstwie intensywnej chry- stianizacji w wydaniu katolickim, która polegać miała na odrzuceniu kurhanowej formy grobu przy jednoczesnym zachowaniu wątku konstruowania obstaw i innych struktur kamiennych. j y y y 9 Wyniki datowań 14C wykonanych w ramach projektu PMMA (D. Błaszczyk – prezentacja wstępnych wyników badań na konferencji „Przeszłość ma przyszłość” w Instytucie Archeologii UW w marcu 2022 r.). 2.1. DANE ARCHEOLOGICZNE Potwierdzenie możliwości pierwszej stano- wiłoby argument wskazujący na odrębność kulturową użytkowników nekropoli z grobami w obstawach kamiennych (chociaż nadal niejednoznaczny). W drugiej wersji, wspierającej koncepcję autochtonicznego charakteru tych populacji, zjawisko to miałoby zaistnieć w efekcie rezygnacji z konstruowania kurhanów (najczęściej ciałopalnych) na rzecz stosowania obstaw kamiennych, bardziej odpowiadających inhumacji wprowadzanej w związku z chrystianizacją (por. Dzik 2012, s. 712–720). Kwestie te, pomimo postępu prac, pozostają nadal nierozstrzygnięte, gdyż ana- lizy materiałów zabytkowych pozyskanych z pojedynczych zespołów cmentarnych lub sąsiadujących ze sobą nekropoli, „birytualnych” pod względem form grobów, wspierają obydwie hipotezy. Starszeństwo kurhanów potwierdzono np. w Zbuczu, pow. bielski i w Korzeniówce Małej, pow. siemiatycki (Dzik [2009–2010] 2011; Krasnodębski, Olczak 2019, s. 61–67), gdzie jednak różnice chronologiczne były bardzo wyraźne (kurhany: koniec IX – 1 poł. X w. i X w., a cmentarzyska z grobami w obstawach: koniec XI–XII w.). Inne pochodzą z tej samej fazy wczesnego śred- niowiecza i nie wykluczają równoległego istnienia obydwu tradycji pogrzebowych, przy założeniu, że formy kurhanowe stanowiły recesywny element obrzędowości. Takie wyniki przyniosły wykopaliska przeprowadzone na Wysoczyźnie Bielskiej w Szczytach-Dzięciołowie, pow. bielski (Olczak, Krasnodębski 2019b, s. 76–77). Podobną wymowę mają także archiwalne materiały z Bacików Dalszych, pow. sie- miatycki (Wysoczyzna Drohiczyńska) analizowane przez Helenę Zoll-Adamikową (1975, s. 53–56), a szczególnie te z badań prowadzonych przez Kazimierza Stołyhwę w Łuzkach, pow. sokołowski (północny skraj Wysoczyzny Siedleckiej). Zarówno w tamtejszych kurhanach, jak i w grobach płaskich z konstrukcjami kamiennymi CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 131 stwierdzono pochówki szkieletowe, a asortyment przedmiotów tworzących inwen- tarze grobowe był w obydwu sektorach podobny i analogicznie datowany (Stołyhwo 1914; Musianowicz 1950/1951)10. Zbliżony schemat zarejestrowano też na północ od Brześcia w dorzeczu Leśnej, gdzie kilka grup kurhanów z pochówkami szkiele- towymi współtworzy zespoły z płaskimi grobami w obstawach. Na razie nie wia- domo, czy sektory te były użytkowane równocześnie, czy powstały we wzajemnym następstwie czasowym, tak jak A.V. Kvâtkovskaâ (1998) sugerowała dla analogicz- nych par stanowisk na północy Białorusi. Poważnym mankamentem jej badań jest to, że wskazany wyżej wniosek nie został zweryfikowany w zestawieniu z danymi dotyczącymi cmentarzysk kurhanowych. Badania tych ostatnich w dorzeczu Leśnej wskazują natomiast, że przejściu od ciałopalenia do inhumacji nie towarzyszyła eliminacja tradycyjnej kurhanowej formy grobu, co czytelne jest na wielu tamtej- szych cmentarzyskach11. Podobne ustalenia poczyniono dla nekropoli w uroczysku Szczekotowo, pow. hajnowski, gdzie jednak starsze kurhany ciałopalne i młodsze szkieletowe tworzyły odrębne sektory (Krasnodębski, Olczak 2019, s. 57–61). 12 Początek osadnictwa słowiańskiego w dorzeczu środkowego Bugu sytuowany jest orientacyjnie w 2 poł. VI–VII w. (Kalaga 2006, s. 178, 186). Tamtejsza schyłkowa faza wczesnego średniowiecza w regionie wiązana jest tradycyjnie z 2 poł. XIII w., chociaż obecnie nie wydaje się to tak oczywiste, jak dotychczas zakładano (Gałecki, Skrzyńska 2021). 10 Pomimo istnienia wiarygodnych opisów lokalizacji stanowiska pochodzących jeszcze z XIX w. i informacji o rozkopaniu tamtejszych grobów poprzedzającym wspomniane wykopaliska (Łuniewski 1883), cmentarzyska nie udało się do tej pory odnaleźć. 11 Np. Chotynowo, Górki, Kustycze, Pokry, Ratajczyce i Wojska w rejonie Kamieńca Litewskiego (por. Korobuškina 1993; Karobuškina 1999). W Górkach i Ratajczycach w sąsiedztwie kurhanów stwierdzono płaskie cmentarzyska z grobami w obstawach kamiennych (por. Kvâtkovskaâ 1998, ryc. 1, nry 132, 139). 10 Pomimo istnienia wiarygodnych opisów lokalizacji stanowiska pochodzących jeszcze z XIX w. i informacji o rozkopaniu tamtejszych grobów poprzedzającym wspomniane wykopaliska (Łuniewski 1883), cmentarzyska nie udało się do tej pory odnaleźć. 11 Np. Chotynowo, Górki, Kustycze, Pokry, Ratajczyce i Wojska w rejonie Kamieńca Litewskiego (por. Korobuškina 1993; Karobuškina 1999). W Górkach i Ratajczycach w sąsiedztwie kurhanów stwierdzono płaskie cmentarzyska z grobami w obstawach kamiennych (por. Kvâtkovskaâ 1998, ryc. 1, nry 132, 139). 12 Początek osadnictwa słowiańskiego w dorzeczu środkowego Bugu sytuowany jest orientacyjnie w 2 poł. VI–VII w. (Kalaga 2006, s. 178, 186). Tamtejsza schyłkowa faza wczesnego średniowiecza w regionie wiązana jest tradycyjnie z 2 poł. XIII w., chociaż obecnie nie wydaje się to tak oczywiste, jak dotychczas zakładano (Gałecki, Skrzyńska 2021). 2.1. DANE ARCHEOLOGICZNE 5b; por. Skrzyńska 2021, s. 71–73; 75–77). W XI–XII w. nastąpić tam mogła reorganizacja struktury zasiedlenia, gdyż starsze fortyfikacje zostały opuszczone, a w niewielkiej odległości od nich zbudowano nowe, inicjując w ten sposób proces formowania się kolejnych skupisk osadniczych (por. ryc. 5b, d; Krasnodębski 2000; 2004; 2006). Ciągłość zasiedlenia potwierdzają tam cmentarzyska kurhanowe, zarówno starsze ciałopalne, jak i młodsze szkieletowe, natomiast w młodszych fazach wczesnego średniowie- cza towarzyszą im te z grobami w obstawach kamiennych i płaskie, na których poświadczone są niespalone pochówki (Dzik 2015a, s. 98–109, 123–129, 133–134; Krasnodębski, Olczak 2019; Olczak, Krasnodębski 2019a; 2019b). Obraz kulturowy Wysoczyzny Siedleckiej (ryc. 5a, b) w IX i X w. współtwo- rzą mikroregiony osadnicze rozlokowane w dolinach rzek, pozostające ze sobą w bezpośredniej relacji przestrzennej i chronologicznej. Charakteryzuje je ciągłość zasiedlenia bez śladów znaczących regresów oraz równomierny przyrost punktów osadniczych, powstających w bezpośrednim sąsiedztwie starszych – „macierzy- stych” (przykłady w: Gałecki, Skrzyńska 2021) lub założeń obronnych, których kilka zbudowano tam w ciągu 2 poł. IX i kilkanaście w X w. (Skrzyńska 2021, s. 75–92). Może to świadczyć o postępującej konsolidacji osadnictwa w ramach tworzących się terytoriów grodowych. Sytuacja ta jest jeszcze lepiej czytelna w XI–XII w., kiedy liczba osad wyraźnie wzrosła – szczególnie przy największych i najstarszych for- tyfikacjach (por. ryc. 5c, d; np. Krzesk – Kalaga 1989, s. 224–239; Niewiadoma – Miśkiewiczowa 1996a, s. 42–65). Powstały też wówczas nowe założenia obronne, stanowiące oparcie dla dalszego rozwoju osadnictwa (przykład – Skrzyńska 2019b). ą p g j (p y y ) Podobne wnioski nasuwają się w odniesieniu do obszarów w dorzeczu górnej Narwi, nad Pulwą i nad strugami współtworzącymi obszar źródłowy Nurca (Kras- nodębski 2011–2012). Od końca VI w. istniał tam kompleks osadniczy w Haćkach (Kobyliński, Szymański 2015), a co najmniej od 3 ćw. IX i w X w. w najlepiej roz- winiętych zespołach osadniczych zbudowane zostały grody (ryc. 5b; por. Skrzyńska 2021, s. 71–73; 75–77). W XI–XII w. nastąpić tam mogła reorganizacja struktury zasiedlenia, gdyż starsze fortyfikacje zostały opuszczone, a w niewielkiej odległości od nich zbudowano nowe, inicjując w ten sposób proces formowania się kolejnych skupisk osadniczych (por. ryc. 5b, d; Krasnodębski 2000; 2004; 2006). Ciągłość zasiedlenia potwierdzają tam cmentarzyska kurhanowe, zarówno starsze ciałopalne, jak i młodsze szkieletowe, natomiast w młodszych fazach wczesnego średniowie- cza towarzyszą im te z grobami w obstawach kamiennych i płaskie, na których poświadczone są niespalone pochówki (Dzik 2015a, s. 14 Z tego regionu znanych jest co najmniej 19 takich nekropoli (Skrzyńska-Jankowska 2014, aneks 2.3, część 3.3). Nie prowadzono tam jednak badań wykopaliskowych. Wiadomo tylko, że na terenie północnej Białorusi cmentarzyska te datowane są od wczesnego średniowiecza aż po fazę wczesnonowożytną (Kvâtkovskaâ 1998; Čaraŭko 2018). Dlatego też nie oznaczono ich na ryc. 5c i d, a zamieszczono jako niebadane na ryc. 3. 13 Na wskazanym obszarze zarejestrowano dotąd co najmniej 84 cmentarzyska kurhanowe (Skrzyńska-Jankowska 2014, aneks 2.3, część 3.1 i 3.2), przy czym na dwudziestu dwóch prowadzono prace wykopaliskowe (Korobuškina 1993). 13 Na wskazanym obszarze zarejestrowano dotąd co najmniej 84 cmentarzyska kurhanowe (Skrzyńska-Jankowska 2014, aneks 2.3, część 3.1 i 3.2), przy czym na dwudziestu dwóch prowadzono prace wykopaliskowe (Korobuškina 1993). 14 Z tego regionu znanych jest co najmniej 19 takich nekropoli (Skrzyńska-Jankowska 2014, aneks 2.3, część 3.3). Nie prowadzono tam jednak badań wykopaliskowych. Wiadomo tylko, że na terenie północnej Białorusi cmentarzyska te datowane są od wczesnego średniowiecza aż po fazę wczesnonowożytną (Kvâtkovskaâ 1998; Čaraŭko 2018). Dlatego też nie oznaczono ich na ryc. 5c i d, a zamieszczono jako niebadane na ryc. 3. 2.1. DANE ARCHEOLOGICZNE Poten- cjał poznawczy opisanego zjawiska ma więc kluczowe znaczenie dla rozstrzygnięcia kwestii homogenności/heterogenności kulturowej wspólnot praktykujących równo- legle (?) obydwie formy obrzędowości. Duże nadzieje w tym zakresie łączyć należy z badaniami cmentarzysk Wysoczyzny Drohiczyńskiej, a możliwości rozpoznania tych zjawisk oferuje np. tamtejszy wczesnośredniowieczny mikroregion osadniczy w górnym biegu rzeki Czarnej, gdzie być może nawet kilku cmentarzyskom z gro- bami w obstawach kamiennych towarzyszą analogicznie datowane kurhany (por. Karwowska red. 2013; 2014; Dzik 2015a, s. 45, 57–76, 138–139, 149–150). Znaczących argumentów przemawiających na korzyść tezy o potencjalnym sko- lonizowaniu tego regionu w XI w. dostarczają wyniki badań nad osadnictwem. Jego rozwój w międzyrzeczu Bugu oraz dolnego i środkowego Nurca, w okresie od VI do XIII w.12, następował bowiem w rytmie zupełnie innym od zaobserwowanego w regionach sąsiednich, tj. wspomnianym już zachodnim Przedpolesiu (dorzeczu Leśnej) oraz na Wysoczyźnie Siedleckiej (Skrzyńska 2021, s. 57–73). Te dwa ostatnie regiony były pod tym względem podobne, gdyż przebieg ich zasiedlenia można definiować w analogicznych etapach: 1. infiltracji (2 poł. VI–VII w.), 2. stabilizacji (VIII–IX w.) oraz 3. konsolidacji, związanej z powstaniem najstarszej sieci grodowej głównie w X w. Charakterystyczna jest dla nich także podfaza reorganizacji struktur zasiedlenia odnotowana dla XI stulecia. 132 KATARZYNA SKRZYŃSKA Obraz kulturowy Wysoczyzny Siedleckiej (ryc. 5a, b) w IX i X w. współtwo- rzą mikroregiony osadnicze rozlokowane w dolinach rzek, pozostające ze sobą w bezpośredniej relacji przestrzennej i chronologicznej. Charakteryzuje je ciągłość zasiedlenia bez śladów znaczących regresów oraz równomierny przyrost punktów osadniczych, powstających w bezpośrednim sąsiedztwie starszych – „macierzy- stych” (przykłady w: Gałecki, Skrzyńska 2021) lub założeń obronnych, których kilka zbudowano tam w ciągu 2 poł. IX i kilkanaście w X w. (Skrzyńska 2021, s. 75–92). Może to świadczyć o postępującej konsolidacji osadnictwa w ramach tworzących się terytoriów grodowych. Sytuacja ta jest jeszcze lepiej czytelna w XI–XII w., kiedy liczba osad wyraźnie wzrosła – szczególnie przy największych i najstarszych for- tyfikacjach (por. ryc. 5c, d; np. Krzesk – Kalaga 1989, s. 224–239; Niewiadoma – Miśkiewiczowa 1996a, s. 42–65). Powstały też wówczas nowe założenia obronne, stanowiące oparcie dla dalszego rozwoju osadnictwa (przykład – Skrzyńska 2019b). Podobne wnioski nasuwają się w odniesieniu do obszarów w dorzeczu górnej Narwi, nad Pulwą i nad strugami współtworzącymi obszar źródłowy Nurca (Kras- nodębski 2011–2012). Od końca VI w. istniał tam kompleks osadniczy w Haćkach (Kobyliński, Szymański 2015), a co najmniej od 3 ćw. IX i w X w. w najlepiej roz- winiętych zespołach osadniczych zbudowane zostały grody (ryc. 2.1. DANE ARCHEOLOGICZNE 98–109, 123–129, 133–134; Krasnodębski, Olczak 2019; Olczak, Krasnodębski 2019a; 2019b). Stabilne i równomierne zasiedlenie dorzecza Leśnej w XI–XII w. także potwier- dzają liczne, szkieletowe lub birytualne cmentarzyska kurhanowe (Korobuškina 1993; Karobuškina 1999)13. Często w ich pobliżu lub w bezpośrednim sąsiedztwie notowane są nekropole z grobami w obstawach kamiennych14. Rozpoznano też tam dwa cmentarzyska z płaskimi grobami szkieletowymi, które datowane są na XI–XII w. (SPB 1990, nr 400v i 914). Sytuacja ta w ogólnym zarysie przypomina obraz osadnictwa Wysoczyzny Drohiczyńskiej w tej samej fazie (Skrzyńska 2021, s. 71–73). Dla analizowanego tutaj problemu istotna jest jednak różnica w zakresie liczby zarejestrowanych cmentarzysk kurhanowych, których wysoka frekwencja CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 133 (wraz z dużą liczbą mogił na pojedynczych stanowiskach) w połączeniu ze zwartą siecią osad datowanych już od VI w. w dorzeczu Leśnej (Karobuškina 2000; Vergej 2005) wskazywać może na trwałość tamtejszego zasiedlenia, a niska (w połączeniu z niewielką liczbą kopców współtworzących pojedyncze nekropolie i nielicznymi osadami) w międzyrzeczu Nurca i Bugu – na jego nierównomierny rozwój co naj- mniej do końca X w. (por. Dzik 2019, s. 92–96). j (p , ) W tym kontekście osadnictwo Wysoczyzny Drohiczyńskiej w starszych fazach wczesnego średniowiecza wyglądało więc inaczej. Obszarem trwale zasiedlonym była wówczas tylko dolina Bugu, gdzie pojedyncze osady (znane głównie dzięki badaniom powierzchniowym) i cmentarzyska kurhanowe (Dzik 2015a, s. 48, 76–78, 82–86, 97–98, 110–111, 116, 122, 137–138, 146–147) odnotowano na północnym brzegu rzeki oraz wzdłuż jej niewielkich dopływów. Dla X w. obserwuje się tam kilka rozwiniętych mikroregionów, które, składając się z sąsiadujących ze sobą punktów osadniczych, nawiązują swym kształtem do sytuacji rozpoznanej po południowej stronie Bugu (por. ryc. 5b). Pozostała część tego obszaru pozostawała w zasadzie niezasiedlona15. Przy zestawieniu map tamtejszego osadnictwa z IX i z X w. (por. ryc. 5a, b) nie obserwuje się zasadniczych zmian, co świadczyć może o jego stag- nacji. Zważywszy na zanik osad z VII–IX w. w rejonie późniejszego grodu drohi- czyńskiego (Musianowicz 1966; 1978; Skrzyńska 2001), można tam dopatrywać się nawet śladów czasowych, lokalnie występujących regresów w dynamice zasied- lenia, polegających na zaniku starszych punktów osadniczych przy jednoczesnym braku śladów świadczących o lokalnym powstawaniu nowych siedlisk, takich, które mogłyby być interpretowane jako świadczące o ponowieniu procesów osadniczych w innym miejscu. W międzyrzeczu Bugu oraz dolnego i środkowego Nurca nie odnotowano jak dotąd żadnego grodziska, które mogłoby być datowane na okres przed XI w., chociaż być może wnioski te nie są ostateczne (por. 16 Grodzisko to nie było jeszcze badane wykopaliskowo, ale pod względem wielkości i układu umocnień jest ono podobne do obiektu obronnego we Włodkach, pow. sokołowski, datowanego na XI–XII w. (Górska i in. 1976, s. 157). Do weryfikacji wykopaliskowej pozostaje też chronologia fortyfikacji w Grodzisku, pow. siemiatycki, które obecnie uznawane jest za przypuszczalnie wczesno- średniowieczne (por. Skrzyńska 2021, ryc. 2.1: C). 15 Wyjątek stanowi tu jedno cmentarzysko usytuowane w północnej części omawianego obszaru, w górnym biegu rzeki Czarnej, które, będąc niepewnie datowanym na VIII–XI w. (Zoll-Adamikowa 1975, s. 70–72, nr 12; według Dzika 2015a, s. 70 – XI w.), mogłoby wskazywać na istnienie tam enklawy starszego osadnictwa. Jej rozkwit datowany jest jednak na młodsze fazy wczesnego średniowiecza (por. ryc. 5c, d). y y 17 Bardzo małe grodzisko, najpewniej strażnica przeprawy, hipotetycznie datowane na 1 poł. XIII w. (Dąbrowski 1955). 2.1. DANE ARCHEOLOGICZNE białostocki; 1 – osady (1a – kontynuacja zasiedlenia od VIII w.; 1b – założone w IX w.; 1c – kontynuacja zasiedlenia od IX w.; 1d – założone w X w.; 1e – kontynuacja zasiedlenia od X w.; 1f – założone w XI w.; 1g – kontynuacja zasiedlenia od XI w.; 1h – założone w XII w.); 2 – grody (2a – zbudo- wane w IX w.; 2b – kontynuacja użytkowania od IX w.; 2c – zbudowane w X w.; 2d – kontynuacja użytkowania od X w.; 2e – zbudowane w XI w.; 2f – kontynuacja użytkowania od XI w.; 2g – zbudowane w XII w.); 3 – cmentarzyska kurhanowe (3a – kontynuacja użytkowania od VIII w.; 3b – założone w IX w., 3c – kontynuacja użytkowania od IX w.; 3d – założone w X w.; 3e – kontynuacja użytkowania od X w.; 3f – założone w XI w.; 3g – kontynuacja użytko- wania od XI w.; 3h – założone w XII w.); 4 – birytualne i szkieletowe cmentarzyska płaskie i w typie tzw. mieszanym (4a – założone w XI w.; 4b – kontynuacja użytkowania od XI w.; 4c – założone w XII w.); 5 – birytualne i szkieletowe Ryc. 5. Wczesnośredniowieczne osadnictwo dorzecza środkowego Bugu Ryc. 5. Wczesnośredniowieczne osadnictwo dorzecza środkowego Bugu a – IX w.; b – X w.; c – XI w.; d – XII w.; WD – Wysoczyzna Drohiczyńska; WS – Wysoczyzna Siedlecka; RB – Rów- nina Bieska; DLP – dorzecze Leśnej i Pulwy; B. – Bobrówka, pow. hajnowski; Ch. – Chłopków, pow. łosicki; Cz. – Czołomyje, pow. siedlecki; D – Dzięcioły, pow. łosicki; Db. – Dobryń Duży, pow. bialski; Dh. – Dołhołęka, pow. bialski; G – Gródek, pow. sokołowski; H. – Haćki, pow. bielski; Hl. – Huszlew, pow. łosicki; K. – Krzesk, pow. siedlecki; Kl. – Klimy pow. łosicki; Kr. – Krasna Wieś, pow. bielski; M. – Mołoczki, pow. bielski; Mg. – Mogielnica, pow. siedlecki; N. – Niewiadoma, pow. sokołowski; P. – Podnieśno, pow. siedlecki; Pk. – Pokry, rej. brzeski; Pn. – Paszkowszczyzna, pow. bielski; W. – Włodki, pow. sokołowski; Wl. – Walim, pow. łosicki; Wz. – Wyłazy, pow. siedle- cki; Z. – Zbucz, pow. hajnowski; Zj. – Zajączki, pow. 2.1. DANE ARCHEOLOGICZNE białostocki; 1 – osady (1a – kontynuacja zasiedlenia od VIII w.; 1b – założone w IX w.; 1c – kontynuacja zasiedlenia od IX w.; 1d – założone w X w.; 1e – kontynuacja zasiedlenia od X w.; 1f – założone w XI w.; 1g – kontynuacja zasiedlenia od XI w.; 1h – założone w XII w.); 2 – grody (2a – zbudo- wane w IX w.; 2b – kontynuacja użytkowania od IX w.; 2c – zbudowane w X w.; 2d – kontynuacja użytkowania od X w.; 2e – zbudowane w XI w.; 2f – kontynuacja użytkowania od XI w.; 2g – zbudowane w XII w.); 3 – cmentarzyska kurhanowe (3a – kontynuacja użytkowania od VIII w.; 3b – założone w IX w., 3c – kontynuacja użytkowania od IX w.; 3d – założone w X w.; 3e – kontynuacja użytkowania od X w.; 3f – założone w XI w.; 3g – kontynuacja użytko- wania od XI w.; 3h – założone w XII w.); 4 – birytualne i szkieletowe cmentarzyska płaskie i w typie tzw. mieszanym (4a – założone w XI w.; 4b – kontynuacja użytkowania od XI w.; 4c – założone w XII w.); 5 – birytualne i szkieletowe Ryc. 5. Wczesnośredniowieczne osadnictwo dorzecza środkowego Bugu a – IX w.; b – X w.; c – XI w.; d – XII w.; WD – Wysoczyzna Drohiczyńska; WS – Wysoczyzna Siedlecka; RB – Rów- nina Bieska; DLP – dorzecze Leśnej i Pulwy; B. – Bobrówka, pow. hajnowski; Ch. – Chłopków, pow. łosicki; Cz. – Czołomyje, pow. siedlecki; D – Dzięcioły, pow. łosicki; Db. – Dobryń Duży, pow. bialski; Dh. – Dołhołęka, pow. bialski; G – Gródek, pow. sokołowski; H. – Haćki, pow. bielski; Hl. – Huszlew, pow. łosicki; K. – Krzesk, pow. siedlecki; Kl. – Klimy pow. łosicki; Kr. – Krasna Wieś, pow. bielski; M. – Mołoczki, pow. bielski; Mg. – Mogielnica, pow. siedlecki; N. – Niewiadoma, pow. sokołowski; P. – Podnieśno, pow. siedlecki; Pk. – Pokry, rej. brzeski; Pn. – Paszkowszczyzna, pow. bielski; W. – Włodki, pow. sokołowski; Wl. – Walim, pow. łosicki; Wz. – Wyłazy, pow. siedle- cki; Z. – Zbucz, pow. hajnowski; Zj. – Zajączki, pow. 2.1. DANE ARCHEOLOGICZNE Skrzyńska 2021, s. 30). Potem również nie ma ich zbyt wiele. Oprócz Drohiczyna i Brańska, „spinających” omawiany region od południa i północy, w stuleciu tym być może zbudowano tylko dwa założe- nia obronne – w Rogawce16 i w Bużyskach, pow. siemiatycki17. Najwyraźniej więc pod względem militarno-administracyjnym terytorium to zostało zorganizowane odmiennie, niż to można obserwować w sąsiednich regionach (Skrzyńska 2021, ryc. 2.1: A i B). 134 KATARZYNA SKRZYŃSKA 0 50 km a b c d a b d 0 50 km Ryc. 5. Wczesnośredniowieczne osadnictwo dorzecza środkowego Bugu a – IX w.; b – X w.; c – XI w.; d – XII w.; WD – Wysoczyzna Drohiczyńska; WS – Wysoczyzna Siedlecka; RB – Rów- nina Bieska; DLP – dorzecze Leśnej i Pulwy; B. – Bobrówka, pow. hajnowski; Ch. – Chłopków, pow. łosicki; Cz. – Czołomyje, pow. siedlecki; D – Dzięcioły, pow. łosicki; Db. – Dobryń Duży, pow. bialski; Dh. – Dołhołęka, pow. bialski; G – Gródek, pow. sokołowski; H. – Haćki, pow. bielski; Hl. – Huszlew, pow. łosicki; K. – Krzesk, pow. siedlecki; Kl. – Klimy pow. łosicki; Kr. – Krasna Wieś, pow. bielski; M. – Mołoczki, pow. bielski; Mg. – Mogielnica, pow. siedlecki; N. – Niewiadoma, pow. sokołowski; P. – Podnieśno, pow. siedlecki; Pk. – Pokry, rej. brzeski; Pn. – Paszkowszczyzna, pow. bielski; W. – Włodki, pow. sokołowski; Wl. – Walim, pow. łosicki; Wz. – Wyłazy, pow. siedle- cki; Z. – Zbucz, pow. hajnowski; Zj. – Zajączki, pow. 2.1. DANE ARCHEOLOGICZNE białostocki; 1 – osady (1a – kontynuacja zasiedlenia od VIII w.; 1b – założone w IX w.; 1c – kontynuacja zasiedlenia od IX w.; 1d – założone w X w.; 1e – kontynuacja zasiedlenia od X w.; 1f – założone w XI w.; 1g – kontynuacja zasiedlenia od XI w.; 1h – założone w XII w.); 2 – grody (2a – zbudo- wane w IX w.; 2b – kontynuacja użytkowania od IX w.; 2c – zbudowane w X w.; 2d – kontynuacja użytkowania od X w.; 2e – zbudowane w XI w.; 2f – kontynuacja użytkowania od XI w.; 2g – zbudowane w XII w.); 3 – cmentarzyska kurhanowe (3a – kontynuacja użytkowania od VIII w.; 3b – założone w IX w., 3c – kontynuacja użytkowania od IX w.; 3d – założone w X w.; 3e – kontynuacja użytkowania od X w.; 3f – założone w XI w.; 3g – kontynuacja użytko- wania od XI w.; 3h – założone w XII w.); 4 – birytualne i szkieletowe cmentarzyska płaskie i w typie tzw. mieszanym (4a – założone w XI w.; 4b – kontynuacja użytkowania od XI w.; 4c – założone w XII w.); 5 – birytualne i szkieletowe CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 135 Zaobserwowane różnice pozwalają wnioskować, że obszar ten pod względem rozwoju kulturowego u schyłku X w. był znacznie zapóźniony w porównaniu z tere- nami sąsiadującymi z nim od wschodu i południa, a jego zasiedlenie można określać jako pozostające wówczas na etapie wstępnym, tj. infiltracji osadniczej o nasileniu podobnym do tego, jakie na Wysoczyźnie Siedleckiej odnotowałam dla VI–VII stulecia (Skrzyńska-Jankowska 2014, s. 250, ryc. 25). Gwałtowny rozkwit osadni- ctwa nastąpił tam w młodszych fazach, tj. w XI i XII w., a więc dopiero w okresie wczesnopaństwowym (ryc. 5c, d). Jego początek sygnalizuje wzniesienie grodu w Drohiczynie, stanowiącego w następnych stuleciach regionalny ośrodek władzy (Skrzyńska-Jankowska 2004; Dzik, Jusupović 2019, s. 86–91). Było to prawdopodob- nie związane z równoczesnym organizowaniem terytorialnym Ziemi Drohickiej jako odrębnej jednostki administracyjnej, która do końca XII w. wraz z sąsiednim teryto- rium brzeskim należała do księstwa wołyńskiego (Jusupović 2007). Na północnym brzegu Bugu ośrodek ten współtworzył kompleks osad otwartych (Musianowicz 1969), co najmniej dwie lub trzy nekropole z płaskimi grobami szkieletowymi (Pawlata 1991; Andrzejewski, Sikora 2011) oraz kurhany – usytuowane co prawda poza głównym zespołem osadniczym Drohiczyna, ale w strefie jego bezpośredniego zaplecza; ciągle jeszcze reprezentujące ciałopalny ryt pogrzebowy (K. Bieńkowska 1980, s. 2.1. DANE ARCHEOLOGICZNE 155; taż 1983, s. 188; taż 1991a; 1991b; 1998; 2003; Chilmon 1981a; 1981b). W 2 poł. XI w. w regionie tym pojawiły się cmentarze z grobami w obstawach kamiennych. Analiza ich rozmieszczenia wskazuje, że zakładano je po obu stronach Bugu w strefach już wcześniej zasiedlonych, a na północ od Drohiczyna przede cmentarzyska z grobami w obstawach kamiennych (5a – założone w XI w.; 5b – kontynuacja użytkowania w XI w.; 5c – założone w XII w.). Na ryc. a i b kolorem brązowym oznaczono stanowiska datowane ogólnie na starsze fazy wczesnego średniowiecza, a na ryc. c i d niebadane cmentarzyska z grobami w obstawach kamiennych. Opracowała K. Skrzyńska Fig. 5. Early medieval settlement in the Middle Bug basin a – 9th c.; b – 10th c.; c – 11th c.; d – 12th c.; WD – Drohiczyn Upland; WS – Siedlce Upland; RB – Bielsk Plain; DLP – Leśna and Pulwa basin; B. – Bobrówka, Hajnówka district; Ch. – Chłopków, Łosice district; Cz. – Czoło- myje, Siedlce district; D – Dzięcioły, Łosice district; Db. – Dobryń Duży, Biała Podlaska district; Dh. – Dołhołęka, Biała Podlaska district; G – Gródek, Sokołów Podlaski district; H. – Haćki, Bielsk Podlaski district; Hl. – Huszlew, Łosice district; K. – Krzesk, Siedlce district; Kl. – Klimy, Łosice district; Kr. – Krasna Wieś, Bielsk Podlaski district; M. – Mołoczki, Bielsk Podlaski district; Mg. – Mogielnica, Siedlce district; N. – Niewiadoma, Sokołów Podlaski district; P. – Podnieśno, Siedlce district; Pk. – Pokry, Brześć district; Pn. – Paszkowszczyzna, Bielsk Podlaski district; W. – Włodki, Sokołów Podlaski district; Wl. – Walim, Łosice district; Wz. – Wyłazy, Siedlce district; Z. – Zbucz, Hajnówka district; Zj. 2.1. DANE ARCHEOLOGICZNE – Zajączki, Białystok district; 1 – settlements (1a – occupied from the 8th c.; 1b – founded in the 9th c.; 1c – occupied from the 9th c.; 1d – founded in the 10th c.; 1e – occupied from the 10th c.; 1f – founded in the 11th c.; 1g – occupied from the 11th c.; 1h – founded in the 12th c.); 2 – fortified settlements (2a – built in the 9th c.; 2b – in use from the 9th c.; 2c – built in the 10th c.; 2d – in use from the 10th c.; 2e – built in the 11th c.; 2f – in use from the 11th c.; 2g – built in the 12th c.); 3 – barrow cemeteries (3a – in use from the 8th c.; 3b – founded in the 9th c., 3c – in use from the 9th c.; 3d – founded in the 10th c.; 3e – in use from the 10th c.; 3f – founded in the 11th c.; 3g – in use from the 11th c.; 3h – founded in the 12th c.); 4 – biritual and inhumation flat cemeteries and of the mixed type (4a – founded in the 11th c.; 4b – in use from the 11th c.; 4c – founded in the 12th c.); 5 – biritual and inhumation cemetery of graves with stone casing (5a – founded in the 11th c.; 5b – in use from the 11th c.; 5c – founded in the 12th c.). Sites dated generally to the older phases of the early Middle Ages are marked in brown in figures a and b; unexplored cemeteries of graves with stone casing are marked in figures c and d. k k i Processing K. Skrzyńska i Processing K. Skrzyńska 136 KATARZYNA SKRZYŃSKA 0 – a – b.1 – b.2 – b.3 – b.4 – c.1 – c.2 – c.3 – d.1 – d.2 – g – e – f – h 50 km Ryc. 6. Lokalizacja cmentarzysk z XI–XIII w. y yp Wg Buko i in. 2020, ryc. 1 (a–d); Skrzyńskiej-Jankowskiej 2014 (e, f); Wójcik 2013, ryc. 50–51 i 86 (g); Buko i in. 2020, tabela 1 (stanowiska oznaczone numerami 1–21); opracowała K. Skrzyńska 2.1. DANE ARCHEOLOGICZNE w dorzeczu środkowego Bugu i w południowej części zlewiska górnej Narwi ł b ł l b k h (b h k ł l b h k Główne typy grobów: a – ciałopalne, jamowe; b – kurhany (b.1 – z pochówkami ciałopalnymi; b.2 – z pochówkami ciałopalnymi i szkieletowymi w osobnych mogiłach; b.3 – z pochówkami szkieletowymi; b.4 – ciałopalne z wtór- nymi pochówkami szkieletowymi); c – cmentarzyska z grobami w obstawach kamiennych (c.1 – ciałopalne [?]; c.2 – birytualne; c.3 – szkieletowe); d – cmentarzyska płaskie (d.1 – birytualne; d.2 – szkieletowe); e – zasięg wystę- powania kurhanów z konstrukcjami kamiennymi; f – zasięg występowania kurhanów bez konstrukcji kamiennych; g – zasięg występowania ceramiki „typu drohiczyńskiego”; h – współczesna granica polsko-białoruska; 1–21 – stanowiska wytypowane do analiz aDNA. y yp Wg Buko i in. 2020, ryc. 1 (a–d); Skrzyńskiej-Jankowskiej 2014 (e, f); Wójcik 2013, ryc. 50–51 i 86 (g); Buko i in. 2020, tabela 1 (stanowiska oznaczone numerami 1–21); opracowała K. Skrzyńska CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 137 wszystkim tam, gdzie nie odnotowano starszego osadnictwa (ryc. 6; por. ryc. 5c, d). Znamienne jest, że takich nekropoli nie stwierdzono w sąsiedztwie grodu i osad w Drohiczynie. Obydwie powyższe obserwacje są zbliżone do ustaleń, jakie poczynił Tomasz Kordala (2005a, s. 236), analizując rozmieszczenie tych cmentarzysk w rejo- nie Płocka. Jednocześnie na obszarze tym stwierdzono ślady wielu osad otwartych, zarejestrowane głównie podczas badań AZP. j g p Ten dynamiczny rozwój, którego rezultatem było trwałe zagospodarowanie mię- dzyrzecza Bugu i Nurca, mógł wiązać się z kolonizacją lub inną formą ukierunko- wanego napływu ludności. Groby w obstawach kamiennych w krótkim czasie stały się tam bowiem najczęściej spotykaną formą grzebania zmarłych. Zważywszy na wyjątkowo synchroniczne ramy chronologiczne, w których sytuowane są inicjalne fazy wszystkich tych nekropoli, można zakładać, że zagospodarowanie obszaru na północ od Drohiczyna miało charakter krótkotrwałego epizodu lub zorganizo- wanej akcji, biorąc pod uwagę liczebność stanowisk – o wyjątkowo intensywnym przebiegu. Oprócz zasiedlenia obszarów do tamtego czasu stanowiących być może pustki osadnicze, wzrosła znacząco liczba osad także po południowej stronie Bugu (Wysoczyzna Siedlecka), gdzie nowa ludność mogła „wtopić się” w strefy osadniczo ukształtowane w starszych fazach wczesnego średniowiecza. Analogiczna sytuacja zaistniała najpewniej na Równinie Bielskiej, a prawdopodobnie również na Wyso- czyźnie Kolneńskiej (Dzik 2021). i y j Weryfikacja tych tez leży już niestety poza sferą możliwości poznawczych samej archeologii, a największy potencjał w tym zakresie mają badania paleogenetyczne. Fig. 6. Location of cemeteries from the 11th–13th centuries in the basin of the middle Bug and in the southern part of the upper Narew catchment area Main types of graves: a – cremation, pit burials; b – barrow (b.1 – with cremation burials; b.2 – with cremation and inhumation burials in separate graves; b.3 – with inhumation burials; b.4 – cremation burials with secondary inhumation burials); c – cemeteries of graves with stone casing (c.1 – cremation [?]; c.2 – biritual; c.3 – inhuma- tion); d – flat cemetery (d.1 – biritual; d.2 – inhumation); e – range of barrows with stone structures; f – range of barrows without stone structures; g – range of “Drohiczyn-type” pottery; h – the modern Polish-Belarus border; 1–21 – sites selected for ancient DNA analyses. After Buko et al. 2020, Fig. 1 (a–d); Skrzyńska-Jankowska 2014 (e, f); Wójcik 2013, Figs 50–51 and 86 (g); Buko et al. 2020, table 1 (sites marked with the numbers 1–21); processing K. Skrzyńska 2.1. DANE ARCHEOLOGICZNE W przeciwieństwie do analiz zawartości izotopu strontu, gdzie miarodajne wyniki w tej kwestii mogłoby przynieść zbadanie wyłącznie najstarszej generacji pochowa- nych, ślady odmienności w aDNA mają charakter trwały i, niezależnie od doraźnych warunków funkcjonowania społeczności, dziedziczone są z pokolenia na pokolenie. 2.2. WYNIKI BADAŃ PALEOGENETYCZNYCH Dysponujemy już wstępnymi danymi dotyczącymi struktury genetycznej wczes- nośredniowiecznych populacji mazowiecko-ruskiego pogranicza (Molak i in. 2019; Buko i in. 2020). Badania przeprowadził zespół prof. dr. hab. Wiesława Bogdano- 138 KATARZYNA SKRZYŃSKA wicza z Muzeum i Instytutu Zoologii PAN18. Zakwalifikowano do nich materiał antropologiczny z 21 stanowisk19, w tym trzech z płaskimi grobami szkieletowymi, sześciu kurhanów, 11 cmentarzysk z grobami w obstawach kamiennych i dwóch grobów o nietypowej konstrukcji. Ogółem pobrano 103 próbki (Skrzyńska, Kras- nodębski 2019, tabela 14.7; Buko i in. 2020, tabela 1), a pozytywne wyniki uzyskano w przypadku 62, z czego 60 – z grobów szkieletowych (Molak i in. 2019, tabela 14.4). Jeśli chodzi o Wysoczyzny Drohiczyńską i Siedlecką, wszystkie one pochodziły z cmentarzysk z grobami w obstawach kamiennych, na Wysoczyźnie Bielskiej – z grobów z obstawami i z kurhanów, a w Dolinie Górnej Narwi – z płaskich cmen- tarzysk szkieletowych (por. ryc. 6). Wyniki analiz porównywano w kilku aspektach, m.in. w odniesieniu do lokalizacji nekropoli w ramach wydzielonych regionów fizjograficznych oraz według kryterium kulturowego, grupując je pod względem różnic w konstrukcji grobów (ryc. 7a) i jednocześnie odnosząc do współczesnych populacji europejskich (ryc. 7b). Kluczowym wnioskiem w przypadku populacji zasiedlającej Wysoczyznę Drohi- czyńską w młodszych fazach wczesnego średniowiecza było następujące stwierdze- nie (Molak i in. 2019, s. 238–239; Buko i in. 2020, s. 595): „analiza korespondencji (ryc. 14.3C) wskazała na relatywnie bliskie podobieństwo wczesnośredniowiecznych populacji pochodzących z terenów na południe od Bugu […] i Doliny Górnej Narwi w odniesieniu do populacji współczesnych. Większe różnice stwierdzono w przy- padku wczesnośredniowiecznej populacji z Równiny Bielskiej, a zupełnie odmienną okazała się wczesnośredniowieczna populacja na Wysoczyźnie Drohiczyńskiej”. ę p p j y y y j Cmentarzyska z grobami w obstawach kamiennych z Wysoczyzny Drohiczyń- skiej charakteryzuje również największa dywersyfikacja haplogrup zidentyfikowa- nych w mitochondrialnym DNA tamtejszych pochówków (por. część poświęcona wynikom badań paleogenetycznych w: Buko i in. 2020, s. 593–598). Różnica ta jest dobrze widoczna zarówno w odniesieniu do kryterium geograficznego (ryc. 8a – Wysoczyzna Drohiczyńska vs. Wysoczyzna Siedlecka, Równina Bielska i Dolina Górnej Narwi), jak też typologicznego (ryc. 8b – pochówki z grobów z obstawami vs. kurhanowe i z grobów płaskich). W opcji „geograficznej”, w przeciwieństwie do pozostałych regionów, makrohaplogrupa HV w populacji Wyżyny Drohiczyńskiej nie dominowała. Jej udział wynosił tam około 6,7%, podczas gdy w pozostałych trzech regionach przekraczał 50%20. y g g j 20 Dolina Górnej Narwi (cmentarzyska płaskie) – około 52,2%; Równina Bielska (razem: cmen- tarzyska w obstawach i kurhany) – około 71,4%; Wysoczyzna Siedlecka (cmentarzyska z grobami w obstawach) – około 64,3%. 18 Dziękuję dr Martynie Molak, która wykonała wymienione analizy, za specjalistyczną konsul- tację tego podrozdziału. 19 W tym ze stanowiska 1 w Szczytach-Dzięciołowie, pow. bielski, gdzie badaniom poddano próby pochodzących z kurhanu i płaskiego grobu w obstawie kamiennej. 18 Dziękuję dr Martynie Molak, która wykonała wymienione analizy, za specjalistyczną konsul- tację tego podrozdziału. 19 W tym ze stanowiska 1 w Szczytach-Dzięciołowie, pow. bielski, gdzie badaniom poddano próby kości pochodzących z kurhanu i płaskiego grobu w obstawie kamiennej. 20 Dolina Górnej Narwi (cmentarzyska płaskie) – około 52,2%; Równina Bielska (razem: cmen- tarzyska w obstawach i kurhany) – około 71,4%; Wysoczyzna Siedlecka (cmentarzyska z grobami w obstawach) – około 64,3%. 18 Dziękuję dr Martynie Molak, która wykonała wymienione analizy, za specjalistyczną konsul- tację tego podrozdziału. 19 W tym ze stanowiska 1 w Szczytach-Dzięciołowie, pow. bielski, gdzie badaniom poddano próby kości pochodzących z kurhanu i płaskiego grobu w obstawie kamiennej. 20 Dolina Górnej Narwi (cmentarzyska płaskie) – około 52,2%; Równina Bielska (razem: cmen- k b h k h ) k ł dl k ( k b 21 Osobny wątek badawczy zarysowuje się w tym przypadku na jeszcze innym polu, gdyż dane z Wysoczyzny Siedleckiej to wyniki analiz prób pobranych wyłącznie z grobów w obstawach kamien- nych (!), a z Równiny Bielskiej – pochówków tego samego typu oraz kurhanowych. Niewykluczone więc, że opisywany wzorzec funeralny, polegający na konstruowaniu obstaw, był dodatkowo wykorzy- stywany jako nowa idea, niezależnie od populacji jego nosicieli. Dobór prób determinowany celami projektu, innymi od prezentowanego tutaj zagadnienia, nie był chyba odpowiedni do jego pogłębionej analizy. Wydaje się bowiem, że do takich badań powinny być wytypowane pochówki najstarsze – teo- retycznie reprezentujące generację założycielską. Należy brać pod uwagę m.in. możliwość, że młodsze groby to pochówki ludności miejscowej, która, współtworząc lokalną społeczność, kulturowo zinte- growała się z przybyszami. 2.2. WYNIKI BADAŃ PALEOGENETYCZNYCH Jest to jeden z wątków przyszłych badań, gdyż wskazana HV jest charakterystyczna dla wschodniej i południowej Europy, zachod- niej Azji oraz dla północnej Afryki (Witas i in. 2020, s. 131–132 – tam literatura). Wynik ten rzuca interesujące światło na kwestię związku populacji z wysokim udziałem HV z autochtonicznym substratem osadniczym, którego pochodzenie CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 139 Ryc. 7. Analiza korespondencji na podstawie frekwencji haplogrup wyższego rzędu między zespołami badanych pochówków wczesnośredniowiecznych („populacja testowa”) a wybranymi współczesnymi populacjami europejskimi a – podział wg typu pochówku; b – podział wg regionów fizjograficznych. Wg Molak i in. 2019, s. 237, ryc. 14.3B i 14.3C; opracowała M. Molak Fig. 7. Correspondence analysis based on the frequency of higher-order haplogroups between the studied assemblages of early medieval burials (“test population”) and selected modern European populations a – division by burial type; b – division by physiographic regions. After Molak et al. 2019, p. 237, Fig. 14.3B and 14.3C; processing M. Molak ędu między zespołami badanych pochówków wczesnośredniowiecznych Ryc. 7. Analiza korespondencji na podstawie frekwencji haplogrup wyższego („populacja testowa”) a wybranymi wsp a – podział wg typu pochówku; b – po Fig. 7. Correspondence analysis based on the frequency of higher-order hap population”) and selected mo a – division by burial type; b – div Fig. 7. Correspondence analysis based on the frequency of higher-orde population”) and selected 140 KATARZYNA SKRZYŃSKA 1.00 0.75 0.50 0.25 0.00 Dolina Górnej Narwi (23) Równina Bielska (7) Wyszoczyzna Drohiczyńska (15) Wyszoczyzna Siedlecka (14) Cmentarzyska kurhanowe (6) Cmentarzyska z grobami w obstawach kamiennych (26) Cmentarzyska płaskie (26) b hg inne I U T J HV a Ryc. 8. Zróżnicowanie frekwencji haplogrup w badanych wczesnośredniowiecznych populacjach dorzecza środkowego Bugu i górnej Narwi a – regiony fizjograficzne; b – typologia konstrukcji grobów; hg – zidentyfikowane haplogrupy. Wg Buko i in. 2020, ryc. 9; opracowała M. Molak Fig. 8. Variation in the frequency of haplogroups in the studied early medieval populations of the 1.00 0.75 0.50 0.25 0.00 Dolina Górnej Narwi (23) Równina Bielska (7) Wyszoczyzna Drohiczyńska (15) Wyszoczyzna Siedlecka (14) Cmentarzyska kurhanowe (6) Cmentarzyska z grobami w obstawach kamiennych (26) Cmentarzyska płaskie (26) b hg inne I U T J HV a Cmentarzyska kurhanowe (6) Cmentarzyska z grobami w obstawach kamiennych (26) Cmentarzyska płaskie (26) b hg inne I U T J HV a kulturowo identyfikowane jest z Polesiem, Wołyniem i wschodnią Małopolską, tj. 2.2. WYNIKI BADAŃ PALEOGENETYCZNYCH obszarami uznawanymi za macierzyste dla Słowian (Baran 1998; Parczewski 1988; 2004; Vergej 1997; Vârgej 2004; Terpilovskij 2004). Współcześnie najwyż- szy udział tej haplogrupy notowany jest na Ukrainie, Białorusi i Słowacji, gdzie jest on znacząco wyższy niż w Polsce (Juras 2012, tabela 27). Spostrzeżenie to ma szczególne znaczenie dla rekonstrukcji osadnictwa Wysoczyzny Siedleckiej oraz Równiny Bielskiej21 (ta druga rozumiana jest jako chronologicznie i przestrzennie związana z makroregionem osadniczym nad Leśną i Pulwą – por. prezentowaną kulturowo identyfikowane jest z Polesiem, Wołyniem i wschodnią Małopolską, tj. obszarami uznawanymi za macierzyste dla Słowian (Baran 1998; Parczewski 1988; 2004; Vergej 1997; Vârgej 2004; Terpilovskij 2004). Współcześnie najwyż- szy udział tej haplogrupy notowany jest na Ukrainie, Białorusi i Słowacji, gdzie jest on znacząco wyższy niż w Polsce (Juras 2012, tabela 27). Spostrzeżenie to ma szczególne znaczenie dla rekonstrukcji osadnictwa Wysoczyzny Siedleckiej oraz Równiny Bielskiej21 (ta druga rozumiana jest jako chronologicznie i przestrzennie związana z makroregionem osadniczym nad Leśną i Pulwą – por. prezentowaną 21 Osobny wątek badawczy zarysowuje się w tym przypadku na jeszcze innym polu, gdyż dane z Wysoczyzny Siedleckiej to wyniki analiz prób pobranych wyłącznie z grobów w obstawach kamien- nych (!), a z Równiny Bielskiej – pochówków tego samego typu oraz kurhanowych. Niewykluczone więc, że opisywany wzorzec funeralny, polegający na konstruowaniu obstaw, był dodatkowo wykorzy- stywany jako nowa idea, niezależnie od populacji jego nosicieli. Dobór prób determinowany celami projektu, innymi od prezentowanego tutaj zagadnienia, nie był chyba odpowiedni do jego pogłębionej analizy. Wydaje się bowiem, że do takich badań powinny być wytypowane pochówki najstarsze – teo- retycznie reprezentujące generację założycielską. Należy brać pod uwagę m.in. możliwość, że młodsze groby to pochówki ludności miejscowej, która, współtworząc lokalną społeczność, kulturowo zinte- growała się z przybyszami. CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 141 wyżej rekonstrukcję tła osadniczego oraz ryc. 3 i 5). Widać także pewne podobień- stwo tych wyników do oznaczeń uzyskanych dla płaskich cmentarzysk w Daniłowie Małym i Surażu, reprezentujących region doliny Górnej Narwi, chociaż obydwie te nekropole także różniły się między sobą na poziomie bezpośredniego wzajem- nego porównania (Olczak i in. 2019, s. 89–90). Odkrycia na grodzisku w Haćkach (Kobyliński, Szymański 2015, s. 114–136) świadczą zaś o tym, że rzeka ta stanowić mogła oś rozwoju osadnictwa zarówno w pradziejach, jak i na początku wczesnego średniowiecza. 2.2. WYNIKI BADAŃ PALEOGENETYCZNYCH [poten- cjalny napływ nowych osadników kultywujących tradycje pogrzebowe odmienne 142 KATARZYNA SKRZYŃSKA od rodzimych]) uznać należy za autochtoniczną dla regionu, znajduje odzwier- ciedlenie także w wynikach analizy korespondencji jej struktury genetycznej ze współczesnymi populacjami europejskimi (por. ryc. 7a). Na ich tle wyraźnie rysuje się odrębność tej grupy, co według specjalistów, w odróżnieniu od pozostałych analizowanych populacji, może wskazywać na ograniczenie lub brak ciągłości gene- tycznej pomiędzy wczesnym średniowieczem a fazą nowożytną (Molak i in. 2019, s. 236–239). Jednoznacznie nasuwa się tutaj kontrast z sąsiednim Polesiem, nie tylko jako homogennym kulturowo obszarem związanym z etnogenezą Słowian (Kuharenko 1968), ale także takim, na którym w świetle analiz antropologicznych i etnograficznych do początku XX w. zachowały się relikty średniowiecznej kultury (Rolecki 2006; Obrębski 2007, s. 140–191, 199–216, 223–234, 253–258, 264–270). Przetrwały one najpewniej w związku ze znaczącym odizolowaniem tej ludności, uwarunkowanym zarówno przez specyfikę naturalnego środowiska, jak też czyn- niki kulturowe związane z usytuowaniem tego regionu począwszy od wczesnego średniowiecza w strefie marginalnej rozwoju cywilizacyjnego. i g j j y yj g Większe zróżnicowanie wewnątrzpopulacyjne zdaje się charakteryzować spo- łeczności użytkujące płaskie cmentarze szkieletowe – pośrednio (Daniłowo Małe) lub bezpośrednio (Suraż) związane z najważniejszymi grodami. Udział haplo- grupy HV w odniesieniu do innych wyodrębnionych w tych populacjach wynosi 50% (ryc. 8b), a I tylko około 3,9%. Wyraźnie też widać, że obok innych oznaczo- nych grup, takich jak wspomniana wyżej haplogrupa I oraz U, T i J, duży udział mają tam haplogrupy tzw. pozostałe, stwierdzone w jednostkowych przypadkach. Ich udział oscyluje na poziomie około 26,9%. Znaczne zróżnicowanie aDNA ludno- ści potwierdza w przypadku tych cmentarzysk obserwacje dotyczące kulturowych i gospodarczych uwarunkowań powstawania i funkcjonowania zespołów osadni- czych o wyższej randze, np. tezę o usytuowaniu ich w węzłowych punktach szlaków komunikacyjnych. W przypadku wskazanych tutaj Daniłowa i Suraża jest to prze- cięcie wodnego szlaku narwiańskiego z lądowym traktem wiodącym z Drohiczyna na Jaćwież i w kierunku Grodna (Skrzyńska 2019a, s. 288; Olczak, Krasnodębski 2020, s. 470–473). Można się domyślać, że zróżnicowanie genetyczne tamtejszej społeczności jest efektem tej sytuacji, spowodowanej najprawdopodobniej przez stałą lub czasową obecność osób obcego pochodzenia. Zostało to bezpośrednio potwierdzone dla Suraża, gdzie pochowano osobnika o haplogrupie M, świadczącej o jego korzeniach rodowych we wschodniej, środkowej lub południowo-wschodniej Azji (Molak i in. 2019, s. 239; Buko i in. 2020, s. 596). j Trudno wyjaśnić, dlaczego populacją najbardziej zróżnicowaną pod względem genetycznym jest ta pochowana na cmentarzyskach z grobami w obstawach kamien- nych. 2.2. WYNIKI BADAŃ PALEOGENETYCZNYCH Wysoki udział HV charakteryzujący tamtejsze testowane populacje wspiera ustalenia poczynione na podstawie źródeł archeologicznych, analizowanych w celu rekonstrukcji procesu zasiedlenia wschodniej części międzyrzecza Bugu i górnej Narwi, a także Bugu oraz Liwca i Krzny (Skrzyńska-Jankowska 2014, s. 237– 299, 301–542, 608–616; Skrzyńska 2021, s. 57–73). Okazuje się bowiem, że regiony te, jako „stare” strefy osadnictwa słowiańskiego, krystalizujące się co najmniej od VI w., charakteryzuje być może także wysoki poziom homogenności biologicznej. Tymczasem bardzo niska frekwencja tej haplogrupy na cmentarzyskach z grobami w obstawach kamiennych z Wysoczyzny Drohiczyńskiej może świadczyć o niewiel- kim udziale rdzennej ludności w użytkujących je społecznościach. Nekropole z obstawami kamiennymi wyraźnie wyróżniają się także w kontekście analizy respektującej drugą z rozpatrywanych zmiennych, tj. podział typologiczny cmentarzysk. Wykazała ona, że zmarłych tam pochowanych charakteryzuje najwięk- sze zróżnicowanie frekwencji poszczególnych haplogrup mitochondrialnego DNA, czytelnie innego od stwierdzonego w przypadku ludności z cmentarzysk płaskich i kurhanów (ryc. 8). Różnica ta jest szczególnie dobrze widoczna przy zestawieniu z pochówkami kurhanowymi, w przypadku których wykryto tylko dominującą haplogrupę HV (około 83,3%) wraz z uzupełniającą ją staroeuropejską I (około 16,7%). Obecnie są one najlepiej reprezentowane na Słowacji i w Czechach oraz na Białorusi i w europejskiej części Rosji (Juras 2012, tabela 27). Ten wyraźny brak zróż- nicowania genetycznego może wskazywać, że społeczności kultywujące kurhanową formę obrzędowości pogrzebowej długo funkcjonowały w biologicznej izolacji. ę ę p g j g j y g j j W odniesieniu do zjawisk kulturowych przejawia się ona głównie w archai- zujących cechach obrzędowości pogrzebowej (Kalaga 2006, s. 186–188; taż 2019, s. 264–267) oraz periodyzacji rozwoju regionalnej wytwórczości garncarskiej, zbież- nej z rekonstruowaną dla międzyrzecza Dniepru i górnego Bugu (Miśkiewicz 1981; Parczewski 1988; Ìoŭ 1992; Iov 1992; Kobylińska 2003; Milân 2004; Vârgej 2004; Vergej 2005). W obydwu tych aspektach jest on wskazywany jako obszar macie- rzysty różnych wzorców kulturowych oraz wyjściowy dla rozprzestrzeniania się osadnictwa, sukcesywnie postępującego z południa i południowego wschodu już w starszych fazach wczesnego średniowiecza. Tezę tę wspiera także analiza form obiektów mieszkalnych i konstrukcji umocnień najstarszych grodów zbadanych na Podlasiu (Skrzyńska-Jankowska 2014, s. 259–299, 340–345, 473–484, 501–519; Skrzyńska 2021, s. 167–184). Ten „hermetyczny” pod względem kulturowym obraz społeczności, którą po około 600 latach obecności na terenie dorzecza środkowego Bugu i górnej Narwi (od VI–VII w. [pierwsza fala osadnicza] do 2 poł. XI w. 22 Znajduje to potwierdzenie w wynikach analiz zawartości izotopów strontu w szkieletach z cmentarzyska w Narojkach, pow. siemiatycki (D. Błaszczyk – prezentacja wstępnych wyników badań na konferencji „Przeszłość ma przyszłość” w Instytucie Archeologii UW w marcu 2022 r.). Wskazują one, że większość zbadanych osób urodziła się i dorastała w najbliższej okolicy. Dotyczy to młodszych pokoleń, ponieważ centralna część cmentarzyska, gdzie przypuszczalnie znajdować się mogą groby najstarsze, nie została rozpoznana wykopaliskowo. Analizy opracowywane są w ramach realizacji projektu PMMA. 2.2. WYNIKI BADAŃ PALEOGENETYCZNYCH Ich usytuowanie na głębokim zapleczu grodu w Drohiczynie oraz niewąt- pliwy związek z osadnictwem otwartym o charakterze wiejskim, możliwość bieżącej, intensywnej wymiany genów czyni mało prawdopodobną. Udział HV i I, uznanych na podstawie wyników badań szkieletów pochodzących z kurhanów za określające populacje autochtoniczne, wynosi tam łącznie około 38,5% (HV – około 30,8%; I – około 7,7%). Pozostałe około 61,5% tworzą haplogrupy U, T, J i „inne” – pod względem częstości występowania mające względem siebie mniej więcej równo- CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 143 rzędną pozycję. Obserwacje te wskazują więc, że społeczność użytkująca te cmen- tarze była pod względem biologicznym dużo bardziej zróżnicowana od tej budują- cej kurhany, chociaż funkcjonowała w tym samym kontekście, współtworząc z nią bazową („wiejską”) tkankę osadniczą regionu22. Pod względem struktury genetycznej populacje użytkowników cmentarzy z grobami w obstawach zdają się być zbliżone do tych użytkujących cmentarze płaskie. Różniły się jednak od nich znacznie w zakresie wielkości udziału poszczególnych składników. Wstępnie można więc stwierdzić, że analizy aDNA wskazują na brak wyraźnych powiązań genetycznych pomiędzy społecznościami „kurhanowymi” i „obstawowymi”. Zakładając, że pierwsi to auto- chtoni, można postawić hipotezę, że drudzy w całości lub w przeważającej części to osadnicy spoza regionu albo ich potomkowie. W pewnym stopniu, pod względem zróżnicowania aDNA byli oni podobni do populacji pochowanych przy głównych grodach, które zapewne w dużej części także współtworzyli obcy przybysze. g p j ę p y y p y y Autorzy badań (por. omówienie wyników badań aDNA w: Buko i in. 2020) zastrzegają jednak, że spostrzeżenia te mają charakter wstępny, gdyż w odniesieniu do liczby znanych regionalnych cmentarzysk seria 62 szkieletów ze zidentyfiko- wanym DNA jest niewspółmiernie mała. Wiarygodność otrzymanych wyników będzie więc weryfikowana w toku dalszych prac, prowadzonych w ramach nowego projektu PMMA (por. przypis 2). p j 23 Cmentarzysko płaskie w Surażu datowane jest od 2 poł. XII do końca XIII w. z dopuszczeniem możliwości sytuowania początków nekropoli w 1 poł. XII w., a schyłku jej użytkowania w XIV – 1 poł. 2.3. ANALIZY ANTROPOLOGICZNE Innych argumentów wspierających tezę o zasiedleniu Wysoczyzny Drohiczyń- skiej przez ludność pochodzącą spoza regionu dostarczyły wyniki badań antropo- logicznych. Ich przedmiotem była m.in. analiza porównawcza szkieletów z dwóch cmentarzysk. Pierwsze z nich to płaska nekropola w Surażu nad Narwią (por. K. Bieńkowska 2005; Olczak i in. 2019, s. 83–90), pod względem cech obrządku pogrzebowego mająca prawdopodobnie charakter przygrodowy (Skrzyńska 2019a, s. 289–292). Druga to datowane na ten sam okres „wiejskie” cmentarzysko z gro- bami w obstawach kamiennych w Czarnej Wielkiej, pow. siemiatycki (Karwowska red. 2013; 2014), usytuowane na północnym skraju Wysoczyzny Drohiczyńskiej, w odległości 43 km na południowy zachód od Suraża. Dla przeprowadzonych analiz istotne znaczenie ma fakt, że obydwie te nekropole przynajmniej od 2 poł. XII w. funkcjonowały synchronicznie23. 23 Cmentarzysko płaskie w Surażu datowane jest od 2 poł. XII do końca XIII w. z dopuszczeniem możliwości sytuowania początków nekropoli w 1 poł. XII w., a schyłku jej użytkowania w XIV – 1 poł. 144 KATARZYNA SKRZYŃSKA Judyta Gładykowska-Rzeczycka, analizując biostrukturę ludności pochowanej w Surażu, wykonała wszechstronne porównania tej populacji z innymi, mniej lub bardziej jej współczesnymi z terenu Polski. W kontekście regionalnym możliwość dokonania takiej analizy stwarzało jedynie wskazane powyżej cmentarzysko w Czar- nej Wielkiej, z którego pochodził wielkościowo odpowiedni zbiór danych, opra- cowanych zresztą przez tę samą antropolog. Jednym z elementów tych badań było zestawienie dotyczące zróżnicowania wielkości czaszek. Efektem są następujące wnioski (wyróżnienie Autorki): „czaszki kobiet z Suraża są najbardziej podobne do wczesnośredniowiecznych czaszek z Kruszwicy z X–XII w., dalej – razem czaszki te przypominają czaszki żeńskie pochodzące z Ostrowa Lednickiego z XII w. Intere- sujące jest to, że czaszki z Suraża d o ś ć z n a c z n i e r ó ż n i ą s i ę od czaszek kobiet z nieodległej Surażowi Czarnej Wielkiej. […] czaszki mężczyzn z Suraża są najbardziej podobne do czaszek mężczyzn z Ostrowa Lednickiego z XII w. Z kolei dane z tych dwóch stanowisk są najbardziej zbliżone do populacji wczesnośrednio- wiecznej z Kruszwicy z X–XII w. […]. Dowodzi tego np. m n i e j s z e p o d o - b i e ń s t w o serii z Suraża, podobnie jak w przypadku kobiet, do najbliższej jej w przestrzeni, a reprezentowanej przez czaszki pochodzące z cmentarzyska w Czar- nej Wielkiej” (Gładykowska-Rzeczycka 2019, s. 180–181). 2.3. ANALIZY ANTROPOLOGICZNE Antropolog upatrywała hipotetycznej przyczyny tych różnic w uwarunkowaniach bytowych obydwu popu- lacji, postulując „konieczność prześledzenia warunków środowiskowych, w jakich przedstawiciele danej ludności żyli, zwłaszcza w wypadku b r a k u p o w i ą z a ń np. między serią szkieletów z Suraża i z Czarnej Wielkiej” (Gładykowska-Rzeczycka 2019, s. 192). Wnioski te były jednak sformułowane przed pojawieniem się możli- wości wykonania analiz aDNA. XV w. (Olczak i in. 2019, s. 87–88). Nekropola z grobami w obstawach kamiennych w Czarnej Wielkiej istniała pomiędzy ostatnią ćwiercią lub tercją XI w. a XVI w. włącznie (Dzik 2014c, s. 226). 2.4. JESZCZE RAZ ARCHEOLOGIA ORAZ DANE HISTORYCZNE I ONOMASTYCZNE Na kulturową odrębność populacji zamieszkującej okolice Drohiczyna w śred- niowieczu wskazuje też obecność naczyń ceramicznych, które ze względu na specy- fikę formy oraz stylistykę opracowania powierzchni stanowią łatwo identyfikowalny typ określany mianem „ceramiki drohiczyńskiej” (Miśkiewiczowa 1996b). Agata Wójcik (2013), rozpatrując kwestie zasięgu występowania tych wyrobów w XI– XIII w., potwierdziła starsze ustalenia (Musianowicz 1962). W świetle dotychcza- sowych badań są one bowiem powszechnie spotykane wyłącznie na Wysoczyźnie Drohiczyńskiej oraz w północnej części Wysoczyzny Siedleckiej, a poza samym Drohiczynem i osadami jego zaplecza znajdowane są głównie na cmentarzyskach z grobami w obstawach kamiennych (ryc. 6g). Naczynia tego typu nie są natomiast charakterystyczne dla Równiny Bielskiej, Doliny Górnej Narwi oraz dla południowej części Wysoczyzny Siedleckiej i Polesia Podlaskiego, gdzie ceramika z młodszych faz CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 145 wczesnego średniowiecza pod względem form i stylistyki nawiązuje bezpośrednio do tradycyjnego kanonu regionalnego (Kalaga 2006, s. 196–236; Olczak, Wójcik 2019, s. 167, ryc. 9.12). Poza linią Nurca na północy i Liwca na południu, liczebność i cha- rakter znalezisk tych naczyń nie różni się od stwierdzonych w całym makroregionie dorzecza środkowej Wisły. Do tej pory zidentyfikowano tam wyłącznie pojedyncze ich egzemplarze, w tym w znaczącej liczbie stanowiące naśladownictwa drohiczyń- skich oryginałów (Wójcik 2014, s. 84, ryc. 5). Podobny jest obraz ich frekwencji na wschodzie: w dorzeczach Leśnej i Muchawca oraz środkowego Niemna (Lysenko 1985, s. 238; Malevskaâ-Malevič 2005, s. 145–146). Badania składu chemicznego mas garncarskich reprezentatywnych wyrobów wykazały, że naczynia te wytwarzano w Drohiczynie (Wójcik 2008). To, czy także tam nastąpiła ewolucja ich form i styli- styki w końcu X w. (Miśkiewicz 1981, s. 39, 41; Malevskaâ-Malevič 2005, s. 102–106, 146), pozostaje w mojej opinii nadal kwestią wymagającą dalszych badań. ) p j j j p ą y g ją ą y Zasięg rozprzestrzenienia ceramiki typu drohiczyńskiego w zestawieniu z roz- mieszczeniem cmentarzysk z grobami w obstawach kamiennych na Wysoczyźnie Drohiczyńskiej jest zbieżny, obejmując także obszar występowania tych nekropoli na południe od Bugu (por. ryc. 6g). Naczynia te pojawiły się w podobny sposób jak cmentarze z obstawami kamiennymi: „nagle”, mając wykształcone formy i konwen- cję stylistyczną (Wójcik 2014, s. 90). Ich udział w zespołach ceramicznych datowa- nych na XI w. wynosi od 40% do 59%, natomiast w pochodzących z XII w. sięga już 85% (Wójcik 2014, s. 80). 24 Na temat kształtowania się takich jednostek, o zróżnicowanej randze w polityczno-admini- stracyjnej organizacji księstw zachodnioruskich – zob. Janeczek 2017. 2.4. JESZCZE RAZ ARCHEOLOGIA ORAZ DANE HISTORYCZNE I ONOMASTYCZNE Poszukiwanie pierwowzorów w innych ośrodkach Słowiańszczyzny Wschodniej i Zachodniej jak dotąd nie przyniosło przekonujących rezultatów, chociaż jako wzorcową wskazuje się ceramikę z Kujaw, Wielkopolski i Mazowsza (m.in. Jaskanis 2008, s. 133–135; Dzik [2010] 2015, s. 129). Interesującą perspektywę dają jednak odkrycia dokonane w Starym Riazaniu nad Oką, gdzie naczynia o identycznych formach i stylistyce zdobień pojawiły się być może jeszcze w 1 poł. XI w. O tym, że były one wytwarzane w miejscowych warsztatach, świadczą relikty pieca garncarskiego, który wraz z naczyniami wewnątrz odkryto na tamtej- szym podgrodziu (Strikalov 2015). Także tam poszukiwania nawiązań do trady- cyjnych, lokalnych wyrobów garncarskich (Wiatycze) nie przyniosły pozytywnych rezultatów. Skłoniło to badaczy do konstatacji, że pojawienie się tej specyficznej wytwórczości związane jest z obecnością obcych rzemieślników, którzy osiedlili się (?) lub zostali osiedleni (?) w Riazaniu w dobie powstania tamtejszego grodu i kształtowania się podległej mu „volosti” (Strikalov 2008). Zbieżność zasięgu występowania tzw. ceramiki drohiczyńskiej oraz cmentarzysk z grobami w obstawach kamiennych na terenie międzyrzecza Bugu i Nurca na północy oraz pomiędzy Toczną, Liwcem i Cetynią na południu (Olczak, Wójcik 2019, s. 156–157, 167) wskazuje, że może to być archeologiczny ślad kształtowa- nia się podobnej jednostki terytorialno-administracyjnej, związanej z ośrodkiem w Drohiczynie24. Jego powstanie w XI w., domyślnie jednoznaczne z początkiem formowania się podległego mu okręgu grodowego, można interpretować jako 146 KATARZYNA SKRZYŃSKA następstwo podziału terytorialnego ziem środkowobużańskich pomiędzy Polskę i Ruś i obustronnym organizowaniem formującego się pogranicza. y g ją g ę p g Wymowę źródeł materialnych wzmacniają ustalenia poczynione na bazie ana- lizy danych historycznych. Kazimierz Pacuski (2019, s. 26–29) zrekonstruował na ich podstawie terytorialny zasięg tej domeny dla 1 poł. XIII – początku XV w., wskazując, że niektóre elementy jego interpretacji odnoszą się również do sytuacji z XI–XII w. (por. też Myśliwski 1994; Sianko 2019). Według wskazanej rekonstruk- cji, na północ od Bugu obszar ten odgradzała rzeka Nurzec, która od północnego zachodu była granicą z Mazowszem, a od północy z ziemią bielską. Wschodni zasięg tego terytorium wyznaczał Nurczyk, a następnie jego granica biegła wodo- działami w kierunku Mielnika, dochodząc do Bugu na wschód od Niemirowa. Na południe od Bugu zachodnią rubież z późniejszą mazowiecką kasztelanią liwską, a także południową z małopolską ziemią łukowską, wyznaczały rzeki Cetynia oraz środkowy i górny Liwiec. Następnie rozgraniczenie sięgało biegu północnej Krzny, gdzie zasięg włości drohiczyńskich mógł obejmować późniejsze średniowieczne dobra Międzyrzec i Stołpno. 2.4. JESZCZE RAZ ARCHEOLOGIA ORAZ DANE HISTORYCZNE I ONOMASTYCZNE Południowo-wschodnia granica z ziemią brzeską rekonstruowana jest natomiast na wododziale Krzny i Bugu, gdzie w zasięgu tery- torium drohiczyńskiego pozostawało dorzecze Tocznej (analogicznie: Dobrowolski 2019) oraz miejscowość Gnojno nad Bugiem. Bardzo podobny obraz dała rekon- strukcja średniowiecznej sieci parafii katolickich i przynależących do nich wsi, którą sporządził Adrian Jusupović (2010), świadcząca o tym, że była ona organizowana w analogicznym zasięgu. Jak zauważył Andrzej Janeczek (2017, s. 137–139), stare terytoria grodowe służyły bowiem do oznaczania granic diecezji kościoła łaciń- skiego, który od 2 poł. XIV w. tworzył na terytorium zachodniej Rusi zręby własnej organizacji terytorialnej. Rekonstrukcje: archeologiczna i historyczna korespondują także z ustaleniami Zygmunta Gałeckiego (2019), który na podstawie źródeł topo- nomastycznych odtworzył sieć średniowiecznych osad służebnych w międzyrzeczu Bugu, Liwca i Krzny. Ich rozmieszczenie interpretować można jako ślad gospodar- czej organizacji południowej części drohiczyńskiego okręgu grodowego, podjętej być może w ostatnim dwudziestoleciu XII w. w związku z czasowym narzuceniem tym terenom polskiej zwierzchności politycznej (szerzej w: Skrzyńska 2019c). Zesta- wienie tych danych wskazuje, że wyniki badań archeologicznych stanowić mogą zasób informacji o najstarszej fazie funkcjonowania historycznej ziemi drohickiej, tj. etapu terytorialnego i osadniczego jej konstytuowania się w 2 poł. XI–XIII w. Ustalenia historyczne i onomastyczne dają natomiast świadectwo o ukształtowanym już obrazie tej struktury w pełnym średniowieczu. 3. PODSUMOWANIE Przedstawione wyniki pozwalają na postawienie hipotezy, że populacja Wyso- czyzny Drohiczyńskiej, która od 2 poł. XI w. zakładała tam cmentarze z grobami w obstawach kamiennych, nie była autochtoniczna i stanowiła obcy substrat napły- wowy. Mogła ona się tam pojawić około połowy wskazanego stulecia, co wiąże się CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 147 przypuszczalnie z powstaniem regionalnego grodu „stołecznego” w Drohiczynie i organizowaniem jego zaplecza osadniczego. Zdefiniowanie chronologicznej rela- cji między powstaniem grodu a zakładaniem tych cmentarzy na obecnym etapie badań nie jest jeszcze możliwe, więc pozostaje ona hipotetyczna. W rezultacie jed- nak, osadnictwo to dało podstawy funkcjonowania „volosti” drohickiej, być może epizodycznie w ostatniej ćwierci XII w. mającej nawet polityczny kształt odrębnego księstwa (Jusupović 2007, s. 25–29, 31–32). Za odmiennością etnokulturową tej ludności, postulowaną na podstawie archeo logicznych badań cmentarzysk tego regionu, przemawiają wyniki badań aDNA. Wskazują one, że ludzie pogrzebani w grobach z obstawami różnili się genetycz- nie od względnie współczesnych im społeczności użytkujących nekropole płaskie i kurhanowe. Tezę tę wspierają także zaobserwowane różnice w wymiarach cza- szek ludności pochowanej na równocześnie funkcjonujących cmentarzach: płaskim w Surażu nad Narwią i z grobami w obstawach kamiennych w Czarnej Wielkiej, usytuowanym w odległości 43 km na południe, na skraju Wysoczyzny Drohiczyń- skiej. Wczesnośredniowieczni Surażanie okazali się być antropologicznie zbliżeni do współczesnej im ludności z Wielkopolski i Kujaw, podczas gdy użytkownicy nekropoli w Czarnej Wielkiej pod względem kraniometrycznym różnili się nie tylko od nich, ale także od wszystkich innych populacji uwzględnionych w zestawieniu J. Gładykowskiej-Rzeczyckiej (2019). y j y j Wskazówek w tym zakresie dostarcza też archeologia. Zasięg zwartego występo- wania ceramiki typu drohiczyńskiego pokrywa się z lokalizacją głównej koncentracji cmentarzysk z grobami w obstawach kamiennych w dorzeczu środkowego Bugu. Wyroby te powszechnie występowały bowiem tylko na Wysoczyźnie Drohiczyńskiej i w północnej części Wysoczyzny Siedleckiej. Stylistyka form i ornamentów tych naczyń nie znajduje bezpośrednich odpowiedników ani na terenie wczesnośrednio- wiecznej Polski, ani innych terenów sąsiednich, chociaż jako pokrewną wskazuje się ceramikę z Mazowsza, Kujaw i Wielkopolski. Najbliższe analogie zidentyfikowano natomiast w Starym Riazaniu nad górną Oką, gdzie jednak naczynia te również są postrzegane jako niezwiązane z lokalną tradycją wytwórczości garncarskiej. Z obserwacjami tymi korespondują także rekonstrukcje historyczne, podjęte w celu zdefiniowania kształtu katolickiej sieci parafialnej oraz oznaczenia całego terytorium ziemi drohickiej w późnym średniowieczu. Ich wyniki uzupełniają bada- nia nad nazwami osad służebnych, które stanowią wątek łączący studia historyczne z archeologiczną analizą zasiedlenia. WYKAZ CYTOWANEJ LITERATURY A n d r z e j e w s k i A . , S i k o r a J . 2009, Drohiczyn średniowieczny i nowożytny w świetle badań z roku 2006, „Podlaskie Zeszyty Archeologiczne”, 5, s. 153–195. A n d r z e j e w s k i A . , S i k o r a J . 2011, Street, square and cemetery in medieval and modern Drohiczyn, „Wratislavia Antiqua. Studia z dziejów Wrocławia”, 13, s. 77–88. B a r a n V. D . 1998, Davnì Slov’âni, Kiїv. B i e n i a k J . 1963, Państwo Miecława. Studium analityczne, Warszawa. B i e ń k o w s k a A . 2010, Wczesnośredniowieczne cmentarzysko typu mazowieckiego w Ski- wach Małych, pow. Siemiatycze, „Podlaskie Zeszyty Archeologiczne”, 6, s. 129–150. B i e ń k o w s k a K . 1980, Drohiczyn, woj. białostockie. Stanowisko XIII, Informator Archeo logiczny. Badania 1979, s. 155. B i e ń k o w s k a K . 1983, Drohiczyn, woj. białostockie. Stanowisko XI, Informator Archeolo- giczny. Badania 1982, s. 187–188. B i e ń k o w s k a K . 1991a, Sprawozdanie z badań wczesnośredniowiecznego kurhanu w Dro- hiczynie, woj. białostockie, „Rocznik Białostocki”, 16, s. 472–480. B i e ń k o w s k a K . 1991b, Sprawozdanie z badań wczesnośredniowiecznego kurhanu na stan. 1 w Janówku Lisowie, woj. białostockie, „Rocznik Białostocki”, 16, s. 469–472. B i e ń k o w s k a K . 1998, Sprawozdanie z ratowniczych badań wykopaliskowych kurhanu wczesnośredniowiecznego w Smarklicach gm. Drohiczyn, „Biuletyn Konserwatorski Woje- wództwa Podlaskiego”, 4, s. 204–208. B i e ń k o w s k a K . 2003, Wczesnośredniowieczny kurhan w Wólce Zamkowej, pow. Siemia- tycze, [w:] M. Dulinicz red. 2003, s. 269–271. B i e ń k o w s k a K . 2005, Cmentarzysko wczesnośredniowieczne w Surażu, woj. podlaskie, „Podlaskie Zeszyty Archeologiczne”, 1, s. 121–166. B u k o A . r e d . 2019, Początki chrześcijaństwa na pograniczu mazowiecko-ruskim w świetle wyników badań wybranych cmentarzysk, Warszawa. B u k o A . , B o g d a n o w i c z W. , K r a s n o d ę b s k i D. , M o l a k M . 3. PODSUMOWANIE Ta ostatnia wskazuje natomiast, że Wysoczyzna Drohiczyńska co najmniej do końca X w. znajdowała się na etapie wstępnej tylko infiltracji osadniczej. Wykrystalizowane mikroregiony, świadczące o punktowej sta- bilizacji zasiedlenia, powstały w południowej części tego obszaru, nad niewielkimi ciekami – dopływami Bugu lub bezpośrednio przy głównej rzece. Radykalna zmiana sytuacji notowana jest w XI–XII w., kiedy całe międzyrzecze Bugu i Nurca stało się strefą trwale zagospodarowaną. Wobec przedstawionych powyżej argumentów należy zmienić podejście do informacji zawartej w Rocznikach Jana Długosza (Długosz 1962, s. 323), gdzie pod datą 1112 znalazł się zapis: „Szczep zaś Jaćwingów, jeśli chodzi o narodowość, język, 148 KATARZYNA SKRZYŃSKA obrzędy, religię i obyczaje, był bardzo podobny do Litwinów, Prusów i Żmudzinów. Uprawiał także bałwochwalstwo. Jego głównym miastem i stolicą był zamek i mia- sto Drohiczyn”. Domysł kronikarza odnoszący się do konkretnej przynależności etnicznej jest oczywiście nieuzasadniony25, ale istotną informacją jest to, że wie- dza o „obcości” tamtejszych mieszkańców mogła dotrwać do XV w. i znalazła swe odzwierciedlenie w ówczesnych źródłach pisanych. 25 Ostateczne podważenie wiarygodności tej informacji – por. Tyszkiewicz 1975, s. 176–191 WYKAZ CYTOWANEJ LITERATURY , S k r z y ń s k a K . 2020, Medieval populations of the Mazovian-Rus‘ frontier in the time of Christianiza- tion, „Archäologisches Korrespondenzblatt”, 50/4, s. 579–600. C h i l m o n K . 1981a, Sprawozdanie z badań wczesnośredniowiecznego kurhanu na st. I w Janówku Lisowie, gm. Drohiczyn, „Rocznik Białostocki”, 15, s. 292–299. C h i l m o n K . 1981b, Nowoodkryte kurhany wczesnośredniowieczne w Janówku Lisowie, gm. Drohiczyn, woj. białostockie, „Rocznik Białostocki”, 15, s. 299–300. Č a r a ŭ k o V. U. 2018, Kamennyâ nadmagilli belaruskaga Padzvinnâ XIV–XVIII stst., [w:] Aktual’nye problemy arhitektury Belorusskogo Podvin’â i sopredel’nyh regionov, V.E. Ovsejčik red., Novopolock, s. 122–131. CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 149 D ą b r o w s k i K . 1955, Grodzisko na łąkach Bużyski, pow. Siemiatycze, „Wiadomości Archeo logiczne”, 22, s. 88–89. D ą b r o w s k i K . 1955, Grodzisko na łąkach Bużyski, pow. Siemiatycze, „Wiadomości Archeo logiczne”, 22, s. 88–89. D ł u g o s z 1962, Jana Długosza Roczniki czyli Kroniki sławnego Królestwa Polskiego, J. Dąbrow- ski red., M. Plezia przekł., IV, Warszawa. D o b r o w o l s k i T. 2019, Osadnictwo dorzecza górnej Tocznej w świetle źródeł historycznych, [w:] Przemiany osadnicze w dorzeczu górnej Tocznej we wczesnym i późnym średniowieczu, J. Kalaga, K. Skrzyńska red., Warszawa, s. 67–110, maszynopis przechowywany w Archi- wum Narodowego Instytutu Dziedzictwa w Warszawie. D u l i n i c z M . 1998, Jaki obrządek pogrzebowy panował na wczesnośredniowiecznym Mazow- szu?, [w:] Studia z dziejów cywilizacji. Studia ofiarowane Profesorowi Jerzemu Gąssow- skiemu w pięćdziesiątą rocznicę pracy naukowej, A. Buko red., Warszawa, s. 103–113. D u l i n i c z M . 2005, Mazowsze w XI–XIII w., [w:] Problemy przeszłości Mazowsza i Podlasia, M. Dulinicz red., Archeologia Mazowsza i Podlasia. Studia i Materiały, III, Warszawa, s. 187–206. D u l i n i c z M . r e d . 2003, Słowianie i ich sąsiedzi we wczesnym średniowieczu, Lublin– Warszawa. D z i k M . 2008, Tomasz Kordala, Wczesnośredniowieczne cmentarzyska szkieletowe na pół- nocnym Mazowszu, Monografie Instytutu Archeologii Uniwersytetu Łódzkiego, t. V, Łódź 2006, „Slavia Antiqua”, 49, s. 253–262. D z i k M . 2009, Co nam powiedzą kamienie? Rzecz o grobowych konstrukcjach na cmenta- rzyskach wczesnośredniowiecznych, Funeralia Lednickie. Spotkanie 11: Metody. Źródła. WYKAZ CYTOWANEJ LITERATURY Dzik red., Materiały V Kongresu Mediewistów Polskich, III, Rzeszów, s. 79–121. nicza w polskich badaniach mediewistycznych, A. Janeczek, M. Parczewski, M. Dzik red., Materiały V Kongresu Mediewistów Polskich, III, Rzeszów, s. 79–121. D z i k M . 2021, Ludwik de Fleury i wczesnośredniowieczne cmentarzyska z grobami w obu- dowach kamiennych na Wysoczyźnie Kolneńskiej, „Archeologia Polski”, 66, s. 119–182. D z i k M . , J u s u p o v i ć A . 2019, Średniowieczny Drohiczyn w świetle źródeł archeolo- gicznych i pisanych, [w:] Sfinks słowiańskiej sfragistyki – plomby „typu drohiczyńskiego” z Czermna na wschodnioeuropejskim tle porównawczym, A. Musin, M. Wołoszyn red., Kraków–Leipzig–Rzeszów–Sankt Petersburg–Warszawa, s. 77–101. G a ł e c k i Z . 2019, Rekonstrukcja średniowiecznego podłoża etnokulturowego dorzecza Tocznej w świetle danych onomastycznych, [w:] Przemiany osadnicze w dorzeczu górnej Tocznej we wczesnym i późnym średniowieczu, J. Kalaga, K. Skrzyńska red., Warszawa, s. 29–66, maszynopis przechowywany w Archiwum Narodowego Instytutu Dziedzictwa w War- szawie. G a ł e c k i Z . , S k r z y ń s k a K . 2021, Wiejskie struktury osadnicze i ich przemiany w czasie: przykład Nowosielca na dawnym pograniczu polsko-rusko-litewskim, „Przegląd Archeo- logiczny”, 69, s. 171–190. g y G i e r l a c h B . 1975, Studia nad archeologią średniowiecznego Mazowsza, Warszawa. G ł a d y k o w s k a - R z e c z y c k a J . 2019, Biostruktura ludności wczesnośredniowiecznego Suraża, [w:] A. Buko red. 2019, s. 179–198. G ó r s k a I . , P a d e r e w s k a L . , P y r g a ł a J. , S z y m a ń s k i W. , G a j e w s k i L . , O k u l i c z Ł . 1976, Grodziska Mazowsza i Podlasia (w granicach dawnego województwa warszawskiego), Wrocław–Warszawa–Kraków. G r z e g o r c z y k A . 2009, Obecność warego-ruska na Mazowszu w świetle źródeł archeolo- gicznych, [w:] Vade Nobiscum. Pokłosie XV ogólnopolskiego zjazdu historyków studentów, II: Materiały z II łódzkiej wiosny historyków 20–22 III 2009, Łódź, s. 11–18. Ì o ŭ A . V. 1992, Ètnìčny sklad nasel’nìctva Zahodnâga Palessâ ŭ X–XIII stst. (pa matèryâlah sel´skih paseliščaŭ), [w:] Staronkì Gìstoryì Belarusì, Mìnsk, s. 27–30. WYKAZ CYTOWANEJ LITERATURY Dokumentacja, W. Dzieduszycki, J. Wrzesiński red., Poznań, s. 257–261. D z i k M . 2010, Grobowe konstrukcje kamienne na późnośredniowiecznych cmentarzyskach w międzyrzeczu Bugu i Narwi – przyczynek do badań, [w:] In silvis, campis… et urbe. Śred- niowieczny obrządek pogrzebowy na pograniczu polsko-ruskim, S. Cygan, M. Glinianowski, P. Kotowicz red., Rzeszów–Sanok, s. 295–304. D z i k M . [2009–2010] 2011, Korzeniówka Mała, woj. podlaskie. Badania w latach 2009–2010, „Światowit”, 8 (49)/B, s. 221–222. D z i k M . 2012, Przemiany w średniowiecznych zwyczajach pogrzebowych w międzyrzeczu środkowego Bugu i górnej Narwi (XI–XV w.), Warszawa, maszynopis pracy doktorskiej przechowywany w Archiwum Wydziału Archeologii Uniwersytetu Warszawskiego. D z i k M . 2014a, Uwagi o funkcji i symbolice konstrukcji wczesnośredniowiecznych grobów w obu- dowach kamiennych, [w:] Grób w przestrzeni, przestrzeń w grobie. Przestrzenne uwarunko- wania w dawnej obrzędowości pogrzebowej, T. Kurasiński, K. Skóra red., Łódź, s. 87–101. D z i k M . 2014b, Rodzaje grobów, [w:] Średniowieczne cmentarzysko w Czarnej Wielkiej, stan. 1, woj. podlaskie (badania 1951–1978), II, H. Karwowska red., Białystok, s. 9–25. D z i k M . 2014c, Analiza chronologiczno-przestrzenna cmentarzyska, [w:] Średniowieczne cmentarzysko w Czarnej Wielkiej, stan. 1, woj. podlaskie (badania 1951–1978), II, H. Kar- wowska red., Białystok, s. 215–226. D z i k M . [2010] 2015, Early medieval stone barrows in the area between the Bug and the Upper Narew Rivers, „Archaeologia Polona”, 48, s. 121–138. g D z i k M . 2015a, Przemiany w średniowiecznych zwyczajach pogrzebowych w międzyrzeczu Bugu i górnej Narwi (XI–XV w.), 1, Rzeszów. g D z i k M . 2015a, Przemiany w średniowiecznych zwyczajach pogrzebowych w międzyrzeczu Bugu i górnej Narwi (XI–XV w.), 1, Rzeszów. g g j D z i k M . 2015b, Przemiany w średniowiecznych zwyczajach pogrzebowych w międzyrzeczu Bugu i górnej Narwi (XI–XV w.), 2, Rzeszów. D z i k M . 2015b, Przemiany w średniowiecznych zwyczajach pogrzebowych w międzyrzeczu Bugu i górnej Narwi (XI–XV w.), 2, Rzeszów. D z i k M . 2019, Cmentarzyska i migracje. Osadnictwo w północnej części pogranicza polsko- -ruskiego w XI – XII wieku w świetle materiałów ze stanowisk sepulkralnych, [w:] Pogra- D z i k M . 2019, Cmentarzyska i migracje. Osadnictwo w północnej części pogranicza polsko- -ruskiego w XI – XII wieku w świetle materiałów ze stanowisk sepulkralnych, [w:] Pogra- 150 KATARZYNA SKRZYŃSKA nicza w polskich badaniach mediewistycznych, A. Janeczek, M. Parczewski, M. WYKAZ CYTOWANEJ LITERATURY p y I o v O . V. 1992, Krugovaâ keramika poselenij IX–XI vv. zapadnoj časti Belorusskogo Poles’â, [w:] Drevnerusskaâ keramika, S.A. Pletneva red., s. 36–47. J a n e c z e k A . 2017, Organizacja terytorialna zachodnich ziem Rusi pomiędzy pełnym i póź- nym średniowieczem, [w:] U źródeł Europy Środkowo-Wschodniej, 1, Od Grodów Czer- wieńskich do linii Curzona, M. Wołoszyn red., Kraków–Leipzig–Rzeszów–Warszawa, s. 107–157. J a s k a n i s D . 2008, Święck. Wczesnośredniowieczny zespół osadniczy na północno-wschodnim Mazowszu, Warszawa. J u r a s A . 2012, Etnogeneza Słowian w świetle badań kopalnego DNA, Poznań, praca dok- torska przechowywana w Archiwum Instytutu Antropologii Uniwersytetu im Adama Mickiewicza. J u s u p o v i ć A . 2007, Tak zwany „Latopis Połocki“ w przekazie Wasilija Tatiščeva. Rola Dro- hiczyna w kontaktach polsko-ruskich drugiej połowy XII wieku, „Studia Źródłoznawcze”, 45, s. 15–32. J u s u p o v i ć A . 2010, Zasięg terytorialny ziemi drohickiej w średniowieczu, Aktual’ny prob- lemi vitčiznănoj ta vsesvitnoj istorii. Zbirnik naukowych prac, Naukovy zapiski Rivnen- skovo deržavnovo gumanitarnovo universitetu, 20, Rivne, s. 77–80. K a l a g a J . 1989, Wczesnośredniowieczny zespół osadniczy w Krzesku-Królowej Niwie, woj. siedleckie, Warszawa, maszynopis pracy doktorskiej przechowywany w Archiwum Wydziału Archeologii Uniwersytetu Warszawskiego. CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 151 K a l a g a J . 2006, Ciałopalny obrządek pogrzebowy w międzyrzeczu Bugu, Liwca i Krzny we wczesnym średniowieczu, Warszawa. y K a l a g a J. 2019, Problem trwania elementów dawnej obrzędowości w dobie konwersji religijnej (XI – 1. połowa XIII w.), [w:] A. Buko red. 2019, s. 259–271. K a r o b u š k i n a T. N . 1999, Nasel´nìctva Belaruskaga Pabužža. X–XIII stst., Mìnsk. K a r o b u š k i n a T. N . 2000, Zemli Sârèdnâga Pabužža, [w:] Gistoryâ Belarusi, 1, Staražytnaâ Belarus’. Ad peršapačatkovaga zasâlennâ da sârèdziny XIII st., M. Čarnâŭski, G. Štyhaŭ red., Mìnsk, s. 195–201. K a r w o w s k a H . r e d . 2013, Średniowieczne cmentarzysko w Czarnej Wielkiej, stan. 1, woj. podlaskie (badania 1951–1978), 1, Białystok. K a r w o w s k a H . r e d . 2014, Średniowieczne cmentarzysko w Czarnej Wielkiej, stan. 1, woj. podlaskie (badania 1951–1978), 2, Białystok. WYKAZ CYTOWANEJ LITERATURY p y K i e r s n o w s k a T. 1992, O pochodzeniu rodu Awdańców, [w:] Społeczeństwo Polski śred- niowiecznej. Zbiór studiów, S.K. Kuczyński red., 5, Warszawa, s. 57–72. K o b y l i ń s k a U. 2003, Ceramika starożytna i wczesnośredniowieczna z grodziska w Kluko- wiczach na Podlasiu, [w:] M. Dulinicz red. 2003, s. 441–452. K o b y l i ń s k i Z . , S z y m a ń s k i W. 2015, Grodzisko w Haćkach na Podlasiu, [w:] Grody średniowiecznego Mazowsza. Księga poświęcona pamięci Marka Dulinicza, M. Żurek, M. Krasna-Korycińska red., Archaeologica Hereditas, 4, Warszawa–Zielona Góra, s. 111– 145. K o r d a l a T. 2000, Z dziejów badań archeologicznych na Mazowszu Płockim, Płock K o r d a l a T. 2003, Podstawy chronologii wczesnośredniowiecznych cmentarzysk szkieletowych na północnym Mazowszu, [w:] M. Dulinicz red. 2003, s. 301–310. K o r d a l a T. 2005a, Wczesnośredniowieczne cmentarzyska szkieletowe na północnym Mazow- szu, [w:] Problemy przeszłości Mazowsza i Podlasia, M. Dulinicz red., Archeologia Mazow- sza i Podlasia. Studia i Materiały, III, s. 223–239. K o r d a l a T. 2005b, W sprawie wczesnośredniowiecznych cmentarzysk szkieletowych na pół- nocnym Mazowszu. Refleksje na marginesie artykułu R. Piotrowskiego (2003), „Archeologia Polski”, 50/1–2, s. 147–153. K o r d a l a T. 2005c, Wczesnośredniowieczne cmentarzyska szkieletowe z grobami w obudo- wach kamiennych na północnym Mazowszu i Podlasiu – próba porównania, „Archaeologia Historica Polona”, 15/2, s. 51–60. K o r d a l a T. 2006, Wczesnośredniowieczne cmentarzyska szkieletowe na Północnym Mazow- szu, Łódź. K o r o b u š k i n a T. N . 1993, Kurgany belorusskogo Pobuž´â. X–XIII vv. Minsk. K o s t r z e w s k i J . 1939–1948, Od mezolitu do okresu wędrówek ludów, [w:] Prehistoria ziem polskich, S. Krukowski, J. Kostrzewski, R. Jakimowicz red., Kraków, s. 118–360. K o w a l c z y k E . 2003, Dzieje granicy mazowiecko-krzyżackiej między Drwęcą a Pisą, War- szawa. K r a s n o d ę b s k i D . 2000, Podlasie na przełomie I i II millenium w świetle wyników naj- nowszych badań archeologicznych, [w:] Osadnictwo i architektura ziem polskich w dobie Zjazdu Gnieźnieńskiego, A. Buko, Z. Świechowski red., Warszawa, s. 179–190. WYKAZ CYTOWANEJ LITERATURY K r a s n o d ę b s k i D . 2003, Fiodora Vasileviča Pokrovskiego obraz archeologiczny Podlasia, [w:] M. Dulinicz red. 2003, s. 63–71. K r a s n o d ę b s k i D . 2004, Archeologiczne zabytki powiatu hajnowskiego, Hajnówka. K r a s n o d ę b s k i D . 2006, Badania Instytutu Archeologii i Etnologii PAN na terenie woj. białostockiego (podlaskiego) w latach 1990–2005, [w:] Stan badań archeologicznych na 152 KATARZYNA SKRZYŃSKA pograniczu polsko-białoruskim od wczesnego średniowiecza po czasy nowożytne, H. Kar- wowska, A. Andrzejewski red., Białystok, s. 63–84. pograniczu polsko-białoruskim od wczesnego średniowiecza po czasy nowożytne, H. Kar- wowska, A. Andrzejewski red., Białystok, s. 63–84. K r a s n o d ę b s k i D . 2011–2012, Wczesnośredniowieczne grody nad górną Narwią i ich związki ze szlakami handlowymi, „Podlaskie Zeszyty Archeologiczne”, 7, s. 153–170. K r a s n o d ę b s k i D . , O l c z a k H . 2019, Kontynuacja osadnictwa czy ponowne wyko- rzystanie starszych nekropoli? Dwa cmentarzyska kurhanowe w Uroczysku Szczekotowo w Puszczy Białowieskiej oraz cmentarzyska kurhanowe i z grobami w obstawach kamien- nych w Zbuczu, [w:] A. Buko red. 2019, s. 55–67. K u c z y ń s k i S . M . [1954–1956] 1956, Nieznany traktat polsko-ruski z r. 1039, „Slavia Antiqua”, 5, s. 255–276. K u h a r e n k o J . V. 1968, Poles’e i ego mesto v ètnogeneze slavân, [w:] Poles’e (lingvistika, arheologiâ, toponimika), Minsk, s. 243–246. Kv â t k o v s k a â A . V. 1998, Âtvâžskie mogil’niki Belarusi (k. XI–XVII vv.), Vilnûs. L e p i o n k a H . 2020, Cmentarzyska ze stelami a praktyki pogrzebowe chłopów litewskich w okresie nowożytnym, [w:] Małe Miasta. Duchowość kanoniczna, M. Zemło red., Biały- stok–Futoma–Supraśl, s. 581–597. L y s e n k o P. F. 1985, Berest’e, Minsk. L y s e n k o P. F. 1991, Dregoviči, Minsk. L y s e n k o P. F. 2007, Otkrytie Berest’â, Minsk. Ł u n i e w s k i T. 1883, Cmentarzyska starożytne w Łużkach i Grodzisku w guberni siedleckiej w powiecie sokołowskim, „Pamiętnik Fizyjograficzny”, 3, s. 477–478. WYKAZ CYTOWANEJ LITERATURY i M a l e v s k a â - M a l e v i č M . V. 2005, Keramika zapadnorusskih gorodov X–XIII vv., Sankt- -Peterburg. M i l â n T. 2004, Rann’oslov’ăns’ka keramìka u Galičinì ta Volinì, „Arheologìčnì doslìdžennă L’vìvs’kogo Unìversitetu”, 7, s. 104–122. M i ś k i e w i c z M . 1981, Mazowsze wschodnie we wczesnym średniowieczu, Warszawa M i ś k i e w i c z M . 1998, Mazowieckie pochówki szkieletowe w XI i XII w., [w:] Studia z dzie- jów cywilizacji. Studia ofiarowane Profesorowi Jerzemu Gąssowskiemu w pięćdziesiątą rocznicę pracy naukowej, A. Buko red., Warszawa, s. 115–118. p y j M i ś k i e w i c z o w a M . 1996a, Wczesnośredniowieczny kompleks osadniczy w Niewiadomej w województwie siedleckim, Warszawa. M i ś k i e w i c z o w a M . 1996b, Wczesnośredniowieczna ceramika typu drohiczyńskiego, [w:] Concordia. Studia ofiarowane Jerzemu Okuliczowi-Kozarynowi w sześćdziesiątą piątą rocz- nicę urodzin, W. Nowakowski red., Warszawa, s. 183–185. M o l a k M . , S u c h e c k a E . , B o g d a n o w i c z W. 2019, Badania genetyczne szcząt- ków kostnych z cmentarzysk na pograniczu mazowiecko-ruskim, [w:] A. Buko red. 2019, s. 233–247. M u s i a n o w i c z K . 1950/1951, Z zagadnień osadnictwa wczesnohistorycznego pow. Sokołów Podlaski, „Wiadomości Archeologiczne”, 17/4, s. 229–250. M u s i a n o w i c z K . 1955, Przyczynki do osadnictwa mazowieckiego na Podlasiu w XII– XIII w., „Światowit”, 21, s. 147–175. M u s i a n o w i c z K . 1960, Granica mazowiecko-drehowicka na Podlasiu we wczesnym śred- niowieczu, „Materiały Wczesnośredniowieczne”, 5, s. 187–230. M u s i a n o w i c z K . 1962, Materiały i problematyka lokalnych grup wczesnośredniowiecznej ceramiki Podlasia, „Światowit”, 24, s. 578–609. M u s i a n o w i c z K . 1966, Ślady osadnictwa z VI–VIII wieku w Drohiczynie, pow. Siemiatycze, „Wiadomości Archeologiczne”, 32, s. 31–53. CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 153 M u s i a n o w i c z K . 1969, Drohiczyn we wczesnym średniowieczu, „Materiały Wczesnośred- niowieczne”, 6, s. 7–235. WYKAZ CYTOWANEJ LITERATURY M u s i a n o w i c z K . 1978, Osadnictwo z VI–VIII wieku w Drohiczynie, w woj. białostockim, „Wiadomości Archeologiczne”, 43/1, s. 83–104. M y ś l i w s k i G . 1994, Powstanie i rozwój granicy liniowej na Mazowszu (XII – poł. XVI w.), „Kwartalnik Historyczny”, 101/3, s. 3–24. y y N o w a k o w s k i W. 2011, „Kurhany Jaćwięgów”. Kilkadziesiąt lat naukowego mitu, „Świa- towit”, IX (L)/B, s. 181–192. O b r ę b s k i J . 2007, Polesie, [w:] Studia etnosocjologiczne, 1, A. Engelking red. i oprac., Warszawa. O k u l i c z o w i e Ł i J . 1963, Dwa wczesnośredniowieczne cmentarzyska z grobami cia- łopalnymi a problem obrządku pogrzebowego na północnym Mazowszu, „Wiadomości Archeologiczne”, 29/2, s. 105–118. g O l c z a k H . , K r a s n o d ę b s k i D. 2019a, Wschodniosłowiańskie szkieletowe cmentarzysko kurhanowe w Uroczysku Jelonka w Puszczy Białowieskiej, [w:] A. Buko red. 2019, s. 67–72. O l c z a k H . , K r a s n o d ę b s k i D . 2019b, Osobliwości obrzędowości funeralnej na przy- kładzie cmentarzyska z grobami w obstawach kamiennych i kurhanami w Szczytach-Dzię- ciołowie, [w:] A. Buko red. 2019, s. 72–77. O l c z a k H . , K r a s n o d ę b s k i D . 2020, Suraż na tle osadnictwa pogranicza mazowie- cko-rusko-litewskiego w okresie od XI do XVI w., [w:] Suraż – średniowieczny gród na pograniczu mazowiecko-rusko-litewskim, D. Krasnodębski, H. Olczak red., Warszawa, s. 469–506. O l c z a k H . , K r a s n o d ę b s k i D . , B i e ń k o w s k a K . 2019, Daniłowo Małe i Suraż: dwa cmentarzyska płaskie nad górną Narwią – nowe spojrzenie na wyniki badań, [w:] A. Buko red. 2019, s. 77–90. O l c z a k H . , Wó j c i k A . 2019, Ceramika naczyniowa z wczesnośredniowiecznych cmen- tarzysk dorzecza środkowego Bugu jako wskaźnik tradycji regionalnych i przemian kultu- rowych, [w:] A. Buko red. 2019, s. 153–168. y P a c u s k i K . WYKAZ CYTOWANEJ LITERATURY 2019, Wschodnia granica Mazowsza w średniowieczu w świetle danych histo- rycznych (X–XV w.), [w:] A. Buko red. 2019, s. 15–32. y y P a r c z e w s k i M . 1988, Najstarsza faza kultury wczesnosłowiańskiej w Polsce, Kraków P a r c z e w s k i M . 1988, Najstarsza faza kultury wczesnosłowiańskiej w Polsce, Kraków. P a r c z e w s k i M 2004 Uwagi o kulturze wczesnosłowiańskiej na ziemiach polskich P a r c z e w s k i M . 1988, Najstarsza faza kultury wczesnosłowiańskiej w Polsce, Kraków. P a r c z e w s k i M . 2004, Uwagi o kulturze wczesnosłowiańskiej na ziemiach polskich, [w:] Wspólnota dziedzictwa kulturowego ziem Białorusi i Polski, A. Kośko red., s. 266–277. z e w s k i M . 2004, Uwagi o kulturze wczesnosłowiańskiej na ziemiach polskich, w:] Wspólnota dziedzictwa kulturowego ziem Białorusi i Polski, A. Kośko red., s. 266–277. P a w l a t a L . 1991, Wyniki badań nowo odkrytego cmentarzyska wczesnośredniowiecznego w Drohiczynie, woj. białostockie, „Rocznik Białostocki”, 17, s. 111–126. P i o t r o w s k i R . 2003, Uwagi na temat chronologii i interpretacji wczesnośredniowiecznych cmentarzysk mazowieckich, „Archeologia Polski”, 48/1–2, s. 165–200. R a u h u t L . 1971, Wczesnośredniowieczne cmentarzyska w obudowie kamiennej na Mazowszu i Podlasiu, „Materiały Starożytne i Wczesnośredniowieczne”, 1, s. 435–656. R a u h u t L . , D ł u g o p o l s k a L . 1972, Wczesnośredniowieczne cmentarzysko szkieletowe w obudowie kamiennej w Łączynie Starym, pow. Przasnysz, „Wiadomości Archeologiczne”, 37/3, s. 320–393. R a u h u t L . , D ł u g o p o l s k a L . 1973, Wczesnośredniowieczne cmentarzysko szkieletowe w Tańsku-Przedborach, pow. Przasnysz, „Wiadomości Archeologiczne”, 38/3–4, s. 383–441. R a u h u t L . , D ł u g o p o l s k a L . 1973, Wczesnośredniowieczne cmentarzysko szkieletowe w Tańsku-Przedborach, pow. Przasnysz, „Wiadomości Archeologiczne”, 38/3–4, s. 383–441. R o l e c k i R . 2006, Czudzin, wieś powiatu łuninieckiego. Stosunki społeczno-gospodarcze, „Białoruskie Zeszyty Historyczne”, 26, s. 146–217. R o l e c k i R . 2006, Czudzin, wieś powiatu łuninieckiego. Stosunki społeczno-gospodarcze, „Białoruskie Zeszyty Historyczne”, 26, s. 146–217. S e d o v V. V. WYKAZ CYTOWANEJ LITERATURY 2000, Žal’niki, „Rossijskaâ Arheologiâ”, 1, s. 7–22. 154 KATARZYNA SKRZYŃSKA S i a n k o P. 2019, Granica mazowiecko-litewska do 1569 roku. Kształtowanie się, rola spo- łeczna. Praca doktorska napisana pod kierunkiem prof. dr. hab. J. Tęgowskiego, maszyno- pis przechowywany w Archiwum Instytutu Historii i Nauk Politycznych Uniwersytetu w Białymstoku. S i k o r a J . 2013, Ethnos or ethos? Some remarks on interpretation of early medieval elite burials in northern Poland, [w:] Scandinavian Culture in Medieval Poland, S. Moździoch, B. Stanisławski, P. Wiszewski red., Wrocław, s. 411–436. S k r z y ń s k a K . 2001, Najstarsze osadnictwo wczesnośredniowieczne Podlasia na przykładzie wybranych stanowisk, [w:] Najstarsze dzieje Podlasia w świetle źródeł archeologicznych, B. Bryńczak, P. Urbańczyk red., Siedlce, s. 169–204. S k r z y ń s k a K . 2019a, Wczesny etap chrystianizacji (XI–XIII w.): pomiędzy historią i archeo logią, [w:] A. Buko red. 2019, s. 273–294. g S k r z y ń s k a K . 2019b, „Nieznane” grodzisko w Dzięciołach, pow. łosicki. Perspektywy badaw- cze, „Wiadomości Archeologiczne”, 70, s. 51–62. S k r z y ń s k a K . 2019c, Dzięcioły-Łosice. Translokacja ośrodków grodowych i protomiejskich na średniowiecznym pograniczu mazowiecko-małopolsko-litewskim, [w:] Przemiany osadni- cze w dorzeczu górnej Tocznej we wczesnym i późnym średniowieczu, J. Kalaga, K. Skrzyń- ska red., Warszawa, s. 244–273, maszynopis przechowywany w Archiwum Narodowego Instytutu Dziedzictwa w Warszawie. S k r z y ń s k a K . 2021, Metryka pogranicza. Sieć grodowa dorzecza środkowego Bugu w 2. poło- wie IX i w X wieku, Warszawa. S k r z y ń s k a - J a n k o w s k a K . 2004, Gród w Drohiczynie – centrum organizacji przestrzeni kulturowej wczesnośredniowiecznego pogranicza polsko-ruskiego, [w:] Stan i zmiany śro- dowiska geograficznego wybranych regionów wschodniej Polski, R. Dobrowolski, S. Terpi- łowski red., Lublin, s. 289–297. S k r z y ń s k a - J a n k o w s k a K . 2014, Dorzecze środkowego Bugu jako pogranicze kulturowe we wcześniejszym średniowieczu (poł. VI–X w.), Warszawa, maszynopis przechowywany w Archiwum Instytutu Archeologii i Etnologii PAN. y g g S k r z y ń s k a K . , K r a s n o d ę b s k i D . 2019, Kryteria selekcji materiałów do analiz aDNA, [w:] A. WYKAZ CYTOWANEJ LITERATURY Buko red. 2019, s. 249–258. S o b o l e v V. J . 2015, Drevnerusskaâ pogrebal’naâ kul’tura Novgorodskoj Zemli: problemy i osobennosti formirovaniâ, „Arheologičeskie vesti”, 21, s. 352–367. S P B 1990, Svod pamâtnikov istorii i kul’tury Belorussii. Brestskaâ oblast’, Minsk. S t a n a s z e k Ł . M . 2019, Charakterystyki antropologiczne populacji pogranicza mazowie- cko-ruskiego na przykładzie wybranych grobów z cmentarzysk wschodniego Mazowsza, [w:] A. Buko red. 2019, s. 213–231. S t o ł y h w o K . 1914, Dziennik badań, rękopis przechowywany w Archiwum Państwowego Muzeum Archeologicznego w Warszawie. S t r i k a l o v I . Û. 2008, Drevnerusskaâ kolonizaciă Srednego Pooč‘â v XI veke, [w:] Trudy (XVIII) Vserossijskogo arheologičeskogo s’ezda v Suzdale 2008 g., II (XVIII), Moskva, s. 404–407. S t r i k a l o v I . Û. 2015, Gončarnyj gorn XI veka iz Staroj Râzani, „Kratkie Soobŝeniâ Instituta Arheologii”, 236, s. 263–271. Š t y h o v G . V. 1971, Arheologičeskaâ karta Belorussii, 2, Pamâtniki železnogo veka i èpohi feodalizma, Minsk. Te r p i l o v s k i j R . V. 2004, Slavăne Podneprov‘â v pervoj polovine I tysăčeletiă n.è., Lublin.i Ty s z k i e w i c z J . 1975, Terytorium Jaćwieży w starszej historiografii, [w:] Prace Instytutu Historycznego Uniwersytetu Warszawskiego, C. Drapińska red., 1, Warszawa, s. 99–207. 155 CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH Vâ r g e j V. S . 2004, Arhealogìâ rannìh slаvân na tèrytoryì Belarusì, [w:] Wspólnota dziedzi- ctwa kulturowego ziem Białorusi i Polski, A. Kośko red., Warszawa, s. 278–304. Ve r g e j V. S . 1997, Poseleniă pražskoj kul‘tury Belorusskogo Poles‘ă, [w:] Trudy VI Meždu- narodnogo kongressa slavănskoj arheologii, Minsk, 3, Ètnogenez i ètnokul‘turnye kontakty Slavăn, Moskva, s. 28–38. Ve r g e j V. S . 2005, Pražskaă kul‘tura v Belarusi, [w:] Archeologia o początkach Słowian, P. Kaczanowski, M. Parczewski red., Kraków, s. 487–502. W i t a s H . , B ł a s z c z y k D . , M o l a k M . 2020, Analiza molekularna, [w:] Przestrzeń osadnicza wczesnośredniowiecznego Radomia, 3, Stanowisko 4. Badania interdyscyplinarne, Łódź–Warszawa, s. 127–136. Wó j c i k A . 2008, Chemical analysis of Early Medieval pottery of the „Drohiczyn type”, „Archaeologia Polona”, 46, s. 271–280. WYKAZ CYTOWANEJ LITERATURY Wó j c i k A . 2013, Wczesnośredniowieczne naczynia typu drohiczyńskiego. Próba charaktery- styki produkcji, Warszawa, maszynopis pracy doktorskiej przechowywany w Archiwum Instytutu Archeologii i Etnologii PAN w Warszawie. Wó j c i k A . 2014, Z badań nad naczyniami typu drohiczyńskiego, [w:] Małe miasta. Per- spektywa archeologiczna, M. Zalewski, M. Zemło red., Acta Collegii Suprasliensis, 17, Lublin, s. 79–91. Z o l l - A d a m i k o w a H . 1975, Wczesnośredniowieczne cmentarzyska ciałopalne Słowian na terenie Polski, 1, Źródła, Wrocław–Warszawa–Kraków–Gdańsk. S u m m a r y An analogous timeframe could be suggested for the cemeteries in the Leśna (Lesnaya) river basin (western Belarus), where these funerary customs could have survived into the early modern period (Kvâtkovskaâ 1998, pp. 159–167; Čaraŭko 2018).i Research on the origins of these cemeteries and an ethno-cultural identification of the population buried there have not resolved doubts, which are expressed in the debate as two opposing views: an “autochthonous” theory assuming a local origin and an “allochthonous” one opting for a foreign origin, either copying or transferring the funerary model from elsewhere in Europe. Neither has been argued conclusively so far. Interdisciplinary research, verifying archaeological inferences with biological and physico-chemical analyses, could be a way of getting past the impasse.hih The first data of this kind was supplied by the research project “The origins of Christianity on the Masovian–Rus’ borderland”, carried out under the National Program for the Development of the Humanities (Buko red. 2019). The emergence, flourishing and decline of the graves with stone casing cemetery tradition in Podlasie was an important aspect of research under this program. Inhuma- tion burials are common in the barrow graves of the eastern part of the region, where they are dated securely to the 11th and 12th centuries, and hypothetically to the end of the 10th c. and to the 11th c. (Korobuškina 1993; Olczak, Krasnodębski 2019a, pp. 67–72; 2019b, pp. 74–77). In the west (around Drohiczyn), inhumations are rare in the barrow graves but prevalent in the flat cemeteries consisting of graves with stone casing. The number of these burial grounds and a consistent dating of the initial phases (second half or fourth quarter of the 11th c.) suggest a short-term episode, either a migration/ colonization or a Christianizing action, introducing a new sepulchral tradition.hl The idea that the graves with stone casing forming flat cemeteries in Podlasie evoke a wave of migration rather than a local substrate is not new in itself (Dzik 2019, pp. 80–88). Soviet and Belarus researchers have long considered the cemeteries to be the burial places of Balts who had undergone Slavization or even of specifically the Yotvingians (Kvâtkovskaâ 1998). In Poland, the idea of migrants from Masovia settling in the region has long dominated the debate (Miśkiewicz 1981, pp. 92–104; 112–113). In opposition to this idea is the theory that this particular sepulchral model emerged on the spot in Podlasie. S u m m a r y Cemeteries of graves with stone casing are typical of the Slav-Balt borderland stretching from the Novgorod Land in the northeast to Masovia in the southwest (Fig. 1). These burial grounds are inter- preted as belonging to a population formally converted to Christianity based on the inhumation rites practiced therein (see Fig. 2). In the oldest phase, cremation burials constituted a regressive wave in these funerary customs. Other themes characteristic of this transitional stage are also evident, appar- ently referring to the apotropaic and resurrection symbolism taking root at that time. Clusterings of these cemeteries have been noted in the northern Masovia and in the Podlasie regions within the territory of Poland (Fig. 3). However, the two differ both in the way individual graves are constructed and in the layout of the cemetery space (Fig. 4). The oldest of the Masovian burial grounds (on the upper Orzyc river) are dated to about the mid 11th c. (Dulinicz 2005, p. 197; Kordala 2005a, p. 227), perhaps even the beginning of the 11th c. according to the latest research (results of 14C dating performed in 2022 – D. Błaszczyk). In Podlasie (region of Drohiczyn, Siemiatycze district, and Bielsk Podlaski), these cemeteries are younger: from the second half or fourth quarter of the 11th c. (Dzik 2012, pp. 428–448, 697; Skrzyńska 2019a, p. 283). The terminal phase of these funerary rites is even more difficult to establish. In northern Masovia, the suggested date falls in the second half of the 12th c. or the beginning of the 13th c. (Kordala 2003, pp. 308–309; 2006, p. 233), although premises exist to assign some of the cemeteries even to the early modern period (the young- est graves without stone structures, according to results of unpublished 14C dating performed in 2022 – D. Błaszczyk). In Podlasie, the burial grounds remained in use at least into the 14th c. (Dzik 156 KATARZYNA SKRZYŃSKA 2012, pp. 733–739), and it is probable that a fading form of this ritual survived also into the 16th–17th centuries (see Dzik 2010). An analogous timeframe could be suggested for the cemeteries in the Leśna (Lesnaya) river basin (western Belarus), where these funerary customs could have survived into the early modern period (Kvâtkovskaâ 1998, pp. 159–167; Čaraŭko 2018).i 2012, pp. 733–739), and it is probable that a fading form of this ritual survived also into the 16th–17th centuries (see Dzik 2010). CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 157 small tributaries (see Fig. 5b). Single settlements appeared also near the mouth of the Nurzec, while the remaining territory from the Bug in the south to the Nurzec in the north could well have been unoccupied in light of the current state of research. Neither is there any fortified settlement there predating the 11th c. (Skrzyńska 2021, p. 30). hf The observed differences indicate that in terms of cultural development the region at the close of the 10th c. lagged well behind the neighbouring regions to the east and south, preceding a boom that came in the 11th and 12th centuries (Fig. 5c, d). The construction of a stronghold at Drohiczyn, which was to be a regional centre of power in the centuries to follow, marked the beginning of this boom (Dzik, Jusupović 2019, pp. 86–91). It was most probably linked to the organization of the later Drohi- czyn Land (see A. Jusupović 2007). Cemeteries of graves with stone casing appeared in the hinterland of the fort in the second half of the 11th c. Distribution analyses indicate that they were situated on both sides of the Bug, in previously settled areas as well as deep in the Drohiczyn Upland where no older settlement was recorded (Fig. 6; see Fig. 5c, d). The effect was a permanent development of the area, regardless of whether as a result of colonization or some other form of targeted population influx. Given the exceptionally synchronous timeframe of the initial phases of the cemeteries in question, it can be assumed that the area north of Drohiczyn was developed in a short-lived dynamic episode or as an organized action. This theory is supported by the results of palaeogenetic and anthropological analyses and indirectly by historical reconstructions, linguistic studies and yet other archaeological data (see below).fi Material from 103 burials from 21 different cemeteries was selected as filling the requirements for palaeogenetic analyses. Positive results were received for 62 samples. The results were compared (Molak et al. 2019; Buko et al. 2020), among others, within a frame of reference supplied by the location of the cemeteries within separate physio-geographical regions and the differences in grave construction (Fig. 7a), while referring simultaneously to modern European populations (Fig. CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH 7b).i A crucial finding for the study of settlement in the Drohiczyn Upland in the 11th–13th centuries was the distinctness of the population from the contemporary inhabitants of the Bielsk Podlaski Plain and the upper Narew river valley. The cemeteries of graves with stone casing near Drohiczyn are also characterized by the greatest diversity of mitochondrial DNA haplogroups. The difference is well attested also in relation to other criteria: geographical (Fig. 8a) and typological (Fig. 8b). With regard to the “geographical” option, the macro-haplogroup HV was not dominant in the Drohiczyn Upland population, unlike what was found for the other regions. Its share was around 6.7% compared to 50% in the other regions. This result is interesting in the context of the relations between people with a high HV and the autochthonous population culturally identified as originating from Polesie, Volhynia and eastern Little Poland. It is of particular significance for a reconstruction of settlement in the Siedlce Upland and the Bielsk Podlaski Plain. The high share of the HV macro-haplogroup in the native populations supports the archaeological inferences concerning settlement processes in the eastern part of the land between the Bug and the upper Narew, as well as the Bug and the Liwiec and Krzna rivers. These regions, where Slavic settlement started to solidify from the 6th century, may also demonstrate a high level of biological homogeneity. Meanwhile the low frequency of this haplogroup noted for cemeteries of graves with stone casing in the Drohiczyn Upland may be indicative of the indigenous population constituting only a small proportion of the communities burying their dead in these cemeteries. h The cemeteries with stone casing graves also stand out clearly when the analysis respects the typological division of the burial grounds. The populations buried in these cemeteries are charac- terized by the greatest variation in the frequency of individual haplogroups of mitochondrial DNA (Fig. 8). The difference is evident in a comparison with the barrow burials where the haplogoup HV was dominant (about 83.3%), complemented by the Old European I group (about 16.7%). A greater inner variability appears to characterize communities using the flat inhumation cemeteries next to the strongholds. The share of haplogroup HV is 50% in these populations, while the Old European I group constitutes only about 3.9%. Other haplogroups total 46.1%. S u m m a r y Cemeteries of graves with stone casing are said to have appeared first in this region, either due to colonization (Kordala 2005c) or conversion to Christianity of the native population (Dzik 2019, pp. 96–99), and were later transferred to Masovia. This idea is untenable in light of new 14C datings, yielding much earlier dates for the Masovian cemeteries compared to those for Podlasie (results of research in 2022 – D. Błaszczyk). y Closely related to the thesis about a migration wave coming from Masovia is the view about the coexistence of two populations: an autochthonous one cultivating barrow burial rites and an incoming group burying their dead in graves with stone casing (Rauhut 1971). The issue remains unresolved despite progress in research, because the finds from “biritual” cemeteries, that is, cemeteries contain- ing both grave forms, stand in support of both: a simultaneous cultivation of both funerary rites and the replacement of barrow graves with graves having a stone casing (see Musianowicz 1950/1951; Zoll‑Adamikowa 1975, pp. 53–56; Dzik [2009–2010] 2011; Krasnodębski, Olczak 2019, pp. 61–67; Olczak, Krasnodębski 2019b, pp. 76–77). Settlement studies in the Drohiczyn Upland have supplied arguments in favour of the potential colonization of the region in the 11th c. With the exception of the valleys of the Bug river and some of its small tributaries the region was settled either poorly or not at all until the 11th c. The cycle of development was similar to that in the neighbouring regions, that is, in the western part of the Leśna (Lesnaya) river basin and in the Siedlce Upland, but the pace was different (Skrzyńska 2021, pp. 57–73). The Siedlce Upland is best investigated in this respect (Fig. 5a, b), demonstrating settlement conti- nuity combined with an evenly growing number of sites from at least the second half of the 6th c., and a series of strongholds, the oldest of which were constructed in the second half of the 9th and in the 10th centuries (Skrzyńska 2021, pp. 75–92), while the younger ones in the 11th–12th centuries (see Fig. 5c, d). Similar conclusions may be inferred for the basins of the upper Narew, Pulwa, Leśna and upper Nurzec rivers (Fig. 5b; Skrzyńska 2021, pp. 71–73; 75–77). Settlement in the Drohiczyn Upland follows a similar pattern – as said above – but only in the valley of the Bug and some of its CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH The considerable aDNA diversification could indicate that the people buried in these cemeteries came from a largely migrant population, probably related to the location of the major strongholds at the junctions of the communication routes. 158 KATARZYNA SKRZYŃSKA However, it is difficult to explain why the population buried in the cemeteries of graves with stone casing is genetically the most varied. The location of these burial grounds deep in the hinterland of the Drohiczyn stronghold makes an ongoing intensive gene exchange not very likely. The share of HV and I haplogroups, which are believed to define an autochthonous population based on the results of analyses of skeletons from the barrow graves, is about 38.5% in total. The remaining approximately 61.5% are made up of other haplogroups, which are roughly equivalent with respect to one another in terms of the frequency of occurrence. Therefore, the results of aDNA analyses apparently show no close genetic ties between the populations buried in the barrows and in the graves with stone casing. Assuming that the former represent a native population, it can be hypothesized that the latter were for the most part settlers from outside the region or their descendants. They were to a certain degree, in terms of the aDNA variability, similar to the population buried around the major strongholds, which were also presumably co-shaped to a large extent by incoming foreigners. p y p g y g g Further supporting evidence for the theory about a foreign component in the population settling the Drohiczyn Upland came from a comparative anthropological analysis of skeletons from a flat cemetery at Suraż and a burial ground with graves with stone casing at Czarna Wielka, situated at the northern edge of the Drohiczyn Upland (Gładykowska-Rzeczycka 2019). It turned out that there are no ties and craniological similarities between these communities living barely 40 km apart. In terms of skull parameters, the burials from Suraż resembled contemporary burials from cemeteries in Great Poland. As for Czarna Wielka, no such correspondence was determined with any of the anthropologi- cally studied populations of medieval Poland.h The cultural distinctness of the population living around Drohiczyn in medieval times is indicated also by the finding of a specific type of ceramics referred to as “Drohiczyn pottery” (Miśkiewiczowa 1996b). According to research by A. CMENTARZYSKA Z GROBAMI W OBSTAWACH KAMIENNYCH Wójcik (2008; 2013; 2014), this pottery type was restricted to the Drohiczyn Upland and the northern parts of the Siedlce Upland in the 11th–13th centuries. Chemical composition analyses of the fabric of these pottery wares indicated a place of production in Drohiczyn. The distribution range of this particular pottery coincides with that of the cemeteries with graves with stone casing forming the so-called Drohiczyn cluster (see Fig. 6:3). The vessels and the cemeteries emerged there at the same time and in a similarly “sudden” manner: the vessels with a fully developed, mature set of shapes and ornaments, and the graves keeping to a strictly defined funerary canon. The distribution range of the two, the pottery and the cemeteries, could be considered as archaeological proof for the formation of a stronghold district corresponding to the territory of Drohiczyn Land of late medieval times (see Pacuski 2019, pp. 26–29). A similar picture was generated by a reconstruction of the medieval Catholic Church parish network and the villages belonging to it (Jusupović 2010). It is important because in western Rus’ the old stronghold territories were used from the second half of the 14th c. to mark out the borders of the dioceses of the Latin Church (Janeczek 2017, pp. 137–139). These findings coincide with the results of a toponomastic analysis with regard to the range of a local network of service settlements (Gałecki 2019).h The presented results of palaeogenetic and anthropological research, and an archaeological analy- sis of settlement development support the idea that the communities setting up cemeteries of graves with stone casing in the Drohiczyn Upland from the second half of the 11th c. constituted a largely foreign population substrate. Their settling in this area could have been linked to the construction of the stronghold in Drohiczyn and the development of its settled hinterland. This would have laid the foundations for a medieval Drohiczyn “volost”, which could have even taken on the political form of a separate principality in the last quarter of the 12th c. Awareness of the cultural distinctness of this population could have survived into the 15th c. and been reflected in Jan Długosz’s account. Translated by Iwona Zych
https://openalex.org/W4206509567	https://www.arsmedica.cl/index.php/MED/article/download/1856/1689	es		Adaptación de una metodología de especificaciones en la calificación y retroalimentación en la enseñanza de cursos de ciencias básicas en una carrera de la salud para mejorar los aprendizajes y la motivación	ARS medica	2,021	cc-by-sa	4,411	Artículo de investigación DOI: doi.org/10.11565/arsmed.v46i4.1856 Adaptación de una metodología de especificaciones en la calificación y retroalimentación en la enseñanza de cursos de ciencias básicas en una carrera de la salud para mejorar los aprendizajes y la motivación Adaptation of a specifications gradings and feedback methodology for basic sciences courses teaching in a health science career to improving learning and motivation Mauricio Cerda1, Karen Gormaz2, Héctor Vega3, Víctor Castañeda2, Diego Varela3, Mariana Casas3, Jocelyn Dunstan4, Milton de la Fuente3, Natasha Kunakov5 Resumen Introducción: la enseñanza de cursos de ciencias básicas en carreras de la salud es un desafío por no estar directa e inmediatamente relacionada con el ámbito profesional. Por otra parte, las condiciones de estrés que ha impuesto el trabajo a distancia requiere de metodologías motivantes, y, que a su vez permitan una evaluación significativa. Objetivos: reportar las adaptaciones metodológicas y los resultados de una adaptación local de la metodología de especificaciones de las calificaciones y retroalimentación del trabajo. Métodos: se aplica una metodología de formación basada en la retroalimentación en el curso de Física para estudiantes de Tecnología Médica (N=106) durante un semestre. Las calificaciones promedio de los estudiantes fueron comparadas con las obtenidas en años anteriores. Para evaluar el desempeño docente se realizaron 2 encuestas a los estudiantes. El cumplimiento de los logros de aprendizaje se midió mediante auto-evaluación (escala likert 1 a 5) al inicio y término de cada uno de los cuatro capítulos. Resultados: las reprobaciones y eliminaciones de estudiantes en el curso fueron menores a años anteriores, siendo las notas significativamente mayores subiendo desde 4,89 a 6,29 (escala de 1 a 7, p<0,001). Los estudiantes se mostraron en un 95% satisfechos con el desempeño docente y finalmente, la auto-evaluación de logros de aprendizaje mostró un aumento promedio de 1 punto. Conclusiones: la metodología de evaluación basada en especificaciones adaptada a dos entregas y con evaluaciones en una escala no-binaria mejoró el rendimiento, los logros de los aprendizajes esperados y la motivación de los estudiantes. Palabras clave: especificaciones; evaluación; ciencias; dominio. Abstract Background: Teaching basic science courses in health careers is a challenge because these courses are not directly linked to professional practice. On the other hand, the stressful conditions imposed by distance work require motivating methodologies and a meaningful evaluation. Objectives: To report the methodological adaptations and the results of a local adaptation of the specifications grading and feedback methodology. Methods: A training methodology based on feedback is applied in the Physics course for Medical Technology students (N = 106) during one semester. We compared the students’ average grades to those obtained in previous years with the same topics. To evaluate the teaching performance, we conducted two student surveys. We measure compliance with learning achievements by self-assessment (Likert scale 1 to 5) at the beginning and end of each of the four chapters. Results: Failures and eliminations of students in the course were lower than previous years, with significantly higher grades from 4.89 to 6.29 (p <0.001). The students were 95% satisfied with the teaching performance, and finally, the self-evaluation of learning achievements showed an average increase of 1 point. Conclusion: The evaluation methodology based on specifications adapted to two deliveries and evaluations on a non-binary scale improved the performance, expected learning achievements, and students’ motivation. Keywords: specification; grading; sciences; learning for mastery. Fecha de envío: 2021-10-27 - Fecha de aprobación: 2021-12-27 (1) Programa de Biología Integrativa, Instituto de Ciencias Biomédicas y Centro de Informática Médica y Telemedicina, Facultad de Medicina, Universidad de Chile. (2) Departamento de Tecnología Médica, Facultad de Medicina, Universidad de Chile. (3) Programa de Fisiología y Biofísica, Instituto de Ciencias Biomédicas, Facultad de Medicina, Universidad de Chile. (4) Iniciativa de Datos e Inteligencia Artificial y Centro de Modelamiento Matemático Facultad de Ciencias Física y Matemáticas, Universidad de Chile. (5) Departamento de Educación en Ciencias de la Salud, Facultad de Medicina, Universidad de Chile. Autor de correspondencia: mauricio.cerda@uchile.cl ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 25 Cerda et al. Introducción En las carreras universitarias relacionadas a la salud, es necesario desarrollar saberes en el ámbito científico general y en temas específicos básicos, por ello, los primeros años tienen cursos de asignaturas tales como química, matemáticas, y física. La enseñanza de estas asignaturas es especialmente desafiante para docentes y alumnos pues no están directamente relacionadas con el ámbito profesional para el cual se están formando. En este contexto, el curso de Física II de la carrera de tecnología médica de la Universidad de Chile, es un curso de primer año, con aproximadamente 100 estudiantes (106 en su edición 2020). Dicho curso posee 5 créditos o 135 horas semestrales con 81 directas y 54 no presenciales. Las horas directas se distribuyen en 2 bloques semanales: 1 hora 30 minutos y 3 horas. El curso está organizado en 4 capítulos: termodinámica, ondas, electromagnetismo, y óptica. Los logros esperados de aprendizaje son: resolución de problemas de física en manera individual y grupal, e Interpretar fenómenos físicos experimentales. Además, posee una actividad final, realizada en conjunto con el curso de matemáticas II que se imparte en paralelo. En el curso participa un equipo de 4 a 6 docentes, 4 ayudantes, y 1 secretaria docente. Al finalizar la edición 2019, la encuesta de satisfacción aplicada a los estudiantes indicó una buena evaluación general, pero una desconexión entre el tipo de preguntas realizadas en las actividades presenciales y las evaluaciones generales de cada capítulo. La metodología de evaluación utilizaba preguntas de alternativas que, si bien agilizaban el proceso, no permitían una real retroalimentación, ni una evaluación del desarrollo de los logros de aprendizaje del curso en los dominios de tecnología en biomedicina, genérico transversal, e investigación. La situación sanitaria durante 2020, obligó a realizar todas las actividades de manera remota, por lo cual, como equipo docente, nos vimos conminados a buscar maneras alternativas de lograr interacción con los estudiantes. Y, paralelamente, lograr un formato de curso que fuera satisfactorio para todos los participantes, docentes y estudiantes, consciente de los reportes de bajo ánimo, y falta de concentración en estudiantes (ELSAM, 2020). Adicionalmente, se incorporó la retroalimentación, que en el ámbito docente ha demostrado ser muy eficiente en el proceso de formar hábitos y actitudes, así como en el aprendizaje de desempeños. Esta es una técnica y se refiere a la entrega de información al estudiante acerca de su desempeño, en cuanto se acerca o aleja de los logros esperados de aprendizaje, todo esto con el propósito de mejorar este desempeño (Ende, 1987). Esta técnica requiere de 2 elementos fundamentales, primero, que el estudiante conozca cuales son los logros de aprendizaje que debe obtener durante el proceso de formación y segundo que se establezca una buena relación docente – estudiante. Considerando lo anterior, se buscó una adaptación al contexto local: con docentes con tiempos acotados, la situación sanitaria, y la reglamentación de la Universidad de Chile. En concreto cada evaluación, incluyendo los trabajos prácticos, ya sea individual o grupal, tuvo un primer momento de evaluación formativa con una instancia de retroalimentación, y luego una evaluación final, calificada, es decir, sumativa. El siguiente trabajo describe los cambios realizados y los resultados obtenidos en el curso de Física II de la carrera de tecnología médica de la Universidad de Chile durante el 2020, reportando las diferencias cuantitativas y cualitativas observadas respecto de versiones anteriores, 2018 y 2019, destacando las lecciones aprendidas para la docencia de asignaturas científicas en carreras de la salud en el contexto chileno del año 2020. Metodología Procedimiento En el curso Física II - 2020 fueron eliminadas todas las evaluaciones tradicionales (certámenes, evaluaciones globales de cada capítulo, examen final), y reemplazadas por desarrollar una tarea semanal. Las tareas consistían en 3 a 4 ejercicios por cada capítulo, más una actividad práctica grupal a realizar en horas no presenciales. Las indicaciones para realizar cada tareas fueron publicadas con 1 mes de anticipación en una guía de trabajo, con 2 fechas de entrega: una entrega formativa, sobre la cual se entregaba retroalimentación individual escrita y posteriormente la entrega final, que era calificada entre 1,0 y 7,0. La calificación en cada uno de los 4 capítulos fue el promedio de las calificaciones de las tareas realizadas, y para asegurar el logro de las Por ello se planteó un cambio en la metodología de evaluación, competencias planificadas, la nota final se calculó como el promedio adaptando los 15 criterios del Specifications Grading (Nilson, 2015), entre ellos evaluar el aprendizaje efectivo y reducir el estrés en estudiantes. Esta metodología plantea la entrega de tareas/trabajos y su evaluación con rúbricas simples (aprueba o no, según las expectativas de logro del curso, que son conocidas) y la posibilidad de que el estudiante pueda volver a entregar su trabajo corregido para una nueva evaluación tantas veces como sea necesario. de los capítulos, siendo condición necesaria que cada capítulo fuera aprobado, es decir, presentara una calificación mayor o igual a 4,0. Se realizó una encuesta voluntaria de desempeño docente intermedia (mitad de semestre) y otra al final del curso. También cada estudiante debía realizar de manera individual y voluntaria, una autoevaluación por logro de aprendizaje en cada capítulo. ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 26 Cerda et al. Tabla 1: Comparación de metodología docente en 2018, 2019 y propuesta en 2020. Aspecto Año 2018, 2019 Año 2020 Modalidad Seminario Presencial con entrega de guía que lleva una pauta de evaluación Virtual sincrónico con entrega de guía que lleva una pauta de evaluación Retroalimentación Sólo se entrega la guía de ejercicios resuelta. Oportunidad de entrega era irregular, promedio 2 semanas. Si, tareas revisadas por ayudante, informes revisados por docentes; en ambos casos se entregaban retroalimentación individual por escrito en 48 horas. Calificación final del curso Promedio de evaluaciones de desarrollo y prácticos y controles. Promedio de calificaciones de capítulos. Calificación capítulo Evaluación Prueba de desarrollo extensa Promedio de actividades de cada capítulo (calificaciones de tareas y prácticos) Examen Prueba de desarrollo extensa, con posible eximición No Análisis estadístico Las tablas de datos se construyeron con Excel. Para realizar comparaciones de medias se utilizó un t-test o el test de Wilcoxon-Mann-Whitney dependiendo de la distribución de los datos. En todos los test se utilizó un p-valor de significancia del 5%. Para realizar los test mencionados se utilizó el paquete estadístico R (Core Team, 2021). A (%) 10 Resultados Desempeño de los estudiantes (2018, 2019 y 2020) En la figura 1 se resume la evolución de las tasas de reprobación, eliminación del curso y el promedio de calificaciones finales. Podemos mencionar que las reprobaciones disminuyeron desde un 7,8% a un 1%, y que las eliminaciones, pese al duro contexto también disminuyeron respecto a 2019 desde un 4% en 2019 a un 1% en 2020. Las notas por otra parte aumentaron de manera sustancial y significativa desde 4,89/7,00 a 6,29/7,00 (p<0,001). B Reprobaciones y eliminaciones Nota final del curso * 7 6 8 5 6 4 4 2 0 6 3 5 3 2 3 1 Reprobaciones 1 1 Eliminaciones 2018 2019 2020 Figura 1: Desempeño de los estudiantes en el curso de Física II. Colores indican años. A Porcentajes de reprobación eliminación (números en barras indican n). B Nota final promedio final del curso. Evaluación del desempeño docente (2020) En el año 2020 durante el desarrollo del curso se realizaron 2 encuestas de desempeño docente, en el intermedio y hacia el final del curso. En ambas encuestas se realizaron las mismas preguntas cuyos resultados resumimos a continuación. En la tabla 2 se muestra el resumen de los resultados. Se puede apreciar que la evaluación de desempeño es mayor en la evaluación final (95%) respecto a la intermedia (88%). Por otra parte, el área de evaluación más baja fue asociada a la dedicación de tiempo (80%) en la encuesta intermedia y la más alta la Gestión del clima del aula (95%). En la encuesta final las evaluaciones por áreas tuvieron una baja dispersión con valores entre 95% a 97%. ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 27 Cerda et al. Tabla 2: Encuestas de desempeño docente 2020. Área Intermedia Final (1 a 4) (1 a 6) Uso de espacios para interacción virtual (3 preguntas) 3,5 (88%) 5,8 (97%) Uso de metodologías participativas y apoyo al trabajo autónomo (4 preguntas) 3,4 (85%) 5,7 (95%) Estrategias de evaluación (4 preguntas) 3,7 (93%) 5,7 (95%) Gestión del clima del aula (2 preguntas) 3,8 (95%) 5,7 (95%) Dedicación de tiempo (1 pregunta) 3,2 (80%) 5,8 (97%) Promedio 3,5 (88%) 5,7 (95%) En la encuesta de desempeño intermedio (N=11) se pidieron señalar tres aspectos positivos a mejorar y 3 aspectos por mejorar. Al analizar estos textos (texto libre) de manera cualitativa, todos los comentarios positivos mencionaron la organización (100%) y con menor cantidad de menciones el clima en el aula (27%), y el tipo de evaluación (18%). Entre los aspectos a mejorar (N=8) se menciona principalmente el tiempo de dedicación (87%) y en menor medida la organización (50%). En la encuesta de desempeño final se pidió indicar aspectos positivos y a mejorar directamente. Los análisis de los aspectos positivos (N=13) muestran que la organización (61%) es lo más destacado, luego el clima en el aula (53%), la metodología de evaluación (15%), y también felicitaciones (23%). En los aspectos a mejorar (N=2) se pide una profundización de un tópico (50%), e incorporar la evaluación con una instancia sincrónica (50%). Adicionalmente, se comparó el desempeño de este equipo docente con los resultados promedios obtenidos por el resto de los cursos de la Facultad de Medicina (FMED) en el mismo periodo, siendo superior al menos en promedio en: • Uso de espacios para interacción virtual (5,8 vs FMED 5,6). • Uso de metodologías participativas y apoyo al trabajo autónomo (5,7 vs FMED 5,5). • Estrategias de evaluación (5,7 vs FMED 5,4). Auto-evaluación de la metodología y cumplimiento de logros de aprendizajes (2020) Durante la versión 2020 del curso Física II se realizaron encuestas de auto-evaluación del aprendizaje voluntarias: la primera en el mes 1 del curso (56/106 respuestas), y luego al inicio y al final de cada capítulo. Capítulo de onda: 57 encuestas iniciales, y 19 finales. Capítulo de electromagnetismo: 14 y 7. Capítulo de óptica: 38 y 17. Dichas encuestas preguntaban a los estudiantes por aspectos metodológicos del curso y por el cumplimiento de los logros de aprendizaje específicos de cada capítulo, al iniciarlo y al terminarlo. Sobre la metodología, se puede apreciar que en general los estudiantes valoraron positivamente la metodología realizada, ver figura 2. Existe un grado de desacuerdo en dos preguntas. La primera “El tiempo dedicado por mi, al curso, está dentro de las horas planificadas” (42/61 en acuerdo y 5/61 en desacuerdo) se puede interpretar como que la modalidad de un trabajo semanal, en un grupo reducido de estudiantes requirió de mayor tiempo semanal, o que al menos tuvieron esa percepción. La segunda pregunta con 2 estudiantes en desacuerdo “Las sesiones sincrónicas han favorecido mi aprendizaje”, resulta interesante pues si bien es un porcentaje menor, se puede interpretar que la retroalimentación asincrónica pudiera reemplazar la interacción en clase con el docente. Figura 2: Valoración de la metodología aplicada en el curso. ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 28 Cerda et al. Se midió cuántos estudiantes indicaron haber efectivamente mejorado en alguno de los logros de aprendizaje propuesto al terminar el capítulo. Es decir, en qué medida el curso, en su conjunto, está favoreciendo su aprendizaje. Con el fin de resumir los resultados, en este análisis tratamos la escala de valoración numéricamente como: Muy en desacuerdo = 1, En desacuerdo = 2, Ni de acuerdo, ni en desacuerdo = 3, De acuerdo = 4, Muy de acuerdo = 5. Luego, si un estudiante indica una valoración del grado de cumplimiento del logro de aprendizaje igual a 1 al inicio del capítulo, y luego igual a 4 al final del capítulo, entonces la mejora es de 3 puntos (4-1). La autoevaluación inicial del promedio del capítulo de ondas fue de 2,98±0,61 y al final 4,11±0,67 (aumento de 1,13±0,73, N=19). La autoevaluación inicial del promedio del capítulo de electromagnetismo fue de 2,76±0,47 y al final 3,96±0,40 (aumento de 1,19±0,48, N=7). La autoevaluación inicial del promedio del capítulo de óptica fue de 3,11±0,73 y al final 4,10±0,76 (aumento de 0,98±1,17, N=17). Los resultados muestran que en general el curso, al menos en términos de auto-evaluación, aumenta, en promedio, en alrededor de 1 punto el nivel de logro de los resultados de aprendizaje. Para resumir el cambio entre el inicio y el término de cada capítulo la Figura 3 ilustra la valoración promedio por estudiante y por capítulo. Auto-evaluación de logros de aprendizaje Auto-evaluación término capítulo (1-5) 5 +2 Ondas Electromagnetismo Óptica +1 +0 2 3 4 En el año 2019 un docente dedicó, en promedio, 5 horas semanales, divididas en 3 horas de docencia directa y 2 horas de planificación (reuniones, preparar material). En el año 2020 estimamos que ese volumen horario aumentó en al menos 1 hora (1/5 o 20%). Esto se logró cubrir principalmente gracias a la colaboración de cuatro ayudantes, que realizan la retroalimentación escrita de cada actividad. Sin embargo, el encargado de curso requirió de un tiempo adicional para coordinar a los ayudantes y manejar correcciones de mayor dificultad, que se estimó en 4 horas semanales (2/5 o 40%). Este tiempo adicional fue requerido para asegurar la calidad de las retroalimentaciones y su entrega oportuna. La metodología propuesta se basa en la confianza en los estudiantes. Si bien se detectaron casos de copias aisladas en las tareas semanales (10 de aproximadamente 2200 entregas o 0,45% de tareas en el semestre), estos fueron sancionados según la reglamentación y se citó a los estudiantes de manera personalizada y sincrónica. Sin embargo, queda la inquietud de hasta qué punto las tareas se realizaban de manera individual y no de manera colectiva. • Termodinámica. Estudio computacional (usando R) sobre proba- 2 1 Cabe señalar que el equipo docente en 2020 estaba compuesto por 6 académicos de diferente formación inicial (físicos, informáticos, bioquímicos), y que conforman una unidad docente con 5 años de experiencia. Esta experiencia interna permitió la flexibilidad para innovar, y dio pie a conversaciones abiertas sobre las inquietudes iniciales. La principal de estas inquietudes fue la dedicación horaria que la metodología realizada requería. En un curso como Física II, se busca desarrollar habilidades científicas en los estudiantes, y realizar experimentos simples y reportarlos de manera clara, es una actividad fundamental en esta línea. Por esto se decidió hacer actividades prácticas en el hogar como parte de cada capítulo: 4 3 Otros aspectos 5 Auto-evaluación inicio capítulo (1-5) Figura 3: Auto-evaluación de los logros de aprendizaje individual por capítulo. Cada punto representa la auto-evaluación de un estudiante. Las líneas discontinuas son referencia cuando no hay mejoras (+0), mejoras de 1 punto (+1) y de 2 puntos (+2). bilidades y conceptos de micro y macro estado (LoPresto, 2010). • Ondas. Estudio de la velocidad de propagación de una onda mecánica en una cuerda mediante la cámara de un teléfono. • Electromagnetismo. Aislamiento electromagnético de un teléfono celular. • Óptica. Construcción de lentes convexos de fabricación casera. En la Figura 4 se muestra un grupo de estudiantes realizando un práctico en su hogar. ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 29 Cerda et al. docente, los estudiantes valoraron muy positivamente la metodología, incluso mejor que otros cursos que son más cercanos al trabajo profesional en una FMED. Se debe recalcar que la adaptación genera trabajo adicional al equipo docente tanto en planificación como ejecución, que debe ser cuidadosamente planificado y comunicado. Tomado en contexto, la adaptación de metodología propuesta es una manera de transitar desde los esquemas tradicionales hacia la metodología de especificaciones y que permite, mediante la práctica, familiarizar y promover el esquema entre los equipos docentes y los estudiantes. Otros aspectos y limitaciones Figura 4: Pasos de una actividad práctica en el hogar. En la figura se muestra la construcción de un lente casero en base a una botella plástica, con estos lentes se exploró el aumento (distancia focal) que generan medios con diferente índice de refracción. Los prácticos y sus informes requieren de experiencia científica, por lo que el equipo decidió que fueran los académicos quienes realizaran la retroalimentación y evaluación, siguiendo la misma dinámica de las tareas semanales realizadas por los ayudantes. Para el equipo docente fue muy gratificante ver la mejora en la calidad de los informes, retroalimentar a los estudiantes y la posterior entrega de versiones corregidas de sus informes. Muestra de ello son las menciones cualitativas en las encuestas intermedias y final del curso (2/11 y 3/13 respectivamente). Discusión Adaptación metodológica La metodología basada en especificaciones y que motiva este trabajo (Nilson, 2015), ya ha sido aplicada satisfactoriamente en dominios del saber tales como: computación (Quintana & Quintana, 2020), veterinaria (Hughes et al., 2020), y nutrición (Pope et al., 2020) como una alternativa metodológica resiliente al trabajo remoto en 2020. La implementación reportada por dichos autores es identificar tareas simples (por ejemplo, asistir a clases, hacer lecturas, entregar ejercicios, entregar informes) y evaluar en una escala binaria (pasa/reprueba) en donde se puede re-enviar cada actividad. La experiencia reportada en este artículo es una adaptación de la metodología de especificaciones, con un número limitado de entregas (2) por actividad y donde las evaluaciones son en una escala (1 a 7). Es decir, es una opción intermedia que permite planificar tanto el volumen como la temporalidad del trabajo realizado, y también entrega evaluaciones en una escala familiar para los estudiantes. Es importante destacar, que ya con la retroalimentación entregada, gran parte de los estudiantes alcanzaban una nota de 7 o muy cercana, lo que explica las notas significativamente mayores del curso. En cuanto al desempeño Se ha reportado en la literatura (Berry et al., 2006) que a nivel universitario el nivel de plagio, en todos sus niveles puede llegar a ser de un 90%. Si bien observamos en el curso de Física II plagio en algunas tareas, esto es claramente inferior (0.45%). Sin embargo, las copias detectadas son copias idénticas y no otras formas parciales de plagio. Futuras adaptaciones deben incorporar esta inquietud del equipo docente, desincentivando el plagio, como el mismo Specifications Grading sugiere (Nilson, 2015). Efectivamente la metodología aplicada muestra mejores resultados que la aplicada en años anteriores, sin embargo, requiere de una mayor dedicación docente. El apoyo brindado por los ayudantes en revisar y dar retroalimentación para las tareas semanales fue parte importante del éxito del curso, de otro modo, el equipo docente se hubiera visto sobrecargado con los plazos. La técnica de retroalimentación, es manejada superficialmente por los docentes/ayudantes, se requiere de una habilitación formal, y estar usándola constantemente para lograr eficiencia en su aplicación. Una limitación importante del trabajo es que en ediciones anteriores del curso no se aplicaron las mismas encuestas, lo que dificulta la comparación. En conclusión, parece relevante que los equipos docentes cohesionados y afiatados pueden lograr soluciones exitosas en tiempos difíciles, como ocurrió con el curso de Física II de la carrera de tecnología médica de la Universidad de Chile durante esta situación sanitaria. Una metodología de formación basada en experiencias externas, puede ser adaptada exitosamente. En este caso, la metodología fue organizar las evaluaciones con dos entregas secuenciales, la primera de ellas con retroalimentación y con evaluaciones en una escala no-binaria, lo cual mejoró el rendimiento general de los estudiantes en un curso de base científica. Esta propuesta permite alcanzar altos niveles de satisfacción de los estudiantes y equipos docentes y una buena auto-evaluación de los logros de aprendizaje. ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 30 Cerda et al. Por ello, el equipo docente cree que el cambio metodológico puede ser mantenido a futuro. Queda por comprobar si los aprendizajes logrados se mantienen en el tiempo. Emanuel E, Wendler D. & Grady C. (2000). What Makes Clinical Research Ethical?. Journal of the American Medical Association 283, 2701-11. Reconocimientos Ende J. (1983). Feedback in clinical medical education. Journal of the American Medical Association 250, 777-81. Agradecemos a todos nuestros estudiantes y ayudantes que participaron y motivaron este trabajo. Todos los autores declaran no tener conflicto de interés en este proyecto. Hughes M, Bertram SM. & Young AM. (2020). Teaching animal behavior online: A primer for the pandemic and beyond. Ethology 127, 14-31. Aspectos éticos LoPresto MC. (2010). A Simple Statistical Thermodynamics Experiment. Physics Teaching 48, 183-185. Se respetaron los principios éticos propuestos por Ezekiel J. Emanuel (Emanuel et al., 2000). La información utilizada en este artículo se manejó de manera totalmente anónima. El análisis fue retrospectivo, por lo que en cada año se aplicó la mejor metodología que el equipo docente podía utilizar para la formación de los profesionales. Referencias Berry P, Thornton B. & Baker R. (2006). Demographics of digital cheating: Who cheats, and what we can do about it. En M. Proceedings of the ninth annual conference of the Southern Association for Information Systems. Davis College of Business. ELSAM. (2020). Accedido en: https://www.uchile.cl/noticias/176250/ el-golpe-de-la-pandemia-en-la-salud-mental-de-las-y-los-estudiantes 8 de octubre de 2021. Nilson LB. (2015). Specifications Grading: Restoring Rigor, Motivating Students, and Saving Faculty Time (1ª edición). Stylus Publishing, United States of America. Pope L, Parker HB & Ultsch S. (2020). Assessment of Specifications Grading in an Undergraduate Dietetics Course, Journal of Nutrition Education and Behavior 52, 439-446. Quintana R. & Quintana C. (2020). When classroom interactions have to go online: the move to specifications grading in a project-based design course, Information and Learning Sciences 121, 525-532. R Core Team. (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ARS MEDICA Revista de Ciencias Médicas Volumen 46 número 4 año 2021 ISSN: 0719-1855 © Dirección de Extensión y Educación Continua, Escuela de Medicina, Pontificia Universidad Católica de Chile. http://arsmedica.cl 31
https://openalex.org/W2798855136	https://europepmc.org/articles/pmc6370796?pdf=render	English	null	Frequency of disturbance alters diversity, function, and underlying assembly mechanisms of complex bacterial communities	bioRxiv (Cold Spring Harbor Laboratory)	2,018	cc-by	10,462	ARTICLE OPEN Frequency of disturbance alters diversity, function, and underlying assembly mechanisms of complex bacterial communities Ezequiel Santillan 1,2, Hari Seshan1,2,4, Florentin Constancias1, Daniela I. Drautz-Moses1 and Stefan zequiel Santillan 1,2, Hari Seshan1,2,4, Florentin Constancias1, Daniela I. Drautz-Moses1 and Stefan Wu Disturbance is known to affect the ecosystem structure, but predicting its outcomes remains elusive. Similarly, community diversity is believed to relate to ecosystem functions, yet the underlying mechanisms are poorly understood. Here, we tested the effect of disturbance on the structure, assembly, and ecosystem function of complex microbial communities within an engineered system. We carried out a microcosm experiment where activated sludge bioreactors operated in daily cycles were subjected to eight different frequency levels of augmentation with a toxic pollutant, from never (undisturbed) to every day (press-disturbed), for 35 days. Microbial communities were assessed by combining distance-based methods, general linear multivariate models, α- diversity indices, and null model analyses on metagenomics and 16S rRNA gene amplicon data. A stronger temporal decrease in α- diversity at the extreme, undisturbed and press-disturbed, ends of the disturbance range led to a hump-backed pattern, with the highest diversity found at intermediate levels of disturbance. Undisturbed and press-disturbed levels displayed the highest community and functional similarity across replicates, suggesting deterministic processes were dominating. The opposite was observed amongst intermediately disturbed levels, indicating stronger stochastic assembly mechanisms. Trade-offs were observed in the ecosystem function between organic carbon removal and both nitriﬁcation and biomass productivity, as well as between diversity and these functions. Hence, not every ecosystem function was favoured by higher community diversity. Our results show that the assessment of changes in diversity, along with the underlying stochastic–deterministic assembly processes, is essential to understanding the impact of disturbance in complex microbial communities. npj Bioﬁlms and Microbiomes (2019) 5:8 ; https://doi.org/10.1038/s41522-019-0079-4 Disturbance is deﬁned in ecology as an event that physically inhibits, injures, or kills some individuals in a community, creating opportunities for other individuals to grow or reproduce.26 When disturbance is long-term or continuous, it is classiﬁed as press- disturbance.27 Disturbance is deemed the main factor inﬂuencing variations in species diversity28 and structuring of ecosystems,27,29 but a clear understanding of its outcomes is lacking.30 Particularly, the intermediate disturbance hypothesis (IDH)31 predicts that diversity should peak at intermediate levels of disturbance due to trade-offs between species’ ability to compete, colonize ecological niches, and tolerate disturbance. www.nature.com/npjbioﬁlms 1Singapore Centre for Environmental Life Sciences Engineering, Nanyang Technological University, Singapore 637551, Singapore; 2Department of Civil and Environmental Engineering, University of California, Davis, CA 95616, USA and 3School of Civil and Environmental Engineering, Nanyang Technological University, Singapore 639798, Singapore Correspondence: Stefan Wuertz (swuertz@ntu.edu.sg) 4Present address: Brown and Caldwell, 9665 Chesapeake Drive, Suite 201, San Diego, CA 92123, USA ARTICLE OPEN Frequency of disturbance alters diversity, function, and underlying assembly mechanisms of complex bacterial communities The IDH has been inﬂuential in ecological theory, as well as in management and conservation,32 but its predictions do not always hold true.28,33 For example, in soil and freshwater bacterial communities, different patterns of diversity were observed with increasing disturbance frequency with biomass destruction34 and removal35 as disturbance type, respectively. Meanwhile, the effect of varying frequencies of non- destructive disturbances on bacterial diversity remains unknown. Furthermore, the IDH predicts a pattern but it is not a coexistence mechanism as it was originally purported to be.36 Hence, its relevance is being debated37,38 with multiple interpretations and simplicity as the main points of critique. To date, the mechanisms Published in partnership with Nanyang Technological University Received: 7 May 2018 Accepted: 11 December 2018 Ecosystem function dynamics and trade-offs Ecosystem function dynamics and trade-offs The undisturbed community (L0) was the only one with complete dissolved organic carbon (COD) removal and nitrate generation without nitrite residuals, while the press-disturbed community (L7) was the only one where nitriﬁcation products were never detected and also had the lowest biomass (Fig. 2). Initially, reactors at the disturbed levels showed an inability to remove all of the 3-CA (with the exception of L1). Such lack of 3-CA degradation was accompanied by a reduction in organic carbon removal in the ﬁrst 3 weeks (Fig. 2a, Supplementary Figure 2a,c,e), and a complete inhibition of nitriﬁcation with subsequent accumulation of ammonium (Fig. 2b, Supplementary Figure 2b,d,f). Removal of 3- CA recovered and was above 95% for all disturbed levels after 28 days (Supplementary Figure 2g), but COD removal was still not 100% despite complete 3-CA removal towards the end of the experiment (Fig. 2c). p y g We analysed changes in the ecosystem function over time by measuring the removal of organic carbon, ammonia, and 3-CA, as well as biomass. Changes in community structure were examined at different levels of resolution using a combination of metage- nomics sequencing and 16S rRNA gene ﬁngerprinting techniques. Such changes were assessed by employing a combination of ordination tools, diversity indices, cluster analysis, univariate and multivariate statistical analyses. We also explored how diversity was related to function, focusing on trade-offs. Furthermore, the role of stochasticity in community assembly was investigated by employing null model techniques from ecology. We hypothesized that time would lead to a decrease in α-diversity at the extreme sides of the disturbance range due to deterministic adaptation to the environment, while less predictable conditions at intermediate disturbance levels would lead to higher α-diversity and stochastic assembly. Consequently, replicates at intermediately disturbed levels should display higher variability in terms of both community structure and function, compared with the ones at the extreme sides of the disturbance range where the opposite (that is, less variability) should occur. Nitriﬁcation was detected on day 21 for L1 and later for other disturbance levels, except for the press-disturbed L7. The dominant NOX component was nitrite, but some nitrate was also produced (Fig. 2d, Supplementary Figure 2h–j). At the end of the study, there was a signiﬁcant negative Spearman’s correlation between organic carbon removal and nitriﬁcation (Supplementary Figure 3a-b) in terms of nitrite (ρ = −0.901) and nitrate production (ρ = −0.697). INTRODUCTION Disturbance is deﬁned in ecology as an event that physically inhibits, injures, or kills some individuals in a community, creating opportunities for other individuals to grow or reproduce.26 When disturbance is long-term or continuous, it is classiﬁed as press- disturbance.27 Disturbance is deemed the main factor inﬂuencing variations in species diversity28 and structuring of ecosystems,27,29 INTRODUCTION Understanding what drives patterns of community succession and structure remains a central goal in ecology1,2 and microbial ecology,3 especially since community diversity and assembly are thought to regulate the ecosystem function.4,5 Assembly pro- cesses can be either stochastic, assuming that all species have equal ﬁtness and that changes in structure arise from random events of ecological drift,6 or deterministic, when communities form as a result of niche diversity shaped by abiotic and biotic factors.7 Deterministic and stochastic assembly dynamics have been proposed to simultaneously act in driving assembly patterns observed in nature.8–12 This has stimulated scientiﬁc discourse including modelling of experimental data13–16 and both observa- tional and manipulative experimentation in a variety of ecosys- tems, like deserts on a global scale,17 groundwater,18 subsurface environments,2,19,20 soil plant–fungi associations,21 rock pools,22 water ponds,23 and sludge bioreactors.15,24,25 These prior studies emphasized the need to understand what governs the relative balance between stochastic and deterministic processes and what conditions would lead to stochastic processes overwhelming deterministic processes, particularly under disturbance.20 To investigate their roles, well-replicated time series experiments are needed.18,25 v t t w Received: 7 May 2018 Accepted: 11 December 2018 1Singapore Centre for Environmental Life Sciences Engineering, Nanyang Technological Engineering, University of California, Davis, CA 95616, USA and 3School of Civil and Environm Correspondence: Stefan Wuertz (swuertz@ntu.edu.sg) 4Present address: Brown and Caldwell, 9665 Chesapeake Drive, Suite 201, San Diego, CA 9 Published in partnership with Nanyang Technological University Understanding what drives patterns of communi structure remains a central goal in ecology1 ecology,3 especially since community diversity a thought to regulate the ecosystem function.4, cesses can be either stochastic, assuming that equal ﬁtness and that changes in structure ar events of ecological drift,6 or deterministic, wh form as a result of niche diversity shaped by a factors.7 Deterministic and stochastic assembly been proposed to simultaneously act in driving a observed in nature.8–12 This has stimulated sc including modelling of experimental data13–16 a tional and manipulative experimentation in a v tems, like deserts on a global scale,17 groundwa environments,2,19,20 soil plant–fungi association water ponds,23 and sludge bioreactors.15,24,25 Th emphasized the need to understand what gov balance between stochastic and deterministic pro conditions would lead to stochastic processe deterministic processes, particularly under d investigate their roles, well-replicated time se are needed.18,25 Received: 7 May 2018 Accepted: 11 December 20 1Singapore Centre for Environmental Life Sciences Engineeri Engineering, University of California, Davis, CA 95616, USA and Correspondence: Stefan Wuertz (swuertz@ntu.edu.sg) 4Present address: Brown and Caldwell, 9665 Chesapeake Drive Published in partnership with Nanyang Technological Univ Understanding what drives patterns of community succession and structure remains a central goal in ecology1,2 and microbial ecology,3 especially since community diversity and assembly are thought to regulate the ecosystem function.4,5 Assembly pro- cesses can be either stochastic, assuming that all species have equal ﬁtness and that changes in structure arise from random events of ecological drift,6 or deterministic, when communities form as a result of niche diversity shaped by abiotic and biotic factors.7 Deterministic and stochastic assembly dynamics have been proposed to simultaneously act in driving assembly patterns observed in nature.8–12 This has stimulated scientiﬁc discourse including modelling of experimental data13–16 and both observa- tional and manipulative experimentation in a variety of ecosys- tems, like deserts on a global scale,17 groundwater,18 subsurface environments,2,19,20 soil plant–fungi associations,21 rock pools,22 water ponds,23 and sludge bioreactors.15,24,25 These prior studies emphasized the need to understand what governs the relative balance between stochastic and deterministic processes and what conditions would lead to stochastic processes overwhelming deterministic processes, particularly under disturbance.20 To investigate their roles, well-replicated time series experiments are needed.18,25 Received: 7 May 2018 Accepted: 11 December 2018 Published in partnership with Nanyang Technological University Published in partnership with Nanyang Technological University E. INTRODUCTION Santillan et al. 2 Fig. 1 Microbial community dynamics across disturbance frequencies and time, as assessed by 16S rRNA gene terminal restriction fragment length polymorphism (T-RFLP) ﬁngerprinting. a Canonical analysis of principal coordinates (CAP, constrained ordination) plot, with disturbance levels as differentiation criteria, shows cluster differentiation for L0 (CAP1 axis) and L7 (CAP2 axis) from intermediately disturbed levels (L1–6). Disturbance levels: L0 [light-green triangles], L1 [blue upside-down triangles], L2 [light-blue open squares], L3 [open red rhombuses], L4 [purple circles], L5 [black crosses], L6 [green x-symbols], and L7 [blue stars]. b Non-metric multidimensional scaling (NMDS, unconstrained ordination) shows temporal dispersion effect. Days: 14 [open triangles], 21 [light-grey upside-down triangles], 28 [dark-grey squares], and 35 [black rhombuses] Fig. 1 Microbial community dynamics across disturbance frequencies and time, as assessed by 16S rRNA gene terminal restriction fragment length polymorphism (T-RFLP) ﬁngerprinting. a Canonical analysis of principal coordinates (CAP, constrained ordination) plot, with disturbance levels as differentiation criteria, shows cluster differentiation for L0 (CAP1 axis) and L7 (CAP2 axis) from intermediately disturbed levels (L1–6). Disturbance levels: L0 [light-green triangles], L1 [blue upside-down triangles], L2 [light-blue open squares], L3 [open red rhombuses], L4 [purple circles], L5 [black crosses], L6 [green x-symbols], and L7 [blue stars]. b Non-metric multidimensional scaling (NMDS, unconstrained ordination) shows temporal dispersion effect. Days: 14 [open triangles], 21 [light-grey upside-down triangles], 28 [dark-grey squares], and 35 [black rhombuses] Published in partnership with Nanyang Technological University RESULTS AND DISCUSSION behind the observed patterns of diversity under disturbance remain to be elucidated.39,40 Overall community dynamics and differentiation of clusters Overall community dynamics and differentiation of clusters Bacterial community structure displayed temporal changes and varied between disturbance levels, as assessed by 16S rRNA gene terminal restriction fragment length polymorphism (T-RFLP) (Fig. 1). Constrained ordination showed a deﬁned cluster separation with 0% misclassiﬁcation error of the outermost levels L0 and L7 from the remaining intermediate levels L1–6 (Fig. 1a). Overall community structure differed over time with a dispersion effect after 14 days (Fig. 1b). Levels across disturbance and time factors showed signiﬁcant differences (PERMANOVA P = 0.003, Supplementary Table 1), with a non-signiﬁcant interaction effect (P = 0.15). Disturbance was the factor responsible for the observed clustering (Fig. 1a) and not heteroscedasticity (PERDIMSP P = 0.35). The objective of this work was to test the effect of disturbance on the bacterial community structure, diversity, and ecosystem function of a complex bacterial system, with emphasis on the underlying assembly mechanisms. We employed sequencing batch bioreactors inoculated with activated sludge from an urban wastewater treatment plant, in a laboratory microcosm setup with eight different frequency levels of augmentation with toxic 3- chloroaniline (3-CA) as disturbance. Triplicate reactors received 3- CA either never (L0, undisturbed), every 7, 6, 5, 4, 3, and 2 days (L1–6, intermediately-disturbed), or every day (L7, press-disturbed) for 35 days. Chloroanilines are toxic and carcinogenic compounds and few bacteria encode the pathways to degrade 3-CA,41 which is also known to inhibit both organic carbon removal and nitriﬁcation in sludge reactors.42 Microcosm studies are useful models of natural systems,43 can be coupled with theory development to stimulate further research,44 and by permitting easier manipulation and replication can allow inference of causal relationships45 and statistically signiﬁcant results.46 Ecosystem function dynamics and trade-offs A Mantel test on Bray–Curtis distance matrixes for both datasets (n = 24) yielded signiﬁcant similarity (r = 0.73, P = 0.002). Procrustes tests of comparisons within ordination methods of PCO (Fig. 3c) and NMDS also yielded signiﬁcant similarities for both datasets (P = 0.002, Supplementary Table 2). Multivariate PERMANOVA tests on the metagenomics dataset produced statistically signiﬁcant results, but with sig- niﬁcant heteroscedasticity as shown by PERMDISP (Supplementary Table 1). We resolved these mean–variance relationship concerns by running a general linear multivariate models (GLMMs) test to ﬁt the data to a negative binomial distribution. Both residuals vs ﬁtted and mean–variance plots supported the choice of a negative binomial distribution for the regression model (Supplementary Figure 5). The analysis of deviance of the regression rejected the null hypothesis of no difference between communities at different disturbance levels, independent of heteroscedasticity (P = 0.0149). Fig. 2 Process performance indicators across disturbance levels. Effects include temporal changes and trade-offs in community function. a, c Percentage of organic carbon as chemical oxygen demand (COD, black circles) and 3-CA (open purple rhombuses) removal for all levels (negative values represent accumulation). c Biomass as volatile suspended solids (VSS, open green squares). b, d Concentration of ammonium (black rhombuses), nitrite (open blue triangles), and nitrate (open red circles) as nitrogen for all levels. Data are from days 7 (a, b) and 35 (c, d) of the study (for all time points sampled, see Supplementary Figure 2). Mean ± s.d. (n = 3) are shown. Undisturbed L0 replicates had consistent organic carbon removal and complete nitriﬁcation, whereas press-disturbed L7 never showed nitriﬁcation and had the lowest ﬁnal biomass. Intermediate levels L1–6 displayed changing functionality with higher s.d. values that increased over time Fig. 3 Community dissimilarity assessed by principal coordinates analysis (PCO) plots for all disturbance levels on T-RFLP datasets on days a 14 and b 35 of the study. Ovals with dashed lines represent 80% similarity calculated by group average clustering. Disturbance levels: L0 [light-green triangles], L1 [blue upside-down triangles], L2 [light-blue open squares], L3 [open red rhombuses], L4 [purple circles], L5 [black crosses], L6 [green x-symbols], and L7 [blue stars]. c Procrustes analysis on PCO at day 35 comparing metagenomics (circles) and T-RFLP (triangles) datasets. Lines unite data points from the same reactor (n = 24). Same colour palette as for disturbance levels. Tests comparing both methods were statistically signiﬁcant (Supplementary Table 2). Ecosystem function dynamics and trade-offs Biomass values on day 35 differed signiﬁcantly among levels with the highest value at L1 and the lowest at L7 Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 E. Santillan et al. 3 Intermediate levels of disturbance displayed increased dissimilarity with time (Fig. 2c). There was a signiﬁcant positive correlation between biomass and nitriﬁcation in terms of nitrite (ρ = 0.466) and nitrate production (ρ = 0.656) (Supplementary Figure 3c-d). To distinguish the effect of disturbance from temporal community dynamics (Fig. 1), community dissimilarity was assessed on the T-RFLP dataset at each time point by ordination analysis using principal coordinates analysis (PCO) (Fig. 3a, b), non-metric multidimensional scaling (NMDS), and canonical analysis of principal coordinates (CAP) with cluster similarity analysis (Supplementary Figure 4). The combination of constrained and unconstrained ordination methods allowed differentiating loca- tion from dispersion effects in community structure.47 L0 was consistently different in all ordination plots and L7 differed after 21 days, both with 0% misclassiﬁcation error at all time points for CAP plots. Dispersion effects within intermediate levels were evident in the unconstrained ordination plots with higher differentiation of biological replicates after 35 days (Fig. 3b), coinciding with the production of nitrite and low levels of nitrate (Fig. 2d). Community differentiation was statistically signiﬁcant from day 21 onwards as supported by PERMANOVA and PERMDISP (Supplementary Table 1). Fig. 2 Process performance indicators across disturbance levels. Effects include temporal changes and trade-offs in community function. a, c Percentage of organic carbon as chemical oxygen demand (COD, black circles) and 3-CA (open purple rhombuses) removal for all levels (negative values represent accumulation). c Biomass as volatile suspended solids (VSS, open green squares). b, d Concentration of ammonium (black rhombuses), nitrite (open blue triangles), and nitrate (open red circles) as nitrogen for all levels. Data are from days 7 (a, b) and 35 (c, d) of the study (for all time points sampled, see Supplementary Figure 2). Mean ± s.d. (n = 3) are shown. Undisturbed L0 replicates had consistent organic carbon removal and complete nitriﬁcation, whereas press-disturbed L7 never showed nitriﬁcation and had the lowest ﬁnal biomass. Intermediate levels L1–6 displayed changing functionality with higher s.d. values that increased over time Metagenomics community analysis validates observations from ﬁngerprint dataset g p β-Diversity patterns observed from 16S rRNA gene amplicon T-RFLP data on day 35 were signiﬁcantly similar to those from shotgun metagenomics data. Published in partnership with Nanyang Technological University Ecosystem function dynamics and trade-offs The undisturbed (L0) and press-disturbed (L7) levels were distinct from each other as well as from the remaining intermediate levels, as supported by multivariate tests (both distance-based and GLMMs). The ordination plots and cluster analyses showed a clear separate clustering for the independent replicates of these two disturbance levels along the experiment, and particularly the constrained ordination plots displayed this with 0% misclassiﬁcation error. Furthermore, the ecosystem function was clearly differentiated between L0 and L7, as well as being consistent across replicates at each level. We contend that the observed clustering is an indication that both the undisturbed and press-disturbed levels favoured deterministic assembly mechanisms, where the selective pressure due to unaltered succession (L0) or sustained toxic-stress (L7) promoted species sorting, resulting in similar community structuring among biological replicates over the course of the experiment. The observed patterns in α-diversity were time-dependent, as diversity decreased over time with respect to the initial sludge inoculum (Fig. 4a, T-RFLP dataset). Such a temporal decrease in diversity was higher at the extreme ends of the disturbance range, resulting in a parabolic pattern on day 35 (Fig. 4b, c). The ﬁnal α-diversity pattern based on Hill number 2D was similar for both T-RFLP and metagenomics methods (Fig. 4b), although the latter showed higher variability. For the metagenomics dataset we also calculated the lower-order Hill numbers (0D, 1D) which give higher weight to less abundant operational taxonomic units (OTUs). They displayed the same parabolic pattern (Fig. 4c). Welch’s ANOVA tests were statistically signiﬁcant for all Hill numbers (P < 0.01, P = 0.022 for 2Dmetagenomics). Additionally, there were strong signiﬁcant correlations between α-diversity and ecosystem function (Supple- mentary Figure 6), focusing on the more robust estimators of microbial diversity 1D and 2D.48 Both 1D and 2D correlated positively with ammonia removal and nitrite generation (Supple- mentary Figure 6a-b), while 2D had a positive correlation with biomass (Supplementary Figure 6c) but a negative correlation with organic carbon removal (Supplementary Figure 6d). g p p Conversely, the communities from intermediately disturbed levels (L1–6) did not form distinct clusters for any particular level through the experiment. Within-treatment dissimilarity among replicates increased over time, with some replicates being more similar to those of other intermediate levels. Concurrently, ecosystem function parameters also displayed within-treatment variability for L1–6. Ecosystem function dynamics and trade-offs Intermediate treatments’ (L1–6) within-treatment dissimilarity increased with time. L0 and L7 clusters consistently displayed higher similarity after 14 days Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 Fig. 3 Community dissimilarity assessed by principal coordinates analysis (PCO) plots for all disturbance levels on T-RFLP datasets on days a 14 and b 35 of the study. Ovals with dashed lines represent 80% similarity calculated by group average clustering. Disturbance levels: L0 [light-green triangles], L1 [blue upside-down triangles], L2 [light-blue open squares], L3 [open red rhombuses], L4 [purple circles], L5 [black crosses], L6 [green x-symbols], and L7 [blue stars]. c Procrustes analysis on PCO at day 35 comparing metagenomics (circles) and T-RFLP (triangles) datasets. Lines unite data points from the same reactor (n = 24). Same colour palette as for disturbance levels. Tests comparing both methods were statistically signiﬁcant (Supplementary Table 2). Intermediate treatments’ (L1–6) within-treatment dissimilarity increased with time. L0 and L7 clusters consistently displayed higher similarity after 14 days npj Bioﬁlms and Microbiomes (2019) 8 Published in partnership with Nanyang Technological University E. Santillan et al. 4 Fig. 4 α-Diversity patterns. a Temporal dynamics of Hill number 2D for abundant OTUs, calculated from T-RFLP data across disturbance levels. b Hill number 2D calculated from T-RFLP (black dashed bars) and metagenomics (grey solid bars) data at days 0 (seed) and 35 (disturbance levels L0–L7). c Hill numbers 0D (black solid bars) and 1D (blue solid bars) from metagenomics data on days 0 (seed) and 35 (L0–L7). Values represent mean ± s.d. (n = 3). Characters above bars indicate Games–Howell post-hoc grouping Fig. 4 α-Diversity patterns. a Temporal dynamics of Hill number 2D for abundant OTUs, calculated from T-RFLP data across disturbance levels. b Hill number 2D calculated from T-RFLP (black dashed bars) and metagenomics (grey solid bars) data at days 0 (seed) and 35 (disturbance levels L0–L7). c Hill numbers 0D (black solid bars) and 1D (blue solid bars) from metagenomics data on days 0 (seed) and 35 (L0–L7). Values represent mean ± s.d. (n = 3). Characters above bars indicate Games–Howell post-hoc grouping Higher α-diversity for intermediately disturbed treatments and diversity-function correlations Deterministic and stochastic patterns of assembly amongst different disturbance levels Niche-structuring at both ends of the disturbance frequency range was suggested by community structure patterns and ecosystem function. Ecosystem function dynamics and trade-offs For example, the conversion of ammonia to NOX products, which was initially hampered when communities were still adapting to degrade 3-CA, was not the same across all equally handled independent replicates. The observed divergence across independent replicates is considered here as a strong indicator of stochasticity in community assembly. Additionally, the lower deviation for L2–L5 from expected β-diversity values estimated via null model analysis indicates a higher role of stochasticity at intermediate disturbance levels. Several processes might be promoting stochastic assembly, like strong feedback processes51 that are linked to density dependence and species interactions,52 priority effects,53 and ecological drift.54 Reactors within this study were designed as closed systems, hence stochastic dispersal processes55 could not affect community assembly. Null model analysis suggests different assembly mechanisms across disturbance frequencies Null model analysis suggests different assembly mechanisms across disturbance frequencies Published in partnership with Nanyang Technological University Null model analysis suggests different assembly mechanisms across disturbance frequencies Overall, stochasticity was stronger for intermediate disturbance levels L2–L5 and also increased with respect to the sludge inoculum through null model analysis that both deterministic and stochastic processes played important roles in controlling community assembly and succession, but their relative importance was time-dependent. The greater role of stochasticity we found on day 35 concurred with higher observed variability in the ecosystem function and structure among replicates for intermediately-disturbed levels. Likewise, previous work on fresh- water ponds tracking changes in producers and animals49 found β-diversity (in terms of dissimilarity) increasing with stochastic processes. These observed patterns are also in accordance with ecological studies proposing deterministic and stochastic pro- cesses balancing each other to allow coexistence,10 with commu- nities exhibiting variations in the strength of stabilization mechanisms and the degree of ﬁtness equivalence among species.9 Thus, it is not sufﬁcient to ask whether communities mirror either stochastic or deterministic processes,8 but also necessary to investigate the combination of such mechanisms that in turn explain the observed community structures along a continuum.9 (sewage to complex synthetic media), immigration rates (open to closed system), and mean cell residence time (low to high). This was a succession scenario in which communities had to adapt to such changes along with the designed disturbance array. For L0 and L7, 2D decreased over time in agreement with deterministically-dominated processes, probably because such levels represented the most predictable environments within our disturbance range. In contrast, intermediate levels either increased or maintained the same 2D over time (after an initial decrease within the ﬁrst 2 weeks), seemingly a case where niche overlap promoted stochastic assembly.8 The emergence of an IDH pattern after time is coherent with ﬁndings in previous microcosm studies using synthetic communities of protists57 and freshwater enrich- ment microbial communities.35 Yet, none of these studies evaluated the relative importance of the underlying assembly mechanisms for the observed diversity dynamics. 1 2 y y Additionally, both 1D and 2D were positively correlated with nitriﬁcation and productivity, suggesting that higher community evenness favours functionality under selective pressure,58 but were negatively correlated with organic carbon removal. Thus, we cannot afﬁrm that more diverse communities have better functionality without considering trade-offs. This supports the notion that higher α-diversity does not necessarily imply a “better” or “healthier” system.56 In addition to the observed changes in OTU diversity, there was an evident variation in ecosystem function along the disturbance range studied (Fig. Null model analysis suggests different assembly mechanisms across disturbance frequencies 2c, d), a similar ﬁnding to that of previous studies with simpler planktonic communities.59 Diversity–disturbance patterns and trade-offs with function Diversity disturbance patterns and trade offs with function We observed the highest α-diversity at intermediate levels as predicted by the IDH,31 both in terms of composition (0D) and abundances (1D, 2D). This ﬁnding is non-trivial in two aspects. First, Svensson et al.32 have shown that most studies ﬁnd support for the IDH by using species richness (0D) rather than evenness or other abundance-related indices (like 1D and 2D). They suggested that low evenness at high disturbance levels could be caused by the dominance of a few disturbance specialists. Second, the use of richness for microbial communities is not reliable48 since it is heavily constrained by the method of measurement,56 which makes it hard to compare results from different studies using this metric. Additionally, for complex communities there is often a huge difference between the abundance of rare and abundant taxa. Hence, for microbial systems, it is reasonable to assess diversity in terms of more robust compound indices rather than richness, the reason why we focused on 1D and 2D for diversity- function analyses. Functional trade-offs are expected under disturbance since organisms need to allocate resources normally used for other functions to recover after a disturbance.60 In our study, commu- nities with higher biomass had lower organic carbon removal efﬁciencies, which together with the trade-offs described for nitriﬁcation, suggest the adoption of different community life- history strategies depending on the frequency of disturbance. The results presented here were all taxonomy-independent since our focus was on diversity, function, and mechanisms of community assembly (phylum-level community changes are provided as supplemental material Supplementary Figure 7). Taxonomy- independent approaches continue to be useful to describe diversity patterns and mechanisms of community assembly.2,61 However, it has been proposed that species’ traits can predict the effects of disturbance and productivity on diversity.62 Hence, further analysis of the different taxa and their genetic potential paired with the observed trade-offs in ecosystem function will aid y Importantly, the observed pattern in α-diversity was time- dependent and resulted in an IDH pattern after 35 days. Temporal dynamics were expected since the sludge community experi- enced an initial perturbation in all reactors after transfer from a wastewater treatment plant to our microcosm arrangement. Null model analysis suggests different assembly mechanisms across disturbance frequencies To test if the observed changes in β-diversity (Figs 1a and 3, Supplementary Figure 4) were due to variations in the underlying stochastic and deterministic mechanisms or due to changes in α- and γ-diversity ratios (α:γ) alone,49 we employed a null model analysis from Kraft et al.50 on the bacterial genus-level metage- nomics datasets on day 0 and day 35. The model estimated null β- diversity values after randomizing the location of each individual within the three independent reactors for each of the eight disturbance treatment levels, while keeping the total quantity of individuals per reactor, the relative abundance of each OTU, and the γ-diversity constant over 10,000 iterations. Under this model, stochastic assembly mechanisms were found to be higher for some intermediately disturbed levels (L2–L5) in terms of stochastic intensity (SI) and standard effect size (SES) values, which corresponded to communities less deviant from the null expecta- tion (Fig. 5). SI was also higher at d35 with respect to the sludge inoculum (d0). y We argue that there were different underlying stochastic–deterministic mechanisms operating in the resulting community assembly along the disturbance range of our study. Similarly, a study on groundwater microbial communities18 found npj Bioﬁlms and Microbiomes (2019) 8 E. Santillan et al. Fig. 5 Inﬂuence of stochastic assembly mechanisms in bacterial communities as assessed by a stochastic intensity and b standard effect size (SES). Both metrics were calculated through null model analysis on the metagenomics genus-level dataset at days 0 (seed) and 35 (disturbance levels L0–L7). Each calculation involved all replicates of each treatment (nseed = 2, nL0–L7 = 3) evaluated over 10,000 null model iterations. SES values closer to zero represent communities less deviant from the null expectation, implying stronger stochastic assembly processes. Overall, stochasticity was stronger for intermediate disturbance levels L2–L5 and also increased with respect to the sludge inoculum E. Santillan et al. 5 Fig. 5 Inﬂuence of stochastic assembly mechanisms in bacterial communities as assessed by a stochastic intensity and b standard effect size (SES). Both metrics were calculated through null model analysis on the metagenomics genus-level dataset at days 0 (seed) and 35 (disturbance levels L0–L7). Each calculation involved all replicates of each treatment (nseed = 2, nL0–L7 = 3) evaluated over 10,000 null model iterations. SES values closer to zero represent communities less deviant from the null expectation, implying stronger stochastic assembly processes. Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 Null model analysis suggests different assembly mechanisms across disturbance frequencies 6) and contend that it should h compound α-diversity indices,48 since the u mechanisms would affect taxa abundance dis The ISH can be further portrayed by notin behind the IDH, namely, that a competition–co would lead to higher diversity at inter disturbance.31 In the context of our study, Fig. 6 Intermediate stochasticity hypothesis ( assembly under varying disturbances. Conceptu the classic relationship between α-diversity including the effect of underlying stochastic processes driving bacterial community assemb ate disturbance regimes result in less predic specialized traits would be less advantageou stochastic equalization of competitive advant higher α-diversity. On the contrary, extreme end conditions are recurrent would select for adapt dominance would result in a lower α-diversity 6 6 Fig. 6 Intermediate stochasticity hypothesis (ISH) for community assembly under varying disturbances. Conceptual representation of the classic relationship between α-diversity and disturbance,31 including the effect of underlying stochastic and deterministic processes driving bacterial community assembly. When intermedi- ate disturbance regimes result in less predictable environments, specialized traits would be less advantageous to taxa, and the stochastic equalization of competitive advantages would lead to higher α-diversity. On the contrary, extreme ends of the range where conditions are recurrent would select for adapted organisms whose dominance would result in a lower α-diversity 6 frequencies reduced the competitive ability of previously domi- nant taxa. Stochastic mechanisms of ecological drift could then play a critical role in shaping the emerging structure of microbial communities3 where random processes of birth, death, and reproduction can have an effect on which of these low abundance taxa will be more beneﬁted as a result of intermediate disturbance frequencies. Drift could also lead to historical contingency and priority effects that are also stochastic,53 where taxa that occupy early the disturbance-opened niches could reduce the availability of resources to other taxa whose abundance will then be limited. Such reasoning could explain why, while higher α-diversity was found at intermediate levels of disturbance in our study, community structure and ecosystem function differed across identically treated replicates. Implications and concluding remarks The implications of this study relate to both process engineering and environmental management. Sludge communities within wastewater treatment are not only model systems in microbial ecology,65 but also a key driver for water sanitation and the environmental impact of anthropogenic water discharges.66 Disturbances could promote stochastic assemblages of the sludge communities, which despite harbouring higher diversity could lead to variable overall ecosystem function. npj Bioﬁlms and Microbiomes (2019) 8 Null model analysis suggests different assembly mechanisms across disturbance frequencies This could be the reason why after similar perturbations the process outcome differs, causing operational problems for water utilities.67 Further- more, cases where disturbance temporally favours stochastic assembly could lead to a different ﬁnal community after the perturbation,27 which could compromise the expected ecosystem function. More research is needed to identify such scenarios in practice. We described how different frequencies of disturbance affected ecosystem function and bacterial community diversity and assembly in a closed microcosm bioreactors system. Communities were assessed through different molecular methods that none- theless yielded very similar patterns. Furthermore, besides the wastewater treatment microbial community, other complex microbial systems (e.g., the gut microbiome) might display similar responses to disturbance. We argue that changes not only in diversity but also in the underlying deterministic–stochastic assembly mechanisms should be evaluated in studies of the effects of disturbance on such systems. For such an assessment, both replication and wide-enough disturbance ranges are key. Additionally, the ISH could be evaluated within open systems to include the effect of dispersal processes. This calls for more studies in microcosm45,68 and mesocosm settings, as well as meta-analysis from full-scale application studies. in the understanding of potential life-history strategies60 and their relationship with community aggregated traits63 in the near future. Merging mechanisms of community assembly and alpha-diversity patterns: an intermediate stochasticity hypothesis Knowing that the validity of the IDH is still under debate37,38 and that many different diversity–disturbance patterns have been reported,28,30,33 we asked whether there is a relationship between the peaked pattern in diversity observed and the underlying stochastic–deterministic processes of community assembly. Under purely stochastic processes, diversity should vary randomly as all species have equal ﬁtness,55 unless some other mechanism acts to prevent this. It is recognized that, beyond empirical pattern description, an understanding of the underlying mechanisms is necessary to comprehend the outcomes of intermediate dis- turbance regimes.30,40,64 We hypothesize that higher α-diversity at intermediate disturbance frequencies is the result of weaker stabilizing mechanisms (niches), which are stronger at extreme ends of the disturbance range. Stochastic mechanisms will produce even assemblages (higher α-diversity) at intermediately disturbed levels, whilst infrequent or too-frequent disturbances will favour some species over others (lower α-diversity). We propose this idea as the intermediate stochasticity hypothesis (ISH, Fig. 6) and contend that it should hold particularly for compound α-diversity indices,48 since the underlying assembly mechanisms would affect taxa abundance distributions. Published in partnership with Nanyang Technological University Null model analysis suggests different assembly mechanisms across disturbance frequencies For the sludge inoculum, this implied changes in reactor volume, frequency of feeding (continuous to batch), type of feeding in the understanding of potential l relationship with community ag future. Merging mechanisms of communi patterns: an intermediate stochast Knowing that the validity of the ID that many different diversity–dis reported,28,30,33 we asked whether the peaked pattern in diversity stochastic–deterministic processes purely stochastic processes, divers species have equal ﬁtness,55 unless prevent this. It is recognized th description, an understanding of necessary to comprehend the o turbance regimes.30,40,64 We hypot intermediate disturbance frequen stabilizing mechanisms (niches), w ends of the disturbance range produce even assemblages (highe disturbed levels, whilst infrequen will favour some species over propose this idea as the interm (ISH, Fig. 6) and contend that compound α-diversity indices,48 s mechanisms would affect taxa abu The ISH can be further portray behind the IDH, namely, that a com would lead to higher diversit disturbance.31 In the context of closed system, colonization would taxa that have an opportunity to g Fig. 6 Intermediate stochasticity h assembly under varying disturbanc the classic relationship between including the effect of underlyin processes driving bacterial commu ate disturbance regimes result in specialized traits would be less a stochastic equalization of compet higher α-diversity. On the contrary, e conditions are recurrent would sele dominance would result in a lower 6 npj Bioﬁlms and Microbiomes (2019) 8 E. Santillan et al. in the understanding of potential life-history st relationship with community aggregated t future. Merging mechanisms of community assembly patterns: an intermediate stochasticity hypoth Knowing that the validity of the IDH is still un that many different diversity–disturbance p reported,28,30,33 we asked whether there is a re the peaked pattern in diversity observed a stochastic–deterministic processes of commun purely stochastic processes, diversity should v species have equal ﬁtness,55 unless some othe prevent this. It is recognized that, beyond description, an understanding of the underly necessary to comprehend the outcomes o turbance regimes.30,40,64 We hypothesize that intermediate disturbance frequencies is the stabilizing mechanisms (niches), which are s ends of the disturbance range. Stochastic produce even assemblages (higher α-diversit disturbed levels, whilst infrequent or too-fre will favour some species over others (low propose this idea as the intermediate stoch (ISH, Fig. Null model analysis on diversity To disentangle the roles of stochastic and deterministic processes as drivers of change in β-diversity it is necessary to incorporate a statistical null model in the analysis,87 which assumes that species interactions are not important for community assembly.88 We employed a null model approach originally applied to woody plants50 and more recently to microbial communities.18 The model deﬁnes β-diversity as the β-partition ðβ ¼ 1 α=γÞ89 and takes into account both composition and relative abundances. To adapt it to handle microbial community data, we considered species as OTUs (genus taxonomic level) and each individual count as one read within the metagenomics dataset. The model randomizes the location of each individual within the three independent reactors for each of the eight disturbance treatment levels, while maintaining the total quantity of individuals per reactor, the relative abundance of each OTU, and the γ-diversity. We applied it to the metagenomics datasets from d0 and d35. Metagenomics sequencing and reads processing Puriﬁed genomic DNA from sludge samples on d0 (inoculum) and d35 (n = 24) were subjected to metagenomics sequencing at the SCELSE sequencing facility (Singapore). Library preparation was performed according to Illumina’s TruSeq Nano DNA Sample Preparation protocol. Libraries were sequenced in one lane on an Illumina HiSeq 2500 sequencer in rapid mode at a ﬁnal concentration of 11 pM and a read-length of 250 bp paired-end. Around 173 million paired-end reads were generated in total and 7.2 ± 0.7 million paired-end reads on an average per sample. Illumina adaptors, short reads, low quality reads or reads containing any ambiguous base were removed using cutadapt (–m 50 –q 20 - --max-n 0, v.1.11).74 Taxonomic assignment of metagenomics reads was done following the method described by Ilott et al.75 High quality reads (99.2 ± 0.09% of the raw reads) were randomly subsampled to an even depth of 12,395,400 for each sample prior to further analysis. They were aligned against the NCBI non-redundant (NR) protein database (March 2016) using DIAMOND (v.0.7.10.59) with default parameters.76 The lowest common ancestor approach implemented in MEGAN Community Edition v.6.5.577 was used to assign taxonomy to the NCBI-NR aligned reads with the following parameters: maxMatches = 25, minScore = 50, min Support = 20, paired = true. On average, 48.2% of the high-quality reads were assigned to cellular organisms, from which in turn 98% were assigned to the bacterial domain. Experimental design We employed sequencing batch microcosm bioreactors (20-mL working volume) inoculated with activated sludge from a full-scale plant and operated for 35 days. The daily complex synthetic feed (adapted from Hesselmann et al.69) included toxic 3-CA at varying frequencies. Eight levels of disturbance were set in triplicate independent reactors (n = 24), which received 3-CA every day (press-disturbed), every 2, 3, 4, 5, 6, or 7 days (intermediately-disturbed), or never (undisturbed). Level numbers were assigned from 0 to 7 (0 for no disturbance, 1 to 7 for low to high disturbance frequency, Supplementary Figure 1). Ecosystem function, in the form of process performance parameters at the end of a cycle, was measured weekly in accordance with Standard Methods70 where appro- priate, and targeted soluble chemical oxygen demand (COD), nitrogen species (ammonium, nitrite, and nitrate ions) and 3-CA, and volatile suspended solids (VSS). On the initial day and from the second week onwards, sludge samples (2 mL) were collected weekly for DNA extraction. The ISH can be further portrayed by noting a key reasoning behind the IDH, namely, that a competition–colonization trade-off would lead to higher diversity at intermediate levels of disturbance.31 In the context of our study, which comprised a closed system, colonization would come from the low abundance taxa that have an opportunity to grow after different disturbance Published in partnership with Nanyang Technological University E. Santillan et al. 7 16S rRNA gene amplicon ﬁngerprinting and processing Bray–Curtis dissimilarity matrixes were computed using square root transformed T-RFLP data and bacterial genus-level taxa tables generated using a metagenomics approach. Mantel tests were then used to determine the strength and signiﬁcance of the Pearson product–moment correlation between complete dissimilarity matrices. Procrustes tests (PROTEST) were also employed as an alternative approach to Mantel tests in order to compare and visualize both matrices on PCO and NMDS ordinations. The resultant m2-value is a statistic that describes the degree of concordance between the two matrices evaluated.86 All these statistical tests were performed using the vegan R-package (functions: procuste, mantel, metaMDS, vegdist). DNA extracted from all sludge samples (n = 99) was analysed by T-RFLP of the 16S rRNA gene using the 530F–1050R primer set targeting V4–V5 regions. Comparison between metagenomics and T-RFLP community datasets Comparison between metagenomics and T-RFLP community datasets Mantel and Procrustes tests85 were applied to compare metagenomics and T-RFLP datasets from all bioreactors on day 35 (n = 24, subsample of the full T-RFLP dataset). Such an approach is valid for the questions asked in this study, since comparisons between NGS and ﬁngerprinting techniques support the use of T-RFLP to detect meaningful community assembly patterns and correlations with environmental variables,61 and such patterns can be validated by NGS on a subset of the ﬁngerprinting dataset.2 DATA AVAILABILITY DNA sequencing data are available at NCBI BioProjects with accession number: 389377. See Supplementary Methods for details on sludge inoculum collection, complex synthetic wastewater preparation, scheme for ecosystem function measurement and sludge collection, chemical analysis, DNA extractions, 16S rRNA gene community ﬁngerprinting, metagenomics library preparation and sequencing, multivariate analyses, alpha-diversity indices, univariate analysis of variance and correlation tests, and null model analysis. Ecosystem function data, R-script for the null model analyses, T-RFLP raw data, and all other relevant data can be publicly accessed on FigShare (https://doi.org/10.6084/m9.ﬁgshare.7369964). Experimental design The PCR program included initial denaturation at 95 °C for 10 min, followed by 30 cycles of denaturation (95 °C, 1 min), annealing (58 °C, 30 s) and extension (72 °C, 1 min), and ﬁnal extension at 72 °C for 7 min. Puriﬁed DNA products were digested using the BsuRI (HaeIII) enzyme through incubating at 37 °C for 16 h. Enzyme inactivation was performed at 80 °C for 20 min. Digested DNA was subjected to T-RFLP on an ABI 3730XL DNA analyser. Sequence alignment ﬁles from T-RFLP runs were assessed for quality control and pre-processed using the software GeneMapper v.5 (Applied Biosystems).71 Peak areas were normalized to the total area per sample72 and de-noised using a conservative ﬂuorescence threshold of 200 units.73 Reporting summary Further information on experimental design is available in the Nature Research Reporting Summary linked to this article. Microbial community analysis and statistical tests All reported P-values for statistical tests in this study were corrected for multiple comparisons using a false discovery rate (FDR) of 10%.78 Community structure was assessed by a combination of ordination methods (PCO, NMDS, CAP) and multivariate tests (PERMANOVA, PERMDISP)79 on Bray–Curtis dissimilarity matrixes constructed from square-root transformed normalized abundance data using PRIMER (v.7). Additionally, GLMMs, which deal with mean–variance relationships,80 were employed using the mvabund R-package81 ﬁtting the metagenomics dataset to a negative binomial distribution, to ensure that the observed differences among groups were due to disturbance levels and not heteroscedasticity. The 500 most abundant genera (97% of total assigned reads abundances) were employed to ensure random distribution of residuals ﬁtted in the model. Signiﬁcance was tested using the anova function in R with PIT-trap bootstrap resampling (n = 999).82 Hill diversity indices83 were employed to measure α-diversity as described else- where,48,84 and calculated for normalized non-transformed relative abundance data. ACKNOWLEDGEMENTS This research was supported by the Singapore National Research Foundation and Ministry of Education under the Research Centre of Excellence Program. We thank C. W. Liew for help with sludge sampling and S.R. Lohar for the library preparations for metagenomics. F. Lauro, R.B.H. Williams, and S. Kjelleberg are acknowledged for their comments on an earlier version of the manuscript. The authors thank M. Holyoak and three anonymous reviewers for their critical and detailed feedback, as well as E.M. Marzinelli for discussions on data transformations and GLMMs. E.S. was partially supported by a Fulbright fellowship. Null model analysis on diversity Adequacy of sequencing depth was corroborated with rarefaction curves at the genus taxonomy level (Supplementary Figure 8) using the rarefy function of the vegan R-package (v.2.5-2). We did not include genotypic information as it was outside the scope of this study, but will do so in future investigations arising from this work. Each step of the null model calculates expected mean α-diversities for each disturbance level and then estimates an expected β-partition. After 10,000 repetitions, the mean and standard deviation of the distribution of random β-partitions for each disturbance level are calculated. The output of this model is a β-deviation or SES, which is the observed β-diversity (βobs) minus the mean of the null distribution of β-diversity values ðβexpÞ, divided by the standard deviation of this distribution (σexp), SES = ðβobs βexpÞ=σexp. We further calculated the SI as the difference between the observed and mean expected β-diversities divided by the observed β- diversity, SI = ðβobs βexpÞ=βobs. Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 REFERENCES W. The intermediate disturbance hypothesis is broadly deﬁned, sub- stantive issues are key: a reply to Sheil and Burslem. Trends Ecol. Evol. 28, 572–573 (2013). 9. Adler, P. B., Hillerislambers, J. & Levine, J. M. A niche for neutrality. Ecol. Lett. 10, 95–104 (2007). 10. Vergnon, R., Dulvy, N. K. & Freckleton, R. P. Niches versus neutrality: uncovering the drivers of diversity in a species-rich community. Ecol. Lett. 12, 1079–1090 (2009). 40. Shea, K., Roxburgh, S. H. & Rauschert, E. S. J. Moving from pattern to process: coexistence mechanisms under intermediate disturbance regimes. Ecol. Lett. 7, 491–508 (2004). 11. Chase, J. M. & Myers, J. A. Disentangling the importance of ecological niches from stochastic processes across scales. Philos. Trans. R. Soc. B Biol. Sci. 366, 2351–2363 (2011). 41. Krol, J. E. et al. Role of IncP-1 beta plasmids pWDL7::rfp and pNB8c in chlor- oaniline catabolism as determined by genomic and functional analyses. Appl. Environ. Microbiol. 78, 828–838 (2012). 12. Fisher, C. K. & Mehta, P. The transition between the niche and neutral regimes in ecology. Proc. Natl. Acad. Sci. U.S.A. 111, 13111–13116 (2014). 42. Falk, M. W. & Wuertz, S. Effects of the toxin 3-chloroaniline at low concentrations on microbial community dynamics and membrane bioreactor performance. Water Res. 44, 5109–5115 (2010). 13. Oﬁteru, I. D. et al. Combined niche and neutral effects in a microbial wastewater treatment community. Proc. Natl. Acad. Sci. U.S.A. 107, 15345–15350 (2010). 14. Jeraldo, P. et al. Quantiﬁcation of the relative roles of niche and neutral processes 43. Jessup, C. M. et al. Big questions, small worlds: microbial model systems in ecology. Trends Ecol. Evol. 19, 189–197 (2004). 14. Jeraldo, P. et al. Quantiﬁcation of the relative roles of niche and neutral processes in structuring gastrointestinal microbiomes. Proc. Natl. Acad. Sci. U.S.A. 109, 9692–9698 (2012). in structuring gastrointestinal microbiomes. Proc. Natl. Acad. Sci. U.S.A. 109, 9692–9698 (2012). 44. Benton, T. G., Solan, M., Travis, J. M. J. & Sait, S. M. Microcosm experiments can inform global ecological problems. Trends Ecol. Evol. 22, 516–521 (2007). 15. Pholchan, M. K., Baptista, J. D., Davenport, R. J., Sloan, W. T. & Curtis, T. P. Microbial community assembly, theory and rare functions. Front. Microbiol. 4, 68 (2013). 45. Drake, J. M. & Kramer, A. M. Mechanistic analogy: how microcosms explain nature. Theor. Ecol. 5, 433–444 (2012). 16. Dini-Andreote, F., Stegen, J. C., van Elsas, J. D. & Salles, J. F. REFERENCES 31. Connell, J. H. Diversity in tropical rain forests and coral reefs. Science 199, 1302–1310 (1978). 1. Stegen, J. C. et al. Quantifying community assembly processes and identifying features that impose them. ISME J. 7, 2069–2079 (2013). 32. Svensson, J. R., Lindegarth, M., Jonsson, P. R. & Pavia, H. Disturbance–diversity models: what do they really predict and how are they tested? Proc. R. Soc. B Biol. Sci. 279, 2163–2170 (2012). 2. Powell, J. R. et al. Deterministic processes vary during community assembly for ecologically dissimilar taxa. Nat. Commun. 6, 8444 (2015). 33. Kershaw, H. M. & Mallik, A. U. Predicting plant diversity response to disturbance: applicability of the intermediate disturbance hypothesis and mass ratio hypothesis. Crit. Rev. Plant Sci. 32, 383–395 (2013). 3. Zhou, J. & Ning, D. Stochastic community assembly: does it matter in microbial ecology? Microbiol. Mol. Biol. Rev. 81, e00002-17 (2017). ecology? Microbiol. Mol. Biol. Rev. 81, e00002-17 (2017). 4. Mouillot, D., Graham, N. A. J., Villeger, S., Mason, N. W. H. & Bellwood, D. R. A functional approach reveals community responses to disturbances. Trends Ecol. Evol. 28, 167–177 (2013). 34. Kim, M., Heo, E., Kang, H. & Adams, J. Changes in soil bacterial community structure with increasing disturbance frequency. Microb. Ecol. 66, 171–181 (2013). 35. Gibbons, S. M. et al. Disturbance regimes predictably alter diversity in an eco- logically complex bacterial system. mBio 7, e01372-16 (2016). 5. Briones, A. & Raskin, L. Diversity and dynamics of microbial communities in engineered environments and their implications for process stability. Curr. Opin. Biotechnol. 14, 270–276 (2003). logically complex bacterial system. mBio 7, e01372-16 (2016). 36. Roxburgh, S. H., Shea, K. & Wilson, J. B. The intermediate disturbance hypothesis: patch dynamics and mechanisms of species coexistence. Ecology 85, 359–371 (2004). 6. Rosindell, J., Hubbell, S. P. & Etienne, R. S. The uniﬁed neutral theory of biodi- versity and biogeography at age ten. Trends Ecol. Evol. 26, 340–348 (2011). 37. Fox, J. W. The intermediate disturbance hypothesis should be abandoned. Trends Ecol. Evol. 28, 86–92 (2013). 7. Silvertown, J. Plant coexistence and the niche. Trends Ecol. Evol. 19, 605–611 (2004). 8. Gravel, D., Canham, C. D., Beaudet, M. & Messier, C. Reconciling niche and neu- trality: the continuum hypothesis. Ecol. Lett. 9, 399–409 (2006). 38. Sheil, D. & Burslem, D. Deﬁning and defending Connell’s intermediate dis- turbance hypothesis: a response to Fox. Trends Ecol. Evol. 28, 571–572 (2013). 39. Fox, J. ADDITIONAL INFORMATION 26. Cain, M., Bowman, W. & Hacker, S. Ecology 3rd edn. (Sinauer Associates Inc., Sunderland, Massachusetts, USA, 2014). Supplementary information accompanies the paper on the npj Bioﬁlms and Microbiomes website (https://doi.org/10.1038/s41522-019-0079-4). Supplementary information accompanies the paper on the npj Bioﬁlms and Microbiomes website (https://doi.org/10.1038/s41522-019-0079-4). 27. Shade, A. et al. Fundamentals of microbial community resistance and resilience. Front. Microbiol. 3, 417 (2012). 28. Mackey, R. L. & Currie, D. J. The diversity–disturbance relationship: is it generally strong and peaked? Ecology 82, 3479–3492 (2001). Competing interests: The authors declare no competing interests. Competing interests: The authors declare no competing interests. 29. Shade, A. et al. Lake microbial communities are resilient after a whole-ecosystem disturbance. ISME J. 6, 2153–2167 (2012). Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 30. Miller, A. D., Roxburgh, S. H. & Shea, K. How frequency and intensity shape diversity–disturbance relationships. Proc. Natl. Acad. Sci. U.S.A. 108, 5643–5648 (2011). AUTHOR CONTRIBUTIONS S.W. and E.S. conceived the study. E.S. and H.S. designed the experiment. S.W. obtained the funding for the study. H.S. developed the process performance analysis pipeline employed. E.S. and H.S. performed the experiments and conducted all laboratory and molecular analyses (except library preparation and sequencing). D.I.D.- Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 E. Santillan et al. 8 23. Lee, J. E., Buckley, H. L., Etienne, R. S. & Lear, G. Both species sorting and neutral processes drive assembly of bacterial communities in aquatic microcosms. FEMS Microbiol. Ecol. 86, 288–302 (2013). M. developed the pipeline for the metagenomics library preparation and sequencing. E.S. performed the T-RFLP, and F.C. the metagenomics bioinformatics analyses. E.S. interpreted the data, generated the results, and elaborated the main arguments in the Introduction and Discussion. F.C. performed the Mantel and Procrustes tests. E.S. and S.W. wrote the manuscript. H.S., F.C. and D.I.D.-M. critically reviewed the manuscript. 24. Grifﬁn, J. S. & Wells, G. F. Regional synchrony in full-scale activated sludge bior- eactors due to deterministic microbial community assembly. ISME J. 11, 500–511 (2017). 25. Zhou, J. Z. et al. Stochastic assembly leads to alternative communities with dis- tinct functions in a bioreactor microbial community. mBio 4, e00584 (2013). Published in partnership with Nanyang Technological University REFERENCES Disentangling mechanisms that mediate the balance between stochastic and deterministic processes in microbial succession. Proc. Natl. Acad. Sci. U.S.A. 112, E1326–E1332 (2015). 46. Prosser, J. I. Replicate or lie. Environ. Microbiol. 12, 1806–1810 (2010). 47. Anderson, M. J. & Willis, T. J. Canonical analysis of principal coordinates: a usefu method of constrained ordination for ecology. Ecology 84, 511–525 (2003). 17. Caruso, T. et al. Stochastic and deterministic processes interact in the assembly of desert microbial communities on a global scale. ISME J. 5, 1406–1413 (2011). 48. Haegeman, B. et al. Robust estimation of microbial diversity in theory and in practice. ISME J. 7, 1092–1101 (2013). p 49. Chase, J. M. Stochastic community assembly causes higher biodiversity in more productive environments. Science 328, 1388–1391 (2010). 18. Zhou, J. Z. et al. Stochasticity, succession, and environmental perturbations in a ﬂuidic ecosystem. Proc. Natl. Acad. Sci. U.S.A. 111, E836–E845 (2014). 50. Kraft, N. J. B. et al. Disentangling the drivers of β diversity along latitudinal and elevational gradients. Science 333, 1755–1758 (2011). 19. Stegen, J. C., Lin, X. J., Konopka, A. E. & Fredrickson, J. K. Stochastic and deter- ministic assembly processes in subsurface microbial communities. ISME J. 6, 1653–1664 (2012). 51. Soto-Ortiz, L. The regulation of ecological communities through feedback loops: a review. Res. Zool. 5, 1–15 (2015). 20. Wang, J. J. et al. Phylogenetic beta diversity in bacterial assemblages across eco- systems: deterministic versus stochastic processes. ISME J. 7, 1310–1321 (2013). 52. Holyoak, M. & Loreau, M. Reconciling empirical ecology with neutral community models. Ecology 87, 1370–1377 (2006). 21. Dumbrell, A. J., Nelson, M., Helgason, T., Dytham, C. & Fitter, A. H. Relative roles of niche and neutral processes in structuring a soil microbial community. ISME J. 4, 337–345 (2010). 53. Fukami, T. Historical contingency in community assembly: integrating niches, species pools, and priority effects. Annu. Rev. Ecol. Evol. Syst. 46, 1–23 (2015). 22. Langenheder, S. & Szekely, A. J. Species sorting and neutral processes are both important during the initial assembly of bacterial communities. ISME J. 5, 1086–1094 (2011). 54. Nemergut, D. R. et al. Patterns and processes of microbial community assembly. Microbiol. Mol. Biol. Rev. 77, 342–356 (2013). Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 E. Santillan et al. 9 55. Hubbell, S. P. The uniﬁed neutral theory of biodiversity and biogeography, Vol. 32. REFERENCES In Monographs in Population Biology 1–375 (Princeton University Press, Prin- ceton, New Jersey, USA, 2001). 76. Buchﬁnk, B., Xie, C. & Huson, D. H. Fast and sensitive protein alignment using DIAMOND. Nat. Methods 12, 59–60 (2015). 77. Huson, D. H. et al. MEGAN Community Edition—interactive exploration and analysis of large-scale microbiome sequencing data. PLoS Comput. Biol. 12, e1004957 (2016). y 56. Shade, A. Diversity is the question, not the answer. ISME J. 11, 1–6 (2017). 57. Violle, C., Pu, Z. C. & Jiang, L. Experimental demonstration of the importance of competition under disturbance. Proc. Natl. Acad. Sci. U.S.A. 107, 12925–12929 (2010). 78. Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57, 289–300 (1995). 58. Wittebolle, L. et al. Initial community evenness favours functionality under selective stress. Nature 458, 623–626 (2009). 79. Clarke, K. R. & Gorley, R. N. PRIMER v7: User Manual/Tutorial (PRIMER-E, 2015). 80. Warton, D. I., Wright, S. T. & Wang, Y. Distance-based multivariate analyses con- found location and dispersion effects. Methods Ecol. Evol. 3, 89–101 (2012). 59. Weithoff, G., Walz, N. & Gaedke, U. The intermediate, disturbance hypothesis— species diversity or functional diversity? J. Plankton Res. 23, 1147–1155 (2001). 81. Wang, Y., Naumann, U., Wright, S. T. & Warton, D. I. mvabund—an R package for model-based analysis of multivariate abundance data. Methods Ecol. Evol. 3, 471–474 (2012). 60. Krause, S. et al. Trait-based approaches for understanding microbial biodiversity and ecosystem functioning. Front. Microbiol. 5, 251 (2014). and ecosystem functioning. Front. Microbiol. 5, 251 (2014). 82. Warton, D. I., Thibaut, L. & Wang, Y. A. The PIT-trap—a “model-free” bootstrap procedure for inference about regression models with discrete, multivariate responses. PLoS One 12, e0181790 (2017). 61. van Dorst, J. et al. Community ﬁngerprinting in a sequencing world. FEMS Microbiol. Ecol. 89, 316–330 (2014). 62. Haddad, N. M. et al. Species’ traits predict the effects of disturbance and pro- ductivity on diversity. Ecol. Lett. 11, 348–356 (2008). 83. Hill, M. O. Diversity and evenness: a uniﬁying notation and its consequences. Ecology 54, 427–432 (1973). 63. Fierer, N., Barberan, A. & Laughlin, D. C. Seeing the forest for the genes: using metagenomics to infer the aggregated traits of microbial communities. Front. Microbiol. 5, 614 (2014). 84. Vuono, D. C. et al. Disturbance and temporal partitioning of the activated sludge metacommunity. ISME J. 9, 425–435 (2015). npj Bioﬁlms and Microbiomes (2019) 8 REFERENCES 64. Huston, M. A. Disturbance, productivity, and species diversity: empiricism vs. logic in ecological theory. Ecology 95, 2382–2396 (2014). 85. Peres-Neto, P. R. & Jackson, D. A. How well do multivariate data sets match? The advantages of a Procrustean superimposition approach over the Mantel test. Oecologia 129, 169–178 (2001). 65. Daims, H., Taylor, M. W. & Wagner, M. Wastewater treatment: a model system for microbial ecology. Trends Biotechnol. 24, 483–489 (2006). 86. Jackson, D. A. PROTEST: a PROcrustean Randomization TEST of community environment concordance. Ecoscience 2, 297–303 (1995). microbial ecology. Trends Biotechnol. 24, 483–489 (2006). 66. Tchobanoglous, G. B., Franklin, L. & Stensel, H. D. Wastewater Engineering: Treatment and Reuse 4th ed. (McGraw Hill, Boston, Massachusetts, USA, 2003). 87. Anderson, M. J. et al. Navigating the multiple meanings of beta diversity: a roadmap for the practicing ecologist. Ecol. Lett. 14, 19–28 (2011). 67. Pholchan, M. K., Baptista, J. D., Davenport, R. J. & Curtis, T. P. Systematic study of the effect of operating variables on reactor performance and microbial diversity in laboratory-scale activated sludge reactors. Water Res. 44, 1341–1352 (2010). 88. Gotelli, N. J. & McGill, B. J. Null versus neutral models: what’s the difference? Ecography 29, 793–800 (2006). 68. Altermatt, F. et al. Big answers from small worlds: a user’s guide for protist microcosms as a model system in ecology and evolution. Methods Ecol. Evol. 6, 218–231 (2015). 89. Tuomisto, H. A diversity of beta diversities: straightening up a concept gone awry. Part 1. Deﬁning beta diversity as a function of alpha and gamma diversity. Eco- graphy 33, 2–22 (2010). 69. Hesselmann, R. P. X., Werlen, C., Hahn, D., van der Meer, J. R. & Zehnder, A. J. B. Enrichment, phylogenetic analysis and detection of a bacterium that performs enhanced biological phosphate removal in activated sludge. Syst. Appl. Microbiol. 22, 454–465 (1999). npj Bioﬁlms and Microbiomes (2019) 8 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. 70. APHA–AWWA–WEF. Standard Methods for the Examination of Water and Waste- water 22nd edn. (AWWA, Washington DC, USA, 2005). 71. Singh, B. K. & Thomas, N. Multiplex-terminal restriction fragment length poly- morphism. Nat. Protoc. 1, 2428–2433 (2006). 72. Kaplan, C. W. & Kitts, C. L. Bacterial succession in a petroleum land treatment unit. Appl. Environ. Microbiol. 70, 1777–1786 (2004). 73. Abdo, Z. et al. Statistical methods for characterizing diversity of microbial com- munities by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes. Environ. Microbiol. 8, 929–938 (2006). 74. Chen, C., Khaleel, S. S., Huang, H. & Wu, C. H. Software for pre-processing Illumina next-generation sequencing short read sequences. Source Code Biol. Med. 9, 8 (2014). © The Author(s) 2019 75. Ilott, N. E. et al. Deﬁning the microbial transcriptional response to colitis through integrated host and microbiome proﬁling. ISME J. 10, 2389–2404 (2016). Published in partnership with Nanyang Technological University npj Bioﬁlms and Microbiomes (2019) 8 Published in partnership with Nanyang Technological University
https://openalex.org/W4200311398	https://www.ssph-journal.org/articles/10.3389/phrs.2021.1604352/pdf	English	null	Diverse Approaches to Creating and Using Causal Loop Diagrams in Public Health Research: Recommendations From a Scoping Review	Public health reviews	2,021	cc-by	11,999	Public Health Reviews REVIEW published: 14 December 2021 doi: 10.3389/phrs.2021.1604352 Diverse Approaches to Creating and Using Causal Loop Diagrams in Public Health Research: Recommendations From a Scoping Review Lori Baugh Littlejohns, Carly Hill and Cory Neudorf* Community Health and Epidemiology, University of Saskatchewan, Saskatoon, SK, Canada Objectives: Complex systems thinking methods are increasingly called for and used as analytical lenses in public health research. The use of qualitative system mapping and in particular, causal loop diagrams (CLDs) is described as one promising method or tool. To our knowledge there are no published literature reviews that synthesize public health research regarding how CLDs are created and used. Methods: We conducted a scoping review to address this gap in the public health literature. Inclusion criteria included: 1) focused on public health research, 2) peer reviewed journal article, 3) described and/or created a CLD, and 4) published in English from January 2018 to March 2021. Twenty-three articles were selected from the search strategy. Results: CLDs were described as a new tool and were based upon primary and secondary data, researcher driven and group processes, and numerous data analysis methods and frameworks. Intended uses of CLDs ranged from illustrating complexity to informing policy and practice. Keywords: scoping review, causal loop diagrams, public health research, methods, complex systems thinking Keywords: scoping review, causal loop diagrams, public health research, methods, complex systems thinking Correspondence: Cory Neudorf cory.neudorf@usask.ca There is a trend in public health research for the application of complex systems thinking methods and tools [1–3]. We conceptualize public health research from this perspective in terms of examining systems that are complex webs of sectors, institutions, people, structures, and interventions that aspire to maintain and improve population health. Furthermore, we value public health research that is “based on the principles of social justice, attention to human rights and equity, evidence-informed policy and practice, and addressing the underlying determinants of health” [4]. Received: 16 July 2021 Accepted: 25 November 2021 Published: 14 December 2021 Edited by: Kasia Czabanowska, Maastricht University, Netherlands Edited by: Conclusion: From our learnings we propose nine recommendations for building knowledge and skill in creating and using CLDs for future public health research. Reviewed by: Robin van Kessel, Maastricht University, Netherlands Brian Li Han Wong, Independent Researcher, London, United Kingdom Citation: Baugh Littlejohns L, Hill C and Neudorf C (2021) Diverse Approaches to Creating and Using Causal Loop Diagrams in Public Health Research: Recommendations From a Scoping Review. Public Health Rev 42:1604352. doi: 10.3389/phrs.2021.1604352 There are published review articles regarding complex systems thinking methods used in public health research and together these paint a broad landscape [2, 3, 5–10]. In this literature, there is clear support for using qualitative system mapping and in particular, causal loop diagrams (CLDs) as analytical tools to embed complex systems thinking. The origins of the use of CLDs emanate from the system dynamics branch of systems science founded by Forrester [11] and CLDs are needed because “we live in a complex of nested feedback loops” [12]. One example of using a CLD in public health research is a study of factors that inﬂuenced health promotion policy and practice in a regional public December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers CLDs in Public Health Research Baugh Littlejohns et al. 3) Why a CLD was selected as a method, 4) How the CLD was created, 5) How the CLD was used, and 6) Recommendations for future research using CLDs. 3) Why a CLD was selected as a method, 4) How the CLD was created, 5) How the CLD was used, and 6) Recommendations for future research using CLDs. 3) Why a CLD was selected as a method, health system [13]. Here, the CLD was useful because “feedback mechanisms can be seen as leverage points to strengthen systems” and to “identify potential opportunities to disrupt or slow down vicious feedback mechanisms or amplify those that are virtuous cycles.” At the time of this study (2018), there were few examples of CLDs in public health literature [14–21]. 4) How the CLD was created, 5) How the CLD was used, and 6) Recommendations for future research using CLDs. Two authors (LBL and CH) extracted verbatim text that aligned with the extraction categories and these were saved to a spreadsheet. Both authors reviewed the spreadsheet in its entirety, discussed individual articles to gain clarity, and wrote summary paragraphs to identify high level themes. Following this, for each article, summary statements were written for the six extraction categories and a table was created. The two authors reviewed each other’s summaries for accuracy and revisions were made. Research Aims Although the literature addressed a range of public health topics, non-communicable disease prevention was most frequently addressed (15/23) and of those, seven were focused on obesity prevention. Table 2 provides a list of research topics. Citation: Finally, directed content analysis was used to interpret extracted data “through systematic classiﬁcation of coding and identifying themes and patterns” [25]. To our knowledge there are no published reviews that synthesize public health research in terms of how CLDs are created and used. We were motivated to conduct a literature review to determine how CLD methodology could be used to identify leverage points in local public health systems to strengthen the response to COVID-19 in Canada. The aim of this paper is to address this gap in the literature and synthesize knowledge from recent innovations for our research and contribute to knowledge development. We posed two research questions: 1) How are CLDs created and used in recent (>2018) public health research? 2) What recommendations emerge regarding how to create and use CLDs in public health research? METHODS We found 23 articles in total that met our inclusion criteria. A list of these articles and summary statements are provided in Table 1. This section answers our ﬁrst research question: How are CLDs created and used in recent (>2018) public health research? The organization of this section mirrors the six data extraction categories indicated above. A scoping review was chosen for this study in order “to examine how research is conducted” and “to provide an overview or map of the evidence” [22]. A narrative synthesis approach was utilized as the topic required exploration more than explanation and human and time resources were limited [23]. Key issues identiﬁed by Byrne [24] to strengthen the review were addressed such as ensuring transparency in search strategy and data extraction, analysis and synthesis. 2) Description of complex systems thinking, Search Strategy Literature was searched using the Scopus and PubMed databases and used the following search terms: causal loop diagram, complex, system thinking, method*, tool, approach, research, and public health. Inclusion criteria were 1) public health research, 2) peer reviewed journal article, 3) described or created a CLD as a research method, and 4) published in English from January 2018 to March 2021. The key objective was to ﬁnd state-of-the-ﬁeld examples of CLDs, therefore, extensive hand searches of references was completed. It is important to note that piloting this search strategy uncovered numerous articles that only mentioned CLDs and did not explicitly meet the criteria of “described or created a CLD as a research method.” While we set out to use PRISMA guidelines we deemed it unnecessary given the search strategy quickly became one of including all articles that meet our inclusion criteria. In terms of research aims found in the 23 articles, four themes emerged: 1) to examine the complexity of a public health topic and illustrate complex systems thinking [26–34]; 2) to discuss the complexity of a public health intervention [35–40]; 3) to describe study protocol and how CLDs were created [41–44]; and 4) to illustrate how CLDs can be used to monitor and track initiatives to improve population health or evaluate impact of interventions [45–48]. Complex Systems Thinking Complex systems thinking was discussed in terms of systems, problems, interventions, and key concepts that drive this type of approach. Several articles indicated that the systems they were studying were complex, for example: 1) Research aim, 1) Research aim, A component of system dynamics modeling to provide insight into the dynamic complexity of the drugged driving environment and trafﬁc safety policy Data Extraction and Analysis First author/citation Research aim Description: complex systems thinking Why CLD w selected a meth Allender [47] To report on insight gathered during development, implementation, and evaluation of the ﬁrst 2 years in a systems-based childhood obesity prevention initiative that was inspired by community based system dynamics Complexity hampers traditional approaches to improving population health; need to conceptualize health as the result of actors and “interdependent elements connected at multiple levels”; initiatives such as obesity prevention need to address feedback loops that can lead to policy resistance, time delays that inﬂuence long term system change, and accumulations and their rates of change To create a visu the causes and childhood obes Araz [45] To analyze potential “real world” impacts of policy interventions on improving roadway safety regarding drugged driving behavior, road environment, and policy through system dynamics modeling Driver behavior was described as a complex system given the dynamic interrelationships and multidimensional variables associated with driving behaviors, policy, environment, and roadway conditions To illustrate vari inﬂuence drugge behaviors, road environment an safety policies Bensberg [35] To describe the establishment of a multi- community chronic disease prevention initiative (Healthy Together Victoria) through a systems thinking lens A way to address complex public health problems; holistic vs. reductionist perspective; the essence of a system is the causal connections between parts and feedback loops To summarize ﬁ illustrate feedba Bradley [34] To report on the importance of employing systems thinking for the prevention and response to COVID-19 Society is a complex adaptive system with interconnected factors impacting the spread of infection; system structure inﬂuences system behaviour; systems change is needed to mitigate COVID-19 To visualize the connections and components of better understan loops and relatio impacting the e of a system imp COVID-19 Brereton [28] To explore the complex causal relationships between children’s health, environment, social, and economic inﬂuences in least developed countries A science that explores how parts connect, react, and interact to increase recognition of non-linearity and cause and effect relationships; to view the “forest and the trees.” A tool to uncov problems that a difﬁcult to view complex system CLD tells a stor cause and effec feedback” and t used to surface models and poli among stakeho Brown [48] To present how a community used a CLD to A method to address complexity To present the r and variables th TABLE 1 \| Summary statements of extracted data (Canada, 2021). CLD was created from data on the most signiﬁcant causes of childhood mortality and a narrative literature review A tool to uncover root problems that are often difﬁcult to view within complex systems; “Each CLD tells a story that links cause and effect through feedback” and that can be used to surface mental models and policy decisions among stakeholders To visualize the causal connections and components of society to better understand feedback loops and relationships impacting the entire status of a system impacted by COVID-19 Data Extraction and Analysis y Study selection was conducted by one author (LBL) while appraisal and duplicate independent data extraction and validation was conducted by two authors (LBL and CH). CN provided input throughout the study and facilitated discussion about any differences. Data extraction followed these six categories: A complex system may be characterized by its heterogeneity (various actors and structures at different levels); its dynamic, interactive, and adaptive nature (its ability to respond to or resist external changes, or changes in the interacting parts); and its emergent properties (arising through interactions between processes or factors that alone do not exhibit such properties) [30]. 2) Description of complex systems thinking, Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers December 2021 \| Volume 42 \| Article 1604352 2 CLDs in Public Health Research Baugh Littlejohns et al. TABLE 1 \| Summary statements of extracted data (Canada, 2021). Data Extraction and Analysis First author/citation Research aim Description: complex systems thinking Why a CLD was selected as a method How the CLD wa created Allender [47] To report on insight gathered during development, implementation, and evaluation of the ﬁrst 2 years in a systems-based childhood obesity prevention initiative that was inspired by community based system dynamics Complexity hampers traditional approaches to improving population health; need to conceptualize health as the result of actors and “interdependent elements connected at multiple levels”; initiatives such as obesity prevention need to address feedback loops that can lead to policy resistance, time delays that inﬂuence long term system change, and accumulations and their rates of change To create a visual model of the causes and effects of childhood obesity Local behavioral (collected using a monitoring system electronic tablets used to support creation of CLD group model bui communities Araz [45] To analyze potential “real world” impacts of policy interventions on improving roadway safety regarding drugged driving behavior, road environment, and policy through system dynamics modeling Driver behavior was described as a complex system given the dynamic interrelationships and multidimensional variables associated with driving behaviors, policy, environment, and roadway conditions To illustrate variables that inﬂuence drugged driving behaviors, road environment and trafﬁc safety policies Researchers revie literature and pub data is determine parameters and ﬂow diagram was create the CLD a quantitative expre were derived for simulation model Bensberg [35] To describe the establishment of a multi- community chronic disease prevention initiative (Healthy Together Victoria) through a systems thinking lens A way to address complex public health problems; holistic vs. Data Extraction and Analysis reductionist perspective; the essence of a system is the causal connections between parts and feedback loops To summarize ﬁndings and illustrate feedback loops CLD was created researchers from analysis interview Bradley [34] To report on the importance of employing systems thinking for the prevention and response to COVID-19 Society is a complex adaptive system with interconnected factors impacting the spread of infection; system structure inﬂuences system behaviour; systems change is needed to mitigate COVID-19 To visualize the causal connections and components of society to better understand feedback loops and relationships impacting the entire status of a system impacted by COVID-19 CLD was created researchers only Brereton [28] To explore the complex causal relationships between children’s health, environment, social, and economic inﬂuences in least developed countries A science that explores how parts connect, react, and interact to increase recognition of non-linearity and cause and effect relationships; to view the “forest and the trees.” A tool to uncover root problems that are often difﬁcult to view within complex systems; “Each CLD tells a story that links cause and effect through feedback” and that can be used to surface mental models and policy decisions among stakeholders CLD was created data on the mos signiﬁcant causes childhood mortal narrative literature Brown [48] To present how a community used a CLD to track the underlying system changes resulting from implementing a healthy eating curriculum in a school A method to address complexity To present the relationships and variables that inﬂuence complex problems CLD was created seven group mod building sessions implementation s were tracked TABLE 1 \| Summary statements of extracted data (Canada, 2021). To illustrate variables that inﬂuence drugged driving behaviors, road environment and trafﬁc safety policies CLD was created by researchers only Highlight potential leverage points in children’s health and enable greater insight for policy and practice Provide a visual example of the dynamic and complex interactions and systems changes needed to address COVID-19 CLD was created from data on the most signiﬁcant causes of childhood mortality and a narrative literature review Highlight potential leverage points in children’s health and enable greater insight for policy and practice CLD was created from seven group model building sessions where implementation strategies were tracked CLD was created by researchers using data from interviews, documents, and ﬁelds notes Enhance theoretical analysis of obesity prevention policy and demonstrate feedback loops and leverage points that either spurred or resisted obesity prevention policy For simulation modelling and results were to inform policy debates Inform simulation scenarios and policy decision-making for group-level obesity interventions To conceptualize the system and inform system dynamic modeling; a means to simulate “what if” scenarios and emergent system behavior to better understand the causal links and variables at play CLD was created after conducting interviews with experts and the causal links found during the interviews were conﬁrmed via literature review A heuristic tool to help document interconnections, virtuous/vicious feedback mechanisms, and leverage points to inform strategies for systems change To increase understanding of the dynamics of complex problems and system behavior, the feedback mechanisms at play, and the determinants of fruit and vegetable intake among children CLD created as part of participatory system dynamics (SD) modelling that included stakeholder interviews and group model building CLD created from data obtained through group model building A step in system dynamic modeling A tool in system dynamic modeling Data Extraction and Analysis To create a visual model of the causes and effects of childhood obesity A component of system dynamics modeling to provide insight into the dynamic complexity of the drugged driving environment and trafﬁc safety policy Provide a visual example of the dynamic and complex interactions and systems changes needed to address COVID-19 To visualize the causal connections and components of society to better understand feedback loops and relationships impacting the entire status of a system impacted by COVID-19 Brown [48] To present how a community used a CLD to track the underlying system changes resulting from implementing a healthy eating curriculum in a school A method to address complexity CLD was created from seven group model building sessions where implementation strategies were tracked Demonstrate how a CLD can be used to measure system changes and evaluate obesity prevention interventions (Continued on following page) To present the relationships and variables that inﬂuence complex problems Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 3 CLDs in Public Health Research Baugh Littlejohns et al. to study complex problems as the manifestation of dynamic interactions among their constituent parts Help monitor and improve the design and (Continued on following page) December 2021 \| Volume 42 \| Article 1604352 Why a CLD was selected as a method Description: complex systems thinking How the CLD was created CLDs were created from stakeholder interviews, documentaries, an ethnography, and a literature review Convey new theoretical insights and implications regarding the interplay of factors for reducing violence escalation and disparity to study complex problems as the manifestation of dynamic interactions among their constituent parts To represent “dynamic hypotheses” about the system structure producing observed outcomes over time; illustrate complex interactions (e.g., interdependence, delays between cause and effect, mutual interaction, and feedback loops reinforcing or counteracting earlier changes) To represent “dynamic hypotheses” about the system structure producing observed outcomes over time; illustrate complex interactions (e.g., interdependence, delays between cause and effect, mutual interaction, and feedback loops reinforcing or counteracting earlier changes) Clarke [37] To examine the dynamics and decisions regarding obesity prevention policy adoption within multi- community chronic disease prevention initiative (Healthy Together Victoria) A non-linear and holistic perspective; appreciation of the multiple, interacting forces guiding policy decisions; understanding system behavior in terms of structures and patterns and feedback mechanisms Crielaard [26] To model social norms regarding body weight and obesity prevalence using system dynamics modeling A complex system is non- linear, is more than the aggregation of its parts, and has feedback loops that inﬂuence emergent system behaviour To conceptualize the system and inform system dynamic modeling; a means to simulate “what if” scenarios and emergent system behavior to better understand the causal links and variables at play Eker [27] To combine quantitative simulation modelling, an interpretivist approach, and a participatory method to examine housing, energy and wellbeing aspects of the UK’s housing stock To understand the dynamic behavior of complex systems or the systems underlying a policy problem and causal feedback thinking and non-linearity among elements; to examine the complexity of interactions between housing, energy and wellbeing A tool in system dynamic modeling CLD created as part of participatory system dynamics (SD) modelling that included stakeholder interviews and group model building For simulation modelling and results were to inform policy debates Gerritsen [41] To describe group model building and system mapping methods used to study fruit and vegetable intake among children and evaluate effectiveness of various tools (graphs over time, cognitive mapping and CLDs) A way to address complex problems characterized as having multiple causes, multilevel contexts, no single solution, and requiring multisectoral action To increase understanding of the dynamics of complex problems and system behavior, the feedback mechanisms at play, and the determinants of fruit and vegetable intake among children CLD created from data obtained through group model building Identify system change actions and increase understanding about complex systems and systems thinking Hassmiller Lich [46] To report on group concept mapping and system dynamics modeling as complimentary methods to address complex problems in evaluation and strategic planning A way to increase understanding of interconnected factors and cause and effect relationships that inﬂuence public health, social, behavioral, or environmental problems To engage stakeholders in identifying and visualizing cause-effect relationships among variables CLD created from data collected through group concept mapping Identify leverage points for strategic planning and intervention scenarios Jalali [38] To increase understanding of the effectiveness of A step in system dynamic modeling Help monitor and improve the design and (Continued on following page) A tool in system dynamic modeling To increase understanding of the dynamics of complex problems and system behavior, the feedback mechanisms at play, and the determinants of fruit and vegetable intake among children Help monitor and improve the design and (Continued on following page) December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers CLDs in Public Health Research Baugh Littlejohns et al. How the CLD was created TABLE 1 \| (Continued) Summary statements of extracted data (Canada, 2021). First author/citation Research aim Description: complex systems thinking Why a CLD was selected as a method How the CLD was created How the CLD was used obesity prevention interventions from an endogenous, organizational behavior or dynamics perspective and use system dynamic modeling methods To study the complexity or dynamics of program success and failure CLD created from interview data and published data implementation of interventions in order to avoid the dynamics that lead to poor outcomes Klement [29] To include individual level factors that inﬂuence COVID-19 to Sahin et al’s (2020) CLD of environmental-health- socio-economic systems of the COVID-19 pandemic To increase understanding the interconnections among parts of a system, feedback loops, system structure and behavior at multiple levels A step in system dynamic modeling CLD created from researcher knowledge, evidence, and assumptions Illustrate the complexity of COVID-19 Knai [30] To demonstrate the application of a complex systems approach to analyze the commercial determinants of health in terms of problem identiﬁcation and policy development Attends to heterogenous stakeholders and interventions, their dynamic interactions at multiple levels, adaptation and emergent system behavior, nonlinearity, feedback loops, and power dynamics in systems in order to inﬂuence systems to be more health promoting To illustrate the complexity of COVID-19 and the effectiveness of public health measures CLD created from researcher knowledge, evidence and assumptions Identify interventions and further research that highlights the interdependence among variables such as market and nonmarket strategies and sectors and how they work together to form system behavior with respect to commercial determinants of noncommunicable diseases Maitland [42] To report on study protocol for applying a ‘whole of system approach’ to evaluate strategies to address childhood obesity To examine complexity, nonlinearity, relationships among variables, and feedback loops To visualize the system by illustrating system components and interconnections that results in a narrative about a problem Study protocol; Did not create a CLD Study protocol; Did not create or use a CLD. Osman [31] To report on applying systems thinking methods and tools to identify interdependence and underlying factors that inﬂuence TB. To identify issues and strategies for improving the community health workers program To support community members and researchers to examine inﬂuencing factors of local systemic problems with respect to chronic disease prevention created from group eling building in each cipating community To highlight feedback loops that are either reinforcing or balancing and identify places to intervene CLD created from document review, key informant interviews, and workshops To illustrate the complexity of COVID-19 and the effectiveness of public health measures (Continued on following page) To understand complex systems and impact and/or consequences of changes to programming How the CLD was created TABLE 1 \| (Continued) Summary statements of extracted data (Canada, 2021). First author/citation Research aim Description: complex systems thinking Why a CLD was selected as a method How the CLD was created How the CLD was used capacity for ongoing action learning Sahin [32] To visualize the complexity in managing the COVID-19 pandemic through a systems lens by identifying the interconnectivity between health, economic, social and environmental aspects A framework to better understand the big picture through identifying the multi-faceted consequences of decisions and to design the most effective strategies to manage the impacts of unintended consequences To identify and illustrate feedback relationships and pinpointed leverage points CLD created from researchers existing knowledge, geographical data, and government documents via four expert workshops To identify leverage points to address COVID-19 Swierad [44] To describe how group model building was conducted and report on the ﬁndings with respect to childhood obesity Obesity is discussed in terms of a complex, multi- level problem To support community members to understand concepts and tools of system dynamics and systems thinking CLD created from group model building To illustrate and increase understanding of childhood obesity as a multifactorial problem (e.g., sociocultural factors), tailor culturally sensitive interventions, and generate hypotheses for further research Urwannachotima [33] To study the dynamic interactions among variables associated with sugar-sweetened beverage tax and dental caries in Thailand To take a whole system perspective of dynamic interactions To visualize dynamic interactions or relationships among variables and their interdependence CLD created from in- depth interviews and group model building The potential of the CLD lies with quantitative modeling and formulating recommendations for intervention TABLE 1 \| (Continued) Summary statements of extracted data (Canada, 2021). CLD created from researchers existing knowledge, geographical data, and government documents via four expert workshops To identify and illustrate feedback relationships and pinpointed leverage points historical values.” They go on to deﬁne the difference between reinforcing and balancing feedback loops. TABLE 2 \| Research topics of reviewed literature (Canada, 2021). How the CLD was created Health systems are complex adaptive systems; many interactions among parts produces system behaviour One component of a larger study to focus on nonlinearity of relationships among factors, feedback loops, and changes in context CLD created from a seminar with diverse experts; utilized ﬁshbone analysis, a 5 whys approach, and afﬁnity diagrams To develop implementation action plans, risk mitigation strategies and track changes in the system Owen [39] To report on applying systems thinking and feedback loops to create a CLD to visualize and understand the dynamic complexity of a successful intervention to address childhood obesity To understand system structure, feedback loops, non-linearity, delays, system behavior, factors that inﬂuence complex problems, and to identify interventions To identify and share understanding of system elements and nonlinear system structures that inﬂuences or dictates system behaviour CLD created from interview data To evaluate project implementation in order to understand leverage points to strengthen systems and/ or create new systems Parmar [40] To study roles of community health volunteers in managing diabetes and hypertension among Syrian refugees and recommend improvements To examine complex adaptive systems in terms of non-linear interactions among multiple actors and processes To understand complex systems and impact and/or consequences of changes to programming CLD created from document review, key informant interviews, and workshops To identify issues and strategies for improving the community health workers program Riley [43] To report on a novel combination of systems methods and tools and systemic inquiry processes in a study of community- based chronic disease prevention An analytic or conceptual lens (to study three organizing principles: interdependent relationships, perspectives and boundaries); systemic inquiry as a process to build To support community members and researchers to examine inﬂuencing factors of local systemic problems with respect to chronic disease prevention CLD created from group modeling building in each participating community To highlight feedback loops that are either reinforcing or balancing and identify places to intervene Description: complex systems thinking CLD created from researcher knowledge, evidence, and assumptions Illustrate the complexity of COVID-19 To identify and share understanding of system elements and nonlinear system structures that inﬂuences or dictates system behaviour To understand complex systems and impact and/or consequences of changes to programming To support community members and researchers to examine inﬂuencing factors of local systemic problems with respect to chronic disease prevention December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers CLDs in Public Health Research Baugh Littlejohns et al. How the CLD was created Research topic Citation Children’s Health [28] Community Violence [36] COVID-19 [29, 32, 34] Driving Behavior [45] Evaluation [46] Housing [27] Noncommunicable disease prevention Commercial Determinants of Health [30] Diabetes/Hypertension/Community Health Workers [40] Healthy Schools [48] Prevention Systems [35, 43] Sugar Sweetened Beverage Tax [33] Obesity Prevention: Organizational dynamics [38] Obesity Prevention: Weight-Related Behavior [26] Obesity Prevention: Children [39, 44, 47] Obesity Prevention: Fruit and Vegetable Intake [41] Obesity Prevention: Policy [37] Obesity Prevention: Whole of system [42] Tuberculosis [31] Another way complex systems thinking was described was with respect to complex problems and interventions. Burrell et al. [36] discussed community violence in terms of embedded contexts and the lack of holistic understanding of such “dynamic complexity.” Complex problems and interventions were often discussed together. The need to move away from “isolated intervention thinking” to systemic interventions to study systems change was highlighted by Knai et al. [30]. All articles built upon the descriptions reported above in some manner when discussing complex systems thinking. Some articles described this as providing “the opportunity to understand, test, and revise our understanding of how the different components in a system work together” [31] and “to study complex problems as the manifestation of dynamic interactions among their constituent parts” [36]. Furthermore, a few articles expanded the discussion to include such concepts as boundary judgement [38, 43, 47], that is, “establishing boundaries to the system is a fundamental starting point to efforts to change systems” [47]. Why Causal Loop Diagrams? CLDs were mostly seen as a means or a tool to examine feedback at play in public health issues. Some articles were explicit [28, 32, 33, 40, 43, 44] while others implied this. Both Riley et al. [43] and Parmar et al. [40] labeled this as “causal loop analysis” and the resulting CLDs were a means to understand systems and potential Following on this, feedback loops in complex systems were explicitly discussed in all articles to some extent. Jalali et al. [38] described these in terms of “causal chains of multiple variables in which changes in each variable could be traced back to its December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 6 TABLE 3 \| How causal loop diagrams were created (Canada, 2021). First Author/Citation Data used for CLD creation Process used for CLD creation Primary data Secondary data Researcher knowledge Researcher created only Researcher created with stakeholder reﬁnement GMB with Stakeholders Allender [47] ✓ ✓ Araz [45] ✓ ✓ Bensberg [35] ✓ ✓ Bradley [34] ✓ ✓ Brereton [28] ✓ ✓ Brown [48] ✓ ✓ Burrell [36] ✓ ✓ ✓ Clarke [37] ✓ ✓ ✓ Crielaard [26] ✓ ✓ ✓ Eker [27] ✓ ✓ Gerritsen [41] ✓ ✓ Hassmiller Lich [46] ✓ ✓ Jalali [38] ✓ ✓ ✓ Klement [29] ✓ ✓ Knai [30] ✓ ✓ Maitland [42] ✓ ✓ Osman [31] ✓ ✓ Owen [39] ✓ ✓ Parmar [40] ✓ ✓ ✓ Riley [43] ✓ ✓ Sahin [32] ✓ ✓ ✓ Swierad [44] ✓ ✓ Urwannachotima [33] ✓ ✓ Baugh Littlejohns et al. CLDs in Public Health Research CLDs in Public Health Research Baugh Littlejohns et al. information and government documents [32], and data from published databases [28, 37, 45]. Literature reviews were undertaken in four articles and these either supplemented primary data [26], secondary data [28, 45], or both [36]. Document and literature review were utilized in four articles [28–30, 45]. “programming.” Using a CLDs was a new tool for some [42, 46] and as one article related, “business as usual” was not working to address obesity [47]. CLDs were also considered a tool to help tell a story. For example, a CLD was thought to support the development of “a concise narrative about a particular problem” [42] and Brereton et al. Processes h h There were three processes used to create CLDs: group model building, researcher created only, and researcher created with stakeholder reﬁnement (Table 3). Group model building (GMB) was the most common process as reported in 11 articles [27, 31–33, 41–44, 46–48]. Urwannachotima et al. [33] described GMB as “an established methodology for engaging stakeholders to gain mutual understanding of complex relationships and to collectively develop comprehensive systems models that represent the cause and effect relationships of a problem.” They go further to explain that “stakeholders are deeply and actively involved in the process of model construction through the exchange, assimilation, and integration of mental models into a holistic system description.” GMB was generally reported to be a process where participants brainstormed and named potential variables, drew connections and feedback loops between the identiﬁed variables, and then mapped these ideas onto a ﬁnal CLD. However, there was a variety of GMB processes used and was often not clearly described in terms of session design and activities. Beyond GMB, Hassmiller Lich et al. [46] discussed group concept mapping and Gerritsen et al. [41] described graphing over time and cognitive mapping. How Were Causal Loop Diagrams Created? There were many combinations of methods used to create CLDs. In this section we present this diversity in terms of 1) data sources, 2) processes, 3) data analysis, 4) frameworks, and 5) diagramming (Table 3). Why Causal Loop Diagrams? [28] stated that “every causal loop tells a story that links cause and effect through feedback.” Data Sources First author/citation How were CLDS primarily intended to be used Inform policy Identify leverage points systems change Inform practice For system dynamic modelling Measure or evaluate Stakeholder engagement to take action To illustrate complexity To inform future research To enhance theory Allender [47] ✓ ✓ Araz [45] ✓ ✓ Bensberg [35] ✓ ✓ ✓ Bradley [34] ✓ ✓ Brereton [28] ✓ ✓ ✓ Brown [48] ✓ Burrell [36] ✓ ✓ ✓ ✓ Clarke [37] ✓ ✓ ✓ ✓ Crielaard [26] ✓ ✓ ✓ Eker [27] ✓ ✓ ✓ Gerritsen [41] ✓ ✓ Hassmiller Lich [46] ✓ ✓ Jalali [38] ✓ ✓ ✓ Klement [29] ✓ Knai [30] ✓ ✓ ✓ ✓ Maitland [42] ✓ ✓ ✓ Osman [31] ✓ ✓ ✓ Owen [39] ✓ ✓ Parmar [40] ✓ ✓ ✓ Riley [43] ✓ ✓ Sahin [32] ✓ ✓ Swierad [44] ✓ ✓ ✓ ✓ Urwannachotima [33] ✓ ✓ ✓ Allender [47] ✓ ✓ Araz [45] ✓ ✓ · Bensberg [35] ✓ ✓ ✓ Bradley [34] ✓ ✓ Brereton [28] ✓ ✓ ✓ Brown [48] ✓ Burrell [36] ✓ ✓ ✓ ✓ Clarke [37] ✓ ✓ ✓ ✓ Crielaard [26] ✓ ✓ ✓ Eker [27] ✓ ✓ ✓ Gerritsen [41] ✓ ✓ Hassmiller Lich [46] ✓ ✓ Jalali [38] ✓ ✓ ✓ Klement [29] ✓ Knai [30] ✓ ✓ ✓ ✓ Maitland [42] ✓ ✓ ✓ Osman [31] ✓ ✓ ✓ Owen [39] ✓ ✓ Parmar [40] ✓ ✓ ✓ Riley [43] ✓ ✓ Sahin [32] ✓ ✓ Swierad [44] ✓ ✓ ✓ ✓ Urwannachotima [33] ✓ ✓ ✓ TABLE 4 \| How causal loop diagrams were intended to be used (Canada, 2021). TABLE 4 \| How causal loop diagrams were intended to be used (Canada, 2021). First author/citation How were CLDS primarily intended to be used Inform policy Identify leverage points systems change Inform practice For system dynamic modelling Measure or evaluate Stakeholder engagement to take action To illustrate complexity To inform future research To enhance theory ✓ • Conducting a literature review to compare causal links uncovered in interview data [26] or a document review [29, 30], CLDs created by researchers only was the second most common process (10/23). Two articles reported that CLDs were presented to stakeholders for reﬁnement [39, 40]. Data Sources First author/citation How were CLDS primarily intended to be used Inform policy Identify leverage points systems change Inform practice For system dynamic modelling Measure or evaluate Stakeholder engagement to take action To illustrate complexity To inform future research To enhance theory Allender [47] ✓ ✓ Araz [45] ✓ ✓ Bensberg [35] ✓ ✓ ✓ Bradley [34] ✓ ✓ Brereton [28] ✓ ✓ ✓ Brown [48] ✓ Burrell [36] ✓ ✓ ✓ ✓ Clarke [37] ✓ ✓ ✓ ✓ Crielaard [26] ✓ ✓ ✓ Eker [27] ✓ ✓ ✓ Gerritsen [41] ✓ ✓ Hassmiller Lich [46] ✓ ✓ Jalali [38] ✓ ✓ ✓ Klement [29] ✓ Knai [30] ✓ ✓ ✓ ✓ Maitland [42] ✓ ✓ ✓ Osman [31] ✓ ✓ ✓ Owen [39] ✓ ✓ Parmar [40] ✓ ✓ ✓ Riley [43] ✓ ✓ Sahin [32] ✓ ✓ Swierad [44] ✓ ✓ ✓ ✓ Urwannachotima [33] ✓ ✓ ✓ Allender [47] ✓ ✓ Araz [45] ✓ ✓ · Bensberg [35] ✓ ✓ ✓ Bradley [34] ✓ ✓ Brereton [28] ✓ ✓ ✓ Brown [48] ✓ Burrell [36] ✓ ✓ ✓ ✓ Clarke [37] ✓ ✓ ✓ ✓ Crielaard [26] ✓ ✓ ✓ Eker [27] ✓ ✓ ✓ Gerritsen [41] ✓ ✓ Hassmiller Lich [46] ✓ ✓ Jalali [38] ✓ ✓ ✓ Klement [29] ✓ Knai [30] ✓ ✓ ✓ ✓ Maitland [42] ✓ ✓ ✓ Osman [31] ✓ ✓ ✓ Owen [39] ✓ ✓ Parmar [40] ✓ ✓ ✓ Riley [43] ✓ ✓ Sahin [32] ✓ ✓ Swierad [44] ✓ ✓ ✓ ✓ Urwannachotima [33] ✓ ✓ ✓ Baugh Littlejohns et al. CLDs in Public Health Research CLDs created by researchers only was the second most • Conducting a literature review to compare causal links TABLE 4 \| How causal loop diagrams were intended to be used (Canada, 2021). Data Sources Both primary and secondary data were used for creating CLDs (Table 3). Most articles reported on primary data collection (18/ 23) and this included interviews [26, 27, 33, 35–40], group model building with stakeholders and/or community members [32, 41, 43, 44, 46, 48], behavioral data [42, 47], ﬁeldnotes [37], and workshops with experts [31]. Twelve articles used primary data only. Secondary data was used in 10 articles [26, 28–30, 32, 36–38, 40, 45] and this consisted of document and/or literature review (Table 3). Of the eighteen articles that reported on primary data collection, six included document review [26, 32, 36–38, 40]. Documents included policy brieﬁngs, reports, consultation papers, and evaluation reports [37], documentaries and ethnographies [36], program data [38], geographical December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 7 CLDs in Public Health Research Baugh Littlejohns et al. TABLE 4 \| How causal loop diagrams were intended to be used (Canada, 2021). Data Sources The range of approaches included: • Completing both a literature review and a document review to identify variables [28, 45], y • Building on an existing CLD [29], and • Using coded interview data to map interactions between key variables [26, 35–38], • Creating a CLD solely from researcher knowledge and expertise [34]. December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers CLDs in Public Health Research Baugh Littlejohns et al. Frameworks l l Several articles applied speciﬁc frameworks to inform research. For example, Allender et al. [47] used Foster- Fishman’s [50] theoretical framework of six elements (i.e., systems norms, ﬁnancial resources, human resources, social resources, regulations, and operations) to study root causes, system interactions, and levers for change. Similarly, Baugh Littlejohns and Wilson’s [5] framework of seven attributes of effective prevention systems (i.e., leadership, resources, health equity paradigm, information, implementation of desired actions, complex systems thinking, collaborative capacity) was used by Bensberg et al. [35] in their study design. Informing practice was also a frequently identiﬁed intended use of CLDs (13/23) (Table 4). For example, Osman et al. [31] stated that their CLD could be used “to develop local action plans for implementation and consider strategies for mitigating possible future risks” and Parmar et al. [40] to develop “strategies to enhance capacities, services, and coordination to improve the health of refugees.” Data Analysis leverage point as well as knowledge and data, resources, workforce, and collaborative relationships that need to be “nudged in the desired direction.” One of the more detailed descriptions of leverage points was from Sahin et al. [32]. They adapted Meadows [51] framework of places to intervene in system to identify shallow or deep leverage points to address the “wicked complexity” of the COVID-19 pandemic. Overall, we found that description was often lacking regarding qualitative data analysis methods used. However, some articles [35, 37, 39] that collected primary data discussed methods described by Kim and Anderson [49]. Others such as Owen et al. [39] created a table to demonstrate how they used coded interview transcript statements to inform their CLD. Steps in the analysis included 1) using coded text to show causal linkages, 2) translating these to cause-and-effect variables, and 3) creating word-and-arrow diagrams for CLD use. Similarly, Brereton and Jagals [28] presented a table to identify variables and describe inﬂuencing links. Inform Policy and Practice Informing policy was a reported intended use of CLDs in twelve articles (Table 4). Some articles were detailed in offering policy directions while others simply stated that the CLD could inform policy. Clarke et al. [37] examined “key inﬂuences on policy processes, and to identify potential opportunities to increase the adoption of recommended policies” with respect to a state government obesity prevention initiative. Other examples include the need for policies to address population growth, family size, and family planning to improve child health [28], housing, energy and wellbeing [27], and sugar-sweetened beverage tax to reduce sugar consumption and dental caries [33]. Informing practice was also a frequently identiﬁed intended use of CLDs (13/23) (Table 4). For example, Osman et al. [31] stated that their CLD could be used “to develop local action plans for implementation and consider strategies for mitigating possible future risks” and Parmar et al. [40] to develop “strategies to enhance capacities, services, and coordination to improve the health of refugees.” Informing policy was a reported intended use of CLDs in twelve articles (Table 4). Some articles were detailed in offering policy directions while others simply stated that the CLD could inform policy. Clarke et al. [37] examined “key inﬂuences on policy processes, and to identify potential opportunities to increase the adoption of recommended policies” with respect to a state government obesity prevention initiative. Other examples include the need for policies to address population growth, family size, and family planning to improve child health [28], housing, energy and wellbeing [27], and sugar-sweetened beverage tax to reduce sugar consumption and dental caries [33]. Illustrate Complexity and Identify Leverage Points Illustrate Complexity and Identify Leverage Points Illustrating complexity was aligned with research aims in several articles (Table 4) and was implicit in the other articles with respect to using CLDs. Identifying leverage points was explicitly discussed in twelve articles. Osman et al. [31] found that key variables and their interactions pointed to strategies to enhance leadership “through a reduction in bureaucracy in the health system.” Similarly, Bensberg et al. [35] identiﬁed leadership as a For System Dynamics Modeling Many articles reported on the use of software for creating the actual diagram. Vensim [31, 35, 37, 39, 40, 44–46], Stella Architect [28], and STICK-E [43] were the three diagrammatic programs used. Further to the actual diagram, there was a wide array of CLD types and degrees of diagram readability. We found that some CLDs were kept quite simple, with fewer variables, arrows, and loops, while others were very complicated. For example, Brereton et al. [28] created a tightly packed and dense color-coded main CLD and six diagrams of various feedback loops to highlight key variables, relationships, and potential leverage points. Overall, we found that key variables in blocks or shapes, labelled arrows and feedback loops, color coding, legends, and clear diagram interpretation descriptions were important aspects for readability. Five articles created CLDs for use in system dynamics modeling [26, 27, 38, 45] (Table 3). This was deﬁned by Araz et al as “a computer-aided approach to model and facilitate analysis of complex system behaviors over time” [45]. They further described the steps in system dynamic modeling, and this was very much in line with other articles: We ﬁrst constructed a causal loop diagram (CLD) informed by the existing literature to present the causal relationships between variables in drugged driving behaviors and trafﬁc safety policies. A stock- ﬂow diagram (SFD) was then used to convert these dynamic processes into quantitative expressions and a simulation tool [45]. Intended Uses of Causal Loop Diagrams There were nine ways that CLDs were intended to be used and these are identiﬁed in Table 4. The following provides examples of each intended use. Mirroring the above descriptions, Crielaard et al. [26] discussed the value of system dynamic modeling in terms of testing policy options from “studying ‘what if’ scenarios using computational modelling approaches.” It was notable that Urwannachotima et al. [33] and Swierad et al. [44] stated that the primary value of CLDs was in quantitative modelling. From Theory to Leverage Points Some articles had strong theoretical coherence with respect to complex systems thinking that was demonstrated in discussions about the reasons for choosing, creating, and using CLDs. We learned that articles were most coherent when they ﬁrst discussed feedback loops from a theoretical perspective and then carried this through to creating CLDs and to using them to identify leverage points for systems change (see for example 30). Overall, the descriptions of feedback in the articles were aligned with the idea that CLDs are “the applications of the loop concept underlying feedback and mutual causality” and that feedback loops are “powerful unifying notions that illuminate the structure of arguments, explanations, and causal views” [56]. Meadows [51] is well-known for explaining that disrupting or amplifying feedback loops can be effective leverage points in systems change. Therefore, we recommend that future research be designed with this theoretical coherence in mind. Measure and Evaluate Systems Change Measure and Evaluate Systems Change Table 4 identiﬁes four articles that used CLDs to help measure and evaluate systems change [31, 39, 42, 48]. For example, Owen et al. [39] reported that “the methods provide a technique to retrospectively evaluate community interventions from a systems perspective and understand the way successful and unsuccessful December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 9 Baugh Littlejohns et al. CLDs in Public Health Research support preliminary recommendations or possible directions to take forward in future research. interventions addressed complexity.” They go further to explain that CLDs go beyond linear cause and effect logic models used in traditional evaluation and lessons regarding unintended consequences provide insights “to increase the chances of success for new prevention initiatives.” Theoretical Frameworks Lewin’s famous statement that “there is nothing so practical as good theory” was salient for what we learned [57]. Few articles used theoretical frameworks in research design or discussed the need to advance theory (i.e., complexity, systems) in public health research. The articles that used frameworks appeared to be more robust especially with respect to embedding theoretical constructs in the resultant CLD (see for example 35). While we appreciate that theory is emerging, we recommend that this be given more emphasis to help continue to build a solid foundation for furthering the application of CLDs in public health research. Boundary Judgements We learned that some articles described in detail theoretical orientations with respect to complex systems thinking while others gave brief explanations. The most frequent concepts regarding complex systems were the inherent dynamic interactions among many entities, factors, variables that illustrate whole system structure and behavior. This is consistent with other public health literature on the topic [52–54]. The difference in descriptions was more a matter of comprehensiveness than deﬁnitions. For example, boundary judgement was not well articulated in the articles. According to Ulrich [55], drawing boundaries builds in selectivity and partiality and therefore transparency is important in study design. Therefore, we recommend that attention be given to deﬁning boundaries to signal a speciﬁc endogenous perspective and a unique, snap-shot-in-time diagram of feedback loops of system behavior [56]. Enhance Stakeholder and Community Participation As discussed above, group model building (GMB) was a frequently reported process to create CLDs and inherent in these processes was the desire for stakeholder and/or community participation and shared understanding (Table 4). Gerritsen et al. [41] stated what many others did, that is, GMB helped people develop an understanding of the system under study and that “participants learn to see causal connections and how these connections result in patterns of behaviour evolving over time.” They hypothesized that resulting plans for system change would be more successful with this fundamental level of participation and understanding. Another article highlighted that GMB brought diverse stakeholders “together to develop a system understanding of the problem, thus paving the way for further collaboration and community action” [44]. Inform Future Research and Enhance Theoretical Perspectives The ﬁnal two intended uses of CLDs were to inform future research and enhance theoretical perspectives (Table 4). These intended uses were not widely discussed and if at all, they were mostly short aspirational statements. However, one example where future research was explicitly discussed was provided by Swierad et al. [44]. Here they reported that “hypotheses” from a CLD of childhood obesity could be used in future research such as “impact of food eaten at school inﬂuencing norms and acceptability of western/packaged food, elasticity of grandparents’ food norms, diversity of grandparents’ ideal body image for children, or beliefs in health of traditional foods.” With respect to using CLDs to enhance theoretical perspectives, Clarke et al. [37] suggested that the CLD “enhanced previously published theoretical analyses of obesity prevention policy decision-making systems by making explicit how underlying feedback loops either spurred policy change or resistance.” Another example is from Burrell et al. [36]. They reported that creating a CLD resulted in “a testable ecologically oriented theory of violence” and “the resulting model conveys new theoretical insights on how racial and economic features of urban settings interact with intrapsychic dimensions to create a self-perpetuating system of violence.” Qualitative Data Analysis Knai et al. [30] pointed out that current public health research “concentrates mainly on a system’s elements rather than the interconnections within it, and this is beginning to reveal its intrinsic limitations.” Some articles described data analysis methods to identify variables and examine interconnections to This section answers our second research question: What recommendations emerge regarding how to create and use CLDs in public health research? We offer nine learnings from the results above and interweave ideas from other research to December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 10 Baugh Littlejohns et al. CLDs in Public Health Research user engagement in deﬁning outcomes for strengthening public health policies and practices. draw CLDs, however, others lacked clear descriptions of the often highly iterative methods and therefore it was difﬁcult to follow a data trail and assess the resultant CLD. We recommend that more clarity be provided as to how researchers innovate in qualitative data analysis to further develop the art and science of creating CLDs. draw CLDs, however, others lacked clear descriptions of the often highly iterative methods and therefore it was difﬁcult to follow a data trail and assess the resultant CLD. We recommend that more clarity be provided as to how researchers innovate in qualitative data analysis to further develop the art and science of creating CLDs. Strengths and Limitations g In terms of limitations, the 23 articles were not considered to be comprehensive. Since completing the study, we found that Mui and others [64] published an article on a community-based system dynamics approach and suggests solutions for improving healthy food access in a low-income urban environment. We also found Savona et al. [65] identiﬁed the views of adolescents regarding the causes of obesity and used CLDs. While this can be considered a limitation, we hope to see a continual building of knowledge and skill in using CLDs in public health research. A strength of this paper is that 23 recent articles were identiﬁed that used CLDs and the depth and breadth of discussion in the articles provided good representation. Having three authors conduct the literature review is also a strength because this afforded a high degree of conﬁdence in reporting results and transparency in search strategy and data extraction, analysis and synthesis. Together the results and recommendations can contribute to informing global public health research by highlighting key considerations to help design research and address public health issues through complex systems thinking. Health Equity We conceptualize public health research to be guided by principles of social justice and human rights to address the goal of reducing health inequities through action on the determinants of health. Although many articles discussed determinants of health, the goal of reducing health inequities was largely absent. Baum et al. [62] discuss the concept of path dependency as “the tendency of institutions to retain policy directions and preferences rather than change or reform them.” They further suggest that disrupting “path dependency that exacerbates health inequities” is critical and we see how CLDs could uncover path dependencies. We recommend that CLDs in public health research should include the examination of leverage points for pro-equity policy and practice. Mixed Methods We found a range of research methods used to create CLDs. Ozawa et al. [58] state that mixed methods research is important because it allows researchers to view problems from multiple perspectives, contextualize information, develop a more complete understanding of a problem, triangulate results, quantify hard-to- measure constructs, provide illustrations of context for trends, examine processes/experiences along with outcomes and capture a macro picture of a system. The Diagram We hypothesize that mixed methods may produce more robust CLDs, however, this needs to be examined. We recommend that future research be undertaken to assess the strengths, limitations, and beneﬁts of using mixed methods and determine what methods create greater conﬁdence in the variables and feedback loops illustrated in CLDs. Senge [63] states that “reality is made up of circles” but often arguments and explanations are linear, therefore, CLDs can provide “a language of interrelationships” to uncover deep patterns in systems. Studying the interrelationships and explanations of each CLD was outside the scope of this paper, however, we learned about some basic elements of reader friendly CLDs. We recommend that the following questions could be used assess CLDs: Are established conventions [56] used effectively for drawing the CLD (e.g., labeling, positive and negative arrows, reinforcing and balancing loops)? Does the diagram illuminate the most signiﬁcant variables, feedback loops or leverage points? How well does the diagram function as an effective medium for presenting ﬁndings to knowledge users? How well does the CLD tell a story of what’s going on in a system? Participatory Action Research p y We found there was a wide range of who was involved in creating CLDs, from researchers only to multiple group model building sessions with stakeholders and community members. We see the latter methodology embedded in the traditions of action research [59] and/or community-based participatory research (CBPR) [60]. The CBPR approach involves “a commitment to conducting research that shares power with and engages community partners in the research process” and is intended “to increase knowledge and understanding of a given phenomena and integrate knowledge gained with interventions and policy and social change” [60]. There was little discussion of CBPR in the articles. We recommend that greater engagement with participatory action research literature be undertaken to embed the theory and philosophy of genuine participation and empowerment in research and action. Knowledge Translation There was limited discussion regarding how exactly CLDs were to be used to enhance evidence-informed policy and practice. Few articles explicitly discussed incorporating knowledge users or those able to use research results. As Sturmberg [61] relates, this requires users who are “deeply interested in understanding the highly interconnected and interdependent nature of the issues.” This led us to think about the importance of knowledge translation (KT) and how to strengthen the use of CLDs. Haynes et al. [6] state that KT needs to be conceptualized as not “a discrete piece of work within wider efforts to strengthen public health, but as integral to and in continual dialogue with those efforts.” We recommend that future public health research using CLDs should articulate KT plans that articulates knowledge December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 11 CLDs in Public Health Research Baugh Littlejohns et al. REFERENCES Immunization Services in Kerala, India. Health Res Pol Sys (2014) 12(1):47. doi:10.1186/1478-4505-12-47 Immunization Services in Kerala, India. Health Res Pol Sys (2014) 12(1):47. doi:10.1186/1478-4505-12-47 16. Agyepong IA, Aryeetey GC, Nonvignon J, Asenso-Boadi F, Dzikunu H, Antwi E, et al. Advancing the Application of Systems Thinking in Health: Provider Payment and Service Supply Behaviour and Incentives in the Ghana National Health Insurance Scheme - a Systems Approach. Health Res Pol Sys (2014) 12: 35. doi:10.1186/1478-4505-12-35 1. Rutter H, Savona N, Glonti K, Bibby J, Cummins S, Finegood D, et al. The Need for a Complex Systems Model of Evidence for Public Health. The Lancet (2017) 390(10112):2602–4. doi:10.1016/s0140-6736(17)31267-9 1. Rutter H, Savona N, Glonti K, Bibby J, Cummins S, Finegood D, et al. The Need for a Complex Systems Model of Evidence for Public Health. The Lancet (2017) 390(10112):2602–4. doi:10.1016/s0140-6736(17)31267-9 2. Carey G, Malbon E, Carey N, Joyce A, Crammond B, Carey A. Systems Science and Systems Thinking for Public Health: a Systematic Review of the Field. BMJ Open (2015) 5(12):e009002. doi:10.1136/bmjopen-2015-009002 2. Carey G, Malbon E, Carey N, Joyce A, Crammond B, Carey A. Systems Science and Systems Thinking for Public Health: a Systematic Review of the Field. BMJ Open (2015) 5(12):e009002. doi:10.1136/bmjopen-2015-009002 17. Biroscak BJ, Schneider T, Panzera AD, Bryant CA, McDermott RJ, Mayer AB, et al. Applying Systems Science to Evaluate a Community-Based Social Marketing Innovation. Soc Marketing Q (2014) 20(4):247–67. doi:10.1177/ 1524500414556649 3. Chughtai S, Blanchet K. Systems Thinking in Public Health: a Bibliographic Contribution to a Meta-Narrative Review. Health Policy Plan (2017) 32(4): 585–94. doi:10.1093/heapol/czw159 3. Chughtai S, Blanchet K. Systems Thinking in Public Health: a Bibliographic Contribution to a Meta-Narrative Review. Health Policy Plan (2017) 32(4): 585–94. doi:10.1093/heapol/czw159 18. Allender S, Owen B, Kuhlberg J, Lowe J, Nagorcka-Smith P, Whelan J, et al. .A Community Based Systems Diagram of Obesity Causes. PLoS ONE (2015) 10: e0129683. doi:10.1371/journal.pone.0129683 4. Canadian Public Health Association. Public Health: A Conceptual Framework. Ottawa (2017). 4. Canadian Public Health Association. Public Health: A Conceptual Framework. Ottawa (2017). 19. Homer J, Milstein B, Wile K, Pratibhu P, Farris R, Orenstein DR. Modeling the Local Dynamics of Cardiovascular Health: Risk Factors, Context, and Capacity. Prev Chronic Dis (2008) 5(2):A63. 5. Baugh Littlejohns L, Wilson A. Strengthening Complex Systems for Chronic Disease Prevention: a Systematic Review. BMC Public Health (2019) 19(1):729. doi:10.1186/s12889-019-7021-9 5. Baugh Littlejohns L, Wilson A. AUTHOR CONTRIBUTIONS Saskatchewan (CH/Dean’s summer research project scholarship). Funding for the open access publication fee will be covered through the Canadian Institutes for Health Research award. The funders were not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication. LB designed the overall research aim and questions and CN provided input throughout the study. Study selection was conducted by LB. Appraisal and duplicate independent data extraction and validation was conducted by two authors (LB and CH). LB and CH completed data analysis and all authors (LB, CH d CN) id d i t i t iti th ﬁ l i t REFERENCES Strengthening Complex Systems for Chronic Disease Prevention: a Systematic Review. BMC Public Health (2019) 19(1):729. doi:10.1186/s12889-019-7021-9 6. Haynes A, Rychetnik L, Finegood D, Irving M, Freebairn L, Hawe P. Applying Systems Thinking to Knowledge Mobilisation in Public Health. Health Res Pol Syst (2020) 18(1):134. doi:10.1186/s12961-020-00600-1 20. Brennan LK, Sabounchi NS, Kemner AL, Hovmand P. Systems Thinking in 49 Communities Related to Healthy Eating, Active Living, and Childhood Obesity. J Public Health Manag Pract (2015) 21:S55–69. doi:10.1097/ PHH.0000000000000248 7. Egan M, McGill E, Penney T, Meier PS, Savona N, de Vocht F, et al. Complex Systems for Evaluation of Public Health Interventions: a Critical Review. The Lancet (2018) 392:S31. doi:10.1016/s0140-6736(18)32053-1 21. Kopelman P, Jebb SA, Butland B. Foresight. Tackling Obesities: Future Choices - Project Report. London: Government Ofﬁce for Science (2007). 8. McGill E, Er V, Penney T, Egan M, White M, Meier P, et al. Evaluation of Public Health Interventions from a Complex Systems Perspective: A Research Methods Review. Soc Sci Med (2021) 272:113697. doi:10.1016/ j.socscimed.2021.113697 22. Munn Z, Peters MDJ, Stern C, Tufanaru C, McArthur A, Aromataris E. Systematic Review or Scoping Review? Guidance for Authors when Choosing between a Systematic or Scoping Review Approach. BMC Med Res Methodol (2018) 18(1):143. doi:10.1186/s12874-018-0611-x 9. McGill E, Marks D, Er V, Penney T, Petticrew M, Egan M. Qualitative Process Evaluation from a Complex Systems Perspective: A Systematic Review and Framework for Public Health Evaluators. Plos Med (2020) 17(11):e1003368. doi:10.1371/journal.pmed.1003368 23. Haddaway NR, Woodcock P, Macura B, Collins A. Making Literature Reviews More Reliable through Application of Lessons from Systematic Reviews. Conservation Biol (2015) 29(6):1596–605. doi:10.1111/cobi.12541 24. Byrne JA. Improving the Peer Review of Narrative Literature Reviews. Res Integr Peer Rev (2016) 1:12. doi:10.1186/s41073-016-0019-2 10. Rusoja E, Haynie D, Sievers J, Mustafee N, Nelson F, Reynolds M, et al. Thinking about Complexity in Health: A Systematic Review of the Key Systems Thinking and Complexity Ideas in Health. J Eval Clin Pract (2018) 24(3): 600–6. doi:10.1111/jep.12856 25. Hsiu-Fang H, Shannon SE. Three Approaches to Qualitative Content Analysis. Qual Health Res (2005) 15(9):1277. doi:10.1177/1049732305276687 26. Crielaard L, Dutta P, Quax R, Nicolaou M, Merabet N, Stronks K, et al. Social Norms and Obesity Prevalence: From Cohort to System Dynamics Models. Obes Rev (2020) 21(9):e13044. doi:10.1111/obr.13044 11. Forrester JW. Learning through System Dynamics as Preparation for the 21st Century. Syst Dyn Rev (2016) 32(3-4):187–203. doi:10.1002/sdr.1571 y y y 12. Forrester J. FUNDING The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. The authors declare that this study received funding from Canadian Institutes for Health Research (LB/Postdoctoral fellowship) and the College of Medicine, University of REFERENCES Mixed Methods Research in Health Systems. Health Policy Plan (2013) 29(3):323–7. doi:10.1093/heapol/czt019 59. Bradbury H, Reason P, Publishing E, Ebsco V. The Sage Handbook of Action Research : Participative Inquiry and Practice. 2nd ed. London: SAGE (2012). p. ed2008. 39. Owen B, Brown AD, Kuhlberg J, Millar L, Nichols M, Economos C, et al. Understanding a Successful Obesity Prevention Initiative in Children under 5 from a Systems Perspective. PLoS One (2018) 13(3):e0195141. doi:10.1371/ journal.pone.0195141 60. B Israel, E Eng, A Schultz, E Parker, editors. Methods in Community-Based Participatory Research in Health. San Fransico, CA: Jossey-Bass (2005). 40. Parmar PK, Rawashdah F, Al-Ali N, Abu Al Rub R, Fawad M, Al Amire K, et al. Integrating Community Health Volunteers into Non-communicable Disease Management Among Syrian Refugees in Jordan: a Causal Loop Analysis. BMJ Open (2021) 11(4):e045455. doi:10.1136/bmjopen-2020-045455 61. Sturmberg J. Without Systems and Complexity Thinking There Is No Progress - or Why Bureaucracy Needs to Become Curious. Int J Health Pol Manag (2021) 10(5):227–80. doi:10.34172/ijhpm.2020.45 62. Baum F, Townsend B, Fisher M. Creating Political Will for Action on Health Equity: Practical Lessons for Public Health Policy Actors. Int J Health Pol Manag (2020) x:1–14. doi:10.34172/ijhpm.2020.233 41. Gerritsen S, Harré S, Rees D, Renker-Darby A, Bartos AE, Waterlander WE, et al. Community Group Model Building as a Method for Engaging Participants and Mobilising Action in Public Health. Int J Environ Res Public Health (2020) 17(10). doi:10.3390/ijerph17103457 63. Senge P. The Fifth Discipline. New York: Doubleday (2006). 42. Maitland N, Williams M, Jalaludin B, Allender S, Strugnell C, Brown A, et al. Campbelltown - Changing Our Future: Study Protocol for a Whole of System Approach to Childhood Obesity in South Western Sydney. BMC Public Health (2019) 19(1):1699. doi:10.1186/s12889-019-7936-1 64. Mui Y, Ballard E, Lopatin E, Thornton RLJ, Pollack Porter KM, Gittelsohn J. A Community-Based System Dynamics Approach Suggests Solutions for Improving Healthy Food Access in a Low-Income Urban Environment. PLoS ONE (2019) 14:e0216985–13. doi:10.1371/journal.pone.0216985 65. Savona N, Macauley T, Aguiar A, Banik A, Boberska M, Brock J, et al. Identifying the Views of Adolescents in Five European Countries on the Drivers of Obesity Using Group Model Building. Eur J Public Health (2021) 31(2):391–6. doi:10.1093/eurpub/ckaa251 43. Riley T, Hopkins L, Gomez M, Davidson S, Chamberlain D, Jacob J, et al. A Systems Thinking Methodology for Studying Prevention Efforts in Communities. Syst Pract Action Res (2020) 34 (5):555–73. 44. REFERENCES Some Basic Concepts in System Dynamics. Boston: Sloan School of Management, Massachusetts Institute of Technology (2009). 27. Eker NZ, Carnohan S, Davies M. Participatory System Dynamics Modelling for Housing, Energy and Wellbeing Interactions. Build Res Inf (2018) 46(7): 738–54. doi:10.1080/09613218.2017.1362919 Management, Massachusetts Institute of Technology (2009). 13. Baugh Littlejohns L, Baum F, Lawless A, Freeman T. The Value of a Causal Loop Diagram in Exploring the Complex Interplay of Factors that Inﬂuence Health Promotion in a Multisectoral Health System in Australia. Health Res Pol Syst (2018) 16:126. doi:10.1186/s12961-018-0394-x 28. Brereton CF, Jagals P. Applications of Systems Science to Understand and Manage Multiple Inﬂuences within Children’s Environmental Health in Least Developed Countries: A Causal Loop Diagram Approach. Int J Environ Res Public Health (2021) 18(6). doi:10.3390/ijerph18063010 14. Paina L, Bennett S, Ssengooba F, Peters DH. Advancing the Application of Systems Thinking in Health: Exploring Dual Practice and its Management in Kampala, Uganda. Health Res Pol Sys (2014) 12(1):41. doi:10.1186/1478-4505- 12-41 29. Klement RJ. Systems Thinking about SARS-CoV-2. Front Public Health (2020) 8(650):585229. doi:10.3389/fpubh.2020.585229 30. Knai C, Petticrew M, Mays N, Capewell S, Cassidy R, Cummins S, et al. Systems Thinking as a Framework for Analyzing Commercial Determinants of Health. Milbank Q (2018) 96(3):472–98. doi:10.1111/1468-0009.12339 15. Varghese J, Kutty VR, Paina L, Adam T. Advancing the Application of Systems Thinking in Health: Understanding the Growing Complexity Governing December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 12 Baugh Littlejohns et al. CLDs in Public Health Research 31. Osman M, Karat AS, Khan M, Meehan S-A, von Delft A, Brey Z, et al. Health System Determinants of Tuberculosis Mortality in South Africa: a Causal Loop Model. BMC Health Serv Res (2021) 21(1):388. doi:10.1186/s12913-021- 06398-0 47. Allender S, Brown AD, Bolton KA, Fraser P, Lowe J, Hovmand P. Translating Systems Thinking into Practice for Community Action on Childhood Obesity. Obes Rev (2019) 20:179–84. doi:10.1111/obr.12865 48. Brown A, Millar L, Hovmand PS, Kuhlberg J, Love P, Nagorcka-Smith P, et al. Learning to Track Systems Change Using Causal Loop Diagrams. Obes Res Clin Pract (2019) 13:73–4. doi:10.1016/j.orcp.2016.10.210 32. Sahin O, Salim H, Suprun E, Richards R, MacAskill S, Heilgeist S, et al. Developing a Preliminary Causal Loop Diagram for Understanding the Wicked Complexity of the COVID-19 Pandemic. Systems (2020) 8(2):20. doi:10.3390/systems8020020 49. Kim H, Andersen DF. REFERENCES Building Conﬁdence in Causal Maps Generated from Purposive Text Data: Mapping Transcripts of the Federal Reserve. Syst Dyn Rev (2012) 28(4):311–28. doi:10.1002/sdr.1480 33. Urwannachotima N, Hanvoravongchai P, Ansah JP. Sugar-sweetened Beverage Tax and Potential Impact on Dental Caries in Thai Adults: An Evaluation Using the Group Model Building Approach. Syst Res (2019) 36(1): 87–99. doi:10.1002/sres.2546 50. Foster-Fishman PG, Nowell B, Yang H. Putting the System Back into Systems Change: a Framework for Understanding and Changing Organizational and Community Systems. Am J Community Psychol (2007) 39(3-4):197–215. doi:10.1007/s10464-007-9109-0 34. Bradley DT, Mansouri MA, Kee F, Garcia LMT. A Systems Approach to Preventing and Responding to COVID-19. EClinicalMedicine (2020) 21: 100325. doi:10.1016/j.eclinm.2020.100325 51. Meadows D. Leverage Points: Places to Intervene in a System. Norwich, Vermont: Donella Meadows Institute (1999). 35. Bensberg M, Joyce A, Wilson E. Building a Prevention System: Infrastructure to Strengthen Health Promotion Outcomes. Int J Environ Res Public Health (2021) 18(4):1618. doi:10.3390/ijerph18041618 52. Mabry PL, Marcus SE, Clark PI, Leischow SJ, Méndez D. Systems Science: a Revolution in Public Health Policy Research. Am J Public Health (2010) 100(7):1161–3. doi:10.2105/AJPH.2010.198176 53. Finegood D. The Complec Science of Obesity. In: J Cawley, editor. Handbook of the Social Science of Obesity. Oxford: Oxford University Press (2011). 36. Burrell M, White AM, Frerichs L, Funchess M, Cerulli C, DiGiovanni L, et al. Depicting "the System": How Structural Racism and Disenfranchisement in the United States Can Cause Dynamics in Community Violence Among Males in Urban Black Communities. Soc Sci Med (2021) 272:113469. doi:10.1016/ j.socscimed.2020.113469 54. Hawe P, Shiell A, Riley T. Theorising Interventions as Events in Systems. Am J Community Psychol (2009) 43(3-4):267–76. doi:10.1007/s10464-009-9229-9 55. Ulrich W. Beyond Methodology Choice: Critical Systems Thinking as Critically Systemic Discourse. J Oper Res Soc (2003) 54(4):325–42. doi:10.1057/palgrave.jors.2601518 37. Clarke B, Kwon J, Swinburn B, Sacks G. Understanding the Dynamics of Obesity Prevention Policy Decision-Making Using a Systems Perspective: A Case Study of Healthy Together Victoria. PLoS ONE (2021) 16(1):e0245535. doi:10.1371/journal.pone.0245535 56. Richardson G. Feedback Thought in Social Science and Systems Theory. Waltham, MA: Pegasus Communications, Inc. (1999). 57. Lewin K. Action Research and Minority Problems. J Soc Issues (1946) 2:34–46. doi:10.1111/j.1540-4560.1946.tb02295.x 38. Jalali MS, Rahmandad H, Bullock SL, Lee-Kwan SH, Gittelsohn J, Ammerman A. Dynamics of Intervention Adoption, Implementation, and Maintenance inside Organizations: the Case of an Obesity Prevention Initiative. Soc Sci Med (2019) 224:67–76. doi:10.1016/j.socscimed.2018.12.021 58. Ozawa S, Pongpirul K. 10 Best Resources on ... PHR is edited by the Swiss School of Public Health (SSPH+) in a partnership with the Association of Schools of Public Health of the European Region (ASPHER)+ REFERENCES Swierad E, Huang TT, Ballard E, Flórez K, Li S. Developing a Socioculturally Nuanced Systems Model of Childhood Obesity in Manhattan’s Chinese American Community via Group Model Building. J Obes (2020) 2020: 4819143. doi:10.1155/2020/4819143 Copyright © 2021 Baugh Littlejohns, Hill and Neudorf. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 45. Araz O, Wilson F, Stimpson J. Complex Systems Modeling for Evaluating Potential Impact of Trafﬁc Safety Policies: a Case on Drug-Involved Fatal Crashes. Ann Operations Res (2020) 291(1-2):37–58. doi:10.1007/s10479-018- 2961-5 46. Hassmiller Lich K, Urban JB, Frerichs L, Dave G. Extending Systems Thinking in Planning and Evaluation Using Group Concept Mapping and System Dynamics to Tackle Complex Problems. Eval Program Plann (2017) 60: 254–64. doi:10.1016/j.evalprogplan.2016.10.008 PHR is edited by the Swiss School of Public Health (SSPH+) in a partnership with the Association of Schools of Public Health of the European Region (ASPHER)+ December 2021 \| Volume 42 \| Article 1604352 Public Health Reviews \| Owned by SSPH+ \| Published by Frontiers 13
https://openalex.org/W1895742341	https://breast-cancer-research.biomedcentral.com/counter/pdf/10.1186/bcr3293	English	null	Comparison of recall, biopsy, and cancer detection rates in the Southern Derbyshire screening programme between 2006 and 2009 using hard-copy mammography and in 2009 to 2012 following the full introduction of soft-copy reporting	Breast cancer research	2,012	cc-by	17,759	O1 Utility of MRI in low and low to moderate density breasts with invasive lobular carcinoma O1 Utility of MRI in low and low to moderate density breasts with invasive lobular carcinoma B Rengabashyam, N Sharma, B Dall g y The Leeds Teaching Hospitals NHS Trust, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):O1 Objective: To determine the feasibility of excluding MRI from the pre- operative diagnostic pathway of invasive lobular carcinoma (ILC) in women with low and low to moderate density breasts on mammography. aOne unable to biopsy due to technical factors, one failure to sample node-B1. bOne failure to sample node-B1. Methods: A total of 179 cases of ILC were diagnosed between 2009 and 2012. Forty-eight cases were identified as low and low to moderate density breasts. The study group includes 32 cases who underwent MRI. Parameters scrutinised include size and number of lesions on mammography, ultrasound and MRI, second-look ultrasound, type of surgery, further surgery and histology. Table 2(abstract O2) IDC ILC US sensitivity 0.73(0.51 to 0.88) 0.73(0.48 to 0.89) US specificity 0.83 (0.68 to 0.92) 0.93 (0.8 to 0.98) US positive predictive value 0.70 (0.48 to 0.86) 0.82 (0.55 to 0.95) There were no statistically significant differences between the two groups. gy Results: Twenty-nine cases had low to moderate density breasts and three had purely low density breasts. Average age of women was 64. Size of lesions ranged between 2 and 50 mm with an average of 20.14 mm. In 25/32 cases (78.12%) conventional imaging matched MRI. MRI identified additional disease in 7/32 (21.8%). This was predominantly in the form of satellites around the index lesion resulting in multifocality in 6/7. Four resulted appropriately in mastectomy. Two led to wider WLE appropriately. In one case, multicentric disease was correctly detected and subjected to mastectomy. Second-look ultrasound was recommended in 4/7 cases. All these cases had low to moderate density breasts on mammography and 6/7 cases measured more than 15 mm in size. Ultrasound matched MRI in one mammographically occult case and was subjected to appropriate WLE. In two cases there was much more disease than anticipated from conventional imaging and MRI (6.25%). lobular cancer (ILC) and invasive ductal cancer (IDC) were reviewed. Preoperative US-guided needle biopsy sampling was performed on abnormal lymph nodes (LN). Conclusion: The previous literature on this topic is inconclusive. Some authors have suggested axillary ultrasound in ILC may be less accurate than in IDC, with a higher false-negative axillary assessment rate. © 2012 various authors, licensee BioMed Central Ltd. All articles published in this supplement are distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. British Society of Breast Radiology Annual Scientific Meeting 2012 These abstracts are available online at http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Open Access MEETING ABSTRACTS Open Access ORAL PRESENTATIONS Table 1(abstract O2) IDC ILC P value Node-positive BC 34% (22/65) 32% (21/65) Preoperative US detection of node-positive disease 54% (12/22) 52% (11/21) 0.5 Abnormal US LN appearance 37% (24/65)a 26% (17/65)b 0.2 False negative rate for axillary US biopsy 18% (4/22) 12% (2/16) 0.5 aOne unable to biopsy due to technical factors, one failure to sample node-B1. bOne failure to sample node-B1. O1 Utility of MRI in low and low to moderate density breasts with invasive lobular carcinoma B Rengabashyam, N Sharma, B Dall The Leeds Teaching Hospitals NHS Trust, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):O1 O1 Utility of MRI in low and low to moderate density breasts with invasive lobular carcinoma Another study concluded that axillary US accuracy rates in ILC were comparable with previous published studies of IDC, used FNA in all cases. We specifically compared accuracy rates of preoperative axillary staging between ILC and IDC in own institution, with 14G needle biopsy as the procedure of choice to sample abnormal nodes. We found that there is no statistical difference in accuracy in US axillary staging between ILC and IDC. Conclusion: Even in low and low to moderate density breasts where mammography has a higher exclusion value, MRI identified additional disease in 21.8% (7/32). O2 Is ultrasound axillary staging less accurate in invasive lobular breast cancer than in ductal breast cancer? P Sankaye, G Porter, J Steel, S Doyle Plymouth Hospitals NHS Trust, Plymouth, UK Breast Cancer Research 2012, 14(Suppl 1):O2 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 O2 Is ultrasound axillary staging less accurate in invasive lobular breast cancer than in ductal breast cancer? O3 Positive predictive value of MRI vacuum biopsies in the diagnosis of nonmass-like lesions of the breast WL Teh, EK Papantoniou, F Ng Northwick Park Hospital, Harrow, UK Breast Cancer Research 2012, 14(Suppl 1):O3 O3 Positive predictive value of MRI vacuum biopsies in the diagnosis of nonmass-like lesions of the breast WL Teh, EK Papantoniou, F Ng Northwick Park Hospital, Harrow, UK Breast Cancer Research 2012, 14(Suppl 1):O3 Objective: To identify whether axillary US is less accurate in invasive lobular breast cancer than in ductal breast cancer. Methods: Randomised cohorts of screening and symptomatic patients were retrospectively identified from histology records of 2010/11. Axillary US of 65 patients with primary breast cancers (BC) from each group of invasive Objective: To evaluate the positive predictive value of MRI scoring for malignant mass-like (ML) and nonmass-like (NML) lesions based on the Page 2 of 14 Page 2 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Results: In the 3-year screening period, 199,082 women were screened. A total of 1,020 women had SDC and 692 further developed IC following screening. Of the 692 IC, 57.8% (391) were TI, 17.7% (120) were FN, 10% (68) were OCC and 14.5% were UCC. After at least 10 years of follow-up, the long-term survival rate (all-cause) for SDC was 81.6%, overall for IC was 72.4% (OR 1.67, P < 0.001), TI was 77.5% (OR 1.00, P = 0.99), FN was 55% (OR 2.36, P < 0.001), OCC was 54.4% (OR 3.17, P < 0.001) and UCC was 87.8% (OR 0.61, P = 0.19). BI-RADS descriptors. To identify MRI characteristics of nonmass lesions which predict malignancy for invasive and non-invasive cancers. Methods: Retrospective analysis of 486 MRI-guided vacuum biopsies performed at Northwick Park Hospital between April 2006 and November 2011. Each lesion was categorised according to BI-RADS lexicon (ML vs. NML lesions) and time-enhancement curves, and given an overall score of MRI 1 to 5 according to overall level of suspicion for malignant disease where MRI 4 and 5 are considered suspicious or diagnostic for malignancy. Biopsy and surgical histology results obtained. Conclusion: Overall survival of IC is significantly different to SDC. However, SDC and TI were not statistically significantly different. FN and OCC had significantly worse long-term survival. Further research is required to identify the underlying cause of poor survival of FN and OCC. O6 O6 Gamma probe and ultrasound-guided percutaneous localisation of the sentinel lymph node in breast cancer patients P Whelehan, A Evans, S Vinnicombe, D Brown, D McLean University of Dundee, UK Breast Cancer Research 2012, 14(Suppl 1):O6 Conclusion: No specific BI-RADS feature predicts for invasive disease in NML lesions. Segmental enhancement is the most common MRI appearance for DCIS. Time-enhancement curves are unhelpful in predicting malignancy in NML lesions. Introduction: A major reason for failure to diagnose axillary lymph node metastasis preoperatively in breast cancer patients is that needle biopsy may not target the sentinel node (SLN). We aimed to address this by testing the accuracy with which we could identify and target the sentinel lymph node percutaneously under combined radioisotope and ultrasound guidance. O4 Use of MRI fusion second-look ultrasound in breast cancer: can MRI US fusion reduce the need for MRI-guided biopsy? SE McWilliams Guy’s and St Thomas’ NHS Foundation Trust, London, UK Breast Cancer Research 2012, 14(Suppl 1):O4 Methods: Ethical approval was obtained. In 48 patients scheduled for surgical sentinel lymph node biopsy (SLNB), following injection of radioisotope, one of three radiologists used a gamma probe in tandem with ultrasound guidance to identify a SLN prior to surgery and mark it with a localising wire. The patients then proceeded to surgical SLNB, guided by radioisotope and blue dye. The surgeon noted whether the wire had correctly marked a SLN. Introduction: Breast MRI is being increasingly used in breast cancer to look for extent of disease, in high-risk screening and in the dense breast. Frequently incidental lesions are detected on MRI that require second-look ultrasound or stereotactic biopsy. Incidental MRI-detected lesions may be occult on conventional imaging and require MRI-guided biopsy. We describe our experiences with US MRI fusion to try and reduce the need for MRI-guided biopsy. y Results: The SLN was correctly localised in 75% of patients (36 of 48; 95% CI = 63 to 87%). A learning curve for the three operators was observed, with a rate of correct localisation in the first five procedures performed by each operator of 67% (10 of 15) rising to 79% in the subsequent 33 (26 of 33; 95% CI = 64 to 93%). Interoperator variations in accuracy were evident, with a success rate of 83% in the most experienced of the three. O6 Methods: At our institution we have introduced an MRI breast biopsy service which is time consuming and expensive. We looked at 10 patients with MRI US fusion technology on our new Hitachi US MRI scanner to see whether lesions were easier to identify combining the MRI and US images, enabling US biopsy to be performed. Conclusion: Percutaneous SLN localisation using combined radioisotope and ultrasound guidance is feasible. Use of this method to guide needle biopsy of the axilla could increase the preoperative diagnosis rate for axillary lymph node metastases in breast cancer patients. Results: Ten patients with a known breast cancer had a further incidental lesion seen on MRI. The patients had an additional supine series on contrast MRI images in addition to the standard prone protocol. The supine images were loaded on the US machine and enabled confident detection of the lesion on US in nine out of 10 patients. P1 Breast Cancer Research 2012, 14(Suppl 1):P1 Abstracts P1 - P16 are educational and not included for publication online Abstracts P1 - P16 are educational and not included for publication online O5 Long-term survival of interval breast cancers in breast cancer screening in Wales YF Fong1, J Evans2, D Brookes2, K Gower Thomas1,2 1Royal Glamorgan Hospital, Llantrisant, UK; 2Breast Test Wales, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):O5 POSTER PRESENTATIONS P1 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P1 Abstracts P1 - P16 are educational and not included for publication online P2 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P2 Abstracts P1 - P16 are educational and not included for publication online P3 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P3 Abstracts P1 - P16 are educational and not included for publication online Conclusion: Using MRI US fusion with one additional MRI series of supine images reduces the need for MRI-guided biopsy enabling US biopsy to be performed, which is cheaper, quicker and more patient acceptable. Long-term survival of interval breast cancers in breast cancer screening in Wales 1* 2 2 1 2 P2 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P2 Abstracts P1 - P16 are educational and not included for publication online P3 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P3 Abstracts P1 - P16 are educational and not included for publication online YF Fong1, J Evans2, D Brookes2, K Gower Thomas1,2 1Royal Glamorgan Hospital, Llantrisant, UK; 2Breast Test Wales, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):O5 P2 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P2 Introduction: Breast Test Wales is part of the NHS Breast Cancer Screening Programme, and oversees the screening programme in Wales. It is successful in identifying asymptomatic cancers; however, interval cancers (IC) still occur between screenings. We aim to evaluate the overall long-term survival of IC and compare that with screen-detected cancers (SDC). Abstracts P1 - P16 are educational and not included for publication online O2 Is ultrasound axillary staging less accurate in invasive lobular breast cancer than in ductal breast cancer? Results: A total of 291 ML and 152 NML lesions, of which there were 150 cancers diagnosed. Positive predictive value of MRI characteristic for malignant mass lesions is 70%. Positive predictive value of MRI characteristic for nonmass lesions is 57%. Segmental enhancement is the most common MRI morphology found in 45% DCIS. No specific features predict for invasive disease in NML lesions. Time-enhancement curves were mainly type 2 (44.6%) and type 3 (52.7%) in malignant ML lesions and unhelpful in predicting malignancy in NML lesions (72% type 2 and 59% type 3 were benign). P3 b Page 3 of 14 P12 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P12 Abstracts P1 - P16 are educational and not included for publication online P13 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P13 Abstracts P1 - P16 are educational and not included for publication online P14 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P14 Abstracts P1 - P16 are educational and not included for publication online P15 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P15 Abstracts P1 - P16 are educational and not included for publication online P16 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P16 Abstracts P1 - P16 are educational and not included for publication online P4 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P4 Abstracts P1 - P16 are educational and not included for publication online P5 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P5 Abstracts P1 - P16 are educational and not included for publication online P6 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P6 Abstracts P1 - P16 are educational and not included for publication online P7 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P7 Abstracts P1 - P16 are educational and not included for publication online P8 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P8 Abstracts P1 - P16 are educational and not included for publication online P9 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P9 Abstracts P1 - P16 are educational and not included for publication online P10 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P10 Abstracts P1 - P16 are educational and not included for publication online P11 Abstracts P1 - P16 are educational and not included for publication online Breast Cancer Research 2012, 14(Suppl 1):P11 Abstracts P1 - P16 are educational and not included for publication online P12 Abstracts P1 - P16 are educational and not included for publication online P17 Breast density measurement for personalised screening JC Sergeant1, S Musa2, M Wilson3, DG Evans3, A Howell3, SM Astley1 1Institute of Population Health, University of Manchester, UK; 2Manchester Medical School, University of Manchester, UK; 3Nightingale Centre and Genesis Prevention Centre, University Hospital of South Manchester, Manchester, UK Breast Cancer Research 2012, 14(Suppl 1):P17 Introduction: Breast density is both a modifiable risk factor for breast cancer and an indicator of the sensitivity of mammography. P3 b Methods: Within BTW, SDC between 1998 and 2001 and IC occurring between 1998 and 2004 but screened between 1998 and 2001 were identified. IC were classified into true interval (TI), false negative (FN), occult (OCC) and unclassified (UCC). BTW receives notification of death of all women that underwent screening. The long-term survival rate was calculated from the date of initial screening and the date of death. Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 P20 P20 Impact of full-field digital mammography on diagnostic work-up and surgical management of mammographic microcalcification SM Bundred1, J Zhou1, S Whiteside1, J Morris2, NJ Bundred2 1University Hospital South Manchester, Manchester, UK; 2University of Manchester, UK Breast Cancer Research 2012, 14(Suppl 1):P20 Impact of full-field digital mammography on diagnostic work-up and surgical management of mammographic microcalcification SM Bundred1, J Zhou1, S Whiteside1, J Morris2, NJ Bundred2 p y Methods: Outcome data from 62,599 routine screening mammograms taken in Greater Manchester between January 2010 and March 2012 were gathered from screening records. Mean recall, biopsy/cytology and cancer detection rates were calculated for analogue and digital screening and compared. After complete conversion to digital screening, the same three measures were obtained on a monthly basis over 2 years. The data were analysed to identify any trends in the measures. 1University Hospital South Manchester, Manchester, UK; 2University of Manchester, UK Introduction: Full-field digital mammography (FFDM) increases detection of benign and malignant calcified breast lesions. Accurate nonoperative diagnosis of malignant impalpable breast lesions minimises numbers of therapeutic surgical procedures. Correct diagnosis of malignant mammo- graphic microcalcification (MM) is important because upgraded lesions require repeat surgical procedures in 57% of cases (NHSBSP Audit 2012). Results: The digital mammography recall rate was 4.69% (2,505/53,444), with analogue significantly lower at 3.58% (328/9,155) (P < 0.001). The digital biopsy/cytology rate was 1.88% (1,005/53,444), with analogue significantly lower at 1.49% (136/9,155) (P = 0.01). The cancer detection rate (per 1,000 screens) was 6.74 (56/9,155) for digital mammography and 6.15 (358/53,444) for analogue; however, this difference was not significant (P = 0.53). The recall rate showed a strong positive correlation with time (r = 0.71, P < 0.001) and the biopsy rate showed a weak positive correlation (r = 0.52 P = 0.011). Cancer detection showed a weak negative correlation with time (r = -0.567 P = 0.05). p g p Methods: Screening and symptomatic women with MM (n = 1,479) were reviewed to determine the impact of FFDM (imaging with FFDM only since April 2010) on the positive predictive value (PPV), diagnostic accuracy and surgical management of MM. Demographic information, preoperative and postoperative diagnosis and number of surgical procedures were recorded for Group 1 (August 2007 to March 2010: n = 711) and Group 2 (April 2010 to May 2011: n = 768). P20 Conclusion: The introduction of digital mammography resulted in higher recall and biopsy rates, but no change in cancer detection. Recall and biopsy rates, whilst still well within recommended levels, appear to be increasing since the introduction of digital mammography; this indicates a need for regular monitoring. y Results: Reduction in PPV of biopsy for MM was observed (Group 1, 42.6%: Group 2, 32.7%; P < 0.0001). Correct or concordant nonoperative diagnosis increased with FFDM (Group 1, 89% vs. Group 2, 95%; P < 0.0001) and was achieved more often at first attempt (Group 1, 80.6% vs. Group 2, 89.5%; P < 0.0001). More lesions under 5 mm were biopsied using FFDM (Group 1, 15%; Group 2, 20.4%; P = 0.008). Accurate preoperative diagnosis of malignancy permitted single-stage surgery in 77.4% Group 2 versus 67.9% Group 1 (P = 0.017). For DCIS cases, similar first-line mastectomy rates were observed (Group 1, 30.3% vs. Group 2, 33.3%, P = NS). Fewer B3/4 lesions upgraded at surgery (Group 1, 48.7% vs. Group 2, 20%; P = 0.011). Changes in recall rate, biopsy rate and cancer detection since the introduction of digital mammography Changes in recall rate, biopsy rate and cancer detection since the introduction of digital mammography * 2. Coleman MP, et al: Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. The Lancet 2010, 377:127-138. 2. Coleman MP, et al: Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. The Lancet 2010, 377:127-138. 2. Coleman MP, et al: Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. The Lancet 2010, 377:127-138. TG Fyall, CRM Boggis, SM Astley, JC Sergeant, J Morris, S Whiteside, M Wilson The University of Manchester, UK Breast Cancer Research 2012, 14(Suppl 1):P18 Introduction: Analogue mammography is gradually being replaced by digital mammography for breast screening. Digital mammography has been shown to be at least as good as analogue mammography at cancer detection, but the picture regarding other outcomes such as recall rate and biopsy rate is less clear. References 1. Office for National Statistics: Statistical Bulletin: Cancer Survival in England: Patients Diagnosed 2005-2009 and Followed up to 2010-2011 London: Office for National Statistics 2011. 1. Office for National Statistics: Statistical Bulletin: Cancer Survival in England: Patients Diagnosed 2005-2009 and Followed up to 2010-2011 London: Office for National Statistics 2011. P18 P18 Changes in recall rate, biopsy rate and cancer detection since the introduction of digital mammography TG Fyall, CRM Boggis, SM Astley, JC Sergeant, J Morris, S Whiteside, M Wilson The University of Manchester, UK Breast Cancer Research 2012, 14(Suppl 1):P18 P3 b Reliable measurement in the screening population could enable personalised screening to maximise early detection of cancer. In the Predicting Risk of Cancer at Screening (PROCAS) study, we are investigating different approaches to measuring breast density, and we present comparative data from one subjective method and two automated volumetric methods. j Methods: The screening mammograms of 4,109 women enrolled in PROCAS were visually assessed independently by two experienced film readers, who recorded their estimates of percentage density on visual analogue scales (VAS). The mammograms were also processed by Quantra™ (Hologic Inc.) and Volpara (Matakina Technology Ltd). They were ranked according to density by each method, and the top 10% and 1% compared. Results: Of the 617 mammograms ranked in the most dense 10% by at least one method, only 127 were high density by all three methods. The overlap between the two volumetric methods was 214; between VAS and Quantra™it was 195 and between VAS and Volpara it was 147. For the Page 4 of 14 Page 4 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Results: In the 3-year screening period, 199,082 women were screened. A total of 1,712 cancers were diagnosed. In total, 1,546 had data available for calculating the NPI. The overall 5-year and 10-year survival was 94% and 82%. See Table 1. 48 mammograms in the top 1% by density, the overlaps were 13, 9 and 7, respectively. 48 mammograms in the top 1% by density, the overlaps were 13, 9 and 7, respectively. p y Conclusion: The lack of overlap between the methods was surprising and has serious implications for the implementation of personalised screening. The optimum measure of breast density in digital mammograms has yet to be identified; different methods may yield the strongest links with cancer risk and sensitivity of mammography, and further research is needed to elucidate these relationships. Conclusion: The overall 5-year and 10-year survival (all cause) has improved even when compared with specific breast cancer survival of recent published data. Our data provide a reference for updating all-cause survival of women diagnosed with breast cancers within a screened population. References P24 g S Taylor-Phillips1, A Clarke1, M Wheaton2, O Kearins3, M Wallis4 1The University of Warwick, Coventry, UK; 2University Hospital Coventry and Warwickshire, Coventry, UK; 3West Midlands Quality Assurance Reference Centre, Birmingham, UK; 4Addenbrookes Hospital, Cambridge, UK Breast Cancer Research 2012, 14(Suppl 1):P24 y g Breast Cancer Research 2012, 14(Suppl 1):P26 Introduction: Currently, breast screening is implemented with X-ray mammography, but new technology such as digital breast tomosynthesis (DBT) may provide significant advantages because it produces three- dimensional breast images with high spatial resolution. To aid assessment of the cost-effectiveness of DBT, we have compared reading times for DBT images and full-field digital mammograms (FFDM) of the same women. Introduction: Currently, breast screening is implemented with X-ray mammography, but new technology such as digital breast tomosynthesis (DBT) may provide significant advantages because it produces three- dimensional breast images with high spatial resolution. To aid assessment of the cost-effectiveness of DBT, we have compared reading times for DBT images and full-field digital mammograms (FFDM) of the same women. Methods: Four consultant radiologists were timed by stopwatch reading FFDM images and then DBT images, with a total of 119 timings of two view cases for each imaging modality. The mammograms used were from women recalled following a screening mammogram and from women who are screened annually due to a family history of breast cancer. The time taken to display the images and to report on them was excluded from the analysis. Introduction: Interpreting mammograms is a repetitive visual task, which may cause fatigue. Current practice in England for digital mammography is that both readers examine batches of mammograms in the same order as one another. This research examines whether there is a potential benefit in the two readers examining the cases in a different order to one another, to ameliorate any effects of fatigue at the system level. Methods: Four consultant radiologists were timed by stopwatch reading FFDM images and then DBT images, with a total of 119 timings of two view cases for each imaging modality. The mammograms used were from women recalled following a screening mammogram and from women who are screened annually due to a family history of breast cancer. The time taken to display the images and to report on them was excluded from the analysis. Methods: NBSS records at one screening centre for 4 years were examined (2007 to 2010, seven readers, >170,000 cases). Role of targeted reassurance breast ultrasound in women with symptoms of a breast lump and normal clinical examination R Gallagher, H Burrell, E Cornford these two techniques in cases of proven invasive lobular carcinoma (ILC). Methods: Thirty-two consecutive cases of ILC underwent 3T breast MR imaging as part of routine staging - retrospective analysis was undertaken. Diffusion-weighted imaging using three b values (0, 500 and 1,000 seconds/mm2) was performed prior to dynamic contrast-enhanced imaging in each case. Post processing included obtaining maximum intensity projections and multiplanar reconstruction. Regions of interest (ROI) were placed over the index (presenting) lesion. In cases of multifocality/multicentricity, the most conspicuous enhancing lesion was interrogated. Enhancement kinetics were evaluated from the dynamic contrast-enhanced images: the percentage of signal intensity increase within the first 2 minutes after administrating contrast agent relative to pre-contrast signal intensity was calculated and the morphology of the curve for 6 minutes after contrast administration was analysed. The same size of ROI was placed over the lesion site on the DWI and the apparent diffusion coefficient (ADC) map was calculated. Introduction: Ultrasound has a well-established role in the triple assessment of breast lumps. The purpose of this study was to review the value of targeted breast ultrasound in women presenting with symptoms of a breast lump but normal clinical examination. Methods: The records of all patients presenting to the symptomatic clinic with a breast lump but with a normal clinical examination over a 6-month period were reviewed. In our institution all such patients undergo targeted ultrasound of the symptomatic area within the breast; and mammography if aged 40 years or over. The results of the ultrasound examination, mammography and histology from any needle biopsy were reviewed. Results: Seven hundred and ninety-nine women were included (mean age = 39, age range 15 to 84). One hundred and two (13%) had an abnormality detected on ultrasound, 92 (90%) of which were at the same site as the patient’s symptoms. Thirty-two patients (4% of the study group) had solid lumps and 70 (9%) had cysts. Five cancers were detected, four of which were at the site of the patient’s symptoms and one of which was an incidental area of DCIS visible on mammography only. Three of the four patients with invasive cancer had a normal mammogram. Methods: The records of all patients presenting to the symptomatic clinic with a breast lump but with a normal clinical examination over a 6-month period were reviewed. P26 A comparison of reading times in full-field digital mammography and digital breast tomosynthesis 1* 2 3 2 4 4 Breast Cancer Research 2012, 14(Suppl 1):P23 SJ Connor1, YY Lim2, C Tate3, H Entwistle2, J Morris4, S Whiteside4, J Sergeant3, M Wilson2, U Beetles2, C Boggis2, F Gilbert5, S Astley3 1Manchester Medical School, University of Manchester, UK; 2Nightingale Breast Centre, University Hospital of South Manchester, Manchester, UK; 3Institute of Population Health, University of Manchester, UK; 4Department of Medical Statistics, University of Manchester, UK; 5School of Clinical Medicine, University of Cambridge, UK P22 3T MR imaging: diffusion-weighted and dynamic contrast-enhanced - relationship of apparent diffusion coefficient value and maximum percentage enhancement in invasive lobular carcinoma of the breast ID Lyburn1,2, HC Russell1,2, J Searle2, C Croucher1,2, DO Hall2 1Thirlestaine Breast Centre, Cheltenham, UK; 2Cheltenham Imaging Centre, Cheltenham, UK Breast Cancer Research 2012, 14(Suppl 1):P22 Breast Cancer Research 2012, 14(Suppl 1):P22 Breast Cancer Research 2012, 14(Suppl 1):P22 P23 Abstract withdrawn P23 Abstract withdrawn P26 A comparison of reading times in full-field digital mammography and digital breast tomosynthesis SJ Connor1, YY Lim2, C Tate3, H Entwistle2, J Morris4, S Whiteside4, J Sergeant3, M Wilson2, U Beetles2, C Boggis2, F Gilbert5, S Astley3 1Manchester Medical School, University of Manchester, UK; 2Nightingale Breast Centre, University Hospital of South Manchester, Manchester, UK; 3Institute of Population Health, University of Manchester, UK; 4Department of Medical Statistics, University of Manchester, UK; 5School of Clinical Medicine, University of Cambridge, UK B t C R h 2012 14(S l 1) P26 P19 The Nottingham Prognostic Index: 5-year and 10-year survival data for all-cause survival within a screened population YF Fong1, J Evans2, D Brookes2, K Gower Thomas1,2 1Royal Glamorgan Hospital, Llantrisant, UK; 2Breast Test Wales, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P19 The Nottingham Prognostic Index: 5-year and 10-year survival data for all-cause survival within a screened population 1* 2 2 1 2 g y p p Conclusion: Nonoperative work-up of MM using FFDM reduced second therapeutic procedures for MM, decreased upgrade of B3/4 lesions at diagnostic surgery, but increased benign nonoperative biopsies for MM. Introduction: The Nottingham Prognostic Index (NPI) is an accepted prognostication tool in the management of breast cancers. The latest overall 5-year and 10-year breast cancer survival has been reported to be 85% [1] and 77% [2]. We applied the NPI to breast cancers diagnosed within Breast Test Wales and present survival data in each NPI category. p g y Methods: All women with screen-detected and interval cancers having had screening between 1998 and 2001 were included. The NPI for each cancer was calculated with the size, nodal status and grade of the primary tumour. Survival data (all cause) was calculated after 10 years of follow-up. Table 1(abstract P19) NPI category Number of cases 5-year survival (%) 10-year survival (%) 1 322 (20.8%) 97 89 2 537 (34.7%) 93 84 3 536 (34.7%) 90 77 4 151 (9.8%) 88 73 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Page 5 of 14 Page 5 of 14 Conclusion: There may be patterns in performance with time since a break. Further research is needed to ascertain whether these patterns remain present in a larger more controlled sample, and whether changing case order could improve overall performance. The Changing case Order to Optimise patterns of Performance in Screening (CO-OPS) randomised controlled trial will be begin recruitment in England soon to address these questions. P25 Role of targeted reassurance breast ultrasound in women with symptoms of a breast lump and normal clinical examination R Gallagher, H Burrell, E Cornford Nottingham Breast Institute, Nottingham, UK Breast Cancer Research 2012, 14(Suppl 1):P25 Introduction: In breast MR imaging, diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) studies are being increasingly utilised in assessing malignancy. We investigated potential relationships between Role of targeted reassurance breast ultrasound in women with symptoms of a breast lump and normal clinical examination R Gallagher, H Burrell, E Cornford Nottingham Breast Institute, Nottingham, UK Breast Cancer Research 2012, 14(Suppl 1):P25 Role of targeted reassurance breast ultrasound in women with symptoms of a breast lump and normal clinical examination R Gallagher, H Burrell, E Cornford In our institution all such patients undergo targeted ultrasound of the symptomatic area within the breast; and mammography if aged 40 years or over. The results of the ultrasound examination, mammography and histology from any needle biopsy were reviewed. Results: Seven hundred and ninety-nine women were included (mean age = 39, age range 15 to 84). One hundred and two (13%) had an abnormality detected on ultrasound, 92 (90%) of which were at the same site as the patient’s symptoms. Thirty-two patients (4% of the study group) had solid lumps and 70 (9%) had cysts. Five cancers were detected, four of which were at the site of the patient’s symptoms and one of which was an incidental area of DCIS visible on mammography only. Three of the four patients with invasive cancer had a normal mammogram. Results: Age range, 34 to 74 years; index lesion size, 4 to 78 mm. For type 2 and type 3 curves, signal increase was 128 to 306%; ADC mean 1.29 × 10-3 mm2/second. In type 1 curves, signal increase was 82 to 93%; ADC mean 0.85 × 10-3 mm2/second. In four cases (with type 1 or type 2 curves) no lesions with restricted diffusion were apparent. pp Conclusion: The percentage signal intensity increase on DCE is higher in cases with measurable ADC values than without. ILC with type 1 or type 2 curves may not be visible on DWI. Conclusion: Ultrasound is a useful diagnostic tool in patients who present with a breast lump, but who have a normal clinical examination. In this study a small number of cancers were detected. The value of a normal or benign ultrasound in reassuring both patient and clinician is more difficult to quantify. Structured evaluation of the effectiveness of an interactive tool for developing interpretation skills in mammography A Williams1, KJ Litton2, A Borley3, P Young2 1School of Medicine, Cardiff University, Cardiff, UK; 2Cardiff and Vale Breast Centre, University Hospital Llandough, Cardiff, UK; 3Velindre Hospital, Velindre NHS Trust, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P30 1St Georges Hospital NHS Trust, London, UK; 2The Royal Surrey County Hospital NHS Foundation Trust, Guildford, UK; 3University of York, UK; 4University of Edinburgh, UK; 5University of Middlesex, UK; Breast Cancer Research 2012, 14(Suppl 1):P30 Introduction: Neoadjuvant chemotherapy is an effective treatment in patients with locally advanced breast cancer. It is being used more frequently and imaging is used to monitor disease response. Imaging response in the axilla was correlated with the final breast histology. Introduction: Neoadjuvant chemotherapy is an effective treatment in patients with locally advanced breast cancer. It is being used more frequently and imaging is used to monitor disease response. Imaging response in the axilla was correlated with the final breast histology. Methods: Patients who had undergone primary chemotherapy, pre and post chemotherapy MRI and surgery from 2009 to present were included. The MRI and histology reports were reviewed. Breast Cancer Research 2012, 14(Suppl 1):P28 Introduction: Lesion Zoo, a computer-based training tool, was developed to give trainee radiologists access to a wide range of possible mammo- graphic appearances. Users describe lesions using terms from the BI-RADS lexicon and are given feedback on how their descriptions compare with those of by a panel of experts. Methods: Patients who had undergone primary chemotherapy, pre and post chemotherapy MRI and surgery from 2009 to present were included. The MRI and histology reports were reviewed. gy p Results: Of 20 patients, 18 had abnormal nodes on the first MRI; of these nine had a complete imaging response and five had axillary disease on final pathology. Of the nine that had abnormal nodes on their final MRI, all but one had metastatic disease on final pathology. Of the nine patients with a response on axillary imaging, two had a complete pathological response in the breast. The other had an average tumour size of 53 mm (range 16 to 120 mm). Of the patients who did not respond radiologically in the axilla, three had a complete pathological response in the breast and the remainder had an average tumour size of 44 mm (range 1.7 to 90 mm). P29 P29 Contrast-enhanced MRI in evaluating the contralateral breast for synchronous malignancy in patients with invasive lobular cancer R Pietrosanu, MA Shah, H Gay, S Heller, M Reddy St Georges Hospital NHS Trust, London, UK Breast Cancer Research 2012, 14(Suppl 1):P29 Breast Cancer Research 2012, 14(Suppl 1):P27 Introduction: Breast density, measured as the proportion of the breast occupied by fibroglandular tissue in a mammogram, is a strong and modifiable risk factor for breast cancer. Area-based estimates made by expert observers are a practical approach, but are subjective. Here we investigate repeatability of visual assessment of percentage breast density. Aim: To evaluate the use of MRI in assessment of the contralateral breast in women with invasive lobular cancer. Methods: Retrospective review of 157 breast MRIs in our local unit from January 2010 to May 2012 was carried out. Of these, 35 women had biopsy- proven invasive lobular cancer. Dotarem-enhanced 1.5T MRI was performed and UK standard reporting criteria were used to grade suspicion in any contralateral lesions alongside the use of contrast enhancement curves. Methods: Seven mammographic film readers re-assessed the density of 100 normal full-field digital mammogram cases for which they had made density estimates at least 1 year previously as part of the Predicting Risk of Cancer at Screening (PROCAS) study. The mammograms for a given reader were selected to show a range in density, by randomly sampling 10 cases from each decile of density assessed by that reader. They were reviewed in similar reading conditions on both occasions using a visual analogue scale to record the assessments. Results: Bilateral MRI demonstrated nine (24%) contralateral lesions graded as MRI 2 or above, warranting further investigation with second- look ultrasound with five proceeding to ultrasound core biopsy. Of these, three were confirmed as malignant, revealing three contralateral invasive lobular cancers and one low-grade ductal carcinoma in situ. Our results demonstrated contralateral malignancy in 8% of women with invasive lobular disease. Results: For the majority of readers the difference in mean density between the two sets of readings was less than 6%, but the largest discrepancy between means was 14.7%. Bland-Altman plots were generated for each reader and showed considerable variation between readings on the two occasions. At best, the limits of agreement were -12.46% to +17.02%, and at worst they were -14.50% to +40.98%. The largest difference between first and second readings for each reader ranged from 26 to 65%. P27 P27 Repeatability of breast density visual assessment L Walshaw1, JC Sergeant2, M Wilson3, S Steed1, N Barr3, U Beetles3, C Boggis3, S Bundred3, S Gadde3, Y Lim3, S Whiteside4, DG Evans3, A Howell3, SM Astley2 1Manchester Medical School, University of Manchester, UK; 2Institute of Population Health, University of Manchester, UK; 3Nightingale Centre and Genesis Prevention Centre, University Hospital of South Manchester, Manchester, UK; 4Department of Medical Statistics, University Hospital of South Manchester, Manchester, UK Breast Cancer Research 2012, 14(Suppl 1):P27 P30 P30 Imaging the breast following neoadjuvant chemotherapy: does improvement of imaging findings in the axilla predict pathological response in the breast? P28 Structured evaluation of the effectiveness of an interactive tool for developing interpretation skills in mammography R Pietrosanu1, H Gay1, A Patel2, L Blot3, M Hartswood4, R Proctor5, P Taylor6, L Wilkinson1 1St Georges Hospital NHS Trust, London, UK; 2The Royal Surrey County Hospital NHS Foundation Trust, Guildford, UK; 3University of York, UK; 4University of Edinburgh, UK; 5University of Middlesex, UK; 6University College London, UK Breast Cancer Research 2012, 14(Suppl 1):P28 P29 Conclusion: Current UK guidance does not routinely recommend MRI screening of the contralateral breast in patients with invasive lobular carcinoma. Studies have shown there is a wide variation in detection of contralateral malignancy from 3 to 18%. Our review showed a relatively high rate of incidental contralateral malignancy of 8%, suggesting that MRI screening of the contralateral breast may have more of a role in this patient group than initially considered. We are now extending this review to include three other hospitals in South West London Breast Cancer Network to evaluate these initial findings further. Conclusion: Although density estimates made by a subset of these readers have been strongly related to cancer risk, the variability in reproducibility calls into question the usefulness of subjective assessment without prior evaluation of reader performance. P30 Imaging the breast following neoadjuvant chemotherapy: does improvement of imaging findings in the axilla predict pathological response in the breast? A Williams1, KJ Litton2, A Borley3, P Young2 1School of Medicine, Cardiff University, Cardiff, UK; 2Cardiff and Vale Breast Centre, University Hospital Llandough, Cardiff, UK; 3Velindre Hospital, Velindre NHS Trust, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P30 P28 P28 Structured evaluation of the effectiveness of an interactive tool for developing interpretation skills in mammography R Pietrosanu1, H Gay1, A Patel2, L Blot3, M Hartswood4, R Proctor5, P Taylor6, L Wilkinson1 1St Georges Hospital NHS Trust, London, UK; 2The Royal Surrey County Hospital NHS Foundation Trust, Guildford, UK; 3University of York, UK; 4University of Edinburgh, UK; 5University of Middlesex, UK; 6University College London, UK Breast Cancer Research 2012, 14(Suppl 1):P28 P24 The time and date that each case was reported was extracted from NBSS. A period of over 1 hour without reporting any cases was defined as a break. Recall rate was compared for the first 10 cases since a break and the 10 cases after that using a within-subjects t test. Results: The reading time was significantly longer for DBT images than for FFDM images, with median times of 66 seconds and 17 seconds, respectively (P < 0.001), with times for FFDM ranging from 4.7 to 109.0 seconds and for DBT from 31.9 seconds to 180.6 seconds. The ratio of reading times DBT: FFDM was approximately 4:1, with readers varying between ratios of 2.4 to 4.9. Results: Each reader examined between 20,080 and 74,028 cases over 4 years, and recall rates ranged from 3.6% to 5.9%. Recall rate was 2.3% higher for the first 10 cases after the break than for the subsequent 10 cases (P = 0.004). The sample was too small to examine effects on cancer detection rate. Page 6 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Conclusion: Reading times were greater for DBT as expected, since the reader has to scroll through the images slices to visualise the entire breast. The images used in this study were likely to have a higher rate of abnormalities than screening mammograms, and to have higher breast density, both of which will increase reading time over routine screening. Results: Analysis of variance shows significant two-way interaction: participants receiving training on calcifications improve more on calcifications than on masses and participants receiving training on masses improve more on masses. Results: Analysis of variance shows significant two-way interaction: participants receiving training on calcifications improve more on calcifications than on masses and participants receiving training on masses improve more on masses. Conclusion: Computer-based training tools provide a valuable addition to conventional training and allow trainees to get rapid access to experience with large numbers of cases. Well-designed tools that provide meaningful feedback on interpretation tasks are likely to be effective in improving performance. P34 y Methods: A retrospective analysis of 165 patients with invasive breast cancer treated with neoadjuvant chemotherapy (NAC) (2005 to 2009). Bilateral cancers were analysed separately (total = 169). Data collected include tumour extent on mammogram and MRI, radiology advice for surgery, surgery performed, and whole tumour size on postsurgical specimens. MRI report recommended wide-local excision (WLE), mastectomy, or WLE with multidisciplinary team discussion (MDTD) based on imaging. We used residual tumour size <40 mm on pathology as a cut-off for suitability of WLE. Introduction: Within the NHSBSP, ladies with normal mammograms may be recalled, if they proffer significant clinical symptoms. A clinical recall may then be initiated by the reader according to local guidelines. The purpose of this study was to determine the recall rate of ladies undergoing clinical recall and the cancer detection rate in this group. Results: Radiology advice regarding surgery was suggested in 167 cases and advice was followed in 142 (85%) cases. Fifty-nine cases were suggested for WLE; 46 (78%) underwent WLE with six requiring further surgery (no further disease in four) and 13 (22%) underwent mastectomy based on patient’s choice or clinical decision. Ninety-one cases were suggested for mastectomy; two (2%) had successful WLE due to patient’s choice (one had therapeutic mammoplasty) and 89 (98%) had mastectomy. Seventy-four were appropriate (inflammatory, multicentric, pathology ≥40 mm, patient choice or surgical cosmesis) and based on pathology 15 may have had inappropriate mastectomies as residual disease <40 mm. Seventeen cases were considered for WLE with MDTD, seven (41%) had successful WLE and 10 (59%) had mastectomy; all were appropriately advised when compared with final pathology. Methods: The entire South West and South Central Health Authority screening population was examined for 2011. The study analysed the number of ladies undergoing breast screening, the number of ladies proceeding to clinical recall and the number of cancers detected following clinical recall. Results: A total of 342,628 women were screened within these two regions in 2011. Of these ladies, 884 were recalled for a clinical recall. This represents an overall clinical recall rate of 0.26% or one clinical recall for every 388 patients screened. Of the 884 patients recalled, 17 of these patients were subsequently diagnosed with breast cancer. This represents one breast cancer detected for every 52 women clinically recalled to assessment. Conclusion: This series suggests that NAC offers more opportunities for breast conservation than are being realised. P35 P35 Annual screening of moderate-risk women: a review of 10 years experience within the NHS Breast Screening Programme ND Forester, CE Holmes, N Sibal, L McLean Department of Breast Radiology, Newcastle upon Tyne, UK Breast Cancer Research 2012, 14(Suppl 1):P35 Introduction: The NHS Breast Screening Programme (NHSBSP) stipulates that film-readers should report a minimum of 5,000 cases per year. However, there is no stipulated maximum caseload. With the increasing number of women being included in the remit of the NHSBSP, the question of a maximum is becoming more relevant. p gy p Breast Cancer Research 2012, 14(Suppl 1):P35 Introduction: We have offered annual breast screening to women under 50 years with moderate risk of breast cancer (as defined by NICE guidelines) within the NHS Breast Screening Programme in Newcastle for over 10 years. We evaluated the process and outcome of this programme in this subpopulation of screened women. g Methods: Over the period 1 April 2008 to 31 March 2011 there were 37 radiologists and 34 advanced practitioners reporting breast screening cases within the West Midlands BSP. The range of screening cases reported during this period was between 35 and 49,053 per reader. Readers were divided into four groups according to total breast screening cases they had reported over the 3-year period (<15,000, 15,000 to 20,000, 20,000 to 25,000 and >25,000), in order to assess film-reader performance by number of cases reported. Only those who reported in the West Midlands for the full 3-year period were included (n = 47). This was to enable direct comparisons with Cornford and colleagues, who reported that performance in readers reporting >25,000 cases (over a 3-year period) may decline [1]. Methods: A retrospective review of women screened between April 2002 and March 2012. The population screened, screening episodes, recall rate and outcomes were assessed. Results: A total of 1,027 moderate risk women were screened over 10 years, resulting in 5,406 screening episodes generated. There were 135 recalls from screening in 126 women, giving a recall rate of 2.5%. Of the 135 recall episodes, 100 were benign on further imaging and clinical examination. Thirty-five underwent biopsy, of which 13 were malignant. This gives a cancer detection rate of 2.4 per 1,000 women screened. y y Results: The group reading 20,000 to 25,000 cases had a significantly lower recall rate than the other three groups (P < 0.01) and a higher PPV (P < 0.01). P32 Film-reading workload in the West Midlands 1* 2 3 1 N Rogers , S Bradley , A Duncan , O Kearins , E O Sullivan 1West Midlands Breast Screening Quality Assurance Reference Centre, Birmingham, UK; 2South Birmingham Breast Screening Service, Birmingham, UK; 3Warwickshire, Solihull & Coventry Breast Screening Service, Coventry, UK Breast Cancer Research 2012, 14(Suppl 1):P32 P31 P31 Does imaging correctly influence the surgical management in the setting of neoadjuvant chemotherapy? H Fatayer, B Kim, N Sharma, DD Manuel, BJG Dall Leeds Teaching Hospitals NHS Trust, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):P31 Structured evaluation of the effectiveness of an interactive tool for developing interpretation skills in mammography Methods: A zoo is a collection of lesions presented to a trainee, who is invited to ascribe features to the lesions and provide a confidence estimate. Trainees receive feedback on their progress, through a detailed comparison of their assessment with that of a panel of experts, and from a learning curve showing how their confidence changes. During this evaluation, two sets of test cases (masses and calcifications) were offered initially as a baseline assessment, followed by training on masses or calcifications, followed by a second assessment after which training was swapped (the group that had initially trained on masses was trained on calcifications and vice versa), and then a third assessment was carried out, allowing a test of the impact of the training on performance. Conclusion: Although MRI is used to predict response to chemotherapy it may underestimate and overestimate disease burden. A radiological response in the axilla does not predict a response in the breast and the Page 7 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 standard to the NHSBSP. Performance will be continually monitored within the West Midlands over the coming years. Reference standard to the NHSBSP. Performance will be continually monitored within the West Midlands over the coming years. Reference two areas need to be assessed independently of each other. Surgical planning must take into consideration the pre-chemotherapy disease status in each area. 1. Cornford , et al: Clin Radiol 2011, 66:103-107. P31 Does imaging correctly influence the surgical management in the setting of neoadjuvant chemotherapy? H Fatayer, B Kim, N Sharma, DD Manuel, BJG Dall Leeds Teaching Hospitals NHS Trust, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):P31 P33 Abstract withdrawn Breast Cancer Research 2012, 14(Suppl 1):P33 Breast Cancer Research 2012, 14(Suppl 1):P33 Breast Cancer Research 2012, 14(Suppl 1):P33 Objective: To evaluate the role of imaging in aiding surgical management and assess whether imaging resulted in unnecessary mastectomies. P34 Clinical recall from the NHS Breast Screening Programme: is it worth doing? M Morgan, J Helsdon, A Thomson Dorset Breast Screening Unit, Poole, UK Breast Cancer Research 2012, 14(Suppl 1):P34 P34 Clinical recall from the NHS Breast Screening Programme: is it worth doing? M Morgan, J Helsdon, A Thomson Dorset Breast Screening Unit, Poole, UK Breast Cancer Research 2012, 14(Suppl 1):P34 P34 Multifocal disease and extensive calcifications may not mandate mastectomy if the patient responds well to chemotherapy. Conclusion: Clinical recall does detect breast cancer but 52 women must be recalled to detect one cancer. These ladies are included in the recall rate of the NHSBSP, but the cancer detection rate in this subset of patients is relatively poor compared with those recalled with abnormal mammography. The NHSBSP should evaluate whether this is an acceptable detection rate or use of resources. An alternative is that these ladies could be directed to their GP, and referred onwards to the symptomatic breast service, when clinically appropriate. P32 Film-reading workload in the West Midlands N Rogers1, S Bradley2, A Duncan3, O Kearins1, E O’Sullivan1 1West Midlands Breast Screening Quality Assurance Reference Centre, Birmingham, UK; 2South Birmingham Breast Screening Service, Birmingham, UK; 3Warwickshire, Solihull & Coventry Breast Screening Service, Coventry, UK Breast Cancer Research 2012, 14(Suppl 1):P32 Assessment of the dynamic range in dynamic contrast-enhanced magnetic resonance imaging breast examinations Objective: To illustrate the radiological features of noncalcified DCIS (NC- DCIS) on mammography, ultrasound and MRI. To highlight the role of MRI in determining extent of disease. AEW Ledger1, M Borri1, M Schmidt1, R Pope2, E Scurr1, T Wallace1, C Richardson2, M Usher2, S Allen2, R Wilson2, K Thomas3, N deSouza1, MO Leach1 Introduction: DCIS accounts for 20% screen-detected cancers and 5% of symptomatic cancers. Eighty to 90% present mammographically as microcalcifications, but 10 to 20% are noncalcified and can be mammo- graphically occult. With reported re-excision rates as high as 65% for breast-conserving surgery in DCIS, accurately determining disease extent on preoperative imaging is important. 1CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden NHS Foundation Trust, Sutton, UK; 2Department of Radiology, Royal Marsden Hospital, Sutton, UK; 3Clinical Research and Development, Royal Marsden Hospital, Sutton, UK B C R h 2012 14(S l 1) P39 Breast Cancer Research 2012, 14(Suppl 1):P39 Introduction: Accurate dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) protocol evaluation is necessary to ensure reliable classification of contrast-agent (CA) uptake curves. This work presents a novel retrospective method to assess the dynamic range that constructs CA uptake curves using enhancement of the internal mammary artery. Methods: Imaging of 117 patients with pure DCIS from 2007 to 2011 was reviewed retrospectively. Fifteen patients with NC-DCIS were identified. Imaging findings were compared with disease extent on postoperative histology. Results: NC-DCIS appeared as follows. Mammography: occult 33%, diffuse increased breast density 33%, focal architectural distortion 13%, well- circumscribed lesion 13%, tubular ductal density 7%. Ultrasound: occult 7%, intraductal lesion 13%, microcystic lesion 13%, solid lesion 53%, ill- defined echo poor focus 13%. MRI: nonmass-like nodular enhancement in a ductal, segmental or regional distribution 83%, amorphous nonmass- like enhancement 17%. MRI best depicted the true extent of disease. Conclusion: Recognition of these imaging features is important for accurate surgical planning. MRI has an important role in accurately delineating disease extent and should be considered in treatment planning for NC-DCIS. Results: NC-DCIS appeared as follows. Mammography: occult 33%, diffuse increased breast density 33%, focal architectural distortion 13%, well- circumscribed lesion 13%, tubular ductal density 7%. Ultrasound: occult 7%, intraductal lesion 13%, microcystic lesion 13%, solid lesion 53%, ill- defined echo poor focus 13%. MRI: nonmass-like nodular enhancement in a ductal, segmental or regional distribution 83%, amorphous nonmass- like enhancement 17%. MRI best depicted the true extent of disease. P36 Survey of clinical recalls from breast screening at Nottingham Breast Institute: implications for practice * Introduction: In our unit, symptomatic women identified at the time of screening (that is, those reporting a breast symptom, or in whom the radiographer notices a breast sign) are recalled for further assessment despite a normal mammogram. Several studies have suggested that the cancer detection rate in this group of patients is low. The aims of this survey were therefore to determine the cancer yield and whether our current practice of recalling women on the basis of clinical history alone is a worthwhile practice. Introduction: Our programme underwent phased implementation of digital mammography, commencing soft-copy reading (SCR) in April 2009. Prior to this, all mammograms -a mixture of direct digital (DDM), computed radiography (CR) or film screen (FS) - were printed and read on multiviewers with the old films. For SCR, prior screens were available to readers but not digitised or mounted. This study was performed to evaluate recall rates, cancer detection and mammographic features assessed and biopsied, before and after SCR was introduced. Methods: Screening episodes between 1 April 2006 and 31 March 2012 were interrogated on NBSS. Women aged 50 to 70 years were included. Screening method, soft-copy or hard-copy reading, women recalled, cancers detected and all biopsies were analysed. Methods: The total number of women screened, and those recalled to assessment, during a 24-month period were retrospectively identified using NBSS. Women recalled on the basis of clinical history alone were identified and information about the screening assessment clinic visit was reviewed. Results: A total of 128,544 screening episodes were recorded in 6 years. The overall recall rate was identical at 2.5% between hard-copy and soft-copy reading. Seventy-two per cent of assessed women underwent biopsy in 2006 to 2009 and 79% in 2009 to 2012. The cancer detection rate was identical at 8.5/1,000 in 2006 to 2009 and 8.54/1,000 in 2009 to 2012. The number of biopsies performed was 18/1,000 women screened for FS, 18.5/ 1,000 for CR and 20/1,000 for DDM. No significant difference was shown in the benign/malignant ratio of calcifications biopsied in women recalled to assessment in 2006 to 2009 compared with 2009 to 2012. Results: Of the 45,940 women screened during this time period, 337 cancers (invasive and DCIS) were diagnosed (7.3 cancers/1,000 women screened). A total of 335 cancers were diagnosed in the 1,177 women recalled because of an abnormal mammogram (28.4%). References 1. Mesurolle B, et al: Mammographically non-calcified ductal carcinoma in situ: sonographic features with pathological correlation in 35 patients. Clin Radiol 2009, 64:628-636. 2. Raza S, et al: Pure ductal carcinoma in situ: a range of MRI features. AJR 2008, 191:689-699. 2. Raza S, et al: Pure ductal carcinoma in situ: a range of MRI features. AJR 2008, 191:689-699. 1. National Collaborating Centre for Primary Care: Familial Breast Cancer: The Classification and Care of Women At Risk of Familial Breast Cancer in Primary, Secondary and Tertiary Care London: NICE 2006, 1-75. P35 The missed cancer rate and cancer detection rate showed no difference. Conclusion: Screening of moderate-risk women is possible within the NHS Breast Screening Programme. The recall rate of 2.5% falls within the acceptable standards for both prevalent and incident screens set in the over-50 years population. As this is annual screening, the overall cancer Conclusion: The findings are not in agreement with previously published data suggesting that these parameters should not be applied as a quality Page 8 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 detection rate is similar to that found in the over-50 years group (approximately 6 per 1,000 over 3 years). Screening is therefore as effective in this increased risk population as the current NHS Breast Screening programme is for over-50 year olds. Reference P37 P37 Imaging features of noncalcified DCIS JM Scudder, J Parikh Guy’s and St Thomas’ NHS Foundation Trust, London, UK Breast Cancer Research 2012, 14(Suppl 1):P37 P36 Survey of clinical recalls from breast screening at Nottingham Breast Institute: implications for practice MZ Mvere, S Tennant Nottingham Breast Institute, Nottingham, UK Breast Cancer Research 2012, 14(Suppl 1):P36 P36 Survey of clinical recalls from breast screening at Nottingham Breast Institute: implications for practice MZ Mvere, S Tennant Nottingham Breast Institute, Nottingham, UK Breast Cancer Research 2012, 14(Suppl 1):P36 P36 Survey of clinical recalls from breast screening at Nottingham Breast Institute: implications for practice Two cancers were detected in the 116 women recalled for clinical history alone (1.72%). The cancer detection rate is therefore 17 times lower in the clinical recall group than in those called for a mammographic abnormality. Conclusion: The cancer detection rate in women recalled on the basis of clinical history alone is low and, given the extra time and resources required in assessing these women, our current practice should be reviewed. Conclusion: No significant difference was demonstrated before and after the introduction of SCR in recall rates or cancer detection. There was a nonsignificant trend towards more biopsies being performed with SCR but without additional cancers detected. Overall cancer detection rate remained high with low recall rates since adopting digital screening. P38 Comparison of recall, biopsy, and cancer detection rates in the Southern Derbyshire screening programme between 2006 and 2009 using hard-copy mammography and in 2009 to 2012 following the full introduction of soft-copy reporting AE Turnbull, S Puri, M Bagnall, J York, S Farmer, N Horsley Breast Unit, Royal Derby Hospital, Derby, UK Breast Cancer Research 2012, 14(Suppl 1):P38 P42 P42 MRI for PIP implant rupture: appearances and rupture rate V Helyar, S McWilliams, C Burke, G Charles-Edwards Guy’s and St Thomas’ NHS Foundation Trust, London, UK Breast Cancer Research 2012, 14(Suppl 1):P42 Introduction: The aim of this study was to assess radiological and pathological features associated with successful ultrasound-guided biopsy of malignant microcalcifications. Methods: Screen-detected breast cancers were reviewed. Cases where microcalcification was the predominant mammographic feature were identified. Ultrasound findings, mammographic appearances, method of biopsy and pathological features were recorded. Introduction: The Poly Implant Prosthèse breast implants contain an unlicensed silicone and are associated with elevated rupture rate. We examined the performance of our MR breast protocol for implant assessment and quantified radiological rupture rate. y Results: There were 348 breast cancers diagnosed in the 24-month study period, 84 cases had microcalcification as the predominant mammographic feature. In 75 cases (89%) an ultrasound scan was performed. Forty-one cases had an ultrasound abnormality and underwent an ultrasound-guided core biopsy. Successful ultrasound-guided core biopsy of malignant microcalcifications was associated with a mammographic size >20 mm (P = 0.002), the presence of an associated mammographic feature (P = 0.03) and invasive disease on surgical resection (P = 0.009). There was no association with calcification morphology, mammographic background pattern or the histological grade of the DCIS. When the ultrasound was normal or the ultrasound-guided core biopsy inconclusive, the patient underwent an additional biopsy typically stereotactic vacuum biopsy. In the 41 cases where ultrasound-guided core was the first biopsy performed, nine cases underwent an additional biopsy prior to the therapeutic surgical procedure, compared with just one case when stereotactic vacuum biopsy was the first procedure (P = 0.005). Methods: This was a retrospective review of 145 patients receiving an implant protocol MR breast between January and April 2012 at Guy’s Hospital. Scans were reported by two consultant radiologists. Patient symptoms, MR findings and origin of surgery were recorded. Data were analysed with descriptive statistics. Results: Twenty-one per cent (30/145) showed unilateral rupture, 46% (14/30) bilateral rupture. Of those with a rupture, only four were symptomatic. Seventy-seven per cent (23/30) of ruptures were intra- capsular, 67% (20/30) showed evidence of silicone adenitis. MR findings show complete agreement with the explantation data gathered to date. Conclusion: These preliminary data support the use of our MR protocol for the assessment of breast implant integrity. Results: Twenty-one per cent (30/145) showed unilateral rupture, 46% (14/30) bilateral rupture. Assessment of the dynamic range in dynamic contrast-enhanced magnetic resonance imaging breast examinations Methods: Routine clinical breast examinations were performed using 3D fat-suppressed spoiled gradient-echo sequences (1.5T). Retrospective analysis was approved by the Clinical Audit Committee. Five different protocols were evaluated (10°, 14° and 18° flip angles (FAs), radial or linear k-space sampling), with seven to 10 patients in each group (n = 45). CA uptake curves were constructed from a standardised axial slice through the right mammary artery, and maximum relative enhancement (Emax) and time- to-peak enhancement (Tmax) were measured for each examination [1]. Emax and Tmax were compared between protocols (ANOVA/Mann-Whitney; P < 0.05 indicating significance). For each protocol, calculated values of Conclusion: Recognition of these imaging features is important for accurate surgical planning. MRI has an important role in accurately delineating disease extent and should be considered in treatment planning for NC-DCIS. Page 9 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Methods: Digital mammograms from 112 consecutively screened women were sourced in each centre. Test sets were enriched with mammograms from each PGMI category and independently scored by four mammo- graphers, each with ≥4 years’ experience, using local PGMI. Each image was individually scored P, G, M, or I. Reasons for scoring less than perfect were documented and each mammogram assigned an overall PGMI score. Test sets were exchanged and the process repeated. maximum relative enhancement (Ecalc) were derived from the Bloch equations and compared with Emax to validate the results. Results: A lower FA and radial k-space sampling resulted in a statistically significant decrease in Emax (P < 0.0001 and P = 0.001, respectively). Radial protocols exhibited greater Tmax than linear protocols at FAs of both 14° (P = 0.025) and 18° (P < 0.0001). Ecalc was found to increase with FA as expected, with good agreement between Ecalc and Emax. Conclusion: Significant differences were found between patient groups with only small alterations in protocol. Observations agreed with expected results, validating this method of retrospective analysis. The dynamic range is optimised at higher flip angles and with linear k-space sampling. Reference Results: Cambridge uses 17 criteria for scoring mammograms less than perfect. Oslo uses similar criteria, but subcategorised, totalling 39 criteria. P43 Review of the Welsh mammographic screening programme for women who underwent mantle irradiation for Hodgkin lymphoma GM Edwards3, E Gallop-Evans2, C Jordan2, G Stevens1, K Gower Thomas1 1Breast Test Wales, Cardiff, UK; 2Velindre Cancer Centre, Cardiff, UK; 3University of Wales Medical School, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P43 Conclusion: Ultrasound-guided core biopsy of malignant microcalcifications is more likely to be successful when mammographic size is larger and invasive disease is present. When ultrasound-guided biopsy is the first-line procedure a second biopsy is significantly more likely prior to therapeutic surgery. Introduction: There is evidence that women who received supra- diaphragmatic radiotherapy (sRT) for Hodgkin lymphoma at a young age are at increased risk of developing breast cancer. This study reports the implementation and results of the breast screening programme for such women in South Wales, which offers screening to women eight or more years following treatment. P42 Of those with a rupture, only four were symptomatic. Seventy-seven per cent (23/30) of ruptures were intra- capsular, 67% (20/30) showed evidence of silicone adenitis. MR findings show complete agreement with the explantation data gathered to date. Conclusion: These preliminary data support the use of our MR protocol for the assessment of breast implant integrity. P43 Review of the Welsh mammographic screening programme for women who underwent mantle irradiation for Hodgkin lymphoma GM Edwards3, E Gallop-Evans2, C Jordan2, G Stevens1, K Gower Thomas1 1Breast Test Wales, Cardiff, UK; 2Velindre Cancer Centre, Cardiff, UK; 3University of Wales Medical School, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P43 P41 P41 Comparing the use of PGMI scoring systems used in the UK and Norway to assess the technical quality of screening mammograms: a pilot study M Boyce1, R Gullen2, D Parashar3, K Taylor1 1Cambridge Breast Unit, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; 2Oslo Universitetssykehus, Ullevål, Oslo, Norway; 3Cambridge Cancer Trials Centre, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK Breast Cancer Research 2012, 14(Suppl 1):P41 Assessment of the dynamic range in dynamic contrast-enhanced magnetic resonance imaging breast examinations There is fair agreement ( = 0.38) between centres in assigning images as acceptable overall (P, G, M) but poor inter-rater agreement within and between centres in further categorising acceptable mammograms as P, G or M ( < 2). Most common faults in Oslo were skin folds, and inadequate pectoralis muscle in Cambridge. Most faults overall were on oblique views. 1. Kuhl, et al: Radiology 1999, 211:101-110. 1. Kuhl, et al: Radiology 1999, 211:101-110. Conclusion: Poor rater agreement and differing faults may be due to the variation in number and interpretation of categories used. Radiographer training may also be an issue. Further research should establish quantitative assessment methods and internationally uniform practice. P40 P40 Use of ultrasound in the assessment of screen-detected malignant microcalcifications MZ Mvere, EJ Cornford, JJ James Nottingham Breast Institute, Nottingham, UK Breast Cancer Research 2012, 14(Suppl 1):P40 MZ Mvere, EJ Cornford, JJ James P45 P45 False negative assessments: an effective quality assurance method DD Manuel, BJG Dall, N Sharma Leeds Teaching Hospital NHS Trust, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):P45 p g Methods: All women with IC having had screening between 1998 and 2001 were included. Classifications of IC were performed during the IC review process. Date of screening and date of diagnosis were used to calculate the interval between screening and diagnosis. Ten-year survival data were obtained on all interval cancers from national register. Objective: To identify the frequency and characteristics of false negative assessment (FNA); an interval cancer with prior recall to assessment. Results: Some 692 interval cancers were diagnosed following screening. In total, 57.8% (391) were TI, 17.7% (120) were FN, 10% (68) were OCC and 14.5% were UCC. A total of 15.6% (108) presented within 1 year of screening, 38.2% (264) in the second year, and 46.1% (319) in the third year since screening. The overall 10-year survival was 72.4%. It was 77.5% for TI and 55% for FN. The odds ratios for cancer presenting in the second year and third year were 1.009 (P = 0.97) and 1.1162 (P = 0.486), respectively, as compared with IC presenting within the first year. Methods: A retrospective audit, over 7 years, between 1 April 2004 and 31 March 2011. Using NBSS databases, we recorded: lesion type, size, further mammographic views, ultrasound, clinical findings, biopsy results and final histology. gy Results: Twenty-nine cases by QARC, 13 true FNA and 16 excluded because contralateral or in a different quadrant. In this period, 220,522 woman were screened and 10,391 (<5%) were referred for assessment. Total cancers detected in this period were 2,343, of which 1,867 were diagnosed at assessment and 476 (20%) were interval cancers. True FNA: 2.7% of all interval cancers. True FNA: 0.6% of all screen-detected cancers. All cases were background mixed density. True FNA: 11/13 incident screen, 2/13 prevalent. Asymmetry/stromal deformity in 7/13 (54%), calcification 4/13 (30%), mass 2/13 (15%). Two cases were early recall (EC) (asymmetry, and mass with calcifications), both incident screens. EC did not contribute in either case as it led to false reassurance. All asymmetry/stromal deformity reassuring on further views, with no correlating ultrasound or clinical abnormality. On review of all cases, only 2/13 were felt not to have had complete triple assessment. Final pathology: two DCIS, four lobular cancers, seven ductal cancers. Comparing the use of PGMI scoring systems used in the UK and Norway to assess the technical quality of screening mammograms: a pilot study 1* 2 3 1 There is a growing argument that many of these younger women are being subjected to an unnecessary invasive test with obvious resource and financial implications. P44 Fibroadenoma: to biopsy or not to biopsy? That is the question AS Karuppiah, WG Bugg, A Juette Norfolk and Norwich University Hospitals NHS Trust, Norwich, UK Breast Cancer Research 2012, 14(Suppl 1):P44 Methods: A retrospective review of images and reports of all the diagnosed breast cancers in women aged 30 and below in the Edinburgh area between 2006 and 2011. Introduction: In 2011, our institution adopted the policy (in line with national guidance) not to biopsy breast masses in patients <25 years if they have benign characteristics on ultrasound. This retrospective audit was performed prior to this decision. We compare the histology result with the USS findings of all patients <25 years who had a breast mass biopsied or excised, between 2003 and 2009, with respect to Stavros criteria. Results: A total of 37 cancers were detected in women of the sample age group. Of these, 35 had definable masses seen on ultrasound, all of which were considered R3 or above, and none of which on imaging review met benign criteria. The remaining two cancers were occult on ultrasound with no definable mass. Methods: Cases were identified from the Histopathology databases. Inclusion criteria include patients <25 years with a breast core biopsy or excision specimen. The ultrasound images were reviewed blind by a consultant breast radiologist and two registrars using Stavros criteria. The team knew one case; a blinded opinion from other consultant breast radiologists was sought. Conclusion: Over a 5-year period, no breast cancers would have been missed in our unit if a policy not to biopsy classic benign solid breast lesions in women aged 30 and below had been implemented. Although only looking at small numbers, this study adds to the growing evidence that biopsies of lesions with classic benign features in women of this age group are extremely unlikely to result in malignant pathology, and that current policy to routinely biopsy may be unnecessary. Results: The cohort included 109 cases, of which two were malignant, three were benign phyllodes and 104 were fibroadenomata. Twenty-three out of 109 cases did not meet Stavros criteria for a benign lesion, which included one malignant spindle cell tumour, one phyllodes and 21 fibroadenomata. P45 Seven cases were discussed at a multidisciplinary meeting. Conclusion: The majority of IC were TI and presented in the third year since screening. FN had worse 10-year survival compared with TI. IC presenting in the third year did not have a statistically significantly worse long-term survival. Reduction of the screening interval would not have an impact on long-term survival outcome. P48 Computer-aided detection of screening breast cancer: a novel approach based on genetic programming F Canavan1, S Harding2, L Gustard3, AM Murphy3, JF Miller4, SL Smith4 1Betsi Cadwaladr University Health Board, Bangor, UK; 2www. machineintelligence.co.uk, York, UK; 3York Teaching Hospital NHS Foundation, York, UK; 4University of York, UK Breast Cancer Research 2012, 14(Suppl 1):P48 Comparing the use of PGMI scoring systems used in the UK and Norway to assess the technical quality of screening mammograms: a pilot study 1* 2 3 1 The one case known to the team was reviewed blind by a consultant survey and the consensus opinion was that this did not meet Stavros criteria. Eighty-five out of 109 cases met Stavros criteria for a benign lesion. Of these, 52 cases had a core biopsy and 33 cases had a surgical excision biopsy. P47 Classification of interval breast cancers and their presentation since screening: impact on long-term survival YF Fong1, J Evans2, D Brookes2, K Gower Thomas1,2 1Royal Glamorgan Hospital, Llantrisant, UK; 2Breast Test Wales, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P47 Comparing the use of PGMI scoring systems used in the UK and Norway to assess the technical quality of screening mammograms: a pilot study 1 2 3 1 Comparing the use of PGMI scoring systems used in the UK and Norway to assess the technical quality of screening mammograms: a pilot study 1* 2 3 1 M Boyce1, R Gullen2, D Parashar3, K Taylor1 1Cambridge Breast Unit, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; 2Oslo Universitetssykehus, Ullevål, Oslo, Norway; 3Cambridge Cancer Trials Centre, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK Breast Cancer Research 2012, 14(Suppl 1):P41 M Boyce1, R Gullen2, D Parashar3, K Taylor1 1Cambridge Breast Unit, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; 2Oslo Universitetssykehus, Ullevål, Oslo, Norway; 3Cambridge Cancer Trials Centre, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK Breast Cancer Research 2012, 14(Suppl 1):P41 y g Methods: Women were identified from the hospital Hodgkin Lymphoma and the Breast Test Wales screening databases. Those treated with sRT below the age of 36 years were included in the review (sRT range 9 to 34 years). Breast cancer and cause of death developing in the cohort were recorded. Results: Ninety-four eligible women were identified who underwent sRT between 1977 and 2004 in whom seven breast cancers developed. Three were found on self-examination, two through the Hodgkin screening programme, one through NHS screening and one had incomplete data. Sixteen women died, three from breast cancer. The mean induction period was 21.4 years between radiation and breast cancer presentation. Conclusion: This review confirms an increased incidence of breast cancer in these irradiated women many years following the therapy. The need for a nationwide screening programme is confirmed although the age at which to offer screening could be reviewed. Introduction: The UK and Norway use PGMI scoring to critique mammo- graphic image quality (IQ). PGMI comprises categories with associated criteria for determining mammograms Perfect, Good, Moderate, Inadequate. Implementation of PGMI may be variable, subjective and interpreted locally, making accurate comparison of performance across countries difficult. We compared PGMI use in Cambridge and Oslo, determining differences and possible contributory factors, enabling suggestions for future research and practice. Page 10 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Page 10 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Introduction: Currently, most women under 30 with solid breast lumps undergo biopsy, despite classic benign features on ultrasound. However, breast cancer is uncommon in this age group and the vast majority have obvious sinister features on ultrasound. P47 Classification of interval breast cancers and their presentation since screening: impact on long-term survival YF Fong1 , J Evans2, D Brookes2, K Gower Thomas1,2 1Royal Glamorgan Hospital, Llantrisant, UK; 2Breast Test Wales, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P47 Conclusion: All nonbenign cases were correctly identified by all reviewers using Stavros criteria. No malignancies were missed in this series. This audit suggests that the biopsy rate can be safely reduced in those <25 years, if the national guidance is followed. Introduction: Interval cancers (IC) present between screenings. All interval cancers within Breast Test Wales (BTW) are classified into true interval (TI), false negative (FN), occult (OCC) and unclassified (UCC). We aim to evaluate the overall number of IC within BTW and their presentation since screening and the impact on long-term survival. P49 , g , , 1Breast Screening Unit, Newcastle, UK; 2Department of Genetics, Newcastle, UK Breast Cancer Research 2012, 14(Suppl 1):P51 Breast Cancer Research 2012, 14(Suppl 1):P51 Breast Cancer Research 2012, 14(Suppl 1):P49 Introduction: The Cancer Reform Strategy [1] and NICE guidelines [2] highlighted the need to identify women at increased genetic risk of breast cancer. Following this, a magnetic resonance imaging (MRI) screening service was established in Newcastle. Computer-aided detection of screening breast cancer: a novel approach based on genetic programming 1* 2 3 3 4 pp g p g g F Canavan1, S Harding2, L Gustard3, AM Murphy3, JF Miller4, SL Smith4 1Betsi Cadwaladr University Health Board, Bangor, UK; 2www. machineintelligence.co.uk, York, UK; 3York Teaching Hospital NHS Foundation, York, UK; 4University of York, UK Breast Cancer Research 2012, 14(Suppl 1):P48 Conclusion: True FNA remains very low. This is because assessment cases have complete triple assessment and MDT discussion. FNA can be used as a valuable educational process and mechanism to ensure consistency and adherence to NHSBSP standards. Introduction: Mammography is the keystone of breast cancer screening. Yet high sensitivity is achieved at the cost of low specificity - only one-third of patients recalled will have breast cancer. Computer-aided detection (CAD) is a potentially valuable tool for assisting the breast radiologist to improve positive prediction values. However, to date, CAD has not reliably altered screening outcomes and the large proportion of false positives remains a drawback. We describe a novel method to improve CAD performance called P46 Benign appearing breast lesions age 30 and below: to biopsy, or not to biopsy? AC Nisbet, R Bradley NHS Lothian, Edinburgh, UK Breast Cancer Research 2012, 14(Suppl 1):P46 P46 Benign appearing breast lesions age 30 and below: to biopsy, or not to biopsy? AC Nisbet, R Bradley NHS Lothian, Edinburgh, UK Breast Cancer Research 2012, 14(Suppl 1):P46 or not to biopsy? AC Nisbet, R Bradley NHS Lothian, Edinburgh, UK Breast Cancer Research 2012, 14(Suppl 1):P46 Page 11 of 14 Page 11 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Cartesian Genetic Programming (CGP); a machine-based learning algorithm, akin to genetic evolution. clearance (ANC). If axillary assessment is normal we perform sentinel node biopsy (SNB) with intraoperative OSNA, unless planned for neoadjuvant chemotherapy or primary medical therapy. In patients with positive OSNA, further surgery is performed as per hospital protocol. Retrospective correlation of preoperative axillary ultrasound and FNA findings with intraoperative OSNA in the SNB group and final histology in the ANC group was performed, with patients diagnosed with invasive breast carcinoma between September 2010 and September 2011. clearance (ANC). If axillary assessment is normal we perform sentinel node biopsy (SNB) with intraoperative OSNA, unless planned for neoadjuvant chemotherapy or primary medical therapy. In patients with positive OSNA, further surgery is performed as per hospital protocol. Computer-aided detection of screening breast cancer: a novel approach based on genetic programming 1* 2 3 3 4 Retrospective correlation of preoperative axillary ultrasound and FNA findings with intraoperative OSNA in the SNB group and final histology in the ANC group was performed, with patients diagnosed with invasive breast carcinoma between September 2010 and September 2011. Methods: A population of 12 CAD programs underwent repeat fitness evaluation of how each performed in classifying breast masses. Each performed a different combination of image manipulations on 26 training mammograms. Output was subjected to a threshold, to produce a binary image predicting a benign or suspicious mass. This was compared with the image labelled by the screening radiologist. The program fitness was determined by accuracy of prediction. Fitter programs were copied and mutated to produce new variants that were re-tested. Multiple programs emerged by evolution and predictions summed to give a single more confident prediction. The confidence level was overlaid as a colour map on the original mammograms. p p Results: See Figure 1. Seventy-seven patients were lymph node-positive (LN+ve) overall (77/185 = 42% population LN+ve). Twenty-seven patients had preoperative LN+ve diagnosis with USS and FNA (27/77 = 35% of LN+ve patients have preoperative diagnosis). Imaging of the OSNA macromet subset is reviewed and examples discussed (n = 27:22 imaging normal, three equivocal, two abnormal). Results: The false positive rate was 7/26 (27%), comparing favourably with current mammography CAD systems (true positives, 13/26; true negatives, 4/26; false negatives, 2/26). Conclusion: This is the first study correlating preoperative imaging with OSNA. Our high rate of preoperative diagnosis is encouraging, but suggestions for improvement are discussed. Conclusion: Our pilot study suggests CGP holds great promise for developing a viable CAD system more suited to breast screening and so warrants further evaluation. P51 Breast screening with magnetic resonance imaging in high-risk women L McLean1, F Douglas2, N Forester1, CE Holmes1 1Breast Screening Unit, Newcastle, UK; 2Department of Genetics, Newcastle, UK Breast Cancer Research 2012, 14(Suppl 1):P51 Should the breast be a check area on standard CT thorax examinations? Introduction: The Belfast Breast Screening Programme serves a population of approximately 25,000 patients. We aimed to analyse radiological and pathological trends between screen-detected and interval breast cancers, and determine our screening lesion miss rate. NL Marshall, L Duddy, J Barry, R Murphy, P Smiddy, M Ryan, D University College Cork, Ireland Breast Cancer Research 2012, 14(Suppl 1):P55 Methods: Using the Quality Assurance Reference Centre (QARC) database patients were identified with screen-detected or interval breast cancers diagnosed via the Belfast Breast Screening Programme over a fixed period. Film packs and operative specimen reports were analysed for radiological and pathological data. Introduction: With increasing cross-sectional imaging, incidental findings (IF) are common across all subspecialities. There are implications for patients and the health service consequent to IFs: anxiety, further radiation and biopsy. Our purpose was to evaluate the malignant assessment (MA) rate for patients referred to the symptomatic breast clinic after a breast IF and to compare this with the general MA rate. Results: Screening identified 75% of breast cancers. Only 30% of screen- detected lesions were palpable by breast clinicians. Sixty-one per cent of lesions were in the upper outer quadrant, with equal left-right distribution. Radiological measurements underestimated lesion size by 22%. There was moderate correlation between lesion size and lymph node status. No other correlations were identified. Twenty-three per cent of interval cancers presented in year 1 following screening, 28% in year 2 and 49% in year 3. They were larger at presentation than screen-detected cancers (29.4 mm vs. 18.2 mm mean size) and of pathologically higher grade (39% vs. 13% grade III). Screen-detected cancers were mostly IDC (63%) or DCIS (18%) subtype. Interval lesions were predominantly IDC subtype (87%). Interval lesions showed more nodes positive per patient. Retrospective review of past screening films of interval-detected cancers showed suspicious features present in 17% of cases. Methods: This retrospective review spanned 2 years. A trawl was performed of breast imaging reports for CT-detected lesions. Demographics, imaging and pathology results were collated. Statistical analysis using Fisher’s exact test was performed to identify demographic factors associated with MAs. The MAs for the IF cohort were then compared with the general clinic population using chi-square analysis. Results: There were 103 patients with IFs (102 women, one man) comprising 49 lesions (30 benign, 19 malignant). In the general service were 10,330 patients with 2,551 lesions (683 malignancies). P50 P50 Preoperative ultrasound and FNA in the diagnosis of axillary involvement in invasive breast cancer: correlation with intraoperative one-stop nucleic acid amplification and final histology EK Jackson, N Howes, A Jones, Z Rayter, A Valencia Bristol Royal Infirmary, Bristol, UK Breast Cancer Research 2012, 14(Suppl 1):P50 50 Preoperative ultrasound and FNA in the diagnosis of axillary involvement in invasive breast cancer: correlation with intraoperative one-stop nucleic acid amplification and final histology EK Jackson, N Howes, A Jones, Z Rayter, A Valencia Bristol Royal Infirmary, Bristol, UK Breast Cancer Research 2012, 14(Suppl 1):P50 Methods: Women were identified from the Regional Genetic Centre database, following assessment by a consultant geneticist as high risk according to the NICE guidelines [2]. Screening comprised annual MRI ± mammography, depending on their age, following the standards set by the NHSBSP [3]. All investigations were double-read by NHSBSP-compliant radiologists. g Results: A total of 142 women underwent 311 screening episodes between January 2008 and June 2012. There were 28 recalls (9%) for second-look ultrasound in 24 women. Thirteen lesions had a core biopsy, from which six (1.9%) malignancies were identified of varying histological type, size and grade. Seven lesions were benign. No ultrasound abnormality was found in the remaining 15 lesions. No interval cancers have developed. Objective: To assess our preoperative pickup of malignant axillary lymph nodes by ultrasound and FNA compared with one-stop nucleic acid amplification (OSNA) and final histology. y Methods: At our unit all patients with invasive breast cancer undergo axillary ultrasound, and those with suspicious or equivocal findings undergo axillary FNA. If FNA is positive we proceed to axillary node Figure 1(abstract P50) Figure 1(abstract P50) Page 12 of 14 Page 12 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 patients are being referred for marker clip placement. This is our experience of the indications for marker clip placement in a large cancer unit using both NACT and NAET for treatment of breast cancers. Methods: An 18-month retrospective audit was performed to see which patients were having marker clip placement. We recorded the indication for marker clip, final surgery and whether the marker clip was used for localisation. A radiological and pathological analysis of screen-detected and interval-detected breast cancers in Belfast 1 1 1 2 1 2 1 2 Breast Cancer Research 2012, 14(Suppl 1):P52 P52 P52 A radiological and pathological analysis of screen-detected and interval-detected breast cancers in Belfast J Malloy1, C Hennell1, L Bamford1,2, K Lowry1,2, L Tong1,2 1Belfast City Hospital, Belfast, UK; 2Northern Ireland Breast Screening Programme (Eastern Unit), Belfast, UK Breast Cancer Research 2012, 14(Suppl 1):P52 References Results: Fifty patients of the 92 had marker clip placement, of which 37 had NACT and 13 NAET. The marker clip was placed in 35/50 as radiology raised the possibility of near or complete pathological response. In 8/50 the marker clip was placed to aid the pathologist because of concern regarding residual low-volume disease in patients having WLE or mastectomy and 7/50 were trial patients. Thirty-eight patients had WLE and the marker clip was localised in 6/38 cases. Ten patients had mastectomy and in two cases no surgery was performed due to proven metastatic disease. Nine patients had complete pathological response. 1. Cancer Reform Strategy London: Department of Health 2007. 1. Cancer Reform Strategy London: Department of Health 2007 2. National Collaborating Centre for Primary Care: Familial Breast Cancer: The Classification and Care of Women At Risk of Familial Breast Cancer in Primary, Secondary and Tertiary Care (Partial Update of NICE Clinical Guideline 14) London: NICE 2006, 1-75. 3. NHS Cancer Screening Programmes: Technical Guidelines for Magnetic Resonance Imaging for the Surveillance of Women at Higher Risk of Developing Breast Cancer. NHSBSP Publications No. 68 2012. 3. NHS Cancer Screening Programmes: Technical Guidelines for Magnetic Resonance Imaging for the Surveillance of Women at Higher Risk of Developing Breast Cancer. NHSBSP Publications No. 68 2012. Conclusion: Due to NACT and NAET, radiologists and pathologists are facing new challenges in localising and identifying residual low-volume disease following completion of treatment. Marker clip placement can play a crucial role in ensuring accurate localisation at the time of surgery and can aid identification of residual disease for the pathologist. P55 Should the breast be a check area on standard CT thorax examinations? P55 Should the breast be a check area on standard CT thorax examinations? NL Marshall, L Duddy, J Barry, R Murphy, P Smiddy, M Ryan, D Hill University College Cork, Ireland Breast Cancer Research 2012, 14(Suppl 1):P55 P50 Conclusion: In our unit, high-risk MRI screening is effective and efficient, with a cancer detection rate of 0.2 per 1,000 women screened and no false negative screens to date, with an acceptable recall rate for second-look ultrasound (within the standard set under the NHSBSP guidelines). This has been possible due to the close liaison between the departments of Genetics and Breast Screening. f patients are being referred for marker clip placement. This is our experience of the indications for marker clip placement in a large cancer unit using both NACT and NAET for treatment of breast cancers. g Methods: An 18-month retrospective audit was performed to see which patients were having marker clip placement. We recorded the indication for marker clip, final surgery and whether the marker clip was used for localisation. P58 Breast MRI quality assurance in practice S Bacon, B Dall, N Sharma, D Manuel, D Wilson St James’s Institute of Oncology, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):P58 Breast MRI quality assurance in practice * p Conclusion: We propose patients staged preoperatively as T2G2 and above should have axillary node biopsy considered despite negative preoperative ultrasound. This may increase the accuracy of axillary preoperative assessment with a subsequent decrease in second operations to the axilla. References Introduction: Breast MRI has been incorporated into the NHSBSP programme and therefore is subject to quality assurance (QA) to NHSBSP standards. The breast screening technical guidelines recommend weekly testing of signal-to-noise ratio (SNR) and suppression effectiveness. We tested a method of implementing these recommendations on a Siemens 1. Aiten E, Osman M: Factors affecting nodal status in invasive breast cancer: a retrospective analysis of 623 patients. Breast J 2010, 16:271-278. 1. Aiten E, Osman M: Factors affecting nodal status in invasive breast cancer: a retrospective analysis of 623 patients. Breast J 2010, 16:271-278. 1.5T Avanto scanner using phantoms and software supplied as standard. Methods: Phantoms were placed in the right and left apertures of the Siemens four-element breast matrix coil and manufacturer QA performed. Suppression effectiveness was measured by acquiring a 3D spoiled gradient echo (FLASH) sequence with and without water suppression. Regions of interest (ROI) were drawn in both phantoms and percentage signal reduction due to water suppression calculated for three slices. SNR was measured using the standard fat-suppressed 3D FLASH sequence acquired three times after waiting 10 minutes to minimise fluid motion artefacts. Subtractions were performed and ROI drawn in both phantoms. SNR was calculated as √2 × (mean signal intensity)/(standard deviation of the subtraction image) for three slices. 2. Hogan BV, Shenoy HG, et al: The use of ultrasound in pre-operative assessment of the axilla in breast cancer [abstract]. Cancer Res 2009, 69(2 Suppl). P57 Predicting risk of malignancy in subgroups of B3 breast lesions ND Forester, M Brotherton, A-M Wason Department of Breast Radiology, Bradford, UK Breast Cancer Research 2012, 14(Suppl 1):P57 Introduction: Heterogeneity and varying malignancy risk makes B3 lesion management difficult. Can histological features predict malignancy risk? Methods: A retrospective review of B3 lesions (April 2005 to March 2010) following 14G biopsy, followed to final pathology. P58 Key phrases from pathology identified; atypia, radial scar/complex sclerosing lesion (RS/CSL), atypical intraductal proliferation (AIDP), atypical ductal/lobular hypertrophy (ADH/ALH), flat epithelial atypia (FEA), lobular in situ neoplasia (LISN), and so forth. Age-adjusted logistic regression to assess risk of malignancy (Stata11). Results: The breast coil passed the manufacturer QA tests. Mean suppression effectiveness was 97.0% (95.6 to 98.0%) for the right phantom and 96.9% (95.4 to 97.7%) for the left. Mean SNR was 98.8 (41.2 to 110.9) for the right phantom and 96.2 (53.8 to 115.4) for the left. The low values of SNR were likely to be due to artefacts from fluid motion on the subtraction images used for noise calculations. Conclusion: The methods proposed allowed an independent measure of suppression effectiveness. More work is needed to ensure a reliable measurement of SNR. P59 Breast clinicians: their contribution to the breast screening programme in England and implications of the current age structure C Swinson Beds & Herts Breast Screening Service, Luton, UK Breast Cancer Research 2012, 14(Suppl 1):P59 Table 1(abstract P57) Pathology Number Odds ratio P value 95% CI Atypia 53 7.48 < 0.001 2.62 to 21.39 AIDP 10 4.87 0.034* 1.13 to 21.01 ADH 11 2.50 0.167 0.68 to 9.24 ALH 6 1.28 0.793 0.21 to 7.93 FEA 8 5.90 0.025* 1.25 to 27.67 CCC 31 5.10 0.001* 1.89 to 13.76 Papillary 14 0.57 0.44 0.14 to 2.38 LISN 30 0.55 0.28 0.18 to 1.64 RS/CSL 13 0.92 0.91 0.23 to 3.74 Epithelial proliferation 20 2.07 0.172 0.73 to 5.88 Mucocele-like 6 1.91 0.48 0.32 to 11.41 Fibroepithelial 10 No malignant diagnoses Haemangioma 1 No malignant diagnoses Spindle cell 2 No malignant diagnoses Significant P value. g , , Breast Cancer Research 2012, 14(Suppl 1):P59 Introduction: Since the role of breast clinician was established in 1987 to support introduction of the NHSBSP, the role has developed and breast clinicians are now important members of the MDT, skilled in the full range of screening work. With a shortage of radiologists this input must continue, but recruitment may be hindered by a restrictive contractual structure. This study set out to establish the contribution of breast clinicians to the screening programme, and their age distribution. g p g g Methods: In July 2012 an email survey of members of the Association of Breast Clinicians was conducted to determine staffing, film-reading workload and age distribution data in English screening units. Should the breast be a check area on standard CT thorax examinations? The IF cohort were more likely to have a lesion (47.6% vs. 28.7%, P < 0.001) and to have a MA (18.4% vs. 6.7%, P < 0.0001). The only demographic factor of the IF group with a statistically significant association with MA was age >65 years (P = 0.0063). Conclusion: Breast screening identifies three-quarters of breast cancers in the screening population. Interval cancers present with increasing frequency through the screening cycle and are faster growing, pathologically more aggressive lesions. Conclusion: While patients with breast IFs are more likely to have a MA, only 47% of those referred actually had a lesion. Age was the only statistically significant factor that correlated with a MA. Whilst a worthy check area clinicians should exercise caution to avoid unnecessary procedures, particularly in younger patients. P53 Abstract withdrawn Breast Cancer Research 2012, 14(Suppl 1):P53 Abstract not submitted for online publication P54 Indications for marker clip in the setting of neoadjuvant/neoendocrine therapy ME Fletcher, BJG Dall, N Sharma Leeds Teaching Hospital NHS Trust, Leeds, UK Breast Cancer Research 2012, 14(Suppl 1):P54 Introduction: We are aware that the use of neoadjuvant therapy (NAC and neoendocrine therapy (NAET) is steadily increasing and mo P53 Abstract withdrawn P53 Abstract withdrawn P56 Node-positive breast cancers diagnosed in 2011 at York Teaching Hospitals NHS Trust: an analysis of the adequacy of current preoperative assessment A Chandra York Hospital NHS Foundation Trust, York, UK Breast Cancer Research 2012, 14(Suppl 1):P56 Breast Cancer Research 2012, 14(Suppl 1):P53 Introduction: Preoperative assessment of the axillary nodal status in primary breast cancer using ultrasonagraphy is now established practice and mandated by the NHSBSP. However, the criteria for nodal biopsy are based on the morphological appearance of the imaged nodes. This study’s poster proposes that the tumour and grade stage of the primary tumour should also influence the threshold for nodal biopsy with the intention of improving preoperative assessment and thus decreasing morbidity Introduction: We are aware that the use of neoadjuvant therapy (NACT) and neoendocrine therapy (NAET) is steadily increasing and more Page 13 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Results: A total of 205 B3 lesions were identified; 112 lesions with subsequent excision biopsy were analysed. Patients had mean age of 56 years (95% CI = 55 to 57 years). Thirty out of 112 lesions were upgraded to B5 diagnosis. Significant P value. Should the breast be a check area on standard CT thorax examinations? * Nine out of 112 had final diagnosis of LCIS. Multivariate analysis of odds ratios for malignancy, after age adjusting, is shown in Table 1. associated with further surgical intervention to the axilla following positive sentinel lymph node biopsy. associated with further surgical intervention to the axilla following positive sentinel lymph node biopsy. Methods: A retrospective analysis of final nodal status of the 296 surgically treated patients diagnosed with primary breast cancer in 2011 was performed. Data including preoperative assessment of tumour size, grade, axillary node status (preoperative and postoperative) and final outcome was collected. Patients with negative preoperative axillary node status were compared with postoperative node status and the proportion requiring further treatment was ascertained. Conclusion: Atypia on core biopsy significantly predicts malignancy, with 7.48 times the odds of malignancy compared with lesions without atypia (P < 0.001; 95% CI = 2.62 to 21.39). Similarly, lesions containing AIDP, FEA and columnar cell change (CCC) have significantly increased odds for malignancy. LISN did not confer an increased risk of malignancy. Stratifying lesions in this way can direct future management. Results: A total of 207 had negative axillary preoperative assessment and underwent sentinel lymph node biopsy (SLNB). Fifty-one had positive sentinel lymph nodes, and of these 9% required further surgery. Sensitivity at the identification of positive axillary nodal disease preoperatively was 0.49 and specificity calculated at 0.99. Preoperative assessment had a positive predictive value of 96% and a negative predictive value of 73%. A positive correlation between axillary nodal involvement and tumour size and grade was observed; that is, 15% in T1G1 tumours to 36% in T3G3 tumours. Patients in the T2G2 group or above were 54% more likely to have a positive SLNB. P62 Mammograms in 35-year-old to 39-year-old symptomatic patients KA Lim, P Young Cardiff and Vale University Health Board, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P62 g Cardiff and Vale University Health Board, Cardiff, UK Breast Cancer Research 2012, 14(Suppl 1):P62 Introduction: Currently, in our institution, all women presenting at the Triple Assessment Symptomatic Breast Clinic over 35 years of age undergo two-view mammograms. Recent recommendations suggest that mammograms should only be performed above the age of 40 [1]. We reviewed our patients to see whether changing our policy would have resulted in any missed cancers. Methods: The aim of this study was to audit the use and effectiveness of ultrasound and stereotaxis guided VACB at a single centre over a 3-year period. We retrospectively identified patients from the Leicestershire Breast Screening Service and Radiology Information Systems records. A total of 152 patients undergoing 157 VACB (n = 157) were identified. Of these, 133 were stereotaxis guided, the remaining 24 ultrasound guided. The overall lesion workload was as follows: asymmetrical densities = 6(4%); mass = 31(20%); parenchymal distortion = 32(20%); and calcification = 88(56%). Methods: All patients on our Breast Cancer Database diagnosed between April 2002 and March 2012, aged between 35 and 39 years at diagnosis, were identified. Patient records were reviewed to determine how the cancer was diagnosed and treated. p y Results: A positive histological diagnosis was achieved in 153 (97%) of the total VACBs performed. Of these patients, 136 (88.9%) had a prior 14G core biopsy, 16 (10.5%) had VACB as a first-line procedure. Outcomes for VACB were then compared against 14G biopsies for individual histological grade, presented as follows - grade on 14G needle (definitive diagnosis percentage on VACB): B1 (83%); B2 (67%); B3 (62%); B4 (76%); and B5c (100%). Overall, 51 (33.6%) patients avoided surgery for benign breast disease (95% CI = 26.1 to 41.1%), and 33 patients (21.7%) had single definitive surgery thereby avoiding multiple procedures (95% CI = 15.2 to 28.3%). Results: Fifty-one patients were identified; 11 were diagnosed elsewhere, and therefore excluded, and one was screen-detected and also excluded. Of the remaining 39 patients, one patient was male. In 33/39 (84.6%) patients, ultrasound was performed of a palpable lesion and mammography did not add any additional diagnostic value. Of the remaining six patients, three had obvious clinical signs of malignancy. P60 Vacuum-assisted core biopsy of the breast: a 3-year single-centre experience Conclusion: Prophylactic mastectomy precludes the use of sentinel node biopsy as the breast is required for this procedure in order for the dye and tracer to be injected into the breast. Screening MRI prior to prophylactic mastectomy is essential in order for a sentinel node to be feasible in case a small tumour is present occult on conventional imaging. Introduction: Vacuum-assisted core biopsy (VACB) of the breast is a minimally invasive technique, used increasingly for the assessment of mammographically and ultrasound detected, nonpalpable breast lesions. The effectiveness of VACB has been demonstrated on lesions both with and without microcalcifications. VACB allows the operator to obtain a sufficient specimen with a single insertion to provide for a more accurate diagnosis. P62 The remaining three patients had malignancy diagnosed as the result of their mammograms. Two (5.1%) had mammographic microcalcification due to intermediate-grade DCIS and one patient had an impalpable cancer, seen on the mammogram, which prompted the diagnostic ultrasound. Conclusion: VACB is a safe, well-tolerated and extremely useful sampling modality which can, in selective cases, accurately determine onward surgical management and avoid unnecessary surgery for benign breast disease. Conclusion: In our institution, over a 10-year period, two cases of intermediate-grade DCIS and one case of Grade 2 invasive ductal carcinoma would have been missed if routine mammography had not been performed in the 35 to 39 year age group. P61 Screening breast MRI in high-risk women undergoing prophylactic mastectomy SE McWilliams Guy’s and St Thomas’ NHS Foundation Trust, London, UK Breast Cancer Research 2012, 14(Suppl 1):P61 Reference P58 Responses were received from 16 units, around 90% of units with breast clinician film- readers. Results: The 16 units were mainly in the south and included some of the largest screening units. Overall, there were 30 breast clinician film- readers, 56 consultant radiologists and 46 radiographer film-readers. Numbers of clinicians in units varied from one to five, representing 9 to 50% of all film-readers. Breast clinicians read from 9 to 66% of films, and Page 14 of 14 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 Breast Cancer Research 2012, Volume 14 Suppl 1 http://breast-cancer-research.com/supplements/14/S1 women. We look at the use of screening MRI in 16 women at our institution prior to surgery to see how this affected management. in 12 units read proportionately more films than expected from their number. Twenty-two (73%) of the clinicians were aged over 50 and 14 (47%) over 55. g y g Methods: Sixteen women aged 25 to 33 years underwent breast MRI prior to prophylactic mastectomy. The women were high risk with proven BRCA 1 or 2 mutations. The images were independently double-reported by two certified board breast radiologists. The patients also underwent bilateral mammogram. Conclusion: Breast clinicians make a substantial contribution to screening units in England, particularly in the south. Retirements over the next 5 to 10 years will add considerably to manpower pressures on the NHSBSP. Recruitment will be dependent upon an attractive career structure and robust training. g Results: Of the 16 women, one patient had a small tumour selected on MRI that was occult on conventional imaging and was able to undergo sentinel node biopsy. In the remaining 15 patients, one small 3 mm lesion was found at surgery and the patient had to have axillary dissection. Conclusion: Prophylactic mastectomy precludes the use of sentinel node biopsy as the breast is required for this procedure in order for the dye and tracer to be injected into the breast. Screening MRI prior to prophylactic mastectomy is essential in order for a sentinel node to be feasible in case a small tumour is present occult on conventional imaging. Results: Of the 16 women, one patient had a small tumour selected on MRI that was occult on conventional imaging and was able to undergo sentinel node biopsy. In the remaining 15 patients, one small 3 mm lesion was found at surgery and the patient had to have axillary dissection. Reference 1. Best Practice Diagnostic Guidelines for Patients Presenting with Breast Symptoms London: Department of Health: Willett AM, Michell MJ, Lee MJR 2010. Introduction: High-risk women with genetic predisposition for breast cancer are increasingly being offered bilateral prophylactic mastectomy for risk reduction. If an incidental tumour is found at surgery, the patient cannot then undergo sentinel node biopsy of the axilla and has to have a full axillary clearance associated with morbidity of lymphoedema often in young Cite abstracts in this supplement using the relevant abstract number, e.g.: Lim and Young: Mammograms in 35-year-old to 39-year-old symptomatic patients. Breast Cancer Research 2012, 14(Suppl 1):P62 Cite abstracts in this supplement using the relevant abstract number, e.g.: Lim and Young: Mammograms in 35-year-old to 39-year-old symptomatic patients. Breast Cancer Research 2012, 14(Suppl 1):P62
https://openalex.org/W2128619130	https://nottingham-repository.worktribe.com/preview/742677/iaai15.pdf	English	null	Using Qualitative Spatial Logic for Validating Crowd-Sourced Geospatial Data	Proceedings of the ... AAAI Conference on Artificial Intelligence	2,015	cc-by	6,352	Copyright c⃝2015, Association for the Advancement of Artiﬁcial Intelligence (www.aaai.org). All rights reserved. Abstract by OSGB (Ordnance Survey 2012) and by OSM (Open- StreetMap 2014) shown in Fig.1. Fig.1(a) shows the repre- sentation of a restaurant Prezzo Ristorante in both OSGB and OSM. In OSGB, the object has geometry shown in light colour and the label ‘PREZZO RISTORANTE, 21-23 FORMAN STREET, RESTAURANT’; in OSM the object has geometry shown in dark colour and the label ‘Prezzo Ristorante, restaurant’. Fig.1(b) shows the representation of a shopping centre in Nottingham city centre. The OSGB dataset contains an object with geometry shown in light colour and the label ‘JOHN LEWIS, 175-182, VICTORIA CENTRE, DEPARTMENT STORE’; the OSM dataset has only one object (shown in dark colour) corresponding to the shopping centre that contains this store along with several others, with the label ‘Victoria Centre’. In order to match ob- jects in the datasets, we need to determine which objects are the same and sometimes (as in the example of John Lewis and Victoria Centre) which objects in one dataset are parts of objects in another. We describe a tool, MatchMaps, that generates sameAs and partOf matches between spatial objects (such as shops, shopping centres, etc.) in crowd-sourced and au- thoritative geospatial datasets. MatchMaps uses reason- ing in qualitative spatial logic, description logic and truth maintenance techniques, to produce a consistent set of matches. We report the results of an initial eval- uation of MatchMaps by experts from Ordnance Sur- vey (Great Britain’s National Mapping Authority). In both the case studies considered, MatchMaps was able to correctly match spatial objects (high precision and recall) with minimal human intervention. tive Spatial Logic for Validating Crowd-Sourced Geospatial Data Hai Nguyen University of Aberdeen Aberdeen, UK hai.nguyen@abdn.ac.uk Natasha Alechina University of Nottingham Nottingham, UK nza@cs.nott.ac.uk Heshan Du University of Nottingham Nottingham, UK hxd@cs.nott.ac.uk Heshan Du University of Nottingham Nottingham, UK hxd@cs.nott.ac.uk Brian Logan University of Nottingham Nottingham, UK bsl@cs.nott.ac.uk Michael Jackson Nottingham Geospatial Institute Nottingham, UK Mike.Jackson@nottingham.ac.uk Brian Logan University of Nottingham Nottingham, UK bsl@cs.nott.ac.uk Michael Jackson Nottingham Geospatial Institute Nottingham, UK Mike.Jackson@nottingham.ac.uk John Goodwin Ordnance Survey Southampton, UK John.Goodwin@ordnancesurvey.co.uk Proceedings of the Twenty-Seventh Conference on Innovative Applications of Artificial Intelligence Proceedings of the Twenty-Seventh Conference on Innovative Applications of Artificial Intelligence Introduction Crowd-sourced or volunteered geospatial data has emerged and developed rapidly in recent years. Crowd-sourced data involves non-specialists in geospatial science, collecting, editing and displaying geospatial data. Compared to au- thoritative data from national mapping agencies, e.g. Ord- nance Survey of Great Britain (OSGB) (Ordnance Sur- vey 2012), crowd-sourced data, e.g. from OpenStreetMap (OSM) (OpenStreetMap 2014), is often less accurate, but could contain more recent and richer user-based information (Jackson, Rahemtulla, and Morley 2010). After it is vali- dated by generating correspondences with authoritative data, this information can then be used to extend and enrich au- thoritative data. Figure 1: a. Prezzo Ristorante; b. John Lewis in Victoria Shopping Centre Establishing correspondences (matches) between spatial objects is a non-trivial task. One of the challenges is that the geometry representations of the same place in differ- ent datasets are usually not exactly the same. The datasets may use different coordinate systems and/or vector rep- resentations. More importantly, the accuracy differs be- tween datasets, and objects may be represented at differ- ent levels of granularity. As an example, consider the ge- ometries of two objects in Nottingham city centre given Figure 1: a. Prezzo Ristorante; b. John Lewis in Victoria Shopping Centre A spatial object in a geospatial dataset has an ID, location information (coordinates and geometry) and meaningful la- bels, such as names or types, and represents an object in the real world. The problem of matching two sets of spatial objects can be stated as follows: given two sets of spatial objects A and B, ﬁnd the set S of all true matches: sen- tences of the form sameAs(a, b) and partOf (a, b) (where a ∈A, b ∈B). A sentence sameAs(a, b) is true if a 3948 Figure 2: a. a buffer; b. three hatched circles are buffered part of (BPT) the solid circle; c. buffered equal (BEQ) and b refer to the same object in the real world. Similarly, partOf (a, b) is true if the object represented by a is part of the object represented by b in the real world. j p y We present a tool, MatchMaps, which generates sameAs and partOf matches between spatial objects. A preliminary version of MatchMaps (previously called GeoMap) was de- scribed in (Du et al. 2013b); this paper presents a version with signiﬁcantly modiﬁed match generation algorithms and the addition of qualitative spatial reasoning. Buffered Equal and PartOf MatchMaps comprises seven main steps summarised be- low. Before we describe MatchMaps and the logical reasoning involved in it, we need to brieﬂy introduce some relevant notions. As mentioned above, the geometries of the same objects in different datasets are not necessarily exactly the same. This is due to differences in representations, coordi- nate systems, and errors or imprecision in crowd-sourced data. Hence we use buffers (ISO Technical Committe 211 2003) to relax equality and the inclusion relation between geometries. 1. Generate disjointness axioms between concepts in TA and TB. A disjointness axiom is a statement of the form C1 ⊓C2 = ∅(the sets of C1 objects and C2 objects are disjoint; for example, Library ⊓Pub = ∅). The disjoint- ness axioms are generated automatically by assuming the disjointness of sibling concepts in each ontology. We also manually generate a small set of axioms that prohibit ob- jects of one type being partOf objects of another type: School ⊑∀partOf .¬Pub (if something is a School, then for all objects it is a part of, they are not Pubs). We use the description logic reasoner Pellet (Sirin et al. 2007) to check that adding a set of disjointness and ‘partOf - disjointness’ axioms DA to TA does not result in incoher- ence (existence of provably unsatisﬁable concepts), sim- ilarly for DB ∪TB. Axioms that cause incoherence are removed from DA and DB, resulting in D = DA ∪DB. This does not require human interaction. This is an auxil- iary step that is needed to facilitate discovering problem- atic matches (such as a sameAs match between a and b where a is a Library and b is a Pub). As shown in Fig. 2(a), by buffering the solid circle x by σ, we obtain a larger circle, denoted as buﬀer(x, σ), where every point is within σ distance from x. A geometry a is a buffered part of (BPT) of x if all points of a are within the buffer of x (Fig. 2(b)). Two geometries are buffered equal (BEQ) with respect to σ, if they fall into the σ-buffer of each other (Fig. 2(c)). If two geometries are BEQ with respect to an appropriate level of tolerance σ, then they probably repre- sent the same real world location, otherwise, they represent different locations. Introduction MatchMaps works by generating candidate matches and checking their consistency with respect to several sets of sentences: on- tology axioms, correspondences between type descriptions across ontologies, disjointness of types, and qualitative spa- tial relations between objects. If an inconsistency is found, a minimal set of statements required to derive it is produced. One of the statements in this set must be retracted to restore consistency. The decision about which statement in the min- imal inconsistent set is incorrect and should be retracted is made by a human expert, as no heuristic for making this de- cision automatically gives sufﬁciently reliable results. Figure 2: a. a buffer; b. three hatched circles are buffered part of (BPT) the solid circle; c. buffered equal (BEQ) of datasets. For example, in the case of Nottingham city cen- tre, the value of σ for the OSGB and OSM datasets is 20 m. Interestingly, OSM positional accuracy (maximal error) has been estimated to be about 20 m in UK (Haklay 2010). MatchMaps The main idea underlying MatchMaps is to generate a set of candidate matches between spatial objects in different datasets (‘assumptions’) and check for their logical consis- tency with respect to other available information. The aim is to produce a maximally consistent subset of candidate matches which is as close as possible to the set of all true matches. In this paper, we present an evaluation of MatchMaps on two case studies. We report the precision (correctness of matches) and recall (how close the set of generated matches is to the set of all true matches) of the set of matches be- tween OSGB and OSM representations of Nottingham and of Southampton city centres relative to ground truth. In both case studies, MatchMaps was able to correctly match spatial objects (precision 90% for Nottingham and 98% for Southampton, recall 84% for Nottingham and 97% for Southampton) with minimal human intervention. The input to MatchMaps is two sets of spatial objects A and B and two ontologies TA and TB deﬁning concepts for describing them. (Note that crowd-sourced datasets do not have formal ontologies, but it is easy to generate simple in- clusion hierarchies from their tags.) The output is a set S of sameAs and partOf matches between objects. Buffered Equal and PartOf As shown in Fig.3(a), there is a clear correspondence between aggre- gated geometries from OSM (dark) and OSGB (light, dot- ted). However, for a single OSM geometry (dark), there can be more than one candidate from OSGB (dotted), as shown in Fig.3(b). We cannot decide which one is correct based only on the similarity of geometries. This type of problem often occurs when matching, for example, ter- raced houses or small shops in a shopping centre. This step does not require human interaction. 7. Validate matches using classiﬁcation. Check for consis- tency of S together with TA ∪TB ∪D ∪M. Restore con- sistency and return the resulting set S. This step requires human interaction since either a match in S or a disjoint- ness axiom may be wrong. 7. Validate matches using classiﬁcation. Check for consis- tency of S together with TA ∪TB ∪D ∪M. Restore con- sistency and return the resulting set S. This step requires human interaction since either a match in S or a disjoint- ness axiom may be wrong. 7. Validate matches using classiﬁcation. Check for consis- tency of S together with TA ∪TB ∪D ∪M. Restore con- sistency and return the resulting set S. This step requires human interaction since either a match in S or a disjoint- ness axiom may be wrong. In addition to Pellet, the LBPT reasoner and the ATMS, the implementation of MatchMaps builds on a number of ex- isting tools. The JTS Topology Suite (Vivid Solutions, Inc. 2014) is used to process two dimensional geometries, and the graphical user interface is implemented using the Open- JUMP libraries (JPP 2014). 4. Generate object matches (sameAs and partOf matches between objects). For each pair of matched geometries from the previous step, we consider associated objects and check them for similarity of labels using a string simi- larity measure. In straightforward cases, when there are two objects a and b with similar geometries and simi- lar labels in different datasets, we add sameAs(a, b) to a set of candidate matches S; or if there is a set of ob- jects {a1, . . . , an} where the union of their geometries is similar to a geometry of a single object b in another dataset, we add partOf (ai, b) to S for every ai. Buffered Equal and PartOf A dif- ﬁcult case is when there is a match between two aggre- gated geometries which contain objects {a1, . . . , an} in one dataset and objects {b1, . . . , bk} in another dataset (many-to-many matching case). When we cannot decide the exact matches automatically using labels and types of objects, we generate all matches which are possibly correct between the objects in the two sets: for each pair ai, bj with similar labels, we generate sameAs(ai, bj), Buffered Equal and PartOf The level of tolerance σ intuitively corresponds to how different geometry representations of the same spatial fea- tures can be. The value of σ can be established empiri- cally by looking at the two datasets side by side and align- ing geometries of several features which are known to be the same in both datasets. The maximal value required to make any two geometric representations of the same object buffered equal (BEQ) gives the value of σ for the given pair 2. Generate terminology matches between concepts in TA and TB of the form C1 ≡C2 where C1 is a concept in TA and C2 is a concept in TB. Currently we generate termi- nology matches automatically using a very simple heuris- tic based on similarity of concept names. For example, 3949 Figure 3: a. matching ‘aggregated’ geometries; b. matching single geometries partOf (ai, bj), partOf (bj, ai). The output of this step is the set S of candidate matches. This step does not require human interaction. 5. Validate matches using LBPT. Check S for consistency using a qualitative spatial logic, Logic of ParT and whole for Buffered geometries (LBPT), which is explained in the next section. If an inconsistency is found, we retract sameAs or partOf matches from S to restore consis- tency. This step is implemented using a dedicated LBPT reasoner and an Assumption-Based Truth Maintenance System (ATMS). This step may require human interaction to decide which matches to remove. Figure 3: a. matching ‘aggregated’ geometries; b. matching single geometries OSGB : Shop ≡OSM : Shop. The set of terminology matches is M. We check coherence of TA ∪TB ∪D ∪M using Pellet. For every set of statements responsible for incoherence, we remove one of the statements in D ∪M. This step requires human interaction, because sometimes we need to decide whether to remove a terminology match or a disjointness axiom to restore coherence. OSGB : Shop ≡OSM : Shop. The set of terminology matches is M. We check coherence of TA ∪TB ∪D ∪M using Pellet. For every set of statements responsible for incoherence, we remove one of the statements in D ∪M. This step requires human interaction, because sometimes we need to decide whether to remove a terminology match or a disjointness axiom to restore coherence. 6. Validate matches using UNA/NPH. Check S for consis- tency with respect to UNA or NPH. Buffered Equal and PartOf UNA refers to Unique Name Assumption: for each dataset, sameAs(a1, a2) does not hold for a1 and a2 with different IDs (each object is represented exactly once). NPH (No PartOf Hierarchy) is a stronger assumption that states that there are no ob- jects b1, b2 in the same set such that partOf (b1, b2) holds. UNA and NPH hold for the OSGB data. However UNA or NPH can be violated in the OSM data. Therefore this check is ‘soft’: if in a crowd-sourced dataset some object is represented twice, or there is a genuine partOf relation- ship determined by human checking, we skip this ‘error’ and do not retract any assumptions. This step is required since even after consistency checks in the previous steps, there may be ‘too many’ matches in S. It is implemented using Pellet, and requires human interaction. This is an optional step, which could be skipped if UNA or NPT is violated frequently in at least one input dataset. 6. Validate matches using UNA/NPH. Check S for consis- tency with respect to UNA or NPH. UNA refers to Unique Name Assumption: for each dataset, sameAs(a1, a2) does not hold for a1 and a2 with different IDs (each object is represented exactly once). NPH (No PartOf Hierarchy) is a stronger assumption that states that there are no ob- jects b1, b2 in the same set such that partOf (b1, b2) holds. UNA and NPH hold for the OSGB data. However UNA or NPH can be violated in the OSM data. Therefore this check is ‘soft’: if in a crowd-sourced dataset some object is represented twice, or there is a genuine partOf relation- ship determined by human checking, we skip this ‘error’ and do not retract any assumptions. This step is required since even after consistency checks in the previous steps, there may be ‘too many’ matches in S. It is implemented using Pellet, and requires human interaction. This is an optional step, which could be skipped if UNA or NPT is violated frequently in at least one input dataset. 3. Generate geometry matches using aggregation and buffering. This is done using standard 2D spatial tools (Vivid Solutions, Inc. 2014) to aggregate, buffer and check for inclusions of their geometries. Buffered Equal and PartOf Rather than matching every single geometry in input datasets, we generate BEQ correspondences between ‘aggregated’ ge- ometries, each of which is obtained by aggregating a non-empty collection of adjacent single geometries. The reason for doing the aggregation step is that sometimes matching every single geometry is impossible while there is a clear match between two aggregates. As shown in Fig.3(a), there is a clear correspondence between aggre- gated geometries from OSM (dark) and OSGB (light, dot- ted). However, for a single OSM geometry (dark), there can be more than one candidate from OSGB (dotted), as shown in Fig.3(b). We cannot decide which one is correct based only on the similarity of geometries. This type of problem often occurs when matching, for example, ter- raced houses or small shops in a shopping centre. This step does not require human interaction. 3. Generate geometry matches using aggregation and buffering. This is done using standard 2D spatial tools (Vivid Solutions, Inc. 2014) to aggregate, buffer and check for inclusions of their geometries. Rather than matching every single geometry in input datasets, we generate BEQ correspondences between ‘aggregated’ ge- ometries, each of which is obtained by aggregating a non-empty collection of adjacent single geometries. The reason for doing the aggregation step is that sometimes matching every single geometry is impossible while there is a clear match between two aggregates. As shown in Fig.3(a), there is a clear correspondence between aggre- gated geometries from OSM (dark) and OSGB (light, dot- ted). However, for a single OSM geometry (dark), there can be more than one candidate from OSGB (dotted), as shown in Fig.3(b). We cannot decide which one is correct based only on the similarity of geometries. This type of problem often occurs when matching, for example, ter- raced houses or small shops in a shopping centre. This step does not require human interaction. 3. Generate geometry matches using aggregation and buffering. This is done using standard 2D spatial tools (Vivid Solutions, Inc. 2014) to aggregate, buffer and check for inclusions of their geometries. Rather than matching every single geometry in input datasets, we generate BEQ correspondences between ‘aggregated’ ge- ometries, each of which is obtained by aggregating a non-empty collection of adjacent single geometries. The reason for doing the aggregation step is that sometimes matching every single geometry is impossible while there is a clear match between two aggregates. Qualitative Spatial Logic In this section we outline the qualitative spatial logic LBPT used to detect errors in matches, and brieﬂy give the intu- itions behind the logic. The logic itself is presented in (Du and Alechina 2014). The language of LBPT contains a set of individ- ual names (for individual geometries), binary predicates NEAR, FAR and BPT, and logical connectives, ¬ (not), ∧(and) and →(implies). As introduced before, BPT and BEQ stand for ‘buffered partOf’ and ‘buffered equal’ respectively. BEQ(a, b) is deﬁned as BPT(a, b) ∧ BPT(b, a). The relations NEAR and FAR are also de- ﬁned using a level of tolerance σ. Two geometries a and b are NEAR if there exist a point pa in a and a point pb in b such that the distance between pa and pb is at most 2σ. The 3950 it). Each node has a label that records the minimal sets of assumptions from which it can be derived. intuition is that if a and b are shifted towards each other by σ they will deﬁnitely be touching. Two geometries a and b are FAR if for any two points pa in a and pb in b, their distance is strictly greater than 4σ. NEAR and FAR are disjoint but not mutually exhaustive relations. For two points which are at the distance of 3σ, they are neither NEAR nor FAR. Whenever a new justiﬁcation is received from the problem-solver, the labels of nodes in the graph are updated to maintain four key properties: • soundness if a set of assumptions is in the label of a for- mula, then the reasoner has found a derivation of this for- mula which only uses those assumptions; y A sound and complete axiomatisation of LBPT consists of the following axioms: A0 All tautologies of classical propositional logic • completeness all ways of deriving the formula discovered by the reasoner so far are included in the label; A0 All tautologies of classical propositional logic A1 BPT(a, a); A2 NEAR(a, b) →NEAR(b, a); A3 FAR(a, b) →FAR(b, a); A4 BPT(a, b) ∧BPT(b, c) →NEAR(c, a); A5 BPT(b, a) ∧BPT(b, c) →NEAR(c, a); A6 BPT(b, a) ∧NEAR(b, c) ∧BPT(c, d) →¬FAR(d, a); A7 NEAR(a, b) ∧BPT(b, c) ∧BPT(c, d) →¬FAR(d, a); MP Modus ponens: φ, φ →ψ ⊢ψ. MP Modus ponens: φ, φ →ψ ⊢ψ. MP Modus ponens: φ, φ →ψ ⊢ψ. Initially, LBPT reasoning was implemented using Pellet. LBPT reasoning rules were encoded as constraints on object properties. However, not all LBPT reasoning rules are ex- pressible in this way, and we also encountered some bugs (for example, non-minimal explanations) in the explana- tions generated using reasoning with role chain inclusions. We therefore implemented a dedicated LBPT reasoner inte- grated with an assumption-based truth maintenance system (ATMS) (de Kleer 1986). The LBPT reasoner derives con- sequences by applying inference rules to previously derived sentences, and the ATMS maintains dependencies between derived consequences and a set of distinguished data known as assumptions. In particular, it maintains all minimal sets of assumptions responsible for the derivation of ⊥(false), referred to as nogoods in ATMS terminology. We therefore implemented a dedicated LBPT reasoner inte- grated with an assumption-based truth maintenance system (ATMS) (de Kleer 1986). The LBPT reasoner derives con- sequences by applying inference rules to previously derived sentences, and the ATMS maintains dependencies between derived consequences and a set of distinguished data known as assumptions. In particular, it maintains all minimal sets of assumptions responsible for the derivation of ⊥(false), referred to as nogoods in ATMS terminology. The LBPT reasoner with an ATMS is implemented in Pop-11 (http://www.cs.bham.ac.uk/research/projects/ poplog/freepoplog.html). The LBPT inference rules are implemented using Poprulebase, a Pop-11 forward chaining rule interpreter. Qualitative Spatial Logic A1 BPT(a, a); • minimality the formula (so far) has not been discovered to be derivable from a strict subset of any set of assump- tions in the label; A3 FAR(a, b) →FAR(b, a); • consistency if a set of assumptions is discovered to be in- consistent, then it is removed from the labels of all nodes (except n⊥). The basic task of the ATMS is to compute sound, com- plete, minimal and consistent labels for datum nodes which correctly reﬂect the sequence of justiﬁcations generated by the reasoner. From the label of a node we can therefore de- termine the minimal sets of assumptions needed to (consis- tently) derive the corresponding datum discovered by the reasoner. From the label of the false node n⊥we can dis- cover the reasons for any inconsistencies in the ontology (i.e., the ATMS nogoods). Experiments and Evaluation In this section, we report the use of MatchMaps to match OSM data (building layer) (OpenStreetMap 2014) to OSGB MasterMap data (Address Layer and Topology Layer) (Ord- nance Survey 2012). The studied areas are in city centres of Nottingham UK and Southampton UK. The Nottingham data was obtained in 2012, and the Southampton data in 2013. The numbers of spatial objects in the case studies are shown in Table 1. The number of OSM objects is smaller in each case, because OSM data often describes a collection of OSGB objects as a whole, for example, OSGB shops as a shopping centre in OSM. The reasoner reasons using inference rules which encode the LBPT axioms. It repeatedly checks whether any of its inference rules are applicable to the assumptions and/or any previously derived sentences, and if so, it sends to the ATMS the consequence of applying the rule together with a justiﬁ- cation consisting of the name of the rule and the antecedents of the rule used to derive the consequence. The newly de- rived consequence can now be used as an input to the rea- soner and the cycle repeats until no new consequences can be derived. The ATMS computes dependency (or derivability) of data from sets of assumptions. In our case statements of matches between objects are treated as assumptions and NEAR and FAR statements as premises (non-retractable statements). To compute derivability, the ATMS builds and maintains a de- pendency or justiﬁcation graph. A node in the graph repre- sents a unique datum (i.e., a formula). Inconsistency is rep- resented by a distinguished node n⊥. Justiﬁcations form the edges of the graph and record the fact that a datum node (the consequent) can be derived from a set of other datum nodes (the antecedents). A node may be the consequent of more than one justiﬁcation (recording the different ways in which it can be derived), and be an antecedent in other jus- tiﬁcations (recording the inferences that can be made using Table 1: Data used for evaluation OSM spatial objects OSGB spatial objects Nottingham 281 13204 Southampton 2130 7678 nogoods retracted BEQ/BPT interactions Nottingham 172 31 3 Southampton 268 114 7 TP FP TN FN Precision Recall Nottingham 177 19 64 21 0.90 0.84 Southampton 1997 21 71 41 0.98 0.97 σ = 20m. If a BEQ or BPT assumption is involved in a derivation of inconsistency and is retracted, then the corre- sponding sameAs or partOf matches are also retracted. performed on an Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00 GHz, 4.00 GB RAM desktop computer. Times are in seconds, averaged over 5 runs. performed on an Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00 GHz, 4.00 GB RAM desktop computer. Times are in seconds, averaged over 5 runs. Table 3 shows the number of nogoods generated by the LBPT reasoner with an ATMS. The number of interactions is the number of times users are asked to take actions or use strategies to resolve problems (a strategy is a heuristic which allows users to retract all ‘similar’ statements at a time, for example, all statements of the form partOf (x, o) for some small object o). As a result of LBPT reasoning and removal of BEQ and BPT assumptions, we also withdraw 1325 sameAs/partOf assumptions for Nottingham and 488 sameAs/partOf assumptions for Southampton. The main objective of evaluation was to establish the precision and recall of MatchMaps. Given the size of the case studies, it was infeasible for domain experts to pro- duce a complete set of ground truth matches manually. In- stead, we computed the ground truth as follows. For each OSM object a, we check all matches which involve a (ei- ther a single sameAs(a, b) match with some b in the OSGB dataset, or several partOf matches involving a) produced by MatchMaps. If the match or matches were determined by a human expert to be correct, a was classiﬁed as ‘Correctly Matched’ (True Positive or TP), otherwise it was classi- ﬁed as ‘Incorrectly Matched’ (False Positive or FP). For a ∈FP, a check was made whether a correct match for a existed; if yes, a was labelled FPsbm. If a was not involved in any matches, a check was made whether a correct match for it existed. If there was no correct match, then a was placed in ‘Correctly Not-matched’ (True Negative or TN), otherwise in ‘Incorrectly Not-matched’ (False Negative or FN). Straightforward matches were checked by a non- expert using guidelines developed in conjunction with a sub- ject matter expert from the Nottingham Geospatial Institute. Table 2: Matching OSM spatial objects to OSGB Table 3: LBPT reasoning nogoods retracted BEQ/BPT interactions Nottingham 172 31 3 Southampton 268 114 7 nogoods retracted BEQ/BPT interactions Nottingham 172 31 3 Southampton 268 114 7 Table 5: UNA and NPH reasoning by Pellet retracted sameAs/partOf interactions Nottingham 151 11 Southampton 199 12 Table 5: UNA and NPH reasoning by Pellet We also performed experiments to evaluate the perfor- mance of the qualitative spatial reasoning step (step 5). In that step, we are primarily interested in consistency of matches generated in step 4 for the many-to-many case, that is, sameAs and partOf matches between two sets of ob- jects {a1, . . . , an} and {b1, . . . , bk} which belong to two buffered equal aggregated geometries. To check consistency of those matches with respect to LBPT, we translate matches to LBPT. For every sameAs(ai, bj), we generate an as- sumption BEQ(g(ai), g(bj)) and for every partOf (ai, bj), we generate an assumption BPT(g(ai), g(bj)), where g(ai) and g(bj) are geometries of ai and bj, respectively. We also generate NEAR and FAR facts between all objects in {a1, . . . , an} and between all objects in {b1, . . . , bk}, from distances between geometries within the same dataset, using retracted sameAs/partOf interactions Nottingham 151 11 Southampton 199 12 nogoods retracted BEQ/BPT interactions Nottingham 172 31 3 Southampton 268 114 7 A subject matter expert at Ordnance Survey (Great Britain’s National Mapping Authority) classiﬁed non-straightforward cases (approximately 10% of the total output of the system for the given datasets). Note that the size of each group is the number of OSM spatial objects in it. For example, for the Victoria Centre in OSM, though there are hundreds of ‘partOf’ matches involving it, it is only counted as one ele- ment in ‘Correctly Matched’. Precision was computed as the ratio of \|TP\| to \|TP\|+\|FP\|, and recall as the ratio of \|TP\| to \|TP\| + \|FN\| + \|FPsbm\|. As shown in Table 2, for both Nottingham and Southampton cases, precision is ≥90% and recall ≥84%. p f p p Using LBPT reasoning to remove these assumptions makes a difference to the running time and to the number of human interactions required to restore consistency in step 6. If we skip the LBPT reasoning step, and check the matching with respect to UNA and NPH assumptions directly, then it would require much more time to calculate explanations for inconsistencies (see Table 4). Table 4: Time for calculating explanations in UNA and NPH step by Pellet with LBPT step without LBPT step Nottingham 16s 812s Southampton 3s 46s A subject matter expert at Ordnance Survey (Great Britain’s National Mapping Authority) classiﬁed non-straightforward cases (approximately 10% of the total output of the system for the given datasets). Note that the size of each group is the number of OSM spatial objects in it. For example, for the Victoria Centre in OSM, though there are hundreds of ‘partOf’ matches involving it, it is only counted as one ele- ment in ‘Correctly Matched’. Precision was computed as the ratio of \|TP\| to \|TP\|+\|FP\|, and recall as the ratio of \|TP\| to \|TP\| + \|FN\| + \|FPsbm\|. As shown in Table 2, for both Nottingham and Southampton cases, precision is ≥90% and recall ≥84%. More importantly, more explanations for inconsistency would be generated by Pellet, since approximately 1800 wrong matches are removed in the LBPT step. The exact number of additional human interactions is not known since an attempt to resolve all inconsistencies was abandoned af- ter 4 hours, but it is signiﬁcantly greater than the number of interactions required with the LBPT step (see Table 5). Table 1: Data used for evaluation Table 1: Data used for evaluation OSM spatial objects OSGB spatial objects Nottingham 281 13204 Southampton 2130 7678 We chose these two datasets for evaluation because they have a reasonable representation in OSM (city centres usu- ally attract more attention from OSM contributors, and a va- riety of buildings and places are represented there) and are of reasonable size. In both cases, we set the value of σ used in geometry matching to be 20 m. The experiments were 3951 Table 2: Matching OSM spatial objects to OSGB Related Work Other tools for ontology matching or data interlinking ex- ist, for example LogMap (Jim´enez-Ruiz and Grau 2011) and KnoFuss (Nikolov, Uren, and Motta 2007). In (Du et al. 2013a), their performance on geospatial datasets is com- pared to that of the preliminary version of MatchMaps (Du et al. 2013b). The precision and recall of MatchMaps (on 3952 de Kleer, J. 1986. An assumption-based TMS. Artiﬁcial Intelligence 28(2):127–162. de Kleer, J. 1986. An assumption-based TMS. Artiﬁcial Intelligence 28(2):127–162. smaller datasets with manually computed ground truth) are much higher than that of LogMap and KnoFuss, mainly because they do not make explicit use of spatial infor- mation. Another feature of MatchMaps which explains its higher precision, is that when an inconsistent set of match- ing statements and axioms is found, MatchMaps does not automatically withdraw one of the statements but instead asks a human expert to make a decision. This approach was adopted because it was discovered empirically that none of the commonly-used heuristics for choosing the statement to withdraw works well in all cases. Du, H., and Alechina, N. 2014. A Logic of Part and Whole for Buffered Geometries. In the 7th European Starting AI Researcher Symposium (STAIRS), 91–100. Du, H.; Alechina, N.; Jackson, M.; and Hart, G. 2013a. Matching Geospatial Instances. In Proceedings of the 8th In- ternational Workshop on Ontology Matching, volume 1111 of CEUR Workshop Proceedings, 239–240. CEUR-WS.org. Du, H.; Alechina, N.; Jackson, M.; and Hart, G. 2013b. Matching Formal and Informal Geospatial Ontologies. In Ge- ographic Information Science at the Heart of Europe, Lecture Notes in Geoinformation and Cartography. Springer. 155– 171. In geospatial information science, several methods have been developed for matching geometries or integrating geospatial data. Kundu (2006) proposed a method for con- ﬂating two polygonal lines, based on a distance measure. Fu and Wu (2008) proposed a method for matching enti- ties in vector spatial data, based on interior intersection for polygons and buffer division for lines. Tong et al (2009) proposed a spatial feature matching method in map conﬂa- tion, based on the calculation of weighted average of posi- tional, shape, directional, and topological measures. How- ever, there is no consensus approach to matching geome- tries, and the computational cost of most methods is high. Crowd-sourced geospatial information introduces additional challenges, since the geometries are typically less accurate and may contain errors. Fu, Z., and Wu, J. 2008. Related Work Entity matching in vector spatial data. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 37(B4):1467 – 72. Haklay, M. 2010. How good is volunteered geographical in- formation? A comparative study of OpenStreetMap and Ord- nance Survey datasets. Environment and Planning B: Plan- ning and Design 37(4):682–703. ISO Technical Committe 211. 2003. ISO 19107:2003 Geo- graphic information – Spatial schema. Technical report, In- ternational Organization for Standardization (TC 211). Jackson, M. J.; Rahemtulla, H.; and Morley, J. 2010. The Synergistic Use of Authenticated and Crowd-Sourced Data for Emergency Response. In 2nd International Workshop on Validation of Geo-Information Products for Crisis Manage- ment (VALgEO), 91–99. For the application described in this paper, standard spa- tial formalisms (see, for example, (Aiello, Pratt-Hartmann, and Benthem 2007)) are not easily applicable since they pre- suppose accurate geometries or regions with sharp bound- aries. Formalisms which consider indeterminate regions, such as (Roy and Stell 2001), presuppose the existence of a core part of a region and a more vague part, while in our application, the real location of the object can be anywhere within the buffer with the same degree of (un)certainty. Jim´enez-Ruiz, E., and Grau, B. C. 2011. LogMap: Logic- Based and Scalable Ontology Matching. In Proceedings of the10th International Semantic Web Conference, volume 7031 of LNCS, 273–288. Springer. JPP. 2014. openjump. http://www.openjump.org. Kundu, S. 2006. Conﬂating two polygonal lines. Pattern Recognition 39(3):363–372. Conclusion Crowd-sourced geospatial datasets contain a wealth of infor- mation. However this information is not entirely reliable and needs to be validated. One way of validating it is to match the objects described in a crowd-sourced dataset to an au- thoritative geospatial data source. We describe an applica- tion, MatchMaps, that generates such matches and checks them for consistency with respect to ontological classiﬁca- tion of the objects and a qualitative spatial logic. We do this by extending existing artiﬁcial intelligence techniques for matching ontologies to the more specialised case of a geospatial ontology (containing location and geometry in- formation), and using a novel qualitative spatial logic to val- idate the matches. An evaluation of MatchMaps shows a high degree of precision and recall for two case studies. One direction of future research is to conduct a user evaluation study with OSGB to determine the amount of human effort required to perform a matching task. Nikolov, A.; Uren, V.; and Motta, E. 2007. KnoFuss: a Com- prehensive Architecture for Knowledge Fusion. In Proceed- ings of the 4th International Conference on Knowledge Cap- ture, 185–186. OpenStreetMap. 2014. The Free Wiki World Map. http://www.openstreetmap.org. Ordnance Survey. 2012. Ordnance Survey. http://www.ordnancesurvey.co.uk. Roy, A., and Stell, J. 2001. Spatial Relations between In- determinate Regions. International Journal of Approximate Reasoning 27(3):205 – 234. Sirin, E.; Parsia, B.; Grau, B. C.; Kalyanpur, A.; and Katz, Y. 2007. Pellet: a Practical OWL-DL Reasoner. Web Semantics: Science, Services and Agents on the World Wide Web 5:51–53. Tong, X.; Shi, W.; and Deng, S. 2009. A probability-based multi-measure feature matching method in map conﬂation. International Journal of Remote Sensing 30(20):5453–5472. References Aiello, M.; Pratt-Hartmann, I. E.; and Benthem, J. F. v. 2007. Handbook of Spatial Logics. Springer. References Aiello, M.; Pratt-Hartmann, I. E.; and Benthem, J. F. v. 2007. Handbook of Spatial Logics. Springer. Vivid Solutions, Inc. 2014. JTS Topology Suite. http://www.vividsolutions.com/jts. 3953
https://openalex.org/W3166472627	https://mdpi-res.com/bookfiles/book/2638/Emerging_Trends_in_TiOsub2sub_Photocatalysis_and_Applications.pdf?v=1685562028	English	null	Emerging Trends in TiO<sub>2</sub> Photocatalysis and Applications	MDPI eBooks	2,020	cc-by	317,995	Emerging Trends in TiO₂ Photocatalysis and Applications Printed Edition of the Special Issue Published in Catalysts Trong-On Do and Sakar Mohan Edited by Emerging Trends in TiO₂ Photocatalysis and Applications Printed Edition of the Special Issue Published in Catalysts www.mdpi.com/journal/catalysts Trong-On Do and Sakar Mohan Edited by Printed Edition of the Special Issue Published in Catalysts www.mdpi.com/journal/catalysts Trong-On Do and Sakar Mohan Edited by MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin Contents Contents About the Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Trong-On Do and Sakar Mohan Editorial: Special Issue on “Emerging Trends in TiO2 Photocatalysis and Applications” Reprinted from: Catalysts 2020, 10, 670, doi:10.3390/catal10060670 . . . . . . . . . . . . . . . . . . 1 Weiwei Fu, Zhiqiang Shi, Helong Bai, Jinyu Dai, Zhiming Lu, Feifei Lei, Deguang Zhang, Lun Zhao and Zongtao Zhang Facile Formation of Anatase Nanoparticles on H-Titanate Nanotubes at Low Temperature for Efﬁcient Visible Light-Driven Degradation of Organic Pollutants Reprinted from: Catalysts 2020, 10, 695, doi:10.3390/catal10060695 . . . . . . . . . . . . . . . . . . 5 Mohan Sakar, Ravikumar Mithun Prakash and Trong-On Do Insights into the TiO2-Based Photocatalytic Systems and Their Mechanisms Reprinted from: Catalysts 2019, 9, 680, doi:10.3390/catal9080680 . . . . . . . . . . . . . . . . . . . 17 Songbo Wang, Feifan Wang, Zhiming Su, Xiaoning Wang, Yicheng Han, Lei Zhang, Jun Xiang, Wei Du and Na Tang Controllable Fabrication of Heterogeneous p-TiO2 QDs@g-C3N4 p-n Junction for Efﬁcient Photocatalysis Reprinted from: Catalysts 2019, 9, 439, doi:10.3390/catal9050439 . . . . . . . . . . . . . . . . . . . 49 Bin Yang, Guoqiang Chen, Huiwen Tian and Lei Wen Improvement of the Photoelectrochemical Performance of TiO2 Nanorod Array by PEDOT and Oxygen Vacancy Co-Modiﬁcation Reprinted from: Catalysts 2019, 9, 407, doi:10.3390/catal9050407 . . . . . . . . . . . . . . . . . . . 65 Soroosh Mortazavian, Ali Saber and David E. James Optimization of Photocatalytic Degradation of Acid Blue 113 and Acid Red 88 Textile Dyes in a UV-C/TiO2 Suspension System: Application of Response Surface Methodology (RSM) Reprinted from: Catalysts 2019, 9, 360, doi:10.3390/catal9040360 . . . . . . . . . . . . . . . . . . . TiO2 Emerging Trends in Photocatalysis and Applications Editors Trong-On Do Sakar Mohan Trong-On Do Sakar Mohan MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin Editors Trong-On Do Laval University Canada Editors Trong-On Do Laval University Canada Sakar Mohan Laval University Canada Editorial Ofﬁce MDPI St. Alban-Anlage 66 4052 Basel, Switzerland Editorial Ofﬁce MDPI St. Alban-Anlage 66 4052 Basel, Switzerland This is a reprint of articles from the Special Issue published online in the open access journal Catalysts (ISSN 2073-4344) (available at: https://www.mdpi.com/journal/catalysts/special issues/ TiO2 photocatal). For citation purposes, cite each article independently as indicated on the article page online and as indicated below: LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. Journal Name Year, Article Number, Page Range. Page Range. ISBN 978-3-03936-706-1 (Pbk) ISBN 978-3-03936-707-8 (PDF) c⃝2020 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications. The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND. Contents 75 Nan-Quan Ou, Hui-Jun Li, Bo-Wen Lyu, Bo-Jie Gui, Xiong Sun, Dong-Jin Qian, Yanlin Jia, Xianying Wang and Junhe Yang Facet-Dependent Interfacial Charge Transfer in TiO2/Nitrogen-Doped Graphene Quantum Dots Heterojunctions for Visible-Light Driven Photocatalysis Reprinted from: Catalysts 2019, 9, 345, doi:10.3390/catal9040345 . . . . . . . . . . . . . . . . . . . 95 Fabi´an M. Mart´ınez, Elim Albiter, Salvador Alfaro, Ana L. Luna, Christophe Colbeau-Justin, Jos´e M. Barrera-Andrade, Hynd Remita and Miguel A. Valenzuela Hydrogen Production from Glycerol Photoreforming on TiO2/HKUST-1 Composites: Effect of Preparation Method Reprinted from: Catalysts 2019, 9, 338, doi:10.3390/catal9040338 . . . . . . . . . . . . . . . . . . . 111 Shinya Higashimoto Titanium-Dioxide-Based Visible-Light-Sensitive Photocatalysis: Mechanistic Insight and Applications Reprinted from: Catalysts 2019, 9, 201, doi:10.3390/catal9020201 . . . . . . . . . . . . . . . . . . . 123 Xiaolan Kang, Sihang Liu, Zideng Dai, Yunping He, Xuezhi Song and Zhenquan Tan Titanium Dioxide: From Engineering to Applications Reprinted from: Catalysts 2019, 9, 191, doi:10.3390/catal9020191 . . . . . . . . . . . . . . . . . . . 145 Oxygen Vacancy Co-Modiﬁcation v Qiang Li, Lifang Wang, Xuhui Fang, Li Zhang, Jingjiu Li and Hongyong Xie Synergistic Effect of Photocatalytic Degradation of Hexabromocyclododecane in Water by UV/TiO2/persulfate Reprinted from: Catalysts 2019, 9, 189, doi:10.3390/catal9020189 . . . . . . . . . . . . . . . . . . . 177 Olivia M. Schneider, Robert Liang, Leslie Bragg, Ivana Jaciw-Zurakowsky, Azar Fattahi, Shasvat Rathod, Peng Peng, Mark R. Servos and Y. Norman Zhou Photocatalytic Degradation of Microcystins by TiO2 Using UV-LED Controlled Periodic Illumination Reprinted from: Catalysts 2019, 9, 181, doi:10.3390/catal9020181 . . . . . . . . . . . . . . . . . . . 191 Morongwa Martha Songo, Richard Moutloali and Suprakas Sinha Ray Development of TiO2-Carbon Composite Acid Catalyst for Dehydration of Fructose to 5-Hydroxymethylfurfural Reprinted from: Catalysts 2019, 9, 126, doi:10.3390/catal9020126 . . . . . . . . . . . . . . . . . . . 201 Yuming He, Huayang Li, Xuelian Guo and Rongbo Zheng Bleached Wood Supports for Floatable, Recyclable, and Efﬁcient Three Dimensional Photocatalyst Reprinted from: Catalysts 2019, 9, 115, doi:10.3390/catal9020115 . . . . . . . Contents . . . . . . . . . . . . 217 Jian Shi, Feng Hui, Jun Yuan, Qinwei Yu, Suning Mei, Qian Zhang, Jialin Li, Weiqiang Wang, Jianming Yang and Jian Lu Ru-Ti Oxide Based Catalysts for HCl Oxidation: The Favorable Oxygen Species and Inﬂuence of Ce Additive Reprinted from: Catalysts 2019, 9, 108, doi:10.3390/catal9020108 . . . . . . . . . . . . . . . . . . . 227 Boxue Zhang, Shengxin Cao, Meiqi Du, Xiaozhou Ye, Yun Wang and Jianfeng Ye Titanium Dioxide (TiO2) Mesocrystals: Synthesis, Growth Mechanisms and Photocatalytic Properties Reprinted from: Catalysts 2019, 9, 91, doi:10.3390/catal9010091 . . . . . . . . . . . . . . . . . . . 243 Sayaka Yanagida, Kentaro Hirayama, Kenichiro Iwasaki and Atsuo Yasumori Adsorption and Photocatalytic Decomposition of Gaseous 2-Propanol Using TiO2-Coated Porous Glass Fiber Cloth Reprinted from: Catalysts 2019, 9, 82, doi:10.3390/catal9010082 . . . . . . . . . . . . . . . . . . . . 271 Yin-Hsuan Chang and Ming-Chung Wu Enhanced Photocatalytic Reduction of Cr(VI) by Combined Magnetic TiO2-Based NFs and Ammonium Oxalate Hole Scavengers Reprinted from: Catalysts 2019, 9, 72, doi:10.3390/catal9010072 . . . . . . . . . . . . . . . . . . . . 287 Bianca Rusinque, Salvador Escobedo and Hugo de Lasa Photocatalytic Hydrogen Production Under Near-UV Using Pd-Doped Mesoporous TiO2 and Ethanol as Organic Scavenger Reprinted from: Catalysts 2019, 9, 33, doi:10.3390/catal9010033 . . . . . . . . . . . . . . . . . . . . 299 Ricardo Rangel Ver´onica Janneth Cede ˜no Jaime Espino Pascual Bartolo-P´erez to 5-Hydroxymethylfurfural Reprinted from: Catalysts 2019, 9, 126, doi:10.3390/catal9020126 . . . . . . . . . . . . . . . . . . . 201 Yuming He, Huayang Li, Xuelian Guo and Rongbo Zheng Bleached Wood Supports for Floatable, Recyclable, and Efﬁcient Three Dimensional Photocatalyst Reprinted from: Catalysts 2019, 9, 115, doi:10.3390/catal9020115 . . . . . . . . . . . . . . . . . . . Contents 217 Jian Shi, Feng Hui, Jun Yuan, Qinwei Yu, Suning Mei, Qian Zhang, Jialin Li, Weiqiang Wang, Jianming Yang and Jian Lu Ru-Ti Oxide Based Catalysts for HCl Oxidation: The Favorable Oxygen Species and Inﬂuence of Ce Additive Reprinted from: Catalysts 2019, 9, 108, doi:10.3390/catal9020108 . . . . . . . . . . . . . . . . . . . 227 Boxue Zhang, Shengxin Cao, Meiqi Du, Xiaozhou Ye, Yun Wang and Jianfeng Ye Titanium Dioxide (TiO2) Mesocrystals: Synthesis, Growth Mechanisms and Photocatalytic Properties Reprinted from: Catalysts 2019, 9, 91, doi:10.3390/catal9010091 . . . . . . . . . . . . . . . . . . . 243 Sayaka Yanagida, Kentaro Hirayama, Kenichiro Iwasaki and Atsuo Yasumori Adsorption and Photocatalytic Decomposition of Gaseous 2-Propanol Using TiO2-Coated Porous Glass Fiber Cloth Reprinted from: Catalysts 2019, 9, 82, doi:10.3390/catal9010082 . . . . . . . . . . . . . . . . . . . . 271 Yin-Hsuan Chang and Ming-Chung Wu Enhanced Photocatalytic Reduction of Cr(VI) by Combined Magnetic TiO2-Based NFs and Ammonium Oxalate Hole Scavengers Reprinted from: Catalysts 2019, 9, 72, doi:10.3390/catal9010072 . . . . . . . . . . . . . . . . . . . . 287 Bianca Rusinque, Salvador Escobedo and Hugo de Lasa Photocatalytic Hydrogen Production Under Near-UV Using Pd-Doped Mesoporous TiO2 and Ethanol as Organic Scavenger Reprinted from: Catalysts 2019, 9, 33, doi:10.3390/catal9010033 . . . . . . . . . . . . . . . . . . . . 299 Ricardo Rangel, Ver´onica Janneth Cede ˜no, Jaime Espino, Pascual Bartolo-P´erez, Geonel Rodr´ıguez-Gattorno and Juan Jos´e Alvarado-Gil Comparing the Efﬁciency of N-Doped TiO2 and N-Doped Bi2MoO6 Photo Catalysts for MB and Lignin Photodegradation Reprinted from: Catalysts 2018, 8, 668, doi:10.3390/catal8120668 . . . . . . . . . . . . . . . . . . . 321 vi vi Hamza M. El-Hosainy, Said M. El-Sheikh, Adel A. Ismail, Amer Hakki, Ralf Dillert, Hamada M. Killa, Ibrahim A. Ibrahim and Detelf W. Bahnemann Highly Selective Photocatalytic Reduction of o-Dinitrobenzene to o-Phenylenediamine over Non-Metal-Doped TiO2 under Simulated Solar Light Irradiation Reprinted from: Catalysts 2018, 8, 641, doi:10.3390/catal8120641 . . . . . . . . . . . . . . . . . . . Contents 333 Osm´ın Avil´es-Garc´ıa, Jaime Espino-Valencia, Rub´ı Romero-Romero, Jos´e Luis Rico-Cerda, Manuel Arroyo-Albiter, Dora Alicia Sol´ıs-Casados and Reyna Natividad-Rangel Enhanced Photocatalytic Activity of Titania by Co-Doping with Mo and W Reprinted from: Catalysts 2018, 8, 631, doi:10.3390/catal8120631 . . . . . . . . . . . . . . . . . . . 345 Irwing M. Ram´ırez-S´anchez and Erick R. Bandala Photocatalytic Degradation of Estriol Using Iron-Doped TiO2 under High and Low UV Irradiation Reprinted from: Catalysts 2018, 8, 625, doi:10.3390/catal8120625 . . . . . . . . . . . . . . . . . . . 365 Yu Ren, Yuze Dong, Yaqing Feng and Jialiang Xu Compositing Two-Dimensional Materials with TiO2 for Photocatalysis Reprinted from: Catalysts 2018, 8, 590, doi:10.3390/catal8120590 . . . . . . . . . . . . . . . . . . . 389 Nick Serpone Heterogeneous Photocatalysis and Prospects of TiO2-Based Photocatalytic DeNOxing the Atmospheric Environment Reprinted from: Catalysts 2018, 8, 553, doi:10.3390/catal8110553 . . . . . . . . . . . . . . . . . . . 415 Kenta Kobayashi, Mai Takashima, Mai Takase and Bunsho Ohtani Mechanistic Study on Facet-Dependent Deposition of Metal Nanoparticles on Decahedral-Shaped Anatase Titania Photocatalyst Particles Reprinted from: Catalysts 2018, 8, 542, doi:10.3390/catal8110542 . . . . . . . . . . . . . . . . . . . 515 Haﬁze Nagehan Koysuren Solid-Phase Photocatalytic Degradation of Polyvinyl Borate Reprinted from: Catalysts 2018, 8, 499, doi:10.3390/catal8110499 . . . . . . . . . . . . . . . . . . . 531 Fei Li, Ming Li, Yi Luo, Ming Li, Xinyu Li, Jiye Zhang and Liang Wang The Synergistic Effect of Pyridinic Nitrogen and Graphitic Nitrogen of Nitrogen-Doped Graphene Quantum Dots for Enhanced TiO2 Nanocomposites’ Photocatalytic Performance Reprinted from: Catalysts 2018, 8, 438, doi:10.3390/catal8100438 . . . . . . . . . . . . . . . . . . . 539 Reo Eguchi, Yuya Takekuma, Tsuyoshi Ochiai and Morio Nagata Improving Interfacial Charge-Transfer Transitions in Nb-Doped TiO2 Electrodes with 7,7,8,8-Tetracyanoquinodimethane Reprinted from: Catalysts 2018, 8, 367, doi:10.3390/catal8090367 . . . . . . . . . . . . . . . . . . . Contents 551 Benjawan Moongraksathum, Jun-Ya Shang and Yu-Wen Chen Photocatalytic Antibacterial Effectiveness of Cu-Doped TiO2 Thin Film Prepared via the Peroxo Sol-Gel Method Reprinted from: Catalysts 2018, 8, 352, doi:10.3390/catal8090352 . . . . . . . . . . . . . . . . . . . 561 Peng Sun, Jun Zhang, Wenxiu Liu, Qi Wang and Wenbin Cao Modiﬁcation to L-H Kinetics Model and Its Application in the Investigation on Photodegradation of Gaseous Benzene by Nitrogen-Doped TiO2 Reprinted from: Catalysts 2018, 8, 326, doi:10.3390/catal8080326 . . . . . . . . . . . . . . . . . . . 571 Hamza M. El-Hosainy, Said M. El-Sheikh, Adel A. Ismail, Amer Hakki, Ralf Dillert, Haﬁze Nagehan Koysuren vii Editorial: Special Issue on “Emerging Trends in TiO2 Photocatalysis and Applications” Trong-On Do 1,* and Sakar Mohan 1,2 o g O o a d Sa a o a 1 Department of Chemical Engineering, Laval University, Quebec G1V 0A6, Canada; sakar.mohan.1@ulava 2 Centre for Nano and Material Sciences, Jain University, Bangalore 562112, India * Correspondence: Trong-On.Do@gch.ulaval.ca g 1 Department of Chemical Engineering, Laval University, Quebec G1V 0A6, Canada; sakar.mohan.1@ulaval.ca 2 Centre for Nano and Material Sciences, Jain University, Bangalore 562112, India * Correspondence: Trong-On.Do@gch.ulaval.ca Received: 4 June 2020; Accepted: 8 June 2020; Published: 13 June 2020 Received: 4 June 2020; Accepted: 8 June 2020; Published: 13 June 2020 It is not an exaggerated fact that the semiconductor titanium dioxide (TiO2) has been evolved as a prototypical material to understand the photocatalytic process and has been demonstrated for various photocatalytic applications such as pollutants degradation, water splitting, heavy metal reduction, CO2 conversion, N2 ﬁxation, bacterial disinfection, etc., as depicted in Figure 1. [1,2] The rigorous photocatalytic studies over TiO2 have paved ways to understand the various chemical processes involved and physical parameters (optical and electrical) required to design and construct diverse photocatalytic systems. [3,4] Accordingly, it has been realized that an eﬀective photocatalyst should have ideal band edge potential, narrow band gap energy, reduced charge recombination, enhanced charge separation, improved interfacial charge transfer, surface-rich catalytic sites, etc. These studies further highlighted that single component catalysts may not be good enough to achieve the required/enhanced photocatalytic process. As a result, many strategies have been developed to design a variety of photocatalytic systems, which include doping, composite formation, sensitization, co-catalyst loading, etc. [5] The doping strategy includes cationic and anionic doping, where it is found that the essential purposeof doping is to tune the band gap energy of the photocatalyst by introducing the new energy levels of the doped elements underneath the conduction band (CB) and above the valence band (VB) of the semiconductor photocatalyst, respectively. On the other hand, the composite formation serves in multiple ways to almost meet all the requirements to achieve a quantum eﬃcient photocatalytic process. The basis of composite formation is found to redesign the charge transport kinetics in the bulk and surface/interface of the integrated photocatalyst systems. These composite systems generally include p-n heterojunction, Z-scheme, etc. Similarly, the mechanism of sensitizing the photocatalysts includes the integration of plasmonic metal nanoparticles, carbon-based materials, 2D materials, quantum dots, and metal organic frameworks to enhance their optical absorption, electrical transportation properties, etc. Catalysts 2020, 10, 670; doi:10.3390/catal10060670 About the Editors Trong-On Do is a full professor in the Department of Chemical Engineering at Laval University, Canada. He received his MSc and PhD from University of P. and M. Curie (France) and carried out postdoctoral research in Prof. G. Bond’s group at Brunel University (UK), and then the French Catalysis Institute (France). He spent two years (1997–1999) in Profs. Hashimoto/Fujishima’s group at KAST under the Japanese STA Fellowship Award. His current research focuses on the design and synthesis of innovative and smart nanomaterials as photocatalysts and their applications in renewable energy and environmental remediation. Sakar Mohan received his PhD from the University of Madras, India in 2015. Later, he joined in the research group of Prof. Trong-On Do, Laval University, Canada for his postdoc. Currently, he is working as an Assistant Professor in the Centre for Nano and Material Sciences, Jain University, India since 2017. His research interest includes photocatalysis, membranes, biodiesel, sensors and biomaterials. ix ix ix catalysts catalysts 1. Sakar, M.; Mithun Prakash, R.; Do, T.O. Insights into the TiO2-based photocatalytic systems and their mechanisms. Catalysts 2019, 9, 680. [CrossRef] 2. Serpone, N. Heterogeneous photocatalysis and prospects of TiO2-based photocatalyticDeNOxing the atmospheric environment. Catalysts 2018, 8, 553. [CrossRef] Conﬂicts of Interest: The authors declare no conﬂict of interest. Figure 1. Overview of TiO2-based various photocatalytic systems and their applications. Figure 1. Overview of TiO2-based various photocatalytic systems and their applications. Figure 1. Overview of TiO2-based various photocatalytic systems and their applications. Towards highlighting the above mentioned diversities in TiO2 photocatalysis, there have been many interesting research works on TiO2, involving material designs for various photocatalytic applications published in this Special Issue. These material systems include TiO2 QDs@g-C3N4 p-n junction, [7] oxygen defective TiO2 nanorod array, [8] TiO2/N-doped graphene QDs, [9] TiO2/HKUST-1, [10] TiO2-Carbon composite, [11] Ru-Ti oxide, [12] TiO2 coated porous glass ﬁber cloth, [13] Ag/Fe3O4/TiO2 nanoﬁbers, [14] Pd-doped TiO2, [15] N-doped TiO2, [16] C/N/S-doped TiO2, [17] Mo/W co-doped TiO2, [18] Fe-doped TiO2, [19] N-doped graphene QDs-TiO2, [20] Nd-doped TiO2, [21] Cu-doped TiO2 thin ﬁlm, [22] surface engineered TiO2, [23] etc., for various photocatalytic applications, such as the degradations of a variety of pollutants, [24–30] biomass reforming, [10] heavy metal reduction, [14] and bacterial disinfections, [22] etc. In addition to these original research papers, some excellent review papers have also been published in this Special Issue, focusing on the various TiO2-based photocatalytic systems and their mechanisms and applications. [1–6] To this end, it is highlighted that future works in TiO2 should involve developing new material systems based on TiO2. For instance, instead of doping N into TiO2, the composition/phase tunable Ti oxy-nitride systems should be developed and so should the Ti oxy-phosphates, oxy-sulfurs, oxy-carbons, etc. From application perspectives, TiO2 should be investigated for its photocatalytic eﬃciencies towards the production of H2/O2 from atmospheric vapor, dark-photocatalytic activities, hydrogen storage, biodiesel productions, etc. However, the research should also be continued on bare TiO2 to achieve an in depth understanding of the photocatalytic mechanisms towards ﬁnding new photocatalytic applications. Editorial: Special Issue on “Emerging Trends in TiO2 Photocatalysis and Applications” [6] Interestingly, the co-catalyst loading serves as an ‘engineered-catalytic-site’ for the speciﬁc redox process to achieve the selective photocatalytic reactions. Furthermore, the unique systems, such as ferroelectric-based photocatalysts, are found to be more interesting as they are governed by their inherent internal electrical ﬁeld and surface polarization properties. For instance, the ferroelectric properties intrinsically facilitate the adsorption of the surrounding molecules, carrier separation, and interfacial charge transfer via band bending phenomenon, etc. Similarly, the inﬂuence of defects in photocatalysis has been well studied over TiO2, where the concepts of “self-doping”, “oxygen vacancy”, “colored TiO2”, etc. have been well addressed in TiO2photocatalysts. Catalysts 2020, 10, 670; doi:10.3390/catal10060670 1 www.mdpi.com/journal/catalysts www.mdpi.com/journal/catalysts Catalysts 2020, 10, 670 Figure 1. Overview of TiO2-based various photocatalytic systems and their applications. References 2 2 Catalysts 2020, 10, 670 3. Higashimoto, S. Titanium-dioxide-based visible-light-sensitive photocatalysis: Mechanistic insight and applications. Catalysts 2019, 9, 201. [CrossRef] pp y 4. Zhang, B.; Cao, S.; Du, M.; Ye, X.; Wang, Y.; Ye, J. Titanium dioxide (TiO2) mesocrystals: Synthesis, growth mechanisms and photocatalytic properties. Catalysts 2019, 9, 91. [CrossRef] 5. Kang, X.; Liu, S.; Dai, Z.; He, Y.; Song, X.; Tan, Z. Titanium dioxide: From engineering to applications. Catalysts 2019, 9, 191. [CrossRef] 6. Ren, Y.; Dong, Y.; Feng, Y.; Xu, J. Compositing two-dimensional materials with TiO2 for photocatalysis. Catalysts 2018, 8, 590. [CrossRef] 7. Wang, S.; Wang, F.; Su, Z.; Wang, X.; Han, Y.; Zhang, L.; Xiang, J.; Du, W.; Tang, N. Controllable fabrication of heterogeneous p-TiO2 QDs@g-C3N4 p-n junction for eﬃcient photocatalysis. Catalysts 2019, 9, 439. [CrossRef] 8. Yang, B.; Chen, G.; Tian, H.; Wen, L. Improvement of the photoelectrochemical performance of TiO array by PEDOT and oxygen vacancy Co-modiﬁcation. Catalysts 2019, 9, 407. [CrossRef] g, ; , ; , ; , p p p array by PEDOT and oxygen vacancy Co-modiﬁcation. Catalysts 2019, 9, 407. [CrossRef] 9. Ou, N.Q.; Li, H.J.; Lyu, B.W.; Gui, B.J.; Sun, X.; Qian, D.J.; Jia, Y.; Wang, X.; Yang, J. Facet-dependent interfacial charge transfer in TiO2/nitrogen-doped graphene quantum dots heterojunctions for visible-light driven photocatalysis. Catalysts 2019, 9, 345. [CrossRef] 10. Martínez, F.M.; Albiter, E.; Alfaro, S.; Luna, A.L.; Justin, C.C.; Barrera-Andrade, J.M.; Remita, H.; Valenzuela, M.A. Hydrogen production from glycerol photoreforming on TiO2/HKUST-1 composites: Eﬀect of preparation method. Catalysts 2019, 9, 338. [CrossRef] 11. Songo, M.M.; Moutloali, R.; Suprakas Sinha, R. Development of TiO2-carbon composite acid catalyst dehydration of fructose to 5-hydroxymethylfurfural. Catalysts 2019, 9, 126. [CrossRef] 12. Shi, J.; Hui, F.; Yuan, J.; Yu, Q.; Mei, S.; Zhang, Q.; Li, J.; Wang, W.; Yang, J.; Lu, J. Ru-Ti oxide based catalysts for HCl oxidation: The favorable oxygen species and inﬂuence of Ce additive. Catalysts 2019, 9, 108. [CrossRef] 13. Yanagida, S.; Hirayama, K.; Iwasaki, K.; Yasumori, A. Adsorption and photocatalytic decomposition of gaseous 2-propanol using TiO2-coated porous glass ﬁber cloth. Catalysts 2019, 9, 82. [CrossRef] 14. Chang, Y.H.; Wu, M.C. Enhanced photocatalytic reduction of Cr(VI) by combined magnetic tio2-based nfs and ammonium oxalate hole scavengers. Catalysts 2019, 9, 72. [CrossRef] Rusinque, B.; Escobedo, S.; de Lasa, H. Photocatalytic hydrogen production under near-uv using Pd-doped mesoporous TiO2 and ethanol as organic scavenger. Catalysts 2019, 9, 33. [CrossRef] 15. References Rusinque, B.; Escobedo, S.; de Lasa, H. Photocatalytic hydrogen production under near mesoporous TiO2 and ethanol as organic scavenger. Catalysts 2019, 9, 33. [CrossRef] 16. Rangel, R.; Cedeño, V.J.; Espino, J.; Bartolo-Pérez, P.; Rodríguez-Gattorno, G.; Alvarado-Gil, J.J. Comparing the eﬃciency of N-doped TiO2 and N-doped Bi2MoO6 photo catalysts for MB and lignin photodegradation. Catalysts 2018, 8, 668. [CrossRef] 17. El-Hosainy, H.M.; El-Sheikh, S.M.; Ismail, A.A.; Hakki, A.; Dillert, R.; Killa, H.M.; Ibrahim, I.A.; Bahnemann, D.W. Highly selective photocatalytic reduction of o-dinitrobenzene to o-phenylenediamine over non-metal-doped TiO2 under simulated solar light irradiation. Catalysts 2018, 8, 641. [CrossRef] 18. Avilés-García, O.; Espino-Valencia, J.; Romero-Romero, R.; Rico-Cerda, J.L.; Arroyo-Albiter, M.; Solís-Casados, D.A.; Natividad-Rangel, R. Enhanced photocatalytic activity of titania by co-doping with Mo and W. Catalysts 2018, 8, 631. [CrossRef] 19. Ramírez-Sánchez, I.M.; Bandala, E.R. Photocatalytic degradation of estriol using iron-doped TiO2 un high and low UV irradiation. Catalysts 2018, 8, 625. [CrossRef] 20. Li, F.; Li, M.; Luo, Y.; Li, M.; Li, X.; Zhang, J.; Wang, L. The synergistic eﬀect of pyridinic nitrogen and graphitic nitrogen of nitrogen-doped graphene quantum dots for enhanced TiO2nanocomposites’ photocatalytic performance. Catalysts 2018, 8, 438. [CrossRef] 21. Eguchi, R.; Takekuma, Y.; Ochiai, T.; Nagata, M. Improving interfacial charge-transfer transitions in Nb-doped TiO2 electrodes with 7,7,8,8-tetracyanoquinodimethane. Catalysts 2018, 8, 367. [CrossRef] 22. Moongraksathum, B.; Shang, J.Y.; Chen, Y.W. Photocatalytic antibacterial eﬀectiven thin ﬁlm prepared via the peroxo sol-gel method. Catalysts 2018, 8, 352. [CrossRef] 23. Kobayashi, K.; Takashima, M.; Takase, M.; Ohtani, B. Mechanistic study on facet-dependent deposition of metal nanoparticles on decahedral-shaped anatasetitaniaphotocatalyst particles. Catalysts 2018, 8, 542. [CrossRef] 24. Mortazavian, S.; Saber, A.; James, D.E. Optimization of photocatalytic degradation of Acid Blue 113 and Acid Red 88 textile dyes in a UV-C/TiO2 suspension system: Application of response surface methodology (RSM). Catalysts 2019, 9, 360. [CrossRef] 3 3 Catalysts 2020, 10, 670 25. Li, Q.; Wang, L.; Fang, X.; Zhang, L.; Li, J.; Xie, H. Synergistic eﬀect of photocatalytic degradation of hexabromocyclododecane in water by UV/TiO2/persulfate. Catalysts 2019, 9, 189. [CrossRef] 26. Schneider, O.M.; Liang, R.; Bragg, L.; Jaciw-Zurakowsky, I.; Fattahi, A.; Rathod, S.; Peng, P.; Servos, M.R.; Norman Zhou, Y. Photocatalytic degradation of microcystins by TiO2 using UV-led controlled periodic illumination. Catalysts 2019, 9, 181. [CrossRef] 27. He, Y.; Li, H.; Guo, X.; Zheng, R. Bleached wood supports for ﬂoatable, recyclable, and eﬃcient th dimensional photocatalyst. Catalysts 2019, 9, 115. [CrossRef] 28. Koysuren, H.N. Received: 13 May 2020; Accepted: 17 June 2020; Published: 19 June 2020 Received: 13 May 2020; Accepted: 17 June 2020; Published: 19 June 2020 Received: 13 May 2020; Accepted: 17 June 2020; Published: 19 June 2020 Abstract: Anatase nanoparticles (5–10 nm) generated on H-titanate nanotube surface (H-titanate/anatase) were prepared by an ingenious and simple method. H-titanate tubes were prepared by a hydrothermal reaction of Ti powder in concentrated NaOH solution and an ion exchange process with HNO3 solution. After that, at a relatively low drying temperature (100 ◦C), a small quantity of anatase nanoparticles were in-situ formed on the H-titanate tubes surface by a surface dehydration reaction. In-situ transformation can form a strong interface coupling between H-titanate and anatase, which is conducive to accelerating charge transfer and improving its photocatalytic activity. In addition, the smaller average crystal size, the large speciﬁc surface areas (BET), the nanotubed and layered structure and the synergistic eﬀect of dual phases would be beneﬁcial to improving the photocatalytic eﬃciency. Keywords: in-situ formation; anatase nanoparticles; H-titanate nanotubes; dual-phase; low temperature Catalysts 2020, 10, 695; doi:10.3390/catal10060695 catalysts catalysts Facile Formation of Anatase Nanoparticles on H-Titanate Nanotubes at Low Temperature for Eﬃcient Visible Light-Driven Degradation of Organic Pollutants Weiwei Fu 1,, Zhiqiang Shi 2, Helong Bai 1, Jinyu Dai 2, Zhiming Lu 2, Feifei Lei 2, Deguang Zhang 2, Lun Zhao 1, and Zongtao Zhang 2,* 1 College of Chemistry, Changchun Normal University, Changchun 130032, China; baihelong@ccsfu.edu 2 State Key Laboratory of Inorganic Synthesis & Preparative Chemistry Jilin University Changchun 130 1 College of Chemistry, Changchun Normal University, Changchun 130032, China; baihelong@ 1 College of Chemistry, Changchun Normal University, Changchun 130032, China; baihelong@ccsfu.edu.cn 2 State Key Laboratory of Inorganic Synthesis & Preparative Chemistry, Jilin University, Changchun 130012, China; shizq17@jlu.edu.cn (Z.S.); daijy16@mails.jlu.edu.cn (J.D.); zmlu18@mails.jlu.edu.cn (Z.L.); leiﬀ18@mails.jlu.edu.cn (F.L.); zhangdg17@mails.jlu.edu.cn (D.Z.) College of Chemistry, Changchun Normal University, Changchun 130032, China; baihelong@ccsfu.edu.cn 2 State Key Laboratory of Inorganic Synthesis & Preparative Chemistry, Jilin University, Changchun 130012 China; shizq17@jlu.edu.cn (Z.S.); daijy16@mails.jlu.edu.cn (J.D.); zmlu18@mails.jlu.edu.cn (Z.L.); leiﬀ18@mails.jlu.edu.cn (F.L.); zhangdg17@mails.jlu.edu.cn (D.Z.) * Correspondence: fuweiwei@ccsfu.edu.cn (W.F.); zhaolun@mail.cncnc.edu.cn (L.Z.); zzhang@jlu.edu.cn (Z.Z.); Tel.: +86-431-8616-8099 (W.F.) www.mdpi.com/journal/catalysts References Solid-phase photocatalytic degradation of polyvinyl borate. Catalysts 2018, 8, 499. [CrossR 28. Koysuren, H.N. Solid-phase photocatalytic degradation of polyvin 28. Koysuren, H.N. Solid-phase photocatalytic degradation of polyvinyl borate. Catalysts 2018, 8, 499. [CrossRef] 29. Sun, P.; Zhang, J.; Liu, W.; Wang, Q.; Cao, W. Modiﬁcation to L-H kinetics model and its application in the investigation on photodegradation of gaseous benzene by nitrogen-doped TiO2. Catalysts 2018, 8, 326. [CrossRef] 30. Fu, W.; Shi, Z.; Bai, H.; Dai, J.; Lu, Z.; Lei, F.; Zhang, D.; Zhao, L.; Zhang, Z. Facile formation of anatase nanoparticles on H-titanate nanotubes at low temperature for eﬃcient visible light-driven degradation of organic pollutants. Catalysts 2020, 10, 695. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 1. Introduction Water contamination resulting from the rapid development of industrialization has attracted worldwide attention. Photocatalytic degradation is the most promising strategy to completely solve the organic pollutants problem [1]. Among all kinds of conversion systems, TiO2-based materials for water pollution are considered to be an environmentally friendly and promising way to eﬃciently utilize solar energy [2,3]. Under UV irradiation, anatase can degrade a broad range of tenacious and toxic organic contaminants in water, and it is nontoxic, relatively cheap and chemically stable. However, its photocatalytic application is limited owing to the rapid recombination of the excited electron–hole pairs, the low visible light activity and low surface area [4,5]. Coupling anatase with another semiconductor favors a narrow band gap and the electron–hole separation [6–9], so as to improve the quantum eﬃciency. Moreover, the synergistic eﬀect between two diﬀerent phases can also enhance the photocatalytic activity [6,10]. It was reported that H-titanate was formed from TiO2 reacting with a concentrated NaOH solution to form titanate and then the ion exchange reacted with a dilute acid solution [11–13]. After calcination at a high temperature, H-titanate can transform into anatase TiO2 [13]. Based on the above formation mechanism of TiO2-H-titanate-TiO2, it is estimated that, at low calcination temperature, a small Catalysts 2020, 10, 695; doi:10.3390/catal10060695 5 Catalysts 2020, 10, 695 amount of anatase TiO2 will be formed in situ on the H-titanate nanotube surface. Several papers have reported the synthesis of titanate/anatase composites; however, most preparation processes were carried out under high pressure and high temperature conditions, which consumed more energy [14–16]. For example, Xiong et al. reported a nitrogen-doped titanate–anatase core–shell nanobelts. In this paper, the titanate–anatase was obtained by calcination at 500 ◦C [14]. Yan et al. demonstrated that the titanate nanotube/anatase nanoparticle composites could be prepared by the hydrothermal method using as-obtained titanate tube dispersed into a HNO3 solution [15]. Herein, we report that a dual-phase photocatalyst (anatase nanoparticles (5–10 nm) was generated in situ on a H-titanate nanotube surface) was obtained via a controllable surface dehydration reaction at low temperature (100 ◦C) and atmospheric pressure, which exhibited a higher visible light photocatalytic activity than P25; pure H-titanate nanotubes and pure anatase. 1. Introduction The eﬃcient visible light photocatalytic activity can be attributed to: (1) In situ transformation can form strong interfacial coupling between H-titanate and anatase, which is favorable to accelerating charge transfer [17]; (2) H-titanate has a layered structure, which is composed of TiO6 octahedra sheets sharing four edges, similar to that of anatase crystals, which is easy to form a heterostructure between anatase and H-titanate [14,18,19]. In the meantime, at a low drying temperature, dual-phase catalyst retained the nanotubed and layered structures, which was beneﬁcial for the high BET surfaces to adsorb organic pollutants and promote the diﬀusion of organic molecules inside the pores. The smaller average crystal size of anatase nanoparticles means a stronger redox ability in the photocatalytic process. Therefore, in the presence of the dual-phase catalyst, rhodamine B (RhB) and methylene blue (MB) can be completely decomposed in a very short time under visible light irradiation. In addition, the synergistic eﬀect would be beneﬁcial to improve the photocatalytic eﬃciency. 2. Results and Discussion Figure 1 shows the X-ray diﬀraction (XRD) patterns of the as-synthesized material. As shown in Figure 1a, there are several diﬀraction peaks located at 2θ = 9.8◦, 24.4◦, 28.4◦, 48.4◦suggesting that the as-prepared sample is layered titanate with a component of H2Ti2O4(OH)2 [20] and a 0.9 nm interlayer distance, which is further conﬁrmed by TEM observation. After drying at 100 ◦C, the crystal structure of H-titanate was well maintained; the diﬀraction peaks of anatase-type TiO2 (JCPDS no. 21-1272) [21] are clearly observed (Figure 1b), showing that dual-phase H-titanate/anatase was obtained at a low temperature, which may be due to the dehydration of the H-titanate. The broad peaks indicate its low crystallinity and nanosized crystallites. With increasing calcination temperature, the peaks of anatase become narrower and sharper (Figure 1c,d). After calcination at 500 ◦C for 5 h, all diﬀraction peaks of H-titanate disappeared (Figure 1e), which suggested that the layered H-titanate was completely transformed into anatase. The apparent sharpening of peaks suggests its high crystallinity. To further conﬁrm the coexistence of these two TiO2 phases in the as-synthesized products, Raman spectroscopy was tested for the H-titanate and dual-phase catalysts (Figure 2). In Figure 2b, the peaks that center at 143 (Eg), 514 (A1g) and 636 cm−1 (Eg) belong to the anatase phase [22], while others match well with the hydrogen titanate phase (shown in Figure 2a). The Raman spectrum in Figure 2b is composed of the characteristic peaks of the H-titanate and anatase phase, which are in good agreement with the XRD analysis. Catalysts 2020, 10, 695 y , , Figure 1. XRD patterns of (a) TiO2-60 (the as-prepared H-titanate tubes), (b) TiO2-100, (c) TiO2-200, (d) TiO2-300, (e) TiO2-500, T is the H-titanate phase and A is the anatase phase. Figure 1. XRD patterns of (a) TiO2-60 (the as-prepared H-titanate tubes), (b) TiO2-100, (c) TiO2-200, (d) TiO2-300, (e) TiO2-500, T is the H-titanate phase and A is the anatase phase. 7 Catalysts 2020, 10, 695 Figure 2. Raman spectra of (a) TiO2-60 and (b) TiO2-100. The detailed characterization and crystal structure of the H-titanate/anatase composite were investigated via transmission electron microscopy (TEM), as shown in Figure 3. H-titanate was formed via the hydrothermal reaction of titanium powder with a concentrated NaOH solution to form sodium titanate and then a subsequent ion exchange reaction with HNO3 solution at room temperature. 2. Results and Discussion In other words, the speciﬁc surface area of the catalysts is enormously increased by the H-titanate nanotubes, which is hoping to enhance the photocatalytic property of the nanomaterials. Figure 3. (a,b) are the TEM images of TiO2-100. (b1) and (b2) are the HRTEM images of TiO2-100 for the enlarged view of the circle and rectangle areas in (b). (c) Reversible nitrogen gas adsorption isotherm for the (c1) TiO2-60, (c2) TiO2-100, (c3) TiO2-500 measured at 77 K. The inset is the pore size distribution of the TiO2-100 photocatalyst calculated via DFT method. Figure 3. (a,b) are the TEM images of TiO2-100. (b1) and (b2) are the HRTEM images of TiO2-100 for the enlarged view of the circle and rectangle areas in (b). (c) Reversible nitrogen gas adsorption isotherm for the (c1) TiO2-60, (c2) TiO2-100, (c3) TiO2-500 measured at 77 K. The inset is the pore size distribution of the TiO2-100 photocatalyst calculated via DFT method. UV-Vis diﬀuse reﬂectance absorption spectra of as-synthesized H-titanate nanotubes (TiO2-60), dual-phase H-titanate/anatase (TiO2-100) and anatase nanoparticles (TiO2-500) are shown in the Figure 4a. As indicated in Figure 4a, in the range of 200-800 nm, the H-titanate nanotubes, dual-phase H-titanate/anatase and anatase samples demonstrate similar absorption. Compared with that of pure H-titanate and anatase, the absorption edge of dual-phase H-titanate/anatase moved toward a longer wavelength. It may be owing to the synergistic eﬀect of H-titanate and anatase, resulting in a narrow band gap, which is a crucial role to the realization of solar energy conversion. The Kubelka-Munk method is often used to estimate the band gap energy (Eg) of as-prepared products [23]. Figure 4b reveals the Tauc plots of (αhν)2 vs. photon energy (hν) of H-titanate tubes, dual-phase H-titanate/anatase and anatase photocatalyst. The band gap (Eg) can be acquired by extending the vertical segment to the hν axis. As shown in Figure 4b, the Eg of the H-titanate, H-titanate/anatase, and anatase is 3.48, 3.30, and 3.36 eV, respectively, which reveals the same results with the ultraviolet-visible diﬀﬀuse reﬂectance spectra (UV-Vis DRS) analysis. The photoluminescence (PL) technique is usually used to investigate the charge carrier separation and transfer processes because PL emission results from the free photogenerated carriers recombination [24]. Figure 4c shows the PL spectra of dual-phase H-titanate/anatase, H-titanate and anatase excited at 315 nm. 2. Results and Discussion The dual-phase H-titanate/anatase catalysts were prepared by in situ generation of anatase nanoparticles on the H-titanate tube surfaces by a controllable surface dehydration reaction at a low drying temperature (100 ◦C). Figure S1a shows images of only the nanotubes, whereas Figure 3b shows that some nanoparticles adhered to the surface of nanotubes, which indicates that a new phase was obtained after drying at 100 ◦C. The average diameter of the as-synthesized nanoparticles is about 5–10 nm. Figure 3b1 (high-resolution transmission electron microscopy (HRTEM) images) distinctly reveals a lattice fringe spacing of 0.35 nm, which corresponds well with the (101) anatase. Figure 3b2 shows that the nanotubes are crystallized of layered H-titanate. It also indicates that the H-titanate retains its morphology of nanotubes, and the anatase reveals the morphology of nanoparticles. There are only particles in Figure S1b, which indicates that when the calcination temperature increased to 500 ◦C, H-titanate nanotubes completely transformed into anatase nanoparticles. The above results coincide with the XRD analysis. At a low drying temperature (100 ◦C), the photocatalyst retained its original nanotube structure and a small amount of anatase nanoparticles were formed on the H-titanate nanotube surfaces. Speciﬁc surface area (BET) is a key factor in photodegradation. High speciﬁc surface area oﬀers more reaction sites for dye molecules and hydroxyl groups. Therefore, the porosities of the samples were determined by N2 sorption. Figure 3c exhibits the N2 sorption isotherms of all as-synthesized products and the DFT (Density Functional Theory) pore size distributions of the dual-phase photocatalyst. The obtained isotherms of the samples are the typical IUPAC type-IV isotherm. We can see that the isotherms decreased with the decrease in the amount of H-titanate nanotubes. The corresponding DFT pore size distribution curve (inset of Figure 3c) of the dual-phase catalyst shows two pore sizes centered at ~1.0 nm and ~12 nm, respectively, which directly proves the layered and nanotubed structure of the dual phase catalyst. BET measurements show that the as-synthesized H-titanate tubes have a surface area of 245 m2·g−1, and dual-phase photocatalyst has a surface area of 174 m2·g−1. As the calcination temperature raised to 500 ◦C, the surface area of the 8 Catalysts 2020, 10, 695 product decreased to 95 m2·g−1 (Table S1). The decrease in surface area may be due to the decreased proportion of layered H-titanate nanotubes. 2. Results and Discussion The emission intensity of dual-phase H-titanate/anatase is much weaker than that of pure H-titanate and anatase, which can be attributed to the formation of H-titanate/anatase heterojunction in two semiconductor interfaces. Diﬀerent band edge 9 Catalysts 2020, 10, 695 positions of H-titanate and anatase can reduce the recombination of the carriers [19,25]. That enables more free electrons and holes to participate in the photocatalytic reactions and accelerates the photocatalytic process. Figure 4. (a) UV/Vis diﬀuse reﬂectance absorption spectra, (b) plots of (αhν)2 versus hν, and (c) PL spectra of TiO2-60, TiO2-100 and TiO2-500. Figure 4. (a) UV/Vis diﬀuse reﬂectance absorption spectra, (b) plots of (αhν)2 versus hν, and (c) PL spectra of TiO2-60, TiO2-100 and TiO2-500. The photocatalytic activities of as-prepared samples were studied by degradation of methyl orange (MO), rhodamine B (RhB) and methylene blue (MB) in aqueous solution under visible light at room temperature. In order to evaluate the photocatalytic eﬃciency, P25 was chosen as the photocatalytic reference material. In general, high speciﬁc surface area can improve the adsorption performance of the materials. Therefore, prior to the photocatalytic degradation studies, the adsorption properties of the materials were investigated. In general, a high degree of surface adsorption was observed within the ﬁrst 30 min of stirring time in dark before attaining the saturation level. In Figure 5a and Figure S2A, P25 shows no obvious adsorption capacity, whereas the as-received dual-phase H-titanate/anatase catalyst shows 90% MB, 25% RhB, 8.9% MO dye adsorption, the H-titanate shows 92% MB, 59% RhB, 9.1% MO adsorption due to their nanotubed and layered structures, larger speciﬁc surface areas, which are beneﬁcial for enhancing the photocatalytic activity. In addition, the electrostatic attraction between the catalysts and the dye molecules plays an important role in the diﬀerent adsorption behavior of several dyes [26]. MB, RhB is cationic and MO is anionic in the aqueous solution, while the nanotubes possess negative surface charge, which is more favorable to absorb MB and RhB. Although both RhB and MB are cationic dyes, the adsorption capacity of nanotubes for MB is much higher than that of RhB, which is due to the diﬀerent molecular structures of dyes. The more linear shape and smaller size of MB molecule mean a weaker steric hindrance during the adsorption process [26]. In order to verify the adsorption of dye on H-titanate/anatase, FTIR analysis was carried out after a MB adsorption and degradation test (Figure S3). 2. Results and Discussion According to the previous report [27], in adsorption test, the peaks at 2926 cm−1 and 666 cm−1 indicated that the MB was adsorbed onto the H-titanate/anatase nanotubes surfaces. After the degradation process, the peaks disappeared, suggesting that MB degraded completely. 10 Catalysts 2020, 10, 695 Figure 5. (a) Photocatalytic degradation of methylene blue (MB) (a1), rhodamine B (RhB) (a2), methyl orange (MO) (a3) over the dual-phase H-titanate/anatase catalysts and photocatalytic degradation of MB (b1), RhB (b2), MO (b3) over P25 under visible light irradiation, (b) cycling experiments of H-titanate/anatase catalysts for RhB degradation under visible light irradiation. Figure 5. (a) Photocatalytic degradation of methylene blue (MB) (a1), rhodamine B (RhB) (a2), methyl orange (MO) (a3) over the dual-phase H-titanate/anatase catalysts and photocatalytic degradation of MB (b1), RhB (b2), MO (b3) over P25 under visible light irradiation, (b) cycling experiments of H-titanate/anatase catalysts for RhB degradation under visible light irradiation. When visible light was turned on, as shown in Figure 5a, RhB was almost decomposed within only 30 min (Figure 5a2 and Figure S4), MO was removed by about 60% (Figure 5a3) within 30 min illumination, and the MB completely decomposed in just 5 min (Figure 5a1) in the test of the dual-phase H-titanate/anatase photocatalyst, whereas P25 showed no signiﬁcant degradation for the dyes in 30 min. Additionally, without any photocatalyst, RhB, MB and MO were hardly degraded by visible light [28–30], which demonstrated the high reactivity of as-synthesized dual-phase nanotubes catalysts. Compared with many reported TiO2-based materials, it also shows higher photocatalytic activity for the degradation of organic dyes under visible light. Xiong et al. reported that under the visible light irradiation, MB degraded completely in about 175 min in the presence of the nitrogen-doped titanate-anatase core-shell nanobelts catalyst [14]. S-doped Na2Ti6O13@TiO2 core-shell nanorods can completely decompose MB in 100 min under visible light [31]. Li et al. demonstrated that within 60 min of UV light irradiation, RhB was completely decomposed in the presence of the double-shell anatase-rutile TiO2 spheres [32]. The report of Pan et al. showed that within 40 min visible light irradiation, MO can be degraded by about 50% in the presence of GQD-TiO2 heterojunctions [33]. Figure S2A shows the photocatalytic activity of all as-synthesized catalysts for MB, RhB and MO under visible light irradiation. Compared with pure H-titanate and anatase catalysts, the dual-phase catalyst exhibited the best photoactivity. 2. Results and Discussion Figure S2B displays the degradation rate of the materials (dual-phase H-titanate/anatase and P25) for MO, RhB under visible light irradiation. It is noted that the dual-phase catalyst displays better degradation eﬃciency than P25. H-titanate/anatase photocatalyst shows 15 times higher eﬃciency for the photodegradation of RhB and 17 times higher eﬃciency for the photodegradation of MO compared to P25 under visible light radiation, which demonstrated the high photocatalytic activity of the dual-phase H-titanate/anatase. In order to evaluate the stability of the dual-phase photocatalyst, several photodegradation tests of RhB under visible light were carried out (Figure 5b). After four cycles of photocatalytic degradation, within about 30 min, the RhB could be decomposed completely, which indicates that the dual-phase H-titanate/anatase has excellent photocatalytic stability. Based on above results, the dual-phase (H-titanate/anatase) catalyst reveals a high visible light photo-degradation ability. It can be expounded by the following involved reasons: (1) In-Situ generation can form a strong interfacial coupling between H-titanate and anatase, which is useful for accelerating charge transfer and improving the photocatalytic activity [17]; (2) The dual-phase catalyst retains the nanotubed and layered structures, and possesses a high BET surface area. Large speciﬁc surface area and pore structure can provide more active sites to adsorb organic pollutants and promote the diﬀusion of organic molecules inside the pores. The layered titanate product has been considered as an excellent adsorbent [34,35]. It would be beneﬁcial to improve the photocatalytic activity. In addition, the meso-nanotubes structure is also conducive to the rapid diﬀusion of quantum, which is formed in the 11 Catalysts 2020, 10, 695 photocatalytic process, further promoting photogenerated charge transport to improve the separation rate [36,37]; (3) The synergetic eﬀect between H-titanate and anatase is one of the major ingredients for its enhanced visible light photocatalytic activity. When two phases combined, a staggered band gap was formed, which lead to the eﬃcient charge separation of the cross phase junction [38]. As shown in Figure 6, under visible light irradiation, anatase in dual-phase can be excited to generate electron-hole pairs. Electrons are excited from the valence band (VB) to the conduction band (CB). According to the energy band data, the CB and the VB energy potentials in anatase are −0.26 and 2.94 eV [39], while those of titanate are −0.50 and 3.03 eV [31]. 3.1. Materials The titanium powder was purchased from Aladdin, Tianjin, China. The NaOH and the HNO3 were purchased from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China. 3. Materials and Methods 3.1. Materials 2. Results and Discussion The photogenerated electrons accumulated on anatase will migrate from the CB of anatase to that of titanate due to the potential diﬀerence. In this way, titanate can eﬀectively collect photogenerated electrons and anatase collect holes. These electrons react with the surface adsorbed O2 to form O2−·, because the CB edge potential of titanate (−0.50 eV) is more negative than the standard redox potential of O2/O2−· (−0.33 eV) [40]. In the mean time, the VB potential of anatase (2.94 eV) is more positive than the standard redox potential of ·OH/OH−(1.99 eV) [41,42], the holes react with OH−to generate ·OH radicals. Then the organic pollutants could be mineralized by the produced ·OH and O2−· radicals. Hence, the above eﬃcient separation of photogenerated electron-hole pairs process improves the photodegradation rate of dyes; (4) The new anatase phase with a small average crystal size means a stronger redox ability because of the quantum size eﬀect [43]. Combining all above factors, the dual-phase H-titanate/anatase photocatalysts displayed high visible light activity. Figure 6. The proposed photocatalytic mechanism of H-titanate/anatase composite. Figure 6. The proposed photocatalytic mechanism of H-titanate/anatase composite. 3. Materials and Methods 3. Materials and Methods 3.4. Photocatalytic Activity Test The photocatalytic activities of the photocatalysts were performed at room temperature in a glass reactor ﬁtted with a Xe lamp (300 W). A 420 nm cut-on ﬁlter was used to ensure that only visible light illuminated the photocatalyst. The reaction liquid was prepared by mixing 0.25 g photocatalysts and 100 mL rhodamine B (or 10 mg/L MB; 10 mg/L MO) aqueous solution (10 mg/L). Then, the suspension was stirred in dark for 30 min to reach adsorption-desorption equilibrium before irradiation. Then it was irradiated under visible light. The suspension (8 mL) was withdrawn from the irradiated solution at preset time intervals and centrifuged to separate the photocatalyst particles, and then the supernatants were analyzed by UV-Vis spectrophotometer (UV-2450, Shanghai, China). 3.3. Catalyst Characterization X-ray powder diﬀraction (XRD) analysis was carried out using a D/Max-2550 X-ray powder diﬀractometer (Tokyo, Japan) with Cu Kα radiation. The ultraviolet-visible diﬀuse reﬂectance spectra of the samples were measured on a UV-Vis-NIR spectrophotometer (Shimadzu U-4100, Shanghai, China) detecting absorption over the range of 200–800 nm. The morphologies of samples were measured on a Tecnai G2 S-Twin F20 transmission electron microscopy (TEM, FEI, Hillsboro, FL, USA). N2 adsorption–desorption isotherms were obtained at 77 K on a Micromeritics ASAP 2020 sorptometer (Norcross, GA, USA). Raman spectra were recorded using a Renishaw InVia Raman spectrometer (London, UK) with a wavelength of 532 nm. Room temperature photoluminescence (PL) spectra with an excitation wavelength of 315 nm were measured on a FLUOROMAX-4 (Beijing, China). 3.2. Catalyst Preparation The photocatalysts were prepared by the following synthetic route. Firstly, 1.5 g titanium powder was mixed with 35 mL NaOH solution (10 M) and stirred at room temperature for 5 h. The mixture was transferred into a Teﬂon-lined autoclave and kept at 150 ◦C for 72 h. The obtained precipitates were washed with deionized water until neutral and dried in an oven (60 ◦C) overnight. Afterwards, the ion-exchange reaction was followed with 0.5 M HNO3 solution for 3 h at room temperature. There were three times ion-exchange reactions for the product. The ﬁnal product was dried at 60 ◦C 12 Catalysts 2020, 10, 695 for 24 h to produce the hydrogen titanate (H-titanate) tubes and then dried at 100 ◦C for 24 h to yield hydrogen titanate/anatase nanotubes. The as-prepared H-titanate was calcined in air at 200, 300 and 500 ◦C for 5 h to get TiO2 products. Henceforth, these samples are referred to as TiO2-60, TiO2-100, TiO2-200, TiO2-300 and TiO2-500, respectively. Author Contributions: Z.Z. and W.F. conceived and designed the experiments; Funding acquisition, L.Z.; Z.S., H.B., J.D., Z.L., F.L. and D.Z. performed the experiments and analyzed the data; W.F. wrote the paper. All authors have read and agreed to the published version of the manuscript. References 1. Cao, S.W.; Low, J.X.; Yu, J.G.; Jaroniec, M. Polymeric Photocatalysts Based on Graphitic Carbon Nitride. Adv. Mater. 2015, 27, 2150–2176. [CrossRef] 2. Zhang, X.; Chen, Y.; Xiao, Y.; Zhou, W.; Tian, G.; Fu, H. Enhanced charge transfer and separation of hierarchical hydrogenated TiO2 nanothorns/carbon nanoﬁbers composites decorated by NiS quantum dots 2. Zhang, X.; Chen, Y.; Xiao, Y.; Zhou, W.; Tian, G.; Fu, H. Enhanced charge transfer and separation of hierarchical hydrogenated TiO2 nanothorns/carbon nanoﬁbers composites decorated by NiS quantum dots for remarkable photocatalytic H2 production activity. Nanoscale 2018, 10, 4041–4050. [CrossRef] for remarkable photocatalytic H2 production activity. Nanoscale 2018, 10, 4041–4050. [CrossRef] 3. Dai, G.; Qin, H.; Zhou, H.; Wang, W.; Luo, T. Template-free fabrication of hierarchical macro/mesoporpous SnS2/TiO2 composite with enhanced photocatalytic degradation of Methyl Orange (MO). Appl. Surf. Sci. 2018, 430, 488–495. [CrossRef] 4. Lyu, J.; Gao, J.; Zhang, M.; Fu, Q.; Sun, L.; Hu, S.; Zhong, J.; Wang, S.; Li, J. Construction of homojunction-adsorption layer on anatase TiO2 to improve photocatalytic mineralization of volatile organic compounds. Appl. Catal. B 2017, 202, 664–670. [CrossRef] 5. Meng, A.; Zhang, L.; Cheng, B.; Yu, J. TiO2−MnOx−Pt Hybrid Multiheterojunction Film Photocatalyst with Enhanced Photocatalytic CO2-Reduction Activity. ACS Appl. Mater. Interfaces 2018, 11, 5581–5589. [CrossRef] [PubMed] 6. Shi, Y.; Li, H.; Wang, L.; Shen, W.; Chen, H. Novel α-Fe2O3/CdS Cornlike Nanorods with Enhanced Ph l i P f ACS A l M t I t f 2012 4 4800 4806 [C R f] [P bM d] 6. Shi, Y.; Li, H.; Wang, L.; Shen, W.; Chen, H. Novel α-Fe2O3/CdS Cornlike Nanorods with Enhanced Photocatalytic Performance. ACS Appl. Mater. Interfaces 2012, 4, 4800–4806. [CrossRef] [PubMed] 7. Ran, J.; Jaroniec, M.; Qiao, S.Z. Cocatalysts in Semiconductor-based Photocatalytic CO2 Reduction: Achievements, Challenges, and Opportunities. Adv. Mater. 2018, 30, 1704649(1)–1704649(31). [CrossRef] y y Achievements, Challenges, and Opportunities. Adv. Mater. 2018, 30, 1704649(1)–1704649(31). [CrossRef] 8. Fu, W.W.; Li, G.D.; Wang, Y.; Zeng, S.J.; Yan, Z.J.; Wang, J.W.; Xin, S.G.; Zhang, L.; Wu, S.W.; Zhang, Z.T. Facile formation of mesoporous structured mixed-phase (anatase/rutile) TiO2 with enhanced visible light photocatalytic activity. Chem. Commun. 2018, 54, 58–61. [CrossRef] 9. Meng, A.; Zhang, L.; Cheng, B.; Yu, J. Dual Cocatalysts in TiO2 Photocatalysis. Adv. Mater. 2019, 31, 1807660(1)–1807660(31). 10. Niu, S.; Zhang, R.; Zhang, Z.; Zheng, J.; Jiao, Y.; Guo, C. In situ construction of the BiOCl/Bi2Ti2O7 heterojunction with enhanced visible-light photocatalytic activity. Inorg. Chem. Front. 2019, 6, 791–798. 4. Conclusions In conclusion, anatase nanoparticles generated on H-titanate nanotubes surfaces were successfully synthesized by an ingenious method. H-titanate tubes were prepared by a hydrothermal synthesis of Ti powder in concentrated NaOH solution and an ion exchange process with HNO3 solution. After that, at a relatively low drying temperature, a small amount of anatase nanoparticles were in-situ formed on the surface of the H-titanate tubes by the surface dehydration reaction. It showed higher photocatalytic activity than pure H-titanate nanotube, anatase, and P25 under visible light (100% rhodamine B (RhB) and methylene blue (MB), 60% methyl orange (MO) degraded in 30 min under visible light irradiation) due to the in-situ transformation, the smaller average crystal size, the nanotubed and layered structure, the large BET surface areas and the synergistic eﬀect of the H-titanate/anatase dual phases, which can accelerate the transfer of electron-hole pairs and inhibit their recombination. This work provides an ingenious and simple method to prepare an eﬃcient visible-light-responsive TiO2-based photocatalyst for solving environment problems. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/10/6/695/s1, Figure S1: TEM images of (a) TiO2-60, (b) TiO2-500, Figure S2: (A) Photocatalytic degradation of RhB over the (a1) dual-phase H-titanate/anatase, (a2) H-titanate, (a3) anatase catalysts; photocatalytic degradation of MB over the (b1) dual-phase H-titanate/anatase, (b2) H-titanate, (b3) anatase catalysts; photocatalytic degradation of MO over the (c1) dual-phase H-titanate/anatase, (c2) H-titanate, (c3) anatase catalysts under visible light irradiation. (B) Photocatalytic kinetic plot of the (a1) P25, (a2) dual-phase H-titanate/anatase for degradation of RhB; photocatalytic kinetic plot of the (b1) P25, (b2) dual-phase H-titanate/anatase for degradation of MO under visible light irradiation. 13 Catalysts 2020, 10, 695 Funding: This work was supported by the Science and Technology Development Planning of Jilin Province, China (No. 20170101098JC). Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References TiO2 NPs Assembled into a Carbon Nanoﬁﬁber Composite Electrode by a One-Step Electrospinning Process for Supercapacitor Applications. Polymers 2019, 11, 899. [CrossRef] [PubMed] 23. Nowak, M.; Kauch, B.; Szperlich, P. Determination of energy band gap of nanocrystalline SbSI using diﬀuse reﬂectance spectroscopy. Rev. Sci. Instrum. 2009, 80, 046107(1)–046107(3). [CrossRef] [PubMed] 24. Zhang, W.F.; Zhang, M.S.; Yin, Z.; Chen, Q. Photoluminescence in anatase titanium dioxide nanocrystals. Appl. Phys. B Laser Opt. 2000, 70, 261–265. [CrossRef] 25. Ng, J.; Xu, S.; Zhang, X.; Yang, H.Y.; Sun, D.D. Hybridized Nanowires and Cubes: A Novel Architecture of a Heterojunctioned TiO2/SrTiO3 Thin Film for Eﬃcient Water Splitting. Adv. Funct. Mater. 2010, 20, 4287–4294. [CrossRef] 26. Wang, Q.; Zhang, B.; Lu, X.; Zhang, X.; Zhu, H.; Li, B. Multifunctional 3D K2Ti6O13 nanobelt-built architectures towards wastewater remediation: Selective adsorption, photodegradation, mechanism insight and photoelectrochemical investigation. Catal. Sci. Technol. 2018, 8, 6180–6195. [CrossRef] 27. Pant, B.; Ojha, G.P.; Kim, H.-Y.; Park, M.; Park, S.-J. Fly-ash-incorporated electrospun zinc oxide nanoﬁﬁbers: Potential material for environmental remediation. Environ. Pollut. 2019, 245, 163–172. [CrossRef] 28. Fu, W.; Ding, S.; Wang, Y.; Wu, L.; Zhang, D.; Pan, Z.; Wang, R.; Zhang, Z.; Qiu, S. F, Ca co-doped TiO2 nanocrystals with enhanced photocatalytic activity. Dalton Trans. 2014, 43, 16160–16163. [CrossRef] 29. Nguyen-Le, M.-T.; Lee, B.-K. Novel fabrication of a nitrogen-doped mesoporous TiO2-nanorod titanate heterojunction to enhance the photocatalytic degradation of dyes under visible light. RSC Adv. 2016, 6, 31347–31350. 30. Niu, F.; Chen, D.; Qin, L.; Zhang, N.; Wang, J.; Chen, Z.; Huang, Y. Facile Synthesis of Highly Eﬃcient p–n Heterojunction CuO/BiFeO3 Composite Photocatalysts with Enhanced Visible-Light Photocatalytic Activity. ChemCatChem 2015, 7, 3279–3289. [CrossRef] 31. Liu, C.; Liang, J.-Y.; Han, R.-R.; Wang, Y.-Z.; Zhao, J.; Huang, Q.-J.; Chen, J.; Hou, W.-H. S-doped Na2Ti6O13@TiO2 core-shell nanorods with enhanced visible light photocatalytic performance. Phys. Chem. Chem. Phys. 2015, 17, 15165–15172. [CrossRef] [PubMed] 32. Li, S.; Chen, J.; Zheng, F.; Li, Y.; Huang, F. Synthesis of the double-shell anatase-rutile TiO2 hollow spheres with enhanced photocatalytic activity. Nanoscale 2013, 5, 12150–12155. [CrossRef] [PubMed] 33. Pan, D.; Jiao, J.; Li, Z.; Guo, Y.; Feng, C.; Liu, Y.; Wang, L.; Wu, M. Eﬃcient Separation of Electron–Hole Pairs in Graphene Quantum Dots by TiO2 Heterojunctions for Dye Degradation. ACS Sustain. Chem. Eng. 2015, 3, 2405–2413. [CrossRef] 34. Tang, Y.; Lai, Y.; Gong, D.; Goh, K.-H.; Lim, T.-T.; Dong, Z.; Chen, Z. References [CrossRef] 11. Bavykin, D.V.; Friedrich, J.M.; Walsh, F.C. Protonated Titanates and TiO2 Nanostructured Materials: Synthesis, Properties, and Applications. Adv. Mater. 2006, 18, 2807–2824. [CrossRef] 12. Mao, Y.; Wong, S.S. Size- and Shape-Dependent Transformation of Nanosized Titanate into A Anatase Titania Nanostructures. J. Am. Chem. Soc. 2006, 128, 8217–8226. [CrossRef] [PubMed] 13. Buchholcz, B.; Haspel, H.; Kukovecz, Á.; Kónya, Z. Low-temperature conversion of titanate nanotubes into nitrogen-doped TiO2 nanoparticles. CrystEngComm 2014, 16, 7486–7492. [CrossRef] 14. Xiong, Z.; Zhao, X. Nitrogen-doped titanate-anatase core-shell nanobelts with exposed {101} anatase facets d h d i ibl li ht h t t l ti ti it J A Ch S 2012 134 5754 5757 [C R f] 14. Xiong, Z.; Zhao, X. Nitrogen-doped titanate-anatase core-shell nanobelts with exposed {101} anatase facets and enhanced visible light photocatalytic activity. J. Am. Chem. Soc. 2012, 134, 5754–5757. [CrossRef] 15. Yan, Y.; Qiu, X.; Wang, H.; Li, L.; Fu, X.; Wu, L.; Li, G. Synthesis of titanate/anatase composites with highly photocatalytic decolorization of dye under visible light irradiation. J. Alloys Compd. 2008, 460, 491–495. [CrossRef] 16. Harsha, N.; Ranya, K.R.; Babitha, K.B.; Shukla, S.; Biju, S.; Reddy, M.L.P.; Warrier, K.G.K. Hydrothermal Processing of Hydrogen Titanate/Anatase-Titania Nanotubes and Their Application as Strong Dye-Adsorbents. J. Nanosci. Nanotechnol. 2010, 10, 1–13. [CrossRef] 17. Wang, P.; Yi, X.; Lu, Y.; Yu, H.; Yu, J. In-situ synthesis of amorphous H2TiO3-modiﬁed TiO2 and its improved photocatalytic H2-evolution performance. J. Colloid Interface Sci. 2018, 532, 272–279. [CrossRef] 18. Cheng, Y.H.; Huang, Y.Z.; Kanhere, P.D.; Subramaniam, V.P.; Gong, D.G.; Zhang, S.; Highﬁeld, J.; Schreyer, M.K.; Chen, Z. Dual-Phase Titanate/Anatase with Nitrogen Doping for Enhanced Degradation of Organic Dye under Visible Light. Chem. Eur. J. 2011, 17, 2575–2578. [CrossRef] 14 Catalysts 2020, 10, 695 19. Cai, J.; Zhu, Y.; Liu, D.; Meng, M.; Hu, Z.; Jiang, Z. Synergistic Eﬀect of Titanate-Anatase Heterostructure and Hydrogenation-Induced Surface Disorder on Photocatalytic Water Splitting. ACS Catal. 2015, 5, 1708–1716. [CrossRef] 20. Zhang, L.; Lin, H.; Wang, N.; Lin, C.; Li, J. The evolution of morphology and crystal form of titanate nanotubes under calcination and its mechanism. J. Alloys Compd. 2007, 431, 230–235. [CrossRef] 21. Xiong, Z.; Dou, H.; Pan, J.; Ma, J.; Xu, C.; Zhao, X.S. Synthesis of mesoporous anatase Ti template method and photocatalysis. CrystEngComm 2010, 12, 3455–3457. [CrossRef] 22. Pant, B.; Park, M.; Park, S.-J. Mohan Sakar 1,2, Ravikumar Mithun Prakash 2 and Trong-On Do 1,* Mohan Sakar 1,2, Ravikumar Mithun Prakash 2 and Trong-On Do 1,* 1 Department of Chemical Engineering, Laval University, Quebec, QC G1V 0A 2 Centre for Nano and Material Sciences, Jain University, Bangalore 562 112, I * Correspondence: Trong-On.Do@gch.ulaval.ca Received: 22 July 2019; Accepted: 7 August 2019; Published: 9 August 2019 1 Department of Chemical Engineering, Laval University, Quebec, QC G1V 0A8, Cana 2 Centre for Nano and Material Sciences, Jain University, Bangalore 562 112, India * Correspondence: Trong-On.Do@gch.ulaval.ca Received: 22 July 2019; Accepted: 7 August 2019; Published: 9 August 2019 1 Department of Chemical Engineering, Laval University, Quebec, QC G1V 0A8, Canada 2 Centre for Nano and Material Sciences, Jain University, Bangalore 562 112, India * Correspondence: Trong-On.Do@gch.ulaval.ca Abstract: Photocatalysis is a multifunctional phenomenon that can be employed for energy applications such as H2 production, CO2 reduction into fuels, and environmental applications such as pollutant degradations, antibacterial disinfection, etc. In this direction, it is not an exaggerated fact that TiO2 is blooming in the ﬁeld of photocatalysis, which is largely explored for various photocatalytic applications. The deeper understanding of TiO2 photocatalysis has led to the design of new photocatalytic materials with multiple functionalities. Accordingly, this paper exclusively reviews the recent developments in the modiﬁcation of TiO2 photocatalyst towards the understanding of its photocatalytic mechanisms. These modiﬁcations generally involve the physical and chemical changes in TiO2 such as anisotropic structuring and integration with other metal oxides, plasmonic materials, carbon-based materials, etc. Such modiﬁcations essentially lead to the changes in the energy structure of TiO2 that largely boosts up the photocatalytic process via enhancing the band structure alignments, visible light absorption, carrier separation, and transportation in the system. For instance, the ability to align the band structure in TiO2 makes it suitable for multiple photocatalytic processes such as degradation of various pollutants, H2 production, CO2 conversion, etc. For these reasons, TiO2 can be realized as a prototypical photocatalyst, which paves ways to develop new photocatalytic materials in the ﬁeld. In this context, this review paper sheds light into the emerging trends in TiO2 in terms of its modiﬁcations towards multifunctional photocatalytic applications. Keywords: TiO2; semiconductors; photocatalysis; redox reactions; band gap engineering; nanostructures Catalysts 2019, 9, 680; doi:10.3390/catal9080680 www.mdpi.com/journal/catalysts References Ultrafast Synthesis of Layered Titanate Microspherulite Particles by Electrochemical Spark Discharge Spallation. Chem. Eur. J. 2010, 16, 7704–7708. [CrossRef] 35. Lim, Y.W.L.; Tang, Y.X.; Cheng, Y.H.; Chen, Z. Morphology, crystal structure and adsorption performance of hydrothermally synthesized titania and titanate nanostructures. Nanoscale 2010, 2, 2751–2757. [CrossRef] [PubMed] 36. Luo, Z.; Poyraz, A.S.; Kuo, C.-H.; Miao, R.; Meng, Y.; Chen, S.-Y.; Jiang, T.; Wenos, C.; Suib, S.L. Crystalline Mixed Phase (Anatase/Rutile) Mesoporous Titanium Dioxides for Visible Light Photocatalytic Activity. Chem. Mater. 2015, 27, 6–17. [CrossRef] 37. Gurulakshmi, M.; Selvaraj, M.; Selvamani, A.; Vijayan, P.; Sasi Rekha, N.R.; Shanthi, K. Enhanced visible-light photocatalytic activity of V2O5/S-TiO2 nanocomposites. Appl. Catal. A 2012, 449, 31–46. [CrossRef] 38. Kho, Y.K.; Iwase, A.; Teoh, W.Y.; Ma¨dler, L.; Kudo, A.; Amal, R. Photocatalytic H2 Evolution over TiO2 Nanoparticles. The Synergistic Eﬀect of Anatase and Rutile. J. Phys. Chem. C 2010, 114, 2821–2829. [CrossRef] 15 Catalysts 2020, 10, 695 39. Meng, A.; Zhang, J.; Xu, D.; Cheng, B.; Yu, J. Enhanced photocatalytic H2-production activity of anatase TiO2 nanosheet by selectively depositing dual cocatalysts on {101} and {001} facets. Appl. Catal. B 2016, 198, 286–294. [CrossRef] 40. Zhang, D.; Wang, Q.; Wang, L.; Zhang, L. Magnetically separable CdFe2O4/graphene catalyst and its enhanced photocatalytic properties. J. Mater. Chem. A 2015, 3, 3576–3585. [CrossRef] 41. Li, K.; Gao, S.; Wang, Q.; Xu, H.; Wang, Z.; Huang, B.; Dai, Y.; Lu, J. In-Situ-Reduced Synthesis of Ti3+ Self-Doped TiO2/g-C3N4 Heterojunctions with High Photocatalytic Performance under LED Light Irradiation. ACS Appl. Mater. Interfaces 2015, 7, 9023–9030. [CrossRef] [PubMed] 42. Yu, J.; Wang, S.; Low, J.; Xiao, W. Enhanced photocatalytic performance of direct Z-scheme g-C3N4-TiO2 photocatalysts for the decomposition of formaldehyde in air. Phys. Chem. Chem. Phys. 2013, 15, 16883–16890. [CrossRef] [PubMed] 43. Liu, Z.Y.; Sun, D.D.; Guo, P.; Leckie, J.O. One-Step Fabrication and High Photocatalytic Activity of Porous TiO2 Hollow Aggregates by Using a Low-Temperature Hydrothermal Method Without Templates. Chem. Eur. J. 2007, 13, 1851–1855. [CrossRef] [PubMed] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 16 16 16 catalysts catalysts 2. Mechanics of TiO2 Photocatalysis 2. Mechanics of TiO2 Photocatalysis Photocatalysis (PC) is the process of performing a chemical reaction in the presence of light and a photoactive catalyst, where the charge carriers (electron hole) get separated by the incident photons with suﬃcient energy and transferred to the respective bands and involved in the redox reactions. The following equations show the reaction mechanism of the photocatalytic process [29,30]. Incident photon: Photocatalyst + hv →e−+ h+ (1) Reduction: 2H+ + 2e−→H2 ΔE = 0 V (2) Oxidation: 2H2O + 4h+ →O2 + 4H+ ΔE = 1.23 V (3) Overall: 2H2O →2H2 + O2 (ΔG = + 237.2 kJmol−1) (4) Incident photon: Photocatalyst + hv →e−+ h+ (1) (1) Reduction: 2H+ + 2e−→H2 ΔE = 0 V (2) Oxidation: 2H2O + 4h+ →O2 + 4H+ ΔE = 1.23 V (3) Overall: 2H2O →2H2 + O2 (ΔG = + 237.2 kJmol−1) (4) (4) As mentioned in the reaction equations, the incident photons generate the photo-induced electron hole (e−/h+) pairs in the semiconductor and the electron involved in the reduction reactions, while the holes are involved in the oxidation reactions. The ﬁrst and foremost prerequisite for a photocatalyst is to have an appropriate band edge potential (valence band/VB, conduction band/CB) to induce the required redox species. Considering the PC process in TiO2, the VB and CB level of TiO2 lies at +2.9 and −0.3 eV, respectively, which leads to the band gap energy of 3.2 eV. It should be noted that the VB and CB level of TiO2 lies at more positive and more negative values in comparison with the standard redox potential of O2/H2O (1.23 eV) and H+/H2 (0 eV) vs. normal hydrogen electrode (NHE), which is one of the more favorable conditions for the photocatalytic redox reactions [31,32]. Apart from the band edge positions, the photocatalytic process also requires enhanced surface reactivity, charge separation, and transportations mechanisms [33]. Upon excitation, the photocatalyst should facilitate the transportation of electrons to the surface, which essentially determines the surface chemistry and reactivity of the photocatalyst. The surface of TiO2 typically contains more defects, which are often found to be oxygen vacancies; the unpaired electrons in such defects are transferred to the conduction band of TiO2 and facilitate the catalytic reactions in the system [34]. 1. Introduction Since the observation of an enhanced electrolysis of water (H2O) molecules into H2 and O2 using TiO2 as photo-anode and Pt as cathode under UV light irradiation, [1] the research on TiO2 is gaining signiﬁcant momentum towards its ‘photocatalytic’ process, which is coined later on. In 1977, Schrauzer and Guth reported the Pt/Rh metal modiﬁed-TiO2 powders for the photocatalytic splitting of water molecules [2]. Followed by such pioneering work in the ﬁeld, a range of semiconducting materials have been explored for the photocatalytic properties towards various photocatalytic applications [3–12]. Accordingly, there has been prompt progress in developing various photocatalytic systems to convert the chemical energy through water splitting [13–16] into H2 and O2 and other associated reactions [17,18]. Speciﬁcally, diverse binary oxide-based photocatalysts have been developed and demonstrated as reliable photocatalysts [19–21]. Despite the emergence of various binary oxide photocatalytic systems, TiO2 is considered as the most promising material due to its unprecedented stability, excellent physiochemical properties with ease of synthesis, availability, and relatively lower cost [22–24]. In addition to this, TiO2 exhibits three polymorphs, namely anatase, rutile, and brookite [25], in which the anatase phase is widely used because of its photocatalytic eﬃciency as its conduction band has been positioned in the appropriate Catalysts 2019, 9, 680; doi:10.3390/catal9080680 17 Catalysts 2019, 9, 680 negative potential, which is the favorable band edge position for redox reactions [26]. Despite such merits and reliable properties, TiO2 lacks in some of the other speciﬁc crucial properties for photocatalysis, such as wide bang gap energy, rapid charge recombination, insuﬃcient transportation, etc. [27]. To surpass such limitations, TiO2 has been modiﬁed in many diﬀerent ways through chemical and physical modiﬁcations, where the former involves doping, composite formation, defects creation, functionalization, plasmonic sensitization, co-catalyst loading, etc., and the other involves size, morphology, and shape modiﬁcations, etc. [28]. In this review, we have essentially focused on the versatile modiﬁcations of TiO2 such as morphology modiﬁcations, doped TiO2, hetero-junctions, Z-scheme, plasmonic, ferroelectric/perovskite, chalcogenides, metal–organic frameworks, carbon-based TiO2, defective TiO2, etc. TiO2 may be the only material that has been used to construct the any given aforementioned photocatalytic systems and investigated for almost all the photocatalytic applications such as dye degradations, pharmaceutical degradations, H2 evolution, O2 evolution, CO2 reduction, heavy metal reduction, N2 ﬁxation, organic synthesis, antimicrobial disinfection, etc. 1. Introduction Unlike other existing reviews, which merely provides TiO2 modiﬁcations such as doping, etc., this review paper gives insights into the modiﬁcations of TiO2 towards developing various photocatalytic systems as a whole, which can be prototyped using other materials. 2. Mechanics of TiO2 Photocatalysis Interestingly, the accumulation of electrons leads to the band bending phenomenon in TiO2 that considerably redesigns the transportation of charges or energy to the surrounding molecules [35]. Charge recombination dynamics is one of the serious issues in a photocatalyst. Regarding TiO2, with its indirect band gap, it is proposed that the recombination process occurs via non-radiative pathways and, thus, the lifetime of charge carriers in TiO2 varies from picoseconds to milliseconds [36,37]. In addition, the observed 18 Catalysts 2019, 9, 680 relatively enhanced PC eﬃciency of TiO2 can also be ascribed to its electron and hole trapping [38]. Generally, the photo-induced charge carriers do not tend to recombine directly due to the factors such as carrier trapping, band bending, etc. Accordingly, it is predicted that the holes in TiO2 can be trapped either at the “bridging” O2−or “surface bound” OH−anions, which results in the generation of O−• and/or OH• centers, respectively. Similarly, the photo-induced electrons can be forced to migrate into the bulk from surface, where they can be delocalized in possible Ti sites. Furthermore, it is also predicted that in TiO2 it is of more possible for bulk trapping rather than surface trapping and thereby TiO2 shows relatively enhanced photocatalytic activities as compared to the other semiconducting oxide-based photocatalysts [38–40]. 3. Versatile Modiﬁcations of TiO2 and Their Photocatalytic Mechanisms 3. Versatile Modiﬁcations of TiO2 and Their Photocatalytic Mechanisms TiO2 as a photocatalyst has been modiﬁed in a variety of ways that generally includes (i) morphological, (ii) defective, (iii) elemental doping (cationic/anionic), (iv) plasmonic metal-loading, composites with (v) binary oxides, (vi) perovskite systems, (vii) metal–organic frameworks, (viii) carbon materials, (ix) chalcogenides, etc. These modiﬁcations essentially lead to development of new photocatalytic systems, enhancing (i) the overall visible light/full-sunlight absorption, (ii) charge separation, (iii) recombination resistance, (iv) charge transportations, and (v) tuning of the band edge potential of the system. Accordingly, the following section presents some of the recent studies that mainly highlight the photocatalytic mechanism/functions in such chemically and physically modiﬁed TiO2. 3.1. Morphology-Dependent Photocatalytic Properties of TiO2 Photocatalysis can be inﬂuenced by the size, shape, and morphology of the photocatalyst due to the spatial conﬁnements of electrons in the system [41,42]. For instance, compared to bulk, the surface reactivity is higher for the nanoparticles, where their high surface area/energy facilitates the enhanced (i) catalytic activity on the surface, (ii) surface adsorption of the molecules, and (iii) promotion of charge carriers to surface. The size parameter also considerably inﬂuences the band-gap energy as well as band-edge position in a photocatalyst. Similarly, the geometrics of photocatalyst also inﬂuences the PC process. For instance, compared to the particles, the one-dimensional nanostructures show improved activity due to the enhanced “delocalization of electrons” in the conduction band of the photocatalyst [43,44]. Further, photocatalysts also demonstrate the crystal-facet-dependent eﬃciencies towards various photocatalytic applications. TiO2 nanocrystals with diﬀerent shapes, as shown in Figure 1a–f, have been synthesized and demonstrated for photo-reforming of methanol into hydrogen under UV light [45]. Figure 1. TEM images of TiO2 NCs synthesized using the precursor TiF4 (a,d), a mixed precursor of TiF4 and TiCl4 (b,e), and TiCl4 (c,f). Those depicted in a−c and d−f are synthesized in the presence of OLAM and 1-ODOL, respectively. (reproduced with permission from ref. [45]). Figure 1. TEM images of TiO2 NCs synthesized using the precursor TiF4 (a,d), a mixed precursor of TiF4 and TiCl4 (b,e), and TiCl4 (c,f). Those depicted in a−c and d−f are synthesized in the presence of OLAM and 1-ODOL, respectively. (reproduced with permission from ref. [45]). 19 Catalysts 2019, 9, 680 In another study, the synthesis of TiO2 solid and hollow nanocubes have been demonstrated, as shown in Figure 2, and applied for the photocatalytic-mediated synthesis of benzimidazole under UV and visible conditions [46]. 3.2. Doped TiO2 Doping can be essentially classiﬁed into two categories, (i) cationic and (ii) anionic doping. Accordingly, TiO2 has been widely modiﬁed through doping under both categories. The cationic and anionic doping in TiO2 leads to the formation of new energy levels underneath the conduction band and above the valence band [54]. The former doping has often been found to reduce the band gap energy and facilitates the visible light absorption and charge separation in TiO2, whereas the latter often helps in shifting of the VB position, mitigates the defects, and enhances the chemical stability of TiO2 [55]. The anionic dopants such as N, C, S, and P have been largely doped in TiO2. Among them, the N doping showed relatively enhanced photocatalytic activity due to the increased stability in the system. Similarly, there are variety of elements doped at the cationic site of TiO2 and explored for their photocatalytic activities under UV-visible light. 3. Versatile Modiﬁcations of TiO2 and Their Photocatalytic Mechanisms Similarly, TiO2 with diﬀerent morphologies such as nanospheres, nanocubes, nanotubes, nanorods, nanoﬂowers, nanosheets, and nanoﬁbers have been synthesized and studied for their photocatalytic applications [47–53]. The size and morphology control over TiO2 photocatalyst exhibit signiﬁcant inﬂuences over their (i) optical properties such as tunable band-gap energy, repositioning of band edge positions, visible light absorption, etc., (ii) electronic properties such as increased carrier lifetime, enhanced photocurrent conduction, reduced recombination, and (iii) surface properties such as enhanced surface energy, porous structures, enhanced surface adsorption, etc. Realizing the photocatalytic phenomenon, these properties are very much important to achieve the enhanced eﬃciencies in the photocatalytic materials. Figure 2. (A) Overall ﬂowchart for fabrication of black hollow nanocubic (BHC)-TiO2 (a–p), (B) Comparison photocatalytic activity of diﬀerent TiO2 nanostructures in the synthesis of benzimidazole under UV and visible conditions; (C) Schematic diagram of the light scattering eﬀect caused by BHC-TiO2 nanocubes (a) and schematic of the proposed mechanism for benzimidazole preparation by BHC-TiO2 architecture (b) (reproduced with permission from ref. [46]). Figure 2. (A) Overall ﬂowchart for fabrication of black hollow nanocubic (BHC)-TiO2 (a–p), (B) Comparison photocatalytic activity of diﬀerent TiO2 nanostructures in the synthesis of benzimidazole under UV and visible conditions; (C) Schematic diagram of the light scattering eﬀect caused by BHC-TiO2 nanocubes (a) and schematic of the proposed mechanism for benzimidazole preparation by BHC-TiO2 architecture (b) (reproduced with permission from ref. [46]). 3.2.1. Anionic Doping inTiO2 Chen et al. reported the origin of visible-light absorption characteristics of C-, N-, and S-doped TiO2 nanomaterials [56]. In their studies, the TiO2-P25 showed the typical band-edge absorption around 390 nm with band gap energy of 3.2 eV, while the C and S doping also showed the same values, however the N-doping showed an absorption around 415 nm with band gap energy of 3.0 eV. Further, their valence band-X ray photoelectron spectra revealed an interesting feature that the doping of C, S, 20 Catalysts 2019, 9, 680 and N created additional states in the TiO2 system, as shown in Figure 3A [56]. These additional states were attributed to the C 2p, S 3p, and N 2p orbitals and they were found to add deeper states into the band gap of TiO2 in the order of C > N > S. Emy et al. reported the band gap engineering in the anionic co-doped TiO2 [57]. According to their investigations, they have explained that in F-doped TiO2, the band gap reduction is mediated by the presence of surface Ti3+ defects underneath the CB, while in N-doped TiO2, the mid-band states have been formed as the N species ﬁll voids as impurities above the VB. On the other hand, the co-doping of N and F into TiO2 leads to the biggest band gap reduction to 2.24 eV from 3.19 eV, where it is attributed to the doping induced creation of defects and shifting of the VB tail towards Fermi level as shown in Figure 3B [57]. Figure 3. (A) Valence band (VB) XPS spectra of pure and (C, S, N)-doped TiO2; (B) proposed band gap engineering structure for all (F, N) doped TiO2 (reproduced with permission from refs. [56,57], respectively). Figure 3. (A) Valence band (VB) XPS spectra of pure and (C, S, N)-doped TiO2; (B) proposed band gap engineering structure for all (F, N) doped TiO2 (reproduced with permission from refs. [56,57], respectively). Based on the available experimental evidences and theoretical results obtained by Wang et al. [58], we have concluded that both the bang gap narrowing and the overlapping of O 2p state with the dopant-induced states strongly aﬀect the photocatalytic activities of anion-doped TiO2. However, Kuznetsov et al. [59] have reported that the visible light absorption happening in these doped-TiO2 may be due to the formation of color centers and may not be due to the band gap narrowing. 3.2.1. Anionic Doping inTiO2 Further, they have also argued that the red shift in the absorption edge could be due to the emergence of color centers and the doping (heavily) may completely lead to the formation of material with completely diﬀerent chemical composition from TiO2 with diﬀerent electronic band structures. However, it should be noted that the anion-doped TiO2 is considered as the second-generation photocatalysts [60]. 3.2.2. Cationic Doping in TiO2 As described, the cationic doping essentially introduces the intra-band energy levels close to the CB of TiO2, which leads to the red shift in the optical property of the system and it is also observed in various cations such as transition metal, [61–63], rare-earth [64–66], and other metals [67–69] doped TiO2. However, the main drawback of the cation doping is the creation of more trapping sites for charge carriers (both electrons and holes) that considerably reduces the eﬃciency of the photocatalyst. This is because the trapped carriers tend to recombine with the respective mobile carriers in the system. The mechanism of cation doping is essentially to tune the Fermi level and electronic structure of d-electron conﬁguration in TiO2, thereby to tune the energy levels to absorb the visible light energy and to enhance the overall photocatalytic eﬃciency of the system as shown in Figure 4a–c [70–72]. 21 21 Catalysts 2019, 9, 680 Figure 4. Band gap engineering in TiO2 via (a) Fe, (b) Ce, (c) Cu doping, showing the formation of dopant energy states underneath the conduction band of TiO2 and associated carrier dynamics (reproduced with permission from refs. [70–72], respectively). Figure 4. Band gap engineering in TiO2 via (a) Fe, (b) Ce, (c) Cu doping, showing the formation of dopant energy states underneath the conduction band of TiO2 and associated carrier dynamics (reproduced with permission from refs. [70–72], respectively). Consequently, there have been many cations doped in TiO2 towards enhancing its PC activities. In such cation doping, TiO2 has been doped with the (i) transition metals such as Sc, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Cd, and W [73–84]; (ii) rare-earth metals such as Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Er, Yb, and La [85–89]; and (iii) other metals such as Li, Mg, Ca, Se, Sr, Al, Sn, and Bi [90–97]. In the case of rare earth elements doping, the electronic conﬁgurations such as 4f, 5d, and 6s are found to be favorable to tune the band edge positions, density of states, and width of VB and CB via altering the crystal, electronic, and optical structures in TiO2 [98–100]. In addition, the rare earth elements tend to form complexes through their f-orbital and form various Lewis-based organic compounds, thereby improving the photocatalytic activities of TiO2 [101,102]. 3.2.2. Cationic Doping in TiO2 For instance, lanthanum (La) leads to the NIR absorption in TiO2 [103], cerium (Ce) owing to its tunable electronic conﬁguration of 4f states, such as 4f 05d0 (Ce4+) and 4f 5d0 (Ce3+), where it leads to the formation of mid-band gap in TiO2 that facilitates the absorption of in the visible region 400–500 nm [104,105]. 3.3. Hetero-Junction TiO2 From their results, they ﬁnally concluded that the band edge positions of the semiconductors involved should be compatible for an eﬀective inter-particle electron injection to happen in the system and, more importantly, the generated holes must be promoted and react highly at the surface to have an improved carrier separation process. heterojunction was not found to be eﬀective and, in fact, it had a tendency to decrease the eﬃciency of TiO2. From their results, they ﬁnally concluded that the band edge positions of the semiconductors involved should be compatible for an eﬀective inter-particle electron injection to happen in the system and, more importantly, the generated holes must be promoted and react highly at the surface to have an improved carrier separation process. Figure 5. Schematic diagram showing the energy band structure and electron-hole pair separation in the (a) p-n heterojunction; (b) non p-n heterojunction; (c) energy diagram illustrating the coupling of two SC in which vectoral electron transfer occurs from the light-activated SC to the non-activated TiO2; (d) diagram depicting the coupling of SC in which vectoral movement of electrons and holes is possible (reproduced with permission from refs. [107,109]). Figure 5. Schematic diagram showing the energy band structure and electron-hole pair separation in the (a) p-n heterojunction; (b) non p-n heterojunction; (c) energy diagram illustrating the coupling of two SC in which vectoral electron transfer occurs from the light-activated SC to the non-activated TiO2; (d) diagram depicting the coupling of SC in which vectoral movement of electrons and holes is possible (reproduced with permission from refs. [107,109]). As aforementioned, the charge transportation mechanism in heterojunction structure is dependent upon the band-edge levels of the semiconductors forming the heterojunction. For instance, the Fe3O4/TiO2 has been widely studied in this direction. Liu et al. [110] reported the 3D ﬂower-like α-Fe2O3@TiO2 core-shell nanostructures, in which the observed photocatalytic eﬃciency was attributed to the interfacial charge transportation. As shown in Figure 6a, where they have irradiated the photocatalyst under UV-visible light, it will excite both the semiconductors. Upon the contact of α-Fe2O3 with TiO2 system, the excited electrons in α-Fe2O3 get injected into the CB of TiO2 due to the relative work function of α-Fe2O3 (5.88 eV) and TiO2 (4.308 eV) system as it leads to the positioning of CB of TiO2 to be positioned below the CB α-Fe2O3. The study by Xia et al. 3.3. Hetero-Junction TiO2 Coupling of TiO2 with other semiconductors, especially narrow band gap semiconductors to form a heterojunction, is considered to be one of the promising strategies to improve the photocatalytic eﬃciencies of the system [106,107]. The selection of semiconductors towards forming the heterojunction should be made in such a way that they have diﬀerent band edge potential and conducting types. For instance, Figure 5a,b depicts the charge transfer mechanisms in the p-n and non p-n junctions between the semiconductors [107]. Such conﬁguration provides several features to the system, such as it helps improve the (i) charge separation, (ii) life time of the charge carriers, (iii) recombination resistance, and (iv) interfacial charge transportations towards the adsorbed molecules [106,107]. The semiconductor that coupled with the host-semiconductor would typically act as a sensitizer. In such cases, it is the sensitizers that get excited and transfer/inject the carriers into the host-semiconductor and, therefore, the VB of the sensitizer should be more cathodic than the VB of TiO2, so that the holes cannot migrate to the TiO2; thereby, the charge separation remains in the system [108]. These kinetics facilitate the phenomenon of electron injections into TiO2 as demonstrated in Figure 5c,d [109]. Based on such thermodynamics of heterojunction formations, Bessekhouad et al. developed Cu2O/TiO2, Bi2O3/TiO2, and ZnMn2O4/TiO2 heterojunctions towards the photocatalytic degradation of multiple organic pollutants Orange II, benzamide, and 4-hydroxybenzoic under UV-visible light [109]. In this study, they have discussed that the CB of Cu2O is positioned at −1.54 eV, which is more negative than the CB of TiO2 (−0.41 eV) that favored the transfer of electrons to TiO2 from Cu2O. Importantly, such electrons-transfer kinetics led to the faster degradation of Orange II molecules as compared to benzamide and 4-hydroxybenzoic molecules as they require more holes oxidation. The same results were also observed in the case of Bi2O3/TiO2 heterojunction. In the case of ZnMn2O4/TiO2 heterojunction, the CB position of ZnMn2O4 is estimated to be +0.062 eV, which is greater than the CB of TiO2. Under such circumstances, the electrons excited to the CB of ZnMn2O4 could not be transferred to TiO2, but the opposite would happen when the TiO2 is excited. However, ZnMn2O4/TiO2 22 Catalysts 2019, 9, 680 heterojunction was not found to be eﬀective and, in fact, it had a tendency to decrease the eﬃciency of TiO2. 3.3. Hetero-Junction TiO2 [111] proposed the charge transfer kinetics in α-Fe2O3@TiO2 system under UV and visible light irradiation separately, as shown in Figure 6b. They explained that under visible light irradiation, the carriers get excited in α-Fe2O3 and transferred to TiO2, whereas no excitation would happen in TiO2 as the system is irradiated by visible light and, subsequently, the charges carrier would be promoted to the surface and perform the photocatalytic redox reaction. On the other hand, it was observed that the system was irradiated under UV light, carriers in TiO2 get excited, and the α-Fe2O3 becomes recombination center of the photo-induced carriers; as a result, α-Fe2O3@TiO2 exhibits relatively poor photocatalytic activity. To address such issues and towards making the α-Fe2O3@TiO2 to work eﬃciently, Lin et al. [112] developed TiO2 with abundant oxygen vacancies via self-doping, which greatly shifted the VB edge position to 2.50 eV (vs. NHE), which is very close to that of α-Fe2O3 (2.48 eV) and unaltered CB position with respect to the CB position of α-Fe2O3, as shown in Figure 6c. However, despite the considerable amount of research that has been done on TiO2-based heterojunction photocatalyst, the carrier dynamics and their transportation, and thereby the photocatalytic process, should be studied in detail [113,114]. 23 Catalysts 2019, 9, 680 Figure 6. (a,b) Schematic diagram of the band edge positions and charge transfer mechanism in various α-Fe2O3@TiO2 photocatalytic systems under UV and visible light irradiation. (c) The presence of abundant oxygen vacancies in TiO2 shifts its VB edge position and aligns it to the VB of Fe2O3 (reproduced with permission from refs. [110–112], respectively). Figure 6. (a,b) Schematic diagram of the band edge positions and charge transfer mechanism in various α-Fe2O3@TiO2 photocatalytic systems under UV and visible light irradiation. (c) The presence of abundant oxygen vacancies in TiO2 shifts its VB edge position and aligns it to the VB of Fe2O3 (reproduced with permission from refs. [110–112], respectively). 3.4. Z-Scheme-Based TiO2 The concept of Z-scheme photocatalytic process is essentially derived from the natural photosynthesis process, which demonstrated a signiﬁcantly enhanced potential towards accomplishing high photocatalytic eﬃciencies [115]. The Z-scheme photocatalyst is typically constructed by coupling two photocatalytic semiconductors, which is likely similar to the conventional heterojunction photocatalyst [116]. However, Z-scheme has a unique mechanism for the injection/transfer of charge carrier into the adjacent semiconductor, as shown in Figure 7a,b [117]. Notably, among the two coupled photocatalysts in Z-scheme, one will be an oxidation and the other will be a reduction photocatalyst. The selection of such oxidation and reduction photocatalyst will be based on the VB and CB edge position, which is dependent upon the speciﬁc applications [118]. As a result of such meticulous construction, Z-scheme systems demonstrate exotic features such as (i) simultaneous strong reduction-oxidation abilities, (ii) spatial separation of reduction and oxidation active sites, (iii) enhanced carrier-separation eﬃciency with high redox abilities, and (iv) extended light absorption range [119,120]. 24 Catalysts 2019, 9, 680 Figure 7. Schematic illustration of the (a) typical heterojunction and (b) Z-scheme photocatalysts (reproduced with permission from ref. [117]). Figure 7. Schematic illustration of the (a) typical heterojunction and (b) Z-scheme photocatalysts (reproduced with permission from ref. [117]). In the Z-scheme-based systems, TiO2 has been largely used as oxidation photocatalyst owing to their low VB position and accordingly, it has been coupled with the other photocatalytic systems such as CdS [121,122], g-C3N4 [123–125], NiS [126], ZnIn2S4 [127], Cu2O [128], and WO3−x [129] owing to their high CB position that act as the reduction photocatalysts. As shown in Figure 7a [117], in the typical heterojunction photocatalyst, the separated electron holes in PCI will be injected into the respective CB and VB of the PCII. In contrast, the charge transfer mechanism in Z-scheme always follows a signature pathway in which the electrons excited to the CB of low VB photocatalyst will be injected into the VB of the high CB photocatalyst (Figure 7b) [117]. As listed above, Figure 8a–d shows the mechanism of various TiO2-based Z-scheme photocatalysts. Interestingly, Fu et al. [128] proposed a Z-scheme system mediated by Ag located at the interface of the TiO2 and Cu2O. 3.4. Z-Scheme-Based TiO2 They observed that the TiO2 and Cu2O coupled photocatalyst demonstrated a relatively poor photocatalytic performance; as a result, they proposed that upon the irradiation of TiO2 and Cu2O, the electrons in the CB of Cu2O get transferred into the TiO2 and meanwhile, the holes in VB of TiO2 get transferred to Cu2O. Such a process essentially led to the depletion of hole density in the VB of TiO2 and it increased in the VB of Cu2O. Under such circumstances, due to the low positive VB edge position of Cu2O, it has insuﬃcient energy to oxidize the OH or H2O molecules. To address such an issue, they introduced Ag into the interfacial contact of TiO2 and Cu2O, as shown in Figure 8e [128]. In this TiO2–Ag–Cu2O system, ﬁrstly, the equilibrium in Fermi levels has been established; thereby, upon irradiation, the excited electrons in the TiO2 CB get injected into Ag and due to the localized electric ﬁeld created by Ag, these electrons are further injected into the Cu2O and enhanced the photocatalytic eﬃciency of the system. Further, they proposed that this system keeps the photo-induced holes on more positive potential (VB of TiO2) and electrons on more negative (CB of Cu2O), which essentially enhance the redox ability as well as the charge separation eﬃciencies of the system as a whole. 25 Catalysts 2019, 9, 680 Figure 8. Charge transfer mechanism in various Z-scheme-based TiO2 photocatalysts, (a) CdS/TiO2, (b) g-C3N4/TiO2, (c) NiS/TiO2, (d) ZnIn2S4/TiO2, and (e) TiO2–Ag–Cu2O (reproduced with permission from refs. [121,124,126–128], respectively). Figure 8. Charge transfer mechanism in various Z-scheme-based TiO2 photocatalysts, (a) CdS/TiO2, (b) g-C3N4/TiO2, (c) NiS/TiO2, (d) ZnIn2S4/TiO2, and (e) TiO2–Ag–Cu2O (reproduced with permission from refs. [121,124,126–128], respectively). 3.5. Plasmonic TiO2 Plasmonic photocatalysis is one of the emerging and interesting concepts in this ﬁeld [130]. These types of photocatalysts make use of the plasmonic nanoparticles to harvest energy in the visible region [131]. It extends the absorption range of the photocatalyst in UV-visible-IR region [132]. The plasmonic nanoparticles also play an important role in alerting the charge transfer mechanism in the host photocatalysts. The plasmon-mediated process in photocatalysts can occur in four diﬀerent ways, (i) direct migration of carriers from the plasmonic particles to photocatalyst, (ii) indirect migration of carriers between the plasmonic particles and photocatalyst via the localized surface plasmon resonance (LSPR), (iii) localized plasmonic heating, and (iv) radiative transfer of photons from the plasmonic particles to the photocatalyst, where these photons will excite the photocatalyst to generate the electron hole pairs in the system [131–133]. However, the origins and functions of plasmonic photocatalysts are under hot debate. Noble metals such as Ag, Au, Pd, and Pt have been integrated with TiO2 to produce the TiO2-based plasmonic photocatalysts. Among them, Ag–TiO2 has been relatively largely studied with diﬀerent conﬁgurations [134–137]. Plasmonic sensitization conventionally happens by the deposition of plasmonic nanoparticles (NPs) onto the surface of the host photocatalyst. However, there have been other conﬁgurations such as core-shell structuring [137], ﬁlling up the plasmonic NPs into the pores of the host photocatalyst, and composite-like formation [135]. As aforementioned, the plasmonic nanoparticles can extend the light absorption in the visible region and they can also substantially inﬂuence the charge transfer kinetics the photocatalyst. However, there are essentially two pathways 26 Catalysts 2019, 9, 680 proposed regarding their charge transfer, which is either from the (i) plasmonic NPs to photocatalyst or (ii) photocatalyst to plasmonic NPs [130]. As a result, it has also been proposed that the scheme of such charge transfer is also determined by the relative band edge potential, conducting type (n/p-type), and work function of the photocatalyst and plasmonic metal, respectively, and also determined by the light source that is used to excite the plasmonic photocatalyst system, as shown in Figure 9a–b [134,138]. Figure 9. (a) Band bending occurs in the metal-semiconductor junction and (b) charge transfers in plasmonic photocatalyst, depending upon the light source irradiated (reproduced with permission from refs. [134,138], respectively). Figure 9. 3.5. Plasmonic TiO2 (a) Band bending occurs in the metal-semiconductor junction and (b) charge transfers in plasmonic photocatalyst, depending upon the light source irradiated (reproduced with permission from refs. [134,138], respectively). As depicted in Figure 9a [138], the work function of the metal nanoparticle with respect to the host semiconductor also directs the course of charge transfer in the plasmonic photocatalyst. For instance, the work function of Au, Ag, and anatase TiO2 has the work function of 5.23, 4.25–4.37, and 5.10 eV, respectively, where the Au–TiO2 and Ag–TiO2 follow the Schottky-junction and Ohmic-junction, respectively, for the charge transfer in the system, as shown in Figure 10a,b [139]. Compared to the Ag and Au, the surface plasmon resonance (SPR) properties of Pt/Pd-deposited TiO2 has been less explored [140]. However, these metal NPs have been explored as a co-catalyst for various photocatalyst systems [141–143]. This is because the plasmonic peak of Pt NPs appears below 450 nm, while the SPR properties of Ag and Au can be well tuned in visible to IR region, and therefore, the Pt and Pd NPs have not been typically used for developing the plasmonic photocatalysts [144–146]. 27 Catalysts 2019, 9, 680 Figure 10. Work function dependent band-bending in (a) Au/TiO2, (b) Ag/TiO2 plasmonic systems (reproduced with permission from ref. [139]). Figure 10. Work function dependent band-bending in (a) Au/TiO2, (b) Ag/TiO2 plasmonic systems (reproduced with permission from ref. [139]). 3.6. Ferroelectrics Modiﬁed TiO2 Ferroelectrics are deﬁned by the spontaneous electric polarization that can be induced by an external electric ﬁeld, where the induced spontaneous polarization will be permanent in the material and it essentially originates from the oﬀ-center displacements of ions in a non-centrosymmetric crystal system [147]. In ferroelectric materials, the internal screening induced by the free carriers and the bulk defects lead to the distribution of charge carriers in the near surface of the material, which essentially creates a space-charge region and band bending in the system [148]. These features greatly help in the photocatalytic process. The bands of ferroelectrics bend at the near the surface or interface region, depending upon the positive or negative spontaneous polarizations, as shown in Figure 11a,b [149]. Figure 11. Schematic diagram of band bending in a ferroelectric material; (a) a surface with negative polarity and (b) a surface with positive polarity. Figure 11. Schematic diagram of band bending in a ferroelectric material; (a) a surface with negative polarity and (b) a surface with positive polarity. For instance, in a negatively polarized surface, the electrons will be depleted from the surface, which leads to a creation of a spatial-charge layer (depletion layer) with “upward” band-bending. On the other hand, in a positively polarized surface, the electrons will be accumulated for screening, which leads to a “downward” band bending in the system along with formation of a spatial accumulation charge layer. Thereby, these interesting features in ferroelectric, along with such deformed migration of charge carriers, largely helpful to exhibit exotic photo-active chemical properties [150,151]. The features such as the spontaneous polarization, deformed migration of carriers, surface charges, band bending process, and the external and/or internal screening eﬀects altogether direct the photo-induced charge carriers in a ferroelectric toward an eﬀective oxidation and reduction reaction for various photocatalytic applications [152–158]. 28 Catalysts 2019, 9, 680 Ferroelectric materials such as BaTiO3 [159–161], BiFeO3 [162,163], PbTiO3 [164] have been successfully integrated with TiO2 to produce ferroelectric-TiO2 photocatalysts. Zhang et al. have explained how the ferroelectric phenomenon inﬂuences the photocatalytic activity of the system, where they demonstrated it using BiFeO3/TiO2 system [162]. They proposed a plausible energy level for the BiFeO3/TiO2 system, as shown in Figure 12a,b. 3.6. Ferroelectrics Modiﬁed TiO2 According to this diagram, the energy levels at the interface of BiFeO3 (BFO) and TiO2 are strongly inﬂuenced by the induced polarization in BiFeO3, where it bends the band of BFO upward when the polarization in negative (i.e., away from the surface) and downward when the polarization is positive (i.e., towards the surface). Under such circumstances, the photo-induced electrons in negative domains are impeded by the energy barrier at the interface; meanwhile, in positive domains, the electrons are moved to the interface, in such a way that it facilitates the photocatalytic activity with enough redox abilities of the excited charge carriers in the system [163]. Figure 12. The energy bands at the BiFeO3/TiO2 interface bend (a) upward and (b) downward corresponding to the applied polarization (reproduced with permission from ref. [162]). Figure 12. The energy bands at the BiFeO3/TiO2 interface bend (a) upward and (b) downward corresponding to the applied polarization (reproduced with permission from ref. [162]). 3.7. Carbon-Based TiO2 Composites 3.7. Carbon-Based TiO2 Composites Carbon-based materials-modiﬁed TiO2 photocatalysts demonstrate signiﬁcant enhancements in the photocatalytic process due to various reasons such as (i) high surface area, (ii) enhanced electrical conductivity, (iii) tunable optical properties, (iv) improved surface adsorption eﬃciency, and (v) controllable structural features [165–167]. These properties essentially help improve the overall properties of the photocatalysts. For instance, the enhanced surface area populates more catalytic-sites on the surface of the catalysts. The enhanced electrical conductivity improves the charge separation and transportation characteristics of the system. The tunable optical properties help activate the photocatalyst under a desirable light source such as visible light and/or sunlight. The improved surface adsorption essentially paves the way for the adsorption of surrounding molecules onto the surface of the photocatalyst that eventually enhances the interfacial interaction of the photocatalyst and molecules. Finally, the controllable structural features of carbon materials such as quantum dots (fullerenes) [168–170], 2D materials (graphene, g-C3N4) [171,172], 1D materials (carbon nanotubes (CNTs), carbon ﬁbers) [173–176], and 3D materials (carbon spheres, ﬂowers) [177,178] oﬀer unique charge transportations and improve the overall eﬃciency of the carbon-based photocatalytic materials. TiO2 has been modiﬁed by the variety of carbon-based materials such as carbon doping, carbon coating, composites with activated carbon, graphene/graphene oxide/reduced-graphene oxide, g-C3N4, CNTs, carbon ﬁbers, anisotropic carbon structures, etc. [165–178]. The general photocatalytic mechanisms of these carbon-based TiO2 systems are summarized in Figure 13a–d [168,179–181] Yu et al. [168], have reported the mechanism of carbon quantum dots (CQDs)-integrated TiO2 towards photocatalytic H2 production. The CQDs play a dual vital role in the improved photocatalytic properties. 29 Catalysts 2019, 9, 680 During the photocatalytic excitation under UV light, the CQDs act as (i) electron reservoirs and (ii) photo-sensitizers. The former role of CQDs essentially plays a role in trapping the photo-generated electrons from the conduction band of TiO2 and facilitates the enhanced process of electrons-holes separation. On the other hand, the latter characteristics of π-conjugated CQDs is to sensitize the TiO2 as similar to the organic dyes, towards making it a visible light active “dyade”-like structure, where it gives the electrons to the CB of TiO2 and leads to the visible light-driven hydrogen production (Figure 13a) [168]. Figure 13. Photocatalytic mechanism in various carbon-TiO2 systems, (a) carbon QD-TiO2, (b) carbon nanotubes (CNT)-TiO2, (c) g-C3N4-rGo-TiO2, (d) rGO-TiO2 (reproduced with permission from refs. [168,179–181], respectively). Figure 13. 3.8. 2D-Transition Metal Chalcogenides Modiﬁed TiO2 3.8. 2D-Transition Metal Chalcogenides Modiﬁed TiO2 It is well established that the large surface-to-volume ratio of 2D nanostructures can provide more surface-active sites for the photocatalytic reactions. The planar structure of 2D materials essentially favors the charge transportations across the interfaces of the catalyst and surrounding phases and thereby it drastically improves the photocatalytic eﬃciencies [186]. Moreover, as compared to other nanostructures, the 2D nanostructures exhibit exotic properties owing to the atomic arrangements with surface atomic elongation and structural-disorder characteristics [187]. These interesting physical structure-induced properties of 2D materials largely contribute in enhancing the photo-stability and chemical durability of the photocatalyst. Furthermore, 2D materials, due to their ﬂat band potential and eﬀective band bending at the interface, help tune the band gap energies and band-edge positions of the photocatalysts [188]. Speciﬁcally, when these 2D materials couple with the other metal and metal oxides, their unique 2D structures serve as a matrix for those integrated materials and enhance the optical and electrical properties of the system as a whole [189–191]. In this direction, the 2D transition metal chalcogenides (2D TMC) with general chemical formula of MX2, M = Mo, or W and X = S, Se, or Te serve as both the independent or composite photocatalytic materials [191]. Accordingly, TiO2 has been modiﬁed with these 2D TMC materials to avail their structural features and unique properties towards various photocatalytic applications. Among the listed 2D TMC materials, the MoS2/TiO2 system has been largely explored for the photocatalytic applications [192–197]. Interestingly, the charge transfer in this system depends upon the photon energy used to excite the system. The Figure 14a,b shows the charge transfer in a MoS2/TiO2 system that irradiated under UV and visible light, respectively [198,199]. When the MoS2/TiO2 system irradiated under UV light, the electrons that were excited in TiO2 will be transferred to the attached MoS2 nanosheets; thereby, this process signiﬁcantly limits the electron hole recombination and promotes carrier separation by eﬀectively transporting to the adsorbed H+ ions to reduce them to produce molecular hydrogen. On the other hand, when the MoS2/TiO2 system is irradiated under visible light, the electron transfer occurs from the MoS2 to TiO2, as shown in Figure 14b [199]. It should be noted that the TiO2 used in this study is doped with N species that facilitates visible light absorption in TiO2 as well. 3.7. Carbon-Based TiO2 Composites Photocatalytic mechanism in various carbon-TiO2 systems, (a) carbon QD-TiO2, (b) carbon nanotubes (CNT)-TiO2, (c) g-C3N4-rGo-TiO2, (d) rGO-TiO2 (reproduced with permission from refs. [168,179–181], respectively). The carbon nanotubes (CNTs), owing to their large electron-storage capacity (per electron for every 32 C-atoms), accept the photo-induced electrons from the supported semiconductor and, thereby, they largely hinder the recombination of charge carriers [179]. It is believed that the excellent conductive nature of the CNTs promotes the electron-hole separation via the formation of a heterojunction between CNTs and semiconductors. For instance, as similar to the carbon QDs, the CNTs also play a dual role in the photocatalytic process. Accordingly, the freely moving electrons in the excited TiO2 get transferred into the CNTs scaﬀolds, where the excess holes in the VB in TiO2 are set to reach and react with the H2O and OH−to generate radicals such OH• as shown in Figure 13b [179]. On the other hand, it is known that TiO2 is UV-driven, but it is observed that the CNTs-TiO2 nanocomposites have become visible light driven, which is attributed to the photo-sensitizing eﬀect of CNTs. In this scenario, the photo-induced electrons in CNTs (sensitizers) get injected into the CB of TiO2 and lead to reducing the adsorbed molecular oxygen to form the superoxide species. In parallel, the holes in these positively charged CNTs react with H2O and form OH• radicals, as shown in Figure 13b. g g Yu et al. [180] have demonstrated that the coupling between TiO2 and g-C3N4 cannot lead to the formation of heterojunction; rather, it always tends to form the Z-scheme-based photocatalyst system. Based on their experiments, they have explained the phenomenon that if TiO2 and g-C3N4 form a heterojunction, then the following scenario will emerge. Under the UV exposure, the photo-induced holes will get transferred from the VB of TiO2 to that of the g-C3N4 and the electrons will get transferred from CB of g-C3N4 to that of the TiO2. As a result, the holes of g-C3N4 cannot oxidize the adsorbed H2O or OH−to form the OH• radicals due to the higher potential of VB of g-C3N4 with respect to the H2O/OH−couple. Such a process eventually leads to the lower oxidation, thereby the photocatalytic 30 Catalysts 2019, 9, 680 eﬃciency of the system is much lower than the TiO2. 3.7. Carbon-Based TiO2 Composites However, the observed photocatalytic eﬃciency of TiO2/g-C3N4 is higher than the individual counterparts, which essentially means that this system forms a direct Z-scheme system without the electron mediator, as shown in Figure 13c [180]. The photocatalytic mechanism in the reduced graphene oxide (rGO)-TiO2 composite has been proposed by Tan et al. [181] as shown in Figure 13d. In the rGO-TiO2 composite, the d and π orbital of TiO2 and rGO, respectively, matches well in their energy levels and they overlap each other well (d-π). As a result, rGO is bound to serve as an electron-collector as well as a transporter towards eﬀectively separating the photo-induced electron-hole pairs, which eventually enhances the lifetime of the charge carriers as well, and thereby the photocatalytic eﬃciency of the rGO-TiO2 system [182–185]. 3.8. 2D-Transition Metal Chalcogenides Modiﬁed TiO2 However, their experimental investigations using ESR and PL demonstrated that the proposed charge is not possible, owing to the low potential of 0.98 V that cannot eﬀectively oxidize the adsorbed surface H2O to produce OH• radicals. Therefore, the photo-generated electrons in the CB of TiO2 might have been transferred and recombined with the holes in MoSe2 VB, leaving the holes in the VB of TiO2 and electrons in the CB of MoSe2 via constructing a ‘direct Z-Scheme’ to augment the photocatalytic redox reactions in the system, as shown in Figure 16b [203]. Similarly, the WS2/TiO2 system has also been explored for various photocatalytic applications and their mechanisms [204–210]. holes in TiO2 VB are likely to migrate into the MoSe2 VB if type-II has been formed. However, their experimental investigations using ESR and PL demonstrated that the proposed charge is not possible, owing to the low potential of 0.98 V that cannot eﬀectively oxidize the adsorbed surface H2O to produce OH• radicals. Therefore, the photo-generated electrons in the CB of TiO2 might have been transferred and recombined with the holes in MoSe2 VB, leaving the holes in the VB of TiO2 and electrons in the CB of MoSe2 via constructing a ‘direct Z-Scheme’ to augment the photocatalytic redox reactions in the system, as shown in Figure 16b [203]. Similarly, the WS2/TiO2 system has also been explored for various photocatalytic applications and their mechanisms [204–210]. Figure 16. Charge transfer mechanism in MoSe2/TiO2 (a) heterojunction and (b) Z-scheme (reproduced with permission from refs. [201,203]). Figure 16. Charge transfer mechanism in MoSe2/TiO2 (a) heterojunction and (b) Z-scheme (reproduced with permission from refs. [201,203]). 3.8. 2D-Transition Metal Chalcogenides Modiﬁed TiO2 Therefore, the coupling of MoS2 with TiO2 promotes the excited electrons to the CB of TiO2 from the CB of MoS2. The further photocatalytic reactions essentially occur via the conventional redox reactions on the surface of the photocatalyst. 31 31 Catalysts 2019, 9, 680 Figure 14. Photocatalytic charge transfer process in MoS2/TiO2 under the irradiation of (a) UV light and (b) visible light (reproduced with permission from refs. [198,199]). Figure 14. Photocatalytic charge transfer process in MoS2/TiO2 under the irradiation of (a) UV light and (b) visible light (reproduced with permission from refs. [198,199]). Zhang et al., have reported the possible charge transfer mechanism in P25-TiO2/MoS2 and P25-TiO2/WS2 systems under UV-visible irradiation [200]. Accordingly, the excited electrons in P25-TiO2/MoS2 migrate from the CB of TiO2 to the CB of MoS2, while it occurs vice versa in the P25-TiO2/WS2 system, as shown in Figure 15a,b [200]. The observed charge transfer mechanism is essentially due to the relative band-edge potentials of the semiconductors involved in the composite. Figure 15. Photocatalytic charge transfer mechanism in (a) P25-TiO2/WS2 and (b) P25-TiO2/MoS2 (reproduced with permission from ref. [200]). Figure 15. Photocatalytic charge transfer mechanism in (a) P25-TiO2/WS2 and (b) P25-TiO2/MoS2 (reproduced with permission from ref. [200]). Similar to the aforementioned systems, there are alternative hypotheses to explain the charge transfer mechanism in MoSe2/TiO2 system. Chu et al. [201] and Shen et al. [202] have proposed that the MoSe2/TiO2 follows the heterojunction mechanism towards the charge transfer process in the system, as shown in Figure 16a [201]. Accordingly, the type-II heterostructure, which formed between MoSe2 and TiO2, facilitates the electron transfer from the CB of MoSe2 to that of TiO2 and reduces the recombination process, prolongs the lifetime of the carriers, and provides an enhanced conductivity in the system towards transporting the carriers to the surrounding for the eﬀective photocatalytic process. On the other hand, Zheng et al. proposed that this system follows the Z-scheme to transfer the charges from the TiO2 to MoSe2 [203]. According to their hypothesis, the MoSe2/TiO2 (nanotubes) photocatalyst could not form a type-II heterojunction. This may be because of the reason that the 32 Catalysts 2019, 9, 680 holes in TiO2 VB are likely to migrate into the MoSe2 VB if type-II has been formed. 3.9. Metal-Organic Framework-TiO2 Composites Metal–organic frameworks (MOFs) are an exotic class of crystalline materials with inherent porous structures. MOFs are constructed using the metal clusters that interconnected by organic ligands built into a 3D networked structure. Their unique properties, such as the well-ordered porosity, very high speciﬁc surface area, and tunable surface chemistry, have made them a promising material for various applications, including photocatalysis. MOFs can be reliable photocatalytic materials due to semiconductor-like properties. In addition, they possess high surface area that largely facilitates enhanced surface catalytic activities; the metal clusters play a role in the eﬀective absorption of incident photons and charge separation, while the ligands favor the charge transportations in the system. However, the major issue in MOFs is the moderate charge separation that considerably reduces the overall photocatalytic eﬃciency of the MOFs [211–216]. Yao et al. proposed the observed superior photocatalytic eﬃciency of TiO2@-NH2-UiO-66 composites towards the degradation of styrene [217]. According to their ﬁndings, (i) the plenty of available interconnected nanopore facilitated the enhanced and rapid diﬀusion of the surrounding styrene molecules into the pores of MOFs, where the encapsulated TiO2 eﬀectively oxidized the molecules with the produced oxidation radical species, and (ii) the linkers in MOFs acted as antenna to augment the light absorption and sensitize the TiO2 and led to the eﬀective absorption of light towards the transportation of charge carriers in the system; thereby, it demonstrated excellent photocatalytic activity [217]. Similarly, the photocatalytic eﬃciency of TiO2/NH2-UiO-66 nanocomposites towards CO2 reduction has been demonstrated by Crake et al. [218]. Based on their observations, the composite of NH2-UiO-66 and TiO2 can lead to the formation of type-II heterojunction. This could essentially be because of the factor that the CB position of NH2-UiO-66 lies at −0.6 eV, while the TiO2 CB lies at a more negative potential at −0.28 eV, as shown in Figure 17a [218]. They have further proposed that the photocatalytic activity of TiO2/NH2-UiO-66 nanocomposites was mainly ruled out by (i) the concentration of TiO2, (ii) the eﬀective charge separation characteristics of NH2-UiO-66, and (iii) the enhanced availability of charge carriers at the interface of the TiO2/NH2-UiO-66 system. Ling et al. have synthesized a ternary nanocomposite composed of TiO2/UiO-66-NH2/graphene oxide and studied 33 Catalysts 2019, 9, 680 towards the photocatalytic dye (RhB) degradation and H2 evolution [219]. 3.9. Metal-Organic Framework-TiO2 Composites They have reported that, under the visible excitation, the electrons tend to transfer from RhB* to CB of MOFs to CB of TiO2 due to the cascading potential of these systems. Under such circumstances, the integrated GO captures the electrons from the CB of TiO2 that eventually enhances charge separation, thereby accelerating the dye removal. On the other hand, the electrons from GO further migrate to the Pt and lead to the H2 production. It is also possible that the electrons from RhB* can get directly injected into Pt and produce H2, as shown in Figure 17b [219]. Similarly, there have been other TiO2/MOFs-based photocatalytic systems reported [220–225]. towards the photocatalytic dye (RhB) degradation and H2 evolution [219]. They have reported that, under the visible excitation, the electrons tend to transfer from RhB* to CB of MOFs to CB of TiO2 due to the cascading potential of these systems. Under such circumstances, the integrated GO captures the electrons from the CB of TiO2 that eventually enhances charge separation, thereby accelerating the dye removal. On the other hand, the electrons from GO further migrate to the Pt and lead to the H2 production. It is also possible that the electrons from RhB* can get directly injected into Pt and produce H2, as shown in Figure 17b [219]. Similarly, there have been other TiO2/MOFs-based photocatalytic systems reported [220–225]. Figure 17. Photocatalytic charge transfer process in (a) TiO2/NH2-UiO-66 for CO2 reduction and (b) TiO2/NH2-UiO-66/GO/Pt for dye removal and H2 production (reproduced with permission from refs. [218,219]). Figure 17. Photocatalytic charge transfer process in (a) TiO2/NH2-UiO-66 for CO2 reduction and (b) TiO2/NH2-UiO-66/GO/Pt for dye removal and H2 production (reproduced with permission from refs. [218,219]). 3.10. Reduced/Defective/Colored TiO2−x Photocatalysts 3.10. Reduced/Defective/Colored TiO2−x Photocatalysts The oﬀ-stoichiometricity in TiO2, which is induced by processes such as self-doping by Ti3+ ions and oxygen vacancy creations (Vo), plays an important role in enhancing the visible light absorption and photocatalytic eﬃciency of the TiO2 materials [226–230]. Based on such a modiﬁcation approach, TiO2 has been synthesized in a variety of “colors” such as black, blue, red, and yellow. The reduced band gap energy in oﬀ-stoichiometric TiO2 essentially originates due to the formation of localized energy states (0.75–1.18eV) underneath the CB minimum of the TiO2 [231]. As compared to any other modiﬁcation strategies, the self-doping and/or oxygen vacancy creation is more favorable for maintaining the intrinsic properties of the TiO2 as well as to introduce the visible light absorption characteristics and enhance the photocatalytic eﬃciencies of TiO2 [232–234]. The ﬁrst black-TiO2 was produced by Chen et al. with band gap energy of around 1.0 eV via high-pressure hydrogenation process in the crystalline TiO2 [235]. The general mechanism for the formation of black TiO2 is broadly attributed to the presence of Ti3+ by self-doping, formation of hydroxyl groups on the surface, oxygen vacancies, Ti-H bonds, and the formation of H-energy states in the mid-gap of the TiO2 band structure, which eventually dispersed the VB in TiO2, as shown in Figure 18a [235]. Zhu et al. synthesized the stable blue TiO2 nanoparticles [236] and proposed the origin that the observed blue color could be due to the high concentration of Ti3+ defects in the bulk and the formation of mid-gap electronic energy states beneath the band gap of TiO2. As a result, the observed enhanced photocatalytic properties were attributed to their unique structural features, which is the disordered-core/ordered-shell-like structure. This essentially means that the TiO2 was stoichiometric at the surface while it was oﬀ-stoichiometric in the core. These features collectively improved the overall photocatalytic eﬃciencies of the blue TiO2 by enhancing the charge separation and transportation, as shown in Figure 18b [236]. 34 Catalysts 2019, 9, 680 Figure 18. Band gap structure of (a) black-TiO2 and (b) blue-TiO2 (reproduced with permission from refs. [235,236]). Figure 18. Band gap structure of (a) black-TiO2 and (b) blue-TiO2 (reproduced with permission from refs. [235,236]). Wu et al. developed ultra-small yellow TiO2 nanoparticles via simple sol-gel process with UV treatment technique. 3.10. Reduced/Defective/Colored TiO2−x Photocatalysts Based on their experimental ﬁndings, the origin of the observed yellow color of TiO2 could be due to titanium vacancies (VTi) and titanium interstitials (Tii) as shown in Figure 19a [237]. Interestingly, Liu et al. prepared the red anatase TiO2 via a gradient co-doping of B-N into the system. It was observed that the band gap energy varied from 1.94 eV on the surface to 3.22 eV in the core, as shown in Figure 19b [238]. Figure 19. (a) Structure of yellow-TiO2 and (b) anatase red TiO2 via gradient B-N co-doping (reproduced with permission from refs. [237,238]). Figure 19. (a) Structure of yellow-TiO2 and (b) anatase red TiO2 via gradient B-N co-doping (reproduced with permission from refs. [237,238]). Ren et al. reported that the NaBH4 reduced TiO2 photocatalysts with a range of colors such as white, light-yellow, light-grey, and dark-grey, which were prepared by varying the concentration of the reducing agent NaBH4, as shown in Figure 20a [239]. The observed color variation was attributed to the self-doping of Ti3+ ions into the TiO2. Similarly, Fan et al. reported the synthesis of TiO2 with white, dark brown, light brown, yellow, light yellow, gray, yellowish gray, and yellowish white color (Figure 20b) that were derived from the amorphous hydrated TiO2 through hydroxylated and N-doping process with a controlled degree of disorders using a heating treatment technique [240]. In this study, the observed color variation was attributed to the heating process that turned the Ti–OH bonds in amorphous TiO2 into the Ti–O bonds that transformed the disordered TiO6 octahedron into a regular 3D structure. As a result, the formed hydroxylated anatase TiO2 with enhanced degree of disorder strongly inﬂuenced the optical transition in TiO2 and narrowed down the band gap energy. Further, these colored TiO2 materials have also demonstrated enhanced photocatalytic eﬃciencies towards the degradation of acid fuchsin under visible light. 35 Catalysts 2019, 9, 680 Figure 20. Photographic images of the (a) chemically reduced TiO2 with increasing concentration of NaBH4 and (b) hydroxylated and N-doped anatase TiO2 that were derived from the amorphous hydrate TiO2 at the increasing processing temperature (reproduced with permission from refs. [239,240]). Figure 20. Photographic images of the (a) chemically reduced TiO2 with increasing concentration of NaBH4 and (b) hydroxylated and N-doped anatase TiO2 that were derived from the amorphous hydrate TiO2 at the increasing processing temperature (reproduced with permission from refs. [239,240]). Funding: This work was supported by the Natural Science and Engineering Research Council of Canada (NSERC) through the Collaborative Research and Development (CRD), Strategic Project (SP), and Discovery Grants (DG). MS gratefully acknowledges the Department of Science and Technology, Govt. of India for the funding support through the DST-INSPIRE Faculty Award [DST/INSPIRE/04/2016/002227, 14-02-2017]. Acknowledgments: We would also like to thank EXP Inc. and SiliCycle Inc. for their support. 4. Summary and Outlook Undoubtedly, TiO2 is indeed an interesting material for various photocatalytic applications. As described, the fundamental photocatalytic process involves the excitation of photo-induced carriers and their successful transfer to the surface to produce the desired redox species towards the designated photocatalytic application. The versatile applications emerge essentially due to the produced redox species with appropriate energy, which is dictated by the band edge potential of the photocatalyst. Since TiO2 inherently meets such requirements, it has been successfully used for various photocatalytic applications. However, TiO2 has limitations such as its wide-band gap, moderate charge separation eﬃciency, etc. To overcome such limitations, TiO2 has been both physically and chemically modiﬁed. Accordingly, herein we provided a glimpse on the various modiﬁcations that were performed on TiO2 towards enhancing its photocatalytic eﬃciencies. These modiﬁcations include morphological modiﬁcations, anionic-cationic doping, heterojunction formations, Z-scheme formations, plasmonic integrations, ferroelectric integrations, carbon-based materials integrations, 2D transition metal chalcogenide integrations, metal–organic framework integrations, and defects inducements in TiO2. We also have discussed the charge transfer mechanism that manifests in these various modiﬁed-TiO2 photocatalytic systems. TiO2 can be a prototype photocatalyst, which can be used to design new photocatalytic materials. The meticulous investigations on TiO2 for their photocatalytic mechanism can be better applied towards its eﬀective applications in photocatalysis. In this direction, the further improvement in TiO2 could be the establishment of techniques to intrinsically modify the TiO2 towards their photocatalytic enhancements. Such known techniques are the inducement of defective structures in TiO2 through self-doping, atoms in interstitial positions, oxygen-, and Ti-vacancies. For instance, instead of doping the N atoms into TiO2, the O atoms can be partially replaced by N atoms to form oxy-nitrides and so the oxy-phosphates, oxy-sulfur, oxy-carbon, etc., can be formed by partially replacing the O atoms with P, S, and C, respectively. These modiﬁcations may lead to the formation of entirely diﬀerent TiO2-based materials with possibly new crystal phase and structure and can exhibit enhanced photocatalytic eﬃciencies. Towards applications, TiO2 can be explored for new photocatalytic processes such as the production of H2/O2 from the atmospheric vapor, dark-photocatalysis, hydrogen storage, biodiesel productions, etc. TiO2 should be consistently explored towards further understanding of their photocatalytic mechanisms and ﬁnding new photocatalytic applications. Funding: This work was supported by the Natural Science and Engineering Research Council of Canada (NSERC) through the Collaborative Research and Development (CRD), Strategic Project (SP), and Discovery Grants (DG). 4. Summary and Outlook MS gratefully acknowledges the Department of Science and Technology, Govt. of India for the funding support through the DST-INSPIRE Faculty Award [DST/INSPIRE/04/2016/002227, 14-02-2017]. 36 Catalysts 2019, 9, 680 Conﬂicts of Interest: The authors declare no conﬂict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. References 1. Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238, 37–38. [CrossRef] 1. Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238, 37–38. [CrossRef] 2. Schrauzer, G.N.; Guth, T.D. Photocatalytic reactions. 1. Photolysis of water and photoreduction of nitrogen 2. Schrauzer, G.N.; Guth, T.D. Photocatalytic reactions. 1. Photolysis of water and photoreduction of nitro on titanium dioxide. J. Am. Chem. Soc. 1977, 99, 7189–7193. [CrossRef] 3. Kiwi, J.; Gratzel, M. Projection, size factors, and reaction dynamics of colloidal redox catalysts mediating light induced hydrogen evolution from water. J. Am. Chem. Soc. 1979, 101, 7214–7217. [CrossRef] 4. Kawai, T.; Sakata, T. Conversion of carbohydrate into hydrogen fuel by a photocatalytic process. Nature 1980, 286, 474–476. [CrossRef] 4. Kawai, T.; Sakata, T. Conversion of carbohydrate into hydrogen fuel by a photocatalytic process. Nature 1980, 286, 474–476. [CrossRef] 5. Sato, S.; White, J.M. Photoassisted water-gas shift reaction over platinized titanium dioxide catalysts. J. Am. Chem. Soc. 1980, 102, 7206–7210. [CrossRef] 6. Bagheri, S.; Youseﬁ, A.T.; Do, T.O. Photocatalytic pathway toward degradation of environmental pharmaceutical pollutants: Structure, kinetics and mechanism approach. Catal. Sci. Technol. 2017, 7, 4548–4569. [CrossRef] 7. Moser, J.; Gratzel, M. Light-induced electron transfer in colloidal semiconductor dispersions: Single vs. dielectronic reduction of acceptors by conduction-band electrons. J. Am. Chem. Soc. 1983, 105, 6547–6555. [CrossRef] 8. Bahnemann, D.; Henglein, A.; Lilie, J.; Spanhel, L. Flash photolysis observation of the absorption spectra of trapped positive holes and electrons in colloidal titanium dioxide. J. Phys. Chem. 1984, 88, 709–711. [CrossRef] 9. Nozik, A.J.; Williams, F.; Nenadovic, M.T.; Rajh, T.; Micic, O.I. Size quantization in small semiconductor particles. J. Phys. Chem. 1985, 89, 397–399. 10. Anpo, M.; Shima, T.; Kodama, S.; Kubokawa, Y. Photocatalytic hydrogenation of propyne with water on small-particle titania: Size quantization eﬀects and reaction intermediates. J. Phys. Chem. 1987, 91, 4305–4310. [CrossRef] 11. Hong, A.P.; Bahnemann, D.W.; Hoﬀmann, M.R. Cobalt (II) tetrasulfophthalocyanine on titanium dioxide. 2. Kinetics and mechanisms of the photocatalytic oxidation of aqueous sulfur dioxide. J. Phys. Chem. 1987, 91, 6245–6251. [CrossRef] 12. Zhang, J.L.; Minagawa, M.; Matsuoka, M.; Yamashita, H.; Anpo, M. Photocatalytic decomposition of NO on Ti-HMS mesoporous zeolite catalysts. Catal. Lett. 2000, 66, 241–243. [CrossRef] 13. Frank, S.N.; Bard, A.J. Heterogeneous photocatalytic oxidation of cyanide ion in aqueous solutions at titanium dioxide powder. J. Am. Chem. Soc. 1977, 99, 303–304. [CrossRef] 14. Halmann, M. References Photoelectrochemical reduction of aqueous carbon dioxide on p-type gallium phosphide in liquid junction solar cells. Nature 1978, 275, 115–116. [CrossRef] 14. Halmann, M. Photoelectrochemical reduction of aqueous carbon liquid junction solar cells. Nature 1978, 275, 115–116. [CrossRef] 15. Anpo, M.; Chiba, K.; Tomonari, M.; Coluccia, S.; Che, M.; Fox, M.A. Photocatalysis on Native and Platinum-Loaded TiO2 and ZnO Catalysts-Origin of Diﬀerent Reactivities on Wet and Dry Metal Oxides. Bull. Chem. Soc. Jpn. 1991, 64, 543–551. [CrossRef] 16. Domen, K.; Naito, S.; Soma, M.; Onishi, T.; Tamaru, K. Photocatalytic decomposition of water vapour on an NiO-SrTiO3 catalyst. J. Chem. Soc. Chem. Commun. 1980, 12, 543–544. [CrossRef] 17. Boonstra, A.H.; Mutsaers, C. Relation between the photoadsorption of oxygen and the number of hydroxyl groups on a titanium dioxide surface. J. Phys. Chem. 1975, 79, 1694–1698. [CrossRef] 18. Yun, C.; Anpo, M.; Kubokawa, Y. UV irradiation-induced ﬁssion of a C=C or C≡C bond adsorbed on TiO2. J. Chem. Soc. Chem. Commun. 1980, 609. [CrossRef] 19. Anpo, M.; Nakaya, H.; Kodama, S.; Kubokawa, Y.; Domen, K.; Onishi, T. Photocatalysis over binary metal oxides. Enhancement of the photocatalytic activity of titanium dioxide in titanium-silicon oxides. J. Phys. Chem. 1986, 90, 1633–1636. [CrossRef] 37 Catalysts 2019, 9, 680 20. Anpo, M.; Kawamura, T.; Kodama, S.; Maruya, K.; Onishi, T. Photocatalysis on titanium-aluminum binary metal oxides: Enhancement of the photocatalytic activity of titania species. J. Phys. Chem. 1988, 92, 438–440. [CrossRef] 21. Dohshi, S.; Takeuchi, M.; Anpo, M. Photoinduced superhydrophilic properties of Ti-B binary oxide thin ﬁlms and their photocatalytic reactivity for the decomposition of NO. J. Nanosci. Nanotechnol. 2001, 1, 337–342. [CrossRef] [PubMed] 22. Liu, Y.; Li, Z.; Green, M.; Just, M.; Li, Y.Y.; Chen, X. Titanium dioxide nanomaterials for photocatalysis. J. Phys. D: Appl. Phys. 2017, 50, 193003. [CrossRef] 23. Jenny, S.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Detlef, W. Bahnemann. Understanding TiO2 photocatalysis: Mechanisms and materials. Chem. Rev. 2014, 114, 9919–9986. 24. Hashimoto, K.; Irie, H.; Fujishima, A. TiO2 photocatalysis: A historical overview and future prospects. Jpn. J. Appl. Phys. 2005, 44, 8269. [CrossRef] 25. Coronado, D.R.; Gattorno, G.R.; Pesqueira, M.E.E.; Cab, C.; de Coss, R.; Oskam, G. Phase-pure TiO2 nanoparticles: Anatase, brookite and rutile. Nanotechnology 2008, 19, 145605. [CrossRef] [PubMed] 26. Zhang, J.; Zhou, P.; Liu, J.; Yu, J. New understanding of the diﬀerence of photocatalytic activity among anatase, rutile and brookite TiO2. Phys. Chem. Chem. Phys. References 2014, 16, 20382–20386. 27. Linsebigler, A.L.; Lu, G.; Yates, J.T., Jr. Photocatalysis on TiO2 surfaces: Principles, mechanisms, and selec results. Chem. Rev. 1995, 95, 735–758. [CrossRef] 28. Girish Kumar, S.; Gomathi Devi, L. Review on modiﬁed TiO2 photocatalysis under UV/visible light: Selected results and related mechanisms on interfacial charge carrier transfer dynamics. J. Phys. Chem. A. 2011, 115, 13211–13241. [CrossRef] 29. de Lasa, H.; Serrano, B.; Salaices, M. Establishing Photocatalytic Kinetic Rate Equations: Basic Principles and Parameters. In Photocatalytic Reaction Engineering; Springer: Boston, MA, USA, 2005. 30. Ravelli, D.; Dondi, D.; Fagnonia, M.; Albini, A. Photocatalysis. A multi-faceted concept for green chemistry. Chem. Soc. Rev. 2009, 38, 1999–2011. [CrossRef] 31. Nakata, K.; Fujishima, A. TiO2 photocatalysis: Design and applications. J. Photochem. Photobio. C Photochem. Rev. 2012, 13, 169–189. [CrossRef] 32. Parrino, F.; De Pasquale, C.; Palmisano, L. Inﬂuence of surface-related phenomena on mechanism, selectivity, and conversion of TiO2-induced photocatalytic reactions. ChemSusChem 2019, 12, 589–602. [CrossRef] [PubMed] 33. Fujishima, A.; Zhang, X.; Tryk, D.A. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2008, 63, 515–582. [CrossRef] 34. Pan, X.; Yang, M.Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective TiO2 with oxygen vacancies: Synthesis, properties and photocatalytic applications. Nanoscale 2013, 5, 3601–3614. [CrossRef] [PubMed] , ; g, Q ; , ; g, ; , J 2 yg y , p p and photocatalytic applications. Nanoscale 2013, 5, 3601–3614. [CrossRef] [PubMed] 35. Scanlon, D.O.; Dunnill, C.W.; Buckeridge, J.; Shevlin, S.A.; Logsdail, A.J.; Woodley, S.M.; Catlow, C.R.; Powell, M.J.; Palgrave, R.G.; Parkin, I.P.; et al. Band alignment of rutile and anatase TiO2. Nat. Mater. 2013, 12, 798–801. [CrossRef] [PubMed] Powell, M.J.; Palgrave, R.G.; Parkin, I.P.; et al. Band alignment of rutile and anatase TiO2. Nat. Mater. 2013, 12, 798–801. [CrossRef] [PubMed] 36. Guo, Q.; Zhou, C.; Ma, Z.; Ren, Z.; Fan, H.; Yang, X. Fundamental Processes in Surface Photocatalysis on TiO2. In Heterogeneous Photocatalysis; Green Chemistry and Sustainable Technology; Colmenares, J., Xu, Y.J., Eds.; Springer: Berlin/Heidelberg, Germany, 2016. 37. Yalavarthi, R.; Naldoni, A.; Kment, S.; Mascaretti, L.; Kmentová, H.; Tomanec, O.; Schmuki, P.; Zboril, R. Radiative and non-radiative recombination pathways in mixed-phase TiO2 nanotubes for PEC water-splitting. Catalysts 2019, 9, 204. [CrossRef] 38. Nguyen, C.C.; Vu, N.N.; Do, T.O. Eﬃcient hollow double-shell photocatalysts for the degradation of organic pollutants under visible light and in darkness. J. Mater. Chem. A 2016, 4, 4413–4419. References [CrossRef] g y , ; , ; , p y g g pollutants under visible light and in darkness. J. Mater. Chem. A 2016, 4, 4413–4419. [CrossRef] 39. Sakar, M.; Nguyen, C.C.; Vu, M.H.; Do, T.O. Materials and mechanisms of photo-assisted chemical reactions under light and dark: Can day-night photocatalysis be achieved? ChemSusChem 2018, 11, 809–820. [CrossRef] g y p under light and dark: Can day-night photocatalysis be achieved? ChemSusChem 2018, 11, 809–820. [CrossRef] 40. Dinh, C.T.; Pham, M.H.; Seo, Y.; Kleitz, F.; Do, T.O. Design of multicomponent photocatalysts for hydrogen production under visible light using water-soluble titanate nanodisks. Nanoscale 2014, 6, 4819–4829. [CrossRef] 41. Dinh, C.T.; Seo, Y.; Nguyen, T.D.; Kleitz, F.; Do, T.O. Controlled synthesis of titanate nanodisks as versatile building blocks for the design of hybrid nanostructures. Angew. Chem. Int. Ed. 2012, 51, 6608–6612. [CrossRef] 38 Catalysts 2019, 9, 680 42. Dinh, C.T.; Nguyen, T.D.; Kleitz, F.; Do, T.O. Shape-controlled synthesis of highly crystalline titania nanocrystals. ACS Nano 2009, 11, 3737–3743. [CrossRef] 43. Ng, J.; Pan, J.H.; Sun, D.D. Hierarchical assembly of anatase nanowhiskers and evaluation of their photocatalytic eﬃciency in comparison to various one-dimensional TiO2 nanostructures. J. Mater. Chem. 2011, 21, 11844–11853. [CrossRef] 44. Conceição, D.S.; Ferreira, D.P.; Graça, C.A.L.; Julio, M.F.; Ilharco, L.M.; Velosa, A.C.; Santos, P.F.; Vieira Ferreira, L.F. Photochemical and photocatalytic evaluation of 1D titanate/TiO2 based nanomaterials. Appl. Surf. Sci. 2017, 392, 418–429. [CrossRef] 45. Gordon, T.R.; Cargnello, M.; Paik, T.; Mangolini, F.; Weber, R.T.; Fornasiero, P.; Murray, C.B. Nonaqueous synthesis of TiO2 nanocrystals using TiF4 to engineer morphology, oxygen vacancy concentration, and photocatalytic activity. J. Am. Chem. Soc. 2012, 134, 6751–6761. [CrossRef] [PubMed] p y y 46. Ziarati, A.; Badiei, A.; Luque, R. Black Hollow TiO2 nanocubes: Advanced nanoarchitectures for eﬃcient visible light photocatalytic applications. Appl. Catal. B Environ. 2018, 238, 177–183. [CrossRef] 47. Dinh, C.T.; Nguyen, T.D.; Kleitz, F.; Do, T.O. A novel single-step route based on solvothermal technique to shape-controlled titanium dioxide nanocrystals. Can. J. Chem. Eng. 2012, 90, 8–17. [CrossRef] 48. Kang, X.; Song, X.Z.; Han, Y.; Cao, J.; Tan, Z. Defect-engineered TiO2 hollow spiny nanocubes for phenol degradation under visible light irradiation. Sci. Rep. 2018, 8, 5904. [CrossRef] [PubMed] 49. Zhou, X.; Liu, N.; Schmuki, P. Photocatalysis with TiO2 nanotubes: “Colorful” reactivity and designing site-speciﬁc photocatalytic centers into TiO2 nanotubes. ACS Catal. 2017, 7, 3210–3235. [CrossRef] 50. Zhao, Q.E.; Wen, W.; Xia, Y.; Wu, J.M. References Photocatalytic activity of TiO2 nanorods, nanowires and nanoﬂowers ﬁlled with TiO2 nanoparticles. Thin Solid Films 2018, 648, 103–107. [CrossRef] 51. Nguyen, C.C.; Vu, N.N.; Do, T.O. Recent advances in the development of sunlight-driven hollow structure photocatalysts and their applications. J. Mater. Chem. A 2015, 7, 8187–8208. [CrossRef] 52. Li, M.; Chen, Y.; Li, W.; Li, X.; Tian, H.; Wei, X.; Ren, Z.; Han, G. Ultrathin anatase TiO2 nanosheets f high-performance photocatalytic hydrogen production. Small 2017, 13, 1604115. [CrossRef] 53. Choi, S.K.; Kim, S.; Lim, S.K.; Park, H. Photocatalytic comparison of TiO2 Nanoparticles and Electrospun TiO2 nanoﬁbers: Eﬀects of mesoporosity and interparticle charge transfer. J. Phys. Chem. C 2010, 114, 16475–16480. [CrossRef] 54. Ribeiro, R.A.P.; de Lazaro, S.R.; de Oliveira, C.R. Band-Gap engineering for photocatalytic applications: Anionic and cationic doping of TiO2 anatase. Curr. Phys. Chem. 2016, 6, 22–27. [CrossRef] 55. Yalçın, Y.; Kılıç, M.; Çınar, Z. The role of non-metal doping in TiO2 photocatalysis. J. Adv. Oxid. Technol. 2016, 13, 281–296. [CrossRef] 56. Chen, X.; Burda, C. The electronic origin of the visible-light absorption properties of C-, N- and S-doped TiO2 nanomaterials. J. Am. Chem. Soc. 2008, 130, 5018–5019. [CrossRef] [PubMed] 57. Emy, M.S.; Sharifah, B.A.H. Eﬀect of band gap engineering in anionic-doped TiO2 photocatalyst. Appl. Surf. Sci. 2017, 391, 326–336. 58. Wang, H.; Lewis, J.P. Second-generation photocatalytic materials: Anion-doped TiO2. J. Phys. Condens. Matter 2006, 18, 421–434. [CrossRef] 59. Kuznetsov, V.N.; Serpone, N. on the origin of the spectral bands in the visible absorption spectra of visible-light-active TiO2 specimens analysis and assignments. J. Phys. Chem. C 2009, 113, 15110–15123. [CrossRef] 60. Serpone, N. Is the band gap of pristine TiO2 narrowed by anion- and cation-doping of titanium dioxide in second-generation photocatalysts? J. Phys. Chem. B 2006, 110, 24287–24293. [CrossRef] 61. Fan, W.Q.; Bai, H.Y.; Zhang, G.H.; Yan, Y.S.; Liu, C.B.; Shi, W.D. Titanium dioxide macroporous materials doped with iron: Synthesis and photo-catalytic properties. CrystEngComm 2014, 16, 116–122. [CrossRef] 62. Chang, S.M.; Liu, W.S. The roles of surface-doped metal ions (V, Mn, Fe, Cu, Ce, and W) in the interfacial behavior of TiO2 photocatalysts. Appl. Catal. B Environ. 2014, 156, 466–475. [CrossRef] 63. Hahn, R.; Stark, M.; Killian, M.S.; Schmuki, P. Photocatalytic properties of in situ doped TiO2-nanotubes grown by rapid breakdown anodization. Catal. Sci. Technol. 2013, 3, 1765–1770. [CrossRef] 64. Ishii, M.; Towlson, B.; Harako, S.; Zhao, X.W.; Komuro, S.; Hamilton, B. References Roles of electrons and holes in the luminescence of rare-earth-doped semiconductors. Electr. Commun. Jpn. 2013, 96, 1–7. [CrossRef] 39 Catalysts 2019, 9, 680 65. Tobaldi, D.M.; Pullar, R.C.; Gualtieri, A.F.; Seabra, M.P.; Labrincha, J.A. Sol-gel synthesis, characterisation and photocatalytic activity of pure, W-, Ag- and W/Ag codoped TiO2 nanopowders. Chem. Eng. J. 2013, 214, 364–375. [CrossRef] 66. de Lima, J.F.; Harunsani, M.H.; Martin, D.J.; Kong, D.; Dunne, P.W.; Gianolio, D.; Kashtiban, R.J.; Sloan, J.; Serra, O.A.; Tang, J.; et al. Control of chemical state of cerium in doped anatase TiO2 by solvothermal synthesis and its application in photocatalytic water reduction. J. Mater. Chem. A 2015, 3, 9890–9898. [CrossRef] 67. Li, H.; Liu, J.; Qian, J.; Li, Q.; Yang, J. Preparation of Bi-doped TiO2 nanoparticles and their visible li photocatalytic performance. Chin. J. Catal. 2014, 35, 1578–1589. [CrossRef] 68. Klaysri, R.; Wichaidit, S.; Tubchareon, T.; Nokjan, S.; Piticharoenphun, S.; Mekasuwandumrong, O.; Praserthdam, P. Impact of calcination atmospheres on the physiochemical and photocatalytic properties of nanocrystalline TiO2 and Si-doped TiO2. Ceram. Int. 2015, 41, 11409–11417. [CrossRef] 69. Zhao, Y.; Liu, J.; Shi, L.; Yuan, S.; Fang, J.; Wang, Z.; Zhang, M. Solvothermal preparation of Sn4+ doped anatase TiO2 nanocrystals from peroxo-metal-complex and their photocatalytic activity. Appl. Catal. B Environ. 2011, 103, 436–443. [CrossRef] 70. Carneiro, J.O.; Azevedo, S.; Fernandes, F.; Freitas, E.; Pereira, M.; Tavares, C.J.; Lanceros-Mendez, S.; Teixeira, V. Synthesis of iron-doped TiO2 nanoparticles by ball-milling process: The inﬂuence of process parameters on the structural, optical, magnetic, and photocatalytic properties. J. Mater. Sci. 2014, 49, 7476–7488. [CrossRef] 71. Makdee, A.; Unwiset, P.; Chanapattharapol, K.C.; Kidkhunthod, P. Eﬀects of Ce addition on the properties and photocatalytic activity of TiO2, investigated by X-ray absorption spectroscopy. Mater. Chem. Phys. 2018, 213, 431–443. [CrossRef] 72. Unwiset, P.; Makdee, A.; Chanapattharapol, K.C.; Kidkhunthod, P. Eﬀect of Cu addition on TiO2 surface properties and photocatalytic performance: X-ray absorption spectroscopy analysis. J. Phys. Chem. Solids 2018, 120, 231–240. [CrossRef] 73. Zhang, D.R.; Liu, H.L.; Han, S.Y.; Piao, W.X. Synthesis of Sc and V-doped TiO2 nanoparticles and photodegradation of rhodamine-B. J. Indus. Eng. Chem. 2013, 19, 1838–1844. [CrossRef] 74. Ould-Chikh, S.; Proux, O.; Afanasiev, P.; Khrouz, L.; Hedhili, M.N.; Anjum, D.H.; Harb, M.; Geantet, C.; Basset, J.; Puzenat, E. Photocatalysis with chromium-doped TiO2: Bulk and surface doping. ChemSusChem 2014, 7, 1361–1371. [CrossRef] [PubMed] j Basset, J.; Puzenat, E. Photocatalysis with chromium-doped TiO2: Bulk and surface doping. References Al-Maliki, F.J.; Al-Lamey, N.H. Synthesis of Tb-doped titanium dioxide nanostructures by sol–gel method f i t l h t t l i li ti J S l G l S i T h l 2017 81 276 283 [C R f] 86. Al-Maliki, F.J.; Al-Lamey, N.H. Synthesis of Tb-doped titanium dioxide nanostructures by sol–gel method for environmental photocatalysis applications. J. Sol-Gel Sci. Technol. 2017, 81, 276–283. [CrossRef] 86. Al-Maliki, F.J.; Al-Lamey, N.H. Synthesis of Tb-doped titanium dioxide nanostructures by sol–gel method 87. Singh, K.; Harish, S.; Kristya, A.P.; Shivani, V.; Archana, J.; Navaneethan, M.; Shimomura, M.; Hayakawa, Y. Erbium doped TiO2 interconnected mesoporous spheres as an eﬃcient visible light catalyst for photocatalytic applications. Appl. Surf. Sci. 2018, 449, 755–763. [CrossRef] 88. Jiang, X.; Li, C.; Liu, S.; Zhang, F.; You, F.; Yao, C. The synthesis and characterization of ytterbium-doped TiO2 hollow spheres with enhanced visible-light photocatalytic activity. RSC Adv. 2017, 7, 24598–24606. [CrossRef] 89. Shwetharani, R.; Sakar, M.; Chandan, H.R.; Geetha Balakrishna, R. Observation of simultaneous photocatalytic degradation and hydrogen evolution on the lanthanum modiﬁed TiO2 nanostructures. Mater. Lett. 2018, 218, 262–265. [CrossRef] 90. Ravishankar, T.N.; Nagaraju, G.; Dupont, J. Photocatalytic activity of Li-doped TiO2 nanoparticles: Synthesis via ionic liquid-assisted hydrothermal route. Mater. Res. Bull. 2016, 78, 103–111. [CrossRef] 91. Shivaraju, H.P.; Midhun, G.; Anil Kumar, K.M.; Pallavi, S.; Pallavi, N.; Behzad, S. Degradation of selected industrial dyes using Mg-doped TiO2 polyscales under natural sun light as an alternative driving energy. Appl. Water Sci. 2017, 7, 3937–3948. [CrossRef] 92. Fu, W.; Ding, S.; Wang, Y.; Wu, L.; Zhang, D.; Pan, Z.; Wang, R.; Zhang, Z.; Qiu, S. F, Ca co-doped T nanocrystals with enhanced photocatalytic activity. Dalton Trans. 2014, 43, 16160–16163. [CrossRef] [PubM 92. Fu, W.; Ding, S.; Wang, Y.; Wu, L.; Zhang, D.; Pan, Z.; Wang, R.; Zhang, Z.; Qiu, S. F, Ca co-doped TiO2 nanocrystals with enhanced photocatalytic activity. Dalton Trans. 2014, 43, 16160–16163. [CrossRef] [PubMed] 93. Xie, W.; Li, R.; Xu, Q. Enhanced photocatalytic activity of Se-doped TiO2 under visible light irradiation. Sci. Rep. 2018, 8, 8752. [CrossRef] [PubMed] 93. Xie, W.; Li, R.; Xu, Q. Enhanced photocatalytic activity of Se-doped TiO2 under visible light irradiation. Sci. Rep. 2018, 8, 8752. [CrossRef] [PubMed] 94. Nguyen, C.C.; Dinh, C.T.; Do, T.O. Hollow Rh/Sr-codoped TiO2 photocatalyst for eﬃcient sunlight-driven organic compound degradation. RSC Adv. 2017, 7, 3480–3487. [CrossRef] 95. References ChemSusChem 2014, 7, 1361–1371. [CrossRef] [PubMed] 75. Deng, Q.R.; Xia, X.H.; Guo, M.L.; Gao, Y.; Shao, G. Mn-doped TiO2 nanopowders with remarkable visible light photocatalytic activity. Mater. Lett. 2011, 65, 2051–2054. [CrossRef] 76. Sun, L.; Zhai, J.; Li, H.; Zhao, Y.; Yang, H.; Yu, H. Study of homologous elements: Fe, Co, and Ni dopant eﬀects on the photoreactivity of TiO2 nanosheets. ChemCatChem 2014, 6, 339–347. [CrossRef] J g y g p eﬀects on the photoreactivity of TiO2 nanosheets. ChemCatChem 2014, 6, 339–347. [CrossRef] 77. Karunakaran, C.; Abiramasundari, G.; Gomathisankar, P.; Manikandan, G.; Anandi, V. Cu-doped TiO2 nanoparticles for photocatalytic disinfection of bacteria under visible light. J. Colloid Interface Sci. 2010, 352, 68–74. [CrossRef] [PubMed] 78. Aware, D.V.; Jadhav, S.S. Synthesis, characterization and photocatalytic applications of Zn-doped TiO2 nanoparticles by sol–gel method. Appl. Nanosci. 2016, 6, 965–972. [CrossRef] 79. Jiang, X.; Gao, Y.; Li, C.; You, F.; Yao, J.; Ji, Y. Preparation of hollow yttrium-doped TiO2 microsph enhanced visible-light photocatalytic activity. Mater. Res. Express 2019, 6, 065510. [CrossRef] 80. Gao, B.; Lim, T.M.; Subagio, D.P.; Lim, T.T. Zr-doped TiO2 for enhanced photocatalytic degradation of bisphenol A. Appl. Catal. A Gen. 2010, 375, 107–115. [CrossRef] 81. Kou, Y.; Yang, J.; Li, B.; Fu, S. Solar photocatalytic activities of porous Nb-doped TiO2 microspheres by coupling with tungsten oxide. Mater. Res. Bull. 2015, 63, 105–111. [CrossRef] 82. Avilés-García, O.; Espino-Valencia, J.; Romero, R.; Rico-Cerda, J.L.; Arroyo-Albiter, M.; Natividad, R. W and Mo doped TiO2: Synthesis, characterization and photocatalytic activity. Fuel 2017, 198, 31–41. [CrossRef] 8 . v és Ga c a, O.; sp o Va e c a, J.; o e o, .; co Ce da, J. .; oyo b te , M.; Nat v dad, . W a d Mo doped TiO2: Synthesis, characterization and photocatalytic activity. Fuel 2017, 198, 31–41. [CrossRef] 83. Hao, H.Y.; He, C.X.; Tian, B.Z.; Zhang, J.L. Study of photocatalytic activity of Cd-doped mes nanocrystalline TiO2 prepared at low temperature. Res. Chem. Intermed. 2009, 35, 705. [CrossRef] 84. Chandan, H.R.; Sakar, M.; Ashesh, M.; Ravishankar, T.N.; Ramakrishnappa, T.; Sergio, R.T.; Geetha Balakrishna, R. Observation of oxo-bridged yttrium in TiO2 nanostructures and their enhanced photocatalytic hydrogen generation under UV/Visible light irradiations. Mater. Res. Bull. 2018, 104, 212–219. 40 Catalysts 2019, 9, 680 85. Štengl, V.; Bakardjieva, S.; Murafa, N. Preparation and photocatalytic activity of rare earth doped TiO2 nanoparticles. Mater. Chem. Phys. 2009, 114, 217–226. [CrossRef] p y 86. References Murashkina, A.A.; Murzin, P.D.; Rudakova, A.V.; Ryabchuk, V.K.; Emeline, A.V.; Detlef, W. Bahnemann. Inﬂuence of the dopant concentration on the photocatalytic activity: Al-doped TiO2. J. Phys. Chem. C 2015, 119, 24695–24703. [CrossRef] 96. Liqiang, J.; Honggang, F.; Baiqi, W.; Dejun, W.; Baifu, X.; Shudan, L.; Jiazhong, S. Eﬀects of Sn dopant on the photoinduced charge property and photocatalytic activity of TiO2 nanoparticles. Appl. Catal. B Environ. 2006, 62, 282–291. [CrossRef] 97. Nan, W.; Xing, L.; Yanling, Y.; Tingting, G.; Xiaoxuan, Z.; Siyang, J.; Tingting, Z.; Yi, S.; Zhiwei, Z. Enhanced photocatalytic degradation of sulfamethazine by Bi-doped TiO2 nano-composites supported by powdered activated carbon under visible light irradiation. Sep. Purif. Technol. 2019, 211, 673–683. g g g g y g J g g photocatalytic degradation of sulfamethazine by Bi-doped TiO2 nano-composites supported by powdered activated carbon under visible light irradiation. Sep. Purif. Technol. 2019, 211, 673–683. 98. Li, W. Inﬂuence of electronic structures of doped TiO2 on their photocatalysis. Phys. Status Solidi R 2015, 9, 10–27. [CrossRef] 99. Boulbar, E.L.; Millon, E.; Leborgne, C.B.; Cachoncinlle, C.; Hakim, B.; Ntsoenzok, E. Optical properties of rare earth-doped TiO2 anatase and rutile thin ﬁlms grown by pulsed-laser deposition. Thin Solid Films 2014, 553, 13–16. [CrossRef] 100. Shwetharani, R.; Sakar, M.; Fernando, C.A.N.; Binas, V.; Geetha Balakrishna, R. Recent advances and strategies applied to tailor energy levels, active sites and electron mobility in titania and its doped/composite analogues for hydrogen evolution in sunlight. Catal. Sci. Technol. 2019, 9, 12–46. [CrossRef] 101. Ma, Y.T.; Li, S.D. Photocatalytic activity of TiO2 nanoﬁbers with doped La prepared by electrospinning method. J. Chin. Chem. Soc. 2015, 62, 380–384. [CrossRef] 102. Borlaf, M.; Colomer, M.T.; de Andrés, A.; Cabello, F.; Serna, R.; Moreno, R. TiO2/Eu3+ thin ﬁlms with high photoluminescence emission prepared by electrophoretic deposition from nanoparticulate sols. Eur. J. Inorg. Chem. 2014, 30, 5152–5159. [CrossRef] 103. Du, J.; Li, B.; Huang, J.; Zhang, W.; Peng, H.; Zou, J. Hydrophilic and photocatalytic performances of lanthanum doped titanium dioxide thin ﬁlms. J. Rare Earth 2013, 31, 992–996. [CrossRef] 104. Maddila, S.; Oseghe, E.O.; Jonnalagadda, S.B. Photocatalyzed ozonation by Ce doped TiO2 catalyst degradation of pesticide Dicamba in water. J. Chem. Technol. Biotechnol. 2016, 91, 385–393. [CrossRef] 105. Choudhury, B.; Borah, B.; Choudhury, A. Extending photocatalytic activity of TiO2 nanoparticles to visible region of illumination by doping of cerium. Photochem. Photobiol. 2012, 88, 257–264. [CrossRef] [PubMed] 41 Catalysts 2019, 9, 680 106. References Vu, T.T.D.; Mighri, F.; Ajji, A.; Do, T.O. Synthesis of titanium dioxide/cadmium sulﬁde nanosphere particles for photocatalyst applications. Ind. Eng. Chem. Res. 2014, 53, 3888–3897. [CrossRef] 107. Wang, H.; Zhang, L.; Chen, Z.; Hu, J.; Li, S.; Wang, Z.; Liu, J.; Wang, X. Semiconductor heterojunction photocatalysts: Design, construction, and photocatalytic performances. Chem. Soc. Rev. 2014, 43, 5234–5244. [CrossRef] [PubMed] 108. Afroz, K.; Moniruddin, M.; Bakranov, N.; Kudaibergenov, S.; Nuraje, N. A heterojunction strategy to improve the visible light sensitive water splitting performance of photocatalytic materials. J. Mater. Chem. A 2018, 6, 21696–21718. [CrossRef] 109. Bessekhouad, Y.; Robert, D.; Weber, J.V. Photocatalytic activity of Cu2O/TiO2, Bi2O3/TiO2 and ZnMn2O4/TiO2 heterojunctions. Catal. Today 2005, 101, 315–321. [CrossRef] 110. Liu, J.; Yang, S.; Wu, W.; Tian, Q.; Cui, S.; Dai, Z.; Ren, F.; Xiao, X.; Jiang, C. 3D Flowerlike α-Fe2O3@TiO2 core-shell nanostructures: General synthesis and enhanced photocatalytic performance. ACS Sustain. Chem. Eng. 2015, 3, 2975–2984. [CrossRef] 111. Xia, Y.; Yin, L. Core-shell structured α-Fe2O3@TiO2 nanocomposites with improved photocatalytic activity in visible light region. Phys. Chem. Chem. Phys. 2013, 15, 18627–18634. [CrossRef] 112. Lin, Z.; Liu, P.; Yan, J.; Yang, G. Matching energy levels between TiO2 and α-Fe2O3 in a core-shell nanoparticle for the visible-light photocatalysis. J. Mater. Chem. A 2015, 3, 14853–14863. [CrossRef] 113. Moniz, S.J.A.; Shevlin, S.A.; Martin, D.J.; Guo, Z.X.; Tang, J. Visible-light driven heterojunction photocatalysts for water splitting—A critical review. Energy Environ. Sci. 2015, 8, 731–759. [CrossRef] 114. Ge, J.; Zhang, Y.; Heo, Y.J.; Park, S.J. Advanced design and synthesis of composite photocatalysts for the remediation of wastewater: A review. Catalysts 2019, 9, 122. [CrossRef] 115. 115. Danlian, H.; Sha, C.; Zeng, G.; Gong, X.; Zhou, C.; Cheng, M.; Xue, W.; Yan, X.; Li, J. Artiﬁcial Z-scheme photocatalytic system: What have been done and where to go? Coordination Chem. Rev. 2019, 385, 44–80. 116. Bard, A.J. Photoelectrochemistry and heterogeneous photo-catalysis at semiconductors. J. Photochem. 1979, 10, 59–75. [CrossRef] 117. Jiang, W.; Zong, X.; An, L.; Hua, S.; Miao, X.; Luan, S.; Wen, Y.; Tao, F.F.; Sun, Z. Consciously constructing heterojunction or direct z-scheme photocatalysts by regulating electron ﬂow direction. ACS Catal. 2018, 8, 2209–2217. [CrossRef] 118. Xu, Q.; Zhang, L.; Yu, J.; Wageh, S.; Al-Ghamdi, A.A.; Jaroniec, M. Direct Z-scheme photocatalysts: Principles, synthesis, and applications. Mater. Today 2018, 21, 1042–1063. [CrossRef] 119. Qi, K.; Cheng, B.; Yu, J.; Ho, W. A review on TiO2-based Z-scheme photocatalysts. Chin. J. Catal. References [CrossRef] 132. Wu, J.; Zhang, Z.; Liu, B.; Fang, Y.; Wang, L.; Dong, B. UV-Vis-NIR-driven plasmonic photocatalysts with dual-resonance modes for synergistically enhancing H2 generation. Sol. RRL 2018, 2, 1800039. [CrossRef] 133. Dinh, C.T.; Hoang, Y.; Kleitz, F.; Do, T.O. Three-dimensional ordered assembly of thin-shell Au/TiO2 hollow nanospheres for enhanced visible-light-driven photocatalysis. Angew. Chem. Int. Ed. 2014, 53, 6618–6623. [CrossRef] [PubMed] 134. He, Y.; Basnet, P.; Hunyadi Murph, S.E.; Zhao, Y. Ag nanoparticle embedded TiO2 composite nanorod arrays fabricated by oblique angle deposition: Toward plasmonic photocatalysis. ACS Appl. Mater. Interfaces 2013, 5, 11818–11827. [CrossRef] [PubMed] 135. Chen, Z.; Fang, L.; Dong, W.; Zheng, F.; Shena, M.; Wang, J. Inverse opal structured Ag/TiO2 plasmonic photocatalyst prepared by pulsed current deposition and its enhanced visible light photocatalytic activity. J. Mater. Chem. A 2014, 2, 824–832. [CrossRef] 136. Dinh, C.T.; Nguyen, T.D.; Kleitz, F.; Do, T.O. A new route to size and population control of silver clusters on colloidal TiO2 nanocrystals. ACS Appl. Mater. Interfaces 2011, 3, 2228–2234. [CrossRef] [PubMed] 137. Hirakawa, T.; Kamat, P.V. Charge separation and catalytic activity of Ag@TiO2 core-shell composite clusters under UV-irradiation. J. Am. Chem. Soc. 2005, 127, 3928–3934. [CrossRef] [PubMed] 138. Luth, H. Solid Surfaces, Interfaces, and Films; Springer: Berlin, Germany, 2001. 139. Kozlov, D.A.; Lebedev, V.A.; Polyakov, A.Y.; Khazova, K.M.; Garshev, A.V. The microstructure eﬀect on the Au/TiO2 and Ag/TiO2 nanocomposites photocatalytic activity. Nanosyst. Phys. Chem. Math. 2018, 9, 266–278. [CrossRef] 140. Leong, K.H.; Chu, H.Y.; Ibrahim, S.; Saravanan, P. Palladium nanoparticles anchored to anatase TiO2 for enhanced surface plasmon resonance-stimulated, visible-light-driven photocatalytic activity. Beilstein J. Nanotechnol. 2015, 6, 428–437. [CrossRef] 141. Tapin, B.; Epron, F.; Especel, C.; Ly, B.K.; Pinel, C.; Besson, M. Study of monometallic Pd/TiO2 catalysts for the hydrogenation of succinic acid in aqueous phase. ACS Catal. 2013, 3, 2327–2335. [CrossRef] 142. Keihan, A.H.; Rasoulnezhad, H.; Mohammadgholi, A.; Sajjadi, A.; Hosseinzadeh, R.; Farhadian, M.; Hosseinzadeh, G. Pd nanoparticle loaded TiO2 semiconductor for photocatalytic degradation of Paraoxon pesticide under visible-light irradiation. J. Mater. Sci. Mater. Electron. 2017, 28, 16718–16727. [CrossRef] 143. Pham, M.H.; Dinh, C.T.; Vuong, G.T.; Do, T.O. General route toward hollow photocatalyst with two cocatalysts separated on two surface sides for hydrogen generation. Phys. Chem. Chem. Phys. 2014, 16, 5937–5941. [CrossRef] 144. Galinska, A.; Walendziewski, J. Photocatalytic water splitting over Pt-TiO2 in the presence of sacriﬁcial reagents. Energy Fuels 2005, 19, 1143–1147. [CrossRef] 145. Nguyen, C.C.; Nguyen, D.T.; Do, T.O. References 2017, 38, 1936–1955. [CrossRef] 120. Li, H.; Tu, W.; Zhou, Y.; Zou, Z. Z-scheme photocatalytic systems for promoting photocatalytic performance: Recent progress and future challenges. Adv. Sci. 2016, 3, 1500389. [CrossRef] 121. Meng, A.; Zhu, B.; Zhong, B.; Zhang, L.; Cheng, B. Direct Z-scheme TiO2/CdS hierarchical photocatalyst for enhanced photocatalytic H2-production activity. Appl. Surface Sci. 2017, 422, 518–527. [CrossRef] 122. Dinh, C.T.; Pham, M.H.; Kleitz, F.; Do, T.O. Design of water-soluble CdS-titanate-nickel nanocomposites for photocatalytic hydrogen production under sunlight. J. Mater. Chem. A 2013, 1, 13308–13313. [CrossRef] 122. Dinh, C.T.; Pham, M.H.; Kleitz, F.; Do, T.O. Design of water soluble CdS titanate nickel nanocomposites for photocatalytic hydrogen production under sunlight. J. Mater. Chem. A 2013, 1, 13308–13313. [CrossRef] 123 Jo W K ; Natarajan T S Inﬂuence of TiO2 morphology on the photocatalytic eﬃciency of direct Z-scheme 123. Jo, W.K.; Natarajan, T.S. Inﬂuence of TiO2 morphology on the photocatalytic eﬃciency of direct Z-sche g-C3N4/TiO2 photocatalysts for isoniazid degradation. Appl. Surf. Sci. 2017, 422, 518–527. 124. Zhou, D.; Chen, Z.; Yang, Q.; Dong, X.; Zhang, J.; Qin, L. In-situ construction of all-solid-state Z-scheme g-C3N4/TiO2 nanotube arrays photocatalyst with enhanced visible-light-induced properties. Solar Energy Mater. Solar Cells 2016, 157, 399–405. [CrossRef] 125. Liao, W.; Murugananthan, M.; Zhang, Y. Synthesis of Z-scheme g-C3N4-Ti3+/TiO2 material: An eﬃcient visible light photoelectrocatalyst for degradation of phenol. Phys. Chem. Chem. Phys. 2015, 17, 8877–8884. [CrossRef] [PubMed] 126. Xu, F.; Zhang, L.; Cheng, B.; Yu, J. Direct Z-scheme TiO2/NiS core-shell hybrid nanoﬁbers with enhanced photocatalytic H2-production activity. ACS Sustain. Chem. Eng. 2018, 6, 12291–12298. [CrossRef] 127. Li, Q.; Xia, Y.; Yang, C.; Lv, K.; Lei, M.; Li, M. Building a direct Z-scheme heterojunction photocatalyst by ZnIn2S4 nanosheets and TiO2 hollowspheres for highly-eﬃcient artiﬁcial photosynthesis. Chem. Eng. J. 2018, 349, 287–296. [CrossRef] 128. Fu, J.; Cao, S.; Yu, J. Dual Z-scheme charge transfer in TiO2-Ag-Cu2O composite for enhanced photocatalytic hydrogen generation. J. Materiomics 2015, 1, 124–133. [CrossRef] 42 Catalysts 2019, 9, 680 129. Pan, L.; Zhang, J.; Jia, X.; Ma, Y.H.; Zhang, X.; Wang, L.; Zou, J.J. Highly eﬃcient Z-scheme WO3-x quantum dots/TiO2 for photocatalytic hydrogen generation. Chin. J. Catal. 2017, 38, 253–259. [CrossRef] 130. Zhang, X.; Chen, Y.L.; Liu, R.S.; Tsai, D.P. Plasmonic photocatalysis. Rep. Prog. Phys. 2013, 76, 046401. [CrossRef] [PubMed] 131. Wang, P.; Huang, B.; Dai, Y.; Whangbo, M.H. Plasmonic photocatalysts: Harvesting visible light with noble metal nanoparticles. Phys. Chem. Chem. Phys. 2012, 14, 9813–9825. References Compliments of conﬁnements: Substitution and dimension induced magnetic origin and band-bending mediated photocatalytic enhancements in 157. Sakar, M.; Balakumar, S.; Saravanan, P.; Bharathkumar, S. Particulates vs. ﬁbers: Dimension featured magnetic and visible light driven photocatalytic properties of Sc modiﬁed multiferroic bismuth ferrite nanostructures. Nanoscale 2016, 8, 1147–1160. [CrossRef] [PubMed] 158. Sakar, M.; Balakumar, S.; Bhaumik, I.; Gupta, P.K.; Jaisankar, S.N. Nanostructured Bi(1-x)Gd(x)FeO multiferroic photocatalyst on its sunlight driven photocatalytic activity. RSC Adv. 2014, 4, 16871–16878. 159. Li, R.; Li, Q.; Zong, L.; Wang, X.; Yang, J. BaTiO3/TiO2 heterostructure nanotube arrays for imp photoelectrochemical and photocatalytic activity. Electrochim. Acta 2013, 91, 30–35. [CrossRef] 160. Küçük, O.; Teber, S.; Kaya, I.C.; Akyildiz, H.; Kalem, V. Photocatalytic activity and dielectric properties of hydrothermally derived tetragonal BaTiO3 nanoparticles using TiO2 nanoﬁbers. J. Alloys Compd. 2018, 765, 82–91. [CrossRef] 161. Li, Q.; Li, R.; Zong, L.; He, J.; Wang, X.; Yang, J. Photoelectrochemical and photocatalytic properties of Ag-loaded BaTiO3/TiO2 heterostructure nanotube arrays. Int. J. Hydrog. Energy 2013, 38, 12977–12983. [CrossRef] 162. Zhang, Y.; Salvador, P.A.; Rohrer, G.S. Ferroelectric-enhanced photocatalysis with TiO2/BiFeO3. In Energy Technology 2014; Wang, C., de Bakker, J., Belt, C.K., Jha, A., Neelameggham, N.R., Pati, S., Prentice, L.H., Tranell, G., Brinkman, K.S., Eds.; The Minerals, Metals & Materials Society: Pittsburgh, PA, USA, 2014. 163. Li, S.; Lin, Y.H.; Zhang, B.P.; Li, J.F.; Nan, C.W. BiFeO3/TiO2 core-shell structured nanocomposites as visible-active photocatalysts and their optical response mechanism. J. Appl. Phys. 2009, 105, 054310. [CrossRef] 164. Liu, G.; Ma, L.; Yin, L.C.; Wan, G.; Zhu, H.; Zhen, C.; Yang, Y.; Liang, Y.; Tan, J.; Cheng, H.M. Selective chemical epitaxial growth of TiO2 islands on ferroelectric PbTiO3 crystals to boost photocatalytic activity. Joule 2018, 2, 1–13. [CrossRef] 165. Leary, R.; Westwood, A. Carbonaceous nanomaterials for the enhancement of TiO2 photocatalysis. Carbon 2011, 49, 741–772. [CrossRef] 166. Khalid, N.R.; Majid, A.; Tahir, M.B.; Niaz, N.A.; Khalid, S. Carbonaceous-TiO2 nanomaterials for photocatalytic degradation of pollutants: A review. Ceram. Int. 2017, 43, 14552–14571. [CrossRef] 167. Sakthivel, S.; Kisch, H. Daylight photocatalysis by carbon-modiﬁed titanium dioxide. Angew. Chem. Int. Ed. 2003, 42, 4908–4911. [CrossRef] [PubMed] 168. Yu, H.; Zhao, Y.; Zhou, C.; Shang, L.; Peng, Y.; Cao, Y.; Wu, L.Z.; Tunga, C.H.; Zhang, T. Carbon quantum dots/TiO2 composites for eﬃcient photocatalytic hydrogen evolution. J. Mater. Chem. A 2014, 2, 3344–3351. [CrossRef] 169. Yu, X.; Liu, J.; Yu, Y.; Zuo, S.; Li, S. References A novel route to synthesize C/Pt/TiO2 phase tunable anatase-rutile TiO2 for eﬃcient sunlight-driven photocatalytic applications. Appl. Catal. B Environ. 2018, 226, 46–52. [CrossRef] 146. Liu, K.; Litke, A.; Su, Y.; van Campenhout, B.G.; Pidko, E.A.; Hensen, E.J.M. Photocatalytic decarboxylation of lactic acid by Pt/TiO2. Chem. Commun. 2016, 52, 11634–11637. [CrossRef] [PubMed] 147. Fang, L.; You, L.; Liu, J. Ferroelectrics in Photocatalysis. In Ferroelectric Materials for Energy Applications; Huang, H., Scott, J.F., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2019. 147. Fang, L.; You, L.; Liu, J. Ferroelectrics in Photocatalysis. In Ferroelectric Materials for Energy Applications; Huang, H., Scott, J.F., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2019. 148. Cui, Y.; Briscoe, J.; Dunn, S. Eﬀect of ferroelectricity on solar-light-driven photocatalytic activity of BaTiO3-inﬂuence on the carrier separation and stern layer formation. Chem. Mater. 2013, 25, 4215–4223. [CrossRef] [ ] 149. Jones, P.M.; Dunn, S. Photo-reduction of silver salts on highly heterogeneous lead zirconate titanate. Nanotechnology 2007, 18, 185702. [CrossRef] 150. Park, S.; Lee, C.W.; Kang, M.G.; Kim, S.; Kim, H.J.; Kwon, J.E.; Park, S.Y.; Kang, C.Y.; Hong, K.S.; Nam, K.T. A ferroelectric photocatalyst for enhancing hydrogen evolution: Polarized particulate suspension. Phys. Chem. Chem. Phys. 2014, 16, 10408–10413. [CrossRef] [PubMed] 43 Catalysts 2019, 9, 680 151. Li, L.; Salvador, P.A.; Rohrer, G.S. Photocatalysts with internal electric ﬁelds. Nanoscale 2014, 6, 24–42. [CrossRef] [PubMed] 152. Huang, H.; Tu, S.; Du, X.; Zhang, Y. Ferroelectric spontaneous polarization steering charge carriers migration for promoting photocatalysis and molecular oxygen activation. J. Colloid Interface Sci. 2018, 509, 113–122. [CrossRef] 153. Cui, Y.; Briscoe, J.; Wang, Y.; Tarakina, N.V.; Dunn, S. Enhanced photocatalytic activity of heterostructured ferroelectric BaTiO3/α-Fe2O3 and the signiﬁcance of interface morphology control. ACS Appl. Mater. Interfaces 2017, 9, 24518–24526. [CrossRef] 154. Fu, Q.; Wang, X.; Li, C.; Sui, Y.; Han, Y.; Lv, Z.; Song, B.; Xu, P. Enhanced photocatalytic activity on polarized ferroelectric KNbO3. RSC Adv. 2016, 6, 108883–108887. [CrossRef] 155. Al-keisy, A.; Ren, L.; Cui, D.; Xu, Z.; Xu, X.; Su, X.; Hao, W.; Doua, S.X.; Du, Y. A ferroelectric photocatalyst Ag10Si4O13 with visible-light photooxidation properties. J. Mater. Chem. A 2016, 4, 10992–10999. [CrossRef] 156. Sakar, M.; Balakumar, S.; Saravanan, P.; Bharathkumar, S. Compliments of conﬁnements: Substitution and dimension induced magnetic origin and band-bending mediated photocatalytic enhancements in Bi1-xDyxFeO3 particulate and ﬁber nanostructures. Nanoscale 2015, 7, 10667–10679. [CrossRef] [PubMed] 156. Sakar, M.; Balakumar, S.; Saravanan, P.; Bharathkumar, S. References Preparation and visible light photocatalytic activity of carbon quantum dots/TiO2 nanosheet composites. Carbon 2014, 68, 718–724. [CrossRef] 169. Yu, X.; Liu, J.; Yu, Y.; Zuo, S.; Li, S. Preparation and visible light photocat dots/TiO2 nanosheet composites. Carbon 2014, 68, 718–724. [CrossRef] 170. Sathish Kumar, M.; Yamini Yasoda, K.; Kumaresan, D.; Kothurkar, N.K.; Batabyal, S.K. TiO2-carbon quantum dots (CQD) nanohybrid: Enhanced photocatalytic activity. Mater. Res. Express 2018, 5, 075502. [CrossRef] 171. Ren, Y.; Dong, Y.; Feng, Y.; Xu, J. Compositing two-dimensional materials with TiO2 for photocatalysis. Catalysts 2018, 8, 590. [CrossRef] 44 Catalysts 2019, 9, 680 172. Baca, M.; Kukułka, W.; Cendrowski, K.; Mijowska, E.; Kale´nczuk, R.J.; Zieli´nska, B. Graphitic carbon nitride and titanium dioxide modiﬁed with 1 D and 2 D carbon structures for photocatalysis. ChemSusChem 2019, 12, 612–620. [CrossRef] [PubMed] 173. Olowoyo, J.O.; Kumar, M.; Jain, S.L.; Babalola, J.O.; Vorontsov, A.V.; Kumar, U. Insights into reinforced photocatalytic activity of the CNT-TiO2 nanocomposite for co2 reduction and water splitting. J. Phys. Chem. C 2019, 123, 367–378. [CrossRef] 174. Shaban, M.; Ashraf, A.M.; Abukhadra, M.R. TiO2 nanoribbons/carbon nanotubes composite with enhanced photocatalytic activity; fabrication, characterization, and application. Sci. Rep. 2018, 8, 781. [CrossRef] 175. Li, X.; Lin, H.; Chen, X.; Niu, H.; Zhang, T.; Liua, J.; Qu, F. Fabrication of TiO2/porous carbo superior visible photocatalytic activity. New J. Chem. 2015, 39, 7863–7872. [CrossRef] , ; , ; , ; , ; g, ; , J ; Q , 2/p superior visible photocatalytic activity. New J. Chem. 2015, 39, 7863–7872. [CrossRef] 176. Kim, S.; Lim, S.K. Preparation of TiO2-embedded carbon nanoﬁbers and their photocatalytic activity in the oxidation of gaseous acetaldehyde. Appl. Catal. B Environ. 2008, 84, 16–20. [CrossRef] 177. Cao, L.; Sahu, S.; Anilkumar, P.; Bunker, C.E.; Xu, J.; Shiral Fernando, K.A.; Wang, P.; Guliants, E.A.; Tackett, K.N.; Sun, Y.P. Carbon nanoparticles as visible-light photocatalysts for eﬃcient CO2 conversion and beyond. J. Am. Chem. Soc. 2011, 133, 4754–4757. [CrossRef] [PubMed] 178. Zhong, J.; Chen, F.; Zhang, J. Carbon-deposited TiO2: Synthesis, characterization, and visible photocatalytic performance. J. Phys. Chem. C 2010, 114, 933–939. [CrossRef] 179. Zhang, W.D.; Bin Xu, B.; Jiang, L.C. Functional hybrid materials based on carbon nanotubes and metal oxides. J. Mater. Chem. 2010, 20, 6383–6391. [CrossRef] 180. Yu, J.; Wang, S.; Lowa, J.; Xiao, W. Enhanced photocatalytic performance of direct Z-scheme g-C3N4-TiO2 photocatalysts for the decomposition of formaldehyde in air. Phys. Chem. Chem. Phys. 2013, 15, 16883–16890. References [CrossRef] [PubMed] 181. Tan, L.L.; Ong, W.J.; Chai, S.P.; Mohamed, A.R. Reduced graphene oxide-TiO2 nanocomposite as a promising visible-light-active photocatalyst for the conversion of carbon dioxide. Nanoscale Res. Lett. 2013, 8, 465. [CrossRef] [PubMed] 182. Tan, L.; Chai, S.; Mohamed, A.R. Synthesis and applications of graphene-based TiO2 photocatalysts. ChemSusChem 2012, 5, 1868–1882. [CrossRef] [PubMed] 183. Nasr, M.; Balme, S.; Eid, C.; Habchi, R.; Miele, P.; Bechelany, M. Enhanced visible-light photocataly performance of electrospun rGO/TiO2 composite nanoﬁbers. J. Phys. Chem. C 2017, 121, 261–269. [CrossR 183. Nasr, M.; Balme, S.; Eid, C.; Habchi, R.; Miele, P.; Bechelany, M. Enhanced visible-light photocatalytic performance of electrospun rGO/TiO2 composite nanoﬁbers. J. Phys. Chem. C 2017, 121, 261–269. [CrossRef] 184. Pan, X.; Zhao, Y.; Liu, S.; Korzeniewski, C.L.; Wang, S.; Fan, Z. Comparing graphene-TiO2 nanowire and 184. Pan, X.; Zhao, Y.; Liu, S.; Korzeniewski, C.L.; Wang, S.; Fan, Z. Comparing graphene-TiO2 nanowire and graphene-TiO2 nanoparticle composite photocatalysts. ACS Appl. Mater. Interfaces 2012, 4, 3944–3950. [CrossRef] [PubMed] 185. Yu, H.; Xiao, P.; Tian, J.; Wang, F.; Yu, J. Phenylamine-functionalized rGO/TiO2 photocatalysts: Spatially separated adsorption sites and tunable photocatalytic selectivity. ACS Appl. Mater. Interfaces 2016, 8, 29470–29477. [CrossRef] [PubMed] 186. Luo, B.; Liu, G.; Wang, L. Recent advances in 2D materials for photocatalysis. Nanoscale 2016, 8, 6904–6920. [CrossRef] [PubMed] 187. Singh, A.K.; Mathew, K.; Zhuang, H.L.; Hennig, R.G. Computational screening of 2D materials for photocatalysis. J. Phys. Chem. Lett. 2015, 6, 1087–1098. [CrossRef] [PubMed] 188. Zhang, C.; Huang, H.; Ni, X.; Zhou, Y.; Kang, L.; Jiang, W.; Chen, H.; Zhong, J.; Liu, F. Band gap reduction in van der Waals layered 2D materials via a de-charge transfer mechanism. Nanoscale 2018, 10, 16759–16764. [CrossRef] [PubMed] 189. Ida, S.; Ishihara, T. Recent progress in two-dimensional oxide photocatalysts for water splitting. J. Phys. Chem. Lett. 2014, 5, 2533–2542. [CrossRef] [PubMed] 190. Nguyen, C.C.; Vu, N.N.; Chabot, S.; Kaliaguine, S.; Do, T.O. Role of CxNy-triazine in photocatalysis for eﬃcient hydrogen generation and organic pollutant degradation under solar light irradiation. Solar RRL 2017, 1, 1700012. [CrossRef] 191. Haque, F.; Daeneke, T.; Kalantar-zadeh, K.; Ou, J.Z. Two-dimensional transition metal oxide and chalcogenide-based photocatalysts. Nano-Micro Lett. 2018, 10, 23. [CrossRef] [PubMed] 192. Guo, L.; Yang, Z.; Marcus, K.; Li, Z.; Luo, B.; Zhou, L.; Wang, X.; Du, Y.; Yang, Y. MoS2/TiO2 heterostructures as nonmetal plasmonic photocatalysts for highly eﬃcient hydrogen evolution. Energy Environ. Sci. 2018, 11, 106–114. [CrossRef] 45 Catalysts 2019, 9, 680 193. References Sun, Y.; Lin, H.; Wang, C.; Wu, Q.; Wang, X.; Yang, M. Morphology-controlled synthesis of TiO2/MoS2 nanocomposites with enhanced visible-light photocatalytic activity. Inorg. Chem. Front. 2018, 5, 145–152. [CrossRef] 194. Dong, Y.; Chen, S.Y.; Lu, Y.; Xiao, Y.X.; Hu, J.; Wu, S.M.; Deng, Z.; Tian, G.; Chang, G.G.; Li, J.; et al. Hierarchical MoS2@TiO2 heterojunctions for enhanced photocatalytic performance and electrocatalytic hydrogen evolution. Chem. Asian J. 2018, 13, 1609. [CrossRef] 195. Zhang, W.; Xiao, X.; Zheng, L.; Wan, C. Fabrication of TiO2/MoS2 composite photocatalyst and its photocatalytic mechanism for degradation of methyl orange under visible light. Can. J. Chem. Eng. 2015, 93, 1594–1602. [CrossRef] 196. Zhao, F.; Rong, Y.; Wan, J.; Hu, Z.; Peng, Z.; Wang, B. MoS2 quantum dots@TiO2 nanotube composites with enhanced photoexcited charges separation and high-eﬃciency visible-light driven photocatalysis. Nanotechnology 2018, 29, 105403. [CrossRef] [PubMed] 197. Liu, X.; Xing, Z.; Zhang, H.; Wang, W.; Zhang, Y.; Li, Z.; Wu, X.; Yu, X.; Zhou, W. Fabrication of 3D mesoporous black TiO2/MoS2/TiO2 nanosheets for visible-light-driven photocatalysis. ChemSusChem 2016, 9, 1118–1124. [CrossRef] [PubMed] 198. Zhu, Y.; Ling, Q.; Liu, Y.; Wang, H.; Zhu, Y. Photocatalytic H2 evolution on MoS2-TiO2 catalysts synthesized via mechanochemistry. Phys. Chem. Chem. Phys. 2015, 17, 933–940. [CrossRef] [PubMed] 199. Liu, X.; Xing, Z.; Zhang, Y.; Li, Z.; Wu, X.; Tan, S.; Yu, X.; Zhu, Q.; Zhou, W. Fabrication of 3D ﬂower-like black N-TiO2-x@MoS2 for unprecedented-high visible-light-driven photocatalytic performance. Appl. Catal. B Environ. 2017, 201, 119–127. [CrossRef] 200. Zhang, J.; Zhang, L.; Ma, X.; Ji, Z. A study of constructing heterojunction between two-dimensional transition metal sulﬁdes (MoS2 and WS2) and (101),(001) faces of TiO2. Appl. Surf. Sci. 2018, 430, 424–437. [CrossRef] 201. Chu, H.; Lei, W.; Liua, X.; Lib, J.; Zheng, W.; Zhu, G.; Li, C.; Pan, L.; Sun, C. Synergetic eﬀect of TiO2 as co-catalyst for enhanced visible light photocatalytic reduction of Cr (VI) on MoSe2. Appl. Catal. A Gen. 2016, 521, 19–25. [CrossRef] 202. Shen, Y.; Ren, X.; Xu, G.; Huang, Z.; Qi, X. Mixed-dimensional TiO2 nanoparticles with MoSe2 nanosheets for photochemical hydrogen generation. J. Mater. Sci. Mater. Electron. 2017, 28, 2023–2028. [CrossRef] 203. Zheng, X.; Yang, L.; Li, Y.; Yang, L.; Luo, S. Direct Z-scheme MoSe2 decorating TiO2 nanotube arr photocatalyst for water decontamination. Electrochim. Acta 2019, 298, 663–669. [CrossRef] 204. Cao, S.; Liu, T.; Hussain, S.; Zeng, W.; Peng, X.; Pan, F. Hydrothermal synthesis, characterization and optical absorption property of nanoscale WS2/TiO2 composites. References Phys. E Low-Dimens. Syst. Nanostruct. 2015, 68, 171–175. [CrossRef] 205. Ren, X.; Qiao, H.; Huang, Z.; Tang, P.; Liu, S.; Luo, S.; Yao, H.; Qi, X.; Zhong, J. Investigating the photocurrent generation and optoelectronic responsivity of WS2-TiO2 heterostructure. Optics Commun. 2018, 406, 118–123. [CrossRef] 206. Cho, E.C.; Chang-Jian, C.W.; Zheng, J.H.; Huang, J.H.; Lee, K.C.; Ho, B.C.; Hsiao, Y.S. Microwave-assisted synthesis of TiO2/WS2 heterojunctions with enhanced photocatalytic activity. J. Taiwan Ins. Chem. Eng. 2018, 91, 489–498. [CrossRef] 207. Wu, Y.; Liu, Z.; Li, Y.; Chen, J.; Zhu, X.; Na, P. Construction of 2D-2D TiO2 nanosheet/layered WS2 heterojunctions with enhanced visible-light-responsive photocatalytic activity. Chin. J. Catal. 2019, 40, 60–69. [CrossRef] 208. Wu, Y.; Liu, Z.; Li, Y.; Chen, J.; Zhu, X.; Na, P. WS2 nanodots-modiﬁed TiO2 nanotubes to enhance visible-light photocatalytic activity. Mater. Lett. 2019, 240, 47–50. [CrossRef] 209. Ho, W.; Yu, J.C.; Lin, J.; Yu, J.; Li, P. Preparation and photocatalytic behavior of MoS2 and WS2 nanocluster sensitized TiO2. Langmuir 2004, 20, 5865–5869. [CrossRef] [PubMed] 210. Dowla Biswas, M.R.U.; Ali, A.; Cho, K.Y.; Oh, W.C. Novel synthesis of WSe2-Graphene-TiO2 ternary nanocomposite via ultrasonic technics for high photocatalytic reduction of CO2 into CH3OH. Ultrason. Sonochem. 2018, 42, 738–746. [CrossRef] 211. Nasalevich, M.A.; van der Veen, M.; Kapteijn, F.; Gascon, J. Metal-organic frameworks as heterogeneous photocatalysts: Advantages and challenges. CrystEngComm 2014, 16, 4919–4926. [CrossRef] 212. Zhang, T.; Lin, W. Metal-organic frameworks for artiﬁcial photosynthesis and photocatalysis. Chem. Soc. Rev. 2014, 43, 5982–5993. [CrossRef] 46 Catalysts 2019, 9, 680 213. Li, Y.; Xu, H.; Ouyang, S.; Ye, J. Metal-organic frameworks for photocatalysis. Phys. Chem. Chem. Phys. 2016, 18, 7563–7572. [CrossRef] 214. Dhakshinamoorthy, A.; Li, Z.; Garcia, H. Catalysis and photocatalysis by metal organic frameworks. Chem. Soc. Rev. 2018, 47, 8134–8172. [CrossRef] 215. Xiao, J.D.; Jiang, H.L. Metal-organic frameworks for photocatalysis and photothermal catalysis. Acc. Chem. Res. 2019, 52, 356–366. [CrossRef] 216. Qiu, J.; Zhang, X.; Feng, Y.; Zhang, X.; Wang, H.; Yao, J. Modiﬁed metal-organic frameworks as photocatalysts. Appl. Catal. B Environ. 2018, 231, 317–342. [CrossRef] 217. Yao, P.; Liu, H.; Wang, D.; Chen, J.; Li, G.; An, T. Enhanced visible-light photocatalytic activity to volatile organic compounds degradation and deactivation resistance mechanism of titania conﬁned inside a metal-organic framework. J. Colloid Interface Sci. 2018, 522, 174–182. [CrossRef] [PubMed] organic compounds degradation and deactivation resistance mechanism of titania conﬁned inside a metal-organic framework. J. Colloid Interface Sci. 2018, 522, 174–182. [CrossRef] [PubMed] 218. References Crake, A.; Christoforidis, K.C.; Kaﬁzas, A.; Zafeiratos, S.; Petit, C. CO2 capture and photocatalytic reduction using bifunctional TiO2/MOF nanocomposites under UV-vis irradiation. Appl. Catal. B Environ. 2017, 210, 131–140. [CrossRef] 219. Ling, L.; Wang, Y.; Zhang, W.; Ge, Z.; Duan, W.; Liu, B. Preparation of a novel ternary composite of TiO2/UiO-66-NH2/Graphene Oxide with enhanced photocatalytic activities. Catal. Lett. 2018, 148, 1978–1984. [CrossRef] 220. Cardoso, J.C.; Stulp, S.; de Brito, J.F.; Flor, J.B.S.; Frem, R.C.G.; Zanoni, M.V.B. MOFs based on ZIF-8 deposited on TiO2 nanotubes increase the surface adsorption of CO2 and its photoelectrocatalytic reduction to alcohols in aqueous media. Appl. Catal. B Environ. 2018, 225, 563–573. [CrossRef] 221. Wang, M.; Wang, D.; Li, Z. Self-assembly of CPO-27-Mg/TiO2 nanocomposite with enhanced performance for photocatalytic CO2 reduction. Appl. Catal. B Environ. 2016, 183, 47–52. [CrossRef] 222. Zhang, B.; Zhang, J.; Tan, X.; Shao, D.; Shi, J.; Zheng, L.; Zhang, J.; Yang, G.; Han, B. MIL-125-NH2@TiO2 Core-shell particles produced by a post-solvothermal route for high-performance photocatalytic H2 production. ACS Appl. Mater. Interfaces 2018, 10, 16418–16423. [CrossRef] 223. Zhao, C.W.; Li, Y.A.; Wang, X.R.; Chen, G.J.; Liu, Q.K.; Ma, J.P.; Dong, Y.B. Fabrication of Cd (II)-MOF-based ternary photocatalytic composite materials for H2 production via gel-to-crystal approach. Chem. Commun. 2015, 51, 15906–15909. [CrossRef] 224. Li, X.; Pi, Y.; Hou, Q.; Yu, H.; Li, Z.; Li, Y.; Xiao, J. Amorphous TiO2@NH2-MIL-125(Ti) homologous MOF encapsulated heterostructures with enhanced photocatalytic activity. Chem. Commun. 2018, 54, 1917–1920. [CrossRef] 225. Xue, C.; Zhang, F.; Chang, Q.; Dong, Y.; Wang, Y.; Hu, S.; Yang, J. MIL-125 and NH2-MIL-125 modiﬁed TiO2 nanotube array as eﬃcient photocatalysts for pollute degradation. Chem. Lett. 2018, 47, 711–714. [CrossRef] 225. Xue, C.; Zhang, F.; Chang, Q.; Dong, Y.; Wang, Y.; Hu, S.; Yang, J. MIL 125 and NH2 MIL 125 modiﬁed TiO2 nanotube array as eﬃcient photocatalysts for pollute degradation. Chem. Lett. 2018, 47, 711–714. [CrossRef] 226. Xing, M.; Fang, W.; Nasir, M.; Ma, Y.; Zhang, J.; Anpo, M. Self-doped Ti3+-enhanced TiO2 nanoparticles with a high-performance photocatalysis. J. Catal. 2013, 297, 236–243. [CrossRef] 226. Xing, M.; Fang, W.; Nasir, M.; Ma, Y.; Zhang, J.; Anpo, M. Self-doped Ti3+-enhanced TiO2 nanoparticles with a high-performance photocatalysis. J. Catal. 2013, 297, 236–243. [CrossRef] 3 227. Zhang, R.; Yang, Y.; Leng, S.; Wang, Q. Photocatalytic activity of Ti3+ self-doped dark TiO2 ultraﬁne nanorods, grey SiO2 nanotwin crystalline, and their composite under visible light. Mater. Res. Express 2018, 5, 045044. References [CrossRef] 228. Liu, X.; Gao, S.; Xu, H.; Lou, Z.; Wang, W.; Huang, B.; Dai, Y. Green synthetic approach for Ti3+ self-doped TiO2−x nanoparticles with eﬃcient visible light photocatalytic activity. Nanoscale 2013, 5, 1870–1875. [CrossRef] [PubMed] 229. Hao, W.; Li, X.; Qin, L.; Han, S.; Kang, S.Z. Facile preparation of Ti3+ self-doped TiO2 nanoparticles and their dramatic visible photocatalytic activity for fast restoration of highly concentrated Cr (VI) eﬄuent. Catal. Sci. Technol. 2019, 9, 2523–2531. [CrossRef] 230. Zheng, Z.; Huang, B.; Meng, X.; Wang, J.; Wang, S.; Lou, Z.; Wang, Z.; Qin, X.; Zhang, X.; Dai, Y. Metallic zinc- assisted synthesis of Ti3+ self-doped TiO2 with tunable phase composition and visible-light photocatalytic activity. Chem. Commun. 2013, 49, 868–870. [CrossRef] [PubMed] 231. Naldoni, A.; Allieta, M.; Santangelo, S.; Marelli, M.; Fabbri, F.; Cappelli, S.; Bianchi, C.L.; Psaro, R.; Santo, V.D. Eﬀect of nature and location of defects on bandgap narrowing in black TiO2 nanoparticles. J. Am. Chem. Soc. 2012, 134, 7600–7603. [CrossRef] [PubMed] 232. Ullattil, S.G.; Narendranath, S.B.; Pillai, S.C.; Periyat, P. Black TiO2 nanomaterials: A review of recent advances. Chem. Eng. J. 2018, 343, 708–736. [CrossRef] 47 47 Catalysts 2019, 9, 680 233. Wang, B.; Shen, S.; Mao, S.S. Black TiO2 for solar hydrogen conversion. J. Materiomics 2017, 3, 96–111. [CrossRef] 234. Chen, X.; Liu, L.; Huang, F. Black titanium dioxide (TiO2) nanomaterials. Chem. Soc. Rev. 2015, 44, 1861–1885. [CrossRef] 235. Chen, X.B.; Liu, L.; Yu, P.Y.; Mao, S.S. Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals. Science 2011, 331, 746–750. [CrossRef] 235. Chen, X.B.; Liu, L.; Yu, P.Y.; Mao, S.S. Increasing solar absorption for ph titanium dioxide nanocrystals. Science 2011, 331, 746–750. [CrossRef] 236. Zhu, Q.; Peng, Y.; Lin, L.; Fan, C.M.; Gao, G.Q.; Wang, R.X.; Xu, A.W. Stable blue TiO2-x nanoparticles for eﬃcient visible light photocatalysts. J. Mater. Chem. A 2014, 2, 4429–4437. [CrossRef] 237. Wu, Q.; Huang, F.; Zhao, M.; Xu, J.; Zhou, J.; Wang, Y. Ultra-small yellow defective TiO2 nanoparticles for co-catalyst free photocatalytic hydrogen production. Nano Energy 2016, 24, 63–71. [CrossRef] 238. Liu, G.; Yin, L.C.; Wang, J.; Niu, P.; Zhen, C.; Xie, Y.; Cheng, H.M. A red a energy conversion. Energy Environ. Sci. 2012, 5, 9603–9610. [CrossRef] 238. Liu, G.; Yin, L.C.; Wang, J.; Niu, P.; Zhen, C.; Xie, Y.; Cheng, H.M. A red anatase TiO2 photocatalyst for solar energy conversion. Energy Environ. Sci. 2012, 5, 9603–9610. [CrossRef] 239. Songbo Wang 1,2, Feifan Wang 1, Zhiming Su 1, Xiaoning Wang 1, Yicheng Han 1, Lei Zhang 1, Jun Xiang 1, Wei Du 1 and Na Tang 1,* 1 Tianjin Key Laboratory of Brine Chemical Engineering and Resource Eco-utilization, College of Chemical Engineering and Materials Science, Tianjin University of Science & Technology, Tianjin 300457, China; wangsongbo@tust.edu.cn (S.W.); wangfeifan4500@sina.com (F.W.); suzhiming0321@sina.com (Z.S.); wangxiaoning@sina.com (X.W.); hanyicheng1998@sina.com (Y.H.); Leizhang@tust.edu.cn (L.Z.); jxiang@tust.edu.cn (J.X.); duwei@tust.edu.cn (W.D.) 2 Tianjin Key Laboratory of Marine Resources and Chemistry, Tianjin University of Science & Technology, Tianjin 300457, China 2 Tianjin Key Laboratory of Marine Resources and Chemistry, Tianjin University of Science & Technology, Tianjin 300457, China j * Correspondence: tjtangna@tust.edu.cn; Tel./Fax: +86-22-60602745 Received: 4 April 2019; Accepted: 6 May 2019; Published: 10 May 2019 Abstract: Photocatalytic technology has been considered to be an ideal approach to solve the energy and environmental crises, and TiO2 is regarded as the most promising photocatalyst. Compared with bare TiO2, TiO2 based p-n heterojunction exhibits a much better performance in charge separation, light absorption and photocatalytic activity. Herein, we developed an eﬃcient method to prepare p-type TiO2 quantum dots (QDs) and decorated graphitic carbonitrile (g-C3N4) nanocomposites, while the composition and structure of the TiO2@g-C3N4 were analyzed by X-ray diﬀraction, Fourier transform infrared spectroscopy, thermogravimetric analysis, transmission electron microscopy, X-ray photoelectron spectroscopy and UV-visible diﬀuse reﬂectance spectroscopy characterizations. The characterization results reveal the surface decorated TiO2 quantum dots is decomposed by titanium glycerolate, which exhibits p-type conductivity. The presence of p-n heterojunction over interface is conﬁrmed, and photoluminescence results indicate a better performance in transfer and separation of photo-generated charge carriers than pure semiconductors and type-II heterojunction. Moreover, the synergy of p-n heterojunction over interface, strong interface interaction, and quantum-size eﬀect signiﬁcantly contributes to the promoted performance of TiO2 QDs@g-C3N4 composites. As a result, the as-fabricated TiO2 QDs@g-C3N4 composite with a p/n mass ratio of 0.15 exhibits improved photo-reactivity of 4.3-fold and 5.4-fold compared to pure g-C3N4 in degradation of organic pollutant under full solar spectrum and visible light irradiation, respectively. Keywords: p-n heterojunction; g-C3N4; TiO2; charge separation; photocatalysis Catalysts 2019, 9, 439; doi:10.3390/catal9050439 www.mdpi.com/journal/catalysts catalysts catalysts References Ren, R.; Wen, Z.; Cui, S.; Hou, Y.; Guo, X.; Chen, J. Controllable synthesis and tunable photocatalytic properties of Ti3+-doped TiO2. Sci. Rep. 2015, 5, 10714. [CrossRef] [PubMed] 240. Fan, C.; Chen, C.; Wang, J.; Fu, X.; Ren, Z.; Qian, G.; Wang, Z. Enhanced photocatalytic activity of hydroxylated and N-doped anatase derived from amorphous hydrate. J. Mater. Chem. A 2014, 2, 16242–16249. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 48 48 Controllable Fabrication of Heterogeneous p-TiO2 QDs@g-C3N4 p-n Junction for Eﬃcient Photocatalysis Songbo Wang 1,2, Feifan Wang 1, Zhiming Su 1, Xiaoning Wang 1, Yicheng Han 1, Lei Zhang 1, Jun Xiang 1, Wei Du 1 and Na Tang 1,* 1. Introduction With the increasing concerns regarding the global environmental and energy-related crises over the past decades, photocatalytic technology has been considered to be an eﬀective approach since the foundation of Fujishima-Honda eﬀect in 1972 [1,2]. The key for an eﬃcient photocatalytic process lies in the design and construction of highly active photocatalyst, which requires a wide light absorption edge, fast transfer and separation of photo-generated charge carriers, and quick surface redox reaction [3–7]. TiO2 is regarded as the most promising photocatalyst due to the advantages such as earth abundance, low price, excellent thermal and chemical stability, and being environmentally friendly [8,9]. Unfortunately, the broad band gap (i.e., 3.2 eV for anatase) means TiO2 can only be excited by ultraviolet (UV) light, which is less than 5% in the solar spectrum. Meanwhile, the fast charge recombination rate in single TiO2 also results in a low quantum eﬃciency. Therefore, it is an Catalysts 2019, 9, 439; doi:10.3390/catal9050439 49 Catalysts 2019, 9, 439 urgently necessary to promote the light absorption and charge separation eﬃciency of TiO2 to meet the requirements of industrial applications [10,11]. Recently, many researchers have focused on the modification of TiO2 towards an improved photocatalytic performance, including for: morphology modulation [12–14], metal or nonmetal doping [15–18], defect engineering [19–22], and fabrication of hetero/homojunction [2,23–27]. In our previous reports, we found that the introduction of titanium vacancies into TiO2 will widen the valence band, which controls the mobility of holes inherently and thus increase charge separation efficiency. Moreover, the introduction of metal vacancies can alter TiO2 from an n-type semiconductor to a p-type semiconductor [28]. Pan et-al. modified p-type TiO2 with n-type TiO2 quantum dots (QDs) to construct p-n homojunction, and the resulted p-n homojunction exhibits significantly high photo-activity compared with pure type TiO2, which is attributed to the formation of large electronic field over the interface [29]. Moreover, besides the influence of p-n homojunction, quantum sized TiO2 can also enhance the charge separation due to the quantum size effect [13,30–32]. However, in these reports, both the titanium defected TiO2 and TiO2 QDs still absorb only UV light, which limits their applications under sunlight. Graphic carbon nitride (g-C3N4) is a promising metal-free photocatalyst in the field of H2 production, organic pollutant degradation, CO2 reduction, and artificial photosynthesize due to the suitable band structure (2.7~2.8 eV), excellent chemical and thermal stability [33–37]. 1. Introduction Usually, g-C3N4 is synthesized by thermal condensation of melamine, urea or other triazine derivatives, but the resultant product exhibits an irregular 2D aggregation structure, leading to a lower surface area and electrical conductivity, as well as an increased charge carrier recombination [38,39]. Both theoretical and experimental results suggest that nanosheet structured g-C3N4 will exhibit a larger surface area and lower charge transfer resistance, so the exfoliation of bulk g-C3N4 into nanosheets has been regarded as an effective approach to promote the photocatalytic performance of g-C3N4 [40,41]. Wang et-al. applied a liquid exfoliation method on bulk g-C3N4 to prepare layer g-C3N4, and the charge transfer resistance decreased by 75% according to the electrochemical impedance spectroscopy (EIS) results, indicating a lower charge recombination [42]. Cheng et-al. conducted an exfoliation treatment on bulk g-C3N4 using a simple thermal oxidation etch method, and g-C3N4 nanosheets with layer thickness of 2 nm and surface area of 306 m2/g were obtained. Compared with bulk g-C3N4, the charge carrier life in the as-prepared g-C3N4 nanosheets is prolonged because of the quantum confinement effect [43]. Li and coworkers also reported the synthesizing of 6–9 atomic thick g-C3N4 nanosheets by a thermal exfoliation approach, the photocatalytic degradation rate is 2.9 times higher than that of bulk g-C3N4 [44]. Therefore, the exfoliation of bulk g-C3N4 to nanosheets can effectively increase the surface area, and thus improve charge transfer and separation. Construction of heterojunction is an effective approach to improve the photocatalytic performance by combining both the advantage of two semiconductors, and the electronic field will enhance the charge transfer and separation across the interface. Type II heterojunction with staggered band alignment is the most widely studied structure. Jiang et-al. placed 5.5 nm sized TiO2 on g-C3N4 nanosheets. Due to the formation of type-II heterojunction, the resultant composites exhibited a Rhodamine B (RhB) degradation rate of 2.5-fold compared to pure g-C3N4 [25]. However, this is due to the fact that the work function of the same types semiconductor is closed to each other, which limits the driving force in type-II heterojunction. Instead, with different conductivity types semiconductor contacts, p-n heterojunction will be formed and the difference of the work function is large enough to realize a more efficient charge carrier separation [26,29]. For example, Wang et-al. reported that Cu2O/TiO2 p-n junction exhibits a much better photoelectrochemical activity than that of pure TiO2 and pure Cu2O [27]. 1. Introduction In this work, based on the improvement of our previous work [26,28,29], nanosized titanium glycerolate (TiGly), precursors of p-type TiO2 quantum dots, were successfully synthesized and in-situ deposited on the surface of g-C3N4, TiO2@g-C3N4 nanosheet p-n heterojunctions were then obtained after calcination in air. This p-n junction can achieve the function of killing three birds with one stone: p-type TiO2 QDs for promoted charge separation, g-C3N4 nanosheets for lower charge resistance and p-n junction for enhanced charge transfer over interface. Therefore, compared with pure g-C3N4 and TiO2/g-C3N4 type-II heterojunction, TiO2 QDs@g-C3N4 p-n heterojunctions exhibit promoted 50 Catalysts 2019, 9, 439 electron-hole separation eﬃciency and excellent photocatalytic performance in degradation of organic pollutant and hydrogen evolution. 2. Results and Discussion The synthesizing process of bulk g-C3N4 and TiO2@g-C3N4 p-n heterojunctions were diagrammatically presented in Scheme 1. Consistent with the literature reported, melamine was calcined in air at 550 ◦C and then bulk g-C3N4 were obtained. Previously, we have demonstrated that titanium glycerolate (TiGly) is the precursor of p-type TiO2, the organic groups will be gradually removed after calcination, and titanium vacancies (VTi) will be introduced into TiO2 during the conjunction process of the remaining Ti-O-Ti parallel lattice chains. Herein, in order to fabricate TiO2@g-C3N4 p-n heterojunction, bulk g-C3N4 is synthesized firstly, and then TiGly nanoparticles were in-situ deposited on the surface of g-C3N4 and labeled as TGC-x according to the added g-C3N4. After calcination in air, g-C3N4 was exfoliated and TiGly nanoparticles were decomposed to p-type TiO2 QDs (labeled as PTC-x). The sharp peak at about 27.4◦in the X-ray diffraction (XRD) pattern of TGC-x (Figure 1a) can be attributed to the periodic accumulation of layers of conjugated aromatic systems in bulk g-C3N4 [45]. In the sample TGC-20 and TGC-40, a weak peak appearing at about 10.2◦, which is the characteristic peaks corresponding to TiGly [28], indicating the presence of TiGly nanoparticles. As for TGC-60, the content of TiGly is too low to be detected. Scheme 1. Schematic fabrication procedures of g-C3N4 and TiO2@g-C3N4 p-n heterojunctions. Figure 1. (a) XRD pattern of bulk g-C3N4, bulk and nanoparticle TiGly and TGC-x before calcination; (b) XRD pattern of g-C3N4 nanosheets, bulk and nanoparticle p-TiO2 and PTC-x after calcination. Scheme 1. Schematic fabrication procedures of g-C3N4 and TiO2@g-C3N4 p-n heterojunctions. Scheme 1. Schematic fabrication procedures of g-C3N4 and TiO2@g-C3N4 p-n heterojunctions. matic fabrication procedures of g-C3N4 and TiO2@g-C3N4 p-n heterojunctions. Figure 1. (a) XRD pattern of bulk g-C3N4, bulk and nanoparticle TiGly and TGC-x before calcination; (b) XRD pattern of g-C3N4 nanosheets, bulk and nanoparticle p-TiO2 and PTC-x after calcination. Figure 1. (a) XRD pattern of bulk g-C3N4, bulk and nanoparticle TiGly and TGC-x before calcination; (b) XRD pattern of g-C3N4 nanosheets, bulk and nanoparticle p-TiO2 and PTC-x after calcination. 51 51 Catalysts 2019, 9, 439 After being calcined at 470 ◦C for 1 h, TiGly nanoparticles will be decomposed into anatase p-type TiO2. As shown in Figure 1b, no diﬀraction peaks of TiGly can be observed, only anatase TiO2 (JCPDS No. 21-1272) and g-C3N4 (JCPDS No. 87-1526) can be observed in PTC-x, with no other visible phases or impurities. 2. Results and Discussion Notably, the full-width at half maxima (FWHM) of peaks at 25.3◦corresponding to (101) planes of p-type TiO2 nanoparticles is broadened from 0.57 to 0.65 compared with bulk p-TiO2, so the average particle size of the TiO2 nanoparticles is smaller than the bulk p-TiO2 according to Scherer Equation. The same phenomenon can also be observed in bulk g-C3N4 and g-C3N4 nanosheets, the peak at 27.4◦corresponding to (002) plane of g-C3N4 nanosheets is broadened and drastically weakened from PTC-60 to PTC-20, which is due to the decreased thickness of nanosheets during the thermal exfoliation process [39]. Moreover, we also calculated the lattice constant of the p-type TiO2 nanoparticles based on XRD patterns, the results indicate that a = b = 3.793 Å, slightly larger than normal anatase TiO2 (a = b = 3.785 Å for JCPDS No. 21-1272), whereas the c axis shrinks from 9.514 Å to 9.488 Å, which is identical with that of titanium defected TiO2 [28]. Thermogravimetric (TG) tests were conducted to quantify the relative content of TiO2 QDs in the as-prepared samples. As shown in Figure 2, the weight loss of the samples below 80 ◦C is due to the removal of surface absorbed water, whereas the weight loss between 80 ◦C and 400 ◦C is attributed to the decomposition of titanium glycerolate as shown in Equation (1) [28,29]. Since bulk g-C3N4 exhibits almost no weight loss in this range, the weight loss of TGC-x can all assigned to the decomposition of TiGly. As shown in Figure 2, the weight loss is 20.03%, 9.77% and 7.98% for TGC-20, TGC-40 and TGC-60, respectively. Therefore, according to Equations (2) and (3), the mass content of TiGly in TGC-x can be calculated as 43.7%, 21.3% and 17.4%, respectively, corresponding to a p-type TiO2 QDs mass content of 29.6%, 12.8% and 10.2% in PTC-20, PTC-40 and PTC-60, respectively. Additionally, there is a sight weight increase after 80 ◦C for all the samples, which may be due to the impurity in the sample gas used in TG tests, but this deviation will not aﬀect the calculated QDs content in the samples. 2. Results and Discussion TiGly (Ti3(C3H4O3)4) + 14O2→3Ti1-xO2 + 12CO2 + 10H2O (1) yTG = Ws WTG (2) yTiO2 = mTiO2 mTiO2 + mg−C3N4 = yTG(1 −WTG) yTG(1 −WTG) + 1 −yTG = yTG −yTGWTG 1 −yTGWTG (3) TiGly (Ti3(C3H4O3)4) + 14O2→3Ti1-xO2 + 12CO2 + 10H2O (1) (1) yTG = Ws WTG (2) (2) yTiO2 = mTiO2 mTiO2 + mg−C3N4 = yTG(1 −WTG) yTG(1 −WTG) + 1 −yTG = yTG −yTGWTG 1 −yTGWTG (3) (3) Among these ﬁgures: yTG is the calculated content of TiGly, %; Ws is the weight loss of TGC-x, %; WTG is the weight loss of pure TiGly, %; mTiO2 is the calculated mass of TiO2 in unit mass of PTC-x, g; mg-C3N4 is the calculated mass of g-C3N4 in unit mass of PTC-x, g; yTiO2 is the calculated mass content of p-type TiO2 QDs in PTC-x, %. Figure 2. TG proﬁles of bulk g-C3N4, TiGly and TGC-x. Figure 2. TG proﬁles of bulk g-C3N4, TiGly and TGC-x. 52 52 Catalysts 2019, 9, 439 High resolution transmission electron microscopy (HRTEM) analyses were conducted to reveal the morphology and composition of the samples. Unlike the densely packed bulk g-C3N4 reported in literatures [39], the as-prepared g-C3N4 exhibits a nanosheet structure, the edges are curled and rough due to the minimizing surface energy (Figure 3a), which provides a lower resistance pathway for charge transfer. As shown in Figure 3b–d, the light contrast nanosheets are g-C3N4 nanosheets whereas the dark contrast are the p-type TiO2 nanoparticles decomposed from TiGly. Figure 3e shows the lattice ﬁngers of 0.35 nm and 0.48 nm in the dark contrast, corresponding to the (101) and (002) planes of anatase TiO2, which reveal the exposure of (010) facet of anatase, consistent with the XRD results [31]. The typical size of TiO2 nanoparticles is 4–5 nm according to the particle size distribution result shown in the inset of Figure 3e, which is accordance with the characteristic size of quantum conﬁnement eﬀect. It is noteworthy that when the p-type TiO2 QDs are in-situ grown on the surface of g-C3N4, there is a strong interaction that exists between them so that the TiO2 QDs cannot be peeled oﬀeven after a 30 min ultra-sonication process. Moreover, the strong interaction implies an obvious interface heterojunction between TiO2 QDs and g-C3N4, which will enhance the transfer of photo generated electrons and holes [25]. 2. Results and Discussion In addition, in agreement with the TG results, the relative content of TiO2 QDs decreases from PTC-20 to PTC-60. Figure 3. TEM images of (a) g-C3N4 nanosheets, (b) PTC-20, (c) PTC-40 and (d) PTC-60, (e) is the enlarged image of PTC-40, inset: size distribution of TiO2 QDs in PTC-40. Figure 3. TEM images of (a) g-C3N4 nanosheets, (b) PTC-20, (c) PTC-40 and (d) PTC-60, (e) is the enlarged image of PTC-40, inset: size distribution of TiO2 QDs in PTC-40. In order to further investigate the composition and interaction between p-type TiO2 QDs and g-C3N4 in the as-prepared composites, Fourier transform infrared spectroscopy (FT-IR) and X-ray photoelectron spectroscopy (XPS) characterizations were conducted. As shown in Figure 4a, for the TGC-x, there are three characteristic bands corresponding to g-C3N4, namely the broad peaks located at 3000–3400 cm−1 assigned to the stretching of N–H bonds, the strong peaks at 1250–1650 cm−1 due to the stretching vibration of C=N heterocycles and C–N bonds, and the band around 808 cm−1 related to the ring vibration of s-triazine [45–48]. Moreover, besides the above three peaks, another three peaks related to TiGly appearing in TGC-x, the peaks located at around 1000–1150 cm−1 were attributed to the alcoholic Ti–O–C stretching mode, the apparent peak at 611 cm−1 was indexed to the stretching mode of Ti–O bonds, and the bands located at 2855–2927 cm−1 were assigned to the glycerol C-H stretching vibration [49]. Besides, the broad peak over 3000–3600 cm−1 was attributed to the presence of physically adsorbed water and glycerol O-H stretching mode. Therefore, by combining with the XRD, TG and FT-IR results, it is proved that the as-prepared TGC-x samples is composed by TiGly and g-C3N4. 53 Catalysts 2019, 9, 439 Figure 4. (a) FT-IR patterns of bulk g-C3N4, TiGly and TGC-x before calcination; (b) FT-IR patterns of layered g-C3N4, p-TiO2 and PTC-x. Figure 4. (a) FT-IR patterns of bulk g-C3N4, TiGly and TGC-x before calcination; (b) FT-IR patterns of layered g-C3N4, p-TiO2 and PTC-x. Upon calcination, the major FT-IR peaks of g-C3N4 almost all remained in PTC-x samples (Figure 4b). However, the bands assigned to C–H and O–H in TiGly disappeared, while only the stretching band of Ti–O remained, indicating the transformation of TiGly to TiO2, further conﬁrming the formation of TiO2 QDs@g-C3N4 heterojunction. 2. Results and Discussion Moreover, with the decrement of g-C3N4 content from PTC-60 to PTC-20, the intensity of s-triazine ring vibration 808 cm−1 also decreases, which is in agreement with the above TG results. It is noteworthy that the stretching vibration of Ti–O–Ti shifted signiﬁcantly towards a lower wavenumber in PTC-x composites, suggesting a strong interaction exists between p-TiO2 QDs and g-C3N4 [25], which is in favor of charge transfer across the interface and thus promotes the photocatalytic performance of the heterojunction. XPS spectrum were recorded to study the status of the C, N, Ti and O elements in the composites. Figure 5a shows the C1s XPS spectra of the samples, there are two main peaks located at 284.8 eV and 288.5 eV, respectively. The peak located at higher binding energy is attributed to the sp3-bonded C of N-C=N2 in g-C3N4, and the peak located at 284.8 eV is due to the surface contaminated carbon during XPS test and sp2-hybridized carbon atoms presented in graphic domains [29]. In the N1s XPS spectra (Figure 5b), a asymmetric proﬁles can be observed in all the samples, with a main peak at 401 eV and a shoulder peak at lower binding energy, the main peak is due to the sp2-hybridized nitrogen (C–N–C), while the shoulder peak is usually attributed to amino functional groups with a hydrogen atom (C–NH) and sp3-hybridized nitrogen (N-[C]3) [25,39]. For the Ti 2p XPS spectra of the PTC-x (Figure 5c), the binding energy of Ti 2p3/2 and Ti 2p1/2 are observed at 458.9 eV and 464.4 eV, respectively, suggests the existence of TiO2 in the samples. Notably, compared with the Ti 2p binding energy of 458.4 eV in n-type TiO2 [29], the Ti 2p binding energy in the as-prepared samples shifted towards a higher binding energy, indicating the existence of titanium vacancies (VTi) and p-type properties of the TiO2 QDs [28]. Figure 5. (a) C1s, (b) N1s and (c) Ti 2p XPS spectra of layered g-C3N4, p-TiO2 and PTC-x, respectively. Figure 5. (a) C1s, (b) N1s and (c) Ti 2p XPS spectra of layered g-C3N4, p-TiO2 and PTC-x, respectively. 2. Results and Discussion As shown in Figure 7b, a blue shift of VB edge is observed in the as-prepared heterojunctions due to the existence of TiO2 nanoparticles, and this tendency becomes more and more obvious with the increment of TiO2, conﬁrming the quantum dots nature of the surface decorated p-type TiO2 nanoparticles [31]. Figure 7. (a) UV-vis DRS spectra and (b) valence band density of states (DOS) of layered g-C3N4, PTC-20, PTC-40 and PTC-60. Fi ( ) UV i DRS d (b) l Figure 7. (a) UV-vis DRS spectra and (b) valence band density of states (DOS) of layered g-C3N4, PTC-20, PTC-40 and PTC-60. As mentioned above, we have demonstrated that the surface deposited TiO2 QDs is abundant in titanium vacancies, while both experimental and density functional theoretical (DFT) results have indicated that metal defected TiO2 is p-type semiconductor [28]. In general, g-C3N4 exhibits n-type conductivity, therefore, p-n heterojunctions will be formed across the interface, which can aﬀord a large electrical ﬁled and more eﬃcient charge separation. Therefore, according to the band gap and valence band (VB) position of pure g-C3N4 and TiO2, the band alignment of the p-n heterojunction (PTC-40) is diagrammed in Scheme 2. Both TiO2 and g-C3N4 can be excited under λ > 365 nm, charge redistribution process will occur to equilibrate the Fermi level (Ef) due to the large diﬀerence of Ef between p-type TiO2 and n-type g-C3N4, the consequence is that the photogenerated electrons migrate to the conduction band (CB) of TiO2 and holes to the g-C3N4 VB. However, only g-C3N4 can be excited under visible light irradiation (λ > 400 nm), and the photogenerated electrons will still transfer to the CB of TiO2, while the holes tend to keep stay in the VB of g-C3N4. In both cases, an eﬃcient electron-hole spatial separation can be realized and the lifetime of charge carriers can also be prolonged, these separated electrons can react with O2 or H2O to form reactive oxygen species (·O2−) or H2, while the holes will oxidize an organic pollutant or sacriﬁcial agent directly. In addition, the presence of metal vacancies will enlarge the width of VB, which controls the mobility of holes inherently [28], and thus the synergy inﬂuence of metal vacancies, quantum conﬁnement eﬀect, and nanosheet structure can eﬀectively promote the charge transfer and separation across the interface. The facilitated charge separation by the p-n heterojunction was confirmed by photoluminescence (PL) spectra. 2. Results and Discussion 54 Catalysts 2019, 9, 439 Catalysts 2019, 9, 439 The O1s XPS spectrum were shown in Figure 6, there is only one symmetrical peak at around 532.7 eV appearing in pure g-C3N4, which is attributed to the loosely bonded oxygen species on the surface (O2, H2O or OH groups), no peaks corresponding to C–O and N–C–O appeared at 531.4 eV, indicating that no O doping process occurred in g-C3N4 during the calcination process [38]. As for the TiO2@g-C3N4 composites, the O1s peaks split into two peaks located at 532.7 eV and 529.8 eV, respectively. The lower binding energy is assigned to the oxygen anions (O2−) in the crystal lattice of anatase [28], and the concentration of TiO2 QDs is proportional to the intensity of this peak. As shown in Figure 6b–d, from PTC-20 to PTC-60, the intensity of this peak becomes obviously weak, conﬁrming the gradually decreased content of TiO2 QDs in the composites. These results are in good agreement with the TEM, TG and FT-IR results, indicating that the composites are composed of g-C3N4 and p-type TiO2 QDs, and that the content of TiO2 QDs decreases from PTC-20 to PTC-60. Figure 6. Fitted O1s XPS spectra of (a) layered g-C3N4, (b) PTC-20, (c) PTC-40 and (d) PTC-60, respectively. Figure 6. Fitted O1s XPS spectra of (a) layered g-C3N4, (b) PTC-20, (c) PTC-40 and (d) PTC-60, respectively. The light absorption properties of the as-prepared samples were characterized by UV-visible diﬀuse reﬂection spectrum (UV-vis DRS). As illustrated in Figure 7a, pure g-C3N4 can absorb light from UV region to visible light up to 465 nm, while p-type TiO2 only absorb UV light up to 400 nm. As for the TiO2 QDs@g-C3N4 composites, there is a remarkable absorption edge extension to the visible region compared with single TiO2 [39]. Moreover, with the increasing content of g-C3N4 form PTC-20 to PTC-60, the optical absorption edge of p-n heterojunction shifts towards a longer wavelength, indicating a stronger light absorption in the visible light region, which is beneﬁcial for the improvement of photocatalytic performance. Meanwhile, the quantum-size eﬀect of the p-type TiO2 55 Catalysts 2019, 9, 439 QDs is illustrated by VB XPS spectra. 2. Results and Discussion As shown in Figure 8, pure g-C3N4 has an emission peak around 450 nm, whereas p-TiO2 exhibits a signal at around 400nm, which is due to the band-band transition, namely the energy corresponding to the emission is close to the excitation energy of g-C3N4 and TiO2, respectively [25,29]. The PL intensity of PTC-x is much lower than pure p-type TiO2 and g-C3N4, indicating a promoted electron-hole separation by the p-n heterojunction. Notably, PTC-40 exhibits the lowest PL intensity among the samples, suggesting a higher charge separation efficiency and a better photocatalytic performance, which also indicates that there exists an approximate p/n ratio between the p-type TiO2 and n-type g-C3N4 in the heterojunction. 56 Catalysts 2019, 9, 439 atalysts 2019, 9, 439 Scheme 2. Schematic illustration of TiO2 QDs@g-C3N4 p-n heterojunction and the pathway for electron-hole transfer and separation over the junction under simulated solar irradiation (λ > 365 nm) and visible light irradiation (λ > 400 nm). Scheme 2. Schematic illustration of TiO2 QDs@g-C3N4 p-n heterojunction and the pathway for electron-hole transfer and separation over the junction under simulated solar irradiation (λ > 365 nm) and visible light irradiation (λ > 400 nm). Figure 8. Steady state PL spectra of layered g-C3N4, PTC-20, PTC-40 and PTC-60. Figure 8. Steady state PL spectra of layered g-C3N4, PTC-20, PTC-40 and PTC-60. Photocatalytic degradation of organic pollutants and photocatalytic water splitting were conducted to evaluate the performance of as-prepared TiO2@g-C3N4 p-n heterojunctions. Figure 9a shows the photodegradation rate of the samples based on pseudo-ﬁrst-order reaction mode. Under simulated sunlight irradiation (λ > 365nm), the degradation rate of MO for PTC-40 with a TiO2 (p) to g-C3N4 (n) mass ratio of 0.15 is 0.52 min−1·g−1, which is the highest among the PTC-x and is 4.3-fold higher than that of pure g-C3N4. Moreover, we also tested the photoactivity with a cut-oﬀﬁlter of λ > 400 nm to evaluate the visible light photocatalytic performance, the reaction rate exhibits the same trends with that under simulated solar spectrum (Figure 9b), and the reaction rate for MO degradation of PTC-40 is 0.125 min−1·g−1, 5.4-fold higher than pure g-C3N4. This result conﬁrms that p-n heterojunction exhibits a better photocatalytic performance than single photocatalyst and that the as-prepared TiO2@g-C3N4 is active under visible light. 57 Catalysts 2019, 9, 439 Figure 9. 3. Experimental 3.1. Materials 3.1. Materials Ethanol, glycerol, melamine and methyl orange (MO) were all purchased from Tianjin Jiangtian Fine Chemical Research Institute. Titanium butoxide (C16H36O4Ti, TBOT), triethanolamine (TEOA) and phenol were purchased from J&K chemical. Milli-Q ultra-pure water with a resistivity larger than 18.2 MΩ·cm was used in all experiments. All the chemicals were reagent grade and used as received. 2. Results and Discussion Pseudo-ﬁrst-order reaction rate in degradation of methyl orange (MO) and phenol under: (a) full solar spectrum and (b) visible light (λ > 400 nm). Figure 9. Pseudo-ﬁrst-order reaction rate in degradation of methyl orange (MO) and phenol under: (a) full solar spectrum and (b) visible light (λ > 400 nm). The photocatalytic H2 evolution reaction rate is shown in Figure 10a, the H2 release rate for g-C3N4, PTC-20, PTC-40 and PTC-60 is 186 μmol·g−1·h−1, 712 μmol·g−1·h−1, 1072 μmol·g−1·h−1 and 838 μmol·g−1·h−1, respectively. It is clearly that the H2 evolution rate of PTC-40 is the fastest, which is 5.8-fold of pure g-C3N4, 1.5-fold of PTC-20 and 1.3-fold of PTC-60. Combining with the PL results and photoactivity, we can get the conclusion that PTC-40 has the most eﬃcient electron-hole separation, and thus exhibits the best photocatalytic performance among the as-fabricated samples. Meanwhile, this phenomenon also indicates that a suitable p/n ratio is required to construct the best p-n heterojunction. In comparison, we also loaded n-type TiO2 QDs on the surface of g-C3N4 to construct a type-II heterojunction and found that the promotion in photoactivity is far from that by p-n heterojunction (1.44-fold vs. 5.8-fold in H2 evolution), illustrating that p-n heterojunctions are more eﬀective in accelerating photogenerated charge carrier separation and promoting the photocatalytic performance. Figure 10. (a) Time course of hydrogen evolution and (b) hydrogen generation rate of the samples. Figure 10. (a) Time course of hydrogen evolution and (b) hydrogen generation rate of the samples. 58 Catalysts 2019, 9, 439 3.3. Preparation of TiO2 QDs@g-C3N4 p-n Heterojunction 3.3. Preparation of TiO2 QDs@g-C3N4 p-n Heterojunction 3.3. Preparation of TiO2 QDs@g-C3N4 p-n Heterojunction As shown in Scheme 1, for the fabrication of TiO2 QDs@g-C3N4 p-n heterojunction, x (x = 20, 40, 60) mg g-C3N4 was dispersed in 148 mL ethanol and sonicated for one hour. After that, 5 mL glycerol was added into the solution. After being stirred for another 20 min, 400 μL TBOT was dropwised into the solution and then titanium glycerolate (TiGly) was in-situ grown on the surface of g-C3N4, the solution was stirred at room temperature for 16 h. The resulted powders (labeled as TGC-x) were washed with water and absolute ethanol for several times, and dried at 70 ◦C overnight. After calcination of TGC-x in air at 470 ◦C for 1 h, TiO2 QDs@g-C3N4 p-n heterojunctions were obtained and labeled as PTC-x (x = 20, 40, 60). As a reference, n-type TiO2 QDs deposited g-C3N4 was synthesized with the same procedure (x = 40 mg) except that glycerol was replaced by 233 μL NH3·H2O for the purpose of triggering the nucleation of TiO2 on g-C3N4 as previously reported (the reference sample was labeled as NTC-40) [29]. 3.2. Preparation of Bulk g-C3N4 Bulk g-C3N4 was synthesized by thermal annealing melamine under air atmosphere. Typically, 5 g of melamine was put into an airtight crucible, then the crucible was placed into a muﬄe furnace and calcined at 550 ◦C for 4 h with a ramping rate of 5 ◦C/min. After being cooled down to room temperature naturally, the obtained yellow powder is bulk g-C3N4. 3.5. Photocatalytic Degradation and Hydrogen Evolution 3.5. Photocatalytic Degradation and Hydrogen Evolution 3.5. Photocatalytic Degradation and Hydrogen Evolution Photodegradation of organic pollutants (phenol and MO) was conducted in an opening quartz chamber (150 mL) vertically irradiated by a 300 W high-pressure xenon lamp (PLS-SXE300, Beijing Perfect Light Co. Ltd., Beijing, China) located on the upper position. The irradiation area was about 20 cm2. The light density was measured using a radiometer (Photoelectric Instrument Factory, Beijing Normal University, Model UV-A, Beijing, China), and the results indicate that the light density at 365 nm and 400 nm was 34.7 mW/cm2 and 32.5 mW/cm2, respectively. Reaction conditions included the following: a temperature of 25 ◦C, solution volume of 100 mL, C0 (phenol) of 400 μmol·L–1; C0 (MO) of 120 μmol·L–1; photocatalyst of 0.1 g·L–1. Prior to the reaction, the suspension was stirred without irradiation for 20 min to achieve an adsorption equilibrium. Samples were withdrawn, centrifuged and analyzed using a U-3010 UV-vis spectrometer. Photocatalytic hydrogen production was carried out in a Pyrex top-irradiation reaction vessel connected to closed glass gas system. 10 mg catalyst dispersed in 120 mL aqueous solution containing TEOA (30 vol.%). The temperature of reaction solution was maintained at 0 ◦C. The resultant hydrogen was analyzed using an oﬀ-line gas chromatography (Bruker 450-GC, CA, USA) equipped with a thermal conductive detector (TCD), 5 Å molecular sieve column, and N2 as carrier gas. 4. Conclusions In this work, p-TiO2 QDs@g-C3N4 p-n heterojunctions were fabricated by in-situ decorating titanium-defected TiO2 QDs on the surface of g-C3N4, in which TiO2 QDs bring up p-type conductivity and g-C3N4 aﬀords for n-type conductivity. The as-prepared p-n heterojunction exhibits higher charge separation eﬃciency and photocatalytic performance in H2 evolution reaction and degradation of organic pollutant than pure g-C3N4 and TiO2/g-C3N4 type-II heterojunction under both UV-light and visible light irradiation, which can be ascribed to the synergy of a large electrical ﬁeld over interface, a strong interface interaction, and the quantum conﬁnement eﬀect. In all the samples, PTC-40 with a p/n mass ratio of 0.15 exhibits the best photocatalytic performance. This work demonstrates that the construction of p-n heterojunction is an eﬀective pathway to accelerate the electron-hole separation that is the key for a highly eﬃcient photocatalyst. Author Contributions: Conceptualization, N.T. and S.W.; methodology, L.Z., W.D. and J.X.; formal analysis, F.W.; investigation, S.W. and F.W.; data curation, F.W., Z.S., X.W. and Y.H.; writing—original draft preparation, F.W.; writing—review and editing, S.W. Funding: This research was funded by the National Natural Science Foundation of China (No. 21808172), Tianjin Municipal Natural Science Foundation (No. 18JCQNJC05800, 18JCZDJC37200), Innovation Fund for Young Talents of TUST and the foundation of Tianjin Key Laboratory of Marine Resources and Chemistry (TUST, No. 201707). 3.4. Characterization of Photocatalysts 3.4. Characterization of Photocatalysts The calcination temperature for the decomposition of titanium glycerolate and the content of TiO2 was determined by thermogravimetric analysis (TGA Q500, TA Instruments, DE, USA) with air gas ﬂow at 50 mL/min in a range of 30–500 ◦C (5 ◦C/min). In order to obtain the composition of the samples, Fourier transform infrared spectroscopy (FT-IR, Bruker Tensor-27 spectrum, Bavaria, Germany) was conducted before and after calcination, the FT-IR spectra were acquired in the range of 400–4000 cm−1 with a resolution of 1 cm−1. All the samples were characterized with an X-ray diﬀractometer (XRD-6100, Shimadzu, Kyoto, Japan) to determine the crystalline properties. The X-ray diﬀractometer was equipped with a Cu Kα radiation at 40 kV and 30 mA at a scanning rate of 5◦/min. The diﬀraction patterns were determined over 2 theta range of 5◦–90◦with a resolution of 2◦/min. To obtain the average size of the crystalline size, Scherrer equation was used (D = 0.9λ/βcosθ), whereas the lattice constants was calculated according to Bragg equation of 2dhklsinθ = λ, where λ is the applied wavelength, θ is the Bragg angle and β is the FWHM value. The chemical states of the as-prepared samples were characterized by X-ray photoelectron spectroscopy (PHI-1600, ULVAC-PHI, Kanagawa, Japan) with Al Kα radiation, and the binding energy was calibrated by the C1s peak (284.8 eV) of the contamination carbon. High resolution transmission electron microscopy (HRTEM) analysis was carried out using a Tecnai G2 F-20 transmission electron microscope (FEI, OR, USA) with a ﬁeld-emission gun operation at 200 kV. The band gap and the light absorption properties of the samples were determined with a UV-vis diffuse reflectance spectrum (U-3010, Hitachi Ltd., Lbarakiken, Japan ) with a 60 mm diameter integrating sphere using BaSO4 as the reflectance sample. Steady-state photoluminescence spectra (PL) spectra were measured by a Fluorolog3-21 (Horiba JobinYvon, NJ, USA) with the excitation light at 325 nm. 59 Catalysts 2019, 9, 439 5. Zhang, H.; Liu, G.; Shi, L.; Liu, H.; Wang, T.; Ye, J. Engineering coordination polymers for photocatalysis. Nano Energy 2016, 22, 149–168. [CrossRef] 4. Wondraczek, L.; Tyystjärvi, E.; Méndez-Ramos, J.; Müller, F.A.; Zhang, Q. Shifting the Sun: Solar Spectral Conversion and Extrinsic Sensitization in Natural and Artiﬁcial Photosynthesis. Adv. Sci. 2015, 2, 1500218. [CrossRef] Funding: This research was funded by the National Natural Science Foundation of China (No. 21808172), Tianjin Municipal Natural Science Foundation (No. 18JCQNJC05800, 18JCZDJC37200), Innovation Fund for Young Talents of TUST and the foundation of Tianjin Key Laboratory of Marine Resources and Chemistry (TUST, No. 201707). 1. Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37–38. [CrossRef] 3. Zhang, N.; Yang, M.-Q.; Liu, S.; Sun, Y.; Xu, Y.-J. Waltzing with the Versatile Platform of Graphene to Synthesize Composite Photocatalysts. Chem. Rev. 2015, 115, 10307–10377. [CrossRef] [PubMed] 1. Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37–38. [CrossRef] 2. Wu, S.-M.; Liu, X.-L.; Lian, X.-L.; Tian, G.; Janiak, C.; Zhang, Y.-X.; Lu, Y.; Yu, H.-Z.; Hu, J.; Wei, H.; et al. Homojunction of Oxygen and Titanium Vacancies and its Interfacial n–p Eﬀect. Adv. Mater. 2018, 30, 1802173. [CrossRef] 3. Zhang, N.; Yang, M.-Q.; Liu, S.; Sun, Y.; Xu, Y.-J. Waltzing with the Versatile Platform of Graphene to Synthesize Composite Photocatalysts. Chem. Rev. 2015, 115, 10307–10377. [CrossRef] [PubMed] 4. Wondraczek, L.; Tyystjärvi, E.; Méndez-Ramos, J.; Müller, F.A.; Zhang, Q. Shifting the Sun: Solar Spectral Conversion and Extrinsic Sensitization in Natural and Artiﬁcial Photosynthesis. Adv. Sci. 2015, 2, 1500218. [CrossRef] 5. Zhang, H.; Liu, G.; Shi, L.; Liu, H.; Wang, T.; Ye, J. Engineering coordination polymers for photocatalysis. Nano Energy 2016, 22, 149–168. [CrossRef] 2. Wu, S.-M.; Liu, X.-L.; Lian, X.-L.; Tian, G.; Janiak, C.; Zhang, Y.-X.; Lu, Y.; Yu, H.-Z.; Hu, J.; Wei, H.; et al. Homojunction of Oxygen and Titanium Vacancies and its Interfacial n–p Eﬀect. Adv. Mater. 2018, 30, 1802173. [CrossRef] References 1. Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37–38. [CrossRef] 2. Wu, S.-M.; Liu, X.-L.; Lian, X.-L.; Tian, G.; Janiak, C.; Zhang, Y.-X.; Lu, Y.; Yu, H.-Z.; Hu, J.; Wei, H.; et al. Homojunction of Oxygen and Titanium Vacancies and its Interfacial n–p Eﬀect. Adv. Mater. 2018, 30, 1802173. [CrossRef] 3. Zhang, N.; Yang, M.-Q.; Liu, S.; Sun, Y.; Xu, Y.-J. Waltzing with the Versatile Platform of Graphene Synthesize Composite Photocatalysts. Chem. Rev. 2015, 115, 10307–10377. [CrossRef] [PubMed] 5. Zhang, H.; Liu, G.; Shi, L.; Liu, H.; Wang, T.; Ye, J. Engineering coordination polymers for photocatalysis. Nano Energy 2016, 22, 149–168. [CrossRef] 60 Catalysts 2019, 9, 439 6. Kapilashrami, M.; Zhang, Y.; Liu, Y.-S.; Hagfeldt, A.; Guo, J. Probing the Optical Property and Electronic Structure of TiO2 Nanomaterials for Renewable Energy Applications. Chem. Rev. 2014, 114, 9662–9707. [CrossRef] [PubMed] 7. Fresno, F.; Portela, R.; Suárez, S.; Coronado, J.M. Photocatalytic materials: recent achievements and n 7. Fresno, F.; Portela, R.; Suárez, S.; Coronado, J.M. Photocatalytic m future trends. J. Mater. Chem. A 2014, 2, 2863–2884. [CrossRef] 7. Fresno, F.; Portela, R.; Suárez, S.; Coronado, J.M. Photocatalytic materials: recent achievements and near future trends. J. Mater. Chem. A 2014, 2, 2863–2884. [CrossRef] , ; , ; , ; , J y future trends. J. Mater. Chem. A 2014, 2, 2863–2884. [CrossRef] future trends. J. Mater. Chem. A 2014, 2, 2863–2884. [CrossRef] 8. Chen, X.; Mao, S.S. Titanium Dioxide Nanomaterials: Synthesis, Properties, Modiﬁcations, and Applicati Chem. Rev. 2007, 107, 2891–2959. [CrossRef] [PubMed] 8. Chen, X.; Mao, S.S. Titanium Dioxide Nanomaterials: Synthesis, Properties, Modiﬁcations, and Applications. Chem. Rev. 2007, 107, 2891–2959. [CrossRef] [PubMed] Chem. Rev. 2007, 107, 2891–2959. [CrossRef] [PubMed] 9 Pan X ; Yang M -Q ; Fu X ; Zhang N ; Xu Y-J Defective TiO2 with oxygen vacancies: synthesis properties Pan, X.; Yang, M.-Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective TiO2 with oxygen vacancies: synthesis, properties d h l i li i N l 2013 5 3601 3614 [C R f] 9. Pan, X.; Yang, M.-Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective TiO2 with oxygen v and photocatalytic applications. Nanoscale 2013, 5, 3601–3614. [CrossRef] 10. Ma, Y.; Wang, X.; Jia, Y.; Chen, X.; Han, H.; Li, C. Titanium Dioxide-Based Nanomaterials for Photocatalytic Fuel Generations. Chem. Rev. 2014, 114, 9987–10043. [CrossRef] 11. References Wang, Z.; Yang, C.; Lin, T.; Yin, H.; Chen, P.; Wan, D.; Xu, F.; Huang, F.; Lin, J.; Xie, X.; et al. Visible-light photocatalytic, solar thermal and photoelectrochemical properties of aluminium-reduced black titania. Energy Environ. Sci. 2013, 6, 3007–3014. [CrossRef] 12. Yang, H.G.; Sun, C.H.; Qiao, S.Z.; Zou, J.; Liu, G.; Smith, S.C.; Cheng, H.M.; Lu, G.Q. Anatase TiO2 single crystals with a large percentage of reactive facets. Nature 2008, 453, 638. [CrossRef] 13. Li, L.; Yan, J.; Wang, T.; Zhao, Z.J.; Zhang, J.; Gong, J.; Guan, N. Sub-10 nm rutile titanium dioxide nanoparticles for eﬃcient visible-light-driven photocatalytic hydrogen production. Nat. Commun. 2015, 6, 5881. [CrossRef] [PubMed] 14. Liu, G.; Yang, H.G.; Pan, J.; Yang, Y.Q.; Lu, G.Q.; Cheng, H.-M. Titanium Dioxide Crystals with Tailored Facets. Chem. Rev. 2014, 114, 9559–9612. [CrossRef] [PubMed] 15. Di Valentin, C.; Pacchioni, G.; Selloni, A. Reduced and n-Type Doped TiO2: Nature of Ti3+ Species. J. Phys. Chem. C 2009, 113, 20543–20552. [CrossRef] 16. Livraghi, S.; Paganini, M.C.; Giamello, E.; Selloni, A.; Di Valentin, C.; Pacchioni, G. Origin of Photoactivity of Nitrogen-Doped Titanium Dioxide under Visible Light. J. Am. Chem. Soc. 2006, 128, 15666–15671. [CrossRef] 17. Li, J.-G.; Büchel, R.; Isobe, M.; Mori, T.; Ishigaki, T. Cobalt-Doped TiO2 Nanocrystallites: Radio-Frequency Thermal Plasma Processing, Phase Structure, and Magnetic Properties. J. Phys. Chem. C 2009, 113, 8009–8015. [CrossRef] 18. Cao, J.; Zhang, Y.; Liu, L.; Ye, J. A p-type Cr-doped TiO2 photo-electrode for photo-reduction. Chem. Commun. 2013, 49, 3440–3442. [CrossRef] 19. Wang, Y.; Han, P.; Lv, X.; Zhang, L.; Zheng, G. Defect and Interface Engineering for Aqueous Electrocatalytic CO2 Reduction. Joule 2018, 2, 2551–2582. [CrossRef] 20. Wang, T.; Liu, L.; Ge, G.; Liu, M.; Zhou, W.; Chang, K.; Yang, F.; Wang, D.; Ye, J. Two-dimensional titanium oxide nanosheets rich in titanium vacancies as an eﬃcient cocatalyst for photocatalytic water oxidation. J. Catal. 2018, 367, 296–305. [CrossRef] 21. Nowotny, M.K.; Bogdanoﬀ, P.; Dittrich, T.; Fiechter, S.; Fujishima, A.; Tributsch, H. Observations of p-type semiconductivity in titanium dioxide at room temperature. Mater. Lett. 2010, 64, 928–930. [CrossRef] 22. Wu, Q.; Huang, F.; Zhao, M.; Xu, J.; Zhou, J.; Wang, Y. Ultra-small yellow defective TiO2 nanoparticles for co-catalyst free photocatalytic hydrogen production. Nano Energy 2016, 24, 63–71. [CrossRef] 23. Dahl, M.; Liu, Y.; Yin, Y. Composite Titanium Dioxide Nanomaterials. Chem. Rev. 2014, 114, 9853–9889. [CrossRef] [PubMed] 24. Li, H.; Zhou, Y.; Tu, W.; Ye, J.; Zou, Z. References State-of-the-Art Progress in Diverse Heterostructured Photocatalysts toward Promoting Photocatalytic Performance. Adv. Funct. Mater. 2015, 25, 998–1013. [CrossRef] 25. Tong, Z.; Yang, D.; Xiao, T.; Tian, Y.; Jiang, Z. Biomimetic fabrication of g-C3N4/TiO2 nanosheets with enhanced photocatalytic activity toward organic pollutant degradation. Chem. Eng. J. 2015, 260, 117–125. [CrossRef] 26. Wang, S.; Huang, C.-Y.; Pan, L.; Chen, Y.; Zhang, X.; Fazale, A.; Zou, J.-J. Controllable fabrication of homogeneous ZnO p-n junction with enhanced charge separation for eﬃcient photocatalysis. Catal. Today 2018. [CrossRef] 27. Wang, M.; Sun, L.; Lin, Z.; Cai, J.; Xie, K.; Lin, C. p–n Heterojunction photoelectrodes composed of Cu2O-loaded TiO2 nanotube arrays with enhanced photoelectrochemical and photoelectrocatalytic activities. Energy Environ. Sci. 2013, 6, 1211–1220. [CrossRef] 61 Catalysts 2019, 9, 439 28. Wang, S.; Pan, L.; Song, J.-J.; Mi, W.; Zou, J.-J.; Wang, L.; Zhang, X. Titanium-Defected Undoped Anatase TiO2 with p-Type Conductivity, Room-Temperature Ferromagnetism, and Remarkable Photocatalytic Performance. J. Am. Chem. Soc. 2015, 137, 2975–2983. [CrossRef] 29. Pan, L.; Wang, S.; Xie, J.; Wang, L.; Zhang, X.; Zou, J.-J. Constructing TiO2 p-n homojunction for photoelectrochemical and photocatalytic hydrogen generation. Nano Energy 2016, 28, 296–303. [CrossRef] 3 30. Ma, L.; Han, H.; Pan, L.; Tahir, M.; Wang, L.; Zhang, X.; Zou, J.-J. Fabrication of TiO2 nanosheets via Ti3+ doping and Ag3PO4 QD sensitization for highly eﬃcient visible-light photocatalysis. RSC Adv. 2016, 6, 63984–63990. [CrossRef] 31. Pan, L.; Zou, J.-J.; Wang, S.; Huang, Z.-F.; Yu, A.; Wang, L.; Zhang, X. Quantum dot self-decorated TiO2 nanosheets. Chem. Commun. 2013, 49, 6593–6595. [CrossRef] 32. Anpo, M.; Kawamura, T.; Kodama, S.; Maruya, K.; Onishi, T. Photocatalysis on titanium-aluminum binary metal oxides: enhancement of the photocatalytic activity of titania species. J. Phys. Chem. 1988, 92, 438–440. [CrossRef] 33. Cao, S.; Yu, J. g-C3N4-Based Photocatalysts for Hydrogen Generation. J. Phys. Chem. Lett. 2014, 5, 2101–2107. [CrossRef] [PubMed] 34. Wang, X.; Blechert, S.; Antonietti, M. Polymeric Graphitic Carbon Nitride for Heterogeneous Photocatalysis. ACS Catal. 2012, 2, 1596–1606. [CrossRef] 35. Low, J.; Cao, S.; Yu, J.; Wageh, S. Two-dimensional layered composite photocatalysts. Chem. Commun. 2014, 50, 10768–10777. [CrossRef] 36. Butchosa, C.; Guiglion, P.; Zwijnenburg, M.A. Carbon Nitride Photocatalysts for Water Splitting: A Computational Perspective. J. Phys. Chem. C 2014, 118, 24833–24842. [CrossRef] 37. Zhang, H.; Zuo, X.; Tang, H.; Li, G.; Zhou, Z. Origin of photoactivity in graphitic carbon nitride and strategies for enhancement of photocatalytic eﬃciency: insights from ﬁrst-principles computations. Phys. Chem. Chem. Phys. 2015, 17, 6280–6288. 48. Zhang, J.; Zhang, M.; Lin, L.; Wang, X. Sol Processing of Conjugated Carbon Nitride Powders for Thin-Film Fabrication. Angew. Chem. Int. Edit. 2015, 54, 6297–6301. [CrossRef] 49. Das, J.; Freitas, F.S.; Evans, I.R.; Nogueira, A.F.; Khushalani, D. A facile nonaqueous route for fabricating titania nanorods and their viability in quasi-solid-state dye-sensitized solar cells. J. Mater. Chem. 2010, 20, 4425–4431. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Received: 25 March 2019; Accepted: 19 April 2019; Published: 30 April 2019 Received: 25 March 2019; Accepted: 19 April 2019; Published: 30 April 201 Abstract: In this study, oxygen vacancy modiﬁed TiO2 nanorod array photoelectrode was prepared by reducing hydrogen atmosphere to increase its free charge carrier density. Subsequently, a p-type conductive poly 3,4-ethylenedioxythiophene (PEDOT) layer was deposited on the surface of oxygen vacancy modiﬁed TiO2, to inhibit the surface states. Meanwhile, a p-n heterojunction formed between PEDOT and TiO2 to improve the separation of photo-induced carriers further. The photocurrent of TiO2 nanorod array increased to nearly 0.9 mA/cm2 after the co-modiﬁcation under standard sunlight illumination, whose value is nearly nine times higher than that of pure TiO2 nanorod array. Thus, this is a promising modiﬁcation method for TiO2 photoanode photoelectrochemical (PEC) performance improving. Keywords: oxygen vacancy; polymeric composites; photoelectrochemistry; co-modification; solar energy conversion Catalysts 2019, 9, 407; doi:10.3390/catal9050407 www.mdpi.com/journal/catalysts Improvement of the Photoelectrochemical Performance of TiO2 Nanorod Array by PEDOT and Oxygen Vacancy Co-Modiﬁcation Bin Yang 1, Guoqiang Chen 2, Huiwen Tian 3,* and Lei Wen 1,* Bin Yang 1, Guoqiang Chen 2, Huiwen Tian 3,* and Lei Wen 1,* 1 National Center for Materials Service Safety, University of Science and Technology Beijing, Beijing 100083, China; binyang@ustb.edu.cn 2 Beijing General Research Institute of Mining & Metallurgy Group, Beijing 100160, China; ccggqq871020@126.com 2 Beijing General Research Institute of Mining & Metallurgy Group, Beijing 100160, China; ccggqq871020@126.com ggqq 3 Key Laboratory of Marine Environmental Corrosion and Bio-fouling, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China * Correspondence: tianhuiwen1983@foxmail.com (H.T.); wenlei@ustb.edu.cn (L.W.); Tel.: +86-010-62333132 (H.T.); +86-0532-82898832 (L.W.) Received: 25 March 2019; Accepted: 19 April 2019; Published: 30 April 2019 References [CrossRef] 38. Huang, Z.-F.; Song, J.; Pan, L.; Wang, Z.; Zhang, X.; Zou, J.-J.; Mi, W.; Zhang, X.; Wang, L. Carbon nitride with simultaneous porous network and O-doping for eﬃcient solar-energy-driven hydrogen evolution. Nano Energy 2015, 12, 646–656. [CrossRef] 39. Zhang, J.-W.; Gong, S.; Mahmood, N.; Pan, L.; Zhang, X.; Zou, J.-J. Oxygen-doped nanoporous carbon nitride via water-based homogeneous supramolecular assembly for photocatalytic hydrogen evolution. Appl. Catal. B-Environ. 2018, 221, 9–16. [CrossRef] 40. Ye, C.; Li, J.-X.; Li, Z.-J.; Li, X.-B.; Fan, X.-B.; Zhang, L.-P.; Chen, B.; Tung, C.-H.; Wu, L.-Z. Enhanced Driving Force and Charge Separation Eﬃciency of Protonated g-C3N4 for Photocatalytic O2 Evolution. ACS Catal. 2015, 5, 6973–6979. [CrossRef] 41. Zhang, S.; Li, J.; Wang, X.; Huang, Y.; Zeng, M.; Xu, J. Rationally designed 1D Ag@AgVO3 nanowire/graphene/protonated g-C3N4 nanosheet heterojunctions for enhanced photocatalysis via electrostatic self-assembly and photochemical reduction methods. J. Mater. Chem. A 2015, 3, 10119–10126. [CrossRef] 42. Yang, S.B.; Gong, Y.J.; Zhang, J.S.; Zhan, L.; Ma, L.L.; Fang, Z.Y.; Vajtai, R.; Wang, X.C.; Ajayan, P.M. Exfoliated Graphitic Carbon Nitride Nanosheets as Eﬃcient Catalysts for Hydrogen Evolution Under Visible Light. Adv. Mater. 2013, 25, 2452–2456. [CrossRef] 43. Niu, P.; Zhang, L.L.; Liu, G.; Cheng, H.M. Graphene-Like Carbon Nitride Nanosheets for Improved Photocatalytic Activities. Adv. Funct. Mater. 2012, 22, 4763–4770. [CrossRef] 44. Xu, H.; Yan, J.; She, X.; Xu, L.; Xia, J.; Xu, Y.; Song, Y.; Huang, L.; Li, H. Graphene-analogue carbon nitride: novel exfoliation synthesis and its application in photocatalysis and photoelectrochemical selective detection of trace amount of Cu2+. Nanoscale 2014, 6, 1406–1415. [CrossRef] 45. Kang, Y.; Yang, Y.; Yin, L.-C.; Kang, X.; Liu, G.; Cheng, H.-M. An Amorphous Carbon Nitride Photocatalyst with Greatly Extended Visible-Light-Responsive Range for Photocatalytic Hydrogen Generation. Adv. Mater. 2015, 27, 4572–4577. [CrossRef] 46. Jun, Y.-S.; Lee, E.Z.; Wang, X.; Hong, W.H.; Stucky, G.D.; Thomas, A. From Melamine-Cyanuric Acid Supramolecular Aggregates to Carbon Nitride Hollow Spheres. Adv. Funct. Mater. 2013, 23, 3661–3667. [CrossRef] 47. Liang, Q.; Li, Z.; Yu, X.; Huang, Z.-H.; Kang, F.; Yang, Q.-H. Macroscopic 3D Porous Graphitic Carbon Nitride Monolith for Enhanced Photocatalytic Hydrogen Evolution. Adv. Mater. 2015, 27, 4634–4639. [CrossRef] 62 Catalysts 2019, 9, 439 © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 63 63 catalysts catalysts 2. Results and Discussion The XRD (X-ray Diﬀraction) patterns of series samples were shown in Figure 1, all diﬀraction peaks of the prepared three photoelectrodes can be indexed as rutile-type and anatase-type TiO2 (JCPDS No. 21-1276, JCPDS No. 21-1272) [21,22]. The characteristic diﬀraction peaks at 2θ = 36.08◦, 54.32, 62.74, and 69.78◦corresponded to the (101), (211), (002), and (112) crystal planes of rutile-type TiO2, and the XRD peaks at 2θ = 63.68◦corresponded to the (204) crystal planes of anatase-type TiO2. The other characteristic diﬀraction peaks at 2θ = 26.57◦, 37.76◦, 51.75◦, and 65.74◦corresponded to the (110), (200), (211), and (301) crystal planes of SnO2 (JCPDS No. 46-1088), which caused by the ﬂuorine doped tin oxide (FTO) conductive glass. So, the prepared TiO2 nanorod array included rutile phases and little anatase phases. The TiO2 nanorod array preparing method in this work was referred to in Liu’s work [17]. The vanished peaks for anatase and rutile TiO2 at 25.4◦and 27.4◦on the XRD curves maybe attributed to the crystal face inhibition eﬀect of the oriented growth nanorod structure, whose results are similar to Liu’s work [17]. # # # ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∗ # # # # # # # # # ∗ # # ∗ ,QWHQVLW\DX #7L2 ∗6Q2 # ∗ +7L23('27 +7L2 7L2 θGHJUHH 5XWLHO $QDWDVH 3') 3') 6Q23') Figure 1. XRD patterns of TiO2, H-TiO2, and H-TiO2-PEDOT. Figure 1. XRD patterns of TiO2, H-TiO2, and H-TiO2-PEDOT. The SEM technique was employed to observe the surface morphologies of the series samples, and the results are shown in Figure 2. As presented in Figure 2A–C, both TiO2, H-TiO2, and H-TiO2-PEDOT appear to have a distinct nanorod structure. The cross-section image of H-TiO2-PEDOT shown in Figure 2D reveals that the TiO2 nanorod is growing vertically on the FTO substrate. The nanorods are tetragonal in shape with square top facets, the expected growth habit for the tetragonal crystal structure. The nanorods are nearly perpendicular to the FTO substrate. After 8 h of growth, the average diameter and length, as determined from SEM, were 90 ± 20 nm and 1 ± 0.2 μm, respectively. 1. Introduction PEDOT is a promising material to modify the TiO2 photoanode to improve its PEC performance [17–20]. Therefore, in this work, we prepared oxygen vacancy modiﬁed TiO2 nanorod array photoanode with high charge mobility capacity. Then, a p-type PEDOT layer was covered on the surface of oxygen vacancy modiﬁed TiO2 photoanode to inhibit the undesirable surface state and construct a p-n heterojunction to accelerate the separation capacity of photo-generated carriers [5]. 1. Introduction TiO2 has been widely investigated in the past few decades since Fujishima and Honda ﬁrst reported its potential in the ﬁelds of photocatalysis and photoelectrochemistry in 1972 [1]. The theoretical limited photocurrent densities of anatase and rutile TiO2 are 1.1 mA/cm2 and 1.8 mA/cm2 under solar light illumination, respectively. [2] Limited by its low solar light utilization rate and high photo-generated carrier recombination rate, many modiﬁcation methods have been researched, such as metal doped [3], non-metal doped [4], and construct heterojunction [5]. Several elements have been introduced into TiO2, such as Fe [6], S [7], and N [8]. Metal and non-metal doping could narrow the bandgap, extend the light absorption range and increase the charge carrier density to improve its photocatalysis performance. However, the introduction of heterogeneous atoms is likely to cause asymmetric doping or impurities, which would serve as recombination centers for the photo-generated electrons and holes, therefore reducing the PEC performance. Many previous research works showed that the formation of surface oxygen vacancy [9–13] could increase the charge carrier density of the semiconductor to improve its PEC performance. Wang et al. [14] obtained a yellowish ZnO with a narrowing band gap by introducing the oxygen vacancies into ZnO crystal, which increased the free charge density of the ZnO, so that the transfer process of the photogenerated charges became feasible. Polymer organic semiconductors with good ﬁlm-forming properties, high conductivity, high visible light transmittance and excellent stability are widely used in the ﬁeld of photoelectrode 65 Catalysts 2019, 9, 407 modiﬁcation. Park et al. [15] used a blend of 100 nm TiO2 scattering particles in PEDOT:PSS (poly 3,4-ethylenedioxythiophene:poly styrenesulfonate) solution to fabricate transparent electrode ﬁlms. When utilized in an organic photovoltaic device, a power conversion eﬃciency of 7.92% was achieved. Sakai et. al. [16] assembled PEDOT and TiO2 layer-by-layer to switch electric conductivity in response to ultraviolet and visible light. PEDOT is a promising material to modify the TiO2 photoanode to improve its PEC performance [17–20]. modiﬁcation. Park et al. [15] used a blend of 100 nm TiO2 scattering particles in PEDOT:PSS (poly 3,4-ethylenedioxythiophene:poly styrenesulfonate) solution to fabricate transparent electrode ﬁlms. When utilized in an organic photovoltaic device, a power conversion eﬃciency of 7.92% was achieved. Sakai et. al. [16] assembled PEDOT and TiO2 layer-by-layer to switch electric conductivity in response to ultraviolet and visible light. 2. Results and Discussion Compared to TiO2, the peak area of Odef in H-TiO2 was enhanced after the hydrogen treatment, indicating the increase of oxygen vacancies from 34.2% to 43.77%, which might improve the PEC performance [16]. Oxygen vacancy concentration refers to the proportion of oxygen vacancy peak area to the total oxygen peak area. Because XPS can only read the distribution of surface elements, the peak area of oxygen (O 1s, Olat, Odef) becomes smaller after PEDOT loading, but the relative content is credible. Next, the peak area of Odef in H-TiO2-PEDOT was reduced further after PEDOT deposition, which can be ascribed to the protection of PEDOT layer. The S was observed in the XPS spectra of H-TiO2-PEDOT indicating PEDOT was introduced successfully, which corresponds to the result of the XPS survey spectra shown in Figure 4C. Because of the low loading amount of PEDOT, noises can be found on the XPS S2p curve. The characteristic peaks of Ti 2p did not shift after the hydrogen treatment, and the deposition of PEDOT (Figure 4D) indicated that the unique TiO2 nanorod structure was preserved. 9R 9R $ 2GHI 2GHI 2GHI 2ODW 2ODW 2ODW 2s ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 +7L23('27 +7L2 7L2 9R +7L23('27 6S ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 % 2V 2V 7LS 7LS &V &V 6S 2V 2V 7L2 +7L23('27 ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 & 7L2 +7L23('27 7LS ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 ' Figure 4. High-resolution XPS spectra of O 1s (A) and S 2p (B) of the TiO2, H-TiO2 and H-TiO2-PEDOT. XPS survey spectra (C), high-resolution XPS spectra of Ti 2p (D) of the TiO2 and H-TiO2-PEDOT. +7L23('27 6S ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 % 9R 9R $ 2GHI 2GHI 2GHI 2ODW 2ODW 2ODW 2s ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 +7L23('27 +7L2 7L2 9R 7L2 +7L23('27 7LS ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 ' 2V 2V 7LS 7LS &V &V 6S 2V 2V 7L2 +7L23('27 ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 & ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 Figure 4. High-resolution XPS spectra of O 1s (A) and S 2p (B) of the TiO2, H-TiO2 and H-TiO2-PEDOT. 2. Results and Discussion The peaks of (101) crystal planes for rutile and (204) for anatase TiO2 can be clearly observed in the HRTEM image inset in Figure 2D, which is in agreement with the XRD results. Meanwhile, PEDOT layer can be observed at the edge area of TiO2 nanorod. Elements distribution of H-TiO2-PEDOT were tested by STEM and STEM-EDS mapping. The STEM mapping shown in Figure 2E reveals the uniform distribution of Ti, O, and S element on the surface of the nanorod, 66 Catalysts 2019, 9, 407 where the S element corresponding to the PEDOT deposition layer. This result indicates that PEDOT layer was successfully deposited on the surface of H-TiO2 photoelectrode. where the S element corresponding to the PEDOT deposition layer. This result indicates that PEDOT layer was successfully deposited on the surface of H-TiO2 photoelectrode. Figure 2. SEM images of (A) TiO2, (B) hydrogen treated TiO2 (H-TiO2), (C) PEDOT modiﬁed hydrogen treated TiO2 (H-TiO2-PEDOT) and (D) cross-section image of H-TiO2-PEDOT. Insert is the HRTEM image of H-TiO2-PEDOT. (E) STEM mapping of H-TiO2-PEDOT. Figure 2. SEM images of (A) TiO2, (B) hydrogen treated TiO2 (H-TiO2), (C) PEDOT modiﬁed hydrogen treated TiO2 (H-TiO2-PEDOT) and (D) cross-section image of H-TiO2-PEDOT. Insert is the HRTEM image of H-TiO2-PEDOT. (E) STEM mapping of H-TiO2-PEDOT. Similar results could be observed on EDS mapping (Figure 3), in which, the O, Ti, and Sn element corresponding to TiO2 nanorod and FTO substrate were evenly distributed throughout all the H-TiO2-PEDOT photoelectrode, besides, C and S elements could be observed simultaneously, which is corresponding to the STEM mapping showed in Figure 2E. Figure 3. EDS mapping of H-TiO2-PEDOT. (A) Scanning area, (B) O Element, (C) C Element, (D) Ti Element, (E) Sn Element, (F) S Element. Figure 3. EDS mapping of H-TiO2-PEDOT. (A) Scanning area, (B) O Element, (C) C Element, (D) Ti Element, (E) Sn Element, (F) S Element. To determine the surface composition and chemical states of the series samples, high-resolution XPS spectra of O 1s and S 2p were used (see Figure 4). The characteristic peaks at 529.8 eV and To determine the surface composition and chemical states of the series samples, high-resolution XPS spectra of O 1s and S 2p were used (see Figure 4). The characteristic peaks at 529.8 eV and 67 Catalysts 2019, 9, 407 530.8 eV correspond to the lattice oxygen (Olat) and the vacancies of O element (Odef). 2. Results and Discussion 7LPHV $ &XUUHQW'HQVLW\P$FP 7L2 +7L23('27 +7L2 % 3RWHQWLDO9 &XUUHQW'HQVLW\P$FP 7L2 +7L2 +7L23('27 ' 7L2 +7L2 +7L23('27 5H>/VF3@[ ,P>/VF3@[ IPLQ ( 7L2 +7L2 +7L23('27 5H>9RF3@[ ,P>9RF3@[ IPLQ & 7L2 +7L2 +7L23('27 =UHORKP =,PDJRKP 5H>9RF3@[ Figure 5. The current density-time curves (I-T) (A), the current density-potential (I-V) (B), AC impedance spectroscopy (EIS) (C), intensity modulated photocurrent spectroscopy (CIMPS) (D) and intensity modulated photovoltage spectroscopy (CIMVS) (E) of series photoelectrodes. Figure 5D shows the CIMPS data of each sample under a monochrome light LED-365 nm with a 5% amplitude. The electron migration time of the sample can be obtained by converting the frequency of the minimum imaginary component into Equation (1), which is shown in the Experimental Section. Electron migration time represents the sum of the photogenerated electron time from excitation to the back electrode FTO and the time of photogenerated holes oxidation of the electrolytes in the electric double layer. Figure 5E is the CIMVS data in the same testing conditions. The electronic lifetime can be obtained by introducing the obtained frequency into Equation (2). From the calculation results shown in Table 1, the electron migration time decreases obviously after hydrogen treatment. However, hydrogen treatment also introduces defects in the surface and bulk phase, which increases the probability of secondary recombination to reduce the lifetime of photogenerated electron holes. After the PEDOT conductive layer deposition, the surface state cannot be oxidized by air, meanwhile, p-n junction can be formed between TiO2 and PEDOT thin ﬁlm. The formation of p-n junction electric ﬁeld accelerates the separation of photogenerated electron holes and reduces the electron migration time. The charge collection eﬃciency of these three samples was also calculated and the calculation process is shown in Equation (3). 2. Results and Discussion XPS survey spectra (C), high-resolution XPS spectra of Ti 2p (D) of the TiO2 and H-TiO2-PEDOT. The PEC performance results of series samples are presented in Figure 5. Figure 5A is the current density-time curve of the series of electrodes, and Figure 5B is the current density-voltage curve of the series of electrodes. The current density-voltage curve shows that the current density of the TiO2 sample at zero bias (vs. Ag/AgCl) is about 0.07 mA/cm2. The current density of H-TiO2 sample at zero bias is about 0.27 mA/cm2. The current density of H-TiO2-PEDOT sample at zero bias is about 0.33 mA/cm2. In the voltage range from −0.5 to 0.5 V, the photocurrent density of sample H-TiO2-PEDOT is higher than that of sample H-TiO2, and the PEC performance of pure TiO2 nanorod array is the worst. Figure 5C is the impedance data of each sample in the absence of light. The arc radius of pure TiO2 is the largest, corresponding to the largest impedance. The arc radius of H-TiO2 is the smallest, corresponding to the smallest impedance. After PEDOT deposition, the arc radius of H-TiO2-PEDOT become larger because of the impedance of PEDOT. After oxygen vacancies modiﬁcation, the arc radius and impedance of the obtained H-TiO2 sample decreases. PEDOT conductive layer coated on the hydrogen treated TiO2 photoelectrode make the arc radius of the obtained H-TiO2-PEDOT further smaller, indicating a smaller impedance of this sample. 68 Catalysts 2019, 9, 407 7LPHV $ &XUUHQW'HQVLW\P$FP 7L2 +7L23('27 +7L2 % 3RWHQWLDO9 &XUUHQW'HQVLW\P$FP 7L2 +7L2 +7L23('27 & 7L2 +7L2 +7L23('27 =UHORKP =,PDJRKP ' 7L2 +7L2 +7L23('27 5H>/VF3@[ ,P>/VF3@[ IPLQ ( 7L2 +7L2 +7L23('27 5H>9RF3@[ ,P>9RF3@[ IPLQ Figure 5. The current density-time curves (I-T) (A), the current density-potential (I-V) (B), AC impedance spectroscopy (EIS) (C), intensity modulated photocurrent spectroscopy (CIMPS) (D) and intensity modulated photovoltage spectroscopy (CIMVS) (E) of series photoelectrodes. 2. Results and Discussion H-TiO2-PEDOT photoanode shows a 37.71% charge collection eﬃciency whose value is higher than that of TiO2 and H-TiO2, indicating that more real hot carriers can be used in the PEC process. Table 1. The calculated data through CIMPS and CIMVS results. Table 1. The calculated data through CIMPS and CIMVS results. fmin(CIMPS) tr fmin(CIMVS) trec ŋ (%) TiO2 172.24 0.924499 154.22 1.032523 10.46 H-TiO2 673.58 0.236402 536.63 0.296733 20.33 H-TiO2-PEDOT 845.47 0.18834 526.63 0.302367 37.71 IPCE of the series of electrodes were tested and the results are shown in Figure 6A. It can be seen that the photoelectric conversion eﬃciency of hydrogen-reduced TiO2 is signiﬁcantly higher than that of non-reduced TiO2. After loading PEDOT on the photoelectrode, the H-TiO2-PEDOT electrodes reducing surface state have more than 60% photoelectric conversion eﬃciency. Figure 6B 69 Catalysts 2019, 9, 407 presents the ultraviolet-visible diﬀuse reﬂectance result of the series photoanodes. It can be seen that the absorption band edge of pure TiO2 is about 400 nm, because of anatase (band gap 3.2 eV) and rutile (band gap 3.0 eV) mix phase. After hydrogen treatment, an indicated absorption can be found from 400 nm to 600 nm, because of the oxygen vacancy energy level formed on the top of the TiO2 valance band. The light absorption capacity of oxygen modiﬁed TiO2 nanorod array did not change after the PEDOT outer layer loading. Comparing with Figure 6A, there is no photocurrent response of H-TiO2-PEDOT photoanode in the wavelength area from 400 nm to 600 nm, indicating that there is no IPCE contribution from oxygen vacancy surface energy level. Figure 6C is a photocurrent-time curve measured continuously for 4 h under 0.5 V (vs. Ag/AgCl) external bias voltage. After 4 h continuous illumination, the photocurrent generating by H-TiO2-PEDOT photoanode decays less than 10% of the initial value, showing acceptable stability. Meanwhile, the oxygen and hydrogen evolution performance were tested during the PEC stability testing for 4 h, and the result shown in Figure 6C indicate that the H-TiO2-PEDOT photoanode can completely split pure water into hydrogen and oxygen under simulated sunlight illumination. 2. Results and Discussion :DYHOHQJWKQP ,3&( 7L2 +7L2 +7L23('27 $ 7L2 +7L2 +7L23('27 $EVDX :DYHOHQJWKQP % DVSURGXFWLRQȝPRO &XUUHQW'HQVLW\P$FP 7LPHPLQ +7L23('27 & 2 + Figure 6. The IPCE curves of series photoelectrodes under 0.5 V (vs. Ag/AgCl) (A), UV-Vis DRS of series photoelectrodes (B), the stability test of H-TiO2-PEDOT under 0.5 V (vs. Ag/AgCl) and corresponding oxygen and hydrogen evolution performance (C). DVSURGXFWLRQȝPRO &XUUHQW'HQVLW\P$FP 7LPHPLQ +7L23('27 & 2 + 7L2 +7L2 +7L23('27 $EVDX :DYHOHQJWKQP % :DYHOHQJWKQP ,3&( 7L2 +7L2 +7L23('27 $ :DYHOHQJWKQP :DYHOHQJWKQP :DYHOHQJWKQP Figure 6. The IPCE curves of series photoelectrodes under 0.5 V (vs. Ag/AgCl) (A), UV-Vis DRS of series photoelectrodes (B), the stability test of H-TiO2-PEDOT under 0.5 V (vs. Ag/AgCl) and corresponding oxygen and hydrogen evolution performance (C). In Table 2, the related research on TiO2 electrodes in recent years is listed. When comparing them, we can see that H-TiO2-PEDOT electrodes presented in this work obtained relatively high PEC performance. Table 2. Statistical list of references. Electrode Light Source Voltage Electrolyte Current Density TiO2 B-NRs [23] Xe lamp 88 mW cm−2 1.1 V 1 M KOH 0.8 mA/cm2 TiO2 nanorod array [24] AM 1.5 100 mW cm−2 0.5 V 0.5 M NaClO4 15 μA/cm2 Carbon Dot/TiO2 Nanorod [25] Xe lamp 88 mW cm−2 0 V 0.1 M NaSO4 + 0.01 M Na2S 0.35 mA/cm2 H:TiO2 nanotube arrays [26] AM 1.5G 100 mW cm−2 0 V 1 M NaOH 0.6 mA/cm2 TiO2 nanotubes [27] UV light 70 mW cm−2 0.2 V 1 M KOH 0.125 mA/cm2 This Work Simulated sunlight 100 mW cm−2 0.5 V 0.1 M NaSO4 + KPi 0.9 mA/cm2 Table 2. Statistical list of references. 2. Results and Discussion The PEC performance improving the mechanism of H-TiO2-PEDOT nanorod photoanode is shown in Figure 7. Firstly, a nanorod array structure of TiO2 was prepared, which provided a unique route for the photogenerated electron transfer and reduced the recombination rate. In addition, after hydrogen treatment, oxygen vacancies formed on the surface of TiO2 nanorod, increasing the concentrations of free charge carriers. Lastly, a PEDOT layer was deposited on the surface of oxygen vacancy modiﬁed TiO2, to inhibit the surface states and improve the separation of photo-induced carriers further by p-n heterojunction formation between PEDOT and TiO2. Thus, more photogenerated holes were transferred to the PEDOT layer and oxidized water, whereas more photogenerated electrons were transferred to the FTO substrate through the TiO2 nanorod to improve the PEC performance of H-TiO2-PEDOT photoanode. 70 atalysts 2019, 9, 407 Figure 7. Schematic diagram of the mechanism of H-TiO2-PEDOT nanorod thin ﬁlm under simulated sunlight. Catalysts 2019, 9, 407 Figure 7. Schematic diagram of the mechanism of H-TiO2-PEDOT nanorod thin ﬁlm under simulated sunlight. 3.2. PEDOT Preparation The PEDOT was coated by H-TiO2 nanorod array through electrodeposition method. Typically, 1 mL 3,4-ethylenedioxythiophene (EDOT) and 20 mmol of sodium dodecyl sulfonate (SDS) were dissolved into 200 mL of deionized water under continuous stirring to prepare precursor solution, the deposition process was carried out in a three-electrode system in the above solution. The as-prepared H-TiO2 photoelectrodes, platinum and Ag/AgCl electrode were served as the working, counter, and reference electrodes, respectively. The electrodeposition was carried out using multi-current steps containing 0.01 s of 1 mA of anodic pulse, 0.004 s of 1 mA of cathodic pulse and 0.5 s of 0 A rest current. This above process is termed as one cycle, and 20 cycles were repeated, the obtained electrode was denoted as H-TiO2-PEDOT. Three-electrode system was used to test the H-TiO2-PEDOT stability with an applied bias of 0.5 V (vs. Ag/AgCl). At the same time, oxygen and hydrogen evolution performance were detected by gas chromatography (97900II) regularly. 3.1. TiO2 Nanorod Array and Oxygen Vacancy Modiﬁed TiO2 Nanorod Preparation 3.1. TiO2 Nanorod Array and Oxygen Vacancy Modiﬁed TiO2 Nanorod Preparation The TiO2 nanorod arrays were prepared through the solvothermal method. In a typical synthesis, 0.5 mL tetrabutyl titanate was dissolved in 15 mL of hydrochloric acid (36.5%) under continuous stirring, and then 15 mL of deionized water was added for another 5 min to obtain a homogenous solution. The mixed solution was then transferred into a 50 mL Teﬂon stainless steel autoclave, then two cleaned FTO substrates were immersed into the mixture and kept at 160 ◦C for 8 h in an oven. After that, the FTO substrates were cleaned with deionized water and then dried under ambient conditions, followed by annealing at 450 ◦C for 2 h with a ramping rate of 10 ◦C/min in air in a muﬄe furnace to obtain TiO2 nanorod array. Then, TiO2 nanorod array was reduced by annealing at 350 ◦C for 0.5 h with a ramping rate of 10 ◦C/min in hydrogen conditions, which was denoted as H-TiO2. 3. Materials and Methods All reagents used in this study were purchased from Aladdin Industrial Corporation (Shanghai, China) with analytical grade. Tetrabutyl titanate, 3,4-ethylenedioxythiophene, and sodium dodecyl sulfonate were not further puriﬁed. 3.4. PEC Performance Testing PEC performance measurements were performed in a traditional three-electrode experimental system using Zahner Zennium Pro Electrochemical Workstation (Zahner, Kronach, Germany). The prepared series photoelectrodes, Ag/AgCl (saturated KCl), and a piece of platinum acted as the working, reference, and counter electrodes, respectively. The series photoelectrodes were illuminated under a standard solar simulator (AM1.5G) (LSE341-Zahner, Kronach, Germany). All tests were performed in 0.1 M Na2SO4 electrolyte. The photocurrent test with time (I-t) curves was measured at a bias potential of 0 V (vs. Ag/AgCl). The linear sweep voltammetry (I-V) curves were measured from −0.5 to 1.5 V (vs. Ag/AgCl) at a scan rate of 0.02 V s−1. The IPCE of the photoelectrodes were tested at 0.5 V (vs. Ag/AgCl) bias potential using an IPCE tester (TLS03-Zahner, Germany). Electrochemical impedance spectroscopy (EIS) tests were performed at OCP vs. Ag/AgCl (saturated KCl) over the frequency range between 105 and 10−1 Hz. Control intensity modulated photocurrent/photovoltage spectroscopy (CIMPS/CIMVS) measured series photoelectrodes with an LED white light source (LSW) from 100 K to 0.1 Hz. The electron transit time (τr) and electron lifetime (τrec) can be obtained by the following Equations: τr = 1/(2π fCIMPS) (1) τrec = 1/(2π fCIMVS) (2) ŋ(%) = (1 −τr/τrec) × 100% (3) where fCIMPS/fCIMVS is the frequency of the minimum imaginary component. τr = 1/(2π fCIMPS) τrec = 1/(2π fCIMVS) ŋ(%) = (1 −τr/τrec) × 100% (2) τrec = 1/(2π fCIMVS) (2) ŋ(%) = (1 −τr/τrec) × 100% (3) where fCIMPS/fCIMVS is the frequency of the minimum imaginary component. 3.3. Characterization The micromorphology of the prepared photoelectrodes was characterized using a ﬁeld emission scanning electron microscope (FE-SEM, Ultra 55, Zeiss, Oberkochen, Germany) and a ﬁeld emission transmission electron microscope (FE-HRTEM, JEM-2100F, Beijing, China). TEM sample was scraping the electrode ﬁlm into powder and ﬁlling the power with alcohol or acetone in a small container. Then a 71 Catalysts 2019, 9, 407 small amount of powder sample was put into it, next, it was placed in an ultrasonic oscillator to vibrate for more than 15 min, and then a copper mesh with supporting ﬁlm was used to gently pull it out from the solution. The elemental compositions of the photoelectrodes were tested through energy dispersive spectroscopy (EDS, X-max, Oxford Instruments, Oxford, England) and scanning transmission electron microscopy (STEM, JEM-2100F, Tokyo, Japan) mapping. X-ray diﬀraction (XRD, D/MAX-2500/PC, Rigaku Co., Tokyo, Japan) was used to identify the crystalline structures of the prepared series photoelectrodes. The elementary composition and bonding information of the materials were analyzed using X-ray photoelectron spectroscopy (XPS; Axis Ultra, Kratos Analytical Ltd., Kratos Analytical, Manchester, England). Characterization of the optical absorption properties of a series of electrodes was done by UV-Vis diﬀuse reﬂectance (TU-1901, Persee Co., Beijing, China). References 1. Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238, 37–38. [CrossRef] [PubMed] 2. Liu, C.; Dasgupta, N.; Yang, P. Semiconductor nanowires for artiﬁcial photosynthesis. Chem. Mater. 2013, 26, 415–422. [CrossRef] 3. Yang, L.; Luo, S.; Li, Y.; Xiao, Y.; Kang, Q.; Cai, Q. High Eﬃcient Photocatalytic Degradation of p-Nitrophenol on a Unique Cu2O/TiO2 p-n Heterojunction Network Catalyst. Environ. Sci. Technol. 2010, 44, 7641. [CrossRef] [PubMed] 4. Kim, H.; Moon, G.; Monllor-Satoca, D.; Park, Y.; Choi, W. Solar Photoconversion Using Graphene/TiO2 Composites: Nanographene Shell on TiO2 Core versus TiO2 Nanoparticles on Graphene Sheet. J. Phys. Chem. C 2012, 116, 1535–1543. [CrossRef] 5. Ghosh, M.; Liu, J.; Chuang, S.; Jana, S. Fabrication of hierarchical V2O5 nanorods on TiO2 nanoﬁbers a their enhanced photocatalytic activity under visible light. ChemCatChem 2018, 10, 3305–3318. [CrossRef 6. Sonawane, R.; Kale, B.; Dongare, M. Preparation and photo-catalytic activity of Fe-TiO2 thin ﬁlms prepared by sol-gel dip coating. Mater. Chem. Phys. 2004, 85, 52–57. [CrossRef] 7. Tian, H.; Zhang, X.; Bu, Y. Sulfur-and carbon-codoped carbon nitride for photocatalytic hydrogen evolution performance improvement. ACS Sustain. Chem. Eng. 2018, 6, 7346–7354. [CrossRef] 8. Zeng, L.; Lu, Z.; Li, M.; Yang, J.; Song, W.; Zeng, D.; Xie, C. A modular calcination method to prepare modiﬁed N-doped TiO2 nanoparticle with high photocatalytic activity. Appl. Catal. B Environ. 2016, 183, 308–316. [CrossRef] 9. Lei, F.; Sun, Y.; Liu, K.; Gao, S.; Liang, L.; Pan, B.; Xie, Y. Oxygen vacancies conﬁned in ultrathin indium oxide porous sheets for promoted visible-light water splitting. J. Am. Chem. Soc. 2014, 136, 6826–6829. [CrossRef] 10. Bu, Y.; Ren, J.; Zhang, H.; Yang, D.; Chen, Z.; Ao, J. Photogenerated-carrier separation along edge dislo of WO3 single-crystal nanoﬂower photoanode. J. Mater. Chem. A 2018, 6, 8604–8611. [CrossRef] 11. Liu, X.; Zhou, K.; Wang, L.; Wang, B.; Li, Y. Oxygen vacancy clusters promoting reducibility and activity of ceria nanorods. J. Am. Chem. Soc. 2009, 131, 3140–3141. [CrossRef] [PubMed] 12. Bu, Y.; Tian, J.; Chen, Z.; Zhang, Q.; Li, W.; Tian, F.; Ao, J. Optimization of the Photo-Electrochemical Performance of Mo-Doped BiVO4 Photoanode by Controlling the Metal-Oxygen Bond State on (020) Facet. Adv. Mater. Interfaces 2017, 4, 1601235. [CrossRef] 13. Wang, S.; Pan, L.; Song, J.; Mi, W.; Zou, J.; Wang, L.; Zhang, X. Titanium-defected undoped anatase TiO2 with p-type conductivity, room-temperature ferromagnetism, and remarkable photocatalytic performance. J. Am. Chem. Soc. 2015, 137, 2975–2983. 4. Conclusions In this study, PEDOT modiﬁed oxygen vacancy-TiO2 nanorod was prepared, oxygen vacancy can improve the charge transfer capacity of TiO2. Meanwhile, the PEDOT could not only serve as the protective layer to inhibit the surface states, but also to fabricate a p-n junction to increase the separation eﬃciency of the photo-generated electrons and holes. Thus, a near 0.9 mA/cm2 photocurrent of TiO2 nanorod array was achieved after oxygen vacancy and PEDOT co-modiﬁcation under standard sunlight illumination. Furthermore, the PEC stability test showed that the photocurrent generating by H-TiO2-PEDOT photoanode decays less than 10% of the initial value after 4 h of continuous illumination. Meanwhile, the H-TiO2-PEDOT photoanode can completely split pure water into hydrogen and oxygen under simulated sunlight illumination. Thus, oxygen vacancy and PEDOT co-modiﬁcation is a promising method for TiO2 photoanode PEC performance improving. Author Contributions: Conceptualization, B.Y.; methodology, H.T., L.W.; experiment and analysis, B.Y., G.C Author Contributions: Conceptualization, B.Y.; methodology, H.T., L.W.; experiment and analysis, B.Y., G.C. Funding: This research received no external funding Funding: This research received no external funding. Funding: This research received no external funding. Acknowledgments: This work was ﬁnancially supported by the National Natural Science Foundation of Ch (Grant Nos. 51679227). Conﬂicts of Interest: The authors declare no conﬂict of interest. 72 Catalysts 2019, 9, 407 References [CrossRef] 14. Wang, J.; Wang, Z.; Huang, B.; Ma, Y.; Liu, Y.; Qin, X.; Zhang, X.; Dai, Y. Oxygen vacancy induced band-gap narrowing and enhanced visible light photocatalytic activity of ZnO. ACS Appl. Mater. Interfaces 2012, 4, 4024–4030. [CrossRef] 15. Park, Y.; Müller-Meskamp, L.; Vandewal, K.; Leo, K. PEDOT: PSS with embedded TiO2 nanoparticles as light trapping electrode for organic photovoltaics. Appl. Phys. Lett. 2016, 108, 253302. [CrossRef] 16. Sakai, N.; Prasad, G.; Ebina, Y.; Takada, K.; Sasaki, T. Layer-by-layer assembled TiO2 nanoparticle/PEDOT-PSS composite ﬁlms for switching of electric conductivity in response to ultraviolet and visible light. Chem. Mater. 2006, 18, 3596–3598. [CrossRef] 17. Liu, B.; Aydil, E. Growth of oriented single-crystalline rutile TiO2 nanorods on transparent conducting substrates for dye-sensitized solar cells. J. Am. Chem. Soc. 2009, 131, 3985–3990. [CrossRef] 18. Wu, F.; Yu, Y.; Yang, H.; German, L.; Li, Z.; Chen, J.; Yang, W.; Huang, L.; Shi, W.; Wang, L.; et al. Simultaneous Enhancement of Charge Separation and Hole Transportation in a TiO2-SrTiO3 Core-Shell Nanowire Photoelectrochemical System. Adv. Mater. 2017, 29, 1701432. [CrossRef] [PubMed] 19. Fonseca, S.; Moreira, T.; Parola, A.; Pinheiro, C.; Laia, C. PEDOT electrodeposition on oriented mesoporous silica templates for electrochromic devices. Sol. Energy Mater. Sol. Cells 2017, 159, 94–101. [CrossRef] 20. Taggart, D.; Yang, Y.; Kung, S.; McIntire, T.; Penner, R. Enhanced thermoelectric metrics in ultra-long electrodeposited PEDOT nanowires. Nano Lett. 2010, 11, 125–131. [CrossRef] 21. Ghosh, M.; Lohrasbi, M.; Chuang, S.; Jana, S. Mesoporous titanium dioxide nanoﬁbers with a signiﬁcantly enhanced photocatalytic activity. ChemCatChem 2016, 8, 2525–2535. [CrossRef] 73 Catalysts 2019, 9, 407 22. Nunes, D.; Pimentel, A.; Santos, L.; Barquinha, P.; Fortunato, E.; Martins, R. Photocatalytic TiO2 nanorod spheres and arrays compatible with ﬂexible applications. Catalysts 2017, 7, 60. [CrossRef] 23. Cho, I.; Chen, Z.; Forman, A.; Kim, D.; Rao, P.; Jaramillo, T.; Zheng, X. Branched TiO2 nanorods for photoelectrochemical hydrogen production. Nano Lett. 2011, 11, 4978–4984. [CrossRef] [PubMed] 24. Wolcott, A.; Smith, W.; Kuykendall, T.; Zhao, Y.; Zhang, J. Photoelectrochemical water splitting using dense and aligned TiO2 nanorod arrays. Small 2009, 5, 104–111. [CrossRef] [PubMed] 25. Bian, J.; Huang, C.; Wang, L.; Hung, T.; Daoud, W.; Zhang, R. Carbon dot loading and TiO2 nanorod length dependence of photoelectrochemical properties in carbon dot/TiO2 nanorod array nanocomposites. ACS Appl. Mater. Interfaces 2014, 6, 4883–4890. [CrossRef] [PubMed] 26. Wang, G.; Wang, H.; Ling, Y.; Tang, Y.; Yang, X.; Fitzmorris, R.; Li, Y. Soroosh Mortazavian 1, Ali Saber 2,* and David E. James 2 Soroosh Mortazavian 1, Ali Saber 2,* and David E. James 2 1 Department of Mechanical Engineering, University of Nevada, Las Vegas, Las Vegas, NV 89154, USA; mortazav@unlv.nevada.edu 2 Department of Civil and Environmental Engineering and Construction, University of Nevada, Las V 2 Department of Civil and Environmental Engineering and Construction, University of Nevada, Las Vegas, Las Vegas, NV 89154, USA; dave.james@unlv.edu Las Vegas, NV 89154, USA; dave.james@unlv.edu * Correspondence: sabersic@unlv.nevada.edu; Tel.: +1-702-285-2836 Received: 19 March 2019; Accepted: 8 April 2019; Published: 14 April 2019 Abstract: Textile industries produce copious amounts of colored wastewater some of which are toxic to humans and aquatic biota. This study investigates optimization of a bench-scale UV-C photocatalytic process using a TiO2 catalyst suspension for degradation of two textile dyes, Acid Blue 113 (AB 113) and Acid Red 88 (AR 88). From preliminary experiments, appropriate ranges for experimental factors including reaction time, solution pH, initial dye concentration and catalyst dose, were determined for each dye. Response surface methodology (RSM) using a cubic IV optimal design was then used to design the experiments and optimize the process. Analysis of variance (ANOVA) was employed to determine signiﬁcance of experimental factors and their interactions. Results revealed that among the studied factors, solution pH and initial dye concentration had the strongest eﬀects on degradation rates of AB 113 and AR 88, respectively. Least-squares cubic regression models were generated by step-wise elimination of non-signiﬁcant (p-value > 0.05) terms from the proposed model. Under optimum treatment conditions, removal eﬃciencies reached 98.7% for AB 113 and 99.6% for AR 88. Kinetic studies showed that a ﬁrst-order kinetic model could best describe degradation data for both dyes, with degradation rate constants of k1, AB 113 = 0.048 min−1 and k1, AR 88 = 0.059 min−1. Keywords: process optimization; response surface methodology; kinetic study; Advanced oxidation processes (AOPs); TiO2 catalyst; textile wastewater catalysts catalysts Catalysts 2019, 9, 360; doi:10.3390/catal9040360 References Hydrogen-treated TiO2 nanowire arrays for photoelectrochemical water splitting. Nano Lett. 2011, 11, 3026–3033. [CrossRef] [PubMed] 27. Li, Y.; Yu, H.; Song, W.; Li, G.; Yi, B.; Shao, Z. A novel photoelectrochemical cell with self-organized TiO2 nanotubes as photoanodes for hydrogen generation. Int. J. Hydrogen Energy 2011, 36, 14374–14380. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 74 74 Optimization of Photocatalytic Degradation of Acid Blue 113 and Acid Red 88 Textile Dyes in a UV-C/TiO2 Suspension System: Application of Response Surface Methodology (RSM) Optimization of Photocatalytic Degradation of Acid Blue 113 and Acid Red 88 Textile Dyes in a UV-C/TiO2 Suspension System: Application of Response Surface Methodology (RSM) www.mdpi.com/journal/catalysts Keywords: process optimization; response surface methodology; kinetic study; Advanced oxidation processes (AOPs); TiO2 catalyst; textile wastewater 1. Introduction Holes, on the other hand, can react with hydroxide ions (OH−) or adsorbed H2O on the catalyst’s surface and generate hydroxyl radicals. Finally, dye molecules will react with the formed radicals. This reaction takes place on the surface of the catalyst particles and will continue until complete mineralization of the organic species [10]. The photocatalytic reactions described above can be summarized as [8]: TiO2 + hν →e−+ h+ (1) e−+O2 →O·− 2 (2) h++Organic →CO2 (3) h++H2O →·OH + H+ (4) ·OH + Organic →CO2 + H2O (5) TiO2 + hν →e−+ h+ (1) e−+O2 →O·− 2 (2) h++Organic →CO2 (3) h++H2O →·OH + H+ (4) ·OH + Organic →CO2 + H2O (5) (1) (4) ·OH + Organic →CO2 + H2O (5) A number of studies have investigated aqueous phase photodegradation of various dyes using TiO2 catalyst [1,7,9,11–18]. Sohrabi and Ghavami (2008) [1] studied photocatalytic degradation of Direct Red 23 using UV/TiO2 system. They reported an increase in dye decomposition rate with increasing TiO2 concentration up to 4.0 g/L; the rate then decreased with further increases in catalyst dose. Juang et al. (2010) [7] investigated photodegradation and mineralization of single and binary Acid Orange 7 (AO7) and Reactive Red 2 (RR2) under UV irradiation in TiO2 suspensions. Their results showed that after 20 min of UV irradiation with 0.5 g/L TiO2, complete removals of single AO7 and RR2 were achieved at pH 6.8. Photocatalytic degradation of Amaranth dye was investigated in a UV-C/TiO2 system by Gupta et al. (2012) [13]. They obtained degradation eﬃciencies of 17%, 26%, 38% and 64% for UV, UV + H2O2, UV + TiO2 and UV + TiO2 + H2O2 systems, respectively, after 100 min irradiation. Barakat (2011) [9] investigated the removal of Procion® yellow H-EXL dye over TiO2 suspension and obtained 100% photodegradation eﬃciency under optimum conditions of pH = 5.0, TiO2 dose = 1.0 g/L and dye concentration = 10 mg/L. Toor et al. (2006) [14] evaluated the photocatalytic degradation of Direct Yellow 12 in a shallow pond slurry using TiO2 suspension under irradiation of UV light using black ﬂuorescent lamps lies in UV-A range. After 1.5 h and under optimum conditions (TiO2 dosage = 2.0 g/L, pH = 4.5 and initial dye concentration = 100 mg/L) complete decolorization was achieved. Khataee et al. 1. Introduction Dyes are widely used in several industries such as textile industry, paper, plastics, food, cosmetics and so forth. [1]. The textile industry has large water consumption and thereby, produces copious amounts of colored wastewater. It has been estimated that 1–20% of total dye consumption is lost during the dying process, which is subsequently introduced to the receiving water bodies [2]. Some dyes are carcinogenic and toxic to humans and aquatic biota [3], requiring appropriate treatments. Methods for color removal are generally divided into three main groups: physical, chemical and biological treatments. Physical methods, such as adsorption and screening, only transfer pollutants from one phase to another; therefore post-treatment is necessary for complete removal of contaminants [4]. The toxic nature and complex molecular structures of many dyes limit their biological degradation [5]. Hence, biological methods are usually not able to treat colored wastewaters [6]. In addition, biological methods have a disadvantage of producing large volumes of sludge [7]. Chemical methods, on the other hand, have demonstrated more promising results [4]. In chemical treatment methods, instead of 75 Catalysts 2019, 9, 360 transferring contaminants from one phase to another, the dyes are converted into harmless substances. Advanced oxidation processes (AOPs) are among the most powerful chemical treatment techniques used for removal of organic compounds. AOPs are characterized by in-situ generation of hydroxyl radicals (•OH), which are strongly oxidizing species (oxidative potential +2.8 V) [6]. Hydroxyl radicals unselectively attack organic molecules to degrade them into simpler and less harmful compounds and ultimately, convert them into CO2, H2O and mineral acids [4,8,9]. Photocatalytic degradation is an advanced oxidation method in which hydroxyl radicals are generated by irradiating UV light on a semiconductor catalyst [10]. In the past several decades, titanium dioxide (TiO2) has been proved to be more eﬃcient for the photocatalytic processes than other semiconductors. It is inexpensive, non-toxic, water-insoluble, highly reactive and photochemically stable [6]. In a photocatalytic reaction, when TiO2 particles are illuminated with a light source having energy greater than its band gap (Eg,TiO2 = 3.2 eV), electrons in the valence band promote to the conduction band, creating electron-hole pairs [6,11]. Formation of electron-hole pairs is a fast-reversible reaction. To prevent this, an electron acceptor, which is dissolved in most cases, is necessary to entrap free electrons and reduce the rate of electron-hole recombination [10]. Photogenerated electrons may also react with dye molecules and reduce them [6]. 1. Introduction (2009) [15] investigated degradation of three azo dyes by UV-A irradiation using immobilized TiO2 and achieved complete decolorization after 6 h at natural pH and an initial dye concentration of 30 mg/L. However, to the best knowledge of authors, there are limited numbers of studies [19] which have assessed interaction eﬀects between operational factors in the photocatalytic degradation process of dyes and optimized the process. 76 Catalysts 2019, 9, 360 There are several classical methods for design and optimization of experiments. For instance, the one-factor-at-a-time method does not consider the interactions among experimental factors [20,21]. The full factorial method, considers interaction eﬀects through a great number of experiments but can be time-consuming and costly in multi-variable systems [20,21]. Response surface methodology (RSM) is a collection of statistical and mathematical methods used for development of a functional relationship between a response of interest and a number of input variables [22]. This method is applied for designing experiments, evaluating the eﬀects of individual operational parameters and their interaction eﬀects and optimizing the parameters, with a signiﬁcant reduction in the number of experiments [23–25]. Several recent studies have optimized the response of various environmental treatment processes using models based on RSM [19,24–33]. For example, Saber et al. (2014, 2017) [26,27] used RSM to optimize Fenton and photo-Fenton processes for treatment of petroleum reﬁnery eﬄuents, Cifuentes et al. (2017) [34] used RSM for simulation of the ethanol’s catalytic steam reforming, Li et al. (2018) [35] used RSM to investigate photocatalytic performance and degradation mechanism of Aspirin by TiO2, Inger et al. (2019) [36] optimized ammonia oxidation using RSM and Aljuboury et al. (2016) [37] optimized TiO2/ZnO photodegradation of petroleum reﬁnery wastewaters by using RSM. The current work, for the ﬁrst time, optimizes experimental conditions for photocatalytic degradation of two anionic textile dyes, Acid Blue 113 (AB 113) and Acid Red 88 (AR 88), in a TiO2 suspension system using UV-C irradiation. A simple enclosed bench-scale batch photoreactor was constructed for this study. Mercury vapor UV-C lamps were positioned over a relatively shallow free surface dye solution to provide suﬃcient light penetration as well as reduce costs of employing quartz tubes to immerse a UV lamp in the solution. In RSM-designed experiments, initial dye concentration, catalyst loading and solution pH were considered as independent parameters. The degradation eﬃciency of dyes was the target response. 1. Introduction Modeling photocatalytic degradation eﬃciency, examining the inﬂuences of several variable parameters on degradation eﬃciency and their interactions and determining optimum conditions for dye removal were conducted using RSM with a cubic IV optimization method. Kinetic studies were also conducted to evaluate dyes’ photodegradation rates under optimum conditions. 2.1. Stage 1: Preliminary Experiments Eﬀects of changing (a) TiO2 dose (g/L) (Initial dye concentration = 50 mg/L, pH = 3.0, Reaction time= 90 min), (b) reaction time (min) (Initial dye concentration = 50 mg/L, pH = 3.0, TiO2 dose = 1.0 g/L), (c) Initial dye concentration (mg/L) (Reaction time = 90 min, pH = 3.0, TiO2 dose = 1.0 g/L) and (d) pH (Initial dye concentration = 50 mg/L, Reaction time = 90 min, TiO2 dose = 1.0 g/L). Error bars show standard deviation of duplicate runs. Figure 1c illustrates that, as expected, increasing dye concentration from 20 mg/L to 200 mg/L decreased removal eﬃciencies in both dyes (from 97.6% to 40% for AR 88 and from 92.6% to 54.4% for AB 113). This was likely because increasing dye concentration in the solution while maintaining a constant catalyst dose caused the ﬁxed number of catalysis sites to be saturated faster [39]. In addition, increased dye concentration probably decreased the light transmittance in the solution. Decreased UV penetration can reduce the activation rate of TiO2 particles and hinder the generation of •OH radicals, resulting in decreased photocatalytic degradation eﬃciencies for both dyes [39]. Figure 1d represents the eﬀects of changing pH from 2.0 to 10.0 on removal eﬃciencies of AB 113 and AR 88. For AB 113, increasing pH from 2.0 to 6.0 caused a marked decrease in removal eﬃciency from 96.7% to 40.3% (56.4% decrease). Further increasing pH above 6.0 up to 9.0 decreased eﬃciency by only an additional 18%. For AR 88, increasing solution pH from 2 to 10 only resulted in 11.8% decrease in degradation eﬃciency. Detailed discussion about the eﬀects of solution pH on each of dyes and possible interactions with other factors are presented next in Section 2.2. In order to investigate the sole contribution of adsorption in removal of dyes from the aqueous solutions, experiments were conducted under dark conditions (i.e., without UV-C radiation) for both AB 113 and AR 88 dyes. After 90 min reaction under dark conditions in the closed photoreactor under the ﬁxed experimental conditions of TiO2 dose = 1.0 g/L, initial dye concentration = 50 mg/L and pH = 3.0, concentrations of both AB 113 and AR 88 remained unchanged when the reacted samples were analyzed by the UV-VIS spectrophotometer. 2.1. Stage 1: Preliminary Experiments 2.1. Stage 1: Preliminary Experiments Figure 1 shows the eﬀects of diﬀerent experimental factors on photocatalytic degradation of AB 113 and AR 88 textile dyes in the ﬁrst stage one-factor-at-a-time preliminary experiments. Figure 1a shows that increasing TiO2 dose up to 2.0 g/L and 1.0 g/L, enhanced removal eﬃciencies of AB 113 and AR 88, respectively. Increases in catalyst dose above these thresholds resulted in decreased removal eﬃciencies. Even though increase of TiO2 particles in the solution might provide more active sites for the dye molecules to be adsorbed and degraded, excessive amounts of catalyst particles might aggregate leading to a decreased number of active sites. Excessive TiO2 doses can also increase the opacity of solution and consequently reduce the penetration of UV light and thus decrease the treatment eﬃciency [1,38]. These results agreed with the previous studies [1,9,14,38] reporting decreased photocatalytic eﬃciency when applying TiO2 dose above an optimum value. Figure 1b shows that increasing reaction time improved removal percentages for both dyes. A higher overall removal eﬃciency was observed for AR 88 compared with AB 113 over the entire reaction time. As observed, only 7.4% and 8.7% increases in removal eﬃciencies were observed after 90 min for AB 113 and AR 88 dyes, respectively, indicating that 90 min was suﬃcient for most of the dye degradation reactions to occur. Hence, a ﬁxed reaction time of 90 min was considered for both dyes in the main experiments (i.e., second stage experiments), while other experimental factors (e.g., initial dye concentration, catalyst dose and pH) were optimized using RSM. 77 Catalysts 2019, 9, 360 Figure 1. Results of preliminary experiments for photocatalytic degradation of Acid Blue (AB) 113 and Acid Red (AR) 88 textile dyes. Eﬀects of changing (a) TiO2 dose (g/L) (Initial dye concentration = 50 mg/L, pH = 3.0, Reaction time= 90 min), (b) reaction time (min) (Initial dye concentration = 50 mg/L, pH = 3.0, TiO2 dose = 1.0 g/L), (c) Initial dye concentration (mg/L) (Reaction time = 90 min, pH = 3.0, TiO2 dose = 1.0 g/L) and (d) pH (Initial dye concentration = 50 mg/L, Reaction time = 90 min, TiO2 dose = 1.0 g/L). Error bars show standard deviation of duplicate runs. Figure 1. Results of preliminary experiments for photocatalytic degradation of Acid Blue (AB) 113 and Acid Red (AR) 88 textile dyes. 2.1. Stage 1: Preliminary Experiments This showed that adsorption onto TiO2 particles did not by itself have a signiﬁcant role in dye removal, demonstrating that the observed removal of dyes in the UV-C/TiO2 system were due to photocatalytic process. 2.2. Stage 2: Process Optimization 2.2.1. Response Surface Plots, Fitted Models and ANOVA 2.2. Stage 2: Process Optimization 2.2.1. Response Surface Plots, Fitted Models and ANOVA Figures 2 and 3 show the results obtained from the 30 experimental runs (i.e., stage 2) for AB 113 and AR 88, respectively, after 90 min reaction time. Analysis of variance (ANOVA) for AB 113 and AR 88 are presented in Tables 1 and 2, respectively. Signiﬁcance of the model terms was evaluated based on computed F-statistic values and their associated p-values. Least-squares cubic regression models 78 Catalysts 2019, 9, 360 were generated by eliminating non-signiﬁcant terms (p-value > 0.05). Reduced cubic models for AB 113 and AR 88 are expressed in the Equations (6) and (7), respectively. g y g g (p ) 3 and AR 88 are expressed in the Equations (6) and (7), respectively. AB 113 Removal (%) = 76.75 −21.30A −17.41B −12.47C + 2.28AB −2.26AC −8.06A2 −11.57B2−5.14A2C −4.44AB2 + 5.23AC2 + 16.52B3 (6) AR 88 removal (%) = 74.21 −2.00A −7.67B −19.61C −2.03AC −1.25BC −11.18B2 −1.63ABC +3.57AB2 −3.99AC2 −3.30B2C −4.44A3 + 7.18B3 −3.95C3 (7) Figure 2. Response surface and contour plots for photocatalytic degradation of AB 113 as a function of (a) C: initial dye concentration (mg/L) and B: TiO2 dose (g/L) (pH = 3.0, reaction time = 90 min), (b) A: pH and B: TiO2 dose (g/L) (initial dye concentration = 50 mg/L, reaction time = 90 min) and (c) A: pH and C: initial dye concentration (mg/L) (TiO2 dose = 1.0 g/L, reaction time = 90 min). AB 113 Removal (%) = 76.75 −21.30A −17.41B −12.47C + 2.28AB −2.26AC −8.06A2 −11.57B2−5.14A2C −4.44AB2 + 5.23AC2 + 16.52B3 (6) (6) AR 88 removal (%) = 74.21 −2.00A −7.67B −19.61C −2.03AC −1.25BC −11.18B2 −1.63ABC +3.57AB2 −3.99AC2 −3.30B2C −4.44A3 + 7.18B3 −3.95C3 (7) Figure 2. Response surface and contour plots for photocatalytic degradation of AB 113 as a function of (a) C: initial dye concentration (mg/L) and B: TiO2 dose (g/L) (pH = 3.0, reaction time = 90 min), (b) A: pH and B: TiO2 dose (g/L) (initial dye concentration = 50 mg/L, reaction time = 90 min) and (c) A: pH and C: initial dye concentration (mg/L) (TiO2 dose = 1.0 g/L, reaction time = 90 min). 79 Catalysts 2019, 9, 360 Figure 3. 2.2.1. Response Surface Plots, Fitted Models and ANOVA For AR 88, Figure 3 shows that removal eﬃciencies greater than 90% occurred for conditions of pH between 2.0 and 4.5, TiO2 dose of 1.0 to 2.0 g/L and dye concentration between 20 and 30 mg/L. As observed from Figure 2 and Equation (6), AB 113 removal eﬃciency decreased with increasing pH (term A) and initial dye concentration (term C), which is consistent with preliminary experiments. pH could aﬀect the speciation of dye molecules and consequently, the electrostatic force between catalysts particles and dye molecules. Since chemical reactions associated with photocatalytic degradation take place on the surface of the catalyst particles [10,38], adsorption of the contaminants’ molecules onto catalyst surface is an essential step for eﬃcient photocatalytic degradation [40]. The point of zero charge (zpc) for TiO2 Degussa P 25 is 6.5 [41]. Hence, the TiO2 surface was positively charged at pH < 6.5 and negatively charged at pH > 6.5. On the other hand, AB 113 is a disulfonate acid dye (having two sulfonated (−SO3−) groups) with an acid dissociation constant (pKa) of 0.5 [38]. Therefore, AB 113 tended to be negatively charged at pH > 0.5. The increased density of positive charges on the surface of TiO2 particles under acidic conditions was likely to be favorable for adsorbing AB 113 [42] and consequently improved photodegradation eﬃciency. The cubic regression model also showed that there were interaction eﬀects between these two factors (i.e., A for pH and C for dye concentration) due to presence of the statistically-signiﬁcant terms AC, A2C and AC2 in the equation. These signiﬁcant interactions indicated that changing pH aﬀected the speciation and ionization state of AB 113 dye molecules as well as TiO2 particles’ surface charge [43]. Eﬀects of catalyst dose (term B) on AB 113 dye degradation were also consistent with preliminary experiments, as well as studies reported in the literature [1,9,14], which is improving removal eﬃciency by increasing TiO2 dose to an optimum point and then reducing the eﬃciency at greater values. Statistically signiﬁcant (p-value < 0.05) interactions between the catalyst dose (term B) and the solution pH (term A) were also observed in ANOVA results. These interactions could be due to the eﬀect of pH on the surface charge of TiO2, aﬀecting the adsorption of dye molecules on its surface. 2.2.1. Response Surface Plots, Fitted Models and ANOVA Response surface and contour plots for photocatalytic degradation eﬃciency of AR 88 as a function of (a) C: initial dye concentration (mg/L) and B: TiO2 dose (g/L) (pH = 3.0, reaction time = 90 min), (b) A: pH and B: TiO2 dose (g/L) (initial dye concentration = 50 mg/L, reaction time = 90 min) and (c) A: pH and C: initial dye concentration (mg/L) (TiO2 dose = 1.0 g/L, reaction time = 90 min). Table 1. Analysis of variance for modiﬁed cubic model obtained for photocatalytic degradation of AB 113 in the UV-C/TiO2 system. Figure 3. Response surface and contour plots for photocatalytic degradation eﬃciency of AR 88 as a function of (a) C: initial dye concentration (mg/L) and B: TiO2 dose (g/L) (pH = 3.0, reaction time = 90 min), (b) A: pH and B: TiO2 dose (g/L) (initial dye concentration = 50 mg/L, reaction time = 90 min) and (c) A: pH and C: initial dye concentration (mg/L) (TiO2 dose = 1.0 g/L, reaction time = 90 min). Table 1. Analysis of variance for modiﬁed cubic model obtained for photocatalytic degradation of AB Table 1. Analysis of variance for modiﬁed cubic model obtained for photocatalytic degradation of AB 113 in the UV-C/TiO2 system. Table 1. Analysis of variance for modiﬁed cubic model obtained for photocatalytic degradation of AB 113 in the UV-C/TiO2 system. Source F-Statistic Value p-Value Model 86.33 <0.0001 A-pH 129.56 <0.0001 B-TiO2 dose (g/L) 18.05 0.0005 C-Initial dye concentration (mg/L) 44.81 <0.0001 AB 4.11 0.0578 AC 4.58 0.0462 A2 20.69 0.0002 B2 50.53 <0.0001 A2C 5.13 0.0360 AB2 4.56 0.0467 AC2 5.87 0.0261 B3 14.68 0.0012 Lack of Fit 0.51 0.8499 80 80 Catalysts 2019, 9, 360 Table 2. Analysis of variance for modiﬁed cubic model obtained for photocatalytic degradation of AR 88 in the UV-C/TiO2 system. Source F-Statistic Value p-Value Model 217.009 <0.0001 A-pH 0.773 0.392 B-TiO2 dose (g/L) 14.338 0.0016 C-Initial dye concentration (mg/L) 92.905 <0.0001 AC 9.611 0.0069 BC 3.301 0.088 B2 90.316 <0.0001 ABC 4.961 0.0406 AB2 6.835 0.0188 AC2 6.427 0.0221 B2C 5.822 0.0282 A3 3.379 0.0847 B3 10.334 0.0054 C3 3.466 0.0811 Lack of Fit 1.312 0.4046 Figure 2 shows that removal eﬃciencies greater than 90% for AB 113 occurred for conditions of pH between 2.0 and 3.0, TiO2 dose of 0.7 to 2.0 g/L and dye concentration of 20 to 65 mg/L. 2.2.1. Response Surface Plots, Fitted Models and ANOVA Since the reaction between hydroxide ions in the solution and holes on the surface of TiO2 particles could generate hydroxyl radicals, an alkaline environment could be favorable for hydroxyl radicals’ generation. However, the electrostatic repulsion between the negatively charged surface of TiO2 and OH−anions in an alkaline environment would hinder the formation of hydroxyl radicals leading to a reduced degradation eﬃciency [44]. Venkatachalam et al. [45] reported that an acidic environment is beneﬁcial for photocatalytic degradation by TiO2, since it minimizes electron-holes recombination and enhances •OH production. In addition, TiO2 particles agglomerate in alkaline conditions, leading to a reduced 81 Catalysts 2019, 9, 360 exposed surface area to the energy source (UV-C light) [46]. This could be another reason for the decreased removal eﬃciency of both dyes at higher pH values. Mohammadzadeh et al. (2015) [38] investigated the photodegradation of AB 113 using ZnO-Ag catalyst under UV illumination. They reported that although there is a stronger electrostatic attraction force between ZnO-Ag and AB 113 at 0.5 < pH < 9.0, which is favorable for photocatalytic degradation reactions, since the catalyst dissolves at pH < 3.0, the overall decolorization was enhanced at higher pH values. Ma et al. (2011) [47] studied photooxidation of three azo dyes including AB 113 using TiO2/H2O2 under vacuum ultraviolet (VUV, λ < 190 nm) irradiation. Consistent with the results of the present study, they found that lower pH values (in their experimental range of 3.0 < pH < 9.0) enhanced AB 113 photodegradation. Figure 3 shows that AR 88 removal eﬃciencies exceeding 90% occurred for dye concentrations between 20 and 53 mg/L and TiO2 doses between 1.0 and 2.0 g/L. In the studied range, pH (term A) was not found to be a signiﬁcant factor in AR 88 removal eﬃciency. Increasing initial AR 88 dye concentration (term C) reduced its degradation eﬃciency. Table 2 and Equation (7) indicate high interactions between all three factors (AC, BC, ABC, AB2, AC2 and B2C). Presence of interaction terms between A and C (i.e., pH and initial dye concentration) in the regression model suggested that these two factors were not completely independent. Hence, it could be inferred that changing pH in a wider range probably would show more intense eﬀects on the response. 2.2.1. Response Surface Plots, Fitted Models and ANOVA 82 Catalysts 2019, 9, 360 Table 3. Summary of ﬁtted models’ characteristics for photocatalytic degradation of AB 113 and AR 88 textile dyes in UV-C/TiO2 system, before and after removing insigniﬁcant terms. Item AB 113 AR 88 Initial Cubic Model Reduced Cubic Model Initial Cubic Model Reduced Cubic Model Standard deviation 3.90 3.78 2.35 2.27 Mean 64.24 64.24 65.35 65.35 Coeﬃcient of variation, % 6.06 5.88 3.60 3.47 PRESS 1205.01 664 1780.39 265.73 R2 0.9890 0.9814 0.9962 0.994 R2adj 0.9682 0.9700 0.9890 0.990 Adequate precision 26.392 34.834 33.969 42.061 Adequate precision was obtained 34.834 and 42.061 for AB 113 and AR 88, respectively. Values greater than 4 for this factor are desirable. High values denote an adequate signal and show that the model can navigate the design space [23]. Comparing values before and after models’ modiﬁcation, it is observed in Table 3 that the modiﬁcations enhanced the signal-to-noise ratio for both dyes, reﬂected as increased adequate precision values. For a speciﬁc model, a lower value for the Predicted Residual Error Sum of Squares (PRESS, see Equation (13) in Section 3.4) is favorable, showing that the model is not overly sensitive to any single data point [49]. Table 3 shows that PRESS values decreased by 81.5% and 567% after modiﬁcation of AB 113 and AR 88 models, respectively, indicating that the cubic models were improved by removing statistically insigniﬁcant data points. 2.2.1. Response Surface Plots, Fitted Models and ANOVA This could also be explained by the electrostatic repulsion between negatively charged surface of TiO2 particles in alkaline environment and negatively charged sulfonic groups (R-SO3−) present in structure of AR 88 [3,43]. Therefore, increased negative surface charge of TiO2 particles due to increased pH could have hindered adsorption of AR 88 onto catalyst surface and consequently, reduced the removal eﬃciency. However, a pKa of 10.7 for AR 88 [48] suggests that AR 88 was not highly ionized in the pH range used in this study. Thus, pH eﬀects on degradation of AR 88 were not very signiﬁcant. Similar to AB 113, changing TiO2 dose (term B) showed an optimum point for AR 88 photodegradation beyond which increasing catalyst dose reduced removal eﬃciency. For AR 88, the pH parameter (term A) was kept in the model because of the signiﬁcant interaction eﬀects between pH and other factors (i.e., AC, AC2, ABC, AB2). The obtained models’ p-values of < 0.0001 demonstrated signiﬁcance of cubic models for both dyes. High lack-of-ﬁt p-values of 0.8499 and 0.4046 for AB 113 and AR 88, respectively, conﬁrmed that both reduced cubic models were statistically signiﬁcant. High calculated F-statistic values and correspondingly low associated p-values for each retained parameter (Tables 1 and 2) indicated highly signiﬁcant eﬀects of each retained speciﬁc parameter or combination of parameters on removal eﬃciency. The potency of experimental variables on changing removal percentages could be graded as pH > initial dye concentration > TiO2 dose for AB 113 and initial dye concentration > TiO2 dose > pH for AR 88. In addition, interaction between pH and the second power of initial dye concentration for AB 113 (AC2) was found to be the most signiﬁcant interaction, while for AR 88 the interaction between pH and initial dye concentration for (AC) was the most signiﬁcant interaction (See Figure S1). A summary of the ﬁtted models’ statistical characteristics for two studied dyes, as well as for the cubic models before modiﬁcation is shown in Table 3. Values of R2adj of 0.9700 for AB 113 and 0.9898 for AR 88 (Table 3) indicate that both reduced models could describe a very large portion of the variance in the design space. Table 3 shows coeﬃcients of variation (standard deviation/mean) of 5.88% and 3.47% for AB 113 and AR 88, respectively, meaning that standard deviations were 5.88% and 3.47% of the mean, respectively. 2.2.2. Optimization In order to verify the accuracy of the reduced cubic models in predicting optimum treatment condition, a third round of experiments was carried out under optimum conditions. Predicted optimum operating conditions and removal eﬃciencies as well as the obtained experimental results are shown in Table 4. Removal eﬃciencies of 98.7% and 99.6% under optimum conditions were achieved for AB 113 and AR 88 dyes, respectively; values reasonably close to 100% removal eﬃciencies predicted by the reduced cubic models. Table 4. Optimum conditions for photocatalytic degradation of AB 113 and AR 88 dyes in UV-C/TiO2 suspension system. Table 4. Optimum conditions for photocatalytic degradation of AB 113 and AR 88 dyes in UV-C/TiO2 suspension system. suspension system. pH Initial Dye Concentration (mg/L) TiO2 Dose (g/L) Predicted Removal Eﬃciency (%) Achieved Removal Eﬃciency (%) AB 113 2.21 43.13 0.98 100% 98.7% AR 88 2.36 22.40 1.22 100% 99.6% Regarding the practical applications of the optimized conditions, it should be noted that although highly acidic conditions—causing high operational costs—were proposed for the complete degradation of both AB 113 and AR 88, Figures 2 and 3 illustrate that high removal eﬃciencies could be achieved in a wider range of operating conditions. For example, having an initial dye concentration of 50 mg/L and a reaction time of 90 min, removal eﬃciencies of almost >80% could be achieved with pH increased to 4.0 and 5.0 for AB 113 and AR 88, respectively. 2.3. Stage 3: Kinetics of Photocatalytic Degradation k1 (min−1) p-Value for k1 Co, model p-Value for ln Co R2 RMSE (mg/L) AB 113 0.048 2.13 × 10−7 41 1.13 × 10−8 0.996 1.72 AR 88 0.059 8.60 × 10−7 20.4 3.70 × 10−7 0.993 1.12 High coeﬃcients of determination (R2AB 113 = 0.996 R2AR 88 = 0.993) and low root mean square error (RMSE) values (RMSEC, AB 113 = 1.72 and RMSEC, AR 88 = 1.12) between the ﬁrst order kinetic models and experimental values shown in Table 5 demonstrated that ﬁrst-order kinetic models were appropriate for observed dye degradation under optimum conditions. AR 88 showed a higher degradation rate compared to AB 113 (k1, AR 88 = 0.059 min−1 with a p-value of 8.60 × 10−7 compared to k1, AB 113 = 0.048 min−1 with a p-value of 2.13 × 10−7) under optimum conditions. This is consistent with the preliminary results shown in Figure 1b indicating a higher removal percentage for AR 88 compared to AB 113 at all reaction times. Ma et al. (2011) [47] reported a pseudo-ﬁrst order rate constant of k = 0.2469 min−1 at pH = 3.0 as the highest degradation rate for AB 113 with an initial dye concentration of 0.0523 mM in the studied pH range of 3.0 to 11.0, using a VUV/TiO2 system, with the VUV lamp immersed in the dye solution and TiO2 dose of 0.5 g/L. After 60 min, 60% of AB 113 was decomposed. The higher reaction rate obtained by Ma et al. (2011) [47] compared to the present study could be due to the application of VUV, with a lower wavelength and thus higher energy compared to UV-C, which potentially enhance the excitation of TiO2 particles. In addition, immersing the VUV lamp inside the dye solution using a quartz tube also provides a better exposure of catalyst particles to the energy source. The present study proposed a more economical approach by using a higher wavelength UV-C (meaning a lower energy, primary emission band 254 nm) source and eliminating the use of quartz-tube through direct radiation of UV on the solution surface. Mohammadzadeh et al. (2015) [38] obtained a pseudo-ﬁrst-order rate constant of 0.007 min−1 for AB 113 photodegradation in a ZnO-Ag/UV system with an immersed UV lamp, under the conditions of initial dye concentration = 40 mg/L, catalyst dose = 0.15 g/L and optimum pH = 8.0. 2.3. Stage 3: Kinetics of Photocatalytic Degradation 2.3. Stage 3: Kinetics of Photocatalytic Degradation In order to evaluate photocatalytic degradation rates of AB 113 and AR 88, kinetic studies were performed for each dye under the optimum experimental conditions. Results are shown in Figure 4 83 Catalysts 2019, 9, 360 and reaction rate constants and model characteristics associated with ﬁtted kinetic models are shown in Table 5. Figure 4. First-order kinetic models for photocatalytic degradation of (a) AB 113 and (b) AR 88 dyes in UV/TiO2 suspension system under optimized conditions for each dye (AB 113: initial dye concentration = 43.13 mg/L, reaction time = 90 min, TiO2 dose = 0.98 g/L, pH = 2.2; AR 88: initial dye concentration = 22.40 mg/L, reaction time = 90 min, TiO2 dose = 1.22 g/L, pH = 2.4). Error bars show standard deviation. Table 5. Characteristics of First order kinetic models for photocatalytic degradation of AB 113 and AR 88 in a UV-C/TiO2 suspension system. Figure 4. First-order kinetic models for photocatalytic degradation of (a) AB 113 and (b) AR 88 dyes in UV/TiO2 suspension system under optimized conditions for each dye (AB 113: initial dye concentration = 43.13 mg/L, reaction time = 90 min, TiO2 dose = 0.98 g/L, pH = 2.2; AR 88: initial dye concentration = 22.40 mg/L, reaction time = 90 min, TiO2 dose = 1.22 g/L, pH = 2.4). Error bars show standard deviation. Table 5. Characteristics of First order kinetic models for photocatalytic degradation of AB 113 and AR 88 in a UV-C/TiO2 suspension system. Figure 4. First-order kinetic models for photocatalytic degradation of (a) AB 113 and (b) AR 88 dyes in UV/TiO2 suspension system under optimized conditions for each dye (AB 113: initial dye concentration = 43.13 mg/L, reaction time = 90 min, TiO2 dose = 0.98 g/L, pH = 2.2; AR 88: initial dye concentration = 22.40 mg/L, reaction time = 90 min, TiO2 dose = 1.22 g/L, pH = 2.4). Error bars show standard deviation. Table 5. Characteristics of First order kinetic models for photocatalytic degradation of AB 113 and AR 88 in a UV-C/TiO2 suspension system. Figure 4. First-order kinetic models for photocatalytic degradation of (a) AB 113 and (b) AR 88 dyes in Table 5. Characteristics of First order kinetic models for photocatalytic degradation of AB 113 and AR 88 in a UV-C/TiO2 suspension system. 2.3. Stage 3: Kinetics of Photocatalytic Degradation After 90 min, almost 50% of AB 113 degraded. Their lower degradation rate compared to the present study might be due to the application of a diﬀerent catalyst at a lower dose. Anandan et al. (2008) [11] studied the photocatalytic degradation of AR 88 using Ag-loaded TiO2 particles (Ag/TiO2) under visible light and compared the photodegradation rates with using unloaded 84 Catalysts 2019, 9, 360 TiO2. They obtained ﬁrst-order rate constants of about 0.006 min−1 and 0.008 min−1 for TiO2 and Ag/TiO2, respectively, using an initial AR 88 concentration of 0.034 mg/L, a TiO2 dose of 0.6 g/L, with no pH adjustments. After 425 min, 55% TOC removal was observed using Ag/TiO2. Konyar et al. (2017) [50] studied photocatalytic degradation of AR 88 using sintered-reticulated ZnO catalyst under UV-A and UV-C radiations, in a quartz tube reactor surrounded by a cylindrical light assembly. They obtained pseudo-ﬁrst order rate constants of about 0.007 min−1 and 0.009 min−1 for photodegradation under UV-A and UV-C radiations, respectively, having initial AR 88 concentration of 50 mg/L and catalyst dose of 40 g/L, without pH adjustment. After 180 min, 60% and 80% color removal percentages were obtained under UV-A and UV-C radiation, respectively. p y As shown in Figure 5, AB 113 is a diazo naphthyl dye and AR 88 is a mono-azo naphthyl dye, having conjugated chromophores responsible for their color. When TiO2 is added to the dye solutions, AB 113 and AR 88 molecules are adsorbed mainly through their sulfonate groups [51]. The main degradation pathway proposed by previous researchers studying photodegradation of naphthyl azo dyes in AOP systems [51,52] is the attack of hydroxyl radicals to the naphthalene ring, forming a hydroxylated naphthyl azo dye which is subsequently cleaved. Additionally, hydroxyl radicals attack the aromatic rings with azo groups resulting in azo bond cleavage. Both these reactions result in chromophoric group destruction [52]. Mohammadzadeh et al. (2015) [38] investigated degradation pathway and reaction byproducts for AB 113 photodegradation using a ZnO-Ag nanophotocatalyst under UV radiation. They showed that cleavage of azo bond during a 90 min photocatalysis reaction resulted in the formation of 4-diazenyl-1-naphthylamine, 1-naphthyldiazene or 5-diazenyl-1-naphthol intermediate compounds, which were gradually converted to CO2 and H2O [38]. Madhavan et al. 3. Materials and Methods 3.1. Materials and Equipment 2.3. Stage 3: Kinetics of Photocatalytic Degradation (2010) [53] proposed a pathway for the TiO2 mediated photocatalytic degradation of AR 88 by investigating reaction intermediates using a mass spectrometer and showed formation of hydroxyamino naphthol (4) and 4-aminonaphthalene sulfonic acid (5) as the intermediate products. In the proposed pathway, they showed that AR 88 photocatalytic degradation was mainly due to the hydroxyl radical attack to the aromatic rings [53]. It is expected that the mechanism of hydroxyl radicals attack to AB 113 and AR 88 molecules in the UV-C/TiO2 system to be similar to the mechanisms proposed by Mohammadzadeh et al. (2015) [38] and Madhavan et al. (2010) [53], consisting of cleavage of azo bonds. Figure 5. Molecular structure of (a) Acid Red 88, pKa = 10.7 and (b) Acid Blue 113, pKa = 0.5. Figure 5. Molecular structure of (a) Acid Red 88, pKa = 10.7 and (b) Acid Blue 113, pKa = 0.5. 3.1. Materials and Equipment Acid Red 88 (AR 88) (also known as Fast Red A or 2-Naphthol Red; CAS number 1658-56-6; molecular formula C20H13N2NaO4S; molecular weight 400.38 g/mole) and Acid Blue 113 (AB113) (Fast Navy Blue 5R; CAS number 3351-05-1; molecular formula C32H21N5Na2O6S2; molecular weight 681.65 g/mole) were purchased from Sigma-Aldrich (St. Louis, MO, USA) (dye content 75%). Both of these dyes are common azo acid dyes, usually applied for wool, nylon, rayon and polyester dyeing [54]. Acid dyes are negatively charged dyes [55], which are protonated in pH values below their acid dissociation constant (pKa). The chemical structure of AR 88 and AB 113 are shown in Figure 5. Titanium dioxide (TiO2) Degussa P25 with an average particle size of 30 nm [14] and surface area of 57 m2/g [11] was purchased from Merck (Kenilworth, NJ, USA) (reagent grade) and used as received. 85 Catalysts 2019, 9, 360 Catalysts 2019, 9, 360 Solutions of 1 M, 0.1 M and 0.01 M of HCl and NaOH were used to adjust solution pH to pre-determined values before initialization of photocatalytic process. Standard buﬀer solutions of pH of 4.0 and 7.0 were used to calibrate the pH meter (Jenway, staﬀordshire, UK) 3045 Ion Analyzer pH meter with a Sentek (Stepney, Australia) single-junction, glass body combination electrode ﬁlled with AgCl). To separate TiO2 particles from treated solutions, 8 mL of treated dye solutions were poured in 15 mL-polypropylene centrifuge tubes (17 mm × 120 mm). A Sigma (St. Louis, MO, USA) 201 centrifuge machine was used at 4000× g rpm for 40 min to separate particles. The supernatant was then decanted and used for analysis. Laboratory scales [Sartorius-AC 121S-00MS (Göttingen, Germany) and Rad Wag-WTB 3000 (Radom, Poland)] with the resolution of 0.001 g were used to measure the mass of dyes and TiO2 particles. All the experiments were carried out using DI water (with an electrical resistivity of 1 MOhm/cm at 25 ◦C). A Rayleigh (Beijing, China) UV1601 UV/VIS spectrophotometer was calibrated against standard dye solution concentrations and used to measure the dye concentration. 3.2. Photoreactor Photocatalytic degradation experiments were conducted in a batch reactor, shown schematically in Figure 6. The reactor setup consisted of two UV-C lamps [each lamp: Philips (Somerset, NJ, USA) TUV G30T8 25PK; 30 W, 0.37 A, 102 V; primary emission 253.7 nm, UV-C radiation 12 W, 10% depreciation during 9000 h; 90 cm length, 28 mm diameter], two 4.5-volt rotary agitators, an aeration pump [Hailea (Guangdong, China) ACO 5505, 6 Watt, air output = 5.5 L/min] with two output tubes and two cylindrical dishes (Schott, Germany) for holding dye solutions, with an inside diameter of 13 cm, height of 7.5 cm and bottom thickness of 5 mm. UV lamps were placed on two concrete columns with a height of 25 cm on the top of dye solutions’ containers. Distance from the UV lamps to the surface of dye solutions was 20.7 cm. To ensure a homogenous stirring of catalyst particles in dye solutions during the photocatalytic process, samples were agitated at 150 rpm using rotary agitators. Aeration pump tubes were placed in sample containers to supply the oxygen demand for photocatalytic reaction with the oxygen ﬂow rate of 3.9 × 10−2 mole/min, as well as to achieve a uniform suspension of TiO2 particles in the aqueous solutions. To prevent UV-C radiation leakage, the reactor setup was covered with a cardboard box of 30 cm × 40 cm × 110 cm dimensions. The internal surface of the box was completely covered with aluminum foil to prevent escape of UV radiation. By reﬂecting the radiation toward the samples, the removal eﬃciency would likely be enhanced. Figure 6. Schematic of the UV-C/TiO2 photoreactor set up: (a) UV-C lamps, (b) agitators, (c) aeration pump, (d) sample containers (e) photoreactor cover. Figure 6. Schematic of the UV-C/TiO2 photoreactor set up: (a) UV-C lamps, (b) agitators, (c) aeration pump, (d) sample containers (e) photoreactor cover. 3.3. Experimental Procedure and Measurements 3.3. Experimental Procedure and Measurements For dye concentrations ranging between 20 to 200 mg L−1, the light absorption versus dye concentration plots at the peak of each dye’s absorption spectrum were linear for both dyes (Figure S2). The extinction coeﬃcient (ε) showed values of 0.20 L mg−1 cm−1 for AR 88 and 0.21 L mg−1 cm−1 for AB 113 (Figure S1). where Io/I is the ratio of incident light to transmitted light, A is light absorbance, ε is the molar absorption coeﬃcient (L mg−1 cm−1), L (cm) is the length of solution that light passes through, which is equal to cell thickness used in spectrophotometer and C is the concentration of solution (mg/L) [57]. For dye concentrations ranging between 20 to 200 mg L−1, the light absorption versus dye concentration plots at the peak of each dye’s absorption spectrum were linear for both dyes (Figure S2). The extinction coeﬃcient (ε) showed values of 0.20 L mg−1 cm−1 for AR 88 and 0.21 L mg−1 cm−1 for AB 113 (Figure S1). Degradation eﬃciency, R (%), was calculated using Equation (10): R(%) = Ci −C f Ci × 100 (10) (10) where Ci and Cf are initial and ﬁnal dye concentrations (mg/L). where Ci and Cf are initial and ﬁnal dye concentrations (mg/L). 3.3. Experimental Procedure and Measurements Fresh dye solutions were prepared by adding appropriate amounts of dye powder and DI water in 500-mL volumetric ﬂasks. The solutions were agitated for 10 min on a rotary shaker to obtain a homogenous dye solution and then poured into the cylindrical reaction dishes. Dye solution pH was 86 Catalysts 2019, 9, 360 adjusted to the predetermined levels and the predetermined masses of TiO2 powder were added to the solutions. The suspensions were immediately placed in the photoreactor and the mechanical agitators and air pump were simultaneously turned on sand worked in the dark for 5 min. After 5 min, the two UV lamps were turned on to initiate the photocatalytic reactions. Reaction time was measured from the beginning of UV irradiation. All the experiments were carried out at room temperature (23 ± 0.1 ◦C). To monitor the eﬀectiveness of the process, light absorbance of the samples was measured by the UV-VIS spectrophotometer at characteristic wavelengths of 505 nm and 565 nm for AR 88 and AB 113, respectively [16,56]. Dye concentrations in treated samples were determined from measured light extinction based on Beer–Lambert’s law as expressed below: adjusted to the predetermined levels and the predetermined masses of TiO2 powder were added to the solutions. The suspensions were immediately placed in the photoreactor and the mechanical agitators and air pump were simultaneously turned on sand worked in the dark for 5 min. After 5 min, the two UV lamps were turned on to initiate the photocatalytic reactions. Reaction time was measured from the beginning of UV irradiation. All the experiments were carried out at room temperature (23 ± 0.1 ◦C). To monitor the eﬀectiveness of the process, light absorbance of the samples was measured by the UV-VIS spectrophotometer at characteristic wavelengths of 505 nm and 565 nm for AR 88 and AB 113, respectively [16,56]. Dye concentrations in treated samples were determined from measured light extinction based on Beer–Lambert’s law as expressed below: log10 (I/Io) = A (8) A = ε·L·C (9) log10 (I/Io) = A (8) A = ε·L·C (9) where Io/I is the ratio of incident light to transmitted light, A is light absorbance, ε is the molar absorption coeﬃcient (L mg−1 cm−1), L (cm) is the length of solution that light passes through, which is equal to cell thickness used in spectrophotometer and C is the concentration of solution (mg/L) [57]. 3.5.1. Preliminary Experiments Experiments and optimization were performed in three stages. In the ﬁrst stage, preliminary experiments were conducted to determine the ranges of experimental factors to be used in the main experiments. In the second stage, the main experiments were conducted to determine optimal experimental conditions using the response surface methodology. Finally, in the third stage, reaction kinetics were investigated under optimized operational conditions. Four independent factors including pH, initial dye concentration, TiO2 dose and reaction time were used to evaluate dye removal eﬃciencies in the preliminary ﬁrst-stage experiments. Adopting a one factor-at-a-time approach, three out of four variables were held constant and the fourth was varied in 4 or 5 levels. Table 6 shows variables and levels used in the preliminary experiments. Two replicate runs were conducted for each combination. Table 6. Experimental factors and levels used in preliminary experiments for photocatalytic degradation of AB 113 and AR 88 in the UV-C/TiO2 system. Table 6. Experimental factors and levels used in preliminary experiments for photocatalytic degradation of AB 113 and AR 88 in the UV-C/TiO2 system. Variables Levels pH 2.0 3.0 * 6.0 9.0 10.0 Initial dye concentration (mg/L) 20 50 * 100 150 200 TiO2 dose (g/L) 0.5 1.0 * 2.0 4.0 - Reaction time (min) 30 60 90 * 120 180 * Fixed value of variable when other factors changed. * Fixed value of variable when other factors changed. 3.4. Statistical Analysis 3.4. Statistical Analysis Apart from R2 and standard deviation as the two well-known statistical analysis measures, analysis of variance (ANOVA) uses other standard factors to evaluate signiﬁcance of a ﬁtted regression model to a data set. These factors include coeﬃcient of variation, adequate precision and predicted residual error sum of squares. Coeﬃcient of variation is the standard deviations which is expressed as the mean percentage: Coeﬃcient of variation (%) =Standard deviation mean ×100 (11) (11) Adequate precision is an indicator for measuring signal to noise ratio of the model, which is calculated as: Adequate precision is an indicator for measuring signal to noise ratio of the model, which is calculated as: maximum predicted response−minimum predicted response Average standard deviation of all predicted response (12) Adequate precision =maximum predicted response−minimum predicted response Average standard deviation of all predicted response (12) (12) Predicted residual error sum of squares (PRESS) is a measure between the ﬁtted values and observed values. From a ﬁtted model, each observation from the data set is removed, the model is reﬁtted and the predicted value at that excluded point is calculated. The PRESS is calculated as: [50] PRESS = n i=1 (yi −ˆy−i)2 (13) (13) where n is the number of data points, yi is the outcome of ith data point and the ˆy−i is the prediction of ith data point from the reﬁtted model excluded ith data [58]. 87 87 Catalysts 2019, 9, 360 Catalysts 2019, 9, 360 The F-test in ANOVA investigates if the variance between the means of two populations are signiﬁcantly diﬀerent. The F-statistic is the ratio of the “between-group variability” to the “within-group variability”, or: variability”, or: F = variation between sample means variation within the samples (14) F = variation between sample means variation within the samples (14) (14) The p-value tests the null hypothesis which expresses that data from all groups are from populations with equal means. In other words, p-value determines that if all the populations really have the same mean, what is the chance that random sampling would result in the means as far apart as observed. The p-value is computed from a comparison of the computed F-statistic to the critical value of the F-statistic for the given number of degrees of freedom. If the null hypothesis of no signiﬁcant diﬀerence between sample manes is true, the F-statistic is expected to be close to 1. A large F-statistic means that the variation among the group means is more than is expected to occur by chance. Therefore, a large F-statistic, if it exceeds the critical F-statistic for a pre-established level of signiﬁcant (typically p < 0.05) can lead to the rejection of the null hypothesis, meaning that the data were not likely to have been sampled from populations with the same mean. 3.5. Experimental Design and Optimization 3.5.3. Kinetic Studies In the third stage, additional experiments were conducted to investigate reaction kinetics under the optimized treatment conditions determined from the cubic IV optimal experimental design. A ﬁrst-order kinetic model can be described as: In the third stage, additional experiments were conducted to investigate reaction kinetics under the optimized treatment conditions determined from the cubic IV optimal experimental design. A ﬁrst-order kinetic model can be described as: ln (C) = ln (Co) −k1t (15) (15) where, t is the reaction time (min), Co and C (mg/L) are the initial dye concentration (mg/L) and dye concentration at time t, respectively and k1 is the ﬁrst-order reaction rate constant (min−1). When evaluating goodness of ﬁt for a kinetic model, root mean square error (RMSE) (Equation (16)) was used along with R2 value to evaluate model validity [60]. where, t is the reaction time (min), Co and C (mg/L) are the initial dye concentration (mg/L) and dye concentration at time t, respectively and k1 is the ﬁrst-order reaction rate constant (min−1). When evaluating goodness of ﬁt for a kinetic model, root mean square error (RMSE) (Equation (16)) was used along with R2 value to evaluate model validity [60]. RMSE = n i=1 Cm −Cexp 2 n (16) (16) where Cexp and Cm are the experimental and calculated values (based on the ﬁtted kinetic models) of dye concentration and n is the number of data points. 3.5.2. Experimental Design Using Response Surface Methodology Experimental design, statistical analyses, mathematical modeling and optimizations were accomplished using Design Expert software (Design-Expert®, V 10, Stat-Ease, Inc., Minneapolis, MN, USA). Similar to the paper by Saber et al. [26], a cubic IV optimal design method was employed to investigate the eﬀects of input factors and their interactions on dye removal percentage (i.e., the target response), in the second experimental stage. Cubic IV optimal design minimizes the integral of prediction variance through the design space and results in a lower prediction variance throughout an area of interest [59]. Ranges of experimental variables were considered based on the results from preliminary experiments. As Table 7 shows, six levels of pH, six levels of TiO2 dose and six levels of initial dye concentration were considered as the independent variables in the optimal design. For each 88 Catalysts 2019, 9, 360 dye, 30 combinations of conditions were developed according to the cubic IV optimal design algorithm. The achieved removal eﬃciency for each run was considered as the target response. In order to account for experimental errors, three replicate runs were conducted for each combination of conditions and the average removal eﬃciency of the three runs was reported for each experimental run. Table 7. Experimental factors and their levels used in cubic IV optimal design for photocatalytic degradation of AB 113 and AR 88. Table 7. Experimental factors and their levels used in cubic IV optimal design for photocata degradation of AB 113 and AR 88. able 7. Experimental factors and their levels used in cubic IV optimal design for photocatalytic egradation of AB 113 and AR 88. Factors Levels A: pH 2.0 2.5 3.0 3.5 4.0 5.0 B: TiO2 dose (g/L) 0.5 1.0 1.5 2.0 2.5 3.0 C: Initial dye concentration (mg/L) 20 50 60 80 115 150 Analysis of variance (ANOVA) was performed on the ﬁtted cubic models for each dye to evaluate the signiﬁcance of the ﬁtted models and to identify the relative signiﬁcance of experimental factors and their interactions on the removal eﬃciency for each dye. Three-dimensional response surface and contour plots were generated based on the cubic least-squares regression models obtained from ANOVA. 4. Conclusions This study optimized photodegradation of AB 113 and AR 88 dyes in a UV-C/TiO2 suspension system using RSM, considering initial dye concentration, solution pH and catalyst dose as variant factors and the removal percentage as the target response. Under optimum conditions, kinetics of photocatalytic degradation of AB 113 and AR 88 were also investigated. Analysis of variance showed that reduced cubic models could well describe the removal of AB 113 and AR 88 dyes. The F-test showed that the solution pH and initial dye concentration were the most important parameters for removal of AB 113 and AR 88, respectively. Although pH was the most signiﬁcant parameter aﬀecting AB 113 removal eﬃciency, it was found to be insigniﬁcant for AR 88 removal. However, pH of the AR 88 solution showed signiﬁcant interactions with the other two factors. Degradation eﬃciencies of 98.7% and 99.6% were achieved under optimum conditions for AB 113 and AR 88, respectively. The present study demonstrated almost complete degradation of AB 113 and AR 88 in 90 min under the optimized conditions obtained using RSM, with ﬁrst-order rate constants of degradation rate constants of k1, AB 113 = 0.048 min−1 and k1, AR 88 = 0.059 min−1. 89 Catalysts 2019, 9, 360 Results show that a UV-C/TiO2 photocatalytic degradation process can be considered as a promising and cost-eﬀective technique for dye removal from textile industry eﬄuents. As a proposal, a dyeing plant could adjust conditions in its eﬄuent to the optimized values to obtain more than 98.7% destruction of waste dye and then adjust eﬄuent pH to neutral values and ﬁlter to remove and recycle the TiO2 particles prior to discharge to a receiving water. Additional batch experiments should be conducted at a bench scale in multi dye solutions to evaluate competition between various dyes, followed by pilot-scale application of this treatment method on a real textile wastewater in a ﬂow-through reactor to evaluate limitations due to incomplete mixing and dispersion. 4. Conclusions Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/4/360/s1, Figure S1: Computed F-values for signiﬁcant (p < 0.05) regression model terms associated with photocatalytic degradation of (a) AB 113 (a) and (b) AR 88 [A: pH, B: TiO2 dose (g/L) and C: Initial dye concentration: (mg/L)]., Figure S2: Absorbance versus dye concentrations graphs for (a) AB 113 and (b) AR 88, showing linearity of data for both dyes in the studied dye concentration between 0 mg/L to 200 mg/L. Author Contributions: Conceptualization, S.M. and A.S.; Data curation, S.M. and A.S.; Formal analysis, S.M., A.S. and D.E.J.; Funding acquisition, D.E.J.; Investigation, S.M.; Methodology, S.M. and A.S.; Software, S.M. and A.S.; Supervision, D.E.J.; Validation, S.M. and A.S.; Visualization, S.M.; Writing—original draft, S.M.; Writing—review & editing, A.S. and D.E.J. Author Contributions: Conceptualization, S.M. and A.S.; Data curation, S.M. and A.S.; Formal analysis, S.M., A.S. and D.E.J.; Funding acquisition, D.E.J.; Investigation, S.M.; Methodology, S.M. and A.S.; Software, S.M. and A.S.; Supervision, D.E.J.; Validation, S.M. and A.S.; Visualization, S.M.; Writing—original draft, S.M.; Writing—review & editing, A.S. and D.E.J. Funding: The publication fees for this article were supported by the UNLV University Libraries Open Article Fund. Authors thank Isfahan University of Technology for partially funding (1391-7) this research. Funding: The publication fees for this article were supported by the UNLV University Libraries Open Art Fund. Authors thank Isfahan University of Technology for partially funding (1391-7) this research. Acknowledgments: Authors would like to acknowledge Amir Taebi for his professional advice and Behnaz Harandizadeh for her help in running experiments. The authors greatly appreciate the anonymous reviewers of this manuscript for their constructive suggestions which improved the quality of this study. Acknowledgments: Authors would like to acknowledge Amir Taebi for his professional advice and Behnaz Harandizadeh for her help in running experiments. The authors greatly appreciate the anonymous reviewers of this manuscript for their constructive suggestions which improved the quality of this study. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Sohrabi, M.R.; Ghavami, M. Photocatalytic degradation of Direct Red 23 dye using UV/TiO2: Eﬀect of operational parameters. J. Hazard. Mater. 2008, 153, 1235–1239. [CrossRef] 2. Houas, A.; Lachheb, H.; Ksibi, M.; Elaloui, E.; Guillard, C.; Herrmann, J.-M. Photocatalytic degradation pathway of methylene blue in water. Appl. Catal. B Environ. 2001, 31, 145–157. Available online: https://ac-els-cdn-com.ezproxy.library.unlv.edu/S0926337300002769/1-s2.0-S0926337300002769-main.pd f?_tid=7e7afe96-d229-11e7-85a6-00000aab0f26&acdnat=1511646312_3be9e82c1ae096d6f42aa8199f88814d (accessed on 25 November 2017). [CrossRef] 3. Konicki, W.; Sibera, D.; Mijowska, E.; Lendzion-Bielu´n, Z.; Narkiewicz, U. Equilibrium and kinetic studies on acid dye Acid Red 88 adsorption by magnetic ZnFe2O4 spinel ferrite nanoparticles. J. Colloid Interface Sci. 2013, 398, 152–160. [CrossRef] [PubMed] 4. Gupta, A.K.; Pal, A.; Sahoo, C. Photocatalytic degradation of a mixture of Crystal Violet (Basic Violet 3) and Methyl Red dye in aqueous suspensions using Ag+ doped TiO2. Dyes Pigments 2006, 69, 224–232. [CrossRef] 5. Han, F.; Kambala, V.S.R.; Srinivasan, M.; Rajarathnam, D.; Naidu, R. Tailored titanium dioxide photocatalysts for the degradation of organic dyes in wastewater treatment: A review. Appl. Catal. A Gen. 2009, 359, 25–40. [CrossRef] onstantinou, I.K.; Albanis, T.A. TiO2-assisted photocatalytic degradation of azo dyes in aqueous solution: 6. Konstantinou, I.K.; Albanis, T.A. TiO2-assisted photocatalytic degradation of azo dyes in aqueous solution: Kinetic and mechanistic investigations: A review. Appl. Catal. B Environ. 2004, 49, 1–14. [CrossRef] 6. Konstantinou, I.K.; Albanis, T.A. TiO2-assisted photocatalytic degradation of azo dyes in aqueous solut Kinetic and mechanistic investigations: A review. Appl. Catal. B Environ. 2004, 49, 1–14. [CrossRef] Kinetic and mechanistic investigations: A review. Appl. Catal. B Environ. 2004, 49, 1–14. [CrossRef] 7. Juang, R.S.; Lin, S.H.; Hsueh, P.Y. Removal of binary azo dyes from water by UV-irradiated degradation TiO2 suspensions. J. Hazard. Mater. 2010, 182, 820–826. [CrossRef] 8. Ahmed, S.; Rasul, M.G.; Martens, W.N.; Brown, R.; Hashib, M.A. Heterogeneous photocatalytic degradation of phenols in wastewater: A review on current status and developments. Desalination 2010, 261, 3–18. [CrossRef] 9. Barakat, M.A. Adsorption and photodegradation of Procion yellow H-EXL dye in textile wastewater over TiO2 suspension. J. Hydro-Environ. Res. 2011, 5, 137–142. [CrossRef] 10. Soutsas, K.; Karayannis, V.; Poulios, I.; Riga, A.; Ntampegliotis, K.; Spiliotis, X.; Papapolymerou, G. Decolorization and degradation of reactive azo dyes via heterogeneous photocatalytic processes. Desalination 2010, 250, 345–350. [CrossRef] 90 Catalysts 2019, 9, 360 11. Anandan, S.; Kumar, P.S.; Pugazhenthiran, N.; Madhavan, J.; Maruthamuthu, P. Eﬀect of loaded silver nanoparticles on TiO2 for photocatalytic degradation of Acid Red 88. Sol. Energy Mater. Sol. References Cells 2008, 92, 929–937. [CrossRef] 12. Zayani, G.; Bousselmi, L.; Pichat, P.; Mhenni, F. Photocatalytic degradation of the Acid Blue 113 textile azo dye in aqueous suspensions of four commercialized TiO(2) samples. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 2008, 43, 202–209. [CrossRef] 13. Gupta, V.K.; Jain, R.; Mittal, A.; Saleh, T.A.; Nayak, A.; Agarwal, S.; Sikarwar, S. Photo-catalytic degradation of toxic dye amaranth on TiO2/UV in aqueous suspensions. Mater. Sci. Eng. C 2012, 32, 12–17. [CrossRef] [PubMed] 14. Toor, A.T.; Verma, A.; Jotshi, C.K.; Bajpai, P.K.; Singh, V. Photocatalytic degradation of Direct Yellow 12 dye using UV/TiO2 in a shallow pond slurry reactor. Dyes Pigments 2006, 68, 53–60. [CrossRef] 15. Khataee, A.R.; Pons, M.N.; Zahraa, O. Photocatalytic degradation of three azo dyes using immobilized TiO2 nanoparticles on glass plates activated by UV light irradiation: Inﬂuence of dye molecular structure. J. Hazard. Mater. 2009, 168, 451–457. [CrossRef] [PubMed] 16. Balachandran, K.; Venckatesh, R.; Sivaraj, R.; Rajiv, P. TiO2 nanoparticles versus TiO2-SiO2 nanocomposites: A comparative study of photo catalysis on acid red 88, Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 2014, 128, 468–474. [CrossRef] [PubMed] 17. Gao, B.; Yap, P.S.; Lim, T.M.; Lim, T.T. Adsorption-photocatalytic degradation of Acid Red 88 by supported TiO2: Eﬀect of activated carbon support and aqueous anions. Chem. Eng. J. 2011, 171, 1098–1107. [CrossRef] 18. Moon, J.; Yun, C.Y.; Chung, K.W.; Kang, M.S.; Yi, J. Photocatalytic activation of TiO2 under visible light using Acid Red 44. Catal. Today 2003, 87, 77–86. [CrossRef] 19. Khataee, A.R.; Zarei, M. Photoelectrocatalytic decolorization of diazo dye by zinc oxide nanophotocatalyst and carbon nanotube based cathode: Determination of the degradation products. Desalination 2011, 278, 117–125. [CrossRef] 20. Cavazzuti, M. Optimization Methods: From Theory to Design Scientiﬁc and Technological Aspects in Mechanics; Springer: New York, NY, USA, 2013. [CrossRef] 21. Antony, J. Taguchi or classical design of experiments: A perspective from a practitioner. Sens. Rev. 2006, 26, 227–230. [CrossRef] 22. Khuri, A.I.; Mukhopadhyay, S. Response surface methodology, Wiley Interdiscip. Rev. Comput. Stat. 2010, 2, 128–149. [CrossRef] 23. Myers, R.H.; Montgomery, D.C.; Anderson-Cook, C.M. Response Surface Methodology: Process and Product Optimization Using Designed Experiments, 3rd ed.; John Wiley & Sons: New York, NY, USA, 2009. 24. Wei, L.; Zhu, H.; Mao, X.; Gan, F. Electrochemical oxidation process combined with UV photolysis for the mineralization of nitrophenol in saline wastewater. Sep. Purif. Technol. 2011, 77, 18–25. [CrossRef] 25. References Sahu, J.N.; Acharya, J.; Meikap, B.C. Response surface modeling and optimization of chromium(VI) removal from aqueous solution using Tamarind wood activated carbon in batch process. J. Hazard. Mater. 2009, 172, 818–825. [CrossRef] [PubMed] 26. Saber, A.; Mortazavian, S.; James, D.E.; Hasheminejad, H. Optimization of Collaborative Photo-Fenton Oxidation and Coagulation for the Treatment of Petroleum Reﬁnery Wastewater with Scrap Iron. Water Air Soil Pollut. 2017, 228. [CrossRef] 27. Saber, A.; Hasheminejad, H.; Taebi, A.; Ghaﬀari, G. Optimization of Fenton-based treatment of petroleum reﬁnery wastewater with scrap iron using response surface methodology. Appl. Water Sci. 2014, 4, 283–290. [CrossRef] 28. Zhu, X.; Tian, J.; Liu, R.; Chen, L. Optimization of fenton and electro-fenton oxidation of biologically treated coking wastewater using response surface methodology. Sep. Purif. Technol. 2011, 81, 444–450. [CrossRef] 29. Benatti, C.T.; Tavares, C.R.G.; Guedes, T.A. Optimization of Fenton’s oxidation of chemical laboratory wastewaters using the response surface methodology. J. Environ. Manag. 2006, 80, 66–74. [CrossRef] [PubMed] 30. Ahmadi, M.; Vahabzadeh, F.; Bonakdarpour, B.; Mofarrah, E.; Mehranian, M. Application of the central composite design and response surface methodology to the advanced treatment of olive oil processing wastewater using Fenton’s peroxidation. J. Hazard. Mater. 2005, 123, 187–195. [CrossRef] 31. Bianco, B.; de Michelis, I.; Vegliò, F. Fenton treatment of complex industrial wastewater: Optimization of process conditions by surface response method. J. Hazard. Mater. 2011, 186, 1733–1738. [CrossRef] [PubMed] 91 Catalysts 2019, 9, 360 32. Bagheri, A.R.; Ghaedi, M.; Asfaram, A.; Bazrafshan, A.A.; Jannesar, R. Comparative study on ultrasonic assisted adsorption of dyes from single system onto Fe3O4 magnetite nanoparticles loaded on activated carbon: Experimental design methodology. Ultrason. Sonochem. 2017, 34, 294–304. [CrossRef] [PubMed] 33. Mohajeri, S.; Aziz, H.A.; Isa, M.H.; Zahed, M.A.; Adlan, M.N. Statistical optimization of process parameters for landﬁll leachate treatment using electro-Fenton technique. J. Hazard. Mater. 2010, 176, 749–758. [CrossRef] [PubMed] [ ] 34. Cifuentes, B.; Figueredo, M.; Cobo, M. Response Surface Methodology and Aspen Plus Integration for the Simulation of the Catalytic Steam Reforming of Ethanol. Catalysts 2017, 7, 15. [CrossRef] 35. Li, L.; Ma, Q.; Wang, S.; Song, S.; Li, B.; Guo, R.; Cheng, X.; Cheng, Q. Photocatalytic Performance and Degradation Mechanism of Aspirin by TiO2 through Response Surface Methodology. Catalysts 2018, 8, 118. [CrossRef] 36. Inger, M.; Dobrzy´nska-Inger, A.; Rajewski, J.; Wilk, M. Optimization of Ammonia Oxidation Using Response Surface Methodology. Catalysts 2019, 9, 249. [CrossRef] 37. References Aljuboury, D.a.A.; Palaniandy, P.; Aziz, H.B.A.; Feroz, S.; Amr, S.S.A. Evaluating photo-degradation of COD and TOC in petroleum reﬁnery wastewater by using TiO2/ZnO photo-catalyst. Water Sci. Technol. 2016, 74, 1312–1325. [CrossRef] 38. Mohammadzadeh, S.; Olya, M.E.; Arabi, A.M.; Shariati, A.; Nikou, M.R.K. Synthesis, characterization and application of ZnO-Ag as a nanophotocatalyst for organic compounds degradation, mechanism and economic study. J. Environ. Sci. 2015, 35, 194–207. [CrossRef] [PubMed] 39. Mohan, D.; Pittman, C.U. Activated carbons and low cost adsorbents for remediation of tri- and hexavalent chromium from water. J. Hazard. Mater. 2006, 137, 762–811. [CrossRef] 40. Alkaim, A.F.; Kandiel, T.A.; Hussein, F.H.; Dillert, R.; Bahnemann, D.W. Enhancing the photocatalytic activity of TiO2 by pH control: A case study for the degradation of EDTA. Catal. Sci. Technol. 2013, 3, 3216. [CrossRef] 41. Hu, C.; Yu, J.C.; Hao, Z.; Wong, P.K. Eﬀects of acidity and inorganic ions on the photocatalytic degradation of diﬀerent azo dyes Appl Catal B Environ 2003 46 35–47 [CrossRef] 40. Alkaim, A.F.; Kandiel, T.A.; Hussein, F.H.; Dillert, R.; Bahnemann, D.W. Enhancing the photocatalytic activity of TiO2 by pH control: A case study for the degradation of EDTA. Catal. Sci. Technol. 2013, 3, 3216. [CrossRef] 41 H C Y J C H Z W PK Eﬀ t f idit d i i i th h t t l ti d d ti f y p y g 41. Hu, C.; Yu, J.C.; Hao, Z.; Wong, P.K. Eﬀects of acidity and inorganic ions on the photocatalytic degradation of diﬀerent azo dyes. Appl. Catal. B Environ. 2003, 46, 35–47. [CrossRef] 41. Hu, C.; Yu, J.C.; Hao, Z.; Wong, P.K. Eﬀects of acidity and inorganic ions diﬀerent azo dyes. Appl. Catal. B Environ. 2003, 46, 35–47. [CrossRef] 42. Shirzad-Siboni, M.; Jafari, S.J.; Giahi, O.; Kim, I.; Lee, S.M.; Yang, J.K. Removal of acid blue 113 and reactive black 5 dye from aqueous solutions by activated red mud. J. Ind. Eng. Chem. 2014, 20, 1432–1437. [CrossRef] 43. Lima, E.C.; Royer, B.; Vaghetti, J.C.P.; Simon, N.M.; da Cunha, B.M.; Pavan, F.A.; Benvenutti, E.V.; Cataluña-Veses, R.; Airoldi, C. Application of Brazilian pine-fruit shell as a biosorbent to removal of reactive red 194 textile dye from aqueous solution. Kinetics and equilibrium study. J. Hazard. Mater. 2008, 155, 536–550. [CrossRef] [PubMed] 44. Habibi, M.H.; Vosooghian, H. Photocatalytic degradation of some organic sulﬁdes as environmental pollutants using titanium dioxide suspension. J. Photochem. Photobiol. A Chem. 2005, 174, 45–52. References [CrossRef] 45. Venkatachalam, N.; Palanichamy, M.; Murugesan, V. Sol-gel preparation and characterization of alkaline earth metal doped nano TiO2: Eﬃcient photocatalytic degradation of 4-chlorophenol. J. Mol. Catal. A Chem. 2007, 273, 177–185. [CrossRef] 46. Fox, M.A.; Dulay, M.T. Heterogeneous photocatalysis. Chem. Rev. 1993, 93, 341–357. [CrossRef] 47. Ma, C.-M.; Hong, G.-B.; Chen, H.-W.; Hang, N.-T.; Shen, Y.-S. Photooxidation Contribution Study on the Decomposition of Azo Dyes in Aqueous Solutions by VUV-Based AOPs. Int. J. Photoenergy 2011, 2011, 1–8. [CrossRef] 48. Saharan, V.K.; Pandit, A.B.; Kumar, P.S.S.; Anandan, S. Hydrodynamic cavitation as an advanced oxidation technique for the degradation of Acid Red 88 dye. Ind. Eng. Chem. Res. 2012, 51, 1981–1989. [CrossRef] 49. Spiess, A.N. qpcR package V1.4-0, Modeling and Analysis of Real-Time PCR Data. 2018. Available online: https://www.rdocumentation.org/packages/qpcR/versions/1.4-0 (accessed on 31 May 2018). Ö 50. Konyar, M.; Yildiz, T.; Aksoy, M.; Yatmaz, H.C.; Öztürk, K. Reticulated ZnO Photocatalyst: Eﬃciency Enhancement in Degradation of Acid Red 88 Azo Dye by Catalyst Surface Cleaning. Chem. Eng. Commun. 2017, 204, 711–716. [CrossRef] 51. Camarillo, R.; Rincón, J. Photocatalytic Discoloration of Dyes: Relation between Eﬀect of Operating Parameters and Dye Structure. Chem. Eng. Technol. 2011, 34, 1675–1684. [CrossRef] 52. Stylidi, M.; Kondarides, D.I.; Verykios, X.E. Visible light-induced photocatalytic degradation of Acid Ora 7 in aqueous TiO2 suspensions. Appl. Catal. B Environ. 2004, 47, 189–201. [CrossRef] 53. Madhavan, J.; Kumar, P.S.S.; Anandan, S.; Grieser, F.; Ashokkumar, M. Degradation of acid red 88 by the combination of sonolysis and photocatalysis. Sep. Purif. Technol. 2010, 74, 336–341. [CrossRef] 92 Catalysts 2019, 9, 360 54. Trotma, E.R. Dyeing and Chemical Technology of Textile Fibres; Griﬃn: London, UK, 1970; Available online: https://www.scribd.com/doc/101550453/Dyeing-and-Chemical-Technology-of-Textile-Fibres (accessed on 3 May 2017). 55. Martínez-Huitle, C.A.; Brillas, E. Decontamination of wastewaters containing synthetic organic dyes by electrochemical methods: A general review. Appl. Catal. B Environ. 2009, 87, 105–145. [CrossRef] 56. Shu, H.Y.; Chang, M.C.; Chen, C.C.; Chen, P.E. Using resin supported nano zero-valent iron particles for decoloration of Acid Blue 113 azo dye solution. J. Hazard. Mater. 2010, 184, 499–505. [CrossRef] [PubMed] 57. Meyers, R. Encyclopedia of Physical Science and Technology; Academic Press: New York, NY, USA, 2002. 58. Zumel, N. Estimating Generalization Error with the PRESS statistic\|Win-Vector Blog, (2014) 12. Available online: http://www.win-vector.com/blog/2014/09/estimating-generalization-error-with-the-press-statistic/ (accessed on 2 April 2019). 59. Anderson, M.J.; Whitcomb, P.J. RSM Simpliﬁed: Optimizing Processes Using Response Surface Methods for Design of Experiments; Productivity Press: New York, NY, USA, 2005. Article Facet-Dependent Interfacial Charge Transfer in TiO2/Nitrogen-Doped Graphene Quantum Dots Heterojunctions for Visible-Light Driven Photocatalysis Keywords: electron transfer; graphene quantum dots; heterojunction; photocatalysis; TiO2 Article Facet-Dependent Interfacial Charge Transfer in TiO2/Nitrogen-Doped Graphene Quantum Dots Heterojunctions for Visible-Light Driven Photocatalysis Nan-Quan Ou 1,†, Hui-Jun Li 1,†, Bo-Wen Lyu 1, Bo-Jie Gui 1, Xiong Sun 1, Dong-Jin Qian 2, Yanlin Jia 3,4,, Xianying Wang 1,5, and Junhe Yang 1,5 1 School of Materials Science and Technology, University of Shanghai for Science and Technology, Shanghai 200093, China; ounanquan@163.com (N.-Q.O.); huijunli0701@126.com (H.-J.L.); bowenlyu0324@163.com (B.-W.L.); bojie_gui@163.com (B.-J.G.); sunxiong1993@163.com (X.S.); jhyang@usst.edu.cn (J.Y.) 1 School of Materials Science and Technology, University of Shanghai for Science and Technology, Shanghai 200093, China; ounanquan@163.com (N.-Q.O.); huijunli0701@126.com (H.-J.L.); bowenlyu0324@163.com (B.-W.L.); bojie_gui@163.com (B.-J.G.); sunxiong1993@163.com (X.S.); jhyang@usst.edu.cn (J.Y.) 2 Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, Fudan University, Shanghai 200433, China; djqian@fudan.edu.cn 3 College of Materials Science and Engineering, Central South University, Changsha 410083, China 4 C ll f M t i l S i d E i i B iji U i it f T h l B iji 100124 Chi 3 College of Materials Science and Engineering, Central South University, Changsha 410083, China 4 College of Materials Science and Engineering, Beijing University of Technology, Beijing 100124, China 5 Shanghai Innovation Institute for Materials, Shanghai 200444, China * Correspondence: xianyingwang@usst.edu.cn (X.W.); jiayanlin@126.com (Y.J.) † These two authors contributed equally to this work. Received: 26 February 2019; Accepted: 8 April 2019; Published: 9 April 2019 Abstract: Interfacial charge transfer is crucial in the eﬃcient conversion of solar energy into fuels and electricity. In this paper, heterojunction composites were fabricated, comprised of anatase TiO2 with diﬀerent percentages of exposed {101} and {001} facets and nitrogen-doped quantum dots (NGQDs) to enhance the transfer eﬃciency of photo-excited charge carriers. The photocatalytic performances of all samples were evaluated for RhB degradation under visible light irradiation, and the hybrid containing TiO2 with 56% {001} facets demonstrated the best photocatalytic activity. The excellent photoactivity of TiO2/NGQDs was owed to the synergistic eﬀects of the following factors: (i) The unique chemical features of NGQDs endowed NGQDs with high electronic conductivities and provided its direct contact with the TiO2 surface via forming Ti–O–C chemical bonds. (ii) The co-exposed {101} and {001} facets were beneﬁcial for the separation and transfer of charge carriers in anatase TiO2. (iii) The donor-acceptor interaction between NGQDs and electron-rich {101} facets of TiO2 could remarkably enhance the photocurrent, thus hindering the charge carriers recombination rate. Extensive characterization of their physiochemical properties further showed the synergistic eﬀect of facet-manipulated electron-hole separation in TiO2 and donor-acceptor interaction in graphene quantum dots (GQDs)/TiO2 on photocatalytic activity. Catalysts 2019, 9, 345; doi:10.3390/catal9040345 www.mdpi.com/journal/catalysts References Available online: https: //www.crcpress.com/RSM-Simplified-Optimizing-Processes-Using-Response-Surface-Methods-for/Whit comb-Anderson/p/book/9781563272974 (accessed on 3 May 2017). 60. Saber, A.; Tafazzoli, M.; Mortazavian, S.; James, D.E. Investigation of kinetics and absorption isotherm models for hydroponic phytoremediation of waters contaminated with sulfate. J. Environ. Manag. 2018, 207, 276–291. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 93 93 1. Introduction Anatase TiO2 is generally considered a better photocatalyst than rutile, mainly due to its attributes of longer exciton diﬀusion length, higher electron mobility, and longer carrier life time [1,2]. The photocatalytic activity of anatase has been revealed to depend closely on the crystal surface [3,4]. Clear pictures have now shown that reduction and oxidation reactions would preferentially occur on {101} and {001} facets, respectively. Furthermore, it has been found that the {101} surface is attractive for electrons in aqueous solutions while excess electrons tend to strongly avoid the {001} surface via surface Catalysts 2019, 9, 345; doi:10.3390/catal9040345 95 Catalysts 2019, 9, 345 science experiments and ﬁrst-principles simulations [5]. The substantial electrons in anatase TiO2 generated via photoexcitation play an important part in many energy-related applications. However, due to intrinsic defects and the fast carrier recombination rate, electron trapping in anatase TiO2 are unavoidable, which hamper the overall photocatalytic activity [6,7]. To solve this problem, fabrication of heterojunctions modiﬁed anatase TiO2 has emerged as a promising method [8–10]. Angus and co-workers reported the development of a hetero-structured material by using pre-formed carbon nitride nanosheets (CNNS) composite with facet-controllable TiO2. The materials possess an excellent CO2 adsorption capacity and charge transfer rate, thus leading to the improvement of the photocatalytic activity of TiO2 [11]. Luca Rimoldi et al. have then reported a method to combine TiO2 with WO3. Due to the admirable properties of WO3, the photocatalytic activity enhanced remarkably [12]. Through a series of experiments and calculations, Latterly Olowoyo et al. have also found that carbon nanotubes (CNTs) can strongly be attached to the {101} facet of TiO2, since the atomic orbitals of anatase overlap with the orbitals of the CNTs [13]. p As a novel class of quantum dots (QDs), graphene quantum dots (GQDs) have currently attracted intensive interest in fabricating new heterojunctions, due to their large surface areas, high electron mobilities, conductivity, and adjustable band gaps [14,15]. These properties ensure discrete electronic levels, which could allow for light-induced electron injection, eﬃcient carrier transfer, and long-lived excited states [16–20]. GQDs have also been explored as the light absorber and heteroatoms-doped GQDs are expected to realize absorption in the visible region [21]. A variety of surface functional groups on GQDs could provide better covalent chemical linking between anatase and GQDs, facilitating charge separation and transfer behaviors [22–24]. Pan et al. 2.1. Structural Characterization The crystalline phases of diﬀerent samples before and after decoration with NGQDs were ﬁrstly identiﬁed. Figure 1 shows the X-ray diﬀraction (XRD) patterns of anatase TiO2 with diﬀerent exposed facets. The diﬀraction peaks appeared around 2θ values of 25.3, 38.6, 48.0, 53.9, and 62.1, assigned to the (101), (112), (200), (105), and (213) crystal planes of anatase titania, respectively [27]. For bare anatase TiO2, with the increase of Hydroﬂuoric acid (HF) volume, the {004} diﬀraction peak was broadened, implying the thickness of the TiO2 along the {001} direction was decreased. Meanwhile, the intensity of the {200} diﬀraction peaks was enhanced, indicative of the increasing side length of the nanoparticles along the {100} direction. According to the two peaks, the percentage of the exposed {001} facet could be estimated, based on the calculation method reported in the literature. Shown in Table 1, the percentage was increased when increasing the HF volume [29]. The calculation method was described in the supporting information, as shown in Figure S1. By comparison, the XRD patterns of TiO2/NGQDs exhibited similar but much lower diﬀraction peaks. The XRD pattern of NGQDs showed a wide weak diﬀraction peak centered at 26.8◦, assigned to the {002} facet [30]. Though the diﬀraction peak of NGQDs was not observed in the composites, which might be due to the relatively low diﬀraction intensity of NGQDs, the peak attributed to the {101} facet of anatase decreased obviously. The possible reason is that the {002} crystal orientation in NGQDs inﬂuenced the {101} surface of TiO2, thus leading to decrease of the peak [31]. Figure 1. XRD patterns of diﬀerent samples (a) without and (b) with nitrogen-doped quantum dots (NGQDs) decoration. Figure 1. XRD patterns of diﬀerent samples (a) without and (b) with nitrogen-doped quantum dots (NGQDs) decoration. Table 1. Structural information of diﬀerent anatase titania samples. p Samples Average thickness (nm) Average length (nm) Percentage of {001} T0 8.9 8.1 12% T1 8.5 13.8 56% T2 6.1 21.6 71% T3 4.3 25.5 85% In Raman spectroscopy, all samples show similar peaks centered at 144, 394, 514, and 636 cm−1 shown in Figure S2. When increasing the addition amount of HF, the intensity of the Eg peak at 144 cm−1 decreased simultaneously. The Eg peak is mainly attributed to the symmetric stretching vibration of O−Ti−O TiO2 [29]. 1. Introduction have found that monodispersed amine-functionalized GQDs anatase TiO2 heterojunctions have an absorption range extended into the visible light region and a much lower carrier recombination rate. They attributed the improved performance to the proper energy position of GQDs/TiO2 23]. Then, Yu and co-workers reported the decoration of GQDs on {001} faceted anatase TiO2 with an exposed percentage of 65%–75%. The experiments show a promotion of photocatalytic hydrogen evolution rate of the composites compared to bare anatase, which might originate from the higher-charge separation eﬃciency. [25] Zheng and co-workers also utilized TiO2 and sulfur, nitrogen co-doped GQDs (SN-GQDs) to develop an eﬃcient photocatalyst for synthesizing H2O2. They testiﬁed that SN-GQDs induced visible light absorption, promoted charge transfer, and provided active sites for ·OOH formation [26]. Recently, Prezhdo et al. built several models of donor-acceptor interaction between GQDs and TiO2 via stacking and covalent bonding, respectively, to provide guidance for subsequent photocatalysis application [16]. These ﬁndings provide strong evidence and also motivation for deeply understanding the models of electron-hole separation dynamics at heterojunction interfaces [3,21,27,28]. There has still been no adequate investigation, considering the inﬂuence of facet-dependent photogenerated charge-carrier separation in anatase TiO2, on the photocatalytic activity of GQDs/TiO2. Since both theoretical and experimental studies had shown that {101} crystal facets are electron-rich while {001} crystal facets are hole-rich in anatase TiO2, it would be desirable to elucidate the diﬀerent donor-acceptor interaction between NGQDs and TiO2 with speciﬁc facet composition, thus providing explicit guidance in constructing heterojunction structures with superior performance. Herein, we have designed a heterojunction composite via depositing nitrogen-doped GQDs on anatase TiO2 with diﬀerent exposure percentages of {001} and {101} facets, combining the advantages of facet and interfacial modiﬁcation to maximize the driving force promoting charge carrier transfer. The visible light-driven dye degradation performances on anatase TiO2 with {101}, {001}, and {001}-{101} facets, and their corresponding heterojunctions with nitrogen-doped GQDs (NGQDs) have been systematically studied. To further understand the electron transfer mechanism, the relationships of the morphology, chemical states, optical, and electrical properties with the photocatalytic activity were intensively analyzed. 96 Catalysts 2019, 9, 345 2. Results and Discussion 2. Results and Discussion 2.1. Structural Characterization 2.1. Structural Characterization A higher percentage of exposed {001} facets generally represented fewer amount of symmetric O−Ti−O stretching vibration modes, thus leading to the decreasing intensity of Eg peak in the Raman spectra. Thus, it could be concluded from the Raman spectra that the exposure percentage of {001} facets increased with the increase of HF volume. Shown in the inset of Figure 97 Catalysts 2019, 9, 345 S2b, two characteristic peaks of D and G band appeared at 1351 cm−1 and 1590 cm−1, respectively, conﬁrming the presence of graphite-like structure in the composites. X-ray photoelectron spectroscopy (XPS) spectra were measured to study the bonding conditions in the heterojunctions, shown in Figure 2. The Ti 2p spectra showed two peaks with the binding energies of 458.9 eV and 464.6 eV, which are assigned to Ti 2p3/2 and 2p1/2 spin-orbital splitting photoelectrons, respectively. The splitting values indicated Ti4+ chemical states in these samples, while no Ti3+ forms were observed [32]. The C 1s spectra could be then ﬁtted into three Gaussian peaks (288.8 eV, 286.1 eV, and 284.8 eV). The peak at 288.8 eV was assigned to the sp2 hybridized carbon in the skeleton of NGQDs and also some carbon contaminants from the ambience. The other two peaks corresponded to the oxygenated carbon, representative of carboxyl carbon (288.8 eV) and hydroxyl carbon (286.1 eV) functional groups, respectively [33–36]. No Ti–C bond related peak (282 eV) was observed, implying that the NGQDs were probably anchored to the surface of TiO2 via Ti–O–C bonds. In some reported studies, functional groups including C–O and COOH were evaluated to identify the existence of Ti–O–C bonding [32,37]. These groups are not that stable and might be converted to the epoxy group during the composite formation process. Figure 2. XPS spectra of diﬀerent TiO2/NGQDs composites. Figure 2. XPS spectra of diﬀerent TiO2/NGQDs composites. The evidence of Ti–O–C bonding formation was further provided in the O 1s XPS spectra. It was ﬁtted into two symmetric peaks. The peaks at 530.3 eV is ascribed to the oxygen in crystal lattice (Ti–O–Ti) and the other peak at 531.6 eV is believed to result from the Ti–O–C bonding, based on previously reported cases [25,32,38]. These results indicated the composite formation of TiO2/NGQDs through the C–O–Ti bonds. The C–O–Ti bonds are capable of mediating the coupling between NGQDs and TiO2, which could promote the interfacial electron transfer. 2.2. Morphology Characterization The inset in (d) is the HRTEM image of T2-NGQDs. The inset in (f) is the autocorrelated HRTEM lattice images recorded from the corresponding selected areas. 2.1. Structural Characterization The N 1s spectra revealed a peak centered at 400.3 eV, which could be assigned to the pyrrolic N (400.5 eV). The nitrogen atoms are mainly introduced by the NGQDs, demonstrating the successful decoration of NGQDs on the surface of TiO2 [30]. Furthermore, the percentages of diﬀerent bonds according to the ﬁtting results of the XPS high-resolution spectra were calculated, shown in Table 2 and Table S1. Similar percentages implied similar bonding and chemical composition in diﬀerent composites. Table 2. Percentages of diﬀerent bonds according to the ﬁtting results of the XPS high-resolution spectra calculated from Figure 2. Bond T0-NGQDs T1-NGQDs T2-NGQDs T3-NGQDs % of Ti 2p 26.57 27.63 27.23 27.69 % of C 1s 21.18 19.4 20.65 18.99 O–C=O/Ti–O–C 288.8 eV 7.91 8.03 7.84 7.83 C–O 286.1 eV 19.53 20.30 19.05 19.26 C=C 284.8 eV 72.56 71.07 73.40 73.34 % of O 1s 51.69 52.33 51.53 52.85 Ti–O–Ti 530.3 eV 77.80 80.97 80.84 80.42 Ti–O–C 531.6 eV 22.20 19.03 19.16 19.58 % of N 1s 0.55 0.64 0.58 0.47 98 Catalysts 2019, 9, 345 2.2. Morphology Characterization The morphology of anatase TiO2 with and without NGQDs decoration were characterized by Transmission Electron Microscope (TEM) and High Resolution Transmission Electron Microscope (HR-TEM). The morphology and crystal facets of TiO2 remained similar before and after NGQDs modiﬁcation, as can be seen from Figure 3 and Figure S3. Figure S3 shows that sample T0 is mostly composed of nanoparticles with a truncated octahedral bipryramid. After adding HF into the reaction system, it could be found that the anatase TiO2 mostly consists of nanoplates and the plate size increases with the increase of HF volume, which is consistent with the XRD calculation results. Figure S4 shows the TEM image of NGQDs, which has an average size of ~2.4 nm. Clear lattice fringes demonstrative of its well-crystalline structure, and the autocorrelated HRTEM lattice image (inset in Figure S4b) show a 0.21 nm lattice fringe assigned to the {100} plane of GQDs [39,40]. The formation of NGQDs/TiO2 heterojunctions could be obviously observed in the TEM images, shown in Figure 3, of which the NGQDs were uniformly decorated on both {001} and {101} facets of TiO2. The NGQDs are shown with red circles. No selective deposition of NGQDs on a speciﬁc facet of anatase TiO2 was found. The autocorrelated HRTEM lattice image (inset in Figure 3c,f) both show 0.21 nm lattice fringes assigned to the {100} plane of GQDs. Meanwhile, the autocorrelated HRTEM lattice image (inset in Figure 3c,f) also show clear lattice fringes of 0.35 nm and 0.19 nm, which could be assigned to the {101} and {001} facet of TiO2, respectively. The HRTEM lattice images of all composites show that the lattices of both NGQDs and TiO2 are simultaneously recognized, revealing good attachment of NGQDs over the TiO2 surface. Figure 3. TEM images of (a) T0-NGQDs, (b,c) T1-NGQDs, (d) T2-NGQDs, and (e–f) T3-NGQDs. The inset in (a) is the HR-TEM image of T0-NGQDs. The inset in (c) is the autocorrelated HRTEM lattice images recorded from the corresponding selected areas. The inset in (d) is the HRTEM image of T2-NGQDs. The inset in (f) is the autocorrelated HRTEM lattice images recorded from the corresponding selected areas. Figure 3. TEM images of (a) T0-NGQDs, (b,c) T1-NGQDs, (d) T2-NGQDs, and (e–f) T3-NGQDs. The inset in (a) is the HR-TEM image of T0-NGQDs. The inset in (c) is the autocorrelated HRTEM lattice images recorded from the corresponding selected areas. 2.3. Optical and Electrical Properties The optical properties of all samples were investigated via the Ultraviolet-Visible (UV-Vis) diﬀuse reﬂection spectroscopy (UV-DRS). All anatase TiO2 demonstrate an absorption threshold near 400 nm in the ultraviolet region. The band gaps of the anatase could be obtained based on the Kubelka–Munk rule, seen from the inset in Figure 4a. The band gaps of T0, T1, T2, and T3 are 3.22, 3.28, 3.30, and 3.32 eV, respectively. Apparently, with the increase of the {001} facet percentage, the light absorption 99 Catalysts 2019, 9, 345 edge was slightly blue shifted [21]. The UV-vis spectrum of NGQDs is shown in Figure S4d, and the band gap was approximately 1.56 eV, according to our previous work [30,41]. After decorating NGQDs, it was found that there was an increasing visible light absorption for all composites. Generally, the resultant extended light absorption was due to the existence of Ti–O–C chemical bonds between GQDs and TiO2 [32,37]. The interaction could improve the interfacial carrier transfer rate, which is beneﬁcial to photocatalysis under visible light irradiation. Meanwhile, the presence of the energy gap of NGQDs further ensured the long-lived excited states and absorbance of solar photons in the broad solar spectrum. p The electrochemical Mott–Schottky experiments of the anatase were then measured (Figure 4c). The plots present a positive slope and the ﬂat band potential values were recalculated vs. NHE (Normal Hydrogen Electrode). Combined with the band gaps of anatase TiO2 and NGQDs, we proposed electronic band structures for all composites, shown in Figure 4d. It was hypothesized that the downshift of conduction band (CB) band level of the anatase might make the electrons less reductive and also weaken the dynamics of the electron transfer rate between TiO2 and NGQDs [42–44]. Figure 4. UV-vis diﬀuse reﬂection spectra (UV-DRS) of (a) bare anatase TiO2, (b) TiO2/NGQDs composites, (c) Mott–Schottky plots of bare anatase TiO2, and (d) band structure diagram of diﬀerent samples and NGQDs. The inset in (a) is the Tauc plot of the corresponding bare anatase TiO2. Figure 4. UV-vis diﬀuse reﬂection spectra (UV-DRS) of (a) bare anatase TiO2, (b) TiO2/NGQDs composites, (c) Mott–Schottky plots of bare anatase TiO2, and (d) band structure diagram of diﬀerent samples and NGQDs. The inset in (a) is the Tauc plot of the corresponding bare anatase TiO2. Moreover, the photogenerated charge carrier separation and transfer rate was determined using photocurrent responses, shown in Figure 5. 2.3. Optical and Electrical Properties Fast and uniform photocurrents with good reproducibility was demonstrated, indicative of relatively reversible photo-responses. Via three on-oﬀcycles under visible light irradiation (>420 nm), it was found that the photocurrent density of bare anatase TiO2 decreased in the order of T2, T3, T1, and T0. This suggests that defects (oxygen vacancies) are possibly formed on the surface of pristine anatase TiO2 with small size under light illumination in our work, which would introduce defect energy levels in the band gap and lead to electron-hole separation under visible light irradiation [25]. Meanwhile, the synergistic eﬀect of {001} and {101} facets would also aﬀect the charge carrier separation eﬃciency, resulting in the diﬀerence of photocurrent. After depositing NGQDs, their corresponding photocurrent responses were all enhanced obviously. The photocurrent density of TiO2/NGQDs composites decreased in the order of T1-NGQDs, T2-NGQDs, T0-NGQDs, and T3-NGQDs. The photocurrent density of sample T1-NGQDs was the highest, about 100 Catalysts 2019, 9, 345 three times larger than that of sample T3-NGQDs. The photocurrent enhancement could be attributed to the promoted separation rate of photogenerated charge carriers, owing to the introduction of NGQDs. The unique chemical features of NGQD endow it with superior carrier mobility and excellent electronic conductivity [45,46]. Moreover, the extended π-electron systems of NGQD provide its suﬃcient contact with the surface of titania, and the formation of Ti–O–C bonding could also facilitate the donor-acceptor interaction [16]. These factors contributed to the apparent enhancement of photocurrent density. We calculated multiple times the photocurrents of the anatase TiO2 with and without NGQDs decoration, shown in Figure 5c. Interestingly, the times decreased with the decrease of exposed {101} facet percentage in TiO2. The phenomenon indicated a noticeable improvement of electron-hole separation eﬃciency between the NGQDs and the electron-rich {101} facet, compared to that between the NGQDs and the hole-rich {001} facet. The diﬀerence demonstrated that eﬃcient electron transfer existed in the interfacial interaction between NGQDs and TiO2 with high {101} facet exposure. Figure 5. Periodic on-oﬀphotocurrent output of (a) bare anatase TiO2, and (b) TiO2/NGQDs composites. (c) The multiple times of the photocurrents of the anatase TiO2 with and without NGQDs decoration. Figure 5. Periodic on-oﬀphotocurrent output of (a) bare anatase TiO2, and (b) TiO2/NGQDs composites. (c) The multiple times of the photocurrents of the anatase TiO2 with and without NGQDs decoration. Electrochemical impedance spectroscopy (EIS) measurements were utilized to investigate the mechanism of photocurrent improvement. 2.4. Photocatalytic Performance 2.4. Photocatalytic Performance The photocatalytic performances of all samples were evaluated for RhB photodegradation under visible light irradiation after achieving absorption-desorption balance in the dark, as seen from Figure 8. Obviously, the photocatalytic degradation eﬃciency of the TiO2/NGQDs heterojunctions was greatly enhanced compared with that of the bare anatase TiO2. Among the composites, approximately 96% of the dye was photo-degraded by the T1-NGQDs within 3 h. The degradation process followed the pseudo-ﬁrst-order kinetics: −ln c c0 = kt (1) (1) where k is equal to the corresponding slope of the ﬁtting line, representing the rate constant indicative of the photocatalytic eﬃciency. To prove the eﬃciency of sample T1-NGQDs, the photocatalytic degradation activity of other three organic pollutants including methylene blue (MB), methyl orange (MO), and phenol (Phe) were compared. It demonstrated that the T1-NGQDs exhibited high photocatalytic activity for degrading MB dye as well as other common organic species (including phenol, colorless aqueous solution), shown in Figure S6. where k is equal to the corresponding slope of the ﬁtting line, representing the rate constant indicative of the photocatalytic eﬃciency. To prove the eﬃciency of sample T1-NGQDs, the photocatalytic degradation activity of other three organic pollutants including methylene blue (MB), methyl orange (MO), and phenol (Phe) were compared. It demonstrated that the T1-NGQDs exhibited high photocatalytic activity for degrading MB dye as well as other common organic species (including phenol, colorless aqueous solution), shown in Figure S6. The calculated k values of diﬀerent samples are shown in Figure 8c (the red column corresponds to the bare anatase TiO2 and the blue column to the composites). Before decorating NGQDs, sample T2 and T3 exhibited nearly the same constant rate, higher than that of T0 and T1. RhB could be easily absorbed on the titania surface with reactive {001} facet exposure, leading to subsequent dye self-photosentization process decomposing RhB under visible light irradiation [51–53]. After decoration of NGQDs, it was found that the values were much higher, and the calculated k of T1-NGQDs reached the highest, about 0.8 h−1. The multiple times of the k values of the anatase TiO2 with and without NGQDs decoration were then calculated, shown in Figure 8d. Notably, it was shown that the times decreased with the increase of the exposed {001} facet percentage in TiO2, which is consistent with the photocurrent variation in Figure 5c. 2.3. Optical and Electrical Properties The semicircle diameter generally indicates the carrier transfer resistance. In Figure 6, all Nyquist plots of the composites presented as semicircle, the corresponding arc radius of EIS Nyquist plots all decreased compared to that of bare anatase TiO2, which is consistent with the photocurrent output. Among all photocatalysts, sample T1-NGQDs exhibited the smallest semicircle while sample T0 the largest. This result is demonstrative of more eﬀective carrier separation and transfer process in the heterojunctions [47,48]. Thus, it could be concluded that the percentage of exposed {001} facet indeed had diﬀerent eﬀects on the electronic properties of the composites, which might be attributed to the diﬀerent interfacial interaction between NGQDs and TiO2 with diﬀerent facet compositions. Figure 6. Electrochemical impedance spectroscopy (EIS) Nyquist plots of diﬀerent TiO2 samples without and with NGQDs decoration. Figure 6. Electrochemical impedance spectroscopy (EIS) Nyquist plots of diﬀerent TiO2 samples without and with NGQDs decoration. 101 Catalysts 2019, 9, 345 Catalysts 2019, 9, 345 Photoluminescence (PL) emission spectrum could help to directly understand the carrier behaviors and observe the radiative recombination of charge carriers [49,50]. Generally, PL emission signals are caused by the photo-induced carrier recombination process. Lower intensity is relevant to better photocatalytic performance. All peaks shape similarly in Figure 7. After decorating NGQDs, the composites exhibited a slight decrease in the PL intensity, compared to the bare anatase, which was probably due to the eﬃcient electron transfer from the CB band of anatase to NGQDs. Thus, the trapping and recombination of charge carriers could be hindered. Figure 7. PL spectra of diﬀerent TiO2 samples without and with NGQDs decoration. Figure 7. PL spectra of diﬀerent TiO2 samples without and with NGQDs decoration. 2.4. Photocatalytic Performance The same law illustrates the photocatalytic degradation performance depends greatly on the interfacial charge carrier separation and transfer rate, demonstrative of the signiﬁcant donor-acceptor interaction between NGQDs and {101} facets. At the same time, it was supposed that the diﬀerent roles of {001} and {101} facets in separating the 102 Catalysts 2019, 9, 345 photogenerated electron-hole are unignorable, since the hybrid containing TiO2 with 56% {001} facets exhibited a higher reaction rate value than that with 12% {001} facets. Figure 8. The photocatalytic degradation of RhB for diﬀerent samples under visible light irradiation. Change of the relative concentration (Ct/C0) of RhB in (a) bare anatase TiO2, (b) TiO2/NGQDs composites as a function of irradiation time up to 180 min. (c) Plot of k (ln(C0/Ct)) values for RhB degradation in diﬀerent samples. (d) The times of the k values of the anatase with and without NGQDs decoration. 5 Photocatalytic Mechanism Figure 8. The photocatalytic degradation of RhB for diﬀerent samples under visible light irradiation. Change of the relative concentration (Ct/C0) of RhB in (a) bare anatase TiO2, (b) TiO2/NGQDs composites as a function of irradiation time up to 180 min. (c) Plot of k (ln(C0/Ct)) values for RhB degradation in diﬀerent samples. (d) The times of the k values of the anatase with and without NGQDs decoration. 2.5. Photocatalytic Mechanism of TiO2 before and after NGQDs decoration, it could be concluded that there is a synergistic eﬀect of facet-manipulated electron-hole separation in TiO2 and donor-acceptor interaction in GQDs/TiO2 on the visible light driven photocatalytic performance. Scheme 1. Schematic illustration of the proposed band alignment and interfacial electron transfer process for the TiO2/NGQDs heterojunction composites under visible light irradiation (λ > 420 nm). Materials and Methods Scheme 1. Schematic illustration of the proposed band alignment and interfacial electron transfer process for the TiO2/NGQDs heterojunction composites under visible light irradiation (λ > 420 nm). Materials and Methods 3.1. Synthesis of Anatase TiO2 The anatase TiO2 with diﬀerent exposed percentages of {001} and {101} facets were prepared via the traditional hydrothermal method. Typically, diﬀerent volumes (0.2, 0.4, and 0.8 mL) of hydroﬂuoric acid (≥40.0%. Sinopharm Chemical Reagent Co.,Ltd, Shanghai, China) were added into the mixture of tetrabutyl titanate (5 mL; ≥99.0%, Aladdin, Shanghai, China) and ethanol (8 mL). After stirring for 30 min, the mixture was then transferred into a 50 mL Teﬂon-lined autoclave and heated at 180 ◦C for 24 h. After that, the products were collected by centrifugation, followed by being rinsed several times with absolute ethyl alcohol and dried at 60 ◦C overnight. The obtained anatase TiO2 co-exposed with {001} and {101} facets were named as T1, T2, and T3 respectively. To prepare TiO2 with the {101} dominating plane, the same procedure was conducted but with 0.4 mL of H2O, which was named T0. 2.5. Photocatalytic Mechanism The schematic representation of the electron-hole separation and transfer in TiO2/NGQDs heterojunction composites during the photocatalytic reaction is shown in Scheme 1. Under visible light irradiation, both NGQDs and anatase TiO2 of avoidable intrinsic defects were capable to generate photo-excited electrons. Obviously, the narrow energy gaps for NGQDs allow for rich hot electrons to produce when the excitation wavelength is larger than 420 nm. According to the measured band potential values, the energy levels in Figure 4d further demonstrate that the band conﬁguration contributes to electron injections from the CB level of TiO2 to the LUMO of NGQDs. The donor-acceptor interaction was greatly promoted due to the full contact and C–O–Ti formation between NGQDs and TiO2. The appropriate band alignments explain for the reason why the photocatalytic activities were all enhanced after depositing NGQDs on the surface of TiO2, compared to that of bare TiO2. p g p On the other hand, for bare TiO2, it was found that the anatase TiO2 with the higher percentage of exposed {001} facet owned a much better photocatalytic performance which is due to the high activity of the {001} facet. However, the reaction rate value (k) achieved the highest for T1-NGQDs rather than T3-NGQDs. The diﬀerent variation trend of k value in bare TiO2 and TiO2/NGQDs heterojunctions imply the possible inﬂuence of facets in the interfacial electron transfer process. It revealed that the photo-excited electron and holes behave diﬀerently in TiO2, where electrons could be easily trapped in the {101} facet while holes tend to run to the {001} facet. Since NGQDs are a good electron transport medium, its deposition on the titania surface with more percentages of exposed {101} facets could result in a higher electron transfer eﬃciency, which is consistent with the increasing multiple times of photocurrent and also k values with and without NGQDs decoration. Simultaneously, the diﬀerent roles of {101} and {001} facets in anatase TiO2 on charge carrier separation is not negligible, considering that the k value of T1-NGQDs is higher than that of T0-NGQDs. By comparing the degradation activity 103 Catalysts 2019, 9, 345 of TiO2 before and after NGQDs decoration, it could be concluded that there is a synergistic eﬀect of facet-manipulated electron-hole separation in TiO2 and donor-acceptor interaction in GQDs/TiO2 on the visible light driven photocatalytic performance. 3.2. Synthesis of NGQDs The NGQDs were prepared according to a one-step hydrothermal process reported by Sun and co-workers [54]. In a typical run, 1.44 g of urea (AR, Aladdin, Shanghai, China) and 1.68 g of citric acid (GR, Aladdin, Shanghai, China) were dissolved in 40 mL of deionized water (DI water). The solution was transferred into a 50 mL Teﬂon-lined stainless autoclave and heated at 180 ◦C for 8 h. The ﬁnal product was centrifuged several times at 10,000 rpm for 5 min with absolute alcohol. The obtained NGQDs precipitate was dried at 80 ◦C for 1 h to obtain the NGQDs powders. 3.3. Synthesis of TiO2/NGQDs Heterojunction Composites 3.5. Photocatalytic Performance The photocatalytic performance of the as-synthesized photocatalysts were examined under a 300 W Xe lamp (PLS-SXE 300/300 UV, Perfect Light, Shanghai, China) equipped with a 420 nm cut-oﬀ ﬁlter as the visible light irradiation source. A total of 15 mg of catalysts were added into 50 mL of a 10.0 mg·L−1 solution of diﬀerent dyes: Rhodamin B (RhB), methylene blue (MB), methyl orange (MO), and phenol (Phe) (Sinopharm Chemical Reagent Co. Ltd., Shanghai, China). The molar concentration was 0.026, 0.031, 0.030, and 0.11 mmol/L, respectively. Before irradiation, the suspension was stirred in the dark to ensure the adsorption-desorption equilibrium of RhB on the surface of the photocatalyst. 3.6. Photoelectrochemical Measurements The photocurrent measurements, electrochemical impedance spectroscopy (EIS) and Mott–Schottky experiments were conducted on an electrochemical analyzer (CHI 660C work station, CHI, Shanghai, China). The employed standard three-electrode conﬁguration included a platinum plate (as the counter electrode), an Ag/AgCl electrode (as the reference electrode), and a working electrode. The working electrodes were prepared as follows: 40 mg of powders and 5 mg of Mg(NO3)2·6H2O (≥98%; Alfa Aesar, Shanghai, China) were dispersed in 100 mL of isopropanol. The suspension was ultrasound for 1 h. A clean SnO2 transparent conductive glass doped with ﬂuorine, FTO (as cathode) facing the stainless-steel anode was then immersed into this suspension. The distance between the two electrodes was ﬁxed at about 5 cm. The Mg2+ adsorbed samples suspension was loaded in a quartz vessel as the electrolyte, and the electrophoresis process was performed at 60 V for 120 s. After the electrophoretic deposition (EPD) process, the prepared electrodes were washed by ethanol and deionized water several times and dried at room temperature. A 350 W xenon lamp with a cut-oﬀ ﬁlter (λ > 420 nm) was used as a light source and placed 20 cm away from the working electrode. The working electrode was immersed in 0.1 M Na2SO4 aqueous solution. The EIS measurements were performed over a range from 0.01 to 1000 Hz at 0.2 V, and the amplitude of the applied potential in each case was 5 mV. 3.4. Characterization The XRD patterns of all samples were recorded with a PAN analytical X’Pert Pro MPD diﬀractometer (Pananalytical, Holland) using Cu-Kα radiation (λ = 0.1541 nm), and the data was collected from 20◦to 80◦(2θ). UV-vis diﬀuse reﬂectance spectroscopy (DRS) were taken at room temperature measured using BaSO4 as the reference on a UV-3150 spectrophotometer (Shimadzu, Kyoto, Japan). The photoluminescence (PL) spectroscopy was performed using a RF-5301pc ﬂuorescence spectroscopy (Shimadzu, Kyoto, Japan). Band gap energies were calculated by analysis of the Tauc-plots resulting from Kubelka–Munk transformation of absorption spectra. High resolution transmission electron microscope (HRTEM) were conducted by a Phillips/FEI Tecnai F20 S-TWIN TEM (Hillsborough, OR, USA) instrument operating at 200 kv. X-ray photoelectron spectroscopy (XPS) measurements were carried out on an ESCALAB 250 Xi (Thermo Scientiﬁc, MA, USA) using non-monochromatized Mg-Kα X-ray as the excitation source. The binding energies for the samples were calibrated by setting the measured binding energy of C 1s to 284.60 eV. The Raman spectra were measured on a LabRAM HR Evolution (Horiba, Tokyo, Japan) at room temperature using the 532 nm line of an argon ion laser as the excitation source. 3.3. Synthesis of TiO2/NGQDs Heterojunction Composites An ultrasonic-hydrothermal method was used to prepare the TiO2/NGQDs heterojunction photocatalysts. A certain amount of TiO2 and GQDs at a low doping level (1.0 wt%) was added into 40 mL of DI water. The suspension solution was placed in an ultrasonic bath for 30 min and then transferred into a 50 mL Teﬂon-lined stainless autoclave by heating at 120 ◦C for another 2 h. After centrifuging several times with DI water and being dried at 60 ◦C overnight, the ﬁnal powders were named as T0-NGQDs, T1-NGQDs, T2-NGQDs, and T3-NGQDs, respectively. 104 Catalysts 2019, 9, 345 Catalysts 2019, 9, 345 3.4. Characterization 4. Conclusions In summary, nitrogen-doped graphene quantum dots were successfully deposited onto the surface of anatase TiO2 with diﬀerent percentages of exposed {001} facets to form TiO2/GQDs heterojunction composites. The photocatalytic performances of the hybrid containing TiO2 with 56% {001} facets exhibited the highest reaction rate value under visible light irradiation. The successful decoration of NGQDs on the TiO2 surface extended the light absorption edge into the visible light region. Due 105 Catalysts 2019, 9, 345 to Ti–O–C formation and high electron conductivity of NGQDs, the photocurrent responses of the composites were all enhanced obviously, compared to that of the bare samples. Meanwhile, the diﬀerent roles of {101} and {001} facets in anatase TiO2 on charge carrier separation was also not negligible. The improved photocatalytic activity was due to the synergistic eﬀect of facet-manipulated electron-hole separation in TiO2 and the remarkable donor-acceptor interaction between NGQDs and the electron-rich {101} facet. The existence of the {101} facet contributed to the interfacial electron transfer that played a vital role in improving the photocatalytic activity of the NGQDs/TiO2 heterojunctions. Furthermore, the existence of both facets in anatase assisted the further enhancement of photocatalytic performance. This work provided a new clue to improve the interfacial charge transfer in faceted semiconductor related heterojunctions. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/4/345/s1, Figure S1: (a) Slab model of anatase TiO2 single crystal. (b) Equilibrium model of anatase TiO2 single crystal. (Calculation method of the percentage of {001} facets). Figure S2: (a) Raman spectra of diﬀerent TiO2 samples without and with NGQDs decoration. (b) Raman spectra of T1 before and after decoration of NGQDs. The inset in Figure S2 b is the enlargement of 1200–1700 cm−1 of T1-NGQDs. Figure S3: TEM images of (a) T0, (b) T1, (c) T2, and (d) T3. Figure S4: (a) TEM image, (b) HRTEM image, (c) AFM image, (d) UV-vis spectra and PL spectra of the GQDs (the excitation wavelength is 365 nm), (e) Raman spectra, and (f) XRD pattern of NGQDs. The inset in (a) is the size distribution of NGQDs. The inset in (b) is the autocorrelated HRTEM lattice images recorded. Figure S5: HRTEM images of the anatase TiO2 decorate with NGQDs. Figure S6: (a) The photocatalytic degradation of diﬀerent pollutants for T1-NGQDs, (b) plot of k values for diﬀerent pollutants degradation in T1-NGQDs. Author Contributions: The manuscript was written through contributions of all authors. N.-Q.O., H.-J.L. 4. Conclusions designed the experiments; N.-Q.O., B.-W.L., B.-J.G. and X.S. performed the experiments and analyzed the data; N.-Q.O. and H.-J.L. wrote the paper; D.-J.Q. revised the paper. X.W., Y.J. and J.Y. acted as supervisor. Funding: This research was funded by National Natural Science Foundation of China (51572173, 51602197, 51771121 and 51702212), Shanghai Municipal Science and Technology Commission (16060502300, 16JC1402200 and 18511110600), Shanghai Academic/Technology Research Leader Program (19XD1422900), Shanghai Eastern Scholar Program (QD2016014). Conﬂicts of Interest: The authors declare no conﬂicts of interest. Conﬂicts of Interest: The authors declare no conﬂicts of interest. References 1. Tang, H.; Prasad, K.; Sanjinès, R.; Schmid, P.E.; Lévy, F. Electrical and optical properties of TiO2 anatase thin ﬁlms. J. Appl. Phys. 1994, 75, 2042–2047. [CrossRef] 1. Tang, H.; Prasad, K.; Sanjinès, R.; Schmid, P.E.; Lévy, F. Electrical and optical properties of TiO2 anatase thin ﬁlms. J. Appl. Phys. 1994, 75, 2042–2047. [CrossRef] 2. Xu, M.; Gao, Y.; Moreno, E.M.; Kunst, M.; Muhler, M.; Wang, Y.; Idriss, H.; Woll, C. Photocatalytic activity of bulk TiO2 anatase and rutile single crystals using infrared absorption spectroscopy. Phys. Rev. Lett. 2011, 106, 138302. [CrossRef] [PubMed] 3. Yu, J.; Low, J.; Xiao, W.; Zhou, P.; Jaroniec, M. Enhanced Photocatalytic CO2-Reduction Activity of Anatase TiO2 by Co-exposed {001} and {101} Facets. J. Am. Chem. Soc. 2014, 136, 8839–8842. [CrossRef] 3. Yu, J.; Low, J.; Xiao, W.; Zhou, P.; Jaroniec, M. Enhanced Photocatalytic CO2-Reduction Activity of TiO2 by Co-exposed {001} and {101} Facets. J. Am. Chem. Soc. 2014, 136, 8839–8842. [CrossRef] 4. Tachikawa, T.; Yamashita, S.; Majima, T. Evidence for crystal-face-dependent TiO2 photocatalysis from single-molecule imaging and kinetic analysis. J. Am. Chem. Soc. 2011, 133, 7197–7204. [CrossRef] [PubMed] 4. Tachikawa, T.; Yamashita, S.; Majima, T. Evidence for crystal-face-dependent TiO2 photocatalysis from single-molecule imaging and kinetic analysis. J. Am. Chem. Soc. 2011, 133, 7197–7204. [CrossRef] [PubMed] 5. Selcuk, S.; Selloni, A. Facet-dependent trapping and dynamics of excess electrons at anatase TiO2 surfaces and aqueous interfaces. Nat. Mater. 2016, 15, 1107–1118. [CrossRef] 4. Tachikawa, T.; Yamashita, S.; Majima, T. Evidence for crystal-face-dependent TiO2 photocatalysis from single-molecule imaging and kinetic analysis. J. Am. Chem. Soc. 2011, 133, 7197–7204. [CrossRef] [PubMed] single-molecule imaging and kinetic analysis. J. Am. Chem. Soc. 2011, 133, 7197–7204. [CrossRef] [PubMed] 5. Selcuk, S.; Selloni, A. Facet-dependent trapping and dynamics of excess electrons at anatase TiO2 surfaces and aqueous interfaces. Nat. Mater. 2016, 15, 1107–1118. [CrossRef] 5. Selcuk, S.; Selloni, A. Facet-dependent trapping and dynamics of excess electrons at anatase TiO2 surfaces and aqueous interfaces. Nat. Mater. 2016, 15, 1107–1118. [CrossRef] 6. Chen, Q.; Ma, W.; Chen, C.; Ji, H.; Zhao, J. Anatase TiO2 mesocrystals enclosed by (001) and (101) facets: Synergistic eﬀects between Ti3+ and facets for their photocatalytic performance. Chem. Eur. J. 2012, 18, 12584–12589. [CrossRef] 7. Hyam, R.S.; Lee, J.; Cho, E.; Khim, J.; Lee, H. Eﬀect of Annealing Environments on Self-Organized TiO2 Nanotubes for Eﬃcient Photocatalytic Applications. Nanosci. Nanotechnol. 2012, 12, 8908–8912. [CrossRef] 8. References C 2012, 116, 19893–19901. [CrossRef] 18. Wang, F.; Wu, Y.; Wang, Y.; Li, J.; Jin, X.; Zhang, Q.; Li, R.; Yan, S.; Liu, H.; Feng, Y.; et al. Construction of novel Z-scheme nitrogen-doped carbon dots/{0 0 1} TiO2 nanosheet photocatalysts for broad-spectrum-driven diclofenac degradation: Mechanism insight, products and eﬀects of natural water matrices. Chem. Eng. J. 2019, 356, 857–868. [CrossRef] 19. Wang, W.; Ni, Y.; Xu, Z. One-step uniformly hybrid carbon quantum dots with high-reactive TiO2 for photocatalytic application. J. Alloys Compd. 2015, 622, 303–308. [CrossRef] 20. Baker, D.R.; Kamat, P.V. Photosensitization of TiO2 Nanostructures with CdS Quantum Dots: Particulate versus Tubular Support Architectures. Adv. Funct. Mater. 2009, 19, 805–811. [CrossRef] 21. Kenrick, J.; Williams, C.A.N.; Yan, X.; Li, L.-S.; Zhu, X. Hot Electron Injection from Graphene Quantum D to TiO2. ACS Nano 2013, 7, 1388–1394. 2 , , 22. Yang, N.; Liu, Y.; Wen, H.; Tang, Z.; Zhao, H.; Li, Y.; Wang, D. Photocatalytic Properties of Graphdiyne and Graphene Modiﬁed TiO2: From Theory to Experiment ACS Nano 2013 7 1504–1512 [CrossRef] 22. Yang, N.; Liu, Y.; Wen, H.; Tang, Z.; Zhao, H.; Li, Y.; Wang, D. Photocatalytic Properties of Graphdiyne Graphene Modiﬁed TiO2: From Theory to Experiment. ACS Nano 2013, 7, 1504–1512. [CrossRef] Graphene Modiﬁed TiO2: From Theory to Experiment. ACS Nano 2013, 7, 1504–1512. [CrossRef] 23. Pan, D.; Jiao, J.; Li, Z.; Guo, Y.; Feng, C.; Liu, Y.; Wang, L.; Wu, M. Eﬃcient Separation of Electron–Hole Pairs in Graphene Quantum Dots by TiO2 Heterojunctions for Dye Degradation. ACS Sustain. Chem. Eng. 2015, 3, 2405–2413. [CrossRef] 24. Ma, Y.; Chen, A.Y.; Xie, X.; Wang, X.; Wang, D.; Wang, P.; Li, H.-J.; Yang, J.; Li, Y. Doping eﬀect and ﬂuorescence quenching mechanism of N-doped graphene quantum dots in the detection of dopamine. Talanta 2018, 196, 563–571. [CrossRef] [PubMed] 25. Yu, S.; Zhong, Y.Q.; Yu, B.Q.; Cai, S.Y.; Wu, L.Z.; Zhou, Y. Graphene quantum dots to enhance the photocatalytic hydrogen evolution eﬃciency of anatase TiO2 with exposed {001} facet. Phys. Chem. Chem. Phys. 2016, 18, 20338–20381. [CrossRef] 26. Zheng, L.; Su, H.; Zhang, J.; Walekar, L.S.; Vafaei Molamahmood, H.; Zhou, B.; Long, M.; Hu, Y.H. Highly selective photocatalytic production of H2O2 on sulfur and nitrogen co-doped graphene quantum dots tuned TiO2. Appl. Catal. B 2018, 239, 475–484. [CrossRef] 27. Jiang, B.; Tian, C.; Zhou, W.; Wang, J.; Xie, Y.; Pan, Q.; Ren, Z.; Dong, Y.; Fu, D.; Han, J.; et al. References Wang, X.; Sun, G.; Li, N.; Chen, P. Quantum dots derived from two-dimensional materials and their 7. Hyam, R.S.; Lee, J.; Cho, E.; Khim, J.; Lee, H. Eﬀect of Annealing Environments on Self-Organized T Nanotubes for Eﬃcient Photocatalytic Applications. Nanosci. Nanotechnol. 2012, 12, 8908–8912. [CrossR Nanotubes for Eﬃcient Photocatalytic Applications. Nanosci. Nanotechnol. 2012, 12, 8908–8912. [CrossRef] 8. Wang, X.; Sun, G.; Li, N.; Chen, P. Quantum dots derived from two-dimensional materials and their applications for catalysis and energy. Chem. Soc. Rev. 2016, 45, 2239–2262. [CrossRef] 8. Wang, X.; Sun, G.; Li, N.; Chen, P. Quantum dots derived from two-dimensional materials and their applications for catalysis and energy. Chem. Soc. Rev. 2016, 45, 2239–2262. [CrossRef] 9. Wegner, K.D.; Hildebrandt, N. Quantum dots: Bright and versatile in vitro and in vivo ﬂuorescence imaging biosensors. Chem. Soc. Rev. 2015, 44, 4792–4834. [CrossRef] 10. Yuan, F.; Li, S.; Fan, Z.; Meng, X.; Fan, L.; Yang, S. Shining carbon dots: Synthesis and biomedical and optoelectronic applications. Nano Today 2016, 11, 565–586. [CrossRef] 106 Catalysts 2019, 9, 345 11. Crake, A.; Christoforidis, K.C.; Godin, R.; Moss, B.; Kaﬁzas, A.; Zafeiratos, S.; Durrant, J.R.; Petit, C. Titanium dioxide/carbon nitride nanosheet nanocomposites for gas phase CO2 photoreduction under UV-visible irradiation. Appl. Catal. B 2019, 242, 369–378. [CrossRef] 12. Olowoyo, J.O.; Kumar, M.; Jain, S.L.; Babalola, J.O.; Vorontsov, A.V.; Kumar, U. Insights into Reinforced Photocatalytic Activity of the CNT–TiO2 Nanocomposite for CO2 Reduction and Water Splitting. J. Phys. Chem. C 2018, 123, 367–378. [CrossRef] 13. Rimoldi, L.; Giordana, A.; Cerrato, G.; Falletta, E.; Meroni, D. Insights on the photocatalytic degradation processes supported by TiO2/WO3 systems. The case of ethanol and tetracycline. Catal. Today 2019, 328, 210–215. [CrossRef] 14. Zheng, X.T.; Ananthanarayanan, A.; Luo, K.Q.; Chen, P. Glowing graphene quantum dots and carbon dots: Properties, syntheses, and biological applications. Small 2015, 11, 1620–1655. [CrossRef] 15. Freeman, R.; Finder, T.; Bahshi, L.; Gill, R.; Willner, I. Functionalized CdSe/ZnS QDs for the detection of nitroaromatic or RDX explosives. Adv. Mater. 2012, 24, 6416–6421. [CrossRef] 16. Long, R.; Casanova, D.; Fang, W.-H.; Prezhdo, O.V. Donor-Acceptor Interaction Determines the Mechanism of Photo-Induced Electron Injection from Graphene Quantum Dots into TiO2: Stacking Supersedes Covalent Bonding. J. Am. Chem. Soc. 2017, 139, 2619–2629. [CrossRef] 17. Wang, W.-S.; Wang, D.-H.; Qu, W.-G.; Lu, L.-Q.; Xu, A.-W. Large Ultrathin Anatase TiO2 Nanosheets with Exposed {001} Facets on Graphene for Enhanced Visible Light Photocatalytic Activity. J. Phys. Chem. References In situ growth of TiO2 in interlayers of expanded graphite for the fabrication of TiO2-graphene with enhanced photocatalytic activity. Chem. Eur. J. 2011, 17, 8379–8387. [CrossRef] 28. Liu, L.; Jiang, Y.; Zhao, H.; Chen, J.; Cheng, J.; Yang, K.; Li, Y. Engineering Coexposed {001} and {101} Facets in Oxygen-Deﬁcient TiO2 Nanocrystals for Enhanced CO2 Photoreduction under Visible Light. ACS Catal. 2016, 6, 1097–1108. [CrossRef] 29. Tian, F.; Zhang, Y.; Zhang, J.; Pan, C. Raman Spectroscopy: A New Approach to Measure the Percentage of Anatase TiO2 Exposed (001) Facets. J. Phys. Chem. C 2012, 116, 7515–7519. [CrossRef] 107 Catalysts 2019, 9, 345 30. Li, H.-J.; Sun, X.; Xue, F.; Ou, N.; Sun, B.-W.; Qian, D.-J.; Chen, M.; Wang, D.; Yang, J.; Wang, X. Redox Induced Fluorescence On–OﬀSwitching Based on Nitrogen Enriched Graphene Quantum Dots for Formaldehyde Detection and Bioimaging. ACS Sustain. Chem. Eng. 2018, 6, 1708–1716. [CrossRef] 31. Qu, A.; Xie, H.; Xu, X.; Zhang, Y.; Wen, S.; Cui, Y. High quantum yield graphene quantum dots decorate TiO2 nanotubes for enhancing photocatalytic activity. Appl. Surf. Sci. 2016, 375, 230–241. [CrossRef] g g q y g p q TiO2 nanotubes for enhancing photocatalytic activity. Appl. Surf. Sci. 2016, 375, 230–241. [CrossRef] 32. Rajender, G.; Kumar, J.; Giri, P.K. Interfacial charge transfer in oxygen deﬁcient TiO2-graphene quantum dot hybrid and its inﬂuence on the enhanced visible light photocatalysis. Appl. Catal. B 2018, 224, 960–972. [CrossRef] 33. Shen, K.; Xue, X.; Wang, X.; Hu, X.; Tian, H.; Zheng, W. One-step synthesis of band-tunable N, S co-doped commercial TiO2/graphene quantum dots composites with enhanced photocatalytic activity. RSC Adv. 2017, 7, 23319–23327. [CrossRef] 34. Wang, W.; Xu, D.; Cheng, B.; Yu, J.; Jiang, C. Hybrid carbon@TiO2 hollow spheres with enhanced photocatalytic CO2 reduction activity. J. Mater. Chem. A 2017, 5, 5020–5029. [CrossRef] 35. Xu, C.; Han, Q.; Zhao, Y.; Wang, L.; Li, Y.; Qu, L. Sulfur-doped graphitic carbon nitride decorated with graphene quantum dots for an eﬃcient metal-free electrocatalyst. J. Mater. Chem. A 2015, 3, 1841–1846. [CrossRef] 36. Han, S.; Hu, X.; Wang, J.; Fang, X.; Zhu, Y. Novel Route to Fe-Based Cathode as an Eﬃcient Bifunctional Catalysts for Rechargeable Zn-Air Battery. Adv. Energy Mater. 2018, 8, 1800955. [CrossRef] 37. Li, H.-J.; Ou, N.-Q.; Sun, X.; Sun, B.-W.; Qian, D.-J.; Chen, M.; Wang, X.; Yang, J. Exploitation of the synergistic eﬀect between surface and bulk defects in ultra-small N-doped titanium suboxides for enhancing photocatalytic hydrogen evolution. References Catal. Sci. Technol. 2018, 8, 5515–5525. [CrossRef] 38. Rimoldi, L.; Pargoletti, E.; Meroni, D.; Falletta, E.; Cerrato, G.; Turco, F.; Cappelletti, G. Concurrent role of metal (Sn, Zn) and N species in enhancing the photocatalytic activity of TiO2 under solar light. Catal. Today 2018, 313, 40–46. [CrossRef] 39. Ding, H.; Yu, S.B.; Wei, J.S.; Xiong, H.M. Full-Color Light-Emitting Carbon Dots with a Surface-State- Controlled Luminescence Mechanism. ACS Nano 2016, 10, 484–491. [CrossRef] 40. Ding, H.; Wei, J.S.; Zhang, P.; Zhou, Z.Y.; Gao, Q.Y.; Xiong, H.M. Solvent-Controlled Synthesis of Highly Luminescent Carbon Dots with a Wide Color Gamut and Narrowed Emission Peak Widths. Small 2018, 14, e1800612. [CrossRef] 41. Yang, H.; Wang, P.; Wang, D.; Zhu, Y.; Xie, K.; Zhao, X.; Yang, J.; Wang, X. New Understanding on Photocatalytic Mechanism of Nitrogen-Doped Graphene Quantum Dots-Decorated BiVO4 Nanojunction Photocatalysts. ACS Omega 2017, 2, 3766–3773. [CrossRef] y g 42. Deming, C.P.; Mercado, R.; Gadiraju, V.; Sweeney, S.W.; Khan, M.; Chen, S. Graphene Quantum Dots-Supported Palladium Nanoparticles for Eﬃcient Electrocatalytic Reduction of Oxygen in Alkaline Media. ACS Sustain. Chem. Eng. 2015, 3, 3315–3323. [CrossRef] 43. Safardoust-Hojaghan, H.; Salavati-Niasari, M. Degradation of methylene blue as a pollutant with N-doped graphene quantum dot/titanium dioxide nanocomposite. J. Clean. Prod. 2017, 148, 31–36. [CrossRef] 44. Wang, S.; Cole, I.S.; Li, Q. Quantum-conﬁned bandgap narrowing of TiO2 nanoparticles by graphene quantum dots for visible-light-driven applications. Chem. Commun. 2016, 52, 9208–9211. [CrossRef] 45. Wang, D.; Chen, J.-F.; Dai, L. Recent Advances in Graphene Quantum Dots for Fluorescence Bioimaging from Cells through Tissues to Animals. Part. Part. Syst. Charact. 2015, 32, 515–523. [CrossRef] 46. Zou, X.; Liu, M.; Wu, J.; Ajayan, P.M.; Li, J.; Liu, B.; Yakobson, B.I. How Nitrogen-Doped Graphene Quantum Dots Catalyze Electroreduction of CO2 to Hydrocarbons and Oxygenates. ACS Catal. 2017, 7, 6245–6250. [CrossRef] 47. Li, H.-J.; Qian, D.J.; Chen, M. Templateless Infrared Heating Process for Fabricating Carbon Nitride Nanorods with Eﬃcient Photocatalytic H2 Evolution. ACS Appl. Mater. Interfaces 2015, 7, 25162–25170. [CrossRef] 48. Yan, M.; Zhu, F.; Gu, W.; Sun, L.; Shi, W.; Hua, Y. Construction of nitrogen-doped graphene quantum dots-BiVO4/g-C3N4Z-scheme photocatalyst and enhanced photocatalytic degradation of antibiotics under visible light. RSC Adv. 2016, 6, 61162–61174. [CrossRef] 49. Liu, J.; Li, J.; Sedhain, A.; Lin, J.; Jiang, H. Structure and Photoluminescence Study of TiO2 Nanoneedle Text along Vertically Aligned Carbon Nanoﬁber Arrays. J. Phys. Chem. C 2008, 112, 17127–17132. Hydrogen Production from Glycerol Photoreforming on TiO2/HKUST-1 Composites: Eﬀect of Preparation Method Fabián M. Martínez 1, Elim Albiter 1, Salvador Alfaro 1, Ana L. Luna 2, Christophe Colbeau-Justin 2, José M. Barrera-Andrade 1, Hynd Remita 2 and Miguel A. Valenzuela 1,* Fabián M. Martínez 1, Elim Albiter 1, Salvador Alfaro 1, Ana L. Luna 2, Christophe Colbeau-Justin 2, José M. Barrera-Andrade 1, Hynd Remita 2 and Miguel A. Valenzuela 1,* 1 Laboratorio de Catálisis y Materiales, ESIQIE-Instituto Politécnico Nacional, México City 07738, Mexico; fmm003@eng.ucsd.edu (F.M.M.); ealbitere@ipn.mx (E.A.); salfaroh@ipn.mx (S.A.); jmanban@yahoo.com.mx (J.M.B.-A.) 1 Laboratorio de Catálisis y Materiales, ESIQIE-Instituto Politécnico Nacional, México City 07738, Mexico; fmm003@eng.ucsd.edu (F.M.M.); ealbitere@ipn.mx (E.A.); salfaroh@ipn.mx (S.A.); jmanban@yahoo.com.mx (J.M.B.-A.) 2 Laboratoire de Chimie Physique, CNRS UMR 8000 Université Paris-Sud, 91405 Orsay, France; aluna0786@gmail.com (A.L.L.); christophe.colbeau-justin@u-psud.fr (C.C.-J.); hynd.remita@u-psud.fr (H.R.) * Correspondence: mavalenz@ipn.mx; Tel.: +52-55-5729-6000 (ext. 55112) 2 Laboratoire de Chimie Physique, CNRS UMR 8000 Université Paris-Sud, 91405 Orsay, France; aluna0786@gmail.com (A.L.L.); christophe.colbeau-justin@u-psud.fr (C.C.-J.); hynd.remita@u-psud.fr (H.R.) * Correspondence: mavalenz@ipn.mx; Tel.: +52-55-5729-6000 (ext. 55112) Received: 12 March 2019; Accepted: 31 March 2019; Published: 4 April 2019 * Correspondence: mavalenz@ipn.mx; Tel.: +52-55-5729-6000 (ext. 55112) Received: 12 March 2019; Accepted: 31 March 2019; Published: 4 April 2019 Abstract: Coupling metal-organic frameworks (MOFs) with inorganic semiconductors has been successfully tested in a variety of photocatalytic reactions. In this work we present the synthesis of TiO2/HKUST-1 composites by grinding, solvothermal, and chemical methods, using diﬀerent TiO2 loadings. These composites were used as photocatalysts for hydrogen production by the photoreforming of a glycerol-water mixture under simulated solar light. Several characterization techniques were employed, including X-ray diﬀraction (XRD), UV-Vis diﬀuse reﬂectance spectroscopy (DRS), infrared spectroscopy (FTIR), and time-resolved microwave conductivity (TRMC). A synergetic eﬀect was observed with all TiO2/HKUST-1 composites (mass ratio TiO2/MOF 1:1), which presented higher photocatalytic activity than that of individual components. These results were explained in terms of an inhibition of the charge carrier (hole-electron) recombination reaction after photoexcitation, favoring the electron transfer from TiO2 to the MOF and creating reversible Cu1+/Cu0 entities useful for hydrogen production. Keywords: hydrogen production; photocatalysis; TiO2-HKUST-1 composites; solar light Catalysts 2019, 9, 338; doi:10.3390/catal9040338 www.mdpi.com/journal/catalysts References [CrossRef] , J ; , J ; , ; , J ; J g, y 2 along Vertically Aligned Carbon Nanoﬁber Arrays. J. Phys. Chem. C 2008, 112, 17127–17132. [CrossRef] 50. Nadica, D.; Abazovic, M.I.C.O.; Dramicanin, M.D. Photoluminescence of Anatase and Rutile TiO2 Particles. J. Phys. Chem. B 2006, 110, 25366–25370. 50. Nadica, D.; Abazovic, M.I.C.O.; Dramicanin, M.D. Photoluminescence of Anatase and Rutile TiO2 Particles. J. Phys. Chem. B 2006, 110, 25366–25370. 108 Catalysts 2019, 9, 345 51. Hu, C.; Zhang, X.; Li, X.; Yan, Y.; Xi, G.; Yang, H.; Bai, H. Au photosensitized TiO2 ultrathin nanosheets with {001} exposed facets. Chem. Eur. J. 2014, 20, 13557–13560. [CrossRef] 52. Li, K.; Xu, Y.; He, Y.; Yang, C.; Wang, Y.; Jia, J. Photocatalytic fuel cell (PFC) and dye self-photosensitization photocatalytic fuel cell (DSPFC) with BiOCl/Ti photoanode under UV and visible light irradiation. Environ. Sci. Technol. 2013, 47, 3490–3497. [CrossRef] 53. Wang, X.; Xia, R.; Muhire, E.; Jiang, S.; Huo, X.; Gao, M. Highly enhanced photocatalytic performance of TiO2 nanosheets through constructing TiO2/TiO2 quantum dots homojunction. Appl. Surf. Sci. 2018, 459, 9–15. [CrossRef] 54. Qu, D.; Zheng, M.; Du, P.; Zhou, Y.; Zhang, L.; Li, D.; Tan, H.; Zhao, Z.; Xie, Z.; Sun, Z. Highly luminescent S, N co-doped graphene quantum dots with broad visible absorption bands for visible light photocatalysts. Nanoscale 2013, 5, 12272–12278. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 109 catalysts catalysts 1. Introduction For this reason, the search for new materials, active and stable in the presence of sunlight, is of great interest [7–9]. Metal organic frameworks (MOFs) are obtained by the self-assembly of metal ions and organic ligands through the formation of covalent bonds or the presence of inter-molecular forces between them [10]. MOFs present a long-range periodic structure with good crystallinity, and MOF-based structures take some unique properties of both organic and inorganic porous materials. They exhibit several advantages such as a high surface area, tunable pore size, easy preparation, ﬂexibility, and structural diversity [11]. CuMOF, also known as HKUST-1, copper-benzene-1,3,5-tricarboxylate (Cu-BTC), MOF-199 or Basolite® C300, was ﬁrst assembled by Chui et al. [12] through the formation of coordination bonds between trimesic acid (H3BTC) and Cu ions [13]. MOFs have been investigated in many ﬁelds, such as sensing, drug delivery, sequestration, separation, molecular transport, electronics, bioreactors, optics, energy production, and catalysis, among others [14]. Applications in photocatalysis have been reported in the last decade, and since then, several articles and reviews have been published focusing on artiﬁcial photosynthesis (i.e., water splitting and CO2 photoreduction) [15,16], organic photosynthesis [17], and pollutants degradation [18,19]. Speciﬁcally, in solar-driven hydrogen evolution with the presence of a sacriﬁcial electron donor, e.g., alcohols, most of the MOFs cannot be used as a stable and eﬃcient photocatalyst for this application individually [20]. Certain modiﬁcations of the pristine MOF, including the decoration of the organic linker or metal center, combination with semiconductors, metal nanoparticles loading, decoration with reduced graphene oxide, sensitization, pyrolyzation, and incorporation with other functional materials, have been tested to increase their activity and stability under visible light [19,21]. Hybrid nanocomposites of semiconductors with MOFs have attracted increased attention because they improve charge transfer mechanisms with a lower charge recombination and more eﬃcient light harvesting [22]. Hybrid nanocomposites based on TiO2 and HKUST-1 are exciting materials which could show synergic eﬀects enhancing photocatalytic activity under visible light. Only a few investigations have reported the synthesis, structure, and properties (i.e., as photocatalysts in hydrogen production) of TiO2/HKUST-1 nanocomposites [23–26]. Particularly, it has been reported that in these nanocomposites, HKUST-1 is transformed to Cu-Cu2O nanoparticles after calcination at 400 ◦C, presenting better rates of hydrogen production in comparison with Cu deposited on TiO2 by conventional methods [24]. There are contradictory results concerning the stability of HKUST-1. 1. Introduction Nowadays, one of the most important necessities of society is the use of natural renewable resources to produce energy, minimizing the use of fossil fuels and reducing the associated harmful pollution produced by their combustion. On the other hand, hydrogen is considered a good candidate as a green energy carrier because it produces null pollution during its combustion, and it can be obtained from renewable sources [1–3]. The use of hydrogen as an energy carrier has several beneﬁts, such as the many diﬀerent storage possibilities, its ability to be converted to other energy forms with ease and to be produced from water with near-zero emissions, and its high conversion eﬃciency [2]. However, there are also severe limitations for the widespread use of hydrogen, for example, as a fuel for transportation. If we are planning to use hydrogen-combustion and hydrogen-fuel-cell vehicles in the future, we must ﬁrst resolve outstanding issues, such as the eﬃcient and safe storage of hydrogen, creating a fueling infrastructure, and reducing its production costs [4]. Certainly, one possibility to reduce the production cost of hydrogen is the use of green energy sources. In this sense, hydrogen production using solar energy can be categorized as: (a) thermal, (b) photovoltaic, (c) bio-photolysis, and (c) photo-electrochemical [5]. Although most of the production methods involve renewable Catalysts 2019, 9, 338; doi:10.3390/catal9040338 111 Catalysts 2019, 9, 338 sources, they are not well understood, and their development implies an increase in production costs and low global eﬃciency [2–5]. Photocatalytic hydrogen generation can be obtained mostly by two diﬀerent approaches: (1) photocatalytic water splitting and (2) photocatalytic reforming of organics [6]. The ﬁrst method relates to the capability of water to be reduced and oxidized by reacting with photogenerated electrons and positive holes, during semiconductor irradiation, in the presence of selected co-catalysts. The second approach is based on the ability of some organic species—namely, sacriﬁcial agents—to donate electrons to the positive holes of the illuminated photocatalyst and be oxidized, generating proton ions, while photogenerated electrons reduce the latter to produce hydrogen in the presence of proper co-catalysts. Glycerol is a sustainable compound that can be used for hydrogen production by photocatalytic reactions (photoreforming). Although this reaction has been studied extensively, the overall performance towards hydrogen evolution is low, and in many cases, a high photocatalytic activity is only achieved with UV-light irradiation. 2.1. Photocatalytic Hydrogen Evolution 2.1. Photocatalytic Hydrogen Evolution Due to the lack of studies regarding the eﬀect of the optimal amount of TiO2 that can be deposited on the HKUST-1, Figure 1 shows the photocatalytic hydrogen evolution rates as a function of TiO2 content. All the experiments were conducted under similar operating conditions, and the H2 production rate was estimated after 8 h of irradiation time. As can be observed, the results demonstrate a synergic photocatalytic activity between HKUST-1 and TiO2, and the best performance corresponds to the composites with 50 wt % TiO2. In the case of the catalyst prepared by grinding (50TiO2 P25/com-HKUST-1) the production rate was 2.9 mmol × g−1 × h−1, 2.4 mmol × g−1 × h−1 for the 50TiO2-ST/com-HKUST-1, and 4.5 mmol × g−1 × h−1 for the 50TiO2 P25/syn-HKUST-1. Note that the photoactivity of the synthesized HKUST-1 and commercial HKUST-1 was insigniﬁcant, and as a comparison, the production rate shown by TiO2 P25 was 1.1 mmol × g−1 × h−1. It is very important to highlight that during the reaction, there was a change in color in the photocatalysts from light blue (original composite color) to reddish brown (spent composite, see Figure 2b), which is indicative that the Cu2+ originally present in the HKUST-1 was partially reduced towards Cu1+ or Cu0 [28,29]. This observation suggests that HKUST-1 assembled with Cu ions and benzene 1,3,5-tricarboxylate ligands (Cu-BTC) is an unstable material when irradiated in an aqueous medium, and functions as a precursor of Cu reduced species interacting with TiO2, as co-catalysts in the production of hydrogen. Note that the highest amounts of CO2 and CH4 were obtained with the 50TiO2 P25/syn-HKUST-1 composite, which comes from the photocatalytic oxidation of an aqueous solution of glycerol. Indeed, it has been proposed that a secondary alcohol photoreforming can produce methane via β-hydride elimination, which could explain the origin of produced methane [30]. Figure 1d compares the amount of hydrogen produced as a function of the irradiation time for the prepared 50TiO2/HKUST-1 composites. Hydrogen production followed almost the same trend with the three photocatalysts. However, a higher photoactivity was observed with the 50TiO2 P25/syn-HKUST-1 composite. It is important to point out that only the 50TiO2 P25/syn-HKUST-1 composite presented long-term activity, and it was evaluated in ﬁve cycles, under simulated solar light. Figure 2a shows the H2 production rate reached during 8 h of reaction time in each cycle. 1. Introduction For example, when this MOF was used in aqueous media, it decomposed after 24 h of reaction [27]. However, TiO2/HKUST-1 composites synthesized using ionic liquids as solvents showed high activity and stability during photo-oxidation/photoreduction reactions [28]. In this context, it would be very useful to know the role played by TiO2-HKUST interactions on the activity and stability in glycerol photoreforming. Therefore, TiO2/HKUST-1 composites were synthesized by employing three methods: the ﬁrst composite was prepared by grinding the commercial 112 Catalysts 2019, 9, 338 reagents, Aeroxide® TiO2 P25 (Evonik, P25), and HKUST-1 (Basolite® C300). These composites were designated as TiO2 P25/com-HKUST-1. The second one was formed by TiO2 prepared by a solvothermal route in the presence of the commercial HKUST-1 (TiO2-ST/com-HKUST-1), and the third composite was prepared by synthesizing HKUST-1 by a chemical route in the presence of TiO2 P25 (TiO2 P25/syn-HKUST-1). Furthermore, the aim of the present work was focused on the eﬀect of the preparation method of TiO2/HKUST-1 composites, as well as the mass ratio TiO2:MOF employed, on their photocatalytic properties for hydrogen production, using glycerol as a sacriﬁcial agent. 2.1. Photocatalytic Hydrogen Evolution It can be seen from the second cycle that the composite shows a reduction on the production rate, and in the ﬁfth cycle the observed reduction was ca. 50% of the production rate observed in the ﬁrst run. An explanation for this unfavorable behavior can be given in terms of a partial reduction of Cu2+ contained in the original HKUST-1 by the photogenerated electrons in the TiO2 conduction band, which was clearly demonstrated by the color change of the original HKUST-1 from light blue (original composite) to reddish brown (spent composite), as shown in Figure 2b. Furthermore, the zone attributed to the d-d spin allowed the transition of the Cu2+ between 500–800 nm (discussed later), which was modiﬁed to a reddish-brown color, as is characteristic of Cu reduced species in the HKUST-1 structure [31]. Interestingly, after each reaction cycle and subsequent washing and purging of the reaction cell, the solid returned to the original light blue color of the composite. This means that the HKUST-1 structure was not completely destroyed, otherwise it would be forming the Cu1+-Cu2+ MOF meta-stable phase with photocatalytic activity to reduce protons to hydrogen. This behavior was previously reported in other applications of 113 Catalysts 2019, 9, 338 KUST-1 [28,29,31,32] however, this is the ﬁrst time that it has been observed in the photocatalyt ydrogen evolution reaction, which requires a systematic and thorough study. Figure 1. Eﬀect of TiO2 content on the H2, CO2, and CH4 production rates of TiO2/HKUST-1 composites, after 8 h irradiation with simulated solar light with (a) TiO2 P25/com-HKUST-1, (b) TiO2-ST/com-HKUST-1, and (c) TiO2 P25/syn-HKUST-1 (d) Hydrogen evolution vs. time for TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (▲), 50TiO2-ST/com-HKUST-1 (•), and 50TiO2 P25/syn-HKUST-1 (■). Figure 2. (a) Photocatalytic stability tests of 50TiO2 P25/syn-HKUST-1, under simulated solar light, (b) UV-Vis diﬀuse reﬂectance spectra (DRS) of fresh and spent 50TiO2 P25/syn-HKUST-1 composites. HKUST-1 [28,29,31,32] however, this is the ﬁrst time that it has been observed in the photocatalytic hydrogen evolution reaction, which requires a systematic and thorough study. Figure 1. Eﬀect of TiO2 content on the H2, CO2, and CH4 production rates of TiO2/HKUST-1 composites, after 8 h irradiation with simulated solar light with (a) TiO2 P25/com-HKUST-1, Figure 1. 2.2. Characterization Figure 3 shows the Fourier-transform infrared (FT-IR) spectra of com-HKUST-1, syn-HKUST-1, and TiO2/HKUST-1 composites. Clearly, the com-HKUST-1 and syn-HKUST-1 spectra are quite similar to those reported in previous works [33–35], which indicates that the method employed for the synthesis of syn-HKUST-1 was eﬀective. In these spectra, several signals appeared in the range from 1300 to 1500 cm−1 and from 1500 to 1700 cm−1, which are associated with the interactions between the carboxylate anion—in symmetric and asymmetric modes respectively—with the metal ion [34,35]. The signals indicated at 1110, 765, and 740 cm−1 are associated with the C-H vibration modes in the aromatic ring [34]. The band at 1060 cm−1 is attributed to the presence of copper coordinated N,N-dimethylformamide (DMF) molecules [33], and the band centered at 507 cm−1 is assigned to the Cu-O stretching mode [35]. On the other hand, all the bands mentioned previously appeared in the spectrum of the 50TiO2 P25/com-HKUST-1 composite, indicating a weak interaction between TiO2 P25 and com-HKUST-1. In the case of the 50TiO2-ST/com-HKUST-1 and 50TiO2 P25/syn-HKUST-1, the bands between 1300 and 1700 cm−1 were less deﬁned, and the signals at 507, 740, and 765 cm−1 were replaced by a broad band (500–900 cm−1) in the case of TiO2-ST/com-HKUST-1 and two bands (500–700 and 765–830 cm−1) in the spectrum of 50TiO2 P25/syn-HKUST-1. These last results could indicate that there is a chemical interaction between TiO2 and HKUST-1 when either component is obtained by a chemical route. Figure 3. Fourier-transform infrared (FT-IR) spectra of (a) syn-HKUST-1 (1), com-HKUST-1 (2) (b) 50TiO2 P25/syn-HKUST-1 (3), 50TiO2 ST/com-HKUST-1 (4), and 50TiO2 P25/com-HKUST-1 (5). Figure 3. Fourier-transform infrared (FT-IR) spectra of (a) syn-HKUST-1 (1), com-HKUST-1 (2) (b) 50TiO2 P25/syn-HKUST-1 (3), 50TiO2 ST/com-HKUST-1 (4), and 50TiO2 P25/com-HKUST-1 (5). The optical properties of HKUST-1 and the TiO2/Cu MOFs composites were investigated by UV-Vis diﬀuse reﬂectance spectroscopy (DRS). As can be seen in Figure 4a, the as-prepared HKUST-1 sample and the commercial HKUST-1 showed a similar spectrum, exhibiting two characteristic absorption bands centered at 300 and 700 nm, similar values to those reported in the literature [36]. Note that one shoulder can also be detected at 375 nm. The ﬁrst band located in the UV region is assigned to π-π* transitions of the ligands and the band in the visible zone is attributed to the d-d spin and allowed transition of the Cu2+ [37]. 2.1. Photocatalytic Hydrogen Evolution Eﬀect of TiO2 content on the H2, CO2, and CH4 production rates of TiO2/HKUST-1 composites, after 8 h irradiation with simulated solar light with (a) TiO2 P25/com-HKUST-1, (b) TiO2-ST/com-HKUST-1, and (c) TiO2 P25/syn-HKUST-1 (d) Hydrogen evolution vs. time for TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (▲), 50TiO2-ST/com-HKUST-1 (•), and 50TiO2 P25/syn-HKUST-1 (■). Figure 1. Eﬀect of TiO2 content on the H2, CO2, and CH4 production rates of TiO2/HKUST-1 composites, after 8 h irradiation with simulated solar light with (a) TiO2 P25/com-HKUST-1, (b) TiO2-ST/com-HKUST-1, and (c) TiO2 P25/syn-HKUST-1 (d) Hydrogen evolution vs. time for TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (▲), 50TiO2-ST/com-HKUST-1 (•), and 50TiO2 P25/syn-HKUST-1 (■). / y ( ) Figure 2. (a) Photocatalytic stability tests of 50TiO2 P25/syn-HKUST-1, under simulated solar light, (b) UV-Vis diﬀuse reﬂectance spectra (DRS) of fresh and spent 50TiO2 P25/syn-HKUST-1 composites. Figure 2. (a) Photocatalytic stability tests of 50TiO2 P25/syn-HKUST-1, under simulated solar light, (b) UV Vi diﬀ ﬂ (DRS) f f h d 50TiO P25/ HKUST 1 i Figure 2. (a) Photocatalytic stability tests of 50TiO2 P25/syn-HKUST-1, under simulated solar light, (b) UV-Vis diﬀuse reﬂectance spectra (DRS) of fresh and spent 50TiO2 P25/syn-HKUST-1 composites. 114 114 Catalysts 2019, 9, 338 Catalysts 2019, 9, 338 2.2. Characterization The shoulder at 375 nm is ascribed to the ligand-to-metal charge transfer (LMCT), and the additional broad absorption band between 500 and 800 nm is assigned to the d-d spin and allowed transition of the Cu2+ (d9) ions [37]. 115 Catalysts 2019, 9, 338 Figure 4. UV-Vis DRS spectra of (a) P25 (1), syn-HKUST-1 (2), com-HKUST-1 (3), and (b) TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (4), 50TiO2 ST/com-HKUST-1 (5), and 50TiO2 P25/syn-HKUST-1 (6). Figure 4. UV-Vis DRS spectra of (a) P25 (1), syn-HKUST-1 (2), com-HKUST-1 (3), and (b) TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (4), 50TiO2 ST/com-HKUST-1 (5), and 50TiO2 P25/syn-HKUST-1 (6). Figure 4b corresponds to the UV-Vis DRS spectra of the TiO2/HKUST-1 composites. In general, all composite photocatalysts showed similar absorption behavior to the pristine HKUST-1. However, it is worth noting a slight change of their absorption edge to the UV zone (350–400 nm), compared to those of commercial and synthesized HKUST-1 (Figure 4a), which can be related to the TiO2-HKUST-1 interaction. On the other hand, the slight diﬀerences in the 50TiO2 ST/com-HKUST-1 composite spectrum (e.g., a lower absorption in the visible region) could be related to a shielding eﬀect by TiO2, partially inhibiting the visible light absorption of the Cu2+ ions in the HKUST-1 structure, because TiO2, in this particular composite, was grown in intimate contact with the commercial HKUST-1. Figure 5a presents the X-ray diﬀraction (XRD) pattern of syn-HKUST-1, which is quite similar to the pattern of com-HKUST-1, showing the main reﬂections peaks at 11.6◦, 13.4◦, 17.4◦and 19◦[38,39]. These results prove that the HKUST-1 structure was successfully obtained using our described preparation method. XRD patterns of 50TiO2 P25/com-HKUST-1, 50TiO2 ST/com-HKUST-1, and 50TiO2 P25/syn-HKUST-1 composites are shown in Figure 5b. These three samples displayed the same reﬂections described earlier, indicating that the HKUST-1 structure was preserved despite the preparation method used. Note that a higher crystallinity is observed in samples 50TiO2 P25/com-HKUST-1 and 50TiO2 P25/syn-HKUST-1 in comparison with sample 50TiO2 ST/com-HKUST-1, which means that a poor crystallization of TiO2 occurred due to the low synthesis temperature (i.e., 100 ◦C) compared to that reported in the literature, above 150 ◦C under solvothermal process [40]. Figure 5. X-ray diﬀraction patterns of (a) as-sensitized HKUST-1 (1), commercial HKUST-1 (2), and (b) 50TiO2 P25/com-HKUST-1 (3), 50TiO2 ST/com-HKUST-1 (4), and 50TiO2 P25/syn-HKUST-1 (5). ▼denotes the peak corresponding to TiO2 anatase. Figure 5. 2.2. Characterization X-ray diﬀraction patterns of (a) as-sensitized HKUST-1 (1), commercial HKUST-1 (2), and (b) 50TiO2 P25/com-HKUST-1 (3), 50TiO2 ST/com-HKUST-1 (4), and 50TiO2 P25/syn-HKUST-1 (5). ▼denotes the peak corresponding to TiO2 anatase. 116 Catalysts 2019, 9, 338 Catalysts 2019, 9, 338 Figure 6a compares the time-resolved microwave conductivity (TRMC) proﬁles of TiO2 P25, syn-HKUST-1, and com-HKUST-1, obtained under a wavelength excitation of 355 nm. The highest signal was exhibited by TiO2 P25, which also presented a long-time decay. It is worth noting that syn-HKUST-1 showed a TRMC signal because it behaves like a semiconductor material; however, it decays faster than TiO2 P25, revealing a short lifetime of photogenerated electrons. Surprisingly, com-HKUST-1, which presented very similar structure and light absorption (see Figure 4a) to those of syn-HKUST-1, displayed a much lower TRMC signal, revealing a great diﬃculty in executing the charge separation after irradiation with UV light of 355 nm. Figure 6. Time-resolved microwave conductivity (TRMC) transient signal of (a,c) P25 (1), syn-HKUST-1 (2), com-HKUST-1 (3) (b,d) TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (4), 50TiO2 ST/com-HKUST-1) (5), and 50TiO2 P25/syn-HKUST-1 (6). The excitation wavelengths were 355 nm (a,b) and 410 nm (c,d). Figure 6. Time-resolved microwave conductivity (TRMC) transient signal of (a,c) P25 (1), syn-HKUST-1 (2), com-HKUST-1 (3) (b,d) TiO2/HKUST-1 composites: 50TiO2 P25/com-HKUST-1 (4), 50TiO2 ST/com-HKUST-1) (5), and 50TiO2 P25/syn-HKUST-1 (6). The excitation wavelengths were 355 nm (a,b) and 410 nm (c,d). On the other hand, by analyzing the TRMC signals of the composites in Figure 6b, the sample prepared by grinding (sample 4) displayed a decay proﬁle quite similar than that obtained with TiO2 P25, indicating that the charge carrier dynamics are mainly due to the TiO2 P25 contribution. Unexpectedly, the 50TiO2 ST/com-HKUST-1 (sample 5) presented a TRMC signal with an Imax value slightly smaller than that of the grinding composite (sample 4), clearly showing a charge carrier separation, but with a short time decay. The decay signal abruptly becomes highly noisy after 70 ns, denoting a charge carrier recombination or electron transfer from TiO2 to HKUST-1. The 50TiO2 ST/com-HKUST-1 (sample 6) did not display a clear TRMC signal, similar to that shown by the com-HKUST-1(Figure 6a), which can be connected with the XRD results, meaning that TiO2 was poorly crystallized. As simulated solar light is being used in the photocatalytic evaluation, it is interesting to see the TRMC signals of the materials under 410 nm excitation. 2.2. Characterization As can be seen in Figure 6c, TiO2 P25, com-HKUST-1 and syn-HKUST-1MOFs presented similar behaviors to those presented under 355 nm 117 Catalysts 2019, 9, 338 excitation, but all the samples showed a lower Imax value. The TiO2 P25 signal can be attributed to the presence of rutile, which has a bandgap of 3.0 eV, making possible the generation of electron-hole pairs under visible-light irradiation. All the composites shown in Figure 6d exhibited a small TRMC signal, which means that they have the capacity to generate electron-hole pairs under 410 nm excitation. 3. Discussion The above results suggest that the photocatalytic performance of TiO2 is improved by the incorporation of HKUST-1, forming a semiconductor-MOF composite regardless the preparation method. Nonetheless, the integration of HKUST-1 with TiO2 P25 (TiO2 P25/syn-HKUST-1) by a chemical method showed greater photocatalytic activity and stability compared with to grinding (TiO2 P25/com-HKUST-1) or the TiO2-ST/com-HKUST-1. At ﬁrst glance, the greater photocatalytic activity shown by the composites in comparison with TiO2 P25 or HKUST-1 is explained by the synergy between the semiconductor and the MOF, inhibiting electron-hole recombination. This cooperative behavior implied the partial reduction of the Cu2+ contained in the MOF, forming Cu1+ species which absorb visible light and could contribute to the proton reduction, as shown in Figure 7. The creation of reversible Cu1+/Cu2+ entities in the composite was attributed to the electron transfer from P25 to HKUST-1 generating in situ species, i.e., Cu1+-Cu2+ MOF, giving rise to improved hydrogen production. Figure 7. Proposed photocatalytic mechanism of glycerol photoreforming with the TiO2 P25/syn-HKUST-1 under solar light irradiation. Figure 7. Proposed photocatalytic mechanism of glycerol photoreforming with the TiO2 P25/syn-HKUST-1 under solar light irradiation. Figure 7. Proposed photocatalytic mechanism of glycerol photoreforming with the TiO2 P25/syn-HKUST under solar light irradiation. Furthermore, it was found that the higher photoactivity and stability shown by the TiO2 P25/syn-HKUST-1 can also be related to a strong interaction between the two components, which was not seen with the other two composites prepared by grinding or mixing poor crystallized TiO2 with commercial HKUST-1. 4.2.3. TiO2 P25 Incorporation During HKUST-1 Synthesis (TiO2 P25/syn-HKUST-1) The synthesis route of synthesized HKUST-1 mainly followed the procedure reported by Tranchemontagne et al. [41] with some modiﬁcations, such as the integration of the TiO2 P25 during the preparation of precursor solution. First, two solutions were prepared: one solution containing 100 mg of trimesic acid (H3BTC) dissolved in 6 mL of a mixture of DMF/EtOH/H2O with a molar ratio (1:1:1), and a second solution contained 200 mg of Cu(OAc)2 × H2O dissolved in 6 mL of the solvent DMF/EtOH/H2O. Both solutions were mixed under magnetic stirring to get a homogeneous solution. Then, 0.2 mL of Et3N was added drop by drop as an oxidant agent to the reaction mixture under magnetic stirring. After that, a given quantity of TiO2 P25 (25, 50 and 75 wt %) was added to the MOFs precursor solution. This suspension was kept under magnetic stirring for 24 h, and the powder was obtained by centrifugation. The solid was washed ﬁve times with 5 mL of DMF to eliminate the residues of any organic compound, and ﬁnally, the solid was dried—at 50 ◦C for 5 h—milled and sieved to get a homogeneous particle size. 4.1. Materials Copper (II) acetate monohydrate (Cu(OAc)2H2O), trimesic acid (H3BTC), N,N-dimethylformamide (DMF), Triethylamine (Et3N), and Titanium (IV) isopropoxide were purchased from Sigma-Aldrich (St. Louis, MO, USA), Ethanol (EtOH) and ammonium hydroxide 30% were purchased from Panreac Chemicals (Chicago, IL, USA), and deionized water (H2O), Aeroxide® TiO2 P25 (Degussa), and Basolite® C300 (Sigma-Aldrich) were used as reference materials without further puriﬁcation. 118 Catalysts 2019, 9, 338 4.2. Preparation Methods 4.2.1. Grinding (TiO2 P25/com-HKUST-1) 4.2.1. Grinding (TiO2 P25/com-HKUST-1) The composites were prepared by grinding the TiO2 P25 and commercial HKUST-1 powders by hand in an agate mortar until a homogeneous light blue color was obtained. Five composites were prepared with a TiO2 content of 25, 50 and 75 wt %. 4.2.2. TiO2 Solvothermal Deposition on Commercial HKUST-1 (TiO2-ST/com-HKUST-1) TiO2 was prepared by mixing 1.14 mL of titanium (IV) isopropoxide with 100 mL of anhydrous ethanol and sonicating this for 5 min. Then, under vigorous magnetic stirring, concentrated nitric acid (70% v/v) was added drop by drop to get a pH around 1. The solution was diluted with 10 mL of distilled water, and 8 mL of ammonium hydroxide was added as a precipitating agent. Subsequently, a given amount of commercial HKUST-1 was added to the suspension under vigorous stirring. The resultant suspension was placed in a homemade PTFE-lined autoclave and sealed hermetically, and then introduced into a convective furnace (Fisher Scientiﬁc, Pittsburgh PA, USA) at 100 ◦C for 24 h. The composite formed was recovered and washed ﬁve times with a mixture of DMF/EtOH/H2O (molar ratio of 1:1:1) to eliminate the residues of any organic compound. Then, the product was dried at 50 ◦C for 5 h. The solid was ground in an agate mortar and sieved using a US 80 mesh to get a homogeneous particle size. Three composites were prepared by this method with a nominal TiO2 content of 25, 50, and 75 wt %. 4.4. Photocatalytic H2 Evolution 4.4. Photocatalytic H2 Evolution The composite powders were evaluated in the hydrogen production reaction using a glycerol-water solution with a volumetric ratio glycerol/water = 1:9. The photocatalytic reaction was carried out in a 25 mL glass cell. The composites (1 g/L) and the glycerol-water mixture was placed in the cell and mixed to form a homogeneous suspension, and then purged with nitrogen to eliminate all dissolved oxygen. Before irradiation, the reaction cell was maintained under stirring for one hour for adsorption/desorption equilibration and then irradiated with a solar simulator (Model 9600, 150 W; Newport Corporation, Irvine, CA, USA) for 8 h or 24 h. The gas mixture (H2, CO2, CH4) produced during the reaction was analyzed in a Perkin–Elmer Gas Chromatograph (Autosystem XL, Waltham, MA, USA). 5. Conclusions A series of TiO2/HKUST-1 composites were successfully prepared using grinding, solvothermal, and chemical methods. All composites showed a higher photocatalytic activity than the individual components, particularly those containing a TiO2/HKUST-1 weight ratio of 1:1. These results demonstrated the eﬀect of the synthesis method of composites on photocatalytic activity and stability. The best performance was obtained with the composite prepared by a chemical route, i.e., the synthesis of HKUST-1 in the presence of TiO2 P25, leading to a strong interaction between the two components. The higher photocatalytic performance of the composites, compared with TiO2 or HKUST-1, was explained regarding a synergy between the semiconductor and the HKUST-1, inhibiting electron-hole recombination. There was experimental evidence of the reversible partial reduction of Cu2+ towards the Cu1+-Cu0 entities contained in HKUST-1, which could indicate the in situ formation of highly active HKUST-1 co-catalysts, improving the photocatalytic activity. Author Contributions: Conceptualization, S.A., and M.A.V.; investigation, F.M.M., and A.L.L.; methodology, C.C.-J., M.A.V., and H.R.; writing—original draft preparation, M.A.V., and S.A.; writing—review and editing, E.A., and J.M.B.-A. Funding: This research was sponsored by “Consejo Nacional de Ciencia y Tecnología” México (Project No. 153356), and “Instituto Politécnico Nacional” (Projects SIP 20194976 and SIP 20196347). Funding: This research was sponsored by “Consejo Nacional de Ciencia y Tecnología” México (Project No. 1533 and “Instituto Politécnico Nacional” (Projects SIP 20194976 and SIP 20196347). Acknowledgments: M.V. and H.R. acknowledge Université Paris-Saclay for the ﬁnancial support through Chaire Jean d’Alembert program and the IRS MOMENTOM (Ininiative de Recherche Stratégique). Acknowledgments: M.V. and H.R. acknowledge Université Paris-Saclay for the ﬁnancial support through the Chaire Jean d’Alembert program and the IRS MOMENTOM (Ininiative de Recherche Stratégique). Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 4.3. Characterization Techniques All composites were characterized by several techniques. X-ray diﬀraction patterns were recorded on a Siemens D-5000 diﬀractometer (Munich, Germany), with a copper anode and Cu-Kα radiation over a 2 theta range of 10–80◦using a step size of 4 ◦/min. FTIR and UV-Vis spectra of powder samples were respectively obtained using a Nicolet system (Nexus 470, Thermo Fisher Scientiﬁc, Waltham, MA, USA) (with KBr pellet samples) and a GBC spectrophotometer (Cintra 20, GBC Scientiﬁc, Hampshire, IL, USA), repectively. The dynamics of the charge carriers in the photocatalysts were studied by the TRMC technique. The TRMC set-up consists of two main components: (1) a pulse light source, which has the objective to photo-excite the samples and (2) microwave source. A Gunn diode of Kα band at 30 GHz was used to generate the incident microwaves. A tunable laser in the range between 220 and 2000 nm (NT342B; EKSPLA, Vilnius, Lithuania) was used as a light source. It was equipped with an optical parametric oscillator (OPO). The laser delivered 8 ns FWMH pulses with a frequency of 10 Hz. The selected excitation wavelengths were 355 and 410 nm, with a light energy density of 747.6 μJ × cm−2 and 2.6 mJ × cm−2, respectively. 119 Catalysts 2019, 9, 338 References Metal–organic frameworks: A new class of porous materials. Microporous Mesoporous Mater. 2004, 73, 3–14. [CrossRef] 11. Nguyen, L.T.L.; Nguyen, T.T.; Nguyen, K.D.; Phan, N.T.S. Metal–organic framework MOF-199 as an eﬃci heterogeneous catalyst for the aza-Michael reaction. Appl. Catal. Gen. 2012, 425–426, 44–52. [CrossRef] Chui, S.S.-Y.; Lo, S.M.-F.; Charmant, J.P.H.; Orpen, A.G.; Williams, I.D. A Chemically Functionalizable 12. Chui, S.S.-Y.; Lo, S.M.-F.; Charmant, J.P.H.; Orpen, A.G.; Williams, I.D. A Chemical Nanoporous Material [Cu3(TMA)2(H2O)3]n. Science 1999, 283, 1148–1150. [CrossRef] 13. Prestipino, C.; Regli, L.; Vitillo, J.G.; Bonino, F.; Damin, A.; Lamberti, C.; Zecchina, A.; Solari, P.L.; Kongshaug, K.O.; Bordiga, S. Local Structure of Framework Cu(II) in HKUST-1 Metallorganic Framework: Spectroscopic Characterization upon Activation and Interaction with Adsorbates. Chem. Mater. 2006, 18, 1337–1346. [CrossRef] 14. Kumar, P.; Vellingiri, K.; Kim, K.-H.; Brown, R.J.C.; Manos, M.J. Modern progress in metal-organic frameworks and their composites for diverse applications. Microporous Mesoporous Mater. 2017, 253, 251–265. [CrossRef] 15. Wen, M.; Mori, K.; Kuwahara, Y.; An, T.; Yamashita, H. Design and architecture of metal organic frameworks for visible light enhanced hydrogen production. Appl. Catal. B Environ. 2017, 218, 555–569. [CrossRef] 16. Zhang, T.; Lin, W. Metal–organic frameworks for artiﬁcial photosynthesis and photocatalysis. Chem. Soc. Rev. 2014, 43, 5982–5993. [CrossRef] [PubMed] 17. Deng, X.; Li, Z.; García, H. Visible Light Induced Organic Transformations Using Metal-Organic-Frameworks (MOFs). Chem. Eur. J. 2017, 23, 11189–11209. [CrossRef] 18. Wang, C.-C.; Li, J.-R.; Lv, X.-L.; Zhang, Y.-Q.; Guo, G. Photocatalytic organic pollutants degradation in metal–organic frameworks. Energy Environ. Sci. 2014, 7, 2831–2867. [CrossRef] 19. Dias, E.M.; Petit, C. Towards the use of metal–organic frameworks for water reuse: a review of the recent advances in the ﬁeld of organic pollutants removal and degradation and the next steps in the ﬁeld. J. Mater. Chem. A 2015, 3, 22484–22506. [CrossRef] 20. Fang, Y.; Ma, Y.; Zheng, M.; Yang, P.; Asiri, A.M.; Wang, X. Metal–organic frameworks for solar energy conversion by photoredox catalysis. Coord. Chem. Rev. 2018, 373, 83–115. [CrossRef] 21. Qiu, J.; Zhang, X.; Feng, Y.; Zhang, X.; Wang, H.; Yao, J. Modiﬁed metal-organic frameworks as photocatalysts. Appl. Catal. B Environ. 2018, 231, 317–342. [CrossRef] 22. Aguilera-Sigalat, J.; Bradshaw, D. Synthesis and applications of metal-organic framework–quantum dot (QD@MOF) composites. Coord. Chem. Rev. 2016, 307, 267–291. [CrossRef] 23. Li, R.; Hu, J.; Deng, M.; Wang, H.; Wang, X.; Hu, Y.; Jiang, H.-L.; Jiang, J.; Zhang, Q.; Xie, Y.; et al. References 1. Andrews, J.; Shabani, B. Re-envisioning the role of hydrogen in a sustainable energy economy. Int. J. Hyd Energy 2012, 37, 1184–1203. [CrossRef] 2. Dincer, I.; Acar, C. Review and evaluation of hydrogen production methods for better sustainability. Int. J. Hydrog. Energy 2015, 40, 11094–11111. [CrossRef] 3. Dodds, P.E.; Staﬀell, I.; Hawkes, A.D.; Li, F.; Grünewald, P.; McDowall, W.; Ekins, P. Hydrogen and fuel cell technologies for heating: A review. Int. J. Hydrog. Energy 2015, 40, 2065–2083. [CrossRef] 4. Liao, C.-H.; Huang, C.-W.; Wu, J.C.S. Hydrogen Production from Semiconductor-based Photocatalysis via W t S litti C t l t 2012 2 490 516 [C R f] 4. Liao, C.-H.; Huang, C.-W.; Wu, J.C.S. Hydrogen Production from Semiconductor-based Photocatalysis via Water Splitting. Catalysts 2012, 2, 490–516. [CrossRef] 4. Liao, C.-H.; Huang, C.-W.; Wu, J.C.S. Hydrogen Production from Semiconductor-based Photocatalysis via Water Splitting. Catalysts 2012, 2, 490–516. [CrossRef] 4. Liao, C.-H.; Huang, C.-W.; Wu, J.C.S. Hydrogen Produc Water Splitting. Catalysts 2012, 2, 490–516. [CrossRef] 5. Bozoglan, E.; Midilli, A.; Hepbasli, A. Sustainable assessment of solar hydrogen production techniques. Energy 2012, 46, 85–93. [CrossRef] 5. Bozoglan, E.; Midilli, A.; Hepbasli, A. Sustainable assessment of solar hydrogen production techniques. Energy 2012, 46, 85–93. [CrossRef] 6. Christoforidis, K.C.; Fornasiero, P. Photocatalytic Hydrogen Production: A Rift into the Future Energy Supply. ChemCatChem 2017, 9, 1523–1544. [CrossRef] 6. Christoforidis, K.C.; Fornasiero, P. Photocatalytic Hydrogen Production: A Rift into the Future Energy Supply. ChemCatChem 2017, 9, 1523–1544. [CrossRef] 7. Babu, V.J.; Vempati, S.; Uyar, T.; Ramakrishna, S. Review of one-dimensional and two-dimensional nanostructured materials for hydrogen generation. Phys. Chem. Chem. Phys. 2015, 17, 2960–2986. [CrossRef] [PubMed] 7. Babu, V.J.; Vempati, S.; Uyar, T.; Ramakrishna, S. Review of one-dimensional and two-dimensional nanostructured materials for hydrogen generation. Phys. Chem. Chem. Phys. 2015, 17, 2960–2986. [CrossRef] [PubMed] 8. Guo, L.; Jing, D.; Liu, M.; Chen, Y.; Shen, S.; Shi, J.; Zhang, K. Functionalized nanostructures for enhanced photocatalytic performance under solar light. Beilstein J. Nanotechnol. 2014, 5, 994–1004. [CrossRef] 8. Guo, L.; Jing, D.; Liu, M.; Chen, Y.; Shen, S.; Shi, J.; Zhang, K. Functionalized nanostructures for enhanced photocatalytic performance under solar light. Beilstein J. Nanotechnol. 2014, 5, 994–1004. [CrossRef] 120 Catalysts 2019, 9, 338 9. Kumar, S.; Kumar, A.; Bahuguna, A.; Sharma, V.; Krishnan, V. Two-dimensional carbon-based nanocomposites for photocatalytic energy generation and environmental remediation applications. Beilstein J. Nanotechnol. 2017, 8, 1571–1600. [CrossRef] 10. Rowsell, J.L.C.; Yaghi, O.M. References Integration of an Inorganic Semiconductor with a Metal–Organic Framework: A Platform for Enhanced Gaseous Photocatalytic Reactions. Adv. Mater. 2014, 26, 4783–4788. [CrossRef] 24. Binh, N.T.; Thu, P.T.; Le, N.T.H.; Tien, D.M.; Khuyen, H.T.; Giang, L.T.K.; Huong, N.T.; Lam, T.D. Study on preparation and properties of a novel photo–catalytic material based on copper–centred metal–organic frameworks (Cu–MOF) and titanium dioxide. Int. J. Nanotechnol. 2015, 12, 447–455. [CrossRef] 25. Abedi, S.; Morsali, A. Ordered Mesoporous Metal–Organic Frameworks Incorporated with Amorphous TiO2 As Photocatalyst for Selective Aerobic Oxidation in Sunlight Irradiation. ACS Catal. 2014, 4, 1398–1403. [CrossRef] 26. Li, R.; Wu, S.; Wan, X.; Xu, H.; Xiong, Y. Cu/TiO2 octahedral-shell photocatalysts derived from metal–organic framework@semiconductor hybrid structures. Inorg. Chem. Front. 2016, 3, 104–110. [CrossRef] 27. Canivet, J.; Fateeva, A.; Guo, Y.; Coasne, B.; Farrusseng, D. Water adsorption in MOFs: Fundamentals and applications. Chem. Soc. Rev. 2014, 43, 5594–5617. [CrossRef] 28. Chen, C.; Wu, T.; Yang, D.; Zhang, P.; Liu, H.; Yang, Y.; Yang, G.; Han, B. Catalysis of photooxidation reactions through transformation between Cu2+ and Cu+ in TiO2–Cu–MOF composites. Chem. Commun. 2018, 54, 5984–5987. [CrossRef] 29. Ahmed, A.; Robertson, C.M.; Steiner, A.; Whittles, T.; Ho, A.; Dhanak, V.; Zhang, H. Cu(I)Cu(II)BTC, a microporous mixed-valence MOF via reduction of HKUST-1. RSC Adv. 2016, 6, 8902–8905. [CrossRef] 30. Bahruji, H.; Bowker, M.; Davies, P.R.; Al-Mazroai, L.S.; Dickinson, A.; Greaves, J.; James, D.; Millard, L.; Pedrono, F. Sustainable H2 gas production by photocatalysis. J. Photochem. Photobiol. Chem. 2010, 216, 115–118. [CrossRef] 121 Catalysts 2019, 9, 338 31. Szanyi, J.; Daturi, M.; Clet, G.; Baer, D.R.; Peden, C.H.F. Well-studied Cu–BTC still serves surprises: Evidence for facile Cu2+/Cu+ interchange. Phys. Chem. Chem. Phys. 2012, 14, 4383–4390. [CrossRef] g y y 32. Fu, Q.; Xie, K.; Tan, S.; Ren, J.M.; Zhao, Q.; Webley, P.A.; Qiao, G.G. The use of reduced copper metal–organic g y y 32. Fu, Q.; Xie, K.; Tan, S.; Ren, J.M.; Zhao, Q.; Webley, P.A.; Qiao, G.G. The use of reduced copper metal–organic frameworks to facilitate CuAAC click chemistry. Chem. Commun. 2016, 52, 12226–12229. [CrossRef] y 33. Loera-Serna, S.; Oliver-Tolentino, M.A.; de Lourdes López-Núñez, M.; Santana-Cruz, A.; Guzmán-Vargas, A.; Cabrera-Sierra, R.; Beltrán, H.I.; Flores, J. Electrochemical behavior of [Cu3(BTC)2] metal-organic framework: 33. Loera-Serna, S.; Oliver-Tolentino, M.A.; de Lourdes López-Núñez, M.; Santana-Cruz, A.; Guzmán-Vargas, A.; Cabrera-Sierra, R.; Beltrán, H.I.; Flores, J. Electrochemical behavior of [Cu3(BTC)2] metal-organic framework: The eﬀect of the method of synthesis. J. Alloys Compd. Shinya Higashimoto Department of Applied Chemistry, Faculty of Engineering, Osaka Institute of Technology, 5-16-1 Omiya, Asahi-ku, Osaka 535-8585, Japan; shinya.higashimoto@oit.ac.jp; Tel.: +81-(0)6-6954-4283 Department of Applied Chemistry, Faculty of Engineering, Osaka Institute of Technology, 5-16-1 Omiya, Asahi-ku, Osaka 535-8585, Japan; shinya.higashimoto@oit.ac.jp; Tel.: +81-(0)6-6954-4283 Received: 15 January 2019; Accepted: 14 February 2019; Published: 22 February 2019 Received: 15 January 2019; Accepted: 14 February 2019; Published: 22 February 2019 Abstract: Titanium dioxide (TiO2) is one of the most practical and prevalent photo-functional materials. Many researchers have endeavored to design several types of visible-light-responsive photocatalysts. In particular, TiO2-based photocatalysts operating under visible light should be urgently designed and developed, in order to take advantage of the unlimited solar light available. Herein, we review recent advances of TiO2-based visible-light-sensitive photocatalysts, classiﬁed by the origins of charge separation photo-induced in (1) bulk impurity (N-doping), (2) hetero-junction of metal (Au NPs), and (3) interfacial surface complexes (ISC) and their related photocatalysts. These photocatalysts have demonstrated useful applications, such as photocatalytic mineralization of toxic agents in the polluted atmosphere and water, photocatalytic organic synthesis, and artiﬁcial photosynthesis. We wish to provide comprehension and enlightenment of modiﬁcation strategies and mechanistic insight, and to inspire future work. Keywords: Titanium dioxide (TiO2); visible-light-sensitive photocatalyst; N-doped TiO2; plasmonic Au NPs; interfacial surface complex (ISC); selective oxidation; decomposition of VOC; carbon nitride (C3N4); alkoxide; ligand to metal charge transfer (LMCT) Catalysts 2019, 9, 201; doi:10.3390/catal9020201 www.mdpi.com/journal/catalysts References 2012, 540, 113–120. [CrossRef] 34. Dhumal, N.R.; Singh, M.P.; Anderson, J.A.; Kiefer, J.; Kim, H.J. Molecular Interactions of a Cu-Based Metal–Organic Framework with a Conﬁned Imidazolium-Based Ionic Liquid: A Combined Density Functional Theory and Experimental Vibrational Spectroscopy Study. J. Phys. Chem. C 2016, 120, 3295–3304. [CrossRef] 35. Borfecchia, E.; Maurelli, S.; Gianolio, D.; Groppo, E.; Chiesa, M.; Bonino, F.; Lamberti, C. Insights into Adsorption of NH3 on HKUST-1 Metal–Organic Framework: A Multitechnique Approach. J. Phys. Chem. C 2012, 116, 19839–19850. [CrossRef] 36. Wen, L.-L.; Wang, F.; Feng, J.; Lv, K.-L.; Wang, C.-G.; Li, D.-F. Structures, Photoluminescence, and Photocatalytic Properties of Six New Metal−Organic Frameworks Based on Aromatic Polycarboxylate Acids and Rigid Imidazole-Based Synthons. Cryst. Growth Des. 2009, 9, 3581–3589. [CrossRef] 37. Wen, L.; Zhao, J.; Lv, K.; Wu, Y.; Deng, K.; Leng, X.; Li, D. Visible-Light-Driven Photocatalysts of Metal–Organic Frameworks Derived from Multi-Carboxylic Acid and Imidazole-Based Spacer. Cryst. Growth Des. 2012, 12, 1603–1612. [CrossRef] 38. Schlichte, K.; Kratzke, T.; Kaskel, S. Improved synthesis, thermal stability and catalytic properties of the metal-organic framework compound Cu3(BTC)2. Microporous Mesoporous Mater. 2004, 73, 81–88. [CrossRef] 39. DeCoste, J.B.; Peterson, G.W.; Schindler, B.J.; Killops, K.L.; Browe, M.A.; Mahle, J.J. The eﬀect of water adsorption on the structure of the carboxylate containing metal–organic frameworks Cu-BTC, Mg-MOF-74, and UiO-66. J. Mater. Chem. A 2013, 1, 11922–11932. [CrossRef] 40. Chen, X.; Mao, S.S. Titanium Dioxide Nanomaterials: Synthesis, Properties, Modiﬁcations, and Applications. Chem. Rev. 2007, 107, 2891–2959. [CrossRef] 41. Tranchemontagne, D.J.; Hunt, J.R.; Yaghi, O.M. Room temperature synthesis of metal-organic frameworks: MOF-5, MOF-74, MOF-177, MOF-199, and IRMOF-0. Tetrahedron 2008, 64, 8553–8557. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 122 catalysts catalysts 1. Introduction Titanium dioxide (TiO2) is one of the most practical and prevalent photo-functional materials, since it is chemically stable, abundant (Ti: 10th highest Clarke number), nontoxic, and cost-effective. In recent years, a great deal of attention has been directed towards TiO2 photocatalysis for useful applications such as photocatalytic mineralization of toxic agents in the polluted atmosphere and water, photocatalytic organic synthesis, and artiﬁcial photosynthesis [1–20]. The TiO2 involving Ti3+ sites that are oxygen-deﬁcient at the impurity level exhibits n-type semiconductor. The photocatalytic activities of TiO2 strongly depend on crystal structures (anatase, brookite, and rutile), crystallinity, crystalline plane, morphology, particle sizes, defective sites, and surface OH groups. The valence band (V.B.) and conduction band (C.B.) of TiO2 consist of O 2p and Ti 3d orbitals, respectively, and their band gap (forbidden band) is circa ~3.0–3.2 eV (~410–380 nm). Photo-irradiation (hv > 3.2 eV) of the TiO2 photocatalyst leads to band gap excitation, resulting in charge separation of electrons into the C.B. and the holes in the V.B. These photo-formed electrons and holes simultaneously work as electron donors and acceptors, respectively, on the photocatalyst surface, thus enabling the photocatalytic reactions. Details are given in other articles and reviews [21–25]. UV light reaching the earth surface represents only a very small fraction (4%) of the solar energy available. Therefore, many researchers have endeavored to design several types of visible-light-responsive photocatalyst. In particular, TiO2-based photocatalysts operating under visible light should be urgently designed and developed, in order to take advantage of the unlimited solar light available. In the late 1990s, Anpo et al. ﬁrst reported that TiO2 doped with Cr, V, and Fe cations by ion implantation operates under visible light irradiation. They exhibited red shift of the band-edge of the 123 Catalysts 2019, 9, 201 TiO2, resulting in decomposition of NO into N2, O2, and N2O [26]. This work accelerated subsequent works for the design and development of visible-light-responsive photocatalysts. Recently, much attention has been paid to visible-light-responsive TiO2 prepared by: doping with nitrogen (N), carbon (C), and sulfur (S) ions etc.; surface plasmonic effects with Au or Ag nanoparticles (NPs); the interfacial surface complex (ISC); coupling with visible-light-sensitive hetero-semiconductors (cadmium sulﬁde, carbon nitride etc.); and dye-sensitized photocatalysts. In fact, some photocatalysts are considered to work under similar principles. Along these backgrounds, this review focuses on the recent advances of the visible-light-sensitive TiO2 photocatalyst. 1. Introduction These advances have been classiﬁed by the origin of charge separation photo-induced in (1) the bulk impurity (N-doping), (2) hetero-junction of metal (Au NPs), and (3) the interfacial surface complex (ISC) (See Figure 1). They have been well characterized by several spectroscopic techniques, and applied for mineralization of volatile organic compounds (VOC), water splitting to produce H2, and ﬁne organic synthesis. Figure 1. Visible-light-sensitive TiO2 photocatalyst modiﬁed by (1) nitrogen-doping, (2) plasmonic Au nanoparticles (NPs), and (3) interfacial surface complex (ISC). Figure 1. Visible-light-sensitive TiO2 photocatalyst modiﬁed by (1) nitrogen-doping, (2) plasmonic Au nanoparticles (NPs), and (3) interfacial surface complex (ISC). 2. Nitrogen-doped TiO2 Photocatalysts In 1986, Sato and co-workers ﬁrst explored the photocatalytic activity of nitrogen-doped TiO2 (N-doped TiO2) photocatalysts for the oxidation of gaseous ethane and carbon monoxide [27]. They found that N-doped TiO2 photocatalyst exhibited a superior photocatalytic activity to pure TiO2 under visible light irradiation. Later, in 2001, Asahi et al. demonstrated visible-light-induced complete photo-oxidation of gaseous CH3CHO (one of VOCs) to CO2 with an N-doped TiO2 photocatalyst [28]. In this section, fundamental synthetic routes, characterizations, and application of photocatalytic reactions are highlighted. 2.1. Synthesis of N-doped TiO2 Photocatalyst 2.1.2. Wet Processes A sol-gel method can be employed for the preparation of N-doped TiO2 powder. Typically, NH3 aq. (NH4OH) is added to a solution of titanium (IV) isopropoxide (TTIP) [31–33] to form titanium hydroxide involving N-species. The precipitate was dried, followed by calcination at ~400–450 ◦C in air to obtain a yellowish TiO2 powder. 2.1. Synthesis of N-doped TiO2 Photocatalyst N-doped TiO2 was prepared by employing several procedures and materials. Details are given in Reference [13]. Preparation methods for N-doped TiO2 photocatalysts can be classiﬁed into two categories: dry processes and wet processes. 124 Catalysts 2019, 9, 201 Catalysts 2019, 9, 201 2.1.1. Dry Processes Typically, N-doped TiO2 powder can be prepared by the nitriﬁcation of TiO2 in an ammonia (NH3) gas ﬂow at high temperature [28,29]. The amount of N doping into the TiO2 can be controlled by annealing temperatures in the range of 550−600 ◦C under an NH3 ﬂow. However, a large number of O vacancies are introduced into the N-doped TiO2 with increasing annealing temperature, since the NH3 decomposes into N2 and H2 at high temperature, and TiO2 is simultaneously reduced by H2 [30]. Figure 2 shows schematics of N-doping into TiO2, accompanied by the formation of oxygen vacancies to exhibit the n-type semiconductor. 7L 7L 7L 1 2 2 2 2 2 K 7L 7L 7L 1 2 2 2 2 7L H K 7L 7L 7L 1 2 2 2 2 7L H D E F 7L 2 2 2 2 2 2 + 7L 7L 7L 2 2 2 2 2 2 7L 2 2 1+ 7L2 1GRSHG7L2 SW\SH QW\SH 7L 7L 7L 1 2 2 2 2 2 K 7L 7L 7L 1 2 2 2 2 2 7L 7L 7L 1 2 2 2 2 2 K 7L 7L 7L 1 2 2 2 2 7L H K 7L 7L 7L 1 2 2 2 2 7L H K 7L 7L 7L 1 2 2 2 2 7L H 7L 7L 7L 1 2 2 2 2 7L H D E F 7L 2 2 2 2 2 2 + 7L 7L 7L 2 2 2 2 2 2 7L 2 2 7L 7L 7L 2 2 2 2 2 2 7L 7L 7L 2 2 2 2 2 2 7L 2 2 1+ 7L2 1GRSHG7L2 SW\SH QW\SH 7L 7L 7L 7L 7L 7L 7L 7L QW\SH QW\SH QW\SH QW\SH 1GRSHG7L2 1GRSHG7L2 Figure 2. When the N3−is replaced with lattice O2−ions in the TiO2 lattice, the hole (h+) is formed in order to compensate for the charge balance (p-type semiconductor) (a). However, an oxygen vacancy is produced by the reduction with H2, which is formed by the decomposition of NH3 to produce an oxygen vacancy and excess electrons (b). As a consequence, N3−doped into TiO2 (N-doped TiO2) involves electrons located at N 2p and Ti 3d sites at impurity levels (n-type semiconductor) (c). 2.2.1. DFT Calculations DFT calculations demonstrated the electronic structures of the N-doped TiO2 photocatalyst (see Figure 3). The substitution of N with lattice O of the N-doped TiO2 exhibits band gap narrowing (circa 0.1 eV) caused by mixing orbitals of N 2p with O 2p, resulting in the negative shift of the valence band edge. On the other hand, the interstitial N is localized to impurity states (N 2p levels) above the V.B. (circa 0.7 eV) in the mid-band gap. Therefore, the oxidation power of photo-induced holes on the N 2p is lower than on the O 2p in the TiO2 lattice. ,QWHUVWLWLDO1GRSHG7L2 7L 7L 2 2 2 2 1 7L 9% &% H9 1S 6XEVWLWXWLRQDO 1GRSHG7L2 7L 7L 1 2 2 2 7L 9% &% H9 1S ,QWHUVWLWLDO1GRSHG7L2 7L 7L 2 2 2 2 1 7L 9% &% H9 1S ,QWHUVWLWLDO1GRSHG7L2 7L 7L 2 2 2 2 1 7L 7L 7L 2 2 2 2 1 7L 9% &% H9 1S 6XEVWLWXWLRQDO 1GRSHG7L2 7L 7L 1 2 2 2 7L 9% &% H9 1S 6XEVWLWXWLRQDO 1GRSHG7L2 7L 7L 1 2 2 2 7L 7L 7L 1 2 2 2 7L 9% &% H9 1S Figure 3. Schematic illustration of structures and their corresponding energy bands for substitutional and interstitial N species in the N-doped TiO2, together with photo-induced electronic processes. 6XEVWLWXWLRQDO 1GRSHG7L2 7L 7L 1 2 2 2 7L 9% &% H9 1S 6XEVWLWXWLRQDO 1GRSHG7L2 7L 7L 1 2 2 2 7L 9% &% H9 1S 6XEVWLWXWLRQDO 1GRSHG7L2 7L 7L 1 2 2 2 7L 7L 7L 1 2 2 2 7L 9% &% H9 1S Figure 3. Schematic illustration of structures and their corresponding energy bands for substitutional and interstitial N species in the N-doped TiO2, together with photo-induced electronic processes. 2.2. N-states in N-doped TiO2 One of the major concerns is to understand the physico-chemical nature of the N species in N-doped TiO2, which are responsible for the visible light sensitivity. They were characterized by density functional theory (DFT) calculations, X-rap photoelectron spectroscopy (XPS), Ultraviolet-visible (UV-vis) and electron paramagnetic resonance (EPR) spectroscopy. 125 Catalysts 2019, 9, 201 Catalysts 2019, 9, 201 2.2.4. Electron Paramagnetic Resonance (EPR) Spectra N species in the N-doped TiO2 are present at either diamagnetic (N−) or paramagnetic (N•) bulk centers, which are responsible for the visible light sensitivity [31,37]. The EPR measurements can detect the paramagnetic (N•) bulk centers (see Figure 5). One type, of three lines with a hyperﬁne tensor (g = 2.006 and A = 32.0 G) splitting by nuclear spin of nitrogen (I = 1), was observed. The signal intensity of N• radicals increased when the light was turned on, while the signal intensity signiﬁcantly decreased when the light was turned off. In general, the paramagnetic interaction between N species and O2 makes EPR signals disappear. However, they were remarkably enhanced in the presence of O2 under λ > 420 nm, while its signal intensity still remained to some extent even after the light was turned off. These results suggest that N-species are located in bulk inside the TiO2, and visible light irradiation of the N-doped TiO2 exhibits effective charge separation to form holes (N• radicals) and electrons, which participate in the oxidation and reduction of reactant molecules, respectively. OLJKWRQ 7LPHPLQ RII I1 I1 RQ RII RQ RII >,,,@ LQYDFXXP LQ$U LQ2 OLJKWRQ 7LPHPLQ RII I1 I1 RQ RII RQ RII >,,,@ LQYDFXXP LQ$U LQ2 Figure 5. Schematic illustration of [I] formation of paramagnetic ·N by the excitation of diamagnetic N−species. Electron paramagnetic resonance (EPR) signal [II] of ·N radicals on N–TiO2, and the relative signal intensity of IN/IN0 [III] under vacuum, in the presence of argon (Ar) or O2 (400 Pa) [37]. IN0 and IN show the intensity due to ·N radicals at the initial and measured time, respectively. OLJKWRQ 7LPHPLQ RII I1 I1 RQ RII RQ RII >,,,@ LQYDFXXP LQ$U LQ2 OLJKWRQ 7LPHPLQ RII I1 I1 RQ RII RQ RII >,,,@ LQYDFXXP LQ$U LQ2 7LPHPLQ 7LPHPLQ Figure 5. Schematic illustration of [I] formation of paramagnetic ·N by the excitation of diamagnetic N−species. Electron paramagnetic resonance (EPR) signal [II] of ·N radicals on N–TiO2, and the relative signal intensity of IN/IN0 [III] under vacuum, in the presence of argon (Ar) or O2 (400 Pa) [37]. IN0 and IN show the intensity due to ·N radicals at the initial and measured time, respectively. 2.2.2. XPS Spectra XPS analysis can conﬁrm the oxidative states of the N species and bonding states in the N-doped TiO2 (See Figure 4I). N 1s XPS peaks at a binding energy in the range of ~396–400 eV showed different oxidative states of the N species. By the combination of the DFT calculations [31], it was identiﬁed that the N 1s XPS peaks at ~396–397 eV are due to the substitution of N with the lattice O of TiO2 [13,34], while those at ~399–400 eV are due to the interstitial N in the form of NOx or NHx [13,31,35,36]. Figure 4. XPS [I] and UV-vis absorption spectra [II] of (a) N-doped TiO2 nanoball ﬁlm [34], and (b) N-doped TiO2 prepared by the sol-gel method [36]. Figure 4. XPS [I] and UV-vis absorption spectra [II] of (a) N-doped TiO2 nanoball ﬁlm [34], and (b) N-doped TiO2 prepared by the sol-gel method [36]. 126 Catalysts 2019, 9, 201 2.2.3. Optical Properties The UV-vis absorption spectra of the N-doped TiO2 are shown in Figure 4II. The N-doped TiO2 with the substitution of N exhibited band gap narrowing from 3.1 to 2.8 eV. On the other hand, the N-doped TiO2 prepared by the sol-gel method exhibited visible light absorption up to 540 nm (2.3 eV), due to the electronic transition from localized N doping level to the C.B. of the TiO2, while band-narrowing was not observed. These results are in good agreement with the DFT calculations. 2.2.4. Electron Paramagnetic Resonance (EPR) Spectra 2.2.4. Electron Paramagnetic Resonance (EPR) Spectra 2.2.5. Photo-Electrochemical Properties Nakamura et al. investigated the photo-electrochemical oxidation power of the N-doped TiO2 by employing several electron donors [38]. Figure 6 shows that the photo-induced hole on the N 2p level can directly oxidize only I−ions under visible light illumination, while I−, SCN−, Br−, and H2O are oxidized by the hole on the V.B. under UV light illumination. Therefore, the oxidation power of the holes induced on the N 2p level is lower than that of those on the O 2p on the V.B. Tang et al. studied the dynamics of photogenerated electrons and holes on the N-doped TiO2 using transient absorption spectroscopy [39]. They concluded that the lack of activity of nanocrystalline N-doped TiO2 ﬁlm for photocatalytic water oxidation is due to rapid electron–hole recombination. 127 Catalysts 2019, 9, 201 Catalysts 2019, 9, 201 On the other hand, Higashimoto et al. investigated the photo-electrochemical reduction power of the N-doped TiO2 (see Figure 7) [33]. When the N-doped TiO2 was photo-excited under visible light irradiation, the photo-induced electrons were accumulated on the oxygen vacancies of TiO2. Subsequently, when various kinds of redox species as electron acceptors were introduced into the photo-charged N–TiO2, the accumulated electrons could reduce O2 molecules, Pt4+, Ag+, and Au3+ ions, but not MV2+, H+, and Cu2+ ions. In principle, the N-doped TiO2 has the potential to reduce H+/H2, but many oxygen vacancies involved in the bulk TiO2 could inﬂuence the drastic charge recombination. In particular, photo-induced electrons trapped at the oxygen vacancies (mainly γ region) could reduce O2 molecules to form such active oxygen species as hydrogen peroxide (H2O2), resulting in further oxidation of organic substrates. Figure 6. Schematic illustration of proposed energy bands for the N-doped TiO2, together with some photo-induced electronic processes. E: equilibrium redox potentials for one electron transfer [38]. Figure 6. Schematic illustration of proposed energy bands for the N-doped TiO2, together with some photo-induced electronic processes. E: equilibrium redox potentials for one electron transfer [38]. Concentrations of VOC are (a) 0.5 M and (b) 50 mM. 2.2.5. Photo-Electrochemical Properties 㽢㽢㽢㽢 1GRSLQJOHYHO γ E 9YV1+(S+ 9% &% α K β H 9LVLEOHOLJKW H9 㽢 $J $J9 +2DQGRURUJDQLFFRPSRXQGV R[LGDWLYHSURGXFWV + +9 >$X&O@ $X&O 9 &X &X9 㽢 2 + +22+9 >3W&O@ 3W&O 9 1GRSLQJOHYHO γ E 9YV1+(S+ 9% &% α K β H 9LVLEOHOLJKW H9 㽢 $J $J9 +2DQGRURUJDQLFFRPSRXQGV R[LGDWLYHSURGXFWV + +9 >$X&O@ $X&O 9 &X &X9 㽢 2 + +22+9 >3W&O@ 3W&O 9 Figure 7. Energy levels for sub-band structures of N-doped TiO2 and photo-induced charge transfer into various kinds of redox species under visible light irradiation. The energy levels of sub-bands at the α, β, and γ potential regions (oxygen vacancies) and N-doping levels are also shown. Oxygen vacancies were estimated from the photo-electrochemical measurements. Signs of circle and cross stand for energetically favorable and unfavorable electron transfers, respectively [33]. Figure 7. Energy levels for sub-band structures of N-doped TiO2 and photo-induced charge transfer into various kinds of redox species under visible light irradiation. The energy levels of sub-bands at the α, β, and γ potential regions (oxygen vacancies) and N-doping levels are also shown. Oxygen vacancies were estimated from the photo-electrochemical measurements. Signs of circle and cross stand for energetically favorable and unfavorable electron transfers, respectively [33]. Figure 7. Energy levels for sub-band structures of N-doped TiO2 and photo-induced charge transfer into various kinds of redox species under visible light irradiation. The energy levels of sub-bands at the α, β, and γ potential regions (oxygen vacancies) and N-doping levels are also shown. Oxygen vacancies were estimated from the photo-electrochemical measurements. Signs of circle and cross stand for energetically favorable and unfavorable electron transfers, respectively [33]. 128 Catalysts 2019, 9, 201 Catalysts 2019, 9, 201 2.3. Application to Photocatalytic Decomposition of Volatile Organic Compounds (VOC) 2.3. Application to Photocatalytic Decomposition of Volatile Organic Compounds (VOC) Time proﬁle for the photocatalytic decomposition of gaseous acetaldehyde on the N-doped TiO2 is shown in Figure 8. The N-doped TiO2 exhibited photocatalytic activity 5 times greater than TiO2 under visible light irradiation, while they exhibited similar activities under UV light irradiation [28]. Figure 8. Photocatalytic decomposition of gaseous acetaldehyde on the N-doped TiO2 photocatalyst. Evolved CO2 concentration (, •, N-doped TiO2; □, ■, TiO2) [28]. Figure 8. 2.2.5. Photo-Electrochemical Properties Photocatalytic decomposition of gaseous acetaldehyde on the N-doped TiO2 photocatalyst. Evolved CO2 concentration (, •, N-doped TiO2; □, ■, TiO2) [28]. Figure 8. Photocatalytic decomposition of gaseous acetaldehyde on the N-doped TiO2 photocatalyst. Evolved CO2 concentration (, •, N-doped TiO2; □, ■, TiO2) [28]. Table 1 shows that the N-doped TiO2 exhibited photocatalytic activity for the decomposition of several kinds of VOC into CO2 under visible light irradiation (λ > 420 nm). It was observed that the N-doped TiO2 exhibited photocatalytic activity for the decomposition of aldehydes, but little activity for alcohol, acid, ketone, and halogene compounds. The vanadium species was deposited on the N-doped TiO2 (VCl3/N-doped TiO2) by impregnation method. As shown in Table 1, VCl3/N-doped TiO2 showed higher photocatalytic activity for the decomposition of all VOC, in particular, acetic acid or acetone by ~13–16 times more than N-doped TiO2. Therefore, it was conﬁrmed that vanadium species worked as the effective co-catalyst. Table 1. Yields of CO2 for the photocatalytic decomposition of various kinds of volatile organic compounds (VOC) in aqueous solutions with N–TiO2 and VCl3/N-doped TiO2 under visible light irradiation (λ > 420 nm) for 3 h [40]. Table 1. Yields of CO2 for the photocatalytic decomposition of various kinds of volatile organic compounds (VOC) in aqueous solutions with N–TiO2 and VCl3/N-doped TiO2 under visible light irradiation (λ > 420 nm) for 3 h [40]. Entry Reactant Molecules Yields of CO2/μmol N-doped TiO2 VCl3/N-doped TiO2 1 methanol a 0.2 1.1 2 ethanol a 0.3 0.5 3 formaldehyde a 4.6 21.6 4 acetaldehyde a 4.1 35.0 5 formic acid a 0.7 4.8 6 acetic acid a 1.2 17.0 7 acetone a 0.7 11.4 8 ethyl acetate a 1.3 10.6 9 dichloromethane b 2.4 4.1 10 trichloromethane b 1.5 4.1 11 1, 1-dichloroethane b 0.7 4.8 12 trans-1, 2-dichloroethylene b 1.0 5.7 Concentrations of VOC are (a) 0.5 M and (b) 50 mM. 129 Catalysts 2019, 9, 201 Furthermore, effects of co-catalysts (48 metal ions using nitrate, sulfate, chloride, acetate, and oxide precursors) deposited on the N-doped TiO2 for the photocatalytic activities were examined (See Figure 9) [40]. The bars marked in yellow exhibited higher photocatalytic activities than the N-doped TiO2 by itself. In particular, N-doped-TiO2-deposited Cu, Fe, V, and Pt oxides exhibited high photocatalytic activities. The local structures of the co-catalysts were characterized by XPS. 2.2.5. Photo-Electrochemical Properties It was observed that Cu loaded N-doped TiO2 involves cuprous oxide (Cu2O) or Cu hydroxides, Fe loaded N-doped TiO2 involves clusters containing Fe–O bonds or Fe2+ hydroxide [41], and Pt loaded N-doped TiO2 involves Pt4+/Pt2+ species [36]. The redox potentials of co-catalysts such as V (+IV/+V), Fe (+II/+III), Cu (+I/+II), and Pt (+III/+IV) were in the range of circa +0.6 to +1.0 V vs. SHE, while the multi-electron reduction of O2 leads to the formation of active oxygen species via O2 + 2H+ + 2e−/ H2O2 (E0 = +0.687 V vs. SHE). Therefore, the co-catalysts, such as Pt, Fe, Cu, and V species, enhance the photocatalytic activity due to the effective electron transfer to O2 (O2 reduction), resulting in the formation of active oxygen species. 17L2 1D&O 0J&O $O12 .&O &D&O 9&O 1+92 .&U2 .0Q2 )H&O &R&O 1L12 &X&O =Q$FD12 7H&O 6E&O 5E&O 6U&O <&O =U&O2 +0R2 5X&O 5K&O 3G&O $J12 &G12 6Q&O %L&O &V&O %D&O +:2 +3W&O +$X&O 3E12 /X$F <E&O 7P$F /D$F &H62 3U$F 1G12 6P12 (X&O G12 7E&O '\&O +R&O (U&O 6HYHUDONLQGVRIPHWDOOLFVDOWVIRUFRFDWDO\VW &2 HYROXWLRQμPRO K 17L2 1D&O 0J&O $O12 .&O &D&O 9&O 1+92 .&U2 .0Q2 )H&O &R&O 1L12 &X&O =Q$FD12 7H&O 6E&O 5E&O 6U&O <&O =U&O2 +0R2 5X&O 5K&O 3G&O $J12 &G12 6Q&O %L&O &V&O %D&O +:2 +3W&O +$X&O 3E12 /X$F <E&O 7P$F /D$F &H62 3U$F 1G12 6P12 (X&O G12 7E&O '\&O +R&O (U&O 6HYHUDONLQGVRIPHWDOOLFVDOWVIRUFRFDWDO\VW &2 HYROXWLRQμPRO K 6HYHUDONLQGVRIPHWDOOLFVDOWVIRUFRFDWDO\VW 6HYHUDONLQGVRIPHWDOOLFVDOWVIRUFRFDWDO\VW Figure 9. Photocatalytic activities for the decomposition of acetic acid under visible light irradiation (λ > 420 nm) on N-doped TiO2, modiﬁed by various kinds of metal species as co-catalysts. Each metal salt used in this study is shown [40]. Figure 9. Photocatalytic activities for the decomposition of acetic acid under visible light irradiation (λ > 420 nm) on N-doped TiO2, modiﬁed by various kinds of metal species as co-catalysts. Each metal salt used in this study is shown [40]. Figure 9. 3.1. What Is Localized Surface Plasmon Resonance (LSPR)? Localized surface plasmon resonance (LSPR) is an optical phenomenon generated by light when it interacts with conductive nanoparticles (NPs) that are smaller than the incident wavelength. The LSPR is induced by the collective oscillations of delocalized electrons in response to an external electric ﬁeld. The resonance wavelength strongly depends on the size and shape of the NPs, the interparticle distance, and the dielectric property of the surrounding medium. The Au and Ag NPs exhibit unique plasmon absorption [52,53]. The plasmonic Ag NPs are considered to be unstable under illumination, and could be applicable to multi-colored rewritable devices. In this section, we focused on stable plasmonic Au NPs exploited for a visible-light-sensitive photovoltaic fuel cell or photocatalyst [54,55]. 3.2. Preparation and Characterization of Au–TiO2 Photocatalyst 2.2.5. Photo-Electrochemical Properties Photocatalytic activities for the decomposition of acetic acid under visible light irradiation (λ > 420 nm) on N-doped TiO2, modiﬁed by various kinds of metal species as co-catalysts. Each metal salt used in this study is shown [40]. 2.4. C3N4-Modiﬁed TiO2 Compared with N-doped TiO2 Several nitrogen sources such as urea, cyanamid, cyanuric acid, and melamine were employed for the preparation of N-containing TiO2 photocatalyst, i.e., the TiO2 surface is modiﬁed with polymerized carbon nitride (C3N4) [42–51]. The structures of the C, N-species strongly depend on their concentrations. If the C, N species are present in only a small amount, they act as a molecular photosensitizer. At higher amounts they form a C3N4 crystalline semiconductor, which chemically binds to TiO2. The C3N4–TiO2 was systematically synthesized by thermal condensation of cyanuric acid on the TiO2 surface [51]. In fact, H2 was evolved from TEA aq. on the C3N4–TiO2 photocatalyst under visible light irradiation, while the N-doped TiO2 did not exhibit H2 production. From characterization of C3N4–TiO2 by Fourier transformed-infrared (FT-IR), XPS, electrochemical measurements, and DFT calculations, the band structures and photo-induced charge separation mechanisms were demonstrated (Figure 10). The C3N4–TiO2 was found to exhibit photo-induced charge separation through the hetero-coupling of semiconductors between C3N4 and TiO2 on the surface. On the other hand, N-doped TiO2 was photo-sensitized by bulk impurity of the N-doping. It can be assumed that many 130 Catalysts 2019, 9, 201 oxygen vacancies promoted the charge recombination, resulting in weak reduction power in the N-doped TiO2. 9% 7L2 &% H9 H DOFRKRO R[LGDWLYHSURGXFWV + + K 9LVLEOHOLJKW 3KRWRFKDUJLQJ 3WVSHFLHV H9 9% &% &1 9% 7L2 &% H9 H DOFRKRO R[LGDWLYHSURGXFWV + + K 9LVLEOHOLJKW 3KRWRFKDUJLQJ 3WVSHFLHV H9 9% &% &1 Figure 10. Photo-induced charge separation on the C3N4 deposited TiO2 surface [51]. DOFRKRO DOFRKRO Figure 10. Photo-induced charge separation on the C3N4 deposited TiO2 surface [51]. 3. Plasmonic Au NPs Modiﬁed TiO2 3. Plasmonic Au NPs Modiﬁed TiO2 3.1. What Is Localized Surface Plasmon Resonance (LSPR)? 3.2.1. Photodeposition (PD) Methods 3.2.1. Photodeposition (PD) Methods By using the photocatalysis of TiO2, metallic Au was deposited on the TiO2 surface, accompanied by the oxidation of methanol [56,57] or ethanol [58]. Typically, TiO2 powder was suspended in a 50 vol. % aqueous methanol in the presence of HAuCl4·6H2O, purged of air with argon. The suspension was photoirradiated with UV light under magnetic stirring. The temperature of the suspension during photoirradiation was maintained at 298K. The Au/TiO2 photocatalyst was centrifuged, washed with distilled water, dried at 393K, and ground in an agate mortar. 3.2.2. Colloid Photodeposition Operated in the Presence of a Hole Scavenger (CPH) Colloidal Au NPs were prepared using the method reported by Frens [59]. In brief, mixtures of an aqueous tetrachloroauric acid (HAuCl4) solution and sodium citrate were heated and boiled for 1 h. The color of the solution changed from deep blue to deep red. The citrate plays a role in the reduction of Au ions, and the capping agent in suppressing the aggregation of Au NPs. The suspension of TiO2 in an aqueous solution of colloidal Au NPs and oxalic acid was then photo-irradiated at λ > 300 nm at 298 K under argon (Ar). The solids were recovered, washed, and dried to produce Au–TiO2. Details are given in Reference [60]. 131 Catalysts 2019, 9, 201 3.2.3. Deposition Precipitation (DP) Method 3.2.3. Deposition Precipitation (DP) Method Deposition–precipitation (DP) methods were employed for the deposition of a gold (III) species on the TiO2 surface [61,62]. The [AuCl(OH)3]−, main species present at pH 8, adjusted by NaOH aq., reacts with hydroxyl groups of the TiO2 surface to form a grafted hydroxyl–gold compound. The catalyst was then recovered, ﬁltered, washed with deionized water, and dried. Finally, the powder was calcined at ~473–673 K in air. 3.2.4. Characterization of the Au–TiO2 Photocatalyst 3.2.4. Characterization of the Au–TiO2 Photocatalyst The Au–TiO2 photocatalysts were typically characterized by the transmittance electron microscope (TEM) for the particle sizes, and UV-vis absorption for optical properties (See Table 2). Table 2. Particle sizes of Au nanoparticles (NPs) and optical properties of the Au–TiO2 prepared by several techniques. Table 2. Particle sizes of Au nanoparticles (NPs) and optical properties of the Au–TiO2 prepared by several techniques. Entry Au Deposition Methods Particle Sizes/nm Top Peak/nm Ref. 1 PD ~10–60 ~530–610 [56–58] 2 CPH ~12–14 ~550–560 [60,63–66] 13 ~550–620 [67] 3 DP ~2–6 ~550–560 [61] < 5 550 [68–70] Kowalska et al. [56,57] reported that Au–TiO2 photocatalysts with different Au particle sizes (~10–60 nm) were prepared by photo-deposition (entry 1). The particle sizes of Au strongly depend on the particle sizes of the TiO2 polycrystalline structure. The top peak of plasmonic absorption was in the range of ~530–610 nm, depending on the particle sizes of the Au NPs. Tanaka and Kominami et al. [60,63–66] reported unique CPH methods for the preparation of Au-TiO2 (entry 2). The particle sizes were uniformed to be ~12–14 nm, which exhibits plasmonic absorption at ~550–560 nm. Thus, colloidal Au NPs were successfully loaded onto TiO2 without change in the original particle size. Furthermore, the top peak of Au plasmon absorption was found to extend towards 620 nm by simple calcinations of the samples. This phenomenon is due to high contact area between TiO2 and Au NPs without change of particle size [66]. Additionally, Naya et al. [67,68] and Shiraishi et al. [69] employed precipitation deposition methods to deposit small Au NPs (~2–6 nm) on TiO2 (entry 3). 3.3. Application of LSPR of Au–TiO2 to Several Photocatalytic Reactions Au NPs deposited on TiO2 have been used as visible-light-responsive photocatalysts for several chemical reactions: decomposition of VOCs, selective oxidation of an aromatic alcohol, direct water splitting, H2 formation from sacriﬁcial aqueous solutions, and reduction of organic compounds (see Table 3). Several research groups concluded that photocatalytic activities are induced by LSPR of the Au NPs. Some research indicates that small Au NPs (~5 nm) effectively work for the reactions [61,69]. Tanaka and Kominami et al. suggest that two types of Au particles of different sizes loaded onto TiO2 exhibit different functionalities. That is, the larger Au particles contribute to strong light absorption, and the smaller Au particles act as a co-catalyst for H2 evolution [63]. 3.2.4. Characterization of the Au–TiO2 Photocatalyst 132 Catalysts 2019, 9, 201 Table 3. Applications to several photocatalytic reactions on the Au–TiO2 photocatalyst. Table 3. Applications to several photocatalytic reactions on the Au–TiO2 photocatalyst. Entry Photocatalytic Reactions Au Deposition Methods References 1 oxidations of 2-propanol and ethanol oxidation of formic acid PD CPH [56–58] [60] 2 oxidation of thiol to disulﬁde DP [67] oxidation of amine to imine DP [68] oxidation of aromatic alcohol to aldehyde CPH [66] DP [69] oxidation of benzene to phenol PD [70] 3 H2 formation from alcohols CPH [63,71] water splitting into H2 and O2 DP CPH [61] [64,72] 4 reduction of nitrobenzene to aniline CPH [65] 3.4. Application to a Photovoltaic Fuel Cell Operating under Visible Light Irradiation 3.4. Application to a Photovoltaic Fuel Cell Operating under Visible Light Irradiation The Au–TiO2 ﬁlms were found to exhibit the behavior of a photovoltaic fuel cell [54,55]. An anodic photocurrent was yielded on the Au−TiO2 ﬁlm as the visible light was irradiated, while the current was observed neither on a TiO2 ﬁlm under visible light irradiation, nor on the Au−TiO2 ﬁlm when the light was turned off. The short-circuit photocurrent density (Jsc) was strongly inﬂuenced by kinds of donors, and the photocurrent efﬁciency was maximized in the presence of Fe2+ ions. Furthermore, the photocurrent action spectra were closely ﬁtted with the absorption spectrum of the Au NPs deposited on the TiO2 ﬁlm (See Figure 11). Figure 11. Short-circuit photocurrent densities [I] vs. apparent formal potential of different donors on the Au−TiO2 photoanode in acetonitrile/ethylene glycol (v/v 60/40) containing 0.1 M LiNO3 and 0.1 M donors; IPCE [II] of the Au−TiO2 ﬁlm in a N2-saturated acetonitrile and ethylene glycol (v/v: 60/40) solution containing 0.1 M FeCl2 and 0.05 M FeCl3 [55]. Figure 11. Short-circuit photocurrent densities [I] vs. apparent formal potential of different donors on the Au−TiO2 photoanode in acetonitrile/ethylene glycol (v/v 60/40) containing 0.1 M LiNO3 and 0.1 M donors; IPCE [II] of the Au−TiO2 ﬁlm in a N2-saturated acetonitrile and ethylene glycol (v/v: 60/40) solution containing 0.1 M FeCl2 and 0.05 M FeCl3 [55]. 3.5. Mechanisms of Charge Separation 3.5. Mechanisms of Charge Separation The mechanism for the Au plasmon-induced charge separation is shown in Figure 12. Visible light irradiation generates the photo-excited state of the Au NPs by LSPR. The photo-excited electrons are injected into the C.B. of TiO2, while the holes abstracted electrons from a donor in the solution. The Au NPs behave like an intrinsic semiconductor, and the Fermi levels of Au NPs and TiO2 are leveled out, resulting in the formation of Schottky barrier at Au–TiO2 junctions. This band model seems to be similar with dye-sensitized photo-anodic electrodes. 133 Catalysts 2019, 9, 201 [I] [II] Figure 12. Schematic illustration [I] and its energy band levels [II] for the photo-induced charge separation on the Au–TiO2 in the presence of donors [55]. [I] [I] [II] [I] Figure 12. Schematic illustration [I] and its energy band levels [II] for the photo-induced charge separation on the Au–TiO2 in the presence of donors [55]. Recently, Furube et al. studied the plasmon-induced charge transfer mechanisms between Au NPs and TiO2 by means of femtosecond visible pump/infrared probe transient absorption spectroscopy [73]. The electron transfer from the Au NPs to the C.B. of TiO2 was conﬁrmed to occur within 50 fs, and that the electron injection yielded 20–50% upon 550 nm laser excitation. 4.1. Dye-Sensitized TiO2 Photocatalysis 4.1. Dye-Sensitized TiO2 Photocatalysis 4.1. Dye-Sensitized TiO2 Photocatalysis Dye sensitized TiO2 photocatalysis was studied in the late 1990s. The Ru complex, [Ru(bipy)3]2+ grafted on the TiO2 surface exhibits visible light absorption [74,75]. In this system, the excitation of the Ru complex induces electron transfer via metal–ligand charge transfer (MLCT). The photo-induced electrons are then transferred onto TiO2, resulting in photocatalytic water splitting to produce H2. The platinum-chloride-modiﬁed TiO2 system was reported by Kisch et al. [76,77]. Photo-irradiation of Pt(IV) chloride exhibits visible-light absorption to generate the active center, (Pt4+(Cl−)4 + hν → Pt3+Cl0(Cl–)3. The photo-induced electrons are transferred from Pt3+ to C.B. of TiO2 as reductive sites, while the Cl0 work as the oxidative sites, resulting in the redox photocatalytic reactions. Important strategies to develop these types of photocatalysts are to design robust sensitizers adjusted with HOMO-LUMO levels. 4.2. Visible-Light-Responsive TiO2 Photocatalyst Modiﬁed by Phenolic Organic Compounds Strong interaction of phenolic groups in organic compounds with Ti–OH of the TiO2 surface probably forms two types of interfacial surface complexes (ISC, Figure 13I), which exhibits visible light absorption via LMCT. The photocatalysis of the ISC is strongly inﬂuenced by the electronic structures of the ISC (Figure 13II): the ISC with EWG exhibits strong oxidizability under visible light irradiation, and it can favorably oxidize the TEA, together with H2 evolution from deaerated TEA aqueous solutions [78]. The visible light response of the ISC is attributed to electronic excitation from the donor levels (0.7 V above V.B.) to the C.B. of TiO2 (see Figure 14). Therefore, the electronic structures of sensitizers strongly inﬂuence the photocatalytic activities. Ikeda et al. [79] demonstrated that a TiO2 photocatalyst modiﬁed with 1,1′-binaphthalene-2,2′-diol (bn(OH)2) exhibited photocatalytic H2 evolution from deaerated TEA aq. under visible light irradiation. Kamegawa et al. [80] designed a 2, 3-dihydroxynaphthalene (2,3-DN)-modiﬁed TiO2 photocatalyst for the reduction of nitrobenzene to aminobenzene under visible light irradiation. 4.2. Visible-Light-Responsive TiO2 Photocatalyst Modiﬁed by Phenolic Organic Compounds 134 Catalysts 2019, 9, 201 R 2 + 2+ +2 7L 7L 2 2 2 2 + 2+ +2 + 2 + 2 2 2 7L + 2 + 2 7L 7L 2 2 2 2 2 7L 5 5 5 5 2 + 2+ +2 7L 7L 2 2 2 7L 7L 2 2 2 2 + 2+ +2 + 2 + 2 2 2 7L + 2 + 2 2 2 7L + 2 + 2 7L 7L 2 + 2 + 2 7L 7L 2 2 2 2 2 7L 5 5 5 5 R: -OCH3 (MC), -C(CH3)3 (BC) 㸸EDG -COOH (BA), -CN (BN), -SO3Na (TN) 㸸EWG [I] [II] 2 + 2+ +2 7L 7L 2 2 2 2 + 2+ +2 + 2 + 2 2 2 7L + 2 + 2 7L 7L 2 2 2 2 2 7L 5 5 5 5 2 + 2+ +2 7L 7L 2 2 2 7L 7L 2 2 2 2 + 2+ +2 + 2 + 2 2 2 7L + 2 + 2 2 2 7L + 2 + 2 7L 7L 2 + 2 + 2 7L 7L 2 2 2 2 2 7L 5 5 5 5 G H 7LPHK DE (YROYHG+ μPRO F I G H 7LPHK DE (YROYHG+ μPRO F I Figure 13. Schematic illustration for the formation of two types of ISCs [I], and photocatalytic H2 evolution [II] from aq. TEA (10 vol. %) on (a) BC/TiO2, (b) MC/TiO2, (c) CA/TiO2, (d) BA/TiO2, (e) BN/TiO2, and (f) TN/TiO2 [78]. BC: 4-t-butyl catechol, MC: 3-methoxy catechol, CA: catecol, BA: 2,3-dihydroxy benzoic acid; BN: 3,4-dihydroxy benzonitrile; TN: tiron. 5 [II] G H 7LPHK DE (YROYHG+ μPRO F I G H 7LPHK DE (YROYHG+ μPRO F I R: 7LPHK 7LPHK [II] [I] Figure 13. Schematic illustration for the formation of two types of ISCs [I], and photocatalytic H2 evolution [II] from aq. TEA (10 vol. %) on (a) BC/TiO2, (b) MC/TiO2, (c) CA/TiO2, (d) BA/TiO2, (e) BN/TiO2, and (f) TN/TiO2 [78]. 4.2. Visible-Light-Responsive TiO2 Photocatalyst Modiﬁed by Phenolic Organic Compounds BC: 4-t-butyl catechol, MC: 3-methoxy catechol, CA: catecol, BA: 2,3-dihydroxy benzoic acid; BN: 3,4-dihydroxy benzonitrile; TN: tiron. Figure 13. Schematic illustration for the formation of two types of ISCs [I], and photocatalytic H2 evolution [II] from aq. TEA (10 vol. %) on (a) BC/TiO2, (b) MC/TiO2, (c) CA/TiO2, (d) BA/TiO2, (e) BN/TiO2, and (f) TN/TiO2 [78]. BC: 4-t-butyl catechol, MC: 3-methoxy catechol, CA: catecol, BA: 2,3-dihydroxy benzoic acid; BN: 3,4-dihydroxy benzonitrile; TN: tiron. 7($7($ H 7L 2 2 7L 2 2 2 1 + + 9LVLEOHOLJKW E 9YV$J$J&O S+ K 3W 9% &% 7L2 䕿 EIE ,6& H9 E++ 'RQRUOHYHO 7($7($ H 7L 2 2 7L 2 2 2 1 + + 9LVLEOHOLJKW E 9YV$J$J&O S+ K 3W 9% &% 7L2 䕿 EIE ,6& H9 E++ 'RQRUOHYHO Figure 14. Schematic illustration of photo-induced charge separation on the BN/TiO2 for H2 evolution from TEA aq. in the presence of Pt as co-catalyst under visible light irradiation [78]. 7L2 7L2 Figure 14. Schematic illustration of photo-induced charge separation on the BN/TiO2 for H2 evolution from TEA aq. in the presence of Pt as co-catalyst under visible light irradiation [78]. On the other hand, the phenolic compounds were degraded on the TiO2 in the presence of O2 under visible light (λ > 420 nm) illumination, producing Cl– and CO2 [81]. The ISC formed by the interaction of phenolic compounds with TiO2 exhibited self-degradation. It was proposed that an electronic transition occurs from the ISC to the C.B. of TiO2 to form active oxygen species, which also participate in the oxidative degradation of phenolic compounds. 4.3. Interfacial-Surface-Complex-Mediated Visible-Light-Sensitive TiO2 Photocatalysts 4.3. Interfacial-Surface-Complex-Mediated Visible-Light-Sensitive TiO2 Photocatalysts The interfacial surface complex (ISC)-mediated visible-light-sensitive TiO2 photocatalyst was applied to selective oxidation of several aromatic alcohols [82–88]. Unlike to the ISC in Figure 14, reactant molecules adsorbed onto the TiO2 surface (ISC) is activated under visible-light irradiation, and they are converted into products. Figure 15 shows reaction time proﬁles for the oxidation of 135 Catalysts 2019, 9, 201 benzyl alcohol in an acetonitrile solution suspended with TiO2 photocatalyst in the presence of O2 under visible light irradiation (λ > 420 nm). This reaction does not proceed without TiO2 or irradiation. It was found that the amount of benzyl alcohol decreased with an increase in the irradiation time, while the amount of benzaldehyde increased. Neither benzoic acid nor CO2 were formed as oxidative products. The yield of benzaldehyde reached circa 95%, and the carbon balance in the liquid phase was circa 95% after photo-irradiation for 4 h. $PRXQWVμPRO &DUERQEDODQFH% 7LPHK D E FG H $PRXQWVμPRO &DUERQEDODQFH% 7LPHK D E FG H $PRXQWVμPRO &DUERQEDODQFH% 7LPHK D E FG H Figure 15. Selective oxidation of benzyl alcohol on TiO2 (50 mg) under visible light irradiation [82]. The initial amount of benzyl alcohol was 50 μmol. Amounts of: benzyl alcohol (a); benzaldehyde (b); benzoic acid (c); CO2 (d); and percentage of total organic compounds in solution (e). 7LPHK 7LPHK 7LPHK Figure 15. Selective oxidation of benzyl alcohol on TiO2 (50 mg) under visible light irradiation [82]. The initial amount of benzyl alcohol was 50 μmol. Amounts of: benzyl alcohol (a); benzaldehyde (b); benzoic acid (c); CO2 (d); and percentage of total organic compounds in solution (e). Photocatalytic oxidation of benzyl alcohol and its derivatives into corresponding aldehydes was carried out with TiO2 under visible light irradiation. Benzyl alcohol and its derivatives substituted by –OCH3, –Cl, –NO2, –CH3, –CF3, and –C(CH3)3 groups were successfully converted to corresponding aldehydes with a high conversion and high selectivity on TiO2, while no other products were observed (See Table 4). 4.3. Interfacial-Surface-Complex-Mediated Visible-Light-Sensitive TiO2 Photocatalysts However, the phenolic compound (entry 9) was deeply oxidized, since it strongly adsorbed on the TiO2 surface [82]. Table 4. Chemoselective photocatalytic oxidation of different kinds of benzylic alcohols on TiO2 [82]. 7L22 9LVLEOHOLJKWK 5 5 2+ 5 5 2 7L22 9LVLEOHOLJKWK 5 5 2+ 5 5 2 Entry R1 R2 Conversion (%) Selectivity (%) 1 H H > 99 > 99 2 H C(CH3)3 > 99 > 99 3 H OCH3 > 99 > 99 4 H CH3 > 99 > 99 5 H Cl > 99 > 99 6 H NO2 > 99 > 99 7 H CF3 > 99 > 99 8 CH3 H > 99 > 99 9 H OH > 85 23 4.3.1. What Is the Origin of the Visible Light Response? The interaction of benzyl alcohol with TiO2 was analyzed by FT-IR spectroscopy (See Figure 16). Characteristic features of the ISC are as follows: (i) a remarkable downward negative band at 3715 cm−1 attributed to the O–H stretching of the terminal OH group; (ii) a new band appeared at circa 1100 cm−1, 136 Catalysts 2019, 9, 201 which is attributed to the C–O stretching of the alkoxide species formed by the interaction of benzyl alcohol with TiO2, while that of benzyl alcohol by itself is 1020 cm−1. [I] [I] [II] 7L 2 2䠉+ 7L 7L2+JURXSV FP + FP v&2 DONR[LGH 2 v&2 FDFP 2+ 7L 2 2䠉+ 7L 7L2+JURXSV FP + FP v&2 DONR[LGH 2 7L 2 2䠉+ 7L 7L2+JURXSV FP + FP v&2 DONR[LGH 2 v&2 FDFP 2+ v&2 FDFP 2+ 2+ Figure 16. FT-IR spectra [I] of benzyl alcohol by itself and benzyl alcohol adsorbed on TiO2; and [II] their peak identiﬁcation [82]. [I] Figure 16. FT-IR spectra [I] of benzyl alcohol by itself and benzyl alcohol adsorbed on TiO2; and [II] their peak identiﬁcation [82]. When the TiO2 was treated by diluted HF (aq), the IR band at 3715 cm−1 on the HF–TiO2 drastically decreased, while the photocatalytic activity signiﬁcantly decreased. The active sites were conﬁrmed to be alkoxide by the interaction of benzyl alcohol with the terminal OH groups of TiO2. TiO2 by itself exhibited absorption only in the UV region, which is attributed to the charge transition from V.B. to C.B. When the benzyl alcohol was adsorbed on TiO2, absorption in the visible region could be observed. 4.3. Interfacial-Surface-Complex-Mediated Visible-Light-Sensitive TiO2 Photocatalysts Photo-induced electron transfer from the hybridized orbital to the C.B. of TiO2 under visible light irradiation [87]. Density maps of V.B., C.B., and hybridized orbital are shown here. 4.3.2. What Makes the High Selectivity for the Photocatalytic Reactions? 4.3.2. What Makes the High Selectivity for the Photocatalytic Reactions? It was observed that benzyl alcohol is adsorbed on TiO2 more favorably than benzaldehyde in a mixture of benzyl alcohol and benzaldehyde under dark conditions. This result indicates that the interaction between benzaldehyde and TiO2 is fairly weak. According to DFT calculations [87,88], the interaction of benzyl alcohol with the TiO2 surface formed a hybridized orbital, while benzaldehyde did not form orbital mixing. Therefore, once benzaldehyde was produced by the oxidation of benzyl alcohol, benzaldehyde was immediately released into the bulk solution, and was not oxidized further to benzoic acid or CO2. 4.3. Interfacial-Surface-Complex-Mediated Visible-Light-Sensitive TiO2 Photocatalysts This absorption in the visible light region is assignable to the ISC through the LMCT (See Figure 17). The action spectra of apparent quantum yield (AQY) plots were ﬁtted with the photo-absorption of TiO2-adsorbed benzyl alcohol, suggesting that visible light absorption directly participated in the photocatalytic reactions. :DYHOHQJWKQP $EVRUEDQFH $4< D E F :DYHOHQJWKQP $EVRUEDQFH $4< D E F /0&7LQWKHDONR[LGH H 9LVLEOHOLJKW K >7L@>2@ & + 2 DONR[LGH /0&7LQWKHDONR[LGH H 9LVLEOHOLJKW K >7L@>2@ & + 2 DONR[LGH H 9LVLEOHOLJKW K >7L@>2@ & + 2 DONR[LGH Figure 17. UV-vis absorption spectra of TiO2 (a), TiO2 adsorbed with benzyl alcohol (b), and apparent quantum yield (AQY) for the formation of benzaldehyde (c); and schematic illustration of photo-induced charge transfer through LMCT in the alkoxide [82]. :DYHOHQJWKQP $EVRUEDQFH $4< D E F :DYHOHQJWKQP $EVRUEDQFH $4< D E F /0&7LQWKHDONR[LGH /0&7LQWKHDONR[LGH Figure 17. UV-vis absorption spectra of TiO2 (a), TiO2 adsorbed with benzyl alcohol (b), and apparent quantum yield (AQY) for the formation of benzaldehyde (c); and schematic illustration of photo-induced charge transfer through LMCT in the alkoxide [82]. DFT calculations [87] indicated the interaction of benzyl alcohol with surface hydroxyl groups on the TiO2 surface, resulting in the formation of alkoxide species. The electron density contour maps for 137 Catalysts 2019, 9, 201 the alkoxide species are shown in Figure 18. The orbital #212 at −0.80 eV forms the V.B. of TiO2, while #218 at +2.25 eV forms the C.B. One type of surface state consisting of the orbital (#215) originates with the alkoxide species ([Ti]–O–CH2–ph) hybridized with the O2p AOs in the V.B. of the TiO2. The energy gap between #215 and #218 (2.8 eV) was conﬁrmed to be the origin for the visible light response. Figure 18. Photo-induced electron transfer from the hybridized orbital to the C.B. of TiO2 under visible light irradiation [87]. Density maps of V.B., C.B., and hybridized orbital are shown here. Figure 18. Photo-induced electron transfer from the hybridized orbital to the C.B. of TiO2 under visible light irradiation [87]. Density maps of V.B., C.B., and hybridized orbital are shown here. Figure 18. 4.3.3. Reaction Mechanisms behind the Selective Photocatalytic Oxidation of Benzyl Alcohol The photocatalytic activities for the oxidation of benzyl alcohol or α, α-d2 benzyl alcohol were investigated. The kinetic isotope effect (KIE) [=kC-H/kCD] was estimated to be 3.9 at 295 K. This result suggests that the process for the α-deprotonation is the rate determining step (RDS) for the overall reaction. From the experimental and theoretical studies by DFT calculations, one of the favorable reaction paths is depicted in Figure 19. When benzyl alcohol interacts with Ti–OH of the TiO2, the alkoxide species (ISC) is formed on a Ti site (3). The ISC was photo-excited under visible light irradiation via LMCT of the ISC, which induces holes (h+) and electrons (e−). Subsequently, the electrons are transferred to O2 to form superoxide anions (the bonding distance between O–O becomes longer), which induces α-deprotonation of the benzyl alcohol (4-5TS). Such hydro-peroxide species would further induce the de-protonation from another benzyl alcohol to form benzaldehyde (7-8TS), resulting in regeneration of the surface terminal OH groups. The consecutive generation of the terminal OH groups would, thus, be one of the key factors for the photocatalytic reactions. 4.4. Photocatalytic Oxidation of Benzyl Amine into Imine Imines are important intermediates for the synthesis of pharmaceuticals and agricultural chemicals. Selective photocatalytic oxidation of benzyl amine into N-benzylidenebenzylamine takes place in the presence of O2 on the TiO2 at room temperature (Scheme 1) [89,90]. Several kinds of benzylic amines were examined, and they were converted into the corresponding imines, yielding circa 38–94% [89]. The origin of the visible light response is due to formation of amine oxide (ISC) through the interaction of benzylic amine onto the surface of TiO2, and the ISC exhibits electronic transition from the localized N 2p orbitals of the amine oxide (ISC) to the C.B. of TiO2. The photo-induced redox catalysis produces benzaldehyde in the presence of O2. Subsequently, the condensation reaction of benzaldehyde with another benzyl amine forms N-benzylidenebenzylamine under dark conditions. 1 1+ 7L22 YLVLEOHOLJKW EHQ]\ODPLQH 1EHQ]\OLGHQHEHQ]\ODPLQH 1 1+ 7L22 YLVLEOHOLJKW EHQ]\ODPLQH 1EHQ]\OLGHQHEHQ]\ODPLQH Scheme 1. Selective oxidation of benzyl amine into N-benzylidenebenzylamine on the TiO2 photocatalyst under visible light irradiation. 1EHQ]\OLGHQHEHQ]\ODPLQH 1EHQ]\OLGHQHEHQ]\ODPLQH EHQ]\ODPLQH EHQ]\ODPLQH Scheme 1. Selective oxidation of benzyl amine into N-benzylidenebenzylamine on the TiO2 photocatalyst under visible light irradiation. 4.3.3. Reaction Mechanisms behind the Selective Photocatalytic Oxidation of Benzyl Alcohol 138 Catalysts 2019, 9, 201 (QHUJ\N-PRO 9LVLEOHOLJKW 9LVLEOHOLJKW 7L &+ 2 2 + 7L 2 2 + 7L &+ 2 2 + 7L 2 2 + 7L 2 2 + 2 &+ 2 7L + + 7L 2 2 + + +& 2 7L 2 + + 7L 2 2 + + +& 2 7L 2+ 7L 2 2 + &+ +2 2 + 7L 2 + 3K&+2+ 2 + 7L &+ 2 2 + 7L 2 + + + DONR[LGH 76 76 76 DONR[LGH 76 76 (QHUJ\N-PRO 9LVLEOHOLJKW 9LVLEOHOLJKW 7L &+ 2 2 + 7L 2 2 + 7L &+ 2 2 + 7L 2 2 + 7L 2 2 + 2 &+ 2 7L + + 7L 2 2 + + +& 2 7L 2 + + 7L 2 2 + + +& 2 7L 2+ 7L 2 2 + &+ +2 2 + 7L 2 + 3K&+2+ 2 + 7L 2 + 3K&+2+ 2 + 7L &+ 2 2 + 7L 2 + + + DONR[LGH 7L &+ 2 2 + 7L 2 + + + DONR[LGH 76 76 76 DONR[LGH 76 76 Figure 19. Possible reaction path for the selective oxidation of benzyl alcohol in the presence of O2 on the TiO2 under visible light irradiation [87]. Figure 19. Possible reaction path for the selective oxidation of benzyl alcohol in the presence of O2 on the TiO2 under visible light irradiation [87]. 5. Conclusions This review focused on some fundamental issues behind the visible-light-sensitive TiO2 photocatalysts, highlighting the bulk and/or surface electronic structures modiﬁed by doping with nitrogen anions; plasmonic Au NPs, and interfacial surface complexes (ISC) and their related photocatalysts. Tailoring the interface and bulk properties, including surface band bending, sub-band structure, surface state distribution, and charge separation, signiﬁcantly reﬂects on the photocatalysis. We hope that this review has provided some useful contributions for the future design and development of novel photocatalytic systems employing TiO2 as well as non-TiO2 semiconductor materials with nanoscale levels. The applications of such photocatalytic systems could not only convert 139 Catalysts 2019, 9, 201 unlimited solar energy into chemical energy, but also protect our environment, leading to sustainable green chemistry. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations NPs nanoparticles ISC interfacial surface complex VOCs volatile organic compounds V.B. valence band C.B. conduction band XPS X-ray photoelectron spectroscopy EPR electron paramagnetic resonance UV-vis Ultraviolet-visible LSPR localized surface plasmon resonance PD photodeposition CPH colloid photodeposition by hole scavenger DP deposition precipitation TEM transmittance electron microscope JSC short-circuit photocurrent IPCE incident photo to current efﬁciency DFT density functional theory MLCT metal to ligand charge transfer FT-IR Fourier transformed-infrared KIE kinetic isotope effect LMCT ligand to metal charge transfer 1. Honda, K.; Fijishima, A. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 37–38. 1. Honda, K.; Fijishima, A. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37–38. h h S d h l l d f b d d 1. Honda, K.; Fijishima, A. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37–38. 2. Inoue, T.; Fujishima, A.; Konishi, S.; Honda, K. Photoelectrocatalytic reduction of carbon dioxide in aqueous 2. Inoue, T.; Fujishima, A.; Konishi, S.; Honda, K. Photoelectrocatalytic reduction of carbon dioxide in aque suspensions of semiconductor powders. Nature 1979, 277, 637–638. [CrossRef] 3. Kamat, P.V. Photochemistry on nonreactive and reactive (semiconductor) surfaces. Chem. Rev. 1993, 267–300. 4. Fox, M.A.; Dulay, M.T. Heterogeneous photocatalysis. Chem. Rev. 1993, 93, 341–357. 5. Hoffman, M.R.; Martin, S.T.; Choi, W.; Bahnemann, D.W. Environmental applications of semiconductor photocatalysis. Chem. Rev. 1995, 95, 69–96. [CrossRef] 6. Fujishima, A.; Rao, T.N.; Tryk, A. Titanium dioxide photocatalysis. J. Photochem. Photobio. C Photochem. Rev. 2000, 1, 1–21. [CrossRef] 7. Anpo, M.; Takeuchi, M. 5. Conclusions Effect of Crystallinity of TiO2 on Its Photo-catalytic Action. Chem. Phys. Lett. 1991, 187, 73–76. [CrossRef] 23. Luttrell, T.; Halpegamage, S.; Tao, J.; Kramer, A.; Sutter, E.; Batzill, M. Why is anatase a better photocatalyst than rutile? - Model studies on epitaxial TiO2 ﬁlms. Sci. Rep. 2014, 4, 4043–4050. [CrossRef] 23. Luttrell, T.; Halpegamage, S.; Tao, J.; Kramer, A.; Sutter, E.; Batzill, M. Why is anatase a better photocatalyst than rutile? - Model studies on epitaxial TiO2 ﬁlms. Sci. Rep. 2014, 4, 4043–4050. [CrossRef] 24. Gordon, T.R.; Cargnello, M.; Paik, T.; Mangolini, F.; Weber, R.T.; Fornasiero, P.; Murray, C.B. Nonaqueous Synthesis of TiO2 Nanocrystals Using TiF4 to Engineer Morphology, Oxygen Vacancy Concentration, and Photocatalytic Activity. J. Am. Chem. Soc. 2012, 134, 6751–6761. [CrossRef] y y 25. Zhang, Z.; Wang, C.-C.; Zakaria, R.J.; Ying, Y. Role of Particle Size in Nanocrystalline TiO2-Based Photocatalysts. J. Phys. Chem. B 1998, 102, 10871–10878. [CrossRef] 26. Anpo, M.; Ichihashi, Y.; Takeuchi, M.; Yamashita, H. Design of unique titanium oxide photocatalysts by an advanced metal ion-implantation method and photocatalytic reactions under visible light irradiation. Res. Chem. Intermed. 1998, 24, 143–149. [CrossRef] 27. Sato, S. Photocatalytic activity of NOx-doped TiO2 in the visible light region. Chem. Phys. Lett. 1986, 123, 126–128. [CrossRef] 28. Asahi, R.; Morikawa, T.; Aoki, K.; Taga, Y. Visible-light photocatalysis in nitrogen-doped titanium oxides. Science 2001, 293, 269–271. [CrossRef] 29. Irie, H.; Watanabe, Y.; Hashimoto, K. Nitrogen-Concentration Dependence on Photocatalytic Activity of TiO2-xNx Powders. J. Phys. Chem. B 2003, 107, 5483–5486. [CrossRef] Shin, C.; Bugli, G.; Djega-Mariadassou, G. Preparation and characterization of titanium oxynitrides with high speciﬁc surface areas J Solid State Chem 1991 95 145 155 [CrossRef] 30. Shin, C.; Bugli, G.; Djega-Mariadassou, G. Preparation and characterization of titanium oxynitrides with high speciﬁc surface areas. J. Solid State Chem. 1991, 95, 145–155. [CrossRef] 30. Shin, C.; Bugli, G.; Djega-Mariadassou, G. Preparation and characterization high speciﬁc surface areas. J. Solid State Chem. 1991, 95, 145–155. [CrossRef] 31. Livraghi, S.; Paganini, M.C.; Giamello, E.; Selloni, A.; Di Valentin, C.; Pacchioni, G. Origin of Photoactivity of Nitrogen-Doped Titanium Dioxide under Visible Light. J. Am. Chem. Soc. 2006, 128, 15666–15671. [CrossRef] 31. Livraghi, S.; Paganini, M.C.; Giamello, E.; Selloni, A.; Di Valentin, C.; Pacchioni, G. Origin of Photoactivity of Nitrogen-Doped Titanium Dioxide under Visible Light. J. Am. Chem. Soc. 2006, 128, 15666–15671. [CrossRef] 32. Wang, J.; Zhu, W.; Zhang, Y.; Liu, S. 5. Conclusions The design and development of highly reactive titanium oxide photocatalysts operating under visible light irradiation. J. Catal. 2003, 216, 505–516. [CrossRef] 8. Chen, X.; Mao, S.S. Titanium Dioxide Nanomaterials: Synthesis, Properties, Modiﬁcations, and Applicati Chem. Rev. 2007, 197, 2891–2959. [CrossRef] Shiraishi, Y.; Hirai, T. Selective organic transformations on titanium oxide-based photocatalysts. J. Photochem. 9. Shiraishi, Y.; Hirai, T. Selective organic transformations on titanium oxide-based photocatalysts. J. Photochem. Photobiol. C Photochem. Rev. 2008, 9, 157–170. [CrossRef] 10. Palmisano, G.; García-López, E.; Marcì, G.; Loddo, V.; Yurdakal, S.; Augugliaro, V.; Palmisano, L. Advances in selective conversions by heterogeneous photocatalysis. Chem. Commun. 2010, 46, 7074–7089. [CrossRef] 11. Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D.W. Understanding TiO2 Photocatalysis: Mechanisms and Materials. Chem. Rev. 2014, 114, 9919–9986. [CrossRef] 12. Ma, Y.; Wang, X.; Jia, Y.; Chen, X.; Han, H.; Li, C. Titanium dioxide-based nanomaterials for: Photocatalytic fuel generations. Chem. Rev. 2014, 114, 9987–10043. [CrossRef] 140 Catalysts 2019, 9, 201 13. Asahi, R.; Morikawa, T.; Irie, H.; Ohwaki, T. Nitrogen-doped titanium dioxide as visiblelight-sensitive photocatalyst: Designs, developments, and prospects. Chem Rev. 2014, 114, 9824–9852. [CrossRef] 14. Lang, X.; Chen, X.; Zhao, J. Heterogeneous visible light photocatalysis for selective organic transformations. Chem. Soc. Rev. 2014, 43, 473–486. [CrossRef] 15. Sang, L.; Zhao, Y.; Burda, C. TiO2 Nanoparticles as Functional Building Blocks. Chem. Rev. 2014, 114, 9283–9318. [CrossRef] 16. Nosaka, Y.; Nosaka, A.Y. Generation and Detection of Reactive Oxygen Species in Photocatalysis. Chem. Rev. 2017, 117, 11302–11336. [CrossRef] 17. Kou, J.; Lu, C.; Wang, J.; Chen, Y.; Xu, Z.; Varma, R.S. Selectivity Enhancement in Heterogeneous Photocatalytic Transformations. Chem. Rev. 2017, 117, 1445–1514. [CrossRef] 18. Prakash, J.; Sun, S.; Swart, H.C.; Gupta, R.K. Noble metals-TiO2 nanocomposites: From fundamental mechanisms to photocatalysis, surface enhanced Raman scattering and antibacterial applications. Appl. Mater. Today 2018, 11, 82–135. [CrossRef] y 19. Wang, W.; Tadé, M.O.; Shao, Z. Nitrogen-doped simple and complex oxides for photocatalysis: A review. Prog. Mater. Sci. 2018, 92, 33–63. [CrossRef] 20. Yamashita, H.; Mori, K.; Kuwahara, Y.; Kamegawa, T.; Wen, M.; Verma, P.; Che, M. Single-site and nano-conﬁned photocatalysts designed in porous materials for environmental uses and solar fuels. Chem. Soc. Rev. 2018, 47, 8072–8096. [CrossRef] 21. Ahmed, A.Y.; Kandiel, T.A.; Oekermann, T.; Bahnemann, D. Photocatalytic Activities of Different Well-deﬁned Single Crystal TiO2Surfaces: Anatase versus Rutile. J. Phys. Chem. Lett. 2011, 2, 2461–2465. [CrossRef] 22. Tanaka, K.; Capule, M.F.V.; Hisanaga, T. 5. Conclusions An efﬁcient two-step technique for nitrogen-doped titanium dioxide synthesizing: Visible-light-induced photodecomposition of methylene blue. J. Phys. Chem. C 2007, 111, 1010–1014. [CrossRef] 33. Higashimoto, S.; Azuma, M. Photo-induced charging effect and electron transfer to the redox species on nitrogen-doped TiO2 under visible light irradiation. Appl. Catal. B Environ. 2009, 89, 557–562. [CrossRef] 34. Wang, H.; Hu, Y. The Photocatalytic Property of Nitrogen-Doped TiO2 Nanoball Film. Int. J. Photoenergy 33. Higashimoto, S.; Azuma, M. Photo-induced charging effect and electron transfer to the redox species on nitrogen-doped TiO2 under visible light irradiation. Appl. Catal. B Environ. 2009, 89, 557–562. [CrossRef] nitrogen-doped TiO2 under visible light irradiation. Appl. Catal. B Environ. 2009, 89, 557–562. [CrossRef] 34. Wang, H.; Hu, Y. The Photocatalytic Property of Nitrogen-Doped TiO2 Nanoball Film. Int. J. Photoenergy 2013. [CrossRef] 34. Wang, H.; Hu, Y. The Photocatalytic Property of Nitrogen-Doped TiO2 Nanoball Film. Int. J. Photoenergy 2013. [CrossRef] 35. Diwald, O.; Thompson, T.L.; Zubkov, T.; Goralski, E.G.; Walck, S.D.; Yates, J.T., Jr. Photochemical Activity of Nitrogen-Doped Rutile TiO2(110) in Visible Light. J. Phys. Chem. B 2004, 108, 6004–6008. [CrossRef] 141 Catalysts 2019, 9, 201 36. Higashimoto, S.; Ushiroda, Y.; Azuma, M.; Ohue, H. Synthesis, characterization and photocatalytic activity of N-doped TiO2 modiﬁed by platinum chloride. Catal. Today 2008, 132, 165–169. [CrossRef] 37. Higashimoto, S.; Ushiroda, Y.; Azuma, M. Mechanism for enhancement of visible light response on nitrogen-doped TiO2 by modiﬁcation with vanadium species. J. Nanosci. Nanotechnol. 2010, 10, 246–251. [CrossRef] 38. Nakamura, R.; Tanaka, T.; Nakato, Y. Mechanism for Visible Light Responses in Anodic Photocurrents at N-Doped TiO2 Film Electrodes. J. Phys. Chem. B 2004, 108, 10617–10620. [CrossRef] 39. Tang, J.; Cowan, A.J.; Durrant, J.R.; Klug, D.R. Mechanism of O2 production from water splitting: Nature of charge carriers in nitrogen doped nanocrystalline TiO2 ﬁlms and factors limiting O2 production. J. Phys. Chem. C 2011, 115, 3143–3150. [CrossRef] 40. Higashimoto, S.; Tanihata, W.; Nakagawa, Y.; Azuma, M.; Ohue, H.; Sakata, Y. Effective photocatalytic decomposition of VOC under visible-light irradiation on N-doped TiO2 modiﬁed by vanadium species. Appl. Catal. A Gen. 2008, 340, 98–104. [CrossRef] 41. Morikawa, T.; Ohwaki, T.; Suzuki, K.; Moribe, S.; Tero-Kubota, S. Visible-light-induced photocatalytic oxidation of carboxylic acids and aldehydes over N-doped TiO2 loaded with Fe, Cu or Pt. Appl. Catal. B Environ. 2008, 83, 56–62. [CrossRef] 42. Sreethawong, T.; Laehsalee, S.; Chavadej, S. Use of Pt/N-doped mesoporous-assembled nanocrystalline TiO2 for photocatalytic H2 production under visible light irradiation. 5. Conclusions Catal. Commun. 2009, 10, 538–543. [CrossRef] 43. Dolat, D.; Quici, N.; Kusiak-Nejman, E.; Morawski, A.W.; Puma, G.L. One-step, hydrothermal synthesis of nitrogen, carbon co-doped titanium dioxide (N,C-TiO2) photocatalysts. Effect of alcohol degree and chain length as carbon dopant precursors on photocatalytic activity and catalyst deactivation. Appl. Catal. B Environ. 2012, 115, 81–89. [CrossRef] 44. Virkutyte, J.; Varma, R.S. Visible light activity of Ag-loaded and guanidine nitrate-doped nano-TiO2: Degradation of dichlorophenol and antibacterial properties. RSC Adv. 2012, 2, 1533–1539. [CrossRef] 45. Cong, Y.; Zhang, J.; Chen, F.; Anpo, M. Synthesis and Characterization of Nitrogen-Doped TiO2Nanophotocatalyst with High Visible Light Activity. J. Phys. Chem. C 2007, 111, 6976–6982. [CrossRef] 46. Yang, X.; Cao, C.; Erickson, L.; Hohn, K.; Maghirang, R.; Klabunde, K. Synthesis of visible-light-ac TiO2-based photocatalysts by carbon and nitrogen doping. J. Catal. 2008, 260, 128–133. [CrossRef] 47. Mitoraj, D.; Kisch, H. On the Mechanism of Urea–Induced Titania Modiﬁcation. Chem. Eur. J. 2010, 261–269. [CrossRef] 48. Chai, B.; Peng, T.; Mao, J.; Li, K.; Zan, L. Graphitic carbon nitride (g-C3N4)-Pt-TiO2 nanocomposite as an efﬁcient photocatalyst for hydrogen production under visible light irradiation. Phys. Chem. Chem. Phys. 2012, 14, 16745–16752. [CrossRef] 49. Han, C.; Wang, Y.; Lei, Y.; Wang, B.; Wu, N.; Shi, Q.; Li, Q. In situ synthesis of graphitic-C3N4 nanosheet hybridized N-doped TiO2 nanoﬁbers for efﬁcient photocatalytic H2 production and degradation. Nano Res. 2015, 8, 1199–1209. [CrossRef] 50. Yan, H.; Yang, H. TiO2-g-C3N4 composite materials for photocatalytic H2 evolution under visible light irradiation. J. Alloy. Comp. 2010, 509, L26–L29. [CrossRef] 51. Higashimoto, S.; Hikita, K.; Azuma, M.; Yamamoto, M.; Takahashi, M.; Sakata, Y.; Matsuoka, M.; Kobayashi, H. Visible Light-Induced Photocatalysis on Carbon Nitride Deposited Titanium Dioxide: Hydrogen Production from Sacriﬁcial Aqueous Solutions. Chin. J. Chem. 2017, 35, 165–172. [CrossRef] 52. Zhang, Q.; Gangadharan, D.T.; Liu, Y.; Xu, Z.; Chaker, M.; Ma, D. Recent advancements in plasmon-enhanced visible light-driven water splitting. J. Materiomics 2017, 3, 33–50. [CrossRef] 53. Ohko, Y.; Tatsuma, T.; Fujii, T.; Naoi, K.; Niwa, C.; Kubota, Y.; Fujishima, A. Multicolour photochromism of TiO2 ﬁlms loaded with silver nanoparticles. Nat. Mater. 2003, 2, 29–31. [CrossRef] [PubMed] 54. Tian, Y.; Tatsuma, T. Plasmon-induced photoelectrochemistry at metal nanoparticles supported on nanoporous TiO2. Chem. Commun. 2004, 0, 1810–1811. [CrossRef] [PubMed] 55. Tian, Y.; Tatsuma, T. Mechanisms and Applications of Plasmon-Induced Charge Separation at TiO2 Films Loaded with Gold Nanoparticles. J. Am. Chem. Soc. 2005, 127, 7632–7637. [CrossRef] [PubMed] 56. 5. Conclusions Kowalska, E.; Abe, R.; Ohtani, B. Visible light-induced photocatalytic reaction of gold-modiﬁed titanium(IV) oxide particles: Action spectrum analysis. Chem. Commun. 2009, 0, 241–243. [CrossRef] [PubMed] 142 Catalysts 2019, 9, 201 57. Kowalska, E.; Mahaney, O.O.P.; Abe, R.; Ohtani, B. Visible-light-induced photocatalysis through surface plasmon excitation of gold on titania surfaces. Phys. Chem. Chem. Phys. 2010, 12, 2344–2355. [CrossRef] [PubMed] 58. Kolinko, P.A.; Selishchev, D.S.; Kozlov, D.V. Visible Light Photocatalytic Oxidation of Ethanol Vapor on Titanium Dioxide Modiﬁed with Noble Metals. Theor. Exp. Chem. 2015, 51, 96–103. [CrossRef] 59. Frens, G. Controlled Nucleation for the Regulation of the Particle Size in Monodisperse Gold Suspensions. Nat. Phys. Sci. 1973, 241, 20–22. [CrossRef] 60. Tanaka, A.; Ogino, A.; Iwaki, M.; Hashimoto, K.; Ohnuma, A.; Amano, F.; Ohtani, B.; Kominami, H. Gold–Titanium(IV) Oxide Plasmonic Photocatalysts Prepared by a Colloid-Photodeposition Method: Correlation Between Physical Properties and Photocatalytic Activities. Langmuir 2012, 28, 13105–13111. [CrossRef] 61. Silva, C.G.; Juarez, R.; Marino, T.; Molinari, R.; Garcia, H. Inﬂuence of Excitation Wavelength (UV or Visible Light) on the Photocatalytic Activity of Titania Containing Gold Nanoparticles for the Generation of Hydrogen or Oxygen from Water. J. Am. Chem. Soc. 2011, 133, 595–602. [CrossRef] [PubMed] 62. Zanella, R.; Delannoy, L.; Louis, C. Mechanism of deposition of gold precursors onto TiO2 during the preparation by cation adsorption and deposition–precipitation with NaOH and urea. Appl. Catal. A Gen. 2005, 291, 62–72. [CrossRef] 63. Tanaka, A.; Sakaguchi, S.; Hashimoto, K.; Kominami, H. Preparation of Au/TiO2 with Metal Cocatalysts Exhibiting Strong Surface Plasmon Resonance Effective for Photoinduced Hydrogen Formation under Irradiation of Visible Light. ACS Catal. 2013, 3, 79–85. [CrossRef] 64. Tanaka, A.; Nakanishi, K.; Hamada, R.; Hashimoto, K.; Kominami, H. Simultaneous and Stoichiometric Water Oxidation and Cr(VI) Reduction in Aqueous Suspensions of Functionalized Plasmonic Photocatalyst Au/TiO2–Pt under Irradiation of Green Light. ACS Catal. 2013, 3, 1886–1891. [CrossRef] 65. Tanaka, A.; Nishino, Y.; Sakaguchi, S.; Yoshikawa, T.; Imamura, K.; Hashimoto, K.; Kominami, H. Functionalization of a plasmonic Au/TiO2 photocatalyst with an Ag co-catalyst for quantitative reduction of nitrobenzene to aniline in 2-propanol suspensions under irradiation of visible light. Chem. Commun. 2013, 49, 2551–2553. [CrossRef] [PubMed] 66. Tanaka, A.; Hashimoto, K.; Kominami, H. A very simple method for the preparation of Au/TiO2 plasmonic photocatalysts working under irradiation of visible light in the range of 600–700 nm. Chem. Commun. 2017, 53, 4759–4762. [CrossRef] 67. Naya, S.; Teranishi, M.; Isobe, T.; Tada, H. 5. Conclusions Light wavelength-switchable photocatalytic reaction by gold nanoparticle-loaded titanium(IV) dioxide. Chem. Commun. 2010, 46, 815–817. [CrossRef] 68. Naya, S.; Kimura, K.; Tada, H. One-Step Selective Aerobic Oxidation of Amines to Imines by Gold Nanoparticle-Loaded Rutile Titanium(IV) Oxide Plasmon Photocatalyst. ACS Catal. 2013, 3, 10–13. [CrossRef] 69. Tsukamoto, D.; Shiraishi, Y.; Sugano, Y.; Ichikawa, S.; Tanaka, S.; Hirai, T. Gold Nanoparticles Located at the Interface of Anatase/Rutile TiO2 Particles as Active Plasmonic Photocatalysts for Aerobic Oxidation. J. Am. Chem. Soc. 2012, 134, 6309–6315. [CrossRef] 70. Zheng, Z.; Huang, B.; Qin, X.; Zhang, X.; Dai, Y.; Whangbo, M.-H. Facile in situ synthesis of visible-light plasmonic photocatalysts M@TiO2 (M = Au, Pt, Ag) and evaluation of their photocatalytic oxidation of benzene to phenol. J. Mater. Chem. 2011, 21, 9079–9087. [CrossRef] 71. Tanaka, A.; Sakaguchi, S.; Hashimoto, K.; Kominami, H. Preparation of Au/TiO2 exhibiting strong surface plasmon resonance effective for photoinduced hydrogen formation from organic and inorganic compounds under irradiation of visible light. Catal. Sci. Technol. 2012, 2, 907–909. [CrossRef] 72. Tanaka, A.; Teramura, K.; Hosokawa, S.; Kominami, H.; Tanaka, T. Visible light-induced water splitting in an aqueous suspension of a plasmonic Au/TiO2 photocatalyst with metal co-catalysts. Chem. Sci. 2017, 8, 2574–2580. [CrossRef] [PubMed] 73. Furube, A.; Du, L.; Hara, K.; Katoh, R.; Tachiya, M. Ultrafast Plasmon-Induced Electron Transfer from Gold Nanodots into TiO2 Nanoparticles. J. Am. Chem. Soc. 2007, 129, 14852–14853. [CrossRef] [PubMed] 74. Borgarello, E.; Kiwi, J.; Pelizzetti, E.; Visca, M.; Grätzel, M. Photochemical cleavage of water by photocatalysis. Nature 1981, 289, 158–160. [CrossRef] 75. Vinodgopal, K.; Hua, X.; Dahlgren, R.L. Photochemistry of Ru(bpy)2(dcbpy)2+ on A12O3 and TiO2 surfaces. an insight into the mechanism of photosensitization. J. Phys. Chem. 1995, 99, 10883–10889. [CrossRef] 143 Catalysts 2019, 9, 201 76. Sakthivel, S.; Kisch, H. Daylight photocatalysis by carbon-modiﬁed titanium dioxide. Angew. Chem. Int. Ed. 2003, 42, 4908–4911. [CrossRef] [PubMed] 77. Macyk, W.; Burgeth, G.; Kisch, H. Photoelectrochemical properties of platinum(IV) chloride surface modiﬁed TiO2. Photochem. Photobiol. Sci. 2003, 2, 322–328. [CrossRef] 77. Macyk, W.; Burgeth, G.; Kisch, H. Photoelectrochemical properties o TiO2. Photochem. Photobiol. Sci. 2003, 2, 322–328. [CrossRef] 78. Higashimoto, S.; Nishi, T.; Yasukawa, M.; Azuma, M.; Sakata, Y.; Kobayashi, H. Photocatalysis of titanium dioxide modiﬁed by interfacial surface complexes (ISC) with different substituted groups. J. Catal. 2015, 329, 286–290. [CrossRef] 79. Ikeda, S.; Abe, C.; Torimoto, T.; Ohtani, B. 5. Conclusions Photochemical hydrogen evolution from aqueous triethanolamine solutions sensitized by binaphthol-modiﬁed titanium(IV) oxide under visible-light irradiation. J. Photochem. Photobiol. A Chem. 2003, 160, 61–67. [CrossRef] 80. Kamegawa, T.; Seto, H.; Matsuura, S.; Yamashita, H. Preparation of hydroxynaphthalene modiﬁed TiO2 via formation of surface complexes and their applications in the photocatalytic reduction of nitrobenzene under visible-light irradiation. ACS Appl. Mater. Interfaces 2012, 4, 6635–6639. [CrossRef] t irradiation. ACS Appl. Mater. Interfaces 2012, 4, 6635–6 ACS Appl. Mater. Interfaces 2012, 4, 6635–6639. [CrossR 81. Kim, S.; Choi, W. Visible-light-induced photocatalytic degradation of 4-chlorophenol and phenolic compounds in aqueous suspension of pure titania: Demonstrating the existence of a surface-complex-mediated path. J. Phys. Chem. B 2005, 109, 5143–5149. [CrossRef] [PubMed] 82. Higashimoto, S.; Kitao, N.; Yoshida, N.; Sakura, T.; Azuma, M.; Ohue, H.; Sakata, Y. Selective photocatalytic oxidation of benzyl alcohol and its derivatives into corresponding aldehydes by molecular oxygen on titanium dioxide under visible light irradiation. J. Catal. 2009, 266, 279–285. [CrossRef] 83. Higashimoto, S.; Okada, K.; Morisugi, T.; Azuma, M.; Ohue, H.; Kim, T.-H.; Matsuoka, M.; Anpo, M. Effect of surface treatment on the selective photocatalytic oxidation of benzyl alcohol infrared study of hydroxy groups on coordinative defect sites. Top. Catal. 2010, 53, 578–583. [CrossRef] 84. Higashimoto, S.; Okada, K.; Azuma, M.; Ohue, H.; Terai, T.; Sakata, Y. Characteristics of the charge transfer surface complex on titanium(IV) dioxide for the visible light induced chemoselective oxidation of benzyl alcohol. RSC Adv. 2012, 2, 669–676. [CrossRef] 85. Higashimoto, S.; Suetsugu, N.; Azuma, M.; Ohue, H.; Sakata, Y. Efﬁcient and selective oxidation of benzylic alcohol by O2 into corresponding aldehydes on a TiO2 photocatalyst under visible light irradiation: Effect of phenyl-ring substitution on the photocatalytic activity. J. Catal. 2010, 274, 76–83. [CrossRef] 86. Higashimoto, S.; Shirai, R.; Osano, Y.; Azuma, M.; Ohue, H.; Sakata, Y.; Kobayashi, H. Inﬂuence of metal ions on the photocatalytic activity: Selective oxidation of benzyl alcohol on iron (III) ion-modiﬁed TiO2, using visible light. J. Catal. 2014, 311, 137–143. [CrossRef] 87. Kobayashi, H.; Higashimoto, S. DFT study on the reaction mechanisms behind the catalytic oxidation of benzyl alcohol into benzaldehyde by O2 over anatase TiO2 surfaces with hydroxyl groups: Role of visible-light irradiation. Appl. Catal. B Environ. 2015, 170, 135–143. [CrossRef] 88. Li, R.; Kobayashi, H.; Guo, J.; Fan, J. Visible-light induced high-yielding benzyl alcohol-to benzaldehyde transformation over mesoporous crystalline TiO2: A self-adjustable photooxidation system with controllable hole-generation. J. Phys. Chem. Received: 17 January 2019; Accepted: 10 February 2019; Published: 19 February 2019 Received: 17 January 2019; Accepted: 10 February 2019; Published: 19 February 2019 Abstract: Titanium dioxide (TiO2) nanomaterials have garnered extensive scientiﬁc interest since 1972 and have been widely used in many areas, such as sustainable energy generation and the removal of environmental pollutants. Although TiO2 possesses the desired performance in utilizing ultraviolet light, its overall solar activity is still very limited because of a wide bandgap (3.0–3.2 eV) that cannot make use of visible light or light of longer wavelength. This phenomenon is a deﬁciency for TiO2 with respect to its potential application in visible light photocatalysis and photoelectrochemical devices, as well as photovoltaics and sensors. The high overpotential, sluggish migration, and rapid recombination of photogenerated electron/hole pairs are crucial factors that restrict further application of TiO2. Recently, a broad range of research efforts has been devoted to enhancing the optical and electrical properties of TiO2, resulting in improved photocatalytic activity. This review mainly outlines state-of-the-art modiﬁcation strategies in optimizing the photocatalytic performance of TiO2, including the introduction of intrinsic defects and foreign species into the TiO2 lattice, morphology and crystal facet control, and the development of unique mesocrystal structures. The band structures, electronic properties, and chemical features of the modiﬁed TiO2 nanomaterials are clariﬁed in detail along with details regarding their photocatalytic performance and various applications. Keywords: TiO2; energy band engineering; morphology modiﬁcation; mesocrystals; application Catalysts 2019, 9, 191; doi:10.3390/catal9020191 Review Review Titanium Dioxide: From Engineering to Applications Xiaolan Kang †, Sihang Liu †, Zideng Dai, Yunping He, Xuezhi Song and Zhenquan Tan * School of Petroleum and Chemical Engineering, Dalian University of Technology, No. 2 Dagong Road, New District of Liaodong Bay, Panjin, Liaoning 124221, China; kxl@mail.dlut.edu.cn (X.K.); wdlsd@mail.dlut.edu.cn (S.L.); xiaodai@mail.dlut.edu.cn (Z.D.); yphe_04@mail.dlut.edu.cn (Y.H.); songxz@dlut.edu.cn (X.S.) www.mdpi.com/journal/catalysts 5. Conclusions C 2011, 115, 23408–23416. [CrossRef] 89. Lang, X.; Ma, W.; Zhao, Y.; Chen, C.; Ji, H.; Zhao, J. Visible-light-induced selective catalytic aerobic oxidation of amines into imines on TiO2. Chem. Eur. J. 2012, 18, 2624–2631. [CrossRef] 89. Lang, X.; Ma, W.; Zhao, Y.; Chen, C.; Ji, H.; Zhao, J. Visible-light-induced selec of amines into imines on TiO2. Chem. Eur. J. 2012, 18, 2624–2631. [CrossRef] 90. Higashimoto, S.; Hatada, Y.; Ishikawa, R.; Azuma, M.; Sakata, Y.; Kobayashi, H. Selective Photocatalytic Oxidation of Benzyl Amine by O2 into N-Benzylidenebenzylamine on TiO2 Using Visible Light. Curr. Org. Chem. 2013, 17, 2374–2381. [CrossRef] © 2019 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 144 Titanium Dioxide: From Engineering to Applications School of Petroleum and Chemical Engineering, Dalian University of Technology, No. 2 Dagong Road, New District of Liaodong Bay, Panjin, Liaoning 124221, China; kxl@mail.dlut.edu.cn (X.K.); wdlsd@mail.dlut.edu.cn (S.L.); xiaodai@mail.dlut.edu.cn (Z.D.); yphe_04@mail.dlut.edu.cn (Y.H.); songxz@dlut.edu.cn (X.S.) * Correspondence: tanzq@dlut.edu.cn; Tel.: +86-427-263-1808 † These authors contributed equally to this work. † These authors contributed equally to this work. 1. Introduction Over the past several decades, the increasing severe energy shortages and environmental pollution have caused great concern worldwide. To achieve sustainable development of society, there is an urgent need to explore environmentally friendly technologies applicable to pollutant recovery and clean energy supplies. In the long-term, solar energy is an inexhaustible source of renewable energy; therefore, developing technologies and materials to enhance solar energy utilization is central to both energy security and environmental stewardship. In 1972, Fujishima and Honda ﬁrst published a study for producing hydrogen on titanium dioxide (TiO2) photoelectrodes under ultraviolet light illumination, which garnered worldwide attention [1,2]. From then on, semiconductor photocatalysis has been considered one of the most promising pathways to address both hydrogen production and pollution abatement. Photocatalysis can be widely used anywhere in the world, providing natural solar light or artiﬁcial indoor illumination is available [3]. Semiconductor materials are often used as photocatalysts [4]. According to band energy theory, the discontinuous band structure of semiconductors is composed of low energy valence bands ﬁlled with electrons, high-energy conduction bands, and band gaps. When the energy of the incident photons equals or exceeds the bandgap, the photoexcitation of electron–hole pairs and the consequential Catalysts 2019, 9, 191; doi:10.3390/catal9020191 www.mdpi.com/journal/catalysts 145 Catalysts 2019, 9, 191 photocatalytic redox reaction take place [5]. The photocatalytic process mainly involves the steps of generation, separation, recombination, and surface capture of photogenerated electrons and hole pairs. Photochemical reactions occur on the surface of a solid catalyst, which includes two half-reaction oxidation reactions of photogenerated holes and reduction reactions of photogenerated electrons [6]. The speciﬁc process that occurs in semiconductors is described in Figure 1. During this process, a large proportion of charge carriers (e−/h+ pairs) recombine quickly at the surface and interior of the bulk material, leading to the dissipation of absorbed energy in the form of light (photon generation) or heat (lattice vibration). Therefore, these charge carriers cannot participate in the subsequent photocatalytic reactions, which is detrimental to the whole process [7]. Figure 1. Photocatalytic process in semiconductor. Figure 1. Photocatalytic process in semiconductor. The electrons and holes that successfully migrate to the surface of the semiconductor without recombining can be involved in the reduction and oxidation reactions, respectively, which are the bases for photodegradation of organic pollutants and photocatalytic water splitting to produce H2 [8]. As excellent oxidizers, the photogenerated holes can mineralize organic pollutants directly. 1. Introduction In addition, the holes can also form hydroxyl radicals (•OH) with strong oxidizing properties. Photoexcited electrons, on the other hand, can produce superoxide radicals (O2•−) and •OH. These free radicals and e−/h+ pairs are highly reactive and can induce a series of redox reactions. In addition, with respect to water splitting, photogenerated electrons can be captured by H+ in water to generate hydrogen, while holes will oxidize H2O to form O2 [9–11]. In general, to increase the activity of photocatalysts and utilize visible light more effectively, several requirements need to be satisﬁed. First, the light absorption process determines the amount of excited charges, which means that more charge carriers are likely to be accumulated on the surface if more light can be absorbed by the photocatalyst. Additionally, considering that ultraviolet (UV) light occupies less than 4% of sunlight’s emission spectrum, while visible light accounts for approximately 40%, a smaller bandgap is necessary for a semiconductor to absorb solar energy across a broad range of spectra. Therefore, improving the optical absorption properties has become a common purpose for photocatalyst design to enhance their overall activity [12]. In addition, the position of conduction bands (CBs) and valence bands (VBs) is critical, which are responsible for the production of active species, such as •OH, HO2•, H2O2, and O2•−. Furthermore, the photogenerated electrons and holes should be transported and separated efﬁciently in the photocatalyst because the fast recombination of charge carriers will otherwise result in low reactivity. Finally, the as-prepared photocatalytic materials and their modiﬁcation processes should be environmentally friendly and economical [13]. 146 Catalysts 2019, 9, 191 Since 1972, TiO2 has been intensively investigated due to its thermal and chemical stability, superhydrophilicity, low toxicity, and natural geologic abundance. Compared with other semiconductor materials, TiO2 is of ubiquitous interest across many research fields and for many applications [14], such as photodegradation of pollutants and hazardous materials, photolysis (splitting) of water to yield H2, artificial photosynthesis, etc. Nevertheless, the poor visible light absorption and fast electron–hole recombination, as well as the sluggish transfer kinetics of the charge carriers to the surrounding media, considerably limit the photocatalytic activities of TiO2. Hence, during the past few decades, much effort has been devoted to overcoming these problems by, for example, reducing e−/h+ pair recombination and improving the optical absorption properties by energy band regulation, morphology control, and the construction of heterogeneous junctions [15]. 1. Introduction g j In this review, we mainly focus on the regulation of the electronic structure and modiﬁcation of the micromorphology of TiO2 nanomaterials to achieve property enhancements that could be applicable to a variety of potential applications. 2. Energy Band Engineering of TiO2 The absorption of incident light and redox potential of TiO2 mainly depend on its energy band conﬁguration [16]. To utilize solar energy more effectively, it is necessary to explore and develop longwave-light-sensitive TiO2 photocatalysts with excellent performance on the basis of energy band engineering [17]. A better understanding of the electronic structure of TiO2 is important for band gap modiﬁcation. The molecular orbital bonding energy diagram in Figure 2 clearly shows the fundamental features of anatase TiO2 [18]. The chemical bonding of anatase TiO2 can be deconstructed into Ti, e.g., Ti t2g (dyz, dxz, and dxy), O pσ (in the Ti3O cluster plane), and O pπ (out of the Ti3O cluster plane). The upper valence bands include three main regions: the σ bonding, which is located at the bottom, is the most stable bond type, and arises from the hybridization of Ti, e.g., O pσ; the hybridization of the O pπ and Ti dyz (or dxz) orbitals constitutes the middle energy region of π bonding; and the higher energy region in the top of the valence bands, which is dominated by the O pπ orbitals. The conduction band is composed of Ti 3d and 4s, and the bottom of the conduction bands is composed of the isolated Ti dxy orbitals [19,20]. For the purpose of narrowing the bandgap of TiO2, three basic approaches of adjusting the VBs or CBs or the continuous modiﬁcation of the VBs and CBs of the anatase are shown in Figure 3. Figure 2. (a) Total and projected densities of states (DOS) of the anatase TiO2 structure and (b) molecular orbital bonding structure for anatase TiO2 [18]. Copyright 2004 The American Physical Society. Figure 2. (a) Total and projected densities of states (DOS) of the anatase TiO2 structure and (b) molecular orbital bonding structure for anatase TiO2 [18]. Copyright 2004 The American Physical Society. 147 Catalysts 2019, 9, 191 Figure 3. Three schemes of the band gap modiﬁcations of TiO2 match the solar spectrum: (a) a higher shift in valence band maximum (VBM); (b) a lower shift in conduction band minimum (CBM); and (c) continuous modiﬁcation of both VBM and CBM. Figure 3. Three schemes of the band gap modiﬁcations of TiO2 match the solar spectrum: (a) a higher shift in valence band maximum (VBM); (b) a lower shift in conduction band minimum (CBM); and (c) continuous modiﬁcation of both VBM and CBM. 2.1. Doping of TiO2 Doping TiO2 with certain earth rare metal ions represents another promising method to prolong the recombination time of charge carriers and improve their separation efﬁciency. The 4f electrons in most rare earth elements can give rise to the formation of a multielectron conﬁguration, which acts as a shallow trap for photogenerated electrons and holes [33]. Furthermore, the use of rare earth metal ion dopants in TiO2 tends to facilitate the utilization of solar light from ultraviolet to infrared light regions. Li et al. prepared a series of Ce-doped TiO2 nanoparticles by the sol–gel method. The characterization results showed that Ce ions entered the TiO2 matrix at Ti sites, leading to the formation of impurity states, as shown in Figure 5. In addition, enhanced separation of the photogenerated charge carriers was also realized due to the coexistence of Ce3+ and Ce4+ dopant ions [34]. doping V, Cr, Mn, and Fe based on a large number of experimental results, and they suggested that cation vacancies may lead to these impurity states [31]. An early work by Borgarello et al. in 1982 reported that Cr3+-doped TiO2 nanoparticles (investigated for properties of photocatalytic hydrogen evolution) exhibit excellent absorption of visible light in the range of 400 to 550 nm. They believed that the 3d electrons of Cr3+ were excited into the conduction band of TiO2, thus inducing a visible light response [32]. Doping TiO2 with certain earth rare metal ions represents another promising method to prolong the recombination time of charge carriers and improve their separation efﬁciency. The 4f electrons in most rare earth elements can give rise to the formation of a multielectron conﬁguration, which acts as a shallow trap for photogenerated electrons and holes [33]. Furthermore, the use of rare earth metal ion dopants in TiO2 tends to facilitate the utilization of solar light from ultraviolet to infrared light regions. Li et al. prepared a series of Ce-doped TiO2 nanoparticles by the sol–gel method. The characterization results showed that Ce ions entered the TiO2 matrix at Ti sites, leading to the formation of impurity states, as shown in Figure 5. In addition, enhanced separation of the photogenerated charge carriers was also realized due to the coexistence of Ce3+ and Ce4+ dopant ions [34]. Figure 5. Band energy structure and charge transfer [34]. Copyright 2017 American Chemical Society. Figure 5. Band energy structure and charge transfer [34]. Copyright 2017 American Chemical Society. 2.1. Doping of TiO2 To extend the visible light response of TiO2 and improve its photocatalytic activities, various modiﬁcation strategies, such as dye sensitization, impurity or intrinsic doping or semiconductor coupling, have been developed [21–23]. Among them, introducing impurity ions into the TiO2 crystal lattice to substitute the host anions and/or cations has earned much attention in the past decade. By means of physical or chemical methods, researchers have been able to introduce a variety of ions into the TiO2 matrix, where they change the band structure of TiO2 by inducing impurity states within the bandgap [2], as shown in Figure 4. In general, ion doping contributes to the improved activities of TiO2 in three ways: (1) by narrowing the bandgap and promoting the adsorption of the main region of the solar spectrum, such as doping with N, S, C, B, etc. [24,25]; (2) by improving the conductivity of TiO2 and the mobility of charge carriers, the increased charge traps can reduce bulk recombination and separate photogenerated electrons and holes more efﬁciently (e.g., Zn, Fe, and Y) [26]; and (3) by altering the conduction band position of TiO2 with certain metal ion dopants, such as Zr4+, Nb5+, and W6+, which further affects the carrier transfer properties [27]. Figure 4. TiO2 nanoparticles with different doping elements [2]. Copyright 2014 American Chemical Society. Figure 4. TiO2 nanoparticles with different doping elements [2]. Copyright 2014 American Chemical Society. TiO2 doping can be doped with a variety of metal ions, including transition metal and rare earth metal ions. For transition metal dopants, such as Fe, Mn, V, Cu, and Cr, both delocalized and localized impurity states will be created within the band gap of TiO2 along the crystal ﬁeld splitting of metal 3d orbitals [28–30]. Mizushima et al. determined impurity levels of 1.9 to 3.0 eV below CBM by 148 Catalysts 2019, 9, 191 doping V, Cr, Mn, and Fe based on a large number of experimental results, and they suggested that cation vacancies may lead to these impurity states [31]. An early work by Borgarello et al. in 1982 reported that Cr3+-doped TiO2 nanoparticles (investigated for properties of photocatalytic hydrogen evolution) exhibit excellent absorption of visible light in the range of 400 to 550 nm. They believed that the 3d electrons of Cr3+ were excited into the conduction band of TiO2, thus inducing a visible light response [32]. Catalysts 2019, 9, 191 Catalysts 2019, 9, 191 Recently, the non-metal doping of nitrogen (N), sulfur (S), carbon (C), ﬂuorine (F), iodine (I), and phosphorus (P) has been extensively studied due to their relatively high photostability and photoelectric properties [39]. However, in comparison to metal-doped TiO2, the role of the non-metal dopants as recombination centers of charge carriers might be minimized. By replacing the oxygen atoms in the TiO2 lattice, the non-metal elements can signiﬁcantly narrow the bandgap and thereby improve the visible light response of TiO2. In addition, impurity states can be formed near the valence band edge alone with non-metal doping, as displayed in Figure 6. Instead of acting as recombination centers, these occupied levels can be regarded as shallow traps that effectively separate photogenerated electron–hole pairs [40]. Figure 6. Comparison of atomic p levels among anions. The band gap of TiO2 is formed between the O 2pπ and Ti 3d states [39]. Copyright 2014 American Chemical Society. Figure 6. Comparison of atomic p levels among anions. The band gap of TiO2 is formed between the O 2pπ and Ti 3d states [39]. Copyright 2014 American Chemical Society. In 2001, Asahi et al. ﬁrst published research on N-doped TiO2 nanomaterials, which initiated a wave of studies related to non-metal-doped photocatalysts [41]. In a similar work, Zhao et al. reported highly active N-doped TiO2 nanotubes for CO2 reduction. Despite the tubular structure with a large surface area providing more surface active sites, the N dopants contributed more to the improved photocatalytic activity. It was found that a redshift of the light absorption and a color center were achieved with N-doped TiO2 nanotubes because N atoms can substitute for the lattice O atoms of TiO2, thereby reducing its bandgap and resulting in a ~4 times higher visible light photocatalytic CO2 reduction activity in comparison to pure TiO2 nanotubes [42]. Irie et al. prepared C-doped TiO2 nanoparticles by oxidizing TiC powder, and the efﬁciency of decomposing gaseous isopropanol under visible light was signiﬁcantly improved [43]. S-doped anatase TiO2 with a high surface area was obtained by Li et al. They treated pure TiO2 using a supercritical strategy and used the materials for methylene blue degradation under visible light irradiation. 2.1. Doping of TiO2 Anandan et al. studied the photodegradation of monocrotophos under visible light irradiation with La-doped TiO2. They associated rapid mineralization with the enhanced separation of electrons and holes by doping La3+ into the TiO2 matrix, which subsequently generated a large number of •OH radicals along with the trapping of excess holes at the surface [35]. In contrast, based on the density functional theory calculation method, Sun et al. worked extensively on the changes of the electronic structure and the photocatalytic activity of TiO2 after introducing substitutional La dopants. Their calculations demonstrate that the enhanced visible light absorption of La–TiO2 mainly arises from adsorbed La on the TiO2 surface rather than from substitutional La doping [36]. Notably, not all kinds of dopants give rise to positive consequences. Chio et al. systematically studied 21 kinds of metal ion-doped TiO2 materials and their application with respect to various photocatalytic reactions [37]. The results associated with model reactions for the photocatalytic reduction of carbon tetrachloride and the photodegradation of chloroform indicated that only the doping of certain ions, such as Fe3+, Ru3+, Re5+, V4+, and Mo5+, increased reactivity. In addition, the study demonstrated that optimizing the content and placement of the dopant ions content play a positive role in affecting photocatalytic activity. Despite the robust photoactivity of certain metal ion-doped TiO2 catalysts, some inevitable problems remain and need to be considered. The metal-doped nanomaterials have been shown to suffer from unstable optical properties and thermal instability, in addition to the need to use expensive ion implantation equipment to produce these enhanced materials [38]. Furthermore, the localized d-electron state formed in the band gap of TiO2 may become the recombination center of photogenerated electron–hole pairs, thereby leading to a decline in the photocatalytic activity. 149 Catalysts 2019, 9, 191 Catalysts 2019, 9, 191 S atoms with large diameters are difﬁcult to dope into the TiO2 lattice, but X-ray photoelectron spectroscopy (XPS) detected the existence of S–Ti–O bonds, which introduced lattice defects, acting as shallow traps for electrons and reducing carrier recombination [44]. Li et al. mixed HIO3 with tetrabutyl titanate and hydrolyzed the samples directly to obtain I-doped TiO2, which signiﬁcantly boosted its visible light performance [45]. Although various non-metal ions are used for doping modiﬁcation of TiO2, N doping is still one of the most widely used methods to modify the electronic structure and to extend light absorption to the visible range [46]. However, researchers have not yet come to a complete agreement regarding the mechanisms associated with the N doping enhancements. In the literature, it is not difﬁcult to ﬁnd studies stating that it is not only the dopant concentration but also the dopant location in the TiO2 lattice (surface or bulk, substitutional, and interstitial) that ultimately determines the photocatalytic properties [17,47]. In the case of N-doped TiO2 nanomaterials, some researchers believe that only the substitution of O2−by N3−with high dopant concentrations can elevate the valence band edge, 150 Catalysts 2019, 9, 191 Catalysts 2019, 9, 191 bringing about the desired band gap narrowing [48,49]. However, others suggest that the doping of N will induce oxygen vacancies in TiO2 and that the enhanced visible light adsorption is associated with the local state induced in the band gap, rather than the generally believed theory that the introduction of N into the TiO2 lattice can reduce its band gap, as shown in Figure 7 [50]. bringing about the desired band gap narrowing [48,49]. However, others suggest that the doping of N will induce oxygen vacancies in TiO2 and that the enhanced visible light adsorption is associated with the local state induced in the band gap, rather than the generally believed theory that the introduction of N into the TiO2 lattice can reduce its band gap, as shown in Figure 7 [50]. Figure 7. (a) Diffuse reﬂectance spectra of the anatase TiO2 nanobelts before and after heat treatment in ammonia gas ﬂow at different temperatures and (b) the band structure of N-doped-TiO2 under visible and UV light irradiation [50]. Copyright © 2009 American Chemical Society. Figure 7. 2.2. Intrinsic Defect Formation Catalysts 2019, 9, 191 (a) Diffuse reﬂectance spectra of the anatase TiO2 nanobelts before and after heat treatment in ammonia gas ﬂow at different temperatures and (b) the band structure of N-doped-TiO2 under visible and UV light irradiation [50]. Copyright © 2009 American Chemical Society. As another widely studied non-metal-doped TiO2, F-doped TiO2 also shows promising potential for photocatalytic applications. Zhang et al. obtained F-doped TiO2 mesocrystals through the topological transformation of TiOF2 precursors. An in situ characterization technique was adopted to detect the doping process. The results showed that the doping of F was accompanied by the formation of oxygen defects, which ensured a higher visible light response [51]. Park et al. added sodium ﬂuoride to aqueous TiO2 suspensions to obtain surface ﬂuorinated TiO2, and a series of characterizations showed that neither an improvement in crystallinity nor a redshift of the band edge was achieved, but the photocatalytic oxidation of phenol and Acid Orange was considerably enhanced. They attributed such photocatalytic improvement to ﬂuorine surface modiﬁcation, which enhances free •OH radical-mediated oxidation pathways [19]. Similar to the doping of N, the reason for the observed high performance upon F doping is still undetermined. Some studies suggest that instead of entering the TiO2 lattice, ﬂuorine ions adsorbed on the surface of TiO2 can increase the wettability and surface acidity, which is beneﬁcial to the adsorptivity and e−/h+ separation of the oxide [20]. Other researchers hold the opinion that a tail state in the band gap of TiO2 is formed by F doping, which favors the more efﬁcient utilization of incident light. Recently, an increasing number of studies proposed that a charge compensation effect induced by F doping brings about the formation of a certain amount of oxygen vacancies and Ti3+ in TiO2, resulting in the enhanced absorption of visible light [52,53]. Although the principle of F doping is not very clear, the proper doping level of F can effectively improve the activity of TiO2. 2.2. Intrinsic Defect Formation 2.2. Intrinsic Defect Formation In 2011, a black TiO2 with a narrowed bandgap (approximately 1.5 eV) and fabricated by hydrogenation reduction was reported to achieve absorption of full spectrum sunlight and improved photocatalytic activity [54]. Unsurprisingly, this discovery has aroused worldwide scientiﬁc interest 151 Catalysts 2019, 9, 191 and paved the way towards intrinsic defect modiﬁcation. Creating intrinsic defects in the TiO2 lattice is a kind of self-structural modiﬁcation that includes surface disorder layers, Ti3+/oxygen vacancy self-doping, formation of surface Ti–OH, and incorporation of doped-Consequentially, considerable changes in surface properties and electronic and crystal structures are often achieved in this process [55–57]. Furthermore, studies in terms of defect engineered TiO2 have conﬁrmed that these intrinsic defects are emerging as a promising attribute for improving the separation of electrons and holes, outperforming, in some cases, other kinds of modiﬁed TiO2 nanomaterials [58]. Since the study by Chen et al., various methods have been developed to induce defects in TiO2, including direct reduction of TiO2; that is, the currently reported H2, Al, Na, Mg, NaBH4, hydrides, imidazoles, etc. can effectively transfer modify pure TiO2 nanomaterials into their defect engineered counterparts under certain conditions [59,60]. In addition, electrochemical reduction and high-energy particle bombardment (such as photon beam and H2 plasma or electron beam) are widely used to induce TiO2 defects. Partial oxidation from low-valence-state Ti species such as TiH2, TiO, TiCl3, TiN, and even Ti foil represents another promising approach, fulﬁlling the needs for highly active TiO2−x photocatalysts [61]. Liu et al. prepared rice-shaped Ti3+ self-doped TiO2−x nanoparticles through mild hydrothermal treatment of TiH2 in H2O2 aqueous solution, and proposed a unique “surface oxide-interface diffusion–redox mechanism” (as shown in Figure 8) to explain the formation process of TiO2−x [62]. The defect types and their formation mechanism in TiO2−x are closely related to the preparation methods. Generally, the Ti–H bond is present only in hydrogen-reduced TiO2−x, while the surface disorder layer causes severe damage to the TiO2 structure. Thus, relatively strong reduction conditions are required, such as high temperature/pressure hydrogen reduction, aluminothermic reduction, hydrogen plasma treatment, etc. Surface Ti–OH, Ti3+, and oxygen vacancies commonly exist in most defective TiO2 nanostructures [63]. Figure 8. (A) Schematic of the formation mechanisms for the rice-shaped Ti3+ self-doped TiO2−x nanoparticles. (B,C) The interface diffusion–redox diagram. The green arrows indicate ion diffusion [62]. Copyrighted 2014 The Royal Society of Chemistry. Figure 8. 2.2. Intrinsic Defect Formation (A) Schematic of the formation mechanisms for the rice-shaped Ti3+ self-doped TiO2−x nanoparticles. (B,C) The interface diffusion–redox diagram. The green arrows indicate ion diffusion [62]. Copyrighted 2014 The Royal Society of Chemistry. The dominant mechanism involved in improving photocatalytic performance by inducing intrinsic defects into TiO2 can be explained, both experimentally and theoretically, to be the regulation of the band structure of TiO2 and boosted charge separation and transport. For black TiO2, band tail states and shallow dopant states can be formed to reduce its band gap and further increase its optical absorption properties. Chen et al. observed a disordered surface layer in black TiO2 nanocrystals after a hydrogenation treatment, as shown in Figure 9. From the high-resolution transmission electron microscopy (HRTEM) spectra, it can be readily observed that the straight lattice fringes are bent at the edge of the particles, and the lattice spacing is no longer uniform, indicating that the hydrotreated 152 Catalysts 2019, 9, 191 black TiO2 nanoparticles possess a “crystal-disordered” core–shell structure. Such a disordered layer is believed to facilitate the introduction of the tail state at the top of the valence band and the bottom of the conduction band, consequently yielding a redshift of the light absorption [54]. Moreover, because the disorder layer exhibits a set of properties that are distinct from those of their crystalline counterparts, rapid charge separation could be realized when the amorphous layer closely contacts crystalline TiO2. The lattice distortions tend to blueshift the VBM while having less impact on CBM. Therefore, the photogenerated holes accumulate in the thin disordered shell and participate in the photocatalytic reactions immediately; electrons are widely spread in both the shell and core regions. This result highlights the strong synergistic effect on charge transfer between the crystalline and disordered parts [64]. Figure 9. (A) Schematic illustration of the structure and electronic DOS of a semiconductor in the form of a disorder-engineered nanocrystal with dopant incorporation. (B) A photo comparing unmodiﬁed white and disorder-engineered black TiO2 nanocrystals. (C,D) HRTEM images of TiO2 nanocrystals before and after hydrogenation, respectively [54]. Copyright 2011 American Association for the Advancement of Science. Figure 9. (A) Schematic illustration of the structure and electronic DOS of a semiconductor in the form of a disorder-engineered nanocrystal with dopant incorporation. (B) A photo comparing unmodiﬁed white and disorder-engineered black TiO2 nanocrystals. (C,D) HRTEM images of TiO2 nanocrystals before and after hydrogenation, respectively [54]. 2.2. Intrinsic Defect Formation Copyright 2011 American Association for the Advancement of Science. For Ti3+/oxygen vacancy incorporation and H-doping in reduced TiO2−x, the hybridization of Ti-3d, O-2p and H-1s orbitals results in the mid-gap states formation below the CBM and the Fermi level’s upshift [65,66]. The extra electrons in either Ti3+ or oxygen vacancies are inclined to occupy the empty states of Ti ions, forming new Ti 3d bands below the CBM. With a further increase in defect concentration, the 3d band shifts deeper and ﬁnally results in multiple bands in the CBM. Moreover, the existence of multiple mid-gap states as well as the associated derivate (surface Ti–OH) can also function as extra carrier trap sites or carrier scavengers to prolong the lifetime of electrons and holes [67]. The high concentration of electron donors will greatly improve the conductivity of materials and promote the transfer of carriers [68]. Wang et al. treated pure white TiO2 with hydrogen plasma to fabricate H-doped black TiO2 for photodegradation of methyl orange under visible light irradiation. The as-prepared samples showed a degradation rate 2.5 times that of the white counterpart [69]. Sinhamahapatra et al. reported a novel controlled magnesiothermic reduction to synthesize reduced TiO2−x under 5% H2/Ar atmosphere [70]. During this process, the band position and band gap, surface defects and oxygen vacancies can be well regulated to maximize the optical adsorption in the visible and infrared regions and minimize the charge recombination centers. As shown in Figure 10, a new controlled magnesium thermal reduction method to synthesize and reduce black TiO2 under 5% H2/Ar atmosphere. The material has the best band gap and band position, oxygen vacancy, surface 153 Catalysts 2019, 9, 191 defect, and charge recombination center, and the optical absorption in visible and infrared regions is improved obviously. These synergistic effects enable the defective TiO2−x with Pt as a co-catalyst to produce H2 at a rate of 43 mmol h−1 g−1 under the full solar wavelength light illumination, superior to other reported photocatalysts for hydrogen production. Figure 10. (a) H2 generation proﬁle, (b) rate (rH2) of hydrogen generation for different samples, and (c) the stability study of the sample BT-0.5 under the full solar wavelength range of light [70]. Copyright 2015 The Royal Society of Chemistry. Figure 10. 3. Morphology Modiﬁcation It is well known that the photocatalytic performance of semiconductors is closely related to their structural and morphological characteristics at the nanoscale, including their size, dimensionality, pore structure and volume, speciﬁc surface area, exposed surface facets, and crystalline phase content [74]. During the past few decades, numerous promising structure engineering strategies have been developed to fabricate highly active photocatalysts with the desired morphology and structure. Among them, particular emphasis has been placed on controlling and optimizing the structural dimensionality of a given semiconductor to improve its photocatalytic efﬁciency. Zero-dimensional TiO2 nanospheres are the most widely studied TiO2-based materials because of their high speciﬁc surface area and attractive pore structures [75–77]. Figure 12 shows a classic ripening approach to synthesize hollow nanospheres [75]. As photocatalytic reactions take place on the surface of the photocatalyst, TiO2 nanoparticles with smaller sizes are inclined to provide more reactive sites, resulting in better photocatalytic performance. Moreover, due to the quantum size effect, the photogenerated electrons and holes in the bulk regions are able to migrate to the surface of TiO2 nanoparticles via shorter distances, thereby considerably reducing the carrier quench rate [78]. TiO2 nanospheres are also good candidates as light captors, and their structural features enable as much light as possible to access the interior, resulting in amazing light harvesting capabilities. However, it should be mentioned that the diffusion length of photogenerated electrons and holes must be longer than the particle size to avoid the recombination of the dominant carriers on the surface of the photocatalyst, which is very important for achieving efﬁcient charge carrier dynamics [79]. One-dimensional (1D) nanostructures, including nanotube (NT), nanorod (NR), nanobelt (NB), and nanowire (NW), have become a popular research topic in recent years. They have been extensively studied because of their distinct optical, electronic and chemical properties. Despite some similar features with nanoparticles, such as quantum conﬁnement effects and large surface area, 1D nanomaterials possess many unique properties, which are hard for other categories of structured materials to achieve. For example, 1D nanostructures restrict the migration of electrons and protons by allowing the lateral conﬁnement of electrons/protons and guide their transport in the axial direction [80,81]. Furthermore, excellent ﬂexibility and mechanical properties enable them to be easily used and recycled. 2.2. Intrinsic Defect Formation (a) H2 generation proﬁle, (b) rate (rH2) of hydrogen generation for different samples, and (c) the stability study of the sample BT-0.5 under the full solar wavelength range of light [70]. Copyright 2015 The Royal Society of Chemistry. To date, numerous strategies, either common or uncommon, have been developed to introduce various kinds of dopants or defects into the TiO2 matrix. However, considering its highly stable nature, most methods are rigorous and energy-consuming, and are contrary to the sustainable and environmentally friendly development criteria. Therefore, an increasing number of studies are dedicated to seek convenient, economical, energy efﬁcient, and environmentally friendly methods for the structural modiﬁcation of TiO2 [71]. In our recent studies, we developed a facile photoreduction strategy to induce intrinsic defects into anatase TiO2 to modulate its band structure, thereby extending the absorption of incident light to the visible region. As shown in Figure 11, the band gap was narrowed to 2.7 eV, and the color changed to earth yellow after the photoreduction treatment. NH4TiOF3 mesocrystals were adopted as precursors, which can release ﬂuorine and nitrogen ions during the topological transformation process. Thus, non-metal ion doping (i.e., F and N ions) was also achieved simultaneously, further improving the transport and separation of photogenerated charge carriers. The as-prepared NF–TiO2−x exhibited excellent photocatalytic degradation and photoelectrochemical efﬁciency under visible light irradiation compared to pristine TiO2 [72,73]. 154 Catalysts 2019, 9, 191 Figure 11. (a) UV–Vis diffuse reﬂectance spectra and (b) Tauc plot for band gap determination [73]. Copyright 2018 Springer Nature Publishing AG. Figure 11. (a) UV–Vis diffuse reﬂectance spectra and (b) Tauc plot for band gap determination [73]. Copyright 2018 Springer Nature Publishing AG. 3. Morphology Modiﬁcation In this regard, 1D TiO2-ordered nanostructures are promising not only for constructing highly active photocatalytic systems but also for building blocks for various (photo)electrochemical devices, such as batteries, fuel cells, solar cells, and photoelectrochemical cells. To further optimize the photocatalytic reactivity of 1D TiO2 nanomaterials, one can precisely regulate the aspect ratio (the ratio of length to diameter) or modify these 1D nanostructures with novel strategies to accelerate electron transport and separation processes, as well as to enhance the capture of incident light; TiO2 nanotubes are examples of these materials [82]. Through the electrochemical anodization process, it is possible to precisely control the tube crystal structure 155 Catalysts 2019, 9, 191 (anatase, rutile, or amorphous) and tube geometry (diameter and length), as shown in Figure 13a, or direct the tube arrangements to obtain a deﬁned tube-to-tube interspace (Figure 13b). For the sake of extending the scope of application, constructing ﬂow through membranes with TiO2 nanotubes is a good choice (Figure 13c). Other modiﬁcations for minimizing charge carrier annihilation and boosting light harvesting are illustrated in Figure 13d–i, ranging from self-decoration to surface alterations to energy band engineering. Figure 12. (A) Schematic illustration (cross-sectional views) of the ripening process and two types (i and ii) of hollow structures. Evolution (TEM images) of TiO2 nanospheres synthesized with 30 mL of TiF4 (1.33 mM) at 180 ◦C with different reaction times: (B) 2 h (scale bar = 200 nm), (C) 20 h (scale bar = 200 nm), and (D) 50 h (scale bar = 500 nm) [75]. Copyright 2004 American Chemical Society. Figure 12. (A) Schematic illustration (cross-sectional views) of the ripening process and two types (i and ii) of hollow structures. Evolution (TEM images) of TiO2 nanospheres synthesized with 30 mL of TiF4 (1.33 mM) at 180 ◦C with different reaction times: (B) 2 h (scale bar = 200 nm), (C) 20 h (scale bar = 200 nm), and (D) 50 h (scale bar = 500 nm) [75]. Copyright 2004 American Chemical Society. TiO2 nanosheets, nanoﬂakes, and thin ﬁlms consist of titania-based two-dimensional nanomaterials, which have ﬂat surfaces and high aspect ratios. The lateral size of some nanomaterials is controllable, ranging from the sub-micrometer or even nanometer level to several tens of micrometers with thicknesses of 1–10 nm. Such structures provide TiO2 nanomaterials with several unique characteristics, such as excellent adhesion to substrates, low turbidity and high smoothness [83]. 3. Morphology Modiﬁcation Furthermore, when exposed to UV light irradiation, TiO2 2D nanomaterials exhibit superhydrophilicity, which leads to a variety of potential applications, such as self-cleaning coatings and electrodes in photoelectronic devices [84]. Notably, considering that photocatalytic reactions always occur on the surface of catalysts, the exposed crystal facets are of great importance in determining the photocatalytic performance. Accordingly, developing TiO2 crystals with different active facets is highly desirable in many applications. In general, TiO2 nanocrystals have three basic low-index exposed facets—{101}, {001}, and {010}—with surface energy relationships of {001}, 0.90 J m−2 > {100}, 0.53 J m−2 > {101}, 0.44 J m−2 [85,86]. Therefore, as the most thermodynamically stable facets, the {001} crystal facet is dominant among most anatase TiO2 nanomaterials, reducing the overall surface energy of the material. In 2008, Yang et al. ﬁrst reported TiO2 single crystals with 47% highly active {001} facets exposed to HF as capping agents [87]. This work has attracted considerable global attention. Since then, TiO2 with various ratios of exposed {001} facets have been successfully fabricated [88]. Meanwhile, other active planes, such as {010}, {111}, and {110}, have also been reported and widely used in water splitting, solar cells, artiﬁcial light synthesis and other ﬁelds, as shown in Figure 14 [89]. Zheng et al. obtained {001} facet-oriented anatase by facile heat treatment of a tetrabutyl titanate, absolute ethanol, and HF mixture. 156 Catalysts 2019, 9, 191 Such a material with 85% {001} facets exhibited much higher photocatalytic activity in comparison to commercial P25 materials [90]. Figure 13. Schematic drawing of (a,b) formation and (c−i) modiﬁcation of anodic nanotube arrays (as discussed in the text) [82]. Copyright 2017 American Chemical Society. Figure 13. Schematic drawing of (a,b) formation and (c−i) modiﬁcation of anodic nanotube arrays (as discussed in the text) [82]. Copyright 2017 American Chemical Society. During the process of photocatalytic reactions, oxidation predominantly occurs in the {001} facets, while reduction occurs in the {101} crystal plane of TiO2 because the {101} facet (with relatively low surface energy) tends to attract more electrons. Electron holes subsequently accumulate in the {001} plane, facilitating the space separation of electron–hole pairs [91]. In addition, Ti atoms of the {001} plane exist mainly in the form of 5-coordination, which can provide more active sites that more readily attract free reactant molecules than the {101} plane. 3. Morphology Modiﬁcation Thus, when a certain proportion of {001} crystal facets are exposed, the photocatalytic activity increases rapidly. Nevertheless, it is not always the case that a higher {001} crystal face exposure ratio results in improved catalytic performance. Studies have reported that the photocatalytic activity is compromised when the proportion of {001} facets exceeds 71% [89]. In addition, faceted TiO2 photocatalysts suffer from weak visible light utilization due to their large band gap. Hence, the modiﬁcation of the electronic structure of faceted TiO2 to fully utilize sunlight and promote the migration and separation of electron/hole pairs is highly desirable. Wang et al. prepared Ti3+ self-doped TiO2 mesoporous nanosheets dominated by {001} facets with supercritical technology. They associated the extended region of incident light absorption with the introduction of Ti3+ [91]. Using an ionic liquid as a surface control agent, Biplab et al. 157 Catalysts 2019, 9, 191 synthesized microporous TiO2 nanocrystals with exposed {001} facets. After depositing Pt on the surface, the hydrogen production rate in visible irradiation was greatly improved [92]. Figure 14. Summary of main shapes and applications (i.e., lithium ion batteries, photocatalytic hydrogen evolution, photodegradation, and solar cells) of anatase, rutile, and brookite TiO2 crystals with their surfaces consisting of different Facets [89]. Copyright 2014 American Chemical Society. Figure 14. Summary of main shapes and applications (i.e., lithium ion batteries, photocatalytic hydrogen evolution, photodegradation, and solar cells) of anatase, rutile, and brookite TiO2 crystals with their surfaces consisting of different Facets [89]. Copyright 2014 American Chemical Society. A three-dimensional TiO2 hierarchical structure based on intrinsic shape-dependent properties has been the central focus of many recent studies. Designed and fabricated 3D TiO2 nanomaterials commonly incorporate interconnected structures, hollow structures and hierarchical superstructures constructed from small dimensional building blocks [93]. Most of these novel structures include larger spatial dimensions and more varied morphologies. The high surface-to-volume ratio provides a more efﬁcient diffusion path for reactant molecules, enabling the contaminant molecules to enter the framework of the photocatalyst for efﬁcient puriﬁcation, separation, and storage. In addition, the unique optical characteristic is of particular interest because many of these architectures have distinctive physicochemical properties favorable for incident light utilization. For example, when light is irradiated onto the surface of the TiO2 hierarchical structure, photons are scattered multiple times, so the probability of the catalyst absorbing photons is increased; this phenomenon is known as the “trapping effect” and is illustrated in Figure 15 [94]. 3. Morphology Modiﬁcation Figure 15. Schematic diagram of the reﬂecting and scattering effects in hierarchical microspheres [94]. Copyright 2014 The Royal Society of Chemistry. Figure 15. Schematic diagram of the reﬂecting and scattering effects in hierarchical microspheres [94]. Copyright 2014 The Royal Society of Chemistry. 158 Catalysts 2019, 9, 191 Catalysts 2019, 9, 191 The hollow structure TiO2 nanomaterials have attracted considerable attention due to their amazing light harvesting ability, low density, and large speciﬁc surface area. The hollow structure, on the one hand, is capable of providing a large amount of space to accommodate more reactant molecules, thereby increasing the effective contact between the catalyst and the reactants. On the other hand, incident light inside the cavity can undergo multiple reﬂections to capture more light, as shown in Figure 16 [95]. Kondo et al. obtained TiO2 hollow nanospheres through hydrothermal and calcination processes with polymer polyethylene cationic balls as templates. The as-prepared photocatalyst had more favorable activity than its commercial counterparts with respect to decomposing isopropanol [96]. In the following work, an ultrathin TiO2 shell-like structure was prepared in a similar manner with a shell thickness of approximately 5 nm. The morphology of the TiO2 hollow materials prepared by the hard template method is relatively uniform, and the composition and thickness of the shells are adjustable. However, the preparation process is complicated and requires multiple execution steps to be realized. Moreover, the hollow structure may be destroyed when the template is removed. Therefore, alternative strategies, including soft templates and non-template methods, have played an increasingly important role in the development of hollow structure TiO2 nanomaterials in recent years. Figure 16. Comparison of photocatalytic activities of titania spheres with solid, sphere-in-sphere, and hollow structures [95]. Copyright 2007 American Chemical Society. Figure 16. Comparison of photocatalytic activities of titania spheres with solid, sphere-in-sphere, and hollow structures [95]. Copyright 2007 American Chemical Society. Li et al. prepared hollow TiO2 nanospheres with high photocatalytic activity by a template-free process. The increased catalytic activity is mainly due to the multiple reﬂections of incident light inside the TiO2 sphere, which extends the optical path [97]. Multichannel TiO2 hollow nanoﬁbers were constructed by Zhao et al. for degrading gaseous acetaldehyde, and the speciﬁc surface area of this material increased rapidly as the number of channels increased. 4. TiO2 Mesocrystals It is widely accepted that for TiO2-based photocatalytic materials, large crystallites result in high structural coherence, which beneﬁts the transfer and separation of electron–hole pair, while the availability of plentiful reaction sites is dependent on obtaining large speciﬁc surface areas. However, producing a structure that simultaneously satisﬁes the requirements of large crystallites and high surface area is extremely challenging. Fortunately, the advent of mesocrystals is a promising material that may meet the challenge [100]. Mesocrystals were ﬁrst proposed by Cölfen and Antonietti in 2005, and since then have received increased attention [101]. Different from the classical single crystals in which the crystal lattice of the entire sample is continuous with no grain boundaries and polycrystals whose units do not have the same orientation, mesocrystals are a new kind of superstructure material that follow a nonclassical crystallization process involving crystallographically ordered assemblies of nanocrystal building blocks. The relevant formation mechanisms of TiO2 mesocrystals reported thus far mainly include topotactic transformation, mineral bridges, nanoparticle alignment with organic matrices, physical ordering, space constraints, and self-similar growth [100]. Different methods may give rise to different structures and morphologies, but the as-prepared TiO2 mesocrystals are usually single-crystal-like structures with high porosity, surface area, and crystallinity; they are considered periodically hierarchical structures that are similar to sophisticated biominerals. All of these features pave the way for a wide range of applications, such as catalysis and energy storage and conversion [102]. Fabrication and modiﬁcation strategies for TiO2 mesocrystals have developed rapidly in recent years. Due to the similar structure between NH4TiOF3 and TiO2, preparing TiO2 mesocrystals through topotactic transformation from NH4TiOF3 represents an innovative process. As illustrated in Figure 17, the critical parameters in the {001} facets of both NH4TiOF3 and TiO2 are quite similar, with an average lattice mismatch of 0.02%. The position of titanium atoms in the {001} plane of TiO2 is similar to NH4TiOF3, but in NH4TiOF3, these are separated by ammonium ions in a lamellar structure. Hence, it is reasonable to use NH4TiOF3 as a starting material, transforming it into TiO2 mesocrystals by thermal decomposition or aqueous hydrolysis with H3BO3 [71]. Based on this mechanism, Majima et al. performed extensive studies on tailoring TiO2 mesocrystals with versatile structures and morphologies, as well as postmodiﬁcations to further improve their photocatalytic efﬁciency. For example, to investigate the anisotropic electron ﬂow in different facets and to maximize their separation during the photocatalytic reaction, Zhang et al. 3. Morphology Modiﬁcation They proposed that the multichannel hollow structures induced both an inner trap effect on gaseous molecules and a multiple-reﬂection effect on incident light, which were the main reasons for the improved photocatalytic activity of TiO2 hollow ﬁbers [98]. Shang et al. synthesized submicron-sized TiO2 hollow spheres from a mixture of TiCl4, alcohols, and acetone by a template-free solvothermal method. Control of the sphere size was achieved by adjusting the ratio of ethanol to acetone. Based on a series of characterizations, they suggested a possible formation mechanism for the hollow structure: the tiny anatase phase TiO2 nanoparticles with poor crystallinity form through a hydrolysis reaction, due to the very high surface energy, and then quickly aggregate to form spheres. The increased water promotes the crystallinity of particles in the spherical shell, while the internal particles dissolve and migrate to the spherical shell, leading to the formation of highly crystalline TiO2 hollow spheres [99]. An intriguing work carried out by Kang et al. to establish hierarchical anatase TiO2 nanocubes with hollow structures has been reported recently. Instead of seeking complicated templates or surfactants, they directly 159 Catalysts 2019, 9, 191 converted NH4TiOF3 mesocrystals to hollow spiny TiO2 with a high speciﬁc area and photodegradation activity [73]. converted NH4TiOF3 mesocrystals to hollow spiny TiO2 with a high speciﬁc area and photodegradation activity [73]. 4. TiO2 Mesocrystals controllably synthesized a speciﬁc facet-dominated TiO2 superstructure with NH4F as an orientation-directing agent [103]. Under UV light irradiation, mesocrystals with different facet ratios showed different reactivity orders in the photooxidation of 4-chlorophenol, i.e., {001} > {101} (by 1.7 times), and photoreduction, i.e., {101} > {001} (by 2–3 times). Moreover, constructing the composite of MoS2 and TiO2 mesocrystals, as well as the co-catalyst selective modiﬁcation on TiO2, also showed the desired separation of photogenerated charge carriers during the hydrogen evolution reaction [104]. In terms of extending the absorption of incident light to the visible region, Zhang et al. tried doping or codoping non-mental elements into the TiO2 matrix to examine the effects on its electronic structure and band gap. An in situ ﬂuorine-doped TiO2 superstructure was recently realized. F doping into TiO2 mesocrystals for the incorporation of active color centers facilitates visible light harvesting and accelerates charge separation for hydrogen generation [51]. They further introduced nitrogen and ﬂuorine codopants into {001} facet-oriented TiO2 mesocrystals during topochemical transformation for photoreduction of Cr(VI) under visible light illumination. The extended optical light absorption could be attributed to doped nitrogen, which introduces the isolated mid-gap state. The high yield of hydroxyl radicals and preferential adsorption are correlated with ﬂuorine doping, as conﬁrmed by the comparison between untreated TiO2 with 160 Catalysts 2019, 9, 191 Catalysts 2019, 9, 191 TiO2 washed in NaOH aqueous solution. The synergistic effect on charge separation and trapping was suggested through a femtosecond time-resolved diffused reﬂectance (TDR) measurement [105]. As shown in Figure 18, the g-C3N4 nanosheet/TiO2 mesocrystal metal-free composite was successfully constructed by Elbanna et al. [106]. The as-prepared sample exhibited an excellent hydrogen evolution rate under visible light irradiation without any noble metal co-catalyst. Then, they further broadened the light capture of the TiO2 mesocrystals to include near-infrared regions. Au nanorods (NRs) with various aspect ratios were loaded onto the surface of TiO2 by the ligand exchange method. Different aspect ratios resulted in different incident light absorption and photogenerated electron transfer. The highest photocatalytic activity of Au NRs and TMC composites reached 924 μmol h−1 g−1 under visible-near-infrared (NIR) light irradiation [107]. Figure 17. Illustration of the oriented transformation of NH4TiOF3 mesophyte to TiO2 (anatase) mesocrystal [71]. Copyright 2008 American Chemical Society. Figure 17. Illustration of the oriented transformation of NH4TiOF3 mesophyte to TiO2 (anatase) mesocrystal [71]. Copyright 2008 American Chemical Society. Figure 18. 4. TiO2 Mesocrystals Representative scheme of electron injection and movement in g-C3N4 NS (31 wt %)/TMC during visible-light irradiation [106]. Copyright 2017 American Chemical Society. Figure 18. Representative scheme of electron injection and movement in g-C3N4 NS (31 wt %)/TMC during visible-light irradiation [106]. Copyright 2017 American Chemical Society. 161 Catalysts 2019, 9, 191 Catalysts 2019, 9, 191 Considering the aforementioned merits of mesocrystal nanomaterials, we recently tried different approaches to further improve the optical absorption properties of TiO2 mesocrystals, in addition to their enhanced transfer and separation properties. Oxygen vacancies and N dopants were successfully introduced into the TiO2 lattice with a facile low temperature calcination process [108], as shown in Figure 19. NH4TiOF3 mesocrystal nanocubes were used as precursors in our system, and topological transformation from NH4TiOF3 to TiO2 facilitated the release and doping of nitrogen. Oxygen vacancies were also readily produced in the inert heating atmosphere. The signiﬁcantly improved photodegradation and photoelectrochemical performance under visible light irradiation may be associated with the unique structure of mesocrystals as well as the introduction of foreign and intrinsic defects. Figure 19. Schematic representation of the synthesis of TiOx nanosheets. X-ray powder diffraction (XRD) pattern of (a) NH4TiOF3 and (b) N/TiO2−x. SEM images of (c,e) NH4TiOF3 and (d,f) N/TiO2−x [108]. Copyright 2019 The Royal Society of Chemistry. Figure 19. Schematic representation of the synthesis of TiOx nanosheets. X-ray powder diffraction (XRD) pattern of (a) NH4TiOF3 and (b) N/TiO2−x. SEM images of (c,e) NH4TiOF3 and (d,f) N/TiO2−x [108]. Copyright 2019 The Royal Society of Chemistry. 5. Separation of Charges Since metals and metal oxides have different working functions, resulting in the formation of a Schottky potential barrier, an effective modiﬁcation method is to deposit precious metals (Ag, Au, or Pt) on the surface of metal oxides. Since metals and metal oxides have different working functions, resulting in the formation of a Schottky potential barrier, an effective modiﬁcation method is to deposit precious metals (Ag, Au, or Pt) on the surface of metal oxides. Choi et al. presented Ag/TiO2 by a photodeposition method [109]. Due to the different transfer rates of interface charges between electrons and holes to redox species in water, excessive charges can accumulate on photocatalysts [110,111]. By depositing Ag, which can provide a temporary home for excessive electrons, the composite utilized the electron storage capacity to promote the separation of electrons and holes to reduce Cr(VI) in the following dark period. Li et al. prepared a sandwich structure with CdS-Au-TiO2 on a ﬂuorine-doped tin oxide (FTO) substrate [112]. In this composite structure, Au nanoparticles not only acted as an electronic relay between CdS quantum dots (QDs) and TiO2 to increase charge separation occurring on a long-time scale but also served as a plasma photosensitizer that prolonged the photoconversion to improve the absorption range of light. The rate of charge transfer and reverse transfer depends on the relative energy of the hot plasma electrons to the Schottky barrier [112]. The PEC performance is represented in Figure 20. 162 Catalysts 2019, 9, 191 Figure 20. (a) Electron relay effect of Au nanoparticles, facilitating the charge transfer from CdS QDs to TiO2 nanorods under the irradiation of incident solar light with a wavelength <525 nm. (b) Plasmonic energy transfer from the excited Au nanoparticles to TiO2 through hot electron transfer under the irradiation of incident solar light with a wavelength >525 nm. CB = conduction band, VB = valence band, EF = Fermi energy level, and Φb = Schottky barrier [112]. Copyright 2014 American Chemical Society. Figure 20. (a) Electron relay effect of Au nanoparticles, facilitating the charge transfer from CdS QDs to TiO2 nanorods under the irradiation of incident solar light with a wavelength <525 nm. (b) Plasmonic energy transfer from the excited Au nanoparticles to TiO2 through hot electron transfer under the irradiation of incident solar light with a wavelength >525 nm. 6. Application of TiO2 Nanomaterials 6. Application of TiO2 Nanomaterials Over the past several years, semiconductors, especially titanium dioxide, have been widely used as photocatalysts. It is well known that there are three main steps associated with the photocatalysis process: (1) generation of electrons and holes after the absorption of photons; (2) separation and migration of the charge; and (3) transition of the charge and reaction between the carriers and the reagent. To date, TiO2 has been mainly applied in the areas of environmental conservation, new energy resources, and so on. In this section, we will focus on recent progress in these photocatalytic applications of TiO2. 6.1. Applications in the Environment 5. Separation of Charges CB = conduction band, VB = valence band, EF = Fermi energy level, and Φb = Schottky barrier [112]. Copyright 2014 American Chemical Society. Precious metal deposition can greatly improve the performance of catalysts, but the scarcity of precious metals dramatically limits this modiﬁcation method and makes it difﬁcult to achieve industrial-scale production. In this case, the search for an inexpensive and efﬁcient doped composite has also attracted much attention. Carbon, abundant on earth, has good electrical conductivity, and its combination with TiO2 can result in excellent photocatalytic performance. Wang et al. demonstrated TiO2–carbon nanoparticles by the sol–gel method and then synthesized core–shell-structured TiO2 and amorphous carbon [113]. This unique morphology and structure result in the modiﬁed TiO2 sample exhibiting enhanced responsiveness and excellent photocatalytic activity. Due to the rapid charge transfer in the carbon shell, both the carrier separation efﬁciency and the photodegradation of pollutants in water is improved. The reduced TiO2 is also more efﬁcient in the production of H2 due to its correct edge position. 6.1.1. Degradation of Aqueous Pollutants Industrial development is often accompanied by pollution of the environment, especially water. Photocatalytic water treatment using heterogeneous semiconductors under visible light is considered an eco-friendly technology. Photocatalysis involves the generation of large numbers of electrons and holes on the surface of TiO2 after the absorption of photons; the photogenerated holes have considerable oxidizing capacity and can degrade almost all organic contaminants including carbon dioxide (CO2). However, due to its own deﬁciencies, such as a wide bandgap and fast recombination of electrons and holes, TiO2 cannot make full use of sunlight to remove the pollutants in water. Wang et al. reported hydrogenation by TiO2 nanosheets with exposed {001} facets maintained by the formation of Ti–H bonds [114]. By annealing the ﬁne-sized pristine hydrothermal product under a high-pressure hydrogen atmosphere, the hydrogenation of F-modiﬁed anatase TiO2 nanosheets (with exposed high 163 Catalysts 2019, 9, 191 percentages of {001} facets) was achieved. Under UV–Vis and visible light irradiation, this material decomposed methylene blue (MB) faster than P25 and pristine TiO2, as shown in Figure 21. Figure 21. Photocatalytic decomposition of MB (a) and •OH generation measurement (b) of TiO2 and TiO2–H under UV–Vis light irradiation. Schematic illustration (c) of the hydrogenation effect on the structural change in TiO2 and TiO2–H [114]. Copyright 2012 The Royal Society of Chemistry. Figure 21. Photocatalytic decomposition of MB (a) and •OH generation measurement (b) of TiO2 and TiO2–H under UV–Vis light irradiation. Schematic illustration (c) of the hydrogenation effect on the structural change in TiO2 and TiO2–H [114]. Copyright 2012 The Royal Society of Chemistry. Plodinec et al. applied black TiO2 nanotube arrays with Ag nanoparticles, which promoted hydrogenation for the degradation of salicylic acid [115]. The photocatalyst can degrade salicylic acid effectively, and its photocatalytic performance far exceeds that of TiO2 nanotubes and commercial TiO2 P25 (the reference material used for the modeling of photocatalytic processes). Ling et al. prepared TiO2 nanoparticles (with diameters of 10–23 nm) that exhibited photocatalytic activity [116]. The initial degradation rate of phenol by a TiO2 nanocatalyst was 6 times higher than that achieved with H2O2 alone, and the addition of H2O2 to TiO2 can increase the initial concentration of hydroxyl radicals and accelerate the degradation rate. Hao et al. developed a TiO2/WO3/GO nanocomposite (via a hydrothermal synthesis), which presented excellent optical absorbance and displayed excellent photocatalytic activity for the degradation of bisphenol A [117]. 6.1.1. Degradation of Aqueous Pollutants In addition to the oxidizing capacity, the photogenerated electrons on TiO2 have strong reducing capacity to remove pollutants, such as Cd(II), Hg(II), As(V), and Cr(VI), from water; these cations can be reduced into less toxic metallic or ion states. Dusadee et al. fabricated a titania-decorated reduced graphene oxide (TiO2·rGO) nanocomposite via a hydrothermal process [110]. Studies on reducing the toxic Cr6+ (hexavalent chromium) ion toxicity using the titanium dioxide x/rGO numerical control have found that photocatalytic reduction of toxic Cr6+ generally increases with the increase in x. In addition, since rGO accelerates electron transport, the combination of photoexcited electrons and holes decreases leads to an increased duration of photocatalytic activity [118]. TiO2 has facilitated many pollutant degradation processes such as the reduction of nitrate, the degradation of acid fuchsin, the decomposition of acetaldehyde, and the dechlorination of CCl4 [119–122]. Due to the continued proliferation of environment pollutants, TiO2 and other nanostructured materials should be vigorously developed in the future to improve the degradation of pollutants by photocatalysis. 6.1.2. Degradation of Air Pollutants Just as industrial and technological developments can result in water pollution, so too can the atmosphere be adversely impacted by toxic pollutants that are emitted from chemical manufacturing plants, power plants, industrial facilities, transportation technologies, etc. Air pollution impacts the health of the global environment and the array of species that live within it, and new techniques are sought to reduce harmful airborne emissions. Highly efﬁcient oxidation and reduction during 164 Catalysts 2019, 9, 191 photocatalysis are considered to be an effective method to degrade inorganic and organic air pollutants to improve air quality [123–125]. Similarly, TiO2 is considered the most promising photocatalyst. Kakeru et al. prepared TiO2 nanoparticles with palladium sub-nanoclusters (<1 nm) using the ﬂame aerosol technique [126]. Under sunlight, these materials can remove NOx at approximately 3 to 7 times the rate of commercial TiO2 (P25, Evonik) (without Pd). Natércia et al. prepared new composite materials of TiO2 (P25) and N-doped carbon quantum dots (P25/NCQD) by a hydrothermal method, which was ﬁrst used as the photooxidation catalyst of NO under the irradiation of ultraviolet and visible light [127]. The experiment showed that the conversion rate of the P25/NCQD composite material (27.0%) was more than twice that of P25 (10%) without modiﬁcation, and the selectivity in visible light increased from 37.4% to 49.3%. The photocatalytic performance of the composite material in the UV region was also better than that of P25. Zeng et al. reported a H2 reduction strategy to produce H–TiO2 materials (with enhanced oxygen vacancy concentrations and distributions) that can promote formaldehyde decomposition in the dark [128]. Research of TiO2-based photocatalysts has also been conducted to facilitate removal of tetrachloroethylene [129], acetone [130], benzene [131], phenol [73], etc. from the atmosphere. 6.2. Applications in Energy 6.2.1. Photocatalytic Hydrogen Generation With the extensive use of nonrenewable fossil fuels, mankind is facing an unprecedented energy crisis. The photogenerated electrons on TiO2 have strong reducing capacity, enabling hydrogen production from the photocatalytic splitting of water. Moreover, hydrogen combustion produces only water and no harmful emissions, and therefore its potential as a truly clean energy source has received considerable attention since it was discovered. Zou et al. reported a self-modiﬁed TiO2 material with paramagnetic oxygen vacancies [132]. For the synthesis of Vo-TiO2 (Vo: denotes a paramagnetic oxygen vacancy), they chose a porous amorphous TiO2 material as a precursor that possessed a high surface area of 543 m2 g−1. The precursor was calcined in the presence of imidazole and hydrochloric acid at an elevated temperature in air to obtain the Vo-TiO2 material [132]. The Vo-TiO2 sample (for H2 evolution from water) used methanol as a sacriﬁcial reagent under visible light (≥400 nm) at room temperature, and the H2 production rate was approximately 115 μmol h−1 g−1, which is substantially higher than that achieved with Vo-Ti3+-TiO2 (32 μmol h−1 g−1). Zhou et al. introduced an ordered mesoporous black TiO2 material that utilized a thermally stable and high surface area mesoporous TiO2 as the hydrogenation precursor for treatment at 500 ◦C [133]. The samples possessed a relatively high surface area of 124 m2 g−1 and exhibited a photo response that extended from ultraviolet to visible light. As shown in Figure 22, the ordered mesoporous black TiO2 material exhibits a high solar-driven hydrogen production rate (136.2 μmol h−1), which is almost twice as high as that of pristine mesoporous TiO2 (76.6 μmol h−1). Zhong et al. constructed a covalently bonded oxidized graphitic C3N4/TiO2 heterostructure that markedly increased the visible light photocatalytic activity for H2 evolution by nearly a factor of approximately 6.1 compared to a simple physical mixture of TiO2 nanosheets and O-g-C3N4 [134]. 6.2.2. Photocatalytic CO2 Reduction into Energy Fuels In addition to reducing water to hydrogen, the photogenerated electrons on TiO2 are capable of generating valuable solar energy fuels, such as CH4, HCO2H, CH2O, CH3OH, and CO2, which are considered highly viable energy sources that can alleviate the problems associated with the production of greenhouse gases from the combustion of fossil fuels. Slamet et al. prepared Cu-doped TiO2 through an improved impregnation method for photocatalytic CO2 reduction [135]. Both the distribution of copper on the catalyst surface and the grain size of copper–titania catalysts (crystallite size of approximately 23 nm) were uniform, and it was determined that Cu doping can greatly enhance the photocatalytic performance of TiO2 with respect to CO2 reduction. Liu et al. found that 165 Catalysts 2019, 9, 191 copper-loaded titania photocatalysts, prepared via a one-pot, sol–gel synthesis method, comprised highly dispersed copper and that CO2 photoreduction exhibited a strong volcano dependence on Cu loading, which reﬂected the transition from 2-dimensional CuOx nanostructures to 3-dimensional crystallites; optimum CH4 production was observed for 0.03 wt.% Cu/TiO2 [136]. Figure 22. Photocatalytic hydrogen evolution of ordered mesoporous black TiO2 (a) and pristine ordered mesoporous TiO2 materials (b). (A) Cycling tests of photocatalytic hydrogen generation under AM 1.5 and visible light irradiation. (B) The photocatalytic hydrogen evolution rates under single-wavelength light and the corresponding QE. The inset enlarges the QE of single-wavelength light at 420 and 520 nm [133]. Copyright 2014 American Chemical Society. Figure 22. Photocatalytic hydrogen evolution of ordered mesoporous black TiO2 (a) and pristine ordered mesoporous TiO2 materials (b). (A) Cycling tests of photocatalytic hydrogen generation under AM 1.5 and visible light irradiation. (B) The photocatalytic hydrogen evolution rates under single-wavelength light and the corresponding QE. The inset enlarges the QE of single-wavelength light at 420 and 520 nm [133]. Copyright 2014 American Chemical Society. 6.2.3. Solar Batteries Since semiconductors absorb photons to produce photonic carriers and the photonic carriers move and separate at the same time, electric energy can be obtained through charge transport. TiO2 can also be applied to dye-sensitized solar cells, Li-ion batteries, Na-ion batteries, and supercapacitors. Liu et al. synthesized a spring-like Ti@TiO2 nanowire array wire that could be used as a photoanode in dye-sensitized solar cells; this conﬁguration exhibited a conversion efﬁciency maintenance rate of more than 95.95% [137]. Another study reported the use of anatase TiO2 nanotubes on rutile TiO2 nanorod arrays as photoanodes in quantum dot-sensitized solar cells, which have a small thickness of 1 μm and an excellent solar energy conversion efﬁciency of approximately 1.04%; this is almost 2.7 times higher than the conversion efﬁciencies measured for solar cells using the original TiO2 nanorod array photoanodes, as shown in Figure 23 [138]. Chen et al. implemented a C@TiO2 nanocomposite as the anode material for lithium-ion batteries, which utilize the esteriﬁcation of ethylene glycol with acetic acid in the presence of potassium chloride. Li-ion batteries utilizing the C@TiO2 nanocomposite anode exhibited excellent rate performance and speciﬁc capacity (237 mA h−1 g−1), and a coulomb efﬁciency (CE) of approximately 100% after 100 cycles [139]. Su et al. synthesized anatase TiO2 via a template approach for use as the anode in Na-ion batteries; use of the template-synthesized TiO2 resulted in better battery performance in comparison to that achieved when amorphous and rutile TiO2 was used as the anode material. Compared to other crystalline phases of titanium dioxide, anatase titanium dioxide produced the highest capacity, 295 mA h−1 g−1, in the second cycle, tested at a current density of 20 mA g−1 [140]. Kim et al. developed a black-colored TiO2 nanotube array synthesized by electrochemical self-doping of an amorphous TiO2 nanotube array and N2 annealing; the material exhibited good stability, high capacitance, and electrocatalytic performance, and is an excellent material for supercapacitors and oxide anodes [141]. 6.2.4. Supercapacitors (d) Schematic conﬁguration for our device showing the interfacial charge transfer and recombination processes [138]. Copyright 2015 The Royal Society of Chemistry. Oth A li ti 6.3. Other Applications 6.3. Other Applications 6.3.1. Antibacterial and Wound Healing It is generally believed that electron–hole pairs formed under light illumination, such as •O2− and •OH, not only destroy all chemical contaminants but also kill microorganisms. Liu et al. proposed a TiO2/Ag2O heterostructure (produced by a facile in situ precipitation route) to enhance antibacterial activities [144]. Yu et al. synthesized a TiO2/BTO/Au heterostructured nanorod arrays (exhibiting piezophototronic and plasmonic effects) by using a simple process that combined hydrothermal and PVD methods. This material can be used as an antibacterial coating for efﬁcient light driven in vitro/in vivo sterilization and wound healing [145]. 6.2.4. Supercapacitors Yang et al. developed a hybrid material, covalently coupled ultraﬁne H–TiO2 nanocrystals/ nitrogen-doped graphene, via the hydrothermal route [142]. Due to the strong interaction between H–TiO2 nanocrystals and NG plates, the high structural stability of the H–TiO2 nanocrystal aggregation is inhibited. At the same time, the NG matrix plays the role of electron conductor and mechanical skeleton, imparting good stability and electrochemical activity on most of the well-dispersed ultraﬁne 166 Catalysts 2019, 9, 191 H-TiO2 nanocrystals [142]. The material exhibited a high reversible speciﬁc capacity of 385.2 F g−1 at 1 A g−1 and excellent cycling stability with 98.8% capacity retention. Parthiban et al. reported a blue titanium oxide (B-TiO2) nanostructure that was applied via a one-pot hydrothermal route and hydrothermal oxidation [143]. The B–TiO2 nanostructure indicated excellent cycling stability with approximately 90.2% capacitance retention after 10,000 charge–discharge cycles. Figure 23. (a) Electron lifetime as a function of Voc for TiO2 NRA and H–TiO2 NRA electrodes with various reaction times. (b) Recombination resistance (Rrec) of the QDSCs made from TiO2 NRAs and H-TiO2 NRAs at various forward biases in the dark. (c) Transient photovoltage responses of CdS–TiO2 NRAs and CdS–H-TiO2 NRAs. The wavelength of the laser pulse was 532 nm. Inset: schematic setup of TPV measurements. (d) Schematic conﬁguration for our device showing the interfacial charge transfer and recombination processes [138]. Copyright 2015 The Royal Society of Chemistry. 3 Other Applications Figure 23. (a) Electron lifetime as a function of Voc for TiO2 NRA and H–TiO2 NRA electrodes with various reaction times. (b) Recombination resistance (Rrec) of the QDSCs made from TiO2 NRAs and H-TiO2 NRAs at various forward biases in the dark. (c) Transient photovoltage responses of CdS–TiO2 NRAs and CdS–H-TiO2 NRAs. The wavelength of the laser pulse was 532 nm. Inset: schematic setup of TPV measurements. (d) Schematic conﬁguration for our device showing the interfacial charge transfer and recombination processes [138]. Copyright 2015 The Royal Society of Chemistry. Figure 23. (a) Electron lifetime as a function of Voc for TiO2 NRA and H–TiO2 NRA electrodes with various reaction times. (b) Recombination resistance (Rrec) of the QDSCs made from TiO2 NRAs and H-TiO2 NRAs at various forward biases in the dark. (c) Transient photovoltage responses of CdS–TiO2 NRAs and CdS–H-TiO2 NRAs. The wavelength of the laser pulse was 532 nm. Inset: schematic setup of TPV measurements. 6.3.2. Drug Delivery Carriers TiO2 has the advantages of nontoxicity, stability, biocompatibility, and natural abundance. The preparation of TiO2 with a high speciﬁc surface area can be advantageous in drug delivery carrier applications. Johan et al. controlled the kinetics of drug delivery from mesoporous titania thin ﬁlms via surface energy and pore size control [146]. Different pore sizes ranging from 3.4 nm 167 Catalysts 2019, 9, 191 to 7.2 nm were achieved by the use of different structural guiding templates and expansive agents. In addition, by attaching dimethyl silane to the pore wall, the surface energy of the pore wall could be altered. The results indicated that the pore size and surface energy had signiﬁcant effects on the adsorption and release kinetics of alendronate [146]. Biki et al. designed silica-supported mesoporous titania nanoparticles (MTN) coated with hyaluronic acid to cure breast cancer by effectively delivering doxorubicin (DOX) to the cancer cells [147]. Guo et al. deposited (onto the surface of MTN) hyaluronic acid and cyclic pentapeptide (ADH-1), which target CD44-overexpressing tumor cells and selectively inhibit the function of N-cadherin, respectively, to overcome the drug resistance of tumors [148]. to 7.2 nm were achieved by the use of different structural guiding templates and expansive agents. In addition, by attaching dimethyl silane to the pore wall, the surface energy of the pore wall could be altered. The results indicated that the pore size and surface energy had signiﬁcant effects on the adsorption and release kinetics of alendronate [146]. Biki et al. designed silica-supported mesoporous titania nanoparticles (MTN) coated with hyaluronic acid to cure breast cancer by effectively delivering doxorubicin (DOX) to the cancer cells [147]. Guo et al. deposited (onto the surface of MTN) hyaluronic acid and cyclic pentapeptide (ADH-1), which target CD44-overexpressing tumor cells and selectively inhibit the function of N-cadherin, respectively, to overcome the drug resistance of tumors [148]. y g Recently, Nakayama et al. found that H2O2-treated TiO2 can enhance the ability to produce reactive oxygen species (ROS) in response to X-ray irradiation [149]. As shown in Figure 24, the atomic packing factor (APF) intensity indicated that hydroxyl radical production in the TiOx (H2O2-treated TiO2) nanoparticles increased in a radiation dose-dependent manner in comparison to that of the non-H2O2-treated TiO2 nanoparticles. This behavior allows H2O2-treated TiO2 nanoparticles to act as potential agents for enhancing the effects of radiation in the treatment of pancreatic cancer. Dai et al. 6.3.2. Drug Delivery Carriers designed and synthesized a novel nanodrug delivery system for the synergistic treatment of lung cancer [150]. They loaded DOX onto H2O2-treated TiO2 nanosheets. In this way, chemotherapy and radiotherapy were combined effectively for the synergistic therapy of cancers. Figure 24. ROS production by the TiOxNPs, PAA-TiOxNPs, and TiO2 NPs under X-ray irradiation. (A) Atomic packing factor (APF) intensity indicating that hydroxyl radical production in the TiOxNPs and the PAA-TiOxNPs increased in a radiation dose-dependent manner, but that of the TiO2 NPs did not. Irradiated radiation doses were 0, 5, 10, and 30 Gy. Data are shown as the mean ± SD from 5 independent experiments. (B) Production and scavenging of ROS by 1 mM vitamin C (Vit. C) or 1 mM glutathione (GSH). Histograms show the mean ± SD calculated from 5 independent experiments. (C) Hydrogen peroxide production from the TiOxNPs under X-ray irradiation [149]. Copyright 2016 Springer Nature Switzerland AG. Figure 24. ROS production by the TiOxNPs, PAA-TiOxNPs, and TiO2 NPs under X-ray irradiation. (A) Atomic packing factor (APF) intensity indicating that hydroxyl radical production in the TiOxNPs and the PAA-TiOxNPs increased in a radiation dose-dependent manner, but that of the TiO2 NPs did not. Irradiated radiation doses were 0, 5, 10, and 30 Gy. Data are shown as the mean ± SD from 5 independent experiments. (B) Production and scavenging of ROS by 1 mM vitamin C (Vit. C) or 1 mM glutathione (GSH). Histograms show the mean ± SD calculated from 5 independent experiments. (C) Hydrogen peroxide production from the TiOxNPs under X-ray irradiation [149]. Copyright 2016 Springer Nature Switzerland AG. Figure 24. ROS production by the TiOxNPs, PAA-TiOxNPs, and TiO2 NPs under X-ray irradiation. 7. Conclusions As discussed in this review article, TiO2-based nanomaterials with wide band gaps have advantages associated with natural geologic abundance, nontoxicity and stability but they also exhibit inherent deﬁciencies and limitations related to ineffective visible light responses and other 168 Catalysts 2019, 9, 191 photocatalytic properties. The present review aimed to summarize key studies related to the marked enhancement of the photocatalytic performance of TiO2 by analyzing its electrical structure and photocatalytic reaction process. We have highlighted TiO2 photocatalysts with well-deﬁned electrical and structure design, as well as tailored facets, dimensions, and remarkable morphologies, which are promising with respect to enhancing the photocatalytic properties of TiO2. All works presented in this review has enabled the authors to obtain an in-depth understanding of the TiO2 photocatalytic process, and the critical design of TiO2 nanostructures with enhanced light absorption, high surface area, desired photostability, and charge carrier dynamics. We hope that this review will guide the future development of more robust TiO2-based photocatalysts for large-scale applications. Finally, photocatalysis technology is one of the most active research ﬁelds in the world in recent years. However, photocatalysis technologies based on TiO2 semiconductor still suffer from several key scientiﬁc and technological problems, such as low solar energy utilization rate, inferior quantum yield, and difﬁcult recovery, which greatly restricts its wide application in industry. The fundamental solution to improve solar energy absorption is energy band engineering, designing and regulating the bandgap to optimize the harvesting of incident photons. Narrow bandgap and direct semiconductor are more likely to make use of low energy light, but they are restricted by very high electron and hole recombination rate and the incompatible band-edge position. High quantum yield is inevitable for an idea photocatalysis in practical solar engineering, but it cannot be achieved simply doping or inducing intrinsic defects. More works are needed to do to search high quantum yield. All of the above problems depend on the deepening of basic research. Although at present, photocatalysis technology is still a long way from large-scale production and application, its huge potential excellent performance provides a good way for our development. In the near future, with the breakthrough of these key issues, the practical application of nano-photocatalytic materials will certainly be realized to improve our environment, provide cleaner energy, and bring more convenience to our daily life. Author Contributions: X.K. and S.L. 7. Conclusions collected references, prepared ﬁgures, and wrote the original draft of the manuscript, they contributed equally to this work; Z.D. and Y.H. collected references and analyzed the data; X.S. gave valuable advice; Z.T. acted as a project director and contributed to subsequent revisions. All authors agreed to the ﬁnal version of the paper. Funding: This research was funded by the National Natural Science Foundation of China grant number 21571028, 21601027, the Fundamental Research Funds for the Central Universities grant number DUT16TD19, DUT17LK33, DUT18LK28 and the Education Department of the Liaoning Province of China grant number LT2015007. Conﬂicts of Interest: The authors declare no conﬂicts of interest. References A multi-faceted concept for green chemistry. Chem. Soc. Rev. 2009, 38, 1999–2011. [CrossRef] [PubMed] 13. Bai, S.; Jiang, J.; Zhang, Q.; Xiong, Y. Steering charge kinetics in photocatalysis: Intersection of materials syntheses, characterization techniques and theoretical simulations. Chem. Soc. Rev. 2015, 44, 2893–2939. [CrossRef] [PubMed] 14. Gao, M.; Zhu, L.; Ong, W.L.; Wang, J.; Ho, G.W. Structural design of TiO2-based photocatalyst for H2 production and degradation applications. Catal. Sci. Technol. 2015, 5, 4703–4726. [CrossRef] Š Š 15. Šuligoj, A.; Arˇcon, I.; Mazaj, M.; Draži´c, G.; Arˇcon, D.; Cool, P.; Štangar, U.L.; Tušar, N.N. Surface modiﬁed titanium dioxide using transition metals: Nickel as a winning transition metal for solar light photocatalysis. J. Mater. Chem. A 2018, 6, 9882. [CrossRef] 16. Hou, Y.; Liu, S.; Zhang, J.; Cheng, X.; Wang, Y. Facile hydrothermal synthesis of TiO2-Bi2WO6 hollow superstructures with excellent photocatalysis and recycle properties. Dalton Trans. 2014, 43, 1025–1031. [CrossRef] [PubMed] 17. Ansari, S.A.; Khan, M.M.; Ansari, M.O.; Cho, M.H. Nitrogen-doped titanium dioxide (N-doped TiO2) for visible light photocatalysis. New J. Chem. 2016, 40, 3000–3009. [CrossRef] 18. Asahi, R.; Taga, Y.; Mannstadt, W.; Freeman, A.J. Electronic and optical properties of anatase TiO2. Phys. Rev. B 2000, 61, 7459–7465. [CrossRef] 19. Park, H.; Choi, W. Effects of TiO2 surface ﬂuorination on photocatalytic reactions and photoelectrochemical behaviors. J. Phys. Chem. B 2004, 108, 4086–4093. [CrossRef] 20. Xu, J.; Ao, Y.; Fu, D.; Yuan, C. Low-temperature preparation of F-doped TiO2 ﬁlm and its photocatalytic activity under solar light. Appl. Surf. Sci. 2008, 254, 3033–3038. [CrossRef] 21. Niu, M.; Cui, R.; Wu, H.; Cheng, D.; Cao, D. Enhancement mechanism of the conversion effﬁciency of dye-sensitized solar cells based on nitrogen-, ﬂuorine-, and iodine-doped TiO2 photoanodes. J. Phys. Chem. C 2015, 119, 13425–13432. [CrossRef] 22. Liu, H.; Li, Y.; Yang, Y.; Mao, M.; Zeng, M.; Lan, L.; Yun, L.; Zhao, X. Highly efﬁcient UV–Vis-infrared catalytic puriﬁcation of benzene on CeMnxOy/TiO2 nanocomposite, caused by its high thermocatalytic activity and strong absorption in the full solar spectrum region. J. Mater. Chem. A 2016, 4, 9890–9899. [CrossRef] 23. Narayan, H.; Alemu, H.; Macheli, L.; Thakurdesai, M.; Rao, T.K. Synthesis and characterization of Y3+-doped TiO2 nanocomposites for photocatalytic applications. Nanotechnology 2009, 20, 255601. [CrossRef] [PubMed] 23. Narayan, H.; Alemu, H.; Macheli, L.; Thakurdesai, M.; Rao, T.K. Synthesis and characterization of Y3+-doped f h l l h l [ f] [ b d] 24. References 1. Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238, 37. [CrossRef] [PubMed] Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 238, 37. [CrossRef] [PubMed] 2. Long, L.; Zhang, A.; Yang, J.; Zhang, X.; Yu, H. A green approach for preparing doped TiO2 single crystals. ACS Appl. Mater. Interfaces 2014, 6, 16712–16720. [CrossRef] 3. Nakata, K.; Fujishima, A. TiO2 photocatalysis: Design and applications. J. Photochem. Photobio. C 2012, 13, 169–189. [CrossRef] 3. Nakata, K.; Fujishima, A. TiO2 photocatalysis: Design and applications. J. Photochem. Photobio. C 2012, 13, 169–189. [CrossRef] 4. Tong, H.; Ouyang, S.; Bi, Y.; Umezawa, N.; Oshikiri, M.; Ye, J. Nano-photocatalytic materials: Possibilities and challenges. Adv. Mater. 2012, 24, 229–251. [CrossRef] 4. Tong, H.; Ouyang, S.; Bi, Y.; Umezawa, N.; Oshikiri, M.; Ye, J. Nano-photocatalytic materials: Possibilities and challenges. Adv. Mater. 2012, 24, 229–251. [CrossRef] 5. Grätzel, M. Photoelectrochemical cells. Nature 2001, 414, 338. [CrossRef] [PubMed] 5. Grätzel, M. Photoelectrochemical cells. Nature 2001, 414, 338. [CrossRef] [PubMed] 5. Grätzel, M. Photoelectrochemical cells. Nature 2001, 414, 33 6. Ma, Y.; Wang, X.; Jia, Y.; Chen, X.; Han, H.; Li, C. Titanium dioxide-based n fuel generations. Chem. Rev. 2014, 114, 9987–10043. [CrossRef] [PubMed] 7. Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D Understanding TiO2 photocatalysis: Mechanisms and materials. Chem. Rev. 2014, 114, 9919–9986. [CrossR 8. Wang, M.; Iocozzia, J.; Sun, L.; Lin, C.; Lin, Z. Inorganic-modiﬁed semiconductor TiO2 nanotube arrays for photocatalysis. Energ. Environ. Sci. 2014, 7, 2182–2202. [CrossRef] 8. Wang, M.; Iocozzia, J.; Sun, L.; Lin, C.; Lin, Z. Inorganic-modiﬁed semiconductor TiO2 nanotube arrays for photocatalysis. Energ. Environ. Sci. 2014, 7, 2182–2202. [CrossRef] 9. Kalyanasundaram, K. Photochemical applications of solar energy: Photocatalysis and photodecomposition of water. Photochemistry 2013, 41, 182–265. 9. Kalyanasundaram, K. Photochemical applications of solar energy: Photocatalysis and photodecomposition of water. Photochemistry 2013, 41, 182–265. 169 Catalysts 2019, 9, 191 10. Liu, B.; Yang, J.; Zhao, X.; Yu, J. The role of electron interfacial transfer in mesoporous nano-TiO2 photocatalysis: A combined study of in situ photoconductivity and numerical kinetic simulation. Phys. Chem. Chem. Phys. 2017, 19, 8866–8873. [CrossRef] 11. Liu, B.; Zhao, X.; Terashima, C.; Fujishima, A.; Nakata, K. Thermodynamic and kinetic analysis of heterogeneous photocatalysis for semiconductor systems. Phys. Chem. Chem. Phys. 2014, 16, 8751–8760. [CrossRef] [PubMed] 12. Ravelli, D.; Dondi, D.; Fagnoni, M.; Albini, A. Photocatalysis. References Wang, X.; Yuan, X.; Wang, D.; Dong, W.; Dong, C.; Zhang, Y.; Lin, T.; Huang, F. Tunable synthesis of colorful nitrogen-doped titanium oxide and its application in energy storage. ACS Appl. Energy Mater. 2018, 1, 876–882. [CrossRef] 25. Niu, P.; Wu, T.; Wen, L.; Tan, J.; Yang, Y.; Zheng, S.; Liang, Y.; Li, F.; Irvine, J.T.S.; Liu, G.; et al. Substitutional carbon-modiﬁed anatase TiO2 decahedral plates directly derived from titanium oxalate crystals via topotactic transition. Adv. Mater. 2018, 30, e1705999. [CrossRef] [PubMed] 26. Zhao, X.; Liu, X.; Yu, M.; Wang, C.; Li, J. The highly efﬁcient and stable Cu, Co, Zn-porphyrineTiO2 photocatalysts with heterojunction by using fashioned one-step method. Dyes Pigments 2017, 136, 648–656. [CrossRef] 27. Hachiya, A.; Takata, S.; Komuro, Y.; Matsumoto, Y. Effects of V-ion doping on the photoelectrochemical properties of epitaxial TiO2(110) thin ﬁlms on Nb-doped TiO2 (110) single crystals. J. Phys. Chem. C 2012, 116, 16951–16956. [CrossRef] 28. Klosek, S.; Raftery, D. Visible light driven V-doped TiO2 photocatalyst and its photooxidation of ethanol. J. Phys. Chem. B 2001, 105, 2815–2819. [CrossRef] 29. Zahid, M.; Papadopoulou, E.L.; Suarato, G.; Binas, V.D.; Kiriakidis, G.; Gounaki, I.; Moira, O.; Venieri, D.; Bayer, I.S.; Athanassiou, A. Fabrication of visible light-induced antibacterial and self-cleaning cotton fabrics using manganese doped TiO2 nanoparticles. ACS Appl. Bio Mater. 2018, 1, 1154–1164. [CrossRef] 170 Catalysts 2019, 9, 191 30. Taguchi, T.; Ni, L.; Irie, H. Alkaline-resistant titanium dioxide thin ﬁlm displaying visible-light-induced superhydrophilicity initiated by interfacial electron transfer. Langmuir 2013, 29, 4908–4914. [CrossRef] p y p y y g 31. Mizushima, K.; Tanaka, M.; Asai, A.; Iida, S.; Goodenough, B. Impurity levels of iron-group ions in TiO2(II). J. Phys. Chem. Solids 1979, 40, 1129–1140. [CrossRef] 32. Borgarello, E.; Kiwi, J.; Grátzel, M.; Pelizzetti, E.; Viscald, M. Visible light induced water cleavage in colloidal solutions of chromium-doped titanium dioxide particles. J. Am. Chem. Soc. 1982, 104, 2996–3002. [CrossRef] 33. Xu, D.; Feng, L.; Lei, A. Characterizations of lanthanum trivalent ions/TiO2 nanopowders catalysis prepared by plasma spray. J. Colloid Interface Sci. 2009, 329, 395–403. [CrossRef] [PubMed] 34. Li, N.; Zhou, X.; Liu, M.; Wei, L.; Shen, Q.; Bibi, R.; Xu, C.; Ma, Q.; Zhou, J. Enhanced visible light photocatalytic hydrogenation of CO2 into methane over a Pd/Ce-TiO2 nanocomposition. J. Phys. Chem. C 2017, 121, 25795–25804. [CrossRef] 35. Anandan, S.; Ikuma, Y.; Murugesan, V. Highly active rare-earth-metal La-doped photocatalysts: Fabrication, characterization, and their photocatalytic activity. Int. J. References Photoenergy 2012, 10, 921412. [CrossRef] 36. Sun, L.; Zhao, X.; Cheng, X.; Sun, H.; Li, Y.; Li, P.; Fan, W. Synergistic effects in La/N codoped TiO2 anatase (101) surface correlated with enhanced visible-light photocatalytic activity. Langmuir 2012, 28, 5882–5891. [CrossRef] [PubMed] 37. Choi, J.; Park, H.; Hoffman, M.R. Effects of single metal-ion doping on the visible-light photoreactivity TiO2. J. Phys. Chem. C 2010, 114, 783–792. [CrossRef] 38. Di Paola, A.; Ikeda, S.; Marcì, G.; Ohtani, B.; Palmisano, L. Transition metal doped and photocatalytic behaviour. Int. J. Photoenergy 2001, 3, 171. [CrossRef] 39. Asahi, R.; Morikawa, T.; Irie, H.; Ohwaki, T. Nitrogen-doped titanium dioxide as visible-light-sensitive photocatalyst: Designs, developments, and prospects. Chem. Rev. 2014, 114, 9824–9852. [CrossRef] 40. Muhich, C.L.; Westcott, J.Y.; Fuerst, T.; Weimer, A.W.; Musgrave, C.B. Increasing the photocatalytic activity of anatase TiO2 through B, C, and N doping. J. Phys. Chem. C 2014, 118, 27415–27427. [CrossRef] 41. Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visible-light photocatalysis in nitrogen-doped titanium oxides. Science 2001, 13, 269–271. [CrossRef] 42. Zhao, Z.; Fan, J.; Wang, J.; Li, R. Effect of heating temperature on photocatalytic reduction of CO2 by N-TiO2 nanotube catalyst. Catal. Commun. 2012, 21, 32–37. [CrossRef] 43. Irie, H.; Watanabe, Y.; Hashimoto, K. Nitrogen-concentration dependence on photocatalytic activity TiO2-xNx powders. J. Phys. Chem. B 2003, 107, 5483–5486. [CrossRef] 44. Li, H.; Zhang, X.; Huo, Y.; Zhu, J. Supercritical preparation of a highly active S-doped TiO2 photocatalyst for methylene blue mineralization. Environ. Sci. Technol. 2007, 41, 4410–4414. [CrossRef] [PubMed] 45. Liu, G.; Zhao, Y.; Sun, C.; Li, F.; Lu, G.Q.; Cheng, H.M. Synergistic effects of B/N doping on the visible-light photocatalytic activity of mesoporous TiO2. Angew. Chem. Int. Ed. 2008, 47, 4516–4520. [CrossRef] [PubMed] 45. Liu, G.; Zhao, Y.; Sun, C.; Li, F.; Lu, G.Q.; Cheng, H.M. Synergistic effects of B/N doping on the visible light photocatalytic activity of mesoporous TiO2. Angew. Chem. Int. Ed. 2008, 47, 4516–4520. [CrossRef] [PubMed] 46. Peighambardoust, N.S.; Asl, S.K.; Mohammadpour, R.; Asl, S.K. Band-gap narrowing and electrochemical properties in N-doped and reduced anodic TiO2 nanotube arrays. Electrochim. Acta 2018, 270, 245–255. [CrossRef] 47. Lynch, J.; Giannini, C.; Cooper, J.K.; Loiudice, A.; Sharp, I.D.; Buonsanti, R. Substitutional or interstitial site-selective nitrogen doping in TiO2 nanostructures. J. Phys. Chem. C 2015, 119, 7443–7452. [CrossRef] 48. Di Valentin, C.; Finazzi, E.; Pacchioni, G.; Selloni, A.; Livraghi, S.; Paganini, M.C.; Giamello, E. N-doped TiO2: Theory and experiment. References Wang, X.; Fu, R.; Yin, Q.; Wu, H.; Guo, X.; Xu, R.; Zhong, Q. Black TiO2 synthesized via magnesiothermic reduction for enhanced photocatalytic activity. J. Nanopart. Res. 2018, 20, 89. [CrossRef] 61. Dong, J.; Han, J.; Liu, Y.; Nakajima, A.; Matsushita, S.; Wei, S.; Gao, W. Defective black TiO2 synthesized via anodization for visible-light photocatalysis. ACS Appl. Mater. Interfaces 2014, 6, 1385–1388. [CrossRef] [PubMed] 62. Liu, X.; Gao, S.; Xu, H.; Lou, Z.; Wang, W.; Huang, B.; Dai, Y. Green synthetic approach for Ti3+ self-doped TiO2-x nanoparticles with efﬁcient visible light photocatalytic activity. Nanoscale 2013, 5, 1870–1875. [CrossRef] [PubMed] 63. Chen, S.; Wang, Y.; Li, J.; Hu, Z.; Zhao, H.; Xie, W.; Wei, Z. Synthesis of black TiO2 with efﬁcient visible-light photocatalytic activity by ultraviolet light irradiation and low temperature annealing. Mater. Res. Bull. 2018, 98, 280–287. [CrossRef] 64. Liu, L.; Yu, P.Y.; Chen, X.; Mao, S.S.; Shen, D.Z. Hydrogenation and disorder in engineered black TiO2. Phys. Rev. Lett. 2013, 111, 065505. [CrossRef] [PubMed] 65. Fujishima, A.; Zhang, X.; Tryk, D.A. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2008, 63, 515–582. [CrossRef] 66. Pan, X.; Yang, M.Q.; Fu, X.; Zhang, N.; Xu, Y.J. Defective TiO2 with oxygen vacancies: Synth and photocatalytic applications. Nanoscale 2013, 5, 3601–3614. [CrossRef] [PubMed] 67. Tan, H.; Zhao, Z.; Niu, M.; Mao, C.; Cao, D.; Cheng, D.; Feng, P.; Sun, Z. A facile and versatile method for preparation of colored TiO2 with enhanced solar-driven photocatalytic activity. Nanoscale 2014, 6, 10216–10223. [CrossRef] 68. Li, S.; Qiu, J.; Ling, M.; Peng, F.; Wood, B.; Zhang, S. Photoelectrochemical characterization of hydrogenated TiO2 nanotubes as photoanodes for sensing applications. ACS Appl. Mater. Interfaces 2013, 5, 11129–11135. [CrossRef] 69. Wang, Z.; Yang, C.; Lin, T.; Yin, H.; Chen, P.; Wan, D.; Xu, F.; Huang, F.; Lin, J.; Xie, X.; Jiang, M. H-doped black titania with very high solar absorption and excellent photocatalysis enhanced by localized surface plasmon resonance. Adv. Funct. Mater. 2013, 23, 5444–5450. [CrossRef] 70. Sinhamahapatra, A.; Jeon, J.-P.; Yu, J.-S. A new approach to prepare highly active and stable black titania visible light-assisted hydrogen production. Energ. Environ. Sci. 2015, 8, 3539–3544. [CrossRef] 71. Zhou, L.; Boyle, D.S.; O’Brien, P. A facile synthesis of uniform NH4TiOF3 mesocrystals and their conversion to TiO2 mesocrystals. J. Am. Chem. Soc. 2008, 130, 1309–1320. [CrossRef] 72. Kang, X.; Han, Y.; Song, X.; Tan, Z. References Chem. Phys. 2007, 339, 44–56. [CrossRef] 49. Wu, Y.; Lazic, P.; Hautier, G.; Perssonb, K.; Ceder, G. First principles high throughput screening of oxynitrides for water-splitting photocatalysts. Energy Environ. Sci. 2013, 6, 157–168. [CrossRef] 50. Wang, J.; Tafen, D.N.; Lewis, J.P.; Hong, Z.; Manivannan, A.; Zhi, M.; Li, M.; Wu, N. Origin of photocatalytic activity of nitrogen-doped TiO2 nanobelts. J. Am. Chem. Soc. 2009, 131, 12290–12297. [CrossRef] 51. Zhang, P.; Tachikawa, T.; Fujitsuka, M.; Majima, T. In situ ﬂuorine doping of TiO2 superstructures for efﬁcient visible-light driven hydrogen generation. ChemSusChem 2016, 9, 617–623. [CrossRef] [PubMed] g j j p g p visible-light driven hydrogen generation. ChemSusChem 2016, 9, 617–623. [CrossRef] [PubMed] 52. Seo, H.; Baker, L.R.; Hervier, A.; Kim, J.; Whitten, J.L.; Somorjai, G.A. Generation of highly n-type oxide using plasma ﬂuorine insertion. Nano Lett. 2011, 11, 751–756. [CrossRef] [PubMed] 53. Yu, W.; Liu, X.; Pana, L.; Li, J.; Liu, J.; Zhang, J.; Li, P.; Chen, C.; Sun, Z. Enhanced visible light photocata degradation of methylene blue by F-doped TiO2. Appl. Surf. Sci. 2014, 319, 107–112. [CrossRef] 54. Chen, X.; Liu, L.; Yu, P.Y.; Mao, S.S. Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals. Science 2011, 331, 746–749. [CrossRef] [PubMed] 171 Catalysts 2019, 9, 191 55. Song, H.; Li, C.; Lou, Z.; Ye, Z.; Zhu, L. Effective formation of oxygen vacancies in black TiO2 nanostructures with efﬁcient solar-driven water splitting. ACS Sustain. Chem. Eng. 2017, 5, 8982–8987. [CrossRef] p g g 56. Hu, Y.H. A highly efﬁcient photocatalyst–hydrogenated black TiO2 for the photocatalytic splitting of water. Angew. Chem. Int. Ed. 2012, 51, 12410–12412. [CrossRef] [PubMed] 56. Hu, Y.H. A highly efﬁcient photocatalyst–hydrogenated black TiO2 fo Angew. Chem. Int. Ed. 2012, 51, 12410–12412. [CrossRef] [PubMed] 57. Lin, L.; Huang, J.; Li, X.; Abass, M.A.; Zhang, S. Effective surface disorder engineering of metal oxide nanocrystals for improved photocatalysis. Appl. Catal. B Environ. 2017, 203, 615–624. [CrossRef] 58. Liu, N.; Haublein, V.; Zhou, X.; Venkatesan, U.; Hartmann, M.; Mackovic, M.; Nakajima, T.; Spiecker, E.; Osvet, A.; Frey, L.; Schmuki, P. “Black” TiO2 nanotubes formed by high-energy proton implantation show noble-metal-co-catalyst free photocatalytic H2-evolution. Nano Lett. 2015, 15, 6815–6820. [CrossRef] 59. Wang, G.; Wang, H.; Ling, Y.; Tang, Y.; Yang, X.; Fitzmorris, R.C.; Wang, C.; Zhang, J.Z.; Li, Y. Hydrogen-treated TiO2 nanowire arrays for photoelectrochemical water splitting. Nano Lett. 2011, 11, 3026–3033. [CrossRef] 60. References Development of photocatalytic self-cleaning glasses utilizing metal ox nanosheets. J. Surf. Finish. Soc. Jpn 2010, 61, 30–35. [CrossRef] 85. Ong, W.-J.; Tan, L.-L.; Chai, S.-P.; Yong, S.-T.; Mohamed, A.R. Highly reactive {001} facets of TiO2-based composites: Synthesis, formation mechanism and characterization. Nanoscale 2014, 6, 1946–2008. [CrossRef] [PubMed] g g g y composites: Synthesis, formation mechanism and characterization. Nanoscale 2014, 6, 1946–2008. [CrossRef] [PubMed] 86. Sun, L.; Zhao, Z.; Zhou, Y.; Liu, L. Anatase TiO2 nanocrystals with exposed {001} facets on graphene sheets via molecular grafting for enhanced photocatalytic activity. Nanoscale 2012, 4, 613–620. [CrossRef] [PubMed] 87. Yang, H.G.; Sun, C.H.; Qiao, S.Z.; Zou, J.; Liu, G.; Smith, S.C.; Cheng, H.M.; Lu, G.Q. Anatase TiO2 single t l ith l t f ti f t Nature 2008 453 638 641 [C R f] 86. Sun, L.; Zhao, Z.; Zhou, Y.; Liu, L. Anatase TiO2 nanocrystals with exposed {001} facets on graphene sheets via molecular grafting for enhanced photocatalytic activity. Nanoscale 2012, 4, 613–620. [CrossRef] [PubMed] 87. Yang, H.G.; Sun, C.H.; Qiao, S.Z.; Zou, J.; Liu, G.; Smith, S.C.; Cheng, H.M.; Lu, G.Q. Anatase TiO2 single crystals with a large percentage of reactive facets. Nature 2008, 453, 638–641. [CrossRef] 88. Tan, Z.; Sato, K.; Takami, S.; Numako, C.; Umetsu, M.; Soga, K.; Nakayama, M.; Sasaki, R.; Tanaka, T.; Ogino, C.; et al. Particle size for photocatalytic activity of anatase TiO2 nanosheets with highly exposed {001} facets. RSC Adv. 2013, 3, 19268–19271. [CrossRef] 89. Liu, G.; Yang, H.G.; Pan, J.; Yang, Y.Q.; Lu, G.Q.; Cheng, H.M. Titanium dioxide crystals with tailored facets. Chem. Rev. 2014, 114, 9559–9612. [CrossRef] 90. Zheng, Z.; Huang, B.; Qin, X.; Zhang, X.; Dai, Y.; Jiang, M.; Wang, P.; Whangbo, M.H. Highly efﬁcient photocatalyst: TiO2 microspheres produced from TiO2 nanosheets with a high percentage of reactive {001} facets. Chemistry 2009, 15, 12576–12579. [CrossRef] 91. Wang, J.; Zhang, P.; Li, X.; Zhu, J.; Li, H. Synchronical pollutant degradation and H2 production on a Ti3+-doped TiO2 visible photocatalyst with dominant (001) facets. Appl. Catal. B Environ. 2013, 134–135, 198–204. [CrossRef] 92. Banerjee, B.; Amoli, V.; Maurya, A.; Sinha, A.K.; Bhaumik, A. Green synthesis of Pt-doped TiO2 nanocrystals with exposed (001) facets and mesoscopic void space for photo-splitting of water under solar irradiation. Nanoscale 2015, 7, 10504–10512. [CrossRef] [PubMed] 93. Fattakhova-Rohlﬁng, D.; Zaleska, A.; Bein, T. Three-dimensional titanium dioxide nanomaterials. Chem. Rev. 2014, 114, 9487–9558. [CrossRef] [PubMed] 94. References A facile photoassisted route to synthesis N, F-codoped oxygen-deﬁcient TiO2 with enhanced photocatalytic performance under visible light irradiation. Appl. Surf. Sci. 2018, 434, 725–734. [CrossRef] 73. Kang, X.; Song, X.-Z.; Han, Y.; Cao, J.; Tan, Z. Defect-engineered TiO2 hollow spiny nanocubes for phenol degradation under visible light irradiation. Sci. Rep. 2018, 8, 5904. [CrossRef] 74. Li, X.; Yu, J.; Jaroniec, M. Hierarchical photocatalysts. Chem. Soc. Rev. 2016, 45, 2603–2636. [CrossRef] [PubMed] 75. Yang, H.G.; Zeng, H.C. Preparation of hollow anatase TiO2 nanospheres via Ostwald ripening. J. Phys. Chem. B 2004, 108, 3492–3495. [CrossRef] 172 Catalysts 2019, 9, 191 76. Pan, J.H.; Zhang, X.; Du, A.J.; Sun, D.D.; Leckie, J.O. Self-etching reconstruction of hierarchically mesoporous F-TiO2 hollow microspherical photocatalyst for concurrent membrane water puriﬁcations. J. Am. Chem. Soc. 2008, 130, 11256–11257. [CrossRef] [PubMed] 77. Cao, J.; Song, X.-Z.; Kang, X.; Dai, Z.; Tan, Z. One-pot synthesis of oleic acid modiﬁed monodispersed mesoporous TiO2 nanospheres with enhanced visible light photocatalytic performance. Adv. Powder Technol. 2018, 29, 1925–1932. [CrossRef] 78. Banerjee, A.N. The design, fabrication, and photocatalytic utility of nanostructured semiconductors: Focus on TiO2-based nanostructures. Nanotechnol. Sci. Appl. 2011, 4, 35–65. [CrossRef] 79. Lee, S.-Y.; Park, S.-J. TiO2 photocatalyst for water treatment applications. J. Ind. Eng. Chem. 2013, 19, 1761–1769. [CrossRef] 80. Pauzauskie, P.J.; Yang, P. Nanowire photonics. Mater. Today 2006, 9, 36–45. [CrossRef] 81. Yan, R.; Gargas, D.; Yang, P. Nanowire photonics. Nature Photonics 2009, 3, 569. [CrossR 82. Zhou, X.; Liu, N.; Schmuki, P. Photocatalysis with TiO2 nanotubes: “Colorful” reactivity and designing site-speciﬁc photocatalytic centers into TiO2 nanotubes. ACS Catal. 2017, 7, 3210–3235. [CrossRef] 82. Zhou, X.; Liu, N.; Schmuki, P. Photocatalysis with TiO2 nanotubes: Colorful reactivity and designing site-speciﬁc photocatalytic centers into TiO2 nanotubes. ACS Catal. 2017, 7, 3210–3235. [CrossRef] 83. Shibata, T.; Sakai, N.; Fukuda, K.; Ebina, Y.; Sasaki, T. Photocatalytic properties of titania nanostructured ﬁlms fabricated from Titania nanosheets. Phys. Chem. Chem. Phys. 2007, 9, 2413–2420. [CrossRef] [PubMed] 83 S bata, ; Saka , N ; ukuda, ; b a, ; Sasak , otocata yt c p ope t es o t ta a a ost uctu ed ﬁlms fabricated from Titania nanosheets. Phys. Chem. Chem. Phys. 2007, 9, 2413–2420. [CrossRef] [PubMed] ﬁlms fabricated from Titania nanosheets. Phys. Chem. Chem. Phys. 2007, 9, 2413–2420. [CrossRef] [PubMed] 84. Shichi, T.; Katsumata, K.-I. Development of photocatalytic self-cleaning glasses utilizing metal oxide 84. Shichi, T.; Katsumata, K.-I. References Xiong, T.; Dong, F.; Wu, Z. Enhanced extrinsic absorption promotes the visible light photocatalytic activity of wide band-gap (BiO)2CO3 hierarchical structure. RSC Adv. 2014, 4, 56307–56312. [CrossRef] 95. Li, H.; Bian, Z.; Zhu, J.; Zhang, D.; Li, G.; Huo, Y.; Li, H.; Lu, Y. Mesoporous titania spheres with tunable chamber stucture and enhanced photocatalytic activity. J. Am. Chem. Soc. 2007, 129, 8406–8407. [CrossRef] [PubMed] 96. Kondo, Y.; Yoshikawa, H.; Awaga, K.; Murayama, M.; Mori, T.; Sunada, K.; Bandow, S.; Iijima, S. Preparation, photocatalytic activities, and dye-sensitized solar-cell performance of submicron-scale TiO2 hollow spheres. Langmuir 2008, 24, 547–550. [CrossRef] [PubMed] 97. Wang, L.; Takayoshi, S.; Ebina, Y.; Kurashima, K.; Watanabe, M. Fabrication of controllable ultrathin hollow shells by layer-by-layer assembly of exfoliated ritania nanosheets on polymer templates. Chem. Mater. 2002, 14, 4827–4832. [CrossRef] 173 Catalysts 2019, 9, 191 98. Zhao, T.; Liu, Z.; Nakata, K.; Nishimoto, S.; Murakami, T.; Zhao, Y.; Jiang, L.; Fujishima, A. Multichannel TiO2 hollow ﬁbers with enhanced photocatalytic activity. J. Mater. Chem. 2010, 20, 5095–5099. [CrossRef] 99. Shang, S.; Jiao, X.; Chen, D. Template-free fabrication of TiO2 hollow spheres and their photocatalytic properties. ACS Appl. Mater. Interfaces 2012, 4, 860–865. [CrossRef] 100. Song, R.Q.; Colfen, H. Mesocrystals–ordered nanoparticle superstructures. Adv. Mater. 2010, 22, 1301–1330. [CrossRef] 101. Colfen, H.; Antonietti, M. Mesocrystals: Inorganic superstructures made by highly parallel crystallization and controlled alignment. Angew. Chem. Int. Ed. 2005, 44, 5576–5591. [CrossRef] [PubMed] 102. Cai, J.; Qi, L. TiO2 mesocrystals: Synthesis, formation mechanisms and applications. Sci. China Chem. 2012, 55, 2318–2326. [CrossRef] 103. Zhang, P.; Tachikawa, T.; Bian, Z.; Majima, T. Selective photoredox activity on speciﬁc facet-dominated TiO2 mesocrystal superstructures incubated with directed nanocrystals. Appl. Catal. B Environ. 2015, 176–177, 678–686. [CrossRef] 104. Zhang, P.; Tachikawa, T.; Fujitsuka, M.; Majima, T. Efﬁcient charge separation on 3D architectures of TiO2 mesocrystals packed with a chemically exfoliated MoS2 shell in synergetic hydrogen evolution. Chem. Commun. 2015, 51, 7187–7190. [CrossRef] [PubMed] 105. Zhang, P.; Fujitsuka, M.; Majima, T. TiO2 mesocrystal with nitrogen and ﬂuorine codoping during topochemical transformation: Efﬁcient visible light induced photocatalyst with the codopants. Appl. Catal. B Environmental 2016, 185, 181–188. [CrossRef] 106. Elbanna, O.; Fujitsuka, M.; Majima, T. g-C3N4/TiO2 Mesocrystals composite for H2 evolution under visible-light irradiation and its charge carrier dynamics. ACS Appl. Mater. Interfaces 2017, 9, 34844–34854. [CrossRef] [PubMed] 107. Ossama Elbanna, S.K. References Mamoru Fujitsuka and Tetsuro Majima, TiO2 mesocrystals composited with gold nanorods for highly efﬁcient visible-NIR-photocatalytic hydrogen production. Nano Energy 2017, 17, 1842. 108. Kang, X.; Song, X.-Z.; Liu, S.; Pei, M.; Wen, W.; Tan, Z. In situ formation of defect-engineered N-doped TiO2 porous mesocrystal for enhanced photo-degradation and PEC performance. Nanoscale Adv. 2019. [CrossRef] 108. Kang, X.; Song, X.-Z.; Liu, S.; Pei, M.; Wen, W.; Tan, Z. In situ formation of defect-engineered N-doped TiO2 porous mesocrystal for enhanced photo degradation and PEC performance Nanoscale Adv 2019 [CrossRef] 109. Choi, Y.; Ko, M.S.; Bokare, A.D.; Kim, D.-H.; Bahnemann, D.W.; Choi, W. Sequential process combination of photocatalytic oxidation and dark reduction for the removal of organic pollutants and Cr(VI) using Ag/TiO2. Environ. Sci. Technol. 2017, 51, 3975–3981. [CrossRef] 110. Kim, S.; Park, H. Sunlight-harnessing and storing heterojunction TiO2/Al2O3/WO3 electrodes for night-time applications. RSC Adv. 2013, 3, 17551–17558. [CrossRef] 111. Tatsuma, T.; Saitoh, S.; Ohko, Y.; Fujishima, A. TiO2-WO3 photoelectrochemical anticorrosion system with an energy storage ability. Chem. Mater. 2001, 13, 2838–2842. [CrossRef] 112. Li, J.; Cushing, S.K.; Zheng, P.; Senty, T.; Meng, F.; Bristow, A.D.; Manivannan, A.; Wu, N. Solar hydrogen generation by a CdS-Au-TiO2 sandwich nanorod array enhanced with Au nanoparticle as electron relay and plasmonic photosensitizer. J. Am. Chem. Soc. 2014, 136, 8438–8449. [CrossRef] [PubMed] 113. Hu, L.; Zhang, Y.; Zhang, S.; Li, B. A transparent TiO2-C@TiO2-graphene free-standing ﬁlm with enhanced visible light photocatalysis. RSC Adv. 2016, 6, 43098–43103. [CrossRef] 113. Hu, L.; Zhang, Y.; Zhang, S.; Li, B. A transparent TiO2-C@TiO2-graphene f visible light photocatalysis. RSC Adv. 2016, 6, 43098–43103. [CrossRef] 114. Wang, W.; Ni, Y.; Lu, C.; Xu, Z. Hydrogenation of TiO2 nanosheets with exposed {001} facets for enhan photocatalytc activity. RSC Adv. 2012, 2, 8286–8288. 115. Plodinec, M.; Grcic, I.; Willinger, M.G.; Hammud, A.; Huang, X.; Panzic, I.; Gajovic, A. Black TiO2 nanotube arrays decorated with Ag nanoparticles for enhanced wisible-light photocatalytic oxidation of salicylic acid. J. Alloys Comp. 2019, 776, 883–896. [CrossRef] 116. Ling, H.; Kim, K.; Liu, Z.; Shi, J.; Zhu, X.; Huang, J. Photocatalytic degradation of phenol in water on as-prepared and surface modiﬁed TiO2 nanoparticles. Catal. Today 2015, 258, 96–102. [CrossRef] 117. Hao, X.; Li, M.; Zhang, L.; Wang, K.; Liu, C. Photocatalyst TiO2/WO3/GO nano-composite with high efﬁcient photocatalytic performance for BPA degradation under visible light and solar light illumination. J. Ind. Eng. Chem. 2017, 55, 140–148. [CrossRef] J g 118. References Khamboonrueang, D.; Srirattanapibul, S.; Tang, I.-M.; Thongmee, S. TiO2·rGO nanocomposite as a photo catalyst for the reduction of Cr6+. Mater. Res. Bull. 2018, 107, 236–241. [CrossRef] 119. Ren, H.-T.; Jia, S.-Y.; Zou, J.-J.; Wu, S.-H.; Han, X. A facile preparation of Ag2O/P25 photocatalyst for selective reduction of nitrate. Appl. Catal. B Environ. 2015, 176–177, 53–61. [CrossRef] 174 Catalysts 2019, 9, 191 120. Fan, C.; Chen, C.; Wang, J.; Fu, X.; Ren, Z.; Qian, G.; Wang, Z. Black hydroxylated titanium dioxide prepared via ultrasonication with enhanced photocatalytic activity. Sci. Rep. 2015, 5, 11712. [CrossRef] 121. Vequizo, J.J.M.; Matsunaga, H.; Ishiku, T.; Kamimura, S.; Ohno, T.; Yamakata, A. Trapping-induced enhancement of photocatalytic activity on brookite TiO2 powders: Comparison with anatase and rutile TiO2 powders. ACS Catal. 2017, 7, 2644–2651. [CrossRef] 122. Kim, S.; Moon, G.-h.; Kim, G.; Kang, U.; Park, H.; Choi, W. TiO2 complexed with dopamine-derived polymers and the visible light photocatalytic activities for water pollutants. J. Catal. 2017, 346, 92–100. [CrossRef] 123. Huang, H.L.; Lee, W.G.; Wu, F.S. Emissions of air pollutants from indoor charcoal barbecue. J. Hazard. M 2016, 302, 198–207. [CrossRef] [PubMed] 124. Rodrigues, S.; Ranjit, K.T.; Uma, S.; Martyanov, I.N.; Klabunde, K.J. Single-step synthesis of a highly active visible-light photocatalyst for oxidation of a common indoor air pollutant: Acetaldehyde. Adv. Mater. 2005, 17, 2467–2471. [CrossRef] 125. Lyu, J.; Zhu, L.; Burda, C. Considerations to improve adsorption and photocatalysis of low concentration air pollutants on TiO2. Catal. Today 2014, 225, 24–33. [CrossRef] 126. Fujiwara, K.; Müller, U.; Pratsinis, S.E. Pd subnano-clusters on TiO2 for solar-light removal of NO. ACS Catal. 2016, 6, 1887–1893. [CrossRef] 127. Martins, N.C.T.; Ângelo, J.; Girão, A.V.; Trindade, T.; Andrade, L.; Mendes, A. N-doped carbon quantum dots/TiO2 composite with improved photocatalytic activity. Appl. Catal. B Environ. 2016, 193, 67–74. [CrossRef] 127. Martins, N.C.T.; Ângelo, J.; Girão, A.V.; Trindade, T.; Andrade, L.; Mendes, A. N-doped carbon quantum dots/TiO2 composite with improved photocatalytic activity. Appl. Catal. B Environ. 2016, 193, 67–74. [CrossRef] 128. Zeng, L.; Song, W.; Li, M.; Zeng, D.; Xie, C. Catalytic oxidation of formaldehyde on surface of / h l h ll l l 128. Zeng, L.; Song, W.; Li, M.; Zeng, D.; Xie, C. Catalytic oxidation of formaldehyde on surface of H-TiO2/H-C-TiO2 without light illumination at room temperature. Appl. Catal. B Environ. 2014, 147, 490–498. [CrossRef] 129. Yamazakia, S.; Tsukamoto, H.; Araki, K.; Tanimura, T.; Tejedor-Tejedor, I.; Anderson, M.A. References Photocatalytic degradation of gaseous tetrachloroethylene on porous TiO2 pellets. Appl. Catal. B Environ. 2001, 33, 109–117. [CrossRef] 130. Hernández-Alonso, M.D.; Tejedor-Tejedor, I.; Coronado, J.M.; Anderson, M.A.; Soria, J. Operando FTIR study of the photocatalytic oxidation of acetone in air over TiO2-ZrO2 thin ﬁlm. Catal. Today 2009, 143, 364–373. [CrossRef] 131. Xu, Y.-J.; Zhuang, Y.; Fu, X. New insight for enhanced photocatalytic Activity of TiO2 by doping carbon nanotubes: A case study on degradation of benzene and methyl orange. J. Phys. Chem. C. 2010, 114, 2669–2676. [CrossRef] 132. Zou, X.; Liu, J.; Su, J.; Zuo, F.; Chen, J.; Feng, P. Facile synthesis of thermal- and photostable titania with paramagnetic oxygen vacancies for visible-light photocatalysis. Chemistry 2013, 1, 2866–2873. [CrossRef] 133. Zhou, W.; Li, W.; Wang, J.Q.; Qu, Y.; Yang, Y.; Xie, Y.; Zhang, K.; Wang, L.; Fu, H.; Zhao, D. Ordered mesoporous black TiO(2) as highly efﬁcient hydrogen evolution photocatalyst. J. Am. Chem. Soc. 2014, 136, 9280–9283. [CrossRef] [PubMed] 134. Zhong, R.; Zhang, Z.; Yi, H.; Zeng, L.; Tang, C.; Huang, L.; Gu, M. Covalently bonded 2D/2D O-g-C3N4/TiO2 heterojunction for enhanced visible-light photocatalytic hydrogen evolution. Appl. Catal. B Environ. 2018, 237, 1130–1138. [CrossRef] 135. Slamet; Nasution, H.W.; Purnama, E.; Kosela, S.; Gunlazuardi, J. Photocatalytic reduction of CO2 on copper-doped Titania catalysts prepared by improved-impregnation method. Catal. Commun. 2005, 6, 313–319. [CrossRef] 136. Liu, D.; Fernández, Y.; Ola, O.; Mackintosh, S.; Maroto-Valer, M.; Parlett, C.M.A.; Lee, A.F.; Wu, J.C.S. On the impact of Cu dispersion on CO2 photoreduction over Cu/TiO2. Catal. Commun. 2012, 25, 78–82. [CrossRef] 137. Liu, G.; Wang, H.; Wang, M.; Liu, W.; Ardhi, R.E.A.; Zou, D.; Lee, J.K. Study on a stretchable, ﬁber-shaped, and TiO2 nanowire array-based dye-sensitized solar cell with electrochemical impedance spectroscopy method. Electrochim. Acta 2018, 267, 34–40. [CrossRef] 138. Liu, B.; Sun, Y.; Wang, X.; Zhang, L.; Wang, D.; Fu, Z.; Lin, Y.; Xie, T. Branched hierarchical photoanode of anatase TiO2 nanotubes on rutile TiO2 nanorod arrays for efﬁcient quantum dot-sensitized solar cells. J. Mater. Chem. A 2015, 3, 4445–4452. [CrossRef] 139. Chen, J.; Li, Y.; Mu, J.; Zhang, Y.; Yu, Z.; Han, K.; Zhang, L. C@TiO2 nanocomposites with impressive electrochemical performances as anode material for lithium-ion batteries. J. Alloys Comp. 2018, 742, 828–834. [CrossRef] 175 Catalysts 2019, 9, 191 140. Su, D.; Dou, S.; Wang, G. Anatase TiO2: Better anode material than amorphous and rutile phases of TiO2 for Na-ion batteries. Chem. Mater. References 2015, 27, 6022–6029. [CrossRef] 141. Kim, C.; Kim, S.; Lee, J.; Kim, J.; Yoon, J. Capacitive and oxidant generating properties of black-colored TiO2 nanotube array fabricated by electrochemical self-doping. ACS Appl. Mater. Interfaces 2015, 7, 7486–7491. [CrossRef] [PubMed] 142. Yang, S.; Lin, Y.; Song, X.; Zhang, P.; Gao, L. Covalently coupled ultraﬁne H-TiO2 nanocrystals/ nitrogen-doped graphene hybrid materials for high-performance supercapacitor. ACS Appl. Mater. Interfaces 2015, 7, 17884–17892. [CrossRef] [PubMed] 143. Pazhamalai, P.; Krishnamoorthy, K.; Mariappan, V.K.; Kim, S.J. Blue TiO2 nanosheets as a high-performance electrode material for supercapacitors. J. Colloid Interface Sci. 2019, 536, 62–70. [CrossRef] [PubMed] 144. Liu, B.; Mu, L.; Han, B.; Zhang, J.; Shi, H. Fabrication of TiO2/Ag2O heterostructure with enhanced photocatalytic and antibacterial activities under visible light irradiation. Appl. Surf. Sci. 2017, 396, 1596–1603. [CrossRef] 145. Yu, X.; Wang, S.; Zhang, X.; Qi, A.; Qiao, X.; Liu, Z.; Wu, M. Heterostructured nanorod array with piezophototronic and plasmonic effect for photodynamic bacteria killing and wound healing. Nano Energy 2018, 46, 29–38. [CrossRef] 146. Karlsson, J.; Atefyekta, S.; Andersson, M. Controlling drug delivery kinetics from mesoporous titania t ﬁlms by pore size and surface energy. Int. J. Nanomed. 2015, 10, 4425–4436. [CrossRef] [PubMed] 147. Gupta, B.; Poudel, B.K.; Ruttala, H.B.; Regmi, S.; Pathak, S.; Gautam, M.; Jin, S.G.; Jeong, J.H.; Choi, H.G.; Ku, S.K.; et al. Hyaluronic acid-capped compact silica-supported mesoporous titania nanoparticles for ligand-directed delivery of doxorubicin. Acta Biomater. 2018, 80, 364–377. [CrossRef] Ku, S.K.; et al. Hyaluronic acid-capped compact silica-supported mesoporous titania nanoparticles for ligand-directed delivery of doxorubicin. Acta Biomater. 2018, 80, 364–377. [CrossRef] g y 148. Guo, Z.; Zheng, K.; Tan, Z.; Liu, Y.; Zhao, Z.; Zhu, G.; Ma, K.; Cui, C.; Wang, L.; Kang, T. Overcoming drug resistance with functional mesoporous titanium dioxide nanoparticles combining targeting, drug delivery and photodynamic therapy. J. Mater. Chem. B 2018, 6, 7750–7759. [CrossRef] 149. Nakayama, M.; Sasaki, R.; Ogino, C.; Tanaka, T.; Morita, K.; Umetsu, M.; Ohara, S.; Tan, Z.; Nishimura, Y.; Akasaka, H.; et al. Titanium peroxide nanoparticles enhanced cytotoxic effects of X-ray irradiation against pancreatic cancer model through reactive oxygen species generation in vitro and in vivo. Radiat. Oncol. 2016, 11, 91. [CrossRef] 150. Dai, Z.; Song, X.Z.; Cao, J.; He, Y.; Wen, W.; Xu, X.; Tan, Z. Dual-stimuli-responsive TiOx/DOX nanodrug system for lung cancer synergistic therapy. RSC Adv. 2018, 8, 21975. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. Synergistic Effect of Photocatalytic Degradation of Hexabromocyclododecane in Water by UV/TiO2/persulfate Qiang Li 1,2, Lifang Wang 1, Xuhui Fang 3, Li Zhang 2,, Jingjiu Li 2 and Hongyong Xie 2, 1 School of Management, Northwestern Polytechnical University, 127 West Youxi Road, Xian 710072, Chi failureend@163.com (Q.L.); lifang@nwpu.edu.cn (L.W.) 2 Research Center of Resource Recycling Science and Engineering, School of Environmental and Materials Engineering, Shanghai Polytechnic University, Shanghai 201209, China; lijingjiu123@163.com 3 Centre Testing International Pinbiao (Shanghai) Co., Ltd., 1996 New Jinqiao Road, Shanghai 201206, China; fangxuwh@126.com 2 Research Center of Resource Recycling Science and Engineering, School of Environmental and Materials Engineering, Shanghai Polytechnic University, Shanghai 201209, China; lijingjiu123@163.com 3 Centre Testing International Pinbiao (Shanghai) Co., Ltd., 1996 New Jinqiao Road, Shanghai 201206, China; fangxuwh@126.com g * Correspondence: zhangli@sspu.edu.cn (L.Z.); hyxie@sspu.edu.cn (H.X.); Tel.: +86-021-5021-1210 (L.Z.); +86-021-5021-1231 (H.X.) * Correspondence: zhangli@sspu.edu.cn (L.Z.); hyxie@sspu.edu.cn (H.X.); Tel.: +86-021-5021-1210 (L.Z.); +86-021-5021-1231 (H.X.) Received: 4 January 2019; Accepted: 31 January 2019; Published: 18 February 2019 Abstract: In this work, the elimination of hexabromocyclododecane (HBCD) is explored by using photodegradation of the UV/TiO2 system, the UV/potassium persulfate (KPS) system, and the homo/heterogeneous UV/TiO2/KPS system. The experimental results show that the dosages of TiO2 and potassium persulfate have optimum values to increase the degradation degree. HBCD can be almost completely degraded and 74.3% of the total bromine content is achieved in the UV/TiO2/KPS homo/heterogeneous photocatalysis, much more than in the UV/persulfate system and the UV/TiO2 system. Roles of radicals SO4•−and OH• in the photocatalysis systems are discussed based on experimental measurements. The high yield of the concentration of bromide ions and decreased pH value indicates that synergistic effects exist in the UV/TiO2/KPS homo/heterogeneous photocatalysis, which can mineralize HBCD into inorganic small molecules like carboxylic acids, CO2 and H2O, thus much less intermediates are formed. The possible pathways of degradation of HBCD in the UV/TiO2/KPS system were also analyzed by GC/MS. This work will have practical application potential in the ﬁelds of pollution control and environmental management. Keywords: hexabromocyclododecane; environmental management; photocatalysis; advanced oxidation processes Catalysts 2019, 9, 189; doi:10.3390/catal9020189 References This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 176 www.mdpi.com/journal/catalysts catalysts catalysts 1. Introduction The photocatalytic technology can be brieﬂy described as follows: under UV irradiation, the electron in TiO2 was excited and transferred from the valence band (VB) to the conduction band (CB), resulting in the formation of high energy electron-hole pairs; the electrons may also react with O2 and generate a superoxide ion (O2•−), while holes were captured by surface hydroxyl groups (OH−) on the photocatalyst surface to yield OH• [14,15]. However, the photogenerated electron-hole pairs are easy to combine within a very short time of 10−9~10−12 s, which results in a lower photocatalytic degradation efﬁciency [16]. In order to solve this problem, scientists have conducted lots of meaningful and in-depth research. For instance, Aronne et al. found that high Ti3+ self-doping TiO2−x not only has a wide range of visible light responses, but also has a low recombination rate of electron–hole pairs [17]; Sannino et al. fabricated hybrid TiO2–acetylacetonate amorphous gel-derived material with stably adsorbed superoxide radical (O2•−) active in oxidative degradation of organic pollutants in the absence of any light irradiation [18,19]. have not been practically used because of the harsh reaction conditions, high energy consumption, low efﬁciency, and secondary pollution [9,10]. Nowadays, advanced oxidation processes (AOPs) have been widely used for the elimination of organic pollutants of water or gas, using the highly reactive chemical species like hydroxyl radicals (OH•) to oxidize most of pollutants into small molecular substances that are harmless to the environment, such as CO2, H2O and so on [11]. Increasing the number of hydroxyl radicals could increase the efﬁciency of the AOPs reactions. Some types of AOPs based on UV, H2O2/UV, O3/UV and H2O2/O3/UV combinations use photolysis of H2O2 and ozone to produce OH•, while the heterogeneous UV/TiO2 photocatalysis and homogeneous photo-Fenton are based on the use of a wide-band gap semiconductor and addition of H2O2 to dissolved iron salts that produce OH• under UV irradiation, respectively [12]. Among AOPs, the UV/TiO2 heterogeneous photocatalysis has gradually attracted the interest of scientists in elimination of toxic pollutions due to its efﬁciency, low-cost and broad applicability [13]. 1. Introduction The photocatalytic technology can be brieﬂy described as follows: under UV irradiation, the electron in TiO2 was excited and transferred from the valence band (VB) to the conduction band (CB), resulting in the formation of high energy electron-hole pairs; the electrons may also react with O2 and generate a superoxide ion (O2•−), while holes were captured by surface hydroxyl groups (OH−) on the photocatalyst surface to yield OH• [14,15]. However, the photogenerated electron-hole pairs are easy to combine within a very short time of 10−9~10−12 s, which results in a lower photocatalytic degradation efﬁciency [16]. In order to solve this problem, scientists have conducted lots of meaningful and in-depth research. For instance, Aronne et al. found that high Ti3+ self-doping TiO2−x not only has a wide range of visible light responses, but also has a low recombination rate of electron–hole pairs [17]; Sannino et al. fabricated hybrid TiO2–acetylacetonate amorphous gel-derived material with stably adsorbed superoxide radical (O2•−) active in oxidative degradation of organic pollutants in the absence of any light irradiation [18,19]. It has been reported that using the strong oxidant of persulfate ion (S2O82−) (with redox potential of 2.05 V) is effective for degrading organic pollutions in water solution through direct chemical oxidation [20]. The S2O82−can be activated via thermal, UV light, or redox decomposition to generate the stronger oxidant of sulfate radicals (E0 = (2.5–3.1) V vs. NHE) [21–23]. It’s worthwhile to note that both the persulfate ion and sulfate radicals (SO4•−) can be dissolved in water, so the free radicals and contaminants in water can be contracted at the molecular level, leading to a higher reaction rate. For instance, Li et al. have found that addition of persulphate to UV/TiO2 could improve the photocatalytic degradation of tetrabromobisphenol A and other pollutants [24–26]. Therefore, it is necessary to combine persulfate and UV/TiO2 photocatalytic techniques to increase the mineralization of HBCD. In this work, degradation of HBCD under UV/TiO2, UV/potassium persulfate (KPS), and UV/TiO2/KPS systems were investigated. Effects of TiO2 and KPS dosage have been examined on degradation degree of HBCD. The photodegradation efﬁciency and the yield of bromide ion were tested to evaluate the mineralization of HBCD. The intermediates were analyzed by GC/MS to study the degradation mechanism. Based on experimental measurements, roles of radicals SO4•−and OH• in the photocatalysis systems were also discussed. 1. Introduction Hexabromocyclododecane (HBCD) is a high bromine content additive ﬂame retardant that is mainly used in polystyrene electrical equipment, insulation boards, resin, polyester fabric, synthetic rubber coating, and so on [1]. Studies have shown that HBCD is a potential endocrine disruptor, and it has immunotoxicity, neurotoxicity, and cytotoxicity [2]. The presence of HBCD was detected in environmental samples such as water, atmosphere, sediment, soil, food, and even in the human body in breast milk and plasma, as it can be enriched through the food chain, causing persistent pollution [3]. The hazard of HBCD and its pollution in the environment have caused widespread concern, and it is of great signiﬁcance to develop a strategy for the elimination of HBCD pollution [4]. However, the molecular structure of HBCD with a ring structure is relatively stable, making it heat-resistant, UV-resistant, and difﬁcult to be decomposed in the natural environment [5]. Methods for eliminating HBCD in the environment include microbial degradation, ultrasonic degradation, chemical reduction, phytoremediation, and mechanical ball milling [6–8]. However, these methods Catalysts 2019, 9, 189; doi:10.3390/catal9020189 www.mdpi.com/journal/catalysts www.mdpi.com/journal/catalysts 177 Catalysts 2019, 9, 189 have not been practically used because of the harsh reaction conditions, high energy consumption, low efﬁciency, and secondary pollution [9,10]. Nowadays, advanced oxidation processes (AOPs) have been widely used for the elimination of organic pollutants of water or gas, using the highly reactive chemical species like hydroxyl radicals (OH•) to oxidize most of pollutants into small molecular substances that are harmless to the environment, such as CO2, H2O and so on [11]. Increasing the number of hydroxyl radicals could increase the efﬁciency of the AOPs reactions. Some types of AOPs based on UV, H2O2/UV, O3/UV and H2O2/O3/UV combinations use photolysis of H2O2 and ozone to produce OH•, while the heterogeneous UV/TiO2 photocatalysis and homogeneous photo-Fenton are based on the use of a wide-band gap semiconductor and addition of H2O2 to dissolved iron salts that produce OH• under UV irradiation, respectively [12]. Among AOPs, the UV/TiO2 heterogeneous photocatalysis has gradually attracted the interest of scientists in elimination of toxic pollutions due to its efﬁciency, low-cost and broad applicability [13]. 2.1. Determination of TiO2 Dosages The amount of catalysts added in the solution needs to be matched to the number of contaminants in the photocatalytic process, so the dosing weight range of the catalyst were determined. Figure 1 shows the effects of different TiO2 dosages on the photodegradation rate of HBCD. Under the condition of no addition of TiO2, the degradation rate was only 21.5% at 180 min. Having increased the catalyst dosage of TiO2 to 100 mg/L, the photocatalytic efﬁciency also increased to 82.93%. Further increasing the dosage of TiO2 more than 100 mg/L, the photocatalytic efﬁciency decreased. The dosage of the 178 Catalysts 2019, 9, 189 addition increased the suspended particles in the solution, and greatly reduced the utilization of light, resulting in the partial catalyst not being fully activated during the photocatalysis, so the photocatalytic efﬁciency decreased [24,27]. In addition, the HBCD adsorbed on the catalysts in dark is less than 10%, so the free radical (OH•) reaction dominates the rate of degradation reaction in the UV/TiO2 system. Figure 1. Effects of different TiO2 dosages on the photodegradation rate of HBCD (the initial concentration of HBCD is 25 mg/L, and KPS dosage is 0 mg/L). Figure 1. Effects of different TiO2 dosages on the photodegradation rate of HBCD (the initial concentration of HBCD is 25 mg/L, and KPS dosage is 0 mg/L). 2.3. Kinetic Analysis of Different Reaction Systems 2.3. Kinetic Analysis of Different Reaction Systems Figure 3 shows the degradation effect of HBCD in the three systems of “UV/TiO2 (TiO2: 100 mg/L)”, “UV/K2S2O8 (KPS: 4 mM)”, and “TiO2 (TiO2: 100 mg/L) + K2S2O8 (KPS: 4 mM)”, respectively. The initial HBCD concentration and light source in the three systems were all the same (25 mg/L, 100 W mercury lamp). The degradation degree of HBCD over time in 180 min is shown in Figure 3a. The degradation degree for the UV/TiO2/KPS photocatalytic system was 87.6% at 90 min, but the degradation rates for the UV/TiO2 photocatalytic system and UV/KPS system were only 56.8% and 52.5% at the same time. The above experimental results show that the degradation effect of UV/TiO2/KPS photocatalytic system on HBCD is far superior to that of the UV/TiO2 system and the UV/KPS system. The kinetic model was used to study the degradation dynamic behavior of the three different systems [30], −ln(C0/C) = kt, where k is the reaction apparent rate constant and t is the light irradiation time. Figure 3b shows the effect of different systems on the kinetics of HBCD under irradiation for 180 min. The three reaction systems are all ﬁt to pseudo-ﬁrst-order kinetics, and the k values for the UV/KPS system, the UV/TiO2 system, and the UV/TiO2/KPS system are 0.0065, 0.0080, and 0.0174 min−1, respectively (Figure 3b). Obviously, the k value of the UV/TiO2/KPS system is far higher than those of the UV/KPS system and the UV/TiO2 system, indicating that the degradation efﬁciency of the UV/TiO2/KPS photocatalytic system is much higher than that of the UV/KPS photocatalytic system and the UV/TiO2 photocatalytic system. The photocatalytic process of the UV/TiO2 system contains an adsorption and free radical (O2•−, OH•, etc.) reaction [21]. The HBCD adsorbed on the catalysts in dark is less than 10%, so the free radical reaction dominates the rate of degradation reaction in the UV/TiO2 system. The UV/KPS system also relies on sulfate radicals (SO4•−, S2O8•−, etc.) excited by UV light to degrade pollutants [22,24]. In the UV/TiO2/KPS system, more free radicals were present and the free radical reaction is more complicated. S2O82−can be excited by photogenerated electrons on the surface of the catalyst to generate sulfate radicals (SO4•−), while SO4•−can react with OH−to produce OH• [21–24]. 2.2. Effect of KPS Dosage Figure 2 shows the effect of different K2S2O8 dosages on the photodegradation rate of HBCD. The addition of K2S2O8 can effectively improve the degradation efﬁciency of HBCD, but its degradation efﬁciency increases ﬁrst and then decreases with further increasing K2S2O8 concentration, and the highest degradation efﬁciency occurs at 4 mM. When the K2S2O8 dosage was more than 4 mM, the degradation efﬁciency of HBCD decreased with the increase of persulfate dosage. When the K2S2O8 concentrations in the solution are between 0 and 4 mM, the main reactions in the UV/TiO2/KPS system are as follows [28]: S2O82−+ UV →2SO4•− (1) TiO2 + UV →e− CB + h+ VB (2) S2O82−+ e− CB →2SO4•− (3) H2O + h+ VB →OH• + H+ (4) (3) (4) The advanced oxidation process relies on the amount of free radicals and is reﬂected in the degradation rate of the contaminants. The strong oxidizing agents of sulfate radicals (SO4•−) and hydroxyl radicals (OH•) generated by the above reactions increase with increasing KPS concentration in the solution. But increasing the KPS dosage further to 8 mM will lead to a surplus of reactants (S2O82−), which may deplete lots of OH• and SO4•−, and lead to the decrease of the degradation degree by the following two reactions [29]: S2O82−+ OH• →S2O8•−+ OH− (5) S2O82−+ SO4•−→SO42−+ S2O8•− (6) (5) (6) 179 179 Catalysts 2019, 9, 189 Figure 2. Effects of different KPS dosages on the photodegradation rate of HBCD (the initial concentration of HBCD is 25 mg/L, and TiO2 dosage is 100 mg/L). Catalysts 2019, 9, 189 Figure 2. Effects of different KPS dosages on the photodegradation rate of HBCD (the initial concentration of HBCD is 25 mg/L, and TiO2 dosage is 100 mg/L). 2.3. Kinetic Analysis of Different Reaction Systems In the three systems, the degradation rates are all determined by the reactions between free radicals and contaminant molecules, while the intensity of the UV light (100 W) and the initial concentration (25.00 mg/L) of the contaminants in the three systems are all the same, so all reaction systems could be in line with pseudo-ﬁrst-order kinetics. Figure S1 (Supporting Information) shows the degradation of HBCD over the UV/TiO2/KPS system with three time cycling uses. The TiO2 photocatalysts could be easily recovered by sedimentation and reused, which would greatly promote their industrial application in eliminating organic pollutants from wastewater. 180 Catalysts 2019, 9, 189 Figure 3. The degradation degree (a) and kinetic linear simulation curves of the removal of HBCD (b) in UV/TiO2, UV/KPS, and UV/TiO2/KPS systems. (The initial concentration of HBCD is 25.0 mg/L, KPS dosage is 4 mg/L, and TiO2 dosage is 100 mg/L). Figure 3. The degradation degree (a) and kinetic linear simulation curves of the removal of HBCD (b) in UV/TiO2, UV/KPS, and UV/TiO2/KPS systems. (The initial concentration of HBCD is 25.0 mg/L, KPS dosage is 4 mg/L, and TiO2 dosage is 100 mg/L). 2.4. The Mineralization Degree of HBCD Measuring the concentration of bromide ion is a practical strategy to evaluate the amount of intermediates and the mineralization degree of HBCD [24]. Figure 4 shows the change of bromide ion concentration during the degradation of HBCD in the UV/TiO2/KPS system. The initial HBCD concentration, TiO2 dosage, and K2S2O8 dosage were 25.0 mg/L, 100 mg/L, and 4 mM, respectively. As can be seen from Figure 4, with the prolongation of degradation time, the concentration of bromide ion in the solution increased continuously. When the reaction was carried out for 3.0 h, the concentration of bromide ions in the solution was 13.8 mg/L, which accounted for 74.3% of the total bromine content of HBCD in the solution. It can be seen that the yield of bromine ion by UV/TiO2/KPS system is much better than that of UV/TiO2 system (12.3 mg/L) and UV/KPS system (11.9 mg/L). The signiﬁcantly increased bromide ion concentration yield indicates that there exist synergistic effects in the UV/TiO2/KPS photocatalysis as described in the previous Formulas (1)–(4), which can mineralize HBCD into inorganic small molecules relatively thorough, thus much fewer intermediates are formed in UV/TiO2/KPS homo/heterogeneous photocatalysis. Figure 5 shows the change of pH in solution over time during HBCD degradation. It can be seen that the pH value of the solution gradually decreases from 6.53 to 3.72 with the increasing of the degradation time within 180 min. It may be due to the partial consumption of OH−in the solution, since OH−can easily react with SO4•−to produce OH•. The CO2 gas generated during the mineralization of HBCD subsequently dissolved in the water, which also lead to a decrease in pH. 181 Catalysts 2019, 9, 189 Catalysts 2019, 9, 189 At the same time, there are some small molecules of carboxylic acids generated in the degradation of HBCD, which also cause the decrease in pH. As mentioned above, the degradation rate of HBCD reached 96.5% when the reaction proceeded to 180 min. It is indicated that the intermediate in the solution is rapidly decomposed into small molecular of carboxylic acids, and further mineralized to CO2 and H2O, so that the pH of the solution continued to decrease as the reaction time prolonged. The increasing concentration of bromide ions in the solution and the decreasing pH value indicate that HBCD is highly mineralized in UV/TiO2/KPS homo/heterogeneous photocatalysis. 2.4. The Mineralization Degree of HBCD At the same time, there are some small molecules of carboxylic acids generated in the degradation of HBCD, which also cause the decrease in pH. As mentioned above, the degradation rate of HBCD reached 96.5% when the reaction proceeded to 180 min. It is indicated that the intermediate in the solution is rapidly decomposed into small molecular of carboxylic acids, and further mineralized to CO2 and H2O, so that the pH of the solution continued to decrease as the reaction time prolonged. The increasing concentration of bromide ions in the solution and the decreasing pH value indicate that HBCD is highly mineralized in UV/TiO2/KPS homo/heterogeneous photocatalysis. Figure 4. The change of bromide ion in water samples from different time points (the initial concentration of HBCD is 25.0 mg/L, KPS dosage is 4 mg/L, and TiO2 dosage is 100 mg/L). Figure 4. The change of bromide ion in water samples from different time points (the initial concentration of HBCD is 25.0 mg/L, KPS dosage is 4 mg/L, and TiO2 dosage is 100 mg/L). Figure 5. The change of pH over time (the initial concentration of HBCD is 25.0 mg/L, KPS dosage is 4 mg/L, and TiO2 dosage is 100 mg/L). Figure 5. The change of pH over time (the initial concentration of HBCD is 25.0 mg/L, KPS dosage is 4 mg/L, and TiO2 dosage is 100 mg/L). 2.5. The Mechanism of Photodegradation of HBCD 2.5. The Mechanism of Photodegradation of HBCD Figure 6 shows the mass spectrum of the intermediates obtained by GC-MS analysis. The solution was sampled during the degradation of HBCD in the UV/TiO2/KPS system at 90 min, with the reaction conditions the same as mentioned above. The mass spectrum of degradation products were tetrabromocyclododecene (A), dibromocyclododecadiene (B), 1,5,9-Cyclododecatriene (C), 1,2-Epoxy-5,9-cyclododecadiene (D), dibromo-epoxy-cyclododecene (E), 4,5-dibromooctanedioic acid (F), and succinic acid (G), respectively [31–33]. 182 Catalysts 2019, 9, 189 Catalysts 2019, 9, 189 Figure 6. Cont. Figure 6. Cont. Figure 6. Cont. 183 Catalysts 2019, 9, 189 Figure 6. The mass spectrum of the intermediates in the UV/TiO2/KPS system at 90 min ((A) Tetrabromocyclododecene; (B) Dibromocyclododecadiene; (C) 1,5,9-Cyclododecatriene; (D) 1,2-Epoxy-5,9-cyclododecadiene; (E) Dibromo-epoxy-cyclododecene; (F) 4,5-dibromooctanedioic acid; (G) succinic acid). Figure 6. The mass spectrum of the intermediates in the UV/TiO2/KPS system at 90 min ((A) Tetrabromocyclododecene; (B) Dibromocyclododecadiene; (C) 1,5,9-Cyclododecatriene; (D) 1,2-Epoxy-5,9-cyclododecadiene; (E) Dibromo-epoxy-cyclododecene; (F) 4,5-dibromooctanedioic acid; (G) succinic acid). Figure 6. The mass spectrum of the intermediates in the UV/TiO2/KPS system at 90 min ((A) Tetrabromocyclododecene; (B) Dibromocyclododecadiene; (C) 1,5,9-Cyclododecatriene; (D) 1,2-Epoxy-5,9-cyclododecadiene; (E) Dibromo-epoxy-cyclododecene; (F) 4,5-dibromooctanedioic acid; (G) succinic acid). By analyzing the degradation products of GC/MS, the possible degradation pathway of HBCD in UV/TiO2/KPS system is determined, as shown in Figure 7. Under the action of active free radicals, two adjacent C–Br bonds in the molecular structure of HBCD undergo cleavage and debromination to form carbon-carbon double bonds, thus the compounds A, B, and C were obtained successively [34]. Compound C can be directly oxidized to D or oxidized to G by double bond cleavage. In addition, the compound B can also be oxidized to form the compound E, or oxidized to F and G by double bond cleavage [35]. The compound F can also be further debrominated and oxidized to form G. Succinic acid (G) is a small molecule, and it can be easily degraded by free radicals (like SO4•−and OH•) into carboxylic acids, CO2, and H2O in the following time. The adsorption of intermediate species on the surface of TiO2 may cover the active sites, which may result in a decrease in catalytic efﬁciency. But in this work, the initial concentration of HBCD is very low (25 mg/L), and under the irradiation of UV irradiation, the surface of the titanium dioxide is hydrophilic, so the organic intermediates are more easily dispersed into the water-methanol mixed solution under strong stirring. 184 Catalysts 2019, 9, 189 Figure 7. 3. Materials and Methods 3.1. Reagents 3.1. Reagents Ethanol (HPLC grade) and acetonitrile (HPLC grade) were supplied by LABSCIENCE (Reno, NV, USA) and TEDIA (Nashville, TN, USA), respectively. HBCD (99.0%), sodium carbonate (99.8%), sodium bicarbonate (99.5%), sodium nitrite (99.0%), potassium persulfate (KPS, 99%), dichloromethane (HPLC grade), anhydrous sodium sulfate (99.0%) and methanol (HPLC grade) were supplied by Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). All reagents were used as received without further puriﬁcation. TiO2 nanoparticles were laboratory-made, as described in the previous literature [26]. The preparation of HBCD stock solution was as follows: accurately weigh 0.05 g of HBCD powder into 100 mL volumetric ﬂasks, and add chromatographically pure methanol to the 100 mL mark. After dissolving, the HBCD stock solution with the concentrations of 500.00 mg/L was obtained, and then it was stored in a refrigerator at 4 ◦C for later use. The HBCD stock solution was diluted by ultrapure water to different concentrations for drawing the peak area-concentration standard curve, and it also be used as pollutants in the photodegradation experiments. 2.5. The Mechanism of Photodegradation of HBCD Possible pathways of degradation of HBCD in UV/TiO2/KPS systems. Figure 7. Possible pathways of degradation of HBCD in UV/TiO2/KPS systems. 3.2. Photodegradation of HBCD Quickly take 2 mL of the sample at intervals of 30 min, place it in a tube containing 2 mL of methanol, mix well by shaking, and ﬁlter through a 0.22 μm ﬁlter. The ﬁltrate was loaded into the sample vial for analysis in a liquid chromatograph. The average of 3 parallel determinations was taken as the concentration of each sample. For the measurement of bromide ion concentration, a sample solution was quenched right after the sample was taken out by using a same volume of 0.2 M sodium nitrite solution. Then supernatant and the TiO2 nanoparticles were separated in the same method. The supernatant was used to measure the concentration of bromide ion. 3.3. Analysis Method Concentrations of HBCD were measured by a high-performance liquid chromatography (HPLC, LC-20AD, Shimadzu, Kyoto, Japan) instrument equipped with UV–vis detector set at 210 nm. The mobile phase was acetonitrile/water (85/15 (v/v)) and the ﬂow rate was maintained at 1.0 mL/min. The HPLC chromatogram of HBCD was shown in Figure S2 (Supporting Information). According to the change of concentration of HBCD before and after degradation of the reaction system, the degradation rate of HBCD was calculated. The calculation was as follows: ηHBCD = C0 −Ct C0 × 100% (7) (7) where C0 represents the initial concentration of HBCD in the reaction system and Ct represents the concentration of HBCD in the system at time t. Concentration of bromide ion was measured by a Diane Ion chromatograph (ICS1100, Dionex, Sunnyvale, CA, USA) with an IonPac AS23 anion analytical column (250 mm × 4.0 mm × 5 μm, Dionex, Sunnyvale, CA, USA) and a Dionex IonPac AG22 anion protective column (50 mm × 4 mm, Dionex, Sunnyvale, CA, USA). The peak area-concentration standard curve of Br ion was plotted using potassium bromide powder as the bromine source. Leaching solution was 4.5 mM Na2CO3 and 1.4 mM NaHCO3 with a ﬂow rate of 1.2 mL/min. The intermediates were qualitatively analyzed by a gas chromatography/mass spectrometry (GC/MS, Shimadzu QP2010 plus). The inlet temperature of GC is 200 degrees with the column type of DB-5MS capillary column (30 m × 0.25 mm × 0.25 μm, Agilent, Santa Clara, CA, USA). The injection volume is 1 μL, and the carrier gas is high purity nitrogen (99.999%). The ion source temperature, the electron bombardment energy, and the scanning mode of the mass spectrometer is 240 ◦C, 70 eV, and full scan mode (15~500 m/z), respectively. The sample solution was pretreated by extraction by dichloromethane and passed through anhydrous sodium sulfate, then it was concentrated by evaporation to about 1 mL under nitrogen, and passed through a 0.45 μm ﬁlter before the GC/MS analysis. 3.2. Photodegradation of HBCD The photoreactor was supplied by Xujiang Electromechanical Plant (XPA-7, Nanjing, China). For determination of TiO2 dosages, a 50 mL HBCD water-methanol mixture solution with the concentration of 25.00 mg/L was added into a quartz tube, and then TiO2 powder with different dosages (0~400 mg/L) were also added into the tube. For determination of the effect of KPS dosage, a 50 mL HBCD water-methanol mixture solution with the concentration of 25.00 mg/L was added into 185 Catalysts 2019, 9, 189 a quartz tube, and then TiO2 powder with the dosage of 100 mg/L and KPS with different dosages (1~8 mM) were also added into the tube. For the kinetic analysis, the photodegradation of HBCD in the three systems of “UV/TiO2 (TiO2: 100 mg/L)”, “UV/K2S2O8 (KPS: 4 mM)”, and “TiO2 (TiO2: 100 mg/L) + K2S2O8 (KPS: 4 mM)” were performed, respectively. In all of the above experimental systems, after all the reagents were completely added, the mixed solution was placed in dark and stirred for 60 min to allow all the reagents to be uniformly mixed and to achieve adsorption equilibrium between TiO2 particles (if any) and HBCD in the solution system. Then, turn on the cooling water and the 100 W mercury light source to start the photocatalytic reaction. Quickly take 2 mL of the sample at intervals of 30 min, place it in a tube containing 2 mL of methanol, mix well by shaking, and ﬁlter through a 0.22 μm ﬁlter. The ﬁltrate was loaded into the sample vial for analysis in a liquid chromatograph. The average of 3 parallel determinations was taken as the concentration of each sample. a quartz tube, and then TiO2 powder with the dosage of 100 mg/L and KPS with different dosages (1~8 mM) were also added into the tube. For the kinetic analysis, the photodegradation of HBCD in the three systems of “UV/TiO2 (TiO2: 100 mg/L)”, “UV/K2S2O8 (KPS: 4 mM)”, and “TiO2 (TiO2: 100 mg/L) + K2S2O8 (KPS: 4 mM)” were performed, respectively. In all of the above experimental systems, after all the reagents were completely added, the mixed solution was placed in dark and stirred for 60 min to allow all the reagents to be uniformly mixed and to achieve adsorption equilibrium between TiO2 particles (if any) and HBCD in the solution system. Then, turn on the cooling water and the 100 W mercury light source to start the photocatalytic reaction. 4. Conclusions Conﬂicts of Interest: The authors declare no conﬂicts of interest. 4. Conclusions Degradation of HBCD is investigated in the UV/TiO2, UV/KPS, and UV/TiO2/KPS system by measurement of the concentrations of HBCD and bromide ion. HBCD can be almost completely degraded and 74.3% of the total bromine content is achieved in the UV/TiO2/KPS homo/heterogeneous photocatalysis, much more than in the UV/KPS system and the UV/TiO2 186 Catalysts 2019, 9, 189 system. The SO4•−produced in persulphate and OH• radicals produced in TiO2 photocatalysis have synergistic effects in the degradation of HBCD in the UV/TiO2/KPS homo/heterogeneous photocatalysis. The high yield of the concentration of bromide ions in the solutions indicates that fewer intermediates are formed in the UV/TiO2/KPS homo/heterogeneous photocatalysis of HBCD. The efﬁcient UV/TiO2/KPS homo/heterogeneous system would provide great impetus to pollution control and environmental management. system. The SO4•−produced in persulphate and OH• radicals produced in TiO2 photocatalysis have synergistic effects in the degradation of HBCD in the UV/TiO2/KPS homo/heterogeneous photocatalysis. The high yield of the concentration of bromide ions in the solutions indicates that fewer intermediates are formed in the UV/TiO2/KPS homo/heterogeneous photocatalysis of HBCD. The efﬁcient UV/TiO2/KPS homo/heterogeneous system would provide great impetus to pollution control and environmental management. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/2/189/s1, Figure S1: Photocatalysis for HBCD degradation in the UV/TiO2/KPS system with three time cycling uses, Figure S2: The HPLC chromatogram of HBCD. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/2/189/s1, Figure S1: Photocatalysis for HBCD degradation in the UV/TiO2/KPS system with three time cycling uses, Figure S2: The HPLC chromatogram of HBCD. Author Contributions: Conceptualization, Q.L.; methodology, L.W.; formal analysis, L.Z.; investigation, X.F.; resources and data curation, J.L.; writing—original draft preparation, Q.L.; writing—review and editing, L.Z.; and H.X. Author Contributions: Conceptualization, Q.L.; methodology, L.W.; formal analysis, L.Z.; investigation, X.F.; resources and data curation, J.L.; writing—original draft preparation, Q.L.; writing—review and editing, L.Z.; and H.X. Funding: This research was funded by Natural Science Foundation of China (No. 21806101), Natural Science Foundation of Shanghai (Nos. 16ZR1412600, 15ZR1401200), Gaoyuan Discipline of Shanghai-Environmental Science and Engineering (Resource Recycling Science and Engineering), Innovation Research Grant (13YZ130) and Leading Academic Discipline Project (J51803) from the Shanghai Education Committee. Funding: This research was funded by Natural Science Foundation of China (No. 21806101), Natural Science Foundation of Shanghai (Nos. 16ZR1412600, 15ZR1401200), Gaoyuan Discipline of Shanghai-Environmental Science and Engineering (Resource Recycling Science and Engineering), Innovation Research Grant (13YZ130) and Leading Academic Discipline Project (J51803) from the Shanghai Education Committee. References 1. Almughamsi, H.; Whalen, M.M. Hexabromocyclododecane and tetrabromobisphenol A alter secretion of interferon gamma (IFN-gamma) from human immune cells. Arch. Toxicol. 2016, 90, 1695–1707. [CrossRef] [PubMed] 2. Alaee, M.; Arias, P.; Sjodin, A.; Bergman, A. An overview of commercially used brominated ﬂame retardants, their applications, their use patterns in different countries/regions and possible modes of release. Environ. Int. 2003, 29, 683–689. [CrossRef] 3. Jeannerat, D.; Pupier, M.; Schweizer, S.; Mitrev, Y.N.; Favreau, P.; Kohler, M. Discrimination of hexabromocyclododecane from new polymeric brominated ﬂame retardant in polystyrene foam by nuclear magnetic resonance. Chemosphere 2016, 144, 1391–1397. [CrossRef] [PubMed] 4. Li, L.; Weber, R.; Liu, J.G.; Hu, J.X. Long-term emissions of hexabromocyclododecane as a chemical of concern in products in China. Environ. Int. 2016, 91, 291–300. [CrossRef] 5. Hunziker, R.W.; Gonsior, S.; Macgregor, J.A.; Desjardins, D. Fate and effect of hexabromocyclododecane in the environment. Organohalogen Compd. 2004, 66, 2300–2305. 6. Stiborova, H.; Vrkoslavova, J.; Pulkrabova, J.; Poustka, J.; Hajslova, J.; Demnerova, K. Dynamics of brominated ﬂame retardants removal in contaminated wastewater sewage sludge under anaerobic conditions. Sci. Total Environ. 2015, 533, 439–445. [CrossRef] [PubMed] 7. Wagoner, E.R.; Baumberger, C.P.; Peverly, A.A.; Peters, D.G. Electrochemical reduction of 1, 2, 5, 6, 9, 10-hexabromocyclododecane at carbon and silver cathodes in dimethylformamide. J. Electroanal. Chem. 2014, 713, 136–142. [CrossRef] 8. Zhang, K.; Huang, J.; Wang, H.; Liu, K.; Yu, G.; Deng, S.B.; Wang, B. Mechanochemical degradation of hexabromocyclododecane and approaches for the remediation of its contaminated soil. Chemosphere 2014, 116, 40–45. [CrossRef] [PubMed] 9. Takigami, H.; Watanabe, M.; Kajiwara, N. Destruction behavior of hexabromocyclododecanes during incineration of solid waste containing expanded and extruded polystyrene insulation foams. Chemosphere 2014, 116, 24–33. [CrossRef] [PubMed] 10. Jondreville, C.; Cariou, R.; Meda, B.; Dominguez-Romero, E.; Omer, E.; Dervilly-Pinel, G.; Le Bizec, B.; Travel, A.; Baeza, E. Accumulation of a-hexabromocyclododecane (alpha-HBCDD) in tissues of fast- and slow-growing broilers (Gallus domesticus). Chemosphere 2017, 178, 424–431. [CrossRef] 11. Guo, Y.G.; Lou, X.Y.; Xiao, D.X.; Xu, L.; Wang, Z.H.; Liu, J.S. Sequential reduction-oxidation for photocatalytic degradation of tetrabromobisphenol A: Kinetics and intermediates. J. Hazard. Mater. 2014, 241–242, 301–306. [CrossRef] [PubMed] 187 Catalysts 2019, 9, 189 12. Saien, J.; Ojaghloo, Z.; Soleymani, A.R.; Rasoulifard, M.H. Homogeneous and heterogeneous AOPs for rapid degradation of Triton X-100 in aqueous media via UV light, nano titania hydrogen peroxide and potassium persulfate. Chem. Eng. J. 2011, 167, 172–182. [CrossRef] 13. Salari, D.; Niaei, A.; Aber, S.; Rasoulifard, M.H. References The photooxidative destruction of CI basic yellow 2 using UV/S2O82−process in a rectangular continuous photoreactor. J. Hazard. Mater. 2009, 166, 61–66. [CrossRef] [PubMed] 14. Zhang, L.; Zhang, Q.H.; Xie, H.Y.; Guo, J.; Lyu, H.L.; Li, Y.G.; Sun, Z.G.; Wang, H.Z.; Guo, Z.H. Electrospun titania nanoﬁbers segregated by graphene oxide for improved visible light photocatalysis. Appl. Catal. B Environ. 2017, 201, 470–478. [CrossRef] 15. Zhang, L.; Li, Y.G.; Xie, H.Y.; Wang, H.Z.; Zhang, Q.H. Efﬁcient mineralization of toluene by W-doped TiO2 nanoﬁbers under visible light irradiation. J. Nanosci. Nanotechnol. 2015, 15, 2944–2951. [CrossRef] 16. Zhang, L.; Li, Y.G.; Zhang, Q.H.; Shi, G.Y.; Wang, H.Z. Fast synthesis of highly dispersed anatase TiO2 nanocrystals in a microﬂuidic reactor. Chem. Lett. 2011, 40, 1371–1373. [CrossRef] y 17. Aronne, A.; Fantauzzi, M.; Imparato, C.; Atzei, D.; De Stefano, L.; D′Errico, G.; Sannino, F.; Rea, I.; Pirozzi, D.; Elsener B ; et al Electronic properties of TiO2-based materials characterized by high Ti3+ self-doping and 17. Aronne, A.; Fantauzzi, M.; Imparato, C.; Atzei, D.; De Stefano, L.; D′Errico, G.; Sannino, F.; Rea, I.; Pirozzi, D.; Elsener, B.; et al. Electronic properties of TiO2-based materials characterized by high Ti3+ self-doping and low recombination rate of electron–hole pairs. RSC Adv. 2017, 7, 2373–2381. [CrossRef] 18. Sannino, F.; Pernice, P.; Imparato, C.; Aronne, A.; D′Errico, G.; Minieri, L.; Perfetti, M.; Pirozzi, D. Hybrid TiO2–acetylacetonate amorphous gel-derived material with stably adsorbed superoxide radical active in oxidative degradation of organic pollutants. RSC Adv. 2015, 5, 93831–93839. [CrossRef] 19. Sannino, F.; Pernice, P.; Minieri, L.; Gamandona, G.A.; Aronne, A.; Pirozzi, D. Oxidative Degradation of Different Chlorinated Phenoxyalkanoic Acid Herbicides by a Hybrid ZrO2 Gel-Derived Catalyst without Light Irradiation. ACS Appl. Mater. Interfaces 2015, 7, 256–263. [CrossRef] 20. Guo, Y.G.; Zhou, J.; Lou, X.Y.; Liu, R.L.; Xiao, D.X.; Fang, C.L.; Wang, Z.H.; Liu, J.S. Enhanced degradation of Tetrabromobisphenol A in water by a UV/base/persulfate system: Kinetics and intermediates. Chem. Eng. J. 2014, 254, 538–544. [CrossRef] 21. Neta, P.; Huie, R.E.; Ross, A.B. Rate constants of inorganic radicals in aqueous-solution. J. Phys. Chem. Ref. Data 1988, 17, 1027–1284. [CrossRef] 22. Xu, J.; Meng, W.; Zhang, Y.; Lei, L.; Guo, C.S. Photocatalytic degradation of tetrabromobisphenol A by mesoporous BiOBr: Efﬁcacy, products and pathway. Appl. Catal. B Environ. 2011, 107, 355–362. [CrossRef] 23. Zhao, J.Y.; Zhang, Y.B.; Quan, X.; Chen, S. Enhanced oxidation of 4-chlorophenol using sulfate radicals generated from zero-valent iron and peroxydisulfate at ambient temperature. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 32. Zhou, D.N.; Chen, L.; Wu, F.; Wang, J.; Yang, F. Debromination of hexabromocyclododecane in aqueous solutions by UV-C irradiation. Fresenius Environ. Bull. 2012, 21, 107–111. References Sep. Purif. Technol. 2010, 71, 302–307. [CrossRef] 24. Li, Q.; Wang, L.F.; Zhang, L.; Xie, H.Y. Rapid degradation of tetrabromobisphenol A under the UV/TiO2/KPS systems in alkaline aqueous solutions. Res. Chem. Intermed. 2018. [CrossRef] 25. Ahmadi, M.; Ghanbari, F.; Moradi, M. Photocatalysis assisted by peroxymonosulfate and persulfate for benzotriazole degradation: Effect of pH on sulfate and hydroxyl radicals. Water Sci. Technol. 2015, 72, 2095–2102. [CrossRef] [PubMed] 26. Xie, H.Y.; Zhu, L.P.; Wang, L.L.; Chen, S.W.; Yang, D.D.; Yang, L.J.; Gao, G.L.; Yuan, H. Photodegradation of benzene by TiO2 nanoparticles prepared by ﬂame CVD process. Particuology 2011, 9, 75–79. [CrossRef] 27. Yang, H.; Zhou, S.L.; Yin, M.L.; Pi, L.L.; Zeng, J.; Yi, B. Parameters effect on photocatalytic kinetics of carbofuran in TiO2 aqueous solution. China Environ. Sci. 2013, 33, 82–87. 28. Varanasi, L.; Coscarelli, E.; Khaksari, M.; Mazzoleni, L.R.; Minakata, D. Transformations of dissolved organic matter induced by UV photolysis, hydroxyl radicals, chlorine radicals, and sulfate radicals in aqueous-phase UV-Based advanced oxidation processes. Water Res. 2018, 135, 22–30. [CrossRef] [PubMed] 29. Li, W.; Jain, T.; Ishida, K.; Liu, H.Z. A mechanistic understanding of the degradation of trace organic contaminants by UV/hydrogen peroxide, UV/persulfate and UV/free chlorine for water reuse. Environ. Sci. Water Res. Technol. 2017, 3, 128–138. [CrossRef] 30. Elmolla, E.S.; Chaudhuri, M. Degradation of amoxicillin, ampicillin and cloxacillin antibiotics in aqueous solution by the UV/ZnO photocatalytic process. J. Hazard. Mater. 2010, 173, 445–449. [CrossRef] [PubMed] 31. Barontini, F.; Cozzani, V.; Cuzzola, A.; Petarca, L. Investigation of hexabromocyclododecane thermal degradation pathways by gas chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 2001, 15, 690–698. [CrossRef] [PubMed] 188 Catalysts 2019, 9, 189 32. Zhou, D.N.; Chen, L.; Wu, F.; Wang, J.; Yang, F. Debromination of hexabromocyclododecane in aqueous solutions by UV-C irradiation. Fresenius Environ. Bull. 2012, 21, 107–111. y 33. Yu, Y.; Zhou, D.; Wu, F. Mechanism and products of the photolysis of hexabromocyclododecane in acetonitrile–water solutions under a UV-C lamp. Chem. Eng. J. 2015, 281, 892–899. [CrossRef] 34. Zhao, Y.Y.; Zhang, X.H.; Sojinu, O.S. Thermodynamics and photochemical properties of alpha, beta, and gamma-hexabromocyclododecanes: A theoretical study. Chemosphere 2010, 80, 150–156. 35. Tso, C.P.; Shih, Y.H. The transformation of hexabromocyclododecane using zerovalent iron nanoparticle aggregates. J. Hazard. Mater. 2014, 277, 76–83. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 32. Zhou, D.N.; Chen, L.; Wu, F.; Wang, J.; Yang, F. Debromination of hexabromocyclododecane in aqueous solutions by UV-C irradiation. Fresenius Environ. Bull. 2012, 21, 107–111. 33. Yu, Y.; Zhou, D.; Wu, F. Mechanism and products of the photolysis of hexabromocyclododecane in acetonitrile–water solutions under a UV-C lamp. Chem. Eng. J. 2015, 281, 892–899. [CrossRef] 34. Zhao, Y.Y.; Zhang, X.H.; Sojinu, O.S. Thermodynamics and photochemical properties of alpha, beta, and gamma-hexabromocyclododecanes: A theoretical study. Chemosphere 2010, 80, 150–156. 35. Tso, C.P.; Shih, Y.H. The transformation of hexabromocyclododecane using zerovalent iron nanoparticle aggregates. J. Hazard. Mater. 2014, 277, 76–83. [CrossRef] [PubMed] y 33. Yu, Y.; Zhou, D.; Wu, F. Mechanism and products of the photolysis of hexabromocyclododecane in acetonitrile–water solutions under a UV-C lamp. Chem. Eng. J. 2015, 281, 892–899. [CrossRef] p g 34. Zhao, Y.Y.; Zhang, X.H.; Sojinu, O.S. Thermodynamics and photochemical properties of alpha, beta, and gamma-hexabromocyclododecanes: A theoretical study. Chemosphere 2010, 80, 150–156. g y y 35. Tso, C.P.; Shih, Y.H. The transformation of hexabromocyclododecane using zerovalent iron nanoparticle aggregates. J. Hazard. Mater. 2014, 277, 76–83. [CrossRef] [PubMed] Article Photocatalytic Degradation of Microcystins by TiO2 Using UV-LED Controlled Periodic Illumination Keywords: Microcystis aeruginosa; microcystin; controlled periodic illumination; titanium dioxide; advanced oxidation process Article Photocatalytic Degradation of Microcystins by TiO2 Using UV-LED Controlled Periodic Illumination Olivia M. Schneider 1,†, Robert Liang 1,2,,†, Leslie Bragg 3, Ivana Jaciw-Zurakowsky 1, Azar Fattahi 1, Shasvat Rathod 1, Peng Peng 4, Mark R. Servos 3 and Y. Norman Zhou 1,2 1 Centre for Advanced Materials Joining, Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada; omschnei@edu.uwaterloo.ca (O.M.S.); ivana.jaciwzurakowsky@edu.uwaterloo.ca (I.J.-Z.); azar.fattahi@uwaterloo.ca (A.F.); s2rathod@edu.uwaterloo.ca (S.R.); nzhou@uwaterloo.ca (Y.N.Z.) 1 Centre for Advanced Materials Joining, Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada; omschnei@edu.uwaterloo.ca (O.M.S.); ivana.jaciwzurakowsky@edu.uwaterloo.ca (I.J.-Z.); azar.fattahi@uwaterloo.ca (A.F.); s2rathod@edu.uwaterloo.ca (S.R.); nzhou@uwaterloo.ca (Y.N.Z.) 1 Centre for Advanced Materials Joining, Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada; omschnei@edu.uwaterloo.ca (O.M.S.); ivana.jaciwzurakowsky@edu.uwaterloo.ca (I.J.-Z.); azar.fattahi@uwaterloo.ca (A.F.); s2rathod@edu.uwaterloo.ca (S.R.); nzhou@uwaterloo.ca (Y.N.Z.) 2 Waterloo Institute of Nanotechnology, University of Waterloo, Waterloo, ON N2L 3G1, Canada 3 Department of Biology, University of Waterloo, Waterloo, ON N2L 3G1, Canada; leslie.bragg@uwaterloo.ca (L.B.); mservos@uwaterloo.ca (M.R.S.) 4 School of Mechanical Engineering and Automation, International Research Institute for Multidisciplinary Science, Beihang University, 37 Xueyuan Rd, Beijing 100191, China; peng.peng@uwaterloo.ca Correspondence: rliang@uwaterloo.ca; Tel.: +1-519-888-4567 (ext. 33326) 4 School of Mechanical Engineering and Automation, International Research Institute for Multidisciplina Science, Beihang University, 37 Xueyuan Rd, Beijing 100191, China; peng.peng@uwaterloo.ca * Correspondence: rliang@uwaterloo.ca; Tel.: +1-519-888-4567 (ext. 33326) † Denotes equal contribution. † Denotes equal contribution. Received: 18 January 2019; Accepted: 11 February 2019; Published: 14 February 2019 Received: 18 January 2019; Accepted: 11 February 2019; Published: 14 February 2019 Abstract: Toxic microcystins (MCs) produced by freshwater cyanobacteria such as Microcystis aeruginosa are of concern because of their negative health and economic impacts globally. An advanced oxidation process using UV/TiO2 offers a promising treatment option for hazardous organic pollutants such as microcystins. The following work details the successful degradation of MC-LA, MC-LR, and MC-RR using a porous titanium–titanium dioxide (PTT) membrane under UV-LED light. Microcystin quantitation was achieved by sample concentration and subsequent LC–MS/MS analysis. The PTT membrane offers a treatment option that eliminates the need for the additional ﬁltration or separation steps required for traditional catalysts. Controlled periodic illumination was successfully used to decrease the total light exposure time and improve the photonic efﬁciency for a more cost-effective treatment system. Individual degradation rates were inﬂuenced by electrostatic forces between the catalyst and differently charged microcystins, which can potentially be adjusted by modifying the solution pH and the catalyst’s isoelectric point. Catalysts 2019, 9, 181; doi:10.3390/catal9020181 catalysts catalysts References 189 1. Introduction Cyanobacteria are a phylum of phototrophic bacteria capable of producing toxic blooms. Microcystis aeruginosa is a common freshwater cyanobacteria which produces microcystins (MC), a group of cyanotoxins with strong hepatotoxic effects. With the increased eutrophication of freshwater resources worldwide, the prevalence of these toxic blooms is a growing concern [1–3]. Over the last several years they have been linked to a variety of both animal and human poisonings globally, including in Canada, Australia, the United Kingdom, China, and Africa [4]. The World Health Organization deems the acceptable level of microcystin-LR, the most common microcystin, in drinking water to be 1 μg L−1 [4]. For example, a review document produced by Health Canada in 2016 concluded that toxic algal blooms impact drinking water safety in the majority of Canadian provinces [5]. These toxic blooms also cause signiﬁcant economic losses in affected areas by impeding tourism and ﬁshing, lowering property values, and requiring expensive preventative strategies and monitoring [2,6,7]. Catalysts 2019, 9, 181; doi:10.3390/catal9020181 www.mdpi.com/journal/catalysts 191 Catalysts 2019, 9, 181 Catalysts 2019, 9, 181 Recent studies show that advanced oxidation processes (AOPs) such as ultraviolet (UV) light/TiO2 photocatalysis can break down microcystins [8–12]. When TiO2 particles are irradiated by UV light, electrons in the valence band are excited to the conduction band, creating electron–hole pairs. These electron–hole pairs can either undergo redox reactions directly with small organic molecules or with water to form reactive oxygen species (ROS) such as hydroxyl radicals, which then participate in redox reactions with small organic molecules [13–20]. UV/TiO2 photocatalysis is a promising AOP for treating microcystins in water because the TiO2 is catalytic, providing a constant oxidant source without having to be replenished, as is the case in UV/H2O2 AOPs. Traditional slurry TiO2 reactors are impractical for water treatment because they require an additional ﬁltration step to remove the catalyst. In order to address this issue, a porous titanium–titanium dioxide membrane was used in this study. The membrane is made of a porous titanium sheet, oxidized, and calcinated to produce TiO2 structures on its surface. In order to show the hydroxyl radical production of the catalyst under UV illumination, the conversion of terephthalic acid (TPA) to 2-hydroxyterephthalic acid (HTPA) was quantiﬁed by ﬂuorescence [16]. The trade-off of using the membrane is that it has less surface area than the equivalent mass of powder TiO2, decreasing adsorption and lowering the degradation rate [21,22]. 2. Results and Discussion 2. Results and Discussion 1. Introduction Previous work also suggests that the photonic efﬁciency of the process can be improved using controlled periodic illumination (CPI) [19,23,24]. The improved efﬁciency under CPI conditions can be compared to a phenomenon called Parrondo’s paradox, where alternating two less favorable conditions results in a more favorable outcome. The improved efﬁciency, when using a catalyst such as the porous titanium–titanium dioxide (PTT) membranes, can be explained by mass-transfer limitations. Because the membrane has a limited surface area for adsorption, the rate of the reaction may be limited by this surface area at high LED duty cycles [25]. In this case, periodically illuminated lighting conditions (within the mass-transfer limit) will be equally effective. In the case of photon-limited reactions (for example, slurry reactors), mass-transfer limitations would not apply because the reaction rate is faster than the adsorption rate. Typical light sources such as mercury and xenon lamps require mechanical shutters to produce CPI, take time to warm up, and lack efﬁciency and reliability. UV-LEDs are ideal light sources in this case because high-frequency CPI can be generated using pulse-width modulation (PWM) through a microcontroller. The microcontroller also allows for easy optimization of the light frequency, including the implementation of dual-frequency proﬁles that may exhibit a synergistic effect. By decreasing the cumulative light exposure time in UV/TiO2 photocatalysis, the efﬁciency of the system and the life span of the light source can be increased without sacriﬁcing performance. A more energy-efﬁcient system would be particularly advantageous for the practical application of larger-scale water treatment. The following study details the removal of MC-LA, MC-LR, and MC-RR from water using UV/TiO2 photocatalytic degradation with porous titanium–titanium dioxide membranes. Degradation under each set of conditions, for individual and cumulative microcystin concentrations, was monitored using LC–MS/MS. Controlled periodic illumination at frequencies of 50, 5, 0.5, and dual 0.05 and 25 Hz were all considered, with the goal of improving the photonic efﬁciency of the AOP. 2.2. TPA Conversion 2.2. TPA Conversion A summary of the TPA conversion under continuous UV, 0.5, 25, and dual 0.05 and 25 Hz can be seen in Figure 1. As expected, k1 was highest for continuous UV light because the cumulative UV-LED exposure was twice that of the other lighting conditions. In order to properly compare degradation relative to the electrical energy consumed, electrical energy per order (EEO) was calculated. Although the dual lighting conditions had the lowest degradation rates, they also had the lowest EEO, making dual lighting the most efﬁcient set of conditions for TPA conversion. The other two frequencies tested also had lower EEO values than that for continuous light, making continuous UV-LED exposure the least efﬁcient of the four conditions tested for TPA conversion. These results are in agreement with Parrondo’s paradox and show that CPI is a viable method for improving the efﬁciency of the photocatalytic AOP. Figure 1. (a) Rate constants and (b) electrical energy per order for terephthalic acid (TPA) conversion under various UV-LED conditions. Figure 1. (a) Rate constants and (b) electrical energy per order for terephthalic acid (TPA) conversion under various UV-LED conditions. 2.1. PTT Membrane Characterization 2.1. PTT Membrane Characterization The PTT membrane characterization results have been described in previous works [16,17]. In summary, Raman spectra and XRD indicated mainly anatase TiO2, with some rutile TiO2 and titanium. The experimental band-gap energy of 3.0 eV also indicated that the PTT membranes were primarily composed of crystalline TiO2. The isoelectric point of the PTT membrane was also determined to be 6.0 using a SurPASS™electrokinetic analyzer. 192 Catalysts 2019, 9, 181 Catalysts 2019, 9, 181 Catalysts 2019, 9, 181 2.4. Degradation of Microcystins under CPI Using UV-LED PWM, the following CPI conditions were examined: 0.5, 5, 50, and dual 0.05 and 25 Hz. The calculated EEO for these conditions, as well as those for continuous UV light for comparison, can be seen in Figure 2. Because MC-LA degraded preferentially to MC-LR and MC-RR, the change in concentration of MC-LR and MC-RR was subtle and lacked linearity (see Table S1 and Figure S1). For this reason, there were signiﬁcantly larger errors associated with their calculated EEO. As a result, the cumulative microcystin EEO was considered when comparing different lighting conditions. As predicted by the TPA results, several of the controlled periodic illumination conditions presented more energy efﬁcient options. Both the 5 and 0.5 Hz UV lighting had signiﬁcantly lower EEO than the continuous UV lighting. Dual frequency lighting, which is an equal combination of 0.05 and 25 Hz, had a comparable EEO to continuous UV. Among the lighting conditions tested, the least degradation overall was observed at 50 Hz (cumulative kapp = −7.83 × 10−4 min−1, see Table S1). Very little degradation of MC-LA and MC-RR occurred, and no degradation of MC-LR occurred (see Figures S1 and S2). Given the low magnitude of degradation and the associated margin of error, it is difﬁcult to discern any trend in degradation between different microcystins under 50 Hz UV illumination, as was done for continuous UV. This signiﬁcant decrease in overall degradation resulted in a correspondingly high EEO for microcystin degradation under 50 Hz UV. In general, the EEO increased as the frequency increased, meaning lower-frequency lighting conditions were more efﬁcient. In dual lighting, the combination of both high and low frequencies balanced each other and resulted in an insigniﬁcant net change relative to continuous UV. Interestingly, the improvement of reaction efﬁciency under CPI agrees with Parrondo’s paradox. In Parrondo’s paradox, alternating between two less favorable conditions yields a more favorable result. In this case, the two less favorable UV-LED conditions were off (which did not contribute to the AOP) and on (which was inefﬁcient). By alternating these two conditions at different frequencies, the time the UV-LED was on decreased by 50%. The favorable outcome was that the EEO under CPI decreased, demonstrating improved efﬁciency. Although the PTT membrane provided a more practical option for water treatment, the reduced surface area relative to a slurry reactor imposed mass-transfer limitations. 2.3. Degradation of Microcystins under Continuous Light Experiments testing the membrane under dark conditions and testing UV exposure without the membrane showed no degradation of the microcystins (see the Supplementary Information). This veriﬁes that the photocatalytic AOP was responsible for the microcystin degradation. All three microcystins showed successful degradation when treated with UV light and PTT membranes, with a cumulative rate of −0.00453 min−1. MC-LA degraded the fastest, with a rate of −0.00841 min−1. Both MC-LR and MC-RR degraded at less than half that rate (−0.00350 min−1 and −0.00332 min−1, respectively). This difference in the degradation rate can be explained by the difference in adsorption, which is inﬂuenced by relative charge. The unmodiﬁed experimental pH was determined to be 5 and remained consistent throughout the course of the experiment. At this experimental pH, each microcystin has a different charge depending on the variable amino acids in the ring structure. A summary of these charges is included in Table 1. At experimental pH the PTT membrane is positively charged. These relative charges can explain the signiﬁcant difference seen between degradation rates for MC-LA and MC-LR or MC-RR. In solution, MC-LA is the most oppositely charged to the PTT membrane and experiences the greatest electrostatic attraction. MC-LR is also negatively charged (though not as strongly) and will experience less signiﬁcant electrostatic attraction. Inversely, MC-RR is positively charged and will experience electrostatic repulsion from the PTT membrane. The increased electrostatic attraction experienced by MC-LA will increase its adsorption onto the PTT membrane and result in a faster degradation rate. Since MC-LR and MC-RR compete with MC-LA for limited adsorption sites on the PTT membrane, the two microcystins that experience less electrostatic attraction will not adsorb as well and will have slower degradation rates. The inﬂuence of these interactions is reﬂected in the relative degradation rates under continuous UV illumination, where MC-LR and MC-RR have degradations rates less than half that of MC-LA. 193 Catalysts 2019, 9, 181 Catalysts 2019, 9, 181 Table 1. Microcystin charge at experimental pH. Compound Charge at pH 5a MC-LA −1.9332 MC-LR −0.9329 MC-RR 0.0567 a Charge was calculated by chemicalize.org. These results are consistent with previous studies by Arlos et al. and Liang et al., which showed that electrostatic forces between the pollutant and catalyst have a signiﬁcant inﬂuence on the degradation rate [16,25]. 2.3. Degradation of Microcystins under Continuous Light The inﬂuence of electrostatic forces demonstrates the importance of considering the pH and the charge of target pollutants when treating water [26]. Degradation rates are highly pH-dependent, so the pH of the water being treated must be considered, especially in practical applications. In future water treatment designs, this information can be used to tune the isoelectric point of the catalyst to improve the degradation of desired pollutants. 3.1. Reagents and Chemicals Microcystins and nodularin (Cayman Chemicals, Ann Arbor, MI, USA) were dissolved in UHPLC-grade methanol (VWR International, Mississauga, ON, Canada) and stored at −20 ◦C. PTT membrane synthesis required hydrogen peroxide (Sigma-Aldrich, St. Louis, MO, USA) and 0.254-mm-thick porous titanium (PTi) sheets (Accumet Materials, Ossining, NY, USA). For LC–MS analysis, HPLC-grade ammonium ﬂuoride (Sigma-Aldrich, St. Louis, MO, USA) and HPLC-grade acetonitrile (Thermo Fisher Scientiﬁc, Waltham, MA, USA) were used. For measuring hydroxyl radical formation, TPA and sodium hydroxide were purchased from Sigma-Aldrich, St. Louis, MO, USA. Water was puriﬁed using a Milli-Q®Integral water puriﬁcation system (EMD Millipore, Burlington, MA, USA) (18.2 mΩ·cm resistivity at 25 ◦C). 2.4. Degradation of Microcystins under CPI Because the rates of adsorption and desorption to the surface of the catalyst were signiﬁcantly slower than the rate of electron–hole pair formation and recombination, the process of adsorption and desorption was rate-limiting [24,27]. This had a signiﬁcant impact on the photonic efﬁciency of the system. In using CPI, the dark period allowed for the equilibration of the untreated pollutant molecules on adsorption sites, without wasting energy [18,23,28]. This improved the photonic efﬁciency of the photocatalytic system. The results indicated that CPI is a viable method for improving the efﬁciency of photocatalytic 194 Catalysts 2019, 9, 181 AOPs used to treat organic pollutants and toxins, though further optimization of conditions is required. Improving the photonic efﬁciency of the process makes it more energy efﬁcient and prolongs the life of the light source. These characteristics are particularly attractive in the water treatment industry because they reduce costs. Figure 2. Electrical energy per order for microcystin degradation under continuous and controlled periodic illumination (CPI) conditions. Figure 2. Electrical energy per order for microcystin degradation under continuous and controlled periodic illumination (CPI) conditions. 3.2. PTT Membrane Synthesis and Characterization PTT membrane synthesis and characterization methods are described in previous works [16,17]. In short, PTi membranes were cut into 5-cm diameter discs, cleaned and oxidized in a hydrogen peroxide solution at 80 ◦C, and then calcined at 600 ◦C. Material characterization methods included micro-Raman spectroscopy (He–Ne laser λ = 632.8 nm, Renishaw, Wotton-under-Edge, UK), scanning electron microscopy (FE-SEM LEO 1550, Carl Zeiss Microscopy, Jena, Germany), and X-ray diffraction (XPERT-PRO, Malvern Panalytical, Malvern, UK). 3.3. Experimental Setup for Microcystin Degradation 3.3. Experimental Setup for Microcystin Degradation A volume of MC-LA, MC-LR, and MC-RR stock solution was evaporated to dryness under nitrogen gas and then reconstituted to 2 μg/L in MilliQ water for a reaction solution. The PTT membranes were suspended in 0.4-L beakers on metal stands, 1.5 cm under the solution surface with a volume of 0.3 L. Many studies of UV/TiO2 photocatalysis use methanol as a carrier solvent when preparing aqueous pollutant solutions [10,16,17,29–35]. More recent studies show that methanol has a signiﬁcant effect on photocatalytic degradation because it acts as a hydroxyl radical scavenger, even at low 195 Catalysts 2019, 9, 181 concentrations [36,37]. In order to replicate the effects of methanol under typical experimental conditions, all reactions were conducted in 0.02% methanol. Reactions took place using a UV-LED source with an average irradiance of 2.18 mW cm−2 under continuous illumination and 1.08 mW cm−2 under a 50% duty cycle (measured 18 cm from the light source using Thorlabs PM100-USB power meter, S120VC 200–1100 nm, 50 mW). Reaction solutions were stirred at 600 rpm. The solution surface was initially 10.5 cm below the light source. A diagram of the reaction setup is shown in Figure 3a with the UV-LED (LED-Engin LZ1, 1000 mA) spectral power distribution peaking at 365 nm and the total radiation included angle of 105◦(90% of the total radiant ﬂux). Figure 3. (a) Experimental setup for UV/TiO2 reactors and (b) UV-LED spectral power distribution and radiation pattern. Abbreviations: PWM—pulse-width modulation, PTT—porous titanium–titanium dioxide. Figure 3. (a) Experimental setup for UV/TiO2 reactors and (b) UV-LED spectral power distribution and radiation pattern. Abbreviations: PWM—pulse-width modulation, PTT—porous titanium–titanium dioxide. Reactions were equilibrated in the dark for 1 h before UV-LED irradiation, with a total reaction time of 6 h. Samples of 4 mL were taken every hour. Each set of conditions was repeated in triplicate. Arduino microcontrollers and LEDSEEDUINO LED current drivers were used to program the UV-LEDs for 0.5, 5, and 50 Hz as well as a dual frequency (0.05 and 25 Hz alternating for equal periods). The pulsed-width modulation script used to program the different conditions can be found in the Supplementary Information. All UV-LED ﬂashing sequences had a duty cycle of 50%. 3.5. Electrical Energy per Order To more accurately compare the efﬁciency of lighting conditions with different duty cycles, electrical energy per order (EEO) was calculated. EEO is the energy in kWh needed to decrease the microcystin or TPA concentration by one order of magnitude in a liter of water. EEO was calculated by the following equations [40,41]: EEO(MC) = 1000·P·t V·log Ci Cf EEO(HTPA) = 1000·P·t V·k2 where P is the power dissipated over the treatment process in kW, t is the reaction time in min, V is the reaction volume in L, k2 is the degradation rate of HTPA, and Ci and Cf are the initial and ﬁnal microcystin concentrations, respectively. where P is the power dissipated over the treatment process in kW, t is the reaction time in min, V is the reaction volume in L, k2 is the degradation rate of HTPA, and Ci and Cf are the initial and ﬁnal microcystin concentrations, respectively. 3.4. Experimental Setup for TPA Conversion The experimental method is derived from previous work [16]. In brief, a solution of 5 mM TPA was made in 6 mM NaOH. Under the same conditions as described in the previous section, 300 mL of this solution was placed in a beaker with a PTT membrane. UV-LED irradiation began after a 1-h dark equilibration, and samples were taken at various time points over 4 h. The following lighting conditions were tested: Continuous, 0.05, 25, and dual 0.05 and 25 Hz. All frequencies were programmed with a duty cycle of 50%. HTPA concentrations were quantiﬁed by ﬂuorescence using a plate reader (SpectraMax M3, Molecular Devices, San Jose, CA, USA) with an excitation wavelength of 315 nm and emission from 350 nm to 550 nm. The intensity value was taken from the peak of the spectrum. HTPA was the ﬁrst degradation product of TPA, so its concentration increased sharply at the beginning of the reaction. As more HTPA was produced, it also degraded into more oxidized products. These rates can be described by the following kinetic model [39]: CHTPA = k1 k2 1 −e−k2t where k1 is the zeroth-order rate of HTPA formation, k2 is the pseudo ﬁrst-order degradation rate of HTPA, CHTPA is the concentration of HTPA in mol L−1, and t is time in minutes. where k1 is the zeroth-order rate of HTPA formation, k2 is the pseudo ﬁrst-order degradation rate of HTPA, CHTPA is the concentration of HTPA in mol L−1, and t is time in minutes. 3.3. Experimental Setup for Microcystin Degradation The duty cycle is the ratio of the time on to the time off, as described by the following equation: D = PW T × 100% D = PW T × 100% where D is the duty cycle expressed as a percentage, PW is the pulse width duration, and T is the period of the wave. where D is the duty cycle expressed as a percentage, PW is the pulse width duration, and T is the period of the wave. At low adsorbate concentrations, the following equation can be used to approximate Langmuir–Hinshelwood kinetics [38]: −r = dC dt = −kappC. The equation can then be rearranged and integrated to give the following [25]: The equation can then be rearranged and integrated to give the following [25]: ln C Co = kappt ln C Co = kappt where C(g L−1) is the analyte concentration at time t (min), Co (g L−1) is the initial analyte concentration at t = 0, and kapp (min−1) is the apparent kinetic rate. The slope of a plot of the equation gives the 196 Catalysts 2019, 9, 181 kapp. OriginLabPro (version 8.0, OriginLab, Northhampton, MA, USA, 2018) was used to complete the linear regression analysis to determine the rates for cumulative and individual compounds. kapp. OriginLabPro (version 8.0, OriginLab, Northhampton, MA, USA, 2018) was used to complete the linear regression analysis to determine the rates for cumulative and individual compounds. 3.4. Experimental Setup for TPA Conversion 4. Conclusions In this study three common microcystins, MC-LA, MC-LR, and MC-RR, were successfully degraded in water using a UV/TiO2 photocatalytic AOP. EEO values calculated for TPA conversion determined continuous UV-LED illumination to be less efﬁcient than all the CPI conditions tested, demonstrating the potential of CPI to improve the efﬁciency of the photocatalytic AOP. Under continuous illumination, the negatively charged MC-LA degraded at more than twice the rate of MC-LR or MC-RR because it preferentially adsorbed onto the positive PTT membrane catalyst. The pH dependence of the degradation rates suggests that the isoelectric point of the catalyst can be tuned to improve the degradation of target pollutants in water of a known pH, given the compound charges. CPI conditions of 0.5, 5, and 50 Hz as well as dual 0.5 and 25 Hz with a 50% duty cycle were also examined for treating microcystins. When considering the cumulative microcystin solution, the 0.5 and 5 Hz CPI conditions were determined to be more efﬁcient than continuous UV light based on the calculated EEO. These results can be explained by mass-transfer limitations, where the rate of adsorption and desorption onto the surface of the catalyst limits photonic efﬁciency. The results of this work indicate that the use of CPI has the potential to improve the energy efﬁciency and light source life span in photocatalytic AOPs and is worth further investigation. Improving these parameters makes photocatalytic AOPs more attractive as a large-scale water treatment solution because they have the potential to greatly decrease costs. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/2/181/s1, Table S1: Calculated degradation rates. Poor ﬁt with the linear regression model due to insigniﬁcant degradation is seen in membrane-only and UV-only conditions, as well as in 50 Hz MC-LR; Figure S1: Change in microcystin concentration over the course of various UV/TiO2 treatments; Figure S2: Linear regression plots for the calculation of degradation rate; Figure S3: Chromatogram demonstrating the separation of MC-LA, MC-LR, MC-RR, and NOD. Author Contributions: Conceptualization, R.L.; methodology, O.M.S. and L.B.; investigation, O.M.S., A.F., and I.J.-Z.; data curation, O.M.S. and S.R.; writing—original draft preparation, O.M.S.; writing—review and editing, O.M.S., A.F., S.R., and R.L.; visualization, R.L.; supervision, Y.N.Z., P.P. and M.R.S.; funding acquisition, Y.N.Z., M.R.S., and R.L. 3.6. Sample Preparation and Analysis Each 4-mL sample was spiked with nodularin (NOD), a toxin similar in structure to microcystins, to 0.8 μg L−1 for an internal standard [1]. The spiked samples were then evaporated to dryness in a Rocket Evaporator (Thermo Scientiﬁc) and reconstituted in 160 μL of UHPLC-grade methanol. Prepared samples were stored at −20 ◦C until analysis. Microcystin quantitation was achieved with LC–MS/MS using an Agilent 1200 HPLC and 3200 quadrupole ion trap (QTRAP) mass spectrometer with electrospray ionization (ABSciex). Speciﬁc mass spectrometry parameters are summarized in Table 2. In order to achieve separation, a Poroshell 120 SB-C18 column (4.6 × 150 mm, 2.7 μm, Agilent Technologies) was used with 0.5 mM ammonium ﬂuoride and acetonitrile (ACN) at 1 mL min−1 and 40 ◦C, with 20-μL sample injections. For the mobile phase gradient, 10% ACN was held for 0.5 min, which was increased to 100% ACN over 4.5 min and held for 1 min. The mobile phase composition was then returned to 10% ACN over 0.5 min and equilibrated for 3.5 min before the next injection. The calibration curves for each microcystin were linear from 0.5 to 500 μg L−1. 197 Catalysts 2019, 9, 181 Table 2. Mass spectrometry parameters for the detection of microcystins. Q1a (Da) Q3b (Da) Time (ms) DPc (volts) EPd (volts) CEe (volts) CXPf (volts) CEPg (volts) Retention Time (min) NOD 825.563 135.3 150 96 12 75 4 40 4.55 MC-LA 911.395 135.2 150 51 12 81 4 36 4.62 MC-LR 995.699 135.1 150 116 12 99 4 36 4.67 MC-RR 519.960 135.2 150 131 7 41 4 26 4.88 a First quadrupole, b third quadrupole, c declustering potential, d entrance potential, e collision energy, f collision cell exit potential, and g collision cell entrance potential. Table 2. Mass spectrometry parameters for the detection of microcystins. 4. Conclusions Funding: This research was funded by the Natural Sciences and Engineering Research Council of Canada (grant number: STPG-494554-2016) through a strategic project grant and the Schwartz–Reisman Foundation through the Waterloo Institute of Nanotechnology—Technion University grant. Acknowledgments: The authors would like to thank the Natural Sciences and Engineering Research Council of Canada, the Schwartz–Reisman Foundation, and Waterloo Institute of Nanotechnology for their ﬁnancial support. Acknowledgments: The authors would like to thank the Natural Sciences and Engineering Research Counci Canada, the Schwartz–Reisman Foundation, and Waterloo Institute of Nanotechnology for their ﬁnancial supp Conﬂicts of Interest: The authors have no conﬂict of interest to declare. Conﬂicts of Interest: The authors have no conﬂict of interest to declare. cyanobacterial blooms and cyanotoxins. Environ. Int. 2013, 59, 303–327. [CrossRef] [PubMed] 3. Umehara, A.; Takahashi, T.; Komorita, T.; Orita, R.; Chio, J.-W.; Takenaka, R.; Mabuchi, R.; Park, H.-D.; Tsutsumi, H. Widespread dispersal and bio-accumulation of toxic microcystins in benthic marine ecosystems. Chemoshpere 2017, 167, 492–500. [CrossRef] [PubMed] References 1. Dawson, R.M. The Toxicology of Microcystins. Toxicon 1998, 36, 953–962. [CrossRef] 2. Merel, S.; Walker, D.; Chicana, R.; Snyder, S.; Baurès, E.; Thomas, O. State of knowledge and concerns on cyanobacterial blooms and cyanotoxins. Environ. Int. 2013, 59, 303–327. [CrossRef] [PubMed] 3. Umehara, A.; Takahashi, T.; Komorita, T.; Orita, R.; Chio, J.-W.; Takenaka, R.; Mabuchi, R.; Park, H.-D.; Tsutsumi, H. Widespread dispersal and bio-accumulation of toxic microcystins in benthic marine ecosystems. Chemoshpere 2017, 167, 492–500. [CrossRef] [PubMed] 198 Catalysts 2019, 9, 181 4. Chambon, P.; Lund, U.; Galal-Gorchev, H.; Ohanian, E. Guidelines for Drinking-Water Quality Volume 2— Health Criteria and Other Supporting Information, 2nd ed.; Kenny, J., Galal-Gorchev, H., Eds.; World Health Organisation: Geneva, Switzerland, 1998; Volume 2. Organisation: Geneva, Switzerland, 1998; Volume 2. 5. Committee on Drinking Water. Cyanobacterial Toxins in Drinking Water; Committee on Drinking Water: Ottawa, ON, Canada, 2016. 6. Wolf, D.; Klaiber, H.A. Bloom and bust: Toxic algae’s impact on nearby property values. Ecol. Econ. 2017, 135, 209–221. [CrossRef] 135, 209–221. [CrossRef] 7. Dyson, K.; Huppert, D.D. Regional economic impacts of razor clam beach closures due to harmful algal 7. Dyson, K.; Huppert, D.D. Regional economic impacts of razor clam beach closures due to harm blooms (HABs) on the Paciﬁc coast of Washington. Harmful Algae 2010, 9, 264–271. [CrossRef] Dyson, K.; Huppert, D.D. Regional economic impacts of razor clam beach closures due to harmful algal blooms (HABs) on the Paciﬁc coast of Washington Harmful Algae 2010 9 264–271 [CrossRef] Dyson, K.; Huppert, D.D. Regional economic impacts of razor clam beach closures due to harmful algal blooms (HABs) on the Paciﬁc coast of Washington. Harmful Algae 2010, 9, 264–271. [CrossRef] 8. Sharma, V.K.; Triantis, T.M.; Antoniou, M.G.; He, X.; Pelaez, M.; Han, C.; Song, W.; O’shea, K.E.; De La Cruz, A.A.; Kaloudis, T.; et al. Destruction of microcystins by conventional and advanced oxidation processes: A review. Sep. Purif. Technol. 2012, 91, 3–17. [CrossRef] 9. Liu, I.; Lawton, L.A.; Bahnemann, D.W.; Liu, L.; Proft, B.; Robertson, P.K.J. The photocatalytic decomposition of microcystin-LR using selected titanium dioxide materials. Chemosphere 2009, 76, 549–553. [CrossRef] [PubMed] 10. Shephard, G.S.; Om, S.S.O.; De Villiers, D.; Engelbrecht, W.J.; E El, G.; Wessels, F.S. Degradation of microcystin toxins in a falling ﬁlm photocatalytic reactor with immobilized titanium dioxide catalyst. Water Res. 2002, 36, 140–146. [CrossRef] 11. Cornish, B.J.P.A.; Lawton, L.A.; Robertson, P.K.J. Hydrogen peroxide enhanced photocatalytic oxidation of microcystin-LR using titanium dioxide. Appl. References Catal. B Environ. 2000, 25, 59–67. [CrossRef] 12. Lawton, L.A.; Robertson, P.K.J.; Cornish, B.J.P.A.; Marr, I.L.; Jaspars, M. Processes inﬂuencing surface interaction and photocatalytic destruction of microcystins on titanium dioxide photocatalysts. J. Catal. 2003, 213, 109–113. [CrossRef] 13. Rizzo, L.; Meric, S.; Guida, M.; Kassinos, D.; Belgiorno, V. Heterogenous photocatalytic degradation kinetics and detoxiﬁcation of an urban wastewater treatment plant efﬂuent contaminated with pharmaceuticals. Water Res. 2009, 43, 4070–4078. [CrossRef] [PubMed] 14. Liang, R.; Hu, A.; Li, W.; Zhou, Y.N. Enhanced degradation of persistent pharmaceuticals found in wastewater treatment efﬂuents using TiO2 nanobelt photocatalysts. J. Nanopart. Res. 2013, 15, 1990. [CrossRef] 15. Martínez, C.; Canle L., M.; Fernández, M.I.; Santaballa, J.A.; Faria, J. Aqueous degradation of diclofenac by heterogeneous photocatalysis using nanostructured materials. Appl. Catal. B Environ. 2011, 107, 110–118. [CrossRef] 16. Arlos, M.J.; Liang, R.; Hatat-Fraile, M.M.; Bragg, L.M.; Zhou, N.Y.; Servos, M.R.; Andrews, S.A. Photocatalytic decomposition of selected estrogens and their estrogenic activity by UV-LED irradiated TiO2 immobilized on porous titanium sheets via thermal-chemical oxidation. J. Hazard. Mater. 2016, 318, 541–550. [CrossRef] [PubMed] 17. Arlos, M.J.; Hatat-Fraile, M.M.; Liang, R.; Bragg, L.M.; Zhou, N.Y.; Andrews, S.A.; Servos, M.R. Photocatalytic decomposition of organic micropollutants using immobilized TiO2 having different isoelectric points. Water Res. 2016, 101, 351–361. [CrossRef] [PubMed] 18. Ku, Y.; Shiu, S.-J.; Wu, H.-C. Decomposition of dimethyl phthalate in aqueous solution by UV–LED/TiO2 process under periodic illumination. J. Photochem. Photobiol. A Chem. 2017, 332, 299–305. [CrossRef] 19. Buechler, K.J.; Nam, C.H.; Zawistowski, T.M.; Noble, R.D.; Koval, C.A. Design and Evaluation of a Novel-Controlled Periodic Illumination Reactor To Study Photocatalysis. Ind. Eng. Chem. Res. 1999, 38, 1258–1263. [CrossRef] 20. Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D.W. Understanding TiO2 Photocatalysis: Mechanisms and Materials. Chem. Rev. 2014, 114, 9919–9986. [CrossRef] 20. Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D.W. Understanding TiO2 Photocatalysis: Mechanisms and Materials. Chem. Rev. 2014, 114, 9919–9986. [CrossRef] 21. Manassero, A.; Satuf, M.L.; Alfano, O.M. Photocatalytic reactors with suspended and immobilized TiO2: Comparative efﬁciency evaluation Chem Eng J 2017 326 29–36 [CrossRef] Understanding TiO2 Photocatalysis: Mechanisms and Materials. Chem. Rev. 2014, 114, 9919–9986. [CrossRef] 21. Manassero, A.; Satuf, M.L.; Alfano, O.M. Photocatalytic reactors with suspended and immobilized TiO2: Comparative efﬁciency evaluation. Chem. Eng. J. 2017, 326, 29–36. [CrossRef] 21. Manassero, A.; Satuf, M.L.; Alfano, O.M. Photocatalytic reactors with suspended and immobilized TiO2: Comparative efﬁciency evaluation. References Chem. Eng. J. 2017, 326, 29–36. [CrossRef] 22. Heged˝us, P.; Szabó-Bárdos, E.; Horváth, O.; Szabó, P.; Horváth, K. Investigation of a TiO2 photocatalyst immobilized with poly(vinyl alcohol). Catal. Today 2017, 284, 179–186. [CrossRef] 23. Sczechowski, J.G.; Koval, C.A.; Noble, R.D. Evidence of critical illumination and dark recovery times for increasing the photoefﬁciency of aqueous heterogeneous photocatalysis. J. Photochem. Photobiol. A Chem. 1993, 74, 273–278. [CrossRef] 24. Tokode, O.; Prabhu, R.; Lawton, L.A.; Robertson, P.K.J. Controlled periodic illumination in semiconductor photocatalysis. J. Photochem. Photobiol. A Chem. 2016, 319–320, 96–106. [CrossRef] 199 Catalysts 2019, 9, 181 25. Liang, R.; Van Leuwen, J.C.; Bragg, L.M.; Arlos, M.J.; Li Chun Fong, L.C.M.; Schneider, O.M.; Peng, P.; Servos, M.R.; Zhou, Y.N. Utilizing UV-LED pulse width modulation on TiO2 advanced oxidation processes to enhance the decomposition efficiency of pharmaceutical micropollutants. Chem. Eng. J. 2019, 361, 439–449. [CrossRef] 26. Friedmann, D.; Mendive, C.; Bahnemann, D. Environmental TiO2 for water treatment: Parameters affectin the kinetics and mechanisms of photocatalysis. Appl. Catal. B Environ. 2010, 99, 398–406. [CrossRef] 26. Friedmann, D.; Mendive, C.; Bahnemann, D. Environmental TiO2 for water treatment: Parameters affecting the kinetics and mechanisms of photocatalysis. Appl. Catal. B Environ. 2010, 99, 398–406. [CrossRef] 27. Memming, R. Photoinduced charge transfer processes at semiconductor electrodes and particles. In Topics in Current Chemistry; Springer: Berlin/Heidelberg, Germany, 1994; Volume 169, pp. 105–181, ISBN 978-3-540-57565-8. 28. Sczechowski, J.G.; Koval, C.A.; Noble, R.D. A Taylor vortex reactor for heterogeneous photocatalysis. Chem. Eng. Sci. 1995, 50, 3163–3173. [CrossRef] 29. Miranda-García, N.; Maldonado, M.I.; Coronado, J.M.; Malato, S. Degradation study of 15 emerging contaminants at low concentration by immobilized TiO2 in a pilot plant. Catal. Today 2010, 151, 107–113. [CrossRef] 30. Miranda-García, N.; Suárez, S.; Sánchez, B.; Coronado, J.M.; Malato, S.; Maldonado, M.I. Photocatalytic degradation of emerging contaminants in municipal wastewater treatment plant efﬂuents using immobilized TiO2 in a solar pilot plant. Appl. Catal. B Environ. 2011, 103, 294–301. [CrossRef] 31. Sun, W.; Li, S.; Mai, J.; Ni, J. Initial photocatalytic degradation intermediates/pathways 17α-ethynylestradiol: Effect of pH and methanol. Chemosphere 2010, 81, 92–99. [CrossRef] [PubMed] 32. Nasuhoglu, D.; Berk, D.; Yargeau, V. Photocatalytic removal of 17α-ethinylestradiol (EE2) and levonorgestrel (LNG) from contraceptive pill manufacturing plant wastewater under UVC radiation. Chem. Eng. J. 2012, 185–186, 52–60. [CrossRef] 33. Kralchevska, R.; Milanova, M.; Bistan, M.; Pintar, A.; Todorovsky, D. The photocatalytic degradation of 17α-ethynylestradiol by pure and carbon nanotubes modiﬁed TiO2 under UVC illumination. Open Chem. Development of TiO2-Carbon Composite Acid Catalyst for Dehydration of Fructose to 5-Hydroxymethylfurfural Morongwa Martha Songo 1,2, Richard Moutloali 2 and Suprakas Sinha Ray 1,2,* 1 DST-CSIR National Centre for Nanostructured Materials, Council for Scientiﬁc and Industrial Research, Pretoria 0001, South Africa; msongo@csir.co.za 2 Department of Applied Chemistry, University of Johannesburg, Doornfontein, Johannesburg 2028, South Africa; rmoutloali@uj.ac.za * Correspondence: rsuprakas@csir.co.za; Tel.: +27-12-841-2388 Received: 28 December 2018; Accepted: 29 January 2019; Published: 31 January 2019 1 DST-CSIR National Centre for Nanostructured Materials, Council for Scientiﬁc and Industrial Research, Pretoria 0001, South Africa; msongo@csir.co.za 2 Department of Applied Chemistry, University of Johannesburg, Doornfontein, Johannesburg 2028, South Africa; rmoutloali@uj.ac.za * Correspondence: rsuprakas@csir.co.za; Tel.: +27-12-841-2388 Received: 28 December 2018; Accepted: 29 January 2019; Published: 31 January 2019 DST CSIR National Centre for Nanostructured Materials, Council for Scientiﬁc and Industrial R Pretoria 0001, South Africa; msongo@csir.co.za 2 Department of Applied Chemistry, University of Johannesburg, Doornfontein, Johannesburg 20 South Africa; rmoutloali@uj.ac.za * Correspondence: rsuprakas@csir.co.za; Tel.: +27-12-841-2388 * Correspondence: rsuprakas@csir.co.za; Tel.: +27-12-841-2388 Received: 28 December 2018; Accepted: 29 January 2019; Published: 31 January 2019 Abstract: A TiO2-Carbon (TiO2C) composite was prepared using the microwave-assisted method and sulfonated using fuming sulfuric acid to produce a TiO2C solid acid catalyst. The prepared solid acid catalyst was characterised using scanning electron microscopy, Brunauer-Emmett-Teller analysis, Fourier transform infrared spectroscopy, and X-ray diffraction. Crystallinity analysis conﬁrmed that TiO2C has an anatase structure, while analysis of its morphology showed a combination of spheres and particles with a diameter of 50 nm. The TiO2C solid acid catalyst was tested for use in the catalytic dehydration of fructose to 5-hydroxymethylfurfural (5-HMF). The effect of reaction time, reaction temperature, catalyst dosage, and solvent were investigated against the 5-HMF yield. The 5-HMF yield was found to be 90% under optimum conditions. The solid acid catalyst is very stable and can be reused for four catalytic cycles. Hence, the material has great potential for use in industrial applications and can be used for the direct conversion of fructose to 5-HMF because of its high activity and high reusability. Keywords: TiO2C composite; acid catalyst; dehydration; fructose; 5-Hydroxymethylfurfural Catalysts 2019, 9, 126; doi:10.3390/catal9020126 www.mdpi.com/journal/catalysts References 2012, 10, 1137–1148. [CrossRef] 34. Marinho, B.A.; de Liz, M.V.; Lopes Tiburtius, E.R.; Nagata, N.; Peralta-Zamora, P.; Iguchi, T.; Kubota, Y.; Fujishima, A. TiO2 and ZnO mediated photocatalytic degradation of E2 and EE2 estrogens. Photochem. Photobiol. Sci. 2013, 12, 678–683. [CrossRef] [PubMed] 35. Fernández, R.L.; McDonald, J.A.; Khan, S.J.; Le-Clech, P. Removal of pharmaceuticals and endocrine disrupting chemicals by a submerged membrane photocatalysis reactor (MPR). Sep. Purif. Technol. 2014, 127, 131–139. [CrossRef] 36. Nosaka, Y.; Nosaka, A.Y. Photocatalysis and Water Puriﬁcation: From Fundamentals to Recent Applications; John Wiley & Sons: Hoboken, NJ, USA, 2013; pp. 3–23. 37. Paul, T.; Miller, P.L.; Strathmann, T.J. Visible-light-mediated TiO2 photocatalysis of ﬂuoroquinolone antibacterial agents. Environ. Sci. Technol. 2007, 41, 4720–4727. [CrossRef] [PubMed] 38. Turchi, C.S.; Ollis, D.F. Photocatalytic degradation of organic w hydroxyl radical attack. J. Catal. 1990, 122, 178–192. [CrossRef] 38. Turchi, C.S.; Ollis, D.F. Photocatalytic degradation of organic water contaminants: Mechanisms involving hydroxyl radical attack. J. Catal. 1990, 122, 178–192. [CrossRef] 39. Cernigoj, U.; Kete, M.; Stangar, U.L. Development of a ﬂuorescence-based method for evaluation of self-cleaning properties of photocatalytic layers. Catal. Today 2010, 151, 46. [CrossRef] 40. Gora, S.; Liang, R.; Zhou, Y.N.; Andrews, S. Settleable engineered titanium dioxide nanomaterials for removal of natural organic matter from drinking water. Chem. Eng. J. 2018, 334, 638–649. [CrossRef] 41. Liang, R.; Li Chun Fong, L.C.M.; Arlos, M.J.; Van Leeuwen, J.; Shahnam, E.; Peng, P.; Servos, M.R.; Zhou, Y.N. Photocatalytic degradation using one-dimensional TiO2 and Ag-TiO2 nanobelts under UV-LED controlled periodic illumination. J. Environ. Chem. Eng. 2017, 5, 4365–4373. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 200 catalysts catalysts 1. Introduction Increase in the usage of fossil fuels for the production of chemicals and energy has caused not only a rise in greenhouse gas emissions into the atmosphere, but also water pollution, thereby contributing to the growing number of serious health issues. These side effects have spurred researchers to look for alternative ways of producing valuable chemicals and energy using sustainable and renewable resources. Biomass has been identiﬁed as a promising resource for the sustainable production of valuable chemical feedstocks and fuels [1]. 5-hydroxymethylfurfural (5-HMF) has been recognised as a key intermediate in the production of biorenewable chemicals [2] and has been classiﬁed as a “platform molecule” because it is an important chemical building block used to derive a variety of chemicals, pharmaceuticals, and furane-based polymers [3]. 5-HMF is usually synthesised from the dehydration of fructose or glucose using homogeneous organic acids or heterogeneous solid acid catalysts [4]. Although the use of homogeneous catalysts produces high fructose conversion and high dehydrations of 5-HMF, there are drawbacks with regard to separation, recycling, and equipment corrosion [5,6]. Heterogeneous acid catalysts such as mesoporous zirconium phosphate [7,8], sulfonated metal oxides [9], magnetite mixed-metal oxides [10], and functionalised silica nanoparticles [11,12] have been used for the dehydration of fructose to 5-HMF. Although the use of some of these materials has promising results, the studies cited pointed out the remaining challenge of ﬁnding catalysts that are highly active, selective, chemically and structurally Catalysts 2019, 9, 126; doi:10.3390/catal9020126 201 Catalysts 2019, 9, 126 stable during repeated use, and do not lead to any side reactions (by-products). Biomass-derived carbonaceous materials were found to be promising candidates for catalytic support application because of their low cost of production and thermal stability [13]. For example, Wang et al. [14] used a sulfonated carbonaceous solid acid catalyst for the dehydration of fructose into 5-HMF in dimethyl sulfoxide (DMSO). Their catalyst was very efﬁcient and effective in that it converted 96.1% of the fructose for a high 5-HMF yield of 93.4%. Zhao et al. [15] further explored a sulfonated carbon sphere solid acid catalyst that converted 100% of the fructose in DMSO solvent at 160 ◦C for 1.5 h to produce 90% 5-HMF. Wang et al. [16] used C-based solid acid catalysts to catalyse dehydration of fructose in DMSO for 1.5 h at 130 ◦C to achieve 91.2% 5-HMF. Guo et al. 1. Introduction [17] used a lignin-derived carbonaceous catalyst to convert fructose into 5-HMF under microwave irradiation in the mixture of DMSO and ionic liquid at 110 ◦C for 10 min for a fructose conversion of 98% and a 5-HMF yieldof 84%. Hu et al. [18] explored the use of a magnetic lignin-derived carbonaceous acid catalyst for the catalytic conversion of fructose into 5-HMF in DMSO solvent. Use of the magnetic C-based catalyst achieved a 100% fructose conversion and a 5-HMF yieldof 81.1% under optimum reaction conditions. Studies have found that the anatase phase of TiO2 can catalyse the dehydration of carbohydrates such as fructose and glucose into 5-HMF and that the performance of the TiO2 catalyst is strongly controlled by its morphological and structural properties [3,19–21]. For example, the use of nanostructured TiO2 remarkably favoured the production of 5-HMF from glucose and fructose [19,20], with its high activity being ascribed to the morphology of the nanoparticles [20]. In this paper, we report the synthesis, characterisation, and application of TiO2-Carbon sphere (TiO2C) composite for the catalysis of the dehydration of fructose into 5-HMF in DMSO solvent. To the best of our knowledge, there have been no reports on the use of TiO2C as a catalyst for the dehydration of fructose into 5-HMF. We synthesised the TiO2C composites using a microwave-assisted method. Because TiO2 has redox acidic sites, Bronsted acid sites were introduced by sulfonating the TiO2C composites with concentrated sulfuric acid. Scanning electron microscopy (SEM), Brunauer-Emmett-Teller (BET) analysis, X-ray diffraction, and Fourier transform infrared (FTIR) spectroscopy were used to characterise the TiO2C acid catalyst. This solid acid catalyst was tested in DMSO solvent as a catalyst for the dehydration of fructose into 5-HMF. To achieve a higher dehydration of 5-HMF, reaction parameters such as reaction temperature, reaction time, and catalyst amount were optimised. Moreover, the ability to recycle the catalyst was studied to evaluate its catalytic stability. 2.1. Materials All chemicals used for synthesising the TiO2 nanoparticles (NPs) and TiO2C composites were purchased from Sigma-Aldrich Corporation (St. Louis, MO, USA). Titanium (IV) isopropoxide (TTIP) and fructose were used as precursors in the preparation of TiO2 and TiO2C composites. Polyvinylpyrrolidone (PVP) was used as a surfactant and ammonium hydroxide was used as a reducing agent for the preparation of TiO2 NPs. Sulfuric acid was used for sulfonating the TiO2C solid acid catalyst. 2.2. Preparation of TiO2 Nanoparticles Using the Sol-Gel Method Ethanol (50 mL) was mixed with 17.79 g of TTIP and stirred for 30 min. PVP (4 g) was added to the mixture, which then was stirred at 70 ◦C until it was completely dissolved. Next, 4 mL of NH3·2H2O was added. The solution was heated at 90 ◦C under reﬂux conditions for 24 h. The resultant sample was washed with water and dried at 100 ◦C for 24 h. 2.3. Preparation of Carbon Spheres Using the Microwave-Assisted Method 2.6. Surface Acid Measurement Tests A total of 100 mg of C_S and TiO2C_S composite acid catalysts was mixed with 50 ml of DI water and 40 ml of 0.01 M NaOH and stirred at room temperature (25 to 27 ◦C) overnight [22]. The solution was ﬁltered and titrated with 0.1 N of oxalic acid using phenolphthalein as the indicator. 2.8. Catalytic Testing Fructose (0.5 g), DMSO (6 ml), and catalyst (0.1 g) were placed in a vessel and allowed to react for 1 h at 120 ◦C using the ultraclave high-performance microwave reactor. After the reaction, the solution was centrifuged to separate the solid catalyst particles from the liquid, and the supernatant liquid was analysed using a Lambda UV-Vis spectrometer Model 750s (PerkinElmer, Shelton, CT, USA) at the wavelength of 284 nm. 2.7. Characterisation Elemental composition and surface morphology analyses were conducted using the AURIGA® scanning electron microscope (Zeiss, Oberkochen, Germany). The crystallinity and phase composition were investigated using the PanAlytical X’Pert Pro (Panalytical Ltd., Eindhoven, The Netherlands). The functional groups of the catalysts were measured in the wavelength range of 550–4000 cm−1 using a Spectrum 100 FTIR spectrometer (PerkinElmer, Waltham, MA, USA). The speciﬁc surface area, pore volume, and pore size were measured using the ASAP 2020 BET analysis system (Micromeritics Instruments Corp., Norcross, GA, USA). 2.5. Preparation of the Sulfonated Carbon and TiO2C Acid Catalysts TiO2C was sulfonated by adding 1 g of TiO2C composite to 25 ml of concentrated sulfuric acid and stirring the solution at 120 ◦C for 24 h under reﬂux conditions. The sulfonated samples, denoted as TiO2C_S, were washed with DI water and dried at 150 ◦C for 24 h. The carbon spheres were sulfonated using the same method and were denoted as C_S. 2.3. Preparation of Carbon Spheres Using the Microwave-Assisted Method The carbon precursor (fructose) was mixed with deionised (DI) water and stirred at room temperature (25 to 27 ◦C) for 2 h. The obtained mixture was transferred into microwavable plastic 202 Catalysts 2019, 9, 126 vessels that were then placed into the ultraclave high-performance microwave reactor and allowed to react at 180 ◦C for 2 h. 2.4. Preparation of TiO2C Composites Using the Microwave-Assisted Method TiO2 nanoparticles were added to 50 mL of DI water. The solution was sonicated at 30 ◦C for at least 1 h, after which the fructose was added. The pH of the solution was adjusted to 2. The solution was stirred at room temperature (25 to 27 ◦C) for 2 h and then microwaved in the ultraclave high-performance microwave reactor for another 2 h at 180 ◦C. The resultant sample was washed three times with DI water, dried at 150 ◦C for 24 h, and then calcined at 500 ◦C for 2 h. Different amounts of TiO2 nanoparticles (1, 5, and 10%) were loaded onto the carbon spheres. For example, to prepare the 1%TiO2C nanocomposite, 0.1 g of TiO2 nanoparticles was added to 9.9 g of fructose, whereas to prepare the 5%TiO2C and 10%TiO2C, the mass ratio of TiO2 nanoparticles to fructose was 0.5:9.5 and 1:9, respectively. The TiO2C composites loaded with 1, 5, and 10% TiO2 were denoted as 1%TiO2C, 5%TiO2C, and 10%TiO2C, respectively. 2.5. Preparation of the Sulfonated Carbon and TiO2C Acid Catalysts 3.1. BET Surface Area and Pore Volume of the Catalysts 3.1. BET Surface Area and Pore Volume of the Catalysts The BET surface area and pore volume of the carbon spheres and the various compositions of the TiO2C composites are presented in Table 1. The carbon spheres had a surface area of 517 m2/g before sulfonation. However, a reduction in surface area was observed after adding TiO2 NPs to the carbon due to TiO2 NPs blocking some of the micropores of the carbon. The TiO2C composite loaded with the lowest dosage of TiO2 NPs (i.e., 1%TiO2C) had a larger surface area and pore volume of 413 m2/g and 0.26 cm3/g, respectively, than the composites loaded with higher amounts of TiO2 (i.e., 5%TiO2C and 10%TiO2C). The surface area of the TiO2C composite decreased when a high loading of TiO2 NPs was introduced onto the carbonaceous support. The 1%TiO2C composite had the highest pore volume of 0.26 cm3/g compared to that of 5%TiO2C and 10%TiO2C composites. Both the surface area and the pore volume were affected by functionalising the carbon spheres and TiO2C with concentrated sulfuric acid. After sulfonation, the surface area of the carbon spheres decreased from 517 to 167 m2/g and that of the 1%TiO2C, 5%TiO2C, and 10%TiO2C composites decreased from 413 to 83, 273 to 61, and 202 to 59 m2/g, respectively. Pore volume also decreased. The reduction in surface area and pore volume might be attributed to the SO3H groups that were attached to the pores of the carbon spheres and TiO2C composites during sulfonation. Other researchers reported similar results [21,22]. For example, Tamborini et al. [21] synthesised sulfonated porous carbon materials and used them for the production of biodiesel. The surface areas of the different synthesised carbons (PC100S and PC200S) were 630 and 695 m2/g, respectively. The sulfonation process decreased the surface area of PC100S to 470 m2/g and that of PC200S to 140 m2/g. The pore volume of the carbon materials also decreased from 0.92 to 0.77 cm3/g and from 1 to 0.34 cm3/g, respectively. Sulfonation corroded the microporosity and mesoporosity of these carbon materials. Liu et al. [22] prepared a carbon-based acid catalyst and used it for the esteriﬁcation of acetic acid with ethanol. After sulfonation, the surface area and pore volume of the activated carbon decreased from 751 to 602 m2/g and from 0.47 to 0.38 cm3/g, respectively. Their ﬁndings suggested that SO3H groups were grafted onto the pore spaces of the activated carbons. 2.10. Hot Filtration Tests The 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S acid catalysts were used to perform hot ﬁltration tests to determine the heterogeneous nature of the acid catalyst. In a typical procedure, 0.5 g of fructose and 0.1 g of TiO2C_S were added to 6 ml of DMSO and allowed to react for 30 min using the ultraclave high-performance microwave reactor. After the reaction, the solution was allowed to cool down and then was centrifuged to separate the catalyst from the reaction mixture. The solution was analysed using a UV-Vis spectrometer at a wavelength of 284 nm. The same reaction was repeated for 1 h using the solution ﬁltrate without adding the substrate. 2.9. Reusability Studies Reusability of the materials was studied using the dehydration of fructose into 5-HMF in DMSO as an illustrative reaction. The C_S and TiO2C_S composite acid catalysts were allowed to react for 1 h in separate vessels and then were separated from the solution by centrifugation. The recovered catalysts were then washed thoroughly with DI water and ethanol and dried for 4 h at 90 ◦C. This process was repeated four times, with the 5-HMF yield calculated each time. 203 Catalysts 2019, 9, 126 2.10. Hot Filtration Tests 3.1. BET Surface Area and Pore Volume of the Catalysts Table 1. Textural properties and hot ﬁltration results of carbon spheres and TiO2C nanocomposites. Catalyst before Sulfonation Catalyst after Sulfonation SBET (m2/g) Vp (cm3/g) SBET (m2/g) Vp (cm3/g) Sulfonic Groups (m·molg−1) a C 517 - C_S 167 - 1.34 1%TiO2C 413 0.26 1%TiO2C_S 83 0.07 1.46 5%TiO2C 273 0.13 5%TiO2C_S 61 0.04 1.55 10%TiO2C 202 0.10 10%TiO2C_S 59 0.03 1.49 C: Carbon; SBET: Surface Area; Vp: Pore Volume; S_: Sulfonated; a Attained by titration with NaOH. Table 1. Textural properties and hot ﬁltration results of carbon spheres and TiO2C nanocomposites. The acid strength of C_S, 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S was calculated to be 1.46, 1.55, 1.49, and 1.36 mmol g−1, respectively, as seen in Table 1. The acid distribution results conﬁrmed that SO3H acid sites were introduced onto the surface of the prepared catalysts, verifying the strong 204 Catalysts 2019, 9, 126 adsorption bands of the S=O group at 1180 and 1008 cm−1, which were associated with the SO3− groups in the FTIR spectra discussed in Section 3.3. The heterogeneous nature of the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S acid catalysts was evaluated using hot ﬁltration tests conducted at optimum conditions. The results showed that leaching of SO3H functional groups did not occur during the dehydration of fructose to 5-HMF. The 5-HMF dehydration was about 80% after 30 min of reaction. The reaction was repeated again for 1 h and showed that the HMF dehydration did not improve beyond 80%. These ﬁndings conﬁrmed that the TiO2C_S acid catalysts are heterogeneous and that the functional groups were strongly attached to the surface of the TiO2C_S catalysts, so leaching did not occur. 3.2. Surface Morphology The SEM images of the neat carbon and neat TiO2C composites are shown in Figure 1a–d, and those of the C_S and TiO2C_S acid catalysts are shown in Figure 1e–h. The neat carbon was spherical and had a smooth surface, with the spheres agglomerated and interconnected (Figure 1a). The neat 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S composites consisted of a combination of interconnected spheres and small TiO2 particles which did not have a deﬁnite shape, and were aggregated (Figure 1b–d). Moreover, a high amount (5 and 10 wt.%) of TiO2 NPs loading promoted the formation of irregular TiO2 particles in the TiO2C composites. After sulfonation, the particles were found to have shrunk, as seen in Figure 1f–h, whereas in a case of C_S the spheres were found to have enlarged as shown in Figure 2e. Figure 1. SEM low magniﬁcation images of (a) neat C, (b) neat 1%TiO2C, (c) neat 5%TiO2C, (d) neat 10%TiO2C, (e) C_S solid acid catalyst, (f) 1%TiO2C_S solid acid catalyst, (g) 5%TiO2C_S solid acid catalyst, and (h) 10%TiO2C_S solid acid catalyst. Figure 1. SEM low magniﬁcation images of (a) neat C, (b) neat 1%TiO2C, (c) neat 5%TiO2C, (d) neat 10%TiO2C, (e) C_S solid acid catalyst, (f) 1%TiO2C_S solid acid catalyst, (g) 5%TiO2C_S solid acid catalyst, and (h) 10%TiO2C_S solid acid catalyst. Figure 2 shows the high magniﬁcation SEM images of (a) C_S solid acid catalyst, (b) 1%TiO2C_S, (c) 5%TiO2C_S, (d) 10%TiO2C_S and (e) sulfonated TiO2. From Figure 2a, we can see that the C_S solid acid catalyst had a smooth surface, whereas the 1%TiO2C_S and 5%TiO2_S showed the presence of small traces of TiO2 particles which were deposited onto the carbon surface. In the 10%TiO2C_S sample (Figure 2d) TiO2 particles were deposited on the surface of the carbon. However, these TiO2 particles were not homogeneously distributed onto the surface of the carbon. In the sulfonated TiO2 sample (Figure 2e); SEM revealed that TiO2 particles had irregular shaped and were agglomerated. 205 Catalysts 2019, 9, 126 Figure 2. SEM high magniﬁcation images of (a) C_S solid acid catalyst, (b) 1%TiO2C_S solid acid catalyst, (c) 5%TiO2C_S solid acid catalyst, (d) 10%TiO2C_S solid acid catalyst and (e) sulfonated TiO2. Figure 2. SEM high magniﬁcation images of (a) C_S solid acid catalyst, (b) 1%TiO2C_S solid acid catalyst, (c) 5%TiO2C_S solid acid catalyst, (d) 10%TiO2C_S solid acid catalyst and (e) sulfonated TiO2. Figure 2. 3.2. Surface Morphology SEM high magniﬁcation images of (a) C_S solid acid catalyst, (b) 1%TiO2C_S solid acid catalyst, (c) 5%TiO2C_S solid acid catalyst, (d) 10%TiO2C_S solid acid catalyst and (e) sulfonated TiO2. The energy-dispersive spectroscopy (EDS) spectra of the neat carbon and neat TiO2C composites are shown in parts (a) to (d) of Figure 3, and those of the carbon spheres and TiO2C after sulfonation are shown in parts (e) to (h) of Figure 3. The EDS spectrum of the carbon spheres contains only C and O peaks, whereas that of the composite samples contains the Ti peak, conﬁrming that TiO2 was present in the samples. The elemental weight percent of C decreased with an increase in TiO2 loading. After sulfonating the carbon, the C, O, and S peaks are present in the spectrum (Figure 3e), indicating that the carbon spheres were neat and no impurities were detected in the sample. The spectra of the TiO2C_S acid catalysts [parts (f) to (h) of Figure 3] showed the presence of C, O, Ti, and S, and the elemental weight percentage of Ti increased with an increase in the loading percentage of TiO2. However, the weight percentage of C decreased with an increase in the loading percentage of TiO2. No other impurity elements were found in the prepared solid acid catalysts. The presence of S in all the solid acid catalysts indicated that the samples were successfully functionalised with sulfuric acid. Elemental composition analysis revealed that all the prepared acid catalysts contained sulfonic acid in the form of SO3H groups [23], further proving the presence of sulfonic groups, as indicated in Table 1. Wang et al. [14] also conﬁrmed the presence of sulfonic groups on the surface of carbon materials after sulfonation. The TiO2C_S composites had C, O, Ti, and S, and their elemental weight percent of Ti 206 Catalysts 2019, 9, 126 increased with an increase in the loading percentage of Ti. However, the weight percent of C decreased with an increase in the loading percentage of Ti. Figure 3. EDS spectra and elemental composition of (a) neat C, (b) neat 1%TiO2C, (c) neat 5% TiO2C, (d) neat 10% TiO2C, (e) C_S solid acid catalyst, (f) 1%TiO2C_S solid acid catalyst, (g) 5% TiO2C_S solid acid catalyst, and (h) 10% TiO2C_S solid acid catalyst. Figure 3. 3.2. Surface Morphology EDS spectra and elemental composition of (a) neat C, (b) neat 1%TiO2C, (c) neat 5% TiO2C, (d) neat 10% TiO2C, (e) C_S solid acid catalyst, (f) 1%TiO2C_S solid acid catalyst, (g) 5% TiO2C_S solid acid catalyst, and (h) 10% TiO2C_S solid acid catalyst. 3.3. Chemical Analysis Using FTIR 3.3. Chemical Analysis Using FTIR The FTIR spectra of the neat, calcined, and sulfonated carbon spheres are shown in Figure 4a. The presence of oxygen groups is demonstrated by the bands at 3000–3600 and 1710 cm−1, which were attributed to –OH stretching and C=O vibrations, respectively [3,24]. The 875–750 cm−1 band was assigned to the aromatic C–H group, and the presence of aromatic rings was conﬁrmed by the band at 1620 cm−1, which was assigned to C=C vibrations [24]. Additional bands at 1180 and 1008 and 1106−1168 cm−1 were observed in the spectrum of the sulfonated carbon spheres. These bands were assigned to the symmetric stretching vibration of S=O groups, which are associated with the SO3− groups [25] and C=S stretching [26], respectively as shown in Figure 4a. These sulfonic peaks in the spectrum of the sulfonated carbon spheres indicate that -SO3H functional groups were attached to the carbon spheres. The FTIR spectra of the neat, calcined, and sulfonated 1%TiO2C, 5%TiO2C, and 10%TiO2C (Figure 4b–d) show bands at 1700 and 1200 cm−1, which were attributed to the C=O and 207 Catalysts 2019, 9, 126 C–O stretching of the carboxyl group [27]. A strong adsorption band of the S=O group at 1008 cm− which was associated with the SO3−groups, was observed. 1HDWFDUERQ &DUERQFDOFDW & &DUERQVXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP 1HDWFDUERQ &DUERQFDOFDW& &DUERQVXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP 1HDW7L2& 7L2 &FDOFDW & 7L2 &VXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP 1HDW7L2& 7L2 &FDOFDW & 7L2 &VXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP ;ĂͿ ;ďͿ 1HDW7L2& 7L2&FDOFDW & 7L2 &VXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP 1HDW7L2& 7L2&FDOFDW & 7L2 &VXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP ;ĐͿ Figure 4. Cont. C–O stretching of the carboxyl group [27]. A strong adsorption band of the S=O group at 1008 cm−1, which was associated with the SO3−groups, was observed. 3.3. Chemical Analysis Using FTIR 1HDWFDUERQ &DUERQFDOFDW& &DUERQVXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHU FP 1HDWFDUERQ &DUERQFDOFDW & &DUERQVXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP ;ĂͿ 7UDQVPLWWDQFH 7UDQVPLWWDQFH :DYHQXPEHUFP :DYHQXPEHUFP :DYHQXPEHUFP 1HDW7L2& 7L2 &FDOFDW & 7L2 &VXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP 1HDW7L2& 7L2 &FDOFDW & 7L2 &VXOIRQDWHG :DYHQXPEHUFP ;ďͿ 7UDQVPLWWDQFH 7UDQVPLWWDQFH :DYHQXPEHUFP :DYHQXPEHUFP 1HDW7L2& 7L2&FDOFDW & 7L2 &VXOIRQDWHG :DYHQXPEHUFP ;ĐͿ 1HDW7L2& 7L2&FDOFDW & 7L2 &VXOIRQDWHG 7UDQVPLWWDQFH :DYHQXPEHUFP 7UDQVPLWWDQFH 7UDQVPLWWDQFH :DYHQXPEHUFP :DYHQXPEHUFP Figure 4. Cont. 208 Catalysts 2019, 9, 126 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 Neat 10% TiO 2C 10% TiO 2C calc at 500 0C 10% TiO 2C sulfonated Transmittance / % Wavenumber (103) / cm-1 0.8 1.2 1.6 2.0 2.4 Neat 10% TiO 2C 10% TiO 2C calc at 500 0C 10% TiO 2C sulfonated Transmittance / % Wavenumber (103) / cm-1 (d) Figure 4. FTIR spectra of neat, calcined and sulfonated (a) carbon, (b) 1%TiO2C, (c) 5%TiO2C, and (d) 10%TiO2C. 0.8 1.2 1.6 2.0 2.4 Neat 10% TiO 2C 10% TiO 2C calc at 500 0C 10% TiO 2C sulfonated Transmittance / % Wavenumber (103) / cm-1 0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 Neat 10% TiO 2C 10% TiO 2C calc at 500 0C 10% TiO 2C sulfonated Transmittance / % Wavenumber (103) / cm-1 (d) Transmittance / % Transmittance / % Figure 4. FTIR spectra of neat, calcined and sulfonated (a) carbon, (b) 1%TiO2C, (c) 5%TiO2C, and (d) 10%TiO2C. 3.4. Structural Characterisation Using XRD The XRD patterns of the calcined carbon spheres and TiO2C composites are shown in Figure 5a. The XRD pattern of the neat carbon spheres showed broad peaks at 25◦and 43◦, which were indexed to the [002] and [100] characteristic phases of amorphous carbonaceous material, whereas the TiO2C sample which was loaded with the least dosage of TiO2 (1%TiO2C), was found to be amorphous, due to the effect of high concentration of the carbon which was present in the sample. The XRD pattern of the crystalline structure of the 5%TiO2C and 10%TiO2C composites had sharp peaks at 25.2◦, 37.83◦, 48.2◦, 54.8◦, 62.2◦, and 70.2◦, which were indexed to the [101], [004], [200], [105], [204], and [116] phases of anatase TiO2 [28]. After sulfonation, the peaks for the carbon spheres and 1%TiO2C shifted, as shown in Figure 5b. Moreover, the acid treatment (sulfonation) caused the [100] plane peak to disappear. The 5%TiO2C and 10%TiO2C composites were not affected by the sulfonation. Figure 5. XRD results for carbon spheres and TiO2C composites (a) that were calcined at 500 ◦C and (b) after sulfonation. Figure 5. XRD results for carbon spheres and TiO2C composites (a) that were calcined at 500 ◦C and (b) after sulfonation. (b) after sulfonation. 3.5. Catalytic Testing 3.5.1. Effect of Different Solvents on the Dehydration of Fructose into 5-HMF 3.5.1. Effect of Different Solvents on the Dehydration of Fructose into 5-HMF Dehydration of fructose into 5-HMF was tested using different alcoholic solvents (such as isopropanol, ethanol, and methanol), water, and DMSO at 120 ◦C for 60 min (Table 2). All the 209 Catalysts 2019, 9, 126 alcoholic solvents produced a low 5-HMF yield of <12% on all the solid catalysts tested for this reaction. Water was also found to be an ineffective solvent. DMSO was the most effective solvent. The HMF yield was 84% for the carbon solid acid catalyst, 91% for 1%TiO2C_S, 92% for 5%TiO2C_S, and 95% for 10%TiO2C_S. Thus, the HMF dehydration increased as the TiO2 content increased. These results suggest that DMSO is a suitable solvent for use in the dehydration of fructose into 5-HMF when using TiO2C solid acid catalysts. DMSO acted both as a solvent and as a reaction mediator; hence a high dehydration of 5-HMF was achieved. Table 2. Dehydration of fructose in different solvents. 3.4. Structural Characterisation Using XRD Catalyst HMF Dehydration (%) [UV-Vis at the Measured Absorbance of 284 nm] Methanol DI Water Ethanol Isopropanol DMSO C 3 5 11 1 84 1%TiO2C 3 6 4 9 91 5%TiO2C 3 1 6 0 92 10%TiO2C 0 4 12 11 95 Reaction conditions: Substrate, fructose; catalyst amount, 0.1 g; temperature, 120 ◦C; time, 60 min. Table 2. Dehydration of fructose in different solvents. Table 3 shows a comparison of the TiO2C composite acid catalyst synthesised in this work with other catalysts reported in published works on dehydration of fructose into 5-HMF in DMSO using the microwave-assisted method. We found that our designed TiO2C_S acid catalysts produced the highest 5-HMF yield. The highest 5-HMF yield that we achieved at 120 ◦C, a temperature lower than those used in other studies, was 91% for 1%TiO2C_S and 92 and 95% for 5%TiO2C_S and 10%TiO2C_S, respectively. To our knowledge, this is the ﬁrst report of the use of TiO2C_S composites as effective solid acid catalysts for the conversion of fructose into 5-HMF using the microwave-assisted method. De et al. [19] used mesoporous TiO2 nanomaterial to catalyse the dehydration of D-fructose into 5-HMF in DMSO solvent under microwave-assisted heating. This reaction was conducted at 130 ◦C for 2 min for a 49.2% 5-HMF yield was achieved. Dutta et al. [15] also used mesoporous TiO2 nanoparticles for the same reaction at 140 ◦C for 5 min and reported a yield of53.4%. Use of carbonaceous acid catalysts for this reaction achieved a 100% conversion of fructose and 90% 5-HMF yield at 160 ◦C for 1.5 h. Wang et al. [16] used carbon-based solid acid catalysts to catalyse the dehydration of fructose into 5-HMF at 130 ◦C for 1.5 h for a 5-HMF yield of about 91.2%. Hu et al. [18] explored a magnetic lignin-derived carbonaceous acid catalyst for the catalysed conversion of fructose into 5-HMF and achieved a 5-HMF yield of 81.1% with 100% fructose conversion. Table 3. Comparison of results of the dehydration of fructose (substrate) into 5-HMF in DMSO using carbonaceous and TiO2-based solid acid catalysts. carbonaceous and TiO2 based solid acid catalysts. Catalyst Catalyst Mass (g) Substrate Mass (g) T (◦C) Time (min) 5-HMF Yield (%) Ref. 3.4. Structural Characterisation Using XRD Uv-Vis a HPLC b C_S 0.1 0.5 120 60 84 - This work 1%TiO2C_S 0.1 0.5 120 60 91 - This work 5%TiO2C-S 0.1 0.5 120 60 92 - This work 10%TiO2C_S 0.1 0.5 120 60 95 - This work TiO2 0.05 0.1 130 2 49.5 47.8 [19] TiO2 0.1 0.05 140 5 53.4 - [20] CS 0.1 0.5 160 90 - 90 [15] C 0.4 0.5 130 90 - 91.2 [16] Magnetic lignin-derived carbon (MLC)-SO3H 0.05 0.1 130 40 - 81.1 [18] a 5-HMF dehydration was calculated by UV-vis at the measured absorbance of 284 nm; b 5-HMF dehydration measured by HPLC. 3.5.2. Effect of Reaction Temperature on HMF Dehydration 3.5.2. Effect of Reaction Temperature on HMF Dehydration The effect of reaction temperature on the catalytic transformation of fructose to HMF was carried out at 25, 60, 80, 100 and 120 ◦C. The reaction was conducted using 1%TiO2C_S, 5%TiO2C_S, The effect of reaction temperature on the catalytic transformation of fructose to HMF was carried out at 25, 60, 80, 100 and 120 ◦C. The reaction was conducted using 1%TiO2C_S, 5%TiO2C_S, 210 Catalysts 2019, 9, 126 10%TiO2C_S, C_S solid acid catalysts and without the catalyst (non-catalytic reaction). The results are shown in Figure 6, which suggests that no 5-HMF was formed when the reaction was carried out in the absence of a catalyst on all the reaction temperatures that were studied. For the carbon solid acid catalyst and reaction temperature of 25 ◦C, the 5-HMF yield was <8%. When the temperature was raised to 60 ◦C, the 5-HMF yield slightly increased to 35%. At 80 ◦C, the 5-HMF yield improved to 71%. The highest 5-HMF yield of 85% was achieved at 120 ◦C. The effect of reaction temperature was also tested using the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts. The 5-HMF yield was low when the reaction was conducted at 25 ◦C, then increased as the reaction temperature increased. The highest HMF yields of 91, 92, and 95% were achieved at 120 ◦C with the use of 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts, respectively. Thus, the best temperature for 5-HMF production using these composite catalysts was 120 ◦C. Compared to the performance of the carbon solid acid catalyst, an improved HMF yield was achieved with the use of 1, 5, and 10%TiO2C_S solid acid catalysts. 0 20 40 60 80 100 40 60 80 100 120 1% TiO 2C_S 5% TiO 2C_S 10% TiO 2C_S C_S No catalyst HMF yield / % Temperature / oC Figure 6. Effect of reaction temperature on fructose dehydration into HMF using sulfonated carbon and TiO2C composite catalysts in DMSO solvent. 0 20 40 60 80 100 40 60 80 100 120 1% TiO 2C_S 5% TiO 2C_S 10% TiO 2C_S C_S No catalyst HMF yield / % o Temperature / oC Figure 6. Effect of reaction temperature on fructose dehydration into HMF using sulfonated carbon and TiO2C composite catalysts in DMSO solvent. 3.5.3. 3.5.2. Effect of Reaction Temperature on HMF Dehydration Effect of Reaction Time on 5-HMF Dehydration The effect of reaction time on the catalysed dehydration of fructose into 5-HMF was studied using different times of 15, 30, 60, and 120 min. The reaction was conducted using the prepared solid acid catalysts and no catalyst (non-catalytic reaction). The results are shown in Figure 7. No 5-HMF was formed when the reaction was carried out without a catalyst for all four reaction times. When the reaction was performed with the C_S solid acid catalyst, increasing the reaction time from 15 to 30 min improved the 5-HMF yield from 25% to 46%. The HMF dehydration increased rapidly to 85% after conducting the reaction for 60 min. However, performing the reaction for 120 min caused a slight drop in the 5-HMF yield to 83%. When conducting the dehydration reaction using 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts, increasing the reaction time from 15 to 30 min increased the 5-HMF yield from 27 to 49%, 28 to 58%, and 29 to 68%, respectively. Increasing the reaction time to 60 min signiﬁcantly improved the 5-HMF yield to >90% for all the TiO2C composite acid catalysts. Finally, for a reaction time of 120 min, the HMF dehydration for 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts was calculated to be 88, 92, and 93%, respectively. 211 Catalysts 2019, 9, 126 y 0 20 40 60 80 100 20 40 60 80 100 120 1%TiO 2C_S 5% TiO 2C_S 10%TiO 2C_S C_S No catalyst HMF yield / % Reaction time / min Figure 7. Effect of reaction time on the dehydration of fructose into HMF in DMSO solvent. 0 20 40 60 80 100 20 40 60 80 100 120 1%TiO 2C_S 5% TiO 2C_S 10%TiO 2C_S C_S No catalyst HMF yield / % Reaction time / min Reaction time / min Figure 7. Effect of reaction time on the dehydration of fructose into HMF in DMSO solvent. Figure 7. Effect of reaction time on the dehydration of fructose into HMF in DMSO solvent. 3.5.4. Effect of Catalyst Amount on HMF Dehydration The effect of the amount of catalyst on the 5-HMF yield was studied, with the results presented in Figure 8. The reaction was carried out using catalyst dosages of 0.02, 0.05, 0.1, and 0.2 g. 3.5.2. Effect of Reaction Temperature on HMF Dehydration The results indicated that no 5-HMF was formed when this reaction was conducted in the absence of a catalyst on all the catalyst dosages that were studied. The 5-HMF yield increased from 34 to 59% when the amount of the C_S solid acid catalyst increased from 0.02 to 0.05 g under the same reaction conditions. An increase in the catalyst dosage to 0.1 g increased the 5-HMF yield to 85%. However, increasing the catalyst dosage to 0.2 g resulted in a reduction of the 5-HMF yield to 82%. The effect of catalyst concentration was also tested using the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts. An increase in the catalyst dosage from 0.02 to 0.1 g drastically improved the 5-HMF yield of the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts to 91, 92, and 95%, respectively. However, when the catalyst dosage increased to 0.2 g, the 5-HMF yield of the 5%TiO2C_S, and 10%TiO2C_S solid acid catalysts slightly decreased to 91 and 92% respectively, whereas in a case of the 1%TiO2C_S the 5-HMF yield increased to 93%. This decrease in 5-HMF dehydration at a high catalyst dosage for the 5%TiO2C_S and 10%TiO2C_S, could be attributed to the excess acid active sites that promote both the dehydration reaction and the formation of by-products such as humins [29]. The 5-HMF yields did not increase with an increase in catalyst dosage for the C_S, 5%TiO2C_S and 10%TiO2C_S solid acid catalysts. 212 Catalysts 2019, 9, 126 alysts 2019, 9, 126 0 20 40 60 80 100 0.05 0.1 0.15 0.2 1% TiO 2C_S 5% TiO 2C_S 10% TiO 2C_S C_S No catalyst HMF yield / % Catalyst dosage / g Figure 8. Effect of catalyst dosage on the dehydration of fructose into HMF in DMSO solvent. Catalyst dosage / g Figure 8. Effect of catalyst dosage on the dehydration of fructose into HMF in DMSO solvent. 3.5.5. Reusability of the Catalyst The reusability of the catalysts was studied, with the results presented in Figure 9. After the ﬁrst run, the 5-HMF yields were 94, 97, and 93% for 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S, respectively. The 5-HMF yield slightly decreased after the fourth run. The 5-HMF yield of the 1%TiO2C_S and 10%TiO2C_S composite acid catalysts decreased by 3% and that of 5%TiO2C_S decreased by 5%. These results indicate that the TiO2C composite acid catalysts are highly stable and can be reused for the dehydration of fructose. 0 20 40 60 80 100 1 2 3 4 1%TiO 2C_S 5%TiO 2C_S 10%TiO 2C_S HMF yield / % Catalytic run Figure 9. Recycling of the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S acid catalysts used in the dehydration of fructose into HMF. Conditions: reaction time = 60 min, reaction temperature = 120 ◦C, catalyst concentration = 0.1 g, amount of fructose = 0.5 g, volume of DMSO = 6 mL). Catalytic run Figure 9. Recycling of the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S acid catalysts used in the dehydration of fructose into HMF. Conditions: reaction time = 60 min, reaction temperature = 120 ◦C, catalyst concentration = 0.1 g, amount of fructose = 0.5 g, volume of DMSO = 6 mL). Figure 9. Recycling of the 1%TiO2C_S, 5%TiO2C_S, and 10%TiO2C_S acid catalysts used in the dehydration of fructose into HMF. Conditions: reaction time = 60 min, reaction temperature = 120 ◦C, catalyst concentration = 0.1 g, amount of fructose = 0.5 g, volume of DMSO = 6 mL). 213 Catalysts 2019, 9, 126 4. Conclusions TiO2C solid acid catalysts were successfully prepared using the microwave-assisted method and then sulfonated with concentrated sulfuric acid. Microwave heating increases the rate of dehydration of fructose. The dehydration of fructose is closely related to the acidity of the catalyst. Among the solvents tested for use in the dehydration of fructose into 5-HMF, DMSO performed the best in terms of 5-HMF dehydration. The reaction temperature, reaction time, and catalyst dosage were found to have an effect on the 5-HMF dehydration. The solid acid catalysts synthesised in this work were highly stable and heterogeneous. Moreover, this study was the ﬁrst time such TiO2C solid acid catalysts were used for dehydrating fructose into 5-HMF. Author Contributions: M.M.S. design the concept and wrote the ﬁrst draft of manuscript. R.M. went through manuscript and provided comments. S.S.R. critically reviewed and corrected the manuscript. Funding: The authors are grateful to the Department of Science and Technology (DST, project no. HGERA8X) and the Council for Scientiﬁc and Industrial Research (CSIR, project no. HGER74p) of South Africa for ﬁnancial support. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Démolis, A.; Essayem, N.; Rataboul, F. Synthesis and applications of alkyl levulinates. ACS Sustain. Chem. Eng. 2014, 2, 1338–1352. [CrossRef] 2. McNeff, C.V.; Nowlan, D.T.; McNeff, L.C.; Yan, B.; Fedie, R.L. Continuous production of 5-hydroxymethylfurfural from simple and complex carbohydrates. Appl. Catal. A Gen. 2010, 384, 65–69. [CrossRef] 3. Qi, X.; Watanabe, M.; Aida, T.M.; Smith, R.L., Jr. Catalytical conversion of fructose and glucose into 5-hydroxymethylfurfural in hot compressed water by microwave heating. Catal. Commun. 2008, 9, 2244–2249. [CrossRef] Wang, J.; Ren, J.; Liu, X.; Xi, J.; Xia, Q.; Zu, Y.; Lu, G.; Wang, Y. Direct conversion of carbohydrates to 4. Wang, J.; Ren, J.; Liu, X.; Xi, J.; Xia, Q.; Zu, Y.; Lu, G.; Wang, Y. Direct conversion of carbohyd 5-hydroxymethylfurfural using Sn-Mont catalyst. Green Chem. 2012, 14, 2506–2512. [CrossRef] g g y 5-hydroxymethylfurfural using Sn-Mont catalyst. Green Chem. 2012, 14, 2506–2512. [CrossRef] 5. Moreau, C.; Finiels, A.; Vanoye, L. Dehydration of fructose and sucrose into 5 hydroxymethylfurfural in the presence of 1-H-3-methyl imidazolium chloride acting both as solvent and catalyst. J. Mol. Catal. A Chem. 2006, 253, 165–169. [CrossRef] 6. Asghari, F.S.; Yoshida, H. Acid catalysed production of 5-hydroxymethyl furfural from d-Fructose in Subcritical Water. Ind. Eng. Chem. Res. 2006, 45, 2163–2173. [CrossRef] 7. Xu, H.; Miao, Z.; Zhao, H.; Yang, J.; Zhao, J.; Song, H.; Liang, N.; Chou, L. Dehydration of fructose into 5-hydroxymethylfurfural by high stable ordered mesoporous zirconium phosphate. Fuel 2015, 145, 234–240. [CrossRef] 8. Jain, A.; Shore, A.M.; Jonnalagadda, S.C.; Ramanujachary, K.V.; Mugweru, A. Conversion Of fructose, glucose and sucrose to 5-hydroxymethyl-2-furfural over mesoporous zirconium phosphate catalyst. Appl. Catal. A Gen. 2015, 489, 72–76. [CrossRef] 9. Kılıc, E.; Yılmaz, S. Fructose dehydration to 5-hydroxymethylfurfural over sulfated TiO2–SiO2, Ti-SBA-15, ZrO2, SiO2, and activated carbon catalysts. Eng. Chem. Res. 2015, 54, 5220–5225. [CrossRef] 10. Wang, S.; Zhang, Z.; Liu, B. Catalytic conversion of fructose and 5-hydroxymethylfurfural into 2,5-Furandicarboxylic acid over a recyclable Fe3O4−CoOx magnetite nanocatalyst. ACS Sustain. Chem. Eng. 2015, 3, 406–412. [CrossRef] 11. Yang, Z.; Qi, W.; Huang, R.; Fang, J.; Su, R.; He, Z. Functionalized silica nanoparticles for conversion of fructose to 5-hydroxymethylfurfural. Chem. Eng. J. 2016, 296, 209–216. [CrossRef] 12. Morales, G.; Paniagua, M.; Melero, J.A.; Iglesias, J. Efﬁcient production of 5-ethoxymethylfurfural from fructose by sulfonic mesostructured silica using DMSO as co-solvent. Catal. Today 2017, 279, 305–316. [CrossRef] 13. Hu, B.; Wang, K.; Wu, L.; Yu, S.H.; Antonietti, M.; Titirici, M.M. References Engineering carbon materials from the hydrothermal carbonization process of biomass. Adv. Mater. 2010, 22, 813–828. [CrossRef] [PubMed] 214 Catalysts 2019, 9, 126 14. Wang, J.; Zhang, Y.; Wang, Y.; Zhu, L.; Cui, H.; Yi, W. Catalytic fructose dehydration to 5-hydroxymethylfurfural over sulfonated carbons with hierarchically ordered pores. J. Fuel Chem. Technol. 2016, 44, 1341–1348. [CrossRef] 15. Zhao, J.; Zhou, C.; He, C.; Dai, Y.; Jia, X.; Yang, Y. Efﬁcient dehydration of fructose to 5 hydroxymethylfurfural over sulfonated carbon sphere solid acid catalysts. Catal. Today 2016, 264, 123–130. [CrossRef] 16. Wang, J.; Xu, W.; Ren, J.; Liu, X.; Lu, G.; Wang, Y. Efﬁcient catalytic conversion of fructose i hydroxymethylfurfural by a novel carbon-based solid acid. Green Chem. 2011, 13, 2678–2681. [CrossRef 17. Guo, F.; Fang, Z.; Zhou, T.J. Conversion of fructose and glucose into 5-hydroxymethylfurfural with lignin-derived carbonaceous catalyst under microwave irradiation in dimethyl sulfoxide–ionic liquid mixtures. Bioresour. Technol. 2012, 112, 313–318. [CrossRef] [PubMed] 18. Hu, L.; Tang, X.; Wu, Z.; Linc, L.; Xu, J.; Xu, N.; Dai, B. Magnetic lignin-derived carbonaceous catalyst for the dehydration of fructose into 5-hydroxymethylfurfural in dimethyl sulfoxide. Chem. Eng. J. 2015, 263, 299–308. [CrossRef] 19. De, S.; Dutta, S.; Patra, A.K.; Bhaumik, A.; Saha, B. Self-assembly of mesoporous TiO2 nanospheres via aspartic acid templating pathway and its catalytic application for 5-hydroxymethyl-furfural synthesis. J. Mater. Chem. 2011, 21, 17505–17510. [CrossRef] 20. Dutta, S.; De, S.; Patra, A.K.; Sasidharan, M.; Bhaumik, A.; Saha, B. Microwave assisted rapid conversion of carbohydrates into 5-hydroxymethylfurfural catalysed by mesoporous TiO2 nanoparticles. Appl. Catal. A Gen. 2011, 409–410, 133–139. [CrossRef] 21. Tamborini, L.H.; Casco, M.E.; Militello, M.P.; Silvestre-Albero, J.; Barbero, C.A.; Acevedo, D.F. Sulfonated porous carbon catalysts for biodiesel production: Clear effect of the carbon particle size on the catalyst synthesis and properties. Fuel Process. Technol. 2016, 149, 209–217. [CrossRef] 22. Liu, X.Y.; Huang, M.; Ma, H.L.; Zhang, Z.Q.; Gao, J.M.; Zhu, Y.L.; Han, X.J.; Guo, X.Y. Preparation of a carbon-based solid acid catalyst by sulfonating activated carbon in a chemical reduction process. Molecules 2010, 15, 7188–7196. [CrossRef] [PubMed] 23. Hou, Q.; Li, W.; Ju, M.; Liu, L.; Chen, Y.; Yang, Q. One-pot synthesis of sulfonated graphene oxide for efﬁcient conversion of fructose into HMF. RCS Adv. 2016, 16, 104016–104024. [CrossRef] 24. Qi, X.; Guo, H.; Li, L.; Smith, R.L., Jr. Acid-catalyzed dehydration of fructose into 5-hydroxymethylfurfu by cellulose derived amorphous carbon. ChemSusChem 2012, 5, 2215–2220. [CrossRef] [PubMed] 25. Yuming He 1, Huayang Li 1, Xuelian Guo 2 and Rongbo Zheng 1,* Yuming He 1, Huayang Li 1, Xuelian Guo 2 and Rongbo Zheng 1,* 1 Yunnan Province Key Laboratory of Wood Adhesives and Glued Products, College of Chemical Engineering, Southwest Forestry University, Kunming 650224, China; hymingin@163.com (Y.H.); Li876770996@outlook.com (H.L.) 1 Yunnan Province Key Laboratory of Wood Adhesives and Glued Products, College of Chemical Engineering, Southwest Forestry University, Kunming 650224, China; hymingin@163.com (Y.H.); Li876770996@outlook.com (H.L.) 1 Yunnan Province Key Laboratory of Wood Adhesives and Glued Products, College of Chemical Engineering, Southwest Forestry University, Kunming 650224, China; hymingin@163.com (Y.H.); Li876770996@outlook.com (H.L.) 2 Wetland college, Southwest Forestry University, Kunming 650224, China; guoxuelian2009@hotmail.com * Correspondence: zhengrbzy@hotmail.com * Correspondence: zhengrbzy@hotmail.com Received: 21 December 2018; Accepted: 18 January 2019; Published: 26 January 2019 Received: 21 December 2018; Accepted: 18 January 2019; Published: 26 January 2019 Abstract: To suppress the agglomeration of a photocatalyst, facilitate its recovery, and avoid photolysis of dyes, various support materials such as ceramic, carbon, and polymer have been investigated. However, these support materials pose the following additional challenges: ceramic supports will settle down at the bottom of their container due to their high density, while the carbon support will absorb the UV-vis light for its black color. Herein, we propose a ﬂoatable, UV transmitting, mesoporous bleached wood with most lignin removal to support P25 nanoparticles (BP-wood) that can effectively, recyclable, three dimensional (3D) photocatalytic degrade dyes such as methylene blue (MB) under ambient sunlight. The BP-wood has the following advantages: (1) The deligniﬁcation makes the BP-wood more porous to not only quickly transport MB solutions upstream to the top surface, but is also decorated with P25 nanoparticles on the cell wall to form a 3D photocatalyst. (2) The deligniﬁcation endows the BP-wood with good UV transmittance to undergo 3D photocatalytic degradation under sunlight. (3) It can ﬂoat on the surface of the MB solution to capture more sunlight to enhance the photodegradation efﬁciency by suppressing the photolysis of MB. (4) It has comparable or even better photocatalytic degradation of 40 mg/L and 60 mg/L MB than that of P25 nanoparticles suspension. (5) It is green, recyclable, and scalable. Keywords: bleached wood support materials; 3D photocatalyst; UV transmittance; ﬂoatable; recyclable www.mdpi.com/journal/catalysts catalysts catalysts References Sevilla, M.; Fuertes, A.B. The production of carbon materials by hydrothermal carbonization of cellul Carbon 2009, 47, 2281–2289. [CrossRef] 26. Sun, Y.; Zhao, J.; Wang, J.; Tang, N.; Zhao, R.; Zhang, D.; Guan, G.; Li, K. Sulfur doped millimetre sized microporous activated carbon spheres derived from sulfonated poly(styrene-devinylbenzene) for CO2 capture. J. Phys. Chem. C 2017, 121, 10000–10009. [CrossRef] 27. Rao, C.N.R. Contribution to the infrared spectra of organosulphur compounds. Can. J. Chem. 1964, 42, 36–42. [CrossRef] 28. Yu, D.; Bo, B.; Yunhua, H. Fabrication of TiO2@yeast-carbon hybrid composites with the raspberry like structure and their synergistic adsorption photocatalysis performance. J. Nanomater. 2013, 2013, 851417. [CrossRef] 29. Kansal, S.K.; Sood, S.; Umar, A.; Mehta, S.K. Photocatalytic degradation of Eriochrome Black T dye using well-crystalline anatase TiO2 nanoparticles. J. Alloys Compd. 2013, 581, 392–397. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 215 1. Introduction In order to rapidly, efﬁciently, and cost-effectively remove dyes from industrial waste water [1], various technologies, such as physical adsorption [2], photocatalytic degradation, chemical oxidation, and membrane ﬁltration, have been implemented, among which photocatalytic degradation has been demonstrated to be of high efﬁciency [3–5]. Heterogeneous photocatalysis is based on the use of UV light with a wavelength shorter than 380 nm to stimulate a semiconductor material (i.e., TiO2 with band gap of ca. 3.2 eV, corresponding to radiation of UV light with a wavelength of about 380 nm) to excite the electrons from the valence band to the conduction band to generate electron–hole pairs, which serve as the oxidizing and reducing agents to photocatalytic degrade dyes [6]. The efﬁciency of TiO2 was reported to be inﬂuenced by many factors, such as crystalline structure [7–10], particle size [10–13], and doping with the other ions [14–17]. However, there are some disadvantages in the use of TiO2 nanoparticle suspension during photocatalytic processes: it tends to agglomerate at high concentrations, and is difﬁcult to separate and recycle from the solution [18,19]. To overcome these disadvantages, TiO2 can be supported on a material that suppresses the agglomeration and facilitates its further recovery. In this context, various support materials, such as ceramic (i.e., molecular sieves, silica, zeolite, and clay) [19–22], carbon (i.e., activated carbon, carbon nanotube, graphene, Catalysts 2019, 9, 115; doi:10.3390/catal9020115 www.mdpi.com/journal/catalysts www.mdpi.com/journal/catalysts 217 Catalysts 2019, 9, 115 and graphite) [18,23–26], and polymer (i.e., chitosan, polyamide, polyester) [27–29] have been investigated. However, these support materials pose additional challenges. For instance, ceramic supports will settle down at the bottom of their container due to their high density, while the dyes in the upper solution will absorb the UV light to have photolysis instead of photocatalytic degradation by the photocatalyst. The carbon support will absorb the UV-vis light due to its black color [30–32]. Thus, it is still a challenge to provide an excellent support material with low density to ﬂoat on the dye solutions’ surface to efﬁciently exploit UV light, 3D porous structure to support photocatalysts nanoparticles, transmit UV light and transport dye upstream. Wood, an earth-abundant, natural, low density, hierarchical, mesoporous material, has been widely used as the template to prepare TiO2 nanomaterials [33,34], the substrate to coat with TiO2 nanoparticle to enhance weathering performance [35], and the support to decorate with palladium nanoparticles for efﬁcient wastewater treatment [36]. 1. Introduction With its mesoporous structure, wood is comprised of numerous long, partially aligned lumens as well as nanochannels along its growth direction, facilitating its ﬂoatation on the solutions’ surface, decoration nanoparticles on the cell wall, and bulk treatment as water ﬂows through the entire mesoporous wood [36]. However, it is difﬁcult to exploit wood as a photocatalyst support due to the 20–30% lignin, whose absorption ranges from 300 nm to 600 nm [31]. The objective of the current study is to test the hypothesis that the bleached wood (i.e., cellulose-based hierarchical porous structure of wood obtained via deligniﬁcation) can also be exploited as an alternative photocatalyst support material. Herein, P25, a commercial TiO2 photocatalyst was coated on three kinds of wood-based supports, namely bleached wood with P25 (BP-wood), half-bleached wood with P25 (HBP-wood), and natural wood (N-wood) with P25 (NP-wood). The ﬁrst two are obtained by removing about 50% lignin and 95% lignin, respectively. Notably, the mesoporous structure of N-wood is well maintained even after removing 95% lignin, and possess transmittance with UV light. The photocatalytic activity of above three wood-based catalysts are investigated in aqueous solution by using methylene blue (MB) dye as a model contaminant under ambient sunlight illumination. The mesoporous structure of the bleached wood is demonstrated to play important roles in the photocatalytic degradation since the wood is composed of 50% vessel channels and 20% ﬁber channels, which provide a pathway to quickly transport MB solution onto the top surface to be photodegraded with the P25 nanoparticles under sunlight illumination. Moreover, the P25 nanoparticles can be penetrated into the wood cell wall to form 3D photocatalitic composites to further enhance the photodegradation with the illumination of transmitted UV light. The experimental result shows that it has better photocatalytic degradation of 60 mg/L MB than that of P25 nanoparticles suspension. 2. Results and Discussion Scheme 1a illustrates the preparation process of BP-wood and the photocatalytic process. In order to remove lignin, the natural basswood is deligniﬁed by H2O2 steam. Scheme 1b demonstrates the approximate photodegradation mechanism of BP-wood-supported catalysts. The degraded materials are continuously transported from the bottom of the BP-wood to the top and inside, forming a 3D catalytic mechanism under the permeation of the light source. As for the control group (P25 nanoparticles are directly added to the MB solution), shown in Scheme 1c. P25 is easily wrapped with the light-absorbing dye in the solution, so that the dye undergoes weak photolysis under illumination, which has a certain degree of inﬂuence on the photocatalytic degradation of P25. 218 Catalysts 2019, 9, 115 b a c Scheme 1. Material preparation and usage. (a) Sketch of BP-wood preparation. (b) Wood absorbs methylene blue (MB) from the bottom of the contact surface and sends MB to the location of lower concentration by capillary action and transpiration of the pipeline. With the provision of ultraviolet light from sunlight, TiO2, once exposed to MB, immediately produces an effective photocatalytic degradation. There is also a degradation process of MB inside the timber pipe and inside the pipe wall. Due to the higher transmission of UV light in BP-wood, the P25 penetrating into the interior of the wood also plays a role of catalyzer. (c) Schematic diagram of a control group-added P25 particles directly. Catalysts 2019, 9, 115 a b c b c Scheme 1. Material preparation and usage. (a) Sketch of BP-wood preparation. (b) Wood absorbs methylene blue (MB) from the bottom of the contact surface and sends MB to the location of lower concentration by capillary action and transpiration of the pipeline. With the provision of ultraviolet light from sunlight, TiO2, once exposed to MB, immediately produces an effective photocatalytic degradation. There is also a degradation process of MB inside the timber pipe and inside the pipe wall. Due to the higher transmission of UV light in BP-wood, the P25 penetrating into the interior of the wood also plays a role of catalyzer. (c) Schematic diagram of a control group-added P25 particles directly. The lignin content can be decreased from 22.5% (N-wood) to 12.3% (Half B-wood), to 1.01% (B-wood) when the H2O2 steam time prolong to 1 h and 4 h, respectively (shown in Figure 1a). The mechanical strength of BP-wood is shown in Figure 1b. 2. Results and Discussion For the wet BP-wood with a thickness of 5 mm, the fracture strength is about 0.4 MPa. It is lower than that of N-wood and dry BP-wood (2.4 MPa), which was strong enough to be carried out in the photocatalytic process. From Figure 1c,d, we can see that massive microscale pores were generated in the cell wall and cell wall corners after deligniﬁcation compared with N-wood, which will provide sites to P25 nanoparticles, and a pathway to quickly transport dye solution upstream to the B-wood’s top surface. Furthermore, P25 nanoparticles dispersed in aqueous solutions were coated on the surface of N-wood, Half B-wood, and B-wood to obtain P25 nanoparticles supported on the N-wood (NP-wood), Half B-wood (HBP-wood ), and B-wood (BP-wood), respectively. As shown in Figure 1e, Raman spectra revealed that peak intensity of B-wood at 1300, 1602, and 1730 cm−1 decreased compared with that of N-wood, which further demonstrated the removal of most of the lignin in B-wood [37]. Moreover, the degradation of the cellulose was negligible, while both lignin and hemicellulose were dramatically removed, as shown in our previous work [38]. The color changed from yellow to white during deligniﬁcation process, shown in Figure 1f. 219 Catalysts 2019, 9, 115 d e NP-wood Half-BP-wood BP-wood 0 5 10 15 20 Lignin content (%) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.1 0.2 0.3 0.4 Stress (MPa) Strain (%) 300 600 900 1200 1500 1800 1122 1730 1602 Raman intensity (a.u.) Wavenumber (cm-1) N-wood B-wood 396 1498 1300 20 %) a c 0.3 0.4 Pa) b f Figure 1. (a) Lignin content and (b) Mechanical strength of BP-wood in wet state. SEM images of cell wall corners and the middle lamella of (c) Natural basswood (N-wood) and (d) B-wood. (e) Raman characterization of N-wood, B-wood. The images were obtained through baseline corrected and normalized. (f) Photos of wood after removal of different content of lignin. NP-wood Half-BP-wood BP-wood 0 5 10 15 20 Lignin content (%) 20 %) a 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.0 0.1 0.2 0.3 0.4 Stress (MPa) Strain (%) 0.3 0.4 Pa) b a d Strain (%) f d c c f e e 300 600 900 1200 1500 1800 1122 1730 1602 Raman intensity (a.u.) Wavenumber (cm-1) N-wood B-wood 396 1498 1300 Figure 1. (a) Lignin content and (b) Mechanical strength of BP-wood in wet state. 2. Results and Discussion SEM images of cell wall corners and the middle lamella of (c) Natural basswood (N-wood) and (d) B-wood. (e) Raman characterization of N-wood, B-wood. The images were obtained through baseline corrected and normalized. (f) Photos of wood after removal of different content of lignin. In order to investigate the water transportation capacity, we designed a dye transportation experiment. N-wood and B-wood are put into a dye solution to observe the distance the dye arrives after a certain time. As shown in Figure 2a, the dye in the B-wood reaches to a larger distance than that of the N-wood, which indicates that the B-wood exhibits better material transportation capabilities than that of N-wood. The speed of dye transportation in the B-wood and the N-wood are determined to be 2.2 and 6.3 mm/min, respectively (Figure 2b). a b N-wood B-wood 0 1 2 3 4 5 6 Speed (mm min-1) N-wood B-wood 0 1 2 3 4 5 6 Speed (mm min-1) Figure 2. (a) The MB transport distance through B-wood and N-wood after 0, 5 min. (b) The MB transport speed through B-wood and N-wood during 5 min. As shown in Figure 3a–d, P25 nanoparticles are not only decorated on the top surface of B-wood ut also penetrate into the cell wall of B-wood due to its mesoporous structure, which results i a Figure 2. (a) The MB transport distance through B-wood and N-wood after 0, 5 min. (b) The MB transport speed through B-wood and N-wood during 5 min. As shown in Figure 3a–d, P25 nanoparticles are not only decorated on the top surface of B-wood, but also penetrate into the cell wall of B-wood due to its mesoporous structure, which results in 220 Catalysts 2019, 9, 115 three-dimensional (3D) P25-wood composites. The light transmittance of wet BP-wood and NP-wood in the range of 200–800 nm are shown in Figure 3e. It is worth noting that, in the ultraviolet range of 300–400 nm, BP-wood still has a light transmittance of 0.5–20%, while NP-wood does not have light transmittance before about 550 nm due to the existence of 22.5% lignin. To demonstrate the 3D photocatalytic features of BP-wood, the P25 nanoparticles coated on the B-wood surface were purposely removed to preserve the P25 nanoparticles which had penetrated into the interior of the B-wood (BI-wood). 2. Results and Discussion The re-plotted linear graph of ln(c0/c) ~ t shown in Figure 4b indicates that the rate constants were determined to be 0.52, 0.41, 0.21, and 0.08 h−1 for P25, BP-wood, HBP-wood, and NP-wood, respectively. Combining with the corresponding lignin content, we can conclude that the photocatalytic activity of wood supported P25 increases with the decreasing of lignin content. Notably, the photolysis of MB with 20 mg/L is very small. After photodegradation, both BP-wood and NP-wood were taken out from the solutions. It is clear that, after the photodegradation, the P25 nanoparticles’ suspension leads to a turbidity inside the entire beaker, which indicates the difﬁculty to be separated and recycled (Figure 4c). As for NP-wood, the solution after photodegradation exhibits yellow color due to the leaching of N-wood [32]. However, it is clean and pollution-free for BP-wood, which reveals the clean and environmental benign. We further exam the BP-wood and NP-wood after photodegradation, shown in Figure 4d,e. Compared with the blue color of the interior of NP-wood, BP-wood appears pure white without MB molecules. That is, the MB molecules inside BP-wood are also photocatalytical degraded, which further demonstrates the 3D photocatalyst feature of BP-wood. As for NP-wood, although there was photocatalytic degradation occurring on the surface, the MB molecules absorbed in the porous wood still remained. a b e d c 0 1 2 3 4 5 6 7 8 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 P25 BP-wood HBP-wood NP-wood Blank ln(C0/C) Time (h) Figure 4. (a) Photo of photocatalytic degradation devices under sunlight. (b) Photodegradation of MB monitored as the normalized concentration change versus irradiation time under sunlight. Photo of (c) MB solutions, (d) BP-wood, and (e) NP-wood after photodegradation of MB solution under sunlight. b 0 1 2 3 4 5 6 7 8 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 P25 BP-wood HBP-wood NP-wood Blank ln(C0/C) Time (h) a a b e c d e c Figure 4. (a) Photo of photocatalytic degradation devices under sunlight. (b) Photodegradation of MB monitored as the normalized concentration change versus irradiation time under sunlight. Photo of (c) MB solutions, (d) BP-wood, and (e) NP-wood after photodegradation of MB solution under sunlight. We also characterize the photocatalytic degradation of the high-concentration MB solution with BP-wood under ambient sunlight. 2. Results and Discussion 10 mg/L methylene blue (MB) aqueous solution was photocatalytically degraded with BI-wood, and B-wood under ambient sunlight. The re-plotted linear graph of ln(c0/c) ~t shown in Figure 3f indicates that the photocatalytic degradation of MB with P25 decorated inside the BP-wood follows roughly the pseudo-ﬁrst-order reaction [7]. The rate constants were determined to be 0.35, 0.21, and 0.18 h−1 for BI-wood, B-wood, and methylene blue (MB) aqueous solution, respectively. That is, the photocatalytic degradation of BI-wood is better than absorption of B-wood and the photolysis of MB, which indicates that the bleached, deligniﬁed wood can be used as 3D photocatalyst support due to its mesoporous structure, and UV transmittance. c d b a e f 200 300 400 500 600 700 800 0 5 10 15 20 25 30 Transmittance (%) Wavelength (nm) BP-wood NP-wood Figure 3. SEM images of B-wood’s (a) top surface and (b–d) cross section decorated with P25 nanoparticles. (e) Optical transmittance of BP-wood and NP-wood. (f) Comparison of photocatalytic degradation of BI-wood with TiO2 removed at the top and B-wood and blank control groups (where C0 is the initial concentration of the dye solution and C is the concentration of dye at corresponding time) [6]. d b b a a c d c f f e 200 300 400 500 600 700 800 0 5 10 15 20 25 30 Transmittance (%) Wavelength (nm) BP-wood NP-wood Wavelength (nm) Wavelength (nm) Figure 3. SEM images of B-wood’s (a) top surface and (b–d) cross section decorated with P25 nanoparticles. (e) Optical transmittance of BP-wood and NP-wood. (f) Comparison of photocatalytic degradation of BI-wood with TiO2 removed at the top and B-wood and blank control groups (where C0 is the initial concentration of the dye solution and C is the concentration of dye at corresponding time) [6]. The photocatalytic properties of BP-wood, HBP-wood, and NP-wood were examined by measuring the photodegradation of 20 mg/L MB under ambient sunlight. As shown in Figure 4a, all wood-based photocatalysts, including BP-wood, HBP-wood, and NP-wood can ﬂoat on the surface of the MB solution, and the P25 nanoparticles coating the top surface of the BP-wood are directly exposed to sunlight. Figure 4b shows the MB photolysis and photodegradation kinetic curves for reactions in 221 Catalysts 2019, 9, 115 which P25, NP-wood, HBP-wood, and BP-wood are used as photocatalysts. Overall, the photocatalytic activity increased when the lignin content decreasing. 2. Results and Discussion From Figure 5a,b, we can see that the photodegradation performance of the BP-wood is comparable to or even better than those of P25 suspension when the concentration of MB increases to 40 mg/L and 60 mg/L, respectively. As we know, MB molecules can be degraded by either photolysis or photocatalytic degradation. With the MB concentration increased, MB molecules will absorb more UV light to be degraded by photolysis, which decreases the photocatalytic degradation efﬁciency of P25 suspension, while the effect on BP-wood is negligible due to its ﬂoatability. Figure 5c further demonstrates that the enhancement factor of the BP-wood versus P25 suspension increased with the increasing of MB concentration. 222 Catalysts 2019, 9, 115 20 40 60 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Enhancement factor Concentration (mg L -1) P25 BPwood a 1 2 3 4 5 4 5 6 7 Time (h) Circle number (N) 20mg L -1 10mg L -1 6 7 ) d 1.2 1.4 ctor c b 0 4 8 12 16 20 24 28 0.0 0.4 0.8 1.2 1.6 2.0 2.4 P25 BP-wood ln(C0/C) Time (h) 0 2 4 6 8 10 12 14 0.0 0.5 1.0 1.5 2.0 2.5 3.0 P25 BP-wood ln(C0/C) Time (h) Figure 5. Photodegradation of MB solutions with (a) 40 mg/L, (b) 60 mg/L monitored as the normalized concentration change versus irradiation time in the presence of P25 and BP-wood under ambient sunlight. (c) Enhancement factor of the use of BP-wood compared with P25. (d) Recycling performance of BP-wood. a 0 4 8 12 16 20 24 28 0.0 0.4 0.8 1.2 1.6 2.0 2.4 P25 BP-wood ln(C0/C) Time (h) b 0 2 4 6 8 10 12 14 0.0 0.5 1.0 1.5 2.0 2.5 3.0 P25 BP-wood ln(C0/C) Time (h) 11c 20 40 60 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Enhancement factor Concentration (mg L -1) P25 BPwood 1.2 1.4 ctor c ( ) 77 d 1 2 3 4 5 4 5 6 7 Time (h) Circle number (N) 20mg L -1 10mg L -1 6 7 ) d Figure 5. Photodegradation of MB solutions with (a) 40 mg/L, (b) 60 mg/L monitored as the normalized concentration change versus irradiation time in the presence of P25 and BP-wood under ambient sunlight. (c) Enhancement factor of the use of BP-wood compared with P25. (d) Recycling performance of BP-wood. 3.1. Materials and Chemicals Basswood was used in this study. P25 was bought from Degussa AG. H2O2, MB, anhydrous ethanol ethanol were bought from Sigma Chemicals (Shanghai, China). 2. Results and Discussion The photodegradation process of BP-wood consists of the following steps. Firstly, the BP-wood ﬂoats on the surface of MB aqueous solutions due to its low density. Secondly, the MB solutions will transport to the top surface of the BP-wood via aligned channels to make contact with P25 nanoparticles. Thirdly, MB molecules will be photocatalytical degraded via P25 nanoparticles under UV light with a wavelength shorter than 380 nm in sunlight. Fourth, MB molecules will continuously accumulate in both top surface and the interior of BP-wood via a concentration gradient to continuous photodegradation. It should be noted that the P25 nanoparticles decorated into the cell wall of BP-wood also exhibit photocatalytic activity since UV light can be transmitted into the interior of BP-wood. Thanks to its large size and 0.4 MPa mechanical strength, BP-wood can be easily recycled to photodegrade the MB solution under ambient sunlight. After the degradation, the BP-wood was taken out and kept under ambient conditions. As shown in Figure 5d, our BP-wood exhibited excellent recyclable performance: during the 5 circles, it takes 6.5, 7.0, 6.8, 7.1, and 6.9 h respectively to achieve photodegradation of a 20 mg/L MB solution. There is no signiﬁcant decline in efﬁciency during the photodegradation process. 3.3. Preparation of NP-Wood and BP-Wood 3.3. Preparation of NP-Wood and BP-Wood The P25 nanoparticles were dispersed in deionized water and ultrasonically dispersed for 10 min. After overnight, the upper 5 mL * 3 g/L P25 suspension was coated on the N-wood, Half-B-wood, and B-wood to form NP-wood, HBP-wood, and BP-wood. 3.5. Characterization Scanning electron microscopy images were determined with a Nova NanoSEM 450, Lincoln, Ne, USA. The accelerating voltage was 15 kV. The UV–vis absorption was measured on Cary 500 Scan UV–vis–NIR spectrophotometer (Harbor, CA, USA). The transmittance of the material comes from the ultraviolet visible spectrophotometer. The UV-visible spectrophotometer model is U-4100 Spectrophotometer, Hitachi (Tokyo, Japan). A universal mechanical test machine was used to measure the mechanical properties with the SUNS UTM-5000 electronic universal testing machine (Shenzhen, China). The size of the test sample was 10 × 1 × 0.5 cm3. And Raman spectra were obtained from LabRam HR Evolution, Horiba, France. 3.4. Photocatalytic Activity Measurement The photocatalytic activity of the aforementioned samples was investigated by placing NP-wood, HBP-wood, and BP-wood on the surface of MB aqueous solution to measure the P25-assisted photodegradation of 100 mL MB aqueous solutions. At the same time, 0 mL and 5 mL * 3 g/L P25 suspension was added into 100 mL MB aqueous solutions as the control group. Then, they were irradiated under ambient sunlight. Finally, the concentration of MB after illuminating for a certain time was monitored by measuring the absorbance of the solutions (which were centrifuged at 2000 rpm to remove P25) at 664 nm. 3.2. Preparation of N-Wood and B-Wood 3.2. Preparation of N-Wood and B-Wood 3.2. Preparation of N-Wood and B-Wood Natural basswood (N-wood) slices with size of π × 20 × 20 × 5 mm3 were obtained by cutting along the direction perpendicular to the growth of wood. Half-B-wood and B-wood were obtained by H2O2 steam deligniﬁcation of above-mentioned N-wood at 100 ◦C for 1 and 4 h, respectively [38]. After rinsed with water, and ethanol for three times, they were dried at 50 ◦C for 4 h. 223 Catalysts 2019, 9, 115 Catalysts 2019, 9, 115 4. Conclusions To summarize, a ﬂoatable, recyclable, efﬁcient, UV light permeable, environmentally friendly, and 3D photocatalyst was easily synthesized through decoration with P25 nanoparticles on both the surface and in the interior of bleached wood. The bleached wood was obtained by removing most lignin from N-wood through H2O2 steam deligniﬁcation. The deligniﬁcation not only endows bleached wood with UV light transmittance, but also provides a highway to transport the MB solution up to the top surface. The as-made BP-wood photocatalyst shows a high photocatalytic degradation of 60 mg/L MB solution under ambient sunlight, better than that of P25 nanoparticles suspension. It also exhibits excellent recyclability due to its large size, ﬂoatability, and 0.4 MPa mechanical strength. The present work opens up an efﬁcacious avenue for designing bleached wood-based recyclable, ﬂoatable, UV permeable, and efﬁcient 3D photocatalyst for environmental pollution. Author Contributions: Conceptualization, R.Z. and Y.H.; methodology, Y.H.; software, R.Z.; validation, R.Z. and X.G.; formal analysis, H.L.; investigation, X.G.; resources, R.Z.; data curation, R.Z.; writing—original draft preparation, Y.H.; writing—review and editing, R.Z.; visualization, R.Z.; supervision, R.Z.; project administration, R.Z. and X.G.; funding acquisition, R.Z. and X.G. Funding: This research was funded by the Joint Special Project of Agricultural Basic Research in Yunnan (2017FG001036). And the APC was funded by the National Natural Science Foundation of China, grant number 41563008, 31100420. Conﬂicts of Interest: The authors declare no conﬂict of interest. 2. Maity, S.K.; Rana, M.S.; Bej, S.K.; Ancheyta-Juárez, J.; Murali Dhar, G.; Prasada Rao, T.S.R. TiO2–ZrO2 mixed oxide as a support for hydrotreating catalyst. Catal. Lett. 2001, 72, 115–119. [CrossRef] 1. Ren, Z.J.; Umble, A.K. Water treatment: Recover wastewater resources locally. Nature 2016, 529, 25. [CrossRef] [PubMed] 1. Ren, Z.J.; Umble, A.K. Water treatment: Recover wastewater resources locally. Nature 2016, 529, 25. [CrossRef] [PubMed] 2. Maity, S.K.; Rana, M.S.; Bej, S.K.; Ancheyta-Juárez, J.; Murali Dhar, G.; Prasada Rao, T.S.R. TiO2–ZrO2 mixed oxide as a support for hydrotreating catalyst. Catal. Lett. 2001, 72, 115–119. [CrossRef] References Al-Ekabi, H., Eds.; Elsevier: Amsterdam, The Netherlands, 1993; p. 169. 11. Yoneyama, H.; Yamanaka, S.; Haga, S. Photocatalytic activities of microcrystalline titania incorporated sheet silicates of clay. J. Phys. Chem. 1989, 93, 4833–4837. [CrossRef] 12. Zhang, Z.; Wang, C.C.; Zakaria, R.; Ying, J. Role of particle size in nanocrystalline TiO2 -based photocatal J. Phys. Chem. B 1998, 102, 10871–10878. [CrossRef] 13. Tsai, S.-J.; Cheng, S. Effect of TiO2 crystalline structure in photocatalytic degradation of phenolic contaminants. Catal. Today 1997, 33, 227–237. [CrossRef] y 14. Paola, A.D.; Ikeda, S.; Marcì, G.; Ohtani, B.; Palmisano, L. Photocatalytic degradation of organic compounds in aqueous systems by transition metal doped polycrystalline TiO2. Catal. Today 2002, 75, 171–176. [CrossRef] 15. Hu, C.C.; Hsu, T.C.; Kao, L.H. One-step cohydrothermal synthesis of nitrogen-doped titanium oxide nanotubes with enhanced visible light photocatalytic activity. Int. J. Photoenergy 2012, 391958. [CrossRef] 15. Hu, C.C.; Hsu, T.C.; Kao, L.H. One-step cohydrothermal synthesis of nitrogen-doped titanium oxide b i h h d i ibl li h h l i i i I J Ph 1 391958 [C R f] 16. Chen, X.; Mao, S.S. Titanium dioxide nanomaterials synthesis, properties, modiﬁcations, and applications. Chem. Rev. 2007, 107, 2891–2959. [CrossRef] [PubMed] 17. Al Dahoudi, N.; Zhang, Q.; Cao, G. Alumina and hafnia ALD Layers for a niobium-doped titanium oxide photoanode. Int. J. Photoenergy 2012, 401393. [CrossRef] 18. Zheng, R.B.; Meng, X.W.; Tang, F.Q. A general protocol to coat titania shell on carbon-based composite cores using carbon as coupling agent. J. Solid State Chem. 2009, 182, 1235–1240. [CrossRef] 19. Hsien, Y.H.; Chang, C.F.; Chen, Y.H.; Cheng, S. Photodegradation of aromatic pollutants in water over TiO2 supported on molecular sieves. Appl. Catal. B Envion. 2001, 31, 241–249. [CrossRef] 19. Hsien, Y.H.; Chang, C.F.; Chen, Y.H.; Cheng, S. Photodegradation of aromatic pollutants in water over TiO2 supported on molecular sieves. Appl. Catal. B Envion. 2001, 31, 241–249. [CrossRef] 20. Sampath, S.; Uchida, H.; Yoneyama, H. Photocatalytic degradation of gaseous pyridine over zeolite-supported titanium dioxide. J. Catal. 1994, 149, 189–194. [CrossRef] 21. Van Grieken, R.; Aguado, J.; López-Muñoz, M.J.; Marugán, J. Synthesis of size-controlled silica-supported TiO2 photocatalysts. J. Photochem. Photobiol. A Chem. 2012, 148, 315–322. [CrossRef] 22. Paul, B.; Martens, W.N.; Frost, R.L. Immobilised anatase on clay mineral particles as a photocatalyst herbicides degradation. Appl. Clay Sci. 2012, 57, 49–54. [CrossRef] 23. Baek, M.H.; Yoon, J.W.; Hong, J.S.; Suh, J.K. References 224 Catalysts 2019, 9, 115 3. Ali, I. New generation adsorbents for water treatment. Chem. Rev. 2012, 112, 5073–5091. [CrossRef] [PubMed] 4. Panizza, M.; Cerisola, G. Direct and mediated anodic oxidation of organic pollutants. Chem. Rev. 2009, 109, f 3. Ali, I. New generation adsorbents for water treatment. Chem. Rev. 2012, 112, 5073–5091. [CrossRef] [PubMed] 4 Panizza M ; Cerisola G Direct and mediated anodic oxidation of organic pollutants Chem Rev 2009 109 3. Ali, I. New generation adsorbents for water treatment. Chem. Rev. 2012, 112, 5073–5091. [CrossRef] [PubMed] 3. Ali, I. New generation adsorbents for water treatment. Chem. Rev. 2012, 112, 5073–5091. [CrossRef] [PubMed] 4. Panizza, M.; Cerisola, G. Direct and mediated anodic oxidation of organic pollutants. Chem. Rev. 2009, 109, 6541–6569. [CrossRef] [PubMed] 4. Panizza, M.; Cerisola, G. Direct and mediated anodic oxidation of organic pollutants. Chem. Rev. 2009, 109, 6541–6569. [CrossRef] [PubMed] 5. Huang, L.; Chen, J.; Gao, T.; Zhang, M.; Li, Y.; Dai, L.; Qu, L.; Shi, G. Reduced graphene oxide membranes for ultrafast organic solvent nanoﬁltration. Adv. Mater. 2016, 28, 8669–8674. [CrossRef] [PubMed] 6. Gratzel, M. Photocatalysis: Fundamentals and Applications; Serpone, N., Pelizzetti, E., Eds.; Wiley: New York, 5. Huang, L.; Chen, J.; Gao, T.; Zhang, M.; Li, Y.; Dai, L.; Qu, L.; Shi, G. Reduced graphene oxide membranes for ultrafast organic solvent nanoﬁltration. Adv. Mater. 2016, 28, 8669–8674. [CrossRef] [PubMed] 6. Gratzel, M. Photocatalysis: Fundamentals and Applications; Serpone, N., Pelizzetti, E., Eds.; Wiley: New York, NY, USA, 1989; p. 123. 7. Zheng, R.B.; Meng, X.W.; Tang, F.Q. Synthesis, characterization and photodegradation study of mixed-ph titania hollow submicrospheres with rough surface. Appl. Surf. Sci. 2009, 255, 5989–5994. [CrossRef] g, R.B.; Meng, X.W.; Tang, F.Q. Synthesis, characterization and photodegradation study of mixed-phase tania hollow submicrospheres with rough surface. Appl. Surf. Sci. 2009, 255, 5989–5994. [CrossRef] 8. Nishimoto, S.I.; Ohtani, B.; Kajiwara, H.; Kagiya, T. ChemInform abstract: Correlation of the crystal structure of titanium dioxide prepared from titanium tetra-2-propoxide with the photocatalytic activity for redox reactions in aqueous propan-2-ol and silver salt solutions. J. Chem. Soc. Faraday Trans. 1985, 16, 61–68. [CrossRef] 9. Fox, M.A.; Dulay, M.T. Heterogeneous photocatalysis. Chem. Rev. 1993, 93, 341–357. [CrossRef] 10. Tanaka, K.; Hisanaga, T.; Rivera, A.P. Photocatalytic Puriﬁcation and Treat Al-Ekabi, H., Eds.; Elsevier: Amsterdam, The Netherlands, 1993; p. 169. 10. Tanaka, K.; Hisanaga, T.; Rivera, A.P. Photocatalytic Puriﬁcation and Treatment of Water and Air; Ollis, D.F. References Application of TiO2-containing mesoporous spherical activated carbon in a ﬂuidized bed photoreactor—Adsorption and photocatalytic activity. Appl. Catal. A Gen. 2013, 45, 222–229. [CrossRef] 24. Chaturvedi, S.; Dave, P.N.; Shah, N.K. Applications of nano-catalyst in new era. J. Saudi Chem. Soc. 2016, 16, 307–325. [CrossRef] 25. Hsieh, S.H.; Chen, W.J.; Wu, C.T. Pt-TiO2/Graphene photocatalysts for degradation of AO7 dye under visible light. Appl. Surf. Sci. 2015, 340, 9–17. [CrossRef] 26. Li, D.; Jia, J.; Zhang, Y.; Wang, N.; Guo, X.; Yu, X. Preparation and characterization of Nano-graphite/TiO2 composite photoelectrode for photoelectrocatalytic degradation of hazardous pollutant. J. Hazard. Mater. 2016, 315, 1–10. [CrossRef] [PubMed] 27. Hamdi, A.; Bouﬁ, S.; Bouattour, S. Phthalocyanine/chitosan-TiO2 photocatalysts: Characterization and photocatalytic activity. Appl. Surf. Sci. 2015, 339, 128–136. [CrossRef] 225 Catalysts 2019, 9, 115 28. Lei, Y.; Zhang, C.; Lei, H.; Huo, J. Visible light photocatalytic activity of aromatic polyamide dendrimer/TiO2 composites functionalized with spirolactam-based molecular switch. J. Colloid Interface Sci. 2013, 406, 178–185. [CrossRef] 29. Mejía, MI.; Marín, JM.; Restrepo, G.; Rios, LA.; Pulgarín, C.; Kiwi, J. Preparation, testing and performance of a TiO2/polyester photocatalyst for the degradation of gaseous methanol. Appl. Catal. B 2010, 94, 166–172. [CrossRef] 30. Zhu, M.W.; Li, Y.J.; Chen, G.; Jiang, F.; Yang, Z.; Luo, X.G.; Wang, Y.B.; Lacey, S.D.; Dai, J.Q.; Wang, C.W.; et al. Tree-inspired design for high-efﬁciency water extraction. Adv. Mater. 2017, 29, 1704107. [CrossRef] 31. Liu, H.; Chen, C.J.; Chen, G.; Kuang, Y.D.; Zhao, X.P.; Song, J.W.; Jia, C.; Xu, X.; Hitz, E.; Xie, H.; et al. High-performance solar steam device with layered channels: Artiﬁcial tree with a reversed design. Adv. Energy Mater. 2017, 8, 1701616. [CrossRef] g g High-performance solar steam device with layered channels: Artiﬁcial tree with a reversed design. Adv. Energy Mater. 2017, 8, 1701616. [CrossRef] 32. Chen, C.; Li, Y.; Song, J.; Yang, Z.; Kuang, Y.; Hitz, E.; Jia, C.; Gong, A.; Jiang, F.; Zhu, J.Y.; et al. Highly ﬂexible and efﬁcient solar steam generation device. Adv. Mater. 2017, 29, 1701756. [CrossRef] 33. Sun, Q.F.; Lu, Y.; Tu, J.C.; Li, J. Bulky macroporous TiO2 photocatalyst with cellular structure via facile wood-template method. Int. J. Photoenergy 2013, 649540. [CrossRef] 33. Sun, Q.F.; Lu, Y.; Tu, J.C.; Li, J. Bulky macroporous TiO2 photocat wood-template method. Int. J. Photoenergy 2013, 649540. [CrossRef] 34. Gao, L.K.; Gan, W.T.; Li, J. Preparation of heterostructured WO3/TiO2 catalysts from wood ﬁbers and its versatile photodegradation abilities. Sci. Rep. Ru-Ti Oxide Based Catalysts for HCl Oxidation: The Favorable Oxygen Species and Inﬂuence of Ce Additive Jian Shi 1,2, Feng Hui 1,2, Jun Yuan 1,2, Qinwei Yu 1,2, Suning Mei 1,2, Qian Zhang 1,2, Jialin Li 1,2, Weiqiang Wang 1,2, Jianming Yang 1,2,* and Jian Lu 1,2,* 1 State Key Laboratory of Fluorine & Nitrogen Chemicals, Xi’an, Shaanxi 710065, China; a38860075@126.com (J.S.); huifeng.hyff@126.com (F.H.); luckyyjjun@163.com (J.Y.); qinweiyu204@163.com (Q.Y.); meisuning@aliyun.com (S.M.); qz450945428@163.com (Q.Z.); lijialin95126@gmail.com (J.L.); wqwang07611@163.com (W.W.) 2 Xi’an Modern Chemistry Research Institute, Xi’an, Shaanxi 710065, China * Correspondence: yangjm204@163.com (J.Y.); lujian204@gmail.com (J.L.); Te * Correspondence: yangjm204@163.com (J.Y.); lujian204@gmail.com (J.L.); Tel.: +86-29-88291367 (J.Y.) Received: 16 December 2018; Accepted: 17 January 2019; Published: 22 January 2019 Abstract: Several Ru-Ti oxide-based catalysts were investigated for the catalytic oxidation of HCl to Cl2 in this work. The active component RuO2 was loaded on different titanium-containing supports by a facile wetness impregnation method. The Ru-Ti oxide based catalysts were characterized by XRD, N2 sorption, SEM, TEM, H2-TPR, XPS, and Raman, which is correlated with the catalytic tests. Rutile TiO2 was conﬁrmed as the optimal support even though it has a low speciﬁc surface area. In addition to the interfacial epitaxial lattice matching and epitaxy, the extraordinary performance of Ru-Ti rutile oxide could also be attributed to the favorable oxygen species on Ru sites and speciﬁc active phase-support interactions. On the other hand, the inﬂuence of additive Ce on the RuO2/TiO2-rutile was studied. The incorporation of Ce by varied methods resulted in further oxidation of RuO2 into RuO2δ+ and a modiﬁcation of the support structure. The amount of favorable oxygen species on the surface was decreased. As a result, the Deacon activity was lowered. It was demonstrated that the surface oxygen species and speciﬁc interactions of the Ru-Ti rutile oxide were critical to HCl oxidation. Keywords: Ru-Ti oxide catalysts; HCl oxidation; oxygen species; Ce incorporation; active phase-support interactions References 2017, 7, 1102. [CrossRef] [PubMed] 35. Zheng, R.B.; Tshabalala, M.A.; Li, Q.Y.; Wang, H.Y. Construction of hydrophobic wood surfaces by room temperature deposition of rutile (TiO2) nanostructures. Appl. Surf. Sci. 2015, 28, 453–458. [CrossRef] 36. Chen, F.J.; Gong, A.S.; Zhu, M.W.; Hu, L.B. Mesoporous, three-dimensional wood membrane decorated with nanoparticles for highly efﬁcient water treatment. ACS Nano 2017, 11, 4275–4282. [CrossRef] [PubMed] 37. Jana, S.; Vanessa, S.; Tobias, K.; Ingo, B. Characterization of wood derived hierarchical cellulose scaffolds for multifunctional applications. Materials 2018, 11, 517. [CrossRef] 37. Jana, S.; Vanessa, S.; Tobias, K.; Ingo, B. Characterization of wood d multifunctional applications. Materials 2018, 11, 517. [CrossRef] 38. Li, H.Y.; Guo, X.L.; He, Y.M.; Zheng, R.B. A green, steam-modiﬁed deligniﬁcation method to low lignin deligniﬁed wood for thick, large, highly transparent wood composites. J. Mater. Res. 2018. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 226 www.mdpi.com/journal/catalysts catalysts catalysts 1. Introduction The treatment of the huge amount of excess hydrogen chloride byproduct has become a challenging and demanding problem in the chlorine-based chemical industry, as a byproduct HCl is environmentally undesirable and has a very restricted market [1,2]. The method using heterogeneously catalyzed HCl oxidation (Deacon process) to recycle chlorine is regarded as a low energy-consuming and sustainable route for the more efﬁcient Cl2 industry [1]. The reaction is exothermic and reversible, which is shown as follows. 4HCl + O2 cat. ↔2Cl2+2H2O ΔHr,298= −28.5 kJ·mol−1 HCl Ru-based catalysts are commonly considered the most active for this process. RuO2 supported on rutile TiO2 and SnO2 (cassiterite) with an excellent activity and outstanding lifetime have been successively reported by Sumitomo [2] and Bayer [3], respectively. Great attention has been paid to Ru-Ti oxide catalysts in various catalytic reactions besides the Deacon process, including oxidation of propane [4], N2O decomposition [5], selective methanation of CO [6], CO oxidation [7], and Catalysts 2019, 9, 108; doi:10.3390/catal9020108 227 www.mdpi.com/journal/catalysts Catalysts 2019, 9, 108 aqueous-phase ketonization of acetic acid [8,9]. The Ce-based catalysts also exhibit Deacon activity and outstanding stability in the process [10,11]. Related studies have demonstrated that CeO2 can accelerate the catalyst reoxidation step by supplying oxygen donor/storage sites [10]. Although the single Ce-based or Cu-based catalyst shows a limited activity, the combination of Ce-Cu can boost the overall HCl oxidation performance [12,13]. The build-up of Ce-Ti oxide can also enhance the oxygen storage capacity [14,15]. The CeO2/TiO2 catalyst showed excellent activity for selective catalytic reduction of NO with NH3, where the Ce-O-Ti species were conﬁrmed to be the active sites [16–19]. The concentration of surface adsorbed oxygen presented a positive correlation with the catalytic activity in the NH3-SCR reaction [16,20]. However, the combination of Ru-Ce-Ti for HCl catalytic oxidation has not been reported to the best of our knowledge. Despite a number of studies concerning supported Ru-based catalysts for one-step HCl oxidation [21,22], there are only a few research studies about the details of active phase-support interactions in the Ru-Ti oxide-based catalysts [2,23]. Moreover, little attention was paid to the inﬂuence of oxygen species in this catalytic process. In addition to the lattice matching and epitaxial growth of RuO2 on the substrate [2,21], the extraordinary performance of Ru-Ti rutile oxide for the Deacon process still requires a more sufﬁcient interpretation. 1. Introduction In this article, the performance of different shaped Ru-Ti oxide based catalysts, resembling the forms of the industrial reality, are compared in the Deacon process. Detailed characterizations are performed to investigate the special oxygen species and interactions of the catalysts. The favorable oxygen species and speciﬁc active phase-support interactions ensure high Deacon activity of RuO2/TiO2-rutile. The role of Ce in the Ru-Ti rutile oxide system was also investigated. The addition of Ce decreased favorable oxygen species and affected the active phase-support interactions between RuO2 and TiO2, by evolving the Ru-O-Ce structure and enhancing the positive charge density of Ru sites. The Deacon activity was lowered as a result. It can be deduced that the electronic interactions between RuO2 and rutile TiO2 are critical for the gas-phase oxidation of HCl to Cl2. The ﬁndings in this work may be a reference value for the design and tailor of Ru-Ti oxide based catalysts toward better Deacon activity. 2.1. Morphology and Phase Structure The crystal structures of the catalysts were analyzed by X-ray diffraction (XRD). The XRD patterns of the supported catalysts only exhibit characteristic diffraction peaks of the TiO2 supports (Figure 1). Due to the low loading and high dispersion of the Ru species, the RuO2 phase were logically not detected. In the XRD pattern of RuO2/TiO(OH)2, much broader peaks can be observed than those of RuO2/TiO2-a, which implies that the support that originated from the TiO(OH)2 precursor has a much lower crystallinity as well as average crystallite size. 228 Catalysts 2019, 9, 108 5X27L22+ 5X27L2D ♦7L2DQDWDVH •7L2UXWLOH ♦ ♦ ♦ ♦ ♦ ♦♦ ♦ ♦ ♦ ♦ ♦ • • • • •• • • • • • • ♦ ,QWHQVLW\DX θ (°) • 5X27L2U Figure 1. XRD patterns of the RuO2/TiO2 catalysts with the supports of: TiO2-rutile, TiO2-anatase, and TiO(OH)2 (as the precursor). ,QWHQVLW\DX Figure 1. XRD patterns of the RuO2/TiO2 catalysts with the supports of: TiO2-rutile, TiO2-anatase, and TiO(OH)2 (as the precursor). The SEM micrographs in Figure 2 show that the anatase TiO2 and the corresponding catalysts (35 to 50 nm) have a smaller particle size than the rutile of 45 to 60 nm. The particle size of rutile TiO2 has not changed much after loading RuO2 (Figures 2a and 2c), while some coagulation can be observed on anatase TiO2 and the average size has increased from 40.3 nm to 48.5 nm (Figures 2b and 2d). The supported catalysts were also scrutinized by TEM. As displayed in Figure 3a, some dark edges and layers were observed on the substrate of RuO2/TiO2-r. They were presumed to be the dispersed RuO2 phase, which was consistent with the literature [2,4]. On the other hand, some aggregation can be observed in RuO2/TiO2-a (Figure 3b). The EDX elemental mapping proves the existence of highly dispersed Ru species on the TiO2-r support (Figure 4). 2.2.1. H2-TPR Analysis Temperature programmed reduction (TPR) by H2 was employed in this study to distinguish speciﬁc oxygen species and estimate the oxygen storage capacity [11,14]. The H2-TPR proﬁles of the bulk RuO2 and supported catalysts are shown in Figure 5. In this scenario, all the supported catalysts were loaded with an Ru content of 0.5 wt%. The bulk RuO2 sample was reduced in the range of 130 to 230 ◦C. It should be noted that the RuO2 sample used in characterization was obtained from RuCl3·3H2O calcined at 350 ◦C for 8 h, the phase of which was veriﬁed by XRD (see Figure S1). The RuO2 phase prepared by this method has a preferential (1 0 1) plane rather than (1 1 0) (PDF #65-2824). It may be the origin of different reduction temperatures as discussed in the previous study [21]. The H2-TPR proﬁle of RuO2/TiO2-r contains three peaks in the range of 90 to 200 ◦C and one peak from 330 to 450 ◦C. The former three peaks are assigned to the reduction of RuO2, while the last one is attributed to the partial reduction of the TiO2 surface [24]. Ru-Ce/TiO2-r and RuO2/TiO2-a exhibit a similar reduction peak of titania, except that the peak shifts to a higher temperature were similar to RuO2/TiO2-r. Note the peak in the range 370 to 500 ◦C of Ru-Ce/TiO2-r includes the reduction of ceria. The H2-TPR proﬁles of pure TiO2 in rutile and anatase phase are shown in Figure S2, which conﬁrms the partial reduction of TiO2 support. ႏ ႏ ႏ 5X&H7L2U 5X2 5X27L2U 5X27L2D ,QWHQVLW\DX 7HPSHUDWXUH°& Figure 5. H2-TPR proﬁles of the bulk and supported RuO2 catalysts. Figure 5. H2-TPR proﬁles of the bulk and supported RuO2 catalysts. In the range of 50 to 200 ◦C, the oxygen species of RuO2 were subjected to reduction. Generally, the reduction temperature of the same phase is related with the particle size. In fact, oxygen species with different reducibility can be a more essential perspective. We know that smaller particles expose more surface species. For our catalysts, RuO2 phase is mainly distributed on the surface of TiO2 support and more surface species mean more surface oxygen species. Among these oxygen species, it is quite probable that species with more coordination numbers are more difﬁcult to reduce. 2.1. Morphology and Phase Structure 澜濕澝澔澜濖澝澔澜濗澝澔澜濘澝澔濈濃濃瀁瀀澳濈濃濃瀁瀀澳濈濃濃瀁瀀澳濈濃濃瀁瀀澳 0HDQQP 3HUFHQWIUHTXHQF\ 3DUWLFOHVL]HQP 0HDQQP 3HUFHQWIUHTXHQF\ 3DUWLFOHVL]HQP 0HDQQP 3HUFHQWIUHTXHQF\ 3DUWLFOHVL]HQP 0HDQQP 3HUFHQWIUHTXHQF\ 3DUWLFOHVL]HQP Figure 2. SEM micrographs of the rutile (a) and anatase TiO2 supports (b) used in this study, RuO2/TiO2-r (c), RuO2/TiO2-a (d). Figure 2. SEM micrographs of the rutile (a) and anatase TiO2 supports (b) used in this study, RuO2/TiO2-r (c), RuO2/TiO2-a (d). 229 Catalysts 2019, 9, 108 澜濕澝澔澜濖澝澔 Figure 3. TEM images of RuO2/TiO2-r (a) and RuO2/TiO2-a (b). Figure 3. TEM images of RuO2/TiO2-r (a) and RuO2/TiO2-a (b). 澜濕澝澔濈濝澔澿煇澥澔澜濖澝澔濃澔澿煇澥澔澜濗澝澔濆濩澔激煇澥澔澜濘澝澔澜濙澝澔 ůĞŵĞŶƚ tĞŝŐŚƚ;йͿ ƚŽŵŝĐ;йͿ tĞŝŐŚƚĞƌƌŽƌ;йͿ dŝ;<Ϳ ϱϲ͘ϳϬ ϯϮ͘ϭϰ Ϭ͘Ϯϰ K;<Ϳ ϯϵ͘ϯϵ ϲϲ͘ϴϭ Ϭ͘Ϯϰ ZƵ;<Ϳ ϯ͘ϵϭ ϭ͘Ϭϱ Ϭ͘ϭϳ Figure 4. TEM-EDX elemental mapping of RuO2/TiO2-r: (a) representative TEM image, (b) Ti (Kα1) green color, (c) O (Kα1) blue color, (d) Ru (Lα1) pink color, and (e) EDX (Energy-dispersive X-ray) result of the selected area. 澜濙澝澔 ůĞŵĞŶƚ tĞŝŐŚƚ;йͿ ƚŽŵŝĐ;йͿ tĞŝŐŚƚĞƌƌŽƌ;йͿ dŝ;<Ϳ ϱϲ͘ϳϬ ϯϮ͘ϭϰ Ϭ͘Ϯϰ K;<Ϳ ϯϵ͘ϯϵ ϲϲ͘ϴϭ Ϭ͘Ϯϰ ZƵ;<Ϳ ϯ͘ϵϭ ϭ͘Ϭϱ Ϭ͘ϭϳ Figure 4. TEM-EDX elemental mapping of RuO2/TiO2-r: (a) representative TEM image, (b) Ti (Kα1) green color, (c) O (Kα1) blue color, (d) Ru (Lα1) pink color, and (e) EDX (Energy-dispersive X-ray) result of the selected area. 230 Catalysts 2019, 9, 108 2.2. Characterization of Oxygen Species and Interfacial Interactions 2.2.1. H2-TPR Analysis 2.2.1. H2-TPR Analysis For sample RuO2/TiO2-r, the peaks (from 90 to 200 ◦C) are evidently distinguished from the other two catalyst samples. We deduce that the three peaks are assigned to the reduction of top oxygen, bridge oxygen, and bulk oxygen of RuO2 with H2, from a low to a high reduction temperature. These three types of oxygen are coordinated to 1, 2, 3 Ru atoms, respectively. The former two oxygen species are located at the surface and are signiﬁcant to the Deacon process [21,25,26]. For sample RuO2/TiO2-a, only one reduction peak was detected in the range from 90 to 200 ◦C, which was 231 Catalysts 2019, 9, 108 attributed to the elimination of bulk oxygen. It could be rationalized in the following way. RuO2 could not grow epitaxially on TiO2-a due to a huge difference of lattice matching. Thus, the RuO2 active phase mainly exists as bigger particles on TiO2-a rather than ﬁlms in RuO2/TiO2-r [2]. In this case, the bulk oxygen species of RuO2 prevailed, which was coordinated to 3 Ru atoms and was the most difﬁcult to reduce. From Figure 5, it can be observed that Ru-Ce/TiO2-r and RuO2/TiO2-a have a better oxygen storage capacity than RuO2/TiO2-r, especially in the high temperature range. attributed to the elimination of bulk oxygen. It could be rationalized in the following way. RuO2 could not grow epitaxially on TiO2-a due to a huge difference of lattice matching. Thus, the RuO2 active phase mainly exists as bigger particles on TiO2-a rather than ﬁlms in RuO2/TiO2-r [2]. In this case, the bulk oxygen species of RuO2 prevailed, which was coordinated to 3 Ru atoms and was the most difﬁcult to reduce. From Figure 5, it can be observed that Ru-Ce/TiO2-r and RuO2/TiO2-a have a better oxygen storage capacity than RuO2/TiO2-r, especially in the high temperature range. 2.2.2. XPS Analysis XPS analysis was performed to study the surface species and electronic structure of the catalyst samples. The survey spectra veriﬁed the complete removal of chlorine in all the catalysts (not shown). The XPS peaks of O 1s were deconvoluted to analyze the different types of O species in the supported catalysts (Figure 6). The O 1s peaks were mainly composed of signals corresponding to the chemisorbed oxygen (Oα) and the lattice oxygen (Oβ1, Oβ2) [27–29]. XPS data of chemisorbed oxygen are listed in Table 1. The proportion of chemisorbed oxygen (Oα) on the surface exhibited a dependence on Ru loading (Figure 6a–c). RuO2/TiO2-r exhibited a higher amount of chemisorbed oxygen with the increase of Ru loading, while a much lower content of chemisorbed oxygen was detected in RuO2/TiO2-a (Figure 6d, Table 1). The Ti 2p core-level spectra of RuO2/TiO2-r also show a relevance with the Ru content (Figure S3), where the XPS peaks are broadened and shifted to a lower binding energy with the increase of ruthenium. The chemical environment change of the Ti sites can be ascribed to the interactions and electronic effects among Ti, Ru, and O atoms. A slight interfacial charge transfer from RuO2 to TiO2-r may lead to the binding energy shift of the Ti 2p peaks [30,31]. (D) α %LQGLQJ(QHUJ\H9 β1 β2 (E) α %LQGLQJ(QHUJ\H9 β1 β2 (F) α %LQGLQJ(QHUJ\H9 β1 β2 (G) α %LQGLQJ(QHUJ\H9 β1 β2 Figure 6. XPS proﬁles and ﬁtting curves of O 1s peaks: RuO2/TiO2-r of (a) 0.3, (b) 0.5, (c) 1.0 wt% Ru, RuO2/TiO2-a of (d) 0.5 wt% Ru. (E) α %LQGLQJ(QHUJ\H9 β1 β2 %LQGLQJ(QHUJ\H9 (G) α %LQGLQJ(QHUJ\H9 β1 β2 %LQGLQJ(QHUJ\H9 %LQGLQJ(QHUJ\H9 %LQGLQJ(QHUJ\H9 %LQGLQJ(QHUJ\H9 Figure 6. XPS proﬁles and ﬁtting curves of O 1s peaks: RuO2/TiO2-r of (a) 0.3, (b) 0.5, (c) 1.0 wt% Ru, RuO2/TiO2-a of (d) 0.5 wt% Ru. 232 232 Catalysts 2019, 9, 108 Table 1. Chemisorbed oxygen (Oα) in the RuO2/TiO2 catalysts. 2.2.2. XPS Analysis Sample Eb of Oα (eV) Oα/OT (%) 0.3 wt%-RuO2/TiO2-r 532.11 18.29 0.5 wt%-RuO2/TiO2-r 532.27 20.12 1.0 wt%-RuO2/TiO2-r 532.57 23.70 0.5 wt%-RuO2/TiO2-a 532.36 16.51 Table 1. Chemisorbed oxygen (Oα) in the RuO2/TiO2 catalysts. The Ru 3d spectra for RuO2-CeO2/TiO2-r catalysts are presented in Figure 7. The signal of Ru 3d5/2 core-level, attributed to RuO2 or RuO2δ+, was detected in the region of 281.2–282.9 eV. Meanwhile, the peaks around 280.5 eV appeared for Ru-2Ce/TiO2-r and Ru-2Ce-C/TiO2-r, which were assigned to Ru0 [3]. The peaks of C 1s and Ru 3d3/2 appeared to overlap [32,33]. The Ru 3d5/2 peaks shifted toward a higher binding energy when the preparation methods were altered. In the spectra of Ce 3d, an overall shift towards lower binding energies was observed in the similar sequence of Ru-2Ce-R/TiO2-r, Ru-2Ce/TiO2-r, and Ru-2Ce-C/TiO2-r (Figure S4). It can be inferred that electrons are transferred from RuO2 to CeO2, which results in the further oxidation of RuO2 into RuO2δ+. As shown in Figure 5, the reduction peaks of Ru-Ce/TiO2-r in the low temperature range (90 to 200 ◦C) are one fewer than those of RuO2/TiO2-r, which is ascribed to the elimination of top oxygen on the RuO2 surface (vide supra). The existence of top oxygen is critical to the Deacon reaction with Ru-based catalysts [25,26]. In RuO2-CeO2/TiO2-r, Ru-O-Ce linkage was likely formed, which induced the decrease of the active sites in the Ru-Ti rutile oxide system. The generation of Ru0 in Ru-2Ce/TiO2-r and Ru-2Ce-C/TiO2-r also indicated the decrease of the active sites for HCl oxidation. The decline of chemisorbed oxygen from XPS data is also consistent with the deduction above (see Figure S5, Table S1). ,QWHQVLW\DX %LQGLQJ(QHUJ\H9 &V5XG 5XG 5X27L2U 5X&H57L2U 5X&H7L2U 5X&H&7L2U Figure 7. XPS spectra and peak ﬁtting curves of Ru 3d of the supported RuO2 and Ru-Ce/Ti oxide catalysts. ,QWHQVLW\DX Figure 7. XPS spectra and peak ﬁtting curves of Ru 3d of the supported RuO2 and Ru-Ce/Ti oxide catalysts. 2.2.3. Raman Analysis The active phase-support interactions were further conﬁrmed by Raman characterization. As shown in Figure 8a, the characteristic bands of rutile TiO2 are observed at 234, 441, and 606 cm−1, which can be assigned to the multiple photon scattering process, the Eg (planar O-O vibration), and A1g (Ti-O stretch) Raman-active modes, respectively [32,34,35]. Since the RuO2 Raman bands overlapped with those of TiO2-r, only the Eg mode of RuO2 at 515 cm−1 could be distinguished [5], which indicates the existence of the RuO2 phase in the catalyst. The declination of Raman signals of TiO2-r was 233 Catalysts 2019, 9, 108 ascribed to the decreased amount of Ti-O-Ti structure, which resulted from the formation of Ru-O-Ti structure with the increase of Ru loading. The relative intensity of the RuO2 Raman signal at 515 cm−1 was enhanced at the same time, which coincided with the change of the active component loading. Moreover, the RuO2 Raman bands shifted from 515 to 507 cm−1 with the increase of Ru loading, while a clear blue-shift towards higher wavenumbers (from 234 cm−1 to 258 cm−1) was observed in the spectrum of the support (Figures 8b and 8c). The Raman shifts suggest that the rutile structure of the TiO2 support was modiﬁed according to the changed RuO2 crystal size and interfacial interactions, substantially as a result of the mechanical strains generated from the differences between the rutile phase of RuO2 and TiO2 support [6]. The formation of Ru-O-Ti linkage corresponded with the results from XPS analysis. The interactions between Ce and Ti were also conﬁrmed by Raman characterization on the set of Ce-containing RuO2/TiO2-r catalysts. The characteristic bands of planar Ti-O vibration and O-O stretch for TiO2-r exhibited a slight blue shift (see Figure S6) in accordance with the binding energy shift of the Ru 3d5/2 core-level peak. It indicated that the chemical environment of the rutile support was also affected by Ce addition. D ,QWHQVLW\DX 5DPDQVKLIWFP 7L2U E ,QWHQVLW\DX 5DPDQVKLIWFP 7L2U F 5DPDQVKLIWFP Figure 8. 2.2.3. Raman Analysis Raman spectra of TiO2-r and RuO2/TiO2-r with 0.1, 0.3, 0.5, and 1.0 wt% Ru loading (a) and partial enlarged views of TiO2-r (b) and RuO2 (c). D ,QWHQVLW\DX 5DPDQVKLIWFP 7L2U E ,QWHQVLW\DX 5DPDQVKLIWFP 7L2U F 5DPDQVKLIWFP ,QWHQVLW\DX 5DPDQVKLIWFP 5DPDQVKLIWFP Figure 8. Raman spectra of TiO2-r and RuO2/TiO2-r with 0.1, 0.3, 0.5, and 1.0 wt% Ru loading (a) and partial enlarged views of TiO2-r (b) and RuO2 (c). 234 234 Catalysts 2019, 9, 108 2.3. Catalytic Performance of Ru-Ti Oxide Based Catalysts 2.3.1. Catalytic Activity of RuO2/TiO2-r Catalyst 2.3.1. Catalytic Activity of RuO2/TiO2-r Catalyst 2.3.2. Comparison of Ru-Ti Oxide Based Catalysts The catalysts without sufﬁxes above are prepared by co-impregnation with Ru and Ce. g Calculated based on the mole of HCl reacted per hour per mole of Ru. Herein, we attempt to explicate the inﬂuence of favorable oxygen species and interfacial interactions of the Ru-Ti oxide system on the Deacon process, by correlating the characterization results with catalytic performances. With the increase of Ru loading, the linkage of Ru-O-Ti appeared to be more abundant, which was conﬁrmed by the results from XPS and Raman spectra. Subsequently, the amount of chemisorbed oxygen species increased. The chemisorbed oxygen is intimately related with coordinatively unsaturated ruthenium atoms [26], which provide critical active sites and promote oxygen activation. It can partially explain the higher activity of RuO2 when loaded on TiO2-r. Since RuO2 can grow epitaxially on the rutile titania, the exposure of more active sites is favored. A series of Ru-Ce-Ti oxide catalysts were prepared by different methods, as described in the experiment. From Table 2, it can be afﬁrmed that the incorporation of Ce reduced the activity of Ru-Ti rutile oxide catalysts. The catalytic activity declined with the increase of Ce loading by comparing the results of No. 4, 5, and 7 in Table 2. When exchanging the impregnation sequence of Ru and Ce, it was found that performing Ru impregnation in the ﬁrst place was beneﬁcial for improving the catalytic performance to some extent (see No. 6, 7, and 8). 2.3.1. Catalytic Activity of RuO2/TiO2-r Catalyst The inﬂuence of Ru loading on HCl conversion is depicted in Figure 9a. It can be noted that the increase of Ru loading contributes to the reaction conversion to a certain extent. The catalysts with 0.5 wt% and 1.0 wt% Ru loadings achieved similar conversion. As presented in Figure 9b, the ratio of O2/HCl also plays a critical role in the reaction, especially when the value is less than 1.0 vol./vol. Since oxygen re-adsorption is recognized as the rate-determining step under lean oxygen condition, a higher O2 partial pressure has been proven to be beneﬁcial for Cl2 production [26]. [+&O 5XORDGLQJEDVHGRQWKHVXSSRUWZW D E [+&O )HHG2+&OUDWLR Figure 9. The inﬂuence of Ru loading at 350 ◦C (a) and feed O2/HCl ratio at 320 ◦C (b) on HCl conversion. Note volumetric ﬂowrate HCl/O2 = 1:2 for (a). 5XORDGLQJEDVHGRQWKHVXSSRUWZW Figure 9. The inﬂuence of Ru loading at 350 ◦C (a) and feed O2/HCl ratio at 320 ◦C (b) on HCl conversion. Note volumetric ﬂowrate HCl/O2 = 1:2 for (a). Figure 10 shows HCl conversion at different reaction temperatures on catalysts with 0.5 wt% and 1.0 wt% Ru loading. For both catalysts, the conversion was improved with the elevation of the reaction temperature. When comparing the two catalysts with different Ru loadings over 300 ◦C, we found that the conversion was not proportional to the loading amount of the active component. It implied that the Ru-speciﬁc activity declined with the increasing loading. As indicated by Figure S3, a slight interfacial charge transfer from RuO2 to TiO2-r shows up with the increase of Ru content. This charge transfer was deduced not to be beneﬁcial to Ru-speciﬁc activity. [+&O 7HPSHUDWXUH°& ZW ZW Figure 10. HCl conversion on different reaction temperatures and Ru loading. [+&O 7HPSHUDWXUH°& ZW ZW Figure 10. HCl conversion on different reaction temperatures and Ru loading. 235 235 Catalysts 2019, 9, 108 2.3.2. Comparison of Ru-Ti Oxide Based Catalysts 2.3.2. Comparison of Ru-Ti Oxide Based Catalysts The catalytic activities for HCl oxidation over Ru-Ti oxide based catalysts using different supports are compared in Table 2. Although RuO2/TiO2-r has the smallest surface area (28 m2·g−1) among all the Ru-Ti oxide based catalysts, its catalytic activity turned out to be the best. Notably, the speciﬁc surface area of the supports seemed to be less critical in the RuO2/TiO2-based catalytic system. The performances of catalysts differed signiﬁcantly when the support was changed, even though the supports were all based on the Ti-O structure. Table 2. Characterization and catalytic activity data. Reaction Temperature No. Catalyst a BET Surface Area c (m2·g−1) HCl Conversion (%) STY d (gCl2·gRu −1·h−1) TOF (h−1) g 350 ◦C 1 RuO2/TiO2-r 28 (28) 93.5 57.3 163.2 2 RuO2/TiO2-a 44 (79) 53.6 32.8 93.4 3 RuO2/TiO(OH)2 b 193 (309) 21.1 12.9 36.7 320 ◦C 4 RuO2/TiO2-r 28 81.0 49.6 141.3 5 Ru-Ce/TiO2-r e 27 74.3 45.5 129.6 6 Ru-2Ce-R/TiO2-r f 26 73.2 44.8 127.6 7 Ru-2Ce/TiO2-r 27 66.3 40.6 115.6 8 Ru-2Ce-C/TiO2-r 27 61.8 37.9 107.9 a 0.5 wt% Ru loading based on the support or support precursor. b The precursor of support is TiO(OH)2. c Determined by N2 adsorption, surface area of the support in brackets. d The space time yield deﬁned as the gram of Cl2 produced per gram of Ru per hour. e The molar ratio of Ru/Ce is 1 and 0.5, denoted as Ru-Ce and Ru-2Ce respectively. f The catalysts prepared by different methods are distinguished by sufﬁxes –R and -C, which refer to the impregnation of TiO2-r support with Ru ﬁrst and Ce ﬁrst, respectively. The catalysts without sufﬁxes above are prepared by co-impregnation with Ru and Ce. g Calculated based on the mole of HCl reacted per hour per mole of Ru. Table 2. Characterization and catalytic activity data. g pp pp p p pp ( )2 c Determined by N2 adsorption, surface area of the support in brackets. d The space time yield deﬁned as the gram of Cl2 produced per gram of Ru per hour. e The molar ratio of Ru/Ce is 1 and 0.5, denoted as Ru-Ce and Ru-2Ce respectively. f The catalysts prepared by different methods are distinguished by sufﬁxes –R and -C, which refer to the impregnation of TiO2-r support with Ru ﬁrst and Ce ﬁrst, respectively. 2.3.3. The Inﬂuence of Ce on the Ru-Ti Rutile Oxide System The characterizations and catalytic performances of Ce-containing RuO2/TiO2-r catalysts further conﬁrmed the signiﬁcance of the active phase-support interactions for the RuO2/TiO2-r system. When Ce was introduced to the TiO2-r support prior to Ru, a greater change of the RuO2 and TiO2-r structure was incurred. The formation of Ti-O-Ce and Ru-O-Ce linkages remarkably affected the interfacial interactions and electronic structure of the RuO2/TiO2-r system, which was corroborated by XPS and Raman characterizations. The new linkages restricted the active sites of coordinatively unsaturated ruthenium and the transport of oxygen species. The amount of chemisorbed oxygen evidently decreased when improving the priority of the introduction of Ce (Figure S4, Table S1). The correlated catalytic activity declined, as presented in Table 2 (No. 4, 6, 7, and 8). 236 Catalysts 2019, 9, 108 The introduction of Ce triggered further oxidation of RuO2 into RuO2δ+ and could be unfavorable for H2O desorption and Cl recombination. Higher positive charge density of Ru sites induced easier adsorption of the reactants. Therefore, the active-phase surface was more likely to be poisoned by adsorbates. Since HCl oxidation proceeds on RuO2 via a Langmuir-Hinshelwood reaction mechanism, adsorbed HCl dehydrogenates through a hydrogen transfer to produce Cl and OH species in on-top positions [26]. The recombination of neighboring on-top Cl atoms to form the desired Cl2 product is regarded as the rate-determining step. Nevertheless, the existence of Ce strengthened the dissociative adsorption of HCl so that the liberation of Cl2 restricted the activity, which was also observed on the IrO2(110) surface [36,37]. Although Ce-based catalysts showed Deacon activity themselves, the active temperature for the Deacon process was generally reaching 430 ◦C or more, considering the higher energy requirements for Cl activation and recombination [10,11]. Therefore, ceria itself contributed to little Deacon activity for RuO2/TiO2-r below 350 ◦C in this study. On the other hand, although Ce provided more reducible oxygen species (Figure 5), it seemed that the oxygen species with a high reduction temperature were not crucial for the Deacon reaction. The oxygen species of support with a lower reduction temperature (330 to 350 ◦C) in RuO2/TiO2-r, as a result of interfacial interactions between the phases, was speculated to be beneﬁcial for facilitating activation and transport of oxygen species for the active phase. Moreover, compared to RuO2/TiO2-r, the most readily reducible oxygen species (90 to 100 ◦C) in Ru-Ce/TiO2-r disappeared. 2.3.3. The Inﬂuence of Ce on the Ru-Ti Rutile Oxide System The on-top oxygen occupying the coordinatively unsaturated ruthenium sites are mentioned in the discussion of the H2-TPR results. The introduction of Ce enhanced the positive charge density of Ru sites, which might cause the easier formation of the bridge and bulk oxygen of RuO2 other than the on-top oxygen. Because of the higher positive charge density of Ru sites, O was more inclined to bond to Ru atoms and the coordination number of O with Ru on the RuO2 surface likely increased. More evidence may be provided by further characterizations on ﬁne structure and corresponding computational studies. 3.3. Catalytic Tests The catalytic oxidation of HCl to Cl2 was investigated in a Hastelloy alloy (HC-276®) ﬁxed-bed reactor with a diameter of 30 mm at an ambient pressure. The upstream lines of the set-up were also made from Hastelloy alloy (HC-276®) in order to prevent the corrosion of the reactor, while the downstream lines were made from Teﬂon® to improve corrosion resistance. In addition, a 25-g shaped cylinder catalyst was loaded into the reactor. Thereafter, the reaction feed (HCl ﬂow = 80 cm3 STP min−1 and volumetric ﬂowrate HCl/O2 = 1:2, unless otherwise speciﬁed) was continuously introduced. The reaction temperature was controlled in the range of 330 to 430 ◦C (±1.0 ◦C). The data obtained by each test were the average of at least three steady-state measurements. The total chlorine balance was conﬁrmed with an accuracy of ±2%. Blank support without a loading active component showed negligible activity under the corresponding reaction condition. The reaction efﬂuent was absorbed by excessive potassium iodide solution and analyzed by iodometry and acid-base titration to measure the generated Cl2 and unreacted HCl. The conversion of HCl was calculated based on the detected results. 3.2. Characterization of Catalysts 3.2. Characterization of Catalysts Powder X-ray diffraction (XRD) patterns were recorded with an Empyrean, PANalytical X-ray diffractometer (Almelo, The Netherlands), with Cu Kα radiation (λ = 0.154056 nm) at 40 kV and 40 mA. The diffraction patterns were taken in the 2θ range of 5 to 90◦with a step size of 0.02◦. Speciﬁc surface areas of the samples were measured by N2 physisorption at 77 K using a Micromeritics ASAP 2020 instrument (Norcross, GA, USA). The surface area was determined by the Brunauer-Emmett-Teller (BET) method. The morphology and particle size of prepared catalysts and supports were studied by a scanning electron microscope (SEM, FEI Quanta 600FEG, operated at 20 kV, Hillsboro, OR, USA) and a transmission electron microscope (TEM, FEI Tecnai G2 F20, operated on 200 kV, Hillsboro, OR, USA). The energy-dispersive X-ray spectroscopy (EDX, Oxford INCA Energy IE350, Oxford, UK) mapping method was applied to determine the elemental distributions of different components in the catalysts. Temperature programmed reduction of hydrogen (H2-TPR) was performed at an AutoChem II 2950 instrument (Micromeritics, Norcross, GA, USA) equipped with a thermal conductivity detector (TCD). Furthermore, 100 mg catalyst was heated and programmed from 50 ◦C to 550 ◦C (or 800 ◦C for the pure support) at a rate of 10 ◦C·min−1 in a gas ﬂow of 5 vol.% H2/Ar of 50 cm3 STP min−1. Raman spectra were collected on a confocal Raman microscope (inVia Raman Microscope, Renishaw plc, Wotton-under-Edge, UK) with a 785 nm laser diode (Renishaw plc, Wotton-under-Edge, UK). X-ray photoelectron spectra (XPS) were recorded on a Thermo ESCALAB 250 spectrometer (Waltham, MA, USA) with a monochromatized Al Kα (1486.6 eV) radiation and a passing energy of 50 eV. The binding energies were calibrated by the C 1s signal of adventitious carbon at 284.8 eV. All the characterizations were performed on catalyst samples with a 0.5% Ru loading unless otherwise speciﬁed. 3.1. Preparation of Catalysts All the reagents were of an analytical grade, supplied by Aladdin (Shanghai, China), and used as received without further puriﬁcation, except for TiO(OH)2 from Tuoboda Titanium Dioxide Products Co. (Wuxi, China). Ru-Ti oxide based catalysts were prepared by a facile wetness impregnation method as follows. First, RuCl3·3H2O was dissolved in a mixed solution with an equal volume of water and ethanol. After 3 min of ultrasonic mixing, support or the support precursor was added and the suspension was stirred for 16 h at room temperature. Then the mixture was evaporated under vacuum in a rotary evaporator. The obtained powder was dried at 120 ◦C for 12 h and was then tableted into cylinders with a diameter of 5 mm. Lastly, the sample were calcined at 350 ◦C in static air for 8 h. The nominal loading of Ru was 0.1, 0.3, 0.5, and 1.0 wt% on the support basis. The supports and support precursors included rutile and anatase TiO2 and TiO(OH)2. Henceforth, the rutile and anatase TiO2 polymorphs are abbreviated as TiO2-r and TiO2-a, respectively. The RuO2-CeO2/TiO2-r catalysts were all loaded with 0.5 wt% Ru on the rutile TiO2 basis. They were prepared by almost the same method as RuO2/TiO2 except that Ce(NO3)3·6H2O was introduced to the solution when dissolving RuCl3·3H2O. The molar ratios of Ru/Ce were 1.0 and 0.5, which were denoted as Ru-Ce/TiO2-r and Ru-2Ce/TiO2-r, respectively. In order to investigate the effect of Ce on the Ru-Ti oxide system, the RuO2-CeO2/TiO2-r catalysts were also prepared by changing the sequence of Ce introduction, where the molar ratios of Ru/Ce were 0.5. The powder from the rotary evaporator (vide supra) was added to the solution of Ce(NO3)3·6H2O. Then the suspension was stirred, evaporated, dried, tableted, and calcined as described before. The impregnation sequence of Ru and Ce was also exchanged for obtaining another catalyst. These two catalysts were distinguished by sufﬁxes -R and -C, namely Ru-2Ce-R/TiO2-r and Ru-2Ce-C/TiO2-r. The catalysts without the sufﬁxes were prepared by co-impregnation with Ru and Ce. 237 Catalysts 2019, 9, 108 4. Conclusions A series of Ru-Ti oxide based catalysts have been investigated for HCl oxidation in this research. It was clariﬁed that the special oxygen species and active phase-support interactions of RuO2/TiO2 were signiﬁcant for the Deacon process. RuO2 ﬁlm grows epitaxially on rutile TiO2. This produces more active sites and oxygen species on the catalyst surface. More importantly, the assembly of RuO2 and rutile TiO2 generates coordinatively unsaturated ruthenium sites and bridge oxygen, which are efﬁcient for the Deacon reaction. The interactions of Ru-Ti were conﬁrmed by characterizations and correlated to the amount of the active component. The reducible oxygen species of the rutile TiO2 may facilitate activation and transport of oxygen species during the active phase. It is inferred that the rutile support is likely involved in the catalytic reaction rather than merely acting as an inert support. On the other hand, the incorporation of Ce altered the electronic structure of the RuO2/TiO2-r system. The formation of Ru-O-Ce linkage decreased the amount of favorable oxygen species and 238 Catalysts 2019, 9, 108 increased the positive charge of Ru sites, which restricted the recombination of Cl atoms and Cl2 elimination. To achieve a better catalytic performance, a more delicate tuning of the RuO2/TiO2-r system by considering moderate positive charge density of Ru sites is required. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/2/108/s1, Figure S1: XRD patterns of the self-made RuO2 and the corresponding intensity line in red from PDF 65-2824, Figure S2: H2-TPR proﬁles of the pure TiO2 in rutile and anatase, Figure S3: XPS proﬁles of Ti 2p for RuO2/TiO2-r with 0.3, 0.5, and 1.0 wt% Ru, Figure S4: XPS spectra of Ce 3d for Ru-Ce/Ti oxide catalysts, Figure S5: XPS spectra of O 1s for the supported RuO2 and Ru-Ce/Ti oxide catalysts, Table S1: Chemisorbed oxygen (Oα) in the supported RuO2 and Ru-Ce/Ti oxide catalysts, Figure S6: Raman spectra of the supported RuO2 and Ru-Ce/Ti oxide catalysts. Author Contributions: Conceptualization, J.S., J.Y. (Jianming Yang) and J.L. (Jian Lu); Data curation, J.S., J.Y. (Jun Yuan), Q.Y., S.M., Q.Z. and W.W.; Formal analysis, J.Y. (Jianming Yang) and Q.Y.; Funding acquisition, J.Y. (Jianming Yang) and J.L. (Jian Lu); Investigation, J.S., F.H. and J.L. (Jialin Li); Methodology, J.S., F.H., J.Y. (Jun Yuan) and S.M.; Resources, J.S., Q.Z., J.L. (Jialin Li) and W.W.; Software, F.H.; Supervision, J.Y. (Jianming Yang) and J.L. (Jian Lu); Validation, J.Y. (Jianming Yang) and J.L. References 1. Hammes, M.; Valtchev, M.; Roth, M.B.; Stöwe, K.; Maier, W.F. A search for alternative Deacon catalysts. Appl. Catal. B Environ. 2013, 132–133, 389–400. [CrossRef] 2. Seki, K. Development of RuO2/Rutile-TiO2 Catalyst for Industrial HCl Oxidation Process. Catal. Surv. A 2010, 14, 168–175. [CrossRef] 3. Amrute, A.P.; Mondelli, C.; Schmidt, T.; Hauert, R.; Pérez-Ramírez, J. Industrial RuO2-Based Deacon Catalysts: Carrier Stabilization and Active Phase Content Optimization. ChemCatChem 2013, 5, 748–756. [CrossRef] 4. Debecker, D.P.; Farin, B.; Gaigneaux, E.M.; Sanchez, C.; Sassoye, C. Total oxidation of propane with a nano-RuO2/TiO2 catalyst. Appl. Catal. A Gen. 2014, 481, 11–18. [CrossRef] 5. Lin, Q.; Huang, Y.; Wang, Y.; Li, L.; Liu, X.Y.; Lv, F.; Wang, A.; Li, W.-C.; Zhang, T. RuO2/rutile-TiO2 5. Lin, Q.; Huang, Y.; Wang, Y.; Li, L.; Liu, X.Y.; Lv, F.; Wang, A.; Li, W.-C.; Zhang, T. RuO2/rut superior catalyst for N2O decomposition. J. Mater. Chem. A 2014, 2, 5178–5181. [CrossRef] 5 , Q ; ua g, ; Wa g, ; , ; u, ; , ; Wa g, ; , W C ; a g, uO2/ ut e superior catalyst for N2O decomposition. J. Mater. Chem. A 2014, 2, 5178–5181. [CrossRef] 6. Martínez Tejada, L.M.; Muñoz, A.; Centeno, M.A.; Odriozola, J.A. In-situ Raman spectrosco 6. Martínez Tejada, L.M.; Muñoz, A.; Centeno, M.A.; Odriozola, J.A. In-situ Raman spectros Ru/TiO2 catalyst in the selective methanation of CO. J. Raman Spectrosc. 2016, 47, 189–197. [C Ru/TiO2 catalyst in the selective methanation of CO. J. Raman Spectrosc. 2016, 47, 189–197. [CrossRe ao, Y.; Jiang, H.; Chen, F. RuO2/TiO2/Pt Ternary Photocatalysts with Epitaxial Heterojunction and Their Jiao, Y.; Jiang, H.; Chen, F. RuO2/TiO2/Pt Ternary Photocatalysts with Epitaxial Heterojunction and The Application in CO Oxidation ACS Catal 2014 4 2249–2257 [CrossRef] 7. Jiao, Y.; Jiang, H.; Chen, F. RuO2/TiO2/Pt Ternary Photocatalysts with Ep Application in CO Oxidation. ACS Catal. 2014, 4, 2249–2257. [CrossRef] 8. Pham, T.N.; Shi, D.; Sooknoi, T.; Resasco, D.E. Aqueous-phase ketonization of acetic acid over Ru/TiO2/carbon catalysts. J. Catal. 2012, 295, 169–178. [CrossRef] 9. Aranda-Pérez, N.; Ruiz, M.P.; Echave, J.; Faria, J. Enhanced activity and stability of Ru-TiO2 rutile for liquid phase ketonization. Appl. Catal. A Gen. 2017, 531, 106–118. [CrossRef] 10. Farra, R.; García-Melchor, M.; Eichelbaum, M.; Hashagen, M.; Frandsen, W.; Allan, J.; Girgsdies, F.; Szentmiklósi, L.; López, N.; Teschner, D. Promoted Ceria: A Structural, Catalytic, and Computational Study. ACS Catal. 2013, 3, 2256–2268. [CrossRef] 11. 4. Conclusions (Jian Lu); Writing – original draft, J.S.; Writing – review & editing, J.S., J.Y. (Jianming Yang) and J.L. (Jian Lu). Funding: The research was funded by Key Research & Development Plan Projects of Shanxi Province (Nos. 2017ZDXM-GY-073, 2017ZDXM-GY-042, 2017ZDXM-GY-070). Conﬂicts of Interest: The authors declare no conﬂicts of interest. References 2010, 276, 141–151. [CrossRef] 22. Teschner, D.; Farra, R.; Yao, L.; Schlögl, R.; Soerijanto, H.; Schomäcker, R.; Schmidt, T.; Szentmiklósi, L.; Amrute, A.P.; Mondelli, C.; et al. An integrated approach to Deacon chemistry on RuO2-based catalysts. J. Catal. 2012, 285, 273–284. [CrossRef] 23. Kondratenko, E.V.; Amrute, A.P.; Pohl, M.-M.; Steinfeldt, N.; Mondelli, C.; Pérez-Ramírez, J. Superior activity of rutile-supported ruthenium nanoparticles for HCl oxidation. Catal. Sci. Technol. 2013, 3, 2555. [CrossRef] 24. Perkas, N.; Zhong, Z.Y.; Chen, L.W.; Besson, M.; Gedanken, A. Sonochemically prepared high dispersed Ru/TiO2 mesoporous catalyst for partial oxidation of methane to syngas. Catal. Lett. 2005, 103, 9–14. [CrossRef] 25. Zweidinger, S.; Crihan, D.; Knapp, M.; Hofmann, J.P.; Seitsonen, A.P.; Weststrate, C.J.; Lundgren, E.; Andersen, J.N.; Over, H. Reaction Mechanism of the Oxidation of HCl over RuO2 (110). J. Phys. Chem. C 2008, 112, 9966–9969. [CrossRef] 26. López, N.; Gómez-Segura, J.; Marín, R.P.; Pérez-Ramírez, J. Mechanism of HCl oxidation (Deacon process) over RuO2. J. Catal. 2008, 255, 29–39. [CrossRef] 27. Yang, S.; Zhu, W.; Wang, X. Inﬂuence of the structure of TiO2, CeO2, and CeO2-TiO2 supports on the activity of Ru catalysts in the catalytic wet air oxidation of acetic acid. RARE Met. 2011, 30, 488–495. [CrossRef] 28. Yang, S.; Feng, Y.; Wan, J.; Zhu, W.; Jiang, Z. Effect of CeO2 addition on the structure and activity of RuO2/γ-Al2O3 catalyst. Appl. Surf. Sci. 2005, 246, 222–228. [CrossRef] 29. Zeng, Y.; Zhang, S.; Wang, Y.; Liu, G.; Zhong, Q. The effects of calcination atmosphere on the catalytic performance of Ce-doped TiO2 catalysts for selective catalytic reduction of NO with NH3. RSC Adv. 2017, 7, 23348–23354. [CrossRef] 30. Kim, J.; Kim, J.Y.; Park, B.G.; Oh, S.J. Photoemission and x-ray absorption study of the electronic structure of SrRu1-xTixO3. Phys. Rev. B 2006, 73. [CrossRef] 31. Wang, S.; Liu, B.; Zhu, Y.; Ma, Z.; Liu, B.; Miao, X.; Ma, R.; Wang, C. Enhanced performance of TiO2 -based perovskite solar cells with Ru-doped TiO2 electron transport layer. Sol. Energy 2018, 169, 335–342. [CrossRef] 32. Nguyen-Phan, T.-D.; Luo, S.; Vovchok, D.; Llorca, J.; Sallis, S.; Kattel, S.; Xu, W.; Piper, L.F.J.; Polyansky, D.E.; Senanayake, S.D.; et al. Three-dimensional ruthenium-doped TiO2 sea urchins for enhanced visible-light-responsive H2 production. Phys. Chem. Chem. Phys. 2016, 18, 15972–15979. [CrossRef] [PubMed] 33. Sui, C.; Niu, X.; Wang, Z.; Yuan, F.; Zhu, Y. References Amrute, A.P.; Mondelli, C.; Moser, M.; Novell-Leruth, G.; López, N.; Rosenthal, D.; Farra, R.; Schuster, M.E.; Teschner, D.; Schmidt, T.; et al. Performance, structure, and mechanism of CeO2 in HCl oxidation to Cl2. J. Catal. 2012, 286, 287–297. [CrossRef] 12. Amrute, A.P.; Larrazábal, G.O.; Mondelli, C.; Pérez-Ramírez, J. CuCrO2 Delafossite: A Stable Copper Catalyst for Chlorine Production. Angew. Chem. Int. Ed. 2013, 52, 9772–9775. [CrossRef] [PubMed] 13. Fei, Z.; Liu, H.; Dai, Y.; Ji, W.; Chen, X.; Tang, J.; Cui, M.; Qiao, X. Efﬁcient catalytic oxidation of HCl to recycle Cl2 over the CuO–CeO2 composite oxide supported on Y type zeolite. Chem. Eng. J. 2014, 257, 273–280. [CrossRef] 239 Catalysts 2019, 9, 108 14. Dutta, G.; Waghmare, U.V.; Baidya, T.; Hegde, M.S.; Priolkar, K.R.; Sarode, P.R. Origin of Enhanced Reducibility/Oxygen Storage Capacity of Ce1-xTixO2 Compared to CeO2 or TiO2. Chem. Mater. 2006, 18, 3249–3256. [CrossRef] 15. Sun, P.; Guo, R.; Liu, S.; Wang, S.; Pan, W. Enhancement of the low-temperature activity of Ce / TiO2 catalyst by Sm modiﬁcation for selective catalytic reduction of NOx with NH3. Mol. Catal. 2017, 433, 224–234. [CrossRef] 16. Fei, Z.; Yang, Y.; Wang, M.; Tao, Z.; Liu, Q.; Chen, X.; Cui, M.; Zhang, Z.; Tang, J.; Qiao, X. Precisely fabricating Ce-O-Ti structure to enhance performance of Ce-Ti based catalysts for selective catalytic reduction of NO with NH3. Chem. Eng. J. 2018, 353, 930–939. [CrossRef] 17. Zhang, Z.; Chen, L.; Li, Z.; Li, P.; Yuan, F.; Niu, X.; Zhu, Y. Activity and SO2 resistance of amorphous CeaTiOx catalysts for the selective catalytic reduction of NO with NH3: In situ DRIFT studies. Catal. Sci. Technol. 2016, 6, 7151–7162. [CrossRef] 18. Ding, J.; Zhong, Q.; Zhang, S. A New Insight into Catalytic Ozonation with Nanosized Ce–Ti Oxides for NOx Removal: Conﬁrmation of Ce–O–Ti for Active Sites. Ind. Eng. Chem. Res. 2015, 54, 2012–2022. [CrossRef] 19. Li, P.; Xin, Y.; Li, Q.; Wang, Z.; Zhang, Z.; Zheng, L. Ce–Ti Amorphous Oxides for Selective Catalytic Reduction of NO with NH3: Conﬁrmation of Ce–O–Ti Active Sites. Environ. Sci. Technol. 2012, 46, 9600–9605. [CrossRef] [PubMed] 20. Shan, W.; Geng, Y.; Zhang, Y.; Lian, Z.; He, H. A CeO2/ZrO2-TiO2 Catalyst for the Selective Catalytic Reduction of NOx with NH3. Catalysts 2018, 8, 592. [CrossRef] 21. Hevia, M.A.G.; Amrute, A.P.; Schmidt, T.; Pérez-Ramírez, J. Transient mechanistic study of the gas-phase HCl oxidation to Cl2 on bulk and supported RuO2 catalysts. J. Catal. 35. Narksitipan, S.; Thongtem, S. Preparation and characterization of rutile TiO2 ﬁlms. J. Ceram. Process. Res. 2012, 13, 35–37. 37. Moser, M.; Mondelli, C.; Amrute, A.P.; Tazawa, A.; Teschner, D.; Schuster, M.E.; Klein-Hoffman, A.; López, N.; Schmidt, T.; Pérez-Ramírez, J. HCl Oxidation on IrO2-Based Catalysts: From Fundamentals to Scale-Up. ACS Catal. 2013, 3, 2813–2822. [CrossRef] 36. Over, H.; Schomäcker, R. What Makes a Good Catalyst for the Deacon Process? ACS Catal. 2013, 3, 1034–1046. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Review Titanium Dioxide (TiO2) Mesocrystals: Synthesis, Growth Mechanisms and Photocatalytic Properties Boxue Zhang, Shengxin Cao, Meiqi Du, Xiaozhou Ye , Yun Wang and Jianfeng Ye Department of Chemistry, College of Science, Huazhong Agricultural University, Wuhan 430070, China; bx1058779150@hotmail.com (B.Z.); shengxincao@hotmail.com (S.C.); dumeiqi@webmail.hzau.edu.cn (M.D.); wangyun@mail.hzau.edu.cn (Y.W.) * C d @ il h d (X Y) ji f @ il h d (J Y) T l 86 27 8728 4018 (J Y Department of Chemistry, College of Science, Huazhong Agricultural University, Wuhan 430070, China; bx1058779150@hotmail.com (B.Z.); shengxincao@hotmail.com (S.C.); dumeiqi@webmail.hzau.edu.cn (M.D.); wangyun@mail.hzau.edu.cn (Y.W.) * Correspondence: xzye@mail.hzau.edu.cn (X.Y); jianfengye@mail.hzau.edu.cn (J.Y.); Tel.: +86-27-8728 4018 (J.Y.) Received: 10 December 2018; Accepted: 11 January 2019; Published: 16 January 2019 * Correspondence: xzye@mail.hzau.edu.cn (X.Y); jianfengye@mail.hzau.edu.cn (J.Y.); Tel.: +86-27-8728 4018 ( Received: 10 December 2018; Accepted: 11 January 2019; Published: 16 January 2019 Abstract: Hierarchical TiO2 superstructures with desired architectures and intriguing physico-chemical properties are considered to be one of the most promising candidates for solving the serious issues related to global energy exhaustion as well as environmental deterioration via the well-known photocatalytic process. In particular, TiO2 mesocrystals, which are built from TiO2 nanocrystal building blocks in the same crystallographical orientation, have attracted intensive research interest in the area of photocatalysis owing to their distinctive structural properties such as high crystallinity, high speciﬁc surface area, and single-crystal-like nature. The deeper understanding of TiO2 mesocrystals-based photocatalysis is beneﬁcial for developing new types of photocatalytic materials with multiple functionalities. In this paper, a comprehensive review of the recent advances toward fabricating and modifying TiO2 mesocrystals is provided, with special focus on the underlying mesocrystallization mechanism and controlling rules. The potential applications of as-synthesized TiO2 mesocrystals in photocatalysis are then discussed to shed light on the structure–performance relationships, thus guiding the development of highly eﬃcient TiO2 mesocrystal-based photocatalysts for certain applications. Finally, the prospects of future research on TiO2 mesocrystals in photocatalysis are brieﬂy highlighted. Keywords: TiO2; photocatalysis; mesocrystals; synthesis; modiﬁcation Catalysts 2019, 9, 91; doi:10.3390/catal9010091 www.mdpi.com/journal/catalysts catalysts catalysts References Activity and deactivation of Ru supported on La1.6Sr0.4NiO4 perovskite-like catalysts prepared by different methods for decomposition of N2O. Catal. Sci. Technol. 2016, 6, 8505–8515. [CrossRef] 34. Balachandran, U.; Eror, N.G. Raman spectra of titanium dioxide. J. Solid State Chem. 1982, 42, 276–282. [CrossRef] 240 Catalysts 2019, 9, 108 35. Narksitipan, S.; Thongtem, S. Preparation and characterization of rutile TiO2 ﬁlms. J. Ceram. Process. Res. 2012, 13, 35–37. 36. Over, H.; Schomäcker, R. What Makes a Good Catalyst for the Deacon Process? ACS Catal. 2013, 3, 1034–1046. [CrossRef] 37. Moser, M.; Mondelli, C.; Amrute, A.P.; Tazawa, A.; Teschner, D.; Schuster, M.E.; Klein-Hoffman, A.; López, N.; Schmidt, T.; Pérez-Ramírez, J. HCl Oxidation on IrO2-Based Catalysts: From Fundamentals to Scale-Up. ACS Catal. 2013, 3, 2813–2822. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 241 eywords: TiO2; photocatalysis; mesocrystals; synthesis; modiﬁcation 1. Introduction Semiconductor-based photocatalysis is well known to be one of the most eﬀective approaches to alleviate the serious conundrums of global energy exhaustion, as well as environmental deterioration, by utilizing the inexhaustible solar energy [1–7]. Among various kinds of semiconductors, Titanium dioxide (TiO2) is the most attractive one as a photocatalyst owing to its high photoreactivity, outstanding chemical stability, easy availability, and cheap price [8–15]. Despite tremendous eﬀorts having been made toward the fabrication of TiO2 materials, as well as the investigation of their photocatalytic properties, real applications of TiO2 in photocatalysis are still largely hampered by the wide band gap of TiO2 (e.g., 3.2 eV for anatase and brookite, 3.0 eV for rutile), which can merely absorb ultraviolet radiation (accounting for < 5% of solar light), and the fast recombination of photoinduced charge carriers, which leads to low quantum eﬃciency [16–21]. It is always a hot topic in the research area of materials chemistry and photocatalysis to manipulate the morphology and architecture of TiO2 to achieve extended light response and facilitate photogenerated electron-hole separation, thus realizing remarkably enhanced photocatalytic activity in various applications [22–26]. Recently, it has been well demonstrated that building highly ordered superstructures from nanocrystal building blocks is very important for fabricating new materials and devices, as this kind of nanoparticle assembly can not only display properties and functions associated with individual nanoparticles, but can also exhibit new collective properties and advanced tunable functions [27–32]. In 243 Catalysts 2019, 9, 91 particular, mesocrystals, a new type of ordered superstructure built from crystallographically oriented nanocrystal subunits, have drawn signiﬁcant research interest since the concept of “mesocrystal” was ﬁrst introduced in 2005 [33,34]. These unique ordered superstructures were initially identiﬁed from the studies of the structural characteristics and growth mechanisms of biominerals, and were proposed to be formed through a non-classical, particle-mediated growth process, namely, mesoscale transformation, rather than the conventional classical, atom/ion-mediated crystallization route (Figure 1). Subsequently, the mesocrystal concept evolved from the classical mesocrystals, which were generated via the aforementioned mesoscale transformation process, to all the hierarchical materials built from crystallographically oriented nanocrystal subunits regardless of the mechanism of formation. 1. Introduction Despite the ﬂourishing emergence of reports on the fabrication of mesocrystals, the history of mesocrystal synthesis is closely related to the continuous exploitation of mesocrystals with new compositions and the persistent development of synthetic procedures having advantages in terms of low cost, convenience in handling, and easiness in compositional and structural control [35–41]. Figure 1. Schematic illustration of the single-crystal formation from classical crystallization, oriented attachment and non-classical crystallization. Reprinted with permission from [33]. Copyright John Wiley & Sons Inc., 2005. Figure 1. Schematic illustration of the single-crystal formation from classical crystallization, oriented attachment and non-classical crystallization. Reprinted with permission from [33]. Copyright John Wiley & Sons Inc., 2005. To date, mesocrystals with a broad range of compositions involving metal oxides (e.g., TiO2 [42–68], ZnO [69–85], Fe2O3 [86–95], CuO [96–101], SnOx [102,103], Co3O4 [104–108], Ag2O [109]), metal chalcogenides (e.g., ZnS [110], PbS [111–113],Ag2S [114], PbSe [115]), metals (e.g., Au [116–118], Ag [119], Cu [120], Pt [121,122], Pd [123]) have been produced, as introduced in some previous reviews [124–126]. Among these mesocrystals, TiO2 mesocrystals are widely accepted to be particularly promising in photocatalytic applications [127–152]. It is noted that the high internal porosity and high surface areas of TiO2 mesocrystals can be beneﬁcial for the adsorption of reagents and provide more active sites for the subsequent photocatalytic reactions, while the well-oriented nanocrystal alignment provides eﬀective conduction pathways and signiﬁcantly enhances charge transport and separation with TiO2 particles [135,153]. Although signiﬁcant attention has been directed to fabricating TiO2 mesocrystals with controlled morphologies, the realization of TiO2 mesocrystals is always a challenging task, probably because the titanium precursors used are highly reactive, and it is rather diﬃcult to precisely control the growth dynamic of TiO2 crystals. Additionally, considering the wide band gap of the pristine TiO2 materials, it is also demanding to modify the mesostructure of TiO2 mesocrystals to realize broadened light absorption, thus achieving highly eﬃcient photocatalysis in various applications. 244 Catalysts 2019, 9, 91 Catalysts 2019, 9, 91 In this review article, we ﬁrst summarize numerous attempts toward the fabrication of TiO2 mesocrystals. Four representative synthetic routes, namely, oriented topotactic transformation, growth on substrates, organic-additive-assisted growth in solution, and direct additive-free synthesis in solution, are presented one by one, with a special focus being channeled towards the underlying mesocrystallization mechanism and its controlling rules. 1. Introduction The construction of doped TiO2 mesocrystals, as well as TiO2 mesocrystal-based heterostructures, is also covered in this review. The potential applications of the resultant TiO2 mesocrystal-based materials in photocatalysis are then introduced to gain a deep understanding of the structure–performance relationships, thus providing useful guidelines for rationally designing and fabricating highly eﬃcient TiO2 mesocrystal-based photocatalysts for certain applications. Finally, some future research directions in the research area are brieﬂy discussed and summarized. 2.1. Oriented Topotactic Transformation 2.1. Oriented Topotactic Transformation Early reports on the fabrication of TiO2 mesocrystals were based on topotactic transformation from pre-synthesized NH4TiOF3 mesocrystals, as the titanium precursors used (e.g., TiCl4, titanium tetrabutoxide (TBOT), titanium tetraisopropanolate (TTIP)) are normally highly reactive, making it rather challenging to manipulate the growth process of TiO2 crystals upon direct syntheses. In 2007, O’Brien’s group disclosed the ﬁrst preparation of TiO2 mesocrystals. In a synthetic procedure, NH4TiOF3 mesocrystals were ﬁrst prepared in the (NH4)2TiF6 and H3BO3 aqueous solution with the assistance of a nonionic surfactant (e.g., Brij 56, Brij 58, or Brij 700). After being washed with H3BO3 solution or sintered in air at 450 ◦C, the as-formed NH4TiOF3 mesocrystals were successfully transformed into anatase TiO2 mesocrystals, with the original platelet-like shapes well preserved [42,43]. Such a topotactic transformation could proceed mainly because of the crystal structure similarity between NH4TiOF3 and anatase TiO2 crystals (less than 0.02% in an average lattice mismatch), and the as-synthesized NH4TiOF3 mesocrystals could thus serve as a crystallographically matched template for the subsequent formation of TiO2 mesocrystals (Figure 2). Owing to the great eﬀectiveness of the methodology, NH4TiOF3 mesocrystals with a variety of morphologies were obtained by simply adjusting the reaction parameters, giving rise to a series of morphology-preserved anatase TiO2 mesocrystals [44,45,137,141,143]. In addition, single-crystalline NH4TiOF3 crystals could also be utilized as a template for the oriented topotactic formation of anatase TiO2 mesocrystals. For instance, by annealing a thin layer of aqueous solution containing TiF4, NH4F, and NH4NO3 on a Si wafer, nanosheet-shaped anatase TiO2 mesocrystals enclosed by a high percentage of (001) facets were produced (Figure 3) [135]. Despite the one-step characteristic of the synthetic process, single-crystalline NH4TiOF3 nanosheets were actually ﬁrst generated in the precursor solution at low annealing temperatures, which could then be easily transformed into anatase TiO2 upon further increase in annealing temperature. With large quantities of N and F elements removed, the volume of the crystals decreased. Pores would form within the particles, resulting in anatase TiO2 mesocrystals consisting of anatase nanocrystals predominantly enclosed by (001) facets. 245 Catalysts 2019, 9, 91 Figure 2. Schematic illustration of oriented topotactic transformation of NH4TiOF3 mesocrystal to anatase TiO2 mesocrystal. The electron diﬀraction (SAED) patterns of the selected area illustrate single-crystal-like diﬀraction behavior for both samples. Reprinted with permission from [43]. Copyright American Chemical Society, 2008. Figure 2. 2.1. Oriented Topotactic Transformation Schematic illustration of oriented topotactic transformation of NH4TiOF3 mesocrystal to anatase TiO2 mesocrystal. The electron diﬀraction (SAED) patterns of the selected area illustrate single-crystal-like diﬀraction behavior for both samples. Reprinted with permission from [43]. Copyright American Chemical Society, 2008. Figure 2. Schematic illustration of oriented topotactic transformation of NH4TiOF3 mesocrystal to anatase TiO2 mesocrystal. The electron diﬀraction (SAED) patterns of the selected area illustrate single-crystal-like diﬀraction behavior for both samples. Reprinted with permission from [43]. Copyright American Chemical Society, 2008. Figure 3. (a) Schematic presentation of oriented topotactic formation of anatase TiO2 mesocrystals with dominant (001) facets; (b) SEM; (c) TEM; and (d) HRTEM images of anatase mesocrystals. The inset displays the related SAED pattern. Reprinted with permission from [135]. Copyright American Chemical Society, 2012. Figure 3. (a) Schematic presentation of oriented topotactic formation of anatase TiO2 mesocrystals with dominant (001) facets; (b) SEM; (c) TEM; and (d) HRTEM images of anatase mesocrystals. The inset displays the related SAED pattern. Reprinted with permission from [135]. Copyright American Chemical Society, 2012. Most recently, Qi’s group proposed a new topotactic transformation method for fabricating anatase TiO2 mesocrystals [154]. In their synthetic procedure, (010)-faceted orthorhombic titanium-containing precursor nanosheet arrays were ﬁrstly synthesized on conducting FTO glass substrate through solvothermally treating 0.1 M K2TiO(C2O4)2 in mixed solvents of deionized water and diethylene glycol. After a further hydrothermal treatment, the as-formed precursor nanosheet arrays could be readily converted to (001)-faceted anatase TiO2 nanosheet arrays. It was revealed that the lattice match between the orthorhombic precursor crystal and the tetragonal anatase crystal accounted for the topotactic transformation from (010)-faceted precursor nanosheets to (001)-faceted anatase TiO2 nanosheets (Figure 4). 246 Catalysts 2019, 9, 91 Figure 4. Schematic presentation of topotactic transformation from (010)-faceted precursor nanosheet arrays to (001)-faceted anatase TiO2 nanosheet arrays on the basis of crystal lattice matchment between orthorhombic precursor crystal and tetragonal anatase crystal. Reprinted with permission from [154]. Copyright Springer, 2017. Figure 4. Schematic presentation of topotactic transformation from (010)-faceted precursor nanosheet arrays to (001)-faceted anatase TiO2 nanosheet arrays on the basis of crystal lattice matchment between orthorhombic precursor crystal and tetragonal anatase crystal. Reprinted with permission from [154]. Copyright Springer, 2017. 2.2. Growth on Substrates As presented above, topotactic transformation has been well demonstrated to be a very useful method to construct TiO2 mesocrystals. However, precursors suitable for such a topotactic transformation are mainly limited to NH4TiOF3, and it is rather diﬃcult to realize the morphological manipulation of the resultant TiO2 mesocrystals at will. Therefore, it is highly desirable to explore facile solution-phase routes toward the direct fabrication of TiO2 mesocrystals, since these kinds of syntheses are normally advantageous in light of their low cost, easy modulation of morphology, and great potential for environmentally benign production of inorganic materials. In 2008, Zeng’s group ﬁrst utilized multiwalled carbon nanotubes (CNTs) as substrate to grow anatase TiO2 mesocrystals with controllable surface coverage [155]. It was revealed that the as-formed [001]-oriented petal-like anatase mesocrystals were uniformly distributed on CNTs, with TiO2 nanocrystal building blocks having diameters in the range of 2–4 nm and mesopores having a very uniform size distribution centered at 2.5 nm. Additionally, by employing graphene nanosheets as a template to control the growth dynamic of TiO2, uniform mesoporous anatase TiO2 nanospheres were successfully generated and anchored on the graphene nanosheets (Figure 5) [156]. It is noteworthy that in comparison to the conventionally generated porous particles constructed by randomly aggregated anatase nanocrystals, the thus-formed mesoporous nanospheres were single-crystal-like. Detailed investigation on the growth process of the mesoporous anatase nanospheres revealed that such a graphene-nanosheet-assisted mesocrystallization route actually involved the nucleation of anatase TiO2 on graphene nanosheets and subsequent oriented aggregation of tiny nanocrystals onto pre-anchored nuclei to reduce the total surface energy of anatase crystals. As a result, mesoporous mesocrystals of anatase TiO2 would ﬁnally form. Moreover, Qi’s group reported the fabrication of two-dimensional (2D) nanoarray structures constructed from mesocrystalline rutile TiO2 nanorods on Ti substrate via a simple solution-phase synthesis [66]. These nanorod arrays were obtained by hydrothermally treating the aqueous solution of TBOT and HCl. It was revealed that during the growth process of the mesocrystalline rutile TiO2 nanorod arrays, stem nanorods were ﬁrst grown onto Ti substrate due to the high concentration of titanium-containing precursors, and with the consumption of the precursors, the resulting low concentration of reactant was responsible for the growth of the tiny nanotips with continuous crystal lattices, resulting in the ﬁnal mesocrystalline rutile TiO2 nanorods with a hierarchical architecture. 247 Catalysts 2019, 9, 91 Figure 5. 2.2. Growth on Substrates (a) SEM, (b) TEM, and (c) HRTEM images of mesoporous anatase TiO2 nanospheres on graphene nanosheets. The inset is the SAED pattern related to a single nanosphere; (d) Schematic illustration of the growth mechanism of mesoporous anatase nanospheres. Reprinted with permission from [156]. Copyright John Wiley & Sons Inc., 2011. Figure 5. (a) SEM, (b) TEM, and (c) HRTEM images of mesoporous anatase TiO2 nanospheres on graphene nanosheets. The inset is the SAED pattern related to a single nanosphere; (d) Schematic illustration of the growth mechanism of mesoporous anatase nanospheres. Reprinted with permission from [156]. Copyright John Wiley & Sons Inc., 2011. 2.3. Organic-Additive-Assisted Growth in Solution Apart from the aforementioned solid templates or substrates, various organic additives could also be utilized to guide the formation of TiO2 mesocrystals. In 2009, Yu’s group ﬁrst prepared hollow-sphere-shaped rutile TiO2 mesocrystals assembled by nanorod subunits via a facile hydrothermal synthesis by using TiCl4 as the titanium source and N, N’-dicyclohexylcarbodiimide (DCC) and L-serine as biological additives (Figure 6) [46]. It was proposed that such hollow-sphere-shaped mesocrystals were actually formed through a distinctive crystallization and transformation process, which involved the appearance of polycrystalline aggregates at the initial stage of reaction, mesoscale transformation to sector-shaped mesocrystals, further transformation of mesocrystals to nanorod bundles upon end-to-end and side-by-side oriented attachment accompanied by assembly of sectors to solid spheres, and ﬁnal generation of hollow spheres via Ostwald ripening. Later on, with the assistance of organic small molecules of glacial acetic acid (HAc) and benzoic acid, rod-like anatase TiO2 mesocrystals were successfully fabricated via a simple solvothermal route [127]. These mesocrystals were proposed to be formed through the well-known oriented attachment, and the mesocrystallization process was found to be carried out under the synergism of hydrophobic bonds, p-p interactions and “mixed-esters-templates”. Furthermore, Gao’s group synthesized spindle-shaped mesoporous anatase TiO2 mesocrystals by utilizing peroxotitanium as the titanium source and polyacrylamide (PAM) as the polymer additive to adjust the growth process of TiO2 [129]. They proposed that these anatase mesocrystals were formed via TiO2-PAM co-assembly, accompanied by an amorphous-to-crystalline transformation. Figure 6. (a) SEM, (b) TEM, and (c) HRTEM images of hollow spheres of rutile TiO2 mesocrystals. The inset in (a) is a magnified SEM image and the inset in (b) shows the related SAED pattern. (d) Schematic illustration of the formation mechanism of the rutile TiO2 mesocrystals. Reprinted with permission from [46]. Copyright American Chemical Society, 2009. Figure 6. (a) SEM, (b) TEM, and (c) HRTEM images of hollow spheres of rutile TiO2 mesocrystals. The inset in (a) is a magnified SEM image and the inset in (b) shows the related SAED pattern. (d) Schematic illustration of the formation mechanism of the rutile TiO2 mesocrystals. Reprinted with permission from [46]. Copyright American Chemical Society, 2009. 248 Catalysts 2019, 9, 91 Catalysts 2019, 9, 91 In 2011, Tartaj’s group developed a method based on inverse microemulsions to produce sub-100 nm sphere-like mesocrystalline nanostructures, which involved a two-stage temperature program [132]. 2.3. Organic-Additive-Assisted Growth in Solution In the ﬁrst stage, the reaction at a low temperature (60 ◦C) triggered inverse microemulsions, resulting in thermal destabilization via forming nanomicellar structures smaller than 100 nm. The subsequent partial hydrolysis of TiOSO4 produced sub-100-nm sphere-shaped TiO2 frameworks through replicating those nanomicellar structures. In the second stage, increasing the reaction temperature to 80 ◦C or higher generated mesocrystalline TiO2 architectures with interstitial porosity partially ﬁlled with surfactants. After the removal of the interstitial surfactants, mesoporosity was generated and uniform spherical-shaped mesocrystalline architectures of anatase TiO2 with particle sizes ranging from 50 to 70 nm were produced ﬁnally. Later on, this method was extended to fabricate spherical-shaped mesoporous anatase TiO2 mesocrystals with a much smaller size of 25 nm [133]. Recently, Zhao’s group reported a facile evaporation-driven oriented assembly method to fabricate mesoporous anatase TiO2 microspheres (~800 nm in diameter) with radially oriented hexagonal mesochannels and single-crystal-like pore walls (Figure 7) [64]. The synthesis started with the liquid-liquid phase separation, which was induced by the preferential evaporation of the solvent of tetrahydrofuran (THF) at a relatively low temperature (40 ◦C), and spherical-shaped PEO-PPO-PEO/TiO2 oligomer composite micelles with PPO segments as the core and titania-associated PEO segments as the shell formed at the liquid-liquid phase interface. Upon further evaporation of THF at 40 ◦C, the concentration of the spherical micelles increased, leading to the formation of uniform mesoporous TiO2 microspheres assembled by composite micelles (step 1 and 2). As the evaporation temperature increased to 80 ◦C, the continuous evaporation of the residual THF and hydrolyzed solvents from TBOT precursor drove the oriented growth of both mesochannels and nanocrystal building blocks from the initially formed spherical composite micelles along the free radial and restricted tangential direction within the TiO2 microspheres (step 3). Radially oriented mesoporous anatase TiO2 microspheres with single-crystal-like pore walls were produced after removal of the triblock copolymer templates ﬁnally (step 4). It is noteworthy that by simply adjusting the reaction parameters, mesoporous, single-crystal-like, olive-shaped, anatase TiO2 mesocrystals constructed by ultrathin nanosheet subunits could also be synthesized [65]. Figure 7. Schematic presentation of the formation process of mesoporous anatase TiO2 microspheres with radially oriented hexagonal mesochannels and single-crystal-like pore walls through evaporation-driven oriented assembly. Reprinted with permission from [64]. Copyright American Chemical Society, 2015. Figure 7. Schematic presentation of the formation process of mesoporous anatase TiO2 microspheres with radially oriented hexagonal mesochannels and single-crystal-like pore walls through evaporation-driven oriented assembly. 2.4. Direct Additive-Free Growth in Solution 2.4. Direct Additive-Free Growth in Solution Considering that the introduction of solid substrates or organic additives into the reaction system is unfavorable for the large-scale production of mesocrystals, it is, therefore, highly desirable to explore facile additive-free synthetic approaches toward functional mesocrystals with controllable crystallinity, porosity, morphology, and architecture. In 2011, Qi’s group reported the ﬁrst additive-free synthesis of nanoporous anatase TiO2 mesocrystals with a spindle-shaped morphology, single-crystal-like structure, and tunable sizes via solvothermal treatment of the solution of TBOT in HAc, followed by calcination in air to remove the residual organics (Figure 8) [47]. These mesocrystals were illustrated to be elongated along the [001] direction, having lengths mainly in the range of 300–450 nm and diameters of 200–350 nm. It was revealed that under the solvothermal conditions, the reaction between TBOT and HAc ﬁrstly generated unstable titanium acetate complexes through ligand exchange/substitution, accompanied by the release of C4H9OH. The subsequent esteriﬁcation reaction between thus-formed C4H9OH and the solvent HAc produced H2O molecules slowly. Then, Ti-O-Ti bonds were formed via both nonhydrolytic-condensation and hydrolysis-condensation processes, resulting in transient amorphous ﬁber-like precursor. As the reaction continued, crystallized ﬂower-like precursor was generated at the expense of the ﬁber-like precursor. This crystallized ﬂower-like precursor acted as a reservoir to continuously release soluble titanium-containing species to generate tiny anatase nanocrystals. These tiny anatase nanocrystals underwent oriented aggregation along the [001] direction, together with some lateral attachment along some side facets of (101) facets, accompanied by the entrapment of in situ produced butyl acetate. As a result, [001]-elongated, spindle-shaped, anatase mesocrystals were produced when the reaction time was long enough. Further calcination in air would remove the butyl acetate residuals, consequently yielding nanoporous anatase TiO2 mesocrystals. Figure 8. (a) SEM and (b) TEM images of nanoporous anatase TiO2 mesocrystals obtained via solvothermal treatment of the solution of TBOT in HAc, followed by thermal treatment in air. The inset is the related SAED pattern of a single mesocrystal. (c) Proposed formation mechanism of nanoporous anatase TiO2 mesocrystals. Reprinted with permission from [47]. Copyright American Chemical Society, 2011. Figure 8. (a) SEM and (b) TEM images of nanoporous anatase TiO2 mesocrystals obtained via solvothermal treatment of the solution of TBOT in HAc, followed by thermal treatment in air. The inset is the related SAED pattern of a single mesocrystal. (c) Proposed formation mechanism of nanoporous anatase TiO2 mesocrystals. Reprinted with permission from [47]. 2.3. Organic-Additive-Assisted Growth in Solution Reprinted with permission from [64]. Copyright American Chemical Society, 2015. 249 Catalysts 2019, 9, 91 Catalysts 2019, 9, 91 2.4. Direct Additive-Free Growth in Solution Copyright American Chemical Society, 2011. 250 Catalysts 2019, 9, 91 Catalysts 2019, 9, 91 After half a month of Qi’s pioneering work, Lu’s group disclosed the fabrication of anatase TiO2 mesocrystals with a single-crystal-like structure, high speciﬁc surface area, preferential exposure of highly reactive (001) crystal facets, and controllable mesoporous network [130]. As shown in Figure 9, by hydrothermal treating the solution of TiOSO4 in tert-butyl alcohol, anatase TiO2 nanocrystals were ﬁrstly generated, the (001) facets of which were preferably adsorbed by SO42−anions. Subsequent oriented attachment of the anatase nanocrystal building blocks created anatase clusters with the (001) facets well protected (step 1). Upon further attachment of the building blocks, anatase TiO2 mesocrystals preferentially exposed by (001) facets and having a disordered mesoporous network were ﬁnally produced (step 2). It is noteworthy that when the growth was conﬁned in a scaﬀold with ordered pore channels, such as mesoporous silica containing 2D (SBA-15, P6mm space group) and three-dimensional (3D) (KIT-6, Ia3d space group) ordered mesopores, the subsequent scaﬀold removal would lead to TiO2 crystals with replicated 2D hexagonal (step 3) or 3D (step 4) ordered network structure, respectively. More interestingly, such a novel methodology could be extended to fabricating mesoporous single-crystal-like structures with other compositions (e.g., ZrO2, CeO2, etc.), thus providing promising materials for various applications. Figure 9. (a) Synthesis of mesoporous single-crystal-like anatase TiO2 mesocrystals. (1) Formation of anatase clusters through oriented attachment of anatase nanocrystal building blocks with (001) facets preferably adsorbed by SO42−ions. (2) Further attachment of the building blocks resulting in mesocrystals with preferential exposed (001) facets and disordered mesoporous structure. Mesocrystals with ordered mesoporous structure were prepared by a conﬁned growth of the anatase crystals in (3) SBA-15 (mesoporous silica with 2D ordered pore channels) and (4) KIT-6 (mesoporous silica with 3D ordered pore channels) followed by scaﬀold removal. TEM images of anatase mesocrystals with disordered mesopores (b), mesoporous mesocrystals grown within SBA-15 (c) and KIT-6 (d) followed by removal of the scaﬀold. The insets in (b–d) show the related SAED and FFT patterns. Reprinted with permission from [130]. Copyright John Wiley & Sons Inc., 2011. Figure 9. (a) Synthesis of mesoporous single-crystal-like anatase TiO2 mesocrystals. (1) Formation of anatase clusters through oriented attachment of anatase nanocrystal building blocks with (001) facets preferably adsorbed by SO42−ions. (2) Further attachment of the building blocks resulting in mesocrystals with preferential exposed (001) facets and disordered mesoporous structure. 2.4. Direct Additive-Free Growth in Solution Mesocrystals with ordered mesoporous structure were prepared by a conﬁned growth of the anatase crystals in (3) SBA-15 (mesoporous silica with 2D ordered pore channels) and (4) KIT-6 (mesoporous silica with 3D ordered pore channels) followed by scaﬀold removal. TEM images of anatase mesocrystals with disordered mesopores (b), mesoporous mesocrystals grown within SBA-15 (c) and KIT-6 (d) followed by removal of the scaﬀold. The insets in (b–d) show the related SAED and FFT patterns. Reprinted with permission from [130]. Copyright John Wiley & Sons Inc., 2011. The above two groups’ fascinating work opened a promising avenue for the facile synthesis of porous anatase mesocrystals. An increasing number of reports of the direct fabrication of TiO2 mesocrystals in solutions without any additives have been disclosed in recent years. For example, Leite’s group proposed a kinetically controlled crystallization process to produce anatase TiO2 mesocrystals with a truncated bipyramidal morphology, which was realized through a nonaqueous sol-gel reaction between TiCl4 and n-octanol [131]. By adopting a similar method to adjust the hydrolysis dynamic of 251 Catalysts 2019, 9, 91 TTIP in an oxalic acid aqueous solution, hierarchical rutile TiO2 mesocrystals were produced [48]. Zhao’s group developed a facile synthetic approach to fabricate regular shaped anatase TiO2 mesocrystals with controllable proportion of (001) and (101) facets [136]. These anatase TiO2 mesocrystals were prepared by solvothermally treating the solution of TTIP in formic acid (FA), and the exposed (101)/(001) ratio could be adjusted via simply varying the duration of solvothermal treatment. Most recently, our group proposed a novel synthetic procedure for producing spindle-shaped, single-crystal-like, anatase TiO2 mesocrystals, which was realized by controlling the hydrolysis rate of TiCl3 in the green solvent PEG-400 (Figure 10) [150]. These mesocrystals constructed by ultraﬁne nanocrystals (~1.5–4.5 nm in size) were revealed to be spindle-shaped and elongated along the [001] direction, having lengths predominantly of 50–85 nm and diameters of 20–40 nm. It was proposed that at the initial stage of the reaction, the chelation of PEG-400 to titanium centers ﬁrstly resulted in the formation of a titanium precursor. This chelated titanium precursor then underwent hydrolysis-condensation reaction in the presence of water to form Ti-O-Ti bonds, accompanied by the gradual oxidation of Ti3+ to Ti4+ by the dissolved oxygen, yielding numerous tiny anatase nanocrystals. 2.4. Direct Additive-Free Growth in Solution They found that SDBS played a vital role in the oriented self-assembly process, and rutile mesocrystals with controllable morphologies were successfully fabricated by varying the adding amount of SDBS. Specifically, uniform octahedral rutile TiO2 mesocrystals 100–300 nm in size were obtained when the titanate/SDBS ratio was set at 0.09, while nanorod-shaped rutile TiO2 mesocrystals were fabricated when the titanate/SDBS ratio increased to 0.15. Interestingly, the morphology and crystalline phase of the TiO2 mesocrystals were demonstrated to be adjustable upon using different counterions to manipulate the growth dynamic of TiO2 [63]. If the conversion of titanate nanowire precursors was carried out in HCl aqueous solution instead of HNO3, dumbbell-shaped rutile TiO2 superstructures composed of loose nanowire subunits were prepared, whereas anatase TiO2 mesocrystals with a quasi-octahedral or truncated-octahedral morphology were obtained from H2SO4 aqueous solution. Such a novel synthetic procedure could also be extendable for the preparation of TiO2 mesocrystals with other crystal phases. For example, by using amorphous titanates as titanium precursor and oxalic acid as structure-directing agent, novel brookite TiO2 mesocrystals were successfully fabricated, as well [157]. Figure 11. (a,b) TEM and (c) HRTEM images of rutile TiO2 mesocrystals formed by conversion of titanate nanowire precursors in HNO3 aqueous solution without any additives. The lower left inset in (b) is an enlarged TEM image, and the upper right inset is the SAED pattern related to the whole particle. (d) Schematic illustration of a tentative mechanism for the formation of rutile TiO2 mesocrystals. Reprinted with permission from [61]. Copyright Royal Society of Chemistry, 2012. Figure 11. (a,b) TEM and (c) HRTEM images of rutile TiO2 mesocrystals formed by conversion of titanate nanowire precursors in HNO3 aqueous solution without any additives. The lower left inset in (b) is an enlarged TEM image, and the upper right inset is the SAED pattern related to the whole particle. (d) Schematic illustration of a tentative mechanism for the formation of rutile TiO2 mesocrystals. Reprinted with permission from [61]. Copyright Royal Society of Chemistry, 2012. 3. Modiﬁcation of TiO2 Mesocrystals 2.4. Direct Additive-Free Growth in Solution These tiny anatase nanocrystals were temporarily stabilized by the solvent PEG-400 molecules and underwent oriented attachment along the [001] direction, together with some lateral attachment along some side facets of (101) facets, resulting in the formation of mesocrystalline anatase aggregates elongated along the [001] direction. It is worth noting that continuous oriented attachment of tiny anatase nanocrystals on the preformed elongated mesocrystalline aggregates occurred when reaction time was prolonged, and well-deﬁned spindle-shaped anatase TiO2 mesocrystals were produced when the reaction time was extended to 5 h. Figure 10. (a) SEM and (b,c) TEM images of anatase TiO2 mesocrystals obtained via hydrolysis reaction of TiCl3 in PEG-400. The insets in (a) are the related particle size distributions of the mesocrystals. (d) SAED pattern recorded on the anatase mesocrystal shown in (c); (e) HRTEM image of anatase mesocrystal; (f) A tentative mechanism for the formation of anatase mesocrystals. Reprinted with permission from [150]. Copyright American Chemical Society, 2017. Figure 10. (a) SEM and (b,c) TEM images of anatase TiO2 mesocrystals obtained via hydrolysis reaction of TiCl3 in PEG-400. The insets in (a) are the related particle size distributions of the mesocrystals. (d) SAED pattern recorded on the anatase mesocrystal shown in (c); (e) HRTEM image of anatase mesocrystal; (f) A tentative mechanism for the formation of anatase mesocrystals. Reprinted with permission from [150]. Copyright American Chemical Society, 2017. In addition to the widely employed titanium sources of TBOT, TTIP, TiOSO4, and TiCl3, it has been well proved that titanate precursors could also be utilized for the fabrication of TiO2 mesocrystals. In 2012, Wei’s group reported the synthesis of unique ultrathin-nanowire-constructed rutile TiO2 mesocrystals through direct transformation from hydrogen titanate nanowire precursors (Figure 11) [61]. These hydrogen titanate nanowire precursors were prepared by hydrothermally treating the anatase TiO2 in KOH solution, followed by acid washing. Then the precipitated hydrogen titanate nanowires were dispersed in HNO3 aqueous solution and kept at 50 ◦C for 7 days, generating single-crystal-like rutile TiO2 mesocrystals having lengths of about 300 nm and diameters 60–80 nm. It was proposed that such rutile mesocrystals were actually formed via face-to-face oriented attachment of ultrathin hydrogen titanate nanowire building blocks, accompanied by the conversion from hydrogen titanate precursor into rutile 252 Catalysts 2019, 9, 91 TiO2. To further modify the morphology of the rutile TiO2 mesocrystals, Wei’s group introduced the surfactant of sodium dodecyl benzene sulfonate (SDBS) into the reaction solution [62]. 3.1. Fabrication of Doped TiO2 Mesocrystals 3.1. Fabrication of Doped TiO2 Mesocrystals As mentioned above, the pristine TiO2 can merely absorb ultra-violet irradiation owing to its wide band gap; continuous efforts have thus been channeled towards developing visible-light-responsive TiO2 photocatalysts for various applications [8–13,16–21]. In addition to the well-known dye sensitization, the modification of TiO2 with impurity doping was demonstrated to exhibit visible-light-responsive photocatalytic reactivity and showed improved stability upon light irradiation [11,16,19]. Considering the novel structural characteristics of TiO2 mesocrystals, the fabrication of metal- or nonmetal-doped TiO2 mesocrystals may give rise to ideal photocatalysts for particle applications, and thus has drawn considerable research interest [158–161]. For example, Majima’s group successfully prepared N-doped anatase TiO2 mesocrystals by solvothermal treatment of the pre-synthesized TiO2 mesocrystals with triethanolamine [158]. Owing to the high internal porosity and high specific surface area of TiO2 mesocrystals, the element of N could diffuse into the pores easily and was adsorbed on the surface. In addition, by stirring TiO2 mesocrystals in NaF aqueous solution at room temperature, F-doped anatase TiO2 mesocrystals could also be fabricated. It was proposed that surface fluorination via ligand exchange between F−and surface OH groups on TiO2 occurred during the stirring process, resulting in the incorporation of F into TiO2 mesocrystals. Combining these two doping strategies together would lead to the formation of N, F-codoped anatase TiO2 mesocrystals without changing the morphology, 253 Catalysts 2019, 9, 91 crystallinestructure, and surface area of TiO2 mesocrystals (Figure 12). Apart from the nonmetal-doped TiO2 mesocrystals, it was demonstrated that metal-doped TiO2 mesocrystals could also be synthesized. Wei’s group prepared pure rutile TiO2 mesocrystals first, and then hydrothermally treated them in aqueous niobium oxalate solution. After a certain period of hydrothermal treatment, homogeneous Nb-doped rutile TiO2 mesocrystals could finally be produced [161]. Figure 12. Proposed synthetic route toward N, F-codoped anatase TiO2 mesocrystals. Reprinted with permission from [158]. Copyright Elsevier, 2016. Figure 12. Proposed synthetic route toward N, F-codoped anatase TiO2 mesocrystals. Reprinted with permission from [158]. Copyright Elsevier, 2016. Recently, the introduction of oxygen vacancies or Ti3+ ions into TiO2 to produce oxygen-deﬁcient/Ti3+ self-doped TiO2 mesostructures has been well accepted to be one of the most eﬃcient ways to extend the light absorption region of TiO2 to visible light [162–166]. Diﬀerent from traditional doping strategies, introducing oxygen vacancies or Ti3+ ions is a unique doping method that can maintain the characteristic nature of TiO2. 3.2. Construction of TiO2 Mesocrystal-Based Heterostructures 3.2. Construction of TiO2 Mesocrystal-Based Heterostructures Apart from the above-mentioned doping strategies, the coupling of TiO2 mesocrystals with appropriate foreign elements to construct TiO2 mesocrystal-based heterostructures is considered to be another eﬀective way to enhance the light absorbance capability as well as inhibit the photoinduced charge carrier recombination [17,18,21]. Hitherto, various kinds of foreign elements have been successfully utilized to modify anatase TiO2 mesocrystals [59,60,169–183]. For example, Sun’s group successfully fabricated spindle-like TiO2/CdS composites by uniformly distributing CdS nanoparticles onto nanoporous anatase mesocrystals via the simple hydrothermal and hot-injection methods [170]. Bian’s group produced CdS quantum dot (QD)-decorated anatase TiO2 mesocrystals preferably enclosed by (001) facets via the facile solvothermal treatment of TiOSO4 in tert-butyl alcohol, followed by modiﬁcation with CdS QDs via a simple ion-exchange treatment [175]. Majima’s group applied a simple photodeposition method to deposit noble metal (Au, Pt) nanoparticles onto the pre-synthesized sheet-like anatase TiO2 mesocrystals and realized the fabrication of novel metal-semiconductor superstructure nanocomposites [169]. Similarly, by adopting by a facile impregnation method, they were also able to deposite Au nanoparticles onto TiO2 mesocrystals and fabricate promising plasmonic photocatalysts [172]. Moreover, to broaden the light-responsive region of TiO2 mesocrystals to near-infrared (NIR) light, they also loaded Au nanorods with controllable size and tunable surface plasmon resonance (SPR) band onto anatase TiO2 mesocrystals through the well-known ligand exchange method [179]. It is noteworthy that in addition to the deposition of guest elements onto the pre-synthesized anatase TiO2 mesocrystals, anatase TiO2 mesocrystals with desired morphologies could also be grown on various kinds of substrates. Tang’s group introduced graphene oxide (GO) nanosheets into the reaction solution of TBOT in HAc. They found that after a solvothermal treatment at elevated temperatures, spindle-shaped anatase TiO2 mesocrystals were successfully grown on the reduced graphene nanosheets [171]. Later on, Lu’s group dispersed a certain amount of graphene into the reaction system of TiOSO4 in tert-butyl alcohol. Upon microwave treatment of the obtained suspension, anatase TiO2 mesocages with a single-crystal-like structure were found to be evenly anchored on graphene nanosheets [59]. Most recently, our group demonstrated that through in situ growth of nanosized defective anatase TiO2-x mesocrystals (DTMCs) on g-C3N4 nanosheets (NSs), a novel 3D/2D DTMC/g-C3N4 NS heterostructure with tight interfaces could be formed (Figure 14) [183]. Figure 14. (a,b) TEM and (c) HRTEM images of 33.3% g-C3N4/DTMCs. The inset is the SAED pattern related to the whole particle. 3.1. Fabrication of Doped TiO2 Mesocrystals At the same time, this kind of doping also improves the electroconductivity of TiO2, thereby facilitating charge transportation within TiO2 particles [162,164,167]. In this regard, great eﬀorts have been made toward preparing oxygen-deﬁcient/Ti3+ self-doped TiO2 mesocrystals [65,136,150,168]. A good example in this area is that Zhao’s group reported a facile evaporation-driven oriented assembly route combined with post thermal treatment in N2 atmosphere to fabricate ultrathin-nanosheet-assembled olive-shaped mesoporous anatase TiO2 mesocrystals (Figure 13) [65]. These mesoporous mesocrystals were illustrated to have high surface area (~189 m2/g), large pore volume (0.56 cm3/g), and abundant oxygen vacancies or unsaturated Ti3+ sites. Additionally, by thermally treating the anatase TiO2 mesocrystals precipitated from the PEG-400/TiCl3 mixed solution in vacuum, our group successfully synthesized Ti3+ self-doped, single-crystal-like, spindle-shaped, anatase TiO2 mesocrystals [150]. Moreover, by reducing the pre-synthesized TiO2 mesocrystals with NaBH4, oxygen-deﬁcient sheet-like anatase TiO2 mesocrystals were also synthesized [168]. Figure 13. (a) Schematic illustration of the growth process of Ti3+ self-doped olive-shaped mesoporous anatase TiO2 mesocrystals through evaporation-driven oriented assembly process; (b) SEM image, (c) TEM image, (d) EPR spectra, and (e) Ti2p XPS core-level spectra of Ti3+ self-doped olive-shaped mesoporous anatase TiO2 mesocrystals. The inset in (c) is the SAED pattern of an individual mesocrystal. Reprinted with permission from [65]. Copyright American Chemical Society, 2015. Figure 13. (a) Schematic illustration of the growth process of Ti3+ self-doped olive-shaped mesoporous anatase TiO2 mesocrystals through evaporation-driven oriented assembly process; (b) SEM image, (c) TEM image, (d) EPR spectra, and (e) Ti2p XPS core-level spectra of Ti3+ self-doped olive-shaped mesoporous anatase TiO2 mesocrystals. The inset in (c) is the SAED pattern of an individual mesocrystal. Reprinted with permission from [65]. Copyright American Chemical Society, 2015. 254 Catalysts 2019, 9, 91 4.1. Bare TiO2 Mesocrystals for Photocatalytic Applications Owing to the novel structural characteristics of mesocrystals, it is speculated the as-synthesized TiO2 mesocrystals can be a promising candidate for photocatalytic applications. Liu’s group ﬁrst reported that the precipitated rod-like anatase TiO2 mesocrystals delivered relatively higher photoreactivity toward the removal of methyl orange (MO) than the corresponding commercial P25 counterpart [127]. They ascribed the remarkably improved photocatalytic activity of the sample to its relatively high surface area, which could provide abundant sites for adsorption capability of MO. Yu’s group proposed that the TiO2 mesocrystals obtained in their additive-free reaction system possessed a well-crystallized rutile phase, low band gap energy and fast electron transfer property, and could exhibit high and stable photocatalytic activity for the removal of NO [128]. Lu’s group evaluated the photoreactivity of the obtained single-crystal-like anatase TiO2 mesocages and found that those unique TiO2 mesocages with 3D ordered mesoporous channels exhibited superior photocatalytic activity toward oxidizing toluene to benzaldehyde and cinnamyl alcohol to cinnamaldehyde relative to that of TiO2 mesocages with 2D ordered mesoporous channels, TiO2 mesocages with disordered mesoporous channels, polycrystalline TiO2, and P25 [130]. Leite’s group claimed that the combination of high surface area and high crystallinity of the recrystallized mesocrystals can be more advantageous in photocatalytic applications than the corresponding disordered aggregate of nanocrystals [131]. Despite of the great eﬀorts mentioned above toward the investigation of the photoreactivity of TiO2 mesocrystals, it wasn’t until 2012 that Majima’s group ﬁrst illustrated the photoelectronic properties of TiO2 superstructures, in order to shed light on the intrinsic relationships between structural ordering and photoreactivity [135]. In their study, plate-like anatase TiO2 mesocrystals synthesized via a topotactic transformation were selected as the target objects. These TiO2 mesocrystals were built from crystallographically ordered anatase TiO2 nanocrystal subunits and had a high surface area and high percentage of exposed highly reactive (001) facets. The photoconductive atomic force microscopy and time-resolved diﬀuse reﬂectance spectroscopy (DRS) were adopted to measure the charge transportation within the anatase mesocrystals, and the obtained results were compared with the reference anatase nanocrystals having similar surface area. It was consequently demonstrated that such a novel structure of anatase mesocrystals could exhibit largely enhanced charge separation and have remarkably long-lived charges, and thus could deliver greatly enhanced photoconductivity and photoreactivity (Figure 15). In 2015, Bian’s group carefully evaluated the inﬂuence of intercrystal misorientation within anatase TiO2 mesocrystals on the photoreactivity of the sample. 3.2. Construction of TiO2 Mesocrystal-Based Heterostructures (d) HAADF-TEM image with elemental mapping of 33.3% g-C3N4/DTMCs. (e) Schematic presentation of the in situ growth of TiO2 mesocrystals on a g-C3N4 nanosheet. Reprinted with permission from [183]. Copyright John Wiley & Sons Inc., 2018. Figure 14. (a,b) TEM and (c) HRTEM images of 33.3% g-C3N4/DTMCs. The inset is the SAED pattern related to the whole particle. (d) HAADF-TEM image with elemental mapping of 33.3% g-C3N4/DTMCs. (e) Schematic presentation of the in situ growth of TiO2 mesocrystals on a g-C3N4 nanosheet. Reprinted with permission from [183]. Copyright John Wiley & Sons Inc., 2018. 255 Catalysts 2019, 9, 91 4. TiO2 Mesocrystals for Photocatalytic Applications 4.1. Bare TiO2 Mesocrystals for Photocatalytic Applications 4.1. Bare TiO2 Mesocrystals for Photocatalytic Applications They concluded that the misorientation of nanocrystal building blocks within anatase mesocrystals was harmful for the eﬀective separation of photogenerated charge carriers and thus largely suppressed the photocatalytic eﬃciencies (Figure 16) [184]. Recently, Hu’s group reported that the photocatalytic properties of anatase TiO2 mesocrystals were actually largely dependent on the interfacial defects of intergrains within the particles [152]. They found that anatase TiO2 mesocrystal photocatalysts exhibited much higher photocatalytic activity toward organic degradation and hydrogen evolution in comparison to single-crystalline crystals and poly crystalline crystals, which can be attributed to the presence of an appropriate number of interfacial defects at the intergrains and the facilitated charge carrier transport across the highly oriented interfaces. Moreover, it is inferred that the photoreactivity of the resultant anatase TiO2 mesocrystal could be further optimized by regulation of defects, which could be simply achieved through annealing in redox atmospheres. 256 talysts 2019, 9, 91 Figure 15. Photodegradation of (a) 4-CP and (b) Cr(VI) using various kinds of TiO2 as catalysts. (c) Time-resolved diﬀuse reﬂectance spectra observed at 200 ns after the laser ﬂash (355-nm) during the photolysis of Meso-TiO2-500 in the absence and presence of 10 mM 4-(methylthio) phenyl methanol (MTPM) as the probe molecule to estimate the lifetime of the charge-separated state in acetonitrile. (d) Diﬀerential time traces of %Abs at 550 nm obtained in the presence of 10 mM MTPM for diﬀerent TiO2 samples in acetonitrile. Reprinted with permission from [135]. Copyright American Chemical Society, 2012. Catalysts 2019, 9, 91 Figure 15. Photodegradation of (a) 4-CP and (b) Cr(VI) using various kinds of TiO2 as catalysts. (c) Time-resolved diﬀuse reﬂectance spectra observed at 200 ns after the laser ﬂash (355-nm) during the photolysis of Meso-TiO2-500 in the absence and presence of 10 mM 4-(methylthio) phenyl methanol (MTPM) as the probe molecule to estimate the lifetime of the charge-separated state in acetonitrile. (d) Diﬀerential time traces of %Abs at 550 nm obtained in the presence of 10 mM MTPM for diﬀerent TiO2 samples in acetonitrile. Reprinted with permission from [135]. Copyright American Chemical Society, 2012. Figure 16. Rates comparison of phenol photodegradation and H2 production upon TiO2 mesocrystals built from well-ordered (red column) and less-ordered (blue column) orientation of nanocrystal subunits. Reprinted with permission from [184]. Copyright American Chemical Society, 2015. Figure 16. 4.1. Bare TiO2 Mesocrystals for Photocatalytic Applications Rates comparison of phenol photodegradation and H2 production upon TiO2 mesocrystals built from well-ordered (red column) and less-ordered (blue column) orientation of nanocrystal subunits. Reprinted with permission from [184]. Copyright American Chemical Society, 2015. 4.2. Doped TiO2 Mesocrystals for Photocatalytic Applications Although a number of reports have demonstrated that TiO2 mesocrystals can exhibit obviously enhanced photocatalytic performance in various applications, their real application is still hampered by the limited light absorbance of the pristine TiO2 with a wide band gap. By utilizing the commonly used doping strategy, the thus-prepared doped TiO2 mesocrystals can therefore become visible-light responsive, thus displaying enhanced visible-light-driven photoreactivity [136,150,158,159,168]. In 2016, Majima’s group investigated the photoreactivity of N, F-codoped anatase TiO2 mesocrystals. They found that, owing to the synergetic eﬀect of N and F doping, the as-prepared product exhibited high visible-light-driven photoreactivity for degradating RhB and 4-nitrophenol (4-NP) [158]. Our group demonstrated that the obtained Ti3+ self-doped anatase TiO2 mesocrystals showed much higher visible-light-driven photoreactivity toward removing NO and Cr (VI) compared with that of Ti3+ self-doped anatase nanocrystal counterparts. Such a photoreactivity enhancement was mainly due to the intrinsic self-doping nature, high crystallinity, as well as high porosity of the anatase 257 Catalysts 2019, 9, 91 mesocrystals (Figure 17) [150]. Most recently, Majima’s group applied femtosecond time-resolved DRS and single-particle photoluminescence (PL) measurements to characterize reduced TiO2 mesocrystals to get deep understanding of the correlation between oxygen deﬁciency, photogenerated charge transfer, and photoreactivity of the material [168]. They conﬁrmed the enhanced light absorption through forming oxygen vacancies did not always result in higher photoreactivity, and an appropriate amount of oxygen vacancies was required to improve the photogenerated charge carrier separation, thus giving rise to optimized photoreactivity. Figure 17. (a) UV-Vis DRS, (b) PL emission spectra, and (c) photocurrent intensity of (i) anatase mesocrystals and (ii) anatase nanocrystals of TiO2 self-doped with Ti3+. (d) Visible-light-driven photodegradation of NO upon (i) anatase mesocrystals and (ii) anatase nanocrystals self-doped with Ti3+. Reprinted with permission from [150]. Copyright John Wiley & Sons Inc., 2017. Figure 17. (a) UV-Vis DRS, (b) PL emission spectra, and (c) photocurrent intensity of (i) anatase mesocrystals and (ii) anatase nanocrystals of TiO2 self-doped with Ti3+. (d) Visible-light-driven photodegradation of NO upon (i) anatase mesocrystals and (ii) anatase nanocrystals self-doped with Ti3+. Reprinted with permission from [150]. Copyright John Wiley & Sons Inc., 2017. Figure 17. 4.1. Bare TiO2 Mesocrystals for Photocatalytic Applications (a) UV-Vis DRS, (b) PL emission spectra, and (c) photocurrent intensity of (i) anatase mesocrystals and (ii) anatase nanocrystals of TiO2 self-doped with Ti3+. (d) Visible-light-driven photodegradation of NO upon (i) anatase mesocrystals and (ii) anatase nanocrystals self-doped with Ti3+. Reprinted with permission from [150]. Copyright John Wiley & Sons Inc., 2017. 4.3. Composited TiO2 Mesocrystals for Photocatalytic Applications 4.3. Composited TiO2 Mesocrystals for Photocatalytic Applications In addition to the aforementioned doping strategy, the coupling of TiO2 mesocrystals with appropriate foreign materials to construct TiO2-mesocrystal-based heterostructures is considered to be another useful methodology to broaden the light absorbance region of the material to visible light or even near-infrared (NIR) light, as well as to facilitate the mobility of photogenerated charge carriers within the particle [169–183]. For example, by utilizing CdS nanocrystals to modify spindle-shaped nanaporous anatase TiO2 mesocrystals, Sun’s group combined the advantages of the individual material, including (1) augmented speciﬁc surface area to provide more absorption and reactive sites; (2) TiO2 mesocrystal substrate with high crystallinity and porosity to facilitate charge transport; (3) uniform distribution of CdS nanocrystals on mesocrystal surface and pores to facilitate charge transfer, and isolate photoinduced electrons and holes in two distinct materials; (4) tight contact between anatase mesocrystals and CdS nanocrystals to minimize the photo-corrosion and leaching oﬀ of CdS nanocrystals; and (5) extension of the photo-response of the material [170]. As expected, this unique spindle-shaped TiO2/CdS photocatalyst exhibited relatively high visible-light-driven activity toward photodegradation of RhB. Bian’s group reported that by decorating CdS QDs onto TiO2 mesocrystals with a high percentage of exposed (001) facets, considerably high visible-light-driven photoreactivity could be achieved when selectively oxidizing various kinds of alcohols to their corresponding aldehydes [175]. Such an enhancement of the photoreactivity could be attributed to CdS QDs with improved photosensitization, porous mesostructure with high surface area, and exposed (001) facets with high surface energy and large quantities of oxygen vacancies, which could promote light absorbance in the visible light region, reactant molecule adsorption and activation, as well as photogenerated charge carrier separation. Majima’s group claimed that superior electron transport and enhanced photoreactivity could be realized upon fabricating noble metal (Au, Pt) 258 Catalysts 2019, 9, 91 nanoparticle-loaded nanoplate-shaped anatase TiO2 mesocrystals [169]. 4.1. Bare TiO2 Mesocrystals for Photocatalytic Applications They proposed that most of the photogenerated electrons could migrate from the dominant surface to the edge of the TiO2 mesocrystal with the reduction reactions mainly occurring at its lateral surfaces containing (101) facets, as illustrated by single-molecule ﬂuorescence spectroscopy. The as-fabricated metal-semiconductor nanocomposites were found to display signiﬁcant enhancement of the photocatalytic reaction rate in organic degradation and hydrogen production. More interestingly, by utilizing Au nanorods to modify anatase TiO2 mesocrystal superstructures, highly eﬃcient photocatalytic hydrogen production under visible-NIR-light irradiation could be obtained [179]. This eﬃcient hydrogen production could be attributed to the SPR of Au nanorods which injected electrons into anatase TiO2 mesocrystals and the facilitated charge transport within mesocrystal particles. Apart from the adjustment of deposited guest particles, it was also demonstrated that eﬃcient defect-state-induced hot electron transfer could be found in the as-prepared Au nanoparticles/reduced TiO2 mesocrystal photocatalysts, which lead to the enhanced photoreactivity of the photocatalyst in removing methylene blue (MB) [182]. Most recently, our group evaluated the photoreactivity of the 3D/2D DTMC/g-C3N4 NS heterostructure with chemically bonded tight interfaces and found that the as-fabricated composite photocatalyst displayed much higher visible-light-driven photoreactivity toward removing the pollutants of MO and Cr(VI) than the corresponding DTMCs and g-C3N4 NSs counterparts (Figure 18) [183]. Systematic characterization results indicated that such an enhancement in the photoredox ability of the composite photocatalyst was based on the direct Z-scheme charge separation, as veriﬁed by the ·OH-trapping experiment. Figure 18. (a) Proposed Z-scheme charge-carrier transfer within DTMC/g-C3N4 composite. (b) XPS valence band spectra and (c) schematic electronic band structures of DTMCs and g-C3N4 NSs. (d) ·OH-trapping PL spectra of DTMCs/g-C3N4 and the corresponding ﬂuorescence intensity upon DTMCs/g-C3N4 in comparison to DTMCs. Reprinted with permission from [183]. Copyright John Wiley & Sons Inc., 2018. Figure 18. (a) Proposed Z-scheme charge-carrier transfer within DTMC/g-C3N4 composite. (b) XPS valence band spectra and (c) schematic electronic band structures of DTMCs and g-C3N4 NSs. (d) ·OH-trapping PL spectra of DTMCs/g-C3N4 and the corresponding ﬂuorescence intensity upon DTMCs/g-C3N4 in comparison to DTMCs. Reprinted with permission from [183]. Copyright John Wiley & Sons Inc., 2018. 5. Summary and Outlook In this paper, we have summarized some recent progress in fabricating TiO2 mesocrystals, with special eﬀorts being directed toward illustrating the underlying mesocrystallization process and its controlling rules. Four representative routes toward the fabrication of TiO2 mesocrystals have been illustrated: oriented topotactic transformation, growth on substrates, organic-additive-assisted growth in solution, and direct additive-free synthesis in solution. In line with the ﬂourishing emergence of reports on the fabrication of TiO2 mesocrystals, the trends of TiO2 mesocrystal synthesis are always related to the continuous exploitation of synthetic procedures having advantages like low cost, convenience in handling, and easiness of compositional and structural control. Apart from the fabrication of bare TiO2 mesocrystals, the construction of doped TiO2 mesocrystals, as well as TiO2 mesocrystal-based heterostructures, are both considered to be promising strategies to further enhance 259 Catalysts 2019, 9, 91 the performance of TiO2 mesocrystals in various applications, and thus have also been covered in this review. Taking into account the novel structural characteristics of TiO2 meoscrystals, such as high crystallinity, high porosity, and oriented nanocrystal assembly, the potential applications of the resultant TiO2 mesocrystal-based materials in photocatalysis have been discussed to gain a deep understanding of the structure-performance relationships, which can provide useful guidelines for designing and fabricating highly eﬃcient TiO2 mesocrystal-based photocatalysts for certain applications. Despite great success having been achieved in the fabrication of TiO2 mesocrystals, the related mesocrystallization process of TiO2 mesocrystals is still not fully understood, and deserves further investigation. It remains an ongoing task to ﬁgure out the speciﬁc reason for the well-ordered alignment of TiO2 nanocrystal building blocks in certain circumstances and develop facile, reproducible, and environmentally benign synthetic approaches toward TiO2 mesocrystals with desired morphologies and architectures. In addition, it should be pointed out that compared with the synthesis of TiO2 mesocrystals, the application of thus-produced TiO2 mesocrystals in photocatalysis is much less explored, suggesting the high demand of a deep investigation into TiO2 mesocrystal-based photocatalysts in various applications. For example, although overall enhancement of photoctalytic activity of TiO2 mesocrystals has been demonstrated in recent years, the real mechanism for the photoreactivity enhancement in certain applications has not yet been fully understood. 5. Summary and Outlook It is a necessity to thoroughly examine the relationship between the structure and photocatalytic properties of TiO2 mesocrystals, which can guide the rational design and fabrication of TiO2 mesocrystals with desired morphologies and architectures to fully satisfy the needs of speciﬁc applications in the future. In addition, the exploration of TiO2 mesocrystal-based photocatalysts in some more challenging application areas, such as selective CO2 reduction, ammonia synthesis, and methanol activation, deserves signiﬁcant research attention to fully excavate their potential in photocatalytic applications. Author Contributions: J.Y. and X.Y. chose the topic; J.Y., X.Y., B.Z., S.C., M.D., and Y.W. wrote and revi the article. Funding: Financial support from National Natural Science Foundation of China (21603079, 21503085), Natural Science Foundation of Hubei Province (2015CFB175, 2015CFB233), Da Bei Nong Group Promoted Project for Young Scholar of HZAU (2017DBN010), and Fundamental Research Funds for the Central Universities (2662015QC042) is gratefully acknowledged. Conﬂicts of Interest: The authors declare no conﬂict of interest. Funding: Financial support from National Natural Science Foundation of China (21603079, 21503085), Natural Science Foundation of Hubei Province (2015CFB175, 2015CFB233), Da Bei Nong Group Promoted Project for Young Scholar of HZAU (2017DBN010), and Fundamental Research Funds for the Central Universities (2662015QC042) is gratefully acknowledged. References Lan, Y.; Lu, Y.; Ren, Z. Mini review on photocatalysis of titanium dioxide nanoparticles and their solar applications. Nano Energy 2013, 2, 1031–1045. [CrossRef] pp gy 10. Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D.W. Understanding TiO h l i M h i d i l Ch R 2014 114 9919 9986 [C R f] pp gy 10. Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D.W. Understanding TiO2 photocatalysis: Mechanisms and materials. Chem. Rev. 2014, 114, 9919–9986. [CrossRef] 11. Asahi, R.; Morikawa, T.; Irie, H.; Ohwaki, T. Nitrogen-doped titanium dioxide as visible-light-sensitive photocatalyst: Designs, developments, and prospects. Chem. Rev. 2014, 114, 9824–9852. [CrossRef] [PubMed] 12. Ma, Y.; Wang, X.; Jia, Y.; Chen, X.; Han, H.; Li, C. Titanium dioxide-based nanomaterials for photocatalytic fuel generations. Chem. Rev. 2014, 114, 9987–10043. [CrossRef] [PubMed] Ma, Y.; Wang, X.; Jia, Y.; Chen, X.; Han, H.; Li, C. Titanium dioxide-based nanomaterials for photocatalytic f l ti Ch R 2014 114 9987 10043 [C R f] [P bM d] 12. Ma, Y.; Wang, X.; Jia, Y.; Chen, X.; Han, H.; Li, C. Titanium dioxide-based n fuel generations. Chem. Rev. 2014, 114, 9987–10043. [CrossRef] [PubMed] 13. Kapilashrami, M.; Zhang, Y.; Liu, Y.-S.; Hagfeldt, A.; Guo, J. Probing the optical property and electronic structure of TiO2 nanomaterials for renewable energy applications. Chem. Rev. 2014, 114, 9662–9707. [CrossRef] [PubMed] 14. Wang, X.; Li, Z.; Shi, J.; Yu, Y. One-dimensional titanium dioxide nanomaterials: Nanowires, nanorods, and nanobelts. Chem. Rev. 2014, 114, 9346–9384. [CrossRef] 15. Li, W.; Wu, Z.; Wang, J.; Elzatahry, A.A.; Zhao, D. A perspective on mesoporous TiO2 materials. Chem. Mater. 2014, 26, 287–298. [CrossRef] 16. Pelaez, M.; Nolan, N.T.; Pillai, S.C.; Seery, M.K.; Falaras, P.; Kontos, A.G.; Dunlop, P.S.M.; Hamilton, J.W.J.; Byrne, J.A.; O’Shea, K.; et al. A review on the visible light active titanium dioxide photocatalysts for environmental applications. Appl. Catal. B Environ. 2012, 125, 331–349. [CrossRef] Park, H.; Park, Y.; Kim, W.; Choi, W. Surface modiﬁcation of TiO2 photocatalyst for environmental applications J. Photochem. Photobiol. C Photochem. Rev. 2013, 15, 1–20. [CrossRef] 17. Park, H.; Park, Y.; Kim, W.; Choi, W. Surface modiﬁcation of TiO2 photocataly J. Photochem. Photobiol. C Photochem. Rev. 2013, 15, 1–20. [CrossRef] 18. Zhang, G.; Kim, G.; Choi, W. Visible light driven photocatalysis mediated via ligand-to-metal charge transfer (LMCT): An alternative approach to solar activation of titania. Energy Environ. Sci. References 2014, 7, 954–966. [CrossRef] 19. Etacheri, V.; Valentin, C.D.; Schneider, J.; Bahnemann, D.; Pillai, S.C. Visible-light activation of TiO2 photocatalysts: Advances in theory and experiments. J. Photochem. Photobiol. C Photochem. Rev. 2015, 25, 1–29. [CrossRef] 20. Gao, M.; Zhu, L.; Ong, W.; Wang, J.; Ho, G.W. Structural design of TiO2-based photocatalyst for H2 production and degradation applications. Catal. Sci. Technol. 2015, 5, 4703–4726. [CrossRef] 21. Colmenares, J.C.; Varma, R.S.; Lisowski, P. Sustainable hybrid photocatalysts: Titania immobilized on carbon materials derived from renewable and biodegradable resources. Green Chem. 2016, 18, 5736–5750. [CrossRef] 22. Zhou, W.; Fu, H. Mesoporous TiO2: Preparation, doping, and as a composite for photocatalysis. ChemCatChem 2013, 5, 885–894. [CrossRef] 23. Wang, M.; Ioccozia, J.; Sun, L.; Lin, C.; Li, Z. Inorganic-modiﬁed semiconductor TiO2 nanotube arrays for photocatalysis. Energy Environ. Sci. 2014, 7, 2182–2202. [CrossRef] 24. Ola, O.; Maroto-Valer, M.M. Review of material design and reactor engineering on TiO2 photocatalysis fo CO2 reduction. J. Photochem. Photobiol. C Photochem. Rev. 2015, 24, 16–42. [CrossRef] 25. Ge, M.; Li, Q.; Cao, C.; Huang, J.; Li, S.; Zhang, S.; Chen, Z.; Zhang, K.; Al-Deyab, S.S.; Lai, Y. One-dimensional TiO2 nanotube photocatalysts for solar water splitting. Adv. Sci. 2017, 4, 1600152. [CrossRef] 26. Zhang, X.; Wang, Y.; Liu, B.; Sang, Y.; Liu, H. Heterostructures construction on TiO2 nanobelts: A powe tool for building high-performance photocatalysts. Appl. Catal. B Environ. 2017, 202, 620–641. [CrossRe ol for building high-performance photocatalysts. Appl. Catal. B Environ. 2017, 202, 620–641. [CrossRef] 27. Mann, S. Self-assembly and transformation of hybrid nano-objects and nanostructures under equilibrium and non equilibrium conditions Nat Mater 2009 8 781 792 [CrossRef] [PubMed] 27. Mann, S. Self-assembly and transformation of hybrid nano-objects and nanostructures and non-equilibrium conditions. Nat. Mater. 2009, 8, 781–792. [CrossRef] [PubMed] 28. Nie, Z.; Petukhova, A.; Kumacheva, E. Properties and emerging applications of self-assembled structures made from inorganic nanoparticles. Nat. Nanotechnol. 2010, 5, 15–25. [CrossRef] g p 29. Talapin, D.V.; Lee, J.-S.; Kovalenko, M.V.; Shevchenko, E.V. Prospects of colloidal nanocrystals for electronic and optoelectronic applications. Chem. Rev. 2010, 110, 389–458. [CrossRef] 30. Liu, J.-W.; Liang, H.-W.; Yu, S.-H. Macroscopic-scale assembled nanowire thin ﬁlms and their functionalities. Chem. Rev. 2012, 112, 4770–4799. [CrossRef] Chem. Rev. 2012, 112, 4770–4799. [CrossRef] 31. Klinkova, A.; Choueiri, R.M.; Kumacheva, E. Self-assembled plasmonic nanostructures. Chem. Soc. Rev. 2014, 43, 3976–3991. [CrossRef] 32. References 1. Keane, D.A.; McGuigan, K.G.; Ibáñez, P.F.; Polo-López, M.I.; Byrne, J.A.; Dunlop, P.S.M.; O’Shea, K.; Dionysiou, D.D.; Pillai, S.C. Solar photocatalysis for water disinfection: Materials and reactor design. Catal. Sci. Technol. 2014, 4, 1211–1226. [CrossRef] 1. Keane, D.A.; McGuigan, K.G.; Ibáñez, P.F.; Polo-López, M.I.; Byrne, J.A.; Dunlop, P.S.M.; O’Shea, K.; Dionysiou, D.D.; Pillai, S.C. Solar photocatalysis for water disinfection: Materials and reactor design. Catal. Sci. Technol. 2014, 4, 1211–1226. [CrossRef] 2. Spasiano, D.; Marotta, R.; Malato, S.; Fernandez-Ibanez, P.; Somma, I.D. Solar photocatalysis: Materials, reactors, some commercial and pre-industrialized applications. A comprehensive approach. Appl. Catal. B Environ. 2015, 170–171, 90–123. [CrossRef] 2. Spasiano, D.; Marotta, R.; Malato, S.; Fernandez-Ibanez, P.; Somma, I.D. Solar photocatalysis: Materials, reactors, some commercial and pre-industrialized applications. A comprehensive approach. Appl. Catal. B Environ. 2015, 170–171, 90–123. [CrossRef] 3. Chen, D.; Zhang, X.; Lee, A.F. Synthetic strategies to nanostructured photocatalysts for CO2 reduction to solar fuels and chemicals. J. Mater. Chem. A 2015, 3, 14487–14516. [CrossRef] 3. Chen, D.; Zhang, X.; Lee, A.F. Synthetic strategies to nanostructured photocatalysts for CO2 reduction to solar fuels and chemicals. J. Mater. Chem. A 2015, 3, 14487–14516. [CrossRef] 4. Marszewski, M.; Cao, S.; Yu, J.; Jaroniec, M. Semiconductor-based photocatalytic CO2 conversion. Mater. Horiz. 2015, 2, 261–278. [CrossRef] 4. Marszewski, M.; Cao, S.; Yu, J.; Jaroniec, M. Semiconductor-based photocatalytic CO2 conversion. Mater. Horiz. 2015, 2, 261–278. [CrossRef] 5. Chen, S.; Takata, T.; Domen, K. Particulate photocatalysts for overall water splitting. Nat. Rev. Mater. 2017, 2, 17050. [CrossRef] 5. Chen, S.; Takata, T.; Domen, K. Particulate photocatalysts for overall water splitting. Nat. Rev. Mater. 2017, 2, 17050. [CrossRef] 6. Zhu, S.; Wang, D. Photocatalysis: Basic principles, diverse forms of implementations and emerging scientiﬁc opportunities. Adv. Energy Mater. 2017, 7, 1700841. [CrossRef] 6. Zhu, S.; Wang, D. Photocatalysis: Basic principles, diverse forms of implementations and emerging scientiﬁc opportunities. Adv. Energy Mater. 2017, 7, 1700841. [CrossRef] 7. Christoforidis, K.C.; Fornasiero, P. Photocatalytic hydrogen production: A rift into the future energy supply. ChemCatChem 2017, 9, 1523–1544. [CrossRef] 7. Christoforidis, K.C.; Fornasiero, P. Photocatalytic hydrogen production: A rift into the future energy supply. ChemCatChem 2017, 9, 1523–1544. [CrossRef] 8. Nakata, K.; Fujishima, A. TiO2 photocatalysis: Design and applications. J. Photochem. Photobiol. C Photochem. Rev. 2012, 13, 169–189. [CrossRef] 8. Nakata, K.; Fujishima, A. TiO2 photocatalysis: Design and applications. J. Photochem. Photobiol. C Photochem. Rev. 2012, 13, 169–189. [CrossRef] 260 Catalysts 2019, 9, 91 9. References Cargnello, M.; Johnston-Peck, A.C.; Diroll, B.T.; Wong, E.; Datta, B.; Damodhar, D.; Doan-Nguyen, V.V.T.; Herzing, A.A.; Kagan, C.R.; Murray, C.B. Substitutional doping in nanocrystal superlattices. Nature 2015, 524, 450–455. [CrossRef] [PubMed] 261 Catalysts 2019, 9, 91 33. Cölfen, H.; Antonietti, M. Mesocrystals: Inorganic superstructures made by highly parallel crystallization and controlled alignment. Angew. Chem. Int. Ed. 2005, 44, 5576–5591. [CrossRef] [PubMed] 34. Cölfen, H.; Antonietti, M. Mesocrystals and Nonclassical Crystallization; John Wiley & Sons: Chichester, UK, 2008. 35. Zhou, L.; O’Brien, P. Mesocrystals: A new class of solid materials. Small 2008, 4, 1566–1574. [CrossRef] [PubMed] 36. Song, R.-Q.; Cölfen, H. Mesocrystals-ordered nanoparticle superstructures. Adv. Mater. 2010, 22, 1301–1330. [CrossRef] 37. Fang, J.; Ding, B.; Gleiter, H. Mesocrystals: Syntheses in metals and applications. Chem. Soc. Rev. 2011, 40, 5347–5360. [CrossRef] [PubMed] 38. Zhou, L.; O’Brien, P. Mesocrystals-properties and applications. J. Phys. Chem. Lett. 2012, 3, 620–628. [CrossRef] 39. Uchaker, E.; Cao, G. Mesocrystals as electrode materials for lithium-ion batteries. Nano Today 2014, 9, 499–524. [CrossRef] 40. Tachikawa, T.; Majima, T. Metal oxide mesocrystals with tailored structures and properties for energy conversion and storage applications. NPG Asia Mater. 2014, 6, e100. [CrossRef] 41. Bergström, L.; Sturm (née Rosseeva), E.V.; Salazar-Alvarez, G.; Cölfen, H. Mesocrystals in biominerals colloidal arrays. Acc. Chem. Res. 2015, 48, 1391–1402. [CrossRef] Zhou, L.; Boyle, D.S.; O’Brien, P. Uniform NH4TiOF3 mesocrystals prepared by an ambient temperature self assembly process and their topotaxial conversion to anatase Chem Commun 2007 144 146 [CrossRef] 42. Zhou, L.; Boyle, D.S.; O’Brien, P. Uniform NH4TiOF3 mesocrystals prepared by an ambient temperat self-assembly process and their topotaxial conversion to anatase. Chem. Commun. 2007, 144–146. [CrossR self assembly process and their topotaxial conversion to anatase. Chem. Commun. 2007, 144 146. [CrossRef] 43. Zhou, L.; Smyth-Boyle, D.; O’Brien, P. A facile synthesis of uniform NH4TiOF3 mesocrystals and their conversion to TiO2 mesocrystals. J. Am. Chem. Soc. 2008, 130, 1309–1320. [CrossRef] [PubMed] 43. Zhou, L.; Smyth-Boyle, D.; O’Brien, P. A facile synthesis of uniform NH4TiOF3 mesocrystals and their conversion to TiO2 mesocrystals. J. Am. Chem. Soc. 2008, 130, 1309–1320. [CrossRef] [PubMed] 44. Feng, J.; Yin, M.; Wang, Z.; Yan, S.; Wan, L.; Li, Z.; Zou, Z. Facile synthesis of anatase TiO2 mesocrystal sheets with dominant {001} facets based on topochemical conversion. CrystEngComm 2010, 12, 3425–3429. [CrossRef] 45. Inoguchi, M.; Afzaal, M.; Tanaka, N.; O’Brien, P. References The poly(ethylene glycol) assisted preparation of NH4TiOF3 mesocrystals and their topotactic conversion to TiO2. J. Mater. Chem. 2012, 22, 25123–25129. [CrossRef] 46. Liu, S.-J.; Gong, J.-Y.; Hu, B.; Yu, S.-H. Mesocrystals of rutile TiO2: Mesoscale transformation, crystallization, and growth by a biologic molecules-assisted hydrothermal process. Cryst. Growth Des. 2009, 9, 203–209. [CrossRef] 47. Ye, J.; Liu, W.; Cai, J.; Chen, S.; Zhao, X.; Zhou, H.; Qi, L. Nanoporous anatase TiO2 mesocrystals: Additive-free synthesis, remarkable crystalline-phase stability, and improved lithium insertion behavior. J. Am. Chem. Soc. 2011, 133, 933–940. [CrossRef] [PubMed] 48. Wang, H.; Liu, Y.; Liu, Z.; Xu, H.; Deng, Y.; Shen, H. Hierarchical rutile TiO2 mesocrystals assembled by nanocrystals-oriented attachment mechanism. CrystEngComm 2012, 14, 2278–2282. [CrossRef] 49. Zhen, M.; Guo, X.; Gao, G.; Zhou, Z.; Liu, L. Rutile TiO2 nanobundles on reduced graphene oxides as anode materials for Li ion batteries. Chem. Commun. 2014, 50, 11915–11918. [CrossRef] 50. Wang, H.; Sun, L.; Wang, H.; Xin, L.; Wang, Q.; Liu, Y.; Wang, L. Rutile TiO2 mesocrystallines with aggregated nanorod clusters: Extremely rapid self-reaction of the single source and enhanced dye-sensitized solar cell performance. RSC Adv. 2014, 4, 58615–58623. [CrossRef] 51. Fu, X.; Wang, B.; Chen, C.; Ren, Z.; Fan, C.; Wang, Z. Controllable synthesis of spherical anatase mesocrystals for lithium ion batteries. New J. Chem. 2014, 38, 4754–4759. [CrossRef] 52. Zhou, Y.; Wang, X.; Wang, H.; Song, Y.; Fang, L.; Ye, N.; Wang, L. Enhanced dye-sensitized solar cells performance using anatase TiO2 mesocrystals with the Wulﬀconstruction of nearly 100% exposed {101} facets as eﬀective light scattering layer. Dalton Trans. 2014, 43, 4711–4719. [CrossRef] [PubMed] 53. Hong, Z.; Zhou, K.; Zhang, J.; Huang, Z.; Wei, M. Facile synthesis of rutile TiO2 mesocrystals with enhanced sodium storage properties. J. Mater. Chem. A 2015, 3, 17412–17416. [CrossRef] 54. Amarilla, J.M.; Morales, E.; Sanz, J.; Sobrados, I.; Tartaj, P. Electrochemical response in aprotic ionic liquid electrolytes of TiO2 anatase anodes based on mesoporous mesocrystals with uniform colloidal size. J. Power Sources 2015, 273, 368–374. [CrossRef] 55. Hong, Z.; Zhou, K.; Huang, Z.; Wei, M. Iso-oriented anatase TiO2 mesocages as a high performance anode material for sodium-ion storage. Sci. Rep. 2015, 5, 11960. [CrossRef] 262 Catalysts 2019, 9, 91 56. Wu, D.; Cao, K.; Wang, H.; Wang, F.; Gao, Z.; Xu, F.; Guo, Y.; Jiang, K. Tunable synthesis of single-crystalline-like TiO2 mesocrystals and their application as eﬀective scattering layer in dye-sensitized solar cells. J. References Colloid Interface Sci. 2015, 456, 125–131. [CrossRef] 57. Wu, Q.; Yang, X.; Zhou, W.; Gao, Q.; Lu, F.; Zhuang, J.; Xu, X.; Wu, M.; Fan, H.J. “Isofacet” anatase TiO2 microcages: Topotactic synthesis and ultrastable Li-ion storage. Adv. Mater. Interfaces 2015, 2, 1500210. [CrossRef] [ ] 58. Hong, Z.; Hong, J.; Xie, C.; Huang, Z.; Wei, M. Hierarchical rutile TiO2 with mesocrystalline structure for Li-ion and Na-ion storage. Electrochim. Acta 2016, 202, 203–208. [CrossRef] 59. Le, Z.; Liu, F.; Nie, P.; Li, X.; Liu, X.; Bian, Z.; Chen, G.; Wu, H.B.; Lu, Y. Pseudocapacitive sodium storage in mesoporous single-crystal-like TiO2-graphene nanocomposite enables high-performance sodium-ion capacitors. ACS Nano 2017, 11, 2952–2960. [CrossRef] [PubMed] 60. Peng, Y.; Le, Z.; Wen, M.; Zhang, D.; Chen, Z.; Wu, H.B.; Li, H.; Lu, Y. Mesoporous single-crystal-like TiO2 mesocages threaded with carbon nanotubes for high-performance electrochemical energy storage. Nano Energy 2017, 35, 44–51. [CrossRef] 61. Hong, Z.; Wei, M.; Lan, T.; Jiang, L.; Cao, G. Additive-free synthesis of unique TiO2 mesocrystals with enhanced lithium-ion intercalation properties. Energy Environ. Sci. 2012, 5, 5408–5413. [CrossRef] 62. Hong, Z.; Wei, M.; Lan, T.; Cao, G. Self-assembled nanoporous rutile TiO2 mesocrystals w morphologies for high rate lithium-ion batteries. Nano Energy 2012, 1, 466–471. [CrossRef] 63. Hong, Z.; Xu, Y.; Liu, Y.; Wei, M. Unique ordered TiO2 superstructures with tunable morphology crystalline phase for improved lithium storage properties. Chem. Eur. J. 2012, 18, 10753–10760. [CrossR Liu, Y.; Wei, M. Unique ordered TiO2 superstructure 64. Liu, Y.; Che, R.; Chen, G.; Fan, J.; Sun, Z.; Wu, Z.; Wang, M.; Li, B.; Wei, J.; Wei, Y.; et al. Radially oriented mesoporous TiO2 microspheres with single-crystal–like anatase walls for high-eﬃciency optoelectronic devices. Sci. Adv. 2015, 1, e1500166. [CrossRef] 65. Liu, Y.; Luo, Y.; Elzatahry, A.A.; Luo, W.; Che, R.; Fan, J.; Lan, K.; Al-Enizi, A.M.; Sun, Z.; Li, B.; et al. Mesoporous TiO2 mesocrystals: Remarkable defects-induced crystallite-interface reactivity and their in situ conversion to single crystals. ACS Cent. Sci. 2015, 1, 400–408. [CrossRef] [PubMed] 66. Cai, J.; Ye, J.; Chen, S.; Zhao, X.; Zhang, D.; Chen, S.; Ma, Y.; Jin, S.; Qi, L. Self-cleaning, broadband and quasi-omnidirectional antireﬂective structures based on mesocrystalline rutile TiO2 nanorod arrays. Energy Environ. Sci. 2012, 5, 7575–7581. [CrossRef] 67. Dai, H.; Zhang, S.; Hong, Z.; Li, X.; Xu, G.; Lin, Y.; Chen, G. References Enhanced photoelectrochemical activity of a hierarchical-ordered TiO2 mesocrystal and its sensing application on a carbon nanohorn support scaﬀold. Anal. Chem. 2014, 86, 6418–6424. [CrossRef] 68. Dai, H.; Zhang, S.; Gong, L.; Li, Y.; Xu, G.; Lin, Y.; Hong, Z. The photoelectrochemical exploration of multifunctional TiO2 mesocrystals and its enzyme-assisted biosensing application. Biosens. Bioelectron. 2015, 72, 18–24. [CrossRef] 69. Li, Z.; Gessner, A.; Richters, J.-P.; Kalden, J.; Voss, T.; Kuebel, C.; Taubert, A. Hollow zinc oxide mesocrystals from an ionic liquid precursor (ILP). Adv. Mater. 2008, 20, 1279–1285. [CrossRef] 70. Liu, Z.; Wen, X.D.; Wu, X.L.; Gao, Y.J.; Chen, H.T.; Zhu, J.; Chu, P.K. Intrinsic dipole-ﬁeld-driven mesoscale crystallization of core-shell ZnO mesocrystal microspheres. J. Am. Chem. Soc. 2009, 131, 9405–9412. [CrossRef] 71. Wu, X.L.; Xiong, S.J.; Liu, Z.; Chen, J.; Shen, J.C.; Li, T.H.; Wu, P.H.; Chu, P.K. Green light stimulates terahertz emission from mesocrystal microspheres. Nat. Nanotechnol. 2011, 6, 103–106. [CrossRef] 72. Distaso, M.; Klupp Taylor, R.N.; Taccardi, N.; Wasserscheid, P.; Peukert, W. Inﬂuence of the counterion on the synthesis of ZnO mesocrystals under solvothermal conditions. Chem. Eur. J. 2011, 17, 2923–2930. [CrossRef] [PubMed] 73. Distaso, M.; Segets, D.; Wernet, R.; Taylor, R.K.; Peukert, W. Tuning the size and the optical properties of ZnO mesocrystals synthesized under solvothermal conditions. Nanoscale 2012, 4, 864–873. [CrossRef] 74. Hosono, E.; Tokunaga, T.; Ueno, S.; Oaki, Y.; Imai, H.; Zhou, H.; Fujihara, S. Crystal growth process of single-crystal-like mesoporous ZnO through a competitive reaction in solution. Cryst. Growth Des. 2012, 12, 2923–2931. [CrossRef] 75. Liu, M.-H.; Tseng, Y.-H.; Greer, H.F.; Zhou, W.; Mou, C.-Y. Dipole ﬁeld guided orientated attachment of nanocrystals to twin-brush ZnO mesocrystals. Chem. Eur. J. 2012, 18, 16104–16113. [CrossRef] [PubMed] 263 Catalysts 2019, 9, 91 76. Sun, S.; Zhang, X.; Zhang, J.; Song, X.; Yang, Z. Unusual designated-tailoring on zone-axis preferential growth of surfactant-free ZnO mesocrystals. Cryst. Growth Des. 2012, 12, 2411–2418. [CrossRef] g y y 77. Waltz, F.; Wissmann, G.; Lippke, J.; Schneider, A.M.; Schwarz, H.-C.; Feldhoﬀ, A.; Eiden, S.; Behrens, P. Evolution of the morphologies of zinc oxide mesocrystals under the inﬂuence of natural polysaccharides. 77. Waltz, F.; Wissmann, G.; Lippke, J.; Schneider, A.M.; Schwarz, H.-C.; Feldhoﬀ, A.; Eiden, S.; Behrens, P. Evolution of the morphologies of zinc oxide mesocrystals under the inﬂuence of natural polysaccharides. Cryst. Growth Des. 2012, 12, 3066–3075. [CrossRef] 78. Wang, H.; Xin, L.; Wang, H.; Yu, X.; Liu, Y.; Zhou, X.; Li, B. References Aggregation-induced growth of hexagonal ZnO hierarchical mesocrystals with interior space: Nonaqueous synthesis, growth mechanism, and optical properties. RSC Adv. 2013, 3, 6538–6544. [CrossRef] 79. Wang, S.-S.; Xu, A.-W. Template-free facile solution synthesis and optical properties of ZnO mesocrystals. Cryst. Eng. Commun. 2013, 15, 376–381. [CrossRef] 80. Peng, Y.; Wang, Y.; Chen, Q.-G.; Zhu, Q.; Xu, A.W. Stable yellow ZnO me photocatalytic activity. CrystEngComm 2014, 16, 7906–7913. [CrossRef] 80. Peng, Y.; Wang, Y.; Chen, Q.-G.; Zhu, Q.; Xu, A.W. Stable yellow ZnO mesocrystals with eﬃcient visible-light photocatalytic activity. CrystEngComm 2014, 16, 7906–7913. [CrossRef] 81. Liu, J.; Hu, Z.-Y.; Peng, Y.; Huang, H.-W.; Li, Y.; Wu, M.; Ke, X.-X.; Tendeloo, G.V.; Su, B.-L. 2D ZnO mesoporous single-crystal nanosheets with exposed {0001} polar facets for the depollution of cationic dye molecules by highly selective adsorption and photocatalytic decomposition. Appl. Catal. B Environ. 2016, 181, 138–145. [CrossRef] 82. Liu, M.-H.; Chen, Y.-W.; Liu, X.; Kuo, J.-L.; Chu, M.-W.; Mou, C.-Y. Defect-mediated gold substitution dop in ZnO mesocrystals and catalysis in CO oxidation. ACS Catal. 2016, 6, 115–122. [CrossRef] 83. Wang, H.; Wang, C.; Chen, Q.; Ren, B.; Guan, R.; Cao, X.; Yang, X.; Duan, R. Interface-defect-mediated photocatalysis of mesocrystalline ZnO assembly synthesized in-situ via a template-free hydrothermal approach. Appl. Surf. Sci. 2017, 412, 517–528. [CrossRef] g g g photocatalysis of mesocrystalline ZnO assembly synthesized in-situ via a template-free hydrothermal approach. Appl. Surf. Sci. 2017, 412, 517–528. [CrossRef] 84. Liu, M.-H.; Chen, Y.-W.; Lin, T.-S.; Mou, C.-Y. Defective mesocrystal ZnO-supported gold catalysts: Facilitating CO oxidation via vacancy defects in ZnO. ACS Catal. 2018, 8, 6862–6869. [CrossRef] 85. Liang, S.; Gou, X.; Cui, J.; Luo, Y.; Qu, H.; Zhang, T.; Yang, Z.; Yang, Q.; Sun, S. Novel cone-like ZnO mesocrystals with coexposed (10-11) and (000-1) facets and enhanced photocatalytic activity. Inorg. Chem. Front. 2018, 5, 2257–2267. [CrossRef] 86. Park, G.-S.; Shindo, D.; Waseda, Y.; Sugimoto, T. Internal structure analysis of monodispersed pseudocubic hematite particles by electron microscopy. J. Colloid Interface Sci. 1996, 177, 198–207. [CrossRef] 87. Ahniyaz, A.; Sakamoto, Y.; Bergström, L. Magnetic ﬁeld-induced assembly of oriented maghemite nanocubes. Proc. Natl. Acad. Sci. USA 2007, 104, 17570–17574. [CrossRef] 88. Fang, X.-L.; Chen, C.; Jin, M.-S.; Kuang, Q.; Xie, Z.-X.; Xie, S.-Y.; Huang, R.-B.; Zheng, L.-S. Single-crystal-like hematite colloidal nanocrystal clusters: Synthesis and applications in gas sensors, photocatalysis and water treatment. J. Mater. Chem. 2009, 19, 6154–6160. [CrossRef] 89. An, Z.; Zhang, J.; Pan, S.; Yu, F. References Facile template-free synthesis and characterization of elliptic a-Fe2O3 superstructures. J. Phys. Chem. C 2009, 113, 8092–8096. [CrossRef] 90. Chen, J.S.; Zhu, T.; Li, C.M.; Lou, X.W. Building hematite nanostructures by oriented attachment. Angew. Chem. Int. Ed. 2011, 50, 650–653. [CrossRef] 91. Ma, J.; Teo, J.; Mei, L.; Zhong, Z.; Li, Q.; Wang, T.; Duan, X.; Lian, J.; Zheng, W. Porous platelike hematite mesocrystals: Synthesis, catalytic and gas-sensing applications. J. Mater. Chem. 2012, 22, 11694–11700. [CrossRef] 92. Duan, X.; Mei, L.; Ma, J.; Li, Q.; Wang, T.; Zheng, W. Facet-induced formation of hematite mesocrystals with improved lithium storage properties. Chem. Commun. 2012, 48, 12204–12206. [CrossRef] 93. Fei, X.; Li, W.; Shao, Z.; Seeger, S.; Zhao, D.; Chen, X. Protein biomineralized nanoporous inorganic mesocrystals with tunable hierarchical nanostructures. J. Am. Chem. Soc. 2014, 136, 15781–15786. [CrossRef] 94. Cai, J.; Chen, S.; Ji, M.; Hu, J.; Ma, Y.; Qi, L. Organic additive-free synthesis of mesocrystalline hematite 94. Cai, J.; Chen, S.; Ji, M.; Hu, J.; Ma, Y.; Qi, L. Organic additive-free synthesis of mesocrystalline hematite nanoplates via two-dimensional oriented attachment. CrystEngComm 2014, 16, 1553–1559. [CrossRef] 95. Agthe, M.; Plivelic, T.S.; Labrador, A.; Bergström, L.; Salazar-Alvarez, G. Following in real time the two-step assembly of nanoparticles into mesocrystals in levitating drops. Nano Lett. 2016, 16, 6838–6843. [CrossRef] 95. Agthe, M.; Plivelic, T.S.; Labrador, A.; Bergström, L.; Salazar-Alvarez, G. Following in real time the two-step assembly of nanoparticles into mesocrystals in levitating drops. Nano Lett. 2016, 16, 6838–6843. [CrossRef] 96. Liu, B.; Zeng, H.C. Mesoscale organization of CuO nanoribbons: Formation of “dandelions”. J. Am. Chem. Soc. 2004, 126, 8124–8125. [CrossRef] y p y g p 96. Liu, B.; Zeng, H.C. Mesoscale organization of CuO nanoribbons: Formation of “dandelions”. J. Am. Chem. Soc. 2004, 126, 8124–8125. [CrossRef] 97. Yao, W.-T.; Yu, S.-H.; Zhou, Y.; Jiang, J.; Wu, Q.-S.; Zhang, L.; Jiang, J. Formation of uniform CuO nanorods by spontaneous aggregation: Selective synthesis of CuO, Cu2O, and Cu nanoparticles by a solid-liquid phase arc discharge process. J. Phys. Chem. B 2005, 109, 14011–14016. [CrossRef] 264 Catalysts 2019, 9, 91 98. Xu, M.; Wang, F.; Ding, B.; Song, X.; Fang, J. Electrochemical synthesis of leaf-like CuO mesocrystals and their lithium storage properties. RSC Adv. 2012, 2, 2240–2243. [CrossRef] 99. Jia, B.; Qin, M.; Zhang, Z.; Cao, Z.; Wu, H.; Chen, P.; Zhang, L.; Lu, X.; Qu, X. References The formation of CuO porous mesocrystal ellipsoids via tuning the oriented attachment mechanism. CrystEngComm 2016, 18, 1376–1383. [CrossRef] 100. Zhang, J.; Cui, Y.; Qin, Q.; Zhang, G.; Luo, W.; Zheng, W. Nanoporous CuO mesocrystals: Low-temperature synthesis and improved structure-performance relationship for energy storage system. Chem. Eng. J. 2018, 331, 326–334. [CrossRef] 101. Hu, J.; Zou, C.; Su, Y.; Li, M.; Han, Y.; Kong, E.S.-W.; Yang, Z.; Zhang, Y. Ultrasensitive NO2 gas sensor based on hierarchical Cu2O/CuO mesocrystals nanoﬂower. J. Mater. Chem. A 2018, 6, 17120–17131. [CrossRef] 102. Zhao, J.; Tan, R.; Guo, Y.; Lu, Y.; Xu, W.; Song, W. SnO mesocrystals: Additive-free synthesis, oxidation, and top-down fabrication of quantum dots. CrystEngComm 2012, 14, 4575–4577. [CrossRef] 103. Chen, S.; Wang, M.; Ye, J.; Cai, J.; Ma, Y.; Zhou, H.; Qi, L. Kinetics-controlled growth of aligned mesocrystalline SnO2 nanorod arrays for lithium-ion batteries with superior rate performance. Nano Research 2013, 6, 243–252. [CrossRef] 104. Liu, Y.; Zhu, G.; Ge, B.; Zhou, H.; Yuan, A.; Shen, X. Concave Co3O4 octahedral mesocrystal: Polymer-mediated synthesis and sensing properties. CrystEngComm 2012, 14, 6264–6270. [CrossRef] 105. Wang, F.; Lu, C.; Qin, Y.; Liang, C.; Zhao, M.; Yang, S.; Sun, Z.; Song, X. Solid state coalescence growth and electrochemical performance of plate-like Co3O4 mesocrystals as anode materials for lithium-ion batteries. J. 105. Wang, F.; Lu, C.; Qin, Y.; Liang, C.; Zhao, M.; Yang, S.; Sun, Z.; Song, X. Solid state coalescence growth and electrochemical performance of plate-like Co3O4 mesocrystals as anode materials for lithium-ion batteries. J. Power Sources 2013, 235, 67–73. [CrossRef] 106. Su, D.; Dou, S.; Wang, G. Mesocrystal Co3O4 nanoplatelets as high capacity anode materials for Li-ion batteries. Nano Res. 2014, 7, 794–803. [CrossRef] 107. Hassen, D.; El-Safty, S.A.; Tsuchiya, K.; Chatterjee, A.; Elmarakbi, A.; Shenashen, M.A.; Sakai, M. Longitudinal hierarchy Co3O4 mesocrystals with high-dense exposure facets and anisotropic interfaces for direct-ethanol fuel cells. Sci. Rep. 2016, 6, 24330. [CrossRef] 108. Cao, W.; Wang, W.; Shi, H.; Wang, J.; Cao, M.; Liang, Y.; Zhu, M. Hierarchical three-dimensional ﬂower-like Co3O4 architectures with a mesocrystal structure as high capacity anode materials for long-lived lithium-ion batteries. Nano Res. 2018, 11, 1437–1446. [CrossRef] 109. Fang, J.; Leufke, P.M.; Kruk, R.; Wang, D.; Scherer, T.; Hahn, H. External electric ﬁeld driven 3D ordering architecture of silver (I) oxide meso-superstructures. Nano Today 2010, 5, 175–182. [CrossRef] 110. References Electrocatalytic properties of h ACS Appl. Mater. Interfaces 2012, 4, 6942–6948. [CrossRef] 122. Zhong, P.; Liu, H.; Zhang, J.; Yin, Y.; Gao, C. Controlled Synthesis of octahedral platinum-based mesocrystals by oriented aggregation. Chem. Eur. J. 2017, 23, 6803–6810. [CrossRef] [PubMed] 123. Huang, X.; Tang, S.; Yang, J.; Tan, Y.; Zheng, N. Etching growth under surface conﬁnement: An eﬀective strategy to prepare mesocrystalline Pd nanocorolla. J. Am. Chem. Soc. 2011, 133, 15946–15949. [CrossRef] [PubMed] 124. Cai, J.; Qi, L. TiO2 mesocrystals: Synthesis, formation mechanisms and applications. Sci. China Chem. 2012, 55, 2318–2326. [CrossRef] 125. Hong, Z.; Wei, M. Recent progress in preparation and lithium-ion storage properties of TiO2 mesocrys J. Chin. Chem. Soc. 2015, 62, 209–216. [CrossRef] 126. Zhang, P.; Tachikawa, T.; Fujitsuka, M.; Majima, T. The development of functional mesocrystals for ene harvesting, storage, and conversion. Chem. Eur. J. 2018, 24, 6295–6307. [CrossRef] 127. Li, L.; Liu, C.-Y. Organic small molecule-assisted synthesis of high active TiO2 rod-like mesocrystals. CrystEngComm 2010, 12, 2073–2078. [CrossRef] 128. Zhang, D.; Li, G.; Wang, F.; Yu, J.C. Green synthesis of a self-assembled rutile mesocrystalline photocatalyst. CrystEngComm 2010, 12, 1759–1763. [CrossRef] 129. Liu, X.; Gao, Y.; Cao, C.; Luo, H.; Wang, W. Highly crystalline spindle-shaped mesoporous anatase titania particles: Solution-phase synthesis, characterization, and photocatalytic properties. Langmuir 2010, 26, 7671–7674. [CrossRef] 130. Bian, Z.; Zhu, J.; Wen, J.; Cao, F.; Huo, Y.; Qian, X.; Cao, Y.; Shen, M.; Li, H.; Lu, Y. Single-crystal-like tita mesocages. Angew. Chem. Int. Ed. 2011, 123, 1137–1140. [CrossRef] 131. Da Silva, R.O.; Gonçalves, R.H.; Stroppa, D.G.; Ramirez, A.J.; Leite, E.R. Synthesis of recrystallized anat TiO2 mesocrystals with Wulﬀshape assisted by oriented attachment. Nanoscale 2011, 3, 1910–1916. [CrossR 132. Tartaj, P. Sub-100 nm TiO2 mesocrystalline assemblies with mesopores: Preparation, characterization, enzyme immobilization and photocatalytic properties. Chem. Commun. 2011, 47, 256–258. [CrossRef] 133. Tartaj, P.; Amarilla, J.M. Multifunctional response of anatase nanostructures based on 25 nm mesocrystal-like porous assemblies. Adv. Mater. 2011, 23, 4904–4907. [CrossRef] 134. Jiao, W.; Wang, L.; Liu, G.; Lu, G.Q.; Cheng, H.-M. Hollow anatase TiO2 single crystals and mesocrystals with dominant {101} facets for improved photocatalysis activity and tuned reaction preference. ACS Catal. 2012, 2, 1854–1859. [CrossRef] 135. Bian, Z.; Tachikawa, T.; Majima, T. Superstructure of TiO2 crystalline nanoparticles yields eﬀective conduc pathways for photogenerated charges. J. Phys. Chem. Lett. 2012, 3, 1422–1427. [CrossRef] 136. Chen, Q.; Ma, W.; Chen, C.; Ji, H.; Zhao, J. References Belman, N.; Israelachvili, J.N.; Li, Y.; Saﬁnya, C.R.; Ezersky, V.; Rabkin, A.; Sima, O.; Golan, Y. Hierarchical superstructure of alkylamine-coated ZnS nanoparticle assemblies. Phys. Chem. Chem. Phys. 2011, 13, 4974–4979. [CrossRef] 111. Querejeta-Fernandez, A.; Hernandez-Garrido, J.C.; Yang, H.; Zhou, Y.; Varela, A.; Parras, M.; Calvino-Gamez, J.J.; Gonzalez-Calbet, J.M.; Green, P.F.; Kotov, N.A. Unknown aspects of self-assembly of PbS microscale superstructures. ACS Nano 2012, 6, 3800–3812. [CrossRef] 112. Simon, P.; Rosseeva, E.; Baburin, I.A.; Liebscher, L.; Hickey, S.G.; Cardoso-Gil, R.; Eychmüller, A.; Kniep, R.; Carrillo-Cabrera, W. PbS-organic mesocrystals: The relationship between nanocrystal orientation and superlattice array. Angew. Chem. Int. Ed. 2012, 51, 10776–10781. [CrossRef] 113. Simon, P.; Bahrig, L.; Baburin, I.A.; Formanek, P.; Röder, F.; Sickmann, J.; Hickey, S.G.; Eychmüller, A.; Lichte, H.; Kniep, R.; et al. Interconnection of nanoparticles within 2D superlattices of PbS/oleic acid thin ﬁlms. Adv. Mater. 2014, 26, 3042–3049. [CrossRef] 114. De la Rica, R.; Velders, A.H. Biomimetic crystallization of Ag2S nanoclusters in nanopore assemblies. J. Am. Chem. Soc. 2011, 133, 2875–2877. [CrossRef] [PubMed] 115. Nagaoka, Y.; Chen, O.; Wang, Z.; Cao, Y.C. Structural control of nanocrystal superlattices using organic guest molecules. J. Am. Chem. Soc. 2012, 134, 2868–2871. [CrossRef] [PubMed] 116. Soejima, T.; Kimizuka, N. One-pot room-temperature synthesis of single-crystalline gold nanocorolla in water. J. Am. Chem. Soc. 2009, 131, 14407–14412. [CrossRef] 116. Soejima, T.; Kimizuka, N. One-pot room-temperature synthe water. J. Am. Chem. Soc. 2009, 131, 14407–14412. [CrossRef] 117. Fang, J.; Du, S.; Lebedkin, S.; Li, Z.; Kruk, R.; Kappes, M.; Hahn, H. Gold mesostructures with tailored surface topography and their self-assembly arrays for surface-enhanced raman spectroscopy. Nano Lett. 2010, 10, 5006–5013. [CrossRef] 265 Catalysts 2019, 9, 91 118. You, H.; Ji, Y.; Wang, L.; Yang, S.; Yang, Z.; Fang, J.; Song, X.; Ding, B. Interface synthesis of gold mesocrystals with highly roughened surfaces for surface-enhanced Raman spectroscopy. J. Mater. Chem. 2012, 22, 1998–2006. [CrossRef] 119. Fang, J.; Ding, B.; Song, X. Self-assembly mechanism of platelike silver mesocrystal. Appl. Phys. Lett. 2007, 91, 083108. [CrossRef] 120. Cao, Y.; Fan, J.; Bai, L.; Hu, P.; Yang, G.; Yuan, F.; Chen, Y. Formation of cubic Cu mesocrystals b solvothermal reaction. CrystEngComm 2010, 12, 3894–3899. [CrossRef] 121. Li, T.; You, H.; Xu, M.; Song, X.; Fang, J. Electrocatalytic properties of hollow coral-like platinum mesocrys ACS Appl. Mater. Interfaces 2012, 4, 6942–6948. [CrossRef] 121. Li, T.; You, H.; Xu, M.; Song, X.; Fang, J. References Anatase TiO2 mesocrystals enclosed by (001) and (101) facets: Synergistic eﬀects between Ti3+ and facets for their photocatalytic performance. Chem. Eur. J. 2012, 18, 12584–12589. [CrossRef] 137. Liu, Y.; Zhang, Y.; Li, H.; Wang, J. Manipulating the formation of NH4TiOF3 mesocrystals: Eﬀects of temperature, surfactant, and pH. Cryst. Growth Des. 2012, 12, 2625–2633. [CrossRef] 138. Aoyama, Y.; Oaki, Y.; Ise, R.; Imai, H. Mesocrystal nanosheet o photocatalyst. CrystEngComm 2012, 14, 1405–1411. [CrossRef] 138. Aoyama, Y.; Oaki, Y.; Ise, R.; Imai, H. Mesocrystal nanosheet of rutile TiO2 and its reaction selectivity as a photocatalyst. CrystEngComm 2012, 14, 1405–1411. [CrossRef] 139. Zhou, L.; Chen, J.; Ji, C.; Zhou, L.; O’Brien, P. A facile solid phase reaction to prepare TiO2 mesocrystals with exposed {001} facets and high photocatalytic activity. CrystEngComm 2013, 15, 5012–5015. [CrossRef] 140. Yao, X.; Liu, X.; Liu, T.; Wang, K.; Lu, L. One-step and large-scale synthesis of anatase TiO2 mesocrystals along [001] orientation with enhanced photocatalytic performance. CrystEngComm 2013, 15, 10246–10254. [CrossRef] 266 Catalysts 2019, 9, 91 141. Guo, Y.; Li, H.; Chen, J.; Wu, X.; Zhou, L. TiO2 mesocrystals built of nanocrystals with exposed {001} facets: Facile synthesis and superior photocatalytic ability. J. Mater. Chem. A 2014, 2, 19589–19593. [CrossRef] 142. Chen, J.; Li, G.; Zhang, H.; Liu, P.; Zhao, H.; An, T. Anatase TiO2 mesocrystals with exposed (001) surface for enhanced photocatalytic decomposition capability toward gaseous styrene. Catal. Today 2014, 224, 216–224. [CrossRef] 143. Fang, Z.; Long, L.; Hao, S.; Song, Y.; Qiang, T.; Geng, B. Mesocrystal precursor transformation strategy for synthesizing ordered hierarchical hollow TiO2 nanobricks with enhanced photocatalytic property. CrystEngComm 2014, 16, 2061–2069. [CrossRef] 144. Lai, L.-L.; Huang, L.-L.; Wu, J.-M. K2TiO(C2O4)2-mediated synthesis of rutile TiO2 mesocrystals and their ability to assist photodegradation of sulfosalicylic acid in water. RSC Adv. 2014, 4, 49280–49286. [CrossRef] 144. Lai, L.-L.; Huang, L.-L.; Wu, J.-M. K2TiO(C2O4)2-mediated synthesis of rutile TiO2 mesocrystals and their ability to assist photodegradation of sulfosalicylic acid in water. RSC Adv. 2014, 4, 49280–49286. [CrossRef] 145. Zhang, P.; Tachikawa, T.; Bian, Z.; Majima, T. Selective photoredox activity on speciﬁc facet-dominated TiO2 145. Zhang, P.; Tachikawa, T.; Bian, Z.; Majima, T. Selective photoredox activity on speciﬁc facet-dominated TiO2 mesocrystal superstructures incubated with directed nanocrystals. Appl. Catal. B 2015, 176–177, 678–686. [CrossRef] 146. Hu, D.; Zhang, W.; Tanaka, Y.; Kusunose, N.; Peng, Y.; Feng, Q. Mesocrystalline nanocomposites of TiO2 polymorphs: Topochemical mesocrystal conversion, characterization, and photocatalytic response. Cryst. References Growth Des. 2015, 15, 1214–1225. [CrossRef] 147. Fu, X.X.; Ren, Z.M.; Fan, C.Y.; Sun, C.X.; Shi, L.; Yu, S.Q.; Qian, G.D.; Wang, Z.Y. Designed fabrication of anatase mesocrystals constructed from crystallographically oriented nanocrystals for improved photocatalytic activity. RSC Adv. 2015, 5, 41218–41223. [CrossRef] 148. Lai, L.-L.; Wu, J.-M. Hollow TiO2 microspheres assembled with rutile mesocrystals: Low-temperature one-pot synthesis and the photocatalytic performance. Ceram. Int. 2015, 41, 12317–12322. [CrossRef] 149. Fu, X.; Fan, C.; Yu, S.; Shi, L.; Wang, Z. TiO2 mesocrystals with exposed {001} facets as eﬃcient photocatalysts. J. Alloys Compd. 2016, 680, 80–86. [CrossRef] 150. Tan, B.; Zhang, X.; Li, Y.; Chen, H.; Ye, X.; Wang, Y.; Ye, J. Anatase TiO2 mesocrystals: Green synthesis, in situ conversion to porous single crystals, and self-doping Ti3+ for enhanced visible light driven photocatalytic removal of NO. Chem. Eur. J. 2017, 23, 5478–5487. [CrossRef] 151. Tang, C.; Liu, L.; Li, Y.; Bian, Z. Aerosol spray assisted assembly of TiO2 mesocrystals into hierarchical hollow microspheres with enhanced photocatalytic performance. Appl. Catal. B 2017, 201, 41–47. [CrossRef] 152. Wang, H.; Chen, Q.; Luan, Q.; Duan, R.; Guan, R.; Cao, X.; Hu, X. Photocatalytic properties dependent on the interfacial defects of intergrains within TiO2 mesocrystals. Chem. Eur. J. 2018, 24, 17105–17116. [CrossRef] [PubMed] 153. Hong, Z.; Dai, H.; Huang, Z.; Wei, M. Understanding the growth and photoelectrochemical properties of mesocrystals and single crystals: A case of anatase TiO2. Phys. Chem. Chem. Phys. 2014, 16, 7441–7447. [CrossRef] [PubMed] 154. Zhang, Y.; Cai, J.; Ma, Y.; Qi, L. Mesocrystalline TiO2 nanosheet arrays with exposed {001} facets: Synthesis via topotactic transformation and applications in dye-sensitized solar cells. Nano Res. 2017, 10, 2610–2625. [CrossRef] 155. Liu, B.; Zeng, H.C. Carbon nanotubes supported mesoporous mesocrystals of anatase TiO2. Chem. Mater. 2008, 20, 2711–2718. [CrossRef] 156. Li, N.; Liu, G.; Zhen, C.; Li, F.; Zhang, L.; Cheng, H.-M. Battery performance and photocatalytic activity of mesoporous anatase TiO2 nanospheres/graphene composites by template-free self-assembly. Adv. Funct. Mater. 2011, 21, 1717–1722. [CrossRef] 157. Zhang, W.; Shen, D.; Liu, Z.; Wu, N.-L.; Wei, M. Brookite TiO2 mesocrystals with enhanced lithium-ion intercalation properties. Chem. Commun. 2018, 54, 11491–11494. [CrossRef] 158. Elbanna, O.; Zhang, P.; Fujitsuka, M.; Majima, T. Facile preparation of nitrogen and ﬂuorine codoped TiO2 mesocrystal with visible light photocatalytic activity. Appl. Catal. B 2016, 192, 80–87. [CrossRef] 158. Elbanna, O.; Zhang, P.; Fujitsuka, M.; Majima, T. Facile preparation of nitrogen and ﬂuorine codoped TiO2 159. References Zhang, P.; Fujitsuka, M.; Majima, T. TiO2 mesocrystal with nitrogen and ﬂuorine codoping during topochemical transformation: Eﬃcient visible light induced photocatalyst with the codopants. Appl. Catal. B 2016, 185, 181–188. [CrossRef] . Primc, D.; Niederberger, M. Synthesis and formation mechanism of multicomponent Sb-Nb:TiO2 mesocrystals Chem. Mater. 2017, 29, 10113–10121. [CrossRef] 161. Lan, T.; Zhang, W.; Wu, N.-L.; Wei, M. Nb-doped rutile TiO2 mesocrystals with enhanced lithium storage properties for lithium ion battery. Chem. Eur. J. 2017, 23, 5059–5065. [CrossRef] 267 Catalysts 2019, 9, 91 162. Naldoni, A.; Allieta, M.; Santangelo, S.; Marelli, M.; Fabbri, F.; Cappelli, S.; Bianchi, C.L.; Psaro, R.; Santo, V.D. Eﬀect of nature and location of defects on bandgap narrowing in black TiO2 nanoparticles. J. Am. Chem. Soc. 2012, 134, 7600–7603. [CrossRef] [PubMed] 163. Pan, X.; Yang, M.-Q.; Fu, X.; Zhang, N.; Xu, Y.-J. Defective TiO2 with oxygen vacancies: Synthesis, properties and photocatalytic applications. Nanoscale 2013, 5, 3601–3614. [CrossRef] [PubMed] 164. Chen, X.; Liu, L.; Huang, F. Black titanium dioxide (TiO2) nanomaterials. Chem. Soc. Rev. 2015, 44, 1861–1 [CrossRef] 164. Chen, X.; Liu, L.; Huang, F. Black titanium dioxide (TiO2) nanomaterials. Chem. Soc. Rev. 2015, 44, 1861 1885. [CrossRef] 165. Ullattil, S.G.; Narendranath, S.B.; Pillai, S.C.; Periyat, P. Black TiO2 nanomaterials: A review of recent [CrossRef] 165. Ullattil, S.G.; Narendranath, S.B.; Pillai, S.C.; Periyat, P. Black TiO2 nanomaterials: A review of recent advances. Chem. Eng. J. 2018, 343, 708–736. [CrossRef] 165. Ullattil, S.G.; Narendranath, S.B.; Pillai, S.C.; Periyat, P. Black TiO2 nanomaterials: A review of rec advances. Chem. Eng. J. 2018, 343, 708–736. [CrossRef] 166. Zhou, W.; Fu, H. Defect-mediated electron-hole separation in semiconductor photocatalysis. Inorg. Chem. Front. 2018, 5, 1240–1254. [CrossRef] 167. Chen, X.; Liu, L.; Yu, P.Y.; Mao, S.S. Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals. Science 2011, 331, 746–750. [CrossRef] [PubMed] 168. Elbanna, O.; Fujitsuka, M.; Kim, S.; Majima, T. Charge carrier dynamics in TiO2 mesocrystals with oxygen vacancies for photocatalytic hydrogen generation under solar light irradiation. J. Phys. Chem. C 2018, 122, 15163–15170. [CrossRef] 169. Bian, Z.; Tachikawa, T.; Kim, W.; Choi, W.; Majima, T. Superior electron transport and photocatalytic abilities of metal-nanoparticle-loaded TiO2 superstructures. J. Phys. Chem. C 2012, 116, 25444–25453. [CrossRef] 170. Gao, P.; Liu, J.; Zhang, T.; Sun, D.D.; Ng, W. Hierarchical TiO2/CdS “spindle-like” composite with high photodegradation and antibacterial capability under visible light irradiation. J. Hazard. Mater. 2012, 229–230, 209–216. [CrossRef] 171. 183. Tan, B.; Ye, X.; Li, Y.; Ma, X.; Wang, Y.; Ye, J. Defective anatase TiO2-x mesocrystal growth in situ on g-C3N4 nanosheets: Construction of 3D/2D Z-scheme heterostructures for highly eﬃcient visible-light photocatalysis. Chem. Eur. J. 2018, 24, 13311–13321. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 184. Chen, F.; Cao, F.; Li, H.; Bian, Z. Exploring the important role of nanocrystals orientation in TiO2 superstructure on photocatalytic performances. Langmuir 2015, 31, 3494–3499. [CrossRef] References Yang, X.; Qin, J.; Li, Y.; Zhang, R.; Tang, H. Graphene-spindle shaped TiO2 mesocrystal composites: Facile synthesis and enhanced visible light photocatalytic performance. J. Hazard. Mater. 2013, 261, 342–350. [CrossRef] 172. Bian, Z.; Tachikawa, T.; Zhang, P.; Fujitsuka, M.; Majima, T. Au/TiO2 superstructure-based plasmonic photocatalysts exhibiting eﬃcient charge separation and unprecedented activity. J. Am. Chem. Soc. 2014, 136, 458–465. [CrossRef] [PubMed] 173. Tachikawa, T.; Zhang, P.; Bian, Z.; Majima, T. Eﬃcient charge separation and photooxidation on cob phosphate-loaded TiO2 mesocrystal superstructures. J. Mater. Chem. A 2014, 2, 3381–3388. [CrossRef] 174. Zhang, P.; Tachikawa, T.; Fujitsuka, M.; Majima, T. Eﬃcient charge separation on 3D architectures of TiO2 mesocrystals packed with a chemically exfoliated MoS2 shell in synergetic hydrogen evolution. Chem. Commun. 2015, 51, 7187–7190. [CrossRef] 175. Li, X.; Wang, J.; Men, Y.; Bian, Z. TiO2 mesocrystal with exposed (001) facets and CdS quantum dots as an active visible photocatalyst for selective oxidation reactions. Appl. Catal. B 2016, 187, 115–121. [CrossRef] 176. Han, T.; Wang, H.; Zheng, X. Gold nanoparticle incorporation into nanoporous anatase TiO2 mesocrystal using a simple deposition-precipitation method for photocatalytic applications. RSC Adv. 2016, 6, 7829–7837. [CrossRef] 177. Yan, D.; Liu, Y.; Liu, C.-Y.; Zhang, Z.-Y.; Niea, S.-D. Multi-component in situ and in-step formation of visible-light response C-dots composite TiO2 mesocrystals. RSC Adv. 2016, 6, 14306–14313. [CrossRef] 178. Tang, H.; Chang, S.; Jiang, L.; Tang, G.; Liang, W. Novel spindle-shaped nanoporous TiO2 coupled graphitic g-C3N4 nanosheets with enhanced visible-light photocatalytic activity. Ceram. Int. 2016, 42, 18443–18452. [CrossRef] 179. Elbanna, O.; Kim, S.; Fujitsuka, M.; Majima, T. TiO2 mesocrystals composited with gold nanorods for highly eﬃcient visible-NIR-photocatalytic hydrogen production. Nano Energy 2017, 35, 1–8. [CrossRef] 180. Elbanna, O.; Fujitsuka, M.; Majima, T. g-C3N4/TiO2 mesocrystals composite for H2 evolution under visible-light irradiation and its charge carrier dynamics. ACS Appl. Mater. Interfaces 2017, 9, 34844–34854. [CrossRef] 181. Yu, X.; Fan, X.; An, L.; Liu, G.; Li, Z.; Liu, J.; Hu, P.A. Mesocrystalline Ti3+-TiO2 hybridized g-C3N4 for eﬃcient visible-light photocatalysis. Carbon 2018, 128, 21–30. [CrossRef] 182. Xue, J.; Elbanna, O.; Kim, S.; Fujitsuka, M.; Majima, T. Defect state-induced eﬃcient hot electron transfer in Au nanoparticles/reduced TiO2 mesocrystal photocatalysts. Chem. Commun. 2018, 54, 6052–6055. [CrossRef] 268 Catalysts 2019, 9, 91 Catalysts 2019, 9, 91 Adsorption and Photocatalytic Decomposition of Gaseous 2-Propanol Using TiO2-Coated Porous Glass Fiber Cloth Sayaka Yanagida 1,, Kentaro Hirayama 2, Kenichiro Iwasaki 2,3 and Atsuo Yasumori 2,3 1 Center for Crystal Science and Technology, University of Yamanashi, 7-32 Miyamae, Kofu 400-8511, Japan 2 Department of Material Science and Technology, Tokyo University of Science, 6-3-1 Niijuku, Katsushika-ku, Tokyo 125-8585, Japan; kentaro1218hr@gmail.com (K.H.); iwasaki@rs.tus.ac.jp (K.I.); yasumori@rs.noda.tus.ac.jp (A.Y.) 3 1 Center for Crystal Science and Technology, University of Yamanashi, 7-32 Miyamae, Kofu 400-8511, Japan 2 1 Center for Crystal Science and Technology, University of Yamanashi, 7-32 Miyamae, Kofu 400-8511, Japan y gy y y J p 2 Department of Material Science and Technology, Tokyo University of Science, 6-3-1 Niijuku, Katsushika-ku Tokyo 125-8585, Japan; kentaro1218hr@gmail.com (K.H.); iwasaki@rs.tus.ac.jp (K.I.); yasumori@rs.noda.tus.ac.jp (A.Y.) 3 Photocatalysis International Research Center, Research Institute for Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda-shi, Chiba 278-8510, Japan Correspondence: syanagida@yamanashi ac jp; Tel : +81 55 220 8723 3 Photocatalysis International Research Center, Research Institute for Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda-shi, Chiba 278-8510, Japan * Correspondence: syanagida@yamanashi.ac.jp; Tel.: +81-55-220-8723 3 Photocatalysis International Research Center, Research Institute for Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda-shi, Chiba 278-8510, Japan * Correspondence: syanagida@yamanashi.ac.jp; Tel.: +81-55-220-8723 * Correspondence: syanagida@yamanashi.ac.jp; Tel.: +81-55-220-8723 Received: 5 December 2018; Accepted: 4 January 2019; Published: 14 January 2019 Received: 5 December 2018; Accepted: 4 January 2019; Published: 14 January 2019 Abstract: Combinations of TiO2 photocatalysts and various adsorbents have been extensively investigated for eliminating volatile organic compounds (VOCs) at low concentrations. Herein, TiO2 and porous glass cloth composites were prepared by acid leaching and subsequent TiO2 dip-coating of the electrically applied glass (E-glass) cloth, and its adsorption and photocatalytic ability were investigated. Acid leaching increased the speciﬁc surface area of the E-glass cloth from 1 to 430 m2/g while maintaining sufﬁcient mechanical strength for supporting TiO2. Further, the speciﬁc surface area remained large (290 m2/g) after TiO2 coating. In the photocatalytic decomposition of gaseous 2-propanol, the TiO2-coated porous glass cloth exhibited higher adsorption and photocatalytic decomposition ability than those exhibited by the TiO2-coated, non-porous glass cloth. The porous composite limited desorption of acetone, which is a decomposition intermediate of 2-propanol, until 2-propanol was completely decomposed to CO2. The CO2 generation rate was affected by the temperature condition (15 or 35 ◦C) and the water content (2 or 18 mg/L); the latter also inﬂuenced 2-propanol adsorption in photocatalytic decomposition. Adsorption and Photocatalytic Decomposition of Gaseous 2-Propanol Using TiO2-Coated Porous Glass Fiber Cloth Both the conditions may change the diffusion and adsorption behavior of 2-propanol in the porous composite. As demonstrated by its high adsorption and photocatalytic ability, the composite (TiO2 and porous glass cloth) effectively eliminates VOCs, while decreasing the emission of harmful intermediates. Keywords: photocatalyst; microporous material; composite; adsorption; air puriﬁcation; TiO2; porous glass Catalysts 2019, 9, 82; doi:10.3390/catal9010082 catalysts catalysts Catalysts 2019, 9, 91 183. Tan, B.; Ye, X.; Li, Y.; Ma, X.; Wang, Y.; Ye, J. Defective anatase TiO2-x mesocrystal growth in situ on g-C3N4 nanosheets: Construction of 3D/2D Z-scheme heterostructures for highly eﬃcient visible-light photocatalysis. Chem. Eur. J. 2018, 24, 13311–13321. [CrossRef] 184. Chen, F.; Cao, F.; Li, H.; Bian, Z. Exploring the important role of nanocrystals orientation in TiO2 superstructure on photocatalytic performances. Langmuir 2015, 31, 3494–3499. [CrossRef] 269 www.mdpi.com/journal/catalysts 1. Introduction The typical composition of E-glass ﬁber is 52–62 wt % SiO2, 16–25 wt % CaO, 12–16 wt % Al2O3, 0–10 wt % B2O3, and 0–5 wt % MgO, along with small amounts of alkaline metal oxides (TiO2 and Fe2O3) and ﬂuorides [20]. When the E-glass ﬁber is corroded in an acid solution [21,22], a porous structure is formed. The pores are formed by the dissolution of Al2O3, B2O3, and the ions of alkaline metals and alkali earth metals, leaving the SiO2 structure. The acid-leached E-glass ﬁber and cloth are the porous materials that had been investigated in previously conducted studies [23–25]. The properties of porous glass cloth, such as the high adsorption capacity for VOCs, high optical transparency, low weight, and sufﬁcient mechanical strength, are considered to be advantageous for photocatalyst support. Very low alkali metal content is another advantage because the diffusion of alkaline metal ions from the glass substrate to the TiO2 photocatalyst deteriorates the photocatalytic activity of TiO2 [26,27]. Kitamura et al. prepared air ﬁlters by combining the TiO2 photocatalyst and the porous glass ﬁber cloth that was prepared by acid leaching [25]. They further examined the ﬁlter’s ability to eliminate gaseous formaldehyde; however, they did not measure the concentration of the decomposition products. Therefore, they could not discuss the contributions of adsorption and photocatalytic decomposition in the elimination process. In this study, a composite material was fabricated from the TiO2 photocatalyst and porous glass ﬁber cloth; further, its adsorption and photocatalytic decomposition ability has been examined with respect to gaseous air-diluted 2-propanol. The TiO2–porous glass cloth composite was prepared by acid leaching the glass cloth and by subsequently dip-coating it using TiO2. The pore formation process was investigated by observing the microstructure, analyzing the chemical composition, and measuring the N2 adsorption of the acid-leached glass cloth. Further, the effects of TiO2 loading on the composite properties were also investigated. Subsequently, the adsorption and photocatalytic properties of the TiO2-coated porous glass cloth were investigated using 300 ppm of gaseous 2-propanol. The concentration changes in the 2-propanol, acetone as an intermediate oxidative product of 2-propanol, and CO2 as a ﬁnal product of 2-propanol decomposition were measured during the adsorption and photocatalytic decomposition. 1. Introduction Indoor air pollution by volatile organic compounds (VOCs) is considered to be a serious health problem. Construction materials emit various VOCs, including formaldehyde, acetaldehyde, toluene, xylene, hexane, acetone, and 2-propanol, which can cause the sick building syndrome even at VOC concentrations lower than 1 ppm [1]. Further, the VOC concentrations in the work spaces are strictly limited to prevent health damage from prolonged exposure. However, the imposed concentration limits of the major solvents are in the range of 10–103 ppm [2]. Titanium dioxide (TiO2) photocatalysts have been extensively investigated in VOC decomposition because of their strong oxidation abilities under ultraviolet (UV) irradiation and because of their ability to decompose various organic compounds [3–6]. However, the VOC decomposition rates are observed to be insufﬁcient at considerably low concentrations because the photocatalytic decomposition reaction tends to be diffusion controlled under such conditions [7]. Catalysts 2019, 9, 82; doi:10.3390/catal9010082 www.mdpi.com/journal/catalysts 271 Catalysts 2019, 9, 82 Catalysts 2019, 9, 82 A promising solution to this problem is to combine TiO2 with an appropriate adsorbate material. Further, the proposed mechanism of VOC elimination by composites can be given as follows. The concentration of VOC molecules can be obtained based on the adherence of VOC molecules in bulk air to the adsorbate surface. The VOC molecules in bulk air are concentrated on the adsorbate surface by adsorption. The adsorbed molecules are diffused to the vicinity of the TiO2 surface and are photo-catalytically decomposed by TiO2 [8,9]. The adsorbate is a porous material, including activated carbon [9,10], zeolite [9,11,12], silica gel [13], or mesoporous silica [14,15]. The composites require a free-standing structure for maintaining sufﬁcient mechanical strength in practical applications. Further, because it is generally difﬁcult to prepare the composites from porous adsorbates, they are used in powder form or are supported on another substrate [12]. Porous glass is an adsorbate candidate exhibiting sufﬁcient mechanical strength for TiO2 support. Thus, a composite photocatalyst of TiO2 and porous glass exhibits superior photocatalytic activity to powder TiO2, or non-porous composite photocatalyst in case of gaseous and aqueous systems [16–19]. The electrically applied glass (E-glass) ﬁber is a mass-produced glass ﬁber with low alkaline metal content. E-glass ﬁber and its cloth are primarily used to form the ﬁber-reinforced plastics. 2.1. Microstructure Analysis and Mechanical Strength Test 2.1. Microstructure Analysis and Mechanical Strength Test The micro-structural change of the E-glass cloth that was prepared by acid leaching and TiO2 coating was investigated using ﬁeld emission scanning electron microscopy (FE–SEM). Hereafter, the unleached and leached samples are referred to as the non-porous and porous glass samples, respectively. Panels (a–c) of Figure 1 denote the cross-sectional images of the non-porous and porous glass ﬁbers comprising cloth samples. After leaching for 1 h, the structure of the glass ﬁbers changed from uniform and dense with no pores (Figure 1a), to a dense core surrounded by a porous surface layer with a thickness of 2–3 μm (Figure 1b). After 3 h of leaching, the glass ﬁber was observed to become completely porous. The SEM images indicate that the porous structure was formed from the surface to the center as the leaching time increased, as previously reported by Tanaka et al. [24]. Panels (d–f) of Figure 1 denote the side views of the non-porous and porous glass ﬁbers. A larger number of cracks developed in the porous ﬁbers when compared to that in the non-porous ﬁbers. The cracks that were observed on the 1-h leached ﬁbers were mainly observed to be perpendicular to the ﬁber elongation direction and propagated only in the porous layer. This indicated that the cracks were generated by the tensile stress in the porous layer. During the drying process, the generation of strong capillary forces in the porous structure could shrink the porous surface layer. The different shrinkage ratios between the core and the porous layer could further result in crack generation. In contrast, the cracks on the 3-h leached ﬁbers progressed along the ﬁber elongation direction and near the ﬁber center. Further, the mechanism of crack generation differed between the samples that were prepared after 1 and 3 h of leaching. The residual stress that was derived from the ﬁber spinning process may explain this type of crack generation, as will be discussed in our subsequent report. Despite the generation of clacking by acid leaching, the ﬂexibility of the porous glass cloth was maintained at a high level. When the 3-h leached glass cloth was curved with a curvature radius of 3 mm, the glass cloth did not break; further, its form was recovered without any deformation. Panels (g–i) of Figure 1 denote the side views of the TiO2-coated ﬁbers. 1. Introduction Further, the effects of temperature and water content on the adsorption and photocatalytic decomposition abilities of the composite were also investigated because these factors affect the photocatalytic decomposition rates of 2-propanol [28–31] and acetone [32,33], and because they may affect the adsorption and diffusion of organic molecules in the porous glass. Finally, based on the experimental results, we discuss the adsorption and photocatalytic decomposition mechanisms of the TiO2-coated porous glass cloth. 272 Catalysts 2019, 9, 82 2.1. Microstructure Analysis and Mechanical Strength Test Agglomerates of the TiO2 nanoparticles are observed on the surfaces, and the coating is observed to be in-homogeneous. Because the glass cloth samples have high ﬁber density, homogeneous coating of each ﬁber by dip-coating the cloth sample would be difﬁcult. Further, the sampling process before FE–SEM observation can cause some damage to the TiO2 coating. Figure S1 depicts the surface and fracture cross-section of TiO2 coating on the porous glass cloth at a position where plenty of TiO2 is loaded and where only little sample damage is observed. The TiO2 coating exhibited a porous structure, and VOCs can penetrate the coating. 2.2. Crystalline Phase of TiO2 The TiO2-coated glass cloth samples, and the TiO2 powder sample that was prepared from TiO2 suspension used in the dip-coating process, were ground with a pestle in an agate mortar, and their X-ray diffraction (XRD) patterns were measured. Figure 2 depicts the XRD patterns of the aforementioned samples. The TiO2 suspension comprised anatase (TiO2) nanoparticles and solvent; therefore, the TiO2 powder sample prepared from the TiO2 suspension exhibited anatase peaks (Figure 2a). The crystalline size that was estimated from Scherrer’s equation using 101 reﬂection is 6.9 nm. The TiO2-coated non-porous glass cloth exhibited no anatase peak (Figure 2b). This can be attributed to the lesser amounts of TiO2, and will be explained in the following elemental analysis in Section 2.3. The broad peaks that can be observed at around 26◦denote the glass halo, whereas the peaks that can be observed at around 42◦denote the graphite oxide generated by incomplete organic molecule combustion. The TiO2-coated 3-h leached porous glass cloth did not exhibit the strongest 101 reﬂection of anatase because it was overwrapped with glass halo at around 24◦(Figure 2c). However, the 220, 105, 211, and 204 reﬂections of anatase can be clearly observed. Based on these results, the TiO2 on the porous sample was observed to be in the anatase form. Further, the TiO2 on the non-porous sample was also in the anatase form because it was prepared in the same manner as the porous sample. 273 Catalysts 2019, 9, 82 It indicates that the use of black light lump (center wavelength: 365 nm), in the following photocatalytic decomposition experiments, is suitable for the photo-excitation of this composite because anatase exhibits a bandgap of 3.2 eV (388 nm) [34]. Figure 1. SEM images of the non-porous and porous glass ﬁbers before (left and center panels) and after (right panel) TiO2 coating. Figure 1. SEM images of the non-porous and porous glass ﬁbers before (left and center panels) and after (right panel) TiO2 coating. &X.α θ GHJUHH ,QWHQVLW\DX D E F Figure 2. Powder X-Ray diffraction (XRD) pattern of the ground samples: (a) dried and heated TiO2 suspension; (b) TiO2-coated non-porous glass cloth; and (c) TiO2-coated porous glass cloth. hkl are reﬂection indices of anatase. Ch i l C i i A l i Figure 1. 2.2. Crystalline Phase of TiO2 SEM images of the non-porous and porous glass ﬁbers before (left and center panels) and after (right panel) TiO2 coating. after (right panel) TiO2 coating. &X.α θ GHJUHH ,QWHQVLW\DX D E F Figure 2. Powder X-Ray diffraction (XRD) pattern of the ground samples: (a) dried and heated TiO2 suspension; (b) TiO2-coated non-porous glass cloth; and (c) TiO2-coated porous glass cloth. hkl are reﬂection indices of anatase. &X.α θ GHJUHH ,QWHQVLW\DX D E F &X.α θ GHJUHH Figure 2. Powder X-Ray diffraction (XRD) pattern of the ground samples: (a) dried and heated TiO2 suspension; (b) TiO2-coated non-porous glass cloth; and (c) TiO2-coated porous glass cloth. hkl are reﬂection indices of anatase. Figure 2. Powder X-Ray diffraction (XRD) pattern of the ground samples: (a) dried and heated TiO2 suspension; (b) TiO2-coated non-porous glass cloth; and (c) TiO2-coated porous glass cloth. hkl are reﬂection indices of anatase. Figure 2. Powder X-Ray diffraction (XRD) pattern of the ground samples: (a) dried and heated TiO2 suspension; (b) TiO2-coated non-porous glass cloth; and (c) TiO2-coated porous glass cloth. hkl are reﬂection indices of anatase. Figure 2. Powder X-Ray diffraction (XRD) pattern of the ground samples: (a) dried and heated TiO2 suspension; (b) TiO2-coated non-porous glass cloth; and (c) TiO2-coated porous glass cloth. hkl are reﬂection indices of anatase. 2.3. Chemical Composition Analysis Figure 3 depicts the chemical compositions of the non-porous and porous glass cloths that have been estimated using the X-ray ﬂuorescence (XRF) measurements. In this ﬁgure, the porous cloths were obtained after leaching for 1, 3, and 12 h. While estimating the compositions of the 3- and 12-h leached glass cloths, sodium oxide was assumed to be zero because the sample pellets were prepared using Na2CO3. With an increase in leaching time, the composition ratios of the alkali earth metal oxides, 274 Catalysts 2019, 9, 82 boron oxide, alkali metal oxides, and alumina decreased, whereas the silica content was maintained to be almost constant. The leaching of the non-silica components from the glass ﬁber resulted in the formation of the observed microporous structures in the glass cloths Panels (b,c,e,f) of Figure 1. The composition almost ceased to change after 3 h of leaching (the time of core disappearance in the acid-leached ﬁbers). 0J2 2WKHUV K K K 6L2 &D2 $O2 %2 &KHPLFDOFRPSRVLWLRQZW K /HDFKLQJWLPHK Figure 3. Chemical compositions of the non-porous (0-h leaching time) and porous glass cloths estimated from the X-ray ﬂuorescence (XRF) measurements. The “Others” category includes TiO2, Na2O, K2O, Fe2O3, and SrO. &KHPLFDOFRPSRVLWLRQZW Figure 3. Chemical compositions of the non-porous (0-h leaching time) and porous glass cloths estimated from the X-ray ﬂuorescence (XRF) measurements. The “Others” category includes TiO2, Na2O, K2O, Fe2O3, and SrO. Further, the TiO2 contents in the TiO2-coated non-porous and porous glass cloths were evaluated using XRF. Figure 4 plots the TiO2 content and weight loss of the glass cloth as functions of the acid leaching time. The weight loss of the glass cloth was calculated as follows: Weight loss (%) = Weight change o f glass cloth a f ter acid leaching Weight o f glass cloth be f ore leaching × 100 :HLJKWORVVRIJODVVFORWK /HDFKLQJWLPHK 7L2 FRQWHQWLQFRPSRVLWHZW Figure 4. TiO2 content in the TiO2-coated porous glass cloth (open squares) and weight loss of the porous glass cloth (ﬁlled circles) versus the acid leaching time. 2.4. N2 Adsorption Figure 5 depicts the N2 adsorption and desorption isotherms of the non-porous and the 3-h leached porous glass cloths before and after TiO2 coating. The non-porous samples (Figure 5a) yielded a type III adsorption isotherm, conﬁrming the absence of micrometer- or nanometer-sized pores [35]. However, the porous samples (Figure 5b) yielded a type I adsorption isotherm, indicating the existence of nanometer-sized pores (<2 nm) [35]. The sample that was prepared in an extended leaching time (12 h) yielded a type I adsorption isotherm, indicating that the pore size will not be signiﬁcantly changed by prolonging the leaching time. Figure 6 plots the speciﬁc surface areas (Sg) of the TiO2-coated and uncoated glass cloth samples versus the acid leaching time. The Sg values of both the samples increased up to approximately 4 h of leaching time and remained almost constant thereafter, indicating that the microporous structure was completely formed after 4 h of leaching. However, in the SEM images, the ﬁber core disappeared after 3 h of leaching (Figure 1c), implying that the microporous structure continued to evolve after the ﬁber center was subjected to acid leaching. After the disappearance of the core, Sg was probably increased by the leaching of the non-silica components that remained in the porous glass cloth and by the precipitation of silica gel in the pores. The latter process is typically observed during the leaching of the phase-separated borosilicate glasses [36,37]. Figure 5. Adsorption (Ads.) and desorption (Des.) isotherms of (a) a non-porous glass cloth and (b) a porous glass cloth before and after TiO2 coating. Figure 5. Adsorption (Ads.) and desorption (Des.) isotherms of (a) a non-porous glass cloth and (b) a porous glass cloth before and after TiO2 coating. As depicted in Figure 5b, the TiO2 coating reduced the adsorbed/desorbed N2 in the low pressure region or the isotherm. This decrease corresponds to the lower Sg value of the TiO2-coated samples when compared to that of the uncoated samples depicted in Figure 5. This decrease cannot be ascribed to the ﬁlling of nanometer-sized pores by TiO2 particles because even the primary particle size of TiO2 (6 nm) is considerably larger than the estimated pore diameter (<2 nm). To elucidate the reason that the Sg decreased during the TiO2 coating process, we investigated the manner in which heat treatment after dip-coating affected the Sg of the samples. 2.3. Chemical Composition Analysis Weight loss (%) = Weight change o f glass cloth a f ter acid leaching Weight o f glass cloth be f ore leaching × 100 :HLJKWORVVRIJODVVFORWK /HDFKLQJWLPHK 7L2 FRQWHQWLQFRPSRVLWHZW /HDFKLQJWLPHK Figure 4. TiO2 content in the TiO2-coated porous glass cloth (open squares) and weight loss of the porous glass cloth (ﬁlled circles) versus the acid leaching time. Both the TiO2 loading amount and weight loss of the glass cloth increased at the maximum leaching time of 3 h. Even if the TiO2 loading amount was unaffected by the leaching time, the TiO2 composition ratio was observed to increase because of the weight loss of the glass cloth. However, the increase in the TiO2 composition ratio exceeded the expected value corresponding to this weight loss, demonstrating the remaining factors that were responsible for this increase in TiO2 composition ratio. As will be discussed subsequently, the TiO2 particles cannot penetrate the pores in the glass ﬁber structure; therefore, the increased volume of pores cannot be explained by the increased loading 275 Catalysts 2019, 9, 82 amount of TiO2. However, the large cracks that are observed in the SEM images (Figure 1e,f) are considered to be the likely support sites of the TiO2 particles in the dip-coating process. Further, the pore formation can also increase the TiO2 content by increasing the surface roughness or the wettability of the glass surface. In contrast, the increase in leaching time from 3 to 12 h decreases the TiO2 amount. However, the factors that can affect the TiO2 loading, weight loss, cracking, and surface conditions were not signiﬁcantly altered in this time range. Therefore, the loading amount of TiO2 was observed to be saturated rather than decreased. 2.4. N2 Adsorption The porous glass cloth that was prepared by 3 h of leaching was subjected to the same heat treatment as that used in the dip-coated samples 276 Catalysts 2019, 9, 82 (300 ◦C for 2 h), and its measured Sg was compared with that of the samples obtained before heat treatment. The heat treatment decreased the Sg from 370 to 200 m2/g, indicating that heating was mainly responsible for the loss of Sg in the TiO2-coated sample. To verify the effects of coating and heating on the porous properties, we investigated the pore-size distribution in the samples. Figure S2 depicts the adsorption isotherms of the uncoated and TiO2-coated porous glass cloths and their pore-size distributions estimated by Saito–Foley ﬁtting [38,39]. The majority of the pores were observed to be less than 1 nm in diameter, and the TiO2 coating and heating process reduced the volume of the large pores. The reduced Sg and enlarged pores in the heat-treated sample may have arisen from dehydration condensation of the silanol groups on the surface of the pores [40]. However, because the subsequent heat treatment ensured that a high speciﬁc surface area of the TiO2-coated glass cloth was retained, the organic-molecule adsorption ability of the TiO2-coated sample should not have been signiﬁcantly degraded. Figure 6. Speciﬁc surface area versus leaching time of the TiO2-coated porous glass cloths (squares) and the uncoated porous glass cloth (circles). Figure 6. Speciﬁc surface area versus leaching time of the TiO2-coated porous glass cloths (squares) and the uncoated porous glass cloth (circles). 2.5. Adsorption and Photocatalytic Ability of 2-Propanol 2.5. Adsorption and Photocatalytic Ability of 2-Propanol The porous sample was prepared by 3 h of leaching. The dark adsorption and photocatalytic deposition experiments were conducted at 15 ◦C and in the presence of low water content (2 mg/L) in a 2-propanol atmosphere. &RQFHQWUDWLRQSSP 89LUUDGLDWLRQWLPHK E SURSDQRO DFHWRQH Δ &2 Δ &2 SSP F D 7LPHPLQ &RQFHQWUDWLRQSSP &RQFHQWUDWLRQSSP 89LUUDGLDWLRQWLPHK 7LPHPLQ Figure 7. (a) Temporal concentration changes of gaseous 2-propanol under dark conditions for the TiO2-coated non-porous (crosses) and porous (open squares) glass cloths. (b,c) Concentrations of gaseous 2-propanol, acetone, and CO2 under subsequent UV light irradiation for the TiO2-coated non-porous and porous glass cloths, respectively. The porous sample was prepared by 3 h of leaching. The dark adsorption and photocatalytic deposition experiments were conducted at 15 ◦C and in the presence of low water content (2 mg/L) in a 2-propanol atmosphere. Further, the effects of temperature and water content on the adsorption and decomposition abilities of the TiO2-coated porous glass cloth were investigated. Figure 8a depicts the trends of the 2-propanol adsorption under dark conditions in 2-propanol atmospheres with both high (18 mg/L) and low (2 mg/L) water contents. Both the experiments were conducted at 35 ◦C. A higher amount of 2-propanol was adsorbed from dry air when compared to that adsorbed from moist air. In the presence of high water contents, the alcohol and water molecules compete to be adsorbed on the silanol sites of the silica surface [41], thereby reducing the amount of adsorbed 2-propanol. The concentration changes of 2-propanol, acetone, and CO2 in dry and moist atmospheres during UV irradiation are depicted in Panels (b,c) of Figure 8. Acetone was not detected under either condition, which was similar to that observed in the experiment that was conducted at 15 ◦C in the presence of a low water content (Figure 7c). Meanwhile, more CO2 was generated at 35 ◦C than at 15 ◦C with low water content. In addition, at 35 ◦C, slightly more CO2 was generated under the high-water atmosphere when compared to that generated under the low-water atmosphere. 2.5. Adsorption and Photocatalytic Ability of 2-Propanol First, the adsorption and photocatalytic decomposition abilities of the TiO2-coated porous glass cloth, that was prepared by 3 h of leaching, and the TiO2-coated non-porous glass cloth were compared at 15 ◦C and a low water content (2 mg/L). Figure 7a plots the 2-propanol concentration versus time under dark conditions. The non-porous sample gradually reduced the concentration of gaseous 2-propanol during the initial 60 min, whereas the porous sample rapidly reduced the 2-propanol concentration within the initial 10 min, reducing it to lower than the detection limit in 20 min. These results indicate the strong absorbency of the porous sample for gaseous 2-propanol. The panels (b,c) of Figure 7 depict the temporal changes of 2-propanol concentration under UV light irradiation for the TiO2-coated non-porous and porous glass cloths, respectively, after the adsorption experiment is conducted in dark conditions. Further, the photocatalytic oxidation decomposition of gaseous 2-propanol tends to desorb acetone from the TiO2 surface; consequently, acetone is formed as a typical decomposition intermediate of this process [33]. The acetone concentration that was desorbed from the TiO2-coated non-porous glass cloth increased to its maximum at 2.5 h of irradiation time and gradually decreased. Meanwhile, the CO2 concentration monotonically increased with the UV light irradiation time. However, acetone was not detected from the porous glass sample, and the CO2 concentration rapidly increased until approximately 1.5 h of the irradiation time; further, it gradually increased up to 4 h of irradiation time. These results indicate that the acetone that was generated on the porous sample was not desorbed from the surface but was decomposed to CO2. Further, a similar reaction process has been reported for the TiO2–zeolite composite [12]. The rapidly increasing amount of CO2 indicates 277 Catalysts 2019, 9, 82 the effectiveness of photocatalytic decomposition during the early stages because of the 2-propanol concentration in the composite. D &RQFHQWUDWLRQSSP 89LUUDGLDWLRQWLPHK 7LPHPLQ E SURSDQRO DFHWRQH Δ &2 Δ &2 SSP F &RQFHQWUDWLRQSSP Figure 7. (a) Temporal concentration changes of gaseous 2-propanol under dark conditions for the TiO2-coated non-porous (crosses) and porous (open squares) glass cloths. (b,c) Concentrations of gaseous 2-propanol, acetone, and CO2 under subsequent UV light irradiation for the TiO2-coated non-porous and porous glass cloths, respectively. 2.5. Adsorption and Photocatalytic Ability of 2-Propanol &RQFHQWUDWLRQSSP 89LUUDGLDWLRQWLPHK SURSDQRO DFHWRQH Δ &2 F E 7LPHPLQ &RQFHQWUDWLRQSSP Δ &2 SSP D Figure 8. (a) Temporal concentration changes of 2-propanol during the dark storage of TiO2-coated porous glass cloth in gaseous 2-propanol with 2-mg/L water content (closed squares) and 18-mg/L water content (open squares). (b,c) The concentration changes of 2-propanol, acetone, and CO2 during UV irradiation of the TiO2-coated non-porous glass cloth in gaseous 2-propanol with 2-mg/L and 18-mg/L water contents, respectively. All the experiments were conducted at 35 ◦C. &RQFHQWUDWLRQSSP 89LUUDGLDWLRQWLPHK SURSDQRO DFHWRQH Δ &2 F E PLQ Δ &2 SSP 7LPHPLQ &RQFHQWUDWLRQSSP D 89LUUDGLDWLRQWLPHK 7LPHPLQ Figure 8. (a) Temporal concentration changes of 2-propanol during the dark storage of TiO2-coated porous glass cloth in gaseous 2-propanol with 2-mg/L water content (closed squares) and 18-mg/L water content (open squares). (b,c) The concentration changes of 2-propanol, acetone, and CO2 during UV irradiation of the TiO2-coated non-porous glass cloth in gaseous 2-propanol with 2-mg/L and 18-mg/L water contents, respectively. All the experiments were conducted at 35 ◦C. Figure 8. (a) Temporal concentration changes of 2-propanol during the dark storage of TiO2-coated porous glass cloth in gaseous 2-propanol with 2-mg/L water content (closed squares) and 18-mg/L water content (open squares). (b,c) The concentration changes of 2-propanol, acetone, and CO2 during UV irradiation of the TiO2-coated non-porous glass cloth in gaseous 2-propanol with 2-mg/L and 18-mg/L water contents, respectively. All the experiments were conducted at 35 ◦C. 278 Catalysts 2019, 9, 82 Catalysts 2019, 9, 82 The rate of CO2 generation by the decomposition of 2-propanol in the TiO2-coated porous glass cloth (Figures 6c and 7b,c) changed after approximately 1.5 h of UV irradiation. No rate change could be observed in the non-porous sample (Figure 6b), conﬁrming that the rate change phenomenon originated from the porous structure. The proposed decomposition model is depicted in Figure 9. 2.5. Adsorption and Photocatalytic Ability of 2-Propanol In the TiO2-coated porous glass cloth, the TiO2 particles are mainly supported on the external surface of the porous glass ﬁbers and do not penetrate the pores. Therefore, the photocatalytic reaction occurs on the external ﬁber surface, whereas the porous interior provides the adsorption sites of 2-propanol, acetone, and water. When the dried sample is inserted into a reactor ﬁlled with gaseous 2-propanol, the 2-propanol and water molecules are competitively adsorbed on the SiO2 walls of the pores. Wu et al. clariﬁed that the water and alcohol molecules emit comparatively large adsorption heat when they are adsorbed onto a bare SiO2 surface than that emitted when they are adsorbed onto or when they form cluster with the molecules that have already being adsorbed onto the SiO2 surface [42]. This observation indicates that the weakly and strongly adsorbed molecules coexist on the SiO2 walls of the pores (Figure 9a). Under UV illumination, the 2-propanol molecules near the TiO2 surface are decomposed to CO2 via the acetone intermediate. In the early phase of photocatalytic decomposition, the weakly adsorbed 2-propanol molecules on the SiO2 walls are preferentially desorbed and diffuse from the interior of the ﬁber to near the TiO2 surface. Further, 2-propanol diffusion provides a sufﬁcient supply of 2-propanol for sustaining CO2 generation at a high rate (Figure 9b). The reaction rate gradually decreases as the supply of weakly adsorbed 2-propanol reduces. In the later phase, the CO2 generation rate is limited by the desorption and diffusion of the strongly adsorbed 2-propanol molecules (Figure 9c). This limiting rate corresponds to the decelerated CO2 generation rate after 1.5 h of light illumination in the photocatalytic decomposition of 2-propanol on the TiO2-coated porous glass cloth (Figures 7c and 8b,c). $FHWRQH &2 :HDNO\DGVRUEHGSURSDQRO 6WURQJO\DGVRUEHGSURSDQRO 6WURQJO\DGVRUEHGZDWHU :HDNO\DGVRUEHGZDWHU D E F 7L2 3RUHLQ6L2 Figure 9. Schematics of (a) adsorption, (b) early-phase photocatalytic oxidation, and (c) late-phase photocatalytic oxidation on the TiO2-coated porous glass cloth. D F Figure 9. Schematics of (a) adsorption, (b) early-phase photocatalytic oxidation, and (c) late-phase photocatalytic oxidation on the TiO2-coated porous glass cloth. Based on this decomposition model, we compared the results of photocatalytic decomposition at 15 ◦C and 35 ◦C under a 2-propanol atmosphere with a low (2 mg/L) water content (Figures 7c and 8b). 2.5. Adsorption and Photocatalytic Ability of 2-Propanol During the early phase of decomposition, the CO2 generation rate was observed to be only slightly higher at 35 ◦C than that at 15 ◦C; however, during the later phase of decomposition, the CO2 decomposition rates of the two conditions were observed to be clearly different. More speciﬁcally, the CO2 generation rates after 2–6 h of UV light irradiation were 18 and 54 ppm/h at 15 ◦C and 35 ◦C, 279 Catalysts 2019, 9, 82 respectively. The higher CO2 generation rate observed at 35 ◦C can be explained by the accelerated diffusion of 2-propanol from inside the porous glass ﬁbers to the TiO2 surface by a higher temperature. The effect of water content was further evaluated from the photocatalytic decomposition results at 35 ◦C in 2-propanol atmospheres with low (2 mg/L) and high (18 mg/L) water contents (Figure 8b,c). After 30 min of light illumination, the CO2 concentrations in the dry and moist atmospheres were observed to be identical; however, the CO2 generation rate changed at CO2 concentrations of 440 and 560 ppm in dry and moist atmospheres, respectively. This result indicates that the CO2 generation rate changed at a later stage of the decomposition process in the moist condition. This difference may be obtained from the condition of 2-propanol adsorption in porous glass ﬁbers. As shown in the decomposition model (Figure 9), the weakly and strongly adsorbed 2-propanol molecules coexisted in the pores. In the moist condition, the abundant water molecules are expected to occupy a large portion of the strong adsorption sites on the bare SiO2 surface, thereby decreasing the ratio of the amount of strongly adsorbed 2-propanol to that of the weakly adsorbed 2-propanol. Because the weakly adsorbed 2-propanol molecules are easily desorbed from the SiO2 walls, a sufﬁcient supply of 2-propanol can continue the generation of CO2 at a constant rate for a certain time. In contrast, in the dry condition, the proportion of strongly adsorbed 2-propanol should be relatively high. Once the weakly adsorbed 2-propanol molecules have been consumed in the early phase, the CO2 generation rate will gradually decrease. We further discuss the reason why the CO2 generation rates were identical during the earliest phase of 2-propanol decomposition. In the surface-reaction limited situation, the rate of mass transportation is generally greater than the rate of surface reaction; therefore, the decomposition rate is considered to be independent of the reactant concentration. 3.2. Preparation of the TiO2-Coated Porous Glass Cloth The organic compounds on the ﬁbers of two pieces of glass cloth (ca. (30 × 35) mm2 × 2, total weight 0.19 g) were eliminated by heating at 500 ◦C for 1 h. The heat-treated cloth samples were immersed in 13.3 ml of 2.5 mol/L HCl aqueous solution at 40 ◦C for 0.5, 1, 3, 4, 6, or 12 h in a screw-capped perﬂuoroalkoxy alkane (PFA) container without stirring. After leaching by HCl, the samples were washed several times in distilled water and were further immersed in 80 ml of distilled water for 10 min. This immersion process was repeated, replacing the water between each immersion, until the pH of the immersing water reached ca. 7 (in this case, after two washes). The samples were washed with ethanol and immersed in sufﬁcient ethanol to perform solvent exchange in the sample pores. After 10 min in ethanol, the samples were dried at 120 ◦C for 10 min to form the porous glass cloth. A reference sample with a non-porous structure was prepared by heating the glass ﬁber cloth at 500 ◦C for 1 h without subsequent acid leaching. This sample was used as the non-porous glass cloth. TiO2 was coated on both the porous and non-porous glass cloths using the conventional dip-coating method. The TiO2 coating solution was prepared by diluting the TiO2 suspension to 1 wt % in 2-propanol. The glass cloth samples were dipped in the diluted TiO2 suspension and were further pulled up at 1.0 mm/s. The dip-coated samples were subsequently dried at room temperature and at 120 ◦C for 10 min; further, they were ﬁnally heated at 300 ◦C for 2 h. The obtained samples were referred to as the TiO2-coated porous and non-porous glass cloths. On the other hand, as a reference in XRD measurement, the TiO2 powder sample was also prepared from the TiO2 suspension by drying at 120 ◦C and subsequently heating at 300 ◦C for 2 h. 2.5. Adsorption and Photocatalytic Ability of 2-Propanol In the mass-transportation limited situation, the rate of surface reaction is greater than the rate of mass transportation, and the decomposition rate is related to the reactant concentration near the reaction site [7]. In our experiments, the concentration of weakly adsorbed 2-propanol in the porous glass ﬁber was expected to be dependent on the water content; however, the initial CO2 generation rate remained constant. This indicated that the photocatalytic decomposition was limited by the surface reaction on TiO2 in this phase. Finally, the effect of water on the radical reaction at the photo-illuminated TiO2 surface is discussed. In the photocatalytic reaction on TiO2, the adsorbed water molecules react with the photogenerated carriers at the TiO2 surface and change to hydroxyl radicals [43]. These hydroxyl radicals further diffuse and decompose the organic compounds near the TiO2 surface; therefore, a certain amount of water vapor can enhance the photocatalytic decomposition rate of organic compounds such as 2-propanol [29,44,45]. Further, excess water vapor prevents the adsorption of organic molecules on the TiO2 surface, thereby decreasing the photocatalytic decomposition rate [29,32,44]. However, as noted above, the initial CO2 generation rates estimated from the CO2 concentration after 30 min of UV irradiation were observed to remain the same under both dry and moist conditions. It indicated that the CO2 generation rates under both conditions were very similar because the amount of water in the gaseous phase was not signiﬁcantly different under both conditions. The majority of the water in the present reactor was expected to be adsorbed on the large Sg of the porous glass ﬁber cloth. Large water adsorption will decrease the water concentration in the gaseous phase, which can be used to plausibly explain the similar CO2 generation rates at high and low water contents in the earliest phase of 2-propanol decomposition. In the later phase of the decomposition, the CO2 generation rates were estimated in the UV light irradiation time ranging from 2 to 6 h. The rates were 54 and 46 ppm/h under low and high water content conditions, respectively. Within this range, the water content did not largely inﬂuence the CO2 generation rates because the rates were limited by the diffusion of the strongly adsorbed 2-propanol molecules. 280 Catalysts 2019, 9, 82 Catalysts 2019, 9, 82 3.1. Materials The glass ﬁber cloth was provided by Arisawa Manufacturing Co. Ltd. (Joetsu, Japan). Figure 10 is a photomicrograph of the glass ﬁber cloth. It has a plain weave structure, and approximately 400 E-glass ﬁbers of 9.1-μm diameters form the warp and weft thread; further, the thread density, the number of warp and the weft thread per certain area of woven fabric was 44 × 32 in 25 × 25 mm2. The thickness of the cloth was 180 μm, and the weight per unit area was 203.5 g/m2. The TiO2 (anatase) nanoparticles were dispersed in 2-propanol (TKD-701, 17.0 wt %, d = 6 nm, TAYCA Co., Osaka, Japan). Further, reagent-grade aqueous hydrochloric acid (HCl; 5 mol/L), ethanol (99.5 wt %), 2-propanol (99.7 wt %), and sodium carbonate (99.5 wt %) were supplied by Wako Pure Chemical Industries Ltd. (Osaka, Japan) and were used without any puriﬁcation. Figure 10. Photograph of the magniﬁed surface of the E-glass cloth. Figure 10. Photograph of the magniﬁed surface of the E-glass cloth. 3. Materials and Methods 3.1. Materials 3.1. Materials 3.3. Characterization For acid leaching (by varying the leaching time from 2 to 12 h), the porous samples were mixed with 25 mg of Na2CO3 as the ﬂux. The pressed pellets were sintered by a 3-step heating process (700 ◦C for 2 h, 800 ◦C for 2 h, and 900 ◦C for 4 h), yielding the glass disks that were required to perform XRF analysis. Further, the porous properties of the samples were determined from the N2 gas adsorption isotherms measured at 77 K (BELLSOAP mini II, BEL Japan Inc., Osaka, Japan). Prior to performing the N2 gas adsorption measurements, all the samples were dried in vacuo at 120 ◦C for 2 h. Their speciﬁc surface areas were calculated using the Brunauer–Emmet–Teller (BET) multi-plot method. To analyze their pore-size distributions, the samples were dried in vacuo at 140 ◦C for 10 h, and the N2 gas adsorption was measured at a very low pressure range (from 10−3 Pa, BELSORP-max-N-VP-CM, BEL Japan Inc., Osaka, Japan). The pore-size distribution was estimated by Saito–Foley ﬁtting using the zeolite Y standard. The adsorption and photocatalytic decomposition properties of 2-propanol were examined using the TiO2-coated porous glass cloth and the TiO2-coated non-porous glass cloth as reference. The samples were cut into divisions of 0.042 ± 0.001 g for performing the experiments. The cut samples were pre-treated by UV–vis light irradiation under a black light (BL) lamp (FL15BLB, Toshiba Lighting & Technology Co., Yokosuka, Japan, peak wavelength: 365 nm, light intensity at 365 nm: 3.5 mW/cm2 at the sample surface) for 24 h and were subsequently heated at 120 ◦C in vacuo to eliminate the adsorbed organic molecules and water. Further, the concentrations of 2-propanol and water vapor were adjusted using the apparatus depicted in Figure 11. The air-diluted 2-propanol vapor that was produced by a calibration gas generator (Permeater PD-1B, GASTECH Co., Ayase, Japan) was mixed with the humid air, that was produced through two steps of water bubbling. The resulting mixed gas contained 300 ppm of 2-propanol and 2 or 18 mg/L of water. The 2-propanol concentration in the mixed gas was conﬁrmed by a gas chromatograph (GC-8A, Shimadzu Co., Kyoto, Japan) using a thermal conductivity detector (TCD), a porous polymer beads column (Sunpak-A, 2 m, 160 ◦C, Shinwa Chemical Industries Ltd., Kyoto, Japan), and He carrier gas (20 mL/min). 3.3. Characterization The surface morphologies of the samples were observed by ﬁeld emission scanning electron microscopy (FE–SEM, Hitachi S-2400 and S-5200, Hitachi High-Technologyies, Tokyo, Japan). Before the FE–SEM observation, the samples were platinum-coated using a sputtering method. The XRD patterns were collected using CuKα radiation (λ = 0.15406 nm, monochromatized by Ni ﬁlter) by 281 Catalysts 2019, 9, 82 an X-ray diffractometer (MiniFlex 600, Rigaku Co., Tokyo, Japan) that was operated at 40 kV and 15 mA. The chemical compositions of the samples were analyzed using an X-ray ﬂuorescence (XRF) spectrometer (ZSX Primus μ, Rigaku Co., Tokyo, Japan). For the XRF measurements, glass disks were prepared from the TiO2-coated and uncoated samples as follows. First, 150 mg of the sample was ground with a pestle in an alumina mortar and was formed into a pellet by uniaxial pressing. For acid leaching (by varying the leaching time from 2 to 12 h), the porous samples were mixed with 25 mg of Na2CO3 as the ﬂux. The pressed pellets were sintered by a 3-step heating process (700 ◦C for 2 h, 800 ◦C for 2 h, and 900 ◦C for 4 h), yielding the glass disks that were required to perform XRF analysis. Further, the porous properties of the samples were determined from the N2 gas adsorption isotherms measured at 77 K (BELLSOAP mini II, BEL Japan Inc., Osaka, Japan). Prior to performing the N2 gas adsorption measurements, all the samples were dried in vacuo at 120 ◦C for 2 h. Their speciﬁc surface areas were calculated using the Brunauer–Emmet–Teller (BET) multi-plot method. To analyze their pore-size distributions, the samples were dried in vacuo at 140 ◦C for 10 h, and the N2 gas adsorption was measured at a very low pressure range (from 10−3 Pa, BELSORP-max-N-VP-CM, BEL Japan Inc., Osaka, Japan). The pore-size distribution was estimated by Saito–Foley ﬁtting using the zeolite Y standard. an X-ray diffractometer (MiniFlex 600, Rigaku Co., Tokyo, Japan) that was operated at 40 kV and 15 mA. The chemical compositions of the samples were analyzed using an X-ray ﬂuorescence (XRF) spectrometer (ZSX Primus μ, Rigaku Co., Tokyo, Japan). For the XRF measurements, glass disks were prepared from the TiO2-coated and uncoated samples as follows. First, 150 mg of the sample was ground with a pestle in an alumina mortar and was formed into a pellet by uniaxial pressing. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/1/82/s1. Figure S1: (a) Surface and (b) fracture cross section of the TiO2 coating on the porous glass cloth, Figure S2: (a) N2 adsorption isotherms of the non-coated and TiO2-coated porous glass cloths and (b) pore-size distributions estimated by Saito–Foley ﬁtting using the adsorption potential for N2 on zeolite Y. Va: adsorbed volume, dp: pore diameter, Vp: pore volume. 3.3. Characterization The water contents in the mixed gas were also conﬁrmed in the gas detector tube (No.6, GASTEC Co., Ayase, Japan). The mixed gas ﬂowed into a gas-tight bag containing two glass vial reactors (diameter: 4 cm; height: 6 cm, volume: approximately 65 mL). After 50 min of gas ﬂow, the pre-treated cloth sample was transferred into one of the vials, and both the vial reactors were immediately sealed with a gas-tight septum. The sealed vials containing the sample, as well as the mixed gas or mixed gas only were transferred to an incubator (maintained at 15 ◦C or 35 ◦C) and left for 1 h in the dark. During this time, 2-propanol was adsorbed without conducting a photocatalytic reaction. The glass vials were illuminated by UV–vis light under a BL lamp (with 1.0 mW/cm2 of light intensity at the sample position with no shielding of the glass vial) for 6 h in the incubator. The concentrations of 2-propanol, acetone, and CO2 in the glass vial were determined at 10- and 30-min intervals during the 2-propanol adsorption and photocatalytic decomposition, respectively, by gas chromatography. 282 Catalysts 2019, 9, 82 &DOLEUDWLRQJDV JHQHUDWRU %XEEOHU 'U\ $LU SURSDQROJDV +XPLGDLU 5HJXODWRU )URZPHWHU :DWHUEDWKR& URYHEDJ ODVVYLDO UHDFWRU ODVVYLDO UHIHUHQFH VDPSOH 6HSWDPFDS Figure 11. Apparatus settings for controlling the water content in gaseous 2-propanol. URYHEDJ ODVVYLDO UHDFWRU ODVVYLDO UHIHUHQFH VDPSOH 6HSWDPFDS Figure 11. Apparatus settings for controlling the water content in gaseous 2-propanol. Author Contributions: Conceptualization, S.Y.; methodology, S.Y. and A.Y.; investigation, K.H., K.I., and S.Y.; writing–original draft preparation, S.Y.; writing–review and editing A.Y. and K.I.; supervision, A.Y. Funding: This work was supported in part by Nippon Sheet Glass Foundation for Materials Science and Engineering. A part of the publication cost was supported by the Gender Equality Ofﬁce, University of Yamanashi. 4. Conclusions A composite material of TiO2 and porous glass cloth was prepared by acid leaching of an E-glass cloth and subsequent dip-coating of a TiO2 photocatalyst. Working inward from the ﬁber surface, acid leaching resulted in the formation of a porous shell structure during an early stage. The ﬁbers were observed to become completely porous after 3 h of leaching; however, the speciﬁc surface area of the acid-leached E-glass cloth continued to increase for another hour. Consequently, the speciﬁc surface area was maximized after 4 h of leaching. The compositional change was observed in the acid-leached E-glass by performing XRF analysis; Al2O3 and B2O3 were eliminated along with the ions of alkaline metals and alkali earth metals, leaving mainly the SiO2 structure. During the photo-catalytic decomposition of 2-propanol, the TiO2-coated porous glass cloth exhibited considerable adsorption ability with respect to 2-propanol and generated CO2 at a higher rate than that of the TiO2-coated non-porous glass cloth. The TiO2-coated porous glass cloth also adsorbed acetone until the decomposition to CO2 was completed. Increasing the temperature from 15 to 35 ◦C clearly increased the CO2 generation rate of the TiO2-coated porous glass cloth because the diffusion rate of the reactant molecules was accelerated. Further, increasing the moisture content from 2 to 18 mg/L at 35 ◦C slightly decreased the amount of adsorbed 2-propanol and delayed the change in the rate-controlling step from surface reaction to mass transportation. This indicated that the competitive adsorption of water and 2-propanol in the porous glass ﬁber decreased the amount of strongly adsorbed 2-propanol in the moist atmosphere. Finally, the porous glass cloth that was prepared from a commercial E-glass cloth provided a sufﬁciently strong TiO2 support with a high speciﬁc surface area. The TiO2-coated porous glass cloth can adsorb and photo-catalytically degrade VOCs such as 2-propanol and acetone. Therefore, it is considered to be a strong candidate for ensuring the practical elimination of gaseous organic pollutants. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/1/82/s1. Figure S1: (a) Surface and (b) fracture cross section of the TiO2 coating on the porous glass cloth, Figure S2: (a) N2 adsorption isotherms of the non-coated and TiO2-coated porous glass cloths and (b) pore-size distributions estimated by Saito–Foley ﬁtting using the adsorption potential for N2 on zeolite Y. Va: adsorbed volume, dp: pore diameter, Vp: pore volume. Author Contributions: Conceptualization, S.Y.; methodology, S.Y. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Kostiainen, R. Volatile organic compounds in the indoor air of normal and sick houses. Atmos. Environ. 1995, 29, 693–702. [CrossRef] 2. Osha Annotated Table Z-1. Available online: http://www.webcitation.org/71UI8WUO9 (accessed on 7 August 2018). 3. Hoffmann, M.R.; Martin, S.T.; Choi, W.; Bahnemann, D.W. Environmental applications of semiconductor photocatalysis. Chem. Rev. 1995, 95, 69–96. [CrossRef] 4. Pichat, P.; Disdier, J.; Hoang-Van, C.; Mas, D.; Goutailler, G.; Gaysse, C. Puriﬁcation/deodorization of indoor air and gaseous efﬂuents by TiO2 photocatalysis. Catal. Today 2000, 63, 363–369. [CrossRef] 5. Wang, S.; Ang, H.M.; Tade, M.O. Volatile organic compounds in indoor environment and photocatalytic oxidation: State of the art. Environ. Int. 2007, 33, 694–705. [CrossRef] [PubMed] Mamaghani, A.H.; Haghighat, F.; Lee, C.-S. Photocatalytic oxidation technology for indoor environment air 6. Mamaghani, A.H.; Haghighat, F.; Lee, C.-S. Photocatalytic oxidation technology for indoor env puriﬁcation: The state-of-the-art. Appl. Catal. B Environ. 2017, 203, 247–269. [CrossRef] Mamaghani, A.H.; Haghighat, F.; Lee, C.-S. Photocatalytic oxidation technology for indoor environment air puriﬁcation: The state-of-the-art Appl Catal B Environ 2017 203 247–269 [CrossRef] g g g y gy puriﬁcation: The state-of-the-art. Appl. Catal. B Environ. 2017, 203, 247–269. [CrossRef] on: The state-of-the-art. Appl. Catal. B Environ. 2017, 20 7. Ohko, Y.; Fujishima, A.; Hashimoto, K. Kinetic analysis of the photocatalytic degradation of gas-phase 2-propanol under mass transport-limited conditions with a TiO2 ﬁlm photocatalyst. J. Phys. Chem. B 1998, 102, 1724–1729. [CrossRef] 8. Takeuchi, M.; Hidaka, M.; Anpo, M. Efﬁcient removal of toluene and benzene in gas phase by the TiO2/y-zeolite hybrid photocatalyst. J. Hazard. Mater. 2012, 237, 133–139. [CrossRef] 9. Yoneyama, H.; Torimoto, T. Titanium dioxide/adsorbent hybrid photocatalysts for photodestruction of organic substances of dilute concentrations. Catal. Today 2000, 58, 133–140. [CrossRef] 10. Ao, C.H.; Lee, S.C. Combination effect of activated carbon with TiO2 for the photodegradation of b pollutants at typical indoor air level. J. Photochem. Photobiol. A Chem. 2004, 161, 131–140. [CrossRef] 11. Mo, J.; Zhang, Y.; Xu, Q.; Yang, R. Effect of TiO2/adsorbent hybrid photocatalysts for toluene decompos in gas phase. J. Hazard. Mater. 2009, 168, 276–281. [CrossRef] 12. Yasumori, A.; Yanagida, S.; Sawada, J. Preparation of a titania/x-zeolite/porous glass composite photocatalyst using hydrothermal and drop coating processes. Molecules 2015, 20, 2349–2363. [CrossRef] [PubMed] 13. Zhang, M.; An, T.; Fu, J.; Sheng, G.; Wang, X.; Hu, X.; Ding, X. Photocatalytic degradation of mixed gaseous carbonyl compounds at low level on adsorptive TiO2/SiO2 photocatalyst using a ﬂuidized bed reactor. 4. Conclusions and A.Y.; investigation, K.H., K.I., and S.Y.; writing–original draft preparation, S.Y.; writing–review and editing A.Y. and K.I.; supervision, A.Y. Funding: This work was supported in part by Nippon Sheet Glass Foundation for Materials Science and Engineering. A part of the publication cost was supported by the Gender Equality Ofﬁce, University of Yamanashi. Conﬂicts of Interest: The authors declare no conﬂict of interest. 283 Catalysts 2019, 9, 82 References Chemosphere 2006, 64, 423–431. [CrossRef] [PubMed] 14. Kang, M.; Hong, W.-J.; Park, M.-S. Synthesis of high concentration titanium-incorporated nanoporous silicates (ti-nps) and their photocatalytic performance for toluene oxidation. Appl. Catal. B Environ. 2004, 53, 195–205. [CrossRef] 15. Tasbihi, M.; Štangar, U.L.; Škapin, A.S.; Risti´c, A.; Kauˇciˇc, V.; Tušar, N.N. Titania-containing mesoporous silica powders: Structural properties and photocatalytic activity towards isopropanol degradation. J. Photochem. Photobiol. A Chem. 2010, 216, 167–178. [CrossRef] 16. Anpo, M.; Aikawa, N.; Kubokawa, Y.; Che, M.; Louis, C.; Giamello, E. Photoluminescence and photocatalytic activity of highly dispersed titanium oxide anchored onto porous vycor glass. J. Phys. Chem. 1985, 89, 5017–5021. [CrossRef] 17. Yamashita, H.; Ichihashi, Y.; Harada, M.; Stewart, G.; Fox, M.A.; Anpo, M. Photocatalytic degradation of 1-octanol on anchored titanium oxide and on TiO2 powder catalysts. J. Catal. 1996, 158, 97–101. [CrossRef] 18. Yamashita, H.; Honda, M.; Harada, M.; Ichihashi, Y.; Anpo, M.; Hirao, T.; Itoh, N.; Iwamoto, N. Preparation of titanium oxide photocatalysts anchored on porous silica glass by a metal ion-implantation method and their photocatalytic reactivities for the degradation of 2-propanol diluted in water. J. Phys. Chem. B 1998, 102, 10707–10711. [CrossRef] 19. Yazawa, T.; Machida, F.; Kubo, N.; Jin, T. Photocatalytic activity of transparent porous glass supported TiO2. Ceram. Int. 2009, 35, 3321–3325. [CrossRef] 20. Wallenberger, F.T. Commercial and experimental glass ﬁbers. In Fiberglass and Glass Technology: Energy-Friendly Compositions and Applications; Wallenberger, F.T., Bingham, P.A., Eds.; Springer: Boston, MA, USA, 2010; pp. 3–90. ISBN 978-1-4419-0736-3. 21. Caddock, B.D.; Evans, K.E.; Masters, I.G. Diffusion behaviour of the core-sheath structure in e-glass ﬁbres exposed to aqueous HCl. J. Mater. Sci. 1989, 24, 4100–4105. [CrossRef] 284 Catalysts 2019, 9, 82 22. Li, H.; Gu, P.; Watson, J.; Meng, J. Acid corrosion resistance and mechanism of E-glass ﬁbers: Boron factor. J. Mater. Sci. 2013, 48, 3075–3087. [CrossRef] 23. Kiwi-Minsker, L.; Yuranov, I.; Siebenhaar, B.; Renken, A. Glass ﬁber catalysts for total oxidation of co and hydrocarbons in waste gases. Catal. Today 1999, 54, 39–46. [CrossRef] 23. Kiwi-Minsker, L.; Yuranov, I.; Siebenhaar, B.; Renken, A. Glass ﬁber ca hydrocarbons in waste gases. Catal. Today 1999, 54, 39–46. [CrossRef] 24. Tanaka, H.; Kuraoka, K.; Yamanaka, H.; Yazawa, T. Development and disappearance of microporous structure in acid treated e-glass ﬁber. J. Noncryst. Solids 1997, 215, 262–270. [CrossRef] 25. Kitamura, T.; Ino, J.; Masuda, R.; Fukuchi, H.; Tougeda, H.; Nippon Sheet Glass, Co. Ltd.; Nippon Muki, Co. Ltd. References Photocatalyst Supporting Glass Fiber Textile, Manufacturing Method of the Same and Air Filter Apparatus Using the Same. Jpn. Kokai Tokkyo Koho (unexamined patent publication) 2004-002176, 8 January 2004. 26. Aubry, E.; Ghazzal, M.N.; Demange, V.; Chaoui, N.; Robert, D.; Billard, A. Poisoning prevention of TiO2 photocatalyst coatings sputtered on soda-lime glass by intercalation of sinx diffusion barriers. Surf. Coat. Technol. 2007, 201, 7706–7712. [CrossRef] 27. Yu, J.; Zhao, X. Effect of substrates on the photocatalytic activity of nanometer TiO2 thin ﬁlms. Mater. Res. Bull. 2000, 35, 1293–1301. [CrossRef] 28. Rekoske, J.E.; Barteau, M.A. Kinetics and selectivity of 2-propanol conversion on oxidized anatase TiO2. J. Catal. 1997, 165, 57–72. [CrossRef] 29. Hager, S.; Bauer, R. Heterogeneous photocatalytic oxidation of organics for air puriﬁcation by near uv irradiated titanium dioxide. Chemosphere 1999, 38, 1549–1559. [CrossRef] 30. Chang, C.-P.; Chen, J.-N.; Lu, M.-C. Characteristics of photocatalytic oxidation of gaseous 2-propanol usin thin-ﬁlm TiO2 photocatalyst. J. Chem. Technol. Biotechnol. 2004, 79, 1293–1300. [CrossRef] 31. Vildozo, D.; Ferronato, C.; Sleiman, M.; Chovelon, J.-M. Photocatalytic treatment of indoor air: Optimization of 2-propanol removal using a response surface methodology (RSM). Appl. Catal. B Environ. 2010, 94, 303–310. [CrossRef] 32. Kim, S.B.; Hong, S.C. Kinetic study for photocatalytic degradation of volatile organic compounds in air using thin ﬁlm TiO2 photocatalyst. Appl. Catal. B Environ. 2002, 35, 305–315. [CrossRef] 33. Coronado, J.M.; Zorn, M.E.; Tejedor-Tejedor, I.; Anderson, M.A. Photocatalytic oxidation of ketones in the gas phase over TiO2 thin ﬁlms: A kinetic study on the inﬂuence of water vapor. Appl. Catal. B Environ. 2003, 43, 329–344. [CrossRef] 34. Reddy, K.M.; Manorama, S.V.; Reddy, A.R. Bandgap studies on anatase titanium dioxide nanopartic Mater. Chem. Phys. 2002, 78, 239–245. [CrossRef] 35. Kaneko, K. Determination of pore size and pore size distribution. J. Membrane Sci. 1994, 96, 59–89. [CrossRef] 36. Elmer, T.H.; Nordberg, M.E.; Carrier, G.B.; Korda, E.J. Phase separation in borosilicate glasses as seen by electron microscopy and scanning electron microscopy. J. Am. Ceram. Soc. 1970, 53, 171–175. [CrossRef] , ; g, ; , ; , J p g y electron microscopy and scanning electron microscopy. J. Am. Ceram. Soc. 1970, 53, 171–175. [CrossRef] 37. Tanaka, H.; Yazawa, T.; Eguchi, K.; Nagasawa, H.; Matsuda, N.; Einishi, T. Precipitation of colloidal silica and pore size distribution in high silica porous glass. J. Noncryst. Solids 1984, 65, 301–309. [CrossRef] 38. Saito, A.; Poley, H.C. 44. Luo, Y.; Ollis, D.F. Heterogeneous photocatalytic oxidation of trichloroethylene and toluene mixtures in air: Kinetic promotion and inhibition, time-dependent catalyst activity. J. Catal. 1996, 163, 1–11. [CrossRef] 45. Bouazza, N.; Lillo-Ródenas, M.; Linares-Solano, A. Photocatalytic activity of TiO2-based materials for the oxidation of propene and benzene at low concentration in presence of humidity. Appl. Catal. B Environ. 2008, 84, 691–698. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). References Argon porosimetry of selected molesular sieves: Experiments and examination of the adapted Horvath-Kawazoe model. Microporous Mater. 1995, 3, 531–542. [CrossRef] 39. Saito, A.; Poley, H.C. Curvature and parametric sensitivity in models for adsorption in micropores. AICh 1991, 37, 429–436. [CrossRef] 40. Zhuravlev, L.T. The surface chemistry of amorphous silica. Zhuravlev model. Colloids Surf. A Physicochem. Eng. Asp. 2000, 173, 1–38. [CrossRef] 41. Chuiko, A.A.; Lobanov, V.V.; Grebenyuk, A.G. Structure of disperse silica surface and electrostatic aspects of adsorption. In Colloidal Silica: Fundamentals and Applications; Bergna, H.E., Roberts, W.O., Eds.; CRC Press: Boca Raton, FL, USA, 2005; pp. 331–360. ISBN 9780824709679. 42. Wu, D.; Guo, X.; Sun, H.; Navrotsky, A. Energy landscape of water and ethanol on silica surfaces. J. Phys. Chem. C 2015, 119, 15428–15433. [CrossRef] 43. Murakami, Y.; Kenji, E.; Nosaka, A.Y.; Nosaka, Y. Direct detection of oh radicals diffused to the gas phase from the uv-irradiated photocatalytic TiO2 surfaces by means of laser-induced ﬂuorescence spectroscopy. J. Phys. Chem. B 2006, 110, 16808–16811. [CrossRef] 285 Catalysts 2019, 9, 82 p p y y J [ ] 45. Bouazza, N.; Lillo-Ródenas, M.; Linares-Solano, A. Photocatalytic activity of TiO2-based materials for the oxidation of propene and benzene at low concentration in presence of humidity. Appl. Catal. B Environ. 2008, 84, 691–698. [CrossRef] Enhanced Photocatalytic Reduction of Cr(VI) by Combined Magnetic TiO2-Based NFs and Ammonium Oxalate Hole Scavengers Yin-Hsuan Chang 1 and Ming-Chung Wu 1,2,3,* 1 Department of Chemical and Materials Engineering, Chang Gung University, Taoyuan 33302, Taiwan; cgu.yinhsuanchang@gmail.com 1 Department of Chemical and Materials Engineering, Chang Gung University, Taoyuan 33302, Taiwan; cgu.yinhsuanchang@gmail.com cgu.yinhsuanchang@gmail.com 2 Green Technology Research Center, Chang Gung University, Taoyuan 33302, Taiwan 3 Division of Neonatology, Department of Pediatrics, Chang Gung Memorial Hospital, Linkou, Taoyuan 33305, Taiwan * Correspondence: mingchungwu@cgu.edu.tw; Tel.: +886-3211-8800 (ext. 3834) Received: 19 December 2018; Accepted: 5 January 2019; Published: 10 January 2019 * Correspondence: mingchungwu@cgu.edu.tw; Tel.: +886-3211-8800 (ext. 3834) * Correspondence: mingchungwu@cgu.edu.tw; Tel.: +886-3211-8800 (ext. 3834) Received: 19 December 2018; Accepted: 5 January 2019; Published: 10 January 2019 Received: 19 December 2018; Accepted: 5 January 2019; Published: 10 January 2019 Abstract: Heavy metal pollution of wastewater with coexisting organic contaminants has become a serious threat to human survival and development. In particular, hexavalent chromium, which is released into industrial wastewater, is both toxic and carcinogenic. TiO2 photocatalysts have attracted much attention due to their potential photodegradation and photoreduction abilities. Though TiO2 demonstrates high photocatalytic performance, it is a difﬁcult material to recycle after the photocatalytic reaction. Considering the secondary pollution caused by the photocatalysts, in this study we prepared Ag/Fe3O4/TiO2 nanoﬁbers (NFs) that could be magnetically separated using hydrothermal synthesis, which was considered a benign and effective resolution. For the photocatalytic test, the removal of Cr(VI) was carried out by Ag/Fe3O4/TiO2 nanoﬁbers combined with ammonium oxalate (AO). AO acted as a hole scavenger to enhance the electron-hole separation ability, thereby dramatically enhancing the photoreduction efﬁciency of Cr(VI). The reaction rate constant for Ag/Fe3O4/TiO2 NFs in the binary system reached 0.260 min−1, 6.95 times of that of Ag/Fe3O4/TiO2 NFs in a single system (0.038 min−1). The optimized Ag/Fe3O4/TiO2 NFs exhibited high efﬁciency and maintained their photoreduction efﬁciency at 90% with a recyclability of 87% after ﬁve cycles. Hence, taking into account the high magnetic separation behavior, Ag/Fe3O4/TiO2 NFs with a high recycling capability are a potential photocatalyst for wastewater treatment. Keywords: TiO2; magnetic property; photocatalyst; reusable; photoreduction catalysts catalysts www.mdpi.com/journal/catalysts Catalysts 2019, 9, 82 Catalysts 2019, 9, 82 © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 286 Catalysts 2019, 9, 72; doi:10.3390/catal9010072 1. Introduction With the advancement of various industries comes serious industrial water pollution. Such wastewaters usually contain a complicated mixture of constituents, often involving the co-existence of multiple contaminants such as heavy metals and organic pollutants. With the development of electroplating, metallurgy, leathermaking and more, heavy metal pollution has become a serious threat to human survival and development. One such heavy metals released into industrial wastewater is Cr(VI), which is both toxic and carcinogenic. It has been the ﬁrst type of carcinogen listed by the World Health Organization’s International Cancer Research Institute since 2012. Cr(VI) is easily accumulated in living organisms and can result in vomiting, liver damage, and severe diarrhea. Compared to Cr(VI), trivalent chromium (Cr(III)) is less toxic and more vital for animals and humans [1,2]. The conventional approach for the reduction or removal of Cr(VI) includes electrochemical precipitation [3,4], adsorption [5,6], bacterial reduction [7,8], ion exchange [9,10], photoreduction [11–16], etc. Compared to the above methods, photocatalytic reactions are considered Catalysts 2019, 9, 72; doi:10.3390/catal9010072 www.mdpi.com/journal/catalysts 287 Catalysts 2019, 9, 72 a clean and promising technology owing to its highly efﬁcient photoreduction of Cr(VI) to the less harmful Cr(III). TiO2 is a well-known photocatalyst widely applied for environmental puriﬁcation due to its advantages, such as its highly active photocatalytic properties, chemical inertness, environmental-friendliness, non-toxicity, and cost-effectiveness [17–24]. It shows great potential in solving the difﬁcult problem of reducing Cr(VI) to Cr(III) in industrial wastewaters. Though TiO2 demonstrates a high photocatalytic performance, it is difﬁcult to recycle following the photocatalytic reaction. Traditional separation approaches such as ﬁltration and centrifugation have been widely adopted. However, the recycling efﬁciency is hindered by the loss of photocatalysts. Considering the secondary pollution caused by the photocatalysts, combining TiO2 with Fe3O4 to form magnetic composite materials for the magnetic separation under modest magnetic ﬁelds has been seen as a benign and effective resolution [25–28]. To date, there have been many facile methods used to synthesize magnetic iron oxides/TiO2 hybrid nanomaterial such as sol–gel, metal–organic chemical vapor deposition, the seed-mediated method, and hydrothermal treatment. In spite of introducing magnetic separation by doping Fe3O4, the photocatalytic performance could be further enhanced by modifying the shape of Fe3O4 to increase the active surface area [29]. In addition, modifying the structure of TiO2 is also a common method used to enhance photocatalytic performance. 1. Introduction Furthermore, combing the ultraﬁne Fe3O4 with one-dimension TiO2 nanoﬁbers can provide a superior charge transport in a one-dimensional direction, and show high activity. A great deal of literature has indicated that incorporating Fe3O4 into TiO2 does not improve the photocatalytic properties of TiO2 as expected [30–32]. The crystallinity of TiO2 depends on the calcination process, which plays a crucial role in the photocatalytic performance. At the same time, calcination also decreases the saturation magnetization of Fe3O4. With the increasing calcination temperature, Fe3O4, which has a superparamagnetic phase, would undergo a phase transition to γ-Fe2O3 and ﬁnally become α-Fe2O3, which has a soft ferromagnetic phase [33]. Another problem is the small bandgap of Fe3O4, which leads to the fast electron-hole pair recombination in Fe3O4/TiO2 composite material [31]. Therefore, in order to enhance the photocatalytic activity and to maintain the magnetic properties, a lot of research has focused on doping metals to obtain the desired effect [34–38]. In particular, doping Ag into TiO2 not only enhances the separation of electron-hole pairs, but also maintains the magnetic performance of the Fe3O4/TiO2 composite material. For the Ag-doped TiO2, the Ag dopants act as the photo-generated electron trapper that enhances the separation of the electron-hole pair and even creates a local electrical ﬁeld to facilitate electron excitation [39–43]. In this study, in order to achieve both a high photocatalytic activity and a high magnetic property, we prepared Ag and Fe3O4 co-doped TiO2 nanoﬁbers (Ag/Fe3O4/TiO2 NFs) via hydrothermal synthesis followed by a calcination treatment. The Ag/Fe3O4/TiO2 NFs were studied systematically through synchrotron X-ray diffractometer, UV-Vis spectroscopy, ﬁeld emission scanning electron microscopy (FESEM), and transmission electron microscopy (TEM). For the photocatalytic test, the removal of Cr(VI) was carried out by Ag/Fe3O4/TiO2 NFs combined with ammonium oxalate (AO). Hence, taking into account the high magnetic separation behavior, Ag/Fe3O4/TiO2 NFs with a high recycling capability are a potential photocatalyst for wastewater treatment. 2. Results Prior to combining magnetic NPs into TiO2, a basic characterization of Fe3O4 was investigated and summarized in Figure 1. The synchrotron X-ray diffractometer was applied to characterize the crystal structure of the Fe3O4 NPs as shown in Figure 1a. The characteristic peaks could be indexed to standard Fe3O4 (JCPDS No. 019-0629). The method used to determine the bandgap of Fe3O4 NPs from the diffusion reﬂectance is shown schematically in Figure 1b. It was calculated according to [F(R)hv]1/2 versus the energy of incident light based on Kubelka–Munk function spectra, F(R). According to Figure 1b, the band gap of Fe3O4 NPs was ~0.8 eV. The magnetic property of the Fe3O4 NPs were investigated using a (Superconducting quantum interference device magnetometer) SQUID 288 Catalysts 2019, 9, 72 at 10 K. The magnetic hysteresis loop shown in Figure 1c indicates the ferromagnetic property that exists in Fe3O4 NPs. The inset of Figure 1d shows the magnetic separation of Fe3O4 NPs from the aqueous dispersion attracted by the Nd-Fe-B magnets. The collected Fe3O4 NPs indicated that it could be controlled by an applied magnetic ﬁeld. Figure 1. (a) Synchrotron X-ray pattern; (b) Tauc plot for the indirect band gap; (c) magnetic hysteresis loop measured at 10 K of as-synthesized Fe3O4 NPs; and (d) the Fe3O4 suspensions before and after magnetic attraction. Figure 1. (a) Synchrotron X-ray pattern; (b) Tauc plot for the indirect band gap; (c) magnetic hysteresis loop measured at 10 K of as-synthesized Fe3O4 NPs; and (d) the Fe3O4 suspensions before and after magnetic attraction. The calcination temperature for the magnetic material is seen as an important factor. For example, as the calcination temperature exceeds 600 ◦C, it results in a phase transformation from magnetite (Fe3O4) to maghemite (γ-Fe2O3) to hematite (α-Fe2O3) conversions. This phase transformation behavior would cause magnetic material to lose its magnetic properties. In order to maintain the ability to magnetically separate the synthesized TiO2, the calcination temperature was ﬁxed at 550 ◦C for the study. Figure 2a shows the synchrotron X-ray patterns Fe3O4/TiO2 with various doping concentrations that depend on the Fe3O4/TiO2 ratio (wt %). The characteristic peaks, which centered at 2θ around 16.71◦, 24.27◦, 24.81◦, 25.31◦, 31.33◦, 34.99◦and 35.71◦, could be indexed to anatase phase TiO2. Pristine TiO2 exhibited characteristic peaks at 2θ around 18.94◦, 19.75◦and 22.01◦that could be identiﬁed as TiO2 low-temperature phase, β-TiO2 monoclinic. 2. Results With the incorporation of Fe3O4, the anatase phase TiO2 became the only phase in the crystal structure, and no characteristic peaks from other phases could be detected. In addition, the radius of an Fe ion (Fe2+ ~ 0.76 Å, Fe3+ ~ 0.64 Å) is slightly smaller than that of a Ti ion (Ti4+ ~ 0.68 Å), indicating that some of doped Fe ion might enter to interstitial voids of TiO2 lattice [30,44]. The Fe ions in the TiO2 lattice would act as carrier traps, leading to the electron-hole recombination. Taking into account the recombination phenomenon, the photocatalytic performance could be affected when Fe ions were incorporated into the catalyst. When the doping concentration reached 25.0 wt %, both anatase TiO2 and Fe3O4 peaks were detected. The excessive Fe3O4 NPs in the Fe3O4-TiO2 lead to non-uniform doping and to the decrease of the crystallinity of TiO2. From the magnetic hysteresis loop shown in Figure 2b, as the amount of Fe3O4 increased, the magnetization increased as well. In addition, the magnetization was proportional to the doping amount. To optimize the doping concentration, the photocatalytic activity was measured by photoreduction of Cr(VI) in K2Cr2O7 aqueous solution under UV-B irradiation. 289 Catalysts 2019, 9, 72 The photoreduction of Cr(VI) using TiO2-based catalyst usually follows Langmuir–Hinshelwood kinetics. It can be mathematically simpliﬁed to ﬁrst-order kinetics in the early stage described as ln(C0/C) = kt, where C0 is the initial concentration of Cr(VI) in K2Cr2O7, C is the remaining Cr(VI) concentration at various times, k is the apparent reaction rate constant, and t is the photodegradation time. The blank experiment was performed under the same conditions but without the existence of the photocatalyst. For the dark experiment, 10.0 wt %-Fe3O4/TiO2 was also tested in dark conditions to observe the adsorption–desorption behavior. From Figure 2c, 10.0 wt %-Fe3O4/TiO2 calcined at 550 ◦C showed the highest photoreduction performance among other Fe3O4/TiO2 photocatalysts due to the highest crystallinity among the Fe3O4/TiO2 series. High crystallinity can hinder the recombination of photoexcited electrons and holes and thus result in high photocatalytic activity. With further increasing the Fe3O4 doping concentration to 15.0 wt % and 25.0 wt %, the excessive dopant might destroy the lattice of TiO2, thus decreasing the crystallinity of TiO2 dramatically and form the impurity phases composed of Fe3O4, γ-Fe2O3 and α-Fe2O3. 2. Results Although 10.0 mol % Ag/Fe3O4/TiO2 possessed the highest reduction performance, after considering the ability to be magnetically separated, we selected the 5.0 mol % doping level as the optimal photocatalyst. We could also observe that the photoreduction for Ag co-doped with 10 wt % Fe3O4/TiO2 showed the higher performance after the incorporation of Ag compared to 10 wt % Fe3O4/TiO2. This enhancement could be interpreted by the energy level theory, namely that the conduction band of Fe3O4 is lower than the conduction band of TiO2, so the conduction band of TiO2 becomes an electron capture position. With the further introduction of Ag into Fe3O4/TiO2, Ag could act as another electron trap to enhance the electron-hole separation ability [45]. Figure 3. Dependence on the amount of Ag (mol %) co-doped with 10.0 wt % Fe3O4/TiO2. (a) Synchrotron X-ray patterns; (b) magnetic hysteresis loop measured at 10 K; and (c) the C/Co curves for the photoreduction of Cr(VI) in K2Cr2O7 aqueous solution under UV-B irradiation using pristine TiO2 and Ag/Fe3O4/TiO2 with various Ag doping concentration calcined at 550 ◦C. Figure 3. Dependence on the amount of Ag (mol %) co-doped with 10.0 wt % Fe3O4/TiO2. (a) Synchrotron X-ray patterns; (b) magnetic hysteresis loop measured at 10 K; and (c) the C/Co curves for the photoreduction of Cr(VI) in K2Cr2O7 aqueous solution under UV-B irradiation using pristine TiO2 and Ag/Fe3O4/TiO2 with various Ag doping concentration calcined at 550 ◦C. Figure 3. Dependence on the amount of Ag (mol %) co-doped with 10.0 wt % Fe3O4/TiO2. (a) Synchrotron X-ray patterns; (b) magnetic hysteresis loop measured at 10 K; and (c) the C/Co curves for the photoreduction of Cr(VI) in K2Cr2O7 aqueous solution under UV-B irradiation using pristine TiO2 and Ag/Fe3O4/TiO2 with various Ag doping concentration calcined at 550 ◦C. After the optimization process, pristine TiO2, 10.0 wt % Fe3O4/TiO2 (Fe3O4/TiO2) and 5.0 mol % Ag/Fe3O4/TiO2 (Ag/Fe3O4/TiO2) were compared. The FESEM images of TiO2-based NFs before and after combining magnetic NPs and Ag are shown in Figure 4. The image shows that the surface of the pristine TiO2 was very clean and smooth (Figure 4a). When incorporated with Fe3O4 NPs, there was no signiﬁcant morphological change for the Fe3O4/TiO2 (Figure 4b). For Ag/Fe3O4/TiO2, the surface became relatively rough and some particles aggregated on it (Figure 4c). On increasing the silver content, the surface charge of TiO2-based material would gradually decrease. 2. Results In addition, all of the Fe3O4/TiO2 showed poorer performance compared to the pristine TiO2, which is in accordance with the XRD spectra. The Fe3+ as carrier traps leading to recombination phenomenon and decreased the photocatalytic performance compared with pristine TiO2. Figure 2. Dependence on the Fe3O4/TiO2 weight ratio (a) Synchrotron X-ray patterns; (b) magnetic hysteresis loop measured at 10 K; and (c) the C/Co curves for the photoreduction of Cr(VI) in K2Cr2O7 aqueous solution under UV-B irradiation using pristine TiO2 and Fe3O4-TiO2 with various doping concentrations calcined at 550 ◦C. Figure 2. Dependence on the Fe3O4/TiO2 weight ratio (a) Synchrotron X-ray patterns; (b) magnetic hysteresis loop measured at 10 K; and (c) the C/Co curves for the photoreduction of Cr(VI) in K2Cr2O7 aqueous solution under UV-B irradiation using pristine TiO2 and Fe3O4-TiO2 with various doping concentrations calcined at 550 ◦C. Ag was co-doped with 10.0 wt % of Fe3O4 into TiO2 to improve the electron-hole separation further. Figure 3a shows the synchrotron X-ray patterns of the Ag/Fe3O4/TiO2 series with various Ag doping concentrations that depended on the amount of Ag (mol %) co-doped with 10 wt % Fe3O4/TiO2. The characteristic peaks of Ag/Fe3O4/TiO2 could all be assigned to anatase phase TiO2 without any Ag signal. The results indicated that the incorporation of Fe3O4 and Ag did not destroy the crystal structure of TiO2. The magnetic hysteresis loop (Figure 3b) illustrates that as the amount of Ag increased, the magnetization decreased. When the excessive Ag dopant was 10.0 mol %, it resulted in a decay of saturation magnetization compared to Fe3O4/TiO2, due to the contribution of the volume of non-magnetic material to the total sample volume. Therefore, the magnetism of the 10.0 mol % Ag/Fe3O4/TiO2 was too low for magnetic separation by adding a magnetic ﬁeld. Figure 3c demonstrates the C/Co curves for photoreduction of Cr(VI) under UV-B irradiation over the Ag/Fe3O4/TiO2 series with different Ag doping concentrations. The blank experiment was also performed under the same conditions but without the presence of the photocatalyst. For the 290 Catalysts 2019, 9, 72 dark experiment, 5.0 mol % Ag/Fe3O4/TiO2 was also tested in dark conditions to eliminate the adsorption–desorption behavior. The 10.0 mol % Ag/Fe3O4/TiO2 showed the highest photoreduction performance, even higher than that of pristine TiO2. 2. Results Sample Fe/Ti (%) Ag/Ti (%) Pristine TIO2 0.0 0.0 Fe3O4/TiO2 2.9 0.0 Ag/Fe3O4/ TiO2 3.1 0.4 Figure 4. SEM images of (a) pristine TiO2; (b) Fe3O4/TiO2 and (c) Ag/Fe3O4/ TiO2. The Kubelka–Munk function spectra of TiO2-based materials are shown in Figure 5a. Pristine TiO2 Figure 4. SEM images of (a) pristine TiO2; (b) Fe3O4/TiO2 and (c) Ag/Fe3O4/ TiO2. Figure 4. SEM images of (a) pristine TiO2; (b) Fe3O4/TiO2 and (c) Ag/Fe3O4/ TiO2. The Kubelka–Munk function spectra of TiO2-based materials are shown in Figure 5a. Pristine TiO2 only showed absorption behavior in the UV range. However, compared to pristine TiO2, the F(R) spectra of Fe3O4/TiO2 and Ag/Fe3O4/TiO2 showed an obvious extension to the visible light region, and the band gap energy also decreased from 3.1 eV to 2.1 eV and 2.0 eV, respectively (Figure 5b). This could be ascribed to the introduction of Fe3O4. During the calcination process, the introduced Fe3+ could exchange with the lattice position of Ti4+ and therefore form an impurity band. Fe3O4/TiO2 and Ag/Fe3O4/TiO2 with a decreased forbidden bandwidth could successfully narrow the band gap for the higher absorption behavior in the visible region. This enhanced absorption behavior could generate a lot of photo-excited electrons and holes for photocatalytic reactions. The Kubelka–Munk function spectra of TiO2-based materials are shown in Figure 5a. Pristine TiO2 only showed absorption behavior in the UV range. However, compared to pristine TiO2, the F(R) spectra of Fe3O4/TiO2 and Ag/Fe3O4/TiO2 showed an obvious extension to the visible light region, and the band gap energy also decreased from 3.1 eV to 2.1 eV and 2.0 eV, respectively (Figure 5b). This could be ascribed to the introduction of Fe3O4. During the calcination process, the introduced Fe3+ could exchange with the lattice position of Ti4+ and therefore form an impurity band. Fe3O4/TiO2 and Ag/Fe3O4/TiO2 with a decreased forbidden bandwidth could successfully narrow the band gap for the higher absorption behavior in the visible region. This enhanced absorption behavior could generate a lot of photo-excited electrons and holes for photocatalytic reactions. Figure 5. (a) Kubelka–Munk function spectra and (b) Tauc plot for the indirect band gap of pristine TiO2, Fe3O4/TiO2 and Ag/Fe3O4/TiO2. Figure 5. (a) Kubelka–Munk function spectra and (b) Tauc plot for the indirect band gap of pristine TiO2, Fe3O4/TiO2 and Ag/Fe3O4/TiO2. The photocatalytic activity test was examined by photoreduction of Cr(VI) to Cr(III). 2. Results With small amounts of Ag dopant, Ag2O and AgO might disperse on the surface of TiO2-based material. When increasing Ag doping concentration, the decrease in surface charge was attributed to an agglomeration of the silver species and a reduction to Ag0 on the TiO2 surface [46]. The EDS-characterized elemental compositions and the corresponding results are listed in Table 1. For Fe3O4/TiO2, the ratio of Fe/Ti and Ag/Ti were ~2.9% and ~0.0%, respectively. After incorporating Ag, the ratio of Fe/Ti was ~3.1%, which was approximately the same as Fe3O4/TiO2, and the ratio of Ag/Ti increased to 0.4%. The corresponding ratios of Fe/Ti and Ag/Ti illustrated the existence of Ag in the Ag/Fe3O4/TiO2, together with the leading component Ti and Fe. The distinct signals of these elements present in the spectrum conﬁrmed the successful inclusion of Ag ions into the host TiO2 lattice. 291 Catalysts 2019, 9, 72 Table 1. The corresponding ratios of Fe/Ti and Ag/Ti for pristine TiO2, Fe3O4/TiO2 and Ag/Fe3O4/ TiO2. Table 1. The corresponding ratios of Fe/Ti and Ag/Ti for pristine TiO2, Fe3O4/TiO2 and Ag/Fe3O4/ TiO2. Sample Fe/Ti (%) Ag/Ti (%) Pristine TIO2 0.0 0.0 Fe3O4/TiO2 2.9 0.0 Ag/Fe3O4/ TiO2 3.1 0.4 Figure 4. SEM images of (a) pristine TiO2; (b) Fe3O4/TiO2 and (c) Ag/Fe3O4/ TiO2. Table 1. The corresponding ratios of Fe/Ti and Ag/Ti for pristine TiO2, Fe3O4/TiO2 and Ag/Fe3O4/ TiO2. Sample Fe/Ti (%) Ag/Ti (%) Pristine TIO2 0.0 0.0 Fe3O4/TiO2 2.9 0.0 Ag/Fe3O4/ TiO2 3.1 0.4 Figure 4. SEM images of (a) pristine TiO2; (b) Fe3O4/TiO2 and (c) Ag/Fe3O4/ TiO2. The Kubelka–Munk function spectra of TiO2-based materials are shown in Figure 5a. Pristine TiO2 only showed absorption behavior in the UV range. However, compared to pristine TiO2, the F(R) spectra of Fe3O4/TiO2 and Ag/Fe3O4/TiO2 showed an obvious extension to the visible light region, and the band gap energy also decreased from 3.1 eV to 2.1 eV and 2.0 eV, respectively (Figure 5b). This could be ascribed to the introduction of Fe3O4. During the calcination process, the introduced Fe3+ could exchange with the lattice position of Ti4+ and therefore form an impurity band. Fe3O4/TiO2 and Ag/Fe3O4/TiO2 with a decreased forbidden bandwidth could successfully narrow the band gap for the higher absorption behavior in the visible region. This enhanced absorption behavior could generate a lot of photo-excited electrons and holes for photocatalytic reactions. 2. Results The photoreduction pathways of Cr(VI) on the surface of TiO2 through UV irradiation can be described by the following reaction sequence (Equations (1)–(6)). After UV light irradiation, photo-excited electron-hole pairs are generated. During the photoreduction reaction of Cr(VI), electrons dominate the entire reaction. Meanwhile, the hole will oxidize H2O to form the reactive oxygen species OH, which will further react with Cr(III) to generate Cr(VI). TiO2 + hv →h+ + e− (1) Cr2O72−+ 14H+ + 6e−→2Cr3+ + 7H2O (2) (1) Cr2O72−+ 14H+ + 6e−→2Cr3+ + 7H2O Cr2O72−+ 14H+ + 6e−→2Cr3+ + 7H2O (2) (2) 292 Catalysts 2019, 9, 72 e−+ h+ →recombination (3) h+ + H2O →·OH + H+ (4) h+ + OH−→·OH (5) 3·OH + Cr3+ →3OH−+ Cr6+ (6) (6) It is unfavorable to reduce Cr(VI) to Cr(III) while Cr(VI) participates in the reaction alone, due to the electron-hole recombination and the oxidation of Cr(III). Figure 6a shows the photoreduction of Cr(VI) over pristine TiO2, Fe3O4/TiO2 and Ag/Fe3O4/TiO2 in a single system for which only Cr(VI) existed in the initial condition. The reduction of Cr(VI) was greatly promoted by the coexistence of ammonium oxalate (AO), and the corresponding results for single systems are also plotted for comparison (Figure 6b). AO is a type of hole scavenger that is widely used for detecting reactive oxygen species during the photocatalytic reaction in order to better understand the reaction mechanism. Therefore, AO would capture the photogenerated holes during the photocatalysis reaction, leaving the photogenerated electrons on the surface of the TiO2-based NFs. With the help of AO, the separation of the electron-hole was greatly facilitated and thus the reduction performance of Cr(VI) was enhanced. The poor enhancement of pristine TiO2 compared with Fe3O4/TiO2 and Ag/Fe3O4/TiO2 could be due to the bandgap of each sample. A decrease in the bandgap for Fe3O4/TiO2 and Ag/Fe3O4/TiO2 resulted in a greater absorption of photons, which was beneﬁcial for the production of electrons and holes required for the photocatalytic reactions. However, the photoexcited electron-hole pair in the Fe3O4/TiO2 and Ag/Fe3O4/TiO2 favored a transfer to Fe3O4. Holes can provide a faster reaction route with AO, rather than recombining with the electron. Further, the residual electron on the surface of Fe3O4/TiO2 and Ag/Fe3O4/TiO2 can reduce Cr(VI) to Cr(III). Therefore, the photoreduction performance for Fe3O4/TiO2 and Ag/Fe3O4/TiO2 showed a dramatic enhancement. 3.2. Synthesis of Ag/Fe3O4/TiO2 NFs The TiO2-based NFs were synthesized by hydrothermal method and crystallized by heat treatment. First, 2.5 g anatase phase TiO2 powder (98%, Sigma-Aldrich, St. Louis, MO, USA), as-synthesized Fe3O4 NPs, and silver nitrate (AgNO3, extra pure, Choneye, Taipei, Taiwan) with various stoichiometric ratios were suspended into separate 62.5 mL of 10 M NaOH. The suspension was dispersed uniformly into an ultrasonic bath. After that, the reactants were transferred into a polytetraﬂuoroethylene-lined autoclave for thermal treatment at 150 ◦C for 24 h to obtain sodium titanate (Na2Ti3O7). Then, various forms of Na2Ti3O7 were washed with 0.10 M hydrochloric acid (HCl, 37%, Sigma-Aldrich, St. Louis, MO, USA) to exchange the sodium ion for protons. Finally, the sodium hydrogen titanate (NaxH2−xTi3O7) was ﬁltered and air dried at 80 ◦C. The dried NaxH2−xTi3O7 was calcined at 550 ◦C for 12 h at a 5 ◦C/min heating rate to obtain magnetic TiO2-based NFs. 2. Results The reaction rate constant for Ag/Fe3O4/TiO2 in binary system achieved 0.260 min−1, which was 6.95 times that of Ag/Fe3O4/TiO2 in a single system at 0.038 min−1. These results conﬁrmed the synergetic promotion effect of ammonium oxalate. Figure 6. Photocatalytic reaction in (a) Cr(VI) single system and (b) Cr(VI) + AO binary system with pristine TiO2, Fe3O4/TiO2, and Ag/Fe3O4/TiO2. The stability and recyclability of the photocatalyst is an important index for practical application. n order to examine the stability and recyclability of Ag/Fe3O4/TiO2, the photoreduction of Cr(VI) as repeated ﬁve times. Each time, the photocatalysts were recycled by adding a magnetic ﬁeld. his exhibited a slight decay of reduction efﬁciency after each cycle, which accounted for the weight ss during every recycle process. After ﬁve cycles, the photoreduction efﬁciency was maintained at 0% (Figure 7a), and the amount of the remaining photocatalyst was 87% (Figure 7b). The stability Figure 6. Photocatalytic reaction in (a) Cr(VI) single system and (b) Cr(VI) + AO binary system with pristine TiO2, Fe3O4/TiO2, and Ag/Fe3O4/TiO2. Figure 6. Photocatalytic reaction in (a) Cr(VI) single system and (b) Cr(VI) + AO binary system with pristine TiO2, Fe3O4/TiO2, and Ag/Fe3O4/TiO2. The stability and recyclability of the photocatalyst is an important index for practical application. In order to examine the stability and recyclability of Ag/Fe3O4/TiO2, the photoreduction of Cr(VI) was repeated ﬁve times. Each time, the photocatalysts were recycled by adding a magnetic ﬁeld. This exhibited a slight decay of reduction efﬁciency after each cycle, which accounted for the weight loss during every recycle process. After ﬁve cycles, the photoreduction efﬁciency was maintained at 90% (Figure 7a), and the amount of the remaining photocatalyst was 87% (Figure 7b). The stability 293 Catalysts 2019, 9, 72 and recyclability tests proved that the Cr(VI) photoreduction efﬁciency over Ag/Fe3O4/TiO2 has consistently high stability and recyclability. Therefore, Ag/Fe3O4/TiO2 is a potential photocatalyst for wastewater treatment. Figure 7. (a) Stability and (b) recyclability test of Ag/Fe3O4/TiO2 for the photocatalytic reduction of Cr(VI) over ﬁve cycles. Figure 7. (a) Stability and (b) recyclability test of Ag/Fe3O4/TiO2 for the photocatalytic reduction of Cr(VI) over ﬁve cycles. 3.1. Synthesis of Fe3O4 Magnetic NPs The synthesis of Fe3O4 NPs was carried out by the co-precipitation method, in which the iron(II) chloride (FeCl2·4H2O, Acros, 99+%) and iron(III) chloride ((FeCl3·6H2O, Acros, 99+%) were used as the raw materials with a molar proportion of 1:2. First, they were dissolved in deionized water and preheated to 60 ◦C. After that, a 10 M sodium hydroxide aqueous solution (NaOH) acting as a precipitation reagent was added into the mixture solution under continuous stirring for 1 h. The Fe3O4 suspension was magnetically separated and washed with deionized water repeatedly until the pH was 7. Finally, the product was air dried at 60 ◦C. 3.4. Photocatalytic Measurement For the measurement of the photoreduction of Cr(VI), 20.0 mg of magnetic TiO2-based photocatalyst was dispersed into 150.0 mL of potassium dichromate (K2Cr2O7, 0.0167 M, Fisher Scientiﬁc, CA, USA) with an initial concentration of 1.0 ppm at ambient conditions. As the control group, 20.0 mg of pristine TiO2 was also dispersed into 150.0 mL K2Cr2O7 with an initial concentration of 1.0 ppm at ambient conditions. The two UV-B light lamps (G15T8E, λmax ~312 nm, 8.0 W, Sankyo Denki, Osaka, Japan) were placed ~10.0 cm above the reaction system. Before exposure to light irradiation, the suspensions were put in the dark for 30 min in order to achieve the adsorption equilibrium and thus minimize the surface adsorption behavior. The concentration of retained Cr(VI) was measured by the diphenylcarbazide method. By comparing the intensity of the Cr(VI) characteristic peak located at λ = 540 nm with the calibration curve examined previously, we can obtain its corresponding concentration. In order to examine the mechanism of Cr(VI) photoreduction, 142.2 μL tert-butanol ((CH3)3COH, ≥99.0%, J.T.Baker, Phillipsburg, NJ, USA) and 24.0 mg ammonium oxalate (C2H8N2O4, 98%, Vetec, trademark of Sigma-Aldrich, St. Louis, MO, USA) were added into K2Cr2O7 in the beginning, respectively. The stability and recyclability of the photocatalysts were measured by cycling experiments. After the Cr(VI) photoreduction for each cycle, the magnetic TiO2 was collected by Nd-Fe-B magnet wrapped with PVC ﬁlm. After removing the magnetic ﬁeld, the magnetic TiO2 was washed three times with ethanol to remove residual ions and molecules and then dried at 80 ◦C. The fresh 1.0 ppm K2Cr2O7 aqueous solution was mixed with the used photocatalyst to perform the second run of photoactivity testing. Similarly, the photocatalyst was recycled to perform the third, fourth, and ﬁfth tests. 4. Conclusions In this study, we successfully synthesized Ag and Fe3O4 co-doped TiO2 NFs using hydrothermal synthesis followed by thermal treatment in order to achieve high photocatalytic performance and a feasible recycle process. The synthesized Ag/Fe3O4/TiO2 exhibited a relatively narrower band gap (2.0 eV) than that of pristine TiO2 (3.1 eV). For the photoreduction of Cr(VI), electrons dominated the photoreduction efﬁciency. The photocatalytic process paired with ammonium oxalate could greatly facilitate the separation of electron-hole pairs and thus enhance the reduction rate of Cr(VI). After ﬁve cycles of the stability and recyclability test, the photoreduction efﬁciency was maintained at 90%, and the amount of remaining photocatalyst was maintained at 87%. Consequently, taking into account the high magnetic separation behavior and the high stability, Ag/Fe3O4/TiO2 showed great potential to be used for practical wastewater treatment. Author Contributions: Y.-H.C. performed the research and analyzed the data; Y.-H.C. and M.-C.W. wrote paper; M.-C.W. was the supervisor and revised the paper. All authors read and approved the ﬁnal manuscrip Funding: This research was funded by the Ministry of Science and Technology, Taiwan (MOST 106-2221-E-182-057-MY3 and MSOT 107-2119-M-002-012), Green Technology Research Center, Chang Gung University (QZRPD181) and Chang Gung Memorial Hospital, Linkou (BMRPC74 and CMRPD2H0171). Acknowledgments: The authors appreciate Wei-Fang Su at National Taiwan University and the Ming-Tao Lee group (TLS BL13A1) at National Synchrotron Radiation Research Center for useful discussion. Conﬂicts of Interest: The authors declare no conﬂict of interest. 3.3. Characterization To observe the crystal structure, the synchrotron X-ray spectra were collected from 5◦to 45◦ of 2θ with a scan rate of 0.02◦/s and a wavelength of ~ 1.025 Å. The Kubelka–Munk function, F(R), spectra were measured and recorded by UV/Vis spectrophotometer (Jacso, V-650, Tokyo, Japan) from 200 to 900 nm wavelength. The magnetic properties of Fe3O4 NPs and magnetic TiO2-based NFs were measured at 10 K temperature using a SQUID magnetometer (MPMS3, Quantum Design, San Diego, CA, USA). The microstructure was characterized by transmission electron microscopy (TEM, spherical-aberration corrected ULTRA-HRTEM, JEM-ARM200FTH, JEOL Ltd., Tokyo, Japan). The morphology and atomic ratio of TiO2-based NFs were measured by FE-SEM (SU8010, Hitachi, Tokyo, Japan) equipped with EDS (XFlash Detector 5030, Bruker AXS, Karlsruhe, Germany). 294 Catalysts 2019, 9, 72 Catalysts 2019, 9, 72 2. Rowbotham, A.L.; Levy, L.S.; Shuker, L.K. Chromium in the Environment: An Evaluation of Exposure of the UK General Population and Possible Adverse Health Effects. J. Toxicol. Environ. Health 2000, 3, 145–178. [CrossRef] Funding: This research was funded by the Ministry of Science and Technology, Taiwan (MOST 106-2221-E-182-057-MY3 and MSOT 107-2119-M-002-012), Green Technology Research Center, Chang Gung University (QZRPD181) and Chang Gung Memorial Hospital, Linkou (BMRPC74 and CMRPD2H0171). 1. Costa, M. Toxicity and Carcinogenicity of Cr(VI) in Animal Models and Humans. Crit. Rev. Toxicol. 1997, 27, 431–442. [CrossRef] [PubMed] Acknowledgments: The authors appreciate Wei-Fang Su at National Taiwan University and the Ming-Tao Lee group (TLS BL13A1) at National Synchrotron Radiation Research Center for useful discussion. References Eng. J. 2017, 314, 622–630. [CrossRef] g g g 2 g y Enhanced Photocatalytic Activity for Cr(VI) Removal under Visible Light. Chem. Eng. J. 2017, 314, 622–630. [CrossRef] 14. Li, Y.; Liu, Z.; Wu, Y.; Chen, J.; Zhao, J.; Jin, F.; Na, P. Carbon Dots-TiO2 Nanosheets Composites for Photoreduction of Cr(VI) under Sunlight Illumination: Favorable Role of Carbon Dots. Appl. Catal. B 2018, 224, 508–517. [CrossRef] 15. Wang, W.; Lai, M.; Fang, J.; Lu, C. Au and Pt Selectively Deposited on {001}-Faceted TiO2 toward SPR Enhanced Photocatalytic Cr(VI) Reduction: The Inﬂuence of Excitation Wavelength. Appl. Surf. Sci. 2018, 439, 430–438. [CrossRef] 16. Ngo, A.; Nguyen, H.; Hollmann, D. Criticial Assessment of the Photocatalytic Reduction of Cr(VI) over Au/TiO2. Catalysts 2018, 8, 606. [CrossRef] 17. Tan, L.-L.; Ong, W.-J.; Chai, S.-P.; Mohamed, A.R. Visible-Light-Activated Oxygen-Rich TiO2 as Next Generation Photocatalyst: Importance of Annealing Temperature on the Photoactivity toward Reduction of Carbon Dioxide. Chem. Eng. J. 2016, 283, 1254–1263. [CrossRef] Busiakiewicz, A.; Kisielewska, A.; Piwo´nski, I.; Batory, D. The Effect of Fe Segregation on the Photocatalytic Growth of Ag Nanoparticles on Rutile TiO (001) Appl Surf Sci 2017 401 378 384 [CrossRef] 18. Busiakiewicz, A.; Kisielewska, A.; Piwo´nski, I.; Batory, D. The Effect of Fe Segregation on the Photocatalytic Growth of Ag Nanoparticles on Rutile TiO2 (001). Appl. Surf. Sci. 2017, 401, 378–384. [CrossRef] 19. Low, J.; Cheng, B.; Yu, J. Surface Modiﬁcation and Enhanced Photocatalytic CO2 Reduction Perform TiO2: A Review. Appl. Surf. Sci. 2017, 392, 658–686. [CrossRef] 20. Tan, L.-L.; Ong, W.-J.; Chai, S.-P.; Mohamed, A.R. Photocatalytic Reduction of CO2 with H2O over Graphene Oxide-Supported Oxygen-Rich TiO2 Hybrid Photocatalyst under Visible Light Irradiation: Process and Kinetic Studies. Chem. Eng. J. 2017, 308, 248–255. [CrossRef] 21. Zhang, Y.; Gu, D.; Zhu, L.; Wang, B. Highly Ordered Fe3+/TiO2 Nanotube Arrays for Efﬁcient Photocataltyic Degradation of Nitrobenzene. Appl. Surf. Sci. 2017, 420, 896–904. [CrossRef] 22. Duan, Y.; Liang, L.; Lv, K.; Li, Q.; Li, M. TiO2 Faceted Nanocrystals on the Nanoﬁbers: Homojunction Ti Based Z-Scheme Photocatalyst for Air Puriﬁcation. Appl. Surf. Sci. 2018, 456, 817–826. [CrossRef] , ; g, ; , ; , Q ; , 2 y j Based Z-Scheme Photocatalyst for Air Puriﬁcation. Appl. Surf. Sci. 2018, 456, 817–826. [CrossRef] 23. Makal, P.; Das, D. Self-Doped TiO2 Nanowires in TiO2-B Single Phase, TiO2-B/Anatase and TiO2-Anatase/Rutile Heterojunctions Demonstrating Individual Superiority in Photocatalytic Activity under Visible and UV Light. Appl. Surf. Sci. References 295 Catalysts 2019, 9, 72 3. Kongsricharoern, N.; Polprasert, C. Electrochemical Precipitation of Chromium (Cr6+) from an Electroplating Wastewater. Water Res. 1995, 31, 109–117. [CrossRef] 4. Hunsom, M.; Pruksathorn, K.; Damronglerd, S.; Vergnes, H.; Duverneuil, P. Electrochemical Treatment of H M l (C 2+ C 6+ Ni2+) f I d i l Efﬂ d M d li f C R d i W R 4. Hunsom, M.; Pruksathorn, K.; Damronglerd, S.; Vergnes, H.; Duverneuil, P. Electrochemical Treatment of Heavy Metals (Cu2+, Cr6+, Ni2+) from Industrial Efﬂuent and Modeling of Copper Reduction. Water Res. 2005, 39, 610–616. [CrossRef] [PubMed] 5. Etemadi, M.; Samadi, S.; Yazd, S.S.; Jafari, P.; Youseﬁ, N.; Aliabadi, M. Selective Adsorption of Cr(VI) Ions from Aqueous Solutions Using Cr6+-Imprinted Pebax/Chitosan/GO/Aptes Nanoﬁbrous Adsorbent. Int. J. Biol. Macromol. 2017, 95, 725–733. [CrossRef] [PubMed] 6. Yogeshwaran, V.; Priya, A.K. Removal of Hexavalent Chromium (Cr6+) Using Different Natural Adsorbents—A Review. J. Chromatogr. Sep. Tech. 2017, 8, 1000392. [CrossRef] 7. Ohtake, H. New Biological Method for Detoxiﬁcation and Removal of Hexavalent Chromium. Water Res. 1992, 25, 395–402. [CrossRef] 8. Bennett, R.M.; Cordero, P.R.F.; Bautista, G.S.; Dedeles, G.R. Reduction of Hexavalent Chromium Using Fungi and Bacteria Isolated from Contaminated Soil and Water Samples. Chem. Ecol. 2013, 29, 320–328. [CrossRef] 8. Bennett, R.M.; Cordero, P.R.F.; Bautista, G.S.; Dedeles, G.R. Reduction of Hexavalent Chromium Using Fungi and Bacteria Isolated from Contaminated Soil and Water Samples. Chem. Ecol. 2013, 29, 320–328. [CrossRef] 9. Xing, Y.; Chen, X.; Wang, D. Electrically Regenerated Ion Exchange for Removal and Recovery of Cr(VI) from Wastewater. Environ. Sci. Technol. 2007, 41, 1439–1443. [CrossRef] 10. Mekatel, H.; Amokrane, S.; Benturki, A.; Nibou, D. Treatment of Polluted Aqueous Solutions by Ni2+, Pb2+, Z Cr+6,Cd+2 and Co+2 Ions by Ion Exchange Process Using Faujasite Zeolite. Procedia Eng. 2012, 33, 52–57. [CrossR 11. Cai, J.; Wu, X.; Zheng, F.; Li, S.; Wu, Y.; Lin, Y.; Lin, L.; Liu, B.; Chen, Q.; Lin, L. Inﬂuence of TiO2 Hollow Sphere Size on Its Photo-Reduction Activity for Toxic Cr(VI) Removal. J. Colloid Interface Sci. 2017, 490, 37–45. [CrossRef] [PubMed] 12. Lu, D.; Yang, M.; Fang, P.; Li, C.; Jiang, L. Enhanced Photocatalytic Degradation of Aqueous Phenol and Cr(VI) over Visible-Light-Driven Tbxoy Loaded TiO2-Oriented Nanosheets. Appl. Surf. Sci. 2017, 399, 167–184. [CrossRef] 13. Wang, L.; Zhang, C.; Gao, F.; Mailhot, G.; Pan, G. Algae Decorated TiO2/Ag Hybrid Nanofiber Membrane with Enhanced Photocatalytic Activity for Cr(VI) Removal under Visible Light. Chem. References 2018, 455, 1106–1115. [CrossRef] 24. Wu, M.-C.; Hsiao, K.-C.; Chang, Y.-H.; Chan, S.-H. Photocatalytic Hydrogen Evolution of Palladium Nanoparticles Decorated Black TiO2 Calcined in Argon Atmosphere. Appl. Surf. Sci. 2018, 430, 407–414. [CrossRef] 296 Catalysts 2019, 9, 72 25. Jing, J.; Li, J.; Feng, J.; Li, W.; Yu, W.W. Photodegradation of Quinoline in Water over Magnetically Separable Fe3O4/TiO2 Composite Photocatalysts. Chem. Eng. J. 2013, 219, 355–360. [CrossRef] p y g 26. Li, Z.-J.; Huang, Z.-W.; Guo, W.-L.; Wang, L.; Zheng, L.-R.; Chai, Z.-F.; Shi, W.-Q. Enhanced Photocatalytic Removal of Uranium(VI) from Aqueous Solution by Magnetic TiO2/Fe3O4 and Its Graphene Composite. 26. Li, Z.-J.; Huang, Z.-W.; Guo, W.-L.; Wang, L.; Zheng, L.-R.; Chai, Z.-F.; Shi, W.-Q. Enhanced Photocatalytic Removal of Uranium(VI) from Aqueous Solution by Magnetic TiO2/Fe3O4 and Its Graphene Composite. Environ. Sci. Technol. 2017, 51, 5666–5674. [CrossRef] [PubMed] 27. Yu, Y.; Yan, L.; Cheng, J.; Jing, C. Mechanistic Insights into TiO2 Thickness in Fe3O4@TiO2-GO Composites for Enroﬂoxacin Photodegradation. Chem. Eng. J. 2017, 325, 647–654. [CrossRef] 28. Liu, M.-C.; Liu, B.; Sun, X.-Y.; Lin, H.-C.; Lu, J.-Z.; Jin, S.-F.; Yan, S.-Q.; Li, Y.-Y.; Zhao, P. Core/Shell Structured Fe3O4@TiO2-DNM Nanospheres as Multifunctional Anticancer Platform: Chemotherapy and Photodynamic Therapy Research. J. Nanosci. Nanotechnol. 2018, 18, 4445–4456. [CrossRef] [PubMed] 29. Wu, W.; Xiao, X.; Zhang, S.; Ren, F.; Jiang, C. Facile Method to Synthesize Magnetic Iron Oxides/TiO2 Hybrid Nanoparticles and Their Photodegradation Application of Methylene Blue. Nanoscale Res. Lett. 2011, 6, 533. [CrossRef] 30. Shojaei, A.F.; Shams-Nateri, A.; Ghomashpasand, M. Magnetically Recyclable Fe3+/TiO2@Fe3O4Nanocomposites Towards Degradation of Direct Blue 71 under Visible-Light Irradiation. IET Micro Nano Lett. 2017, 12, 161–165. [CrossRef] 31. Jia, X.; Dai, R.; Lian, D.; Han, S.; Wu, X.; Song, H. Facile Synthesis and Enhanced Magnetic, Photocatalytic Properties of One-Dimensional Ag@Fe3O4-TiO2. Appl. Surf. Sci. 2017, 392, 268–276. [CrossRef] 32. Wang, H.; Fei, X.; Wang, L.; Li, Y.; Xu, S.; Sun, M.; Sun, L.; Zhang, C.; Li, Y.; Yang, Q.; et al. Magnetically Separable Iron Oxide Nanostructures-TiO2 Nanoﬁbers Hierarchical Heterostructures: Controlled Fabrication and Photocatalytic Activity. New J. Chem. 2011, 35, 1795–1802. [CrossRef] g g g g g y Separable Iron Oxide Nanostructures-TiO2 Nanoﬁbers Hierarchical Heterostructures: Controlled Fabrication and Photocatalytic Activity. New J. Chem. 2011, 35, 1795–1802. [CrossRef] 33. Jafari, A.; Shayesteh, S.F.; Salouti, M.; Boustani, K. Effect of Annealing Temperature on Magnetic Phase Transition in Fe3O4 Nanoparticles. J. Magn. Magn. Mater. 2015, 379, 305–312. [CrossRef] 34. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). p g p 46. Mogal, S.; Gandhi, V.G.; Mishra, M.; Tripathi, S.; Shripathi, T.; Joshi, P.; Shah, D. Single-Step Synthesis of Silver-Doped Titanium Dioxide: Inﬂuence of Silver on Structural, Textural, and Photocatalytic Properties. Ind. Eng. Chem. Res. 2014, 53, 5749–5758. [CrossRef] 45. Zhan, J.; Zhang, H.; Zhu, G. Magnetic Photocatalysts of Cenospheres Coated with Fe3O4/TiO2 Core/Shell Nanoparticles Decorated with Ag Nanopartilces. Ceram. Int. 2014, 40, 8547–8559. [CrossRef] Photocatalytic Hydrogen Production Under Near-UV Using Pd-Doped Mesoporous TiO2 and Ethanol as Organic Scavenger Bianca Rusinque, Salvador Escobedo and Hugo de Lasa Bianca Rusinque, Salvador Escobedo and Hugo de Lasa * Chemical Reactor Engineering Centre (CREC), Faculty of Engineering, Western University, London, ON N6A 5B9, Canada; brusinqu@uwo.ca (B.R.); selfa.iq@gmail.com (S.E.) * Correspondence: hdelasa@uwo.ca; Tel.: +1-519-661-2149 Received: 20 November 2018; Accepted: 21 December 2018; Published: 2 January 2019 Bianca Rusinque, Salvador Escobedo and Hugo de Lasa * Chemical Reactor Engineering Centre (CREC), Faculty of Engineering, Western University, London, ON N6A 5B9, Canada; brusinqu@uwo.ca (B.R.); selfa.iq@gmail.com (S.E.) * Correspondence: hdelasa@uwo.ca; Tel.: +1-519-661-2149 Recei ed 20 No ember 2018 Accepted 21 December 2018 Published 2 Januar 2019 Chemical Reactor Engineering Centre (CREC), Faculty of Engineering, Western University, London, ON N6A 5B9, Canada; brusinqu@uwo.ca (B.R.); selfa.iq@gmail.com (S.E.) * Correspondence: hdelasa@uwo.ca; Tel.: +1-519-661-2149 Received: 20 November 2018; Accepted: 21 December 2018; Published: 2 January 2019 Abstract: Photocatalysis can be used advantageously for hydrogen production using a light source (near-UV light), a noble metal-doped semiconductor and an organic scavenger (2.0 v/v% ethanol). With this end, palladium was doped on TiO2 photocatalysts at different metal loadings (0.25 to 5.00 wt%). Photocatalysts were synthetized using a sol-gel method enhancing morphological properties with a soft template precursor. Experiments were carried out in the Photo-CREC Water II reactor system developed at CREC-UWO (Chemical Reactor Engineering Centre- The University of Western Ontario) Canada. This novel unit offers hydrogen storage and symmetrical irradiation allowing precise irradiation measurements for macroscopic energy balances. Hydrogen production rates followed in all cases a zero-order reaction, with quantum yields as high as 30.8%. Keywords: Photo-CREC Water II reactor; Palladium; TiO2; Hydrogen production; Quantum Yiel Catalysts 2019, 9, 33; doi:10.3390/catal9010033 www.mdpi.com/journal/catalysts catalysts catalysts References Wang, W.-K.; Chen, J.-J.; Gao, M.; Huang, Y.-X.; Zhang, X.; Yu, H.-Q. Photocatalytic Degradation of Atrazine by Boron-Doped TiO2 with a Tunable Rutile/Anatase Ratio. Appl. Catal. B 2016, 195, 69–76. [CrossRef] 35. Rossi, G.; Pasquini, L.; Catone, D.; Piccioni, A.; Patelli, N.; Paladini, A.; Molinari, A.; Caramori, S.; O’Keeffe, P.; Boscherini, F. Charge Carrier Dynamics and Visible Light Photocatalysis in Vanadium-Doped TiO2 Nanoparticles. Appl. Catal. B 2018, 237, 603–612. [CrossRef] p pp 36. Wu, M.-C.; Wu, P.-Y.; Lin, T.-H.; Lin, T.-F. Photocatalytic Performance of Cu-Doped TiO2 Nanoﬁbers Treated by the Hydrothermal Synthesis and Air-Thermal Treatment. Appl. Surf. Sci. 2018, 430, 390–398. [CrossRef] 37. Avilés-García, O.; Espino-Valencia, J.; Romero-Romero, R.; Rico-Cerda, J.; Arroyo-Albiter, M.; Solís-Casados, D.; Natividad-Rangel, R. Enhanced Photocatalytic Activity of Titania by Co-Doping with Mo and W. Catalysts 2018, 8, 631. [CrossRef] 38. Ramírez-Sánchez, I.; Bandala, E. Photocatalytic Degradation of Estriol Using Iron-Doped TiO2 under High and Low UV Irradiation. Catalysts 2018, 8, 625. [CrossRef] 39. Ali, T.; Ahmed, A.; Alam, U.; Uddin, I.; Tripathi, P.; Muneer, M. Enhanced Photocatalytic and Antibacterial Activities of Ag-Doped TiO2 Nanoparticles under Visible Light. Mater. Chem. Phys 2018, 212, 325–335. [CrossRef] 40. Mandari, K.K.; Do, J.Y.; Police, A.K.R.; Kang, M. Natural Solar Light-Driven Preparation of Plasmonic Resonance-Based Alloy and Core-Shell Catalyst for Sustainable Enhanced Hydrogen Production: Green Approach and Characterization. Appl. Catal. B 2018, 231, 137–150. [CrossRef] 41. Chen, S.-H.; Chan, S.-H.; Lin, Y.-T.; Wu, M.-C. Enhanced Power Conversion Efﬁciency of Perovskite Solar Cells Based on Mesoscopic Ag-Doped TiO2 Electron Transport Layer. Appl. Surf. Sci. 2019, 469, 18–26. [CrossRef] 42. Wu, M.-C.; Liao, Y.-H.; Chan, S.-H.; Lu, C.-F.; Su, W.-F. Enhancing Organolead Halide Perovskite Solar Cells Performance through Interfacial Engineering Using Ag-Doped TiO2 Hole Blocking Layer. Sol. RRL 2018, 2, 1800072. [CrossRef] 43. Akel, S.; Dillert, R.; Balayeva, N.; Boughaled, R.; Koch, J.; El Azzouzi, M.; Bahnemann, D. Ag/Ag2O as a Co-Catalyst in TiO2 Photocatalysis: Effect of the Co-Catalyst/Photocatalyst Mass Ratio. Catalysts 2018, 8, 647. [CrossRef] 44. Wen, L.; Liu, B.; Zhao, X.; Nakata, K.; Murakami, T.; Fujishima, A. Synthesis, Characterization, and Photocatalysis of Fe-Doped TiO2: A Combined Experimental and Theoretical Study. Int. J. Photoenergy 2012, 2012, 368750. [CrossRef] 297 Catalysts 2019, 9, 72 298 1. Introduction Hydrogen is a key energy carrier that will likely play an important role in the transportation sector by 2050 [1]. It is considered an environmentally friendly energy vector due to its zero CO2 and zero noxious gas emissions when combusted [2]. Through a photocatalytic water-splitting process, hydrogen can be produced using water and a light source as primary resources [3]. Furthermore, a sacriﬁcial organic agent is required to allow the photocatalytic reaction to occur, forming the desired products [4]. Common sacriﬁcial agents include methanol, triethanolamine, ethanol, acids and inorganic compounds [5]. Ethanol, as one of the most-investigated sacriﬁcial agents, provides high quantum efﬁciencies and will be used in this work as scavenger. The use of ethanol as a sacriﬁcial agent is advantageous given that it can be easily produced from renewable biomass (fermentation processes), making it available and inexpensive [6]. Photocatalytic hydrogen production with sacriﬁcial organic agents proceeds as follows: (a) absorbed photons surpass the energy band gap and generate excited electron-hole pairs [7], and (b) photoexcited electron-hole pairs can be separated due to the sacriﬁcial agent presence. This allows the formation of hydrogen with minimum electron-hole pair recombination [8] and (c) hydroxyl groups from dissociated water lead OH· radical formation and contribute to the conversion of the scavenger [9]. The “in-series-parallel” reaction network was described in detailed by our research team in [10]. Titanium dioxide (TiO2) is a well-known photocatalyst capable of absorbing light and producing electron-hole pairs to accelerate the rate of a water-splitting reaction [11]. TiO2 has been the most used material due to its stability, resistance to corrosion, cleanliness (no pollutant), availability in nature and inexpensiveness compared to other semiconductors [12]. It can be found in three Catalysts 2019, 9, 33; doi:10.3390/catal9010033 www.mdpi.com/journal/catalysts 299 Catalysts 2019, 9, 33 allotropic phases—anatase, rutile, and brookite—where the anatase phase is the most photoactive phase reported [13]. Doping noble metals on TiO2 allows (a) increasing the efﬁciency of the hydrogen evolution reaction, (b) narrowing the band gaps, and (c) improving the optoelectronic semiconductor properties [14]. Furthermore, doping TiO2 with noble metals such as Pd helps to promote energy levels near the band edges or mid-gap states, which may reduce the effective bandgap energy [15]. Therefore, the effect of Pd on TiO2 can be attributed to the shift of Fermi levels in the composite material [16–20]. 2. Results and Discussion 2. Results and Discussion 2.1. Photocatalyst Characterization 1. Introduction One of the most important parameters in photocatalytic reactors and photocatalysts is the quantum yield (QY). This parameter establishes the process efﬁciency relating the photogenerated radical rate over the absorbed photons rate [21]. Using the QY, the doped photocatalysts of the present study were evaluated. Previous studies by our research team considered Pt doped on TiO2 obtaining modest quantum yield efﬁciencies in the 0.7–8.0% range when using undoped DP-25 and 1.0% Pt-impregnated DP-25, respectively [22]. However, the present work emphasizes the photocatalytic hydrogen production using doped Pd, as a much less expensive dopant on mesoporous TiO2. This semiconductor material, Pd–TiO2, is employed under near-UV light, in the Photo-CREC Water II reactor unit developed at CREC (Chemical Reactor Engineering Centre). Results obtained are of signiﬁcant value due to the high quantum yields obtained in the system. 2.1.2. Pulse Hydrogen Chemisorption Table 3 reports hydrogen chemisorption showing the effect of metal loading on metal dispersion. When Pd is used as a dopant, it is shown that higher metal loadings lead to reduced metal dispersion. Table 3. Chemisorption analysis: metal dispersion. Table 3. Chemisorption analysis: metal dispersion. Photocatalyst Metal Dispersion (%) F-127–0.25 wt% Pd–TiO2 500 ◦C 75 F-127–0.50 wt% Pd–TiO2 500 ◦C 27 F-127–1.0 wt% Pd–TiO2 500 ◦C 26 F-127–2.5 wt% Pd–TiO2 500 ◦C 12 F-127–5.0 wt% Pd–TiO2 500 ◦C 8 2.1.1. Brunauer–Emmett–Teller (BET) Surface Area Using a BET surface area unit (Micrometrics, ASAP 2010), the photocatalysts were analyzed with nitrogen. Furthermore, the adsorption–desorption isotherms of type IV were generated and the Vp total photocatalyst pore volume was also calculated with the liquid nitrogen adsorbed at the P/Po relative pressure of 0.99 [23]. Table 1 reports the speciﬁc surface area, the average pore diameter and the speciﬁc pore volume for mesoporous TiO2 prepared using F-127 template. It is shown that when using this template, the morphological properties of TiO2 are improved as follows: (a) speciﬁc surface areas are increased, (b) average pore diameters are augmented, and (c) speciﬁc pore volumes are increased. Furthermore, one can also notice that the F-127–TiO2, displays both higher porosity and speciﬁc surface area than the TiO2 itself. Table 1. Surface area and pore diameter using template Pluronic F-127. Photocatalyst SBET (m2 g−1) DpBJH (4VpBJH/SBET) (nm) VpBJH (cm3g−1) Degussa P-25 59 7.5 0.25 F-127–TiO2 500 ◦C 140 17.5 0.61 Table 1. Surface area and pore diameter using template Pluronic F-127. According to Table 2, the best results in speciﬁc surface area were obtained with the mesoporous TiO2 photocatalysts. These photocatalysts display a clear increment of the speciﬁc surface area and speciﬁc pore volume (Dp) when compared to Degussa P-25 (commercial titania photocatalyst that is commonly used in photocatalytic reactions). However, when using doped Pd on TiO2, there was a modest reduction in speciﬁc surface area and a mild change in pore diameter attributed to a possible and moderate blocking of the TiO2 pores with Pd [24]. 300 Catalysts 2019, 9, 33 Table 2. Surface area and pore diameter using palladium. Table 2. Surface area and pore diameter using palladium. p g p Photocatalyst SBET (m2 g−1) Dp BJH (4VpBJH/SBET) (nm) VpBJH (cm3g−1) Anatase 11 7.3 0.05 Rutile 5 4.7 0.05 Degussa P-25 59 7.5 0.25 F-127–TiO2-500 ◦C 140 17.5 0.61 F-127–0.25 wt% Pd–TiO2 500 ◦C 131 16.5 0.53 F-127–0.50 wt% Pd–TiO2 500 ◦C 124 16.8 0.52 F-127–1.0 wt% Pd–TiO2 500 ◦C 123 21.2 0.65 F-127–2.5 wt% Pd–TiO2 500 ◦C 122 19.9 0.60 F-127–5.0 wt% Pd–TiO2 500 ◦C 119 18.9 0.56 The Barrett–Joyner–Halenda (BJH) method was also used to determine the pore size distribution, by utilizing N2 as an adsorbate and as a desorption isotherm. In all cases, a bimodal pore volume distribution was observed. The largest pore sizes in the 16–35 nm range were achieved with F-127–1.0 wt% Pd–TiO2-500 ◦C. 2.1.3. X-Ray Diffraction (XRD) XRDs overlapped for comparison. A = anatase, Pd = palladium. 2.1.3. X-Ray Diffraction (XRD) Figure 1 reports XRD diffractograms for TiO2 synthesized using the F-127 template and with varying palladium loadings. Anatase and rutile XRDs are reported as references. In this respect, one can observe that anatase peaks are at the 2θ diffraction angles of 25◦, 38◦, 48◦, 54◦, 63◦, 69◦, 70.5◦ and 75◦corresponding to the planes (101), (004), (200), (105), (204), (116), (220) and (215) [JCPDS No. 73-1764]. For rutile, there are a couple of noticeable peaks at 54◦and 67◦corresponding to the planes (201) and (301) [JCPDS No. 34-0180]. Furthermore, Figure 1 also shows that the XRD diffractograms for Pd-doped TiO2 were consistent where a signiﬁcant anatase XRD peak was observed. The nature of the desirable anatase peaks in this semiconductor was conﬁrmed with a 99.7% anatase from Aldrich reference sample [25]. Anatase for all photocatalyst was the dominant TiO2 crystalline phase assumed as 100% with no rutile being present. Pd peaks were also identiﬁed and recorded at 40.12◦(111) and 46.66◦(200) at the 2θ angles [JCPDS No. 87-0638]. One should observe that, in principle, a third peak at 2θ = 68.1◦(220) might be recorded when using Pd as a dopant. However, this peak may overlap with anatase and as a result cannot be used for Pd identiﬁcation [26]. On the other hand, the crystallite sizes for each photocatalyst were determined using the Scherrer equation. On this basis, the mesoporous photocatalysts displayed crystallite sizes between 9 and 14 nm. Lattice parameters of the tetragonal anatase unit cell were also calculated and are reported in Appendix B. 301 Catalysts 2019, 9, 33 $ $ $ $ $ $ $ $ $ $ $ $ 3G $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ ,QWHQVLW\DX WKHWDGHJUHH $QDWDVH 5XWLOH 7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 $ $ $ $ $ 3G $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ 3G 3G 3G 3G 3G 3G 3G 3G 3G Figure 1. X-ray diffraction (XRD) diffractograms of photocatalyst doped with Pd. XRDs overlapped for comparison. A = anatase, Pd = palladium. ,QWHQVLW\DX Figure 1. X-ray diffraction (XRD) diffractograms of photocatalyst doped with Pd. 2.1.4. Band Gap Figure 2 reports that increasing the Pd content slightly augments the band gap. However, and regarding the observed results, one can see in all cases a signiﬁcant reduction in the band gaps for TiO2 doped with lower Pd loadings versus the band gaps for undoped TiO2. The best band gaps achieved were 2.51 for 0.25 wt% Pd and 2.55 eV for 0.50–1.00 wt% Pd–TiO2. Figure 2. Effect of Pd loading on the optical band gap. Figure 2. Effect of Pd loading on the optical band gap. By applying the Kubelka–Munk (K–M) model and following the Tauc plot methodology, the band gaps were determined. Figure 3 reports the changes of the “(αhv)1/2” function versus the photon energy “hν”, with α representing the absorption coefﬁcient, h being the Planck constant (6.34 × 1034 J s/photon) and v denoting the radiation frequency. It should also be noted that v = c/λ, where c is the speed of light under vacuum (3.00 × 108 m/s2). If the straight-line methodology is applied for the band gap calculation as shown with the red line, one can see that the intersection of this line with the abscissa provides the wavelength corresponding to the semiconductor band gap. Furthermore, the Tauc plots (Figure 3) were developed for Pd-doped TiO2 photocatalysts using the F-127 template and at a 500 ◦C calcination temperature. 302 Catalysts 2019, 9, 33 0.00 1.00 2.00 3.00 4.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 (αhv)0.5 hv (eV) Figure 3. Band gap calculation using the Tauc plot methodology and the straight-line extrapolation for 1.0 wt% Pd–TiO2. Catalysts 2019, 9, 33 0.00 1.00 2.00 3.00 4.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 (αhv)0.5 hv (eV) Figure 3. Band gap calculation using the Tauc plot methodology and the straight-line extrapolation for 1.0 wt% Pd–TiO2. 2.1.5. X-Ray Photoelectron Spectroscopy (XPS) 2.1.5. X-Ray Photoelectron Spectroscopy (XPS) The 1.0 wt% Pd–TiO2 photocatalyst was analyzed using the XPS technique. Figure 4 shows the XPS for Pd(0) and PdO. In each case, one can see double peaks as follows: (a) Pd (0) 3d5/2, with 334.70 eV and 3d3/2 at 339.96 eV binding energies, (b) PdO 3d5/2 with 336.46 eV and 3d3/2 at 341.72 eV. This yields a consistent 48.6% of Pd (0) and 51.4% of PdO, at the two binding energy ranges considered. Figure 4. High-resolution X-ray photoelectron spectroscopy (XPS) spectra for 1.00 wt% Pd–TiO2. 2.2. Macroscopic Radiation Energy Balance (MREB) Photocatalytic reactors operate based on emitted photons. These photons are absorbed by a circulating semiconductor slurry suspension. To be able to establish the absorbed radiation in the Photo-CREC Water II Reactor, one must develop a macroscopic radiation balance for accurate energy efﬁciency calculations [27]. The macroscopic balance estimates the photons absorbed as the difference between the incident photons and the combined scattered and transmitted photons [28]. Pa = Pi −Pbs −Pt (1) (1) where Pa is the rate of absorbed photons, for which it is desired to be as high as possible; Pi is the rate of photons reaching the reactor at the inner Pyrex glass surface and is calculated according to Equation (1) in Einstein s−1; Pbs is the rate of backscattered photons; and Pt is the rate of transmitted photons. All these variables can be expressed using the Einstein s−1 units. Furthermore: Pi = P0 −Pa−wall (2) (2) with P0 in Einstein s−1 being the rate of photons emitted by the lamps as per Pa-wall in Einstein s−1, which accounts for the rate of backscattered photons absorbed by the Pyrex glass walls. In addition, P0 can be calculated as: In addition, P0 can be calculated as: P0 = λ2 λ1 λ L 0 2π 0 q(θ, Z, λ)r dθ dz dλ (3) (3) where q (θ, z, λ) is the radiative ﬂux (J s−1 m−3), λ represents the wavelength (nm), r stands for the radial coordinate (m), z denotes the axial coordinate (m), h is the Planck’s constant (J s), and c represents the speed of light (m s−1). The term q (θ, z, λ) is determined using the spectrometer. where q (θ, z, λ) is the radiative ﬂux (J s−1 m−3), λ represents the wavelength (nm), r stands for the radial coordinate (m), z denotes the axial coordinate (m), h is the Planck’s constant (J s), and c represents the speed of light (m s−1). The term q (θ, z, λ) is determined using the spectrometer. Furthermore, when photocatalytic experiments are performed in the Photo-CREC Water II (PCW-II) reactor, photons are absorbed and scattered in the reacting medium. As a result, a backscattering has to be accounted for. 2.1.4. Band Gap Note: Full lines represent Pd (0) at (a) 3d5/2 and (c) 3d3/2. Broken lines represent PdO at (b) 3d5/2 and (d) 3d3/2. Figure 4. High-resolution X-ray photoelectron spectroscopy (XPS) spectra for 1.00 wt% Pd–TiO2. Note: Full lines represent Pd (0) at (a) 3d5/2 and (c) 3d3/2. Broken lines represent PdO at (b) 3d5/2 and (d) 3d3/2. It was also observed that titanium and oxygen were present as major components in the photocatalyst mesoporous support as TiO2 species. Titanium was detected at a binding energy 303 Catalysts 2019, 9, 33 position of 454.45 eV, while oxygen was identiﬁed at 525.85 eV. These bands fell outside the Pd and PdO binding energies as shown in Figure 4, avoiding any possible inadequate band assignment. PdO binding energies as shown in Figure 4, avoiding any possible inadequate band assignment. In conclusion, the XPS data of Figure 4 shows the signiﬁcant Pd (0) availability, and points towards possible future improvements of the synthesized photocatalyst via enhanced Pd reduction. In conclusion, the XPS data of Figure 4 shows the signiﬁcant Pd (0) availability, and points towards possible future improvements of the synthesized photocatalyst via enhanced Pd reduction. 2.2. Macroscopic Radiation Energy Balance (MREB) A possible approach to calculate backscattering is to establish the difference between Pi and the rate of photons transmitted when the catalyst concentration approaches zero (Pt\|C→0+): Pbs = Pi −Pt \|c→0+ (4) (4) Equation (4) assumes that photons are backscattered on the TiO2 particle layer close to the inner surface of the transparent Pyrex walls surface. Equation (4) also assumes that no other backscattered photons contribute to Pbs. Additionally, for Pt determination, Equation (5) considers that transmitted radiation can be deﬁned as the addition of normal scattered photons and forward scattered photons: Pt = Pns −Pf s (5) Pt = Pns −Pf s (5) One should note that (Pfs + Pns) can be measured by employing aluminum polished collimators, which capture radiation reaching the measuring point, with large view angles [28]. Thus, to assess Pa as in Equation (1), macroscopic balances using near-UV light were established at the central axial position using a 0.15 g/L photocatalyst concentration. Figure 5 reports measurements for various TiO2 photocatalysts with different metal loadings. 304 Catalysts 2019, 9, 33 According to Table 4 and Figure 5, one can observe that additions of Pd on TiO2 show that (a) lower Pd levels (0.25 to 1.00 wt%) lead to an increased Pa and high absorption efﬁciencies compared to undoped TiO2, and (b) higher Pd levels (2.50 and 5.00 wt% Pd) give smaller Pa and reduced absorption efﬁciency. These ﬁndings are in line with an increased rate of transmitted photons when using low Pd loadings as well as incremental photon backscattering when using high Pd loadings. Table 4. Absorbed photon rates on TiO2 photocatalysts at 0.15 g/L of photocatalyst concentration. Table 4. Absorbed photon rates on TiO2 photocatalysts at 0.15 g/L of photocatalyst concentration. Near-UV Light Pa (Einstein/s) TiO2 3.11 × 106 0.25 wt% Pd 3.18 × 10−6 0.50 wt% Pd 3.52 × 10−6 1.00 wt% Pd 5.11 × 10−6 2.50 wt% Pd 3.77 × 10−6 5.00 wt% Pd 3.76 × 10−6 0.0 20.0 40.0 60.0 80.0 100.0 TiO2 0.25% 0.50% 1.00% 2.50% 5.00% Absoption Efficiency (%) Photocatalyst Loading (wt%) Figure 5. Absorption efﬁciency on TiO2 photocatalysts at different metal loadings under near-UV light. 2.3. Hydrogen Production Figure 5. Absorption efﬁciency on TiO2 photocatalysts at different metal loadings under near-UV light. Figure 5. Absorption efﬁciency on TiO2 photocatalysts at different metal loadings under near-UV light. 2.3. Hydrogen Production 2.3.1. Effect of Palladium Loadings 2.3.1. Effect of Palladium Loadings Palladium was used as co-catalyst to dope the structure of the TiO2 photocatalyst. This metal enhances the hydrogen production, as compared to the undoped mesoporous TiO2. Nobel metal crystallites reduce the band gap and facilitate electron capture [29]. As a result, Pd reduces the recombination between holes and electrons, promoting better photocatalytic water-splitting performances [30]. Figure 6 reports the inﬂuence of Pd on TiO2 in terms of cumulative hydrogen volume. 7L2) ZW3G7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 +\GURJHQ9ROXPHFP 673 ,UUDGLDWLRQWLPHK 7L2) ZW3G7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 ZW3G7L2 +\GURJHQ9ROXPHFP 673 ,UUDGLDWLRQWLPHK Figure 6. Cumulative hydrogen volume using Pd at different metal loadings (0.25, 1.50, 1.00, 2.50 and 5.00 wt%). Conditions: photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±3.0%. Figure 6. Cumulative hydrogen volume using Pd at different metal loadings (0.25, 1.50, 1.00, 2.50 and 5.00 wt%). Conditions: photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±3.0%. 305 Catalysts 2019, 9, 33 Catalysts 2019, 9, 33 Figure 6 shows there is a maximum volume of 140 cm3 STP (standard temperature and pressure) of hydrogen produced in six hours when using 1.00 wt% Pd on TiO2. This volume is slightly higher than the maximum volume of hydrogen produced when using 0.25 wt% and 0.50 wt% Pd–TiO2, and three times the volume of hydrogen obtained for undoped TiO2. One should also note as well that this volume is close to the 128 cm3 STP of hydrogen produced when platinum is used as a dopant under the same reaction conditions but with a much larger metal loading (5.00 wt% Pt) on TiO2 [31]. Furthermore, one should note that the 140 cm3 STP of hydrogen produced in six hours with 1.00 wt% Pd on TiO2 decreased up to 60 cm3 STP when using higher Pd loadings (2.50 wt% Pd and 5.00 wt% Pd). The macroscopic radiation energy balance provides an explanation showing that at the higher Pd loadings, there is increased irradiation backscattering, with greater irradiation being reﬂected and, as a consequence, light absorption being reduced. 2.3.1. Effect of Palladium Loadings This is in contrast with the lower than 1.00 wt% Pd loadings evaluated, where the absorption efﬁciency, as well as the rate of transmitted photons, increases. Thus, a diminished irradiation absorption given by 2.50 wt% Pd and 5.00 wt% Pd negatively affects the photocatalyst performance [32]. In agreement with this, at the lower palladium loadings studied (0.25, 0.50 and 1.00 wt%) good metal dispersion, mildly affected speciﬁc surface area and pore structure were achieved [33]. On the other hand, for 2.50 and 5.00 wt% Pd–TiO2, poorer metal dispersion with larger metal crystallite sizes were observed, with this being in line with the lower photocatalytic activity [34]. In all cases, palladium-doped TiO2 showed a consistent steady linear trend. The hydrogen production rate displayed consistent zero-order kinetics, with no noticeable photocatalytic decay. This material is stable for extended irradiation periods and no apparent deactivation for 24 h following an “in series-parallel” reaction mechanism shown in detail in [22]. These results show that palladium at 1.00 wt% loading can produce valuable hydrogen yields, with this being an excellent replacement for platinum. As well, Pd can be considered more advantageous than Pt, given that Pd is less expensive (only 20–25% of the cost of platinum). Furthermore, and given the premise of nominal 1.00 wt% Pd–TiO2, photocatalyst X-ray ﬂuorescence spectrometry (XRF) was used to conﬁrm the nominal loading. The observed XRF value was 1.17 wt% Pd on mesoporous TiO2. 2.3.2. Effect of Catalyst Concentration on Hydrogen Production 2.3.2. Effect of Catalyst Concentration on Hydrogen Production Considering that 1.00 wt% Pd–TiO2 showed the best performance in terms of hydrogen production, additional experiments were carried out to determine the inﬂuence of the catalyst concentration during photoreaction. Figure 7 displays four different slurry concentrations of the 1.00 wt% Pd-TiO2 photocatalyst: 0.15, 0.30, 0.50 and 1.00 g/L. These experiments were studied during 6 h of irradiation. One can observe that the runs with 1.00 g/L showed the highest hydrogen production. Thus, given these results, it can be considered that when higher photocatalyst slurry concentrations are used, more photocatalyst electron-holes are provided, with this promoting better hydrogen production. As a result, it was observed that the hydrogen production rate increased 54% when photocatalyst concentration was augmented seven times from 0.15 to 1.00 g/L. However, despite this hydrogen production increase, this could be considered a modest improvement only, given that the photocatalyst needed and the related cost was signiﬁcantly augmented. Therefore, a photocatalyst concentration of 0.15 g/L was considered as a best choice and was selected for further experimentation. 306 Catalysts 2019, 9, 33 +\GURJHQ9ROXPHFP 673 ZW3G7L2)J/ ZW3G7L2)J/ ZW3G7L2)J/ ZW3G7L2)J/ ,UUDGLDWLRQWLPHK Figure 7. Cumulative hydrogen production using 1.0 wt% Pd–TiO2 at different catalyst concentrations (0.15, 0.30, 0.50 and 1.0 g/L). Conditions: 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±4.0%. Catalysts 2019, 9, 33 +\GURJHQ9ROXPHFP 673 ZW3G7L2)J/ ZW3G7L2)J/ ZW3G7L2)J/ ZW3G7L2)J/ ,UUDGLDWLRQWLPHK Figure 7. Cumulative hydrogen production using 1.0 wt% Pd–TiO2 at different catalyst concentrations (0.15, 0.30, 0.50 and 1.0 g/L). Conditions: 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±4.0%. 2.3.3. Effect of Photo-CREC Water II Atmosphere using Argon and CO2 2.3.3. Effect of Photo-CREC Water II Atmosphere using Argon and CO2 Before starting water-splitting runs, the reactor gas chamber was purged with an inert gas to remove the oxygen from the air, avoiding combustion reactions. Argon was used initially as the inert gas given this is heavier than oxygen facilitating its displacement [35]. On the other hand, CO2 was also used in separate runs in the reactor gas chamber to determine its possible inﬂuence on water dissociation reactions. 2.3.2. Effect of Catalyst Concentration on Hydrogen Production According to Figure 8, using argon as an inert gas and utilizing 1.00 wt% Pd–TiO2, yielded 140 cm3 STP of hydrogen after six hours of irradiation. On the other hand, when a CO2 atmosphere was employed, only 80 cm3 STP of hydrogen was obtained. One should note that under a CO2 atmosphere and due to the competition of the CO2 photoreduction with the hydrogen production, a lower net hydrogen formation can be explained. E ZW3G7L2&2 ZW3G7L2&2 ZW3G7L2&2 ,UUDGLDWLRQWLPHK D ZW3G7L2$U ZW3G7L2$U ZW3G7L2$U +\GURJHQ9ROXPHFP 673 ,UUDGLDWLRQWLPHK Figure 8. Cumulative hydrogen production using xPd–TiO2 (x = 1.00, 2.50 and 5.00 wt%) and under two atmospheres: (a) argon and (b) CO2. Conditions: photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: (a) ±3.0%, (b) ±3.4%. Regarding CO2 during the six h of irradiation, it was observed that it steadily augmented und n argon atmosphere reaching 0 4 cm3 STP On the other hand when the runs were performed under E ZW3G7L2&2 ZW3G7L2&2 ZW3G7L2&2 ,UUDGLDWLRQWLPHK D ZW3G7L2$U ZW3G7L2$U ZW3G7L2$U +\GURJHQ9ROXPHFP 673 ,UUDGLDWLRQWLPHK Figure 8. Cumulative hydrogen production using xPd–TiO2 (x = 1.00, 2.50 and 5.00 wt%) and under two atmospheres: (a) argon and (b) CO2. Conditions: photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: (a) ±3.0%, (b) ±3.4%. Regarding CO2 during the six h of irradiation, it was observed that it steadily augmented under an argon atmosphere reaching 0.4 cm3 STP. On the other hand, when the runs were performed under a CO2 atmosphere, the CO2 increment was limited to 0.01 cm3 STP. These ﬁndings support the view that there is competition between CO2 photoreduction and CO2 formation via ethanol OH· radical Regarding CO2 during the six h of irradiation, it was observed that it steadily augmented under an argon atmosphere reaching 0.4 cm3 STP. On the other hand, when the runs were performed under a CO2 atmosphere, the CO2 increment was limited to 0.01 cm3 STP. 2.3.4. Effect of Sacriﬁcial Agent Concentration As a scavenger, ethanol offers important advantages, such as the photogeneration of electron-holes, limiting electron-site recombination and improving photocatalytic activity. Ethanol can donate electrons to scavenge the valence holes and suppresses the reverse reaction [37]. Experiments were performed at 1.00 wt% Pd and three ethanol concentrations (1.0, 2.0, 4.0 v/v%) under an argon atmosphere and with 0.15 g/L of photocatalyst concentration. This was done to evaluate the effect of the ethanol concentration on hydrogen production. Figure 9 reports the inﬂuence of increasing ethanol from 2.0–4.0% on hydrogen production rates. +\GURJHQ9ROXPHFP 673 ZW3G7L2(WKDQRO ZW3G7L2(WKDQRO ZW3G7L2(WKDQRO ZW3G7L2(WKDQRO5 ZW3G7L2(WKDQRO5 ,UUDGLDWLRQWLPHK Figure 9. Hydrogen volume using 1.00 wt% Pd at 1.0, 2.0 and 4.0 v/v% ethanol. Conditions: photocatalyst concentration 0.15 g/L, argon atmosphere, pH = 4 ± 0.05 and near-UV light, R = repeat. Standard deviation: ±6.5%. ,UUDGLDWLRQWLPHK Figure 9. Hydrogen volume using 1.00 wt% Pd at 1.0, 2.0 and 4.0 v/v% ethanol. Conditions: photocatalyst concentration 0.15 g/L, argon atmosphere, pH = 4 ± 0.05 and near-UV light, R = repeat. Standard deviation: ±6.5%. As shown in Figure 9, the highest hydrogen formation rate was obtained at the highest ethanol concentration. However, these important ethanol concentrations changes did not inﬂuence hydrogen production signiﬁcantly. This was particularly true between 2.0 and 4.0 v/v% ethanol concentration. Therefore, 2.0 v/v% was considered fully adequate and was the selected concentration of the ethanol scavenger used for further studies. 2.3.2. Effect of Catalyst Concentration on Hydrogen Production These ﬁndings support the view that there is competition between CO2 photoreduction and CO2 formation via ethanol OH· radical 307 Catalysts 2019, 9, 33 scavenging. It is assumed that these gas phase CO2 ﬁndings could be also be inﬂuenced by the enhanced CO2 solubility in water–ethanol [36]. 2.3.4. Effect of Sacriﬁcial Agent Concentration 2.3.5. By-Products Formation There are several by-products generated from the water-splitting reaction in the gas phase. Detected by-products include methane, ethane, acetaldehyde and CO2. To quantify these by-products, gas samples were taken hourly from the gas port located in the storage tank. They were analyzed using a Shimadzu gas chromatograph (GC) unit. All the experiments were repeated at least three times to secure reproducibility. One can thus see that as soon as the photo-redox reaction starts, all these by-products, together with hydrogen, increase progressively as is shown in Figure 10. In the liquid phase, ethanol was also measured using a Shimadzu HPLC. One can observe in Figure 11 a balanced consumption-formation of ethanol, with a net stable ethanol concentration. This occurs when hydrogen is being produced using the 1.00 wt% Pd–TiO2 photocatalyst. The observed trends could be considered a promising result, showing that none or little additional scavenger is required in subsequent runs once the initial ethanol is fed to the Photo-CREC Water II reactor unit. 308 Catalysts 2019, 9, 33 ZW3G7L2 &29ROXPHFP 673 ZW3G7L2 &+9ROXPHFP 673 ZW3G7L2 &+29ROXPHFP 673 ,UUDGLDWLRQWLPHK F D ,UUDGLDWLRQWLPHK ZW3G7L2 &+9ROXPHFP 673 E G Figure 10. Hydrocarbon proﬁles of (a) carbon dioxide (CO2), (b) methane (CH4), (c) acetaldehyde (C2H4O) and (d) ethane (C2H6) at 1.00 wt% Pd. Conditions: Photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, argon atmosphere, pH = 4 ± 0.05 and near-UV light. Standard deviation: (a) ±4.1%, (b) ±4.7%, (c) ±5.1%, (d) ±6.3%. 2.3.5. By-Products Formation ZW3G7L2 PO ZW3G7L2 &+9ROXPHFP 673 ,UUDGLDWLRQWLPHK ZW3G7L2 &+9ROXPHFP 673 E G ZW3G7L2 &29ROXPHFP 673 ZW3G7L2 &+9ROXPHFP 673 ZW3G7L2 &+29ROXPHFP 673 ,UUDGLDWLRQWLPHK F D ,UUDGLDWLRQWLPHK ZW3G7L2 &+9ROXPHFP 673 E G Figure 10. Hydrocarbon proﬁles of (a) carbon dioxide (CO2), (b) methane (CH4), (c) acetaldehyde (C2H4O) and (d) ethane (C2H6) at 1.00 wt% Pd. Conditions: Photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, argon atmosphere, pH = 4 ± 0.05 and near-UV light. Standard deviation: (a) ±4.1%, (b) ±4.7%, (c) ±5.1%, (d) ±6.3%. ZW3G7L2 (WKDQRO9ROXPHPO 7LPHK Figure 11. Ethanol changes with irradiation time. Conditions: photocatalyst concentration 0.15 g/L, argon atmosphere, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±3.0%. Figure 10. Hydrocarbon proﬁles of (a) carbon dioxide (CO2), (b) methane (CH4), (c) acetaldehyde (C2H4O) and (d) ethane (C2H6) at 1.00 wt% Pd. Conditions: Photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, argon atmosphere, pH = 4 ± 0.05 and near-UV light. Standard deviation: (a) ±4.1%, (b) ±4.7%, (c) ±5.1%, (d) ±6.3%. Figure 10. Hydrocarbon proﬁles of (a) carbon dioxide (CO2), (b) methane (CH4), (c) acetaldehyde (C2H4O) and (d) ethane (C2H6) at 1.00 wt% Pd. Conditions: Photocatalyst concentration 0.15 g/L, 2.0 v/v% ethanol, argon atmosphere, pH = 4 ± 0.05 and near-UV light. Standard deviation: (a) ±4.1%, (b) ±4.7%, (c) ±5.1%, (d) ±6.3%. ZW3G7L2 (WKDQRO9ROXPHPO 7LPHK Figure 11. Ethanol changes with irradiation time. Conditions: photocatalyst concentration 0.15 g/L, argon atmosphere, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±3.0%. ZW3G7L2 (WKDQRO9ROXPHPO 7LPHK Figure 11. Ethanol changes with irradiation time. QYH• = moles o f H•/s moles o f photons absorbed by the photocatalyst/s (6) 2.3.5. By-Products Formation Conditions: photocatalyst concentration 0.15 g/L, argon atmosphere, 2.0 v/v% ethanol, pH = 4 ± 0.05 and near-UV light. Standard deviation: ±3.0%. 2.4. Quantum Yield (QY) evaluation The quantum yield (QY) is the most important parameter to establish the energy utilization efﬁciency in photocatalytic reactors [38]. In terms of hydrogen production, quantum yield can be deﬁned as the hydrogen radical production rate over the absorbed photon rate on the photocatalyst surface. According to this deﬁnition, QY can be determined as follows: QYH• = moles o f H•/s moles o f photons absorbed by the photocatalyst/s (6) (6) 309 Catalysts 2019, 9, 33 Catalysts 2019, 9, 33 Equation (6) is equivalent to: %QY = dNH dt Pa × 100 (7) (7) where dNH dt represents the rate of moles of hydrogen radicals formed at any time during the photocatalyst irradiation. where dNH dt represents the rate of moles of hydrogen radicals formed at any time during the photocatalyst irradiation. To use Equation (7) the assessment of Pa or the moles of absorbed photons is required. This can be accomplished by using the macroscopic radiation energy balance (MREB) in the Photo-CREC Water II reactor as proposed by Escobedo et al. [39]. Appendix C provides a calculation sample to assess the QY. 2.4.2. Effect of Catalyst Concentration on Quantum Yields Considering the QY% observed for the 1.00 wt% Pd–TiO2 during hydrogen production, further QY% evaluations were developed by changing the photocatalyst concentration in the slurry. Table 6 and Figure 13 report the QY% obtained, by augmenting the photocatalyst concentration, under the following conditions: (a) 2.0 v/v% ethanol as scavenger organic compound, (b) pH = 4 ± 0.05 and (c) near-UV light irradiation. Table 6. Quantum yield for 1.00 wt% Pd–TiO2 photocatalyst at different photocatalyst concentrations in the slurry. Table 6. Quantum yield for 1.00 wt% Pd–TiO2 photocatalyst at different photocatalyst concentrations in the slurry. in the slurry. Catalyst Concentration (g/L) QY (%) 0.15 10.9 0.30 14.5 0.50 22.4 1.00 30.8 4< 7LPHK J/3G7L2&89 J/3G7L2&89 J/3G7L2&89 J/3G7L2&89 Figure 13. QY% at various irradiation times using near-UV irradiation and 0.15, 0.30, 0.50 and 1.00 g/L photocatalyst concentrations. Note: Loading was1.00 wt% Pd on TiO2. Figure 13. QY% at various irradiation times using near-UV irradiation and 0.15, 0.30, 0.50 and 1.00 g/L photocatalyst concentrations. Note: Loading was1.00 wt% Pd on TiO2. Figure 13 provides QY% for different photocatalyst concentrations. Here, it was again observed that there was a noticeable increase of the QY% in the ﬁrst hour of irradiation, followed by a stable QY% in the next 5 h of irradiation. Constant QY% during the 1 to 6-h irradiation period was assigned to the steady hydrogen formation rate, linked to consistent zero-order reaction kinetics with no photocatalyst activity decay observed in all cases. 2.4.1. Effect of Pd Addition on Quantum Yields The quantum yield evaluation for different TiO2 photocatalysts involves rigorous macroscopic radiation energy balances. These calculations require the assessment of the Pt transmitted, the Pi incident, and the Pbs backscattered photons using the macroscopic radiation energy balance as described in Section 2.2. With this information and using Equation (1), the Pa was calculated. Furthermore, for every experiment and once the lamp is turned on, the rate of moles of hydrogen can be established. On this basis, QY% can be calculated using Equation (7). Table 5 and Figure 12 report QY% for the mesoporous photocatalysts doped with palladium at different metal loadings (0.25, 0.50, 1.00, 2.50 and 5.00 wt%) under the following conditions: (a) photocatalyst slurry concentrations of 0.15 g/L, (b) 2.0 v/v% ethanol, (c) pH = 4 ± 0.05 and (d) near-UV light. Table 5. Quantum yield (QY) for the Pd–TiO2 photocatalyst when using 0.15 g/L. All reported data are average values of three repeats. Table 5. Quantum yield (QY) for the Pd–TiO2 photocatalyst when using 0.15 g/L. All reported data are average values of three repeats. Semiconductor QY (%) F–127 TiO2 5.0 F-127–0.25 wt% Pd–TiO2 13.7 F-127–0.50 wt% Pd–TiO2 12.8 F-127–1.00 wt% Pd–TiO2 10.9 F-127–2.50 wt% Pd–TiO2 9.6 F-127–5.00 wt% Pd–TiO2 8.5 4< 7LPHK 7L2& 3G7L2& 3G7L2& 3G7L2& 3G7L2& 3G7L2& Figure 12. QY% at various irradiation times under near-UV light and 0.15 g/L of photocatalyst concentration, and using Pd at different loadings (0.25, 0.50, 1.00, 2.50 and 5.00 wt%). Figure 12. QY% at various irradiation times under near-UV light and 0.15 g/L of photocatalyst concentration, and using Pd at different loadings (0.25, 0.50, 1.00, 2.50 and 5.00 wt%). 310 Catalysts 2019, 9, 33 Catalysts 2019, 9, 33 There is a signiﬁcant increase of QY% with 0.25, 0.50 and 1.00 wt% Pd–TiO2, whereas higher Pd loadings led to a decrease of QY%. These results are in line with the QY% of 8% reported by Escobedo when Pt addition proceeds [39]. Figure 12 reports that QY% displays consistent trends for Pd-doped TiO2 photocatalysts: (a) during the ﬁrst hour of irradiation, QY% increased progressively until it reached a stable value; and (b) during the following six hours of irradiation, QY% remained unchanged, with this showing a steady performance of the photocatalysts under study. 3. Experimental Methods The photocatalysts of the present study were synthesized using the sol–gel methodology and doped with palladium. Different techniques were utilized to characterize the doped semiconductors as follows: (a) BET for speciﬁc surface area, (b) chemisorption for crystallite size, (c) x-ray diffraction 311 Catalysts 2019, 9, 33 for crystallographic structure and (d) UV-vis absorption for band gaps. The prepared semiconductors were evaluated in a Photo-CREC Water II reactor unit. 3.1. Photocatalyst Synthesis The sol–gel method can be used for photocatalyst synthesis by converting monomers into colloids (sol phase), and thus promoting a gel structure formation [40]. The sol–gel method for TiO2 synthesis can be modiﬁed, leading to improvements in photocatalyst structural properties such as particle diameter and surface area. Therefore, this also leads to improved photocatalytic activity [41]. Some copolymers, such as Pluronic® F-127 and Pluronic® P-123, formed by chains of ethylene oxide and propylene oxide, can be used for TiO2 synthesis as soft templates. These templates optimize the pore structure network during semiconductor preparation, enhancing pore size distribution, enlarging the surface area, controlling the purity, homogeneity, and morphology of mesoporous materials [42]. Rusinque shows that the Pluronic F-127 template has a greater impact than the Pluronic P-123 template on TiO2 photoactivity, increasing the hydrogen production up to 86% [31]. Thus, considering the Pluronic F-127 advantage over Pluronic P-123 for hydrogen production, further experiments were carried out using only copolymer Pluronic F-127. The sol–gel method adopted used the following reagents: (a) ethanol USP (C2H5OH) from commercial alcohols, (b) hydrochloric acid (HCl, 37% purity), (c) Pluronic F-127, (d) anhydrous citric acid, (e) titanium (IV) isopropoxide, and (f) palladium (II) chloride (PdCl2, 99.9% purity). All the reagents were obtained from Sigma Aldrich, with photocatalyst preparation effected according to the methodology proposed by Guayaquil et al. [43]. Figure 14 describes the sol–gel synthesis as follows: (a) Step 1: In 400 mL of ethanol, 33 g of hydrochloric acid and 20 g of Pluronic F-127 were added until dissolution, under continuous stirring for 1 h. (b) Step 2: 6.30 g of citric acid were dissolved in 20 mL of water for posterior addition to the initial suspension to mix them together for 1 h in order to set the pH at 0.75. (c) Step 3: 28.5 g of titanium (IV) isopropoxide was dissolved in ethanol and added dropwise to the mixture. Finally, palladium (II) chloride was incorporated at different loadings (0.25 to 5.00 wt% Pd). (d) Step 4: The resulting sol–gel suspension was stirred for 24 h and then calcined at 500 ◦C for 6 hours under an air atmosphere. The copolymer was evaporated during the thermal treatment and an ordered mesoporous titanium framework was formed [23]. Figure 14. Photocatalyst preparation process describing the four steps considered for Pd-doped mesoporous. Figure 14. Photocatalyst preparation process describing the four steps considered for Pd-doped mesoporous. 3.2. Equipment The Photo-CREC Water II (PCW-II) reactor is a novel unit used for water splitting reactions and therefore, hydrogen production. It is a 5.7 L slurry batch reactor conﬁgured with two concentric tubes: (a) an inner tube made from transparent borosilicate (Pyrex) and (b) an outer tube made from opaque polyethylene. The ﬂuorescent lamp is placed inside this inner Pyrex tube. Furthermore, the 312 Catalysts 2019, 9, 33 suspended photocatalyst ﬂows in the annular space between the outer polyethylene tube and the inner Pyrex transparent tube which only absorbs 5%) of the near-UV light emitted by the lamp [44]. See Appendix A for a detailed lamp characterization. The PCW-II unit is equipped with a storage feed tank where the photocatalyst suspension is always kept sealed under agitation. This tank has 2 ports for periodic liquid and gas phase sampling. Figure 15 describes the main components of PCW-II: (a) the Photo-CREC Water II Reactor, (b) the centrifugal pump, (c) the sealed storage tank, and (d) the electrical circuit powering the near-UV light lamp. Figure 15. Schematic representation of the Photo-CREC Water II Reactor with a H2 Mixing/Storage Tank: (A) partial longitudinal cross-section of the PCW- II unit showing the down ﬂow circulation of the slurry in the annular channel, (B) overall view of PCW-II showing windows, near UV lamp and recirculation pump (C) hydrogen storage tank with its components, (D) detail of a photocatalyst particle. Figure 15. Schematic representation of the Photo-CREC Water II Reactor with a H2 Mixing/Storage Tank: (A) partial longitudinal cross-section of the PCW- II unit showing the down ﬂow circulation of the slurry in the annular channel, (B) overall view of PCW-II showing windows, near UV lamp and recirculation pump (C) hydrogen storage tank with its components, (D) detail of a photocatalyst particle. The emitted radiation spectra of the lamp used inside the Photo-CREC Water II was established using a Stellar Net EPP2000-25 spectrometer (StellarNet Inc.). The light source is a polychromatic black light blue (BLB) Ushio UV lamp (15 W, 0.305 A, 55 V) with a spectral peak at 368 nm in the 300–420 nm emission range [45]. 3.5. Analytical Techniques 3.5. Analytical Techniques The gas phase was analyzed with a Shimadzu GC2010 gas chromatograph using argon (Praxair 99.999%) as gas carrier. It has 2 detectors, a ﬂame ionization detector (FID) and a thermal conductivity detector (TCD). This unit was equipped with a HayeSepD 100/120 mesh packed column (9.1 m × 2 mm × 2 μm nominal SS) used for the separation of hydrogen from air. This equipment detects hydrogen (H2), carbon monoxide (CO), carbon dioxide (CO2), methane (CH4) and other hydrocarbon organic species. A Shimadzu HPLC model UFLC (ultra-fast liquid chromatography) system was utilized to characterize the liquid phase. This analytical technique allows the liquid mobile phase (0.1% H3PO4) to transport the sample through a column (Supelcogel C-610H 30cm × 7.8mm ID) containing a stationary phase. It selectively separates individual compounds (i.e., ethanol) from water for further detection. This quantitative analysis is performed by employing the RID (refractive index detector) 10A due to polar nature of ethanol. Both the GC and the HPLC analytical techniques were used simultaneously. Samples were taken different irradiation times. 3.4. Hydrogen Production 3.4. Hydrogen Production Pd-doped TiO2 photocatalysts were evaluated using the Photo-CREC Water II reactor equipped with the BLB near-UV lamp for 6 hours of continuous irradiation. This lamp was turned on 30 min before initiating the photoreaction. The hydrogen storage/mixing tank was loaded with 6000 mL of water. Ethanol was used as an organic scavenger and the pH was adjusted to 4 ± 0.05 with H2SO4 [2M] keeping the photoreaction under acidic conditions, which favours available H+ for water splitting process [39]. Following this step, the photocatalyst was loaded at a speciﬁc weight concentration ensuring that most of the radiation was absorbed in the slurry medium. The photocatalyst was subjected to sonication, which reduces the formation of particle agglomerates and promotes homogeneous mixing. Argon gas was circulated to guarantee an inert atmosphere at the beginning of the reaction. 3.3. Photocatalyst Characterization Photocatalyst speciﬁc surfaces areas were determined using a BET surface area analyzer (Micrometrics, ASAP 2010) at −195 ◦C. Each photocatalyst was degassed at 300 ◦C during a period of 3 h. The BET analysis was developed using nitrogen to generate the adsorption–desorption equilibrium isotherms and to establish the isotherm inﬂection point. The BJH (Barrett–Joyner–Halenda) method was used to determine the pore size distribution, by utilizing the desorption isotherm with N2 as an adsorbate. 313 Catalysts 2019, 9, 33 Catalysts 2019, 9, 33 By using the Micromeritics AutoChem II Analyzer for pulse chemisorption, one can calculate the fraction of dispersed metal and average active metal crystallite size [46]. Furthermore, to identify the phases of a crystalline material, X-ray diffraction (XRD) was used [47]. The XRD spectra were analyzed in a Rigaku Rotating Anode X-Ray Diffractometer (Rigaku) perated at 45 kV and 160 mA. The scans were taken between 20–80◦, with a step size of 0.02◦and a dwell time of 2 s/step. In order to determine the characteristic band gap associated to each photocatalyst an UV-VIS-NIR spectrophotometer (Shimadzu UV-3600) was used [48]. BaSO4 was utilized as a reference sample. Kubelka–Munk (K–M) developed a Tauc plot methodology that was followed to establish the corresponding band gaps [49]. X-ray photoelectron spectroscopy (XPS) analysis was also used to identify the elemental composition and the chemical state of each element in the synthesized photocatalyst [50]. 4. Conclusions Funding: This research was funded by Natural Sciences and Engineering Research Council of Canada (NSERC) and the University of Western Ontario, grant given to Hugo de Lasa. Funding: This research was funded by Natural Sciences and Engineering Research Council of Canada (NSE and the University of Western Ontario, grant given to Hugo de Lasa. Acknowledgments: We would like to gratefully thank Florencia de Lasa who assisted with the editing and the drafting of the graphical abstract of the present article. Acknowledgments: We would like to gratefully thank Florencia de Lasa who assisted with the editing and the drafting of the graphical abstract of the present article. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 4. Conclusions (a) The TiO2 mesoporous photocatalysts of the present study were prepared using a F-127 template and following a sol–gel methodology. It was found that the mesoporous prepared using a F-127 template displayed a good photocatalytic performance. (a) The TiO2 mesoporous photocatalysts of the present study were prepared using a F-127 template and following a sol–gel methodology. It was found that the mesoporous prepared using a F-127 template displayed a good photocatalytic performance. (b) The prepared Pd–TiO2 photocatalysts were characterized using BET, XRD, UV-VIS and XPS. On this basis it was proven that energy band gaps were signiﬁcantly affected with Pd addition, and that binding energies showed signiﬁcant contribution of the Pd (0) on the doped-palladium TiO2. (c) Macroscopic radiation energy balances were successfully employed to establish photon absorption rates and radiation absorption efﬁciencies in the PCW-II unit. For the Pd–TiO2 semiconductors, photon absorption efﬁciencies were in the 45 and 60% range under near-UV light. (c) Macroscopic radiation energy balances were successfully employed to establish photon absorption rates and radiation absorption efﬁciencies in the PCW-II unit. For the Pd–TiO2 semiconductors, photon absorption efﬁciencies were in the 45 and 60% range under near-UV light. 314 Catalysts 2019, 9, 33 (d) The formation of hydrogen using Pd–TiO2 photocatalysts followed, in all cases, steady zero-order kinetics with no apparent photocatalyst activity decay. (d) The formation of hydrogen using Pd–TiO2 photocatalysts followed, in all cases, steady zero-order kinetics with no apparent photocatalyst activity decay. (d) The formation of hydrogen using Pd–TiO2 photocatalysts followed, in all cases, steady zero-order kinetics with no apparent photocatalyst activity decay. (e) The prepared Pd–TiO2 photocatalysts under near UV-light were shown to be adequate for hydrogen production reaching up to 210 cm3 STP when using the 1.00 wt%-Pd on TiO2. This photocatalyst showed a best QY% of 30.8%. (e) The prepared Pd–TiO2 photocatalysts under near UV-light were shown to be adequate for hydrogen production reaching up to 210 cm3 STP when using the 1.00 wt%-Pd on TiO2. This photocatalyst showed a best QY% of 30.8%. (e) The prepared Pd–TiO2 photocatalysts under near UV-light were shown to be adequate for hydrogen production reaching up to 210 cm3 STP when using the 1.00 wt%-Pd on TiO2. This photocatalyst showed a best QY% of 30.8%. Author Contributions: Conceptualization, investigation and supervision, H.d.L.; proposed methodology and supervision, S.E.; validation, formal analysis and writing, B.R. Nomenclature Nomenclature CO2 Carbon dioxide CH4 Methane C2H6 Ethane C2H4O Acetaldehyde c Speed of light (3.0 × 108 m/s) Dp Pore diameter (cm) e- Electron h+ Hole h Planck’s constant (6.63 × 1034 J/s) Ebg Energy band gap (eV) Eav Average energy of a photon (kJ/mol photon) F-127 Poly (ethylene oxide)/poly (propylene oxide)/poly (ethylene oxide) H• Hydrogen radical H2O Water I(λ) Intensity of light (W/cm2) OH- Hydroxide ions OH• Hydroxide radicals P-123 Poly (ethylene glycol)-block-poly (propylene glycol)-block-poly (ethylene glycol) P0 Rate of photons emitted by the BLB lamp (einstein/s) Pa Rate of absorbed photons (einstein/s) Pa-wall Rate of photons absorbed by the inner pyrex glass (einstein/s) Pbs Rate of backscattered photons exiting the system (einstein/s) Pd Palladium PdCl2 Palladium II chloride PEO Poly (ethylene oxide) Pfs Rate of forward-scattered radiation (einstein/s) Pi Rate of photons reaching the reactor inner surface (einstein/s) Pns Rate of transmitted non-scattered radiation (einstein/s) PPO Poly (propylene oxide) Pt Rate of transmitted photons (einstein/s) Pt Platinum q (θ, z, λ, t) Net radiative ﬂux over the lamp emission spectrum (μW/cm2) t Time (h) TiO2 Titanium dioxide V Total volume of the gas chamber (5716 cm3) W Weight (g) Wt% Weight percent (% m/m) 315 Catalysts 2019, 9, 33 Greek symbols θ Diffraction angle, also scattering angular angle (o) λ Wave length (nm) ϕ Quantum Yield Efﬁciency (%) Acronyms BJH Barrett–Joyner–Halenda model BLB Black light blue lamp BET Brunauer–Emmett–Teller Surface Area Method CB Conduction band DP25 Degussa P25 (TiO2) JCPDS International Centre for Diffraction Data MIEB Macroscopic Irradiation Energy Balance PCW-II Photo CREC Water II reactor PC Photocatalyst concentration STP Standard temperature and pressure (273 K and 1 atm) UV Ultraviolet VB Valence band Bg Band gap Appendix A. Lamp Characterization Greek symbols θ Diffraction angle, also scattering angular angle (o) λ Wave length (nm) ϕ Quantum Yield Efﬁciency (%) Acronyms BJH Barrett–Joyner–Halenda model BLB Black light blue lamp BET Brunauer–Emmett–Teller Surface Area Method CB Conduction band DP25 Degussa P25 (TiO2) JCPDS International Centre for Diffraction Data MIEB Macroscopic Irradiation Energy Balance PCW-II Photo CREC Water II reactor PC Photocatalyst concentration STP Standard temperature and pressure (273 K and 1 atm) UV Ultraviolet VB Valence band Bg Band gap Appendix A. Lamp Characterization Appendix A. Lamp Characterization Figure A1 reports the spectrum of the polychromatic BLB Ushio near-UV lamp, with an observed output power of 1.61 W and an average of 325.1 kJ/photon mole of emitted photon energy. Nomenclature T:FP :DYHOHQJKWQP Figure A1. Near-UV Lamp Irradiation Spectrum. Figure A1. Near-UV Lamp Irradiation Spectrum. Figure A1. Near-UV Lamp Irradiation Spectrum. The average emitted photon energy was calculated using the recorded irradiation spectra as follows [51]: λ ( ) ( ) Eav = λmax λmin I (λ) E(λ) dλ λmax λmin I (λ) dλ (A1) (A1) where, where, E(λ) = hc λ (A2) (A2) With h being the Planck constant (6.34 × 10−34 J s/photon), c representing the speed of light in a vacuum (3.00 × 108 m/s2) and λ denoting the wavelength expressed in nanometers (nm). I is the emitted photons intensity (W/cm2), assessed as I (λ) ≈q (θ, z, λ, t) dλ and measured with a 316 Catalysts 2019 9 33 Catalysts 2019, 9, 33 spectrophotoradiometer. The irradiance is represented by q (θ, z, λ, t) dλ and given by the lamps spectra as shown in Figure A2. The average emitted photon Energy was calculated as shown in Equation (A1) Eav = λmax λmin I (λ) E(λ) dλ λmax λmin I (λ) dλ = λmax λmin hc λ ∗q (θ, z, λ, t) dλ λmax λmin q (θ, z, λ, t) dλ (A3) Eav = 5.36 × 10−19J/mol photon = 325.1 KJ/mol photo (A4) (A4) Regarding the PCW-II, the axial distribution of the radiative ﬂux was determined. Figure A2 reports the near-UV lamp axial radiation distribution. One can observe that the radiation proﬁle shows no signiﬁcant changes in radiation levels in the central section of the PCW-II. On the other hand, signiﬁcant radiation decay can be seen approaching the endpoints of the lamp [52]. Regarding the PCW-II, the axial distribution of the radiative ﬂux was determined. Figure A2 reports the near-UV lamp axial radiation distribution. One can observe that the radiation proﬁle shows no signiﬁcant changes in radiation levels in the central section of the PCW-II. On the other hand, signiﬁcant radiation decay can be seen approaching the endpoints of the lamp [52]. [ [ [ [ [ [ [ T(LQVWHLQV /HQJWKFP Figure A2. Near-UV Lamp Axial Distribution. /HQJWKFP Figure A2. Near-UV Lamp Axial Distribution. Appendix B. Semiconductor Crystallite Sizes and Lattice Parameters Appendix B. Semiconductor Crystallite Sizes and Lattice Parameters Appendix C. Quantum Yield Calculation Appendix C. Quantum Yield Calculation As stated in Section 2.4, QY% can be deﬁned as the number of moles of hydrogen radical produced per absorbed photons on the photocatalyst surface: %QY = dNH dt Pa × 100 (A5) (A5) where dNH dt Pa represents the rate of moles of hydrogen radicals formed and Pa stands for the moles of photons absorbed. As well, and according to the Macroscopic Irradiation Energy Balances (MIEB) in the Photo-CREC Water Reactor II, Pa was calculated as follows: Pa = Pi −Pbs −Pt (A6) (A6) where, Pi is the rate of photons reaching the reactor at the inner reactor surface, Pbs represents the rate of backscattered photons, and Pt is the rate of transmitted photons (Einstein s−1). A sample calculation is given below considering a hydrogen production rate of 0.2494 μmol/cm3 h using: (a) 1.0 wt.% Pd-TiO2, (b) a photocatalyst concentration of 1.0 g/L, (c) ethanol at 2.0 v/v%, (d) pH = 4 ± 0.05, (e) near-UV Light, (f) gas phase volume in the reactor of 5716 cm3 and (g) Pa = 2.57 × 10−6 Einstein/s. QYH• = 2 ∗(0.2494 × 10−6 mol/cm3h) ∗(5716 cm3) ∗(6.022 × 1023 photon/mol H2) ∗(1h/3600s) 2.57 × 1017 photon/s %QYH• = 30.8% (A7) QYH• = 2 ∗(0.2494 × 10−6 mol/cm3h) ∗(5716 cm3) ∗(6.022 × 1023 photon/mol H2) ∗(1h/3600s) 2.57 × 1017 photon/s %QYH• = 30.8% (A7) QYH• = 2 ∗(0.2494 × 10−6 mol/cm3h) ∗(5716 cm3) ∗(6.022 × 1023 photon/mol H2) ∗(1h/3600s) 2.57 × 1017 photon/s %QYH• = 30.8% (A7) (A7) %QYH• = 30.8% Appendix B. Semiconductor Crystallite Sizes and Lattice Parameters The crystallite sizes were determined using the Scherrer equation as reported in the enclosed Table A1. On this basis the mesoporous photocatalysts displayed crystallite sizes between 9 and 14 nm. Table A1. Photocatalyst Crystallite Sizes. Photocatalyst Crystallite Size (nm) TiO2 9 TiO2 0.25 wt% Pd 500 ◦C 11 TiO2 0.50 wt% Pd 500 ◦C 11 TiO2 1.00 wt% Pd 500 ◦C 11 TiO2 2.50 wt% Pd 500 ◦C 13 TiO2 5.00 wt% Pd 500 ◦C 14 Table A1. Photocatalyst Crystallite Sizes. Furthermore, the calculated a, b and c lattice constants of the tetragonal anatase unit cell are shown in Table A2 indicating that pure anatase was successfully obtained with the phase structures maintained at α = β = γ = 90◦angles. These resulting a, b, and c parameters are in closed agreement with those reported in the literature [53]. Note that lattice parameters a = b ̸= c and these were calculated for Anatase phase (h k l) = (1 0 1). 317 Catalysts 2019, 9, 33 Table A2. Lattice Parameters for TiO2 and Pd doped TiO2. Table A2. Lattice Parameters for TiO2 and Pd doped TiO2. Photocatalyst a = b c 2θ (deg) d (Å) TiO2 [53] 3.7821 9.5022 25.33 3.5139 TiO2 500 ◦C (our study) 3.7679 9.5002 25.41 3.5025 TiO2 0.25 wt% Pd 500 ◦C 3.7832 9.4833 25.33 3.5139 TiO2 0.50 wt% Pd 500 ◦C 3.7858 9.4737 25.31 3.5155 TiO2 1.00 wt% Pd 500 ◦C 3.7825 9.5099 25.32 3.5147 TiO2 2.50 wt% Pd 500 ◦C 3.7748 9.4713 25.38 3.5065 TiO2 5.00 wt% Pd 500 ◦C 3.7691 9.4809 25.41 3.5025 pendix C. Quantum Yield Calculation References Photosensitised dissociation of water using dispersed suspensions of n type s J. Chem. Soc. Faraday Trans. 1 Phys. Chem. Condens. Phases 1982, 12, 3659–3669. [CrossRef] 10. Escobedo Salas, S. Photocatalytic Water Splitting Using a Modiﬁed Pt-TiO2. Kinetic Modeling and Hydrogen Production Efﬁciency. Ph.D. Thesis, The University of Western Ontario, London, ON, Canada, August 2013. scobedo Salas, S. Photocatalytic Water Splitting Using a Modiﬁed Pt TiO2. Kinetic Modeling and Hydrogen roduction Efﬁciency. Ph.D. Thesis, The University of Western Ontario, London, ON, Canada, August 2013. 11. Khan, M.M.; Adil, S.F.; Al-Mayouf, A. Metal oxides as photocatalysts. J. Saudi Chem. Soc. 2015, 19, 462–464. [CrossRef] 12. Haider, A.J. Exploring potential Environmental applications of TiO2 Nanoparticles. Energy Procedia 2017, 119, 332–345. [CrossRef] 13. Chin, W.L.; Low, F.W.; Chong, S.W.; Hamid, S.B.A. An Overview: Recent Development of Titanium Dioxide Loaded Graphene Nanocomposite Film for Solar Application. Curr. Org. Chem. 2015, 19, 1882–1895. 14. Yang, J. Roles of Cocatalysts in Photocatalysis and Photoelectrocatalysis. Acc. Chem. Res. 2013, 46, 1900–1909. [CrossRef] 15. Moslah, C.; Kandyla, M.; Mousdis, G.A.; Petropoulou, G.; Ksibi, M. Photocatalytic Properties of Titanium Dioxide Thin Films Doped with Noble Metals (Ag, Au, Pd, and Pt). Phys. Status Solidi Appl. Mater. Sci. 2018, 215, 1–7. [CrossRef] 16. García-Zaleta, D.S.; Torres-Huerta, A.M.; Domínguez-Crespo, M.A.; García-Murillo, A.; Silva-Rodrigo, R.; González, R.L. Inﬂuence of Phases Content on Pt/TiO2, Pd/TiO2 Catalysts for Degradation of 4-Chlorophenol at Room Temperature. J. Nanomater. 2016, 2016, 1805169. [CrossRef] 17. Subramanian, V.; Wolf, E.E.; Kamat, P.V. Catalysis with TiO2/Gold Nanocomposites. Effect of Metal Particle Size on the Fermi Level Equilibration. J. Am. Chem. Soc. 2004, 126, 4943–4950. [CrossRef] 18. Santara, B.; Pal, B.; Giri, P.K. Signature of strong ferromagnetism and optical properties of Co doped TiO2 nanoparticles. J. Appl. Phys. 2011, 110, 114322. [CrossRef] 19. Khairy, W.; Zakaria, M. Effect of metal-doping of TiO2 nanoparticles on their photocatalytic activities toward removal of organic dyes. Egypt. J. Pet. 2014, 23, 419–426. [CrossRef] 19. Khairy, W.; Zakaria, M. Effect of metal-doping of TiO2 nanoparticles o removal of organic dyes. Egypt. J. Pet. 2014, 23, 419–426. [CrossRef] 20. Sobana, N.; Muruganadham, M.; Swaminathan, M. Nano-Ag particles doped TiO2 for efﬁcient photodegradation of Direct azo dyes. J. Mol. Catal. A Chem. 2006, 258, 124–132. [CrossRef] 21. Cassano, A.E.; Martin, C.A.; Brandi, R.J.; Alfano, O.M. Photoreactor Analysis and Design: Fundamentals and Applications. Ind. Eng. Chem. Res. 1995, 34, 2155–2201. [CrossRef] 22. References 1. Ramesohl, S.; Merten, F. Energy system aspects of hydrogen as an alternative fuel in transport. Energy Po 2006, 34, 1251–1259. [CrossRef] 2. Barreto, L.; Makihira, A.; Riahi, K. The hydrogen economy in the 21st century: A sustainable developm scenario. Int. J. Hydrogen Energy 2003, 28, 267–284. [CrossRef] 3. Maeda, K.; Teramura, K.; Lu, D.; Takata, T.; Saito, N.; Inoue, Y.; Domen, K. Photocatalyst releasing hydrogen from water. Nature 2006, 440, 295. [CrossRef] [PubMed] 3. Maeda, K.; Teramura, K.; Lu, D.; Takata, T.; Saito, N.; Inoue, Y.; Domen, K. Photocatalyst releasing hydrogen from water. Nature 2006, 440, 295. [CrossRef] [PubMed] 4. Gali´nska, A. Photocatalytic Water Splitting over Pt−TiO2 in the Presence of Sacriﬁcial Reagents. Energy Fuels 2005, 19, 1143–1147. [CrossRef] 4. Gali´nska, A. Photocatalytic Water Splitting over Pt−TiO2 in the Presence of Sacriﬁcial Reagents. Energy Fuels 2005, 19, 1143–1147. [CrossRef] 5. Wang, M.; Shen, S.; Li, L.; Tang, Z.; Yang, J. Effects of sacriﬁcial reagents on photocatalytic hydrogen evolution over different photocatalysts. J. Mater. Sci. 2017, 52, 5155–5164. [CrossRef] 5. Wang, M.; Shen, S.; Li, L.; Tang, Z.; Yang, J. Effects of sacriﬁcial reagents on photocatalytic hydrogen evolution over different photocatalysts. J. Mater. Sci. 2017, 52, 5155–5164. [CrossRef] 6. López, C.R.; Melián, E.P.; Méndez, J.A.O.; Santiago, D.E.; Rodríguez, J.M.D.; Díaz, O.G. Comparative study of alcohols as sacriﬁcial agents in H2production by heterogeneous photocatalysis using Pt/TiO2 catalysts. J. Photochem. Photobiol. A Chem. 2015, 312, 45–54. [CrossRef] 6. López, C.R.; Melián, E.P.; Méndez, J.A.O.; Santiago, D.E.; Rodríguez, J.M.D.; Díaz, O.G. Comparative study of alcohols as sacriﬁcial agents in H2production by heterogeneous photocatalysis using Pt/TiO2 catalysts. J. Photochem. Photobiol. A Chem. 2015, 312, 45–54. [CrossRef] 7. Mills, A. An overview of semiconductor photocatalysis. J. Photochem. Photobiol. A Chem. 1997, 108, 1–35. [CrossRef] 7. Mills, A. An overview of semiconductor photocatalysis. J. Photochem. Photobiol. A Chem. 1997, 108, 1–35. [CrossRef] 318 Catalysts 2019, 9, 33 8. Abe, R. Signiﬁcant effect of iodide addition on water splitting into H2 and O2 over Pt-loaded TiO2 photocatalyst: Suppression of backward reaction. Chem. Phys. Lett. 2003, 371, 360–364. [CrossRef] p y pp y 9. Mills, A. Photosensitised dissociation of water using dispersed suspensions of n-type semiconducto p y pp y 9. Mills, A. Photosensitised dissociation of water using dispersed suspensions of n-type semicond J. Chem. Soc. Faraday Trans. 1 Phys. Chem. Condens. Phases 1982, 12, 3659–3669. [CrossRef] 9. Mills, A. References Escobedo, S.; Serrano, B.; Calzada, A.; Moreira, J.; de Lasa, H. Hydrogen production using a platinum modiﬁed TiO2 photocatalyst and an organic scavenger. Kinetic modeling. Fuel 2016, 181, 438–449. [CrossRef] 23. Yu, J.C.; Wang, X.; Fu, X. Pore-Wall Chemistry and Photocatalytic Activity of Mesoporous Titania Molecular Sieve Films. Chem. Mater. 2004, 16, 1523–1530. [CrossRef] 24. Pan, X.; Xu, Y.J. Defect-mediated growth of noble-metal (Ag, Pt, and Pd) nanoparticles on TiO2 with oxygen vacancies for photocatalytic redox reactions under visible light. J. Phys. Chem. C 2013, 117, 17996–18005. [CrossRef] 25. Zhang, J.; Zhou, P.; Liu, J.; Yu, J. New understanding of the difference of photocatalytic activity among anatase, rutile and brookite TiO2. Phys. Chem. Chem. Phys. 2014, 16, 20382–20386. [CrossRef] [PubMed] 26. Rodriguez-Vindas, D. Synthesis of palladium with different nanoscale structures by sputtering deposition onto ﬁber templates. J. Nanophotonics 2008, 2, 021925. [CrossRef] 27. Moreira, J.; Serrano, B.; Ortiz, A.; de Lasa, H.; de Lasa, H. Evaluation of Photon Absorption in an Aqueous TiO2 Slurry Reactor Using Monte Carlo Simulations and Macroscopic Balance. Ind. Eng. Chem. Res. 2010, 49, 10524–10534. [CrossRef] 28. Salaices, M.; Serrano, B.; de Lasa, H.I. Experimental evaluation of photon absorption in an aqueous TiO2 slurry reactor. Chem. Eng. J. 2002, 90, 219–229. [CrossRef] 29. Thornton, J.M.; Raftery, D. Efﬁcient photocatalytic hydrogen production by platinum-loaded carbon-doped d i i d i l ACS A l M t I t f 2012 4 2426 2431 [C R f] 29. Thornton, J.M.; Raftery, D. Efﬁcient photocatalytic hydrogen production by platinum-loaded carbon-doped cadmium indate nanoparticles. ACS Appl. Mater. Interfaces 2012, 4, 2426–2431. [CrossRef] 30. Yoshida, H.; Hirao, K.; Nishimoto, J.; Shimura, K.; Kato, S.; Itoh, H.; Hattori, T. Hydrogen production from methane and water on platinum loaded titanium oxide photocatalysts. J. Phys. Chem. C 2008, 112, 5542–5551. [CrossRef] 319 Catalysts 2019, 9, 33 31. Rusinque, B. Hydrogen Production by Photocatalytic Water Splitting Under Near-UV and Visible Light Using Doped Pt and Pd TiO2. Master Thesis, The University of Western Ontario, London, ON, Canada, September 2018. p 32. Zhang, N.; Liu, S.; Fu, X.; Xu, Y.J. Synthesis of M@TiO2 (M = Au, Pd, Pt) core-shell nanocomposites with tunable photoreactivity. J. Phys. Chem. C 2011, 115, 9136–9145. [CrossRef] 33. Riyapan, S.; Boonyongmaneerat, Y.; Mekasuwandumrong, O.; Yoshida, H.; Fujita, S.; Arai, M.; Panpranot, J. Improved catalytic performance of Pd/TiO2in the selective hydrogenation of acetylene by using H2-treated sol-gel TiO2. J. Mol. Catal. A Chem. References 2014, 383–384, 182–187. [CrossRef] 34. Akbayrak, S.; Tonbul, Y.; Özkar, S. Nanoceria supported palladium(0) nanoparticles: Superb catalyst in dehydrogenation of formic acid at room temperature. Appl. Catal. B Environ. 2017, 206, 384–392. [CrossRef] 34. Akbayrak, S.; Tonbul, Y.; Özkar, S. Nanoceria supported palladium(0) nanoparticles: Superb catalyst in dehydrogenation of formic acid at room temperature. Appl. Catal. B Environ. 2017, 206, 384–392. [CrossRef] 35. Borodin, V.B.; Tsygankov, A.A.; Rao, K.K.; Hall, D.O. Hydrogen production byAnabaena variabilisPK84 35. Borodin, V.B.; Tsygankov, A.A.; Rao, K.K.; Hall, D.O. Hydrogen production byAnabaena variabilisPK84 under simulated outdoor conditions. Biotechnol. Bioeng. 2000, 69, 478–485. [CrossRef] 36. Dalmolin, I.; Skovroinski, E.; Biasi, A.; Corazza, M.L.; Dariva, C.; Oliveira, J.V. Solubility of carbon dioxide in binary and ternary mixtures with ethanol and water. Fluid Phase Equilib. 2006, 245, 193–200. [CrossRef] 37. Puangpetch, T.; Sreethawong, T.; Yoshikawa, S.; Chavadej, S. Hydrogen production from photocatalytic water splitting over mesoporous-assembled SrTiO3nanocrystal-based photocatalysts. J. Mol. Catal. A Chem. 2009, 312, 97–106. [CrossRef] 38. Ibrahim, H.; de Lasa, H. Novel photocatalytic reactor for the destruction of airborne pollutants reaction kinetics and quantum yields. Ind. Eng. Chem. Res. 1999, 38, 3211–3217. [CrossRef] 39. Escobedo, S.; Serrano, B.; de Lasa, H. Serrano and H. de Lasa. Quantum Yield with Platinum Modiﬁed TiO Photocatalysts for Hydrogen Prodcution. Appl. Catal. B. Environ. 2013, 140, 523–536. [CrossRef] 40. Guo, S.P.; Li, J.C.; Xu, Q.T.; Ma, Z.; Xue, H.G. Recent achievements on polyanion-type compounds for sodium-ion batteries: Syntheses, crystal chemistry and electrochemical performance. J. Power Sources 2017, 361, 285–299. [CrossRef] 41. Brinker, J.; Schere, G.W. Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing; Academic Press Inc.: San Diego, CA, USA, 1990. 42. He, X. Recent Progress in Fabrication of Nanostructured Carbon Monolithic Materials; Elsevier: Oxford, UK, 2017. 43 G il S J F S R l B V l dé P l PJ d L H Ph l i h d 42. He, X. Recent Progress in Fabrication of Nanostructured Carbon Monolithic Materials; Elsevier: Oxford, UK, 2017. 43. Guayaquil-Sosa, J.F.; Serrano-Rosales, B.; Valadés-Pelayo, P.J.; de Lasa, H. Photocatalytic hydrogen e, X. Recent Progress in Fabrication of Nanostructured Carbon Monolithic Materials; Elsevier: Oxford, UK, 2017. 43. Guayaquil-Sosa, J.F.; Serrano-Rosales, B.; Valadés-Pelayo, P.J.; de Lasa, H. Photocatalytic hydro production using mesoporous TiO2 doped with Pt. Appl. Catal. B Environ. 2017, 211, 337–348. [CrossRe 44. De Lasa, H.; Serrano, B.; Salaices, M. Photocatalytic Reaction Engineering; Springer Scicence: New York, USA, 2005. 45. Comparing the Efﬁciency of N-Doped TiO2 and N-Doped Bi2MoO6 Photo Catalysts for MB and Lignin Photodegradation Ricardo Rangel 1,, Verónica Janneth Cedeño 1, Jaime Espino 1, Pascual Bartolo-Pérez 2, Geonel Rodríguez-Gattorno 2 and Juan José Alvarado-Gil 2 Ricardo Rangel 1,, Verónica Janneth Cedeño 1, Jaime Espino 1, Pascual Bartolo-Pérez 2, Geonel Rodríguez-Gattorno 2 and Juan José Alvarado-Gil 2 1 División de estudios de Posgrado, Facultad de Ingeniería Química, Universidad Michoacana de S.N.H. Morelia Z.P. 58060, Michoacán, Mexico; cedegarci@gmail.com (V.J.C.); jespinova@yahoo.com.mx (J.E.) 2 CINVESTAV-IPN, Unidad Mérida. Mérida Z.P.97310, Yucatán, Mexico; pascual@cinvestav.mx (P.B.-P.); geonelr@gmail.com (G.R.-G.); jjag09g@gmail.com (J.J.A.-G.) * Correspondence: rrangel@umich.mx; Tel.: +52-443-327-3584 1 División de estudios de Posgrado, Facultad de Ingeniería Química, Universidad Michoacana de S.N.H. Morelia Z.P. 58060, Michoacán, Mexico; cedegarci@gmail.com (V.J.C.); jespinova@yahoo.com.mx (J.E.) 2 CINVESTAV-IPN Unidad Mérida Mérida Z P97310 Yucatán Mexico; pascual@cinvestavmx (PB -P); Received: 27 November 2018; Accepted: 13 December 2018; Published: 19 December 2018 Received: 27 November 2018; Accepted: 13 December 2018; Published: 19 December 2018 Received: 27 November 2018; Accepted: 13 December 2018; Published: 19 December 2018 Abstract: In this study, we tested the efﬁciency of nitrogen-doped titanium dioxide (N-TiO2) and nitrogen-doped bismuth molybdate (N-Bi2MoO6) compounds as photocatalysts capable of degrading methylene blue and lignin molecules under irradiation with ultraviolet (UV) and visible light (VIS). Moreover, we compared TiO2 and Bi2MoO6 catalysts with N-TiO2 and N-Bi2MoO6 compounds using chemical coprecipitation. The catalysts were prepared starting from Ti(OCH2CH2CH3)4, Bi(NO3)3·5H2O, and (NH4)6Mo7O24 reagents. N-doping was achieved in a continuous reﬂux system, using ethylene diamine as a nitrogen source. The resulting materials were characterized using Scanning Electron Microscopy (SEM), X-Ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and X-ray photoelectron spectroscopy (XPS). Additionally, we observed the decrease in particle size after processing the compounds in the reﬂux system. The results regarding photocatalytic degradation tests show a remarkable effect for nitrogen doped samples, achieving 90% of lignin degradation. Keywords: photocatalysis; Titanium dioxide; bismuth molybdate; lignin; UV light; visible light catalysts catalysts Catalysts 2018, 8, 668; doi:10.3390/catal8120668 www.mdpi.com/journal/catalysts References Ushio. UV-B Blacklight & Blacklight Blue; Catalogue: Cypress, CA, USA. 46. AutoChem 2920 Automated Catalyst Characterization System Operator’s Manual; AutoChem: Norcross, GA, USA, 2014. 47. Warren, B.E. X-Ray Diffraction; Dover Publications: New York, NY, USA, 1990. 48. UV-VIS-NIR Spectrophotometer; Shimadzu: Tokyo, Japan, 2018. 48. UV-VIS-NIR Spectrophotometer; Shimadzu: Tokyo, Japan, 2018. 49. Slav, A. Optical characterization of TiO2 -Ge nanocomposite ﬁlms obtained by reactive magnetron sputter Dig. J. Nanomater. Biostructures 2011, 6, 915–920. Dig. J. Nanomater. Biostructures 2011, 6, 915–920. 50. Briggs, D. X: X-Ray Photoelectron Spectroscopy. In Handbook of Adhesion, 2nd ed.; 2005; pp. 621–622 51. Serrano, B.; Ortíz, A.; Moreira, J.; de Lasa, H.I. Energy efﬁciency in photocatalytic reactors for the full span of reaction times. Ind. Eng. Chem. Res. 2009, 48, 9864–9876. [CrossRef] 51. Serrano, B.; Ortíz, A.; Moreira, J.; de Lasa, H.I. Energy efﬁciency in photo reaction times. Ind. Eng. Chem. Res. 2009, 48, 9864–9876. [CrossRef] 52. Salaices, M.; Serrano, B.; De Lasa, H. Photocatalytic conversion of Organic pollutants extinction coefﬁcients and quantum efﬁencies. Ind. Eng. Chem. Res. 2001, 40, 5455–5464. [CrossRef] 53. Treacy, J.P.W.; Hussain, H.; Torrelles, X.; Grinter, D.C.; Cabailh, G.; Bikondoa, O.; Nicklin, C.; Selcuk, S.; Selloni, A.; Lindsay, R.; et al. Geometric structure of anatase Ti O2(101). Phys. Rev. B 2017, 95, 1–7. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). y p article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 320 1. Introduction Advanced oxidation processes (AOP) are commonly seen as an alternative to degraded environmental water pollutants, based on their effectiveness, high reactivity, non-selectivity, and their extensive variety of applications [1]. Since its inception [2], titanium oxide remains one of the most effective and versatile compounds for photocatalytic applications, even if traditional compounds like ZnO [3], CdS [4], SnO2 [5], ZnS [6], and BiVO4 [7] have been tested to improve their efﬁciency or performance in comparison to TiO2. Another aspect commonly used to enhance photocatalytic yield is the process manipulation that produces different structural features of TiO2 particles, such as speciﬁc surface area or shape. Some of these manipulations include chemical coprecipitation [8], sol-gel combustion method [9], spray drying [10], or microwave heating [11]. Most of those studies were used to reduce or control the crystal size and had the purpose of increasing the volume/length ratio to maximize the exposed surface, thus increasing the active sites to achieve the catalytic process more effectively, in terms of reactivity [12]. In the majority of reported studies, the results are marginal in comparison to TiO2. Only few compounds, including ZnO or Bi2MoO6, have exhibited promising results in enhancing the catalytic activity in at least one order of magnitude [13,14]. ZnO is as a robust candidate for wastewater treatment, due to its similarity with TiO2, in aspects such as charge carrier dynamics upon bandgap excitation and the generation of reactive oxygen species in 321 Catalysts 2018, 8, 668 aqueous suspensions. However, inherent details like its wide band gap and massive charge carrier recombination has limited their massive usage [3]. Bismuth molybdate (Bi2MoO6) has been profusely studied in the past as a catalyst in oxidation reactions [15]. At the present time, Bi2MoO6 has attracted a lot of attention in photocatalysis research, due to their effectiveness to degrade organic pollutants under UV irradiation. The Bi2MoO6 morphology [16,17] and production methods include sol-gel [18], solvothermal synthesis [19], and co-precipitation [20], among others. An alternative method to increase the catalytic activity of some photocatalytic materials is through the inclusion of some relevant dopant elements, either by using transition metal cations to replace metal sites [21] or by inserting nonmetal anions like F [22], C [23], or N [24]. The purpose is to create new electronic states between the valence and conduction bands, facilitating the electron conduction to enhance the photocatalytic efﬁciency and also favoring the electron-hole formation. 1. Introduction In addition, those new states promote the shifting towards the visible region of the electromagnetic spectrum. In this regard, nitrogen-doping has been considered one of most effective approaches to improve photocatalytic activity of TiO2 in the visible region [25]. The methods currently reported to achieve it superﬁcially or by replacing atoms include ion implantation [26], physical vapor deposition [27], and spray pyrolysis [28], as well as variants of the known process of sol-gel. An alternative method to include nitrogen is the reﬂuxing methodology, which consists of inducing intimate contact between the nitrogen source precursor (hydrazine, urea, ethylene diamine, etc.) and the catalyst inside a closed system of reﬂux at constant temperature. As a result of that methodology, it is possible to obtain catalysts containing nitrogen [29]. The amount of nitrogen that can be introduced depends on factors including the type of precursor, the reﬂux temperature, and the time involved. The purpose of the present research was to study the efﬁciency of titanium oxide and bismuth molybdate, both nitrogen-doped, to be tested as photocatalysts capable of degrading the methylene blue (MB) dye, which is considered as a model molecule for degradation studies. Additionally, we studied the photodegradation of lignin using both kinds of compounds. Lignin is a compound commonly found in wastewater from the wood process manufacturing. Usually, lignin is partially solubilized with acid and their product is concentrated and burned. Even so, some residual amount of lignin remains bonded to cellulose, which in many cases, is discarded to body waters affecting the environment and altering the photosynthetic processes. Our hypothesis is that nitrogen doping on TiO2 and Bi2MoO6 will make the photocatalytic oxidation process more efﬁcient, this achieving the lignin degradation. 2.1. Scanning Electron Microscopy (SEM) The morphology of samples was analyzed using Scanning Electron Microscopy (SEM). The Figure 1a displays a SEM image of TiO2, which shows spherical and deﬁned morphology. The N-doped TiO2 compound, presented in Figure 1b, shows agglomerates of nano particles in a range of 0.1 to 0.5 μm. The image 1c of Bi2MoO6 displays laminar particles of an N-doped Bi2MoO6 compound. Figure 1d exhibits deﬁned, elongated bar-like shaped particles with smooth edges. 322 Catalysts 2018, 8, 668 Figure 1. Scanning Electron Microscopy (SEM) photo-micrograph obtained at 5000X of (a) titanium dioxide (TiO2), (b) nitrogen-doped titanium dioxide (N-TiO2), (c) bismuth molybdate (Bi2MoO6), and (d) nitrogen-doped bismuth molybdate (N-Bi2MoO6) Bi2MoO6 catalysts. Figure 1. Scanning Electron Microscopy (SEM) photo-micrograph obtained at 5000X of (a) titanium dioxide (TiO2), (b) nitrogen-doped titanium dioxide (N-TiO2), (c) bismuth molybdate (Bi2MoO6), and (d) nitrogen-doped bismuth molybdate (N-Bi2MoO6) Bi2MoO6 catalysts. 2.2. X-ray Diffraction Analysis X-ray analyses were carried out in order to capture pristine TiO2 and Bi2MoO6 catalysts. I addition, they were carried out for their corresponding nitrogen doped compounds. For TiO2 compounds, X-ray diffraction (XRD) patterns are displayed in Figure 2. It can be observed that this sample matches well with the anatase phase reported for TiO2. For the N-doped TiO2 sample, wider peaks are observed, as are changes in the intensity of the (101) diffraction plane, revealing that after N-doping through the reﬂux system, TiO2 particles become smaller in size, providing the characteristic spectrum of an amorphous material. Figure 3 shows the diffraction pattern for Bi2MoO6 compounds, which reveals how the orthorhombic phase, also known as γ-Bi2MoO6 phase, was stabilized in these compounds. The crystallite size was calculated for every system using the Scherrer equation, which is as follows: L = kλ WcosΘ (1) (1) where k is a constant that depends on particle morphology (usually k = 1.0 for cubic or nearly-cubic systems); λ is the Cu, Kα radiation (nm); W is the full width at half-maximum (rad); and Θ is the diffraction angle (deg). The (101) reﬂection was used to perform calculations through Equation (1) for TiO2 compounds, while the (131) reﬂection was used to obtain the crystallite size for Bi2MoO6 compounds. The results were 8.6 nm for TiO2, 4.01 nm for N-doped TiO2, 16.14 nm for γ-Bi2MoO6. and 7.92 nm for N-doped Bi2MoO6. Brieﬂy, it appeared that nitrogen doping inhibited the crystal growth. where k is a constant that depends on particle morphology (usually k = 1.0 for cubic or nearly-cubic systems); λ is the Cu, Kα radiation (nm); W is the full width at half-maximum (rad); and Θ is the diffraction angle (deg). The (101) reﬂection was used to perform calculations through Equation (1) for TiO2 compounds, while the (131) reﬂection was used to obtain the crystallite size for Bi2MoO6 compounds. The results were 8.6 nm for TiO2, 4.01 nm for N-doped TiO2, 16.14 nm for γ-Bi2MoO6. and 7.92 nm for N-doped Bi2MoO6. Brieﬂy, it appeared that nitrogen doping inhibited the crystal growth. 323 Catalysts 2018, 8, 668 Figure 2. X-ray diffraction (XRD) patterns of TiO2 and N-TiO2. Figure 3. XRD patterns for Bi2MoO6 and Bi2MoO6-N catalysts. 2.3. Speciﬁc Surface Area Determination The textural properties of TiO2 Bi2MoO6 and N-doped catalysts are summarize Catalysts 2018, 8, 668 Figure 2. X-ray diffraction (XRD) patterns of TiO2 and N-TiO2. 2.2. X-ray Diffraction Analysis Figure 2. X-ray diffraction (XRD) patterns of TiO2 and N-TiO2. Figure 2. X-ray diffraction (XRD) patterns of TiO2 and N-TiO2. g y ( ) p 2 2 Figure 3. XRD patterns for Bi2MoO6 and Bi2MoO6-N catalysts. 2.3. Speciﬁc Surface Area Determination 2.3. Speciﬁc Surface Area Determination The textural properties of TiO2, Bi2MoO6, and N-doped catalysts are summarized in Table 1. It was appreciated that after doping TiO2 and Bi2MoO6, their surface area changed in comparison to the pristine compounds. A dramatic change was observed for the TiO2 compound which, after doping, decreased their BET area by 37%, while speciﬁc surface for Bi2MoO6 area was reduced to 47%. Apparently, the nitration process in both systems entails amorphization and pore reduction. Table 1. Speciﬁc surface area for TiO2, Bi2MoO6, TiO2-N, and Bi2MoO6-N catalysts. Compound BET Surface Area (m2/g) TiO2 117.0 N-TiO2 73.7 Bi2MoO6 3.8 N-Bi2MoO6 2.0 Table 1. Speciﬁc surface area for TiO2, Bi2MoO6, TiO2-N, and Bi2MoO6-N catalysts. Through the reﬂux procedure, carried out for at least 2 h, the constant movement of the particles in intimate contact with the ethylene-diamine and 1-hexanol resulted in a separation of the initially 324 Catalysts 2018, 8, 668 obtained particles, modifying the size of aggregates towards smaller values, which impacted the speciﬁc surface area. The resulting particles were smaller on average, especially in comparison to pristine undoped TiO2 or Bi2MoO6 compounds, in the order of nanometers. While this process occurred, nitrogen molecules were ﬁxed on the surface blocking pores and contributed to an apparent decrease in the speciﬁc surface area. 2.4. Diffuse Reﬂectance Measurements Because the use of visible energy is necessary to test these systems, it is important to determine if there are any energy gap reductions after conducting the doping process. Thus, the diffuse reﬂectance spectra were obtained (Figure 4) and transformed into F(R), with a magnitude proportional to the extinction coefﬁcient through the Kubelka-Munk function. In this case, R is the absolute reﬂectance of the sampled layer. Therefore, it is inferred that by using the following equation, the energy gap could be obtained graphically. 2 [F(R)hν]2 = C(hν −Eg) [F(R)hν]2 = C(hν −Eg) (2) (2) where Eg is the energy gap for every sample. Figure 4. Reﬂectance diffuse measurements for (a) N-TiO2, (b) TiO2, (c) N-Bi2MoO6, and (d) Bi2MoO6, compounds. Figure 4. Reﬂectance diffuse measurements for (a) N-TiO2, (b) TiO2, (c) N-Bi2MoO6, and (d) Bi2MoO6, compounds. The values obtained for the band gap are summarized in Table 2, where it is appreciated that, in both cases, a decreasing occurs after the nitrogen doping. These results indicate the possibility of using the developed compounds as a photocatalyst in the visible region of the electromagnetic spectrum, which will be demonstrated later in this work. Table 2. Gap values for TiO2, Bi2MoO6, N-TiO2, and N-Bi2MoO6 catalysts. Compound Experimental Gap, eV Reported Gap, eV TiO2 3.17 3.20 N-TiO2 2.96 Bi2MoO6 2.84 2.90 N-Bi2MoO6 2.73 2.5. X-ray Photoelectron Spectroscopy (XPS) 2.5. X-ray Photoelectron Spectroscopy (XPS) 2.5. X-ray Photoelectron Spectroscopy (XPS) XPS analyses were performed in samples with the purpose of establishing present elements and especially to determine if nitrogen doping is detected in TiO2 or Bi2MoO6 doped compounds. Figure 5 shows the XPS spectra for TiO2 and TiO2-N compounds. The lower image, in this ﬁgure, corresponds to the pristine TiO2, where the peaks O1s and Ti 2p are pointed out, as well as the C 1s. This last peak corresponds to small traces of carbon. In the upper part of Figure 5, it can be seen the corresponding image to N-TiO2. On it, O1s, Ti 2p which were detected, and the N 1s signal was also found. In addition, a high-resolution scanning analysis was from 390 to 410 eV was performed, 325 Catalysts 2018, 8, 668 aimed at demonstrating the presence of nitrogen, which is shown as inset in the same ﬁgure. The XPS results for the Bi2MoO6 compounds are included in Figure 6, where the XPS spectra was acquired from 0–700 eV. The image shows the 3p3/2, 3d, and 4p states for Mo and Bi 4f, but also the O 1s and N 1 transitions are pointed out. In the upper left, an inset corresponding to the high-resolution energy window for N has been included, in order to emphasize their presence. In this way, for both cases, it is demonstrated that the procedure to introduce or impregnate nitrogen has been carried out successfully. Our results are in agreement with those results reported by other authors regarding the position of the N peak [26–28]. Figure 5. X-ray Photoelectron Spectroscopy (XPS) analysis for TiO2 and N-TiO2 catalysts. Figure 6. XPS analysis for Bi2MoO6 and N-Bi2MoO6 catalysts. Figure 5 X-ray Photoelectron Spectroscopy (XPS) analysis for TiO2 and N-TiO2 catalysts Figure 5. X-ray Photoelectron Spectroscopy (XPS) analysis for TiO2 and N-TiO2 catalysts. Figure 6. XPS analysis for Bi2MoO6 and N-Bi2MoO6 catalysts. Figure 6. XPS analysis for Bi2MoO6 and N-Bi2MoO6 catalysts. Figure 6. XPS analysis for Bi2MoO6 and N-Bi2MoO6 catalysts. The quantiﬁcation of the elements through the integration under the curve of each of their corresponding reﬂections gave us the following values summarized in Table 3. It is observed that titanium oxide compounds are closer to the TiO2 stoichiometry in comparison to Bi2MoO6 compounds. 3. Photocatalytic Activity Tests 3. Photocatalytic Activity Tests Figures 7–9 show photocatalytic activity measurements and solutions containing TiO2 or Bi2MoO6 compounds. The reaction under ultraviolet (UV) and visible light (VIS) was evaluated separately for both, speciﬁcally methylene blue and lignin samples, which were measured to have an optical absorption at 660 nm and 289 nm, respectively. Pristine TiO2 or Bi2MoO6 compounds were tested for MB degradation for about 120 min in the UV range (Figure 7). The degradation reached for TiO2 was 80%, while for Bi2MoO6 it was nearly 70%. On the other hand, both compounds showed an improved degradation behavior when doped with nitrogen, reaching 90% of MB degradation using N-Bi2MoO6 and 93% for N-TiO2. After the MB tests, both Bi2MoO6 and TiO2 compounds were studied for lignin degradation, which is a more complex organic structure, when compared to MB. The degradation attained for Bi2MoO6 was about 62%; the degradation attained for TiO2 was about 70% (Figure 8). In the case of nitrogen-doped compounds, the degradation was 82% for Bi2MoO6–N and 93% for N-TiO2. Figure 8 summarizes the results obtained for lignin degradation that used TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under visible light. Regarding the TiO2 compound, it was less sensitive to VIS irradiation in such a way that only 12% degradation for TiO2 and 25% for Bi2MoO6 were found. These results reveal the complex nature of the lignin molecule, which is hard to degrade and frequently results in compounds derived from this degradation, such as formic acid, ketones, and aldehydes, among others [30–33]. An interesting result was found for lignin degradation when using N-TiO2 and N-Bi2MoO6 catalysts (Figure 9), where a 32% degradation for N-TiO2 catalyst and 38% for N-Bi2MoO6 compound can be observed. Interestingly, nitrogen doping on both compounds, has been beneﬁcial. However, the nitrogen doping in TiO2 makes this material more efﬁcient, which can be related to the reduction of the energy gap, which would be able to provide additional electronic states among conduction and valence bands. -20 0 20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0 N-Bi2MoO6 Bi2MoO6 N-TiO2 C/C0 Time (min) TiO2 Figure 7. Methylene blue degradation in presence of TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under visible light (VIS) energy irradiation. -20 0 20 40 60 80 100 0.0 0.2 0.4 0.6 0.8 1.0 N-Bi2MoO6 Bi2MoO6 N-TiO2 C/C0 Time (min) TiO2 Figure 7. Table 3. Atomic percent values from XPS analyses for TiO2, Bi2MoO6, N-TiO2, and N-Bi2MoO6 catalysts. Table 3. Atomic percent values from XPS analyses for TiO2, Bi2MoO6, N-TiO2, and N-Bi2MoO6 catalysts. Compound Ti2p O1s N1s Bi Mo TiO2 22.73 67.77 - - - N-TiO2 27.01 67.05 5.94 - - Bi2MoO6 - 61.86 - 24.12 14.02 N-Bi2MoO6 - 52.3 13.38 25.29 9.04 3. Photocatalytic Activity Tests 2.5. X-ray Photoelectron Spectroscopy (XPS) The reason is that usually bismuth tends to segregate close to the surface in the calcination stage, altering in some proportion the desired stoichiometry, as has been discussed in previous works [19,20]. It is important to note that the nitrogen content for N-Bi2MoO6 is larger than the value obtained for N-TiO2. However, it will be shown in the next section that N-TiO2 exhibits 326 Catalysts 2018, 8, 668 better performance in terms of catalytic activity. This means that the proportion of nitrogen has speciﬁc effects for every compound, as will be demonstrated latter. Table 3. Atomic percent values from XPS analyses for TiO2, Bi2MoO6, N-TiO2, and N-Bi2MoO6 catalysts. 3. Photocatalytic Activity Tests Methylene blue degradation in presence of TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under visible light (VIS) energy irradiation. 327 Catalysts 2018, 8, 668 y , , -20 0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Time (min) C/Co N-TiO2 N-Bi2MoO6 TiO 2 Bi2MoO6 Figure 8. Lignin degradation in presence of TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under ultraviolet (UV) energy irradiation. -20 0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Time (min) C/Co N-TiO2 N-Bi2MoO6 TiO 2 Bi2MoO6 Figure 8. Lignin degradation in presence of TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under ultraviolet (UV) energy irradiation. -20 0 20 40 60 80 100 120 0.5 0.6 0.7 0.8 0.9 1.0 Time (min) C/Co N-TiO2 N-Bi2MoO6 Bi2MoO6 TiO 2 -20 0 20 40 60 80 100 120 0.5 0.6 0.7 0.8 0.9 1.0 Time (min) C/Co N-TiO2 N-Bi2MoO6 Bi2MoO6 TiO 2 Figure 9. Lignin degradation in presence of TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under VIS energy irradiation. Time (min) Figure 9. Lignin degradation in presence of TiO2, N-TiO2, Bi2MoO6, and N-Bi2MoO6 catalysts under VIS energy irradiation. 4.1. Synthesis of TiO2 and Bi2MoO6 Catalysts 4.1. Synthesis of TiO2 and Bi2MoO6 Catalysts A TiO2 material was based on colloidal TiO2 obtained from Ti[OCH(CH3)2]4 hydrolysis and stirring the suspension to obtain the gel. The resulting material was ﬁltered and oven dried at 100 ◦C and ﬁnally calcined at 450 ◦C. The Bi2MoO6 compound was obtained using chemical coprecipitation starting with high purity Bi(NO3)3·5H2O and (NH4)6Mo7O24, diluted in distilled water and adjusting the pH 7.0 using either NH4OH or HNO3. Afterward, the precipitated Bi2MoO6 was washed and oven-dried at 100 ◦C and calcined at 400 ◦C. 4.2. Synthesis of N-doped TiO2 and N-doped Bi2MoO6 4.2. Synthesis of N-doped TiO2 and N-doped Bi2MoO6 In order to obtain N-doped Bi2MoO6 and N-doped TiO2 catalysts, both compounds separately, were added with 25 mL of a 2 M ethylene diamine solution and 150 mL of 1-hexanol, in a continuous reﬂuxing system, stirring it for 2 h. Afterward, N-doped Bi2MoO6 and N-doped TiO2 materials were ﬁltered, washed several times with water, and dried in an oven at 200 ◦C. Upon drying, both compounds yielded an intense yellowish powder. 328 Catalysts 2018, 8, 668 4.4. Photocatalytic Evaluation Catalytic activity was tested as previously described [30,31]. Brieﬂy, the reaction was carried out in a batch micro reactor provided with an oxygen ﬂow, to generate superoxide radicals and prevent electron recombination. The solution was previously stirred for 20 min without the presence of light in order to reach a stable MB or lignin absorbance on the photocatalyst surface. Then, the solution was irradiated with the source of light, magnetically stirred, and air was introduced once the reaction system was started. For MB studies, 0.1 g of every catalyst were placed into a beaker containing an aqueous solution of 0.2 g/L of MB. Regarding the lignin degradation, 0.2 g of this compound (Sigma-Aldrich, PM = 28,000 g/mol) was mixed with 15 mL of NaOH (Sigma-Aldrich, México city, México); 0.05 M was used to dissolve the samples, where then the samples were placed into a beaker containing 500 mL of deionized water. The pH = 8 was adjusted to enhance the photo catalytic reaction, according to reference [32]. For every catalyst, the reaction was achieved for 80 min. Samples were taken from the reactor system at 20 min intervals to follow the course of the reaction. Samples were centrifuged for 5 min at 220 rpm, in order to separate the catalyst from the solution to determine the progress of the reaction. A UV light lamp was used with a wavelength of 365 nm for the photocatalytic reaction. In the case of the visible light experiments, the irradiation was performed using a UV-VIS light source of 200 W, provided with a Xe arc lamp (Oriel). The concentration was monitored through a UV-VIS Hach Dr/4000u spectrophotometer at a wavelength of 289 nm for lignin and 660 nm for MB. 4.3. Characterization Details 4.3. Characterization Details The obtained compounds were characterized using X-ray diffraction (Siemens, D-5000 model), operating at 30 keV and 20 mA, with a step size 0.02◦/min from 10 to 70◦(2θ). The images were obtained in a SEM JSM-6400 JEOL Noran Instruments, at 20 keV and 10−6 Torr. The diffuse reﬂectance spectra (R) data were obtained using a UV-visible spectrophotometer (AvaSpec-2048), equipped with an integrating sphere (Ocean Optics, Mod. ISP-50-8-R-GT), equipped with a deuterium halogen light source (Mod. AvaLight DH-S-BAL). Speciﬁc surface area was measured in a Micrometrics Gemini 2060 RIG-100, model at 77 K using the BET method. For the XPS analyses, samples were excited with Al and Ka X-rays with an energy of 1486.6 eV. The spectrometer was calibrated using the Cu 2p3/2 (932.4 eV) and Cu 3p3/2 (74.9 eV) lines. Binding energy calibration was based on C 1s at 284.6 eV. 5. Discussion After the reﬂuxing process, the dissolution and recrystallization processes of a dispersed solid under reﬂux (nitrogen doping) was expected. It was substantiated by the well-known Ostwald’s ripening. Often, the Ostwald’s ripening includes the large crystals growing at the cost of smaller ones (i.e., coarsening). In the present case, the presence of the amine might have changed the expected growth by favoring the nucleation of new parties as part of the recrystallizing processes. According to Classical Nucleation Theory (CNT), a phase transition (i.e., the crystallization of a new phase within another) can be rationalized as result of two main opposed contributions. On one hand, the driving force for the process is universally identiﬁed in chemical reactions with the chemical potential difference (Δμ); on the other hand, the work spent to form the new surface was related to the new phase (associated with the interface energy, σ, and the area created, A) [34]. Hence, the Gibbs free energy for homogeneous nucleation was, ΔG = Δμ + Aσ, where spontaneous nucleation will depend on the balance between both energies’ contribution. In a heterogeneous nucleation, a surface area already exists and acts as nucleation site with lower contribution from the second term. Therefore, ΔGheterogeneous becomes a fraction of ΔGhomogeneous [35]. This explains the well-known “coarsening” phenomena. However, the coarsening should compete with nucleation of new particles, as the amine acts as a surfactant and also decreases the second term by lowering the interfacial energy. Therefore, diminishing the particle size is intuitively expected under present conditions. 329 Catalysts 2018, 8, 668 Catalysts 2018, 8, 668 Furthermore, we cannot discard a possible amorphization process as result of the nitrogen inclusions within the matrix of the solids. Recrystallization accompanied nitrogen impuriﬁcation might cause displacements of Wyckoff positions expected for the spatial groups of both solids; this could occur with the consequent strain increase that would change the Full wide half maximum (FWHM) of the reﬂections. The XPS analyses have demonstrated the incorporation of nitrogen in both N-TiO2 and N-Bi2MoO6 compounds. The nitrogen signal is located close to 400 eV. However, it is worth mentioning that some differences arise when this peak is closely analyzed for every compound. In the case of nitrogen doping for the TiO2 compound, one peak is located at 397.3 eV and another in 400.7 eV. 5. Discussion The ﬁrst is attributed to substitutional or interstitial impurities (corresponding to Ti-N bonds that substitute O by N in the lattice); the other transition, located in 400.7 eV, can be attributed to molecularly chemisorbed (superﬁcial) nitrogen. Regarding the Bi2MoO6 catalyst, something similar occurs when the high-resolution peak that corresponds to nitrogen is analyzed, as the peak found at 396 eV corresponds to shallow surface nitrogen, while the nitrogen signal at 398 eV can be assigned to interstitial nitrogen. In general, nitrogen doping has been beneﬁcial in most cases, because the nitrogen doped samples showed the best degradation performance in comparison to the pristine Bi2MoO6 or TiO2 catalysts. As expected, lignin degradation was more difﬁcult to carry out in comparison to MB, due to the complexity of the lignin molecule. However, the N-TiO2 sample showed a 90% degradation for lignin when using UV radiation. In the case of experiments carried out using visible radiation, it was found that by using an N-Bi2MoO6 compound, 30% degradation was attained; in the case of N-TiO2, however, a 35% of lignin degradation was reached. 6. Conclusions TiO2 and Bi2MoO6 N-doped photocatalysts were successfully synthesized and our XPS analyses demonstrate that nitrogen doping was carried out efﬁciently. It was also found that the N-TiO2 catalyst exhibited a better performance in terms of MB or lignin degradation. Even if N-TiO2 and N-Bi2MoO6 catalysts provide good efﬁciency for MB and lignin degradation, nitrogen doped TiO2 is the best catalyst to degrade lignin. It is demonstrated that nitrogen doping in both compounds, is an effective way to improve their degradation performance. It was also shown that nitrogen doping provides the possibility of using both catalysts under visible light. Author Contributions: Conceptualization, R.R.; methodology, V.C.; formal analysis, P.B.; investigation, V.C.; resources, J.E.; data curation, J.E.; writing—original draft preparation, R.R.; writing—review and editing, G.R.; and J.A. Funding: This research received no external funding. Funding: This research received no external funding. Acknowledgments: R. Rangel acknowledges ﬁnancial support from CIC-UMSNH under project 2018. Also thanks to W. Cahuich from Cinvestav-IPN for SEM images. Also to LANBIO of Cinvestav, Merida for allowing acces to their facilities. Acknowledgments: R. Rangel acknowledges ﬁnancial support from CIC-UMSNH under project 2018. Also thanks to W. Cahuich from Cinvestav-IPN for SEM images. Also to LANBIO of Cinvestav, Merida for allowing acces to their facilities. Conﬂicts of Interest: The authors declare not having conﬂict of interest. Conﬂicts of Interest: The authors declare not having conﬂict of interest. References Spray dried TiO2/WO3 heterostructure for photocatalytic applications with residual activity in the dark. Appl. Catal. B Environ. 2018, 226, 311–323. [CrossRef] 11. Nunes, D.; Pimentel, A.; Pinto, J.V.; Calmeiro, T.R.; Nandy, S.; Barquinha, P.; Pereira, L.; Carvalho, P.A.; Fortunato, E.; Martins, R. Photocatalytic behavior of TiO2 ﬁlms synthesized by microwave irradiation. Cat. Today 2016, 278, 262–270. [CrossRef] 12. Ramakrishnan, V.M.; Natarajan, M.; Santhanam, A.; Asokan, V.; Velauthapillai, D. Size controlled synthesis of TiO2 nanoparticles by modiﬁed solvothermal method towards effective photo catalytic and photovoltaic applications. Mater. Res. Bull. 2018, 97, 351–360. [CrossRef] 13. Rangel, R.; Cedeño, V.; Ramos-Corona, A.; Gutierrez, R.; Alvarado-Gil, J.J.; Ares, O.; Bartolo-Perez, P.; Quintana, P. Tailoring surface and photocatalytic properties of ZnO and nitrogen-doped ZnO nanostructures using microwave-assisted facile hydrothermal synthesis. Appl. Phys. 2017, 123, 552. [CrossRef] 13. Rangel, R.; Cedeño, V.; Ramos-Corona, A.; Gutierrez, R.; Alvarado-Gil, J.J.; Ares, O.; Bartolo-Perez, P.; Quintana, P. Tailoring surface and photocatalytic properties of ZnO and nitrogen-doped ZnO nanostructures g Quintana, P. Tailoring surface and photocatalytic properties of ZnO and nitrogen-doped ZnO nanostructures 14. Phuruangrat, A.; Dumrongrojthanath, P.; Thongtem, S.; Thongtem, T. Synthesis and characterization of visible light-driven W-doped Bi2MoO6 photocatalyst and its photocatalytic activities. Mater. Lett. 2017, 194, 114–117. [CrossRef] 15. Rangel, R.; Maya, R.; García, R. Novel [Ce1−XLaxO2, La2−yCeyO3]/Bi2Mo0.9W0.1O6 Catalysts for CO Oxidation at low temperature. Catal. Sci. Technol. 2012, 2, 639–642. [CrossRef] 16. Geng, B.; Wei, B.; Gao, H.; Xu, L. Ag2O nanoparticles decorated hierarchical Bi2MoO6 microspheres for efﬁcient visible light photocatalysts. J. Alloys Compd. 2017, 699, 783–787. [CrossRef] 17. Guo, J.; Shi, L.; Zhao, J.; Wang, Y.; Yuan, X. Enhanced visible-light photocatalytic activity of Bi2MoO6 nanoplates with heterogeneous Bi2MoO6-x and Bi2MoO6 core-shell structure. App. Catal. B Environ. 2018, 224, 692–704. [CrossRef] 18. Umapathy, V.; Manikandan, A.; Antony, S.A.; Ramu, P.; Neeraja, P. Structure, morphology and opto-magnetic properties of Bi2MoO6 nano-photocatalyst synthesized by sol-gel method. Trans. Nonferrous Metals Soc. China 2015, 25, 3271–3278. [CrossRef] 19. Bi, J.; Wu, L.; Li, J.; Li, Z.; Wang, X.; Fu, X. Simple solvothermal routes to synthesize nanocrystalline Bi2MoO photocatalysts with different morphologies. Acta Mater. 2007, 55, 4699–4705. [CrossRef] Bi, J.; Wu, L.; Li, J.; Li, Z.; Wang, X.; Fu, X. Simple solvothermal routes to synthesize nanocry photocatalysts with different morphologies. Acta Mater. 2007, 55, 4699–4705. [CrossRef] 20. Martínez-de la Cruz, A.; Obregón Alfaro, S. Synthesis and characterization of γ-Bi2MoO6 prepared by co-precipitation: Photoassisted degradation of organic dyes under vis-irradiation. J. Mol. References 1. Fernández, C.; Larrechi, M.S.; Callao, M.P. An analytical over-view of processes for removing organic dyes from wastewater efﬂuents. Trends Anal. Chem. 2010, 29, 1202–1211. [CrossRef] 2. Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972, 37–38. [CrossRef] [PubMed] 3. Kumar, G.; Rao, K.S.R.K. Zinc oxide based photocatalysis: Tailoring Surface–Bulk structure and related interfacial charge carrier dynamics for better environmental applications. RSC Adv. 2015, 5, 3306–3351. [CrossRef] 3. Kumar, G.; Rao, K.S.R.K. Zinc oxide based photocatalysis: Tailoring Surface–Bulk structure and related interfacial charge carrier dynamics for better environmental applications. RSC Adv. 2015, 5, 3306–3351. [CrossRef] 4. Huang, Y.; Sun, F.; Wu, T.; Wu, Q.; Huang, Z.; Su, H.; Zhang, Z. Photochemical preparation of CdS hollow microspheres at room temperature and their use in visible-light photocatalysis. J. Solid State Chem. 2011, 184, 644–648. [CrossRef] 4. Huang, Y.; Sun, F.; Wu, T.; Wu, Q.; Huang, Z.; Su, H.; Zhang, Z. Photochemical preparation of CdS hollow microspheres at room temperature and their use in visible-light photocatalysis. J. Solid State Chem. 2011, 184, 644–648. [CrossRef] 330 Catalysts 2018, 8, 668 5. Abdelkadera, E.; Nadjia, L.; Naceur, B.; Noureddine, B. SnO2 foam grain-shaped nanoparticles: Synthesis, characterization and UVA light induced photocatalysis. J. Alloys Compd. 2016, 679, 408–419. [CrossRef] g p y y p 6. Ye, Z.; Kong, L.; Chen, F.; Chen, Z.; Lin, Y.; Liu, C. A comparative study of photocatalytic activity of ZnS 6. Ye, Z.; Kong, L.; Chen, F.; Chen, Z.; Lin, Y.; Liu, C. A comparative study of photoc photocatalyst for degradation of various dyes. Optik 2018, 164, 345–354. [CrossRef] photocatalyst for degradation of various dyes. Optik 2018, 164, 345–354. [CrossRef] 7. Ullaha, S.; Ferreira-Neto, E.P.; Hazra, C.; Parveen, R.; Rojas-Mantilla, H.D.; Calegaro, M.L.; Serge-Correales, Y.E.; Rodrigues-Filho, U.P.; Ribeiro, S.J.L. Broad spectrum photocatalytic system based on BiVO4 and NaYbF4:Tm3+ upconversion particles for environmental remediation under UV-vis-NIR illumination. Appl. Catal. B Environ. 2019, 243, 121–135. [CrossRef] 8. Sanchez-Martinez, A.; Ceballos-Sanchez, O.; Koop-Santa, C.; López-Mena, E.R.; Orozco-Guareño, E.; García-Guaderrama, M. N-doped TiO2 nanoparticles obtained by a facile coprecipitation method at low temperature. Ceram. Int. 2018, 44, 5273–5283. [CrossRef] 9. Moustakas, N.G.; Kontos, A.G.; Likodimos, V.; Katsaros, F.; Boukos, N.; Tsoutsou, D.; Dimoulas, A.; Romanos, G.E.; Dionysiou, D.D.; Falaras, P. Inorganic-organic core-shell titania nanoparticles for efﬁcient visible light activated photocatalysis. Appl. Catal. B Environ. 2013, 130–131, 14–24. [CrossRef] 10. Khan, H.; Rigamonti, M.G.; Patience, G.S.; Bofﬁto, D.C. References Catal. A Chem. 2010, 320, 85–91. [CrossRef] 21. Jin, S.; Hao, H.; Gan, Y.; Guo, W.; Li, H.; Hu, X.; Hou, H.; Zhang, G.; Yan, S.; Gao, W.; et al. Preparation and improved photocatalytic activities of Ho3+/Yb3+ co-doped Bi2MoO6. Mater. Chem. Phys. 2017, 199, 107–112. [CrossRef] 22. Yu, Ch.; Wu, Z.; Liu, R.; Dionysiou, D.D.; Yang, K.; Wang, Ch.; Liu, H. Novel ﬂuorinated Bi2MoO6 nanocrystals for efﬁcient photocatalytic removal of water organic pollutants under different light source illumination. App. Catal. B Environ. 2017, 209, 1–11. [CrossRef] 23. Xing, Y.; Gao, X.; Ji, G.; Liu, Z.; Du, C. Synthesis of carbon doped Bi2MoO6 for enhanced photocatalytic performance and tumor photodynamic therapy efﬁciency. Appl. Surf. Sci. 2019, 465, 369–382. [CrossRef] 331 Catalysts 2018, 8, 668 24. Asashi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visible light photocatalysis in nitrogen-doped titanium oxide. Science 2001, 293, 269–271. [CrossRef] [PubMed] 25. Wang, J.; Tafen, D.; Lewis, J.; Hong, Z.; Manivannan, A.; Zhi, M.; Wu, N. Origin of photocatalytic activity of nitrogen-doped TiO2 nanobelts. J. Am. Chem. Soc. 2009, 131, 12290–12297. [CrossRef] [PubMed] 26. Yoshida, T.; Niimi, S.; Yamamoto, M.; Ogawa, S.; Nomoto, T.; Yagi, S. Characterization of nitrogen ion implanted TiO2 photocatalysts by XAFS and XPS. Nucl. Instrum. Methods Phys. Res. Sect. B: Beam Interact. Mater. At. 2015, 365, 79–81. [CrossRef] 27. Manova, D.; Franco-Arias, L.; Hofele, A.; Alani, I.; Kleiman, A.; Asenova, I.; Decker, U.; Marquez, A.; Mändl, S. Nitrogen incorporation during PVD deposition of TiO2:N thin ﬁlms. Surf. Coat. Technol. 2017, 312, 61–65. [CrossRef] , ; , ; , ; , ; , ; , ; , ; q , ; Mändl, S. Nitrogen incorporation during PVD deposition of TiO2:N thin ﬁlms. Surf. Coat. Technol. 2017, 312, 61–65. [CrossRef] 28. Boningaria, T.; Reddy-Inturia, S.N.; Suidan, D.; Smirniotisa, P.G. Novel one-step synthesis of nitrogen-doped TiO2 by ﬂame aerosol technique T for visible-light photocatalysis: Effect of synthesis parameters and secondary nitrogen (N) source. Chem. Eng. J. 2018, 350, 324–334. [CrossRef] en (N) source. Chem. Eng. J. 2018, 350, 324–334. [CrossR 29. Tran, V.A.; Truong, T.T.; Pham-Phan, T.A.; Nguyen, T.N.; Huynh, T.V.; Agresti, A.; Pescetelli, S.; Le, T.K.; Carlo, A.; Lund, T.; et al. Application of nitrogen-doped TiO2 nano-tubes in dye-sensitized solar cells. Appl. Surf. Sci. 2017, 399, 515–522. [CrossRef] 30. Rangel, R.; López-Mercado, G.J.; Bartolo-Pérez, P.; García, R. Nanostructured-[CeO2, La2O3, C]/T catalysts for lignin photodegradation. Sci. Adv. Mater. 2012, 4, 573–578. [CrossRef] 31. Highly Selective Photocatalytic Reduction of o-Dinitrobenzene to o-Phenylenediamine over Non-Metal-Doped TiO2 under Simulated Solar Light Irradiation Hamza M. El-Hosainy 3,4, Said M. El-Sheikh 1, Adel A. Ismail 1,2,, Amer Hakki 3, Ralf Dillert Hamada M. Killa 5, Ibrahim A. Ibrahim 1 and Detelf W. Bahnemann 3,6, Hamza M. El-Hosainy 3,4, Said M. El-Sheikh 1, Adel A. Ismail 1,2,, Amer Hakki 3, Ralf Dillert 3, Hamada M. Killa 5, Ibrahim A. Ibrahim 1 and Detelf W. Bahnemann 3,6, 1 Department of Nanomaterials and Nanotechnology, Central Metallurgical R & D Institute, Cairo 11421, Egypt; selsheikh2001@gmail.com (S.M.E.-S.); ibrahimahmedcmrdi01@gmail.com (I.A.I 1 Department of Nanomaterials and Nanotechnology, Central Metallurgical R & D Institute, Cairo 11421, Egypt; selsheikh2001@gmail.com (S.M.E.-S.); ibrahimahmedcmrdi01@gmail.com (I.A.I.) 2 Nanotechnologyand and Advanced Materials Program, Energy & Building Research Center, Kuwait Institute for Scientiﬁc Research (KISR), P.O. Box 24885, Safat 13109, Kuwait 3 Institut für Technische Chemie, Leibniz Universität Hannover, Callinstr. 3, D-30167 Hannover, Germany; hamzaelhosainy@gmail.com (H.M.E.-H.); a.hakki@abdn.ac.uk (A.H.); dillert@iftc.uni-hannover.de (R.D.) 4 Institute of Nanoscience & Nanotechnology, Kafrelsheikh University, Kafrelsheikh 33516, Egypt 5 Faculty of Science, Zagazig University, Zagazig 44519, Egypt; hamadakilla48@gmail.com 6 Laboratory “Photoactive Nanocomposite Materials”, Saint-Petersburg State University, Ulyanovskaya str. 1, Peterhof, Saint-Petersburg 198504, Russia Department of Nanomaterials and Nanotechnology, Central Metallurgical R & D Institute, Cairo 11421, Egypt; selsheikh2001@gmail.com (S.M.E.-S.); ibrahimahmedcmrdi01@gmail.com (I.A.I.) 2 Nanotechnologyand and Advanced Materials Program, Energy & Building Research Center, Kuwait Institute for Scientiﬁc Research (KISR), P.O. Box 24885, Safat 13109, Kuwait 3 Institut für Technische Chemie, Leibniz Universität Hannover, Callinstr. 3, D-30167 Hannover, Germany; hamzaelhosainy@gmail.com (H.M.E.-H.); a.hakki@abdn.ac.uk (A.H.); dillert@iftc.uni-hannover.de (R.D.) 4 Institute of Nanoscience & Nanotechnology, Kafrelsheikh University, Kafrelsheikh 33516, Egypt 5 Faculty of Science, Zagazig University, Zagazig 44519, Egypt; hamadakilla48@gmail.com 6 Laboratory “Photoactive Nanocomposite Materials”, Saint-Petersburg State University, Ulyanovskaya str. 1, Peterhof, Saint-Petersburg 198504, Russia ( ), , , 3 Institut für Technische Chemie, Leibniz Universität Hannover, Callinstr. 3, D-30167 Hannover, Germany; hamzaelhosainy@gmail.com (H.M.E.-H.); a.hakki@abdn.ac.uk (A.H.); dillert@iftc.uni-hannover.de (R.D.) 4 Institute of Nanoscience & Nanotechnology, Kafrelsheikh University, Kafrelsheikh 33516, Egypt 5 Faculty of Science, Zagazig University, Zagazig 44519, Egypt; hamadakilla48@gmail.com 6 Laboratory “Photoactive Nanocomposite Materials”, Saint-Petersburg State University, Ulyanovskaya str. 1, Peterhof, Saint-Petersburg 198504, Russia * Correspondence: aaismail@kisr.edu.kw (A.A.I.); bahnemann@iftc.uni-hannover.de (D.W.B.) Received: 21 November 2018; Accepted: 6 December 2018; Published: 9 December 2018 Abstract: Photocatalytic reduction and hydrogenation reaction of o-dinitrobenzene in the presence of oxalic acid over anatase-brookite biphasic TiO2 and non-metal-doped anatase-brookite biphasic TiO2 photocatalysts under solar simulated light was investigated. Highly Selective Photocatalytic Reduction of o-Dinitrobenzene to o-Phenylenediamine over Non-Metal-Doped TiO2 under Simulated Solar Light Irradiation Compared with commercial P25 TiO2, the prepared un-doped and doped anatase-brookite biphasic TiO2 exhibited a high selectivity towards the formation of o-nitroaniline (85.5%) and o-phenylenediamine ~97%, respectively. The doped anatase-brookite biphasic TiO2 has promoted photocatalytic reduction of the two-nitro groups of o-dinitrobenzene to the corresponding o-phenylenediamine with very high yield ~97%. Electron paramagnetic resonance analysis, Transient Absorption Spectroscopy (TAS) and Photoluminescence analysis (PL) were performed to determine the distribution of defects and the ﬂuorescence lifetime of the charge carriers for un-doped and doped photocatalysts. The superiority of the doped TiO2 photocatalysts is accredited to the creation of new dopants (C, N, and S) as hole traps, the formation of long-lived Ti3+ defects which leads to an increase in the ﬂuorescence lifetime of the formed charge carriers. The schematic diagram of the photocatalytic reduction of o-dinitrobenzene using the doped TiO2 under solar light was also illustrated in detail. eywords: photocatalysis; non-metal- doped TiO2; nitroaromatic compounds; reduction; selectivity Catalysts 2018, 8, 641; doi:10.3390/catal8120641 www.mdpi.com/journal/catalysts References Rangel, R.; García-Espinoza, J.D.; Espitia-Cabrera, I.; Alvarado-Gil, J.J.; Quintana, P.; Bartolo-Pérez, P.; Trejo-Tzab, R. Synthesis of Mesoporous of NyTi1-xCexO2-y Structures and its Visible Light Induced Photocatalytic Performance. Nano 2013, 8, 1350051–1350061. [CrossRef] 32. Dahm, A.; Lucian, A. Titanium dioxide catalyzed photodegradation of lignin in industrial efﬂuents. Ind. Eng. Chem. Res. 2004, 43, 7996–8000. [CrossRef] 33. Gazi, S.; Hung Ng, W.K.; Ganguly, R.; Putra, A.M.; Hirao, H.; Soo, H.S. Selective photocatalytic C–C bond cleavage under ambient conditions with earth abundant vanadium complexes. Chem. Sci. 2015, 6, 7130–7142. [CrossRef] [PubMed] 34. La Mer, V.K; Dinegar, R.H. Theory, Production and Mechanism of Formation of Monodispersed Hydrosols. Ind. Eng. Chem. 1950, 72, 4847–4854. 35. Wnek, W.J. The simulation of precipitation kinetics. Powder Technol. 1978, 20, 289–293. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 332 1. Introduction The great challenge for the modern chemical industry is to drive chemical reactions employing a sustainable, green, and eco-friendly process using renewable energy sources. Therefore, the development of new strategies to obtain ﬁne chemicals in a fast, clean, and efﬁcient approach is of great signiﬁcance and still requires considerable efforts. Heterogeneous photocatalysis is one of the promising and eco-friendly approaches that satisﬁes these requirements. This method is one of the important processes that encourages the use of sunlight as the source for chemical conversion processes [1–4]. Moreover, the photo-induced organic transformations by solar-driven Catalysts 2018, 8, 641; doi:10.3390/catal8120641 www.mdpi.com/journal/catalysts 333 Catalysts 2018, 8, 641 photocatalysis can produce speciﬁc products with high selectivity and lower cost compared to conventional methods [5–7]. The reduction of nitroaromatic compounds to the corresponding amino compounds is one of these photo-induced organic transformations with TiO2 is [3–6] which has attracted signiﬁcant attention because of the importance of these amino compounds as intermediates of numerous valuable compounds such as dyes and medicines [8]. Most of the reported studies in this ﬁeld have used alcohols as reaction media, hole scavenger, and hydrogen source for photocatalytic hydrogenation of nitroaromatics using TiO2 photocatalyst under inert gas atmosphere [9,10]. However, the oxidation products of alcohols may react with the reduction product of the nitro compounds which affect the selectivity and the yield of the desired amino compound. Thus, of other hole scavengers, oxalic acid is preferred since it is easily dissolved in water, which can be used as reaction media, and CO2 is the only byproduct of its oxidation [4,11–13]. TiO2 (P-25) was employed as a photocatalyst for the photoreduction of dinitro compound o-dinitrobenzene to corresponding mono and diamino compounds (o-nitroaniline and o-phenylenediamine) under solar light irradiation [14]. The results showed formation of low yield and selectivity of 55% o-nitroaniline and 30% o-phenylenediamine. y y p y On the other hand, the reduction of nitroaromatic compounds by non-metal-doped TiO2 under solar light is rarely investigated. One example is the use of N-doped TiO2 together with KI for the reduction of o-nitrophenol in the existence of methanol under solar light [3]. The results revealed that N-TiO2 has a low efﬁciency for reduction of nitroaromatic compounds under solar light irradiation. The phase type of TiO2 also has a great effect on the photocatalytic selectivity. 1. Introduction Rutile TiO2 displays higher activity and selectivity than anatase TiO2 and P25 TiO2 for selective hydrogenation of nitroaromatic compounds [15]. Furthermore, compared with rutile TiO2, P25 TiO2 has the ability to complete photocatalytic reduction of m-dinitrobenzene to the corresponding m-phenylenediamine in the deaerated aqueous iso-propanol under 4 h of UV light irradiation [16]. Herein, we report for the ﬁrst time the use of anatase-brookite biphasic TiO2 and non-metal (C, N and S)-doped anatase-brookite biphasic TiO2 for the selective hydrogenation of o-dinitrobenzene to the corresponding o-nitroaniline and o-phenylenediamine under solar simulator light, respectively. To the best of our knowledge, there are no reports showing the high selectivity and formation of o-phenylenediamine by using non-metal-doped anatase-brookite biphasic TiO2 under solar light irradiation. The expected schematic diagram and mechanism for the photocatalytic reduction of o-dinitrobenzene was also interpreted. 2. Results and Discussion Similar to our previous work [17], XRD analysis proved the formation of anatase and brookite biphase TiO2 with compositions ~75% and 25%, respectively. The surface area and pore size for un-doped and doped samples amounted to 226.2 and 85.1 m2 g−1 and 2.2 nm and 3.6 nm, respectively. Therefore, these results show the prepared TiO2 samples have a mesoporous structure. The particle size for un-doped and doped samples were 5–10 nm and 10–15 nm, respectively. The XPS analysis has proved the existence of C, N, S in the doped sample. UV-Vis. spectroscopy displayed a red absorption shift for the doped sample, reﬂecting that the band gap value of doped TiO2 sample decreased from 3.1 to 2.9 eV. Herein, the feasibility of using these materials for photocatalytic reduction of o-dinitrobenzene in aqueous oxygen-free solutions under solar light irradiation was also conducted. 2.1. Reduction of O-Dinitrobenzene to O-Nitroaniline and O-Phenylenediamine Firstly, initial experiments for reduction of o-dinitrobenzene were carried out with an aqueous solution containing TiO2 samples in the existence of oxalic acid as a hole scavenger. Figure 1 represents the time-dependent change in the concentration of o-dinitrobenzene and its photocatalytic products in its aqueous solution containing either un-doped (a) or doped TiO2 (b) in presence of oxalic acid as hole scavenger during the irradiation with solar simulated light. It is clearly observed that the concentration of o-dinitrobenzene dramatically decreases with increasing of the photoirradiation time for both T and DT samples. 9 h were needed to achieve the complete conversion of o-dinitrobenzene when employing T as the photocatalyst, whereas only 7 h were enough in the case 334 Catalysts 2018, 8, 641 of DT sample. Interestingly, the concentration of corresponding monoamino compound (o-nitroaniline) increases gradually with prolonged photoirradiation time when employing the un-doped photocatalyst (T). However, the photo-catalytically produced (o-nitroaniline) undergoes further reduction and hydrogenation to produce the corresponding diamino product (o-phenylenediamine) when DT was employed as the photocatalyst. Yield and selectivity of o-nitroaniline and o-phenylenediamine employing either T or DT samples are displayed in Figure 2a,b, respectively. In the case of T sample, o-nitroaniline is only selective as a result of hydrogenation of o-dinitrobenzene (Figure 2a). The yield and selectivity boost with the increase of photoirradiation time reaching ~88.5% within 13 h (see Figure 2a). 2. Results and Discussion On the other hand, DT photocatalyst shows a higher yield and selectivity ~97% of o-phenylenediamine as a result of reduction and hydrogenation of the two-nitro groups of the o-dinitrobenzene after only 9 h irradiation as displayed in Figure 2b. By comparison, the commercial P25 TiO2 was tested for photocatalytic reduction of o-dinitrobenzene under solar simulator light. The results showed the formation of non-selective reduction products from ≈13% o-nitroaniline and ≈86.5% o-phenylenediamine within 13 h under solar simulator light as displayed in Figure 2c. Irradiation of aqueous solution containing o-dinitrobenzene with TiO2 samples and oxalic acid under solar light produced o-nitroaniline and o-phenylenediamine as a reduction product. On the other hand, no reduction products were obtained from the aqueous solution containing o-dinitrobenzene without using photocatalyst or light and/or oxalic acid, respectively. This means that these parameters are essential for reduction of o-dinitrobenzene to the corresponding o-phenylenediamine. Figure 1. Time-dependent change in the concentration of substrate and products in aqueous solution of (a) un-doped TiO2 (T sample) Note Conc.: Concentration and (b) (C, N, S) -doped TiO2 (DT sample) in the presence of oxalic acid as hole scavenger during photoirradiation under simulated solar light, Note Conc.: Concentration and reaction conditions: 25 mg TiO2 samples, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar. Figure 1. Time-dependent change in the concentration of substrate and products in aqueous solution of (a) un-doped TiO2 (T sample) Note Conc.: Concentration and (b) (C, N, S) -doped TiO2 (DT sample) in the presence of oxalic acid as hole scavenger during photoirradiation under simulated solar light, Note Conc.: Concentration and reaction conditions: 25 mg TiO2 samples, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar. Figure 1. Time-dependent change in the concentration of substrate and products in aqueous solution of (a) un-doped TiO2 (T sample) Note Conc.: Concentration and (b) (C, N, S) -doped TiO2 (DT sample) in the presence of oxalic acid as hole scavenger during photoirradiation under simulated solar light, Note Conc.: Concentration and reaction conditions: 25 mg TiO2 samples, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar. 2. Results and Discussion 335 41 Yield and Selectivity for (a) T, (b) DT samples and (c) Selectivity for P25 compar eaction conditions: 25 mg TiO2 samples, 50 μmol o-dinitrobenzene, 250 μm nized water, Ar. Catalysts 2018, 8, 641 Figure 2. Yield and Selectivity for (a) T, (b) DT samples and (c) Selectivity for P25 compared with other samples, reaction conditions: 25 mg TiO2 samples, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar. Figure 2. Yield and Selectivity for (a) T, (b) DT samples and (c) Selectivity for P25 compared with other samples, reaction conditions: 25 mg TiO2 samples, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar. 336 Catalysts 2018, 8, 641 It is well known that the light-induced six-electron reduction of a one-nitro group of the nitroaromatic compound to the corresponding amino compound in the presence of TiO2 occurs via a sequence of electron transfer, protonation, and dehydrogenation reactions [18]. Thus, the complete reduction of two-nitro groups to two-amino groups requires twelve electrons and twelve protons. This usually occurs via the formation of hydroxylamine and/or nitrosobenzene as intermediates. However, neither nitrosobenzene nor N-phenyl hydroxylamine were detected in our cases. This might be explained by the fact that DT photocatalyst expedites the conversion of the nitro-to-amine through hydrogenation reactions (i.e., via hydrogen species derived from oxalic acid). This inhibits side reactions and facilitates selective o-phenylenediamine production. Therefore, with DT sample, photoirradiation leads to complete transformation of o-dinitrobenzene to the corresponding o-phenylenediamine with high yield and selectivity. It is also important to mention that the reduction of the second nitro group is usually more difﬁcult that the ﬁrst one and therefore it requires stronger reducing agent. The doped anatase/brookite biphasic TiO2 (DT sample) showed the high ability to complete the reduction of the two-nitro group of the dinitro compound to diamino compound (o-phenylenediamine). Therefore, compared with commercial P25 TiO2, un-doped and doped samples formed a selective reduction product from o-nitroaniline and o-phenylenediamine, respectively. 2. Results and Discussion The observed difference in the selectivity of the photocatalytic conversion of o-dinitrobenzene employing the un-doped and doped materials can be attributed to the following different factors: Firstly, this can be accredited to decrease in the band gap for the doped sample (2.9 eV) compared to the un-doped one (3.1 eV) 1 to enhancement its absorption capacity under solar simulator light (see UV-Vis. analysis, Figure S1) [17]. By non-metal doping, the O2P orbitals of TiO2 mixes with the dopants 2P orbitals of C, N and S forming a new mid-gap above the valence band of TiO2 (see Scheme 1, see XPS analysis, Figure S2) which leads to decrease its band gap. Brieﬂy, as illustrated in our previous work [17], XPS analysis revealed C, N and S are doped with TiO2 and carbon is also located on the surface (see Figure S2). Figure 2a illustrates the presence peaks of S2p with binding energy located at 168.5 eV for S6+. Besides, Figure S2b displays the N1s peaks for the doped sample. It is clear that there are two constituent peaks at around 399.7 and 401.8 eV, without the peak at 396–397 eV deﬁnitely assigned to substitutional nitrogen. In the meantime, a peak observed at around 401 eV was credited to interstitial N-doping. Moreover, non-metal dopants lead to formation of Ti3+ defects. This is due to the charge difference between N (-3) and O (-2) when N atoms bonded to Ti atoms [19]. The different electronic interactions of Ti with N anions may result in partial electron transfer from the N to Ti which may form Ti3+ defects. The formation of these Ti3+ species was veriﬁed using XPS and electron paramagnetic resonance (EPR) analyses (see Figures 3 and 4). From XPS analysis of the doped sample, the Ti2p spectrum revealed a slight negative shift of the two peaks at 457.7 eV (Ti2p3/2) and 463.4 eV (Ti2p1/2) with respect to Ti4+ (458 eV, Ti2p3/2 and 463.7 eV, Ti2p1/2) in un-doped sample (see Figure 3). This shift revealed the formation of Ti3+ species [20]. This new Ti3+ species/ defects also enhance the electronic states for TiO2 by the formation of isolated defect energy level from Ti3+ below the bottom of conduction band for TiO2 as displayed in Scheme 1 [21]. 2. Results and Discussion From EPR analysis, for the doped DT sample, the resonances at g values of less than 2.0 (1.96–1.92) are attributed to photogenerated electrons stabilized in Ti cations located at crystallization defects as shown in Figure 4. These trapped electrons could reduce Ti4+, cause the formation of Ti3+ paramagnetic species [22]. In general, the surface Ti3+ has considerably lesser g factors value than those found in bulk. Additionally, the signal shapes for surface Ti3+ is commonly broad, but in the inner (bulk) Ti3+ has a narrow axially symmetric signal [23]. Thus, the g value of 1.92 was credited to the surface Ti3+ species. Moreover, the g value of 1.943 and 1.961 was associated with the formation of bulk Ti3+ [24–26]. Therefore, from XPS and EPR analyses, we can assume that the band gap for the doped sample decreased not only by non-metal dopants (C, N, and S) but also via formation of Ti3+ defects. Subsequently, all these new-formed bands lead to enhancement of the absorption capacity of the doped sample under solar simulator light. Consequently, this leads to enhancing the photocatalytic activity of the doped sample for complete reduction of o-dinitrobenzene to the corresponding o-phenylenediamine. 337 Catalysts 2018, 8, 641 Figure 3. XPS detailed scans in the energy regions of Ti2p for T and DT samples. Figure 4. EPR spectra of T and DT samples, the DT sample recorded at dark and under UV irradiat (after 5 min) at room temperature. Instrument setting: operating at 9.41 GHz ﬁeld modulat modulation amplitude: 0.2 mT, power: 10 mW, gain: 5. Figure 3. XPS detailed scans in the energy regions of Ti2p for T and DT samples. Figure 3. XPS detailed scans in the energy regions of Ti2p for T and DT samples. Figure 4. EPR spectra of T and DT samples, the DT sample recorded at dark and under UV irradiation (after 5 min) at room temperature. Instrument setting: operating at 9.41 GHz ﬁeld modulation. modulation amplitude: 0.2 mT, power: 10 mW, gain: 5. Figure 4. EPR spectra of T and DT samples, the DT sample recorded at dark and under UV irradiation (after 5 min) at room temperature. Instrument setting: operating at 9.41 GHz ﬁeld modulation. modulation amplitude: 0.2 mT, power: 10 mW, gain: 5. Figure 4. EPR spectra of T and DT samples, the DT sample recorded at dark and under UV irradiation (after 5 min) at room temperature. 2. Results and Discussion Instrument setting: operating at 9.41 GHz ﬁeld modulation. modulation amplitude: 0.2 mT, power: 10 mW, gain: 5. Secondly, these new electronic states act as electron-hole traps which leads to an increase in their lifetime by reducing the electron-hole recombination, resulting in an enhancement of the photocatalytic activity for the doped sample. The lifetimes and charge carrier trapping can by determined using laser ﬂash photolysis [27]. The absorption time proﬁle noticed at the selected wavelength (600 nm) for the un-doped and doped samples is shown in Figure 5. It can be clearly noticed that the initial decay for the un-doped T sample is faster than that of the doped DT sample. This can be attributed to presence of the long-lived Ti3+ species. Moreover, the amount of the generated charge carriers upon irradiating the doped sample is higher than that formed in the un-doped one. Overall, the doped sample demonstrates the utmost signiﬁcant charge generation and the maximum number of hole–electron 338 Catalysts 2018, 8, 641 pairs available to participate in surface redox reactions with adsorbed species. This ﬁnding is in very good agreement with the photoreactivity results, as the above doped sample seemed to be the most photoactive for the studied photoreduction reaction. Combining the results of XPS, EPR and TAS analyses, it can be deduced that the non-metal dopants caused the formation of surface/ bulk Ti3+ in DT sample. Therefore, this new defect results in an enhancement in the absorption capacity of the material for complete the reduction of o-dinitrobenzene to the corresponding o-phenylenediamine under solar simulated light. Another evidence on the effect of doping on the charge carrier’s lifetime can be gained from the PL analysis. Figure 6 shows the PL spectra of T and DT samples at excitation wavelength (259 nm) using Xe lamp at room temperature for further evidence out ﬁnding results. The PL spectra for both samples are similar with different intensities. The PL spectra of T and DT samples revealed several emission peaks, the maximum and centered one at about 470 nm, which were referred to as the shallow energy level excitonic PL phenomenon [28]. Moreover, we can notice that the PL intensity of the doped sample decreased compared to the un-doped one. The lower PL signals for the doped sample may indicate the lower electron-hole recombination rate and the higher separation efﬁciency and this result agrees with TAS measurements. 2. Results and Discussion From the above, it can be concluded that the enhancement of the photocatalytic activity of the doped sample using solar light irradiation is not only due to the formation of a mid-gap level via non-metal dopants (C, N, S) above the valence band, but also due to the formation of isolated defect energy level (Ti3+) below the bottom of the conduction band of doped sample. This ﬁnding leads to a decrease of the band gap, decrease of the charge recombination, and increase of the life time of the charge carriers for the doped sample and consequently leads to enhancement of the complete reduction of o-dinitrobenzene to the corresponding o-phenylenediamine as shown in Scheme 1. Thirdly, hydrogen species (maybe associated with Ti3+) derived from oxalic acid facilitates the complete reduction of o-dinitrobenzene (see Scheme 1). Moreover, the formed Ti3+ atoms act as active sites for the reduction of o-dinitrobenzene and o-nitroaniline to the corresponding o-phenylenediamine. These surface Ti3+ atoms behave as an adsorption site for o-dinitrobenzene and o-nitroaniline via an electron donation and as a trapping site for photogenerated electron formed in conduction band (see Scheme 1) [15,16,29]. Therefore, these avenues facilitate the achievement of the reduction process of the two-nitro group of o-dinitrobenzene to the corresponding o-phenylenediamine. Finally, the high crystallinity and the mesoporosity leads to improvement of the photocatalytic activity of the doped sample (see Figure S3) [17]. This attribute to the doped sample was calcined at 450 ◦C. Up to calcination, the organic remains in TiO2 matrix disintegrated and formed a highly mesoporous material with pore-size diameter 3.6 nm compared with the un-doped T sample with pore-size diameter about 2.2 nm [17]. This mesoporous nature for the doped sample facilitated the adsorption capacity of the nitro aromatic compounds. Consequently, this enhanced the photocatalytic activity of the doped sample for complete reduction of o-dinitrobenzene under solar stimulator light. For all the above-mentioned reasons, it can be shown that the doped sample has versatile properties and great ability for highly selective photocatalytic reduction of o-dinitrobenzene to the corresponding o-phenylenediamine under solar simulator light. On the other hand, one of the essential parameters in photocatalytic applications in an aqueous medium is the stability and reusability of the prepared samples at the end of the reaction. The stability of the doped sample can be investigated by monitoring the UV-Vis analysis at the end of the reaction. 2.3. Photocatalysts Preparation The preparation procedure of un-doped anatase/brookite biphase TiO2 and (C, N, S)-doped anatase/brookite biphase TiO2 was published [17]. Un-doped anatase/brookite biphase TiO2 was produced via sol-gel method using TBOT as a TiO2 source and P123 as a directing agent. Then, the prepared TiO2 powder was mixed with thiourea in a weight ratio of 1:1 and calcined in a covered vessel at 450 ◦C for 1 h to get (C, N, S) -doped anatase/ brookite biphase TiO2. The obtained samples were donated as T and DT for un-doped anatase/ brookite biphase TiO2 and (C, N, S)-doped anatase/ brookite biphase TiO2. 2. Results and Discussion The results show that there is no change in the reﬂectance behavior of the doped sample as shown in Figure S1. This clearly revealed DT sample has good stability. Furthermore, the reusability for the doped sample was investigated after four cycles (Figure 7). The o-phenylenediamine yield is slightly decreased owing to a little amount of photocatalyst loss during product separation. 339 Catalysts 2018, 8, 641 Scheme 1. Suggested mechanism for the effect of non-metal dopants (C, N, S) and Ti3+ surface defects in the photocatalytic conversion of o-dinitrobenzene to the corresponding o-phenylenediamine under solar light. Scheme 1. Suggested mechanism for the effect of non-metal dopants (C, N, S) and Ti3+ surface defects in the photocatalytic conversion of o-dinitrobenzene to the corresponding o-phenylenediamine under solar light. Figure 5. Absorption time proﬁle of T and DT samples at 600 nm. Figure 6. PL spectra of T and DT samples. Figure 5. Absorption time proﬁle of T and DT samples at 600 nm. Figure 6. PL spectra of T and DT samples. 340 Catalysts 2018, 8, 641 Figure 7. Reusability and photocatalytic efﬁciency of the DT aqueous solution for reduction of o-dinitrobenzene to corresponding o-phenylenediamine under solar simulated light irradiation after 24 h, reaction conditions: 25 mg DT sample, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar Experimental section. Catalysts 2018, 8, 641 Figure 7. Reusability and photocatalytic efﬁciency of the DT aqueous solution for reduction of o-dinitrobenzene to corresponding o-phenylenediamine under solar simulated light irradiation after 24 h, reaction conditions: 25 mg DT sample, 50 μmol o-dinitrobenzene, 250 μmol oxalic acid, 5 cm3 deionized water, Ar Experimental section. 2.2. Materials and Chemicals The triblock copolymer surfactant poly (ethylene glycol)-poly (propylene glycol)- poly (ethylene glycol) (P-123, M wt. ~5800), titanium tert-butoxide Ti(OC(CH3)3)4 (TBOT), thiourea (≥99%), Triton-X 100, polyethylene glycol (10,000 MW), sodium sulfate (>99%), oxalic acid dihydrate (≥99%), dichloromethane (High-performance liquid chromatography (HPLC) grade, >99.9%), ethanol (99.8%), o-dinitrobenzene (≥99%), o-nitroaniline (98%) and o-phenylene diamine (99%) were purchased from Sigma-Aldrich, Darmstadt, Germany and were used as received. 2.4. Sample Characterization EPR spectra were recorded at room temperature on a MiniScope X-band EPR spectrometer (MS400 Magnettech GmbH, Berlin, Germany) operating at 9.41 GHz ﬁeld modulation. modulation amplitude: 0.2 mT, power: 10 mW, gain: 5. The experimental EPR spectra acquisition and simulation were carried out. The surface chemical composition of the samples was determined using X-ray Photoelectron Spectroscopy, Thermo Fisher Scientiﬁc K-Alpha XPS system (Waltham, MA, USA) with X-ray source –Al Ka micro-focused mono-chromator. The binding energies of surface adventitious carbon calibrated to the C1s peak at 284.4 ± 0.1 eV. Spectroﬂuorophotometer (RF-5301 PC, Shmidzu, Tokyo, Japan) was used to determine the photoluminescence (PL) spectra of the samples at room temperature with excitation wavelength 259 nm. Nanosecond diffuse reﬂectance transient absorption spectroscopy measurements were performed using an experimental set-up as reported previously [30]. For measurements, all powders were purged for 1 2 h with N2 prior to the measurements. 2.5. Photocatalytic Reaction Procedure 2.5. Photocatalytic Reaction Procedure The photocatalytic reactions were carried out in a sealed glass snap-cap bottle (23 mm in diameter and 75 mm in length) with contentious stirring. 25 mg TiO2 (un-doped or doped or P25) were 341 Catalysts 2018, 8, 641 suspended in 5 cm3 of deionized water containing 50 μmol of the o-dinitrobenzene and ~250 μmol oxalic acid. The mixture was stirred in the glass snap-cap bottle in the dark with Ar being purged for 15 min to remove molecular oxygen. Then the mixture was irradiated for 24 h using solar simulator (SOL1200 lamp, UV (A) was measured by Dr. K Hönle UV (A)-detector (Munich, Germany) to be 20 mW/cm2). Afterward, the excess amount of oxalic acid was neutralized by adding desired amount of NH4OH followed by extraction of the reactant and products from the aqueous phase by dichloromethane to be quantitatively and qualitatively analyzed by Gas Chromatograph-Mass Spectrometry (GC/MS) and GC with Flame Ionization Detector (GC-FID), respectively, after ﬁltration through 0.2 μm ﬁlter. Shimadzu GC/MS-QP 5000 (Tokyo, Japan) equipped with a 30 m Rxi-5ms (d = 0.32 mm) capillary column with operating temperatures programmed: injection temperature 310 ◦C, oven temperature 120 ◦C (hold 2 min) from 120 to 280 ◦C at a rate of 10 ◦C min−1, 280 ◦C (hold 15 min) in splitless mode, injection volume was 3.0 μL with helium as a carrier gas was used to qualitative analysis. Shimadzu GC 2010 (Tokyo, Japan) equipped with a Rtx-5 (d = 0.25 mm) capillary column and an FID detector was used to deﬁne the concentration of the reactant and of the products. Operating temperatures programmed: injection temperature 250 ◦C, oven temperature 70 ◦C (hold 2 min) from 70 to 280 ◦C at a rate of 10 ◦C min−1, in splitless mode. Injection volume was 2.0 μL with nitrogen as the carrier gas. The concentrations of the reactant, besides the products, were evaluated according to the calibration curves prepared with authentic standards. Author Contributions: Conceptualization, H.M.E.-H., S.M.E.-S., A.H. and A.A.I.; methodology, H.M.E.-H.; formal analysis and data curation, H.M.E.-H., A.H. and S.M.E.-S.; writing—original draft preparation, H.M.E.-H., S.M.E.-S. and A.A.I.; writing—review and editing, H.M.E.-H., R.D., S.M.E.-S., A.H. and A.A.I.; resources, D.W.B. and S.M.E.-S.; supervision, H.M.K., I.A.I. and D.W.B. Acknowledgments: This work was supported by short cycle 5, Science & Technology Development Fund in Egypt (STDF fellowship) under Grant no. ID 12282. H. El-Hosainy acknowledges Institut für Technische Chemie, Leibniz Universität Hannover, Germany for hosting him during the current research work. The publication of this article was funded by the Open Access Fund of the Leibniz Universität Hannover. Funding: This work was supported by short cycle 5, Science & Technology Development Fund in Egypt (STDF fellowship) under Grant no. ID 12282. Conﬂicts of Interest: The authors declare no conﬂicts of interest. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/8/12/641/s1, Figure S1: (a) UV-Vis absorption spectra (b) Tauc plots of modiﬁed Kubelka-Munk function of samples T, and DT before and after reusing., Figure S2: XPS detailed scans in the energy regions of (a) S2p, (b) N1s and (c) C1s for DT sample, Figure S3: (a) Low angle XRD patterns and (b) Wide angle XRD patterns for the T and DT samples. 3. Conclusions Mesostructured anatase-brookite biphase un-doped TiO2 and (C, N, S) doped anatase-brookite biphase TiO2 photocatalysts have various selectivities towards the reduction of o-dinitrobenzene in aqueous solution in the presence of oxalic acid as a hole scavenger under solar simulator light. Compared with commercial P25 TiO2, the un-doped material showed a good selectivity (85.5%) towards the reduction of just the one-nitro group, i.e., towards the production of o-nitroaniline. On the other hand, (C, N, S) the doped sample displayed a high selectivity (97%) towards the complete reduction of the two-nitro group in o-dinitrobenzene to the corresponding o-phenylenediamine. The superiority of the doped TiO2 photocatalysts is attributed to the formation of new dopants (C, N, S) as hole traps, the formation of Ti3+ defects and increase in the lifetime of the charge carriers, which leads to enhancement of the absorption capacity under solar simulator light. Furthermore, the surface Ti3+ atoms of doped TiO2 act as the adsorption site for nitroaromatics and the trapping site for photogenerated electrons formed on the conduction band. This ﬁnding leads to the acceleration of rapid nitro-to-amine reduction/hydrogenation and the complete formation of o-phenylenediamine, while suppressing side reactions. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/8/12/641/s1, Figure S1: (a) UV-Vis absorption spectra (b) Tauc plots of modiﬁed Kubelka-Munk function of samples T, and DT before and after reusing., Figure S2: XPS detailed scans in the energy regions of (a) S2p, (b) N1s and (c) C1s for DT sample, Figure S3: (a) Low angle XRD patterns and (b) Wide angle XRD patterns for the T and DT samples. Author Contributions: Conceptualization, H.M.E.-H., S.M.E.-S., A.H. and A.A.I.; methodology, H.M.E.-H.; formal analysis and data curation, H.M.E.-H., A.H. and S.M.E.-S.; writing—original draft preparation, H.M.E.-H., S.M.E.-S. and A.A.I.; writing—review and editing, H.M.E.-H., R.D., S.M.E.-S., A.H. and A.A.I.; resources, D.W.B. and S.M.E.-S.; supervision, H.M.K., I.A.I. and D.W.B. Funding: This work was supported by short cycle 5, Science & Technology Development Fund in Egypt (STDF fellowship) under Grant no. ID 12282. Acknowledgments: This work was supported by short cycle 5, Science & Technology Development Fund in Egypt (STDF fellowship) under Grant no. ID 12282. H. El-Hosainy acknowledges Institut für Technische Chemie, Leibniz Universität Hannover, Germany for hosting him during the current research work. The publication of this article was funded by the Open Access Fund of the Leibniz Universität Hannover. Conﬂicts of Interest: The authors declare no conﬂicts of interest. 342 Catalysts 2018, 8, 641 References 1. Ferry, J.L.; Glaze, W.H. Photocatalytic reduction of nitro organics over illuminated titanium dioxide: Role of the TiO2 surface. Langmuir 1998, 14, 3551–3557. [CrossRef] 2. Shiraishi, Y.; Hirai, T. Selective organic transformations on titanium oxide-based photocatalysts. J. Photochem. Photobiol. C Photochem. Rev. 2008, 9, 157–170. [CrossRef] 3. Wang, H.; Yan, J.; Chang, W.; Zhang, Z.; Wang, H. Practical synthesis of aromatic amines by photocatalytic reduction of aromatic nitro compounds on nanoparticles N-doped TiO2. Catal. Commun. 2009, 10, 989–994. [CrossRef] 4. Imamura, K.; Iwasaki, S.; Maeda, T.; Hashimoto, K.; Ohtani, B.; Kominami, H. Photocatalytic reduction of nitrobenzenes to aminobenzenes in aqueous suspensions of titanium (IV) oxide in the presence of hole scavengers under deaerated and aerated conditions. Phys. Chem. Chem. Phys. 2011, 13, 5114–5119. [CrossRef] [PubMed] 5. Hakki, A.; Dillert, R.; Bahnemann, D.W. Photocatalysis as an Auspicious Synthetic Route towards Nitrogen Containing Organic Compounds. Curr. Org. Chem. 2013, 17, 2482–2502. [CrossRef] 6. Ragaini, F.; Cenini, S.; Gasperini, M. Reduction of nitrobenzene to aniline by CO/H2O, catalysed by Ru3(CO)12/chelating diimines. J. Mol. Catal. A Chem. 2001, 174, 51–57. [CrossRef] 7. Longo, C.; Alvarez, J.; Fernández, M.; Pardey, A.J.; Moya, S.A.; Baricelli, P.; Mdleleni, M.M. Water as hydride source in the reduction of nitrobenzene to aniline catalyzed by cis-[Rh(CO)2(2-picoline)2](PF6) in aqueous 2-picoline under CO atmosphere: Kinetics study. Polyhedron 2000, 19, 487–493. [CrossRef] 8. Blaser, H.-U.; Steiner, H.; Studer, M. Selective Catalytic Hydrogenation of Functionalized Nitroarenes: An Update. ChemCatChem. 2009, 1, 210–221. [CrossRef] 9. Mahdavi, F.; Bruton, T.C.; Li, Y. Photoinduced reduction of nitro compounds on semiconductor particles. J. Org. Chem. 1993, 58, 744–746. [CrossRef] 10. Tada, H.; Ishida, T.; Takao, A.; Ito, S. Drastic enhancement of TiO2-photocatalyzed reduction of nitrobenzene by loading Ag clusters. Langmuir 2004, 20, 7898–7900. [CrossRef] 11. Li, Y.; Wasgestian, F. Photocatalytic reduction of nitrate ions on TiO2 by oxalic acid. J. Photochem. Photobiol. A 1998, 112, 255–259. [CrossRef] 12. Kominami, H.; Nakaseko, T.; Shimada, Y.; Furusho, A.; Inoue, H.; Murakami, S.; Kera, Y.; Ohtani, B. Selective photocatalytic reduction of nitrate to nitrogen molecules in an aqueous suspension of metal-loaded titanium (IV) oxide particles. Chem. Commun. 2005, 23, 2933–2935. [CrossRef] [PubMed] 13. Kominami, H.; Furusho, A.; Murakami, S.; Inoue, H.; Kera, Y.; Ohtani, B. Effective Photocatalytic Reduction of Nitrate to Ammonia in an Aqueous Suspension of Metal Loaded Titanium (IV) Oxide Particles in the Presence of Oxalic Acid. Catal. Lett. 2001, 76, 31–34. [CrossRef] 14. References Zand, Z.; Kazemi, F.; Hosseini, S. Development of chemoselective photoreduction of nitro solar light and blue LED irradiation. Tetrahedron Lett. 2014, 55, 338–341. [CrossRef] 15. Shiraishi, Y.; Togawa, Y.; Tsukamoto, D.; Tanaka, S.; Hirai, T. Highly efﬁcient and selective hydrogenation of nitroaromatics on photoactivated rutile titanium dioxide. ACS Catal. 2012, 2, 2475–2481. [CrossRef] 16. Kaur, J.; Pal, B. 100% selective yield of m-nitroaniline by rutile TiO2 and m-phenylenediamine by P25-TiO2 during m-dinitrobenzene photoreduction. Catal. Commun. 2014, 53, 25–28. [CrossRef] 17. El-Sheikh, S.M.; Zhang, G.; El-Hosainy, H.M.; Ismail, A.A.; Dionysiou, D.D. High performance, sulfur, nitrogen and carbon doped mesoporous anatase-brookite TiO2 photocatalyst for the removal of microcystin-LR under visible light irradiation. J. Hazard. Mater. 2014, 280, 723–733. [CrossRef] 18. Hakki, A.; Dillert, R.; Bahnemann, D. Photocatalytic conversion of nitroaromatic compounds in th of TiO2. Catal. Today 2009, 144, 154–159. [CrossRef] 19. Wang, J.; Li, H.; Li, H.; Zou, C. Mesoporous TiO2−xAy (A = N, S) as a visible-light-response photocatal Solid State Sci. 2010, 12, 490–497. [CrossRef] 20. Etacheri, V.; Seery, M.K.; Hinder, S.J.; Pillai, S.C. Highly Visible Light Active TiO2−xNx Heterojunction Photocatalysts. Chem. Mater. 2010, 22, 3843–3853. [CrossRef] y 21. Justicia, I.; Ordejon, P.; Canto, G.; Mozos, J.L.; Fraxedes, J.; Battiston, G.A. Designed Self-Doped Titanium Oxide Thin Films for Efﬁcient Visible-Light Photocatalysis. Adv. Mater. 2002, 14, 1399–1402. [CrossRef] 21. Justicia, I.; Ordejon, P.; Canto, G.; Mozos, J.L.; Fraxedes, J.; Battiston, G.A. Designed Self-Doped Titanium O id Thi Fil f Efﬁi Vi ibl Li h Ph l i Ad M t 2002 14 1399 1402 [C R f] 22. Yang, G.; Jiang, Z.; Shi, H.; Xiao, T.; Yan, Z. Preparation of highly visible-light active N-doped TiO2 photocatalyst. J. Mater. Chem. 2010, 20, 5301–5309. [CrossRef] 343 Catalysts 2018, 8, 641 23. Xiong, L.-B.; Li, J.-L.; Yang, B.; Yu, Y. Ti3+ in the surface of titanium dioxide: Generation, properties and photocatalytic application. J. Nanometer. 2012, 2012. [CrossRef] 24. Tan, H.; Zhao, Z.; Niu, M.; Mao, C.; Cao, D.; Cheng, D.; Feng, P.; Sun, Z. A facile and versatile method for preparation of colored TiO2 with enhanced solar-driven photocatalytic activity. Nanoscale 2014, 6, 10216–10223. [CrossRef] [PubMed] 25. Yan, Y.; Han, M.; Konkin, A.; Koppe, T.; Wang, D.; Andreu, T.; Chen, G.; Vetter, U.; Ramon Morante, J.; Schaaf, P. Slightly hydrogenated TiO2 with enhanced photocatalytic performance. J. Mater. Chem. A 2014, 2, 12708–12716. [CrossRef] 26. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Enhanced Photocatalytic Activity of Titania by Co-Doping with Mo and W Osmín Avilés-García 1, Jaime Espino-Valencia 1,, Rubí Romero-Romero 2, José Luis Rico-Cerda 1, Manuel Arroyo-Albiter 1, Dora Alicia Solís-Casados 2 and Reyna Natividad-Rangel 2, Osmín Avilés-García 1, Jaime Espino-Valencia 1,, Rubí Romero-Romero 2, José Luis Rico-C Manuel Arroyo-Albiter 1, Dora Alicia Solís-Casados 2 and Reyna Natividad-Rangel 2, 1 Facultad de Ingeniería Química, Universidad Michoacana de San Nicolás de Hidalgo, Edif. V1, Ciudad Universitaria, Morelia 58060, Michoacán, Mexico; agosmin@gmail.com (O.A.-G.); jlceri@yahoo com mx (J L R C ); albitmanuel@gmail com (M A A) jlceri@yahoo.com.mx (J.L.R.-C.); albitmanuel@gmail.com (M.A.-A) jlceri@yahoo.com.mx (J.L.R. C.); albitmanuel@gmail.com (M.A. A) 2 Centro Conjunto de Investigación en Química Sustentable, UAEMéx-UNAM, Universidad Autónoma del Estado de México, Km 14.5 Carretera Toluca-Atlacomulco, San Cayetano, Piedras Blancas, 50200 Toluca, MEX, Mexico; rubiromero99@gmail.com (R.R.R.); solis_casados@yahoo.com.mx (D.A.S.-C.) * Correspondence: rnatividadr@uaemex.mx (R.N.-R.); jespinova@yahoo.com.mx (J.E.-V.); Tel.: +52-722-2766610 (ext. 7723) (R.N.-R.); +52-443-3223500 (ext. 2002) (J.E.-V.) j y (J ) g ( ) 2 Centro Conjunto de Investigación en Química Sustentable, UAEMéx-UNAM, Universidad Autónoma del Estado de México, Km 14.5 Carretera Toluca-Atlacomulco, San Cayetano, Piedras Blancas, 50200 Toluca, MEX, Mexico; rubiromero99@gmail.com (R.R.R.); solis_casados@yahoo.com.mx (D.A.S.-C.) * Correspondence: rnatividadr@uaemex.mx (R.N.-R.); jespinova@yahoo.com.mx (J.E.-V.); Tel.: +52-722-2766610 (ext. 7723) (R.N.-R.); +52-443-3223500 (ext. 2002) (J.E.-V.) j y g 2 Centro Conjunto de Investigación en Química Sustentable, UAEMéx-UNAM, Universidad Autónoma del Estado de México, Km 14.5 Carretera Toluca-Atlacomulco, San Cayetano, Piedras Blancas, 50200 Toluca, MEX, Mexico; rubiromero99@gmail.com (R.R.R.); solis_casados@yahoo.com.mx (D.A.S.-C.) Estado de México, Km 14.5 Carretera Toluca-Atlacomulco, San Cayetano, Piedras Blancas, 50200 Tolu MEX, Mexico; rubiromero99@gmail.com (R.R.R.); solis_casados@yahoo.com.mx (D.A.S.-C.) * Correspondence: rnatividadr@uaemex.mx (R.N.-R.); jespinova@yahoo.com.mx (J.E.-V.); Tel.: +52-722-2766610 (ext. 7723) (R.N.-R.); +52-443-3223500 (ext. 2002) (J.E.-V.) g ( ) y ( ) * Correspondence: rnatividadr@uaemex.mx (R.N.-R.); jespinova@yahoo.com.mx (J.E.-V.); Tel.: +52-722-2766610 (ext. 7723) (R.N.-R.); +52-443-3223500 (ext. 2002) (J.E.-V.) Received: 22 October 2018; Accepted: 19 November 2018; Published: 6 December 2018 Abstract: Various W and Mo co-doped titanium dioxide (TiO2) materials were obtained through the EISA (Evaporation-Induced Self-Assembly) method and then tested as photocatalysts in the degradation of 4-chlorophenol. The synthesized materials were characterized by thermogravimetric analysis (TGA), Fourier transform infrared (FTIR) spectroscopy, X-ray diffraction (XRD), Raman spectroscopy (RS), N2 physisorption, UV-vis diffuse reﬂectance spectroscopy (DRS), X-ray photoelectron spectroscopy (XPS), and transmission electron microscopy (TEM). The results showed that the W-Mo-TiO2 catalysts have a high surface area of about 191 m2/g, and the presence of an anatase crystalline phase. Enhanced Photocatalytic Activity of Titania by Co-Doping with Mo and W The co-doped materials exhibited smaller crystallite sizes than those with one dopant, since the crystallinity is inhibited by the presence of both species. In addition, tungsten and molybdenum dopants are distributed and are incorporated into the anatase structure of TiO2, due to changes in red parameters and lattice expansion. Under our experimental conditions, the co-doped TiO2 catalyst presented 46% more 4-chlorophenol degradation than Degussa P25. The incorporation of two dopant cations in titania improved its photocatalytic performance, which was attributed to a cooperative effect by decreasing the recombination of photogenerated charges, high radiation absorption capacity, high surface areas, and low crystallinity. When TiO2 is co-doped with the same amount of both cations (1 wt.%), the highest degradation and mineralization (97% and 74%, respectively) is achieved. Quinones were the main intermediates in the 4-chlorophenol oxidation by W-Mo-TiO2 and 1,2,4-benzenetriol was incompletely degraded. Keywords: W-Mo dopants; titanium dioxide; nanoparticles; photocatalytic activity Catalysts 2018, 8, 631; doi:10.3390/catal8120631 www.mdpi.com/journal/catalysts References Grabstanowicz, L.R.; Gao, S.; Li, T.; Rickard, R.M.; Rajh, T.; Liu, D.-J.; Xu, T. Facile oxidative conversion of TiH2 to high-concentration Ti3+-self-doped rutile TiO2 with visible-light photoactivity. Inorg. Chem. 2013, 52, 3884–3892. [CrossRef] [PubMed] 27. Bahnemann, D.; Henglein, A.; Lilie, J.; Spanhel, L. Flash photolysis observation of the absorption spectra of trapped positive holes and electrons in colloidal titanium dioxide. J. Phys. Chem. 1984, 88, 709–711. [CrossRef] 28. Li, D.; Jia, J.; Zheng, T.; Cheng, X.; Yu, X. Construction and characterization of visible light active Pd nano-crystallite decorated and C-N-S-co-doped TiO2 nanosheet array photoelectrode for enhanced photocatalytic degradation of acetylsalicylic acid. Appl. Catal. B 2016, 188, 259–271. [CrossRef] p y g y y pp 29. Kaur, J.; Singh, R.; Pal, B. Inﬂuence of coinage and platinum group metal co-catalysis for the photocatalytic reduction of m-dinitrobenzene by P25 and rutile TiO2. J. Mol. Catal. A Chem. 2015, 397, 99–105. [CrossRef] 30. Schneider, J.; Nikitin, K.; Wark, M.; Bahnemann, D.W.; Marschall, R. Improved charge carrier separation in barium tantalate composites investigated by laser ﬂash photolysis. Phys. Chem. Chem. Phys. 2016, 18, 10719–10726. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 344 catalysts catalysts 1. Introduction Among the advanced oxidation processes (AOPs), heterogeneous photocatalysis is considered as an efﬁcient method for the degradation of organic pollutants in water and air [1,2]. Numerous semiconductors have been investigated as photocatalysts. Among the semiconductors used, titanium dioxide (TiO2) is the most promising and widely studied material for photocatalytic applications due to its chemical stability, high efﬁciency, photostability, high oxidizing power, abundance, nontoxicity, and low cost [3,4]. The main quality of TiO2 is attributed to oxidative power of hydroxyl radicals generated when the electrons are photoexcited by UV light absorption [5]. However, the recombination of the charge pair (holes and electrons) should not be ignored because it decreases the photoactivity. In order to avoid this and improve the photocatalytic activity, the photogenerated charges must be Catalysts 2018, 8, 631; doi:10.3390/catal8120631 345 Catalysts 2018, 8, 631 trapped on TiO2 surface, thus reducing recombination [6]. In this context, several techniques such as doping with metals and non-metals [7], dye sensitization [8], deposition with noble metals [9], and coupled semiconductor [10] have been assessed. Among these studies, doping has shown positive effects on titania because it gives unique electronic and structural properties that translate into better activity. Doping with non-metallic ions usually introduces energy levels above the valence band of the semiconductor for photon absorption in the visible-light region [11]. Alternatively, the use of dopant metals promotes charge transfer and separation of photogenerated charges [12]. In addition, concentration and distribution of dopant ions in TiO2 are factors that must be considered for a good photocatalytic performance [13]. The incorporation of two types of cations into TiO2 lattice and its photocatalytic performance has been reported in several studies [14,15]. Estrellan et al. [14] reported that when the TiO2 is co-doped with iron and niobium, the photocatalytic efﬁciency is improved due to synergistic actions between the doping species, which favor the e−/h+ generation and reduce the recombination rate. Shi et al. [15] reported enhanced photoactivity of titania with iron and cerium by co-doping, which is due to cooperative effects of both dopants, by broadening the absorption spectrum and retarding the recombination of the photogenerated charges. The synthesis of the catalysts was carried out by Evaporation-Induced Self-Assembly (EISA) method, which allows obtaining mesoporous structures with high surface areas [16]. 2. Results and Discussion 2. Results and Discussion 1. Introduction On the other hand, 4-chlorophenol oxidation was chosen as the reaction to evaluate the synthesized materials, because it is considered as a model molecule for photocatalytic evaluations [17]. However, it is important to mention that the removal of chlorophenols can be carried out by other techniques, such as hydrodechlorination, which has been the subject of many investigations [18,19]. 4-chlorophenol is a pollutant commonly found in the efﬂuents from industries related to insecticides, dyes, plastics, herbicides, detergents, wood preservatives, and petroleum reforming [20]. In addition, it is classiﬁed by the U.S. Environmental Protection Agency (USEPA) as a very toxic pollutant in these efﬂuents, since it causes damage to human health and aquatic environments, so its effective removal is of great interest [21]. In this work, samples of TiO2 co-doped with transition metals tungsten (W) and molybdenum (Mo) were synthesized by the EISA method. In addition, these materials have not been reported previously. The resulted solids were characterized and their photocatalytic activity evaluated in the degradation and mineralization of 4-chlorophenol. This was conducted with the main objective of demonstrating a synergistic effect in terms of improved photoactivity by co-doping compared to W-doped and Mo-doped TiO2 catalysts. Results were compared to those obtained with commercial Degussa P25 TiO2. 2.1. Photocatalysts Characterization Thermal analysis (TGA) of the synthesized titania without dopants is shown in Figure 1. A weight loss (9%) from room temperature to 200 ◦C can be observed, which is assigned to desorption of water and residual organic solvents [22]. Subsequently, a signiﬁcant weight loss (34%) between 200 ◦C and 300 ◦C is attributed to the elimination of P123 organic surfactant. Finally, the dehydroxylation process is observed above 300 ◦C. The thermogravimetric graph of the TiO2 sample co-doped with 1 wt.% tungsten and 1 wt.% molybdenum is shown in Figure 1. For comparison purposes, it can be observed that the incorporation of dopant cations into titania reduces the percentage of total weight loss, which resulted in a better thermal stability, as previously observed by Hussain et al. [23]. The applied heat treatment (400 ◦C as maximum temperature) ensured the complete elimination of the organic surfactant in all synthesized samples. On the other hand, infrared analysis (FTIR) before and after heat treatment were carried out to corroborate the P123 elimination (see Figure 2). Before thermal treatment, the characteristic signals of the surfactant by IR at 1090, 1640, 2850, and 3250 cm−1 are attributed to the C-H, H-O-H, O-H, and C-C bonds vibrations, respectively [24]. After heat treatment, these 346 Catalysts 2018, 8, 631 characteristic bands disappear, which conﬁrms the complete elimination of the organic compound incorporated during the synthesis. Figure 1. Thermogravimetric analysis of pure TiO2 and TiO2 co-doped with 1 wt.% tungsten and molybdenum (DW1M1). Figure 1. Thermogravimetric analysis of pure TiO2 and TiO2 co-doped with 1 wt.% tungsten and molybdenum (DW1M1). Figure 2. Infrared spectra of pure TiO2 sample before and after thermal treatment. Figure 2. Infrared spectra of pure TiO2 sample before and after thermal treatment. Figure 3 shows the X-ray diffraction (XRD) patterns of all synthesized samples. The materials exhibited reﬂections at 2θ ≈25.4◦, 37.8◦, 48.1◦, 54.2◦, 55.2◦, and 62.6◦, which are representative of the crystalline planes (101), (004), (200), (105), (211), and (204) of the anatase TiO2 phase, respectively [25]. The average crystallite sizes according to the Scherrer equation, as well as the lattice distortion of the samples are shown in Table 1. The crystal size of TiO2 (8.6 nm) and DM2 (8.6 nm) is higher compared to the co-doped titania samples, whose values are in the range between 6.6 nm and 8.3 nm. 2.1. Photocatalysts Characterization This suggests that the inclusion of tungsten and molybdenum into the TiO2 matrix suppresses crystalline growth. No relationship was observed due to the increase in weight percentage of a second cation by 347 Catalysts 2018, 8, 631 co-doping with respect to the average crystallite size. The lattice parameters and the unit cell volume of the materials are shown in Table 1. As can be seen, the incorporation of W and Mo in the titania by co-doping further increases the c-axis in comparison with the mono-doping, which indicates a greater expansion of the crystal lattice. Since the ion radii of Mo6+ (0.062 nm) and W6+ (0.060 nm) are very similar to that of Ti4+ (0.0605 nm), they can replace titanium within the anatase structure [26]. These changes in the lattice parameters are associated with lattice distortion due to the inclusion of dopants in the TiO2 matrix. Figure 3. X-ray diffraction patterns of all prepared catalysts. Figure 3. X-ray diffraction patterns of all prepared catalysts. Figure 3. X-ray diffraction patterns of all prepared catalysts. Table 1. Crystal size, lattice distortion and lattice parameters of the synthesized materials. Sample Average Crystallite Size (nm) Lattice Distortion (ε) Lattice Parameters a (nm) c (nm) V (nm3) TiO2 8.6 0.0183 0.377 0.931 0.132 DM2 8.6 0.0183 0.378 0.933 0.133 DW0.3M1.7 7.7 0.0207 0.378 0.982 0.140 DW0.7M1.3 6.6 0.0238 0.378 0.943 0.135 DW1M1 8.3 0.0191 0.379 0.937 0.134 DW1.3M0.7 7.3 0.0216 0.378 0.947 0.136 DW1.7M0.3 7.1 0.0224 0.379 0.959 0.138 DW2 8.1 0.0195 0.379 0.933 0.134 Figure 4 shows Raman spectra (RS) of TiO2, W-TiO2, Mo-TiO2, and W-Mo-TiO2 samples. As observed in Figure 4a, some peaks centered at 144, 197, 397, 517, and 640 cm−1 are assigned to the Eg(1), Eg(2), B1g(1), (A1g + B1g(2)), and Eg(3) vibration modes, respectively, corresponding to anatase crystalline phase [27]. This suggests that, after doping with W and/or Mo cations, the anatase phase is maintained on the TiO2 surface, which is in accordance with the XRD results. The position and intensity of the most intense Raman peak at 144 cm−1 change when titania is mono-doped and co-doped with these kinds of cations (see Figure 4b), which is attributed to changes in lattice parameters and unit cell volume by XRD analysis (see Table 1) [28]. This shift Raman is related to the increase in the c lattice parameter and the lattice expansion, which is due to the incorporation of W and/or Mo dopant ions into anatase, generating structural distortion [29]. In addition, the absence of new Raman bands in 348 Catalysts 2018, 8, 631 the spectra implies that there are no additional phases, which means that the dopants have been incorporated and they are well distributed in the crystal structure of TiO2 [30]. Figure 4. (a) Raman spectra and (b) Raman peak at 144 cm−1 of all synthesized materials. h d d h d d b f ll d l Figure 4. (a) Raman spectra and (b) Raman peak at 144 cm−1 of all synthesized mate Figure 4. (a) Raman spectra and (b) Raman peak at 144 cm−1 of all synthesized materials. The nitrogen adsorption-desorption isotherms and pore size distributions of all prepared materials are shown in Figures 5 and 6, respectively. The samples exhibited type IV isotherm, which is characteristic of mesoporous materials [31] according to the IUPAC (International Union of Pure and Applied Chemistry) classiﬁcation. Figure 3. X-ray diffraction patterns of all prepared catalysts. Type H2 hysteresis loop with monomodal pore size distribution is attributed to porous structures formed by interconnected networks. Table 2 presents the speciﬁc surface area, average pore diameter, and pore volume of all samples. The speciﬁc surface area of the co-doped samples is higher than the doped ones (DW2 and DM2), which is probably due to the incorporation of W and Mo in the crystal structure as well as the decrease in crystallinity. The maximum speciﬁc surface area of 191 m2/g is obtained with sample DW0.7M1.3. No relationship was observed between the surface area and the concentration of a second dopant cation, however, when W or Mo is incorporated in a small amount (0.33 wt.%) by co-doping (DW1.7M0.3 and DW0.3M1.7 samples), the speciﬁc surface area increases by approximately 15% with respect to mono-doped TiO2. High surface areas may offer more adsorption sites for photocatalytic reactions, which could result in better photoactivity [32]. Additionally, the average pore diameter of the co-doped catalysts remained 349 Catalysts 2018, 8, 631 constant (5.6 nm) and with a value between that of the mono-doped catalysts (4.6 nm and 6.1 nm), with the exception of sample DW1M1 that exhibited the highest average pore diameter of 6.6 nm and the maximum pore volume of 0.391 cm3/g. Figure 5. N2 adsorption-desorption isotherms of undoped, doped, and co-doped TiO2 with different weight percentages of W and Mo ions. Figure 6. Pore size distributions of undoped, doped, and co-doped TiO2 with different weight percentages of W and Mo ions. Table 2. Textural properties of TiO2, W-TiO2, Mo-TiO2, and W-Mo-TiO2 catalysts. Catalyst Speciﬁc Surface Area (m2/g) Average Pore Diameter (nm) Pore Volume (cm3/g) TiO2 144 6.1 0.328 DM2 161 6 1 0 354 Figure 5. N2 adsorption-desorption isotherms of undoped, doped, and co-doped TiO2 with different weight percentages of W and Mo ions. Figure 5. N2 adsorption-desorption isotherms of undoped, doped, and co-doped TiO2 with different weight percentages of W and Mo ions. Figure 6. Pore size distributions of undoped, doped, and co-doped TiO2 with different weight percentages of W and Mo ions. Figure 6. Pore size distributions of undoped, doped, and co-doped TiO2 with different weight percentages of W and Mo ions. Table 2. Textural properties of TiO2, W-TiO2, Mo-TiO2, and W-Mo-TiO2 catalysts. Figure 3. X-ray diffraction patterns of all prepared catalysts. Catalyst Speciﬁc Surface Area (m2/g) Average Pore Diameter (nm) Pore Volume (cm3/g) TiO2 144 6.1 0.328 DM2 161 6.1 0.354 DW0.3M1.7 181 5.6 0.362 DW0.7M1.3 191 5.6 0.365 DW1M1 172 6.6 0.391 DW1.3M0.7 185 5.6 0.352 DW1.7M0.3 188 5.6 0.352 DW2 160 4.6 0.307 yst Speciﬁc Surface Area (m2/g) Average Pore Diameter (nm) Pore Volume (cm3/g) The catalysts were analyzed by UV-vis diffuse reﬂectance spectroscopy (DRS) to estimate their band gap energy. Figure 7a shows the UV-vis absorption spectra for all materials. As can be seen, the concentration of molybdenum cations in the titania gradually shifts the absorption edge towards 350 Catalysts 2018, 8, 631 the long wavelength region. Figure 7b shows the Tauc plots from the Kubelka-Munk function [33]. The band gap and its corresponding wavelength are presented in Table 3 for all synthesized samples. TiO2 doped with 2 wt.% tungsten did not show a signiﬁcant reduction in band gap energy (3.08 eV). However, titania doped with 2 wt.% molybdenum showed a slight shift towards visible radiation absorption at 452 nm and a band gap energy of 2.74 eV. The incorporation and increase of tungsten by co-doping does not favor the band gap reduction, since the “d” orbitals of W are located into the conduction band of TiO2, making it difﬁcult to generate energy levels under this band [34]. On the other hand, the increase of molybdenum by co-doping favors the band gap reduction due to charge transfer transitions between the “d” orbitals of molybdenum and TiO2 located under the conduction band. The incorporation of this type of dopant cations (W and Mo) does not exhibit cooperative effects by co-doping towards an effective reduction of the band gap in anatase, so that its absorption of energy for electronic excitation is between the limits of mono-doped materials (DW2 and DM2 samples) at 403–452 nm. Figure 7. (a) UV-vis absorbance spectra and (b) band gap energies (Tauc plots) of synthesized catalysts. -vis absorbance spectra and (b) band gap energies (Tauc plots) of synthesized catalysts. Figure 7. (a) UV-vis absorbance spectra and (b) band gap energies (Tauc plots) of synthesized catalysts. Table 3. Band gap energy and wavelength of synthesized materials. Table 3. Band gap energy and wavelength of synthesized materials. Figure 3. X-ray diffraction patterns of all prepared catalysts. Sample Band Gap (eV) Wavelength (nm) TiO2 3.16 392 DM2 2.74 452 DW0.3M1.7 2.78 446 DW0.7M1.3 2.80 443 DW1M1 2.87 432 DW1.3M0.7 2.90 428 DW1.7M0.3 3.01 412 DW2 3.08 403 X-ray photoelectron spectroscopy (XPS) was used to determine the chemical state of the elements on the surface of the synthesized materials. Figure 8 shows the XPS spectra of Ti 2p in samples TiO2, DM2, DW2, and DW1M1. Two peaks located with binding energies at 458.1 eV and 463.8 eV are assigned to Ti 2p3/2 and Ti 2p1/2 states, respectively, and these correspond to Ti4+ into TiO2 lattice. A slight shift towards higher binding energies at 0.3 eV after doping and co-doping with W and Mo cations is evidence that these dopant metals are part of the anatase crystalline structure [35]. The O 1s XPS spectra for the sample co-doped with 1 wt.% W and 1 wt.% Mo is shown in Figure 9a. The peak at 529.8 eV is attributed to the Ti-O bond in TiO2, whereas that at 532.2 eV is assigned to surface hydroxyl groups. In photocatalysis, these hydroxyl groups play an important role, since they react with the holes generated during photoexcitation to produce hydroxyl radicals, which degrade organic compounds [36]. Figure 9b presents the XPS spectrum in the Mo 3d region for sample DW1M1. In this 351 Catalysts 2018, 8, 631 Catalysts 2018, 8, 631 ﬁgure, two main peaks, attributed to Mo 3d3/2 and Mo 3d5/2, show contributions of Mo6+ at 233.2 eV and 234.7 eV, and Mo5+ at 232.8 eV and 234.3 eV. The percentage of these two species in the catalyst surface is 77.5% and 22.5% for Mo6+ and Mo5+, respectively. The XPS spectrum shown in Figure 9c corresponds to the W 4f region. As it is observed, the surface of sample DW1M1 is composed of 91.7% W6+ and 8.3% W5+ at 36.0 eV and 34.3 eV, respectively. These signals conﬁrm the presence of both W and Mo dopants on the surface layers of TiO2, which are involved in the photocatalytic reactions. ﬁgure, two main peaks, attributed to Mo 3d3/2 and Mo 3d5/2, show contributions of Mo6+ at 233.2 eV and 234.7 eV, and Mo5+ at 232.8 eV and 234.3 eV. The percentage of these two species in the catalyst surface is 77.5% and 22.5% for Mo6+ and Mo5+, respectively. The XPS spectrum shown in Figure 9c corresponds to the W 4f region. Figure 3. X-ray diffraction patterns of all prepared catalysts. As it is observed, the surface of sample DW1M1 is composed of 91.7% W6+ and 8.3% W5+ at 36.0 eV and 34.3 eV, respectively. These signals conﬁrm the presence of both W and Mo dopants on the surface layers of TiO2, which are involved in the photocatalytic reactions. Figure 8. XPS spectra of Ti 2p for undoped, doped, and co-doped TiO2 samples with W and Mo. Figure 8 XPS spectra of Ti 2p for undoped doped and co-doped TiO2 samples with W and Mo Figure 8. XPS spectra of Ti 2p for undoped, doped, and co-doped TiO2 samples with W and Mo. Figure 8. XPS spectra of Ti 2p for undoped, doped, and co-doped TiO2 samples with W and Mo. Figure 9. XPS spectra of (a) O 1s, (b) Mo 3d, and (c) W 4f for co-doped TiO2 sample with W and Mo cations. Figure 9. XPS spectra of (a) O 1s, (b) Mo 3d, and (c) W 4f for co-doped TiO2 sample with W and Mo cations. Figure 10 shows the transmission electron micrographs, the high resolution transmission micrographs, as well as the electron diffraction patterns of mono-doped samples DM2 and DW2, 352 Catalysts 2018, 8, 631 and co-doped sample DW1M1. The TEM images in Figure 10a,d,g for mono-doped and co-doped titania with Mo and W cations show individual nanoparticles from ﬁve to 12 nm in size. The HRTEM micrographs for samples DM2 and DW2 in Figure 10b,e, exhibit interlayer spacing (0.35 nm) that corresponds to the (101) plane of anatase crystalline structure. In the HRTEM image for sample co-doped with 1 wt.% of both cations (Figure 10h); the (101) and (004) planes of anatase phase can be observed [37]. Furthermore, the SAED patterns shown in Figure 10c,f,i indicate a sequence of rings, which are characteristic of the crystalline planes (101), (004), (200), and (105). All these results are consistent with those obtained by the XRD analysis. Figure 10. TEM images, HRTEM images and SAED patterns of samples (a–c) DM2, (d–f) DW2, and (g–i) DW1M1. Figure 10. TEM images, HRTEM images and SAED patterns of samples (a–c) DM2, (d–f) DW2, and (g–i) DW1M1. In order to visualize the dopant atoms in the anatase crystalline structure, Figure 11 shows STEM-HAADF micrographs with corrected aberration of synthesized catalysts DW1, DM1, and DW1M1. Figure 3. X-ray diffraction patterns of all prepared catalysts. It can be observed in Figure 11a (W-mono-doped TiO2) that the tungsten atoms are well distributed and they are part of well-deﬁned atomic columns within the anatase phase. However, when titania is mono-doped with molybdenum atoms (Figure 11b), these can have two locations: (i) As being part of the crystalline structure with a good distribution (yellow circles), and (ii) forming some dispersed agglomerates that are not part of the atomic structure (orange circles). On the other hand, when TiO2 is co-doped with W and Mo, the dispersion of both atoms inside the crystalline lattice is favored (atomic agglomerations disappear) (see Figure 11c). 353 Catalysts 2018, 8, 631 n-corrected STEM-HAADF images of TiO2 mono-doped with (a) W or (b) Mo, and th cations. Figure 11. Aberration-corrected STEM-HAADF images of TiO2 mono-doped with (a) W or (b) Mo, and (c) co-doped with both cations. 2.2. Photocatalytic Tests 2.2. Photocatalytic Tests In this case, the potentials of the Mo6+/Mo5+ and W6+/W5+ pairs are +0.4 V and −0.03 V, respectively [41,42], and they are more positive than the potential of the TiO2 conduction band (−0.51 V) [43], so their electronic capture is favored when both are present in titania. The prepared catalysts showed no 4CP removal due to adsorption, and likewise, the photolysis effect was negligible. degradation of 4-chlorophenol (4CP). Figure 12 shows the photodegradation proﬁles with 100 min of reaction as total time. It can be seen that pure synthesized TiO2 and Degussa P25 exhibit the lowest percentages of 4CP degradation. With respect to the co-doped TiO2 samples, these show better degradation than the mono-doped materials (DM2 and DW2), which might be because of synergistic effects by the two doping species, thus improving their photoactivity. In addition, the co-doped samples exhibited higher surface areas and smaller crystallite sizes than the mono-doped catalysts, which increase the sites density for substrates adsorption during photocatalytic activity [38]. The degradation is gradually improved when the amount of tungsten increases from 0.3 wt.% to 1 wt.%, achieving more than 97% for the DW1M1 sample. Concentrations above 1 wt.% W in co-doped TiO2 do not favor photocatalytic activity, which results in less degradation. While the samples co-doped with high molybdenum concentration (DW0.3M1.7 and DW0.7M1.3) exhibited higher radiation absorption, the signiﬁcant improvement in photoactivity is very likely due to suitable concentration of both dopants. The co-doped TiO2 with W:Mo = 1:1 ratio exhibits a synergistic effect between the species towards the best photocatalytic activity, since it presents a good radiation absorption capacity (see Figure 7) that favors the generation of e−/h+ pairs, as well as the ability to reduce the recombination processes of the photogenerated charges [39]. The process to inhibit the recombination (Equation (2)) after the e−/h+ pair generation (Equation (1)) is shown in Equations. (3) and (4). The electrons can be trapped by the most stable states of W and Mo, the W6+ and Mo6+ species, respectively, reaching a local charge compensation that leads to a longer lifetime of the generated holes [40]. In this way, the holes can be trapped by the adsorbed H2O or by the -OH groups on the TiO2 surface (Equations (5) and (6)) (see Figure 9a), allowing the generation of more •OH radicals, which are responsible for the oxidation of organic compounds (Equation (7)). 2.2. Photocatalytic Tests All synthesized samples of pure titania, mono-doped, and co-doped with molybdenum and tungsten cations, as well as the commercial catalyst Degussa P25 were evaluated in the photocatalytic 354 Catalysts 2018, 8, 631 degradation of 4-chlorophenol (4CP). Figure 12 shows the photodegradation proﬁles with 100 min of reaction as total time. It can be seen that pure synthesized TiO2 and Degussa P25 exhibit the lowest percentages of 4CP degradation. With respect to the co-doped TiO2 samples, these show better degradation than the mono-doped materials (DM2 and DW2), which might be because of synergistic effects by the two doping species, thus improving their photoactivity. In addition, the co-doped samples exhibited higher surface areas and smaller crystallite sizes than the mono-doped catalysts, which increase the sites density for substrates adsorption during photocatalytic activity [38]. The degradation is gradually improved when the amount of tungsten increases from 0.3 wt.% to 1 wt.%, achieving more than 97% for the DW1M1 sample. Concentrations above 1 wt.% W in co-doped TiO2 do not favor photocatalytic activity, which results in less degradation. While the samples co-doped with high molybdenum concentration (DW0.3M1.7 and DW0.7M1.3) exhibited higher radiation absorption, the signiﬁcant improvement in photoactivity is very likely due to suitable concentration of both dopants. The co-doped TiO2 with W:Mo = 1:1 ratio exhibits a synergistic effect between the species towards the best photocatalytic activity, since it presents a good radiation absorption capacity (see Figure 7) that favors the generation of e−/h+ pairs, as well as the ability to reduce the recombination processes of the photogenerated charges [39]. The process to inhibit the recombination (Equation (2)) after the e−/h+ pair generation (Equation (1)) is shown in Equations. (3) and (4). The electrons can be trapped by the most stable states of W and Mo, the W6+ and Mo6+ species, respectively, reaching a local charge compensation that leads to a longer lifetime of the generated holes [40]. In this way, the holes can be trapped by the adsorbed H2O or by the -OH groups on the TiO2 surface (Equations (5) and (6)) (see Figure 9a), allowing the generation of more •OH radicals, which are responsible for the oxidation of organic compounds (Equation (7)). On the other hand, the redox potentials of the doping species involved should not be left aside. 2.2. Photocatalytic Tests On the other hand, the redox potentials of the doping species involved should not be left aside. In this case, the potentials of the Mo6+/Mo5+ and W6+/W5+ pairs are +0.4 V and −0.03 V, respectively [41,42], and they are more positive than the potential of the TiO2 conduction band (−0.51 V) [43], so their electronic capture is favored when both are present in titania. The prepared catalysts showed no 4CP removal due to adsorption, and likewise, the photolysis effect was negligible. W-Mo-TiO2 + hv →e−+ h+ (1) e−+ h+ →TiO2 (2) W6+ + e−→W5+ (3) Mo6+ + e−→Mo5+ (4) H2O + h+ →•OH + H+ (5) —OH + h+ →•OH (6) •OH + 4CP →→CO2 + H2O + Cl− (7) W-Mo-TiO2 + hv →e−+ h+ (1) e−+ h+ →TiO2 (2) W6+ + e−→W5+ (3) Mo6+ + e−→Mo5+ (4) H2O + h+ →•OH + H+ (5) —OH + h+ →•OH (6) •OH + 4CP →→CO2 + H2O + Cl− (7) •OH + 4CP →→CO2 + H2O + Cl− (7) The photocatalytic degradation proﬁles of 4CP were adjusted to a pseudo-ﬁrst-order kinetics [44]. Apparent kinetic constants as well as half-life for all catalysts are summarized in Table 4. It can be seen that the maximum degradation is achieved by using the titania co-doped at 1 wt.% of both cations with a 4CP half-life of approximately 20 min, where about 75% of the initial concentration of 4CP is degraded in 40 min. 355 Catalysts 2018, 8, 631 Figure 12. 4-Chlorophenol photodegradation proﬁles over TiO2, Mo-TiO2, W-TiO2, W-Mo-TiO2, and Degussa P25 catalysts. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Table 4. Removal percentages, apparent kinetic constant, half-time and regression coefﬁcient for 4CP degradation. Figure 12. 4-Chlorophenol photodegradation proﬁles over TiO2, Mo-TiO2, W-TiO2, W-Mo-TiO2, and Degussa P25 catalysts. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Table 4. Removal percentages, apparent kinetic constant, half-time and regression coefﬁcient for 4CP degradation. Figure 12. 4-Chlorophenol photodegradation proﬁles over TiO2, Mo-TiO2, W-TiO2, W-Mo-TiO2, and Degussa P25 catalysts. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Table 4. Removal percentages, apparent kinetic constant, half-time and regression coefﬁcient for 4CP degradation. 2.2. Photocatalytic Tests Catalyst 4CP Degradation (%) k × 102 (min−1) t1/2 (min) r2 DM2 76 1.31 52.8 0.97 DW0.3M1.7 86 1.98 34.9 0.99 DW0.7M1.3 93 2.75 25.2 0.99 DW1M1 97 3.49 19.9 0.99 DW1.3M0.7 93 2.73 25.4 0.99 DW1.7M0.3 92 2.51 27.6 0.99 DW2 75 1.26 54.9 0.95 TiO2 67 1.04 66.8 0.90 Degussa P25 51 0.67 102.9 0.97 The percentage of mineralization, which is estimated by the total organic carbon content (TOC), is an important parameter of the degree of deep or complete oxidation of 4-chlorophenol to carbon dioxide and water (Equation (8)) [45]. C6H5ClO + W-Mo-TiO2 + hv →Intermediates →CO2 + H2O + Cl− (8) (8) Figure 13 shows the 4CP mineralization percentages using the TiO2 catalysts mono and co-doped with W and/or Mo. W-Mo-co-doped TiO2 with a tungsten weight ratio of W/(W + Mo) = 0.5 showed the highest percentage of mineralization achieving more than 74%. On the other hand, the mono-doped samples (DM2 and DW2) exhibited average mineralization percentages of 50% and 53%, respectively. 356 Catalysts 2018, 8, 631 atalysts 2018, 8, 631 Figure 13. TOC reduction percentages and amount of tungsten dopant on mono-doped and co-doped TiO2. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Figure 13. TOC reduction percentages and amount of tungsten dopant on mono-doped and co-dope TiO2. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Figure 13. TOC reduction percentages and amount of tungsten dopant on mono-doped and co-doped TiO2. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Table 5 shows the photocatalytic performance of the synthesized sample DW1M1 in comparison with other reported co-doped TiO2 materials. It can be seen that, although the reaction conditions are a little different, the synthesized material of W-Mo-TiO2 exhibits a higher degradation/mineralization percentage of 4-chlorophenol, as well as enhanced kinetic parameters. Table 5. Photocatalytic performance of W-Mo-co-doped TiO2 compared to other reported co-doped TiO2 materials. TiO2 materials. Photo-Catalyst [4CP]0 × 105 (mol/L) Radiation 4CP Degradation (%) 4CP Mineralization (%) k × 102 (min−1) Ref. 2.2. Photocatalytic Tests W-Mo-TiO2 15.56 UV 97 (100 min) 74 (100 min) 3.49 - B-N-TiO2 5.00 UV 98 (120 min) - 3.41 [46] N-F-TiO2 7.78 VL 72 (300 min) - - [47] C-W-TiO2 50.00 VL - 57 (300 min) - [48] The intermediate compounds formed during the photocatalytic degradation of 4-chlorophenol were determined by HPLC. Figure 14 shows the concentration proﬁles of the main intermediaries when the catalyst DW1M1 is used. The name and structure of these identiﬁed intermediaries are summarized in Table 6. The aromatic ring of 4-chlorophenol can be attacked by hydroxyl radicals in several positions, and for this reason some mechanistic pathways are possible towards the formation of these species (see Figure 15). Furthermore, the generation of hydroxylated compounds is clear evidence of the involvement of hydroxyl radicals in the degradation of 4CP [49]. 357 Catalysts 2018, 8, 631 Figure 14. Concentration proﬁles of intermediates formed during 4CP photodegradation over W-Mo-TiO2. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Table 6. Intermediate compounds formed in the photocatalytic degradation of 4-chlorophenol over TiO2 co-doped with W and Mo. Figure 14. Concentration proﬁles of intermediates formed during 4CP photodegradation over W-Mo-TiO2. [4CP]0 = 15.56 × 10−5 M, Ccat = 2 × 10−4 kg/L, pH0 = 2, T = 298 K. Table 6. Intermediate compounds formed in the photocatalytic degradation of 4-chlorophenol over TiO2 co-doped with W and Mo. Table 6. Intermediate compounds formed in the photocatalytic degradation of 4-chlorophenol over TiO2 co-doped with W and Mo. Intermediate Compound ID Name Molecular Structure HQ Hydroquinone 1,4-benzenediol 1,4-dihydroxybenzene BQ Benzoquinone 1,4-benzoquinone p-benzoquinone Quinone 4CC 4-chlorocatechol 4-chloro-1,2-benzenediol BT 1,2,4-benzenetriol Hydroxyhydroquinone DHB 1,2-dihydroxybenzene 1,2-benzenediol Catechol Pyrocatechol P Phenol Hydroxybenzene 4CR 4-chlororesorcinol 1,3-dihydroxy-4-chlorobenzene Intermediate Compound ID Name Molecular Structure HQ Hydroquinone 1,4-benzenediol 1,4-dihydroxybenzene BQ Benzoquinone 1,4-benzoquinone p-benzoquinone Quinone 4CC 4-chlorocatechol 4-chloro-1,2-benzenediol BT 1,2,4-benzenetriol Hydroxyhydroquinone DHB 1,2-dihydroxybenzene 1,2-benzenediol Catechol Pyrocatechol P Phenol Hydroxybenzene 4CR 4-chlororesorcinol 1,3-dihydroxy-4-chlorobenzene Molecular Structure Molecular Structure 358 358 Catalysts 2018, 8, 631 Figure 15. Proposed photocatalytic degradation pathway of 4-chlorophenol by W-Mo-TiO2. Figure 15. Proposed photocatalytic degradation pathway of 4-chlorophenol by W-Mo-TiO2. According to Figure 15, the oxidation process of 4CP is favored towards the quinone route (routes I and II) with maximum concentrations in the ﬁrst minutes of reaction. Hydroquinone is generated by hydroxylation and dechlorination of 4CP in para position of the aromatic ring, which can be oxidized to benzoquinone. 2.2. Photocatalytic Tests Additionally, benzoquinone can be formed from the hydroxyphenyl radical via route I [50]. 4CP hydroxylation in ortho position generates 4-chlorocatechol (route III). The para-dechlorination of 4CC forms the 1,2-dihydroxyphenyl radical [51], which, when combined with a proton and an electron, produces 1,2-dihydroxybenzene. The peak with the highest concentration of 1,2,4-benzenetriol is found at 60 min, because its formation is after hydroxylation of hydroquinone (route II) or 4-chlorocatechol (route III). The oxidation process of 4CP can be initiated by the attack of the C-Cl bond (route IV) to form the hydroxyphenyl radical [52], which subsequently combined with a proton generates phenol. On the other hand, DHB (1,2-dihydroxybenzene) can also be formed by ortho-hydroxylation of phenol. However, this oxidation route is not favored due to low concentrations of phenol and 1,2-dihydroxybenzene observed during photocatalytic tests (see Figure 14). Similarly, meta-hydroxylation of 4CP generates very little 4-chlororesorcinol concentration (route V). Finally, the aromatic ring of these intermediate compounds is opened by subsequent oxidations to form aliphatic species (organic acids), which are later mineralized to carbon dioxide and water according to Equation (8) [53]. Titanium (IV) butoxide [Ti(OC4H9)4, 97%], ammonium metatungstate [(NH4)6H2W12O40·xH2O] and ammonium heptamolybdate [(NH4)6Mo7O24·4H2O] served as metal precursors Ti, W, and Mo, respectively. Pluronic P123 [EO20PO70EO20] was used as a surfactant. Ethanol (C2H6O, 99.5%) and 3.2. Preparation of Undoped, Doped and Co-Doped TiO2 3.2. Preparation of Undoped, Doped and Co-Doped TiO2 3.2. Preparation of Undoped, Doped and Co-Doped TiO2 Pure TiO2 was prepared by mixing ethyl alcohol and Ti(OC4H9)4 under stirring for 15 min. This mixture was added to the P123 surfactant with constant stirring for 30 min and ﬁnally nitric acid was dropwise incorporated to the reaction media. The homogenous mixture was kept under vigorous stirring for 3 h. The molar ratio of reagents Ti(OBu)4:C2H6O:P123: HNO3 was maintained constant at 1:18.71:0.018:3.55. To prepare co-doped TiO2, tungsten and/or molybdenum precursors were added before incorporating the HNO3. The resulting solution was dried in a rotary evaporator until a solid was formed. This was subsequently calcined at 300 ◦C for 1 h (surfactant removal) and then at 400 ◦C for 4 h with a heating rate of 1 ◦C/min. Various samples with different weight percentages of dopant were synthetized. However, the total amount of dopant (s) was kept constant at 2 wt.% with respect to titania. The catalysts were labeled as DWAMB. W stands for tungsten and M for molybdenum. In addition, A and B are the weight percentages in samples. The DM2 and DW2 catalysts contain 2 wt.% molybdenum and 2 wt.% tungsten, respectively. Commercial TiO2 (Degussa P25) was used for comparison. 3.1. Chemicals Titanium (IV) butoxide [Ti(OC4H9)4, 97%], ammonium metatungstate [(NH4)6H2W12O40·xH2O] and ammonium heptamolybdate [(NH4)6Mo7O24·4H2O] served as metal precursors Ti, W, and Mo, respectively. Pluronic P123 [EO20PO70EO20] was used as a surfactant. Ethanol (C2H6O, 99.5%) and 359 Catalysts 2018, 8, 631 nitric acid (HNO3, 70%) were employed as solvents and catalysts, respectively. 4-chlorophenol (ClC6H4OH, 99%) and deionized water were used during the photocatalytic tests. All reagents were of analytical grade and supplied by Sigma-Aldrich (St. Louis, MO, USA). Methanol and acetonitrile (HPLC grade) were used to prepare the mobile phases during the chromatographic analysis. 3.3. Characterization of Photocatalysts The temperature for the removal of the surfactant during heat treatment was determined by thermogravimetry (TGA 4000, Pyris, Perkin Elmer, Waltham, MA, USA) with N2 gas ﬂow at 20 mL/min in a range of 30 to 700 ◦C (5 ◦C/min). Before and after heat treatment, Infrared spectroscopy (FT-IR, Perkin Elmer Spectrum Two, Waltham, MA, USA) was used. The FT-IR spectra were acquired in the range of 4000–1000 cm−1 and using a resolution of 1 cm−1. All synthesized samples were analyzed with an X-ray diffractometer (Bruker Advance D8, Billerica, MA, USA) to determine the crystalline properties. Cu Kα radiation at 1.5406 Å was used as the X-ray source. The voltage and current applied were 30 kV and 25 mA, respectively. The diffraction patterns were determined over 2θ range of 20◦–70◦with a resolution of 0.02◦2θ/16 s. The average crystallite size (D) was estimated by the Scherrer equation D = 0.9λ/β cosθ, the lattice distortion (d) by d = β/4 tgθ, and the lattice parameters according to the Bragg equation 2dhkl sinθ = λ, where λ is the applied wavelength, θ is the Bragg angle, and β is the FWHM value. Raman spectroscopy was also used to determine the structural properties. For this purpose, a Renishaw micro-Raman spectrometer, provided with a laser of 514 nm, was used. Average pore diameters and speciﬁc surface areas of the synthesized materials were obtained through nitrogen adsorption–desorption isotherms on a Quantachrome (Boynton Beach, FL, USA) Autosorb-1 at 77 K by BJH and BET methods, respectively. Before measurements, the materials were degassed out for 2 h at 200 ◦C. The band gaps of all samples were determined through UV-vis Diffuse Reﬂectance Spectroscopy (DRS) using a Perkin Elmer spectrophotometer (Lambda 35, Waltham, MA, USA) equipped with an integration sphere (Labsphere rsape-20). The Kubelka-Munk function α = (1 −R)2/2R was used to transform the reﬂectance spectra into absorption spectra, which were then used to estimate the band gap energies by constructing Tauc plots of hv vs. (αhv)2 and extrapolating the linear part to (αhv)2 equal to zero. The chemical states of the surface elements present in the materials were examined by X-ray photoelectron spectroscopy (JEOL JPS-9200 spectrometer, Akishima, TYO, Japan) with an Al Kα X-ray source. The energy of C 1s at 284.6 eV was used as a reference for charge correction during the estimation of the binding energies. Acknowledgments: Authors are grateful to CCIQS from UAEM-UNAM and CIMAV-Mexico for the granted support. The technical support of Gustavo López Téllez, Alfredo Rafael Vílchis Néstor, Uvaldo Hernández, Citlalit Martínez and Carlos Elías Ornelas Gutiérrez is also acknowledged. 3.4. Photocatalytic activity and analysis The photoactivity of all catalysts was evaluated in the degradation of 4-chlorophenol. The constant reaction conditions used were 2 × 10−4 kg/L of catalyst loading, 20 × 10−6 kg/L of 4CP and temperature of 298 K. A UV lamp was used as a radiation source with a wavelength and intensity of 254 nm and 4500 μW/cm2, respectively. The 4-chlorophenol solution was adjusted to an initial pH of 2 and stirred at 1000 rpm to keep the catalyst in suspension. Aliquots were withdrawn from the reactor every 20 min and subsequently centrifuged to remove the catalyst. The solutions were then analyzed on a UV-vis spectrophotometer (Perkin Elmer Lambda 25, Waltham, MA, USA) to determine the amount of 4-chlorophenol at 280 nm, according to its absorbance peak. The total organic carbon concentration in the solution was measured by using a Shimadzu TOC-L analyzer to evaluate the 4CP mineralization. The identiﬁcation of intermediate compounds during 4CP degradation was carried out by HPLC (Varian 230, isocratic mode, Santa Clara, CA, USA). Ascentis Express C18 column (2.7 μm, 3 cm × 4.6 mm, Sigma-Aldrich) with a mobile phase methanol/water (20/80 v/v, 1.0 mL min-1) and Eclipse XDB-C18 column (5 μm, 15 cm × 4.6 mm, Agilent, Santa Clara, CA, USA) with a mobile phase acetonitrile/water (10/90 v/v, 0.6 mL/min) were used at 25 ◦C. 3.3. Characterization of Photocatalysts 360 360 Catalysts 2018, 8, 631 Catalysts 2018, 8, 631 High resolution-transmission electron microscopy (HR-TEM, JEOL-2100 at 200 kV, Akishima, TYO, Japan) and selected area electron diffraction (SAED) were used to determine the particle size and crystalline structure. High-angle annular dark-ﬁeld scanning microscopy with a spherical aberration corrector (ac-HAADF-STEM, JEOL 2200FS+CS, Akishima, TYO, Japan) was used to visualize the doping atoms in the crystalline structures. 3.4. Photocatalytic activity and analysis nding: This research was funded by PROMEP-Mexico (ﬁnancial support through project 103.5/13/S257) and NACYT-Mexico (project 269093). Scholarship 378292 by CONACYT. Author Contributions: J.E.-V. and R.N.-R. designed the experiments and supervised the project; O.A.-G. performed the experiments and analyzed the data; D.A.S.-C. contributed with characterization of materials; O.A.-G. wrote the manuscript; All authors contributed to a review of the manuscript before submission. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Finˇcur, N.L.; Krsti´c, J.B.; Šibul, F.S.; Šoji´c, D.V.; Despotovi´c, V.N.; Bani´c, N.D.; Agbaba, J.R.; Abramovi´c, B.F. Removal of alprazolam from aqueous solutions by heterogeneous photocatalysis: Inﬂuencing factors, intermediates, and products. Chem. Eng. J. 2017, 307, 1105–1115. [CrossRef] 2. Verbruggen, S.W. TiO2 photocatalysis for the degradation of pollutants in gas phase: From morphological design to plasmonic enhancement. J. Photochem. Photobiol. C Photochem. Rev. 2015, 24, 64–82. [CrossRef] 3. Linsebigler, A.L.; Lu, G.; Yates, J.T. Photocatalysis on TiO2 surfaces: Principles, mechanisms, and selected results. Chem. Rev. 1995, 95, 735–758. [CrossRef] 4. Fujishima, A.; Zhang, X.; Tryk, D.A. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2008, 63, 515–582. [CrossRef] 4. Fujishima, A.; Zhang, X.; Tryk, D.A. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2008, 63, 515–582. [CrossRef] 5. Kong, M.; Li, Y.; Chen, X.; Tian, T.; Fang, P.; Zheng, F.; Zhao, X. Tuning the relative concentration ratio of bulk defects to surface defects in TiO2 nanocrystals leads to high photocatalytic efﬁciency. J. Am. Chem. Soc. 2011, 133, 16414–16417. [CrossRef] 6. Liqiang, J.; Honggang, F.; Baiqi, W.; Dejun, W.; Baifu, X.; Shudan, L.; Jiazhong, S. Effects of Sn dopant on the photoinduced charge property and photocatalytic activity of TiO2 nanoparticles. Appl. Catal. B Environ. 2006, 62, 282–291. [CrossRef] 7. Yang, G.; Yan, Z.; Xiao, T. Low-temperature solvothermal synthesis of visible-light-responsive S-doped TiO2 nanocrystal. Appl. Surf. Sci. 2012, 258, 4016–4022. [CrossRef] 8. Ding, H.; Sun, H.; Shan, Y. Preparation and characterization of mesoporous SBA-15 supported dye-sensitized TiO2 photocatalyst. J. Photochem. Photobiol. A Chem. 2005, 169, 101–107. [CrossRef] g p p pp TiO2 photocatalyst. J. Photochem. Photobiol. A Chem. 2005, 169, 101–107. [CrossRef] 9. Wu, Y.; Liu, H.; Zhang, J.; Chen, F. Enhanced photocatalytic activity of nitrogen-doped titania by deposi with gold. J. Phys. Chem. C 2009, 113, 14689–14695. [CrossRef] 9. Wu, Y.; Liu, H.; Zhang, J.; Chen, F. Enhanced photocatalytic ac with gold. J. Phys. Chem. C 2009, 113, 14689–14695. [CrossRef] 10. Bayati, M.R.; Golestani-Fard, F.; Moshfegh, A.Z. Photo-degradation of methelyne blue over V2O5–T nano-porous layers synthesized by micro arc oxidation. Catal. Lett. 2010, 134, 162–168. [CrossRef] 11. Yang, G.; Jiang, Z.; Shi, H.; Xiao, T.; Yan, Z. Preparation of highly visible-light active N-doped TiO2 photocatalyst. J. Mater. Chem. 2010, 20, 5301–5309. [CrossRef] 12. Wilke, K.; Breuer, H.D. The inﬂuence of transition metal doping on the physical and photocatalytic properties of titania. J. Photochem. Photobiol. A Chem. 1999, 121, 49–53. 4. Conclusions TiO2 nanoparticles were prepared by mono and co-doping with molybdenum and tungsten cations by the Evaporation-Induced Self-Assembly (EISA) method. All synthesized materials presented anatase crystalline phase and larger speciﬁc surface areas than the commercial photocatalyst Degussa P25. The presence of a second cation by co-doping increased the surface area and decreased the crystallinity, as well as changes in lattice parameters due to structural distortion, since the doping species were successfully incorporated into TiO2 lattice. The dopant cations exhibited oxidation states with valence 5+ and 6+ in the titania. Low tungsten concentrations by co-doping favored the reduction of the anatase band gap. Raman analysis showed that no additional phases by co-doping were created, and both dopants are distributed inside the anatase. All synthesized materials exhibited photocatalytic activity in the oxidation of 4-chlorophenol and all of them outperformed commercial TiO2 P25. Synergistic effect with an appropriate concentration of both dopant cations improved photocatalytic activity by reducing the recombination of photogenerated charges and by increasing the absorption of radiation to promote their generation. The experiments performed at ambient temperature and during 100 min showed that titania, co-doped with 1 wt.% W and 1 wt.%, Mo exhibited the best photoactivity among the catalysts studied, with 97% and 74% of degradation and mineralization of 4-chlorophenol, respectively. The oxidation route of 4-chlorophenol was favored towards quinones as main intermediates via hydroxyl radicals. Remaining TOC (26%) is attributed to 1,2,4-benzenetriol as secondary intermediate and aliphatic compounds. Author Contributions: J.E.-V. and R.N.-R. designed the experiments and supervised the project; O.A.-G. performed the experiments and analyzed the data; D.A.S.-C. contributed with characterization of materials; O.A.-G. wrote the manuscript; All authors contributed to a review of the manuscript before submission. Conﬂicts of Interest: The authors declare no conﬂict of interest. 361 Catalysts 2018, 8, 631 References [CrossRef] 13. Choi, W.; Termin, A.; Hoffmann, M.R. The role of metal ion dopants in quantum-sized TiO2: Correlation between photoreactivity and charge carrier recombination dynamics. J. Phys. Chem. 1994, 98, 13669–13679. [CrossRef] 14. Estrellan, C.R.; Salim, C.; Hinode, H. Photocatalytic activity of sol–gel derived TiO2 co-doped with iron a niobium. React. Kinet. Catal. Lett. 2009, 98, 187–192. [CrossRef] 15. Shi, Z.; Lai, H.; Yao, S.; Wang, S. Photocatalytic activity of Fe and Ce co-doped mesoporous TiO2 cata under UV and visible light. J. Chin. Chem. Soc. 2012, 59, 614–620. [CrossRef] 16. Soler-Illia, G.D.A.; Louis, A.; Sanchez, C. Synthesis and characterization of mesostructured titania-based materials through evaporation-induced self-assembly. Chem. Mater. 2002, 14, 750–759. [CrossRef] 17. Al-Ekabi, H.; Serpone, N. Kinetics studies in heterogeneous photocatalysis. I. Photocatalytic degradation of chlorinated phenols in aerated aqueous solutions over titania supported on a glass matrix. J. Phys. Chem. 1988, 92, 5726–5731. [CrossRef] 18. Chang, W.; Kim, H.; Lee, G.Y.; Ahn, B.J. Catalytic hydrodechlorination reaction of chlorophenols by Pd nanoparticles supported on graphene. Res. Chem. Int. 2016, 42, 71–82. [CrossRef] 19. Ruiz-García, C.; Heras, F.; Calvo, L.; Alonso-Morales, N.; Rodriguez, J.J.; Gilarranz, M.A. Platinum and N-doped carbon nanostructures as catalysts in hydrodechlorination reactions. Appl. Catal. B Environ. 2018, 238, 609–617. [CrossRef] 20. Hernandez, S.R.; Kergaravat, S.V.; Pividori, M.I. Enzymatic electrochemical detection coupled to multivariate calibration for the determination of phenolic compounds in environmental samples. Talanta 2013, 106, 399–407. [CrossRef] 21. Janda, V.; Svecova, M. By-products in drinking water disinfection. Chem. List 2000, 94, 905–90 22. Li, B.; Zhao, Z.; Gao, F.; Wang, X.; Qiu, J. Mesoporous microspheres composed of carbon-coated TiO2 nanocrystals with exposed {001} facets for improved visible light photocatalytic activity. Appl. Catal. B Environ. 2014, 147, 958–964. [CrossRef] 362 Catalysts 2018, 8, 631 23. Hussain, S.T.; Siddiqa, A.; Siddiq, M.; Ali, S. Iron-doped titanium dioxide nanotubes: A study of electrical, optical, and magnetic properties. J. Nanopart. Res. 2011, 13, 6517–6525. [CrossRef] p g p p p 24. Lan, X.; Wang, L.; Zhang, B.; Tian, B.; Zhang, J. Preparation of lanthanum and boron co-doped TiO2 by diﬁd l l th d d t d th i h t t l ti ti it C t l T d 2014 224 163 170 [C R f] 24. Lan, X.; Wang, L.; Zhang, B.; Tian, B.; Zhang, J. Preparation of lanthanum and boron co-doped TiO2 by modiﬁed sol–gel method and study their photocatalytic activity. Catal. References Today 2014, 224, 163–170. [CrossRef] 24. Lan, X.; Wang, L.; Zhang, B.; Tian, B.; Zhang, J. Preparation of lanthanum and boron co-doped TiO2 by 25. Hsieh, C.-T.; Fan, W.-S.; Chen, W.-Y.; Lin, J.-Y. Adsorption and visible-light-derived photocatalytic kinetics of organic dye on Co-doped titania nanotubes prepared by hydrothermal synthesis. Sep. Purif. Technol. 2009, 67, 312–318. [CrossRef] 26. Neši´c, J.; Manojlovi´c, D.D.; Anđelkovi´c, I.; Dojˇcinovi´c, B.P.; Vuli´c, P.J.; Krsti´c, J.; Rogli´c, G.M. Preparation, characterization and photocatalytic activity of lanthanum and vanadium co-doped mesoporous TiO2 for azo-dye degradation. J. Mol. Catal. A Chem. 2013, 378, 67–75. [CrossRef] 27. Ohsaka, T. Temperature dependence of the Raman spectrum in anatase TiO2. J. Phys. Soc. Jpn. 1980, 48, 1661–1668. [CrossRef] 28. Štengl, V.; Velická, J.; Maˇríková, M.; Grygar, T.M. New generation photocatalysts: how tungsten inﬂuences the nanostructure and photocatalytic activity of TiO2 in the UV and visible regions. ACS Appl. Mater. Int. 2011, 3, 4014–4023. [CrossRef] 29. Wu, Y.; Zhang, J.; Xiao, L.; Chen, F. Properties of carbon and iron modiﬁed TiO2 photocatalyst synthesized at low temperature and photodegradation of acid orange 7 under visible light. Appl. Surf. Sci. 2010, 256, 4260–4268. [CrossRef] Shojaie, A.F.; Loghmani, M.H. La3+ and Zr4+ co-doped anatase nano TiO2 by sol-microwave method. Chem. Eng. J. 2010, 157, 263–269. [CrossRef] 31. Liu, D.; Wu, Z.; Tian, F.; Ye, B.-C.; Tong, Y. Synthesis of N and La co-doped TiO2/AC photocatalyst by microwave irradiation for the photocatalytic degradation of naphthalene. J. Alloys Compd. 2016, 676, 489–498. [CrossRef] 32. Wang, Q.; Jiang, H.; Zang, S.; Li, J.; Wang, Q. Gd, C., N and P quaternary doped anatase-TiO2 nano- photocatalyst for enhanced photocatalytic degradation of 4-chlorophenol under simulated sunlight irradiation. J. Alloys Compd. 2014, 586, 411–419. [CrossRef] 33. Kubelka, P.; Munk, F. Ein beitrag zur optik der farbanstriche. Z. Tech. Phys. 1931, 12, 593–601. 34. Gutiérrez-Alejandre, A.; Ramírez, J.; Busca, G. The electronic structure of oxide-supported tun catalysts as studied by UV spectroscopy. Catal. Lett. 1998, 56, 29–33. [CrossRef] 35. Li, J.; Li, B.; Li, J.; Liu, J.; Wang, L.; Zhang, H.; Zhang, Z.; Zhao, B. Visible-light-driven photocatalyst of La–N-codoped TiO2 nano-photocatalyst: Fabrication and its enhanced photocatalytic performance and mechanism. J. Ind. Eng. Chem. 2015, 25, 16–21. [CrossRef] 36. Zhang, J.; Xu, L.J.; Zhu, Z.Q.; Liu, Q.J. Synthesis and properties of (Yb, N)-TiO2 photocatalyst for degradation of methylene blue (MB) under visible light irradiation. Mater. Res. Bull. 2015, 70, 358–364. [CrossRef] 36. References Zhang, J.; Xu, L.J.; Zhu, Z.Q.; Liu, Q.J. Synthesis and properties of (Yb, N)-TiO2 photocatalyst for degradation of methylene blue (MB) under visible light irradiation. Mater. Res. Bull. 2015, 70, 358–364. [CrossRef] 37. Khan, H.; Berk, D. Characterization and mechanistic study of Mo+6 and V+5 codoped TiO2 as a photocatalyst. J Photochem Photobiol A Chem 2014 294 96 109 [CrossRef] 37. Khan, H.; Berk, D. Characterization and mechanistic study of Mo+6 and V+5 codoped TiO2 as a photocatalyst. J. Photochem. Photobiol. A Chem. 2014, 294, 96–109. [CrossRef] 37. Khan, H.; Berk, D. Characterization and mechanistic study of M J. Photochem. Photobiol. A Chem. 2014, 294, 96–109. [CrossRef] 38. Wang, Q.; Xu, S.; Shen, F. Preparation and characterization of TiO2 photocatalysts co-doped with iron (III) and lanthanum for the degradation of organic pollutants. Appl. Surf. Sci. 2011, 257, 7671–7677. [CrossRef] 39. Wang, Z.; Chen, C.; Wu, F.; Zou, B.; Zhao, M.; Wang, J.; Feng, C. Photodegradation of rhodamine B under 38. Wang, Q.; Xu, S.; Shen, F. Preparation and characterization of TiO2 photocatalysts co-doped with iron (III) and lanthanum for the degradation of organic pollutants. Appl. Surf. Sci. 2011, 257, 7671–7677. [CrossRef] 39. Wang, Z.; Chen, C.; Wu, F.; Zou, B.; Zhao, M.; Wang, J.; Feng, C. Photodegradation of rhodamine B un visible light by bimetal codoped TiO2 nanocrystals. J. Hazard. Mater. 2009, 164, 615–620. [CrossRef] 40. Wang, S.; Bai, L.N.; Sun, H.M.; Jiang, Q.; Lian, J.S. Structure and photocatalytic property of Mo-doped T nanoparticles. Powder Technol. 2013, 244, 9–15. [CrossRef] 41. Luo, S.-Y.; Yan, B.-X.; Shen, J. Enhancement of photoelectric and photocatalytic activities: Mo doped TiO2 thin ﬁlms deposited by sputtering. Thin Solid Films 2012, 522, 361–365. [CrossRef] 42. Tae Kwon, Y.; Yong Song, K.; In Lee, W.; Jin Choi, G.; Rag Do, Y. Photocatalytic behavior of WO3-loaded TiO2 in an oxidation reaction. J. Catal. 2000, 191, 192–199. [CrossRef] 43. Park, H.; Park, Y.; Kim, W.; Choi, W. Surface modiﬁcation of TiO2 photocatalyst for environmental applications. J. Photochem. Photobiol. C Photochem. Rev. 2013, 15, 1–20. [CrossRef] 44. Wang, Z.; Liu, X.; Li, W.; Wang, H.; Li, H. Enhancing the photocatalytic degradation of salicylic acid by using molecular imprinted S-doped TiO2 under simulated solar light. Ceram. Int. 2014, 40, 8863–8867. [CrossRef] molecular imprinted S-doped TiO2 under simulated solar light. Ceram. Int. 2014, 40, 8863–8867. [CrossRef] 45. Sharma, S.; Mukhopadhyay, M.; Murthy, Z.V.P. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Irwing M. Ramírez-Sánchez 1 and Erick R. Bandala 2,3,* 1 Department of Civil, Architectural and Environmental Engineering, The University of Texas at Austin, Austin, TX 78712, USA; irwingmoises@gmail.com g g 2 Desert Research Institute (DRI), 755 E. Flamingo Road, Las Vegas, NV 89119-7363, USA 2 Desert Research Institute (DRI), 755 E. Flamingo Road, Las Vegas, NV 89119-7363, USA 3 Graduate Program Hydrologic Sciences University of Nevada Reno NV 89557 USA 2 Desert Research Institute (DRI), 755 E. Flamingo Road, Las Vegas, NV 89119-7363, USA 3 Graduate Program Hydrologic Sciences, University of Nevada, Reno, NV 89557, USA * Correspondence: erick bandala@dri edu; Tel : + 1-(702)-862-5395 3 Graduate Program Hydrologic Sciences, University of Nevada, Reno, NV 89557, USA * Correspondence: erick.bandala@dri.edu; Tel.: + 1-(702)-862-5395 * Correspondence: erick.bandala@dri.edu; Tel.: + 1-(702)-862-5395 Received: 29 September 2018; Accepted: 28 November 2018; Published: 5 December 2018 Received: 29 September 2018; Accepted: 28 November 2018; Published: 5 December 2018 Abstract: Iron-doped TiO2 nanoparticles (Fe-TiO2) were synthesized and photocatalitically investigated under high and low ﬂuence values of UV radiation. The Fe-TiO2 physical characterization was performed using X-ray Powder Diffraction (XRD), Brunauer–Emmett–Teller (BET) surface area analysis, Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), Diffuse Reﬂectance Spectroscopy (DRS), and X-ray Photoelectron Spectroscopy (XPS). The XPS evidenced that the ferric ion (Fe3+) was in the TiO2 lattice and unintentionally added co-dopants were also present because of the precursors of the synthetic method. The Fe3+ concentration played a key role in the photocatalytic generation of hydroxyl radicals (•OH) and estriol (E3) degradation. Fe-TiO2 accomplished E3 degradation, and it was found that the catalyst with 0.3 at.% content of Fe (0.3 Fe-TiO2) enhanced the photocatalytic activity under low UV irradiation compared with TiO2 without intentionally added Fe (zero-iron TiO2) and Aeroxide® TiO2 P25. Furthermore, the enhanced photocatalytic activity of 0.3 Fe-TiO2 under low UV irradiation may have applications when radiation intensity must be controlled, as in medical applications, or when strong UV absorbing species are present in water. Keywords: iron-doped TiO2; photocatalytic activity; low UV irradiation; hydroxyl radical; estrio Photocatalytic Degradation of Estriol Using Iron-Doped TiO2 under High and Low UV Irradiation Irwing M. Ramírez-Sánchez 1 and Erick R. Bandala 2,3,* Catalysts 2018, 8, 625; doi:10.3390/catal8120625 References Rate parameter estimation for 4-chlorophenol degradation by UV and organic oxidants. J. Ind. Eng. Chem. 2012, 18, 249–254. [CrossRef] 45. Sharma, S.; Mukhopadhyay, M.; Murthy, Z.V.P. Rate parameter estimation for 4-chlorophenol degradation by UV and organic oxidants. J. Ind. Eng. Chem. 2012, 18, 249–254. [CrossRef] 363 Catalysts 2018, 8, 631 46. Yuan, J.; Wang, E.; Chen, Y.; Yang, W.; Yao, J.; Cao, Y. Doping mode, band structure and photocatalytic mechanism of B–N-codoped TiO2. Appl. Surf. Sci. 2011, 257, 7335–7342. [CrossRef] p 47. Li, X.; Zhang, H.; Zheng, X.; Yin, Z.; Wei, L. Visible light responsive N-F-codoped TiO2 photocatalysts for the degradation of 4-chlorophenol. J. Environ. Sci. 2011, 23, 1919–1924. [CrossRef] 48. Neville, E.M.; Mattle, M.J.; Loughrey, D.; Rajesh, B.; Rahman, M.; MacElroy, J.M.D.; Sullivan, J.A.; Thampi, K.R. Carbon-Doped TiO2 and Carbon, Tungsten-Codoped TiO2 through Sol–Gel Processes in the Presence of Melamine Borate: Reﬂections through Photocatalysis. J. Phys. Chem. C 2012, 116, 16511–16521. [CrossRef] 49. Shamaila, S.; Sajjad, A.K.L.; Chen, F.; Zhang, J. Synthesis and characterization of mesoporous-TiO2 with enhanced photocatalytic activity for the degradation of chloro-phenol. Mater. Res. Bull. 2010, 45, 1375–1382. [CrossRef] 50. Theurich, J.; Lindner, M.; Bahnemann, D.W. Photocatalytic degradation of 4-chlorophenol in aerated aqueous titanium dioxide suspensions: A kinetic and mechanistic study. Langmuir 1996, 12, 6368–6376. [CrossRef] 51. Gaya, U.I.; Abdullah, A.H.; Zainal, Z.; Hussein, M.Z. Photocatalytic treatment of 4-chlorophenol in aqueous ZnO suspensions: Intermediates, inﬂuence of dosage and inorganic anions. J. Hazard. Mater. 2009, 168, 57–63. [CrossRef] [PubMed] 52. Lipczynska-Kochany, E.; Kochany, J.; Bolton, J.R. Electron paramagnetic resonance spin trapping detection of short-lived radical intermediates in the direct photolysis of 4-chlorophenol in aerated aqueous solution. J. Photochem. Photobiol. A Chem. 1991, 62, 229–240. [CrossRef] 53. Pozan, G.S.; Kambur, A. Removal of 4-chlorophenol from wastewater: Preparation, characterization and photocatalytic activity of alkaline earth oxide doped TiO2. Appl. Catal. B Environ. 2013, 129, 409–415. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 364 www.mdpi.com/journal/catalysts catalysts catalysts 1. Introduction In recent years, society and the scientiﬁc community have concerned of Emerging Contaminants (ECs, also called Contaminants of Emerging Concern), which are chemicals that threaten the environment, human health, and water safety and are not currently covered by existing local or international water quality regulations [1]. ECs include chemical species such as algae toxins, illegal drugs, industrial compounds, ﬂame retardants, food additives, nanoparticles, pharmaceuticals (human and veterinary), personal care products, pesticides, biocides, steroids, synthetic and natural hormones, and surfactants [2]. Natural hormones (e.g., estrone (E1), 17β-estradiol (E2), and estriol (E3)) as ECs are susceptible of persisting and bioaccumulating in the environment, and could induce endocrine disruption in humans and wildlife (vertebrates [3–5] and invertebrates [6,7]). Natural attenuation, drinking water puriﬁcation, and conventional municipal wastewater treatment processes are either incapable or only partially capable of removing estrogens from water [8]. As result, water treatment techniques are being developed to manage, reduce, degrade, and mineralize low-concentrated ECs (including natural estrogen) in drinking and wastewater [9]. Advanced Oxidation Processes (AOPs) are promising techniques to treat ECs in aqueous phase, which include well-known processes such as Fenton and Fenton-like processes, UV/H2O2, ozonation, and photocatalysis using semiconductors, peroxone processes (H2O2/O3), and cavitation [10,11]. Although there are many known AOPs, since Coleman’s Catalysts 2018, 8, 625; doi:10.3390/catal8120625 www.mdpi.com/journal/catalysts 365 Catalysts 2018, 8, 625 work [12], photocatalysis using titanium dioxide (TiO2) has been identiﬁed as one of the most effective methods to degrade estrogens in water [13]. Several reports recognized that TiO2 can degrade estrogens, which prevents increases in estrogenic activity in water [14,15] and partially or completely mineralizing estrogens [14,16]. Titanium dioxide is the most commonly used photocatalyst because of its reasonable optical and electronic properties, good photocatalytic activity, insolubility in water, chemical and photochemical stability, nontoxicity, low cost, and high efﬁciency in pollutant mineralization [17–20]. However, the band gap energy (Eg) of TiO2, frequently reported as 3.2 eV [21], restrains the photocatalytic activation to energy sources with a portion of spectrum emission below 387.5 nm [22]. In general the photocatalytic mechanism is as shown in Figure 1. According to Density Functional Theory (DFT) computations, the valence band (VB) and conduction band (CB) of pure TiO2 are mainly composed of O2p orbitals and Ti3d orbitals, respectively. Hence, the Fermi level (EF) is located in the middle of the band gap (BG), indicating that VB is full ﬁlled while CB is empty [23]. 1. Introduction When using photons with energy higher than 3.2 eV, photoexcitation of the semiconductor promotes electrons from VB to CB creating a charge vacancy or hole (h+) in the VB. The h+ in the VB can react with hydroxide ion to form hydroxyl radical (•OH) or can also be ﬁlled by donor absorbed organic molecule (OMads). Photogenerated electrons in the CB can be transferred to acceptor of electrons and bring about •OH. Figure 1. Photocatalytic mechanism of TiO2 for •OH generation. Where Eg: Band gap energy; E: photon energy; OMads: adsorbed organic molecule; and OMoxi: oxidized organic molecule. Figure 1. Photocatalytic mechanism of TiO2 for •OH generation. Where Eg: Band gap energy; E: photon energy; OMads: adsorbed organic molecule; and OMoxi: oxidized organic molecule. Consequently, reducing the photon energy needed for TiO2 photoactivation has been the focus of the scientiﬁc community until now. Doping is one of the techniques that has been tested to control or modify the surface properties or internal structure of TiO2. Doping introduces a foreign element into TiO2 to cause an impurity state in the band gap. The most frequently used doping materials are transition-metal cations (e.g., Cr, V, Fe, and Ni) at Ti sites, and anions (e.g., N, S, and C) at O sites [24]. Among anion- and cation-dopants, the ferric ion (Fe3+) is one of the most often used because the ionic radius of Fe3+ (0.69 A) is similar to Ti4+ (0.745 A) [25]. Therefore, Fe3+ can be easily incorporated into the TiO2 crystal lattice. The main reported effects of iron-doped TiO2 is a rapid increase in photocatalytic activity that increases with increased Fe doping, which then reaches a maximum value, and ﬁnally decreases with 366 Catalysts 2018, 8, 625 further increased Fe content [23,26–37]. However, detrimental effects have been also reported because of high Fe content [38,39] or agglomerated Fe-TiO2 nanoparticles [40,41]. Although several theoretical and experimental Fe-TiO2 studies have been developed, the trade-off between doping ratio and radiation intensity is scarcely mentioned. Furthermore, Fe-TiO2 photocatalyst has rarely been considered to be a useful technique for the degradation of E3 [42]. In this work, Fe-TiO2 nanoparticles were synthesized to increase the understanding of the relationship between doping ratio and radiation intensity for hydroxyl radical (•OH) generation and E3 degradation. Therefore, we investigated the photocatalytic degradation of E3 using Fe-TiO2 under high and low UV irradiation. 1. Introduction We highlight the term low UV irradiation to avoid confusion with the term “photocatalytic processes under visible light” because we did not intentionally use UV cutoff ﬁlters for the experiments. Table 1. Surface elemental composition determined by XPS. Table 1. Surface elemental composition determined by XPS. Material Atomic % of Elements (at.%) Ti2p O1s C1s Fe2p S2p N1s Zero-iron TiO2 24.4 52.9 21.3 0 1.4 - 0.3 Fe-TiO2 23.8 51.1 22.9 0.3 1.1 0.8 0.6 Fe-TiO2 23.9 53.1 22.5 0.6 - - 1.0 Fe-TiO2 23.5 52.5 20.6 1 1.5 0.9 High-resolution XPS spectra for the iron region (Figure 3) was studied only for 1.0 Fe-TiO2 because no Fe2p signals were detected for zero-iron TiO2, 0.3 Fe-TiO2, or 0.6 Fe-TiO2. The deconvolution of high-resolution XPS spectra (Figure 3) was developed for previously reported peaks of Fe2+ and Fe3+ [43]. Shirley baseline was subtracted before peak ﬁtting. The Gaussian–Lorentzian mix function was used with a 40% factor. Charge compensation was set by the O1s peak charge with −0.58 eV. As a result, the correlation between the experimental signal and the theoretic model (Σχ2) was 8.43 × 10−2. Figure 3. High-resolution XPS spectra for the iron region for 1.0 Fe-TiO2. Figure 3. High-resolution XPS spectra for the iron region for 1.0 Fe-TiO2. According to the theoretical model (sum of ﬁtting peaks), both Fe3+ and Fe2+ were present in the lattice of 1.0 Fe-TiO2. We suggest that Fe3+ was incorporated into the lattice of TiO2 to form Ti–O–Fe bonds, because the ionic radius of Fe3+ (0.69 A) is similar to the ionic radius of Ti4+ (0.745 A) [25]. The XPS technique detected Fe2+ because Fe3+ underwent reduction to Fe2+ during XPS measurement in vacuum [44]. According to the theoretical model (sum of ﬁtting peaks), both Fe3+ and Fe2+ were present in the lattice of 1.0 Fe-TiO2. We suggest that Fe3+ was incorporated into the lattice of TiO2 to form Ti–O–Fe bonds, because the ionic radius of Fe3+ (0.69 A) is similar to the ionic radius of Ti4+ (0.745 A) [25]. The XPS technique detected Fe2+ because Fe3+ underwent reduction to Fe2+ during XPS measurement in vacuum [44]. The band gap energy (Eg) obtained with the Kubelka–Monk method (Figure 4) for Aeroxide® TiO2 P25 was 3.2 eV, which is consistent with the value reported previously [45]. 2.1. Characterization of Iron-Doped TiO2 2.1. Characterization of Iron-Doped TiO2 Figure 2 shows X-ray Photoelectron Spectroscopy (XPS) general spectra of TiO2 without added Fe (zero-iron TiO2) and Fe-TiO2 materials (b, c, and d). For the experimental condition used, Fe did not affect the bonding structure between titanium and oxygen because the main peaks for all samples were Ti2p and O1s with the proportion 1:2.2, which is in agreement with the atomic formula of TiO2. Figure 2. X-ray Photoelectron Spectroscopy (XPS) general spectra for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). Figure 2. X-ray Photoelectron Spectroscopy (XPS) general spectra for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). XPS detected unintentionally added elements such as carbon, sulfur, and nitrogen (Table 1) as co-dopants of zero-iron TiO2 and Fe-TiO2, which were introduced into TiO2 via precursors of the synthesis. Carbon and sulfur could come from sodium dodecyl sulfate (SDS), and nitrogen could come from iron (III) nitrate (Fe(NO3)3·9H2O) and HNO3, all of them used in the synthesis process. 367 Catalysts 2018, 8, 625 Table 1. Surface elemental composition determined by XPS. Table 1. Surface elemental composition determined by XPS. For Aeroxide® TiO2 P25 Eg, red-shifts were detected as 0.22, 0.24, 0.25, and 0.3 eV for zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1 0 Fe TiO respectively which is consistent with values reported by Shi et al of 0 25 eV [46] The band gap energy (Eg) obtained with the Kubelka–Monk method (Figure 4) for Aeroxide® TiO2 P25 was 3.2 eV, which is consistent with the value reported previously [45]. For Aeroxide® TiO2 P25 Eg, red-shifts were detected as 0.22, 0.24, 0.25, and 0.3 eV for zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2, respectively, which is consistent with values reported by Shi et al. of 0.25 eV [46] and with density functional theory calculations that suggested the hybridized band of Ti3d and Fe3d reduces Eg approximately 0.3–0.5 eV [44], or 0.2–0.34 eV [47]. The band gap energy (Eg) obtained with the Kubelka–Monk method (Figure 4) for Aeroxide® TiO2 P25 was 3.2 eV, which is consistent with the value reported previously [45]. For Aeroxide® TiO2 P25 Eg, red-shifts were detected as 0.22, 0.24, 0.25, and 0.3 eV for zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2, respectively, which is consistent with values reported by Shi et al. of 0.25 eV [46] and with density functional theory calculations that suggested the hybridized band of Ti3d and Fe3d reduces Eg approximately 0.3–0.5 eV [44], or 0.2–0.34 eV [47]. 368 Catalysts 2018, 8, 625 Figure 4. Band gap energy (Eg) by the Kubelka–Monk method. Zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). Figure 4. Band gap energy (Eg) by the Kubelka–Monk method. Zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). For zero-iron TiO2, Eg for Fe-TiO2 materials (Table 2) decreased as long as the Fe content increased, so the Fe content generated red-shift. For Aeroxide® TiO2 P25 Eg, the red-shift of Fe-TiO2 agreed with previously reported values, but it agreed less for zero-iron TiO2. Therefore, red-shift was not only related to Fe content, but also to the synthesis method and unintentionally co-doped TiO2. Table 2. Structural and optical properties of zero-iron TiO2, and Fe-TiO2. Table 2. Structural and optical properties of zero-iron TiO2, and Fe-TiO2. Table 1. Surface elemental composition determined by XPS. Material Eg Anatase: Rutile Particle Size Surface Area Pore Size High UV Low UV eV nm % nm m2 g−1 nm % % Aeroxide® TiO2 P25 3.2 * 387.5 * 80:20 * 21 * 50 ± 15 * 17.5 * 36.4 0.8 Zero-iron TiO2 2.98 416.1 73.1:26.9 6.6 66.5 8.4 99.26 7.64 0.3 Fe-TiO2 2.96 418.9 77.9:21.1 6.9 77.6 1.2 99.40 8.21 0.6 Fe-TiO2 2.95 420.3 78.8:21.2 7.1 73.0 1.4 99.42 8.77 1.0 Fe-TiO2 2.90 427.6 76.3:23.7 6.9 83.1 9.4 99.43 10.63 * According to the manufacturer. XRD patterns in Figure 5 revealed zero-iron TiO2 and Fe-TiO2 materials had both anatase and rutile phases. No XRD Fe2O3 peaks (2θ equal to 33.0◦, 35.4◦, 40.7◦, 43.4◦, and 49.2◦) were observed, concluding that Fe3+ replaced Ti4+ in the TiO2 crystal framework [48,49]. The synthesis method allowed uniform distribution of Fe within TiO2. The anatase:rutile phase ratio calculated by Spurr and Myers’ method showed that zero-iron TiO2 and Fe-TiO2 materials were a mixture of anatase and rutile phases (Table 2). The amount of anatase was less in Fe-TiO2 materials than in Aeroxide® TiO2 P25. The smaller proportion of anatase could lead to a reduction of photocatalytic activity because the anatase phase has higher photocatalytic activity than rutile TiO2 [50,51]. However, it is accepted that the optimal photocatalytic activity of TiO2 is reached with an optimal mixture of anatase and rutile phases [52]. Moreover, the increased anatase proportion in 0.3 Fe-TiO2 and 0.6 Fe-TiO2 compared 369 Catalysts 2018, 8, 625 with zero-iron TiO2 could improve photocatalytic activity. The increased anatase proportion was attributable to Fe doping disturbing the arrangements of TiO2 phases [53]. This trend has also been observed when Fe-doped TiO2 was synthesized using sol-gel [54] or co-precipitation methods [32]. Figure 5. XRD patterns for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d), where A is Anatase and R is Rutile phases. Figure 5. XRD patterns for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d), where A is Anatase and R is Rutile phases. The average particle size of Fe-TiO2 materials obtained by Scherrer’s formula was 6.9 nm, which is less than the particle size of Aeroxide® TiO2 P25 (Table 2). Fe-TiO2 materials should increase photocatalytic activity because of their higher surface area and the short migration distance of the photogenerated charge carriers (electron/hole (e−/h+)) from the bulk material to the surface. Table 1. Surface elemental composition determined by XPS. Further BET analysis (Figure 6) conﬁrmed that average surface area of Fe-TiO2 materials was 77.9 m2 g−1, higher than zero-iron TiO2 and Aeroxide® TiO2 P25. BET isotherms followed a type IV shape according to the Langmuir classiﬁcation, which is associated with the characteristics of mesoporous material [55]. The observed hysteresis is probably due to gas cooperative adsorption or condensation inside the pores of material [56]. BET analysis showed pore sizes (Table 2) were in the mesoporous range (2–50 nm, according to IUPAC classiﬁcation) for zero-iron TiO2 and 1.0 Fe-TiO2, and the microporous range (0.2–2 nm, according to IUPAC classiﬁcation) for 0.3 Fe-TiO2 and 0.6 Fe-TiO2. Mesoporous pore size should facilitate the mass transfer of reactants and products in the reaction system, so photocatalytic improvement based on this property could improve zero-iron TiO2 and Fe-TiO2 materials with respect to Aeroxide® TiO2 P25 [31]. Patra et al. [49] developed a similar nanoparticle synthesis procedure, which generated surface area values ranging from 126 to 385 m2 g−1 and mesoporous size distribution values ranging from 3.1 to 3.4 nm. Particles obtained in our work were different, probably because of the application of a mild thermal treatment and the use of SDS at critical micelle concentration as a template. Figure 7 shows SEM images of agglomerated and assembled nanoparticles of zero-iron TiO2. The different amounts of Fe in the TiO2 lattice changed neither the particle size nor the morphology of the zero-iron TiO2. Although the average pore size allowed an increase of the superﬁcial area, agglomeration could lead to lower photocatalytic activity. 370 atalysts 2018, 8, 625 Figure 6. Brunauer–Emmett–Teller (BET) isotherms for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). Figure 7. SEM image of zero-iron TiO2 after mechanical grinding and sonication. Transmission electron microscopy (TEM) images conﬁrmed nanoparticle clusters and partic izes of zero-iron TiO2 (Figure 8b) and 0.3 Fe-TiO2 (Figure 8a) between 5 and 10 nm (between 1.2 an .4 nm according to Scherrer’s formula). The lattice fringe spacing was 0.35 nm, as shown in Figure 8 which was consistent with the d-spacing (101) of anatase [25]. The lattice ﬁngers of the nanoparticl howed that Fe-TiO2 materials were highly crystallized. Catalysts 2018, 8, 625 Figure 6. Brunauer–Emmett–Teller (BET) isotherms for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). Figure 6. Table 1. Surface elemental composition determined by XPS. Brunauer–Emmett–Teller (BET) isotherms for zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). Figure 7. SEM image of zero-iron TiO2 after mechanical grinding and sonication. Figure 7. SEM image of zero-iron TiO2 after mechanical grinding and sonication. Figure 7. SEM image of zero-iron TiO2 after mechanical grinding and sonication. Transmission electron microscopy (TEM) images conﬁrmed nanoparticle clusters and particle sizes of zero-iron TiO2 (Figure 8b) and 0.3 Fe-TiO2 (Figure 8a) between 5 and 10 nm (between 1.2 and 9.4 nm according to Scherrer’s formula). The lattice fringe spacing was 0.35 nm, as shown in Figure 8b, which was consistent with the d-spacing (101) of anatase [25]. The lattice ﬁngers of the nanoparticles showed that Fe-TiO2 materials were highly crystallized. 371 Catalysts 2018, 8, 625 (a) (b) Figure 8. Transmission electron microscopy (TEM) image of 0.3 Fe-TiO2 (a) and zero-iron TiO2 (b). (a) (a) (b) (b) (a) Figure 8. Transmission electron microscopy (TEM) image of 0.3 Fe-TiO2 (a) and zero-iron TiO2 (b). 2.2. Characterization of Irradiation Source Figure 9 shows the emission spectra of irradiation sources used in this study. Using the main peaks reported for a ﬂuorescent lamp (Figure 9a), the calibration of the spectrometer generated an R2 value equal to 0.999. The emission spectrum of the GE F15T8 BLB lamp (Figure 9b) was in the 356–410 nm range. However, the emission spectrum of the GE F15T8 D lamp (Figure 9c) was continuous broadband between 380 and 750 nm. The light intensity of the GE F15T8 lamp was reported to be between 3440 μW cm−2 [57] and 4000 μW cm−2 [58], from which 6% was UV radiation [59]. The intensity of the GE F15T8 lamp was 1500 μW cm−2. This lamp has an internal coating that absorbs 78% of visible light (as speciﬁed by the manufacturer) in the spectrum below 400 nm, as shown in Figure 9b. Therefore, the GE F15T8 BLB and GE F15T8 D lamps were designated as high and low UV irradiation sources, respectively. Figure 9. Emission spectrum and intensity graph of the irradiation source of Tecnolite ﬂuorescent lamp (a), GE F15T8 BLB lamp (b), and GE F15T8 D lamp (c). Figure 9. Emission spectrum and intensity graph of the irradiation source of Tecnolite ﬂuorescent lamp (a), GE F15T8 BLB lamp (b), and GE F15T8 D lamp (c). 372 Catalysts 2018, 8, 625 Because Eg of Aeroxide® TiO2 P25 is 3.2 eV (387.5 nm), see Figure 9, both the GE F15T8 BLB and GE F15T8 D lamps emitted photons that could photoactivate Aeroxide® TiO2 P25. However, the proportion of the emission spectrum that Aeroxide® TiO2 P25 could use for photocatalytic activity was different. An approximation of the amount of radiative intensity used for photocatalytic activity was obtained with the area under the curve-spectrum below the Eg value. Consequently, Aeroxide® TiO2 P25 could take advantage of 36.4% of the emission spectrum of the GE F15T8 BLB lamp and 0.8% of the emission spectrum of the GE F15T8 D lamp. Table 2 lists amount of radiative spectrum used by zero-iron TiO2 and Fe-TiO2 materials according to each Eg. 2.2. Characterization of Irradiation Source g g Based on morphological and crystalline structure analysis, the favorable characteristics to enhance photocatalytic activity of Fe-TiO2 material are effective insertion of the Fe3+ ion into the TiO2 lattice, red-shift (2.90–2.96 eV), nanoparticle size (6.9–7.1 nm), speciﬁc surface area (73.0–83.1 nm), pore size (1.2–9.4 nm), and radiation absorbance below the equivalent Eg wavelength (8.21–10.63% of daylight lamp spectrum). Its main disadvantageous characteristics are expected to be high particle agglomeration and lower anatase phase compared with zero-iron TiO2. Further, photocatalytic activity is very sensitive to crystalline array and particle size and shape; differences in the density of hydroxyl groups on the particle surface and the number of water molecules hydrating the surface; the surface area and surface charge; differences in the number and nature of trap sites; the dopant concentration, localization, and chemical state of the dopant ions; radiation intensity; particle aggregation and superﬁcial charge; and scavenger species in media [39,60]. Consequently, material characterization alone could not predict photocatalytic activity [28]. Therefore, in this research, we used the N,N-dimethyl-p-nitrosoaniline (pNDA) probe and E3 to evaluate the photocatalytic activity by following •OH production, which is one of the most signiﬁcant reactive oxygen species (ROS), and E3, which is an EC. 2.3. Hydroxyl Radical Generation under High and Low UV Irradiation Catalyst at.% Load High UV Irradiation Low UV Irradiation k1 R2 r0 k1 R2 r0 mg L−1 min−1 μM•OH min−1 min−1 μM•OH min−1 TiO2 Aeroxide® P25 - 20 0.06 0.988 0.49 0.012 0.989 0.105 Zero-iron TiO2 0 320 0.056 0.993 0.49 0.005 0.973 0.045 0.3 Fe-TiO2 0.3 320 0.067 0.998 0.58 0.004 0.990 0.042 0.6 Fe-TiO2 0.6 320 0.031 0.998 0.28 0.002 0.999 0.025 1.0 Fe-TiO2 1 320 0.004 0.987 0.04 0.00002 0.891 0.0002 Figure 10. •OH generation (initial velocity) of zero-iron TiO2 and Fe-TiO2 under high UV irradiation (a) and low UV Irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2 at pH 6 ± 0.1, and 20 ◦C. Figure 10. •OH generation (initial velocity) of zero-iron TiO2 and Fe-TiO2 under high UV irradiation (a) and low UV Irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2 at pH 6 ± 0.1, and 20 ◦C. Figure 10. •OH generation (initial velocity) of zero-iron TiO2 and Fe-TiO2 under high UV irradiation (a) and low UV Irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2 at pH 6 ± 0.1, and 20 ◦C. When high UV irradiation was used, the maximum r0 was 0.58 μM•OH min−1 for 0.3 Fe-TiO2. The enhancement in photocatalytic activity of 0.3 at.% Fe-TiO2, compared with zero-iron TiO2 was by the extended lifetime values of the photogenerated charge carriers (e−and h+) produced by Fe3+ ions, which played a role as charge carriers trapped at or near the particle surface. The trapping mechanisms are shown in Equations (2)–(5) [62]. Fe3+ + ecd−→Fe2+ electron trap (2) Fe2+ + Ti4+ →Fe3+ + Ti3+ migration (3) Fe3+ +hvb + →Fe4+ hole trap (4) Fe4+ + OH−→Fe3+ + •OH migration (5) (2) (5) The mechanism suggested for •OH generation is shown in Figure 11. When TiO2 contains a Fe3+ ion, the Fe3d orbitals split into two bands, one is a hybrid band (A2g) and one is midgap band (T2g), which induce a new localized BG state [23]. Therefore, when TiO2 absorbs photons with energy less than 3.2 eV, photoexcitation of the semiconductor promotes an electron from the VB to the midgap band (T2g), also called a shallow trap, creating an electron-hole pair. The hole in the valence band (VB) can react with hydroxide ions to form •OH, absorbed organic molecules, or trap Fe3+ following Equations (4) and (5). 2.3. Hydroxyl Radical Generation under High and Low UV Irradiation 2.3. Hydroxyl Radical Generation under High and Low UV Irradiation The generation of •OH was measured using pNDA, which is a well-characterized •OH scavenger as mentioned in Section 3.5. In brief, pNDA undergoes bleaching when reacting with •OH according to Muff et al. mechanism of the oxidation of pNDA by •OH [61]. In this work, pNDA bleaching followed a pseudo-ﬁrst-order equation, so the apparent rate constant was calculated by ln(C/C0) = k1t, where C0 is the initial concentration, C is the reaction concentration at a given time, and k1 is the pseudo-ﬁrst-order reaction rate constant. The slope of the plot after applying a linear ﬁt represents the rate constant, k1. Because the relationship between pNDA bleaching and •OH production follows a 1:1 stoichiometry [61], the steady-state of •OH generation ([•OH]ss) can be considered equal to the initial velocity (r0) according to Equation (1) and reported in Table 3: [pNDA] dt t=0 = r0 = [•OH]ss (1) (1) Fe-TiO2 materials showed a similar anatase:rutile phase ratio, particle size, and speciﬁc surface area, and therefore the variation in r0 values was due to the difference of Fe content inside TiO2. The generation of •OH radicals (r0) was feasible using zero-iron TiO2, Fe-TiO2 materials, and Aeroxide® TiO2 P25 under both high (Figure 10a) and low UV irradiation (Figure 10b). 373 Catalysts 2018, 8, 625 Table 3. •OH generation rate of zero-iron TiO2 and Fe-TiO2. Catalyst at.% Load High UV Irradiation Low UV Irradiation k1 R2 r0 k1 R2 r0 mg L−1 min−1 μM•OH min−1 min−1 μM•OH min−1 TiO2 Aeroxide® P25 - 20 0.06 0.988 0.49 0.012 0.989 0.105 Zero-iron TiO2 0 320 0.056 0.993 0.49 0.005 0.973 0.045 0.3 Fe-TiO2 0.3 320 0.067 0.998 0.58 0.004 0.990 0.042 0.6 Fe-TiO2 0.6 320 0.031 0.998 0.28 0.002 0.999 0.025 1.0 Fe-TiO2 1 320 0.004 0.987 0.04 0.00002 0.891 0.0002 Figure 10. •OH generation (initial velocity) of zero-iron TiO2 and Fe-TiO2 under high UV irradiation (a) and low UV Irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2 at pH 6 ± 0.1, and 20 ◦C. Table 3. •OH generation rate of zero-iron TiO2 and Fe-TiO2. Table 3. •OH generation rate of zero-iron TiO2 and Fe-TiO2. 2.3. Hydroxyl Radical Generation under High and Low UV Irradiation Additionally, photogenerated electrons in the midgap band (T2g) can be transferred to Fe3+ following a dark redox reaction at the interface, as suggested by Neubert et al. [63] and consequently bring about •OH. 374 Catalysts 2018, 8, 625 Figure 11. Photocatalytic mechanism of Fe-TiO2 and •OH generation. Eg is band gap energy, E is photon energy, OMads is adsorbed organic molecule, OMoxi is oxidized organic molecule. Figure 11. Photocatalytic mechanism of Fe-TiO2 and •OH generation. Eg is band gap energy, E is photon energy, OMads is adsorbed organic molecule, OMoxi is oxidized organic molecule. Figure 11. Photocatalytic mechanism of Fe-TiO2 and •OH generation. Eg is band gap energy, E is photon energy, OMads is adsorbed organic molecule, OMoxi is oxidized organic molecule. Increasing the Fe3+ doping content of Fe-TiO2 to 0.6 and 1.0 at.%, Fe-TiO2 was unfavorable to the photocatalytic activity because the additional Fe3+ doping in the TiO2 sample inhibited the extended lifetime of charge carriers, acted as recombination sites and consequently decreased the photocatalytic efﬁciency [29], as proposed in Equations (6)–(9) [39]. Fe2+ + hvb + →Fe3+ recombination (6) Fe4+ + ecd−→Fe3+ recombination (7) Fe4+ + Fe2+ →2Fe3+ recombination (8) Fe4+ + Ti3+ →Fe3+ + Ti4+ recombination (9) (6) (9) When low UV irradiation conditions were used, the r0 values for zero-iron TiO2 and Fe-TiO2 materials were lower than the value estimated for Aeroxide® TiO2 P25. Compared with the effects of high UV irradiation, the reduction in r0 value observed was related both to pNDA adsorption of UV-visible radiation (lowered the number of photons available to activate the photocatalyst), and the augmented Fe content, which increased the recombination rate. 2.4. Photocatalytic Degradation of Estriol under High and Low UV Irradiation 2.4. Photocatalytic Degradation of Estriol under High and Low UV Irradiation E3 photocatalytic degradation curves are shown in Figure 12a,b using both high and low UV irradiation, respectively. In both cases, E3 photocatalytic degradation followed a pseudo-ﬁrst-order model and the rate constant, k1 (Table 4), was obtained by ﬁtting experimental data to ln ([E3]/[E30]) = k1t. Fe content inﬂuenced k1 for both high and low UV irradiation. 375 Catalysts 2018, 8, 625 Figure 12. Photocatalytic degradation of E3 under high UV irradiation (a), and low UV irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, 1.0 Fe-TiO2, and Aeroxide® TiO2 P25; at pH 6 ± 0.1, and 20 ◦C. Figure 12. Photocatalytic degradation of E3 under high UV irradiation (a), and low UV irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, 1.0 Fe-TiO2, and Aeroxide® TiO2 P25; at pH 6 ± 0.1, and 20 ◦C. Table 4. Kinetic values of E3 degradation using zero-iron TiO2 and Fe-TiO2. Table 4. Kinetic values of E3 degradation using zero-iron TiO2 and Fe-TiO2. Catalyst Load High UV Irradiation Low UV Irradiation k1 R2 r0,E3 k1 R2 r0,E3 mg L−1 min−1 μME3 min−1 min−1 μME3 min−1 TiO2 Aeroxide® P25 20 0.021 0.996 0.21 0.0029 0.992 0.030 Zero-iron TiO2 320 0.007 0.997 0.069 0.0045 0.991 0.040 0.3 Fe-TiO2 320 0.009 0.994 0.090 0.0050 0.992 0.042 0.6 Fe-TiO2 320 0.011 0.997 0.099 0.0034 0.999 0.030 1.0 Fe-TiO2 320 0.003 0.979 0.027 0.0016 0.987 0.012 Figure 13 shows the pseudo-ﬁrst-order rate constant (k1) of E3 photocatalytic degradation. In general, the photocatalytic activity ﬁrst increased and then decreased as the Fe concentration increased, which is similar to the behavior found with the •OH probe in Section 2.3 and has been previously reported using other organic molecules [23,29,64]. Figure 13. Photocatalytic reaction rate (k1) for degradation of E3 under high UV irradiation (a), and low UV irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2; at pH 6 ± 0.1, and 20 ◦C. Figure 13. Photocatalytic reaction rate (k1) for degradation of E3 under high UV irradiation (a), and low UV irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2; at pH 6 ± 0.1, and 20 ◦C. Under high UV irradiation (Figure 13a), 0.6 Fe-TiO2 k1 was higher than for zero-iron TiO2, 0.3 Fe-TiO2, and 1.0 Fe-TiO2. 2.4. Photocatalytic Degradation of Estriol under High and Low UV Irradiation The increase in photocatalytic performance of 0.6 Fe-TiO2 was related with the increase in the lifetime of electron-hole pairs because Fe created additional energy levels near the conduction band of TiO2, as the mechanism suggests in Figure 11. 376 Catalysts 2018, 8, 625 Under low UV irradiation (Figure 13b), zero-iron TiO2, 0.3 Fe-TiO2, and 0.6 Fe-TiO2 showed more photocatalytic activity than Aeroxide® TiO2 P25 because those materials had enhanced superﬁcial properties, such as particle size, and superﬁcial area, as mentioned in Section 2.1. Furthermore, 0.3 Fe-TiO2 enhanced photocatalytic activities with k1 values as high as 0.005 min−1. The high photocatalytic activity of 0.3 Fe-TiO2 was due to the synergistic effect of unintentionally added co-dopants, superﬁcial properties, and Fe content that increased the lifetime of photogenerated charge carriers and the efﬁciency of electron transfer. The photocatalytic degradation rate of E3 using Aeroxide® TiO2 P25 was reported to be 0.25 min−1 [65], 0.134 min−1 [66], and 0.12 min−1 [67]. However, the experimental setups and catalyst loads were different. Besides these few studies, E3 degradation using Fe-TiO2 nanoparticles is scarcely reported. Only comparing magnitudes of k1, the ﬁrst-order rates to degrade pharmaceuticals using Fe-TiO2 nanoparticles were 0.001 min−1 for ibuprofen, 0.0015 min−1 for carbamazepine, and 0.0014 min−1 for sulfamethoxazole [68], which are in the order of magnitude obtained in this work (see Table 4). Regarding unintentionally added co-dopants, Fe-TiO2 co-doping demonstrated a synergistic effect to increase photocatalytic activity under visible light for sulfur [69], nitrogen [44], and FexTi1-xO2-yNy co-doping [70]. Surface properties of the material, such as a particle size (6.9 nm) and surface area (77.6 m2 g−1), also facilitated the mass transfer between interface, E3, and sub-products. The relationship between the •OH radical system and E3 kinetic degradation was determined via linear ﬁt between •OH initial rate generation (r0,OH) and initial E3 degradation (r0,E3). In general, the procedure to correlate r0,OH and r0,E3 was ﬁrst to sort pair values (r0,OH, r0,E3), and then ﬁt the data to linear regression, as shown Figure 14a,b. Figure 14. Correlation between •OH initial rate generation (r0,OH) and initial E3 degradation (r0,E3) under high UV irradiation (a), and low UV irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2; at pH 6 ± 0.1, and 20 ◦C. Figure 14. 2.4. Photocatalytic Degradation of Estriol under High and Low UV Irradiation Correlation between •OH initial rate generation (r0,OH) and initial E3 degradation (r0,E3) under high UV irradiation (a), and low UV irradiation (b); where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2; at pH 6 ± 0.1, and 20 ◦C. Under high UV irradiation, the linear ﬁt correlation was r0,E3 = 0.091 r0,OH + 0.040 with R2 = 0.197. Under low UV irradiation, the linear ﬁt correlation was r0,E3 = 0.066 r0,OH + 0.012 with R2 = 0.975. The correlation between the pair (r0,OH, r0,E3) under high UV irradiation was too low to be considered a linear relationship. We suggest the low correlation was because not only •OH caused E3 degradation, but holes (h+) or other reactive oxygen species also caused E3 degradation. However, a linear relationship under low UV irradiation was attributable to •OH being the main reactive oxygen species responsible for photocatalytic activity. Therefore, the contribution of h+ to photocatalytic activity was lower because oxidation power was lower due to reduced Eg. This suggestion supports the mechanisms proposed in Figure 11, in which adding Fe into the lattice of TiO2 reduced the Eg with a consistent reduction of redox potential, as mentioned by others [28]. The main mechanism of E3 degradation under low UV irradiation was via electron (e−) transfer to give rise •OH. Additionally, the enhanced photocatalytic activity of 0.3 Fe-TiO2 under low UV 377 Catalysts 2018, 8, 625 irradiation provides evidence that the trapping-recombination mechanism of Fe-TiO2 can be controlled by irradiation intensity. Therefore, we suggest that there is a trade-off between irradiation intensity, the trapping-recombination rate, and •OH production that is worthy of further research. The efﬁciency resource of the Fe-TiO2/Low UV system was obtained through dimensional analysis of the slope of the linear ﬁt of data shown in Figure 14b. The units of slope are E3 moles degraded per •OH mol generated at initial time, so 0.662 E3 molecules underwent degradation when one •OH was generated for the photocatalytic system independent of Fe doping content in TiO2. A sustainable process was also achieved, for which 0.3 Fe-TiO2 since absorbed 8.21% of emission spectra of the lamp below the equivalent Eg wavelength over 0.8% or 7.64% of Aeroxide® TiO2 P25 and zero-iron TiO2, respectively. 2.5. Relationship between Fe Content and Kinetic Constant 2.5. Relationship between Fe Content and Kinetic Constant Photonic efﬁciency has been suggested to increase linearly with the doping ratio due to the formation of the charge carrier trapping centers, while it concurrently decreases quadratically with the doping ratio because to the creation of recombination centers [71]. Alternatively, we suggest an empirical relationship between the E3 degradation pseudo-ﬁrst-order rate constant (k1) and Fe content (at.%) in TiO2, as described in Equation (10): k1(δ) = c e−ke(δ+α) −e−ka(δ+α) (10) (10) where k1 is the pseudo-ﬁrst-order constant, ke is the electron trap constant, ka is the electron recombination constant, δ at.% is the Fe doping amount in TiO2, and c and α are system constants. To solve the model described in Equation (10), a numerical approximation by root-mean-square error minimization method was used according to Equation (11): ε = 1 n ∑ i [k1.i] −[k1.i] (11) (11) where [k1.i] is the theoretical k1 value, [k1.i] is the experimental k1 value, n is the number of data, and ε is the root-mean-square error. The solution of Equation (10) was performed by simultaneously solving ke, ka, c, and α using Excel Solver® (Frontline Systems, NV, US). As an example, photocatalytic degradation of E3 under low UV irradiation was ﬁtted to Equation (10), as shown in Figure 15. where [k1.i] is the theoretical k1 value, [k1.i] is the experimental k1 value, n is the number of data, and ε is the root-mean-square error. The solution of Equation (10) was performed by simultaneously solving ke, ka, c, and α using Excel Solver® (Frontline Systems, NV, US). As an example, photocatalytic degradation of E3 under low UV irradiation was ﬁtted to Equation (10), as shown in Figure 15. The empirical model solved in Equation (12) shows that electron trap constant (ke) overcome electron recombination (ka) before optimal catalyst load. This model could lead to experimental work using iron-doped TiO2 in which the optimal content of Fe gives rise to the maximum E3 degradation. k1(δ) = −1.99 e−2.81(δ+0.197) −e−2.78(δ+0.197) (12) (12) 378 atalysts 2018, 8, 625 Figure 15. Experimental relationship between pseudo ﬁrst order constant and at.% content; where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2; at pH 6 ± 0.1; and 20 ◦C. . Materials and Methods Catalysts 2018, 8, 625 Figure 15. 3.1. Reagents Sigma-Aldrich (St. Louis, MO, USA) supplied estriol (E3, C18H24O3, ≥97%), titanium isopropoxide (TTIP, Ti[OCH(CH3)2]4, 97%), N,N-Dimethyl-4-nitrosoaniline (pNDA, also called RNO, C8H10N20, 97%), sodium dodecyl sulfate (SDS), and iron (III) nitrate (Fe(NO3)3·9H2O, >99.99%). Aeroxide® TiO2 P25 (formerly Degussa P25 with 50 ± 15 m2 g−1 of the speciﬁc surface area, 21 nm of average particle size, 80:20 of anatase:rutile ratio according to the manufacturer) granted by Evonik Industries (Essen, Germany) was the photocatalytic standard. Fremont (CA, USA) supplied HNO3, H2SO4, absolute ethanol, HPLC-grade methanol, and HPLC-grade water. All chemicals were used as received. 2.5. Relationship between Fe Content and Kinetic Constant Experimental relationship between pseudo ﬁrst order constant and at.% content; where zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2; at pH 6 ± 0.1; and 20 ◦C. Materials and Methods 3. Materials and Methods 3.1. Reagents 3.2. Photoreactor Setup Figure 16 depicts the photoreactor, which was a cylindrical water-jacketed glass vessel (318 mL) with 102 mm and 63 mm of interior height and diameter, respectively. The horizontal and vertical position of the photoreactor was constant for all experiments. Lamps were set horizontally and centered above the photoreactor. Two 15 W GE F15T8 BLB lamps (also called black-light lamps, Boston, MA, USA) supplied high UV irradiation, and two 15 W GE F15T8 D lamps (also called daylight lamps) provided low UV irradiation. The overall system was in a closed box to avoid the effects of sunlight or any artiﬁcial radiation sources. Lamp emission spectra were measured using a lab-made spectrophotometer using a CMOS webcam with a diffraction grating of 1000 lines mm−1 [72,73]. Emission spectra calibration of the spectrophotometer was developed using a 9 W ﬂuorescent lamp (Tecnolite, Jalisco, Mexico). The temperature of all experiments was set at 20 ◦C using a thermostatic bath with recirculation (Polystat, Cole-Palmer, Vernon Hills, IL, USA). An optical ﬁlter was not used in the experiments, so visible light condition was not simulated. 379 Catalysts 2018, 8, 625 Figure 16. Scheme of photoreactor used for experiments: glass reactor (1), testing solution (2), temperature probe (3), spin bar (4), lamps (5), an optical ﬁlter (if needed) (6), stirring plate (7), cooling fan (8), horizontal position template (9), and lab jack lifting platform (10). Figure 16. Scheme of photoreactor used for experiments: glass reactor (1), testing solution (2), temperature probe (3), spin bar (4), lamps (5), an optical ﬁlter (if needed) (6), stirring plate (7), cooling fan (8), horizontal position template (9), and lab jack lifting platform (10). 3.3. Synthesis of Materials The synthesis method of iron-doped TiO2 (Fe-TiO2) materials followed the hydrothermal sol-gel synthetic approach proposed by Patra et al. with some differences in precursor and thermal treatment [49]. Our synthesis method used iron (III) nitrate instead of FeCl3 and absolute ethanol instead of isopropyl alcohol. The thermal treatment was a programmed cycle of 31 h (increasing ramp-drying-increasing ramp-calcination-decreasing ramp) instead of direct calcination for 6 h. First, solution A was prepared by dissolving 1.44 g of SDS in 10 mL of deionized water. Then, four different solutions B were prepared to dissolve iron (III) nitrate in 2 mL of absolute ethanol (≥99.8 %) and 3 mL of TTIP was added slowly. The amounts of iron (III) nitrate were 0, 0.4, 4.3, and 42.6 mg of Fe(NO3)3·9H2O identiﬁed as zero-iron TiO2, 0.3 Fe-TiO2, 0.6 Fe-TiO2, and 1.0 Fe-TiO2, respectively. Once ready, solution A was continuously stirred and solution B was slowly dropped into solution A. The pH of the resulting mixture was adjusted to 1 using concentrated HNO3 and stirred for 3 h. The mixture was kept at 3 ◦C for 36 h. The precipitated solid was collected by ﬁltration using Whatman Quantitative Filter Paper Grade 42. The materials were simultaneously dried and calcinated with a programmed thermal treatment (Isotemp® Programmable Mufﬂe Furnace, Fisher Scientiﬁc, Dubuque, IA, USA) following ﬁrst the temperature increase from ambient temperature to 353 K, with a temperature ramp of 1 K min−1 that was held for 720 min. The temperature was then increased from 353 K to 773 K with a temperature ramp of 1 K min−1 that was held for 360 min. Finally, the temperature was decreased from 773 K to 353 K with a temperature ramp of −1 K min−1, and then the furnace was turned off. The materials were washed with 50:50 methanol-water and dried to 377 K overnight. 3.4. Materials Characterization X-ray photoelectron spectroscopy (XPS) was performed using a Thermo Fisher Scientiﬁc K-Alpha X-ray photoelectron spectrometer (Waltham, MA, USA) with a monochromatized Al Kα X-ray source (1487 V). The deconvolution of high-resolution XPS spectra was developed using the software XPSpeak 4.1. (Raymund W.M. Kwok, Shatin, Hong Kong). 380 Catalysts 2018, 8, 625 Catalysts 2018, 8, 625 UV-visible reﬂectance spectroscopy was obtained with Video–Barrelino integrating sphere coupled to Cary 50 spectrophotometer (Varian Inc, Palo Alto, CA, USA). Diffuse reﬂectance spectra were transformed using the Kubelka–Munk method to obtain Eg of zero-iron TiO2 and Fe-TiO2 materials. Kubelka–Munk method plots (F(R)hv)1/2 versus hv, draws a tangent at the inﬂection point on the curve and estimates Eg with the hv value at the intersection with abscissa. In this case, F(R) is a reﬂectance function equal to (1 −R)2/2R, R is the reﬂectance percentage, h is the Planck’s constant, and v is frequency. XRD patterns were recorded in a Siemens D-5000 diffractometer (Munich, Germany) using Cu Kα radiation (λ = 1.54060 Å) from 10◦to 85◦. The procedure for phase identiﬁcation used the QualX2.0 software with database developed by Altomare et al. [74]. The cards used for identiﬁcation were 00-901-5929, 00-900-1681, and 00-900-4140 for anatase, rutile, and brookite, respectively. The quantiﬁcation phases followed the method proposed by Spurr and Myers according to Equation (13): f = 1 1 + 1.26 IR IA (13) (13) where f is the anatase percentage, IA is intensity at a diffraction angle 2θ of 25.36◦, and IR is intensity at a diffraction angle 2θ of 27.46◦[75]. where f is the anatase percentage, IA is intensity at a diffraction angle 2θ of 25.36◦, and IR is intensity at a diffraction angle 2θ of 27.46◦[75]. The particle size was estimated by Scherrer’s formula described in Equation (14), where β is the full width at half of the maximum of the diffraction peaks (radians), k is the shape constant, λ is the wavelength of the incident Cu Kα radiation (λ = 1.54060 Å), θ is the Bragg’s angle (radians), and D is the particle size (Å). D = k λ β cos θ (14) (14) Brunauer–Emmett–Teller (BET) isotherms were obtained in Nova Station A equipment (Quantachrome Instruments, Boynton Beach, FL, USA). The surface morphology was observed by SEM in a JEOL ultrahigh resolution ﬁeld emission electron microscope JSM-7800 F (JEOL, Tokyo, Japan) with 20 kV accelerating voltage, and 3 mm WD. 3.4. Materials Characterization Transmission electron microscopy (TEM) images were obtained in a JEM-2100 LaB6 electron microscope (JEOL, Tokyo, Japan). 3.5. Hydroxyl Radical Generation In this study, pNDA bleaching was selected as an •OH probe because pNDA was useful for measuring the photocatalytic performance of TiO2 [51,76,77] because of the following advantages: (1) it is selective of the reaction of pNDA with •OH [78]; (2) its high reaction rate with •OH on the order of 1010 M−1 s−1 [51,79]; (3) its easy application through observable bleaching at 440 nm following Beer’s Law, in which pNDA bleaching a yellowish solution to transparent; and (4) its 1:1 stoichiometry, meaning that one •OH can bleach one pNDA molecule [51,80–82]. The pNDA absorption (Figure 17) measurements were obtained using a UV-visible spectrophotometer (Hatch DR/4000U, Loveland, CO, USA) at 440 nm following Beer-Lambert law. The pNDA test solution was 10 μM initial concentration and pH 6.0 ± 0.1 adjusted using NaOH or HCl when needed. No buffer solutions were used because they can compete for •OH. Final pH was veriﬁed at the end of tests to discharge pH-pNDA bleaching. 381 Catalysts 2018, 8, 625 Figure 17. Structural formula and absorbance spectrum of N,N-dimethyl-p-nitrosoaniline (pNDA). Figure 17. Structural formula and absorbance spectrum of N,N-dimethyl-p-nitrosoaniline (pNDA). Figure 17. Structural formula and absorbance spectrum of N,N-dimethyl-p-nitrosoaniline (pNDA) The photocatalytic standard was Aeroxide® TiO2 P25, and the load was 20 mg L−1. The choice of catalyst load was based on our previous work on •OH generation of Aeroxide® TiO2 P25 [16]. For zero-iron TiO2 and Fe-TiO2 materials, the catalyst load used was 320 mg L−1, which produced a •OH generation rate under high UV irradiation to set a baseline. Catalyst load differences were attributable to the aggregation of lab-made TiO2, superﬁcial properties, and optical properties of suspensions, as shown in Figure 18. Figure 18. Suspension transmittance of Fe-TiO2 material and Aeroxide TiO2 P25; where zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). Figure 18. Suspension transmittance of Fe-TiO2 material and Aeroxide TiO2 P25; where zero-iron TiO2 (a), 0.3 Fe-TiO2 (b), 0.6 Fe-TiO2 (c), and 1.0 Fe-TiO2 (d). The photocatalytic experiments were conducted as follows. First, a pNDA test solution was set at 20 ◦C, the catalyst was added, and the suspension was mixed for 20 min without radiation. To evaluate the adsorption of pNDA on TiO2, an aliquot was withdrawn and centrifuged. Then, the system was fully illuminated, and aliquots were withdrawn after speciﬁc periods. 3.7. Analytical Methods The E3 concentration was monitored using an HPLC system (Waters 1515; Milford, MA, USA) equipped with a UV detector (Waters 2787) that has an injection volume of 20 μL. The analytical method was performed in isocratic analytical mode using an Inertsil® ODS-3 column (GL Science, Tokyo, Japan; 150 mm × 4.6 mm, 5 μm) thermostated at 25 ◦C. The wavelength was at 280 nm according to E3 maximum absorbance. The mobile phase was methanol (49%) and deionized water (51%) at a ﬂow rate of 1 mL min−1. The retention time of E3 was 10 min, and the limit of E3 detection was 0.1 μM (0.029 mg L−1). The detection limit was obtained by developing two calibration curves: the ﬁrst between 10 and 0.1 and second between 1 and 0.01. Both calibration curves followed area = 2928[E3] with R2 = 0.9899, but areas below 0.1 were not detected. 3.6. Photolysis and Photocatalytic Degradation of E3 3.6. Photolysis and Photocatalytic Degradation of E3 3.6. Photolysis and Photocatalytic Degradation of E3 The initial E3 concentration was 10 μM because (1) this research was part of a project focused on the removal of E3 in water using sequentially coupled membrane ﬁltration; (2) the solubility limit of E3 in water was previously reported to be 11.1 μM [83], and 45.1 μM [8,84], and (3) the sensitivity of the analytical techniques used in this work. The E3 solution was prepared to dissolve 2.88 mg of E3 in 1 L of deionized water by stirring at room conditions in the dark for six hours. Working solutions were stored in an amber ﬂask. Each photocatalytic experiment used 100 mL of E3 working solution. Initial pH was adjusted to obtain a similar surface charge of TiO2 [85]. Depending on the initial water conditions, the initial pH value was adjusted to 6.0 ± 0.1 using NaOH or HCl when needed. A dark period (no radiation) was allowed for 20 min. Then, similar experimental conditions were carried out as described in Section 3.5. Additionally, the aliquots withdrawn from suspension were ﬁltered using a 0.1 μm syringe ﬁlter (MillexVV, Millipore, Billerica, MA, USA). A blank experiment without irradiation and TiO2 photocatalyst was conducted for comparison. The blank experiment showed that E3 cannot be degraded in absences of either TiO2 or UV light. Once the catalyst was loaded and after the dark phase, no adsorption of E3 was detected near the detection limit of HPLC. Author Contributions: Funding acquisition, E.R.B.; Investigation, I.M.R.-S.; Project administration, E.R.B.; Supervision, E.R.B.; Writing—original draft, I.M.R.-S.; Writing—review & editing, E.R.B. 3.5. Hydroxyl Radical Generation Each sample was centrifuged at 6000 rpm for 15 min (Biofuge Primo, Sorvall, Hanau, Germany) and measured in the UV-visible spectrophotometer. Once the catalyst load was used and after the dark phase, no adsorption of pNDA was detected near the detection limit of UV-visible spectrophotometer. 382 Catalysts 2018, 8, 625 Catalysts 2018, 8, 625 4. Conclusions This study provided an understanding of the relationship between the Fe doping ratio and radiation intensity for •OH generation and estriol (E3) degradation. The main results were that: • E3 degradation using 0.3 Fe-TiO2 was feasible and can be improved by controlling irradiation intensity which was found closely related with light absorption and the catalytic reaction rate; • the synthesis method and thermal treatment allowed nanoparticles with large superﬁcial areas and the incorporation of iron ions into the TiO2 lattice.; and • E3 degradation using 0.3 Fe-TiO2 was feasible and can be improved by controlling irradiation intensity which was found closely related with light absorption and the catalytic reaction rate; • the synthesis method and thermal treatment allowed nanoparticles with large superﬁcial areas and the incorporation of iron ions into the TiO2 lattice.; and • E3 degradation using 0.3 Fe-TiO2 was feasible and can be improved by controlling irradiation intensity which was found closely related with light absorption and the catalytic reaction rate; y y g p y ; • the synthesis method and thermal treatment allowed nanoparticles with large superﬁcial areas and the incorporation of iron ions into the TiO2 lattice.; and • changes in trapping recombination centers could be controlled with irradiation intensity to enhance the photocatalytic activity. • changes in trapping recombination centers could be controlled with irradiation intensity to enhance the photocatalytic activity. Therefore, our ﬁndings provide the opportunity to reconsider studies in which iron-doped TiO2 impaired photocatalytic activity and to improve an application in which irradiation should be controlled. For example, Fe-TiO2 can potentially be applied to medical uses in which low irradiation intensity should be used to avoid adverse effects in humans or wildlife, which has also been suggested by others [86]. In the ﬁeld of water treatment, we propose that Fe-TiO2 is an efﬁcient material that could harvest low-energy photons to degrade and mineralize dyes [87], biocides [88], pharmaceuticals [89], industrial chemicals [90], and estrogens—as shown in this study—to create an energetically green water treatment process. Author Contributions: Funding acquisition, E.R.B.; Investigation, I.M.R.-S.; Project administration, E.R.B.; Supervision, E.R.B.; Writing—original draft, I.M.R.-S.; Writing—review & editing, E.R.B. 383 Catalysts 2018, 8, 625 Catalysts 2018, 8, 625 Funding: This manuscript is based on work supported in part by ConTex postdoctoral program, which is an initiative of the University of Texas System and Mexico’s National Council of Science and Technology (CONACYT). 4. Conclusions The research was partially funded by CONACYT under Project CB-2011/168285. The APC was funded by the Institutional Open Access Program (IAOP) between The University of Texas at Austin and Desert Research Institute (DRI) at Nevada. Acknowledgments: The Aeroxide® P25 Evonik catalyst used for this work was provided by Intertrade S.A. de C.V., the supplier of Evonik Industries in Mexico. The authors thank L. Lartundo-Rojas, Raul Borja Urbi, Hugo Martinez Gutiérrez, and Joao Jairzinho Salinas Camargo for assistance in XPS spectroscopy, TEM images, SEM images, and absorption isotherms, respectively, all of whom are from Centro de Nanociencias y Micro y Nanotecnología (CNMN) of IPN, Mexico. The authors thank M.A. Quiroz Alfaro for his excellent technical help and for his permission to use materials and equipment at the UDLAP’s electrochemical lab. The authors also thank Nicole Damon (DRI) for her editorial review. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Dulio, V.; van Bavel, B.; Brorström-Lundén, E.; Harmsen, J.; Hollender, J.; Schlabach, M.; Slobodnik, J.; Thomas, K.; Koschorreck, J. Emerging pollutants in the EU: 10 years of NORMAN in support of environmental policies and regulations. Environ. Sci. Eur. 2018, 30, 5. [CrossRef] 2. Mandaric, L.; Celic, M.; Marcé, R.; Petrovic, M. Introduction on Emerging Contaminants in Rivers and Their Environmental Risk. In Emerging Contaminants in River Ecosystems: Occurrence and Effects under Multiple Stress Conditions; Petrovic, M., Sabater, S., Elosegi, A., Barceló, D., Eds.; Springer International Publishing: Cham, Switzerland, 2016; pp. 3–25. ISBN 978-3-319-29376-9. 3. Houtman, C.J.; Legler, J.; Thomas, K. Effect-Directed Analysis of Complex Environmental Contamination; Brack, W., Ed.; Springer: Berlin/Heidelberg, Germany, 2011; pp. 237–265. ISBN 978-3-642-18384-3. 3. Houtman, C.J.; Legler, J.; Thomas, K. Effect-Directed Analysis of Complex Environmental Contamination; Brack, W., Ed.; Springer: Berlin/Heidelberg, Germany, 2011; pp. 237–265. ISBN 978-3-642-18384-3. 4. Dimogerontas, G.; Liapi, C. Endocrine Disruptors (Xenoestrogens): An Overview. In Plastics in Dentistry and Estrogenicity: A Guide to Safe Practice; Eliades, T., Eliades, G., Eds.; Springer: Berlin/Heidelberg, Germany, 2014; pp. 3–48. ISBN 978-3-642-29687-1. 4. Dimogerontas, G.; Liapi, C. Endocrine Disruptors (Xenoestrogens): An Overview. In Plastics in Dentistry and Estrogenicity: A Guide to Safe Practice; Eliades, T., Eliades, G., Eds.; Springer: Berlin/Heidelberg, Germany, 2014; pp. 3–48. ISBN 978-3-642-29687-1. Estrogenicity: A Guide to Safe Practice; Eliades, T., Eliades, G., Eds.; Springer: Berlin/Heidelberg, Germany, 2014; pp. 3–48. ISBN 978-3-642-29687-1. pp 5. Hileman, B. Environmental Estrogens linked to Reproductive Abnormalities, Cancer. Chem. Eng. News Arch. 1994, 72, 19–23. [CrossRef] rat, N.; Rieradevall, M.; Barata, C.; Munné, A. The combined use of metrics of biological quality and omarkers to detect the effects of reclaimed water on macroinvertebrate assemblages in the lower part of a 6. Prat, N.; Rieradevall, M.; Barata, C.; Munné, A. The combined use of metrics of biological quality and biomarkers to detect the effects of reclaimed water on macroinvertebrate assemblages in the lower part of a polluted Mediterranean river (Llobregat River, NE Spain). Ecol. Indic. 2013, 24, 167–176. [CrossRef] polluted Mediterranean river (Llobregat River, NE Spain). Ecol. Indic. 2013, 24, 167–176. [CrossRef] 7. Souza, M.S.; Hallgren, P.; Balseiro, E.; Hansson, L.A. Low concentrations, potential ecological consequences: Synthetic estrogens alter life-history and demographic structures of aquatic invertebrates. Environ. Pollut. 2013, 178, 237–243. [CrossRef] 8. Silva, C.P.; Otero, M.; Esteves, V. Processes for the elimination of estrogenic steroid hormones from water: A review. Environ. Pollut. 2012, 165, 38–58. [CrossRef] 9. References Rodriguez-Narvaez, O.M.; Peralta-Hernandez, J.M.; Goonetilleke, A.; Bandala, E.R. Treatment techn for emerging contaminants in water: A review. Chem. Eng. J. 2017, 323, 361–380. [CrossRef] G ˛agol, M.; Przyjazny, A.; Boczkaj, G. Wastewater treatment by means of advanced oxidation processes based on cavitation—A review. Chem. Eng. J. 2018, 338, 599–627. [CrossRef] 10. G ˛agol, M.; Przyjazny, A.; Boczkaj, G. Wastewater treatment by means of adv on cavitation—A review. Chem. Eng. J. 2018, 338, 599–627. [CrossRef] 11. Boczkaj, G.; Fernandes, A. Wastewater treatment by means of advanced oxidation processes at basic pH conditions: A review. Chem. Eng. J. 2017, 320, 608–633. [CrossRef] 12. Coleman; Eggins, B.; Byrne, J.A.; Palmer, F.L.; King, E. Photocatalytic degradation of 17-β-oestradiol on immobilised TiO2. Appl. Catal. B Environ. 2000, 24, L1–L5. [CrossRef] 13. Ramirez-Sanchez, I.M.; Mendez-Rojas, M.A.; Bandala, E.R. CHAPTER 25 Photocatalytic Degradation of Natural and Synthetic Estrogens with Semiconducting Nanoparticles. In Advanced Environmental Analysis: Applications of Nanomaterials; The Royal Society of Chemistry: London, UK, 2017; Volume 2, pp. 153–177. ISBN 978-1-78262-906-1. 14. Ohko, Y.; Iuchi, K.; Niwa, C.; Tatsuma, T.; Nakashima, T.; Iguchi, T.; Kubota, Y.; Fujishima, A. 17β-Estradiol Degradation by TiO2 Photocatalysis as a Means of Reducing Estrogenic Activity. Environ. Sci. Technol. 2002, 36, 4175–4181. [CrossRef] 384 Catalysts 2018, 8, 625 15. Coleman, H.M.; Routledge, E.J.; Sumpter, J.P.; Eggins, B.R.; Byrne, J.A. Rapid loss of estrogenicity of steroid estrogens by UVA photolysis and photocatalysis over an immobilised titanium dioxide catalyst. Water Res. 2004, 38, 3233–3240. [CrossRef] 16. Ramírez-Sánchez, I.M.; Tuberty, S.; Hambourger, M.; Bandala, E.R. Resource efﬁciency analysis for photocatalytic degradation and mineralization of estriol using TiO2 nanoparticles. Chemosphere 2017, 184, 1270–1285. [CrossRef] 17. Hashimoto, K.; Irie, H.; Fujishima, A. Photocatalysis: A Historical Overview and Future Prospects. Jpn. J. Appl. Phys. 2005, 44, 8269–8285. [CrossRef] 18. Fujishima, A.; Zhang, X.; Tryk, D.A. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2008, 63, 515–582. [CrossRef] 19. Tong, A.Y.C.; Braund, R.; Warren, D.S.; Peake, B.M. TiO2-assisted photodegradation of pharmaceuticals—A review. Cent. Eur. J. Chem. 2012, 10, 989–1027. [CrossRef] 20. Cassaignon, S.; Colbeau-Justin, C.; Durupthy, O. Titanium dioxide in photocatalysis. In Nanomaterials: A Danger or a Promise?: A Chemical and Biological Perspective; Springer: London, UK, 2013; pp. 153–188. ISBN 9781447142133. 21. Augugliaro, V.; Loddo, V.; Pagliaro, M.; Palmisano, G.; Palmisano, L. Clean by Light Irradiation: Practical Applications of Supported TiO2; RSC Publishing: Cambridge, UK, 2010; ISBN 1847558704. 22. References Etacheri, V.; Di Valentin, C.; Schneider, J.; Bahnemann, D.; Pillai, S.C. Visible-light activation of TiO2 photocatalysts: Advances in theory and experiments. J. Photochem. Photobiol. C Photochem. Rev. 2015, 25, 1–29. [CrossRef] 23. Wen, L.; Liu, B.; Zhao, X.; Nakata, K.; Murakami, T.; Fujishima, A. Synthesis, Characterization, and Photocatalysis of Fe-Doped TiO2: A Combined Experimental and Theoretical Study. Int. J. Photoenergy 2012, 2012, 1–10. [CrossRef] 24. Yu, H.; Irie, H.; Hashimoto, K. Conduction band energy level control of titanium dioxide: Toward an efﬁcien visible-light-sensitive photocatalyst. J. Am. Chem. Soc. 2010, 132, 6898–6899. [CrossRef] 25. Choi, W.; Termin, A.; Hoffmann, M.R. The role of metal ion dopants in quantum-sized TiO2: Correlation between photoreactivity and charge carrier recombination dynamics. J. Phys. Chem. 1994, 98, 13669–13679. [CrossRef] 26. Kaur, T.; Sraw, A.; Wanchoo, R.K.; Toor, A.P. Visible–Light Induced Photocatalytic Degradation of Fungicide with Fe and Si Doped TiO2 Nanoparticles. Mater. Today Proc. 2016, 3, 354–361. [CrossRef] 27. Zhao, B.; Mele, G.; Pio, I.; Li, J.; Palmisano, L.; Vasapollo, G. Degradation of 4-nitrophenol (4-NP) using Fe-TiO2 as a heterogeneous photo-Fenton catalyst. J. Hazard. Mater. 2010, 176, 569–574. [CrossRef] 28. Yalçın, Y.; Kılıç, M.; Çınar, Z. Fe+3-doped TiO2: A combined experimental and computational approach to the evaluation of visible light activity. Appl. Catal. B Environ. 2010, 99, 469–477. [CrossRef] 29. Cai, L.; Liao, X.; Shi, B. Using Collagen Fiber as a Template to Synthesize TiO2 and Fex/TiO2 Nanoﬁbers and Their Catalytic Behaviors on the Visible Light-Assisted Degradation of Orange II. Ind. Eng. Chem. Res. 2010, 49, 3194–3199. [CrossRef] 30. Li, J.; Xu, J.; Dai, W.L.; Li, H.; Fan, K. Direct hydro-alcohol thermal synthesis of special core-shell structured Fe-doped titania microspheres with extended visible light response and enhanced photoactivity. Appl. Catal. B Environ. 2009, 85, 162–170. [CrossRef] 31. Tong, T.; Zhang, J.; Tian, B.; Chen, F.; He, D. Preparation of Fe3+-doped TiO2 catalysts by controlled hydrolysis of titanium alkoxide and study on their photocatalytic activity for methyl orange degradation. J. Hazard. Mater. 2008, 155, 572–579. [CrossRef] 32. Ambrus, Z.; Balázs, N.; Alapi, T.; Wittmann, G.; Sipos, P.; Dombi, A.; Mogyorósi, K. Synthesis, structure and photocatalytic properties of Fe(III)-doped TiO2 prepared from TiCl3. Appl. Catal. B Environ. 2008, 81, 27–37. [CrossRef] 33. Cong, Y.; Zhang, J.; Chen, F.; Anpo, M.; He, D. References Preparation, photocatalytic activity, and mechanism of TiO C d d ith it d i (III) J Ph Ch C 2007 111 10618 10623 [C R f] 33. Cong, Y.; Zhang, J.; Chen, F.; Anpo, M.; He, D. Preparation, photocatalytic activity, and mechanism of nano-TiO2 Co-doped with nitrogen and iron (III). J. Phys. Chem. C 2007, 111, 10618–10623. [CrossRef] g, ; g, J ; , ; p , ; , p , p y y, nano-TiO2 Co-doped with nitrogen and iron (III). J. Phys. Chem. C 2007, 111, 10618–10623. [CrossRef] 34. Adán, C.; Bahamonde, A.; Fernández-García, M.; Martínez-Arias, A. Structure and activity of nanosized iron-doped anatase TiO2 catalysts for phenol photocatalytic degradation. Appl. Catal. B Environ. 2007, 72, 11–17. [CrossRef] 34. Adán, C.; Bahamonde, A.; Fernández-García, M.; Martínez-Arias, A. Structure and activity of nanosized iron-doped anatase TiO2 catalysts for phenol photocatalytic degradation. Appl. Catal. B Environ. 2007, 72, 11–17. [CrossRef] 385 Catalysts 2018, 8, 625 35. Yamashita, H.; Harada, M.; Misaka, J.; Takeuchi, M.; Neppolian, B.; Anpo, M. Photocatalytic degradation of organic compounds diluted in water using visible light-responsive metal ion-implanted TiO2 catalysts: Fe ion-implanted TiO2. Catal. Today 2003, 84, 191–196. [CrossRef] p 36. Li, X.; Yue, P.-L.; Kutal, C. Synthesis and photocatalytic oxidation propert nanosemiconductor particles. New J. Chem. 2003, 27, 1264. [CrossRef] 36. Li, X.; Yue, P.-L.; Kutal, C. Synthesis and photocatalytic oxidation properties of iron doped titanium dioxide nanosemiconductor particles. New J. Chem. 2003, 27, 1264. [CrossRef] 37. Zhang, Z.; Wang, C.-C.; Zakaria, R.; Ying, J.Y. Role of Particle Size in Nanocrystalline TiO2-Based Photocatalysts. J. Phys. Chem. B 1998, 102, 10871–10878. [CrossRef] 38. Litter, M.I.; Navío, J.A. Photocatalytic properties of iron-doped titania semiconductors. J. Photochem. Photo A Chem. 1996, 98, 171–181. [CrossRef] 39. Fàbrega, C.; Andreu, T.; Cabot, A.; Morante, J.R. Location and catalytic role of iron species in TiO2:Fe photocatalysts: An EPR study. J. Photochem. Photobiol. A Chem. 2010, 211, 170–175. [CrossRef] y y 40. Seabra, M.P.; Salvado, I.M.M.; Labrincha, J.A. Pure and (zinc or iron) doped titania powders prepared by sol-gel and used as photocatalyst. Ceram. Int. 2011, 37, 3317–3322. [CrossRef] 41. Abazovi´c, N.D.; Mirenghi, L.; Jankovi´c, I.A.; Bibi´c, N.; Šoji´c, D.V.; Abramovi´c, B.F.; ˇComor, M.I. Synthesis and characterization of rutile TiO2 nanopowders doped with iron ions. Nanoscale Res. Lett. 2009, 4, 518–525. [CrossRef] 42. Geissen, V.; Mol, H.; Klumpp, E.; Umlauf, G.; Nadal, M.; van der Ploeg, M.; van de Zee, S.E.A.T.M.; Ritsema, C.J. References Emerging pollutants in the environment: A challenge for water resource management. Int. Soil Water Conserv. Res. 2015, 3, 57–65. [CrossRef] 43. Lin, T.C.; Seshadri, G.; Kelber, J.A. A consistent method for quantitative XPS peak analysis of thin oxide ﬁlms on clean polycrystalline iron surfaces. Appl. Surf. Sci. 1997, 119, 83–92. [CrossRef] 44. Xing, M.; Wu, Y.; Zhang, J.; Chen, F. Effect of synergy on the visible light activity of B, N and Fe co-doped TiO2 for the degradation of MO. Nanoscale 2010, 2, 1233. [CrossRef] [PubMed] 45. Lopez, R.; Gomez, R. Band-gap energy estimation from diffuse reﬂectance measurements on sol-gel and commercial TiO2: A comparative study. J. Sol-Gel Sci. Technol. 2012, 61, 1–7. [CrossRef] 46. Shi, J.; Chen, G.; Zeng, G.; Chen, A.; He, K.; Huang, Z.; Hu, L.; Zeng, J.; Wu, J.; Liu, W. Hydrothermal synthesis of graphene wrapped Fe-doped TiO2 nanospheres with high photocatalysis performance. Ceram. Int. 2018, 44, 7473–7480. [CrossRef] 47. Yu, J.; Xiang, Q.; Zhou, M. Preparation, characterization and visible-light-driven photocatalytic activity of Fe-doped titania nanorods and ﬁrst-principles study for electronic structures. Appl. Catal. B Environ. 2009, 90, 595–602. [CrossRef] 48. Goswami, P.; Ganguli, J.N. Evaluating the potential of a new titania precursor for the synthesis of mesoporous Fe-doped titania with enhanced photocatalytic activity. Mater. Res. Bull. 2012, 47, 2077–2084. [CrossRef] 49. Patra, A.K.; Dutta, A.; Bhaumik, A. Highly ordered mesoporous TiO2-Fe2O3 mixed oxide synthesized by sol-gel pathway: An efﬁcient and reusable heterogeneous catalyst for dehalogenation reaction. ACS Appl. Mater. Interfaces 2012, 4, 5022–5028. [CrossRef] [PubMed] 50. Luttrell, T.; Halpegamage, S.; Tao, J.; Kramer, A.; Sutter, E.; Batzill, M. Why is anatase a better photocatalyst than rutile?—Model studies on epitaxial TiO2 ﬁlms. Sci. Rep. 2014, 4, 4043. [CrossRef] [PubMed] 51. Zang, L.; Qu, P.; Zhao, J.; Shen, T.; Hidaka, H. Photocatalytic bleaching of p-nitrosodimethylaniline in TiO2 aqueous suspensions: A kinetic treatment involving some primary events photoinduced on the particle surface. J. Mol. Catal. A Chem. 1997, 120, 235–245. [CrossRef] 52. Othman, S.H.; Abdul Rashid, S.; Mohd Ghazi, T.I.; Abdullah, N. Fe-Doped TiO2 Nanoparticles Produced via MOCVD: Synthesis, Characterization, and Photocatalytic Activity. J. Nanomater. 2011, 2011, 1–8. [CrossRef] 52. Othman, S.H.; Abdul Rashid, S.; Mohd Ghazi, T.I.; Abdullah, N. Fe-Doped TiO2 Nanoparticles Produced via MOCVD: Synthesis, Characterization, and Photocatalytic Activity. J. Nanomater. 2011, 2011, 1–8. [CrossRef] 53. Teoh, W.Y.; Amal, R.; Mädler, L.; Pratsinis, S.E. Flame sprayed visible light-active Fe-TiO2 for MOCVD: Synthesis, Characterization, and Photocatalytic Activity. References J. Nanomater. 2011, 2011, 1–8. [CrossRef] 53. Teoh, W.Y.; Amal, R.; Mädler, L.; Pratsinis, S.E. Flame sprayed visible light-active Fe-TiO2 for photomineralisation of oxalic acid. Catal. Today 2007, 120, 203–213. [CrossRef] 53. Teoh, W.Y.; Amal, R.; Mädler, L.; Pratsinis, S.E. Flame sprayed visible light-active Fe-TiO2 for photomineralisation of oxalic acid. Catal. Today 2007, 120, 203–213. [CrossRef] 54. Pongwan, P.; Inceesungvorn, B.; Wetchakun, K.; Phanichphant, S.; Wetchakun, N. Highly efﬁcient visible-light-induced photocatalytic activity of Fe-doped TiO2 nanoparticles. Eng. J. 2012, 16, 143–151. [CrossRef] 55. Kruk, M.; Jaroniec, M. Gas adsorption characterization of ordered organic-inorganic nanocomposite materials. Chem. Mater. 2001, 13, 3169–3183. [CrossRef] 56. Limousin, G.; Gaudet, J.P.; Charlet, L.; Szenknect, S.; Barthès, V.; Krimissa, M. Sorption isotherms: A review on physical bases, modeling and measurement. Appl. Geochem. 2007, 22, 249–275. [CrossRef] 386 Catalysts 2018, 8, 625 57. Carvalho, T.C.; La Cruz, T.E.; Tábora, J.E. A photochemical kinetic model for solid dosage forms. Eur. J. Pharm. Biopharm. 2017, 120, 63–72. [CrossRef] [PubMed] p 58. Daugherty, J.P.; Hixon, S.C.; Yielding, K.L. Direct in vitro photoafﬁnity labeling of DNA with daunorubicin, d i i d bid BBA S t N l i A id P t i S th 1979 565 13 21 [C R f] p 58. Daugherty, J.P.; Hixon, S.C.; Yielding, K.L. Direct in vitro photoafﬁnity labeling of DNA with daunorubicin, adriamycin, and rubidazone. BBA Sect. Nucleic Acids Protein Synth. 1979, 565, 13–21. [CrossRef] 58. Daugherty, J.P.; Hixon, S.C.; Yielding, K.L. Direct in vitro photoafﬁnity labeling of DNA with daunorubicin, 59. Hartman, P.E.; Biggley, W.H. Breakthrough of ultraviolet light from various brands of ﬂuorescent lamps: Lethal effects on DNA repair-defective bacteria. Environ. Mol. Mutagen. 1996, 27, 306–313. [CrossRef] 59. Hartman, P.E.; Biggley, W.H. Breakthrough of ultraviolet light from various brands of ﬂuorescent lamps: L h l ff DNA i d f i b i E i M l M t 1996 27 306 313 [C R f] 60. Serpone, N. Relative photonic efﬁciencies and quantum yields in heterogeneous photocatalysis. J. Photochem. Photobiol. A Chem. 1997, 104, 1–12. [CrossRef] 61. Muff, J.; Bennedsen, L.R.; Søgaard, E.G. Study of electrochemical bleaching of p-nitrosodimethylaniline and its role as hydroxyl radical probe compound. J. Appl. Electrochem. 2011, 41, 599–607. [CrossRef] 61. Muff, J.; Bennedsen, L.R.; Søgaard, E.G. Study of electrochemical bleaching of p nitrosodimethylaniline and its role as hydroxyl radical probe compound. J. Appl. Electrochem. 2011, 41, 599–607. [CrossRef] 62. References Zhu, J.; Zheng, W.; He, B.; Zhang, J.; Anpo, M. Characterization of Fe-TiO2 photocatalysts synthesized by hydrothermal method and their photocatalytic reactivity for photodegradation of XRG dye diluted in water. J. Mol. Catal. A Chem. 2004, 216, 35–43. [CrossRef] 63. Neubert, S.; Mitoraj, D.; Shevlin, S.A.; Pulisova, P.; Heimann, M.; Du, Y.; Goh, G.K.L.; Pacia, M.; Kruczała, K.; Turner, S.; et al. Highly efﬁcient rutile TiO2 photocatalysts with single Cu(II) and Fe(III) surface catalytic sites. J. Mater. Chem. A 2016. [CrossRef] 64. Zhou, M.; Yu, J.; Cheng, B. Effects of Fe-doping on the photocatalytic activity of mesoporous TiO2 powd prepared by an ultrasonic method. J. Hazard. Mater. 2006, 137, 1838–1847. [CrossRef] Coleman, H.M.; Vimonses, V.; Leslie, G.; Amal, R. Removal of contaminants of concern in water using advance oxidation techniques Water Sci Technol 2007 55 301–306 [CrossRef] 65. Coleman, H.M.; Vimonses, V.; Leslie, G.; Amal, R. Removal of contaminants o advance oxidation techniques. Water Sci. Technol. 2007, 55, 301–306. [CrossRef] 66. Coleman, H.M.; Abdullah, M.I.; Eggins, B.R.; Palmer, F.L. Photocatalytic degradation of 17[beta]-oestradiol, oestriol and 17[alfa]-ethynyloestradiol in water monitored using ﬂuorescence spectroscopy. Appl. Catal. B Environ. 2005, 55, 23–30. [CrossRef] 67. Coleman, H.M.; Chiang, K.; Amal, R. Effects of Ag and Pt on photocatalytic degradation of endocrine disrupting chemicals in water. Chem. Eng. J. 2005, 113, 65–72. [CrossRef] 68. Lin, L.; Wang, H.; Jiang, W.; Mkaouar, A.R.; Xu, P. Comparison study on photocatalytic oxidation of pharmaceuticals by TiO2-Fe and TiO2-reduced graphene oxide nanocomposites immobilized on optical ﬁbers. J. Hazard. Mater. 2017, 333, 162–168. [CrossRef] [PubMed] 69. Hamadanian, M.; Reisi-Vanani, A.; Behpour, M.; Esmaeily, A.S. Synthesis and characterization of Fe,S-codoped TiO2 nanoparticles: Application in degradation of organic water pollutants. Desalination 2011, 281, 319–324. [CrossRef] 70. Naik, B.; Parida, K.M. Solar Light Active Photodegradation of Phenol over a FexTi1-xO2-yNy Nanophotocatalyst. Ind. Eng. Chem. Res. 2010, 49, 8339–8346. [CrossRef] 71. Bloh, J.Z.; Dillert, R.; Bahnemann, D.W. Zinc Oxide Photocatalysis: Inﬂuence of Iron and Titanium Doping and Origin of the Optimal Doping Ratio. ChemCatChem 2013, 5, 774–778. [CrossRef] 72. Lorenz, R.D. A simple webcam spectrograph. Am. J. Phys. 2014, 82, 169–173. [CrossRef] 73. Widiatmoko, E.; Widayani; Budiman, M.; Abdullah, M.; Khairurrijal. A simple spectrophotometer using common materials and a digital camera. Phys. Educ. 2011, 46, 332–339. [CrossRef] 74. Altomare, A.; Corriero, N.; Cuocci, C.; Falcicchio, A.; Moliterni, A.; Rizzi, R. QUALX2.0: A qualitative phase analysis software using the freely available database POW_COD. J. Appl. References Crystallogr. 2015, 48, 598–603. [CrossRef] 75. Spurr, R.A.; Myers, H. Quantitative Analysis of Anatase-Rutile Mixtures with an X-ray Diffractometer. Anal. Chem. 1957, 29, 760–762. [CrossRef] 76. Kim, C.; Park, H.J.; Cha, S.; Yoon, J. Facile detection of photogenerated reactive oxygen species in TiO2 nanoparticles suspension using colorimetric probe-assisted spectrometric method. Chemosphere 2013, 93, 2011–2015. [CrossRef] 77. Simonsen, M.E.; Muff, J.; Bennedsen, L.R.; Kowalski, K.P.; Søgaard, E.G. Photocatalytic bleaching of p-nitrosodimethylaniline and a comparison to the performance of other AOP technologies. J. Photochem. Photobiol. A Chem. 2010, 216, 244–249. [CrossRef] 78. Kraljic, I.; Trumbore, C.N. p-Nitrosodimethylaniline as an OH radical scavenger in radiation chemistry. J. Am. Chem. Soc. 1965, 87, 2547–2550. [CrossRef] 387 Catalysts 2018, 8, 625 79. Farhataziz, A.B.R. Selected Speciﬁc Rates of Reactions of Transients from Water in Aqueous Solutions III: Hydroxyl Radical and Perhydroxyl Radical and Their Radical Ions; U.S. Department of Commerce: Washington, DC, USA, 1977. 80. Martínez-Huitle, C.A.; Quiroz, M.A.; Comninellis, C.; Ferro, S.; De Battisti, A. Electrochemical incineration of chloranilic acid using Ti/IrO2, Pb/PbO2 and Si/BDD electrodes. Electrochim. Acta 2004, 50, 949–956. [CrossRef] 81. Bors, W.; Michel, C.; Saran, M. On the nature of biochemically generated hydroxyl radicals. Studies using the bleaching of p-nitrosodimethylaniline as a direct assay method. Eur. J. Biochem. 1979, 95, 621–627. [CrossRef] [PubMed] 82. Barashkov, N.N.; Eisenberg, D.; Eisenberg, S.; Shegebaeva, G.S.; Irgibaeva, I.S.; Barashkova, I.I. Electrochemical chlorine-free AC disinfection of water contaminated with Salmonella typhimurium bacteria. Russ. J. Electrochem. 2010, 46, 306–311. [CrossRef] 83. Hurwitz, A.R.; Liu, S.T. Determination of aqueous solubility and pKa values of estrogens. J. Pharm. Sci. 1977, 66, 624–627. [CrossRef] 84. Ying, G.G.; Kookana, R.S.; Ru, Y.J. Occurrence and fate of hormone steroids in the environment. Environ. Int. 2002, 28, 545–551. [CrossRef] 85. Fernández-Ibáñez, P.; De Las Nieves, F.J.; Malato, S. Titanium Dioxide/Electrolyte Solution Interface: Elect Transfer Phenomena. J. Colloid Interface Sci. 2000, 227, 510–516. [CrossRef] 86. George, S.; Pokhrel, S.; Ji, Z.; Henderson, B.L.; Xia, T.; Li, L.; Zink, J.I.; Nel, A.E.; Mädler, L. Role of Fe doping in tuning the band gap of TiO2 for the photo-oxidation-induced cytotoxicity paradigm. J. Am. Chem. Soc. 2011, 133, 11270–11278. [CrossRef] 87. Bhatu, M.N.; Lavand, A.B.; Malghe, Y.S. Visible light photocatalytic degradation of malachite green using modiﬁed titania. J. Mater. Res. Technol. 2018. [CrossRef] 88. Tabasideh, S.; Maleki, A.; Shahmoradi, B.; Ghahremani, E.; McKay, G. Sonophotocatalytic degradation of diazinon in aqueous solution using iron-doped TiO2 nanoparticles. Sep. Purif. Review Compositing Two-Dimensional Materials with TiO2 for Photocatalysis Yu Ren 1,2, Yuze Dong 1,2, Yaqing Feng 1,2,* and Jialiang Xu 1,3,* 1 School of Chemical Engineering and Technology, Tianjin University, Yaguan Road 135, Tianjin 300350, China; renyu9505@163.com (Y.R.); yuze441295@tju.edu.cn (Y.D.) 2 Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, China 3 School of Materials Science and Engineering, Nankai University, Tongyan Road 38, Tianjin 300350, China * Correspondence: yqfeng@tju.edu.cn (Y.F.); jialiang.xu@nankai.edu.cn (J.X.) Received: 12 November 2018; Accepted: 23 November 2018; Published: 28 November 2018 1 School of Chemical Engineering and Technology, Tianjin University, Yaguan Road 135, Tianjin 300350, China; renyu9505@163.com (Y.R.); yuze441295@tju.edu.cn (Y.D.) 2 Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, China 3 School of Materials Science and Engineering, Nankai University, Tongyan Road 38, Tianjin 300350, China * Correspondence: yqfeng@tju.edu.cn (Y.F.); jialiang.xu@nankai.edu.cn (J.X.) Received: 12 November 2018; Accepted: 23 November 2018; Published: 28 November 2018 1 School of Chemical Engineering and Technology, Tianjin University, Yaguan Road 135, Tianjin 300350, China; renyu9505@163.com (Y.R.); yuze441295@tju.edu.cn (Y.D.) 2 Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, China 3 School of Materials Science and Engineering, Nankai University, Tongyan Road 38, Tianjin 300350, China * Correspondence: yqfeng@tju.edu.cn (Y.F.); jialiang.xu@nankai.edu.cn (J.X.) Received: 12 November 2018; Accepted: 23 November 2018; Published: 28 November 2018 Abstract: Energy shortage and environmental pollution problems boost in recent years. Photocatalytic technology is one of the most effective ways to produce clean energy—hydrogen and degrade pollutants under moderate conditions and thus attracts considerable attentions. TiO2 is considered one of the best photocatalysts because of its well-behaved photo-corrosion resistance and catalytic activity. However, the traditional TiO2 photocatalyst suffers from limitations of ineffective use of sunlight and rapid carrier recombination rate, which severely suppress its applications in photocatalysis. Surface modiﬁcation and hybridization of TiO2 has been developed as an effective method to improve its photocatalysis activity. Due to superior physical and chemical properties such as high surface area, suitable bandgap, structural stability and high charge mobility, two-dimensional (2D) material is an ideal modiﬁer composited with TiO2 to achieve enhanced photocatalysis process. In this review, we summarized the preparation methods of 2D material/TiO2 hybrid and drilled down into the role of 2D materials in photocatalysis activities. Keywords: photocatalysis; 2D materials; TiO2; composite References Technol. 2017, 189, 186–192. [CrossRef] 89. Aba-Guevara, C.G.; Medina-Ramírez, I.E.; Hernández-Ramírez, A.; Jáuregui-Rincón, J.; Lozano-Álvarez, J.A.; Rodríguez-López, J.L. Comparison of two synthesis methods on the preparation of Fe, N-Co-doped TiO2 materials for degradation of pharmaceutical compounds under visible light. Ceram. Int. 2017, 43, 5068–5079. [CrossRef] 90. Hemmati Borji, S.; Nasseri, S.; Mahvi, A.; Nabizadeh, R.; Javadi, A. Investigation of photocatalytic degradation of phenol by Fe(III)-doped TiO2 and TiO2 nanoparticles. J. Environ. Health Sci. Eng. 2014, 12, 101. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 388 catalysts catalysts Table 1. Mechanism for heterogeneous photocatalysis on TiO2. Table 1. Mechanism for heterogeneous photocatalysis on TiO2. Primary Process Characteristic Times charge-carrier generation TiO2 + hv →hVB+ + ecb− (fs) charge-carrier trapping hVB+ + >TiIVOH →{>TiIVOH}•+ fast (10 ns) ecb−+ >TiIVOH →{>TiIIIOH} shallow trap (100 ps) (dynamic equilibrium) ecb−+ >TiIV →>TiIII deep trap (10 ns) (irreversible) charge-carrier recombination ecb−+ {>TiIVOH}•+ →>TiIVOH slow (100 ns) hVB+ + {>TiIIIOH} →TiIVOH fast (10 ns) interfacial charge transfer {>TiIVOH}•+ + Red →>TiIVOH + Red•+ slow (100 ns) etr−+ Ox →TiIVOH + Ox•− very slow (ms) Where >TiOH represents the primary hydrated surface functionality of TiO2, ecb−is a conduction band (CB) electron, etr−is a trapped conduction band electron, hVB+ is a valence band (VB) hole, Red is an electron donor, Ox is an electron acceptor, {>TiIVOH}•+ is the surface-trapped VB hole (i.e., surface-bound hydroxyl radical), and {>TiIIIOH} is the surface-trapped CB electron. Upon light irradiation, electrons transfer from VB to CB of TiO2, while both electrons and holes can be trapped by primary hydrated surface functionality of TiO2, achieving the separation of photo induced electrons and holes. At the same time, the recombination between electrons and holes exits, which competes with charge-carrier trapping process. The competition has thus a negative effect on later interfacial charge transfer. Deliberating on TiO2 photocatalysis process, some drawbacks exit as following: (1) The wide bandgap of TiO2 (3.2 eV) means that photons with adequate energy can only excite electrons in the VB to the CB of TiO2, which limits its effective use of sunlight (UV region, λ ≤387 nm); (2) The recombination of excited electrons and holes is inevitable while time for carrier recombination is much shorter than that for charge transfer. Therefore, the effective function of photoexcitation is suppressed greatly. g p pp g y Considering the above two factors, the improvement of the photocatalytic efﬁciency of TiO2 can be obtained through two aspects: the improvement of solar light utilization efﬁciency and the suppression of recombination of electron and hole pairs. In this text, surface modiﬁcation and hybridization of TiO2 such as noble metal loading [26–29] and semiconductor heterojunction [30–32] are effective methods to enhance the photocatalytic performance. The Schottky barrier formed at the interface between the noble metal material and TiO2 can effectively promote the separation of photogenerated carriers. Similarly, the heterojunction structure can form a matching energy level at the semiconductor interface to suppress the recombination of photogenerated carriers. Table 1. Mechanism for heterogeneous photocatalysis on TiO2. However, the opportunities of improvements in photocatalysis performances offered by these attempts are narrow, and thus limited their commercial and efﬁcient application. In the past decade, two-dimensional (2D) materials have attracted more and more attention because of the ﬂexible preparation methods, low price and superior physical and chemical properties. In particular, their high surface area, suitable bandgap, structural stability and high charge mobility [33–36] endow these 2D materials with remarkable performances for applications in photocatalysis [37–41]. When combined with TiO2, not only the utilization of sunlight is improved, but also the matching between energy levels is formed to inhibit the recombination, and the large speciﬁc surface area provides support and active sites for the reaction. In this review, we summarize the recent advances of 2D material-TiO2 composites, including synthesis methods, properties, and catalytic behaviors. Furthermore, the photocatalytic mechanism is deliberated in detail to elaborate the role of 2D materials in the photocatalytic processes. 1. Introduction With the massive consumption of fossil energy and serious environmental pollution problems, there is an urgent need for clean energy and more efﬁcient ways to decompose pollutants. Photocatalysis is an advanced technology that uses photon energy to convert chemical reactions occurring under harsh conditions into reactions under mild conditions by appropriate photocatalyst, and thus emerged as recognizable ﬁelds such as hydrogen generation [1–4], sewage treatment [5–7], harmful gas removal [8,9], organic pollutant degradation [10–13] and carbon dioxide reduction [14–16]. Since the ﬁrst report that TiO2 electrode was applied for hydrogen production by Fujishima and Honda in 1972 [17], TiO2 has attracted numerous attention in photocatalysis as a typical n-type semiconductor [18–21]. Being non-toxic, inexpensive, highly stable [22–24], TiO2 is widely investigated in photocatalytic ﬁelds. Hoffman proposed the following general mechanism (Table 1) for heterogeneous photocatalysis on TiO2 [25]. With the massive consumption of fossil energy and serious environmental pollution problems, there is an urgent need for clean energy and more efﬁcient ways to decompose pollutants. Photocatalysis is an advanced technology that uses photon energy to convert chemical reactions occurring under harsh conditions into reactions under mild conditions by appropriate photocatalyst, and thus emerged as recognizable ﬁelds such as hydrogen generation [1–4], sewage treatment [5–7], harmful gas removal [8,9], organic pollutant degradation [10–13] and carbon dioxide reduction [14–16]. Since the ﬁrst report that TiO2 electrode was applied for hydrogen production by Fujishima and Honda in 1972 [17], TiO2 has attracted numerous attention in photocatalysis as a typical n-type semiconductor [18–21]. Being non-toxic, inexpensive, highly stable [22–24], TiO2 is widely investigated in photocatalytic ﬁelds. Hoffman proposed the following general mechanism (Table 1) for heterogeneous photocatalysis on TiO2 [25]. Catalysts 2018, 8, 590; doi:10.3390/catal8120590 389 www.mdpi.com/journal/catalysts www.mdpi.com/journal/catalysts Catalysts 2018, 8, 590 2. 2D-Material Modiﬁed TiO2 Based on the mobile dimension of electronics, it can be divided into zero-dimensional (0D) materials, one-dimensional (1D) materials, two-dimensional (2D) material and three-dimensional (3D) materials [36], while 2D materials represent an emerging class of materials that possess sheet-like structures with the thickness of only single or a few atom layers [42]. Compared with the bulk structures, the ultrathin 2D structure exhibits superior properties such as modiﬁcation of energy level 390 Catalysts 2018, 8, 590 and larger adjustable surface area. The excellent properties of 2D materials make them widely used in many aspects [43–45]. When composited with TiO2, the synergistic effect of the two can signiﬁcantly improve the photocatalytic activity and thus 2D materials is ideal for TiO2 photocatalysis. 2.1. Graphene Modiﬁed TiO2 Since the ﬁrst isolation by Geim and Novoselov in 2004, graphene has attracted signiﬁcant attention [46–49]. Graphene is a 2D honeycomb construction consisting of carbon atoms. The thickness of graphene is only 0.335 nm, which is the thickness of a carbon atom layer. In the sp2 hybrid distribution form, each carbon atom contributes an unbonded π electron, which can delocalize freely throughout the carbon atom ‘net’ to form an extended π bond. This construction endows graphene excellent properties such as high charge mobility (200,000 cm2 V−1 s−1), high thermal conductivity (5000 W m−1 K−1), and large surface area [35], which is ideal for applications in sensors [50], energy conversion and storage [37], polymer composites [51], drug delivery systems [52], and environmental science [53]. When composited with TiO2, graphene can accept photoinduced electrons from TiO2 and thus greatly enhances the efﬁciency of carriers’ separation [54–58]. 2.1.1. The Synthesis of Graphene/TiO2 Composites 2.1.1. The Synthesis of Graphene/TiO2 Composites Graphite oxide and graphene oxide (GO) intermediates are widely used in the process of combining graphene with other materials [59]. The most widely used technique is chemical reduction of GO as shown in Figure 1, which is usually conducted by Hummers’ method [60]. Graphite is added to a strongly oxidizing solution such as HNO3, KMnO4, and H2SO4 to prepare graphite oxide and the oxygen-containing groups are introduced into the surface or edge of the graphite during the process. The sheets of graphite oxide were exfoliated to obtain GO. The presence of oxygen-containing groups allows GO to provide more surface modiﬁcation active sites and larger speciﬁc surface areas for synthetic graphene-based composites. GO can be converted to reduced graphene oxide (RGO) by chemical reduction to remove these oxygen-containing group. During this process, the number of oxygen-containing groups on the GO decreases drastically, and the conjugated structure of the graphene base will be effectively restored. The presence of oxygen functionalities in GO allows interactions with the cations and provides reactive sites for the nucleation and growth of nanoparticles, which results in the rapid growth of various graphene-based composites. The preparation methods for graphene/TiO2 composites are divided into ex-situ hybridization and in-situ growth, the difference between which is the process of TiO2 formation. • Ex-situ hybridization. The common procedure for ex-situ hybridization is to mix GO and modified TiO2 with physical process such as ultrasound sonication and heat treatments. Rahmatollah et al. [62] reported a facile one-step solvothermal method to synthesize the TiO2-graphene composite sheets by dissolving different mass ratios of GO and TiO2 nanoparticles in anhydrous ethanol solution. Ultrasound irradiation was used to disperse the GO. Finally, a six-fold enhancement was observed in the photocurrent response compared to the improved photoelectrochemical performance (3%) with the pure TiO2. Florina et al. [63] prepared graphene/TiO2-Ag based composites as electrode materials. Similarly, GO suspensions were mixed with prepared TiO2-Ag nanoparticles in NaOH solution. The suspensions were sonicated, dried and subjected to thermal treatment. However, the control of modiﬁcation between the TiO2 and graphene may lead to a decreased interaction between these two parts [64]. • Ex-situ hybridization. The common procedure for ex-situ hybridization is to mix GO and modified TiO2 with physical process such as ultrasound sonication and heat treatments. Rahmatollah et al. 2.1.1. The Synthesis of Graphene/TiO2 Composites [62] reported a facile one-step solvothermal method to synthesize the TiO2-graphene composite sheets by dissolving different mass ratios of GO and TiO2 nanoparticles in anhydrous ethanol solution. Ultrasound irradiation was used to disperse the GO. Finally, a six-fold enhancement was observed in the photocurrent response compared to the improved photoelectrochemical performance (3%) with the pure TiO2. Florina et al. [63] prepared graphene/TiO2-Ag based composites as electrode materials. Similarly, GO suspensions were mixed with prepared TiO2-Ag nanoparticles in NaOH solution. The suspensions were sonicated, dried and subjected to thermal treatment. However, the control of modiﬁcation between the TiO2 and graphene may lead to a decreased interaction between these two parts [64]. In-situ growth. The in-situ growth method is widely used to prepare graphene-based composite materials, and the method can effectively avoid clustering of nanoparticles on the surface of graphene. According to different preparation process, it might be divided into reduction method, electrochemical deposition method, hydrothermal method and sol-gel method. • Reduction method. Usually, in a reduction method, GO and TiO2 metal salts are mixed as precursors. By controlling the hydrolysis of the precursor, TiO2 crystal nucleus grows on GO, • Reduction method. Usually, in a reduction method, GO and TiO2 metal salts are mixed as precursors. By controlling the hydrolysis of the precursor, TiO2 crystal nucleus grows on GO, 391 Catalysts 2018, 8, 590 while GO is reduced to obtain graphene-based TiO2 composite materials [65]. In addition to the chemical reduction method, other commonly used reduction methods are photocatalytic reduction [66] and microwave-assisted chemical reduction [67]. • Electrochemical deposition method. In an electrochemical deposition method, graphene or reduced graphene is used as a working electrode in a dielectric solution containing a metal precursor or its compound [68]. • Electrochemical deposition method. In an electrochemical deposition method, graphene or reduced graphene is used as a working electrode in a dielectric solution containing a metal precursor or its compound [68]. • Hydrothermal/solvothermal method. A hydrothermal/solvothermal method is commonly used for preparing inorganic nanomaterials. It is generally carried out in a dispersion of GO. Under high temperature and high pressure, GO and titanium salt precursor are reduced simultaneously [69,70]. • Hydrothermal/solvothermal method. A hydrothermal/solvothermal method is commonly used for preparing inorganic nanomaterials. It is generally carried out in a dispersion of GO. Under high temperature and high pressure, GO and titanium salt precursor are reduced simultaneously [69,70]. Sol-gel method. 2.1.1. The Synthesis of Graphene/TiO2 Composites The sol-gel method takes titanium alkoxide or titanium chloride as precursors, and it can be uniformly bonded with oxygen group on graphene, polycondensed to form a gel. Then TiO2 nanoparticles are formed through calcining [71,72]. The sol-gel method can obtain loaded nanoparticles with higher uniformity of dispersion. • Sol-gel method. The sol-gel method takes titanium alkoxide or titanium chloride as precursors, and it can be uniformly bonded with oxygen group on graphene, polycondensed to form a gel. Then TiO2 nanoparticles are formed through calcining [71,72]. The sol-gel method can obtain loaded nanoparticles with higher uniformity of dispersion. Figure 1. Preparation of graphene by chemical reduction of graphene oxide synthesized by Hummers’ method. Reprinted with permission from [61]. Copyright 2011, Wiley-VCH. Figure 1. Preparation of graphene by chemical reduction of graphene oxide synthesized by Hummers’ method. Reprinted with permission from [61]. Copyright 2011, Wiley-VCH. 2.1.2. The Role of Graphene in TiO2 Photocatalysis 2.1.2. The Role of Graphene in TiO2 Photocatalysis 2.1.2. The Role of Graphene in TiO2 Photocatalysis Due to the large bandgap, the photocatalysis process of pure TiO2 can only be activated under UV light. Thus, the hybridization of graphene and TiO2 is essential to ensure a broad light stimulation process. In graphene/TiO2 system, electrons ﬂow from TiO2 to graphene through interface because of the higher Fermi level of TiO2. Then graphene gains excess negative charges while TiO2 has positive charges, leading to a space charge layer at the interface which is regarded as Schottky junction. The Schottky junction can serve as an electron trap to efﬁciently capture the photoinduced electrons [73] and thus enhance the photocatalysis activity. Meanwhile, the Schottky barrier also acts as the main obstruction for the electron transport from the graphene to TiO2. Under visible light, electrons on Fermi level of graphene are irradiated and the Schottky barrier has to overcome to ensure the injection of electrons to conduct band of TiO2. In the UV light irradiation process, graphene plays a role as electron acceptor and thus promotes the separation of electron-hole pairs [54] (Figure 2). Different interface interactions have been extensively studied [55,56]. Compared with 0D-2D Degussa P25 (TiO2)/graphene and 1D-2D TiO2 nanotube/graphene composites, the 2D-2D TiO2 nanosheet/graphene hybrid demonstrates higher photocatalytic activity toward the degradation of rhodamine B and 2,4-dichlorophenol under the UV irradiation [56]. The intimate and uniform contact between the two sheets-like nanomaterials allowed for the rapid injection of photogenerated electrons from the excited TiO2 into graphene across the 2D-2D interface while achieving effective electron-hole 392 Catalysts 2018, 8, 590 pair separation and promoted radical’s generation. In another example of RGO–TiO2 hybrid, by having a narrower bandgap, the photo-response range of RGO–TiO2 nanocomposites clearly extended from UV (~390 nm) to visible light (~480 nm), which offered a better utilization of visible light [55]. Raman spectra and other characterization revealed that the narrow bandgap was attributed to the Ti–O–C bond between the two components, and thus caught an intimate interaction between TiO2 nanoparticles and RGO sheets. What’s more, the up-conversion photoluminescence (UCPL) effect of RGO assists the light absorption, and enabled the efﬁcient utilization of both UV light and visible light (Figure 3). It is worth to note that the surface area of RGO–TiO2 was smaller than that of pure TiO2 (P25), which revealed that the enhanced photocatalytic activity of RGO-TiO2 was relevant to the improved conductivity and bandgap structure other than their surface area. 2.1.2. The Role of Graphene in TiO2 Photocatalysis RGO nanosheet can play a role in both charge transfer and active sites after doping with heteroatoms. TiO2/nitrogen (N) doped reduced graphene oxide (TiO2/NRGO) nanocomposites was applied to photoreduction of CO2 with H2O vapor in the gas-phase under the irradiation of a Xe lamp (the wavelength range of 250–400 nm) [57]. Compared with TiO2, TiO2/NRGO composites exhibited a narrower bandgap due to chemical bonding between TiO2 and the speciﬁc sites of N-doped graphene. In the photoreduction of carbon dioxide, the function of nitrogen atoms varied in different chemical environments. The pyridinic-N and pyrrolic-N worked as active sites for CO2 capture and activation while quaternary-N worked as an electron-mobility activation region for the effective transfer of photogenerated electrons from the CB of the TiO2 [57] (Figure 4). The results reveal that the doped atoms can act as basic sites for anchoring target molecular, adjusting the electronic properties and local surface reactivity of graphene. Figure 2. Photocatalytic mechanisms of graphene-TiO2 composite under (a) visible light (b) UV light. Reprinted with permission from [54]. Copyright 2013, Elsevier. Figure 2. Photocatalytic mechanisms of graphene-TiO2 composite under (a) visible light (b) UV light. Reprinted with permission from [54]. Copyright 2013, Elsevier. Figure 2. Photocatalytic mechanisms of graphene-TiO2 composite under (a) visible light (b) UV light. Reprinted with permission from [54]. Copyright 2013, Elsevier. p p [ ] py g , Figure 3. (a) Schematic of up-conversion photoluminescence (UCPL) mechanism for reduced graphene oxide (RGO)–TiO2 nanocomposite under visible light (hν∼2.6 eV) irradiation; (b,c) Schematics of proposed mechanism of Rh. B photodegradation. Reprinted with permission from [55]. Copyright 2017, Springer. Figure 3. (a) Schematic of up-conversion photoluminescence (UCPL) mechanism for reduced graphene oxide (RGO)–TiO2 nanocomposite under visible light (hν∼2.6 eV) irradiation; (b,c) Schematics of proposed mechanism of Rh. B photodegradation. Reprinted with permission from [55]. Copyright 2017, Springer. Figure 3. (a) Schematic of up-conversion photoluminescence (UCPL) mechanism for reduced graphene oxide (RGO)–TiO2 nanocomposite under visible light (hν∼2.6 eV) irradiation; (b,c) Schematics of proposed mechanism of Rh. B photodegradation. Reprinted with permission from [55]. Copyright 2017, Springer. 393 Catalysts 2018, 8, 590 Figure 4. Reaction mechanisms for photoreduction of CO2 with H2O over TiO2/NRGO-300 samples. Reprinted with permission from [57]. Copyright 2017, Elsevier. NRGO: nitrogen doped reduced graphene oxide. Figure 4. Reaction mechanisms for photoreduction of CO2 with H2O over TiO2/NRGO-300 samples. Reprinted with permission from [57]. Copyright 2017, Elsevier. NRGO: nitrogen doped reduced graphene oxide. Figure 4. 2.2. Graphdiyne Modiﬁed TiO2 Graphdiyne (GD) is a new carbon allotrope in which the benzene rings are conjugated by 1,3-diyne bonds to form a 2D planar network structure and features both sp and sp2 carbon atoms. Since the successful synthesis by Li et al. [75], GD has evoked signiﬁcant interest in various scientiﬁc ﬁelds because of unique mechanical, chemical and electrical properties [38,42,76–80]. GD shows potential for photocatalysis with its large surface area as well as high charge mobility. GD features an intrinsic bandgap and exhibits semiconducting property with a measured conductivity of 2.516 × 10−4 S·m−1 and was predicted to be the most stable structure among various diacetylenic non-natural carbon allotropes [81]. It also provides highly active sites for catalysis. Furthermore, GD with diacetylene linkage can be chemically bonded with TiO2 [82–85]. Therefore, the TiO2-graphdiyne composites can greatly improve the photocatalytic activity, and thus their application in photocatalysis has been explored recently [83,84,86]. 2.1.2. The Role of Graphene in TiO2 Photocatalysis Reaction mechanisms for photoreduction of CO2 with H2O over TiO2/NRGO-300 samples. Reprinted with permission from [57]. Copyright 2017, Elsevier. NRGO: nitrogen doped reduced graphene oxide. Except for dimension factor and bonding interaction between graphene and TiO2, a linkage is introduced to graphene/TiO2 system to achieve better interfacial contact as well. A N-doping Graphene-TiO2 composite nano-capsule for gaseous HCHO degradation was reported [58]. It indicated that wrapping with dopamine on the surface of TiO2 enhanced interfacial contact between TiO2 and melamine-doped graphene (MG) sheets, thus promoting the separation and mobility of photoinduced electrons and holes in TiO2@MG-D. The dopamine acted as bridge between TiO2 and MG, creating numerous migration channels for charges and restraining the recombination of electrons and holes (Figure 5). The introduction of linkage can effectively improve the weak interfacial contact and overcome the long distance of electron transport between the graphene and TiO2, leading to raised separation and mobility of photoinduced electrons and holes and thus higher photocatalytic activity. Figure 5. Schematic illustrations for dopamine bridged Melamine-Graphene/TiO2 nanocapsule and photocatalytic degradation process of HCHO. Reprinted with permission from [58]. Copyright 2018, Elsevier. Figure 5. Schematic illustrations for dopamine bridged Melamine-Graphene/TiO2 nanocapsule and photocatalytic degradation process of HCHO. Reprinted with permission from [58]. Copyright 2018, Elsevier. Despite of electron accepter and electron storage, graphene can also act as a transport bridge between photocatalysts. For example, in the 2D ternary BiVO4/graphene oxide (GO)/TiO2 system, both the BiVO4 and the TiO2 were connected to GO forming a p-n heterogeneous structure. The CB of BiVO4 was more negative than that of GO and the CB of GO was more negative than that of TiO2; thus, the electrons generated from the CB of BiVO4 can transfer to the GO and then the electron further moved to the conduction band of TiO2 (Figure 6). Therefore, the GO can enhance the effective separation of the photo-generated electron-hole pairs due to its superior electrical conductivity. Meanwhile, the large surface area of the GO is also beneﬁcial for dye attachment [74]. 394 Catalysts 2018, 8, 590 Figure 6. Photodegradation mechanism of BiVO4/TiO2/GO photocatalyst. Reprinted with permission from [74]. Copyright 2017, Elsevier. Figure 6. Photodegradation mechanism of BiVO4/TiO2/GO photocatalyst. Reprinted with permission from [74]. Copyright 2017, Elsevier. 2.2.2. The Role of GD in TiO2 Photocatalysis Wang et al. [84] were the ﬁrst to combine GD with TiO2 for the enhancement of TiO2 photocatalysis. The resultant GD-P25 composites exhibited higher visible light photocatalytic activity than those of the bare P25, P25-CNT (titania-carbon nanotube), and P25-GR (graphene) materials. By changing the weight percent of GD in the hybrid, the photocatalytic activity of P25-GD can be adjusted. It was speculated that the formation of chemical bonds between P25 and GD can effectively decrease the bandgap of P25 and extended its absorbable light range [84]. Namely, electrons in VB of TiO2 can easily migrate to impurity band which is attributed to the insertion of carbon p-orbitals into the TiO2 bandgap, and then transfer to CB of TiO2 thus enhancing the photo-response activity. In order to further explore the role of GD, Yang et al. [83] investigated the chemical structures and electronic properties of TiO2-GD and TiO2-GR composites employing ﬁrst-principles density functional theory (DFT) calculations. The results revealed that for the TiO2 (001)-GR composite, O and atop C atoms could form C–O σ bond, which acted as a charge transfer bridge at the interface between TiO2 and GR. Besides the C–O σ bond, another Ti-C π bond is also formed in TiO2 (001)-GD composite, which makes GD combine with TiO2 tightly and therefore enhances the charge transfer. In addition, calculated Mulliken charge for the surface of TiO2 (001)-GD and TiO2 (001)-GR suggested a stronger electrons’ capture ability of former (Figure 8). The calculated results were in accordance with theoretical prediction that TiO2 (001)-GD composites showed the highest photocatalysis performance among 2D carbon-based TiO2 composites, conﬁrming that GD could become a promising competitor in the ﬁeld of photocatalysis. After that, Dong et al. prepared GD-hybridized nitrogen-doped TiO2 nanosheets with exposed (001) facets (GD-NTNS) [86]. The doped N and incorporated GD efﬁciently narrowed the bandgap compared with pure TiO2 and widened response range towards light from UV light to 420 nm visible light. The activity of the GD-NTNS photocatalyst presented the most superior performance compared with bare TiO2 nanosheets (TNS) and nitrogen-doped TiO2 nanosheets (NTNS) and GR-NTNS. Figure 8. Plots of electron density difference at the composites interfaces: (a) TiO2 (001)-GD; (b) TiO2 (001)-GR; (c) Mulliken charge of GD or GR (graphene) surface in the composites. Reprinted with permission from [83]. Copyright 2013, American Chemical Society. Figure 8. 2.2.1. The Synthesis of GD/TiO2 Composites The general preparation of GD ﬁlm is through a coupling reaction in which hexaethynylbenzene (HEB) acts as precursor and copper foil serves as catalysis. Meanwhile, the copper foil provides a large planar substrate for the directional polymerization growth of the GD ﬁlm (Figure 7). Despite of ﬁlm, GD with different morphologies such as nanotube arrays, nanowires, nanowalls and nanosheets have been also prepared for diverse applications [87,88]. Figure 7. Preparation of graphdiyne (GD) ﬁlm. Figure 7. Preparation of graphdiyne (GD) ﬁlm. 395 Catalysts 2018, 8, 590 Ex-situ hydrothermal method is commonly used in preparation of GD/TiO2 composites [83,84,86]. In general, the GD and TiO2 are prepared separately. Then the pre-prepared GD and TiO2 are mixed in H2O/CH3OH solvent. After stirring to obtain a homogeneous suspension, the suspension is placed in Teﬂon sealed autoclave and heated to combine the TiO2 and GD. Being rinsed and dried, the GD/TiO2 composites are obtained. 2.2.2. The Role of GD in TiO2 Photocatalysis 2.2.2. The Role of GD in TiO2 Photocatalysis Plots of electron density difference at the composites interfaces: (a) TiO2 (001)-GD; (b) TiO2 (001)-GR; (c) Mulliken charge of GD or GR (graphene) surface in the composites. Reprinted with permission from [83]. Copyright 2013, American Chemical Society. The mechanisms of photocatalysis enhancement by introducing GD remain to be understood. In general, with a lower Fermi level than the conduction band minimum of TiO2, GD can be regarded as an electron pool which accept electrons excited from TiO2 [84,89,90] (Figure 9). As a result, it prompts the charge carriers’ separation and prevents electron-hole recombination. Moreover, GD can generate an impurity band and thus broaden the visible light absorption in TiO2-GD composites [91–93]. 396 Catalysts 2018, 8, 590 Figure 9. Schematic illustration for the possible mechanism of the visible light-driven photocatalytic degradation for the GD-NTNS composites. Reprinted with permission from [86]. Copyright 2018, Springer. 2.3. C3N4 Modiﬁed TiO2 Figure 9. Schematic illustration for the possible mechanism of the visible light-driven photocatalytic degradation for the GD-NTNS composites. Reprinted with permission from [86]. Copyright 2018, Springer. 2.3. C3N4 Modiﬁed TiO2 Graphitic carbon nitride (g-C3N4) is a 2D polymer material which shows broad application prospects in many ﬁelds, given the simple synthesis, rich source, along with unique electronic structure, good thermal stability and chemical stability. Its graphene-like structure is composed of triazine (C3N3) or tri-s-striazine (C6N7) allotropes units (Figure 10). The tri-s-striazine unit structure is more stable and thus draws in extensive studies [34]. Since the ﬁrst report of g-C3N4 for water decomposition, g-C3N4 has attracted wide attention in photocatalyst [40]. The bandgap of g-C3N4 (2.6–2.7 eV) is moderate and the substantial nitrogen sites and ordered units structure endue g-C3N4 an ideal material to composite with TiO2. Figure 10. Triazine (a) and tri-s-striazine (b) allotropes units of g-C3N4; (c) The synthesis of g-C3N4. Figure 10. Triazine (a) and tri-s-striazine (b) allotropes units of g-C3N4; (c) The synthesis of g-C3N4. Figure 10. Triazine (a) and tri-s-striazine (b) allotropes units of g-C3N4; (c) The synthesis of g-C3N4 2.3.1. The Synthesis of g-C3N4/TiO2 Composites 2.3.1. The Synthesis of g-C3N4/TiO2 Composites In general, the synthesis of g-C3N4/TiO2 composites can be also divided into ex-situ method and in-situ method. In general, the synthesis of g-C3N4/TiO2 composites can be also divided into ex-situ method and in-situ method. 2.2.2. The Role of GD in TiO2 Photocatalysis In this process, the solid mixture of TiO2 and pure urea or melamine or dicyandiamide powder are calcinated under ﬁxed temperature to obtain g-C3N4/TiO2 composites. Before calcination, the two components should be evenly dispersed by sonication [101], stirring [102], or grounding [103]. Recently, Tan et al. [104] reported another facile one-step way to prepare nanostructured g-C3N4/TiO2 composite. As seen in Figure 11, melamine was at the bottom of the crucible while P25 was on the top of a cylinder put in the crucible. After a 4-h vapor deposition process, nanostructured g-C3N4/TiO2 composite was obtained. • When TiO2 was used as substrates, calcination is widely used for the convenience and easy operation. In this process, the solid mixture of TiO2 and pure urea or melamine or dicyandiamide powder are calcinated under ﬁxed temperature to obtain g-C3N4/TiO2 composites. Before calcination, the two components should be evenly dispersed by sonication [101], stirring [102], or grounding [103]. Recently, Tan et al. [104] reported another facile one-step way to prepare nanostructured g-C3N4/TiO2 composite. As seen in Figure 11, melamine was at the bottom of the crucible while P25 was on the top of a cylinder put in the crucible. After a 4-h vapor deposition process, nanostructured g-C3N4/TiO2 composite was obtained. Figure 11. Vapor deposition process in the preparation of g-C3N4/TiO2 composite. Reprinted with permission from [104]. Copyright 2018, Elsevier. Figure 11. Vapor deposition process in the preparation of g-C3N4/TiO2 composite. Reprinted with permission from [104]. Copyright 2018, Elsevier. 2.2.2. The Role of GD in TiO2 Photocatalysis • In the ex-situ way, both g-C3N4 and TiO2 materials are pre-prepared, which can be integrated through physical process such as ball milling [94], solvent evaporation [95,96], etc. Though physical process is easy to operate under moderate conditions, some ﬂaws also exist such as ununiformly dispersing and unstable structure. • The in-situ method uses one of the materials as a substrate and then the other material grows on the surface of the substrate. For g-C3N4/TiO2 composites, both materials can be regarded as substrates. • The in-situ method uses one of the materials as a substrate and then the other material grows on the surface of the substrate. For g-C3N4/TiO2 composites, both materials can be regarded as substrates. 397 Catalysts 2018, 8, 590 • When used as substrates, g-C3N4 is pre-prepared by calcinations of precursors. Solvothermal/ hydrothermal method is most common for the next step. After mixing g-C3N4 and titanates in a certain solvent, the solution is well dispersed and sealed in the Teﬂon-lined autoclave, followed by a solvothermal/hydrothermal treatment [97–99]. Furthermore, Atomic Layer Deposition (ALD) was applied to form thin TiO2 ﬁlms on g-C3N4 substrates. ALD involves the surface of a substrate exposed alternately to alternating precursor ﬂow. Then the precursor molecule reacts with the surface in a self-limiting way, which guarantees that the reaction stops as all the reactive sites on the substrate reacted with the precursors. It is an effective way to control the thickness and homogeneity of deposited layer [100]. • When used as substrates, g-C3N4 is pre-prepared by calcinations of precursors. Solvothermal/ hydrothermal method is most common for the next step. After mixing g-C3N4 and titanates in a certain solvent, the solution is well dispersed and sealed in the Teﬂon-lined autoclave, followed by a solvothermal/hydrothermal treatment [97–99]. Furthermore, Atomic Layer Deposition (ALD) was applied to form thin TiO2 ﬁlms on g-C3N4 substrates. ALD involves the surface of a substrate exposed alternately to alternating precursor ﬂow. Then the precursor molecule reacts with the surface in a self-limiting way, which guarantees that the reaction stops as all the reactive sites on the substrate reacted with the precursors. It is an effective way to control the thickness and homogeneity of deposited layer [100]. • When TiO2 was used as substrates, calcination is widely used for the convenience and easy operation. 2.3.2. The Role of g-C3N4 in Photocatalysis With a moderate bandgap of ~2.7 eV, g-C3N4 shows ability of photocatalyst under visible light, in contrast to TiO2, which owns a large bandgap of 3.2 eV (Figure 12). However, because of the rapid recombination of photogenerated electron-hole pairs, the synergistic effect between g-C3N4 and TiO2 plays important roles. In a photocatalyst system of g-C3N4/TiO2 composites, the CB electrons of g-C3N4 transfer to the CB of TiO2 and the VB holes of TiO2 transfer to the VB of g-C3N4, which is a typical Type II system [41]. The electron/hole conduction mechanism can effectively separate electrons and holes, and thus enhances the separation efﬁciency and inhibit the recombination. Figure 12. Bandgaps of TiO2, monolayer g-C3N4 and bulk g-C3N4. Figure 12. Bandgaps of TiO2, monolayer g-C3N4 and bulk g-C3N4. 398 Catalysts 2018, 8, 590 The structure plays a vital role in enhancing photocatalysis efﬁciency. g-C3N4 nanosheets (NS)-TiO2 mesocrystals (TMC) composites was prepared by in-situ process [105]. Compared with bulk g-C3N4/TMC composites, the H2 evolution rate of g-C3N4 (NS)/TMC was about six times higher, which was possibly due to a larger surface area of g-C3N4 (NS)/TMC (57.4 m2g−1) than that of bulk g-C3N4/TMC (34.3 m2g−1). What’s more, the g-C3N4 nanosheets owned a lower surface defect density, given the surface defects normally is seen as recombination centers for photoinduced electrons and holes. However, surface area is not the unparalleled factor of promoted efﬁciency of photocatalyst, taking the fact that the surface area of g-C3N4 NS (31 wt%)/TMC (57.4 m2g−1) and g-C3N4 NS (31 wt%)/P25 (52.3 m2g−1) was nearly the same, as the H2 evolution rate of g-C3N4 (NS)/TMC was about 7 times higher. Further research indicated that the tight interface between g-C3N4 NS and TMC facilitated the charge transfer, which is a ﬂexible way to promote solar energy utilization of g-C3N4/TiO2 photocatalyst. Other structures like core-shell was lucubrated to create high photocatalytic activity towards many dyes [106]. After in-situ calcination and growth of cyanamide on the surface of TiO2, a multiple direction contact structure of TiO2@g-C3N4 hollow core@shell heterojunction photocatalyst (HTCN-1) was synthesized. The g-C3N4 nanosheets grew on the surface of TiO2 caused closer contact between TiO2 and g-C3N4 and a larger interfacial area, as conﬁrmed by XPS analysis [106]. Compared with another core-shell type TiO2@g-C3N4 (C-T) with unidirectional contact structures [107], HTCN-1 possessed higher efﬁciency in the charge separation and enhanced charge transfer. 2.3.2. The Role of g-C3N4 in Photocatalysis It demonstrated that multiple direction contact resulted in a large interfacial area, which would provide sufﬁcient channels for efﬁcient and rapid charge transfer (Figure 13) [106]. In another core-shell structure of g-C3N4/TiO2 hybrid, Ag was introduced as interlayers to participate in electrical conduction and bridge the gap between g-C3N4 and TiO2, facilitating the separation of photoexcited charge and reducing the recombination of the photogenerated electron hole (Figure 14) [108]. The surface area of the samples didn’t change much upon the introduction of Ag (228.4 m2g−1 and 210.3 m2g−1 for Ag/TiO2 microspheres and nonsilver containing TiO2, respectively). It was worth noting that low content of g-C3N4 (2%) in g-C3N4/Ag/TiO2 microspheres had a larger surface area but lower photocatalytic activity than the g-C3N4 (4%)/Ag/TiO2 microsphere sample [108]. The possible reason was that high content of g-C3N4 can generate more electron-hole pairs, leading to a higher photocatalytic activity. However, the g-C3N4 (6%)/Ag/TiO2 microsphere sample showed decreased photocatalytic activity due to reduced surface area, which limited the contact between the catalyst and pollutant and thus lowered the photocatalytic reaction. It reﬂects that proper surface area is needed to provide both active sites and reaction sites. The doping of g-C3N4 is another viable way to realize structure modiﬁcation process. Sulfur was introduced to g-C3N4 nanostructures, and their photocatalytic performance was studied for decomposition of MO dye under visible light. The degradation efﬁciency over g-C3N4-TiO2 composites (CNT) reached 61% within 90 min, while S-C3N4-TiO2 composites (SCNT) reached nearly 100% within the same period [109]. SEM image showed a more transparent and thinner layer of S-C3N4 compared with g-C3N4 when composited with TiO2, leading to an enhanced visible light absorption capability. On the other hand, unique bar-like structure of SCNT provided a pathway for carriers and isolate photon absorption with carriers’ collection in perpendicular directions. Meanwhile, TiO2 nanoparticles were more evenly dispersed on and inside S-C3N4 substrate in SCNT sample, which is beneﬁcial for the interfacial carriers’ transportation between S-C3N4 layer and TiO2 particle [109]. Calculations revealed that the modiﬁed electronic structure with elevation of CB and VB values owing to doped sulfur, contributed to a higher driving force from CB of S-C3N4 to CB of TiO2 and thus promoted the separation efﬁciency of electron-hole pairs (Figure 15). The doping of sulfur alternated both the structure and level distribution of C3N4, causing excellent separation efﬁciency of electron-hole pair when contacted with TiO2. 2.3.2. The Role of g-C3N4 in Photocatalysis 399 399 Catalysts 2018, 8, 590 Figure 13. Structure of HTCN-1 (a) and C-T (b). Reprinted with permission from [106]. Copyright 2018, Elsevier. Figure 13. Structure of HTCN-1 (a) and C-T (b). Reprinted with permission from [106]. Copyright 2018, Elsevier. Figure 14. Photocatalytic mechanism scheme of g-C3N4/Ag/TiO2 microspheres under visible light irradiation (>420 nm). Reprinted with permission from [108]. Copyright 2014, American Chemical Society. Figure 14. Photocatalytic mechanism scheme of g-C3N4/Ag/TiO2 microspheres under visible light irradiation (>420 nm). Reprinted with permission from [108]. Copyright 2014, American Chemical Society. Figure 15. Mechanism of fast charge transfer at the interface between (a) C3N4-TiO2 and (b) S-C3N4/TiO2. Reprinted with permission from [109]. Copyright 2017, Elsevier. Figure 15. Mechanism of fast charge transfer at the interface between (a) C3N4-TiO2 and (b) S-C3N4/TiO2. Reprinted with permission from [109]. Copyright 2017, Elsevier. 400 Catalysts 2018, 8, 590 Catalysts 2018, 8, 590 2.4. MoS2 Modiﬁed TiO2 2.4. MoS2 Modiﬁed TiO2 2D layered transition metal chalcogenides (TMCs) nanostructures spark a research boom due to its unique physical and chemical properties compared with other 2D materials. The usual formula of TMCs is MX2, while M is transition metal and X is chalcogenide element, namely, S, Se, or Te. Because of the typical 2D structure with high surface-to-volume ratio and missing coordination at edge (Figure 16), TMCs exhibits high chemical sensitivity [36]. Considering its versatile physicochemical properties, TMCs can be applied in catalyst [41], energy storage [39], and biology [110]. Some TMCs such as WS2 [111], TiS2 [112] are also used in TiO2 photocatalysis. Among TMCs, MoS2 show extraordinary potential as semiconductors owing to its thickness dependent bandgap and natural abundance. When bulk MoS2 are stripped into a single layer or several layers of nanosheets, the indirect bandgap (1.3 eV) can be converted to a direct bandgap (1.8 eV) [113] and show excellent performance in photocatalysis after compositing with TiO2 [114]. Besides, its high surface-to-volume ratio makes up for the limitation of the low theoretical speciﬁc capacity of TiO2. The synergy between MoS2 and TiO2 endows the TiO2/MoS2 composite superior performance compared to their single material. Figure 16. Structure (a) and solution-based preparation (b) of 2D layered transition metal chalcogenides (TMCs) nanosheets based on top-down and bottom-up approaches. Reprinted with permission from [36]. Copyright 2018, American Chemical Society. Figure 16. Structure (a) and solution-based preparation (b) of 2D layered transition metal chalcogenides (TMCs) nanosheets based on top-down and bottom-up approaches. Reprinted with permission from [36]. Copyright 2018, American Chemical Society. 2.4.1. The Synthesis of MoS2/TiO2 Composites Similar to the synthesis methods of graphene/TiO2 composite, the synthesis of MoS2/TiO2 composites is also divided into ex-situ methods and in-situ methods. For the in-situ method, TiO2 and MoS2 are synthesized separately, then the two are combined by various methods, 401 Catalysts 2018, 8, 590 such as hydrothermal/solvothermal assembly [115,116], mechanical method [117], drop-casting [118], or sol–gel [119], which can be also applied for in-situ methods [120,121]. The ex-situ method is simple and inexpensive, but the two compounds have poor dispersion and show weak interactions. Despite the same process as ex-situ method, there are chemical vapor deposition [122] and co-reduction precipitation [123] in in-situ process. Among them, the hydrothermal method is simple, easy to operate, and has good controllability, and thus is most commonly used in the preparation of MoS2/TiO2 composite materials. The in-situ reduction method uses one of the materials as a substrate and then coats or loads the other material. This involves the molybdenum disulﬁde as substrate or TiO2 as a substrate. The following paragraphs will discuss the two kinds of composites. • MoS2 as substrate. In this process, MoS2 are pre-prepared as substrate for the in-situ growth of TiO2. Hydrothermal method is widely used in which tetrabutyl titanate serves as titanate source [124,125]. Recently, another approach has been developed to synthesize MoS2@TiO2 composites. Ren et al. [126] reported TiO2-modiﬁed MoS2 nanosheet arrays by the ALD process, coating a thin layer of TiO2 on both the edge and basal planes of TiO2 (Figure 17). It provides a new insight for the combination of sites at the basal planes of TiO2. • MoS2 as substrate. In this process, MoS2 are pre-prepared as substrate for the in-situ growth of TiO2. Hydrothermal method is widely used in which tetrabutyl titanate serves as titanate source [124,125]. Recently, another approach has been developed to synthesize MoS2@TiO2 composites. Ren et al. [126] reported TiO2-modiﬁed MoS2 nanosheet arrays by the ALD process, coating a thin layer of TiO2 on both the edge and basal planes of TiO2 (Figure 17). It provides a new insight for the combination of sites at the basal planes of TiO2. • TiO2 composite as substrate. For coated MoS2/TiO2 composites, TiO2 are usually substrates. Liu et al. [127] reported a N-TiO2-x@MoS2 core-shell heterostructure composite. TBT and urea were used to prepare N-doped TiO2 microspheres (N-TiO2) with a smooth surface by hydrothermal method. 2.4.1. The Synthesis of MoS2/TiO2 Composites Considering the growth of molybdenum sulﬁde on the TiO2 substrate, speciﬁc morphology and growth sites of TiO2 is needed. Sun et al. [128] took a targeted etching route to control the morphology of TiO2/MoS2 nanocomposites. Hollow microspheres structured TiO2/MoS2 showed a higher dye degradation activity due to a larger proportion of interface, compared to TiO2/MoS2 nanocomposites of yolk-shell structures. Other structures such as nanobelts and nanotubes have also been developed [129,130]. In addition to the morphology, the formation of a speciﬁc crystal structure of TiO2 as a substrate has also got attention to prepare high performance MoS2/TiO2 composites [130,131]. He et al. [130] reported a few-layered 1T-MoS2 coating on Si doped TiO2 nanotubes (MoS2/TiO2 NTs hybrids) through hydrothermal process. Because of the higher catalytic activity of 1T phase of MoS2 and Si doped TiO2, MoS2/TiO2 NTs hybrids nanocomposites exhibited excellent photocatalytic activity. Figure 17. (a) Schematic illustration and (b) TEM image of the ALD TiO2 coating on pristine MoS2. Reprinted with permission from [126]. Copyright 2017, Wiley-VCH. Figure 17. (a) Schematic illustration and (b) TEM image of the ALD TiO2 coating on pristine MoS2. Reprinted with permission from [126]. Copyright 2017, Wiley-VCH. Figure 17. (a) Schematic illustration and (b) TEM image of the ALD TiO2 coating on pristine MoS2. Reprinted with permission from [126]. Copyright 2017, Wiley-VCH. 2.4.2. The Role of MoS2 in TiO2 Photocatalysis 2.4.2. The Role of MoS2 in TiO2 Photocatalysis During the photocatalysis process, electrons transfer through the interface between TiO2 and MoS2, and therefore the contact between the two is vital for photocatalytic activity. A strategy for construction of 3D semiconductor heterojunction structure by TiO2 and 2D-structured MoS2 is proposed to achieve increase of active sites and decrease of electron-hole pair combination [127,132]. For example, a 3D ﬂower-like N-TiO2-x@MoS2 was obtained by hydrothermal method. Considering that the smooth TiO2 nanosphere shows poor afﬁnity when coated with MoS2 nanosheets, TiO2 was doped with N and 402 Catalysts 2018, 8, 590 Ti3+. X-ray photoelectron spectroscopy (XPS) shows the existence of electronic interactions between MoS2 and N-TiO2-x and the strong heterostructure effect between the MoS2 nanoﬂower and N-TiO2-x nanosphere [127]. Another study of 3D TiO2@MoS2 revealed that the formation of Ti-S bonds made TiO2 nanoarrays ﬁrmly grasp MoS2, thus affording a marvelous mechanical stability for the integrated architectures [133]. Different phase of MoS2 exhibits various chemical and physical properties when combined with TiO2. MoS2 has two main phases, namely the metallic 1T phase and semiconducting 2H phase. As for 2H phase, the active site with catalytic activity is located at the edge of the MoS2 layers and the basal surface of MoS2 is catalytically inactive [134]. Therefore, the 1T phase of MoS2 with active sites on both edge and basal planes attracts researchers’ attention in recent years [118,125,131]. A typical schematic of MoS2/TiO2 composites for photocatalytic hydrogen production is shown in Figure 18. The 1T-MoS2 nanosheets not only provide extra reaction sites on the basal plane, but also play a role in electron delivery. Because of the active site distributing on the edge of 2H-MoS2 nanosheets, the photogenerated electron from TiO2 needs a long-distance move before reacted with H2O. This leaded to a lower diffusion rate compared with 1T-MoS2/TiO2 composites and thus enhanced the separation efﬁciency of electron-hole pairs. Therefore, the 1T-MoS2/TiO2 composites exhibited excellent photocatalytic activity as the hydrogen production rate of 1T-MoS2/TiO2 was 5 and 8 times higher than those of bare TiO2 and 1T-MoS2/TiO2 [125]. In another research, 1T-MoS2 coated onto TiO2 (001) composite (MST) was synthesized. DFT calculations suggested a closer distance between the interface electrons and MoS2 surface than that of TiO2 [131] (Figure 19). Therefore the photo-induced electrons can easily transfer to the conducting channel of MoS2. Furthermore, the introduction of 1T-MoS2 prolonged the carrier lifetime remarkedly. 2.4.2. The Role of MoS2 in TiO2 Photocatalysis All the factors led to an enhanced photocatalytic activity. To further inhibit the recombination of electron-hole pairs, cocatalyst such as graphene is applied to MoS2/TiO2 system [115,135,136]. Xiang et al. employed TiO2/MoS2/graphene composite as photocatalyst [135]. In this system, photo-inducted electrons transfer from VB to CB of TiO2. Then the electrons are further injected into the graphene sheets or MoS2 nanoparticles. What is more, graphene sheets can be seen as electrons transport ‘highway’ through which electrons move from VB of TiO2 to MoS2 (Figure 20). The cocatalyst of MoS2 and graphene enhances the interfacial charge transfer rate, inhibits the recombination of electron-hole pairs and offers a host of active site for adsorption and reaction. Han et al. constructed 3D MoS2/P25/graphene-aerogel networks. In addition to the above-mentioned advantages, 3D graphene porous architecture has a highly porous ultraﬁne nanoassembly network structure, excellent electric conductivity, and the maximization of accessible sites [115]. Recently, a 3D double-heterostructured photocatalyst was constructed by connecting a TiO2-MoS2 core-shell nanosheets (NSs) on a graphite ﬁber (GF@MoS2-TiO2) [136]. Mechanism of photocatalytic decomposition of dyes under both visible light and UV light was discussed (Figure 21). Anatase TiO2 has a wide band gap (2.96 eV), while the band gap of MoS2 is 1.8 eV. Because of the moderate bandgap of MoS2, the electrons can be irradiated from VB to CB of MoS2 and then inject into CB of TiO2 or transfer to graphene through intimate double-heterojunction contact under visible light. Graphene acts as electrons accepter under both circumstance, leading to a high rate of charge separation and thus depress the charge recombination. The contact interfaces and synergy among graphene, TiO2 and MoS2 play an important role in the superior photocatalytic activities. While the transfer of electrons are paid special attention, the role of capturing the holes are often ignored. To solve this problem, a TiO2/WO3@MoS2 (TWM) hybrid Z-scheme photocatalytic system was structured. TiO2 and WO3 have the appropriate energy level matching to form the Z-scheme, while the position of VB in WO3 is lower than the VB of TiO2, and the CB of WO3 is between the CB and VB of TiO2 [137]. Under UV light irradiation, the VB electrons of all three parts are excited to corresponding CB level. 2.4.2. The Role of MoS2 in TiO2 Photocatalysis The excited electrons on CB of TiO2 then transfer to CB of MoS2 for H2 evolution, meanwhile the excited electrons on CB of WO3 were inject to the VB of TiO2 (Figure 22). This procedure suppressed the recombination of photoinduced electrons and holes in TiO2, 403 Catalysts 2018, 8, 590 and therefore the photogenerated electrons and holes can be efﬁciently separated, which further leads to effective photocatalytic activity [137]. Figure 18. Schematic illustrating charge-transfer behavior and H2 evolution active sites for (a) 1T-MoS2/ TiO2 and (b) 2H-MoS2/TiO2. Reprinted with permission from [125]. Copyright 2014, Springer. Figure 18. Schematic illustrating charge-transfer behavior and H2 evolution active sites for (a) 1T-MoS2/ TiO2 and (b) 2H-MoS2/TiO2. Reprinted with permission from [125]. Copyright 2014, Springer. Figure 19. (a) The charge density difference, (b) electrostatic potential and differential charge density of the MoS2/TiO2(001) junction; (c) Planar-averaged differential electron density Dr(z) for MoS2/TiO2(001); (d) Photocatalytic mechanism for 1T-MoS2/TiO2. Reprinted with permission from [131]. Copyright 2017, the Owner Societies. Figure 19. (a) The charge density difference, (b) electrostatic potential and differential charge density of the MoS2/TiO2(001) junction; (c) Planar-averaged differential electron density Dr(z) for MoS2/TiO2(001); (d) Photocatalytic mechanism for 1T-MoS2/TiO2. Reprinted with permission from [131]. Copyright 2017, the Owner Societies. 404 Catalysts 2018, 8, 590 alysts 2018, 8, 590 Figure 20. Schematic illustration of the charge transfer in TiO2/MoS2/graphene composites. Reprinted with permission from [135]. Copyright 2012, American Chemical Society. Figure 21. Structure (a) and schematic diagram of electron-hole separation mechanism upon UV (b) and visible light (c) excitation for 3D graphene@MoS2-TiO2 composites. Reprinted with permission from [136]. Copyright 2017, Elsevier. Figure 20. Schematic illustration of the charge transfer in TiO2/MoS2/graphene composites. Reprinted with permission from [135]. Copyright 2012, American Chemical Society. Figure 20. Schematic illustration of the charge transfer in TiO2/MoS2/graphene composites. Reprinted with permission from [135]. Copyright 2012, American Chemical Society. Figure 20. Schematic illustration of the charge transfer in TiO2/MoS2/graphene composites. Reprinte with permission from [135]. Copyright 2012, American Chemical Society. Figure 21. Structure (a) and schematic diagram of electron-hole separation mechanism upon UV (b) and visible light (c) excitation for 3D graphene@MoS2-TiO2 composites. Reprinted with permission from [136]. Copyright 2017, Elsevier. Figure 21. Structure (a) and schematic diagram of electron-hole separation mechanism upon UV (b) and visible light (c) excitation for 3D graphene@MoS2-TiO2 composites. Reprinted with permission from [136]. Copyright 2017, Elsevier. Figure 22. References 1. Lei, J.M.; Peng, Q.X.; Luo, S.P.; Liu, Y.; Zhan, S.Z.; Ni, C.L. A nickel complex, an efﬁcient cocatalyst for both electrochemical and photochemical driven hydrogen production from water. Mol. Catal. 2018, 448, 10–17. [CrossRef] 2. Wang, Z.; Jin, Z.; Yuan, H.; Wang, G.; Ma, B. Orderly-designed Ni2P nanoparticles on g-C3N4 and UiO-6 efﬁcient solar water splitting. J. Colloid Interface Sci. 2018, 532, 287–299. [CrossRef] [PubMed] 3. Vaiano, V.; Lara, M.A.; Iervolino, G.; Matarangolo, M.; Navio, J.A.; Hidalgo, M.C. Photocatalytic H2 production from glycerol aqueous solutions over ﬂuorinated Pt-TiO2 with high {001} facet exposure. J. Photochem. Photobiol. A 2018, 365, 52–59. [CrossRef] 3. Vaiano, V.; Lara, M.A.; Iervolino, G.; Matarangolo, M.; Navio, J.A.; Hidalgo, M.C. Photocatalytic H2 production from glycerol aqueous solutions over ﬂuorinated Pt-TiO2 with high {001} facet exposure. J. Photochem. Photobiol. A 2018, 365, 52–59. [CrossRef] 4. Liu, Y.L.; Yang, C.L.; Wang, M.S.; Ma, X.G.; Yi, Y.G. Te-doped perovskite NaTaO3 as a promising photocatalytic material for hydrogen production from water splitting driven by visible light. Mater. Res. Bull. 2018, 107, 125–131. [CrossRef] 4. Liu, Y.L.; Yang, C.L.; Wang, M.S.; Ma, X.G.; Yi, Y.G. Te-doped perovskite NaTaO3 as a promising photocatalytic material for hydrogen production from water splitting driven by visible light. Mater. Res. Bull. 2018, 107, 125–131. [CrossRef] 5. Cai, Q.; Hu, J. Effect of UVA/LED/TiO2 photocatalysis treated sulfamethoxazole and trimethoprim containing wastewater on antibiotic resistance development in sequencing batch reactors. Water Res. 2018, 140, 251–260. [CrossRef] [PubMed] 5. Cai, Q.; Hu, J. Effect of UVA/LED/TiO2 photocatalysis treated sulfamethoxazole and trimethoprim containing wastewater on antibiotic resistance development in sequencing batch reactors. Water Res. 2018, 140, 251–260. [CrossRef] [PubMed] 6. Lee, C.G.; Javed, H.; Zhang, D.; Kim, J.H.; Westerhoff, P.; Li, Q.; Alvarez, P.J.J. Porous electrospun ﬁbers embedding TiO2 for adsorption and photocatalytic degradation of water pollutants. Environ. Sci. Technol. 2018, 52, 4285–4293. [CrossRef] [PubMed] 7. Zhou, Y.; Li, W.; Wan, W.; Zhang, R.; Lin, Y. W/Mo co-doped BiVO4 for photocatalytic treatment of polymer-containing wastewater in oilﬁeld. Superlattice Microstruct. 2015, 82, 67–74. [CrossRef] 8. Yu, J.; Wang, S.; Low, J.; Xiao, W. Enhanced photocatalytic performance of direct Z-scheme g-C3N4-TiO2 photocatalysts for the decomposition of formaldehyde in air. Phys. Chem. Chem. Phys. 2013, 15, 16883–16890. [CrossRef] [PubMed] 9. Abou Saoud, W.; Assadi, A.A.; Guiza, M.; Bouzaza, A.; Aboussaoud, W.; Ouederni, A.; Soutrel, I.; Wolbert, D.; Rtimi, S. Funding: This research was funded by the National Natural Science Foundation of China (21761132007, 21773168, and 51503143), the National Key R&D Program of China (2016YFE0114900), Tianjin Natural Science Foundation (16JCQNJC05000), Innovation Foundation of Tianjin University (2016XRX-0017), and Tianjin Science and Technology Innovation Platform Program (No. 14TXGCCX00017). 2.4.2. The Role of MoS2 in TiO2 Photocatalysis (a) Schematic illustration for the growth of MoS2 nanosheets (b) Schematic diagram of the photocatalytic H2 generation over the ternary TiO2/WO3@MoS2 heterostructure composite. Reprinted with permission from [137]. Copyright 2017, Elsevier. Figure 22. (a) Schematic illustration for the growth of MoS2 nanosheets (b) Schematic diagram of the photocatalytic H2 generation over the ternary TiO2/WO3@MoS2 heterostructure composite. Reprinted with permission from [137]. Copyright 2017, Elsevier. 405 Catalysts 2018, 8, 590 3. Conclusions The coupling between TiO2 and 2D material has proven to be an efﬁcient approach to enhanced photocatalytic activity. Different methods vary the structures and surface contact of the hybrid and thus can modify the carrier separation process. The synergistic effects show that 2D material plays a vital role in photocatalysis when composited with TiO2. First, 2D material can act as electrons accepter or bridge to conduct photoinduced electrons, and therefore represses the recombination of carriers efﬁciently. Second, the gigantic surface of 2D material provides substantial active sites for substrate capture and reaction, not to mention rapid electrons transfer rate. Third, the 2D material can be decorated to obtain expected properties, for example, non-metal doping to adjust the energy level, speciﬁc crystal structure to short the pathway for interfacial charge transfer, and defects or introduced functional group for substrate trapping. What’s more, the interfacial heterojunction can adjust energy level to broaden light response range and improve solar utilization. To further enhance the separation efﬁciency of electron-hole pairs, other photocatalysts are introduced to construct co-catalyst systems among which Z-scheme system can raise the hole trapping rate to some extent, and thus offers a new point to improve the separation of carriers. All factors mentioned above highlight the critical role of 2D material in photocatalyst and the 2D material/TiO2 hybrid is worth to get further insight for a wider range of applications. Funding: This research was funded by the National Natural Science Foundation of China (21761132007, 21773168, and 51503143), the National Key R&D Program of China (2016YFE0114900), Tianjin Natural Science Foundation (16JCQNJC05000), Innovation Foundation of Tianjin University (2016XRX-0017), and Tianjin Science and Technology Innovation Platform Program (No. 14TXGCCX00017). Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References Study of synergetic effect, catalytic poisoning and regeneration using dielectric barrier discharge and photocatalysis in a continuous reactor: Abatement of pollutants in air mixture system. Appl. Catal. B Environ. 2017, 213, 53–61. [CrossRef] 406 Catalysts 2018, 8, 590 10. Zhang, J.; Ma, Z. Ag3VO4/AgI composites for photocatalytic degradation of dyes and tetracycline hydrochloride under visible light. Mater. Lett. 2018, 216, 216–219. [CrossRef] 11. Huang, C.; Chen, H.; Zhao, L.; He, X.; Li, W.; Fang, W. Biogenic hierarchical MIL-125/TiO2@SiO2 derived from rice husk and enhanced photocatalytic properties for dye degradation. Photochem. Photobiol. 2018, 94, 512–520. [CrossRef] 12. Kumar, M.; Mehta, A.; Mishra, A.; Singh, J.; Rawat, M.; Basu, S. Biosynthesis of tin oxide nanoparticles using psidium guajava leave extract for photocatalytic dye degradation under sunlight. Mater. Lett. 2018, 215, 121–124. [CrossRef] 13. Dashairya, L.; Sharma, M.; Basu, S.; Saha, P. Enhanced dye degradation using hydrothermally synthesized nanostructured Sb2S3/rGO under visible light irradiation. J. Alloy Compd. 2018, 735, 234–245. [CrossRef] 14. Xiao, L.; Lin, R.; Wang, J.; Cui, C.; Wang, J.; Li, Z. A novel hollow-hierarchical structured Bi2WO6 with enhanced photocatalytic activity for CO2 photoreduction. J. Colloid Interface Sci. 2018, 523, 151–158. [CrossRef] [PubMed] 15. Nie, N.; Zhang, L.; Fu, J.; Cheng, B.; Yu, J. Self-assembled hierarchical direct Z-scheme g-C3N4/ZnO microspheres with enhanced photocatalytic CO2, reduction performance. Appl. Surf. Sci. 2018, 441, 12–22. [CrossRef] 16. Xia, P.; Zhu, B.; Yu, J.; Cao, S.; Jaroniec, M. Ultrathin nanosheet assemblies of graphitic carbon nitride enhanced photocatalytic CO2 reduction. J. Mater. Chem. A 2017, 5, 3230–3238. [CrossRef] 17. Fujishima, A.; Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1 238, 37–38. [CrossRef] [PubMed] 18. Verbruggen, S.W. TiO2 photocatalysis for the degradation of pollutants in gas phase: From morphological design to plasmonic enhancement. J. Photochem. Photobiol. C Photochem. Rev. 2015, 24, 64–82. [CrossRef] 19. Zhang, H.; Wang, Z.; Li, R.; Guo, J.; Li, Y.; Zhu, J.; Xie, X. TiO2 supported on reed straw biochar as an adsorptive and photocatalytic composite for the efﬁcient degradation of sulfamethoxazole in aqueous matrices. Chemosphere 2017, 185, 351–360. [CrossRef] [PubMed] 20. Geltmeyer, J.; Teixido, H.; Meire, M.; Van Acker, T.; Deventer, K.; Vanhaecke, F.; Van Hulle, S.; De Buysser, K.; De Clerck, K. TiO2 functionalized nanoﬁbrous membranes for removal of organic (micro)pollutants from water. Sep. Purif. Technol. 2017, 179, 533–541. [CrossRef] p f 21. Shayegan, Z.; Lee, C.-S.; Haghighat, F. References TiO2 photocatalyst for removal of volatile organic compounds in gas phase—A review. Chem. Eng. J. 2018, 334, 2408–2439. [CrossRef] 22. Drunka, R.; Grabis, J.; Jankovica, D.; Krumina, A.; Rasmane, D. Microwave-assisted synthesis and photocatalytic properties of sulphur and platinum modiﬁed TiO2 nanoﬁbers. IOP Conf. Sér. Mater. Sci. Eng. 2015, 77, 012010. [CrossRef] 23. Dong, C.; Song, H.; Zhou, Y.; Dong, C.; Shen, B.; Yang, H.; Matsuoka, M.; Xing, M.; Zhang, J. Sulfur nanoparticles in situ growth on TiO2 mesoporous single crystals with enhanced solar light photocatalytic performance. RSC Adv. 2016, 6, 77863–77869. [CrossRef] 24. Wang, S.; Pan, L.; Song, J.J.; Mi, W.; Zou, J.J.; Wang, L.; Zhang, X. Titanium-defected undoped anatase TiO2 with p-type conductivity, room-temperature ferromagnetism, and remarkable photocatalytic performance. J. Am. Chem. Soc. 2015, 137, 2975–2983. [CrossRef] [PubMed] 25. Hoffmann, M.R.; Choi, W.; Bahnemann, D.W. Environmental applications of semiconductor photocatalysis. Chem. Rev. 1995, 95, 69–96. [CrossRef] 26. Ma, Y.; Li, Z. Coupling plasmonic noble metal with TiO2, for efﬁcient photocatalytic transfer hydrogenation: M/TiO2, (M = Au and Pt) for chemoselective transformation of cinnamaldehyde to cinnamyl alcohol under visible and 365 nm UV Light. Appl. Surf. Sci. 2018, 452, 279–285. [CrossRef] 27. Matos, J.; Llano, B.; Montana, R.; Poon, P.S.; Hidalgo, M.C. Design of Ag/and Pt/TiO2-SiO2 nanomaterials for the photocatalytic degradation of phenol under solar irradiation. Environ. Sci. Pollut. Res. Int. 2018, 25, 18894–18913. [CrossRef] [PubMed] 28. Roberto, F.; Orc, M.B.; Luisa, D.; Giuseppe, C.; Leonardo, P.; Salvatore, S. Au/TiO2-CeO2 catalysts for photocatalytic water splitting and VOCs oxidation reactions. Catalysts 2016, 6, 121. [CrossRef] 29. Md Saad, S.K.; Ali Umar, A.; Ali Umar, M.I.; Tomitori, M.; Abd Rahman, M.Y.; Mat Salleh, M.; Oyama, M. Two-dimensional, hierarchical Ag-doped TiO2 nanocatalysts: Effect of the metal oxidation state on the photocatalytic properties. ACS Omega 2018, 3, 2579–2587. [CrossRef] 407 Catalysts 2018, 8, 590 30. Meryam, Z.; Benoit, S.; Mounira, M.; Ramzi, B.; Yu, W.; Wu, M.; Olivier, D.; Li, Y.; Su, B. ZnO quantum dots decorated 3DOM TiO2 nanocomposites: Symbiose of quantum size effects and photonic structure for highly enhanced photocatalytic degradation of organic pollutants. Appl. Catal. B Environ. 2016, 199, 187–198. [CrossRef] 31. Luisa, M.P.; Sergio, M.-T.; Sónia, A.C.; Josephus, G.B.; José, L.F.; Adrián, M.T.; Silvab, J.L. Photocatalytic activity of functionalized nanodiamond-TiO2 composites towards water pollutants degradation under UV/Vis irradiation. Appl. Surf. Sci. 2018, 458, 839–848. [CrossRef] 32. Prabhu, S.; Cindrella, L.; Kwon, O.J.; Mohanraju, K. References Photoelectrochemical and photocatalytic activity of TiO2-WO3 heterostructures boosted by mutual interaction. Mater. Sci. Semicond. Process. 2018, 88, 10–19. [CrossRef] 33. Zhang, X.; Lai, Z.; Tan, C.; Zhang, H. Solution-processed two-dimensional MoS2 nanosheets: Preparation, hybridization, and applications. Angew. Chem. Int. Ed. Engl. 2016, 55, 8816–8838. [CrossRef] [PubMed] 34. Thomas, A.; Fischer, A.; Goettmann, F.; Antonietti, M.; Müller, J.O.; Schlögl, R.; Carlsson, J.M. Graphitic carbon nitride materials: Variation of structure and morphology and their use as metal-free catalysts. J. Mater. Chem. 2008, 18, 4893–4908. [CrossRef] 35. Hisatomi, T.; Kubota, J.; Domen, K. Recent advances in semiconductors for photocatalytic and photoelectrochemical water splitting. Chem. Soc. Rev. 2014, 43, 7520–7535. [CrossRef] [PubMed] 36. Han, J.H.; Kwak, M.; Kim, Y.; Cheon, J. Recent advances in the solution-based preparation of two-dimensional l d t iti t l h l id t t Ch R 2018 118 6151 6188 [C R f] [P bM d] 36. Han, J.H.; Kwak, M.; Kim, Y.; Cheon, J. Recent advances in the solution-based preparation of two-dimensional layered transition metal chalcogenide nanostructures. Chem. Rev. 2018, 118, 6151–6188. [CrossRef] [PubMed] 37. Bonaccorso, F.; Colombo, L.; Yu, G.; Stoller, M.; Tozzini, V.; Ferrari, A.C.; Ruoff, R.S.; Pellegrini, V. 2D materials. Graphene, related two-dimensional crystals, and hybrid systems for energy conversion and storage. Science 2015, 347, 1246501. [CrossRef] [PubMed] 38. Jin, Z.; Yuan, M.; Li, H.; Yang, H.; Zhou, Q.; Liu, H.; Lan, X.; Liu, M.; Wang, J.; Sargent, E.H.; et al. Graphdiyne: An efﬁcient hole transporter for stable high-performance colloidal quantum dot solar cells. Adv. Funct. Mater. 2016, 26, 5284–5289. [CrossRef] 39. Muller, G.A.; Cook, J.B.; Kim, H.S.; Tolbert, S.H.; Dunn, B. High performance pseudocapacitor based on layered metal chalcogenide nanocrystals. Nano Lett. 2015, 15, 1911–1917. [CrossRef] [PubMed] 40. Wang, X.; Maeda, K.; Thomas, A.; Takanabe, K.; Xin, G.; Carlsson, J.M.; Domen, K.; Antonietti, M. A metal-free polymeric photocatalyst for hydrogen production from water under visible light. Nat. Mater. 2009, 8, 76–80. [CrossRef] [PubMed] 41. Su, T.; Shao, Q.; Qin, Z.; Guo, Z.; Wu, Z. Role of interfaces in two-dimensional photocatalyst for water splitting. ACS Catal. 2018, 8, 2253–2276. [CrossRef] 42. Huang, C.; Li, Y.; Wang, N.; Xue, Y.; Zuo, Z.; Liu, H.; Li, Y. Progress in research into 2D graphdiyne-based materials. Chem. Rev. 2018, 118, 7744–7803. [CrossRef] 43. Fiori, G.; Francesco, B.; Iannaccone, G.; Palacios, T.; Neumaier, D.; Seabaugh, A.; Banerjee, S.K.; Colombo, L. Electronics based on two-dimensional materials. Nat. Nanotechnol. References Enhanced photocatalytic activit graphene oxide nanocomposites. J. Nanopart. Res. 2017, 19. [CrossRef] 56. Sun, J.; Zhang, H.; Guo, L.H.; Zhao, L. Two-dimensional interface engineering of a Titania-graphene nanosheet composite for improved photocatalytic activity. ACS Appl. Mater. Interfaces 2013, 5, 13035–13041. [CrossRef] [PubMed] 57. Lin, L.Y.; Nie, Y.; Kavadiya, S.; Soundappan, T.; Biswas, P. N-doped reduced graphene oxide promoted nano TiO2 as a bifunctional adsorbent/photocatalyst for CO2 photoreduction: Effect of N species. Chem. Eng. J. 2017, 316, 449–460. [CrossRef] 58. Zhu, M.; Muhammad, Y.; Hu, P.; Wang, B.; Wu, Y.; Sun, X.; Tong, Z.; Zhao, Z. Enhanced interfacial contact of dopamine bridged melamine-graphene/TiO2, nano-capsules for efﬁcient photocatalytic degradation of gaseous formaldehyde. Appl. Catal. B Environ. 2018, 232, 182–193. [CrossRef] 59. Xiang, Q.; Yu, J.; Jaroniec, M. Graphene-based semiconductor photocatalysts. Chem. Soc. Rev. 2012, 41, 782–796. [CrossRef] [PubMed] 60. Hummers, J.; Offeman, R.E. Preparation of graphitic oxide. J. Am. Chem. Soc. 1958, 80, 1339. [CrossRef] 61. Bai, H.; Li, C.; Shi, G. Functional composite materials based on chemically converted graphene. Adv. Mater. 2011, 23, 1089–1115. [CrossRef] [PubMed] 62. Rahimi, R.; Zargari, S.; Sadat Shojaei, Z. Photoelectrochemical investigation of TiO2-graphene nanocomposites. In Proceedings of the International Electronic Conference on Synthetic Organic Chemistry, Tehran, Iran, 18 November 2014; pp. 1–10. 63. Pogacean, F.; Rosu, M.C.; Coros, M. Graphene/TiO2-Ag based composites used as sensitive electrode materials for amaranth electrochemical detection and degradation. J. Electrochem. Soc. 2018, 165, B3054–B3059. [CrossRef] 64. Kuila, T.; Bose, S.; Mishra, A.K.; Khanra, P.; Kim, N.H.; Lee, J.H. Chemical functionalization of graphene and its applications. Prog. Mater Sci. 2012, 57, 1061–1105. [CrossRef] 65. Zhang, J.; Xiong, Z.; Zhao, X.S. Graphene–metal–oxide composites for the degradation of dyes under visible light irradiation. J. Mater. Chem. 2011, 21, 3634–3640. [CrossRef] 66. Fan, W.; Lai, Q.; Zhang, Q.; Wang, Y. Nanocomposites of TiO2 and reduced graphene oxide as efﬁcient photocatalysts for hydrogen evolution. J. Phys. Chem. C 2011, 115, 10694–10701. [CrossRef] 66. Fan, W.; Lai, Q.; Zhang, Q.; Wang, Y. Nanocomposites of TiO2 and reduced graphene oxide as efﬁcient photocatalysts for hydrogen evolution. J. Phys. Chem. C 2011, 115, 10694–10701. [CrossRef] 67. Fu, C.; Chen, T.; Qin, W.; Lu, T.; Sun, Z.; Xie, X.; Pan, L. Scalable synthesis and superior performance of TiO2-reduced graphene oxide composite anode for sodium-ion batteries. Ionics 2015, 22, 555–562. [CrossRef] 68. Yan, Y.; Zhang, X.; Mao, H.; Huang, Y.; Ding, Q.; Pang, X. References 2014, 9, 768–779. [CrossRef] [PubMed] 43. Fiori, G.; Francesco, B.; Iannaccone, G.; Palacios, T.; Neumaier, D.; Seabaugh, A.; Banerjee, S.K.; Colombo, L. Electronics based on two-dimensional materials. Nat. Nanotechnol. 2014, 9, 768–779. [CrossRef] [PubMed] 44. Cao, X.; Tao, C.; Zhang, X.; Zhao, W.; Zhang, H. Solution-processed two-dimensional metal dichalcogenide- based nanomaterials for energy storage and conversion. Adv. Mater. 2016, 28, 6167–6196. [CrossRef] [PubMed] 45. Deng, D.; Novoselov, K.S.; Fu, Q.; Zheng, N.; Tian, Z.; Bao, X. Catalysis with two-dimensional materials and their heterostructures. Nat. Nanotechnol. 2016, 11, 218–230. [CrossRef] [PubMed] 46. Novoselov, K.S.; Geim, A.K.; Morozov, S.V.; Jiang, D.; Zhang, Y.; Dubonos, S.V.; Grigorieva, I.V.; Firsov, A.A. Electric ﬁeld effect in atomically thin carbon ﬁlms. Science 2004, 306, 666–669. [CrossRef] [PubMed] 47. Lee, C.; Wei, X.; Kysar, J.W.; Hone, J. Measurement of the elastic properties and intrinsic strength of monolayer graphene. Science 2008, 321, 385–388. [CrossRef] [PubMed] 48. Tang, B.; Hu, G.; Gao, H.; Hai, L. Application of graphene as ﬁller to improve thermal transport property of epoxy resin for thermal interface materials. Int. J. Heat Mass Transf. 2015, 85, 420–429. [CrossRef] 49. Tang, B.; Guoxin, H.; Gao, H. Raman spectroscopic characterization of graphene. Appl. Spectrosc. Rev. 2010, 45, 369–407. [CrossRef] Wang, Q.; Arash, B. A review on applications of carbon nanotubes and graphenes as nano-resonator sensors Comput. Mater. Sci. 2014, 82, 350–360. [CrossRef] 51. Mittal, G.; Dhand, V.; Rhee, K.Y.; Park, S.-J.; Lee, W.R. A review on carbon nanotubes and graphene as ﬁllers in reinforced polymer nanocomposites. J. Ind. Eng. Chem. 2015, 21, 11–25. [CrossRef] 408 Catalysts 2018, 8, 590 52. Zhang, Q.; Wu, Z.; Li, N.; Pu, Y.; Wang, B.; Zhang, T.; Tao, J. Advanced review of graphene-based nanomaterials in drug delivery systems: Synthesis, modiﬁcation, toxicity and application. Mater. Sci. Eng. C Mater. Biol. Appl. 2017, 77, 1363–1375. [CrossRef] [PubMed] 53. Chabot, V.; Higgins, D.; Yu, A.; Xiao, X.; Chen, Z.; Zhang, J. A review of graphene and graphene oxide sponge: Material synthesis and applications to energy and the environment. Energy Environ. Sci. 2014, 7, 1564–1596. [CrossRef] 54. Hu, G.; Tang, B. Photocatalytic mechanism of graphene/titanate nanotubes photocatalyst under visible-light irradiation. Mater. Chem. Phys. 2013, 138, 608–614. [CrossRef] 55. Chen, Y.; Dong, X.; Cao, Y.; Xiang, J.; Gao, H. Enhanced photocatalytic activities of low-bandgap TiO2-reduced graphene oxide nanocomposites. J. Nanopart. Res. 2017, 19. [CrossRef] 55. Chen, Y.; Dong, X.; Cao, Y.; Xiang, J.; Gao, H. References Hydroxyapatite/gelatin functionalized graphene oxide composite coatings deposited on TiO2 nanotube by electrochemical deposition for biomedical applications. Appl. Surf. Sci. 2015, 329, 76–82. [CrossRef] 69. Anjusree, G.S.; Nair, A.S.; Nair, S.V.; Vadukumpully, S. One-pot hydrothermal synthesis of TiO2/graphene nanocomposites for enhanced visible light photocatalysis and photovoltaics. RSC Adv. 2013, 3, 12933–12938. [CrossRef] 70. Shi, M.; Shen, J.; Ma, H.; Li, Z.; Lu, X.; Li, N.; Ye, M. Preparation of graphene–TiO2 composite by hydrothermal method from peroxotitanium acid and its photocatalytic properties. Colloids Surf. Physicochem. Eng. Aspects 2012, 405, 30–37. [CrossRef] 71. Zhang, X.Y.; Li, H.P.; Cui, X.L.; Lin, Y. Graphene/TiO2 nanocomposites: Synthesis, characterization and application in hydrogen evolution from water photocatalytic splitting. J. Mater. Chem. 2010, 20, 2801–2806. [CrossRef] 72. Farhangi, N.; Chowdhury, R.R.; Medina-Gonzalez, Y.; Ray, M.B.; Charpentier, P.A. Visible light active Fe doped TiO2, nanowires grown on graphene using supercritical CO2. Appl. Catal. B Environ. 2011, 110, 25–32. [CrossRef] 409 Catalysts 2018, 8, 590 73. Peng, R.; Liang, L.; Hood, Z.D.; Boulesbaa, A.; Puretzky, A.; Ievlev, A.V.; Come, J.; Ovchinnikova, O.S.; Wang, H.; Ma, C.; et al. In-plane heterojunctions enable multiphasic two-dimensional (2D) MoS2 nanosheets as efﬁcient photocatalysts for hydrogen evolution from water reduction. ACS Catal. 2016, 6, 6723–6729. [CrossRef] 74. Zhu, Z.; Han, Q.; Yu, D.; Sun, J.; Liu, B. A novel p-n heterojunction of BiVO4/TiO2/GO composite for enhanced visible-light-driven photocatalytic activity. Mater. Lett. 2017, 209, 379–383. [CrossRef] 75. Li, G.; Li, Y.; Liu, H.; Guo, Y.; Li, Y.; Zhu, D. Architecture of graphdiyne nanoscale ﬁlms. Chem. Commun. 2010, 46, 3256–3258. [CrossRef] [PubMed] 76. Matsuoka, R.; Sakamoto, R.; Hoshiko, K.; Sasaki, S.; Masunaga, H.; Nagashio, K.; Nishihara, H. Crystalline graphdiyne nanosheets produced at a gas/liquid or liquid/liquid interface. J. Am. Chem. Soc. 2017, 139, 3145–3152. [CrossRef] [PubMed] 77. Yin, X.P.; Wang, H.J.; Tang, S.F.; Lu, X.L.; Shu, M.; Si, R.; Lu, T.B. Engineering the coordination environment of single-atom platinum anchored on graphdiyne for optimizing electrocatalytic hydrogen evolution. Angew. Chem. Int. Ed. Engl. 2018, 57, 9382–9386. [CrossRef] [PubMed] 78. Parvin, N.; Jin, Q.; Wei, Y.; Yu, R.; Zheng, B.; Huang, L.; Zhang, Y.; Wang, L.; Zhang, H.; Gao, M.; et al. Few-layer graphdiyne nanosheets applied for multiplexed real-time DNA detection. Adv. Mater. 2017, 29. [CrossRef] [PubMed] 79. He, J.; Wang, N.; Cui, Z.; Du, H.; Fu, L.; Huang, C.; Yang, Z.; Shen, X.; Yi, Y.; Tu, Z.; et al. References Hydrogen substituted graphdiyne as carbon-rich ﬂexible electrode for lithium and sodium ion batteries. Nat. Commun. 2017, 8, 1172. [CrossRef] [PubMed] 80. Wang, K.; Wang, N.; He, J.; Yang, Z.; Shen, X.; Huang, C. Preparation of 3D architecture graphdiyne nanosheets for high-performance sodium-ion batteries and capacitors. ACS Appl. Mater. Interfaces 2017, 9, 40604–40613. [CrossRef] [PubMed] 81. Diederich, F. Carbon scaffolding-building acetylenic all-carbon and carbon-rich compounds. Nature 1994, 369, 199–207. [CrossRef] 82. Xue, Y.; Li, Y.; Zhang, J.; Liu, Z.; Zhao, Y. 2D graphdiyne materials: Challenges and opportunities in energy ﬁeld. Sci. China Chem. 2018, 61, 765–786. [CrossRef] 83. Yang, N.; Liu, Y.; Wen, H.; Tang, Z.; Zhao, H.; Li, Y.; Wang, D. Photocatalytic properties of graphdiyne a graphene modiﬁed TiO2: From theory to experiment. ACS Nano 2013, 7, 1504–1512. [CrossRef] [PubMe aphene modiﬁed TiO2: From theory to experiment. ACS Nano 2013, 7, 1504–1512. [CrossRef] [PubMed] 84. Wang, S.; Yi, L.; Halpert, J.E.; Lai, X.; Liu, Y.; Cao, H.; Yu, R.; Wang, D.; Li, Y. A novel and highly efﬁci photocatalyst based on P25-graphdiyne nanocomposite. Small 2012, 8, 265–271. [CrossRef] [PubMed] 85. Jia, Z.; Li, Y.; Zuo, Z.; Liu, H.; Huang, C.; Li, Y. Synthesis and properties of 2D carbon-graphdiyne. Acc. Chem. Res. 2017, 50, 2470–2478. [CrossRef] [PubMed] 86. Dong, Y.; Zhao, Y.; Chen, Y.; Feng, Y.; Zhu, M.; Ju, C.; Zhang, B.; Liu, H.; Xu, J. Graphdiyne-hybridized N-doped TiO2 nanosheets for enhanced visible light photocatalytic activity. J. Mater. Sci. 2018, 53, 8921–8932. [CrossRef] 87. Li, G.; Li, Y.; Qian, X.; Liu, H.; Lin, H.; Chen, N.; Li, Y. Construction of tubular molecule aggregations of graphdiyne for highly efﬁcient ﬁeld emission. J. Phys. Chem. C 2011, 115, 2611–2615. [CrossRef] 88. Zhou, J.; Gao, X.; Liu, R.; Xie, Z.; Yang, J.; Zhang, S.; Zhang, G.; Liu, H.; Li, Y.; Zhang, J.; et al. Synthesis of graphdiyne nanowalls using acetylenic coupling reaction. J. Am. Chem. Soc. 2015, 137, 7596–7599. [CrossRef] [PubMed] 89. Li, Y.; Xu, L.; Liu, H.; Li, Y. Graphdiyne and graphyne: From theoretical predictions to practical construction. Chem. Soc. Rev. 2014, 43, 2572–2586. [CrossRef] [PubMed] 90. Thangavel, S.; Krishnamoorthy, K.; Krishnaswamy, V.; Raju, N.; Kim, S.J.; Venugopal, G. Graphdiyne–ZnO nanohybrids as an advanced photocatalytic material. J. Phys. Chem. C 2015, 119, 22057–22065. [CrossRef] 91. Long, M.; Tang, L.; Wang, D.; Li, Y.; Shuai, Z. Electronic structure and carrier mobility in graphdiyne s and nanoribbons: Theoretical predictions. ACS Nano 2011, 5, 2593–2600. References [CrossRef] [PubMed] 92. Qi, H.; Yu, P.; Wang, Y.; Han, G.; Liu, H.; Yi, Y.; Li, Y.; Mao, L. Graphdiyne oxides as excellent substrate for electroless deposition of pd clusters with high catalytic activity. J. Am. Chem. Soc. 2015, 137, 5260–5263. [CrossRef] [PubMed] 93. Zhang, X.; Zhu, M.; Chen, P.; Li, Y.; Liu, H.; Li, Y.; Liu, M. Pristine graphdiyne-hybridized photocatalysts using graphene oxide as a dual-functional coupling reagent. Phys. Chem. Chem. Phys. 2015, 17, 1217–1225. [CrossRef] [PubMed] 410 Catalysts 2018, 8, 590 94. Zhou, J.; Zhang, M.; Zhu, Y. Photocatalytic enhancement of hybrid C3N4/TiO2 prepared via ball milling method. Phys. Chem. Chem. Phys. 2015, 17, 3647–3652. [CrossRef] [PubMed] 95. Gu, L.; Wang, J.; Zou, Z.; Han, X. Graphitic-C3N4-hybridized TiO2 nanosheets with reactive {001} facets to enhance the UV- and visible-light photocatalytic activity. J. Hazard. Mater. 2014, 268, 216–223. [CrossRef] [PubMed] 96. Li, H.; Gao, Y.; Wu, X.; Lee, P.H.; Shih, K. Fabrication of heterostructured g-C3N4/Ag-TiO2 hybrid photocatalyst with enhanced performance in photocatalytic conversion of CO2 under simulated sunlight irradiation. Appl. Surf. Sci. 2017, 402, 198–207. [CrossRef] 97. Li, K.; Huang, Z.; Zeng, X.; Huang, B.; Gao, S.; Lu, J. Synergetic effect of Ti3+ and oxygen doping on enhancing photoelectrochemical and photocatalytic properties of TiO2/g-C3N4 heterojunctions. ACS Appl. Mater. Interfaces 2017, 9, 11577–11586. [CrossRef] [PubMed] 98. Zhang, C.; Zhou, Y.; Bao, J.; Fang, J.; Zhao, S.; Zhang, Y.; Sheng, X.; Chen, W. Structure regulation of ZnS@g-C3N4/TiO2 nanospheres for efﬁcient photocatalytic H2 production under visible-light irradiation. Chem. Eng. J. 2018, 346, 226–237. [CrossRef] 99. Song, J.; Wang, X.; Ma, J.; Wang, X.; Wang, J.; Xia, S.; Zhao, J. Removal of microcystis aeruginosa and microcystin-LR using a graphitic-C3N4/TiO2 ﬂoating photocatalyst under visible light irradiation. Chem. Eng. J. 2018, 348, 380–388. [CrossRef] g 100. Ricci, P.C.; Laidani, N.; Chiriu, D.; Salis, M.; Carbonaro, C.M.; Corpino, R. ALD growth of metal oxide on carbon nitride polymorphs. Appl. Surf. Sci. 2018, 456, 83–94. [CrossRef] 101. Ren, B.; Wang, T.; Qu, G.; Deng, F.; Liang, D.; Yang, W.; Liu, M. In situ synthesis of g-C3N4/TiO2 heterojunction nanocomposites as a highly active photocatalyst for the degradation of orange ii under visible light irradiation. Environ. Sci. Pollut. Res. 2018, 25, 19122–19133. [CrossRef] [PubMed] 102. Wang, X.; Wang, F.; Bo, C.; Cheng, K.; Wang, J.; Zhang, J.; Song, H. Promotion of phenol photodecomposition and the corresponding decomposition mechanism over g-C3N4/TiO2 nanocomposites. Appl. Surf. Sci. 2018, 453, 320–329. [CrossRef] 103. References Engineering the edges of MoS2 (WS2) crystals for direct exfoliation into monolayers in polar micromolecular solvents. J. Am. Chem. Soc. 2016, 138, 14962–14969. [CrossRef] [PubMed] 115. Han, W.; Zang, C.; Huang, Z.; Zhang, H.; Ren, L.; Qi, X.; Zhong, J. Enhanced photocatalytic activities of three-dimensional graphene-based aerogel embedding TiO2 nanoparticles and loading MoS2 nanosheets as co-catalyst. Int. J. Hydrogen Energy 2014, 39, 19502–19512. [CrossRef] 116. Nimbalkar, D.B.; Lo, H.-H.; Ramacharyulu, P.V.R.K.; Ke, S.C. Improved photocatalytic activity of rGO/MoS2 nanosheets decorated on TiO2 nanoparticles. RSC Adv. 2016, 6, 31661–31667. [CrossRef] 117. Wang, D.; Xu, Y.; Sun, F.; Zhang, Q.; Wang, P.; Wang, X. Enhanced photocatalytic activity of TiO2 under sunlight by MoS2 nanodots modiﬁcation. Appl. Surf. Sci. 2016, 377, 221–227. [CrossRef] 118. Pi, Y.; Li, Z.; Xu, D.; Liu, J.; Li, Y.; Zhang, F.; Zhang, G.; Peng, W.; Fan, X. 1T-phase MoS2 nanosheets on TiO2 nanorod arrays: 3D photoanode with extraordinary catalytic performance. ACS Sustain. Chem. Eng. 2017, 5, 5175–5182. [CrossRef] 119. Yu, Y.; Wan, J.; Yang, Z.; Hu, Z. Preparation of the MoS2/TiO2/HMFS ternary composite hollow microﬁbres with enhanced photocatalytic performance under visible light. J. Colloid Interface Sci. 2017, 502, 100–111. [CrossRef] [PubMed] 120. Yuan, Y.-J.; Ye, Z.J.; Lu, H.W.; Hu, B.; Li, Y.-H.; Chen, D.Q.; Zhong, J.S.; Yu, Z.T.; Zou, Z.G. Constructing anatase TiO2 nanosheets with exposed (001) facets/layered MoS2 two-dimensional nanojunctions for enhanced solar hydrogen generation. ACS Catal. 2015, 6, 532–541. [CrossRef] 121. Liu, X.; Xing, Z.; Zhang, H.; Wang, W.; Zhang, Y.; Li, Z.; Wu, X.; Yu, X.; Zhou, W. Fabrication of 3D mesoporous black TiO2/MoS2/TiO2 nanosheets for visible-light-driven photocatalysis. ChemSusChem 2016, 9, 1118–1124. [CrossRef] [PubMed] 122. He, H.; Lin, J.; Fu, W.; Wang, X.; Wang, H.; Zeng, Q.; Gu, Q.; Li, Y.; Yan, C.; Tay, B.K.; et al. MoS2/TiO2 edge-on heterostructure for efﬁcient photocatalytic hydrogen evolution. Adv. Funct. Mater. 2016, 6, 1600464. [CrossRef] 123. Van Haandel, L.; Geus, J.W.; Weber, T. Direct synthesis of TiO2-supported MoS2 nanoparticles by reductive coprecipitation. ChemCatChem 2016, 8, 1367–1372. [CrossRef] 124. Ren, X.; Qi, X.; Shen, Y.; Xiao, S.; Xu, G.; Zhang, Z.; Huang, Z.; Zhong, J. 2D co-catalytic MoS2 nanosheets embedded with 1D TiO2 nanoparticles for enhancing photocatalytic activity. J. Phys. D Appl. Phys. 2016, 49, 315304. [CrossRef] 125. Bai, S.; Wang, L.; Chen, X.; Du, J.; Xiong, Y. Chemically exfoliated metallic MoS2 nanosheets: A promising supporting co-catalyst for enhancing the photocatalytic performance of TiO2 nanocrystals. Nano Res. References Wu, D.; Li, J.; Guan, J.; Liu, C.; Zhao, X.; Zhu, Z.; Ma, C.; Huo, P.; Li, C.; Yan, Y. Improved photoelectric performance via fabricated heterojunction g-C3N4/TiO2/HNTS loaded photocatalysts for photodegradation of ciproﬂoxacin. J. Ind. Eng. Chem. 2018, 64, 206–218. [CrossRef] 104. Tan, Y.; Shu, Z.; Zhou, J.; Li, T.; Wang, W.; Zhao, Z. One-step synthesis of nanostructured g-C3N4/TiO2 composite for highly enhanced visible-light photocatalytic H2 evolution. Appl. Catal. B Environ. 2018, 230, 260–268. [CrossRef] 105. Elbanna, O.; Fujitsuka, M.; Majima, T. g-C3N4/TiO2 mesocrystals composite for H2 evolution under visible-light irradiation and its charge carrier dynamics. ACS Appl. Mater. Interfaces 2017, 9, 34844–34854. [CrossRef] [PubMed] 106. Guo, N.; Zeng, Y.; Li, H.; Xu, X.; Yu, H.; Han, X. Novel mesoporous TiO2@g-C3N4 hollow core@shell heterojunction with enhanced photocatalytic activity for water treatment and H2 production under simulated sunlight. J. Hazard. Mater. 2018, 353, 80–88. [CrossRef] [PubMed] 107. Jiang, Y.; Li, F.; Liu, Y.; Hong, Y.; Liu, P.; Ni, L. Construction of TiO2 hollow nanosphere/g-C3N4 composites with superior visible-light photocatalytic activity and mechanism insight. J. Ind. Eng. Chem. 2016, 41, 130–140. [CrossRef] 108. Chen, Y.; Huang, W.; He, D.; Situ, Y.; Huang, H. Construction of heterostructured g-C3N4/Ag-TiO2 microspheres with enhanced photocatalysis performance under visible-light irradiation. ACS Appl. Mater. Interfaces 2014, 6, 14405–14414. [CrossRef] [PubMed] 109. Zhao, Y.; Xu, S.; Sun, X.; Xu, X.; Gao, B. Unique bar-like sulfur-doped g-C3N4/TiO2 nanocomposite: Excellent visible light driven photocatalytic activity and mechanism study. Appl. Surf. Sci. 2018, 436, 873–881. [CrossRef] 110. Cheng, L.; Liu, J.; Gu, X.; Gong, H.; Shi, X.; Liu, T.; Wang, C.; Wang, X.; Liu, G.; Xing, H.; et al. Pegylated WS2 nanosheets as a multifunctional theranostic agent for in vivo dual-modal CT/photoacoustic imaging guided photothermal therapy. Adv. Mater. 2014, 26, 1886–1893. [CrossRef] [PubMed] . Zheng, L.; Xiao, X.; Li, Y.; Zhang, W. Enhanced photocatalytic activity of TiO2 nanoparticles using WS2/g-C3N4 hybrid as co-catalyst. Trans. Nonferrous Met. Soc. China 2017, 27, 1117–1126. [CrossRef] 112. He, H.Y. Solvothermal synthesis and photocatalytic activity of s-doped TiO2 and TiS2 powders. Res. Chem. Intermed. 2010, 36, 155–161. [CrossRef] 411 Catalysts 2018, 8, 590 113. Shen, M.; Yan, Z.; Yang, L.; Du, P.; Zhang, J.; Xiang, B. MoS2 nanosheet/TiO2 nanowire hybrid nanostructures for enhanced visible-light photocatalytic activities. Chem. Commun. 2014, 50, 15447–15449. [CrossRef] [PubMed] 114. Hai, X.; Chang, K.; Pang, H.; Li, M.; Li, P.; Liu, H.; Shi, L.; Ye, J. References 2014, 8, 175–183. [CrossRef] 126. Kim, Y.; Jackson, D.H.K.; Lee, D.; Choi, M.; Kim, T.W.; Jeong, S.-Y.; Chae, H.J.; Kim, H.W.; Park, N.; Chang, H.; et al. In situ electrochemical activation of atomic layer deposition coated MoS2 basal planes for efﬁcient hydrogen evolution reaction. Adv. Funct. Mater. 2017, 27, 1701825. [CrossRef] 127. Liu, X.; Xing, Z.; Zhang, Y.; Li, Z.; Wu, X.; Tan, S.; Yu, X.; Zhu, Q.; Zhou, W. Fabrication of 3D ﬂower-like black n- TiO2−x@MoS2 for unprecedented-high visible-light-driven photocatalytic performance. Appl. Catal. B Environ. 2017, 201, 119–127. [CrossRef] 128. Sun, Y.; Lin, H.; Wang, C.; Wu, Q.; Wang, X.; Yang, M. Morphology-controlled synthesis of TiO2/MoS2 nanocomposites with enhanced visible-light photocatalytic activity. Inorg. Chem. Front. 2018, 5, 145–152. [CrossRef] 129. Pu, S.; Long, D.; Wang, M.Q.; Bao, S.J.; Liu, Z.; Yang, F.; Wang, H.; Zeng, Y. Design, synthesis and photodegradation ammonia properties of MoS2@TiO2 encapsulated carbon coaxial nanobelts. Mater. Lett. 2017, 209, 56–59. [CrossRef] 130. He, H.Y. Efﬁcient hydrogen evolution activity of 1T- MoS2/Si-doped TiO2 nanotube hybrids. Int. J. Hydrogen Energy 2017, 42, 20739–20748. [CrossRef] 412 Catalysts 2018, 8, 590 131. Han, H.; Kim, K.M.; Lee, C.W.; Lee, C.S.; Pawar, R.C.; Jones, J.L.; Hong, Y.R.; Ryu, J.H.; Song, T.; Kang, S.H.; et al. Few-layered metallic 1T-MoS2/TiO2 with exposed (001) facets: Two-dimensional nanocomposites for enhanced photocatalytic activities. Phys. Chem. Chem. Phys. 2017, 19, 28207–28215. [CrossRef] [PubMed] 132. Lin, T.; Kang, B.; Jeon, M.; Huffman, C.; Jeon, J.; Lee, S.; Han, W.; Lee, J.; Lee, S.; Yeom, G.; et al. Controlled layer-by-layer etching of MoS2. ACS Appl. Mater. Interfaces 2015, 7, 15892–15897. [CrossRef] [PubMed] 133. Yu, L.; Xie, Y.; Zhou, J.; Li, Y.; Yu, Y.; Ren, Z. Robust and selective electrochemical reduction of CO2: The case of integrated 3D TiO2@MoS2 architectures and Ti–S bonding effects. J. Mater. Chem. A 2018, 6, 4706–4713. [CrossRef] 134. Jaramillo, T.F.; Jorgensen, K.P.; Bonde, J.; Nielsen, J.H.; Horch, S.; Chorkendorff, I. Identiﬁcation of active edge sites for electrochemical H2 evolution from MoS2 nanocatalysts. Science 2007, 317, 100–102. [CrossRef] [PubMed] 135. Xiang, Q.; Yu, J.; Jaroniec, M. Synergetic effect of MoS2 and graphene as cocatalysts for enhanced photocatalytic H2 production activity of TiO2 nanoparticles. J. Am. Chem. Soc. 2012, 134, 6575–6578. [CrossRef] [PubMed] 136. Hu, X.; Zhao, H.; Tian, J.; Gao, J.; Li, Y.; Cui, H. Synthesis of few-layer MoS2 nanosheets-coated TiO2 nanosheets on graphite ﬁbers for enhanced photocatalytic properties. Sol. Energy Mater. Sol. Nick Serpone PhotoGreen Laboratory, Dipartimento di Chimica, Università di Pavia, via Taramelli 12, 27011 Pavia, Italy; nick.serpone@unipv.it; Tel.: +1-514-489-9551 Received: 18 October 2018; Accepted: 5 November 2018; Published: 16 November 2018 Abstract: This article reviews the efforts of the last two decades to deNOxify the atmospheric environment with TiO2-based photocatalytic materials supported on various cementitious-like substrates. Prior to undertaking this important aspect of applied photocatalysis with metal-oxide emiconductor photocatalysts, however, it is pertinent to describe and understand the fundamentals of Heterogeneous Photocatalysis. The many attempts done in a laboratory setting to degrade (deNOxify) the major components that make up the NOx, namely nitric oxide (NO) and nitrogen dioxide (NO2), but most importantly the efforts expended in deNOxifying the real environment upon depositing titania-based coatings on various model and authentic infrastructures, such as urban roads, highway noise barriers, tunnels, and building external walls among others, are examined. Both laboratory and outdoor experimentations have been performed toward NOx being oxidized to form nitrates (NO3−) that remain adsorbed on the TiO2-based photocatalytic surfaces (except in tunnels—indoor walls) but get subsequently dislodged by rain or by periodic washings of the infrastructures. However, no serious considerations have been given to the possible conversion of NOx via photocatalytic reduction back to N2 and O2 gases that would restore the atmospheric environment, as the adsorbed nitrates block the surface-active sites of the photocatalyst and when washed-off ultimately cause unduly damages to the environment. Keywords: photocatalysis; deNOxing; Titania; photophysics; metal oxides; environment catalysts catalysts Catalysts 2018, 8, 553; doi:10.3390/catal8110553 References Cells 2017, 172, 108–116. [CrossRef] 137. Zhao, J.; Zhang, P.; Fan, J.; Hu, J.; Shao, G. Constructing 2D layered MoS2 nanosheets-modiﬁed z-scheme TiO2/WO3 nanoﬁbers ternary nanojunction with enhanced photocatalytic activity. Appl. Surf. Sci. 2018, 430, 466–474. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 413 www.mdpi.com/journal/catalysts 1. Introduction The atmosphere is a very complex matrix that, in addition to nitrogen (N2) and oxygen (O2) gases, so important to human survival, consists of various other pollutant gases albeit at much lower concentrations: (i) carbonaceous oxides (COx); (ii) nitrogen oxides (NOx); (iii) sulfur oxides (SOx); (iv) various hydrocarbons (HCs); and (v) particulate matter. These pollutants originate from both natural sources (e.g., volcanic eruptions, wildﬁres, lightning, and natural degradation of forests among others) and anthropogenic areas (fertilizers and livestock, farms, and urban areas), stationary sources (e.g., industries, power plants, and sewage treatment plants), and mobile sources (e.g., automobiles, trucks, buses, motorcycles, ships, and airplanes) (see Figure 1) [1]. The natural sources of chemical pollutants, however, are of lesser concerns as they are part of the natural environment equilibrium, contrary to the anthropogenic sources that keep increasing in number and concentration with the ever increasing global human population and society’s continuous increasing demand for energy and associated technological advances. Undoubtedly, the most important sources of air pollutants implicate the combustion of fossil fuels to produce energy (residential heating and electricity-generating power plants), together with major metallurgical industries, cement/construction industries, and the transportation sector. Figure 1 also identiﬁes the primary pollutants from various sources: carbon monoxide, sulfur dioxide, ammonia, Catalysts 2018, 8, 553; doi:10.3390/catal8110553 www.mdpi.com/journal/catalysts 415 Catalysts 2018, 8, 553 volatile organic compounds (VOCs), particulates and, relevant to the present article, the two major NOx agents (NO and NO2). Subsequently, through various interacting events that involve the Sun’s radiation, secondary pollutants are generated, among which are sulfur trioxide, ozone, hydrogen peroxide, and sulfuric and nitric acids (the causes of acid rain). Another class of air pollutants generated from internal combustion engines and industrial fumes that react in the atmosphere with sunlight produce secondary pollutants that, in combination with the primary emissions, create photochemical smog [2]. +62+12 1DWXUDO VRXUFHV 8UEDQ 5XUDO VRXUFHV 6WDWLRQDU\ VRXUFHV 7UDQVSRUW 3ULPDU\3ROOXWDQWV 6HFRQGDU\3ROOXWDQWV &2621+92&V3DUWLFXODWHV 1212 12[ 62+221+ 3DUWLFXODWHV $FLG5DLQ Figure 1. Graphic illustrating the natural and anthropogenic sources of atmospheric pollution together with primary and secondary atmospheric pollutants from various sources that ultimately lead to the formation of acid rain. Figure 1. Graphic illustrating the natural and anthropogenic sources of atmospheric pollution together with primary and secondary atmospheric pollutants from various sources that ultimately lead to the formation of acid rain. Pre-industrial concentrations of atmospheric nitrogen oxides have increased steadily from about 280 ppbv to ca. OH + CO →HO•CO 1. Introduction Nitrogen dioxide also reacts with hydrocarbon molecules present in VOCs to produce yet another pollutant (peroxyacetyl nitrates; PAN), a component of photochemical smog that is mostly responsible for eye irritation and is more damaging to plants than ozone [8]. (a) (b) URDGWUDQVSRUW HQHUJ\SURGXFWLRQ GLVWULEXWLRQ FRPPHUFLDOLQVWLWXWLRQDO KRXVHKROGV HQHUJ\XVHLQLQGXVWU\ QRQURDGWUDQVSRUW LQGXVWULDOSURFHVVHV DJULFXOWXUH 3HUFHQW 3HUFHQW URDGYHKLFOHV QRQURDGHTXLSPHQW HOHFWULFLW\JHQHUDWLRQ IRVVLOIXHOFRPEXVWLRQ LQGXVWULDOSURFHVVHV ZDVWHGLVSRVDO ILUHV PLVFHOODQHRXV Figure 2. (a) Percent emission of NOx agents from various sources for 2011 in the European Union (Source: European Union emission inventory report 1990–2011 under the UNECE Convention on Long-range Trans-boundary Air Pollution (LRTAP)). Reproduced from Ref. [4]. (b) Percent emission of NOx agents for 2005 in the United States (Source: U.S. Environmental Protection Agency, Air Emission Sources, 4 November 2009). Reproduced from Ref. [5]. (b) 3HUFHQW URDGYHKLFOHV QRQURDGHTXLSPHQW HOHFWULFLW\JHQHUDWLRQ IRVVLOIXHOFRPEXVWLRQ LQGXVWULDOSURFHVVHV ZDVWHGLVSRVDO ILUHV PLVFHOODQHRXV (a) URDGWUDQVSRUW HQHUJ\SURGXFWLRQ GLVWULEXWLRQ FRPPHUFLDOLQVWLWXWLRQDO KRXVHKROGV HQHUJ\XVHLQLQGXVWU\ QRQURDGWUDQVSRUW LQGXVWULDOSURFHVVHV DJULFXOWXUH 3HUFHQW (b) (a) Figure 2. (a) Percent emission of NOx agents from various sources for 2011 in the European Union (Source: European Union emission inventory report 1990–2011 under the UNECE Convention on Long-range Trans-boundary Air Pollution (LRTAP)). Reproduced from Ref. [4]. (b) Percent emission of NOx agents for 2005 in the United States (Source: U.S. Environmental Protection Agency, Air Emission Sources, 4 November 2009). Reproduced from Ref. [5]. The NOx family of pollutants (NO, NO2, N2O, and their derivatives) causes a wide range of health issues. Nitric oxide (NO) spreads to all parts of the respiratory system because of its low solubility in water, while the health effects of NO2 are related to its ability to dissolve in moisture to produce HNO3 acid—a strong mineral acid. Some of the acute health effects include eye irritation (stinging and watering), throat irritation (pungent smell, stinging nose, and coughing), lung irritation (coughing, wheezing and tight chest—difﬁculty in breathing), and asthma triggered in asthmatics [9,10]. The most serious acute effects occur after signiﬁcant exposure to NO2 causing: (a) acute pulmonary edema—ﬂuid from damaged lung tissue pours out into air spaces preventing air from getting to deeper lung thereby causing choking (asphyxia); and (b) other chronic health effects such as asthma and obliterative bronchiolitis, in which the smallest air passages (the bronchioles) are seriously scarred and become distorted and blocked. 1. Introduction 320 ppbv until a decade ago (2010), with estimated annual emissions of 13.8 Tg of N per year (teragrams; 1012 g), of which ca. 70% is produced by nitriﬁcation and denitriﬁcation processes in undisturbed terrestrial environments and world’s oceans, and ca. 3 Tg of N per year (~ 8%) from agricultural tillage, fertilizer use, and animal wastes [3]. The NOx gases are formed in large measure in gasoline/diesel combustion engines and in power plants that use fossil fuels to produce electricity via high-temperature combustion/oxidation of the fuel’s nitrogen with air oxygen. Initially, only NO is formed followed by formation of NO2 after combustion in the exhaust and in the atmosphere in the presence of more O2. Figure 2a reports the 2011 levels of NOx emissions in the European Union [4], while Figure 2b reports the 2005 NOx emission levels in the United States [5]. The major anthropogenic sources of nitrogen oxides are combustion engines (transportation sector) and the electricity/heating sectors. Most of the tropospheric ozone is formed when NOx, CO and VOCs react in the atmosphere in the presence of sunlight, and, although they might originate in urban areas, airstreams can carry the NOx far from its sources causing ozone formation in less populated regions. Globally, a VOC whose atmospheric concentration has increased greatly during the last century (viz., methane) contributes to the formation of ozone [6]. A series of complex reactions that involve a VOC (e.g., CO) in the formation of ozone implicates oxidation of this VOC by a hydroxyl radical (•OH) [7] ﬁrst to yield the radical species HO•CO (Equation (1)), which subsequently reacts with oxygen to produce the hydroperoxy radical HOO• (Equation (2)) that later reacts with NO to give NO2 (Equation (3)); the latter photolyzes in sunlight to NO and atomic O(3P) (Equation (4)), which by reaction with oxygen yields ozone (O3; Equation (5)). (1) OH + CO →HO•CO 416 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 HO•CO + O2 →HOO• + CO2 (2) HOO• + NO →•OH + NO2 (3) NO2 + hν →NO + O(3P) (4) O(3P) + O2 →O3 (5) (4) (5) While the chemistry involving other VOCs might be more complex, the critical step that leads to ozone formation remains nonetheless the oxidation of NO to NO2 by HOO• radicals. 1. Introduction Consequently, no one questions the need for NOx-free clean air as essential to maintain/enhance an individual’s health, and to maintain the integrity of the surrounding environment. In this regard, transformation of the two major NOx species (NO and NO2), indeed their suppression, has become a necessity as they underwrite (with the VOCs) the formation of hazardous secondary air pollutants and the accompanying photochemical smog. Two cities where photochemical smog is not insigniﬁcant are the Greater Los Angeles (LA) area in the United States and Beijing, China. Home to nearly 19 million people and located in a geological basin conﬁned by the Paciﬁc Ocean and mountains, LA is the basin of considerable pollution caused by its car-centric culture, its bustling industries and ports, its sprawling development, and its sunny climate with often stagnant winds. It was only in the 1950s that hydrocarbons and NOx were recognized as the source of photochemical smog (Figure 3); however, with the implementation of mandatory catalytic converters in automobiles in the last two decades, smog has been attenuated somewhat. 417 Catalysts 2018, 8, 553 Figure 3. Photochemical smog appearing in the Greater Los Angeles area, USA. (Source: United States Geological Survey); see, e.g., https://serc.carleton.edu/eet/aura/case_study.html (accessed 10 November 2018). Figure 3. Photochemical smog appearing in the Greater Los Angeles area, USA. (Source: United States Geological Survey); see, e.g., https://serc.carleton.edu/eet/aura/case_study.html (accessed 10 November 2018). With its rapid growth and home to nearly 20 million people, and being an important industrial hub, Beijing is a city where poor air quality has been for decades a regrettable fact of everyday life owing to the presence of signiﬁcant quantities of particulate matter and photochemical smog, as (Figure 4) experienced by the author in the early 1990s; in subsequent trips several years later, however, this author experienced signiﬁcant improvements of air quality but by no means have the pollutants and smog been totally eliminated. Figure 4. Photochemical smog appearing on January 2016 in Beijing, China. Reproduced from https: //i1.wp.com/dnnsociety.org/wp-content/uploads/2016/01/beijing.jpg?resize=845%2C450&ssl=1 (accessed 10 November 2018). Figure 4. Photochemical smog appearing on January 2016 in Beijing, China. Reproduced from https: //i1.wp.com/dnnsociety.org/wp-content/uploads/2016/01/beijing.jpg?resize=845%2C450&ssl=1 (accessed 10 November 2018). There have been many attempts to remediate the occurrence of NOx and VOC species in polluted urban environments with TiO2-based photocatalytic cementitious-like materials and photocatalytic coatings (paints) on various supports [11]. 1. Introduction Several studies report on the performance of titania deposited on, or otherwise incorporated into cementitious substrates toward the minimization, if not suppression, of air pollutants (see, for example, Refs. [12–18]). Laboratory studies have shown, rather conclusively, that NOx can be oxidized to nitrate anions [19,20], while VOCs can be converted into CO2 and H2O [21]. Of some concern, however, are studies that demonstrate the formation of harmful intermediates (e.g., nitrous acid, HONO), which are far more harmful to human health than either NO or NO2 during the disposal of NOx [22,23]. Not least is the potential that nitrates (NO3−) produced and deposited on the TiO2 particulate surface in the disposal of NOx may be implicated in reNOxiﬁcation reactions; that is, back to NOx [24–26] and formation of ozone [26] that would forestall the application of TiO2-based photocatalytic surfaces to improve the quality of urban air environments. 418 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 The objective of this review article is to examine the various attempts at eliminating NOx species in the urban environment produced mostly by vehicular trafﬁc through application of commercially available titania-based photocatalytic materials, coatings and paints in tunnels, highways, highway noise barriers, and urban roads. However, before tackling that discussion, we describe brieﬂy some fundamentals from basic research that underpin this TiO2-based photocatalytic technology. In its pristine or modiﬁed form, TiO2 has been the most popular and most extensively investigated photocatalyst, and is the primary source of modern third generation composite photoactive materials [27,28]. Photocatalytic processes occurring in heterogeneous systems are complex and multifarious starting from the absorption of photons by the solid photocatalyst, and ending with the evolution of reaction products. This complexity is particularly reﬂected in the terminology used to describe various characteristics of heterogeneous photocatalysis, which, although it has come to some maturity in recent years, continues to undergo extensive developments through efforts of many researchers from the ﬁelds of catalysis, photochemistry and materials science, among others. Accordingly, prior to tackling the many deNOxing efforts in cleaning up the atmospheric environment, an important aspect of applied photocatalysis with metal-oxide semiconductor photocatalysts, it is imperative to appreciate and understand some of the fundamentals underlying Heterogeneous Photocatalysis (following Sections 2 and 3)—the primary approach in these efforts. 2. Some Fundamentals of Heterogeneous Photocatalysis 2. Some Fundamentals of Heterogeneous Photocatalysis Historically, Heterogeneous Photocatalysis is an interdisciplinary ﬁeld at the intersection of Chemistry and Physics. It rests on four basic pillars (Figure 5): (a) heterogeneous catalysis; (b) photochemistry; (c) molecular/solid-state spectroscopy; and (d) materials science of semiconductor photocatalysts of interest (in the present context: metal oxides). Accordingly, it is worth looking into each of these pillars to assess how they have shaped present-day knowhow, particularly Pillars (a)–(c) (Section 2) together with the photophysics of metal-oxide semiconductors (Section 3). Figure 5. The four pillars that have had a great impact on the development of heterogeneous photocatalysis. Figure 5. The four pillars that have had a great impact on the development of heterogeneous photocatalysis. Currently, photocatalysis is best described as a change in the rate of a chemical reaction or its initiation under the action of ultraviolet, visible, or infrared radiation in the presence of a substance that absorbs light and is involved in the chemical transformation of the reaction partners; the photocatalyst is the substance that causes, by absorption of ultraviolet, visible, or infrared radiation, the chemical transformation of the reaction partners, repeatedly coming into intermediate chemical interactions with them and regenerating its chemical composition after each cycle of such interactions [29]. 419 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 2.1. Inﬂuence of Catalysis on Photocatalysis—Comparisons and Contrasts Some researchers consider photocatalysis a component of the ﬁeld of Catalysis. However, some postulates that are typical of traditional catalysis are somewhat antagonistic to photocatalysis. For instance, by analogy with catalysis, one of the postulates would suggest that a photocatalytic process should favor only thermodynamically allowed chemical reactions. From this point of view, light should cause a decrease of the potential energy barrier relative to the dark catalytic reaction and thus accelerate the establishment of chemical equilibrium between reagents and products in the heterogeneous system. Photoreactions that take place in heterogeneous systems and are thermodynamically unfavorable (e.g., photolysis of water) are classiﬁed as non-photocatalytic but photosynthetic, despite similarities in all major steps of the photoprocesses. The sole reason for both types of photoreactions is the free energy of the actinic light. In fact, the action of the free energy of light turns the system into a thermodynamically open system so that, by deﬁnition, the concept of thermodynamic equilibrium is not applicable. 2. Some Fundamentals of Heterogeneous Photocatalysis The deﬁnitions given to photocatalysis and photocatalysts are very similar to the deﬁnitions of catalysis and catalyst commonly used in conventional (thermal) catalysis. This similarity suggests a strong impact of catalysis to the ﬁeld of photocatalysis. According to current thinking, photocatalysis is viewed as an alteration of the reaction rate in the presence of a substance that interacts repeatedly with reagents subsequent to which its original state is restored after each reaction cycle just like in catalytic processes. Thus, photocatalysis could, in principle, be considered as catalysis involving the action of light. Formally speaking then, the transformation of a reagent in a catalytic process may be exempliﬁed by a simple chemical reaction (e.g., Equation (6)): Cat + R →Cat + P (6) (6) where Cat is the catalyst, R is the reagent, and P is the product. For a photocatalytic process, we need only involve the interaction of the system with light (Equation (7)): Cat + R + hν →Cat + P (7) (7) Cat + R + hν →Cat + P Cat + R + hν →Cat + P Certain similarities between catalysis and photocatalysis support such considerations, as both are characterized by the alteration of: (i) the reaction rate; and (ii) the reaction pathway; as well as (iii) by the essential role of adsorption of the reagent R and desorption of the product P in the (photo)catalytic cycle. In addition, the red-shift of the spectral limit of a photocatalytic process, in contrast to a catalyzed photochemical process, can be taken as an analog of the decrease of the activation energy required for the catalyzed reaction compared to the non-catalyzed reaction. Not surprisingly then, some terminology, major characteristics, and quantitative parameters typically used in catalysis have been adapted to the ﬁeld of photocatalysis. For example, the terms activity and selectivity of a photocatalyst, as well as the terms surface-active center, turnover number and turnover frequency are examples transferred from catalysis to photocatalysis. Within the present context, the most important feature in photocatalysis borrowed from catalysis is the notion of a surface-active center that can best be described as a surface regular site or a surface defect site capable of initiating chemical transformations of the reactants [29]. In heterogeneous photocatalysis, surface-active centers (sites) initiate a chemical sequence subsequent to the centers being activated by absorption of (light) photons by the photocatalyst. That is, before the reaction cycle, the surface-active centers (S) on the photocatalyst are initially in their inactive ground state unable to react with other molecules. However, once the solid photocatalyst has been electronically excited by absorption of photons with the appearance of free charge carriers, electrons (e−) and holes (h+), or excitons (e◦), this excitation energy may be localized on the surface-active centers to yield centers in their chemically active excited states (S+ and S−; Equation (8)), which can also form by interaction with the excitons (Equation (9)), or otherwise the photons may excite the 420 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 surface-active sites directly to give S* (Equation (10)). The surface-active centers in their electronically activated excited state are then capable of initiating surface chemical reactions. surface-active sites directly to give S* (Equation (10)). The surface-active centers in their electronically activated excited state are then capable of initiating surface chemical reactions. Cat + R + hν →Cat + P S + e−→S− (8a) S + h+ →S+ (8b) S + e◦→S−+ h+ (9a) S + e◦→S+ + e− (9b) S + hν →S* (10) (8a) (10) The photoactivated states of the surface-active centers can decay back to the ground state of the centers through different physical relaxation pathways: (a) by recombination of the charge carriers trapped at surface-active centers with free charge carriers of the opposite sign, or with free excitons (Equations (11) and (12)); (b) by thermal ionization when the activation energy of ionization is comparable to kT (Equation (13)); (c) photo-ionization in the spectral range of photexcitation corresponding to the absorption band of the surface-active centers (Equation (14)); and (d) by spontaneous deactivation (Equation (15)). S+ + e−→S (11a) S−+ h+ →S (11b) S+ + e◦→S + h+ (12a) S−+ e◦→S + e− (12b) S+ →S + h+ (13a) S−→S + e− (13b) S+ + hν →S + h+ (14a) S−+ hν →S + e− (14b) S* →S + hν’ (15) (15) Concurrently, chemical reactions of molecules with the photoactivated states of the surface-active centers (S) also lead to the return of the centers back to their ground state (S; Equation (16)). S + R →S + P (16) (16) In summary, there are two states of surface-active centers in heterogeneous photocatalysis: (i) the initial inactive ground state S; and (ii) the chemically active excited state(s) (S+, S−or S). Both physical relaxation and chemical interactions of these excited state(s) with molecules return the centers back to their initial ground state. 2.1.1. Is the Process Photocatalytic or Stoichiometric? 2.1.1. Is the Process Photocatalytic or Stoichiometric? The photocatalytic cycle begins by absorption of photons causing the activation of surface-active centers, followed by chemical reactions that terminate by desorption of product(s) from the active centers, and restoring the centers back to their initial ground state. This is essential in determining quantitatively the parameters borrowed from catalysis: namely, the turnover frequency (TOF) and the turnover number (TON) [29]. TOF refers to the number of photoinduced transformations (product 421 Catalysts 2018, 8, 553 formed or reactant consumed) per catalytic site per unit time as expressed by Equation (17); Na is the number of catalyst active sites, and N is the number of photocatalytic transformations. TOF = 1 Na dN dt (17) (17) It is important to recognize that TOF considers the number of surface-active centers in their initial ground state before photactivation. When the number of such active sites is unknown, the surface area is often used to normalize the number of turnovers—this is known as the areal turnover frequency [29]. In catalysis, TOF describes the activity of the catalytic centers. In photocatalysis, however, the rate of a chemical transformation of a molecule in a photocatalytic process depends on the light intensity, so that TOF is also light intensity-dependent and is generally taken as a characteristic feature of a photocatalyst. TOF can be used to compare the activities of various photocatalysts only if the photocatalysts were photoactivated under identical conditions. This is extremely important! Otherwise, any comparison of the activities of various photocatalysts will have no physical/chemical meaning. Another turnover quantity taken from conventional catalysis is TON that describes the number of times an overall reaction (the photochemical transformation) goes through a photocatalytic cycle for a given period, t (Equation (18)). Again, it is important to recognize that it is the number of photocatalytic centers in their ground state that is relevant when assessing TON in photocatalysis. TON = 1 Na t 0 dN dt dt = N Na (18) (18) Since the photocatalytic reaction rate, and thus the number of photoinduced chemical transformations, depends on light intensity, TON cannot be considered an intrinsic property of a photocatalyst. Nonetheless, TON is an important parameter in experimental studies of heterogeneous photochemical reactions as it shows whether the photochemical transformation is catalytic. 2.1.1. Is the Process Photocatalytic or Stoichiometric? Where TON > 1, the active center is restored back to its initial ground state after the chemical transformation cycle and is ready to initiate subsequent cycles: the photochemical process is then said to be photocatalytic. When TON ≤1 after the ﬁrst cycle, the active center is completely deactivated and cannot initiate subsequent cycles of the chemical transformations; this means that the surface-active center only interacts with one molecule. In this case, the process is a stoichiometric process, as encountered in photostimulated adsorptions of molecules on solid surfaces. If a stoichiometric surface reaction were to occur on the same surface-active center as a photocatalytic reaction, it would provide a method with which to account for the number of surface-active centers and thus aid in establishing the magnitude of TON. For instance, if a photostimulated adsorption of a molecule is the ﬁrst step in a complex photocatalytic process that concludes the photocatalytic cycle at the photoadsorption step, then it is possible to determine the number of photoadsorbed molecules that corresponds to the number of surface-active centers involved in the photocatalytic reaction. This method was used by Emeline and coworkers [30] to assess quantitatively TON for the photooxidation of hydrogen by oxygen and the photoreduction of oxygen in the presence of hydrogen over solid particles of ZrO2. The authors showed that the photostimulated adsorption of oxygen on zirconia, which by default is a stoichiometric reaction, is the ﬁrst step in the photooxidation of hydrogen. Therefore, the maximal number of photoadsorbed oxygen molecules gives the number of surface-active centers for the photooxidation of hydrogen. They determined that TON > 1 for this process, and consequently the process was deemed to be a photocatalytic process. 2.1.2. Mechanistic Implications Another common feature between catalytic and photocatalytic processes are the two general mechanisms of surface chemical reactions: (a) the Eley–Rideal mechanism (ER); and (b) the Langmuir–Hinshelwood (LH) mechanism. The Eley–Rideal mechanism assumes that molecules in the bulk solution phase interact with surface species (either pre-adsorbed molecules or surface-active 422 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 centers), whereas the Langmuir–Hinshelwood mechanism presupposes that the reaction involves pre-adsorbed molecules and surface species (either pre-adsorbed molecules or surface-active centers). For decades, the LH mechanism was the favorite mechanism among experimentalists (although erroneously) because of the experimentally observed dependence of the rate of the photocatalytic process on reagent concentration (Equation (19)). centers), whereas the Langmuir–Hinshelwood mechanism presupposes that the reaction involves pre-adsorbed molecules and surface species (either pre-adsorbed molecules or surface-active centers). For decades, the LH mechanism was the favorite mechanism among experimentalists (although erroneously) because of the experimentally observed dependence of the rate of the photocatalytic process on reagent concentration (Equation (19)). dC dt = kKC 1 + KC (19) (19) where k is the apparent rate constant, K is the apparent equilibrium constant, and C is the concentration of reactant molecules. This equation correlates with the Langmuir isotherm (Equation (20)): where k is the apparent rate constant, K is the apparent equilibrium constant, and C is the concentration of reactant molecules. This equation correlates with the Langmuir isotherm (Equation (20)): θ = KLC 1 + KLC (20) (20) where KL is the adsorption/desorption equilibrium constant (KL = kads/kdes) provided the rate of the process is proportional to the surface coverage (θ) by pre-adsorbed molecules (Equation (21)). That is, the adsorption/desorption equilibrium is not disturbed during the photoprocess. where KL is the adsorption/desorption equilibrium constant (KL = kads/kdes) provided the rate of the process is proportional to the surface coverage (θ) by pre-adsorbed molecules (Equation (21)). That is, the adsorption/desorption equilibrium is not disturbed during the photoprocess. dC dt = kθ (21) (21) If K = KL, then by default the observation that the rate dependence followed Equation (19) was taken as an indicator that the process followed the LH mechanism. However, for many heterogeneous photoprocesses, it was established conclusively that K ̸= KL and K depended on light intensity. 2.1.2. Mechanistic Implications Later, Emeline and coworkers [31,32] demonstrated that the dependence expressed by Equation (19) was also consistent with the ER mechanism, considering that the lifetime of the activated state of the surface-active centers is limited (see above). Moreover, if the physical relaxation pathway of the excited state of the active center involved recombination or photo-ionization, then K became dependent on light intensity in accord with experimental results (Figures 6 and 7). Figure 6. Dependence of the apparent constants k and KL on light intensity for the photostimulated adsorption of oxygen on ZrO2. Reproduced with permission from Emeline et al. [31]; Copyright 1998 by the American Chemical Society. Figure 6. Dependence of the apparent constants k and KL on light intensity for the photostimulated adsorption of oxygen on ZrO2. Reproduced with permission from Emeline et al. [31]; Copyright 1998 by the American Chemical Society. The approach based on a limited lifetime of the excited state of active centers is applicable to both ER and LH mechanisms, provided a quasi-steady-state is valid for the concentration of active centers 423 Catalysts 2018, 8, 553 in the activated state. The alternative view (i.e., LH mechanism) is also based on a quasi-steady-state approach for the concentration of adsorbed molecules and on a strong disruption of the adsorption equilibrium in accord with experimental data [33,34]. Kinetic measurements alone, as typically conducted in photocatalytic studies, cannot distinguish unambiguously which mechanism is operative in a photocatalytic process. Whatever the actual mechanism (ER or LH), however, the role of adsorption of a molecule on the surface is not insigniﬁcant and in fact is typical in heterogeneous photocatalysis. The only difference occurs in the ER mechanism, whereby photostimulated adsorption plays an important role rather than dark adsorption, with the latter so crucial to the LH mechanism. Figure 7. Dependence of the apparent constants k and KL on light intensity for the photodegradation of phenol over TiO2. Reproduced with permission from Emeline et al. [32]. Copyright 2000 by Elsevier B.V. (License No. 4452260016035). Figure 7. Dependence of the apparent constants k and KL on light intensity for the photodegradation of phenol over TiO2. Reproduced with permission from Emeline et al. [32]. Copyright 2000 by Elsevier B.V. (License No. 4452260016035). In summary, heterogeneous catalysis has had a non-insigniﬁcant impact on the ﬁeld of heterogeneous photocatalysis. Nonetheless, a serious discrepancy between these two phenomena remains. 2.1.2. Mechanistic Implications In catalysis, the catalyst is in its electronic ground state and is in thermodynamic equilibrium with the environment at a given temperature; it accelerates both direct and back reactions, and promotes a faster establishment of the reaction equilibrium. By contrast, in heterogeneous photocatalysis, the photocatalyst becomes active in an electronically excited state that is not in thermodynamic equilibrium with its surroundings, so that the sole relaxation pathway back to its ground state is acceleration of the direct reaction, because the photocatalytic heterogeneous system is an open system under a permanent ﬂow of light energy as occurs in photochemistry. Inﬂuence of Photochemistry on Photocatalysis—Comparisons and Contrasts 2.2. Inﬂuence of Photochemistry on Photocatalysis—Comparisons and Contrasts Photochemistry of heterogeneous systems is also an important pillar of photocatalysis, as it considers thermodynamically open systems. Thus, photocatalysis can also induce reactions that lead to light energy conversion and storage through formation of higher energy products, as occurs in natural photosynthesis [35,36]. Concepts taken from photochemistry suggest that the photocatalytic process occurs through electronically excited states of the photocatalyst prompted by light absorption. It should be noted that photocatalyzed reactions are often indistinguishable from photosensitized reactions examined in photochemistry [37]. Historically, many photocatalytic reactions were considered photosensitized reactions, so that what was once called a photosensitizer is now referred to as a photocatalyst. Two major characteristics of photosensitized reactions are also typical of photocatalytic processes: (i) a red-shift of the spectral limit of the photoreaction; and (ii) the alteration of the reaction pathway compared to that of the photochemical reaction. The red-shift of the spectral limit of a photoprocess and alteration of the reaction pathway are distinguishable ﬁngerprints of heterogeneous 424 Catalysts 2018, 8, 553 photocatalytic reactions. Enhancing the ability of photocatalysts to sensitize photochemical reactions, especially toward visible light, is a major challenge in applied heterogeneous photocatalysis. Photochemistry is concerned with the chemical effects of ultraviolet, visible, or infrared radiation while the photocatalyst is a substance that can produce on absorption of light chemical transformations of the reaction partners subsequent to formation of an excited state of the photocatalyst, which interacts repeatedly with the reaction partners forming reaction intermediates and regenerates itself after each cycle of such interactions [37]. Hence, a photocatalytic process (Equation (22)) is similar to a photochemical reaction (Equation (23)). Cat + hν →Cat + R (22) R + hν →R* →P (23) (22) R + hν →R* →P (23) where Cat and Cat* denote the photocatalyst in the ground state and in its electronically excited states, respectively; R is the reagent in its ground electronic state; R* is the reagent in the electronically excited state formed by absorption of photons; and P is the product of the photochemical/photocatalytic transformation. Thus, a photocatalytic process could be viewed as a particular case of a photochemical reaction, whereby the photocatalyst plays the double role as one of the reagents and as one of the products subsequent to restoring its initial ground state. 2.2.1. Photocatalytic Versus Photochemical Processes There are far more mechanistic similarities between photocatalytic and photochemical reactions than there are between photocatalyzed and catalyzed reactions. Both the photochemical reaction and the photocatalytic reaction require absorption of photons to form electronically excited states of one of the reaction partners, which cause distortions of the initial thermodynamic equilibrium and both reactions then occur through non-equilibrium states. This means that the subsequent sequence of molecular transformations can only proceed in one direction to restore the equilibrium state of the system and there is no path for a back reaction. Both photoreactions are characterized by the existence of physical relaxation pathways of the electronically excited states: radiative and non-radiative relaxation processes. For a solid photocatalyst, these relaxation processes are excitonic decay, free charge carrier recombination, and charge carrier trapping. In addition, all photocatalytic processes obey the general law of photochemistry: only absorbed photons initiate chemical transformations, and one absorbed photon results in only one elementary transformation. Accordingly, photocatalysis can be viewed as catalysis of a photochemical reaction. Consequently, three major characteristics used in photocatalysis have been adopted from photochemistry: (a) spectral sensitivity of a photoprocess; (b) photoactivity of a photocatalyst; and (c) selectivity of a photocatalyst. Spectral Sensitivity of a Photoprocess The spectral sensitivity of a photoprocess is described by the spectral range wherein photon absorption by the system initiates a chemical transformation; for a photocatalytic process this is the red spectral limit that corresponds to the lowest energy photons to initiate photocatalytic cycles. Note that the blue spectral limit of all photocatalysts expands into the vacuum UV region and so cannot be determined under typical conditions. We have often noted that the photocatalysis literature claims (albeit incorrectly) that photons with energy lower than bandgap of the photocatalyst cannot initiate chemical reactions, and so the red spectral limit of a photocatalytic process is taken as the bandgap energy. Theoretically, this is true only for ideal solids with no defect states, although even for an ideal solid structure there are always regular surface states (e.g., the Tamm and Shockley states) whose energies of photoexcitation could be lower than the bandgap energy. In practice, however, one deals with real solid photocatalysts whose structures possess various imperfections: for example, intrinsic point defects, impurity defects, dislocations, and defect surface states (e.g., add-atoms, corners, 425 Catalysts 2018, 8, 553 edges, steps etc.). Photoexcitation of such defect states results in a signiﬁcant shift of the red limit of the photocatalytic effect toward lower energy photons compared to the bandgap energy—for the case of TiO2, although the bandgap energies are 3.0 eV for the rutile polymorph and 3.2 eV for the anatase counterpart, the red limits for both oxidative and reductive pathways are 2.2 eV (λ = 560 nm), well below bandgap. edges, steps etc.). Photoexcitation of such defect states results in a signiﬁcant shift of the red limit of the photocatalytic effect toward lower energy photons compared to the bandgap energy—for the case of TiO2, although the bandgap energies are 3.0 eV for the rutile polymorph and 3.2 eV for the anatase counterpart, the red limits for both oxidative and reductive pathways are 2.2 eV (λ = 560 nm), well below bandgap. Accordingly, from a practical application of heterogeneous photocatalysis, the longer is the wavelength of the red spectral limit of the photocatalytic process, the larger is the fraction of sunlight that can be exploited for a higher overall activity of the photocatalytic system. Spectral Sensitivity of a Photoprocess Increasing the spectral sensitivity of a photocatalyst toward visible light is a major challenge in applied photocatalysis, currently being examined through modiﬁcation of pristine photocatalysts by metal and non-metal doping and by physical modiﬁcations to form intrinsic defects, which give rise to extended extrinsic absorption of visible light. A detailed assessment of this issue and corresponding problems have been described elsewhere [38–40]. Photoactivity of a Photocatalyst: The Quantum Yield Φ The photonic efﬁciency describes the ratio of the rate of the photoreaction 426 Catalysts 2018, 8, 553 measured at t = 0 (initial rates) to the rate of incident photons within a given wavelength range [29]. Consequently, these parameters provide an estimate of the overall photoactivity of the photocatalysts, but provide no information regarding light absorption, photoexcitation mechanisms, and speciﬁc details as to the efﬁciencies of excitation transfer to the surface or to the initiation of surface chemical transformation as they denote engineering efﬁciencies. We cannot overemphasize that the above parameters used in photocatalysis and borrowed from photochemistry characterize the efﬁciency of photocatalysts, unlike TON and TOF from catalysis which are light intensity-dependent. Photochemical parameters may also depend on the intensity of photoexcitation; however, experimental conditions can be controlled such as to obtain a linear dependence of the reaction rate on light intensity as established by Emeline and coworkers [30,31] to obtain the maximal photoefﬁciency of photocatalysts when all the quantum and photonic parameters become independent of light intensity. The quantum yield value (and other photochemical parameters) is essentially governed by the reaction rate of a heterogeneous photocatalytic reaction that depends on the stationary surface concentration of charge carriers (either electrons or holes) (Equations (28) and (29)). R + e−→P (28a) R + h+ →P (28b) d[R]/dt = k [R] [e] (29a) d[R]/dt = k [R] [h] (29b) R + e−→P (28a) R + h+ →P (28b) d[R]/dt = k [R] [e] (29a) d[R]/dt = k [R] [h] (29b) ( q ( ) ( )) R + e−→P (28a) R + h+ →P (28b) d[R]/dt = k [R] [e] (29a) d[R]/dt = k [R] [h] (29b) Selectivity of a Photocatalyst Selectivity of a Photocatalyst Selectivity of a Photocatalyst The selectivity of a photocatalyst is the ability of a photocatalyst to drive a photoprocess toward a certain reaction product. Emeline and coworkers [41–44] established two possible causes for the selectivity of a photocatalyst when the rate of the surface photochemical reaction is dictated by the surface concentration of photocarriers (electrons for reduction, holes for oxidation). The ﬁrst originates from the ratio between the concentrations of electrons and holes at the surface of a photoactive material, so that alteration of this ratio caused by various factors leads to changes in the ratio between the efﬁciencies of surface redox reactions for molecules displaying both electron-acceptor and electron-donor behavior. Photoactivity of a Photocatalyst: The Quantum Yield Φ Photoactivity of a Photocatalyst: The Quantum Yield Φ The photoactivity of photocatalysts describes the ability of a solid to transform the absorbed actinic light into a chemical sequence, for which the quantum yield (Φ) is its most relevant parameter. In heterogeneous photocatalysis, Φ is the ratio between the number of molecules that have reacted (Nm) to the number of photons actually absorbed (Nhν) at a given wavelength of the actinic light (Equation (24)), provided the photocatalytic process has reached a stationary state [29], otherwise Φ can be expressed by the differential form (Equation (25)): Φ = Nm Nhν (24) Φ = dNm dt dNhν dt (25) (24) (25) Typically, Equation (25) is used since the reaction rate, and thus Φ, depends on such parameters as time, temperature, wavelength of the actinic light, concentration of reagent, light intensity, and the solution pH (among others). Regrettably, the rate of absorption of photons and the number of absorbed photons remain inaccessible quantities in heterogeneous photocatalysis because of experimental limitations. Consequently, the photonic yield (ξ) was introduced and deﬁned in a manner similar to the quantum yield (Φ) with the main difference being that reference is made to the number of photons of a given wavelength of the actinic light (N◦hν) incident on the photoreactor, and not on the number of photons absorbed by the photocatalyst under stationary conditions, as expressed by Equation (26) or in differential form by Equation (27) [28]. N ξ = Nm No hν (26) ξ = dNm dt dNo hν dt (27) (26) (27) Both Φ and ξ are deﬁned as in photochemistry [37]; the former is more appropriate, however, since only absorbed photons can initiate interfacial chemical reactions, thus Φ is an intrinsic characteristic of a photocatalyst. Nonetheless, ξ may be a more practical parameter to assess experimentally as it depends on the number of incident photons and not on the number of photons actually absorbed by the photocatalyst; ξ is also an intrinsic characteristic of a photocatalyst. In the ﬁeld of Chemical Engineering, the parameters often used are quantum efﬁciency and photonic efﬁciency; the former describes the rate of a given photophysical or photochemical process divided by the total photon ﬂux absorbed [29]; it applies especially when using polychromatic radiation to activate the photocatalyst. Photoactivity of a Photocatalyst: The Quantum Yield Φ For example, if a given reaction product (Pi) were formed by a reduction pathway during the photostimulated surface reaction with reagent (R) interacting with both electrons (e) and holes (h), the surface selectivity toward the formation of the given product (SPi) would then be expressed by Equation (30): ke,i and kh,j are the rate constants for a given reaction pathway. A similar expression can be formulated for the products formed by the oxidation reaction pathway. SPi = d[Pi] dt d[R] dt = ke,i[e] ∑i ke,i[e] + ∑j kh,j[h] (30) (30) By default, it is assumed that the reaction rate does not depend on reagent concentration, which would correspond to saturation of the rate dependence on reagent concentration for the LH-like kinetics. Equation (30) clearly indicates that the selectivity of the photocatalyst is determined by the ratio between the surface concentrations of electrons and holes. The second origin of selectivity is related to the formation of surface-active sites of a different type, which may display different activity and selectivity. Different types of surface-active sites can be formed through photoexcitation of localized surface and sub-surface electronic states (intrinsic defects and/or dopant states), or through formation on different types of surface structures (e.g., dominating surface plane). This type of selectivity is clearly observed for molecules that demonstrate only electron-donor or electron-acceptor behavior, although it is typical of any surface photochemical reaction. As such, the selectivity of the surface toward formation of a given reaction product can be expressed by Equation (31), which is determined by the ratio of a given rate constant characteristic 427 Catalysts 2018, 8, 553 of a given sort of active site to the sum of rate constants corresponding to all possible types of active sites on the surface of the photocatalyst (here rate constants are apparent constants and implicitly include the surface concentrations of active sites). Obviously, this factor (ratio of rate constants) is also signiﬁcant for the selectivity in the earlier scenario (Equation (30)). of a given sort of active site to the sum of rate constants corresponding to all possible types of active sites on the surface of the photocatalyst (here rate constants are apparent constants and implicitly include the surface concentrations of active sites). Obviously, this factor (ratio of rate constants) is also signiﬁcant for the selectivity in the earlier scenario (Equation (30)). Photoactivity of a Photocatalyst: The Quantum Yield Φ SPi = d[Pi] dt d[R] dt = ke,i ∑i ke,i (31) (31) Unlike macro crystals with their clearly manifested crystal faces, the dominant surface planes in nanoparticles of photocatalysts are strongly affected by the method and conditions of their synthesis. Concomitantly, in the course of photoprocesses, various planes of photocatalyst particles may display different selectivities, which may be due to different atomic structures, or magnitude and distribution of surface charge, or to a dominant type of defects (e.g., surface-active centers, among others) [45–49]. Hence, altering the nanoparticle surface structure via formation of various dominating faces with strong selectivity toward either oxidation or reduction can alter the overall selectivity of the photocatalyst. 2.2.2. Correlation between Activity and Selectivity The correlation between activity and selectivity of a photocatalyst has been demonstrated by Emeline et al. [50] in the photodegradation of 4-chlorophenol taking place over irradiated TiO2 with formation of three primary intermediates: (i) benzoquinone formed by a reductive pathway; (ii) 4-chlorocatechol formed by an oxidative pathway; and (iii) hydroquinone produced by both oxidative and reductive pathways. Figure 8 shows that the selectivity toward formation of hydroquinone and the activities of the photocatalyst are strongly correlated (r = 0.984; p = 0.0004). U S $FWLYLW\&O3K2+55PD[ 6HOHFWLYLW\+4 U S $FWLYLW\&O3K2+ 55PD[ 6HOHFWLYLW\+4 $FWLYLW\&O3K2+55PD[ Figure 8. Correlation between activities of six tested TiO2 photocatalysts in the photodegradation of 4-chlorophenol and their selectivity toward formation of hydroquinone. Reproduced with permission from Emeline et al. [50]. Copyright 2011 by Elsevier B.V. (License No.: 4452260940331). According to the balance of charge, as expressed by Gerisher [51], a true (photo)catalytic process is characterized by the equality of the rates of consumption of electrons and consumption of holes in the overall reaction (Equation (32)). That is, d[e] dt = d[h] dt (32) (32) This charge balance is a required condition for the effective photocatalytic process, or else the deviation from catalytic equilibrium (Equation (32)) would result in the transformation of charge balance according to Equation (33) [38,52]. d[e] dt + d[F] dt = d[h] dt + d[V] dt (33) (33) 428 Catalysts 2018, 8, 553 where F and V denote, respectively, electrons and holes trapped in bulk defects (color centers), which accelerate bulk charge carrier recombination and decrease the activity of the photocatalyst. Since formation of hydroquinone consumes both electrons and holes, these reaction pathways create a favorable condition for the photodegradation of the phenol being truly photocatalytic by suppressing bulk recombination. Therefore, the higher is the selectivity of the photocatalyst surface toward formation of hydroquinone, the higher is the activity of the photocatalyst during the photodegradation. In general, this rule can be formulated thus: a higher activity of photocatalysts can be expected provided both reductive and oxidative pathways occur with equally high efﬁciency. 2.3. Inﬂuence of Molecular Spectroscopy on Heterogeneous Photocatalysis 2.3. Inﬂuence of Molecular Spectroscopy on Heterogeneous Photocatalysis Another major pillar of heterogeneous photocatalysis is molecular spectroscopy of adsorbed molecules. Most pioneering studies focused on understanding the changes in molecular structure induced by adsorption of molecules on solid surfaces [53–55], with light being an active factor in transforming these adsorbates. The effects of photodissociation and photodesorption of adsorbed molecules have been established together with the photoadsorption of simple molecules on dispersed semiconductors [55,56]. The photocatalytic cycle begins with the absorption of light quanta by the solid photocatalyst and culminates with the chemical transformations of surface-adsorbed molecules, ultimately evolving reaction products into either the gaseous or liquid phase. The role of the photocatalyst and corresponding photophysical events taking place in solids are often treated in a simplistic manner. The ensemble of particles that absorb photons is the light harvesting system, whereby the photocatalyst particle is both a sensitizer and the source of intermediates—i.e., photoelectrons and photoholes. In addition, the intrinsic, that is, the fundamental absorption of light by the solids is of primordial importance in photocatalysis; this was a reasonable approach in most studies that were oriented on mechanistic investigations of chemical reactions, or else on practical applications of heterogeneous photocatalysis. The complexity and variety of photophysical processes in solid photocatalysts, together with the interdependence between physical and chemical events at the microparticles’ and nanoparticles’ surfaces, must always be kept in mind, even in applied heterogeneous photocatalysis. Accordingly, next we describe some relevant events that precede surface chemical reactions on the photocatalyst particle (Figure 9). Figure 9. Schematic illustration of a photocatalytic semiconductor nanoparticulate system in converting light energy into a solar fuel (H2) from the water splitting process. Figure 9. Schematic illustration of a photocatalytic semiconductor nanoparticulate system in converting light energy into a solar fuel (H2) from the water splitting process. Figure 9. Schematic illustration of a photocatalytic semiconductor nanoparticulate system in converting light energy into a solar fuel (H2) from the water splitting process. 2.3.1. Absorption of Light by Solid Photocatalysts Absorption of electromagnetic radiation represents the transfer of energy from an electromagnetic ﬁeld to a material (the photocatalyst) or to a molecular entity [29]. 2.3. Inﬂuence of Molecular Spectroscopy on Heterogeneous Photocatalysis The transformation of light energy 429 Catalysts 2018, 8, 553 into the energy of electronic excitations (electrons, holes) of a solid photocatalyst is fundamental in heterogeneous photocatalysis, with these excitations being created as a result of absorption of photons in a solid during some time (lifetime of charge carriers) before relaxation (recombination of charge carriers), i.e., before transformation of the electronic energy into thermal energy. Electron–hole pairs in solids then transfer the energy within some sub-surface space, the depth of which depends on both the linear absorption coefﬁcient of the photocatalyst and the diffusion length of the photocarriers and, in some particular cases, on particle size. 2.3.2. Quantities that Describe Light Absorption in Heterogeneous Photocatalysis Equation (39) can also be obtained from Equation (40) which expresses the independence of the absorption coefﬁcient α from the light intensity, 430 Catalysts 2018, 8, 553 (40) dI = −α I(x) dx (40) dI = −α I(x) dx where dI is the fraction of light absorbed by a thin layer dx a distance × from the plate surface. The density of absorbed photons (Equation (41)) can be deduced from Equations (39) and (40). If the light beam of intensity I0 were given in units of photons per unit area per second, then the units of dI/dx are cm−3 s−1. where dI is the fraction of light absorbed by a thin layer dx a distance × from the plate surface. The density of absorbed photons (Equation (41)) can be deduced from Equations (39) and (40). If the light beam of intensity I0 were given in units of photons per unit area per second, then the units of dI/dx are cm−3 s−1. dI dx = αI0e−αx (41) (41) The absorption coefﬁcient α is a spectrally dependent parameter (α = α(λ)), whose magnitude varies in a wide range for different solids and can reach values up to ca. 106 cm−1 for the fundamental absorption band but can be very small for extrinsic absorption bands of solids. Where absorption of light is due to defects or to structural imperfections (i.e., in the extrinsic absorption region for a transparent solid), the linear absorption coefﬁcient α is proportional to the concentration of defects n (Equation (42)); σa is the absorption cross section of a defect. α = σa n (42) (42) α = σa n Equation (42) represents Beer’s law (for defects in solids); the law fails at very high defect concentrations (typically > 1018 cm−3) when the interaction between defects in the solid becomes signiﬁcant. Taking into account Equations (42) and (37)–(39), a comparison with corresponding formulas in UV-Vis absorption spectroscopy of solvents indicates that the wavelength-dependent quantity σa becomes the extinction coefﬁcient that characterizes the optical properties of a given defect in the solid. The absorption cross-section, σa, for optical transitions in defects that are allowed by appropriate selection rules can reach values of ~ 10−16 cm2. Consequently, α ≈102 cm−1 for n ≈1017–1018 cm−3; this is typical of imperfect photocatalyst particles. 2.3.2. Quantities that Describe Light Absorption in Heterogeneous Photocatalysis Dispersed (powdered) photocatalysts in contact with gaseous or liquid phases in irregular shaped reactors are typically the light absorbing media, and so require consideration of the absorption and scattering of light together with a few quantities that we now outline for the absorption of light used in experimental photocatalytic reactions. Absorption of light is manifested by the decrease of the energy of the light beam when it passes through a medium because of the transformation of the photon’s energy into another energy form. Experience has taught us that IR + IT < I0 for light absorbing substances and so in accord with the energy conservation law, the intensities of the incident (I0), transmitted (IT) and specular reﬂected (IR) beams are given by Equation (34), I0 = IR + IT + IA (34) (34) where IA is the intensity of the absorbed light beam. For moderate intensities of the incident light beam and in the absence of additional input of energy to the substance, we have (Equation (35)) IR = R I0 (35a) IT = T I0 (35b) IA = A I0 (35c) The dimensionless coefﬁcients R (here it denotes the reﬂectance), T (transmittance) and A (absorbance) in Equations (34) and (35a)–(35c) satisfy Equation (36), R + T + A = 1 (36) R + T + A = 1 (36) Typically, absorbance can be determined optically using a dual-beam spectrophotometric technique such that the ratio IT/I0 is detected, while reﬂectance is compensated using a reference sample. From experience, we have that (Equation (37)) IT = (1 −R)I0e−αd (37) (37) And, from Equations (35a)–(35c) and (37), we obtain for A (Equation (38)) that A = (1 −R) 1 −e−αd (38) (38) which describes the Lambert–Bugger law for absorbance by a solid parallel plate of thickness d; the quantity α is the linear (Naperian) absorption coefﬁcient (in cm−1). In general (Equation (39)), which describes the Lambert–Bugger law for absorbance by a solid parallel plate of thickness d; the quantity α is the linear (Naperian) absorption coefﬁcient (in cm−1). In general (Equation (39)), I(x) = (1 −R)I0e−αx (39) (39) where I(x) is the intensity of the light beam a distance × from the illuminated surface of the optically uniform plate. 2.3.2. Quantities that Describe Light Absorption in Heterogeneous Photocatalysis When the nominal concentration of regular entities (atoms, ions) in solids (n ≈1022 cm−3) is multiplied by σa ≈10 −16 cm2 leads to a realistic estimation of the absorption coefﬁcient (α ≈106 cm−1) for fundamental absorption bands, despite breaching the conditions of Beer’s law validity. Variation of α with wavelength of the actinic light is an important factor in the activity of solid photocatalysts and other spectrally dependent parameters. 2.3.3. Absorbance and Reﬂectance of Photocatalysts in Powdered Form 2.3.3. Absorbance and Reﬂectance of Photocatalysts in Powdered Form Powdered solids with sufﬁciently high BET surface areas (SBET = 10–100 m2 g−1) are typically used in photocatalytic studies and correspond to characteristic mean particle sizes in the microscale to nanoscale comparable to the wavelength of the actinic light. Contrary to the optically uniform plate noted earlier, light scattering now plays a major role in the action of light on powdered photocatalysts. Because of light scattering, a powdered photocatalyst in pellet form is now opaque to light compared with the optically uniform plate of the same material, even in the spectral region where absorbance is very low (extrinsic absorption region). Consequently, most of the light is reﬂected back in various directions and gives rise to diffuse reﬂectance. The light intensity decreases with distance from the illuminated side of the pellet. In this case, the Lambert–Bugger law is not valid in its original form (Equation (6)) so that in practice the energy conservation law (Equation (34)) should be applied to powdered materials with some restraint since care must be taken in measuring the diffuse reﬂected light. Diffuse reﬂectance spectroscopy has proven a useful method in experimental measurements of the optical absorption of powdered metal-oxide photocatalysts. For instance, the absorbance A of a sample and the diffuse reﬂectance coefﬁcient, R, are related by Equation (43) for zero transmittance (i.e., for T = 0); in practice, this is valid when the thickness of the powdered sample is around 3–5 mm. A = 1 −R (43) 431 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 UV-Vis-NIR diffuse reﬂectance spectra, R(λ), are usually recorded with spectrophotometers equipped with an integrating sphere assembly and a standard reference sample of known diffuse reﬂectance spectrum. Accordingly, R(λ) or A(λ) spectra provide the spectral information about the initial state of a powdered sample. Additional spectral information can be obtained from difference diffuse reﬂectance spectra, ΔR(λ), when changes in reﬂection/absorption occur from controllable treatments of the photocatalyst (e.g., annealing or illumination in a vacuum or in the presence of gases) (Equation (44)). ΔR(λ) = R1(λ) −R2(λ) (44) (44) where R1(λ) and R2(λ) are the diffuse reﬂectance spectra measured, respectively, before and after the treatment. The case where ΔR(λ) > 0 corresponds to the treatment-induced absorption spectrum. Hence, from Equation (43) we obtain (Equation (45)): where R1(λ) and R2(λ) are the diffuse reﬂectance spectra measured, respectively, before and after the treatment. 2.3.3. Absorbance and Reﬂectance of Photocatalysts in Powdered Form The case where ΔR(λ) > 0 corresponds to the treatment-induced absorption spectrum. Hence, from Equation (43) we obtain (Equation (45)): ΔR(λ) = ΔA(λ) = A2 −A1 (45) (45) Thus, alteration of the diffuse reﬂectance coefﬁcient in photocatalytic experiments corresponds to changes in the absorbance of the photocatalyst sample, which for practical applications is more important than changes in the absorption coefﬁcient (such as quantum yield measurements) as it provides the information about the spectral behavior of the whole heterogeneous system. Difference diffuse reﬂectance spectra (ΔR(λ)) not only increase signiﬁcantly the accuracy of measurements but make information accessible with regard to absorption spectral shapes induced by whatever treatment the photocatalyst is subjected to. Hence, measurements of a set of induced absorption spectra of different spectral shapes allow for a numerical analysis of the spectral data that could be very important for absorption spectra consisting of several overlapping single absorption bands. 2.4. Intrinsic and Extrinsic Absorption of Light by Semiconductor Photocatalysts 3.1. Intrinsic Structural Point Defects in Metal Oxides Point defects related to anion (oxygen) and cation (metal) vacancies in sub-lattices are the main types of defects in metal-oxide semiconductors/insulators. Structure, together with the optical and EPR properties of families of both oxygen vacancies (Vo) and cation vacancies (Vm) are now fairly understood for wide bandgap metal oxides such as MgO, Al2O3, and ZrO2 (among others) [59–67]. 2.4. Intrinsic and Extrinsic Absorption of Light by Semiconductor Photocatalysts Two different spectral regions of light absorption can be distinguished in the ﬁeld of optics and in the photophysics of the solid state: (i) intrinsic or fundamental absorption; and (ii) extrinsic absorption of light. Intrinsic light absorption is due to photoinduced electronic transitions between occupied delocalized states in the valence band and unoccupied delocalized states in the conduction band of semiconductor photocatalysts [57,58]. The intensity of the inter-band photoexcited transitions, characterized by the coefﬁcients α and σa, is determined by selection rules similar to those for atomic photoexcitation and by the additional requirement for a quasi-momentum conservation. As a case in point, photoinduced electronic transitions in TiO2 from the top of the valence band (mostly oxygen p-states) to the bottom of the conduction band (mostly titanium d-states) are allowed transitions (p →d transitions) as are the p →s transitions in alkali halides. The intensities of photoinduced transitions from the VB to the CB in alkali halides are much greater than those observed in TiO2. The reason for this difference in solids may be found in the momentum conservation law for electrons that requires the fulﬁllment of the condition expressed by Equation (46): (46) k1 −k2 = khν (46) where k1 and k2 are the wave vectors of electrons in the states between which the transition occurs and khν is the wave vector of the photon which, when absorbed by the solid, causes the transition. To the extent that the momentum of photons resulting in the electronic transition in a typical spectral range of photoexcitation of a solid is much smaller than the quasi-momentum of electrons occupying the band states, the condition expressed by Equation (44) can be simpliﬁed to Equation (47) with the proviso that khν ≈0. Δk = k1 −k2 ≈0 (47) Δk = k1 −k2 ≈0 (47) 432 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 The transition between states in the VB and CB bands that follows Equation (46) is referred to as a direct transition (i.e., the transition preserves the momentum of the electron) and is characterized by high intensity at the edge of the fundamental absorption. Where the positions of the maxima do not coincide in k-space (typical of TiO2), direct transition between these states is then forbidden because momentum conservation is not preserved. Nevertheless, photoinduced electronic transitions between such states can occur via involvement of phonons that possess the required momentum. 2.4. Intrinsic and Extrinsic Absorption of Light by Semiconductor Photocatalysts Accordingly, (48) Δk ± khω ≈0 (48) where khω is the wave vector of either the absorbed or the emitted phonon. Such transition is referred to as an indirect transition characterized by lower intensity as it requires a three-body interaction of the electron, the photon and the phonon. Extrinsic light absorption in solids may originate from the photoexcitation of such defect states as: (i) zero-dimensional (0-D) intrinsic point defects (e.g., vacancies, interstitials) and impurity atoms/ions; (ii) one-dimensional (1-D) linear defects (e.g., dislocations); and (iii) two-dimensional (2-D) states (e.g., intrinsic surface states). Special interest in heterogeneous photocatalysis concerns the surface absorption of light associated with the electronic excitation of surface states, for which the energy of absorption corresponds to the extrinsic spectral region of the solids. In this regard, the greater the speciﬁc surface area is, the greater is the impact of surface absorption into extrinsic absorption. The important issue here is that the generation of charge carriers that may induce surface chemical reactions occurs only at the surface of the solid photocatalyst. To recap, the various absorption bands corresponding to photoexcitation of different types of defects, impurities, and surface states can cover a wide spectral range of extrinsic light absorption. This photoexcitation generates free charge carriers or surface-localized excited states that can initiate interfacial chemical processes. Accordingly, the red spectral limit of a heterogeneous photochemical or photocatalytic reaction corresponds to photon energy much less than the bandgap energy of the solid photocatalysts. The photoactivity in the extrinsic absorption spectral region is typical of most metal-oxide photocatalysts, including TiO2. 3. Photophysical Processes in Metal-Oxide Photocatalysts 3. Photophysical Processes in Metal-Oxide Photocatalysts 3.1. Intrinsic Structural Point Defects in Metal Oxides 3.1.1. Defects Related to Oxygen Vacancies (Vo) The optical absorption by F+ and F centers and electron processes subsequent to absorption of light is of some importance in metal oxides, because the energy of a photostimulated electron transition from the ground state to the exited state and the position of the exited state within the bandgap relative to the bottom of the CB band determine the formation of free electrons in the CB band. The optical absorption by F+ and F centers and electron processes subsequent to absorption of light is of some importance in metal oxides, because the energy of a photostimulated electron transition from the ground state to the exited state and the position of the exited state within the bandgap relative to the bottom of the CB band determine the formation of free electrons in the CB band. An explanation of the formation of free charge carriers by photoexcitation of F or F+ centers had been proposed in two earlier studies by Kuznetsov and coworkers [68,69]; it is illustrated schematically in Figure 11. It was implied that the ﬁrst photophysical event is the optically- activated electronic transition from the ground state F (or F+) center to its excited state (F)* (or (F+)). Free electrons in the conduction band then result from a thermally-activated electron transition from the F (or (F+)) state to the conduction band, while the free holes in the valence band appear through a thermally-activated capture of an electron from the valence band by the lower level of the excited F (or (F+)) center. After the photoformation of electron–hole pairs through such a scheme, the F (or (F+)) center returns to its initial ground state that ensures the stability of these processes during the photoexcitation events. The thermal energy needed for the occurrence of such processes is about 0.1 eV for the 2.95 eV absorption band and ~ 0.5 eV for the 2.56 eV band that signiﬁcantly exceeded the energy of phonons at the temperatures employed (bandgap energy Eg of VLA rutile was 3.05 eV at 90 K and 3.01 eV at 290 K). A recent study based on experimental results and literature data further led Kuznetsov and coworkers [70] to hypothesize that, following the prime optical excitation of defects in the solid (Ti3+ centers), the heat released during the non-radiative electron transitions dissipated into the nearest neighborhood of these centers with consequences (see Section 5). 3.1.1. Defects Related to Oxygen Vacancies (Vo) There are three main types of defects in the oxygen sub-lattice of MgO related to oxygen vacancies, the so-called F-type centers: (i) doubly charged (with respect to the lattice) and optically silent anion oxygen vacancies (F2+ centers); (ii) anion vacancies with one trapped electron (F+ centers); and (iii) two trapped electrons (F centers) [59–62]. A main feature of F+ and F centers is a strong localization of the electron(s) within the vacancy. Semi-empirical quantum chemical calculations based on the INDO approach [61] have shown that the optimized geometry of oxygen vacancies with both bare and trapped electrons is characterized by a displacement of the nearest-neighbor cations in a direction away from the vacancy. Atomic relaxation is largest for F2+ centers (6.5% of regular Mg–O distance), whereas cation displacement for neutral F centers is <2% [61]. Calculations of ionic MgO crystals have shown that electrons trapped by Vo are indeed localized in the vacancy, and that the effective charge of F+ and F centers in the ground state was −1.002 and −2.002, respectively, while the changes in 433 Catalysts 2018, 8, 553 the effective charges of Mg and O ions surrounding Vo were inconsequential with respect to regular ones [61]. In the wurtzite ZnO lattice, the value and direction of the relaxations of Zn atoms adjacent to an oxygen vacancy strongly depend on the charge state of the defect. For the F center (Voo), the Zn atoms are displaced inward by approximately 11–12%, while, for the F2+ center (Vo2+), an outward relaxation of about 19–23% occurs (Figure 10) [64]. Figure 10. Ball and stick representation of the local atomic relaxations around the oxygen vacancy in ZnO in the (0), (1+), and (2+) charged states. Reproduced with permission from Seebauer and Kratzer [64]. Copyright 2007 by Elsevier B.V. (License No.: 4452270159193). Figure 10. Ball and stick representation of the local atomic relaxations around the oxygen vacancy in ZnO in the (0), (1+), and (2+) charged states. Reproduced with permission from Seebauer and Kratzer [64]. Copyright 2007 by Elsevier B.V. (License No.: 4452270159193). 3.1.1. Defects Related to Oxygen Vacancies (Vo) Localized non-equilibrated excitation of the phonon subsystem was equivalent to energies up to 1 eV. Experimentally determined positions of the absorption band maxima of F and F+ centers in several other metal oxides have been reported [60,61,71,72]. In accord with experimental results, theoretical calculations placed the ground state levels of both F and F+ centers in MgO (bandgap, 7.8 eV) at 3 eV above the top of the VB band [61]; this means that the exited states of these centers are near the bottom of the CB band (experimentally, ca. 0.06 eV below CB; see Ref. [61] and references therein). In the case of TiO2, the F centers have been located at 2.9 and 2.55 eV [61]. The ground state of the F center in α-Al2O3 is at 5.3 eV from the VB band, and thus the excited state F lies within the conduction band manifold [61]. The ground state of the F+ center is distant from the VB band by 3.1 eV so that the exited state (F+), corresponding to the absorption band at 6.3 eV, lies within the CB band. Indeed, irradiation at 6.1 eV within the spectral range corresponding to the F+ center’s absorption band produces photoconductivity down to at least 10 K (see Ref. [63] and references therein). For other F+-center’s exited states corresponding to the 5.4 eV and 4.8 eV 434 Catalysts 2018, 8, 553 absorption bands are 1.0 eV and 1.7 eV distant, respectively, from the CB band. Thus, UV irradiation in the absorption bands of F and F+ centers in MgO or F centers in α-Al2O3 result in the photo-ionization of the defects and ultimately initiation of surface reactions in the absence of any signiﬁcant number of bulk electron traps. Figure 11. Illustration of the relative positions of the energy levels of: (a) F◦centers; and (b) hole traps and electron traps (e.g., Ti3+ centers) within the band gap of VLA TiO2 rutile. The scheme represents a simplistic description of the formation of charge carriers via the intrinsic absorption at 3.05 eV and absorption in the band at 2.33 eV. For simplicity, only three electron and three-hole traps are shown in (b). The blue arrows in (b) indicate the thermostimulated release of photoholes, while the red arrows indicate the recombination of holes in the valence band with electron Ti3+ centers; Reproduced with permission from Kuznetsov et al. [68]. 3.1.1. Defects Related to Oxygen Vacancies (Vo) Copyright 2014 by the American Chemical Society. Figure 11. Illustration of the relative positions of the energy levels of: (a) F◦centers; and (b) hole traps and electron traps (e.g., Ti3+ centers) within the band gap of VLA TiO2 rutile. The scheme represents a simplistic description of the formation of charge carriers via the intrinsic absorption at 3.05 eV and absorption in the band at 2.33 eV. For simplicity, only three electron and three-hole traps are shown in (b). The blue arrows in (b) indicate the thermostimulated release of photoholes, while the red arrows indicate the recombination of holes in the valence band with electron Ti3+ centers; Reproduced with permission from Kuznetsov et al. [68]. Copyright 2014 by the American Chemical Society. Absorption spectra of strongly defective crystals of MgO and Al2O3 revealed additional absorption bands that were ascribed to aggregates of oxygen vacancies [61,73–76], the main types being F2 centers (i.e., 2 Vos with 4 trapped electrons), and F2+ and F22+ centers (two Vos with three and two trapped electrons, respectively). Under irradiation with fast neutrons (E > 1.2 MeV) [75] or bombardment by Cu+ or Ti+ ions (E = 30 KeV) [74], F and F+ centers were shown to be the dominating defects in the oxygen sub-lattice. It is evident that the family of intrinsic defects related to oxygen vacancies yields in toto broad absorption spectra covering the UV, visible and near-IR spectral regions. A full understanding of the pathways of photoactivation of wide bandgap metal oxides requires a detailed examination of the photoconversion F+ →F. In his 1982 review article, Crawford called attention to a puzzling feature of the behavior of F+ centers in Al2O3 when irradiated into the 4.8 eV and 5.4 eV absorption bands, which decreased the number of F+ centers and increased the number of F centers (absorption band at 6.0 eV) [63]. Photoconversions F+ →F were also found in MgO crystals and were accompanied by formation of hole centers [61–63], which excludes thermal ionization of the excited (F+) center followed by further electron trapping by another F+ center. Hole formation was connected with the spontaneous electron capture from an O2−adjacent to the (F+)-center’s empty ground state [63]. Semi-empirical calculations have also offered explanations for the F+ → F photoconversions [61,62]. 3.1.1. Defects Related to Oxygen Vacancies (Vo) Theory predicted the existence of several quasi-local energy levels of the F+ center in the upper part of the VB band, which consist mainly of atomic orbitals of O2−ions surrounding the F+ center. Under optical excitation, electron transition from these quasi-local levels to 435 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 the ground state of the F+ center yielded an F center and a hole in the VB band [61,62]; ultimately, the hole was trapped either by a cation vacancy Vm or by some impurity. the ground state of the F+ center yielded an F center and a hole in the VB band [61,62]; ultimately, the hole was trapped either by a cation vacancy Vm or by some impurity. Predictable schemes of the photogeneration of electrons and holes on excitation of F-type centers in wide bandgap oxides can be summarized by Equations (49)–(51c). F + hν →F →F+ + e− cb (49) F+ + hν →F+* →F2+ + e− cb (50) F+ + hν →F+* (51a) F+* + (O2−)vb →F + (O−)vb (51b) (F+)ql + hν + (O2−)vb →F + (O−)vb (51c) F + hν →F* →F+ + e− cb (49) F+ + hν →F+* →F2+ + e− cb (50) F+ + hν →F+* (51a) F+* + (O2−)vb →F + (O−)vb (51b) (F+)ql + hν + (O2−)vb →F + (O−)vb (51c) (49) Equations (49) and (50) represent cases of the photoionization of defect centers with formation of free electrons in the CB band. Equation (51b) corresponds to electron capture from an O2−ion adjacent to the F+-center, whereas Equation (51c) represents electron excitation from quasi-local levels of an F+-center (i.e., F+ql) in the VB band accompanied by formation of an F center. q Emeline and coworkers [77] reported a single maximum at 4.6 eV in the spectral dependence of the quantum yield of photoadsorption of O2 on powdered MgO, whereas Kuznetsov et al. [78] and Lisachenko [79] reported the maximum to occur at 5.0 eV. Spectral dependences of Φ of photo-adsorption of H2 and CH4 on powdered MgO displayed maxima at 4.6–4.55 eV [77] and a well-resolved shoulder at 3.75 eV. The photogeneration of holes on excitation within the spectral range, corresponding to the absorption band of F+ centers with maximum at 4.9 eV, was likely responsible for the photoadsorption; however, the spectral feature at 3.75 eV found no counterpart in the absorption spectra of F+ and F centers. 3.1.1. Defects Related to Oxygen Vacancies (Vo) Only the absorption band of an F2 center at 3.63 eV was near the feature at 3.75 eV, but nothing is known about the processes that followed the photon absorption by this center. Note that hole trap point defects related to cation vacancies in MgO absorbed light in the region 2.2–2.35 eV [67], far from the spectral feature just described. Spectral efﬁciencies of the photoadsorption of O2 and the photodissociation of N2O on Al2O3 displayed a maximum at 5.6–5.4 eV and a shoulder at ~ 4.8 eV [79], in good agreement with the absorption bands of F+ centers at 5.4 and 4.8 eV; as noted earlier, however, direct photoionization of these centers in Al2O3 in this spectral region appeared unlikely. Although optical properties of F-type centers in wide bandgap metal oxides have been studied sufﬁciently both theoretically and experimentally, interpretation of the data on the spectral response subsequent to irradiation, even for simple photoreactions such as photoadsorption of O2 and H2, has encountered some problems. Indeed, the spectral dependencies of the quantum yield of O2 and H2 photoadsorption in the visible and near-bandgap spectral regions for powdered TiO2 reported by Cherkashin’s group [80,81], by Emeline and coworkers [77,82], and by Komaguchi et al. [83] have been digitized; results are illustrated in Figures 12 and 13. The majority of the action spectra appear as a sum of two bands, the main band occurring at 3.0–2.9 eV with a half-bandwidth of 0.15–0.3 eV, independent of the type of gas used (O2, H2 or CH4) and irrespective of the crystalline structure of TiO2 (rutile or anatase). Spectra 2 and 3 in Figure 12 and 3 and 4 in Figure 13 display an additional band (seen as a shoulder) with maximum at ca. 2.5–2.7 eV. The action spectra of the photogeneration of paramagnetic O2−• (Figure 12, Curve 4) and [O–O2] species (Figure 13, Curve 5) show broad spectra with a single maximum around 2.6–2.8 eV. The action spectra of the photogeneration of electrons (photoadsorption of O2 or photoformation of O2−•) or of the photogeneration of holes (photoadsorption of H2 and CH4 or the photoformation of [O−O2]) originate from excitation of intrinsic defects, which have a biographical origin because all samples examined were nominally pure (undoped). 3.1.1. Defects Related to Oxygen Vacancies (Vo) Note that the rutile specimen displaying the broad action spectrum (Figure 13, Curve 5) had been strongly reduced by a H2 heat treatment at 773 K; 436 Catalysts 2018, 8, 553 it showed an absorption spectrum with a well-resolved shoulder at 2.7 eV and a broad absorption in the near-IR region (after bleaching the initial absorption in the presence of O2) [83]. Figure 12. Spectral dependencies of the quantum yields of photostimulated adsorption of O2 on the surface of TiO2 specimens: Degussa P25 pre-treated at 600 K (Curve 1) and at 850 K (Curve 1a) [82], rutile (Aldrich) pretreated at 600 K (Curve 2) [82], rutile pretreated at 820 K (Curve 3) [80]. Spectral dependence of the photogeneration of O2−• species at 77 K on the surface of rutile reduced by a H2 heat treatment at 773 K (Curve 4) [81]. Figure 12. Spectral dependencies of the quantum yields of photostimulated adsorption of O2 on the surface of TiO2 specimens: Degussa P25 pre-treated at 600 K (Curve 1) and at 850 K (Curve 1a) [82], rutile (Aldrich) pretreated at 600 K (Curve 2) [82], rutile pretreated at 820 K (Curve 3) [80]. Spectral dependence of the photogeneration of O2−• species at 77 K on the surface of rutile reduced by a H2 heat treatment at 773 K (Curve 4) [81]. Figure 12. Spectral dependencies of the quantum yields of photostimulated adsorption of O2 on the surface of TiO2 specimens: Degussa P25 pre-treated at 600 K (Curve 1) and at 850 K (Curve 1a) [82], rutile (Aldrich) pretreated at 600 K (Curve 2) [82], rutile pretreated at 820 K (Curve 3) [80]. Spectral dependence of the photogeneration of O2−• species at 77 K on the surface of rutile reduced by a H2 heat treatment at 773 K (Curve 4) [81]. Figure 13. Spectral dependencies of the quantum yields of photostimulated adsorption of H2 (Curves 1, 1a, 2, and 3) and CH4 (Curve 4) on the surface of TiO2 specimens: Degussa P25 pre-treated at 600 K (Curve 1) and at 850 K (Curve 1a) [82], rutile (Aldrich) pre-treated at 600 K (Curve 2) [82], rutile (Curves 3 and 4) [77]. Spectral dependence of the efﬁciency of photogeneration of a hole species [O–O2] at 77 K on the surface of anatase pretreated at 800 K (Curve 5) [81]. Figure 13. 3.1.1. Defects Related to Oxygen Vacancies (Vo) Spectral dependencies of the quantum yields of photostimulated adsorption of H2 (Curves 1, 1a, 2, and 3) and CH4 (Curve 4) on the surface of TiO2 specimens: Degussa P25 pre-treated at 600 K (Curve 1) and at 850 K (Curve 1a) [82], rutile (Aldrich) pre-treated at 600 K (Curve 2) [82], rutile (Curves 3 and 4) [77]. Spectral dependence of the efﬁciency of photogeneration of a hole species [O–O2] at 77 K on the surface of anatase pretreated at 800 K (Curve 5) [81]. Four principal reasons led Kuznetsov and Serpone [84] to model the photoactivation of (modiﬁed) TiO2 in the visible region: (1) the spectral position of the bands constituting the action spectra (i.e., bands at 3.0–2.9 and 2.5–2.7 eV); (2) the coincidence of the bands for the photo- generation of electrons and holes with the latter allowing the authors to propose that photoexcitation of intrinsic defects leads to the simultaneous generation of electrons and holes (electron–hole pairs); (3) the spectral position of the absorption bands at 2.95–2.75 eV and at 2.55–2.50 eV in the spectra of reduced TiO2 attributed to F centers [84]; and (4) the known regularities of photoexcitation of F type centers and consequent physical processes. In this regard, the authors [84] proposed a simple reasonable mechanism to interpret the experimental results reported in Figures 12 and 13. In accord with their study of TiO2 specimens, 437 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 the ﬁrst photophysical event of photoinduced absorption and photoreactions was light absorption by the F center that led to the formation of the corresponding excited state F (Equation (52)). the ﬁrst photophysical event of photoinduced absorption and photoreactions was light absorption by the F center that led to the formation of the corresponding excited state F* (Equation (52)). 3.1.1. Defects Related to Oxygen Vacancies (Vo) F + hνAB →F* (52) F* →F+ + e− cb (53a) F* + Δ →F+ + e− cb (53b) F+ + (O2−)vb →F + (O−)vb (54) F* + (O2−)vb →F + (O−)vb + e− cb (55) F + hνAB →F* (52) F* →F+ + e− cb (53a) F* + Δ →F+ + e− cb (53b) F+ + (O2−)vb →F + (O−)vb (54) F* + (O2−)vb →F + (O−)vb + e− cb (55) (52) F* →F+ + e− cb If the ground state of the F center lay within the bandgap near the VB band, then F* should lie within or close to the CB band. The energy difference (Ebg – hνAB) was only about 0.2–0.3 eV between the Ebg of the rutile/anatase TiO2 and the AB1 band, and 0.55–0.60 eV between the Ebg and the AB2 band. The excited F* center transforms spontaneously into the F+ center and a conduction band electron (Equation (53a)), or else it does so through thermal stimulation (Equation (53b)). Formation of a hole supposes the capture of an electron by the F+ center from the VB band, i.e., from the O2−ion (Equation (54)). Equation (55) describes the synchronous formation of an electron–hole pair. Hence, Equations (52)–(55) infer the photogeneration of both CB band electrons and VB band holes under visible light irradiation, and recovery of the centers of light absorption during irradiation. 3.1.2. Defects Related to Cation Vacancies (Vm) The main type of defects in the cation sub-lattice of metal oxides are defects related to cation vacancies, Vm, which have one or two holes trapped, i.e., V−or V◦centers, respectively. The atomic and electronic structure, the mechanism of optical excitation, and thus optical properties of hole V-type centers are deﬁned by hole localization (trapping) at one of several equivalent oxygen ions surrounding the metal vacancy. The term bound small polaron also applies to emphasize the interconnection of carrier (hole) trapping at a single site and the stabilization of the hole at this site by a lattice distortion. INDO calculations [61] showed that, in the V−center of MgO, the O−ion is closer to the cation vacancy by 3% of the Mg–O distance in the perfect crystal. In the V◦center, the two O−ions that can capture two holes are located on the opposite side of the cation vacancy; they are shifted inward toward the vacancy by 2% of the regular Mg–O distance. It is clear from the brief description of optical properties of photocatalytically active metal oxides that, in addition to the photoexcitation of solids in the fundamental absorption bands, there are many other pathways for the generation of electrons and holes. Concurrently, the same defects can play either negative or positive roles with respect to heterogeneous photocatalysis. For instance, they can act as recombination centers that reduce the concentration of photocarriers. However, being localized at the surface of photocatalyst particles, the defects are more likely to play the role of surface-active centers in photocatalytic processes. 3.1.4. Trapping of Carriers by Defects Carrier trapping by a defect is a good example of a perfect inelastic collision of carriers with the defect. The effectiveness of carrier trapping in solids is characterized by the phenomenological trapping cross-section, σtr (Equation (56)). dN dt = σtrvn (56) (56) where dN/dt is the number of carriers trapped by the defect per unit time; v is the mean velocity of the carriers (v ≈107 cm s−1); and n is the concentration of thermal carriers (or stationary concentration of photocarriers). The quantity dN/dt has dimensions and notion of frequency; its inverse can be treated as the lifetime of a defect (unoccupied by a carrier) relative to a trapped carrier. where dN/dt is the number of carriers trapped by the defect per unit time; v is the mean velocity of the carriers (v ≈107 cm s−1); and n is the concentration of thermal carriers (or stationary concentration of photocarriers). The quantity dN/dt has dimensions and notion of frequency; its inverse can be treated as the lifetime of a defect (unoccupied by a carrier) relative to a trapped carrier. Trapping cross-sections are determined by the properties of the defects; they vary within a rather wide range from 10−21–10−20 cm2 (trapping is energetically favorable but an activation barrier exists) p to 10−11–10−10 cm2 (strongly attractive Coulomb center; scattering of excess energy is effective). Typical values of σtr lie in the range between 10−16 and 10−15 cm2 for neutral defects. 3.1.5. Stationary Concentration of Photocarriers and Band-to-Band Recombination Figure 14 outlines the generation and recombination of carriers of various kinds that determine the stationary concentration of photocarriers in wide bandgap solids [85]. In general, both electrons and holes generated in wide bandgap solids on absorption of photons with hν > Ebg initially transit into states at some energy distance from the bottom of the CB and the top of the VB, respectively (Step 1 in Figure 14a). Thermal equilibrium between the crystal lattice and the photocarriers is established as a result of relaxation of the energy and momentum of carriers within the timescale of ~ 10−10 s (Steps 2 and 2′). Subsequent lowering of the energy of carriers occurs via recombination and trapping. For ideal (non-defective) solids, only radiative and nonradiative band-to-band recombination is possible (Steps 3 and 3′). 3.1.3. Photogeneration, Recombination and Trapping of Charge Carriers in Photoactive Solids 3.1.3. Photogeneration, Recombination and Trapping of Charge Carriers in Photoactive Solids Photoexcitation of solid photocatalysts in their fundamental absorption bands is the most appropriate type of photoexcitation in heterogeneous photocatalysis, as both electrons (reductants) and holes (oxidants) are generated. Concurrently, the generation of carriers also occurs as a result of light irradiation at wavelengths corresponding to extrinsic absorption bands, i.e., in the longer wavelength spectral region with respect to the fundamental absorption edge of the solids. The same is true of dopant-sensitized photocatalysts. Recombination of electrons and holes can occur by two pathways: (i) direct recombination of free electrons from the conduction band with free holes from the valence band; and (ii) indirectly through the participation of defects (recombination centers). In the latter case, sequential trapping 438 Catalysts 2018, 8, 553 of free carriers of opposite charge by recombination centers takes place in each recombination cycle. The quasi-stationary concentrations of carriers established under stationary illumination of the photocatalyst depends on the concentration and trapping cross-section of the recombination centers. When photogeneration of carriers takes place in the bulk of the photocatalyst particle (a typical case), photoelectrons and photoholes reach the surface via diffusion. On the other hand, a fraction of the carriers is captured by various carrier traps, both in the bulk and at the surface of the solid particle in sufﬁciently high number, provided that trapping cross sections of one carrier takes place relatively slow (so-called deep traps of the color center type). Deep traps that capture the carriers may be centers of the V-type or of the F-type, depending on the charge of the captured carrier. Surface traps of this kind play the role of photocatalytic active centers. In relation to the chemical interaction with molecules, it is useful to distinguish two states of such centers—the inactive state (without trapped carriers) and the active state (trap is occupied by either an electron or a hole). (e−+ h+) →TiO2 (band-to-band recombination) (58) Kv Kv Kv Kv Kv Kv K K K K K H H H H H (FE (FE (YE (YE 5 F V K K H H DE ,, K HWU KROHUHFRPELQDWLRQqU (FE (YE (WU H H H H H HWU SKRQRQWKHUPDOLRQL]DWLRQqWK HWU SKRWRQSKRWRLRQL]DWLRQqSK F qUqWKqSK Kv , ,,, , ,, ,,, Kv Kv Kv Kv Kv Kv K K K K K H H H H H (FE (FE (YE (YE 5 F V K K H H DE ,, K HWU KROHUHFRPELQDWLRQqU (FE (YE (WU H H H H H HWU SKRQRQWKHUPDOLRQL]DWLRQqWK HWU SKRWRQSKRWRLRQL]DWLRQqSK F qUqWKqSK Kv , ,,, , ,, ,,, Kv Kv Kv Kv K K K H H H (FE (YE 5 D Kv Kv K K H H (FE (YE F V K K H H E E ,, K HWU KROHUHFRPELQDWLRQqU (FE (YE (WU H H H H HWU SKRQRQWKHUPDOLRQL]DWLRQqWK HWU SKRWRQSKRWRLRQL]DWLRQqSK F qUqWKqSK Kv , ,,, , ,, ,,, F Figure 14. Schemes illustrating the processes of: recombination (a); and trapping (b) of photocarriers and events leading to discharge of the defect (for distinctness for defect with captured electron): Step 1: band-to-band optical transition with “hot” electron and “hot” hole generation, where the corresponding initial levels of excited electron and hole lay above the “bottom” of conduction band (Ec) and below the “selling” of the valence band (Ev); Steps 2 and 2′: thermal relaxation of hot carriers; Steps 3 and 3′: radiative and nonradiative band-to-band recombination; Steps 4 and 4′: radiative and nonradiative recombination via defects; Steps 5 and 5’: trapping of carriers with formation of color centers; and Steps 6 and 6′: trapping and de-trapping of carriers by shallow traps (for further details, see text). (c) various stages I, II, and III regarding recombination, thermal ipomnization and photoionization, respectively. Adapted from Artemiev and Ryabchuk [85]. 3.1.4. Trapping of Carriers by Defects The nonradiative electron transitions CB → VB in wide bandgap solids (Ebg ≥3 eV) have low probability since participation of many phonons with energy of 0.1 eV is required in the process. In general, the higher is Ebg, the higher is the probability of radiative transitions in solids with emission of photons manifested as an inter-band or edge luminescence. At the same time, the probability of radiative band-to-band transitions is rather low, since the momentum conservation law (see above) requires that the condition similar to that for band-to-band direct transitions be satisﬁed. Typically, band-to-band luminescence in wide bandgap solids with photon energy close to Ebg (Step 3, Figure 14a) is detected at moderate intensities of the exciting light for rather perfect crystals, for which competitive radiative recombination of carriers through defects (recombination centers, R) is suppressed, or in the case of intense photoexcitation of the crystal when a high concentration of photocarriers is achieved in the solid. 439 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 A more detailed treatment of band-to-band recombination was outside the scope of this review article, but sufﬁces to note that band-to-band recombination in line with band-to-band optical excitation in TiO2-based heterogeneous photocatalysis is often treated following Equations (57) and (58). TiO2 + hν →(e−+ h+) (57) TiO2 + hν →(e−+ h+) (57) (57) (e−+ h+) →TiO2 (band-to-band recombination) (58) (e−+ h+) →TiO2 (band-to-band recombination) (58) (e−+ h+) →TiO2 (band-to-band recombination) (58) (58) With regard to the effectiveness of band-to-band recombination, one should use caution in believing the bimolecular stage described by Equation (58) as being real (not symbolic) in kinetic studies of heterogeneous photocatalytic reactions, even though it can easily explain experimental non-linear dependencies of reaction rates on light intensities, which typically follow a square-root dependence. Kv Kv Kv Kv Kv Kv K K K K K H H H H H (FE (FE (YE (YE 5 F V K K H H DE ,, K HWU KROHUHFRPELQDWLRQqU (FE (YE (WU H H H H H HWU SKRQRQWKHUPDOLRQL]DWLRQqWK HWU SKRWRQSKRWRLRQL]DWLRQqSK F qUqWKqSK Kv , ,,, , ,, ,,, Figure 14. Schemes illustrating the processes of: recombination (a); and trapping (b) of photocarriers and events leading to discharge of the defect (for distinctness for defect with captured electron): Step 1: band-to-band optical transition with “hot” electron and “hot” hole generation, where the corresponding initial levels of excited electron and hole lay above the “bottom” of conduction band (Ec) and below the “selling” of the valence band (Ev); Steps 2 and 2′: thermal relaxation of hot carriers; Steps 3 and 3′: radiative and nonradiative band-to-band recombination; Steps 4 and 4′: radiative and nonradiative recombination via defects; Steps 5 and 5’: trapping of carriers with formation of color centers; and Steps 6 and 6′: trapping and de-trapping of carriers by shallow traps (for further details, see text). (c) various stages I, II, and III regarding recombination, thermal ipomnization and photoionization, respectively. Adapted from Artemiev and Ryabchuk [85]. 3.1.7. Trapping of Carriers with Formation of Centers Similar to Color Centers 3.1.7. Trapping of Carriers with Formation of Centers Similar to Color Centers In addition to the recombination centers R, other types of centers can be distinguished in wide bandgap solids—e.g., the color centers of the V- and F-type and shallow (or thermal) traps (Figure 14b). Discrimination between centers is rather arbitrary. Following the work of Siline and Trukhin [86], we may assume that a center captures photocarriers with a probability p and loses them with probability q (units: number of events per unit time). The rate of formation of defects in a state with a trapped carrier dn/dt (units: cm−3 s−1) is then given by Equation (59) with N0 deﬁning the initial concentration of empty trap centers. d dn dt = p(N0 −n) −qn (59) (59) Equation (59) leads to the kinetics of formation of ﬁlled traps n(t) given by Equation (60), and to the stationary concentration of ﬁlled traps n∞by Equation (61), n(t) = N0 p p + q(1 −e−(p+q)t) (60) n(t) = N0 p p + q(1 −e−(p+q)t) (60) n∞= N0 p p + q (61) (60) n∞= N0 p p + q (61) n∞= N0 p p + q (61) Three possible events can occur for a carrier trapped at a trap center (Figure 14c): 1. Stage I, recombination via trapping of carriers of the opposite sign; 2. Stage II, thermal emission of carriers into the corresponding band; and 3. Stage III, photo-ionization of the trapped carrier by absorption of photons. 3. Stage III, photo-ionization of the trapped carrier by absorption of photons. Accordingly, q in Equation (61) is given by, Accordingly, q in Equation (61) is given by, Accordingly, q in Equation (61) is given by, q = qr + qth + qph (62) (62) q = qr + qth + qph For simplicity, if the process deﬁned by Stage III were neglected (i.e., qph = 0), and if q = (qr + qth) >> p, and qr >> qth, then n∞<< N0. In other words, ﬁlled traps do not accumulate in a solid in numbers compared with the initial number of empty traps N0. Typically, these deep trap centers with high cross-section of carrier trapping that do not accumulate in illuminated solids are the recombination centers. 3.1.6. Recombination of Carriers via Defects Recombination of photocarriers via recombination centers is the main pathway of carrier recombination in imperfect wide bandgap solids. In this case, a given recombination center R subsequently captures a free electron and a free hole, or vice versa, in a single recombination cycle (Steps 4 and 4′ in Figure 14a). In an alternative approach (without the concept of holes), recombination via defects means a consequent two-step transition of electrons from the CB to an empty state of the defect and from the defect to an empty state of the VB, or from the defect to the VB and then from the CB to an empty state of the recombination center R. The excess energy of the electron is dissipated at both steps of the recombination. Typically, the dissipation of excess energy at one of the two stages 440 Catalysts 2018, 8, 553 occurs via a nonradiative pathway with the assistance of existing phonons. Normally, the probability of nonradiative transitions increases as the number of emitted phonons decreases so that the energy levels corresponding to efﬁcient recombination centers typically lie near the middle of the energy gap in wide bandgap solids. Radiative transitions at any stage of the two-step recombination via a center R are responsible for recombination luminescence in solids. 3.1.7. Trapping of Carriers with Formation of Centers Similar to Color Centers In contrast to deep centers, carriers trapped in shallow traps with energy ~ kT from the CB (electrons) or from the VB (holes) can be detrapped via thermal emission with probability qth such that qth ~ exp(–E/κT) >> qr. Deep centers with a low probability of trapping a second carrier (p ≈q) accumulate in illuminated solids in sufﬁciently high number (n∞≈No). Historically, they have been referred to as color centers. The ultimate concentration of color centers can reach values of n∞= 1017–1018 cm−3 for some samples of wide bandgap solids. The term color center originated from the accumulation of such traps in crystals that display absorption bands in the visible spectral region. The result of these traps is that transparent (white when powder) solids become colored under UV illumination, with the color tending to be rather stable at ambient temperature. Such solids can be discolored by calcination, or by thermal bleaching (probability of this process given by qth ∼= exp(−E/κT) is high), or partly by photobleaching (qph is also high) via illumination of the crystalline specimen with light at the wavelengths corresponding to the absorption bands of the color centers. 441 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 3.1.8. Lifetime and Concentration of Free Charge Carriers 3.1.8. Lifetime and Concentration of Free Charge Carriers A major factor affecting processes in heterogeneous photocatalysis is the stationary (surface) concentration of charge carriers. As such, a consideration of some basic approaches that describe the processes responsible for the charge carrier concentration is worth noting. The spatially uniform photogeneration of charge carriers occurs at a constant rate of generation, g, in some space of the solid bulk. The rate of the temporary alteration of the charge carrier concentration is then given by Equation (63), dn dt = g −n τ (63) (63) where τ is the lifetime of the free charge carrier, independent of the charge carrier concentration. In the case of several types of trapping and recombination centers, the lifetime of the charge carriers can then be expressed by Equation (64), provided that the various centers do not interact with each other. Otherwise, the determination of the lifetime becomes rather more complex [87]. 1 τ = ∑ i 1 τi (64) (64) where τi = 1/σtr vNi is the lifetime of the charge carrier with respect to trapping by the defects of a given i-sort; σtr is the trapping cross-section; v is the velocity of the charge carrier; and Ni is the concentration of the defects of a given sort. After an initial period of irradiation, recombination centers and trapping centers are ﬁlled with charge carriers to a stationary level; the lifetime of the charge carriers becomes pseudo-constant and its value determines the stationary concentration of the free charge carriers. Thus, dn/dt = 0, and the stationary concentration of charge carriers is then (Equation (65)), n∞= gτ (65) (65) n∞= gτ Obviously, the higher the concentration of recombination centers and the larger the trapping cross-section are, the shorter is the lifetime and the smaller is the concentration of free charge carriers. For semiconductors and isolators, the lifetime of photoinduced charge carriers can vary in a wide range from picoseconds (10−12 s) to milliseconds (10−3 s). When stationary conditions are established for the charge carriers, one fraction of the carriers remains in the delocalized state, while the other fraction is trapped by the defects. The relation between charge carriers obeys the electroneutrality principle (Equation (66)), ne + ne tr = nh + nh tr (66) (66) where netr and nhtr are the concentrations of electrons and holes, respectively, trapped by the defects. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) As discussed earlier, nitrogen oxides (NOx) are major atmospheric pollutants that play an important role in atmospheric chemistry, and have been the object of a signiﬁcant number of investigations toward their minimization, if not complete removal from the environment. The concentration of NOx in polluted urban air is around 100 ppbv, whereas in the unpolluted troposphere, it ranges from 10 to 500 pptv [88]. Recall that NOx are emitted primarily from artiﬁcial sources (e.g., trafﬁc, coal burning boilers, thermal power plants, and industries of various sorts) and from natural sources (e.g., biological degradation in soil and from lighting thunder). NOx participate in various environmental processes: for instance, in the formation of acid rain; in the greenhouse effect in synergy with sulfur oxides; in the formation of photochemical smog in the presence of CO and VOCs; in the depletion of stratospheric ozone; and in the formation of peroxyacetyl nitrates (PAN), all of which have negative effects on ecosystems and lead to non-insigniﬁcant human health issues. With regard to the latter, NOx pollutants cause problems in the respiratory tract that include lung edema and the reduction of the oxygen-caring capacity of blood—e.g., in the transformation of hemoglobin to methemoglobin. No wonder then that signiﬁcant efforts have been expended to reduce environmental NOx agents back to N2 via a thermal technology using a variety of reductants (e.g., CO, hydrocarbons, H2 and NH3) in what is known as Selective Catalytic Reduction (SCR). While reduction of NO occurs around 100 ◦C in the presence of H2 and a Pd-supported catalyst [89], other reactions require signiﬁcantly greater temperatures. In fact, reduction of NO to N2 through selective catalytic reduction with NH3, and thus potentially treat NOx agents, the costs of the SCR technology for the construction and operation of a facility to treat NOx pollutants, together with the required consumption of energy, may prove prohibitively high. Nonetheless, despite the many efforts to eliminate the NOx emitted from the various sources noted earlier by SCR, the fact remains that the concentration of NO in air in Japan was nearly constant throughout the 1980s, and was often higher than the air quality standard set for NO2, principally along heavily trafﬁcked roads in densely populated areas [90]. This led to the development of a new technology for the disposal of NOx at sub-ppm level from air and from trafﬁcked roads and tunnels, and other environmental sources that emit NOx. 3.1.8. Lifetime and Concentration of Free Charge Carriers ere netr and nhtr are the concentrations of electrons and holes, respectively, trapped by the defects. The lifetime of a trapped charge carrier is determined by the efﬁciency of either thermo- or photo-ionization of the center or by the efﬁciency of the recombination event. When thermo-ionization dominates, the lifetime τth (Equation (67)) varies from picoseconds for shallow traps with depth comparable to kT, to inﬁnite time for color centers in wide bandgap insulators. τth = 1 qth ≈10−12e( E kT ) (67) (67) However, when the lifetime of the trapped charge carrier is determined by recombination (Equation (68)) then 1 1 τr = 1 qr = 1 σrvn (68) (68) 442 442 Catalysts 2018, 8, 553 where σr is the recombination cross-section, v is the velocity of the charge carrier of the opposite sign, and ni is the concentration of the charge carrier of the opposite sign. The higher is the concentration of the opposite charge carrier, the shorter is the lifetime of the trapped charge carrier. Charge carrier trapping and recombination processes determine the stationary concentration of charge carriers at the surface. Moreover, considering that surface defects with trapped charge carriers can act as surface-active centers that initiate surface chemical sequences, the lifetime of trapped charge carriers on such surface defects corresponds to the lifetime of the chemically active states of the surface-active centers. Recombination processes then return the surface-active centers to their (initial) chemically inactive ground states. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Recognition that plants and micro-organisms can easily consume nitrite (NO2−) and nitrate (NO3−) ions as raw materials for nitrogen assimilation provided a further impetus to examine alternative technologies to achieve a practical removal of dilute NOx agents from the environment using sunlight (UV-Visible) radiation at signiﬁcantly lower costs. In this regard, Takeuchi and Isubuki [91] investigated the dry deposition of NOx onto the ground and found that the rate of adsorption of NOx on some soil particles was enhanced by photoillumination. Of the metal oxides constituting the soil particles, TiO2 showed the highest activity for NOx adsorption under photoillumination with ca. 60% of NOx being captured as nitric acid (HNO3) on the surface of TiO2 particulates. Accordingly, the authors thought that the photocatalytic oxidation of NOx to HNO3 by illuminated TiO2 might be most advantageous to treat dilute environmental NOx, as any extra reactants such as NH3 were not required and HNO3 could be trapped on the surface as nitrates. 443 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 One of the ﬁrst studies to examine the fate of one of the NOx agents, namely NO, in the presence of (Degussa) P-25 TiO2 exposed to UV irradiation was reported in 1984 by Courbon and Pichat [92] who exposed isotopically labeled N18O at 295 K in the dark to pre-oxidized and pre-reduced TiO2 powder; subsequent to UV illumination resulted in three phenomena: photoadsoprtion, photoexchange, and photodecomposition of NO to yield N2O and, to a lesser extent, N2. The formation of N2O + N2 corresponded to a photodecomposition of ca. 15% of the NO pressure (decrease) for a pre-reduced titania sample and ca. 20% for a pre-oxidized titania; N2 formed only at the beginning and the percent N216O produced was initially greater for the pre-oxidized titania sample. This early study [92] conﬁrmed that illumination of TiO2 with UV light considerably increased the ease of detachment of surface oxygen atoms, as the isotopic hetero-exchange of N18O occurred at room temperature, while it required higher temperatures in the absence of bandgap (3.2 eV) illumination of the mostly anatase TiO2. Adsorbed oxygen species were involved, as pre-oxidized titania exhibited higher initial efﬁciency; however, the instantaneous exchange with a pre-reduced titania sample in H2 at 723 K showed that detachment also involved surface oxygen atoms that were replenished from NO. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Another aspect of this study was the corroboration of the direct involvement of O−species in photocatalytic oxidations over TiO2 and other n-type semiconductors, since NO and O2 played similar roles in yielding dissociated oxygen species active in both oxidation and oxygen isotopic exchange. A later study (1985) by Hori and coworkers [93] demonstrated that NO2−ions are oxidized to NO3−with or without O2 in aqueous suspensions of some semiconductor powders (Ag2O, PbO, anatase TiO2, Si, ZnO, SnO2, CdS, and Bi2O3) under bandgap illumination; with TiO2, 96% of nitrite was oxidized to nitrate in the presence of oxygen. Along similar lines, Anpo and coworkers [94] found that Cu+ ions on SiO2 (Cu+/SiO2 catalyst) could decompose NO molecules photocatalytically and stoichiometrically into N2 and O2 at 275 K, which they attributed to the signiﬁcant role played by the excited state of the Cu+ species; the photoreaction involved an electron transfer from the excited state of the Cu+ ion into an anti-bonding π orbital of the NO molecule within the lifetime of its excited state. The relationship between the local structures of Cu+ ions in zeolite and their photocatalytic reactivity in the decomposition of NOx into N2 and O2 at 275 K was reviewed by Anpo and coworkers [95] after which Anpo’s group [96] reported on the metal ion-implantation of TiO2 with various transition metal ions that subsequent to calcination in oxygen at ca. 723 K resulted in a large shift of the absorption edge of TiO2 toward visible light regions depending on the amount and type of metal ions implanted; the resulting metal ion-implanted TiO2s proved active in the photocatalytic decomposition of NO to N2, O2 and N2O at 275 K under irradiation with visible light at wavelengths longer than 450 nm. Following their 1989 report [91], Ibusuki and Takeuchi [97] examined the photocatalytic destructive oxidation of NO to NO3−using a mixture (200–250 mg) of TiO2, activated carbon (AC) and Fe2O3 particles located in a ﬂow-type photochemical reactor system (Figure 15) that was photo-illuminated by a cylindrical bank of 12 black lights (wavelength: 300–400 nm) [97]. The AC and Fe2O3 had a remarkable effect in increasing the catalytic activity for NOx removal, likely due to their high adsorptive activity for NO and NO2. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The authors inferred that photo-illuminated TiO2 generated reactive oxygen species that oxidized NO and NO2, respectively, to NO2 and NO3−, while activated carbon trapped NO2 to allow enough time for TiO2 to oxidize NO2 to NO3−; it appears that Fe2O3 acted as a promoter for more NO/NO2 molecules to be adsorbed on the surface of the titania photocatalyst [97]. In a further study, the Takeuchi group [98] examined the use of TiO2 to eliminate NOx in open air with the photocatalyst being activated by sunlight, but noted, however, that in so doing desorption of NO2 occurred during the oxidative removal of NO; the NO2 also needed to be suppressed as it is also a regulated pollutant. Although NOx adsorb on activated carbon to be oxidized ultimately to NO3−, development of an activated carbonaceous photocatalytic material proved difﬁcult. Accordingly, recognizing that thin ﬁlms have many micropores they designed and prepared TiO2 thin ﬁlm photocatalysts by a dip-coating process using titanium alkoxide as the TiO2 precursor and the polymer additive polyethylene glycol (PEG) of different molecular masses (PEG-300, PEG-600, PEG-1000) to 444 Catalysts 2018, 8, 553 give TiO2-PX ﬁlms with thicknesses of 1.0, 0.5 and 0.25 μm after calcination of the ﬁlms at 450 ◦C for 1 h deposited on silica-coated glass plates. Figure 15. Schematic diagram of the ﬂow-type reactor used for the heterogeneous photocatalytic reductions of NOx agents. Experimental conditions: 1–2 ppm NO/NO2; ﬂow rate, 500 mL min−1 for 5–10 h at different relative humidity (dry to 72%); reactor volume, 126 mL; pure O2 or puriﬁed air; reaction temperature, ca. 310 K. Reproduced with permission from Ibusuki and Takeuchi [97]. Copyright 1994 by Elsevier B.V. (License No.: 4452271448487). Figure 15. Schematic diagram of the ﬂow-type reactor used for the heterogeneous photocatalytic reductions of NOx agents. Experimental conditions: 1–2 ppm NO/NO2; ﬂow rate, 500 mL min−1 for 5–10 h at different relative humidity (dry to 72%); reactor volume, 126 mL; pure O2 or puriﬁed air; reaction temperature, ca. 310 K. Reproduced with permission from Ibusuki and Takeuchi [97]. Copyright 1994 by Elsevier B.V. (License No.: 4452271448487). Table 1 summarizes the extent of NO removal [98]. Adsorbed NO was photooxidized to NO2 by the thin ﬁlms, while the produced NO2 was re-photooxidized to NO3−before it desorbed from the ﬁlm surface. Table 1. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Extent of NO removal over 1.0 μm thick TiO2-PX thin ﬁlms irradiated at 365 nm (illuminated area, 100 cm2; light irradiance, 0.38 mW cm−2) for 12 h in a ﬂow-type reactor (ﬂow rate, 1.5 L min−1; initial concentration of NO, 1.0 ppm; dry air). Photocatalytic Thin Film BET Surface Area (m2 g−1) Average NO Removal (%) TiO2-P0 112 71 TiO2-P300 104 65 TiO2-P600 118 70 TiO2-P1000 141 81 Following reports that TiO2 prepared by high-temperature hydrolysis of titanium tetra-alkoxides, Ti(OR)4, in a hydrocarbon solvent was very active toward the photocatalytic dehydrogenation of iso-propanol in aqueous media under deaerated conditions [99] and mineralization of acetic acid under aerated conditions [100], Hashimoto et al. [101] prepared TiO2 by the hydrolysis of titanium alkoxide in a hydrocarbon solvent, followed by calcination at various temperatures; the titania calcined at 300 ◦C proved most active for the photocatalytic oxidation of NO (Table 2), in comparison with P-25 titania. The photocatalytic oxidation was carried out in a ﬁxed bed continuous ﬂow Pyrex-glass reactor under atmospheric pressure with the TiO2 (0.12 g) UV-irradiated with a 10-Wh black light; air contained 10 ppm of NO; ﬂow rate, 110 mL min−1. IR spectral results indicated that UV irradiation promoted the oxidation of NO in the presence of oxygen to yield nitrate species, while the data from ESR measurements for oxygen radicals showed that UV irradiation increased the number of O2−• adsorbed on the surface of titania in the presence of oxygen. These O2−• species vanished simultaneously with their exposure to NO, whereas the spectral intensity of the radical generated from secondary products of O−showed no change. The number of O2−• radical anions generated by UV irradiation reﬂected the photocatalytic oxidative activity of titania toward the oxidation of NO. The rate of formation of O2−• and the number of free electrons induced by UV irradiation decreased signiﬁcantly with an increase in post-calcination temperature (Table 2) [101]. 445 Catalysts 2018, 8, 553 Table 2. Crystal size and surface area of the titania photocatalysts together with the rate of formation and the quantities of the O2−• radical anions . 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Photocatalyst Calcination T (◦C) Crystal Type Crystal Size (nm) BET Surface Area (m2 g−1) Rate of O2−• Formation (μmol min−1) [O2−•] (μmol g−1) TiO2 300 Anatase 10 133 2.9 7.2 TiO2 550 Anatase 18 78 0.80 3.7 TiO2 700 Anatase 26 34 0.06 0.25 TiO2 800 Anatase + Rutile 47 8 0.01 0.07 Adapted with permission from Hashimoto et al. [101]. Copyright 2000 by Elsevier Science S.A. (License No.: 4453260419845). A photocatalytic reaction that takes place in a gas/solid reactor necessitates both the exposure of the catalysts to light irradiation and good contact between reactants and catalyst. In this regard, Lim and coworkers [102] noted that a two-dimensional ﬂuidized-bed photoreactor not only increased the contact of catalyst and gas, but also enhanced UV light penetration compared with a packed bed reactor in which light could not easily penetrate the interior of the catalyst bed, so that it was important to design and fabricate a ﬂuidized-bed photoreactor with higher light throughputs and lower pressure drops. Accordingly, they used: (i) an annular ﬂow-type photoreactor; and (ii) a modiﬁed two-dimensional ﬂuidized-bed photoreactor to examine the photocatalytic decomposition of NO. In the ﬁrst case, two serial annular ﬂow photoreactors were used to increase contact time between the gas and the photocatalyst (Figure 16). Figure 16. Schematic diagram of the annular ﬂow-type photoreactor composed of two quartz glass tubes (height, 500 mm; diameters, 12 mm and 20 mm). Reproduced with permission from Lim and coworkers [102]. Copyright 2000 by Elsevier Science S.A. (License No.: 4452641174439). Figure 16. Schematic diagram of the annular ﬂow-type photoreactor composed of two quartz glass tubes (height, 500 mm; diameters, 12 mm and 20 mm). Reproduced with permission from Lim and coworkers [102]. Copyright 2000 by Elsevier Science S.A. (License No.: 4452641174439). By comparison, the modiﬁed two-dimensional ﬂuidized-bed reactor (Figure 17) consisted of an annular-type reactor made of a larger quartz glass tube (internal diameter, 30 mm; height, 400 mm) in which a small diameter quartz tube (inner diameter, 20 mm; height, 375 mm) was located at the center of the larger tube such that the thickness of the annulus in the bed was 5 mm [102]. A quartz ﬁlter (100-mesh size) was used to distribute a uniform ﬂuidization of the catalyst; a square mirror box surrounded the photoreactor to minimize loss of light irradiation and to improve utilization of reﬂected and deﬂected light. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) In their study [102], the authors examined the effects of gas-residence time, initial NO concentration, reaction temperature and UV light source on the photocatalytic decomposition of NO carried out in the annular ﬂow-type reactor. P-25 titania powder was used to cover a quartz tube (430 mm) by dipping it into a stirred 5% TiO2 slurry solution and then air-dried for 24 h, after which the TiO2-coated quartz tube was ﬁred in a high-temperature furnace at 400 ◦C for 1 h; TiO2 coating was repeated several times until the amount of TiO2 deposited on the quartz tube reached 0.10 g. The quartz tube had been sandblasted previously to create a granular texture to anchor the ﬁne TiO2 446 Catalysts 2018, 8, 553 powder. In addition, precursor solutions for coating TiO2 on silica gel were prepared using titanium ethoxide as a precursor to prepare the TiO2 sample. Figure 17. Modiﬁed two-dimensional ﬂuidized-bed photoreactor. Reproduced with permission from Lim and coworkers [102]. Copyright 2000 by Elsevier Science S.A. (License No.: 4452641174439). Figure 17. Modiﬁed two-dimensional ﬂuidized-bed photoreactor. Reproduced with permission from Lim and coworkers [102]. Copyright 2000 by Elsevier Science S.A. (License No.: 4452641174439). A gas stream (200 mL min−1) of 138 ppmv NO in He in the annular ﬂow-type reactor was irradiated by four UV lamps without TiO2 photocatalyst at ambient temperature for 140 min with no variation in the NO concentration; in the presence of TiO2, however, irradiation led to the decomposition of NO with formation of NO2, N2O and N2 products [102]. The reaction rate followed the power law R = Ro In with n = 0.48 and n = 0.87 depending on the UV intensity of the germicidal white lamp (254 nm) and the ﬂuorescent black lamp (365 nm), respectively (gas ﬂow rate, 100 mL min−1; TiO2 loading, 0.1 g; reaction temperature, 311 K; initial NO concentration, 50 ppm). Adsorption of nitrate on the surface of the photocatalyst increased with irradiation time leading to the deactivation of the photocatalyst. The decomposition of NO decreased linearly on increasing the initial NO concentration and on decreasing the residence time of gas in the photoreactor, so that it was necessary to increase the residence time of the gaseous reactant to provide effective contact of UV light, gaseous reactant and photocatalyst to obtain higher NO decomposition in the annular photoreactor. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) In the modiﬁed two-dimensional ﬂuidized-bed photoreactor, four reaction conditions (without TiO2/SiO2 and UV lamp on/off, with TiO2/SiO2 and UV lamp-on/off) were tested to conﬁrm whether the decomposition of NO really took place by a photocatalytic process. Indeed, in the presence of TiO2/SiO2 and UV lamp-on, the NO concentration decreased indicating that it was in fact decomposed [101]. Decomposition of NO increased with decreasing initial NO concentration and increasing gas-residence time; the reaction rate increased with increasing UV light intensity. Clearly, the modiﬁed photoreactor displayed efﬁcient contact between photocatalyst and reactant gas with good transmission of UV-light and, consequently, increased the NO decomposition efﬁciency (>70%) compared with the annular ﬂow-type photoreactor. Hence, the former photoreactor was an effective tool with which to carry out signiﬁcant NO decomposition with efﬁcient utilization of photon energy [102]. Anpo and coworkers [96] had earlier prepared a TiO2 photocatalyst that subsequent to the implantation of Cr ion and upon irradiation with visible light (>450 nm) decomposed NO into N2, O2, and N2O under O2-free conditions. Additionally, the Cr ion-implanted TiO2 catalyst displayed the exact same photocatalytic efﬁciency as the original TiO2 catalyst, albeit under UV irradiation. As a follow-up to this study, Nakamura et al. [103] examined the role of oxygen vacancies in the removal of NO under an oxidative atmosphere using a commercial TiO2 (Ishihara ST-01; 100% anatase; crystallite size, 7 nm; nominal speciﬁc surface area, 300 m2 g−1) and hydrogen plasma-treated TiO2 powders; the latter was photoactive up to 600 nm without a decrease in UV light activity. Reactions to remove 1.0 ppm of NO were carried out in a Pyrex glass ﬂow reactor (500 cm3) with irradiation from a 447 Catalysts 2018, 8, 553 300-W Xe light source; the UVA (315–400 nm) irradiance at the photocatalyst surface was 0.03 mW cm−2 (ﬂow rate, 1500 mL min−1; TiO2 loading, 0.20 g; total pressure, 760 Torr)—no removal of NO occurred without the metal oxide photocatalyst. NO was converted mainly to NO3−(also less than 2% NO2−formed) by oxidation over the TiO2 powder; NO3−ions accumulated on the catalyst surface. Electrons trapped in oxygen vacancies in the plasma-treated TiO2 were detected under visible light irradiation (F-type color centers; ESR measurements displayed a signal at g = 2.004) with the number being proportional to the percent of NOx removed, which suggested that the number of trapped electrons determined the activity of the photocatalytic oxidation of NO to NO3−. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The anatase TiO2 catalyst with the larger surface area, wider bandgap, and numerous surface –OH groups (Table 3) exhibited the highest photocatalytic reactivity in the decomposition of NO, which the authors deduced that these were the principal factors that affected photocatalytic efﬁciency. The increased bandgap of JRC-TiO-4 was accompanied by a shift in the conduction band edge to higher negative energies, thus moving the redox potential to more negative values thereby enhancing photocatalytic reactivity. Moreover, surface –OH groups and/or physisorbed H2O also played a signiﬁcant role in the photocatalytic reactions through the facile formation of reactive •OH radicals. The intensity of the incident light is also an important factor that affects the kinetics of the photocatalytic decomposition. The quantum efﬁciency of the photocatalytic reaction was higher at the lower intensities of the incident light, and lower at higher intensities of the incident UV light; in addition, the efﬁciency of conversion of NO increased with increase in the O2 ﬂow rate. (a) (b) -5&7L2-5&7L2-5&7L2-5&7L2 12&RQYHUVLRQ Figure 19. (a) Flow system for the photocatalytic reaction of NOx. Conditions: 150 mg TiO2; NO reactant gas (NO + He), 10 ppm; ﬂow rate, 100 mL min−1; irradiation time, 2 h; Toshiba SHL-100UV high-pressure Hg lamp; color ﬁlter, UV-27 (λ > 270 nm). (b) Conversion of the photocatalytic decomposition of NO on the standard reference TiO2 photocatalysts at room temperature; adapted from Ref. [104]. Reproduced with permission from Zhang et al. [104]. Copyright 2001 by Academic Press (License No.: 4452650162755). (a) (b) -5&7L2-5&7L2-5&7L2-5&7L2 12&RQYHUVLRQ (b) (a) Figure 19. (a) Flow system for the photocatalytic reaction of NOx. Conditions: 150 mg TiO2; NO reactant gas (NO + He), 10 ppm; ﬂow rate, 100 mL min−1; irradiation time, 2 h; Toshiba SHL-100UV high-pressure Hg lamp; color ﬁlter, UV-27 (λ > 270 nm). (b) Conversion of the photocatalytic decomposition of NO on the standard reference TiO2 photocatalysts at room temperature; adapted from Ref. [104]. Reproduced with permission from Zhang et al. [104]. Copyright 2001 by Academic Press (License No.: 4452650162755). 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The visible-light photocatalytic activity of the plasma-treated TiO2 was due to photoexcitation of the F-type color centers with energy levels within the forbidden bandgap of the metal oxide (see Figure 18). 7L2DQDWDVH 9DOHQFH%DQG &RQGXFWLRQ%DQG H9 H9 H9 H9 2YDFDQFLHV Figure 18. Proposed band structure model for the anatase TiO2 with oxygen vacancies. Figure 18. Proposed band structure model for the anatase TiO2 with oxygen vacancies. The Anpo group [104] investigated the photocatalytic decomposition of NOx (NO and NO2) on ﬁve well-characterized standard reference ultraﬁne powdered TiO2 photocatalysts (grain size, 0.02–1 μm) denoted TiO2 (JRC-TIO-2, -3, -4, and -5) supplied by the Catalysis Society of Japan (properties summarized in Table 3) in a large-scale continuous ﬂow reaction system (Figure 19a) with high efﬁciency. Special attention was expended on the effects of pretreatment and reaction conditions on the reaction and conversion rates of NO. The authors established that surface hydroxyl groups played a signiﬁcant role as active sites in the decomposition of NO. Table 3. Physicochemical properties of the standard reference TiO2 photocatalysts (JRC-TiO-2, -3, -4, and -5) supplied by the catalysis Society of Japan . Table 3. Physicochemical properties of the standard reference TiO2 photocatalysts (JRC-TiO-2, -3, -4, and -5) supplied by the catalysis Society of Japan . Catalysts (JRC-TiO-) Surface Area (m2 g−1) Acid Concentration (μmol g−1) Relative –OH Concentration Bandgap (eV) 2 (anatase) 16 6 1.0 3.47 3 (anatase) 51 22 1.6 3.32 4 (anatase) 49 5 3.0 3.50 5 (rutile) 3 7 3.1 3.09 Reproduced with permission from Zhang et al. [104]. Copyright 2001 by Academic Press (License No.: 4453251363124). When used on a large scale for long periods in photoreactions, photocatalysts tended to lose, albeit gradually, their photocatalytic activity. In Anpo’s study [104], after 2 h, conversion of NO for each photocatalyst leveled off and dropped to between 0.25 and 0.20 the photocatalytic activity observed initially, indicating a decline in photocatalytic activity of the TiO2 in the decomposition of NO in the absence of O2 and/or H2O. Reaction products in the ﬂow reaction system were N2, O2, and N2O, just as occurred in a closed reaction system. The JRC-TIO-4 photocatalyst displayed the highest photocatalytic activity for the conversion of NO, while for the other three there were small differences: 448 Catalysts 2018, 8, 553 JRC-TIO-4 (11%) >>> −3 (~ 2%) > −5 (1.8%) > −2 (1%) (see Figure 19b). 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The activity of the JRC-TiO-4 photocatalyst was also tested by Tanaka and coworkers [105] in the photoassisted selective catalytic reduction of NO with ammonia (photo-SCR) at low temperature over irradiated TiO2 in a ﬂow reactor; the process was efﬁcient and the adsorbed ammonia reacted with NO under irradiation of TiO2 (Figure 20); note the nearly identical kinetics of formation of both N2 and N2O. The total amount of N2 formed was 0.23 mmol gcat−1, consistent with the amount of ammonia (0.24 mmol gcat−1) adsorbed over TiO2 in equilibrium at 323 K. The kinetic experiment carried out under differential conditions in the pressure range 300 < p(NO), p(NH3) < 2000 ppm, and the presence of excess O2 affected the evolution rate of N2 which depended only on partial pressure of NO; kinetics were ﬁrst order on NO, and zeroth order on O2 and NH3, which strongly suggested that the rate-determining step was adsorption of NO to the irradiated TiO2 adsorbing ammonia molecules [105]. To the extent that the selective catalytic reduction (SCR) with ammonia is a downhill reaction, it also proceeded in the dark at low temperature with a 20% conversion of NO. However, photoirradiation caused a remarkable enhancement of the activity: the evolution rate of N2 gradually increased attaining a steady rate at ca. 80% conversion after 2 h of irradiation. To achieve a further understanding of surface reactions involved in TiO2-based photocatalysis, Dalton and coworkers [106] examined two titania samples (one of unknown source) using X-ray photoelectron spectroscopy and Raman spectroscopy to investigate the NOx adsorbate reaction at the surface of these two TiO2 substrates. The NOx gas was composed of 109 ± 5 ppm of NOx, 449 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 21.0 ± 0.4% O2, the remaining ca. 79% being N2; dry air was the mixer gas to dilute the NOx (NOx concentration, 10–100 ppm) during the reaction performed under UV exposure for 6 and 48 h in a glass vessel (ca. 3 mm thick; Figure 21) that allowed > 80% transmission of the radiation at λ = 320 nm. Formation of NO3−did not vary signiﬁcantly with either exposure time or NOx concentration. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The authors [106] proposed a stepwise mechanism (Figure 22) in which the surface hydroxyls increased the efﬁcacy of the process and participated by reacting with NOx molecules to yield nitrate ions formed indirectly via initial reductive (formation of O2−• radical anions by conduction band electrons) and oxidative (formation of •OH radicals by valence band holes) processes. 1 12 7LPHRQVWUHDPPLQ 12SSP Figure 20. Outlet concentration of N2 and N2O in the SCR of NO with ammonia at 323 K under irradiation. Conditions: TiO2 loading, 1.2 g; volume of catalyst bed, 1.5 mL; irradiation, 300-W ultra-high pressure Xe lamp reﬂection by a cold mirror; composition of reaction gas, 1000 ppm NO + 5% O2, and balance was Ar gas; ﬂow rate, 100 mL min−1; kN2 = 0.026 ± 0.003 min−1, kN2O = 0.021 ± 0.003 min−1. Adapted with permission from Tanaka et al. [105]. Copyright 2002 by the Royal Society of Chemistry (License No.: 4452660060880). 1 12 7LPHRQVWUHDPPLQ 12SSP 1 12 7LPHRQVWUHDPPLQ 12SSP Figure 20. Outlet concentration of N2 and N2O in the SCR of NO with ammonia at 323 K under irradiation. Conditions: TiO2 loading, 1.2 g; volume of catalyst bed, 1.5 mL; irradiation, 300-W ultra-high pressure Xe lamp reﬂection by a cold mirror; composition of reaction gas, 1000 ppm NO + 5% O2, and balance was Ar gas; ﬂow rate, 100 mL min−1; kN2 = 0.026 ± 0.003 min−1, kN2O = 0.021 ± 0.003 min−1. Adapted with permission from Tanaka et al. [105]. Copyright 2002 by the Royal Society of Chemistry (License No.: 4452660060880). Figure 21. Apparatus for NOx gas removal. Shown are: (i) UV exposure box; (ii) gas ﬂow meters; (iii) NOx gas and air mixer gas in; (iv) gas out; (v) water bubbler for air mixer gas; (vi) glass reaction vessel; and (vii) sodium hydroxide bubbler for excess NOx removal. The system consisted of ﬂow meters to allow an NOx concentration of between 10 and 100 ppm when used in conjunction with the air mixer gas; a water bubbler to allow the reaction to be studied with wet or dry gas’ a second bubbler containing aqueous NaOH was used after the reaction vessel to remove unreacted NOx. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Suggested mechanism for the photocatalyzed oxidative removal of NOx over the irradiated surface of the two TiO2 photocatalytic substrates. GHVRUEHG QLWUDWHV Figure 22. Suggested mechanism for the photocatalyzed oxidative removal of NOx over the irradiated surface of the two TiO2 photocatalytic substrates. Dalton et al. [106] concluded that TiO2 was effective at converting NOx agents to NO3−and that XPS proved useful in quantifying the efﬁciency of the reaction, while Raman spectroscopy was a quick and simple way of ascertaining the surface crystal structure of the titania. XPS conﬁrmed only one oxidation state of Ti on the untreated TiO2 materials; however, the O1s peak indicated the presence of two additional components of TiO2: Ti–OH and Ti–OH2. After exposure to UV radiation, XPS spectra revealed nitrogen peaks attributable to organic species (also present before reaction), to some unreacted NO adsorbed on the surface, and to nitrate anions. Reactive nitrogen (NOy) in the atmosphere consists of the sum of the two NOx oxides (NO + NO2) and all compounds produced by atmospheric oxidation of NOx that include the minor species: HNO3, HNO2, the nitrate radical NO3•, N2O5, peroxynitric acid HNO4, peroxyacetyl nitrate (PAN) (CH3C(O)OONO2) and its homologs, and peroxyalkyl nitrates (RC(O)OONO2) [107]. Such compounds can be regarded as reservoirs of NO2 but apparently play no critical role in the formation of ozone O3 that the precursors NO2 and NO do. The oxidative removal of NO over irradiated TiO2 catalyst was examined by Devahasdin and coworkers [107] at source levels (5–60 ppm) in a thin-ﬁlm photoreactor systems (see Figure 23); the process involved a series of oxidation steps through the action of photoformed •OH radicals (NO →HNO2 →NO2 →HNO3). Light intensity increased the capability to oxidize NO (from 0 to 0.8 mW cm−2); the selectivity for NO2 increased with light intensity for 5 ppm inlet NO but remained constant for 40 ppm inlet NO. The steady-state conversion of NO increased with relative humidity from 0 to 50% leveling off at higher relative humidity; the ratio of NO2−to NO3−from spent catalyst liquor decreased with irradiation time until steady state was reached. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Reproduced with permission from Dalton et al. [106]. Copyright 2002 by Elsevier Science Ltd. (License No.: 4452660699057). 7LPHRQVWUHDPPLQ Figure 20. Outlet concentration of N2 and N2O in the SCR of NO with ammonia at 323 K under irradiation. Conditions: TiO2 loading, 1.2 g; volume of catalyst bed, 1.5 mL; irradiation, 300-W ultra-high pressure Xe lamp reﬂection by a cold mirror; composition of reaction gas, 1000 ppm NO + 5% O2, and balance was Ar gas; ﬂow rate, 100 mL min−1; kN2 = 0.026 ± 0.003 min−1, kN2O = 0.021 ± 0.003 min−1. Adapted with permission from Tanaka et al. [105]. Copyright 2002 by the Royal Society of Chemistry (License No.: 4452660060880). Figure 21. Apparatus for NOx gas removal. Shown are: (i) UV exposure box; (ii) gas ﬂow meters; (iii) NOx gas and air mixer gas in; (iv) gas out; (v) water bubbler for air mixer gas; (vi) glass reaction vessel; and (vii) sodium hydroxide bubbler for excess NOx removal. The system consisted of ﬂow meters to allow an NOx concentration of between 10 and 100 ppm when used in conjunction with the air mixer gas; a water bubbler to allow the reaction to be studied with wet or dry gas’ a second bubbler containing aqueous NaOH was used after the reaction vessel to remove unreacted NOx. Reproduced with permission from Dalton et al. [106]. Copyright 2002 by Elsevier Science Ltd. (License No.: 4452660699057). Figure 21. Apparatus for NOx gas removal. Shown are: (i) UV exposure box; (ii) gas ﬂow meters; (iii) NOx gas and air mixer gas in; (iv) gas out; (v) water bubbler for air mixer gas; (vi) glass reaction vessel; and (vii) sodium hydroxide bubbler for excess NOx removal. The system consisted of ﬂow meters to allow an NOx concentration of between 10 and 100 ppm when used in conjunction with the air mixer gas; a water bubbler to allow the reaction to be studied with wet or dry gas’ a second bubbler containing aqueous NaOH was used after the reaction vessel to remove unreacted NOx. Reproduced with permission from Dalton et al. [106]. Copyright 2002 by Elsevier Science Ltd. (License No.: 4452660699057). 450 Catalysts 2018, 8, 553 9DOHQFH%DQG &RQGXFWLRQ%DQG H9 K H H H K K 2DGV 2DGV 12x DGV 12 DGV 2+ DGV 2+DGV 12DGV 12DGV 12 DGV 2+DGV +2 GHVRUEHG QLWUDWHV 7L2 Figure 22. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The system consisted of a thin-ﬁlm photoreactor coated with Degussa P-25 TiO2 (typical loading, 1.0 mg cm2) irradiated with two 8 W or 25 W black lights from both sides; light intensity of 25 W bulbs varied with a dimming electronic ballast; reactor setup and light sources were contained in insulated chamber for control of temperature; light intensity measured with UVA radiometer (range, 320–390 nm) placed inside the reactor; NO and NO2 measured with a chemiluminescent NO–NOx gas analyzer; initial NO concentrations were 28.5 and 472 ppm; NO2 and NO3−measured by chromatography. Reproduced with permission from Devahasdin and et al. [107]. Copyright 2003 by Elsevier Science B.V. (License No.: 4452661132706). Figure 23. Experimental setup used in the disposal of NOx. The system consisted of a thin-ﬁlm photoreactor coated with Degussa P-25 TiO2 (typical loading, 1.0 mg cm2) irradiated with two 8 W or 25 W black lights from both sides; light intensity of 25 W bulbs varied with a dimming electronic ballast; reactor setup and light sources were contained in insulated chamber for control of temperature; light intensity measured with UVA radiometer (range, 320–390 nm) placed inside the reactor; NO and NO2 measured with a chemiluminescent NO–NOx gas analyzer; initial NO concentrations were 28.5 and 472 ppm; NO2 and NO3−measured by chromatography. Reproduced with permission from Devahasdin and et al. [107]. Copyright 2003 by Elsevier Science B.V. (License No.: 4452661132706). Toma and workers [108] reported using a test chamber built speciﬁcally for the TiO2 (Degussa P-25; in powder or pellet form) photocatalytic decrease of NOx. The experimental device consisted of three parts: (a) a chamber where gaseous NOx were prepared in situ by chemical reaction of Cu powder with a dilute solution of HNO3; (b) an environmental chamber; and (c) a NOx analyzer. The pollutants were subsequently injected at ambient temperature into the environmental chamber (volume, ~0.4 m3) until the concentration of NOx reached 1–2 ppmv. A fan ensured homogenization of the gaseous pollutants in the environmental chamber. A polycarbonate photoreactor (100 mm × 100 mm × 50 mm box) equipped with a 70 × 70 mm Plexiglas window allowed light transmission from a 15-W daylight lamp (30% UVA, 4% UVB) placed inside the environmental chamber and crossed by the NOx ﬂow (ﬂow rate, 0.6 L min−1); NOx concentrations were continuously monitored with a chemiluminescence NOx analyzer. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) p y q y Transient behavior of TiO2 during the ﬁrst 2 h of operation with the system setup of Figure 23 (conditions: space time: 12 s; inlet concentration, 40 ppm; light source, two 8-W black lamps; relative humidity, 50%; TiO2 loading: 1.07 mg cm−2) revealed that initially the conversion of NO was very high (ca. 95% after 0.5–3 min of irradiation depending on TiO2 loading) and decreased approaching steady state after 6 h of operation; all the nitrogen was accounted for in the gas phase: NO out (26 ppm) + NO2 out (14 ppm) = NO in (40 ppm). Conversion of NO was 35%; gas phase mass balance showed no N2O formed in the reaction system under steady-state conditions [106]; NO2 selectivity remained constant at 100% for 40 ppm inlet NO and increased with light intensity for 5 ppm inlet NO with a 50% relative humidity, which suggested that for 40 ppm inlet NO at steady state, all the NO should have been converted to NO2. However, the authors [107] believed that, for the 5 ppm inlet NO, the true steady state had not yet been reached, so that increasing light intensity caused the HNO3 to dissociate back to NO2 and •OH and to promote NO2 selectivity from 82% to 95%. The latter inference called attention for the ﬁrst time to the possible reNOxiﬁcation of the nitrates produced in the deNOxiﬁcation of the environment (see Section 5). 451 Catalysts 2018, 8, 553 Figure 23. Experimental setup used in the disposal of NOx. The system consisted of a thin-ﬁlm photoreactor coated with Degussa P-25 TiO2 (typical loading, 1.0 mg cm2) irradiated with two 8 W or 25 W black lights from both sides; light intensity of 25 W bulbs varied with a dimming electronic ballast; reactor setup and light sources were contained in insulated chamber for control of temperature; light intensity measured with UVA radiometer (range, 320–390 nm) placed inside the reactor; NO and NO2 measured with a chemiluminescent NO–NOx gas analyzer; initial NO concentrations were 28.5 and 472 ppm; NO2 and NO3−measured by chromatography. Reproduced with permission from Devahasdin and et al. [107]. Copyright 2003 by Elsevier Science B.V. (License No.: 4452661132706). Figure 23. Experimental setup used in the disposal of NOx. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) For small TiO2 powder quantities, conversion rates increased proportionally reaching maximal value at 0.2 g loading of TiO2; at higher quantities of TiO2 the decrease in NOx remained constant and independent of TiO2 powder amount. After 30 min of UV irradiation (surface, 54 cm2), conversion rates were about 32–35% and 15–18%, respectively, when the mass of the catalyst varied from 0.2 to 1.2 g; maximal conversion was reached at 3.7–4 mg cm−2 of TiO2 powder [108]. Exposing a TiO2 pellet surface (mass, 0.4 to 1.2 g; 54 cm2) to UV radiation from one side only led to a photocatalytic conversion of ca. 28–30% of NO and 10–12% of NOx (Figure 24); the conversion efﬁciency increased with the surface area of the pellet. The amount of compressed TiO2 powder and the thickness of the pellet had little inﬂuence on the extent of NOx decomposition. Anatase TiO2 showed better efﬁciency for the photocatalytic decrease of NOx relative to rutile TiO2, accounting for only 10% and 5%, respectively, for NO and NOx removal. We noted above that the photocatalytic decomposition of NO over TiO2 reported in some of the literature led to the formation of N2O as the main reaction product [92,94,98] with minor N2, NO2 and O2 products. Only Anpo’s group [96] reported the selectivity of NO photodecomposition over a TiO2 photocatalyst to yield N2O and N2, and no other products. According to the views of Bowering and 452 Catalysts 2018, 8, 553 coworkers [109], use of only TiO2 as the catalyst is not ideal for removing NO from the atmosphere as N2O itself is also a regulated pollutant. As such, Tanaka et al. [105] reported that photoassisted selective catalytic reduction (photo-SCR) of NO over TiO2 with NH3 as a reductant was very selective towards N2 formation, with relatively small amounts of N2O. As NH3 is also a pollutant, it would need to be eliminated from the exhaust gas, thereby causing an increase in overall costs of a system. Figure 24. Comparative decrease of the NOx removal performed on TiO2 in the form of powder and pellets (plane surface, 54 cm2). Reproduced with permission from Toma et al. [108]. Copyright 2004 by Springer-Verlag (License No.: 4453250060579). Figure 24. Comparative decrease of the NOx removal performed on TiO2 in the form of powder and pellets (plane surface, 54 cm2). Reproduced with permission from Toma et al. [108]. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Copyright 2004 by Springer-Verlag (License No.: 4453250060579). Car exhaust and industrial emissions are mostly controlled using selective catalytic reduction (SCR) to convert NOx to N2. Accordingly, Bowering et al. [109] used CO as the reductant in eliminating NO photocatalytically with Degussa P-25 TiO2 in a continuous ﬂow reactor (Figure 25) with the objective to convert the NOx preferentially into N2 gas. The authors added TiO2 powder (ca. 0.2 g) to acidiﬁed triply deionized water (TDW; 6 mL of 0.05 M HNO3 in 500 mL of TDW) yielding a dispersion that was stirred for 12 h and then dried at 70 ◦C for 48 h, after which the resulting powders were calcined for 2 h at 120, 200, 450 or 600 ◦C. Subsequently, 25-mL fractions of the dispersion were evaporated at 70 ◦C onto degreased borosilicate glass slides; an amount of TiO2 powder (~ 1 mg) was deposited on the slides and then calcined following the same methodology as for the powders. Figure 25. Schematic of the photoreactor used for testing the photocatalytic behavior of the various P-25 TiO2 catalytic samples (see text). Reproduced with permission from Bowering et al. [109]. Copyright 2005 by Elsevier B.V. (License No.: 4452670211528). Figure 25. Schematic of the photoreactor used for testing the photocatalytic behavior of the various P-25 TiO2 catalytic samples (see text). Reproduced with permission from Bowering et al. [109]. Copyright 2005 by Elsevier B.V. (License No.: 4452670211528). 453 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 The effect of calcination temperatures on the composition and crystallite sizes of P-25 photocatalysts, and the effect of pretreatment temperature on rate of NO conversion and selectivity for N2 formation for NO decomposition and reduction reactions are presented in Table 4 [109]. Pretreatment (calcination) temperatures caused no appreciable change in phase composition; original composition (ca. 77 vol.% anatase, 23 vol.% rutile) was maintained even after treatment at 600 ◦C. The photocatalytic activity for both decomposition and reduction reactions decreased with increasing pretreatment temperature, which was attributed to removal of surface hydroxyl species that acted as active sites for reaction. The only products observed in the decomposition reactions were N2 and N2O; the selectivity for nitrogen formation remained constant (ca. 23%) regardless of pretreatment temperature. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) However, the presence of CO in the reaction gas had a dramatic effect on selectivity of the reactions with N2 selectivity as high as 65%; in addition, an increase in the CO/NO ratio led to increased selectivity for N2 formation. Table 4. Effect of calcination temperatures on the composition and crystallite sizes of P-25 photocatalysts, and of the pretreatment temperature on the rate of NO conversion and selectivity for N2 formation for NO decomposition and reduction reactions . Calcination T (◦C) Rutile (vol. %) Crystallite Size (nm) BET Surface Area (m2 g−1) Rate of NO Conversion (μmol h−1 gcat−1) Selectivity for N2 Formation (%) Anatase Rutile Dec. Red. Dec. Red. 70 23.0 28.0 66.3 51.92 1210 657 21 46 120 23.2 28.8 64.8 50.69 1107 560 25 49 200 23.0 28.2 62.7 49.87 983 467 21 48 450 23.0 28.3 59.6 49.54 550 243 26 26 600 28.0 30.5 69.6 48.24 430 240 30 25 Adapted with permission from Bowering et al. [109]; Copyright 2005 by Elsevier B.V. (License No.: 4452670211528). It is likely that under UV illumination electron transfer occurred from electron trapped centers into antibonding orbitals of adsorbed NO molecules, resulting in their decomposition and formation of N(ads) and O(ads) surface species, which can then scan the TiO2 surface and react with other surface species (e.g., NO(ads), N(ads), O(ads)) to form N2O, NO2, O2 and N2. To the extent that neither O2 nor NO2 was detected led the authors [109] to deduce that Reactions (69) and (70) did not occur on P-25 surfaces under decomposition conditions; the main surface reaction was Reaction (71), as N2O was the major reaction product under these conditions. NO(ads) + O(ads) →NO2(ads (69) O(ads) →O2(ads) (70) NO(ads) + N(ads) →N2O(ads) (71) NO(ads) + O(ads) →NO2(ads (69) O(ads) →O2(ads) (70) NO(ads) + N(ads) →N2O(ads) (71) (69) O(ads) →O2(ads) (70) NO(ads) + N(ads) →N2O(ads) (71) (71) In the presence of CO on the photocatalyst surface, and under UV illumination, other reactions are possible between adsorbed CO and NO molecules together with reactions of CO with N(ads) and O(ads) atoms. No reaction occurred in the dark and under UV illumination without TiO2 indicating that both TiO2 and UV are required for adsorbed NO and CO species to react. Under decomposition conditions, the major reaction product was N2O (~ 75%) with N2 being the minor product (ca. 25%). 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) On the other hand, under reduction conditions selectivity for N2 formation increased (ca. 48%) at the pretreatment temperatures of 70 and 120 ◦C. However, at higher pretreatment temperatures, the selectivity was similar to that achieved in the absence of CO suggesting that the surface N2 forming reaction was favored on a titanium surface rich in hydroxyl groups [109]. Germane to the previous study [109], Roy and coworkers [110] examined a photocatalytic route to destroy NOx by developing a new Pd ion-substituted TiO2 system (Ti1−xPdxO2−δ) with which to reduce NO in the presence of CO via creation of redox adsorption sites and using anion oxygen 454 Catalysts 2018, 8, 553 vacancies on titania; the optimal Pd2+ ion concentration was 1 at.% in TiO2 (anatase). Apparently, even though both NO and CO competed for the same Pd2+ adsorption sites, reduction of NO to N2O was two orders of magnitude higher with the Ti0.99Pd0.01O1.99 photocatalyst under ambient conditions than unsubstituted TiO2; using UV irradiation with a 125-W high-pressure Hg lamp and an inlet 5000 ppm of NO in a ﬂow-type reactor, the rate of reduction of NO was 0.53 μmol g−1 s−1. pp yp g The photocatalytic deNOxing activities of TiO2, N-doped TiO2, Fe-loaded N-doped TiO2, and Pt-loaded N-doped TiO2 exposed to irradiation from monochrome LED lamps at various wavelengths have been investigated in some detail by Yin and coworkers [111], unlike many studies that have typically used 100–500 Watt high-pressure Hg or otherwise Xe light sources to activate titania-based photocatalysts. Bare TiO2−xNx (denoted TiON) powders were prepared by treating 20-nm Ishihara ST-01 anatase TiO2 in an NH3 atmosphere at 600 ◦C for 3 h, followed by annealing at 300 ◦C for 2 h in humid air (N content, ca. 0.25 at.%); for comparison, the ST-01 anatase TiO2 powder was heat-treated in air at 500 ◦C for 1 h (S-TiO2)—BET speciﬁc surface areas were 57.7 m2 g−1 and 100.7 m2 g−1, respectively, for TiON and S-TiO2 powders. The Fe- and Pt-loaded N-doped TiO2 systems were prepared by dispersing TiON powder in a HNO3 aqueous solution containing either Fe(NO3)3 or Pt(NH3)2(NO3)2 at ambient temperature, followed by stirring for 1 h, heated at 150 ◦C to remove the water, and then calcined at 300 and 400 ◦C for 2 h, respectively (loading of Fe and Pt, 0.5 wt.%; systems denoted TiONFe and TiONPt; BET areas were 61.1 m2 g−1 and 59.0 m2 g−1, respectively). 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) The speciﬁcs of irradiation from the four LED sources were (wavelength, irradiance): (i) red light LED (627 nm, 72.76 μmol m−2 s−1); (ii) green light LED (530 nm, 125.12 μmol m−2 s−1); (iii) blue light LED (445 nm, 76.22 μmol m−2 s−1); and (iv) UV light LED (390 nm, 73.70 μmol m−2 s−1). Different samples showed different wavelength dependencies; for instance, S-TiO2 displayed excellent activity at 390 nm but very weak activity at 445 nm, while TiON showed excellent UV light (390 nm) and visible light-induced photocatalytic activity on exposure to 445 nm and 530 nm irradiation (Figure 26) [111]. By comparison, both TiONFe and TiONPt showed excellent deNOxing abilities even under red light irradiation at 627 nm. Speciﬁcally, TiONPt showed the highest deNOx abilities at all light wavelength ranges: about 37.8%, 36.8%, 28.2%, and 16.0% of NOx was removed, respectively, under continuous irradiation by monochromatic light at 390 nm (UV LED), 445 nm (blue LED), 530 nm (green LED), and 627 nm (red LED). Figure 26. Photocatalytic deNOx abilities of samples under irradiation by various LED light sources: (a) S-TiO2; (b) TiON; (c) TiONFe; and (d) TiONPt. Reproduced with permission from Yin et al. [111]. Copyright 2008 by the American Chemical Society. Figure 26. Photocatalytic deNOx abilities of samples under irradiation by various LED light sources: (a) S-TiO2; (b) TiON; (c) TiONFe; and (d) TiONPt. Reproduced with permission from Yin et al. [111]. Copyright 2008 by the American Chemical Society. 455 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 Because of the newly formed N2p level within the bandgap of titania above the O2p valence band, N-doped titania displayed an extrinsic bandgap smaller than the intrinsic bandgap of titania such that TiON absorbed visible light. In addition, TiON displayed signiﬁcant chemiluminescence emission attributed to formation of singlet oxygen 1O2 relative to undoped S-TiO2 which failed to display any light emission. For comparison, TiONFe and TiONPt also displayed relatively high chemiluminescence emission, albeit lower than TiON; the latter showed chemiluminescence intensity increases in the order: UV < blue < green < red. For the undoped S-TiO2 sample, the study of Yin et al. [111] showed a correlation between very weak chemiluminescence emission intensities and very low visible-light induced photocatalytic activity that they attributed to its relatively large intrinsic bandgap. By contrast, for TiON, TiONFe, and TiONPt, results demonstrated that the deNOx ability decreased with an increase in chemiluminescence emission intensity. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) 456 Catalysts 2018, 8, 553 3. Fe and Pt loading improved the photocatalytic activity of N-doped TiO2 under not only UV light but also long-wavelength visible-light irradiation (λ = 530 nm and λ = 627 nm). 3. Fe and Pt loading improved the photocatalytic activity of N-doped TiO2 under not only UV light but also long-wavelength visible-light irradiation (λ = 530 nm and λ = 627 nm). 4. Pt-loaded, N-doped titania possessed the best visible light- and UV-induced photocatalytic activity. In addition, Fe- and Pt-loaded N-doped titania exhibited relatively high quantum yields of deNOxing under long-wavelength LED light irradiation. In their extensive 2009 review article on the catalytic abatement of NOx in the environment, Roy and coworkers [112] focused mostly on thermal methods in the presence of suitable reducing agents, and brieﬂy gave a short account of the alternative photocatalytic methodology at ambient conditions; summarized was also some of the earlier work reported by selected researchers noting that direct photocatalytic decomposition of NO would yield N2 and O2, which would indeed be the ideal outcome and sole products if that could be realized. Unfortunately, as noted above, different conditions and different titania-based photocatalysts lead to signiﬁcantly different results that are worth recalling brieﬂy. For instance, 1. Anpo and coworkers showed that metal ion-implanted TiO2 decomposed NO photo- catalytically to N2, O2 and N2O at 275 K under irradiation with visible light at wavelengths longer than 450 nm [96]. 2. Lim et al. [102] found that the photocatalytic decomposition of NO over Degussa P-25 TiO2 in an annular ﬂow type reactor produces NO2, N2O and N2, with the efﬁciency increasing with light intensity and residence time and decreasing with initial NO concentration. 2. Lim et al. [102] found that the photocatalytic decomposition of NO over Degussa P-25 TiO2 in an annular ﬂow type reactor produces NO2, N2O and N2, with the efﬁciency increasing with light intensity and residence time and decreasing with initial NO concentration. 3. Bowering et al. [109] showed that the photocatalytic activity of Degussa P-25 TiO2 toward deNOxing decreased with increasing pretreatment temperature. 3. Bowering et al. [109] showed that the photocatalytic activity of Degussa P-25 TiO2 toward deNOxing decreased with increasing pretreatment temperature. 4. Roy and coworkers [110] reported that reduction of NO over the catalyst Ti1−xPdxO2−δ was two orders of magnitude greater than unsubstituted TiO2. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Nonetheless, the photocatalytic deNOx activity of TiON was nearly the same as that of S-TiO2 under 390 nm (UV LED) and 627 nm (red LED) irradiation. However, TiON exhibited greater activity than S-TiO2 under 445 nm and 530 nm irradiation, but lower than TiONFe and TiONPt samples under every type of LED light irradiation. The authors ascribed this variation to different band structures and to the presence of Fe and Pt loaded onto the surface of the TiON. Mechanistically, the sequence of events that led to deNOxing by these four titania samples was summarized [110] by the series of Reactions (72)–(77). Subsequent to irradiation of the titania that yields conduction band electrons (e−) and valence band holes (h+), formation of singlet oxygen 1O2 (Equation (76)) competes with formation of superoxide radical anions (Equation (73)) and hydroxyl radicals (Equation (74)) in air (molecular oxygen; relative humidity, ca 25%). Fe and Pt loading on TiON increased charge transfer and charge separation on the surface of the TiONFe and TiONPt photocatalysts [111]. TiO2 + hν →e−+ h+ (72) e−+ O2 →O2−• (73) O2−• + H+ →HOO• (74) H2O + h+ →•OH + H+ (75) O2−• + h+ →1O2 (76) O2−• + h+ →3O2 (77) (72) (77) In the present context, deNOxing reportedly occurred by oxidation of the NOx molecules via the active oxygen •OH and O2−• species (Equations (78)–(80)), whereby NO is converted to NO2 and subsequently to NO3−ions. NO + 2 •OH →NO2 + H2O (78) NO2 + •OH →NO3−+ H+ (79) NO + O2−• →NO3− (80) (78) (79) (80) In a NOx atmosphere, the NOx molecules adsorb onto the photocatalyst’s surface and then interact with the superoxide radical anions O2−• to form NO3−(Equation (80)), as a result of which the NOx molecules consume O2−• and delay singlet oxygen formation (Equation (76)), thereby causing the chemiluminescence emission intensity in the NOx atmosphere to be much lower than in air. To recapitulate, Yin and coworkers [111] deduced that: 1. Nanosized titania exhibited very low deNOx ability under visible light irradiation, irrespective of their excellent UV light-induced (390 nm) deNOx ability. 1. Nanosized titania exhibited very low deNOx ability under visible light irradiation, irrespective of their excellent UV light-induced (390 nm) deNOx ability. 2. N-doped titania displayed excellent photocatalytic activity under 445 nm and 530 nm light irradiation. 2. N-doped titania displayed excellent photocatalytic activity under 445 nm and 530 nm light irradiation. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Schematic ﬂow diagram of the hybrid PCR + HC/SCR reactor system. Reproduced with permission from Heo et al. [114]. Copyright 2013 by the American Chemical Society. Figure 27. Schematic ﬂow diagram of the hybrid PCR + HC/SCR reactor system. Reproduced with permission from Heo et al. [114]. Copyright 2013 by the American Chemical Society. Table 5. Effect of catalysts on the performance of the PCR system at 150 ◦C. Feed gas composition: 200 ppm NO, 134 ppm C12H26, 6% O2, 10% H2O, and N2 balance. Total ﬂow rate = 500 mL min−1; TiO2 (anatase nanopowder) . Catalyst Yield of Reductants (%) NOx Conversion (%) 2 BaY + Total OHCs NH3 TiO2 35 (84 a) - 43 V2O5/TiO2 60 - 29 Au/TiO2 24 - 43 Ag/TiO2 18 - 55 Pt/TiO2 2 - 22 CuCoY 19 - 37 BaY 10 - 55 Ag/Al2O3 16 6 54 Blank PCR (no catalyst) 45 - - a Yield of OHCs in the absence of NO. Reproduced with permission from Heo et al. [114]. Copyright 2013 by the American Chemical Society. a Yield of OHCs in the absence of NO. Reproduced with permission from Heo et al. [114]. Copyright 2013 by the American Chemical Society. To the extent that the deNOx performance of the conventional HC/SCR catalyst was enhanced by both OHCs and NHx-containing reductants, the two representative PCR catalysts V2O5/TiO2 and Ag/Al2O3 were chosen by the authors [114] for further examination in the PCR + HC/SCR hybrid system: the V2O5/TiO2 PCR for its superior OHC formation and the Ag/Al2O3 PCR for its formation of NH3 and possibly organo-nitrogen compounds as precursors of NH3. Table 6 lists the conversions of NO and NOx, the conversion of NOx to N2, and the yields of NO2, N2O, and NH3 during the reduction of NO with the PCR + HC/SCR system [114]. The NOx conversion to N2 was estimated from the conversion of NOx and the yields of NO2, N2O, and NH3 by the mass balance of nitrogen. The selectivity of the Ag/Al2O3 PCR + HC/SCR system for N2 was 94%, 91%, and 82% at 200, 250, and 300 ◦C, respectively. The slight decrease N2 in selectivity of the Ag/Al2O3 PCR + HC/SCR system at 300 ◦C was ascribed to increased formation of NH3 by reaction of NO with OHCs over the HC/SCR reactor, since the PCR readily converted dodecane to OHCs. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) Direct NO decomposition into N2 and N2O occurred via dissociation of NO in the presence of UV radiation at room temperature yielding N2, N2O and O2 with the O2 evolved reacting with NO to give NO2 that is adsorbed by the catalyst upon formation. Prolonged NO2 adsorption makes the surface inactive for NO dissociation; NO dissociation resumed when CO was passed to scavenge the evolved dissociated O2 [110]. On the other hand, the seminal review article by Skalska et al. [113] presented an extensive survey of NOx emission control technologies for three major anthropogenic emission sources: power plants, vehicles and the chemical industry, and further described new and alternative methods such as a hybrid system of SCR (selective catalytic reduction) and O3 injection, fast SCR, and electron beam gas treatment, among others. Also described was the inﬂuence of NOx on the environment and human health. The main focus was put on NOx control methods applied in the combustion of fossil fuels in power stations and mobile vehicles, together with methods used in the chemical industry; the authors emphasized the implementation of ozone and other oxidizing agents in NOx oxidation. Following these footsteps, Heo and coworkers [114] combined photocatalysis and SCR with hydrocarbons as reducing agents (HC/SCR) to improve the activity and durability of deNOx catalysts. The authors developed a photocatalytic HC/SCR system that exhibited high deNOx performance (54.0−98.6% NOx conversion) at low temperatures (150−250 ◦C) using dodecane as the HC reductant over a hybrid SCR system that included a photocatalytic reactor (PCR) and a dual-bed HC/SCR reactor (Figure 27). The PCR generated the highly active oxidants O3 and NO2 from O2 and NO in the feed stream, followed by subsequent formation of the highly efﬁcient reducing oxygenated hydrocarbon (OHC), NH3, and organo-nitrogen compounds. These reductants were key in enhancing the low-temperature deNOx performance of the dual-bed HC/SCR system containing Ag/Al2O3 and CuCoY in the front and rear bed of the reactor, respectively (Table 5). Moreover, the OHCs proved particularly effective for both NOx reduction and NH3 formation over the Ag/Al2O3 catalyst, while NH3 and organo-nitrogen compounds were effective for the reduction of NOx over CuCoY. The photocatalytic assisted hybrid HC/SCR system demonstrated an overall deNOxing performance 457 Catalysts 2018, 8, 553 comparable to that of the NH3/SCR, thus its potential as a promising alternative to the current urea/SCR technology [114]. Figure 27. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Energy-related emissions of nitrogen oxides continue to increase worldwide, standing close to 110 Mt in 2015 with the transportation sector accounting for 52%, followed by industry (26%) and power generation (14%). China (23 Mt) and the United States (13 Mt) accounted for ca. 33% of global NOx emissions that year. According to the International Energy Agency [115], power generation in 2015 was a major source of worldwide emissions of nitrogen oxides (14% of total NOx) with coal being the principal fuel responsible for 70% of those NOx emissions; burning oil to generate electricity also produced signiﬁcant quantities of NOx. Natural gas-ﬁred plants emitted fewer air pollutants than either coal-ﬁred or oil-ﬁred power plants; however, in 2015, gas-ﬁred power generation emitted close to 20% of NOx, while biomass played a negligible role in global power generation, although, in relative terms, it performed only slightly better than coal-ﬁred plants for NOx emissions. Manufacturing industries and other transformation sectors (e.g., reﬁning and mining) accounted for ca. 30% of NOx (28 Mt) in 2015. Process-related NOx emissions were mostly released in cement making (1.5 kt of NOx per Mt of cement that accounted for >50% of global process-related NOx emissions) followed by pulp and paper production (1.2 kt NOx Mt−1 of paper) (see Figure 28) [115]. Considering combustion and process emissions from a regional perspective, China was the largest emitter of NOx, accounting for nearly 40% followed by the United States (11%). However, the United States witnessed considerable decreases in NOx emissions in decades prior to 2015, while NOx emissions from Chinese and Indian industries increased signiﬁcantly. Together with ammonia, NOx and SO2 are the main precursors to formation of acid rain, which affects soil and water (with adverse impact on vegetation and animal life) and accelerates the deterioration of equipment and cultural heritage [115]. The presence of NOx and volatile organic compounds (VOCs) in the environment leads to formation of ground-level ozone (O3) under sunlight. Thus, the considerable research interest witnessed over the last 2–3 decades to attenuate the extensive presence of NOx in the environment is not surprising. 4. Applied Photocatalysis: Laboratory-Scale deNOxing of NOx Agents (NO & NO2) At 400 ◦C, the HC/SCR system alone completely reduced NOx with high N2 selectivity up to 96% so that the PCR could be turned off and bypassed to save energy at temperatures above 400 ◦C. 458 Catalysts 2018, 8, 553 Table 6. Conversion of NO and yields of NO2, N2O, and NH3 during the reduction of NO over the PCR + HC/SCR system. Feed gas composition: 200 ppm NO, 134 ppm C12H26, 6% O2, 10% H2O, and N2 balance. Gas hourly space velocity of the HC/SCR monolith reactor = 16,500 h−1 . System Temperature (◦C) NO (%) NO2 (%) N2O (%) NH3 (%) Total NO (%) Estimated N2 (%) HC/SCR only 256 13 0 1 0 13 13 303 62 0 1 1 62 60 400 100 0 3 2 100 96 500 53 7 1 1 46 44 Ag/Al2O3 PCR + HC/SCR 200 63 0 4 0 63 59 250 99 0 5 4 99 91 300 100 0 5 13 100 82 V2O5/TiO2 PCR + HC/SCR 200 42 0 4 1 42 38 250 95 0 5 0 95 91 Reproduced with permission from Heo et al. [114]. Copyright 2013 by the American Chemical Society. Key to the successful demonstration of this advanced deNOx process was the unique design and functionality of the PCR, which led to three major conclusions [114]: (1) PCR with catalysts was very efﬁcient for both OHC formation and reduction of NOx because of its dual function: in situ UV-induced formation of OHC and conversion of NOx over the catalysts; (2) blank PCR (no catalyst) was very efﬁcient for oxidation of NO to NO2 and HC to OHC, but was inefﬁcient for converting NOx because of the absence of a catalyst; and (3) Ag/Al2O3 PCR (with BaY + Ag/Al2O3) produced OHC and NH3 as intermediates that could be used subsequently to further convert NOx in a downstream reactor containing a dual-bed catalyst such as Ag/Al2O3 (for OHC/SCR) and CuCoY (for NH3/SCR). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment In a 1999 Technical Bulletin, the United States Environmental Protection Agency (EPA) [116] described the various components that make up the NOx pollutants, together with their properties, some of the health concerns, and how the environmental NOx could be abated and controlled by external combustion—pollution prevention methods and add-on control technologies—that is, by non-photocatalytic technologies. In this regard, 459 Catalysts 2018, 8, 553 methods to reduce thermally the NOx emissions at the origins include improved combustion techniques (e.g., fuel denitrogenation, modiﬁcation to combustion methods, modiﬁcation of operating conditions, and tail-end control processes) and installation of low-NOx burners in process heaters and industrial heat and electricity generation plants. Another primary combustion technology is a ﬂuidized-bed combustion technology for solid fuels, while end-of-pipe technologies focus on the removal of NOx from ﬂue gases by means of either physical separation or chemical reactions before their release to the atmosphere. Selective catalytic reduction (SCR; NH3 as the reductant; presence of a combination of TiO2 and V/W oxides as catalysts (Equations (81) and (82)) or selective non-catalytic reduction systems (SNR; urea or ammonia (Equations (83) and (84)) can signiﬁcantly reduce NOx in the ﬂue gas of stationary sources into N2 and H2O. methods to reduce thermally the NOx emissions at the origins include improved combustion techniques (e.g., fuel denitrogenation, modiﬁcation to combustion methods, modiﬁcation of operating conditions, and tail-end control processes) and installation of low-NOx burners in process heaters and industrial heat and electricity generation plants. Another primary combustion technology is a ﬂuidized-bed combustion technology for solid fuels, while end-of-pipe technologies focus on the removal of NOx from ﬂue gases by means of either physical separation or chemical reactions before their release to the atmosphere. Selective catalytic reduction (SCR; NH3 as the reductant; presence of a combination of TiO2 and V/W oxides as catalysts (Equations (81) and (82)) or selective non-catalytic reduction systems (SNR; urea or ammonia (Equations (83) and (84)) can signiﬁcantly reduce NOx in the ﬂue gas of stationary sources into N2 and H2O. ELRPDVV JDV FRDO RLO NWSHU0WRH Figure 28. Global average NOx emissions from various sources in 2015 in kilo tonnes (kt) per million tonnes oil equivalent (Mtoe). Plot made from selected data from the International Energy Agency [115]. ELRPDVV JDV FRDO RLO NWSHU0WRH NWSHU0WRH Figure 28. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Usage of TiO2-based photocatalytic material on roadway surfaces to convert nitrogen oxides NOx to nitrate: ﬁnished roadway with the coated surface showing a lighter color—photo was courtesy of Fujita Road Construction Co., Ltd. to the authors of Ref. [117]. Reproduced with permission from Fujishima et al. [117]. Copyright 2008 by Elsevier B.V. (License No.: 4452671501245). Figure 29. Usage of TiO2-based photocatalytic material on roadway surfaces to convert nitrogen oxides NOx to nitrate: ﬁnished roadway with the coated surface showing a lighter color—photo was courtesy of Fujita Road Construction Co., Ltd. to the authors of Ref. [117]. Reproduced with permission from Fujishima et al. [117]. Copyright 2008 by Elsevier B.V. (License No.: 4452671501245). The 2013 article by Hanus and Harris [118] entertained some innovations of nanotechnology for the construction industry, most noteworthy being improvements in concrete strength, durability and sustainability being achieved with use of metal/metal-oxide nanoparticles and engineered nanoparticles (carbon nanotubes and carbon nanoﬁbers), as well as environment-responsive anticorrosion coatings formed using nano-encapsulation techniques. The 2013 article by Hanus and Harris [118] entertained some innovations of nanotechnology for the construction industry, most noteworthy being improvements in concrete strength, durability and sustainability being achieved with use of metal/metal-oxide nanoparticles and engineered nanoparticles (carbon nanotubes and carbon nanoﬁbers), as well as environment-responsive anticorrosion coatings formed using nano-encapsulation techniques. For their part, Fresno and coworkers [119] described achievements, near-future trends and critically assessed many photocatalytic materials on the basis of knowledge accumulated in pre-2014 years as to which materials or multicomponent systems, among the multitude of developments, could be taken as a ready consolidated technology or else as more likely to become a real alternative in the short term. Germane to this, they noted that a photocatalyst could be incorporated during material manufacturing either as an additive, most often the case of construction materials, or as a coating on an already conformed cementitious surface with the mechanical resistance of the coating, the optimal amount of photocatalyst (TiO2) and binder, their impact on the properties of the materials and their long-term performance and aesthetic durability being factors to consider in their applications as carbonation can lead to deactivation after several months of use, a point we shall emphasize later with regard to a most celebrated example: the Dives in Misericordia church in Rome (see below) [120] built with Italcementi’s TX-Active® photocatalytic concrete [121] for self-cleaning purposes and for reduction of NOx pollution (among others). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Global average NOx emissions from various sources in 2015 in kilo tonnes (kt) per million tonnes oil equivalent (Mtoe). Plot made from selected data from the International Energy Agency [115]. Reduction of NOx in the environment may not only involve SCR and SNR technologies, but also and of particular relevance herein is the TiO2-based photocatalytic technology that is the subject of this section. (a) Selective Catalytic Reduction (SCR) elective Catalytic Reduction (SCR) 4 NO + 4 NH3 + O2 + Cat. →4 N2 + 6 H2O (81) 2 NO2 + 4 NH3 + O2 + Cat. →3 N2 + 6 H2O (82) (81) (82) Conditions: Temperature: 300–400 ◦C; typical efﬁciencies: about 80% (b) Selective Non-catalytic Reduction (SNR) (b) Selective Non-catalytic Reduction (SNR) (b) Selective Non-catalytic Reduction (SNR) 4 NO + 4 NH3 + O2 →4 N2 + 6 H2O (83) 4 NO + 2 (NH2)2CO + O2 →4 N2 + 4 H2O + 2 CO2 (84) (83) (84) Conditions: Temperature: 900–1000 ◦C; 40–60% reduction is obtained. Several review articles have appeared in the last decade [117–119] that described, among others, some of the early attempts in the abatement of NOx agents under indoor and outdoor experimental conditions. The 2008 article by Fujishima et al. [117] offered an overview of some highlights of TiO2 photocatalysis, reviewed some of its origins, and indicated some useful applications: self-cleaning surfaces, water puriﬁcation, air puriﬁcation, self-sterilizing surfaces, anti-fogging surfaces, heat dissipation and heat transfer, anticorrosion applications, environmentally friendly surface treatment, photocatalytic lithography, photochromism of metal oxides, and microchemical systems. In the air puriﬁcation application (i.e., deNOxing), the 2008 article brieﬂy noted that some Japanese companies were considering covering roads with the TiO2 photocatalyst, and removing the NOx from automobile exhaust with sunlight using TiO2-coated road bricks, prepared by mixing 460 Catalysts 2018, 8, 553 colloidal TiO2 solutions with cement: the photo-road technology applied to no less than 14 different locations in Japan, one of which was the 7th belt highway in Tokyo (surface area covered, ca. 300 m2); NOx removed from this testing area was ca. 50–60 mg per day, equivalent to NOx discharged by 1000 automobiles (Figure 29); however, no relative efﬁciency was provided with respect to total NOx in the environment. Figure 29. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Incorporation of anatase titania into the wearing layer, and the use of a double-layered concrete with addition of the photocatalyst to the top layer were two of the possible strategies noted for this application to reduce NOx. Most importantly, the authors [119] emphasized the need for further assessments of the durability of the photoactive coatings, and their capability of abating other air pollutants (e.g., the VOCs). 461 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 Along these lines, the discussion below will ﬁrst emphasize the results from TiO2-based photocatalytic deNOxing the environment with the photocatalyst deposited on cementitious substrates in an indoor laboratory setting using small photoreactors set in ﬂow-through systems and UV light sources. This is then followed by deNOxing results from various outdoor settings in wide open air environments with sunlight as the light source to activate the photocatalytic surfaces. 5.1. Indoor DeNOxing Environment Tests with TiO2 Photocatalytic Cementitious Surfaces One of the ﬁrst articles in the search for means to reduce NOx from the various emission sources, which affected the air quality in Japan because of relatively high concentrations of NOx, particularly along heavily trafﬁcked roads in densely populated areas, was that of Ibusuki and Takeuchi in 1994 [94]. They proposed a new non-thermal technology for removing NOx at sub-ppm level from the air of trafﬁcked roads, tunnels and other environmental emission sources (see Section 4) following their earlier ﬁndings [91] that NOx deposited onto the ground caused an enhancement of the rate of adsorption of NOx on soil particles upon photoillumination, with TiO2 exhibiting the highest activity for NOx adsorption—about 60% of NOx was captured as nitric acid on the TiO2 surface. Their follow-up laboratory experiments indicated that up to 90% of NO could be removed using mixtures of TiO2, activated carbon (AC), and iron oxide (Fe2O3) particulates in a ﬂow-through reactor [96]. p g Along similar lines, at the 2000 JIPEA World Congress, Murata and coworkers [122] reported the development of interlocking cementitious bricks loaded with TiO2 (NOXER©) for the oxidative removal of NOx under sunlight UV radiation, humidity, and NOx concentrations that paralleled roadside environments. This novel technology at the time was implemented in a couple of Tokyo suburbs. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Simulated emission gas consisted of mixing dry air, dilute NO gas (1.0 ppm); ﬂow rate was 3 L min−1; air space between the brick and the Pyrex glass window was 5 mm; temperature, 25 ◦C; relative humidity RH, 50%; UV intensity from 2 black lights (300–400 nm), 6 W m−2. Reproduced from Murata et al. [122]. Evidence of the effectiveness of NOx abatement was also carried out by the Italcementi Group using their TX-Active® photocatalytic TiO2 deposited on a suitable substrate [121,123] and subsequently placed in a small reactor (top in Figure 32 [121]) of predetermined volume in which NOx gas was introduced, and then diluted with air to reach a certain pollutant concentration; the schematic of the ﬂow-through reactor assembly is also displayed (bottom of Figure 32). The efﬁcacy against NOx gases was also demonstrated during the project PICADA with tests conducted at the European Laboratory of Ispra (Italy), inside an Indoortron—an environmental chamber with people access and characterized by such controlled parameters as temperature, relative humidity, air quality and air exchange rate—to also study the fate of various other internal VOC contaminants. Figure 33 illustrates the time course of the removal of NOx with the reactor assembly of Figure 32. Evidently, complete NOx removal under the ISO conditions occurred within ca. 6 h of light (UV) irradiation in the presence of TX Active® in the absence of which no changes in NOx concentration occurred. Tests on the best formulations of a white and a grey photocatalytic paint with TX Active®, chosen for tunnel renovation (see below), showed a NOx abatement capacity of 88–90% after only 60 min under UV light irradiation in the reactor assembly of Figure 32; however, under similar conditions, the same paints necessitated nearly 26 h of UV irradiation to decolorize 70% of a rhodamine-B dye stain on the paint’s surface [124]. An otherwise similar reactor assembly was used by Martinez and coworkers [125] to examine the degradation of NO present in the air by means of a photocatalytic oxidation process based on TiO2 nanoparticles that had been incorporated in a polymer-matrix-based coating. The experimental setup consisted of a ﬂow type reactor (Figure 34) adapted from the ISO 22197-1 standard; the ﬁnal products detected were NO2 in the gas phase and nitrate ions adsorbed on the photocatalytic surface. The photocatalyst (anatase-TiO2) was a commercial slurry solution available from Evonik (Aerodisp® W740X). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The permeable interlocking bricks (Figure 30) were prepared by mixing aggregates, cement, TiO2 powder and water in an appropriate template, and then cured at ambient temperature for one month, after which they were used for the photocatalytic indoor tests in a ﬂow-through small PVC reactor (conditions in Figure 31). After 12 h of UV irradiation, ca. 88% of NO was removed at a relative humidity (RH) of 10%, decreasing to 52% at 80% RH, whereas the quantity of NO removed increased with increase in UV intensity (from 10% at 0 W m−2 to 88% at 12 W m−2). Varying the NO concentration from 0.05 to 5.0 ppm showed that in the 0.05–0.15 ppm range of NO—a range similar to roadside levels—a constant removal of ca 90% was observed decreasing to ca. 45% at 5 ppm of NO [122]. In an outdoor test in which NOx from the roadside (ca. 0.5 ppm) was passed through the PVC reactor exposed to natural sunlight (T = 17 ◦C and RH = 47%) led to ca. 80% decrease in NOx. Figure 30. Paving brick for NOx removal; dimensions of brick: 10 cm × 20 cm; thickness of the surface layer containing TiO2: 5–7 mm. Reproduced from Murata et al. [122]. Figure 30. Paving brick for NOx removal; dimensions of brick: 10 cm × 20 cm; thickness of the surface layer containing TiO2: 5–7 mm. Reproduced from Murata et al. [122]. Figure 30. Paving brick for NOx removal; dimensions of brick: 10 cm × 20 cm; thickness of the surface layer containing TiO2: 5–7 mm. Reproduced from Murata et al. [122]. 462 Catalysts 2018, 8, 553 Figure 31. Indoor ﬂow-through PVC reactor system for evaluating the performance of a brick in removing NOx from the environment. Simulated emission gas consisted of mixing dry air, dilute NO gas (1.0 ppm); ﬂow rate was 3 L min−1; air space between the brick and the Pyrex glass window was 5 mm; temperature, 25 ◦C; relative humidity RH, 50%; UV intensity from 2 black lights (300–400 nm), 6 W m−2. Reproduced from Murata et al. [122]. Figure 31. Indoor ﬂow-through PVC reactor system for evaluating the performance of a brick in removing NOx from the environment. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment = 60 mm; length = 300 mm; high transparency to UV-A radiation; low adsorption capacity) used by Martinez and coworkers [125] to test the photocatalytic TiO2 coatings toward the abatement of NOx pollutants; coated and control samples (100 × 50 mm2) were placed in the median plane of the reactor using a PTFE holder; gas circulated through the semi-cylindrical space between test piece and upper part of the reactor. Light source was a 300-W OSRAM Ultravitalux bulb with an emission spectrum close to that of daylight (light intensity = 5.8 W m−2). Other experimental conditions: ﬂow rate, 1.5 L min−1; initial NO concentrations, 400–2000 ppb; relative humidity, 0–74%; temperature, 25 ◦C). Reproduced with permission from Martinez et al. [125]. Copyright 2011 by Elsevier Ltd. (License No.: 4452680979256). Figure 34. Cylindrical reactor (borosilicate-glass; dia. = 60 mm; length = 300 mm; high transparency to UV-A radiation; low adsorption capacity) used by Martinez and coworkers [125] to test the photocatalytic TiO2 coatings toward the abatement of NOx pollutants; coated and control samples (100 × 50 mm2) were placed in the median plane of the reactor using a PTFE holder; gas circulated through the semi-cylindrical space between test piece and upper part of the reactor. Light source was a 300-W OSRAM Ultravitalux bulb with an emission spectrum close to that of daylight (light intensity = 5.8 W m−2). Other experimental conditions: ﬂow rate, 1.5 L min−1; initial NO concentrations, 400–2000 ppb; relative humidity, 0–74%; temperature, 25 ◦C). Reproduced with permission from Martinez et al. [125]. Copyright 2011 by Elsevier Ltd. (License No.: 4452680979256). Using a speciﬁcally-developed test apparatus (Figure 35), Staub de Melo and Triches [126] assessed the efﬁciency of a photocatalytic mortar under no less than 27 different environmental conditions: varying the relative air humidity (30%, 50% and 70%), the UVA radiation (10, 25 and 40 W m−2), pollutant mass ﬂow rate (1, 3 and 5 L min−1), and initial concentration of NO (20 ppmv). Results showed that the higher were the levels of UVA radiation, the better was the performance of the mortar in degrading NOx. By contrast, at higher relative humidity levels and ﬂow rate caused a decrease in photocatalytic activity, which showed that environmental conditions have a signiﬁcant impact on the efﬁciency of the photocatalytic mortar in the degradation of NOx. (a) (b) Figure 35. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The coatings were primarily formulated for the surface treatment of building materials. To identify the possible inﬂuence of the nature of the substrate on the photocatalytic efﬁciency, various types of substrates were tested. The coatings were applied to: (i) mortars; (ii) glass plates; and (iii) non-absorbent cardboard materials. The wet thickness of coatings was 40 μm (other conditions are reported in Figure 34). 463 Catalysts 2018, 8, 553 Figure 32. (top) Indoor reactor used to measure NOx abatement using the TX-Active photocatalytic TiO2–based cementitious substrate Photograph reproduced from Borgarello Ref. [121]. (bottom) Schematic of the ﬂow-through reactor assembly; reproduced from Guerrini and Peccati [123]. Figure 32. (top) Indoor reactor used to measure NOx abatement using the TX-Active photocatalytic TiO2–based cementitious substrate Photograph reproduced from Borgarello Ref. [121]. (bottom) Schematic of the ﬂow-through reactor assembly; reproduced from Guerrini and Peccati [123]. QR7;$FWLYH 7;$FWLYH ,UUDGLDWLRQWLPHPLQ 12[ QR7;$FWLYH 7;$FWLYH ,UUDGLDWLRQWLPHPLQ 12[ Figure 33. Plots illustrating the immediate destruction of NOx upon turning the light on and after 60 min of lamp stabilization of the chamber (recirculation tests). Plot made from data reported in Borgarello [121]. ,UUDGLDWLRQWLPHPLQ Figure 33. Plots illustrating the immediate destruction of NOx upon turning the light on and after 60 min of lamp stabilization of the chamber (recirculation tests). Plot made from data reported in Borgarello [121]. Results indicated that irradiation for 60 min and relative humidity of 60% led to ca. 25% of an initial concentration of 8.61 μmol of NO to be degraded at a ﬂow rate of 1.5 L min−1 [125]. On mortar and glass substrates, the inﬂuence of increasing humidity on the degradation rates depended on the nature of the substrate and on initial NO concentrations; no signiﬁcant inﬂuence of humidity was observed at initial NO concentrations of 400 and 1000 ppb, while a signiﬁcant decrease in the kinetics 464 Catalysts 2018, 8, 553 was seen with a decrease of humidity at higher initial concentrations of NO (1500 and 2000 ppb). Generation of NO2 on mortar was very low, because of good adsorption capacities of the supporting substrate. On glass, NOx degradation rates decreased strongly on generation of NO2 owing to competition between pollutant and humidity (water) for the adsorption sites [125]. Figure 34. Cylindrical reactor (borosilicate-glass; dia. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment (a) Photoreactor in a ﬂow-through assembly for the measurement of photocatalytic activity of a photocatalytic mortar for the abatement of NOx; and (b) photocatalytic mortar applied to a Precast Concrete Paving sample. Reproduced with permission from Staub de Melo and Triches [126]. Copyright 2012 by Elsevier Ltd. (License No.: 4452681429886). (a) (b) (b) (a) Figure 35. (a) Photoreactor in a ﬂow-through assembly for the measurement of photocatalytic activity of a photocatalytic mortar for the abatement of NOx; and (b) photocatalytic mortar applied to a Precast Concrete Paving sample. Reproduced with permission from Staub de Melo and Triches [126]. Copyright 2012 by Elsevier Ltd. (License No.: 4452681429886). The photocatalytic mortar was produced using Portland cement with Pozzolan (CP II Z 32); the catalyst was a nanometric rutile TiO2 bar (dia. = 10 nm; length = 40 nm; 98% purity; speciﬁc surface 465 Catalysts 2018, 8, 553 area = 150 ± 10 m2 g−1; real density = 4.23 g cm−3). Mortars with addition of 3%, 6% and 10% TiO2 were investigated; layers of photocatalytic mortar with thicknesses of 3, 6 and 10 mm were applied to samples of precast concrete paving (PCP). The gas system consisted of dry air and 500 ppmv of NO stabilized with N2 gas, which simulated a polluted atmosphere. Maximal removal of NOx was 50% for an initial 20 ppmv of NO, relative humidity of 50%, a ﬂow rate of 1.0 L min−1 and a UVA irradiation (10 W m−2) period of 25 min. A cost to photocatalytic efﬁciency evaluation led to a 3-mm coated mortar incorporating 3% of TiO2. For the application of such materials, the authors [126] suggested that locations with lower relative humidity, high incidence solar radiation and little air mass movement should be sought in the ﬁeld, as they would provide better conditions to achieve high efﬁciency of the TiO2-coated precast concrete paving materials. An examination of the past literature shows that different types of photoreactors have been used in standardization methods to quantify the activity of photocatalysts in air remediation with commercially available photocatalytic materials. Classically, the degradation of NOx has been a major subject of investigations because relatively simple and inexpensive chemiluminescence instruments are available to quantify NOx and because NO2 is of crucial importance for urban air quality. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Nonetheless, when investigating the photodegradation of NOx in laboratory settings, only NO was used because of its facile detection, its lower ability to adsorb on reactor surfaces, and because of slower dark reactions that might occur on photocatalytic surfaces [127]. To reduce the time for establishing adsorption equilibrium and to increase precision of the NOx data from low-sensitive instruments, unrealistically high NO concentrations (500–1000 ppbv) have been commonly used; and as analyses of reaction products require more sophisticated instrumentation, no reaction products (e.g., nitrite and nitrate) other than NO2 are quantiﬁed. To the extent that NO2 is an intermediate in the photocatalytic oxidation of NO by reaction with O2−•/HOO• radicals, the photocatalytic removal of NOx, not just NO, is quantiﬁed by different standardization methods that are still under development (as we speak). In this regard, Ifang and coworkers [127] demonstrated that transport limitations can lead to an underestimation of the activity, if fast heterogeneous reactions were investigated in bed photoreactors. When using stirred tank photoreactors, complex secondary chemistry can lead to an overestimation of the photocatalytic remediation of NOx, if NO2 were also present, not to mention that the quantities used for ranking the activity of photocatalysts in air remediation in the different methods currently used are not independent of experimental conditions, so that any inter- comparison between different methods or extrapolation to atmospheric conditions is a futile exercise. Consequently, the authors [127] proposed a modiﬁed method for quantifying air remediation activity of photocatalytic surfaces that would overcome such problems. The method is based on a bed ﬂow reactor (Figure 36) that can easily be adapted to the ISO method. The extent of degradation of NOx on photocatalytic surfaces in continuous stirred tank reactors can be signiﬁcantly inﬂuenced: (a) by the gas-phase Leigthon chemistry (Equations (85)–(87)); (b) by unwanted wall losses of reactive agents and products; and (c) by heterogeneous formation of products on reactor walls (e.g., HONO) [127]. In such a photoreactor, short reaction times of only a few seconds and more homogeneous inert surfaces (no fan, among others) should minimize the aforementioned issues. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment NO2 + hν →NO + O(3P) (85) O(3P) + O2 →O3 (86) NO + O3 →NO2 + O2 (87) (85) (87) The bed ﬂow photoreactor was constructed of a single block of Teﬂon in which photocatalytic samples up to 40 cm × 5 cm × 1 cm could be investigated; recommended light sources were two 20-W UVA ﬂuorescence lamps (300–500 nm, λmax = 370 nm; length, 57 cm) mounted at variable distances to the reactor to adjust the irradiance level measured by a calibrated spectroradiometer [127]. Because of differences in photoactivity of commercial photocatalysts toward NOx (and VOCs), the authors suggested that at least one compound from each class be examined in standardization methods for 466 Catalysts 2018, 8, 553 air remediation. Moreover, to avoid saturation problems, laboratory experiments would have to be performed under relevant atmospheric conditions (i.e., for RH = 50%; reactant concentrations ≤100 ppb), and, as NO2 is of much greater environmental importance compared to NO, the use of NO2 as the test reactant was strongly recommended [127]. Figure 36. Modiﬁed bed ﬂow photoreactor with movable injector and turbulence barriers. Reproduced with permission from Ifang et al. [127]. Copyright 2014 by Elsevier Ltd. (License No.: 4452690260303). Figure 36. Modiﬁed bed ﬂow photoreactor with movable injector and turbulence barriers. Reproduced with permission from Ifang et al. [127]. Copyright 2014 by Elsevier Ltd. (License No.: 4452690260303). On their part, Zouzelka and Rathousky [128] investigated the photocatalytic activity of two commercial titania-based products: (1) Protectam FN2 that consisted of ca. 74% of Evonik’s Aeroxide P-25 TiO2 powder and 26% of an inorganic binder; and (2) Aeroxide P-25 TiO2 powder as photocatalytic coatings (10 μm thick) on concrete and plaster supports toward the abatement of NO and NO2. Photocatalytic experiments on the coatings were performed in two types of ﬂow reactors, one with laminar ﬂow while the other with an ideally-mixed ﬂow (Figure 37), under real world conditions in terms of temperature, relative humidity, irradiation intensity and pollutant concentrations. Results showed that the photocatalytic process reduced signiﬁcantly the concentration of both NOx agents in the air. Figure 37. Experimental laminar ﬂow (a); and ideally-mixed ﬂow (b) reactors used to examine the photocatalytic oxidation of low concentrations of gas NOx streams at some speciﬁed humidity. Reproduced with permission from Zouzelka and Rathousky [128]. Copyright 2017 by The Authors (open access license). Figure 37. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Experimental laminar ﬂow (a); and ideally-mixed ﬂow (b) reactors used to examine the photocatalytic oxidation of low concentrations of gas NOx streams at some speciﬁed humidity. Reproduced with permission from Zouzelka and Rathousky [128]. Copyright 2017 by The Authors (open access license). The decrease in the concentration of NOx achieved in the steady-state for an inlet concentration of NO and NO2 of 0.1 ppmv, corresponding to highly polluted urban air, was up to 75 μmol m−2 h−1 (for the Protectam FN2) and 50 μmol m−2 h−1 (for the Aeroxide P-25 TiO2) at a ﬂow rate of 3000 cm3 min−1 and a relative humidity of 50%. Because of a conspicuous lack of data regarding the performance of photocatalytic coatings over long periods, the authors [128] also examined aged photocatalytic FN2 coatings on a 300-m2 concrete noise barrier that had been exposed to heavily-trafﬁcked (ca. 30,000 vehicles a day) thoroughfare in Prague where the NOx concentration reached 30–40 μg m−3, often exceeding the permitted NO2 limit of 40 μg m−3 (or 0.021 ppmv). 467 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 Experimentally, the area of irradiated photocatalytic surface was 50 cm2 (5 cm × 10 cm); ﬂow rate of air mixture was 3000 cm3 min−1; total volume of air treated in 24 h was 4.32 × 106 cm3; volume of puriﬁed air and area of irradiated photocatalytic surface were the same in both photoreactors, although the reactors differed substantially in volume/irradiated area (65 times greater for the ideally-mixed ﬂow reactor); volume of the ideally-mixing ﬂow reactor was 5200 cm3 (18 × 32 × 9 cm); the free volume of the laminar ﬂow reactor was 80 cm3 (5 × 32 × 0.5 cm); and linear streaming velocity of the gas was 0.2 m s−1. The bandgaps of the TiO2 in the two materials were 3.2 eV for the P-25 sample and 3.05 eV for the FN2 sample, the red-shift in the latter being attributed to the effect of the binder on titania [128]. Comparison of the photocatalytic performance of P-25 and FN2 materials reported in Figure 38 shows the reaction rate with the FN2 coating to be greater than for the P-25, even though the quantity of TiO2 in the FN2 coating was lower. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment For instance, the reaction rate with the FN2 coating on concrete in both laminar and ideally-mixed ﬂow reactors at an inlet NO concentration of 1.0 ppmv was 40% and 49% higher, respectively, than the corresponding reaction rate of P-25 which, according to the authors [128], was likely due to the nearly twofold larger surface area of the FN2 specimen (82 m2 g−1) relative to the P-25 sample (47 m2 g−1). Figure 38. Comparison of the photocatalytic reaction rates in the degradation of NOx with the FN2 and P-25 in both ideally-mixed and laminar ﬂow reactors for an inlet NO2 concentration of 0.1 ppmv and relative humidity (RH) of 50%. FN2 (black) and P-25 (blue) coated on concrete in laminar ﬂow reactor. FN2 (red) and P25 (green) coated on plaster in ideally-mixed ﬂow reactor. Reproduced with permission from Zouzelka and Rathousky [128]. Copyright 2017 by The Authors (open access license). Figure 38. Comparison of the photocatalytic reaction rates in the degradation of NOx with the FN2 and P-25 in both ideally-mixed and laminar ﬂow reactors for an inlet NO2 concentration of 0.1 ppmv and relative humidity (RH) of 50%. FN2 (black) and P-25 (blue) coated on concrete in laminar ﬂow reactor. FN2 (red) and P25 (green) coated on plaster in ideally-mixed ﬂow reactor. Reproduced with permission from Zouzelka and Rathousky [128]. Copyright 2017 by The Authors (open access license). Figure 38. Comparison of the photocatalytic reaction rates in the degradation of NOx with the FN2 and P-25 in both ideally-mixed and laminar ﬂow reactors for an inlet NO2 concentration of 0.1 ppmv and relative humidity (RH) of 50%. FN2 (black) and P-25 (blue) coated on concrete in laminar ﬂow reactor. FN2 (red) and P25 (green) coated on plaster in ideally-mixed ﬂow reactor. Reproduced with permission from Zouzelka and Rathousky [128]. Copyright 2017 by The Authors (open access license). Evidently, the inlet concentration of NO had a substantial inﬂuence on the reaction rate in both reactors, as evidenced on comparing the data for the FN2 coating in the laminar ﬂow reactor in which the reaction rate was approximately proportional to the inlet NO concentration; in other words, for an inlet NO concentration of 1.0 ppmv, the rate was about ten times greater than for a concentration of 0.1 ppmv [128]. With regard to the effect of NO2 concentration used (0.1 ppmv) the trends differed from those for NO. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment First, the difference in the reaction rate at the beginning of the reaction and in the steady-state was much smaller for both P-25 and FN2 coatings when applied on plaster than on concrete. On plaster, the reaction rate was practically unchanged, whereas on concrete the decrease in rate was between 40% and 60%. Moreover, the steady-state reaction rate (after a 24 h period) on the coatings applied on plaster was consistently 1.5–1.8 times higher in comparison with the coatings on concrete. Another important issue emphasized by Zouzelka and Rathousky [128] was the durability of the performance of the photocatalytic coatings under real-world conditions, as exempliﬁed by 468 Catalysts 2018, 8, 553 the commercial photocatalytic coating Protectam FN2 that maintained relatively high efﬁciency in removing NOx from contaminated air even after two years under the harsh conditions noted above; this was likely due to good mechanical properties of the binder. Moving from small benchtop laboratory photoreactors to actual ﬁeld studies to investigate the efﬁciencies (activities) of photocatalytic concrete/mortar with embedded titania photocatalysts, simulation chambers have also proven suitable to test photocatalytic materials under controlled environmental conditions, as they represent a mid-way step between laboratory and actual environ-ment (a sort of pilot plant scale). Additionally, simulation chambers offer the opportunity to examine the impact of photocatalytic surfaces, also on secondary air chemistry, in contrast to the smaller ﬂow-type reactors. In this regard, within the framework of the Life + project PhotoPAQ (2010–2014) that will be described later (Section 5.2), Mothes and coworkers [25] tested the behavior of O3, NOx, and selected VOCs (toluene and isoprene) on a photocatalytic cementitious coating material (with Italcementi’s TX-Active®) under UV irradiation and atmospherically relevant conditions (relative humidity, temperature, and realistic pollutant concentration) using the aerosol chamber LEipziger Aerosol Kammer (LEAK) at the Leibniz Institute for Tropospheric Research (TROPOS)—a 19 m3 aerosol cylindrical chamber with a surface-to-volume ratio of 2.0 m−1 and illuminated with UV lamps (λ = 300–400 nm) with an average light intensity of 11 W m−2. The TX-Active® material was mixed with ultrapure water and subsequently applied manually on both sides of sand-blasted glass plates supported by a home-made device to obtain a rough but uniform surface (ﬁnal thickness, ca. 3–4 mm), after which the material was cured for ca. 1 month. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Comparison of the blank and boosted materials revealed that UV irradiation had a signiﬁcant inﬂuence on NOx owing to the photocatalytic properties of TiO2 embedded in the boosted material. The small increase of NO at the beginning of the light period was attributed by the authors [25] to photolysis of NO2 according to the Leighton sequence described by Reactions (85)–(87). Results from the chamber studies also showed a photocatalytic effect on O3, evidenced by an increase of the geometric uptake coefﬁcient from 5.2 × 10−6 for the inactive to 7.7 × 10−6 for the active material under irradiation. Measured ﬁrst-order rate constants for NOx under irradiation ranged from 2.6 × 10−4 s−1 to 5.9 × 10−4 s−1, signiﬁcantly higher compared to the range for the inactive materials (7.3–9.7 × 10−5 s−1), thus demonstrating the photocatalytic effect. However, no signiﬁcant photo- catalytic degradation was observed for the VOCs (toluene, isoprene); the upper limit uptake coefﬁcient for both was only 5.0 × 10−7. Small carbonyl (C1–C5) gas-phase compounds were identiﬁed when using the photocatalytically active material, a result of the photocatalytic degradation of the organic additives [25]. In contrast to the uptake observed for pure O3, a clear photocatalytic formation of O3 was observed (ko(O3) ≈5 × 107 molecules cm−3 s−1) during the experiments with NOx (RH ≥50%). The authors further noted the necessity for detailed studies of heterogeneous reactions on such surfaces under more complex simulated atmospheric conditions as enabled by simulation chambers. Although many studies have been done on TiO2 photocatalytic cementitious substrates to deNOxify the environment—as we discuss below—nearly all studies neglected to consider the inter- relationship between the metal-oxide photocatalyst and the non-negligible chemistry that might occur on the cementitious support—in fact, this inter-relationship is particularly relevant for an appropriate understanding of deNOxiﬁcation processes. In this regard, the investigations carried out by Macphee and Folli [129] on Photocatalytic concretes—The interface between photocatalysis and cement chemistry have opened up the proverbial Pandora’s Box in which they addressed: (i) the photocatalytic mechanisms applicable to atmospheric depollution; (ii) the inﬂuence of doping the metal oxide; and (iii) the application of TiO2-based photocatalysts to concrete. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The cleaned coated glass plates were then inserted into a special aluminum rack installed inside the chamber—the area of the photocatalytically active surface was 6.65 m2 (boosted specimen)—a blank specimen devoid of TX-Active® titania was also made for comparison. Both specimens were ﬂushed with clean air (200 L min−1) with the UV lamps turned ON for ca. 30 h to remove any potentially adsorbed pollutants; RH varied between dry air and 50%—the photocatalytic tests were performed following the 2007 ISO-22197-1 standard methodology. Results obtained for the NO and NOx degradation experiments together with formation of O3 at three RHs and at 293 K are displayed in Figure 39; dark period ca. 120 min, irradiation period also 120 min, initial concentrations of NO and NOx = 40 ppb [25]. Figure 39. NOx investigations under dark and light conditions in LEAK simulation chamber (RH = 10%, 50% and 70%; T = 293 K) to compare the behavior of the boosted (solid lines) and blank materials (dashed lines) for the time-dependent changes in the concentrations of NO, NOx, and O3 akin to environmental quantities. Reproduced with permission from Mothes et al. [25]. Copyright 2016 by Springer-Verlag Berlin Heidelberg (License No.: 4453250415918). C h b h i f NO d i h d k i d NO h d l li h l diff Figure 39. NOx investigations under dark and light conditions in LEAK simulation chamber (RH = 10%, 50% and 70%; T = 293 K) to compare the behavior of the boosted (solid lines) and blank materials (dashed lines) for the time-dependent changes in the concentrations of NO, NOx, and O3 akin to environmental quantities. Reproduced with permission from Mothes et al. [25]. Copyright 2016 by Springer-Verlag Berlin Heidelberg (License No.: 4453250415918). Contrary to the behavior of NOx during the dark period, NO showed only a slightly different behavior at the different RHs; NO’s behavior was nearly constant at 50% RH and 70% RH, whereas Contrary to the behavior of NOx during the dark period, NO showed only a slightly different behavior at the different RHs; NO’s behavior was nearly constant at 50% RH and 70% RH, whereas 469 Catalysts 2018, 8, 553 at 10% RH a slight decrease was seen that was attributed to slightly adsorbed NO on the specimens’ surface [25]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment (a) Schematic illustration of pH changes and light penetration proﬁle in the ageing concrete; (b) optical microscope image of the cross section of a concrete specimen containing TiO2 under X-polarized light—note the surface exhibits a thin calcite layer that appears in a lighter colour; (c) optical microscope image of the cross section of a concrete specimen containing TiO2 under ﬂuorescent light—note that the calcite layer now appears dark green indicative of low porosity; and (d) SEM image of the top surface of a concrete specimen containing TiO2 (same sample as in (b,c)) wherein the surface consists of a layer of small, closely spaced, euhedral calcite crystals—no TiO2 photocatalytic clusters are exposed. Reproduced with permission from Macphee and Folli [129]. Copyright 2016 by Elsevier Ltd. (License No.: 4453091259877). Figure 40. (a) Schematic illustration of pH changes and light penetration proﬁle in the ageing concrete; (b) optical microscope image of the cross section of a concrete specimen containing TiO2 under X-polarized light—note the surface exhibits a thin calcite layer that appears in a lighter colour; (c) optical microscope image of the cross section of a concrete specimen containing TiO2 under ﬂuorescent light—note that the calcite layer now appears dark green indicative of low porosity; and (d) SEM image of the top surface of a concrete specimen containing TiO2 (same sample as in (b,c)) wherein the surface consists of a layer of small, closely spaced, euhedral calcite crystals—no TiO2 photocatalytic clusters are exposed. Reproduced with permission from Macphee and Folli [129]. Copyright 2016 by Elsevier Ltd. (License No.: 4453091259877). The alkali nature of the cementitious surface also has signiﬁcant consequences on the chemistry taking place at the surface when NOx molecules are adsorbed, even in the absence of any catalyst and under non-irradiating conditions, as both NO and NO2 (NOx molecules) can undergo an oxidative reaction with the alkali to yield nitrite and nitrate ions (Equations (88) and (89)) [130]. These two reactions must be considered whenever photocatalytic efﬁciencies are assessed in deNOxiﬁcation processes. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment These authors further emphasized some of the points described earlier that the catalyst efﬁciency is inﬂuenced by several factors, none of which are negligible: for instance: (i) energy and intensity of the activating radiation incident on the photocatalyst; (ii) the number and relative energy positions of the electronic states in the photocatalyst, deﬁned by the crystal structure and the redox potentials associated with the required redox processes; (iii) the charge carrier mobility within the semiconductor’s conduction and valence bands; (iv) the kinetics of charge transfer processes; and (v) the accessible catalyst surface for adsorption of the pollutants, oxygen, and water [129]. Of greater importance, they noted that the chemistry of the cement environment is quite different from the ambient conditions normally prevalent in environmental photocatalysis, which will have a non-insigniﬁcant inﬂuence on the normal behavior of the catalysts in the concrete [129]. In freshly mixed cement, the high pH and the high ionic strength aqueous mix containing multiply charged ions dramatically modiﬁes the surface chemistry and behavior of TiO2 dispersions, and thus will have a strong impact on dispersion behavior and adsorption properties, as the mode of adsorption and strength of binding between an adsorbing molecule and a surface are conditioned by their relative charges. Additionally, pH changes during the early stages of mixing and ageing the cementitious substrate will also have an impact on the band edge positions in the semiconductor photocatalyst as well as the oxidation chemistry of NOx. The alkalinity of wet cement/concrete is typically greater than pH 13, which can change rapidly at the surface as the concrete sets and the formwork is removed (Figure 40a) [129]. The highly alkaline aqueous phase can react with atmospheric CO2 to produce CO32−ions, causing the pH to decrease and induces surface deposition of calcite—a carbonate mineral and the most stable polymorph of calcium carbonate—that can obstruct deposited photocatalysts. Figure 40b–d displays the microscopic details of the surface calcite layer that obstructs TiO2 clusters present in the hardened cementitious matrix. 470 Catalysts 2018, 8, 553 Figure 40. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment 2 NO2 + 2 OH−→NO2−+ NO3−+ H2O (88) NO2 + NO + 2 OH−→2 NO2−+ H2O (89) (88) (89) To address the geographical limitations of UV intensity and recognizing that sunlight is the light source to activate the photocatalyst in practical applications of photocatalytic concrete in deNOxiﬁcation, the Macphee group [129,131] also investigated M-doped titania (M = W, Nb) photo-catalysts toward removal of NOx because unmodiﬁed TiO2 releases a signiﬁcant quantity of toxic NO2 in the deNOxiﬁcation process, a problem rarely raised in several previous studies as NOx abatement is commonly assessed only on the basis of NO removal rather than NO2 or total NOx removal. Accordingly, to highlight the problem, they performed a systematic study on a number of commercial titania powders and doped titania specimens by determining both the photocatalytic activity (i.e., the photonic efﬁciency ξ [29], as deﬁned by Equation (90)) and the selectivity (S deﬁned by Equation (91)) toward formation of NO2 and NO3−in the oxidation of NO. ξ = (Cdark −Cirr)Vp ρoART (90) (90) 471 Catalysts 2018, 8, 553 S = ξNOx ξNO (91) S = ξNOx ξNO (91) where Cdark and Cirr are the concentrations of the species of interest under dark and irradiated conditions, respectively; V is the volumetric ﬂow rate; p is the pressure; ρo is the photon ﬂux impinging on the photocatalyst surface; A is the irradiated area; R the gas constant; and T is the Kelvin temperature. In the case of W-doped TiO2, Figure 41 demonstrates that while doping titania with tungsten signiﬁcantly improved the selectivity toward formation of NO3−, it considerably reduced the photonic efﬁciency [131]. This raised the question as to how one could evaluate various photocatalysts toward NOx abatement when both activity and selectivity needed to be considered. To resolve this matter, the authors [129,131] deﬁned a new ﬁgure of merit: the DeNOx index (Equation (92)), which was derived by assigning a toxicity value to both NO and NO2 and then expressing the change in total toxicity rather than the concentration change of the individual NOx gases [129,131]: ξdeNOx = ξNO −3ξNO2 (92) ξdeNOx = ξNO −3ξNO2 (92) The deNOx index is positive if the photocatalyst lowers the NOx toxicity. The deNOx index is positive if the photocatalyst lowers the NOx toxicity. The deNOx index is positive if the photocatalyst lowers the NOx toxicity. The deNOx index is positive if the photocatalyst lowers the NOx toxicity. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Figure 41. Plots illustrating the photonic efﬁciency of NOx removal and nitrate selectivity of W-modiﬁed TiO2 illuminated under broadband radiation. Adapted with permission from Bloh et al. [131]. Copyright 2014 by the Royal Society of Chemistry (license ID: 4453100929571). Figure 41. Plots illustrating the photonic efﬁciency of NOx removal and nitrate selectivity of W-modiﬁed TiO2 illuminated under broadband radiation. Adapted with permission from Bloh et al. [131]. Copyright 2014 by the Royal Society of Chemistry (license ID: 4453100929571). Figure 42a illustrates the results from several photocatalyst compositions and demonstrates a positive deNOx index at nominal W contents > 4.2 at.% for the W-doped TiO2 [129]. A range of commercial TiO2 photocatalysts were tested under comparable conditions giving a DeNOx index between 0 and −4000 representing poor nitrate selectivity and, most importantly, meaning that the more toxic NO2 was generated [129]. These data were further supported by analyses on powder and mortar samples incorporating W-doped and Nb-doped TiO2, as displayed in Figure 42b [129]. Interestingly, although undoped P-25 TiO2 has been the workhorse in photocatalysis, it displayed a net negative effect owing to its high activity but low nitrate selectivity; that is, it efﬁciently converted NO to NO2 rather than to NO3−ions. By contrast, both W- and Nb-doped TiO2 showed a higher tendency towards NO3−and thus are better suited as remediation photocatalysts, whether in powdered form or embedded in mortars. 472 Catalysts 2018, 8, 553 (a) (b) Figure 42. (a) deNOx index measured using W-doped TiO2 under broad band radiation simulating sunlight; and (b) deNOx data on powder and mortar samples containing W-and Nb-doped TiO2 (broad band and visible light exposure). Reproduced with permission from Macphee and Folli [129]. Copyright 2016 by Elsevier Ltd. (License No.: 4453091259877). (a) (a) (b) (b) Figure 42. (a) deNOx index measured using W-doped TiO2 under broad band radiation simulating sunlight; and (b) deNOx data on powder and mortar samples containing W-and Nb-doped TiO2 (broad band and visible light exposure). Reproduced with permission from Macphee and Folli [129]. Copyright 2016 by Elsevier Ltd. (License No.: 4453091259877). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment NO →HNO2 →NO2 →NO3− (93) (93) Although this oxidation sequence is thought to be the preferred reaction with respect to air quality, as claimed by the authors [132], a far better process to improve air quality would be the reduction of NOx back to N2 and O2 as can be achieved by selective catalytic reduction (SCR) and selective non-catalytic reduction (SNCR) technologies (see above). Regardless, in TiO2-based photocatalytic methods, both oxidative entities (valence band holes) and reductive entities (conduction band electrons) are generated upon sunlight UV activation of the TiO2 photocatalyst, so that the oxidation sequence (Reaction (93)) may also lead to intermediates—for example, ozone and peroxyacyl nitrates—in addition to those in the sequence (Reactions (94)–(96)) produced in surface processes in the NO2 production regime, and those produced in the sequence (Reactions (97)–(100)) in surface processes in the N2O5 production regime when ozone is present [133], all of which could be released into the atmospheric environment not forgetting that NO2 is nearly three times more toxic than NO. 2 NO2(ads) ←→NO3− (ads) + NO+ (ads) (94) NO+ (ads) + O2− (ads) ←→NO2− (ads) (95) NO2− (ads) + NO2(ads) ←→NO3− (ads) + NO(gas) (96) TiO2 + O3 →O-TiO2 + O2 (97) NO2(ads) + O-TiO2 →NO3 −TiO2 (98) NO3(ads) + NO2(g) →N2O5(ads) (99) N2O5(ads) →N2O5(g) (100) 2 NO2(ads) ←→NO3− (ads) + NO+ (ads) (94) NO+ (ads) + O2− (ads) ←→NO2− (ads) (95) NO2− (ads) + NO2(ads) ←→NO3− (ads) + NO(gas) (96) TiO2 + O3 →O-TiO2 + O2 (97) NO2(ads) + O-TiO2 →NO3 −TiO2 (98) NO3(ads) + NO2(g) →N2O5(ads) (99) N2O5(ads) →N2O5(g) (100) 2 NO2(ads) ←→NO3− (ads) + NO+ (ads) (94) NO+ (ads) + O2− (ads) ←→NO2− (ads) (95) NO2− (ads) + NO2(ads) ←→NO3− (ads) + NO(gas) (96) TiO2 + O3 →O-TiO2 + O2 (97) NO2(ads) + O-TiO2 →NO3 −TiO2 (98) NO3(ads) + NO2(g) →N2O5(ads) (99) N2O5(ads) →N2O5(g) (100) (94) (100) Notwithstanding the above sequences, accumulation of NO3−on catalytic surfaces must be managed since adsorbed NO3−ions block catalytic sites from further NO/NO2 adsorption—thus, affecting catalytic activity—as the nitrates back react to form NO2 thereby causing nitrate selectivity to be diminished. This calls attention to the notion that reNOxiﬁcation may be an important event in deNOxiﬁcation processes. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Continuing their interesting studies on photocatalytic concrete, in their most recent article, the Macphee group [132] examined the effect of photocatalyst placement as regards photocatalyst efﬁciencies in concrete technology, and pointed out that even though the technology represents a well-established concept and notwithstanding the signiﬁcant opportunities for air quality improve-ments to be derived from the considerable concrete surfaces exposed to the atmosphere, especially in cities highly polluted by vehicle exhaust and industrial emissions, photocatalytic concretes have so far remained in the investigative sphere, rather than in the mainstream of applications. As with any commercial new technology, the likely barriers for widespread implementation may well be cost effectiveness that might emanate from the photocatalyst impact and the challenges in measuring directly what the impact of this new concrete technology may actually be on air quality. The challenges are indeed very complex. Accordingly, the authors [132] placed photocatalytic efﬁciencies into context by comparing performances of conventional photo-catalyst dispersions in surface mortar coatings vis-à-vis photocatalysts supported on surface-exposed aggregates as well as on the nature and impact of catalyst binding to aggregate supports. However, as we discuss below, the efﬁciencies in real-world 473 Catalysts 2018, 8, 553 environmental applications differ signiﬁcantly from the efﬁciencies experienced in a laboratory setting where small photo-reactors have been used to measure the efﬁcacies of various TiO2 photocatalyst concrete specimens indoor, wherein experimental conditions can be controlled in contrast to outdoor environmental conditions, which are not only widely different, but are also uncontrollable conditions that one experiences in the real world. Nonetheless, their ﬁndings in photocatalyst placement as regards efﬁcacies of photocatalytic concrete are worth noting [132]. environmental applications differ signiﬁcantly from the efﬁciencies experienced in a laboratory setting where small photo-reactors have been used to measure the efﬁcacies of various TiO2 photocatalyst concrete specimens indoor, wherein experimental conditions can be controlled in contrast to outdoor environmental conditions, which are not only widely different, but are also uncontrollable conditions that one experiences in the real world. Nonetheless, their ﬁndings in photocatalyst placement as regards efﬁcacies of photocatalytic concrete are worth noting [132]. In treating NOx gases by TiO2-based photocatalytic substrates, the currently accepted sequence is represented by Reaction (93); the greater is the degree of conversion of NO and NO2 to nitrates, the greater is the catalytic activity of the substrates. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Accordingly, NO3−must be removed periodically by washings [132] at least on a weekly basis for certain infrastructures (tunnels, noise barriers, and buildings, among others) but especially for road tunnels where rain has no impact; this maintenance would carry increased costs. Moreover, where rain is sparse in the summer months as often occurs in Southern Italy and California (USA), such periodic washings may well be necessary. In addition, in conventional photo-catalytic concretes, nitrate washings via condensed atmospheric moisture represents but a mild risk from acidiﬁcation of the cement (carbonated at the near surface) which may inﬂuence leaching-induced damage over time [132]. In their earlier articles [129,131], Macphee’s group addressed various factors that limit photocatalytic efﬁciencies in cement-based systems in which the metal-oxide photocatalyst was incorporated in the mortar/concrete. They have now proposed to separate the chemistry taking place on the photocatalyst surface from the chemistry occurring on the cementitious support, so as to 474 Catalysts 2018, 8, 553 enhance photocatalytic efﬁciency by depositing the photocatalyst on the surface of the cementitious aggregate (Figure 43) [132]; that would require some sort of binder which unfortunately would also add to the complexity of the events as the binder may not only photodegrade but may also bring about changes to the electronic structure of the photocatalyst [128]. enhance photocatalytic efﬁciency by depositing the photocatalyst on the surface of the cementitious aggregate (Figure 43) [132]; that would require some sort of binder which unfortunately would also add to the complexity of the events as the binder may not only photodegrade but may also bring about changes to the electronic structure of the photocatalyst [128]. Figure 43. Conventional mortar/concrete with the photocatalyst embedded within the aggregate (mixed), and aggregate-supported conﬁguration of the photocatalyst (blue) applications in proposed concrete technology whereby the photocatalyst is supported on sort of binder that bonds both the photocatalyst and the cementitious support. Reproduced with permission from Yang et al. [132]. Copyright 2018 by The Authors (open access license). Fi 43 C ti l t / t ith th h t t l t b dd d ithi th t Figure 43. Conventional mortar/concrete with the photocatalyst embedded within the aggregate (mixed), and aggregate-supported conﬁguration of the photocatalyst (blue) applications in proposed concrete technology whereby the photocatalyst is supported on sort of binder that bonds both the photocatalyst and the cementitious support. Reproduced with permission from Yang et al. [132]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The photonic efﬁciency ξNO increased initially with 475 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 increasing TiO2 mass fractions but became steady around 0.76%. Concomitantly, the ξNOx and nitrate selectivity (S%) increased with increasing mass fraction of TiO2, which led to a corresponding decrease of the rate of production of NO2. Compared with the NOx removal results for pure TiO2, the estimated ξNO for the QST specimens was ca. 0.6%, while nitrate selectivity was ca. 40%; the supported TiO2 particles presented higher photonic efﬁciency for NO removal, but lower nitrate selectivity owing to reduced NO2 removal efﬁciency. Figure 45. Concentration proﬁles for NO, NO2 and NOx (NO + NO2) during the photocatalytic oxidation of NO on TiO2 hydrosol (0.1 g) photocatalysts. Reproduced with permission from Yang et al. [132]. Copyright 2018 by the authors (open access license). Figure 45. Concentration proﬁles for NO, NO2 and NOx (NO + NO2) during the photocatalytic oxidation of NO on TiO2 hydrosol (0.1 g) photocatalysts. Reproduced with permission from Yang et al. [132]. Copyright 2018 by the authors (open access license). Clearly, the photocatalytic performance depended on the TiO2 loading with the activity increasing with TiO2 content up to ca. 0.76%, while nitrate selectivity increased across the range of TiO2 loading used, and was strongly conditioned by the negative inﬂuence of Ti–O–Si bonds on selectivity and the effect of dilution as TiO2 loading was increased. Washing to remove weakly bonded TiO2 led to a decrease of the photocatalytic activity ξNO by about 30%, even though nearly half of the TiO2 had been lost (only 0.34% remaining); however, selectivity was not signiﬁcantly affected [132]. While such a low photocatalytic activity could be explained, in part, by fast charge carrier recombination and charge carrier trapping dynamics within intra-bandgap defect states in TiO2 photocatalytic substrates, mechanical loss of titania particulates from the support, coupled with the observed activity under controlled indoor experimental conditions (small ﬂow-through reactor), as reported by the authors [132], do not bode well to convince implementation of the photocatalytic concrete technology to outdoor real-world environments where, as noted earlier, the conditions are different and uncontrollable. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Copyright 2018 by The Authors (open access license). Figure 43. Conventional mortar/concrete with the photocatalyst embedded within the aggregate (mixed), and aggregate-supported conﬁguration of the photocatalyst (blue) applications in proposed concrete technology whereby the photocatalyst is supported on sort of binder that bonds both the photocatalyst and the cementitious support. Reproduced with permission from Yang et al. [132]. Copyright 2018 by The Authors (open access license). Figure 43. Conventional mortar/concrete with the photocatalyst embedded within the aggregate (mixed), and aggregate-supported conﬁguration of the photocatalyst (blue) applications in proposed concrete technology whereby the photocatalyst is supported on sort of binder that bonds both the photocatalyst and the cementitious support. Reproduced with permission from Yang et al. [132]. Copyright 2018 by The Authors (open access license). The aggregate used to support the prepared TiO2 photocatalyst was quartz sand (1–2 mm diameter) that was subsequently treated in aqueous solution of Ca(OH)2 followed by carbonation with CO2 (aggregates denoted QST) [132]. The aggregates were placed in the ﬂow-through reactor displayed in Figure 44 (monolayer; area, 5 × 10−3 m2), after which they were irradiated with a 500-W Xe-lamp solar illuminator (photon ﬂux at the aggregates, 3.05 μmol m−2 s−1); other conditions were: 1 ppm NO gas in synthetic air; relative humidity, 40%; temperature, 25 ◦C; volumetric ﬂow rate, 5 × 10−5 m3 s−1 through the reactor. Figure 44. Schematic illustration of photoreactor used in testing the photocatalytic degradation of NOx gases. Reproduced with permission from Yang et al. [132]. Copyright 2018 by The Authors (open access license). Figure 44. Schematic illustration of photoreactor used in testing the photocatalytic degradation of NOx gases. Reproduced with permission from Yang et al. [132]. Copyright 2018 by The Authors (open access license). Three stages were identiﬁed by the authors in the process of converting NO to nitrate: (i) stabilization of NO concentrations on the QST specimens in the dark for achieving an adsorption/desorption equilibrium; (ii) the photocatalytic conversion of NO under illumination; and (iii) recovery of NO concentrations after illumination was terminated [132]. Results showed that the concentration of NO decreased rapidly on illumination, remaining nearly constant for a while, whereas the concentration of NO2 increased signiﬁcantly during this time; stopping the illumination caused the NO2 concentration to go back to zero (Figure 45). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The results of the few studies reported herein with regard to the extent of removal of NOx gases (NO and NO2) are summarized in Table 7 [25,97,121,122,124–126,128,132,134] for indoor laboratory experiments carried out with small bed ﬂow-through reactors employing a variety of different conditions from one study to the next. Various standardization protocols have been proposed in the last few years that have led to signiﬁcant variation in the results—as should have been expected. 476 Catalysts 2018, 8, 553 Table 7. Percent removal of NOx gases carried out with TiO2-based photocatalytic protocols in an indoor laboratory setting using small ﬂow-through reactors under a variety of conditions. Study Gas TiO2-Based Photocatalytic Removal of NOx Gases Year Study Was Carried Out Refs 1 NOx 60–90% 1994 [97] 2 NO 52–88% 2000 [122] 45–90% 3 NOx 88–90% 2009, 2011 [121,124] 4 NO 25% 2011 [125] 5 NOx 50% 2012 [126] 6 NOx 40–60% 2017 [128] 7 NOx 35%; 66% 2016 [25] 8 NO 44% 2018 [132] NOx 16% 9 NO 80% 2006 [134] 5 2 DeNOxing the OUTDOOR Environment with TiO2 Photocatalytic Cementitious Materials 5.2. DeNOxing the OUTDOOR Environment with TiO2 Photocatalytic Cementitious Materials 5.2. DeNOxing the OUTDOOR Environment with TiO2 Photocatalytic Cementitious Materials As shown above, many studies have been reported, albeit not all have been cited herein, on the photocatalytic abatement of NOx in a laboratory setting wherein experimental conditions could be controlled. Relatively speaking, however, only a few ﬁeld trials have been conducted outdoors in open air environment under real-world conditions; as we have noted, the conditions are not only different but, more importantly, they also cannot be controlled. As an example of the latter, some reduction of NOx levels was achieved at the laboratory scale and in outdoor ﬁeld experiments in Antwerp (Belgium) where a surface of ca. 10,000 m2 was covered with photocatalytic pavement blocks—researchers that did the study were unable to conclude the extent to which NOx was reduced because the measurement period was too short [135]. Additionally, an experiment to test the efﬁcacy of photocatalytic TiO2-mortar panels to degrade NOx in a setting that involved artiﬁcial canyon streets showed NOx concentrations to be reduced by as much as 37–82% depending on the conditions [136]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The church was erected with three huge, totally white sails made of precast photocatalytic concrete blocks (Figure 47) that would ensure unparalleled and time-enduring white color to the built elements thanks to the self-cleaning properties of the ﬁnal TX-Active® surfaces [140,141]. The photocatalytic cement-based product used was expected to maintain its aesthetic characteristics unchanged over time, especially the color, even in the presence of aggressive urban environments. Laboratory pilot-scale tests demonstrated that NOx abatement with photocatalytic cement-based products was impacted by temperature, relative humidity, contact time of NOx with the surface (ﬂow velocity, height of the air ﬂow above the sample, among others); it was also noted that reduction efﬁciency increased with longer contact times (larger surfaces, lower velocities, and higher turbulence/mixing), higher temperatures and lower relative humidity—however, in real situations, conditions such as high temperatures, no wind and no rain present the largest risk of ozone formation [142]. Figure 46. Laboratory tests of the anatase TiO2 photocatalytic cement-based products used in the precast concrete panels in the sails of the Dives in Misericordia Church (Rome): (a) under dark conditions; and (b) under UV irradiation for 7 h. Note the increased reduction of NO and NO2 under UV irradiation. Reproduced with permission from Cassar [139]. Copyright 2004 by the Materials Research Society (License No.: 4453250907572). Figure 46. Laboratory tests of the anatase TiO2 photocatalytic cement-based products used in the precast concrete panels in the sails of the Dives in Misericordia Church (Rome): (a) under dark conditions; and (b) under UV irradiation for 7 h. Note the increased reduction of NO and NO2 under UV irradiation. Reproduced with permission from Cassar [139]. Copyright 2004 by the Materials Research Society (License No.: 4453250907572). Figure 46. Laboratory tests of the anatase TiO2 photocatalytic cement-based products used in the precast concrete panels in the sails of the Dives in Misericordia Church (Rome): (a) under dark conditions; and (b) under UV irradiation for 7 h. Note the increased reduction of NO and NO2 under UV irradiation. Reproduced with permission from Cassar [139]. Copyright 2004 by the Materials Research Society (License No.: 4453250907572). Guerrini and Corazza [142] claimed that a building element containing cement to which TiO2 had been added was capable of maintaining its aesthetic appearance unaltered in time (see also [121]), thus contributing to reduce the dirtiness of surfaces exposed to speciﬁc polluted environments. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Likewise, a study carried out in 2006 in Bergamo (Italy) over a four-week period in a segment of a local street covered with photocatalytic paving stones showed a 30–40% reduction in the NOx concentration compared to a similar section left untreated [137]; elsewhere, NOx reduction levels of 20 to 50% have been quoted that depended on weather and trafﬁc conditions [138]. A study done in the city of Segrate (Italy) [137] where a concrete road segment (7000 m2) was constructed using a thin-layer of a photocatalytic mortar was said to display a 57% NOx abatement level, while a similar study in an industrial section near the town of Calusco (Italy) reported a NOx abatement level of 45% for an 8000 m2 pavement built with photocatalytic concrete blocks. By contrast, ﬁeld studies carried out in an artiﬁcial model street canyon in Petosino, Italy [15] and in the Leopold II tunnel in Brussels (Belgium) [14] reported photocatalytic reductions of NOx well below measurement precision errors (1–2%). An “indoor” ﬁeld study of the Umberto I Tunnel (Rome), for which the walls were coated with a photocatalytic paint and a new UV-Visible lighting system installed, showed more than 20% reduction of NOx [138]. The latter two ﬁeld studies [14,138] are discussed in some detail below. Clearly, the many discrepancies in NOx reduction levels when using photocatalytic paints, photocatalytic paving bricks, or photocatalytic mortar/concrete on streets, highways, or roadside noise barriers: (i) call attention to highly differing prevailing atmospheric conditions in ﬁeld experiments; (ii) call into question how these ﬁeld experiments were carried out; and (iii) call into question how the empirical results were interpreted. For that reason, we now describe the various attempts carried out using TiO2-based photocatalytic substrates/aggregates in a few countries and discuss their ﬁndings with an emphasis on three major undertakings funded wholly or in part by the European Union: the PICADA Project (2002–2006); the Life+-funded Project PhotoPAQ (2010–2014); and the LIFE MINOX-STREET Project (2014–2018). One of the ﬁrst applications of photocatalytic cement-based materials for self-cleaning purposes was Richard Meier’s Dives in Misericordia Church project in Rome where Italcementi served as the principal technical sponsor, analyzed the concrete mix (Figure 46) [139], and provided continuous 477 Catalysts 2018, 8, 553 supervision during the construction period (1999–2001). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment A look at the photographs illustrated in Figure 48, however, shows that the long-term effectiveness of anatase TiO2 photocatalytic cement-based external walls of the church failed the test of time. 478 Catalysts 2018, 8, 553 Figure 47. Dives in Misericordia Church, Rome, constructed of TiO2 self-cleaning and depolluting TX Active cement (inaugurated in 2003). Reproduced from https://es.i-nova.net/content?articleId=96804 or https://www.archdaily.com/20105/church-of-2000-richard-meier. Figure 47. Dives in Misericordia Church, Rome, constructed of TiO2 self-cleaning and depolluting TX Active cement (inaugurated in 2003). Reproduced from https://es.i-nova.net/content?articleId=96804 or https://www.archdaily.com/20105/church-of-2000-richard-meier. Another important sponsorship of the CTG-Italcementi Group was the renovation of the Umberto I Tunnel located in the center of Rome (dimensions: length, 347.7 m; width, 17 m; height: 8.5 m) for which renovation, carried out in the Summer of 2007, was deemed necessary because, among others, of the griminess of the indoor vault that was coated with an oily thickness of smog (Figure 49a) [123]. After a thorough clean-up, the tunnel walls were coated with a photocatalytic cement-based paint, with the gray paint applied on both sides of the tunnel (up to 1.80 m height from the road surface) and with a white paint for the remaining surface (total surface, 9000 m2)—it is not clear, however, what the source of the paints was (Cimax Ecosystem Paint by Calci Idrate Marcellina srl, a photovoltaic product patented by Italcementi [123], or Airlite paint, as claimed by Ref. [143]). Pollution and weather conditions were expected to be less variable in the indoor vault of the tunnel than in a typical outdoor environment, thus the researchers thought would facilitate the evaluation of the photocatalytic depolluting action of the photocatalytic grey and white paints with a reduced number of parameters and making the interpretation of the data less complex. During the ﬁrst campaign of July 2007, the extent of NOx abatement was determined to be 20–25% in the center of the tunnel, whereas in the second campaign of September–October 2007, the NOx abatement level was estimated at 51–64% based on a statistical approach [138]. Regardless, Figure 49b [123] displays a photograph of the renovated tunnel before inauguration day of 9 September 2007, while the photographs in Figure 50 show the characteristics of the Umberto I Tunnel in Rome nearly 11 years after the renovation. Evidently, the long-term effectiveness of the photocatalytic product is also an issue here, not to mention the degradation (blisters) of the painted cementitious walls. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment 479 Catalysts 2018, 8, 553 (a) (b) (c) Catalysts 2018, 8, 553 (a) (b) (c) (d) (e) Figure 48. (a) Google Earth photograph of the Dives in Misericordia Church in Rome, Italy (unloade Spring of 2017); and (b–e) photographs of the Dives in Misericordia Church taken on 24 February 201 (Copyright by N. Serpone). Note the breaking-up of the cementitious layer on the outside sails. (a) (b) (c) (b) (c) (d) (e) (d) (e) Figure 48. (a) Google Earth photograph of the Dives in Misericordia Church in Rome, Italy (unloaded Spring of 2017); and (b–e) photographs of the Dives in Misericordia Church taken on 24 February 2018 (Copyright by N. Serpone). Note the breaking-up of the cementitious layer on the outside sails. 480 Catalysts 2018, 8, 553 (a) (b) Figure 49. (a) Conditions of the Umberto I tunnel in Rome (Italy) before the 2007 renovation; and (b) tunnel after renovation but before inauguration day of 9 September 2007. Reproduced from Guerrini and Peccati [123]. (a) (a) (a) (b) (b) Figure 49. (a) Conditions of the Umberto I tunnel in Rome (Italy) before the 2007 renovation; and (b) tunnel after renovation but before inauguration day of 9 September 2007. Reproduced from Guerrini and Peccati [123]. As the adsorbed nitrates (NO3−) formed subsequent to the oxidative conversion of the NOx gases on the surface of the photocatalytic cement-based paints cannot be washed off from the tunnel’s indoor walls, a problem arises with regard to the chemistry that occurs on the walls as accumulation of adsorbed NO3−on the catalytic surfaces block the catalytic sites from further NO/NO2 adsorption—thus, not only is the catalytic activity of the photocatalytic paint affected, but also the NO3−may be reduced catalytically back to NO2. In other words, reNOxiﬁcation is likely to occur subsequent to the deNOxiﬁcation process, unless the adsorbed nitrates are removed regularly by washings [132], not possible by natural means in a tunnel conﬁguration unless done manually and periodically by appropriate maintenance of the infrastructure (best done on a weekly basis, also because the lamps will be covered with dirt and dust from the trafﬁc (see, e.g., Figure 57), thereby diminishing the UV irradiance of the lamps—compare Figures 49 and 50). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The principal goal of the European funded PICADA project (Photocatalytic Innovative Coverings Applications for Depollution Assessment) [134] was the development of a range of innovative materials that could easily be applied to various structures (buildings, tunnels, streets, roadside noise barriers, etc.) and that possessed despoiling and depolluting capabilities such as the TiO2 semiconductor photocatalyst (in powder format) that had earlier shown excellent photocatalytic properties toward the oxidative/reductive destruction of organic contaminants in aqueous media and in air (VOCs), removal of toxic metals (Hg and Pb), recovery of precious metals (Ag, Au, Pt, Pd, and Rh) [144–148], and self-cleaning capabilities when deposited on glass and other supports 481 Catalysts 2018, 8, 553 (in colloidal format). The PICADA project involved a consortium (eight partners) that assembled both industry and research institutes from Greece, France, Italy, Denmark and Great Britain, and whose two main objectives were: (i) to develop and optimize industrial formulations of TiO2 and application methodologies; and (ii) to establish local behavior models under different exposure conditions in realistic urban environments [134]. (in colloidal format). The PICADA project involved a consortium (eight partners) that assembled both industry and research institutes from Greece, France, Italy, Denmark and Great Britain, and whose two main objectives were: (i) to develop and optimize industrial formulations of TiO2 and application methodologies; and (ii) to establish local behavior models under different exposure conditions in realistic urban environments [134]. Figure 50. Photographs showing the status of the Umberto I Tunnel in Rome nearly 11 years after the renovation. Photographs taken 4 March 2018 at 17:00 (Copyright by N. Serpone). V b d d d d h h l d d T O Figure 50. Photographs showing the status of the Umberto I Tunnel in Rome nearly 11 years after the renovation. Photographs taken 4 March 2018 at 17:00 (Copyright by N. Serpone). Figure 50. Photographs showing the status of the Umberto I Tunnel in Rome nearly 11 years after the renovation. Photographs taken 4 March 2018 at 17:00 (Copyright by N. Serpone). Various cement-based and ready-to-use products within this project involved nano-sized TiO2 and suitable additives/binders with products to be developed with regard to desoiling and depolluting efﬁciencies, ease-of-use, durability, and cost effectiveness; Table 8 gives a brief description of these products. 482 Catalysts 2018, 8, 553 Table 8. Preparations and properties of the range of water-based products developed within the PICADA Project. Table 8. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Laboratory tests of these products toward NOx destruction were carried out in a 0.45-m glass chamber equipped with a UV lamp 50 cm above the eight petri dishes and a fan to circulate the inlet air/NOx mixture (NOx, 200 ppb); the materials displayed signiﬁcant performance toward destruction of NOx (especially NO) that disappeared completely at the end of the experiments. By comparison, tests toward the destruction of NO2 were carried out in a 20 L glass chamber with one petri dish per experiment; UV lamp was located 25 cm above the dish, air/NO2 mix (NO2, 200 ppb) was recirculated within a closed circuit; nitrite and nitrate ions were detected. Materials again performed well toward destruction of NO2 with an efﬁciency ca. 10 times greater than a control material without TiO2 [134]. Subsequently, three of the products (the 10-mm thick B1 mortar, the 1-mm thick B2 mineral coating, and the translucent C1 coating) were coated on a 4-m2 glass surface and tested in a much larger chamber (23 ◦C; relative humidity, 50%) equipped with a fan to keep the air/NOx (NOx, 200 ppb) circulating; lamp was located 150 cm above the samples. Results of the latter tests showed that in all cases ca. 80% of NO was destroyed. The TiO2-treated B1 mortar was also tested in a pilot-scale three-canyon streets (dimensions: 18 m long, 2 m wide and 5 m high) near Paris; NOx gas was emitted from an engine into the test and reference canyons by a perforated pipe. The B1 mortar reduced NOx by 40–80% relative to the reference canyon without the B1 test material [134]; another report with more details is also available [149]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Preparations and properties of the range of water-based products developed within the PICADA Project. Products Notation Precursor Materials Properties Applications Dry mix (water added on site) B1 Sand; White ordinary Portland cement; lime; TiO2; mineral binder 10 mm thick decorative mortar Façade coatings; Interior applications B2 Filler; White ordinary Portland cement; TiO2; mineral binder 1 mm mineral rendering; 100 μm thick paint Thin decorative mineral façade coating; paint; Interior applications Ready-to-use C1 Siloxane binder; TiO2; CaCO3 ﬁller Translucent; resistant to photocatalytic effect of TiO2 Coating C2 Siloxane binder, TiO2 Opaque coating Paints C3 Acrylic binder; TiO2 Opaque coating Paints; Indoor applications C4 Silicate binder; TiO2 Opaque coating Paints Laboratory tests of these products toward NOx destruction were carried out in a 0.45-m glass chamber equipped with a UV lamp 50 cm above the eight petri dishes and a fan to circulate the inlet air/NOx mixture (NOx, 200 ppb); the materials displayed signiﬁcant performance toward destruction of NOx (especially NO) that disappeared completely at the end of the experiments. By comparison, tests toward the destruction of NO2 were carried out in a 20 L glass chamber with one petri dish per experiment; UV lamp was located 25 cm above the dish, air/NO2 mix (NO2, 200 ppb) was recirculated within a closed circuit; nitrite and nitrate ions were detected. Materials again performed well toward destruction of NO2 with an efﬁciency ca. 10 times greater than a control material without TiO2 [134]. Subsequently, three of the products (the 10-mm thick B1 mortar, the 1-mm thick B2 mineral coating, and the translucent C1 coating) were coated on a 4-m2 glass surface and tested in a much larger chamber (23 ◦C; relative humidity, 50%) equipped with a fan to keep the air/NOx (NOx, 200 ppb) circulating; lamp was located 150 cm above the samples. Results of the latter tests showed that in all cases ca. 80% of NO was destroyed. The TiO2-treated B1 mortar was also tested in a pilot-scale three-canyon streets (dimensions: 18 m long, 2 m wide and 5 m high) near Paris; NOx gas was emitted from an engine into the test and reference canyons by a perforated pipe. The B1 mortar reduced NOx by 40–80% relative to the reference canyon without the B1 test material [134]; another report with more details is also available [149]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Gurol [150] examined the results of the PICADA project and concluded (at that time) that there were several unanswered questions that needed to be addressed in a laboratory environment with regard to reaction rates, mechanisms, reaction products, fate of reaction products, and types of pollutants; the effect of various variables that include TiO2 particle size, type of TiO2, percent loading of TiO2 in the mixture, thickness of the mixture required (penetration distance of reactions), temperature, humidity, and concentration of pollutants; identiﬁcation and quantiﬁcation of reaction products under various indoor and outdoor conditions to establish that no hazardous chemicals were released from photocatalytic reactions; determination of the useful life of the TiO2-containing material; and evaluation of the effectiveness of TiO2 in a colored matrix as all buildings are not necessarily painted white. In addition, laboratory experiments ought to be conducted in a fully-controlled system that can operate under steady-state and continuous-ﬂow conditions (as opposed to batch lab experiments) to be able to simulate realistic conditions. Furthermore, Gurol [150] recommended that a conceptual process model be developed to describe mass transport and reaction of various pollutants under various realistic scenarios for outdoor atmospheric conditions, together with evaluating the sensitivity of the model to all possible variables and atmospheric conditions. Whether these issues were considered and examined is typically not disclosed by industries, as they too often maintain that the results are proprietary. Nonetheless, additional concerns with regard to outdoor applications remain to this day: (i) the long-term effectiveness of TiO2-containing materials; (ii) the possible desorption of pollutants at night time; (iii) the decrease of the effectiveness of TiO2 over time if adsorbed reaction products and 483 Catalysts 2018, 8, 553 chemicals are not washed off from the surface; and (iv) formation of ozone during the degradation of NO2 as reported by Maggos et al. [151]. Field studies undertaken in 2005–2009 in the Netherlands by an international panel of air quality experts sought innovative solutions to improve air quality on and around motorways in densely populated bottlenecks (hot spots) along some of the Dutch motorways. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The large 484 Catalysts 2018, 8, 553 ﬂuctuations in the individual values cast doubt as to whether the results were signiﬁcant [153]. It must be noted that dirt, dust and oily layer on the sidewalks caused by the trafﬁc and pedestrians had a non-insigniﬁcant impact on reducing the level of NOx emitted. By contrast, Cristal Global (producer of CristalACTiV™; anatase TiO2 nanoparticulate powder or colloidal sols) carried out a trial in 2006 in which TiO2 was incorporated into paving slabs laid down at either side along the length of a street (covering ca. 1200 m2) in the Borough of Camden (London, UK); a chemiluminescence analysis of NOx showed a reduction of ca. 20% (no other details were given) [154]. A ﬁeld trial conducted at a train station in Manila (Philippines) by coating a 4100-m2 exterior wall with TiO2-based photocatalytic paint (Cristal Global) showed that about 26 g of NOx per 100 m2 of painted surface was removed; it was also claimed that each painted square meter could remove 80 g of NOx per year [154,155]. In another trial that ran over a four-year period, a 135 m2 wall in London was treated with a Cristal photocatalytic paint; the company claimed a reduction of 60% of NO and 20% of NO2 in the vicinity of the wall (again short in details regarding the exact physical steps taken in the trials). In addition, a depolluting 12-month trial in an indoor car park in 2007 in Paris, in which two Cristal specialist paints were used with a very active photocatalyst within the matrix of the coating, revealed (analysis of monthly nitrate accumulation) reduction levels of NOx between 53% and 99% depending on paint type and lighting levels [154]. However, another trial carried out, this time in a courtyard behind the Central St. Martin’s College of Art & Design (London, UK) by Cristal Global with one of its photocatalytic coatings, showed that, subsequent to chemiluminescence monitoring of NO, NO2 and NOx for nearly two years after the application of the photocatalytic coating, NOx reductions amounted to 35–65% depending on time of year and local weather conditions [154]. With regard to processes that might take place on paint surfaces, Laufs and coworkers [156] examined the photocatalytic reactions of nitrogen oxides (NOx = NO + NO2) on commercial TiO2-doped façade paints in a ﬂow-tube photoreactor under simulated atmospheric conditions in a laboratory setting. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment To this end, two series of practical trials were conducted using four panels with different TiO2 coatings ﬁtted to an existing noise barrier along the A1 highway at Terschuur, and on a later erected TiO2-coated porous concrete noise barrier along the A28 motorway at Putten (Figure 51); measurements of NO2 and NOx performed under various weather conditions at both locations showed very low conversion rates of the NOx gases [152] that were attributed to the short contact time between air and barrier, to the relatively unfavorable meteorological conditions (wind direction and light intensity), to the high relative humidity, and to frequent low temperatures in the Netherlands. The experts concluded that improved air quality with TiO2 coatings on concrete had not been demonstrated! Figure 51. Example of a TiO2-coated concrete noise barrier erected along the A1 and A28 highways in Terschuur and Putten, the Netherlands. Reproduced from Ref. [152]. Figure 51. Example of a TiO2-coated concrete noise barrier erected along the A1 and A28 highways in Terschuur and Putten, the Netherlands. Reproduced from Ref. [152]. Road rehabilitation of Petersbergerstraße in Fulda (Hesse, Germany) was undertaken to investigate possible effects on the concentration of NO2 emission by TiO2-coated photocatalytic paving stones installed on sidewalks (ca. 800 m) on both sides of the street [153]. Measurements were carried out at two points (Points 1 and 2) across from each other and taken in June 2010 (Point 1) and in October 2010 (Point 2) at two different heights: 10 cm and 3 m above the road surface; for comparison, similar measurements were undertaken beyond the 800-m stretch as a reference. The average trafﬁc volume on this street amounted to about 24,000–30,000 vehicles per day; proportion of light and heavy trucks and buses was 3.2% and 2.1% and 0.2%, whereas the car share was 88%, and motorcycles 6.5%. The whole TiO2-coated surface was ca. 4500 m2, a relatively small proportion compared to the remaining road surface of ca. 15,000 m2. Relative to the reference section, the NO2-reducing effectiveness appeared somewhat greater directly above the pavement surface (at 10 cm) than at the 3 m height; at Point 2 the effectiveness was, respectively, 17% and 9%, whereas at Point 1 the effectiveness was 3.5% at 10 cm above ground while the NO2 level remained virtually unchanged at 3 m above. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Both NO and NO2 were rapidly converted photocatalytically, albeit only on the photocatalytic paints but not on non-catalytic reference paints. Nitrous acid (HONO) was formed in the dark on all paints examined; however, HONO decomposed efﬁciently under irradiation only on photocatalytic samples, so that photocatalytic paint surfaces did not represent a daytime source of HONO, contrary to pure TiO2 surfaces. Formation of adsorbed nitric acid/nitrate anion (HNO3/NO3−) occurred with near unity yield. The mechanism proposed by Laufs et al. for the photocatalytic reactions of NO, NO2, and HONO carried out on photocatalytic paint surfaces in a laboratory setting is displayed in Figure 52 [156]. Figure 52. Postulated mechanism for the photocatalytic reactions of NO, NO2 and HONO on photoactive paint surfaces. Reproduced with permission from Laufs et al. [156]. Copyright 2010 by Elsevier Ltd. (License No.: 4453150462958). Figure 52. Postulated mechanism for the photocatalytic reactions of NO, NO2 and HONO on photoactive paint surfaces. Reproduced with permission from Laufs et al. [156]. Copyright 2010 by Elsevier Ltd. (License No.: 4453150462958). 485 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 A full-scale outdoor ﬁeld demonstration of air purifying pavement in Hengelo (The Netherlands) was carried out by Ballari and Brouwers [157] on the full width of a street surfaced with a concrete pavement containing C-doped TiO2 (Kronos International; 4% w/w TiO2 water suspension; 50 L of suspension covered a 750 m2 surface; TiO2 loading was 2.67 g m−2) sprayed over a length of 150 m; for comparison, another part of the street (ca. 100 m) was paved with normal paving blocks; outdoor monitoring was done during 26 days over a period of more than a year. Prior and during the ﬁeld measurements, the used blocks were examined simultaneously in a laboratory setting (small ﬂow-through reactor) to assess performance. The ﬁrst coating applied May 2010 gave good results in the laboratory setting (7.7% of NO and 6.9% of NOx under visible light illumination) and in the ﬁeld. NOx levels were sampled at different heights: 5 cm (near the active surface), 30 cm (car exhaust height) and 1.5 m (the breathing zone) to assess the extent of deNOxing as a function of distance from the active surface. Unfortunately, the TiO2 photocatalytic coating vanished after the blocks were exposed outdoors for 2.5 months due to normal wear, to vehicular trafﬁc, to weather, and/or else due to solid dust and dirt deposits on the surface. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment A second coating was subsequently applied to the surfaced road in September 2010; after being exposed for 1.5 months to the street environment, the photocatalytic performance returned to values of the ﬁrst coating (laboratory testing). On average, the extent of NOx converted (to nitrates) outdoors determined by the chemiluminescence technique was 19.2 ± 17.8% (daily readings) and 28.3 ± 20.0% (afternoon readings)—note the high ﬂuctuations in the samplings, not typically reported by many researchers; under ideal weather conditions (high radiation; low relative humidity) the decrease of NOx was 45% [157]. y We have seen thus far that although NOx levels can be reduced effectively with TiO2-based photocatalytic surfaces in an indoor laboratory setting, signiﬁcant variable results have been more the rule than the exception with regard to NOx reduction levels in an outdoor urban setting. For instance, NOx reduction levels of 40–80% [134,136,149], 26–66% [137], and ca. 19% [157] have been reported in various outdoor ﬁeld trials; however, as we have seen in the ﬁeld trial in Putten (The Netherlands) [152], the measured NOx reductions were at or well below detection limits. Such variations are likely the result: (i) of limited contact between the pollutant NOx and the photo-catalytic surface; (ii) of variable features of ﬁeld sites; (iii) of prevailing local atmospheric conditions (wind velocity, wind direction, relative humidity, light intensity, etc.); (iv) of the time of measurements; and (v) of the nature and source of TiO2 photocatalytic products and their associated reactivity upon photo-activation. Attempts to reduce NOx levels in an external “indoor” environment such as a road tunnel, as was the case with the Umberto I tunnel in Rome [123,138], were expected to bring about certain beneﬁts: reduced health issues of pedestrians that use the tunnel, no need for a ventilation system, and decrease of the contribution of pollutants to the surrounding areas. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment To ascertain the beneﬁts of photocatalysis toward NOx reductions in such indoor infrastructures and as part of ﬁeld studies within the PhotoPAQ program, Gallus and coworkers [14] undertook an extensive ﬁeld study of the Leopold II tunnel in Brussels (2011–2013) where photocatalytic cementitious coating materials (ﬁrst campaign; Italcementi’s TX-Active® Skim Coat) were applied on the side walls and ceiling of a test section (73 m long) of one tube of the Leopold II tunnel (Figure 53) [158], followed by a monitoring campaign and later by a third campaign in which the same section was extended to 160 m and covered by a novel photocatalytically more active mortar (Italcementi’s TX-Active Skim Coat Boosted) [14]. Installed UV lighting (wavelength range, 315–420 nm) had an average irradiance on the active surfaces of 0.6 ± 0.3 W m−2 (TX-Active; second campaign) and 1.6 ± 0.8 W m−2 (TX-Active Boosted; third campaign). 486 Catalysts 2018, 8, 553 Figure 53. (upper) Condition of the Leopold II tunnel after coating the walls and also showing the lighting system; and (lower) schematic representation of the test sites in the Leopold II tunnel during the PhotoPAQ ﬁeld trials. Reproduced with permission from Boonen et al. [158]. Copyright 2015 by Elsevier Ltd. (License No.: 4453110422490). Figure 53. (upper) Condition of the Leopold II tunnel after coating the walls and also showing the lighting system; and (lower) schematic representation of the test sites in the Leopold II tunnel during the PhotoPAQ ﬁeld trials. Reproduced with permission from Boonen et al. [158]. Copyright 2015 by Elsevier Ltd. (License No.: 4453110422490). Most interesting in the ﬁeld trial of Gallus et al. [14] was the highly recommendable experimental approach used to assess the level of photocatalytic reduction of NOx gases inside the tunnel based on the realization that changes in atmospheric compositions, trafﬁc ﬂow, and dilution (i.e., different wind speeds) occur inside the road tunnel. Consequently, the authors monitored NOx concentrations and normalized them to the photocatalytically inert tracer CO2 gas that was emitted by the vehicular trafﬁc at the same time as the NOx (and others) to yield NOx/CO2 concentration ratios; the instruments used throughout the campaigns had typical detection limits of 1–2 ppb, precisions of ~ 1% and accuracies of ~ 7% and ~ 10% for NO and NO2, respectively; CO2 levels were measured by a nondispersive infrared absorption technique: detection limit, 2 ppm; precision, 1%; accuracy, 7%. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The NOx/CO2 ratios so obtained from the slopes of plots of NOx against CO2 were independent of the absolute pollution level and thus of the emissions from the vehicular source and variable dilution inside the tunnel. Comparison of such slopes obtained in a control setting (dark conditions or otherwise under light illumination but without the TiO2 photocatalyst) relative to a setting in which the slopes of NOx versus CO2 plots obtained under light illumination should show signiﬁcant variations if NOx were photocatalytically converted (to nitrates and others, such as HONO). Moreover, changes in the NOx/CO2 ratio—e.g., between upwind and downwind of the active section—could then be attributed to a photocatalytic remediation of the pollutants. The demonstration of NOx levels at different heights 487 Catalysts 2018, 8, 553 within the tunnel were also relevant in their approach; in this regard, Figure 54 shows NOx levels at Site 2 (under normal trafﬁc conditions, only air at this site was in contact with the photocatalytic surfaces) were sampled at 1.1 m and 3.2 m above the street level and then plotted against the corresponding data at the tunnel ceiling (4.4 m above the street level) [14]. No signiﬁcant gradients were observed at Site 2 under normal driving conditions with similar concentrations at the different heights above the street (note the slopes of the lines are nearly unity for data at the two heights). Figure 54. Plot of NOx concentration (10-min averages) monitored at 1.1 m and 3.2 m above the ground surface against the corresponding data at the tunnel ceiling (4.4 m); data recorded at Site 2 (downwind; see Figure 53 (lower)) during the third campaign (160 m section). Reproduced with permission from Gallus et al. [14]. Copyright 2014 by Elsevier Ltd. (License No.: 4453111244188). Figure 54. Plot of NOx concentration (10-min averages) monitored at 1.1 m and 3.2 m above the ground surface against the corresponding data at the tunnel ceiling (4.4 m); data recorded at Site 2 (downwind; see Figure 53 (lower)) during the third campaign (160 m section). Reproduced with permission from Gallus et al. [14]. Copyright 2014 by Elsevier Ltd. (License No.: 4453111244188). Figure 54. Plot of NOx concentration (10-min averages) monitored at 1.1 m and 3.2 m above the ground surface against the corresponding data at the tunnel ceiling (4.4 m); data recorded at Site 2 (downwind; see Figure 53 (lower)) during the third campaign (160 m section). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Comparison of all upwind (Site 1) and downwind (Site 2) data from the second campaign showed a minor formation of NOx (70 m test section; TX-Active used), whereas a small reduction was inferred at Site 2 from the third campaign (160-m section; TX-Active Boosted used) in qualitative accord with laboratory experiments on sample plates exposed to tunnel air. In the latter experiments, the authors [14] observed photocatalytic formation of NOx on dirty tunnel samples when the TX-Active material was used (second campaign), whereas only a very small reduction in NOx was measured on sample plates exposed to tunnel air when TX-Active Boosted was used. An examination of the combined errors of the NOx/CO2 ratios together with the precision of the duplicate instruments used (±2% for NOx) (see Table 9) led the authors to deduce that the observed differences in results from the second and third tunnel campaigns were insigniﬁcant. To substantiate whether there was a potentially small photocatalytic reduction of NOx from the upwind/downwind data from the third campaign, the authors [14] further examined the inﬂuence of reaction time on the photocatalytic NOx remediation in the tunnel, in which a greater photocatalytic reduction was expected at lower wind speeds because of longer reaction times (residence times). However, no photocatalytic reduction of NOx was observed on comparing upwind and downwind data for both sets of wind speeds (or reaction times), thereby precluding any (expected) increase in reduction of NOx pollution at the lower wind speed. Comparing the NOx/CO2 ratios obtained with lights ON (active) and lights OFF (non-active) also showed no quantiﬁable remediation beyond experimental uncertainty, as also observed when only low wind speed data were used for which highest reduction was expected (Table 9). Accordingly, in view of the precision errors in the data analysis, the authors concluded that the extent of photocatalytic NOx remediation in the 160 m more active tunnel section from the third campaign was at best ≤2% [14]. A theoretical model showed that, in accord with the 20% NOx reduction observed in the 350 m long tunnel in Rome [123,138], a reduction upper limit of ≤20%, would have been possible in the 160 m test section of the Leopold II tunnel conditions as relative humidity (RH, 50%), wind speed (1 m s−1) and UVA light irradiance (10 W m−2) were optimal, and deactivation of the tunnel’s photocatalytic surfaces was disregarded. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Reproduced with permission from Gallus et al. [14]. Copyright 2014 by Elsevier Ltd. (License No.: 4453111244188). Plotting the NOx data from the downwind Site 2 of the third campaign against CO2 data yielded a slope that gave a NOx/CO2 ratio of (3.09 ± 0.04) × 10−3 (Figure 55). More comprehensive results obtained under various conditions are summarized in Table 9 [14]. Plotting the NOx data from the downwind Site 2 of the third campaign against CO2 data yielded a slope that gave a NOx/CO2 ratio of (3.09 ± 0.04) × 10−3 (Figure 55). More comprehensive results obtained under various conditions are summarized in Table 9 [14]. Figure 55. Plot of NOx against CO2 data (10-min averages) at Site 2 during the third campaign. Reproduced with permission from Gallus et al. [14]. Copyright 2014 by Elsevier Ltd. (License No.: 4453111244188). Figure 55. Plot of NOx against CO2 data (10-min averages) at Site 2 during the third campaign. Reproduced with permission from Gallus et al. [14]. Copyright 2014 by Elsevier Ltd. (License No.: 4453111244188). 488 Catalysts 2018, 8, 553 Table 9. Average NOx/CO2 ratios obtained from experimental results in the photocatalytic remediation of NOx in the Leopold II tunnel in Brussels; standard errors represent 2σ . Table 9. Average NOx/CO2 ratios obtained from experimental results in the photocatalytic remediation of NOx in the Leopold II tunnel in Brussels; standard errors represent 2σ . Average NOx/CO2 Ratios from Site 1 and Site 2 in the Second and Third Campaigns with Photocatalytically-Active Surfaces and with Lights ON Campaign Photocatalytic Material (TiO2) Site 1—upwind (×10−3) Site 2—downwind (×10−3) Second; 09/2011 TX-Active 3.03 ± 0.06 3.14 ± 0.07 Third; 01/2013 TX-Active Boosted 3.18 ± 0.08 3.10 ± 0.05 Average NOx/CO2 Ratios for High and Low Wind Speeds with Lights ON in the Third Campaign Wind speed Site 1—upwind (×10−3) Site 2—downwind (×10−3) Low <2 m s−1 3.24 ± 0.13 3.23 ± 0.12 High >2 m s−1 2.93 ± 0.20 2.95 ± 0.10 Average NOx/CO2 Ratios at Downwind Site 2 with Lights ON or OFF in the Third Campaign Data used Lights OFF Lights ON All 3.09 ± 0.06 3.10 ± 0.05 Low wind speed <2 m s−1 3.25 ± 0.10 3.23 ± 0.12 High wind speed >2 m s−1 2.99 ± 0.11 2.95 ± 0.10 Adapted from Gallus et al. [14]. Copyright Elsevier Ltd. (License No.: 4453120115447). 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Unfortunately, deactivation of these surfaces was not insigniﬁcant under the heavily polluted tunnel conditions as demonstrated by laboratory experiments, not to mention that UVA irradiances of 0.6 and 1.6 W m−2 were far below the targeted value above 4 W m−2, which was therefore a contributing factor in the deactivation phenomenon and further decreased the photocatalytic activity. Moreover, typical wind speed (ca. 3 m s−1 during daytime) and the cold and humid (RH, 70–90%) conditions during the third campaign of January 2013 also caused a decrease of the activity of the photocatalytic material. Interestingly, another simple model calculation [14] that used uptake kinetics determined from laboratory experiments under the polluted tunnel conditions indicated an upper limit of only if such 0.4% for the photocatalytic NOx remediation, in fairly good 489 Catalysts 2018, 8, 553 accord with the experimental tunnel results of <2%. Not to be neglected, photocatalytic degradation of NO led to signiﬁcant formation of NO2, particularly under humid conditions, thereby lowering the expected NOx reduction even further. Thus far, the description of some ﬁeld trials carried out under a variety of experimental approaches and conditions that could not be controlled, as the trials were performed in an open air environment, have led to a large variation in results associated with the extent of TiO2-based photocatalytic removal of NOx gases. Accordingly, it is worth summarizing the results, which are collected in Table 10 [14,15,134,136–139,149,152–157,159,160]. Clearly, the results are all over the place. Table 10. Percent removal of NOx gases in ﬁeld trials carried out with TiO2-based photocatalytic protocols in an open outdoor environment and under a variety of conditions. Table 10. Percent removal of NOx gases in ﬁeld trials carried out with TiO2-based photocatalytic protocols in an open outdoor environment and under a variety of conditions. Field Trial Gas Extent of Photocatalytic Removal Structures Year Ref. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment 1 NOx 37–82% Model street canyon 2008 [136] 2 NOx 30–40% Local street 2007 [137] 3 NOx 20–50% Street 2012 [138] 4 NOx 45–57% Street 2007 [137] 5 NOx ≤2% Model street canyon, Tunnel 2015, 2015 [14,15] 6 NOx > 20% Tunnel 2012 [138] 7 (a) NO 55% Church external wall 2004 [139] NO2 32% Church external wall 8 NOx 20–25% Tunnel 2012 [138] NOx 51–64% Tunnel 9 NOx 40–80% Model street canyon 2006, 2008 [134,136,149] 10 NOx Not-measurable Highways’ noise barriers 2010, 2009 [152,160] 11 NO2 0–17% Sidewalks/street 2012 [153] 12 NOx 20% Street 2006 [154] 13 NO 60% Building external wall 2006, 2012 [154,155] NOx 20% Building external wall 14 NOx 53–99% Car park 2006 [154] 15 NOx 35–65% Building external wall 2006 [154] 16 NOx 1–37% Street 2013 [157] NOx 8–48% Street 17 NOx 26–66% Real urban street canyon 2007 [137] 18 NOx 25–30% Model street canyon 2010 [159] 19 NOx 40–80% Model street canyon 2006, 2008 [136,149] 20 NOx 19% Real urban street canyon 2013 [157] 21 NOx Not-measurable Street 2012 [153] (a) Photocatalytic material (TX-Active) tested in a laboratory setting. It has been reported time and again that photocatalysis, especially with TiO2-based commercial products, could degrade environmental pollutants as shown, within the present context, by the removal of NOx gases from the environment not only in a laboratory setting (Table 7), but to some extent also outdoors under environmental conditions (Table 10). Taking artiﬁcial model street canyons as examples for the removal of NOx, we have seen that the reported extent of NOx removal varies from 25–30% to 40–80% and 37–82%, while in a real urban street canyon values of 19% and 26–66% have been reported (Table 10). By contrast, similar TiO2-based products have shown no effect in removing environmental NOx gases in urban streets or on the highways’ noise barriers. All the outdoors ﬁeld trials expected the NOx to be transformed into nitrates on the photocatalytic surfaces ultimately to be desorbed when raining. However, there are also reports that other intermediate species are likely to form also such as nitrous acid (HONO) [156,161–164], which is far more toxic than the NOx pollutants, and not least is the potential for reNOxiﬁcation and formation of ozone from the reaction of adsorbed nitrates with reducing agents (TiO2 conduction band electrons) [164,165]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Plots of the 10 min averaged data from the active canyon against the reference can for NO, NO2 and NOx gases during nighttime and daytime periods. Reproduced with permission fr Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 44531208873 Table 11. Slopes from the plots of daytime and nighttime results for NO, NO2 and NOx gases from active canyon versus the reference canyon; a negative value implies a photocatalytic remediation th t ll t t Figure 57. Plots of the 10 min averaged data from the active canyon against the reference canyon for NO, NO2 and NOx gases during nighttime and daytime periods. Reproduced with permission from Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453120887380). Table 11. Slopes from the plots of daytime and nighttime results for NO, NO2 and NOx gases from the Figure 57. Plots of the 10 min averaged data from the active canyon against the reference canyon for NO, NO2 and NOx gases during nighttime and daytime periods. Reproduced with permission from Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453120887380). Table 11. Slopes from the plots of daytime and nighttime results for NO, NO2 and NOx gases from the active canyon versus the reference canyon; a negative value implies a photocatalytic remediation of h ll Figure 57. Plots of the 10 min averaged data from the active canyon against the reference canyon for NO, NO2 and NOx gases during nighttime and daytime periods. Reproduced with permission from Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453120887380). Table 11. Slopes from the plots of daytime and nighttime results for NO, NO2 and NOx gases from the active canyon versus the reference canyon; a negative value implies a photocatalytic remediation of that pollutant . Table 11. Slopes from the plots of daytime and nighttime results for NO, NO2 and NOx gases from the active canyon versus the reference canyon; a negative value implies a photocatalytic remediation of that pollutant . Gaseous Pollutant Daytime Results Nighttime Results (Day—Night) × 100 (% ± 2σ) NO 0.953 ± 0.016 0.988 ± 0.029 −3.5 ± 3.3 NO2 0.982 ± 0.013 0.966 ± 0.014 +1.6 ± 1.9 NOx 0.984 ± 0.011 0.970 ± 0.014 +1.4 ± 1.8 Reproduced with permission from Gallus et al. [15]. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment For these very reasons and as part of the PhotoPAQ investigative program, Gallus and coworkers [15] structured a two-step campaign to investigate the fate of NOx gases outdoors in two artiﬁcial model street canyons in Petosino (Italy) (see Figure 56). In the ﬁrst campaign, both model canyons had their side walls covered with a photocatalytically inactive ﬁbrous cement, while in the second campaign one of the canyons had its side walls and ground surface covered with a photocatalytic cementitious coating material (Italcementi’s TX-Active Skim Coat Boosted), while 490 Catalysts 2018, 8, 553 the other was used as the reference canyon; the experimental approach used was otherwise similar to the approach used to examine the fate of NOx in the Leopold II tunnel [14]. The results of their investigations are reproduced herein to demonstrate, what we consider the best approach, their highly recommended protocol and their actual results, not simply the authors’ conclusions [15]. Figure 56. Image and schematic of the two artiﬁcial model street (reference and active) canyons with dimensions 5 × 5 × 53 m (width × height × length) at an Italcementi industrial site in Petosino near Bergamo, Italy. Reproduced with permission from Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453120887380). Figure 56. Image and schematic of the two artiﬁcial model street (reference and active) canyons with dimensions 5 × 5 × 53 m (width × height × length) at an Italcementi industrial site in Petosino near Bergamo, Italy. Reproduced with permission from Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453120887380). Results from monitoring and analyzing the NO, NO2 and NOx gases (10-min averaged values) from the active canyon were plotted against the results obtained in the reference canyon (Figure 57), with the expectation that differences in the respective slopes (Table 11) between daytime data and nighttime data would reﬂect the photocatalytic effect independent of any artiﬁcial differences. Perusal of the slopes that are summarized in Table 11 indicates that only for the conversion of NO was there a hint of a photocatalytic effect (−3.5 ± 3.3%). In addition, comparison of the daytime and nighttime results led the authors [15] to infer that less than 2% (that is, ≤2%) of the NOx was converted photocatalytically to nitrates (Table 12). 491 Catalysts 2018, 8, 553 Figure 57. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment In the tunnel case, the cementitious surfaces were allowed to cure in the dark for several days, whereas photocatalytic and non-photocatalytic surfaces in the canyons were freshly exposed to sunlight UV radiation. 3. In the tunnel case, the cementitious surfaces were allowed to cure in the dark for several days, whereas photocatalytic and non-photocatalytic surfaces in the canyons were freshly exposed to sunlight UV radiation. Figure 58. Sample of a photoactive TiO2-based cementitious surface before and after one week in the Leopold II tunnel in Brussels. Courtesy of Dr. Falk Mothes of the Leibnitz Institute for Tropospheric Research (TROPOS), Germany. Figure 58. Sample of a photoactive TiO2-based cementitious surface before and after one week in the Leopold II tunnel in Brussels. Courtesy of Dr. Falk Mothes of the Leibnitz Institute for Tropospheric Research (TROPOS), Germany. Even though they expected to conﬁrm NOx removal levels of ca. 20%, if not greater (see Table 10), as reported in earlier ﬁeld trials in open atmospheric environments, the ﬁndings that there was no signiﬁcant photocatalytic remediation of NOx (and VOCs, O3, and particulate matter)—removal level of ca. 2% or less—was somewhat enigmatic to say the least [15], albeit consistent with non-measurable NOx reduction levels found in highways (The Netherlands) and in a real urban street (Fulda, Germany). Such minute NOx reductions were likely due to low surface-to-volume ratios (Sactive/V) of the open structures; note that the Sactive/V ratio is a very signiﬁcant factor that reﬂects the available active surface versus the space volume occupied by gaseous pollutants, i.e., the limiting heterogeneous uptake of a species onto a solid surface [166,167], as is also the associated deposition velocity. Evidently, other factors must have played a role, and/or some differences in experimental approaches might aid in explaining such very low levels of NOx reductions, namely: 1. Although the overall geometry of the PhotoPAQ canyon model was otherwise similar to earlier model street canyons, the dimensions were different with respect to smaller model sites (e.g., 5 m wide canyon versus the earlier 2.4-m wide canyons that led to unrealistically higher Sactive/V ratios and as such to higher NOx reduction levels of 20–80% (Table 10) versus ≤2%. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453121427505). Table 12. Average concentrations of NO, NO2, and NOx in both canyons for daytime (06:00–20:30) and nighttime (20:30–06:00) and relative concentration differences; negative values imply a photocatalytic effect in the active canyon . Table 12. Average concentrations of NO, NO2, and NOx in both canyons for daytime (06:00–20:30) and nighttime (20:30–06:00) and relative concentration differences; negative values imply a photocatalytic effect in the active canyon . Concentrations (ppb ± 2σ) NO NO2 NOx All data (1 −(Ref./Act.) × 100 [%] 2 ± 17 −0.7 ± 6.1 −0.4 ± 6.0 Daytime (1 −(Ref./Act.) × 100 [%] 0.6 ± 11.5 0.2 ± 6.1 0.3 ± 6.0 Nighttime (1 −(Ref./Act.) × 100 [%] 23 ± 150 −2.4 ± 5.9 −2.0 ± 5.9 Adapted with permission from Gallus et al. [15]. Copyright 2015 by Springer-Verlag Berlin Heidelberg (License No.: 4453121427505). The ﬁeld trial carried out in the Leopold II tunnel in Brussels found that the photocatalytic active surface was deactivated [14]. Accordingly, laboratory tests were undertaken to verify whether deactivation might also have occurred in the canyon trial campaign by taking samples of the active surface in the canyon (before and after) and using a ﬂow-through reactor to examine their photocatalytic 492 Catalysts 2018, 8, 553 behavior toward degradation of the gas pollutants. Results showed that no deactivation of the canyon surfaces had occurred. Consequently, the authors [15] surmised that this was due to several evident differences in the model canyon campaign versus those from the Leopold II tunnel; that is, 1. Much lower pollution levels exist in the canyon than in the tunnel; in the latter case, the greater pollutant level (very dirty ambient conditions–dirt/dust/grime/air pollutants) blocked the active sites (see Figure 58). 1. Much lower pollution levels exist in the canyon than in the tunnel; in the latter case, the greater pollutant level (very dirty ambient conditions–dirt/dust/grime/air pollutants) blocked the active sites (see Figure 58). ( g ) 2. Much higher UV irradiance was available in the canyon (sunlight UV in the canyon versus an artiﬁcial UV light system in the tunnel) causing the pollutants to degrade faster. g Much higher UV irradiance was available in the canyon (sunlight UV in the canyon versus an artiﬁcial UV light system in the tunnel) causing the pollutants to degrade faster. 3. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment [15] recommended sampling NOx at a distance of 3 m above the photoactive surface for urban network stations, contrary to earlier ﬁeld trials where sampling of NOx was done but a few centimeters from the active surface (5 cm to 1.5 m [148]; 30 cm to 1 m [137]; 50 cm [166,167])—indeed, since conversion of NOx gases to nitrates requires the NOx to be adsorbed onto the photocatalytically active surface, it is obvious that close to the surface is where the greatest change in NOx concentration will be felt—this means that, as you stroll on a photocatalytic street surface, your feet will experience a healthier environment than your head. 4. High reductions of NOx levels in earlier ﬁeld trials [134,137,149,157,166,167] likely reﬂected some differences between active sites and reference sites in terms of the quantity of NOx artiﬁcially injected into the active and reference canyons [134,136,149,166,167], whereas the ﬁeld trial in the Petosino model canyon the NOx gases were those present in the homogeneous industrial environment [15]. 5. Where the active and reference model street canyons were distant from each other [137,157] may also have resulted in non-insigniﬁcant differences in the quantity of NOx artiﬁcially injected and in pollution dispersion (e.g., different wind speeds and different wind direction), thus leading to non-insigniﬁcant uncertainties in the results. 5. Where the active and reference model street canyons were distant from each other [137,157] may also have resulted in non-insigniﬁcant differences in the quantity of NOx artiﬁcially injected and in pollution dispersion (e.g., different wind speeds and different wind direction), thus leading to non-insigniﬁcant uncertainties in the results. With regard to the latter issue, model studies have re-evaluated the daytime results from the experimental street canyon in Bergamo and determined that the upper limit of photocatalytic NOx remediation was more like 4–14% [160], rather than the claimed 26–66% [137], because of: (a) strong differences in the vehicle NOx emissions at the active and reference sites with much higher vehicle ﬂeet density at the reference site; and (b) different dispersion conditions (geometry of sites, micrometeorology). However, Gallus et al. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Note that the large reduction levels of NOx on photocatalytic TiO2 surfaces experienced in laboratory settings (Table 7) are also explained by this surface-to-volume ratio factor which is considerably greater: typically, 167 m−1 for a small ﬂow-through reactor in a laboratory setting versus 0.6 m−1 for the model street canyon in the open environment in the industrial zone of Petosino (Italy). If one were to scale down the earlier results [134,136,149,166,167] to real urban street canyon conditions, it would result in only approximately 5% NOx reduction. 2. Previous ﬁeld trials considered results of monitoring the fate of NOx pollutants during daytime hours only, contrary to other ﬁeld trials where the whole diurnal data were considered. NOx levels during nighttime were often comparable to those during daytime (higher emissions during the 2. Previous ﬁeld trials considered results of monitoring the fate of NOx pollutants during daytime hours only, contrary to other ﬁeld trials where the whole diurnal data were considered. NOx levels during nighttime were often comparable to those during daytime (higher emissions during the 493 Catalysts 2018, 8, 553 day were often compensated by stronger convective vertical dilution) so that the above estimate of ~ 5% based on daytime data was further reduced to ca. 3%, comparable to expectations typical of a main urban street [15]. 3. The distance at which the NOx gases were sampled from the photoactive surface is another important factor. Gallus et al. [15] recommended sampling NOx at a distance of 3 m above the photoactive surface for urban network stations, contrary to earlier ﬁeld trials where sampling of NOx was done but a few centimeters from the active surface (5 cm to 1.5 m [148]; 30 cm to 1 m [137]; 50 cm [166,167])—indeed, since conversion of NOx gases to nitrates requires the NOx to be adsorbed onto the photocatalytically active surface, it is obvious that close to the surface is where the greatest change in NOx concentration will be felt—this means that, as you stroll on a photocatalytic street surface, your feet will experience a healthier environment than your head. 3. The distance at which the NOx gases were sampled from the photoactive surface is another important factor. Gallus et al. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment [15] contended that, if in the model studies one were to assume a reasonable wind speed of 1 m s−1 and a realistic photocatalytic deposition velocity of NOx on the active surface of 0.3 cm s−1, the upper limit of 4% [160] would be reduced further to <2% considering diurnal averages and transport limitations and more in line with real environmental results [152,153,168]. Clearly, the above factors and discussion call attention to the necessity of some care into how the data are treated and how the data are interpreted. Several TiO2-based photocatalytic materials were also tested in 2015 by Pujadas and coworkers [169] of CIEMAT (Madrid, Spain) within the framework of the 2013–2018 LIFE MINOx-STREET European project. Tests were ﬁrst carried out in a laboratory setting to choose the most photoactive TiO2-based material (fresh sample showed 45% reduction of NO in a ﬂow-type reactor, dropping to 20% for a used street sample caused by trafﬁc and ageing over a month) so designed as to be used on bituminous mixtures with which to examine the effect of reducing NOx levels in the two-way Boulevard Paseo de la Chopera, a real urban setting in the Municipality of Alcobendas (Madrid)—conditions were: ground level wind speed, 2 m s−1; relative humidity, <65%; solar radiation, >400 W m−2; TiO2-based coated active area, 1000 m2; length of section, ca. 60 m. The experimental approach was similar to that used in the PhotoPAQ project in that the daytime and nighttime NOx collected data in the median strip at two active sites (Site 2 and Site 3) and at the inactive site (Site 4) were plotted against NOx data collected at the inactive Site 1; the plots yielded the slopes reported in Table 13 [169]. The similarities between the data collected in daytime and at nighttime during the most optimal measurement periods before and after application of the photocatalytic coating led the researchers to deduce that there was no photocatalytic reduction of NO levels, in line with observations made by others in a real urban setting (see above). 494 Catalysts 2018, 8, 553 Table 13. Slopes from correlation plots of NO data (ppb ± 1σ) from three sampling sites (two active, one inactive) against inactive Site 1 for NO concentrations before and after implementation of the photocatalytic coating. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment Diurnal Nocturnal Sampling Site Before Applying TiO2-Based Coating After Applying TiO2-Based Coating Before Applying TiO2-Based Coating After Applying TiO2-Based Coating 2 (active) 1.05 ± 0.08 1.07 ± 0.08 1.014 ± 0.005 0.983 ± 0.004 3 (active) 1.41 ± 0.17 1.03 ± 0.06 0.977 ± 0.008 0.915 ± 0.004 4 (inactive) 0.89 ± 0.06 0.90 ± 0.07 0.985 ± 0.005 0.946 ± 0.025 A real scale examination of NOx depollution by TiO2-based photocatalytic sidewalk pavement and façade was performed by Pujadas and coworkers of CIEMAT [170] in an urban environment that resembled a model street canyon, albeit narrower, in which they also found no signiﬁcant NOx reduction during their experiments in the sidewalk model in open air, even though in a laboratory setting the extent of NOx removal by the photocatalytic material was 65%. Nonetheless, a small ambient NOx reduction effect was observed on the photocatalytic façade (mimicking a building brick wall), although for only a very short time and very close to the active surface (e.g., <10 cm), and then only under very speciﬁc ambient and meteorological conditions. At a presentation of their research at a conference in Barcelona (Spain) that was performed within the MINOx-STREET project, Palacios and coworkers [171] described results on urban street experiments, as well as experiments with model sidewalks and model building façades. Two principal conclusions are worth noting from that conference: 1. A great difﬁculty was experienced in urban scenarios to establish a possible cause-effect relationship between any observed ambient NOx reductions in the presence of photocatalytic surfaces. In road bituminous pavement (the Alcobendas experiment) and sidewalk scenarios, the NOx removal effect of photocatalytic materials had not been unequivocally demonstrated experimentally; however, in the case of the façade scenario that effect was documented, albeit the effect was seen only at distances very close to the wall surface. 2. Physicochemical characteristics of heterogeneous photocatalysis, such as low quantum yield for absorption of solar radiation by TiO2, heterogeneous molecular processes, and the high dependence on ambient conditions helped to explain the weak macroscopic effects observed in open urban air. Consequently, a low NOx concentration reduction was only attained very close to photoactive surfaces with poor global incidence on ambient air. 3. 6. Concluding Remarks One of the objectives of this review article is to provide an introduction to the basic approaches and terminology inherently and commonly used in heterogeneous photocatalysis. Relevant characteristic features of solid semiconductor/insulator photocatalysts are examined, together with some initial and subsequent events that follow the absorption of photons by these solids. The relationship(s) between traditional heterogeneous catalysis, on the one hand, and photochemistry and molecular spectroscopy, on the other hand, with heterogeneous photocatalysis have been made. The closer interconnection of photocatalysis to photochemistry is emphasized. Where similar symbols are used to mean different things, they are nonetheless fully described in the context used. The other objective is to examine closely the various attempts made to apply the photocatalytic technology—albeit brieﬂy with regard to the published literature on environmental remediation of contaminated ecosystems—through incorporation of the extensively investigated TiO2 photocatalyst into various cementitious substrates (concrete, mortar, plaster, paints, etc.) applied to various infrastructures (highways, urban streets, building external walls and road tunnels, among others) and through the use of various coating materials. With regard to the latter aspect, we have witnessed the external conditions of the Dives in Misericordia Church and the Umberto I tunnel in Rome (Italy) over a decade later from when the photoactive coatings were ﬁrst applied to these structures. Evidently, deactivation of TiO2-based photoactive surfaces may become an issue over time so that, as Gallus et al. [15] aptly noted, experiments should be carried out at the ﬁeld site of interest in small scale experiments exposed to the prevailing atmospheric conditions of the site—and not in a laboratory setting miles away from the application—before giving any consideration to apply the photocatalytic technology to surfaces of large infrastructures. Only when these surfaces display high activity towards NOx (NO and/or NO2) removal, in particular, and degradation of VOC pollutants, in general, and no strong deactivation of the photoactive surface occurs under the prevailing speciﬁc ﬁeld conditions, should photocatalysis be advocated as a possible technology to reduce NOx and VOC levels in open urban environments. We have also seen that the photocatalytic technology has demonstrated signiﬁcant removal of NOx pollutants in a laboratory setting, whereas the overall results from real-scale application in an open air environment have been, to say the least, highly variable, but quite disappointing at best even under appropriate approaches in carrying out the experiments. 5. Applied Photocatalysis: Prospective Attempts at DeNOxing the Atmospheric Environment The CIEMAT group [171] also investigated the NOx depolluting effect of TiO2-based photocatalytic materials in a medium-scale tunnel reactor under semi-controlled conditions using 200 ppbv of NO and compared the results with those from a real-scale outdoor tunnel (street in Alcobendas) and from a laboratory-scale reactor—Table 14 summarizes the parameters and the extent of NO removed in all three cases [171]. 3. The CIEMAT group [171] also investigated the NOx depolluting effect of TiO2-based photocatalytic materials in a medium-scale tunnel reactor under semi-controlled conditions using 200 ppbv of NO and compared the results with those from a real-scale outdoor tunnel (street in Alcobendas) and from a laboratory-scale reactor—Table 14 summarizes the parameters and the extent of NO removed in all three cases [171]. Table 14. Parameters and results of NO removal in a medium-scale tunnel reactor (UVA irradiance, >40 W m−2; relative humidity, <30%; dimensions, 0.4 × 0.4 × 10 m; photoactive surface, 0.4 × 10 m) compared to a real-scale tunnel and a small laboratory ﬂow-through reactor. Table 14. Parameters and results of NO removal in a medium-scale tunnel reactor (UVA irradiance, >40 W m−2; relative humidity, <30%; dimensions, 0.4 × 0.4 × 10 m; photoactive surface, 0.4 × 10 m) compared to a real-scale tunnel and a small laboratory ﬂow-through reactor. Parameters Outdoor Real-Scale Tunnel Reactor Outdoor Medium-Scale Tunnel Reactor Laboratory Flow-through Reactor Deposition velocity (m s−1) 5.25 × 10−3 2.05 × 10−3 5.25 × 10−3 Length (m) 60 100 0.1 Air velocity (m s−1) 1 0.33 0.20 Photoactive surface (m2) 1000 4 0.005 Sactive/V (m−1) 0.067 2.5 200 Residence time (s−1) 60 30.3 0.5 First order rate coefﬁcient (s−1) 3.4 × 10−4 5.07 × 10−3 1.1 Average yield of NO removed (%) ~ 2 15 ± 4 41 495 Catalysts 2018, 8, 553 6. Concluding Remarks Such deviations from a laboratory setting to a real outdoor environmental setting is due to several factors that advantage the laboratory results: (1) a small ﬂow-through photoreactor was used in the laboratory tests against a vast open environment reactor system; (2) the photocatalytic surface used in the laboratory was a clean surface against an outdoor surface that in relatively short time is ultimately covered with dust, ﬁlth etc., thereby blocking the surface active sites where the reaction between the NOx and the oxidizing entities at these sites takes place—while rain might wash off the nitrates, the dust, and the ﬁlth from the open outdoor, that is not the case where the indoor walls of tunnels are concerned; (3) the rate of ﬂow of the NOx gases could be controlled in the laboratory tests, unlike the open environment where no control of wind speed and/or its direction are possible; (4) the UV/Visible radiance could be controlled in the laboratory versus lack of control of the sunlight impinging on the photocatalytic surfaces; (5) unlike the open environment, in laboratory tests, the ﬂow rate of the NOx gases in the photoreactor could be controlled which, therefore, affected the deposition velocity of the NOx onto the photocatalytic surface (i.e., a competition between horizontal ﬂow versus vertical ﬂow) as well as the residence time spent by the NOx gases within the reactor; (6) the Sactive/volume ratio in the laboratory reactor was far more advantageous than it could be in the open environment; (7) the relative humidity within the laboratory reactor could also be controlled, unlike the open environmental reactor; and (8) laboratory tests were carried out for a relatively short period (minutes to a few hours) within which the photocatalytic surface retained its integrity versus the open environment where the wished-for expectations were that the photocatalytic surfaces should remain active for months if not years and beyond. 496 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 Nonetheless, despite the disadvantages experienced in the open environment, there is no reason to throw away the baby with the bath water. It sufﬁces to carry out cost-efﬁciency analyses (as they may be geographically different) and compare them with other technologies available out there—for example, selective catalytic reduction (SCR) and selective non-catalytic reduction (SNCR) together with natural means (e.g., trees on road sides). 6. Concluding Remarks Photocatalysis may yet be found more attractive on a cost-efﬁciency basis compared to other, perhaps more expensive methods if extra costs of photocatalytic surfaces associated with application to normal urban surfaces (roads, paints, roof tiles, etc.) were minimized by the industrial sector in the future, and if photocatalytic materials were applied when urban infrastructures were being renovated [15]. It cannot be overemphasized that the effective use of photocatalysts in cementitious substrates in the deNOxiﬁcation of the environment requires serious considerations of some fundamental physical and chemical notions prior to any attempt at implementation of the TiO2-based photo-catalytic methodology, and as appropriately emphasized by Macphee and Folli [129], the following concerns are worth noting and sharing: 1. Photocatalysis is a surface phenomenon that is impacted by the chemistry of the immediate environment, so that the cementitious photocatalytic surface must be engineered to maximize light absorption by the photocatalytic surface and for the reactants to access the surface. 2. Redox potentials of pollutants considered for elimination must match the semiconductor band edge positions (i.e., ﬂatband potentials) of the conduction and valence bands for a successful deNOxiﬁcation as the band edges are pH-dependent, especially for a metal-oxide photo-catalyst. 3. More fundamental research is needed toward optimizing photonic efﬁciencies of visible-light activation of photocatalysts if anion- and/or cation-doped photocatalysts were used, as geographical locations may be limiting the usage of photocatalytic concrete if conventional pristine TiO2-based surfaces were used. 4. There is a need to maximize both the dispersion of photocatalytic particles in the cementitious substrate and the photoactive surface area toward target applications and agglomeration of particles in a porous structure should be minimized as it could block access of larger pollutant molecules to the internal photoactive surface. 5. Oxidative removal of NOx pollutants by their conversion to nitrates occurs only if they are adsorbed on the photocatalytic surface; recall that the adsorption mode of molecules on the photoactive surface is highly pH-dependent (surface charge) and could thus affect the oxidative efﬁciency. 6. 7. Recommendations In practical applications of photocatalysis, conventional TiO2-based photocatalytic surfaces have been used to oxidize NOx to nitrate species; the latter species do not desorb spontaneously and consequently deactivate or block the surface-active centers of the photocatalyst from carrying out the next cycles. To avoid such deactivation, the nitrates (or nitric acid) should typically be washed away by rain [172]; however, the nitric acid is corrosive and could pollute the soil when its concentration at the site becomes too high. A promising way to resolve this problem, which the users of applied photocatalysis have failed to consider but known to occur in a laboratory setting since the ﬁrst report by Courbon et al. [92] in 1984, would be to change the selectivity of the photocatalytic reaction so that the NOx gases are converted back to N2 and O2 by some photoreduction pathway as reported some time ago by Anpo and coworkers [94,95], who used Cu+ ions in SiO2 or in zeolite to effect the photocatalytic reduction of NOx. No deactivation of the active sites would occur for this photoreduction reaction since nitrogen and oxygen readily desorb from the surface [173]. The selectivity toward the photoreduction of NO could be improved greatly by reducing the hexacoordinated Ti4+ species (TiO6 octahedra) to tetracoordinated Ti4+ species (TiO4 tetrahedra) [174], as successfully achieved by depositing isolated TiO4 clusters inside cavities of zeolite-Y using ion beam implantation [174,175]. In a more recent article, Wu and van de Krol [176] proposed a novel strategy to change the photocatalytic selectivity of TiO2 by creating a large and stable concentration of oxygen vacancies in TiO2 nanoparticles through thermal reduction in a reducing atmosphere; these oxygen vacancies were stabilized by doping the TiO2 nanoparticles with an electron acceptor-type dopant such as Fe3+ which also greatly enhanced the activity of the photoreduction process. The authors [176] further showed that with this strategy NO was indeed photoreduced to N2 and O2 and that photooxidation of NO was largely suppressed. Moreover, photoreducing Fe3+ to Fe2+ provided a recombination pathway that suppressed nearly quantitatively the formation of NO2 and consequently enhanced the selectivity of the reaction for N2 formation [176]. The authors also alluded to formation of N2 and O2 via two different routes. One route would see a small amount of tetrahedrally coordinated Ti formed in the Fe-doped TiO2 samples, which Anpo et al. 6. Concluding Remarks The deNOx index that combines photonic efﬁciency and selectivity into one environmental impact parameter for a given photocatalytic material should be considered whenever screening candidate photocatalytic materials; for instance, if this index for a photoactive surface were negative shows that the oxidative removal of NO favors formation of the more toxic product NO2 rather than the NO3−species, as demonstrated for a conventional unselective TiO2 (e.g., bare P-25 titania) that exhibited a large negative deNOx index [129], a sign that the catalytic process was ineffective. Consequently, both photocatalyst selectivity and photo-catalyst activity must be assessed to reduce emission of harmful by-products. 7. On the more fundamental side, the thermodynamics (energetics) are the most valuable tool in understanding and designing selective photocatalytic processes. Recall that, after the rapid e−/h+ pair recombination, the next fastest process is charge carrier trapping. This calls attentions that not only is the semiconductor band edge positions relevant, but also charge transfer from traps to adsorbate molecules need to be fully appreciated in interpreting photocatalytic processes as also the engineering and design of band structures of single semiconductors or semiconductor composites. 7. On the more fundamental side, the thermodynamics (energetics) are the most valuable tool in understanding and designing selective photocatalytic processes. Recall that, after the rapid e−/h+ pair recombination, the next fastest process is charge carrier trapping. This calls attentions that not only is the semiconductor band edge positions relevant, but also charge transfer from traps to adsorbate molecules need to be fully appreciated in interpreting photocatalytic processes as also the engineering and design of band structures of single semiconductors or semiconductor composites. Additionally, using a tunnel as an example of an infrastructure in need of eliminating NOx agents, Gallus et al. [14] also expressed the notion that a reasonable photocatalytic remediation of NOx in a 497 Catalysts 2018, 8, 553 road tunnel can only be obtained under suitable/special experimental conditions, as photocatalytic remediation will not be effective where the photoactive materials are strongly deactivated under highly polluted conditions. Accordingly, photocatalytic materials should always be tested in small-scale bed-ﬂow reactors on site using: (1) the UVA irradiance equivalent to that of the tunnel; and (2) the NOx present in ambient air sucked from the tunnel as the reactant. Only if NOx were efﬁciently decomposed under these conditions, would larger scale applications in a road tunnel be worth considering. 6. Concluding Remarks Regardless of the possible deactivation of photo-catalytic materials under high pollution conditions, model calculations have shown that applications of such materials are unlikely to result in signiﬁcant remediation of NOx to nitrates at low UVA irradiances and where high wind speeds and/or high relative humidity prevail. With regard to NOx conversions in road tunnels, this calls further attention to the need for a careful characterization of tunnel conditions, for a quantiﬁcation of possible deactivation of the photocatalytic materials, and for performing simple upper limit model calculations of expected NOx reductions, together with a cost–beneﬁt analysis well before considering application of photocatalytic materials and installation of costly UV lighting systems. These same considerations also apply to pollutant remediation in other infrastructures. 7. Recommendations [175,177] claimed as the active site for the catalytic decomposition of NO to N2 and O2 at Ti-modiﬁed zeolites. As most TiIV ions at the TiO2 surface are ﬁvefold-coordinated, a single oxygen vacancy created at or near the surface could lead to a fourfold-coordinated Ti4+ center; however, this would require a strong reduction of the Ti–O bond length that would be possible only at very high oxygen vacancy concentrations, which the authors [177] deemed an unlikely pathway and proposed the other route that implicated oxygen vacancies acting 498 Catalysts 2018, 8, 553 as the catalytic centers through the capture of the oxygen side of NO as illustrated in Figure 59 and summarized in Reactions (101)–(104); the associated experimental data are also displayed in Figure 59. y g p yg g marized in Reactions (101)–(104); the associated experimental data are also displayed in Figure 59. Vo++ (surf) + 2 e−+ NO(g) →Osurf −N (101) 2 Osurf −N →2 Osurf + N2(g) (102) 2 Osurf →2 Vox + O2(g) (103) Vox + 2h+ →Vo++ (104) (a) (b) Figure 59. (a) Cartoon illustrating the possible pathway to reduce NO to N2 and O2 gases through the involvement of oxygen vacancies; and (b) photocatalytic conversion of NO to N2 and O2 over 1% Fe-doped TiO2 under irradiation with UV light; concentration of NO: 100 ppm in He. Reproduced with permission from Ref. [176]. Copyright 2012 by the American Chemical Society. Vo++ (surf) + 2 e−+ NO(g) →Osurf −N (101) 2 Osurf −N →2 Osurf + N2(g) (102) 2 Osurf →2 Vox + O2(g) (103) Vox + 2h+ →Vo++ (104) (b) (a) (b) (b) (a) Figure 59. (a) Cartoon illustrating the possible pathway to reduce NO to N2 and O2 gases through the involvement of oxygen vacancies; and (b) photocatalytic conversion of NO to N2 and O2 over 1% Fe-doped TiO2 under irradiation with UV light; concentration of NO: 100 ppm in He. Reproduced with permission from Ref. [176]. Copyright 2012 by the American Chemical Society. Figure 59. (a) Cartoon illustrating the possible pathway to reduce NO to N2 and O2 gases through the involvement of oxygen vacancies; and (b) photocatalytic conversion of NO to N2 and O2 over 1% Fe-doped TiO2 under irradiation with UV light; concentration of NO: 100 ppm in He. Reproduced with permission from Ref. [176]. Copyright 2012 by the American Chemical Society. Summing Reactions (101)–(104) yields the overall Reaction (105). 7. Recommendations 2 Vo++ 2 NO(g) + 4 hν →N2(g) + O2(g) (105) (105) Although the conversion efﬁciency was somewhat modest (ca. 4.5% after 1050 min correspond-ing to a TON of ∼2 NO molecules per O vacancy site), the Fe-doped TiO2 photocatalyst showed no signs of deactivation as the NO conversion centers were not blocked by nitrate species [168], contrary to standard deNOx TiO2-based photocatalysts that have to be washed away periodically. In a most recent article, Cao and coworkers [178] investigated the adsorption of NO and the consequent reactions on differently treated rutile TiO2(110) surfaces using polarization/azimuth-resolved infrared reﬂection absorption spectroscopy. Apparently, surface defects (e.g., oxygen vacancies, Vo) and reconstructions on TiO2(110) had a strong effect on the reaction pathways of NO →N2O conversion (N2O is laughing gas). The pathway proposed involved a defect-free oxidized TiO2(110) surface in which two NO molecules are adsorbed on adjacent surface-pentacoordinated Ti (Ti5c) sites ﬁrst, which then couple to form a cis-(NO)2/Ti&Ti dimer through the N−N bond of the dimer, and then are converted to N2O species (or perhaps even to N2 gas) [178]. Clearly, much fundamental research in TiO2-based photocatalysis needs to be undertaken in the optic toward applications to environmental deNOxiﬁcation, with special attention and efforts directed at titania doped with Fe, Cr, Co and Ni dopants that may yet prove interesting [176]. As a case in point, a recent article by Kuznetsov and coworkers [70] examined possible additional speciﬁc channels of photoactivation of solid semiconductors with regard to thermo-/photo-stimulated bleaching of photoinduced Ti3+ color centers in visible-light-active (VLA) photo-chromic rutile TiO2, which an optical emission spectroscopic analysis had shown to contain 99.4 at.% Ti and 0.2 at.% Al 499 Catalysts 2018, 8, 553 as the principal impurity, together with 0.09 at.% Fe, 0.05 at.% Sn, 0.04 at.% Nb, and 0.03 at.% Cr as minor impurities. Considering that the prime photophysical process of photostimulated bleaching of Ti3+ color centers is absorption of light quanta by the Ti3+ centers, the authors [70] found that no selectivity of photostimulated bleaching of a certain type of Ti3+ centers could be ascertained, and that photogenerated holes captured at a set of traps were also participants in the photostimulated bleaching of these color centers. 7. Recommendations Based on current ﬁndings and earlier results, the authors hypothesized that the heat released during nonradiative electron transitions, following the prime photophysical processes of excitation and ionization of Ti3+ centers, dissipates in the nearest neighborhood of the Ti3+ centers and that localized nonequilibrated excitation of the phonon subsystem leads to thermal detrapping of the photoholes with different depths up to 1 eV. Subsequent recombination of free holes with trapped electrons from Ti3+ centers leads to the observable photostimulated bleaching of the color centers [70]. Based on experimental evidence, the authors further argued that following absorption of vis–NIR light by the color centers, the subsequent release of thermal energy accompanying nonradiative electron transitions provides an additional speciﬁc channel to photoactivate the VLA rutile TiO2, in particular, and possibly other photocolorable metal-oxide semiconductors as well. Following their interest of the photophysics of color centers in VLA rutile titania ceramics and titania powder resulting from the photoformation and separation of charge carriers, Kuznetsov et al. [179] noted that the action spectrum of the photoformation of Ti3+ centers at very low temperatures (90 K) accorded fully with the absorption spectra of intrinsic defects that consisted of a set of individual absorption bands that they attributed to several different Ti3+ centers. Analysis of the dependencies of the photoformation of separate centers on the wavelength of illumination and light exposure, which provided extraction of speciﬁc Ti3+ centers, led the authors to identify Ti3+-based centers with excessive negative charge that formed at signiﬁcantly high concentration upon maximal exposure of the titania specimens to Vis-light illumination: (2Ti3+ + Vo2+) ←→(Tiδ+ + Vo2+) with 3 > δ > 2. They also showed from thermoprogrammed annealing (TPA) that the spectra of Ti3+ color centers in the range 90–500 K consisted of a set of ﬁrst-order peaks corresponding to traps, whose depths ranged from ~ 0.2 eV (peak at 130 K in the powder specimen) to 1.06 eV (peak at 455 K in the ceramics specimen). The highest rate of recombination of holes released to the valence band with Ti3+ centers—an event attributed to Tiδ+ centers—provided TPA spectra that clearly manifested the existence of shallow traps. 7. Recommendations In addition, mass spectrometric experiments on the photoadsorption of molecular oxygen and photodesorption of photoadsorbed oxygen from the surface of powdered VLA titania specimens provided further evidence of the photoformation of electrons and holes in VLA TiO2 under Vis-light illumination, and allowed the authors [179] to determine the kinetics of photo- desorption of O2 under orange light illumination subsequent to photoadsorption of O2 stimulated by blue light excitation. Those experiments provided further proof of the occurrence of another speciﬁc channel toward the photoactivation of VLA TiO2 via photoexcitation of photoinduced Ti3+ color centers. It is important to recognize that Ti3+-based centers (i.e., Tiδ+ centers) appeared after many other Ti3+ centers had already been formed. In other words, such Ti3+-based centers appeared at high density of Ti3+ centers (see below). Accordingly, speciﬁc properties of Ti3+-related centers responsible for the (extrinsic) absorption bands at 1.56 eV and 1.26 eV were postulated to account for the excess negative charge characteristic of such Ti3+-based centers. In line with the work of Déak and coworkers [180], the two adjacent Ti3+ centers located near a single oxygen vacancy forming a (2Ti3+ + Vo2+) complex (Figure 60a) were taken by Kuznetsov et al. [179] as extra charged Ti3+ centers when compared to isolated Ti3+ centers (Figure 60b). The two Ti3+ species in the (2Ti3+ + Vo2+) complex can, in principle, disproportionate to (Ti2+ + Ti4+ + Vo2+) so that, in accord with the more generally accepted view, these extra-negatively charged Ti3+ centers are best referred to as Tiδ+ centers for which 3 > δ > 2. Clearly, the appearance of such Tiδ+ centers, whether photogenerated or resulting from the removal of the structural oxygen during a reduction event, is of lower probability because of the electrostatic repulsion of the two trapped electrons and the well-known instability of such centers to oxidation. Consequently, 500 Catalysts 2018, 8, 553 the formation and increase in the concentration of such photoinduced Tiδ+ centers appears to occur only at high density of photogenerated Ti3+ centers that ensue upon prolonged exposure to Vis-light illumination in the later stages of photocoloration (i.e., formation of Ti3+ color centers). Figure 60. Cartoons representing the meaning of: (a) extra-negatively charged Ti3+ as being the complex formed between two adjacent Ti3+ species neighboring an oxygen vacancy (Vo) that can be viewed as a Tiδ+ center with 3 > δ > 2; and (b) isolated Ti3+ centers. 7. Recommendations Reproduced with permission from Kuznetsov et al. [70]. Copyright 2018 by Elsevier, B.V. (License No.: 4453641492289). Figure 60. Cartoons representing the meaning of: (a) extra-negatively charged Ti3+ as being the complex formed between two adjacent Ti3+ species neighboring an oxygen vacancy (Vo) that can be viewed as a Tiδ+ center with 3 > δ > 2; and (b) isolated Ti3+ centers. Reproduced with permission from Kuznetsov et al. [70]. Copyright 2018 by Elsevier, B.V. (License No.: 4453641492289). Figure 60. Cartoons representing the meaning of: (a) extra-negatively charged Ti3+ as being the complex formed between two adjacent Ti3+ species neighboring an oxygen vacancy (Vo) that can be viewed as a Tiδ+ center with 3 > δ > 2; and (b) isolated Ti3+ centers. Reproduced with permission from Kuznetsov et al. [70]. Copyright 2018 by Elsevier, B.V. (License No.: 4453641492289). Germane to the above, the work of Déak and coworkers [180] showed that the ﬁrst case scenario is that two self-trapped electrons in the (Vo2+ + 2e) complexes are located at two equivalent ﬁrst neighbors of the oxygen vacancy (extra-negatively charged Ti3+ →Tiδ+ centers; Figure 60a), while in the second scenario both electrons are more remote from the Vo2+ vacancy and are not in the same plane as the vacancy (isolated Ti3+ centers; Figure 60b); the energies of the vertical transitions of these self-trapped electrons to the conduction band are ca. 1.1 eV. Following this reasoning, a question arose as to why the growth of the number of such Tiδ+ centers was observed only under Vis-light illumination. This led Kuznetsov and coworkers [70] to focus attention on the differences in the spatial photoexcitation events that occur in the microparticle when illuminated in the UV and Vis spectral regions. Such differences had not heretofore been considered in the literature; their views of the events that occur under UV and Visible light illumination are summarized in Figure 61. 501 Catalysts 2018, 8, 553 Figure 61. Illustration of the absorption of light quanta and formation of: (1) electron–hole pairs; (2) electron; and (3) hole transport and their localization at traps; together with (4) electron–hole recombination under: (a) UV irradiation; and (b) Vis-light irradiation of a microcrystalline particle of VLA TiO2. 7. Recommendations In (a), the black stars denote the photogenerated electron–hole pairs, the blue stars the F-type centers (electron trapped in oxygen vacancy); the green circles the Ti3+ color centers, while the red circles refer to the trapped holes. In (b), the blue stars denote the F-type centers (electron trapped in oxygen vacancy), the green circles the Ti3+ color centers, while the red circles denote the trapped holes. Note the Tiδ+ centers in (b). Reproduced with permission from Kuznetsov et al. [70]. Copyright 2018 by Elsevier, B.V. (License No.: 4453641492289). Figure 61. Illustration of the absorption of light quanta and formation of: (1) electron–hole pairs; (2) electron; and (3) hole transport and their localization at traps; together with (4) electron–hole recombination under: (a) UV irradiation; and (b) Vis-light irradiation of a microcrystalline particle of VLA TiO2. In (a), the black stars denote the photogenerated electron–hole pairs, the blue stars the F-type centers (electron trapped in oxygen vacancy); the green circles the Ti3+ color centers, while the red circles refer to the trapped holes. In (b), the blue stars denote the F-type centers (electron trapped in oxygen vacancy), the green circles the Ti3+ color centers, while the red circles denote the trapped holes. Note the Tiδ+ centers in (b). Reproduced with permission from Kuznetsov et al. [70]. Copyright 2018 by Elsevier, B.V. (License No.: 4453641492289). When hν > Eg, light quanta are absorbed spontaneously in solids in an arbitrary manner, each time producing e-h pairs at new spatial sites (black stars in Figure 61a) for which charge carrier transport and localization in the microparticle are determined by the distribution of charge carrier traps (Processes 2 and 3 in Figure 61a). At moderate UV-light irradiances, the authors [70] supposed that since every subsequent photoformation of charge carriers and trapping event occur in other spatial sites, a signiﬁcant density of Ti3+ centers would not be reached. However, photoformation of electrons and holes can also be achieved on illumination in the Vis region at hν ≤Eg when light quanta are absorbed by the native point defects (i.e., F (or F+) centers); the latter are limited in number and are located at deﬁnite sites in the microparticle (Figure 61b). 7. Recommendations Moreover, because photoexcitation of F (or F+) centers can produce charge carriers followed by their subsequent decay to their initial electronic states, as proposed in earlier studies [68,69], repetitive absorption of light quanta and photogeneration of electrons and holes occurs each time at the same spatial sites (F or F+ centers) in the microparticle (blue stars in Figure 61b). Such considerations then lead to the reasonable inference that transport of carriers and occupation of traps (processes 2 and 3 in Figure 61b) start repetitively at the same sites in the microparticle. In that case, ﬁlling of the nearest neighbor F (or F+) center traps facilitates the attainment of a high density of Ti3+ centers and the consequent formation of the Tiδ+ centers [70]. The above notwithstanding regarding the TiO2-based technology, people intending to deNOxify the environment must ﬁrst come to appreciate and understand the rich chemistry of nitrogen oxides, in general, and NO and NO2, in particular, in a homogeneous phase and in hetero- geneous media. For instance, NO dimerizes to N2O2 upon condensing to a liquid, although the association is weak and reversible [181]. In addition, to the extent that the enthalpy of formation of NO is endothermic, NO can easily undergo disproportionation back to its constituent elements N2 and O2 as might occur in catalytic converters—for example, Reaction (106) occurs over the zeolite Cu2+-ZSM-5 [182]. 2 NO →N2 + O2 (106) (106) 2 NO →N2 + O2 Nitric oxide is also thermodynamically unstable at 25 ◦C and 1 atm; under pressure, it decomposes readily in the temperature range 30–50 ◦C to yield NO2 and N2O (Reaction (107)) and may react either as NO2 or as N2O3 [183]. 3 NO →N2O + NO2 (107) (107) 3 NO →N2O + NO2 502 Catalysts 2018, 8, 553 Catalysts 2018, 8, 553 When exposed to atmospheric oxygen, nitric oxide converts instantly to NO2 (Reaction (108)) [181], which likely occurs via the intermediate ON–O–O–NO. 2 NO + O2 →2 NO2 (108) 2 NO + O2 →2 NO2 (108) In water, NO reacts with oxygen and water to form nitrous acid HONO (Reaction (109)). 4 NO + O2 + 2 H2O →4 HONO (109) 4 NO + O2 + 2 H2O →4 HONO (109) Since both NO2 and NO are radical species, they combine to form the intensely blue dinitrogen trioxide N2O3 (Reaction (110)) [184]. 7. Recommendations NO + NO2 ⇄ON −NO2 (110) (110) Both the brown gas nitrogen dioxide, NO2, and the colorless gas dinitrogen tetroxide, N2O4, exist in a strongly temperature-dependent equilibrium (Reaction (111)) for which ΔH = −57.23 kJ mol−1, with NO2 being favored at higher temperatures, while N2O4 predominates at lower temperatures. 2 NO2 ⇄N2O4 (111) (111) Because of the relatively weak N–O bond in NO2, nitrogen dioxide is a relatively good oxidizing agent in aqueous media (Reaction (112); nearly comparable to Br2 gas), which makes the mixed oxides NO2 and N2O4—also known as nitrous fumes—react vigorously if not explosively with several compounds, particularly with hydrocarbons via hydrogen abstraction as a ﬁrst step (Reaction (113)) [181]. N2O4 + 2 H+ + 2 e−→2 HONO E0 = 1.07 V (112) NO2 + RH →R• + HONO (113) (113) In aqueous media, NO2 also hydrolyzes to form nitrous acid and nitric acid via Reaction (114), which is one of the steps in the industrial production of nitric acid from ammonia via the Ostwald process [185]. H2O By contrast, not only did NO2 signiﬁcantly inﬂuence the adsorption of VOCs and mineralization on the TiO2 surface, but once the threshold surface coverage of NO2 was reached at room temperature, the NO2 adsorbed reactively on the TiO2 surface by evolving NO in the gas phase. Quantitative measurements performed downstream of the reactor led the authors [187] to propose a new mechanism expressed by the Reactions (115)–(118) for the adsorption of NO2 on TiO2 at room temperature under dry air conditions: 2 NO2(ads) ⇄NO2−+ NO+ (ads) (115) NO + O2−(TiO2 Surface latice) ⇄NO2− (ads) (116) NO2− (ads) + NO2(ads) ⇄NO3− (ads) + NO(g) (117) th l b l ti f NO d ti i th (R ti (118)) 2 NO2(ads) ⇄NO2−+ NO+ (ads) (115) NO + O2−(TiO2 Surface latice) ⇄NO2− (ads) (116) NO2− (ads) + NO2(ads) ⇄NO3− (ads) + NO(g) (117) h l b l i f NO d i i h (R i (118)) 2 NO2(ads) ⇄NO2−+ NO+ (ads) (115) NO + O2−(TiO2 Surface latice) ⇄NO2− (ads) (116) NO2− (ads) + NO2(ads) ⇄NO3− (ads) + NO(g) (117) According to this sequence, the global reaction of NO2 adsorption is then (Reaction (118)): 3 NO2(ads) ⇄2 NO3− (ads) + NO(g) (118) 3 NO2(ads) ⇄2 NO3− (ads) + NO(g) Accordingly, the proposed NO2 adsorption mechanism on TiO2 at room temperature in the dark [187], together with the experimental observations, could be summarized thus: (i) three NO2 molecules adsorb on TiO2, produce two NO3−ions on the TiO2 surface, and evolve one NO molecule in the gas phase; (ii) the ratio between consumed NO2, TPD desorbed NO2 subsequent to adsorption, and NO produced during NO2 adsorption = 3:2:1; (iii) the NO2 adsorption time (i.e., the TiO2 surface coverage) signiﬁcantly modiﬁed the nature of the adsorbed species at ambient temperature; (iv) the NO formation time was controlled principally by the surface coverage of NO2−and NO3−ions, rather than by the NO2 inlet concentration; and (v) at higher NO2 gas phase concentrations (greater than 35 ppm) at room temperature, the total amount of consumed NO2 decreased as a result of self-poisoning of the sites by adsorbed NO3−species. H2O H2O H2O (114) H2O 2 NO2 ⇄N2O4 →HONO + HNO3 (114) (114) 2 NO2 ⇄N2O4 →HONO + HNO3 (114) Although Reaction (114) is negligibly slow at the low concentrations of NO2 characteristically encountered in the ambient atmosphere, it does proceed upon uptake of NO2 onto surfaces to produce gaseous HONO in outdoor and indoor environments [186]. Several studies have examined the interactions of NO2 on the TiO2 surface under various experimental conditions, such as different NO2 partial pressures and various temperatures in the range 323–573 K [102,106,108,172]. All these studies reported production of NO in the gas phase, and formation of nitrates on the TiO2 surface, albeit under photocatalytic conditions. However, other aspects that seem to have been overlooked by many are the potential speciﬁc interactions between the two NOx molecules and the TiO2 surface under dark conditions, which need to be re-emphasized constantly. In this regard, in their 2003 FTIR study carried out in the dark in borosilicate glass vessels, Finlayson-Pitts et al. [186] discovered that the loss of gaseous NO2 was accompanied by formation of HONO, NO and N2O; further FTIR studies also revealed the formation of HNO3, N2O4 and NO2+ species, which led them to hypothesize that the symmetric form of the NO2 dimer, N2O4, is taken up on the surface and isomerizes to the asymmetric form, ONONO2, with the latter undergoing autoionization to NO+NO3−. Apparently, it is the latter intermediate species that react with water to generate HONO and surface-adsorbed HNO3. Subsequently, NO is generated by secondary reactions of HONO on the highly acidic surface. The authors further noted that a key aspect of this chemistry 503 Catalysts 2018, 8, 553 is that in the atmospheric boundary layer where human exposure occurs and many measurements of HONO and related atmospheric constituents (e.g., ozone) are made, a major component for this heterogeneous chemistry is the surface of buildings, roads, soils, vegetation and other materials [186]. A more recent investigation by Sivachandrian and coworkers [187] on the adsorption of NO and NO2 molecules on the metal oxide TiO2 at ambient temperature, speciﬁcally carried out under dark experimental conditions to avoid any photocatalytic interference, showed no signiﬁcant adsorption of NO on TiO2. Funding: This research received no external funding. Funding: This research received no external funding. Acknowledgments: We are grateful to Angelo Albini for his gracious hospitality in the PhotoGreen Laboratory at the University of Pavia, Italy. Acknowledgments: We are grateful to Angelo Albini for his gracious hospitality in the PhotoGreen Laboratory at the University of Pavia, Italy. Conﬂicts of Interest: The author declares no conﬂict of interest. Conﬂicts of Interest: The author declares no conﬂict of interest. References 1. Engineering Alliance, Inc. Air Quality Services, Types_of_sources_02-2012. Available online: https://www. eaincglobal.com/air-quality/attachment/types_of_sources_02-2012/ (accessed on 12 October 2018). 2. Smog. Available online: https://en.wikipedia.org/wiki/Smog (accessed on 12 October 201 3. Popescu, F.; Ionel, I. Anthropogenic Air Pollution Sources. In Air Quality; Kumar, A., Ed.; InTech Europe: Rijeka, Croatia, August 2010; Chapter 1; pp. 1–22. ISBN 978-953-307-131-2. Available online: http://www. intechopen.com/books/airquality/anthropogenic-air-pollution-sources (accessed on 15 July 2018). 3. Popescu, F.; Ionel, I. Anthropogenic Air Pollution Sources. In Air Quality; Kumar, A., Ed.; InTech Europe: Rijeka, Croatia, August 2010; Chapter 1; pp. 1–22. ISBN 978-953-307-131-2. Available online: http://www. h /b k / l / h ll ( d l ) 4. European Union Emission Inventory Report 1990–2011 under the UNECE Convention on Long-Range Trans-Boundary Air Pollution (LRTAP). Available online: http://www.icopal-noxite.co.uk/nox-problem/ nox-pollution.aspx (accessed on 12 October 2018). 4. European Union Emission Inventory Report 1990–2011 under the UNECE Convention on Long-Range Trans-Boundary Air Pollution (LRTAP). Available online: http://www.icopal-noxite.co.uk/nox-problem/ nox-pollution.aspx (accessed on 12 October 2018). 5. United States Environmental Protection Agency. Air Emission Sources. 4 November 2009. Available onl http://www.epa.gov/air/emissions/index.htm (accessed on 12 October 2018). 6. Tropospheric Ozone. Available online: https://en.wikipedia.org/wiki/Tropospheric_ozone#cite_note-5 (accessed on 12 October 2018). 6. Tropospheric Ozone. Available online: https://en.wikipedia.org/wiki/Tropospheric_ozone#cite_note-5 (accessed on 12 October 2018). 7. Reeves, C.E.; Penkett, S.A.; Bauguitte, S.; Law, K.S.; Evans, M.J.; Bandy, B.J.; Monks, P.S.; Edwards, G.D.; Phillips, G.; Barjat, H.; et al. Potential for photochemical ozone formation in the troposphere over the North 7. Reeves, C.E.; Penkett, S.A.; Bauguitte, S.; Law, K.S.; Evans, M.J.; Bandy, B.J.; Monks, P.S.; Edwards, G.D.; Phillips, G.; Barjat, H.; et al. Potential for photochemical ozone formation in the troposphere over the North Atlantic as derived from aircraft observations during ACSOE. J. Geophys. Res. 2002, 107, 4707. [CrossRef] Phillips, G.; Barjat, H.; et al. Potential for photochemical ozone formation in the troposphere over the North Atlantic as derived from aircraft observations during ACSOE. J. Geophys. Res. 2002, 107, 4707. [CrossRef] 8 Deziel C How Is Photochemical Smog Formed? Available online: https://sciencing com/photochemical- Atlantic as derived from aircraft observations during ACSOE. J. Geophys. Res. 2002, 107, 4707. [CrossRe 8. Deziel, C. How Is Photochemical Smog Formed? Available online: https://sciencing.com/photochemical- smog-formed-6505511.html (accessed on 25 April 2017). 9. Health Effects of Nitrogen Oxides, Department of Employment, Economic Development and Innovation, Queensland Government, Australia. Available online: https://www.dnrm.qld.gov.au/data/assets/pdf_ ﬁle/0020/212483/2-health-effects-of-nitrogen-dioxide.pdf (accessed on 12 October 2018). ﬁle/0020/212483/2-health-effects-of-nitrogen-dioxide.pdf (accessed on 12 October 2018). 10. McCarron, G. H2O Similarly, Haubrich and coworkers [188] examined the interaction of NO2 on rutile TiO2(110) and discovered that the presence of NO2 and water led to the formation of multilayers under dark conditions of nitric acid, HNO3, contrary to exposure of the surface to pure water after saturation of the surface with 200 mTorr of NO2; no further growth of the AP-XPS (ambient pressure X-ray photoelectron spectroscopy) nitrate signals occurred under the latter conditions. Apparently, formation of HNO3 requires weakly adsorbed NO2 molecules, an important ﬁnding with important implications in environmental processes since their study [188] conﬁrmed that metal oxides facilitate the formation of nitric acid under ambient humidity (in the dark) conditions typically encountered in atmospheric environments. It is evident that there is much more that needs to be investigated to fully understand whatever events occur in and on the TiO2 (and others) semiconductor photocatalyst. Laboratory experiments using models and solar simulators are just the beginning, after which what is needed is to bring the laboratory outdoors using actual environmental quantities of NOx as the reagents and humid air as prevails in the environment being investigated for application of the photocatalytic substrates. A careful study of the levels of NOx or its products at various distances from the photocatalytic surface, both vertically and horizontally, without precluding determination of all other environmental factors is also needed. The deNOx index has shown that metal-doped TiO2 systems are also worth investigating further. 504 Catalysts 2018, 8, 553 References Formation of indoor nitrous acid (HONO) by light-induced NO2 heterogeneous reactions with white wall paint. Environ. Sci. Pollut. Res. 2014, 21, 9259–9269. [CrossRef] [PubMed] 22. Bartolomei, V.; Sörgel, M.; Gligorovski, S.; Alvarez, E.G.; Gandolfo, A.; Strekowski, R.; Quivet, E.; Held, A.; Zetzsch, C.; Wortham, H. Formation of indoor nitrous acid (HONO) by light-induced NO2 heterogeneous reactions with white wall paint Environ Sci Pollut Res 2014 21 9259–9269 [CrossRef] [PubMed] 23. Langridge, J.M.; Gustafsson, R.J.; Grifﬁths, P.T.; Cox, R.A.; Lambert, R.M.; Jones, R.L. Solar driven nitrous acid formation on building material surfaces containing titanium dioxide: A concern for air quality in urban areas? Atmos. Environ. 2009, 43, 5128–5131. [CrossRef] 24. Ndour, M.; Conchon, P.; D’Anna, B.; Ka, O.; George, C. Photochemistry of mineral dust surface as a potential atmospheric renoxiﬁcation process. Geophys. Res. Lett. 2009, 36, 1–4. [CrossRef] 25. Mothes, F.; Böge, O.; Herrmann, H. A chamber study on the reactions of O3, NO, NO2 and selected VOCs with a photocatalytically active cementitious coating material. Environ. Sci. Pollut. Res. 2016, 23, 15250–15261. [CrossRef] [PubMed] 26. Monge, M.E.; George, C.; D’Anna, B.; Doussin, J.-F.; Jammoul, A.; Wang, J.; Eyglunent, G.; Solignac, G.; Daële, V.; Mellouki, A. Ozone formation from illuminated titanium dioxide surfaces. J. Am. Chem. Soc. 2010, 132, 8234–8235. [CrossRef] [PubMed] 27. Emeline, A.V.; Kuznetsov, V.N.; Ryabchuk, V.K.; Serpone, N. On the way to the creation of next generation photoactive materials. Environ. Sci. Pollut. Res. 2012, 19, 3666–3675. [CrossRef] [PubMed] 28. Serpone, N.; Emeline, A.V. Semiconductor photocatalysis—Past, present, and future outlook. J. Phys. Chem. Lett. 2012, 3, 673–677. [CrossRef] [PubMed] 29. Braslavsky, S.E.; Braun, A.M.; Cassano, A.E.; Emeline, A.V.; Litter, M.I.; Palmisano, L.; Parmon, V.N.; Serpone, N. Glossary of terms used in photocatalysis and radiation catalysis (IUPAC Recommendations 2011). Pure Appl. Chem. 2011, 83, 931–1014. [CrossRef] 30. Emeline, A.V.; Panasuk, A.V.; Sheremetyeva, N.; Serpone, N. Mechanistic studies of the formation of different states of oxygen on irradiated ZrO2 and the photocatalytic nature of photoprocesses from determination of turnover numbers. J. Phys. Chem. B 2005, 109, 2785–2792. [CrossRef] [PubMed] 31. Emeline, A.V.; Rudakova, A.V.; Ryabchuk, V.K.; Serpone, N. Photostimulated reactions at the surface of wide bandgap metal oxides {ZrO2 and TiO2}: Interdependence of rates of reactions on pressure- concentration and on light intensity. J. Phys. Chem. B 1998, 102, 10906–10916. [CrossRef] 32. Emeline, A.V.; Ryabchuk, V.K.; Serpone, N. Factors affecting the efﬁciency of a photocatalyzed process in aqueous metal-oxide dispersions. References Air Pollution and human health hazards: A compilation of air toxins acknowledged by the gas industry in Queensland’s Darling Downs. Int. J. Environ. Stud. 2018, 75, 171–185. [CrossRef] 11. Chen, J.; Poon, C.-S. Photocatalytic construction and building materials: From fundamentals to applications. Build. Environ. 2009, 44, 1899–1906. [CrossRef] 12. Mendoza, C.; Valle, A.; Castellote, M.; Bahamonde, A.; Faraldos, M. TiO2 and TiO2–SiO2 coated cement: Comparison of mechanic and photocatalytic properties. Appl. Catal. B 2015, 178, 155–164. [CrossRef] 13. Pei, C.C.; Leung, W.W.F. Photocatalytic oxidation of nitrogen monoxide and o-xylene by TiO2/ZnO/Bi2O3 nanoﬁbers: Optimization, kinetic modeling and mechanisms. Appl. Catal. B 2015, 174–175, 515–525. [CrossRef] 14. Gallus, M.; Akylas, V.; Barmpas, F.; Beeldens, A.; Boonen, E.; Boréave, A.; Cazaunau, M.; Chen, H.; Daële, V.; Doussin, J.F.; et al. Photocatalytic depollution in the Leopold II tunnel in Brussels: NOx abatement results. Build. Environ. 2015, 84, 125–133. [CrossRef] 15. Gallus, M.; Ciuraru, R.; Mothes, F.; Akylas, V.; Barmpas, F.; Beeldens, A.; Bernard, F.; Boonen, E.; Boréave, A.; Cazaunau, M.; et al. Photocatalytic abatement results from a model street canyon. Environ. Sci. Pollut. Res. 2015, 22, 18185–18196. [CrossRef] [PubMed] 16. Sikkema, J.K.; Ong, S.K.; Alleman, J.E. Photocatalytic concrete pavements: Laboratory investigation of NO oxidation rate under varied environmental conditions. Constr. Build. Mater. 2015, 100, 305–314. [CrossRef] 17. Gandolfo, A.; Bartolomei, V.; Gomez-Alvarez, E.; Tlili, S.; Gligorovski, S.; Kleffmann, J.; Wortham, H. The effectiveness of indoor photocatalytic paints on NOx and HONO levels. Appl. Catal. B 2015, 166–167, 84–90. [CrossRef] 18. Martinez, T.; Bertron, A.; Escadeillas, G.; Ringot, E.; Simon, V. BTEX abatement by photocatalytic TiO2- bearing coatings applied to cement mortars. Build. Environ. 2014, 71, 186–192. [CrossRef] 19. Folli, A.; Pade, C.; Hansen, T.B.; De Marco, T.; Macphee, D.E. TiO2 photocatalysis in cementitious systems: Insights into self-cleaning and depollution chemistry. Cem. Concr. Res. 2012, 42, 539–548. [CrossRef] 505 Catalysts 2018, 8, 553 20. Karapati, S.; Giannakopoulou, T.; Todorova, N.; Boukos, N.; Antiohos, S.; Papageorgiou, D.; Chaniotakis, E.; Dimotikali, D.; Trapalis, C. TiO2 functionalization for efﬁcient NOx removal in photoactive cement. Appl. Surf. Sci. 2014, 319, 29–36. [CrossRef] 21. Mo, J.; Zhang, Y.; Xu, Q.; Lamson, J.J.; Zhao, R. Photocatalytic puriﬁcation of volatile organic compounds in indoor air: A literature review. Atmos. Environ. 2009, 43, 2229–2246. [CrossRef] 22. Bartolomei, V.; Sörgel, M.; Gligorovski, S.; Alvarez, E.G.; Gandolfo, A.; Strekowski, R.; Quivet, E.; Held, A.; Zetzsch, C.; Wortham, H. References Prospect for distinguishing between the two kinetic models. J. Photochem. Photobiol. A Chem. 2000, 133, 89–97. [CrossRef] 33. Ollis, D.F. Kinetics of liquid phase photocatalyzed reactions: An illuminating approach. J. Phys. Chem. B 2005, 109, 2439–2444. [CrossRef] [PubMed] 34. Mills, A.; Wang, J.; Ollis, D.F. Kinetics of liquid phase semiconductor photoassisted reactions: Supporting observations for a pseudo-steady-state model. J. Phys. Chem. B 2006, 110, 14386–14390. [CrossRef] [PubMed] 35. Graetzel, M. (Ed.) Energy Resources through Photochemistry and Catalysis; Academic Press: New York, NY, USA, 1983; p. 573. 36. Clayton, R.K. Photosynthesis: Physical mechanisms and chemical patterns. IUPAB Biophys. Ser. 1980, 4, 36. Clayton, R.K. Photosynthesis: Physical mechanisms and chemical patterns. IUPAB Biophys. Ser. 1980, 4, 281. 37. Braslavsky, S.E. Glossary of terms used in photochemistry. (IUPAC Recommendations 2006). Pure Appl. Chem. 2007, 79, 293–465. [CrossRef] 37. Braslavsky, S.E. Glossary of terms used in photochemistry. (IUPAC Recommendations 20 Pure Appl. Chem. 2007, 79, 293–465. [CrossRef] 38. Emeline, A.V.; Sheremetyeva, N.V.; Khomchenko, N.V.; Kuzmin, G.N.; Ryabchuk, V.K.; Teoh, W.Y.; Amal, R. Spectroscopic studies of pristine and ﬂuorinated nano-ZrO2 in photostimulated heterogeneous processes. J. Phys. Chem. C 2009, 113, 4566–4583. [CrossRef] J. Phys. Chem. C 2009, 113, 4566–4583. [CrossRef] 39. Serpone, N. Is the band gap of pristine TiO2 narrowed by anion- and cation-doping of titanium dioxid second-generation photocatalysts? J. Phys. Chem. B 2006, 110, 24287–24293. [CrossRef] [PubMed] 39. Serpone, N. Is the band gap of pristine TiO2 narrowed by anion- and cation-doping of titanium dioxide in second-generation photocatalysts? J. Phys. Chem. B 2006, 110, 24287–24293. [CrossRef] [PubMed] 40. Serpone, N.; Emeline, A.V.; Kuznetsov, V.N.; Ryabchuk, V.K. Visible-light-active titania photocatalysts. The case of N-doped TiO2s—Properties and some fundamental issues. Int. J. Photoenergy 2008, 1, 1–19. 39. Serpone, N. Is the band gap of pristine TiO2 narrowed by anion- and cation-doping of titanium dioxide in second-generation photocatalysts? J. Phys. Chem. B 2006, 110, 24287–24293. [CrossRef] [PubMed] 40 Serpone N ; Emeline A V; Kuznetsov VN ; Ryabchuk VK Visible-light-active titania photocatalysts second-generation photocatalysts? J. Phys. Chem. B 2006, 110, 24287–24293. [CrossRef] [PubMed] 40. Serpone, N.; Emeline, A.V.; Kuznetsov, V.N.; Ryabchuk, V.K. Visible-light-active titania photocatalysts. The case of N-doped TiO2s—Properties and some fundamental issues. Int. J. Photoenergy 2008, 1, 1–19. 40. Serpone, N.; Emeline, A.V.; Kuznetsov, V.N.; Ryabchuk, V.K. Visible-light-active titania photocatalysts. The case of N-doped TiO2s—Properties and some fundamental issues. Int. J. Photoenergy 2008, 1, 1–19. 506 Catalysts 2018, 8, 553 41. References Serpone, N.; Emeline, A.V. Modeling heterogeneous photocatalysis by metal-oxide nanostructured semiconductor and insulator materials; Factors that affect the activity and selectivity of photocatalysts. Res. Chem. Intermed. 2005, 31, 391–432. [CrossRef] Res. Chem. Intermed. 2005, 31, 391–432. [CrossRef] 42. Emeline, A.V.; Serpone, N. Spectral selectivity of photocatalyzed reactions on the surface of titanium dioxide nanoparticles. J. Phys. Chem. B 2002, 106, 12221–12226. [CrossRef] 42. Emeline, A.V.; Serpone, N. Spectral selectivity of photocatalyzed reactions nanoparticles. J. Phys. Chem. B 2002, 106, 12221–12226. [CrossRef] 43. Emeline, A.V.; Zhang, X.; Jin, M.; Murokami, T.; Fujishima, A. Spectral dependences of the activity and selectivity of N-doped TiO2 in photodegradation of phenols. J. Photochem. Photobiol. A Chem. 2009, 207, 13–19. [CrossRef] 44. Emeline, A.V.; Kuzmin, G.N.; Serpone, N. Quantum yields and their wavelength-dependence in the photoreduction of O2 and photooxidation of H2 on a visible-light-active N-doped TiO2 system. Chem. Phys. Lett. 2008, 454, 279–283. [CrossRef] 45. Murakami, N.; Chiyoya, T.; Tsubota, T.; Ohno, T. Switching redox site of photocatalytic reaction on titanium(IV) oxide particles modiﬁed with transition-metal ion controlled by irradiation wavelength. Appl. Catal. A 2008, 348, 148–152. [CrossRef] 46. Baye, E.; Murakami, N.; Ohno, T. Exposed crystal surface-controlled TiO2 nanorods having rutile phase from TiCl3 under hydrothermal conditions. J. Mol. Catal. A Chem. 2009, 300, 72–79. [CrossRef] 47. Murakami, N.; Kurihara, Y.; Tsubota, T.; Ohno, T. Shape-controlled anatase titanium(IV) oxide particles prepared by hydrothermal treatment of peroxo titanic acid in the presence of polyvinyl alcohol. J. Phys. Chem. C 2009, 113, 3062–3069. [CrossRef] 48. Baye, E.; Ohno, T. Exposed crystal surface-controlled rutile TiO2 nanorods prepared by hydrothermal treatment in the presence of poly(vinyl)pyrrolidone. Appl. Catal. B Environ. 2009, 91, 634–639. [CrossRef] 49. Tachikawa, T.; Yamashita, S.; Majima, T. Evidence for crystal-face-dependent TiO2 photocatalysis from single-molecule imaging and kinetic analysis. J. Am. Chem. Soc. 2011, 133, 7197–7204. [CrossRef] [PubMed] 50. Emeline, A.V.; Zhang, X.; Murakami, T.; Fujishima, A. Activity and selectivity of photocatalysts in photodegradation of phenols. J. Hazard. Mater. 2012, 211–212, 154–160. [CrossRef] [PubMed] 51. Gerisher, H.; Heller, A. The role of oxygen in photooxidation of organic molecules on semiconductor particles. J. Phys. Chem. 1991, 95, 5261–5267. [CrossRef] 52. Emeline, A.V.; Kataeva, G.V.; Panasuk, A.V.; Ryabchuk, V.K.; Sheremetyeva, N.; Serpone, N. Effect of surface photoreactions on the photocoloration of a wide band gap metal oxide: Probing whether surface reactions are photocatalytic. J. Phys. Chem. B 2005, 109, 5175–5185. [CrossRef] [PubMed] 53. Terenin, A.N. References Kuznetsov, V.N.; Glazkova, N.I.; Mikhaylov, R.V.; Serpone, N. Additional speciﬁc channel of photo-activation of solid semiconductors. A revisit of the thermo-/photo-stimulated bleaching of photo-induced Ti3+ color centers in visible-light-active photochromic rutile titania. J. Phys. Chem. C 2018, 122, 13294–13303. [CrossRef] g p 71. Zecchina, A.; Lofthouse, M.G.; Stone, F.S. Reﬂectance spectra of surface states in magnesium oxide and calcium oxide. J. Chem. Soc. Faraday Trans. 1 Phys. Chem. Condens. Phases 1975, 71, 1476–1490. [CrossRef] 72. Zecchina, A.; Stone, F.S. Reﬂectance spectra of surface states in strontium oxide and barium oxide. J. Chem. Soc. Faraday Trans. 1 Phys. Chem. Condens. Phases 1976, 72, 2364–2374. [CrossRef] 73. Kristianpoller, N.; Rehavi, A.; Shmilevicha, A.; Weiss, D.; Chen, R. Radiation effects in pure and doped Al2O3 crystals. Nucl. Instrum. Methods Phys. Res. B 1998, 141, 343–346. [CrossRef] 74. Kortov, V.S.; Vainshtein, I.A.; Vokhmintsev, A.S.; Gavrilov, N.V. Spectroscopic characteristics of anionic centers in α-Al2O3 crystals bombarded by Cu+ and Ti+. J. Appl. Spectrosc. 2008, 75, 452–455. [CrossRef] 75. Izerrouken, M.; Benyahia, T. Absorption and photoluminescence study of Al2O3 single crystal irradiated with fast neutrons. Nucl. Instrum. Methods Phys. Res. B 2010, 268, 2987–2990. [CrossRef] 76. Itou, M.; Fujiwara, A.; Uchino, T. Reversible photoinduced interconversion of color centers in α-Al2O3 prepared under vacuum. J. Phys. Chem. C 2009, 113, 20949–20957. [CrossRef] 77. Emeline, A.V.; Kuzmin, G.N.; Purevdorj, D.; Ryabchuk, V.K.; Serpone, N. Spectral dependencies of the quantum yield of photochemical processes on the surface of wide band gap solids. 3. Gas/Solid systems. J. Phys. Chem. B 2000, 104, 2989–2999. [CrossRef] 78. Kuznetsov, V.N.; Lisachenko, A.A. Spectral manifestation of wide-band oxide own defects in photo- stimulated surface reactions. Russ. J. Phys. Chem. 1991, 65, 1328–1334. 79. Lisachenko, A. Photon-driven electron and atomic processes on solid-state surface in photoactivated spectroscopy and photocatalysis. J. Photochem. Photobiol. A Chem. 2008, 196, 127–137. [CrossRef] 80. Zakharenko, V.S.; Cherkashin, A.E.; Volodin, A.M.; Keier, N.P. Spectral dependence of oxygen and carbon monoxide photoadsorption on rutile. React. Kinet. Catal. Lett. 1979, 10, 325–332. [CrossRef] 81. Volodin, A.M.; Cherkashin, A.E.; Zakharenko, V.S. Inﬂuence of physically adsorbed oxygen on the separation of electron-hole pairs on anatase irradiated by visible light. React. Kinet. Catal. Lett. 1979, 11, 103–106. [CrossRef] 82. Emeline, A.V.; Smirnova, L.G.; Kuzmin, G.N.; Basov, L.L.; Serpone, N. Spectral dependence of quantum yields in gas/solid heterogeneous photosystems. Inﬂuence of anatase/rutile content on the photo- stimulated adsorption of dioxygen and dihydrogen on titania. J. Photochem. Photobiol. References Optical investigations of the adsorption of gas molecules. Uchenye Zapiski Leningrad. Gosudarst. Univ. Ser. Fiz. Nauk 1939, 5, 26–40. (In Russian) 54. Terenin, A.N. Optical investigations of activated adsorption. Z. Fiz. Khim. 1940, 14, 1362–1369. (In Russ 55. Kasparov, K.Ya.; Terenin, A. Optical investigations of activated adsorption. I. Photodecomposition of NH3 adsorbed on catalysts. Acta Physicochim. USSR 1941, 15, 343–365. 56. Terenin, A.N.; SoloIIitzyn, Y. Action of light on the gas adsorption by solids. Discuss. Faraday Soc. 1959, 28, 28–35. [CrossRef] 57. Ashcroft, N.W.; Mermin, N.D. Solid State Physics; Holt, Rinehart and Winston: New York, NY, USA, 197 58. Pankove, J.I. Optical Processes in Semiconductors; Dover Publications: New York, NY, USA, 1971. 59. Stoneham, A.M. Theory of Defects in Solids; Clarendon Press: Oxford, UK, 1975. 60. Henderson, B.; Werts, J.E. Defects in the Alkaline Earth Oxides; Taylor & Francis Ltd.: London, UK, 1977; p. 61. Kotomin, E.A.; Popov, A.I. Radiation-induced point defects in simple oxides. Nucl. Instrum. Methods Phys. Res. B 1998, 141, 1–15. [CrossRef] 62. Popov, A.I.; Kotomin, E.A.; Maier, J. Basic properties of the F-type centers in halides, oxides and perovskites Nucl. Instrum. Methods Phys. Res. B 2010, 268, 3084–3089. [CrossRef] m. Methods Phys. Res. B 2010, 268, 3084–3089. [CrossRe 63. Crawford, H.J. Recent developments in Al2O3 color-center research. Sem 63. Crawford, H.J. Recent developments in Al2O3 color-center research. Semicond. Insul. 1982, 5, 599–620. 64. Seebauer, E.G.; Kratzer, M.C. Charged point defects in semiconductors. Mater. Sci. Eng. R 2006, 55, 57–149. [CrossRef] 65. Schirmer, O.F. O– bound small polarons in oxide materials. J. Phys. Condens. Matter 2006, 18, R667–R704. [CrossRef] 66. Schirmer, O.F. Holes bound as small polarons to acceptor defects in oxide materials: Why are their thermal ionization energies so high? J. Phys. Condens. Matter 2011, 23, 334218. [CrossRef] [PubMed] 507 Catalysts 2018, 8, 553 67. Dolgov, S.A.; Kärner, T.; Lushchik, A.; Maaroos, A.; Nakonechnyi, S.; Shablonin, E. Trapped hole centers in MgO single crystals. Phys. Solid State 2011, 53, 1244–1252. [CrossRef] 68. Kuznetsov, V.N.; Emeline, A.V.; Glazkova, N.I.; Mikhaylov, R.V.; Serpone, N. Real-time in situ monitoring of optical absorption changes in visible-light-active TiO2 under light irradiation and temperature-programmed annealing. J. Phys. Chem. C 2014, 118, 27583–27593. [CrossRef] 69. Kuznetsov, V.N.; Ryabchuk, V.K.; Emeline, A.V.; Mikhaylov, R.V.; Rudakova, A.V.; Serpone, N. Thermo- and photo-stimulated effects on the optical properties of rutile titania ceramic layers formed on titanium substrates. Chem. Mater. 2013, 25, 170–177. [CrossRef] 70. References Removal of low concentration nitrogen oxides through photoassisted heterogene catalysis. J. Mol. Catal. 1994, 88, 93–102. [CrossRef] 98. Negishi, N.; Takeuchi, K.; Ibusuki, T. Surface structure of the TiO2 thin ﬁlm photocatalyst. J. Mater. Sci. 1998, 33, 5789–5794. [CrossRef] 99. Kominami, H.; Matsuura, T.; Iwai, K.; Ohtani, B.; Nishimoto, S.; Kera, Y. Ultra-highly active titanium(IV) oxide photocatalyst prepared by hydrothermal crystallization from titanium(IV) alkoxide in organic solvents. Chem. Lett. 1995, 24, 693–694. [CrossRef] 100. Kominami, H.; Kato, J.; Kohno, M.; Kera, Y.; Ohtani, B. Photocatalytic mineralization of acetic acid in aerated aqueous suspension of ultra-highly active titanium(IV) oxide prepared by hydrothermal crystallization in toluene. Chem. Lett. 1996, 25, 1051–1052. [CrossRef] 101. Hashimoto, K.; Wasada, K.; Toukai, N.; Kominami, H.; Kera, Y. Photocatalytic oxidation of nitrogen monoxide over titanium(IV) oxide nanocrystals large size areas. J. Photochem. Photobiol. A Chem. 2000, 136, 103–109. [CrossRef] 102. Lim, T.H.; Jeong, S.M.; Kim, S.D.; Gyenis, J. Photocatalytic decomposition of NO by TiO2 particles. J. Photochem. Photobiol. A Chem. 2000, 134, 209–217. [CrossRef] 103. Nakamura, I.; Negishi, N.; Kutsuna, S.; Ihara, T.; Sugihara, S.; Takeuchi, K. Role of oxygen vacancy in the plasma-treated TiO2 photocatalyst with visible light activity for NO removal. J. Mol. Catal. A Chem. 2000, 161, 205–212. [CrossRef] 104. Zhang, J.; Ayusawa, T.; Minagawa, M.; Kinugawa, K.; Yamashita, H.; Matsuoka, M.; Anpo, M. Investigations of TiO2 photocatalysts for the decomposition of NO in the ﬂow system: The role of pretreatment and reaction conditions in the photocatalytic efﬁciency. J. Catal. 2001, 198, 1–8. [CrossRef] p y y 105. Tanaka, T.; Teramura, K.; Arakaki, K.; Funabiki, T. Photoassisted NO reduction with NH3 over TiO2 photocatalyst. Chem. Commun. 2002, 22, 2742–2743. [CrossRef] 106. Dalton, J.S.; Janes, P.A.; Jones, N.G.; Nicholson, J.A.; Hallam, K.R.; Allen, G.C. Photocatalytic oxidation of NOx gases using TiO2: A surface spectroscopic approach. Environ. Pollut. 2002, 120, 415–422. [CrossRef] 107. Devahasdin, S.; Fan, C., Jr.; Li, K.; Chen, D.H. TiO2 photocatalytic oxidation of nitric oxide: Transient behavior and reaction kinetics. J. Photochem. Photobiol. A Chem. 2003, 156, 161–170. [CrossRef] 108. Toma, F.L.; Bertrand, G.; Klein, D.; Coddet, C. Photocatalytic removal of nitrogen oxides via titanium dioxide. Environ. Chem. Lett. 2004, 2, 117–121. [CrossRef] 109. Bowering, N.; Walker, G.S.; Harrison, P.G. Photocatalytic decomposition and reduction reactions of ni oxide over Degussa P25. Appl. Catal. B Environ. 2006, 62, 208–216. [CrossRef] 110. Roy, S.; Hegde, M.S.; Ravishankar, N.; Madras, G. References A. Chem. 2002, 148, 99–104. 83. Komaguchi, K.; Maruoka, T.; Nakano, H.; Imae, I.; Ooyama, Y.; Harima, Y. Electron-transfer reaction of oxygen species on TiO2 nanoparticles induced by sub-band-gap illumination. J. Phys. Chem. C 2010, 114, 1240–1245. [CrossRef] 84. Kuznetsov, V.N.; Serpone, N. On the origin of the spectral bands in the visible absorption spectra of visible-light-active TiO2 specimens: Analysis and assignments. J. Phys. Chem. C 2009, 113, 15110–15123. [CrossRef] 85. Artemiev, Y.M.; Ryabchuk, V.K. Introduction to Heterogeneous Photocatalysis (A Textbook); Saint Petersburg State University: Saint Petersburg, Russia, 1999. (In Russian) 86. Siline, A.R.; Trukhin, A.N. Point Defects and Elementary Excitations in Crystalline and Non-Crystalline SiO2; Zinatne: Riga, Latvia, 1985. 87. Ryvkin, S.M. Photoelectric Effects in Semiconductors; Consultants Bureau: New York, NY, USA, 1964. 88. Delany, A.C.; Dickerson, R.R.; Melchior, F.L.; Wartburg, A.F. Modiﬁcation of a commercial NOx detector high sensitivity. Rev. Sci. Instrum. 1982, 12, 1899–1902. [CrossRef] 89. Wolf, C.A.; Nieuwenhuys, B.E. The NO–H2 reaction over Pd(III). Surf. Sci. 2000, 469, 196–203. [CrossRef] 508 Catalysts 2018, 8, 553 90. Environmental Agency of Japan. Kankyo Hakusho; State of Environment in Japan: Tokyo, Japan, 1991. (In Japanese) 91. Takeuchi, K.; Ibusuki, T. Heterogeneous Photochemical Reactions and Processes in the Troposphere. In Encyclopedia of Environmental Control Technology; Cheremisonoff, P.N., Ed.; Gulf Publishing: Houston, TX, USA, 1989; p. 279. 92. Courbon, H.; Pichat, P. Room-temperature interaction of N18O with ultraviolet-illuminated titanium diox J. Chem. Soc. Faraday Trans. 1 1984, 80, 3175–3185. [CrossRef] 93. Hori, Y.; Nakatsu, A.; Suzuki, S. Heterogeneous photocatalytic oxidation of NO2 in aqueous suspen various semiconductor powders. Chem. Lett. 1985, 14, 1429–1432. [CrossRef] 94. Anpo, M.; Nomura, T.; Kitao, T.; Giamello, E.; Murphy, D.; Che, M.; Fox, M.A. Approach to de-NOx-ing photocatalysis. II. Excited state of copper ions supported on silica and photocatalytic activity for NO decomposition. Res. Chem. Intermed. 1991, 15, 225. [CrossRef] 95. Anpo, M.; Matsuoka, M.; Hanou, K.; Mishima, H.; Yamashita, H.; Patterson, H.H. The relationship between the local structure of copper(I) ions on Cu+/zeolite catalysts and their photocatalytic reactivities for the decomposition of NOx into N2 and O2 at 275 K. Coord. Chem. Rev. 1998, 171, 175–184. [CrossRef] 96. Anpo, M.; Ichihashi, Y.; Takeuchi, M.; Yamashita, H. Design of unique titanium dioxide photocatalysts by an advanced metal-ion implantation method and photocatalytic reactions under visible light irradiation. Res. Chem. lntermed. 1998, 24, 143–149. [CrossRef] 97. Ibusuki, T.; Takeuchi, K. References Creation of redox adsorption sites by Pd2+ ion substitution in nano-TiO2 for high photocatalytic activity of CO oxidation, NO reduction, and NO decomposition. J. Phys. Chem. C 2007, 111, 8153–8160. [CrossRef] 509 Catalysts 2018, 8, 553 111. Yin, S.; Liu, B.; Zhang, P.; Morikawa, T.; Yamanaka, K.-I.; Sato, T. Photocatalytic oxidation of NOx under visible LED light irradiation over nitrogen-doped titania particles with iron or platinum loading. J. Phys. Chem. C 2008, 112, 12425–12431. [CrossRef] 112. Roy, S.; Hegde, M.S.; Madras, G. Catalysis for NOx abatement. Appl. Energy 2009, 86, 2283–2297. [CrossR 112. Roy, S.; Hegde, M.S.; Madras, G. Catalysis for NOx abatement. Appl. Energy 2009, 86, 2283–2297. [CrossRef] 113. Skalska, K.; Miller, J.S.; Ledakowicz, S. Trends in NOx abatement: A review. Sci. Total Environ. 2010, 408, 3976–3989. [CrossRef] [PubMed] y g y pp gy 113. Skalska, K.; Miller, J.S.; Ledakowicz, S. Trends in NOx abatement: A review. Sci. Total Environ. 2010, 408, 3976–3989. [CrossRef] [PubMed] 114. Heo, I.; Kim, M.K.; Sung, S.; Nam, I.-S.; Cho, B.K.; Olson, K.L.; Li, W. Combination of photocatalysis and HC/SCR for improved activity and durability of deNOx catalysts. Environ. Sci. Technol. 2013, 47, 3657–3664. [CrossRef] [PubMed] [CrossRef] [PubMed] 115. International Energy Agency. 2016 Energy and Air Pollution; World Energy Outlook Special Report: Paris, France, 2016. 116. Nitrogen Oxides (NOx), Why and How They Are Controlled; Technical Bulletin; Report EPA-456/F-99-006R; United States Environmental Protection Agency: Research Triangle Park, NC, USA, 1999. 117. Fujishima, A.; Zhang, X.; Tryk, D.A. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2008, 63, 515–582. [CrossRef] 118. Hanus, M.J.; Harris, A.T. Nanotechnology innovations for the construction industry. Prog. Mater. Sci. 2013, 58, 1056–1102. [CrossRef] 119. Fresno, F.; Portela, R.; Suarez, S.; Coronado, J.M. Photocatalytic materials: Recent achievements and near future trends. J. Mater. Chem. A 2014, 2, 2863–2884. [CrossRef] 120. Church of 2000/Richard Meier & Partners. Available online: https://www.archdaily.com/20105/church-of- 2000-richard-meier (accessed on 12 October 2018). 121. Borgarello, E. TX Active® Principio Attivo Fotocatalitico—APPROFONDIMENTO TECNICO; Italcementi, Italcementi Group: Bergamo, Italy, October 2009; Available online: https://www.construction21.org/italia/ data/sources/users/62/txactiveapprofondimentoottobre2009ita.pdf (accessed on 12 October 2018). g , p ; , Italcementi Group: Bergamo, Italy, October 2009; Available online: https://www.construction21.org/italia/ data/sources/users/62/txactiveapprofondimentoottobre2009ita.pdf (accessed on 12 October 2018). 122. Murata, Y.; Kamitami, K.; Takeuchi, K. Air purifying blocks based on photocatalysis. In Proceedings of the JIPEA World Congress, Tokyo, Japan, 17–21 September 2000. 123. Guerrini, G.L.; Peccati, E. TUNNEL “UMBERTO I” IN ROME Monitoring Program Results; Report No. References In Proceedings of the RILEM International Symposium on Photocatalysis ‘Environment and Construction Materials’, Florence, Italy, 8–9 October 2007; pp. 179–186. International Symposium on Photocatalysis ‘Environment and Construction Materials’, Florence, Italy, 8–9 October 2007; pp. 179–186. 138. Guerrini, G.L. Photocatalytic performances in a city tunnel in Rome: NOx monitoring results. Constr. Build. Mater. 2012, 27, 165–175. [CrossRef] 139. Cassar, L. Photocatalysis of cementitious materials: Clean buildings and clean air. MRS Bull. 2004, 29, 328–331. [CrossRef] 140. Cassar, L.; Pepe, C.; Tognon, G.; Guerrini, G.L.; Amadelli, R. White cement for architectural concrete possessing photocatalytic properties. In Proceedings of the 11th International Congress on the Chemistry of Cement, Durban, South Africa, 11–16 May 2003; Volume 4, p. 12. 141. Guerrini, G.L.; Plassais, A.; Pepe, C.; Cassar, L. Use of photocatalytic cementitious materials for self-cleaning applications. In Proceedings of the RILEM International Symposium on Photocatalysis ‘Environment and Construction Materials’, Florence, Italy, 8–9 October 2007; pp. 219–226. 142. Guerrini, G.L.; Corazza, F. White cement and photocatalysis Part 1: Fundamentals. In Proceedings of the First Arab International Conference and Exhibition on The Uses of White Cement, Cairo, Egypt, 28–30 April 2008; Available online: https://www.researchgate.net/publication/266358310_WHITE_CEMENT_ AND_PHOTOCATALYSIS_PART_1_FUNDAMENTALS (accessed on 29 September 2018). First Arab International Conference and Exhibition on The Uses of White Cement, Cairo, Egypt, 28–30 April 2008; Available online: https://www.researchgate.net/publication/266358310_WHITE_CEMENT_ AND_PHOTOCATALYSIS_PART_1_FUNDAMENTALS (accessed on 29 September 2018). 143. NOx Gas. Available online: http://www.airlite.com/air-quality/nox-gas/ (accessed on 28 September 20 144. Borgarello, E.; Harris, R.; Serpone, N. Photochemical deposition and photorecovery of gold using semiconductor dispersions. A practical application. Nouv. J. Chim. 1985, 9, 743–747. 145. Borgarello, E.; Terzian, R.; Serpone, N.; Pelizzetti, E.; Barbeni, M. Photocatalyzed transformation of cyanide to thio-cyanate by rhodium-loaded cadmium sulﬁde in alkaline aqueous sulﬁde media. Inorg. Chem. 1986, 25, 2135–2137. [CrossRef] 146. Borgarello, E.; Serpone, N.; Emo, G.; Harris, R.; Pelizzetti, E.; Minero, C. Light-induced reduction of Rh(III) and Pd(II) on TiO2 dispersions, and the selective photochemical separation and recovery of Au(III), Pt(IV), and Rh(III) from dilute solutions. Inorg. Chem. 1986, 25, 4499–4503. [CrossRef] 147. Serpone, N.; Borgarello, E.; Barbeni, M.; Pelizzetti, E.; Pichat, P.; Herrmann, J.-M.; Fox, M.A. Photo-chemical reduction of gold(III) on semiconductor dispersions of TiO2 in the presence of cyanide ions: Disposal of CN−with H2O2. J. Photochem. 1987, 36, 373–388. [CrossRef] 148. Serpone, N.; Ah-You, Y.K.; Tran, T.P.; Harris, R.; Pelizzetti, E.; Hidaka, H. References 24; CTG-Italcementi Group: Bergamo, Italy, 22 April 2008; Available online: http://www.tiocem.pl/ﬁles/ references/TX_Active_Tunnel_Umberto_I_ENG.pdf (accessed on 28 September 2018). 124. Guerini, G.L. Case Study: The Italcementi TX Active® Story. Cristal Global Conference, London, UK, 17 November 2011. Available online: http://www.cristalactiv.com/uploads/speaker/Case_Study_The_ Italcementi_TX_Active_Story_Gian_Luca_Guerrini.pdf (accessed on 12 October 2018). 17 November 2011. Available online: http://www.cristalactiv.com/uploads/speaker/Case_Study_The_ Italcementi_TX_Active_Story_Gian_Luca_Guerrini.pdf (accessed on 12 October 2018). 125. Martinez, T.; Bertron, A.; Ringot, E.; Escadeillas, G. Degradation of NO using photocatalytic coatings applied to different substrates. Build. Environ. 2011, 46, 1808–1816. [CrossRef] 126. Staub de Melo, J.V.; Trichês, G. Evaluation of the inﬂuence of environmental conditions on the efﬁciency of photocatalytic coatings in the degradation of nitrogen oxides (NOx). Build. Environ. 2012, 49, 117–123. [CrossRef] 127. Ifang, S.; Gallus, M.; Liedtke, S.; Kurtenbach, R.; Wiesen, P.; Kleffmann, J. Standardization methods for testing photocatalytic air remediation materials: Problems and solution. Atmos. Environ. 2014, 91, 154–161. [CrossRef] 128. Zouzelka, R.; Rathousky, J. Photocatalytic abatement of NOx pollutants in the air using commercial functional coating with porous morphology. Appl. Catal. B Environ. 2017, 217, 466–476. [CrossRef] 129. Macphee, D.E.; Folli, A. Photocatalytic concretes—The interface between photocatalysis and cement chemistry. Cem. Concr. Res. 2016, 85, 48–54. [CrossRef] 130. Horgnies, M.; Dubois-Brugger, I.; Gartner, E.M. NOx de-pollution by hardened concrete and the inﬂuence of activated charcoal additions. Cem. Concr. Res. 2012, 42, 1348–1355. [CrossRef] 131. Bloh, J.Z.; Folli, A.; Macphee, D.E. Photocatalytic NOx abatement: Why the selectivity matters. RSC Adv. 2014, 4, 45726–45734. [CrossRef] 132. Yang, L.; Hakki, A.; Wang, F.; Macphee, D.E. Photocatalyst efﬁciencies in concrete technology: The effect of photocatalyst placement. Appl. Catal. B Environ. 2018, 222, 200–208. [CrossRef] 133. Erme, K.; Raud, J.; Jogi, I. Adsorption of nitrogen oxides on TiO2 surface as a function of NO2 and N2O5 fraction in the gas phase. Langmuir 2018, 34, 6338–6345. [CrossRef] [PubMed] 510 Catalysts 2018, 8, 553 134. Ofﬁcial Presentation—Innovative Façade Coatings with De-soiling and De-polluting Properties. In The PICADA Project—Photocatalytic Innovative Coverings Applications for Depollution Assessment; EC Project No. GRD1-2001-40449; GTM Construction: Nanterre, France, 2006. 135. Boonen, E.; Beeldens, A. Recent photocatalytic applications for air puriﬁcation in Belgium. Coatings 2005, 4, 553–573. [CrossRef] 136. Maggos, T.; Plassais, A.; Bartzis, J.G. Photocatalytic degradation of NOx in a pilot street canyon conﬁgura using TiO2-mortar panels. Environ. Monit. Assess. 2008, 136, 35–44. [CrossRef] [PubMed] 137. Guerrini, G.L.; Peccati, E. Photocatalytic cementitious roads for de-pollution. References AM1 simulated sunlight photoreduction and elimination of Hg(II) and CH3Hg(II) chloride salts from aqueous suspensions of titanium dioxide. Sol. Energy 1987, 39, 491–498. [CrossRef] 149. PICADA PROJECT—Performance Process Protocol, Workpackage 7 January 2006. Available online: http://www.picada-project.com/domino/SitePicada/Picada.nsf/1f9d19927a32e752c12569ab002c7ff8/ 50905a4f28b6ae58c12571320033f015/$FILE/D20.pdf (accessed on 18 September 2018). 150. Gurol, M.D. Photocatalytic Construction Materials and Reduction in Air Pollutants, San Diego State University, San Diego, CA, USA, March 2006. Available online: https://www.csus.edu/calst/FRFP/ PHOTO-CATALYTIC.pdf (accessed on 18 September 2018). 151. Maggos, Th.; Kotzias, D.; Bartzis, J.G.; Leva, P.; Bellintani, A.; Vasilakos, C. Investigations of TiO2-containing construction materials for the decomposition of NOx in environmental chambers. In Proceedings of the 5th International Conference on Urban Air Quality, Valencia, Spain, 29–31 March 2005. 152. The 2010 Report Dutch Air Quality Innovation Programme Concluded. Available online: https://laqm. defra.gov.uk/documents/Dutch_Air_Quality_Innovation_Programme.pdf (accessed on 29 September 2018). 153. Jacobi, S. NO2-Reduzierung Durch Photokatalytisch Wirksame Oberﬂächen? Modellversuch Fulda, (Hesse, Germany) 2012. Available online: https://www.hlnug.de/ﬁleadmin/dokumente/das_hlug/jahresbericht/ 2012/jb2012_059-066_I2_Jacobi_ﬁnal.pdf (accessed on 30 September 2018). 511 Catalysts 2018, 8, 553 154. Photocatalytic Titanium Dioxide—A Demonstrated and Proven Technology (Cristal ACTiv™), Cristal Global, London, UK. Available online: http://www.cristalactiv.com/uploads/case/casePhotocatalysis%20- %20English.pdf (accessed on 1 October 2018). 155. Burton, A. Titanium dioxide photocleans polluted air. Environ. Health Perspect. 2012, 120, A229. [CrossRef] [PubMed] 156. Laufs, S.; Burgeth, G.; Duttlinger, W.; Kurtenbach, R.; Maban, M.; Thomas, C.; Wiesen, P.; Kleffmann, J. Conversion of nitrogen oxides on commercial photocatalytic dispersion paints. Atmos. Environ. 2010, 44, 2341–2349. [CrossRef] 157. Ballari, M.M.; Brouwers, H.J.H. Full scale demonstration of air-purifying pavement. J. Hazard. Mater. 2013, 254–255, 406–414. [CrossRef] [PubMed] 158. Boonen, E.; Akylas, V.; Barmpas, F.; Boreave, A.; Bottalico, L.; Cazaunau, M.; Chen, H.; Daele, V.; De Marco, T.; Doussin, J.F.; et al. Construction of a photo- catalytic de-polluting ﬁeld site in the Leopold II tunnel in Brussels. J. Environ. Manag. 2015, 155, 136–144. [CrossRef] [PubMed] 159. Fraunhofer. Clean Air by Airclean®. 2010. Available online: http://www.ime.fraunhofer.de/content/ dam/ime/de/documents/AOe/2009_2010_Saubere%20Luft%20durch%20Pﬂastersteine_s.pdf (accessed on 11 May 2015). 160. Tera. In Situ Study of the Air Pollution Mitigating Properties of Photocatalytic Coating, Tera Environement, (Contract Number 0941C0978), Report for ADEME and Rhone-Alpe region, France. Available online: http://www.air-rhonealpes.fr/site/media/telecharger/651413 (accessed on 11 May 2015). 161. Gustafsson, R.J.; Orlov, A.; Grifﬁths, P.T.; Cox, R.A.; Lambert, R.M. Reduction of NO2 to nitrous acid on illuminated titanium dioxide aerosol surfaces: Implications for photocatalysis and atmospheric chemistry. Chem. Commun. 2006, 3936–3938. [CrossRef] [PubMed] 162. Ndour, M.; D’Anna, B.; George, C.; Ka, O.; Balkanski, Y.; Kleffmann, J.; Stemmler, K.; Ammann, M. References Photoenhanced uptake of NO2 on mineral dust: Laboratory experiments and model simulations. Geophys. Res. Lett. 2008, 35, L05812. [CrossRef] 163. Beaumont, S.K.; Gustafsson, R.J.; Lambert, R.M. Heterogeneous photochemistry relevant to the troposphere: H2O2 production during the photochemical reduction of NO2 to HONO on UV-illuminated TiO2 surfaces. Chem. Phys. Chem. 2009, 10, 331–333. [CrossRef] [PubMed] 164. Monge, M.E.; D’Anna, B.; George, C. Nitrogen dioxide removal and nitrous acid formation on titanium oxide surfaces—An air quality remediation process? Phys. Chem. Chem. Phys. 2010, 12, 8991–8998. [CrossRef] [PubMed] 165. Mothes, F.; Herrmann, H. Lab and ﬁeld studies on photocatalysis to improve urban air quality—Results from the PhotoPAQ project. In Proceedings of the Life MINOx-STREET Project Ending Meeting: Results and Conclusions, CIEMAT, Madrid, Spain, 21 March 2018. 166. Flassak, T. Numerical simulation of the depollution effectiveness of photocatalytic coverings in street canyons. In Proceedings of the Photocatalysis: Science and Application for Urban Air Quality, The 2012 LIFE+ PhotoPaq Conference, Proticcio, Island of Corsica, France, 14–17 May 2012. 167. Bolte, G.; Flassak, T. Numerische simulation der wirksamkeit photokatalytisch aktiver betonoberﬂächen. In Proceedings of the Internationale Baustofftagung 18, Ibausil (Proceedings), Bauhaus-University Weimar, Weimar, Germany, 12–15 September 2012; Fischer, H.-B., Bode, K.-A., Beuthan, C., Eds.; Bauhaus-University Weimar: Weimar, Germany, 2012. 168. Pujadas, M.; Palacios, M.; Nunez, L.; German, M.; Fernandez-Pampillon, J.; Iglesias, J.D.; Santiago, J.L. Real scale demonstration of the depolluting capabilities of a photocatalytic pavement in a real urban area. In Proceedings of the 17th International Conference on Harmonization within Atmospheric Dispersion Modeling for Regulatory Purposes, Budapest, Hungary, 9–12 May 2016. 169. Pujadas, M.; Palacios, M.; Nunez, L.; German, M.; Fernandez-Pampillon, J.; Sanchez, B.; Santiago, J.L.; Sanchez, B.; Munos, R.; Moral, F.; et al. Real scale tests of the depolluting capabilities of a photocatalytic sidewalk pavement and a façade in an urban scenario. In Proceedings of the 18th International Conference on Harmonization within Atmospheric Dispersion Modelling for Regulatory Purposes, Bologna, Italy, 9–12 October 2017. 170. Palacios, M.; Pujadas, M.; Nunez, L.; Sanchez, B.S.; Santiago, J.L.; Martilli, A.; Suarez, S.; Cabrero, B.S. Monitoring and modeling NOx removal efﬁciency of photocatalytic materials: A strategy for urban air quality management. In Proceedings of the Life-Platform Meeting on Air, Barcelona, Spain, 26–27 September 2017. 512 Catalysts 2018, 8, 553 171. Pujadas, M.; Palacios, M.; Nunez, L.; Fernandez-Pampillon, J.; German, M. Characterization of the NOx depolluting effect of photocatalytic materials in a medium-scale tunnel reactor. References In Proceedings of the Air Quality Meeting, Barcelona, Spain, 12–16 March 2018. 172. Wang, H.; Wu, Z.; Zhao, W.; Guan, B. Photocatalytic oxidation of nitrogen oxides using TiO2 loading on woven glass fabric. Chemosphere 2007, 66, 185–190. [CrossRef] [PubMed] 173. Anpo, M.; Zhang, S.G.; Mishima, H.; Matsuoka, M.; Yamashita, H. Design of photocatalyst encapsulated within the zeolite framework and cavities for the decomposition of NO into N2 and O2 at normal temperature. Catal. Today 1997, 39, 159–168. [CrossRef] 174. Anpo, M.; Takeuchi, M.; Ikeue, K.; Dohshi, S. Design and development of titanium oxide photocatalysts operating under visible and UV light irradiation: The applications of metal ion-implantation techniques to semiconducting TiO2 and Ti/zeolite catalysts. Curr. Opin. Solid State Mater. Sci. 2002, 6, 381–388. [CrossRef] 175. Yamashita, H.; Ichihashi, Y.; Zhang, S.G.; Matsumura, Y.; Souma, Y.; Tatsumi, T.; Anpo, M. Photocatalytic decomposition of NO at 275 K on titanium oxide catalysts anchored within zeolite cavities and framework. Appl. Surf. Sci. 1997, 121, 305–309. [CrossRef] 176. Wu, Q.; van de Krol, R. Selective photoreduction of nitric oxide to nitrogen by nanostructured TiO2 photocatalysts: Role of oxygen vacancies and iron dopant. J. Am. Chem. Soc. 2012, 134, 9369–9375. [CrossRef] [PubMed] 177. Hu, Y.; Martra, G.; Zhang, J.; Higashimoto, S.; Coluccia, S.; Anpo, M. Characterization of the local structures of Ti-MCM-41 and their photocatalytic reactivity for the decomposition of NO into N2 and O2. J. Phys. Chem. B 2006, 110, 1680–1685. [CrossRef] [PubMed] 178. Cao, Y.; Yu, M.; Qi, S.; Ren, Z.; Yan, S.; Hu, S.; Xu, M. Nitric oxide reaction pathways on rutile TiO2(110): The inﬂuence of surface defects and reconstructions. J. Phys. Chem. C 2018. [CrossRef] 179. Kuznetsov, V.N.; Glazkova, N.I.; Mikhaylov, R.V.; Kozhevina, A.V.; Serpone, N. Photophysics of color centers in visible-light-active rutile titania. Evidence of the photoformation and trapping of charge carriers from advanced diffuse reﬂectance spectroscopy and mass spectrometry. Catal. Today 2018. [CrossRef] 180. Déak, P.; Aradi, B.; Frauenheim, T. Quantitative theory of the oxygen vacancy and carrier self-trapping in bulk TiO2. Phys. Rev. B 2012, 86, 195206. [CrossRef] 181. Cotton, F.A.; Wilkinson, G. Advanced Inorganic Chemistry, 5th ed.; John Wiley & Sons: New York, USA, 1988. 182. Iwamoto, M.; Furukawa, H.; Mine, Y.; Uemura, F.; Mikuriya, S.-I.; Kagawa, S. Copper(II) ion-exchanged ZSM-5 zeolites as highly active catalysts for direct and continuous decomposition of nitrogen monoxide. J. Chem. Soc. Chem. Commun. 1986, 1272–1273. [CrossRef] 183. Mechanistic Study on Facet-Dependent Deposition of Metal Nanoparticles on Decahedral-Shaped Anatase Titania Photocatalyst Particles Kenta Kobayashi 1, Mai Takashima 1,2,, Mai Takase 3 and Bunsho Ohtani 1,2 1 Graduate School of Environmental Science, Hokkaido University, Sapporo 060-0810, Japan; kobayashi.k@cat.hokudai.ac.jp (K.K.); ohtani@cat.hokudai.ac.jp (B.O.) 2 Institute for Catalysis, Hokkaido University, Sapporo 001-0021, Japan 3 Graduate School of Engineering, Muroran Institute of Technology, Mizumoto-cho, Muroran 050-8585, Japan; mai@mmm.muroran-it.ac.jp Correspondence: takashima m@cat hokudai ac jp; Tel : +81-11-706-9130 * Correspondence: takashima.m@cat.hokudai.ac.jp; Tel.: +81-11-706-9130 * Correspondence: takashima.m@cat.hokudai.ac.jp; Tel.: +81-11-706-9130 Received: 11 October 2018; Accepted: 9 November 2018; Published: 13 November 2018 Received: 11 October 2018; Accepted: 9 November 2018; Published: 13 November 20 Received: 11 October 2018; Accepted: 9 November 2018; Published: 13 November 2018 Abstract: Facet-selective gold or platinum-nanoparticle deposition on decahedral-shaped anatase titania particles (DAPs) exposing {001} and {101} facets via photodeposition (PD) from metal-complex sources was reexamined using DAPs prepared with gas-phase reaction of titanium (IV) chloride and oxygen by quantitatively evaluating the area deposition density on {001} and {101} and comparing with the results of deposition from colloidal metal particles in the dark (CDD) or under photoirradiation (CDL). The observed facet selectivity, more or less {101} preferable, depended mainly on pH of the reaction suspensions and was almost non-selective at low pH regardless of the deposition method, PD or CDL, and the metal-source materials. Based on the results, the present authors propose that facet selectivity is attributable to surface charges (zeta potential) depending on the kind of facets, {001} and {101}, and pH of the reaction mixture and that this concept can explain the observed facet selectivity and possibly the reported facet selectivity without taking into account facet-selective reaction of photoexcited electrons and positive holes on {101} and {001} facets, respectively. Keywords: decahedral-shaped anatase titania particles; {001} and {101} facets; facet-selective metal photodeposition; pH dependence; zeta potential; facet-selective reaction catalysts catalysts Catalysts 2018, 8, 542; doi:10.3390/catal8110542 www.mdpi.com/journal/catalysts References Chiu, K.W.; Savage, P.D.; Wilkinson, G.; Williams, D.J. Nitrosation of alkenes by nitric oxide: Crystal structures of bis-(1-nitroso-2-nitro-cyclohexane) and bis-(1-nitroso-2-nitro-1-phenylethane). Polyhedron 1985, 4, 1941–1945. [CrossRef] 184. Greenwood, N.N.; Earnshaw, A. Chemistry of the Elements, 2nd ed.; Butterworth-Heinemann: Oxford, UK, 1997. 185. Thiemann, M.; Scheibler, E.; Wiegand, K.W. Nitric Acid, Nitrous Acid, and Nitrogen Oxides. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim, Germany, 2005. 186. Finlayson-Pitts, B.J.; Wingen, L.M.; Sumner, A.L.; Syomin, D.; Ramazan, K.A. The heterogeneous hydrolysis of NO2 in laboratory systems and in outdoor and indoor atmospheres: An integrated mechanism. Phys. Chem. Chem. Phys. 2003, 5, 223–242. [CrossRef] 187. Sivachandrian, L.; Thevenet, F.; Gravejat, P.; Rousseau, A. Investigation of NO and NO2 adsorption mechanisms on TiO2 at room temperature. Appl. Catal. B Environ. 2013, 142–143, 196–204. [CrossRef] 188. Haubrich, J.; Quiller, R.G.; Benz, L.; Liu, Z.; Friend, C.M. In Situ ambient pressure studies of the chemistry of NO2 and water on rutile TiO2(110). Langmuir 2010, 26, 2445–2451. [CrossRef] [PubMed] © 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 513 1. Introduction Since the fate of electron-positive hole pairs is limited to an alternative, chemical reaction or recombination, the pairs that are not used to liberate photocatalytic-reaction products must disappear by mutual recombination [4]. Based on a simple kinetic assumption, the overall efﬁciency of electron-hole utilization is regulated by the ratio of the rate of this alternative; the efﬁciency must be high and low when the ratio is high and low, respectively. A frequently found misconception in papers on photocatalysis is low efﬁciency (low activity) being attributed only to faster recombination; slower electron/positive hole transfer to a substrate(s) can reduce efﬁciency/activity even though recombination occurs at a constant rate. Another misconception involves the recognition of charge recombination as a counter backward process of charge separation, i.e., charge recombination occurs because charges are not spatially separated. Thus, charge separation, a possible physical process just after photoabsorption, has been conveniently used in the interpretation of results of chemical reactions detected in chemical analysis without any support or evidence. Anyway, if electron-positive hole pairs are created in the bulk of photocatalyst particles, charge separation would be expected to occur only when there is an internal electric ﬁeld (IEF) in each particle; it seems impossible to separate negative and positive charges without an IEF overcoming the attractive electrostatic force between them. However, as described above, such an IEF does not seem to exist in not heavily (or negligibly) doped semiconducting materials, and even if there is an IEF from the surface to the bulk of a particle, a charge, electron or positive hole, separated to the bulk cannot react with a surface-adsorbed substrate(s). Therefore, when charge separation induced by an IEF is expected, a photocatalyst particle must have (i) two kinds of surfaces with different potentials and (ii) a smoothly changing bulk structure from one surface to the other surface forming a potential slope, i.e., an IEF in the bulk, though such a ﬁne structure, especially providing (ii), seems unrealistic. Titanium (IV) oxide (titania) is one of the most promising photocatalyst materials and is well known to exhibit high level of photocatalytic activity in various kinds of heterogeneous photocatalytic reactions. Three kinds of crystalline polymorphs of titania (with negligible occurrence of TiO2(B) [5,6]) have been found: Anatase, rutile and brookite as natural minerals in characteristic octahedral, complexed faceted and hexagonal plate-like crystal shapes, respectively. 1. Introduction The term “charge separation” is one of the most attractive and convenient terms and/or concepts for researchers in the ﬁeld of heterogeneous photocatalysis [1,2]. Since charge separation, i.e., spatial separation of a photoexcited electron in the conduction band and a positive hole in the valence band, is a kind of physical process and since the separated charges recombine easily with each other within a very short time period if there is no subsequent chemical reaction consuming those charges, direct observation of the single physical process of charge separation seems practically impossible. Time-resolved pump-probe spectroscopy using a femtosecond laser system has been reported to show such charge separation, i.e., accumulation of trapped photoexcited electrons was completed within a laser pulse and only the decay (disappearance) of those separated charges could be followed. To the best of the authors’ knowledge, however, there has been no reported evidence of the primary step of charge separation itself. Then, why has “charge separation” been believed to occur in photocatalyst materials under photoirradiation? One possible reason is speculation assuming a mechanism as an analogy of a photoelectrochemical reaction of semiconductor electrodes, in which there is a space charge layer, i.e., an electric ﬁeld in a semiconductor being in contact with the electrolyte; for n-type semiconductors such as metal oxides, positive holes and photoexcited electrons are made to migrate 515 Catalysts 2018, 8, 542 to the surface and the bulk of an electrode, respectively [2,3]. It should be noted that the depth of the space charge layer depends on the donor density of the electrode material, and it is known that ordinary (non-doped) semiconductor particles such as “white” titanium (IV) oxide particles have a negligible donor density to make the depth larger than the particle size, i.e., there may not be a space charge layer in those particles. Another reason for believing the occurrence of “charge separation” is simple; this concept is very convenient for interpretation of results of photocatalytic reactions; for example, high and low quantum efﬁciencies or photocatalytic activities have been explained by high and low extents of charge separation without showing direct evidence as described above. What we observe (or can observe) is the results of chemical reactions that follow physical processes, photoexcitation and possible charge separation. 1. Introduction For natural anatase crystals, though they are predominantly octahedral crystals exposing eight equivalent {101} facets, decahedral-shaped crystals exposing an additional two {001} facets have occasionally been found. On the other hand, when titania particles were prepared in various procedures, octahedral-shaped anatase particles have been rarely found except for reports from the authors’ group [7,8], presumably because natural crystals might grow very slowly for satisfying the thermodynamic requirement to expose only the most stable (lowest energy) facets, {101}. However, detailed analysis revealed that decahedral-shaped anatase particles (DAPs) existed in titania particles, e.g., Evonik (previously Degussa) P25 [9], though selective preparation of DAPs had not been reported before the publications of independent works by Yang and coworkers [10] and the present authors’ group [11] using a hydrothermal reaction with a structure-controlling agent (SCA) and using a gas-phase reaction of titanium (IV) chloride and oxygen, respectively. It has been believed that DAPs exposing two {001} and eight {101} facets have a high level of photocatalytic activity because photoexcited electrons and positive holes migrate to the {101} and {001} facets, respectively, resulting in efﬁcient charge separation [12–23]. However, since such “charge 516 Catalysts 2018, 8, 542 separation” cannot be observed directly as mentioned above, the reason seems to be just speculation or a hypothesis proposed on the basis of microscopic observations of metal- and metal oxide-deposited DAPs through photocatalytic reduction and oxidation from their precursors on {101} and {001} facets, respectively [12–20]. There are also reports about facet-selective metal/metal-oxide deposition on the surface of rutile titania and other metal-oxide particles with a polyhedral shape [24–27]. These results have been attributed to facet-selective reduction and oxidation on {101} and {001} facets, i.e., as far as the authors know, all reports except for one [28] have indicated that photoexcited electrons and positive holes migrate selectively to {101} and {001} facets, respectively. 1. Introduction It seems that this “facet-selective redox (FSR) hypothesis” has become established, and studies are now focusing on the possible mechanism, e.g., facet-dependent band positions [12–14], though the basic assumptions that (i) electron-hole pairs are created in the bulk and then move to the surface and (ii) electrons and holes migrate to different facets depending on the band energies seem inconsistent with (a) observation by femtosecond pump-probe laser spectroscopy of electrons and holes being trapped quickly in the surface states [29] and (b) the fact that charges, electrons and holes, cannot undergo separation and/or directional migration to escape from their electrostatic attraction without an electric ﬁeld in the bulk of each particle. However, before discussing the mechanism of FSR, experimental results of photocatalytic metal/metal-oxide deposition leading to the FSR hypothesis may have several problems. For example, (a) the effect of possibly remaining SCAs, used in order to prepare facetted particles, is neglected, (b) the facet-selective deposition has been evaluated only qualitatively using a few scanning electron microscopic images, without showing the number (or volume) ratio of deposits depending on the type of facets and (c) there has been no discussion of the possible migration of metal/metal-oxide deposits after photodeposition. On the basis of the above-mentioned background, FSR on facetted anatase titania particles was re-examined in this study using (1) DAPs prepared by gas-phase reaction of titanium (IV) chloride and oxygen without using SCAs through (2) quantitative analysis by counting the number of deposited particles in order to evaluate the surface-density ratio of facet-selective deposition via an ordinary photocatalytic reaction and (3) photoassisted deposition from colloidal metal particles; points (1), (2) and (3) were introduced/employed to solve (or suppress) problems (a), (b) and (c), respectively. 2. Results and Discussion 2. Results and Discussion 2.1. Decahedral-Shaped Anatase Titania Particles (DAPs) Used in This Study The sample particles, DAPs, were prepared by a previously reported procedure, coaxial-ﬂow gas-phase reaction of titanium (IV) chloride and oxygen as schematically shown in Figure 1. One of the features of this DAP sample is its high level of purity, possibly because only titanium, chlorine and oxygen are involved in the preparation system, i.e., no SCAs and even hydrogen sources such as water or organic compounds are included. Although the reason why DAPs exposing only {101} and {001} facets are selectively prepared has not been clariﬁed yet, it is speculated that DAPs are liberated as a lower surface area/volume ratio only exposing two kinds of facets ({101} and {001}) as the lowest and appreciably low surface energy [30,31], based on the observation that the observed oblateness (ratio of the short side to the long side of the {101} trapezoid (b/a: see Materials and Methods 3.5.)) was ca. 0.7 and it was not changed even when the preparation conditions were modiﬁed (See Materials and Methods). The lowest surface area (nm2)-volume (nm3) ratio (SV ratio; 8.2 nm−1) of ideal decahedral particles is expected to be obtained with oblateness of ca. 0.5 assuming the same surface energy of {101} and {001} facets. This oblateness of the lowest SV ratio seems smaller than the observed ratio, though a plot of SV ratio as a function of oblateness seems parabolic as shown in Section 3.5 and the SV ratio at 0.7 oblateness, 8.9 nm−1, seems not so high compared with the SV ratio at 0.5. Although there is still a possibility that the sole by-product of the gas-phase reaction, chlorine, remained on the {001} facets even after thorough washing with water and this led to lowering of the {001} surface energy and high 517 Catalysts 2018, 8, 542 Catalysts 2018, 8, 542 oblateness, it can be stated that the particle shape is not controlled by the surface energy (stability) of each particle, not surface modiﬁers, and this sample seems suitable for study on FSR behavior. In any case, even if the observed oblateness was 0.7, more than 70% of the entire exposed surface area of a DAP was {101}. 2. Results and Discussion oblateness, it can be stated that the particle shape is not controlled by the surface energy (stability) of each particle, not surface modiﬁers, and this sample seems suitable for study on FSR behavior. In any case, even if the observed oblateness was 0.7, more than 70% of the entire exposed surface area of a DAP was {101}. Figure 1. Schematic representation of the DAP synthesis process by coaxial-ﬂow gas-phase reaction of titanium (IV) chloride and oxygen. Figure 1. Schematic representation of the DAP synthesis process by coaxial-ﬂow gas-phase reaction of titanium (IV) chloride and oxygen. Figure 2 shows a representative electron-microscopic image of the sample. In this image, all of the particles have a sharp-edged decahedral shape with sizes of ca. 50–200 nm. Ridges in these observed DAPs looked white due to the “edge effect”, i.e., secondary-electron emission occurs preferably at sharp edges. In other words, the DAP samples used in this study have sharp edges. Other images for the samples showed a similar trend except for a few particles with a non-decahedral shape, which might be rutile or non-crystalline titania particles as described below. The crystallite size of anatase evaluated from the XRD patterns with the Scherrer equation was ca. 75 nm (d101: 71 nm, d004: 68 nm and d200: 87 nm), which was almost the same as the above-mentioned particle size evaluated from SEM images. This rough coincidence suggests that each DAP appearing in SEM images was a single crystal anatase particle. Figure 2. FE-SEM image of DAPs synthesized by coaxial-ﬂow gas-phase reaction. Figure 2. FE-SEM image of DAPs synthesized by coaxial-ﬂow gas-phase reaction. The crystalline content of the DAP sample used was shown by Rietveld analysis of the diffraction pattern to be 89% anatase and 4% rutile with a 7% non-crystal component. 2.2. Deposition of Metal Particles on DAPs 2.2. Deposition of Metal Particles on DAPs 2.2. Deposition of Metal Particles on DAPs 2.2. Deposition of Metal Particles on DAPs SEM images of gold and platinum-deposited DAPs through photodeposition (PD; (a) and (d)) from precursors and colloid deposition in the dark (CDD; (b) and (e)) and that under photoirradiation (CDL; (c) and (f)) are shown in Figure 3. The roughly estimated particle sizes of both gold and platinum deposits, shown in Figure 3, were in the range of 4–12 nm (Table 1) and no distinct difference in the size depending on the kind of facets was observed. For metal particles photodeposited from their 518 Catalysts 2018, 8, 542 precursors, the size was 4–5 nm regardless of the kind of metals, though the possibility of small (<1 nm) particle formation could not be excluded due to the resolution of SEM analysis in this study. The average particle size of the original gold colloid was ca. 12 nm and this was not changed by loading under photoirradiation and in the dark, while the original size of platinum colloid particles was ca. 5 nm. The two-times larger size after deposition in the dark is attributable to possible aggregation of a few platinum particles, though each of them was observed as one particle. precursors, the size was 4–5 nm regardless of the kind of metals, though the possibility of small (<1 nm) particle formation could not be excluded due to the resolution of SEM analysis in this study. The average particle size of the original gold colloid was ca. 12 nm and this was not changed by loading under photoirradiation and in the dark, while the original size of platinum colloid particles was ca. 5 nm. The two-times larger size after deposition in the dark is attributable to possible aggregation of a few platinum particles, though each of them was observed as one particle. Figure 3. Representative FE-SEM images of DAPs deposited with gold (a–c) and platinum (d–f) by photodeposition (PD) from their precursors (HAuCl4 and H2PtCl6, respectively) (a,d), deposition from their colloids in the dark (CDD; b,e) and under photoirradiation (CDL;c,f). Figure 3. Representative FE-SEM images of DAPs deposited with gold (a–c) and platinum (d–f) by photodeposition (PD) from their precursors (HAuCl4 and H2PtCl6, respectively) (a,d), deposition from their colloids in the dark (CDD; b,e) and under photoirradiation (CDL;c,f). Table 1. Summary of the results for deposition of metals on DAPs. Table 1. 2.3. Mechanism of Metal Deposition on DAPs and FSR Although it has not been discussed so far, the mechanism of photocatalytic metal-particle deposition does not seem to be straightforward since the formation of deposited metal particles composed of a large number of metal atoms requires a process to make metal-metal bonds. At least three steps may be included. (1) The ﬁrst step is reduction of precursor metal-complex anions (AuCl4− and PtCl62−) adsorbed on the surface of photocatalyst particles by photoexcited electrons along with oxidation of a sacriﬁcial hole scavenger, methanol in the present case, by positive holes. (2a) The next step is migration of low-valent or metal atoms and/or clusters to grow into metal particles (as detected in SEM analysis). It is expected that the larger the size of meal clusters or nanoparticles is, the lower is the possibility of surface migration. (2b) The next step is reduction of metal precursor anions on the surface of formed metal nanoparticles being in contact with titania. Both steps (2a) and (2b) are probable, and facet-selective deposition is regulated by (i) facet-selective migration of photoexcited electrons and positive holes in step (1), (ii) facet-selective adsorption of metal precursor anions in step (1/2b) and/or (iii) surface migration and facet-selective attachment (ﬁxation) of metal clusters/nanoparticles in steps (2a/2b). At least one of them, if there are no other possible reasons, can lead to a value of s that is different from unity, though previous reports suggested reason (i) without showing evidence that excludes the possibility of (ii) and (iii). In order to check the possibility of (ii) and (iii), deposition of metal particles on DAPs from gold and platinum colloid solutions was examined in the dark (CDD) and under photoirradiation (CDL). It is well known that colloidal particles prepared using citric acid, which was used in this study, are stabilized by coverage of their surface with citric acid and repulsion with negative charges with carboxylate groups, i.e., the metal-particle surface is negatively charged. For CDD, as shown in Table 1, the area deposition density (D) was lower than that with PD presumably because there seems to be no driving force for colloid particles to settle on the surface (other than interaction with possible protonated hydroxyl groups on the DAP surface) and a higher concentration (2 wt %) of the colloid solution was necessary to count the number of deposited metal-particles. On the other hand, CDL gave a high D compared to that with CDD. 2.2. Deposition of Metal Particles on DAPs Summary of the results for deposition of metals on DAPs. Entry Source Amount 1 (wt %) Medium Size 2/nm pH 3 D{001} 4/10−4 nm−2 D{101} 5/10−4 nm−2 S 6 PD 7(Au) HAuCl4 0.5 MeOH 8 5 — 9 3.9 6.4 0.61 CDD 10(Au) Au colloid 2.0 water 12 — 9 0.065 0.37 0.18 CDL 11(Au) Au colloid 0.5 water 12 7.6 0.011 0.25 0.04 PD 7(Pt) H2PtCl6 0.5 MeOH 8 4 7.2 4.5 16.0 0.28 PD 7(Pt/CA) H2PtCl6 0.5 MeOH 8 + CA 12 5 2.5 5.1 8.2 0.62 CDD 10(Pt) Pt colloid 2.0 water 10 — 9 0.12 0.48 0.24 CDL 11(Pt) Pt colloid 0.5 water 5 7.0 0.76 16.0 0.05 CDL 11(Pt) Pt colloid 2.0 water 5 8.9 0.676 22.0 0.03 CDL 11(Pt/CA) Pt colloid 0.5 water +CA 12 4 2.4 6.6 9.0 0.73 1 Amount as metal. 2 Roughly estimated average size of metal deposits assuming a spherical shape. 3 Measured after deposition. 4 Area deposition density of metal deposits for {001} facets. 5 Area deposition density of metal deposits for {101} facets. 6 Facet selectivity (=D{001}/D{101}). 7 Photodeposition. 8 50 vol % aqueous methanol. 9 Not measured. 10 Colloid deposition in the dark. 11 Colloid deposition under UV irradiation. 12 Citric acid (0.1 mol L−1). 1 Amount as metal. 2 Roughly estimated average size of metal deposits assuming a spherical shape. 3 Measured after deposition. 4 Area deposition density of metal deposits for {001} facets. 5 Area deposition density of metal deposits for {101} facets. 6 Facet selectivity (=D{001}/D{101}). 7 Photodeposition. 8 50 vol % aqueous methanol. 9 Not measured. 10 Colloid deposition in the dark. 11 Colloid deposition under UV irradiation. 12 Citric acid (0.1 mol L−1). As a general trend, gold and platinum particles were deposited preferably on {101} facets. In order to evaluate the facet selectivity quantitatively, facet selectivity (s) was deﬁned as the ratio of number (not volume) density per unit area, i.e., area density, of metal deposits on {001} (D{001}) and {101} (D{001}) facets determined by counting more than 100 DAPs in several SEM images for each sample (See Materials and Methods). Since the exposed surface was predominantly {101} facets for DAPs used in this study, a comparison without consideration of the surface areas of the two kinds of facets does not seem to make sense. 2.2. Deposition of Metal Particles on DAPs 519 Catalysts 2018, 8, 542 For the photodeposited samples, the s values of gold and platinum (in the absence of citric acid) were 0.61 and 0.28, respectively (Table 1), both of which were below 1, i.e., photodeposition proceeded preferably on {101} facets. Although these selectivities were not 0%-or-100%, the tendency of preferential photodeposition on {101} facets seems consistent with previously reported results. 2.3. Mechanism of Metal Deposition on DAPs and FSR 2.3. Mechanism of Metal Deposition on DAPs and FSR 2.3. Mechanism of Metal Deposition on DAPs and FSR Based on the fact that hydrogen and carbon dioxide production was detected and the fact that the particle size of CDL-deposited platinum was almost half of that of CDD-deposited platinum, oxidative decomposition of surface-covering citric acid to expose the bare metal surface proceeded as a counter reaction of hydrogen evolution by photoexcited electrons. One of the interesting features is that s was less than unity in all cases (PD, CDD and CDL), suggesting that the facet selectivity was governed by the metal-nanoparticle ﬁxation (attachment) process, not the redox process. Another interesting feature is that CDL gave very low s, i.e., high {101} selectivity, which was induced by lowering D{001} and enhancement of D{101} in gold and platinum deposition results, respectively. It seems that these results cannot be interpreted by the conventional mechanism including facet-selective migration of electrons and positive holes if the above-mentioned photocatalytic oxidative decomposition of citric acid on the surface of metal particles induces deposition and no migration of metal particles occurs after the deposition. 2.4. Inﬂuence of pH on Metal Deposition on DAPs Then, what governs the observed facet selectivity? One possible candidate is the surface charge depending on the kind of facets; in all of the deposition methods, PD, CDD and CDL, negatively charged substances, metal-precursor anions or metal colloid particles, are deposited on the surface and such deposition must be inﬂuenced by the surface charge of DAPs. 520 Catalysts 2018, 8, 542 Catalysts 2018, 8, 542 Figure 4 shows representative SEM images of platinum-deposited DAPs obtained by (a) PD and (b) CDL under acidic conditions. Citric acid was chosen for acidiﬁcation of the reaction medium since colloidal metal particles contained this acid as a surface stabilizer. As easily seen in those images, platinum particles were deposited both on {001} and {101} facets and this tendency was evaluated by counting more than 100 DAPs as shown in Table 1 (PD(Pt/CA) and CDL(Pt/CA)); the values of s were greatly increased by the acidiﬁcation to 0.62 and 0.73, respectively, which were almost non-selective. On the other hand, a markedly low s value, high {101} selectivity, was observed when pH of the colloidal suspension was basic (pH = 8.9) by using a higher concentration of platinum colloid (2 wt %) (Table 1). Thus, s depended strongly on pH of suspensions in both PD and CDL platinization. Figure 4. 2.3. Mechanism of Metal Deposition on DAPs and FSR FE-SEM images of metal-deposited DAPs under acidic conditions by: (a) photodeposition; (b) colloid deposition under UV irradiation. Figure 4. FE-SEM images of metal-deposited DAPs under acidic conditions by: (a) photodeposition; (b) colloid deposition under UV irradiation. The effect of pH (acidiﬁcation) is summarized in Figure 5a. The plots for PD and CDL platinization processes in Figure 5a seem to show resemblance, i.e., their pH dependence may originate from the same effect. The drastic decrease in the s value along with pH was caused by the drastic (Note that the D plots are shown in a logarithmic scale.) decrease in D{001}, while D{101} was constant or slightly increased with an increase in pH raise regardless of the procedure, PD or CDL, and deposition amount, 0.5 or 2.0 wt %. Figure 5. (a) pH dependence of facet selectivity (s) and deposition density (D) in PD (open symbols) and CDL (closed symbols) platinum deposition. For deposition-density plots, circles and squares correspond to the density of {001} and {101} facets, respectively. Plots in the dotted oval reﬂect samples deposited with a 2 wt % platinum-colloid solution. (b) Hypothetical pH-dependent zeta-potential curves for {101} and {001} facets. A dashed line is a rough sketch of actual zeta-potential measurement of a DAP sample giving point of zero charge at ca. 7. It should be noted that the DAP sample used in this zeta-potential change was prepared under slightly different reaction conditions. Figure 5. (a) pH dependence of facet selectivity (s) and deposition density (D) in PD (open symbols) and CDL (closed symbols) platinum deposition. For deposition-density plots, circles and squares correspond to the density of {001} and {101} facets, respectively. Plots in the dotted oval reﬂect samples deposited with a 2 wt % platinum-colloid solution. (b) Hypothetical pH-dependent zeta-potential curves for {101} and {001} facets. A dashed line is a rough sketch of actual zeta-potential measurement of a DAP sample giving point of zero charge at ca. 7. It should be noted that the DAP sample used in this zeta-potential change was prepared under slightly different reaction conditions. 521 Catalysts 2018, 8, 542 Catalysts 2018, 8, 542 One of the possible reasons for such pH dependence is pH-dependent and facet-dependent surface charges and, assuming negatively charged species PtCl6−anions or negatively charged platinum colloidal particles are approaching the surface for platinum deposition, changes in surface charge by pH for {101} and {001} facets are different. 2.3. Mechanism of Metal Deposition on DAPs and FSR It is well known that there are appreciable amounts of hydroxyl groups on the surfaces of metal-oxide particles, and protonation/deprotonation depending on pH of a surrounding medium gives protonated (–OH2+), neutral (–OH) and deprotonated (–O−) forms, resulting in the observed zeta-potential curves; the surface of titania particles is positively charged at low pH, decreased by pH increase and then negatively charged at high pH. A hypothesis is that a zeta-potential curve for {001} facets is shifted to the lower-pH side from that of {101} facets as depicted in Figure 5b, being consistent with the actual zeta-potential curve for a DAP sample shown in the ﬁgure as a rough sketch. As has been reported for the facet-selective surface charge for bismuth oxybromide [26], different facets with different surface energies may have different surface charges. The hypothetical zeta-potential curves for {101} and {001} facets can reasonably interpret the observed pH-dependent s and deposition density (D) shown in Figure 5a as follows. At low pH, both {101} and {001} facets are positively charged with protonated surface hydroxyls (–OH2+) to induce attraction of negatively charged precursor anions/colloidal particles and thereby non facet-selective deposition occurs at pH = 2.4–2.5. At neutral pH, the average charge on {001} facets is decreased, while {101} facets are still positively charged to decrease s at pH ca. 7. A further increase in pH leads to negatively charged (–O−) {001} facets to give negligible facet selectivity at pH = 8.9. At higher pH, it is expected that both {101} and {001} facets are negatively charged, resulting in negligible deposition densities. Under such high pH conditions, however, large platinum particles were formed and the number of deposits was very small (data not shown), presumably because precursor small metal particles or colloidal particles were aggregated by neutralization of surface negative charges by sodium cations. Although there has been no experimental evidence for the above-mentioned hypothetical zeta-potential curves due to the lack of a technique for measuring zeta-potentials of each facet on a DAP, the observed facet selectivity in PD and CDL-induced deposition of platinum nanoparticles on DAPs can be consistently explained without taking “charge separation” (FSR) into account. 2.3. Mechanism of Metal Deposition on DAPs and FSR Since there also seems to be no direct evidence for FSR, i.e., speculated only from the position of metal and/or metal oxide deposition, the above-mentioned effect of facet-dependent variation in surface charge may still be a possible reason for the facet-selective (or non-selective) deposition of metals as an alternative of FSR. 2.5. Inﬂuence of Stirring Operation on the Deposition Density of Metal Nanoparticles on DAPs When the above-mentioned interaction between the charged surface (facets) and precursor anions or metal nanoparticles is assumed, the possible detachment of the photodeposited metal nanoparticles should be examined. Figure 6 shows the effect of post-irradiation stirring in the dark on Ds for (a) gold and (b) platinum deposition. Although the plots seemed rather scattered, the densities were decreased by post-irradiation stirring except for platinum deposition on {101} facets. For gold deposition, the densities on both facets were decreased, but it seemed that gold deposits on {001} facets tended to be detached faster than those on {101} facets to result in a lower facet-selectivity value. On the other hand, platinum deposits were more stable than gold deposits and the trend of decrease in D was not obvious for both {101} and {001} facets. The observation of a decrease in the number of deposited metal particles by post-irradiation stirring in the dark indicates that the deposited metal particles can be detached and thereby possibly aggregated to larger particles. 522 Catalysts 2018, 8, 542 Figure 6. Change in deposition density (D) by post-irradiation stirring of PD processes for (a) gold and (b) platinum. Open and closed circles correspond to D of {001} and {101} facets, respectively. Figure 6. Change in deposition density (D) by post-irradiation stirring of PD processes for (a) gold and (b) platinum. Open and closed circles correspond to D of {001} and {101} facets, respectively. Figure 7 shows the change in the particle size distribution of gold nanoparticles deposited on each facet of DAPs with the time of stirring in the dark after PD. Although only ca. 200 gold particles were counted in each distribution, the number of which seems too small for reliable statistical analysis, the shape of the distribution patterns seemed to change with stirring time. However, a change in the average particle size on each facet was not obvious and thereby the detachment of deposited gold particles proceeded almost homogeneously for both {101} and {001} facets. Being consistent with the results showing that {001} facets tend to release gold nanoparticles faster than do {101} facets as shown in Figure 6, the average gold-nanoparticle size on {001} facets was a slightly smaller than that on {101} facets, i.e., {101} facets might be able to keep larger particles than those can be kept on {001} facets. Figure 7. 2.5. Inﬂuence of Stirring Operation on the Deposition Density of Metal Nanoparticles on DAPs Summary of results on PD deposition of metals on DAPs without magnetic stirring. Table 2. Summary of results on PD deposition of metals on DAPs without magnetic stirring. Entry Source Amount 1 (wt %) Medium Size 2/nm pH 3 D{001} 4/10−4 nm−2 D{101} 5/10−4 nm−2 S 6 PD 7(Au) HAuCl4 2.0 MeOH 8 5 — 9 0.28 0.75 0.38 PD 7(Pt) H2PtCl6 2.0 MeOH 8 4 7.2 1.3 3.3 0.39 PD 7(Pt) [Pt(NH3)4]Cl2 2.0 MeOH 8 5 7.4 2.0 3.5 0.58 1 Amount as metal. 2 Roughly estimated average size of metal deposits assuming spherical shape. 3 Measured after deposition. 4 Area deposition density of metal deposits for {001} facets. 5 Area deposition density of metal deposits for {101} facets. 6 Facet selectivity (= D{001}/D{101}). 7 Photodeposition. 8 50 vol % aqueous methanol. 9 Not measured. 1 Amount as metal. 2 Roughly estimated average size of metal deposits assuming spherical shape. 3 Measured after deposition. 4 Area deposition density of metal deposits for {001} facets. 5 Area deposition density of metal deposits for {101} facets. 6 Facet selectivity (= D{001}/D{101}). 7 Photodeposition. 8 50 vol % aqueous methanol. 9 Not measured. Another feature seen in Table 2 is that when a cationic precursor, [Pt(NH3)4]2+, was used, the Ds for platinum were comparable or even higher than those for deposition with an anionic precursor, PtCl62−. This fact suggests that the precursor for deposition which is affected by the surface charge, based on the assumption that D is governed by the surface charge, is small metal particles/clusters, not source metal-complex ions, e.g., PtCl62−or [Pt(NH3)4]2+; in the initial stage of PD, small metal particles/clusters are created followed by migration of these precursors with possible particle growth to be ﬁxed on the surface, and the position of deposition in the second step is regulated by surface charges depending on the kind of facets, {101} or {001}. It should be noted that this proposed mechanism does not exclude the possibility of FSR in the ﬁrst step, and even though facet-selective metal deposition is observed, it seems that this does not prove the occurrence of FSR. 2.5. Inﬂuence of Stirring Operation on the Deposition Density of Metal Nanoparticles on DAPs Change in distribution of gold-particle size on (a) {001} and (b) {101} facets by stirring in the dark for (top) 0 h, 2 h, 13 h, 24 h and (bottom) 48 h. Frequency was standardized to be 100% for the 0-h samples and summation of distribution is proportional to the number of remaining gold nanoparticles. Figures in square brackets show average particle size in the unit of nm. Figure 7. Change in distribution of gold-particle size on (a) {001} and (b) {101} facets by stirring in the dark for (top) 0 h, 2 h, 13 h, 24 h and (bottom) 48 h. Frequency was standardized to be 100% for the 0-h samples and summation of distribution is proportional to the number of remaining gold nanoparticles. Figures in square brackets show average particle size in the unit of nm. 523 Catalysts 2018, 8, 542 Thus, the results suggested that the distribution of metal nanoparticles on {001} and {101} facets depends on the size and that metal particles that have been deposited can be detached during the process of PD. 2.6. Photodeposition of Metal Particles on DAPs under No Stirring Conditions Since, as described in the preceding section, metal particles deposited in PD processes can be detached from the DAP surface, further experiments using PD without magnetic stirring were performed to reduce the detachment, and the results are summarized in Table 2. In these experiments, DAPs were ﬁxed on a glass plate and irradiated in aqueous methanol containing metal complexes (2 wt % as metal) (See Materials and Methods). Ds were evaluated neglecting DAPs without any metal deposits because unlike PD under the condition of magnetic stirring, only the surface layer of the DAP ﬁlm absorbs light for metal deposition. Even though such modiﬁcation in the D evaluation scheme was adopted and a higher concentration of the metal source was used, the actual densities were comparable or even lower than those of PD with magnetic stirring (Table 1). In all of the cases shown in Table 2, s was in the middle range, 0.4–0.6, i.e., ambiguous facet selectivity. Although it is difﬁcult to compare the facet selectivity with that obtained for deposition with magnetic stirring (Table 1) and although facet selectivity for gold deposition without stirring was even decreased, the above-mentioned ambiguous facet selectivity again suggested less probable FSR. Table 2. 3.1. Preparation and Characterization of DAP Samples DAP samples were prepared by a gas-phase reaction of titanium (IV) chloride (TiCl4; Wako, Tokyo, Japan) and oxygen (O2) using a coaxial-ﬂow gas-phase reactor as reported previously (modiﬁed from the original procedure [32]). A brief description of the procedure is as follows. An argon (Ar; >99.99%; puriﬁed by a Shimadzu (Kyoto, Japan) GLC Click-on Triple (hydrocarbon, oxygen and moisture) trap) stream (100 mL min−1, 453 K) containing 1 vol % TiCl4 (Wako, Tokyo, Japan)(quantitatively introduced by a syringe feeder) vapor and an O2 stream (>99.5%, dried and puriﬁed by a Shimadzu (Kyoto, Japan) GLC Click-on Combi (hydrocarbon and moisture) trap; 800 mL min−1) are introduced into a quartz 524 Catalysts 2018, 8, 542 reactor tube as inside and outside, respectively, coaxial ﬂow and heated from platinum foil (3.0 cm), wrapped around the quartz reactor tube, the temperature of which is kept at 1473 K by infrared lamps (Advance Riko VHT-E44, Yokohama., Japan; totally 2 kW maximum). The preheated outside O2 ﬂow is heated by the 1473-K wrapped platinum foil prior to the inside TiCl4/Ar stream and expanded toward the center to react with TiCl4 (TiCl4 + O2 →TiO2 + 2Cl2). The resultant white titania smoke ﬂows at the center of the reactor tube (see Figure 1) and is collected by a glass-ﬁber ﬁlter thimble (Whatman high-purity glass microﬁber extraction thimble, 25 mm × 90 mm, Tokyo, Japan). The white product is washed with water ﬁve times to remove possibly adsorbed chlorine and then freeze-dried (EYELA FDU-2100, Tokyo, Japan) under vacuum (<10 Pa) for 24 h. reactor tube as inside and outside, respectively, coaxial ﬂow and heated from platinum foil (3.0 cm), wrapped around the quartz reactor tube, the temperature of which is kept at 1473 K by infrared lamps (Advance Riko VHT-E44, Yokohama., Japan; totally 2 kW maximum). The preheated outside O2 ﬂow is heated by the 1473-K wrapped platinum foil prior to the inside TiCl4/Ar stream and expanded toward the center to react with TiCl4 (TiCl4 + O2 →TiO2 + 2Cl2). The resultant white titania smoke ﬂows at the center of the reactor tube (see Figure 1) and is collected by a glass-ﬁber ﬁlter thimble (Whatman high-purity glass microﬁber extraction thimble, 25 mm × 90 mm, Tokyo, Japan). The white product is washed with water ﬁve times to remove possibly adsorbed chlorine and then freeze-dried (EYELA FDU-2100, Tokyo, Japan) under vacuum (<10 Pa) for 24 h. 3.1. Preparation and Characterization of DAP Samples y p The DAP samples were characterized by X-ray diffractometry (XRD) and scanning electron microscopy (SEM) with a Rigaku SmartLab X-ray diffractometer with CuKα radiation (40 kV, 30 mA, Rigaku, Akishima, Japan) and a JEOL JSM-7400F microscope (JEOL, Akishima, Japan), respectively. The details of XRD measurements are as follows: A DAP sample and 20 wt % nickel oxide (NiO; Wako, Tokyo, Japan) as an internal crystalline standard [33] were mixed thoroughly in an agate mortar, and the XRD pattern of the mixture was recorded with a scanning rate of 1.0◦min−1 and steps of 0.008◦in the 2θ range of 10–90◦. Recorded diffractograms were analyzed using the software PDXL 2 (Version 2.6.1.2, Rigaku, Akishima, Japan) including a RIETAN-FP Rietveld analysis package [33]. Crystallite size, i.e., primary particle size, was estimated by the Scherrer equation with corrected average peak width of anatase 101, 004 and 200 peaks at 2θ of ca. 25.4◦, 37.8◦and 48.0◦, respectively. The detailed conditions and procedure for SEM analysis of the DAP samples are described in the following sections. 3.2. Photodeposition of Metal Nanoparticles on DAP Samples from Metal Complexes 3.2. Photodeposition of Metal Nanoparticles on DAP Samples from Metal Complexes In the process of PD of platinum and gold, a 30-mL solution of 50 vol % aqueous methanol (Wako, Tokyo, Japan) containing hydrogen hexachloroplatinum(IV) (H2PtCl6, Wako, Tokyo, Japan) or aqueous hydrogen tetrachlorogold(III) (HAuCl4, Wako, Tokyo, Japan) (0.5 wt % (or 2.0 wt %) as metal) was poured in a glass tube containing 0.015 g of DAP. In some experiments, tetraammineplatinum (II) chloride ([Pt(NH3)4]Cl2, Wako, Tokyo, Japan) was used instead of H2PtCl6, and citric acid (Wako, Tokyo, Japan; 0.1 mol L−1 in a suspension) was added to acidify the suspension. The suspension was sonicated to be homogenized, deaerated by argon bubbling, and then irradiated by a 400-W mercury arc (>290 nm; Eiko-sha 400) with vigorous magnetic stirring at 1000 rpm. After 15-min irradiation and 2-h irradiation for platinum and gold, respectively, pH of the suspension was measured using a pH meter (Horiba pH meter LAQUA twin, Kyoto, Japan) and the powder was recovered by centrifugation, washed three times with Milli-Q water, and freeze-dried under vacuum (<10 Pa) for 24 h. 3.3. Photodeposition of Metal Nanoparticles on DAP Samples from Metal Colloids 3.3. Photodeposition of Metal Nanoparticles on DAP Samples from Metal Colloids Platinum and gold colloid solutions were prepared following the reported procedures for platinum [34] and gold [35], respectively. For platinum colloid, a 196-mL portion of an aqueous H2PtCl6 solution (0.30 mmol L−1) was heated to be reﬂuxed by a mantle heater (MS-ES-3, As one, Osaka, Japan) under magnetic stirring. Then 4.0 mL of aqueous sodium-citrate solution (0.84 mol L−1) was added and the reaction mixture was kept boiling for 45 min. After being cooled down rapidly in an ice bath, excess citric acid and inorganic salts in the resultant colloidal solutions were removed by being passed through an ion exchange resin (Organo Amberlite MB-1, Tokyo, Japan)-packed column. For gold colloid, a 202-mL portion of an aqueous HAuCl4 solution (1.0 mmol L−1) was heated to be reﬂuxed, and then 24 mL of sodium-citrate solution (0.039 mol L−1) was added followed by maintenance of reﬂux for 30 min. The workup procedure was the same as that for the above-mentioned platinum colloid preparation. Deposition of platinum nanoparticles and deposition of gold nanoparticles under CDL or CDD were performed using 0.015 g and 0.044 g, respectively, of DAP suspended in a 5.0-mL colloid solution containing the required amount of the metal. The reaction mixture was kept at 298 K with magnetic 525 Catalysts 2018, 8, 542 stirring under photoirradiation, with the same setup as that for PD, or in the dark. The resultant powder was recovered by centrifugation, washed three times with Milli-Q water, and freeze-drying for 24 h. 3.4. Photodeposition of Metal Nanoparticles on DAP Samples without Agitation A 0.20-mL portion of a sonicated DAP suspension (10 mg mL−1) was poured onto a glass plate and dried in vacuum at ambient temperature for 24 h. The DAP-coated glass plate was immersed in 50 vol % aqueous methanol containing a metal complex, H2PtCl6, HAuCl4 or [Pt(NH3)4]Cl2 (2 wt % as metal), and irradiated by a mercury arc at >290 nm for 2 h. After metal deposition, the DAP-coated glass plate was dried at room condition without washing procedure. 3.5. Evaluation of Deposition Densities and Facet Selectivity of Metal Deposition The metal nanoparticle-deposited DAPs were analyzed by electron microscopy using a ﬁeld emission-type scanning electron microscope (FE-SEM; JEOL JSM-7400M, Yokohama, Japan) in a mode of secondary electron image (SEI) with operating conditions of 5.0–10.0-kV electron-acceleration voltage, 10.0-μA current and 3–6-mm working distance. 4. Conclusions As described above, metal-nanoparticle deposition was reexamined using DAPs synthesized by gas-phase reaction of titanium (IV) chloride and oxygen in the absence of a so-called SCA to obtain the following three signiﬁcant aspects. One is that the frequently reported almost perfect (0%-or-100%) facet selectivity for photocatalytic deposition, PD, i.e., reductive metal deposition and oxidative metal-oxide deposition on {101} and {001} facets, respectively, due to FSR could not be observed, at least for the DAP samples used in this study, though there seemed to be a tendency of {101}-selective deposition. The difference from previously reported results may be due to (i) the conclusion of 0%-or-100% selectivity in previous studies by using only one or a few microscopic images matching the FSR concept or (ii) an appreciable difference in the DAP surface structures, e.g., our samples being inﬂuenced by the negligibly remaining chlorine or the surfaces of previous samples prepared through liquid-phase processes being covered by an SCA or the others. The second aspect is that the change in the observed facet selectivity with different reaction conditions was similar to that in deposition of metals, gold and platinum, from CDD or CDL. The FSR concept cannot be applied to CDD since photoexcitation of titania is not induced in this process. Although the detailed mechanism, at least why metal deposition from colloids was enhanced by photoirradiation, has not yet been clariﬁed, the colloid-stabilizing agent citric acid was decomposed along with hydrogen and carbon dioxide evolution and thereby oxidative decomposition/removal of citric acid covering colloidal metal particles may lead to deposition. In such a case, oxidative deposition of metal nanoparticles should be observed on {001} facets, not the actually observed {101} facets, according to the FSR concept. The third aspect is pH-dependent change in facet selectivity of platinum-nanoparticle deposition in both PD and CDL processes; {101}-preferable facet selectivity in neutral and basic pH conditions became ambiguous at low pH. Based on the assumption that {001} facets are more acidic, i.e., easily releasing protons to bear negative surface charges, than are {101} facets and that metal nanoparticles are created by the assembly of small atomic or cluster-sized metal precursors, which migrate on the surface in detachment and re-attachment cycles, it is thought that the difference in surface charges depending on the kind of facets and deposition conditions, e.g., pH, accounts for the observed facet selectivity in all of the PD, CDD and CDL processes. Acknowledgments: FE-SEM analyses of samples were carried out using a JEOL JSM-7400F electron microscope at Global Facility Center, Creative Research Institution, Hokkaido University Technical assistance and support for construction and maintenance of the gas-phase reactor for DAP synthesis and the other instrumental setups by the Technical Division of Institute for Catalysis, Hokkaido University are acknowledged. 3.3. Photodeposition of Metal Nanoparticles on DAP Samples from Metal Colloids Evaluation of D and s was performed by counting the number, not volume, of metal deposits per unit area in FE-SEM images as follows. First, the number of deposited metal nanoparticles (N{001} and N{101}) and total area of deposited facets (S{001} and S{101}) were measured using several SEM images for {001} and {101} facets, respectively. In order to keep statistical reliability and reproducibility, more than 100 metal-deposited DAPs were counted. For the area measurement, the following equations were used to estimate the area of each facet, S{001} and S{101}, with the measured lengths of two ridges of a DAP, long (a) and short (b) sides of a {101} trapezoid (Figure 8a). Then the Ds on each facet, D{001} and D{101}, were obtained as N/S, and s was calculated as D{001}/D{101}. Values of s of more than 1, 1 and less than 1 mean {001} selective, non selective and {101} selective, respectively. S{001} = b2 (1) S{001} = b2 (1) S{101} = (a + b) × h/2 = (a + b) × (a −b) × tan 69.7◦/4 (2) (2) (a) (b) Figure 8. (a) Assumed dimension of a DAP and (b) surface area-volume (SV) ratio as a function of oblateness, b/a, of an ideal {101} trapezoid. The grey part corresponds to the oblateness range giving an SV ratio within excess 10% of the minimum value (8.2 nm−1). The surface area-volume (SV) ratio, Stotal/V, of a DAP was calculated using the dimension shown Figure 8b. (b) (a) (b) (a) (b) Figure 8. (a) Assumed dimension of a DAP and (b) surface area-volume (SV) ratio as a function of oblateness, b/a, of an ideal {101} trapezoid. The grey part corresponds to the oblateness range giving an SV ratio within excess 10% of the minimum value (8.2 nm−1). The surface area-volume (SV) ratio, Stotal/V, of a DAP was calculated using the dimension shown in Figure 8b. The surface area-volume (SV) ratio, Stotal/V, of a DAP was calculated using the dimension shown in Figure 8b. 526 Catalysts 2018, 8, 542 Catalysts 2018, 8, 542 Funding: This research was partly supported by a Grant-in-Aid for Scientiﬁc Research (A) (Grant Numb 15H0220106 and 18H0392308) from Japan Society for the Promotion of Science (JSPS). Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Ohtani, B. Preparing Articles on Photocatalysis—Beyond the Illusions, Misconceptions, and Speculation. Chem. Lett. 2008, 37, 216–229. [CrossRef] Chem. Lett. 2008, 37, 216–229. [CrossRef] 2. Ohtani, B. Revisiting the fundamental physical chemistry in heterogeneous photocatalysis: Its thermodynam and kinetics Phys Chem Chem Phys 2014 16 1788 1797 [CrossRef] [PubMed] 2. Ohtani, B. Revisiting the fundamental physical chemistry in heterogeneous photocatalysis: Its thermodynamics and kinetics. Phys. Chem. Chem. Phys. 2014, 16, 1788–1797. [CrossRef] [PubMed] netics. Phys. Chem. Chem. Phys. 2014, 16, 1788–1797. [Cro 3. Ohtani, B. Photocatalysis A to Z—What we know and what w Photobiol. C Photochem. Rev. 2010, 11, 157–178. [CrossRef] 3. Ohtani, B. Photocatalysis A to Z—What we know and what we do not know in a scientiﬁc sense. J. P 3. Ohtani, B. Photocatalysis A to Z—What we know and what we do not know in a scientiﬁc sense. J. Photochem. Photobiol. C Photochem. Rev. 2010, 11, 157–178. [CrossRef] 3. Ohtani, B. Photocatalysis A to Z What we know and what we do not know in a scientiﬁc sense. J. Photochem. Photobiol. C Photochem. Rev. 2010, 11, 157–178. [CrossRef] y Photobiol. C Photochem. Rev. 2010, 11, 157–178. [CrossRef] 4. Ohtani, B. Titania Photocatalysis beyond Recombination: A Critical Review. Catalysts 2013, 3, 942–953. [CrossRef] 5. Banﬁeld, J.F.; Veblen, D.R.; Smith, D.J. The identiﬁcation of naturally occurring TiO2(B) by structure determination using high-resolution electron microscopy, image simulation, and distance-least-squares reﬁnement. Am. Mineralogist 1991, 76, 343–353. 6. Banﬁeld, J.F.; Bischoff, B.L.; Anderson, M.A. TiO2 accessory minerals: Coarsening, and transformation kinetics in pure and doped synthetic nanocrystalline materials. Chem. Geol. 1993, 110, 211–231. [CrossRef] 7. Amano, F.; Yasumoto, T.; Prieto-Mahaney, O.-O.; Uchida, S.; Shibayama, T.; Ohtani, B. Photocatalytic activity of octahedral single-crystalline mesoparticles of anatase titanium (IV) oxide. Chem. Comm. 2009, 2311–2313. [CrossRef] [PubMed] 8. Amano, F.; Yasumoto, T.; Prieto-Mahaney, O.-O.; Uchida, S.; Shibayama, T.; Terada, Y.; Ohtani, B. Highly Active Titania Photocatalyst Particles of Controlled Crystal Phase, Size, and Polyhedral Shapes. Top. Catal. 2010, 53, 455–461. [CrossRef] 9. Ohtani, B.; Iwai, K.; Nishimoto, S.-I.; Sato, S. Role of Platinum Deposits on Titanium (IV) Oxide Particles: Structural and Kinetic Analyses of Photocatalytic Reaction in Aqueous Alcohol and Amino Acid Solutions. J. Phys. Chem. B 1997, 101, 3349–3359. [CrossRef] 10. Yang, H.G.; Sun, C.H.; Qiao, S.Z.; Zou, J.; Liu, G.; Smith, S.C.; Cheng, H.M.; Lu, G.Q. Anatase TiO2 single crystals with a large percentage of reactive facets. Nature 2008, 453, 638. 4. Conclusions On the basis of these aspects, it can be concluded that the concept of FSR, facet-selective reaction of photoexcited electrons and positive holes, does not seem to be necessary to explain the change in facet selectivity observed in this study and that the results can be consistently interpreted by the possible surface charges depending on the kind of facets, {001} and {101}. Since the procedure and conditions for preparation of DAPs in this study were actually different from those used in previous studies, there might be another mechanism for the reported facet selectivity. Furthermore, the results of this study, non 0%-or-100% facet selectivity, do not suggest that only 0%-or-100% selectivity results were chosen in previous studies since possibly different surface structures of DAPs used in previous studies might have led to 0%-or-100% selectivity. However, the present authors propose here that the above-mentioned surface charge-dependent deposition of platinum and gold nanoparticles can explain the results shown in this article and may be expanded to the previously reported results. Author Contributions: All authors: draft preparation, writing and editing. K.K.: performing all of the experiments and data analyses. M.T. (Mai Takase) and B.O.: design, maintenance and instruction of the gas-phase reaction for preparation of DAPs. M.T. (Mai Takashima): project management and maintenance of FE-SEM. B.O.: project leading and fund acquisition. Funding: This research was partly supported by a Grant-in-Aid for Scientiﬁc Research (A) (Grant Numbers: 15H0220106 and 18H0392308) from Japan Society for the Promotion of Science (JSPS). Acknowledgments: FE-SEM analyses of samples were carried out using a JEOL JSM-7400F electron microscope at Global Facility Center, Creative Research Institution, Hokkaido University Technical assistance and support for construction and maintenance of the gas-phase reactor for DAP synthesis and the other instrumental setups by the Technical Division of Institute for Catalysis, Hokkaido University are acknowledged. 527 Catalysts 2018, 8, 542 References [CrossRef] [PubMed] 11. Amano, F.; Prieto-Mahaney, O.-O.; Terada, Y.; Yasumoto, T.; Shibayama, T.; Ohtani, B. Decahedral Single-Crystalline Particles of Anatase Titanium (IV) Oxide with High Photocatalytic Activity. Chem. Mater. 2009, 21, 2601–2603. [CrossRef] 12. Ye, L.Q.; Liu, J.Y.; Tian, L.H.; Peng, T.Y.; Zan, L. The replacement of {101} by {010} facets inhibits the photocatalytic activity of anatase TiO2. Appl. Catal. B. Environ. 2013, 134, 60–65. [CrossRef] 13. Zheng, Z.; Huang, B.; Lu, J.; Qin, X.; Zhang, X.; Dai, Y. Hierarchical TiO2 Microspheres: Synergetic Effect of {001} and {101} Facets for Enhanced Photocatalytic Activity. Chem. Eur. J. 2011, 17, 15032–15038. [CrossRef] [PubMed] 14. Meng, A.; Zhang, J.; Xu, D.; Cheng, B.; Yu, J. Enhanced photocatalytic H2-production activity of anatase TiO2 nanosheet by selectively depositing dual-cocatalysts on {101} and {001} facets. Appl. Catal. B. Environ. 2016, 198, 286–294. [CrossRef] 15. Sun, D.; Yang, W.; Zhou, L.; Sun, W.; Li, Q.; Shang, J.K. The selective deposition of silver nanoparticles onto {101} facets of TiO2 nanocrystals with co-exposed {001}/{101} facets, and their enhanced photocatalytic reduction of aqueous nitrate under simulated solar illumination. Appl. Catal. B. Environ. 2016, 182, 85–93. [CrossRef] 16. Murakami, N.; Kurihara, Y.; Tsubota, T.; Ohno, T. Shape-Controlled Anatase Titanium (IV) Oxide Particles Prepared by Hydrothermal Treatment of Peroxo Titanic Acid in the Presence of Polyvinyl Alcohol. J. Phys. Chem. C 2009, 113, 3062–3069. [CrossRef] 17. Chamtouri, M.; Kenens, B.; Aubert, R.; Lu, G.; Inose, T.; Fujita, Y.; Masuhara, A.; Hofkens, J.; Uji-i, H. Facet-Dependent Diol-Induced Density of States of Anatase TiO2 Crystal Surface. ACS Omega 2017, 2, 4032–4038. [CrossRef] [PubMed] 18. Liu, C.; Han, X.; Xie, S.; Kuang, Q.; Wang, X.; Jin, M.; Xie, Z.; Zheng, L. Enhancing the Photocatalytic Activity of Anatase TiO2 by Improving the Speciﬁc Facet-Induced Spontaneous Separation of Photogenerated Electrons and Holes. Chem. Asian. J. 2012, 8, 282–289. [CrossRef] [PubMed] 528 Catalysts 2018, 8, 542 19. Wang, W.; Lai, M.; Fang, J.; Lu, C. Au and Pt selectively deposited on {001}-faceted TiO2 toward SPR enhanced photocatalytic Cr(VI) reduction: The inﬂuence of excitation wavelength. Appl. Surf. Sci. 2018, 439, 430–438. [CrossRef] 20. Murakami, N.; Kawakami, S.; Tsubota, T.; Ohno, T. Dependence of photocatalytic activity on particle size of a shape-controlled anatase titanium (IV) oxide nanocrystal. J. Mol. Catal. A. Chem. 2012, 358, 106–111. [CrossRef] 21. Xiong, Z.; Lei, Z.; Chen, X.; Gong, B.; Zhao, Y.; Zhang, J.; Zheng, C.; Wu, J.C.S. References CO2 photocatalytic reduction over Pt deposited TiO2 nanocrystals with coexposed {101} and {001} facets: Effect of deposition method and Pt precursors. Catal. Commun. 2017, 96, 1–5. [CrossRef] 22. Jiang, Z.; Ding, D.; Wang, L.; Zhang, Y.; Zan, L. Interfacial effects of MnOx-loaded TiO2 with exposed {001} facets and its catalytic activity for the photoreduction of CO2. Catal. Sci. Technol. 2017, 7, 3065–3072. [CrossRef] 23. Ohno, T.; Sarukawa, K.; Matsumura, M. Crystal faces of rutile and anatase TiO2 particles and their roles in photocatalytic reactions. New. J. Chem. 2002, 26, 1167–1170. [CrossRef] 24. Li, R.; Zhang, F.; Wang, D.; Yang, J.; Li, M.; Zhu, J.; Zhou, X.; Han, H.; Li, C. Spatial separation of photogenerated electrons and holes among {010} and {110} crystal facets of BiVO4. Nat. Commun. 2013, 4, 1432. [CrossRef] [PubMed] 25. Zhou, C.; Wang, S.; Zhao, Z.; Shi, Z.; Yan, S.; Zou, Z. A Facet-Dependent Schottky-Junction Electron Shuttle in a BiVO4{010}–Au–Cu2O Z-Scheme Photocatalyst for Efﬁcient Charge Separation. Adv. Funct. Mater. 2018, 28, 1801214. [CrossRef] 26. Guo, Y.; Siretanu, I.; Zhang, Y.; Mei, B.; Li, X.; Mugele, F.; Huang, H.; Mul, G. pH-Dependence in facet-selective photo-deposition of metals and metal oxides on semiconductor particles. J. Mater. Chem. A 2018, 6, 7500–7508. [CrossRef] 27. Zhang, Q.; Li, R.; Li, Z.; Li, A.; Wang, S.; Liang, Z.; Liao, S.; Li, C. The dependence of photocatalytic activity on the selective and nonselective deposition of noble metal cocatalysts on the facets of rutile TiO2. J. Catal. 2016, 337, 36–44. [CrossRef] 28. Wenderich, K.; Klaassen, A.; Siretanu, I.; Mugele, F.; Mul, G. Sorption-Determined Deposition of Platinum on Well-Deﬁned Platelike WO3. Angew. Chem. Int. Edit. 2014, 53, 12476–12479. [CrossRef] 29. Ohtani, B.; Bowman, R.M.; Colombo, D.P., Jr.; Kominami, H.; Noguchi, H.; Uosaki, K. Femtosecond Diffuse Reﬂectance Spectroscopy of Aqueous Titanium (IV) Oxide Suspension: Correlation of Electron-Hole Recombination Kinetics with Photocatalytic Activity. Chem. Lett. 1998, 27, 579–580. [CrossRef] 30. Lazzeri, M.; Vittadini, A.; Selloni, A. Structure and energetics of stoichiometric TiO2 anatase surfaces. Phys. Rev. B 2001, 63, 155409. [CrossRef] 31. Lazzeri, M.; Vittadini, A.; Selloni, A. Erratum: Structure and energetics of stoichiometric TiO2 anatase surfaces. Phys. Rev. B 2002, 65, 119901. [CrossRef] 32. Janczarek, M.; Kowalska, E.; Ohtani, B. Decahedral-shaped anatase titania photocatalyst particles: Synthesis in a newly developed coaxial-ﬂow gas-phase reactor. Chem. Eng. J. 2016, 289, 502–512. [CrossRef] 33. Wei, Z.; Kowalska, E.; Ohtani, B. Haﬁze Nagehan Koysuren Department of Environmental Engineering, Ahi Evran University, Kirsehir 40100, Turkey; hnkoysuren@gmail.com; Tel.: +90-386-280-3855 Received: 28 September 2018; Accepted: 24 October 2018; Published: 26 October 2018 Department of Environmental Engineering, Ahi Evran University, Kirsehir 40100, Turkey; hnkoysuren@gmail.com; Tel.: +90-386-280-3855 Abstract: In this study, polymer composites based on polyvinyl borate (PVB) with titanium dioxide (TiO2) nanoparticles were prepared through the condensation reaction of polyvinyl alcohol and boric acid in the presence of TiO2 nanoparticles. The solid-phase photocatalytic degradation of the polymer composites under UV light irradiation was investigated and compared with that of the pure PVB with the aid of weight loss measurements. The introduction of the photocatalyst nanoparticles in PVB enhanced the solid-phase photocatalytic degradation of the polymer matrix under UV light irradiation. The structural and morphological properties of PVB/TiO2 composites were analyzed by transmission electron microscopy (TEM), scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), thermogravimetric analysis (TGA), and UV-Vis spectroscopy, respectively. FTIR analysis revealed that PVB synthesis was successfully carried out in the presence of the photocatalyst nanoparticles. According to the morphological analyses, TiO2 nanoparticles were well dispersed in the PVB matrix. Keywords: solid-phase photocatalytic degradation; polyvinyl borate; titanium dioxide Catalysts 2018, 8, 499; doi:10.3390/catal8110499 www.mdpi.com/journal/catalysts References Inﬂuence of Post-Treatment Operations on Structural Properties and Photocatalytic Activity of Octahedral Anatase Titania Particles Prepared by an Ultrasonication-Hydrothermal Reaction. Molecules 2014, 19, 19573–19587. [CrossRef] [PubMed] 34. Kajita, M.; Kuwabara, T.; Hasegawa, D.; Yagi, M. Element-saving preparation of an efﬁcient electrode catalyst based on self-assembly of Pt colloid nanoparticles onto an ITO electrode. Green Chem. 2010, 12, 2150–2152. [CrossRef] 35. Frens, G. Controlled Nucleation for Regulation of Particle-Size in Monodisperse Gold Suspensions. Nature 1973, 20–22. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 529 1. Introduction With the speeding up of the industrial process based on the plastic technology, a large amount of plastic waste is directly discharged into the nature, inducing the accumulation of large amounts of toxic organic compounds in our daily lives and bringing with it an enormous threat to human health [1]. Most of the plastic based organic pollutants are burned for disposal, producing toxic gases. Recycling is applied to some plastics, but the recycling cost is high and requires high technology equipment [2]. The degradation of the plastic waste has attracted more attention in recent years and most of the studies have focused mainly on biodegradation. However, certain types of plastic materials are biodegradable under both aerobic and anaerobic conditions and the degradation rate of the most used plastics under natural environmental conditions is too low to apply in practice [3]. Most of the plastics are very difﬁcult to degrade by the ordinary processing methods [1]. The solid-phase photocatalytic process is an ideal method to degrade plastic waste by using the inexhaustible solar light energy [1]. The solid-phase photocatalytic degradation process is based on solar energy corresponding to the band energy of the photocatalyst and the following photo-generated electron transfer. There are many kinds of semiconductor photocatalysts and titanium dioxide (TiO2) is the most commonly used among them because of its high photostability, non-toxicity, low cost, and high activity [4]. TiO2 as a photocatalyst has attracted great attention in various ﬁelds like air and waste water treatment, hydrogen fuel production, metal anti-corrosion, antibacterial activity, and self-puriﬁcation. TiO2 and other semiconductor atoms possess a valence band, which is occupied with stable energy electrons, and a conduction band, which is empty. The band gap energy, which is the energy difference between the top energy state of the valence band and the bottom energy state of the conduction band in semiconductors, is utilized to emit light inside TiO2 to induce a redox reaction on its surface, which is 531 Catalysts 2018, 8, 499 known as the photocatalytic reaction [5]. When photons of sun light with energy greater than the band gap of TiO2 are absorbed by TiO2 semiconductor, one electron from the valance band rises into the conduction band, generating a photoinduced electron-hole pair. This electron-hole pair transfers to the photocatalyst surface, where it reacts with the surface-absorbed molecules like water and oxygen to form active radicals. 1. Introduction These active radicals can degrade the plastic based organic pollutants through oxidation reactions [2]. Plastic waste is discharged into nature, where it is directly subjected to sunlight in open air. Therefore, it is noteworthy to study the solid-phase photocatalytic degradation of plastics under the sunlight. The significance of immobilization of the TiO2 photocatalyst in polymer matrix for the solid-phase photocatalytic degradation of plastics is highlighted by the recent studies. In literature, the photocatalytic degradation of polystyrene [6,7], polyaniline [8], polyvinyl chloride [9,10], low-density polyethylene [2], polyvinyl alcohol [3], and poly (methyl methacrylate) [11] was investigated in the presence of TiO2 nanoparticles. The solid-phase photocatalytic degradation rate of plastics including TiO2 nanoparticles was much faster than the simple photolysis of pure plastics. Polyvinyl borate is a synthetic polymer prepared through the condensation reaction of polyvinyl alcohol and boric acid. Polyvinyl alcohol is crosslinked with boric acid to improve the thermal and mechanical properties of the polymer. In addition, polyvinyl alcohol is a water-soluble polymer and crosslinking the polymer with boric acid enhances the moisture resistance of the matrix [12]. A few studies have been performed on crosslinking of polyvinyl alcohol with boric [4,12–14]. In contrast to polyvinyl alcohol, PVB is a non-environmental polymer and it degradation is also signiﬁcant in terms of human health and the environment. There is no study in the literature on the solid-phase photocatalytic degradation of PVB. In this study, PVB composites were prepared through the condensation reaction of polyvinyl alcohol and boric acid in the presence of TiO2 nanoparticles. The photocatalytic activity of PVB/TiO2 composites was evaluated by investigating the solid-phase photocatalytic degradation of the polymer matrix in the ambient air under ultraviolet light irradiation. 2. Results and Discussion Figure 1 illustrates the FTIR spectra of pure PVA and PVB/TiO2 composite, including 10 wt.% of the photocatalyst nanoparticles. The broad absorption band around 3200 cm−1 was related to the O–H group of polyvinyl alcohol, which formed complexes with boron-containing oxyanions during the crosslinking reactions [13]. On the other hand, this broad band could also be related to the stretching band of O–H, which might be due to unreacted O–H groups of polyvinyl alcohol (Figure 1a) [14]. The absorption peaks at 2923 cm−1 and 1338 cm−1 are related to the stretching bond of C–H (Figure 1a) and the peak at 1724 cm−1 was assigned to the stretching bond of C=O (Figure 1a) [14]. The absorption peaks at 1299 cm−1 and 1133 cm−1 were assigned to the stretching vibrations of B–O–C bonds. This absorption peak provided strong evidence for the condensation reaction between polyvinyl alcohol and boric acid to synthesize PVB (Figure 1a) [14]. In addition, the peak at 1430 cm−1 was attributed to the stretching vibration of the B–O bond, which might be due to the unreacted boric acid (Figure 1a). Figure 1b illustrates FTIR spectrum of the composite. According to Figure 1a,b, the spectrum of the composite matched with that of the pure polymer. Characteristic absorption peaks of PVB, the stretching vibrations of B–O–C bonds, were also observed on the spectrum of the composite (Figure 1b). Different from the spectrum of pure PVB, there was a broad absorption band between 600 cm−1 and 900 cm−1, which was attributed to Ti–O stretching vibrations (Figure 1b) [15]. This absorption band proved the presence of TiO2 nanoparticles in PVB matrix. The small absorption peak of pure PVB, present at around 1600 cm−1 and related to physically absorbed moisture, could also be seen on the spectrum of the composite (Figure 1a,b) [15]. This peak intensity became wider and increased. Hence, TiO2 contribution might increase the moisture content of the composite. 532 Catalysts 2018, 8, 499 Figure 1. Fourier transform infrared spectroscopy (FTIR) spectra of (a) pure polyvinyl borate (PVB) and (b) PVB/TiO2 (10 wt.%) composite. Figure 1. Fourier transform infrared spectroscopy (FTIR) spectra of (a) pure polyvinyl borate (PVB) and (b) PVB/TiO2 (10 wt.%) composite. Figure 2 shows TGA curves of pure polymer and the composite, containing 10 wt.% of the photocatalyst nanoparticles. The TGA curve of the pure polymer contains two main degradation steps between 300–400 ◦C. 2. Results and Discussion One of these degradation steps is the large step, corresponding to deacetylation reactions, and the other is the small step, corresponding to chain scission reactions [12]. TGA curves revealed that the photocatalyst nanoparticles slightly enhanced the thermal stability of PVB. The onset temperature of degradation is higher than that of pure PVB. Pure PVB and PVB/TiO2 composite exhibited total weight losses of 42% and 39%, respectively, in the temperature range between 250 ◦C and 450 ◦C. Figure 2. Thermogravimetric analysis (TGA) curves of (a) pure PVB and (b) PVB/TiO2 (10 wt.%) composite. Figure 2. Thermogravimetric analysis (TGA) curves of (a) pure PVB and (b) PVB/TiO2 (10 wt.%) composite. SEM analyses of the composites, containing 5 wt.%, 10 wt.%, and 15 wt.% of TiO2, respectively, were carried out to visualize the distribution of the photocatalyst nanoparticles in PVB matrix. Figure 3 illustrates SEM images of the composites. TiO2 aggregates, which consisted of hundreds of nanoparticles, could be seen on the composite surfaces, which suggested that TiO2 nanoparticles tended to agglomerate. During the synthesis of the composite, insufﬁcient mixing might be performed to lead to a partial aggregated morphology. TEM images supported the stated thought (Figure 4). TiO2 aggregates were well distributed in the polymer matrix and the size of most aggregates was less than 500 nm. According to SEM and TEM images, the polymer matrix held TiO2 nanoparticles in intimate contact form, which is also important for enhanced photocatalytic activity. 533 Catalysts 2018, 8, 499 Figure 3. Scanning electron microscopy (SEM) images of (a) PVB/TiO2 (5 wt.%) composi (b) PVB/TiO2 (10 wt.%) composite and (c) PVB/TiO2 (15 wt.%) composite. Figure 3. Scanning electron microscopy (SEM) images of (a) PVB/TiO2 (5 wt.%) composite, (b) PVB/TiO2 (10 wt.%) composite and (c) PVB/TiO2 (15 wt.%) composite. 534 Catalysts 2018, 8, 499 Figure 4. Transmission electron microscopy (TEM) images of PVB/TiO2 (10 wt.%) composite. Figure 4. Transmission electron microscopy (TEM) images of PVB/TiO2 (10 wt.%) composite. Figure 5 presents the solid-phase photocatalytic degradation tendency of the pure polymer and the composites under air atmosphere. The weight loss rate of the composites was higher than that of PVB. The weight loss of the pure polymer and the composites increased continuously with UV irradiation. Pure PVB resulted in insigniﬁcant weight loss, almost 2 wt.%, after 240 h under UV irradiation, which could be ascribed to the boron containing chain structure of the pure polymer [4]. 3. Materials and Methods Polyvinyl alcohol (PVA) with the molecular weight between 89,000 and 98,000 g/mol, boric acid and titanium dioxide (TiO2, anatase, <25 nm) were obtained from Sigma-Aldrich (Munich, Germany). Pure PVB, which did not contain the photocatalyst nanoparticles, was synthesized through the condensation reaction of PVA and boric acid according to the procedure given in the literature [4]. In detail, 4.0 g of PVA was dissolved in 100 mL of distilled water and the solution was heated up to 80 ◦C under stirring. At the same time, 4.0 g of boric acid was dissolved in 100 mL of distilled water and the solution was kept under stirring at room temperature. Afterward, the boric acid solution was fed into the polyvinyl alcohol solution. The mixture was maintained at 80 ◦C under stirring for half an hour, which resulted in the formation of PVB in gel form. PVB in solid form was obtained after drying the gel polymer in an oven at 120 ◦C [4]. PVB/TiO2 composites, including 5, 10, and 15 wt.% of TiO2 nanoparticles, were synthesized with the same procedure followed to prepare pure PVB. Different from the given synthesis procedure, the photocatalyst nanoparticles were fed into the PVA solution prior to mixing with the boric acid solution. Fourier transform infrared (FTIR) spectra of pure polymer and the composite, containing 10 wt.% of TiO2 nanoparticles, were recorded on a Thermo Scientiﬁc FTIR Nicolet 380 (Nicolet Thermo Corporation, Edina, MN, USA) in the wavenumber range between 600 and 4000 cm−1. The thermogravimetric analyses (TGA) of the pure polymer and the composite, including 10 wt.% of TiO2, were performed with a Setaram Labsys TGA/DTA thermogravimetric analyzer (Setaram Instrumentation, Ankara, Turkey) under nitrogen atmosphere at the heating rate 5 ◦C/min. The morphology of the pure polymer and PVB/TiO2 composites, including 5, 10, and 15 wt.% of TiO2, respectively, were studied in a QUANTA 400F model ﬁeld emission scanning electron microscope (FE-SEM) (Thermo Fisher, Hillsboro, OR, USA). A FEI-Tecnai G2 Spirit Biotwin model conventional transmission electron microscope (CTEM) (Thermo Fisher, Hillsboro, OR, USA) was used for transmission electron microscopy (TEM) analysis of the composite, containing 10 wt.% of TiO2. For this purpose, PVB/TiO2 composite sample was grinded into powder form. Then, the powder sample was dispersed in ethanol and the dispersion was dropped on carbon coated copper grids. 2. Results and Discussion TGA results also supported the stated thought. Boron containing chain structure might be the reason for enhanced degradation stability and thermal stability. For the composites with 5 wt.%, 10 wt.% and 15 wt.% of TiO2 nanoparticles, the weight loss values were 8.8 wt.%, 11.2 wt.%, 17.9 wt.%, respectively, after 240 h of UV irradiation. Photoinduced weight loss of the composites increased in parallel with TiO2 content, which demonstrated the effectiveness of the photocatalyst. It was thought that the strong chain structure of PVB lead to the low weight loss values in the composites. UV-Vis spectroscopy was performed to investigate the optical property of TiO2 and the composite with 10 wt.% of TiO2 nanoparticles. According to Figure 6, TiO2 absorbed the majority of the incoming light between 250 and 350 nm. On the other hand, the composite absorbed the majority of the incoming light below 400 nm. UV-Vis spectrum of pure TiO2 exhibits a characteristic absorption band at around 322 nm, which was attributed to the characteristic Ti–O–Ti stretching vibrations [16]. Figure 5. Weight loss of (a) pure PVB, (b) PVB/TiO2 (5 wt.%) composite, (c) PVB/TiO2 (10 wt.%) composite and (d) PVB/TiO2 (15 wt.%) composite under UV irradiation as the function of time in air. Figure 5. Weight loss of (a) pure PVB, (b) PVB/TiO2 (5 wt.%) composite, (c) PVB/TiO2 (10 wt.%) composite and (d) PVB/TiO2 (15 wt.%) composite under UV irradiation as the function of time in air. 535 Catalysts 2018, 8, 499 Figure 6. UV-Vis absorbance spectrum of (a) TiO2 and (b) PVB/TiO2 (10 wt.%). Figure 6. UV-Vis absorbance spectrum of (a) TiO2 and (b) PVB/TiO2 (10 wt.%). References 1. Xing, M.; Qiu, B.; Li, X.; Zhangin, J. TiO2/Graphene Composites with Excellent Performance in Photocatalysis. In Nanostructured Photocatalysts Advanced Functional Materials; Yamashita, H., Li, H., Eds.; Springer International Publishing: Zurich, Switzerland, 2016; pp. 23–68. ISBN 978-3-319-26079-2. g pp 2. Zan, L.; Fa, W.; Wang, S. Novel Photodegradable Low-Density Polyethylene-TiO2 Nanocomposite Film. Environ. Sci. Technol. 2006, 40, 1681–1685. [CrossRef] [PubMed] 2. Zan, L.; Fa, W.; Wang, S. Novel Photodegradable Low-Density Environ. Sci. Technol. 2006, 40, 1681–1685. [CrossRef] [PubMed] 3. He, C.H.; Gong, J. The preparation of PVA–Pt/TiO2 composite nanoﬁber aggregate and the photocatalytic degradation of solid-phase polyvinyl alcohol. Polym. Degrad. STable 2003, 81, 117–124. [CrossRef] 4. Koysuren, O.; Koysuren, H.N. Photocatalytic activity of polyvinyl borate/titanium dioxide composite UV light degradation of organic pollutants J Macromol Sci A 2018 55 401 407 [CrossRef] 4. Koysuren, O.; Koysuren, H.N. Photocatalytic activity of polyvinyl borate/titanium dioxide composites for UV light degradation of organic pollutants. J. Macromol. Sci. A 2018, 55, 401–407. [CrossRef] 5. Lee, S.Y.; Park, S.J. TiO2 photocatalyst for water treatment applications. J. Ind. Eng. Chem. 2013, 19, 1761–1769. [CrossRef] 6. Fa, W.; Guo, L.; Wang, J.; Guo, R.; Zheng, Z.; Yang, F. Solid-phase photocatalytic degradation of polystyre with TiO2/Fe(St)3 as catalyst. J. Appl. Polym. Sci. 2013, 128, 2618–2622. [CrossRef] 7. Zan, L.; Tian, L.; Liu, Z.; Peng, Z. A new polystyrene–TiO2 nanocomposite ﬁlm and its photocataly degradation. Appl. Catal. A-Gen. 2004, 264, 237–242. [CrossRef] dation. Appl. Catal. A-Gen. 2004, 264, 237–242. [CrossRe 8. Zhang, L.; Liu, P.; Su, Z. Preparation of PANI–TiO2 nanocomposites and their solid-phase photocatalytic degradation. Polym. Degrad. Stab. 2006, 91, 2213–2219. [CrossRef] 9. Yang, C.; Gong, C.; Peng, T.; Deng, K.; Zan, L. High photocatalytic degradation activity of the polyvinyl chloride (PVC)-vitamin C (VC)-TiO2 nano-composite ﬁlm. J. Hazard. Mater. 2010, 178, 152–156. [CrossRef] [PubMed] 10. Cho, S.; Choi, W. Solid-phase photocatalytic degradation of PVC-TiO2 polymer composites. J. Photochem. Photobiol. A 2001, 143, 221–228. [CrossRef] 11. Koysuren, O.; Koysuren, H.N. Photocatalytic activities of poly (methyl methacrylate)/titanium diox nanoﬁber mat. J. Macromol. Sci. A 2018, 54, 80–84. [CrossRef] 12. Geng, S.; Shah, F.U.; Liu, P.; Antzutkin, O.N.; Oksman, K. Plasticizing and crosslinking effects of bor additives on the structure and properties of poly (vinyl acetate). RSC Adv. 2017, 7, 7483–7491. [CrossRe 13. Lawrence, M.B.; Desa, J.A.E.; Aswal, V.K.; Rai, R. Properties of poly (vinyl alcohol)-borax gel doped with neodymium and praseodymium. Bull. Mater. Sci. 2014, 37, 301–307. [CrossRef] 14. 3. Materials and Methods The solid-phase photocatalytic degradation of the pure polymer and the composite samples (1.0 g) was carried out in the ambient air using a 30 W UV lamp (254 nm, Philips, Istanbul, Turkey). All samples were weighed before and after UV irradiation to evaluate the weight loss of PVB through the solid-phase photocatalytic degradation. Each photocatalytic degradation experiment used a triplicate set of samples. The UV–Vis absorption spectrum of TiO2 and PVB/TiO2 composite, containing 10 wt.% of TiO2 nanoparticles, were carried out by a Genesys 10S spectrophotometer (Thermo Fisher, Hillsboro, OR, USA) in the wavelength of 200–800 nm. 536 Catalysts 2018, 8, 499 4. Conclusions PVB/TiO2 polymer composites were synthesized through the condensation reaction of polyvinyl alcohol and boric acid in the presence of TiO2 nanoparticles. FTIR analysis veriﬁed PVB synthesis. TGA results revealed the improvement in thermal stability of PVB with TiO2 contribution. According to SEM and TEM analyses, TiO2 nanoparticles in aggregate structure illustrated good dispersion in PVB matrix. Adding TiO2 nanoparticles in PVB matrix enhanced the solid-phase photocatalytic degradation of the polymer matrix under UV light irradiation. Funding: This research received no external funding. Acknowledgments: I thank Ozcan Koysuren for comments that greatly improved the manuscript. Acknowledgments: I thank Ozcan Koysuren for comments that greatly improved the manuscript Conﬂicts of Interest: The author declares no conﬂict of interest. Conﬂicts of Interest: The author declares no conﬂict of interest. The Synergistic Effect of Pyridinic Nitrogen and Graphitic Nitrogen of Nitrogen-Doped Graphene Quantum Dots for Enhanced TiO2 Nanocomposites’ Photocatalytic Performance Fei Li 1, Ming Li 2, Yi Luo 1, Ming Li 1,, Xinyu Li 1, Jiye Zhang 3 and Liang Wang 2, 1 College of Science, Guilin University of Technology, Guilin 541004, China; m17753101516@163.com (F.L adam2513@163.com (Y.L.); lixinyu5260@163.com (X.L.) y 2 Institute of Nanochemistry and Nanobiology, School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China; limingdobest@163.com 2 Institute of Nanochemistry and Nanobiology, School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China; limingdobest@163.com g y g g 3 School of Materials Science and Engineering, Shanghai University, Shanghai 200444, China; jychang@shu.edu.cn g y g g 3 School of Materials Science and Engineering, Shanghai University, Shanghai 200444, China; jychang@shu.edu.cn * Correspondence: liming928@163.com (M.L.); wangl@shu.edu.cn (L.W.); Tel.: +86-773-369-6613 (M.L.); +86-021-661-3526 (L.W.) Received: 9 September 2018; Accepted: 30 September 2018; Published: 4 October 2018 Received: 9 September 2018; Accepted: 30 September 2018; Published: 4 October 2018 Abstract: In this study, nitrogen-doped graphene quantum dots (N-GQDs) and a TiO2 nanocomposite were synthesized using a simple hydrothermal route. Ammonia water was used as a nitrogen source to prepare the N-GQDs. When optically characterized by UV-vis, N-GQDs reveal stronger absorption peaks in the range of ultraviolet (UV) light than graphene quantum dots (GQDs). In comparison with GQDs/TiO2 and pure TiO2, the N-GQDs/TiO2 have signiﬁcantly improved photocatalytic performance. In particular, it was found that, when the added amount of ammonia water was 50 mL, the content of pyridinic N and graphitic N were as high as 22.47% and 31.44%, respectively. Most important, the photocatalytic activity of N-GQDs/TiO2-50 was about 95% after 12 min. The results illustrated that pyridinic N and graphitic N play a signiﬁcant role in photocatalytic performance. Keywords: N-doped graphene quantum dots; TiO2; photocatalytic performance; pyridinic N; graphitic N www.mdpi.com/journal/catalysts References Yanase, I.; Ogaware, R.; Kobayashi, H. Synthesis of boron carbide powder from polyvinyl borate precursor. Mater. Lett. 2009, 63, 91–93. [CrossRef] 15. Riaz, N.; Bustam, M.A.; Chong, F.K.; Man, Z.B.; Khan, M.S.; Shariff, A.M. Photocatalytic Degradation of DIPA Using Bimetallic Cu-Ni/TiO2 Photocatalyst under Visible Light Irradiation. Sci. World J. 2014, 2014, 1–8. [CrossRef] 16. Sarmah, S.; Kumar, A. Photocatalytic activity of polyaniline-TiO2 nanocomposites. Indian J. Phys. 2011, 85, 713–726. [CrossRef] © 2018 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 537 1. Introduction In recent decades, increasing environmental pollution has attracted more and more attention, especially the discharge of dye wastewater from factories. It is therefore appropriate to ﬁnd an effective, low-cost and pollution-free replacement for traditionally problematic energy production. Photocatalysis could be one of the most effective measures to solve the problems of energy shortage and environmental pollution [1–4]. In many semiconductor metal oxide materials, for example, titanium dioxide (TiO2) is extensively used as a photocatalyst [5–7], due to its beneﬁcial characteristics. It is inexpensive, non-poisonous, and has excellent chemical and physical stability [5,8]. Although it has so many superior properties, use of TiO2 as a photocatalyst is limited by some disadvantages in practical application, such as a wide band gap (3.2 eV) and a high electron-hole recombination rate, which leads to low photocatalytic efﬁciency [3,9–11]. To perfect the photocatalytic activity of TiO2, various measures were utilized, such as many ions being doped into the lattice of TiO2 [12], sensitization via absorbed molecules [13–15], compound with other materials [16,17], and the surface being coated with other cocatalysts possessing excellent performance [18–21]. Among the methods mentioned above, surface loading with other cocatalysts is relatively facile and effective in enhancing the photocatalytic activity of TiO2. Although some auxiliary catalysts can improve the photocatalytic performance of Catalysts 2018, 8, 438; doi:10.3390/catal8100438 539 Catalysts 2018, 8, 438 TiO2, for instance Pt, Au and Ag, their high cost limits their application [22,23]. Therefore, it would be signiﬁcant to ﬁnd highly efﬁcient, simple and eco-friendly cocatalysts which enhance the photocatalytic performance of TiO2. Graphene quantum dots (GQDs) are a novel kind of 0D carbon nanomaterial with dimensions below 10 nm. In addition to all the properties of graphene, GQDs also have unique edge effects and quantum conﬁnement [24]. GQDs are widely used in various ﬁelds due to their excellent physical and chemical properties. They are used in photovoltaic devices [25], catalysis [26–30], drug delivery [31], and cell imaging [32–35]. GQDs are environmentally friendly materials with strong anti-chemical corrosion and anti-ultraviolet (UV) irradiation capabilities. Pure GQDs display low catalytic activity due to their high exciton binding energy [36]. In many past studies, GQDs as auxiliary catalysts effectively improved the photocatalytic performance of TiO2 [37], and some reports showed that the doped GQDs displayed excellent effects on improving the photocatalytic performance of TiO2, for example, when nitrogen [38], sulfur [3] and nitrogen and sulfur co-doped [39]. 1. Introduction However, there is little work on the effect of different N-bonding structure for the photocatalytic performance of nitrogen-doped GQDs (N-GQDs). g p In our study, N-GQDs with different N contents were synthesized by a facile hydrothermal stratagem using different volumes of ammonia water and GQDs. N-GQDs were attached tightly to the surface of TiO2 with a facile hydrothermal method. The photocatalytic performance of N-GQDs/TiO2 was tested by introducing methyl orange (MO). A possible mechanism for improving photocatalytic performance was also investigated and analyzed by comparing the photocatalytic effect of N-GQDs/TiO2 and pure TiO2. All the results showed that N-GQDs effectively improved the photocatalytic performance of TiO2, in which pyridinic N and graphitic N play a decisive role. This work may provide a new perspective for the future study of complexes based on N-GQDs. 2. Results and Discussion Unless otherwise speciﬁed, the N-GQDs-50 with the best optical performance was selected for various characterizations. 2.1. Morphology and Structural Characterization of GQDs and N-GQDs 2.1. Morphology and Structural Characterization of GQDs and N-GQDs Figure 1a–d displays the transmission electron microscopy (TEM) and high resolution transmission electron microscopy (HRTEM) images of GQDs and N-GQDs-50. Figure 1a,b shows the TEM images of GQDs and N-GQDs-50, and the size distribution is homogeneous. The size of GQDs and N-GQDs-50 ranged from 2–16 nm and 1–5 nm, the average diameter was 8.66 nm and 3.12 nm respectively (insert in Figure 1a,b). The reduction in the size of N-GQDs-50 was likely to be due to further decomposition of GQDs during the subsequent hydrothermal reaction (refer to the experiment for details). The HRTEM images show that the plane lattice spacing of GQDs and N-GQDs-50 was 0.21 nm, which is similar to the in-plane lattice spacing of graphite (002) [40,41]. Figure 1e shows a TEM image of TiO2 nanoparticles with the thin film. After the hydrothermal reaction, the N-GQDs were compounded on the surface of TiO2. With the oxygen-containing functional groups of N-GQDs, the hydroxyl functional groups of TiO2 may be able to construct functional and relatively stable composites. XRD was employed to determine the crystalline structure of GQDs and N-GQDs-50. It was clear from the pattern (Figure 2a) of pure TiO2 that there were two types of TiO2, namely anatase and rutile. Peaks at 2θ = 25.6◦, 37.18◦, 48.25◦, 54.02◦, 55.24◦and 62.7◦represented (101), (004), (200), (105), (211) and (116) planes of anatase. Others at 41.44◦and 56.82◦, represented (110) and (114) of rutile, which identified with P25. The XRD pattern of N-GQDs-50 showed the peaks of N-GQDs/TiO2-50 were the same as that of TiO2, indicating that the structure of TiO2 was not affected by N-GQDs. FT-IR spectroscopy can also characterize samples. As shown in Figure 2b, broad absorption bands at 480–700 cm−1 were associated with stretching vibrations of Ti-O-Ti and Ti-O-C. The peak at 1380 cm−1 was related to nitrate ion and the peak at 1633 cm−1 was due to δH2O vibration of the water molecule [42]. The figures of FT-IR indicated that N-GQDs was successfully coupled with TiO2. 540 Catalysts 2018, 8, 438 Figure 1. TEM images of GQDs (a), N-GQDs-50 (b), TiO2 (e) and N-GQDs/TiO2-50 (f), insets are of corresponding lateral size distribution. HRTEM images of GQDs (c) and N-GQDs-50 (d). Figure 1. TEM images of GQDs (a), N-GQDs-50 (b), TiO2 (e) and N-GQDs/TiO2-50 (f), insets are of corresponding lateral size distribution. HRTEM images of GQDs (c) and N-GQDs-50 (d). Figure 2. 2.1. Morphology and Structural Characterization of GQDs and N-GQDs XRD pattern (a) of pure TiO2 and N-GQDs/TiO2-50; FT-IR spectra (b) of pure TiO2, N-GQDs-50 and N-GQDs/TiO2-50 composites. Figure 2. XRD pattern (a) of pure TiO2 and N-GQDs/TiO2-50; FT-IR spectra (b) of pure TiO2, N-GQDs-50 and N-GQDs/TiO2-50 composites. To further investigate the composition of GQDs and N-GQDs, XPS (X-ray photoelectron spectroscopy) measurement was employed. Figure 3a shows the full XPS spectra of GQDs, N-GQDs-50 and N-GQDs-100. Peaks can be seen at approximately 284 eV (C 1s), 399.08 eV (N 1s) and 531.08 eV (O 1s) in all the samples. Compared to GQDs, the intensity of N 1s peak N-GQDs-50 and N-GQDs-100 was relatively enhanced, indicating that the N was successfully doped into the GQDs through the hydrothermal reaction with ammonia water. The results in Table 1 further show that the N content of N-GQDs-50 and N-GQDs-100 was higher than GQDs, and the content of N-GQDs-50 was the highest, reaching 10.64%, also indicating that N-GQDs were synthesized successfully. The high-resolution 541 Catalysts 2018, 8, 438 spectrum of N 1s region of N-GQDs-50 and N-GQDs-100 was divided into three peaks at 398.9 eV (pyridinic N), 399.6 eV (pyrrolic N) and 401.5 eV (graphitic N) [25,43]. Figure 3. The full XPS spectra (a) of GQDs, N-GQDs-50 and N-GQDs-100. High-resolution N 1s spectrum of N-GQDs-50 (b) and N-GQDs-100 (c). Figure 3. The full XPS spectra (a) of GQDs, N-GQDs-50 and N-GQDs-100. High-resolution N 1s spectrum of N-GQDs-50 (b) and N-GQDs-100 (c). Table 1. The atomic percent (%) GQDs, N-GQDs-50 and N-GQDs-100 from XPS data. Table 1. The atomic percent (%) GQDs, N-GQDs-50 and N-GQDs-100 from XPS data. Samples C (at%) O (at%) N (at%) GQDs 73.51 18.82 7.67 N-GQDs-50 71.41 17.95 10.64 N-GQDs-100 70.36 19.22 10.42 2.2. Optical Properties 2.2. Optical Properties Optical properties were used to characterize the physical nature of carbon-based materials. As‘shown in Figure 4, the optical absorption ability of GQDs (0.07 mg/mL) (Figure 4a) and N-GQDs-50 (0.07 mg/mL) (Figure 4b) was investigated by UV-vis spectrometer. GQDs displayed a wide absorption peak at 400–500 nm, which was similar to previous studies [44,45]. Compared to GQDs, N-GQDs-50 was also detected as having a strong absorption peak in the UV region at approximately 344 nm. Obviously, differences between GQDs and N-GQDs-50 in UV-vis spectra indicated that GQDs doped with N atom resulted in a strong absorption peak in the UV range. Figure 4. The UV-vis absorption spectra of the GQDs (a) and N-GQDs-50 (b). Figure 4. The UV-vis absorption spectra of the GQDs (a) and N-GQDs-50 (b). According to the results of UV-vis absorption spectrum, the excitation wavelength of 310–390 nm was chosen for photoluminescent (PL) measurement. In this experiment, both GQDs and 542 Catalysts 2018, 8, 438 N-GQDs-50 displayed excitation-independent PL behaviors, which were contrary to previous work on carbon-based ﬂuorescent materials [41]. The emission peak of GQDs (Figure 5a) was approximately 540 nm and N-GQDs-50 (Figure 5b) was observed at approximately 520 nm. The excitation wavelength of 350 nm, with an emission peak of N-GQDs had blue shifted 20 nm compared to the GQDs in Figure 5c, which was most likely due to the reduction in size of N-GQDs-50 (see inserts of Figure 1a,b). This phenomenon is consistent with the trend observed by other quantum dots due to the quantum conﬁnement effect at smaller particle size [46]. Figure 5d shows the PLE spectrum of N-GQDs-50 under the emission wavelength at 520 nm. One of the peaks was observed at 388 nm, which was in accordance with the PL results. N-GQDs-50 displayed excitation-independent PL behaviors, which were contrary to previous work on carbon-based ﬂuorescent materials [41]. The emission peak of GQDs (Figure 5a) was approximately 540 nm and N-GQDs-50 (Figure 5b) was observed at approximately 520 nm. The excitation wavelength of 350 nm, with an emission peak of N-GQDs had blue shifted 20 nm compared to the GQDs in Figure 5c, which was most likely due to the reduction in size of N-GQDs-50 (see inserts of Figure 1a,b). This phenomenon is consistent with the trend observed by other quantum dots due to the quantum conﬁnement effect at smaller particle size [46]. 2.2. Optical Properties Figure 5d shows the PLE spectrum of N-GQDs-50 under the emission wavelength at 520 nm. One of the peaks was observed at 388 nm, which was in accordance with the PL results. Figure 5. The PL spectra of GQDs (a) and N-GQDs-50 (b) under excitation with 310–390 nm. (c) Comparison of GQDs with N-GQDs under irradiation with 360 nm wavelength. (d) PLE spectra of N-QDs-50 when ﬁxing emission wavelength at 520 nm. (e) Typical electronic transitions of triple carbenes in the optical spectrum of N-GQDs. (f) Comparison of N-GQDs-50 with N-GQDs/TiO2-50 under irradiation with 330 nm wavelength. Figure 5. The PL spectra of GQDs (a) and N-GQDs-50 (b) under excitation with 310–390 nm. (c) Comparison of GQDs with N-GQDs under irradiation with 360 nm wavelength. (d) PLE spectra of N-QDs-50 when ﬁxing emission wavelength at 520 nm. (e) Typical electronic transitions of triple carbenes in the optical spectrum of N-GQDs. (f) Comparison of N-GQDs-50 with N-GQDs/TiO2-50 under irradiation with 330 nm wavelength. The photoluminescence excitation (PLE) spectra showed that luminescence from N-GQDs could correspond to transitions at 278 (4.46 eV) and 388 (3.2 eV) nm, which could have been the result of transition between the σ and π orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO), as shown in Figure 5e. Fluorescence performance was improved by the combination with photo-generated carriers. However, one way to improve photocatalytic performance was to inhibit the recombination of photo-generated carriers so that they could react with organic pollutants on The photoluminescence excitation (PLE) spectra showed that luminescence from N-GQDs could correspond to transitions at 278 (4.46 eV) and 388 (3.2 eV) nm, which could have been the result of transition between the σ and π orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO), as shown in Figure 5e. Fluorescence performance was improved by the combination with photo-generated carriers. However, one way to improve photocatalytic performance was to inhibit the recombination of photo-generated carriers so that they could react with organic pollutants on 543 Catalysts 2018, 8, 438 the surface of photocatalysts. Thus, Figure 5f shows that N-GQDs/TiO2-50 possesses excellent photocatalytic activity. The multiplicity of carbine ground-state was connected with energy differences (δE) between the σ and π orbital. According to previous reports, δE should be less than 1.5 eV [47]. In our study, δE of N-GQDs was 1.26 eV, which demonstrated that δE was within the theoretical value. 2.3. 2.2. Optical Properties Photocatalytic Activity and Possible Mechanism for Improving Photocatalytic Activity The concentration C/C0 of undegraded MO was used to indicate photocatalytic performance of different catalysts. MO without a catalyst degrades differently under UV light in Figure 6a,b. Pure TiO2 nanoparticles displayed ﬁne photocatalytic activity by UV irradiation, and the degradation rate of MO reached approximately 57% within 12 min. The photocatalytic performance of GQDs/TiO2 was higher than pure TiO2, which reached about 65%. Although the photocatalytic performance of GQDs/TiO2 was preferable to that of pure TiO2, the effect was not satisfactory. MO degradation by N-GQDs/TiO2 was much higher than that of other catalysts. In particular, the degradation of MO by N-GQDs/TiO2-50 reached 95% within 12 min, indicating that the content of graphitic N played a signiﬁcant function on photocatalytic activity. As shown in Table 2, with the increase of ammonia water content, the content of pyrrolic N obviously increased, while the content of pyridinic N and graphitic N were lessened. The content of pyridinic N and graphitic N of N-GQDs-50 were higher than that of N-GQDs-100, up to 22.47% and 31.4%, respectively. Figure 6. (a) The relationship between the concentration of undegraded MO and illumination time for different photocatalysts. (b) The relationship between the logarithm of C/C0 and irradiation time of different photocatalysts. Figure 6. (a) The relationship between the concentration of undegraded MO and illumination time for different photocatalysts. (b) The relationship between the logarithm of C/C0 and irradiation time of different photocatalysts. Table 2. The content (%) of doped-N and the different N species of N-GQDs-50 and N-GQDs-100. Table 2. The content (%) of doped-N and the different N species of N-GQDs-50 and N-GQDs-100. Samples N (at%) Pyridinic N (at%) Pyrrolic N (at%) Graphitic N (at%) N-GQDs-50 10.64 22.47 44.09 31.44 N-GQDs-100 10.42 16.38 53.03 30.59 To further study the ability and stability of photocatalytic MO degradation by N-GQDs/TiO2 composites, the cyclic stability experiment of photocatalytic degradation of MO by N-GQDs/TiO2-50 was investigated (Figure 7a). After ﬁve cycles, the N-GQDs/TiO2-50 was yet to show a good photocatalytic effect. As shown in Figure 7b, the photocatalytic performance of N-GQDs/TiO2-50 and pure TiO2 was slightly reduced, but the photocatalytic activity was still excellent, reaching over 90%. Their results show that the photocatalytic activity of TiO2 could be greatly enhanced by modifying TiO2 with effective methods. 544 Catalysts 2018, 8, 438 Figure 7. Scheme 1. The possible photocatalyst mechanism of N-GQDs/TiO2 under UV light 3.3. Preparation of N-GQDs/TiO2 N-GQDs/TiO2 composites were synthesized by a simple hydrothermal method. Weighed 0.4 g Degussa P25 TiO2 was dispersed into 200 mL N-GQDs aqueous solution (0.2 mg/mL) by ultrasonication (500 W, 40 kHz) for 30 min. Then the solution was transferred into Teﬂon-lined steel autoclave and then heated at 180 ◦C for 24 h. After cooling to room temperature, the product was washed three times by centrifugation with deionized water and anhydrous alcohol, then the collected sediment was dried at 70 ◦C in air to obtain the N-GQDs/TiO2-50 composite. The GQDs/TiO2 and N-GQDs/TiO2-100 composites were also synthesized under the same conditions for comparison. 3.2. Preparation of N-GQDs GQDs were prepared using a simple hydrothermal method [48]. Then N-GQDs samples (N-GQDs-50, N-GQDs-100, -numbers represent the volume of the added ammonia water) were synthesized by a simple hydrothermal method. Brieﬂy, 0.1 g GQDs was dispersed in 50 mL H2O, and added to different volumes of ammonia water (50 mL and 100 mL). The mixed homogeneous solution was transferred into a Teﬂon-lined steel autoclave and then heated at 180 ◦C for 12 h. After cooling to room temperature, the obtained solution was ﬁltered with a 0.22 μm ﬁlter membrane, and the ﬁltered solution was dialyzed for 24 h using a 3500 Da dialysis bag to remove excess ions. Finally, the obtained N-GQDs were dried at 70 ◦C in air for the subsequent experiment. 3.4. Photocatalytic Activity Measurements The photocatalytic performance of the obtained samples was explored by degrading MO in quartz tubes at the UV light irradiation of a 600 W mercury lamp, and the photocatalytic experiment was carried out at room temperature. The distance between solution and lamp was 10 cm. The experimental procedure is as described: 50 mg samples were dissolved in 50 mL MO (5 mg/L). The resulting solution was roughened without treatment for an hour to achieve adsorption and desorption equilibrium between the catalyst and MO. Then the solution was put under the mercury lamp for illumination with magnetic stirring and 4 mL solution was removed every 3 min to a centrifuge for 5 min (8000 rpm) to remove catalyst particles. The concentration of MO after centrifugation was measured by a UV/vis/near infrared (NIR) spectrometer. The expression formula of degradation rate of MO is C0−C C0 . The concentration of undegraded MO can be expressed as C/C0. In this study, C represents the concentration of MO after irradiation, C0 represents the original concentration of MO before irradiation. 2.2. Optical Properties (a) Recycle stability of the photocatalytic decomposition of MO by N-GQDs/TiO2-50. (b) Repetitive photocatalytic decomposition of MO for TiO2 and N-GQDs/TiO2-50 photocatalysts. Figure 7. (a) Recycle stability of the photocatalytic decomposition of MO by N-GQDs/TiO2-50. (b) Repetitive photocatalytic decomposition of MO for TiO2 and N-GQDs/TiO2-50 photocatalysts. Scheme 1 explains the probable mechanism of degradation of MO by N-QGDs/TiO2 composites. As an n-type semiconductor, TiO2 was able to create electron-hole pairs [44]. Under UV irradiated light, the electron of TiO2 transferred from the valence band to the conduction band to form an electro-hole. Electro-holes reacted with absorbed O2/OH- to produce ·O2/·OH so as to degrade MO. N-GQDs attached to the surface of TiO2, absorbing UV light and raising the excitation of electrons. The excited electrons then transferred to the conduction band of TiO2. With TiO2 as the base of catalytic reaction, the N-GQDs as an unexceptionable electron migration area on the surface of TiO2 could effectively and rapidly transmit photogenerated electrons, inhibiting the fast binding of photogenerated electron-hole pairs, and thus greatly improved the catalytic efﬁciency of TiO2. Compared with GQDs, N-GQDs showed a strong absorption peak in the UV region. Therefore, N-GQDs/TiO2 displayed strong photocatalytic activity by UV light. The oxygen in MO also combined with electros on N-GQDs to generate ·O2−, which may have played a signiﬁcant role in photocatalytic activity [6]. As the main N-binding conﬁguration, pyridine N only existed at the edge of the GQDs, which could be used as the oxygen-reduction active site to enhance the activity of a catalyst. Furthermore, the graphitic N was the electron transfer site [43]. Thus, N-GQDs/TiO2 displayed good photocatalytic performance, and the pyridinic N and graphitic N played a signiﬁcant position in photocatalytic performance. Scheme 1. The possible photocatalyst mechanism of N-GQDs/TiO2 under UV light. Scheme 1. The possible photocatalyst mechanism of N-GQDs/TiO2 under UV light. 3. Experimental Section 3.1. Chemicals All reagents were not processed further. Pyrene (C6H6), anhydrous alcohol (C2H5OH), sodium hydroxide (NaOH), nitric acid (HNO3) ammonia water, titanium dioxide (TiO2) were purchased 545 Catalysts 2018, 8, 438 from reagent agent (manufacturer, city, country). Deionized water was used in all the experimental processes. All the chemicals were purchased from shanghai, China. 4. Conclusion In this study, we synthesized N-GQDs/TiO2 composites by two facile hydrothermal methods. The results show that N-GQDs/TiO2 exhibit excellent photocatalytic performance, and the ability to degrade MO for cyclic stability. In addition, the photocatalytic activity of N-GQDs/TiO2 is associated with the content of graphitic N and the higher content, the better of photocatalytic activity. In particular, it was found that when the amount of ammonia water added was 50 mL and the contents of pyridinic N and graphitic N were as high as 22.47% and 31.44%, respectively. The photocatalytic performance of N-GQDs/TiO2-50 reached about 95% in 12 min. This accomplishment may provide a new perspective for the future study of composities based on N-GQDs. Author Contributions: Correspondence author, M.L. and L.W.; Data curation, F.L.; investigation, X.L. and J.Z.; methodology, F.L. and M.L. and Y.L.; supervision, M.L. (correspondence author) and L.W.; writing (original draft), F.L.; writing (review and editing), L.W. Acknowledgments: This work was ﬁnancially supported by the National Natural Science Foundation of China (Nos. 11764011, 21671129, 51472241, 21571124) and Natural Science Foundation of Guangxi Province (No. 2016GXNSFAA380008, 2017GXNSFBA198216), P. R. China. Acknowledgments: This work was ﬁnancially supported by the National Natural Science Foundation of China (Nos. 11764011, 21671129, 51472241, 21571124) and Natural Science Foundation of Guangxi Province (No. 2016GXNSFAA380008, 2017GXNSFBA198216), P. R. China. Conﬂicts of Interest: The authors declare no competing ﬁnancial interest. 3.5. Characterization Morphology of samples was measured by transmission electron microscopy (TEM, HT7700, Hitachi, Tokyo, Japan), and X-ray photoelectron spectroscopy (XPS) data were characterized by an ESCALAB 250Xi electron spectrometer (ThermoFisher Scientiﬁc, Waltham, MA, USA) with Al Kα Radiation (1486.6 eV). X-ray diffraction (XRD) patterns of samples were recorded within 5–80◦(2θ) using a Rigaku D/MAX 2550 diffractometer (Rigaku, Tokyo, Japan) carried out at 40 kV and 100 mA. Fourier transform infrared spectroscopy (FT-IR) was investigated using a Perkin-Elmer spectrum. The UV-vis absorption spectrum was measured by using a UV/vis/NIR spectrometer (Perkin-Elmer, Lambda 750, PerkinElmer, Shelton, CT, USA). The photoluminescent (PL) and photoluminescence excitation (PLE) spectra (Carry Eclipse Fluorescence Spectrophotometer, Agilent Technologies Ltd., Cheadle, UK) were carried out using a ﬂuorescence spectrophotometer. 546 Catalysts 2018, 8, 438 References 1. Kumar, S.G.; Devi, L.G. Review on modiﬁed TiO2 photocatalysis under UV/visible light: selected results and related mechanisms on interfacial charge carrier transfer dynamics. J. Phys. Chem. Coruña 2011, 115, 13211–13241. [CrossRef] [PubMed] 2. Cheng, H.; Fuku, K.; Kuwahara, Y.; Mori, K.; Yamashita, H. Harnessing single-active plasmonic nanostructures for enhanced photocatalysis under visible light. J. Mater. Chem. Coruña 2015, 3, 5244–5258. [CrossRef] 3. Luo, Y.; Li, M.; Hu, G.; Tang, T.; Wen, J.; Li, X.; Wang, L. Enhanced photocatalytic activity of sulfur-doped graphene quantum dots decorated with TiO2 nanocomposites. Mater. Res. Bull. 2018, 97, 428–435. [CrossRef] 4. Bhatia, S.; Verma, N. Photocatalytic activity of ZnO nanoparticles with optimization of defects. Mater. Res. Bull. 2017, 95, 468–476. [CrossRef] 5. Wang, X.; Kaﬁzas, A.; Li, X.; Moniz, S.J.A.; Reardon, P.J.T.; Tang, J.; Parkin, I.P.; Durrant, J.R. Transient absorption spectroscopy of anatase and rutile: The impact of morphology and phase on photocatalytic activity. J. Phys. Chem. C 2015, 119, 10439–10447. [CrossRef] y y 6. Konstantinou, I.K.; Albanis, T.A. TiO2-assisted photocatalytic degradation of azo dyes in aqueous solution: kinetic and mechanistic investigations: A review. Appl. Catal. B Environ. 2004, 49, 1–14. [CrossRef] 7. Ruggieri, F.; Antonio D’Archivio, A.; Fanelli, M.; Santucci, S. Photocatalytic degradation of linuron in aqueous suspensions of TiO2. RSC Adv. 2011, 1, 611–618. [CrossRef] 8. Wei, H.; Wang, L.; Li, Z.; Ni, S.; Zhao, Q. Synthesis and photocatalytic activity of one-dimensional CdS@TiO2 core-shell heterostructures. Nano-Micro Lett. 2011, 3, 6–11. [CrossRef] 8. Wei, H.; Wang, L.; Li, Z.; Ni, S.; Zhao, Q. Synthesis and photocatalytic act core-shell heterostructures. Nano-Micro Lett. 2011, 3, 6–11. [CrossRef] 9. Ruggieri, F.; Di Camillo, D. Electrospun Cu-, W- and Fe-doped TiO2 nanoﬁbres for photocatalytic degradation of rhodamine 6G. J. Nanopart. Res. 2013, 15, 1982. [CrossRef] 9. Ruggieri, F.; Di Camillo, D. Electrospun Cu-, W- and Fe-doped TiO2 nan of rhodamine 6G. J. Nanopart. Res. 2013, 15, 1982. [CrossRef] 10. Sawunyama, P.; Yasumori, A.; Okada, K. The nature of multilayered T by a sol-gel process. Mater. Res. Bull. 1998, 33, 795–801. [CrossRef] 10. Sawunyama, P.; Yasumori, A.; Okada, K. The nature of multilayered TiO2-based photocatalytic ﬁlms prepared by a sol-gel process. Mater. Res. Bull. 1998, 33, 795–801. [CrossRef] 11. Mori, R.; Takahashi, M.; Yoko, T. 2D spinodal phase-separated TiO2 ﬁlms prepared by sol–gel process and photocatalytic activity. Mater. Res. Bull. 2004, 39, 2137–2143. [CrossRef] 12. Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, T. References Visible-light photocatalysis in nitrogen-doped titani oxides. Science 2001, 293, 269–271. [CrossRef] [PubMed] 12. Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, T. Visible- oxides. Science 2001, 293, 269–271. [CrossRef] [PubMed] 13. Mori, K.; Kawashima, M.; Yamashita, H. Visible-light-enhanced Suzuki-miyaura coupling reaction by cooperative photocatalysis with an Ru-Pd bimetallic complex. Chem. Commun. 2014, 50, 14501–14503. [CrossRef] [PubMed] 14. Yuan, Y.-P.; Yin, L.-S.; Cao, S.-W.; Xu, G.-S.; Li, C.-H.; Xue, C. Improving photocatalytic hydrogen production of metal-organic framework UiO-66 octahedrons by dye-sensitization. Appl. Catal. B Environ. 2015, 168–169, 572–576. [CrossRef] 547 Catalysts 2018, 8, 438 15. Wang, P.; Wang, J.; Ming, T.; Wang, X.; Yu, H.; Yu, J.; Wang, Y.; Lei, M. Dye-sensitization-induced visible-light reduction of graphene oxide for the enhanced TiO2 photocatalytic performance. ACS Appl. Mater. Interfaces 2013, 5, 2924–2929. [CrossRef] [PubMed] 16. Sadhu, S.; Poddar, P. Template-free fabrication of highly-oriented single-crystalline 1D-rutile TiO2- MWCNT composite for enhanced photoelectrochemical activity. J. Phys. Chem. C 2014, 118, 19363–19373. [CrossRef] 17. Wan, L.; Long, M.; Zhou, D.; Zhang, L.; Cai, W. Preparation and characterization of freestanding hierarchical porous TiO2 monolith modiﬁed with graphene oxide. Nano-Micro Lett. 2012, 4, 90–97. [CrossRef] porous TiO2 monolith modiﬁed with graphene oxide. Nano-Micro Lett. 2012, 4, 90–97. [CrossRef] 18. Yu, H.; Liu, R.; Wang, X.; Wang, P.; Yu, J. Enhanced visible-light photocatalytic activity of Bi2WO6 po ous O2 o o t od ed w t g ap e e o de Nano Micro ett 0 , , 90 9 [C oss e ] 18. Yu, H.; Liu, R.; Wang, X.; Wang, P.; Yu, J. Enhanced visible-light photocatalytic activity of Bi2WO6 18. Yu, H.; Liu, R.; Wang, X.; Wang, P.; Yu, J. Enhanced visible-light photocatalytic activity of Bi2W nanoparticles by Ag2O cocatalyst. Appl. Catal. B Environ. 2012, 111–112, 326–333. [CrossRef] 19. Wang, P.; Xia, Y.; Wu, P.; Wang, X.; Yu, H.; Yu, J. Cu(II) as a general cocatalyst for improved visible-light photocatalytic performance of photosensitive ag-based compounds. J. Phys. Chem. C 2014, 118, 8891–8898. [CrossRef] 20. Gu, Y.; Xing, M.; Zhang, J. Synthesis and photocatalytic activity of graphene based doped TiO2 nanocomposites. Appl. Surf. Sci. 2014, 319, 8–15. [CrossRef] 21. Aleksandrzak, M.; Adamski, P.; Kukułka, W.; Zielinska, B.; Mijowska, E. Effect of graphene thickness on photocatalytic activity of TiO2-graphene nanocomposites. Appl. Surf. Sci. 2015, 331, 193–199. [CrossRef] 22. Seh, Z.W.; Liu, S.; Low, M.; Zhang, S.-Y.; Liu, Z.; Mlayah, A.; Han, M.-Y. References Janus Au-TiO2 Photocatalysts with Strong Localization of Plasmonic Near-Fields for Efﬁcient Visible-Light Hydrogen Generation. Adv. Mater. 2012, 24, 2310–2314. [CrossRef] [PubMed] 23. Zheng, Z.; Huang, B.; Qin, X.; Zhang, X.; Dai, Y.; Whangbo, M.-H. Facile in situ synthesis of visible-light plasmonic photocatalysts M@TiO2 (M = Au, Pt, Ag) and evaluation of their photocatalytic oxidation of benzene to phenol. J. Mater. Chem. 2011, 21, 9079–9087. [CrossRef] p 24. Wang, L.; Wu, B.; Li, W.; Wang, S.; Li, Z.; Li, M.; Pan, D.; Wu, M. Amphiphilic graphene quantum dots as self-targeted ﬂuorescence probes for cell nucleus imaging. Adv. Biosyst. 2018, 2, 1700191. [CrossRef] 25. Pan, D.; Jiao, J.; Li, Z.; Guo, Y.; Feng, C.; Liu, Y.; Wang, L.; Wu, M. Efﬁcient separation of electron-hole pairs in graphene quantum dots by TiO2 heterojunctions for dye degradation. ACS Sustain. Chem. Eng. 2015, 3, 2405–2413. 26. Fei, H.; Ye, R.; Ye, G.; Gong, Y.; Peng, Z.; Fan, X.; Samuel, E.L.G.; Ajayan, P.M.; Tour, J.M. Boron- and nitrogen-doped graphene quantum dots/graphene hybrid nanoplatelets as efﬁcient electrocatalysts for oxygen reduction. ACS Nano 2014, 8, 10837–10843. [CrossRef] [PubMed] 27. Tang, L.; Wang, J.; Jia, C.; Lv, G.; Xu, G.; Li, W.; Wang, L.; Zhang, J.; Wu, M. Simulated solar driven catalytic degradation of psychiatric drug carbamazepine with binary BiVO4 heterostructures sensitized by graphene quantum dots. Appl. Catal. B Environ. 2017, 205, 587–596. [CrossRef] 28. Yeh, T.-F.; Teng, C.-Y.; Chen, S.-J.; Teng, H. Nitrogen-doped graphene oxide quantum dots as photocatalysts for overall water-splitting under visible light illumination. Adv. Mater. 2014, 26, 3297–3303. [CrossRef] [PubMed] 29. Li, F.; Sun, L.; Luo, Y.; Li, M.; Xu, Y.; Hu, G.; Li, X.; Wang, L. Effect of thiophene S on the enhanced ORR electrocatalytic performance of sulfur-doped graphene quantum dot/reduced graphene oxide nanocomposites. RSC Adv. 2018, 8, 19635–19641. [CrossRef] 30. Sun, L.; Luo, Y.; Li, M.; Hu, G.; Xu, Y.; Tang, T.; Weng, J.; Li, X.; Wang, L. Role of pyridinic-N for nitrogen-doped graphene quantum dots in oxygen reaction reduction. J. Colloid Interface Sci. 2017, 508, 154–158. [CrossRef] [PubMed] 31. Dong, J.; Wang, K.; Sun, L.; Sun, B.; Yang, M.; Chen, H.; Wang, Y.; Sun, J.; Dong, L. Application of graphene quantum dots for simultaneous ﬂuorescence imaging and tumor-targeted drug delivery. Sens. Actuators B Chem. 2018, 256, 616–623. [CrossRef] 32. Wang, L.; Li, W.; Wu, B.; Li, Z.; Pan, D.; Wu, M. References Room-temperature synthesis of graphene quantum dots via electron-beam irradiation and their application in cell imaging. Chem. Eng. J. 2017, 309, 374–380. [CrossRef] 33. Wang, L.; Wu, B.; Li, W.; Li, Z.; Zhan, J.; Geng, B.; Wang, S.; Pan, D.; Wu, M. Industrial production of ultra-stable sulfonated graphene quantum dots for Golgi apparatus imaging. J. Mater. Chem. B 2017, 5, 5355–5361. [CrossRef] 34. Wang, L.; Li, W.; Li, M.; Su, Q.; Li, Z.; Pan, D.; Wu, M. Ultrastable amine, sulfo cofunctionalized graphene quantum dots with high two-photon ﬂuorescence for cellular imaging. ACS Sustain. Chem. Eng. 2018, 6, 4711–4716. [CrossRef] 548 Catalysts 2018, 8, 438 35. Li, W.; Li, M.; Liu, Y.; Pan, D.; Li, Z.; Wang, L.; Wu, M. Three-minute ultrarapid microwave-assisted synthesis of bright ﬂuorescent graphene quantum dots for live cell staining and white LEDs. ACS Appl. Nano Mater. 2018, 1, 1623–1630. [CrossRef] 36. Li, L.; Yan, X. Colloidal graphene quantum dots. J. Phys. Chem. Lett. 2010, 1, 2572–2576. [CrossRef] 37. Rajender, G.; Kumar, J.; Giri, P.K. Interfacial charge transfer in oxygen deﬁcient TiO2-graphene quantum dot hybrid and its inﬂuence on the enhanced visible light photocatalysis. Appl. Catal. B Environ. 2018, 224, 960–972. [CrossRef] 38. Safardoust-Hojaghan, H.; Salavati-Niasari, M. Degradation of methylene blue as a pollutant with N-dop graphene quantum dot/titanium dioxide nanocomposite. J. Clean. Prod. 2017, 148, 31–36. [CrossRef] 39. Tian, H.; Shen, K.; Hu, X.; Qiao, L.; Zheng, W. N, S co-doped graphene quantum dots-graphene-TiO2 nanotubes composite with enhanced photocatalytic activity. J. Alloys Compd. 2017, 691, 369–377. [CrossRef] nanotubes composite with enhanced photocatalytic activity. J. Alloys Compd. 2017, 691, 369–377. [CrossRef] 40. Zhang, B.-X.; Gao, H.; Li, X.-L. Synthesis and optical properties of nitrogen and sulfur co-doped graphene quantum dots. New J. Chem. 2014, 38, 4615–4621. [CrossRef] p p y y y p 40. Zhang, B.-X.; Gao, H.; Li, X.-L. Synthesis and optical properties of nitrogen and sulfur co-doped graphene quantum dots. New J. Chem. 2014, 38, 4615–4621. [CrossRef] q 41. Wang, Y.; Zhang, L.; Liang, R.-P.; Bai, J.-M.; Qiu, J.-D. Using graphene quantum dots as photoluminescent b f t i ki i A l Ch 2013 85 9148 9155 [C R f] [P bM d] 41. Wang, Y.; Zhang, L.; Liang, R.-P.; Bai, J.-M.; Qiu, J.-D. Using graphene quantum dots as photoluminescent probes for protein kinase sensing. Anal. Chem. 2013, 85, 9148–9155. [CrossRef] [PubMed] 42. Taibi, M.; Ammar, S.; Jouini, N.; Fiévet, F.; Molinié, P.; Drillon, M. Improving Interfacial Charge-Transfer Transitions in Nb-Doped TiO2 Electrodes with 7,7,8,8-Tetracyanoquinodimethane Reo Eguchi 1,2,, Yuya Takekuma 1,2, Tsuyoshi Ochiai 2,3,4 and Morio Nagata 1, 1 Graduate School of Engineering, Tokyo University of Science, 12-1, Ichigayafunagawara, Shinjuku-ku, Tokyo 162-0826, Japan; tytakekuma@gmail.com Tokyo 162 0826, Japan; tytakekuma@gmail.com 2 Photocatalyst Group, Research and Development Department, Local Independent Administrative Agency Kanagawa Institute of industrial Science and TEChnology (KISTEC), 407 East Wing, Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan; pg-ochiai@newkast.or.jp 3 Materials Analysis Group, Kawasaki Technical Support Department, KISTEC, Ground Floor East Wing, Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan 4 Photocatalysis International Research Center, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan y , J p ; y g 2 Photocatalyst Group, Research and Development Department, Local Independent Administrative Agency Kanagawa Institute of industrial Science and TEChnology (KISTEC), 407 East Wing, Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan; pg-ochiai@newkast.or.jp y J p y g 2 Photocatalyst Group, Research and Development Department, Local Independent Administrative Agency Kanagawa Institute of industrial Science and TEChnology (KISTEC), 407 East Wing, Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan; pg-ochiai@newkast.or.jp 3 Materials Analysis Group, Kawasaki Technical Support Department, KISTEC, Ground Floor East Wing, Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan y p y g 2 Photocatalyst Group, Research and Development Department, Local Independent Administrative Agen Kanagawa Institute of industrial Science and TEChnology (KISTEC), 407 East Wing, Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan; pg-ochiai@newkast.or.j Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan; pg-ochiai@newkast.or.j 3 Materials Analysis Group, Kawasaki Technical Support Department, KISTEC, Ground Floor East Wing Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan 4 Photocatalysis International Research Center Tokyo University of Science 2641 Yamazaki Noda Innovation Center Building, KSP, 3-2-1 Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0012, Japan 4 Photocatalysis International Research Center, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan * Correspondence: rikadai2018@gmail.com (R.E.); nagata@ci.kagu.tus.ac.jp (M.N.); l ( ) ( ) rrespondence: rikadai2018@gmail.com (R.E.); nagata@ * Correspondence: rikadai2018@gmail.com (R.E.); nagata@ci.kagu.tus.ac.jp (M.N.); Tel.: +81-3-5228-8311 (R.E. & M.N.); Fax: +81-3-5261-4631 (R.E. & M.N.) p g ( ); g g jp ( ); Tel.: +81-3-5228-8311 (R.E. & M.N.); Fax: +81-3-5261-4631 (R.E. Improving Interfacial Charge-Transfer Transitions in Nb-Doped TiO2 Electrodes with 7,7,8,8-Tetracyanoquinodimethane & M.N.) Received: 25 July 2018; Accepted: 28 August 2018; Published: 30 August 2018 Abstract: Interfacial charge-transfer (ICT) transitions involved in charge-separation mechanisms are expected to enable efﬁcient photovoltaic conversions through one-step charge-separation processes. With this in mind, the charge-transfer complex fabricated from TiO2 nanoparticles and 7,7,8,8-tetracyanoquinodimethane (TCNQ) has been applied to dye-sensitized solar cells. However, rapid carrier recombination from the conduction band of TiO2 to the highest occupied molecular orbital (HOMO) of TCNQ remains a major issue for this complex. In this study, to inhibit surface- complex recombinations, we prepared Nb-doped TiO2 nanoparticles with different atomic ratios for enhanced electron transport. To investigate the effects of doping on electron injection through ICT transitions, these materials were examined as photoelectrodes. When TiO2 was doped with 1.5 mol % Nb, the Fermi level of the TiO2 electrode shifted toward the conduction band minimum, which improved electron back-contact toward the HOMO of TCNQ. The enhancement in electron transport led to increases in both short circuit current and open circuit voltage, resulting in a slight (1.1% to 1.3%) improvement in photovoltaic conversion efﬁciency compared to undoped TiO2. Such control of electron transport within the photoelectrode is attributed to improvements in electron injection through ICT transitions. Keywords: photovoltaic conversion; interfacial charge-transfer transition; 7,7,8,8-tetracyanoquinodimethane; Nb-doped TiO2 Catalysts 2018, 8, 367; doi:10.3390/catal8090367 www.mdpi.com/journal/catalysts References Layered nickel hydroxide salts: Synthesis, characterization and magnetic behaviour in relation to the basal spacing. J. Mater. Chem. 2002, 12, 3238–3244. [CrossRef] 43. Xu, Y.; Mo, Y.; Tian, J.; Wang, P.; Yu, H.; Yu, J. The synergistic effect of graphitic N and pyrrolic N for the enhance photocatalytic performance of nitrogen-doped graphene/TiO2 nanocomposites. Appl. Catal. B Environ. 2016, 181, 810–817. [CrossRef] 44. Dong, Y.; Chen, C.; Zheng, X.; Gao, L.; Cui, Z.; Yang, H.; Guo, C.; Chi, Y.; Li, C.M. One-step and high yield simultaneous preparation of single- and multi-layer graphene quantum dots from CX-72 carbon black. J. Mater. Chem. 2012, 22, 8764–8766. [CrossRef] 45. Ye, R.; Xiang, C.; Lin, J.; Peng, Z.; Huang, K.; Yan, Z.; Cook, N.P.; Samuel, E.L.G.; Hwang, C.-C.; Ruan, G.; et al. Coal as an abundant source of graphene quantum dots. Nat. Commun. 2013, 4, 2943. [CrossRef] [PubMed] 46. Melnikov, D.V.; Chelikowsky, J.R. Quantum conﬁnement in phosphorus-doped silicon nanocrystals. P Rev. Lett. 2004, 92, 046802. [CrossRef] [PubMed] , , [ ] [ ] 47. Peng, J.; Gao, W.; Gupta, B.K.; Liu, Z.; Aburto, R.-R.; Ge, L.; Song, L.; Alemany, L.B.; Zhan, X.; Gao, G.; et al. Graphene quantum dots derived from carbon ﬁbers Nano Lett 2012 12 844 849 [CrossRef] [PubMed] 47. Peng, J.; Gao, W.; Gupta, B.K.; Liu, Z.; Aburto, R.-R.; Ge, L.; Song, L.; Alemany, L.B.; Zhan, X.; Gao, G.; e Graphene quantum dots derived from carbon ﬁbers. Nano Lett. 2012, 12, 844–849. [CrossRef] [PubMed] eng, J.; Gao, W.; Gupta, B.K.; Liu, Z.; Aburto, R. R.; Ge, L.; Song, L.; Alemany, L.B.; Zhan, X.; Gao, G.; et al. raphene quantum dots derived from carbon ﬁbers. Nano Lett. 2012, 12, 844–849. [CrossRef] [PubMed] 48. Wang, L.; Wang, Y.; Xu, T.; Liao, H.; Yao, C.; Liu, Y.; Li, Z.; Chen, Z.; Pan, D.; Sun, L.; Wu, M. Gram-scale synthesis of single-crystalline graphene quantum dots with superior optical properties. Nat. Commun. 2014, 5, 5357. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 549 catalysts catalysts 1. Introduction Interfacial charge-transfer (ICT) transitions between inorganic semiconductors and π-conjugated organic compounds are characteristic electronic transitions that enable direct photoinduced charge separation. Due to this feature, ICT transitions are applicable to photovoltaic conversions [1]. To date, dicyanomethylene-based compounds, such as tetracyanoethylene (TCNE) and 7,7,8,8- tetracyanoquinodimethane (TCNQ), form surface complexes with TiO2 that have been reported to absorb visible light due to ICT transitions from the π-conjugated system to the conduction band of TiO2 [2–4]. Although photovoltaic effects due to direct electron injection have been investigated using these surface complexes as photoanodes in photoelectrochemical cells, the photovoltaic Catalysts 2018, 8, 367; doi:10.3390/catal8090367 551 Catalysts 2018, 8, 367 conversion efﬁciencies under AM1.5 illumination are quite low (under 2%) when compared to those of dye-sensitized solar cells (DSSCs) [5]. To date, photovoltaic conversions and electron injections between surface complexes, such as bis(dicyanomethylene) compounds (TCNX) [TCNE, TCNQ, and 11,11,12,1 2-tetracyanonaphtho-2, 6-quinodimethane (TCNAQ)] and TiO2, have been studied theoretically by Fujisawa et al. using density functional theory (DFT) on the basis of Marcus theory, which revealed that the structure and formation mechanism of the surface complex need to be considered to control interfacial electronic transitions and carrier recombinations by adjusting the electron afﬁnity of TCNX [6–8]. In addition, they also demonstrated that carrier recombinations from the conduction band of TiO2 to the highest occupied molecular orbital (HOMO) levels of these compounds occur quite rapidly, which, as geminate recombinations, are more severe than in DSSCs. Hence, the rapid carrier recombinations of surface complexes hinder the use of ICT transitions in photovoltaic conversion [9] and is the most signiﬁcant problem faced. To overcome rapid electron recombination, the properties of surface complexes require further investigation through different approaches that include investigating the interactions between TCNX and modiﬁed-TiO2 photoanodes. Zaban et al. studied the suppression of electron recombination in surface complexes aided by a thin SrTiO3 coating layer on TiO2 that improved electron injection and electron transport [10]. On the other hand, many researchers reported that the TiO2 photoanode in a DSSC is one of the most important components affecting photovoltaic performance because it acts as the support for dye molecules as well as the electron-transport region. Furthermore, TiO2 can, in principal, be n-type-doped to enhance charge collection and electron-transport efﬁciency within the TiO2 layer [11]. 1. Introduction Among n-type-doped TiO2 systems, Nb-doped TiO2 photoanodes have been studied to improve electron conductance and injection because Nb has one more electron than Ti (IV) [12–18]. Recently, Lin et al. reported the inﬂuence of TiO2 doped with Group V-b metal atoms on the photovoltaic performance of dye-sensitized solar cells [18]. Although V, Nb, and Ta belong to the same group (V-b) and have one more electron than Ti, DSSCs based on Nb-doped TiO2 showed the best photovoltaic performance as a result of the creation of donor levels, which increased the concentration of the carriers. According to this study, the charge transport and conductivity for Nb-doped TiO2 were superior to those of V- and Ta-doped TiO2. Therefore, Nb is a superior doping element. In this study, to inhibit rapid electron recombinations in surface complexes, we examined the effects of Nb-doped TiO2 electrodes with TCNQ on photovoltaic performance. The effects of different amounts of Nb-doping on the photovoltaic properties of surface complexes were evaluated by means of incident photon-to-current efﬁciency (IPCE) spectroscopy and by acquiring current-density–voltage (J-V) curves. TiO2 doped with 1.5 mol % Nb exhibited improved Jsc and Voc values, resulting in a 17% improvement in photoconversion efﬁciency compared to undoped TiO2. 2.1. Structural Characterization of Nb-Doped TiO2 Nb-doped TiO2 nanoparticles were formed by a hydrothermal method. Figure 1 displays the X-ray diffraction (XRD) patterns of undoped and Nb-doped TiO2 samples with varying Nb contents. The anatase and rutile phases of TiO2 are distinct, and the intensity of the peak corresponding to the anatase phase increased with increasing Nb content in the sample, while that of the rutile phase decreased (Figure 1a). Hence, doping the TiO2 lattice with Nb was observed to enhance the growth of the anatase TiO2 phase while hindering the formation of the rutile TiO2 phase [19]. The average crystallite size of each phase was calculated by the Scherrer-equation [20]: D = kλ B cos θ (1) D = kλ B cos θ (1) 552 Catalysts 2018, 8, 367 where D is the crystallite size, k is a dimensionless shape factor that has a typical value of 0.94, λ is the Cu-Kα X-ray wavelength (1.5406 Å), θ is the Bragg angle in degrees, and B is the full-width-at-half-maximum (FWHM) of the peak. The anatase and rutile crystallinities were determined from their (101) and (110) peak heights, respectively; crystallite size are listed in Table S1. Incorporation of the Nb dopant into the TiO2 structure typically decreased crystallite growth; indeed, the crystallite size of the anatase phase slowly decreased from 10.1 nm to 8.8 nm with increased doping, led by the effect of Nb on nucleation following doping into the TiO2 lattice. A similar effect of different oxide additives on the TiO2 phase transformation was reported by Yanagida et al. [19]. In addition, the diffraction peaks shifted to lower θ values with increasing Nb content as a result of the larger radius of Nb5+ (0.64 Å) compared to that of Ti4+ (0.61 Å) [15], in accordance with the Bragg equation: λ = 2dsinθ (Figure 1b). Figure 1. (a) X-ray diffraction (XRD) patterns of TiO2 with varying Nb contents (A: Anatase, R: Rutile). (b) XRD patterns between 2θ values of 40◦and 60◦. Figure 1. (a) X-ray diffraction (XRD) patterns of TiO2 with varying Nb contents (A: Anatase, R: Rutile). (b) XRD patterns between 2θ values of 40◦and 60◦. Figure 2 displays the X-ray photoelectron spectroscopy (XPS) spectra of the undoped and Nb-doped TiO2 samples with varying Nb contents. Nb 3d3/2 and 3d5/2 peaks are evident in the spectra of the Nb-doped TiO2 samples, and their intensities increased with increasing Nb content (Figure 2a). 2.1. Structural Characterization of Nb-Doped TiO2 The Nb 3d3/2 and 3d5/2 peaks are located at binding energies of 209 and 206 eV, respectively. Nb-doping also caused the Ti 2p1/2 and 2p3/2 peaks to shift slightly toward higher binding energies, which is attributable to the higher electronegativity of Nb (1.6) compared to that of Ti (1.54) (Figure 2b) [15]. The peaks in the O 1s spectra correspond to Ti4+–O bonds; the positions of these peaks show similar trends to those observed for Ti due to increases in both lattice oxygen and Nb5+–O bonds in these samples. Figure 2d displays a double-band structure, with the main peak (29.4 eV) related to O 1s electron binding in TiO2 and the other peak, at a higher binding energy (531 eV), attributed to OH groups on the surfaces of these samples. Typically, OH groups are beneﬁcial for TCNQ anchoring [21]. 553 Catalysts 2018, 8, 367 Figure 2. High-resolution (a) Nb 3d, (b) Ti 2p, and (c) O 1s X-ray photoelectron spectroscopy (XPS) spectra of TiO2 with varying Nb contents. (d) O 1s spectral ﬁtting, revealing the presence of peaks related to OH groups. Figure 2. High-resolution (a) Nb 3d, (b) Ti 2p, and (c) O 1s X-ray photoelectron spectroscopy (XPS) spectra of TiO2 with varying Nb contents. (d) O 1s spectral ﬁtting, revealing the presence of peaks related to OH groups. 2.2. Optical Properties of Nb-Doped TiO2 The energy of the CBM of the 1.5 mol % Nb-doped TiO2 was found to be 0.37 eV higher than that of the Fermi level, while that of the undoped TiO2 was 0.46 eV higher (Table 1). The CBM of the sample containing 1.5 mol % Nb was therefore 0.09 eV lower in energy than that of undoped TiO2. The observed lowering of the CBM of the sample containing 1.5 mol % and 3.0 mol % Nb due to donor levels induced by these dopants [18] is attributable to higher electron transitions between the Nb-doped TiO2 electrode and the anchoring TCNQ. On the other hand, the CBM for the sample containing 5.0 mol % Nb exhibited a large offset between its CBM energy and EF, compared to the undoped TiO2, due to intra-band transport [15]. (VBMs) were directly determined from the photoemission spectra by linear extrapolation of the onsets of the valence-band emissions [12]. In the case of the undoped TiO2 sample, the VBM was found to be located 2.6 eV below the Fermi level. With the optical bandgap determined to be 3.06 eV, we deduce that EF is 0.46 eV lower than the energy of the conduction band minimum (ECBM). The calculated positions of the conduction band minimum (CBM) with respect to the Fermi level are summarized in Table 1. The energy of the CBM of the 1.5 mol % Nb-doped TiO2 was found to be 0.37 eV higher than that of the Fermi level, while that of the undoped TiO2 was 0.46 eV higher (Table 1). The CBM of the sample containing 1.5 mol % Nb was therefore 0.09 eV lower in energy than that of undoped TiO2. The observed lowering of the CBM of the sample containing 1.5 mol % and 3.0 mol % Nb due to donor levels induced by these dopants [18] is attributable to higher electron transitions between the Nb-doped TiO2 electrode and the anchoring TCNQ. On the other hand, the CBM for the sample containing 5.0 mol % Nb exhibited a large offset between its CBM energy and EF, compared to the undoped TiO2, due to intra-band transport [15]. Figure 3. XPS spectra and valence-band maxima of the Nb-doped TiO2 with Nb contents of (a) 0, (b) 1.5, (c) 3.0 mol %, and (d) 5.0 mol %. 2.2. Optical Properties of Nb-Doped TiO2 Examination of the electronic structures of the Nb-doped TiO2 electrodes is essential to understand the ICT transition from the highest occupied molecular orbital (HOMO) of TCNQ to the conduction band of TiO2. XPS and calculated Eg values for the valence band can provide electronic-potential information. The reﬂectance spectra displayed in Figure S1a enable the bandgaps of the samples to be determined. Eg values were calculated after converting the reﬂectance data into the equivalent absorption coefﬁcients using Equation (1), as shown in Figure S1b and listed in Table 1. Eg increased from 3.06 eV to 3.14 eV as the Nb content was increased from 0 mol % to 5.0 mol %, which is ascribable to strong hybridization between the Ti 3d and Nb 3d states that forms a d-type conduction band [22]. The observed increase in bandgap with increasing Nb content can be also explained by the decrease in the amount of rutile phase TiO2 formed (Table S1); indeed, the bandgap of anatase is 3.2 eV, which is larger than that of rutile (3.0 eV). Table 1. Optical bandgap values for TiO2 samples with different Nb contents. Nb Content (mol %) 0 1.5 3.0 5.0 Eg (eV) 3.06 3.07 3.11 3.14 ECBM–EF (eV) 0.46 0.37 0.41 0.64 Table 1. Optical bandgap values for TiO2 samples with different Nb contents. Figure 3 shows the photoemission spectra of TiO2 samples with different Nb contents, in which the binding energies are referenced against the Fermi energy level (EF). Two peaks are evident in each photoemission spectrum, one centered at 6.4 eV and the other at 8.2 eV. These peaks are attributed to emissions from π- and σ-type O 2p orbitals, respectively. The positions of the valence band maxima 554 Catalysts 2018, 8, 367 (VBMs) were directly determined from the photoemission spectra by linear extrapolation of the onsets of the valence-band emissions [12]. In the case of the undoped TiO2 sample, the VBM was found to be located 2.6 eV below the Fermi level. With the optical bandgap determined to be 3.06 eV, we deduce that EF is 0.46 eV lower than the energy of the conduction band minimum (ECBM). The calculated positions of the conduction band minimum (CBM) with respect to the Fermi level are summarized in Table 1. 2.2. Optical Properties of Nb-Doped TiO2 The Gaussian ﬁtted electron emissions from the π (green) and σ (purple) O 2p orbitals are also shown in (a). Figure 3. XPS spectra and valence-band maxima of the Nb-doped TiO2 with Nb contents of (a) 0, (b) 1.5, (c) 3.0 mol %, and (d) 5.0 mol %. The Gaussian ﬁtted electron emissions from the π (green) and σ (purple) O 2p orbitals are also shown in (a). 2.3. Photovoltaic Performance of Interfacial Charge-Transfer (ICT) Photoconversion Devices The photovoltaic performance of ICT-transition devices based on the undoped and Nb-doped TiO2 electrodes with varying Nb contents under solar illumination (AM 1.5 G, 100 mW/cm2) are displayed in Figure 4a, with the photovoltaic performance parameters summarized in Table 2. Jsc was observed to increase from 4.5 mA/cm2 to 5.7 mA/cm2 at an Nb content of 3.0 mol %, which is attributable to improved electronic transitions between the nanoparticles and the anchoring TCNQ associated with the negative CBM-energy shift (Table 1). On the other hand, Voc increased slightly, from 0.40 V to 0.41 V, as the dopant content was increased to 1.5 mol % due to the increased gap between EF and the 555 Catalysts 2018, 8, 367 redox potential of the I−/I3−couple. This is also attributable to Nb-doped TiO2-surface passivation, according to electrochemical impedance spectroscopy (EIS) (Nyquist plots, Figure S2) [23,24]. However, increased Nb doping has a negative inﬂuence on photovoltaic behavior at levels over 5.0 mol %; indeed, Voc decreased from 0.40 V to 0.36 V, which is ascribable to a detrimental conduction-band-edge effect. Indeed, the XRD result shows the decrease in rutile crystallization in 5.0 mol % Nb-doped TiO2 (Table S1) led to defects in crystallization, and affected electron recombination such the degradation of Voc parameter. Figure 4. (a) Photocurrent-density–voltage curves and (b) incident photon-to-current conversion efﬁciency spectra of Interfacial charge-transfer (ICT) photoconversion devices based on the undoped and Nb-doped TiO2 electrodes prepared in this study. Figure 4. (a) Photocurrent-density–voltage curves and (b) incident photon-to-current conversion efﬁciency spectra of Interfacial charge-transfer (ICT) photoconversion devices based on the undoped and Nb-doped TiO2 electrodes prepared in this study. Table 2. Photovoltaic-performance parameters of Interfacial charge-transfer (ICT) photoconversion devices based on the undoped and Nb-doped TiO2 electrodes prepared in this study. Table 2. Photovoltaic-performance parameters of Interfacial charge-transfer (ICT) photoconversion devices based on the undoped and Nb-doped TiO2 electrodes prepared in this study. Table 2. 2.2. Optical Properties of Nb-Doped TiO2 Photovoltaic-performance parameters of Interfacial charge-transfer (ICT) photoconversion devices based on the undoped and Nb-doped TiO2 electrodes prepared in this study. Sample Jsc (mA/cm2) Voc (V) FF (%) H (%) Nb 0 mol % 4.5 0.40 63 1.1 Nb 1.5 mol % 5.5 0.41 59 1.3 Nb 3.0 mol % 5.7 0.38 57 1.2 Nb 5.0 mol % 4.4 0.36 56 0.87 Figure 4b reveals that the device containing the 1.5 mol % Nb-doped TiO2 electrode exhibits enhanced IPCE compared to that of the undoped electrode. Indeed, the absorption edge was observed to shift from 730 nm to 750 nm. Nb has been reported to create oxygen vacancies in TiO2 that act as active sites, resulting in a photoresponse red shift [25]. However, the lowest IPCE was obtained in the device containing the 5.0 mol % Nb-doped TiO2 electrode, as revealed by the J-V curves. According to previous studies on the fabrication of DSSCs based on TiO2 doped with Nb at concentrations up to 5.0 mol %, Feng et al. and Huang et al. reported that the highest photovoltaic efﬁciency was observed for Nb concentrations of 2.0 mol % and 5.0 mol %, respectively [15,16]. However, our results demonstrated that Nb-doping of 1.5 mol % produced the highest efﬁciency. Considering the behavior of Jsc with doping, we can conclude that the optimum Nb concentration for reaching the best efﬁciency is in the range of 1.5–3.0 mol %. In addition, the error bar of photovoltaic efﬁciency in this study is shown in Figure S3. 556 Catalysts 2018, 8, 367 3.2. Cell Fabrication The Nb-doped TiO2 pastes were synthesized following the procedure reported by Ito et al. [26]. The ﬁnal screen-printing pastes correspond to 18 wt % TiO2, 9 wt % ethyl cellulose and 73 wt % terpineol. Two kinds of pure ethyl cellulose (EC) powders, i.e., EC (10 mPas, Kanto Chemical Co., Tokyo, Japan) and EC (45 mPas, Kanto Chemical Co., Tokyo, Japan) were dissolved before usage in ethanol to yield 10 wt % solutions. Then, 0.325 g of EC (10 mPas) and 0.175 g of EC (45 mPas) of these 10 wt % ethanolic mixtures were added to a round-bottomed rotovap ﬂask containing 1 g pure TiO2 (obtained from a previously prepared precipitate) and 4.05 g of terpineol (Kanto Chemical Co., Tokyo, Japan), and diluted with approximately 100 ml of ethanol. This mixture was then ultrasonicated for 48 h. Ethanol and water were removed from these TiO2/ethyl cellulose solutions using a rotary-evaporator. The ﬁnal formulations of the pastes were made with a three-roll mill (Exakt, Nagase Screen Printing Research Co., Aichi, Japan). The Nb-doped TiO2 electrodes were fabricated by screen-printing pastes onto glass substrates coated with transparent conducting F-doped SnO2 (FTO), Nippon Sheet Glass Co., Tokyo, Japan) with a sheet resistance of 10 Ω sq−1, followed by sintering at 500 ◦C for 30 min. The thickness of the TiO2 electrode was set to 8 μm. The TiO2 electrodes were immersed in a 1 mM solution of TCNQ (Tokyo Kasei Kogyo Co., Tokyo, Japan) in acetonitrile at 60 ◦C for 24 h. Photovoltaic cells were fabricated using the TCNQ-treated TiO2 electrode (active area: 4 mm × 4 mm), a Pt-sputtered FTO glass counter electrode (Geomatec Co., Kanagawa, Japan), an I−/I3−redox couple electrolyte (1 M LiI (Sigma-Aldrich Co., St. Louis, MO, USA) and 0.025 M I2 (Kanto Chemical Co.) in acetonitrile), and a spacer ﬁlm (thickness: 30 μm). 3.1. Preparation of Nb-Doped TiO2 3.1. Preparation of Nb-Doped TiO2 The Nb-doped TiO2 nanoparticles were synthesized following the procedure described by Nikolay et al. [12]. These syntheses used Ti and Nb precursors and hydrothermal processes. To prepare Nb-doped TiO2 (0, 1.5, 3.0, or 5.0 mol %) nanoparticles, 16.4 mL of titanium tetraisopropoxide (Kanto Chemical Co., Tokyo, Japan) was mixed with 0, 60, 120, or 200 μL of niobium ethoxide (Wako Co., Tokyo, Japan), respectively, after which 2.64 mL of acetic acid (Kanto Chemical Co., Tokyo, Japan) was added under stirring condition with a Teﬂon stirrer blade for 15 min at room temperature. The mixture was dropped into 68.4 mL of deionized water while stirred at about 800 rpm. After stirring for 1 h, 2.35 mL of 65% nitric acid (Kanto Chemical Co., Tokyo, Japan) was added to the solution. The solution temperature was increased to 80 ◦C over 40 min and then held at 80 ◦C for 80 min under reﬂux conditions with intensive stirring. The nanoparticles were hydrothermally grown using the prepared colloidal solution in a Teﬂon-lined mini-autoclave at 180 ◦C for 12 h, after which 0.52 mL of 65% nitric acid was added to the colloidal solution, followed by ultrasonication with stirring for 1 h. The prepared mixture was ﬁnally washed three times with ethanol by centrifugation. 3. Experimental Section 3.1. Preparation of Nb-Doped TiO2 3.1. Preparation of Nb-Doped TiO2 3.3. Characterization Incident photon-to-current efﬁciency (IPCE) spectra were acquired using a Hypermonolight system (M10, Bunkoukeiki Co., Tokyo, Japan) with a calibrated silicon photodiode (Bunkoukeiki Co., Tokyo, Japan). Current-density–voltage (J-V) curves were recorded using a potentiostat (1287A potentiostat/galvanostat, Ametek Co., St. Berwyn, PA, USA) under 100 mW/cm2 AM 1.5 G simulated sunlight produced by a solar simulator (Yamashita Denso Co., Tokyo, Japan). Electrochemical impedance spectroscopy (EIS) was performed using a potentiostat equipped with calculation software (1255B frequency-response analyzer, Ametek Co, Tokyo, Japan). The thicknesses of the ﬁlms were measured using a surface roughness proﬁlometer (SURFCOM1440D, Accretech Co., Tokyo, Japan). Crystal structures were determined by X-ray diffraction (XRD, Ultima X-ray diffractometer, Rigaku Co., Yamanashi, Japan). The electronic structures and chemical states of the TiO2 electrodes were investigated by X-ray photoelectron spectroscopy (XPS) (JPS-9010MC, 557 Catalysts 2018, 8, 367 Catalysts 2018, 8, 367 Nihondensi Co., Tochigi, Japan). The binding energies were calibrated against the C 1s peak at 284.60 eV. For all XPS measurements, undoped TiO2 and Nb-doped TiO2 were deposited on carbon sheets. UV-vis spectroscopy was performed using a U-3900H spectrometer (Hitachi Co., Tokyo, Japan) in reﬂectance mode, and the spectra were analyzed using the Kubelka-Munk formalism to convert reﬂectance into the equivalent absorption coefﬁcient, αKM [27–29]: αKM = (1 −R∞)2 2R∞ (2) (2) where: R∞is the reﬂectance of an inﬁnitely thick sample with respect to the reference at each wavelength. 4. Conclusions In this study, we examined the abilities of Nb-doped TiO2 electrodes with 7,7,8,8- tetracyanoquinodimethane (TCNQ) to inhibit electron recombinations in surface-complexes. The 1.5 mol % Nb-doped TiO2 electrode exhibited improved photovoltaic performance and superior short circuit current and open circuit voltage, resulting in 1.3% photoconversion efﬁciency, which is 17% higher than that of the undoped photoelectrode. This improvement is ascribed to enhanced electron injection resulting from a shift in the Fermi level of the TiO2 electrode toward the conduction band minimum, and the effect of passivation, as revealed by Electrochemical impedance spectroscopy. As expected, these experimental data for Nb-doped TiO2 with TCNQ reveal that semiconductor modiﬁcation can be used to achieve efﬁcient photovoltaic conversion through Interfacial charge-transfer transitions by suppressing surface-complex carrier recombinations and improving electron transport. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/8/9/367/s1. Table S1: Crystallite sizes of TiO2 samples with various Nb contents; Table S2: Fitted EIS spectra of TiO2 samples with varying Nb contents. Figure S1: (a) Reﬂectance spectra and (b–e) optical bandgaps of TiO2 samples with varying Nb contents; Figure S2: Electrochemical impedance spectra (Nyquist plots) of the undoped and Nb-doped TiO2 electrodes; Figure S3: Error bars for Conversion efﬁciency of DSSCs employing different Nb dopants. Author Contributions: R.E., Y.T., T.O., and M.N. participated in the study design and conducted the experiments. Data were collected and analyzed by R.E. and Y.T. The manuscript was written by R.E., M.N. and T.O. provided valuable input and advice regarding the manuscript. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂicts of interest. 5. Manzhos, S.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Isotopic Substitution as a Strategy to Control Non-Adiabatic Dynamics in Photoelectrochemical Cells: Surface Complexes between TiO2 and Dicyanomethylene Compounds. J. Appl. Phys. 2012, 51, 10NE03. [CrossRef] References 1. Fujisawa, J.-I. Large Impact of Reorganization Energy on Photovoltaic Conversion Due to Interfacial 1. Fujisawa, J.-I. Large Impact of Reorganization Energy on Photovoltaic Conversion Due to Interfacial 1. Fujisawa, J.-I. Large Impact of Reorganization Energy on Photovoltaic Conversion Due to Interfacial Charge-Transfer Transitions. Phys. Chem. Chem. Phys. 2015, 17, 12228–12237. [CrossRef] [PubMed] Charge-Transfer Transitions. Phys. Chem. Chem. Phys. 2015, 17, 12228–12237. [CrossRef] [PubMed] 2. Xagas, A.P.; Bernard, M.C.; Hugot-Le Goff, A.; Spyrellis, N.; Loizos, Z.; Falaras, P. Surface Modiﬁcation and Photosensitisation of TiO2 Nanocrystalline Films with Ascorbic Acid. J. Photochem. Photobiol. A 2000, 132, 115–120. [CrossRef] 2. Xagas, A.P.; Bernard, M.C.; Hugot-Le Goff, A.; Spyrellis, N.; Loizos, Z.; Falaras, P. Surface Modiﬁcation and Photosensitisation of TiO2 Nanocrystalline Films with Ascorbic Acid. J. Photochem. Photobiol. A 2000, 132, 115–120. [CrossRef] 3. Tae, E.L.; Lee, S.H.; Lee, J.K.; Yoo, S.S.; Kang, E.J.; Yoon, K.B. A Strategy to Increase the Efﬁciency of the Dye-Sensitized TiO2 Solar Cells Operated by Photoexcitation of Dye-to-TiO2 Charge-Transfer Bands. J. Phys. Chem. B 2005, 109, 22513–22522. [CrossRef] [PubMed] 3. Tae, E.L.; Lee, S.H.; Lee, J.K.; Yoo, S.S.; Kang, E.J.; Yoon, K.B. A Strategy to Increase the Efﬁciency of the Dye-Sensitized TiO2 Solar Cells Operated by Photoexcitation of Dye-to-TiO2 Charge-Transfer Bands. J. Phys. Chem. B 2005, 109, 22513–22522. [CrossRef] [PubMed] 4. Manzhos, S.; Jono, R.; Yamashita, K.; Fujisawa, J.-I.; Nagata, M.; Segawa, H. Study of Interfacial Charge Transfer Bands and Electron Recombination in the Surface Complexes of TCNE, TCNQ, and TCNAQ with TiO2. J. Phys. Chem. C 2011, 115, 21487–21493. [CrossRef] 4. Manzhos, S.; Jono, R.; Yamashita, K.; Fujisawa, J.-I.; Nagata, M.; Segawa, H. Study of Interfacial Charge Transfer Bands and Electron Recombination in the Surface Complexes of TCNE, TCNQ, and TCNAQ with TiO2. J. Phys. Chem. C 2011, 115, 21487–21493. [CrossRef] 5. Manzhos, S.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Isotopic Substitution as a Strategy to Control Non-Adiabatic Dynamics in Photoelectrochemical Cells: Surface Complexes between TiO2 and Dicyanomethylene Compounds. J. Appl. Phys. 2012, 51, 10NE03. [CrossRef] 5. Manzhos, S.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Isotopic Substitution as a Strategy to Control Non-Adiabatic Dynamics in Photoelectrochemical Cells: Surface Complexes between TiO2 and Dicyanomethylene Compounds. J. Appl. Phys. 2012, 51, 10NE03. [CrossRef] 558 Catalysts 2018, 8, 367 6. Fujisawa, J.-I.; Nagata, M.; Hanaya, M. Charge-Transfer Complex Versus [Sigma]-Complex Formed between TiO2 and Bis (Dicyanomethylene) Electron Acceptors. Phys. Chem. Chem. Phys. 2015, 17, 27343–27356. [CrossRef] [PubMed] 7. References Michinobu, T.; Satoh, N.; Cai, J.; Li, Y.; Han, L. Novel Design of Organic Donor-Acceptor Dyes without Carboxylic Acid Anchoring Groups for Dye-Sensitized Solar Cells. J. Mater. Chem. C 2014, 2, 3367–3372. [CrossRef] 8. Fujisawa, J.-I.; Hanaya, M. Electronic Structures of TiO2–TCNE, –TCNQ, and –2,6-TCNAQ Surface Complexes Studied by Ionization Potential Measurements and DFT Calculations: Mechanism of the Shift of Interfacial Charge-Transfer Bands. Chem. Phys. Lett. 2016, 653, 11–16. [CrossRef] 9. Wang, Y.; Hang, K.; Anderson, N.A.; Lian, T. Comparison of Electron Transfer Dynamics in Molecule- to-Nanoparticle and Intramolecular Charge Transfer Complexes. J. Phys. Chem. B 2003, 107, 9434–9440. [CrossRef] 10. Hod, I.; Shalom, M.; Tachan, Z.; Rühle, S.; Zaban, A. SrTiO3 Recombination-Inhibiting Barrier Layer for Type II Dye-Sensitized Solar Cells. J. Phys. Chem. C 2010, 114, 10015–10018. [CrossRef] 11. Zhang, S.; Yang, X.; Numata, Y.; Han, L. Highly Efﬁcient Dye-S Challenges. Energy Environ. Sci. 2013, 6, 1443–1464. [CrossRef] 11. Zhang, S.; Yang, X.; Numata, Y.; Han, L. Highly Efﬁcient Dye-Sensitized Solar Cells: Progress and Future Challenges. Energy Environ. Sci. 2013, 6, 1443–1464. [CrossRef] 12. Nikolay, T.; Larina, L.; Shevaleevskiy, O.; Ahn, B.A. Electronic Structure Study of Lightly Nb Electrode for Dye-Sensitized Solar Cells. Energy Environ. Sci. 2011, 4, 1480–1486. [CrossRef] Nikolay, T.; Larina, L.; Shevaleevskiy, O.; Ahn, B.A. Electronic Structure Study of Lightly Nb-Doped TiO2 13. Ghartavol, H.M.; Mohammadi, M.R.; Afshar, A.; Chau-Nan Hong, F.; Jeng, Y.-R. Efﬁcient Dye-Sensitized Solar Cells Based on CNT-Derived TiO2 Nanotubes and Nb-Doped TiO2 Nanoparticles. RSC Adv. 2016, 6, 101737–101744. [CrossRef] Solar Cells Based on CNT-Derived TiO2 Nanotubes and Nb-Doped TiO2 Nanoparticles. RSC Adv. 2016, 6, 101737–101744. [CrossRef] 14. Liu, W.; Wang, H.-G.; Wang, X.; Zhang, M.; Guo, M. Titanium Mesh Supported TiO2 Nanowire Arrays/Nb-Doped TiO2 Nanoparticles for Fully Flexible Dye-Sensitized Solar Cells with Improved Photovoltaic Properties. J. Mater. Chem. C 2016, 4, 11118–11128. [CrossRef] 15. Su, H.; Huang, Y.-T.; Chang, Y.-H.; Zhai, P.; Hau, N.Y.; Cheung, P.C.H.; Yeh, W.-T.; Wei, N.T.-C.; Feng, S.-P. The Synthesis of Nb-Doped TiO2 Nanoparticles for Improved-Performance Dye Sensitized Solar Cells. Electrochim. Acta 2015, 182, 230–237. [CrossRef] 16. Lü, X.; Mou, M.X.; Wu, J.; Zhang, D.; Zhang, L.; Huang, F.; Xu, F.; Huang, S. Improved-Performance Dye-Sensitized Solar Cells Using Nb-Doped TiO2 Electrodes: Efﬁcient Electron Injection and Transfer. Adv. Funct. Mater. 2010, 20, 509–515. [CrossRef] 17. Kim, S.G.; Ju, M.J.; Choi, I.T.; Choi, W.S.; Choi, H.-J.; Baek, J.-B.; Kim, H.K. Nb-Doped TiO2 Nanoparticles for Organic Dye-Sensitized Solar Cells. RSC Adv. 27. Kubelka, P. New Contributions to the Optics of Intensely Light-Scattering Materials. Part I. J. Opt. Soc. Am. 1948, 38, 448–457. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 29. Yang, L.; Miklavcic, S.J. Revised Kubelka–Munk Theory. III. A General Theory of Light Propagation in Scattering and Absorptive Media. J. Opt. Soc. Am. A 2005, 22, 1866–1873. [CrossRef] 28. Yang, L.; Kruse, B. Revised Kubelka–Munk Theory. I. Theory and Application. J. Opt. Soc. Am. A 2004, 21, 1933–1941. [CrossRef] References 2013, 3, 16380–16386. [CrossRef] 18. Liu, J.; Duan, Y.; Zhou, X.; Lin, Y. Inﬂuence of VB Group Doped TiO2 on Photovoltaic Performance of Dye-Sensitized Solar Cells. Appl. Surf. Sci. 2013, 277, 231–236. [CrossRef] 19. Hishita, S.; Mutoh, I.; Koumoto, K.; Yanagida, H. Inhibition Mechanism of the Anatase- Transformation by Rare Earth Oxides. Ceram. Int. 1983, 9, 61–67. [CrossRef] 20. Cullity, B.D. Elements of X-Ray Diffraction. Am. J. Phys. 1957, 25, 394–395. [CrossRef] 21. Jono, R.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Theoretical Study of the Surface Complex between TiO2 and TCNQ Showing Interfacial Charge-Transfer Transitions. J. Phys. Chem. Lett. 2011, 2, 1167–1170. [CrossRef] [PubMed] 22. Taro, H.; Hideyuki, K.; Koichi, Y.; Hiroyuki, N.; Yutaka, F.; Shoichiro, N.; Naoomi, Y.; Akira, C.; Hiroshi, K.; M h O t l El t i B d St t f T t C d t Nb D d A t TiO 22. Taro, H.; Hideyuki, K.; Koichi, Y.; Hiroyuki, N.; Yutaka, F.; Shoichiro, N.; Naoomi, Y.; Akira, C.; Hiroshi, K.; Masaharu, O.; et al. Electronic Band Structure of Transparent Conductor: Nb-Doped Anatase TiO2. Appl. Phys. Express 2008, 1, 111203. Masaharu, O.; et al. Electronic Band Structure of Transparent Conductor: Nb-Doped Anatase TiO2. Appl. Phys. Express 2008, 1, 111203. 23. Bisquert, J. Theory of the Impedance of Electron Diffusion and Recombination in a Thin Layer. J. Phys. Chem. B 2002, 106, 325–333. [CrossRef] 24. Zhang, Z.; Zakeeruddin, S.M.; O’Regan, B.C.; Humphry-Baker, R.; Grätzel, M. Inﬂuence of 4-Guanidinobutyric Acid as Coadsorbent in Reducing Recombination in Dye-Sensitized Solar Cells. J. Phys. Chem. B 2005, 109, 21818–21824. [CrossRef] [PubMed] 25. Karvinen, S. The Effects of Trace Elements on the Crystal Properties of TiO2. Solid State Sci. 2003, 5, 811–819. [CrossRef] 26. Ito, S.; Murakami, T.N.; Comte, P.; Liska, P.; Grätzel, C.; Nazeeruddin, M.K.; Grätzel, M. Fabrication of Thin Film Dye Sensitized Solar Cells with Solar to Electric Power Conversion Efﬁciency over 10%. Thin Solid Films 2008, 516, 4613–4619. [CrossRef] 559 Catalysts 2018, 8, 367 27. Kubelka, P. New Contributions to the Optics of Intensely Light-Scattering Materials. Part I. J. Opt. Soc. Am. 1948, 38, 448–457. [CrossRef] 28. Yang, L.; Kruse, B. Revised Kubelka–Munk Theory. I. Theory and Application. J. Opt. Soc. Am. A 2004, 21, 1933–1941. [CrossRef] 29. Yang, L.; Miklavcic, S.J. Revised Kubelka–Munk Theory. III. A General Theory of Light Propagation in Scattering and Absorptive Media. J. Opt. Soc. Am. A 2005, 22, 1866–1873. Photocatalytic Antibacterial Effectiveness of Cu-Doped TiO2 Thin Film Prepared via the Peroxo Sol-Gel Method Benjawan Moongraksathum 1, Jun-Ya Shang 1 and Yu-Wen Chen 1,2,* 1 Department of Chemical and Materials Engineering, National Central University, Jhong-Li 3 bmoongraksathum@gmail.com (B.M.); alice19940430@gmail.com (J.-Y.S.) 2 Department of Chemistry, Tomsk State University, 36 Lenin Prospekt, Tomsk 634050, Russia * Correspondence: ywchen@cc.ncu.edu.tw; Tel.: +886-3422-7151 (ext. 34203) Received: 23 July 2018; Accepted: 20 August 2018; Published: 27 August 2018 1 Department of Chemical and Materials Engineering, National Central University, Jhong-Li 32001, T bmoongraksathum@gmail.com (B.M.); alice19940430@gmail.com (J.-Y.S.) 2 Department of Chemistry, Tomsk State University, 36 Lenin Prospekt, Tomsk 634050, Russia * Correspondence: ywchen@cc.ncu.edu.tw; Tel.: +886-3422-7151 (ext. 34203) Received: 23 July 2018; Accepted: 20 August 2018; Published: 27 August 2018 Received: 23 July 2018; Accepted: 20 August 2018; Published: 27 August 2018 Abstract: Cu-doped titanium dioxide thin ﬁlms (Cu/TiO2) were prepared on glass substrate via peroxo sol-gel method and dip-coating process with no subsequent calcination process for the degradation of organic dye and use as an antibacterial agent. The as-prepared materials were characterised using transmission electron microscopy (TEM), X-ray diffraction (XRD), scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS). For photocatalytic degradation of methylene blue in water, the samples were subjected to Ultraviolet C (UVC) and visible light irradiation. Degraded methylene blue concentration was measured using UV-Vis spectrophotometer. The antibacterial activities of the samples were tested against the gram-negative bacteria Escherichia coli (ATCC25922). Copper species were present in the form of CuO on the surface of modiﬁed TiO2 particles, which was conﬁrmed using TEM and XPS. The optimal observed Cu/TiO2 weight ratio of 0.5 represents the highest photocatalytic activities under both UVC and visible light irradiation. Moreover, the same composition remarkably exhibited high antibacterial effectiveness against E. coli after illumination with ultraviolet A. The presence of CuO on TiO2 signiﬁcantly enhanced photocatalytic activities. Therefore, active Cu-doped TiO2 can be used as a multipurpose coating material. Keywords: antibacterial; copper oxide; photocatalyst; titanium dioxide; thin ﬁlm; visible light Catalysts 2018, 8, 352; doi:10.3390/catal8090352 www.mdpi.com/journal/catalysts References [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 560 catalysts catalysts 1. Introduction Photocatalysis has garnered plenty of attention from the scientiﬁc community in recent decades, resulting in various commercialized products having photocatalytic functions. Among photocatalysts, titanium dioxide (TiO2) has received the greatest interest because of its remarkable stability and non-toxicity. Modiﬁcation of TiO2 has been extensively studied to improve its physical and chemical properties and to overcome the limitation of TiO2 in photocatalytic processes. Modiﬁed TiO2 has been deployed both environmentally and hygienically, including in the photocatalytic decomposition of organic pollutants [1–5], in self-cleaning materials [6–9], and as an antibacterial agent of photo-induced photocatalytic reactions [10–13]. The activity of TiO2 nanoparticles is due to the oxidative stress and/or the production of reactive oxygen species (ROS), including hydroxyl radical (OH•) and hydrogen peroxide (H2O2) under UV light irradiation; therefore, TiO2 is used as an antimicrobial agent. The produced ROS can cause cell membrane damage, cell cycle cessation, DNA damage and lipid peroxidation in microorganisms via direct contact between cells and nanoparticles, thereby resulting in cell death [14–16]. Several transition metals are toxic to various microbial pathogens. In addition to considerable commercialization in this ﬁeld, this ﬁnding has led to widespread research on the use of such materials 561 Catalysts 2018, 8, 352 as practical antimicrobial agents. Among several transition metals, copper has gained considerable attention, both as dispersions and, in the case of elemental copper and as alloys [10,17,18]. For use as antibacterial agents, copper and its compounds achieve antibacterial activity by the accumulation of copper ions within cells, eventually causing degradation of cell membranes [19–21]. Using titanium tetrachloride (TiCl4) as a precursor and H2O2 as a peptizing agent, previous studies have proposed a peroxo sol-gel method for the synthesis of neutral TiO2 sol [22–25]. The advantages of this method include not needing a calcination process to obtain the anatase structure of TiO2, as well as the ability to use H2O2 as an oxidizing agent. It results in the formation of TiO2 nanoparticles dispersed in neutral, stable, and transparent sol. The purpose of this study was to prepare Cu-doped TiO2 thin ﬁlms using the peroxo sol-gel method, thus determining how the addition of Cu to TiO2 inﬂuenced its antibacterial effectiveness, as well as the photocatalytic degradation of methylene blue (MB) aqueous solution under either UVC or visible light irradiation. 2.1. Characteristic of Cu-Doped TiO2 Particles The peroxo sol-gel method was used to prepare Cu-doped TiO2 sol by direct addition of the precursor of copper (Cu(NO3)2·3H2O) during the heating of TiO2 sol. To obtain powder nanoparticles, the as-prepared sols were further dried at 70 ◦C for several days. The X-ray diffraction (XRD) patterns of samples are shown in Figure 1. The diffraction peaks of the TiO2 and a series of Cu-modiﬁed TiO2 were located at the same positions and showed a similar pattern. The diffraction peaks located at 2θ = 25.31◦, 37.80◦, 48.05◦, 53.89◦, 55.06◦, 62.69◦, 68.76◦, and 75.03◦corresponded to the anatase phase of (101), (004), (200), (105), (211), (204), (116), and (215), respectively (JCPDS 21-1272). Furthermore, no additional peaks of copper oxide or other forms were found, implying that copper oxides were either highly dispersed with little TiO2 particles or the amount of Cu dopant was below the detection level of the technique. The size of the crystallite was calculated by the Scherrer equation [26,27]: L = 0.9λ β cos θ (1) (1) where λ is the X-ray wavelength (0.1540 nm), β is the full width at half maximum (FWHM), θ is the diffraction angle, and L is the average crystallite size. The results are listed in Table 1. The presence of copper slightly decreased the crystallite size of TiO2, which was attributed to the inhibition of titania condensation and the crystallization in the Cu-doped system [21]. ȱ ȱ Figure 1. X-ray diffraction patterns of TiO2 and a series of Cu-doped TiO2 powder (* = anatase). Figure 1. X-ray diffraction patterns of TiO2 and a series of Cu-doped TiO2 powder (* = anatase). 562 Catalysts 2018, 8, 352 Table 1. Crystallite sizes of TiO2 and a series of Cu-doped TiO2. Table 1. Crystallite sizes of TiO2 and a series of Cu-doped TiO2. Sample Weight Ratio of Cu:TiO2 Crystallite Size (nm) TiO2 0:100 14.96 0.1Cu/TiO2 0.1:100 9.12 0.3Cu/TiO2 0.3:100 9.40 0.5Cu/TiO2 0.5:100 9.21 0.7Cu/TiO2 0.7:100 9.60 1Cu/TiO2 1:100 10.16 Morphology of the as-prepared samples was investigated using TEM and HRTEM. Figure 2 depicts TEM images of TiO2 and 0.5Cu/TiO2 particles. The morphology of the TiO2 particles prepared via the peroxo sol-gel method is best described as an elliptical shape with particle size of 40–60 nm and 15–30 nm for long and short axes, respectively [3,7,12,23–25]. In Figure 2b, HRTEM image displayed the lattice fringe of TiO2 of 0.327 nm, corresponding to the anatase (101) plane. 2.1. Characteristic of Cu-Doped TiO2 Particles Figure 2c shows 0.5Cu/TiO2 particles having the particle size ranging from 20 to 30 nm and 5 to 10 nm for long and short axes, respectively. In addition, magniﬁed view clearly identiﬁed some small copper nanoparticles (≤4nm) deposited on the surface of the TiO2 nanoparticles (see Figure 2c). Therefore, elliptical anatase TiO2 nanoparticles can be synthesised via the peroxo sol-gel method without a subsequent annealing process, and the presence of copper could decrease the particles size and crystallite size of TiO2, attributed to phase deterioration of the TiO2 anatase [21]. ȱ ȱ (a)ȱ (b)ȱ (c)ȱ Figure 2. TiO2 particle: (a) TEM image and (b) HRTEM micrograph representing lattice fringe attributed to the (101) plane of titania; (c) TEM image and magniﬁed view of 0.5Cu/TiO2 particle. (c)ȱ ȱ (b)ȱ ȱ (a)ȱ (b)ȱ (a)ȱ (c)ȱ Figure 2. TiO2 particle: (a) TEM image and (b) HRTEM micrograph representing lattice fringe attributed to the (101) plane of titania; (c) TEM image and magniﬁed view of 0.5Cu/TiO2 particle. 2.2. Characterization of Cu-Doped TiO2 Thin Film 2.2. Characterization of Cu-Doped TiO2 Thin Film XPS was performed to investigate the electronic state of each element in TiO2 and Cu-doped TiO2 thin ﬁlms. Figure 3a–c shows the XPS spectra of Ti 2p, O 1s and Cu 2p in the TiO2 and 0.5Cu/TiO2 thin ﬁlms. The peaks of Ti 2p1/2 and Ti 2p3/2 in pure TiO2 were located at 464.6 and 458.9 eV, respectively, corresponding to the tetravalent state (Ti4+) [28]. The characteristic peaks of Ti 2p did not change in the presence of copper, indicating that the cations in the Cu-doped TiO2 ﬁlm are all in the Ti4+ state. Figure 3b shows the XPS spectra of the O 1s region for TiO2 and 0.5Cu/TiO2 ﬁlms. The binding state of O 1s region of TiO2 was deconvoluted into two peaks centred at 530.7 eV and 531.7 eV, which were ascribed to lattice oxide ions in TiO2 and hydroxyl groups on the surface, respectively (see Figure 3b and Table 2) [22]. The characteristic Cu 2p3/2 and Cu 2p1/2 peaks were observed at 934.9 eV and 954.9 eV, respectively, which were consistent with those of the Cu2+ cations [29]. The presence of Cu ions could capture the photogenerated carriers to accelerate the separation of charge carriers, 563 Catalysts 2018, 8, 352 subsequently transferring them to the surface of TiO2 thin ﬁlm, resulting in the improvement in the photocatalytic activity of the Cu-doped TiO2 [30]. ȱ ȱ ȱ (a)ȱ (b)ȱ (c)ȱ Figure 3. X-ray photoelectron spectroscopy (XPS) spectra of (a) Ti 2p, (b) O1s of TiO2 and 0.5Cu/TiO2, and (c) Cu 2p of 0.5Cu/TiO2. ȱ (b)ȱ ȱ (a)ȱ ȱ (c)ȱ (a)ȱ (a)ȱ (b)ȱ (c)ȱ Figure 3. X-ray photoelectron spectroscopy (XPS) spectra of (a) Ti 2p, (b) O1s of TiO2 and 0.5Cu/TiO2, and (c) Cu 2p of 0.5Cu/TiO2. Table 2. O1s XPS data and the fraction of total area of TiO2 and 0.5Cu/TiO2 thin ﬁlms. Sample Lattice O2− Ti-OH BE (eV) Fraction (%) BE (eV) Fraction (%) TiO2 530.7 81.48 531.7 18.52 0.5Cu/TiO2 530.6 73.72 531.6 26.28 Wettability measurements were performed using a customized in-house contact angle meter. To measure the water contact angle (WCA), a 5 μL DI water drop was dripped on the ﬁlms. The TiO2 ﬁlm prepared via the peroxo sol-gel method showed hydrophilicity with an average WCA of 6.4◦. After doping with copper, the average WCA of the TiO2 dramatically increased from 6.4◦to 35.9◦ (see Figure 4). 2.2. Characterization of Cu-Doped TiO2 Thin Film In general, elemental copper exhibits super hydrophilicity, whereas copper oxide (e.g., CuO and Cu2O) shows hydrophobicity. This ﬁnding conﬁrms the presence of CuO (Cu2+) in the Cu-doped TiO2 ﬁlms, which was in accordance with the XPS [31,32]. ȱ ȱ ȱ Figure 4. Water contact angle of TiO2 and Cu-doped TiO2 ﬁlms. Figure 4. Water contact angle of TiO2 and Cu-doped TiO2 ﬁlms. 2.3. Photocatalytic Degradation of MB Aqueous Solution 2.3. Photocatalytic Degradation of MB Aqueous Solution Prior to the photocatalytic activity, all samples were immersed into the set-up reactor and kept in the dark for 1 h to attain equilibrium adsorption of MB. The photocatalytic degradation of MB in water under UVC and visible light irradiation is shown in Figure 5. The highest photocatalytic activity under both UVC and visible irradiation was shown for the modiﬁed TiO2 with the weight ratio Cu:TiO2 = 0.5:100 (0.5Cu/TiO2). The amount of copper beyond a certain loading decrease the photocatalytic activity of TiO2 due to the Cu light absorption [33]. Furthermore, larger doping of copper resulted in CuO acting as a recombination centre [34]. 564 Catalysts 2018, 8, 352 ȱ ȱ ȱ ȱ ȱ Figure 5. Photocatalytic activities of TiO2 and Cu–doped TiO2 ﬁlms under UV light illumination (top) and visible light illumination (bottom). ȱ ȱ Figure 5. Photocatalytic activities of TiO2 and Cu–doped TiO2 ﬁlms under UV light illumination (top) and visible light illumination (bottom). The photodegradation of MB was ﬁtted to the Langmuir–Hinshelwood model. The slope of ln (C0/C) plotted versus irradiation time (h) indicates the reaction rate constant of the sample. Under UVC and visible light irradiation, the photocatalytic activity was the highest for 0.5Cu/TiO2, with a rate constant of 0.737 h−1 and 0.160 h−1, respectively (Table 3). Higher photocatalytic activity was attributed to the photoelectron transfer of the conduction band (CB) of TiO2 to CuO, leaving the hole on TiO2 to take part in oxidation reaction. In other words, under UV irradiation, the presence of CuO can slow down the recombination of electron/hole pairs in TiO2, as shown in the following equations [34,35]. Cu2+ + e−CB →Cu+ (2) Cu+ + (O2, H2O2 or other ROS) →Cu2+ + e− (3) (2) (3) CuO nanoparticles deposited on the TiO2 surface received the photogenerated electrons from the CB of TiO2 to form Cu+ ion as shown in Equation (2), and Cu+ ions could be re-oxidised to Cu2+ by the ROS species present in the surrounding media (3). CuO nanoparticles deposited on the TiO2 surface received the photogenerated electrons from the CB of TiO2 to form Cu+ ion as shown in Equation (2), and Cu+ ions could be re-oxidised to Cu2+ by the ROS species present in the surrounding media (3). Table 3. Rate constant of the reaction from pseudo-ﬁrst order kinetics under visible light irradiation Table 3. 2.4. Photocatalytic Antibacterial Effectiveness of Cu-doped TiO2 Thin Film 2.4. Photocatalytic Antibacterial Effectiveness of Cu-doped TiO2 Thin Film 2.4. Photocatalytic Antibacterial Effectiveness of Cu-doped TiO2 Thin Film 2.4. Photocatalytic Antibacterial Effectiveness of Cu-doped TiO2 Thin Film Under irradiation with UVA light, the antibacterial activity of the samples was tested against Escherichia coli (ATCC25922). E. coli is present as a normal intestinal ﬂora and is commonly found in contaminated drinking water. After a UVA radiation of 3 h at a low UVA intensity of 33 μW/cm2, the 0.5Cu/TiO2 coating showed high antibacterial effectiveness of >99% when tested against E. coli. In contrast, TiO2 displayed an average effectiveness of 61.20% when tested against E. coli (see Figure 6). Antibacterial activities of Cu-doped TiO2 could be attributed to the production of ROS species (e.g., O2−, OH•, and H2O2) using TiO2, as well as CuO nanoparticles that trigger oxidative stress and cell damage in bacteria [36]. In addition, released Cu ions (Cu2+) increased intracellular ROS in bacteria using the following pathway [17,21]: Cu+ + H2O2 →Cu2+ + OH−+ OH• (4) (4) H2O2 is a byproduct of normal metabolism of oxygen in bacterial cells. The accumulation of ROS dramatically increased, eventually causing cell death [10,17–21,36]. Therefore, Cu-doped TiO2 showed antibacterial effectiveness, even though a low-intensity of UVA light source was applied. (b)ȱ ȱ ȱ (a)ȱ (b)ȱ Figure 6. Antibacterial effectiveness (%) against E. coli following periods of illumination with 33 μW/cm2, UVA radiation of 1 and 3 h (a), and counts of viable E. coli after incubation (1:103 dilution) (b). ȱ (a)ȱ (a)ȱ (b)ȱ Figure 6. Antibacterial effectiveness (%) against E. coli following periods of illumination with 33 μW/cm2, UVA radiation of 1 and 3 h (a), and counts of viable E. coli after incubation (1:103 dilution) (b). 3. Materials and Methods 3.1. Materials 3.1. Materials TiCl4 (purity > 99.9%) and H2O2 (30% in water) were purchased from Showa Chemicals Industry, Ltd. (Tokyo, Japan). NH4OH was purchased from Merck Co. (Kenilworth, NJ, USA). Copper (II) nitrate trihydrate (Cu(NO3)2·3H2O) was purchased from Sigma-Aldrich (St. Louis., MO, USA). Distilled water was used throughout the experiments. 2.3. Photocatalytic Degradation of MB Aqueous Solution Rate constant of the reaction from pseudo-ﬁrst order kinetics under visible light irradiation. Samples Rate Constant (k, h−1) Under UV light Under Visible Light TiO2 0.449 0.070 0.1Cu/TiO2 0.687 0.066 0.3Cu/TiO2 0.546 0.098 0.5Cu/TiO2 0.737 0.160 0.7Cu/TiO2 0.524 0.146 1Cu/TiO2 0.317 0.076 565 Catalysts 2018, 8, 352 3.2. Preparation of TiO2 and Cu-Doped TiO2 Sol The typical procedure of preparing TiO2 and Cu/TiO2 sols went as follows: Three milliliters of TiCl4 were added dropwise into 150 mL 1 M HCl aqueous solution (under magnetic stirring) and kept in an ice bath so as to maintain the temperature at 0 ◦C for 30 min. An aqueous solution of 1 M NH4OH was then added dropwise to form the white hydrated titanium oxide gel Ti(OH)4. The pH of the solution was adjusted to 8.0 by adding the required amount of ammonia solution. After aging and stirring for 30 min, a Ti(OH)4 cake was ﬁltered and washed with distilled water until no chloride ions were detected (QUANTOFIX®). An amount of the as-prepared gel was re-dispersed in distilled water 566 Catalysts 2018, 8, 352 under magnetic stirring to form a milky solution. An aqueous solution of H2O2 was added dropwise to the solution under vigorous stirring for 1 h. The resultant solution was heated at 95 ◦C for 4 h under magnetic stirring. under magnetic stirring to form a milky solution. An aqueous solution of H2O2 was added dropwise to the solution under vigorous stirring for 1 h. The resultant solution was heated at 95 ◦C for 4 h under magnetic stirring. For preparation of Cu-doped TiO2 sol, the copper precursor was added into the TiO2 sol during heating at 95 ◦C. The solid content of the TiO2 in the solution was 1.0 wt %; the molar ratio of H2O2:TiO2 was 4:1; and the weight ratio of Cu:TiO2 was: 0.1:100, 0.3:100, 0.5:100, 0.7:100, and 1:100. These ratios were denoted as 0.1Cu/TiO2, 0.3Cu/TiO2, 0.5Cu/TiO2, 0.7Cu/TiO2, and 1Cu/TiO2, respectively. 3.5. Photocatalytic Degradation of MB Aqueous Solution 3.5. Photocatalytic Degradation of MB Aqueous Solution 3.5. Photocatalytic Degradation of MB Aqueous Solution The photocatalytic activity of the samples was determined by inducing the decomposition of MB under irradiation with either UVC or visible light. An aqueous solution of MB (40 mL) with a concentration of 10 mg/L was loaded onto a quartz glass plate. The samples were horizontally immersed into the MB solution. Before the photocatalytic activity was measured, the reactor was kept in the dark under magnetic stirring for 1 h to achieve the saturation adsorption of MB on the coatings. The catalysts were irradiated by either two 9 W UVC lamps (wavelength = 254 nm, TUV PL-L 18W/4P 1CT/25, Philips) or two 18 W compact ﬂuorescent lamps (1200 lumen, Philips (Pro UV Lamps Ltd., Bucks, UK)) equipped with UV cut-off ﬁlters with a cut-off wavelength of 410 nm. The distance from the lamp to the surface of the solution was 15 cm, and the concentration of the aqueous solution of MB was determined at intervals of 1 h using UV-vis spectrophotometer (JASCO V-630 (Japan Spectroscopic Company, Tsukuba, Japan)). The wavelength selected for the measurements was 663 nm, which is the characteristic maximum absorption wavelength of MB. 3.3. Preparation of Films The sols were aged at room temperature for 8 h before deposition on glass. The ﬁlms were prepared using the dip-coating method. Soda lime glass was used as the substrate. The total coating surface area of each glass substrate was 40 cm2. The glass substrate was cleaned using a commercial dishwashing detergent. Subsequently, it was ultrasonicated for 30 min in 1 M NH4OH solution, thoroughly rinsed with distilled water and oven-dried at 60 ◦C. The glass substrate was vertically dipped into the as-prepared sol with a withdrawal speed of 30 cm/min for 7 times. The thickness of the ﬁlms was 250–300 nm, as measured by SEM. 3.4. Characterization of Cu-Doped TiO2 Particles and Thin Films All samples were air dried at 80 ◦C for 1 h before further characterisation. The XRD patterns of the samples were determined using a Siemens D500 powder diffractometer (Siemens, Westborough, MA, USA) with Cu Kα source (λ = 1.5405 Å). The morphology and structure of the samples were investigated using transmission electron microscopy (TEM) on a JEM-2000 EX II (JEOL, Tokyo, Japan) operated at an accelerating voltage of 120 kV and high resolution TEM (HRTEM) on a JEOL JEM-2010 (JEOL, Tokyo, Japan) operated at 200 kV. The lattice spacing of the samples was measured using Gatan Digital Micrograph software. The chemical composition and chemical state of the samples were determined using X-Ray photoelectron spectroscopy (XPS) with a Thermo VG Scientiﬁc Sigma Prob spectrometer (Thermo Fisher Scientiﬁc, Logan, UT, USA). The XPS spectra were collected using Al Kα radiation at a voltage of 20 kV and current of 30 mA. The binding energy (BE) was calibrated using contaminant carbon (C1S = 284.6 eV). The peaks of each spectrum were organised using XPSPEAK software (Thermo Fisher Scientiﬁc, Logan, UT, USA); Shirley type background and 30:70 Lorentzian/Gaussian peak shape were adopted during the deconvolution. The thickness of the ﬁlms was measured using scanning electron microscopy (SEM) (Hitachi-3000, Tokyo, Japan). 3.6. Study of Antibacterial Activity Before testing, all samples were sterilized and activated using two 9 W UVC light (λ = 254 nm, TUV PL-L 18W/4P 1CT/25, Philips (Pro UV Lamps Ltd., Bucks, UK)) for 1 h. In order to quantitatively evaluate the antibacterial activity of the coatings, 0.4 mL of bacterial suspension was added onto each coating. Next, the test pieces were covered with 4 × 4 cm of adhesive ﬁlm (with a transparency of >85% at 340–380 nm). Covered with the adhesive ﬁlm, the test piece was placed into a Petri dish and exposed to two UVA lamps (λ = 365 nm, PL-L 36W/09/4P, Philips (Pro UV Lamps Ltd., Bucks, UK)) for either 1 h or 3 h. The irradiance of the UVA intensity was measured at 33 μW/cm2, using a UV light meter (model UV-340A, Lutron (Lutron Electronic Enterprise Co., Taipei, Taiwan)). After UVA irradiation, the bacterial suspension on each coating was washed off and diluted with phosphate buffer saline (PBS). Bacterial colony-forming units (CFUs) were enumerated by plating serial dilutions (1:10–1:105). The number of surviving bacterial colonies was counted (CFU/mL) after incubation at 37 ◦C for 24 h. The experiments were repeated three times for each sample type; therefore, three parallel CFU values were obtained for each type of sample. The antibacterial effectiveness was calculated according to the following equation [12,25]: C(%) = A −B A × 100% (5) (5) where C represents antibacterial effectiveness, A is the average number of colonies formed in the blank control group (CFU/mL), and B is the average number of colonies formed in the experimental group (CFU/mL). where C represents antibacterial effectiveness, A is the average number of colonies formed in the blank control group (CFU/mL), and B is the average number of colonies formed in the experimental group (CFU/mL). 4. Conclusions 0.5Cu-doped TiO2 nanoparticles can be successfully prepared via the peroxo sol-gel method without needing further calcination. The CuO nanoparticles, having a particle size of <4 nm, were deposited on the TiO2 surface. The photocatalytic activity was the highest for 0.5Cu/TiO2, with a rate constant of 0.737 h−1 and 0.160 h−1 under UVC and visible light irradiation, respectively. Moreover, the 0.5Cu/TiO2 coating showed high antibacterial effectiveness of >99% against E. coli after illumination with 33 μW/cm2 UVA radiation for 3 h. Therefore, the presence of CuO signiﬁcantly enhanced the photocatalytic activity as well as antibacterial effect of TiO2. Therefore, the Cu-coped TiO2 materials prepared via the peroxo sol-gel method can be an alternative and promising solution to increasing environmental contamination. Author Contributions: B.M. and Y.-W.C. designed the experiments; B.M. and J.-Y.S. performed the experiments, analyzed the data and contributed material characterization and analysis; B.M. wrote the paper and Y.-W.C. supervised the project. Funding: This research was funded by MOST 107-0205-2511. Funding: This research was funded by MOST 107-0205-2511. Acknowledgments: This research was supported by the Ministry of Science and Technology, Taiwan. Conﬂicts of Interest: The authors declare no conﬂicts of interest. Conﬂicts of Interest: The authors declare no conﬂicts of interest. 1. Vinodgopal, K.; Wynkoop, D.; Kamat, P. Environmental photochemistry on semiconductor surfaces: Photosensitized degradation of a textile azo dye, acid orange 7, on TiO2 particles using visible light. Environ. Sci. Technol. 1996, 30, 1660–1666. [CrossRef] 3.6. Study of Antibacterial Activity The test method of the coatings against E. coli’s (ATCC25922) antibacterial activity was modiﬁed from the certiﬁcate “JIS Z 2801: 2000 (E)—Antimicrobial products-Test for antimicrobial activity 567 Catalysts 2018, 8, 352 and efﬁcacy” and “TN-050—Standard on nano anti-bacterial coating.” The strains were grown on tryptic soy agar and diluted to 5.5 × 106–6.0 × 106 cells/mL using distilled water. The bacterial concentrations were measured from the optical density reading at 600 nm (OD600). The as-prepared sols were deposited on a 5 cm × 5 cm glass substrate using the dip-coating technique and a catalyst content of 0.2 mg/cm2. Bare glass substrates were used as controls. Before testing, all samples were sterilized and activated using two 9 W UVC light (λ = 254 nm, TUV PL-L 18W/4P 1CT/25, Philips (Pro UV Lamps Ltd., Bucks, UK)) for 1 h. In order to quantitatively evaluate the antibacterial activity of the coatings, 0.4 mL of bacterial suspension was added onto each coating. Next, the test pieces were covered with 4 × 4 cm of adhesive ﬁlm (with a transparency of >85% at 340–380 nm). Covered with the adhesive ﬁlm, the test piece was placed into a Petri dish and exposed to two UVA lamps (λ = 365 nm, PL-L 36W/09/4P, Philips (Pro UV Lamps Ltd., Bucks, UK)) for either 1 h or 3 h. The irradiance of the UVA intensity was measured at 33 μW/cm2, using a UV light meter (model UV-340A, Lutron (Lutron Electronic Enterprise Co., Taipei, Taiwan)). After UVA irradiation, the bacterial suspension on each coating was washed off and diluted with phosphate buffer saline (PBS). Bacterial colony-forming units (CFUs) were enumerated by plating serial dilutions (1:10–1:105). The number of surviving bacterial colonies was counted (CFU/mL) after incubation at 37 ◦C for 24 h. The experiments were repeated three times for each sample type; therefore, three parallel CFU values were obtained for each type of sample. and efﬁcacy” and “TN-050—Standard on nano anti-bacterial coating.” The strains were grown on tryptic soy agar and diluted to 5.5 × 106–6.0 × 106 cells/mL using distilled water. The bacterial concentrations were measured from the optical density reading at 600 nm (OD600). The as-prepared sols were deposited on a 5 cm × 5 cm glass substrate using the dip-coating technique and a catalyst content of 0.2 mg/cm2. Bare glass substrates were used as controls. References 1. Vinodgopal, K.; Wynkoop, D.; Kamat, P. Environmental photochemistry on semiconductor surfaces: Photosensitized degradation of a textile azo dye, acid orange 7, on TiO2 particles using visible light. Environ. Sci. Technol. 1996, 30, 1660–1666. [CrossRef] 568 Catalysts 2018, 8, 352 2. Mohamed, M.; Al-Esaimi, M. Characterization, adsorption and photocatalytic activity of vanadium-doped TiO2 and sulfated TiO2 (rutile) catalysts: Degradation of methylene blue dye. J. Mol. Catal. A Chem. 2006, 255, 53–61. [CrossRef] 3. Moongraksathum, B.; Chen, Y.W. CeO2–TiO2 mixed oxide thin ﬁlms with enhanced photocatalytic degradation of organic pollutants. J. Sol-Gel Sci. Technol. 2017, 82, 772–782. [CrossRef] 4. Chong, M.; Jin, B.; Chow, C.; Saint, C. Recent developments in photocatalytic water treatment technolo A review Water Res 2010 44 2997 3027 [CrossRef] [PubMed] 4. Chong, M.; Jin, B.; Chow, C.; Saint, C. Recent developments in photocatalytic water treatment technology: A. review. Water Res. 2010, 44, 2997–3027. [CrossRef] [PubMed] 4. Chong, M.; Jin, B.; Chow, C.; Saint, C. Recent developments in pho A. review. Water Res. 2010, 44, 2997–3027. [CrossRef] [PubMed] 5. Fujishima, A.; Zhang, X.; Tryk, D. TiO2 photocatalysis and related surface phenomena. Surf. Sci. Rep. 2 63, 515–582. [CrossRef] 6. Guan, K. Relationship between photocatalytic activity, hydrophilicity and self-cleaning effect of TiO2/SiO2 ﬁlms. Surf. Coat. Technol. 2005, 191, 155–160. [CrossRef] 7. Moongraksathum, B.; Chen, Y.W. Preparation and characterization of SiO2–TiO2 neutral sol by peroxo sol-gel method and its application on photocatalytic degradation. J. Sol-Gel Sci. Technol. 2016, 77, 288–297. [CrossRef] 8. Sakai, N.; Fujishima, A.; Watanabe, T.; Hashimoto, K. Quantitative evaluation of the photoinduced hydrophilic conversion properties of TiO2 thin ﬁlm surfaces by the reciprocal of contact angle. J. Phys. Chem. B 2003, 107, 1028–1035. [CrossRef] 9. Watanabe, T.; Fukayama, S.; Miyauchi, M.; Fujishima, A.; Hashimoto, K. Photocatalytic activity and photo-induced wettability conversion of TiO2 thin ﬁlm prepared by sol-gel process on a soda-lime glass. J. Sol-Gel Sci. Technol. 2000, 19, 71–76. [CrossRef] 10. Sunada, K.; Watanabe, T.; Hashimoto, K. Bactericidal activity of copper-deposited TiO2 thin ﬁlm under weak UV light illumination. Environ. Sci. Technol. 2003, 37, 4785–4789. [CrossRef] [PubMed] 11. Pelaez, M.; Nolan, N.; Pillai, S.; Seery, M.; Falaras, P.; Kontos, A.; Dunlop, P.; Hamilton, J.; Byrne, J.; O’Shea, K.; et al. A review on the visible light active titanium dioxide photocatalysts for environmental applications. Appl. Catal. B 2012, 125, 331–349. [CrossRef] 12. Moongraksathum, B.; Chen, Y.W. Anatase TiO2 co-doped with silver and ceria for antibacterial application. Catal. References Moongraksathum, B.; Chien, M.Y.; Chen, Y.W. Antiviral and antibacteria prepared by the peroxo sol–gel method. J. Nanosci. Nanotechnol.. Accepted. 25. Moongraksathum, B.; Chien, M.Y.; Chen, Y.W. Antiviral and antibacterial effects of silver-doped TiO2 prepared by the peroxo sol–gel method. J. Nanosci. Nanotechnol.. Accepted. 26. Monshi, A.; Foroughi, M.R.; Monshi, M.R. Modiﬁed Scherrer equation to estimate more accurately nano-crystallite size using XRD. World J. Nano Sci. Eng. 2012, 2, 154–160. [CrossRef] 27. Alexander, L.; Klug, H.P. Determination of crystallite size with the X–ray spectrometer. J. Appl. Phys. 1950, 21, 137–142. [CrossRef] 28. Dake, L.S.; Lad, R.J. Electronic and chemical interactions at aluminum/TiO2 (110) interfaces. Surf. Sci. 1993, 289, 297–306. [CrossRef] 29. Su, J.; Li, Z.; Zhang, Y.; Wei, Y.; Wang, X. N-Doped and Cu-doped TiO2-B nanowires with enhanced photoelectrochemical activity. RSC Adv. 2016, 6, 16177–16182. [CrossRef] 30. Wang, S.; Meng, K.; Zhao, L.; Jiang, Q.; Lian, J. Superhydrophilic Cu-doped TiO2 thin ﬁlm for solar-driven photocatalysis. Ceram. Int. 2014, 40, 5107–5110. [CrossRef] 30. Wang, S.; Meng, K.; Zhao, L.; Jiang, Q.; Lian, J. Superhydroph photocatalysis. Ceram. Int. 2014, 40, 5107–5110. [CrossRef] 31. Eshaghi, A.; Eshaghi, A. Preparation and hydrophilicity of TiO2 sol–gel derived nanocomposite ﬁlms modiﬁed with copper loaded TiO2 nanoparticles. Mater. Res. Bull. 2011, 46, 2342–2345. [CrossRef] 32. Xu, Y.; Li, J.A.; Yao, L.F.; Li, L.H.; Yang, P.; Huang, N. Preparation and characterization of Cu-doped thin ﬁlms and effects on platelets adhesion. Surf. Coat. Technol. 2015, 261, 436–441. [CrossRef] p f , , [ ] 33. Behnajady, M.; Shokri, M.; Taba, H.; Modirshahla, N. Photocatalytic activity of Cu doped TiO2 nanoparticles d f d d i [C f] 33. Behnajady, M.; Shokri, M.; Taba, H.; Modirshahla, N. Photocatalytic activity of Cu doped TiO2 nanopar and comparison of two main doping procedures. Micro Nano Lett. 2013, 8, 345–348. [CrossRef] j y, ; , ; , ; , y y p 2 and comparison of two main doping procedures. Micro Nano Lett. 2013, 8, 345–348. [CrossRef] 34. Moniz, S.J.A.; Tang, J. Charge transfer and photocatalytic activity in CuO/TiO2 nanoparticle heterojunctions synthesised through a rapid, one-pot, microwave solvothermal route. ChemCatChem 2015, 7, 1659–1667. [CrossRef] 35. Janczarek, M.; Kowalska, E. On the origin of enhanced photocatalytic activity of copper-modiﬁed titania in the oxidative reaction systems. Catalysts 2017, 7, 317. [CrossRef] 36. Applerot, G.; Lellouche, J.; Lipovsky, A.; Nitzan, Y.; Lubart, R.; Gedanken, A.; Banin, E. References Today 2018, 310, 68–74. [CrossRef] 13. Verdier, T.; Coutand, M.; Bertron, A.; Roques, C. Antibacterial activity of TiO2 photocatalyst alo coatings on E. coli: The inﬂuence of methodological aspects. Coatings 2014, 4, 670–686. [CrossRef] Verdier, T.; Coutand, M.; Bertron, A.; Roques, C. Antibacterial activity of TiO2 photocatalyst alone or in coatings on E. coli: The inﬂuence of methodological aspects. Coatings 2014, 4, 670–686. [CrossRef] 14. Maness, P.; Smolinski, S.; Blake, D.M.; Huang, Z.; Wolfrum, E.J.; Jacoby, W.A. Bactericidal activity of photocatalytic TiO2 reaction: Toward an understanding of its killing mechanism. Appl. Environ. Microb. 1999, 65, 4094–4098. photocatalytic TiO2 reaction: Toward an understanding of its killing mechanism. Appl. Environ. Microb. 1999, 65, 4094–4098. 15. Castro, C.; Sanjines, R.; Pulgarin, C.; Osorio, P.; Giraldo, S.A.; Kiwi, J. Structure-reactivity relations of the Cu-cotton sputtered layers during E. coli inactivation in the dark and under light. J. Photochem. Photobiol. A 2010, 216, 295–302. [CrossRef] 16. Richardson, S.D.; Thruston, A.D.; Collette, T.W.; Ireland, J.C. Identiﬁcation of TiO2/UV disinfection byproducts in drinking water. Environ. Sci. Technol. 1996, 30, 3327–3334. [CrossRef] 17. Leyland, N.; Podporska-Carroll, J.; Browne, J.; Hinder, S.; Quilty, B.; Pillai, S. Highly Efﬁcient F, Cu doped TiO2 anti-bacterial visible light active photocatalytic coatings to combat hospital-acquired infections. Sci. Rep. 2016, 6, 24770. [CrossRef] [PubMed] 18. Litter, M. Heterogeneous photocatalysis Transition metal ions in photocatalytic systems. Appl. Catal. B Environ. 1999, 23, 89–114. [CrossRef] 19. Espirito Santo, C.; Quaranta, D.; Grass, G. Antimicrobial metallic copper surfaces kill Staphylococcus haemolyticus via membrane damage. MicrobiologyOpen 2012, 1, 46–52. [CrossRef] [PubMed] 20. Grass, G.; Rensing, C.; Solioz, M. Metallic copper as an antimicrobial surface. Appl. Environ. Microbiol. 2011, 77, 1541–1546. [CrossRef] [PubMed] 21. Rtimi, S.; Pulgarin, C.; Kiwi, J. Recent developments in accelerated antibacterial inactivation on 2D Cu-Titania surfaces under indoor visible light. Coatings 2017, 7, 20. [CrossRef] 22. Sasirekha, N.; Rajesh, B.; Chen, Y.W. Synthesis of TiO2 sol in a neutral solution using TiCl4 as a precursor and H2O2 as an oxidizing agent. Thin Solid Films 2009, 518, 43–48. [CrossRef] 23. Chen, Y.W.; Chang, J.Y.; Moongraksathum, B. Preparation of vanadium–doped titanium dioxide neutral sol and its photocatalytic applications under UV light irradiation. J. Taiwan Inst. Chem. Eng. 2015, 52, 140–146. [CrossRef] 569 Catalysts 2018, 8, 352 24. Moongraksathum, B.; Hsu, P.T.; Chen, Y.W. Photocatalytic activity of ascorbic acid-modiﬁed TiO2 sol prepared by the peroxo sol-gel method. J. Sol-Gel Sci. Technol. 2016, 78, 647–659. [CrossRef] y g 25. Article Modiﬁcation to L-H Kinetics Model and Its Application in the Investigation on Photodegradation of Gaseous Benzene by Nitrogen-Doped TiO2 Peng Sun, Jun Zhang, Wenxiu Liu, Qi Wang and Wenbin Cao * School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing 100083, China; ustbsunpeng@163.com (P.S.); zhangjunustb@foxmail.com (J.Z.); liuwenxiu@outlook.com (W.L.); wangqi15@ustb.edu.cn (Q.W.) School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing 100083, China; ustbsunpeng@163.com (P.S.); zhangjunustb@foxmail.com (J.Z.); liuwenxiu@outlook.com (W.L.); wangqi15@ustb.edu.cn (Q.W.) gq * Correspondence: wbcao@ustb.edu.cn; Tel.: +86-010-6233-2457 p ; Received: 11 July 2018; Accepted: 6 August 2018; Published: 9 August 2018 Received: 11 July 2018; Accepted: 6 August 2018; Published: 9 August 2018 Abstract: In this paper, the Langmuir-Hinshelwood (L-H) model has been used to investigate the kinetics of photodegradation of gaseous benzene by nitrogen-doped TiO2 (N-TiO2) at 25 ◦C under visible light irradiation. Experimental results show that the photoreaction coefficient kpm increased from 3.992 × 10−6 mol·kg−1·s−1 to 11.55 × 10−6 mol·kg−1·s−1 along with increasing illumination intensity. However, the adsorption equilibrium constant KL decreased from 1139 to 597 m3·mol−1 when the illumination intensity increased from 36.7 × 104 lx to 75.1 × 104 lx, whereas it was 2761 m3·mol−1 in the absence of light. This is contrary to the fact that KL should be a constant if the temperature was fixed. This phenomenon can be attributed to the breaking of the adsorption-desorption equilibrium by photocatalytically decomposition. To compensate for the disequilibrium of the adsorption-desorption process, photoreaction coefficient kpm was introduced to the expression of KL and the compensation form was denoted as Km. KL is an indicator of the adsorption capacity of TiO2 while Km is only an indicator of the coverage ratio of TiO2 surface. The modified L-H model has been experimentally verified so it is expected to be used to predict the kinetics of the photocatalytic degradation of gaseous benzene. Keywords: modiﬁed L-H model; N-TiO2; photocatalytic degradation; benzene catalysts catalysts Catalysts 2018, 8, 326; doi:10.3390/catal8080326 References Understanding the antibacterial mechanism of CuO nanoparticles: Revealing the route of induced oxidative stress. Small 2012, 8, 3326–3337. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 570 1. Introduction Gaseous benzene released from paints, artiﬁcial panel or furniture is threatening to human health, particularly for children. However, the gaseous benzene in indoor air is difficult remedy with traditional methods due to its low concentration (ppm or ppb level) [1–3]. However, TiO2 can decompose gaseous benzene under ultraviolet light irradiation, thus it has attracted growing attention [4–8]. In fact, the photodegradation of gaseous benzene by TiO2 photocatalyst is a heterogeneous reaction occurring at a gas-solid interface, and the reaction rate is strongly affected by the environmental factors, particularly illumination intensity [9–11]. So the kinetic study of photocatalytic reaction is important for revealing the effect of these factors on the photocatalytic reaction rate. The heterogeneous reaction includes two consecutive steps. Firstly, the reactants are adsorbed on the surface of the photocatalysts and secondly, the photocatalytic reaction commences. Generally, the adsorption rate is slower than the photocatalytic reaction rate. So the overall photocatalytic reaction rate is mainly dominated by the adsorption rate. Furthermore, the adsorption rate can be equivalently expressed using the coverage ratio of the adsorbed reactants on the surface of the photocatalysts [12–15]. So the photocatalytic reaction rate r can be expressed as Equation (1) [16–18], which is widely known as the original L-H model. r = −dc dt = kpθ (1) r = −dc dt = kpθ (1) 10.3390/catal8080326 www.mdpi.com/journal/catalysts 571 Catalysts 2018, 8, 326; doi:10.3390/catal8080326 571 www.mdpi.com/journal/catalysts Catalysts 2018, 8, 326 where c is the concentration of the reactant, t is the photocatalytic reaction time, θ is the coverage ratio of pollutants on the TiO2 surface, kp is photoreaction coefﬁcient. where c is the concentration of the reactant, t is the photocatalytic reaction time, θ is the coverage ratio of pollutants on the TiO2 surface, kp is photoreaction coefﬁcient. According to Langmuir adsorption theory, the coverage ratio is related to adsorption capacity and the concentration of the reactant. KL was deﬁned as adsorption equilibrium constant to measure the adsorption capacity of TiO2 and coverage ratio θ can be expressed as Equation (2) according to adsorption theory [19]. θ = KLc 1 + KLc (2) (2) Input θ from Equation (2) to Equation (1), the photoreaction coefﬁcient r can be expressed as Equation (3) [20–22], K r = kp KLc 1 + KLc (3) (3) Equation (3) is the much known expression of L-H model and has been widely used in investigating the kinetics of photocatalytic reactions. 1. Introduction Lin et al. [23] studied the photocatalytic degradation pathway of dimethyl sulfide. They used original and derivative L-H models to study the kinetics under different temperatures and found that temperature can enhance photocatalytic activity. Dhada et al. [24] investigated the photocatalytic degradation of benzene by TiO2 under sunshine and UV light. They found that UV light can promote photocatalytic reaction than visible light due to its higher energy of the photons. Cheng et al. [25] studied the photocatalytic degradation of benzene. They found that higher temperature, illumination intensity and humidity can promote the reaction rate greatly. The works mentioned above are focused in revealing the effect of environmental factors such as illumination intensity, the amount of the photocatalyst and some processing parameters on the photodegradation ratio. However, the effect of illumination intensity on adsorption equilibrium coefﬁcient of gaseous pollutant was neglected in most articles. In liquid phase photocatalysis, some authors have reported their research on the effect of the illumination intensity on both the photoreaction coefﬁcient and the adsorption coefﬁcient [26–29]. Du [30] found that the value of the adsorption coefﬁcient calculated from the L-H model was illumination intensity-dependent in photodegradation of liquid dimethyl phthalate (DMP). Coincidentally, it has also been found that the adsorption coefﬁcient has been affected by light intensity in the gaseous photocatalytic reactions [31,32]. Brosillon [31] studied the kinetic model of photocatalytic degradation of butyric acid, and they found that the adsorption coefﬁcient KR can be expressed as Equation (4) KR = (krLHCRads0 + kd1 + k′d2I)K kd1 + k′d2I (4) (4) where krLH is the reaction rate of the reaction between ·OH and reactants, kd1, k′d2 is the decomposition rate of ·OH in the routes of ·OH →OH−+ h+ and ·OH + e−→OH−, I is the light intensity, K is the adsorption constant without light irradiation. Their results indicate that the adsorption coefﬁcient in gas photocatalytic reaction is a function of light intensity, which is not reasonable as the adsorption coefﬁcient should be a constant under a ﬁxed temperature. And, the parameter krLH, kd1, k′d2 are difﬁcult to calculate as the concentration of ·OH is difﬁcult to accurately measure [33] during the process of photocatalytic degradation of benzene and its concentration changes during the progression of the photocatalytic reaction. So this model is not applicable to predict the concentration of the reactant at different reaction times under different illumination intensities. 1. Introduction He [32] investigated the degradation of benzene by mesoporous TiO2 and also found that the adsorption coefﬁcient could be affected by light intensity. They attributed it to the decrease of available active sites as the increased photo-induced radicals will occupy more of the active sites under higher illumination intensity. However, the effect of photocatalytic decomposition of the adsorbed benzene by the increased radicals on the adsorption coefﬁcient was not considered. So, it’s necessary to accurately describe the relationship between the adsorption coefﬁcient and the illumination intensity in gaseous photocatalytic reactions. where krLH is the reaction rate of the reaction between ·OH and reactants, kd1, k′d2 is the decomposition rate of ·OH in the routes of ·OH →OH−+ h+ and ·OH + e−→OH−, I is the light intensity, K is the adsorption constant without light irradiation. Their results indicate that the adsorption coefﬁcient in gas photocatalytic reaction is a function of light intensity, which is not reasonable as the adsorption coefﬁcient should be a constant under a ﬁxed temperature. And, the parameter krLH, kd1, k′d2 are difﬁcult to calculate as the concentration of ·OH is difﬁcult to accurately measure [33] during the process of photocatalytic degradation of benzene and its concentration changes during the progression of the photocatalytic reaction. So this model is not applicable to predict the concentration of the reactant at different reaction times under different illumination intensities. He [32] investigated the degradation of benzene by mesoporous TiO2 and also found that the adsorption coefﬁcient could be affected by light intensity. They attributed it to the decrease of available active sites as the increased photo-induced radicals will occupy more of the active sites under higher illumination intensity. However, the effect of photocatalytic decomposition of the adsorbed benzene by the increased radicals on the adsorption coefﬁcient was not considered. So, it’s necessary to accurately describe the relationship between the adsorption coefﬁcient and the illumination intensity in gaseous photocatalytic reactions. 572 Catalysts 2018, 8, 326 Catalysts 2018, 8, 326 In the present work, the effect of illumination intensity on photoreaction coefﬁcient kpm and adsorption equilibrium coefﬁcient has been studied under a constant 25 ◦C. Photoreaction coefﬁcient was introduced as the modiﬁcation to KL and the compensation Km was used to replace KL in the original L-H model. The modiﬁed L-H model can reveal the interaction between the adsorption, desorption and photo-oxidation process. 1. Introduction The results showed that the Km and kpm can be obtained under different illumination intensity at 25 ◦C, thus the concentration at different reaction times can be predicted. 2.1. Characterization of the N-TiO2 Photocatalysts 2.1. Characterization of the N-TiO2 Photocatalysts The N-TiO2 catalysts were characterized by X-ray diffraction (XRD), Transmission electron microscopy (TEM), UV-Vis spectra (UV-Vis) and X-ray photoelectron spectroscopy (XPS) and the results were illustrated in Figure 1. Figure 1a shows the XRD patterns of N-TiO2. It is clear that all the diffraction peaks were indexed to that of anatase TiO2 (JCPDS no. 21-1272). The crystal size calculated by Scherrer’s Equation was also around 10.2 nm. Figure 1b shows the morphology of the N-TiO2 powders. It can be found that the prepared sample was composed of spherical TiO2 and the size was ranged from 9 to 12 nm, which is in consistent with the calculated result. The light absorption spectrum was measured by UV-Vis spectrum and was shown in Figure 1c. It is well known that the bandgap of pristine anatase is 3.2 eV, while the light absorption has been extended into the ranged of 400 to 600 nm of as-prepared N-TiO2. And its bandgap energy was 2.9 eV shown in the inset of Figure 1c calculated by using the method in other works [34,35]. The chemical state of N1s was also investigated by XPS and the result was shown in Figure 1d. Only one peak located at 399.9 eV can be found, which can be attributed to the interstitial doping of nitrogen into TiO2 lattice with Ti–O–N bond [36]. (a) (b) Figure 1. Cont. (b) (b) (a) Figure 1. Cont. 573 Catalysts 2018, 8, 326 (c) (d) Figure 1. Characterization of N-TiO2 catalysts (a) XRD patterns, (b) TEM image, (c) UV-Vis spectrum, (d) N1s binding energy peak. (d) (c) (d) (c) Figure 1. Characterization of N-TiO2 catalysts (a) XRD patterns, (b) TEM image, (c) UV-Vis spectrum, (d) N1s binding energy peak. 2.2. Kinetic Study of Photocatalytic Degradation of Benzene under Different Illumination Intensity Figure 2 shows the variation of benzene concentration with photocatalytic degradation time under different illumination intensities. It shows that the concentration of benzene remained almost unchanged during the ﬁrst hour without light irradiation, indicating that adsorption and desorption processes of benzene on TiO2 surface have reached equilibrium, thus the decrease of benzene after illumination can be ascribed to the photocatalytic degradation process. 2.1. Characterization of the N-TiO2 Photocatalysts When it was illuminated for 4 h under different illumination intensity of 36.7 × 104, 46.9 × 104, 61.7 × 104 and 75.1 × 104 lx, the removal ratio of benzene was 72.1%, 84%, 90% and 92.4%, respectively. The removal ratio increased dramatically under higher illumination intensity, indicating that illumination intensity can promote the photocatalytic degradation performance. Figure 2. Variation of benzene concentration vs. photocatalytic degradation time under different illumination intensity. Figure 2. Variation of benzene concentration vs. photocatalytic degradation time under different illumination intensity. 574 Catalysts 2018, 8, 326 Catalysts 2018, 8, 326 During the photocatalytic degradation process, the amount of degraded benzene per unit time can be calculated by Equation (5). d Δn = rV = −dc dt V (5) (5) where Δn is the amount of degraded benzene per unit time, r is the photocatalytic degradation rate, V is the volume of the reactor, c is concentration of benzene and t is photocatalytic degradation time. The detailed form of r is shown by original L-H model in Equation (3) [32,37], so after inputting r from Equation (3) to Equation (5), we can get Δn = −dc dt V = kp KLc 1 + KLc (6) (6) In Equation (6), kp is the photoreaction coefﬁcient of the whole reaction system and is related to the mass of the catalysts. So the photocatalytic degradation rate coefﬁcient per unit mass kpm can be expressed in Equation (7) k kpm = kp m (7) (7) Input kpm from Equation (7) into Equation (6), then we can get −dc dt V = mkpm KLc 1 + KLc (8) (8) So the relationship between dc and dt can be expressed in Equation (9) − V mkpm 1 + KLc KLc dc = dt (9) (9) The relationship between c and t can be obtained after making integration to Equation (9), that is The relationship between c and t can be obtained after making integration to Equation (9), that is − V kpmm c c0 1 + KLc KLc dc = t 0 dt (10) (10) The result of Equation is t = V mkpm (c0 −c) + 1 KL (ln c0 −ln c) (11) (11) After rearranging in terms of 1/(c0 −c), the linear form of Equation (11) is obtained. 2.1. Characterization of the N-TiO2 Photocatalysts ln(c0/c) c0 −c = m V kpmKL t c0 −c −KL (12) (12) In Equation (12), it can be found that ln(c0/c)/(c0 −c) and t/(c0 −c) is a linear relationship, and the slope and intercept of the line is mkpmKL/V and KL respectively. Figure 3 shows the plots of ln(c0/c)/(c0 −c) vs. t/(c0 −c) under different illumination intensity. According to the obtained slopes and intercepts, the values of kpm and KL were calculated and summarized in Table 1. And the standard deviation R2 for each case were also listed in Table 1. 575 talysts 2018, 8, 326 Figure 3. Plots of ln(c0/c)/(c0 −c) vs. t/(c0 −c) under different illumination intensity. (solid points: experimental results; solid line(curve): ﬁtted results). Catalysts 2018, 8, 326 Figure 3. Plots of ln(c0/c)/(c0 −c) vs. t/(c0 −c) under different illumination intensity. (solid points: experimental results; solid line(curve): ﬁtted results). Table 1. Calculated Kpm and KL under different illumination intensity using the original L-H model. Table 1. Calculated Kpm and KL under different illumination intensity using the original L-H mode Illumination Intensity/104 lx kpm/ 10−6 mol·kg−1·s−1 KL/ m3·mol−1 R2 36.7 3.992 1139 0.9981 46.9 5.731 1064 0.9847 61.7 8.589 791 0.9961 75.1 11.55 597 0.9674 It can be seen from Table 1 that kpm was calculated as 3.992 × 10−6, 5.371 × 10−6, 8.589 × 10−6, 11.55 × 10−6 mol·kg−1·s−1 corresponding to the illumination intensity of 36.7 × 104, 46.9 × 104, 61.7 × 104, and 75.1 × 104 lx, respectively. And kpm increased greatly with increases in illumination intensity, which means that the photodegradation rate of benzene can be signiﬁcantly promoted by increasing the illumination intensity in our experiment conditions. It’s reasonable that the increased illumination intensity means more photon irradiated on TiO2 surface, that can produce more ·OH, which is the main radical in photocatalytic reaction. According to other works [32], photoreaction rate coefﬁcient kpm depends on illumination intensity in a power law kpm = αIn (13) (13) The value of intensity coefﬁcient α and exponent n was 2.24 × 10−14 and 1.482 obtained by using the results in Table 1. And the value of adsorption constant KL was decreased from 1139 m3·mol−1 to 597 m3·mol−1 when the illumination intensity was increased from 36.7 × 104 lx to 75.1 × 104 lx. That is, KL varied with the variation of the illumination intensity. 2.1. Characterization of the N-TiO2 Photocatalysts However, the adsorption constant KL is related to the temperature and should be a constant as the temperature of the reactor was carefully maintained at 25 ◦C according to Langmuir adsorption theory. So the obtained results are inconsistent with the basic fact that the KL should be kept unchanged if the temperature was ﬁxed for a certain adsorption-desorption balance, which shows that original L-H model cannot be used to describe the photocatalysis processes accurately. Generally, it is widely recognized that the photocatalytic degradation of gaseous chemicals mainly includes two steps, gas adsorption on the surface of the photocatalyst and photodegradation. After the gas chemicals were adsorbed on the surface of the photocatalyst, certain amount of the adsorbed molecules were decomposed by photocatalytic degradation. 576 Catalysts 2018, 8, 326 Catalysts 2018, 8, 326 However, the original L-H model only considers the adsorption and desorption equilibrium of the gas molecules on the surface of the photocatalyst. So the amount of the adsorbed benzene molecules Δna and lost desorbed benzene molecules Δnd of N-TiO2 surface per unit time can be deﬁned as Equation (14) and Equation (15) respectively [38]. Δna = kac(1 −θ)S (14) Δnd = kdθS (15) (14) Δnd = kdθS (15) (15) where ka and kd is adsorption and desorption constant of benzene and is all thermodynamic constant. When adsorption and desorption process reach equilibrium, there is Δna = Δnd, and the detailed form is shown in Equation (16). kac(1 −θ)S = kdθS (16) (16) So coverage ratio θ and adsorption equilibrium constant KL can be obtained [19] θ = kac kd + kac = ka kd c 1 + ka kd c (17) KL = ka kd (18) θ = kac kd + kac = ka kd c 1 + ka kd c (17) (17) kd + kac 1 + ka kd c ( ) KL = ka kd (18) KL = ka kd (18) (18) KL is thermodynamically constant due to ka and kd being thermodynamic constants, and is an indication of adsorption ability of the catalysts. While in photocatalytic reaction, the degradation process would cause the decrease of benzene on TiO2 surface, which is equivalent to the increase in the desorption rate of benzene molecules. So the equilibrium between adsorption and desorption process would be broken. 2.1. Characterization of the N-TiO2 Photocatalysts However, adsorption equilibrium constant KL is only related to ka and kd in Equation (18), which make it impossible to reveal the effect of degradation process on the equilibrium. Therefore, original L-H model based on Langmuir adsorption theory is not entirely suitable for the photocatalytic degradation of benzene and necessary modiﬁcation should be applied to original L-H model for better understanding kinetics of the photocatalysis process. 2.3. Modication to the L-H Model and Kinetic Results under Different Illumination Intensity In the photocatalytic reaction, there are three processes: Adsorption, desorption and the photocatalytic degradation process. The photocatalytic degradation process will cause decrease of benzene molecules on interface, so the amount of lost benzene molecules Δnb is the sum of desorbed and photocatalytic degraded benzene molecules per unit time. Δnb = kdθS + kpmθS (19) Δnb = kdθS + kpmθS (19) Combing Equation (13) and (18), the coverage ratio θ becomes Combing Equation (13) and (18), the coverage ratio θ becomes θ = kac kd + kpm + kac = ka kd+kpm c 1 + ka kd+kpm c = Kmc 1 + Kmc (20) Km = ka kd + kpm = kd kd + kpm ka kd = kd kd + kpm KL (21) (20) Km = ka kd + kpm = kd kd + kpm ka kd = kd kd + kpm KL (21) (21) ka/(kd + kpm) can be deﬁned as coverage coefﬁcient Km in Equation (21). The coverage coefﬁcient Km is a function of ka, kd and kpm, so Km is not thermodynamic constant due to kpm is photodynamic. The value of Km is equal to that of KL while there is no light due to kpm is zero without irradiation. And the value of kpm will increase greatly under high illumination intensity, thus will result in a decrease of Km, which is in accordance with the experimental results in Table 1. ka/(kd + kpm) can be deﬁned as coverage coefﬁcient Km in Equation (21). The coverage coefﬁcient Km is a function of ka, kd and kpm, so Km is not thermodynamic constant due to kpm is photodynamic. The value of Km is equal to that of KL while there is no light due to kpm is zero without irradiation. 2.1. Characterization of the N-TiO2 Photocatalysts And the value of kpm will increase greatly under high illumination intensity, thus will result in a decrease of Km, which is in accordance with the experimental results in Table 1. 577 Catalysts 2018, 8, 326 Catalysts 2018, 8, 326 The original L-H model can be modiﬁed by using Km to replace KL in Equation (13) and (14) there is The original L-H model can be modiﬁed by using Km to replace KL in Equation (13) and (14) there is V 1 The original L-H model can be modiﬁed by using Km to replace KL in Equation (13) and ( there is t = V mkpm (c0 −c) + 1 Km (ln c0 −ln c) (22) ln(c0/c) c0 −c = m V kpmKmt 1 c0 −c −Km (23) (22) (23) The expression form of Equation (23) is similar to that of original L-H model except coverage coefﬁcient Km and equilibrium coefﬁcient KL. KL in original L-H model is an indicator of adsorption capacity of TiO2, while Km is the indicator of the amount of benzene on TiO2 surface. The parameters kpm and Km can be obtained through the plots of ln(c0/c)/(c0 −c) vs. t/(c0 −c) which were shown in Figure 4a and the results were listed in Table 2. And after taking reciprocal on both sides of Equation (21), the linear relationship exists between 1/Km and kpm can be found in Equation (24) and was shown in Figure 4b. Then the values of ka, kd and KL can also be obtained and summarized in Table 2. The value of KL in modiﬁed L-H model is 2629 m3·mol−1 under different illumination intensity at 25 ◦C, which is consistent with Langmuir adsorption theory. The value of ka and kd is constant in a given temperature at 25 ◦C and the relationship of kpm and I is revealed in Equation (14), thus Km under different illumination intensity can be obtained by Equation (21). Therefore, the concentration c at different photocatalytic reaction time t under different illumination intensity I can be predicted from Equation (22) 1 k k a given temperature at 25 C and the relationship of kpm and I is revealed in Equation (14), thus K under different illumination intensity can be obtained by Equation (21). 2.1. Characterization of the N-TiO2 Photocatalysts Therefore, the concentration at different photocatalytic reaction time t under different illumination intensity I can be predicted fro Equation (22) 1 Km = kpm ka + kd ka (2 (a) Figure 4. Cont. 578 1 Km = kpm ka + kd ka 1 Km = kpm ka + kd ka (24) (24) (a) Figure 4. Cont. Figure 4. Cont. 578 Catalysts 2018, 8, 326 (b) Figure 4. The relationship of the kinetic parameters in modiﬁed L-H model (a) The linear of ln(c0/c)/ (c0 −c) vs. t/(c0 −c), (b) The linear of 1/Km vs. kpm (solid points: experimental results; solid line(curve): Fitted results). (b) (b) Figure 4. The relationship of the kinetic parameters in modiﬁed L-H model (a) The linear of ln(c0/c)/ (c0 −c) vs. t/(c0 −c), (b) The linear of 1/Km vs. kpm (solid points: experimental results; solid line(curve): Fitted results). Table 2. Results of modiﬁed L-H model under different illumination intensity. Table 2. Results of modiﬁed L H model under different illumination intensity. Illumination Intensity/104 lx Photoreaction Coefﬁcient kpm/10−6 mol·kg−1·s−1 Coverage Coefﬁcient Km/m3·mol−1 Adsorption Constant ka/m3·kg−1·−1 Desorption Constant kd/mol·kg−1·s−1 Adsorption Equilibrium Constant KL/m3·mol−1 36.7 3.992 1139 9.242 × 10−3 3.514 × 10−6 2629 46.9 5.731 1064 61.7 8.589 791 75.1 11.55 597 2.4. The Adsorption Equilibrium Constant KL Obtained by Using Adsorption Theory In fact, the adsorption equilibrium constant KL is thermodynamically constant and can be used to evaluate the adsorption ability. In Langmuir adsorption theory, the adsorption equilibrium constant KL without light irradiation can be obtained as follow [39–41]: c0 (cT −c0)V = c0 cmV + 1 KLcmV (25) (25) where cT is total concentration of benzene ﬁlled into the reactor, c0 is initial concentration of gaseous benzene after adsorption equilibrium, cm is the maximum concentration that can be adsorbed by N-TiO2. It is obvious that there is a linear relationship between c0/(cT −c0)V and c0 in Equation (25). By ﬁlling different volume of benzene into reactor, cT and c0 can be measured after adsorption equilibrium and were summarized in Table 3. The plot of c0/(cT −c0)V vs. c0/cmV was shown in Figure 5. The slope and intercept of the linear is 1/cmV and 1/KLcmV, respectively. The value of KL was 2761 m3·mol−1, which is an indicator of the adsorption ability of benzene of N-TiO2 at 25 ◦C. 2.1. Characterization of the N-TiO2 Photocatalysts where cT is total concentration of benzene ﬁlled into the reactor, c0 is initial concentration of gaseous benzene after adsorption equilibrium, cm is the maximum concentration that can be adsorbed by N-TiO2. It is obvious that there is a linear relationship between c0/(cT −c0)V and c0 in Equation (25). By ﬁlling different volume of benzene into reactor, cT and c0 can be measured after adsorption equilibrium and were summarized in Table 3. The plot of c0/(cT −c0)V vs. c0/cmV was shown in Figure 5. The slope and intercept of the linear is 1/cmV and 1/KLcmV, respectively. The value of KL was 2761 m3·mol−1, which is an indicator of the adsorption ability of benzene of N-TiO2 at 25 ◦C. 579 Catalysts 2018, 8, 326 Figure 5. Linear relationship between c0/(cT −c0) and c0 (solid points: experimental results; solid line(curve): ﬁtted results). Figure 5. Linear relationship between c0/(cT −c0) and c0 (solid points: experimental results; solid line(curve): ﬁtted results). Table 3. Concentration of benzene before and after adsorption equilibrium at 25 ◦C. Table 3. Concentration of benzene before and after adsorption equilibrium at 25 ◦C Total Concentration Filled into the Reactor ct/ppm Initial Concentration after Adsorption Equilibrium c0/ppm 15 7.79 18.75 10.29 22.5 13.41 26.25 16.17 30 19.56 2.5. Veriﬁcation of the Modiﬁed L-H Model 2.5. Veriﬁcation of the Modiﬁed L-H Model 3.2. Photocatalytic Reaction System The schematic setup of the photocatalytic reaction system is illustrated in Figure 7. The cylindrical reactor with 15 cm in height and 10 cm in diameter was made of 316 L stainless steel. The temperature of the reactor were maintained at 25 ◦C by a bath circulator. A xenon lamp with a cut-off ﬁlter of 420 nm was used as the visible light illumination source. The illumination intensity could be adjusted at the range of 0 to 80 × 104 lx. A quartz window was mounted on the reactor for light irradiation. A gas chromatography (GC-2014, Shimadzu, Kyoto, Japan) was connected to the reactor to measure the concentrations of charged benzene in the reactor. The gas chromatography was equipped with Rtx-wax capillary column (Shimadzu) with 60 m in length, 0.53 mm in internal diameter and 1.0 μm in thickness. 2.5. Veriﬁcation of the Modiﬁed L-H Model To verify the modiﬁed L-H model, the photodegradation of benzene under the illumination intensity of 23.8 × 104 lx was carried out by ﬁxing other conditions except the initial concentration of benzene was 14.81 ppm. In this case, the calculated kpm and Km is 2.101 × 10−6 mol·kg−1·s−1 mol and 1645 m3·mol−1 respectively. By inputting the values of kpm and Km into Equation (22), the predicted concentration variation of benzene vs. irradiation time was obtained, which is shown in Figure 6 (denoted with the black solid line). The experimentally measured concentration of the benzene was denoted with red solid squares in Figure 6. It is clearly seen that the theoretical prediction shows very good agreement with the experimental results. So the modiﬁed L-H model can be used to predict benzene concentration under different illumination intensities at a constant temperature. 580 Catalysts 2018, 8, 326 Figure 6. The predicted and measured concentration of benzene vs. time. Figure 6. The predicted and measured concentration of benzene vs. time. 3.1. Preparation and Characterization of Samples Nanocrystalline N-TiO2 powders were prepared by hydrothermal method following the route used in our previous work [42]. The phase of the nano powders was determined by X-ray diffraction (XRD) with Cu Kα source in the 2θ ranging from 20 to 80◦. The morphology of N-TiO2 was characterized by Transmission electron microscopy (TEM, Hitachi, Jeol 200CX, Tokyo, Japan). UV-Vis spectra of the as-prepared sample was measured by Pgeneral UV-1901 instrument. The valence state of N was characterized by X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientic, Escalab 250, Waltham, MA, USA). Then the N-TiO2 catalysts were dispersed into alcohol with ultrasonic wave of 50 kHz by an ultrasonicato (S6103, Aladdin, Shanghai, China) for two hours. After dispersing, the suspension was spray-coated on the surface of a SiO2 glass substrate (5 cm × 5 cm) and the amount of coated N-TiO2 catalysts was 30 mg. The N-TiO2 coated glass was dried in air under 60 ◦C for 2 h. 3.3. Photocatalytic Reaction Procedures The N-TiO2 loaded glass was put into the photocatalytic reaction chamber. After a leakage check, the reactor was pumped to a vacuum of 0.1 atmosphere pressure, then the reactor was irradiated for 24 h under 254 nm ultraviolet light to clean the possible pollutants that may be adsorbed on the surface of the photocatalysts and the reactor as well. After a certain volume of benzene was charged/ﬂushed into the reactor, clean air (N2:O2 = 80%:20%) was ﬂushed into the reactor until the inner pressure was balanced with the atmospheric pressure. The concentration of benzene was set at 30 ppm as much as possible. Then the reactor was kept in dark for 60 min to reach the balance of adsorption-desorption. After that, the xenon lamp was turned on to make the irradiation through the quartz window, while the illumination intensity was adjusted at 36.7 × 104, 46.9 × 104, 61.7 × 104 and 75.1 × 104 lx by 581 Catalysts 2018, 8, 326 adjusting the distance between the light source and the sample. The concentration of the benzene in the reactor was measured and recorded every 30 min. The temperature of the reactor was maintained at 25 ◦C by a bath circulator. Figure 7. Schematic illustration of the photocatalytic reaction. Figure 7. Schematic illustration of the photocatalytic reaction. 4. Conclusions The L-H model has been used to investigate the kinetics of photodegradation of gaseous benzene by N-TiO2 at 25 ◦C under visible light irradiation. Experimental data indicates that the adsorption equilibrium constant KL calculated according to the L-H model decreased from 1139 to 597 m3·mol−1 when the illumination intensity was increased from 36.7 × 104 lx to 75.1 × 104 lx, whereas it was 2761 m3·mol−1 when in absence of light. This is contrary to the fact that KL should be a constant if the reaction temperature was ﬁxed. The benzene molecules adsorbed on the surface of the N-TiO2 were dynamically photodegraded by the photocatalyst and thus the equilibrium of adsorption-desorption was broken would account for that. Photoreaction coefﬁcient kpm was introduced in the L-H model to compensate the disequilibrium of the adsorption-desorption caused by photodecomposition. Experiment result shows that kpm is proportional to the light intensity I1.482. As a result, the new parameter Km (ka/(kd + kpm)) is closely related to the light intensity. Therefore, the concentration variation of benzene c vs irradiation time t under different light intensity I can be predicted. Author Contributions: In this paper, P.S., J.Z. and W.L. designed the experiments; P.S., J.Z. and W.L. performed the experiments; P.S., J.Z., W.L. and Q.W. analyzed the data; the manuscript was written by P.S. and edited by W.C. Funding: This research was funded by the National Key Research and Development Plan of China [Grant Nos. 2016YFC0700901, 2016YFC0700607]. Acknowledgments: This work is ﬁnancially supported by the National Key Research and Development Plan China [Grant Nos. 2016YFC0700901, 2016YFC0700607]. Acknowledgments: This work is ﬁnancially supported by the National Key Research and Development Plan of China [Grant Nos. 2016YFC0700901, 2016YFC0700607]. Conﬂicts of Interest: The authors declare no conﬂict of interest. 3. Ye, C.Z.; Ariya, P.A. Co-adsorption of gaseous benzene, toluene, ethylbenzene, m-xylene (btex) and SO2 on recyclable Fe3O4 nanoparticles at 0–101% relative humidities. J. Environ. Sci. 2015, 31, 164–174. [CrossRef] [PubMed] 2. Jiang, N.; Hui, C.-X.; Li, J.; Lu, N.; Shang, K.-F.; Wu, Y.; Mizuno, A. Improved performance of parallel surface/packed-bed discharge reactor for indoor VOCs decomposition: Optimization of the reactor structure. J. Phys. D Appl. Phys. 2015, 48, 40. [CrossRef] References 1. Sui, H.; Zhang, T.; Cui, J.; Li, X.; Crittenden, J.; Li, X.; He, L. Novel off-gas treatment technology to remove volatile organic compounds with high concentration. Ind. Eng. Chem. Res. 2016, 55, 2594–2603. [CrossRef] 2. Jiang, N.; Hui, C.-X.; Li, J.; Lu, N.; Shang, K.-F.; Wu, Y.; Mizuno, A. Improved performance of parallel surface/packed-bed discharge reactor for indoor VOCs decomposition: Optimization of the reactor structure. J. Phys. D Appl. Phys. 2015, 48, 40. [CrossRef] 582 Catalysts 2018, 8, 326 4. Zeng, L.; Lu, Z.; Li, M.; Yang, J.; Song, W.; Zeng, D.; Xie, C. A modular calcination method to prepare modiﬁed N-doped TiO2 nanoparticle with high photocatalytic activity. Appl. Catal. B Environ. 2016, 183, 308–316. [CrossRef] 5. Ren, L.; Mao, M.; Li, Y.; Lan, L.; Zhang, Z.; Zhao, X. Novel photothermocatalytic synergetic effect leads to high catalytic activity and excellent durability of anatase TiO2 nanosheets with dominant {001} facets for benzene abatement. Appl. Catal. B Environ. 2016, 198, 303–310. [CrossRef] pp 6. Yadav, H.M.; Kim, J.-S. Solvothermal synthesis of anatase TiO2-graphene oxide nanocomposites and their photocatalytic performance. J. Alloys Compd. 2016, 688, 123–129. [CrossRef] 6. Yadav, H.M.; Kim, J.-S. Solvothermal synthesis of anatase TiO2-graphene o photocatalytic performance. J. Alloys Compd. 2016, 688, 123–129. [CrossRef] performance. J. Alloys Compd. 2016, 688, 123–129. [Cros 7. Fujimoto, T.M.; Ponczek, M.; Rochetto, U.L.; Landers, R.; Tomaz, E. Photocatalytic oxidation of selected gas-phase VOCs using UV light, TiO2, and TiO2/Pd. Environ. Sci. Pollut. Res. 2017, 24, 6390–6396. [CrossRef] [PubMed] 8. Wongaree, M.; Chiarakorn, S.; Chuangchote, S.; Sagawa, T. Photocatalytic performance of electrospun CNT/TiO2 nanoﬁbers in a simulated air puriﬁer under visible light irradiation. Environ. Sci. Pollut. Res. 2016, 23, 21395–21406. [CrossRef] [PubMed] 9. Sabbaghi, S.; Mohammadi, M.; Ebadi, H. Photocatalytic degradation of benzene wastewater using PANI-TiO2 nanocomposite under UV and solar light radiation. J. Environ. Eng. 2016, 142, 05015003. [CrossRef] 10. Fang, J.; Chen, Z.; Zheng, Q.; Li, D. Photocatalytic decomposition of benzene enhanced by the heating effect of light: Improving solar energy utilization with photothermocatalytic synergy. Catal. Sci. Technol. 2017, 7, 3303–3311. [CrossRef] 11. Lan, L.; Li, Y.; Zeng, M.; Mao, M.; Ren, L.; Yang, Y.; Liu, H.; Yun, L.; Zhao, X. Efﬁcient UV-Vis-Infrared light-driven catalytic abatement of benzene on amorphous manganese oxide supported on anatase TiO2 nanosheet with dominant {001} facets promoted by a photothermocatalytic synergetic effect. Appl. Catal. B Environ. 2017, 203, 494–504. [CrossRef] 12. References Melián, E.P.; Díaz, O.G.; Araña, J.; Rodríguez, J.M.D.; Rendón, E.T.; Melián, J.A.H. Kinetics and adsorption comparative study on the photocatalytic degradation of o-, m- and p-cresol. Catal. Today 2007, 129, 256–262. [CrossRef] 13. Chen, M.; Bao, C.; Cun, T.; Huang, Q. One-pot synthesis of ZnO/oligoaniline nanocomposites with improved removal of organic dyes in water: Effect of adsorption on photocatalytic degradation. Mater. Res. Bull. 2017, 95, 459–467. [CrossRef] 14. Zhi, Y.; Li, Y.; Zhang, Q.; Wang, H. Zno nanoparticles immobilized on ﬂaky layered double hydroxides as photocatalysts with enhanced adsorptivity for removal of acid red g. Langmuir 2010, 26, 15546–15553. [CrossRef] [PubMed] 15. Dong, W.; Lee, C.W.; Lu, X.; Sun, Y.; Hua, W.; Zhuang, G.; Zhang, S.; Chen, J.; Hou, H.; Zhao, D. Synchronous role of coupled adsorption and photocatalytic oxidation on ordered mesoporous anatase TiO2-SiO2 nanocomposites generating excellent degradation activity of rhb dye. Appl. Catal. B Environ. 2010, 95, 197–207. [CrossRef] 16. Kim, S.B.; Hong, S.C. Kinetic study for photocatalytic degradation of volatile organic compounds in air using thin ﬁlm TiO2 photocatalyst. Appl. Catal. B Environ. 2002, 35, 305–315. [CrossRef] 17. Golshan, M.; Zare, M.; Goudarzi, G.; Abtahi, M.; Babaei, A.A. Fe3O4@hap-enhanced photocatalytic degradation of Acid Red73 in aqueous suspension: Optimization, kinetic, and mechanism studies. Mater. Res. Bull. 2017, 91, 59–67. [CrossRef] 18. Deng, X.-Q.; Liu, J.-L.; Li, X.-S.; Zhu, B.; Zhu, X.; Zhu, A.-M. Kinetic study on visible-light photocatalytic removal of formaldehyde from air over plasmonic Au/TiO2. Catal. Today 2017, 281, 630–635. [CrossRef] 19. Langmuir, I. The constitution and fundamental properties of solids and liquids. Part I. Solids. J. Am. Chem. Soc. 18. Deng, X.-Q.; Liu, J.-L.; Li, X.-S.; Zhu, B.; Zhu, X.; Zhu, A.-M. Kinetic study on visible-light photocatalytic removal of formaldehyde from air over plasmonic Au/TiO2. Catal. Today 2017, 281, 630–635. [CrossRef] 19. Langmuir, I. The constitution and fundamental properties of solids and liquids. Part I. Solids. J. Am. Chem. Soc. 1916, 38, 2221–2295. [CrossRef] 20. Liu, P.; Yu, X.; Wang, F.; Zhang, W.; Yang, L.; Liu, Y. Degradation of formaldehyde and benzene by TiO2 photocatalytic cement based materials. J. Wuhan Univ. Technol.-Mater. Sci. Ed. 2017, 32, 391–396. [CrossRef] 21. Yuzawa, H.; Aoki, M.; Otake, K.; Hattori, T.; Itoh, H.; Yoshida, H. Reaction mechanism of aromatic ring hydroxylation by water over platinum-loaded titanium oxide photocatalyst. J. Phys. Chem. C 2012, 116, 25376–25387. [CrossRef] 22. Einaga, H.; Mochiduki, K.; Teraoka, Y. References Li, C.X.; Jin, H.Z.; Yang, Z.Z.; Yang, X.; Dong, Q.Z.; Li, T.T. Preparation and photocatalytic properties of mesporous RGO/TiO2 composites. J. Inorg. Mater. 2017, 32, 357–364. 36. Xu, J.; Liu, Q.; Lin, S.; Cao, W. One-step synthesis of nanocrystalline N-doped TiO2 powders and their photocatalytic activity under visible light irradiation. Res. Chem. Intermed. 2013, 39, 1655–1664. [CrossRef] 37. Wang, J.; Ruan, H.; Li, W.; Li, D.; Hu, Y.; Chen, J.; Shao, Y.; Zheng, Y. Highly efﬁcient oxidation of gaseous benzene on novel Ag3VO4/TiO2 nanocomposite photocatalysts under visible and simulated solar light irradiation. J. Phys. Chem. C 2012, 116, 13935–13943. [CrossRef] 38. Cong, Y.; Zhang, J.; Chen, F.; Anpo, M.; He, D. Preparation, photocatalytic activity, and mechanism of nano- TiO2 co-doped with nitrogen and iron (iii). J. Phys. Chem. C 2007, 111, 10618–10623. [CrossRef] 39. Foo, K.Y.; Hameed, B.H. Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 2010, 156, 2–10. [CrossRef] 40. Wei, D.; Li, S.; Fang, L.; Zhang, Y. Effect of environmental factors on enhanced adsorption and photocatalytic regeneration of molecular imprinted TiO2 polymers for ﬂuoroquinolones. Environ. Sci. Pollut. Res. 2018, 25, 6729–6738. [CrossRef] [PubMed] 41. Li, Z.; Kim, J.K.; Chaudhari, V.; Mayadevi, S.; Campos, L.C. Degradation of metaldehyde in water by nanoparticle catalysts and powdered activated carbon. Environ. Sci. Pollut. Res. 2017, 24, 17861–17873. [CrossRef] [PubMed] 41. Li, Z.; Kim, J.K.; Chaudhari, V.; Mayadevi, S.; Campos, L.C. Degradation of metaldehyde in water by nanoparticle catalysts and powdered activated carbon. Environ. Sci. Pollut. Res. 2017, 24, 17861–17873. [CrossRef] [PubMed] 42. Xu, J.; Sun, P.; Zhang, X.; Jiang, P.; Cao, W.; Chen, P.; Jin, H. Synthesis of N-doped TiO2 with different nitrogen 41. Li, Z.; Kim, J.K.; Chaudhari, V.; Mayadevi, S.; Campos, L.C. Degradation of metaldehyde in water by nanoparticle catalysts and powdered activated carbon. Environ. Sci. Pollut. Res. 2017, 24, 17861–17873. [CrossRef] [PubMed] 42. Xu, J.; Sun, P.; Zhang, X.; Jiang, P.; Cao, W.; Chen, P.; Jin, H. Synthesis of N-doped TiO2 with different nitrogen concentrations by mild hydrothermal method. Mater. Manuf. Processes 2014, 29, 1162–1167. [CrossRef] 42. Xu, J.; Sun, P.; Zhang, X.; Jiang, P.; Cao, W.; Chen, P.; Jin, H. Synthesis of N-doped TiO2 with different nitrogen concentrations by mild hydrothermal method. Mater. Manuf. Processes 2014, 29, 1162–1167. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. References Photocatalytic oxidation processes for toluene oxidation over TiO2 catalysts. Catalysts 2013, 3, 219. [CrossRef] 583 Catalysts 2018, 8, 326 23. Lin, Y.-H.; Hsueh, H.-T.; Chang, C.-W.; Chu, H. The visible light-driven photodegradation of dimethyl sulﬁde on S-doped TiO2: Characterization, kinetics, and reaction pathways. Appl. Catal. B Environ. 2016, 199, 1–10. [CrossRef] 24. Dhada, I.; Nagar, P.K.; Sharma, M. Photo-catalytic oxidation of individual and mixture of benzene, toluene and p-xylene. Int. J. Environ. Sci. Technol. 2016, 13, 39–46. [CrossRef] 25. Cheng, L.; Kang, Y.; Li, G. Effect factors of benzene adsorption and degradation by nano-TiO2 immobilized on diatomite. J. Nanomaterials 2012, 2012, 6. [CrossRef] 26. Ollis, D.F. Kinetics of liquid phase photocatalyzed reactions: An illuminating approach. J. Phys. Chem 2005, 109, 2439–2444. [CrossRef] [PubMed] 27. Xu, Y.; Langford, C.H. Variation of langmuir adsorption constant determined for TiO2-photocatalyzed degradation of acetophenone under different light intensity. J. Photochem. Photobiol. A Chem. 2000, 133, 67–71. [CrossRef] 28. Giovannetti, R.; Rommozzi, E.; D’Amato, C.; Zannotti, M. Kinetic model for simultaneous adsorption/ photodegradation process of alizarin red s in water solution by nano-TiO2 under visible light. Catalysts 2016, 6, 84. [CrossRef] 29. Silva, C.G.; Faria, J.L. Effect of key operational parameters on the photocatalytic oxidation of phenol by nanocrystalline sol-gel TiO2 under uv irradiation. J. Mol. Catal. A Chem. 2009, 305, 147–154. [CrossRef] 30. Du, E.; Zhang, Y.X.; Zheng, L. Photocatalytic degradation of dimethyl phthalate in aqueous TiO2 suspension: A modiﬁed langmuir–hinshelwood model. React. Kinet. Catal. Lett. 2009, 97, 83–90. [CrossRef] 31. Brosillon, S.; Lhomme, L.; Vallet, C.; Bouzaza, A.; Wolbert, D. Gas phase photocatalysis and liquid phase photocatalysis: Interdependence and inﬂuence of substrate concentration and photon ﬂow on degradation reaction kinetics. Appl. Catal. B Environ. 2008, 78, 232–241. [CrossRef] 32. He, F.; Li, J.; Li, T.; Li, G. Solvothermal synthesis of mesoporous TiO2: The effect of morphology, size and calcination progress on photocatalytic activity in the degradation of gaseous benzene. Chem. Eng. J. 2014, 237, 312–321. [CrossRef] 33. Soltani, T.; Lee, B.-K. Novel and facile synthesis of Ba-doped BiFeO3 nanoparticles and enhancement of their magnetic and photocatalytic activities for complete degradation of benzene in aqueous solution. J. Hazard. Mater. 2016, 316, 122–133. [CrossRef] [PubMed] 34. Chen, S.-H.; Hsiao, Y.-C.; Chiu, Y.-J.; Tseng, Y.-H. A simple route in fabricating carbon-modiﬁed titania ﬁlms with glucose and their visible-light-responsive photocatalytic activity. Catalysts 2018, 8, 178. [CrossRef] 35. References This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). 584 MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel: +41 61 683 77 34 Fax: +41 61 302 89 18 www.mdpi.com ISBN 978-3-03936-707-8
https://openalex.org/W2179217092	https://www.mdpi.com/2073-4360/7/11/1516/pdf?version=1447756911	English	null	Amino Acid-Modified Polyethylenimines with Enhanced Gene Delivery Efficiency and Biocompatibility	Polymers	2,015	cc-by	10,759	Article Qin-Fang Zhang 1,2, Chao-Ran Luan 1, Dong-Xiao Yin 1, Ji Zhang 1,, Yan-Hong Liu 1, Qi Peng 3, Yong Xu 4 and Xiao-Qi Yu 1, Received: 25 August 2015 ; Accepted: 10 November 2015 ; Published: 17 November 2015 Academic Editor: Jianxun Ding Received: 25 August 2015 ; Accepted: 10 November 2015 ; Published: 17 November 2015 Academic Editor: Jianxun Ding 1 Key Laboratory of Green Chemistry and Technology (Ministry of Education), College of Chemistry, Sichuan University, Chengdu 610064, China; qfzhang_scu@163.com (Q.-F.Z.); 2014222030150@stu.scu.edu.cn (C.-R.L.); 1501110342@pku.edu.cn (D.-X.Y.); yanhongliu@scu.edu.cn (Y.-H.L.) 1 Key Laboratory of Green Chemistry and Technology (Ministry of Education), College of Chemistry, Sichuan University, Chengdu 610064, China; qfzhang_scu@163.com (Q.-F.Z.); 2014222030150@stu.scu.edu.cn (C.-R.L.); 1501110342@pku.edu.cn (D.-X.Y.); yanhongliu@scu.edu.cn (Y.-H.L.) y g 2 National Key Laboratory of Shock Wave and Detonation Physics, Institute of Fluid Physics, Chinese Academy of Engineering Physics, Mianyang 621900, China 3 3 West China School of Medicine, Sichuan University, Chengdu 610041, China; 2012151621064@stu.scu.edu.cn 4 Humanwell Healthcare (Group) Co., LTD., 666 Gaoxin Road, East Lake High-Tech Development Zone, Wuhan 430075, China; peter.xu@renfu.com.cn * Correspondence: jzhang@scu.edu.cn (J.Z.); xqyu@scu.edu.cn (X.-Q.Y.); Tel.: +86-28-8541-8580 (X.-Q.Y.); Fax: +86-28-8541-5886 (X -Q Y) Abstract: The development of gene delivery vectors with high efﬁciency and biocompatibility is one of the key points of gene therapy. A series of polycations were prepared from polyethylenimine (PEI) with several amino acids or their analogs. The target polymers have different charge and hydrophilic/hydrophobic properties, which may affect their performance in the gene transfection process. Gel retardation and DLS assays showed that these polymers may condense DNA into nanoparticles with positive zeta potentials and proper sizes for cellular uptake. Luciferase reporter gene transfection results revealed their higher transfection efﬁciency than PEI; especially in the presence of serum, in which up to 23 times higher efﬁciency was achieved by employing glycolic acid-grafted PEI. Moreover, it was found that the degree of substitution on PEI has an apparent inﬂuence on the transfection, and the balance between electron-positive/negative groups largely affects the delivery process. The higher serum tolerance was also proven by BSA adsorption, ﬂow cytometry and confocal microscopy assays. Results demonstrate that such type of polycations may serve as promising non-viral gene delivery vectors. Keywords: gene delivery; polyethylenimine; non-viral gene vector; biocompatibility 1. Introduction As an important type of non-viral gene delivery vectors, cationic polymers have received attention for their good stability, easy preparation, and modiﬁcation [1,2]. Cationic polymers can efﬁciently condense DNA via electrostatic interaction, and their cationic properties may also facilitate the contact between the polymer/DNA complex (polyplex) and negatively charged cell membrane, leading to better cell uptake. The leading polymeric materials for gene delivery include polyethylenimine (PEI) [3,4], poly(L-lysine) [5], poly(tertiaryamine methacrylate) [3], and polyamidoamine [6], etc. Among these polymers, PEI was most widely studied due to its easy availability. High molecular weight (HMW) PEI, especially that with the Mw of 25 kDa, has relatively higher tranfection Polymers 2015, 7, 2316–2331; doi:10.3390/polym7111516 www.mdpi.com/journal/polymers Polymers 2015, 7, 2316–2331 efﬁciency (TE) both in vitro and in vivo. This made it the golden standard for the design of novel polymeric vectors [7–9]. However, along with the higher TE, signiﬁcant cytotoxicity and limited biocompatibility are also found in HMW PEI-mediated gene delivery [10,11]. Thus, to develop novel PEI-based materials with both high TE and improved biocompatibility is of great importance. In recent years, large efforts were made to overcome the shortcomings of PEI [12–17]. Wang et al. showed that the introduction of hydrophobic dodecyl chains to PEI would improve the TE by enhanced escape of DNA from the endosome to the cytoplasm, but unfortunately the biocompatibility was not effectively improved [12]. Ramezani and co-workers revealed that alkyl-oligoamine derivatives of PEI without reduction of primary amine quantity is an effective strategy for the balance of hydrophobic-hydrophilic property and for the improvement of TE while maintaining low toxicity [13]. Zhuo found that the incorporation of a hydroxyl-enriched “skin” would afford PEI-g-5-ethyl-5-(hydroxymethyl)-1,3-dioxan-2-oxo (PEI-g-EHDO) with remarkably improved biocompatibility and stronger resistance against the serum-associated detrimental effects, such as protein adsorption, particle aggregation, and polycation-protein exchange [11]. In addition, poly(ethyleneglycol) (PEG) has been shown to have the ability to shield the positive charges on the complex, improving the serum tolerance [14–16]. However, the strong charge shielding and volume exclusion from such a polymer coating would interfere with the polycation/DNA complexation and hinder their efﬁcient internalization into cells [17]. With these concerns, the modiﬁcation of PEI, aiming for higher TE together with improved biocompatibility is not an easy task. The balances between the hydrophobic and hydrophilic properties and between the DNA binding and release abilities are of much signiﬁcance. 1. Introduction Taking clues from these studies, the present investigation was undertaken with the objective to incorporate hydrophobic, hydrophilic, or charged elements in the polymeric structure and evaluate their effects on the TE, cytotoxicity, and serum resistance. For this purpose, we selected serine (Ser), glycolic acid (Ga), glycine (Gly), N,N-bis(21-aminoethyl)glycine (Deta) and leucine (Leu) to modify 25 kDa PEI with various degrees of substitution (DS). Among these novel polymers, compared to PEI, Ser-PEI introduces hydrophilic hydroxyl groups without the reduction of amines; Ga-PEI replaces some amines with hydroxyls; Gly-PEI maintains the amount of amines; Deta-PEI may increase the primary amine groups; while Leu-PEI introduces hydrophobic alkyl groups. Using HEK293, HeLa and U-2OS cells as models, experiments revealed that these modiﬁed polymers showed better biocompatibility, serum resistance, and TE. 2.2. Preparation of Polymers N-(tert-Butoxycarbonyl)-carboxylic acid (2 mmol), EDCI (2.4 mmol), HOBT (2.4 mmol), and DIEA (2.4 mmol) in dry methylene chloride (50 mL) was stirred for 0.5 h in ice bath. Then, the desired amount of PEI (25 kDa) based on molar ratios solved in dry methylene chloride was added and the reaction mixture was stirred for two days at room temperature. After completion of the reaction, MeOH/HCl solution was added to remove the protecting t-butyloxycarbonyl (Boc) group. The residue was dissolved in a small amount of water and dialyzed (MWCO 8000–14,000 kDa) against deionized water for three days. The product was obtained as a white or pale-yellow solid after lyophilization. Yield: 46%–62%. 2.3. Polymer Characterization Leu-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 0.86 ppm (d, –CH3), 1.58–1.64 ppm (m, –CHCH3), 2.68–3.60 ppm (m, PEI–H), 3.9 ppm (m, –CHNH2). The calculation of DS of the polymers: Taking Ga-PEI for example, the characteristic singlet of δ 3.9 ppm represents the 2H on CH2OH on the substituents, while the broad multiplet represents the C–H of PEI (4H for each ethylenimine unit). Consequently, if the DS were 100%, the area ratio of the two peaks would be 1:2. Therefore, the DS may be calculated by (real peak area ratio)/(1:2). For Ga-PEI1, the DS would be (1:25)/(1:2) = 0.08, i.e., 8%. 2.1. Materials Unless otherwise stated, all chemicals and reagents were obtained commercially and used without further puriﬁcation. Absolute chloroform (CHCl3) and dichloromethane (CH2Cl2) were distilled after being dried with calcium hydride (CaH2). Column chromatography was performed using 200–300 mesh silica gel or 200–300 mesh Al2O3. All aqueous solutions were prepared from deionized or distilled water. The 1H NMR spectra was measured on a Bruker AM400 NMR spectrometer (Zurich, Switzerland). Proton chemical shifts of NMR spectra were given in ppm relative to internal reference TMS (1H, 0.00 ppm). N-(tert-Butoxycarbonyl)-L-serine, N-(tert-Butoxycarbonyl)-L-Glycine and N-(tert-Butoxycarbonyl)-L-leucine were purchased from Aladdin Industrial Corporation (Shanghai, China), MicroBCA protein assay kit was obtained from Pierce (Rockford, IL, USA). Luciferase assay kit was purchased from Promega (Madison, WI, USA). Endotoxin-free plasmid puriﬁcation kit was purchased from TIANGEN (Beijing, China). Cell counting Kit-8 (CCK-8) was purchased from Dojindo Laboratories (Kumamoto, Japan). The plasmids used in the study were pGL-3 (Promega, Madison, WI, USA) coding for luciferase DNA, and pEGFP-N1 (Clontech, Palo Alto, CA, USA) coding for EGFP DNA. The Dulbecco1s modiﬁed Eagle1s medium (DMEM) and fetal bovine serum were purchased from Invitrogen Co. (Chengdu, China). Cy5™was obtained from Molecular Probe (Mirus, 2317 Polymers 2015, 7, 2316–2331 Madison, WI, USA). HEK293T human embryonic kidney cell lines, HeLa cell lines, and U-2OS human osteosarcoma cancer cells were purchased from Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). N1,N”-di-Boc-N,N-bis(21-aminoethyl)glycine was synthesized as previously reported [18]. 2.3. Polymer Characterization 1H NMR spectra in D2O were obtained on a Bruker AV 400-MHz instrument at 25 ˝C. The molecular weights (Mw) and polydispersity (PDI, Mw/Mn) of prepared cationic polymers were determined by a gel permeation chromatography (GPC) system, which consisted of a Waters 515 pump, a Linear 7.8 mm ˆ 300 mm column (Waters Corp, Milford, MA, USA), an 18-angle laser scattering instrument (Wyatt Technology Corporation, Santa Barbara, CA, USA), and an OPTILAB DSP interferometric refractometer (Wyatt Technology Corporation, Santa Barbara, CA, USA). NaAc (0.3 M, pH 4.4) passed through a 0.02 µm ﬁlm ﬁlter was used as the eluent. A ﬂow rate of 0.5 mL/min was applied. Ser-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.50 ppm (m, PEI–H and –CH(NH2)CH2OH), 3.76 ppm (s, –CH2OH). Ser-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.50 ppm (m, PEI–H and –CH(NH2)CH2OH), 3.76 ppm (s, –CH2OH). Ga-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.52–3.25 ppm (m, PEI–H), 3.94 ppm (s, –CH2OH). Gly-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.51 ppm (m, PEI–H), 3.67 ppm (m, –CH2NH2). Deta-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.13 ppm (m, –CH2CH2NH2, –CONCH2CH2), 3.25 ppm (m, –COCH2, –CONCH2). Leu-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 0.86 ppm (d, –CH3), 1.58–1.64 ppm (m, –CHCH3), 2.68–3.60 ppm (m, PEI–H), 3.9 ppm (m, –CHNH2). Ser-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.50 ppm (m, PEI–H and –CH(NH2)CH2OH), 3.76 ppm (s, –CH2OH). Ga-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.52–3.25 ppm (m, PEI–H), 3.94 ppm (s, –CH2OH). Gly-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.51 ppm (m, PEI–H), 3.67 ppm (m, –CH2NH2). Deta-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.13 ppm (m, –CH2CH2NH2, –CONCH2CH2), 3.25 ppm (m, –COCH2, –CONCH2). Leu-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 0.86 ppm (d, –CH3), 1.58–1.64 ppm (m, –CHCH3), 2.68–3.60 ppm (m, PEI–H), 3.9 ppm (m, –CHNH2). Ser-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.50 ppm (m, PEI–H and –CH(NH2)CH2OH), 3.76 ppm (s, –CH2OH). Ga-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.52–3.25 ppm (m, PEI–H), 3.94 ppm (s, –CH2OH). Gly-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.51 ppm (m, PEI–H), 3.67 ppm (m, –CH2NH2). Deta-PEI: 1H NMR (400 MHz, D2O, TMS): δ = 2.64–3.13 ppm (m, –CH2CH2NH2, –CONCH2CH2), 3.25 ppm (m, –COCH2, –CONCH2). 2.8. Transfection Procedure Gene transfection of a series of complexes was investigated in U-2OS, HeLa and HEK293 cells. Cells were seeded in 24-well plates (10 ˆ 104 cells/well for U-2OS, 8.5 ˆ 104 cells/well for HeLa and HEK293) and grown to reach 70%–80% cell conﬂuence at 37 ˝C for 24 h in 5% CO2. Before transfection, the medium was replaced with a serum-free or a 10% serum-containing culture medium containing polymer/pDNA (1 µg) complexes at various weight ratios. After 4 h under standard incubator conditions, the medium was replaced with fresh medium containing serum and incubated for another 20 h. For ﬂuorescent microscopy assays, cells were transfected by complexes containing pEGFP-N1. After 24 h incubation, GFP-expressed cells were observed with an inverted ﬂuorescence microscope (Nikon Eclipse TE 2000E) equipped with a cold Nikon camera at a magniﬁcation of 100ˆ. Control transfection was performed in each case using a commercially available transfection reagent bPEI (25 kDa) based on the standard conditions speciﬁed by the manufacturer. For luciferase assays, cells were transfected by complexes containing pGL-3. For a typical assay in a 24-well plate, 24 h post transfection as described above, cells were washed with cold PBS and lysed with 100 µL 1ˆ Lysis reporter buffer (Promega, Madison, WI, USA). The luciferase activity was measured by microplate reader (Model 550, Bio-Rad, Hercules, CA, USA). Protein content of the lysed cell was determined by BCA protein assay. Gene transfection efﬁciency was expressed as the relative ﬂuorescence intensity per mg protein (RLU/mg protein). All the experiments were done in triplicate. 2.4. Ampliﬁcation and Puriﬁcation of Plasmid DNA pGL-3 and pEGFP-N1 plasmids were used. The former was seed as the luciferase reporter gene, which was transformed in M109 Escherichia coli, and the latter was used as the enhanced green ﬂuorescent protein reporter gene, which was transformed in E. coli DH5α. Both plasmids were ampliﬁed in E. coli grown in LB medium at 37 ˝C and 220 rpm overnight. The plasmids were puriﬁed by an EndoFree Tiangen™Plasmid Kit. Then, the puriﬁed plasmids were dissolved in TE (Tris+EDTA) buffer solution and stored at ´80 ˝C. The integrity of plasmids was conﬁrmed by agarose gel electrophoresis. The purity of plasmids was determined by the ratio of ultraviolet (UV) absorbances at 260 to 280 nm, and the result was approximately 1.91. 2318 Polymers 2015, 7, 2316–2331 2.7. Cell Culture Human osteosarcoma (U-2OS) cells, HeLa cells, and human embryonic kidney transformed HEK293 cells were incubated in DMEM with 10% (v/v) fetal bovine serum (FBS) and 1% antibiotics (penicillin-streptomycin, 10,000 U¨ mL´1) at 37 ˝C in a humidiﬁed atmosphere containing 5% CO2. The medium was replenished every other day. 2.6. Particle Size and ζ-Potential Measurement in Water Zeta potential (ζ-potential) and particle size were measured by a Nano-ZS ZEN3600 apparatus (Malvern Instruments) at 25 ˝C. This instrument is equipped with a red laser of wavelength 630 nm and measures the electrophoretic mobility of the particles using phase analysis of scattered light in an experimental set up similar to Laser Doppler Velocimetry (M3PALS technique, Malvern Instruments Ltd., Worcestershire, UK). Polyplexes with various w/w ratios were prepared by adding 1 mg of pUC-19 to the appropriate volume of the polymer solution (in PBS). Then, the solution of polyplexes was incubated at 37 ˝C for 0.5 h and diluted with deionized water to 1 mL prior to measurement. Data were shown as mean ˘ standard deviation (SD) based on triplicate independent experiments. 2.5. Agarose Gel Retardation Assay Polyplexes at different w/w ratios (weight ratio of polymer relative to pDNA) were prepared by adding an appropriate volume of the polymer solution to 5 µL of pUC 19 (0.025 mg/mL). The obtained complex solution was then diluted to the total volume of 15 µL. After incubation at 37 ˝C for 0.5 h, the polyplexes were electrophoresed on a 1% (w/v) agarose gel containing GelRed™in Triseacetate (TAE) running buffer at 120 V for 0.5 h. Then, DNA was visualized under an ultraviolet lamp using a Vilber Lourmat imaging system. 2.10. Protein Adsorption Assay In brief, 1 mL of polymer solution (1 mg/mL) was added to 1 mL of bovine serum albumin (BSA) solution (2 mg/mL). After shaking at 37 ˝C for 0.5 h, the polymer adsorbed BSA to form white cotton-shaped precipitates, which were removed by centrifugation. The supernatant was carefully collected and the concentration of BSA in it was determined by BCA protein assay. The protein adsorbed on the polyplexes was calculated using the following equation: q “ pCi ´ Csq ˆ V{m (1) (1) where Ci and Cs are the initial BSA concentration and the BSA concentration in the supernatant after adsorption experiments, respectively; V is the total volume of the solution (2 mL); and m is the weight of the polymer (1 mg) added into the solution. where Ci and Cs are the initial BSA concentration and the BSA concentration in the supernatant after adsorption experiments, respectively; V is the total volume of the solution (2 mL); and m is the weight of the polymer (1 mg) added into the solution. 2.11. Cellular Uptake of Plasmid DNA The cellular uptake of the polymer/ﬂuorescein-labeled DNA complexes was analyzed by ﬂow cytometry. The Label IT Cy5 Labeling Kit was used to label pDNA with Cy5 according to the manufacturer’s protocol. Brieﬂy, U-2OS cells were seeded in 12-well plates (2 ˆ 105 cells/well) and allowed to attach and grow for 24 h. For transfection in the absence of serum, the medium was exchanged with serum-free medium. As for transfection in the presence of serum, the medium was exchanged with serum-containing medium. Cells were incubated with Cy5 labeled DNA complexes (2 µg DNA/well, optimal weight ratio of each sample) in media for 4 h at 37 ˝C. Subsequently, the cells were washed with 1ˆ PBS containing heparin (120 U/mL) and harvested with 0.25% Trypsin/EDTA and resuspended in PBS. Mean ﬂuorescence intensity was analyzed using a FC 500 ﬂow cytometer (Beckman Coulter, Brea, CA, USA). Cy5-labeled plasmid DNA uptake was measured in the FL4 channel using the red diode laser (633 nm). Data from 10,000 events were gated using forward and side-scatter parameters to exclude cell debris. The ﬂow cytometer was calibrated for each run to obtain a background level of ~1% for control samples (i.e., untreated cells). 2.9. Cytotoxicity Assay Toxicity toward Human osteosarcoma (U-2OS) cells, HeLa cells, and human embryonic kidney transformed HEK293 cells was determined by cell counting kit-8 (CCK-8). About 7000 cells per well were seeded in 96-well plates and cultured overnight for 70%–80% cell conﬂuence. The medium was replaced with 50 µL of fresh medium, to which 50 µL polyplexes at various concentrations were 2319 Polymers 2015, 7, 2316–2331 added to achieve a ﬁnal volume of 100 µL. Twenty-four hours later, 10 µL CCK-8 mixed in 90 µL PBS was added to each well for additional 1 h incubation. The absorbance was measured in an ELISA plate reader (model 550, BioRad, Hercules, CA, USA) at a wavelength of 450 nm. The metabolic activity of the polyplex-treated cells was expressed as a relative to untreated cell controls taken as 100% metabolic activity. In addition, the cell viability of PEI was performed with a CCK-8 assay as unmodiﬁed control. 3.1. Synthesis and Characterization of Acylated Polymers The aim of this study was to rationally design and evaluate a class of potentially safe non-viral gene delivery vectors based on modiﬁed PEI. As shown in Scheme 1, target polymers were prepared from PEI 25 kDa and a series of functional acids with different amino densities and hydrophilic/hydrophobic properties. Except glycolic acid (Ga), other acids with primary amine group(s) need to be protected by di-tert-butyl dicarbonate (Boc2O) before their reaction with PEI. The ﬁnal polymers were obtained by dialysis (MWCO: 8000–14,000) against water for 3 days and lyophilization. Adjusting the feeding ratio of acid carboxyls to amines on PEI led to different DS (for example, feeding ratios of 0.1, 0.2, 0.3, and 0.5 resulted in Ga-PEI1 to Ga-PEI4, respectively), which could be calculated from the speciﬁc peak integrals in the 1H NMR spectra (Figure 1, also see the Experimental Section). The calculated DS are listed in Table 1. Considering the steric effect and the reactivity diversity of 1˝/2˝/3˝ amines in PEI, the DS is always lower than the carboxyl/amine feeding ratio. The molecular weights of target polymers were measured by GPC, and results in Table 1 show that the Mw increased with the rise of DS. Polymers 2015, 7, page–page (for example, feeding ratios of 0.1, 0.2, 0.3, and 0.5 resulted in Ga-PEI1 to Ga-PEI4, respectively), which could be calculated from the specific peak integrals in the 1H NMR spectra (Figure 1, also see the Experimental Section). The calculated DS are listed in Table 1. Considering the steric effect and the reactivity diversity of 1°/2°/3° amines in PEI, the DS is always lower than the carboxyl/amine feeding ratio. The molecular weights of target polymers were measured by GPC, and results in Table 1 show that the Mw increased with the rise of DS. Polymers 2015, 7, page–page (for example, feeding ratios of 0.1, 0.2, 0.3, and 0.5 resulted in Ga-PEI1 to Ga-PEI4, respectively), which could be calculated from the specific peak integrals in the 1H NMR spectra (Figure 1, also see the Experimental Section). The calculated DS are listed in Table 1. Considering the steric effect and the reactivity diversity of 1°/2°/3° amines in PEI, the DS is always lower than the carboxyl/amine feeding ratio. 3.1. Synthesis and Characterization of Acylated Polymers The molecular weights of target polymers were measured by GPC, and results in Table 1 show that the M increased with the rise of DS carboxylic acid H2N N H N NH2 HN NH2 x y z bPEI bPEI O n = NH2 OH Ser-PEI OH NH2 N NH2 NH2 NH2 Ga-PEI Gly-PEI Deta-PEI Leu-PEI HOOC Scheme 1. Cationic polymers modified from branched 25 kDa polyethylenimine (PEI). Scheme 1. Cationic polymers modiﬁed from branched 25 kDa polyethylenimine (PEI). carboxylic acid H2N N H N NH2 HN NH2 x y z bPEI bPEI O n = NH2 OH Ser-PEI OH NH2 N NH2 NH2 NH2 Ga-PEI Gly-PEI Deta-PEI Leu-PEI HOOC Scheme 1. Cationic polymers modified from branched 25 kDa polyethylenimine (PEI). NH2 Leu-PEI NH2 L PEI Ser-PEI Scheme 1. Cationic polymers modified from branched 25 kDa polyethylenimine (PEI). Scheme 1. Cationic polymers modiﬁed from branched 25 kDa polyethylenimine (PEI). Scheme 1. Cationic polymers modified from branched 25 kDa polyethylenimine (PEI). gure 1 1H NMR spectra of Ga PEI The degree of substitution (DS) could be calculated by Figure 1. 1H NMR spectra of Ga-PEI. The degree of substitution (DS) could be calculated by the ratio between the peak areas. The singlet of δ 3.9 ppm represents CH2OH on the substituents, while the broad multiplet represents the C–H near the amino groups of PEI. Table 1. Characterization of the polymers. DS was obtained from 1H NMR and represents the DS of the entire amino groups in PEI M and polydispersity (PDI) were obtained from GPC Figure 1. 1H NMR spectra of Ga-PEI. The degree of substitution (DS) could be calculated by the ratio between the peak areas. The singlet of δ 3.9 ppm represents CH2OH on the substituents, while the broad multiplet represents the C–H near the amino groups of PEI. Table 1. Characterization of the polymers. DS was obtained from 1H NMR and represents the DS of Figure 1. 1H NMR spectra of Ga-PEI. The degree of substitution (DS) could be calculated by the ratio between the peak areas. The singlet of δ 3.9 ppm represents CH2OH on the substituents, while the broad multiplet represents the C–H near the amino groups of PEI. g p g ( ) y between the peak areas. The singlet of δ 3.9 ppm represents CH2OH on the substituents, while the broad multiplet represents the C–H near the amino groups of PEI. 2.12. Confocal Laser Scanning Microscopy (CLSM) Analysis U-2OS cells were seeded at a density of 2 ˆ 104 cells per well in a 35 mm confocal dish (Φ = 15 mm), 24 h prior to transfection. For transfection in the absence of serum, the medium was exchanged with serum-free medium. As for transfection in the presence of serum, the medium was exchanged with serum-containing medium. Cells were incubated with Cy5-labeled DNA complexes (2 µg DNA/well, optimal weight ratio of each sample) in media for 4 h at 37 ˝C. Subsequently, cells were rinsed twice with PBS (pH 7.4) to remove complexes that were not taken up by cells, ﬁxed with 4% paraformaldehyde (dissolved with PBS buffer) for 10 min, nuclear staining was done with DAPI. The CLSM observation was performed using Leica TCS SP5 at excitation wavelengths of 405 nm for DAPI (blue), and 633 nm for Cy5 (red), respectively. 2320 Polymers 2015, 7, 2316–2331 3.1. Synthesis and Characterization of Acylated Polymers Table 1 Characterization of the polymers DS was obtained from 1H NMR and represents the DS of Figure 1. 1H NMR spectra of Ga-PEI. The degree of substitution (DS) could be calculated by the ratio between the peak areas. The singlet of δ 3.9 ppm represents CH2OH on the substituents, while the broad multiplet represents the C–H near the amino groups of PEI. Figure 1. 1H NMR spectra of Ga-PEI. The degree of substitution (DS) could be calculated by the ratio between the peak areas. The singlet of δ 3.9 ppm represents CH2OH on the substituents, while the broad multiplet represents the C–H near the amino groups of PEI. 6.8 9.3 4 8 S (%) M 6.8 9 3 2321 Polymers 2015, 7, 2316–2331 Table 1. Characterization of the polymers. DS was obtained from 1H NMR and represents the DS of the entire amino groups in PEI. Mw and polydispersity (PDI) were obtained from GPC. Table 1. Characterization of the polymers. DS was obtained from 1H NMR and represents the DS of the entire amino groups in PEI. Mw and polydispersity (PDI) were obtained from GPC. Table 1. Characterization of the polymers. DS was obtained from 1H NMR and represents the DS of the entire amino groups in PEI. Mw and polydispersity (PDI) were obtained from GPC. Polymer DS (%) Mw (kDa) PDI Ser-PEI1 6.8 31.9 1.67 Ser-PEI2 9.3 32.6 1.66 Ser-PEI3 14.8 35.8 1.80 Ser-PEI4 19.0 37.0 1.71 Ga-PEI1 8.0 29.4 1.55 Ga-PEI2 10.0 31.0 1.63 Ga-PEI3 15.9 32.6 1.68 Ga-PEI4 18.2 33.3 1.68 Gly-PEI1 4.0 28.5 1.54 Gly-PEI2 8.7 30.5 1.58 Gly-PEI3 13.3 32.7 1.60 Gly-PEI4 25.0 35.2 1.70 Deta-PEI1 6.5 33.2 1.88 Deta-PEI2 10.5 36.0 1.77 Deta-PEI3 12.5 37.3 1.96 Deta-PEI4 15.4 42.1 1.90 Leu-PEI1 7.0 32.3 1.56 Leu-PEI2 10.3 34.6 1.71 Leu-PEI3 13.3 36.6 1.78 Leu-PEI4 21.1 46.4 2.37 3.2. Formation of Polymer/DNA Complexes (Polyplexes) Polymers 2015, 7, page–page Table 1. Cont. Polymer DS (%) Mw (kDa) PDI Deta-PEI1 6.5 33.2 1.88 Deta-PEI2 10.5 36.0 1.77 Deta-PEI3 12.5 37.3 1.96 Deta-PEI4 15.4 42.1 1.90 Leu-PEI1 7.0 32.3 1.56 Leu-PEI2 10.3 34.6 1.71 3.2. Formation of Polymer/DNA Complexes (Polyplexes) Leu-PEI1 7.0 Leu-PEI2 10.3 3.2. Formation of Polymer/DNA Complexes (Polyplexes) Leu-PEI1 7.0 Leu-PEI2 10.3 Condensation of DNA into nano-sized particles is a prerequisite for efﬁcient delivery of DNA into cells [13]. 3.1. Synthesis and Characterization of Acylated Polymers The binding strength of the modiﬁed PEIs to plasmid DNA was determined by gel retardation assay at various weight ratios (polymer/DNA, w/w), and the binding ability was indicated by the ratio at which full retardation was achieved [19]. Figure 2 shows that the complete DNA retardation induced by these new materials could be observed at a w/w ratio of 0.8, and the DS did not affect the DNA condensation ability [20]. Among the ﬁve modiﬁcations, Deta-PEI shows slightly lower binding ability. The stability of the polyplexes was also studied, and it was found that DNA could be gradually released from the polyplex with the increase of heparin (Figure S1). However, a greater amount of heparin than the polymer was needed for release, indicating a tight binding of polymer to DNA. Moreover, DNA protection against nuclease was investigated, and the results are shown in Figure S2. After the treatment of DNase, naked DNA was degraded and no band was observed on the gel. On the contrary, after condensation with the polymers, DNA could resist DNase and be released by subsequent addition of heparin. Such results suggest that these polymers may well protect DNA from degradation in the circumstance with nuclease [21]. Leu-PEI3 13.3 36.6 1.78 Leu-PEI4 21.1 46.4 2.37 3.2. Formation of Polymer/DNA Complexes (Polyplexes) Condensation of DNA into nano-sized particles is a prerequisite for efficient delivery of DNA into cells [13]. The binding strength of the modified PEIs to plasmid DNA was determined by gel retardation assay at various weight ratios (polymer/DNA, w/w), and the binding ability was indicated by the ratio at which full retardation was achieved [19]. Figure 2 shows that the complete DNA retardation induced by these new materials could be observed at a w/w ratio of 0.8, and the DS did not affect the DNA condensation ability [20]. Among the five modifications, Deta-PEI shows slightly lower binding ability. The stability of the polyplexes was also studied, and it was found that DNA could be gradually released from the polyplex with the increase of heparin (Figure S1). However, a greater amount of heparin than the polymer was needed for release, indicating a tight binding of polymer to DNA. Moreover, DNA protection against nuclease was investigated, and the results are shown in Figure S2. After the treatment of DNase, naked DNA was degraded and no band was observed on the gel. 3.1. Synthesis and Characterization of Acylated Polymers On the contrary, after condensation with the polymers, DNA could resist DNase and be released by subsequent addition of heparin. Such results suggest that these polymers may well protect DNA from degradation in the circumstance with nuclease [21]. Figure 2. Agarose gel electrophoresis of plasmid DNA (pUC 19) complexed with the cationic polymers at different weight ratios. In each image, the first lane is DNA control. Since cells typically take up particles ranging from micrometers to nanometers [22], it is Figure 2. Agarose gel electrophoresis of plasmid DNA (pUC 19) complexed with the cationic polymers at different weight ratios. In each image, the ﬁrst lane is DNA control. Figure 2. Agarose gel electrophoresis of plasmid DNA (pUC 19) complexed with the cationic polymers at different weight ratios. In each image, the first lane is DNA control. Since cells typically take up particles ranging from micrometers to nanometers [22] it is Figure 2. Agarose gel electrophoresis of plasmid DNA (pUC 19) complexed with the cationic polymers at different weight ratios. In each image, the ﬁrst lane is DNA control. n endocy asure the 2322 Polymers 2015, 7, 2316–2331 Since cells typically take up particles ranging from micrometers to nanometers [22], it is necessary for the cationic polymers to compact nucleic acids into nanoparticles with proper zeta potentials, which also play an important role in endocytosis and material cytotoxicity [23]. A dynamic light scattering (DLS) assay was used to measure the diameters and zeta potential of polyplexes formed by the modiﬁed-PEI and plasmid DNA at w/w ratios of 0.5, 2, 4, and 8, and unmodiﬁed PEI was used for comparison. The DS of each polymer was chosen according to the optimized DS from the luciferase gene transfection experiments (Section 3.3). As shown in Figure 3, in general, similar to PEI, these polymers may condense DNA into particles with diameters ranging from 100 to 450 nm, and the zeta-potential of the polyplexes reversed to positive at w/w ratio of ~1. The particle size decreased whereas the zeta potential increased with the rise of weight ratio. Polyplexes formed from Ga-PEI or Leu-PEI, in which hydroxyl or hydrophobic groups were introduced, showed relatively lower zeta potentials. Meanwhile, polyplex derived from Deta-PEI had the highest amino density, leading to the highest positive charge. However, the particle size seemed to be unaffected by the density of amino groups. 3.1. Synthesis and Characterization of Acylated Polymers Deta-PEI only gave the smallest particles at a relatively higher w/w ratio of 8. The inﬂuence of DS on polyplex size and potential was also studied by using Ga-PEI as model, and the results (Figure S3) show that with the rise of DS, an increase in particle size and a decreased zeta potential were observed. This may be logically explained by the positive charge screening from the increased amount of hydroxyl groups. Polymers 2015, 7, page–page and the zeta-potential of the polyplexes reversed to positive at w/w ratio of ~1. The particle size decreased whereas the zeta potential increased with the rise of weight ratio. Polyplexes formed from Ga-PEI or Leu-PEI, in which hydroxyl or hydrophobic groups were introduced, showed relatively lower zeta potentials. Meanwhile, polyplex derived from Deta-PEI had the highest amino density, leading to the highest positive charge. However, the particle size seemed to be unaffected by the density of amino groups. Deta-PEI only gave the smallest particles at a relatively higher w/w ratio of 8. The influence of DS on polyplex size and potential was also studied by using Ga-PEI as model, and the results (Figure S3) show that with the rise of DS, an increase in particle size and a decreased zeta potential were observed. This may be logically explained by the positive charge screening from the increased amount of hydroxyl groups. Figure 3. Particle size (a) and zeta potential (b) of polyplexes obtained from different modified PEIs by dynamic light scattering (DLS). Data represent mean ± SD (n = 3). Figure 3. Particle size (a) and zeta potential (b) of polyplexes obtained from different modiﬁed PEIs by dynamic light scattering (DLS). Data represent mean ˘ SD (n = 3). Figure 3. Particle size (a) and zeta potential (b) of polyplexes obtained from different modified PEIs by dynamic light scattering (DLS). Data represent mean ± SD (n = 3). Figure 3. Particle size (a) and zeta potential (b) of polyplexes obtained from different modiﬁed PEIs by dynamic light scattering (DLS). Data represent mean ˘ SD (n = 3). 3.3. In Vitro Gene Transfection 3.3. In Vitro Gene Transfection In Hela cells, up to 23 times higher TE than PEI was achieved. In each cell line, the TE of Ga-PEI was higher than Ser-PEI, which also has hydroxyl groups but maintains the amount of amino groups. This indicates that the balance between electron-positive/negative groups, which also called the “hydroxyl effect” [11,28] or the charge screening effect [29,30], is important for the transfection by such types of polycations, especially in the serum circumstance. On the other hand, Gly-PEI also showed good TE, but the relative TE compared to PEI was lower than that obtained without serum age at the TE of PEI dramatically decreases in serum-supplem ment of non-viral gene vectors that may retain their effi tudy, the TEs of the newly prepared polymers with optim 10% serum. As shown in Figure 5a, compared to the res polymers modified with hydroxyls showed better seru EI gave six times higher TE than PEI, while the value wa PEI also showed excellent TE toward other cell lines incl e 5c). In Hela cells, up to 23 times higher TE than PEI wa PEI was higher than Ser-PEI, which also has hydroxyl gr Figure 4. Luciferase gene expression transfected by polyplexes at different weight ratios in comparison with 25 kDa PEI (w/w = 1.4, N/P = 10) in U-2OS cells in the absence of serum. In each group of columns, the w/w are 2, 3, 4, and 5, consecutively. Data represent mean ± SD (n = 3). Figure 4. Luciferase gene expression transfected by polyplexes at different weight ratios in comparison with 25 kDa PEI (w/w = 1.4, N/P = 10) in U-2OS cells in the absence of serum. In each group of columns, the w/w are 2, 3, 4, and 5, consecutively. Data represent mean ˘ SD (n = 3). amount of amino groups. This indicates that the balance between electron-positive/negative groups, which also called the “hydroxyl effect” [11,28] or the charge screening effect [29,30], is important for the transfection by such types of polycations, especially in the serum circumstance. On the other hand, Gly-PEI also showed good TE, but the relative TE compared to PEI was lower than that obtained h Figure 5. Luciferase gene expression transfected in the presence of 10% serum in different cell lines. (a) U-2OS; (b) HeLa; (c) HEK293. Data represent mean ± SD (n = 3). 3.3. In Vitro Gene Transfection 3.3. In Vitro Gene Transfection This indicates that the balance between electron-positive/negative groups, which also called the “hydroxyl effect” [11,28] or the charge screening effect [29,30], is important for the transfection by such types of polycations, especially in the serum circumstance. On the other hand, Gly-PEI also showed good TE, but the relative TE compared to PEI was lower than that obtained without serum. Luciferase reporter gene was used to quantitatively assess the in vitro TE of these modified-PEIs in U-2OS cells. Figure 4 shows the relative TE of these polymers at various w/w ratios (2, 3, 4, 5) in comparison with PEI (25 kDa) at its optimal weight ratio of 1.4 (N/P ratio = 10) [24]. The conjugation of amino acids on PEI definitely enhanced the luciferase expression, and glycine-modified Gly-PEI showed the best TE, which was about 10 times higher than PEI. The DS also affects their gene delivery behavior: for each polymer, about 10% DS is most suitable for the transfection. 8 Figure 4. Luciferase gene expression transfected by polyplexes at different weight ratios in comparison with 25 kDa PEI (w/w = 1.4, N/P = 10) in U-2OS cells in the absence of serum. In each group of columns, the w/w are 2, 3, 4, and 5, consecutively. Data represent mean ± SD (n = 3). Figure 4. Luciferase gene expression transfected by polyplexes at different weight ratios in comparison with 25 kDa PEI (w/w = 1.4, N/P = 10) in U-2OS cells in the absence of serum. In each group of columns, the w/w are 2, 3, 4, and 5, consecutively. Data represent mean ˘ SD (n = 3). Polymers 2015, 7, page–page It is known that the TE of PEI dramatically decreases in serum-supplemented medium [25–27]. Thus, the development of non-viral gene vectors that may retain their efficiency in serum is very important. In this study, the TEs of the newly prepared polymers with optimal DS were also tested in the presence of 10% serum. As shown in Figure 5a, compared to the results without the use of serum (Figure 4), polymers modified with hydroxyls showed better serum tolerance than PEI. For example, Ga-PEI gave six times higher TE than PEI, while the value was four in the absence of serum. Further, Ga-PEI also showed excellent TE toward other cell lines including HeLa (Figure 5b) and HEK293 (Figure 5c). 3.3. In Vitro Gene Transfection 3.3. In Vitro Gene Transfection Luciferase reporter gene was used to quantitatively assess the in vitro TE of these modified-PEIs in U-2OS cells. Figure 4 shows the relative TE of these polymers at various w/w ratios (2, 3, 4, 5) in comparison with PEI (25 kDa) at its optimal weight ratio of 1.4 (N/P ratio = 10) [24]. The conjugation of amino acids on PEI definitely enhanced the luciferase expression, and glycine-modified Gly-PEI showed the best TE, which was about 10 times higher than PEI. The DS also affects their gene delivery behavior: for each polymer, about 10% DS is most suitable for the transfection. Luciferase reporter gene was used to quantitatively assess the in vitro TE of these modiﬁed-PEIs in U-2OS cells. Figure 4 shows the relative TE of these polymers at various w/w ratios (2, 3, 4, 5) in comparison with PEI (25 kDa) at its optimal weight ratio of 1.4 (N/P ratio = 10) [24]. The conjugation of amino acids on PEI deﬁnitely enhanced the luciferase expression, and glycine-modiﬁed Gly-PEI showed the best TE, which was about 10 times higher than PEI. The DS also affects their gene delivery behavior: for each polymer, about 10% DS is most suitable for the transfection. It is known that the TE of PEI dramatically decreases in serum-supplemented medium [25–27]. Thus, the development of non-viral gene vectors that may retain their efﬁciency in serum is very important. In this study, the TEs of the newly prepared polymers with optimal DS were also tested in the presence of 10% serum. As shown in Figure 5a, compared to the results without the use of serum (Figure 4), polymers modiﬁed with hydroxyls showed better serum tolerance than PEI. For example, Ga-PEI gave six times higher TE than PEI, while the value was four in the absence of serum. Further, Ga-PEI also showed excellent TE toward other cell lines including HeLa (Figure 5b) and HEK293 (Figure 5c). In Hela cells, up to 23 times higher TE than PEI was achieved. In each cell line, the TE of 2323 Polymers 2015, 7, 2316–2331 y y g 3 3 I Vit G T Ga-PEI was higher than Ser-PEI, which also has hydroxyl groups but maintains the amount of amino groups. Polymers 2015, 7, 2316–2331 Polymers 2015, 7, 2316–2331 In addition, enhanced green ﬂuorescent protein (EGFP) expression assays were performed to directly visualize the infected cells expressing the pEGFP-Nl reporter gene. Figure 6 shows the density of transfected cells by modiﬁed polymers under the optimal weight ratios obtained in luciferase assays. The results indicate that the transfection mediated by these polymers was more effective than that obtained in the experiment involving PEI. In addition, the same assay was carried out in the presence of serum (Figure S4) or in other cell lines (HEK293, Figures S5 and S6; HeLa, Figures S7 and S8). All results further demonstrate the good serum tolerance of these vectors. Polymers 2015, 7, page–page Figure 6. Fluorescence microscope images of pEGFP-transfected U-2OS cells by the polycations at their optimal weight ratio. Figure 6. Fluorescence microscope images of pEGFP-transfected U-2OS cells by the polycations at their optimal weight ratio. Figure 6. Fluorescence microscope images of pEGFP-transfected U-2OS cells by the polycations at their optimal weight ratio. Figure 6. Fluorescence microscope images of pEGFP-transfected U-2OS cells by the polycations at their optimal weight ratio. 3.3. In Vitro Gene Transfection 3.3. In Vitro Gene Transfection 25 kDa PEI (w/w = 1.4, N/P = 10) was used as control. In addition enhanced green fluorescent protein (EGFP) expression assays were performed to Figure 5. Luciferase gene expression transfected in the presence of 10% serum in different cell lines. (a) U-2OS; (b) HeLa; (c) HEK293. Data represent mean ˘ SD (n = 3). 25 kDa PEI (w/w = 1.4, N/P = 10) was used as control. Figure 5. Luciferase gene expression transfected in the presence of 10% serum in different cell lines. (a) U-2OS; (b) HeLa; (c) HEK293. Data represent mean ± SD (n = 3). 25 kDa PEI (w/w = 1.4, N/P = 10) was used as control. Figure 5. Luciferase gene expression transfected in the presence of 10% serum in different cell lines. (a) U-2OS; (b) HeLa; (c) HEK293. Data represent mean ˘ SD (n = 3). 25 kDa PEI (w/w = 1.4, N/P = 10) was used as control. ing the p olymers u 2324 3.4. Biocompatibility Studies Th ti l h 3.4. Biocompatibility Studies The negatively charged cellular and blood components would also interact with the cationic polyplexes, leading to inherent toxicity [31]. Transfection involving polycations is usually accompanied by some degree of cell toxicity, which may limit their use as gene delivery vectors [32]. The cytotoxicities of these PEI derivatives were studied and compared with that of unmodified 25 kDa bPEI by CCK-8 assay at various concentrations. The effect of DS was first investigated. For Ser-PEI and Ga-PEI, the viability of U-2OS cells slightly increased with the rise of DS (Figure S9), indicating that the introduction of hydroxyl groups might benefit the biocompatibility of the polymeric materials [14,19,28]. Meanwhile, for other PEI derivatives, such a DS-depended trend was not found [13]. The cytotoxicities of these five polymers with their optimized DS, which were the same with those used in DLS studies, were subsequently measured within different cell lines, and the results are shown in Figure 7. As anticipated, these modified-PEI conjugates exhibited obviously lower cytotoxicity than PEI, especially in normal cells HEK293. Leu-PEI with hydrophobic side chains showed higher cytotoxicity than other polymers, suggesting that hydrophobic modification is prone to induce toxicity, which may come from the enhanced interaction with the cell membrane [33–35]. The negatively-charged cellular and blood components would also interact with the cationic polyplexes, leading to inherent toxicity [31]. Transfection involving polycations is usually accompanied by some degree of cell toxicity, which may limit their use as gene delivery vectors [32]. The cytotoxicities of these PEI derivatives were studied and compared with that of unmodiﬁed 25 kDa bPEI by CCK-8 assay at various concentrations. The effect of DS was ﬁrst investigated. For Ser-PEI and Ga-PEI, the viability of U-2OS cells slightly increased with the rise of DS (Figure S9), indicating that the introduction of hydroxyl groups might beneﬁt the biocompatibility of the polymeric materials [14,19,28]. Meanwhile, for other PEI derivatives, such a DS-depended trend was not found [13]. The cytotoxicities of these ﬁve polymers with their optimized DS, which were the same with those used in DLS studies, were subsequently measured within different cell lines, and the results are shown in Figure 7. As anticipated, these modiﬁed-PEI conjugates exhibited obviously lower cytotoxicity than PEI, especially in normal cells HEK293. 3.5. Cellular Uptake and Intracellular Distribution h h f h 3.5. Cellular Uptake and Intracellular Distribution To gain insight into the transfection mechanisms promoted by the title polymers, the cellular uptake of DNA was analyzed by flow cytometry. After 4 h incubation of polyplexes with U-2OS cells, the percentage of cells positive for Cy5-labelled pDNA was calculated and shown in Figure 9a. All of the polyplexes showed good cellular uptake, and about 90% of tested cells were positive for Cy5-labled DNA. Although Gly-PEI gave the best TE in luciferase assay (Figure 4), its cellular uptake was not the highest. Since cellular uptake is only one of the several barriers in the gene delivery process [28,40], we speculate that Gly-PEI may deliver the nucleic acid with higher intracellular efficiency. Moreover, the cellular uptake experiments were also carried out in the presence of 10% serum in the same cell line. The results shown in Figure 9b reveal that the cellular uptake of all polyplexes was negatively affected. However, it’s clearly shown that Ga-PEI gave the least uptake decrease, which might be attributed to its higher charge screening effect (Figure 3b) and better protein adsorption resistance (Figure 8). This also contributes to its higher serum tolerance than other polymers (Figure 5). To gain insight into the transfection mechanisms promoted by the title polymers, the cellular uptake of DNA was analyzed by ﬂow cytometry. After 4 h incubation of polyplexes with U-2OS cells, the percentage of cells positive for Cy5-labelled pDNA was calculated and shown in Figure 9a. All of the polyplexes showed good cellular uptake, and about 90% of tested cells were positive for Cy5-labled DNA. Although Gly-PEI gave the best TE in luciferase assay (Figure 4), its cellular uptake was not the highest. Since cellular uptake is only one of the several barriers in the gene delivery process [28,40], we speculate that Gly-PEI may deliver the nucleic acid with higher intracellular efﬁciency. Moreover, the cellular uptake experiments were also carried out in the presence of 10% serum in the same cell line. The results shown in Figure 9b reveal that the cellular uptake of all polyplexes was negatively affected. However, it’s clearly shown that Ga-PEI gave the least uptake decrease, which might be attributed to its higher charge screening effect (Figure 3b) and better protein adsorption resistance (Figure 8). This also contributes to its higher serum tolerance than other polymers (Figure 5). Polymers 2015, 7, 2316–2331 The serum-induced inhibition in gene transfection largely depends on the non-speciﬁc interaction between the polyplex and negatively charged serum protein [36–38]. The resistance against protein adsorption may favor the competitive approach of DNA cargo onto cell membrane, resulting in enhanced endocytosis and gene expression [39]. To verify this assumption, bovine serum albumin (BSA) was used as a model protein to simulate non-speciﬁc protein adsorption onto the surfaces of the polymers. As shown in Figure 8, modiﬁed PEI showed far lower protein adsorption than the unmodiﬁed one, especially for Ga-PEI. Such higher resistance against the negatively charged protein may contribute to its higher relative TE in serum condition (Figure 5). Polymers 2015, 7, page–page Figure 8. Comparison of protein adsorption between the polymers. 1 Milliliter of polymer solution (1 mg/mL) was added to 1 mL bovine serum albumin (BSA) solution (2 mg/mL) and co-incubated with shaking at 37 °C for 0.5 h prior to measurement. Data represent mean ± SD (n = 3). 5 C ll l U k d I ll l Di ib i Figure 8. Comparison of protein adsorption between the polymers. 1 Milliliter of polymer solution (1 mg/mL) was added to 1 mL bovine serum albumin (BSA) solution (2 mg/mL) and co-incubated with shaking at 37 ˝C for 0.5 h prior to measurement. Data represent mean ˘ SD (n = 3). Figure 8. Comparison of protein adsorption between the polymers. 1 Milliliter of polymer solution (1 mg/mL) was added to 1 mL bovine serum albumin (BSA) solution (2 mg/mL) and co-incubated with shaking at 37 °C for 0.5 h prior to measurement. Data represent mean ± SD (n = 3). Figure 8. Comparison of protein adsorption between the polymers. 1 Milliliter of polymer solution (1 mg/mL) was added to 1 mL bovine serum albumin (BSA) solution (2 mg/mL) and co-incubated with shaking at 37 ˝C for 0.5 h prior to measurement. Data represent mean ˘ SD (n = 3). 3.4. Biocompatibility Studies Th ti l h 3.4. Biocompatibility Studies Leu-PEI with hydrophobic side chains showed higher cytotoxicity than other polymers, suggesting that hydrophobic modiﬁcation is prone hich w 2325 Polymers 2015, 7, 2316–2331 to induce toxicity, which may come from the enhanced interaction with the cell membrane [33–35]. Among the ﬁve polymers, Ga-PEI showed the lowest cytotoxicity. This might be attributed to the replacement of amines with hydroxyls, which would shield the positive surface charge of the PEI (Figure 3b). Polymers 2015, 7, page–page than the unmodified one, especially for Ga-PEI. Such higher resistance against the negatively charged protein may contribute to its higher relative TE in serum condition (Figure 5). Figure 7. Cytotoxicity of the polycations at different concentrations toward (a) U-2OS; (b) HeLa and (c) HEK293 cells. Data represent mean ± SD (n = 3). Figure 7. Cytotoxicity of the polycations at different concentrations toward (a) U-2OS; (b) HeLa and (c) HEK293 cells. Data represent mean ˘ SD (n = 3). Figure 7. Cytotoxicity of the polycations at different concentrations toward (a) U-2OS; (b) HeLa and (c) HEK293 cells. Data represent mean ± SD (n = 3). Figure 7. Cytotoxicity of the polycations at different concentrations toward (a) U-2OS; (b) HeLa and (c) HEK293 cells. Data represent mean ˘ SD (n = 3). 2326 Polymers 2015, 7, 2316–2331 3.5. Cellular Uptake and Intracellular Distribution h h f h 3.5. Cellular Uptake and Intracellular Distribution Subsequently, the internalization and intracellular location of the delivered DNA (Cy5-labelled) by these polymers at the optimal transfection weight ratio were studied in U-2OS cells with CLSM. The nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI, blue). As shown in Figure 10, in the absence of serum, all polyplexes may effectively deliver a considerable amount of DNA (red) into the peri-nucleic region as well as nuclei after 4 h transfection. However, the red ﬂuorescence decreased for the PEI-mediated transfection in the presence of serum. Meanwhile, there was no obvious change of the ﬂuorescence density in the transfection mediated by the modiﬁed PEIs. For Ga-PEI mediated transfection, many more red signals were found in the nuclei, indicating its potential as non-viral gene delivery vector with higher TE, especially in the serum circumstance. 2327 Polymers 2015, 7, 2316–2331 decrease, which m protein adsorption other polymers (Figure 5). Figure 9. Cellular uptake of polyplexes at optimal weight ratio in U-2OS cells quantified by flow cytometry analysis. (a) Without serum; (b) In the presence of 10% serum. Data represent mean ± SD (n = 3). Figure 9. Cellular uptake of polyplexes at optimal weight ratio in U-2OS cells quantiﬁed by ﬂow cytometry analysis. (a) Without serum; (b) In the presence of 10% serum. Data represent mean ˘ SD (n = 3). Polymers 2015, 7, page–page in the absence of serum, all polyplexes may effectively deliver a considerable amount of DNA (red) into the peri-nucleic region as well as nuclei after 4 h transfection. However, the red fluorescence decreased for the PEI-mediated transfection in the presence of serum. Meanwhile, there was no obvious change of the fluorescence density in the transfection mediated by the modified PEIs. For Ga-PEI mediated transfection many more red signals were found in the nuclei indicating its other polymers (Figure 5). Polymers 2015, 7, page–page in the absence of serum all polyplexes may effectively deliver a considerable amount of DNA (red) Figure 9. Cellular uptake of polyplexes at optimal weight ratio in U-2OS cells quantified by flow cytometry analysis. (a) Without serum; (b) In the presence of 10% serum. Data represent mean ± SD (n = 3). Figure 9. Cellular uptake of polyplexes at optimal weight ratio in U-2OS cells quantiﬁed by ﬂow cytometry analysis. (a) Without serum; (b) In the presence of 10% serum. Data represent mean ˘ SD (n = 3). 3.5. Cellular Uptake and Intracellular Distribution h h f h 3.5. Cellular Uptake and Intracellular Distribution into the peri-nucleic region as well as nuclei after 4 h transfection. However, the red fluorescence decreased for the PEI-mediated transfection in the presence of serum. Meanwhile, there was no obvious change of the fluorescence density in the transfection mediated by the modified PEIs. For Ga PEI mediated transfection many more red signals were found in the nuclei indicating its 12 by t ese po y e s at t e opti a t a s ectio eig t atio e e studied i U OS ce s it C SM The nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI, blue). As shown in Figure 10, Figure 10. CLSM images of U-2OS cells transfected with Cy5-labelled DNA by polymers at their respective optimal weight ratio in the absence (upper) and presence (lower) of serum. PEI as control (w/w = 1.4, N/P = 10). For each triad image, left: cell nuclei stained by DAPI (blue); middle: Cy5-labeled pDNA (red); right: merged image. 4. Conclusions Figure 10. CLSM images of U-2OS cells transfected with Cy5-labelled DNA by polymers at their respective optimal weight ratio in the absence (upper) and presence (lower) of serum. PEI as control (w/w = 1.4, N/P = 10). For each triad image, left: cell nuclei stained by DAPI (blue); middle: Cy5-labeled pDNA (red); right: merged image. Figure 10. CLSM images of U-2OS cells transfected with Cy5-labelled DNA by polymers at their respective optimal weight ratio in the absence (upper) and presence (lower) of serum. PEI as control (w/w = 1.4, N/P = 10). For each triad image, left: cell nuclei stained by DAPI (blue); middle: Cy5-labeled pDNA (red); right: merged image. 4. Conclusions Figure 10. CLSM images of U-2OS cells transfected with Cy5-labelled DNA by polymers at their respective optimal weight ratio in the absence (upper) and presence (lower) of serum. PEI as control (w/w = 1.4, N/P = 10). For each triad image, left: cell nuclei stained by DAPI (blue); middle: Cy5-labeled pDNA (red); right: merged image. ophilic/hy er sizes an 2328 Polymers 2015, 7, 2316–2331 Supplementary Materials: Supplementary materials can be found at www.mdpi.com/2073-4360/7/11/1516/s1. Supplementary Materials: Supplementary materials can be found at www.mdpi.com/2073-4360/7/11/1516/ Acknowledgments: This work was ﬁnancially supported by the National Science Foundation of China (Nos. 21472131, 21232005, J1310008 and J1103315), and the Specialized Research Fund for the Doctoral Program of Higher Education in China (20120181130006). Author Contributions: Qin-Fang Zhang, Ji Zhang and Xiao-Qi Yu designed the studies. Qin-Fang Zhang performed the chemical and biological experiments, Chao-Ran Luan and Dong-Xiao Yin performed the chemical experiments, Yan-Hong Liu and Qi Peng performed the biological experiments, Qin-Fang Zhang, Ji Zhang, Yong Xu and Xiao-Qi Yu prepared the manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest. 4. Conclusions Branched PEI (25 kDa) was modiﬁed with several amino acids or other functional acids to give the derivative polymers with various DS. Different carboxylic acids were employed to endue the new polymers with different charge and hydrophilic/hydrophobic properties. These polymers may condense DNA into nanoparticles with proper sizes and zeta potentials. In vitro experiments showed that compared to PEI, they could give higher TE and lower cytotoxicity. Up to 23 times higher TE than PEI was achieved by employing Ga-PEI as transfection reagent. The effects of DS and substituted groups were also investigated, and 10% DS was found to be most suitable for transfection. Ga-PEI, in which some amino groups were replaced by hydroxyls, showed the highest serum tolerance and also the highest TE in several cell lines with the presence of serum. BSA adsorption and ﬂow cytometry assays also proved its higher biocompatibility. Such results may guide a way to design PEI derivatives as non-viral gene vectors with higher efﬁciency and lower toxicity. References 1. Hunt, K.K.; Vorburger, S.A. Gene therapy: Hurdles and hopes for cancer treatment. Science 2002, 297, 415–416. [CrossRef] [PubMed] 1. Hunt, K.K.; Vorburger, S.A. Gene therapy: Hurdles and hopes for cancer treatment. Science 2002, 297, 415–416. [CrossRef] [PubMed] 2. Niidome, T.; Huang, L. Gene therapy progress and prospects: Nonviral vectors. Gene Ther. 2002, 9, 1647–1652. [CrossRef] [PubMed] 2. Niidome, T.; Huang, L. Gene therapy progress and prospects: Nonviral vectors. Gene Ther. 2002, 9, 1647–1652. [CrossRef] [PubMed] 3. Xu, F.J.; Yang, W.T. Polymer vectors via controlled/living radical polymerization for gene delivery. Prog. Polym. Sci. 2011, 36, 1099–1131. [CrossRef] 3. Xu, F.J.; Yang, W.T. Polymer vectors via controlled/living radical polymerization for gene delivery. Prog. Polym. Sci. 2011, 36, 1099–1131. [CrossRef] 4. Bisht, H.S.; Manickam, D.S.; You, Y.; Oupicky, D. Temperature-controlled properties of DNA complexes with poly(ethylenimine)-graft-poly(N-iso-propylacrylamide). Biomacromolecules 2006, 7, 1169–1178. [CrossRef] [PubMed] 5. Wagner, E.; Ogris, M.; Zauner, W. Polylysine-based transfection systems utilizing receptor-mediated delivery. Adv. Drug Deliv. Rev. 1998, 30, 97–113. [PubMed] 6. Kukowska-Latallo, J.F.; Bielinska, A.U.; Johnson, J.; Spindler, R.; Tomalia, D.A.; Baker, J.R., Jr. Efﬁcient transfer of genetic material into mammalian cells using starburst polyamidoamine dendrimers. Proc. Natl. Acad. Sci. USA 1996, 93, 4897–4902. [CrossRef] [PubMed] 7. Mintzer, M.A.; Simanek, E.E. Nonviral vectors for gene delivery. Chem. Rev. 2009, 109, 259–302. [CrossRef] [PubMed] 8. Baker, A.; Saltik, M.; Lehrmann, H.; Killisch, I.; Mautner, V.; Lamm, G.; Christofori, G.; Cotton, M. Polyethylenimine (PEI) is a simple, inexpensive and effective reagent for condensing and linking plasmid DNA to adenovirus for gene delivery. Gene Ther. 1997, 4, 773–782. [CrossRef] [PubMed] 9. Boussif, O.; Lezoualch, F.; Zanta, M.A.; Mergny, M.D.; Scherman, D.; Demeneix, B.; Behr, J.P. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: Polyethylenimine. Proc. Natl. Acad. Sci. USA 1995, 92, 7297–7301. [CrossRef] [PubMed] 10. Tian, H.Y.; Lin, L.; Chen, J.; Chen, X.S.; Park, T.G.; Maruyama, A. RGD targeting hyaluronic acid coating system for PEI-PBLG polycation gene carriers. J. Control. Release 2011, 155, 47–53. [CrossRef] [PubMed] 11. Luo, X.H.; Huang, F.W.; Qin, S.Y.; Wang, H.F.; Feng, J.; Zhang, X.Z.; Zhuo, R.X. A strategy to improve serum-tolerant transfection activity of polycation vectors by surface hydroxylation. Biomaterials 2011, 32, 9925–9939. [CrossRef] [PubMed] 2329 Polymers 2015, 7, 2316–2331 12. Yan, X.B.; Zhang, Y.; Zhang, H.; Wang, P.G.; Chua, X.; Wang, X. Amphiphilic polyethylenimine (PEI) as highly efﬁcient non-viral gene carrier. Org. Biomol. Chem. 2014, 12, 1975–1982. References [CrossRef] [PubMed] 13. Dehshahri, A.; Oskuee, R.K.; Shier, W.T.; Hateﬁ, A.; Ramezani, M. Gene transfer efﬁciency of high primary amine content, hydrophobic, alkyl-oligoamine derivatives of polyethylenimine. Biomaterials 2009, 30, 4187–4194. [CrossRef] [PubMed] 14. Clements, B.A.; Bai, J.; Kucharski, C.; Farrell, L.L.; Lavasanifar, A.; Ritchie, B.; Ghahary, A.; Uludag, H. RGD conjugation to polyethyleneimine does not improve DNA delivery to bone marrow stromal cells. Biomacromolecules 2006, 7, 1481–1488. [CrossRef] [PubMed] 15. Dong, X.; Tian, H.Y.; Chen, L.; Chen, J.; Chen, X.S. Biodegradable mPEG-b-P(MCC-g-OEI) copolymers for efﬁcient gene delivery. J. Control. Release 2011, 152, 135–142. [CrossRef] [PubMed] 16. Yang, C.; Ong, Z.Y.; Yang, Y.Y.; Ee, P.L.; Hedrick, J.L. Novel biodegradable block copolymers of poly(ethylene glycol) (PEG) and cationic polycarbonate: Effects of PEG conﬁguration on gene delivery. Macromol. Rapid Commun. 2011, 32, 1826–1833. [CrossRef] [PubMed] 17. Petersen, H.; Merdan, T.; Kunath, K.; Fischer, D.; Kissel, T. Poly(ethylenimine-co-L-lactamide-co- succinamide): A biodegradable polyethylenimine derivative with an advantageous pH-dependent hydrolytic degradation for gene delivery. Bioconjug. Chem. 2002, 13, 812–821. [CrossRef] [PubMed] 18. Makino, A.; Hara, E.; Hara, I.; Ozeki, E.; Kimura, S. Size control of core-shell-type polymeric micel nanometer precision. Langmuir 2014, 30, 669–674. [CrossRef] [PubMed] 19. Tripathi, S.K.; Yadav, S.; Gupta, K.C.; Kumar, P. Synthesis and evaluation of N-(2,3-dihydroxypropyl)-PEIs as efﬁcient vectors for nucleic acids. Mol. Biosyst. 2012, 8, 1426–1434. [CrossRef] [PubMed] 20. Layek, B.; Haldar, M.K.; Sharma, G.; Lipp, L.; Mallik, S.; Singh, J. Hexanoic acid and polyethylene glycol double grafted amphiphilic chitosan for enhanced gene delivery: Inﬂuence of hydrophobic and hydrophilic substitution degree. Mol. Pharm. 2014, 11, 982–994. [CrossRef] [PubMed] 20. Layek, B.; Haldar, M.K.; Sharma, G.; Lipp, L.; Mallik, S.; Singh, J. Hexanoic acid and polyethylene glycol double grafted amphiphilic chitosan for enhanced gene delivery: Inﬂuence of hydrophobic and hydrophilic substitution degree. Mol. Pharm. 2014, 11, 982–994. [CrossRef] [PubMed] 21. Xun, M.-M.; Xiao, Y.-P.; Zhang, J.; Liu, Y.-H.; Peng, Q.; Guo, Q.; Wu, W.-X.; Xu, Y.; Yu, X.-Q. Low molecular weight PEI-based polycationic gene vectors via Michael addition polymerization with improved serum-tolerance. Polymer 2015, 65, 45–54. [CrossRef] 21. Xun, M.-M.; Xiao, Y.-P.; Zhang, J.; Liu, Y.-H.; Peng, Q.; Guo, Q.; Wu, W.-X.; Xu, Y.; Yu, X.-Q. Low molecular weight PEI-based polycationic gene vectors via Michael addition polymerization with improved serum-tolerance. Polymer 2015, 65, 45–54. [CrossRef] y 22. Liu, Y.M.; Reineke, T.M. Hydroxyl stereo chemistry and amine number within poly(glycoamidoamine)s affect intracellular DNA delivery. J. Am. Chem. Soc. 2005, 127, 3004–3015. References [CrossRef] [PubMed] 22. Liu, Y.M.; Reineke, T.M. Hydroxyl stereo chemistry and amine number within poly(glycoamidoamine)s affect intracellular DNA delivery. J. Am. Chem. Soc. 2005, 127, 3004–3015. [CrossRef] [PubMed] 23. Petersen, H.; Fechner, P.M.; Martin, A.L.; Kunath, K.; Stolnik, S.; Roberts, C.J.; Fischer, D.; Davies, M.C.; Kissel, T. Polyethylenimine-graft-poly(ethylene glycol) copolymers: Inﬂuence of copolymer block structure on DNA complexation and biological activities as gene delivery system. Bioconjug. Chem. 2002, 13, 845–854. [CrossRef] [PubMed] 24. Xiu, K.M.; Yang, J.J.; Zhao, N.N.; Li, J.S.; Xu, F.J. Multiarm cationic star polymers by atom transfer radical polymerization from β-cyclodextrin cores: Inﬂuence of arm number and length on gene delivery. Acta Biomater. 2013, 9, 4726–4733. [CrossRef] [PubMed] 25. Dai, F.; Liu, W. Enhanced gene transfection and serum stability of polyplexes by PDMAEMA-polysulfobetaine diblock copolymers. Biomaterials 2011, 32, 628–638. [CrossRef] [PubMed] 26. He, Y.; Cheng, G.; Xie, L.; Nie, Y.; He, B.; Gu, Z. Polyethyleneimine/DNA polyplexes with reduction-sensitive hyaluronic acid derivatives shielding for targeted gene delivery. Biomaterials 2013, 34, 1235–1245. [CrossRef] [PubMed] 27. Xu, P.; Quick, G.K.; Yeo, Y. Gene delivery through the use of a hyaluronate-associated intrace degradable crosslinked polyethyleneimine. Biomaterials 2009, 30, 5834–5843. [CrossRef] [PubMed] 28. Zhang, Q.-F.; Yi, W.-J.; Wang, B.; Zhang, J.; Ren, L.; Chen, Q.-M.; Guo, L.; Yu, X.-Q. Linear polycations by ring-opening polymerization as non-viral gene delivery vectors. Biomaterials 2013, 34, 5391–5401. [CrossRef] [PubMed] 29. Chen, C.-K.; Jones, C.H.; Mistriotis, P.; Yu, Y.; Ma, X.; Ravikrishnan, A.; Jiang, M.; Andreadis, S.T.; Pfeifer, B.A.; Cheng, C. Poly(ethylene glycol)-block-cationic polylactide nanocomplexes of differing charge density for gene delivery. Biomaterials 2013, 34, 9688–9699. [CrossRef] [PubMed] 30. Tian, H.; Guo, Z.; Lin, L.; Jiao, Z.; Chen, J.; Gao, S.; Zhu, X.; Chen, X. pH-responsive zwitterionic copolypeptides as charge conversional shielding system for gene carriers. J. Control. Release 2014, 174, 117–125. [CrossRef] [PubMed] 31. Fischer, D.; Li, Y.X.; Ahlemeyer, B.; Krieglstein, J.; Kissel, T. In vitro cytotoxicity testing of polycations: Inﬂuence of polymer structure on cell viability and hemolysis. Biomaterials 2003, 24, 1121–1131. [CrossRef] 2330 Polymers 2015, 7, 2316–2331 32. Tao, X.; Wu, R.; Xia, Y.; Huang, H.; Chai, W.; Feng, T.; Gan, Y.; Zhang, W. Biotemplated fabrication of Sn@C anode materials based on the unique metal biosorption behavior of microalgae. ACS Appl. Mater. Interfaces 2014, 6, 3969–3702. [CrossRef] [PubMed] 33. Chang, H.; Wang, H.; Shao, N.; Wang, M.; Wang, X.; Cheng, Y. Surface-engineered dendrimers with a diaminododecane core achieve efﬁcient gene transfection and low cytotoxicity. References Bioconjugate Chem. 2014, 25, 342–350. [CrossRef] [PubMed] 34. Zhang, Q.-F.; Wang, B.; Yin, D.-X.; Zhang, J.; Wu, W.-X.; Yu, Q.-Y.; Yu, X.-Q. Linear TACN-based cationic polymers as non-viral gene vectors. RSC Adv. 2014, 4, 59164–59174. [CrossRef] 35. Yi, W.-J.; Zhang, Q.-F.; Zhang, J.; Liu, Q.; Ren, L.; Chen, Q.-M.; Guo, L.; Yu, X.-Q. Cyclen-based lipidic oligomers as potential gene delivery vehicles. Acta Biomater. 2014, 10, 1412–1422. [CrossRef] [PubMed] 35. Yi, W. J.; Zhang, Q. F.; Zhang, J.; Liu, Q.; Ren, L.; Chen, Q. M.; Guo, L.; Yu, X. Q. Cyclen based lipidic oligomers as potential gene delivery vehicles. Acta Biomater. 2014, 10, 1412–1422. [CrossRef] [PubMed] 36. Funhoff, A.M.; van Nostrum, C.F.; Lok, M.C.; Fretz, M.M.; Crommelin, D.J.A.; Hennink, W.E. Poly(3-guanidinopropyl methacrylate): A novel cationic polymer for gene delivery. Bioconjug. Chem. 2004, 15, 1212–1220. [CrossRef] [PubMed] 37. Son, S.; Kim, W.J. Biodegradable nanoparticles modiﬁed by branched polyethylenimine for plasmid DNA delivery. Biomaterials 2010, 31, 133–143. [CrossRef] [PubMed] 38. Jones, C.H.; Chen, C.-K.; Jiang, M.; Fang, L.; Cheng, C.; Pfeife, B.A. Synthesis of cationic polylactides with tunable charge densities as nanocarriers for effective gene delivery. Mol. Pharm. 2013, 10, 1138–1145. [CrossRef] [PubMed] 39. Liu, Z.; Janzena, J.; Brooks, D.E. Adsorption of amphiphilic hyperbranched polyglycerol derivatives onto human red blood cells. Biomaterials 2010, 31, 3364–3373. [CrossRef] [PubMed] 40. Gabrielson, N.P.; Pack, D.W. Acetylation of polyethylenimine enhances gene delivery via weakened polymer/DNA interactions. Biomacromolecules 2006, 7, 2427–2435. [CrossRef] [PubMed] © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). 2331 2331
https://openalex.org/W4220779623	https://kneesurgrelatres.biomedcentral.com/track/pdf/10.1186/s43019-022-00140-8	English	null	Agreement and accuracy of radiographic assessment using a decision aid for medial Oxford partial knee replacement: multicentre study	Knee surgery & related research	2,022	cc-by	7,240	Abstract Background: Indication for mobile-bearing partial knee replacement (PKR) is made on the basis of a radiological decision aid. This study aimed to reveal the inter-rater reproducibility and accuracy of the decision aid when used by experienced surgeons. Patients and methods: Anonymised radiographic image sets (anteroposterior, lateral, varus/valgus stress in 20° knee flexion, and skyline views) from 20 consecutive patients who underwent knee replacement were assessed by 12 experienced surgeons. Agreements of each section and accuracy were compared by intra-operative inspection of the status of the anterior cruciate ligament (ACL) and medial and lateral cartilage according to the protocol of Radio- graphic Assessment for Medial Oxford PKR. Fleiss’ kappa (κ) values were used as a statistical measure. Results: Full-thickness medial cartilage had the best agreement between the surgeons (κ = 94.7%) and best accuracy (94.2%). Although functioning ACL (90.8%), intact cartilage (91.7%) and full-thickness lateral cartilage defects (86.1%) were accurately diagnosed, diagnoses of deficient ACL (up to 42.5%) and partial-thickness lateral cartilage defects (11.7%) were poor; they were sometimes misdiagnosed as being intact. Moreover, agreement of lateral and valgus stress radiographs regarding intact MCL function, as well as the overall decision, was considered to be inadequate (κ = 0.47, 0.58 and 0.51, respectively). Conclusions: Although the radiological aid is useful for selection of patients who are likely to be suitable for PKR, surgeons should still carefully assess the lateral weight-bearing area for partial-thickness loss and deficiency of the ACL because they were sometimes overlooked by surgeons using radiographs. MRI will be helpful to improve the accuracy of determination of Oxford PKR indication. Keywords: Operation, Indication, Unicompartmental knee arthroplasty, Radiography Agreement and accuracy of radiographic assessment using a decision aid for medial Oxford partial knee replacement: multicentre study Takafumi Hiranaka1,8* , Ryosuke Furuhashi2, Kenichiro Takashiba3, Takao Kodama4, Kazuhiko Michishita5, Hiroshi Inui6 and Eita Togashi7 © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Open Access Open Access Knee Surgery & Related Research Knee Surgery & Related Research Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 https://doi.org/10.1186/s43019-022-00140-8 1) Space between the femoral and tibial bony surface This study aims to reveal the inter-rater reproducibility and accuracy of the decision aid when used by experienced surgeons. and reduced mortality, as well as superior patient satis- faction [1–3]. On the other hand, national registry data revealed that the revision rate was higher for PKR than for TKR, even after adjustment for the pre-operative con- ditions [2]. Patient selection is reported to be the key to successful PKR. Intact lateral cartilage and intact ante- rior cruciate ligament (ACL) are necessary conditions for successful application of PKR [4]. In addition to this, Kozinn and Scott [5] proposed a strict indication recom- mending that patients over 82 kg, younger than 60 years, who are extremely physically active or who perform heavy labour, or who have chondrocalcinosis or exposed bone in the patellofemoral joint should be contraindi- cated. As a result, the usage of PKR has been reported to be just 9.3% in Sweden [6], 8% in the USA [7], 11.2% in the United Kingdom [8] and 5.4% in Australia [9]. How- ever, these indications are for fixed-bearing PKR. Else- where, Liddle et al. reported that the best result could be achieved when the usage of PKR ranged between 40% and 60% of all cases of knee arthroplasty [10]. As PKR has been assumed to be a technically demanding opera- tion, increasing the number of PKRs under validated indication could improve the post-operative survivorship after PKR [11], this difference can be achieved by patient selection. In their examination of the best indication for PKR, White et al. introduced the concept of anteromedial osteoarthritis (AMOA) to show indication for mobile- bearing PKR [12]. Knees that have bone-on-bone (full- thickness cartilage defect) in the medial compartment, intact cartilage in the lateral compartment, a functionally normal medial collateral ligament (MCL) and anterior cruciate ligament (ACL), and acceptable patellofemo- ral joint arthritis are considered to have AMOA [4, 13]. AMOA can reportedly be detected by plain varus and valgus radiographs. A radiological decision aid was intro- duced to enable sophisticated decision-making regarding mobile-bearing PKR [14]. Hamilton et al. reported that clinical results were slightly better in patients who met the criteria compared with those who did not, and that approximately half of all patients met the criteria [15]. The decision was made by a single senior surgeon, how- ever, and its interrater reliability was not evaluated in the paper. 1) Space between the femoral and tibial bony surface 1) Space between the femoral and tibial bony surface and reduced mortality, as well as superior patient satis- faction [1–3]. On the other hand, national registry data revealed that the revision rate was higher for PKR than for TKR, even after adjustment for the pre-operative con- ditions [2]. Patient selection is reported to be the key to successful PKR. Intact lateral cartilage and intact ante- rior cruciate ligament (ACL) are necessary conditions for successful application of PKR [4]. In addition to this, Kozinn and Scott [5] proposed a strict indication recom- mending that patients over 82 kg, younger than 60 years, who are extremely physically active or who perform heavy labour, or who have chondrocalcinosis or exposed bone in the patellofemoral joint should be contraindi- cated. As a result, the usage of PKR has been reported to be just 9.3% in Sweden [6], 8% in the USA [7], 11.2% in the United Kingdom [8] and 5.4% in Australia [9]. How- ever, these indications are for fixed-bearing PKR. Else- where, Liddle et al. reported that the best result could be achieved when the usage of PKR ranged between 40% and 60% of all cases of knee arthroplasty [10]. As PKR has been assumed to be a technically demanding opera- tion, increasing the number of PKRs under validated indication could improve the post-operative survivorship after PKR [11], this difference can be achieved by patient selection. In their examination of the best indication for PKR, White et al. introduced the concept of anteromedial osteoarthritis (AMOA) to show indication for mobile- bearing PKR [12]. Knees that have bone-on-bone (full- thickness cartilage defect) in the medial compartment, intact cartilage in the lateral compartment, a functionally normal medial collateral ligament (MCL) and anterior cruciate ligament (ACL), and acceptable patellofemo- ral joint arthritis are considered to have AMOA [4, 13]. AMOA can reportedly be detected by plain varus and valgus radiographs. A radiological decision aid was intro- duced to enable sophisticated decision-making regarding mobile-bearing PKR [14]. Hamilton et al. reported that clinical results were slightly better in patients who met the criteria compared with those who did not, and that approximately half of all patients met the criteria [15]. The decision was made by a single senior surgeon, how- ever, and its interrater reliability was not evaluated in the paper. In addition, the accuracy of the decision aid was not described. 1) Space between the femoral and tibial bony surface On the contrary, in the case of an ACL deficiency, the tibia moves forward and the cartilage of the posterior tibial plateau could be worn out. The exposed bony surfaces contact each other even in the knee flexion position, and eventually the MCL will be shortened; the varus is therefore not correctable. Such knees are unsuitable for PKR. f 4) Functionally normal MCL If the MCL is functional and not contracted, it retains its original length and any varus deformity will be correctable; this also implies that the ACL is intact [12]. If the ACL is intact, although the MCL will shrink during knee extension because of carti- lage wear, its length can be restored in knee flexion, when the condyles with intact cartilage thickness are facing each other. Consequently, the MCL never contracts and the varus is therefore correctable. On the contrary, in the case of an ACL deficiency, the tibia moves forward and the cartilage of the posterior tibial plateau could be worn out. The exposed bony surfaces contact each other even in the knee flexion position, and eventually the MCL will be shortened; the varus is therefore not correctable. Such knees are unsuitable for PKR. 5) Acceptable patellofemoral joint 5) Acceptable patellofemoral joint Medial facet OA, with or without bone loss, and lat- eral facet OA without bone loss are accepted. Lateral facet OA with bone loss, grooving or subluxation is unacceptable [18]. 6) Overall decision on PKR suitability All sections are rated “yes” or “no”. Knees with “yes” for all sections were considered to be suitable for mobile-bearing PKR; they were otherwise rated as unsuitable. Introductionh There is increasing interest in partial knee replacement (PKR) as it has been reported to be an efficient treat- ment option for knee osteoarthritis (OA). Advantages over total knee replacement (TKR) include faster recov- ery, deeper flexion angle, fewer systemic complications *Correspondence: Takafumi.hiranaka@gmail.com 8 Department of Orthopaedic Surgery and Joint Surgery Centre, Takatsuki General Hospital, 1‑3‑13, Kosobe‑Cho, Takatsuki City, Osaka 561‑1115, Japan p Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Page 2 of 9 Hiranaka et al. Knee Surgery & Related Research 1) Space between the femoral and tibial bony surface In addition, the accuracy of the decision aid was not described. This study aims to reveal the inter-rater reproducibility and accuracy of the decision aid when used by experienced surgeons. Medial bone-on-bone is evaluated on varus radio- graphs and indicates a full-thickness cartilage defect in the medial compartment. A space between the femoral and tibial bony surface can imply a partial- thickness cartilage defect that would rule out PKR. 2) Functionally intact ACL The location (if present) of a bony erosion may be seen on lateral radiographs. When the ACL is intact, the erosion is located anteriorly or is not seen. If it locates and/or extends posteriorly, the ACL would be deficient [16]. i 3) Full-thickness lateral cartilage Lateral cartilage thickness was assessed on valgus stress radiographs. We considered the lateral car- tilage to be intact if the lateral joint space was fully retained. Any osteophytes on the lateral condyle were ignored because their existence has not been reported to affect the clinical outcome [17]. pf 4) Functionally normal MCL If the MCL is functional and not contracted, it retains its original length and any varus deformity will be correctable; this also implies that the ACL is intact [12]. If the ACL is intact, although the MCL will shrink during knee extension because of carti- lage wear, its length can be restored in knee flexion, when the condyles with intact cartilage thickness are facing each other. Consequently, the MCL never contracts and the varus is therefore correctable. On the contrary, in the case of an ACL deficiency, the tibia moves forward and the cartilage of the posterior tibial plateau could be worn out. The exposed bony surfaces contact each other even in the knee flexion position, and eventually the MCL will be shortened; the varus is therefore not correctable. Such knees are unsuitable for PKR. f 4) Functionally normal MCL If the MCL is functional and not contracted, it retains its original length and any varus deformity will be correctable; this also implies that the ACL is intact [12]. If the ACL is intact, although the MCL will shrink during knee extension because of carti- lage wear, its length can be restored in knee flexion, when the condyles with intact cartilage thickness are facing each other. Consequently, the MCL never contracts and the varus is therefore correctable. The radiological decision aidhi This study was approved by the institutional review board of our hospital, and written informed consent was obtained from all patients. This study included 20 consecutive patients who underwent single-sided total The decision aid contains five criteria based on plain anteroposterior and lateral radiographs along with varus and valgus stress radiographs at 20° flexion [14]: Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Page 3 of 9 Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Hiranaka et al. Knee Surgery & Related Research Regarding the lateral femoral condyle, the ability to diag- nose it as intact or defective (partial, full-thickness or both) was assessed. knee replacement (TKR) or (PKR) in the corresponding author’s hospital in August 2019. Pre-operative radiogra- phy sets including anteroposterior (AP), true lateral, val- gus and varus stress at 20° flexion and skyline view were anonymised and prepared. Statistical analysis We recruited 12 experienced surgeons with extensive knowledge and skills who are domestic instructors in the use of mobile-bearing PKR. The average years of experi- ence of surgery and average number of PKRs performed per year of the participating surgeons were 26.3 years (15–43 years) and 72.5 cases (24–150 cases), respectively. The radiographic datasets were sent electronically to the participants, and rating was performed in each hospi- tal. The reviewers rated “yes” or “no” for each section on the basis of the 20 radiograph sets, and an overall deci- sion of suitability or non-suitability for PKR was made on the basis of the rates of the five sections. The results were sent back to the corresponding author’s hospital for analysis. Fleiss’ kappa values were calculated to evaluate the relia- bility of agreement between the raters. Calculations were performed using Microsoft Excel (Microsoft Corp., Red- mond, WA). The predictability of ACL status, lateral car- tilage status and overall PKR suitability were compared using repeated measures analysis of variance followed by Bonferroni’s multiple comparison test. Moreover, Pear- son’s correlation coefficient between the predictabilities and surgeons’ years of experience and the number of surgeries were calculated. Analysis was performed using easy R (EZR; Jichi Medical University, Japan) running on R (R Foundation for Statistical Computing, Vienna, Aus- tria) [20]. The condition of the ACL and the cartilage at the medial femoral condyle, medial tibial plateau and lat- eral femoral condyle (weight-bearing area) was assessed intra-operatively (Table 1). The assessment was done by the corresponding author and another assistant surgeon. If their diagnoses differed, re-assessment was performed. If the decision was still different, the corresponding author’s decision was used for analysis. If the ACL was graded as normal, with synovial damage or with longi- tudinal split, it was considered to be functioning; it was otherwise classified as deficient [19]. If the cartilage was rated as normal or as having superficial damage, it was deemed to be intact; it was otherwise rated as defective. Regarding the lateral femoral condyle, a full-thickness cartilage defect at the lateral edge was ignored because the lesion was in the non-weight-bearing area. Radio- graphic decisions were validated on the basis of intra- operative findings.h Results On the basis of intra-operative inspection, 8 of the 20 cases were considered suitable for PKR (40%). Interest- ingly, the percentage of knees which the respective raters considered to be suitable for PKR varied between 45% and 75% with a mean of 60%, showing a higher percent- age than the suitability based on macroscopic findings. These values were not significantly correlated with the rates of PKR usage (r = 0.09, P = 0.77).h The details of the decisions based on the radiographs and the intra-operative findings are presented in Table 2. The accuracy of the diagnosis of the respective sections is summarised in Table 3. The most accurately diagnosed section was medial bone-on-bone, indicating full-thick- ness cartilage loss on both condyles. Although a full- thickness defect of the medial condyle was detected in every case intra-operatively, two cases were not outlined as bone-on-bone. These cases had a severe loss of bone with grooving in the medial tibial plateau. Consequently, there was still some space between the bones due to non- conformity between the articular surfaces (Fig. 1). i The accuracy of the estimation of the status of medial femoral and tibial condyles, lateral femoral condyles and ACL (functioning or deficient) was evaluated on the basis of intra-operative inspection, and the percentage of knees whose condition was correctly predicted was calculated. Table 1 Status of ACL and lateral cartilage ACL status Lateral cartilage status Normal ACL functioning (ACLF) Normal Intact Synovial damage Superficial damage Longitudinal split Partial-thickness defect Defective Friable and fragmented ACL deficient (ACLD) Full-thickness defect Absent Bone loss Table 1 Status of ACL and lateral cartilage Hiranaka et al. Results Knee Surgery & Related Research (2022) 34:13 Page 4 of 9 Table 2 Details of the decision for each section in each case BnB, medial bone on bone; ACL, functionally intact ACL; LC, intact lateral cartilage; MCL, functionally intact MCL; PFJ, acceptable patellofemoral joint change Case BnB ACL LC MCL PFJ Overall ACL status Lateral cartilage PKR suitability Correct diagnosis (%) 1 12 12 7 11 10 8 Functioning Normal Yes 67 2 12 2 2 9 4 0 Deficient Full-thickness defect No 100 3 12 12 12 12 12 12 Deficient Normal No 100 4 12 7 12 10 12 7 Functioning Normal Yes 58 5 12 12 7 9 12 7 Deficient Partial-thickness defect No 42 6 12 12 12 12 10 11 Deficient Partial-thickness defect No 8 7 12 7 11 12 10 7 Deficient Partial-thickness defect No 42 8 12 8 11 12 12 9 Functioning Superficial damage Yes 75 9 5 6 3 7 10 0 Deficient Full-thickness defect No 100 10 12 4 10 10 12 3 Deficient Superficial damage No 75 11 12 11 12 12 12 11 Functioning Superficial damage Yes 92 12 12 11 12 12 12 11 Functioning Normal Yes 92 13 12 12 12 12 12 12 Functioning Normal Yes 100 14 12 12 12 12 12 12 Functioning Normal Yes 100 15 12 11 12 12 12 12 Functioning Normal Yes 100 16 12 7 11 9 11 6 Deficient Superficial damage No 50 17 12 9 11 8 6 5 Functioning Partial-thickness defect No 58 18 12 12 12 11 12 11 Deficient Partial-thickness defect No 8 19 12 5 0 8 5 2 Deficient Full-thickness defect No 83 20 5 4 9 8 11 0 Functioning Normal Yes 0 Table 2 Details of the decision for each section in each case Table 3 Accuracy of diagnosis of the intra-operative inspection Joint status Radiographs used Accuracy (%) Medial bone-on-bone Varus stress 94.2 ACL functioning Lateral 80.8 Valgus stress 90.8 Lateral + valgus stress 72.5 ACL deficient Lateral 34.2 Valgus stress 17.5 Lateral + valgus stress 42.5 Normal lateral cartilage Valgus stress 91.7 Lateral cartilage defect Valgus stress 39.6 Partial-thickness defect Valgus stress 86.1 Full-thickness defect Valgus stress 11.7 were not accurately detected (40%). Although full-thick- ness defects were well outlined (86%), partial-thickness defects were poorly recognised (12%). Results Table 3 Accuracy of diagnosis of the intra-operative inspection The overall agreement of the suitability of a case for PKR was moderate (κ = 0.51). The best agreement was found in the medial bone-on-bone section, followed by full-thickness lateral cartilage on valgus stress radio- graphs, and acceptable change in patellofemoral disease on skyline view. The functioning ACL showed inadequate agreement (κ = 0.47) (Table 4). Details of each subject are presented in Table 3. On the whole, suitable cases were well diagnosed except for one extreme case, which was unsuitable for PKR (Fig. 1), despite a functioning ACL and lateral cartilage. On the other hand, unsuitable cases had a tendency to be diag- nosed as suitable, mainly due to thickness cartilage defects of the lateral condyle being only partial, or a defi- cient ACL (Figs. 2, 3).h A functioning ACL showed good rates of diagnosis based on valgus stress radiographs (90%) followed by lat- eral radiographs (80%). In contrast, ACL deficiency was not sufficiently diagnosed using lateral radiographs (34%) or valgus stress radiographs (18%). If the combination of both radiographs was used to make the decision, the accuracy increased to 43%, but this was still inadequate. The percentage of correctly predicted suitability var- ied between surgeons (45% to 75%). There were no significant correlations of the predictability with expe- rience as a surgeon, annual number of TKA and PKR performed, and percentage of PKR usage (0.130, −0.098 and −0.025, respectively). The predictably of ACL sta- tus and lateral cartilage status varied between 50–70% and 65–75%, respectively. Similarly, the predictability Regarding the lateral cartilage, intact lateral cartilage was well diagnosed (92%), while lateral cartilage defects Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Page 5 of 9 Fig. 1 An extreme case (case 20). Despite the severe bone loss in the medial compartment, the ACL and the lateral cartilage were intact. Some doctors judged this case as having no medial bone-on-bone appearance Fig. 1 An extreme case (case 20). Despite the severe bone loss in the medial compartment, the ACL and the lateral cartilage were intact. Some doctors judged this case as having no medial bone-on-bone appearance Discussionhi Table 4 Inter-rater agreements of decisions for each section Criterion Kappa 95% CI Overall agreement (%) Medial bone-on-bone 0.89 0.75–1.00 94.7 Functionally intact ACL 0.47 0.26–0.67 77.3 Full-thickness lateral cartilage 0.68 0.52–0.85 84.2 Functionally normal MCL 0.58 0.40–0.77 79.2 Acceptable patello-femoral joint 0.69 0.51–0.86 84.3 Overall PKR indication 0.51 0.31–0.70 75.3 of ACL and lateral cartilage status did not show signifi- cant correlation with experience or amount of surgery. The lateral cartilage status was significantly better pre- dicted than ACL status and overall status (versus ACL status P < 0.001, versus overall PKR suitability P = 0.01). Table 4 Inter-rater agreements of decisions for each section Criterion Kappa 95% CI Overall agreement (%) Medial bone-on-bone 0.89 0.75–1.00 94.7 Functionally intact ACL 0.47 0.26–0.67 77.3 Full-thickness lateral cartilage 0.68 0.52–0.85 84.2 Functionally normal MCL 0.58 0.40–0.77 79.2 Acceptable patello-femoral joint 0.69 0.51–0.86 84.3 Overall PKR indication 0.51 0.31–0.70 75.3 Table 4 Inter-rater agreements of decisions for each section Criterion Kappa 95% CI Overall agreement (%) Table 4 Inter-rater agreements of decisions for each section Criterion Kappa 95% CI Overall Table 4 Inter-rater agreements of decisions for each section This is the first report concerning agreement of the indi- cation of PKR between surgeons. Some reports sug- gest that only 5–8% of candidates for knee arthroplasty are eligible for PKR [21, 22]. On the other hand, Ham- ilton et al. [15] reported that around 50% of patients were found to be suitable for PKR using a decision aid. However, the decision was made by a single rater in their study, meaning there was insufficient evaluation of inter-rater variation. In our study, 12 experienced sur- geons were recruited to perform the evaluation. Despite the expertise of the raters, the overall PKR decision rate varied between 45% and 75%. Interestingly, the rate was higher than the actual rate of PKR suitability in this case series (40%). This variation might be caused by the accu- racy of diagnosis and inter-rater variations. of ACL and lateral cartilage status did not show signifi- cant correlation with experience or amount of surgery. The lateral cartilage status was significantly better pre- dicted than ACL status and overall status (versus ACL status P < 0.001, versus overall PKR suitability P = 0.01). of ACL and lateral cartilage status did not show signifi- cant correlation with experience or amount of surgery. Discussionhi The lateral cartilage status was significantly better pre- dicted than ACL status and overall status (versus ACL status P < 0.001, versus overall PKR suitability P = 0.01). Regarding the accuracy, all raters scored a higher per- centage of the knees suitable for PKR compared with the actual intra-articular condition (the status of the ACL Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Page 6 of 9 Fig. 2 A misleading case (case 6). Most surgeons diagnosed no erosion and correctable varus along with retained lateral cartilage, despite a deficient ACL and the presence of a partial-thickness defect of the lateral femoral condyle Fig. 2 A misleading case (case 6). Most surgeons diagnosed no erosion and correctable varus along with retained lateral cartilage, despite a deficient ACL and the presence of a partial-thickness defect of the lateral femoral condyle and lateral cartilage). Although functioning ACLs were accurately detected with excellent agreement, deficient ACLs were sometimes misdiagnosed as functioning on both lateral and valgus radiographs. This indicates that ACL deficiency would be correctly diagnosed. However, even if a radiograph shows an ACL as functioning, it is possible that the surgeon may find ACL deficiency during the operation. Mancuso et al. reported that there are two types of pathology regarding OA and ACL deficiency: primary ACL deficiency followed by secondary OA, and secondary ACL rupture caused by developed primary OA [23]. The latter can maintain the characteristics of OA with a functioning ACL despite the deterioration of the ACL. Such a deficiency is difficult to recognise using functional radiographs. valgus stress radiographs did not predict cartilage degen- eration, and the Osteoarthritis Research Society Interna- tional (OARSI) macroscopic grade did not correlate with the lateral cartilage thickness apparent on valgus stress radiographs [25]. It is understandable that contacting the surrounding full-thickness cartilage area hides partial- thickness defects under valgus stress conditions. In addi- tion, a lesion might be located in a minimal area, and consequently, the defect can be detected only when each (femur and tibial) lesion meets during the stress radio- graph. Consequently, the presence of a normal lateral joint space does not always confirm intact lateral carti- lage. MRI could be helpful to evaluate the lateral carti- lage as well as the condition of the meniscus. Discussionhi Although it can exaggerate the disease and incurs additional cost, it can improve the accuracy of the patient selection for PKR and could diminish the failure rate [17].fi Although full-thickness defects of the lateral cartilage were accurately diagnosed, partial-thickness defects of the lateral cartilage have been poorly detected on pre- operative radiographs [24]. Waldstein et al. reported that We found insufficient inter-rater agreement regard- ing the lateral radiographs, which is a reflection of the Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Page 7 of 9 Fig. 3 Another deceptive case (case 18). The erosion seemed to locate anteriorly. The lateral cartilage appeared normal and varus looked well corrected. The ACL was also deficient, and lateral cartilage was partially defective Fig. 3 Another deceptive case (case 18). The erosion seemed to locate anteriorly. The lateral cartilage appeared normal and varus looked well corrected. The ACL was also deficient, and lateral cartilage was partially defective between 45% and 75%. The lateral cartilage was more cor- rectly predicted than overall suitability and ACL status, but decisions were not perfect. Moreover, the predict- ability was not influenced by the years of experience or volume of surgery. Although all raters were experienced surgeons, the decision is considered not to be affected by the surgeon’s familiarity with the surgery.h ACL condition. Keyes et al. reported that, if there is no posterior tibial erosion or the erosion is not seen on the lateral radiographs, there is a 95% probability that the ACL would be functionally intact [16]. However, as the decision aid does not mention how to judge a case without obvious erosion, this can cause some misinter- pretation. In addition, it is not always easy to recognise erosion owing to image quality or overlapping condyles. The boundary of the erosion, and whether or not it is located in the posterior region, is not clear. More quan- titative criteria based on large numbers of clinical cases will be needed. Again, MRI could improve the accuracy of the ACL status evaluation, if it is available and the cost is tolerable for patients. Further study might be valuable in which the accuracy and reliability of the patient selec- tion for PKR is evaluated on the basis of not only radiog- raphies, but also MRI. ACL condition. Keyes et al. Discussionhi A surgeon makes a decision comprehensively using additional informa- tion from both radiological and clinical investigations. Regarding the surgeons who participated in this study, 6 out of 12 perform MRI routinely and another 5 obtain it only in selected patients. Thus, in total, 11 out of 12 sur- geons used MRI for decision-making at least in dubious cases. Most importantly, surgeons do not always under- take PKR, even if a patient is determined to be suitable for it. The reason for the individual surgeon’s decision in each case was not evaluated. Despite these limitations, this study provides valuable information for sophisticated decision-making in selection of suitable patients and improvement of PKR outcomes. judgement of the next case, and another study is needed to assess this. Thirdly, the number of patients was rela- tively small (20 patients), and no power analysis was per- formed because the statistics were descriptive and no comparison was conducted in this study. Nevertheless, the limitation of decision-making using radiography was recognised by this study. Finally, the decision was made on the basis of only five types of radiographs. Some other radiographs, such as Rosenberg’s view or long-stand- ing radiographs, are frequently used. A surgeon makes a decision comprehensively using additional informa- tion from both radiological and clinical investigations. Regarding the surgeons who participated in this study, 6 out of 12 perform MRI routinely and another 5 obtain it only in selected patients. Thus, in total, 11 out of 12 sur- geons used MRI for decision-making at least in dubious cases. Most importantly, surgeons do not always under- take PKR, even if a patient is determined to be suitable for it. The reason for the individual surgeon’s decision in each case was not evaluated. Despite these limitations, this study provides valuable information for sophisticated decision-making in selection of suitable patients and improvement of PKR outcomes. Consent for publication g 12. White SH, Ludkowski PF, Goodfellow JW (1991) Anteromedial osteoarthri- tis of the knee. J Bone Joint Surg Br 73-B:582–586 12. White SH, Ludkowski PF, Goodfellow JW (1991) Anteromedial osteoarthri- tis of the knee. J Bone Joint Surg Br 73-B:582–586 Not applicable. 13. Berend KR, Berend ME, Dalury DF et al (2015) Consensus statement on indications and contraindications for medial unicompartmental knee arthroplasty. J Surg Orthop Adv 24:252–256. https://doi.org/10.3113/jsoa. 2015.0252 References 1. Price AJ, Webb J, Topf H et al (2001) Rapid recovery after Oxford uni- compartmental arthroplasty through a short incision. J Arthroplasty 16:970–976. https://doi.org/10.1054/arth.2001.25552 2. Liddle AD, Judge A, Pandit H, Murray DW (2014) Adverse outcomes after total and unicompartmental knee replacement in 101,330 matched patients: a study of data from the National Joint Registry for England and Wales. Lancet 384:1437–1445. https://doi.org/10.1016/S0140-6736(14) 60419-0 Availability of data and materials No data or materials are provided. Availability of data and materials 9. Orthopaedic A, Association Australian Orthopaedic Association National Joint Replacement Registry. https://aoanjrr.sahmri.com/documents/ 10180/668596/Hip%2C+Knee+%26+Shoulder+Arthroplasty/c287d2a3- 22df-a3bb-37a2-91e6c00bfcf0 9. Orthopaedic A, Association Australian Orthopaedic Association National Joint Replacement Registry. https://aoanjrr.sahmri.com/documents/ 10180/668596/Hip%2C+Knee+%26+Shoulder+Arthroplasty/c287d2a3- 22df-a3bb-37a2-91e6c00bfcf0 No data or materials are provided. Acknowledgements 3. Liddle ADD, Pandit H, Judge A, Murray DWDW (2015) Patient-reported outcomes after total and unicompartmental knee arthroplasty Joint Registry for England and Wales. Bone Joint J 97:793–801. https://doi.org/ 10.1302/0301-620X.97B6 We would like to thank Drs. Kenjiro Yoshida (Yoshida Orthopaedic Clinic, Osaka), Shigenobu Fukushima (Saiseikai Fukushima General Hospital, Yama- gata), Shingo Nitta (Anshin Hospital, Hyogo), Ichiro Tatsumi (Shonan Kamakura General Hospital, Kanagawa), Hidetoshi Hamaguchi (Sapporo Hospital, Hokkaido), Satoshi Takei (Hankai Hospital, Osaka) and Keinosuke Ryu (Nihon University Hospital, Tokyo) for data collection and analysis. We thank Benjamin Phillis at the Clinical Study Support Center, Wakayama Medical University for proofreading and editing. We also thank Professor Toshio Shimokawa at the Clinical Study Support Center, Wakayama Medical University for data analysis and statistics. 4. Goodfellow JW, Kershaw CJ, Benson MK, O’Connor JJ (1988) The Oxford Knee for unicompartmental osteoarthritis. The first 103 cases. J Bone Joint Surg Br 70:692–701. https://doi.org/10.1302/0301-620X.70B5.31925 63 4. Goodfellow JW, Kershaw CJ, Benson MK, O’Connor JJ (1988) The Oxford Knee for unicompartmental osteoarthritis. The first 103 cases. J Bone Joint Surg Br 70:692–701. https://doi.org/10.1302/0301-620X.70B5.31925 63 5. Kozinn SC, Scott R (1989) Unicondylar knee arthroplasty. J Bone Joint Surg Am 71:145–150 5. Kozinn SC, Scott R (1989) Unicondylar knee arthroplasty. J Bone Joint Surg Am 71:145–150 6. The Swedish National Board of Health And Welfare Swedish knee arthro- plasty register. In: 2018. http://myknee.se/pdf/SVK_2018_Eng_1.0.pdf. 6. The Swedish National Board of Health And Welfare Swedish knee arthro- plasty register. In: 2018. http://myknee.se/pdf/SVK_2018_Eng_1.0.pdf. Author details 1 Department of Orthopaedic Surgery and Joint Surgery Centre, Takatsuki General Hospital, 1‑3‑13 Kosobe, Takatski City, Osaka 569‑1192, Japan. 2 Depart- ment of Orthopaedic Surgery, Japanese Red Cross Hamamatsu Hospital, 1088‑1 Kobayashi, Hamakita‑ku, Hamamatsu City, Shizuoka 434‑8533, Japan. 3 Department of Joint Reconstruction Center, Souseikai Fukuoka Mirai Hos- pital, 3‑5‑1, Kashiiteriha, Higashi‑ku, Fukuoka City, Fukuoka 813‑0017, Japan. 4 Department of Orthopaedic Surgery, Japan Community Healthcare Organiza- tion Saitama Medical Center, 4‑9‑3 Kitaurawa Urawa‑ku, Saitama 330‑0074, Japan. 5 Department of Orthopaedic Surgery, Japan Community Healthcare Organization Yugawara Hospital, 2‑21‑6 Chuo, Yugawara, Kanagawa 259‑0396, Japan. 6 Department of Orthopaedic Surgery, Faculty of Medicine, The Univer- sity of Tokyo, 7‑3‑1, Hongo, Bunkyo‑ku, Tokyo 113‑0033, Japan. 7 Department of Orthopedic Surgery, Yamagata Tokushukai Hospital, 2‑3‑51 Kiyozumimachi, Yamagata city, Yamagata 990‑0834, Japan. 8 Department of Orthopaedic Sur- gery and Joint Surgery Centre, Takatsuki General Hospital, 1‑3‑13, Kosobe‑Cho, Takatsuki City, Osaka 561‑1115, Japan. Received: 16 November 2021 Accepted: 10 February 2022 Received: 16 November 2021 Accepted: 10 February 2022 Authors’ contributions 7. Riddle DL, Jiranek WA, McGlynn FJ (2008) Yearly incidence of unicompart- mental knee arthroplasty in the United States. J Arthroplasty 23:408–412. https://doi.org/10.1016/j.arth.2007.04.012 7. Riddle DL, Jiranek WA, McGlynn FJ (2008) Yearly incidence of unicompart- mental knee arthroplasty in the United States. J Arthroplasty 23:408–412. https://doi.org/10.1016/j.arth.2007.04.012 TH contributed patient data, performed data analysis and wrote the manu- script. RF, KT, TK, KM, HI, ET and TT performed the data analysis and revised the manuscript. All authors read and approved the final manuscript. 8. National Joint Registry NJR report. https://reports.njrcentre.org.uk/knees- primary-procedures-components/K15v1NJR?reportid=5CF7C49E-FAEE- 466C-89F0-4BEEFCFB2264&defaults=DC__Reporting_Period__Date_ Range=%22MAX%22,R__Filter__Implant_Type_Name=%22Knee-Bicon dylar%22 Declarations 10. Liddle AD, Pandit H, Judge A, Murray DW (2015) Optimal usage of uni- compartmental knee arthroplasty. Bone Joint J 97-B:1506–1511. https:// doi.org/10.1302/0301-620X.97B11.35551 10. Liddle AD, Pandit H, Judge A, Murray DW (2015) Optimal usage of uni- compartmental knee arthroplasty. Bone Joint J 97-B:1506–1511. https:// doi.org/10.1302/0301-620X.97B11.35551 Discussionhi reported that, if there is no posterior tibial erosion or the erosion is not seen on the lateral radiographs, there is a 95% probability that the ACL would be functionally intact [16]. However, as the decision aid does not mention how to judge a case without obvious erosion, this can cause some misinter- pretation. In addition, it is not always easy to recognise erosion owing to image quality or overlapping condyles. The boundary of the erosion, and whether or not it is located in the posterior region, is not clear. More quan- titative criteria based on large numbers of clinical cases will be needed. Again, MRI could improve the accuracy of the ACL status evaluation, if it is available and the cost is tolerable for patients. Further study might be valuable in which the accuracy and reliability of the patient selec- tion for PKR is evaluated on the basis of not only radiog- raphies, but also MRI. There are some limitations to this study. Firstly, the results may change according to the selection of patients. Each patient has individual characteristics. This case series included an extreme case (Fig. 1) with a function- ing ACL and lateral cartilage with severe bone loss and grooving. In addition, the percentage of the series suit- able for PKR was lower than typical; in general, 50–60% of patients who are candidates for knee arthroplasty undergo PKR in our institution. However, we recruited patients prospectively to exclude selection bias. Secondly, there was no evaluation of inter-rater reproducibility. It is possible that a previously assessed case may affect the Interestingly, the predictability of ACL and lateral cartilage status and overall PKR suitability was varied, Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Page 8 of 9 judgement of the next case, and another study is needed to assess this. Thirdly, the number of patients was rela- tively small (20 patients), and no power analysis was per- formed because the statistics were descriptive and no comparison was conducted in this study. Nevertheless, the limitation of decision-making using radiography was recognised by this study. Finally, the decision was made on the basis of only five types of radiographs. Some other radiographs, such as Rosenberg’s view or long-stand- ing radiographs, are frequently used. Ethics approval and consent to participate This study was approved by the institutional review board in Takatsuki General Hospital (2019-17), and written informed consent was obtained in each patients. 11. Badawy M, Espehaug B, Indrekvam K et al (2014) Higher revision risk for unicompartmental knee arthroplasty in low-volume hospitals. Acta Orthop 85:342–347. https://doi.org/10.3109/17453674.2014.920990 11. Badawy M, Espehaug B, Indrekvam K et al (2014) Higher revision risk for unicompartmental knee arthroplasty in low-volume hospitals. Acta Orthop 85:342–347. https://doi.org/10.3109/17453674.2014.920990 Funding There is no funding regarding this study. Availability of data and materials No data or materials are provided. dam/zb-minisites/oxford-partial-knee-hcp/documents/oxford-decision- aid-flyer.pdf Waldstein W, Bou Monsef J, Buckup J, Boettner F (2013) The value of valgus stress radiographs in the workup for medial unicompartmental arthritis. Clin Orthop Relat Res 471:3998–4003. https://doi.org/10.1007/ s11999-013-3212-3 25. Waldstein W, Bou Monsef J, Buckup J, Boettner F (2013) The value of valgus stress radiographs in the workup for medial unicompartmental arthritis. Clin Orthop Relat Res 471:3998–4003. https://doi.org/10.1007/ s11999-013-3212-3 Competing interests One of the authors received honorarium for presentations and instructions from Zimmer Biomet. 14. ZimmerBiomet Radiographic Assessment for Medial Oxford ® Partial Knee Replacement (PKR). https://www.oxfordpartialknee.net/content/ Page 9 of 9 Hiranaka et al. Knee Surgery & Related Research (2022) 34:13 Hiranaka et al. Knee Surgery & Related Research dam/zb-minisites/oxford-partial-knee-hcp/documents/oxford-decision- aid-flyer.pdf dam/zb-minisites/oxford-partial-knee-hcp/documents/oxford-decision- aid-flyer.pdf 15. Hamilton TW, Pandit HG, Lombardi AV et al (2016) Radiological decision aid to determine suitability for medial unicompartmental knee arthro- plasty: development and preliminary validation. Bone Joint J 98-B:3–10. https://doi.org/10.1302/0301-620X.98B10.BJJ-2016-0432.R1 16. Keyes GW, Carr AJ, Miller RK, Goodfellow JW (1992) The radiographic clas- sification of medial gonarthrosis. Acta Orthop Scand 63:497–501. https:// doi.org/10.3109/17453679209154722 17. Hamilton TW, Choudhary R, Jenkins C et al (2017) Lateral osteophytes do not represent a contraindication to medial unicompartmental knee arthroplasty: a 15-year follow-up. Knee Surg Sports Traumatol Arthrosc 25:652–659. https://doi.org/10.1007/s00167-016-4313-9 g 18. Beard DJ, Pandit H, Ostlere S et al (2007) Pre-operative clinical and radiological assessment of the patellofemoral joint in unicompartmental knee replacement and its influence on outcome. J Bone Joint Surg Br 89-B:1602–1607. https://doi.org/10.1302/0301-620X.89B12.19260 19. Hiranaka T, Hida Y, Tanaka T et al (2020) Validation of the macroscopic anterior cruciate ligament status using the Oxford classification system in relation to cartilage defects on the medial tibial plateau in osteoarthritic knees. J Knee Surg. https://doi.org/10.1055/s-0040-1721032 g p g 20. Kanda Y (2013) Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 48:452–458. https:// doi.org/10.1038/bmt.2012.244 g 21. Ritter MA, Faris PM, Thong AE et al (2004) Intra-operative findings in varus osteoarthritis of the knee. An analysis of pre-operative alignment in potential candidates for unicompartmental arthroplasty. J Bone Joint Surg Br 86:43–47 21. Ritter MA, Faris PM, Thong AE et al (2004) Intra-operative findings in varus osteoarthritis of the knee. An analysis of pre-operative alignment in potential candidates for unicompartmental arthroplasty. J Bone Joint Surg Br 86:43–47 22. Stern SH, Becker MW, Insall JN (1993) Unicondylar knee arthroplasty: an evaluation of selection criteria. Clin Orthop Relat Res 286:143–148. https://doi.org/10.1097/00003086-199301000-00021 22. Stern SH, Becker MW, Insall JN (1993) Unicondylar knee arthroplasty: an evaluation of selection criteria. Clin Orthop Relat Res 286:143–148. https://doi.org/10.1097/00003086-199301000-00021 p g 23. Mancuso F, Dodd CA, Murray DW, Pandit H (2016) Medial unicompart- mental knee arthroplasty in the ACL-deficient knee. J Orthop Traumatol 17:267–275. https://doi.org/10.1007/s10195-016-0402-2 23. Mancuso F, Dodd CA, Murray DW, Pandit H (2016) Medial unicompart- mental knee arthroplasty in the ACL-deficient knee. J Orthop Traumatol 17:267–275. https://doi.org/10.1007/s10195-016-0402-2 g 24. Tashiro Y, Matsuda S, Okazaki K et al (2014) The coronal alignment after medial unicompartmental knee arthroplasty can be predicted: usefulness of full-length valgus stress radiography for evaluating correctability. Knee Surg Sports Traumatol Arthrosc 22:3142–3149. https://doi.org/10.1007/ s00167-014-3248-2 25. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from:
https://openalex.org/W4255716886	https://bmccomplementmedtherapies.biomedcentral.com/track/pdf/10.1186/1472-6882-12-S1-P240	English	null	P02.184. Abstract withdrawn	BMC complementary and alternative medicine	2,012	cc-by	258	POSTER PRESENTATION Open Access BMC Complementary and Alternative Medicine 2012, 12(Suppl 1):P240 http://www.biomedcentral.com/1472-6882/12/S1/P240 BMC Complementary and Alternative Medicine 2012, 12(Suppl 1):P240 http://www.biomedcentral.com/1472-6882/12/S1/P240 © 2012 licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract withdrawn Published: 12 June 2012 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit © 2012 licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit © 2012 licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
https://openalex.org/W2901037659	https://www.journals.vu.lt/literatura/article/download/12185/10764	Russian	null	Anos Achmatovos kūrybos recepcija lietuvių egzodo literatūroje (Henrikas Radauskas, Alfonsas Nyka-Niliūnas, Tomas Venclova)	Literatura/Literatūra	2,018	cc-by	14,572	ISSN 0258-0802. ISSN 1648-1143 (Online). LITERATŪRA 2018 60(2) DOI: https://doi.org/10.15388/Litera.2018.2.12037 ISSN 0258-0802. ISSN 1648-1143 (Online). LITERATŪRA 2018 60(2) DOI: https://doi.org/10.15388/Litera.2018.2.12037 1 Balys Sruoga išvertė tris Achmatovos eilėraščius „planuotam savo poezijos vakarui Maskvoje kaip Ser gejaus Prokofjevo dainų tekstus“: „Pilkaakis karalius“, „Atminta Saulės“, „Saulė kambarį prikūpino“ (Sruoga 1998, 784). 2 Achmatova, A. 1964. Poezija [redakcinė komi sija: E. Mieželaitis (pirm.); vertė S. Nėris, J. Vačiūnai tė, T. Venclova]. Vilnius: Vaga; Achmatova, A. 1994. Erškėtis žydi = Shipovnik cvetet: eilėraščiai ir poema. Vert.: Janina Degutytė, Jonas Jakštas, Jolanta Jonynaitė, Jonas Strielkūnas, Judita Vaičiūnaitė, Tomas Venclova. Vilnius : Vyturys. ANOS ACHMATOVOS KŪRYBOS RECEPCIJA LIETUVIŲ EGZODO LITERATŪROJE (HENRIKAS RADAUSKAS, ALFONSAS NYKA-NILIŪNAS, TOMAS VENCLOVA) Aliona Sofija Ivinskaja Vilniaus universitetas Aliona Sofija Ivinskaja Aliona Sofija Ivinskaja Straipsnyje nagrinėjami trijų lietuvių egzodo poetų – Henriko Radausko, Alfonso Nykos-Niliūno ir Tomo Venclovos atsiminimai, dienoraščiai, vertimai, dedikacijos, poezija. Šiuos lietuvių egzodo poezijos atstovus vargu ar galima būtų priskirti kokiai nors vienai literatūrinei srovei. Todėl Anos Achmatovos kūrybos recepcija bus apžvelgiama atskirai kiekvieno iš pasirinktų autorių kūrybinio palikimo kontekste. Remiantis recepcijos koncepciją, suvokimo procesas nėra pastovus faktas, tai besikeičiantis, dinamiškas reiškinys, kuriame svarbiausia yra priimanti literatūra, todėl nagrinėjant Achmatovos kūrybos recepciją, svarbu atskleisti lietuvių poetų kūrybos ypatumus, jų meno suvoki mą, poezijos vertinimus. Raktiniai žodžiai: Achmatova, recepcija, Venclova, Radauskas, Nyka-Niliūnas, egzodas, lietuvių literatūra. Keywords: Akhmatova, reception, Venclova, Radauskas, Nyka-Niliūnas, exodus, Lithuanian litera ture. tūrologai savo tyrinėjimuose pastebėdavo Salomėjos Nėries, Juditos Vaičiūnaitės, Tomo Venclovos ir Achmatovos kūrybi nės paraleles. Tai Danutės Sabromienės straipsnis „Ana Achmatova Juditos Vai čiūnaitės vertimuose“ (Сабромене 2004), Gintarės Bernotienės monografija Apie žo džių sandūros tikslumą (Bernotienė 2016), Vinco Kuzmicko straipsnis „S. Nėris – A. Achmatovos lyrikos vertėja“ (Kuzmic kas 1974), Tomo Venclovos „Atsiminimai apie Aną Achmatovą“ (Venclova 2013b), Nuo to laiko, kai 1916 m. B. Sruoga išver tė tris Anos Achmatovos eilėraščius į lietu vių kalbą1, praėjo daugiau negu 100 metų. Per tą laiką rusų poetės eilėraščius vertė Salomėja Nėris, Janina Degutytė, Jonas Jakštas, Jolanta Jonynaitė, Jonas Striel kūnas, Judita Vaičiūnaitė, Tomas Venclo va, Sigitas Geda, Gintautas Iešmantas, Jonas Juškaitis, Ramutė Skučaitė, Aldona Sapežinskaitė ir kt., išleistos jos poezijos rinktinės lietuvių kalba2. Lietuvos litera 136 Viktorijos Daujotytės monografijos Para šyta moterų (Daujotytė 2001) bei Janina Degutytė: gyvenimo ir kūrybos apybraiža (Daujotytė 1984), kur mokslininkė mini ir apie A. Achmatovos įtaką S. Neries bei J. Degutytės kūrybai. Viktorijos Daujotytės monografijos Para šyta moterų (Daujotytė 2001) bei Janina Degutytė: gyvenimo ir kūrybos apybraiža (Daujotytė 1984), kur mokslininkė mini ir apie A. Achmatovos įtaką S. Neries bei J. Degutytės kūrybai. Viktorijos Daujotytės monografijos Para šyta moterų (Daujotytė 2001) bei Janina Degutytė: gyvenimo ir kūrybos apybraiža (Daujotytė 1984), kur mokslininkė mini ir apie A. Achmatovos įtaką S. Neries bei J. Degutytės kūrybai. ros procesas rašė, kad recepcijos centre yra priimanti literatūra: Nacionalinė literatūra – pirmoji žodinės kūrybos bendrumo pakopa, individuali, spontaniškai iškylanti iš žodžio materijos, psichinės sanklodos ir istorinės situacijos. <…> Ji turi savyje “ selekcijos aparatą”, atsirenka iš aplinkinių literatūrų, kas ati tinka jos prigimtį ir reikalinga jai augti. Ne išgarsėjusio kūrinio objektyvi vertė, o receptuojančios literatūros reikmės bei raidos sąlygos lemia, ar tas reiškinys bus priimtas, ar atmestas <…>. Aliona Sofija Ivinskaja Tad recepci jos, vertimai ir poveikiai priklauso ne duo dančios, o priimančios literatūros plotams. <…>. Sudėtingame literatūrinių ryšių že mėlapyje ji stovi centre kaip veikiantis su bjektas“. Achmatovos „atėjimui“ į lietuvių kul tūrinę terpę gali būti taikomas recepcinis požiūris. Moderniosios literatūros teore tikų V. Zinčenko, V. Zusman, Z. Kirnoze darbuose recepcija vadinamas „vienos literatūros įsiskverbimas į kitos literatū ros pasaulį“, kuris įvyko dėl vidinių ir iš orinių priežasčių ir paruošė „priimančiąją aplinką“ „išorinio impulso priėmimui“ (Зинченко; Зусман; Кирнозе 2002, 83). Lietuvių literatūrologė Aušra Jurgutienė recepciją įvardina kaip komparatyvistikos metodologinę inovaciją: „<...> šiuolaiki nėje komparatyvistikoje iš esmės pasikeitė recepcijos samprata, įgydama griežtesnę metodologinę reikšmę“ (Jurgutienė 2006, 64). Recepcijos kaip suvokimo estetikos samprata atsirado Konstancos mokyklos veikloje. Jurgutienė siūlo lietuviškai ją vadinti „skaitymo estetika“. Svarbu, kad „recepcijos arba suvokimo estetika litera tūros tyrinėtojus paskatino ne mechaniškai derinti formalistinius estetinius ir kultūri nius istorinius tyrimo aspektus, bet išpro vokavo esminį literatūros istorinių tyrimų lūžį: nuo grožinių kūrinių istorijos pereiti prie jų skaitymo arba estetinių recepcijų lyginamojo istorinio tyrimo“ (Jurgutienė 2001, 66). Remiantis šiuolaikinėmis re cepcijos koncepcijomis, suvokimo proce sas nėra pastovus faktas, tai besikeičiantis, dinamiškas reiškinys, kuriam ištirti reika lingas kompleksinis požiūris pasitelkiant istorinį sinchroninį ir kultūrinį diachroninį metodus. V. Kubilius savo monografijoje Lietuvių literatūra ir pasaulinės literatū 4 J. Blekaitis straipsnyje „Balys Sruoga“ rašo: „Savo profesoriaus Mūnchene A. Kutcherio pavyzdžiu Sruoga įveda ir skaito atskirą teatro istorijos bei teorijos kursą, vėliau įsteigia ir Teatro seminarą – šeštadieninių kultūrinių diskusijų, dažnai prasiliejančių už teatro ribų, forumą. Iš vyresniųjų čia daug popiečių praleido Pulgis Andriušis, Henrikas Radauskas, Bronys Raila, Kazys Jankauskas, Antanas Gustaitis, Antanas Rūkas, o pasku tiniais metais buvo palankę seminarą ir tik pradėję tada iškilti tokie jauni poetai, kaip Alfonsas Nyka-Niliūnas ir Kazys Bradūnas“ (Blekaitis 1978, 407). 3 Septyni S. Nėries Achmatovos eilėraščių vertimai buvo aptikti ir publikuojami nuo 1946 m. (Nėris 1984, 295). 1916–1917 m. Balio Sruogos atlikti trijų Achma tovos eilėraščių vertimai buvo paskelbti jo Raštuose tik 1998 m. (Sruoga 1998, 783-784). 5 Leono Karsavino (Levas Karsavinas, 1882– 1952) sesuo, garsi balerina Tamara Karsavina (1885– 1978) gyvendama Sankt Peterburge reguliariai lankėsi legendinėje kavinėje „Brodiačiaja sobaka“ («Бродячая собака»). 1914 metais ten įvyko unikalus renginys, skirtas Tamarai Karsavinai. Tame vakare buvo pristaty tas poezijos rinkinys Buket dlia Tamary Karsavinoj. Į rinkinį įėjo Sidabrinio amžiaus poetų eilėraščiai, skir ti Karsavinai. Žr.: Букет для Тамары Карсавиной: [сборник стихов / сост. и макет: В. Киселёв]. Москва: Композитор, 1998. 81, [2] c.: ил., портр. Rinkinyje be kitų autorių, yra A. Achmatovos ir N. Gumiliovo eilė raščiai: „Tamare Platonovne Karsavinoj“ («Тамаре Платоновне Карсавиной»), „Dolgo molili o tance my vas, no molili naprasno“ («Долго молили о танце мы вас, но молили напрасно»). (Kubilius 1983, 22) Lietuvių egzodo kritikas Rimvydas Šil bajoris straipsnyje apie Henriką Radauską cituoja tokius Boriso Pasternako žodžius: „Poeto biografija – tai, kas atsitinka jo skaitytojams“, norėdamas pabrėžti, kad poeto kūryba negali egzistuoti be priiman čios pusės, tų, kas tą poeziją skaitydami suvokia, ir tas suvokimas nėra pastovus, nes kaskart tas įsiskverbimas ir įsisavini mas bus kitoks (Šilbajoris 1994, 462). Taigi, nagrinėjant Achmatovos kūry bos recepciją, svarbu atskleisti lietuvių poetų kūrybos ypatumus, jų estetinį meno suvokimą, poezijos vertinimus, nes, kaip rašė V. Kubilius, „poetas negimsta nei ro mantiku, nei realistu. <...> Bet mūsų am žiuje poetinę sąmonę formuoja jau ne tiek visuotiniai literatūriniai judėjimai, kurie kaskart darosi vis menkesni, kiek pačių kūrėjų nusistatymai ir polinkiai“ (Kubilius 1989, 10). Lietuvos egzodo literatūra iš panašaus likimo literatūrų išsiskiria tuo, kad lietu vių egzodas užsitęsė ilgus dešimtmečius, 137 apima net penkias rašančiųjų kartas. Ši literatūra atliko svarbų vaidmenį lietuvių literatūros raidoje, reiškė laisvą ir nepri klausomą kūrėjo žodį, prisidėjo prie tar pliteratūrinio dialogo Vakarų pasaulyje (Šilbajoris 1993, 1-2). Šio darbo tikslas – pabandyti išaiškinti, kokia buvo Achmato vos kūrybos recepcija lietuvių egzodo lite ratūroje, kodėl lietuvių poetai domėjosi jos kūryba, skaitė bei vertė jos poeziją, kaip ją vertino. raskalionnym gornom“ ir itališkų eilėraš čių ciklas“ (Churginas, Geda 1987, 23). Kas gi paskatino Radauską domėtis rusų literatūra bei Achmatovos kūryba? Manau, kad tai galėjo būti Balys Sruoga, kai Henrikas Radauskas lankė jo Teatro seminarą4 (Churginas, Geda 1987, 23). Tai galėjo būti ir jo žmona Vera Sotniko vaitė-Radauskienė (1903–1973), gimusi Šauliuose. Jaunystėje ji mokėsi garsioje Imperatoriškojoje Rusijos baleto moky kloje, bet, susižeidusi koją, į šokėjas ne beiškilo, tapo baleto kritike. Kaune ji su sipažino su rusų emigrantais, bendravo su profesoriumi Leonu Karsavinu5, filosofu Vasilijumi Sezemanu. Ji tikriausiai įvedė ir Radauską „ton kultūriškai stimuliuojan čion aplinkon, kurioje imlus ir talentingas poetas galėjo nemažai pasisemti“ (Blekai tis 1977a, 45-46). Pati Radauskienė, pasak Jurgio Blekaičio, buvo labai aktyvi, godi Darbe bus nagrinėjami trijų lietuvių eg zodo poezijos atstovų – Henriko Radaus ko, Alfonso Nykos-Niliūno, Tomo Ven clovos atsiminimai, dienoraščiai, vertimai, dedikacijos, poezija. 6 Ivaras Ivaskas (1927–1992) – estų išeivijos lite ratūrologas, poetas, dailininkas. Nuo 1951 iki 1970 m. Henrikas Radauskas artimai bendravo su Ivaru Ivasku bei latvių poete Astride Ivaska. Henrikas Radauskas (1910–1970) Su Achmatovos kūryba Henrikas Radaus kas susipažino dar būdamas Lietuvoje. Achmatovos eilėraščius skaitė originalo kalba, nes tarpukariu lietuviškai buvo iš spausdintas tik vienas Achmatovos eilė raštis „Choroni choroni menia veter“, kurį 1928 m. Salomėja Nėris išvertė ir pava dino „Palydėk“3 (Nėris 1984, 439). Rusų kalbą Radauskas išmoko dar vaikystėje, kai 1914 m., prasidėjus I pasauliniam ka rui, jis su šeima pasitraukė į Sibirą, kur lankė rusišką pradinę mokyklą (Blekaitis 1977b, 73-74). Henrikas Radauskas buvo gerai susipažinęs su rusų literatūra. Pasak Aleksandro Churgino, „lenkų ir rusų poe zijoje jis jautėsi kaip namie <...>. Iš rusų poetų labiausiai vertino Mandelštamą, Achmatovą, Gumiliovą. Iš Gumiliovo jam labiausiai patiko eilėraštis „On stoit pred 138 knygų skaitytoja, meno, ypač teatro en tuziastė, ji buvo „Henrikui vieno asmens “kultūrinė aplinka” (Blekaitis 1977a, 46). Skaityti Radauską skatino motina, Ama lija Radauskienė-Kieragga, jų namuose buvo daug knygų ir iki Spalio perversmo išleistų žurnalų. Panevėžio gimnazijos bibliotekoje, kur mokėsi Radauskas, buvo sutelkta daug tokių autorių kaip Ibsenas, Hugo, Dickensas, Goethe, Hauptmannas, Šekspyras ir t.t. Čia jis galėjo susipažinti ir su rusų poezijos „sidabro amžiaus“ po etais – Bloku, Briusovu, Balmontu ir Bal trušaičiu (Blekaitis 1977b, 79). kiškai, rusiškai, lenkiškai. „Na Jorge, kaip tas Pasternako?“ – ragina ir mane, nes vie ną eilėraštį, iš kažkur išsirašęs, moku, ku rio nėra jo rinktinėje knygoje. Bet dažniau siai ir dešimtis kartų skaitytų bei iš Henriko girdėtų aš ištisai neatsimenu, tik fragmen tais, o jis, rodos, galėtų kalbėti eilėmis visą vakarą. Pereina į Rilkę, Achmatovą, Man delštamą, Tuwimą, Heinę... kiškai, rusiškai, lenkiškai. „Na Jorge, kaip tas Pasternako?“ – ragina ir mane, nes vie ną eilėraštį, iš kažkur išsirašęs, moku, ku rio nėra jo rinktinėje knygoje. Bet dažniau siai ir dešimtis kartų skaitytų bei iš Henriko girdėtų aš ištisai neatsimenu, tik fragmen tais, o jis, rodos, galėtų kalbėti eilėmis visą vakarą. Pereina į Rilkę, Achmatovą, Man delštamą, Tuwimą, Heinę... (Blekaitis 1977a, 37) (Blekaitis 1977a, 37) Įvairiais metais Radausko laiškuose, amžininkų atsiminimuose, draugų dieno raščiuose randame poeto pasisakymus apie Achmatovą bei jos kūrybos vertinimus. Radauskas pirkdavo Achmatovos knygas, gaudavo jas iš draugų, pažįstamų, leidy klų. 1952 m. Čikagoje Radauskas įsigijo Achmatovos knygą, ir norėjo nusiusti savo bičiuliui Ivarui Ivaskui6: „Achmatovos knygą čia įsigijau prieš penkias ar šešias savaites (deja, su labai prasta įžanga). Šeš tadienį Jums tą knygą nusiųsiu“ (Radaus kas 2009, 40). Veikiausiai čia kalbama apie Achmatovos knygą Izbrannye stichotvo renija 1952 m. išleista Niujorke Čechovo vardo leidykloje (Ахматова 1952). 8 1952 m. Čechovo vardo leidykla be autorių ži nios išleido Achmatovos ir Zoščenkos rinktines. Plačiau apie Čechovo leidyklos veiklą žr.: Базанов, П. Н. 2014. Издательство имени Чехова. Новый журнал. 276. 7 Achmatovos eilėraščio originale pavadinimas ra šomas: «Подкапризовая Дорога» (Ахматова 1990: I, 183). Podkaprizovaja – nuo žodžio kaprizas. Carskoje selo parke yra du statiniai – Didelis ir Mažas kaprizai – dvi arkos, išlenktos virš kelio, kurios buvo pastatytos Jekaterinos II įsakymu. Henrikas Radauskas (1910–1970) Toliau Radauskas kritikuoja šį rinkinį, kuriame labai daug korektūros klaidų, kartu pa brėždamas, kad net tai negali sumenkinti Achmatovos poezijos: ( ) Radauskas turėjo mėgstamų, cituojamų poetų sąrašą. Iš senesniųjų prancūzų – tai „Verlaine’as, Mallarme ir Valery. <...> Iš vokiečių – ir gal iš visų kraštų aukščiau siai – Raineris Maria Rilke. Kiti – Hof mannstahlis, George. Iš lenkų – pirmiausia Tuwimas, po jo Lechonis. Iš rusų – Paster nakas, Mandelštamas, Achmatova“ (Ble kaitis 1977a, 28). Toks, rašo J. Blekaitis, „Radausko Parnaso gyventojų – neabejo tinai ne visas – sąrašas. To ekskliuzyvaus klubo narius verta sužymėti, nes jie nėra kaprizingų nuotaikų ar laikinų susižavė jimų žybtelėjimai, bet nuolatinė atranka pagal minėtą griežtą kriterijų. Tai, ką Ra dauskas vertino, to jis ir pats savo poezijo je siekė“ (Blekaitis 1977a, 28). Achmatova buvo Radausko mėgstamų poetų sąraše, ir, kadangi jo poetiniai pomėgiai buvo pasto vūs, jis nuolat domėjosi jos kūryba. Ra dauskas Amerikoje, dirbdamas varginantį fizinį darbą, laisvalaikiu aptardavo su žmo na ir draugais Literatūros lankuose ir kitur paskelbtus straipsnius, naujas knygas. Vie ną tokį vakarą prisimena J. Blekaitis: <...> Rinkinys neblogas jau vien dėl to, kad ši poetė labai nedaug teparašė vidutiniškų eilėraščių. Tik gaila, kad knygoje apstu primityvių, tiesiog kvailų korektūros klai dų, dėl kurių tiesiog neįmanoma iššifruoti prasmės (подкапризовая [?!] дорога). (Radauskas 2009, 40) Čia kalbama apie eilėraštį „Odni glia diatsia v laskovye vzory“, kurį Achma Radauskienė turi daug teatrinių prisimini mų <...> O Henrikas ima atmintinai sakyti savo mėgstamiausių poetų eilėraščius – vo 139 tova parašė 1936 m. Pirmą kartą jis buvo išspausdintas 1940 m. žurnale Leningrad (Ахматова 1990: I, 400). Minima eilėraš tyje „Podkaprizovaja Doroga“7 Carsko je selo parke buvo mėgstamiausia Ach matovos ir jo draugo, poeto ir literatūros kritiko, Nikolajaus Nedobrovo susitikimų vieta. Jam Achmatova ir skiria šitą eilėraš tį. Radauskas pastebėjo, kad geografinis pavadinimas „Podkaprizovaja Doroga“ leidinyje buvo išspausdintas iš mažosios raidės. Čechovo vardo leidykloje, veiku sioje Niujorke nuo 1951 iki 1956 m., buvo išleistos 178 knygos tų autorių, kurių kūri niai dėl cenzūros negalėjo būti išleisti Ru sijoje. Medžiagą knygoms, dažniausiai be autorių sutikimo ir žinios, leidykla imdavo iš savilaidinių leidinių bei senos periodi kos, kur įsiveldavo daug klaidų8. Pasipik tinimas klaidomis Achmatovos knygoje at siranda iš būdingos Radauskui nuostatos, kad poezija, tapyba, muzika bei viskas, kas siejama su menu, turi būti tobula: kritiškai vertina jų turinį. 1966 m. 9 Turima omenyje: Ахматова, Анна. 1965–1983. Собрание сочинений. Т. 1–3. Ред. Г. и Н. Струве и Б. Филиппов. Вашингтон; Париж. Henrikas Radauskas (1910–1970) sausio 1 dieną Radauskas rašo: „Vakar gavau pir mąjį Achmatovos tomą9 (vėl išleido Fil lipow-Struve) su senais gražiais jos por tretais, beveik pamirštais per 25 metus“ (Radauskas 2009, 220). Iš amžininkų at siminimų yra žinoma, kad Achmatova ne buvo patenkinta Glebo Struvės leidiniais. T. Venclova atsimena: Achmatova piktinosi Georgijum Ivanovu, Sergejum Makovskiu, jos nežavėjo ir Gle bas Struve. Amerikiečių profesorius Glebas Struvė <...> kartu su Borisu Filipovu leido Rusijos valdžios nemalonėje esančių poetų raštus. Jiedu atliko itin svarbų darbą, bet su daugybe korektūros ir kitokių klaidų, netei singų atribucijų, ne tų tekstų ir t.t. Tos klai dos, ypač klaidos biografijoje, Achmatovą labai pykino (mūsų susitikimo metu ant jos stalo kaip tik gulėjo Struvės ir Filipovo iš leistas pirmasis Gumiliovo raštų tomas). (Venclova 2013b, 510) (Venclova 2013b, 510) 1966 m. sausio 21 dieną Radauskas laiške toliau dalijasi savo įspūdžiais apie anksčiau minėtą Achmatovos poezijos rin kinį: Radauskas yra iš tų menininkų, kurie meną apgaubia ypatingo šventumo aureole. To kie menininkai visur: gamtoje, aplinkoje, buityje ieško grožio – estetinių emocijų, meninės prasmės. Jiems grožis, kurio kvin tesencija yra menas, tampa aukščiausia vertybe, o meno kūryba – vienintele gyve nimo prasme. Achmatovos tome daug nežinomų vėlyvųjų eilėraščių, surinktų iš įvairiausių šaltinių, daugiausiai turbūt iš žurnalų. Patriotiniai (1941–1945) eilėraščiai silpniausi, bet tarp vėlyvųjų tikrai daug gerų. Antrajame tome bus spausdinami jos darbai apie Puškiną, atsiminimai ir po 1964 (!) parašyti eilė raščiai. Pirmajame tome daug įspūdingų jos portretų, ypač geri (ankstyvieji), tapy ti Anenkovo ir Altmano, ir vienas Tyrsos, vaizduojantis ją jau vidutinio amžiaus. Mo diglianio piešinys neatspindi tikrosios Ach matovos ir apskritai yra silpnas. 11 Zikaras Juozas (1881–1944) – skulptorius. Kūrė portretinius biustus, figūrines kompozicijas, reljefines skulptūras, antkapinius paminklus. Medalionų, medalių ir monetų autorius. 1924 m. ir 1936 m. sukūrė lito ir cento monetų seriją (Biliukevičiūtė 2014, 600). 10 Mackevičius Jonas (1872–1954) – dailininkas, tapytojas. Nuo 1907 m. dalyvavo parodose. 1926 m. ir 1928 m. Kaune surengė individualias parodas. Plačiau apie J. Mackevičių žr.: Jurkus, P. 1954. Dail. Jono Mac kevičiaus palikimas. Aidai. 8. 380. 12 Naujoji Romuva 1931 m., Nr. 31, p. 740 – iš spausdintas Henriko Radausko eilėraštis „Apie žmogų kristale“. (Girdzijauskas 1993, 5-6) (Girdzijauskas 1993, 5-6) Laiškuose I. Ivaskui Radauskas ana lizuoja gautus Achmatovos rinkinius bei (Radauskas 2009, 220) 140 Taigi, Radauskas ne tik puikiai žinojo Achmatovos poeziją, bet ir domėjosi jos portretais, buvo susikūręs savo Achma tovos įvaizdį. Kadangi Radauskas mini, kad portretai beveik užmiršti per 25 me tus, vadinasi jis susipažino su jos portre tais dar būdamas Lietuvoje. Pavyzdžiui, 1931 m. iliustruotame savaitiniame kul tūros gyvenimo žurnale Naujoji Romuva buvo išspausdintas garsus Natano Altma no Achmatovos portretas. Menotyrininko Justino Vienožinskio straipsnyje „Dai lininkų J. Mackevičiaus10 ir J. Zikaro11 meno darbų paroda“ apie Achmatovos portretą rašoma: Įdomu, kad amžininkai pastebėdavo Radausko žmonos Veros Radauskienės pa našumą į Achmatovą. Taip Alfonsas Nyka- Niliūnas dienoraščiuose rašė: Šiandien Sandra su rituališka rimtim pri siminė p. Radauskienę (kuriai vasario mė nesy būtų suėję šimtas metų) <...> Ji buvo nuostabus žmogus, gyvenęs šalia tokio pat nuostabaus širšės Radausko. Savo stotu, laikysena ir dar kažkuo ji priminė Ach matovą. Mūsų artimesnės pažinties pra džioje man, tada dar visai jaunam žmogui, šiek tiek trukdė jos išdidi (hautaine) laiky sena. Bet tai buvo mano kaltė: aš jos nesu pratau (vėliau mes susidraugavome). (Nyka-Niliūnas 2009, 123-124) (Nyka-Niliūnas 2009, 123-124) (Nyka-Niliūnas 2009, 123-124) Charakteringas modernisto Altmano Ach matovos portretas vėl duoda komponuotą ir podraug realų vaizdą – čia visos spal vinės dėmės ir atskirų fragmentų meniškoji struktūra vienodai traktuotos, formos taip komponuotos, kad vieną kitą papildo, pa stiprina, o visos drauge stilingo ornamento pavidalu sudaro vieną meno daiktą... Galimas daiktas, kad Achmatovos por tretai matyti, kaip pats Radauskas rašo, prieš 25 metus dar nepriklausomoje Lietu voje, patraukė jo dėmesį, nes priminė jam gyvenimą tėvynėje. Nepamirškime ir jo domėjimosi menais, jis buvo tikras meno žinovas ir vertintojas. Pasak Nykos-Niliū no, „menai jam buvo absoliutus pasaulio centras“ (Nyka-Niliūnas 1978, 13). (Vienožinskis 1931, 228) Naujojoje Romuvoje buvo spausdinami ir Henriko Radausko eilėraščiai12, tad tikė tina, kad jis matė ir šį, išspausdintą žurna le, Achmatovos portretą. Iš J. Vienožinskio straipsnio konteksto matome, kad Achma tovos vardas buvo jau gerai žinomas Lie tuvos skaitytojui. Nykos-Niliūno dienoraščiuose yra pri siminimas, datuotas 1966 m. gruodžio 9 d., kuris gali būti siejamas su Radausko Ach matovos kaip vertėjos vertinimu: Su abiem Radauskais Pennsylvania Avenue knygyne. Iš Henriko bendradarbio gavę Sa lomėjos eilėraščių knygą rusų kalba (Stiho tvorenija i poemy). Jie abu jai nepalankūs ir abu, tarytum susitarę, sako, kad rusiški vertimai labai dažnai geriau skambą už lie tuviškus originalus. 13 Randall Jarrell (1914–1965), Amerikos poetas, kritikas, eseistas, vertė H. Radausko poeziją („Žiemos pasaka“, „Gaisras Panoptikume“, „Veneros gimimas“) (Blekaitis 1994, 306). (Nyka-Niliūnas 2002, 439) (Nyka-Niliūnas 2002, 439) (Nyka-Niliūnas 2002, 439) S. Nėries rinktinė Stichotvoreniia i po emy buvo išleista 1953 m., joje buvo ir 12 Achmatovos vertimų (Нерис 1953). Ma nau, kad tas faktas galėjo turėti įtakos Ra dausko rinktinės vertinimui. 141 natūralus, taip pat kartais ryškėja darbo pėdsakai jo kūryboje, o kartais jis roman tiškas ir retoriškas, iš dalies taip yra jo “Жемчуга”)“ (Radauskas 2009, 59-60). Laiškuose Ivaskui yra Radausko pasi sakymų apie Achmatovos kūrybą. Antai, 1952 m. jis apibūdina Achmatovos stilių kaip klasikinį, bet visiškai modernų: „Tuo tarpu Achmatova, – <...> nepaisant “kla sikinio stiliaus”, yra “visiškai moderni”“ (Radauskas 2009, 45). Radauskas pastebi, kad Achmatovos modernumas slypi ne ei lėraščio formoje ar žodžių eksperimentuo se, jos klasikiniuose posmuose sutelktos didelės prasmės, viename jos poetiniame žodyje aidi visas pasaulis. Išskirtinis ir Ra dausko poezijos modernumas. Dėl to, kad jo poezijoje beveik nepastebimi tragiškų istorinių įvykių, užgriuvusių pasaulį, at garsiai, jis „atrodė svetimas ir sunkiai su prantamas skaitytojams, pripratusiems po eziją laikyti vien tiesioginiu emociniu tam tikro laiko balsu“ (Šilbajoris 1994, 463). Bet iš tiesų jo poezijos šiuolaikiškumas ir modernumas tai vidinė įtampa, vaizdų, garsų, idėjų įvairovė, įprastų daiktų, kurie yra vaizduotės visiškai pertvarkyti: „Jam, menininkui, rūpi ne moraliniai ir sociali niai imperatyvai, gimstantys žmogui su siduriant su metafiziniais egzistencijos aspektais, o natūralios žmogaus patirties turtingumas bei įvairovė“ (Šilbajoris 1994, 463). 1964 m. Radauskas nagrinėjo R. Jarel lo13 apžvalgą, kurioje lietuvių poetas rado per daug didžių asmenybių amerikiečių poezijoje, bei siūlė į ją pažvelgti per tokią prizmę: „Ko verti visi tie Frostai, Eliotai arba Poundai, kai juos sulyginti, pavyz džiui, su Rilke, Pasternaku, Mandelštamu, arba kad ir su Tuwimu ar Achmatova!“ (Radauskas 2009, 212). 1964 m. Radauskas nagrinėjo R. Jarel lo13 apžvalgą, kurioje lietuvių poetas rado per daug didžių asmenybių amerikiečių poezijoje, bei siūlė į ją pažvelgti per tokią prizmę: „Ko verti visi tie Frostai, Eliotai arba Poundai, kai juos sulyginti, pavyz džiui, su Rilke, Pasternaku, Mandelštamu, arba kad ir su Tuwimu ar Achmatova!“ (Radauskas 2009, 212). 1964 m. Radauskas nagrinėjo R. Jarel lo13 apžvalgą, kurioje lietuvių poetas rado per daug didžių asmenybių amerikiečių poezijoje, bei siūlė į ją pažvelgti per tokią prizmę: „Ko verti visi tie Frostai, Eliotai arba Poundai, kai juos sulyginti, pavyz džiui, su Rilke, Pasternaku, Mandelštamu, arba kad ir su Tuwimu ar Achmatova!“ (Radauskas 2009, 212). Apie Achmatovos mirtį, Radauskas rašė: Apie Achmatovos mirtį, Radauskas rašė: Vakar perskaičiau, kad mirė Achmatova (prieš dieną ar dvi). (Nyka-Niliūnas 2002, 439) Jos vardas dviejose vietose buvo pakeistas į „Unna“. Nebepri simenu, ar esu Jums rašęs, kad prieš kelis mėnesius, dvi valandos prieš jos išvykimą iš Paryžiaus, Achmatovą ištiko širdies smū gis. Tuo tarpu Achmadulina kuo sveikiau sia ir plepėjo, kad „Paryžius gali būti aukš tinamas tik poemose“. (Radauskas 2009, 223) Radauskas suprato užplūdusius Ach matovą Paryžiuje išgyvenimus, o ironija viešniai iš Tarybų Sąjungos – poetei Bellai Achmadulinai – sustiprina tai. Paaiškinsiu, kad 1965 m. birželio mėn. 5 d. Achmato vai Oksfordo universitete buvo suteiktas garbės daktaro vardas. Grįždama ji trims dienoms (birželio 18–20 dd.) apsistojo Paryžiuje. Achmatova aplankė vietas, ku riose daugiau nei prieš 50 metų gyveno povestuvinės kelionės metu, susitikinėjo su tada dar jaunu dailininku Amedėju Mo diljaniu. Paryžiuje Achmatova išgyveno Tais pačiais metais Radauskas lygina Rilkės ir Achmatovos talentą: „Kad Rilke yra parašęs ir prastų eilėraščių, manęs be veik nestebina: daugelis genialių poetų ir tokių yra sukūrę. Nors, pavyzdžiui, Ach matovos kūryboje visai niekam tikusių ei lių gal ir nerasime (arba vos kelias), nors jos talentas mažesnis negu Rilke’s“ (Ra dauskas 2009, 37). Radauskas lygina Achmatovos kūrybą ne tik su Rilke, bet ir su Gumiliovu: „La bai gera yra Achmatova, – mano požiūriu geresnė už Gumiliovą (šis dažnokai ne 142 Ir kartą, rudenėjant, svetimšalę Aš, saulei leidžiantis, susitikau, Ir jūroje šiltoj mes nusimaudėm. Keistoki josios rūbai man atrodė Ir dar keistesnės lūpos; žodžiai jos – Tartum rugsėjo naktį žvaigždės – krito. Lieknoji plaukioti mane mokino Ir ant stangrių bangų ji mano kūną Dar nepatyrusį ranka prilaikė Dažnai stovėdama vandens žydrynėj, Jinai iš lėto su manim kalbėjo, Ir man atrodė, kad viršūnės miško Palengva ūšia, arba šnera žvyras, Arba sidabriniu balsu vamzdelis Apie atsiskyrimą toly gieda17 (Achmato a 1953 2) jaunystės prisiminimus bei pajuto kontras tą tarp sovietinio gyvenimo uždarumo ir laisvo gyvenimo, tai matyti iš jos pačios replikos: „Jeigu Jūs žinotumėte, kas tai!... štai taip sėdėti... o aplinkui visi tie žmonės, tas jaunimas... įeina, išeina, juokiasi, links mi, gyvybingi, nerūpestingi“14 (Адамович 2014, 469-470). Radausko kūrybiniame palikime yra ir vertimų. Jis vertė J. Baltrušaitį, B. Paster naką, A. Achmatovą, A. Bloką, I. Babelį, J. W. Goethę, H. Heinę, Cz. Miłoszą (Sat kauskytė 2001a, 407). 1953 m. Literatūros lankuose pasiro dė Achmatovos eilėraščio „V to vremia ja gostila na zemle“ vertimas, pasirašytas slapyvardžiu Jonas Pelėnas15. Šis eilėraš tis pirmą kartą buvo išspausdintas 1914 m. Achmatovos rinkinyje Čiotki pavadini mu „Otryvok iz poemy“. 17 «Однажды поздним летом иностранку / Я встретила в лукавый час зари, / И вместе мы купались в теплом море, / Ее одежда странной мне казалась, / Еще страннее – губы, а слова – / Как звезды падали сентябрьской ночью. / И стройная меня учила пла вать, / Одной рукой поддерживая тело, / Неопытное на тугих волнах» (Ахматова 1990: I, 158). naši į Puškiną. Achmatova žavėjosi Puškinu, tyrinėjo jo kūrybą. Tomas Venclova prisimena: „...kai kalbėjau apie rusų poetus, kurie man yra svarbūs. Leidau sau pasakyti, kad Puškinas dabar kažkaip mažiau aktualus. Į tai Anna Andrejevna atkirto: “Puškino Jūs nelieskit“. Ir perskaitė ištisą pamokslą apie tai, kad Puškino nega lima liesti“ (Venclova 2013b, 512). 14 «Если бы вы знали, что это!...вот так сидеть...а вокруг все эти люди, эта молодежь... входят, выходят, смеются, веселые, оживленные, беспечные». 15 Vienas iš Henriko Radausko slapyvardžių – Jonas Pelėnas (Lietuviškieji slapyvardžiai 2004, 725). 16 Achmatovos eilėraštyje Mūza apibūdi nama kaip „tamsiaveidė“ («смуглая») ir „svetimša lė“ («иностранка»). Mūzos apibūdinimas siejamas su Achmatovos eilėraščiu „V Carskom sele“ («В Царском селе») (1911): „Smuglyi otrok brodil po alle jam, / U oziornych grustil beregov“ ( «Смуглый отрок бродил по аллеям, / У озерных грустил берегов»). 1958 m. eilėraštis buvo išspausdintas Achmatovos rinktinėje Stichotvorenija (Стихотворения) pavadi nimu „Puškin“ («Пушкин») (Ахматова 1990: I, 370). „Tamsiaveidė“ ir „svetimšalė“ Achmatovos Mūza pa 14 «Если бы вы знали, что это!...вот так сидеть...а вокруг все эти люди, эта молодежь... входят, выходят, смеются, веселые, оживленные, беспечные». 15 Vienas iš Henriko Radausko slapyvardžių – Jonas Pelėnas (Lietuviškieji slapyvardžiai 2004, 725). Alfonsas Nyka-Niliūnas (1919– 2015) nuojančios ir grojančios kalvos“ (Radauskas 1999, 124). Palyginimui Achmatovos eilės: „Il’ golosom serebrianym volynka / Vdali pojot o večere razluk“. Radausko eilėrašty je „Dovanos“ rašoma apie nepaprastą poeto likimą ir Mūzą: „...mūzos neša dovanų – / Neišsipildančių sapnų, / Neišsibaigian čių troškimų, / Ir vandenyno balsą kimų, / Ir laimę, – ji dar neatėjo, – / Ir laivo burę, pilną vėjo“ (Radauskas 1999, 90). Achma tovos eilėraščiuose Mūza liūdna, verkianti, pranašaujanti išsiskyrimo džiaugsmą: „Kak vestnika nebesnogo, molila / Ja devušku pe čal’nuiu togda: / Skaži, skaži, začem ugasla pamiat’ / I, tak tomitel’no laskaia sluch, / Ty otniala blaženstvo povtoren’ia?“. Arba ki tame Achmatovos eilėraštyje „V Carskom sele“ Mūza vaizduojama kaip Jekaterinos parko skulptūra: „cholodnyi, belyi“, „moj mramornyi dvojnik“ (Ахматова 1990: I, 26). Radausko eilėraštyje „Mūza“ ji taip pat akmeninė: „Kaip statula išauga kamba ry“, su „<...> aklų akmeninų veidu“, kaip „akmuo paduoda vyrui stilių ir lentelę“ (Ra dauskas 1999, 24). Alfonso Nykos-Niliūno dienoraščiuose ne kartą užsimenama apie Achmatovos poe ziją, knygas, atskirus eilėraščius (Nyka- Niliūnas 2002; Nyka-Niliūnas 2009). Iš tų ištraukų galima pabandyti suprasti, kokią vietą rusų poetės kūryba užima Nykos-Ni liūno poetiniame pasaulyje. Į Vakarus jis pasitraukė 1944 m. Iki 1949 m. gyveno Vokietijoje, o paskui apsigyveno Balti morėje (JAV). Artimai bendravo su Jonu Aisčiu, Henriku Radausku, Jurgiu Blekai čiu. 1986 m. Baltimorėje Nyka-Niliūnas parašė eilėraštį „Autobiografija 1986“, kuriame filosofiškai pasakoja apie savo gyvenimą, išskirdamas jame tris etapus. Pirma eilėraščio dalis – vaikystė: „Gimiau ir augau dar tėvų / Fantazijos sukurtoje valstybėje. / Turėjau savo / Namus ir savo vardą“ (Nyka-Niliūnas 1996, 414). Vai kystę Nyka-Niliūnas praleido tėvų namuo se, Nemeikščiuose, netoli Utenos, kur jam buvo įskiepyta meilė literatūrai, poezijai, gimtajam žodžiui. Vaikystės vaizdai lydėjo poetą visą gyvenimą, įkvėpdavo jo filoso finį būties ir trukmės pasaulio suvokimą. Vaikystės pasaulis buvo jo pradžia ir pa baiga, jo kūrybos atspirties taškas. Antras jo gyvenimo etapas – emigracija atsispindi antroje eilėraščio dalyje: Savo straipsnyje apie vertimo meną Radauskas rašo, kad vertėjas labai gerai turi būti išsinagrinėjęs verčiamo autoriaus kūrybą, įsijausti į autoriaus asmenybę. Vertime svarbus „ne vien žodinis ir pras minis tikslumas (nors tai irgi turi nemažą reikšmę), bet ir verčiamo autoriaus tempe ramentas, jo veidas, jo balsas, jo literatūri nė maniera, jo stilius“ (Radauskas 1940a, 111). Radauskas teigė, „kad geriau visai neversti tokių autorių, kurių visa dvasinė struktūra, pasaulėžiūra ir temperamentas yra griežtai priešingi paties vertėjo simpa tijoms ir palinkimams“ (Radauskas 1940b, 256). (Nyka-Niliūnas 2002, 439) Vėliau jis įėjo į ciklą „Epičeskie motivy“ kartu su dar dviem eilėraščiais „Pokinuv rosči rodi ny sviasčennoi“ ir „Smerkaetsia, i v nebe temno-sinem. Pirmą kartą visi trys eilėraš čiai kaip vienas ciklas buvo išspausdinti 1943 m. Achmatovos rinktinėje Izbrannoe (Ахматова 1990: I, 395-396). (Achmatova 1953, 2) (Achmatova 1953, 2) Kodėl Radauskas pasirinko būtent šitą eilėraštį vertimui, nuorodų nepavyko aptik ti. Galbūt Radauskui buvo artima eilėraščio tema – poetas, poezijos gimimas, Mūza. Gamtos vaizdai – jūra, žvaigždės, naktis, žemė, saulėlydis kaip neatsiejamas kūry bos, poezijos gimimo kontekstas eilėraštyje galėjo patraukti Radausko dėmesį. Radaus ko kūryboje svarbiausios temos yra: gam ta, grožis, poetas ir menas (Girdzijauskas 1994, 518-543). Su Achmatovos eilėraščiu susišaukia kai kurios Radausko eilės. Pa vyzdžiui, eilėraštyje „Dainos gimimas“: „Ir vamzdį paimu, ir groju, ir dainuoju / Su vėju ir paukščiu ir su medžiu baltuoju, / Ir ūžia debesys nežemiškos spalvos / Virš tos dai Achmatovos eilėraštyje lyrinė herojė sutinka savo Mūzą16: 143 nuojančios ir grojančios kalvos“ (Radauskas 1999, 124). Palyginimui Achmatovos eilės: „Il’ golosom serebrianym volynka / Vdali pojot o večere razluk“. Radausko eilėrašty je „Dovanos“ rašoma apie nepaprastą poeto likimą ir Mūzą: „...mūzos neša dovanų – / Neišsipildančių sapnų, / Neišsibaigian čių troškimų, / Ir vandenyno balsą kimų, / Ir laimę, – ji dar neatėjo, – / Ir laivo burę, pilną vėjo“ (Radauskas 1999, 90). Achma tovos eilėraščiuose Mūza liūdna, verkianti, pranašaujanti išsiskyrimo džiaugsmą: „Kak vestnika nebesnogo, molila / Ja devušku pe čal’nuiu togda: / Skaži, skaži, začem ugasla pamiat’ / I, tak tomitel’no laskaia sluch, / Ty otniala blaženstvo povtoren’ia?“. Arba ki tame Achmatovos eilėraštyje „V Carskom sele“ Mūza vaizduojama kaip Jekaterinos parko skulptūra: „cholodnyi, belyi“, „moj mramornyi dvojnik“ (Ахматова 1990: I, 26). Radausko eilėraštyje „Mūza“ ji taip pat akmeninė: „Kaip statula išauga kamba ry“, su „<...> aklų akmeninų veidu“, kaip „akmuo paduoda vyrui stilių ir lentelę“ (Ra dauskas 1999, 24). Alfonsas Nyka-Niliūnas (1919– 2015) Tai, kad Radauskas vertė Achmato vos eiles, yra svarbus faktas, įrodantis, jog jos kūryba ir pasaulėžiūra buvo jam svarbi, suprantama ir artima. Dabar aš gyvenu Savo paties lagamine, susidėliojęs Reikalingiausius kasdienybei daiktus, – Miniatiūrinius baldus ir knygų Lentynėles su savo Discours de la methode, Su savo Sein und Zeit, su savo Baigiančiom susidėvėti Kaukėm... (Nyka-Niliūnas 1996, 414) Įsidėmėtina, kad eilėraščio dalis apie emigraciją prasideda žodžiu „dabar“, kuris 144 atspindi poeto požiūrį į laiko ir gyvenimo trukmę, amžinybės bei akimirkos suvoki mą. Nykos-Niliūno žodyje „dabar“ slypi jo siekimas rasti save, savo kelią kūryboje. Į „dabarties lagaminą“ jis susideda praeities momentus ir patirtis, visa tai, kas lemia jo gyvenimą. Kaip rašo Rita Tūtlytė, „įsitrau kęs į trukmės potyrį ir trukmės veiksmą Nyka-Niliūnas fiksuoja tai, kas trukmės požiūriu svarbu dabartiškume, kas galbūt nusės sąmonėn kaip svarbus patyrimas“ (Tūtlytė 2010, 92). Šiai minčiai iliustruoti R. Tūtlytė pateikia A. Nykos-Niliūno die noraščio ištrauką apie Anos Achmatovos eilėraščius: „Du nuostabūs Achmatovos eilėraščiai („Pesnia poslednej vstreči“ ir „On liubil“) man padėjo baigti dieną, kuri šiaip būtų praėjusi beveik nepastebėta ir neišlikusi atminty. Mes patys irgi tik taip išliekame arba neišliekame“ (Nyka-Niliū nas 2002, 282). Kaip buvo sutarta, Hania18 mus nuvedė į siauroje, beveik iki langų prikritusioje ru denio lapų gatvelėje (reikėjo briste bristi) esantį privatų antikvariatą. Iš tikrųjų tai jos draugų iš Varšuvos butas, kuriame parda vinėjamos jų pažįstamų sunešamos knygos lenkų (daugiausia), prancūzų, rusų ir vo kiečių kalbomis. <...> Knygų buvo tikrai daug, beveik visos naujos. Jų gausa man atėmė apetitą, ir aš nusipirkau tik keturias nedideles knygeles: Pawlikowskos Surowy jedwab, Achmatovos Belaja staja, Andre Gide’o Paludes ir Paulio Valery Album de vers anciens. Kaip buvo sutarta, Hania18 mus nuvedė į siauroje, beveik iki langų prikritusioje ru denio lapų gatvelėje (reikėjo briste bristi) esantį privatų antikvariatą. Iš tikrųjų tai jos draugų iš Varšuvos butas, kuriame parda vinėjamos jų pažįstamų sunešamos knygos lenkų (daugiausia), prancūzų, rusų ir vo kiečių kalbomis. <...> Knygų buvo tikrai daug, beveik visos naujos. Jų gausa man atėmė apetitą, ir aš nusipirkau tik keturias nedideles knygeles: Pawlikowskos Surowy jedwab, Achmatovos Belaja staja, Andre Gide’o Paludes ir Paulio Valery Album de vers anciens. (Nyka-Niliūnas 2002, 69-70) Achmatovos knygas jam dovanodavo ir draugai. 1966 m. gruodžio 9 d. poeto dienoraščiuose yra įrašas apie gautą do vanų Achmatovos rinkinį Beg vremeni19 (Nyka-Niliūnas 2002, 428). Nykos-Niliū no svarstymuose apie moterų poeziją užsi menama ir apie Achmatovą: Moterų poezija. 18 Ona Lukauskaitė-Poškienė (1906–1983) – poetė, memuaristė. 1933 m. išėjo eilėraščių rinkinys Brangiau sios pėdos, 1939 eilėraščių knyga Eilėraščių kraitis. 1946 m. rengė Atsišaukimą į pasaulio tautas. 1946 m. nuteista ir iki 1955 m. kalinta lageriuose. Nuo 1976 m. kartu su T. Venclova, Karoliu Garucku, ir kt. dalyvavo Helsinkio grupės veikloje. Poetė mėgo Achmatovą, Gu miliovą. lenkų poetę Mariją Pavlikovskają. Jų eilėraš čius vertė į lietuvių kalbą (Striogaitė 2001, 299). 19 1965 m. Maskvoje leidykloje „Sovetskij pisatel’“ buvo išleistas Achmatovos rinkinis Beg vremeni. 20 Vladislavas Chodasevičius (1886–1939), rusų poetas, kritikas. Poeto tėvai gimė Lietuvoje, jo mo tina, Sofja įskiepijo sūnui meilę Lietuvai. Virš jos lo vos kabėjo auksinis Aušros Vartų Marijos paveikslėlis. Kiekvieną rytą ji atvesdavo į kambarį sūnų, skaityda vo jam „Poną Tadą“. V. Chodasevičius domėjosi len kų literatūra, parašė monografiją apie A. Mickevičiaus „Poną Tadą“ (Ходасевич 1996: II, 309-310.) Paveiktas N. Gumiliovo, S. Uchtomskio ir dar apie 100 inteligen tų sušaudymu, 1921 m. emigravo iš Sovietų Rusijos, padėdamas Jurgio Baltrušaičio, kuris parūpino jam ir M. Barberovai užsienio pasus (Шубинский 2011). Alfonsas Nyka-Niliūnas (1919– 2015) Aš vis buvau linkęs galvo ti, kad tokia kategorija yra mūsų kartais tiesiog liguisto palinkimo metodologiškai skirstyti, klasifikuoti, uniformuoti. Bet esa ma šitokio tokio pagrindo sakyti, kad vis dėlto yra moterų poezija. Ji skiriasi nuo vyriškosios savo neapsirinkamai kitokia dikcija, vidiniu (dvasios) ritmu, specifine atmosfera, betarpiškumu, esamu arčiau būties, antiformalizmu. Geriau įsiskaičius ir įsiklausius pasidaro aišku, kad Marceli nos Desbordes-Valmore, Emily Dickinson, Annos Achmatovos, Mary Noel ar mūsų Salomėjos Nėries poezijos nebūtų galėjęs Nyka-Niliūnas įsitikinęs, kad net nežy mūs įvykiai praeityje gali turėti įtakos as menybės formavimuisi: „Įsidėmėtinas mo mentas net ir sąmonės patamsiuose lieka aktualus ir kuria gyvenimą. Išliekame taip ir tokie, kokius atsiminimus surenkame“ (Tūtlytė 2010, 92). i Eilėraštyje „Autobiografija 1986“, ra šydamas apie „lagaminą“, kuriame lyri nis herojus laiko svarbiausius sau daiktus, t.y. svarbiausius momentus, iš kurių susi deda jo dabartis, užsimena ir apie knygų lentynėlę. Šitoje knygų lentynėlėje, kartu su Rene Dekartu ir Martinu Heideggeriu, manau, galėjo būti ir Anos Achmatovos poezijos rinktinė. Nyka-Niliūnas domė josi Achmatovos kūryba, pirko ir skaitė rusų poetės knygas. Apie tai yra įrašų po eto dienoraščiuose. Pavyzdžiui, 1943 m., dar gyvendamas Lietuvoje, Nyka-Niliūnas prisimena: 145 parašyti vyras, kaip lygiai Baudelaire’o, Mallarme, Rilkės, G. Benno, B. Pasterna ko, R. Lowellio ar mūsų Aisčio – moteris. (Nyka-Niliūnas 2002, 287) parašyti vyras, kaip lygiai Baudelaire’o, Mallarme, Rilkės, G. Benno, B. Pasterna ko, R. Lowellio ar mūsų Aisčio – moteris. (Nyka-Niliūnas 2002, 287) jos pamėgdžiotojas, galėtų būti sutrumpin tai „išverstas“ kad ir šitaip: Aš išmokiau moteris kalbėti ir rašyti... / Viešpatie, kaip man dabar jas sustabdyti?!“ (Nyka-Niliūnas 2009, 124). (Nyka-Niliūnas 2002, 287) Taigi Nyka-Niliūnas charakterizuoja Achmatovos poeziją kaip moterišką, su spe cifiškai moteriška dikcija, ritmu, tematika ir t.t. Achmatovos kūrybą, ypač ankstyvąją, kritikai nagrinėjo moterų poezijos proble mos kontekste. Poetas, kritikas Vladisla vas Chodasevičus20 straipsnyje „“Ženskie” stichi“ rašė, kad vienos talentingiausių rusų poetės Anos Achmatovos kūryba yra sintezė tarp moterų poezijos, su būdinga jai tematika ir poetika, ir poezijos plačiąja to žodžio prasme21 (Ходасевич 1996: II, 210). Nyka-Niliūnas domėjosi moterų rašy mo problema. Pripažindamas, kad moterų rašymas skiriasi nuo vyrų savo tematika, stiliumi, poetine išraiška, jis matė ir kitą pusę. Jis kritikuoja poečių menką rašymą, pavyzdžiui, bandant „mėgdžioti“, perdėtai dauginti savo tekstus: „Anos Achmatovos eilėraštis “Epigrama”22, nukreiptas prieš Nyka-Niliūnas kaip ir Henrikas Radaus kas ne tik skaitė, bet ir atmintinai mokėjo Achmatovos eilėraščius. Tai įrodo literatū riniai Nykos-Niliūno ir Henriko Radausko žaidimai, kuriuose reikėjo parodyti savo poetinę erudiciją. 1966 m. gegužės 27 d. tiek, / Kad, Dieve, jau sunku priversti jas nutilti!“ (Dau jotytė 2001, 8). 21 «В наши дни самая одаренная из русских поэтесс, Анна Ахматова, создала как бы синтез между «женской» поэзией и поэзией в точном смыс ле слова. Но этот синтез лишь кажущийся: Ахмато ва очень умна: сохранив тематику и многие приемы женской поэзии, она коренным образом переработа ла и то и другое в духе не женской, а общечеловече ской поэтики». 20 Vladislavas Chodasevičius (1886–1939), rusų poetas, kritikas. Poeto tėvai gimė Lietuvoje, jo mo tina, Sofja įskiepijo sūnui meilę Lietuvai. Virš jos lo vos kabėjo auksinis Aušros Vartų Marijos paveikslėlis. Kiekvieną rytą ji atvesdavo į kambarį sūnų, skaityda vo jam „Poną Tadą“. V. Chodasevičius domėjosi len kų literatūra, parašė monografiją apie A. Mickevičiaus „Poną Tadą“ (Ходасевич 1996: II, 309-310.) Paveiktas N. Gumiliovo, S. Uchtomskio ir dar apie 100 inteligen tų sušaudymu, 1921 m. emigravo iš Sovietų Rusijos, padėdamas Jurgio Baltrušaičio, kuris parūpino jam ir M. Barberovai užsienio pasus (Шубинский 2011). 21 «В наши дни самая одаренная из русских поэтесс, Анна Ахматова, создала как бы синтез между «женской» поэзией и поэзией в точном смыс ле слова. Но этот синтез лишь кажущийся: Ахмато ва очень умна: сохранив тематику и многие приемы женской поэзии, она коренным образом переработа ла и то и другое в духе не женской, а общечеловече ской поэтики». 22 1958 m parašyta „Epigrama“, įeinanti į „Amato paslapčių“ ciklą, V. Daujotytės knygoje Parašyta mote rų išversta taip: „Ar galėjo Bičė kurti tarsi Dantė, / Arba Laura reikšti meilės viltį? / Aš išmokiau moteris kalbėti „Ontologizuotas“ Pasternakas prieš feminizuotą Bloką. Jelabuga prieš Domodedovo. (Nyka-Niliūnas 2009, 28) Stalas. Balta užuolaida mėnulio šviesoje. Geltoni Aukso amžiaus Kūnų spalvos jurginai. Stalas. Bet mūsų nebėra. Alfonsas Nyka-Niliūnas (1919– 2015) Nykos-Niliūno dienoraščiuose yra įrašas: Vaikščiodami su Radausku, pavargę nuo „rimtų“ literatūrinių ginčų, išsigalvojame įvairiausių literatūrinių žaidimų. Vienas jų yra toks: vienam iš mūsų padeklamavus eilėraščio strofą, kitas tuoj pat turi atsakyti sekančia. Šiandieninis žaidimas baigėsi ly giomis. Mano šios dienos propozicija buvo Achmatova: Твой белый дом и тихий сад оставлю. / Да будет жизнь пустынна и светла. / Тебя, тебя в моих стихах про славлю, / Как женщина прославить не могла. Jis nė nemirktelėjęs tęsia. И ты подругу помнишь дорогую / В тобою созданном для глаз ее раю, / А я товаром редкостным торгую – / Твою любовь и нежность продаю <...> (Nyka-Niliūnas 2002, 430-431) (Nyka-Niliūnas 2002, 430-431) Po keturiasdešimt vienerių metų, 2007 m. Nyka-Niliūnas dienoraštyje cituoja ištrauką iš to paties Achmatovos eilėraščio: „Dvilypis Annos Achmatovos epitetas gy venimui: pustynna i svetla“ (Nyka-Niliūnas 2009, 274). Tai citata iš 1913 m. sukurto Achmatovos eilėraščio „Tvoi belyi dom i tichii sad ostavliu“ iš ciklo „Baltieji paukš čiai“ (Ахматова 1990: I, 73). Nykos–Ni liūno kūryboje susivienija filosofinis mąs tymas ir poetinis pasisakymas, poetas siekė žodyje sutelkti pasaulio suvokimą, per me 22 1958 m parašyta „Epigrama“, įeinanti į „Amato paslapčių“ ciklą, V. Daujotytės knygoje Parašyta mote rų išversta taip: „Ar galėjo Bičė kurti tarsi Dantė, / Arba Laura reikšti meilės viltį? / Aš išmokiau moteris kalbėti 22 1958 m parašyta „Epigrama“, įeinanti į „Amato paslapčių“ ciklą, V. Daujotytės knygoje Parašyta mote rų išversta taip: „Ar galėjo Bičė kurti tarsi Dantė, / Arba Laura reikšti meilės viltį? / Aš išmokiau moteris kalbėti 146 „Ontologizuotas“ Pasternakas prieš feminizuotą Bloką. Jelabuga prieš Domodedovo. (Nyka-Niliūnas 2009, 28) „Ontologizuotas“ Pasternakas prieš feminizuotą Bloką. Jelabuga prieš Domodedovo. (Nyka-Niliūnas 2009, 28) taforą ir epitetą suvienyti emociją ir pras mę. Idėja, mintis jo poezijoje randa išraiš ką stebinamai taikliose metaforose. Kaip pastebi J. Blekaitis, „minties ir emocijos nuostabi vienuma, neturinti sau pavyzdžio visoje mūsų poezijoje, yra jo kūrybos jėga ir išskirtinis bruožas“ (Blekaitis 1985, 19). Achmatovos epitetas gyvenimui patrau kė Nykos-Niliūno dėmesį dar ir todėl, kad atitiko lietuvių poeto pasaulio pajautimą: absoliuti tuštuma, vienatvė, kaip poezijos gimimo sąlyga, prasmės ieškojimai, vedan tis į nebūtį, žodis, kuris aukščiau nei eg zistencija, tikrovės ir sapno sąlygiškumas, palikto namo ilgesys ir laiko negrįžtamumo pojūtis, viso pasaulio dvilypumas (Kubilius 1996, 497-484). Egzistencinis viso pasaulio pajautimas išreiškiamas vienoje metaforo je, vienoje akimirkoje: (Nyka-Niliūnas 2009, 28) Nykos-Niliūno, Cvetajevos ir Achma tovos palyginimo centre – karnavalas bei karnavalo kaukės. „Baigiančios susidėvėti kaukės“, kaip neatsiejama gyvenimo dalis, minimos ir Nykos-Niliūno eilėraštyje „Au tobiografija 1986“. Literatūrinės kaukės, kurias žmogus dėvi visa gyvenimą, svar bios ir Achmatovai. Achmatovos kūryboje skirtingais gyvenimo tarpais išryškėdavo moters, žmonos, motinos, profesionalios poetės vaidmenys (Михайлова 2009, 73). Skirtingais metais ji identifikavo save, kaip asmenybė ir poetė, su Sapfo, Dante, Puški nu, su kai kuriais istoriniais, biblijiniais ir literatūriniais personažais – Žana D’Ark, didikė Morozova, Melchola, Loto žmona, Salomeja, Kleopatra, karaliumi Lyru, kara liene Gertruda, su mitologiniais herojais – Didona, Kasandra, Fedra (Цивьян 1989, 29-33). Nyka-Niliūnas apibūdino Achma tovos literatūrines kaukes kaip „karnava lo kaukes“, o Cvetajevos – kaip žmogaus „tragiškojo likimo kaukes“. Achmatova savo kūryboje pasirinkdavo „karnavalines“, tai yra žinomų istorijos, literatūros, kul tūros veikėjų vaidmenis. Manau, kad toki įspūdį Nykai-Niliūnui padarė Achmatovos „Poema be Herojaus“ kurioje rusų poetė dramatiškai apdainavo savo amžininkus, dėvinčius įvairias karnavalo kaukes. Nyka- Niliūnas nevienareikšmiškai vertina šį Ach matovos poemą. Jis nesutinka su kritikais, kurie laiko „Poemą be Herojaus“ Achmato vos geriausiu kūriniu. Nykos-Niliūno ma nymu, Achmatovos „poema ne visada pa kyla virš privačios komunikacijos lygio“, ji skirta ne visai visuomenei, o „žinantiems“, „inicijuotiems“, „dalyviams“; iššifruoti, už j j Balta užuolaida mėnulio šviesoje. Geltoni Aukso amžiaus Kūnų spalvos jurginai. Stalas. Aplinkui susirinkę kėdės (Po paskutinės vakarienės). Miręs Vidudienio, gėlių ir vyno kvapas. Pajuodę Letos topoliai ir Fornarinos Kakta. Ištroškęs gerti Vienišas ąsotis. Bet mūsų nebėra. Balta užuolaida mėnulio šviesoje. Geltoni Aukso amžiaus Kūnų spalvos jurginai. Stalas. Aplinkui susirinkę kėdės (Po paskutinės vakarienės). Miręs Vidudienio, gėlių ir vyno kvapas. Pajuodę Letos topoliai ir Fornarinos Kakta. Ištroškęs gerti Vienišas ąsotis. Bet mūsų nebėra. Balta užuolaida mėnulio šviesoje. Geltoni Aukso amžiaus Kūnų spalvos jurginai. 28 A. Achmatovos ir I. Berlino susitikimai apra šyti I. Berlino atsiminimuose “Meetings with Russian Writers in 1945 and 1956” knygoje: I. Berlin. Personal impressions. Oxford University Press, 1982. Berlino atsiminimai apie rusų poetę yra A. Naimano knygoje Рассказы о Анне Ахматове. Nyka-Niliūnas galėjo su (Nyka-Niliūnas 1989, 268) (Nyka-Niliūnas 1989, 268) 2001 m. vasarą Nykos-Niliūno dieno raštyje atsirado įrašas, kuriame jis glaustai palygino poetę Mariną Cvetajevą ir Ach matovą: 2001 m. vasarą Nykos-Niliūno dieno raštyje atsirado įrašas, kuriame jis glausta palygino poetę Mariną Cvetajevą ir Ach matovą: Cvetajeva ir Achmatova: paralelės ir prieštaros. Gyvenimo ir karnavalo paralelė. Gyvenimas prieš karnavalą. Tragiška kaukė prieš karnavalo kaukę. Pasaulis prieš erotiką, intelektas prieš emociją, būtis prieš laiką. Universalus individualizmas prieš estetizuotą egoizmą. matovą: Cvetajeva ir Achmatova: paralelės ir prieštaros. Gyvenimo ir karnavalo paralelė. Gyvenimas prieš karnavalą. Tragiška kaukė prieš karnavalo kaukę. Pasaulis prieš erotiką, intelektas prieš emociją, būtis prieš laiką. Universalus individualizmas prieš estetizuotą egoizmą. matovą: Cvetajeva ir Achmatova: paralelės ir prieštaros. Gyvenimo ir karnavalo paralelė. Gyvenimas prieš karnavalą. Tragiška kaukė prieš karnavalo kaukę. Pasaulis prieš erotiką, intelektas prieš emociją, būtis prieš laiką. Universalus individualizmas prieš estetizuotą egoizmą. 147 pildyti, kas nepasakyta, eilinis skaitytojas turi pats“ (Nyka-Niliūnas 2003, 266-267). Su tuo Nykos-Niliūno pastebėjimu negali ma nesutikti, juo labiau, kad pati Achmato va apie savo „Poemą“ sakydavo taip: „Po ema bus neįdomi ir nesuprantama tiems, kurie nežino kai kurių tų metų Peterburgo gyvenimo aplinkybių“23 (Ахматова 1990: I, 350). Nyka-Niliūnas nesižavi poema, nes „struktūrinį jos pagrindą sudaro nebe širdies ir žodžio sąmokslas, bet tam tikras scenarijus, paremtas vizualiuoju išdėstymu, daugiau negu žodžiai reiškiančiomis pauzė mis, iškalbingomis retorinėmis elipsėmis, vardais, dedikacijomis, net motto“ (Nyka- Niliūnas 2003, 266-267). Pati Achmatova stebėjosi, kad kritikai pernelyg sureikšmi na „Poemą“ ir bando ją pateikti kaip kone reikšmingiausią jos kūrinį. Iš tikrųjų Ach matova, rašydama „Poemą be Herojaus“ svarstė, ar tai nebus libretas baletui (kaip tiksliai pastebėjo Nyka-Niliūnas – „tam ti kras scenarijus“), kuriam muziką turėjo pa rašyti jos jaunystės draugas, kompozitorius Arturas Lurjė. Achmatova rašė: „Ir vienas Dievas žino ką aš rašiau: ar libretą baletui ar kino scenarijų“ 24 (Ахматова 1990: I, 354). Š kalba, tyla, tremtis ir akmuo ir sudaro tas linijas. Nykos-Niliūno eilėraštyje „Sunkiai krito paskutinis“ yra eilutės kurios, susi šaukia su Achmatovos eilėraščių „Kak be lyi kamen’ v glubine kolodca“ (Ахматова 1990: I, 116). Nykos-Niliūno eilėraštyje apjungiami tokie reiškiniai kaip kalba, tyla, tremtis ir akmuo: „Ir tik dabar aš sužinojau, / Kas yra tremtis: kapų paminklai / Tyli sve tima kalba, / Ir man dabar / Reikės rašyti amžinai ex Ponto. / Žinant, kad silurinis akmuo kieme, / Krisdamas į šulinį, / Kalba taip, kaip mes“ (Nyka-Niliūnas 1989, 247). 25 Vertimas – Juditos Vaičiūnaitės iš rinkinio (Ach matova 1964, 25). Šilbajoris straipsnyje „Filosofinių di mensijų poezija“ pateikia pažodinį šito eilėraščio dviejų eilučių vertimą: „Lyg akmuo baltas šulinio gelmėje / Manyje guli viens prisiminimas“ (Šilbajoris 1985). 27 Чуковская, Л. 1976, 1980. Записки об Анне Ах матовой. В 2-х т. Париж: «YMCA-Press»; Чуковская, Л. 1997. Записки об Анне Ахматовой. В 3-х т. Москва: Согласие. 26 Anatolijus Naimanas, poetas, vertėjas, nuo 1963 m. Achmatovos literatūrinis sekretorius, parašė atsiminimus apie Achmatovą (Найман 1989). 23 «Тем же, кто не знает некоторые «петербург ские обстоятельства», поэма будет непонятна и не интересна». 24 «И один Бог знает, что я писала: то ли балет ное либретто, то ли киношный сценарий». нравственным, и сослалась еще на довод Модильяни: “Разве мне было бы приятно узнать, что кто-то мо жет вызвать тень моей покойной матери?” – “А впро чем, – закончила она, – возьмите словарь Брокгауза на букву С и прочтите статью Владимира Соловьева „Спиритизм“, очень толковую”» (Найман 1989, 56). 31 «Дело было в Комарове, зимой, мы с Бродским и Мариной Басмановой, его подругой, зашли к Ах матовой в гости. Заговорили о спиритизме, я расска зал, что двое моих приятелей клянутся, что вызвали духов Гете и Лебедева-Кумача, те явились одновре менно и застряли в дверях». 29 Naimano knygos vertimas į anglų kalbą, apie kurį kalba Nyka-Niliūnas, pasirodė 1991 m.: Naiman, Ana toly. 1991. Remembering Anna Akhmatova / translated by Wendy Rosslyn. New York: Henry Holt. 32 1940 m. Džambulo Džambajevo eilėraščių ver timai į lietuvių kalbą buvo išspausdinti 2-me žurnalo Raštai tome ( 171-173). sipažinti su Berlino atsiminimais apie Achmatovą ir iš minėtos anksčiau Naimano knygos. y y y y 30 «Она [Ахматова] сказала, что относится к столоверчению враждебно, считая его занятием без (Nyka-Niliūnas 1989, 268) Anos Achmatovos eilėraštyje, kaip įžvelgė Šilbajoris, visi Nykos-Niliūno elementai susijungia su Achmatovos nutiesta linija tarp akmens, gyvenimo ir atminties: „Lyg balzganas akmuo, šulny nugrimzdęs, / Taip manyje prisiminimas guli. / Aš nepajėgiu ir nenoriu krimstis: / Lyg džiaugsmą saugau sielvartą didžiulį“25. Nyką-Niliūną domino ir memuarinė literatūra apie Aną Achmatovą. Jis skaitė Anatolijaus Naimano „Rasskazy o Anne Achmatovoj“26, Achmatovos daugiametės draugės Lidijos Čukovskajos „Zapiski ob Anne Achmatovoj“27 ir Isaiah Berlino at siminimus apie pokalbius su Achmatova 1945 ir 1956 metais28. Apie tai yra įrašai Rimvydas Šilbajoris straipsnyje „Filo sofinių dimensijų poezija“ recenzuodamas Nykos-Niliūno eilėraščių rinkinį Žiemos te ologija, išleistą Čikagoje 1985 m., įžvelgė Anos Achmatovos ir Nykos-Niliūno kūry binių paralelių. Nykos-Niliūno eilėraščiuo se yra vaizdinių, kuriuose „pavieniai reiški niai susisiekia vienas su kitu ir sudaro tam tikras izotopijas — linijas, apjungiančias šios poetiškos žemės paviršiaus įvairias plo tmes...“ (Šilbajoris 1985). Tokie reiškiniai, kaip vanduo, upė, mirtis, paukštis, meilė, 148 poeto dienoraščiuose. Pavyzdžiui, 1997 m. gegužės 20 d. Nyka-Niliūnas rašo: „Be veik dvi dienas buvau užsiėmęs Anatolijaus Naimano atsiminimais Remembering Anna Akhmatova (angliškai Rasskazy o Anne Achmatovoj vertimas)29. Knygoje daug įdo mių, naujas perspektyvas atskleidžiančių faktų iš pirmų rankų“ (Nyka-Niliūnas 2003, 492). Įdomu, kad Nyka-Niliūnas kritiškai vertina informaciją, pateiktą atsiminimuo se, svarstydamas: „ką aš iš jos sužinojau arba tariuosi sužinojęs?“ (Nyka-Niliūnas 2003, 492). Skaitydamas atsiminimus, Ny ka-Niliūnas suvokė, kad Achmatovos „man kadaise taip imponavusi asmeninė mistika, jos estetinis ir emocinis “šventumas”, jos intelektualinės egzistencijos legenda etc. – viskas jau buvo tapę mitais“. Poetas pabrė žė, kad tai, „savaime aišku, jokia prasme nepaliečia jos poezijos (kuri jau buvo para šyta)“ (Nyka-Niliūnas 2003, 492). Tai, kuo žavėjosi Nyka-Niliūnas, kas labiausiai jam imponavo Achmatovos asmenybėje tapo mitu, legenda. Bet svarbiausias, manau, yra poeto pastebėjimas, kad kas bebūtų parašy ta atsiminimuose apie Achmatovą, nepakeis jo požiūrio į jos poeziją. poeto dienoraščiuose. Pavyzdžiui, 1997 m. gegužės 20 d. Nyka-Niliūnas rašo: „Be veik dvi dienas buvau užsiėmęs Anatolijaus Naimano atsiminimais Remembering Anna Akhmatova (angliškai Rasskazy o Anne Achmatovoj vertimas)29. Knygoje daug įdo mių, naujas perspektyvas atskleidžiančių faktų iš pirmų rankų“ (Nyka-Niliūnas 2003, 492). Įdomu, kad Nyka-Niliūnas kritiškai vertina informaciją, pateiktą atsiminimuo se, svarstydamas: „ką aš iš jos sužinojau arba tariuosi sužinojęs?“ (Nyka-Niliūnas 2003, 492). Skaitydamas atsiminimus, Ny ka-Niliūnas suvokė, kad Achmatovos „man kadaise taip imponavusi asmeninė mistika, jos estetinis ir emocinis “šventumas”, jos intelektualinės egzistencijos legenda etc. – viskas jau buvo tapę mitais“. sipažinti su Berlino atsiminimais apie Achmatovą ir iš minėtos anksčiau Naimano knygos. 29 Naimano knygos vertimas į anglų kalbą, apie kurį kalba Nyka-Niliūnas, pasirodė 1991 m.: Naiman, Ana toly. 1991. Remembering Anna Akhmatova / translated by Wendy Rosslyn. New York: Henry Holt. 30 «Она [Ахматова] сказала, что относится к столоверчению враждебно, считая его занятием без 33 «Последняя роза. Мне с Морозовою класть поклоны, / С падчерицей Ирода плясать, / С дымом улетать с костра Дидоны, / Чтобы с Жанной на ко стер опять. / Господи! Ты видишь, я устала / Воскре сать, и умирать, и жить. / Все возьми, но этой розы алой / Дай мне свежесть снова ощутить. Комарово». 34 «Так начнется двадцать первый, золотой, / на тропинке, красным солнцем залитой, / на вопросы и проклятия в ответ, / обволакивая паром этот свет. / Но на Марсовое поле дотемна / Вы придете оди нешенька-одна, / в синем платье, как бывало уж не раз, / но навечно без поклонников, без нас». (Nyka-Niliūnas 1989, 268) Poetas pabrė žė, kad tai, „savaime aišku, jokia prasme nepaliečia jos poezijos (kuri jau buvo para šyta)“ (Nyka-Niliūnas 2003, 492). Tai, kuo žavėjosi Nyka-Niliūnas, kas labiausiai jam imponavo Achmatovos asmenybėje tapo mitu, legenda. Bet svarbiausias, manau, yra poeto pastebėjimas, kad kas bebūtų parašy ta atsiminimuose apie Achmatovą, nepakeis jo požiūrio į jos poeziją. Tai buvo Komarove, žiemą, mes su Brods kiu ir jo drauge Marina Basmanova užė jome į svečius pas Achmatovą. Prasidėjo pokalbis apie spiritizmą, aš papasakojau, kad du mano bičiuliai prisiekia, kad iškvie tė Gėtės ir Lebedevo-Kumačo dvasias, jie atėjo vienu metu ir įstrigo tarpduryje31. (Найман 1989, 56) (Найман 1989, 56) Matyt, norėdamas sustiprinti taip pa tikusį jam komišką momentą, Nyka-Ni liūnas pakeitė sovietinį poetą Vasilijų Lebedevą-Kumačą į daug įspūdingesnį Džambulą Džambajevą, mitologizuotą ka zachų poetą, kurio poezijos vertimai į rusų kalbą buvo, o originalų – ne (Кибальник 2015, 90). Verta pridurti, kad buvo ir Džambulo neegzistuojančių eilėraščių vertimų į lietuvių kalbą. Apie tą faktą Nyka-Niliūnas galėjo žinoti, nes vertimai buvo išspausdinti dar 1940 m., kai jis dar gyveno Lietuvoje32. Tame pačiame žurna lo Raštai numeryje buvo išspausdintas ir H. Radausko straipsnis apie dailiosios lite ratūros vertimo meną (Radauskas 1940b). Lidijos Čukovskajos prisiminimus Ny ka-Niliūnas vadina „fascinuojančia kny ga“, kurioje aprašomi ne tik rusų poetai Achmatova, Pasternakas bet ir sovietme čio kasdienis gyvenimas (Nyka-Niliūnas 2003, 143-144). Kalbant apie Achmatovą, Nyka-Niliūnas pastebi, kad atsiminimuose Sprendžiant iš Naimano atsiminimų, Nykai-Niliūnui labai patiko “neliečiantis Achmatovos” komiškas momentas: „kaip spiritistų iškviesti Goethe ir Džambulas, norėdami vienas kitą pralenkti, įstringa duryse....“ (Nyka-Niliūnas 2003, 492). Naimano atsiminimuose, Achmatova ne pritarė Brodskio ir Naimano pasiūlymui surengti spiritizmo seansą30: 149 ji vaizduojama „<...> gal net didesnė negu natūroje: išdidi, kiekvienu momentu žinanti savo vertę, paskendusi savyje, nuolat sle giama savo pačios enigmos“ (Nyka-Niliū nas 2003, 143-144). ji vaizduojama „<...> gal net didesnė negu natūroje: išdidi, kiekvienu momentu žinanti savo vertę, paskendusi savyje, nuolat sle giama savo pačios enigmos“ (Nyka-Niliū nas 2003, 143-144). Poeto komentaras: „Pavargusi Achma tovos fotografija ir eilėraštis (taip pat šiek tiek pavargęs)“ leidžia suprasti, kad šita me, viename iš paskutiniųjų poetės eilėraš čių, atskleidžiami Achmatovos kūrybos bei gyvenimo svarbiausi dalykai. Achmatovos eilėraštis „Posledniaja roza“ yra atsakas į Josifo Brodskio eilėraštį „A. A. Achma tovoj“ („Zakričiat i zachlopočiut petuchi“ 1962 m.). (Nyka-Niliūnas 1989, 268) Brodskio dedikuotame tekste yra eilutės kur Achmatova, vienintelė iš „sidabrinio amžiaus“ poetų įžengs į dvide šimt pirmą amžių – į aukso amžių: Isaiah Berlino atsiminimai pastūmėjo Nyką-Niliūną ironiškam filosofiškam pa mąstymui, ar gali poetas paveikti, pakeisti pasaulį: Isaiah Berlin: Achmatova tikėjusi, kad ju dviejų susitikimai Leningrade buvę tikroji Šaltojo karo priežastis. Jonas Aistis buvo giliausiai įsitikinęs, kad jo valstybės teori jos, įgyvendintos, išspręstų visas politines Lietuvos, o gal net ir viso pasaulio proble mas; kad jis yra ne tik didžiausias Lietuvos poetas, bet taip pat ir didžiausias valstybės teoretikas. Keisti žmonės tie poetai! Tak načnetsia dvadcat’ pervyi, zolotoj, na tropinke, krasnym solncem zalitoj, na voprosy i prokliatija v otvet, obvolakivaja parom etot svet. No na Marsovoje pole dotemna Vy pridiote odiniošen’ka-odna, v sinem platje, kak byvalo už ne raz, no navečno bez poklonnikov, bez nas. (Бродский 1992, 75)34 (Nyka-Niliūnas 2009, 117) (Nyka-Niliūnas 2009, 117) Poeto ir jo poezijos vieta pasaulyje yra svarbi tema kiekvienam poetui. Ny ka-Niliūnas savo dienoraštyje, manau, neatsitiktinai užrašė vieną iš paskutiniųjų Achmatovos eilėraščių „Posledniaja roza“ (1962): (Бродский 1992, 75)34 Eilėraštyje „Posledniaja roza“ poetė apmąsto savo gyvenimą, kuriame svarbiau sias dalykas buvo ir tebėra poezija, simbo liškai vaizduojama kaip rožės žiedas. Rožių žiedai Achmatovos bei kitų poetų-simbo listų tradicijoje simbolizuoja poetus bei jų eilėraščius (Серова 2002, 69-78). Šitą tradiciją matome jau anksčiau minėtame rinkinyje Buket dlia Karsavinoj, išleistame 1914 m. Eilėraščio pavadinime „paskutinė rožė“ – tai Achmatovos, paskutinės liku sios iš „sidabrinio amžiaus“ poetų, kūryba. Nyka-Niliūnas, sužinojęs apie Achmatovos mirtį, užrašė savo dienoraštyje, kad mirė paskutinė iš „sidabrinio amžiaus“ didi poe Posledniaja roza Mne s Morozovoju klast’ poklony, S padčericei Iroda pliasat’, S dymom uletat’ s kostra Didony, Čtoby s Žannoi na kostior opiat’. Gospodi! Ty vidiš, ja ustala Voskresat’, i umirat’, i žit’. Vsio voz’mi, no etoi rozy aloj Dai mne svežest’ snova osčiutit’. Komarovo 1962 (Nyka-Niliūnas 2009, 492-493)33 Posledniaja roza Mne s Morozovoju klast’ poklony, S padčericei Iroda pliasat’, S dymom uletat’ s kostra Didony, Čtoby s Žannoi na kostior opiat’. Gospodi! Ty vidiš, ja ustala Voskresat’, i umirat’, i žit’. Vsio voz’mi, no etoi rozy aloj Dai mne svežest’ snova osčiutit’. Komarovo 1962 (Nyka-Niliūnas 2009, 492-493)33 150 2001b, 536). Su Achmatovos kūryba jis susipažino dar mokykloje: „Achmatovos vardą, beje, sužinojau <...> iš to paties šal tinio, iš kurio apie jį sužinodavo dauguma to laiko moksleivių, tai yra iš Andrejaus Ždanovo rašinio. Jos eilėraščiai, aišku, mane irgi sudomino“ (Venclova 2013b, 508). ą ( , ) 36 Į Achmatovos rinkinį Poezija įėjo T. Venclovos vertimai: „Palūžo balsas, siela nepalūžo“ (1912), „Tą penktąjį metų laiką“ (1913), „Nei vyno, nei vandens negersim dviese“ (1913), „Vienuma“ (1914), „O meilės valandą“ (1915), „Bet juk kažkur lengviau“ (1915), „Iš „Epinių nuotrupų“ (1914–1916), „Tebekeliauja žodžiai. Greit atskries jie“ (1916), „Neregėtas ruduo virš pasau lio išaukštino bonią“ (1922), „Voronežas“ (1936), „Dan tė“ (1936), „Kas žvelgia į akis – į šiltą aukštį“ (1936), „Kūryba“ (1936), „O man nebūtinos ginkluotos odės“ (1940), „Sugrįžimas“ (1941), „Priešistorija“ (1945), „Iš „Šiaurės elegijų“ (1943–1953), „Nemiegant“ (1946) (Achmatova 1964). 35 Tomas Venclova emigravo į JAV 1977 m. po to, kai 1975 m. parašė Atvirą laišką, bei 1976 m. įsijungė į Lietuvos Helsinkio grupės veiklą. Atvirame laiške (Nyka-Niliūnas 2003, 51) (Nyka-Niliūnas 2003, 51) Nykos-Niliūno įžvalga apie 1977 m. JAV išleistą Venclovos poezijos rinkinį 98 eilėraščiai, kelia klausimą, kodėl Achma tovos kūryba tapo svarbi Tomui Venclovai? Venclova parašė: „Komunistinė ideologija man tolima ir, mano manymu, didele dalimi klaidinga. Jos abso liutus viešpatavimas atnešė mūsų šaliai daug nelaimių. Informaciniai barjerai ir represijos, taikomos kitaip ma nantiems, stumia visuomenę į stagnaciją, o šalį į atsili kimą“ (Venclova 1991, 19). Venclova parašė: „Komunistinė ideologija man tolima ir, mano manymu, didele dalimi klaidinga. Jos abso liutus viešpatavimas atnešė mūsų šaliai daug nelaimių. Informaciniai barjerai ir represijos, taikomos kitaip ma nantiems, stumia visuomenę į stagnaciją, o šalį į atsili kimą“ (Venclova 1991, 19). (Nyka-Niliūnas 1989, 268) Mokykloje ir universitete Venclova susipažino su „savilaidiniais griežtai drau džiamų poetų nuorašais“ – Achmatovos, Cvetajevos, Gumiliovo, Mandelštamo ei lėraščiais (Venclova 2013b, 508). Venclo va asmeniškai pažinojo Achmatovą. Sep tintojo dešimtmečio pradžioje Maskvoje, lietuvių poetas buvo susitikęs su ja daug kartų, pristatė Achmatovai jos eilėraščių vertimų į lietuvių kalbą rinktinę. 1964 m. išleistoje Achmatovos Poezijos knygoje buvo 18 paties Tomo Venclovos vertimų (Ахматова 1964)36. tė: „1966 kovo 7d. Užvakar Domodedovo sanatorijoje prie Maskvos mirė Anna Ach matova, paskutinė iš didžiojo rusų XX a. poetų ketverto (Achmatova – Pasternakas – Mandelštamas – Cvetajeva). Requiem ae ternam...“ (Nyka-Niliūnas 2002, 423). Nyka-Niliūnas išvardino poetus, apie ku rios Achmatova 1961 m. parašė eilėraštyje „Mes – ketvertas“ iš ciklo „Vainikas miru siems“ (Ахматова 1990: I, 253). Tą „ke tvertą“ sieja, kaip rašo Viktorija Daujotytė, „Vidinė likimų dramaturgija – šie keturi poetai, artimi pagal gimimo laiką, artimi ir pagal likimų linijas. Likimų jungtį gal labiausiai ir pajuto Ana Achmatova, išta rusi neatšaukiamą formulę: “Nas četvero”“ (Daujotytė 2016, 17). 1978 m. rugsėjo 5 d. Nyka-Niliūnas rašo: Pagaliau perskaičiau Tomo Venclovos 98 eilėraščius. Pirmiausia krinta į akį tai, kad, nepaisant čionykštėje spaudoje skaitytų au toriaus griežtai provakarietiškų pasisakymų (o gal aš juos netiksliai interpretavau), jo poezija jokiu būdu ne vakarietiška, sakyčiau, savo esme greičiau antivakarietiška (tatai, be abejo, jokia yda), ir jos dvasinė tėvynė ne Londonas, Paryžius ar Roma, bet Gumilio vo, Achmatovos, Pasternako ir Mandelšta mo Sankt Peterburgas (Leningradas). Emigracijoje Tomas Venclova patobu lino 1964-jų vertimus, ir 1991 m. rinkinyje Pašnekesys žiemą publikavo naujuosius variantus. Pavyzdžiui, Venclova pateikė visą Achmatovos eilėraščio „Voronežas“, skirto Osipui Mandelštamui, vertimą. 1964-aisiais šio eilėraščio vertimas dėl Tomas Venclova (g. 1937) Iki emigracijos35 Venclova jau buvo žino mas kaip poetas, 1972 m. išėjo jo poezi jos rinktinė Kalbos ženklas (Satkauskytė 151 „Amato paslaptys“ (Achmatova 1964). Dar gyvendamas Lietuvoje Venclova ėmė domėtis struktūrine poetika, semiotika, dalyvavo pirmosiose Tartu universiteto semiotinės mokyklos stovyklose. Jis pir masis aprašė lietuvių autorių tekstus, tai kydamas struktūrinės analizės principus straipsnyje „Erdvė ir laikas K. Donelaičio Metuose; Pastabos apie J. Baltrušaičio po etiką“ (Satkauskytė 2001b, 537). Venclo vos moksliniai interesai, manau, iš dalies paskatino domėtis bei tyrinėti Achmato vos eilėdarą. cenzūros buvo išspausdintas be dedikaci jos ir paskutinių keturių eilučių: „O čia, poeto tremtinio buveinėj, / Kur lankosi tai Mūzos, tai mirtis, / Tėra naktis, / Naktis, kuri jau niekad nepraeina (Venclova 1991, 251). T. Venclova atsiminimuose rašo, kad Achmatova pati padiktavo jam trūkstamas eilutes: cenzūros buvo išspausdintas be dedikaci jos ir paskutinių keturių eilučių: „O čia, poeto tremtinio buveinėj, / Kur lankosi tai Mūzos, tai mirtis, / Tėra naktis, / Naktis, kuri jau niekad nepraeina (Venclova 1991, 251). T. Venclova atsiminimuose rašo, kad Achmatova pati padiktavo jam trūkstamas eilutes: Išsaugojau dar vieną jos autografą. Tai Mandelštamui skirtas eilėraštis „Vorone žas“. Jis buvo spausdinamas be paskutinių jų keturių eilučių, kurios neįtiko cenzūrai. Vertime aišku, tų eilučių irgi nebuvo, apie jas nė nežinojau, nors nujaučiau, kad eilė raštis susijęs su Mandelštamu. Achmatova eilėraštį man padiktavo, o paskui pasirašė37. (Venclova 2013b, 512) Emigracijoje Venclovos tyrinėjimų centru tapo XX a. pradžios rusų poezija. Tie tyrinėjimai sudėti jo knygose Neus tojčivoe ravnovesie: vosem’ russkich poe tičeskich tekstov39 ir Sobesedniki na piru: Statji o russkoj literature40. Venclova ban dė įminti ir Achmatovos poetinio žodžio paslaptį, mokėsi iš jos poezijos meno, jo disertacijoje buvo skyrius apie Achmato vos poeziją (Mockūnas 1997, 699). 1994 m. Lietuvoje išėjo Anos Achma tovos rinktinė Erškėtis žydi, kuriame buvo ir Venclovos vertimai (Achmatova 1994). Judita Vaičiūnaitė interviu J. Jakštui papa sakojo apie šio rinkinio rengimą: „Baigiu tvarkyti Anos Achmatovos eilėraščių ver timus. <...> Pridėsiu ir Tomo Venclovos vertimus. Atrodo, kitais metais ši verstinė knyga bus išleista“38 (Jakštas 2004, 6). 1964 m. Tomas Venclova, negalėjo dėl cenzūros rinktis, kokius Achmatovos eilė raščius versti, bet iš patvirtintų cenzūros eilėraščių jis pasirinko Achmatovos kūri nius apie poeziją, kūrybą, poetinio žodžio paslaptį. Pavyzdžiui, jis išvertė eilėraščius „Tebekeliauja žodžiai. Greit atskries jie“, „Bet juk kažkur lengviau“, „Subrėško, ir į tamsmėlynį dangų“, „Kūryba“ ir „O man nebūtinos ginkluotos odės“ iš ciklo Venclova gerai pažinojo Achmatovos poeziją, dėstė rusų poezijos kursą Yale’io universiteto studentams. 37 Pirmas autografas: Achmatova ant 1964 m. kny gos – Achmatovos eilėraščių vertimai į lietuvių kalbą – užrašė: «Томашу Венцлова тайные от меня самой мои стихи» (Venclova 2013b, 511). 38 J. Vaičiūnaitės ir J. Jakšto pokalbis vyko 1989 m. A. Achmatovos dvikalbę poezijos rinktinę Erškėtis žydi „Vyturio“ leidykla išleido tik 1994 m. 39 Венцлова, Т. 1986. Неустойчивое равновесие: восемь русских поэтических текстов. New Haven; Yale Сепter for International and Аrea Studies. Knygo je yra skyrius apie Achmatovą: «А. А. Ахматова. Из цикла “Ташкентские страницы”». 40 Т. Венцлова, T. 1997. Собеседники на пиру: Статьи о русской литературе. Vilnius: Baltos lankos. 39 Венцлова, Т. 1986. Неустойчивое равновесие: восемь русских поэтических текстов. New Haven; Yale Сепter for International and Аrea Studies. Knygo je yra skyrius apie Achmatovą: «А. А. Ахматова. Из цикла “Ташкентские страницы”». 40 Т. Венцлова, T. 1997. Собеседники на пиру: Статьи о русской литературе. Vilnius: Baltos lankos. Tomas Venclova (g. 1937) Maža to, manau, galima teigti, kad rusų poetės kūryba įsi skverbė į Venclovos poetinį pasaulį. Jis vertė Achmatovos poeziją, tobulino jau išverstus eilėraščius, nes vertimai buvo Venclovai svarbi literatūrinė misija, jam labiausiai rūpėjo autoriai, kurių niekas nevertė dėl to, kad jie buvo draudžiami ir nepriimtini tuometinei valdžiai ir cen zūrai. Pasak poeto, dar svarbesnis buvo asmeninis motyvas: „visi eilėraščiai, ku 152 riuos verčiau, man patiko, be kai kurių ne įsivaizdavau savo kasdienybės“ (Venclova 2006, 5). Venclova parašė eilėraščių, skirtų Achmatovai („Po paskaitos“) ir jos sušau dytam vyrui, poetui Nikolajui Gumilio vui („Kavaleristas prie Seinų“) (Venclova 2005, 43; Venclova 2013a, 14). <...> Eilėraštis sapnuojasi pats sau. Mes turime sudrumsti tuos sapnus, išvesti jį iš ten – tiksliau, iš nieko – į sintaksės, fonetikos aikštės. Ir tai sunku. Čia lemia ne kalba, bet, įtariu, kažkas už ją aukštesnis. Jis auga lyg miglotas gaudesys ir baigus darbą įkvepia jam dvasią. <...> Bet mes to neaprėpsim. Mums belieka detalės: tarkime, garnys pakrantėj, neišvaizdi Depression Gothic arka ar piktžolės po balkonu, kur būta aptriušusių auksuotų raidžių: Deus conservat omnia. Ir to gana, kad eilės sutalpintų visą meilę, kurią esi patyręs. Kad paguostų net mirusius. Įdomus Tomo Venclovos Achmatovos portretas poetiškai pateiktas eilėraštyje „Po paskaitos“: „<...> Pasaulis ėmė keistis. / <...> ryškėjo kambarys – / aikštingo, išdi daus neturto būstas / nebesančioj šaly, ki tam laike, / ir moteris, nugrimzdusi į krėslą, / sena, apsunkus, bet keistai grakšti, / kurią kadaise pažinau“ (Venclova 2005, 43). Ei lėraštyje yra įsivaizduojamas pokalbis su Achmatova apie poetinio žodžio paslaptį. Poezijos vertimų knygoje Kitaip yra pra tarmė „Anna Achmatova“, kur poetas rašo: „Mano eilėraščio “Po paskaitos” herojė yra Achmatova: įsivaizduojamas jos monolo gas parašytas mano paties, bet drįstu many ti, kad jo mintys poetei buvo nesvetimos“ (Achmatova 2006, 263). Pokalbyje tarp poetų išryškėja požiūris į poezijos, poeti nio žodžio gimimą. Lyrinė herojė iš dalies sutinka su autoriaus dėstomais studentams teiginiais apie poeziją: (Venclova 2005, 43-44) Eilėraštyje vyksta poetų dialogas (nors iš tikrųjų jis tėra autoriaus įsivaizduoja mas), kuris atskleidžia Achmatovos poeti nio meistriškumo svarbą Tomui Venclovai. Tomas Venclova įtraukdavo Achmatovos poeziją į savo eilėraščius, cituodamas ir perfrazuodamas jos eiles. Venclovos ei lėraštyje „Po paskaitos“ minimą reikš minga Achmatovai frazę „Deus conservat omnia“41 (Dievas viską saugo), kuri tapo „Poemos be Herojaus“ epigrafu. 41 Deus conservat omnia – lotyniškas užrašas ant Fontano rūmų, kur Achmatova gyveno, kai rašė „Po emą be Herojaus“: «Девиз в гербе на воротах дома, в котором я жила, когда писала поэму» (Ахматова 1990: I, 304). 42 Eilėraštį „Subrėško, ir į tamsmėlynį dangų“ iš ciklo „Epinės nuotrupos“ išvertė T. Venclova: „Ant tilto, tarp rūdijančių turėklų / Keli vaikai mosavo pirštinaitėm / Ir lesino godžias ir margas antis – / Jos vartės eketėj lyg rašalinėj“ (Achmatova 1994, 111-113). Tomas Venclova (g. 1937) Josifas Brodskis rašo apie Tomo Ven clovos poezijos intertekstualumą: „Esama dar vienos poeto pareigos, paaiškinančios jo atsidavimą formai, – tai pareiga pirmta kams, poetinės kalbos kūrėjams, kurių įpė dinis jis yra. Ji pasireiškia kaip bet kokio sąmoningo rašytojo nuojauta, jog jis turi rašyti taip, kad būtų suprastas savo pirm takų – tų, iš kurių mokėsi poetinės kal bos“ (Brodskij 1999, 117). Venclovos ei <...> jog priebalsiai pažįsta vienas kitą ir susišaukia, lyg nuklydę girion medžiotojai – dusliai, bet įsakmiai; jog balsiai leidžiasi ir kopia laiptais; jog skiemenį paantrina skiemuo, <...> Bet tai ne viskas, nes eilutėj slepias vos beįskaitomos garsyno mįslės: kadaise – uždraustas dievybės vardas, o šiandien žodžiai, ženklinantys daiktus, menkus, tačiau poetui pravarčius kaip amuletai. <...> (Venclova 2005, 41-42) (Venclova 2005, 41-42) Toliau herojė teigia, kad yra šis tas dau giau, mįslė, kurią reikia įminti: 153 ma“ įvardino, kas būdinga didžiam poetui: „Kiekvienas didis poetas turi savąjį, vidinį, idiosinkratinį landšaftą, kurio fone sąmo nėje – arba, ko gero, pasąmonėje – skamba jo balsas <...> Venclova – šiaurės poetas, gimęs ir užaugęs Baltijos pajūryje, jo pei zažas monochrominis, vyrauja pilkos ir niūrios spalvos, taigi tiesiog šviesa iš vir šaus, sutirštėjusi iki tamsumos“ (Brodskij 1999, 124). Manau, kad Sankt Peterbur gas, miestas prie Baltijos jūros, su šiaurės peizažais, pavaizduotas bei apdainuotas Gumiliovo, Achmatovos, Mandelštamo, Pasternako eilėse, atitiko Venclovos poe tinį peizažą, tapo, pasak Nykos-Niliūno, jo poezijos dvasinė tėvynė. Venclovos poezi ja skirta išsilavinusiam skaitytojui, kuris sugeba atpažinti mitologinius simbolius, citatas. Apie savo poeziją jis rašo: lėraštyje, dedikuotame Josifui Brodskiui, „Kartaginoje po daugelio metų“, minimas Achmatovos mirties laikmetis bei perfra zuojamos rusų poetės eilės: Dar apsunkina pojūčius kovo pradžia, dar tamsuoja sode luošas Hermis, pusiau prisidengęs šiurkščia lentine visata, dar ne sykį iš svetimo posmo atskris, iš miglos atskardens margasparnių sužvarbusių ančių būrys ant šarmoto vandens. luošas Hermis, pusiau prisidengęs šiurkščia lentine visata, dar ne sykį iš svetimo posmo atskris, (Venclova 1998, 13) (Venclova 1998, 13) Čia perfrazuojamos Achmatovos eilės iš eilėraščio „Smerkajetsia i v nebe tem no-sinem“ iš ciklo „Epičeskie motivy“: „I na mostu, skvoz’ ržavuju rešiotku / Proso vyvaja ruki v rukavičkach, / Kormili deti pestrych žadnych utok, / Čto kuvyrkalis’ v prorubi černil’noj“42 (Ахматова 1990: I, 160). Iš šio Achmatovos „posmo“, iš pra eities „miglos“ atsirado Venclovos eilėraš tyje „ančių būrys ant šarmoto vandens“43. Mano eilėraščius, matyt, ne tiek jau daug žmonių skaito. Jie gan sunkūs, painūs. 43 Achmatovos ir Venclovo eilėraščių paralelės minimos Donatos Mitaitės knygoje apie T. Venclovos kūrybą. Žr.: (Mitaitė 2002). Tomas Venclova (g. 1937) Tai vadinamoji poesia docta, mokyta poezija, kokia praktikavo renesanso ar baroko po etai. Vienas dalykas man rūpi – kad eilėse būtu aiški prasmė. Siekiu, kad eilėraštis tu rėtų aiškią kompoziciją, pradžią pabaigą, kad jame būtų išsakytos mintys. Tų minčių judėjimą ne visada lengva sekti, bet jų ten visada yra. Achmatova mirė kovo 5 dieną, o To mas Venclova tada buvo Maskvoje ir da lyvavo atsisveikinimo ceremonijoje (Ven clova 2013b, 517). 1995 m. Tomas Venclova parašė atsi minimus apie susitikimus su Ana Achma tova (Venclova 2013b): „Šitie atsiminimai fragmentiški ir negilūs. <...> man atrodo, kad reikia išsaugoti viską, visas smulk menas apie didžius žmones“ (Venclova 2013b, 517). Josifas Brodskis straipsnyje „Poezija kaip pasipriešinimo realybei for (Klioštoraitytė 2004, 37) Svarbu paminėti, kad Venclovos ir Achmatovos susitikimuose dažnai buvo kalbama apie Lietuvą. Achmatova prisi mindavo Salomėją Nėrį ir Liudmilą Ma linauskaitę-Eglę, kurių eilėraščius vertė į rusų kalbą (Venclova 2013b, 509). Poetė pasakojo Venclovai, kad 1914 m. buvo Vilniuje, kai lydėjo vyrą, Nikolajų Gumi liovą, į frontą, meldėsi „Vilniaus šventovė je – Aušros Vartuose“. Jai patiko lietuvių liaudies skulptūros religine tematika, Ko marove (vasarvietėje netoli Leningrado) ji 154 turėjo knygą apie rūpintojėlius (Venclova 2013b, 514). ta ne visiems skaitytojams, o „žinantiems“, „inicijuotiems“, „dalyviams“. Nykos-Niliūno eilėraštyje „Sunkiai kri to paskutinis“ yra eilutės kurios, susišau kia su Achmatovos eilėraščių „Kak belyj kamen’ v glubine kolodca“. Anos Achma tovos eilėraštyje Nykos-Niliūno izotopi jos - linijos, jungiančios tokius reiškinius, kaip vanduo, upė, mirtis, paukštis, meilė, kalba, tyla, tremtis ir akmuo susijungia su rusų poetės nutiesta linija tarp akmens, gy venimo ir atminties. Achmatovos poezija Nykai-Niliūnui buvo nekintama vertybė, lietuvių poetui svarbus Achmatovos poe zijos pasaulėvaizdis, jos eilėraščiuose jis rasdavo „minties ir emocijos vienumą“, o tai atitiko ir jo paties meninę raišką. Apibendrinant galima pasakyti, kad susipažinę su Achmatovos kūryba dar Lietuvoje, H. Radauskas, A. Nyka-Niliū nas ir T. Venclova domėjosi Achmatovos poezija ir emigracijoje, jos kūryba tapo jų poetinio pasaulio dalimi. Kiekvienas iš šių poetų įsisavindavo Achmatovos kūry bą kitaip. Achmatova buvo H. Radausko mėgs tamiausių poetų sąraše, daug Achmatovos eilėraščių jis mokėjo atmintinai. H. Ra dauskas vertė Achmatovą, jo poezijoje įžvelgiamos kūrybinės paralelės su Ach matovos eilėraščiais. Galima teigti, kad pati Achmatova (poetės portretai) ir jos poezija tapo H. Radausko estetinio pasau lio dalimi, kai menas laikomas vienintele vertybe ir absoliučiu pasaulio centru. Ra dauskas perprato Achmatovos kūrybos modernumo savitumą, nes jo paties kūryba buvo moderni bei unikali. Radauskui buvo artimos Achmatovos poezijos temos – po etas, poezijos gimimas, Mūza – tai matyti iš jo vertimo ir originalių tekstų. Tomas Venclova asmeniškai pažinojo rusų poetę, parašė eilėraščių, skirtų Ach matovai („Po paskaitos“) ir jos sušaudytam vyrui, poetui Nikolajui Gumiliovui („Kava leristas prie Seinų“). Jis vertė Achmatovos poeziją, tobulino jau išverstus eilėraščius, nes vertimai buvo Venclovai svarbi literatū rinė misija. Sankt Peterburgas, miestas prie Baltijos jūros, su šiaurės peizažais, pavaiz duotas bei apdainuotas Gumiliovo, Achma tovos, Mandelštamo, Pasternako eilėse, ati tiko Venclovos poetinį peizažą, tapo, pasak Nykos-Niliūno, jo poezijos dvasine tėvyne. A. Nykos-Niliūno dienoraščiai liudija, kad poetas nuolat domėjosi rusų poetės kūryba: skaitė jos poeziją, amžininkų at siminimus, atmintinai žinojo jos eilėraš čius. LITERATŪRA Achmatova, A. 1953. Aš žemėj svečiavausi tuo metu. Vertė Jonas Pelėnas. Literatūros lankai: Ne periodinis poezijos, prozos ir kritikos žodis. Buenos Aires. 3, 2. Jakštas, J. 2004. Judita Vaičiūnaitė: „Stengiuo si savo dvasia išlikti laisva“ (interviu). Literatūra ir menas. Vilnius: AB „Spauda“. 2987, 6-7. Jurgutienė, A. 2006. Recepcija – komparatyvis tikos metodologinė inovacija. Acta litteraria com parative. T. 1: Kultūros intertekstai. Vilnius : VPU leidykla, 64-73. Achmatova, A. 1964. Poezija. Vertė: S. Nėris, J. Vaičiūnaitė, T. Venclova. Vilnius: Vaga. Achmatova, A. 1994. Erškėtis žydi. Eilėraščiai ir poema. Vilnius: Vyturys. Klioštoraitytė, R. 2004. Rašyti eiles galima tik gimtąja kalba. Knygos šventė Bolonijoje. Pašneke sys su poetu T. Venclova. Kultūros barai. Vilnius: Všį “Kultūros barų” leidykla. 11, 37. Bernotienė, G. 2016. Apie žodžių sandūros tiks lumą. Juditos Vaičiūnaitės lyrikos vertimai ir ver tinimai. Vilnius: Lietuvių literatūros ir tautosakos institutas. Kubilius, V. 1989. Salomėjos Nėries Lyrika. Vil nius: Vaga. Biliukevičiūtė, J. 2014. Zikaras Juozas. Visuoti nė lietuvių enciklopedija. XXV. Vilnius: Mokslo ir enciklopedijų leidybos centras, 600-601. Kubilius, V. 1996. XX amžiaus literatūra: Lietu vių literatūros istorija. Vilnius: Alma littera, Lietu vių literatūros ir tautosakos institutas. Blekaitis, J. 1977a. Radauskas. Metmenys Chi cago, Illinois: M. Morkūno spaustuvė. 33, 13-67. Kuzmickas, V. 1974. S. Nėris – A. Achmatovos lyrikos vertėja. Lietuvių poetikos tyrinėjimai. Litera tūra ir kalba. T. 13. Vilnius: Vaga. Blekaitis, J. 1977b. Henriko Radausko jaunos dienos. Pagal brolio Bruno Radausko atsiminimus. Metmenys. M. Morkūno spaustuvė, Chicago, Illinois. 33, 68-84. Lietuviškieji slapyvardžiai, 2004. Lietuviškos spaudos iki 1990 m. slapyvardžių sąvadas. T. 2. Autoriai. Sudarė ir parengė Jonas Mačiulis. Vilnius: Petro ofsetas.i Blekaitis, J. 1978. Balys Sruoga: Gyvenimo ir kūrybos kelio bendrieji bruožai. Aidai. 9 lapkritis, 402-411. Prieiga internetu: http://www.aidai.eu/in dex.php?option=com_content&view=article&id=2 421:pa&catid=186:197809&Itemid=216 [2018 08 11]. Mitaitė, D. 2002. Tomas Venclova: biografijos ir kūrybos ženklai. Vilnius: Petro ofsetas. Mockūnas, L. 1997. Paskutinieji iš Lietuvos. Lietuvių egzodo literatūra 1945–1990. Vilnius: Vaga. 690-700. Blekaitis, J. 1985. Nyka-Niliūnas: „Apie savo pasaulį poezijoje“. Metmenys. Kūryba ir analizė. Chicago, Illinois: M. Morkūno spaustuvė. 49, 3-37. Nėris, S. 1984. Raštai. Vilnius: Vaga. Antras to mas. Poezija 1940–1945. Vertimai. Blekaitis, J., 1994. Henriko Radausko jaunos dienos. Pagal brolio Bruno Radausko atsimini mus. Radauskas. Apie kūrybą ir save. Recenzijos ir straipsniai. Henrikas Radauskas atsiminimuose ir kritikoje. Vilnius: Baltos lankos, 367-395. Nyka-Niliūnas, A. 1978. Žodis pomirtinei Hen riko Radausko knygai. Radauskas, H. Eilėraščiai (1965–1970): pomirtinė poezijos knyga [parengė Jurgis Blekaitis]. Chicago (Ill.): Saulius, 7-23. Nyka-Niliūnas, A. 1989. Būties erozija. Eilėraš čiai 1937–1984. Vilnius: Vaga. (Klioštoraitytė 2004, 37) Bandydamas atsakyti į jam svarbius estetinius ir filosofinius klausimus, Nyka- Niliūnas įtraukė Achmatovos asmenybę ir kūrybą į savo apmąstymus apie gyvenimą, poezijos svarbą. Venclova citavo ir perfrazavo Achma tovos eiles, įtraukė jas į savo eilėraščius. Jo poezija, vadinamoji poesia docta, skir ta išsilavinusiam skaitytojui, kuris sugeba atpažinti mitologinius simbolius, citatas. Venclova bandė įminti Achmatovos poeti nio žodžio paslaptį, mokėsi iš jos poezijos meno. Nyka-Niliūnas nevienareikšmiškai ver tino poetės „Poemą be Herojaus“. Nykos- Niliūno manymu, Achmatovos poema skir 155 LITERATŪRA Brodskij, J. 1999. Poetas ir proza. Vilnius: Bal tos lankos, 115-125. Nyka-Niliūnas, A. 1996. Eilėraščiai. 1937– 1996. Vilnius: Baltos lankos. Churginas A., Geda S., 1987. „Jei apdainuoji gi rias, tebūnie jos konsulo vertos“. Poetų Aleksio Chur gino ir Sigito Gedos pokalbis apie Henriką Radauską ir jo kūrybą. Švyturys. Vilnius: Tiesa. 21, 22-25. Nyka-Niliūnas, A. 2002. Dienoraščio fragmen tai 1938–1975. Vilnius: Baltų lankų leidyba. Nyka-Niliūnas, A. 2003. Dienoraščio fragmen tai 1976–2000. Vilnius: Baltų lankų leidyba. Daujotytė V. 1984. Janina Degutytė: Gyvenimo ir kūrybos apybraiža. Vilnius: Vaga. Nyka-Niliūnas, A. 2009. Dienoraščio fragmen tai 2001–2009 ir Papildymai 1940–2000. Vilnius: Baltų lankų leidyba. Daujotytė, V. 2001. Parašyta moterų. Vilnius: Alma littera. Daujotytė, V. 2016. Poezija – sunki, bet išverčia ma. Mes ketvertas. Sudarė ir vertė Ramutė Skučaitė. Vilnius: Santara, 14-19. Radauskas, H. 1940a. Dailiosios literatūros verti mo menas Tarybų Sąjungoj ir LTSR-oj. Raštai. Kau nas: Lietuvos TSR Valstybinė leidykla. 1, 111-115. Girdzijauskas, J. 1993. Eilėraščio grožis ir pras mė. Radauskas, H. Pasauliu netikiu, o pasaka tikiu. Kaunas: Šviesa, 5-22. Radauskas, H. 1940b. Dailiosios literatūros verti mo menas Tarybų Sąjungoj ir LTSR-oj. Raštai. Kau nas: Lietuvos TSR Valstybinė leidykla. 2, 254-258. 156 Radauskas, H. 1999. Eilėraščiai. Vilnius: Baltos lankos. Straipsniai apie literatūrą ir kultūrą. Vilnius: Lietu vių literatūros ir tautosakos institutas, 507-518. Straipsniai apie literatūrą ir kultūrą. Vilnius: Lietu vių literatūros ir tautosakos institutas, 507-518. Vienožinskis, J. 1931. Dailininkų J. Mackevi Vienožinskis, J. 1931. Dailininkų J. Mackevi čiaus ir J. Zikaro meno darbų paroda. Naujoji Romu va. Kaunas: J. Keliuotis. 10, 227-229. Radauskas, H. 2009. Henriko Radausko laiškai Ivarui Ivaskui. Sudarė S. Gaižiūnas; iš vokiečių k. vertė Raminta Gamziukaitė-Mažiulienė. Vilnius: Pa sviręs pasaulis. čiaus ir J. Zikaro meno darbų paroda. Naujoji Romu va. Kaunas: J. Keliuotis. 10, 227-229. va. Kaunas: J. Keliuotis. 10, 227-229. Адамович, Г. 2014. Мои встречи с Ахмато Адамович, Г. 2014. Мои встречи с Ахмато Satkauskytė, D. 2001a. Radauskas. Lietuvių lite ratūros enciklopedija. Vilnius: Lietuvių literatūros ir tautosakos institutas, 406-407. вой. Ахматова А. Я научила женщин говорить. Москва: АСТ, 469-470. Ахматова, А. 1952. Избранные стихотворе ния. Нью-Йорк: Издательство имени Чехова. Satkauskytė, D. 2001b. Venclova. Lietuvių lite ratūros enciklopedija. Vilnius: Lietuvių literatūros ir tautosakos institutas, 536-537. Ахматова, А. 1990. Сочинения в 2-х т. Мо сква: Правда. Sruoga, B. 1998. Raštai. Vilnius: Alma Littera. Penktas tomas. Pirma knyga. Vertimai. Бродский, И. 1992. Форма времени. Сти хотворения, эссе, пьесы. В 2-х т. Стихотворения. Минск: Эридан. T. 1. Striogaitė, D. 2001. Ona Lukauskaitė-Poškienė. Lietuvių literatūros enciklopedija. LITERATŪRA Vilnius: Lietuvių literatūros ir tautosakos institutas, 299.i Зинченко, В. Г., Зусман В. Г., Кирнозе, З. И., 2002. Методы изучения литературы. Систем ный подход. Учебное пособие для студентов, аспирантов, преподавателей-филологов. Москва: Флинта; Наука. Кибальник, С. 2015. Миф о Джамбуле (по материалам современной казахстанской печати). Известия Уральского федерального университе та. Сер. 2, Гуманитарные науки. 2 (139), 89-99. М й Г 2009 А A Зинченко, В. Г., Зусман В. Г., Кирнозе, З. И., 2002. Методы изучения литературы. Систем ный подход. Учебное пособие для студентов, аспирантов, преподавателей-филологов. Москва: Флинта; Наука. Šilbajoris, R. 1985. Filosofinių dimensijų poe zija. Aidai.5 rugsėjis-spalis. Prieiga internetu: http:// www.aidai.eu/index.php?option=com_content&task =view&id=7921&Itemid=503 [2018 08 10]. Кибальник, С. 2015. Миф о Джамбуле (по материалам современной казахстанской печати). Кибальник, С. 2015. Миф о Джамбуле (по материалам современной казахстанской печати). Известия Уральского федерального университе та. Сер. 2, Гуманитарные науки. 2 (139), 89-99. Šilbajoris, R. 1993. Lietuvių egzodo literatūra visos lietuvių literatūros raidoj. Draugas. Mokslas, menas, literatūra. Chicago, Illinois: Lietuvių katali kų spaudos draugija. lapkričio 13 d. 220 (39), 1-2. Известия Уральского федерального университе та. Сер. 2, Гуманитарные науки. 2 (139), 89-99. Михайлова, Г. 2009. Анна Aхматова в по исках «самости». Literatūra. Vilnius: Vilniaus universiteto leidykla. 51(2), 73-89. Šilbajoris, R. 1994. Radauskas – modernis tas. Radauskas. Apie kūrybą ir save. Recenzijos ir straipsniai. Henrikas Radauskas atsiminimuose ir kritikoje. Vilnius: Baltos lankos, 462-479. Найман, А. 1989. Рассказы о Анне Ахмато вой. Москва: Художественная литература. Найман, А. 1989. Рассказы о Анне Ахмато вой. Москва: Художественная литература. Нерис, С. 1953. Стихотворения и поэмы. Tūtlytė, R. 2010. Alfonsas Nyka-Niliūnas: lai kas ir trukmė. Alfonsas Nyka-Niliūnas: poetas ir jo pasaulis. Kūrybos tyrinėjimai ir archyvinė medžia ga. Vilnius: Lietuvių literatūros ir tautosakos insti tutas, 79-99. Москва: Гос. изд-во художественной литературы. Сабромене, Д. 2004. А. Ахматова в перево дах Ю. Вайчюнайте. Meninis Tekstas. Suvokimas. Analizė. Interpretacija. Mokslinių str. rinkinys. Vilnius: VPU leidykla. 4, 128-136. Vaičiūnaitė, J. 1989. A. Achmatova. Iš anksty vosios lyrikos. Įžanga. Pergalė. Vilnius: Periodika. 6, 125. Серова, М. 2002. «Цветы» в поэтическом мире Анны Ахматовой. Архетипические струк туры художественного сознания: Сб. статей. 3-й вып. Екатеринбург: Изд-во Уральского ун- та, 69-78. Режим доступа: http://ahmatova.niv. ru/ahmatova/kritika/serova-cvety-v-poeticheskom- mire-ahmatovoj.htm (см. 07 09 2018). Venclova, T. 1991a. Pašnekesys žiemą. Vilnius: Vaga. Venclova, T. 1991b. Vilties formos: Eseistika ir publicistika. Vilnius: Lietuvos rašytojų sąjungos lei dykla. Venclova, T. 1998. Reginys iš alėjos. Vilnius: Baltos lankos. Ходасевич, В. 1996. Собрание сочинений в 4 т. Москва: «Согласие». II. Venclova, T. 2005. Sankirta. Eilėraščiai. Lietu vos rašytojų sąjungos leidykla. REFERENCES brother Bruno Radauskas‘s memories]. Radauskas. Apie kūrybą ir save. Recenzijos ir straipsniai. Henri kas Radauskas atsiminimuose ir kritikoje. [Radaus kas. About creativity and self. Reviews and articles. Henrikas Radauskas in memories and criticism]. Vil nius: Baltos lankos Publ., 367-395. Adamovich, G. 2014. Moi vstrechi s Akhma tovoi. [My meetings with Akhmatova]. Ia nauchila zhenshchin govorit‘. [Akhmatova, A. I taught women to speak]. Moscow: AST Publ., 469-470. Akhmatova, A. 1952. Izbrannye stikhotvoreniia. [Selected poems]. New York: Chekhov Publishing House. Brodskii, I. 1992. Forma vremeni. Stikhotvore niia, esse, p‘esy. In 2 vol. [The form of time. Poems, essays, plays]. Minsk: Eridan Publ. Vol. 1. Akhmatova, A. 1953. Aš žemėj svečiavausi tuo metu. Vertė Jonas Pelėnas. [I came to visit earth that time]. Translated by Jonas Pelėnas. Literatūros lan kai: Neperiodinis poezijos, prozos ir kritikos žodis. Buenos Aires Publ. 3, 2. time]. Translated by Jonas Pelėnas. Literatūros lan kai: Neperiodinis poezijos, prozos ir kritikos žodis. Buenos Aires Publ. 3, 2. Brodskij, J. 1999. Poetas ir proza. [Poet and pro se].Vilnius: Baltos lankos, 115-125. Churginas A., Geda S., 1987. „Jei apdainuoji girias, tebūnie jos konsulo vertos“. Poetų Aleksio Churgino ir Sigito Gedos pokalbis apie Henriką Ra dauską ir jo kūrybą. [“If rural lays we sing, let those lays be worthy of a consul‘s-ear“. The conversation between poets Aleksis Churginas and Sigitas Geda about Henrikas Radauskas and his work]. Švyturys. Vilnius: Tiesa Publ. 21, 22-25. Akhmatova, A. 1964. Poezija. [Poetry]. Transla ted by: S. Nėris, J. Vaičiūnaitė, T. Venclova. Vilnius: Vaga Publ. Akhmatova, A. 1990. Sochineniia v 2 t. [Works in 2 vol.]. Moscow: Pravda Publ. Akhmatova, A. 1994. Erškėtis žydi. Eilėraščiai ir poema. [Eglantine in Bloom. Poems]. Vilnius: Vy turys Publ. Daujotytė V. 1984. Janina Degutytė: Gyvenimo ir kūrybos apybraiža [Janina Degutytė. Life and cre ativity outline]. Vilnius: Vaga Publ. Bernotienė, G. 2016. Apie žodžių sandūros tiks lumą. Juditos Vaičiūnaitės lyrikos vertimai ir verti nimai. [About word accuracy. Juditos Vaičiūnaitės lyric translations and evaluations]. Vilnius: Lietuvių literatūros ir tautosakos institutas Publ. Daujotytė, V. 2001. Parašyta moterų. [Written by women]. Vilnius: Alma littera Publ. Daujotytė, V. 2016. Poezija – sunki, bet išver čiama. Mes ketvertas. Sudarė ir vertė R. Skučaitė. [Poetry is hard, but translatable. Four of us. Com piled and translated by R. Skučaitė].Vilnius: Santara Publ., 14-19. Biliukevičiūtė, J. 2014. Zikaras Juozas. Visuo tinė lietuvių enciklopedija. XXV. [Universal Lithu anian Encyclopedia. XXV]. Vilnius: Mokslo ir enci klopedijų Publ., 600-601. Blekaitis, J. 1977a. Radauskas. Metmenys. LITERATŪRA Цивьян, Т. 1989. Кассандра, Дидона, Федра. Античные героини – зеркала Ахматовой. Лите ратурное обозрение. 5. 29-33. Режим доступа: http://ahmatova.niv.ru/ahmatova/kritika/civyan- kassandra-didona-fedra-antichnye-geroini.htm (см. 05 09 2018). Venclova, T. 2006. Anna Achmatova. Kitaip. Po ezijos vertimų rinktinė. Lietuvos rašytojų sąjungos leidykla. 263. Venclova, T. 2013a. Eilėraščiai. Metai. Vilnius: Lietuvos rašytojų sąjunga. 12, 10-14. Шубинский, В. 2011. Владислав Ходасевич: чающий и говорящий. С.-Петербург: Вита Нова. Venclova, T. 2013b. Atsiminimai apie Anną Achmatovą. Tomas Venclova. Pertrūkis tikrovėje. 157 REFERENCES Nyka-Niliūnas, A. 2002. Dienoraščio fragmen tai 1938–1975. [Diary fragments 1938–1975]. Vil nius: Baltos lankos Publ. Nyka-Niliūnas, A. 2003. Dienoraščio fragmen tai 1976–2000. [Diary fragments 1976–2000]. Vil nius: Baltos lankos Publ. Kubilius, V. 1989. Salomėjos Nėries Lyrika. [Sa lomėja Nėris lyrics]. Vilnius: Vaga Publ. Nyka-Niliūnas, A. 2009. Dienoraščio fragmen tai 2001–2009 ir Papildymai 1940–2000. [Diary fragments 2001–2009 and Additions 1940–2000]. Vilnius: Baltos lankos Publ. Kuzmickas, V. 1974. S. Nėris – A. Achmatovos lyrikos vertėja. Lietuvių poetikos tyrinėjimai. [S. Nėris – A. Akhmatova‘s lyric translator. Lithuanian poetic explorations]. Literatūra ir kalba. Vilnius: Vaga Publ. Vol. 13. Radauskas, H. 1940a. Dailiosios literatūros ver timo menas Tarybų Sajungoj ir LTSR-oj. [The art of translating fictional literature in the Soviet Union and the LSSR]. Raštai. [Works]. Kaunas: Lietuvos TSR Valstybinė Publ. 1, 111-115. Lietuviškieji slapyvardžiai, 2004. Lietuviškos spaudos iki 1990 m. slapyvardžių sąvadas. [Lithu anian press until 1990 pseudonyms collection]. Vol. 2. Autoriai. Sudarė ir parengė Jonas Mačiulis. [Authors. Compiled and prepared by Jonas Mačiu lis]. Vilnius: Petro ofsetas Publ. Radauskas, H. 1940b. [The art of translating fictional literature in the Soviet Union and the LSSR- oj]. Raštai. [Works]. Kaunas: Lietuvos TSR Valsty binė Publ. 2, 254-258. Mikhailova, G. 2009. Anna Akhmatova v poiskakh «samosti» [Anna Akhmatova in search of „self“]. Literatūra. Vilnius: Vilniaus universitetas Publ. 51(2), 73-89. Radauskas, H. 1999. Eilėraščiai. [Poems]. Vil nius: Baltos lankos Publ. Radauskas, H. 2009. Henriko Radausko laiškai Ivarui Ivaskui. [Henrikas Radauskas‘s letters to Ivar Ivaskas]. Sudarė S. Gaižiūnas; iš vokiečių k. vertė R. Gamziukaitė-Mažiulienė. [Compiled by S. Gai žiūnas; translated by R. Gamziukaitė-Mažiulienė]. Vilnius: Pasviręs pasaulis Publ. Mitaitė, D. 2002. Tomas Venclova: biografijos ir kūrybos ženklai. [Tomas Venclova: Signs of Bio graphy and Creative Work]. Vilnius: Petro ofsetas Publ. Mockūnas, L. 1997. Paskutinieji iš Lietuvos. [The last from Lithuania]. Lietuvių egzodo literatū ra 1945–1990. [Lithuanian exodus literature 1945– 1990]. Vilnius: Vaga Publ., 690-700. Sabromene, D. 2004. A. Akhmatova v perevo dah Ju. Vajchjunajte. Meninis Tekstas. Suvokimas. Analizė. Interpretacija. Mokslinių str. rinkinys. [Artistic Text. Reception. Analysis. Interpretation. Scientific article collection]. Vilnius: VPU Publ. 4, 128-136. Naiman, A. 1989. Rasskazy o Anne Akhmatovoi. [Stories about Anna Akhmatova]. Moscow: Khudozhestvennaia literatura Publ. Nėris, S. 1953. Stikhotvoreniia i poemy. [Po ems]. Moscow: Khudozhestvennaia literatura Publ. Satkauskytė, D. 2001a. Radauskas. Lietuvių li teratūros enciklopedija. [Encyclopedia of lithuanian literature]. Vilnius: Lietuvių literatūros ir tautosakos institutas Publ., 406-407. Nėris, S. 1984. Raštai. [Works]. Vilnius: Vaga Publ. Vol. 2. Poezija 1940–1945. Vertimai. [Poetry 1940–1945. Translations]. REFERENCES Chi cago, Illinois: M. Morkūno Publ. 33, 13-67. Girdzijauskas, J. 1993. Eilėraščio grožis ir pras mė. [The beauty and meaning of the poem]. Radaus kas, H. Pasauliu netikiu, o pasaka tikiu. [Radauskas, H. I do not believe in the world. I believe in a fairy tale]. Kaunas: Šviesa Publ., 5-22. Blekaitis, J. 1977b. Henriko Radausko jaunos dienos. Pagal brolio Bruno Radausko atsiminimus. [Henrikas Radauskas‘s young days. According to brother Bruno Radauskas‘s memories]. Metmenys. M. Morkūno press, Chicago, Illinois. 33, 68-84. Jakštas, J. 2004. Judita Vaičiūnaitė: „Stengiuosi savo dvasia išlikti laisva“ (interviu). [Judita Vaičiū naitė: „I try to keep my spirit free“ (interview)]. Lite ratūra ir menas. Vilnius: Spauda Publ. 2004-02-06. 2987, 6-7. Blekaitis, J. 1978. Balys Sruoga: Gyvenimo ir kūrybos kelio bendrieji bruožai. [Balys Sruoga: The common paths of life and creativity]. Aidai. 9 November. Available at: http://www.aidai.eu/index. php?option=com_content&view=article&id=2421: pa&catid=186:197809&Itemid=216. Accessed: 11 August 2018. Jurgutienė, A. 2006. Recepcija – komparatyvis tikos metodologinė inovacija. [Reception - metho dological innovation of comparative analysis]. Acta litteraria comparative. T. 1: Kultūros intertekstai. Vilnius: VPU Publ., 64-73. Blekaitis, J. 1985. Nyka-Niliūnas: „Apie savo pasaulį poezijoje“. [Nyka-Niliūnas: „About your world in poetry“]. Metmenys. Chicago, Illinois: M. Morkūno Publ. 49, 3-37. Khodasevich, V. 1996. Sobranie sochinenii v 4 t. [Works in 4 vol.]. Mosсow: «Soglasie» Publ. Vol. II. Kibal‘nik, S. 2015. Mif o Dzhambule (po ma terialam sovremennoi kazakhstanskoi pechati). [The Myth of Jambul (based on the materials of the mo dern Kazakh press]. Izvestiia Ural‘skogo federal‘no Blekaitis, J. 1994. Henriko Radausko jaunos dienos. Pagal brolio Bruno Radausko atsiminimus. [Henrikas Radauskas‘s young days. According to 158 by Jurgis Blekaitis)]. Chicago (Ill.): Saulius Publ., 7-23. go universiteta. Ser. 2. [Ural Federal University. Se ries. 2]. Gumanitarnye nauki Publ. 2 (139), 89-99. go universiteta. Ser. 2. [Ural Federal University. Se ries. 2]. Gumanitarnye nauki Publ. 2 (139), 89-99. Nyka-Niliūnas, A. 1989. Būties erozija. Eilėraš čiai 1937–1984. [The erosion of existence. Poems 1937–1984]. Vilnius: Vaga Publ. Klioštoraitytė, R. 2004. Rašyti eiles galima tik gimtąja kalba. Knygos šventė Bolonijoje. Pašneke sys su poetu T. Venclova. [Write verse only in your native language. Book celebration in Bologna. Con versation with poet T. Venclova]. Kultūros barai. Vilnius:Všį “Kultūros barų” Publ. 11, 37. Nyka-Niliūnas, A. 1996. Eilėraščiai. 1937– 1996. [Poems. 1937–1996]. Vilnius: Baltos lankos Publ. Kubilius, V. 1996. XX amžiaus literatūra: Lietu vių literatūros istorija. [Literature of the 20th Centu ry: History of Lithuanian Literature]. Vilnius: Alma littera, Lietuvių literatūros ir tautosakos institutas Publ. REFERENCES Satkauskytė, D. 2001b. Venclova. Lietuvių lite ratūros enciklopedija. [Encyclopedia of lithuanian literature]. Vilnius: Lietuvių literatūros ir tautosakos institutas Publ., 536-537. Nyka-Niliūnas, A. 1978. Žodis pomirtinei Hen riko Radausko knygai. [The word for the postmor tal book of Henrikas Radauskas]. Radauskas, H. Eilėraščiai (1965–1970): pomirtinė poezijos knyga (parengė Jurgis Blekaitis). [Radauskas, H. Poems (1965–1970): The postmortal poetry book (written Nyka-Niliūnas, A. 1978. Žodis pomirtinei Hen riko Radausko knygai. [The word for the postmor tal book of Henrikas Radauskas]. Radauskas, H. Eilėraščiai (1965–1970): pomirtinė poezijos knyga (parengė Jurgis Blekaitis). [Radauskas, H. Poems (1965–1970): The postmortal poetry book (written Serova, M. 2002. «Tsvety» v poeticheskom mire Anny Akhmatovoi. Arkhetipicheskie struktury khu dozhestvennogo soznaniia: Sbornik statei. 3 vyp. 159 [„Flowers“ in the poetic world of Anna Akhmatova. Archetypal structures of artistic consciousness: Col lection of articles. The third issue]. Ekaterinburg: Publishing house of the Ural University. 69-78. Available at: http://ahmatova.niv.ru/ahmatova/kri tika/serova-cvety-v-poeticheskom-mire-ahmatovoj. htm Accessed: 7 September 2018. Tūtlytė, R. 2010. Alfonsas Nyka-Niliūnas: lai kas ir trukmė. [Alfonsas Nyka-Niliūnas: time and duration]. Alfonsas Nyka-Niliūnas: poetas ir jo pa saulis. Kūrybos tyrinėjimai ir archyvinė medžiaga. [Alfonsas Nyka-Niliūnas: The poet and his world. Creative writing and archival material]. Vilnius: Lietuvių literatūros ir tautosakos institutas Publ., 79-99. Shubinskii, V. 2011. Vladislav Khodasevich: chaiushchii i govoriashchii. [Vladislav Khodase vich: Hoping and Speaking]. St. Petersburg: Vita Nova Publ. Vaičiūnaitė, J. 1989. A. Achmatova. Iš ankstyvo sios lyrikos. Įžanga. [A. Akhmatova. From the ear ly lyrics. Introduction]. Pergalė. Vilnius: Periodika Publ. 6, 125. Sruoga, B. 1998. Raštai. [Works]. Vilnius: Alma Littera Publ. Vol. 5. First book. Vertimai. [Trans lations]. Venclova, T. 1991a. Pašnekesys žiemą. [Conver sation in the winter].Vilnius: Vaga Publ. Venclova, T. 1991b. Vilties formos: Eseistika ir publicistika. [Forms of Hope: Essays]. Vilnius: Lie tuvos rašytojų sąjungos Publ. Striogaitė, D. 2001. Ona Lukauskaitė-Poškienė. Lietuvių literatūros enciklopedija. [Encyclopedia of Lithuanian Literature]. Vilnius: Lietuvių literatūros ir tautosakos institutas Publ., 299.i Venclova, T. 1998. Reginys iš alėjos. [View from the alley]. Vilnius: Baltos lankos Publ. Šilbajoris, R. 1985. Filosofinių dimensijų poezi ja. [Poetry of philosophical dimensions]. Venclova, T. 2005. Sankirta. Eilėraščiai. [Inter sections. Poems]. Lietuvos rašytojų sąjungos Publ. Aidai 1985. 5 September-October. Available at: http://www.aidai.eu/index.php?option=com_conten t&task=view&id=7921&Itemid=503. Accessed: 10 August 2018. Venclova, T. 2006. Anna Akhmatova. Kitaip. Poezijos vertimų rinktinė. [Otherwise. Collection of poetry in translation]. Lietuvos rašytojų sąjungos Publ., 263. Šilbajoris, R. 1993. Lietuvių egzodo literatūra visos lietuvių literatūros raidoj. [Lithuanian exodus literature in the development of all Lithuanian lite rature]. Draugas. REFERENCES Mokslas, menas, literatūra. Chica go, Illinois: Lietuvių katalikų spaudos draugija Publ. 220 (39), 1-2. Venclova, T. 2013a. Eilėraščiai. [Poems]. Metai. Vilnius: Lietuvos rašytojų sąjungos Publ. 12, 10-14. Venclova, T. 2013b. Atsiminimai apie Anną Achmatovą. [Remembering Anna Akhmatova]. To mas Venclova. Pertrūkis tikrovėje. Straipsniai apie literatūrą ir kultūrą. [A break in reality. Articles about Literature and Culture]. Vilnius: Lietuvių lite ratūros ir tautosakos institutas Publ., 507-518. Šilbajoris, 1994. R. Radauskas – modernistas. [Radauskas – modernist]. Radauskas. Apie kūrybą ir save. Recenzijos ir straipsniai. Henrikas Radauskas atsiminimuose ir kritikoje. [Radauskas. About creati vity and the self. Reviews and articles. Henrikas Ra dauskas in memories and criticism]. Vilnius: Baltos lankos Publ., 462-479. Vienožinskis, J. 1931. Dailininkų J. Mackevi čiaus ir J. Zikaro meno darbų paroda. [Art exhibition of painters J. Mackevičius and J. Zikaras]. Naujoji Romuva. Kaunas: J. Keliuotis Publ. 10, 227-229. Tsiv‘ian, T. 1989. Kassandra, Didona, Fedra. Antichnye geroini – zerkala Akhmatovoi. [Cassan dra, Dido, Phaedra. Antique heroines – the mirrors of Akhmatova]. Literaturnoe obozrenie. 5, 29-33. Available at: http://ahmatova.niv.ru/ahmatova/kriti ka/civyan-kassandra-didona-fedra-antichnye-geroi ni.htm. Accessed: 5 September 2018. Zinchenko, V., Zusman, V., Kirnoze, Z. 2002. Metody izucheniia literatury. Sistemnyi podkhod. Uch. posobie dlia studentov, aspirantov, prepoda vatelei-filologov. [Methods of studying literature. Systemic approach. Textbook for students, gradua te students, teachers, philologists]. Moscow: Flinta; Nauka Publ. 160 Aliona Sofija Ivinskaja S u m m a r y is not a constant fact – it is a changing, dynamic phe nomenon. The theory of reception considers the most important aspect – the accepting side – therefore, it is important to analyze the horizons of the expectations of these Lithuanian poets, their aesthetic perception of art and the way they evaluate poetry. This article takes into account the “horizon of expectation” of the Lithuanian poets, their aesthetic attitudes, per ception of art in general and poetry in particular, as well as some features of their poetic creativity. This article discusses the memoirs, diaries, transla tions, dedications, poetry and critical works of three Lithuanian exodus poets: Henrikas Radauskas, Al fonsas Nyka-Niliūnas and Tomas Venclova. These Lithuanian poets of the exodus are hardly possible to associate with any kinds of literature movements. Therefore, the reception of Akhmatova’s creative work will be reviewed separately in the context of the creative heritage of the selected authors. Under the new concept of reception, the perception process РЕЦЕПЦИЯ ТВОРЧЕСТВА АННЫ АХМАТОВОЙ В ЛИТЕРА ЛИТОВСКОГО ЗАРУБЕЖЬЯ (ГЕНРИКАС РАДАУСКАС, АЛЬФОНСАС НИКА-НИЛЮНАС, ТОМАС ВЕНЦЛОВА) Алена София Ивинская Алена София Ивинская Р е з ю м е Алена София Ивинская Р е з ю м е В свете современной теории рецепции процесс восприятия не является фактом, данным раз и навсегда, это динамический процесс, в котором центральное место занимает принимающая сто рона. В статье учитывается «горизонт ожидания» литовских поэтов, их эстетические установки, восприятие искусства вообще и поэзии в частно сти, а также некоторые особенности их поэтиче ского творчества. В статье рассмотрены воспоминания, дневники, переводы, дедикации, поэтические тексты трех поэтов литовского зарубежья – Генриха Радауска са, Альфонсаса Ника-Нилюнаса и Томаса Вен цловы. Этих представителей литературы вряд ли можно отнести к какому-либо одному литератур ному течению, поэтому их рецепция творчества Ахматовой рассматривается в контексте творче ства каждого из названных поэтов в отдельности. Адрес автора: Vilniaus universitetas Baltijos kalbų ir kultūrų institutas Universiteto g. 5 LT-01513 Vilnius Lietuva E-mail: alionaivi@yahoo.com Получено: 2018, сентябрь Принято: 2018, октябрь Получено: 2018, сентябрь Принято: 2018, октябрь 161
https://openalex.org/W2912921754	https://eejournal.ktu.lt/index.php/elt/article/download/22733/9796	English	null	Study on Transient Electromagnetic Response of High Resistivity Goafs and Its Application	Elektronika ir elektrotechnika	2,019	cc-by	2,852	ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 http://dx.doi.org/10.5755/j01.eie.25.1.22733 Manuscript received 26 May, 2018; accepted 11 November, 2018. This research was funded by National Natural Science Foundation (No. 41674133) and sub-item Major National Science and Technology (No. 2016YFC060110204). Maofei Li1, Shucai Liu1, Benyu Su1, Yongsheng Ma2, Qilong Sun2 1The School of Resource and Geoscience, China University of Mine and Technology, Xuzhou 221116, Jiangsu Province, China 2The Key Laboratory of Unconventional Petroleum geology, CGS subenyu@cumt.edu.cn Maofei Li1, Shucai Liu1, Benyu Su1, Yongsheng Ma2, Qilong Sun2 1The School of Resource and Geoscience, China University of Mine and Technology, Xuzhou 221116, Jiangsu Province, China 2The Key Laboratory of Unconventional Petroleum geology, CGS subenyu@cumt.edu.cn difference between goaf and souring rock of the earth. In recent years, geophysical workers in China have performed a lot of researches on the exploration of goafs, but the jobs are mainly focused on the low resistivity goafs filled with water [8]–[10]. The reason is that the responses of high resistivity target bodies detected by electromagnetic methods are not clearly observed. However, identified high resistivity goafs unfilled with water is the major challenge to construction of ground infrastructure in the future. So it is necessary for the detection of high resistivity goafs. Compared with other geophysical method, Transient Electromagnetic Method (TEM) has the following traits, convenient operating equipment, high efficiency and effectiveness. Based on above advances of transient electromagnetic method, we try to detect high resistance goaf by transient electromagnetic method. 1Abstract—Based on the characteristics of mine goafs, the geo-electrical models are established. In order to study the response characteristics of goaf by transient electromagnetic method, we have performed three-dimensional numerical simulations in three cases, geo-electrical model of mine geology without goaf, geo-electrical model of mine geology with low resistivity goaf and geo-electrical model of mine geology with high resistivity goaf. Moreover, the physical experiments have also been performed for the above three cases. Furthermore, we compare numeral simulation results and physical experiment results. Finally, we obtain the conclusion that although the response of high resistivity goafs is weaker than that of low resistivity, its response characters still can indicate the existing of goaf. Hence, above work actually signifies that shallow high resistivity goafs can be detected in a reliable way by transient electromagnetic method. Index Terms—Transient electromagnetic method; Time-Depth conversion; High resistance goafs; Physical modelling. Study on Transient Electromagnetic Response of High Resistivity Goafs and Its Application Maofei Li1, Shucai Liu1, Benyu Su1, Yongsheng Ma2, Qilong Sun2 1The School of Resource and Geoscience, China University of Mine and Technology, Xuzhou 221116, Jiangsu Province, China 2The Key Laboratory of Unconventional Petroleum geology, CGS subenyu@cumt.edu.cn A. 1D Case of Numerical Simulation Because of continuous coal mining in China, the problems of coal mine goafs is becoming more and more serious and it has brought a series of social problems [1]. However, the most serious goafs are in shallow earth and without collapse which are major danger potential for the safety production of coal mine. Especially, the goafs were formed by early private coal mines. Furthermore, these goafs are shallow burial and scattered distribution, hence it is very difficult to confirm the boundaries and locations. Usually, they will bring great damage to later coal mine and ground infrastructure construction [2]–[4]. The rectangular as transmitter is employed to perform 1D numerical simulation of transient electromagnetic method. It can be thought to be formed by finite small vertical magnetic dipoles [11]. Hence, first we compute the electromagnetic response of the model in the frequency domain [12], and then we convert it from frequency domain to the time domain by Gaver-Stehfest (G-S) transform method [13]. For the numerical simulation, we choose two earth models formed by high resistivity layers and low resistivity layers and the parameters of the models as shown in Table I and Table II, respectively. From Table I and Table II, we can see that we perform the numerical simulation by only changing the thickness of the first layer. Four side lengths of transmitter rectangular are all 600 meter and the current is 10 A. Besides, the measurement point is central of the rectangular. Generally speaking, the goafs are mainly classified into the high resistivity goafs unfilled with water and the low resistance goafs filled with water. The electrical properties of two kinds of goafs are quite different from the surrounding rock [5]–[7]. Therefore, the electromagnetic methods can effectively detect goafs by the resistivity TABLE I. THE PARAMETERS OF LOW RESISTIVITY. Resistivity (Ω∙m) Thickness (m) First layer 100 10, 30, 50, 70, 90, 110, 150 Second layer 0.1 10 Third layer 100 31 ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 TABLE II. THE PARAMETERS OF HIGH RESISTIVITY. A. 1D Case of Numerical Simulation Resistivity (Ω∙m) Thickness (m) First layer 100 1, 10, 30, 50, 70, 90, 110, 150 Second layer 100000 10 Third layer 100 the thickness of the first layer is more than 110 m, the difference between model with high resistivity goaf and background field is so small that we cannot recognize the existence of the goaf, especially in the case of presence of noise. (a) (b) Fig. 2. Difference curves between response of model and background field: a) Difference curves between response of model with low resistivity goaf and background field; b) Difference curves between response of model with high resistivity goaf and background field. Figure 1(a) is the numerical simulation results of model with the low resistivity goaf and Fig. 1(b) is the numerical simulation results of model with the high resistivity goaf. Compared with the Fig. 1(a) and Fig. 1(b), the electromagnetic response of model with low resistivity goaf are more sensitivity than that of model with high resistivity. However, we can improve the sensitivity of the model with high resistivity goaf by subtracting background field [14]. (a) (b) Fig. 1. Voltage curve as a function of time: a) voltage curve of model with low resistivity goaf; (b) voltage curve of model with high resistivity. ( ) (a) (b) Fig. 1. Voltage curve as a function of time: a) voltage curve of model with low resistivity goaf; (b) voltage curve of model with high resistivity. Fig. 2. Difference curves between response of model and background field: a) Difference curves between response of model with low resistivity goaf and background field; b) Difference curves between response of model with high resistivity goaf and background field. ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 background response from the two curves, as shown in Fig. 5. (a) (b) Fig. 3. The models of different goaf. Fig. 4. Response curve of TEM. This figure shows that the response value of high resistivity goaf is less than the response value of low resistivity goaf. Furthermore, in the early stage, the response value of high resistivity goaf is negative. However, the curve becomes positive in the late stage. For low resistivity goaf the response characteristic is opposite to the high resistance goaf and much higher in absolute value. The detected results, as shown in Fig. 4, are usually background response from the two curves, as shown in Fig. 5. converted into apparent resistivity and detected depth. Actually, electrical voltage value described in the vertical axis is converted into apparent resistivity by the (1), and the time value is changed to detected depth by the (3) [15] (a) (b) Fig. 3. The models of different goaf. (a) (b) Fig 3 The models of different goaf (a) 2/3 5/3 2 . 4 5 / L s r M t V S             (1) (1) In the above formula,  is magnetic permeability, 7 0 4 10 / H m      for the free air; 2 = M IL , where transmitter electrical current is I (A) and L is the side length of transmitting rectangle coil (m); V is the induced electromotive force (V); r S is the effective area of rectangle receiving coil (m2), and t is measurement time. Fig. 5. Response difference curve of TEM. (b) Fig. 3. The models of different goaf. (b) Fig. 3. The models of different goaf. Fig. 5. Response difference curve of TEM. Fig. 4. Response curve of TEM. According to the research of Jiang B. Y. B. 3D Case of Numerical Simulation Based on the 1D modelling results, we have performed the 3D numerical simulation in the following parts. According to the characteristic of goafs, two 3D models are designed. A high resistivity goaf unfilled with water and a low resistivity goaf filled with water are in the coal seam, as shown in Fig. 3, respectively. Volumes of two goafs are all 100 m × 100 m × 20 m. Besides, the size of transmitting coil is 600 m × 600 m, the transmitter current is 10 A and receivers are set in the centre of the coil. (b) Fig. 1. Voltage curve as a function of time: a) voltage curve of model with low resistivity goaf; (b) voltage curve of model with high resistivity. Figure 2 shows the curves of the difference between models electromagnetic response with goaf present and the background field. In particular, Fig. 2(a) corresponds to the model of low resistivity goaf which shows that the positive peak appears in early period. In an opposite way, Fig. 2(b) corresponds to the model of high resistivity goaf which shows that the negative peak appears in early period and the positive peak appears in the late period. In particular, when The response curves of transient electromagnetic method of above models are shown in Fig. 4. Compared with background response, we cannot detect any anomalous response behaviour from the curves, as shown in Fig. 4. However, useful results are found after subtracting the 32 REFERENCES [1] Du Kun et al., “Comprehensive evaluation of underground goaf risk and engineering application”, Journal of Central South University, vol. 42, no. 9, pp. 2803–2811, 2011. (b) [2] H. Poor, An Introduction to Signal Detection and Estimation. New York: Springer-Verlag, 1985, ch. 4. [2] H. Poor, An Introduction to Signal Detection and Estimation. New York: Springer-Verlag, 1985, ch. 4. Fig. 6. Physical model: a) Top view of physical model; b) Side view of physical model. [3] Wang Jilin et al., “Numerical analysis of effect of abandoned goaf foundation deformation on ground buildings”, Journal of Mining & Safety Engineering, vol. 25, no. 4, pp. 477–480, 2008. According to the physical model mentioned above, the induced voltage is measured, as shown at Fig. 7 which is composed by voltage measured by different time. [4] Liu Jinghua et al., “The geophysical exploration about exhausted area and sinking area in coal mine”, Journal of China Coal Society, vol. 30, no. 6, pp. 715–719, 2005. Fig. 7. Measured induced voltage. Fig. 7. Measured induced voltage. [5] Xu Ping, Mao Xianbiao, Zhang Minxia, Zhou Yuejin, Yu Bangyong. “Safety analysis of building foundations over old goaf under additional stress from building load and seismic actions”, International Journal of Mining Science and Technology, vol. 24, no. 5, pp. 713–718, 2014. DOI: 10.1016/j.ijmst.2014.03.030. [6] Xue Guoqiang et al., “Feasibility analysis of transient electromagnetic method for detectiong underground cave”, Journal of the University of Petroleum, vol. 28, no. 5, pp. 135–138, 2004. [7] Yu Chuantao, Liu Hongfu, Gao Jianping, “The experimental study on the Dual frequency induced polarization method detecting coal mine gob”, Progress in Geophysics, vol. 23, no. 5, pp. 1603–1609, 2008. [8] Chen Weiying, Xue Guoqiang, “Application on coal-mine voids detection with multi-device TEM technology”, Progress in Geophysics, vol. 28, no. 5, pp. 2709–2717, 2013. Fig. 7. Measured induced voltage. p y pp [9] Fan Tao, “Application of TEM detected goaf in coal mine with high- resistivity and shallow layer”, Int. Conf. Machine Intelligence (ICMI 2011), 2011. Usually, the value of early time is larger than that of late time. And then, based on measured voltage and measuring time, the apparent resistivity and time-depth conversion of (1), (2) and (6) are adopted, and the apparent resistivity profile is drawn, as shown in Fig. 8 and from which we can [10] Xie Haijun, “Research on TEM detection technology for water- accumlating gob in coal mines”, Ph.D. ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 (1998) [16], the equation for the velocity of propagation of the electromagnetic (EM) waves in the earth is given by the following formula   2 1/2 1 1 1 2 1/2 2 1 ( 2) 1 , 2 i i C V C C C a C                                 (2) (2) 1 2 3 (2 3)!! ( ) 1 ( ) , 4 4 !( 1)! 2 k k k C k k                   (3) 2 0 3 (2 1)!! ( ) ( ) , 4 !( 1)! 2 k k k C k k          (4) 2 = / 4 , i i a t  (5) (3) Fig. 4. Response curve of TEM. (4) This figure shows that the response value of high resistivity goaf is less than the response value of low resistivity goaf. Furthermore, in the early stage, the response value of high resistivity goaf is negative. However, the curve becomes positive in the late stage. For low resistivity goaf, the response characteristic is opposite to the high resistance goaf and much higher in absolute value. 2 = / 4 , i i a t  (5) (5) here i  is the reciprocal of apparent resistivity at moment it ;  is the coefficient of proportionality. Then the depth of the it can be calculated by (6) . i i i h V t   (6) (6) The detected results, as shown in Fig. 4, are usually 33 ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 III. PHYSICAL MODELLING find high resistivity anomalous geological bodies which are the two tunnels and one cavity. An existing model which shown in Fig. 6 is used to do physical experiments the model includes three parts from the roof to the floor. Here, both of the top part and bottom part are Quaternary clay which resistivity is low. Besides, the middle part is filled with high resistivity coal seam. A transient electromagnetic equipment named “Terra-TEM” is used to measure data. The line with red colour is measured path shown in Fig. 6(a), and the interval between two neighbour measuring points is 50 cm. The size of rectangle transmitting coil is 50 cm × 50 cm and the number of turns is 5. The size of rectangle receiving coil is 20 cm × 20 cm and the number of turns is 20. Fig. 8. Apparent resistivity section of physical model. Fig. 8. Apparent resistivity section of physical model. (a) (b) Fig. 6. Physical model: a) Top view of physical model; b) Side view of physical model. (a) Compared with models and measuring results, we can conclude that the geological model and measuring result match very well. IV. CONCLUSIONS According to the above presented research work, we can conclude the following: 1. Numerical modelling results show that the response characteristics of transient electromagnetic method of high resistivity goaf are weaker than that of low resistivity goaf. However, the response of high resistivity goaf can still be recognized by simple processing of the received voltage curves. 2. Physical modelling results show that high resistivity anomalous objects can be effectively detected. Therefore, it appears that our research can be useful for the application of transient electromagnetic method (TEM) in the exploration of the earth. (b) REFERENCES dissertation, China University of Geosciences, Beijing, 2009. [11] B. Su, J. Yu, C. Sheng, “Maxwell-equations based on mining transient electromagnetic method for coal mine-disaster water 34 ELEKTRONIKA IR ELEKTROTECHNIKA, ISSN 1392-1215, VOL. 25, NO. 1, 2019 detection”, Elektronika ir Elektrotechnika, vol. 23, no. 3, 2017. DOI: 10.5755/j01.eie.23.3.18326. detection”, Elektronika ir Elektrotechnika, vol. 23, no. 3, 2017. DOI: 10.5755/j01.eie.23.3.18326. using the Gaver‐Stehfest inverse Laplace transform method”, Geophysics, vol. 47, no. 1, pp. 47–50, 1982. DOI: 10.1190/1.1441280. [12] Li Jianhui, Zhu Ziqiang, Liu Shucai, “The transient electromagnetic field that excited by rectangular transmitting loop based on Gaver- Stehfest algorithm”, Oil Geophysical Prospecting, vol. 46, no. 3, pp. 489–492, 2011. [15] Li Maofei, “Study on the transient electromagnetic interpretation method based on the inversion model of seismic wave impedance”, Ph.D. dissertation, China University of Mining and Technology, 2017. [13] M. N. Nabighian, Ed., Electromagnetic Methods in Applied Geophysics. Society of Exploration Geophysicists, 1991. DOI: 10.1190/1.9781560802686. [16] Niu Zhilian, “Principle of time domain electromagnetic method”, M.S. thesis, The Publishing House of Central South University, 2007, pp. 22–25. [14] J. H. Knight, A. P. Raiche, “Transient electromagnetic calculations H. Knight, A. P. Raiche, “Transient electromagnetic calculation 35
https://openalex.org/W4361884521	https://figshare.com/articles/journal_contribution/Supplementary_Table_S1_from_Abiraterone_Treatment_in_Castration-Resistant_Prostate_Cancer_Selects_for_Progesterone_Responsive_Mutant_Androgen_Receptors/22458644/1/files/39909929.pdf	English	null	Supplementary Table S2. from Abiraterone Treatment in Castration-Resistant Prostate Cancer Selects for Progesterone Responsive Mutant Androgen Receptors	null	2,023	cc-by	185	Supplementary Table S1. AR T878A Mutations in genomic DNA from CRPC patients failing abiraterone or ketoconazole Patient Treatment Source T878A Frequency Read Depth BI-1 ketoconazole bone marrow 18.25% 977790 bone marrow 20.57% 769872 bone marrow 21.33% 771930 bone marrow 19.92% 1431218 bone marrow 22.01% 7592920 bone marrow 23.16% 5398942 bone marrow 21.80% 4312842 BI-2 abiraterone bone marrow 17.26% 11056 bone marrow 17.98% 26360 bone marrow 14.12% 8698 448-6 abiraterone/dutasteride liver biopsy 18.16% 389660 liver biopsy 21.00% 336294 liver biopsy 17.90% 208292 liver biopsy 14.55% 1055220 Supplementary Table S1. AR T878A Mutations in genomic DNA from CRPC patients failing abiraterone or ketoconazole Patient Treatment Source T878A Frequency Read Depth BI-1 ketoconazole bone marrow 18.25% 977790 bone marrow 20.57% 769872 bone marrow 21.33% 771930 bone marrow 19.92% 1431218 bone marrow 22.01% 7592920 bone marrow 23.16% 5398942 bone marrow 21.80% 4312842 BI-2 abiraterone bone marrow 17.26% 11056 bone marrow 17.98% 26360 bone marrow 14.12% 8698 448-6 abiraterone/dutasteride liver biopsy 18.16% 389660 liver biopsy 21.00% 336294 liver biopsy 17.90% 208292 liver biopsy 14.55% 1055220 mentary Table S1. AR T878A Mutations in genomic DNA from CRPC patients biraterone or ketoconazole
https://openalex.org/W4321791553	https://www.frontiersin.org/articles/10.3389/fendo.2023.1154170/pdf	English	null	Corrigendum: Genetically predicted C-reactive protein mediates the association between rheumatoid arthritis and atlantoaxial subluxation	Frontiers in endocrinology	2,023	cc-by	628	TYPE Correction PUBLISHED 23 February 2023 DOI 10.3389/fendo.2023.1154170 Incorrect Author Name In the published article, an author name was incorrectly written as Wengting Wang. The correct spelling is Wenting Wang. The authors apologize for this error and state that this does not change the scientiﬁc conclusions of the article in any way. The original article has been updated. KEYWORDS Mendelian randomization, rheumatoid arthritis, C-reactive protein, atlantoaxial subluxation, upper cervical instability Front. Endocrinol. 14:1154170. doi: 10.3389/fendo.2023.1154170 Front. Endocrinol. 14:1154170. doi: 10.3389/fendo.2023.1154170 COPYRIGHT © 2023 Yuan, Xiong, Zhang, Feng, Zhang, Wang and Tang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. COPYRIGHT © 2023 Yuan, Xiong, Zhang, Feng, Zhang, Wang and Tang. This is an open-access article distributed under the terms of the C eati e Commons Att ib tion License A Corrigendum on Genetically predicted C-reactive protein mediates the association between rheumatoid arthritis and atlantoaxial subluxation (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. by Yuan J, Xiong X, Zhang B, Feng Q, Zhang J, Wang W and Tang J (2022) Front. Endocrinol. 13:1054206. doi: 10.3389/fendo.2022.1054206 OPEN ACCESS OPEN ACCESS APPROVED BY Frontiers Editorial Ofﬁce, Frontiers Media SA, Switzerland CORRESPONDENCE Jia Tang tj676533@163.com Wenting Wang wangwentingtn@sina.com †These authors have contributed equally to this work SPECIALTY SECTION This article was submitted to Bone Research, a section of the journal Frontiers in Endocrinology RECEIVED 30 January 2023 ACCEPTED 09 February 2023 PUBLISHED 23 February 2023 CITATION Yuan J, Xiong X, Zhang B, Feng Q, Zhang J, Wang W and Tang J (2023) Corrigendum: Genetically predicted C-reactive protein mediates the association between rheumatoid arthritis and atlantoaxial subluxation. Front. Endocrinol. 14:1154170. doi: 10.3389/fendo.2023.1154170 APPROVED BY Frontiers Editorial Ofﬁce, Frontiers Media SA, Switzerland CORRESPONDENCE Jia Tang tj676533@163.com Wenting Wang wangwentingtn@sina.com †These authors have contributed equally to this work Jiaqin Yuan 1,2†, Xiaoqin Xiong 3†, Bin Zhang 4†, Qingyuan Feng 5, Jinglin Zhang 6, Wenting Wang 7* and Jia Tang 8* Jiaqin Yuan 1,2†, Xiaoqin Xiong 3†, Bin Zhang 4†, Qingyuan Feng 5, Jinglin Zhang 6, Wenting Wang 7* and Jia Tang 8* 1Department of Orthopedics, The Second People’s Hospital of Yibin, Yibin, Sichuan, China, 2Department of Orthopedics, Yibin Hospital, West China Hospital of Sichuan University, Yibin, Sichuan, China, 3Department of Pediatrics, The Afﬁliated Hospital of Southwest Medical University, Luzhou, China, 4Rheumatism Immunity Branch, Weifang People’s Hospital, Weifang, Shandong, China, 5Wuxi School of Medicine, Jiangnan University, Wuxi, Jiangsu, China, 6Department of Occupational Disease, Yibin Center for Disease Control and Prevention, Yibin, Sichuan, China, 7Department of Anesthesiology, The Second Afﬁliated Hospital of Hainan Medical University, Haikou, China, 8Department of Pediatrics, Daping Hospital, Army Medical University, Chongqing, China CITATION Yuan J, Xiong X, Zhang B, Feng Q, Zhang J, Wang W and Tang J (2023) Corrigendum: Genetically predicted C-reactive protein mediates the association between rheumatoid arthritis and atlantoaxial subluxation. Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Frontiers in Endocrinology 01 frontiersin.org
https://openalex.org/W3014293451	https://europepmc.org/articles/pmc7240478?pdf=render	English	null	SiO2 NPs-PQ/PMMA Photopolymer Material Doped with a High-Concentration Photosensitizer for Holographic Storage	Polymers	2,020	cc-by	7,231	Received: 5 March 2020; Accepted: 1 April 2020; Published: 4 April 2020 Abstract: Dispersing nanoparticles and increasing the photosensitizer concentration have been regarded as eﬀective approaches for improving the performance of a holographic storage material. In this paper, SiO2 nanoparticle (NP)-dispersed PQ/PMMA nanocomposite material (SiO2 NP-PQ/PMMA) with a high PQ doping concentration was prepared. By introducing the co-monomer methyl isobutyl ketone (MIBK) that comes from an SiO2 NP colloidal solution, the concentration of PQ in the system increased to 1.2 wt %. We investigated the performance of polarization holographic recordings in both traditional PQ/PMMA and nanocomposite material SiO2 NP-PQ/PMMA with the orthogonally polarized signal and reference waves. With the dispersion of the SiO2 NPs colloidal solution and the increase in the PQ concentration, diﬀraction eﬃciency and photoinduced birefringence were multiplied. In addition, high-quality holographic image reconstruction was achieved by our homemade material. Keywords: nanocomposite photopolymer material; holographic data storage; polarization holography; holographic storage material Polymers 2020, 12, 816; doi:10.3390/polym12040816 www.mdpi.com/journal/polymers Ying Liu 1,2,* , Fenglan Fan 3 and Xiaodi Tan 4,* Ying Liu 1,2,* , Fenglan Fan 3 and Xiaodi Tan 4,* 1 Research Institute for Frontier Science, Beihang University, No. 37th, Xueyuan Road, Haidian District, Beijing 100091, China g j g 3 Department of Chemistry and Chemical Engineering, Hebei Normal Universitry for Nationalities, Chengde 067000, Hebei, China; ﬄ619@163.com g 4 Fujian Provincial Key Laboratory of Photonics Technology, College of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350007, Fujian, China 4 Fujian Provincial Key Laboratory of Photonics Technology, College of Photonic and Electronic Engineering Fujian Normal University, Fuzhou 350007, Fujian, China * Correspondence: liu_ying@buaa.edu.cn (Y.L.); xtan@fjnu.edu.cn (X.T.); Tel.: +86-1326-971-1843 (Y.L.) * Correspondence: liu_ying@buaa.edu.cn (Y.L.); xtan@fjnu.edu.cn (X.T.); Tel.: +86-1326-971-1843 (Y.L.) 1. Introduction Holographic storage material is a key factor restricting the development of holographic storage technology. The storage capacity, data transmission rate, storage life, and stability of a holographic storage device are closely related to the performance of the recording material [1]. PQ (phenanthrenequinone) is sensitive to light and often used as a holographic storage material photosensitizer. PQ/PMMA (phenanthrenequinone-doped poly(methyl methacrylate)) photopolymer has polarization sensitivity, controllable thickness, good optical quality, and low shrinkage (contraction coeﬃcient < 10−5), all of which make it suitable as a holographic storage material for multidimensional storage [2–4]. However, due to the low PQ saturation concentration dissolved in MMA and to the poor photosensitivity associated with a limited quantum of PQ molecular reactions [5], PQ/PMMA is unsatisfactory regarding diﬀraction eﬃciency and polarization sensitivity. y g g y p y In general, there are two main ways to increase the number of reactions of PQ molecules. One is to increase the amount of PQ molecules reacted by adding nano-components to introduce the interdiﬀusion of nanoparticles and PQ molecules. In 2001, Vaia R. A. et al. [6] ﬁrst incorporated gold www.mdpi.com/journal/polymers 2 of 10 of 10 Polymers 2020, 12, 816 Polymers 2020, 12, x FOR nanoparticles into photopolymers. Suzuki N. et al. [7–9] incorporated a series of non-metal oxides (SiO2, TiO2, and ZrO2) into the methacrylate organic photopolymer system. It was found that, by adjusting the blending ratio, the refractive index modulation of the material could be improved, and the incorporation of nanoparticles helps to improve the stability of the material and reduce the shrinkage rate [10–13]. The research of the doping system has unique research value. p Q , [ ] p g nanoparticles into photopolymers. Suzuki N. et al. [7–9] incorporated a series of non-metal oxides (SiO2, TiO2, and ZrO2) into the methacrylate organic photopolymer system. It was found that, by adjusting the blending ratio, the refractive index modulation of the material could be improved, and the incorporation of nanoparticles helps to improve the stability of the material and reduce the shrinkage rate [10–13]. The research of the doping system has unique research value. The other is to directly increase the concentration of the photosensitizer. Since the solubility of PQ in an MMA solution at room temperature is only 0.7 wt %, H. Liu et al. [14] reported a method for increasing the PQ concentration by increasing the prepolymerization temperature. 1. Introduction The results showed that about 1.0 wt % PQ was dissolved in the MMA solution at a reaction temperature of 60 ◦C. Mahilny et al. [15] demonstrated that the concentration of PQ can be increased to 4 mol % by casting the liquid solution directly onto the substrate and drying it to a solid state, but the method is only suitable for relatively low thickness (50–180 µm). Recently, Fenglan Fan et al. [16,17] proposed chemically modifying the material components to prepare a photopolymer material with PQ-loaded co-monomer, which improved the solubility of the photoinitiator in the photopolymer and thus promoted the optical properties of the material. g [ ] p g y q The other is to directly increase the concentration of the photosensitizer. Since the solubility of PQ in an MMA solution at room temperature is only 0.7 wt %, H. Liu et al. [14] reported a method for increasing the PQ concentration by increasing the prepolymerization temperature. The results showed that about 1.0 wt % PQ was dissolved in the MMA solution at a reaction temperature of 60 °C. Mahilny et al. [15] demonstrated that the concentration of PQ can be increased to 4 mol % by casting the liquid solution directly onto the substrate and drying it to a solid state, but the method is only suitable for relatively low thickness (50–180 μm). Recently, Fenglan Fan et al. [16,17] proposed chemically modifying the material components to prepare a photopolymer material with PQ-loaded co-monomer, which improved the solubility of the photoinitiator in the photopolymer and thus promoted the optical properties of the material. In this paper, a kind of SiO2 nanoparticle-dispersed PQ/PMMA composite holographic storage material (SiO2 NP-PQ/PMMA) containing a high photosensitizer PQ concentration was prepared. Photosensitizer PQ has certain solubility in an MIBK solution (a protective solution of doped SiO2 nanoparticles), which is considered a photopolymer co-monomer. The addition of an MIBK solution with a certain solubility to photosensitizer PQ increases the PQ concentration (to 1.2 wt %) in photopolymer materials. The inﬂuence of the newly introduced solution on the reaction of the system and the improvement of the performance of the holographic storage material were analyzed. In addition, a series of experiments were carried out on the prepared SiO2 NP-PQ/PMMA nanocomposites for the material reaction and holographic diﬀraction characteristics. 1. Introduction p p p p In this paper, a kind of SiO2 nanoparticle-dispersed PQ/PMMA composite holographic storage material (SiO2 NP-PQ/PMMA) containing a high photosensitizer PQ concentration was prepared. Photosensitizer PQ has certain solubility in an MIBK solution (a protective solution of doped SiO2 nanoparticles), which is considered a photopolymer co-monomer. The addition of an MIBK solution with a certain solubility to photosensitizer PQ increases the PQ concentration (to 1.2 wt %) in photopolymer materials. The influence of the newly introduced solution on the reaction of the system and the improvement of the performance of the holographic storage material were analyzed. In addition, a series of experiments were carried out on the prepared SiO2 NP-PQ/PMMA nanocomposites for the material reaction and holographic diffraction characteristics. 2. Material Preparation 2. Material Preparation Concentration ratio of each component in the prepared sample. Since the viscosity of MIBK dopants is higher than that of the MMA solution, excessive SiO2 NP colloidal solution will aﬀect the copolymerization, leading to many small bubbles inside the material. Photopolymer samples with an SiO2 NP doping concentration of 3 wt % were chosen. The sample battle was ultrasonically shaken in an ultrasonic cleaner to form a uniform multi-component solution. Subsequently, the homogeneously mixed solution was placed in a magnetically stirred, constant-temperature water bath. The temperature was continuously raised to 60 ◦C to start the prepolymerization process and was stabilized at 60 ◦C for an appropriate period until the homogeneous solution became viscous (glycerol viscosity). The viscous solution was poured into a speciﬁc glass mold. The mold was placed in a blast oven at 60 ◦C for about 40 h, until the material was completely cured. Finally, the mold was removed and placed in a refrigerator for 2 h. The prepared SiO2 NP-PQ/PMMA material was bulk with millimeter-level dimensions. It had good optical transparency, and the color was light yellow. As the concentration of photosensitizer and material thickness increased, the sample color deepened. 2. Material Preparation 2. Material Preparation Figure 1 shows chemical structures of the major components in the photopolymer sample. MMA and methyl isobutyl ketone (MIBK, from an SiO2 NP colloidal solution) were regarded as the co-monomers, PQ was used as the photosensitizer, and 2,2-azobisisobutyronitrile (AIBN) was employed as the thermo-initiator. The SiO2 NP colloidal solution contains 30% SiO2, 69.5% MIBK, and 0.5% CH4O. The size of each SiO2 NP is about 20 nm. In this paper, the SiO2 NP colloidal solution contained the nano-doped components. Figure 1 shows chemical structures of the major components in the photopolymer sample. MMA and methyl isobutyl ketone (MIBK, from an SiO2 NP colloidal solution) were regarded as the co- monomers, PQ was used as the photosensitizer, and 2,2-azobisisobutyronitrile (AIBN) was employed as the thermo-initiator. The SiO2 NP colloidal solution contains 30% SiO2, 69.5% MIBK, and 0.5% CH4O. The size of each SiO2 NP is about 20 nm. In this paper, the SiO2 NP colloidal solution contained the nano-doped components. Figure 1. Chemical structures of the major components in the photopolymer sample. Figure 1. Chemical structures of the major components in the photopolymer sample. Figure 1. Chemical structures of the major components in the photopolymer sample. Figure 1. Chemical structures of the major components in the photopolymer sample. In our fabricating process, since PQ molecules have solubility in both the MMA monomer and MIBK, an introduced SiO2 NP colloidal solution can increase the dissolved concentration of the PQ In our fabricating process, since PQ molecules have solubility in both the MMA monomer and MIBK, an introduced SiO2 NP colloidal solution can increase the dissolved concentration of the PQ 3 of 10 Polymers 2020, 12, 816 molecule. The monomer MMA and SiO2 NPs were mixed ﬁrstly with a weight ratio of 100:3. After weighing, the SiO2 NP colloidal solution, the monomer MMA, the photosensitizer PQ, and the thermal initiator AIBN were mixed in a clean reaction bottle at a certain mass ratio in a dark room at room temperature. The proportion of each component in the mass ratio of MMA is shown in Table 1 (SiO2 NPs/MIBK = 3:7). Table 1. Concentration ratio of each component in the prepared sample. Sample MMA (wt %) SiO2 (wt %) PQ (wt %) AIBN (wt %) 1 100 3 0.7 1 2 0 1 3 3 1 4 3 1.1 5 3 1.15 6 3 1.2 Table 1. 3 2 FT IR Spectra Measurements 3.2. FT-IR Spectra Measurements 3. . I Spectra Measurements Infrared spectroscopy can obtain photochemistry reaction information about the samples. We measured the FT-IR absorption spectrum of PQ and the monomers before and after photo-irradiation to investigate whether new photoproducts were formed by adding the SiO2 NP colloidal solution. Figure 3 shows the FT-IR spectra of unexposed and exposed PQ/SiO2 NP colloidal solution (bottom, middle of Figure 3) and exposed PQ/SiO2 NP colloidal solution/MMA samples (top of Figure 3), which were obtained with a Nicolet 6700 Infrared Spectrometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Since the main component of the doped SiO2 NP colloidal solution was MIBK, we considered MIBK as a co monomer Infrared spectroscopy can obtain photochemistry reaction information about the samples. We measured the FT-IR absorption spectrum of PQ and the monomers before and after photo-irradiation to investigate whether new photoproducts were formed by adding the SiO2 NP colloidal solution. Figure 3 shows the FT-IR spectra of unexposed and exposed PQ/SiO2 NP colloidal solution (bottom, middle of Figure 3) and exposed PQ/SiO2 NP colloidal solution/MMA samples (top of Figure 3), which were obtained with a Nicolet 6700 Infrared Spectrometer (Thermo Fisher Scientiﬁc Inc., Waltham, MA, USA). Since the main component of the doped SiO2 NP colloidal solution was MIBK, we considered MIBK as a co-monomer. Polymers 2020, 12, x FOR PEER REVIEW 5 of 10 we considered MIBK as a co monomer. The FT-IR spectra of the PQ/MIBK solution before and after exposure are mostly the same; n obvious new bond was formed. We suggest that there was no photopolymerization reaction betwee the MIBK solution and the PQ molecules. The main photoproduct was still formed by the reaction o MMA molecules and PQ molecules. Within the broad absorption of 2800–3000 cm−1, all samples ha a broad absorption, which was derived from the C–H bond stretching (CH3 and CH2 stretching). A strong absorption peak at about 1745 cm−1 resulted from a carbonyl group commonly contained i the MIBK and MMA units. The absorption band at about 1230 and 939cm−1 was mainly due to the C O–C group of the photoproduct formed by the reaction of the group in the PQ molecule and the viny group in the monomer molecule. These phenomena indicate that the added MIBK solution had n effect on the photochemical reaction of the photosensitizer PQ molecule or of the monomeric MMA molecule. Figure 3. 3.1. UV–Vis Spectra Measurements The optical absorption of the material had an important impact on the holographic performance. According to Beer’s law, the absorbance A is proportional to the product of concentration c of the light-absorbing substance and the optical path length b of the absorption cell (A = ε•c•b). For the UV–Vis spectroscopic technique, absorption can be indicated as A = −lg (IT/I0), in which I0 is the intensity of the incident light entering the substance and It is the intensity of the transmitted light emitted from the back substance. In order to determine the eﬀect of the introduced SiO2 NP colloidal solution and the high concentration PQ on the optical absorption of the prepared SiO2 NP-PQ/PMMA material, optical absorption was measured using a TU-1901 dual-beam UV–Vis spectrophotometer (PERSEE, Beijing, China) at 25 ◦C. The thickness of the sample was 1.5 mm. The plot of absorbance versus the wavelength is shown in Figure 2. The absorption trend of the SiO2 NP-PQ/PMMA sample shows no obvious change in the absorption curve of the conventional PQ/PMMA photopolymer material. The absorption in the short-wave direction was strong and substantially zero when λ > 600 nm, shown in Figure 2a. A red laser with a wavelength of 632.8 nm (which did not cause a change in the optical properties of the material) was used as the probe beam for the experiment with a refractive index modulation change caused by the photoinduced anisotropy of the material. As depicted in Figure 2b, the sample has a certain absorption at 532 nm, and by increasing PQ concentration, the absorption value of the material has little change at 532 nm. Thus, in the subsequent holographic recording experiments, a green laser of 532 nm was selected as the recording light to induce an anisotropic reaction in the sample material. 4 of 10 4 of 10 Polymers 2020, 12, 816 Polymers 2020, 12, x FO (a) (b) Figure 2. (a) UV–visible absorption spectrum of SiO2 NP-PQ/PMMA sample; (b) partially amplified UV–Vis absorption spectrum. Figure 2. (a) UV–visible absorption spectrum of SiO2 NP-PQ/PMMA sample; (b) partially ampliﬁed UV–Vis absorption spectrum. (b) (a) (b) (a) Figure 2. (a) UV–visible absorption spectrum of SiO2 NP-PQ/PMMA sample; (b) partially amplified UV–Vis absorption spectrum. Figure 2. (a) UV–visible absorption spectrum of SiO2 NP-PQ/PMMA sample; (b) partially ampliﬁed UV–Vis absorption spectrum. 3.3. Holographic Diﬀraction Characteristics Photosensitizer PQ molecules possess a high conjugate coplanar structure [18]. Before the illumination, PQ molecules with diﬀerent orientations were randomly distributed in the material. When the material is exposed by a linearly polarized wave, dependent on the polarization state, PQ molecules with a certain orientation have a greater chance of reacting with monomers on certain regions [5,19]. The orientation of the PQ molecules that was parallel to the illuminated wave polarization state had a greater chance of reacting. Double bonds in the carbonyl functional group of PQ molecules were excited by illuminated photons. PQ molecules became radicals. PQ radicals reacted with MMA monomers that had carbonic double bonds on the vinyl functional group. The photoproduct ﬁnally formed [20,21]. This photoreaction can cause the polarization distribution. More photosensitizer PQ molecules can participate in the reaction when more PQ molecules are doped within a certain range. Thus, the concentration of photosensitizer PQ plays an important role on polarization holographic recording. Furthermore, a mutual diﬀusion process occurred in the SiO2 NP-PQ/PMMA photopolymer sample, in which the dispersed NPs played a positive role in enhancing the diﬀraction eﬃciency formation [22]. Along with the consumption of PQ molecules and the formation of photoproducts, the embedded SiO2 NPs introduced a multicomponent diﬀusion process. As a result, the SiO2 NP composition gradually increased the amount of PQ molecules participating in the reaction, and the amount of photoproduct increased correspondingly. Therefore, the addition of the SiO2 NP colloidal solution improved the saturated diﬀraction eﬃciency of the material. Figure 4 shows the experimental setup of the holographic diﬀraction characteristic measurements. In the experiment, the collimated laser (532 nm), from a diode-pumped solid-state Nd:YAG laser, was split into a vertically polarized wave (s-pol., signal wave) and a horizontally polarized wave (p-pol., reference wave) by a polarization beam splitter (PBS). We recorded diﬀraction gratings by two orthogonal linearly polarized waves. They were incident to the material symmetrically and received by two photodetectors. In the recording stage, the volume polarization hologram was written at a cross-angle of 30◦, and each beam intensity was 22 mW. The shutters were used to control the holographic recording time. The recording process and the reconstruction process were separated after each period of exposure. In the recording process, Shutters 1 and 2 were opened for 4 s, while Shutter 3 was closed. In the reconstruction process, Shutter 2 closed and Shutter 3 opened. 3 2 FT IR Spectra Measurements 3.2. FT-IR Spectra Measurements FT-IR absorption spectra of unexposed PQ/MIBK, exposed PQ/MIBK, and exposed PQ/MIBK/MMA. Figure 3. FT-IR absorption spectra of unexposed PQ/MIBK, exposed PQ/MIBK, and exposed PQ/MIBK/MMA. Figure 3. FT-IR absorption spectra of unexposed PQ/MIBK, exposed PQ/MIBK, and exposed PQ/MIBK/MMA. Figure 3. FT-IR absorption spectra of unexposed PQ/MIBK, exposed PQ/MIBK, and exposed PQ/MIBK/MMA. 3.3. Holographic Diffraction Characteristics Photosensitizer PQ molecules possess a high conjugate coplanar structure [18]. Before the illumination, PQ molecules with different orientations were randomly distributed in the material. When the material is exposed by a linearly polarized wave, dependent on the polarization state, PQ molecules with a certain orientation have a greater chance of reacting with monomers on certain The FT-IR spectra of the PQ/MIBK solution before and after exposure are mostly the same; no obvious new bond was formed. We suggest that there was no photopolymerization reaction between the MIBK solution and the PQ molecules. The main photoproduct was still formed by the reaction of MMA molecules and PQ molecules. Within the broad absorption of 2800–3000 cm−1, all samples had a broad absorption, which was derived from the C–H bond stretching (CH3 and CH2 stretching). 5 of 10 Polymers 2020, 12, 816 A strong absorption peak at about 1745 cm−1 resulted from a carbonyl group commonly contained in the MIBK and MMA units. The absorption band at about 1230 and 939cm−1 was mainly due to the C–O–C group of the photoproduct formed by the reaction of the group in the PQ molecule and the vinyl group in the monomer molecule. These phenomena indicate that the added MIBK solution had no eﬀect on the photochemical reaction of the photosensitizer PQ molecule or of the monomeric MMA molecule. A strong absorption peak at about 1745 cm−1 resulted from a carbonyl group commonly contained in the MIBK and MMA units. The absorption band at about 1230 and 939cm−1 was mainly due to the C–O–C group of the photoproduct formed by the reaction of the group in the PQ molecule and the vinyl group in the monomer molecule. These phenomena indicate that the added MIBK solution had no eﬀect on the photochemical reaction of the photosensitizer PQ molecule or of the monomeric MMA molecule. 3.3. Holographic Diﬀraction Characteristics The original reference wave irradiated the material to retrieve the grating for 0.4 s. At this moment, we obtained the corresponding diﬀraction signal after the 4 s recording. The recording process and the reconstruction process were carried out on a continuous loop until the power of reconstructed wave was saturated. Figure 5 shows the temporal evolution of the orthogonal linearly-grating diﬀraction eﬃciency for the SiO2 NP-PQ/PMMA samples containing diﬀerent photosensitizer concentrations. The diﬀraction eﬃciency η is deﬁned as I+1/ (I0 + I+1) in this paper, and I0 and I+1 are the intensities of the transmitted and the 1st-order diﬀracted wave, respectively. In Figure 6, PQ1.0, PQ1/SiO2, PQ1.1/SiO2, PQ1.15/SiO2, and PQ1.2/SiO2 indicate Sample 2, Sample 3, Sample 4, Sample 5, and Sample 6, respectively. The saturated diﬀraction eﬃciency of the recorded grating was improved compared with the undoped sample (PQ1.0). With the increase in PQ concentration, the saturation diﬀraction eﬃciency increased correspondingly. The saturation diﬀraction eﬃciency was increased to nearly 4%. The diﬀraction eﬃciency of PQ1.0 increased faster than that of PQ1/SiO2, at low exposure times. This is because the presence of MIBK monomers caused a decrease of MMA monomers in the same region, while PQ 6 of 10 Polymers 2020, 12, 816 monomers did not increase. Hence, compared with PQ1.0, fewer PQ molecules were involved in the reaction for PQ1/SiO2 at low exposure times. Subsequently, due to the continuous consumption of PQ, the eﬀect of dynamic redistribution from NPs was obvious. More PQ molecules were involved in the reaction, which caused the diﬀraction eﬃciency to increase. y , , irradiated the material to retrieve the grating for 0.4 s. At this moment, we obtained the corresponding diffraction signal after the 4 s recording. The recording process and the reconstruction process were carried out on a continuous loop until the power of reconstructed wave was saturated. Figure 4. Experimental setup for holographic diffraction efficiency measurement, where HWP: half- wave plate; PBS: polarization beam splitter; PD: photo detector; M: mirror; s-pol.: vertical polarization light; p-pol.: horizontally polarized light. Figure 4. Experimental setup for holographic diﬀraction eﬃciency measurement, where HWP: half-wave plate; PBS: polarization beam splitter; PD: photo detector; M: mirror; s-pol.: vertical polarization light; p-pol.: horizontally polarized light. Figure 4. Experimental setup for holographic diffraction efficiency measurement, where HWP: half- wave plate; PBS: polarization beam splitter; PD: photo detector; M: mirror; s-pol.: vertical polarization light; p-pol.: horizontally polarized light. 3.3. Holographic Diﬀraction Characteristics Figure 5 shows the temporal evolution of the orthogonal linearly-grating diffraction efficien or the SiO2 NP-PQ/PMMA samples containing different photosensitizer concentrations. T iffraction efficiency η is defined as I+1/ (I0 + I+1) in this paper, and I0 and I+1 are the intensities of ansmitted and the 1st-order diffracted wave, respectively. In Figure 6, PQ1.0, PQ1/SiO2, PQ1.1/Si Q1.15/SiO2, and PQ1.2/SiO2 indicate Sample 2, Sample 3, Sample 4, Sample 5, and Sample espectively. The saturated diffraction efficiency of the recorded grating was improved compar with the undoped sample (PQ1.0). With the increase in PQ concentration, the saturation diffract fficiency increased correspondingly. The saturation diffraction efficiency was increased to nea %. The diffraction efficiency of PQ1.0 increased faster than that of PQ1/SiO2, at low exposure tim his is because the presence of MIBK monomers caused a decrease of MMA monomers in the sa egion, while PQ monomers did not increase. Hence, compared with PQ1.0, fewer PQ molecules w nvolved in the reaction for PQ1/SiO2 at low exposure times. Subsequently, due to the continuo onsumption of PQ, the effect of dynamic redistribution from NPs was obvious. More PQ molecu were involved in the reaction, which caused the diffraction efficiency to increase. PQ1.0). W Figure 4. Experimental setup for holographic diffraction efficiency measurement, where HWP: half- wave plate; PBS: polarization beam splitter; PD: photo detector; M: mirror; s-pol.: vertical polarization light; p-pol.: horizontally polarized light. Figure 4. Experimental setup for holographic diﬀraction eﬃciency measurement, where HWP: half-wave plate; PBS: polarization beam splitter; PD: photo detector; M: mirror; s-pol.: vertical polarization light; p-pol.: horizontally polarized light. olved in the reaction for PQ1/SiO2 at low exposure times. Subsequently, due to the continu sumption of PQ, the effect of dynamic redistribution from NPs was obvious. More PQ molecu e involved in the reaction, which caused the diffraction efficiency to increase. Figure 5 shows the temporal evolution of the orthogonal linearly-grating diffraction efficie the SiO2 NP-PQ/PMMA samples containing different photosensitizer concentrations. T raction efficiency η is defined as I+1/ (I0 + I+1) in this paper, and I0 and I+1 are the intensities of nsmitted and the 1st-order diffracted wave, respectively. In Figure 6, PQ1.0, PQ1/SiO2, PQ1.1/S 1.15/SiO2, and PQ1.2/SiO2 indicate Sample 2, Sample 3, Sample 4, Sample 5, and Sample pectively. The saturated diffraction efficiency of the recorded grating was improved compa h the undoped sample (PQ1.0). With the increase in PQ concentration, the saturation diffract ciency increased correspondingly. 3.3. Holographic Diﬀraction Characteristics The saturation diffraction efficiency was increased to nea The diffraction efficiency of PQ1.0 increased faster than that of PQ1/SiO2, at low exposure tim s is because the presence of MIBK monomers caused a decrease of MMA monomers in the sa ion, while PQ monomers did not increase. Hence, compared with PQ1.0, fewer PQ molecules w Figure 5. The dependence of the orthogonal linearly-grating diffraction efficiency on exposure time. Figure 5. The dependence of the orthogonal linearly-grating diﬀraction eﬃciency on exposure time. s is because the presence of MIBK monomers caused a decrease of MMA monomers in the s ion, while PQ monomers did not increase. Hence, compared with PQ1.0, fewer PQ molecules w Figure 5. The dependence of the orthogonal linearly-grating diffraction efficiency on exposure time. Figure 5. The dependence of the orthogonal linearly-grating diﬀraction eﬃciency on exposure time. involved in the reaction for PQ1/SiO2 at low exposure times. Subsequently, due to the continuous consumption of PQ, the effect of dynamic redistribution from NPs was obvious. More PQ molecules were involved in the reaction, which caused the diffraction efficiency to increase. From the above experimental analysis, it can be deemed that the concentration of photosensitizer is important for photopolymer materials and that nanoparticle doping can improve the performance of the material. It is feasible to use a nanoparticle protection solution to increase the concentration of photosensitizer and form a two-monomer composite system to improve the holographic properties of the material. The concentration of the photosensitizer PQ in the PQ/PMMA photopolymer system increases based on the introduction of favorable nano-components, and the material is further optimized. 3.4. Photoinduced Birefringence The sample developed into anisotropy by the oriented photoproduct molecules in the sample. , p y We explored the photoinduced birefringence of the photopolymer sample pumped by a diode pumped solid state (DPSS) Nd:YAG laser (λ = 532 nm). The experimental setup is schematically shown in Figure 6. The photoinduced birefringence was investigated with a continuous He-Ne laser (λ = 632.8 nm) as the probe light, which is far from the photopolymer absorption band. In the experimental detection, the intensity of the pumping wave incident on the sample surface was 20mW controlled by the attenuator, and the diameter of the spot was 6 mm; the power of the probe light incident on the material surface after passing through Polarizer 1 was 0.8 mW, and the diameter of the spot was 2 mm. The cross-angle θ1 between the two beams was set to 6°. Firstly, the photopolymer was illuminated by a separate probe wave, and no transmitted wave was received. The simple was isotropic. After the linearly-polarized pumping wave was turned on, the intensity of the transmitted probe wave gradually increased. The sample developed into anisotropy by the oriented photoproduct molecules in the sample. Figure 6. Diagram of photoinduced birefringence measurement device: θ = 6°, P0, P1, and P2 are horizontal polarizers, negative 45° polarizers and positive 45° polarizers, respectively. Figure 6. Diagram of photoinduced birefringence measurement device: θ = 6◦, P0, P1, and P2 are horizontal polarizers, negative 45◦polarizers and positive 45◦polarizers, respectively. Figure 6. Diagram of photoinduced birefringence measurement device: θ = 6°, P0, P1, and P2 are horizontal polarizers, negative 45° polarizers and positive 45° polarizers, respectively. Figure 6. Diagram of photoinduced birefringence measurement device: θ = 6◦, P0, P1, and P2 are horizontal polarizers, negative 45◦polarizers and positive 45◦polarizers, respectively. Figure 7 shows the temporal evolution of the sample photoinduced birefringence, in which PQ0.7/SiO2 indicates Sample 1. At the initial stage of pumping exposure, the photoinduced birefringence of the sample increased rapidly with the cumulative exposure energy, and PQ molecules formed double-substituted phenanthrene photoproducts. The photoinduced birefringence slowly changed after a certain period of time, when most of the PQ molecules, whose directions are the same or similar to that of the polarized light field, have absorbed the energy and when photolysis has occurred. 3.4. Photoinduced Birefringence Photoinduced birefringence ∆n is one of the key parameters of polarization holographic material. For a PQ/PMMA material system, photoinduced birefringence is mainly caused by the structural rearrangements induced by the photochemical reaction of PQ molecules [18,23]. The photoinduced birefringence is given by (ignore the absorption) [24]: Polymers 2020, 12, 816 structural rearrang photoinduced bire 7 of 10 3]. The ∆n = n1 −n2 = λ πdarcsin r IT I0sin22θ0 (1) T 1 2 2 0 0 I λ Δn = n - n = arcsin πd I sin 2θ (1) (1) (1) where n1, n2, d, λ, I0, IT, and θ0 are the refractive index along and perpendicular to the polarization direction of the pump light, the thickness of the photopolymer sample, the wavelength of the pumping laser, the intensity of the probe wave before pumping exposure, the intensity of the probe wave after pumping exposure, and the angle between the polarization direction of the green laser and that of the red laser, respectively. where n1, n2, d, λ, I0, It, and θ0 are the refractive index along and perpendicular to the polarization direction of the pump light, the thickness of the photopolymer sample, the wavelength of the pumping laser, the intensity of the probe wave before pumping exposure, the intensity of the probe wave after pumping exposure, and the angle between the polarization direction of the green laser and that of the red laser, respectively. We explored the photoinduced birefringence of the photopolymer sample pumped by a diode pumped solid state (DPSS) Nd:YAG laser (λ = 532 nm). The experimental setup is schematically shown in Figure 6. The photoinduced birefringence was investigated with a continuous He-Ne laser (λ = 632.8 nm) as the probe light, which is far from the photopolymer absorption band. In the experimental detection, the intensity of the pumping wave incident on the sample surface was 20mW controlled by the attenuator, and the diameter of the spot was 6 mm; the power of the probe light incident on the material surface after passing through Polarizer 1 was 0.8 mW, and the diameter of the spot was 2 mm. The cross-angle θ1 between the two beams was set to 6◦. Firstly, the photopolymer was illuminated by a separate probe wave, and no transmitted wave was received. The simple was isotropic. After the linearly-polarized pumping wave was turned on, the intensity of the transmitted probe wave gradually increased. 3.5. Application Experiments 3.5. Application Experiments 3.5. Application Experiments We recorded and reconstructed a real image on the SiO2 NP-PQ/PMMA photopolymers by our polarization holographic recording system [17] to better exhibit the material application of the polarization holography. In the experiment, the signal wave and reference wave of polarization holographic recording were s-polarized and p-polarized, respectively. The image was recorded in a sample (MMA/SiO2 NPs/PQ = 100:3:1.2) approximately 10 × 10 × 15 mm. The original image for polarization holographic recording was uploaded onto spatial light modulators (SLM). Figure 8a shows the original transmitted image that was directly transmitted and transformed to a digital signal by a picture receiver (CMOS). Figure 8b shows the reconstructed image that was recorded in the SiO2 NP-PQ/PMMA sample by a polarization holographic recording. Both images were 300 × 300 pixels. The reconstructed image was reconstructed faithfully and had a clear contrast and high fidelity. The information polarization recording competence of the SiO2 NP-PQ/PMMA material was shown. The feasibility and potential of our material for recording polarization multiplexed holograms was thus demonstrated. We recorded and reconstructed a real image on the SiO2 NP-PQ/PMMA photopolymers by our polarization holographic recording system [17] to better exhibit the material application of the polarization holography. In the experiment, the signal wave and reference wave of polarization holographic recording were s-polarized and p-polarized, respectively. The image was recorded in a sample (MMA/SiO2 NPs/PQ = 100:3:1.2) approximately 10 × 10 × 15 mm. The original image for polarization holographic recording was uploaded onto spatial light modulators (SLM). Figure 8a shows the original transmitted image that was directly transmitted and transformed to a digital signal by a picture receiver (CMOS). Figure 8b shows the reconstructed image that was recorded in the SiO2 NP-PQ/PMMA sample by a polarization holographic recording. Both images were 300 × 300 pixels. The reconstructed image was reconstructed faithfully and had a clear contrast and high ﬁdelity. The information polarization recording competence of the SiO2 NP-PQ/PMMA material was shown. The feasibility and potential of our material for recording polarization multiplexed holograms was thus demonstrated. We recorded and reconstructed a real image on the SiO2 NP-PQ/PMMA photopolymers by our polarization holographic recording system [17] to better exhibit the material application of the polarization holography. In the experiment, the signal wave and reference wave of polarization holographic recording were s-polarized and p-polarized, respectively. 3.5. Application Experiments 3.5. Application Experiments 3.5. Application Experiments The image was recorded in a sample (MMA/SiO2 NPs/PQ = 100:3:1.2) approximately 10 × 10 × 15 mm. The original image for polarization holographic recording was uploaded onto spatial light modulators (SLM). Figure 8a shows the original transmitted image that was directly transmitted and transformed to a digital signal by a picture receiver (CMOS). Figure 8b shows the reconstructed image that was recorded in the SiO2 NP-PQ/PMMA sample by a polarization holographic recording. Both images were 300 × 300 pixels. The reconstructed image was reconstructed faithfully and had a clear contrast and high fidelity. The information polarization recording competence of the SiO2 NP-PQ/PMMA material was shown. The feasibility and potential of our material for recording polarization multiplexed holograms was thus demonstrated. (a) (b) Figure 8. Image reconstruction results in polarization holography: (a) original transmitted image; (b) reconstructed image. (a) (b) Figure 8. Image reconstruction results in polarization holography: (a) original transmitted image; (b) reconstructed image. Figure 8. Image reconstruction results in polarization holography: (a) original transmitted image; (b) reconstructed image. (a) (a) (b) (b) (b) (b) (a) (a) Figure 8. Image reconstruction results in polarization holography: (a) original transmitted image; (b) reconstructed image Figure 8. Image reconstruction results in polarization holography: (a) original transmitted image; (b) reconstructed image. Figure 8. Image reconstruction results in polarization holography: (a) original transmitted image; (b) reconstructed image. 3.4. Photoinduced Birefringence With the same composition of dispersed SiO2 NPs in the sample, accompanied by an Figure 7 shows the temporal evolution of the sample photoinduced birefringence, in which PQ0.7/SiO2 indicates Sample 1. At the initial stage of pumping exposure, the photoinduced birefringence of the sample increased rapidly with the cumulative exposure energy, and PQ molecules formed double-substituted phenanthrene photoproducts. The photoinduced birefringence slowly changed after a certain period of time, when most of the PQ molecules, whose directions are the same or similar to that of the polarized light ﬁeld, have absorbed the energy and when photolysis has occurred. With the same composition of dispersed SiO2 NPs in the sample, accompanied by an increased PQ molecule concentration, the material achieved a high ∆n value. In addition, the doping of nano-components also enhanced the photoinduced birefringence of the material. When doped with nanoparticles, the photoinduced birefringence curve of the sample with a 0.7 wt % PQ concentration was similar to that of the sample with a 1.0 wt % PQ concentration without nanoparticles doping. 8 of 10 without without Polymers 2020, 12, 816 concentration was nanoparticles dopi concentration was nanoparticles dopi Figure 7. Temporal evolution of photoinduced birefringence. Figure 7. Temporal evolution of photoinduced birefringence. Figure 7. Temporal evolution of photoinduced birefringence. Figure 7. Temporal evolution of photoinduced birefringence. Figure 7. Temporal evolution of photoinduced birefringence. Figure 7. Temporal evolution of photoinduced birefringence. References 1. Murciano, A.; Carretero, L.; Blaya, S.; Madrigal, R.F.; Fimia, A. Experimental Study of Multiplexed Holographic Gratings Recorded in a Photopolymerizable Silica Glass. Appl. Phys. B 2006, 83, 619–622. [CrossRef] 2. Tolstik, E.; Romanov, O.; Matusevich, V.; Tolstik, A.; Kowarschik, R. Formation of self-trapping waveguides in bulk PMMA media doped with Phenanthrenequinone. Opt. Express 2014, 22, 3228–3233. [CrossRef] [PubMed] 3. Manukhin, B.G.; Chivilikhin, S.A.; Schelkanova, I.J.; Andreeva, N.V.; Materikina, D.A.; Andreeva, O.V. Reversible and irreversible alterations of the optical thickness of PQ/PMMA volume recording media samples. Part I: Experiment. Appl. Opt. 2017, 56, 7351–7357. [CrossRef] [PubMed] 3. Manukhin, B.G.; Chivilikhin, S.A.; Schelkanova, I.J.; Andreeva, N.V.; Materikina, D.A.; Andreeva, O.V. Reversible and irreversible alterations of the optical thickness of PQ/PMMA volume recording media samples. Part I: Experiment. Appl. Opt. 2017, 56, 7351–7357. [CrossRef] [PubMed] p pp p 4. Liu, P.; Chang, F.; Zhao, Y.; Li, Z.; Sun, X. Ultrafast volume holographic storage on PQ/PMMA photopolymers with nanosecond pulsed exposures. Opt. Express 2018, 26, 1072–1082. [CrossRef] 4. Liu, P.; Chang, F.; Zhao, Y.; Li, Z.; Sun, X. Ultrafast volume holographic storage on PQ/PMMA photopolymers with nanosecond pulsed exposures. Opt. Express 2018, 26, 1072–1082. [CrossRef] 5. Mahilny, U.; Troﬁmova, A.; Nazarov, S.; Tolstik, A.; Heintzmann, R.; Tolstik, E. Highly concentrated phenanthrenequinone-polymethylmethacrylate composite for thick reﬂection holograms recording at 532 nm. Opt. Mater. Express 2016, 6, 3427–3437. [CrossRef] 5. Mahilny, U.; Troﬁmova, A.; Nazarov, S.; Tolstik, A.; Heintzmann, R.; Tolstik, E. Highly concentrated phenanthrenequinone-polymethylmethacrylate composite for thick reﬂection holograms recording at 532 nm. Opt. Mater. Express 2016, 6, 3427–3437. [CrossRef] 6. Vaia, R.A.; Dennis, C.L.; Natarajan, L.V.; Tondiglia, V.P.; Tomlin, D.W.; Bunning, T.J. One step, Micrometer-scale organization of nano- and mesoparticles using holographic photopolymerization: A generic technique. Adv. Mater. 2001, 13, 1570–1574. [CrossRef] 7. Suzuki, N.; Tomita, Y.; Kojima, T. Holographic recording in TiO2 nanoparticle-dispersed methacrylate photopolymer ﬁlms. Appl Phys Lett. 2002, 81, 4121–4123. [CrossRef] 8. Suzuki, N.; Tomita, Y.; Ohmori, K.; Hidaka, M.; Chikama, K. Highly transparent ZrO2 nanoparticle-dispersed acrylate photopolymers for volume holographic recording. Opt. Express 2006, 14, 12712–12719. [CrossRef] 8. Suzuki, N.; Tomita, Y.; Ohmori, K.; Hidaka, M.; Chikama, K. Highly transparent ZrO2 nanoparticle-dispersed acrylate photopolymers for volume holographic recording. Opt. Express 2006, 14, 12712–12719. [CrossRef] 9. Tomita, Y.; Nishibiraki, H. Improvement of holographic recording sensitivities in the green in SiO2 9. Tomita, Y.; Nishibiraki, H. 4. Conclusions In this work, we provided an eﬀective pathway to overcoming the PQ solubility limitation and to simultaneously introducing an NP modulation of PQ/PMMA. The introduction of an SiO2 NP colloidal solution increased the concentration of PQ to 1.2 wt %. The newly introduced solution caused no reaction in the system. Compared with the undoped PQ/PMMA material, the saturated diﬀraction 9 of 10 Polymers 2020, 12, 816 eﬃciency of the grating formed by SiO2 NP-PQ/PMMA nanocomposites under a polarized hologram recorded by orthogonal linear polarized light increased from 0.6% to nearly 4%. It is thus shown that controlling the doping of nanoparticles and the concentration of the photosensitizer can improve the performance of holographic photopolymer storage material. Therefore, a solution with higher compatibility can be selected as a composite component of the system to improve the performance of holographic photopolymer storage materials. Author Contributions: Conceptualization, Y.L.; methodology, Y.L. and F.F.; software, Y.L. and F.F.; validation, Y.L. and F.F.; formal analysis, Y.L. and F.F.; investigation, Y.L. and F.F.; resources, X.T.; data curation, Y.L.; writing—original draft preparation, Y.L.; writing—review and editing, Y.L. and X.T.; visualization, Y.L. and F.F.; supervision, X.T.; project administration, Y.L. and X.T.; funding acquisition, Y.L. and X.T. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Major Scientiﬁc Research Project of Zhejiang Lab, grant number 2019MB0AE03, the National Key R&D Program of China, grant number 2018YFA0701800, and the National Natural Science Foundation of China, grant number 61703025. Conﬂicts of Interest: The authors declare no conﬂict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. References Improvement of holographic recording sensitivities in the green in SiO2 nanoparticle-dispersed methacrylate photopolymers doped with pyrromethene dyes. Appl. Phys. Lett. 2003, 83, 410–412. [CrossRef] 10. Li, C.; Cao, L.; He, Q.; Jin, G. Holographic Kinetics for Mixed Volume Gratings in Gold Nanoparticles Doped Photopolymer. Opt. Express 2014, 22, 5017–5028. [CrossRef] 11. Hata, E.; Mitsube, K.; Momose, K.; Tomita, Y. Holographic nanoparticle-polymer composites based on step-growth thiol-ene photopolymerization. Opt. Mater. Express 2011, 1, 207–222. [CrossRef] 12. Tomita, Y.; Urano, H.; Fukamizu, T.; Kametani, Y.; Nishimura, N.; Odoi, K. Nanoparticle-polymer composite volume holographic gratings dispersed with ultrahigh-refractive-index hyperbranched polymer as organic nanoparticles. Opt. Lett. 2016, 41, 1281–1284. [CrossRef] [PubMed] p p 13. Liu, P.; Zhao, Y.; Li, Z.; Sun, X. Improvement of ultrafast holographic performance in silver nanoprisms dispersed photopolymer. Opt. Express 2018, 26, 6993–7004. [CrossRef] [PubMed] p p 13. Liu, P.; Zhao, Y.; Li, Z.; Sun, X. Improvement of ultrafast holographic performance in silver nanoprisms dispersed photopolymer. Opt. Express 2018, 26, 6993–7004. [CrossRef] [PubMed] 10 of 10 Polymers 2020, 12, 816 14. Liu, H.; Yu, D.; Li, X.; Luo, S.; Jiang, Y.; Sun, X. Diﬀusional enhancement of volume gratings as an optimized strategy for holographic memory in PQ-PMMA photopolymer. Opt. Express 2010, 18, 6447–6454. [CrossRef] [PubMed] 15. Mahilny, U.V.; Marmysh, D.N.; Stankevich, A.I.; Tolstik, A.L.; Matusevich, V.; Kowarschik, R. Holo volume gratings in a glass-like polymer material. Appl. Phys. B 2005, 82, 299–302. [CrossRef] 16. Fan, F.; Liu, Y.; Hong, Y.; Zang, J.; Kang, G.; Zhao, T.; Tan, X.; Shimura, T. Volume polarization holographic recording in phenanthrenequinone doped poly(MMA-Co-BzMA) photopolymer. Chem. Lett. 2018, 47, 520–523. [CrossRef] 17. Fan, F.; Liu, Y.; Hong, Y.; Zang, J.; Wu, A.; Zhao, T.; Kang, G.; Tan, X.; Shimura, T. Improving the polarization-holography performance of PQ/PMMA photopolymer by doping with THMFA. Opt. Express 2018, 26, 17794–17803. [CrossRef] 18. Li, C.; Cao, L.; Wang, Z.; Jin, G. Hybrid polarization-angle multiplexing for volume holography in gold nanoparticle-doped photopolymer. Opt. Lett. 2014, 39, 6891. [CrossRef] 19. Kawatsuki, N. Photoalignment and Photoinduced Molecular Reorientation of Photosensitive Materials. Chem. Lett. 2011, 40, 548. [CrossRef] 20. Chen, P.L. Phenanthrenequinone-doped copolymers for holographic data storage. Opt. Eng. 2009, 48, 035802. [CrossRef] 21. Steckman, G.J.; Shelkovnikov, V.; Berezhnaya, V.; Gerasimova, T.; Solomatine, I.; Psaltis, D. Holographic recording in a photopolymer by optically induced detachment of chromophores. Opt. Lett. 2000, 25, 607. [CrossRef] [PubMed] 22. Tomita, Y.; Suzuki, N. References Holographic manipulation of nanoparticle distribution morphology in nanoparticle-dispersed photopolymers. Opt. Lett. 2005, 30, 839–841. [CrossRef] [PubMed] 23. Troﬁmova, A.V.; Stankevich, A.I.; Mogil’nyi, V.V. Phenanthrenequinone–polymethylmethacrylate composite for polarization phase recording. J. Appl. Spectrosc. 2009, 76, 585. [CrossRef] 24. Szukalski, A.; Haupa, K.; Miniewicz, A.; Mysliwiec, J. Photoinduced Birefringence in PMMA Polymer Doped with Photoisomerizable Pyrazoline Derivative. J. Phys. Chem. C 2015, 119, 10007–10014. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W2798955091	https://europepmc.org/articles/pmc6399406?pdf=render	English	null	Anti-BP180 Autoantibodies Are Present in Stroke and Recognize Human Cutaneous BP180 and BP180-NC16A	bioRxiv (Cold Spring Harbor Laboratory)	2,018	cc-by	5,286	Edited by: Karin Loser, University of Münster, Germany Edited by: Karin Loser, University of Münster, Germany University of Münster, Germany Reviewed by: Takashi Hashimoto, Osaka University, Japan Hiroaki Iwata, Hokkaido University, Japan Cezary Kowalewski, Medical University of Warsaw, Poland Correspondence: Li Li lilipumch2007@sina.com Bin Peng pengbin3@hotmail.com Reviewed by: Takashi Hashimoto, Osaka University, Japan Hiroaki Iwata, Hokkaido University, Japan Cezary Kowalewski, Medical University of Warsaw, Poland Reviewed by: Takashi Hashimoto, Osaka University, Japan Hiroaki Iwata, Hokkaido University, Japan Cezary Kowalewski, Medical University of Warsaw, Poland Design: One hundred patients with stroke and 100 matched healthy controls were randomly selected for measurement of anti-BP180/BP230 IgG autoantibodies by enzyme-linked immunosorbent assay (ELISA), salt-split indirect immunoﬂuorescence (IIF), and immunoblotting against human cutaneous BP180 and BP180-NC16A. Results: Anti-BP180 autoantibodies were found in 14 (14.0%) patients with stroke and 5 (5.0 %) of controls by ELISA (p < 0.05). Sera from 13 (13.0%) patients with stroke and 3 (3.0%) controls reacted with 180-kDa proteins from human epidermal extract (p < 0.05). 11 (11.0%) of stroke and 2 (2.0%) of control sera recognized the human recombinant full length BP180 and NC16A (p < 0.05). The anti-BP180-positive patients were signiﬁcantly younger than the negative patients at the time of stroke (p < 0.001). Specialty section: This article was submitted to Autoimmune and Autoinﬂammatory Disorders, a section of the journal Frontiers in Immunology Received: 26 October 2018 Accepted: 28 January 2019 Published: 26 February 2019 Specialty section: This article was submitted to Autoimmune and Autoinﬂammatory Disorders, Specialty section: This article was submitted to Autoimmune and Autoinﬂammatory Disorders, a section of the journal Frontiers in Immunology Conclusion: Development of anti-BP180 autoantibodies occurs at a higher frequency after stroke, suggesting BP180 as a relatively common autoantigen after stroke and providing novel insights into BP pathogenesis in aging. Received: 26 October 2018 Accepted: 28 January 2019 Published: 26 February 2019 Keywords: BP180, anti-BP180 autoantibodies, BP180-NC16A, bullous pemphigoid, stroke Anti-BP180 Autoantibodies Are Present in Stroke and Recognize Human Cutaneous BP180 and BP180-NC16A Yanan Wang 1, Xuming Mao 2, Di Wang 1, Christoph M. Hammers 3, Aimee S. Payne 2, Yiman Wang 1, Hongzhong Jin 1, Bin Peng 4 and Li Li 1* 1 Department of Dermatology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy Medical Science, Beijing, China, 2 Department of Dermatology, University of Pennsylvania, Philadelphia, PA, United States, 3 Department of Dermatology, University of Luebeck, Luebeck, Germany, 4 Department of Neurology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy Medical Science, Beijing, China ORIGINAL RESEARCH published: 26 February 2019 doi: 10.3389/ﬁmmu.2019.00236 Keywords: BP180, anti-BP180 autoantibodies, BP180-NC16A, bullous pemphigoid, stroke Objective: Current evidence has revealed a signiﬁcant association between bullous pemphigoid (BP) and neurological diseases (ND), including stroke, but the incidence of BP autoantibodies in patients with stroke has not previously been investigated. Our study aimed to assess BP antigen-speciﬁc antibodies in stroke patients. Objective: Current evidence has revealed a signiﬁcant association between bullous pemphigoid (BP) and neurological diseases (ND), including stroke, but the incidence of BP autoantibodies in patients with stroke has not previously been investigated. Our study aimed to assess BP antigen-speciﬁc antibodies in stroke patients. Immunoblotting Protein extract preparations, polyacrylamide gel electrophoresis, and immunoblotting were performed as previously described (12). Brieﬂy, human foreskin samples were treated with 1 M NaCl (24 h) and the epidermis were subjected to Cell Lysis Solution (Thermo Fisher Scientiﬁc, Massachusetts, U.S.A). Following homogenization, ice-incubation (30 min) and centrifugation (4,700 rpm, 4◦C, 15 min), the supernatant was collected and mixed with loading buﬀer. Primers were designed according to the full length BP180/NC16A DNA sequence to amplify the target gene fragments by PCR and sequence for FLAG tag was incorporated into the cDNA. Then the human full length BP180/NC16A gene fragments were subcloned into the pcDNA3.1 mammalian expression vector. HEK 293 cells were transiently transfected with the plasmids and lipofectamine (Life Technologies, Carlsbad, CA, USA) as per the manufacturer’s instructions, followed by lysis of the cells in a lysis buﬀer (50 mM Tris (PH8.0), 300 mM NaCl, 1% Triton X-100, 1 mM DTT, 5% glycerol). The proteins in lysates were puriﬁed using FLAG peptide aﬃnity chromatography and peptide elution. Expression of the target proteins were conﬁrmed by western-blot (anti-FLAG tag antibody) and quantiﬁed by the protein quantiﬁcation assay kit with ﬁnal concentrations of 600–1,000 µg/ml. Proteins were subjected to 8% SDS- PAGE gels under denaturing conditions, transferred onto a PVDF membrane (Thermo Fisher Scientiﬁc, Massachusetts, U.S.A), and incubated with 0.5 ug/ul human serum samples as the primary antibody and anti-human lgG-HRP (Abcam, Cambridge, Britain) as the secondary antibody. Membranes were developed with detection solution (Merck KGaA, Darmstadt, Germany) and the protein side of the membrane was exposed to an image analysis system (Tanon, Shanghai, China). Anti- human Collagen XVII antibody (Abcam, Cambridge, Britain) (primary antibody) and goat anti-rabbit IgG H&L (HRP) antibodies (Abcam, Cambridge, Britain) (secondary antibody) were used as the positive control to detect BP180-protein in immunoblots. Stroke is one of the most common forms of ND, which often presents as a life-threatening condition and is the second leading cause of death worldwide. It is characterized by an acute onset and considerably severe tissue damage in the brain. Recently, a study including 12,607 patients with ﬁrst-ever stroke has revealed that 38 (0.3%) patients developed BP in a median of 3.5 years, while only eight people (0.06%) had BP in a median of 3.7 years in the control group (10). Immunoblotting In addition, a population-based case- control study has shown that there was a 2-fold increase in risk of developing BP in those with acute ischemic stroke in the UK (11). In the current study, we aimed to quantitatively determine the level and estimate the positive rate of BP autoantibodies in stroke patients, which may shed light on the mechanism for the high incidence of BP in the ND. BP180 and BP230 Enzyme-Linked Immunosorbent Analysis (ELISA) Anti-BP180/BP230 IgG autoantibodies in the sera samples of patients and healthy controls were detected by commercially available ELISAs for human BP180 (NC16A domain) IgG (MEASACUP BP180, MBL, Japan) and BP230 N-and C-terminal domain IgG (BP230 ELISA kit, MBL, Japan), according to the manufacturer’s instructions (based on a cut-oﬀ value >9 U/ml). A signiﬁcant association between BP and neurological diseases (ND) (stroke, dementia, Parkinson disease, epilepsy, and schizophrenia) has been fully supported by a series of previous studies (7–9). Development of BP autoantibodies against both the skin and neuronal forms of antigens in BP with ND might be associated with aging-related dysfunction of the immune system in the elderly population. Alternatively, the autoimmune response might be triggered by chronic inﬂammatory changes or tissue damage in ND, which exposes antigens from the brain to the immune system. Citation: Wang Y, Mao X, Wang D, Hammers CM, Payne AS, Wang Y, Jin H, Peng B and Li L (2019) Anti-BP180 Autoantibodies Are Present in Stroke and Recognize Human Cutaneous BP180 and BP180-NC16A. Bullous pemphigoid (BP) is an autoimmune blistering skin disorder most commonly found in the elderly (1). It is characterized by circulating and tissue-bound autoantibodies directed against two hemidesmosomal components: the transmembrane BP180 (collagen XVII, BPAG2) protein, and the plakin family protein BP230 (BPAG1). The skin-speciﬁc BP180 molecules locate primarily at the basement membrane zone (BMZ) (2), and the neuronal form of BP180 was observed mainly in cytoplasm, which is highly expressed in the hypoglossal nucleus, oculomotor nucleus, and pyramidal cells of the hippocampal regions in human brain (3). Expression of tissue-speciﬁc February 2019 \| Volume 10 \| Article 236 Frontiers in Immunology \| www.frontiersin.org Anti-BP180 Autoantibodies and Stroke Wang et al. isoforms of BP180, such as the skin and neuronal BP180, is attributable to alternative splicing and variation in translational start sites of the gene transcripts, although the exact diﬀerence between these isoforms was not clearly demonstrated (3, 4). Anti- BP180 IgG autoantibodies play a key role in blister formation and correlate with disease activity in BP, particularly at the time of diagnosis and at disease ﬂare (2, 5). These autoantibodies mainly target immunodominant epitopes of BP180 localized in the extracellular non-collagenous 16A (NC16A) domain (2). The majority of BP patients also react with the intracellular antigen BP230, which is thought to result from a secondary immune reaction after tissue damage (6). of 69 and 74%, respectively. Patients with neurological diseases (other cerebrovascular disease, Parkinson’s disease, dementia, multiple sclerosis, and amyotrophic lateral sclerosis) and skin diseases (bullous skin disease, dermatitis, eczema, and other autoimmune diseases) were excluded from the control group after reviewing medical records. Normal human foreskin was obtained from patients receiving circumcision in the Department of Urology. Frontiers in Immunology \| www.frontiersin.org RESULTS BP180 but Not BP230 Autoantibodies Are Signiﬁcantly Elevated in Patients After Stroke Compared to Unaffected Controls Sera from patients with stroke (n = 100) (including cerebral infarction and cerebral hemorrhage) and healthy controls (n = 100) were collected to examine anti-BP180/BP230 IgG antibodies by ELISA (cut-oﬀvalue >9 U/ml). The positive rate of anti-BP180 antibody in the stroke cohort (14, 14.0%) was signiﬁcantly higher than that in controls (5, 5.0%) (P = 0.03) (Tables 1, 2). All anti-BP180 IgG positive patients (14 stroke samples and ﬁve healthy controls) were further examined by immunoblotting against human epidermal extract, human recombinant full length BP180, and human recombinant NC16A (Tables 1, 2). Sera from 13(13.0%) stroke patients and 3 (3.0%) healthy controls reacted with a 180-kDa protein from the human epidermal extract (P = 0.016) (Figure 1A). Sera from 11 (11.0%) patients with stroke and 2(2.0 %) healthy controls recognized both of the human recombinant full length BP180 (P = 0.018) (Figure 1B) and human recombinant NC16A (P = 0.018) (Figure 1C). Anti-BP180 positive sera were further tested by salt-split IIF, and only one patient with stroke revealed IgG antibody binding on the epidermal side of BMZ (Figure 2). CI, Cerebral infarction; CH, Cerebral hemorrhage; IIF, Indirect immunoﬂuorescence; +, Positive; –, Negative. BP180, Human recombinant full length cutaneous BP180; NC16A, Human recombinant cutaneous BP180-NC16A. BP180 ELISA based on a cut-off value >9 U/ml. CI, Cerebral infarction; CH, Cerebral hemorrhage; IIF, Indirect immunoﬂuorescence; +, Positive; –, Negative. BP180, Human recombinant full length cutaneous BP180; NC16A, Human recombinant cutaneous BP180-NC16A. BP180 ELISA based on a cut-off value >9 U/ml. The positive rate of anti-BP230 antibodies had no statistical diﬀerence between the stroke (14, 14.0%) and control groups (15, 15.0%) by ELISA. Patient Samples p Following the principles of the Declaration of Helsinki this study was approved by the Ethical Committee of Peking Union Medical College Hospital and informed consent was obtained from all patients and unaﬀected individuals. Patients with stroke were from the Department of Neurology, Peking Union Medical College Hospital. These included 82 patients with ischemic, 5 with hemorrhagic, and 13 with both ischemic and hemorrhagic stroke, who had been diagnosed with stroke for 1 week to 19 years before our study. The patients were recruited for review and serum samples were collected during July 2014 to October 2016. Stroke was diagnosed based on medical history, clinical symptoms, and results of neuroimaging in the Department of Neurology. The age- and sex- matched control group was comprised of patients attending the hospital for surgery (benign neoplasms resection) in the Department of General Surgery from 2014 to 2016. The average age of stroke and control patients is 66 years for both populations, with a proportion of male patients February 2019 \| Volume 10 \| Article 236 Frontiers in Immunology \| www.frontiersin.org 2 Anti-BP180 Autoantibodies and Stroke Wang et al. TABLE 1 \| Comparison of the BP autoantibody positive rates between stroke and control. Item Stroke Control P-value n 100 100 BP180 ELISA 14(14.0%) 5(5.0%) 0.03* IMMUNOBLOTTING Human epidermal extract 13(13.0%) 3(3.0%) 0.016* Human recombinant full length BP180 11(11.0%) 2(2.0%) 0.018* Human recombinant NC16A 11(11.0%) 2(2.0%) 0.018* Salt split IIF 1(1.0%) 0(0.0%) 0.316 ELISA, enzyme-linked immunosorbent assay; IIF, indirect immunoﬂuorescence. Asterisk “” denotes a statistical signiﬁcance. TABLE 1 \| Comparison of the BP autoantibody positive rates between stroke and control. arison of the BP autoantibody positive rates between stroke and Salt Split Indirect Immunoﬂuorescence (IIF) Five micro molar frozen non-ﬁxed sections of human skin (treated with 1 M NaCl) were blocked (1% BSA in PBS) and human sera in 1:4 to 1:320 dilutions were used as primary antibodies. Rabbit anti-human IgG-FITC (Abcam, Cambridge, Britain) was used as the secondary antibody. The skin sections were then mounted with glycerol/PBS (2:1, pH 9.0) and observed under a ﬂuorescence microscope. A clear linear immunostaining on the BMZ was considered positive, while no ﬂuorescence was considered negative. Statistical Analysis The data involved in the statistical analysis included qualitative analysis, such as BP180 antibody (positive/negative), sex (male/female), complication (yes/no), and frequency of attack (single/ multiple); and quantitative analysis (such as age, BP180/BP230 antibody titer). Speciﬁcally, qualitative data were statistically analyzed by chi square test and logistic regression analysis, and the quantitative data were statistically analyzed by t-test and rank sum test. TABLE 2 \| Immunological testing results of anti-BP180 positive patients. No. Diagnosis BP180 ELISA (U/ml) Immunoblotting Salt-split IF Human epidermal extract BP180 NC16A 1 CI and CH 17 + + + – 2 CI 19 + – – – 3 CI 23 + + + – 4 CI 19 + – – – 5 CI 15 + + + – 6 CI 30 + + + – 7 CI 17 + + + IgG+ 8 CH 23 + + + – 9 CI 22 + + + – 10 CI 10 + + + – 11 CI 18 + + + – 12 CI 30 + + + – 13 CI 10 – – – – 14 CI 16 + + + – 15 Control 11 + + – – 16 Control 11 + + – – 17 Control 10 – – + – 18 Control 16 – – – – 19 Control 13 + – + – CI, Cerebral infarction; CH, Cerebral hemorrhage; IIF, Indirect immunoﬂuorescence; +, Positive; –, Negative. BP180, Human recombinant full length cutaneous BP180; NC16A, Human recombinant cutaneous BP180-NC16A. BP180 ELISA based on a cut-off value >9 U/ml. TABLE 2 \| Immunological testing results of anti-BP180 positive patients. Younger Stroke Patients Are Signiﬁcantly More Likely to Develop BP180 Serum Autoreactivity Than Older Stroke Patients FIGURE 2 \| Positive result of salt split IIF. The IgG autoantibodies in the sera from a stroke patients bind to the epidermal side of BMZ (linear deposition of antibodies as shown in green). According to statistical analysis, the average age of the anti-BP180 positive group (60.1 years) was signiﬁcantly lower than that of the anti-BP180 negative group (69.0 years; P < 0.001). Among them, the proportion of patients younger than 60 years in the anti-BP180 positive group (8/14, 57.1%) was signiﬁcantly higher than that of the anti-BP180 negative patients (19/86, 24.4%; P = 0.006). The duration of follow up after ﬁrst stroke attack for the BP180 positive group (7.0 ± 2.94 years) was signiﬁcantly shorter than that of the anti-BP180 negative group (10.4 ± 6.05 years; P < 0.001). There was no signiﬁcant diﬀerence in sex, complications, and stroke attack times between the two groups (Table 3). FIGURE 2 \| Positive result of salt split IIF. The IgG autoantibodies in the sera from a stroke patients bind to the epidermal side of BMZ (linear deposition of antibodies as shown in green). The Appearance of Serum BP180 Autoantibodies Is Not Associated With Clinical Onset of BP During Short-Term Follow up group (12.2 ± 2.39 U/ml; P = 0.024). The medical records of anti-BP180 positive patients and controls were reviewed and all patients were followed up until October 2017. In the 1-3- year follow-up period, neither stroke patients nor the controls developed BP-like skin lesions. The 1-3-year survival rate of anti BP180 positive patients with stroke and control were both 100%. Anti-BP180 autoantibody titers were signiﬁcantly higher in the stroke group (19.2 ± 6.07 U/ml) compared to those of control February 2019 \| Volume 10 \| Article 236 Frontiers in Immunology \| www.frontiersin.org 3 Wang et al. Anti-BP180 Autoantibodies and Stroke FIGURE 1 \| Autoantibodies in the sera of stroke patients react with BP180 and BP180-NC16A in immunoblotting. Serum antibodies from a patient with bullous pemphigoid (BP), a stroke patient (Stroke), or anti-human BP180 antibody (Control) but not a negative control serum (Negative) recognized a 180-kDa protein from human epidermal extract (A), the human recombinant full length BP180 (B), and human recombinant BP180-NC16A (C). FIGURE 1 \| Autoantibodies in the sera of stroke patients react with BP180 and BP180-NC16A in immunoblotting. Serum antibodies from a patient with bullous pemphigoid (BP), a stroke patient (Stroke), or anti-human BP180 antibody (Control) but not a negative control serum (Negative) recognized a 180-kDa protein from human epidermal extract (A), the human recombinant full length BP180 (B), and human recombinant BP180-NC16A (C). Frontiers in Immunology \| www.frontiersin.org DISCUSSION Previously we have shown that a signiﬁcant proportion of serum samples obtained from patients with BP and ND could react with both the human skin and neuronal forms of BP antigens, supporting the existence of BP antigens in the brain of ND patients (12, 13). In the present study, we evaluate the BP autoAb levels in 100 Chinese patients with stroke and 100 unaﬀected controls, demonstrating that BP autoantibodies are detectable at a higher frequency in stroke patients relative to unaﬀected controls (14 vs. 5%) and they react with human cutaneous BP180 and BP180-NC16A. The 5% incidence of BP autoantibodies in our control group is consistent with a Previously we have shown that a signiﬁcant proportion of serum samples obtained from patients with BP and ND could react with both the human skin and neuronal forms of BP antigens, supporting the existence of BP antigens in the brain of ND patients (12, 13). In the present study, we evaluate the BP autoAb levels in 100 Chinese patients with stroke and 100 unaﬀected controls, demonstrating that BP autoantibodies are detectable at a higher frequency in stroke patients relative to unaﬀected controls (14 vs. 5%) and they react with human cutaneous BP180 and BP180-NC16A. The 5% incidence of BP autoantibodies in our control group is consistent with a previous study in 337 healthy Americans (11/297, 3.7%) by ELISA (14). These observations raise the possibility that BP 180 acts as a shared autoantigen in both stroke and BP. We speculate that severe damage or alterations in the human central nervous system (CNS) during the course of stroke could release or expose the neuronal isoform of BP180, thus triggering an immune reaction that may eventually lead to BP and cutaneous damage (12, 13). February 2019 \| Volume 10 \| Article 236 Frontiers in Immunology \| www.frontiersin.org 4 Anti-BP180 Autoantibodies and Stroke Wang et al. TABLE 3 \| Demographic characteristics of anti-BP180 negative or positive patients in the stroke group. DISCUSSION Variable BP180 (+) BP180 (–) P-value n 14 86 Sex, M/F 11/3 (79/21%) 58/28(67/33%) 0.540 Age, mean ± sd 60.1 ± 11.20 69.0 ± 11.67 <0.001 <60 years, n (%) 8 (57.1) 19 (24.4) 0.006* 60–70 years, n (%) 4 (28.6) 36 (46.2) 0.395 ≥75 years, n (%) 2 (14.3) 23 (29.5) 0.508 Stroke attack times ≥2,n (%) 6 (42.9) 21 (24.4) 0.194 Duration after ﬁrst attack ≥1 years, n (%) 9 (64.3) 49 (57.0) 0.607 Duration after ﬁrst attack (y), median 7.0 ± 2.94 10.4 ± 6.05 <0.001* The numbers and percentages of anti-BP180 negative (–) or positive (+) were listed in the table, respectively. The P-values were calculated by comparing these numbers between these two groups. Asterisk “” denotes a statistical signiﬁcance/. BP180, Human recombinant full length cutaneous BP180. TABLE 3 \| Demographic characteristics of anti-BP180 negative or positive patients in the stroke group. lesions (14). More recently, Kokkonen et al. showed BP180 autoantibodies were found in 18% of patients with Alzheimer’s disease and 3% of controls (P = 0.019), while none of them had BP-like lesions (17). The reason why anti-BP antibody positive patients had no BP-like lesions may be due to a few possibilities. First, titers of anti-BP antibodies in these subjects may be too low to cause cutaneous lesions. Second, some patients may be misdiagnosed because of atypical lesions. In about 20% of BP patients, atypical lesions arise as prurigo-like nodules, intertrigo- like pemphigoid, and localized forms. In about 10–20% of BP patients, disease onset is preceded by a prodromal phase of weeks to months with pruritus, excoriations, and eczematous lesions, and some patients never develop blisters (19). Third, the time of follow-up is not long enough and the study sample is not large enough to develop BP. Future studies with longer clinical follow-up and larger samples will be necessary to clarify the association of BP autoantibody development with overt clinical disease. The numbers and percentages of anti-BP180 negative (–) or positive (+) were listed in the table, respectively. The P-values were calculated by comparing these numbers between these two groups. Asterisk “” denotes a statistical signiﬁcance/. BP180, Human recombinant full length cutaneous BP180. The positive rate of anti-BP180 ELISA in stroke and control groups was a little higher than that of immunoblotting in our study (Table 1). DISCUSSION A meta-analysis described ELISA as a quantitative test with high sensitivity and speciﬁcity (87 and 98–100%, respectively) for diagnosis of BP (20), while immunoblotting was a semi-quantitative test. Therefore, the ELISA positive stroke patient with low titers may have a negative test by immunoblotting. It is also important to note that for BP180, 7.8% of BP sera react exclusively to regions of BP180 outside of the NC16A region, which would not be identiﬁed by using the commercially available BP180 NC16A ELISA test. Thus, a negative test should be closely followed up with DIF and/or IIF (21). There was no ELISA negative patients that were positive for IIF in our study. In our study, only one stroke patient positive both for BP autoantibodies by ELISA and immunoblotting (human BP180 and BP180-NC16A) showed binding of IgG autoantibodies to the epidermal side of the BMZ, consistent with works by Messingham et al. (15), Tuusa et al. (16), and Kokkonen et al. (17) in Alzheimer’s disease, multiple sclerosis, and Parkinson’s disease, respectively. Our results suggest that that these autoantibodies were not able to bind the conformational epitope of skin BP180 but the linear protein epitope due to release of the neuronal form, indicating that epitopes bound by BP patient antibodies and stroke patient antibodies do diﬀer. The stroke patient antibodies may bind to epitopes only present in the denatured skin BP180, except for the one case. Anti-BP180 autoantibody values were signiﬁcantly higher in the stroke group (19.2 ± 6.07 U/ml) compared to those in the control group by ELISA (12.2 ± 2.39 U/ml; P = 0.024). As the anti-BP180 antibodies are correlated with the disease activity of BP, we considered that the diﬀerence between the two groups is of clinical signiﬁcance and may herald the development of BP in the future. Jedlickova et al. found that male stroke patients were more likely to have BP (22). In our study the proportion of male anti-BP180 positive patients (11/14, 84.6%) showed no signiﬁcant diﬀerence compared to controls (4/5, 80.0%; P > 0.999). Notably, we found that anti-BP180 positive stroke patients (60.1years) were signiﬁcantly younger than anti-BP180 negative stroke patients (69.0 years; P < 0.001), suggesting that young age might be a risk factor for stroke patients to develop BP. Frontiers in Immunology \| www.frontiersin.org ACKNOWLEDGMENTS The authors are grateful to Nan Yang, for her outstanding assistance. We also thank biorxiv (www.biorxiv.org) for publishing a pre-print of this manuscript (23). AUTHOR CONTRIBUTIONS YaW performed most of the experiments, analyzed the data, and wrote the manuscript. LL, BP, and HJ conceived the project and supervised the study. LL, YaW, and DW designed the experiments. XM interpreted data and revised the manuscript. LL, BP, CH, AP, and YiW discussed the results and revised the manuscript. REFERENCES One gene but diﬀerent proteins and diseases: the complexity of dystonin and bullous pemphigoid antigen 1. Exp Dermatol. (2016) 25:10–6. doi: 10.1111/exd.12877 19. Danescu S, Chiorean R, Macovei V, Sitaru C, Baican A. Role of physical factors in the pathogenesis of bullous pemphigoid: case report series and a comprehensive review of the published work. J Dermatol. (2016) 43:134–40. doi: 10.1111/1346-8138.13031 7. Lai YC, Yew YW, Lambert WC. Bullous pemphigoid and its association with neurological diseases: a systematic review and meta-analysis. J Eur Acad Dermatol Venereol. (2016) 30:2007–15. doi: 10.1111/jdv.13660 7. Lai YC, Yew YW, Lambert WC. Bullous pemphigoid and its association with neurological diseases: a systematic review and meta-analysis. J Eur Acad Dermatol Venereol. (2016) 30:2007–15. doi: 10.1111/jdv.13660 20. Muglia C, Bronsnick T, Kirkorian AY, Cha J. Questioning the speciﬁcity and sensitivity of ELISA for bullous pemphigoid diagnosis. Cutis (2017) 99:E27–e30. 8. Pietkiewicz P, Gornowicz-Porowska J, Bowszyc-Dmochowska M, Bartkiewicz P, Dmochowski M. Bullous pemphigoid and neurodegenerative diseases: a study in a setting of a Central European university dermatology department. Aging Clin Exp Res. (2016) 28:659–63. doi: 10.1007/s40520-015-0459-4 21. Keller JJ, Kittridge AL, Debanne SM, Korman NJ. Evaluation of ELISA testing for BP180 and BP230 as a diagnostic modality for bullous pemphigoid: a clinical experience. Arch Dermatol Res. (2016) 308:269–72. doi: 10.1007/s00403-016-1631-1 9. Khosravani S, Handjani F, Alimohammadi R, Saki N. Frequency of neurological disorders in bullous pemphigoid patients: a cross-sectional study. Int Sch Res Notices (2017) 2017:6053267. doi: 10.1155/2017/6053267 22. Jedlickova H, Hlubinka M, Pavlik T, Semradova V, Budinska E, Vlasin Z. Bullous pemphigoid and internal diseases - a case- control study. Eur J Dermatol. (2010) 20:96–101. doi: 10.1684/ejd. 2010.0805 10. Shen AL, Lin HL, Lin HC, Tseng YF, Hsu CY, Chou CY. Increased risk of bullous pemphigoid after ﬁrst-ever stroke: a population-based study. Neurodegener Dis. (2017) 17:166–70. doi: 10.1159/000469710 23. Wang Y, Mao X, Wang D, Hammers CM, Payne AS, Wang Y, et al. Anti- BP180 autoantibodies are present in stroke and recognize humancutaneous BP180 and BP180-NC16A. (2018) BioRxiv [PrePrint]. doi: 10. 1101/313692 11. Langan SM, Groves RW, West J. The relationship between neurological disease and bullous pemphigoid: a population-based case-control study. J Invest Dermatol. (2011) 131:631–6. doi: 10.1038/jid.2010.357 12. Chen J, Li L, Chen J, Zeng Y, Xu H, Song Y, et al. Sera of elderly bullous pemphigoid patients with associated neurological diseases recognize bullous pemphigoid antigens in the human brain. Gerontology (2011) 57:211–6. FUNDING further clarify why BP is signiﬁcantly associated with aging and neurologic diseases. This study was funded by National Natural Science Foundation of China (81371731), Milstein Medical Asian American Partnership foundation (2017, Dermatology), and the Education Reform Projects of Peking Union Medical College (No. 2016zlgc0106). DISCUSSION The proportion of patients younger than 60 years in the anti-BP180 positive patients (8/14, 57.1%) was signiﬁcantly higher than that in the anti-BP180 negative patients (19/86, 24.4%; P = 0.006) (Table 3). The duration of follow-up after ﬁrst stroke attack of anti-BP180 positive patients (7.0 ± 2.94 years) was signiﬁcantly shorter than that of anti- BP180 negative patients (10.4 ± 6.05 years; P < 0.001), further supporting that younger stroke patients with shorter duration after ﬁrst attack are more likely to develop BP antibodies. This may be due to strong immune responses in the early stage after stroke, whereas there is down-regulation of the autoimmune response in the late or recovery stage of stroke (18). We conclude that anti-BP180 autoantibodies could be detected at a higher rate in stroke patients than age- and sex- matched controls, supporting that BP180 could serve as a shared autoantigen in both stroke and BP. Our data suggest young age in stroke could be a risk factor for later developing BP and provide a molecular mechanism for BP associated with ND (23). The present study is limited by a relatively short follow- up time so that, although long enough to detect signiﬁcant diﬀerences in serum reactivity, may not have been long enough to allow for the subsequent development of BP skin lesions. Because of the low incidence of BP, a larger sample size is also necessary to detect the clinical onset of BP after stroke. In the future, cell and animal models may be used to help During our follow-up, neither the anti-BP180/BP230 positive stroke patients nor the controls exhibited BP-like skin lesions, in accordance with a previous study that showed none of the anti-BP180/BP230 antibody positive individuals had BP-like skin February 2019 \| Volume 10 \| Article 236 Frontiers in Immunology \| www.frontiersin.org 5 Anti-BP180 Autoantibodies and Stroke Wang et al. REFERENCES 15. Messingham KA, Aust S, Helfenberger J, Parker KL, Schultz S, Mckillip J, et al. Autoantibodies to collagen XVII are present in parkinson’s disease and localize to tyrosine-hydroxylase positive neurons. J Invest Dermatol. (2016) 136:721–3. doi: 10.1016/j.jid.2015. 12.005 1. Kridin K, Ludwig RJ. The growing incidence of bullous pemphigoid: overview and potential explanations. Front Med. (2018) 5:220. doi: 10.3389/fmed.2018.00220 1. Kridin K, Ludwig RJ. The growing incidence of bullous pemphigoid: overview and potential explanations. Front Med. (2018) 5:220. doi: 10.3389/fmed.2018.00220 2. Liu Y, Li L, Xia Y. BP180 is critical in the autoimmunity of bullous pemphigoid. Front Immunol. (2017) 8:1752. doi: 10.3389/ﬁmmu.2017.01752 2. Liu Y, Li L, Xia Y. BP180 is critical in the autoimmunity of bullous pemphigoid. Front Immunol. (2017) 8:1752. doi: 10.3389/ﬁmmu.2017.01752 16. Tuusa J, Lindgren O, Tertsunen HM, Nishie W, Kokkonen N, Huilaja L, et al. BP180 autoantibodies target diﬀerent epitopes in multiple sclerosis or alzheimer’s disease than in bullous pemphigoid. J Invest Dermatol. (2018) 139, 293–299. doi: 10.1016/j.jid.2018. 09.010 3. Seppänen A, Suuronen T, Hofmann SC, Majamaa K, AlafuzoﬀI. Distribution of collagen XVII in the human brain. Brain Res. (2007) 1158:50–6. doi: 10.1016/j.brainres.2007.04.073 3. Seppänen A, Suuronen T, Hofmann SC, Majamaa K, AlafuzoﬀI. Distribution of collagen XVII in the human brain. Brain Res. (2007) 1158:50–6. doi: 10.1016/j.brainres.2007.04.073 j 4. Seppänen A. Collagen XVII: a shared antigen in neurodermatological interactions? Clin Dev Immunol. (2013) 2013:240570. doi: 10.1155/2013/240570 17. Kokkonen N, Herukka SK, Huilaja L, Kokki M, Koivisto AM, Hartikainen P, et al. Increased levels of the bullous pemphigoid BP180 autoantibody are associated with more severe dementia in Alzheimer’s disease. J Invest Dermatol. (2017) 137:71–6. doi: 10.1016/j.jid.2016.09.010 4. Seppänen A. Collagen XVII: a shared antigen in neurodermatological interactions? Clin Dev Immunol. (2013) 2013:240570. doi: 10.1155/2013/240570 5. Hammers CM, Stanley JR. Mechanisms of disease: pemphigus and bullous pemphigoid. Annu Rev Pathol. (2016) 11:175–97. doi: 10.1146/annurev-pathol-012615-044313 5. Hammers CM, Stanley JR. Mechanisms of disease: pemphigus and bullous pemphigoid. Annu Rev Pathol. (2016) 11:175–97. doi: 10.1146/annurev-pathol-012615-044313 18. Fu Y, Liu Q, Anrather J, Shi FD. Immune interventions in stroke. Nat Rev Neurol. (2015) 11:524–35. doi: 10.1038/nrneurol.2015.144 6. Kunzli K, Favre B, Choﬄon M, Borradori L. One gene but diﬀerent proteins and diseases: the complexity of dystonin and bullous pemphigoid antigen 1. Exp Dermatol. (2016) 25:10–6. doi: 10.1111/exd.12877 6. Kunzli K, Favre B, Choﬄon M, Borradori L. REFERENCES doi: 10.1159/000315393 Conﬂict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. 13. Li L, Chen J, Wang B, Yao Y, Zuo Y. Sera from patients with bullous pemphigoid (BP) associated with neurological diseases recognized BP antigen 1 in the skin and brain. Br J Dermatol. (2009) 160:1343–5. doi: 10.1111/j.1365-2133.2009.09122.x Copyright © 2019 Wang, Mao, Wang, Hammers, Payne, Wang, Jin, Peng and Li. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 14. Wieland CN, Comfere NI, Gibson LE, Weaver AL, Krause PK, Murray JA. Anti-bullous pemphigoid 180 and 230 antibodies in a sample of unaﬀected subjects. Arch Dermatol. (2010) 146:21–5. doi: 10.1001/archdermatol.2009.331 February 2019 \| Volume 10 \| Article 236 Frontiers in Immunology \| www.frontiersin.org 6
W4220831046.txt	http://repositori.upf.edu/bitstream/10230/54947/1/Zapata_im_medi.pdf	en		Mediterranean thinking in migration studies: A methodological regionalism approach	International migration	2,022	cc-by	10,052	\| Revised: 20 December 2021 \| Accepted: 22 February 2022 DOI: 10.1111/imig.12995 ORIGINAL ARTICLE Mediterranean thinking in migration studies: A methodological regionalism approach Ricard Zapata-Barrero Interdisciplinary Research Group on Immigration, GRITIM-UPF, Department of Social and Political Science, Pompeu Fabra University, Barcelona, Catalonia, Spain Correspondence Ricard Zapata-Barrero, Social and Political Sciences Department, University Pompeu Fabra, Ramon Trias Fargas, 25-27, 08002 Barcelona, Catalonia, Spain. Email: ricard.zapata@upf.edu Funding information Erasmus+ Jean Monnet Network Program Abstract The core reflection of this article is to explore the potential of using the Mediterranean as a category of analysis for migration studies, what epistemological and ontological effects this may have and how this could be done. To better capture this focus, I will speak about Med-Thinking. This Mediterranean scale of analysis invites us to follow methodological regionalism. In practice, this means that we need to encourage scholars to channel their findings towards the development of a Mediterranean approach. To develop an integrative and reflexive regional approach, I will first point out the background premises of such a normative effort and second, I will present the main analytical tools framing this Med-Thinking approach by articulating four main dimensions: epistemic solipsism, holism, homeostatic and positive thinking on migration. In the end, I will suggest some paths to promote the continued application of the Med- Thinking approach for the development of Mediterranean migration studies. I NTRO D U C TI O N : FR A M I N G TH E M E D -T H I N K I N G A PPROAC H The core reflection of this article is to explore the potentialities of using the Mediterranean as a category of analysis for migration studies, what epistemological and ontological effects this may have and how this could be done from a critical thinking perspective. To better capture this focus, I will speak about Mediterranean thinking (Med-Thinking).1. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. International Migration published by John Wiley & Sons Ltd on behalf of International Organization for Migration. International Migration. 2022;00:1–14. wileyonlinelibrary.com/journal/imig \| 1 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Received: 22 July 2021 \| ZAPATA-BARRERO This Mediterranean scale of analysis invites us to contribute to an emerging scale of analysis between the global and local migration research, 2. and shape the contours of methodological regionalism in migration studies. This is a direct appeal to conduct most of the Mediterranean migration research findings towards a region-making process, detaching what is distinctive from this region of the world. The background premise is that even today, if we can say there is an undergirding research on Mediterranean migration, from a Med-Thinking point of view, findings are disarticulated and lost in a conceptual limbo. This is because there is no Mediterranean portfolio that frames the outcomes within a regional approach. In seeking to point out a series of epistemological and methodological pitfalls in our current understanding of migration in the Mediterranean region, this article could be useful for a research community which is often faced with a fragmented Mediterranean migration research area. In practice this means that we need to encourage scholars to channel their research towards the development of a Mediterranean approach. To develop an integrative and reflexive regional approach, the following sections will point out, first the background premises of such a normative effort and second it will present the main analytical tools framing this Med- Thinking approach by articulating four main dimensions: epistemic solipsism, holism, homeostatic and positive thinking on migration. In the end, I will suggest some paths to promote the continued application of the Med- Thinking approach for the development of Mediterranean migration studies. BAC KG RO U N D PR E M I S E S : CO NTE X T UA L A N D A N A LY TI C A L BA S I S O F M E D -TH I N K I N G So far, migration has been one of the biggest and most divisive political questions of the twenty-first century, and it seems likely to be a central part of our politics going forward, especially in the Mediterranean area where several continents and migration regimes meet. Population movements constitute the bedrock of Mediterranean history (Chambers & Curti, 2008; Pace & Roccu, 2020), but controlling it is a key feature of current Mediterranean history. Today, the Mediterranean is considered as the most diasporic region in the world (Gallant, 2016, 205), as the world's least peaceful region (Global Peace Index, 2018) and by far the world's deadliest zone (UNHCR and IOM latest reports). It is also a space where migration is governed by a multipolar geo-politics where the Mediterranean plays the role of a chess game with multiple key players and migrant pawns (Garelli et al., 2018; Gillespie & Volpi, 2019). Current EU politics is basically re-active by creating a hostile environment for Med-movers that are unable to meet regular human migration corridors (Fekete, 2018; Pécoud, 2020), while criminalizing those social servants working for migrant rescue (Paoletti, 2011; Stierl, 2016). The Mediterranean is definitively viewed as a border space full of filters and conditions basically dominated by a unipolar EU Thalassocracy (Abulafia, 2014) dominance of all the Mediterranean space by concentrating power in the EU and European states. This current history of restrictions and lockdowns of mobility creates a ‘disquieting account’ in Chambers’ terms (2008, 3). This is because presentism and event-based concerns prevails with a crisis narrative that generates an atmosphere of uncertainties and insecurity (Kirtsoglou & Tsimouris, 2016; Wolff & Hadj-Abdou, 2019), which often fuels xenophobic and neo-nationalist discourses in contemporary Europe (Michael, 2021), racial Europeanization (Goldber, 2006) and religious and anti-European fundamentalisms outside the European Mediterranean territory (Adamson & Tsourapas, 2020; Natter, 2018; Thiollet, 2011; Tsourapas, 2018). It also invites us to link Mediterranean migration with postcolonial critical thinking (Mayblin & Turner, 2021), where it becomes clear that Europe is governed by race and religion drivers in its politics of containment (Garelli & Tazzioli, 2020). Within this current scenario, Med-thinking is an epistemological and ontological invitation to have a second- order reflection on the way migration research is producing and using knowledge. This article is a methodological exercise of positionality and reflexivity taking the Mediterranean as a category of analysis. These two epistemic attitudes activate self-awareness and critical thinking, and it even invites us to be questioned by the substantial ethical question of who benefits from research outcomes (Zapata-Barrero & Yalaz, 2019). These second-order 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2 \| 3 rationales are crucial to identify the key features of Med-Thinking. Reflexivity leads us towards a reflection on the scope of research not only in space (holistic thinking) but also in time (homeostatic thinking). Positionality also questions the self-systemic process that only legitimatizes the knowledge that oneself produces for the benefits of oneself. This is better captured with what I call ‘epistemic solipsism’. Positionality also invites us to have an axiological reflection about how we focus migration issues. It is a fact that the mainstream rationale frames Med- movers as a problem. Positive thinking is then much needed. The main purpose of this article is also to conduct these rationales towards the normative dimension of Mediterranean migration research. In fact, this normativity of the Mediterranean as a dynamic region fabric probably belongs to the most well-known ‘romantic yearnings’ of the present/past of this geographical space. The ‘imaginary’ of peace, stability, prosperity and common values of free mobility of people/goods/information is of great importance, since it constitutes the utopian skyline of most governance instruments being deployed since the beginning of the 1995 Mediterranean Partnership Framework process (the so-called Barcelona process), the first EU initiative of region-making.3. Yet, this is only an ideal scenario, as this is not ‘imaginary’, yearnings are not happening within any European state nor within the European Union itself, which is full of migration-related cleavages. The fact is that in this twenty-seventh anniversary, the failure diagnosis prevails because it has not managed to shape a positive view of Med-mobility, as it is always placed in the basket of security. This EU Mediterranean partnership was initially (and ideally) a multilateral affair that shifted to a much more bilateral one, where most relations are dominated by conditionality or the so-called more-for-more rhetoric or an incentive-based approach that began to be applied during the second decade of the twenty-first century.4. Taking this regional making as the normative horizon of Med-Thinking, allow me in the next section to further attempt to draw this Mediterranean methodological regionalism, before entering to the different dimensions that fashion the Med-Thinking approach. At the end, I will emphasize how far postcolonialism clusters most of the dimensions shaping the Med-Thinking approach and suggest some application paths so as to promote the continued normative development of Mediterranean migration studies. M E D -T H I N K I N G A S A M E TH O D O LO G I C A L R EG I O N A L LE N S Med-Thinking is beforehand an invitation to frame migration research with a particular methodological regional lens. Regional building theories are often based on a number of shared markers and a set of assumptions under which a ‘region’ makes sense and is thus given a shared meaning (Pace, 2006; 27). The premise is that a region is not a natural entity, but rather a political and social construct. There are broadly speaking two different ways of explaining the making of regions. Materialist theories emphasize the resource basis such as commonly shared characteristics, like geography, language, religion, economic ties and institutions. Ideationist scholars, on the other hand, have argued that although material factors matter, regions are above all ‘imagined communities’. This means that regions are socially constructed entities, created by common narratives and shared spaces that may contribute to creating a sense of belonging. Med-Thinking combines both. Sympathizing with most of Pace's focus on regional identity building, the fact is that material factors become intelligible only in light of ideational factors (Pace, 2006; 10). The spatial dimension is then key for Med-Thinking. Region-making is the sum of relations, connections, embodiments, narratives and practices, involving a large series of different scales of actors (Jones & Paasi, 2013). This relational view usually links politics, territory and power (Martin, 2009; Paasi & Metzger, 2017). It sketches different networks that belong to different state jurisdictions and even political regimes and religious/nationalist ideologies. Regions are envisaged as complicated constellations of materiality, agency, social relations and power. This ‘relational thinking’ is discussed by multiple geographers (Jones & Passi, 2013; Varro & Lagendijk, 2012) and suggests that the Med-region should not be seen as a bounded nested territory but as a net space of nodal places 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MEDITERRANEAN THINKING IN MIGRATION STUDIES \| ZAPATA-BARRERO with given narratives, practices and perceptions. Geographers often use region-making in descriptive terms in general as a set of spatial relations, but less on the added value of this region-making in normative terms, which consist of building a space of shared value making and even identity making. This methodological regionalism grounds the basic dimensions shaping Med-Thinking. Let me start with epistemic solipsism. E PI S TE M I C S O LI P S I S M I N M E D ITE R R A N E A N M I G R ATI O N S T U D I E S : EU RO - C E NTR I S M I S O U R TA RG E T Solipsism is a view that the world is ‘my world’ in the sense that nothing exists independently of how I interpret the world. For me it is the epistemological foundation of Western-centrism, Eurocentrism and even ‘white’ supremacy that leads the current EU Thalassocracy. It is very close to methodological individualism (Heath, 2020), but applied to the politics of knowledge of the EU and European states. A good example of solipsism is the credo that ‘there is no civilization outside of White Euro-America’ (Hicks, 2020; 45). Epistemic solipsism shows us how knowledge production is related to self-interest. The Med-Thinking approach develops its first dimension as a radical criticism to the epistemic solipsism governing our views about Mediterranean migration and the geo-migration politics that follows. As an epistemological position, solipsism is driven by power relations and holds that knowledge of migrants outside one's own vantage point is unsure, unstable and can even, in its radical form, threaten our own worldviews. For Med-Thinking, solipsism is an epistemological aggression and a channel of legitimation of current hostile Mediterranean migration policies that have difficulties passing the evidence-based approach, contextual dependent and ideology laden (Baldwin-Edwards et al., 2018). Even if we may adhere to the importance of placing human rights and democracy, gender, equality, diversity recognition and interculturalism as Mediterranean fixed values, liberal nationalism, religious, cultural and even racial narratives tend to create a fluidity of interpretations of these values. Solipsism even has an ethical sense of selfishness, which takes the form of egocentrism. By targeting solipsism, we can identify assumptions that nurture our own vantage point, legitimating one's own position but rejecting the others as ‘less credible’ and even ‘dangerous’ for our ‘western standards’. Med- Thinking is an outcome of the critique of methodological nationalism applied at the regional/ Mediterranean level. It is an answer to the current diagnosis of how certain Mediterranean migration research is performed: state-centred, dominated by narratives/practices/perceptions of crisis and security, dominated by top-down approaches, presentism, event-driven, highly Euro-centric and White/Christian supremacist. This Euro- centric view generates the geopolitical environment for an ‘Europeanization’ of the Mediterranean (the current Thalassocracy) fixing binary categorizations and monolithic constructs of the Mediterranean, which we need to problematize. Here, we can reflect on the diagnosis done by Arendt in the Preface to the first edition (1950) of The Origin of Totalitarianism, when she stated that: ‘the subterranean stream of Western history has finally come to the surface and usurped the dignity of our tradition. This is the reality in which we live’ (1976; ix). Against this epistemic solipsism, Med-Thinking seeks to pursue a bottom-up polycentric view of Mediterranean migration research by learning to widen our current often ‘biased or unilateral’ perspectives, providing a ‘variable focal length’ (see Zapata-Barrero & Faustini, 2019). It is an epistemological cognitive denunciation on how far national narratives (and EU and European States policies) govern interpretations of the current trans-Mediterranean population dynamics and governance. It is important to capture the chronopolitics (Hicks, 2020) behind this Eurocentrism in order to shape the Med-Thinking approach. This also entails that we probably need to reassess the current Mediterranean migration research agenda, which is overly focused on a European view of crisis and instability in the region. This can allow us to interpret the so-called EuroMed policy as being nothing more than a Europeanization of the Mediterranean and a historical continuation of the hegemonic roman imperial and colonial ‘Mare Nostrum’ legacy. Along these lines, Chambers 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 \| 5 (2008, 20) reminds us that legislating against migrants reveals a war against the political articulation of difference. Most EU Mediterranean policies since then have, accordingly, been focused on achieving the necessary conditions for long-term political and economic stability in this ‘bitterly contested and fractured geopolitical space’ (Giaccaria & Minca, 2011, 352). This means to articulate an answer to basic questions in migration studies that explodes the empty, homogeneous continuum of current historical knowledge and include what has been silenced, occluded, marginalized, culturally repressed and physically eradicated from a solipsist Europe, as most postcolonial thought denounces (Chambers & Curti, 1996; Pace & Roccu, 2020). As a political theoretical project, Med-Thinking is concerned with the decolonization of the West representation of the Mediterranean, of taking the Mediterranean as a continual interweaving of cultural roots and historical routes. Chambers even speaks about the ‘epistemological violence’ of liberal thought, deposited in the implicit knot of race and civilization (Chambers, 2008, 28). H O LI S TI C TH I N K I N G : TH E CO N C E P T A N D IT S A PPLI C ATI O N FO R G U I D I N G M E D ITE R R A N E A N M I G R ATI O N S T U D I E S ‘Holism’ is a concept that has become an umbrella term that is used to encompass anything that is non- reductionist or dualist, monolithic construct. Holism is sometimes spelled as ‘wholism’, to accentuate its semantic connections to the ‘whole’ (Shoroff, 2011, 244–5). It pictures an organic system view, if we cannot understand the whole by knowing the components in isolation from each other. In sociology, Durkheim developed a concept of holism which he set as opposite to the notion that a society was nothing more than a simple collection of individuals. Ecology is the leading and most important approach to holism, as it tries to include biological, chemical, physical and economic views in a given area. The complexity grows with the area, so that it is necessary to reduce the characteristic of the view in other ways, for example to specific time duration. Epistemologically holism can prevent solipsism since it invites the researcher to look at the plurality of angles and narratives of a given topic. Holistic thinkers believe that events or social dynamics are products of external forces and situations. This involves that the knowledge we may produce needs to also be framed within a given relational space, and that migration-related topics cannot be analysed out of a contextual reference framework. Taking Foucault (2004) reflections on space, holistic thinking considers space as ‘emplacements’ rather than ‘localizations’. Emplacements are sites that can only be thought of if they are in relation to other sites. Emplacements are context-based locations. Epistemologically it can be contrasted to monism and atomism, namely and respectively, the credo that there is just one truth and principle of a given event, and the credo that there is just one indivisible component as a unit of analysis, the individual, the nation, the religion. In common usage, this all-encompassing concept emphasizes the interconnectedness of its components within a systemic approach. This also involves including as many aspects as we can from the given topic, we are analysing ethical, political, economic, legal, sociological, psychological, anthropological, etc. Holistic thinking requires both interdisciplinarity and intersectionality. This combines mainstream patterns, although it is also open to explore other ways of understanding a given fact, trying to go beyond mainstream narratives. In this case, holistic thinking often falls within critical thinking because looking at a topic from different angles also allows for weighing up arguments and drawing conclusions. Applied to Med-Thinking, holism assumes a multiscale approach for generating understanding, from the individual to collective associations, institutions, cities, sub-regions, etc. The new UN Global Compact on migration (2018) is a perfect example of holistic thinking. It combines a global approach of migration and a view of the different types of relationships between many elements in a complex system. As a methodological approach, a good analogy could be a puzzle. A puzzle can be overwhelming if you just have a bunch of jumbled pieces and try to put them together without guidance (Terry, 2016). The Mediterranean region-making is one of the needed frames of reference required for holistic thinking. 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MEDITERRANEAN THINKING IN MIGRATION STUDIES \| ZAPATA-BARRERO Going from concept to applications, holism has also been one approach proposed at the EU Parliament on 23 March 2016 to deal precisely with Mediterranean migration affairs: ‘on the situation in the Mediterranean and the need for a holistic EU approach to migration’ (2015). In spite of being endorsed by the majority of the EU Parliament, this approach has had little policy continuity and less traction by researchers. It is a fact that the concept appears five times in the document and in a descriptive form, without any serious conceptual definition. It seems to mean ‘broad approach’ that tries to go beyond short-termism, but it incorporates some key analytical understandings of holism as a system where different kinds of policies interconnect. As the preliminary document grounding this EU proposal, the S&D research paper released in December 2015 stated that holism was ‘one that embodies the principle of solidarity, thus enabling Member States to share responsibility fairly and to maintain a focus on human rights’ (Socialists & Democrats, 2015, 4). This holistic initiative includes solidarity and the fair-sharing of responsibility, border management and visa policy, developing safe and lawful routes for asylum seekers and refugees into the EU, developing a strategy for cooperating with third countries, developing adequate legal economic migration, examining the use of funding in the area of migration and implementing the Common European Asylum System. Connecting all these actions within a holistic umbrella expresses the wish to encapsulate, within a single box concept, all aspects of the Mediterranean migration system, together and in conjunction with one another as the most effective way to handle migration. The message was clear: that concrete Mediterranean migration policies cannot be done in isolation from each other and that a systemic view was necessary. This would also involve more coordination from agencies and offices in charge of Mediterranean migration and development aid. Among the several measures, we can summarize that in the short term this involves a humanitarian response to Mediterranean migration and in a medium term, greater investment and enforcement against smuggling and trafficking networks, and to look at alternative forms of admission so that people no longer resort to smugglers. The debate turned around very different and probably dispersed and often contradictory claims taking the shape of policy priorities. This illustrates that holism has then had some application, even if we may discuss if this holistic approach passes the filter of solipsism, so important for the Med-Thinking approach. E X PLO R I N G TH E CO N C E P T O F H O M EOS TA S I S A N D CO N S I D E R I N G IT S I M PLI C ATI O N S FO R M E D ITE R R A N E A N M I G R ATI O N S T U D I E S There is a widespread tendency in migration studies to focus research on social, political, economic, cultural changes, leaving aside and even discrediting what remains after a process of change and have a more permanent temporality. This is probably due to the fact that most background research assumes that migration and the diversity that follows is one of the key drivers of change today. These initial assumed premises blur most of the current research focus and are probably one of the key rationales why short-termism prevails over long-term focus in Mediterranean migration. This event-driven prevalence over more longue durée shows that there is often a lack of a genuine philosophy of history in most Mediterranean migration research. This philosophy of history may invite us to zoom in on the time frame of what is being researched to place a given topic within an historical process. Within this background, Med-Thinking is claiming for some sort of Copernic turn, inviting the researcher to focus on continuities over changes. In philosophical terms, we can state for a more Parmenidean reading over a more Heraclitan reading of Mediterranean migration research. The Parmenides approach fixes the focus of what remains over change, while the Heraclitan approach points towards what changes over what remains. The stasis of Parmenides is then contrasted to the flux of Heraclitus. This is the basis of homeostatic thinking. It is much more related to the Braudelian's effort to approach the Mediterranean looking at longue durée events in contrast to those that only have a short-term life and are event-driven, with little impact on future developments. In its definitional format, the concept of homeostasis comes from biology and physiology. Digging into this origin may help us clarify its use in shaping Med-Thinking. The term derives from Greek roots meaning ‘similar’ (homeo, not to be confused by homo:the same) and ‘a state of stability’. In biology it also designates ‘the ability 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 \| 7 to maintain a stable internal state that persists despite changes in the word outside’ (Lanese, 2019). Broadly speaking, it designates a state of equilibrium (balance between opposing pressures), as in an organism or cell, maintained by self-regulating processes (Miller-Keane Encyclopedia, 2003). In fact, self-regulation is seen as a homeostatic mechanism in a given body or system. There is also an ontological dimension of the concept. For instance, Rodolfo (2000) referred to ‘homeostasis’ as ‘to any process that living things use to actively maintain fairly stable conditions necessary for survival’. This ontological background may have many applications in social sciences. It can even be related to ‘resilience’, since homeostatic systems may have primarily evolved to help organisms maintain optimal function in different stressing environments and shocking situations (Biggs et al., 2016). A homeostatic society maintains its stability despite competing political, economic and cultural factors (Rodolfo, 2000). Some scientists theorize that homeostasis primarily provides a ‘quiet background’ for cells, tissues and organs to communicate with each other. This ‘quiet background’ is often linked to equilibrium and balance. The word element-stasis implies static but also fixed or an unmoving state. In socio-psychological terms it may refer to how a person under conflicting stresses and motivations can maintain a stable psychological condition. Taken historically, it is much more related to continuity rather than changes, but adds that this continuity helps us to understand the present. The application into the social sciences may be multifaceted. We may use homeostasis as a framework to understand how people and patterns maintain stability/routines despite disruptions. As a perspective for social science research, homeostasis can help underline how important it is to box fixed points in a changing environment, since these fixities in a process of change may help to give meaning to these changes and even help to draw different scenarios. If the focus is on what is stable, fixed in a given dynamic system, we may, for instance, have the seeds of identity building, a Mediterranean identity, based on historical continuities and fixities. Med-Thinking is then characterized by powerful homeostatic tendencies. If we take a multi-layered view of history, following Braudel philosophy, the temporality of events is dominated by different historical strands that may overlap in the present. This involves that we can critically explore past structural events that have been silenced by politics and ideology, as it is often underlined by postcolonial thinkers. In this case, the making of silence, of ‘political historical amnesia’, is incorporated into the Med-Thinking agenda. This homeostatic perspective is being used by some demographers (Wilson & Airey, 1999) and economists (Damasio & Damasio, 2016) looking at functional variables that may help to understand and explain fluctuations and regime transitions. I contend that if Mediterranean migration is looking at objectifiable drivers that may help us to understand and explain complex patterns of human mobility, diversity and governance, we need a framework of thought that tries to capture homeostatic regimes beyond variations, uncertainties and indomitable factors. This can be a valuable approach that can help give light to the normative dimension of Med-Thinking as well. This also implies that beyond variation Med-Thinking seeks to encode temporally long-term variations. One well-known and often quoted homeostatic thinking is the famous ‘invisible hand’ that guides the market capitalist economy (Damasio & Damasio, 2016; 129). Invisible and often escaping conscious rationality factors that give us light to understand current patterns and probably may help us to find ways of taming changes and variations, uncertainties that often paralyzes our understanding. This dimension captures one traditional meaning of homeostasis calling attention to a non-conscious form of physiological control which operates automatically without awareness or deliberation on the part of the organism. Applying a homeostatic approach to Mediterranean migration provides us with a temporal context. Beyond communication channels, human mobility and translocal relations, what increasingly connects the past and the present can be clustered by colonialism. Here we welcome the postcolonial studies, as most of the colonized countries are now countries of emigration and the main producers of the largest diaspora of the world. Med-Thinking invites us to enter in the domain of uprooted geographies (Chambers, 2008). Postcolonialism is accounting for and combating the residual effects of colonialism and empires on culture, society and politics. It is by nature a critical process of raising awareness that today there is still a cultural legacy of colonialism in most of the policies governing Mediterranean migration relations. Power and violence are not something of the past, but still remain under 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MEDITERRANEAN THINKING IN MIGRATION STUDIES \| ZAPATA-BARRERO subtle security narratives legitimating control policies and building a hostile Mediterranean environment. As Hicks has recently highlighted in his preface: ‘violence is not some past act, to be judged by the supposed standards of the past, but an ongoing event’ (2020; xiv). In connecting the past and present, the homeostatic approach can help us to bring to surface what has been buried into a political amnesia. Homeostatic thinking applied to Mediterranean migration studies can also be a useful approach to promote an epistemological reflection of how far categories that apparently objectify real facts are just mere political ideological constructions reinforcing current Mediterranean power relations (Thalassocracy). This view can help us analyse current issues of the migration research agenda from other angles and perspectives, listening more to what is often labelled as ‘the others’ or ‘subalterns’ in Gramscian terms (often used by postcolonial studies). For instance, the same Chambers (2008, 7) suggests interpreting today's xenophobia concentrated in European countries as a failure and unwillingness to work through a still largely unconscious European past in which colonialism and Empire were (and still are) distilled into national configurations of ‘identity’, ‘culture’, ‘modernity’ and ‘progress’. In his stimulating work, Gilroy (2005; 2) argues that ‘political conflicts which characterize multicultural societies can take a very different aspect if they are understood to exist firmly in a context supplied by imperial and colonial history’. Let us say, following this application, that the accumulated effects of discriminatory acts in the past may have led to a contemporary environment that is itself discriminatory. Following Said's tradition of thought, Med-Thinking sees colonialism not as a closed past chapter but rather as a contemporary presence moulding and modifying the horizon of Mediterranean migration possibilities. P OS ITI V E TH I N K I N G I N M E D ITE R R A N E A N M I G R ATI O N R E S E A RC H The word migration is usually seen as a ‘dirty word’. Positive thinking is the practice of focusing on the good in any given situation. It is an emotional attitude that expects results that will benefit you. At the core of this notion, there is then an initial solipsism that we need to dispel. Considering the normative horizon of Med-Thinking, it is very difficult to contribute to region-making through negative inputs. For centuries, migration has fostered global trade links, shaped nations and societies, fuelled human endeavours and enabled skills and cultures to be shared across the globe. As world leaders recognized in the UN Global Compact on Migration in December 2018, migration ‘is a source of prosperity, innovation and sustainable development in our globalized world’. The tricky question is whether migration can be explored in good or bad, in pros and cons terms. Applying most of the dimensions of Med-Thinking we have already introduced, we tend to assess that these questions are not worth asking, since these binomial reference frameworks may have many answers depending on which vantage point we formulate it. If we focus on the potential answers rather than the formulation of the question, in general, it always tends to be answered in Eurocentric terms. Good and bad for whom? European states? Countries of origin? Africa/Balkans? Morocco/Lebanon? The migrant and his/her family? These questions invite us to think ethically as well (Zapata-Barrero & Yalaz, 2020). So the best way to properly target this question, and then to confirm the adequacy of this positive thought for shaping the Med-Thinking approach, is to keep the holistic and homeostatic approaches alive, as well as the epistemic solipsism proviso when answering. Apart from its intrinsic dualist nature, I also had initial doubts if this dimension needed to be included given its instrumentalist nature. The fact is that we can jump from pros to cons very easily. To enter into this debate's path, we need to carefully think as to avoid falling within a dialectical approach. This means to refrain from following the rationale that if there is a group of people that think migration is bad in terms of crime, for instance, the positive side would be that migration provides security, which is also false. The relation is much more complex and related to different forms of articulating the consequences and impact of migration. It is not the same to assess migration in positive terms from the country of reception than from the vantage point of the sending country; it is not the same to build a research design from the position of the migrant than from a particular state's interests. What is also a common trend is that this query of positive thinking on migration is always focused within a consequentialist 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 \| 9 angle. The question of whether migration is good or bad touches on a lot more than just economics. There are social, cultural and moral arguments used to debate for and against migration, as well as human security concerns raised by letting people move back and forth (Zapata-Barrero & Gabrielli, 2017). The list of pros is quite long and dispersed: increased economic output, potential transnational entrepreneurs, increased economic demand and growth, a possible solution for an ageing population, more flexible labour market, solving a skills shortage, filling undesirable job vacancies, cosmopolitan creative and innovative society. The mainstream rationale is to link the positive side with cultural, economic, social development arguments, innovation and creativity. In this case, the negative view of migration is formulated in normative terms of cohesion erosion, social conflict and economic instability and even ontological national survival. Often this debate inevitably includes discussions about stereotypes, prejudices and myths of migration. These negative misperceptions contaminate the public narrative and even affect practices and perceptions. The counter positive argument here takes the shape of providing evidence-b ased arguments countervailing these myths. For instance, providing evidence that migrants do not steal and increase criminality, do not collapse public services or occupy jobs that nationals may need. The focus on diversity advantages is already an open debate. This approach is usually used as a direct attack against the view of diversity as a direct source of conflict and disadvantages, as a ‘diversity dividend’ of private/social economic benefits (Syrett & Sepulveda, 2011). In seeking to realize the diversity dividend, city public promoters have for instance pursued a number of different strategies, from the branding of cities as vibrant, multicultural/cosmopolitan locations to attract investors, tourists, events and high skilled workers (Musterd & Murie, 2010; Rath, 2007), to the promotion of, for example, ethnic businesses (Ram & Jones, 2008), diaspora trading networks (Kitching et al., 2009; Kuznetsov & Sabel, 2006), ethnic quarters and festivals (Shaw, 2007). This literature often links diversity–a dvantages production with intercultural policies (Zapata-B arrero, 2019, Ch. 8). If we look at the Intercultural Cities Program of the Council of Europe, it incorporated a competitive call for cities in March 2015 to select the best practical initiatives on diversity advantage challenges. The way they define diversity advantage is broad, but it fits the strand we are shaping for Med-T hinking. ‘Recognizing that diversity is not a threat – it can bring competitive benefits for businesses, organizations and communities if managed competently and in the spirit of inclusion’, ‘Embracing diversity is not a gimmick for the branding of a business, organization or city but a philosophy of governance, management and decision making’ (Council of Europe, 2015; 3). The diversity–a dvantage approach of interculturalism (Wood & Landry, 2008) certainly emerges assuming the economic development hypothesis leading the debate. We are always searching for stories about the inspiring deeds of immigrants. For instance, giving visibility to artists with migrant origin, writers or even well-k nown football players with a migrant and working-class origin. Histories of successful migrants’ entrepreneurs could also belong to this Med-t hinking approach to migration studies. CO N C LU D I N G R E M A R K S FO R FU RTH E R D I S C U S S I O N . P OTE NTI A LITI E S O F TH E A PPLI C ATI O N O F TH E M E D -T H I N K I N G A PPROAC H This article is an invitation to explore the analytical, methodological and normative possibilities afforded by a Mediterranean approach for migration studies. First, it briefly presents the background premises on the historical, social and geopolitical dimensions that make the Mediterranean region uniquely positioned to develop critical thinking on a predominantly (North and West) unilateral European approach to knowledge production, cautioning at the same time against the trappings of idealized, utopian or ‘romantic yearnings’ about the Mediterranean. Second, and more importantly, the article lays out the methodological and epistemological dimensions of a Mediterranean approach: against Euro-centrism (‘epistemic solipsism’), holistic and multiscale or better, multi- layered approach; focusing on positive thinking understood as an analytical choice to purposively avoid ‘crises’, ‘threat’ and ‘security’ laden framings of migration; favouring continuity over or within change (homeostasis). 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MEDITERRANEAN THINKING IN MIGRATION STUDIES \| ZAPATA-BARRERO The fact that the Mediterranean can and should be used as a category of analysis to enhance Mediterranean migration studies is then beyond doubt. What maybe is less obvious and probably needs further discussion is the guiding normative chart of this article: how far can Med-Thinking contribute normatively to Mediterranean region-making. First of all, we may initially argue that there are probably many ways of Mediterranean region-making which are not necessarily related to migration, but to interchange commodities, knowledge and culture, economic and social circulation within all the shores of the Mediterranean. This is true. But we cannot leave aside from this list that human mobility is one of the most powerful resources for Mediterranean region-making today. This path is clearly under-researched. Med-Thinking can help to carry this on, since their various dimensions have potential to strengthen the necessary epistemological and ontological views for shaping a normative approach. One further reflection may request the role of postcolonial thinking within this Med-Thinking normative approach, since, as we have seen, it clusters almost all the four dimensions of Med-thinking. Obviously, postcolonialism provides strong arguments against epistemic solipsism and enhances the potentialities of holistic, homeostatic and positive thinking. Shaping Med-Thinking, postcolonialism can help to frame an alternative approach to Mediterranean migration research and can be at the foundation for innovative thinking in Mediterranean migration studies. Med-Thinking could be interpreted as a regional application of postcolonialism applied to migration studies. Given this conceptual background, the current Mediterranean system of filters and blockage of population mobility can be interpreted as different ways of putting a stitch in the wheel of region-making. Mobility as a factor of region-making is not new. In fact, it is at the core of the constructivist approach of identity building followed by European citizenship (Favell & Recchi, 2009). Focusing on transnational relations, there are so many themes related to mobility and migration altogether: transnational relations of migrants, trans-local relations through cities, civil society actors networking in the Mediterranean to save lives, but also civil society relations through different intercultural programmes, such as those promoted by the Anna Lindh Foundation,5. whose mission is to deploy the cultural pillar of the Barcelona Process. The normative dimension that brings us Med-Thinking may have many understandings and policy applications. The easiest way to begin is to look at identifying the key features of Mediterranean migration that may help to understand the specificities of this geographical area in a global context. For instance, we can identify key dimensions of this particular region, such as the migrants and associated moving population (such as refugees) and the long array of institutional, collective and individual practices around them, transnational dimensions that may take not only individual but also multiscale dimensions (translocalism, transregionalism, etc.), migrant transnational entrepreneurship, the presence of unaccompanied children, migrants with an irregular status, but also the colonial legacy still present for understanding European policies towards Mediterranean migration. Not only issues related to cities and NGO networks, corridors but also shared views and concerns, and even cosmopolitan values related to an intercultural Mediterranean, may also help to regionalize topics instead of viewing it within a territorial vacuum. We can even go on to the still under-researched issues of Mediterranean citizenship without falling to the ‘Mediterraneanism’, namely looking at a ‘Mediterranean race’ mostly used by Mussolini's fascism. Here citizenship studies may help us to see if a regional understanding of citizenship is possible. This seems to follow a communitarian view of a region, rather than the current liberal one governed by Mediterranean unipolar geo-migration politics. Human mobility remains a key feature of Mediterranean region-making. I follow the perspective of considering mobility essential in the making of societies (Cresswell, 2006; Urry, 2007). In a similar vein, trans-M ed human mobility connecting Mediterranean cities could be considered another factor building the Med-region (Zapata-B arrero, 2020b). I am very sympathetic to the particular debate of the Mediterranean scale of governance, putting city networks at the core (Kramsch, 2004), transnational urbanism (Smith, 2005) and translocal spatial geographies (Brickell & Datta, 2011) applied at city levels (Christou, 2011) or simply translocality (Greiner & Sakdapolrak, 2013), which I understand both as physical and virtual. It is not only a way to focus on the link between space/people, space/governance and then an invitation to rethink the role of space and 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 \| 11 territory in migration governance, but also to stress the new Med-spatial geographies. Here we can also place the debate that focuses on translocalism as a driver for new values making, and the promotion of a new sense of belonging based on diverse city attachments (Smith & Eade, 2008), and even to frame this debate within cosmopolitanism (Smith, 1999). As Cooke asserts and from which we get inspired, Med-Thinking can be summarized as an approach that seeks to homogenize diversity (Cooke, 1999, 292). Within this folder we can place a list of factors that may help region- making: reiterated relations, routines, patterns, fixities and how these factors interact in a comprehensive way to shape regional ecological system. We can take the geographical or the international relations description which points more on the geopolitical dimension of this geographical area (Collyer, 2016). What I have tried to defend here is a path for building a normative approach (Med-Thinking) capable of contributing to the development of Mediterranean migration studies. This debate is much needed since it will allow us to place on to the Mediterranean migration research agenda the importance of ontological and epistemological questions in the process of building knowledge, insisting that the variety of legitimate approaches and views is probably what makes the Mediterranean unique. As a last resort, Med-T hinking may contribute to be more sensitive to the concepts that shape Mediterranean migration studies and make visible their inner hierarchies and power differentials within the Mediterranean region. This conceptual thinking (Zapata-B arrero et al., 2022) applied to Med-T hinking would probably help us to decentralize Mediterranean migration knowledge production and governance. To raise awareness of how far the EU and the power of its states shape an imperialist Thalassocracy governing the Mediterranean, and to place into the agenda other Mediterranean fixities such as cities, which are older than states in the Mediterranean (Leontidou, 2019), it can also invite us to carve a much more polyhedric approach and look at other sources of information and other theoretical frameworks that may contribute to build knowledge around Mediterranean migration. The list of potential research avenues applying the Med-t hinking approach in migration studies is then vast and promising. AC K N OW L E D G E M E N T S This work is supported by a 3-year (2019–2022) Erasmus+ Jean Monnet Network Program (Project Reference: 611260-EPP-1-2019-1-ES-EPPJMO-NETWORK) entitled ‘Mapping European Mediterranean Migration Studies’ (Acronym: EUMedMi). DATA AVA I L A B I L I T Y S TAT E M E N T In references I have placed all DOI for articles and mentioned and given all retrieved from website when it is free access. ORCID Ricard Zapata-Barrero https://orcid.org/0000-0002-3478-1330 E N D N OT E S 1. This approach was first articulated in the editorial of a special issue I edited for Comparative Migration Studies (Zapata- Barrero, 2020a), and then further developed in a EuroMedMig Working-Paper (Zapata-Barrero, 2020b). It is inspired by the seminal work of Cooke (1999) and the basic development done by Chambers (2008) in its stimulating book on Mediterranean Crossings. 2. Some initial works are devoting increasing attention in multi-scale settings, and the regional arrangements is receiving prominent interest; however, it is still an under-research scale of analysis in migration governance. See some seminal works in Bisong (2021), Geddes et al. (2019), Lavenex and Piper (2021) and Rother and Piper (2015). 3. In November 1995, following a European Council decision, a Euro-Mediterranean Conference of Foreign Affairs Ministers was held in the Spanish city of Barcelona. It marked the launch of the Euro-Mediterranean Partnership (EuroMed), also known as the Barcelona Process for short, after the name of the city in which the decision was taken. It was the EU's first 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MEDITERRANEAN THINKING IN MIGRATION STUDIES \| ZAPATA-BARRERO comprehensive policy for the region. There is a huge literature on Euromed policy and the EuroMed partnerhip initiative. I just quote two that seems to be better offer an introductory descriptive overview: Gillespie (2013) and Jünemann (2004). 4. Concretely it was first emphasized in May 2011 in a joint communication on ‘a new response to a changing Neighborhood’ European Commission and High representative Joint communication, a new response to a changing Neighborhood, Brussels, 25/05/2011, COM(2011) 303. 5. http://www.annalindhfoundation.org/ REFERENCES Abulafia, D. (2014) Thalassocracies. In: Horden, P. & Kinoshita, S. (Eds.) A companion to Mediterranean history. Chichester: Wiley Blackwell, pp. 139–153. Adamson, F.B. & Tsourapas, G. (2020) The migration state in the global south: nationalizing, developmental, and neoliberal models of migration management. International Migration Review, 54(3), 853–882. Available from: https://doi. org/10.1177/0197918319879057 Arendt, H. (1976) Totalitarianism: part three of the origins of totalitarianism. San Diego, CA: A Harvest Book/Harcourt. Baldwin-Edwards, M., Blitz, B.K. & Crawley, H. (2018) The politics of evidence-based policy in Europe’s ‘migration crisis’. Journal of Ethnic and Migration Studies, 45(12), 2139–2155. Available from: https://doi.org/10.1080/1369183X.2018.1468307 Biggs, R., Schlüter, M. & Schoon, M.L. (2016) Principles for building resilience: sustaining ecosystem services in social- ecological systems. Cambridge: Cambridge University Press. Bisong, A. (2021) Invented, invited and instrumentalised spaces: conceptualising non-state actor engagement in regional migration governance in West Africa. Journal of Ethnic and Migration Studies, 1–19. Available from: https://doi. org/10.1080/1369183X.2021.1972570 Brickell, K. & Datta, A. (Eds.) (2011) Translocal geographies: spaces, places, connections. Farnham: Ashgate. Chambers, I. (2008) Mediterranean crossings: the politics of an interrupted modernity. Durham and London: Duke University Press. Chambers, I. & Curti, L. (Eds.) (1996) The postcolonial question: common skies, divided horizons. London: Routledge. Chambers, I. & Curti, L. (2008) Migrating modernities in the Mediterranean. Postcolonial Studies, 11(4), 387–399. Available from: https://doi.org/10.1080/13688790802456077 Christou, A. (2011) Translocal spatial geographies. Multi-sited encounters of Greek migrants in Athens, Berlin, and New York. In Brickell, K. & Datta, A. (Eds.) Translocal geographies: spaces, places, connections. Farnham: Ashgate or Simply Translocality, pp. 145–162. Collyer, M. (2016) Geopolitics as a migration governance strategy: European Union bilateral relations with Southern Mediterranean countries. Journal of Ethnic and Migration Studies, 42(4), 606–624. Available from: https://doi. org/10.1080/1369183X.2015.1106111 Cooke, M. (1999) Mediterranean thinking: from Netizen to Medizen. The Geographical Review, 89(2), 290–3 00. Available from: https://doi.org/10.1111/j.1931-0 846.1999.tb00220.x Council of Europe (2015) Diversity advantage challenge. Council of Europe. Available from: https://edoc.coe.int/en/living- together-diversity-and-freedom-in-europe/6555-diversity-advant age-challenge-selected-initiatives.html [Accesed 3rd February 2022] Cresswell, T. (2006) On the move. Mobility in the modern Western world. London: Routledge. Damasio, A. & Damasio, H. (2016) Exploring the concept of homeostasis and considering its implications for economics. Journal of Economic Behaviour & Organization, 126, 125–129. Available from: https://doi.org/10.1016/j.jebo.2015.12.003 EU Parliament (2015) On the situation in the Mediterranean and the need for a holistic EU approach to migration (2015/2095(INI)). Available from: https://www.europarl.europa.eu/doceo/document/A-8-2016-0 066_EN.html Favell, A. & Recchi, E. (Eds.) (2009) Pioneers of European integration: citizenship and mobility in the EU. Cheltenham: Edward Elgar. Fekete, L. (2018) Migrants, borders and the criminalisation of solidarity in the EU. Race and Class, 59(4), 65–83. Available from: https://doi.org/10.1177/0306396818756793 Foucault, M. (2004) Des espaces autres. Empan, 2(2), 12–19. Available from: https://doi.org/10.3917/empa.054.0012 Gallant, T.W. (2016) Afterword: writing Mediterranean diasporas after the transnational turn. In: Isabella, M. & Zanou, K. (Eds.) Mediterranean diasporas: politics and ideas in the long 19th century. London: Bloomsbury, pp. 205–209. Garelli, G., Sciurba, A., & Tazzioli, M. (2018) Introduction: Mediterranean movements and the reconfiguration of the military-h umanitarian border in 2015. Antipode, 50(3), 662–672. Available from: https://doi.org/10.1111/ anti.12385 Garelli, G., & Tazzioli, M. (2020) Rescuing, kidnapping, and criminalising. Migration containment in the Mediterranean. International Journal of Migration and Border Studies, 6(4), 1. Geddes, A., Espinoza, M.V., Hadj-Abdou, L. & Brumat, L. (2019) The dynamics of regional migration governance. Northampton, MA: Edward Elgar. 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 12 \| 13 Giaccaria, P. & Minca, C. (2011) The Mediterranean alternative. Progress in Human Geography, 35(3), 345–365. Available from: https://doi.org/10.1177/0309132510376850 Gillespie, R. (Ed.) (2013) The Euro-Mediterranean partnership: political and economic perspectives. Oxon: Routledge. Gillespie, R. & Volpi, F. (Eds.) (2019) Routledge handbook of Mediterranean politics. London: Routledge. Gilroy, P. (2005) Postcolonial melancholia. New York: Columbia University Press. Global Peace Index (2018) Measuring peace in a complex world. Sydney, NSW: Institute for Economics & Peace. Goldber, D.T. (2006) Racial Europeanization. Ethnic and Racial Studies, 29(2), 331–364. Available from: https://doi. org/10.1080/01419870500 465611 Greiner, C. & Sakdapolrak, P. (2013) Translocality: concepts, applications and emerging research perspectives. Geography Compass, 7(5), 373–384. Available from: https://doi.org/10.1111/gec3.12048 Heath, J. (2020) Methodological individualism. In: Zalta, E.N. (Ed.) The stanford encyclopedia of philosophy (summer edition). Available from: https://plato.stanford.edu/archives/sum2020/entries/methodological-individualism/ [Accesed 3rd February 2022]. Hicks, D. (2020) The British museums the Benin bronzes, colonial violence and cultural restitution. London: Pluto Press. Jones, M. & Paasi, A. (2013) Guest editorial: regional world(s): advancing the geography of regions. Regional Studies, 47(1), 1–5. Available from: https://doi.org/10.1080/003434 04.2013.746437 Jünemann, A. (Ed.) (2004) Euro-Mediterranean relations after September 11: international, regional, and domestic dynamics. London: Frank Cass. Kirtsoglou, E. & Tsimouris, E. (2016) “Il était un petit navire”: the refugee crisis, neo-orientalism, and the production of radical alterity. Journal of Modern Greek Studies, Occasional paper 9. Available from: https://www.press.jhu.edu/sites/ default/files/MGS_OP09.pdf?utm_medium =papers &utm_source =jhup&utm_campaign=MGSOP&utm_content=9 [Accesed 3rd February 2022]. Kitching, J., Smallbone, D. & Athayde, R. (2009) Ethnic diasporas and business competitiveness: minority-owned enterprises in London. Journal of Ethnic and Migration Studies, 35(4), 689–705. Available from: https://doi.org/10.1080/13691 830902765368 Kramsch, O.T. (2004) Towards a Mediterranean scale of governance. Twenty-first-century urban networks across the ‘Inner Sea’. In: Kramsch, O.T. & Hooper, B. (Eds.) Cross-border governance in the European Union. London: Routledge, pp. 191–208. Kuznetsov, Y. & Sabel, C. (2006) International migration of talent, diaspora networks and development: overview of main issues. In: Kuznetsov, Y. (Ed.) Diaspora networks and the international migration of skills. Washington, DC: World Bank Institute, pp. 3–20. Lanese, N. (2019) What is homeostasis? Live Science, July 15. Available from: https://www.livescience.com/65938 -h omeos tasis.html#:-: text=Homeos tasis is the abilit y to.proces s energy and ultimately survive [Accesed 3rd February 2022]. Lavenex, S. & Piper, N. (2021) Regions and global migration governance: perspectives ‘from above’, ‘from below’ and ‘from beyond’. Journal of Ethnic and Migration Studies, 1–18. Available from: https://doi.org/10.1080/1369183X.2021.1972564 Leontidou, L. (2019) Mediterranean city. In: Orum, A.M. (Ed.) The Wiley-Blackwell encyclopedia of urban and regional studies. Chichester: Wiley Blackwell, pp. 1176–1184. Martin, J. (2009) Phase space: geography, relational thinking, and beyond. Progress in Human Geography, 33(4), 487–506. Available from: https://doi.org/10.1177/0309132508101599 Mayblin, L. & Turner, J. (2021) Migration studies and colonialism. Cambridge: Polity Press. Michael, C.M. (2021) Migration and the crisis of democracy in contemporary Europe. Cham: Palgrave Macmillan. Miller-Keane Encyclopedia (2003) Homeostatic mechanism. In: Miller-Keane encyclopedia and dictionary of medicine, nursing, and allied health, 7th edition. Available from: https://medical-dictionary.thefreedictionar y.com/Homeostatic+mechanism [Accesed 3rd February 2022]. Musterd, S. & Murie, A. (Eds.) (2010) Making competitive cities. Chichester: Wiley. Natter, K. (2018) Rethinking immigration policy theory beyond ‘Western liberal democracies’. Comparative Migration Studies, 6(1), 4. Available from: https://doi.org/10.1186/s40878-018-0 071-9 Paasi, A. & Metzger, J. (2017) Foregrounding the region. Regional Studies, 51(1), 19–3 0. Available from: https://doi. org/10.1080/003434 04.2016.1239818 Pace, M. (2006) The politics of regional identity: meddling with the Mediterranean. London, New York: Routledge. Pace, M. & Roccu, R. (2020) Imperial Pasts in the EU’s approach to the Mediterranean. Interventions, 22(6), 671–685. Available from: https://doi.org/10.1080/1369801X.2020.1749702 Paoletti, E. (2011) The migration of power and North-South inequalities. Basingstoke: Palgrave Macmillan. Pécoud, A. (2020) Death at the border: revisiting the debate in Light of the Euro-Mediterranean migration crisis. American Behavioral Scientist, 64(4), 379–388. Available from: https://doi.org/10.1177/000276 4219882987 Ram, M. & Jones, T. (2008) Ethnic minority business in the UK: a review of research and policy developments. Environment and Planning C: Government and Policy, 26(2), 352–374. Rath, J. (Ed.) (2007) Tourism, ethnic diversity and the city. New York: Routledge. 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License MEDITERRANEAN THINKING IN MIGRATION STUDIES \| ZAPATA-BARRERO Rodolfo, K. (2000) “What is homeostasis?”, Scientific American (January 31th). Available from: https://www.scientificameri can.com/article/what-is-homeostasis/ [Accesed 3rd February 2022]. Rother, S. & Piper, N. (2015) Alternative regionalism from below: democratizing ASEAN's migration governance. International Migration, 53(3), 36–49. Available from: https://doi.org/10.1111/imig.12182 Shaw, S. (2007) Ethnic quarters in the cosmopolitan-creative city. In: Richards, G. & Wilson, J. (Eds.) Tourism, creativity and development. London: Routledge, pp. 189–200. Shoroff, F.H. (2011) Conceptualizing holism in international interdisciplinary critical perspective. Social Theory & Health, 9(3), 244–255. Available from: https://doi.org/10.1057/sth.2011.6 Smith, M.P. (1999) Transnationalism and the city. In: Beauregard, R. & Body-Gendrot, S. (Eds.) The urban moment: cosmopolitan essays on the late 20th century city. London: Sage Publications, pp. 119–139. Smith, M.P. (2005) Transnational urbanism revisited. Journal Ethnic and Migration Studies, 31(2), 235–244. Available from: https://doi.org/10.1080/13691830420 00339909 Smith, M.P. & Eade, J. (2008) Transnational ties: cities, migrations and identities. In: Smith, M.P. & Eade, J. (Eds.) Transnational ties: cities, migrations and identities. London: Transaction, pp. 3–13. Socialists and Democrats (2015) Results of the survey on the “Situation in the Mediterranean and the need for a holistic approach to migration”. S&D Research Paper, European Parliament December. Stierl, M. (2016) A sea of struggle –activist border interventions in the Mediterranean Sea. Citizenship Studies, 20(5), 561– 578. https://doi.org/10.1080/13621025.2016.1182683 Syrett, S. & Sepulveda, L. (2011) Realising the diversity dividend: population diversity and urban economic development. Environment and Planning A, 43(2), 487–504. Available from: https://doi.org/10.1068/a43185 Terry, P. (2016) Three principles of holistic thinking. Available from: https://www.linkedin.com/pulse/three-principles-holis tic-thinking-paul-terry [Accesed 3rd February 2022]. Thiollet, H. (2011) Migration as diplomacy: labor migrants, refugees, and Arab regional politics in the oil-rich countries. International Labor and Working-Class History, 79(1), 103–121. Tsourapas, G. (2018) The politics of migration in modern Egypt: strategies for regime survival in autocracies. Cambridge: Cambridge University Press. United Nations (2018) Global compact for safe, orderly and regular migration. Resolution adopted by the General Assembly on 19 December 2018. Available from: https://www.iom.int/global-compact-migration [Accesed 3rd February 2022]. Urry, J. (2007) Mobilities. Cambridge: Polity. Varro, K. & Lagendijk, A. (2012) Conceptualizing the region – in what sense relational? Regional Studies, 47(1), 18–28. Available from: https://doi.org/10.1080/003434 04.2011.602334 Wilson, C. & Airey, P. (1999) How can a homeostatic perspective enhance demographic transition theory? Population Studies, 53(2), 117–128. Available from: https://doi.org/10.1080/00324720308 076 Wolff, S. & Hadj-Abdou, L. (2019) Mediterranean migrants and refugees: historical and political continuities and discontinuities. In: Gillespie, R. & Volpi, F. (Eds.) Routledge handbook of Mediterranean politics. London; New York: Routledge, 2018, pp. 382–393. Wood, P.H. & Landry, C. (2008) The intercultural city: planning for diversity advantage. London: Earthscan. Zapata-Barrero, R. (2019) Intercultural citizenship in the post-multicultural era. London: Sage Swifts Publishing. Zapata-Barrero, R. (2020a) Editorial: “Mediterranean thinking” for mapping a Mediterranean migration research agenda. Comparative Migration Studies, 8(6), 1–18. Available from: https://doi.org/10.1186/s40878-019-0170-2 Zapata-Barrero, R. (2020b) Rescaling Mediterranean migration governance: setting a research agenda that establishes the centrality of cities for region-making. EuroMedMig Working Papers Series, Number 3 (June). Available from: http://hdl. handle.net/10230/4 4978 [Accesed 3rd February 2022]. Zapata-Barrero, R. & Faustini, L. (Eds.) (2019) Mediterranean migration research: variable focal length. Comparative Migration Studies, 7(26). Zapata-Barrero, R. & Gabrielli, L. (2017) Ethics and the securitization of migration: revisiting the current policy framework. In: Bourbeau, P. (Ed.) Handbook on migration and security. Cheltenham: Edward Elgar, pp. 125–143. Zapata-B arrero, R., Jacobs, D. & Kastoryano, R. (Eds.) (2022) Contested concepts in migration studies. London: Routledge. Zapata-Barrero, R. & Yalaz, E. (2020) Qualitative migration research ethics: a roadmap for migration scholars. Qualitative Research Journal, 20(3), 269–279. Available from: https://doi.org/10.1108/QRJ-02-2020-0 013 How to cite this article: Zapata-Barrero, R. (2022) Mediterranean thinking in migration studies: A methodological regionalism approach. International Migration, 00, 1–14. Available from: https://doi. org/10.1111/imig.12995 14682435, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/imig.12995 by Readcube (Labtiva Inc.), Wiley Online Library on [20/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 14
https://openalex.org/W2549000275	https://kclpure.kcl.ac.uk/ws/files/60967013/journal.pone.0165467.PDF	English	null	Central Coherence in Eating Disorders: A Synthesis of Studies Using the Rey Osterrieth Complex Figure Test	PloS one	2,016	cc-by	7,213	Citation for published version (APA): Lang, K., Roberts, M., Harrison, A., Lopez, C., Goddard, E., Khondoker, M., Treasure, J., & Tchanturia, K. (2016). Central Coherence in Eating Disorders: A Synthesis of Studies Using the Rey Osterrieth Complex Figure Test. PLoS ONE, 11(11), Article e0165467. https://doi.org/10.1371/journal.pone.0165467 Citation for published version (APA): Lang, K., Roberts, M., Harrison, A., Lopez, C., Goddard, E., Khondoker, M., Treasure, J., & Tchanturia, K. (2016). Central Coherence in Eating Disorders: A Synthesis of Studies Using the Rey Osterrieth Complex Figure Test. PLoS ONE, 11(11), Article e0165467. https://doi.org/10.1371/journal.pone.0165467 Citing this paper Pl h C t g t s pape Please note that where the full-text provided on King's Research Portal is the Author Accepted Manuscript or Post-Print version this may differ from the final Published version. If citing, it is advised that you check and use the publisher's definitive version for pagination, volume/issue, and date of publication details. And where the final published version is provided on the Research Portal, if citing you are again advised to check the publisher's website for any subsequent corrections. Citation for published version (APA): Lang, K., Roberts, M., Harrison, A., Lopez, C., Goddard, E., Khondoker, M., Treasure, J., & Tchanturia, K. (2016). Central Coherence in Eating Disorders: A Synthesis of Studies Using the Rey Osterrieth Complex Figure Test. PLoS ONE, 11(11), Article e0165467. https://doi.org/10.1371/journal.pone.0165467 General rights General rights Copyright and moral rights for the publications made accessible in the Research Portal are retained by the authors and owners and it is a condition of accessing publications that users recognize and abide by the legal requirements associ ral rights for the publications made accessible in the Research Portal are retained by the authors and/or other copyright condition of accessing publications that users recognize and abide by the legal requirements associated with these right •Users may download and print one copy of any publication from the Research Portal for the purpose of private study or research. •You may not further distribute the material or use it for any profit-making activity or commercial gain •You may freely distribute the URL identifying the publication in the Research Portal •Users may download and print one copy of any publication from the Research Portal for the purpose of private study •You may not further distribute the material or use it for any profit-making activity or commercial gain y y p g y •You may freely distribute the URL identifying the publication in the Research Portal Take down policy If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we will remove access to the work immediately and investigate your claim. p y If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we w the work immediately and investigate your claim. Download date: 24. Oct. 2024 RESEARCH ARTICLE Results Poor global processing was observed across all current eating disorder sub-groups and in unaffected relatives. There was no difference in performance between recovered AN and HC groups. Data Availability Statement: Due to potentially identifying patient information, data is available from the first and last author on request. Please contact either kate.lang@kcl.ac.uk or kate. tchnaturia@kcl.ac.uk. Method Data was collated from 984 participants in total. Anorexia Nervosa, Bulimia Nervosa, recov- ered Anorexia Nervosa, unaffected family members and healthy controls were compared using the Rey Figure. Copyright: © 2016 Lang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. OPEN ACCESS Citation: Lang K, Roberts M, Harrison A, Lopez C, Goddard E, Khondoker M, et al. (2016) Central Coherence in Eating Disorders: A Synthesis of Studies Using the Rey Osterrieth Complex Figure Test. PLoS ONE 11(11): e0165467. doi:10.1371/journal.pone.0165467 Katie Lang1, Marion Roberts1, Amy Harrison2, Carolina Lopez1,3, Elizabeth Goddard1, Mizan Khondoker1, Janet Treasure1, Kate Tchanturia1,4* Katie Lang1, Marion Roberts1, Amy Harrison2, Carolina Lopez1,3, Elizabeth Goddard1, Mizan Khondoker1, Janet Treasure1, Kate Tchanturia1,4* 1 King’s College London (KCL), Psychological Medicine, Institute of Psychiatry, Psychology & Neuroscience, London, United Kingdom, 2 Regents School of Psychotherapy & Psychology, Faculty of Humanities, Arts & Social Sciences, Regent’s University, London, United Kingdom, 3 Department of Pediatrics and Child Surgery East, Faculty of Medicine, University of Chile, Santiago, Chile, 4 Illia State University, Department of Psychology, Tbilisi, Georgia a1111 * kate.tchanturia@kcl.ac.uk Background Large variability in tests and differences in scoring systems used to study central coherence in eating disorders may lead to different interpretations, inconsistent findings and between study discrepancies. This study aimed to address inconsistencies by collating data from several studies from the same research group that used the Rey Osterrieth Complex Figure Test (Rey Figure) in order to produce norms to provide benchmark data for future studies. Editor: Cosimo Urgesi, Universita degli Studi di Udine, ITALY Received: May 6, 2016 Accepted: October 12, 2016 Published: November 2, 2016 * kate.tchanturia@kcl.ac.uk Central Coherence in Eating Disorders: A Synthesis of Studies Using the Rey Osterrieth Complex Figure Test Katie Lang1, Marion Roberts1, Amy Harrison2, Carolina Lopez1,3, Elizabeth Goddard1, Mizan Khondoker1, Janet Treasure1, Kate Tchanturia1,4* Introduction Research conducted with individuals with Anorexia Nervosa (AN) has highlighted the pres- ence of an inefficient cognitive processing style [1,2]. This is now thought to be a shared char- acteristic (to varying degrees) across eating disorder sub-types, and a likely contributor to their pathogenesis [3]. Competing Interests: The authors have declared that no competing interests exist. Central coherence is one such area of interest in the neuropsychological study of eating dis- orders (ED). Historically, central coherence is grounded in Gestalt psychology, which hypothe- sises that integrated coherent structures form the basis of processing and perception [4]. It also posits that the structured whole is different from the sum of the configural parts. Following gestalt theory, Navon’s global precedence hypothesis theorised that information processing fol- lows a hierarchical network, from global structures to more local structures, whereby global ele- ments take precedence. Importantly, this theory suggests that global and local processing are not independent, and interact to provide the whole unit [5]. Research has highlighted that individuals with eating disorders (ED) may not process infor- mation in the same hierarchical fashion describedabove. There is a wealth of research demon- strating that adults with AN have poor global processing (for a systematic reviews see (Lang et al., 2014; Lopez et al., 2008)), often with superior detail focused processing [6]. This profile of weak central coherence is also present once weight has been restored and in children and adolescents, albeit in an attenuated form [7–9]. Furthermore, this processing style seems to have a familial component, as unaffected relatives of those with AN, such as mothers and sis- ters, also demonstrate poor global processing [9–11]. Although there is less research, studies with individuals with Bulimia Nervosa (BN) have also demonstrated poor global processing in comparison to healthy controls (HCs); suggesting that this processing style is a trans-diagnostic characteristic amongst EDs [9,12,13]. Though poor global processing does not seem to be specific to the EDs and can be seen as a common mechanism across other psychiatric disorders such as Autism Spectrum Disorder [14], Schizophrenia [15] and Obsessive Compulsive Disorder [16], it does appear to be present most consistently in ED, and certainly the pattern of heritability of these traits seems stronger [17,18]. Central Coherence in Eating Disorders: A Data Synthesis the NIHR or the Department of Health. Kate Tchanturia would also like to thank the Swiss Anorexia Foundation (34-16). the NIHR or the Department of Health. Kate Tchanturia would also like to thank the Swiss Anorexia Foundation (34-16). Conclusions This is the largest dataset reported in the literature and supports previous studies implicat- ing poor global processing across eating disorders using the Rey Figure. It provides robust normative data useful for future studies. Funding: The authors would like to thank the National Institute for Health Research (NIHR) [Mental Health Biomedical Research Centre and/or Dementia Biomedical Research Unit] at South London & Maudsley NHS Foundation Trust for financial support. The views expressed are those of the authors and not necessarily those of the NHS, 1 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Participants & procedure Data were drawn from six previously published studies and two unpublished datasets carried out within the Eating Disorders Research Unit, King’s College London [6,7,9,21–23]. The col- lated dataset consisted of 984 female participants (age ranging between 18–65 years) with either AN (N = 364), BN (N = 125), or individuals who had recovered from AN (N = 107), unaffected relatives (N = 85 mother and N = 30 sisters) of individuals with AN, and HCs (N = 273). Data were collected between 2007 and 2014. In all studies, participants with an ED were recruited from the specialist ED serviceof the South London and Maudsley NHS Trust and the community. HCs were recruited via advertisements in the local community. Semi-structuredinterviews were used across the different studies, such as the SCID [24] and the EATATE [25] phenotype interviewto determine the different ED diagnoses of the participants. Individuals with AN (either restricting subtype or binge/purge subtype) were included in the studies if they had a body mass index (BMI) of 17.5 or below. The BN group consisted of individuals with a binge/purge frequency of more than twice a week for the past three months in line with DSM-IV criteria. The recovered group consisted of female participants who had recovered from AN and reported restored, regular menstruation for at least the previous year, did not report clinically significant scores (4 or above) on the Eating Disorders Examination Questionnaire [26], and had a BMI of 18.5 or above for at least the previous year. Unaffected relatives were N = 85 mothers of adolescent daughters with AN and N = 30 sis- ters of adult sisters with AN with no current or prior history of an ED or other psychiatric ill- ness themselves. There were no significant differences on any of the indices of the Rey between unaffected mothers or sisters and therefore the two groups were collapsed into one ‘unaffected relatives’ group. HCs had no current or prior history of psychiatric illness. They were excluded if they reported a family history of an ED. Exclusion criteria for all groups were a learning difficulty, neurological impairment, or psychosis. All studies had ethical approval from the NHS national committee (ethics numbers: 020/05; 12/LO/2015; CREC/07/08-67, 08/H0606/58; 08/H0306/ 66), and all participants had given written informed consent. Measures. Demographic information: Each of the collated studies had used a questionnaire to collect demographic information. Introduction A current methodologicalproblem in the field of neuropsychology of ED is the large vari- ability in the number of tests used to measure central coherence together with differences in scoring methods, which may be contributing to conflictingresults between studies. It is there- fore of upmost importance that future studies employ robust neuropsychological tests and valid scoring methods to allow for accurate assessment of central coherence in ED. The Rey Osterrieth Complex Figure Test (Rey Figure, [19]) is a popular neuropsychological measure of central coherence used in the ED and wider psychiatric field. It is a pencil and paper task whereby the participant is asked to make a direct copy of a complex figure. The way in which the participant draws the figure can offer insights into their processing style. Histori- cally, participants were asked to make a direct copy of the shape and also one from memory and both the copy and the delayed recall were used to measure central coherence. However, the most direct measure of central coherence is now considered to be collected from the direct copy using Booth’s (2006) scoring system [20]. This scoring system incorporates both the order in which the participant chooses to draw the elements (whether preference is shown to global or detailed elements) and the style in which they are drawn in (fragmented or coherent). Full administration and scoring instructions can be found at www.katetchanturia.com (http:// www.katetchanturia.com/#!clinical-work-packages—protocols/cff4). Studies employing the Rey Figure have consistently demonstrated poor global processing in AN, BN, weight restored AN and also healthy unaffected relatives of those with AN compared PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 2 / 11 Central Coherence in Eating Disorders: A Data Synthesis to HCs [2,7–9,12]. With this in mind, the present study aims to collate data on the Rey Figure in ED from several studies containing participants with AN, BN, recovered AN, unaf- fected family members and healthy controls, using Booth’s (2006) scoring system. Collating data in this fashion rather than performing a meta-analysis will allow us to provide normative comparison data for future studies. A secondary aim of the study is to investigate possible pre- dictors of central coherence across the different ED sub-types. PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Participants & procedure This included: age, medication status, duration of illness, number of years of education and ethnicity. The Structured Clinical Interview for DSM-IV (SCID, [24]) screening questionnaire was implemented to screen for the presence of possible psychiatric illnesses within both the ED and HC groups. The SCID is a semi-structuredinterview that probes for symptoms of depression, anxiety, OCD, substance abuse and ED pathology. It is widely used and recommended for research protocols in psychiatric populations. EATATE phenotype interview [25]: A semi-structureddiagnostic interview that assesses life time course of psychiatric disorders. 3 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Central Coherence in Eating Disorders: A Data Synthesis Self-reported anxiety and depression were measured using either the Hospital Anxiety and Depression Scale (HADS [27], 89% of participants) or the Depression, Anxiety and Stress Scale (DASS [28] 11% of participants). As different measures of anxiety and depression were employed between the studies, outcome scores were standardised by calculating z scores for the depression and anxiety scale of each measure. Rey Osterrieth Complex Figure Test [19]: The Rey Figure is a pencil and paper neuropsycho- logical task commonly used in both the ED and wider psychiatric field as a measure of gestalt processing. The drawing strategy adopted by the participant is used as a measure of central coherence. Given the task instructions to copy the figure “as carefully as you can”, performance in terms of accuracy is benefitted from taking a more global approach. Participants are asked to start drawing using a black coloured pencil, and the experimenter changes the coloured pen- cil each time one of the 18 elements are completed or if the participant begins drawing a new element without completing the previous one. The specific order in which the pencils are used (black, green, purple, brown, dark blue, pink, light blue, red, yellow and orange), aids with scor- ing of the Rey Figure once the drawing has been completed. The administration of all partici- pants Rey Figures were also video recorded. This further aids with the reliability of the scoring, and allows for each Rey Figure to be scored by a second and independent researcher. The Rey Figure was scored according to a slightly modifiedversion of Booth’s (2006) scor- ing method (amendments to the scoring method were discussed with R. Booth prior to admin- istration). Participants & procedure The order index was modifiedso that the detail of the diagonal line of element 18 was not required. Two modifications to the style index were made. Firstly, whereas full exten- sion of the vertical line (element 5) was previously required, this was modifiedso that only extending the line above or below would result in full marks. Secondly, the requirement to also draw element 15 with element 13 was also eliminated. This method incorporates both the order in which the participant chooses to draw the elements (whether preference is shown to global or detailed elements) and the style in which they are drawn in (fragmented or coherent). Order index and style index are computed and added to give the Central Coherence Index (CCI). Booth’s (2006) scoring method is hierarchical, whereby higher scores are awarded when preference is shown to global elements. Higher scores on this measure are indicative of better global or holistic processing. An outline of Booth’s scoring system is provided in Table 1 and Fig 1. A second researcher co-rated 10% of the Rey figure data for each dataset and the reliabil- ity of the datasets ranged from 0.71–0.97. Statistical analysis. Histograms were used to assess the data for normality. Both the order and the Central Coherence Indices of the Rey Figure were negatively skewed and so reverse log transformations (whereby scores are first reversed and then a log transformation is applied) were employed to normalise the data when testing for group differences. Regression analysis was employed to investigate group differences in order index, style index and CCI. A secondary regression analysis was employed to assess potential predictors of CCI within each ED group. Due to the testing of multiple variables, we used a more conserva- tive significance level of 1%, to minimise Type I errors. Cohen’s d was calculated to provide effect sizes (>0.5 = moderate; >0.8 = large). Data was analysed using the statistical package IBM SPSS version 22.00. Demographic data Table 2 displays demographic information for each group. In general, clinical groups were in their mid-late 20s, and had a long duration of illness (more than 9 years). Those with current AN were significantly underweight (mean BMI 15.7) while all other groups had a BMI within 4 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Central Coherence in Eating Disorders: A Data Synthesis Table 1. Scoring system for Rey Figure according to Booth (2006). Category Element Description Global element (Score = 4) 2 Large rectangle 13 Side of large triangle attached to large rectangle Global internal element (Score = 3) 3 Diagonal cross 4 Horizontal midline of large rectangle 5 Vertical midline of large rectangle 16 Horizontal line with sides of large triangle Local perimeter element (Score = 1) 1 Vertical cross 9 Small triangle above large rectangle 14 Diamond 17 Horizontal cross 18 Square attached to large rectangle Local internal element (Score = 0) 6 Small rectangle 7 Small horizontal line above small rectangle 15 Vertical line within side of large triangle 8 Four parallel lines 10 Small vertical line with large rectangle 11 Circle with three dots 12 Five parallel lines doi:10.1371/journal.pone.0165467.t001 Table 1. Scoring system for Rey Figure according to Booth (2006). the normal range. There was a significant different in age between the HC group and the AN recovered group and unaffected relative groups, therefore analyses were adjusted for age. Central Coherence in Eating Disorders: A Data Synthesis Table 2. Participant demographics. Age BMI Illness duration Years of education** Medication (% yes) Anxiety* Depression* AN (N = 364) 26.1 (8.1) p = .290 d = 0.01 15.6 (2.1) p < .000 d = 2.98 9.0 (7.8) 15.4 (2.4) p < .001 d = 0.39 46% 0.6 (1.3) p < .001 d = 1.16 0.6 (0.4) p < .001 d = 3.0 BN (N = 125) 27.0 (7.7) p = .857 d = 0.01 21.6 (2.3) p = .780 d = 0.0 10.3 (7.6) 15.8 (2.5) p = .097 d = 0.22 20.7% 0.6 (0.7) p < .001 d = 2.11 0.3 (0.7) p < .001 d = 1.76 AN Recovered (N = 107) 29.4 (10.6) p = .026 d = 0.28 20.8 (2.0) p = .001 d = 0.42 9.4 (7.3) 16.2 (2.5) P = .792 d = 0.04 8.3% -0.6 (0.5) p < .001 d = 0.0 0.0 (0.5) p < .001 d = 1.4 Unaffected Relative (N = 115) 44.5 (13.3) p < .000 d = 1.75 23.0 (3.2) p = .001d = 0.59 - 15.8 (3.0) p = .139d = 0.20 - -0.2 (0.9) p < .001 d = 0.62 -0.3 (0.9) p = .001 d = 0.51 HC (N = 273) 26.9 (8.4) 21.6 (1.9) - 16.3 (2.2) - -0.6 (0.5) -0.6 (0.4) AN = Anorexia Nerovsa; BN = Bulimia Nervosa; AN recovered = Recovered Anorexia Nervosa; HC = Healthy Control; BMI = Body Mass Index. Z score calculated from the Hospital Anxiety and Depression Scale or the Depression, Anxiety and Stress Scale. P value and effect size when compared to HC group. Table 2. Participant demographics. AN = Anorexia Nerovsa; BN = Bulimia Nervosa; AN recovered = Recovered Anorexia Nervosa; HC = Healthy Control; BMI = Body Mass Index. Z score calculated from the Hospital Anxiety and Depression Scale or the Depression, Anxiety and Stress Scale. *P value and effect size when compared to HC group. Rey Figure: Group comparisons Table 4 displays the means, transformed means (used for analysis where the data were skewed) and standard deviations and medians for the Rey Figure outcome variables. AN, BN and unaf- fected relative groups differed significantly from HC on all three Rey Figure outcome variables. No significant differences were found for the AN recovered group. Fig 2 depicts a radar chart of the effects sizes of each of the indices of the Rey for each group. Predictors of central coherence within ED subtype Regression analyses were conducted within each ED group to investigate possible predictors of central coherence (age, illness duration, BMI, years of education and anxiety and depression). Separate regressions were performed within each sub-group as data for some of the covariates were not available for the HC sample. None of these variables significantly predicted CCI outcome in AN or AN recovered. The lack of statistical significance for these analyses may be due to the small sample within some of the sub-groups. Rey Figure: Normative data Medians and interquartile ranges for each outcome of the Rey Figure stratified by clinical group are displayed in Table 3. Fig 1. Scoring elements for the Rey Figure. doi:10.1371/journal.pone.0165467.g001 l 0165467 N b 2 2016 5 / 11 Fig 1. Scoring elements for the Rey Figure. doi:10.1371/journal.pone.0165467.g001 Fig 1. Scoring elements for the Rey Figure. PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 5 / 11 doi:10.1371/journal.pone.0165467.t002 AN = Anorexia Nerovsa; BN = Bulimia Nervosa; AN recovered = Recovered Anorexia Nervosa; HC = Healthy Control; BMI = Body Mass Index. Z score calculated from the Hospital Anxiety and Depression Scale or the Depression, Anxiety and Stress Scale. *P value and effect size when compared to HC group. Discussion This study assessed central coherence in ED by collating data from previously conducted stud- ies within the same research group that used the Rey Figure with psychologists trained to use a consistent administration and scoring system to measure central coherence [20], and applied it Table 3. Medians and interquartile ranges for normative data for the Rey Figure. Normative data Order Index median Style Index median Central Coherence Index median AN (N = 364) 2.17 (1.50–2.50) 1.50 (0.93–1.67) 1.38 (1.02–1.61) BN (N = 125) 2.17 (1.83–2.54) 1.33 (1.00–1.67) 1.36 (1.09–1.60) AN Recovered (N = 107) 2.17 (1.78–2.67) 1.33 (0.90–1.67) 1.41 (1.05–1.66) Unaffected Relative (N = 115) 2.17 (1.50–2.50) 1.33 (0.98–1.67) 1.34 (1.02–1.53) HC (N = 273) 2.33 (1.97–2.67) 1.50 (1.16–1.67) 1.50 (1.21–1.68) AN = Anorexia Nerovsa; BN = Bulimia Nervosa; AN recovered = Recovered Anorexia Nervosa; HC = Healthy Control Table 3. Medians and interquartile ranges for normative data for the Rey Figure. PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 6 / 11 Central Coherence in Eating Disorders: A Data Synthesis Table 4. Means and transformed means (where data were skewed) and Standard Deviations for the Rey Figure analysis. Order Index Style Index Style Index transformed Central Coherence Index Central Coherence Index transformed* AN (N = 364) 1.98 (0.72) p < .001 d = 0.42 1.32 (0.50) 0.47 (0.30) p < .004 d = 0.11 1.28 (0.43) 0.51 (0.24) p < .001 d = 0.26 BN (N = 125) 2.16 (0.58) p = .026 d = 0.17 1.30 (0.43) 0.50 (0.26) p = .003 d = 0.26 1.31 (0.34) 0.50 (0.20) p = .001 d = 0.23 AN Recovered (N = 107) 2.15 (0.66) p = .780 d = 0.18 1.30 (0.46) 0.49 (0.27) p = .090 d = 0.21 1.32 (0.40) 0.49 (0.23) p = .187 d = 0.18 Unaffected Relative (N = 115) 1.95 (0.64) p < .001 d = 0.50 1.27 (0.47) 0.51 (0.28) p < .001 d = 0.29 1.27 (0.39) 0.52 (0.22) p < .001 d = 0.32 HC (N = 273) 2.26 (0.61) 1.39 (0.43) 0.44 (0.22) 1.40 (0.36) 0.45 (0.22) Data have been transformed using reverse log transformations, therefore value of mean has been reversed. P value and effect size when compared to HC group s (where data were skewed) and Standard Deviations for the Rey Figure analysis. s and transformed means (where data were skewed) and Standard Deviations for the Rey Figure analysis. Table 4. Means and transformed means (where data were skewed) and Standard Deviati systematically and reliably to individuals with AN, BN, recovered AN, unaffected relatives of AN and HCs. Secondly, we aimed to provide benchmark data using this robust measure and consistent scoring method with a large number of participants across different ED diagnoses. The results of this study were consistent with previous research, demonstrating poorer global processing in AN, BN and unaffected AN relatives in comparison to HCs as evidenced by lower scores on central coherence indices [6,10,22,29–31]. However, the findings are dis- crepant with a number of studies that report recall and copy data from the Rey figure as a mea- sure of central coherence [32]. Contrary to other findings, this study did not however, find any differences between recovered AN and the HC group [33]. Central Coherence in Eating Disorders: A Data Synthesis more global processing style than the AN group. Although this data lends support to the notion that such characteristics may be influenced by starvation, we are limited in making further interpretations due to the cross-sectional design of this study. Studies employing longitudinal designs are needed to examine this thought further. One rather curious finding from this study are the rather contradicting findings of poorer global processing in unaffected relatives of those with AN and better global processing in recovered ANs in comparison to currently ill ANs. One of these findings lends support to such characteristics representing an underlying vulnerability, whereas the other suggests they are a consequence of starvation. One possible explanation for high levels of global processing in the recovered AN group in this study is that it is likely that a large proportion of these individuals received Cognitive Remediation Therapy (CRT) as part of their treatment package. Involve- ment in such cognitive training in combination with weight-gain may therefore have signifi- cantly impacted on this groups processing style, either through increasing insight into alternative strategies or by learning to employ alternative ways of processing information. Such explanations of the current findings are speculative; however they are supported by the findings of RCTs demonstrating cognitive improvement following CRT in adults with AN [34]. Furthermore, this study did not find any significant correlations between clinical variables and central coherence index. Such findings may indicate that poor global processing is inde- pendent of such factors and maybe an underlying trait. The findings from this study are important for the field of ED for several reasons. Firstly, as the Rey Figure is one of the most widely used neuropsychological measures of CC in ED, this study provides much needed normative data that can be used as a benchmark for clinicians and future studies. These data address the methodologicalflaws in the current literature, which have led to inconsistent results and discrepant findings between studies. Furthermore, employ- ing a consistent administration and scoring system and reporting the data in a systematic fash- ion helps to synthesise knowledge, identify consistencies and provides robust comparison data for future studies. This is an important issue in clinical trials where neuropsychological assess- ments are used as an outcome measure and it may not be possible to recruit a healthy control group. PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 The difference in effect size between currently ill ANs and recovered can be observedin Fig 2, with recovered ANs demonstrating a Data have been transformed using reverse log transformations, therefore value of mean has been reversed. P value and effect size when compared to HC group Analysis performed using regression analysis. doi:10.1371/journal.pone.0165467.t004 Fig 2. Radar chart of effects sizes of each of the indices of the Rey for each group in comparison with HC group. doi:10.1371/journal.pone.0165467.g002 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 7 / 11 *Data have been transformed using reverse log transformations, therefore value of mean has been reversed. P value and effect size when compared to HC group doi:10.1371/journal.pone.0165467.t004 systematically and reliably to individuals with AN, BN, recovered AN, unaffected relatives of AN and HCs. Secondly, we aimed to provide benchmark data using this robust measure and consistent scoring method with a large number of participants across different ED diagnoses. g g g The results of this study were consistent with previous research, demonstrating poorer global processing in AN, BN and unaffected AN relatives in comparison to HCs as evidenced by lower scores on central coherence indices [6,10,22,29–31]. However, the findings are dis- crepant with a number of studies that report recall and copy data from the Rey figure as a mea- sure of central coherence [32]. Contrary to other findings, this study did not however, find any differences between recovered AN and the HC group [33]. The difference in effect size between currently ill ANs and recovered can be observedin Fig 2, with recovered ANs demonstrating a Fig 2. Radar chart of effects sizes of each of the indices of the Rey for each group in comparison with HC group. doi:10.1371/journal.pone.0165467.g002 Fig 2. Radar chart of effects sizes of each of the indices of the Rey for each group in comparison with HC group. doi:10.1371/journal.pone.0165467.g002 Fig 2. Radar chart of effects sizes of each of the indices of the Rey for each group in comparison with HC group. d i 10 1371/j l 0165467 002 Fig 2. Radar chart of effects sizes of each of the indices of the Rey for each group in comparison with HC group. Radar chart of effects sizes of each of the indices of the Rey for each group in comparison with HC group. 7 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Central Coherence in Eating Disorders: A Data Synthesis randomised controlled trials have shown improvements to cognitive processing following CRT in individuals with AN [35]. However, due to the limited amount of research in BN, CRT has not been routinely implemented with this group. The results of the present study provide strong evidence to suggest that CRT could also be of potential benefit for BN, as suggested by Dingemans et al., (2013) [36]. Furthermore, unhelpful cognitive styles within families, such as lower levels of global pro- cessing highlighted here are likely to negatively impact on the responses of carers to ED and may produce maladaptive behaviours, such as a rigid parenting style and an inability to see the wider context of the problem, that may inadvertently serve to maintain the ED. These findings indicate that including family members in treatments such as CRT may be beneficialfor help- ing to raise awareness of cognitive processing styles within families, and could result in more adaptive coping to ED pathology. In support of this a small pilot study demonstrated that a module of CRT developed for collaborative use with individuals with AN and their carers was feasible [37,38]. The strengths of this study include its large sample size and inclusion of ED subtypes and unaffected relatives. Additionally, the collated data presented here holds advantages over that of a meta-analysis as the raw data from each study was collated and analysed, compared to the use of sample averages in a meta-analysis. This analysis allowed for the production of means and medians that can be used as benchmark data for future studies. Collating the data in this fashion has been done previously in AN with regards to set-shifting,and have proved to be extremely useful for the ED field [1,39]. As the results for this study were obtained by combining data from several studies one potential weakness is the lack of a consistent measure of ED pathology or intelligence. The studies from which the data were collated were investigating different hypotheses and employed different measures to do so. It was therefore not possible to report on these charac- teristics in the present study. In terms of limitations, it is also acknowledged that this paper reports on a re-analysis of previously published datasets. The fact that this study only considers data from one research group could also be seen as a potential weakness, for example as men- tioned above a number of patients within the studies are likely to have been offered CRT as part of their treatment package, which may have contributed to the outcomes in the recovered AN group and may therefore not necessarily generalise to other recovered AN populations. However, this is also a strength of the study as we can be sure that the Rey was administered and scored consistently, adding validity and reliability to this studies results. In conclusion, this study provides robust data from a large sample size for perhaps the most widely used measure of central coherence in the ED literature. Reporting data in this synthe- sised fashion provides reliable normative data useful for future research studies within ED and the wider psychiatric field. Acknowledgments The Authors would like to thank the National Institute for Health Research (NIHR) [Mental Health Biomedical Research Centre and/or Dementia Biomedical Research Unit] at South Lon- don & Maudsley NHS Foundation Trust for financial support. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Kate Tchanturia would also like to thank the Swiss Anorexia Foundation (34–16). Author Contributions Conceptualization: KL KT. The reliable data presented in the form of means standard deviations and medians reported in this manuscript can therefore be used for comparisons. Secondly, the different pro- cessing profiles identified across the ED sub-types provide strong support for the Rey Figure as a sensitive measure of central coherence within ED. The use of a scoring system to derive the central coherence index based on measures of order and style produces a reliable outcome. We suggest that in order to avoid inconsistencies, the use and development of alternative measures and scoring systems should be discouraged. One factor that may contribute to the differences in outcomes produced between studies using Booth’s scoring system is the scoring of the style index. This index relies a little more on experimenter’s judgment and therefore may be vulnerable to error or bias. It is therefore rec- ommended that researchers are fully trained in the administration and scoring of the Rey Figure and clear guidelines are given as to what constitutes certain scores. It is also recom- mended that as well as video recording the administration of Rey Figures, future research stud- ies should also attempt to co-rate a minimum of 10% of their data and perform reliability checks to be reported in published manuscripts. This paper also provides the largest dataset for the Rey Figure with both individuals with BN and also unaffected family members of those with AN. The findings demonstrated poorer global processing in these groups compared to HC comparisons, and a similar global process- ing style to what is observedin AN. Such findings provide support for the development and use of remediation treatments with both clinical groups. Cognitive Remediation Therapy (CRT) directly targets biased cognitive processing, such as poor global processing, and 8 / 11 References 1. Tchanturia K, Davies H, Roberts M, Harrison A, Nakazato M, Schmidt U, et al. Poor cognitive flexibility in eating disorders: examining the evidence using the Wisconsin Card Sorting Task. PloS One. 2012; 7 (1):e28331. doi: 10.1371/journal.pone.0028331 PMID: 22253689 2. Lang K, Lopez C, Stahl D, Tchanturia K, Treasure J. Central coherence in eating disorders: An updated systematic review and meta-analysis. World J Biol Psychiatry. 2014; 15(8):586–98. doi: 10. 3109/15622975.2014.909606 PMID: 24882144 3. Treasure J, Schmidt U. The cognitive-interpersonal maintenance model of anorexia nervosa revisited: a summary of the evidence for cognitive, socio-emotional and interpersonal predisposing and perpetu- ating factors. J Eat Disord. 2013; 1(1):1. 4. Wagemans J, Elder JH, Kubovy M, Palmer SE, Peterson MA, Singh M, et al. A century of Gestalt psy- chology in visual perception: I. Perceptual grouping and figure–ground organization. Psychol Bull. 2012; 138(6):1172. doi: 10.1037/a0029333 PMID: 22845751 5. Navon D. Forest before trees: The precedence of global features in visual perception. Cognit Psychol. 1977; 9(3):353–83. 6. Harrison A, Tchanturia K, Treasure J. Measuring state trait properties of detail processing and global integration ability in eating disorders. World J Biol Psychiatry. 2011; 12(6):462–72. doi: 10.3109/ 15622975.2010.551666 PMID: 21554026 7. Lopez C, Tchanturia K, Stahl D, Treasure J. Weak central coherence in eating disorders: a step towards looking for an endophenotype of eating disorders. J Clin Exp Neuropsychol. 2009; 31(1):117– 25. doi: 10.1080/13803390802036092 PMID: 18608648 8. Lang K, Lloyd S, Khondoker M, Simic M, Treasure J, Tchanturia K. Do Children and Adolescents with Anorexia Nervosa Display an Inefficient Cognitive Processing Style? PloS One. 2015; 10(7): e0131724. doi: 10.1371/journal.pone.0131724 PMID: 26133552 9. Roberts ME, Tchanturia K, Treasure JL. Is attention to detail a similarly strong candidate endopheno- type for anorexia nervosa and bulimia nervosa? World J Biol Psychiatry. 2013; 14(6):452–63. doi: 10. 3109/15622975.2011.639804 PMID: 22263673 10. Lang K, Treasure J, Tchanturia K. Is inefficient cognitive processing in anorexia nervosa a familial trait? A neuropsychological pilot study of mothers of offspring with a diagnosis of anorexia nervosa. World J Biol Psychiatry. 2015;1–8. 11. Galimberti E, Fadda E, Cavallini MC, Martoni RM, Erzegovesi S, Bellodi L. Executive functioning in anorexia nervosa patients and their unaffected relatives. Psychiatry Res. 2013; 208(3):238–44. doi: 10.1016/j.psychres.2012.10.001 PMID: 23122554 12. Lopez C, Tchanturia K, Stahl D, Treasure J. Central coherence in eating disorders: a systematic review. Psychol Med. 2008; 38(10):1393–404. doi: 10.1017/S0033291708003486 PMID: 18447964 13. Writing – review& editing: MR AH CL EG JT. Writing – review& editing: MR AH CL EG JT. Conceptualization: KL KT. 9 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Central Coherence in Eating Disorders: A Data Synthesis PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 References Assessment of eating disorders: Interview or self-report questionnaire? Int J Eat Disord. 1994; 16(4):363–70. PMID: 7866415 27. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983; 67 (6):361–70. PMID: 6880820 28. Lovibond PF, Lovibond SH. The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther. 1995; 33(3):335–43. PMID: 7726811 29. Goddard E, Carral-Ferna´ndez L, Denneny E, Campbell IC, Treasure J. Cognitive flexibility, central coherence and social emotional processing in males with an eating disorder. World J Biol Psychiatry. 2014; 15(4):317–26. doi: 10.3109/15622975.2012.750014 PMID: 23336111 30. Tenconi E, Santonastaso P, Degortes D, Bosello R, Titton F, Mapelli D, et al. Set-shifting abilities, cen- tral coherence, and handedness in anorexia nervosa patients, their unaffected siblings and healthy controls: exploring putative endophenotypes. World J Biol Psychiatry. 2010; 11(6):813–23. doi: 10. 3109/15622975.2010.483250 PMID: 20486870 31. Aloi M, Rania M, Caroleo M, Bruni A, Palmieri A, Cauteruccio MA, et al. Decision making, central coherence and set-shifting: a comparison between Binge Eating Disorder, Anorexia Nervosa and Healthy Controls. BMC Psychiatry. 2015; 15(1):1. 32. Stedal K, Rose M, Frampton I, Landrø NI, Lask B. The neuropsychological profile of children, adoles- cents, and young adults with anorexia nervosa. Arch Clin Neuropsychol. 2012; 27(3):329–37. doi: 10. 1093/arclin/acs032 PMID: 22414676 33. Lindner SE, Fichter MM, Quadflieg N. Decision-making and planning in full recovery of anorexia ner- vosa. Int J Eat Disord. 2012; 45(7):866–75. doi: 10.1002/eat.22025 PMID: 22552824 34. Tchanturia K, Lounes N, Holttum S. Cognitive remediation in anorexia nervosa and related conditions: a systematic review. Eur Eat Disord Rev. 2014; 22(6):454–62. doi: 10.1002/erv.2326 PMID: 25277720 35. Tchanturia K, Lloyd S, Lang K. Cognitive remediation therapy for anorexia nervosa: current evidence and future research directions. Int J Eat Disord. 2013; 46(5):492–5. doi: 10.1002/eat.22106 PMID: 23658098 36. Dingemans AE, Danner UN, Donker JM, Aardoom JJ, Van Meer F, Tobias K, et al. The effectiveness of cognitive remediation therapy in patients with a severe or enduring eating disorder: a randomized controlled trial. Psychother Psychosom. 2013; 83(1):29–36. doi: 10.1159/000355240 PMID: 24281361 37. Lang K, Treasure J, Tchanturia K. Acceptability and feasibility of self-help Cognitive Remediation Ther- apy For Anorexia Nervosa delivered in collaboration with carers: A qualitative preliminary evaluation study. Psychiatry Res. 2015; 225(3):387–94. doi: 10.1016/j.psychres.2014.12.008 PMID: 25555418 38. Hutchison S, Roberts A, Lask B. Family cognitive remediation therapy for child and adolescent anorexia nervosa. References Darcy AM, Fitzpatrick KK, Manasse SM, Datta N, Klabunde M, Colborn D, et al. Central coherence in adolescents with bulimia nervosa spectrum eating disorders. Int J Eat Disord. 2015; 48(5):487–93. doi: 10.1002/eat.22340 PMID: 25146149 14. Hill EL. Executive dysfunction in autism. Trends Cogn Sci. 2004; 8(1):26–32. PMID: 14697400 . Hill EL. Executive dysfunction in autism. Trends Cogn 15. Pantelis C, Barber FZ, Barnes TR, Nelson HE, Owen AM, Robbins TW. Comparison of set-shifting ability in patients with chronic schizophrenia and frontal lobe damage. Schizophr Res. 1999; 37 (3):251–70. PMID: 10403197 16. Head D, Bolton D, Hymas N. Deficit in cognitive shifting ability in patients with obsessive-compulsive disorder. Biol Psychiatry. 1989; 25(7):929–37. PMID: 2720006 17. Snitz BE, MacDonald AW, Carter CS. Cognitive deficits in unaffected first-degree relatives of schizo- phrenia patients: a meta-analytic review of putative endophenotypes. Schizophr Bull. 2006; 32 (1):179–94. doi: 10.1093/schbul/sbi048 PMID: 16166612 10 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 Central Coherence in Eating Disorders: A Data Synthesis 18. Cavedini P, Zorzi C, Piccinni M, Cavallini MC, Bellodi L. Executive dysfunctions in obsessive-compul- sive patients and unaffected relatives: searching for a new intermediate phenotype. Biol Psychiatry. 2010; 67(12):1178–84. doi: 10.1016/j.biopsych.2010.02.012 PMID: 20381015 19. Osterrieth P. The test of copying a complex figure: A contribution to the study of perception and mem- ory. Arch Psychol. 1944; 30:286–356. 20. Booth R. Local-global Processing and Cognitive Style in Autism Spectrum Disorder and Typical Devel- opment. 2006. 21. Roberts ME, Tchanturia K, Treasure JL. Exploring the neurocognitive signature of poor set-shifting in anorexia and bulimia nervosa. J Psychiatr Res. 2010; 44(14):964–70. doi: 10.1016/j.jpsychires.2010. 03.001 PMID: 20398910 22. Lopez CA, Tchanturia K, Stahl D, Treasure J. Central coherence in women with bulimia nervosa. Int J Eat Disord. 2008; 41(4):340–7. doi: 10.1002/eat.20511 PMID: 18306347 23. Harrison A, Tchanturia K, Naumann U, Treasure J. Social emotional functioning and cognitive styles in eating disorders. Br J Clin Psychol. 2012; 51(3):261–79. doi: 10.1111/j.2044-8260.2011.02026.x PMID: 22803934 24. First MB, Spitzer RL, Gibbon M, Williams JB. Structured clinical interview for DSM-IV axis I disorders- patient edition (SCID-I/P, Version 2.0). N Y Biom Res Dep N Y State Psychiatr Inst. 1995; 722. 25. Anderluh M, Tchanturia K, Rabe-Hesketh S, Collier D, Treasure J. Lifetime course of eating disorders: design and validity testing of a new strategy to define the eating disorders phenotype. Psychol Med. 2009; 39(01):105–14. 26. Fairburn CG, Beglin SJ. PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016 References Cogn Remediat Ther CRT Eat Weight Disord Lond Routledge. 2014; 147. 39. Tchanturia K, Harrison A, Davies H, Roberts M, Oldershaw A, Nakazato M, et al. Cognitive flexibility and clinical severity in eating disorders. Plos One. 2011; 6(6):e20462. doi: 10.1371/journal.pone. 0020462 PMID: 21698277 11 / 11 PLOS ONE \| DOI:10.1371/journal.pone.0165467 November 2, 2016
https://openalex.org/W2112137594	https://bmccardiovascdisord.biomedcentral.com/counter/pdf/10.1186/1471-2261-3-1	English	null	Management of congestive heart failure: a gender gap may still exist. Observations from a contemporary cohort	BMC cardiovascular disorders	2,003	cc-by	5,404	BioMed Central BioMed Central Open Research article Management of congestive heart failure: a gender gap may still exist. Observations from a contemporary cohort Jason M Burstein*1, Raymond Yan2, Iris Weller3 and Beth L Abramson4 Open Access Received: 6 November 2002 Accepted: 5 February 2003 Received: 6 November 2002 Accepted: 5 February 2003 This article is available from: http://www.biomedcentral.com/1471-2261/3/1 © 2003 Burstein et al; licensee BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL. © 2003 Burstein et al; licensee BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribut media for any purpose, provided this notice is preserved along with the article's original URL. Abstract Background: Unlike other cardiovascular diseases the incidence and prevalence of congestive heart failure (CHF) continues to increase. While gender differences in coronary artery disease have been well described, to date, there has been a relative paucity of similar data in patients with CHF. We conducted a pilot study to evaluate the profile and management of patients with CHF at a tertiary care centre to determine if a gender difference exists. Methods: A chart review was performed at a tertiary care centre on consecutive patients admitted with a primary diagnosis of CHF between June 1997 and 1998. Co-morbidity, diagnostic investigations, and management of CHF were recorded. Comparisons between male and female patients were conducted. Results: One hundred and forty five patients were reviewed. There were 80 male (M) and 65 female (F) patients of similar age [71.6 vs. 71.3 (M vs. F), p = NS]. Male patients were more likely to have had a previous myocardial infarction (66% vs. 35%, p < 0.01) and revascularization (41% vs. 20%, p < 0.05), and had worse left ventricular ejection fraction (LVEF) than women, [median LVEF 3 vs. 2 (M vs. F), p < 0.01]. Male patients were more likely to have a non-invasive assessment of left ventricular (LV) function [85% vs. 69%, (M vs. F), p < 0.05]. A logistic regression analysis suggests that amongst those without coronary disease, males were more likely to receive non-invasive testing. There were no differences in the use of prescribed medications, in this cohort. Conclusions: This pilot study demonstrated that there seem to be important gender differences in the profile and management of patients with CHF. Importantly women were less likely to have an evaluation of LV function. As assessment of LV function has significant implications on patient management, this data justifies the need for larger studies to assess gender differences in CHF profile and treatment. talisation in the elderly. In the province of Ontario alone, there are approximately 13,000 admissions annually for CHF. One-year case fatality for patients with CHF in On- tario (1994–1997) was over 30 % for both men and for women. [7] In stark contrast to other cardiovascular Page 1 of 7 (page number not for citation purposes) Diagnostic investigations The use of echocardiogram, multiple gated cardiac blood pool imaging (MUGA), and/or cardiac catheterization, ei- ther during the current admission, or comments from pre- vious studies, was recorded. With respect to non-invasive assessment of left ventricular function, the analysis in- cluded current studies or previous studies as noted in the chart. Grade/Etiology f The presence of hypertension, diabetes mellitus, previous myocardial infarctions, and revascularization [coronary artery bypass grafting (CABG) or percutaneous translumi- nal coronary angioplasty (PTCA); past and present] was noted. Coronary disease (CAD) was defined as a history of myocardial infarction, CABG, PTCA, or chart documenta- tion of "CAD" as assessed by the attending physician. The ejection fraction (if documented in the patient chart) was recorded. Previous admissions for CHF were noted. If there was a result of an angiogram, echocardiogram, or nuclear test, the left ventricular ejection fraction (LVEF) was graded 1 to 4 based on the test result. [LVEF > 60% = grade 1, LVEF 40–59% = grade 2, LVEF 20–39% = grade 3, and LVEF 0–19% = grade 4.] Other risk factors for CHF such as valvular heart disease, body mass index, left ven- tricular hypertrophy, and alcohol abuse were not record- ed, as they were not uniformly noted on charts. The management and prognosis in heart failure patients will vary depending on the etiology, and it is critical to as- sess LV function and look for precipitating factors, as rec- ommended by Canadian guidelines. [17] It is extremely important to assess for possible gender differences in the profile, presentation and management of patients with heart failure to ensure optimal medical care for both men and women, given the possible differences in pathophys- iology and presentation of disease. Although there is a large body of literature on "gender discrepancies" in the management of coronary disease (CAD) [18]-[29] there is a relative paucity of data on the evaluation of heart failure patients and gender. One recent study found that women with CHF are less likely to undergo assessment of LV func- tion.[30] There is also evidence that women receive less Angiotensin Converting Enzyme (ACE) inhibitors in the treatment of CHF.[11,29,30] http://www.biomedcentral.com/1471-2261/3/1 BMC Cardiovascular Disorders 2003, 3 diseases, the incidence and prevalence of CHF continue to increase, for reasons including an aging population and better treatment of coronary disease and hypertension. Patients with heart failure create broad reaching manage- ment issues for both primary care physicians and specialists. tients with a primary discharge diagnosis of CHF (ICD codes #428.0) between June 1997 and June 1998 were identified. Patients were enrolled during their first hospi- tal admission during the study period. Subsequent admis- sions, if any, were not analyzed. An unbiased observer (R.Y.) reviewed the charts and recorded predetermined variables, if present, in the following categories: grade/eti- ology, diagnostic investigations, and management of CHF. (see below) Gender differences may exist between patients with heart failure. [8–13] Although the overall prevalence is similar, women make up a greater percentage of patients in those over 75 years of age.[3,4] The syndrome results from car- diac dysfunction which is either systolic (reduced ejection fraction and ventricular dilatation) or diastolic (impaired myocardial relaxation leading to elevated end-diastolic pressures with normal cardiac chamber size). Women may have more hypertension, diabetes, and diastolic dys- function then men, and less coronary disease. [14,15] Pre- vious epidemiological studies such as Framingham and the National Health and Nutrition Examination Survey re- port better survival in women with CHF.[2,5] These stud- ies did not assess left ventricular (LV) function, and therefore, probably captured more women with diastolic dysfunction, which bodes a better prognosis. More recent studies enrolling only patients with systolic dysfunction, have reported that women, in fact, have higher mortality rates.[16] Statistical analysis Differences between males and females were assessed us- ing χ2 tests (Fischer's Exact Test in the case of expected cell counts less than 5) for categorical variables and Wilcoxon Rank Sum tests for ordinal variables. Logistic regression analysis was used to assess the association between gender Management The triage of patients [medical floor vs. cardiology ward vs. coronary care unit (CCU)/intensive care unit (ICU)] and the use of cardiac monitors were recorded. In our in- stitution patients are uniformly monitored in the CCU and ICU, however, on the cardiology floor monitored beds are at the discretion of the treating physician. The use of ACE inhibitors, β-blockers, diuretics, and digoxin, were recorded based on discharge medications and inpatient medication records. Given the growing epidemic of CHF, it is important to evaluate if gender differences exist in the etiology and management of CHF. We conducted a pilot study to deter- mine if gender differences exist in the etiology, diagnostic work-up, or management of patients admitted to tertiary care teaching hospital with a diagnosis of CHF. Background h Congestive heart failure (CHF), a common medical con- dition associated with high mortality and morbidity, has an enormous impact on public health. [1–6] In Canada, as in the United States, CHF is the leading cause of hospi- Page 1 of 7 (page number not for citation purposes) Page 1 of 7 (page number not for citation purposes) http://www.biomedcentral.com/1471-2261/3/1 Page 2 of 7 (page number not for citation purposes) Methods Among those with CAD, sex was not related to non- invasive testing, OR (95%CI) 1.05 (0.29–3.79), whereas among those with no CAD, males were more likely to re- ceive non-invasive testing, OR (95% CI) 4.48 (1.15– 17.50), suggesting that both CAD and sex are independ- ent factors in referral practice. and non-invasive assessment of left ventricular function. Covariates assessed were age, presence of diabetes melli- tus, hypertension, and coronary artery disease. In order to assess the fit of the model, deviances from fitting the mod- el with and without the covariate were compared. The dif- ferences in the deviances were assumed to follow a chi- square distribution. All analyses were done using the SPSS and SAS® statistical packages. and non-invasive assessment of left ventricular function. Covariates assessed were age, presence of diabetes melli- tus, hypertension, and coronary artery disease. In order to assess the fit of the model, deviances from fitting the mod- el with and without the covariate were compared. The dif- ferences in the deviances were assumed to follow a chi- square distribution. All analyses were done using the SPSS and SAS® statistical packages. Methods The study took place at St. Michael's hospital, a tertiary care teaching centre in Toronto, Canada. Consecutive pa- Page 2 of 7 (page number not for citation purposes) Page 2 of 7 (page number not for citation purposes) BMC Cardiovascular Disorders 2003, 3 http://www.biomedcentral.com/1471-2261/3/1 http://www.biomedcentral.com/1471-2261/3/1 BMC Cardiovascular Disorders 2003, 3 Table 1: Gender differences in profile and management Male N = 80 Female N = 65 Etiology Previous Myocardial Infarction 66% 35% †† Hypertension 58% 55% Diabetes Mellitus 41% 33% Median LVEF 3 2 †† History of CABG or PTCA 41% 20% ††† Coronary Artery Disease 70% 37% † Diagnostic Investigations Echo/MUGA 85% 69% ††† Echo/MUGA (patients with CAD) 84% 83% Echo/MUGA (patients with no CAD) 88% 61%††† Cardiac Catheterization 25% 22% Triage Cardiology Ward 48% 37% CCU/ICU 39% 32% †Monitored Bed 59% 40% ††† †p < 0.001, †p < 0.01, ††† p < 0.005 Table 1: Gender differences in profile and management †p < 0.001, †p < 0.01, ††† p < 0.005 In this cohort, male patients were more likely than wom- en to have a non-invasive (echo or nuclear) assessment of LV function during their hospital stay [85% and 69% re- spectively, p < 0.05]. The odds ratio (OR) and 95% Con- fidence Interval (CI) for males was 2.53 (1.12,5.65). Because more men than women had diagnosed coronary artery disease, and because more patients with coronary artery disease had a non-invasive assessment of their left ventricular function OR (95% CI) 2.13 (0.96,4.76)], we did an additional logistic regression analysis to assess whether coronary artery disease (CAD) confounds the re- lation between sex and non-invasive assessment of left ventricular function. The addition of CAD resulted in a model with a poorer fit than the model with sex alone. Nevertheless there is some indication of a small con- founding influence, since the odds ratios for both CAD and SEX were altered towards the null value, OR (95%CI) 1.68 (0.72–3.91), 2.14 (0.91–5.01), respectively. Howev- er, it also suggests that sex is a stronger influence on refer- ral practice. Because more men than women had CAD, it is not clear in this analysis whether it was the patients' sex or the presence of CAD that influenced the physicians' de- cisions. To further explore this we stratified by CAD status to assess the relation between SEX and non-invasive test- ing. Results Similar numbers of male and female patients had a cardi- ac catheterization during this admission [25% vs. 22% respectively)]. With respect to triage, in this cohort of patients, there were no significant differences in the rates of admission to a cardiology floor (48% M, 37% F) or to a CCU/ICU (39% M, 32% F). Male patients were more likely to receive a monitored bed (CCU/ICU or cardiology ward) (59% M, 40% F, p < 0.05). In terms of prescribed medications there were no statistical differences in the use of ACE inhibitors, beta-blockers, diuretics, or digoxin (see table 2). Further- more in the subgroup of patients with grade 3 to 4 LV function there were no differences in the use of ACE in- hibitors, beta-blockers, or digoxin (see table 3). Little is known on gender differences in the etiology of CHF. In our cohort of patients with CHF, of those patients tested, women had preserved LV function as compared to their male counterparts. This is consistent with other re- ports that have noted that women are more likely to have had CHF despite having less severe LV dysfunction. Mendes et al. [12] evaluated over one thousand, five hun- dred patients who had undergone coronary angiography and found that women with more symptoms of CHF were found to have better LV function. Female patients had smaller end-diastolic volumes, despite similar LV end-di- astolic pressures, implicating diastolic dysfunction as the etiology of the heart failure. These retrospective data, along with the data from this current study, raises the Results One hundred and fifty three patients fulfilled entry crite- ria. Of these, 145/153 (95%) charts were available for re- view. There were 80 male (M) patients, and 65 female (F) patients in this group. The average age did not differ be- tween men and women [71.6 ± 11.1 (SD) years vs. 71.3 ± 13.8 (SD) years (M vs. F), p = NS]. The range of ages for women and men respectively was 34 to 89 and 41 to 93. Men had worse left ventricular function than women, [median LVEF 3 vs. 2 (M vs. F), p < 0.01]. There was no dif- ference in the rates of previous admission to hospital for CHF [49% M vs. 57% F)]. Gender differences in the profile and management of pa- tients appear in Table 1. Male patients were more likely than females to have had a previous myocardial infarction (66% vs. 35%, p < 0.01). Male patients were more likely to have had bypass surgery or angioplasty during this ad- mission or previously [41% vs. 20% (M vs. F), p < 0.05]. Correspondingly, males had a higher rate of CAD than fe- males [70% vs. 37% (M vs. F), p < 0.001]. However, there were no gender differences seen in rates of hypertension [58% M vs. 55%] or diabetes [40% M vs. 31% F] in this population. Results Page 3 of 7 (page number not for citation purposes) BMC Cardiovascular Disorders 2003, 3 http://www.biomedcentral.com/1471-2261/3/1 Table 2: Drug treatment in CHF patients Medication Class Male n = 80 Female n = 65 ACE inhibitors 56(70%) 41/(63%) Digoxin 38 (48%) 25 (38%) B-Blockers 32 (40%) 18 (28%) Diuretics 63 (79%) 49 (75%) P = NS for all comparisons Table 3: Drug treatment in CHF patients with Grade 3 or 4 LV Function Medication Class Male n = 46 Female n = 20 ACE inhibitors 36 (78%) 15 (75%) Digoxin 15 (33%) 10 (50%) B-Blockers 14 (33%) 9 (45%) P = NS for all comparisons http://www.biomedcentral.com/1471-2261/3/1 http://www.biomedcentral.com/1471-2261/3/1 BMC Cardiovascular Disorders 2003, 3 Table 2: Drug treatment in CHF patients Medication Class Male n = 80 Female n = 65 ACE inhibitors 56(70%) 41/(63%) Digoxin 38 (48%) 25 (38%) B-Blockers 32 (40%) 18 (28%) Diuretics 63 (79%) 49 (75%) P = NS for all comparisons Table 3: Drug treatment in CHF patients with Grade 3 or 4 LV Function Medication Class Male n = 46 Female n = 20 ACE inhibitors 36 (78%) 15 (75%) Digoxin 15 (33%) 10 (50%) B-Blockers 14 (33%) 9 (45%) P = NS for all comparisons Table 3: Drug treatment in CHF patients with Grade 3 or 4 LV Function Age, and a diagnosis of hypertension or diabetes did not confound the relation between sex and non-invasive testing. group differed in rates of coronary disease (previous myo- cardial infarction and revascularization), this does not ful- ly explain the lack of assessment of female patients. In those without coronary disease, a logistic regression anal- ysis found men more likely to undergo non-invasive test- ing than women. The absence of CAD should not negate the necessity of LV assessment since the causes of CHF are multifactorial. Therefore, the finding that men in our co- hort had more CAD does not justify their more frequent assessment of LV function as compared to women. Impor- tantly women were under-evaluated, regardless of a gen- der comparison, with only 69% having a non-invasive assessment of LV function with an admission diagnosis of CHF. This lack of evaluation should be highlighted since assessment of LV function is probably the most important diagnostic and prognostic step in evaluation of this com- mon patient population. Age, and a diagnosis of hypertension or diabetes did not confound the relation between sex and non-invasive testing. Page 4 of 7 (page number not for citation purposes) http://www.biomedcentral.com/1471-2261/3/1 BMC Cardiovascular Disorders 2003, 3 need for ongoing evaluation of gender differences in the etiology of heart failure. in men and women. Although there were no differences seen in medication use, the sample size was not large, and there may have been differences, which were not apparent due to lack of statistical power. Although this is consistent with other data from Ontario[7], it is in contrast to other preliminary data from eastern Canada[40] that show gen- der discrepancies in the treatment of heart failure, and therefore, raises the possibility of a regional variation in prescribing practices. Accounting for differences in systolic and diastolic dys- function may be explained partly in differing underlying causes of heart failure. This study, consistent with other larger studies, demonstrated that men with CHF are more likely to have had a prior myocardial infarction. Although not seen in this study, possibly due to lack of statistical power, others have noted gender differences in the rates of hypertension and diabetes in heart failure patients, both potential causes of diastolic dysfunction.[12] In both the Framingham study and the Studies of Left Ventricular Dysfunction (SOLVD) trial women were more likely to have hypertension and diabetes than their male counter- parts.[2,5,16,31] The possibility exists that there may be myocardial properties and/or hormonal environments unique to women contributing to heart failure. In a study by Carroll et al examining LVH caused by aortic stenosis, women had smaller, thicker-walled ventricles despite sim- ilar outflow obstruction, suggesting that female ventricles may respond differently to a pressure-overload state.[9] Reis et al, has recently shown that women in heart failure trials on estrogen, have a greater survival rate than those not on hormone replacement.[32] While intriguing and hypothesis generating, many of these studies have been retrospective, and underscore the need for further evalua- tion of women with heart failure, from bench to bedside. As in any chart review, there are inherent limitations to our study. The use of hospital charts may be limited by re- call bias, as tests performed outside of a hospital admis- sion, may not have been recorded on the chart. However, it is reasonable that efforts are made to obtain this infor- mation and convey it on the chart during an admission to a teaching hospital. A recent assessment of LV function would influence the need for obtaining an echocardio- gram on a current admission. http://www.biomedcentral.com/1471-2261/3/1 However in this pilot study, men and women were equally as likely to have been hos- pitalized previously for CHF, and the chart was reviewed for such an assessment during the previous admission. It is also possible that the use of ICD codes underestimated the number of CHF admissions, as the institution in which this study was conducted has several thousand ad- missions yearly in the cardiac program. Despite these lim- itations, this survey offers a "snap shot" of current practices in such an institution. Lastly, although this study may have been underpowered to detect certain differences such as the use of ACE inhibitors, it raises important ques- tions that must be addressed in clinical practice in the future. Limited data exists on the triage and management of pa- tients with CHF. In this particular group, there were no gender differences noted in the triage of patients. Al- though men were more likely to be assigned a monitored bed, they also had a higher incidence of known coronary disease. Men in this cohort were also more likely to have undergone revascularization procedures; this possibly is also explained by the higher incidence of known coronary disease. Although not statistically significant, there was a trend for women to be admitted to a general medical floor as opposed to a cardiology floor or CCU. This cannot be explained by age difference, because in contrast to other heart failure populations, the group admitted was of sim- ilar age. Although the literature suggests a gender bias in the evaluation and treatment of coronary disease [33–38], there is a paucity of data on the differences between diag- nostic procedures in men and women with CHF. This data is compelling evidence however for a more formal evalu- ation of gender differences in the diagnostic work-up of heart failure. Discussion The most important finding in this study is that, among heart failure patients with no known history of coronary artery disease, women were less likely to be referred for non-invasive assessment of left ventricular function. We also found that the heart failure patients with coronary ar- tery disease were more likely to be referred for non-inva- sive testing. Eighty-five percent of men as compared to 69% of women had non-invasive assessments of LV function (via echo or nuclear imaging). Although the men and women in this Page 4 of 7 (page number not for citation purposes) Page 4 of 7 (page number not for citation purposes) http://www.biomedcentral.com/1471-2261/3/1 References 24. Maynard C, Althouse R, Cequeira M, Olsufka M and Kennedy JW Un- derutilization of thrombolytic therapy in eligible women with acute myocardial infarction. Am J Cardiol 1991, 68:529-530 25 J k ER A d WA P k WF IV V h L C l RS d 24. Maynard C, Althouse R, Cequeira M, Olsufka M and Kennedy JW Un- derutilization of thrombolytic therapy in eligible women with acute myocardial infarction. Am J Cardiol 1991, 68:529-530 25. Jackson ER, Anderson WA, Peacock WF IV, Vaught L, Carley RS and Wilson AG Effect of a patient's sex on the timing of thrombo- lytic therapy. Ann Emerg Med 1996, 27:8-15 1. Smith WM Epidemiology of congestive heart failure. Am J Cardiol 1985, 55:3A-8A 2. Ho KK, Anderson KM, Kannel WB, Groosman W and Levy D Sur- vival after the onset of congestive heart failure in Framing- ham Heart Study Subjects Circulation. Circ 1993, 88:107-115 25. y J 25. Jackson ER, Anderson WA, Peacock WF IV, Vaught L, Carley RS and Wilson AG Effect of a patient's sex on the timing of thrombo- lytic therapy. Ann Emerg Med 1996, 27:8-15 y j 3. Gilium RF Heart Failure in the United States 1970–85. Am Heart J 1987, 113:1043-1045 3. Gilium RF Heart Failure in the United States 1970–85. Am Heart J 1987, 113:1043-1045 26. 26. Weaver WD, White HD, Wilcox RG, Aylward PE, Morris D and au- erci A Comparisons of characteristics and outcomes among women and men with acute myocardial infarction treated with thrombolytic therapy. JAMA 1996, 275:777-782 J 4. Ghali JK, Cooper R and Ford E Trends in rates for heart failure in the United States 1973–1986: evidence for increasing pop- ulation prevalence. Arch Intern Med 1990, 769-773 p J 27. Abramson BL, Weller I, Stewart DE and Liu P Chest pain: women still present later (abstract). The Canadian Journal of Cardiology 1998, 14:84F-84F p 5. Schocken DD, Arriata MI, Laever PE and Ross EA Prevalence and mortality rate of congestive heart failure in the United States. J Am Coll Cardiol 1992, 20:301-306 28. Wenger NK Exclusion of the elderly and women from coro- nary trials. JAMA 1992, 268:1460-1461 J 6. Massie BM and Nihir BS Evolving trends in the epidemiologic factors of heart failure: Rationale for preventive strategies and comprehensive disease management. Am Heart J 1997, 133:703-712 29. References Sueta CA, Metts A, Griggs TR, Borders VC and Simpson RJ ACE-I use and LV function in the elderly admitted with heart fail- ure: gender differences (abstract). J Am Coll Cardiol 1997, 29:17136 7. Tu JV and Zhang H Congestive Heart Failure outcomes in On- tario. In Cardiovascular Health and Services in Ontario: An ICES Atlas (Ed- ited by: Naylor CD, Slaughter PM) Toronto: Institute for Clinical Evaluative Sciences 1999, 11-122 30. Clinical Quality Improvement Network Mortality risk and pat- terns of practice in 4606 acute care patients with congestive heart failure: the relative importance of age, sex, and medi- cal therapy. Arch Intern Med 1996, 156:1669-1673 8. Adams KF Jr, Sueta CA, Gheorghiade M, O'Connor CM, Schwartz TA and Koch GG Gender differences in survival in advanced heart failure. Insights from the FIRST Study. Circulation 1999, 99:1816-1821 py 31. The SOLVD Investigators Effect of enalapril on survival in pa- tients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991, 325:293-302 9. Carroll JD, Carroll EP, Feldman T, Ward DM, Lang RM, McKaughey D and Karp RB Sex-associated differences in left ventricular function in aortic stenosis of the elderly. Circ 1992, 86:1099- 1107 g g J 32. Reis SE, Holubkov R, Young JB, White BG, Cohn JN and Feldman AM Estrogen is associated with improved survival in aging wom- en with congestive heart failure: analysis of the Vesnarinone studies. J Am Coll Cardiol 2000, 36:529-533 10. Lindenfeld J, Krause-Steinrauf H and Salerno J Where are all the women with heart failure? J Am Coll Cardiol 1997, 30:1417-1419 J 33. Tobin JN Sex Bias in Considering Coronary Bypass Surgery. Ann Intern Med 1987, 107:19-25 J , 11. Petrie MC, Dawson NF, Murdoch DR, Davie AP and McMurray JJV Failure of women's hearts. Circ 1999, 99:2334-2341 34. Hutchinson LA, Pasternack PF, Baumann FG, Grossi EA, Riles TS and Lamparello PJ Is there detrimental gender bias in preoperative cardiac management of patients undergoing vascular surgery? Circulation 1994, 90:220-223 12. Mendes LA, Davidoff R, Cupples LA, Ryan TJ and Jacobs AK Conges- tive heart failure in patients with coronary artery disease: the gender paradox. Am Heart J 1997, 134:207-212 g y 35. Barrett-Connor E Sex differences in coronary heart disease. Why are women so superior? The 1995 Ancel Keys Lecture. Circulation 1997, 95:252-264 g p J 13. References Zucker DR, Griffith JL, Beshansky JR and Selker HP Presentations of acute myocardial infarction in men and women. J Gen Intern Med 1997, 12:79-87 36. Wenger NK Gender, coronary artery disease, and coronary bypass surgery. Ann Intern Med 1990, 112:557-558 14. Dhinflrt FM, Kotsis JB, Yusuf S, Quinones MA, Pitt B and Stewart D Diabetes mellitus: a predictor of morbidity and mortality in the Studies of Left Ventricular Dysfunction (SOLVD) Trials and Registry. Am J Cardiol 1996, 77:1017-1020 37. Bickell NA, Pieper KS, Lee KL, Mark DB, Glower DD and Pryor DB Referral patterns for coronary artery disease treatment: gender bias or good clinical judgment? Ann Intern Med 1992, 116:791-797 15. Johnstone D, Limacher M, Stewart D, Liang CS, Ekelund L and Her- man M Clinical characteristics of patients in the Studies of pa- tients in the Studies of Left Ventricular Dysfunction. American Journal of Cardiology 1992, 70:894-900 38. Khan SS, Nessim S, Gray R, Czer LS, Chaux A and Matloff J In- creased mortality of women in coronary artery bypass sur- gery: evidence for referral bias. Ann Intern Med 1990, 112:561- 567 J gy 16. The SOLVD Investigators Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med 1992, 327:685-691 39. Hillis GS, Tent RJ, Winton P, MacLeod AM and Jennings KP Angi- otensin converting enzyme inhibitors in the management of cardiac failure: are we ignoring the evidence? Q J Med 1995, 89:145-152 17. Johnstone DE, Alnoor A, Arnold M, Berstein V, Bourassa M and Bro- phy J Diagnosis and management of Heart Failure. Can J Cardiol 1994, 10:613-631 40. Howlett JG, Stanlely J, MacDonald M and Johnstone DE The gender gap in optimal care for congestive heart failure at a Canadian university teaching hospital (abstract). The Canadian Journal of Cardiology 2000, 16:44B 18. Ayanian JZ and Epstein AM Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med 1991, 325:221-225 gy 41. Roger VL, Farkouh ME, Weston SA, Reeder Gs, Jacobsen SJ and Zins- meister AR Sex differences in evaluation and outcome of un- stable angina. JAMA 2000, 283:646-652 g J 19. Jagal SB and Naylor CD Sex Differences in the Use of Invasive Coronary Procedures in Ontario. Can J Cardiol 1994, 10:239-244 y J 20. Competing Interests 23. Schwartz LM, Fisher ES, Tosteson ANA, Woloshin S, Chang C and Virnig BA Treatment and health outcomes of women and men in a cohort with coronary artery disease. Arch Intern Med 1997, 157:1545-1551 p g None declared. http://www.biomedcentral.com/1471-2261/3/1 BMC Cardiovascular Disorders 2003, 3 This important pilot data justifies the need for larger stud- ies to assess gender discrepancies in heart failure. It also highlights the clinical problem of optimal evaluation and management of women with this common condition. 21. Hachamovitch R, Berman DS, Kiat H, Bairey-Merz CN, Cohen I and Cabico JA Gender-related differences in clinical management after exercise nuclear testing. J Am Coll Cardiol 1996, 26:1457- 1464 This important pilot data justifies the need for larger stud- ies to assess gender discrepancies in heart failure. It also highlights the clinical problem of optimal evaluation and management of women with this common condition. 22. 22. Gregor RD, Bata IR, Eastwood BJ, Garner JB, Guernsey JR and Mac- kenzie BR Gender differences in the presentation, treatment, and short term mortality of acute chest pain. Clin Invest Med 1994, 17:551-562 Conclusion In summary, there have been few studies to date that have analysed gender differences in assessment and manage- ment of heart failure. In this cohort of patients, women had less frequent non-invasive evaluations of left ven- tricular function than their male counterparts. Significant- ly, LV function is the most important prognostic marker for patient outcome. Furthermore, a difference in non-in- vasive evaluation has significant implications in patient care, as the management of diastolic and systolic dysfunc- tion varies. This pilot study was underpowered to show gender differences in treatment such as the use of ACE in- hibitors. A recent study by Roger et al.[41] examining sex differences in unstable angina patients demonstrated that female patients with unstable angina had less cardiac di- agnostic procedures but, perhaps paradoxically, had bet- ter long term outcomes when adjusting for baseline characteristics. Our study also demonstrates a diagnostic discrepancy in the work-up of patients with CHF. Although concerning, it remains to be seen if this gender difference translates into differences in patient outcomes. The last area studied in this cohort was the use of prescrip- tion medications. Previous reports have noted underutili- zation of ACE inhibitors in both sexes, with some reports noting a sex disparity.[30,39] Our data showed no statis- tical difference between prescription use of ACE inhibitors Page 5 of 7 (page number not for citation purposes) http://www.biomedcentral.com/1471-2261/3/1 Page 6 of 7 (page number not for citation purposes) References Shaw LJ, Miller DD, Romeis JC, Kargl D, Younis LT and Chaitman BR Gender Differences in the Non-invasive Evaluation and Man- agement of Patients with Suspected Coronary Artery Disease. Ann Intern Med 1994, 120:559-566 Page 6 of 7 (page number not for citation purposes) Page 6 of 7 (page number not for citation purposes) BMC Cardiovascular Disorders 2003, 3 http://www.biomedcentral.com/1471-2261/3/1 http://www.biomedcentral.com/1471-2261/3/1 BMC Cardiovascular Disorders 2003, 3 Pre-publication history Pre-publication history p y The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1471-2261/3/1/prepub http://www.biomedcentral.com/1471-2261/3/1/prepub Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 7 of 7 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge
https://openalex.org/W4367678812	https://www.frontiersin.org/articles/10.3389/fbioe.2023.1081072/pdf	English	null	Tailoring the HHx monomer content of P(HB-co-HHx) by flexible substrate compositions: scale-up from deep-well-plates to laboratory bioreactor cultivations	Frontiers in bioengineering and biotechnology	2,023	cc-by	8,848	OPEN ACCESS OPEN ACCESS EDITED BY Anthony Sinskey, MIT Dept of Biology, United States REVIEWED BY Takeharu Tsuge, Tokyo Institute of Technology, Japan Maciej Guzik, Polish Academy of Sciences, Poland CORRESPONDENCE Sebastian L. Riedel, riedel@tu-berlin.de, sebastian.riedel@bht-berlin.de †These authors have contributed equally to this work and share ﬁrst authorship RECEIVED 26 October 2022 ACCEPTED 18 April 2023 PUBLISHED 02 May 2023 CITATION Santolin L, Thiele I, Neubauer P and Riedel SL (2023), Tailoring the HHx monomer content of P(HB-co-HHx) by ﬂexible substrate compositions: scale-up from deep-well-plates to laboratory bioreactor cultivations. Front. Bioeng. Biotechnol. 11:1081072. doi: 10 3389/fbioe 2023 1081072 OPEN ACCESS EDITED BY Anthony Sinskey, MIT Dept of Biology, United States REVIEWED BY Takeharu Tsuge, Tokyo Institute of Technology, Japan Maciej Guzik, Polish Academy of Sciences, Poland CORRESPONDENCE Sebastian L. Riedel, riedel@tu-berlin.de, sebastian.riedel@bht-berlin.de †These authors have contributed equally to this work and share ﬁrst authorship RECEIVED 26 October 2022 ACCEPTED 18 April 2023 PUBLISHED 02 May 2023 CITATION Santolin L, Thiele I, Neubauer P and Riedel SL (2023), Tailoring the HHx monomer content of P(HB-co-HHx) by ﬂexible substrate compositions: scale-up from deep-well-plates to laboratory bioreactor cultivations. Front. Bioeng. Biotechnol. 11:1081072. doi: 10.3389/fbioe.2023.1081072 EDITED BY Anthony Sinskey, MIT Dept of Biology, United States REVIEWED BY Takeharu Tsuge, Tokyo Institute of Technology, Japan Maciej Guzik, Polish Academy of Sciences, Poland CORRESPONDENCE Sebastian L. Riedel, riedel@tu-berlin.de, sebastian.riedel@bht-berlin.de †These authors have contributed equally to this work and share ﬁrst authorship RECEIVED 26 October 2022 ACCEPTED 18 April 2023 PUBLISHED 02 May 2023 Lara Santolin 1†, Isabel Thiele1†, Peter Neubauer1 and Sebastian L. Riedel1,2 1Technische Universität Berlin, Institute of Biotechnology, Chair of Bioprocess Engineering, Berlin, Germany, 2Berliner Hochschule für Technik, Department VIII – Mechanical Engineering, Event Technology and Process Engineering, Laboratory of Environmental and Bioprocess Engineering, Berlin, Germany Santolin L, Thiele I, Neubauer P and Riedel SL (2023), Tailoring the HHx monomer content of P(HB-co-HHx) by ﬂexible substrate compositions: scale-up from deep-well-plates to laboratory bioreactor cultivations. The enhanced material properties exhibited by the microbially synthetized polyhydroxyalkanoate (PHA) copolymer poly(hydroxybutyrate- co-hydroxyhexanoate) [P(HB-co-HHx)] evidence that this naturally biodegrading biopolymer could replace various functionalities of established petrochemical plastics. In fact, the thermal processability, toughness and degradation rate of P(HB-co-HHx) can be tuned by modulating its HHx molar content enabling to manufacture polymers à-la-carte. We have developed a simple batch strategy to precisely control the HHx content of P(HB-co-HHx) to obtain tailor-made PHAs with deﬁned properties. TYPE Original Research PUBLISHED 02 May 2023 DOI 10.3389/fbioe.2023.1081072 TYPE Original Research PUBLISHED 02 May 2023 DOI 10.3389/fbioe.2023.1081072 TYPE Original Research PUBLISHED 02 May 2023 DOI 10.3389/fbioe.2023.1081072 KEYWORDS bioplastic, PHA, polyhydroxyalkanoate, poly(hydroxybutyrate-co-hydroxyhexanoate), Ralstonia eutropha, substrate-ﬂexible, biodegradable, scale-up frontiersin.org Frontiers in Bioengineering and Biotechnology OPEN ACCESS By adjusting the ratio of fructose to canola oil as substrates for the cultivation of recombinant Ralstonia eutropha Re2058/pCB113, the molar fraction of HHx in P(HB-co-HHx) could be adjusted within a range of 2–17 mol% without compromising polymer yields. The chosen strategy proved to be robust from the mL-scale in deep-well-plates to 1-L batch bioreactor cultivations. Front. Bioeng. Biotechnol. 11:1081072. doi: 10.3389/fbioe.2023.1081072 COPYRIGHT © 2023 Santolin, Thiele, Neubauer and Riedel. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 1 Introduction From the poles to the deep ocean basins, plastic pollution has reached every remote corner of our planet. While marine and freshwater ecosystems are threatened by up to 23 million tons of plastics entering the oceans each year, the petrochemical plastics industry is thriving and plastic production could reach over 600 million tons produced by 2030 (Borelle et al., 2020; MacLeod et al., 2021). To counter this threat, legislations must be enacted to curb plastic waste generation and promote the transition to more environmentally friendly yet competitive materials (Gutschmann et al., 2022a). A key role in this group is played by polyhydroxyalkanoates (PHAs), microbially produced bioplastics that are stored by various microorganisms from various carbon sources as energy and carbon storage compounds. The most common type of PHA is the homopolymer polyhydroxybutyrate (PHB) that was shown to degrade in various environments to CO2 and water (Narancic and O’Connor, 2019). However, this thermoplastic is very crystalline and has a 01 Frontiers in Bioengineering and Biotechnology Frontiers in Bioengineering and Biotechnology frontiersin.org 10.3389/fbioe.2023.1081072 Santolin et al. high melting point (175°C) which is close to its degradation temperature, making the processing window too small and limiting its practical application (Noda et al., 2005). To be of a real value as replacement for commodity plastics, copolymerization of HB units with longer chain length monomers is often targeted, which reduces the melting temperature and weakens the crystalline structure by steric hindrance (Grigore et al., 2019). Poly(hydroxybutyrate-co- hydroxyhexanoate) (P(HB-co-HHx)) is one of such copolymers and, while biodegradability of PHAs in each environment is strongly affected by the monomer composition and post-processing, this copolymer also shows full biodegradability in soil and seawater (Narancic and O’Connor, 2019; Riedel and Brigham, 2020; Amasawa et al., 2021). Many efforts have been made to genetically modify the PHA operon of the model organism for PHA production, Ralstonia eutropha, in order to produce P(HB-co-HHx) from related carbon sources, where HHx precursors are generated from intermediates of the ß-oxidation of fatty acids, like palm oil (Budde, et al., 2011b; Riedel et al., 2014), as well as from unrelated carbon sources such as sucrose and CO2 (Arikawa et al., 2017; Tanaka et al., 2021). The strain R. 2.4.1 Evaluation of suitable C/N ratio In the ﬁrst series of experiments, growth at four different carbon concentrations: 0.5%, 1%, 1.5%, and 2% (w v−1) of fructose and total carbon equivalent concentrations of canola oil: 0.25%, 0.5%, 0.75%, and 1% (w v−1) was evaluated with a ﬁxed amount of 0.744 g L−1 urea as nitrogen source, resulting in C/N ratios of about 5, 11, 17, and 22 (g g−1). For increased C/N ratios of about 22, 45, 68, and 90 (g g−1), the same carbon concentrations were tested with 0.186 g L−1 urea. 2.4 Deep-well-plate cultivations 24-deep-well-plates with square shape wells and a maximum volume of 11 mL (Duetz-MTPS, Adolf Kühner AG, Switzerland) were used in this study. To ensure identical cultivation conditions for the deep-well-plate replicates, 50 mL of each media with each chosen fructose to canola oil ratio was prepared and inoculated from TSB overnight cultures (see above), and then 3 mL of culture was transferred into each of the wells. To obtain a deﬁned and comparable canola oil concentration in the different wells, the medium was pre-emulsiﬁed with gum arabic (GA) before sterilization using an adapted method from Budde et al. (2011a): each medium was prepared by mixing the phosphate buffer, water and K2SO4 with the desired amount of canola oil and adding GA to a ﬁnal concentration 0.3% (w v−1). The mixture was homogenized with an Ultra-Turrax T25 (IKA-Werke GmbH and Co. KG, Germany) for 1 min at 8,000 rpm. After emulsifying the oil, the media was autoclaved, and the remaining media components were added from sterile stocks. GA was chosen as the emulsiﬁer as it has been shown not to support growth of R. eutropha (see Supplementary Figure S1). Plates were incubated for 72 h at 30°C and 225 rpm in an orbital shaker with 50 mm amplitude. Culture volume and incubation conditions were chosen according to manufacturer’s instructions to ensure sufﬁcient oxygen supply. Biological triplicates were performed for each condition. In the current study we report a very simple batch strategy that enabled to precisely control the HHx molar content in tailor-made P(HB-co-HHx) copolymers employing varying mixtures of fructose and canola oil with a ﬁxed ﬁnal carbon content for better comparison. An upscale of the method from the mL- to the L-scale starting in deep-well-plates and moving on to shake ﬂasks and ﬁnally to lab-scale bioreactors proves the robustness of our approach. 1 Introduction eutropha Re2058/pCB113 was engineered with an heterologous PHA synthase (phaC2 from Rhodococcus aethivorans) and an enoyl-coA hydratase (phaJ1 from Pseudomonas aeruginosa) to accumulate P(HB-co-HHx) when fed with raw materials containing fatty acids (Budde, et al., 2011a; Riedel et al., 2012; Saad et al., 2021; Gutschmann et al., 2022b), whereas it accumulates only PHB when fed with sugars (Santolin et al., 2021). The utilization of oily substrates by the strain is realized via the natural secretion of lipases that mediate the hydrolysis of the triacylglycerols forming natural emulsions that may also be stabilized by extracellular polysaccharides (Gutschmann et al., 2021). The great interest reported on P(HB-co-HHx) relates with the possibility of tailoring the physiochemical properties of this bioplastic targeting speciﬁc applications by adjusting the HHx molar fraction of the copolymer (Selli et al., 2022). pCB113 was streaked from a cryoculture on a TSB agar plate and incubated for 3–4 days at 30°C. A single colony from the plate was used to inoculate 10 mL TSB using a 125-mL Ultra Yield Flask (Thomson Instrument Company, United States), equipped with an AirOtop membrane (Thomson Instrument Company, United States). TSB was always supplemented with 10 μg mL−1 gentamycin sulfate and 200 μg mL−1 kanamycin sulfate. The preculture was incubated at 30°C and 200 rpm shaking speed for approximately 17 h or until an OD600 of 5 was reached. The main cultures in MSM were inoculated to an initial OD600 of 0.05. Fructose or canola oil (Edeka Zentrale AG and Co. KG, Germany) were used as carbon sources and urea was used as the sole nitrogen source in the MSM. The explicit amounts are described in the text. All chemicals were purchased from Carl Roth GmbH and Co. KG (Germany) unless stated otherwise. 2.1 Bacterial strain All experiments were conducted with the engineered R. eutropha strain Re2058/pCB113 that produces the copolymer P(HB-co-HHx) when grown on oleaginous feedstocks (Budde, et al., 2011b). The strain was stored in 20% (v v-1) glycerol at −80°C. 2.3 Calculation of C/N ratio Speciﬁc carbon and nitrogen concentrations (g L−1) and carbon- to-nitrogen ratios [C/N (g g−1)] were used for all experiments. See Supplementary Material for calculations. Frontiers in Bioengineering and Biotechnology 2.5 Shake ﬂask cultivations Four different mixtures of fructose and canola oil were selected from the deep-well-plate cultivations and upscaled to 100 mL cultures, applying the exact same cultivation strategy at this scale. Cultivations were performed in 500-mL DURAN bafﬂed ﬂasks (DWK Life Science GmbH, Germany) sealed with AirOtop membranes to ensure sufﬁcient oxygen supply. These cultivations, performed in biological triplicates, were incubated at 30°C and 200 rpm in an orbital shaker (50 mm amplitude) for 72 h and sampling was performed every 24 h. The cultivations were then repeated doubling the amount of carbon content available to 10 g L−1 but maintaining the C/N ratio by also doubling the urea concentration to 0.92 g L−1. 3 Results P(HB-co-HHx) production with fructose and canola oil mixtures was upscaled to 1-L bioreactors using six Multifors2 parallel benchtop bioreactors with two six-blade Rushton impellers (Infors AG, Switzerland). The cultivation temperature was kept constant at 30°C and the pH was maintained at 6.8 ± 0.1 using 1 M H3PO4 and 2 M NaOH for pH control. The initial stirring speed was set to 200 rpm, whereas the initial ﬂow rate was set to 0.05 vvm. Via an automatized cascade, aeration was increased up to 0.5 vvm and later stirring was increased up to 1,500 rpm in order to prevent DO values from dropping below 40%. Foam was mechanically broken as described previously (Riedel et al., 2012). Six different mixtures of fructose and canola oil, all yielding a ﬁnal carbon content of 10 g L−1 and a C/N ratio of 22 (g g−1) were used to produce sufﬁcient amounts of P(HB-co-HHx) copolymers with varying HHx monomer content for polymer characterization. 3.1 Determination of a suitable C/N ratio in deep-well-plate cultivations The choice of a suitable C/N ratio is crucial for matching the monomer composition of P(HB-co-HHx) to the substrate mixture supplied as, with excess carbon sources, only the preferred substrate will be used so that the effects of different mixing ratios will be negligible. Different C/N ratios, in the range of 5–90 (g g−1) were investigated in deep-well-plates with a working volume of 3 mL and the results are shown in Figure 1. Increasing ﬁnal biomass and PHA content values were observed with increasing C/N ratios from 5 to 22 (Figures 1A, B). For fructose, no increase in these values was observed when the C/N ratio was further increased above 22 (Figure 1C), indicating that the added fructose was not consumed. In the case of canola oil (Figure 1D), a stagnation of the achieved biomass was only observed above a C/N ratio of 68 (g g−1). PHA values showed, that cells growing on fructose accumulated only up to about 50 wt% of PHA, while cells growing on canola oil were able to accumulate up to 80 wt% of PHA, which explains why, with canola oil, more carbon source was consumed and ﬁnal biomass values increased with increasing C/N ratios above 22 (g g−1). As expected, the ﬁnal CDW values were about four times higher in the ﬁrst series of experiments (Figures 1A, B), where the added urea concentration was four times higher than in the second series of experiments (Figures 1C, D), with the same C/N ratio. 2.8 Determination of molecular weight characteristics of the produced PHA Molecular weight distribution of the PHA polymers was determined by size exclusion chromatography (SEC) from CDW samples as described previously (Thiele et al., 2021). 2.4.2 Evaluation of different fructose to canola oil mixtures The PHA content and composition of the dried cells were determined using a methanolysis protocol and gas chromatography as previously described (Bartels et al., 2020). Residual cell dry weight (RCDW) was deﬁned as CDW minus PHA content in g L−1. A regular distribution of different ratios of canola oil to fructose as well as each sole carbon source were tested to determine their effect on cell growth and PHA accumulation and composition. Seven different mixtures of fructose and canola oil, namely, 1:0, 5:1, 2:1, 1:1, 0.5:1, 0.2:1, and 0:1 [carbon ratio fructose to canola oil (g g−1)], all yielding a ﬁnal carbon content of 5 g L−1 were used in combination with 0.46 g L−1 urea to reach the selected C/N ratio of 22 (g g−1). During bioreactor cultivations, fructose and NH3 concentrations were measured from the supernatant of centrifuged samples. For fructose measurement, 750 µL of supernatant was washed twice by mixing with 750 µL of cold hexane in a 2-mL Eppendorf tube and shaking for 15 min in an overhead shaker (Rotator Drive STR4, StuartScientiﬁc, Cole-Parmer, Germany). Centrifugation was performed at 8,000 × g for 2 min and the bottom phase was collected. The washed supernatant was then ﬁltered through an 0.2 µm PES syringe ﬁlter and fructose concentration determined via HPLC-RID. Chromatography was performed with 20 µL injection volume at 80°C for 62 min on an Agilent Hi-Plex Ca column. The eluent was DI H2O with an 0.6 mL min−1 ﬂux. Unﬁltered and unwashed supernatant was measured using the Cedex Bio HT Analyzer (Cedex Bio HT Analyzer, Roche Diagnostics International AG, Switzerland) to determine NH3 consumption. 2.2 Seed train Tryptic soy broth (TSB) media, agar plates and mineral salt media (MSM) compositions have been described previously (Gutschmann et al., 2019). Ralstonia eutropha Re2058/ 02 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 Frontiers in Bioengineering and Biotechnology frontiersin.org 3.3 Upscaling and optimization of mixed substrate cultivations to shake ﬂask scale sources were to be completely consumed in order to determine the effects of different mixture ratios on the HHx content. Four of the previously tested mixtures, namely, 5:1, 1:1, 0.5:1, and 0:1 [carbon ratio fructose to canola oil (g g−1)], were scaled up, following the same cultivation strategy (C/N ratio and total carbon content), to 100 mL working volume in shake ﬂask cultivations. When utilizing the same ﬁnal carbon content as in deep-well-plates cultivations (5 g L−1), comparable biomass values were obtained (see Supplementary Figure S3). Nevertheless, in the three mixtures with the higher canola oil contents, a decrease in the PHA content between 48–72 h was observed showing also an increased HHx content at the end of the cultivation in comparison with the previous deep-well-plate experiments. To avoid premature degradation of the PHA granules and to achieve higher ﬁnal yields, it was decided to double the used carbon content to 10 g L−1 while maintaining the C/N ratio of 22 (g g−1) (see Figure 3). With this approach, comparable results were achieved in terms of PHA content and composition as with the deep-well-plate cultivations, while the ﬁnal biomass yield was approximately doubled (see Table 1). 2.7 Analytical methods For quantiﬁcation, the entire 3 mL culture was taken from the deep-well-plates at the end of the cultivations, while for cultivations in shake ﬂasks and bioreactors, 5 mL samples were taken at each sampling point. For cell dry weight (CDW) determination the samples were collected in pre-weighed 15-mL tubes and centrifuged for 15 min and 4°C at 8,000 × g. The pellets were washed with 3.5 mL cold deionized (DI) water and 1.5 mL cold hexane to remove residual oil and then resuspended again in 2 mL DI water, frozen at −80°C and dried for 48 h by lyophilization (Gamma 1–20, Martin Christ Gefriertrocknungsanlagen GmbH, Germany). For the following experiments with fructose and canola oil mixtures, a C/N ratio of 22 (g g−1) was chosen, as both carbon 03 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 FIGURE 1 Evaluation of C/N ratios in 3-mL deep-well-plates for R. eutropha Re2058/pCB113 cultivations using fructose or canola oil as carbon source and urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) achieved after 72 h of cultivation with different C/N ratios (g g−1) are shown. Left graphs show the results of low C/N ratios achieved with 0.744 g L−1 urea and different (w v−1) concentrations of fructose (A) and total-carbon-equivalent canola oil concentrations (B). Right graphs show the results of high C/N ratios obtained with the same fructose (C) and canola oil (D) concentrations and 0.186 g L−1 urea. CDW error bars indicate standard deviation from biological triplicates. PHA and HHx error bars represent standard deviation from duplicate measurements of pulled samples [in (C,D) the scarce amount of sample was only sufﬁcient for a single measurement]. FIGURE 1 FIGURE 1 Evaluation of C/N ratios in 3-mL deep-well-plates for R. eutropha Re2058/pCB113 cultivations using fructose or canola oil as carbon source and urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) achieved after 72 h of cultivation with different C/N ratios (g g−1) are shown. Left graphs show the results of low C/N ratios achieved with 0.744 g L−1 urea and different (w v−1) concentrations of fructose (A) and total-carbon-equivalent canola oil concentrations (B). 2.7 Analytical methods Right graphs show the results of high C/N ratios obtained with the same fructose (C) and canola oil (D) concentrations and 0.186 g L−1 urea. CDW error bars indicate standard deviation from biological triplicates. PHA and HHx error bars represent standard deviation from duplicate measurements of pulled samples [in (C,D) the scarce amount of sample was only sufﬁcient for a single measurement]. Frontiers in Bioengineering and Biotechnology 3.2 Evaluation of the impact of different mixtures of fructose and canola oil on the P(HB-co-HHx) composition in deep-well- plate cultivations Seven fructose and canola oil mixtures, all with a total carbon content of 5 g L−1, were used to test the effect of varying substrate ratios on the molar compositions of P(HB-co-HHx) in 24-deep- well-plates with a working volume of 3-mL (see Figure 2). Depending on the amount of canola oil, the HHx content increased linearly from 0 mol% when no oleaginous feedstock was available to 16 mol% with canola oil as the sole carbon source (see Supplementary Figure S2 for linear correlation). Comparable ﬁnal biomass values over 4 g L-1 with 65–90 wt% PHA were observed with all mixtures except when fructose was supplied as the sole carbon source. With a maximum accumulation of 57 wt% PHA, the ﬁnal biomass values of pure fructose cultures showed the lowest CDW accumulation. Frontiers in Bioengineering and Biotechnology 04 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 FIGURE 2 Impact of different fructose to canola oil mixture ratios on the composition of P(HB-co-HHx) in 3-mL deep-well-plate cultivations with R. eutropha Re2058/pCB113 using urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (wt%) and HHx content of PHA (mol%) achieved after 72 h with different mixtures with a ﬁnal carbon content of 5 g L−1 and a C/N ratio of 22 g g−1 are displayed. The carbon ratio (g g−1) of fructose to canola oil is indicated for each mixture. CDW error bars indicate standard deviation from biological triplicates. PHA and HHx error bars represent standard deviation from duplicate measurements of pulled samples. FIGURE 2 FIGURE 2 Impact of different fructose to canola oil mixture ratios on the composition of P(HB-co-HHx) in 3-mL deep-well-plate cultivations with R. eutropha Re2058/pCB113 using urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (wt%) and HHx content of PHA (mol%) achieved after 72 h with different mixtures with a ﬁnal carbon content of 5 g L−1 and a C/N ratio of 22 g g−1 are displayed. The carbon ratio (g g−1) of fructose to canola oil is indicated for each mixture. CDW error bars indicate standard deviation from biological triplicates. PHA and HHx error bars represent standard deviation from duplicate measurements of pulled samples. 3.6 Characterization of P(HB-co-HHx) copolymers Molecular weight characteristics of the produced copolymers after 72 h of cultivation in shake ﬂask and bioreactor scale were determined by size exclusion chromatography. A decrease of the molecular weight with increasing scale was noticeable, whereas no marked decrease in Mw was observed with increasing HHx contents (Table 2). Only the shake ﬂask cultivations containing 5 g L−1 carbon as substrate showed a slight decrease in the Mw of about 10% with increasing HHx molar fraction. When the ﬁnal carbon content was doubled from 5–10 g L−1 in shake ﬂask cultivations, a clear decrease in the ﬁnal Mw, up 35% was observed. In bioreactor cultivations, where higher PHA contents per CDW were reached compared to shake ﬂask cultivations, the lowest molecular weights around 3.5 × 105 Da were obtained. The polydispersity index Đ, around 2.5, was comparable along the scales. During bioreactor cultivations no clear decrease in the Mw was observed over time (see Supplementary Table S1). 3.4 Upscaling the production of molar- speciﬁc P(HB-co-HHx) to 1-L bioreactor cultivations comparable throughout all mixtures, whereby the values were slightly higher with higher canola oil contents. PHA values were comparable at all scales, with a general increase in polymer accumulation (from 60%–90%) measured with increasing canola oil contents. A linear correlation between the canola oil content in each mixture and the ﬁnal HHx molar content was observed at all scales (see Supplementary Figure S2) showing that it was possible to tailor P(HB-co-HHx) by only applying different fructose and canola oil mixtures. When higher polymer yields were obtained (shake ﬂask with 10 g L−1 CC vs. shake ﬂask with 5 g L−1 CC and bioreactor vs. shake ﬂask with 10 g L−1 CC), slightly lower molar HHx contents were observed. To demonstrate the scalability of our approach, 1-L bioreactor cultivations were carried out transferring the cultivation strategy from shake-ﬂask cultivations (constant C/N ratio and total carbon content). Again, a linear correlation was found between the amount of canola oil and fructose used and the HHx content obtained (see Supplementary Figure S2). With ﬁnal biomass values between 9–13 g L−1 and PHA contents between 60–88 wt%, the ﬁnal yields were slightly better than in the previous shake ﬂask cultures (Figure 4). The data showed that fructose was not consumed at least during the ﬁrst 24 h of cultivation, suggesting that canola oil was consumed ﬁrst, and the cells later switched to fructose as the non-preferred carbon source (Figure 5). In the three cultures with the highest fructose ratios, residual fructose concentrations were measured between 3–5 g L-1, indicating that the cells would have reached slightly higher PHA contents with slightly lower HHx molar contents if the cultivations had been operated for a longer period. In these three cultures nitrogen depletion also set in later than in cultures with higher canola oil ratios (see Supplementary Figure S4). Frontiers in Bioengineering and Biotechnology frontiersin.org 3.5 Comparison between deep-well-plate-, shake ﬂask- and bioreactor-scale When comparing the ﬁnal CDW values at all scales investigated, a slight increase in biomass yields was observed when moving from deep-well-plates to shake ﬂasks and from shake ﬂasks to bioreactor cultivations (see Table 1). Accordingly, when twice the total amount of carbon was added in the second shake ﬂask run, the ﬁnal CDWs approximately doubled. Within each scale, biomass values were 05 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 FIGURE 3 100-mL shake ﬂask cultivations with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen source. Final yields of cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) after 72 h are shown in (A) as well as values every 24 h for each cultivation with 5: 1 (B), 1: 1 (C), 0.5: 1 (D), and 0: 1 (E) fructose to canola oil ratio. All cultivations had a ﬁnal carbon content of 10 g L−1 with a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. Error bars indicate standard deviation from biological triplicates. FIGURE 3 100-mL shake ﬂask cultivations with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen source. Final yields of cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) after 72 h are shown in (A) as well as values every 24 h for each cultivation with 5: 1 (B), 1: 1 (C), 0.5: 1 (D), and 0: 1 (E) fructose to canola oil ratio. All cultivations had a ﬁnal carbon content of 10 g L−1 with a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. Error bars indicate standard deviation from biological triplicates. ake ﬂask cultivations with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen FIGURE 3 100-mL shake ﬂask cultivations with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen source. 3.5 Comparison between deep-well-plate-, shake ﬂask- and bioreactor-scale Final yields of cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) after 72 h are shown in (A) as well as values every 24 h for each cultivation with 5: 1 (B), 1: 1 (C), 0.5: 1 (D), and 0: 1 (E) fructose to canola oil ratio. All cultivations had a ﬁnal carbon content of 10 g L−1 with a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. Error bars indicate standard deviation from biological triplicates. Frontiers in Bioengineering and Biotechnology frontiersin.org 4 Discussion eutropha Re2058/pCB113 using mixtures of fructose and canola oil as carbon source and urea as nitrogen source with an applied C/N ratio of 22 g g-1. For deep-well-plate cultivations, CDW measurements represent means from triplicate cultivations and PHA and HHx measurements represent means from duplicate measurements of pulled samples. For shake ﬂask cultivations, measurements represent means of triplicate cultivations. ± are indicating standard deviation. Fructose: Canola oil [g g−1] Scale Total carbon [g L−1] CDW [g L−1] PHA [wt%] HHx [mol%] 1: 0 3-mL deep-well-plate 5 3.4 ± 0.2 56.9 ± 2.0 0.0 10: 1 1-L bioreactor 10 9.7 66.5 2.3 5: 1 3-mL deep-well-plate 5 4.1 ± 0.8 63.9 ± 3.7 3.7 ± 0.3 100-mL shake ﬂask 5 3.9 ± 0.3 57.3 ± 4.7 4.1 ± 0.5 100-mL shake ﬂask 10 7.1 ± 0.1 61.3 ± 1.3 4.4 ± 0.2 1-L bioreactor 10 9.0 58.6 4.3 2: 1 3-mL deep-well-plate 5 4.8 ± 0.3 78.1 ± 0.9 5.7 ± 0.2 1-L bioreactor 10 10.5 60.4 6.5 1: 1 3-mL deep-well-plate 5 5.1 ± 0.8 79.7 ± 1.8 8.9 ± 0.2 100-mL shake ﬂask 5 4.6 ± 0.4 63.6 ± 2.5 9.2 ± 0.2 100-mL shake ﬂask 10 8.1 ± 0.3 69.7 ± 2.1 8.9 ± 0.2 1-L bioreactor 10 10.9 70.4 7.5 0.5: 1 3-mL deep-well-plate 5 4.0 ± 0.5 76.4 ± 1.9 11.2 ± 0.2 100-mL shake ﬂask 5 4.7 ± 0.2 65.3 ± 3.0 14.6 ± 0.2 100-mL shake ﬂask 10 9.0 ± 0.5 73.1 ± 2.4 12.6 ± 0.3 1-L bioreactor 10 12.3 75.7 11.4 0.2: 1 3-mL deep-well-plate 5 4.6 ± 1.3 77.6 ± 4.8 14.3 ± 0.7 0: 1 3-mL deep-well-plate 5 4.4 ± 0.8 89.7 ± 2.3 16.1 ± 0.7 100-mL shake ﬂask 5 4.0 ± 0.3 66.5 ± 2.6 21.9 ± 1.4 100-mL shake ﬂask 10 8.6 ± 0.6 72.9 ± 1.3 16.5 ± 0.8 1-L bioreactor 10 12.9 88.0 14.3 et al., 2016; Purama et al., 2018). Using CPKO in combination with oil palm tree trunk sap in shake ﬂask cultivations, Murugan et al. obtained from 31 up to 68 wt% of PHA at 4.2–7.1 g L−1 CDW, and comparatively higher HHx molar ratios from 14–27 mol% (Murugan et al., 2016). Frontiers in Bioengineering and Biotechnology 4 Discussion scales (7–13 g L−1 CDW and 60–88 wt% PHA respectively). The strategy succeeded in tuning HHx monomer contents from 2–17 mol%, showing a linear correlation, validated at all scales, with the canola oil content in each mixture. We have developed a simple and robust batch strategy to control the molar HHx content in P(HB-co-HHx) from the 3-mL deep-well- plate to 1-L bioreactor scale. Since R. eutropha Re2058/ pCB113 produces the copolymer only when grown on oleaginous feedstocks but not when using sugars (Budde et al., 2011b), mixtures of fructose and canola oil were chosen to tune the monomer composition of P(HB-co-HHx). While most published studies to date vary the concentration of one substrate while using a ﬁxed concentration of the respective sugar or oleaginous feedstock (Murugan et al., 2016; Murugan et al., 2017; Purama et al., 2018), we orientated our studies on consistently using the same total carbon content and C/N ratio for all mixtures, achieving comparable CDWs and PHA contents with all mixtures that were reproducible along all It has been shown that carbon sources containing a higher abundance of MCFAs (medium-chain fatty acids, 6–12 carbons) lead to a higher incorporation of HHx precursors than using plant oils holding LCFAs (long-chain fatty acids, 13–21 carbons) like canola oil as used in this study (Mifune et al., 2008; Budde et al., 2011a). Per fatty acid, only one molecule of 3HHx-CoA can be formed, thus shorter fatty acids lead to a lower ratio of 3HB-CoA to 3HHx-CoA as fewer acetyl-CoA molecules are released from β-oxidation (Riedel et al., 2014). Date seed oil with 19.1% C12:0 (lauric acid) or crude palm kernel oil (CPKO) containing 3% C8:0, 3% C10:0%, and 48% C12:0 produced P(HB-co-HHx) with 39 and 44 mol% HHx, respectively (Murugan 06 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 TABLE 1 Comparison of biomass, PHA content and composition obtained among all studied scales with R. eutropha Re2058/pCB113 using mixtures of fructose and canola oil as carbon source and urea as nitrogen source with an applied C/N ratio of 22 g g-1. For deep-well-plate cultivations, CDW measurements represent means from triplicate cultivations and PHA and HHx measurements represent means from duplicate measurements of pulled samples. For shake ﬂask cultivations, measurements represent means of triplicate cultivations. ± are indicating standard deviation. TABLE 1 Comparison of biomass, PHA content and composition obtained among all studied scales with R. frontiersin.org 4 Discussion 10.3389/fbioe.2023.1081072 FIGURE 5 Comparison of CDW, PHA, and HHx accumulation- and fructose FIGURE 4 Final yields of 1-L bioreactor cultivations for tailor-made P(HB- co-HHx) production with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) after 72 h are shown for each mixture with a ﬁnal carbon content of 10 g L−1 and a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. FIGURE 4 Final yields of 1-L bioreactor cultivations for tailor-made P(HB- co-HHx) production with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) after 72 h are shown for each mixture with a ﬁnal carbon content of 10 g L−1 and a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. FIGURE 4 FIGURE 4 Final yields of 1-L bioreactor cultivations for tailor-made P(HB- co-HHx) production with R. eutropha Re2058/pCB113 using fructose and canola oil mixtures as carbon source and urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%) and HHx content of PHA (HHx; mol%) after 72 h are shown for each mixture with a ﬁnal carbon content of 10 g L−1 and a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. When PHA storage is triggered by nutrient limitation or stress conditions, more HB precursors will be incorporated into the polymer, gradually levelling off the relative concentration of HHx monomers. Additionally, oil is preferred over sugars which contributes to the higher contents of HHx at the beginning of the cultivation when this substrate is being consumed. Monitoring of the fructose concentration in the medium showed that fructose was not consumed at least in the ﬁrst 24 h of cultivation, which was also observed in other studies (Murugan et al., 2017). 4 Discussion In a follow up study, three substrate mixtures of palm olein and fructose were chosen, to obtain P(HB-co-HHx) with lowered HHx contents from 4–15 mol% in bioreactor cultivations (Murugan et al., 2017). Here, the effect of increasing the sugar to oil ratio to effectively lower the HHx fraction was proved as palm tree trunk sap, containing only 17% fructose of the total sugars with a large fraction of glucose, was replaced by pure fructose. Ralstonia eutropha is only able to metabolize fructose and no glucose (Sichwart et al., 2011). Further, date molasses, containing over 50% fructose, and date seed oil mixtures were also used in bioreactor cultivations reaching varying CDW concentrations from 1.7–6.9 g L−1 CDW, up to 49 wt% of PHA and broader HHx molar ratios from 2–28 mol% (Purama et al., 2018). fraction of P(HB-co-HHx) on a molecular level was presented by controlling the expression of the phaJ gene of R. eutropha, involved in the generation of HHx precursors, showing that copolymers with HHx molar contents ranging from 2.8–10.7 mol% could be obtained (Miyahara et al., 2021). Arikawa et al. recently reported the tailored production of P(HB-co-HHx) with HHx contents up to 36 mol% by the deletion of the ß-ketothiolase gene together with the overexpression of the (R)-speciﬁc enoyl-Coa hydratase and PhaC synthase (Arikawa and Sato, 2022). As in the studies mentioned above, an increase in the molar HHx content was observed in mixtures with increasing oleaginous substrate concentrations. Furthermore, in accordance with the literature (Budde et al., 2011a; Riedel et al., 2015), a decrease in the HHx content was measured over the course of the cultivation. Whether growing on a pure oleaginous feedstock or in combination with fructose, during the growth phase less HB precursors will be formed in comparison to the production phase as acetyl-coA ﬂows into the TCA cycle and less HB precursors are formed. In this context, it is reported that high intracellular CoA concentrations inhibit PhaA, leading to a slower rate of HB-CoA synthesis (Oeding and Schlegel, 1973). All abovementioned studies, even if they achieved lower biomass yields due to the use of less urea as nitrogen source (about half of this study) and less comparability along the tested mixtures, are based on the same principle and prove that it is plausible to extend our strategy to other feedstocks. Another different approach to control the HHx 07 frontiersin.org Santolin et al. Frontiers in Bioengineering and Biotechnology 4 Discussion The fact that no dissolved oxygen peak (or dropping of the stirring cascade) was observed around the timepoint when the strain started to consume fructose (data not shown) as the second preferred carbon source suggests a smooth transition from one substrate to the other with the strain presumably being able to assimilate both canola oil and fructose simultaneously. In general, a higher PHA content of up to 88 wt% was obtained when the carbon source was of oleaginous origin, whereas only about 60 wt% PHA could be obtained with fructose alone as substrate. This is due to the fact that the utilization of fructose in this strain is less efﬁcient than the utilization of oleaginous feedstocks. After the conversion of fructose to two pyruvate molecules via the Entner-Doudoroff pathway, one molecule of CO2 is released for the conversion to each acetyl-CoA by the pyruvate-dehydrogenase while no carbon in the form of CO2 is lost in the ß-oxidation of oils. In addition, the strain Re2058/ pCB113 was engineered to utilize plant oils efﬁciently, boosting the synthesis of HHx monomers (Budde et al., 2011b). frontiersin.org FIGURE 5 Measurements represent means from duplicate measurements. ± are indicating minimum and maximum values. Fructose: Canola oil [g g−1] Scale Total carbon [g L−1] Mw × 105 [Da] Mn × 105 [Da] Đ [-] 10: 1 1-L bioreactor 10 3.51 1.42 2.47 5: 1 100-mL shake ﬂask 5 6.2 ± 0.1 2.5 ± 0.3 2.5 ± 0.2 100-mL shake ﬂask 10 4.0 ± 0.3 1.6 ± 0.2 2.5 ± 0.2 1-L bioreactor 10 4.0 ± 0.1 1.6 ± 0.1 2.4 ± 0.0 2: 1 1-L bioreactor 10 3.2 ± 0.0 1.3 ± 0.0 2.4 ± 0.0 1: 1 100-mL shake ﬂask 5 6.1 ± 0.1 2.8 ± 0.1 2.2 ± 0.0 100-mL shake ﬂask 10 4.8 ± 0.7 1.8 ± 0.1 2.6 ± 0.3 1-L bioreactor 10 3.5 ± 0.0 1.5 ± 0.0 2.3 ± 0.0 0.5: 1 100-mL shake ﬂask 5 5.8 ± 0.1 2.7 ± 0.1 2.2 ± 0.0 100-mL shake ﬂask 10 4.8 ± 0.3 2.2 ± 0.3 2.2 ± 0.2 1-L bioreactor 10 3.6 ± 0.1 1.7 ± 0.1 2.2 ± 0.1 0: 1 100-mL shake ﬂask 5 5.6 ± 0.1 2.5 ± 0.3 2.5 ± 0.2 100-mL shake ﬂask 10 4.2 ± 0.8 1.6 ± 0.9 2.1 ± 0.1 1-L bioreactor 10 3.4 ± 0.1 1.5 ± 0.0 2.3 ± 0.0 compositions. The strategy, which allows the HHx content to be adjusted between 2–17 mol%, proved to be scalable from the mL-scale in deep-well-plates to the L-scale in bioreactors. At all scales, high PHA contents of over 60 wt% were obtained with comparable molecular weight properties. However, to increase the overall yields of tailor-made P(HB-co-HHx), a fed-batch process needs to be developed for the different substrate mixtures. can be explained by evaporation during sterilization of the media and thereby a concentration of the carbon sources, as well as by a slight misestimation of the carbon content of canola oil (Table 1). In this study, no signiﬁcant change in the molecular weight with different HHx fractions was observed when the carbon source concentration was 10 g L−1. In the shake ﬂask cultivation, where a lower substrate concentration of 5 g L−1 was used (Supplementary Figure S3), a decrease of the molecular weight of the copolymer with increasing HHx fraction was observed, contradicting the results obtained by Murugan et al. (2017) who obtained an increasing Mw. Purama et al. Author contributions SR, IT, and LS contributed to the conception and design of the study. IT and LS carried out the experiments and analysis of the data. IT and LS prepared the ﬁrst draft of the manuscript. SR and PN were responsible for the project administration and funding acquisition. All authors contributed to the manuscript revision, read, and approved the submitted version. FIGURE 5 (2018) reported an overall decreasing trend of molecular weights between 8.3 and 5.8 × 105 Da with increasing HHx fractions between 5 and 28 mol%. Additionally, they observed a narrower polydispersity index of 1.7 compared to our study, which was around 2.17–2.61. Generally, the bulkier HHx is assumed to reduce the synthase turnover rate leading to lower molecular weights (Murugan et al., 2017). Moreover, when the substrate concentration was doubled, the glycerol concentration from oil cleavage by secreted lipases of R. eutropha also increased. Glycerol reportedly acts as a chain terminator and could thus be the cause of the lower molecular weight observed during these cultivations (Ashby et al., 2012). Frontiers in Bioengineering and Biotechnology Data availability statement The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation. FIGURE 5 Comparison of CDW, PHA, and HHx accumulation- and fructose consumption curves during 1-L bioreactor cultivations of R. eutropha Re2058/pCB113 for tailor-made P(HB-co-HHx) production using mixtures with decreasing fructose to canola oil ratios as carbon source and urea as nitrogen source. Cell dry weight (CDW; g L−1), PHA content of CDW (PHA; wt%), HHx content of PHA (HHx; mol%) and fructose concentration (Fructose; g L−1) values are shown over the course of the cultivation for each mixture with a ﬁnal carbon content of 10 g L−1 and a C/N ratio of 22 g g−1. The carbon ratio of fructose to canola oil (g g−1) is indicated for each mixture. When higher polymer yields were obtained (shake ﬂasks with 5 g L−1 carbon content vs. shake ﬂasks with 10 g L−1 carbon content and shake ﬂasks with 10 g L−1 carbon content vs. bioreactor cultivations), slightly lower molar HHx contents were observed. This could be due to the degradation of the HB-rich polymer ends after 48 h in the ﬁrst case, which was observed when only half of carbon was applied, presumably due to depletion of the carbon sources. In the second case, a faster growth in the bioreactor in comparison to the shake ﬂasks supported by a better overall physiological state (pH- and O2-control) may have enabled the cells to further consume the fructose present in the media, leading to more HB-monomers being incorporated to the polymer chain, thus decreasing the ﬁnal HHx content. During cultivations in the bioreactor, CDW values were obtained that were higher than the theoretical yield, which 08 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 TABLE 2 Molecular weight characterization of samples after 72 h of 100-mL shake ﬂask and 1-L bioreactor cultivations with R. eutropha Re2058/pCB113 using mixtures of fructose and canola oil and urea as nitrogen source with an applied C/N ratio of 22 g g-1. Mw = weight-average molecular weight, Mn = number- average molecular weight, Đ = polydispersity index. Measurements represent means from duplicate measurements. ± are indicating minimum and maximum l TABLE 2 Molecular weight characterization of samples after 72 h of 100-mL shake ﬂask and 1-L bioreactor cultivations with R. eutropha Re2058/pCB113 using mixtures of fructose and canola oil and urea as nitrogen source with an applied C/N ratio of 22 g g-1. Mw = weight-average molecular weight, Mn = number- average molecular weight, Đ = polydispersity index. frontiersin.org References ABA3656 Narancic, T., and O’Connor, K. E. (2019). Plastic waste as a global challenge: Are biodegradable plastics the answer to the plastic waste problem? Microbiology 165 (2), 129–137. doi:10.1099/mic.0.000749 Budde, C. F., Riedel, S. L., Hübner, F., Risch, S., Popović, M. K., Rha, C., et al. (2011a). Growth and polyhydroxybutyrate production by Ralstonia eutropha in emulsiﬁed plant oil medium. Appl. Microbiol. Biotechnol. 89 (5), 1611–1619. doi:10.1007/s00253-011- 3102-0 Noda, I., Green, P. R., Satkowski, M. M., and Schechtman, L. A. (2005). Preparation and properties of a novel class of polyhydroxyalkanoate copolymers. Biomacromolecules 6 (2), 580–586. doi:10.1021/bm049472m Budde, C. F., Riedel, S. L., Willis, L. B., Rha, C., and Sinskey, A. J. (2011b). Production of poly (3-hydroxybutyrate-co-3-hydroxyhexanoate) from plant oil by engineered Ralstonia eutropha strains. Appl. Environ. Microbiol. 77 (9), 2847–2854. doi:10.1128/ AEM.02429-10 Oeding, V., and Schlegel, H. G. (1973). β-Ketothiolase from Hydrogenomonas eutropha H16 and its signiﬁcance in the regulation of poly-β-hydroxybutyrate metabolism. Biochem. 134 (1), 239–248. doi:10.1042/bj1340239 Grigore, M. E., Grigorescu, R. M., Iancu, L., Ion, R. M., Zaharia, C., and Andrei, E. R. (2019). Methods of synthesis, properties and biomedical applications of polyhydroxyalkanoates: A review. J. Biomater. Sci. Polym. Ed. 30 (9), 695–712. doi:10.1080/09205063.2019.1605866 Grigore, M. E., Grigorescu, R. M., Iancu, L., Ion, R. M., Zaharia, C., and Andrei, E. R. (2019). Methods of synthesis, properties and biomedical applications of polyhydroxyalkanoates: A review. J. Biomater. Sci. Polym. Ed. 30 (9), 695–712. doi:10.1080/09205063.2019.1605866 Gutschmann, B., Bock, M. C. E., Jahns, S., Neubauer, P., Brigham, C. J., and Riedel, S. L. (2021). Untargeted metabolomics analysis of Ralstonia eutropha during plant oil cultivations reveals the presence of a fucose salvage pathway. Sci. Rep. 11 (14267), 14267. doi:10.1038/s41598-021-93720-9 Purama, R. K., Al-Sabahi, J. N., and Sudesh, K. (2018). Evaluation of date seed oil and date molasses as novel carbon sources for the production of poly(3Hydroxybutyrate-co- 3Hydroxyhexanoate) by Cupriavidus necator Re2058/pCB113. Ind. Crops Prod. 119, 83–92. doi:10.1016/j.indcrop.2018.04.013 Gutschmann, B., Bock, M. C. E., Jahns, S., Neubauer, P., Brigham, C. J., and Riedel, S. L. (2021). Untargeted metabolomics analysis of Ralstonia eutropha during plant oil cultivations reveals the presence of a fucose salvage pathway. Sci. Rep. 11 (14267), 14267. doi:10.1038/s41598-021-93720-9 Riedel, S. L., Bader, J., Brigham, C. J., Budde, C. F., Yusof, Z. A. M., Rha, C., et al. (2012). Production of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) by Ralstonia eutropha in high cell density palm oil fermentations. Biotechnol. Bioeng. 109 (1), 74–83. References Gutschmann, B., Schiewe, T., Weiske, M. T. H., Neubauer, P., Hass, R., and Riedel, S. L. (2019). In-line monitoring of polyhydroxyalkanoate (PHA) production during high- cell-density plant oil cultivations using photon density wave spectroscopy. Bioengineering 6 (3), 85. doi:10.3390/bioengineering6030085 Amasawa, E., Yamanishi, T., Nakatani, J., Hirao, M., and Sato, S. (2021). Climate change implications of bio-based and marine-biodegradable plastic: Evidence from poly (3-hydroxybutyrate-co-3-hydroxyhexanoate). Environ. Sci. Technol. 55 (5), 3380–3388. doi:10.1021/acs.est.0c06612 MacLeod, M., Arp, H. P. H., Tekman, M. B., and Jahnke, A. (2021). The global threat from plastic pollution. Science 373 (6550), 61–65. doi:10.1126/science.abg5433 Arikawa, H., Matsumoto, K., and Fujiki, T. (2017). Polyhydroxyalkanoate production from sucrose by Cupriavidus necator strains harboring csc genes from Escherichia coli W. Appl. Microbiol. Biotechnol. 101 (20), 7497–7507. doi:10.1007/s00253-017-8470-7 Mifune, J., Nakamura, S., and Fukui, T. (2008). Targeted engineering of Cupriavidus necator chromosome for biosynthesis of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) from vegetable oil. Can. J. Chem. 86 (6), 621–627. doi:10.1139/V08-047 Arikawa, H., and Sato, S. (2022). Impact of various ß-ketothiolase genes on PHBHHx production in Cupriavidus necator H16 derivatives. Appl. Microbiol. Biotechnol. 106, 3021–3032. doi:10.1007/s00253-022-11928-9 Miyahara, Y., Hiroe, A., Sato, S., Tsuge, T., and Taguchi, S. (2021). “Microbial polyhydroxyalkanoates (PHAs): From synthetic Biology to industrialization,” in Biopolymers for biomedical and biotechnological applications. Editors B. Rehm and M. F. Moradali (Wiley Online Library, Wiley-WCH, Weinheim), 231–264. doi:10.1002/9783527818310.ch8 Ashby, R. D., Solaiman, D. K. Y., Strahan, G. D., Zhu, C., Tappel, R. C., and Nomura, C. T. (2012). Glycerine and levulinic acid: Renewable co-substrates for the fermentative synthesis of short-chain poly(hydroxyalkanoate) biopolymers. Bioresour. Technol. 118, 272–280. doi:10.1016/j.biortech.2012.05.092 Murugan, P., Chhajer, P., Kosugi, A., Arai, T., Brigham, C. J., and Sudesh, K. (2016). Production of P(3HB-co-3HHx) with controlled compositions by recombinant Cupriavidus necator Re2058/pCB113 from renewable resources. Clean. (Weinh) 44 (9), 1234–1241. doi:10.1002/clen.201500714 Bartels, M., Gutschmann, B., Widmer, T., Grimm, T., Neubauer, P., and Riedel, S. L. (2020). Recovery of the PHA copolymer P(HB-co-HHx) with non-halogenated solvents: Inﬂuences on molecular weight and HHx-content. Front. Bioeng. Biotechnol. 8, 944. doi:10.3389/fbioe.2020.00944 Murugan, P., Gan, C. Y., and Sudesh, K. (2017). Biosynthesis of P(3HB-co-3HHx) with improved molecular weights from a mixture of palm olein and fructose by Cupriavidus necator Re2058/pCB113. Int. J. Biol. Macromol. 102, 1112–1119. doi:10. 1016/j.ijbiomac.2017.05.006 Borelle, S. B., Ringma, J., Law, K. L., Monnahan, C. C., Lebreton, L., McGivern, A., et al. (2020). Predicted growth in plastic waste exceeds efforts to mitigate plastic pollution. Science 369 (6510), 1515–1518. doi:10.1126/SCIENCE. Funding The simple and robust approach presented in this study using mixtures of fructose and canola oil can be used to produce P(HB-co-HHx) with precisely controlled This research was supported by the German Federal Ministry of Education and Research, grant number 031B0833A. 09 frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 Acknowledgments All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. We thank Professor Anthony Sinskey from MIT for providing the engineered R. eutropha strain used in this study. We thank Björn Gutschmann for helpful discussions. We acknowledge support by the German Research Foundation and the Open Access Publication Fund of TU Berlin. Conﬂict of interest The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbioe.2023.1081072/ full#supplementary-material The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Thiele, I., Rose, J. H., Gutschmann, B., Tofaily, S., Schöttel, J., and Berlin, T. U. (2021). Scale-up of the downstream process for polyhydroxyalkanoate copolymer P(HB-co- HHx): Extraction with nonhalogenated solvents. J. Sib. Fed. Univ. Biol. 14 (4), 454–464. doi:10.17516/1997-1389-0364 Riedel, S. L., Lu, J., Stahl, U., and Brigham, C. J. (2014). Lipid and fatty acid metabolism in Ralstonia eutropha: Relevance for the biotechnological production of value-added products. Appl. Microbiol. Biotechnol. 98 (4), 1469–1483. doi:10.1007/ s00253-013-5430-8 Selli, F., Hufenus, R., Gooneie, A., Halis, U., and Perret, E. (2022). Structure – property relationship in melt-spun poly(hydroxybutyrate-co-3-hexanoate) monoﬁlaments. Polymers 14, 200–217. doi:10.3390/polym14010200 Sichwart, S., Hetzler, S., Bröker, D., and Steinbüchel, A. (2011). Extension of the substrate utilizationrangeofRalstonia eutropha strainH16by metabolic engineering to includemannose and glucose. Appl. Environ. Microbiol. 77 (4), 1325–1334. doi:10.1128/AEM.01977-10 Santolin, L., Waldburger, S., Neubauer, P., and Riedel, S. L. (2021). Substrate-ﬂexible two-stage fed-batch cultivations for the production of the PHA copolymer P(HB-co- HHx) with Cupriavidus necator Re2058/pCB113. Front. Bioeng. Biotechnol. 9, 623890–623912. doi:10.3389/fbioe.2021.623890 Saad, V., Gutschmann, B., Grimm, T., Widmer, T., Neubauer, P., and Riedel, S. L. (2021). Low-quality animal by-product streams for the production of PHA- biopolymers: Fats, fat/protein-emulsions and materials with high ash content as low-cost feedstocks. Biotechnol. Lett. 43 (3), 579–587. doi:10.1007/s10529-020- 03065-y Tanaka, K., Yoshida, K., Orita, I., and Fukui, T. (2021). Biosynthesis of poly(3- hydroxybutyrate-co-3-hydroxyhexanoate) from CO2 by a recombinant Cupriavidus necator. Bioengineering 8 (11), 179–210. doi:10.3390/bioengineering8110179 References doi:10.1002/bit.23283 Gutschmann, B., Huang, B., Santolin, L., Thiele, I., Neubauer, P., and Riedel, S. L. (2022a). Native feedstock options for the polyhydroxyalkanoate industry in europe: A review. Microbiol. Res. 264 (127177), 127177. doi:10.1016/j.micres. 2022.127177 Riedel, S. L., and Brigham, C. J. (2020). “Inexpensive and waste raw materials for PHA production,” in The handbook of polyhydroxyalkanoates. Editor M. Koller (CRC Press Taylor and Francis Group, CRC Press, Boca Raton), 203–221. doi:10.1201/ 9780429296611-10 Gutschmann, B., Maldonado Simões, M., Schiewe, T., Schröter, E. S., Münzberg, M., Neubauer, P., et al. (2022b). Continuous feeding strategy for polyhydroxyalkanoate production from solid waste animal fat at laboratory- and pilot-scale. Microb. Biotechnol. 00, 295–306. doi:10.1111/1751-7915.14104 Riedel, S. L., Jahns, S., Koenig, S., Bock, M. C. E., Brigham, C. J., Bader, J., et al. (2015). Polyhydroxyalkanoates production with Ralstonia eutropha from low quality waste animal fats. J. Biotechnol. 214, 119–127. doi:10.1016/j.jbiotec. 2015.09.002 10 Frontiers in Bioengineering and Biotechnology frontiersin.org Santolin et al. 10.3389/fbioe.2023.1081072 Selli, F., Hufenus, R., Gooneie, A., Halis, U., and Perret, E. (2022). Structure – property relationship in melt-spun poly(hydroxybutyrate-co-3-hexanoate) monoﬁlaments. Polymers 14, 200–217. doi:10.3390/polym14010200 Saad, V., Gutschmann, B., Grimm, T., Widmer, T., Neubauer, P., and Riedel, S. L. (2021). Low-quality animal by-product streams for the production of PHA- biopolymers: Fats, fat/protein-emulsions and materials with high ash content as low-cost feedstocks. Biotechnol. Lett. 43 (3), 579–587. doi:10.1007/s10529-020- 03065-y Saad, V., Gutschmann, B., Grimm, T., Widmer, T., Neubauer, P., and Riedel, S. L. (2021). Low-quality animal by-product streams for the production of PHA- biopolymers: Fats, fat/protein-emulsions and materials with high ash content as low-cost feedstocks. Biotechnol. Lett. 43 (3), 579–587. doi:10.1007/s10529-020- 03065-y Santolin, L., Waldburger, S., Neubauer, P., and Riedel, S. L. (2021). Substrate-ﬂexible two-stage fed-batch cultivations for the production of the PHA copolymer P(HB-co- HHx) with Cupriavidus necator Re2058/pCB113. Front. Bioeng. Biotechnol. 9, 623890–623912. doi:10.3389/fbioe.2021.623890 11 frontiersin.org Frontiers in Bioengineering and Biotechnology
https://openalex.org/W4361263982	https://figshare.com/articles/journal_contribution/Supplementary_Figure_5_from_Nutlin-3a_Activates_p53_to_Both_Down-regulate_Inhibitor_of_Growth_2_and_Up-regulate_i_mir-34a_mir-34b_i_and_i_mir-34c_i_Expression_and_Induce_Senescence/22373691/1/files/39818928.pdf	Malagasy	null	Supplementary Figure 5 from Nutlin-3a Activates p53 to Both Down-regulate Inhibitor of Growth 2 and Up-regulate <i>mir-34a, mir-34b,</i> and <i>mir-34c</i> Expression, and Induce Senescence	null	2,023	cc-by	6	Supplementary Figure S5 Supplementary Figure S5
W4220943822.txt	https://www.researchsquare.com/article/rs-1387715/latest.pdf	en		Genetic and Immunological Characterization of Commercial Infectious Bronchitis Virus (IBV) Vaccines Used in Korea	Research Square (Research Square)	2,022	cc-by	6,460	Genetic and Immunological Characterization of Commercial Infectious Bronchitis Virus (IBV) Vaccines Used in Korea Jeongho Park (  jhp@kangwon.ac.kr ) Kangwon National University https://orcid.org/0000-0002-0340-9181 Ji Seung Jung Kangwon National University Rangyeon Lee Kangwon National University Sung-il Yoon Kangwon National University Geun-Shik Lee Kangwon National University Hann Woo Sung Kangwon National University Hyuk Moo Kwon Kangwon National University Research Article Keywords: Posted Date: March 22nd, 2022 DOI: https://doi.org/10.21203/rs.3.rs-1387715/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/19 Abstract Infectious bronchitis virus (IBV) is the avian coronavirus that was first isolated eight decades ago and it has remained as a major contagious pathogen in the poultry industry. Although vaccination is considered the most effective preventive strategy, continuously emerging variants prohibit vaccine efficacy and cross protection. In the last two decades, because nephropathogenic strains have been prevalent, the basic structure of numerous live attenuated vaccines was based on them. For a better understanding of their antigenic and immunogenic properties, this study obtained three commercial IBV vaccines commercially available in Korea and analyzed their genetic and immunogenic features. S1 gene sequence comparison with other field strains clustered those vaccine strains into the GI-19 or GI-15 lineage. The genetic variance of vaccine strains from their parental viruses was mostly detected in HVRI. Although each vaccine strain targets both respiratory and nephrotropic IBVs, their amino acid residues were not comparable at multiple locations. Conversely, antigenic stimulation with commercial vaccines and regional IBV variants was not sufficient to modify major immune cell phenotypes. Our study suggests that vaccine selection should be carefully considered according to their structural background because genetic variance can still exist even during the manufacturing process. Further investigation of the precise immune response by each vaccine strain can improve quarantine strategies. Introduction Infectious bronchitis virus (IBV) is the first isolated coronaviruses in the 1930s and has caused enormous economic damage globally. Among its four major structural proteins, the nucleocapsid (N), membrane (M), envelope (E), and spike (S) proteins, the S glycoprotein mediates the core events of infection, such as host invasion, host range, cell tropism, and induction of host immune responses. The S glycoprotein is further cleaved into two subunits. The S1 subunit mediates host cell attachment, which initiates immune responses, and the S2 subunit bridges the viral membrane with the S1 subunit. RNA viruses mutate faster than do DNA viruses and positive-sense single-stranded RNA viruses such as IBV have the highest mutation and substitution rates among all RNA viruses [1, 2, 10, 32]. Specifically, genetic mutation and recombination of IBV occur continuously in hypervariable regions (HVRs) of the S1 glycoprotein because of poor viral polymerase proofreading mechanism. HVRs are divided into three parts: HVR I (amino acid residues 38 ~ 67), HVR II (amino acid residues 91 ~ 141), and HVR III (amino acid residues 274 ~ 387). This results in the continuous emergence of variants, which prohibits vaccine efficacy and cross protection [1, 2, 25, 32]. The IBV variants are clustered into six genetically divergent groups and are further classified into 32 lineages that include genotype I (GI), with 27 lineages, and other five genotypes that are dependent on the S1 glycoprotein sequence [38]. IBV infection prevention can be managed by optimizing the biosecurity of the housing environment. For example, ventilation and light management and the control of flock size can reduce the risk of IBV spread in farms [8]. However, the most effective strategy for protection against this virus is thought to be achieved by vaccination. Compare with other farm animals, the life-span of broilers is much shorter and intensive vaccination is given at a very young age. Because of the heterogenicity of variants, IBV vaccines Page 2/19 have been developed based on the antigenicity of regional isolates. The original IBV vaccine was designed from the Mass serotype, which belongs to the GI-1 lineage and had been considered a core strain in the IBV vaccine platform. However, novel variants, such as nephropathogenic strains, were detected that induce a complicated pathogenesis. They initially invade the respiratory epithelium where they begin replicating. Then, the pathogen penetrates deeper epithelial layer and further spreads to remote organs, such as the kidney, liver, and spleen via blood vessels [30]. These novel IBVs originated from the QX strain from China, and QX-like strains were further spread to European and Asian countries. Most of them are clustered into the GI-19 lineage [21, 40, 41]. This prompted a debate regarding the benefit of heterologous vs. homologous vaccination because of possible divergent and unexpected immunity to heterogeneous field strains. Moreover, the risk of vaccine-escape variants exist [11]. Thus, the introduction of the protectotype concept, which combines two heterologous vaccine strains for an additive protection effect has been suggested [15]. The major IBV vaccine type is live attenuated, which is constituted according to emerging variants. Therefore, a critical step in vaccine development is the identification of field strains, antigenicity of which determines vaccine structure. Both inappropriate vaccine selection and a poor application strategy in the field tend to increase the risk of generation of unexpected variants. For instance, massive-delivery methods lower vaccine efficiency which can lead to the mutation and recombination of field viruses and vaccine strains. Although a proper vaccine strain is chosen, its possible reversion to virulence can accelerate novel virus derivation [12, 13]. Consequently, frequent vaccine alterations and lowered cross protection can render protection difficult. After the initial IBV outbreak of 1986, live attenuated vaccines derived from the Mass strain (H120 and Ma5) were widely used in Korea up to the late 1990s. The subsequent emergence of nephrotropic strains prompted the development of the K2 vaccine strain, which is based on a Korean native strain, KM91. In the 2000s, a novel respiratory strain (KI strain) was detected in Korea and other new variants were reported thereafter [7, 34]. We also analyzed the genetic feature of the novel IBV strains, K046-12 and K047-12, which were correlated with other Korean isolates and QX-like strains, respectively [24]. Despite the development of numerous vaccines for long-lasting immunity and cross protection against multiple viral strains, the repeated emergence of variants has increased the demand for restored vaccine development. For better control of IBV and execution of vaccine strategies, evaluation of the genetic stability and subpopulations of commercial vaccines is necessary. To this end, we analyzed the genetic sequence of the S1 glycoprotein in three commercial vaccine strains. Using a phylogenetic analysis, we examined the genetic closeness between vaccine strains and their parental strains and regional variants. The genetic stability of vaccine strains was evaluated by comparing their amino acid sequences. Moreover, the immune responses elicited by antigens from vaccine strains and regional variants were examined. Materials And Methods Vaccines and viruses Page 3/19 Three commercial live-attenuated IBV vaccines produced by different manufacturers were collected. These vaccines included the K40/09, the Kr/D85/06, and a recombination of the KM91 and QX IBV strains. Here, they were coded as A, B and C, respectively, and their features and GenBank accession numbers are shown in Table 1. As previously reported, the lyophilized vaccines were reconstituted in PBS for the sequence analysis [39]. In addition, Vaccine B was propagated in 9 ~ 11-day-old specific pathogen free (SPF) embryonated chicken eggs (Sungmin Farm, Korea) and a single passage of Vaccine B was performed. Table 1 List of the commercial vaccine strains selected for this study. The parental strains, target, and GenBank accession numbers of each vaccine are indicated. Name Parental strain Target GenBank ID Vaccine A K40/09 Respiratory and renal variants OM685193 Vaccine B D85/06 Respiratory and renal variants OM685194 Vaccine B_single passage D85/06 Respiratory and renal variants OM685195 Vaccine C KM91 & QX-like Respiratory and renal variants OM685196 PCR and cloning of the S1 gene Total RNA of each vaccine was isolated using the TRIzol® Reagent (Invitrogen™) and then reverse transcribed into cDNA with AccuPower® RT PreMix (BIONEER) using an S1000™ Thermal Cycler (BioRad) according to the manufacturer’s instructions. Each reaction was carried out in a final volume of 20 µl and the resultant cDNA was immediately used for conventional PCR or stored at -70°C for later use. The S1 gene of the vaccines was amplified by conventional PCR using AccuPower® PCR PreMix (Bioneer). The primer sets (forward: 5′-TAG TGA CCC TTT TGT GTG CAC TAT-3′ and reverse: 5′-GTT TGT ATG TAC TCA TCT GTA AC-3′) were used as reported previously [23]. PCR was performed using a PCR Thermal Cycler Dice™ Touch (TaKaRa Bio Inc., Australia) and the following conditions: predenaturation at 94°C for 5 min; followed by 35 cycles of denaturation (94°C, 90 s), annealing (45°C, 30 s), and polymerization (72°C, 90 s); and a postpolymerization step was at 72°C for 3 min. The amplified sequences were analyzed by 1% agarose gel electrophoresis. PCR products from the three vaccines and the single-passaged Vaccine B were purified using a the Gel Extraction kit (QIAGEN), ligated into the TOPO TA pCR®2.1 vector and transformed into One Shot®TOP10 competent E.coli (Invitrogen), as recommended by the manufacturer. Transformants were grown in Luria-Bertani (LB) medium supplemented with kanamycin (50 ng/µl). Plasmid purification was performed using a DNA-spin Plasmid DNA purification kit (iNtRON) and the isolates were sent to Macrogen Inc. (Seoul, South Korea) for sequencing using the universal primers M13F (5′-GTA AAA CGA CGG CCA GT-3′) and M13R-pUC (5′-CAG GAA ACA GCT ATG AC-3′). Sequence comparisons and phylogenetic analysis Page 4/19 The sequences received from Macrogen were analyzed using the Lasergene package version 10 (DNASTAR, Inc., USA) and deposited in GenBank as follows: Vaccine A, OM685193; Vaccine B, OM685194; Vaccine B_single passage, OM685195; Vaccine C, OM685196. The nucleotide sequences were aligned via the Clustal W method using the DNASTAR 2.0 MegAlign program and the phylogenetic tree was constructed via the neighbor-joining method with 1,000 bootstrap replicates using MEGA X software (version 10.2.2). The nucleotide sequences of the S glycoproteins, for comparison, were downloaded from the NCBI GenBank database as follows: Conn48725 (FJ899692.1), Beaudette (M95169.1), M41 (AY561711.1), H120 (FJ888351.1), K046-12 (MK618758.1), BP-CaKII (MF924724.1), QXIBV (AF193423.1), K40/09 (HM486957.1), K047-12 (MK618759.1), ArkDPI (AF006624.1), GX2-98 (AY251816.1), K210-02 (AY257068.1), and Kr/D85/06 (EF621400.1). The chromatograms were obtained and reviewed manually using BioEdit version 7.2.5 to detect the presence of viral subpopulations in the vaccines. Immune cell preparation and culture Total splenocytes were collected from 4 ~ 5-week-old SPF chickens (Sungmin Farm, Korea). After RBS lysis, cells were seeded into 96-well U-bottom plates (5 x 105 cells/well) and cultured for 4 ~ 5 days with Concanavalin A (5 µg/ml) (Sigma Aldrich, MO) and chicken IL-2 (10 ng/ml) (Kingfisher Biotech, MN, USA) in RPMI-1640 supplemented with 10% FBS. For antigenic stimulation, vaccines and IBV variants suspension (K046-12 and K047-12, 10 µL/mL) were added to the culture. The animal work was approved by the Institutional Animal Care and Use Committee (IACUC) of Kangwon National University (No: KW211021-6). Cell staining and flow cytometry Cultured splenocytes were harvested and stained with the LIVE/DEAD Fixable Aqua Dead Cell Stain (L3457; Thermo Fisher) and antibodies specific for CD4 (CT-4, SouthernBiotech), CD8α (CT-8, SouthernBiotech), CD44 (AV6, SouthernBiotech), Monocyte/Macrophage (KUK01, SouthernBiotech), MHCII (2G11, SouthernBiotech), Bu-1 (AV20, SouthernBiotech). Stained cells were analyzed using flow cytometry (Cytoflex, Beckman Coulter). Statistics Dunnett’s test with ANOVA was used to assess the statistical significance of the results obtained from the viral-antigen-stimulated groups compared with the untreated group. Comparisons among the stimulated groups were analyzed by one-way ANOVA. Significance was set as P < 0.05. Results Cloning of the S1 glycoprotein gene from commercial vaccine strains Page 5/19 The PCR products of the S1 glycoprotein gene from three different commercial vaccine strains and the single-passaged commercial vaccine strain were inserted into the pCR®2.1 vector, which contains a kanamycin resistance ORF region (Fig. 1A). The inserted PCR product (~ 1,400 bp) was verified by digestion of cloned plasmids with EcoRI, followed by gel electrophoresis (Fig. 1B). This procedure confirmed that the suitable preparation for the DNA sequencing of the S1 gene. Classification of vaccine strains and IBV variants based on S1 glycoprotein gene Using the determined sequences of the full-length S1 gene in commercial vaccines (Fig. S1), we compared genetic relatedness to their parental strains and ten different field strains, which were included in five distinct lineages. Moreover, two Korean isolates (K046-12 and K047-12; the genetic features of which were reported by us in 2019) [24] were added in the clustering. The S1 sequences of Vaccines A and C were grouped in the GI-19 lineage, which includes nephrotropic strains such as the QX-strain and its regional variant, K047-12. As expected, Vaccine A was genetically close to its parental strain, K40/09, and its passaged strain, K40/09 CE50. In contrast, Vaccine C was genetically more related to the K40/09 strain compared with the KM91 and QX-like strains. Vaccine B and its single-passaged form were grouped in the GI-15 lineage together with their parental strain, D85/06. Otherwise, none of the commercial vaccine strains were allotted to the conventional respiratory GI-1 lineage, which comprises the Connecticut, Beaudette, M41, and H120 strains (Fig. 2). Although most of the recent vaccines target nephrotropic strains that induce a complicated pathology [30], the recent variant K046-12 was genetically remote from the three commercial vaccine strains. This phylogenetic analysis indicated that genetic features of the S1 gene of IBV in commercial vaccines are clustered according to their parental strains, but do not completely cover emerging variants. Amino acid variance in S1 glycoprotein among vaccine and parental strains The S1 glycoprotein consists of 520 amino acid residues and induces a neutralizing antibody response. Because of frequent genetic modification in the HVRs, viral clearance in the host is prohibited. The major HVRs of the S1 glycoprotein have multiple locations, i.e., at amino acid residues 38 ~ 67, 91 ~ 141, and 274 ~ 387 [3, 25]. To evaluate the genetic modification of vaccines from their parental strains, we compared the amino acid arrangement in HVRs and HVR-adjacent regions. We found that Vaccine A, which originated from a nephrotropic virus, showed polymorphism within HVRs compared with the parental strain K40/09 and its passaged strain, K40/09 CE50. The amino acid at position 83 was an asparagine (Asn) in the K40/09 strains, whereas it was an aspartic acid (Asp) in the Vaccine A strain. Similarly, the amino acids at positions 97 and 365 were serine (Ser) and isoleucine (Ile) in the K40/09 strains, whereas they were phenylalanine (Phe) and valine (Val) in the Vaccine A strain. Conversely in the HVRIII-adjacent area (412 amino acid residues), all three vaccine strains carried a glutamic acid (Glu), with a point mutation detected in Vaccine A exclusively. This indicates that the difference from K40/09 was more obvious in Vaccine A than changes occurring after 50 passages (Fig. 3A). Amino acid variance in Vaccine B was mostly observed adjacent to HVRI at position 8, 11, and 13, compared with its parental strain, D85/06. AVR1/08 is an attenuated vaccine strain that was passaged 89 times [7], and its genetic similarity in the upstream of S1 glycoprotein was higher with D85/06 than it was with Vaccine B. The Page 6/19 amino acids encoded were valine (Val), isoleucine (Ile), and phenylalanine (Phe) for D85/06 and AVR1/08, whereas the Vaccine B strains carried leucines (Leu) or cysteine (Cys). Moreover, at the 56th position, D85/06 encoded serine (Ser) but other strains encoded tyrosine (Tyr). In addition, a single passage of Vaccine B yielded a change from glutamic acid (Glu) to glycine (Gly) at HVRII (amino acid residue number 96) (Fig. 3B). Despite the overall similarity in the whole S1 genomic sequence, the vaccine strains under study were composed of multiple polymorphisms exclusively among HVRs and the upstream of S1 glycoprotein. Genetic variance of the S1 glycoprotein among commercial vaccines and new variants Next, we addressed the genetic similarity of the S1 gene among the three commercial vaccine strains. Dissimilarity was detected before HVRI and II at the 83rd, 257th, and 262nd amino acid positions. Although Vaccines A and C were genetically closer in terms of the whole S1 protein sequence (Fig. 2), Vaccines A and B exhibited an identical amino acid sequence near HVRI and II (Fig. 4A). Furthermore, at the 315th amino acid position, Vaccine C showed a second peak that encoded serine (Ser), whereas the major peak of each of three strains encoded cysteine (Cys). This suggests the existence of two viral subpopulations in a vaccine strain (Fig. 4B). Because the three vaccines are designed to deal with regional variants, we examined whether these vaccines are genetically close to emerging Korean variants, the genetic features of which were previously analyzed by us [24]. The amino acid sequences of HVRI (up to the 130th position) from K046-12 and K047-12 were aligned with those of the three commercial vaccine strains. Heterogeneity was most frequently observed around the 30 ~ 50th and 120 ~ 130th amino acid positions (Fig. 5A). To further analyze the genetic proximity regarding HVRI, we drew another phylogenetic tree based on the amino acid sequences of this region. We found that Vaccines A and C were grouped with the nephrotropic variant, K047-12, whereas Vaccine B was less associated. Moreover, none of the vaccine strains showed genetic closeness to the respiratory variant, K046-12 (Fig. 5B). The result of this analysis agrees with the phylogenetic tree of whole S1 gene sequences, which implies that HVRI variation is a dominant factor that attributes genetic characteristics to both field and vaccine strains. Immune responses by commercial vaccines and regional variants We recently reported the action of the nephrotropic variant K047-12 on the innate immune response and identified key immunological factors that can govern the acute response in chicken embryonic kidney (CEK) cells [20]. The avian spleen, similar to that of mammals, is a major secondary lymphoid tissue in which various immune cells, such as lymphocytes and myeloid cells regulate adaptive immunity. In the spleen, dendritic cells and macrophages recognize antigens and present them to lymphocytes, i.e., T and B cells. Activated lymphocytes differentiate into effector cells, thus initiating both cellular and humoral immune responses [14, 36]. We examined the manner in which viral antigens in vaccine strains and regional variants affect the immune cell phenotype. Total splenocytes from 4 ~ 5 week-old SPF chickens were stimulated with Concanavalin A and chicken IL-2 for 4 ~ 5 days in the presence of the three vaccine strains or the regional variants (K046-12 and K047-12). To examine the memory T cell population, which Page 7/19 potentially interacts with viral antigens, we stained the surface of CD4+ and CD8+ T cells with an antiCD44 antibody. None of the treatments changed the frequencies of memory CD4+ and CD8+ T cells (Fig. 6A). To assess the antigen-presenting capacity of macrophages, monocytes were stained with an antibody to MHCII molecule. Although IBVs induced a bit more activated macrophages to a slightly greater extent, significant differences between the variant and vaccine strains were hardly observed. A similar trend was observed for B cell frequency (Fig. 6B and C). These data illustrate the contention that different antigenic stimulations in vitro are not sufficient to change the surface phenotype of major immune cells. Further functional analyses, such as those of intracellular molecules and cytokine regulation within each cell type, are warranted to evaluate these issues. Discussion The first IBV vaccine was developed based on the Mass serotype and was as the only licensed vaccine for many years. Since the isolation of novel IBV serotypes from the GI-19 lineage, vaccine construction has been updated depending on emerging variants. In addition, two or more serotypes are combined in vaccine programs. Currently, numerous IBV vaccines are available and the selection of a proper vaccine strain against regional variants is critical [16]. To better evaluate vaccine stability and safety, it is necessary to identify genetic variation from the parental strains as well as the immune response elicited by vaccination. Thus, we obtained three vaccines that are commercially available in Korea and assessed their genetic sequence encoding the S1 protein together with that of field strains. Subsequently, we investigated the genetic variance of the three vaccine strains compared with the parental strains using S1 gene sequences. The vaccine strains exhibited genetic modification primarily in HVRI and its adjacent region. Near HVRII, genetic variance was also detected. However, compared with the regional variants, antigenic stimulation with the vaccine strains did not significantly change the phenotype of major immune cells. Vaccines A and C were clustered into the GI-19 lineage, whereas Vaccine B and its passaged strain were included in the GI-15 lineage. Because the majority of emerging IBVs are classified as GI-19, we observed genetic distance between Vaccines A and C and other field strains. The parental strain of Vaccine A, K40/09, was developed from a QX-like strain that was reported in 2011 and is thought to be generated by recombination with a KM91-like strain [22]. When chickens are immunized with K40/09, cross protection against both respiratory and renal IBV strains was shown [17]. Based on our phylogenetic analysis, Vaccine A maintained genetic stability compared with its parental strains. Vaccine A was also close to the new regional variant, K047-12, in terms of the S1 gene sequence. HVRI includes the receptor-binding domain, which determines the interaction between the virus and respiratory tissues. Hence, genetic modification of amino acid residues in this region can generate novel variants [29, 32]. The HVRI amino acid sequences differed between K40/09 and Vaccine A to a certain extent, although genetic variation of HVRI was less observed after 50 passages. Therefore, whether Vaccine A is sufficient to provide broad immunity to novel variants warrants further exploration. Moreover, the potential risk of recombination with field viruses should be evaluated. Page 8/19 KM91 is a native Korean variant that was first isolated in 1991, and KM-like variants have been isolated that exhibit recombination with QX-like strains [6, 20]. According to the manufacturer, Vaccine C is a recombinant of the QX and KM91 strains that targets both respiratory and nephropathogenic IBVs; however it is difficult to determine the parental strain in this case. Our analysis showed that Vaccine C also belonged to the GI-19 lineage, which confirmed that this vaccine stems from the KM91 and QX-like strains. Considering the genetic analysis of the S1 glycoprotein and the information provided by the manufacturer, we suggest that Vaccine C shares a genetic background with Vaccine A. However, the efficacy of the vaccines to new regional respiratory variants, such as K046-12, which belongs to the GI-1 lineage, remains to be examined. The D85/06 strain was isolated from Korea in 2006 and clustered into the K-I type together with other respiratory IBV strains. The AVR1/08 strain was developed from the D85/06 strain via 89 passages. A point mutation was observed at the 56th amino acid position of AVR1/08 after 47 passages, which attenuated the virulence of the strain [5, 19]. Vaccine B was introduced from AVR1/08 and showed genetic modification at the upstream of HVRI. We performed a single passage of Vaccine B to evaluate its genetic stability and detected a point mutation in HVRI after this procedure. According to a previous report, AVR1/08 yielded broad protection against both regional respiratory and renal IBVs [7]. However, our phylogenetic analysis revealed that Vaccine B was grouped into the GI-15 lineage, which is genetically separated from the major nephropathogenic strains from the GI-19 lineage. Moreover, Vaccine B was genetically distant from respiratory strains, such as M41 or Beaudette and the regional isolate, K046-12. The investigation of the genetic variance among the three vaccine strains revealed a genetic difference was only observed outside HVRs with the exception of the second peak at the 315th amino acid position in Vaccine C. To estimate the protective efficacy of vaccines against novel IBVs, we aligned the HVRI amino acid sequences of the three commercial vaccines with those of the regional isolates K046-12 and K047-12. A major modification was observed around upstream of the HVRI. Of note, the phylogenetic distance of HVRI was comparable with phylogenetic results obtained for the whole S1 gene sequence. To some extent, our results suggest that protectotype conception is applicable to the three vaccine strains, whereas an alternative vaccine for respiratory strain is necessary to provide broader protection against emerging variants. The induction of adaptive immunity is accelerated when an IBV-specific epitope stimulates the cytotoxic T lymphocyte (CTL) response. An S1 glycoprotein-associated peptide can work as a CTL epitope. When chicken splenocytes were activated with a peptide epitope, IFN-γ production and CD8+ T cell proliferation were facilitated. Furthermore, administration of this epitope induced a DNA vaccine-like effect and protected the host from the IBV challenge [33, 37]. The major purpose of vaccination is to induce specific immune responses against an invading pathogen. However, IBV variants are likely to escape host immune surveillance by modifying their antigenicity because genetic modification of the S1 glycoprotein alters the affinity of the virus for host cell receptors. As shown previously, protection can be achieved by adoptive transfer of memory CD8+ T cells. The key protective action was mediated by memory T cells that had experienced the IBV antigen [27]. IBV variants with dissimilar amino acid residues in the S1 Page 9/19 glycoprotein are less likely to express a coherent epitope. However, we found that the S1 antigenic variance was not sufficient to modify the immune cell phenotype. When splenocytes were stimulated with Concanavalin A, memory CD8+ (CD8+CD44+) T cells occupied more than 50% of the total CD8+ T cell; however, the proportion of memory CD4+ (CD4+CD44+) T cells was < 20% of the total CD4+ T cell. Moreover, the frequencies of each subset of T cells were not changed by antigenic stimulation with IBV variants or vaccine strains. This implies that an immune-boosting strategy needs to be considered during vaccine development. Although phenotypic change was not observed here, during infection, inhibitory receptors (IRs) are expressed on surface of T cells and regulates T cell activation together with T cell receptor (TCR) signal. IR activation can be mediated by macrophage-originated cytokine, such as IL-27. Consequently, the T cell phenotype can be shaped expression of IRs, such as PD-1, CTLA-4, TIGIT, and LAG-3 [4, 9]. Although studies of chicken IRs have yet to be well established, and detection antibodies are not currently available, a recent study reported an analysis of chicken IRs such as chPD-1 and chPD-L1. This is useful for identifying unknown features of T cells during IBV infection [31]. Antigen presentation and cytokine production by macrophages are required for the induction of pathogen-specific T cell activity. For example, IFN-γ production in CTLs is mediated by M1 macrophage generation during intracellular pathogen infection [26]. The role of macrophages in IBV infection is also crucial. Upon infection with the M41 variant, the viability and phagocytic function of macrophages are inhibited. Conversely, overall innate immunity was enhanced such as antimicrobial activity, toll like receptor (TLR) activation, and type I IFN or pro-inflammatory cytokine induction [35]. We detected a minimal number of activated macrophages that expressed MHCII molecules on their surface, probably because of the optimal inducing condition for myeloid cells was not provided. Therefore, a comprehensive analysis of viral antigen-induced macrophage activity needs to be carried out using bone marrow-derived myeloid cells and stimulating cytokines. Moreover, revised IBV vaccines that boost MHCII expression on myeloid cells for enhanced antigen presentation can be applied [18]. This will provide improved antigen-specific T cell proliferation and afford immunity geared toward novel variants. Together with cellular immunity, humoral response by B cells is required. IBV-specific antibody-secreting cells (ASCs) are maximally activated around 10 days after IBV infection. However, during in vitro culture, splenic ASCs are transiently observed only when restimulated with the viral antigen [28]. In our splenocyte culture system, viral antigens enhanced B cell frequency but the level of the response did not rely on antigens. Therefore, kinetic observations and modified culture conditions need to be further considered. The inconsistent antigenic phenotype of IBV variants is a major hurdle for IB prevention, despite the continuous vaccine development of vaccines. The majority of IBV vaccines are of the live-attenuated type, which requires stability during the manufacturing process. In the host, vaccines should be safe and provide long-lasting immunity to multiple variants. Throughout this study, we identified the genetic features and stability of IBV vaccines that are currently available in Korea. In addition, the possible host immune response was examined after the stimulation of host cells with the viral antigen. Our study provides guidance for the evaluation of commercial vaccines and suggestions for the new IBV vaccine development. Page 10/19 Declarations Acknowledgments This study was supported by funding from a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (no. 2022R1C1C1002793). References 1. Belouzard S, Millet JK, Licitra BN, Whittaker GR (2012) Mechanisms of coronavirus cell entry mediated by the viral spike protein. Viruses 4:1011-1033 2. Cavanagh D (1983) Coronavirus IBV Giycopolypeptides: Size of Their Polypeptide Moieties and Nature of Their Oligosaccharides. J gen Virol 64:1187-1191 3. Cavanagh D, Davis PJ, Mockett AA (1988) Amino acids within hypervariable region 1 of avian coronavirus IBV (Massachusetts serotype) spike glycoprotein are associated with neutralization epitopes. Virus research 11:141-150 4. Chen DY, Wolski D, Aneja J, Matsubara L, Robilotti B, Hauck G, de Sousa PSF, Subudhi S, Fernandes CA, Hoogeveen RC, Kim AY, Lewis-Ximenez L, Lauer GM (2020) Hepatitis C virus-specific CD4+ T cell phenotype and function in different infection outcomes. The Journal of clinical investigation 130:768-773 5. Choi K-S, Jeon W-J, Lee E-K, Kye S-J, Park M-J, Kwon J-H (2011) Development of an attenuated vaccine strain from a korean respiratory type infectious bronchitis virus. Korean Journal of Veterinary Research 51:193-201 6. Choi KS, Lee EK, Jeon WJ, Park MJ, Kim JW, Kwon JH (2009) Pathogenicity and antigenicity of a new variant of Korean nephropathogenic infectious bronchitis virus. J Vet Sci 10:357-359 7. Choi KS, et al. (2011) Development of an attenuated vaccine strain from a korean respiratory type infectious bronchitis virus. Korean J Vet Res 51:193-201 8. De Wit J, Swart W, Fabri T (2010) Efficacy of infectious bronchitis virus vaccinations in the field: association between the α-IBV IgM response, protection and vaccine application parameters. Avian Pathology 39:123-131 9. DeLong JH, O’Hara Hall A, Rausch M, Moodley D, Perry J, Park J, Phan AT, Beiting DP, Kedl RM, Hill JA, Hunter CA (2019) IL-27 and TCR Stimulation Promote T Cell Expression of Multiple Inhibitory Receptors. ImmunoHorizons 3:13-25 10. Duffy S, Shackelton LA, Holmes EC (2008) Rates of evolutionary change in viruses: patterns and determinants. Nature Reviews Genetics 9:267-276 11. Franzo G, Legnardi M, Tucciarone CM, Drigo M, Martini M, Cecchinato M (2019) Evolution of infectious bronchitis virus in the field after homologous vaccination introduction. Vet Res 50:92 12. Gallardo RA (2021) Infectious bronchitis virus variants in chickens: evolution, surveillance, control and prevention. Austral journal of veterinary sciences 53:55-62 Page 11/19 13. Guzmán M, Hidalgo H (2020) Live Attenuated Infectious Bronchitis Virus Vaccines in Poultry: Modifying Local Viral Populations Dynamics. Animals 10:2058 14. Hirakawa R, Nurjanah S, Furukawa K, Murai A, Kikusato M, Nochi T, Toyomizu M (2020) Heat Stress Causes Immune Abnormalities via Massive Damage to Effect Proliferation and Differentiation of Lymphocytes in Broiler Chickens. Frontiers in Veterinary Science 7 15. Jackwood MW, Jordan BJ, Roh HJ, Hilt DA, Williams SM (2015) Evaluating Protection Against Infectious Bronchitis Virus by Clinical Signs, Ciliostasis, Challenge Virus Detection, and Histopathology. Avian Dis 59:368-374 16. Jordan B (2017) Vaccination against infectious bronchitis virus: A continuous challenge. Vet Microbiol 206:137-143 17. Kim BY, Lee DH, Jang JH, Lim TH, Choi SW, Youn HN, Park JK, Lee JB, Park SY, Choi IS, Song CS (2013) Cross-protective immune responses elicited by a Korean variant of infectious bronchitis virus. Avian Dis 57:667-670 18. Larsen FT, Guldbrandtsen B, Christensen D, Pitcovski J, Kjærup RB, Dalgaard TS (2020) Pustulan activates chicken bone marrow-derived dendritic cells in vitro and promotes ex vivo CD4+ T cell recall response to infectious bronchitis virus. Vaccines 8:226 19. Lee E-K, Jeon W-J, Lee Y-J, Jeong O-M, Choi J-G, Kwon J-H, Choi K-S (2008) Genetic diversity of avian infectious bronchitis virus isolates in Korea between 2003 and 2006. Avian diseases 52:332-337 20. Lee HC, Jeong S, Cho AY, Kim KJ, Kim JY, Park DH, Kim HJ, Kwon JH, Song CS (2021) Genomic Analysis of Avian Infectious Bronchitis Viruses Recently Isolated in South Korea Reveals Multiple Introductions of GI-19 Lineage (QX Genotype). Viruses 13 21. Lee HJ, Youn HN, Kwon JS, Lee YJ, Kim JH, Lee JB, Park SY, Choi IS, Song CS (2010) Characterization of a novel live attenuated infectious bronchitis virus vaccine candidate derived from a Korean nephropathogenic strain. Vaccine 28:2887-2894 22. Lim TH, Lee HJ, Lee DH, Lee YN, Park JK, Youn HN, Kim MS, Lee JB, Park SY, Choi IS, Song CS (2011) An emerging recombinant cluster of nephropathogenic strains of avian infectious bronchitis virus in Korea. Infect Genet Evol 11:678-685 23. Lim TH, Kim MS, Jang JH, Lee DH, Park JK, Youn HN, Lee JB, Park SY, Choi IS, Song CS (2012) Live attenuated nephropathogenic infectious bronchitis virus vaccine provides broad cross protection against new variant strains. Poultry Science 91:89-94 24. Moon H-W, Sung HW, Kwon HM (2019) Genomic characteristics of natural recombinant infectious bronchitis viruses isolated in Korea. Korean Journal of Veterinary Research 59:123-132 25. Moore KM, Jackwood MW, Hilt DA (1997) Identification of amino acids involved in a serotype and neutralization specific epitope within the s1 subunit of avian infectious bronchitis virus. Arch Virol 142:2249-2256 26. Park J, Hunter CA (2020) The role of macrophages in protective and pathological responses to Toxoplasma gondii. Parasite immunology 42:e12712 Page 12/19 27. Pei J, Briles WE, Collisson EW (2003) Memory T cells protect chicks from acute infectious bronchitis virus infection. Virology 306:376-384 28. Pei J, Collisson EW (2005) Specific antibody secreting cells from chickens can be detected by three days and memory B cells by three weeks post-infection with the avian respiratory coronavirus. Developmental and comparative immunology 29:153-160 29. Promkuntod N, van Eijndhoven RE, de Vrieze G, Gröne A, Verheije MH (2014) Mapping of the receptorbinding domain and amino acids critical for attachment in the spike protein of avian coronavirus infectious bronchitis virus. Virology 448:26-32 30. Reddy VR, Trus I, Desmarets LM, Li Y, Theuns S, Nauwynck HJ (2016) Productive replication of nephropathogenic infectious bronchitis virus in peripheral blood monocytic cells, a strategy for viral dissemination and kidney infection in chickens. Veterinary research 47:1-19 31. Reddy VRAP, Mwangi W, Sadigh Y, Nair V (2019) In vitro Interactions of Chicken Programmed Cell Death 1 (PD-1) and PD-1 Ligand-1 (PD-L1). Frontiers in Cellular and Infection Microbiology 9 32. Rohaim MA, El Naggar RF, Abdelsabour MA, Mohamed MHA, El-Sabagh IM, Munir M (2020) Evolutionary Analysis of Infectious Bronchitis Virus Reveals Marked Genetic Diversity and Recombination Events. Genes 11:605 33. Seo SH, Wang L, Smith R, Collisson EW (1997) The carboxyl-terminal 120-residue polypeptide of infectious bronchitis virus nucleocapsid induces cytotoxic T lymphocytes and protects chickens from acute infection. Journal of virology 71:7889-7894 34. Song CS (2017) An Attenuated Infectious bronchitis virus and An Infectious bronchitis vaccine using the same. 35. Sun X, Wang Z, Shao C, Yu J, Liu H, Chen H, Li L, Wang X, Ren Y, Huang X (2021) Analysis of chicken macrophage functions and gene expressions following infectious bronchitis virus M41 infection. Veterinary research 52:1-15 36. Sutton KM, Morris KM, Borowska D, Sang H, Kaiser P, Balic A, Vervelde L (2021) Characterization of Conventional Dendritic Cells and Macrophages in the Spleen Using the CSF1R-Reporter Transgenic Chickens. Frontiers in Immunology 12 37. Tan L, Liao Y, Fan J, Zhang Y, Mao X, Sun Y, Song C, Qiu X, Meng C, Ding C (2016) Prediction and identification of novel IBV S1 protein derived CTL epitopes in chicken. Vaccine 34:380-386 38. Valastro V, Holmes EC, Britton P, Fusaro A, Jackwood MW, Cattoli G, Monne I (2016) S1 gene-based phylogeny of infectious bronchitis virus: An attempt to harmonize virus classification. Infect Genet Evol 39:349-364 39. van Santen VL, Toro H (2008) Rapid selection in chickens of subpopulations within ArkDPI-derived infectious bronchitis virus vaccines. Avian Pathol 37:293-306 40. Wang Y, Wang Y, Zhang Z, Fan G, Jiang Y, Liu X, Ding J, Wang S (1998) Isolation and identification of glandular stomach type IBV (QX IBV) in chickens. Chinese Journal of Animal Quarantine 15:1-3 41. Worthington KJ, Currie R, Jones RC (2008) A reverse transcriptase-polymerase chain reaction survey of infectious bronchitis virus genotypes in Western Europe from 2002 to 2006. Avian pathology Page 13/19 37:247-257 Figures Figure 1 Cloning of the S1 glycoprotein from commercial vaccines. (A) The S1 gene was inserted into the pCR®2.1 vector using EcoRI restriction sites. (B) The inserted PCR product was verified using the EcoRI restriction enzyme. The four lower bands correspond to the inserted S1 glycoprotein in vaccine strains, and the upper bands are the cloning vectors. Page 14/19 Figure 2 Phylogenetic tree analysis of the S1 gene sequences in vaccines and variants. Vaccine A and its parent strain are indicated by triangles (▲), Vaccine B and its parent strain are indicated by squares (■), and vaccine C and its parent strains are indicated by circles (●). The five different lineages of the strains are indicated in bold font. The tree was created using neighbor-joining method in MEGA, version 10.2.2, with 1000 bootstrap replicates. Page 15/19 Figure 3 Generic modification of commercial vaccines from parental strains. The amino acids encoded by the S1 gene were compared at multiple spots. (A) Amino acid variance of Vaccine A from K90/09 and K40/09 CE50. (B) AVR1/08 and D85/06 were compared with Vaccine B before and after a single passage. Each color stands for the specific polarities of the amino acids, as follows: Purple, non-polar; green, neutral polar; red, acidic polar; and blue, basic polar. Page 16/19 Figure 4 Amino acid variance among vaccine strains. (A) Differences in amino acids among Vaccines A, B and C. (B) Chromatogram of the heterogenicity of the S1 gene. The secondary nucleotide peak observed at the 315th amino acid residue is indicated by an arrowhead. Each color stands for the specific polarities of the amino acids, as follows: Purple, non-polar; green, neutral polar; red, acidic polar; and blue, basic polar. Page 17/19 Figure 5 Genetic similarity of HVRs among the three vaccine strains and the regional variants. (A) The amino acid sequences encoded by the upstream of S1 gene were aligned. (B) Phylogenetic analysis based on the amino acid sequence of HVRI. Page 18/19 Figure 6 Immune cell phenotypic changes induced by viral antigen stimulation. The frequencies of memory CD4+ and CD8+ T cells (A), MHCII+ monocytes (B), and B cells (C) are shown. Total splenocytes from SPF chickens were stimulated in the presence or absence of viral antigens from vaccine or variant strains. Representative and pooled data obtained (mean ± SEM) from 4 independent experiments. *Significant differences between indicated groups (P<0.05). Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. SipplementaryMaterial.docx Page 19/19
https://openalex.org/W2728391758	https://nvlvet.com.ua/index.php/journal/article/download/1171/1169	Ukrainian	null	Поширеність та причини виникнення хвороб копитець у корів	Naukovij vìsnik Lʹvìvsʹkogo nacìonalʹnogo unìversitetu veterinarnoï medicini ta bìotehnologìj ìmenì S.Z. G̀žicʹkogo	2,017	cc-by	3,834	Citation: Khomyn, N.M., Mysak, A.R., Iglitskej, I.I., Pritsak, V.V. (2017). Prevalance and causes of diseases occurrence of cows hooves. Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 19(77), 22–26. http://nvlvet.com.ua/ УДК 619:617.58:636.2 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Науковий вісник Львівського національного університету ветеринарної медицини та біотехнологій імені С.З. Ґжицького Scientific Messenger of Lviv National University of Veterinary Medicine and Biotechnologies named after S.Z. Gzhytskyj doi:10.15421/nvlvet7706 ISSN 2518–7554 print ISSN 2518–1327 online http://nvlvet.com.ua/ Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Науковий вісник Львівського національного університету ветеринарної медицини та біотехнологій імені С.З. Ґжицького Scientific Messenger of Lviv National University of Veterinary Medicine and Biotechnologies named after S.Z. Gzhytskyj doi:10.15421/nvlvet7706 ISSN 2518–7554 print ISSN 2518–1327 online http://nvlvet.com.ua/ Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Науковий вісник Львівського національного університету ветеринарної медицини та біотехнологій імені С.З. Ґжицького Scientific Messenger of Lviv National University of Veterinary Medicine and Biotechnologies named after S.Z. Gzhytskyj doi:10.15421/nvlvet7706 ISSN 2518–7554 print ISSN 2518–1327 online Н.М. Хомин, А.Р. Мисак, І.І. Ігліцький, В.В. Пріцак nadiakhomyn@ukr.net Н.М. Хомин, А.Р. Мисак, І.І. Ігліцький, В.В. Пріцак nadiakhomyn@ukr.net Львівський національний університет ветеринарної медицини та біотехнологій імені С.З. Ґжицького, вул. Пекарська, 50, м. Львів, 79010, Україна У статті наведено аналіз літературних даних щодо частоти виникнення та етіологічних факторів хвороб копитець у корів, зокрема різних клінічних форм пододерматитів. Встановлено, що на погіршення якості копитцевого рогу, що приз- водить до виникнення хвороб копитець, впливають: незадовільні умови утримання, незбалансована (дефіцитна за пожив- ними та мінеральними речовинами) годівля, деформація копитець, відсутність моціону, інсоляції та коректуючої розчист- ки. Вагоме місце в патології копитець високопродуктивних корів належить деформаціям рогу копитець. Ввстановлено, що визначальний вплив на утворення деформованих копитець мають похибки у годівлі (надмірне згодовування концентрованих кормів – за висококонцентратної годівлі), в утриманні (відсутність або недостатність активного моціону, використання бетонної, цегляної та щілинної підлоги тощо) та розведенні тварин з деформованими копитцями, неправильна форма яких успадковується. До ортопедичних захворювань відносять пододерматити, ламініти, виразку м’якуша (виразку Рустерго- льца), тілому, а також гнійно-некротичні ураження, що. здебільшого протікають у вигляді флегмони вінчика та м’якуша, причиною яких є проникнення патогенної мікрофлори через пошкоджену шкіру, рідше внаслідок метастазування інфекту. Однак, незважаючи на значну кількість праць, присвячених проблемі ортопедичної патології, невизначеною залишилась низка питань стосовно етіопатогенезу окремих клінічних форм пододерматитів, у зв’язку з чим в Україні проводять нау- кові дослідження у цьому напрямку. Тому перспективним є поглиблене вивчення окремих ланок патогенезу деяких клінічних форм пододерматитів задля розробки тактики та методу лікування ортопедично хворих тварин. Ключові слова ׃худоба, копитця, пододерматит, ламініт, виразка Рустергольца, тілома, флегмона вінчика, артрит копитцевого суглобу Ключові слова ׃худоба, копитця, пододерматит, ламініт, виразка Рустергольца, тілома, флегмона вінчика, артрит копитцевого суглобу. Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 Н.М. Хомин, А.Р. Мысак, И.И. Иглицкий, В.В. Прицак nadiakhomyn@ukr.net Львовский национальный университет ветеринарной медицины и биотехнологий имени С.З. Гжицкого, ул. Пекарская, 50, г. Львов, 79010, Украина В статье приведен анализ литературных данных о частоте возникновения и этиологических факторах болезней конеч- ностей у коров, в том числе различных клинических форм пододерматитов. Установлено, что на ухудшение качества копитцевого рога, что приводит к возникновению болезней конечностей, влияют неудовлетворительные условия содержа- ния, несбалансированная кормление, дефицитное по питательным и минеральными веществами, деформация конечностей, отсутствие моциона, инсоляции и корректирующей расчистки. Важное место в патологии конечностей высокопродукти- вных коров принадлежит деформациям рога конечностей. Установлено, что определяющее влияние на образование дефо- рмированных конечностей имеют погрешности в кормлении (чрезмерное скармливание концентрированных кормов – при висококонцентратном кормления), в содержании (отсутствие или недостаточность активного моциона, использование бетонных, кирпичных и щелевых полов) и разведении животных с деформированными копытцами, неправильная форма которых наследуется. К ортопедическим заболеваниям относят пододерматит, Ламинит, язву мякоти (язву Рустерголь- ца), тело, а также гнойно-некротические поражения, которое в основном протекает в виде флегмоны венчика и мякиша, Citation: Khomyn, N.M., Mysak, A.R., Iglitskej, I.I., Pritsak, V.V. (2017). Prevalance and causes of diseases occurrence of cows hooves. LNUVMBT named after S.Z. Gzhytskyj, 19(77), 22–26. sak, A.R., Iglitskej, I.I., Pritsak, V.V. (2017). Prevalance and causes of diseases occurrence of cows hooves. Scientific Messenger fter S.Z. Gzhytskyj, 19(77), 22–26. Citation: h Citation: Khomyn, N.M., Mysak, A.R., Iglitskej, I.I., Pritsak, V.V. (2017). Prevalance and causes of diseases occurrence of cows hooves. Scientific Messenger NUVMBT named after S.Z. Gzhytskyj, 19(77), 22–26. Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 22 22 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 причиной которых является проникновение патогенной микрофлоры через поврежденную кожу, реже вследствие метас- тазирования инфекта. Однако, несмотря на значительное количество работ, посвященных проблеме ортопедической патологии, неопределенным остался ряд вопросов по этиопатогенеза отдельных клинических форм пододерматит, в связи с чем в Украине проводят научные исследования в этом направлении. Поэтому перспективным является углубленное изуче- ние отдельных звеньев патогенеза некоторых клинических форм пододерматитов с целью разработки тактики и метода лечения ортопедически больных животных. причиной которых является проникновение патогенной микрофлоры через поврежденную кожу, реже вследствие метас- тазирования инфекта. Однако, несмотря на значительное количество работ, посвященных проблеме ортопедической патологии, неопределенным остался ряд вопросов по этиопатогенеза отдельных клинических форм пододерматит, в связи с чем в Украине проводят научные исследования в этом направлении. Поэтому перспективным является углубленное изуче- ние отдельных звеньев патогенеза некоторых клинических форм пододерматитов с целью разработки тактики и метода лечения ортопедически больных животных. Ключевые слова: коровы, копытца. пододерматит, ламинит, язва Рустергольца, тилома, флегмона венчика, артрит копитцевого сустава, одержание, кормление, моцион. Prevalance and causes of diseases occurrence of cows hooves N.M. Khomyn, A.R. Mysak, I.I. Iglitskej, V.V. Pritsak nadiakhomyn@ukr.net Lviv National University of Veterinary Medicine and Biotechnologies named after S.Z. Gzhytskyi, Pekarska Str., 50, Lviv, 79010, Ukraine The article presents an analysis the literature data on the frequency of and etiologic factors of disease of cows hooves, including different clinical forms of pododermatitis. It is established that the deterioration of the quality hoof horn, which causes diseases of hooves influencing on poor conditions of keeping, unbalanced feeding, by deficiency of nutrients and minerals, deformation hooves, lack of motions, insolation and corrective clearing. An important place in the pathology of highly productive cows hooves strains belonging to deformations of hooves horn. It was established that determining influence on the formation of deformed hooves have errors in feeding (excessive feeding of concentrated feed - for high concentrate feeding), in keeping (absence or lack of active consti- tutional, the use of concrete, brick and slot floors, etc.) and breeding of animals with deformed hooves, irregular shape of which are inherited. To orthopedic diseases include diseases: pod dermatitis, laminitis, ulcers of pulp (Rusterholts ulcer), body, and also puru- lent necrotic lesion, which mostly occur in the form of rim phlegmon and pulp, caused by an infiltration of pathogenic microflora through damaged skin, rarer due to metastasis of infect. However, despite the significant number of works, devoted to the problem of orthopedic pathology, remained undefined a number of issues regarding pathogenesis of some clinical forms of pododermatitis, in connection with that in Ukraine research in this direction were conducted. So perspective is depth study of some links of pathogenesis, some clinical forms of pododermatitis to develop tactics and treatment method of orthopedic sick animals. Key words: cattle hooves pododermatitis laminitis ulcers Rusterholts tiloma rim phlegmon arthritis of hoof joint Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Кулинича, за умови прив’язного утримання тварин хвороби копитець, зокрема гнійні пододерматити, виявляють у 3,2% від наявного поголів’я, за прив’язного на чавунній щі- линній підлозі – у 3,6%, а на суцільнй бетонній – у 2,6% (Kulynych, 2012). Іншою важливою причиною виникнення та розви- тку патології копитець у високопродуктивних корів є гіпокінезія. Вона сприяє порушенню обміну речовин у тваринному організмі, знижує засвоєння корму, особливо мінеральних речовин і вітамінів, погіршує стан сухожилково-зв’язкового апарату, кровопоста- чання пальців, зокрема сприяє порушенню циркуляції крові в ділянці «периферичного серця», що призво- дить до розвитку патологічних процесів (Pan'ko, 2000). Навіть за умов своєчасного внесення підстилки та- ке утримання призводить до збільшення вологи в копитцевому розі й виключає його стирання, що спричиняє надмірне відростання рогу підошви з по- дальшим розвитком деформації; процес ускладнюєть- ся відсутністю регулярної коректуючої розчистки (Demchuk et al., 1996). Виникненню і перебігу хвороб копитець сприяє низький рівень годівлі, пов’язаний з недоодержанням організмом тварин поживних та мінеральних речовин, а також незбалансовані раціони годівлі (Kadrik, 1996). Негативний вплив на якість копитцевого рогу та ви- никнення захворювань копитець має дисбаланс у мінеральному обміні, зменшення вмісту в крові каль- цію, фосфору, зниження активності лужної фосфатази тощо (Borysevych and Homyn, 2002). Здебільшого причинами хвороб копитець, зокрема пододерматитів, є безприв’язне утримання корів і молодняку, а також безприв’язно-боксове утримання тварин на щілинних підлогах у поєднанні з гіпокінезі- єю. Підлоги, виготовлені з залізобетону, порушують рівномірний розподіл маси тіла на всю площу підош- ви копитець і зазвичай призводять до розвитку прихо- ваних, а також клінічно виражених хвороб у цій час- тині пальців. Розвитку цих захворювань зокрема сприяє низька якість решіток підлоги, їхня деформа- ція та нерівне укладання, що призводить до надмірно- го стирання рогу підошви та розвитку надалі пододе- рматитів (Kogut and Borysevych, 2003). В умовах системного остеопорозу у корів виникає розтяг пальцевих згиначів з надривами в місцях приє- днання до кістки, що призводить до незворотного порушення постави кінцівок та супутніх хвороб копи- тець. Розвиток такої патології відбувається поступово й спочатку не позначається на загальному стані тва- ринного організму (Chumachenko, 1997). Утримання тварин на гладкій і слизькій бетонній підлозі, здебільшого з наявністю ухилу понад 6° сприяє ковзанню, падінню тварин та виникненню травм. Для зменшення негативного впливу перфоро- ваних залізобетонних підлог на копитця корів засто- совують полімерне покриття. За показниками щільно- сті, теплопровідності, теплоємкості такі підлоги за- ймають проміжне місце між бетоном та деревом, що значно зменшує вибракування тварин через хвороби копитець (Demchuk et al., 1996). Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 В умовах системного остеопорозу, пов’язаного здебільшого з дефіцитом в організмі тварин мінера- льних речовин, зокрема кальцію, з гіподинамією та відсутністю інсоляції, виникає розтяг згиначів пальця з надривами у місцях приєднання до кістки, що приз- водить до незворотного порушення постави кінцівок та деформації копитець, а також до виникнення супу- тніх хвороб дистального відділу пальців у худоби (Borysevych and Homyn, 2002). відпочинку (з 86,0 до 28,8 хв) та збільшення кількості таких періодів протягом доби (від 8.8 до 15,8). Вказані зміни поведінки корів спричиняють збільшення кількості бурситів у 15,8 раза, гострих ламінітів – у 6, виразок підошви – у 3, дерматитів у ділянці пальців – у 2 і більше разів порівняно з утриманням тварин на такій же підлозі з великою кількістю соломи (4– 5 кг/гол./добу як підстилки) (Kozij, 2007). відпочинку (з 86,0 до 28,8 хв) та збільшення кількості таких періодів протягом доби (від 8.8 до 15,8). Вказані зміни поведінки корів спричиняють збільшення кількості бурситів у 15,8 раза, гострих ламінітів – у 6, виразок підошви – у 3, дерматитів у ділянці пальців – у 2 і більше разів порівняно з утриманням тварин на такій же підлозі з великою кількістю соломи (4– 5 кг/гол./добу як підстилки) (Kozij, 2007). Утримання високопродуктивних тварин на бетонній підлозі з недостатньою кількістю підстилки призводить до нерівномірного розвитку рогу латерального та медіального копитця на тазових кінцівках, що зумовлює підвищення захворюваності в ділянці пальця. У корів з виразками підошви збільшується висота м’якуша латерального копитця на 37,2%, ширина – на 6,9%, а у тварин, хворих на міжпальцевий та папіломатозний пальцевий дерматити, вірогідно вираженою є різниця висоти м’якуша й ширини копитець (Kozij, 2007). За остеомаляції тварини з широкою поставою гру- дних кінцівок підводять тазові кінцівки під черево, що супроводжується порушенням фізіологічного розподілу опорно-силових навантажень у тканинах копитець і виникненням деформації та пододермати- тів (Molokanov, 1997). Встановлено, що значна кількість патологічних процесів у ділянці копитець пов’язана з умовами утримання тварин (Molokanov, 1997). Утримання корів на глибокій незмінній підстилці і, особливо, невчасне її внесення призводить до роз- витку масових гнійно-некротичних процесів у ділянці копитець. Постійна мацерація тканин гноївкою зни- жує опір шкіри і створює сприятливі умови для про- никнення мікроорганізмів, особливо бактерій некрозу, які, як відомо, населяють передшлунки жуйних і пос- тійно виділяються з фекаліями. Не випадково серед усіх захворювань ділянки пальця некробактеріозні ураження складають 23–35% (Pan'ko et al., 1998). Згідно з дослідженнями С.М. Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 24 Вступ масового характеру, є однією з причин недоотриман- ня у корів значної кількості молока (в середньому 4 кг за добу, або 5% за лактацію), а також сприяє виник- ненню хвороб копитець (Luk'janovskij, 1992; Molokanov, 1997). Як відомо, серед хірургічної патології в корів важ- ливе місце належить хворобам копитець, які викли- кають у тварин болючість, кульгавість і, як наслідок, зниження молочної продуктивності, неплідність, що здебільшого призводять до передчасного вибракову- вання (до 27%) цінних у продуктивному й племінно- му відношенні тварин (Molokanov, 1997; Pan'ko, 2000; Borysevych and Homyn, 2002). Борисевич В.Б. (2002), вивчаючи етіологію, пато- генез, клінічні особливості та патологоанатомічні зміни деформованих копитець встановив, що дефор- мація копитець є складним патологічним процесом, в основі якого лежать порушення фізіологічного розпо- ділу опорно-силових навантажень у структурах копи- тець (Borysevych and Homyn, 2002). Встановлено, що найчастіше хвороби копитець у корів (55–75%) виникають в перші 1–3 тижні після отелення (Smilie, 1996). Однак серед ортопедичних захворювань найбільш поширеною формою ураження копитець, як стверджують вчені, є пододерматити та ламініти, що становить 51–64%, дерматити міжпаль- цевого склепіння – 20–32%, флегмони вінчика – 6– 10% та артрити копитцевого суглоба – 7–8% (Smilie, 1996). Аналізуючи етіологічні та патогенетичні механіз- ми деформованих копитець, вчений встановив, що визначальний вплив на утворення деформованих ко- питець мають похибки у годівлі (надмірне згодову- вання концентрованих кормів – за висококонцентрат- ної годівлі), в утриманні (відсутність або недостат- ність активного моціону, використання бетонної, цегляної та щілинної підлоги тощо) та розведенні тварин з деформованими копитцями, неправильна форма яких успадковується (Borysevych and Homyn, 2002). Вагоме місце в патології копитець високопродук- тивних корів, згідно з твердженнями іншого автора, належить деформаціям рогу копитець (54,2%) (Kozij, 2007). Згідно з дослідженнями українських вчених дефо- рмація копитець за гіпокінезії зумовлена порушенням здебільшого відтоку крові, лімфи і циркуляції тка- нинної рідини (Kozij et al., 2000). Як показали дослідження, деформація копитець – одне з найбільш поширених уражень. Значна кількість авторів відрізняє деформацію копитець від їх хвороб і вважає етіологічним фактором ураження порушення співвідношення між швидкістю відростання та сти- рання копитцевого рогу. Деформація розвивається поступово й спочатку не викликає змін з боку загаль- ного стану тваринного організму, однак, набуваючи Виникненню деформації копитець сприяє також зміна постави кінцівок, спричинена остеодистрофією (Homyn and Borysevych, 2001). Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 23 Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 23 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Однією з причин виникнення патології копитець, згідно з дослідженнями українських і зарубіжних вчених, є інтенсивна білкова годівля, яка сприяє по- рушенню вітамінно-мінерального живлення та поси- ленню відростання копитцевого рогу (Smilie, 1996; Kulynych, 2012). За умов згодовування великої кількості силосу, цукрового буряку, концентрованих кормів, злакових трав у тваринному організмі накопичується велика кількість органічних кислот, особливо, молочної. Всмоктуючись у кров, вони призводять до Встановлено, що прив’яне утримання корів на бетонній підлозі з обмеженою кількістю підстилки призводить до зменшення часу їхнього відпочинку (з 12,6 до 7, 6 год.), середньої тривалості одного періоду Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 24 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 декомпенсованого ацидозу та порушення обмінних процесів. В основі шкіри копитець на рівні мікроциркуляторного русла порушується проникність капілярів, спостерігається екстравазація рідкої частини крові, що супроводжується місцями відшаруванням рогу копитець (Smilie, 1996; Sorokina and Matveev, 2000). падків, а з масою тіла більше ніж 600 кг – у 39,7. У 44,4% тварин зі специфічною виразкою м’якуша про- дуктивність за лактацію перевищує 4500 кг молока (Timofeev and Gimranov, 2005). У 20,5% клінічно здорових корів виявлені екзосто- зи на поверхні підошви як латеральної, так і медіаль- ної копитцевих кісток, а у тварин з виразкою Рустер- гольца вона відмічається у 75,4%, що дає підстави розглядати екзостози як одну з причин виникнення цього захворювання (Mishhenko and Mishhenko, 2007). У країнах з високорозвиненим молочним скотарством, згідно з дослідженнями зарубіжних вчених, основною причиною пододерматитів вважають ацидотичний стан організму, який діагностують за величиною рН рубцевого вмісту (Moore et al., 1991). Відомі етіологічні фактори пододерматитів у корів, хворих на ендометрит і мастит (Molokanov, 1997). У країнах з високорозвиненим молочним скотарством, згідно з дослідженнями зарубіжних вчених, основною причиною пододерматитів вважають ацидотичний стан організму, який діагностують за величиною рН рубцевого вмісту (Moore et al., 1991). Відомі етіологічні фактори пододерматитів у корів, хворих на ендометрит і мастит (Molokanov, 1997). Дослідження щодо вивчення тіломи проведені у багатьох країнах. Встановлено, що ураження складає 12–30% від усієї ортопедичної патології у лактуючих корів. У 87,5–99,4% випадків гіперплазія шкіри між- пальцевого склепіння локалізується на тазових і лише у 0,5–12% – на грудних кінцівках (Mishhenko and Mishhenko, 2007). Доведено, що гнійний пододерматит у корів час- тіше виникає за умови деформації копитець з ознака- ми розпаду рогової підошви. У господарствах з прив’язним способом утримання корів рогова підош- ва в хворих тварин розпадається на сироподібну масу, в в умовах безприв’язного утримання – ріг сухий і тирсоподібний (Kulynych, 2012). Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 Основною причиною виникнення тіломи більшість дослідників вважає посилене навантаження на шкіру міжпальцевого склепіння, що призводить до надмір- ного її розтягнення та травмування. Найбільша кіль- кість випадків кульгавості, зумовленої ускладненням тіломи, спостерігається на початку пасовищного пері- оду. У корів до 4-річного віку живою масою 400– 500 кг хворобу реєструють в поодиноких випадках, тимчасом як у корів віком понад 9 років тілому від- мічають у 56–63%, а з живою масою більше ніж 650 кг – у 32–41% корів (Molokanov, 1997). Крім того, встановлено, що внаслідок несвоєчасної коректуючої розчистки, деформації, порушення умов утримання через тривале перебування тварин у при- міщенні (зимово-стійловий період), рівень годівлі в ділянці підошви копитець накопичуються мікроско- пічні гриби, які проявляють кератолітичні властивос- ті, руйнуючи кератин копитець (Kulynych, 2012). Значна кількість публікацій присвячена гнійно- некротичним ураженням копитець, які часто виника- ють як ускладнення проникаючих ран у ділянці паль- ців (Mishhenko and Mishhenko, 2007). Гнійно- некротичні процеси здебільшого протікають у вигляді флегмони вінчика та м’якуша, причиною яких є про- никнення патогенної мікрофлори через пошкоджену шкіру, рідше внаслідок метастазування інфекту, а також похибок в утриманні тварин (надмірна скупче- ність тварин на обмеженій площі, невдала конструк- ція підлог, порушення санітарно-гігієнічних норм, незбалансована годівля по усіх показниках, необхід- них для життєдіяльності тваринного організму, що знижує його резистентність (Molokanov, 1997; Chumachenko, 1997). Причиною виникнення гнійних пододерматитів, як показали дослідження, найчастіше є травми, які виникають внаслідок надмірного навантаження на одиницю площі поверхні підошви копитець, з розвит- ком надалі асептичного запалення, відшаруванням рогу м’якуша і накопиченням ексудату, найчастіше, у ділянці облямівки з подальшим проникненням у цей сегмент основи шкіри мікрофлори із зовнішнього середовища (коки, диплококи, кишкова паличка, бак- терії некрозу тощо (Kogut and Borysevych, 2003). Іншим етіологічним чинником виникнення гнійно- го пододерматиту є прокол підошви з подальшим інфікуванням травмованої ділянки патогенними мік- роорганізмами, який виникає здебільшого у пасовищ- ний період утримання тварин (Borysevych et al., 2004). Однак, незважаючи на значну кількість праць, присвячених проблемі ортопедичної патології, неви- значеною залишилась низка питань стосовно етіопа- тогенезу окремих клінічних форм пододерматитів, у зв’язку з чим в Україні продовжуються наукові дос- лідження в цьому напрямку. Виразковий пододерматит частіше виявляють у весняний та осінній періоди, рідше – влітку та взимку. У 91,6% випадків виразка локалізується на латераль- ному і лише у 8,4% – на підошві медіального копитця. Основною причиною виникнення виразки м’якуша вважається посилене навантаження на центральну частину підошви копитця у зв’язку з невчасною коректуючою розчисткою копитець. Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 За таких умов відбувається стиснення основи шкіри підошви (м’якуша) між каудо-аксіальним краєм копитцевої кістки та підошвою рогового башмака (Timofeev and Gimranov, 2005; Mishhenko and Mishhenko, 2007). Висновки 1. В умовах стійлового утримання корів хвороби копитець вважають одними з найпоширеніших захво- рювань. 1. В умовах стійлового утримання корів хвороби копитець вважають одними з найпоширеніших захво- рювань. р 2. Основними етіологічними факторами хвороб копитець є похибки в утриманні тварин, зниження рівня годівлі та незбалансованість кормів раціонів, відсутність регулярного активного моціону, інсоляції та належної і регулярної ортопедичної розчистки копитець. Виразку Рустергольца у корів віком до 4 років з масою 300–400 кг та середньорічним надоєм 3500 кг молока реєструють в поодиноких випадках. Із збіль- шенням віку, живої маси і продуктивності збільшу- ється імовірність виникнення захворювання: у корів віком понад 9 років виразку реєструють у 37,5% ви- Scientific Messenger LNUVMBT named after S.Z. Gzhytskyj, 2017, vol. 19, no 77 25 Науковий вісник ЛНУВМБТ імені С.З. Ґжицького, 2017, т 19, № 77 3. Надалі плануємо проводити дослідження щодо вивчення окремих ланок патогенезу деяких клініч- них форм пододерматитів з метою розробки тактики та методу лікування ортопедично хворих тварин. 3. Надалі плануємо проводити дослідження щодо вивчення окремих ланок патогенезу деяких клініч- них форм пододерматитів з метою розробки тактики та методу лікування ортопедично хворих тварин. Cerkov', 69–70 (in Russian). Cerkov', 69–70 (in Russian). Mishhenko, V.A., Mishhenko, A.B. (2007). Bolezni konechnostej u vysokoproduktivnyh korov. Veterinarnyj konsul'tant. 6, 11–13 (in Russian). Pan'ko, I.S., Tyhonjuk, L.A., Nagornyj, V.V. (1998). Nekrobakteriozni urazhennja kincivok u koriv. Visnyk Bilocerkiv. derzh. agrar. un-tu. 5(2), 197–200 (in Ukrainian). Стаття надійшла до редакції 15.03.2017 Бібліографічні посилання Borysevych, B., Homyn, N. (2002). Etiopatogenez urazhen' kopytec' u koriv v umovah stijlovogo utrymannja. Veterynarna medycyna Ukrai'ny. 12, 32– 34 (in Ukrainian). Pan'ko, I.S. (2000). Novi pidhody do vyvchennja prychyn ta profilaktyky hvorob ratyc' u vysokoproduktyvnyh koriv. Visnyk Bilocerkiv. derzh. agrar. un-tu. Bila Cerkva. 13(1), 19–22 (in Ukrainian). Borysevych, V.B., Kogut, N.V., Homyn, N.M. (2004). Poverhnevyj ta glybokyj gnijnyj pododermatyt u koriv. Visnyk Bilocerkiv. derzh. agrar. un-tu. Bila Cerkva. 28, 8–14 (in Ukrainian). Sorokina, I.A., Matveev, L.V. (2000). Gnojno- nekroticheskie processy v oblasti pal'cev u krupnogo rogatogo skota. Veterinarija. 2, 43–45 (in Russian). Demchuk, M.V. Chornyj, M.V., Vysokos, M.P. (1996). Gigijena tvaryn. K.: Urozhaj (in Ukrainian). Timofeev, C.B., Gimranov, V.V. (2005). Rasprostranenie jazvennyh processov v oblasti pal'cev u krupnogo rogatogo skota. Veterinarija. 7, 43–45 (in Russian). Kogut, N., Borysevych, V. (2003). Vynyknennja pododermatytiv u velykoi' rogatoi' hudoby vnaslidok travmuvannja kincivok. Veterynarna medycyna Ukrai'ny. 2, 29–30 (in Ukrainian). Homyn, N., Borysevych, B. (2001). Rol' osteodystrofii' u vynyknenni pododermatytu u velykoi' rogatoi' hudoby. Vet. medycyna Ukrai'ny. 6, 30–31 (in Ukrainian). Kozij, V.I., Ivchenko, V.M., Stadnyk, P.O. (2000). Dejaki pytannja etiologii' zahvorjuvan' u diljanci pal'cja u vysokoproduktyvnyh koriv. Visnyk Bilocerkiv. derzh. agrar. un-tu. Bila Cerkva. 13(I), 49–53 (in Ukrainian). Chumachenko, V.Ju. (1997). Rezystentnist' tvaryn i faktory, shho vplyvajut' na i'i' stan. Veterynarna medycyna Ukrai'ny. 3, 23–25 (in Ukrainian). Kadrik, K. (1996). On the mechanism of elementary acid – base balance disorders in cows. Estonian Veterinary Review Supplemental – Acta Veterinaria Baltica. Estonia,Tartu, 16–22. Kozij, V.I. (2007). Laminit u vysokoproduktyvnyh koriv: Avtoref. dys… dokt. vet. nauk: 16.00.05. – Bila Cerkva, 36 (in Ukrainian). Moore, J.N., Allen, D.Jr., Clark, E.S. (1991). Pathophysiology of acute laminitis. Equine Pract. 13(5), 29–30. Kulynych, S.M. (2012). Urazhennja kopytec' u koriv, sprychyneni keratomicetamy: Avtoref. dys… dokt. vet. nauk: 16.00.05. K., 36 (in Ukrainian). ( ) Luk'janovskij, V.A. (1992). Vlijanie sanitarno- gigienicheskih uslovij na patologiju kopytec u korov. Veterinarija. 1, 17–20 (in Russian). Smilie, R. (1996). Prevalence of lesions associated with subclinical laminits in first-lactation cows from herds with high milk productions. JAVMA. 208(9), 1445– 1451. Molokanov, V.A. (1997). Jetiopatogenez zabolevanij kopytec u vysokoproduktivnyh korov. Problemy hirurgicheskoj patologii sel'skohozjajstvennyh zhivotnyh: Tez. dokl. Vsesojuz. nauch. konf. – Belaja Стаття надійшла до редакції 15.03.2017 Стаття надійшла до редакції 15.03.2017
https://openalex.org/W4388731018	https://biosignaling.biomedcentral.com/counter/pdf/10.1186/s12964-023-01335-6	English	null	Epigenetic-focused CRISPR/Cas9 screen identifies (absent, small, or homeotic)2-like protein (ASH2L) as a regulator of glioblastoma cell survival	Cell communication and signaling	2,023	cc-by	10,482	© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom‑ mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Epigenetic‑focused CRISPR/Cas9 screen identifies (absent, small, or homeotic)2‑like protein (ASH2L) as a regulator of glioblastoma cell survival Ezgi Ozyerli‑Goknar1,2,3, Ezgi Yagmur Kala1, Ali Cenk Aksu1, Ipek Bulut1, Ahmet Cingöz1, Sheikh Nizamuddin2,3, Martin Biniossek4, Fidan Seker‑Polat1, Tunc Morova5,6, Can Aztekin5, Sonia H. Y. Kung6, Hamzah Syed1,5,7, Nurcan Tuncbag1,5,8, Mehmet Gönen1,5,9, Martin Philpott10, Adam P. Cribbs10, Ceyda Acilan1,5, Nathan A. Lack1,5,6, Tamer T. Onder1,5, H. T. Marc Timmers2,3 and Tugba Bagci‑Onder1,5* Abstract Background Glioblastoma is the most common and aggressive primary brain tumor with extremely poor prognosis, highlighting an urgent need for developing novel treatment options. Identifying epigenetic vulnerabilities of cancer cells can provide excellent therapeutic intervention points for various types of cancers. Method In this study, we investigated epigenetic regulators of glioblastoma cell survival through CRISPR/Cas9 based genetic ablation screens using a customized sgRNA library EpiDoKOL, which targets critical functional domains of chromatin modifiers. Results Screens conducted in multiple cell lines revealed ASH2L, a histone lysine methyltransferase complex subunit, as a major regulator of glioblastoma cell viability. ASH2L depletion led to cell cycle arrest and apoptosis. RNA sequenc‑ ing and greenCUT&RUN together identified a set of cell cycle regulatory genes, such as TRA2B, BARD1, KIF20B, ARID4A and SMARCC1 that were downregulated upon ASH2L depletion. Mass spectrometry analysis revealed the interaction partners of ASH2L in glioblastoma cell lines as SET1/MLL family members including SETD1A, SETD1B, MLL1 and MLL2. We further showed that glioblastoma cells had a differential dependency on expression of SET1/MLL family members for survival. The growth of ASH2L-depleted glioblastoma cells was markedly slower than controls in orthotopic in vivo models. TCGA analysis showed high ASH2L expression in glioblastoma compared to low grade gliomas and immuno‑ histochemical analysis revealed significant ASH2L expression in glioblastoma tissues, attesting to its clinical relevance. Therefore, high throughput, robust and affordable screens with focused libraries, such as EpiDoKOL, holds great prom‑ ise to enable rapid discovery of novel epigenetic regulators of cancer cell survival, such as ASH2L. Conclusion Together, we suggest that targeting ASH2L could serve as a new therapeutic opportunity for glioblastoma. Keywords Epigenetic, Chromatin modifiers, CRISPR/Cas9 screen, sgRNA library, ASH2L, Glioblastoma Correspondence: Tugba Bagci‑Onder tuonder@ku.edu.tr Full list of author information is available at the end of the article Correspondence: Tugba Bagci‑Onder tuonder@ku.edu.tr Full list of author information is available at the end of the article Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 https://doi.org/10.1186/s12964-023-01335-6 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 https://doi.org/10.1186/s12964-023-01335-6 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 https://doi.org/10.1186/s12964-023-01335-6 Cell Communication and Signaling Open Access Open Access Reagents and cell lines All reagents and cell lines are described in Supplementary Information. RRID numbers for used cell lines, antibod- ies, organisms, plasmids, or tools are given in the related method sections. Background targeted by sgRNAs leading to higher proportion of null mutations. Glioblastoma represents 48.3% of malignant primary brain tumors and occurs with 3.1/100 000 incidence per year [1]. Presence of necrotic foci, high prolifera- tion rate, invasiveness, and highly angiogenic feature are hallmarks of glioblastoma, contributing to its very high lethality. Median survival with standard care for patients is 14.6 months; and only 5.5% of patients can survive past 5 years after diagnosis [2]. Therefore, deeper understanding of genetic and epigenetic vulner- abilities of glioblastoma is crucial to design more effec- tive therapies. Here, we introduce a customized Epigenetic Domain- specific Knock Out Library (EpiDoKOL), which targets functional domains of key epigenetic modifiers. Using EpiDoKOL, we performed drop-out screens on multiple cell lines and identified ASH2L as an indispensable gene for glioblastoma cell survival. ASH2L is a trithorax group family member that functions within the SET1/MLL family methyltransferase complexes specific for histone H3 lysine 4 (H3K4) methylation including MLL1, MLL2, MLL3, MLL4, SET1A and SET1B; and acts as a cofac- tor, supporting active gene transcription [16–19]. Here, we show that ASH2L directly regulates cell cycle-related genes and facilitates tumor cell survival both in vitro and in vivo. Together, our results suggest suitability of domain-targeted chromatin-focused CRISPR library screens for the identification of novel and druggable epi- genetic vulnerabilities in glioblastoma. The de-regulation of gene expression associated with tumor suppression or progression through epigenetic changes have been reported [3, 4]. DNA methylation, his- tone modifications and chromatin remodeling are major epigenetic alterations, which broadly affect cell pheno- type. Histone proteins are prone to a variety of dynamic posttranslational modifications (e.g. phosphorylation, acetylation, methylation, ubiquitination), which modu- late their affinity for the DNA or form new binding sites for protein modules, supporting the euchromatin (active) or heterochromatin (repressed) state [5]. These modifica- tions are written, read or removed by unique proteins to regulate gene transcription, DNA repair, replication and chromosome condensation [6]. Abnormal DNA meth- ylation and distinct histone modification patterns due to aberrant activity of epigenetic modifiers are frequently encountered in tumor cells and affect their drug response and growth [4, 7]. Therefore, identifying epigenetic vul- nerabilities of cancer cells can provide excellent thera- peutic intervention points for various types of cancers. © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom‑ mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 2 of 16 Generation of EpiDoKOL Epigenetic Domain-specific Knock Out Library (EPI- DOKOL) targeting 251 critical chromatin modifier enzymes was developed by pooling sgRNAs designed against functional domains of target genes using CCtop tool [20]. The library consists of 1628 gene-targeting sgRNAs in total, in addition to 80 non-targeting sgR- NAs retrieved from GECKO Library as negative con- trol. 5 sgRNAs were picked for each domain of the targeted chromatin modifier genes. Designed sgRNAs were cloned to pLentiGuide (RRID:Addgene_52963) and plentiCRISPRv2 (RRID:Addgene_52961) backbones [21] by GIBSON assembly. To determine sgRNA dis- tribution in plasmid pools, next generation sequenc- ing was performed on MISeq at MIT BioMicro Center. Library composition, sgRNA design, cloning and sequencing procedures are detailed in Supplementary Information. Sequences of sgRNAs of EpiDoKOL are available in Supplementary Table 1. CRISPR/Cas9 system is a widely used genome editing tool based on the delivery of gene–complementary syn- thetic guide RNA (sgRNA) with 3´-protospacer-asso- ciated motif (PAM) into the cell [8]. Cas9 endonuclease binds sgRNA and initiates site-specific cleavage of DNA, which ultimately results in insertions and/or deletions within targeted genes. CRISPR/Cas9-based functional genetic screens were shown to be more potent than shRNA screens [9, 10]. Genome-wide CRISPR knockout screens enabled the identification of essentiality genes [10] and vulnerabilities to drugs such as TRAIL [11], ATR inhibitors [12] and Ras inhibitors [13] in various cancers. However, labor-intensive and expensive nature of genome-wide screens with low signal/noise ratio increased attention for focused libraries [14]. Addition- ally, different gene targeting approaches have been devel- oped, since targeting the 5′ exons as a default practice may produce in-frame functional variants of proteins. Therefore, as an alternative approach, “domain spe- cific targeting” was developed [15], where the genomic regions encoding the functional protein domains were Histone extraction Histone were acid-extracted, and protein concentra- tion was determined with Pierce BCA Protein Assay Kit (Thermo Scientific, 23225). Histone extraction and western blotting details are described in Supplementary Information. Cas9 activity assay pBabe-hygro-GFP expressing T98G and U373 cells were seeded to cell culture plates and transduced with control sgRNAs g-NT1 and g-NT2 and GFP-targeting sgRNAs g-T1 and g-T2 separately. sgRNA sequences are listed in Supplementary Table 2. Percentage of GFP(+) cells were assessed by Flow cytometry. Details are described in Supplementary Information. Control cells (g-NT infected) and cells carrying sgRNAs against candidate genes were seeded as 750 cells/well in triplicates in 6-well plates 7 days post-transduction. Colonies were grown for 14 days with fresh medium and then stained with Crystal Violet. Number of colo- nies were counted using ImageJ Software (NIH Image, Bethesda, MD, USA). Staining procedure is detailed Supplementary Information. Viral packaging and transduction Lentiviral particles from plentiCRISPRv2 and retroviral particles from pBabe-hygro-GFP (RRID:Addgene_61215) were produced in 293T cells as described [22, 23]. For stable cell line generation, cells were seeded at 1.5 × 106 cells per plate in 10 cm plates and were transduced Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 3 of 16 when 50–70% confluency was reached. For transfec- tion MLL1 (Ambion, 107890) and WDR5 (Ambion, 136959) siRNAs were utilized. Transfection procedure is detailed in Supplementary Information. with virus containing media supplemented with prota- mine sulfate (10 µg/ml). Transduced cells were selected by antibiotics. Details are described in Supplementary Information. Cell viability and apoptosis assays Cell viabilities and Caspase 3/7 activities were meas- ured via Cell Titer-Glo (CTG) Luminescent Cell Viability Assay (Promega, USA) or Caspase-Glo® 3/7 (Promega, USA) respectively, according to manufacturer’s instruc- tions using a plate reader (BioTek’s Synergy H1, VT, USA). Annexin V staining was performed with Muse® AnnexinV&Dead Cell Kit (Luminex, MCH100105) according to the manufacturer’s instructions and ana- lyzed by Muse Cell Analyzer (Merck, Darmstadt, Ger- many). Western Blots for apoptotic markers; cleaved PARP (Abcam, ab74290, USA) and caspase3 (Cell Signal- ing, 5A1E) were also performed. Details of cell viability, caspase activation assays, Annexin V staining, and West- ern Blotting are described in Supplementary Information. GFP competition assay Screen was performed as two biological replicates. U373 and T98G cells were seeded and transduced with Epi- DoKOL virus with MOI = 0.4 to ensure that each cell takes up a single sgRNA. After transduction, a propor- tion of cells were pelleted to serve as a reference point for baseline sgRNA distribution while remaining cells were selected with puromycin. At day 30, cells were counted and pelleted. Collected pellets were stored at -80 °C and genomic was DNA isolated using MN Nucleospin Tis- sue kit (Macherey-Nagel, Germany) according to manu- facturers’ protocol. Isolated DNA was used for nested PCR. Sequences of external and internal PCR primers are listed in Supplementary Table 3. Samples were sent to Illumina Hiseq2500 RAPID sequencing to Vincent J. Coates Genomic Sequencing Laboratory of University of Berkeley. NGS results were analyzed using Python programming language and Model-based Analysis of Genome-wide CRISPR-Cas9 Knockout (MAGeCK) (version 0.5.8) [24]. P < 0.05 cutoff was applied to gene- level analysis to identify significantly depleted genes. EpiDoKOL screen sequencing data are deposited to the NCBI GEO database with the accession number GSE201657. Detailed information on screening proce- dure, sequencing and statistical analysis are provided in Supplementary Information. Seven days post-transduction, cells were seeded in 6-well plates together with their GFP-stable counterparts at equal proportion (150,000 cells each) and at final density of 300,000 cells per well for the GFP competition assay. At initial seeding day (day 0) and on days 2, 5, 7 and 9, the cells were detached with trypsin and the GFP-positive and negative cell ratio was determined by flow cytometry. Cells were analyzed by BD Accuri C6 (BD Biosciences, USA) (excitation 488 nm, emission 530/575 nm) record- ing 10,000 events per sample. sgRNA cloning for validation of hits To knock-out candidate essentiality genes, top depleted sgRNAs sequences for ASH2L, RBX1, SSRP1 were derived from EpiDoKOL and individually cloned to pLentiCRISPRv2 plasmid. All sgRNA sequences used for validation experiments are listed in Supplementary Table 2. Cloning procedure is detailed in Supplementary Information. Immunohistochemistry Brain Glioblastoma tissue microarray (US Biomax -GL806f) was stained for ASH2L. Aperio AT2 Scan- ner (Leica Biosystems) was used for imaging slides and Aperio ImageScope (Leica Biosystems) program with Positive Pixel Count Algorithm was used for digital scor- ing as further described in Supplementary Information. To identify epigenetic modifier genes indispensable for cell survival, we generated a customized epigenome-wide domain-targeted pooled sgRNA library. In Epigenetic Domain-specific Knockout Library (EpiDoKOL), sgRNAs were designed against gene regions encoding functional catalytic domains of chromatin modifiers. EpiDoKOL consists of 1750 sgRNAs in total, targeting 250 different genes, and 80 non-targeting controls (Fig. 1A). Molecu- lar cloning of sgRNAs for pooled libraries into two dif- ferent backbones (pLentiGuide and pLentiCrisprv2) was achieved, leading to 800X coverage and successful lenti- viral packaging (Fig. 1B, Supplementary Fig. 1A, B). TCGA Gene expression profiles of “glioblastoma” (GBM) and “brain lower grade glioma” (LGG) tumors were pre- processed by the pipeline of The Cancer Genome Atlas (TCGA) consortium. Classical, Mesenchymal, Neural, and Proneural subgroups were identified. The log2-trans- formed gene expression values were compared against each other using the Wilcoxon rank sum test. Detailed explanation is available in Supplementary Information. Quantitative RT‑PCR List of primers can be found in Supplementary Table 4. Experimental details are available in Supplementary Information. siRNA transfection Glioblastoma cells (5 × 104) were seeded on 6-well plates. After 16 h, cells were transfected with siRNA Cells were collected 14 days post-transduction, fixed with cold 70% Ethanol, stained with Muse Cell Cycle Reagent Ozyerli‑Goknar et al. Cell Communication and Sign (The Muse® Cell Cycle Kit, Luminex, M analyzed with Muse Cell Analyzer (M Germany). Staining and cell cycle ana detailed in Supplementary Information. Quantitative RT‑PCR List of primers can be found in Supple Experimental details are available in Information. TCGA Gene expression profiles of “glioblasto “brain lower grade glioma” (LGG) tu processed by the pipeline of The Canc (TCGA) consortium. Classical, Mese and Proneural subgroups were identifie formed gene expression values were c each other using the Wilcoxon rank s explanation is available in Supplementa Immunohistochemistry Brain Glioblastoma tissue microarr -GL806f) was stained for ASH2L. A ner (Leica Biosystems) was used for im Aperio ImageScope (Leica Biosystem Positive Pixel Count Algorithm was use ing as further described in Supplementa RNA sequencing U373 cells were transduced with ASH2 Cell pellets were collected as triplicate transduction for RNA isolation. Lib and sequencing was performed at Uni (Oxford, UK). RNAseq and statistical an are detailed in Supplementary Informat Quantitative mass spectrometry of the A Sample preparation, GFP-affinity puri analysis were performed as reported [25 and cytoplasmic proteins were isolated f cells harboring ASH2L-GFP and used purification with GFP-Trap agarose Chromotek). Purified proteins were and tryptic peptides were analyzed by n MS with a high performance nanoflow based ms/ms system (Thermo Fisher Sc Page 4 of 16 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 (The Muse® Cell Cycle Kit, Luminex, MCH100106) and analyzed with Muse Cell Analyzer (Merck, Darmstadt, Germany). Staining and cell cycle analysis procedure is detailed in Supplementary Information. Genome localization experiments by greenCUT&RUN Genome localization analysis of GFP-tagged ASH2L was performed by greenCUT&RUN with the combination of enhancer-MNase and LaG16-MNase, as described [26]. Sequencing libraries were prepared as described [25] and sequenced in Illumina, HiSeq 3000 platform. Detailed experimental procedure and bioinformatic analyses of NGS are available in Supplementary Information. In vivo tumor growth All in vivo experiments were approved by the institu- tional ethical committee of Koç University. 6–8-week-old non-obese diabetic/severe combined immunodeficiency (NOD/SCID) mice were used for orthotopic tumor mod- els as described in Supplementary Information. Tumors were monitored using IVIS Lumina III (Perkin Elmer, USA). Quantification of tumor progression was per- formed with GraphPad PRISM software (San Diego, CA, USA). RNA sequencing U373 cells were transduced with ASH2L or NT sgRNAs. Cell pellets were collected as triplicates at 14 days post transduction for RNA isolation. Library preparation and sequencing was performed at University of Oxford (Oxford, UK). RNAseq and statistical analysis procedures are detailed in Supplementary Information. As Cas9-based genetic ablation may be suboptimal in certain tumor cell lines due to chromosomal aber- rations and mutational burden, we first examined the efficiency of Cas9-mediated cleavage efficiency in glio- blastoma cell lines. We first transduced U373 and T98G with GFP-encoding lentiviruses and then introduced g-NT1, g-NT2 (non-targeting sgRNAs), g-T1 and g-T2 (sgRNAs targeting GFP) along with Cas9 in a single len- tiviral vector. Flow cytometry indicated that GFP+ cells were reduced to 20% of the total population at 12 days post-transduction in g-T1 or g-T2 transduced cells. No change in GFP signal was observed with g-NT1 and g-NT2 vectors, indicating that Cas9-activity is optimal within 2 weeks (Fig. 1C, D). We then continued with Epi- DoKOL screen in both U373 and T98G cells. To identify common regulators of glioblastoma cell survival, cells were transduced with EpiDoKOL at a low multiplicity Quantitative mass spectrometry of the ASH2L interactome Sample preparation, GFP-affinity purification and data analysis were performed as reported [25]. Briefly, nuclear and cytoplasmic proteins were isolated from glioblastoma cells harboring ASH2L-GFP and used for GFP-affinity purification with GFP-Trap agarose beads (#gta-200, Chromotek). Purified proteins were on-bead digested and tryptic peptides were analyzed by nanoflow-LC-MS/ MS with a high performance nanoflow-HPLC Orbitrap based ms/ms system (Thermo Fisher Scientific). The raw data files were analyzed with MaxQuant software (ver- sion 1.5.3.30). The obtained protein files were analyzed by Perseus software (MQ package, version 1.6.12). Detailed protocol is available in Supplementary Information. Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 5 of 16 and assessed their effects with several functional experiments (Fig. 2A). Accordingly, colony form- ing abilities of U373 and T98G cells with individual knockouts of RBX1, ASH2L, SSRP1 genes markedly decreased compared to control cells that received g-NT (Fig. 2B). To assess the effect of hit genes in a heterogenous population, we performed GFP com- petition assays, where GFP-negative cells were trans- duced with sgRNA against a hit gene and mixed in a 1:1 ratio with gNT-transduced GFP-positive cells. (i g p g ) Fig. 1 Epigenetic Domain-specific Knock Out Library (EpiDoKOL) screen identified common essentiality genes in T98G and U373 cells. A Composition of EpiDoKOL including target molecules, their functions and related epigenetic modifications. B Library generation procedure. Figure created with BioRender.com. C Cas9 activity assay of U373 and T98G cells. Microscopic images of T98G and U373 cells transduced with indicated sgRNAs 18 days post-transduction. D Flow cytometric analysis of U373 and T98G cells transduced with indicated sgRNAs (g-NT1, g-NT2: non-targeting sgRNAs; g-T1 and g-T2: sgRNAs targeting GFP), up to 12 days post-transduction. E Schematic of EpiDoKOL screening procedure. Figure created with BioRender.com. F sgRNA density plots from cells transduced with pLentiCRISPRv2 plasmid containing EpiDoKOL. Cell pellets collected before puromycin selection. G Log2 counts of sgRNAs at initial and final time points in T98G and U373 cells. H Waterfall plots for Log2fold changes of genes after screening T98G cells with EpiDoKOL for a month. Domain and gene-based analysis revealed overlapping essentiality hits. I Waterfall plots for Log2fold changes of genes after screening U373 cells with EpiDoKOL for a month. Domain and gene-based analysis revealed overlapping essentiality hits. Mageck p values were calculated with negative binomial model fitting. Other p values were determined by two-tailed Student’s t-test P < 0.05, P < 0.01, *P < 0.001 RNA sequencing Monitoring both the GFP+ and GFP- fraction over 16 days, we observed gradual decreases in the cells transduced with selected sgRNA in U373 and T98G cells (Fig. 2C), and U87MG cells, as a third glioblas- toma cell line (Supplementary Fig. 2A). The most sig- nificant phenotype was observed for U373 cells and we further delineated the effects of RBX1, ASH2L, SSRP1 ablation on glioblastoma cell fitness using this cell line. Depletion of ASH2L, RBX1 or SSRP1 genes slowed down the growth of cells as gauged by cell viability assays (Fig. 2D); and induced apoptosis as revealed by elevated caspase 3/7 activity (Fig. 2E) and AnnexinV positivity (Fig. 2F). Thus, apoptotic cell death is likely to contribute to the reduced cell numbers observed upon ASH2L, RBX1 and SSRP1 ablation. These proof- of-principle experiments illustrated that EpiDoKOL is a practical functional genomics tool that enables iden- tification of epigenetic modifiers important for cancer cell fitness. of infection (MOI = 0.4) with 500 × coverage to ensure single sgRNA intake and proper sgRNA representation (Fig. 1E). Following antibiotic selection, transduced cells were cultured for 30 days by maintaining the sgRNA cov- erage at each population doubling. Using standard library preparation protocols, deep sequencing and analysis [27], we compared the sgRNA composition in initial and final timepoints. Histogram of median normalized read counts for all sgRNA plotted for U373 and T98G cells confirmed normal distribution of log2-transformed nor- malized counts, indicating that no bias was introduced during cloning or transduction steps (Fig. 1F). We then compared the depletion scores of gene-targeting versus non-targeting sgRNAs to evaluate the overall screen effi- ciency. Essential gene targeting sgRNAs were depleted significantly, whereas no change in non-targeting control sgRNA abundance was observed (Fig. 1G). Altogether, these analyses illustrated that EpiDoKOL preserves nor- mal distribution of sgRNAs in glioblastoma cells and reveals potential essential genes for cell survival. EpiDoKOL screen in T98G and U373 cells identified common regulators of glioblastoma cell survival g g As EpiDoKOL was composed of multiple sgRNAs target- ing a functional domain of a single gene, it was possible that one gene was targeted by more than 10 sgRNAs in total, depending on the domain composition. There- fore, we undertook two different analysis approaches; at domain-level or gene-level and determined com- mon depleted genes using median normalization scores (Fig. 1H, I). Accordingly, several epigenetic modifiers were found to be commonly depleted in glioblastoma cell lines. Along with epigenetic modifiers that were pre- viously implicated in cancer cell fitness, such as CHD1, CHD4, DNMT3B, ELP3, SUPT16H, SUV39H2; novel hits ASH2L, RBX1 and SSRP1 were discovered as essential genes for both U373 and T98G cells suggesting common regulatory role (Supplementary Fig. 1C). Fig. 2 Effects of candidate genes on glioblastoma cell fitness were validated with functional assays in vitro. A Scheme of validation experiments for novel EpiDoKOL essentiality hits. TD: post-transduction day. B Representative images of long-term clonogenic assay of cells infected with sgRNAs against selected hits and statistical analysis. Quantification of colonies was performed by ImageJ software. C Results of GFP competition flow cytometric assay for selected hits. Day 0 refers to the day of the cell seeding at TD 7. D-F Cell viability (D), Caspase 3/7 activity (E) and AnnexinV analysis (F) conducted on U373 cells upon depletion of ASH2L, RBX1 and SSRP1 genes. G Western blot analysis of ASH2L protein levels upon transduction of U373 cells with g-ASH2L or g-NT. H Western Blot analysis for cleaved Caspase3 and PARP in U373 cells at day 6 and 14 post-transduction with g-NT control or g-ASH2L. I Western Blot analysis for H3K4 mono and trimethylation levels in U373 cells 6 days post-transduction with g-NT control or g-ASH2L. P values determined by two-tailed Student’s t-test P < 0.05, P < 0.01, P < 0.001 (See figure on next page.) ASH2L is critical for cell cycle regulation We explored genes bound by ASH2L on chromatin and differentially expressed upon ASH2L depletion as major targets of ASH2L. By overlapping RNAseq and greenCUT&RUN results, we generated a list of genes that may be directly linked with the essential func- tion of ASH2L. When ASH2L peaks on promoter- TSSs were overlapped with DEGs from RNA-seq data, a total 150 of up-regulated and 350 down-regulated genes were identified, which together constitute a “critical gene list” (Fig. 3E). GSEA conducted on this list led to the identification of G2/M checkpoint, E2F targets and mitotic spindle hallmark gene sets as enriched (Fig. 3F-G). Heatmap for differential expres- sion of cell cycle and mitosis related genes as well as greenCUT&RUN IGV tracks for selected genes are shown as examples (Fig. 3H-I, Supplementary Fig. 3D). Based on our finding of cell cycle regulatory genes as ASH2L targets, we conducted cell cycle analysis on ASH2L-depleted U373 and U87MG cells. Propidium iodide staining revealed decreased number of cells in the G0/G1 and accumulation at S and G2/M phases of cell cycle upon ASH2L knockout, resembling mitotic arrest (Fig. 3J, Supplementary Fig. 2D). Collectively, these results suggest that ASH2L plays a critical role in maintenance of glioblastoma cell fitness through regu- lating cell cycle progression. Regulation of the transcriptome by ASH2L prompted us to analyze its localization on chromatin using CUT&RUN in glioblastoma cells. Using the default cri- teria of the HOMER (FDR < 0.0001 for calling the peaks and fold changes ≥ 4 and p value < 0.001 against con- trol for filtering the peaks), 13,360 and 1964 chromatin peaks were called in U373 and U87MG cells, respectively (Fig. 3D). To find the commonalities between these two cell lines, initially all the peaks were merged, and cover- age was compared. In total, 13,782 merged peaks were identified, of which, 117 (0.85% of total merged peaks, 5.96% of the U87MG peaks) and 4493 (32.6% of total merged peaks and 33.6% of the U373 peaks) were found unique for the respective cell lines. ASH2L is essential for glioblastoma cell survival and regulates histone methylation and transcription ASH2L functions as a cofactor within MLL family of methyltransferase complexes to trigger his- tone H3 lysine 4 (H3K4) methylation [16–19], and its loss was previously associated with reduction of H3K4 trimethylation. Consistently, we observed reduced mono and trimethylation of H3K4 upon ASH2L deple- tion in U373 and U87MG cells (Fig. 2I, Supplementary Fig. 2C). Since H3K4me3 is a well-known euchromatin mark acting globally, we investigated gene expression changes upon ASH2L knock-out to gain a mechanistic understanding for its essential role. We performed RNA- seq comparing U373 cells transduced with control and ASH2L sgRNAs (Fig. 3A). ASH2L knockout resulted in 461 up-regulated (FDR < 0.05, p-value < 0.05 and log2- fold-change ≥ 1) and 1076 down-regulated (FDR < 0.05 and log2-fold-change ≤ -1) genes. Number of downregu- lated genes was higher than upregulated ones, consist- ent with the association between H3K4 trimethylation and gene activity (Fig. 3B). GSEA revealed several nega- tively and positively enriched pathways with significant normalized enrichment scores (Fig. 3C, Supplementary Fig. 3A). Downregulated genes upon ASH2L depletion significantly overlapped with cell cycle and mitosis-asso- ciated gene sets, whereas upregulated genes enriched for metabolism, glycolysis, and hypoxia pathways. Expres- sion of representative downregulated hits were validated by qRT-PCR (Supplementary Fig. 3B). ASH2L is essential for glioblastoma cell survival and regulates histone methylation and transcription Among the genes validated for their essential function in glioblastoma, ASH2L was the least studied in can- cer and its ablation gave a stronger phenotype; there- fore we further investigated its role in glioblastoma. We knocked out ASH2L in U373 and U87MG cells and showed successful depletion of ASH2L in protein level (Fig. 2G, Supplementary Fig. 2B). Extensive cleav- age of caspase-3 and PARP was observed indicating To validate the function of these novel hits in glio- blastoma, we individually introduced selected sgRNAs Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 6 of 16 Page 6 of 16 Ozyerli Goknar et al. Cell Communication and Signaling (2023) 21:328 Fig. 1 (See legend on previous page.) Fig. 1 (See legend on previous pag e ) Fig. 1 (See legend on previous page.) Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 7 of 16 indicating direct targets of ASH2L in glioblastoma cells (Supplementary Fig. 3C). potent induction of apoptosis upon ASH2L depletion (Fig. 2H). ASH2L functions as a cofactor within MLL family of methyltransferase complexes to trigger his- tone H3 lysine 4 (H3K4) methylation [16–19], and its loss was previously associated with reduction of H3K4 trimethylation. Consistently, we observed reduced mono and trimethylation of H3K4 upon ASH2L deple- tion in U373 and U87MG cells (Fig. 2I, Supplementary Fig. 2C). Since H3K4me3 is a well-known euchromatin mark acting globally, we investigated gene expression changes upon ASH2L knock-out to gain a mechanistic understanding for its essential role. We performed RNA- seq comparing U373 cells transduced with control and ASH2L sgRNAs (Fig. 3A). ASH2L knockout resulted in 461 up-regulated (FDR < 0.05, p-value < 0.05 and log2- fold-change ≥ 1) and 1076 down-regulated (FDR < 0.05 and log2-fold-change ≤ -1) genes. Number of downregu- lated genes was higher than upregulated ones, consist- ent with the association between H3K4 trimethylation and gene activity (Fig. 3B). GSEA revealed several nega- tively and positively enriched pathways with significant normalized enrichment scores (Fig. 3C, Supplementary Fig. 3A). Downregulated genes upon ASH2L depletion significantly overlapped with cell cycle and mitosis-asso- ciated gene sets, whereas upregulated genes enriched for metabolism, glycolysis, and hypoxia pathways. Expres- sion of representative downregulated hits were validated by qRT-PCR (Supplementary Fig. 3B). potent induction of apoptosis upon ASH2L depletion (Fig. 2H). (See figure on next page.) Fig. 3 ASH2L transcriptionally regulates cell cycle via direct promoter interactions with G2/M-checkpoint and E2F target gene sets. A Schematic of RNAseq and greenCUT&RUN experiments. B Volcano plot of RNAseq data showing differentially expressed genes (DEGs) in ASH2L depleted U373 cells compared to controls on the 14 day post-transduction. C Normalized enrichment score results of gene set enrichment analysis (GSEA) for all gene sets available from MSigDB v7.5. Top 5 biological processes enriched in downregulated or upregulated genes upon ASH2L knockout are shown. P values were calculated by hypergeometric test. D greenCUT&RUN analysis in U373 cells revealing genomic localization of ASH2L at differentially expressed gene promoters. E Venn diagram and volcano plot of overlapped RNAseq and greenCUT&RUN data to show proportion of differentially expressed genes bound by ASH2L in U373 cells. Genes bound by ASH2L and differentially expressed upon ASH2L depletion are now called as critical genes, denoted by blue (for downregulated) and yellow (for upregulated) dots on volcano plot. F Normalized enrichment score and FDR-qval results of GSEA on critical genes for all gene sets available from MSigDB v7.5. Some of the negatively enriched pathways related to cell cycle are highlighted. G Top 10 biological processes enriched in regulatory genes upon ASH2L knockout. H RNAseq heatmaps of critical genes. I greenCUT&RUN representative igv plots of critical genes. J Flow cytometric cell cycle analysis of ASH2L-depleted U373 by PI staining on post-transduction day 14 and its statistical analysis. P values determined by two-tailed Student’s t-test P < 0.05, P < 0.01, P < 0.001 ASH2L is critical for cell cycle regulation Conversely, 94.1% of U87MG peaks and 66.4% of U373 peaks were common Glioblastoma cells have differential dependency on SET1/ MLL family of transcription factors interacting with ASH2L To address the essential role of ASH2L for glioblastoma cell survival further, we focused our attention to chroma- tin modifying complexes, specifically, SET1/MLL family of histone methyltransferases SET1 (SETD1A), SET1B (SETD1B), MLL1 (KMT2A), MLL2 (KMT2B), MLL3 (KMT2C) and MLL4 (KMT2D) [28]. Each of these meth- yltransferases can form distinct multi-subunit complexes with diverse functions. All SET1/MLL families contain four core subunits, WDR5, RBBP5, ASH2L and DPY30, which are collectively named as the WRAD module. Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 8 of 16 Page yerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 ig. 2 (See legend on previous page.) Fig. 2 (See legend on previous page.) Fig. 2 (See legend on previous page.) Fig. 2 (See legend on previous page.) Fig. 2 (See legend on previous page.) Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 9 of 16 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 derived primary glioblastoma cells, GBM4. GBM4 cells, upon transduction with g-ASH2L, exhibited effi- cient reduction of ASH2L expression (Fig. 5A), and displayed significantly lower proliferation and sphere formation ability (Fig. 5B, C). Encouraging results obtained in primary GBM cells led us to investigate the role of ASH2L in glioblastoma growth in vivo. We utilized orthotopic xenograft models using Luciferase (Fluc)-expressing U87MG cells, stably transduced with control sgRNA or ASH2L sgRNA. Following trans- duction, cells were intracranially injected and tumor growth was monitored until 33 days post-implanta- tion (Fig. 5D). Accordingly, U87MG cells with ASH2L knockout could not form tumors as compared to the control cells (Fig. 5E). Efficient block of tumor form- ing capacity by lack of ASH2L encouraged us to check available patient data. Indeed, ASH2L expression was higher in glioblastoma in comparison to low grade gliomas (LGG-Grade 3 and Grade 3); ASH2L was expressed in all glioblastoma subtypes, but enriched for proneural subtype, based on TCGA (Fig. 5F-G). We next examined ASH2L protein in patient samples by immunohistochemistry utilizing a glioblastoma tis- sue microarray with 80 cores representing 40 different glioblastoma cases. 70% of glioblastoma specimens had strong ASH2L expression, whereas nonmalignant tissue specimens had none/low expression (Fig. 5H, I, Supplementary Fig. 5). ASH2L is critical for cell cycle regulation Collectively, these clinically rel- evant results illustrate that ASH2L plays a critical role in glioblastoma progression. Other subunits of SET1/MLL family complexes are spe- cific to one or a few SET1/MLL families and involved in chromatin recruitment [29]. Being part of WRAD mod- ule, ASH2L acts as a cofactor to support trimethylation [16–19]. To assess which SET1/MLL family complexes ASH2L is mainly part of in glioblastoma cells, we per- formed quantitative mass spectrometry (MS) of the ASH2L interactome for U373 and U87MG cell lines. Cells were transduced with inducible N-terminal GFP fusions of ASH2L gene encoding N-GFP-ASH2L fusion protein. Nuclear and cytoplasmic proteins were extracted, GFP- affinity purified and subjected to tandem MS (Fig. 4A). Accordingly, we identified common members of MLL- family and SETD1A/B complexes (DPY30, WDR5, ASH2L, RBBP5, HCFC1); as well as with MLL-specific (KMT2A, KMT2B, KMT2C, KMT2D, NCOA6, KDM6A, MEN1, PSIP1, PAGR1, PAXIP) and SETD1A/B-specific (SETD1A, SETD1B, WDR82, BOD1, CXXC1) complex members in nuclear fraction. Based on LFQ values, we determined the stoichiometry of interacting proteins (Fig. 4B-D). While in cytoplasmic fraction, interaction was evident only with SETD1A/B complex (Supplemen- tary Fig. 4A, B); in nuclear fraction ASH2L interacted with MLL family and SETD1A/B members with simi- lar stoichiometry (Fig. 4B-D). To assess the functional importance of SET1/MLL family complex members for glioblastoma cell viability, we performed siRNA mediated knock-down of WDR5 and KMT2A (MLL1) in U373 and U87MG cells and checked their colony forming ability (Fig. 4E). Downregulation of MLL1 significantly reduced colony forming potential of both U373 and U87MG cells, whereas WDR5 silencing affected only U87MG cells (Fig. 4F). Taken together, these results implicate ASH2L- MLL1 complex as a specific dependency in GBM cell proliferation. Discussion Epigenome-directed treatment strategies interfer- ing with abnormal DNA methylation, acetylation or chromatin remodeling patterns of glioblastoma are heavily investigated. For example, clinical outcome of conventional treatment with Temozolomide (TMZ) is tightly dependent on DNA tumors’ methylation status. Because patients with methylated MGMT gene have ASH2L is essential for tumor progression in vivo ASH2L is essential for tumor progression in vivo To assess the role of ASH2L in more clinically-relevant settings, we tested the effect of ASH2L KO in patient To assess the role of ASH2L in more clinically-relevant settings, we tested the effect of ASH2L KO in patient Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 10 of 16 Fig. 3 (See legend on previous page.) Fig. 3 (See legend on previous page.) Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 11 of 16 Page 11 of 16 epigenetic silencing of multiple tumor suppressor genes [43]. An essential role for DNA methyltrans- ferase DNMT3B in cancer cell survival has previously been illustrated [44]. SUV39H2 is a potential onco- gene in lung adenocarcinoma [45] and osteosarcoma [46]. ELP3 can initiate tumorigenesis and regeneration in the intestine [47]. Finally, WDR5 upregulation pro- motes proliferation, self-renewal and chemoresistance in bladder cancer via H3K4 trimethylation [48]. Iden- tification of these already-established cancer fitness genes by EpiDoKOL attests to the power and utility of our screening strategy. survival advantage [30], synthetic inhibitors of MGMT were tested in glioblastoma patients (NCT00613093) [31]. HDAC inhibitors also hold promise for bet- ter clinical outcome [32, 33]. Valproic acid with TMZ and radiotherapy (NCT00302159) gave promising results [34]. Inhibitors of chromatin remodeling com- plex, namely Oliparib and Veliparib are FDA-approved and clinically investigated for glioblastoma as single agents or in combination with chemo/radio therapy (NCT01390571, NCT03212274, and NCT02152982, NCT03581292, NCT01514201 respectively) with promising results [35, 36]. Based on the ever-growing importance of the epi- genetic mechanisms in tumor progression and ther- apy response, we investigated epigenetic regulators of glioblastoma cell survival through a functional screen approach using our custom-generated targeted library, EpiDoKOL designed against functional domains of epi- genetic modifier proteins. sgRNAs targeting essential domains generate the strongest lethality phenotypes and thus provide a strategy to rapidly define the pro- tein domains required for cancer dependence [37]. Targeting functional domains results in higher pro- portion of null mutations as opposed to conventional CRISPR approaches targeting 5’ exon of genes which might create truncated proteins that retain function- ality. We identified ASH2L, CHD1, CHD4, DNMT3B, ELP3, RBX1, SSRP1, SUPT16H, SUV39H2 as essential regulators of the viability of multiple cell lines. Some of the gene hits from EpiDoKOL screen were in concord- ance with previous reports. Fig. 4 Glioblastoma cells have differential dependency to SET1/MLL family of transcription factors interacting with ASH2L. A Scheme of tandem mass spectrometry experiments performed on glioblastoma cells. B, C Proteomic analyses of GFP-tagged ASH2L in U373 (B) and U87MG (C) cells. Volcano plots of significant interactors of GFP-ASH2L isolated from nuclear extracts are shown. Stoichiometry plots of bound SET1/MLL family members (MLL family specific: yellow, SETD1A/B specific: red, MLL & SETD1A/B common members: blue) are depicted. All interactors are normalized to the GFP-ASH2L bait. Results shown represent Intensity Based Absolute Quantification with standard deviations. D Distribution of ASH2L common interactors in terms of subcellular localization in U373 and U87MG cells. E qRT-PCR analysis for mRNA levels upon siRNA mediated knockdown of WDR5 and KMT2A genes in U373 and U87MG cells. F Representative images of long-term clonogenic assay upon siRNA mediated knockdown of WDR5 and KMT2A genes and statistical analysis. Quantification of colonies was performed by ImageJ software. P values determined by two-tailed Student’s t-test P < 0.05, P < 0.01, P < 0.001 (See figure on next page.) ASH2L is essential for tumor progression in vivo To exemplify, suppression of SSRP1 decreases proliferation of malignant glioma through modulation of MAPK signaling [38]. SSRP1 also facilitates hepatocellular carcinoma malignancy [39]. Inhibition of FACT complex eliminates tumor initiating cells and increases survival in preclinical studies [40]. RBX1 is E3 ubiquitin ligase and its repres- sion attenuates tumor proliferation through cell cycle arrest, programmed cell death and senescence [41]. Essentiality of chromatin remodeling factor CHD1 is demonstrated in PTEN-deficient cancer [42]. CHD4 has oncogenic functions in initiating and maintaining In our experiments, we chose to focus on charac- terizing the role of ASH2L, one of the four well-con- served core subunits of SET1/MLL family complexes, since it has not been previously studied in glioblas- toma. Emerging evidence implicates ASH2L in regu- lating cell proliferation and fate. ASH2L enhances Tbx1, Pax7, or Mef2‐mediated transcriptional activ- ity in embryogenesis and stem cell differentiation by increasing H3K4me3 levels [49, 50]. ASH2L has also oncogenic properties [51]. ASH2L interacts with MYC to enhance MYC‐mediated gene transcription and cooperates with activated H-RAS to transform rat embryo fibroblasts [51, 52]. It participates in the cross- talk between H2B ubiquitination and H3K4 methyla- tion to modulate gene transcription [53]. Additionally, ASH2L has been shown to play important roles in the progression of multiple tumors. ASH2L upregu- lates GATA3‐induced transcription of ESR1 in breast cancer cells [51]. ASH2L is recruited to the promoter region of apoptosis‐related genes, co‐activating p53 to promote apoptosis in colorectal cancer [54]. ASH2L is highly expressed in cervical cancer, and its depletion inhibits HeLa cell proliferation [51]. ASH2L is involved in promotion of endometrial cancer progression via upregulation of PAX2 transcription [55]. ASH2L drives proliferation and sensitivity to bleomycin and other genotoxins in Hodgkin’s lymphoma and testicular cancer cells [56]. Low expression of ASH2L protein correlates with a favorable outcome in acute myeloid leukemia [57]. Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 12 of 16 4 (See legend on previous page.) Fig. 4 (See legend on previous page.) Page 13 of 16 Page 13 of 16 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Ozyerli‑Goknar et al. ASH2L is essential for tumor progression in vivo Cell Communication and Signaling (2023) 21:328 In line with previous evidence, we observed the fol- lowing upon ASH2L depletion: 1) colony formation and proliferation of the cells were reduced; 2) apop- tosis was induced; 3) transcription dynamics were altered; 4) cell cycle distribution of cells was altered and G2/M transition of cell cycle was blocked; 5) growth of patient-derived primary cells were inhib- ited; and 6) tumor forming capacity of glioblastoma cells was inhibited. The RNA-seq data of ASH2L- depleted cells showed a high number of downregu- lated genes consistent with the necessity of the WRAD core complex to establish H3K4me1 and H3K4me3 at active enhancers and promoters, respectively [58]. We observed downregulation of MYC target genes (SMARCC1, NOP56) upon ASH2L depletion consist- ent with previous reports. Additionally, cell cycle, mitotic spindle and E2F transcription factor associated genes [59] (e.g. BARD1, PDS5B, RAD21, SMARCC1, KMT5A, ROCK1, NF1, RICTOR) were downregu- lated upon ASH2L depletion. The pathways related to apoptosis phenotype, such as UV-response pathway genes and NfκB/TNF-signaling related genes were also downregulated. Together, we generated a model for the role of ASH2L in regulating multiple cellular pheno- types in glioblastoma cells, such as cell cycle, mitotic spindle formation and apoptosis (Fig. 5J). Accordingly, the SET1/MLL family complex and H3K4 tri-methyl- transferase have been reported to participate in the coordination of cell cycle progression and glioblas- toma proliferation [60]. Similarly, knockout of ASH2L resulted in G2/M arrest accompanied with a prolifera- tion defect in hematopoietic progenitor cells [61, 62]. Moreover, greenCUT&RUN analysis revealed that ASH2L directly occupied the core promoter regions of these downregulated cell cycle regulatory genes imply- ing a direct regulation. On the other hand, in RNAseq analysis a small but considerable number (337) of upregulated genes were identified upon ASH2L depletion and those genes were also direct targets of ASH2L. These upregulated genes were enriched in metabolism; EMT, glycolysis and hypoxia related gene sets and needs further investigation to decipher their surprising negative regulation by ASH2L. Quantitative mass spectrometry of the ASH2L interac- tome for multiple cell lines revealed ASH2L interaction with MLL family and SETD1A/B members with similar stoichiometry in nucleus. We observed that glioblas- toma cells had differential dependency to SET1/MLL family of epigenetic factors, since cells showed differen- tial response to MLL1 and WDR5 ablations. Essentiality of WDR5 together with ASH2L for glioblastoma cells emphasizes the importance of WRAD module for cancer cell fitness. (See figure on next page.) Fig. 5 ASH2L is essential for glioblastoma tumor growth in vivo. A qRT-PCR analysis of ASH2L expression levels upon transduction of GBM4 primary glioblastoma cells with g-ASH2L or g-NT. B Representative images of GBM4 primary neurospheres 16 days post-transduction. C Cell viability analysis of GBM4 cells 1, 3 and 5 days after seeding. D Scheme for in vivo validation of ASH2L essentiality for GBM via intracranial injection of g-ASH2L or g-NT transduced U87MG cells (n = 5 per group) and bioluminescence detection. E Representative images of 3 mice taken during bioluminescence measurements at days 0-12-33 were illustrated. Bioluminescence signal of tumors formed by g-ASH2L or g-NT transduced U87MG cells were compared 33 days post-injection. F The boxplots displaying the ASH2L gene expression ratio (RSEM normalization values) in low grade gliomas (LGG, grade 2 and grade 3) and GBM based on TCGA. G The boxplots displaying the ASH2L gene expression ratio in GBM subtypes based on TCGA. H Representative core images from Brain Glioblastoma tissue microarray (TMA) stained with anti-ASH2L antibody. Scale bar 500 μm. I Percentage of ASH2L-positive cores were shown (n = 80 cores). P values determined by two-tailed Student’s t-test P < 0.05, P < 0.01, *P < 0.001. J Model of ASH2L essentiality for glioblastoma cell survival. ASH2L together with DPY30, RBBP5 and WDR5 forms WRAD module, which acts as cofactor of SET1/MLL family transcription factors (MLL1/2, MLL3/4, SETD1A, SETD1B) to bind promoters of target genes and induce methylation of H3K4, a mark for euchromatin state. Active RNA polymerase binds to open chromatin to initiate transcription of genes involved in cell cycle progression, regulation of mitotic spindles and survival of cells (e.g. TRA2B, BARD1, KIF20B, ARID4A, SMARCC1). Deregulation of ASH2L levels interferes with cell cycle and leads to cell cycle arrest and apoptosis ASH2L is essential for tumor progression in vivo RBBP5, another component of the module, has recently been classified as a novel oncogene involved in gliomagenesis. Amplification and overexpression of the RBBP5 gene were found in patient-derived tumor samples [63]. ASH2L-ablated glioblastoma cells had reduced tumor forming capacity in vivo and ASH2L expression was high in glioblastoma tissues, attesting to ASH2L’s clini- cal relevance. Though we illustrated the essentiality of ASH2L for cell fitness in multiple established cell lines and a primary GBM line, whether this essentiality stems from proteins’ function in SET1/MLL family complexes and to what extent each family member contributes to cell fitness remain elusive and will be subject to further investigation. Conclusion Despite the complexity and heterogeneous nature of cancer, epigenetic therapies hold great promise for improved survival of patients due to their potential of resetting the cancer epigenome. Detection of epigenetic factors modulating tumor survival via high throughput, Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 14 of 16 ig. 5 (See legend on previous page.) Fig. 5 (See legend on previous page.) Fig. 5 (See legend on previous page.) Fig. 5 (See legend on previous page.) Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 Page 15 of 16 Page 15 of 16 robust and affordable screens such as EpiDoKOL holds great potential to enable rapid discovery of novel cancer- related mechanisms and development of effective thera- pies. Designing effective strategies to inhibit ASH2L, identified through our targeted screens as a novel regu- lator of tumor cell survival, can lead to development of successful therapeutic approaches for glioblastoma. Author details 1 Koç University Research Center for Translational Medicine (KUTTAM), Rumelifeneri Yolu, Sarıyer, Istanbul 34450, Türkiye. 2 German Cancer Research Center (DKFZ), German Cancer Consortium (DKTK) Partner Site Freiburg, Heidelberg, Germany. 3 Department of Urology, Medical Center-University of Freiburg, Freiburg, Germany. 4 Institute for Molecular Medicine and Cell Research, University of Freiburg, Freiburg, Germany. 5 Koç University School of Medicine, Istanbul, Türkiye. 6 Vancouver Prostate Centre, University of British Columbia, Vancouver, BC, Canada. 7 Biostatistics, Bioinformatics and Data Management Lab, KUTTAM, Istanbul, Türkiye. 8 Department of Chemical and Biological Engineering, Koç University, Istanbul, Türkiye. 9 Department of Industrial Engineering, Koç University, Istanbul, Türkiye. 10 Botnar Research Centre, Nuffield Department of Orthopedics, Rheumatology and Musculoskel‑ etal Sciences, University of Oxford, Oxford, UK. Availability of data and materials 12. Ruiz S, et al. A genome-wide CRISPR screen identifies CDC25A as a deter‑ minant of sensitivity to ATR inhibitors. Mol Cell. 2016. https://doi.org/10. 1016/j.molcel.2016.03.006. 12. Ruiz S, et al. A genome-wide CRISPR screen identifies CDC25A as a deter‑ minant of sensitivity to ATR inhibitors. Mol Cell. 2016. https://doi.org/10. 1016/j.molcel.2016.03.006. RNA sequencing data is deposited to the NCBI GEO database with the accession number GSE201657. Mass spectrometry data is deposited to the ProteomeX‑ change Consortium via the PRIDE partner repository (PXD033358). greenCUT&RUN sequencing datasets is deposited to the Sequence Read Archive (SRA) portal of the NCBI with accession ID PRJNA828380. List of gene names, domains and sequences of sgRNAs of EpiDoKOL are available in Supplementary Table 1. 13. Krall EB, et al. KEAP1 loss modulates sensitivity to kinase targeted therapy in lung cancer. Elife. 2017. https://doi.org/10.7554/eLife.18970. 13. Krall EB, et al. KEAP1 loss modulates sensitivity to kinase targeted therapy in lung cancer. Elife. 2017. https://doi.org/10.7554/eLife.18970. 14. Miles LA, Garippa RJ, Poirier JT. Design, execution, and analysis of pooled in vitro CRISPR/Cas9 screens. FEBS J. 2016. https://doi.org/10.1111/febs.13770. 14. Miles LA, Garippa RJ, Poirier JT. Design, execution, and analysis of pooled in vitro CRISPR/Cas9 screens. FEBS J. 2016. https://doi.org/10.1111/febs.13770. 15. Shi J, et al. Discovery of cancer drug targets by CRISPR-Cas9 screening of protein domains. Nat Biotechnol. 2015;33:661–7. 15. Shi J, et al. Discovery of cancer drug targets by CRISPR-Cas9 screening of protein domains. Nat Biotechnol. 2015;33:661–7. Funding Financial support was obtained from The Scientific and Technological Research Council of Turkey (TUBITAK) (1003-216S461, TBO), the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) through pro‑ ject TI688/1-1 (HTMT), and Medical Research Council (MRC) career develop‑ ment fellowship (MR/V010182/1, APC). 10. Hart T, et al. High-resolution CRISPR screens reveal fitness genes and genotype-specific cancer liabilities. Cell. 2015. https://doi.org/10.1016/j. cell.2015.11.015. 10. Hart T, et al. High-resolution CRISPR screens reveal fitness genes and genotype-specific cancer liabilities. Cell. 2015. https://doi.org/10.1016/j. cell.2015.11.015. 11. Heigwer F, et al. CRISPR library designer (CLD): software for multispecies design of single guide RNA libraries. Genome Biol. 2016. https://doi.org/ 10.1186/s13059-016-0915-2. 11. Heigwer F, et al. CRISPR library designer (CLD): software for multispecies design of single guide RNA libraries. Genome Biol. 2016. https://doi.org/ 10.1186/s13059-016-0915-2. Ethics approval and consent to participate All in vivo experiments were approved by the institutional ethical committee of Koç University. 17. Steward MM, et al. Molecular regulation of H3K4 trimethylation by ASH2L, a shared subunit of MLL complexes. Nat Struct Mol Biol. 2006. https://doi. org/10.1038/nsmb1131. 17. Steward MM, et al. Molecular regulation of H3K4 trimethylation by ASH2L, a shared subunit of MLL complexes. Nat Struct Mol Biol. 2006. https://doi. org/10.1038/nsmb1131. Acknowledgements The authors acknowledge the use of the services and facilities of the Koç University Research Center for Translational Medicine (KUTTAM), funded by the Presidency of Strategy and Budget, Türkiye. We also thank to Ebru Yilmaz, Nareg Pınarbaşı-Değirmenci and Dr. Özlem Yedier-Bayram for their assistance with experimental analysis and image design. 5. Khorasanizadeh S. The nucleosome: from genomic organization to genomic regulation. Cell. 2004. https://doi.org/10.1016/S0092-8674(04) 00044-3. 5. Khorasanizadeh S. The nucleosome: from genomic organization to genomic regulation. Cell. 2004. https://doi.org/10.1016/S0092-8674(04) 00044-3. 6. Strahl BD, Allis CD. The language of covalent histone modifications. Nature. 2000;403:41–5. 6. Strahl BD, Allis CD. The language of covalent histone modifications. Nature. 2000;403:41–5. 7. Bolden JE, Peart MJ, Johnstone RW. Anticancer activities of histone dea‑ cetylase inhibitors. Nat Rev Drug Discov. 2006;5:769–84. 7. Bolden JE, Peart MJ, Johnstone RW. Anticancer activities of histone dea‑ cetylase inhibitors. Nat Rev Drug Discov. 2006;5:769–84. References 1. Ostrom QT, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2012–2016. Neuro-Oncology. 2019. https://doi.org/10.1093/neuonc/ noz150. activity and essentiality validation experiments. Supplementary Table 3. Sequences of external and internal PCR primers used for Nested-PCR. Supplementary Table 4. Sequences qPCR primers used for RNAseq validation. 2. Preusser M, et al. Current concepts and management of glioblastoma. Ann Neurol. 2011;70:9–21. 3. Ozyerli-Goknar E, Bagci-Onder T. Epigenetic deregulation of apoptosis in cancers. Cancers. 2021. https://doi.org/10.3390/cancers13133210. 4. Esteller M. Epigenetics in cancer. - main article. N Engl J Med. 2008;358:1148–59. 4. Esteller M. Epigenetics in cancer. - main article. N Engl J Med. 2008;358:1148–59. Consent for publication 18. Patel A, Dharmarajan V, Vought VE, Cosgrove MS. On the mechanism of multiple lysine methylation by the human mixed lineage leukemia pro‑ tein-1 (MLL1) core complex. J Biol Chem. 2009. https://doi.org/10.1074/ jbc.M109.014498. 18. Patel A, Dharmarajan V, Vought VE, Cosgrove MS. On the mechanism of multiple lysine methylation by the human mixed lineage leukemia pro‑ tein-1 (MLL1) core complex. J Biol Chem. 2009. https://doi.org/10.1074/ jbc.M109.014498. Authors’ contributions Study design: TBO, EOG, FSP, CA, NAL, TTO; data generation: EOG, EYK, ACA, IB, AC, SN, MB, SHYK, MP, APC; data analysis: EOG, EYK, ACA, IB, APC, SN, TM, HS, NT, MG; data interpretation: TBO, EOG, HTMT; supplied reagents: CAA, TTO, HTMT; manuscript draft: EOG and TBO; approved final manuscript: all authors. 8. Mojica FJM, Díez-Villaseñor C, García-Martínez J, Almendros C. Short motif sequences determine the targets of the prokaryotic CRISPR defence system. Microbiology. 2009;155:733–40. 9. Evers B, et al. CRISPR knockout screening outperforms shRNA and CRISPRi in identifying essential genes. Nat Biotechnol. 2016. https://doi.org/10. 1038/nbt.3536. 9. Evers B, et al. CRISPR knockout screening outperforms shRNA and CRISPRi in identifying essential genes. Nat Biotechnol. 2016. https://doi.org/10. 1038/nbt.3536. Supplementary Information The online version contains supplementary material available at https://doi. org/10.1186/s12964-023-01335-6. Received: 2 May 2023 Accepted: 26 September 2023 Additional file 1: Supplementary Figure 1. Quality Check of Epi‑ DoKOL in plasmid level. Supplementary Figure 2. Effects of candidate genes on U87MG cell fitness. Supplementary Figure 3. RNAseq and greenCUT&RUN analyses in glioblastoma cells. Supplementary Figure 4. Cytoplasmic interactions of ASH2L in glioblastoma cells. Supplemen‑ tary Figure 5. ASH2L immunohistochemistry in glioblastoma. Sup‑ plementary Table 1. Sequences and targets of sgRNAs of EpiDoKOL. Supplementary Table 2. All sgRNA sequences used for EpiDoKOL Cas9 activity and essentiality validation experiments. Supplementary Table 3. Sequences of external and internal PCR primers used for Nested-PCR. Supplementary Table 4. Sequences qPCR primers used for RNAseq validation. Additional file 1: Supplementary Figure 1. Quality Check of Epi‑ DoKOL in plasmid level. Supplementary Figure 2. Effects of candidate genes on U87MG cell fitness. Supplementary Figure 3. RNAseq and greenCUT&RUN analyses in glioblastoma cells. Supplementary Figure 4. Cytoplasmic interactions of ASH2L in glioblastoma cells. Supplemen‑ tary Figure 5. ASH2L immunohistochemistry in glioblastoma. Sup‑ plementary Table 1. Sequences and targets of sgRNAs of EpiDoKOL. Supplementary Table 2. All sgRNA sequences used for EpiDoKOL Cas9 activity and essentiality validation experiments. Supplementary Table 3. Sequences of external and internal PCR primers used for Nested-PCR. Supplementary Table 4. Sequences qPCR primers used for RNAseq validation. Declarations 16. Dou Y, et al. Regulation of MLL1 H3K4 methyltransferase activity by its core components. Nat Struct Mol Biol. 2006. https://doi.org/10.1038/ nsmb1128. 16. Dou Y, et al. Regulation of MLL1 H3K4 methyltransferase activity by its core components. Nat Struct Mol Biol. 2006. https://doi.org/10.1038/ nsmb1128. Competing interests p g APC is co-founder and an employee of Caeruleus Genomics Ltd and is an inventor of several patents related to sequencing technologies filed by Oxford University Innovations. 19. Li Y, et al. Structural basis for activity regulation of MLL family methyl‑ transferases. Nature. 2016. https://doi.org/10.1038/nature16952. 19. Li Y, et al. Structural basis for activity regulation of MLL family methyl‑ transferases. Nature. 2016. https://doi.org/10.1038/nature16952. Page 16 of 16 Page 16 of 16 Ozyerli‑Goknar et al. Cell Communication and Signaling (2023) 21:328 20. Stemmer M, Thumberger T, Del Sol Keyer M, Wittbrodt J, Mateo JL. CCTop: an intuitive, flexible and reliable CRISPR/Cas9 target prediction tool. PLoS One. 2015. https://doi.org/10.1371/journal.pone.0124633. 43. Xia L, et al. CHD4 has oncogenic functions in initiating and maintaining epigenetic suppression of multiple tumor suppressor genes. Cancer Cell. 2017. https://doi.org/10.1016/j.ccell.2017.04.005. 21. Sanjana NE, Shalem O, Zhang F. Improved vectors and genome-wide libraries for CRISPR screening. Nat Methods. 2014. https://doi.org/10. 1038/nmeth.3047. 44. Beaulieu N, et al. An essential role for DNA methyltransferase DNMT3B in cancer cell survival. J Biol Chem. 2002. https://doi.org/10.1074/jbc.M2047 34200. 22. Bagci-Onder T, et al. Real-time imaging of the dynamics of death receptors and therapeutics that overcome TRAIL resistance in tumors. Oncogene. 2012;32:2818–27. 45. Zheng Y, et al. Identification of SUV39H2 as a potential oncogene in lung adenocarcinoma. Clin Epigenetics. 2018. https://doi.org/10.1186/ s13148-018-0562-4. 23. Onder TT, et al. Chromatin-modifying enzymes as modulators of repro‑ gramming. Nature. 2012;483:598–602. 46. Piao L, et al. Histone methyltransferase SUV39H2 serves oncogenic roles in osteosarcoma. Oncol Rep. 2019. https://doi.org/10.3892/or.2018.6843. 46. Piao L, et al. Histone methyltransferase SUV39H2 serves oncogenic roles in osteosarcoma. Oncol Rep. 2019. https://doi.org/10.3892/or.2018.6843. 47. Ladang A, et al. Elp3 drives Wnt-dependent tumor initiation and regen‑ eration in the intestine. J Exp Med. 2015. https://doi.org/10.1084/jem. 20142288. 24. Li W, et al. MAGeCK enables robust identification of essential genes from genome-scale CRISPR/Cas9 knockout screens. Genome Biol. 2014. https://doi.org/10.1186/s13059-014-0554-4. 47. Ladang A, et al. Elp3 drives Wnt-dependent tumor initiation and regen‑ eration in the intestine. J Exp Med. 2015. https://doi.org/10.1084/jem. 20142288. 25. Nizamuddin S, et al. Integrating quantitative proteomics with accurate genome profiling of transcription factors by greenCUT&RUN. Nucleic Acids Res. 2021. https://doi.org/10.1093/nar/gkab038. 48. Chen X, et al. Upregulated WDR5 promotes proliferation, self-renewal and chemoresistance in bladder cancer via mediating H3K4 trimethylation. Sci Rep. 2015. https://doi.org/10.1038/srep08293. g pi g p y g Acids Res. 2021. https://doi.org/10.1093/nar/gkab038. 26. Koidl S, Timmers HTM. Competing interests greenCUT&RUN: efficient genomic profiling of GFP-tagged transcription factors and chromatin regulators. Curr Protoc. 2021. https://doi.org/10.1002/cpz1.266. 49. Chang NC, et al. The dystrophin glycoprotein complex regulates the epigenetic activation of muscle stem cell commitment. Cell Stem Cell. 2018. https://doi.org/10.1016/j.stem.2018.03.022. 50. Stoller JZ, et al. Ash2l interacts with Tbx1 and is required during early embryogenesis. Exp Biol Med. 2010. https://doi.org/10.1258/ebm.2010. 009318. 27. Yedier-Bayram O, et al. EPIKOL, a chromatin-focused CRISPR/Cas9-based screening platform, to identify cancer-specific epigenetic vulnerabilities. bioRxiv. 2021. https://doi.org/10.1101/2021.05.14.444239. g p y pi p g bioRxiv. 2021. https://doi.org/10.1101/2021.05.14.444239. 51. Lüscher-Firzlaff J, et al. The human trithorax protein hASH2 functions as an oncoprotein. Cancer Res. 2008. https://doi.org/10.1158/0008-5472. CAN-07-3158. 28. Shilatifard A. The COMPASS family of histone H3K4 methylases: mechanisms of regulation in development and disease pathogenesis. Annu Rev Biochem. 2012. https://doi.org/10.1146/annurev-bioch em-051710-134100. 52. Ullius A, et al. The interaction of MYC with the trithorax protein ASH2L promotes gene transcription by regulating H3K27 modification. Nucleic Acids Res. 2014. https://doi.org/10.1093/nar/gku312. 29. Cenik BK, Shilatifard A. COMPASS and SWI/SNF complexes in devel‑ opment and disease. Nat Rev Genet. 2021. https://doi.org/10.1038/ s41576-020-0278-0. 53. Wu L, et al. ASH2L regulates ubiquitylation signaling to MLL: trans-regu‑ lation of H3 K4 methylation in higher eukaryotes. Mol Cell. 2013. https:// doi.org/10.1016/j.molcel.2013.01.033. 53. Wu L, et al. ASH2L regulates ubiquitylation signaling to MLL: trans-regu‑ lation of H3 K4 methylation in higher eukaryotes. Mol Cell. 2013. https:// doi.org/10.1016/j.molcel.2013.01.033. 30. Esteller M, et al. Inactivation of the DNA-repair gene MGMT and the clini‑ cal response of gliomas to alkylating agents. N Engl J Med. 2000. https:// doi.org/10.1056/nejm200011093431901. 54. Mungamuri SK, Wang S, Manfredi JJ, Gu W, Aaronson SA. Ash2L enables P53-dependent apoptosis by favoring stable transcription pre-initiation complex formation on its pro-apoptotic target promoters. Oncogene. 2015. https://doi.org/10.1038/onc.2014.198. 54. Mungamuri SK, Wang S, Manfredi JJ, Gu W, Aaronson SA. Ash2L enables P53-dependent apoptosis by favoring stable transcription pre-initiation complex formation on its pro-apoptotic target promoters. Oncogene. 2015. https://doi.org/10.1038/onc.2014.198. 31. Quinn JA, et al. Phase II trial of temozolomide plus O6-benzylguanine in adults with recurrent, temozolomide-resistant malignant glioma. J Clin Oncol. 2009. https://doi.org/10.1200/JCO.2008.18.8417. p g 55. Zeng K, et al. ASH2L is involved in promotion of endometrial cancer pro‑ gression via upregulation of PAX2 transcription. Cancer Sci. 2020. https:// doi.org/10.1111/cas.14413. 55. Zeng K, et al. ASH2L is involved in promotion of endometrial cancer pro‑ gression via upregulation of PAX2 transcription. Cancer Sci. 2020. https:// doi.org/10.1111/cas.14413. 32. Sharma SV, et al. Competing interests Bralten LBC, et al. Integrated genomic profiling identifies candidate genes implicated in glioma-genesis and a novel LEO1-SLC12A1 fusion gene. Genes Chromosom Cancer. 2010. https://doi.org/10.1002/gcc.20760. 63. Bralten LBC, et al. Integrated genomic profiling identifies candidate genes implicated in glioma-genesis and a novel LEO1-SLC12A1 fusion gene. Genes Chromosom Cancer. 2010. https://doi.org/10.1002/gcc.20760. 40. Dermawan JKT, et al. Pharmacological targeting of the histone chaper‑ one complex FACT preferentially eliminates glioblastoma stem cells and prolongs survival in preclinical models. Cancer Res. 2016. https://doi.org/ 10.1158/0008-5472.CAN-15-2162. Competing interests A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell. 2010;141:69–80. 33. Romani M, Pistillo MP, Banelli B. Epigenetic targeting of gliobla 33. Romani M, Pistillo MP, Banelli B. Epigenetic targeting of glioblastoma. Front Oncol. 2018. https://doi.org/10.3389/fonc.2018.00448. Front Oncol. 2018. https://doi.org/10.3389/fonc.2018.00448. 56. Constantin D, Widmann C. ASH2L drives proliferation and sensitiv‑ ity to bleomycin and other genotoxins in Hodgkin’s lymphoma and testicular cancer cells. Cell Death Dis. 2020. https://doi.org/10.1038/ s41419-020-03231-0. 34. Krauze AV, et al. A phase 2 study of concurrent radiation therapy, temo‑ zolomide, and the histone deacetylase inhibitor valproic acid for patients with glioblastoma. Int J Radiat Oncol Biol Phys. 2015. https://doi.org/10. 1016/j.ijrobp.2015.04.038. 57. Butler JS, et al. Low expression of ASH2L protein correlates with a favora‑ ble outcome in acute myeloid leukemia. Leuk Lymphoma. 2017. https:// doi.org/10.1080/10428194.2016.1235272. 35. Halford SER, et al. Results of the OPARATIC trial: a phase I dose escalation study of olaparib in combination with temozolomide (TMZ) in patients with relapsed glioblastoma (GBM). J Clin Oncol. 2018. https://doi.org/10. 1200/jco.2017.35.15_suppl.2022. 58. Jiang H. The complex activities of the SET1/MLL complex core subunits in development and disease. Biochim Biophys Acta Gene Regul Mech. 2020. https://doi.org/10.1016/j.bbagrm.2020.194560. 36. Su JM, et al. A phase I trial of veliparib (ABT-888) and temozolomide in children with recurrent CNS tumors: a pediatric brain tumor consortium report. Neuro Oncol. 2014. https://doi.org/10.1093/neuonc/nou103. 59. Malumbres M. Cyclin-dependent kinases. Genome Biol. 2014. https://doi. org/10.1186/gb4184. 59. Malumbres M. Cyclin-dependent kinases. Genome Biol. 2014. https://doi. org/10.1186/gb4184. 37. Munoz DM, et al. CRISPR screens provide a comprehensive assessment of cancer vulnerabilities but generate false-positive hits for highly amplified genomic regions. Cancer Discov. 2016. https://doi.org/10.1158/2159- 8290.CD-16-0178. 60. Beilharz TH, et al. Coordination of cell cycle progression and mitotic spindle assembly involves histone H3 lysine 4 methylation by set1/ COMPASS. Genetics. 2017. https://doi.org/10.1534/genetics.116. 194852.f 38. Liao J, et al. SSRP1 silencing inhibits the proliferation and malignancy of human glioma cells via the MAPK signaling pathway. Oncol Rep. 2017. https://doi.org/10.3892/or.2017.5982. 61. Lüscher-Firzlaff J, et al. Hematopoietic stem and progenitor cell prolifera‑ tion and differentiation requires the trithorax protein Ash2l. Sci Rep. 2019. https://doi.org/10.1038/s41598-019-44720-3. p g 39. Ding Q, et al. SSRP1 contributes to the malignancy of hepatocellular car‑ cinoma and is negatively regulated by MIR-497. Mol Ther. 2016. https:// doi.org/10.1038/mt.2016.9. g 62. Liang W, et al. Loss of the epigenetic regulator Ash2l results in desintegra‑ tion of hepatocytes and liver failure. Int J Clin Exp Pathol. 2016;9:5167–75.ii 63. Publisher’s Note 41. Jia L, Soengas MS, Sun Y. ROC1/RBX1 E3 ubiquitin ligase silencing suppresses tumor cell growth via sequential induction of G2-M arrest, apoptosis, and senes‑ cence. Cancer Res. 2009. https://doi.org/10.1158/0008-5472.CAN-08-4671. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. 42. Zhao D, et al. Synthetic essentiality of chromatin remodelling factor CHD1 in PTEN-deficient cancer. Nature. 2017. https://doi.org/10.1038/ nature21357.
https://openalex.org/W2284022987	http://www.scielo.br/pdf/cadsc/v23n2/1414-462X-cadsc-23-2-198.pdf	English	null	Standard obesogenic diet: the impact on oral health in children and teenagers at the Recôncavo Baiano - Brazil	Cadernos Saúde Coletiva	2,015	cc-by	6,037	DOI: 10.1590/1414-462X201500020038 DOI: 10.1590/1414-462X201500020038 Original Article Dieta de padrão obesogênico: o impacto na saúde bucal em crianças e em adolescentes no Recôncavo Baiano, Brasil Rita de Cássia Ribeiro Silva1, Luce Alves da Silva2, Roberto Paulo Correia de Araújo3, Felipe Fagundes Soares4, Rosemeire Leovigildo Fiaccone5, Maria Cristina Teixeira Cangussu6 Rita de Cássia Ribeiro Silva1, Luce Alves da Silva2, Roberto Paulo Correia de Araújo3, Felipe Fagundes Soares4, Rosemeire Leovigildo Fiaccone5, Maria Cristina Teixeira Cangussu6 Standard obesogenic diet: the impact on oral health in children and teenagers at the Recôncavo Baiano - Brazil Dieta de padrão obesogênico: o impacto na saúde bucal em crianças e em adolescentes no Recôncavo Baiano, Brasil 5 PhD in Statistics, Lancaster University (LANCASTER). Professor Associate of Instituto de Matemática of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil. 6PhD in Public Health São Paulo University (USP) Professor Associate of Faculdade de Odontologia of Universidade Federal da Bahia (UFBA) Salvador (BA) 1 PhD in Public Health. Professor Associate of Departamento de Ciências da Nutrição of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil. 3 Doctorate in Dentistry. Professor Associate of Instituto de Ciências da Saúde of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil. 6 PhD in Public Health, São Paulo University (USP). Professor Associate of Faculdade de Odontologia of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil. Correspondence: Maria Cristina Teixeira Cangussu – Rua Desembargador Baldoíno Andrade, 211/702 – Chame-chame – CEP: 40157-180 – Salvador (BA), Brasil - Email: cangussu@ufba.br ( ), 5 PhD in Statistics, Lancaster University (LANCASTER). Professor Associate of Instituto de Matemática of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil. 6 PhD in Public Health, São Paulo University (USP). Professor Associate of Faculdade de Odontologia of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil ▄ ▄INTRODUCTION only a few studies have evaluated their influence on dental caries prevalence13-16. Therefore, the objective of this study was to fill this void by evaluating the influence of dietary patterns on the occurrence of dental caries in the population of children in a municipality of northeastern Brazil. It is hoped that such studies can provide information on the design and implementation of strategies to promote good oral health and nutritional habits. The most important changes that have been observed in dietary patterns are the increased consumption of foods with high energy density, including saturated fat and simple carbohydrates, instead of the consumption of foods of plant origin, which present low caloric density and are sources of micronutrients and fibers1,2. These eating habits, which are related to obesity, may also result in higher prevalence of dental caries because of the frequent intake of sucrose, an important etiologic factor of this dental disease3-7. The American Dietetic Association8 asserts that nutrition is a key to good oral health, and that nutrition in combination with foods have a direct influence on the progression of dental caries. Although the epidemiological data on dental caries have shown a significant decline in its prevalence in schoolchildren under eighteen years old, morbidity is still the greatest problem of oral health in Brazil. The results of the National Oral Health Survey - SB Brazil 2010 show the northeast region with a Decay, Missing and Filled Teeth index - DMFT of 2.63 and the state capital of Bahia with an even better DMFT of 1,7,9,10. Approximately 50% of the twelve-year old children were caries free. Changes in dietary pattern facilitate the development of dental caries, even if individuals are exposed to protective factors, such as the widespread use of fluoride11. Resumo O objetivo deste estudo é avaliar a influência de padrões alimentares sobre a ocorrência de cárie dentária. Realizou-se um estudo transversal de base populacional, com uma amostra de 1.439 crianças de 6 a 13 anos, matriculadas em escolas públicas da cidade de São Francisco do Conde, no Estado da Bahia. Os exames de cárie dentária foram realizados por dois cirurgiões-dentistas (Kappa=0,72) no ambiente escolar, sob luz natural, com o auxílio de uma espátula de madeira e gaze. A ingestão alimentar foi avaliada por meio do questionário qualitativo-quantitativo de frequência alimentar. Um questionário sobre as condições demográficas e socioeconômicas foi aplicado nas famílias. A análise multivariada de Poisson foi realizada para fins estatísticos. Considerando ambas as dentições, as prevalências de cárie foram 82,14 e 37,3% nas dentições permanentes e decíduas, respectivamente. Os padrões alimentares identificados foram classificados em “padrão obesogênico”, e os resultados mostraram uma associação estatisticamente significativa entre o padrão alimentar obesogênico e a cárie dentária “prudente”. A proporção de alunos com cárie foi 1,4 vez maior entre aqueles que consomem frequentemente alimentos constituídos de uma dieta obesogênica padrão (RP=1,40 e IC95%=1,04-1,96) em comparação com aqueles que consomem, com menor frequência, esses tipos de alimentos. Uma associação positiva entre o padrão alimentar obesogênico e a cárie dentária foi encontrada nos estudantes Portanto, a promoção de hábitos alimentares saudáveis e de hábitos nutricionais devem ser incentivados como forma de prevenção da cárie dentária em crianças e em adolescentes. Palavras-chave: cárie dentária; padrões de consumo de alimentos; crianças; adolescentes Abstract The aim of this study is to evaluate the influence of dietary patterns on the occurrence of dental caries. Using a population-based, cross-sectional study design, we assessed a sample of 1.439 children aged 6-13 years who were enrolled in public schools in the municipality of São Francisco do Conde, Bahia state, Brazil. Oral examinations for dental caries were performed by two dental surgeons (Kappa = 0.72) in the school environment under natural light with the aid of a wooden spatula and gauze. Dietary intake was assessed using the qualitative-quantitative food frequency questionnaire. A questionnaire on the demographic and socioeconomic conditions was applied to the families. The Poisson multivariate analysis was performed for statistical purposes. Considering both dentitions, the prevalence of caries was 82.14% and 37.3% in the permanent and deciduous dentitions, respectively. The identified dietary patterns were classified into “standard obesogenic” and “prudent.” The results showed statistically significant association between the obesogenic dietary pattern and dental caries; the proportion of students with dental caries was 1.4 times higher among those who most frequently consumed foods that constitute a standard obesogenic diet (PR = 1.40, 95% CI: 1.04-1.96) compared to those who less frequently consumed these types of foods. A positive association between the obesogenic dietary pattern and dental caries was found in the students investigated. Therefore, the promotion of healthy eating habits and nutritional habits should be encouraged to prevent dental caries in children and adolescents. Keywords: dental caries; food consumption patterns; children; adolescents. words: dental caries; food consumption patterns; children 1 PhD in Public Health. Professor Associate of Departamento de Ciências da Nutrição of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil 3 Doctorate in Dentistry. Professor Associate of Instituto de Ciências da Saúde of Universidade Federal da Bahia (UFBA) - Salvador (BA), Brasil. Correspondence: Maria Cristina Teixeira Cangussu – Rua Desembargador Baldoíno Andrade, 211/702 – Chame-chame – CEP: 40157-180 – Salvador (BA), Brasil - Email: cangussu@ufba.br Financial support: Fundação de Amparo à Pesquisa do Estado da Bahia – FAPESB, project nº 7638/2009. Conflict of interests: None. Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 198 Obesogenic dietary pattern and dental caries Data collection Oral examinations were performed according to the WHO criteria10,13 by two qualified dental surgeons (Kappa = 0.72) in the school environment under natural light with the aid of a wooden spatula and gauze. To determine the severity of dental caries, the following indices were used: DMFT (permanent teeth) and DMFT (deciduous teeth). Decayed teeth (c), teeth indicated for extraction (e) and filled teeth (o) were included, whereas extracted teeth for which there was difficulty in identifying the cause of extraction, such as either decay or natural process of tooth exfoliation, were excluded. The prevalence of dental caries was considered as the outcome variable, i.e., the occurrence of at least one decayed tooth or a previous history of dental caries in the primary or permanent dentition. Study design/Population/Sample Study design/Population/Sample A cross-sectional study design was used to assess 6- to 13-year-old children living in São Francisco do Conde (SFC), a municipality located in the metropolitan area of Salvador, northeastern Brazil. This municipality has 33.183 inhabitants and a high urbanization rate (80.2%). The city council is the largest local employer. Out of the 417 municipalities in Bahia state, São Francisco do Conde ranks third in the economic development index, but far lower positions regarding social (30th), education (139th) and health (178th) development indexes.. Sample size estimation was carried out using data from the Education Department of the municipality of SFC in 2010. Of the 3.734 students enrolled, 2,649 students were from rural areas and 1.085 students were from urban areas. These students were distributed across 22 schools in the municipal school system. To minimize travel costs and the time required for subject recruitment, the nine schools with 150 or more students were included in the sample. All students aged 6-13 years in each selected school were eligible for the study. Because the prevalence of respiratory allergies varies from 15% to 40%, sample sizes of 531 and 834 students from the urban and rural areas, respectively, were selected after taking into account a 3% error and a 95% confidence level. We added 10% to the total sample size to compensate for the loss of students who refused to participate in the study. Most studies use food alone or associated with micronutrients to study the association between dietary pattern and chronic diseases, including the occurrence of dental caries. However, food and nutrients are consumed together. The World Health Organization - WHO suggests that in nutritional epidemiology, the assessment of dietary intake in populations should be based on eating patterns12 for better prediction of the risk of diseases than on the analysis of food and nutrients separately. To determine the dietary patterns, factor analysis with the principal components extraction method was used. Although dietary patterns have been widely investigated as the predictors of chronic diseases, Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 199 Rita de Cássia Ribeiro Silva, Luce Alves da Silva, Roberto Paulo Correia de Araújo, Felipe Fagundes Soares, Rosemeire Leovigildo Fiaccone, M ce Alves da Silva, Roberto Paulo Correia de Araújo, Felipe Fagundes Soares, Rosemeire Leovigildo Fiaccone, Maria Cristina Teixeira Cangussu Confounding variablesh The adjustment variables used in the study included the following: gender (male, female (reference category), age (<10 years, ≥ 10 years (reference category), education of the caregiver (≤ 4th grade, >4th grade (reference category), place of residence (urban, rural (reference category), per capita income ( <1 minimum salary (SM), ≥1 SM (reference category), number of people living in the household (>3, ≤3 (reference category), and age of the caregiver (<30 years, > 30 years (reference category). To assess the anthropometric status, the WHO tables19 with percentile values of body mass index [BMI = weight (kg)/height (m)2] according to age and sex were used as the reference. For classification of the anthropometric status, we used the WHO19 guidelines: underweight (<3rd percentile); normal weight (≥3rd percentile and <85th percentile, reference category); overweight (≥ 85th percentile and <97th percentile); and obese (≥ 97th percentile). For analysis, the overweight and obese categories were aggregated. Therefore, children with excess weight were at or above the 85th percentile. Procedures eggs, sausages in general, breads, cereals (rice, pasta, cakes, flours), roots and tubers, legumes, vegetables, fruits, greens, sauces, artificial sweeteners. The frequency of consumption of foods or food groups was summarized with a single value for each student, and the formula used for obtaining this value is as follows: (Σ frequency of consumption of foods contained in the food group)/No. of food group * maximum frequency of consumption according to the FFQ14,18. For the energy intake, the nutrition program Virtual Nutri version 1.0 was used. Data were collected from August to December 2010 by a qualified and trained team. The principals of the selected schools received an invitation letter requesting the participation of the schools in the research. This letter contained information about the objectives and methodology of the research. Additional meetings for clarification on the research and persuasion of subjects were conducted, and informed consent was obtained from both the schools and the parents. The parents were invited to attend the school interviews at which the questionnaires were administered, and the information was recorded in appropriate forms. ▄ ▄RESULTS From the total number of initially selected students (1,500 students), 61 (4.0%) students withdrew from the study or discontinued their participation (the reasons included refusal to participate, family moved to another town, or the child was transferred to another school), thereby 1,439 students of both sexes aged 6-13 years remained in the study. Lacking data; Measured in Brazilian minimum salary (BSM). Value in 2010= R$510.00 (Brazilian reais; equivalent to US$ 290.7) Among the children and adolescents surveyed, 45.6% were female and 50.5% were over 10 years old. Among the students’ caregivers, 96.2% were female; 50.6% were aged 30-39 years; 26.7% attended primary school –1st to 4th grade; 26.6% attended secondary school – 5th to 8th grade; and 4.6% were illiterate. It is worth noting that 72.5% of the households received between 2 and 3 minimum salaries. Additional information is provided in Table 1. drinks, soft drinks, typical preparations, bakery items, fast food, oils and fats, milk/dairy products, and cereals. The indices for internal consistency (a >0.60) indicated an acceptable level of measurement accuracy for the two standards, thereby ensuring satisfactory internal consistency of the dimensions in the questionnaire used. The following foods and food groups were not considered in the analysis due to low commonality (h2<20): chicken, fish, eggs, sausages, bread, vegetables, sauces, and sweeteners. Considering both dentitions, the data indicate that 82.14% of participants presented dental caries activity; there was an increase in dental caries prevalence with age in the permanent dentition (p<0.001), and a decrease in dental caries prevalence with age in the deciduous dentition (p<0.001). The mean DMFT index was 0.94 (±1.55) (Table 2). After adjusting for anthropometric characteristics, such as the student’s age, maternal age, family income, maternal education, school location and energy intake, there was a statistically significant association between the obesogenic dietary pattern and dental caries; the proportion of students with dental caries was 1.4 times higher among those who most frequently consumed foods that constitute a standard obesogenic diet (PR = 1.40, 95% CI: 1.04-1.96) compared with those who least frequently consumed these types of foods. Moreover, there was no association between the prudent dietary pattern and the presence of dental caries (PR = 0.88, 95% CI: 0.64-1.20) (Table 4). Statistical analysis Characteristics of the study population – children and adolescents aged 7-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 each consumption pattern were dichotomised as Percentile > P75 versus Percentile ≤ P75. Table 1. Characteristics of the study population – children and adolescents aged 7-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 Variables n % Age() <10 years 697 49.5 ≥10 years 712 50.5 Gender Male 783 54.4 Female 656 45.6 Caregiver’s age (in years)() 19-24 18 1.3 25-29 281 20.8 30-39 683 50.6 ≥40 369 27.3 Education of caregiver () illiterate 63 4.6 1st to 4th grade 367 26.7 5th to 8th grade 366 26.6 Secondary school or higher 580 32.2 Per capita income (*) ≤1 BMS 215 15.6 2-3 BMS 997 72.5 ≥4 BMS 164 12.0 Place of residence Urban 486 33.8 Rural 953 66.2 Lacking data; *Measured in Brazilian minimum salary (BSM). Value in 2010= R$510.00 (Brazilian reais; equivalent to US$ 290.7) Poisson multivariate analysis was used for statistical purposes. The magnitude of the association between dietary patterns and dental caries was expressed in terms of prevalence ratio (PR) and corresponding 95% confidence intervals (95% CI). For the statistical analyses, two-tailed tests and significance level of 5% were used. The choice of confounding variables in the model was based on the information existing in the available literature. The Statistical Package for the Social Sciences - SPSS, v. 17.0 was used for statistical analyses. Ethical issues The study protocol was submitted to the Ethics Committee of the School of Nutrition, Federal University of Bahia, which commented favorably on the relevance of research ethics under the 006-06/CEP record. The parents and caregivers signed the informed consent form, thereby allowing the participation of children in this research. Statistical analysis The Quantitative Food Frequency Questionnaire (QFFQ), developed by Slater et al.14, was adapted to the population and adopted to evaluate the frequency of food intake in students in Sao Paulo. The QFFQ is an instrument that has been validated by Voci et al.17, and it comprises 132 food items. For the present study, regional foods that were originally not a part of the validated questionnaire were included in the QFFQ. This new questionnaire is under the process of validation. The rate of consumption of these regional food items was divided into the following five categories: never consumed = 0; consumed one to three times a month = 1; consumed one to two times a week = 2; consumed two to four times a week = 3; and consumed four or more times a week = 4. For analysis, these food items were grouped into 23 categories according to the similarity of nutritional content: sugars/sweets, typical dishes of the Brazilian cuisine (feijoada: a black bean stew with pork; “drover beans”: a dish made with beans, cassava flour, sausage, garlic, onions, bacon and eggs; acarajé: a dish made of bean seeds made into balls and fried in palm oil), artificial drinks, soft drinks, confectionery, fast food, oils/fats, milk and dairy products, meat, chicken, fish, For the processing and construction of the database, Epi Info version 6.04 was used (Centers for Disease Control and Prevention, Atlanta, United States), and duplicate data entry was adopted, after reviewing the questionnaires and correcting the errors because of coding, which was initially performed in the field. i The characteristics of the population were identified by descriptive analysis using categorised data of prevalence. Food consumption patterns were derived from Principal Component Analysis (PCA)20. Prior to performing exploratory factor analysis, the coefficient of Kaiser-Mayer-Olkin (KMO) was estimated, and Bartlett’s test of sphericity was used to indicate the appropriateness of factor analysis. Principal component analysis was performed, followed by orthogonal rotation (Varimax) to examine the exploratory factor structure (pattern). The number of factors extracted was based on the interpretability of data, as recommended by Hearty and Gibney21. The internal consistency of the dimensions of QFFQ was assessed and considered acceptable at levels of Cronbach’s alpha > 0.65. The scores of Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 200 Obesogenic dietary pattern and dental caries Table 1. ▄ ▄RESULTS The confidence factorial index confirmed by the KMO coefficient (0.873) and Bartlett’s test of sphericity (p<0.001) showed that the values for the correlations between the items were sufficient and adequate for performing factor analysis. From this analysis, two patterns were obtained, which represented 45.70% of the total variance, and these patterns determined the classification of the foods as “obesogenic” and “prudent” (Table 3). The obesogenic pattern was composed of sugars/sweets, artificial Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 201 ce Alves da Silva, Roberto Paulo Correia de Araújo, Felipe Fagundes Soares, Rosemeire Leovigildo Fiaccone, Maria Cristina Teixeira Cangussu de Cássia Ribeiro Silva, Luce Alves da Silva, Roberto Paulo Correia de Araújo, Felipe Fagundes Soares, Rosemeire Leovigildo Fiaccone, Maria Table 2. Prevalence and severity of dental caries in the permanent dentition and deciduous dentition in children and adolescents aged 6-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 Table 2. Prevalence and severity of dental caries in the permanent dentition and deciduous dentition in children and adolescents aged 6-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 Table 2. Prevalence and severity of dental caries in the permanent dentition and deciduous dentition in children and adolescents aged 6-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 Age n CPO-D SD (CPO-D) Healthy (%) – * Healthy (%) – Permanent Dentition Deciduous Dentition 7 173 0.31 0.69 80.35 29.48 8 265 0.61 1.00 67.55 28.30 9 239 0.85 1.33 59.00 35.00 10 290 0.89 1.27 56.48 55.25 11 253 1.23 1.54 49.01 73.52 12 143 1.27 1.79 49.65 86.01 13 76 1.80 2.22 43.42 94.74 Total 1,439 0.91 1.41 59.02 52.20 Mantel-Haenszel’s linear association test (χ2 p<0.001) Table 3. Distribution of factorial changes in the food consumption patterns of children and adolescents aged 6-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 Table 3. ▄ ▄RESULTS Distribution of factorial changes in the food consumption patterns of children and adolescents aged 6-13 years and enrolled in the public schools of the municipality of São Francisco do Conde, Bahia state, 2010 p y Foods and groups of foods Obesogenic dietary pattern Healthy dietary pattern h2 Sugars and sweets 0.698 - 0.503 Artificial drinks (juices) 0.484 - 0.234 Sodas 0.445 0.226 Typical preparations 0.601 - 0.396 Pastry 0.603 - 0.366 Fast food 0.740 0.554 Oil and fats 0.568 - 0.436 Milk and dairy products 0.615 - 0.434 Meats - 0.483 0.240 Cereals and derivatives (rice, cakes, pastas, flour, cassava) 0.548 - 0.441 Roots and tubercles - 0.653 0.464 Legumes - 0.793 0.629 Fruits 0.493 0.438 Leafy vegetables 0.701 0.502 % Variance explained 28.62 17.08 % Variance cumulative 45.70 ▄ ▄DISCUSSION Association between dietary patterns and caries experiences in children and adolescents aged 6-13 years and enrolled in the public schools of São Francisco do Conde, Bahia state, 2011 y p p g y p schools of São Francisco do Conde, Bahia state, 2011 Variables Univariate Model (Crude) Model (adjusted) PR (CI 95%) p PR (CI 95%)* p Obesogenic dietary pattern Percentile <75% 1 0.057 1 Percentile ≥75% 1.31(0.99-1.74) 1.41(1.04-1.96) 0.035 Prudent dietary pattern Percentile <75% 0.91(0.69-1.20) 0.527 0.88(0.64-1.20) 0.439 Percentile ≥75% 1 1 Anthropometric status Not overweight 0.74(0.53-1.03) 0.075 Overweight 1 Age <10 years 1.83(1.43-2.36) 0.000 ≥10 years 1 Education of caregiver ≤ 4th grade 0.96(0.73-1.26) 0.784 >4th grade 1 Per capita income <1 minimum salary (BMS) 0.84(0.60-1.18) 0.320 ≥1 minimum salary (BMS) 1 Age of caregiver <30 years 1.76(1.30-2.38) 0.001 ≥30 years 1 Place of residence Urban 0.75(0.58-0.97) 0.034 Rural 1 Estimates adjusted for anthropometric characteristics, such as student age, age of caregiver, per capita income, education of caregiver, household location, and energy intake Estimates adjusted for anthropometric characteristics, such as student age, age of caregiver, per capita income, education of caregiver, household location, and energy intake Estimates adjusted for anthropometric characteristics, such as student age, age of caregiver, per capita income, education of ca energy intake opometric characteristics, such as student age, age of caregiver, per capita income, education of caregiver, household location, a Estimates adjusted for anthropometric characteristics, such as student age, age of caregiver, per capita income, education of caregiver, household location, and energy intake factors. It is generally considered that the risk of dental caries is higher if there is high frequency of sugar consumption and the sugar forms can be retained in the mouth for long periods22. Sucrose is the most cariogenic sugar, as it produces extracellular polysaccharides that facilitate bacterial adhesion to the teeth and limit the diffusion of acid and buffer onto the plaque. Therefore, the role of sucrose in the risk of dental caries is well known. Although high energy density foods such as snacks (fast foods, chips) are not sweet, they are potentially cariogenic due to their starch content. Starch has a very high cariogenic potential, and its hydrolysis produces fermentative sugars that promote the formation of an acidogenic plaque, which favors the development of dental caries. Consequently, snacks have become an important risk factor for the development of dental caries in children and adolescents24. ▄ ▄DISCUSSION year 2000, which included a mean DMFT index value equal to or smaller than 3 in children up to 12 years old was achieved in São Francisco do Conde. Furthermore, there was an increase in dental caries prevalence with age, and this result not only highlights the cumulative prevalence of dental caries and the severity indicators used in dentistry, but also indicates the inability of the health services to control this dental disease in the population through its range of policies, without adopting a high-risk strategy9 . In this study, considering both dentitions, the prevalence of dental caries was 82.14% and 37.3% in the permanent and deciduous dentitions, respectively, and the mean DMFT index was 0.94 (±1.55). Although the DMFT index value was acceptable, the high percentage of students with untreated cavities in this study is disturbing, and it indicates the need for greater access to dental services. The DMFT index value of dental caries in the 12-year-olds of the municipality of SFC (1.27) was lower than the national average of dental caries in the last survey conducted in 2003, in which the DMFT index was 2.10, and close to the value of the capital of the state, Salvador10. Additionally, the prevalence of dental caries in the municipality of SFC was lower than that of the northeast region, in which the mean DMFT index was 2.6310, and those of many other representative epidemiological surveys conducted in different Brazilian regions19. The data demonstrate that the World Health Organization22 goals for the This study sought to evaluate the influence of dietary patterns on dental caries activity in the context of a multifactorial disease. The following two food consumption patterns were identified among the children and adolescents in São Francisco do Conde: “prudent” and “obesogenic”. In this study, an association was found between the obesogenic dietary pattern and dental caries. The proportion of students with dental caries was 1.4 times higher among those who more frequently consumed foods Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 202 Obesogenic dietary pattern and dental caries Table 4. Association between dietary patterns and caries experiences in children and adolescents aged 6-13 years and enrolled in the public schools of São Francisco do Conde, Bahia state, 2011 Table 4. ▄ ▄REFERENCES 10. Brasil. Ministério da Saúde. Projeto SB Brasil 2010: nota para a imprensa. Brasília: Ministério da Saúde; 2010. 1. Feferbaum R, Abreu LC, Leone C. Fluid intake patterns: an epidemiological study among children and adolescents in Brazil. BMC Public Health. 2012;12(1):1005. http://dx.doi.org/10.1186/1471-2458-12-1005. PMid:23167254. 11. World Health Organization, Food and Agriculture Organization of the United Nations – FAO. Preparation and use of food-based dietary guidelines. Geneva: WHO; 1998. WHO Technical Report Series, 880. 2. Catteau C, Trentesaux T, Delfosse C, Rousset MM. Consumption of fruit juices and fruit drinks: impact on the health of children and teenagers, the dentist’s point of view. Arch Pediatr. 2012;19(2):118-24. http://dx.doi. org/10.1016/j.arcped.2011.11.011. PMid:22206891. 12. Perera I, Ekanayake L. Relationship between dietary patterns and dental caries in Sri Lankan adolescents. Oral Health Prev Dent. 2010; 8(2):165-72. PMid:20589251. 3. Bawadi HA, Khader YS, Haroun TF, Al-Omari M, Tayyem RF. The association between periodontal disease, physical activity and healthy diet among adults in Jordan. J Periodontal Res. 2011;46(1):74-81. http://dx.doi. org/10.1111/j.1600-0765.2010.01314.x. PMid:20860591. 13. World Health Organization. Oral health surveys. Basic methods. Geneva: WHO; 1997. 14. Slater B, Philippi ST, Fisberg RM, Latorre MR. Validation of a semi- quantitative adolescent food frequency questionnaire applied at a public school in São Paulo, Brazil. Eur J Clin Nutr. 2003;57(5):629-35. http:// dx.doi.org/10.1038/sj.ejcn.1601588. PMid:12771963. 4. Burt BA, Kolker JL, Sandretto AM, Yuan Y, Sohn W, Ismail AI. Dietary patterns related to caries in a low-income adult population. Caries Res. 2006; 40(6):473-80. http://dx.doi.org/10.1159/000095645. PMid:17063017. 15. Touger-Decker R, van Loveren C. Sugars and dental caries. Am J Clin Nutr. 2003;78(4):881S-92S. PMid:14522753. 5. Kolker JL, Yuan Y, Burt BA, Sandretto AM, Sohn W, Lang SW, et al. Dental caries and dietary patterns in low-income African American children. Pediatr Dent. 2007;29(6):457-64. PMid:18254414. 16. Piovesan C, Antunes JL, Guedes RS, Ardenghi TM. Influence of self- perceived oral health and socioeconomic predictors on the utilization of dental care services by schoolchildren. Braz Oral Res. 2011;25(2):143-9. http://dx.doi.org/10.1590/S1806-83242011005000004. PMid:21359493. 6. Jaghasi I, Hatahet W, Dashash M. Dietary patterns and oral health in schoolchildren from Damascus, Syrian Arab Republic. East Mediterr Health J. 2012;18(4):358-64. PMid:22768698. 17. Voci SM, Slater B, Silva MV, Marchioni DML, Latorre MRDO. Calibration study of the Food Frequency Questionnaire for Adolescents (AFFQ). Cien Saude Colet. 2011;16(4):2335-43. http://dx.doi.org/10.1590/S1413- 81232011000400033. PMid:21584475. 7. Levine RS, Nugent ZJ, Rudolf MC, Sahota P. Dietary patterns, toothbrushing habits and caries experience of schoolchildren in West Yorkshire, England. Community Dent Health. 2007;24(2):82-7. PMid:17615822. 8. ▄ ▄CONCLUSION A positive association (PR = 1.40, 95% CI: 1.04-1.96) was found between obesogenic dietary patterns and the development of dental caries. Therefore, the promotion of healthy eating habits and nutritional habits should be encouraged to prevent dental caries in children and adolescents. It is recommended that municipal oral health policy should include essential measures, such as monitoring of oral diseases, as well as programs and services based on the principles of fairness, identifying potential interventions to minimize inequalities in oral health and their effects. Considering the vulnerability of school children to the adoption of poor eating habits, schools should be designed as privileged environments for the promotion of oral health in young people. ▄ ▄DISCUSSION Therefore, this study could not infer causal relationships because this type of study does not consider the temporal sequence between exposure and effect. However, the results of this study are in agreement with the biological plausibility and the results reported in prospective studies which support the theory that obesogenic dietary patterns are associated with dental caries25 . Some of the limitations of the use of factor analysis to identify food consumption patterns include the arbitrary method of extracting the number of factors retained and the method of rotation of the correlation matrix involved in the factor analysis. However, these decisions were made taking into account the objectives of the research and the interpretability of the data, as recommended by Hearty and Gibney21. Furthermore, the oral examinations were conducted in the school environment under natural light with the aid of formation3,25. These factors indicate a variety of social problems that have not been addressed, and they thereby demonstrate the inequalities in oral health in many poor regions of the country25. The oral examinations did not include radiographic examination. As a result, in particular, the interproximal carious lesions could not be diagnosed, which may have underestimated the prevalence of dental caries. equa t es o a ea t a y poo eg o s o t e cou t y . It is important to note the limitations of this study and the fact that this was a cross-sectional study. Therefore, this study could not infer causal relationships because this type of study does not consider the temporal sequence between exposure and effect. However, the results of this study are in agreement with the biological plausibility and the results reported in prospective studies which support the theory that obesogenic dietary patterns are associated with dental caries25 . Some of the limitations of the use of factor analysis to identify food consumption patterns include the arbitrary method of extracting the number of factors retained and the method of rotation of the correlation matrix involved in the factor analysis. However, these decisions were made taking into account the objectives of the research and the interpretability of the data, as recommended by Hearty and Gibney21. Furthermore, the oral examinations were conducted in the school environment under natural light with the aid of a wooden spatula and gauze, according to the WHO criteria13. 10. Brasil. Ministério da Saúde. Projeto SB Brasil 2010: nota para a imprensa. Brasília: Ministério da Saúde; 2010. ▄ ▄DISCUSSION More recently, it has been documented that sugary drinks are an important risk factor for the development of dental caries in children1,22. In addition to an unhealthy dietary pattern, there are multiple factors that affect the prevalence of dental caries in populations, and these factors include access to dental products containing fluoride, access to a fluoridated water supply, access to oral health services, and the adoption of a proper brushing technique to control biofilm that constitute a standard obesogenic diet (PR = 1.40, 95% CI: 1.04‑1.96) compared with those who less frequently consumed these types of foods. These results held true even after adjusting for anthropometric status, such as the student’s age, maternal age, family income, maternal education, school location and energy intake. The findings support the hypothesis that obesogenic dietary patterns have the potential to affect oral health in this age group, even in areas where the subjects are exposed to fluoride. Only a few studies demonstrating the influence of dietary patterns on dental caries activity have been reported in the literature. Slater et al.14, after using the same type of analysis (Factorial Analysis) to investigate dietary patterns, concluded that the dietary pattern characterized by sweets and sugar products, breads and pastries emerged as a significant predictor of dental caries. A positive association between obesogenic dietary patterns and dental caries has also been reported5,6,23. These authors directly associated a particular standard diet based on extrinsic sugars with the emergence and progression of carious lesions. It is a fact that dental caries prevalence in this age group depends on a combination of factors, including the type and frequency of carbohydrate consumption, the microbial composition of the dental biofilm, and salivary Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 203 Rita de Cássia Ribeiro Silva, Luce Alves da Silva, Roberto Paulo Correia de Araújo, Felipe Fagundes Soares, Rosemeire Leovigildo Fiaccone, M ce Alves da Silva, Roberto Paulo Correia de Araújo, Felipe Fagundes Soares, Rosemeire Leovigildo Fiaccone, Maria Cristina Teixeira Cangussu formation3,25. These factors indicate a variety of social problems that have not been addressed, and they thereby demonstrate the inequalities in oral health in many poor regions of the country25. It is important to note the limitations of this study and the fact that this was a cross-sectional study. ▄ ▄REFERENCES Touger-Decker R, Mobley CC, American Dietetic Association. Position of the American Dietetic Association: oral health and nutrition. J Am Diet Assoc. 2007;107(8):1418-28. http://dx.doi.org/10.1016/j.jada.2007.06.003. PMid:17712930. 18. Neumann AICP, Shirassu MM, Fisberg RM. Consumo de alimentos de risco e proteção para doenças cardiovasculares entre funcionários públicos. Rev Nutr. 2006;19(1):19-28. http://dx.doi.org/10.1590/S1415-52732006000100002. 19. Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ. 2007;85(9):660-7. http://dx.doi. 9. Narvai PC. Cárie dentária e flúor: uma relação do século XX. Cien Saude Colet. 2000;5:381-92. 204 Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 Obesogenic dietary pattern and dental caries org/10.2471/BLT.07.043497. PMid:18026621. org/10.2471/BLT.07.043497. PMid:18026621. privadas de Salvador, Bahia, Brasil, em 2001. Pesqui Odontol Bras. 2002;16(4):379-84. http://dx.doi.org/10.1590/S1517-74912002000400017. PMid:12612780. privadas de Salvador, Bahia, Brasil, em 2001. Pesqui Odontol Bras. 2002;16(4):379-84. http://dx.doi.org/10.1590/S1517-74912002000400017. PMid:12612780. 20. Newby PK, Tucker KL. Empirically derived eating patterns using factor or cluster analysis: a review. Nutr Rev. 2004;62(5):177-203. http://dx.doi. org/10.1111/j.1753-4887.2004.tb00040.x. PMid:15212319. 24. Sohn W, Burt BA, Sowers MR. Carbonated soft drinks and dental caries in the primary dentition. J Dent Res. 2006;85(3):262-6. http://dx.doi. org/10.1177/154405910608500311. PMid:16498075. 21. Hearty AP, Gibney MJ. Comparison of cluster and principal component analysis techniques to derive dietary patterns in Irish adults. Br J Nutr. 2009;101(4):598-608. http://dx.doi.org/10.1017/S0007114508014128. PMid:18577300. 25. Zaborskis A, Milciuviene S, Narbutaite J, Bendoraitiene E, Kavaliauskiene A. Caries experience and oral health behaviour among 11 - 13-year- olds: an ecological study of data from 27 European countries, Israel, Canada and USA. Community Dent Health. 2010;27(2):102-8. PMid:20648887. 25. Zaborskis A, Milciuviene S, Narbutaite J, Bendoraitiene E, Kavaliauskiene A. Caries experience and oral health behaviour among 11 - 13-year- olds: an ecological study of data from 27 European countries, Israel, Canada and USA. Community Dent Health. 2010;27(2):102-8. PMid:20648887. 22. Fédération Dentaire Internationale, World Health Organization. Global goals for oral health in the year 2000. Fédération Dentaire Internationale. Int Dent J. 1982;32(1):74-7. PMid:6951815. 23. Cangussu MCT, Castellanos RA, Pinheiro MF, Albuquerque SR, Pinho C. Cárie dentária em escolares de 12 e 15 anos de escolas públicas e Received on: Mar. 06, 2015 Accepted on: June 15, 2015 Cad. Saúde Colet., 2015, Rio de Janeiro, 23 (2): 198-205 205
https://openalex.org/W4390778348	https://journalofbigdata.springeropen.com/counter/pdf/10.1186/s40537-023-00867-5	English	null	Prediction of flight departure delays caused by weather conditions adopting data-driven approaches	Journal of big data	2,024	cc-by	12,176	© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the mate- rial. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creativecommons.org/licenses/by/4.0/. RESEARCH Open Access Seongeun Kim1,2 and Eunil Park3,4* Seongeun Kim1,2 and Eunil Park3,4* Abstract In this study, we utilize data-driven approaches to predict flight departure delays. The growing demand for air travel is outpacing the capacity and infrastructure avail- able to support it. In addition, abnormal weather patterns caused by climate change contribute to the frequent occurrence of flight delays. In light of the extensive net- work of international flights covering vast distances across continents and oceans, the importance of forecasting flight delays over extended time periods becomes increasingly evident. Existing research has predominantly concentrated on short-term predictions, prompting our study to specifically address this aspect. We collected datasets spanning over 10 years from three different airports such as ICN airport in South Korea, JFK and MDW airport in the United States, capturing flight informa- tion at six different time intervals (2, 4, 8, 16, 24, and 48 h) prior to flight departure. The datasets comprise 1,569,879 instances for ICN, 773,347 for JFK, and 404,507 for MDW, respectively. We employed a range of machine learning and deep learning approaches, including Decision Tree, Random Forest, Support Vector Machine, K-nearest neighbors, Logistic Regression, Extreme Gradient Boosting, and Long Short-Term Memory, to pre- dict flight delays. Our models achieved accuracy rates of 0.749 for ICN airport, 0.852 for JFK airport, and 0.785 for MDW airport in 2-h predictions. Furthermore, for 48-h predictions, our models achieved accuracy rates of 0.748 for ICN airport, 0.846 for JFK airport, and 0.772 for MDW airport based on our experimental results. Consequently, we have successfully validated the accuracy of flight delay predictions for longer time frames. The implications and future research directions derived from these findings are also discussed. 1 Department of Semiconductor and Display Engineering, Sungkyunkwan University, Seoul 03063, Republic of Korea 2 Samsung Electronics, Gyeonggi, Republic of Korea 3 Department of Interaction Science, Sungkyunkwan University, 25‑2 Sungkyunkwan‑ro, Jongno‑gu, Seoul 03063, Republic of Korea 4 Teach Company, 25‑2 Sungkyunkwan‑ro, Jongno‑gu, Seoul 03063, Republic of Korea Keywords: Flight delay, Delay prediction weather, Machine learning, LSTM Correspondence: eunilpark@skku.edu 1 Department of Semiconductor and Display Engineering, Sungkyunkwan University, Seoul 03063, Republic of Korea 2 Samsung Electronics, Gyeonggi, Republic of Korea 3 Department of Interaction Science, Sungkyunkwan University, 25‑2 Sungkyunkwan‑ro, Jongno‑gu, Seoul 03063, Republic of Korea 4 Teach Company, 25‑2 Sungkyunkwan‑ro, Jongno‑gu, Seoul 03063, Republic of Korea Kim and Park Journal of Big Data (2024) 11:11 https://doi.org/10.1186/s40537-023-00867-5 Kim and Park Journal of Big Data (2024) 11:11 https://doi.org/10.1186/s40537-023-00867-5 Journal of Big Data Prediction of flight departure delays caused by weather conditions adopting data‑driven approaches Seongeun Kim1,2 and Eunil Park3,4 Introduction With the increasing demand for air travel, the number of air passengers has significantly increased. The global air passenger transport market doubles every 15 years [1]. For example, as of February 2023, the revenue passenger kilometer in Asia Pacific and North America has increased by 105.4% and 25.1% relative to that in 2022, respectively. Despite a temporary decline in passenger traffic during the Covid-19 pandemic, the number of air passengers has steadily increased over the past few decades [2]. Kim and Park Journal of Big Data (2024) 11:11 Page 2 of 25 Table 1 Number of eligible passengers for compensation versus the number of total passengers Year Number of passenger in Europe (millions) Rate of compensation (%) Number of eligible passengers (millions) Evolution (2017/2016) (%) 2017 1020 1.58 16 14.20 2016 970 1.46 14 15.60 2015 918 1.33 12 16.10 Table 2 Different types of delays Bold valuesindicate the greatest results Delay type Portion (%) Mean and standard deviation of delay time (min) CarrierDelay 35.30 42.08 (64.98) WeatherDelay 3.86 69.81 (100.79) NASDelay 38.18 31.40 (41.93) SecurityDelay 0.40 30.40 (40.68) LateAircraftDelay 22.27 51.27 (58.03) Meeting the increasing demand for air travel and ensuring efficient supply chain operations require the development of aviation infrastructure. This includes expanding airport facilities, updating airline fleets, and implementing effective air schedule man- agement. Addressing these issues is crucial to provide a seamless and reliable travel experience for passengers. However, a significant challenge in delivering satisfactory ser- vices is the frequent occurrence of unexpected flight delays and cancellations [3]. According to Tileagă and Oprisan [4], the number of compensation cases due to delayed flight schedules is increasing steadily. Table 1 shows that the number of com- pensation recipients for air delays and cancellations is steeply increasing every year. Flight delays have significant economic consequences for both airlines and passengers, rendering it a notable issue within the aviation industry. Table 2 show the types and proportion of delays from 2010 to 2021 at the John F. Ken- nedy International Airport (JFK). It reveals that weather-related delays account for a small proportion of delays (3.86%). However, weather-related delays were longer than other types of delays, with an average delay time of 69.81 min and a standard deviation of 100.79 min [5]. The frequency of abnormal weather phenomena that are known to contribute to an increase in flight delays [6] is on the rise worldwide. Introduction In addition, the regional climate determined by geographical location plays a significant role in flight operations [7]. For example, in South Korea, the total rainfall period is concentrated from July to Septem- ber each year, with approximately 42.5% rainfall in July, 27.4% in August, and 12.8% in September. In addition, the region is directly affected by typhoons at the end of August through early September every year. While previous studies on flight delay prediction have often incorporated weather information [8–10], the majority of these studies have centered around predicting delays within relatively short timeframes, typically within thresholds of 15 min or up to 4 h, primarily tailored to airline services. However, the unique context of international flights Kim and Park Journal of Big Data (2024) 11:11 Page 3 of 25 covering vast distances across continents and oceans, with flight durations spanning from as little as 10 h to as long as 20 h, underscores the necessity for delay prediction over more extended timeframes. Therefore, this study aims to predict flight delays over more extended timeframes (2 to 48 h) based on weather data. We focus on three well-known international airports: Incheon International Airport in South Korea (ICN), John F. Kennedy International Air- port (JFK), and Chicago Midway International Airport (MDW) in the United States. In addition, we use weather information from the meteorological agencies located at each airport. “Background and related work” section reviews previous research in this area, whereas “Methodology” section presents the machine learning and deep learning mod- els along with the evaluation methods utilized in the study. The experimental procedures and the comparison of the results across the models are presented in “Implementation and result” section. “Discussion and concluding remarks” section concludes this paper by presenting the interpretation of the results, noteworthy findings, limitations, and sug- gestions for future research. Background and related work Several studies have been conducted to forecast flight departure delays using various statistical methods, machine learning, and deep learning techniques. Table 3 provides a summary of prior flight delay detection research based on machine learning and neural network approaches. Researchers [9, 11, 12] have utilized Bayesian modeling, clustering, classification, and regression with diverse datasets from different regions. The time span of the data varied, ranging from 1 month to 5 years, and the airports under investigation differed as well. Khaksar and Sheikholeslami [9] identified parameters that enable effective estimation of delays. They used Bayesian modeling, decision tree, cluster classification, random forest, and hybrid methods. They used 2,825,647 data for US airlines and 15,428 data for Ira- nian airlines. They realized an accuracy of approximately 70%. h Al-Tabbakh et al. [11] analyzed the flight delay patterns using four decision tree clas- sifiers, including Decisionstump, J48, Random Forest, and REPTree. They utilized 512 data from a brief duration of 1 month, i.e., January 2018. The findings revealed that among the classifiers evaluated for the Egypt Airline dataset, REPTree attained the high- est accuracy score of 80.3%. Ye et al. and Atlioğlu et al. [12, 13] conducted flight delay prediction via supervised learning methods, whereas [12] employed multiple linear regression, a support vec- tor machine, extremely randomized trees, and LightGBM. They used 105,993 data and reported the highest accuracy of 86.53%. Atlioğlu et al. [13] studied 11 machine learning models using data obtained follow- ing feature selection and transformation. They used 8086 data and achieved F1-scores of approximately 81%. Certain researchers predict airline delay using neural networks and hybrid models [8, 10, 14]. Kim et al. [8] investigated the effectiveness of deep learning models in predict- ing air traffic delays. Daily sequences of departure and arrival flight delays for individual airports were modeled using the long short-term memory (LSTM) and recurrent neural network (RNN) architecture. The accuracy of RNN improves with deeper architectures, Page 4 of 25 Kim and Park Journal of Big Data (2024) 11:11 Table 3 Summary of prior prediction of flight delay Sources Method Datasets Data period Delay time (min) Results Machine learning Khaksar and Sheikholeslami [9] Bayesian modeling, decision tree, cluster classification, random forest, and hybrid method US and Iranian airline US: 6 months, Iran: 16 months 0–15, 15–60, 60+ Accuracy more than 70% Al-Tabbakh et al. Background and related work [11] Decision tree, random forest, and REPTree Egypt airline Jan 2018 (1 month) – Accuracy around 80.3% Ye et al. [12] Multiple linear regression, support vector machine, extremely randomized trees, and LightGBM Nanjing Lukou airline Mar 1st 2017 to Feb 28th 2018 15+ Accuracy of 86.53% Atlioglu et al. [13] 11 machine learning models. CART, KNN, GBM, XGB, and LGBM Dammam King Fahd Inter- national Airport Jan 1st 2017 to Dec 9th 2019 15+ Accuracy around 82% Neural network Kim et al. [8] LSTM, RNN ATL, LAX, ORD, DFW, DEN, JFK, SFO, CLT, LAS, PHX Jan 2010–Aug 2015 15+, 30+ Accuracy of 90.95% Qu et al. [10] CBAM-CondenseNet and SimAM-CNN-MLSTM The Civil Aviation Admin- istration of the China East China Regional Administra- tion (ECRA) Mar 2018–May 2019 15–60, 60–120, 120–240, 240+ Accuracy of 89.8%, 91.36% Yazdi et al. [14] Stack denoising autoen- coder- levenberg marquart model, SAE-LM, SDA The Bureau of Transporta- tion Statistics of United State Department of Trans- portation For 5 years 15+ Accuracy of 96%, 86%, 89% Kim and Park Journal of Big Data (2024) 11:11 Page 5 of 25 exhibiting the highest performance with an accuracy of 90.95% on the Atlanta air traffic data. exhibiting the highest performance with an accuracy of 90.95% on the Atlanta air traffic data. Qu et al. [10] analyzed and predicted flight delays using a convolutional neural net- work (CNN) and RNN models that are well-suited for classification problems in the field of deep learning. They improved the CondenseNet network by incorporating CBAM modules within the CNN-based CondenseNet algorithm to develop CBAMCon- denseNet. Additionally, they constructed a CNN-MLSTM network based on the CNN model and injected the SimAM module to enhance the attention of flight chain data. They used 36,287 data of China and achieved the highest accuracy score of 91.36%. h Yazdi et al. [14] designed the proposed model to output optimized results by incorpo- rating a technique based on stack denoising autoencoder to account for the noisy flight delay data. They constructed SAE-LM based on an autoencoder and LM algorithm. The stacked denoising autoencoder is based on only denoising autoencoder. They utilized a comprehensive dataset spanning 5 years of US flight operations, comprising a total of 3,601,679 data points. The results demonstrated that the proposed model exhibited enhanced accuracy compared with the RNN model, highlighting its effectiveness in pre- dicting flight delays. Background and related work While numerous researchers have utilized state-of-the-art machine learning and deep learning techniques to study weather-related takeoff delays from vari- ous angles, the majority of studies have focused on predicting delays within a time cri- terion of approximately 15 min. There has been limited exploration and prediction of flight delays exceeding 2 h. l By employing established research methodologies, it is feasible to aggregate the out- comes of short-term predictions to generate long-term forecasts. Nevertheless, it’s vital to recognize that repeated predictions may introduce inaccuracies. When assessing the practical utility of such models, the ability to predict aviation delays over extended time intervals based on input data widens the scope of possibilities for long-haul flights and diverse flight schedules. This expanded capability offers benefits not only from the per- spective of airport resource management but also in various other aspects. Therefore, there is a pressing need for research that focuses on machine learning and neural net- work models capable of forecasting the distant future using authentic long-term differ- ential data. Hence, in this study, our objective is to specifically address and forecast flight delays of more than 2 h. Classification models • Support Vector Machine (SVM): SVM is a powerful supervised learning model that can be used for various tasks such as classification, regression, and anomaly detec- tion. It aims to find a decision boundary that maximizes the separation between two classes while satisfying certain conditions. SVM can handle both linear and non-lin- ear classification problems by using different kernel functions [21]. It determines the side of the decision boundary to which a data point belongs, allowing it to effectively classify data. Although it may be slower and less interpretable due to the require- ment for multiple combination tests, it offers the advantage of being applicable to both categorical and numerical prediction problems, with minimal vulnerability to outlier data. Additionally, it is less susceptible to overfitting and more user-friendly compared to neural networks [22, 23]. • Support Vector Machine (SVM): SVM is a powerful supervised learning model that can be used for various tasks such as classification, regression, and anomaly detec- tion. It aims to find a decision boundary that maximizes the separation between two classes while satisfying certain conditions. SVM can handle both linear and non-lin- ear classification problems by using different kernel functions [21]. It determines the side of the decision boundary to which a data point belongs, allowing it to effectively classify data. Although it may be slower and less interpretable due to the require- ment for multiple combination tests, it offers the advantage of being applicable to both categorical and numerical prediction problems, with minimal vulnerability to outlier data. Additionally, it is less susceptible to overfitting and more user-friendly compared to neural networks [22, 23]. • K-Nearest Neighbors (KNN): KNN is a classification algorithm that operates based on the principle of similarity. It assigns a class label to a given data point by consider- ing the labels of its “k” nearest neighbors in the feature space. The distance between data points is typically calculated using the Euclidean distance measurement method [24]. It offers several advantages, such as high accuracy and the ability to exclude outlier data from consideration by using only the top k closest data points. Further- more, it does not rely on assumptions about the data since it is based on existing data. Classification models We used the following machine learning models and LSTM neural network to predict flight takeoff delays. The LSTM model boasts the advantage of effectively managing time-series data, but it comes with the drawback of requiring considerably more com- plex and powerful hardware. From this standpoint, machine learning (ML) models allow predictions at the individual time-unit level and are notably more computationally effi- cient when compared to the LSTM model. • Decision Tree (DT): DT is a type of supervised learning model that classifies or regresses data by applying a set of classification rules. The resulting model has a tree- like structure, hence the name ‘Decision Tree.’ Pruning techniques can be applied • Decision Tree (DT): DT is a type of supervised learning model that classifies or regresses data by applying a set of classification rules. The resulting model has a tree- like structure, hence the name ‘Decision Tree.’ Pruning techniques can be applied Kim and Park Journal of Big Data (2024) 11:11 Page 6 of 25 to enhance the model’s generalization performance and prevent overfitting, ensuring that it performs effectively on unknown data. Grid search can be used to find the optimal parameter values for the DT model, optimizing its performance [15]. It does not necessitate data preprocessing, such as normalization or handling missing values and outliers. It also has the capability to simultaneously handle both numerical and categorical variables. However, it has the limitation of considering only one variable at a time, which can make it challenging to capture interactions between variables. Moreover, the shape of the resulting decision tree can exhibit significant variations with minor differences in the data [16, 17]. f • Random Forest (RF): RF is an ensemble algorithm that trains multiple DT models and combines their results to make predictions. The method entails the random selection of a subset of features from the complete feature set to build one decision tree, followed by the selection of another random feature subset to create additional decision trees. Multiple decision trees are generated using this process. The final prediction is made by choosing the most frequently occurring prediction from these multiple decision trees [18]. This approach is versatile as it can be applied to both classification and regression problems. It is particularly effective in handling large- scale data and mitigates the issue of overfitting by reducing model noise, ultimately improving model accuracy [19, 20]. Classification models However, it has the disadvantage of increased processing time as the dataset size grows, as it needs to compare with all existing data points, and it may require significant memory usage for large datasets [22, 25]. i • Logistic Regression (LR): LR is one of the simplest classification models. It predicts the probability of data belonging to a certain category as a value between 0 and 1 and classifies it into the category with a higher probability [26]. It has the advantage of being less complex and faster due to linear combinations, making it easy to interpret the results. However, it may suffer a reduction in learning ability when dealing with Kim and Park Journal of Big Data (2024) 11:11 Page 7 of 25 non-linear relationships and can be sensitive to outliers and anomalies, which are its disadvantages [27, 28]. • Extreme Gradient Boosting (XGB): XGB is an algorithm implemented using the boosting technique. It supports both regression and classification problems and exhibits suitable performance and resource efficiency. It is characterized by strong durability with its built-in overfitting regularization function [29, 30]. i • Long Short-Term Memory (LSTM): LSTM networks are a type of RNN that can learn the order dependence in sequence prediction problems. RNNs are modified by adding a memory cell that can store information for an extended period. LSTM was proposed as a solution to address the issue of vanishing gradients in RNN when processing long sequential data [31]. However, it has the drawback of being compu- tationally intensive and having a complex model structure due to the incorporation of forget gates, input gates, and output gates [32–34]. i • Long Short-Term Memory (LSTM): LSTM networks are a type of RNN that can learn the order dependence in sequence prediction problems. RNNs are modified by adding a memory cell that can store information for an extended period. LSTM was proposed as a solution to address the issue of vanishing gradients in RNN when processing long sequential data [31]. However, it has the drawback of being compu- tationally intensive and having a complex model structure due to the incorporation of forget gates, input gates, and output gates [32–34]. Evaluation methods To evaluate the performance of each classifier, we calculated the confusion matrix and measured the accuracy, precision, recall, and F-score. Table 4 is the confusion matrix, a 2 × 2 matrix representation of classification results. The number of correctly classified instances is the sum of the diagonals of the matrix, while all other instances are incorrectly classified. Each item in the confusion matrix includes the following four indicators. The first indicator is True Positive (TP), which signifies that the predicted value is positive when the actual value is positive. The second indicator is True Negative (TN), indicating that the predicted value is negative when the actual value is negative. The third indicator is False Positive (FP), denoting that the predicted value is positive when the actual value is negative. Lastly, the fourth indicator is False Negative (FN), showing that the predicted value is negative when the actual value is positive [35]. Accuracy serves as “a metric for assessing the overall performance of each model by computing the ratio of correctly classified samples to the total number of samples” [36]. However, in situations with a significant imbalance between positive and negative samples, accuracy may not provide a suitable evaluation measure. Precision presents “the proportion of true positive cases among all predicted positive cases” [37], while recall computes “the ratio of correctly predicted positive samples to the total number of true positive samples” [38]. F1-score represents “a balanced measure that combines both precision and recall” [39]. Table 4 Confusion matrix Classified as delayed Classified as not delayed Actual delayed True positive (TP) False negative (FN) Actual not delayed False positive (FP) True negative (TN) Kim and Park Journal of Big Data (2024) 11:11 Page 8 of 25 Implementation and result Data description and analysis We collected three datasets including flight and weather information of Incheon Inter- national Airport in South Korea (ICN) [40], John F. Kennedy International Airport (JFK) [41], and Chicago Midway International Airport (MDW) [42] in the United States. The flight information [43, 44] is organized by all flight-related features, including scheduled departure time, actual departure time, and delay type. The weather infor- mation is the officially introduced regional weather feature. The flight information scheduled from 2010 to 2021 was examined, spanning a total of 11 years. The weather information corresponding to the same period was also collected. For the experiment, weather and flight information were merged with a time difference for data preproc- essing to predict flights based on weather conditions. The merged datasets include the attributes listed in Tables 5 and 6. Among these attributes, the airline, flight number, and destination were not used in the actual model training. Additionally, since the features wind direction (e.g., NW, WNW) and condition (e.g., Cloudy, Windy) are categorical data, they were transformed into one-hot encoding before being included in the training dataset. ICN dataset In situations where the scheduled departure time differs by more than 1 h, we classify the data as delayed. The ICN dataset comprises 1,562,029 instances of normal flights and Table 5 Incheon International Airport’s attributes list Attribute name Description Mean (Std) Min Max Time (year) 2010–2021 (e.g. 2020) – – – Airline Unique carrier [e.g. KE (Korean Air)] – – – Flight number Flight number (e.g. KE831) – – – Destination Destination (e.g. Asiana Airlines) – – – Planned departure time Planned departure time (e.g. 10:30) – – – Actual departure time Actual departure time (e.g. 11:30) – – – Result status Takeoff intime or delay status (e.g. cancel- lation) – – – Delay type Delay type (e.g. weather-snow) – – – Wind direction (deg) Wind direction (deg) (e.g. 10) 204.3 (109.7) 0 360 Wind velocity (KT) Wind velocity (KT) (e.g. 5) 7.4 (4.4) 0 49 Meteorological range Visible distance (e.g. 10,000) 8311 (2699.4) 0 11000 Cloud cover Cloud cover (e.g. 3) 2.9 (1.8) 1 9 Cloud form Cloud form (e.g. 5) – – – The height of the cloud ceiling (FT) The height of the cloud ceiling (FT) (e.g. 15,000) 7390.2 (6950.7) 0 24000 Temperature (celcius) Temperature (celcius) (e.g. − 7) 12.5 (10.49) − 17.2 36.6 Dew point temperature (celcius) Dew point temperature (celcius) (e.g. − 18.8) 5.6 (11.4) − 28.4 26.5 Sea-level pressure (hPA) The pressure of the atmosphere at the sea level (e.g. 1023.8) 1016.6 (8.4) 981.3 1040.4 Station pressure (hPa) Station pressure (hPa) (e.g. 1022.8) 1015.8 (8.4) 980.6 1039.5 Rainfall (mm) Rainfall (mm) (e.g. 0.3) 0.3 (1.4) 0 66 Page 9 of 25 Kim and Park Journal of Big Data (2024) 11:11 Table 6 John F. Kennedy International Airport, and Chicago Midway International Airport’s attributes list Attribute mame Description Mean (Std) Min Max Time (year) 2010–2021 (e.g. 2020) – – – Airline Unique carrier [e.g. AA (American Airlines)] – – – Flight number Flight number (e.g. AA2000) – – – Destination Destination (e.g. JFK) – – – Planned departure time Planned departure time (e.g. 1622) – – – Actual departure time Actual departure time (e.g. 1634) – – – Result status Takeoff intime or delay status (e.g. 1) – – – Delay type Delay type (e.g. WeatherDelay) – – – Wind direction Wind direction (e.g. NW, WNW) – – – Wind speed Wind speed (e.g. 3) 10.5 (5.3) 0 51 Wind gust Wind gust (e.g. ICN dataset 24) 5.3 (10.9) 0 75 Temperature (celcius) Temperature (celcius) (e.g. 34) 51.5 (20.5) − 21 103 Dew point temperature (celcius) Dew point temperature (celcius) (e.g. 31) 39.9 (19.5) − 32 79 Humidity Humidity (e.g. 92) 67.7 (17.2) 0 100 Pressure (hPa) Pressure (hPa) (e.g. 29.96) 29.3 (0.3) 0 30.2 Precipitation (mm) Precipitation (mm) (e.g. 0.1) 0.006 (0.046) 0 2 Condition Condition (e.g. Cloudy, Windy) – – – 7850 instances of delayed flights caused by weather conditions. To achieve a balanced distribution between normal and delayed cases, we randomly sampled an equal number of normal and delayed flight instances. To address the absence of certain features in the cases, we utilized a data interpolation method that was previously validated in a research study [45]. Due to the hourly-based nature of the ICN weather information, there were instances of missing features. To fill these gaps, we employed a linear interpolation tech- nique to estimate the values for the unmeasured time periods. The interpolated data comprises 953 data points, which accounts for 0.9% of the total 105,192 data points. Fur- thermore, we included flight takeoff results with time differences as additional features. To fulfill the objectives of the present study, we implemented a time difference criterion and utilized combined flight and weather cases. The time differences were categorized into intervals of 2, 4, 8, 16, 24, and 48 h. JFK dataset and MDW dataset Similar to the ICN dataset, we created delayed flight instances for the JFK and MDW datasets based on the time difference between the scheduled and actual departure times. The JFK dataset consisted of 763,930 normal cases and 9417 delayed cases attributed to weather conditions, while the MDW dataset comprised 398,945 normal cases and 5562 delayed flight instances. Similar to the approach followed for the ICN dataset, we con- ducted down-sampling procedures to achieve a 1:1 ratio of normal and delayed cases. In both the JFK and MDW datasets, the weather information consists of several cat- egorical features, such as wind direction and condition details. To incorporate these features into our data-driven approaches for machine learning and neural network frameworks, we employed a one-hot encoding technique. This encoding method allows us to represent the categorical variables as binary vectors, facilitating their utilization Kim and Park Journal of Big Data (2024) 11:11 Page 10 of 25 in the models. Additionally, we included flight takeoff results with time differences as one of the features in the dataset. Subsequently, both the JFK and MDW datasets with weather information were merged. Experiment Figure 1 shows the flow chart of our overall approach. For machine learning models, we input the data sampled following the process as mentioned above, while we stack the sampled data to create time-series data and input them to the LSTM model. To begin, we partitioned the dataset into subdata and testing subsets in an 80:20 ratio. Subsequently, we further divided the subdata into training and validation subsets in an 80:20 ratio, resulting in a distribution of the training, validation, and test datasets with a ratio of 67:13:20. Table 7 presents the number of datasets used for training, validation, and testing. Fig. 1 Flow charts for a machine learning, and b LSTM models Table 7 Summary of the employed datasets in training, validation, and test sessions Airport Train Validation Test Total ICN 10551 (67.2%) 2009 (12.8%) 3140 (20.0%) 15700 (100%) JFK 12756 (68.7%) 2041 (11.0%) 3767 (20.3%) 18564 (100%) MDW 7476 (68.6%) 1196 (11.0%) 2225 (20.4%) 10897 (100%) Fig. 1 Flow charts for a machine learning, and b LSTM models Fig. JFK dataset and MDW dataset 1 Flow charts for a machine learning, and b LSTM models Table 7 Summary of the employed datasets in training, validation, and test sessions Airport Train Validation Test Total ICN 10551 (67.2%) 2009 (12.8%) 3140 (20.0%) 15700 (100%) JFK 12756 (68.7%) 2041 (11.0%) 3767 (20.3%) 18564 (100%) MDW 7476 (68.6%) 1196 (11.0%) 2225 (20.4%) 10897 (100%) able 7 Summary of the employed datasets in training, validation, and test sessions in the models. Additionally, we included flight takeoff results with time differences as one of the features in the dataset. Subsequently, both the JFK and MDW datasets with weather information were merged. in the models. Additionally, we included flight takeoff results with time differences as one of the features in the dataset. Subsequently, both the JFK and MDW datasets with weather information were merged. in the models. Additionally, we included flight takeoff results with time differences as one of the features in the dataset. Subsequently, both the JFK and MDW datasets with weather information were merged. Experiment Figure 1 shows the flow chart of our overall approach. For machine learning models, we input the data sampled following the process as mentioned above, while we stack the sampled data to create time-series data and input them to the LSTM model. To begin, we partitioned the dataset into subdata and testing subsets in an 80:20 ratio. Subsequently, we further divided the subdata into training and validation subsets in an 80:20 ratio, resulting in a distribution of the training, validation, and test datasets with a ratio of 67:13:20. Table 7 presents the number of datasets used for training, validation, and testing. Page 11 of 25 Kim and Park Journal of Big Data (2024) 11:11 Table 8 Tested parameters in DT Parameters Value Max depth 2, 4, 6, 8, 10, 12, 14, 16, 18, 20 Min impurity decrease 0.0001, 0.0005, 0.001, 0.005, 0.01 Min samples split 2, 3, 4, 5 Min samples leaf 1, 2, 3, 4, 5 Table 9 Tested parameters in LSTM model Parameters Value Layer 1, 2, 3 Learning rate 0.0001, 0.0003, 0.0005, 0.001, 0.005 Epoch 300, 400, 500, 600, 700 Time series 2 h, 3 h, 4 h, 5 h, 6 h Table 8 Tested parameters in DT Table 9 Tested parameters in LSTM model Parameters Value Layer 1, 2, 3 Learning rate 0.0001, 0.0003, 0.0005, 0.001, 0.005 Epoch 300, 400, 500, 600, 700 Time series 2 h, 3 h, 4 h, 5 h, 6 h All experiments were conducted on a single GeForce RTX 3080 Ti 10GB GPU and implemented using Python 3.6 as the programming language. We performed a grid search to determine the optimal hyperparameters, including learning rates, number of epochs, number of layers, and number of stacked time-series data. We selected the most optimal parameters for the best performance. Tables 8 and 9 show the list of hyperpa- rameters for DT and LSTM used in the grid search. In the case of the LSTM model, the training parameters varied for each airport dataset. The ICN dataset had 2,385 param- eters, while the JFK and MDW datasets had 2,833 parameters. Flight delay prediction Tables 10, 11 and 12 show the prediction results of flight departure delays based on weather data using various models. The results were obtained corresponding to a total of six different time differences (2, 4, 8, 16, 24, and 48 h). ff Table 10 summarizes the results of the ICN dataset. The RF model reported the high- est accuracy score of 0.749 with a time difference of 2 h. Except for the DT model that showed the best recall performance of 0.700, the RF model displayed superior perfor- mance in other metrics. For the JFK airport dataset with a time difference of 2 h, the LSTM model achieved the highest accuracy score of 0.852 (Table 11). In terms of recall for predicting flight delays, the DT model outperformed all other models (0.826), whereas in terms of precision of prediction of on-time flights, the RF model outperformed all other models (0.835). Nonetheless, the LSTM model demonstrated superior performance in other evaluation metrics. The result corresponding to the MDW airport dataset for a time difference of 2 h is presented in Table 12. The LSTM model achieved the highest accuracy score of 0.785. Although the DT model exhibited the best performance in terms of recall (0.759), the LSTM model outperformed the other models in all other evaluation metrics. Flight delay prediction Kim and Park Journal of Big Data (2024) 11:11 Page 12 of 25 Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.678 0.691 0.675 0.683 0.095 0.318 0.687 0.702 0.678 0.690 0.091 0.318 Delayed 0.665 0.681 0.673 0.672 0.696 0.684 RF Normal 0.744 0.726 0.806 0.764 2.187 16.242 0.745 0.719 0.826 0.769 2.136 16.242 Delayed 0.768 0.679 0.721 0.782 0.659 0.715 SVM Normal 0.641 0.625 0.750 0.682 3.591 530.255 0.641 0.626 0.749 0.682 3.595 545.541 Delayed 0.666 0.525 0.587 0.666 0.528 0.589 KNN Normal 0.662 0.673 0.666 0.669 0.003 30.892 0.649 0.663 0.643 0.653 0.004 32.803 Delayed 0.651 0.658 0.655 0.635 0.656 0.645 LR Normal 0.598 0.612 0.591 0.601 0.095 0.318 0.525 0.535 0.574 0.554 0.069 0.637 Delayed 0.583 0.605 0.594 0.512 0.472 0.491 Algorithm Time difference: 2 h Time difference: 4 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.688 0.704 0.676 0.690 0.112 0.318 0.681 0.693 0.681 0.687 0.099 0.318 Delayed 0.671 0.700 0.685 0.669 0.681 0.675 RF Normal 0.749 0.729 0.814 0.769 2.254 16.242 0.735 0.717 0.802 0.757 2.231 16.242 Delayed 0.776 0.680 0.725 0.760 0.665 0.710 SVM Normal 0.651 0.631 0.774 0.695 3.625 458.280 0.646 0.629 0.756 0.687 3.452 474.522 Delayed 0.686 0.522 0.593 0.672 0.529 0.592 KNN Normal 0.641 0.655 0.637 0.646 0.003 60.510 0.652 0.662 0.661 0.661 0.004 31.847 Delayed 0.628 0.646 0.637 0.642 0.643 0.643 LR Normal 0.595 0.600 0.635 0.617 0.085 0.318 0.583 0.591 0.613 0.602 0.094 0.637 Delayed 0.589 0.552 0.570 0.575 0.553 0.563 XGB Normal 0.721 0.715 0.759 0.736 0.150 1.274 0.707 0.700 0.753 0.725 0.125 1.274 Delayed 0.728 0.680 0.703 0.716 0.659 0.686 LSTM Normal 0.644 0.620 0.776 0.689 490.4 3.503 0.609 0.602 0.679 0.638 490.8 0.318 Delayed 0.687 0.509 0.584 0.618 0.537 0.575 Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.678 0.691 0.675 0.683 0.095 0.318 0.687 0.702 0.678 0.690 0.091 0.318 Delayed 0.665 0.681 0.673 0.672 0.696 0.684 RF Normal 0.744 0.726 0.806 0.764 2.187 16.242 0.745 0.719 0.826 0.769 2.136 16.242 Delayed 0.768 0.679 0.721 0.782 0.659 0.715 SVM Normal 0.641 0.625 0.750 0.682 3.591 530.255 0.641 0.626 0.749 0.682 3.595 545.541 Delayed 0.666 0.525 0.587 0.666 0.528 0.589 KNN Normal 0.662 0.673 0.666 0.669 0.003 30.892 0.649 0.663 0.643 0.653 0.004 32.803 Delayed 0.651 0.658 0.655 0.635 0.656 0.645 LR Normal 0.598 0.612 0.591 0.601 0.095 0.318 0.525 0.535 0.574 0.554 0.069 0.637 Delayed 0.583 0.605 0.594 0.512 0.472 0.491 Time difference: 8 h Kim and Park Journal of Big Data (2024) 11:11 Page 13 of 25 ( ) Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) XGB Normal 0.727 0.714 0.784 0.747 0.123 0.955 0.714 0.706 0.759 0.732 0.123 1.274 Delayed 0.745 0.668 0.704 0.724 0.665 0.693 LSTM Normal 0.587 0.581 0.669 0.622 488.4 0.318 0.540 0.531 0.797 0.637 490.1 3.503 Delayed 0.595 0.502 0.545 0.566 0.274 0.369 Algorithm Time difference: 24 h Time difference: 48 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.676 0.704 0.672 0.688 0.093 0.637 0.7680 0.692 0.679 0.685 0.102 0.637 Delayed 0.670 0.702 0.685 0.668 0.681 0.674 RF Normal 0.743 0.724 0.808 0.764 2.230 16.561 0.748 0.721 0.830 0.772 2.465 19.427 Delayed 0.769 0.674 0.718 0.786 0.661 0.718 SVM Normal 0.647 0.625 0.784 0.695 3.922 572.930 0.631 0.619 0.732 0.671 3.846 592.357 Delayed 0.688 0.502 0.581 0.650 0.525 0.580 KNN Normal 0.651 0.652 0.651 0.005 31.529 0.648 0.660 0.650 0.655 0.004 35.032 0.641 Delayed 0.632 0.630 0.631 0.636 0.646 0.641 LR Normal 0.547 0.564 0.528 0.545 0.089 0.637 0.554 0.565 0.574 0.569 0.098 0.637 Delayed 0.533 0.568 0.550 0.542 0.532 0.537 XGB Normal 0.705 0.702 0.739 0.720 0.150 1.274 0.703 0.693 0.758 0.724 0.138 1.274 Delayed 0.708 0.669 0.688 0.716 0.646 0.679 LSTM Normal 0.580 0.586 0.591 0.588 493.3 3.185 0.551 0.548 0.666 0.601 494.4 3.503 Delayed 0.574 0.569 0.572 0.556 0.433 0.487 Kim and Park Journal of Big Data (2024) 11:11 Page 14 of 25 Algorithm Time difference: 2 h Time difference: 4 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.787 0.819 0.751 0.783 0.055 0.637 0.790 0.827 0.745 0.784 0.049 0.637 Delayed 0.759 0.826 0.791 0.758 0.837 0.795 RF Normal 0.843 0.835 0.864 0.849 0.993 21.019 0.850 0.838 0.877 0.857 1.058 20.064 Delayed 0.852 0.821 0.836 0.864 0.822 0.842 SVM Normal 0.650 0.643 0.709 0.675 5.253 618.153 0.638 0.646 0.650 0.648 4.914 632.166 Delayed 0.658 0.588 0.621 0.630 0.626 0.628 KNN Normal 0.712 0.749 0.659 0.701 0.008 70.382 0.722 0.761 0.667 0.711 0.005 40.045 Delayed 0.682 0.768 0.722 0.691 0.780 0.732 LR Normal 0.581 0.597 0.560 0.578 0.107 0.637 0.573 0.594 0.527 0.558 0.118 0.637 Delayed 0.566 0.603 0.584 0.556 0.622 0.587 XGB Normal 0.779 0.783 0.785 0.784 0.164 2.548 0.769 0.772 0.778 0.775 0.127 2.548 Delayed 0.774 0.772 0.773 0.765 0.760 0.762 LSTM Normal 0.852 0.831 0.882 0.856 560.0 4.140 0.829 0.826 0.833 0.829 564.4 4.140 Delayed 0.876 0.822 0.848 0.833 0.826 0.830 Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.796 0.826 0.764 0.793 0.049 0.637 0.800 0.827 0.770 0.798 0.051 0.637 Delayed 0.770 0.831 0.799 0.775 0.831 0.802 RF Normal 0.843 0.835 0.865 0.850 1.018 20.382 0.840 0.832 0.863 0.847 1.091 20.064 Delayed 0.853 0.820 0.836 0.850 0.817 0.833 SVM Normal 0.643 0.656 0.635 0.646 5.503 766.242 0.642 0.661 0.618 0.639 6.148 802.548 Delayed 0.630 0.651 0.640 0.625 0.667 0.645 KNN Normal 0.724 0.763 0.669 0.713 0.011 48.726 0.725 0.758 0.681 0.717 0.008 48.726 Delayed 0.692 0.782 0.735 0.697 0.771 0.733 LR Normal 0.594 0.617 0.545 0.579 0.112 0.637 0.582 0.605 0.529 0.564 0.124 0.637 Delayed 0.575 0.645 0.608 0.563 0.638 0.598 Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.796 0.826 0.764 0.793 0.049 0.637 0.800 0.827 0.770 0.798 0.051 0.637 Delayed 0.770 0.831 0.799 0.775 0.831 0.802 RF Normal 0.843 0.835 0.865 0.850 1.018 20.382 0.840 0.832 0.863 0.847 1.091 20.064 Delayed 0.853 0.820 0.836 0.850 0.817 0.833 SVM Normal 0.643 0.656 0.635 0.646 5.503 766.242 0.642 0.661 0.618 0.639 6.148 802.548 Delayed 0.630 0.651 0.640 0.625 0.667 0.645 KNN Normal 0.724 0.763 0.669 0.713 0.011 48.726 0.725 0.758 0.681 0.717 0.008 48.726 Delayed 0.692 0.782 0.735 0.697 0.771 0.733 LR Normal 0.594 0.617 0.545 0.579 0.112 0.637 0.582 0.605 0.529 0.564 0.124 0.637 Delayed 0.575 0.645 0.608 0.563 0.638 0.598 Time difference: 8 h Kim and Park Journal of Big Data (2024) 11:11 Page 15 of 25 ( ) Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) XGB Normal 0.776 0.785 0.776 0.780 0.132 2.548 0.778 0.783 0.782 0.783 0.121 2.548 Delayed 0.767 0.777 0.772 0.772 0.773 0.772 LSTM Normal 0.814 0.829 0.790 0.809 565.3 4.140 0.799 0.773 0.843 0.807 565.1 4.140 Delayed 0.802 0.838 0.820 0.829 0.756 0.791 Algorithm Time difference: 24 h Time difference 48 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.779 0.816 0.733 0.772 0.056 0.637 0.784 0.821 0.740 0.779 0.051 0.955 Delayed 0.747 0.826 0.785 0.753 0.831 0.790 RF Normal 0.837 0.822 0.869 0.845 1.081 21.975 0.846 0.830 0.880 0.854 1.067 20.064 Delayed 0.854 0.803 0.827 0.865 0.811 0.837 SVM Normal 0.618 0.637 0.589 0.612 6.033 824.841 0.625 0.641 0.608 0.624 6.403 890.127 Delayed 0.601 0.649 0.624 0.610 0.644 0.626 KNN Normal 0.723 0.755 0.678 0.714 0.008 51.911 0.721 0.752 0.681 0.714 0.011 43.949 Delayed 0.695 0.770 0.730 0.695 0.764 0.728 LR Normal 0.562 0.593 0.463 0.520 0.125 0.637 0.565 0.588 0.504 0.543 0.112 0.637 Delayed 0.542 0.666 0.598 0.547 0.629 0.585 XGB Normal 0.778 0.787 0.777 0.782 0.121 2.229 0.773 0.777 0.782 0.779 0.123 2.548 Delayed 0.769 0.779 0.774 0.770 0.764 0.767 LSTM Normal 0.778 0.780 0.771 0.776 568.2 4.140 0.736 0.724 0.761 0.742 569.4 4.140 Delayed 0.776 0.785 0.780 0.748 0.710 0.729 B ld l i di h l Kim and Park Journal of Big Data (2024) 11:11 Page 16 of 25 p Algorithm Time difference: 2 h Time difference: 4 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.731 0.741 0.702 0.721 0.073 0.955 0.722 0.750 0.659 0.702 0.045 0.955 Delayed 0.721 0.759 0.740 0.701 0.784 0.740 RF Normal 0.762 0.748 0.784 0.766 0.997 17.834 0.766 0.750 0.792 0.771 1.011 17.197 Delayed 0.777 0.741 0.759 0.784 0.741 0.762 SVM Normal 0.587 0.588 0.558 0.573 3.716 542.994 0.600 0.596 0.599 0.598 3.715 564.968 Delayed 0.586 0.615 0.600 0.604 0.601 0.602 KNN Normal 0.642 0.645 0.620 0.632 0.007 73.248 0.646 0.649 0.623 0.636 0.009 29.936 Delayed 0.640 0.664 0.652 0.643 0.668 0.656 LR Normal 0.571 0.570 0.548 0.558 0.079 0.637 0.581 0.579 0.567 0.573 0.076 0.955 Delayed 0.594 0.582 0.583 0.595 0.589 0.578 XGB Normal 0.716 0.718 0.703 0.710 0.178 2.866 0.715 0.717 0.702 0.709 0.142 3.185 Delayed 0.714 0.728 0.721 0.713 0.728 0.720 LSTM Normal 0.785 0.755 0.849 0.799 404.2 2.548 0.756 0.744 0.786 0.765 411.7 2.548 Delayed 0.824 0.719 0.768 0.769 0.725 0.746 Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.716 0.731 0.677 0.703 0.042 0.955 0.726 0.752 0.667 0.707 0.046 1.274 Delayed 0.704 0.755 0.729 0.705 0.783 0.742 RF Normal 0.767 0.75 0.775 0.767 1.070 17.834 0.774 0.752 0.812 0.781 1.331 17.516 Delayed 0.774 0.758 0.766 0.800 0.737 0.767 SVM Normal 0.613 0.619 0.574 0.595 4.172 612.102 0.615 0.619 0.584 0.601 3.862 637.898 Delayed 0.609 0.652 0.630 0.612 0.646 0.629 KNN Normal 0.643 0.657 0.585 0.619 0.006 27.389 0.667 0.677 0.629 0.652 0.006 29.618 Delayed 0.632 0.701 0.664 0.659 0.705 0.681 Ablation study We conducted training on the ICN dataset with identical parameters and training strate- gies, except for the exclusion of linear interpolation, while examining a time difference of 2 h. The results, as depicted in Table 17, reveal a slight reduction in overall performance, ranging from 1 to 2%, when interpolation was omitted. It is noteworthy that the interpo- lated data constitutes only 0.9% (953 out of 105,192) of the entire dataset, which lends credibility to the decision to incorporate linear interpolation in our research. Flight delay prediction (1 to 24 h, hourly) Tables 13, 14 and 15 provide an hourly breakdown of model accuracy from 1 h to 24 h, utilizing the same three datasets for ICN, JFK, and MDW airports, along with average training and testing times. The hyperparameters that yielded the best performance in the prior experiments were applied. Across all three airport datasets, the highest accu- racy was observed at a 1-h time difference, with a declining trend in performance as the time difference increased. The magnitude of performance decline from 1 h to 24 h for each model is detailed in Table 16. Notably, the Random Forest model exhibited the least performance degradation, with a decrease of only − 3.6%, while the SVM model showed the most significant performance decline, with an average decrease of − 16.1%. Machine learning models completed their training in just a few seconds, while LSTM required several 100 s, indicating it was approximately 100 times more time-consuming. In terms of testing time, it ranged from as low as 1 ms to a maximum of around 1.3 ms. Time difference: 8 h Kim and Park Journal of Big Data (2024) 11:11 Page 17 of 25 Table 12 (continued) Algorithm Time difference: 8 h Time difference: 16 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) LR Normal 0.578 0.577 0.554 0.565 0.071 1.911 0.605 0.605 0.587 0.596 0.080 0.637 Delayed 0.572 0.601 0.589 0.606 0.624 0.615 XGB Normal 0.714 0.723 0.687 0.704 0.140 3.185 0.726 0.731 0.706 0.719 0.150 3.185 Delayed 0.707 0.741 0.723 0.721 0.745 0.733 LSTM Normal 0.741 0.736 0.759 0.747 409.3 2.548 0.718 0.701 0.772 0.735 410.8 2.548 Delayed 0.746 0.723 0.735 0.741 0.664 0.700 Algorithm Time difference: 24 h Time difference: 48 h Accuracy Precision Recall F1-score Train (s) Test (us) Accuracy Precision Recall F1-score Train (s) Test (us) DT Normal 0.703 0.747 0.683 0.713 0.045 0.955 0.727 0.744 0.684 0.713 0.049 0.955 Delayed 0.712 0.773 0.741 0.712 0.769 0.740 RF Normal 0.773 0.762 0.787 0.774 1.081 18.153 0.772 0.759 0.790 0.774 1.128 16.879 Delayed 0.784 0.758 0.771 0.785 0.754 0.769 SVM Normal 0.600 0.611 0.528 0.566 3.974 595.223 0.614 0.619 0.577 0.597 4.377 630.573 Delayed 0.591 0.670 0.628 0.610 0.652 0.630 KNN Normal 0.663 0.667 0.641 0.654 0.006 27.389 0.664 0.679 0.609 0.642 0.008 26.433 Delayed 0.660 0.685 0.672 0.651 0.717 0.683 LR Normal 0.597 0.601 0.558 0.579 0.084 0.955 0.596 0.598 0.563 0.580 0.096 0.637 Delayed 0.594 0.635 0.614 0.594 0.628 0.611 XGB Normal 0.731 0.737 0.710 0.723 0.134 3.503 0.725 0.732 0.703 0.717 0.152 3.503 Delayed 0.725 0.751 0.738 0.719 0.747 0.733 LSTM Normal 0.714 0.697 0.768 0.731 415.2 2.548 0.712 0.719 0.702 0.710 416.1 2.548 Delayed 0.736 0.659 0.696 0.705 0.722 0.713 Kim and Park Journal of Big Data (2024) 11:11 Page 18 of 25 Discussion and concluding remarks For predicting flight takeoff delays using weather information for the airports of ICN, JFK, and MDW, machine learning and LSTM models were employed. Based on the pre- diction results for the three regions, the RF model demonstrated the highest perfor- mance for the ICN airport, while the LSTM model exhibited the highest performance for JFK and MDW airports, with a minimum time difference of 2 h. The accuracy scores were 0.749 for ICN, 0.852 for JFK, and 0.785 for MDW airports. Moreover, the RF model also displayed the best performance with high accuracy for all three airports, with a maximum time difference of 48 h; the accuracy scores were 0.748 for ICN, 0.846 for JFK, and 0.772 for MDW airports. Moreover, when assessing test times, all of the mod- els require less than 2 ms, which makes them suitable for real-time predictions. These findings confirm the feasibility of predicting flight takeoff delays using weather data col- lected 2 h prior to the scheduled departure time. Our analysis incorporated datasets spanning from 2011 to 2021, encompassing a long time period. This extensive dataset allowed us to leverage both actual flight operation data and weather information for our analysis. By utilizing these comprehensive datasets, our pro- posed models exhibited outstanding performance in predicting delayed flights across three different datasets. The utilization of a long-term dataset facilitated robust predictions and enhanced the reliability of our models. Furthermore, the approaches we developed can be applied to various other transportation-related domains, including ocean vessel delays, vehi- cle operation restrictions, and outdoor construction work stoppages. In these application areas, early-stage warnings play a crucial role in mitigating potential risks to human safety and property damage. By leveraging our proposed models, it becomes feasible to anticipate and prepare for potential disruptions, enabling proactive measures to be taken in advance. This can significantly contribute to minimizing the adverse impacts associated with delays and restrictions in these transportation-related sectors. The presented implications not- withstanding, it is important to acknowledge the presence of notable limitations. One such limitation is the significant influence of national and regional factors on weather conditions, rendering it challenging to generalize the results to other locations. The generalization of findings beyond the specific context may not be straightforward owing to these variations. Furthermore, the performance of the ICN airport dataset was relatively lower compared with the JFK and MDW airport datasets. Comparison with prior approaches We conducted a performance comparison between our models and a prior research model [8]. Using the same JFK airport dataset, we compared our research’s Random Forest and LSTM models with the prior research model’s LSTM model. Our Random Forest model achieved an accuracy of 84.3% with a 2-h time difference and 84.6% with a 48-h time difference. In contrast, the LSTM model in our research achieved an accuracy of 85.2% with a 2-h time difference and 73.6% with a 48-h time differ- ence. It’s worth noting that the previous model exhibited a performance of 86.51% at a short time interval of 15 min. Feature importance To determine the features with a substantial impact on our models, we conducted fea- ture importance analysis. We chose the Random Forest and LSTM models, which dem- onstrated the best performance. For the Random Forest model, we made use of the built-in feature importance function, whereas for the LSTM model, we employed exter- nal algorithms using loss data. Consequently, in the case of Random Forest, higher val- ues correspond to greater feature importance, whereas for LSTM, lower values signify reduced importance. Considering the results of the ICN airport dataset, Random Forest attributed the highest importance to temperature, dew point, and weather phenomena in that order, while LSTM assigned the highest importance to temperature, wind speed, weather phenomena, and local pressure. Notably, temperature was identified as the most crucial feature in both models (Table 18). For the JFK airport dataset, Random Forest identified pressure, temperature, and dew point as the most important features, while LSTM emphasized pressure, precipitation, and wind speed as the top influential factors. Notably, pressure was recognized as the most crucial feature in both models for this dataset (Table 19). In the case of the MDW airport dataset, Random Forest indicated that pressure, humidity, and temperature were the top features in terms of importance, while LSTM emphasized pressure, precipitation, and wind speed as the most influential factors. Notably, pressure was consistently identified as the most important feature in both mod- els for this dataset (Table 20). Kim and Park Journal of Big Data (2024) 11:11 Page 19 of 25 Discussion and concluding remarks This discrepancy in performance could be attrib- uted to several factors, including the presence of missing features in the dataset. Discussion and concluding remarks The absence Kim and Park Journal of Big Data (2024) 11:11 Page 20 of 25 Table 13 Results of ICN airport (accuracy from 1 to 24 h) Algorithm 1 h 2 h 3 h 4 h 5 h 6 h 7 h 8 h 9 h 10 h 11 h 12 h 13 h DT 0.688 0.688 0.687 0.681 0.686 0.678 0.674 0.678 0.677 0.675 0.680 0.675 0.661 RF 0.750 0.749 0.748 0.735 0.746 0.738 0.741 0.744 0.751 0.743 0.743 0.751 0.739 SVM 0.697 0.651 0.686 0.646 0.653 0.654 0.652 0.641 0.660 0.645 0.641 0.638 0.640 KNN 0.659 0.641 0.654 0.652 0.652 0.646 0.646 0.662 0.651 0.641 0.644 0.635 0.638 LR 0.669 0.595 0.659 0.583 0.589 0.592 0.582 0.598 0.571 0.573 0.547 0.537 0.548 XGB 0.734 0.721 0.733 0.707 0.715 0.726 0.718 0.727 0.717 0.714 0.712 0.722 0.712 LSTM 0.622 0.644 0.609 0.609 0.624 0.602 0.579 0.587 0.573 0.580 0.571 0.562 0.539 Algorithm 14 h 15 h 16 h 17 h 18 h 19 h 20 h 21 h 22 h 23 h 24 h Avg train (s) Avg test (us) DT 0.686 0.677 0.687 0.675 0.680 0.689 0.684 0.682 0.685 0.680 0.687 0.011 0.318 RF 0.737 0.743 0.745 0.747 0.739 0.739 0.756 0.745 0.738 0.746 0.743 3.613 20.701 SVM 0.643 0.628 0.641 0.636 0.651 0.651 0.628 0.640 0.641 0.640 0.647 6.295 970.382 KNN 0.643 0.630 0.649 0.653 0.629 0.629 0.641 0.655 0.640 0.657 0.641 0.004 47.134 LR 0.521 0.568 0.525 0.554 0.527 0.527 0.528 0.546 0.554 0.544 0.547 0.074 0.637 XGB 0.715 0.710 0.714 0.714 0.715 0.715 0.716 0.713 0.706 0.716 0.705 0.211 1.592 LSTM 0.558 0.563 0.540 0.564 0.561 0.561 0.571 0.571 0.548 0.545 0.580 468.7 184.713 Kim and Park Journal of Big Data (2024) 11:11 Page 21 of 25 Table 14 Results of JFK airport (accuracy from 1 to 24 h) Algorithm 1 h 2h 3 h 4 h 5 h 6 h 7 h 8 h 9 h 10 h 11 h 12 h 13 h DT 0.831 0.787 0.828 0.790 0.826 0.806 0.804 0.796 0.789 0.791 0.791 0.790 0.799 RF 0.882 0.843 0.877 0.850 0.869 0.853 0.854 0.843 0.846 0.848 0.845 0.842 0.846 SVM 0.801 0.650 0.782 0.638 0.728 0.701 0.698 0.643 0.662 0.678 0.671 0.665 0.638 KNN 0.803 0.712 0.792 0.722 0.772 0.756 0.738 0.724 0.731 0.739 0.736 0.724 0.730 LR 0.706 0.581 0.665 0.573 0.642 0.617 0.606 0.594 0.582 0.584 0.605 0.588 0.560 XGB 0.859 0.779 0.851 0.769 0.822 0.801 0.801 0.776 0.788 0.783 0.786 0.782 0.779 LSTM 0.848 0.852 0.842 0.829 0.828 0.826 0.817 0.814 0.809 0.812 0.804 0.803 0.800 Algorithm 14 h 15 h 16 h 17 h 18 h 19 h 20 h 21 h 22 h 23 h 24 h Avg train (s) Avg test (us) DT 0.799 0.790 0.800 0.786 0.797 0.784 0.781 0.796 0.797 0.784 0.779 0.081 0.796 RF 0.844 0.844 0.840 0.836 0.844 0.832 0.841 0.839 0.839 0.846 0.837 1.809 21.503 SVM 0.646 0.659 0.642 0.620 0.650 0.616 0.653 0.651 0.628 0.610 0.618 11.877 1289.089 KNN 0.727 0.724 0.725 0.728 0.724 0.708 0.722 0.727 0.722 0.707 0.723 0.009 57.606 LR 0.577 0.565 0.582 0.575 0.577 0.559 0.603 0.608 0.569 0.557 0.562 0.128 0.531 XGB 0.774 0.779 0.778 0.772 0.783 0.771 0.786 0.779 0.769 0.767 0.778 0.194 2.655 LSTM 0.795 0.809 0.799 0.804 0.803 0.796 0.794 0.786 0.785 0.785 0.778 554.2 4.513 Page 22 of 25 Kim and Park Journal of Big Data (2024) 11:11 Table 15 Results of MDW airport (accuracy from 1 to 24 h) Algorithm 1 h 2 h 3 h 4 h 5 h 6 h 7 h 8 h 9 h 10 h 11 h 12 h 13 h DT 0.755 0.731 0.743 0.722 0.763 0.740 0.738 0.716 0.745 0.726 0.731 0.743 0.720 RF 0.811 0.762 0.787 0.766 0.791 0.783 0.803 0.767 0.789 0.767 0.779 0.783 0.783 SVM 0.735 0.587 0.716 0.600 0.690 0.651 0.664 0.613 0.627 0.616 0.642 0.630 0.606 KNN 0.728 0.642 0.727 0.646 0.710 0.675 0.698 0.643 0.661 0.692 0.688 0.674 0.667 LR 0.698 0.571 0.655 0.581 0.641 0.612 0.607 0.578 0.566 0.595 0.590 0.589 0.575 XGB 0.797 0.716 0.769 0.715 0.776 0.741 0.750 0.714 0.731 0.715 0.755 0.743 0.745 LSTM 0.785 0.785 0.769 0.756 0.773 0.746 0.744 0.741 0.750 0.749 0.737 0.727 0.734 Algorithm 14 h 15 h 16 h 17 h 18 h 19 h 20 h 21 h 22 h 23 h 24 h Avg train (s) Avg test (us) DT 0.717 0.731 0.726 0.728 0.726 0.693 0.719 0.730 0.731 0.709 0.728 0.052 0.899 RF 0.777 0.776 0.774 0.773 0.765 0.756 0.766 0.790 0.782 0.756 0.773 1.243 23.820 SVM 0.584 0.615 0.615 0.596 0.608 0.572 0.600 0.622 0.563 0.580 0.600 4.500 884.045 KNN 0.661 0.663 0.667 0.659 0.665 0.653 0.646 0.658 0.670 0.652 0.663 0.008 48.090 LR 0.546 0.584 0.605 0.574 0.580 0.587 0.581 0.604 0.577 0.559 0.597 0.089 0.899 XGB 0.732 0.723 0.726 0.714 0.717 0.707 0.725 0.741 0.728 0.700 0.731 0.177 4.045 LSTM 0.716 0.728 0.718 0.724 0.713 0.729 0.723 0.715 0.709 0.727 0.714 391.6 3.596 Kim and Park Journal of Big Data (2024) 11:11 Page 23 of 25 Table 16 Comparison of accuracy levels between 1 and 24 h Table 16 Comparison of accuracy levels between 1 and 24 h Algorithm ICN (%) JFK (%) MDW (%) Average (%) DT − 1.7 − 6.3 − 6.9 − 5.0 RF − 0.9 − 5.1 − 4.7 − 3.6 SVM − 7.2 − 22.8 − 18.4 − 16.1 KNN − 2.7 − 10.0 − 8.9 − 7.2 LR − 18.2 − 20.4 − 14.5 − 17.7 XGB − 4.0 − 9.4 − 8.3 − 7.2 LSTM − 6.8 − 8.3 − 9.0 − 10.1 Table 17 Ablation study on linear interpolation in the ICN dataset with a time difference of 2 h Algorithm With linear interpolation Without linear interpolation Table 17 Ablation study on linear interpolation in the ICN dataset with a time difference of 2 h Algorithm With linear interpolation Without linear interpolation Accuracy Precision Recall F1-score Accuracy Precision Recall F1-score DT Normal 0.688 0.704 0.676 0.690 0.680 0.695 0.672 0.683 Delayed 0.671 0.700 0.685 0.665 0.688 0.677 RF Normal 0.749 0.729 0.814 0.769 0.740 0.724 0.800 0.760 Delayed 0.776 0.680 0.725 0.762 0.678 0.718 SVM Normal 0.651 0.631 0.774 0.695 0.600 0.597 0.684 0.638 Delayed 0.686 0.522 0.593 0.605 0.511 0.554 KNN Normal 0.641 0.655 0.637 0.646 0.635 0.643 0.653 0.648 Delayed 0.628 0.646 0.637 0.627 0.616 0.622 LR Normal 0.595 0.600 0.635 0.617 0.545 0.553 0.590 0.571 Delayed 0.589 0.552 0.570 0.534 0.496 0.514 XGB Normal 0.721 0.715 0.759 0.736 0.688 0.681 0.737 0.708 Delayed 0.728 0.680 0.703 0.696 0.635 0.664 of these features may have impacted the overall performance of the models. Discussion and concluding remarks Future research endeavors should focus on addressing these limitations by exploring more comprehensive datasets and improving data collection methods to minimize missing features. This would enhance the generalizability and accuracy of the models in predicting flight delays. Table 17 Ablation study on linear interpolation in the ICN dataset with a time difference of 2 h Algorithm With linear interpolation Without linear interpolation Accuracy Precision Recall F1-score Accuracy Precision Recall F1-score DT Normal 0.688 0.704 0.676 0.690 0.680 0.695 0.672 0.683 Delayed 0.671 0.700 0.685 0.665 0.688 0.677 RF Normal 0.749 0.729 0.814 0.769 0.740 0.724 0.800 0.760 Delayed 0.776 0.680 0.725 0.762 0.678 0.718 SVM Normal 0.651 0.631 0.774 0.695 0.600 0.597 0.684 0.638 Delayed 0.686 0.522 0.593 0.605 0.511 0.554 KNN Normal 0.641 0.655 0.637 0.646 0.635 0.643 0.653 0.648 Delayed 0.628 0.646 0.637 0.627 0.616 0.622 LR Normal 0.595 0.600 0.635 0.617 0.545 0.553 0.590 0.571 Delayed 0.589 0.552 0.570 0.534 0.496 0.514 XGB Normal 0.721 0.715 0.759 0.736 0.688 0.681 0.737 0.708 Delayed 0.728 0.680 0.703 0.696 0.635 0.664 Table 18 Feature importance of ICN airport Bold valuesindicate the greatest results Feature importance Wind speed Visibility Weather phenomena Temperature Dew point Sea-level pressure Local pressure RF 0.062 0.057 0.073 0.078 0.076 0.071 0.069 LSTM 20.718 20.724 20.721 20.715 21.006 20.932 20.721 Table 19 Feature importance of JFK airport Bold valuesindicate the greatest results Feature importance Temperature Dew point Humidity Wind speed Wind gust Pressure Precipitation RF 0.183 0.167 0.155 0.109 0.037 0.184 0.005 LSTM 68.161 22.531 43.579 10.509 35.987 8.548 9.584 of these features may have impacted the overall performance of the models. Discussion and concluding remarks Future research endeavors should focus on addressing these limitations by exploring more comprehensive Delayed 0.728 0.680 0.703 0.696 0.635 0.664 Table 18 Feature importance of ICN airport Bold valuesindicate the greatest results Feature importance Wind speed Visibility Weather phenomena Temperature Dew point Sea-level pressure Local pressure RF 0.062 0.057 0.073 0.078 0.076 0.071 0.069 LSTM 20.718 20.724 20.721 20.715 21.006 20.932 20.721 Table 19 Feature importance of JFK airport Bold valuesindicate the greatest results Feature importance Temperature Dew point Humidity Wind speed Wind gust Pressure Precipitation RF 0.183 0.167 0.155 0.109 0.037 0.184 0.005 LSTM 68.161 22.531 43.579 10.509 35.987 8.548 9.584 Table 18 Feature importance of ICN airport Bold valuesindicate the greatest results Feature importance Wind speed Visibility Weather phenomena Temperature Dew point Sea-level pressure Local pressure RF 0.062 0.057 0.073 0.078 0.076 0.071 0.069 LSTM 20.718 20.724 20.721 20.715 21.006 20.932 20.721 Table 18 Feature importance of ICN airport of these features may have impacted the overall performance of the models. Future research endeavors should focus on addressing these limitations by exploring more comprehensive datasets and improving data collection methods to minimize missing features. This would enhance the generalizability and accuracy of the models in predicting flight delays. of these features may have impacted the overall performance of the models. Future research endeavors should focus on addressing these limitations by exploring more comprehensive datasets and improving data collection methods to minimize missing features. This would enhance the generalizability and accuracy of the models in predicting flight delays. Page 24 of 25 Kim and Park Journal of Big Data (2024) 11:11 Table 20 Feature importance of MDW airport Bold valuesindicate the greatest results Feature importance Temperature Dew point Humidity Wind speed Wind gust Pressure Precipitation RF 0.165 0.158 0.170 0.109 0.029 0.176 0.004 LSTM 36.448 30.777 26.572 13.893 78.303 11.887 11.957 In future research, our aim is to develop a more robust model that incorporates geo- graphic information, enabling its application to other airports beyond the specific data- sets analyzed in this study. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Funding Funding This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2023S1A5A8075518). This research was supported by the MSIT (Ministry of Science and ICT), Korea, under the ICAN (ICT Challenge and Advanced Network of HRD) support program (IITP-2023-RS-2023-00259497) supervised by the IITP (Institute for Information & Communications Technology Planning & Evaluation). Received: 9 June 2023 Accepted: 14 December 2023 Received: 9 June 2023 Accepted: 14 December 2023 Received: 9 June 2023 Accepted: 14 December 2023 References 1. Economics-IATA: air passenger market analysis 2014. 2014. y 2. Economics-IATA: air passenger market analysis 2023. 2023. https://www.iata.org/en/iata-repository/publications/ economic-reports/air-passenger-market-analysis2/. 2. Economics-IATA: air passenger market analysis 2023. 2 economic-reports/air-passenger-market-analysis2/. 2. Economics-IATA: air passenger market analysis 2023. 2023. economic-reports/air-passenger-market-analysis2/. 3. Efthymiou M, Njoya ET, Lo PL, Papatheodorou A, Randall D. The impact of delays on customers’ satisfaction: an empirical analysis of the British airways on-time performance at Heathrow airport. J Aerosp Technol Manag. 2018;11:e0219. 3. Efthymiou M, Njoya ET, Lo PL, Papatheodorou A, Randall D. The impact of delays on customers’ satisfaction: an empirical analysis of the British airways on-time performance at Heathrow airport. J Aerosp Technol Manag. 2018;11:e0219. 4. Tileagă C, Oprisan O. Flights delay compensation 261/2004: a challenge for airline companies? In: Organizations and Performance in a complex world: 26th international economic conference of Sibiu (IECS) 26. Springer; 2021. p. 335–44. 4. Tileagă C, Oprisan O. Flights delay compensation 261/2004: a challenge for airline companies? In: Organizations and Performance in a complex world: 26th international economic conference of Sibiu (IECS) 26. Springer; 2021. p. 335–44. 5. U.S.D. of transportation, airline on-time performance data. 2023. https://www.transtats.bts.gov/tables.asp?QO_VQ= EFD &QO_anzr=Nv4yv0r. 6. Sim G-M, Kim Y-S, Jung M-P, Kim J-W, Park M-S, Hong S-H, Kang K-K. Changes in the frequency of abnormal weather events in South Korea in recent years. J Korean Soc Clim Change. 2018;9(4):461–70. 6. Sim G M, Kim Y S, Jung M P, Kim J W, Park M S, Hong S H, Kang K K. Changes in the events in South Korea in recent years. J Korean Soc Clim Change. 2018;9(4):461–70 y g 7. Lee J-W, Yoo H-I, Kim G-H. Analysis of South Korea’s heavy rain characteristics from 2006 to 2015 using AWS data. In: Proceedings of the Korean meteorological society conference 2016 p 521–2 7. Lee J-W, Yoo H-I, Kim G-H. Analysis of South Korea’s heavy rain characteristics from 2006 to 2015 using AWS data. In: Proceedings of the Korean meteorological society conference. 2016. p. 521–2. 8. Kim YJ, Choi S, Briceno S, Mavris D. A deep learning approach to flight delay prediction. In: 2016 IEEE/AIAA 35th digital avionics systems conference (DASC). IEEE; 2016. p. 1–6. y 9. Khaksar H, Sheikholeslami A. Airline delay prediction by machine learning algorithms. Scientia Iranica. 2019;26(5):2689–702. 10. Qu J, Wu S, Zhang J. Flight delay propagation prediction based on deep learning. Mathematics. 2023;11(3):494. Author contributions Author contributions Kim contributed to the design, implementation, and analysis of the research with the examination of the manuscript. Kim and Park wrote and revised the manuscript. Park approved the final version of the manuscript. Competing interests h h d l h Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. References nal conference on signal and image processing. IEEE; 2014. p. 171–8 international conference on signal and image processing. IEE g g p g p 20. Ardiansyah D, Mantoro T, Syafei WA. Potential classification prediction of solar and wind energy in Indonesia using machine learning with random forest algorithm. In: 2022 5th international conference of computer and informatics engineering (IC2IE). IEEE; 2022. p. 297–302. g g ( ) ; p 21. Lee J, Park E. D-HRSP: dataset of helpful reviews for service providers. Telemat Inform. 2023;82:102001. 22. Fadhil IM, Sibaroni Y. Topic classification in Indonesian-language tweets using fast-text feature expansion with sup- port vector machine (SVM). In: 2022 international conference on data science and its applications (ICoDSA). IEEE; 2022. p. 214–9. p 23. Charan PVS, Ramkumar G. Black fungus classification using Adaboost with SVM-based classifier and compare accuracy with support vector machine. In: 2022 5th international conference on contemporary computing and informatics (IC3I). IEEE; 2022. p. 1895–901. p 24. Hwang S, Ahn H, Park E. iMovieRec: a hybrid movie recommendation method based on a user-image-item model. Int J Mach Learn Cybern. 2023;14:3205–16. y 25. Auleria M, Arrahmah AI, Saputra DE. A review on KN nearest neighbour based classification for object recog In: 2021 international conference on data science and its applications (ICoDSA). 2021; IEEE. p. 274–80. y 25. Auleria M, Arrahmah AI, Saputra DE. A review on KN nearest neighbour based classification for object recognition. I 2021 i t ti l f d t i d it li ti (IC DSA) 2021 IEEE 274 80 In: 2021 international conference on data science and its app In: 2021 international conference on data science and its applications (ICoDSA). 2021; IEEE. p. 274–80. 26. Kim S, An C, Cha J, Kim D, Park E. D-visa: a dataset for detecting visual sentimen the IEEE/CVF international conference on computer vision. 2023. p. 3051–9. 26. Kim S, An C, Cha J, Kim D, Park E. D-visa: a dataset for detecting visual sentiment from art images. In: Proceedings of the IEEE/CVF international conference on computer vision. 2023. p. 3051–9. 27. Akoulih M, Tigani S, Saadane R, Tazi A. Electrocoagulation based chromium removal efficiency classification using logistic regression. Appl Sci. 2020;10(15):5179. logistic regression. Appl Sci. 2020;10(15):5179. 28. Guan X, Zhang J, Chen S. Logistic regression based on statistical learning model with linearized kernel for classifica- tion. Comput Inform. 2021;40(2):298–317. 29. Paleczek A, Grochala D, Rydosz A. References In: 2020 interna- tional conference on computer science, engineering and applications (ICCSEA). IEEE; 2020. p. 1–5. 45. Panda B, Adhikari RK. A method for classification of missing values using data mining techniques. In: 2020 interna- tional conference on computer science, engineering and applications (ICCSEA). IEEE; 2020. p. 1–5. References Artificial breath classification using XGBoost algorithm for diabetes detection. Sen- sors. 2021;21(12):4187. 30. Liang H, Li J, Wu H, Li L, Zhou X, Jiang X. Mammographic classification of breast cancer microcalcifications through extreme gradient boosting. Electronics. 2022;11(15):2435. g g ; ( ) 31. Lee S, Jeong D, Park E. MultiEmo: multi-task framework for emoji prediction. Knowl-Based Syst. 2022;242: 108437.i 1. Lee S, Jeong D, Park E. MultiEmo: multi-task framework for emoj 32. Hur Y. Malaysian name-based ethnicity classification using LSTM. KSII Trans Internet Inf Syst. 2022;16(12):3855–67. 33. Zerrouki N, Houacine A, Harrou F, Bouarroudj R, Cherifi MY, Sun Y. Exploiting deep learning-based LSTM classification for improving hand gesture recognition to enhance visitors’ museum experiences. In: 2022 international conference on innovation and intelligence for informatics, computing, and technologies (3ICT). IEEE; 2022. p. 451–6. ji p g p g for improving hand gesture recognition to enhance visitors’ museum experiences. In: 2022 international co on innovation and intelligence for informatics, computing, and technologies (3ICT). IEEE; 2022. p. 451–6. 34. Madanan M, Venugopal A, Velayudhan NC. A hybrid anomaly based intrusion detection methodology using IWD for LSTM classification. In: 2020 IEEE international conference on advanced networks and telecommunications systems (ANTS). IEEE; 2020. p. 1–5. 35. Lee S, Kim J, Kim D, Kim KJ, Park E. Computational approaches to developing the implicit media bias dataset: assess- ing political orientations of nonpolitical news articles. Appl Math Comput. 2023;458:128219. 36. Lee S, Kim J, Park E. Can book covers help predict bestsellers using machine learning approaches? Telemat Inform. 2023;78: 101948. 37. Park E. CRNet: a multimodal deep convolutional neural network for customer revisit prediction. J Big Data. 2023;10(1):1–10. 38. Oh S, Ji H, Kim J, Park E, del Pobil AP. Deep learning model based on expectation–confirmation theory to predict customer satisfaction in hospitality service. Inform Technol Tour. 2022;24(1):109–26. Yu H, Park E. A harmless webtoon for all: an automatic age-restriction prediction system for webtoon contents. Telemat Inform. 2023;76: 101906. 39. Yu H, Park E. A harmless webtoon for all: an automatic age-restriction prediction system for webtoon contents. Telemat Inform. 2023;76: 101906. 40. Incheon airport weather. https://data.kma.go.kr/data/air/selectAmosRltmList.do?pgmNo=575 &tabNo=1. City weather. https://www.wunderground.com/history/daily/us/ny 42. Chicago City weather. https://www.wunderground.com/history/daily/us/il/chicago/KMDW.l 43. Incheon air port flight. https://www.airport.kr/co/ko/cpr/statisticOfDelay.do. 44. United States Department of Transport. https://www.transtats.bts.gov/tables.asp?QO_VQ=EFD &QO_anzr=Nv4yv 0r. 45. Panda B, Adhikari RK. A method for classification of missing values using data mining techniques. References 11 Al T bb kh SM El Z h d H M hi l i t h i f l i f E ti fli ht d l J S i R S i 10. Qu J, Wu S, Zhang J. Flight delay propagation prediction based on deep learning. Mathematics. 2023;11(3):494. 11. Al-Tabbakh SM, El-Zahed H. Machine learning techniques for analysis of Egyptian flight delay. J Sci Res Sci. 2018 35( t 1) 390 9 10. Qu J, Wu S, Zhang J. Flight delay propagation prediction based on deep learning. Mathematics. 2023;11(3):494. 11. Al-Tabbakh SM, El-Zahed H. Machine learning techniques for analysis of Egyptian flight delay. J Sci Res Sci. 2018;35(part 1):390–9. Page 25 of 25 Kim and Park Journal of Big Data (2024) 11:11 12. Ye B, Liu B, Tian Y, Wan L. A methodology for predicting aggregate flight departure delays in airports based on supervised learning. Sustainability. 2020;12(7):2749. g y 13. Atlioğlu MC, Bolat M, Şahin M, Tunali V, Kilinç D. Supervised learning approaches to flight delay prediction. Sakarya Univ J Sci. 2020;24(6):1223–31. 14. Yazdi MF, Kamel SR, Chabok SJM, Kheirabadi M. Flight delay prediction based on deep learning and Levenberg–Mar- quart algorithm. J Big Data. 2020;7:1–28. q g g 15. Lee J, Cha J, Park E. Data-driven approaches into political orientation and news outlet discrimination: the case of news articles in south korea. Telemat Inform. 2023;85: 102066. 15. Lee J, Cha J, Park E. Data-driven approaches into political orienta news articles in south korea. Telemat Inform. 2023;85: 102066. news articles in south korea. Telemat Inform. 2023;85: 102066 16. Gao Z, Gatpandan MP, Gatpandan PH. Classification decision tree algorithm in predicting students’ course prefer- ence. In: 2021 2nd international symposium on computer engineering and intelligent communications (ISCEIC). IEEE; 2021. p. 93–7. ; p 17. Sharma A, Sharma M, Dwivedi R. Improved decision tree classification (IDT) algorithm for social media data. In: 2021 10th international conference on system modeling & advancement in research trends (SMART). IEEE; 2021. p. 155–7. y g p 18. Kim E, Ji H, Kim J, Park E. Classifying apartment defect repair tasks in South Korea: a machine learning approach. J Asian Archit Build Eng. 2022;21(6):2503–10. y g p Asian Archit Build Eng. 2022;21(6):2503–10. g 19. Soumya A, Kumar GH. Classification of ancient epigraphs into different periods using random forests. In: 2014 fifth international conference on signal and image processing. IEEE; 2014. p. 171–8. Publisher’s Note ub s e s ote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
https://openalex.org/W2889971187	https://europepmc.org/articles/pmc6164118?pdf=render	English	null	C-Type Natriuretic Peptide (CNP) Inhibition of Interferon-γ-Mediated Gene Expression in Human Endothelial Cells In Vitro	Biosensors	2,018	cc-by	9,250	Received: 30 July 2018; Accepted: 11 September 2018; Published: 14 September 2018 Abstract: Cardiovascular diseases, including atherosclerosis, now account for more deaths in the Western world than from any other cause. Atherosclerosis has a chronic inﬂammatory component involving Th1 pro-inﬂammatory cytokines such as IFN-γ, which is known to induce endothelial cell inﬂammatory responses. On the other hand CNP, which acts via its receptors to elevate intracellular cGMP, is produced by endothelium and endocardium and is upregulated in atherosclerosis. It is believed to be protective, however its role in vascular inﬂammation is not well understood. The aim of this study was to investigate the effects of CNP on human endothelial cell inﬂammatory responses following IFN-γ stimulation. Human umbilical vein endothelial cells were treated with either IFN-γ (10 ng/mL) or CNP (100 nm), or both in combination, followed by analysis by ﬂow cytometry for expression of MHC class I and ICAM-1. IFN-γ signiﬁcantly increased expression of both molecules, which was signiﬁcantly inhibited by CNP or the cGMP donor 8-Bromoguanosine 3’,5’-cyclic monophosphate (1 µm). CNP also reduced IFN-γ mediated kynurenine generation by the IFN-γ regulated enzyme indoleamine-2,3-deoxygenase (IDO). We conclude that CNP downmodulates IFN-γ induced pro-inﬂammatory gene expression in human endothelial cells via a cGMP-mediated pathway. Thus, CNP may have a protective role in vascular inﬂammation and novel therapeutic strategies for CVD based on upregulation of endothelial CNP expression could reduce chronic EC inﬂammation. Keywords: C-type natriuretic peptide; endothelium; inflammation; cardiovascular; interferon gamma; indoleamine-2,3-dioxygenase C-Type Natriuretic Peptide (CNP) Inhibition of Interferon-γ-Mediated Gene Expression in Human Endothelial Cells In Vitro Amy Day 1,2, Zoe Jameson 1,2, Carolyn Hyde 3, Bigboy Simbi 2, Robert Fowkes 2,* and Charlotte Lawson 1,* 1 Cardiovascular and Inﬂammation Biology Group, Comparative Biomedical Sciences, Royal Veterinary College, Royal College Street London, NW1 0TU, UK; Amy.Day@gstt.nhs.uk (A.D.); zjameson3@rvc.ac.uk (Z.J.) 1 Cardiovascular and Inﬂammation Biology Group, Comparative Biomedical Sciences, Royal Veterinary College, Royal College Street London, NW1 0TU, UK; Amy.Day@gstt.nhs.uk (A.D.); zjameson3@rvc.ac.uk (Z.J.) j ( ) 2 Endocrine Signalling Group, Royal Veterinary College, University of London, Royal College Street, London NW1 0TU, UK; bsimbi@rvc.ac.uk j 2 Endocrine Signalling Group, Royal Veterinary College, University of London, Royal College Street, London NW1 0TU, UK; bsimbi@rvc.ac.uk 2 Endocrine Signalling Group, Royal Veterinary College, University of London, Royal College Street, London NW1 0TU, UK; bsimbi@rvc.ac.uk 3 Bio-Analysis Centre, London Bioscience Innovation Centre, Royal College Street, London NW1 0NH, UK; cali@b-ac.co.uk 3 Bio-Analysis Centre, London Bioscience Innovation Centre, Royal College Street, London NW1 0NH, UK; cali@b-ac.co.uk * Correspondence: chlawson@rvc.ac.uk (C.L.); rfowkes@rvc.ac.uk (R.F.); Tel.: +44-( * Correspondence: chlawson@rvc.ac.uk (C.L.); rfowkes@rvc.ac.uk (R.F.); Tel.: +44-(0)20-7468-1216 (C.L.) biosensors biosensors biosensors biosensors www.mdpi.com/journal/biosensors 1. Introduction According to the latest statistics published by the American Heart Association cardiovascular disease (CVD) is the largest cause of death worldwide and accounts for over 750,000 deaths in the USA annually [1,2]. The importance of the endothelium is well established for the initial development and subsequent advancement of CVD, having a large inﬂuence on maintenance of blood vessel tone via production of nitric oxide and other vasoactive factors. Furthermore, a healthy endothelium is responsible for the maintenance of an anti-thrombotic environment via the production of tissue factor pathway inhibitors and thrombomodulin [3,4]. There is now compelling evidence that chronic systemic Biosensors 2018, 8, 86; doi:10.3390/bios8030086 www.mdpi.com/journal/biosensors 2 of 12 Biosensors 2018, 8, 86 inﬂammation also has a major impact on progression of CVD, with accelerated secondary CVD being noted in obese individuals, and as a risk factor for patients with diabetes or autoimmunity [1]. There is very strong evidence that CD4+ T helper type 1 (Th1) cells are present in early atheromatous lesions and contribute to lesion progression [5], including production of the pro-inﬂammatory cytokine, interferon gamma (IFN-γ). IFN-γ has many pro-inﬂammatory effects on vascular endothelial cells (EC), most notably enhancing the expression of adhesion molecules involved in ﬁrm adhesion of both monocytes (VCAM-1), and other leukocyte populations (ICAM-1 [6]). It also increases expression of MHC class I and MHC Class II in human EC [7], which may lead to further activation of antigen speciﬁc CD8 and CD4+ T cells, thus contributing to plaque progression. IFN-γ also has a number of immuno-regulatory functions, one of the most well described is its upregulation of indoleamine-2,3-deoxygenase (IDO), an inducible enzyme that reduces tryptophan availability by catalyzing its breakdown to kynurenine for reviews see [8–11]. This reduction in bioavailable tryptophan in the local microenvironment of the inﬂamed vasculature could potentially reduce the activity of highly metabolic cells such as T lymphocytes and myeloid cells and therefore reduce the chronic immune response. Atrial- and B-Type natriuretic peptides (ANP and BNP, respectively) are well characterised hormones that exert profound effects on the cardiovascular system, as well as having established anti-inﬂammatory roles in the endothelium and other tissues [12–15]. C-type natriuretic peptide (CNP) is a 22 amino acid peptide, the third member of the natriuretic peptide family, identiﬁed ﬁrst in extracts from porcine brain [16]. 1. Introduction It is known to have actions on central regulation of vasoactive hormones such as vasopressin and adrenocorticotrophin hormone release, but has also been shown to have direct vasodilatory effects, in particular on smooth muscle relaxation. Coupled with the ﬁnding that it is less well expressed in heart than other family members and its abundance in endothelium, it has been hypothesized to be a third, so called, endothelial derived hyperpolarizing factor (EDHF), alongside nitric oxide and prostacyclin [17,18]. CNP has also been described to have broadly “anti-inﬂammatory” effects on endothelial cells in vitro [19–21], to be anti-ﬁbrotic [22–24] and to be cardioprotective [25–28]. We hypothesized that as CNP has been shown to be abundantly expressed in endothelial cells and to have broadly anti-inﬂammatory effects on EC it can act as a protective brake, speciﬁcally acting on pro-inﬂammatory molecule expression to limit leukocyte emigration and IDO expression to limit subsequent activation after an inﬂammatory insult. We measured the effect of CNP administration on endothelial pro-inﬂammatory gene expression including ICAM-1, MHCI, MHCII and IDO activity after treatment of human umbilical vein endothelial cells with IFN-γ. 2.1. Materials All reagents were from Sigma (Poole, UK) unless otherwise stated. All tissue culture reagents were from Sigma or Gibco (ThermoFisher, Loughborough, UK), and plastics were from Nunc (ThermoFisher; Loughborough, UK). 2.3. Flow Cytometry HUVEC were treated on 24 well plates for up to 72 h as above. The supernatants were removed and snap frozen at −80 ◦C for further analysis. Cells were harvested by trypsinization. Cells were centrifuged and the pellet resuspended in 2 mL PBS divided between four 5 mL FACS tubes and re-centrifuged. Pellets were resuspended in 50 µL PBS and primary antibodies added at predetermined concentrations (anti-human ICAM-1 clone 6.5B5 (a gift from Professor DO Haskard, Imperial College UK); anti-MHC class I clone W6/32 (ATCC); MHC class II clone L243 (ATCC)). Cells were held on ice for 30 min, washed with 2 mL ice cold PBS, centrifuged, pellets were resuspended in 50 µL cold PBS with predetermined concentration of FITC-conjugated Goat anti-mouse-Ig antibody (Jackson; Stratech Luton, UK) and incubated on ice for 30 min followed by a wash with 2 mL ice cold PBS, and centrifugation. Pellets were resuspended in 0.5 mL PBS/0.5% formaldehyde and held at 4 ◦C before analysis by ﬂow cytometry using a FACS CANTO II with FACS DIVA software (BD Biosciences; Oxford, UK). The ﬂow cytometer was calibrated daily with CS&T beads (BD Biosciences) according to the manufacturer’s instructions. Cell populations were gated on FSC/SSC and using a negative control of FITC-conjugated secondary antibody alone. 2.5. Mass Spectrometry for Tryptophan Metabolites 2.5. Mass Spectrometry for Tryptophan Metabolites Four metabolites of tryptophan were prepared at 1 mg/mL for use as standards Kynurenine (K), 2,3-Pyridinecarboxylic acid (PCA), 3 Hydroxy DL Kynerenine (HDLK) and 3 Hydroxyanthranilic acid (HAA) (All from Sigma). K, PCA and HAA were dissolved in 500 µL of water and 500 µL methanol. HDLK was dissolved in 600 µL methanol, 400 µL water and 1 µL formic acid. Standards were; 10 µg/mL, 100 ng/mL and 10 ng/mL. g g g Samples were thawed at room temperature and 10 µL of each sample was diluted with 495 µL of water and 495 µL of methanol. A Shimadzu LCMS8040 was used—a triple quadruple mass spectrometer with high sensitivity, high speed and high reliability. Analysis of both positive and negative ions is possible in the same experiment due to the ultrafast polarity switching capabilities. LabSolutions software was used for the acquisition and analysis of data. A Phenomenex (Macclesﬁeld, UK) Kinetex PFP (50 × 2.1 mm i.d., 2.6 µm, 100 Å) with mobile phases A (0.1% formic acid in water, v/v) and B (0.1% formic acid in Methanol, v/v) was used. The column was kept at 40 ◦C. All analytes were detected in positive ion multiple reaction monitoring (MRM) mode. The ﬂow rate, at all times, was 0.25 mL/min. The column efﬂuent was delivered to the mass spectrometer with no split. If not otherwise noted, an injection volume of 1 µL was used. All MS parameters were optimized by the auto optimization program. The ESI source was operated under standard conditions of Nebulising Gas at 3 L/min, DL temperature at 250 ◦C, a heat block temperature at 400 ◦C and drying gas ﬂow at 15 L/min. 2.4. cGMP Assay Cells were treated with CNP in quadruplicate in 24 well plates for 30 min at 37 ◦C before removal of culture medium and lysis and fixation with 750 µL 100% Ethanol for 5 min at RT before freezing at −20 ◦C until analysis using cGMP EIA (R&D, Oxford, UK) according to the manufacturer’s instructions. 2.2. Cell Culture Umbilical cords were collected from The Royal London Hospital with approval from the East London Research Ethics Committee and according to the Declaration of Helsinki. Human umbilical vein endothelial cells (HUVEC) were isolated according to a modiﬁed method of Jaffe [29] and maintained as described previously [30]. Cells were seeded onto plates pre-coated with gelatin at 2 × 105 cells/well (6 well) or 5 × 104 cells/well (24 well) plates as appropriate. They were allowed to adhere overnight before treatment with IFN-γ 10 ng/mL (Insight Biotech, Wembley, UK), 100 nm CNP (Sigma, Poole, UK) for up to 72 h, 1 µm 8-bromo-cGMP (Sigma, Poole, UK) or 100 µm 8-(4-Chlorophenylthio)-guanosine 3’,5’-cyclic monophosphate (8-CPT-cGMP; Sigma, Poole, UK), for 48 h. 3 of 12 Biosensors 2018, 8, 86 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC As shown in Figure 1, and as has been described previously in the literature [31,32] IFN-γ induced 4.9 ± 1.2-fold increase in ICAM-1 expression above basal levels in HUVEC after 24 h treatment, increasing to 7.4 ± 0.9-fold after 48 h and maintained for at least 72 h. Alone, CNP had no effect on ICAM-1 cell surface expression; however, it caused a signiﬁcant reduction in the IFN-γ mediated response at both 24 h (to 2.6 ± 0.06-fold, p < 0.05) and 48 h (to 4.3 ± 0.7-fold, p < 0.05) with the trend continuing out to 72 h treatment (3.7 ± 1.0-fold, p = 0.24). Similarly, CNP did not affect basal MHC class I expression when administered alone to HUVEC for up to 72 h (Figure 2) but signiﬁcantly reduced IFN-γ mediated increased surface expression after 48h co-treatment (6.4 ± 1.2-fold vs. 3.2 ± 0.6-fold, p < 0.05) and 72 h (7.8 ± 2.5-fold vs. 3.3 ± 0.6-fold p < 0.05). In contrast (Figure 3), although IFN-γ increased the expression of MHC class II on the surface of HUVEC at each time-point (2.8 ± 1.1-fold, 5.5 ± 1.4-fold, 17.2 ± 6.0-fold, at 24 h, 48 h and 72 h, respectively), CNP failed to signiﬁcantly alter these responses. As shown in Figure 1, and as has been described previously in the literature [31,32] IFN-γ induced 4.9 ± 1.2-fold increase in ICAM-1 expression above basal levels in HUVEC after 24 h treatment, increasing to 7.4 ± 0.9-fold after 48 h and maintained for at least 72 h. Alone, CNP had no effect on ICAM-1 cell surface expression; however, it caused a significant reduction in the IFN-γ mediated response at both 24 h (to 2.6 ± 0.06-fold, p < 0.05) and 48 h (to 4.3 ± 0.7-fold, p < 0.05) with the trend continuing out to 72 h treatment (3.7 ± 1.0-fold, p = 0.24). Similarly, CNP did not affect basal MHC class I expression when administered alone to HUVEC for up to 72 h (Figure 2) but significantly reduced IFN-γ mediated increased surface expression after 48h co-treatment (6.4 ± 1.2-fold vs. 3.2 ± 0.6-fold, p < 0.05) and 72 h (7.8 ± 2.5-fold vs. 3.3 ± 0.6-fold p < 0.05). 2.6. Data Analysis Post-acquisition analysis of ﬂow cytometry data was using FACS DIVA II software (BD Biosciences, Oxford, UK), FloJo v10 (FLoJo LLC, Ashland, OR, USA) and Flowing Software v2.5.1 (University of Turku, Finland) and data is presented as fold increase in median ﬂuorescence intensity over untreated cells; mean ± SEM. All statistical analyses were performed using Prism 7 (GraphPad Software Inc., CA, USA). One-way ANOVA followed by Bonferroni post-test or Independent T tests were used as appropriate, with p < 0.05 () considered as statistically signiﬁcant. Mann–Whitney tests were used to 4 of 12 Biosensors 2018, 8, 86 analyze data that were not normally distributed. All experiments were performed on at least three separate isolates of HUVEC and data are presented as mean ± SEM. biosensors 2018, 18, x FOR PEER REVIEW 4 of 12 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC In contrast (Figure 3), although IFN-γ increased the expression of MHC class II on the surface of HUVEC at each time-point (2.8 ± 1.1-fold, 5.5 ± 1.4-fold, 17.2 ± 6.0-fold, at 24 h, 48 h and 72 h, respectively), CNP failed to significantly alter these responses. Figure 1. ICAM-1 expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for ICAM-1 after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which range from 820 to 3089; mean ± SEM) for ICAM-1 on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates ( p < 0.05, ** p < 0.01, significantly different from IFN-γ alone). Figure 1. ICAM-1 expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for ICAM-1 after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which range from 820 to 3089; mean ± SEM) for ICAM-1 on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, signiﬁcantly different from IFN-γ alone). Fi 1 ICAM 1 i i HUVEC ft i b ti ith IFN l i bi ti ith Figure 1. ICAM-1 expression in HUVEC after incubation with IFN-γ alone or in combinations with Figure 1. ICAM-1 expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for ICAM-1 after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. 3. Results 3. Results 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which range from 820 to 3089; mean ± SEM) for ICAM-1 on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, significantly different from IFN-γ alone). Figure 1. ICAM-1 expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for ICAM-1 after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which range from 820 to 3089; mean ± SEM) for ICAM-1 on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, signiﬁcantly different from IFN-γ alone). 5 of 12 5 of 12 Biosensors 2018, 8, 86 biosensors 2018, 18, x F Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for MHC-I after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 2079 to 6549; mean ± SEM) for MHC-I on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, significantly different from IFN-γ alone). Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for MHC-I after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 2079 to 6549; mean ± SEM) for MHC-I on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, signiﬁcantly different from IFN-γ alone). Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for MHC-I after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 2079 to 6549; mean ± SEM) for MHC-I on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, significantly different from IFN-γ alone). Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for MHC-I after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 2079 to 6549; mean ± SEM) for MHC-I on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, signiﬁcantly different from IFN-γ alone). Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for MHC-I after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 2079 to 6549; mean ± SEM) for MHC-I on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, significantly different from IFN-γ alone). Figure 2. MHC class I expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for MHC-I after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 2079 to 6549; mean ± SEM) for MHC-I on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates (* p < 0.05, ** p < 0.01, signiﬁcantly different from IFN-γ alone). 6 of 12 6 of 12 Biosensors 2018, 8, 86 biosensors 2018 18 x F Figure 3. MHC class II expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for MHC-II after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 187 to 216; mean ± SEM) for MHC-II on untreated HUVEC or after 24 h (B), 48 h (C) 72 h (D) t t t ith IFN l i bi ti ith CNP 3 HUVEC i l t Figure 3. MHC class II expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for MHC-II after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inﬂammatory Molecules on the Surface of HUVEC. 3.1. CNP Reduces IFN-γ Mediated Expression of Pro-Inflammatory Molecules on the Surface of HUVEC (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 187 to 216; mean ± SEM) for MHC-II on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates. Figure 3. MHC class II expression in HUVEC after incubation with IFN-γ alone or in combinations Figure 3. MHC class II expression in HUVEC after incubation with IFN-γ alone or in combinations with CNP for up to 72 h. (A) Representative flow cytometry histograms for MHC-II after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 187 to 216; mean ± SEM) for MHC-II on untreated HUVEC or after 24 h (B), g p γ with CNP for up to 72 h. (A) Representative ﬂow cytometry histograms for MHC-II after incubation of HUVEC alone (Red) or with IFN-γ (Blue), CNP (Orange) or IFN-γ and CNP (Green) for the indicated times. (B–D) Mean Fluorescence Intensity (expressed as fold increase over MFI of untreated cells, which ranged from 187 to 216; mean ± SEM) for MHC-II on untreated HUVEC or after 24 h (B), 48 h (C) or 72 h (D) treatment with IFN-γ alone or in combination with CNP; n = 3 HUVEC isolates. 48 h (C) or 72 h (D) treatment with IFN-γ 3.2. CNP Induces cGMP Release in HUVEC 3.2. CNP Induces cGMP Release in HUVEC CNP exerts the vast majority of its effects via the guanylyl cyclase B (GC-B) receptor and cGMP generation [33,34]. We therefore measured cGMP accumulation in HUVEC stimulated with 0 or 100 nm CNP in the presence of 1 mm IBMX. As shown (Figure 4A), CNP stimulated total cGMP accumulation in HUVEC isolates, to 2.5 ± 0.3-fold above basal (p = 0.015). To determine whether the observed effects of CNP on IFN-γ-induced ICAM-1 and MHC class I expression in HUVEC were mediated via cGMP, we used a cell permeable cGMP agonist (8-Br-cGMP) to mimic the observed cGMP increase in response to CNP. As shown (Figure 4B,C), 1 mm 8-Br-cGMP caused an identical inhibition of the IFN-γ-mediated increases in ICAM-1 (7.6 ± 1.04-fold vs. 2.4 ± 0.4-fold, p < 0.001) and MHC class I expression (6.6 ± 0.8-fold vs. 2.5 ± 0.5-fold, p < 0.001). However, similar to CNP, 8-Br- cGMP failed to alter the effects of IFN-γ on MHC class II expression (Figure 4D). 8-CPT-cGMP also reduced IFN-γ mediated increase in MHC class I (Supplementary Figure S1). Collectively, these data CNP exerts the vast majority of its effects via the guanylyl cyclase B (GC-B) receptor and cGMP generation [33,34]. We therefore measured cGMP accumulation in HUVEC stimulated with 0 or 100 nm CNP in the presence of 1 mm IBMX. As shown (Figure 4A), CNP stimulated total cGMP accumulation in HUVEC isolates, to 2.5 ± 0.3-fold above basal (p = 0.015). To determine whether the observed effects of CNP on IFN-γ-induced ICAM-1 and MHC class I expression in HUVEC were mediated via cGMP, we used a cell permeable cGMP agonist (8-Br-cGMP) to mimic the observed cGMP increase in response to CNP. As shown (Figure 4B,C), 1 mm 8-Br-cGMP caused an identical inhibition of the IFN-γ-mediated increases in ICAM-1 (7.6 ± 1.04-fold vs. 2.4 ± 0.4-fold, p < 0.001) and MHC class I expression (6.6 ± 0.8-fold vs. 2.5 ± 0.5-fold, p < 0.001). However, similar to CNP, 8-Br-cGMP failed to alter the effects of IFN-γ on MHC class II expression (Figure 4D). 8-CPT-cGMP also reduced IFN-γ mediated increase in MHC class I (Supplementary Figure S1). Collectively, these data support a role for GC-B/cGMP signaling in attenuating the effects of IFN-γ in HUVEC. 48 h (C) or 72 h (D) treatment with IFN-γ 3.2. CNP Induces cGMP Release in HUVEC (B–D) Effect of 8-bromo-cGMP on IFN-γ mediated ICAM-1 (B), MHC-I (C), and MHC-II (D) expressed as fold increase over MFI of untreated cells expression after 48hr treatment (which ranged from 572 to 819, 1313 to 2012, 176 to 230, for ICAM-1, MHC-I and MHC-II, respectively). n = 3 HUVEC isolates (* p < 0.05, *** p < 0.001, signiﬁcantly different from IFN-γ alone). 48 h (C) or 72 h (D) treatment with IFN-γ 3.2. CNP Induces cGMP Release in HUVEC (B–D) Effect of 8-bromo-cG on IFN-γ mediated ICAM-1 (B), MHC-I (C), and MHC-II (D) expressed as fold increase over MF untreated cells expression after 48hr treatment (which ranged from 572 to 819, 1313 to 2012, 17 230, for ICAM-1, MHC-I and MHC-II, respectively). n = 3 HUVEC isolates (* p < 0.05, *** p < 0. significantly different from IFN-γ alone). Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimulated cGMP accumulation in HUVEC isolates, after 30 min stimulation with 100 nm CNP in the presence of 1 mm IBMX. Data shown are means ± SEM pooled from 6 independent isolates (n = 6), each performed in duplicate (* p = 0.015, signiﬁcantly different from Basal). (B–D) Effect of 8-bromo-cGMP on IFN-γ mediated ICAM-1 (B), MHC-I (C), and MHC-II (D) expressed as fold increase over MFI of untreated cells expression after 48hr treatment (which ranged from 572 to 819, 1313 to 2012, 176 to 230, for ICAM-1, MHC-I and MHC-II, respectively). n = 3 HUVEC isolates (* p < 0.05, *** p < 0.001, signiﬁcantly different from IFN-γ alone). Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimula cGMP accumulation in HUVEC isolates, after 30 min stimulation with 100 nm CNP in the prese of 1 mm IBMX. Data shown are means ± SEM pooled from 6 independent isolates (n = 6), e performed in duplicate (* p = 0.015, significantly different from Basal). (B–D) Effect of 8-bromo-cG on IFN-γ mediated ICAM-1 (B), MHC-I (C), and MHC-II (D) expressed as fold increase over MF untreated cells expression after 48hr treatment (which ranged from 572 to 819, 1313 to 2012, 17 230, for ICAM-1, MHC-I and MHC-II, respectively). n = 3 HUVEC isolates (* p < 0.05, *** p < 0. significantly different from IFN-γ alone). Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimulated cGMP accumulation in HUVEC isolates, after 30 min stimulation with 100 nm CNP in the presence of 1 mm IBMX. Data shown are means ± SEM pooled from 6 independent isolates (n = 6), each performed in duplicate (* p = 0.015, signiﬁcantly different from Basal). 48 h (C) or 72 h (D) treatment with IFN-γ 3.2. CNP Induces cGMP Release in HUVEC 7 of 12 7 o Biosensors 2018, 8, 86 7 of 12 biosensors 2018, 18, x FOR PEER REVIEW 7 of Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimulated cGMP accumulation in HUVEC isolates, after 30 min stimulation with 100 nm CNP in the presence of 1 mm IBMX. Data shown are means ± SEM pooled from 6 independent isolates (n = 6), each performed in duplicate (* p = 0.015, significantly different from Basal). (B–D) Effect of 8-bromo-cGMP on IFN-γ mediated ICAM-1 (B), MHC-I (C), and MHC-II (D) expressed as fold increase over MFI of untreated cells expression after 48hr treatment (which ranged from 572 to 819, 1313 to 2012, 176 to 230, for ICAM-1, MHC-I and MHC-II, respectively). n = 3 HUVEC isolates (* p < 0.05, *** p < 0.001, significantly different from IFN-γ alone) Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimulated cGMP accumulation in HUVEC isolates, after 30 min stimulation with 100 nm CNP in the presence of 1 mm IBMX. Data shown are means ± SEM pooled from 6 independent isolates (n = 6), each performed in duplicate (* p = 0.015, signiﬁcantly different from Basal). (B–D) Effect of 8-bromo-cGMP on IFN-γ mediated ICAM-1 (B), MHC-I (C), and MHC-II (D) expressed as fold increase over MFI of untreated cells expression after 48hr treatment (which ranged from 572 to 819, 1313 to 2012, 176 to 230, for ICAM-1, MHC-I and MHC-II, respectively). n = 3 HUVEC isolates (* p < 0.05, *** p < 0.001, signiﬁcantly different from IFN-γ alone). Biosensors 2018, 8, 86 biosensors 2018, Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimul Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimulated Figure 4. cGMP mediated effects on cell surface receptor expression in HUVEC. (A) CNP-stimula cGMP accumulation in HUVEC isolates, after 30 min stimulation with 100 nm CNP in the prese of 1 mm IBMX. Data shown are means ± SEM pooled from 6 independent isolates (n = 6), e performed in duplicate (* p = 0.015, significantly different from Basal). 4 Discussion 4. Discussion 4. Discussion Here, we have shown that CNP, a natriuretic peptide shown to be broadly cardioprotective [35] and known to be highly expressed by the endothelium [36] is able to downregulate IFN-γ-mediated gene expression in human endothelial cells in vitro. In particular, we have shown that CNP significantly inhibited the IFN-γ increase in ICAM-1 expression on the cell surface of HUVEC. ICAM- 1 is an adhesion molecule involved in firm adhesion and trans-endothelial migration of leukocytes including neutrophils, monocytes and both T and B cells, and is also important for immunological synapse formation during T cell activation (for review see [37]), so by downregulating expression to basal levels CNP may prevent leukocyte accumulation during vascular inflammation, that could lead to the early stages of fatty streak formation. Likewise, by reducing ICAM-1 expression on the endothelium, immunological synapse formation will be compromised which could limit T cell accumulation. Thus, development of the pro-inflammatory environment leading to lesion formation will be reduced by some degree. ICAM-1 is a ligand for β-2 integrin family members, including αx/β2 (CD11c/CD18), the α-subunit (CD11c) having recently been shown to be upregulated in IDO/ApoE−/− mice [38]. However, VCAM-1 has been shown to be an alternative ligand for CD11c [39], which Polyzos et al. [40] recently demonstrated was upregulated in ApoE−/− mice and aortic endothelial cells treated chronically with 1-MT to inhibit IDO. Polyzos et al. [40] also showed that CCL2 was increased. Taken together these could explain the increased macrophage accumulation after IDO blockade in these two animal models. Interestingly, CNP has previously been shown to downregulate secretion of CCL2 from THP-1 human macrophage cell line in vitro [41] and there are differences in expression patterns of VCAM-1 in murine and human endothelial cells [7,42]. Future work should examine the effect of CNP on other IFNγ-mediated responses including expression of VCAM-1 and CCL2, or the Here, we have shown that CNP, a natriuretic peptide shown to be broadly cardioprotective [35] and known to be highly expressed by the endothelium [36] is able to downregulate IFN-γ-mediated gene expression in human endothelial cells in vitro. In particular, we have shown that CNP signiﬁcantly inhibited the IFN-γ increase in ICAM-1 expression on the cell surface of HUVEC. 3 3 CNP Inhibits IFN γ Induced Upregulation of Tryptophan Metabol 3.3. CNP Inhibits IFN-γ Induced Upregulation of Tryptophan Metabolism 3.3. CNP Inhibits IFN γ Induced Upregulation of Tryptophan Metabolism IFN-γ has previously been shown to upregulate expression of indolamine 2,3-dioxygen (IDO), an inducible enzyme found at sites of immune privilege and thought to be important exerting the anti-microbial effects of IFN-γ [11]. The enzyme works to reduce tryptophan availabi catalyzing the first step of the pathway to kynurenine, which can be measured by mass spectrom in vivo in plasma or in vitro in cell culture medium. As shown in Figure 5, IFN-γ stimulation indu a significant increase in kynurenine in the culture supernatant and this was significantly reduced co-treatment with CNP after 48 h (100.0 ± 5.3% vs. 24.5 ± 2.4%, p < 0.01). This suggests that as we potentially downregulating leukocyte adhesion and trans-endothelial migration during endothe inflammation it may have an opposing effect of maintaining function of those leukocytes which able to adhere and migrate at inflammatory sites in the vasculature IFN-γ has previously been shown to upregulate expression of indolamine 2,3-dioxygenase (IDO), an inducible enzyme found at sites of immune privilege and thought to be important for exerting the anti-microbial effects of IFN-γ [11]. The enzyme works to reduce tryptophan availability, catalyzing the ﬁrst step of the pathway to kynurenine, which can be measured by mass spectrometry in vivo in plasma or in vitro in cell culture medium. As shown in Figure 5, IFN-γ stimulation induced a signiﬁcant increase in kynurenine in the culture supernatant and this was signiﬁcantly reduced by co-treatment with CNP after 48 h (100.0 ± 5.3% vs. 24.5 ± 2.4%, p < 0.01). This suggests that as well as potentially downregulating leukocyte adhesion and trans-endothelial migration during endothelial inﬂammation it may have an opposing effect of maintaining function of those leukocytes which are able to adhere and migrate at inﬂammatory sites in the vasculature. 8 of 12 8 of 12 Biosensors 2018, 8, 86 biosensors 2018, 18, x F Figure 5. CNP downregulates IFN-γ-mediated tryptophan metabolism in HUVEC. (A) Representative chromatogram of kynurenine production following IFN-γ treatment. (B) Kynurenine production from untreated HUVEC or after 24 h and 48 h treatment with IFN-γ alone or in combination with CNP. Data shown are representative from a single HUVEC isolate, expressed as % IFN-γ -stimulated kyunurenine production), and performed in duplicate. ( p < 0.01, significantly different from IFN-γ alone). Figure 5. CNP downregulates IFN-γ-mediated tryptophan metabolism in HUVEC. 4 Discussion 4. Discussion ICAM-1 is an adhesion molecule involved in ﬁrm adhesion and trans-endothelial migration of leukocytes including neutrophils, monocytes and both T and B cells, and is also important for immunological synapse formation during T cell activation (for review see [37]), so by downregulating expression to basal levels CNP may prevent leukocyte accumulation during vascular inﬂammation, that could lead to the early stages of fatty streak formation. Likewise, by reducing ICAM-1 expression on the endothelium, immunological synapse formation will be compromised which could limit T cell accumulation. Thus, development of the pro-inﬂammatory environment leading to lesion formation will be reduced by some degree. ICAM-1 is a ligand for β-2 integrin family members, including αx/β2 (CD11c/CD18), the α-subunit (CD11c) having recently been shown to be upregulated in IDO/ApoE−/−mice [38]. However, VCAM-1 has been shown to be an alternative ligand for CD11c [39], which Polyzos et al. [40] recently demonstrated was upregulated in ApoE−/−mice and aortic endothelial cells treated chronically with 1-MT to inhibit IDO. Polyzos et al. [40] also showed that CCL2 was increased. Taken together these could explain the increased macrophage accumulation after IDO blockade in these two animal models. Interestingly, CNP has previously been shown to downregulate secretion of CCL2 from THP-1 human macrophage cell line in vitro [41] and there are differences in expression patterns of VCAM-1 in murine and human endothelial cells [7,42]. Future work should examine the effect of CNP on other IFNγ-mediated responses including expression of VCAM-1 and CCL2, or the effects of CNP on leukocyte adhesion and trans-endothelial migration. effects of CNP on leukocyte adhesion and trans-endothelial migration. We have also demonstrated that CNP reduced IFN-γ-mediated increases in MHC I expression. This could additionally contribute to a brake on CD8+ T cell immune responses in the inflamed environment. Cytotoxic T cells have been identified in advanced atherosclerotic lesions and may contribute up to 50% of the lymphocyte population [43]. Activated CD8+ T cells have been shown to efficiently migrate into the intima of both healthy and diseased arteries in itro [43] Interestingly we We have also demonstrated that CNP reduced IFN-γ-mediated increases in MHC I expression. This could additionally contribute to a brake on CD8+ T cell immune responses in the inﬂamed environment. Cytotoxic T cells have been identiﬁed in advanced atherosclerotic lesions and may contribute up to 50% of the lymphocyte population [43]. 3 3 CNP Inhibits IFN γ Induced Upregulation of Tryptophan Metabol 3.3. CNP Inhibits IFN-γ Induced Upregulation of Tryptophan Metabolism (A) Representative chromatogram of kynurenine production following IFN-γ treatment. (B) Kynurenine production from untreated HUVEC or after 24 h and 48 h treatment with IFN-γ alone or in combination with CNP. Data shown are representative from a single HUVEC isolate, expressed as % IFN-γ -stimulated kyunurenine production), and performed in duplicate. ( p < 0.01, signiﬁcantly different from IFN-γ alone). Figure 5. CNP downregulates IFN-γ-mediated tryptophan metabolism in HUVEC. (A) Representative chromatogram of kynurenine production following IFN-γ treatment. (B) Kynurenine production from untreated HUVEC or after 24 h and 48 h treatment with IFN-γ alone or in combination with CNP. Data shown are representative from a single HUVEC isolate, expressed as % IFN-γ -stimulated kyunurenine production), and performed in duplicate. ( p < 0.01, significantly different from IFN-γ alone). Figure 5. CNP downregulates IFN-γ-mediated tryptophan metabolism in HUVEC. (A) Representative chromatogram of kynurenine production following IFN-γ treatment. (B) Kynurenine production from untreated HUVEC or after 24 h and 48 h treatment with IFN-γ alone or in combination with CNP. Data shown are representative from a single HUVEC isolate, expressed as % IFN-γ -stimulated kyunurenine production), and performed in duplicate. ( p < 0.01, signiﬁcantly different from IFN-γ alone). 4 Discussion 4. Discussion Activated CD8+ T cells have been shown to efﬁciently migrate into the intima of both healthy and diseased arteries in vitro [43]. Interestingly, 9 of 12 9 of 12 Biosensors 2018, 8, 86 we did not see an effect of CNP on IFN-γ induced MHC class II expression, suggesting that CNP exerts a selective inhibition of CD8 T cell/NK cell accumulation, whilst allowing CD4+ T cell activity. There is evidence that certain subsets of CD4+ T cells (Th2-like, Treg) may be beneﬁcial in reducing atherosclerotic lesion formation [3], thus it could be argued that a selective reduction in CD8 vs. CD4 T cell accumulation via CNP’s reduction in gene expression mediated by IFN-γ is an early protective response by the endothelium. CNP exerts the vast majority of its effects via the GC-B receptor, and the generation of cGMP [33,34]. In our current study, not only did we conﬁrm the presence of functional GC-B receptors in HUVEC, as described previously [44], but were able to mimic the dampening effects of CNP on IFN-γ responsiveness by using the cell permeable analogue, 8-Br-cGMP and with an additional cGMP analogue 8-CPT-cGMP, which is a more membrane permeant molecule. Interestingly, 8-CPT-cGMP only partially recapitulated the effects on MHC-I and ICAM-I seen in the presence of either CNP or 8-Br-cGMP. It is unclear as to why 8-CPT-cGMP failed to inhibit INF-y-stimulated ICAM-1 expression in HUVECs. However, previous studies suggest that the sub-cellular localisation of cGMP generation in cardiomyocytes and HUVEC can inﬂuence biological responsiveness [44,45]. Therefore, it is possible that CNP, 8-Br-cGMP and 8-CPT-cGMP treatments lead to spatially distinct increases in cGMP within HUVECs that may alter endothelial cell responsiveness. HUVEC also express the NPR-C receptor, which has been strongly implicated in the role that CNP performs to maintain vascular homeostasis [46]. Even though NPR-C lacks intrinsic guanylyl cyclase and, therefore, does not directly enhance cGMP production, previous studies have shown that activation of NPR-C by natriuretic peptides may also stimulate nitric oxide production, leading to cGMP generation via soluble guanylyl cyclase [47]. Therefore, although our data implicate GC-B/cGMP signaling as a mechanism for the anti-inﬂammatory effects of CNP in HUVEC, it is perfectly conceivable that additional, NPR-C-mediated mechanisms may also contribute to these effects. We have used HUVEC as model human endothelial cells in this study because umbilical cords are a readily available, ethically uncontroversial source of human vascularized tissue. 4 Discussion 4. Discussion HUVEC have been used for more than 40 years and their responses to a range of stimuli including IFN-γ are very well characterized and have been shown to be comparable to adult human large artery endothelium [48]. In conclusion, CNP is able to downregulate IFN-γ-mediated gene expression in the endothelium, which could limit vascular inﬂammation by directing speciﬁc T cell subsets into developing atherosclerotic lesions, ultimately affecting atheroma progression. Further work is required to determine whether modulation of endogenously expressed endothelial CNP or addition of exogenous CNP is of potential therapeutic value to treat atherosclerosis in vivo. Supplementary Materials: The following are available online at http://www.mdpi.com/2079-6374/8/3/86/s1, Figure S1: ICAM-1 and MHC Class I expression in HUVEC after incubation with IFN-γ alone or in combinations with 8-CPT cGMP for 48 h. Author Contributions: Conceptualization C.L., R.F.; experimental data and analysis A.D., Z.J., C.H., B.S., C R.F.; preparation of the manuscript C.L., R.F. Author Contributions: Conceptualization C.L., R.F.; experimental data and analysis A.D., Z.J., C.H., B.S., C.L., R.F.; preparation of the manuscript C.L., R.F. Funding: BBSRC Project grant BBD0015601 awarded to RCF. Conﬂicts of Interest: The authors declare no conﬂict of interest. 1. Mozaffarian, D.; Benjamin, E.J.; Go, A.S.; Arnett, D.K.; Blaha, M.J.; Cushman, M.D.; Das, S.R.; De Ferranti, S.; Cushman, M.; Despr’s, J.P.; et al. Executive Summary: Heart Disease and Stroke Statistics-2016 Update: A Report From the American Heart Association. Circulation 2016, 133, 447–454. [CrossRef] [PubMed] 2. Benjamin, E.J.; Virani, S.S.; Callaway, C.W.; Chamberlain, A.M.; Chang, A.R.; Cheng, S.; Chiuve, S.E.; Cushman, M.; Delling, F.N.; Deo, R. Heart Disease and Stroke Statistics-2018 Update: A Report From the American Heart Association. Circulation 2018, 137, e67–e492. [CrossRef] [PubMed] References 1. Mozaffarian, D.; Benjamin, E.J.; Go, A.S.; Arnett, D.K.; Blaha, M.J.; Cushman, M.D.; Das, S.R.; De Ferranti, S.; Cushman, M.; Despr’s, J.P.; et al. Executive Summary: Heart Disease and Stroke Statistics-2016 Update: A Report From the American Heart Association. Circulation 2016, 133, 447–454. [CrossRef] [PubMed] 2. Benjamin, E.J.; Virani, S.S.; Callaway, C.W.; Chamberlain, A.M.; Chang, A.R.; Cheng, S.; Chiuve, S.E.; Cushman, M.; Delling, F.N.; Deo, R. Heart Disease and Stroke Statistics-2018 Update: A Report From the American Heart Association. Circulation 2018, 137, e67–e492. [CrossRef] [PubMed] 10 of 12 Biosensors 2018, 8, 86 10 of 12 3. Hansson, G.K.; Libby, P.; Tabas, I. Inﬂammation and plaque vulnerability. J. Intern. Med. 2015, 278, 483–493. [CrossRef] [PubMed] 4. Martin, F.A.; Murphy, R.P.; Cummins, P.M. Thrombomodulin and the vascular endothelium: Insights into functional, regulatory, and therapeutic aspects. Am. J. Physiol. Heart Circ. Physiol. 2013, 304, H1585–H1597. [CrossRef] [PubMed] 5. Ketelhuth, D.F.; Hansson, G.K. Cellular immunity, low-density lipoprotein and atherosclerosis: Break of tolerance in the artery wall. Thromb. Haemost. 2011, 106, 779–786. [CrossRef] [PubMed] 6. Thornhill, M.H.; Wellicome, S.M.; Mahiouz, D.L.; Lanchbury, J.S.S.; Kyan-Aung, U.; Haskard, D.O. Tumor necrosis factor combines with IL-4 or IFN-gamma to selectively enhance endothelial cell adhesiveness for T cells. The contribution of vascular cell adhesion molecule-1-dependent and -independent binding mechanisms. J. Immunol. 1991, 146, 592–598. [PubMed] 7. Lidington, E.A.; Moyes, D.L.; McCormack, A.M.; Rose, M.L. A comparison of primary endothelial cells and endothelial cell lines for studies of immune interactions. Transpl. Immunol. 1999, 7, 239–246. [CrossRef] 8. Mbongue, J.C.; Nicholas, D.A.; Torrez, T.W.; Kim, N.-S.; Firek, A.F.; Langridge, W.H.R. The Role of Indoleamine 2,3-Dioxygenase in Immune Suppression and Autoimmunity. Vaccines 2015, 3, 703–729. [CrossRef] [PubMed] . Mellor, A.L.; Keskin, D.B.; Johnson, T.; Chandler, P.; Munn, D.H. Cells expressing indoleam 2,3-dioxygenase inhibit T cell responses. J. Immunol. 2002, 168, 3771–3776. [CrossRef] [PubMed] 10. Mellor, A.L.; Munn, D.H. Tryptophan catabolism and T-cell tolerance: Immunosuppression by starvation? Immunol. Today 1999, 20, 469–473. [CrossRef] 11. Taylor, M.W.; Feng, G.S. Relationship between interferon-gamma, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J. 1991, 5, 2516–2522. [CrossRef] [PubMed] 12. Curry, F.-R.E.; Clark, J.F.; Jiang, Y.; Kim, M.-H.; Adamson, R.H.; Simon, S.I. The role of atrial natriuretic peptide to attenuate inﬂammation in a mouse skin wound and individually perfused rat mesenteric microvessels. Physiol. Rep. 2016, 4, e12968. [CrossRef] [PubMed] 13. Khoury, E.E.; Kinaneh, S.; Aronson, D.; Amir, O.; Ghanim, D.; Volinsky, N.; Azzam, Z.; Abassi, Z. References Natriuretic peptides system in the pulmonary tissue of rats with heart failure: Potential involvement in lung edema and inﬂammation. Oncotarget 2018, 9, 21715–21730. [CrossRef] [PubMed] g 14. Kuhn, M. Endothelial actions of atrial and B-type natriuretic peptides. Br. J. Pharmacol. 2012, 166, 522–531. [CrossRef] [PubMed] 15. Okamoto, A.; Nojiri, T.; Konishi, K.; Tokudome, T.; Miura, K.; Hosoda, H.; Hino, J.; Miyazato, M.; Kyomoto, Y.; Asai, K. Atrial natriuretic peptide protects against bleomycin-induced pulmonary ﬁbrosis via vascular endothelial cells in mice: ANP for pulmonary ﬁbrosis. Respir. Res. 2017, 18, 1. [CrossRef] [PubMed] 16. Sudoh, T.; Minamino, N.; Kangawa, K.; Matsuo, H. C-type natriuretic peptide (CNP): A new member of natriuretic peptide family identiﬁed in porcine brain. Biochem. Biophys. Res. Commun. 1990, 168, 863–870. [CrossRef] 17. Ahluwalia, A.; Hobbs, A.J. Endothelium-derived C-type natriuretic peptide: More than just a hyperpolarizing factor. Trends Pharmacol. Sci. 2005, 26, 162–167. [CrossRef] [PubMed] 18. Scotland, R.S.; Ahluwalia, A.; Hobbs, A.J. C-type natriuretic peptide in vascular physiology and disease. Pharmacol. Ther. 2005, 105, 85–93. [CrossRef] [PubMed] 19. Izumiya, Y.; Araki, S.; Usuku, H.; Rokutanda, T.; Hanatani, S.; Ogawa, H. Chronic C-Type Natriuretic Peptide Infusion Attenuates Angiotensin II-Induced Myocardial Superoxide Production and Cardiac Remodeling. Int. J. Vasc. Med. 2012, 2012, 246058. [CrossRef] [PubMed] 20. Scotland, R.S.; Cohen, M.; Foster, P.; Lovell, M.; Mathur, A.; Ahluwalia, A.; Hobbs, A.J. C-type natriuretic peptide inhibits leukocyte recruitment and platelet-leukocyte interactions via suppression of P-selectin expression. Proc. Natl. Acad. Sci. USA 2005, 102, 14452–14457. [CrossRef] [PubMed] 21. Qian, J.; Haruno, A.; Asada, Y.; Nishida, T.; Saito, Y.; Matsuda, T.; Ueno, H. Local expression of C-type natriuretic peptide suppresses inﬂammation, eliminates shear stress-induced thrombosis, and prevents neointima formation through enhanced nitric oxide production in rabbit injured carotid arteries. Circ. Res. 2002, 91, 1063–1069. [CrossRef] [PubMed] 22. Hu, P.; Wang, J.; Hu, B.; Lu, L.; Xuan, Q.; Qin, Y. Increased urinary C-type natriuretic peptide excretion may be an early marker of renal tubulointerstitial ﬁbrosis. Peptides 2012, 37, 98–105. [CrossRef] [PubMed] 11 of 12 11 of 12 Biosensors 2018, 8, 86 23. Li, Z.; Liu, Y.; Li, G.; Li, B.; Liu, Y.; Li, X.; Liu, A. Inhibitory effects of C-type natriuretic peptide on the differentiation of cardiac ﬁbroblasts, and secretion of monocyte chemoattractant protein-1 and plasminogen activator inhibitor-1. Mol. Med. Rep. 2015, 11, 159–165. [CrossRef] [PubMed] 24. Ichiki, T.; Boerrigter, G.; Huntley, B.K.; Sangaralingham, S.J.; McKie, P.M.; Harty, G.J.; Harders, G.E.; Burnett, J.C., Jr. References Differential expression of the pro-natriuretic peptide convertases corin and furin in experimental heart failure and atrial ﬁbrosis. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2013, 304, R102–R109. [CrossRef] [PubMed] 25. Burley, D.S.; Hamid, S.A.; Baxter, G.F. Cardioprotective actions of peptide hormones in myocardial ischemia. Heart Fail. Rev. 2007, 12, 279–291. [CrossRef] [PubMed] 26. Soeki, T.; Kishimoto, I.; Okumura, H.; Tokudome, T.; Horio, T.; Mori, K.; Kangawa, K. C-type natriuretic peptide, a novel antiﬁbrotic and antihypertrophic agent, prevents cardiac remodeling after myocardial infarction. J. Am. Coll Cardiol. 2005, 45, 608–616. [CrossRef] [PubMed] 27. Ichiki, T.; Schirger, J.A.; Huntley, B.K.; Brozovich, F.V.; Maleszewski, J.J.; Sandberg, S.M.; Sangaralingham, S.J.; Park, S.J.; Burnett, J.C., Jr. Cardiac ﬁbrosis in end-stage human heart failure and the cardiac natriuretic peptide guanylyl cyclase system: Regulation and therapeutic implications. J. Mol. Cell Cardiol. 2014, 75, 199–205. [CrossRef] [PubMed] 28. Sangaralingham, S.J.; McKie, P.M.; Ichiki, T.; Scott, C.G.; Heublein, D.M.; Chen, H.H.; Bailey, K.R.; Redﬁeld, M.M.; Rodeheffer, R.J.; Burnett, J.C., Jr. Circulating C-type natriuretic peptide and its relationship to cardiovascular disease in the general population. Hypertension 2015, 65, 1187–1194. [CrossRef] [PubMed] 29. Jaffe, E.A.; Nachman, R.L.; Becker, C.G.; Minick, C.R. Culture of human endothelial cells derived from umbilical veins. Identiﬁcation by morphologic and immunologic criteria. J. Clin. Investing 1973, 52, 2745–2756. 28. Sangaralingham, S.J.; McKie, P.M.; Ichiki, T.; Scott, C.G.; Heublein, D.M.; Chen, H.H.; Bailey, K.R.; Redﬁeld, M.M.; Rodeheffer, R.J.; Burnett, J.C., Jr. Circulating C-type natriuretic peptide and its relationship to cardiovascular disease in the general population. Hypertension 2015, 65, 1187–1194. [CrossRef] [PubMed] 29. Jaffe, E.A.; Nachman, R.L.; Becker, C.G.; Minick, C.R. Culture of human endothelial cells derived from umbilical veins. Identiﬁcation by morphologic and immunologic criteria. J. Clin. Investing 1973, 52, 2745–2756. [CrossRef] [PubMed] 30. Lawson, C.; Ainsworth, M.; Yacoub, M.; Rose, M. Ligation of ICAM-1 on endothelial cells leads to expression of VCAM-1 via a nuclear factor-kappaB-independent mechanism. J. Immunol. 1999, 162, 2990–2996. [PubMed] 31. Pober, J.S.; Gimbrone, M.A.; Lapierre, L.A.; Mendrick, D.; Fiers, W.; Rothlein, R.; Springer, T. A Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J. Immunol. 1986, 137, 1893–1896. [PubMed] 32. Thornhill, M.H.; Haskard, D.O. IL-4 regulates endothelial cell activation by IL-1, tumor necrosis factor, or IFN-gamma. J. Immunol. 1990, 145, 865–872. [PubMed] 33. Del Ry, S. C-type natriuretic peptide: A new cardiac mediator. Peptides 2013, 40, 93–98. [CrossRef] [P 33. Del Ry, S. References C-type natriuretic peptide: A new cardiac mediator. Peptides 2013, 40, 93–98. [CrossRef] [PubMed] 34. Potter, L.R. Regulation and therapeutic targeting of peptide-activated receptor guanylyl cyclases. Pharmacol. Ther. 2011, 130, 71–82. [CrossRef] [PubMed] 34. Potter, L.R. Regulation and therapeutic targeting of peptide-activated receptor guanylyl cyclases. Pharmacol. Ther. 2011, 130, 71–82. [CrossRef] [PubMed] 35. Hobbs, A.; Foster, P.; Prescott, C.; Scotland, R.; Ahluwalia, A. Natriuretic peptide receptor-C regulates coronary blood ﬂow and prevents myocardial ischemia/reperfusion injury: Novel cardioprotective role for endothelium-derived C-type natriuretic peptide. Circulation 2004, 110, 1231–1235. [CrossRef] [PubMed] 36. Chauhan, S.D.; Nilsson, H.; Ahluwalia, A.; Hobbs, A.J. Release of C-type natriuretic peptide accounts for the biological activity of endothelium-derived hyperpolarizing factor. Proc. Natl. Acad. Sci. USA 2003, 100, 1426–1431. [CrossRef] [PubMed] [ ] [ ] 37. Lawson, C.; Wolf, S. ICAM-1 signaling in endothelial cells. Pharmacol. Rep. 2009, 61, 22–32. [CrossRef] 38. Cole, J.E.; Astola, N.; Cribbs, A.P.; Goddard, M.E.; Park, I.; Green, P.; Davies, A.H.; Williams, R.O.; Feldmann, M.; Monaco, C. Indoleamine 2,3-dioxygenase-1 is protective in atherosclerosis and its metabolites provide new opportunities for drug development. Proc. Natl. Acad. Sci. USA 2015, 112, 13033–13038. [CrossRef] [PubMed] 39. Sadhu, C.; Ting, H.J.; Lipsky, B.; Hensley, K.; Garcia-Martinez, L.F.; Simon, S.I.; Staunton, D.E. CD11c/CD18: Novel ligands and a role in delayed-type hypersensitivity. J. Leukoc. Biol. 2007, 81, 1395–1403. [CrossRef] [PubMed] 40. Polyzos, K.A.; Ovchinnikova, O.; Berg, M.; Baumgartner, R.; Agardh, H.; Pirault, J.; Gister, A.; Assinger, A.; Laguna-Fernandez, A.; Back, M. Inhibition of indoleamine 2,3-dioxygenase promotes vascular inﬂammation and increases atherosclerosis in Apoe-/- mice. Cardiovasc. Res. 2015, 106, 295–302. [CrossRef] [PubMed] 41. Osawa, H.; Yamabe, H.; Kaizuka, M.; Tamura, N.; Tsunoda, S.; Baba, Y.; Shirato, K.; Tateyama, F.; Okumura, K. C-Type natriuretic peptide inhibits proliferation and monocyte chemoattractant protein-1 secretion in cultured human mesangial cells. Nephron 2000, 86, 467–472. [CrossRef] [PubMed] 12 of 12 Biosensors 2018, 8, 86 42. Dansky, H.M.; Barlow, C.B.; Lominska, C.; Sikes, J.L.; Kao, C.; Weinsaft, J.; Cybulsky, M.I.; Smith, J.D. Adhesion of monocytes to arterial endothelium and initiation of atherosclerosis are critically dependent on vascular cell adhesion molecule-1 gene dosage. Arterioscler. Thromb. Vasc. Biol. 2001, 21, 1662–1667. [CrossRef] [PubMed] 43. Gewaltig, J.; Kummer, M.; Koella, C.; Cathomas, G.; Biedermann, B.C. Requirements for CD8 T-cell migration into the human arterial wall. Hum. Pathol. 2008, 39, 1756–1762. [CrossRef] [PubMed] 44. Rautureau, Y.; Gowers, I.; Wheeler-Jones, C.P.D.; Baxter, G.F. References C-type natriuretic peptide regulation of guanosine-3′,5′-cyclic monophosphate production in human endothelial cells. Auton. Autacoid Pharmacol. 2010, 30, 185–192. [CrossRef] [PubMed] 45. Subramanian, H.; Froese, A.; Jonsson, P.; Schmidt, H.; Gorelik, J.; Nikolaev, V.O. Distinct submembrane localisation compartmentalises cardiac NPR1 and NPR2 signalling to cGMP. Nat. Commun. 2018, 9, 2446. [CrossRef] [PubMed] 46. Moyes, A.J.; Khambata, R.S.; Villar, I.; Bubb, K.J.; Baliga, R.S.; Lumsden, N.G.; Xiao, F.; Gane, P.J.; Rebstock, A.-S.; Worthington, R.J.; et al. Endothelial C-type natriuretic peptide maintains vascular homeostasis. J. Clin. Investing 2014, 124, 4039–4051. [CrossRef] [PubMed] 47. Costa, M.A.; Elesgaray, R.; Balaszczuk, A.M.; Arranz, C. Role of NPR-C natriuretic receptor in nitric oxide system activation induced by atrial natriuretic peptide. Regul. Pept. 2006, 135, 63–68. [CrossRef] [PubMed] 47. Costa, M.A.; Elesgaray, R.; Balaszczuk, A.M.; Arranz, C. Role of NPR-C natriuretic receptor in nitric oxide system activation induced by atrial natriuretic peptide. Regul. Pept. 2006, 135, 63–68. [CrossRef] [PubMed] 48. Mcdouall, R.M.; Batten, P.; Mccormack, A.; Yacoub, M.H.; Rose, M.L. MHC class II expression on human 48. Mcdouall, R.M.; Batten, P.; Mccormack, A.; Yacoub, M.H.; Rose, M.L. MHC class II expression on human heart microvascular endothelial cells: Exquisite sensitivity to interferon-gamma and natural killer cells. Transplantation 1997, 64, 1175–1180. [CrossRef] [PubMed] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
https://openalex.org/W4386285154	https://www.mdpi.com/1999-4923/15/9/2244/pdf?version=1693392548	English	null	β-Cyclodextrin Modified Hydrogels of Kappa-Carrageenan for Methotrexate Delivery	Pharmaceutics	2,023	cc-by	12,896	Article β-Cyclodextrin Modiﬁed Hydrogels of Kappa-Carrageenan for Methotrexate Delivery Nataliya Kochkina 1,* , Marianna Arinina 2 , Valery Kulichikhin 2 and Irina Terekhova 1,* ria Nikitina 1, Nataliya Kochkina 1,* , Marianna Arinina 2 , Valery Kulichikhin 2 and Irin 1 G.A. Krestov Institute of Solution Chemistry of RAS, 153045 Ivanovo, Russia 2 A.V. Topchiev Institute of Petrochemical Synthesis of RAS, 119991 Moscow, Russia * Correspondence: nek@isc-rus.ru (N.K.); ivt@isc-ras.ru (I.T.) Abstract: This work is aimed at developing a kappa-carrageenan (kCR) gel with increased methotrex- ate (MTX) content. β-Cyclodextrin (βCD), which is able to inclusion complex formation with MTX, has been used to increase the drug concentration in the hydrogel. The rheological behavior of the designed gels was investigated and the inﬂuence of MTX and βCD on the viscoelastic properties of kCR gel was studied in detail. The effect of βCD and its concentration on the MTX-releasing rate from the kCR gels was examined. The properties of kappa- and iota-carrageenans loaded with MTX were compared and the differences observed were explained in terms of different binding afﬁnities of MTX to these polymers. The obtained gels provided desirable viscoelastic properties useful for topical application. Keywords: carrageenan; methotrexate; cyclodextrin; gel Article Article 1. Introduction Pharmacological hydrogels are soft dosage forms which are used to treat a wide variety of diseases. Hydrogels are considered as three-dimensional polymeric networks insoluble in water due to the presence of chemical crosslinks (tie-points, junctions), or physical crosslinks, such as entanglements or crystallites [1]. These crosslinked polymer structures are capable of imbibing large amounts of water or aqueous solutions of biologically and pharmacologically active substances. From a pharmaceutical point of view, these hydrogels must be chemically and biologically safe as well as biocompatible and stable during storage, have speciﬁed rheological characteristics, and ensure drug bioavailability. Additionally, one of the modern requirements of hydrogels for biomedical and pharmaceutical usage is the ability to sustain release of the active pharmaceutical ingredient. This property of the gels gives the opportunity to increase the duration of the therapeutic effect and, consequently, to reduce the frequency of medicine application. pharmaceutics pharmaceutics pharmaceutics pharmaceutics Citation: Nikitina, M.; Kochkina, N.; Arinina, M.; Kulichikhin, V.; Terekhova, I. β-Cyclodextrin Modiﬁed Hydrogels of Kappa-Carrageenan for Methotrexate Delivery. Pharmaceutics 2023, 15, 2244. https://doi.org/10.3390/ pharmaceutics15092244 Among the polymers capable for gelling, carrageenans (CRs) are natural, non-toxic and biocompatible polysaccharides produced from red seaweeds [2,3]. CRs, being lin- ear sulphated galactans, exist in several different forms based on their sulfate content (Figure 1). Among the CR family, only kappa-carrageenan (kCR) and iota-carrageenan (iCR) possess gelling capability, suitable physicochemical properties, high water-holding capacity, and good adhesion to the skin and mucosal surfaces [3,4]. Moreover, they exhibit antiviral (against coronaviruses, dengue virus, herpes simplex virus, vesicular stomatitis virus, human immunodeﬁciency virus, inﬂuenza virus, human papillomavirus, etc.) and immunomodulatory activity [5,6]. The US Food and Drug Administration has generally recognized kCR and iCR as safe for topical application and consumption [6]. For instance, the efﬁcacy and safety of a CR-based gel to prevent human papillomavirus infection have been recently demonstrated by Laurie et al. [7]. Nasally administered iCR improved out- comes in COVID-19 patients [6]. Furthermore, CR gels can be used for transmucosal and transdermal delivery of biologically active substances, tissue engineering, and regenera- tive medicine [3–5,8–16]. For instance, triamcinolone acetonide has been loaded in kCR Received: 12 July 2023 Revised: 14 August 2023 Accepted: 25 August 2023 Published: 30 August 2023 The topical use of MTX may help to weaken or eliminate a number of serious side effects caused by the oral administration of MTX. Topical gels with MTX have been extensively studied [22–25]. There are some examples. Carbomer gels bearing MTX have showed improved topical delivery intended for effec- tive management of psoriasis [22,23]. MTX-loaded chitin [24] and chitosan/hyaluronan [25] nanogels have been formulated for its topical use in psoriasis Recently, we proposed iCR gels for methotrexate (MTX) delivery [20]. MTX (Figure 1) is a therapeutic agent having a wide range of applications, including treatment of oncologi- cal and autoimmune diseases [21]. The topical use of MTX may help to weaken or eliminate a number of serious side effects caused by the oral administration of MTX. Topical gels with MTX have been extensively studied [22–25]. There are some examples. Carbomer gels bearing MTX have showed improved topical delivery intended for effective management of psoriasis [22,23]. MTX-loaded chitin [24] and chitosan/hyaluronan [25] nanogels have been formulated for its topical use in psoriasis. [25] nanogels have been formulated for its topical use in psoriasis. This work, being a continuation of our previous study [20], focuses on kCR gels, which for the first time are proposed for MTX delivery. To the best of our knowledge, only one publication [26] concerning MTX-loaded magnetic kCR/chitosan hydrogels has been found in the literature. The MTX encapsulation efficiency in this gel has been in- creased by increasing its magnetite and chitosan contents [26] This work, being a continuation of our previous study [20], focuses on kCR gels, which for the ﬁrst time are proposed for MTX delivery. To the best of our knowledge, only one publication [26] concerning MTX-loaded magnetic kCR/chitosan hydrogels has been found in the literature. The MTX encapsulation efﬁciency in this gel has been increased by increasing its magnetite and chitosan contents [26]. creased by increasing its magnetite and chitosan contents [26]. Since MTX is poorly soluble in water [27–29], the employment of solubilizing agents is necessary to increase the aqueous solubility and, consequently, the content of MTX in kCR hydrogel. Cyclodextrins (CDs), being native cyclic oligosaccharides, are widely used as solubilizers due to their ability to include guest molecules into a hydrophobic inner cavity and form inclusion (or host–guest) complexes [30,31]. Among native CDs, only βCD exhibits a more pronounced solubilizing effect on MTX [20,32,33]. Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/pharmaceutics Pharmaceutics 2023, 15, 2244. https://doi.org/10.3390/pharmaceutics15092244 Pharmaceutics 2023, 15, 2244 2 of 15 e engi- gel for ocular delivery [8]. Yermak et al. [9] proposed CR gel beads for ophthalmic and oral delivery of echinochrome. Moreover, CR hydrogels have been used to administer acetaminophen orally to patients who have difﬁculty swallowing conventional formula- tions such as tablets or capsules [9,10]. There are kCR gels available for the delivery of β-carotene [11], silver nanoparticles [12], metformin hydrochloride [13], zaltoprofen [14], lidocaine [15], and brimonidine tartrate [16]. Hydrogels of CRs have been extensively studied as wound dressing materials because of their high water-holding capacity and biocompatibility [17]. Curcumin hydrogel ﬁlm based on CR has been developed and pro- posed as a functional wound dressing material [18]. CR-based hydrogel ﬁlm reinforced with sulfur nanoparticles and grapefruit seed extract has been designed for wound healing application [19]. been loaded in kCR gel for ocular delivery [8]. Yermak et al. [9] proposed CR gel beads for ophthalmic and oral delivery of echinochrome. Moreover, CR hydrogels have been used to administer acetaminophen orally to patients who have difficulty swallowing conventional formulations such as tablets or capsules [9,10]. There are kCR gels available for the delivery of β-carotene [11], silver nanoparticles [12], metformin hydrochloride [13], zaltoprofen [14], lidocaine [15], and brimonidine tartrate [16]. Hydrogels of CRs have been extensively studied as wound dressing materials because of their high wa- ter-holding capacity and biocompatibility [17]. Curcumin hydrogel film based on CR has been developed and proposed as a functional wound dressing material [18]. CR-based hydrogel film reinforced with sulfur nanoparticles and grapefruit seed extract has been designed for wound healing application [19]. MTX kCR iCR Figure 1. Chemical structure of MTX, kCR and iCR. Figure 1. Chemical structure of MTX, kCR and iCR. kCR iCR Figure 1. Chemical structure of MTX, kCR and iCR. Figure 1. Chemical structure of MTX, kCR and iCR. Figure 1. Chemical structure of MTX, kCR and iCR. Figure 1. Chemical structure of MTX, kCR and iCR. Recently, we proposed iCR gels for methotrexate (MTX) delivery [20]. MTX (Figure 1) is a therapeutic agent having a wide range of applications, including treatment of on- cological and autoimmune diseases [21]. 2.3. Rheological Measurements The rheological properties of the gels were measured by means of HAAKE MARS 60 Rheometer (Thermo Fisher Scientiﬁc, Dreieich, Germany) using a cone-plate (CP) mea- suring cell (CP 20/1◦). All rheological measurements were performed at 25 ◦C and 37 ◦C in duplicate. Temperature was controlled with accuracy of 0.1 ◦C by the Peltier elements. p p y y The gels were kept in the measuring cell for 15 min before the measurements for the temperature stabilization. At each temperature, the samples were measured in the following regimes: (1) Oscillation to obtain frequency dependences of the storage and loss moduli in the linear viscoelasticity range. The strain value was 0.5% and the frequency varied from 0.05 to 100 Hz; (2) Shear rate control mode to obtain ﬂow curves; the shear rate was increased from 0.004 to 5000 s−1 in a step-wise mode, with a duration of deformation of 30 s at every shear rate step. 2.2. Preparation of Hydrogels The kCR hydrogel was prepared by dissolving kCR powder in distilled water at 80 ◦C for 20 min under magnetic stirring until the powder was fully swollen and solubilized. Then, the solution was slowly cooled to room temperature. To obtain kCR/MTX gel, the saturated solution of MTX (2 × 10−4 M) was obtained and then used to prepare the gel as described above. To prepare the kCR/βCD gel, βCD (1 wt.%) was preliminarily dissolved in distilled water and then kCR powder was added to this solution. The protocol for kCR/MTX/βCD gel was as follows: βCD solution was prepared and used to obtain a saturated MTX solution, to which, kCR was added. The kCR concentration in all hydrogels under study was 1.25 wt.%. The content of the gels under study is reported in Table S1I. All gels were held for 24 h before tests to let the kCR swell properly and form a homogeneous solution. 2. Materials and Methods 2.1. Materials kCR, MTX, and βCD were purchased from Sigma-Aldrich (Moscow, Russia) and used without additional puriﬁcation. The content of cations in kCR was as follows: K+ (2.6 wt.%), Na+ (6.6 wt.%), and Ca2+ (7.7 wt.%). All other chemicals (Na2HPO4, KH2PO4, NaOH) were of analytical reagent grade and used without previous puriﬁcation. Double distilled water was used for gels and buffer solutions preparation. The pH of the solutions was determined by means of Five Easy pH-meter (Mettler Toledo, Columbus, OH, USA) standardized using reference solutions. In this connec- tion, βCD can be used to increase the MTX concentration in the kCR gel. Thus, the pur- pose of the present work was to design and characterize the kCR gels with MTX and βCD content, and the following aspects were investigated: (i) the influence of MTX (as active pharmaceutical ingredient) and βCD (as solubilizer) on the rheological behavior of kCR Since MTX is poorly soluble in water [27–29], the employment of solubilizing agents is necessary to increase the aqueous solubility and, consequently, the content of MTX in kCR hydrogel. Cyclodextrins (CDs), being native cyclic oligosaccharides, are widely used as solubilizers due to their ability to include guest molecules into a hydrophobic inner cavity and form inclusion (or host–guest) complexes [30,31]. Among native CDs, only βCD exhibits a more pronounced solubilizing effect on MTX [20,32,33]. In this connection, βCD can be used to increase the MTX concentration in the kCR gel. Thus, the purpose of the present work was to design and characterize the kCR gels with MTX and βCD content, and the following aspects were investigated: (i) the inﬂuence of MTX (as active pharmaceutical ingredient) and βCD (as solubilizer) on the rheological behavior of kCR gel; (ii) the possible interactions of MTX and βCD with the kCR network; (iii) the release of MTX from the gels and transmembrane permeability in vitro. Pharmaceutics 2023, 15, 2244 3 of 15 3 of 15 It should be also emphasized that the physicochemical and functional properties of CR gels are determined by the polymer structure, and, therefore, for iCR and kCR gels, they can be different. As is known, kCR contains only one sulfate group per disaccharide (Figure 1) and forms rigid gels with high swelling ability. On the contrary, iCR has two sulfate groups per disaccharide-repeating unit (Figure 1) and form more soft gels [1,2]. The carrageenan gels with various rheological characteristics could display different pharmaco- logical properties. Hence, it was interesting to compare the gels of iCR and kCR with MTX. The inﬂuence of the number of sulfate groups in the CR structure on the binding afﬁnity to MTX was considered herein. 2.4. Scanning Electron Microscopy (SEM) The SEM photos of the freeze-dried gels were made using the Quattro S microscope (Thermo Fisher Scientiﬁc), operating at 5 kV. Pharmaceutics 2023, 15, 2244 4 of 15 2.5. 1H NMR A Bruker-AV-500 1H NMR spectrometer was used. The constant temperature of 25 ◦C was maintained with the help of Bruker BVT-3000 temperature controller. Deuterated water (isotopic purity is 99.9%) was used as solvent in these experiments. 2.6. FTIR Spectroscopy 2.6. FTIR Spectroscopy FTIR spectra of the freeze-dried gels were acquired using the Fourier transform infrared spectrometer Vertex 80 v (Munich, Germany). Spectra were recorded in the range of 400–4000 cm−1. A KBr disc was prepared for each sample. 2.7. Dynamic Light Scattering Particle size distribution in the samples was determined by the dynamic light scat- tering (DLS) method using a Zetasizer Nano ZS analyzer (Malvern Instruments, Malvern, UK). Each sample was ﬁltered through a 0.45 µm ﬁlter and analysed after 24 h of sample preparation. Measurements were performed at 25 ◦C in the 173-degree backscattering mode. Concentration of the reagents was as follows: 0.01 wt.% for kCR, 2 × 10−4 M for MTX, and 1 wt.% for βCD. Each sample was measured three times and the average value was taken. 3. Results and Discussion kCR, kCR/MTX, kCR/βCD, and kCR/MTX/βCD were prepared and characterized with the aim of revealing the effect of the additives (MTX, βCD, and MTX/βCD inclusion complexes) on the physicochemical and pharmacological properties of the aforementioned gels designed for topical administration. The preliminary tests allowed for the establishment of the optimal concentration of kCR. It was found that gels with a kCR content <1.25 wt.% undergo a gel–sol transition temperature below 35 ◦C (Figure S2I), thus limiting their topical application. On the other hand, gels with a kCR content ≥1.25 wt.% were stable at the physiological temperatures (Figure S2I), but their consequent high rigidity also restricted their topical use. Therefore, the gel with a kCR content of 1.25 wt.% was selected as the most suitable for the purpose of this study. 2.9. In Vitro Transmembrane Permeation Study The in vitro permeation study was performed using a vertical Franz diffusion cell (PermeGear Inc., Hellertown, PA, USA). The receptor compartment was ﬁlled with 5 mL of phosphate buffer and stirred at 500 rpm. Gel (1 g) was placed on the donor compartment. A polyethersulfone membrane with 0.45 µm pore size was used as model membrane. An aliquot of 0.5 mL was taken from the receptor compartment at the predetermined time points and then replaced with the same volume of release medium to maintain sink conditions. The experiments were performed at constant temperature maintained at 32 ◦C. All samples were further analyzed spectrophotometrically (UV-1800, Shimadzu) at 258 nm. 2.8. Release Study The in vitro release of MTX from the kCR hydrogels without and with βCD was determined in phosphate buffer pH 7.4 (0.04 M Na2HPO4·12H2O, 0.03 M NaH2PO4·2H2O, 0.17 M NaCl) as release medium. The samples were stored in a hermetically sealed cuvette at 37 ◦C. At each time point, the absorbance of the solution was measured using a UV–vis spectrophotometer (UV-1800, Shimadzu, Tokyo, Japan) at 258 nm. MTX concentration was calculated using a previously obtained calibration curve (Figure S1I). 3.1. Linear Viscoelastic Properties Rheological properties play an important role in the design of topical pharmaceutical formulations. In this work, the rheological characteristics of the gels were obtained under Pharmaceutics 2023, 15, 2244 5 of 15 maceuti 5 of 15 maceuti various regimes of shearing for the evaluation of MTX, βCD, and MTX/βCD inclusion complex regarding the structure of kCR gel. under various regimes of shearing for the evaluation of MTX, βCD, and MTX/βCD in clusion complex regarding the structure of kCR gel. At first, the viscoelastic properties of the samples were measured in the guaranteed various regimes of shearing for the evaluation of MTX, βCD, and MTX/βCD inclusion complex regarding the structure of kCR gel. under various regimes of shearing for the evaluation of MTX, βCD, and MTX/βCD in clusion complex regarding the structure of kCR gel. At first, the viscoelastic properties of the samples were measured in the guaranteed p g g g At ﬁrst, the viscoelastic properties of the samples were measured in the guaranteed linear viscoelastic domain. Figure 2a displays the frequency sweep of the storage modulus (G′) and loss modulus (G′′) for kCR, kCR/MTX, kCR/βCD, and kCR/MTX/βCD gels at 25 ◦C. The results show that G′ > G′′ for all the systems. Moreover, both modules are very weakly dependent on the applied frequency. Such viscoelastic behavior is typical for strong gels [34]. It is believed that kCR is able to form gels with a three-dimensional network structure. In these systems, junction zones are the helical aggregates, which are formed via a conformational transition (caused by decrease of temperature) of random coils of the polymer to the double helices [35]. The junction zones in kCR gel can also be the aggregates of double helices containing 2–10 polysaccharide strands [36]. This particularity determines the formation of rigid kCR gels and distinguishes it from softer gels of iCR, for which the coil-to-double helix transition process shows second-order kinetics, indicating only dimerization [37]. At first, the viscoelastic properties of the samples were measured in the guaranteed linear viscoelastic domain. Figure 2a displays the frequency sweep of the storage mod ulus (G′) and loss modulus (G″) for kCR, kCR/MTX, kCR/βCD, and kCR/MTX/βCD gel at 25 °C. The results show that G′ > G″ for all the systems. Moreover, both modules ar very weakly dependent on the applied frequency. Such viscoelastic behavior is typica for strong gels [34]. 3.1. Linear Viscoelastic Properties It is believed that kCR is able to form gels with a three-dimensiona network structure. In these systems, junction zones are the helical aggregates, which ar formed via a conformational transition (caused by decrease of temperature) of random coils of the polymer to the double helices [35]. The junction zones in kCR gel can also b the aggregates of double helices containing 2–10 polysaccharide strands [36]. This par ticularity determines the formation of rigid kCR gels and distinguishes it from softer gel of iCR, for which the coil-to-double helix transition process shows second-order kinetics indicating only dimerization [37]. 10 0 10 1 10 2 10 0 10 1 10 2 10 3 ω, s -1 G' G'' kCR kCR/MTX kCR/βCD kCR/MTX/βCD 37 °C G', G'', Pa (b) 10 0 10 1 10 2 10 1 10 2 10 3 ω, s -1 25 °C G' G'' kCR kCR/MTX kCR/βCD kCR/MTX/βCD G', G'', Pa (a) 10 0 10 1 10 2 10 1 10 2 10 3 ω, s -1 25 °C G' G'' kCR kCR/MTX kCR/βCD kCR/MTX/βCD G', G'', Pa (a) 10 0 10 1 10 2 10 0 10 1 10 2 10 3 ω, s -1 G' G'' kCR kCR/MTX kCR/βCD kCR/MTX/βCD 37 °C G', G'', Pa (b) Figure 2. Frequency dependence of the storage (G′) and loss (G″) moduli at 25 °C (a) and 37 °C (b) Figure 2. Frequency dependence of the storage (G′) and loss (G′′) moduli at 25 ◦C (a) and 37 ◦C (b). (b) (a) Figure 2. Frequency dependence of the storage (G′) and loss (G″) moduli at 25 °C (a) and 37 °C (b). Figure 2. Frequency dependence of the storage (G′) and loss (G′′) moduli at 25 ◦C (a) and 37 ◦C (b). As shown in Figure 2a, the additives used (MTX, βCD, or MTX/βCD inclusion complexes) slightly affect the G′ values, without influencing the G″ values at 25 °C However, the effect of these additives on G′ and G″ becomes more pronounced at 37 °C (Figure 2b). This fact points out that the lability of the three-dimensional network of kCR is changed with the temperature rise and this promotes the interactions of kCR with MTX, βCD, or MTX/βCD. As shown in Figure 2a, the additives used (MTX, βCD, or MTX/βCD inclusion com- plexes) slightly affect the G′ values, without inﬂuencing the G′′ values at 25 ◦C. 3.1. Linear Viscoelastic Properties However, the effect of these additives on G′ and G′′ becomes more pronounced at 37 ◦C (Figure 2b). This fact points out that the lability of the three-dimensional network of kCR is changed with the temperature rise and this promotes the interactions of kCR with MTX, βCD, or MTX/βCD. β β It should be noted that the difference between the G′ values obtained for kCR/MTX gel at 25 °C and 37 °C is less than for kCR/βCD and kCR/MTX/βCD gels (Figure 2). Thi means that the MTX molecules are more capable of stabilizing the structure of kCR gel making it less sensitive to the temperature rise. Such a stabilizing effect is apparently caused by a decrease in the mobility of the polymeric chains due to the interactions be tween MTX and kCR. MTX has carboxylic and amino groups in the structure (Figure 1) which are able to conduct ionization (pKa1 = 3.22, pKa2 = 4.53 and pKa3 = 5.62 [38]). kCR contains –OH and –SO3− groups (Figure 1). Thus, binding of MTX with kCR can occu mainly through the electrostatic interactions, H-bonds formation, van der Waals forces and hydrophobic interactions [39]. These multiple interactions were also observed be tween iCR and MTX [20]. It can be hypothesized that MTX incorporation both in junction β It should be noted that the difference between the G′ values obtained for kCR/MTX gel at 25 ◦C and 37 ◦C is less than for kCR/βCD and kCR/MTX/βCD gels (Figure 2). This means that the MTX molecules are more capable of stabilizing the structure of kCR gel, making it less sensitive to the temperature rise. Such a stabilizing effect is apparently caused by a decrease in the mobility of the polymeric chains due to the interactions between MTX and kCR. MTX has carboxylic and amino groups in the structure (Figure 1), which are able to conduct ionization (pKa1 = 3.22, pKa2 = 4.53 and pKa3 = 5.62 [38]). kCR contains –OH and –SO3−groups (Figure 1). Thus, binding of MTX with kCR can occur mainly through the electrostatic interactions, H-bonds formation, van der Waals forces, and hydrophobic interactions [39]. These multiple interactions were also observed between iCR and MTX [20]. It can be hypothesized that MTX incorporation both in junction zones and interjunction connections in kCR gel increases the stability of its structure and dramatically decreases the mobility of the biopolymer macromolecules. 3.1. Linear Viscoelastic Properties Meanwhile, βCD also has a stabilizing effect on the kCR gel, as in Figure 2, even if it is less pronounced than with MTX. In a previous work, the authors conﬁrmed the binding Pharmaceutics 2023, 15, 2244 6 of 15 of CDs to a kCR via hydrogen bonding [40]. A rearrangement of random kCR coils in the presence of CDs, which lead to a more uniform distribution of kCR molecules in water, has been postulated according to the rheological data and SEM images [40,41]. It has been also demonstrated by Wang et al. [42] that hydroxypropyl-β-CD affected the kCR gelation mainly by facilitating the ordering of kCR coils and preventing the aggregation of kCR helices. Hydroxypropylated β-CD enabled gel formation as well as gel melting at a slightly higher temperature, compared with the “empty” kCR gel [42]. g p p p y g Conversely, the kCR/MTX/βCD gel is characterized by lower G′ values than the pure kCR gel (Figure 2). This effect is pronounced at 37 ◦C. Moreover, the kCR/MTX/βCD gel demonstrates the smallest gap between G′ and G′′ at 37 ◦C, conﬁrming the gel structure in the presence of MTX/βCD inclusion complexes. Apparently, the presence of MTX/βCD complexes results in the formation of a more labile structure of the kCR gel. It is important to note that the kCR/MTX/βCD gel still has G′ > G′′ at 37 ◦C, retaining the gel at human body temperature. 3.2. Steady Flow Properties The possibility of application of the topical gels is determined by their ﬂow behavior. Therefore, we explored the rheological properties of the gels under steady shear. Figure 3 displays the ﬂow viscosity of the samples versus the shear stress. One can see a similarity in the behavior of all the gels tested at 25 ◦C, whereas the noticeable difference in the rheolog- ical behavior of the samples was observed at 37 ◦C. Addition of MTX, βCD, or MTX/βCD inclusion complexes to kCR gel does not change its non-Newtonian pseudoplastic behavior, but essentially affects its yield shear stress and apparent viscosity. 5, x FOR PEER REVIEW 10 1 10 2 10 -2 10 0 10 2 10 4 25°C kCR kCR/MTX kCR/βCD kCR/MTX/βCD η, Pa·s τ, Pa 10 0 10 1 10 2 10 -3 10 -1 10 1 10 3 kCR kCR/MTX kCR/βCD kCR/MTX/βCD τ, Pa 37°C η, Pa·s Figure 3. Dependence of the apparent viscosity on the shear stress for the gels under s Figure 3. Dependence of the apparent viscosity on the shear stress for the gels under study. 10 1 10 2 10 -2 10 0 10 2 10 4 25°C kCR kCR/MTX kCR/βCD kCR/MTX/βCD η, Pa·s τ, Pa 10 0 10 1 10 2 10 -3 10 -1 10 1 10 3 kCR kCR/MTX kCR/βCD kCR/MTX/βCD τ, Pa 37°C η, Pa·s Figure 3. Dependence of the apparent viscosity on the shear stress for the gels u Figure 3. Dependence of the apparent viscosity on the shear stress for the gels under study. Table 1. Estimated rheological model constants and statistical parameters . The inﬂuence of the additives on the rheological behavior of kCR gels at 37 ◦C (Figure S3I) was assessed by using several mathematical models: Table 1. Estimated rheological model constants and statistical parameters . 3.2. Steady Flow Properties h l lkl d l h d l d The inﬂuence of the additives on the rheological behavior of kCR gels at 37 ◦C (Figure S3I) was assessed by using several mathematical models: Herschel-Bulkley Model Bingham Model Casson Mo P k P R P P R P P Herschel–Bulkley model τ0, Pa k, Pa sn n R2 τ0, Pa ηp, Pa s R2 τ0, Pa ηp, Pa s 2.1 ± 0.5 3.8 ± 0.3 0.37± 0.01 0.99 12 ± 2 0.020 ± 0.001 0.82 2.4 ± 0.2 0.12 ± 0.01 6.4 ± 3.7 13.6 ± 3.7 0.21± 0.03 0.95 24 ± 3 0.019 ± 0.002 0.63 4.1 ± 0.3 0.10 ± 0.01 2.2 ± 1.0 8.9 ± 0.8 0.29± 0.01 0.99 17 ± 3 0.025 ± 0.002 0.77 2.8 ± 0.3 0.14 ± 0.01 0.7 ± 0.2 2.0 ± 0.1 0.45± 0.01 0.99 7 ± 2 0.026 ± 0.001 0.87 1.8 ± 0.1 0.13 ± 0.01 —experimental data for 37 °C. As can be seen from Table 1, the Herschel–Bulkley equation provides the τ = τ0 + kγn (1) Bingham model τ = τ0 + ηpγ (2) Casson model τ0, Pa ηp, Pa s R τ0, Pa ηp, Pa s 9 12 ± 2 0.020 ± 0.001 0.82 2.4 ± 0.2 0.12 ± 0.01 5 24 ± 3 0.019 ± 0.002 0.63 4.1 ± 0.3 0.10 ± 0.01 τ = τ0 + kγn (1) ± 0.0 0 0 (1) 6.4 ± 3.7 13.6 ± 3 2.2 ± 1.0 8.9 ± 0 Bingham model 9 7 ± 2 0.026 ± 0.001 0.87 1.8 ± 0.1 0.13 ± 0.01 7 °C. m Table 1, the Herschel–Bulkley equation provides the τ = τ0 + ηpγ (2) ± 0.1 0.45± 0.01 0.99 7 ± 2 0.026 ± 0.001 0.87 1.8 ± 0.1 0.13 ± 0.01 experimental data for 37 °C. τ = τ0 + ηpγ (2) ± 0.0 (2) As can be seen from Table 1, the Herschel–Bulkley equation provides the m asson model Casson model Casson model e gels (R2 were the highest). The yield shear stress (τ0) of a m um stress needed to start the flow. The consistency coeffi τ0.5 = τ00.5 + (ηpγ)0.5 (3) of a coeffi (3) close to the average viscosity of materials in the entire shear rate range [43]. Acc the obtained results, the addition of MTX or βCD increased the values of τ0 and influence of MTX is more significant. On the contrary, minimum values of τ0 a obtained for kCR/MTX/βCD gel. Thus, kCR/MTX/βCD gel is more spreadable h l d d where τ is the shear stress (Pa); τ0 is the yield shear stress (Pa); k is the consistency coefﬁcient (Pa·sn); γ is the shear rate (s−1); ηp is plastic viscosity (Pa·s); n is the ﬂow coefﬁcient. The rheological parameters and correlation coefﬁcients (R2) speciﬁc to these models are listed in Table 1. close to the average viscosity of materials in the entire shear rate range [43]. Acc the obtained results, the addition of MTX or βCD increased the values of τ0 and influence of MTX is more significant. On the contrary, minimum values of τ0 a obtained for kCR/MTX/βCD gel. Thus, kCR/MTX/βCD gel is more spreadable h l d id i where τ is the shear stress (Pa); τ0 is the yield shear stress (Pa); k is the consistency coefﬁcient (Pa·sn); γ is the shear rate (s−1); ηp is plastic viscosity (Pa·s); n is the ﬂow coefﬁcient. The rheological parameters and correlation coefﬁcients (R2) speciﬁc to these models are listed in Table 1. 7 of 15 Pharmaceutics 2023, 15, 2244 Table 1. Estimated rheological model constants and statistical parameters . Table 1. Estimated rheological model constants and statistical parameters . Casson model Gel Herschel-Bulkley Model Bingham Model Casson Model τ0, Pa k, Pa·sn n R2 τ0, Pa ηp, Pa·s R2 τ0, Pa ηp, Pa·s R2 kCR 2.1 ± 0.5 3.8 ± 0.3 0.37± 0.01 0.99 12 ± 2 0.020 ± 0.001 0.82 2.4 ± 0.2 0.12 ± 0.01 0.84 kCR/MTX 6.4 ± 3.7 13.6 ± 3.7 0.21± 0.03 0.95 24 ± 3 0.019 ± 0.002 0.63 4.1 ± 0.3 0.10 ± 0.01 0.64 kCR/βCD 2.2 ± 1.0 8.9 ± 0.8 0.29± 0.01 0.99 17 ± 3 0.025 ± 0.002 0.77 2.8 ± 0.3 0.14 ± 0.01 0.76 kCR/MTX/βCD 0.7 ± 0.2 2.0 ± 0.1 0.45± 0.01 0.99 7 ± 2 0.026 ± 0.001 0.87 1.8 ± 0.1 0.13 ± 0.01 0.91 —experimental data for 37 ◦C. As can be seen from Table 1, the Herschel–Bulkley equation provides the most appro- priate ﬁt for all the gels (R2 were the highest). The yield shear stress (τ0) of a material is deﬁned as the minimum stress needed to start the ﬂow. The consistency coefﬁcient (k) is close to the average viscosity of materials in the entire shear rate range [43]. According to the obtained results, the addition of MTX or βCD increased the values of τ0 and k, and the inﬂuence of MTX is more signiﬁcant. On the contrary, minimum values of τ0 and k were obtained for kCR/MTX/βCD gel. Thus, kCR/MTX/βCD gel is more spreadable than the other gels under consideration. g The ﬂow index values (n < 1) conﬁrm that all systems are pseudoplastic ﬂuids. The n values are decreased with the addition of MTX or βCD to kCR gel and they are increased for the kCR/MTX/βCD system. The obtained data indicate that MTX and βCD strengthen the gel structure; however, after overcoming the yield stress, the structure breaks more easily under the action of high-speed shear. Inclusion complexes MTX/βCD, on the contrary, reduce both the gel strength and sensitivity of the gel structure to shear deformations imposed on the samples in this experiment. It should be noted that the k value for the kCR/MTX/βCD gel is lower than that of the kCR/MTX sample, so the former will be easier to apply on the skin or tissues. 3.3. SEM Analysis SEM analysis was performed to study the surface morphology of the freeze-dried gels. The micrographs reported in Figure 4 demonstrate that the gels hold a porous structure, a denser and more even texture with smaller pore sizes, which could be due to the interactions between the kCR helixes and MTX. In the case of kCR/βCD gel, the structure is non-uniform with a lower quantity of pores as compared with the pure kCR sample. A similar effect of βCD on the kCR gel has been reported by Yuan C. et al. [40]. Larger pores have been observed in kCR/MTX/βCD compared to the pure kCR gel. REVIEW 8 of 16 kCR kCR kCR/MTX 500 μm 100 μm 500 μm 100 μm Figure 4. Cont. 100 μm kCR/MTX Figure 4. Cont. 8 of 15 8 of 15 Pharmaceutics 2023, 15, 2244 kCR/MTX kCR/βCD kCR/MTX/βCD Figure 4. SEM images of freeze-dried gels under study. 3.4.1. H NMR and FTIR Spectroscopy T l th t f th i t ti b t kCR d th dditi (MTX 500 μm 500 μm 500 μm 100 μm 100 μm 100 μm Figure 4. SEM images of freeze-dried gels under study. 3.4. 1H NMR and FTIR Spectroscopy To reveal the nature of the interactions between kCR and the additives (MTX an βCD) i h l h 1H NMR i i d 1H NMR f MTX kCR/MTX kCR/MTX kCR/βCD 500 μm 500 μm kCR/βCD 500 μm 100 μm kCR/MTX/βCD 500 μm kCR/MTX/βCD 500 μm 100 μm Figure 4 SEM images of freeze dried gels under study Figure 4. SEM images of freeze-dried gels under study. g g g d I 3.4. 1H NMR and FTIR Spectroscopy As one can see, more pronounced changes were observed for protons H25, H28 d H29 located near polar side groups of MTX (Figure 1). More probably, these polar oups are involved in the binding with kCR via hydrogen bonding. 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 H8 H8 H18,20 H18,20 H17,21 H17,21 H25 H25 MTX kCR/MTX H13 H13 H15 H15 H29 H29 H28 H28 Figure 5. 1H NMR spectra of MTX (blue color), hydrogels of kCR (red color) and kCR/MTX (black color) at 25 °C. Figure 5. 1H NMR spectra of MTX (blue color), hydrogels of kCR (red color) and kCR/MTX (black color) at 25 ◦C. ups are involved in the binding with kCR via hydrogen bonding. Binding of MTX with kCR was additionally studied in solution. Chemical shift changes of MTX protons induced by interactions with kCR were measured and given in Figure 6. As one can see, more pronounced changes were observed for protons H25, H28 and H29 located near polar side groups of MTX (Figure 1). More probably, these polar groups are involved in the binding with kCR via hydrogen bonding. 0.06 -0.06 Δδ, ppm iCR kCR 8 13 15 17,21 18,20 25 28 29 0.03 0.00 -0.03 gure 6. Chemical shift changes of MTX protons induced by the presence of kCR (0.4 wt.%) and R (4 wt.%) in solution (D2O, 25 °C). 0.06 -0.12 -0.09 -0.06 Δδ, ppm iCR kCR 8 13 15 17,21 18,20 25 28 29 0.03 0.00 -0.03 Figure 6. Chemical shift changes of MTX protons induced by the presence of kCR (0.4 wt.%) and iCR (4 wt.%) in solution (D2O, 25 ◦C). -0.06 ure 6. Chemical shift changes of MTX protons induced by the presence of kCR (0.4 wt.%) and (4 wt.%) in solution (D2O, 25 °C). Figure 6. Chemical shift changes of MTX protons induced by the presence of kCR (0.4 wt.%) and iCR (4 wt.%) in solution (D2O, 25 ◦C). Figure 6. Chemical shift changes of MTX protons induced by the presence of kCR (0.4 wt.%) and iCR (4 wt %) in solution (D2O 25 °C) -0.12 -0.09 It was interesting to evaluate the role of –SO3− groups in CR molecule in the binding h MTX. To this aim, the binding affinity of MTX to kCR and iCR [20] in solution was mpared. g g g d I 3.4. 1H NMR and FTIR Spectroscopy 3.4.1. H NMR and FTIR Spectroscopy To reveal the nature of the interactions between kCR and the additives (MTX and βCD) in the gels, the 1H NMR experiments were carried out. 1H NMR spectra of MTX in its pure form and incorporated in the kCR gel are shown in Figure 5. As is evident from To reveal the nature of the interactions between kCR and the additives (MTX and βCD) in the gels, the 1H NMR experiments were carried out. 1H NMR spectra of MTX in its pure form and incorporated in the kCR gel are shown in Figure 5. As is evident from the comparative analysis of the 1H NMR spectra, the signals from MTX protons H25, H28, and H29 located near the polar amino and carboxylic groups of MTX (Figure 1) are upﬁeld shifted. It seems that the polar fragment of the MTX molecule participates in the binding with kCR. Unfortunately, the 1H NMR spectra of kCR gels with βCD were not suitable for analysis due to the overlapping of their main characteristic signals. Binding of MTX with kCR was additionally studied in solution. Chemical shift changes of MTX protons induced by interactions with kCR were measured and given in Figure 6. As one can see, more pronounced changes were observed for protons H25, H28 and H29 located near polar side groups of MTX (Figure 1). More probably, these polar groups are involved in the binding with kCR via hydrogen bonding. Pharmaceutics 2023, 15, 2244 9 of 15 in the ere not 9 of 15 in the ere not y pp g g Figure 5. 1H NMR spectra of MTX (blue color), hydrogels of kCR (red color) and kCR/MTX (black color) at 25 °C. 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 H8 H8 H18,20 H18,20 H17,21 H17,21 H25 H25 MTX kCR/MTX kCR H13 H13 H15 H15 H29 H29 H28 H28 Figure 5. 1H NMR spectra of MTX (blue color), hydrogels of kCR (red color) and kCR/MTX (black color) at 25 ◦C. ure 5. 1H NMR spectra of MTX (blue color), hydrogels of kCR (red color) and kCR/MTX (black or) at 25 °C. Binding of MTX with kCR was additionally studied in solution. Chemical shift anges of MTX protons induced by interactions with kCR were measured and given in gure 6. g g g d I 3.4. 1H NMR and FTIR Spectroscopy As follows from the obtained Δδ values (Figure 6), more significant chemical ft changes of MTX protons were observed in the presence of 0.4 wt.% iCR. iCR has two It was interesting to evaluate the role of –SO3−groups in CR molecule in the binding with MTX. To this aim, the binding afﬁnity of MTX to kCR and iCR [20] in solution was compared. As follows from the obtained ∆δ values (Figure 6), more signiﬁcant chemical shift changes of MTX protons were observed in the presence of 0.4 wt.% iCR. iCR has two sulfate groups per disaccharide repeating unit (Figure 1). Consequently, the number of binding sites with MTX is higher in iCR compared with kCR having only one –SO3−group per disaccharide unit. iCR (4 wt.%) in solution (D2O, 25 C). It was interesting to evaluate the role of –SO3− groups in CR molecule in the binding with MTX. To this aim, the binding affinity of MTX to kCR and iCR [20] in solution was compared. As follows from the obtained Δδ values (Figure 6), more significant chemical shift changes of MTX protons were observed in the presence of 0.4 wt.% iCR. iCR has two Interactions of kCR with MTX and βCD in solid state were studied by FTIR spec- troscopy. To this end, FTIR spectra of freeze-dried gels under study were recorded and analyzed (Figure S4I). As it was observed, main characteristic bands of pure kCR, MTX and βCD were remained unchanged in the spectra of the gels. This fact points out the absence of chemical interaction between kCR network and additives (MTX and βCD) in the solid state. 3.5. DLS Upon addition of βCD to kCR solution, the tendency of kCR to aggregate is reduced, and one broad peak with the mode at 105 nm is observed (Figure 7). A similar peak corresponding to a hydrodynamic diameter of 136 nm appears for kCR solution with MTX content. Most likely, the revealed changes in the particle size distribution are due to the conformational rearrangement of biopolymer macromolecules caused by their possible interactions with the additives. A broadening of the peak induced by decreasing temperature of kCR solution has been reported by Abad et al. [45]. This phenomenon has been assigned to the appearance of particles with higher rigidity due to the coil-to- helix transition of kCR. Apparently, MTX and βCD can interact with kCR via multiple noncovalent interactions, promoting the helical conformation of the biopolymer. p g p y The bimodal particle size distribution was detected for kCR solution in the presence of MTX/βCD inclusion complexes (Figure 7). It seems that MTX/βCD complexes have no considerable inﬂuence on the kCR aggregation. It is more likely that βCD and MTX have higher afﬁnity to interact with each other than with kCR. Therefore, the behavior of kCR in the presence of the complexes is similar to the behavior of pure kCR. However, a decrease in the intensity of the peak corresponding to large particles and an increase in the intensity of peak representing the individual kCR coils should be noted. This fact can be explained by the ability of the complexes to prevent aggregation of the kCR coils. 3.5. DLS As one can see from Figure 7, addition of MTX (3 × 10−4 M) or βCD (0.009 M) to kCR solution (0.01 wt.%) causes the disappearence of the peak corresponding to the existence of large aggregates. Upon addition of βCD to kCR solution, the tendency of kCR to aggregate is reduced, and one broad peak with the mode at 105 nm is observed (Figure 7). A similar peak corresponding to a hydrodynamic diameter of 136 nm appears for kCR solution with MTX content. Most likely, the revealed changes in the particle size distribution are due to the conformational rearrangement of biopolymer macromolecules caused by their possible interactions with the additives. A broadening of the peak induced by decreasing temperature of kCR solution has been reported by Abad et al. [45]. This phenomenon has been assigned to the appearance of particles with higher rigidity due to the coil-to- helix transition of kCR. Apparently, MTX and βCD can interact with kCR via multiple noncovalent interactions, promoting the helical conformation of the biopolymer. The bimodal particle size distribution was detected for kCR solution in the presence of MTX/βCD inclusion complexes (Figure 7). It seems that MTX/βCD complexes have no considerable inﬂuence on the kCR aggregation. It is more likely that βCD and MTX have higher afﬁnity to interact with each other than with kCR. Therefore, the behavior of kCR in the presence of the complexes is similar to the behavior of pure kCR. However, a decrease in the intensity of the peak corresponding to large particles and an increase in the intensity Figure 7. Size distribution spectra of the samples under study at 25 °C. Figure 7. Size distribution spectra of the samples under study at 25 ◦C. Figure 7. Size distribution spectra of the samples under study at 25 °C. Figure 7. Size distribution spectra of the samples under study at 25 ◦C. As one can see from Figure 7, addition of MTX (3 × 10−4 M) or βCD (0.009 M) to kCR solution (0.01 wt.%) causes the disappearence of the peak corresponding to the existence of large aggregates. Upon addition of βCD to kCR solution, the tendency of kCR to ag- gregate is reduced, and one broad peak with the mode at 105 nm is observed (Figure 7). A similar peak corresponding to a hydrodynamic diameter of 136 nm appears for kCR so- lution with MTX content. 3.5. DLS DLS is the most popular method for determining polymer sizes in solution. DLS measurements of the dilute kCR solutions without and with the additives under study were performed with the aim of revealing the inﬂuence of MTX, βCD, or MTX/βCD inclusion Pharmaceutics 2023, 15, 2244 10 of 15 n. DLS r study complexes on the aggregation behavior and conformation of kCR macromolecules. The obtained results are shown in Figure 7. As one can see, pure kCR solution has a bimodal intensity size distribution with the peaks corresponding to hydrodanamic diameters of 91 nm and 503 nm. A similar size distribution for kCR solution has been previously reported by Antonov et al. [44]. The authors pointed out that these two peaks probably originate from kCR macromolecules (coils) and their aggregates. clusion complexes on the aggregation behavior and conformation of kCR macromole- cules. The obtained results are shown in Figure 7. As one can see, pure kCR solution has a bimodal intensity size distribution with the peaks corresponding to hydrodanamic diameters of 91 nm and 503 nm. A similar size distribution for kCR solution has been previously reported by Antonov et al. [44]. The authors pointed out that these two peaks probably originate from kCR macromolecules (coils) and their aggregates. Figure 7. Size distribution spectra of the samples under study at 25 °C. Figure 7. Size distribution spectra of the samples under study at 25 ◦C. Figure 7. Size distribution spectra of the samples under study at 25 °C. As one can see from Figure 7, addition of MTX (3 × 10−4 M) or βCD (0.009 M) to kCR solution (0.01 wt.%) causes the disappearence of the peak corresponding to the existence of large aggregates. Upon addition of βCD to kCR solution, the tendency of kCR to ag- gregate is reduced, and one broad peak with the mode at 105 nm is observed (Figure 7). A similar peak corresponding to a hydrodynamic diameter of 136 nm appears for kCR so- lution with MTX content. Most likely, the revealed changes in the particle size distribu- tion are due to the conformational rearrangement of biopolymer macromolecules caused by their possible interactions with the additives. A broadening of the peak induced by decreasing temperature of kCR solution has been reported by Abad et al. [45]. This Figure 7. Size distribution spectra of the samples under study at 25 ◦C. 3.5. DLS Most likely, the revealed changes in the particle size distribu- tion are due to the conformational rearrangement of biopolymer macromolecules caused by their possible interactions with the additives. A broadening of the peak induced by decreasing temperature of kCR solution has been reported by Abad et al. [45]. This As one can see from Figure 7, addition of MTX (3 × 10−4 M) or βCD (0.009 M) to kCR solution (0.01 wt.%) causes the disappearence of the peak corresponding to the existence of large aggregates. Upon addition of βCD to kCR solution, the tendency of kCR to aggregate is reduced, and one broad peak with the mode at 105 nm is observed (Figure 7). A similar peak corresponding to a hydrodynamic diameter of 136 nm appears for kCR solution with MTX content. Most likely, the revealed changes in the particle size distribution are due to the conformational rearrangement of biopolymer macromolecules caused by their possible interactions with the additives. A broadening of the peak induced by decreasing temperature of kCR solution has been reported by Abad et al. [45]. This phenomenon has been assigned to the appearance of particles with higher rigidity due to the coil-to- helix transition of kCR. Apparently, MTX and βCD can interact with kCR via multiple noncovalent interactions, promoting the helical conformation of the biopolymer. As one can see from Figure 7, addition of MTX (3 × 10−4 M) or βCD (0.009 M) to kCR solution (0.01 wt.%) causes the disappearence of the peak corresponding to the existence of large aggregates. Upon addition of βCD to kCR solution, the tendency of kCR to ag- gregate is reduced, and one broad peak with the mode at 105 nm is observed (Figure 7). A similar peak corresponding to a hydrodynamic diameter of 136 nm appears for kCR so- lution with MTX content. Most likely, the revealed changes in the particle size distribu- tion are due to the conformational rearrangement of biopolymer macromolecules caused by their possible interactions with the additives. A broadening of the peak induced by decreasing temperature of kCR solution has been reported by Abad et al. [45]. This As one can see from Figure 7, addition of MTX (3 × 10−4 M) or βCD (0.009 M) to kCR solution (0.01 wt.%) causes the disappearence of the peak corresponding to the existence of large aggregates. 3.6. Release Study Figure 3 and Table ow that kCR/MTX/βCD gel is less viscous and more spreadable than kCR/MTX at 37 Most likely, it facilitates the diffusion of MTX. Moreover, the additional reason for observed phenomenon is the inclusion complex formation occuring between MTX βCD in solution [20,32]. The MTX content is larger in kCR/MTX/βCD than in /MTX. Consequently, the higher concentration gradient determines the faster release TX from kCR/MTX/βCD than kCR/MTX gel. However, the concentration gradient is a key factor governing the release rate. To confirm this, we prepared kCR/MTX/βCD kCR/MTX gels with the same MTX concentration and found that the release from irst of the mentioned gels proceeds faster (Figure S5I). Moreover, we believe that the ity of MTX is stronger to βCD than to kCR. This assumption is supported by the osite rheological behavior of kCR/MTX and kCR/MTX/βCD gels at 37 °C (Figure 3). βCD incorporation in the kCR gels induces a high release of MTX. For instance, full MTX release (100 wt.%) is achieved in 4 h in the presence of 1 wt.% βCD. The obtained results are in accordance with the rheological properties of these gels. Figure 3 and Table 1 show that kCR/MTX/βCD gel is less viscous and more spreadable than kCR/MTX at 37 ◦C. Most likely, it facilitates the diffusion of MTX. Moreover, the additional reason for the observed phenomenon is the inclusion complex formation occuring between MTX and βCD in solution [20,32]. The MTX content is larger in kCR/MTX/βCD than in kCR/MTX. Consequently, the higher concentration gradient determines the faster release of MTX from kCR/MTX/βCD than kCR/MTX gel. However, the concentration gradient is not a key factor governing the release rate. To conﬁrm this, we prepared kCR/MTX/βCD and kCR/MTX gels with the same MTX concentration and found that the release from the ﬁrst of the mentioned gels proceeds faster (Figure S5I). Moreover, we believe that the afﬁnity of MTX is stronger to βCD than to kCR. This assumption is supported by the opposite rheological behavior of kCR/MTX and kCR/MTX/βCD gels at 37 ◦C (Figure 3). Inclusion complexes of CDs are water-soluble due to the availability of the external –OH groups surrounding the macrocyclic cavity. Thus, MTX inserted into βCD displays higher afﬁnity to an aqueous environment compared with the uncomplexed MTX. The MTX release rate increases with βCD content in kCR/MTX/βCD gel from 0.2 wt.% to 1.0 wt.%. 3.6. Release Study The ability of kCR gels to release the MTX was investigated in vitro in phosphate buffer at 37 ◦C. Figure 8 shows the release proﬁles of MTX from hydrogels. The release rate from kCR/MTX gel is relatively low, −65 wt.% of the drug during 6 h. It can be seen Pharmaceutics 2023, 15, 2244 11 of 15 se en 11 of 15 se en for comparison that kCR and iCR gels of the same composition released MTX, reaching 50 wt.% and 35 wt.%, respectively, after 2.5 h. The faster release from kCR gel can be explained by the revealed weaker binding of MTX with kCR than with iCR. t.% and 35 wt.%, respectively, after 2.5 h. The faster release from kCR gel can be ex- ed by the revealed weaker binding of MTX with kCR than with iCR. for comparison that kCR and iCR gels of the same composition released MTX, reaching 50 wt.% and 35 wt.%, respectively, after 2.5 h. The faster release from kCR gel can be explained by the revealed weaker binding of MTX with kCR than with iCR. wt.% and 35 wt.%, respectively, after 2.5 h. The faster release from kCR gel can be ex- ned by the revealed weaker binding of MTX with kCR than with iCR. 0 50 100 150 200 250 300 350 400 0 20 40 60 80 100 kCR/MTX/βCD_0.2% kCR/MTX/βCD_0.5% kCR/MTX/βCD_1% kCR/MTX/βCD_1.4% kCR/MTX iCR/MTX Cumulative MTX released (wt.%) Time, min 8 R l fil f MTX f CR l (1 25 t %) i h h t b ff ( H 7 4) t 37 °C Figure 8. Release proﬁles of MTX from CR gels (1.25 wt.%) in phosphate buffer (pH = 7.4) at 37 ◦C. l f l f X f C l ( %) h h b ff ( ) C Figure 8. Release proﬁles of MTX from CR gels (1.25 wt.%) in phosphate buffer (pH = 7.4) at 37 ◦C. re 8. Release profiles of MTX from CR gels (1.25 wt.%) in phosphate buffer (pH = 7.4) at 37 °C. βCD incorporation in the kCR gels induces a high release of MTX. For instance, full X release (100 wt.%) is achieved in 4 h in the presence of 1 wt.% βCD. The obtained lts are in accordance with the rheological properties of these gels. 3.6. Release Study However, further increase in βCD concentration up to 1.4 wt.% has no inﬂuence on the MTX release rate. This behavior is in accordance with the increase in the concentra- tion of the inclusion complexes, which are formed between MTX and βCD in solution. The dependence of the inclusion complexes concentration on βCD concentration in solution was obtained considering the previously determined binding constant (K = 736 M−1 [20]), and given in Figure S6I. The concentration of the inclusion complexes increases with a βCD amount up to 1 wt.%, and after that, it is not changed. This fact could be taken into account for prediction of the MTX release rate from the gels with the variable βCD content. p g The release of MTX from kCR gels of different composition was mathematically described using different kinetic models such as: p g The release of MTX from kCR gels of different composition was mathematically described using different kinetic models such as: zero-order model (4) Qt/Q∞= K0t ﬁrst-order model Pharmaceutics 2023, 15, 2244 12 of 15 12 of 15 log Qt/Q∞= −K1·t/2.303 (5) log Qt/Q∞= −K1·t/2.303 (5) Higuchi model Higuchi model Qt/Q∞= KH·t1/2 (6) Qt/Q∞= KH·t1/2 (6) Hixson–Crowell model Q∞1/3 −Qt1/3 = KHC·t (7) (7) Korsemeyer–Peppas model Qt/Q∞= KKP·tn (8) Qt/Q∞= KKP·tn (8) where Qt/Q∞represents the fractional drug release; K0, K1, KH, KHC, and KKP are the kinetic constants of each mathematical model; n is release exponent. The best ﬁtting was revealed considering the values of the correlation coefﬁcients (R2) summarized in Table 2. The Korsemeyer–Peppas model was the most appropriate model to predict the MTX release from the gel, as indicated by the highest R2. In this model, the n value is used to determine the release mechanism. Generally, the n values below 0.45 correspond to Fickian diffusion, while values between 0.45 and 0.89 reveal anomalous diffusion (non-Fickian) [46]. Based on Table 2, n < 0.45 values indicate the release of MTX from all gels under study is Fickian diffusion. The presence of βCD in the gels does not affect the mode of MTX release. Table 2. Modeling of MTX release behavior from kCR gels of different composition. 3.6. Release Study Model R2 kCR/MTX kCR/MTX/βCD_0.2% kCR/MTX/βCD_0.5% iCR/MTX/βCD_1% iCR/MTX/βCD_1.4% Zero-order 0.735 0.939 0.619 0.565 0.662 First-order 0.455 0.802 0.473 0.486 0.575 Higuchi 0.832 0.982 0.836 0.803 0.850 Hixson-Crowell 0.549 0.859 0.5242 0.512 0.604 Korsemeyer-Peppas 0.931 (n = 0.44) 0.977 (n = 0.41) 0.910 (n = 0.41) 0.934 (n = 0.08) 0.962 (n = 0.10) 3.7. Permeation Study Table 2. Modeling of MTX release behavior from kCR gels of different composition. eling of MTX release behavior from kCR gels of different composition. 4. Conclusions Significant differences in the permeation profiles of MTX from kCR and iCR were ved (Figure 9). This is in accordance with the faster release of MTX from kCR gel re 8) and revealed the weaker binding affinity of MTX to kCR (Figure 6). The area under the plasma drug concentration–time curve (AUC) obtained from e 9 can be used to predict in vivo AUC. Calculated values of AUC0–6.7h are given in 2SI. This attempt is based on a linear in vivo–in vitro correlation that has been d by M. Klitgaard et al. [49] when comparing the area under the in vitro drug per- on–time curve to the AUC of the plasma concentration–time profile obtained from n vivo study. Based on these results, the permeation method was proposed as a ising tool for estimating the in vivo performance. l i Gels based on biocompatible kCR have been easily prepared and used for MTX entrap- ment. MTX concentration in the gel can be increased by means of βCD, which forms stable inclusion complexes with MTX. Moreover, stabilization of MTX in the gel can be achieved due to the inclusion complex formation with βCD. It was demonstrated that the addition of βCD to the gel formulations has no signiﬁcant effects on the viscoelastic properties of the gels; meanwhile, it affects the MTX release from the gels. The MTX release proﬁles from kCR/MTX and kCR/MTX/βCD gels with variable βCD content were obtained and mathematically analyzed. It was found that MTX release from all gels follows Fickian diffusion, which can be controlled by increasing the βCD concentration in the formulation. Various behaviors of kCR/MTX and iCR/MTX gels were demonstrated and explained by the different binding afﬁnity of MTX to CRs, having one and two –SO3−groups. Gels based on biocompatible kCR have been easily prepared and used for MTX en- ment. MTX concentration in the gel can be increased by means of βCD, which forms e inclusion complexes with MTX. Moreover, stabilization of MTX in the gel can be ved due to the inclusion complex formation with βCD. 3.7. Permeation Study Prediction of membrane permeability is necessary in the development of pharmaceu- tical formulations. MTX permeation through an artiﬁcial polyethersulfone microporous membrane (0.45 µm) that mimics the skin [47,48] was studied. It was found that the cumulative amount of free MTX released from kCR/MTX gel and passed through the membrane during 6 h was 16 wt.%. Complexation with βCD increases the MTX release rate, as is shown in Figure 8, but at the same time, reduces MTX permeation (Figure 9). It was demonstrated by means of 1H NMR spectroscopy that inclusion complexes of MTX with βCD are able to pass through the model polyethersulfone membrane. However, in comparison with the free MTX, the permeation rate of the inclusion complexes is lower due to their larger size. Permeation is governed by the stability constant of the complexes of MTX with βCD and decreases with the rise of βCD concentration (Figure 9). Therefore, gels of the basis of kCR provided the sustained release of MTX. Signiﬁcant differences in the permeation proﬁles of MTX from kCR and iCR were observed (Figure 9). This is in accordance with the faster release of MTX from kCR gel (Figure 8) and revealed the weaker binding afﬁnity of MTX to kCR (Figure 6). The area under the plasma drug concentration–time curve (AUC) obtained from Figure 9 can be used to predict in vivo AUC. Calculated values of AUC0–6.7h are given in Table 2SI. This attempt is based on a linear in vivo–in vitro correlation that has been found by M. Klitgaard et al. [49] when comparing the area under the in vitro drug permeation– time curve to the AUC of the plasma concentration–time proﬁle obtained from the in vivo Pharmaceutics 2023, 15, 2244 13 of 15 s e 13 of 15 s e study. Based on these results, the permeation method was proposed as a promising tool for estimating the in vivo performance. herefore, gels of the basis of kCR provided the sustained release of MTX. study. Based on these results, the permeation method was proposed as a promising tool for estimating the in vivo performance. erefore, gels of the basis of kCR provided the sustained release of MTX. 0 100 200 300 400 0 5 10 15 Cumulative MTX permeated (wt.%) Time, min kCR/MTX kCR/MTX/βCD_1% kCR/MTX/βCD_1.4% iCR/MTX re 9. 3.7. Permeation Study In vitro permeation profiles of MTX from gels on the basis of kCR and iCR (1.25 wt.%) at C. Figure 9. In vitro permeation proﬁles of MTX from gels on the basis of kCR and iCR (1.25 wt.%) at 32 ◦C. 9. In vitro permeation profiles of MTX from gels on the basis of kCR and iCR (1.25 wt.%) at Figure 9. In vitro permeation proﬁles of MTX from gels on the basis of kCR and iCR (1.25 wt.%) at 32 ◦C. References 1. Peppas, N.A.; Bures, P.; Leobandung, W.; Ichikawa, H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000, 50, 27–46. [CrossRef] [PubMed] 2. Hilliou, L. Structure–Elastic Properties Relationships in Gelling Carrageenans. Polymers 2021, 13, 4120. [CrossRef] [PubMed] 3. Zhu, B.; Ni, F.; Xiong, Q.; Yao, Z. Marine oligosaccharides originated from seaweeds: Source, preparation, structure, physiological activity and applications. Crit. Rev. Food Sci. Nutr. 2021, 61, 60–74. [PubMed] 2. Hilliou, L. Structure–Elastic Properties Relationships in Gelling Carrageenans. Polymers 2021, 13, 4120 p p g g y 3. Zhu, B.; Ni, F.; Xiong, Q.; Yao, Z. Marine oligosaccharides originated from seaweeds: Source, preparation, structure, physiological activity and applications. Crit. Rev. Food Sci. Nutr. 2021, 61, 60–74. [PubMed] 4. Jingjing, L.; Xiudan, Z.; Jianbo, W.; Yitao, W.; Chunming, W. Review for carrageenan-based pharmaceutical biomaterials: Favourable physical features versus adverse biological effects. Carbohydr. Polym. 2015, 121, 27–36. M.; Ruiz-Caro, R.; Veiga, M.D. Carrageenan: Drug Delivery Systems and Other Biomedical Applications 583. g 6. Frediansyah, A. The antiviral activity of iota-, kappa-, and lambda-carrageenan against COVID-19: A critical review. Clin. Epidemiol. Glob. Health 2021, 12, 100826. [CrossRef] p 7. Laurie, C.; El-Zein, M.; Botting, S.; Tota, J.E.; Tellier, P.P.; Coutlee, F.; Burchell, A.N.; Franco, E.L. Efﬁcacy and safety of a self-applied carrageenan-based gel to prevent human papillomavirus infection in sexually active young women (CATCH study): An exploratory phase IIB randomised, placebo-controlled trial. EClinicalMedicine 2023, 8, 102038. [CrossRef] p y p p 8. Khan, M.S.; Ravi, P.; Mir, S.; Rawat, P. Optimization and in vivo evaluation of triamcinolone acetonide loaded in situ gel prepared using reacted tamarind seed xyloglucan and kappa-carrageenan for ocular delivery. Int. J. Biol. Macromol. 2023, 233, 123533. [CrossRef] 9. Yermak, I.M.; Gorbach, V.I.; Karnakov, I.A.; Davydova, V.N.; Pimenova, E.A.; Chistyulin, D.A.; Isakov, V.V.; Glazunov, V.P. Carrageenan gel beads for echinochrome inclusion: Inﬂuence of structural features of carrageenan. Carbohydr. Polym. 2021, 15, 118479. 10. Miyazaki, S.; Ishitani, M.; Takahashi, A.; Shimoyama, T.; Itoh, K.; Attwood, D. Carrageenan gels for oral sustained delivery of acetaminophen to dysphagic patients. Biol. Pharm. Bull. 2011, 34, 164–166. [CrossRef] 11. Hezaveh, H.; Muhamad, I.I. Modiﬁcation and swelling kinetic study of kappa-carrageenan-based hydrogel for controlled release study. J. Taiwan Inst. Chem. Eng. 2013, 44, 182–191. 12. Azizi, S.; Mohamad, R.; Abdul Rahim, R.; Mohammadinejad, R.; Bin Ariff, A. Hydrogel beads bio-nanocomposite based on Kappa-Carrageenan and green synthesized silver nanoparticles for biomedical applications. Int. J. Biol. Macromol. 2017, 104, 423–431. References [PubMed] [ ] 13. Zhao, J.; Sun, C.; Li, H.; Dong, X.; Zhang, X. Studies on the physicochemical properties, gelling behavior and drug release performance of agar/κ-carrageenan mixed hydrogels. Int. J. Biol. Macromol. 2020, 154, 878–887. [PubMed] 13. Zhao, J.; Sun, C.; Li, H.; Dong, X.; Zhang, X. Studies on the physicochemical properties, gelling behavior and drug release performance of agar/κ-carrageenan mixed hydrogels. Int. J. Biol. Macromol. 2020, 154, 878–887. [PubMed] 14. Sonawane, R.O.; Patil, S.D. Fabrication and statistical optimization of starch-κ-carrageenan cross-linked hydrogel composite for 14. Sonawane, R.O.; Patil, S.D. Fabrication and statistical optimization of starch-κ-carrageenan cross-link extended release pellets of zaltoprofen. Int. J. Biol. Macromol. 2018, 120, 2324–2334. [PubMed] ded release pellets of zaltoprofen. Int. J. Biol. Macromol. 2018, 120, 2324–2334. [PubMed] 15. Rasool, A.; Ata, S.; Islam, A.; Rizwan, M.; Azeem, M.K.; Mehmood, A.; Mahmood, H.A. Kinetics and controlled release of lidocaine from novel carrageenan and alginate-based blend hydrogels. Int. J. Biol. Macromol. 2020, 147, 67–68. lidocaine from novel carrageenan and alginate-based blend hydrogels. Int. J. Biol. Macromol. 2020, 147, 67–68. 16. Xu, H.; Liu, Y.; Jin, L.; Chen, X.; Chen, X.; Wang, Q.; Tang, Z. Preparation and Characterization of Ion-Sensitive Brimonidine T I Si G l f O l D li Ph i l 2023 16 90 [C R f] g g y g J , , 16. Xu, H.; Liu, Y.; Jin, L.; Chen, X.; Chen, X.; Wang, Q.; Tang, Z. Preparation and Characterization of Ion-Sensitive Brimonidine T t t I Sit G l f O l D li Ph ti l 2023 16 90 [C R f] g g y g J 16. Xu, H.; Liu, Y.; Jin, L.; Chen, X.; Chen, X.; Wang, Q.; Tang, Z. Preparation and Characterization of Ion-Sensitive Brimonidine Tartrate In Situ Gel for Ocular Delivery. Pharmaceuticals 2023, 16, 90. [CrossRef] J g Q g p Tartrate In Situ Gel for Ocular Delivery. Pharmaceuticals 2023, 16, 90. [CrossRef] 17. Yegappan, R.; Selvaprithiviraj, V.; Amirthalingam, S.; Jayakumar, R. Carrageenan based hydrogels for drug delivery, tissue engineering and wound healing. Carbohydr. Polym. 2018, 198, 385–400. g g g y y 18. Postolovic, K.; Ljujic, B.; Kovacevic, M.; Dordevic, S.; Nikolic, S.; Zivanovic, S.; Stanic, Z. Optimization, characterization, and evaluation of carrageenan/alginate/poloxamer/curcumin hydrogel ﬁlm as a functional wound dressing material. Mater. Today Commun. 2022, 31, 103528. 19. Jaiswal, L.; Shankar, S.; Rhim, J.W. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 4. Conclusions It was demonstrated that ddition of βCD to the gel formulations has no significant effects on the viscoelastic Supplementary Materials: The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/pharmaceutics15092244/s1, Figure S1I: Calibration curve for spectrophotometric determination of MTX in phosphate buffer; Figure S2I: Temperature depen- dence of the storage (G′) and loss (G′′) moduli for the kCR gels (a −1 wt.% kCR; b −1.25 wt.% kCR); Figure S3I: Flow curves for gels under study; Figure S4I: FTIR spectra of freeze-dried gels under study; Figure S5I: Release proﬁles of MTX from kCR gels (1.25 wt.%) in phosphate buffer (pH = 7.4) at 37 ◦C (concentration of MTX in the gels is the same); Figure S6I: Concentration of the inclusion complexes of MTX with βCD formed in solution versus βCD concentration; Table S1I: Content of the components in gels on the basis of kCR; Table S2I: AUC0–6.7h in vitro of permeated MTX from the gels under study. Author Contributions: Conceptualization, I.T. and N.K.; methodology, M.A., N.K. and I.T.; investiga- tion, M.N., N.K. and M.A.; resources, I.T. and V.K.; data curation, N.K., I.T. and V.K.; writing—original draft preparation, N.K. and M.N.; writing—review and editing, I.T.; supervision, I.T. and V.K.; project administration, I.T.; funding acquisition, I.T. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Russian Science Foundation, grant number No. 22-23-00891. I f d C t St t t N t li bl Funding: This research was funded by Russian Science Foundation, grant number No. 22-23-00891. Funding: This research was funded by Russian Science Foundation, grant number No. 22-23-00891. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data is contained within the article or Supplementary Materials. Acknowledgments: We thank “The Upper Volga Region Centre of Physicochemical Research” (Ivanovo, Russian Federation) for equipment provided for NMR, DLS and SEM experiments. 14 of 15 14 of 15 Pharmaceutics 2023, 15, 2244 References Carrageenan-based functional hydrogel ﬁlm reinforced with sulfur nanoparticles and grapefruit seed extract for wound healing application. Carbohydr. Polym. 2019, 224, 115191. 20. Kochkina, N.; Nikitina, M.; Agafonov, M.; Delyagina, E.; Terekhova, I. iota-Carrageenan hydrogels for methotrexate delivery. J. Mol. Liq. 2022, 368, 120790. 21. Kozminski, P.; Halik, P.; Chesori, R.; Gniazdowska, E. Overview of Dual-Acting Drug Methotrexate in Different Neurological Diseases, Autoimmune Pathologies and Cancers. Int. J. Mol. Sci. 2020, 21, 3483. [CrossRef] [PubMed] o s , ; , ; C eso , ; G o s , O e e o c g g e o e e e e Ne o og c Diseases, Autoimmune Pathologies and Cancers. Int. J. Mol. Sci. 2020, 21, 3483. [CrossRef] [PubMed] se ses, o e o og es C ce s J , , 3 83 [C oss e ] [ e ] ipathi, P.; Kumar, A.; Jain, P.K.; Patel, J.R. Carbomer gel bearing methotrexate loaded lipid nanocontai g 22. Tripathi, P.; Kumar, A.; Jain, P.K.; Patel, J.R. Carbomer gel bearing methotrexate loaded lipid nanocon topical delivery intended for effective management of psoriasis. Int. J. Biol. Macromol. 2018, 120, 1322– topical delivery intended for effective management of psoriasis. Int. J. Biol. Macromol. 2018, 120, 1322 1334. [CrossRef] 23. Bahramizadeh, M.; Bahramizadeh, M.; Kiafar, B.; Jafarian, A.H.; Nikpoor, A.R.; Hatamipour, M.; Esmaily, H.; Rezaeemehr, Z.; Golmohammadzadeh, S.; Moosavian, S.A.; et al. Development, characterization and evaluation of topical methotrexate-entrapped deformable liposome on imiquimod-induced psoriasis in a mouse model. Int. J. Pharm. 2019, 569, 118623. [CrossRef] [PubMed] 24. Panonnummal, R.; Sabitha, M. Anti-psoriatic and toxicity evaluation of methotrexate loaded chitin nanogel in imiquimod induced i d l I t J Bi l M l 2018 110 245 258 [C R f] [P bM d] p y g p 23. Bahramizadeh, M.; Bahramizadeh, M.; Kiafar, B.; Jafarian, A.H.; Nikpoor, A.R.; Hatamipour, M.; Esmaily, H.; Rezaeemehr, Z.; Golmohammadzadeh, S.; Moosavian, S.A.; et al. Development, characterization and evaluation of topical methotrexate-entrapped deformable liposome on imiquimod-induced psoriasis in a mouse model. Int. J. Pharm. 2019, 569, 118623. [CrossRef] [PubMed] p q p 24. Panonnummal, R.; Sabitha, M. Anti-psoriatic and toxicity evaluation of methotrexate loaded chitin nanog mice model. Int. J. Biol. Macromol. 2018, 110, 245–258. [CrossRef] [PubMed] 25. Wang, Y.; Fu, S.; Lu, Y.; Lai, R.; Liu, Z.; Luo, W.; Xu, Y. Chitosan/hyaluronan nanogels co-delivering methotrexate and 5-aminolevulinic acid: A combined chemo-photodynamic therapy for psoriasis. Carbohydr. Polym. 25. Wang, Y.; Fu, S.; Lu, Y.; Lai, R.; Liu, Z.; Luo, W.; Xu, Y. Chitosan/hyaluronan nanogels co-deliv 5-aminolevulinic acid: A combined chemo-photodynamic therapy for psoriasis. Carbohydr. Polym. 2022 References 2022, 277, 118819. [CrossRef] Pharmaceutics 2023, 15, 2244 15 of 15 15 of 15 26. Mahdavinia, G.R.; Mosallanezhad, A.; Soleymani, M.; Sabzi, M. Magnetic- and pH-responsive κ-carrageenan/chitosan complexes for controlled release of methotrexate anticancer drug. Int. J. Biol. Macromol. 2017, 97, 209–217. [CrossRef] g J [ ] 27. Youseﬁ, G.; Foroutan, S.M.; Zarghi, A.; Shafaati, A. Synthesis and Characterization of Methotrexate Polyethylene Glycol Esters as a Drug Delivery. Syst. Chem. Pharm. Bull. 2010, 58, 147–153. [CrossRef] g y y 28. Chadha, R.; Arora, P.; Kaur, R.; Saini, A.; Singla, M.L.; Jain, D.S. Characterization of solvatomorphs of methotrexate using thermoanalytical and other techniques. Acta Pharm. 2009, 59, 245–257. [CrossRef] y q 29. Ouyang, L.; Ma, L.; Jiang, B.; Li, Y.; He, D.; Guo, L. Synthesis of novel dendrimers having aspartate grafts and their ability to enhance the aqueous solubility of model drugs. Eur. J. Med. Chem. 2010, 45, 2705–2711. [CrossRef] 30. Phatsawee, J.; Noriko, O.; Thorsteinn, L. Cyclodextrins: Structure, physicochemical properties and pharmaceutical applications. Int. J. Farm. 2018, 535, 272–284. 30. Phatsawee, J.; Noriko, O.; Thorsteinn, L. Cyclodextrins: Structure, physicochemical properties and pharmaceutical applications. Int. J. Farm. 2018, 535, 272–284. 31 Periasamy R Cyclodextrin-based molecules as hosts in the formation of supramolecular complexes and their practical 31. Periasamy, R. Cyclodextrin-based molecules as hosts in the formation of supramolecular compl applications—A review. J. Carbohydr. Chem. 2021, 40, 135–155. [CrossRef] pp 32. Kritskiy, I.; Kumeev, R.; Volkova, T.; Shipilov, D.; Kutyasheva, N.; Grachev, M.; Terekhova, I. Selective binding of methotrexate to monomeric, dimeric and polymeric cyclodextrins. New J. Chem. 2018, 42, 14559–14567. [CrossRef] p y y J 33. Giri, B.R.; Yang, H.S.; Song, I.S.; Choi, H.G.; Cho, J.H.; Kim, D.W. Alternative Methotrexate Oral Formulation: Enhanced Aqueous Solubility, Bioavailability, Photostability, and Permeability. Pharmaceutics 2022, 14, 2073. [CrossRef] [PubMed] y y y y 34. Ross-Murphy, S.B.; Shatwell, K.P. Polysaccharide strong and weak gels. Biorheology 1993, 30, 217–227. [CrossRef] [PubMed] 35. Whistler, R.L. Industrial Gums. In Polysaccharides and Their Derivatives, 3rd ed.; Academic Press: New York, NY, USA, 1993; pp. 145–180. 34. Ross-Murphy, S.B.; Shatwell, K.P. Polysaccharide strong and weak gels. Biorheology 1993, 30, 217–227. 35. Whistler, R.L. Industrial Gums. In Polysaccharides and Their Derivatives, 3rd ed.; Academic Press: New York, NY, USA, 1993; pp. 145–180. pp 36. Rees, D.A.; Williamson, F.B.; Frangou, S.A.; Morris, E.R. Fragmentation and modiﬁcation of iota-carrageenan and characterisation of the polysaccharide order-disorder transition in solution. Eur. J. Biochem. 1982, 122, 71–79. References [CrossRef] p y Pereira, M.C.; Wyn-Jones, E.; Morris, E.R.; Ross-Murphy, S.B. Characterisation of interchain associati l i b l i l i d l i C b h d P l 2 103 113 [C R f] 37. Pereira, M.C.; Wyn-Jones, E.; Morris, E.R.; Ross-Murphy, S.B. Characterisation of interchain association in polysaccharide solutions by ultrasonic relaxation and velocity. Carbohydr. Polym. 1982, 2, 103–113. [CrossRef] 37. Pereira, M.C.; Wyn-Jones, E.; Morris, E.R.; Ross-Murphy, S.B. Characterisation of interchain association in polysaccharide solutions by ultrasonic relaxation and velocity. Carbohydr. Polym. 1982, 2, 103–113. [CrossRef] 38. Szakacs, Z.; Noszal, B. Determination of dissociation constants of folic acid, methotrexate, and other photolabile pteridines by pressure-assisted capillary electrophoresis. Electrophoresis 2006, 27, 3399–3409. [CrossRef] [PubMed] p p y p 39. Alves, L.; Medronho, B.F.; Antunes, F.E.; Romano, A.; Miguel, M.G.; Lindman, B. On the role of hydrophobic interactions in cellulose dissolution and regeneration: Colloidal aggregates and molecular solutions. Colloids Surf. A Physicochem. Eng. Asp. 2015, 483, 257–263. [CrossRef] 40. Yuan, C.; Sang, L.; Wang, Y.; Cui, B. Inﬂuence of cyclodextrins on the gel properties of kappa-carrageenan. Food Chem. 2018, 266, 545–550. [CrossRef] 41. Yuan, C.; Du, L.; Zhang, G.; Jin, Z.; Liu, H. Inﬂuence of cyclodextrins on texture behavior and freeze-thaw stability of kappa- carrageenan gel. Food Chem. 2016, 210, 600–605. [CrossRef] , Y.; Yuan, C.; Liu, Y.; Xu, D.; Cui, B. The inﬂuence of a hydroxypropyl-beta-cyclodextrin composite on t 42. Wang, Y.; Yuan, C.; Liu, Y.; Xu, D.; Cui, B. The inﬂuence of a hydroxypropyl-beta-cyclodextrin c kappa-carrageenan. Food Hydrocoll. 2019, 90, 276–284. [CrossRef] 42. Wang, Y.; Yuan, C.; Liu, Y.; Xu, D.; Cui, B. The inﬂuence of a hydroxypropyl- kappa-carrageenan. Food Hydrocoll. 2019, 90, 276–284. [CrossRef] pp g y 43. Pang, B.; Wang, S.; Chen, W.; Hassan, M.; Lu, H. Effects of ﬂow behavior index and consistency coefﬁcient on hydrodynamics of power-law ﬂuids and particles in ﬂuidized beds. Powder Technol. 2020, 366, 249–260. [CrossRef] 43. Pang, B.; Wang, S.; Chen, W.; Hassan, M.; Lu, H. Effects of ﬂow behavior index and consistency coefﬁci ﬂ ﬂ d h l f 44. Antonov, Y.A.; Zhuravleva, I.L.; Cardinaels, R.; Moldenaers, P. Macromolecular complexes of lysozyme with kappa carrageenan. Food Hydrocoll. 2018, 74, 227–238. [CrossRef] y Um, I.C.; Ki, C.S.; Kweon, H.Y.; Lee, K.G.; Ihm, D.W.; Park, Y.H. Wet spinning of silk polymer II. Effect of dra characteristics and properties of ﬁlament. Int. J. Biol. Macromol. 2004, 34, 107–119. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. 49. Klitgaard, M.; Müllertz, A.; Berthelsen, R. Estimating the Oral Absorption from Self-Nanoemulsifying Drug Delivery Systems Using an In Vitro Lipolysis-Permeation Method. Pharmaceutics 2021, 13, 489. [CrossRef] References [CrossRef] 45. Um, I.C.; Ki, C.S.; Kweon, H.Y.; Lee, K.G.; Ihm, D.W.; Park, Y.H. Wet spinning of silk polymer II. Effect of drawing on the structure characteristics and properties of ﬁlament. Int. J. Biol. Macromol. 2004, 34, 107–119. [CrossRef] 46. Peppas, N.A.; Sahlin, J.J. A simple equation for the description of solute release. III. Coupling of diffusion and relaxation. Int. J. Pharm. 1989, 57, 169–172. [CrossRef] 47. Fiorentin-Ferrari, L.D.; Celant, K.M.; Goncalves, B.C.; Teixeira, S.M.; Slusarski-Santana, V.; Modenes, A.N. Fabrication and characterization of polysulfone and polyethersulfone membranes applied in the treatment of ﬁsh skin tanning efﬂuent. J. Clean. Prod. 2021, 294, 126127. [CrossRef] 48. Neupane, R.; Boddu Sai, H.S.; Renukuntla, J.; Babu, R.J.; Tiwari, A. Alternatives to Biological Skin in Permeation Studies: Current Trends and Possibilities. Pharmaceutics 2020, 12, 152. [CrossRef] 49. Klitgaard, M.; Müllertz, A.; Berthelsen, R. Estimating the Oral Absorption from Self-Nanoemulsifying Drug Delivery Systems Using an In Vitro Lipolysis-Permeation Method. Pharmaceutics 2021, 13, 489. [CrossRef] 49. Klitgaard, M.; Müllertz, A.; Berthelsen, R. Estimating the Oral Absorption from Self-Nanoemulsifyin Using an In Vitro Lipolysis-Permeation Method. Pharmaceutics 2021, 13, 489. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W2118259673	https://jmedicalcasereports.biomedcentral.com/track/pdf/10.1186/1752-1947-1-115	English	null	Unsuspected Pneumocystis pneumonia in an HIV-seronegative patient with untreated lung cancer: circa case report	Journal of medical case reports	2,007	cc-by	3,513	Case report Address: Guangzhou Institute of Respiratory Disease, First Affiliated Hospital of Guangzhou Medical University, Guangzhou 510012, China. Email: Cai Chuang* - skinblack1966@yahoo.com.cn; Xie Zhanhong - Drxiezhanhong@163.com; Gu Yinyin - gyblgyy@126.com; Zeng Qingsi - qingsi21@21cn.com; Zhong Shuqing - drzhongshuqing@163.com; Zhong Nanshan - nanshan@vip.163.com * Corresponding author Published: 28 October 2007 Journal of Medical Case Reports 2007, 1:115 doi:10.1186/1752-1947-1-115 Received: 22 June 2007 Accepted: 28 October 2007 This article is available from: http://www.jmedicalcasereports.com/content/1/1/115 © 2007 Chuang et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Received: 22 June 2007 Accepted: 28 October 2007 Received: 22 June 2007 Accepted: 28 October 2007 g ; This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Background: Patients with solid malignant tumours are at increased risk of Pneumocystis jiroveci infection from immunosuppression as a result of chemotherapy and/or radiotherapy, but active Pneumocystis pneumonia (PCP) in untreated lung cancer is uncommon. Case presentation: A 43-year-old woman presented with prolonged fever, progressive dyspnoea, diffuse alveolar and interstitial infiltrates. Malignant cells were found on sputum cytology, confirming the diagnosis of lung cancer. She had been treated with corticosteroids and antibiotics but did not receive chemotherapy or radiotherapy. Pneumocystis jiroveci was later found in the sputum but she proved to be HIV negative. Conclusion: Unsuspected PCP can occur in chemotherapy and radiotherapy-naïve, HIV- seronegative patients with lung cancer. The complex clinicoradiological manifestations of PCP with underlying lung cancer can lead to delay in diagnosis and may worsen the prognosis. Page 1 of 5 (page number not for citation purposes) Open A Case report Unsuspected Pneumocystis pneumonia in an HIV-seronegative patient with untreated lung cancer: circa case report Cai Chuang*, Xie Zhanhong, Gu Yinyin, Zeng Qingsi, Zhong Shuqing and Zhong Nanshan Open Access Background Chest CT displaying diffuse nodular parenchymal infiltrates with partial confluence Figure 2 Chest CT displaying diffuse nodular parenchymal infiltrates with partial confluence. Diffuse bronchioloalveolar carcinoma (BAC) complicated by acute respiratory failure was suspected, malignancy was confirmed with evidence of malignant cells repeatedly found in sputum smears examined by a lung pathologist (Dr Gu YY). Intubation and invasive mechanical ventila- tion was strongly recommended because of her critical condition, but was declined by the patient and her family. Non-invasive ventilation (CPAP) was given as an alterna- tive to treat the respiratory failure, with imipenem/cilastin to target the suspected bacterial pneumonia. Her orthop- noea and high fever remained despite treatment, with peripheral WBC elevated to 22.6 × 109/L, granulocytes 95%, lymphocytes 5%, yet repeated blood and sputum cultures for bacteria were negative. Empiric vancomycin was added for possible gram-positive cocci infections to no avail. Chest radiograph (Fig. 1) revealed diffuse nodular and patchy consolidations with predominance in the lower lung fields, and right hilar expansion. Chest CT scanning (Fig. 2) demonstrated diffuse alveolar and interstitial infiltrates with partial confluence and adenopathy in both hilar regions. CT angiography was negative for pulmonary embolism. Diffuse bronchioloalveolar carcinoma (BAC) complicated by acute respiratory failure was suspected, malignancy was confirmed with evidence of malignant cells repeatedly found in sputum smears examined by a lung pathologist (Dr Gu YY). Intubation and invasive mechanical ventila- tion was strongly recommended because of her critical condition, but was declined by the patient and her family. Non-invasive ventilation (CPAP) was given as an alterna- tive to treat the respiratory failure, with imipenem/cilastin to target the suspected bacterial pneumonia. Her orthop- noea and high fever remained despite treatment, with peripheral WBC elevated to 22.6 × 109/L, granulocytes 95%, lymphocytes 5%, yet repeated blood and sputum cultures for bacteria were negative. Empiric vancomycin was added for possible gram-positive cocci infections to no avail. Chest radiograph showing diffuse nodular and patchy pulmo- nary infiltrates, and right hilar widening Figure 1 Chest radiograph showing diffuse nodular and patchy pulmo- nary infiltrates, and right hilar widening. Four days after admission, clusters of Pneumocystis jiroveci cysts were unexpectedly and repeatedly identified in her sputa using methamine silver stain (Fig. 3). The patient was therefore diagnosed with PCP. Background p A 43-year-old nonsmoking female was referred to our hospital for persistent fever, progressive dyspnoea, and diffuse alveolar and interstitial infiltrates. The patient complained of intermittent mild to moderate fever, pro- gressive dyspnoea, increasing exportation and paroxysmal wheezing for 6 weeks. Her past history was unremarkable, without known exposure to occupational or environmen- tal hazards. She was initially diagnosed as community acquired pneumonia in her local hospital, treated with intravenous levofloxacin, which was escalated to cephatri- axone when her symptoms worsened. Intravenous dex- amethasone (10 mg to 20 mg prn) with aminophylline was administered irregularly to relieve her dyspnoea and Pneumocystis jiroveci pneumonia (PCP, formerly known as Pneumocystis carinii pneumonia) has been increas- ingly reported as a severe opportunistic infection in HIV- seronegative patients with solid tumours (brain, lung, breast and ovarian cancer), as a sequel to severe immuno- suppression from chemotherapy and/or radiotherapy with or without corticosteroids [1-5]. We present a case of unsuspected PCP in an HIV-seroneg- ative patient with untreated lung cancer, manifested as persistent fever, progressive dyspnoea, and diffuse alveo- lar and interstitial infiltrates. Page 1 of 5 (page number not for citation purposes) Page 1 of 5 (page number not for citation purposes) http://www.jmedicalcasereports.com/content/1/1/115 Journal of Medical Case Reports 2007, 1:115 Chest CT displaying diffuse nodular parenchymal infiltrates with partial confluence Figure 2 Chest CT displaying diffuse nodular parenchymal infiltrates with partial confluence. wheezing. She developed orthopnoea with high fever and copious gel-like phlegm four weeks after the onset of symptoms, so she was transferred to our hospital. On admission, she displayed orthopnoea, cyanosis, finger clubbing and nonpitting edoema in the lower extremities. Vital signs: body temperature 38.8°C, HR 124 beats/min, respiratory rate 32 breath/min, and Bp 124/76 mmHg. Fine crackles were audible at both bases with resonant wheezing throughout inspiration and expiration. Superfi- cial lymph nodes, heart and abdomen were unremarka- ble. Laboratory workup: complete blood count revealed WBC elevation with WBC 15.4 ×109/L, granulocytes 86%, lym- phocytes 14%, RBC and PLT were in normal range. Serum liver and renal biochemistry as well as electrolytes were unremarkable. Serum CEA 13.4 µg/L (0–5 µg/L), D-dimer 680 ng/mL (< 200 ng/mL). Arterial blood gas analysis (nasal oxygen, 3 L/min): pH 7.42, PaO26.54 kPa, PaCO2 3.86 kPa. Chest CT displaying diffuse nodular parenchymal infiltrates with partial confluence Figure 2 Chest CT displaying diffuse nodular parenchymal infiltrates with partial confluence. Background Oral TMP-SMX (2 double-strength tab, tid, as intravenous pentamdine or trimethoprim/sulfamethoxazole was not available in Guangzhou), with intravenous methylprednisolone 40 mg qd as adjunctive therapy, was initiated with the origi- nal antibiotic regimen still continued. Further investiga- tions found that her peripheral blood CD4 T lymphocyte was 189/µL with CD4/CD8 ratio as 1.4: 1, C reactive pro- tein (CRP) was 26 mg/L (0–8 mg/L), serum lactate dehy- drogenase (LDH) was 371 IU/L (35–90 IU/L). She tested negative for HIV and cytomegalovirus. Her family mem- bers including her husband and children were also HIV- seronegative. When 7 days' treatment with co-trimoxazole Chest radiograph showing diffuse nodular and patchy pulmo- nary infiltrates, and right hilar widening Figure 1 Chest radiograph showing diffuse nodular and patchy pulmo- nary infiltrates, and right hilar widening. Chest radiograph showing diffuse nodular and patchy pulmo- nary infiltrates, and right hilar widening Figure 1 Chest radiograph showing diffuse nodular and patchy pulmo- nary infiltrates, and right hilar widening. Page 2 of 5 (page number not for citation purposes) Page 2 of 5 (page number not for citation purposes) Journal of Medical Case Reports 2007, 1:115 http://www.jmedicalcasereports.com/content/1/1/115 case is the first to document active unsuspected PCP in an untreated, HIV- seronegative patient with lung cancer. Methamine silver stain demonstrating clusters of Pneumo- cystis jiroveci cysts in the sputum (×100) Figure 3 Methamine silver stain demonstrating clusters of Pneumo- cystis jiroveci cysts in the sputum (×100) Though fever, dry cough and progressive dysponea with pulmonary infiltrates of alveolar pattern are typical of PCP, these manifestations are nonspecific. Definitive diagnosis of PCP requires identification of Pneumocystis jiroveci in respiratory secretions, bronchial alveolar lavage fluid or biopsied lung tissues [1,10]. When the presenta- tion is complicated by the presence of underlying lung cancer, the diagnosis of PCP becomes more challenging as the clinical picture becomes more atypical and confusing. In the present case, the differential diagnosis for the fever, orthopnoea and diffuse alveolar and interstitial infiltrates, should include other superimposed infections, pulmo- nary embolism and broncholymphatic dissemination of the neoplasm. Both PCP and advanced lung cancer might have contributed to progressive dyspnoea and refractory respiratory failure, as well as the radiological pattern of infiltrates, but the manifestation of copious glutinous sputum is atypical of PCP, and points to lung cancer, most possibly BAC. Background Methamine silver stain demonstrating clusters of Pneumo- cystis jiroveci cysts in the sputum (×100) Figure 3 Methamine silver stain demonstrating clusters of Pneumo- cystis jiroveci cysts in the sputum (×100) failed to alleviate her respiratory distress and high fever, her family opted to withdraw her from therapy and obtained her discharge. She died of respiratory failure three days later. Although this HIV-seronegative patient was radiotherapy and chemotherapy naïve, she displayed impaired cellular immunity with significant reduction of peripheral CD4 T lymphocytes. This lymphocytic subset is believed to play a crucial role in the pathogenesis of PCP by orchestrating host immune response responsible for eradication of Pneumocystis jiroveci. Reduction of peripheral CD4 T lym- phocytes below 200/µL is associated with a higher inci- dence of PCP in both AIDS and non-AIDS patients [1,10,11]. The mechanism for CD4 T lymphocyte decline without lymphopenia in this patient was unclear, the pre- ceding exposure to high dose dexamethasone might have been an important contributing factor, since corticoster- oids can reduce CD4 T lymphocytes numbers and inhibit their functions [1,2,10], thus an important risk factor for colonization of Pneumocystis jiroveci in patients with chronic obstructive pulmonary disease or lung cancer [8]. And chronic steroid therapy in patients with brain tumours, connective collagen diseases, or even in child with asthma, is frequently associated with active PCP, as a complication of severe immunosuppression from pro- longed steroid use [1,2,8,12]. In addition, host cellular immunity may be compromised by neoplasms which inhibit the activation of T lymphocytes [13]. References Mahindra AK, Grossman SA: Pneumocystis carinii pneumonia in HIV negative patients with primary brain tumors. J Neuroon- col 2003, 63:263-265. 3. Velcheti V, Govindan R: Pneumocystis pneumonia in a patient with non-small cell lung cancer (NSCLC) treated with peme- trexed containing regimen. Lung Cancer 2007, 57:240-242. g g g 4. Klippstein A, Schneider CP, Sayer HG, Hoffken K: Pneumocystis carinii pneumonia as a complication of bendamustine mono- therapy in a patient with advanced progressive breast can- cer. J Cancer Res Clin Oncol 2003, 129:316-319. 5. Watanabe M, Aoki Y, Kurata H, Tanaka K: Pneumocystis carinii pneumonia in a patient with stage IV ovarian cancer. Gynecol Oncol 2002, 87:225-227. Discussion The patient presented with progressive dyspnoea, copious glutinous sputum, diffuse alveolar and interstitial infil- trates on chest radiograph and CT. These are consistent with the clinicoradiological features of the diffuse form of BAC [6]. In addition, with the cytological evidence of malignancy, a clinical diagnosis of diffuse form of BAC could be reached. However, as the definitive diagnosis of BAC requires histological evidence of malignancy from resected or biopsied lung tissues, whereas in the present case, invasive procedures such as transbronchial lung biopsy were precluded due to the patient's critical condi- tion and lack of invasive mechanical ventilation to ensure reliable oxygenation. So, for this patient, the appropriate diagnosis is lung cancer, most likely, diffuse form of BAC. PCP is relatively uncommon in lung cancer, as a sequel to immunosuppression, it had been reported sporadically in lung cancer patients [3,7]. All these patients had under- gone chemotherapy or radiotherapy with or without cor- ticosteroids, before the onset of PCP. Recently, Maskell et al revealed asymptomatic colonization of Pneumocystis jiroveci in lung cancer patients by fibrobronchoscopy, and de la Horra et al identified 12 cases of unsuspected sub- clinical Pneumocystis jiroveci infection using polymerase chain reaction in fixed lung samples from hospitalized patients died of lung carcinoma untreated by radiother- apy or chemotherapy [8,9]. To our knowledge, the present Severe cellular immunodeficiency resulting from systemic corticosteroids treatment must have led to activation of latent Pneumocystis jiroveci infection in our patient, as recent evidence suggests that approximately 20% of patients with chronic lung diseases including lung cancer are asymptomatic carriers of Pneumocystis jiroveci [8], which progressed to severe active PCP following high- Page 3 of 5 (page number not for citation purposes) Page 3 of 5 (page number not for citation purposes) http://www.jmedicalcasereports.com/content/1/1/115 http://www.jmedicalcasereports.com/content/1/1/115 Journal of Medical Case Reports 2007, 1:115 Competing interests dose dexamethasone treatment, since other transmission pathways, such as nosocomial infection or cross-transmis- sion among family members were less likely. The author(s) declare that they have no competing inter- ests. Authors' contributions As this case demonstrates, significant increase of LDH and CRP reflecting severe systematic inflammation, and severe hypoxemia requiring mechanical ventilation may be asso- ciated with a poor clinical outcome. These parameters have been reported to be prognostic predictors of PCP [1,10,14]. Drs C.C, X.ZH, Z.SQ, Z.NS were responsible for clinical management of the patient, and collection and interpreta- tion of clinical data, Dr G.YY participated in the interpre- tation of histology and microbiology data, Dr Z.QS was in charge of the radiological manifestations of lung cancer and Pneumocystis jiroveci pneumonia. All participated in the discussion of the present case and contributed to the drafting of the manuscript. Possible resistance to co-trimoxazole by infective Pneu- mocystis jiroveci, concurrent bacterial infections, progres- sion of advanced carcinoma especially by broncholymphatic dissemination might also have con- tributed to the treatment failure. Acknowledgements We are indebted to the patient and her family for their understanding and written consent to allow us to publish the present data. Despite the increasing awareness of active PCP in severely immunosuppressed non-AIDS patients, unsuspected PCP cases are still not uncommon due to an increase in the sus- ceptible population, and the complexities of diagnosis, especially when masked by the manifestations of underly- ing diseases and their complications. For example, the clinicoradiological manifestations of our patient might have been reasonably and convincingly explained by BAC and concurrent bacterial infections if Pneumocystis jiroveci had not been found in the sputum. Similarly, 9 cases of unsuspected active PCP were retrospectively iden- tified in 50 infants with PCP as presenting manifestation of severe combined immunodeficiency [15]. Under such circumstances, close monitoring of peripheral CD4 T lym- phocyte count and serial examination of respiratory sam- ples for Pneumocystis jiroveci might be helpful for earlier diagnosis and improving prognosis. We are also indebted to Professor Jeanne Bell, from University of Edin- burgh for her precious opinion on the manuscript. The patient's next of kin (the husband) has given written informed consent for the publication of this case report and any accompanying images. A copy of the written consent is ready for the review by the editor-in-chief of this journal. References References 1. Russian DA, Levine SJ: Pneumocystis carinii pneumonia in patients without HIV infection. Am J Med Sci 2001, 321:56-65. 2. Mahindra AK, Grossman SA: Pneumocystis carinii pneumonia in HIV negative patients with primary brain tumors. J Neuroon- col 2003, 63:263-265. 3. Velcheti V, Govindan R: Pneumocystis pneumonia in a patient with non-small cell lung cancer (NSCLC) treated with peme- trexed containing regimen. Lung Cancer 2007, 57:240-242. 4. Klippstein A, Schneider CP, Sayer HG, Hoffken K: Pneumocystis carinii pneumonia as a complication of bendamustine mono- therapy in a patient with advanced progressive breast can- cer. J Cancer Res Clin Oncol 2003, 129:316-319. 5. Watanabe M, Aoki Y, Kurata H, Tanaka K: Pneumocystis carinii pneumonia in a patient with stage IV ovarian cancer. Gynecol Oncol 2002, 87:225-227. 6. Gotway MB: The diffuse form of bronchioloalveolar carci- noma. Clinl Pulm Med 2003, 10:59-62. 7. Remiszewski P, Slodkowska J, Wiatr E, Zych J, Radomski P, Rowinska- Zakrzewska E: Fatal infection in patients treated for small cell lung cancer in the Institute of Tuberculosis and Chest Dis- eases in the years 1980–1994. Lung Cancer 2001, 31:101-110. 8. Maskell NA, Waine DJ, Lindley A, Pepperell JC, Wakefield AE, Miller RF, Davies RJ: Asymptomatic carriage of Pneumocystis jiroveci in subjects undergoing bronchoscopy: a prospective study. Thorax 2003, 58:594-597. 9. de la Horra C, Varela JM, Fernandez-Alonso J, Medrano FJ, Respaldiza N, Montes-Cano MA, Calderon EJ: Association between human pneumocystis infection and small-cell lung carcinoma. Eur J Clin Invest 2004, 34:229-235. 10. Thomas CF Jr, Limper AH: Pneumocystis pneumonia. N Engl J Med 2004, 350:2487-2498. 11. Mansharamani NG, Balachandran D, Vernovsky I, Garland R, Koziel H: Peripheral blood CD4 + T-lymphocyte counts during Pneumocystis carinii pneumonia in immunocompromised patients without HIV infection. Chest 2000, 118:712-720. 12. Lemaire CM, Browning JC, Hsu S: Medical Pearl: Pneumocystis pneumonia prophylaxis for patients on chronic systemic cor- ticosteroids. J Am Acad Dermatol 2006, 55:124-125. 13. Evans C, Dalgleish AG, Kumar D: Immune suppression and colorectal cancer. Aliment Pharmacol Ther 2006, 24:1163-1177. 1. Russian DA, Levine SJ: Pneumocystis carinii pneumonia in patients without HIV infection. Am J Med Sci 2001, 321:56-65. 1. Russian DA, Levine SJ: Pneumocystis carinii pneumonia in patients without HIV infection. Am J Med Sci 2001, 321:56-65. 2. Mahindra AK, Grossman SA: Pneumocystis carinii pneumonia in HIV negative patients with primary brain tumors. J Neuroon- col 2003, 63:263-265. 2. Conclusion The present case demonstrated the co-morbidity of PCP in a chemotherapy and radiotherapy-naïve, HIV-seronega- tive lung cancer patient with grave clinical outcome. Prompt diagnosis of PCP as a sequel of lung cancer can be difficult because of overlap in the clinicoradiological manifestations, and in the disturbed cellular immunolog- ical profile. Clinical and microbiological evaluations are indicated in those at high risk of cellular immunity defi- ciency, including subjects who undergo prolonged treat- ment with high dose corticosteroids. 6. Gotway MB: The diffuse form of bronchioloalveolar carci- noma. Clinl Pulm Med 2003, 10:59-62. 7. Remiszewski P, Slodkowska J, Wiatr E, Zych J, Radomski P, Rowinska- Zakrzewska E: Fatal infection in patients treated for small cell lung cancer in the Institute of Tuberculosis and Chest Dis- eases in the years 1980–1994. Lung Cancer 2001, 31:101-110. 8. Maskell NA, Waine DJ, Lindley A, Pepperell JC, Wakefield AE, Miller RF, Davies RJ: Asymptomatic carriage of Pneumocystis jiroveci in subjects undergoing bronchoscopy: a prospective study. Thorax 2003, 58:594-597. y 9. de la Horra C, Varela JM, Fernandez-Alonso J, Medrano FJ, Respaldiza N, Montes-Cano MA, Calderon EJ: Association between human pneumocystis infection and small-cell lung carcinoma. Eur J Clin Invest 2004, 34:229-235. 14. Mansharamani NG, Garland R, Delaney D, Koziel H: Management and outcome patterns for adult pneumocystis carinii pneu- monia, 1985 to 1995: comparison of HIV-associated cases to other immunocompromised states. Chest 2000, 118:704-711. Journal of Medical Case Reports 2007, 1:115 p 15. Berrington JE, Flood TJ, Abinun M, Galloway A, Cant AJ: Unsus- pected Pneumocystis carinii pneumonia at presentation of severe primary immunodeficiency. Arch Dis Child 2000, 82:144-147. List of abbreviations 10. Thomas CF Jr, Limper AH: Pneumocystis pneumonia. N Engl J Med 2004, 350:2487-2498. List of abbreviations BAC = bronchioloalveolar carcinoma List of abbreviations BAC = bronchioloalveolar carcinoma 11. Mansharamani NG, Balachandran D, Vernovsky I, Garland R, Koziel H: Peripheral blood CD4 + T-lymphocyte counts during Pneumocystis carinii pneumonia in immunocompromised patients without HIV infection. Chest 2000, 118:712-720. CRP = C reactive protein p 12. Lemaire CM, Browning JC, Hsu S: Medical Pearl: Pneumocystis pneumonia prophylaxis for patients on chronic systemic cor- ticosteroids. J Am Acad Dermatol 2006, 55:124-125. LDH = lactate dehydrogenase J 13. Evans C, Dalgleish AG, Kumar D: Immune suppression and colorectal cancer. Aliment Pharmacol Ther 2006, 24:1163-1177. 13. Evans C, Dalgleish AG, Kumar D: Immune suppression and colorectal cancer. Aliment Pharmacol Ther 2006, 24:1163-1177. Page 4 of 5 (page number not for citation purposes) Page 4 of 5 (page number not for citation purposes) http://www.jmedicalcasereports.com/content/1/1/115 http://www.jmedicalcasereports.com/content/1/1/115 http://www.jmedicalcasereports.com/content/1/1/115 Journal of Medical Case Reports 2007, 1:115 Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 5 of 5 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge 14. Mansharamani NG, Garland R, Delaney D, Koziel H: Management and outcome patterns for adult pneumocystis carinii pneu- monia, 1985 to 1995: comparison of HIV-associated cases to other immunocompromised states. Chest 2000, 118:704-711. 15. Berrington JE, Flood TJ, Abinun M, Galloway A, Cant AJ: Unsus- pected Pneumocystis carinii pneumonia at presentation of severe primary immunodeficiency. Arch Dis Child 2000, 82:144-147.
https://openalex.org/W3044656463	https://bmcgenomdata.biomedcentral.com/track/pdf/10.1186/s12863-020-00885-9	English	null	Characterization, identification and expression profiling of genome-wide R-genes in melon and their putative roles in bacterial fruit blotch resistance	BMC genomic data	2,020	cc-by	9,372	Islam et al. BMC Genetics (2020) 21:80 https://doi.org/10.1186/s12863-020-00885-9 Islam et al. BMC Genetics (2020) 21:80 https://doi.org/10.1186/s12863-020-00885-9 Open Access Characterization, identification and expression profiling of genome-wide R- genes in melon and their putative roles in bacterial fruit blotch resistance Md. Rafiqul Islam1,2, Mohammad Rashed Hossain1,3, Denison Michael Immanuel Jesse1, Hee-Jeong Jung1, Hoy-Taek Kim1, Jong-In Park1 and Ill-Sup Nou1* * Correspondence: nis@scnu.ac.kr; nis@sunchon.ac.kr p ; 1Department of Horticulture, Sunchon National University, Suncheon, Jeonnam 57922, Republic of Korea Full list of author information is available at the end of the article Abstract Background: Bacterial fruit blotch (BFB), a disease caused by Acidovorax citrulli, results in significant economic losses in melon. The causal QTLs and genes for resistance to this disease have yet to be identified. Resistance (R)- genes play vital roles in resistance to plant diseases. Since the complete genome sequence of melon is available and genome-wide identification of R-genes has been performed for this important crop, comprehensive expression profiling may lead to the identification of putative candidate genes that function in the response to BFB. Results: We identified melon accessions that are resistant and susceptible to BFB through repeated bioassays and characterized all 70 R-genes in melon, including their gene structures, chromosomal locations, domain organizations, motif distributions, and syntenic relationships. Several disease resistance-related domains were identified, including NBS, TIR, LRR, CC, RLK, and DUF domains, and the genes were categorized based on the domains of their encoded proteins. In addition, we profiled the expression patterns of the genes in melon accessions with contrasting levels of BFB resistance at 12 h, 1 d, 3 d, and 6 d after inoculation with A. citrulli. Six R- genes exhibited consistent expression patterns (MELO3C023441, MELO3C016529, MELO3C022157, MELO3C022146, MELO3C025518, and MELO3C004303), with higher expression levels in the resistant vs. susceptible accession. Conclusion: We identified six putative candidate R-genes against BFB in melon. Upon functional validation, these genes could be targeted for manipulation via breeding and biotechnological approaches to improve BFB resistance in melon in the future. Keywords: BFB, Candidate gene, Expression, Resistance, Melon, NBS-LRR, qRT-PCR B, Candidate gene, Expression, Resistance, Melon, NBS-LRR, qRT-PCR * Correspondence: nis@scnu.ac.kr; nis@sunchon.ac.kr Background cabbage [39], maize [40], wheat [41], Arabidopsis thali- ana [42], and apple [43]. g Melon (Cucumis melo L.) is a highly diversified eudicot diploid (2n = 2x = 24) cucurbitaceous crop with a gen- ome size of approximately 375 Mb [1]. Melon is eco- nomically important and ranks as the 9th most cultivated horticultural crop in terms of worldwide pro- duction [2, 3]. Its sweet, musky-flavored, fleshy fruit is rich in vitamins, minerals, and health-promoting antioxi- dants, including ascorbic acid, carotene, folic acid, and potassium [4–6]. An improved assembly and annotation of the melon (Cucumis melo L.) reference genome identified 70 R- genes in melon [1, 44, 45]. In the current study, we in- vestigated the expression patterns of R-genes throughout the melon genome in response to the BFB-causing bac- terium A. citrulli in melon accessions contrasting in BFB resistance. The aim of this study was to identify putative candidate R-genes that confer resistance to BFB in melon. p Melon is vulnerable to various biotic and abiotic stresses [7, 8]. Bacterial fruit blotch (BFB) is a devastat- ing disease of melon caused by Acidovorax citrulli, an aerobic, mesophilic, gram-negative, rod-shaped seed- borne bacterium belonging to the beta subdivision of the Proteobacteria [9]. BFB has been reported in many countries and poses a serious threat to melon, as well as other cucurbit crops including prickly paddy melon, cit- ron melon, cucumber, pumpkin, squash, several types of gourds, and watermelon [10–16]. BFB causes water- soaked lesions to form on cotyledons and leaves, leading to collapse and death. The lesions on fruits are small (~ 1 cm diameter), irregular, and often sunken, progressing through the rind. The lesions then become necrotic, causing decay and cracks in the fruit. These lesions ex- pose the plant to secondary infections and cause A. citrulli to colonize the pulp, eventually allowing the seed to become contaminated [17]. BFB causes 80–100% losses in production under favorable environmental situ- ations, especially during the rainy season and in regions with highly fluctuating temperatures [18, 19]. Although BFB is of great concern to farmers and seed companies, strategies for managing this disease are limited; chemical control measures are environmentally hazardous and only partially effective, and resistant commercial culti- vars have not yet been developed [13, 20–24]. Host re- sistance represents the most cost effective and environmentally friendly approach for managing BFB [12]. Genome-wide melon R-genes and their chromosomal distribution The latest version (v3.6.1) of the whole-genome se- quence of the melon double haploid line DHL90 was constructed using an improved assembly and annotation. This sequence contains 70 R-genes [1, 44]. We retrieved genomic information for these 70 R-genes, including their coding sequences and deduced amino acid se- quences, from the cucurbit genome database (http:// cucurbitgenomics.org). Detailed genomic information about these R-genes, including their locations on chro- mosomes, is provided in Table 1. Chromosomal map- ping of the 70 R-genes revealed that they are distributed across all 12 melon chromosomes, with 1 to 12 genes per chromosome (Fig. 1; Table 1). Chromosome 9 (Chr09) contains the most R-genes (12), followed by Chr12 and Chr01 (10 and 9 genes, respectively). Chr11 contains the fewest R-genes (2), followed by Chr03 and Chr07 (3 genes each). The genes appear to be clustered, particularly in the telomere regions of chromosomes such as chr09, chr01, and chr04 (Fig. 1). Background However, no QTL or R-gene for this disease has thus far been identified in melon. Efforts to develop BFB-resistant melon genotypes would be greatly en- hanced by the identification of functional R-genes. © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Islam et al. BMC Genetics (2020) 21:80 Islam et al. BMC Genetics (2020) 21:80 Page 2 of 13 Gene structures, domain organizations, and motif distribution of R-genes in melon We analyzed the exon–intron structures of all 70 melon R-genes by comparing their coding sequences with the corresponding genomic sequences using the online tool GSDS2.0 (http://gsds.cbi.pku.edu.cn/). The highest num- ber of exons (22) was in the gene MELO3C013803, followed by 18 in MELO3C007367 (Additional file 1: Fig. S1). Among the 70 R-genes, 21 were mono-exonic, while 12 and 4 genes were bi- and tri-exonic, respectively. Genomic studies have provided insight into the evolu- tion of R-genes, which play important roles in the plant immune system in response to various pathogens and insects [25]. Plant R-genes encode proteins containing domains such as Nucleotide-binding site (NBS), Leucine-rich repeat (LRR), Toll/interleukin-1 receptor (TIR), Coiled-coil (CC), and Receptor-like kinase (RLK) domains [26–32]. These domains are involved in patho- gen recognition, signaling, and plant innate immunity responses [26, 27, 29, 31–35]. R-genes have been identi- fied in the genomes of plant species including water- melon [36], cucumber [25], rice [37, 38], Chinese We analyzed the conserved domains of the 70 melon R- genes using the Conserved Domain Database (CDD) at https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi. We analyzed the conserved domains of the 70 melon R- genes using the Conserved Domain Database (CDD) at https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi. We detected several disease resistance-related domains encoded by these R-genes, such as NB-ARC (Nucleotide- binding adaptor shared by APAF-1, R proteins, and CED-4), LRR (Leucine-rich repeat), TIR (Toll/interleukin-1 receptor), CC (Coiled-coil), and RLK (Receptor-like kinase) domains. The R-genes were grouped into different classes based on We detected several disease resistance-related domains encoded by these R-genes, such as NB-ARC (Nucleotide- binding adaptor shared by APAF-1, R proteins, and CED-4), LRR (Leucine-rich repeat), TIR (Toll/interleukin-1 receptor), CC (Coiled-coil), and RLK (Receptor-like kinase) domains. The R-genes were grouped into different classes based on Islam et al. BMC Genetics (2020) 21:80 Page 3 of 13 Table 1 Information about R-genes throughout the melon genome including chromosomal positions, lengths, and annotated descriptions Sl. Gene IDa Chr. Gene structures, domain organizations, and motif distribution of R-genes in melon BMC Genetics (2020) 21:80 Page 4 of 13 Table 1 Information about R-genes throughout the melon genome including chromosomal positions, lengths, and annotated descriptions (Continued) Sl. Gene IDa Chr. Gene structures, domain organizations, and motif distribution of R-genes in melon Number Position on chromosome CDS- length (bp) Peptide length (AA) Strand Description Start End 1 MELO3C023580.2 chr01 33,386,823 33,390,698 687 288 – Disease-resistance protein RGA2-like 2 MELO3C023579.2 chr01 33,395,126 33,397,789 2664 887 – Disease-resistance protein RGA2-like isoform X1 3 MELO3C023578.2 chr01 33,410,087 33,414,749 1158 385 – Disease-resistance protein 4 MELO3C023577.2 chr01 33,419,963 33,423,566 2715 904 – Disease-resistance protein RGA2-like 5 MELO3C023441.2 chr01 34,457,351 34,462,055 2766 921 – Receptor-kinase, putative 6 MELO3C023440.2 chr01 34,462,521 34,463,915 1338 445 – LRR receptor-like serine/threonine-protein kinase G 7 MELO3C023439.2 chr01 34,468,416 34,473,193 3207 1068 – LRR receptor-like serine/threonine-protein kinase G 8 MELO3C023438.2 chr01 34,474,924 34,475,353 336 111 + LRR receptor-like serine/threonine-protein kinase G 9 MELO3C023437.2 chr01 34,475,729 34,476,367 354 117 + Receptor-kinase, putative 10 MELO3C029319.2 chr02 4,111,584 4,115,605 717 238 + NBS-LRR type resistance protein 11 MELO3C015353.2 chr02 985,162 987,242 1737 578 + Disease-resistance protein RGA2-like 12 MELO3C015354.2 chr02 990,582 993,823 3240 1080 + Disease-resistance protein RGA2-like 13 MELO3C029505.2 chr02 7,359,371 7,363,388 765 254 – TMV resistance protein N-like 14 MELO3C010346.2 chr02 17,481,683 17,485,283 1593 530 + TMV resistance protein N 15 MELO3C010827.2 chr03 30,596,169 30,600,072 3663 1054 – Receptor-kinase, putative 16 MELO3C010826.2 chr03 30,600,299 30,603,794 3054 1071 – Receptor-kinase, putative 17 MELO3C010825.2 chr03 30,604,364 30,611,770 6069 2022 – Receptor-kinase, putative 18 MELO3C009695.2 chr04 30,097,463 30,100,144 2682 893 + Disease-resistance protein 19 MELO3C009694.2 chr04 30,103,601 30,106,071 2391 796 + Disease-resistance protein 20 MELO3C009693.2 chr04 30,110,724 30,113,156 2358 786 + Disease-resistance protein 21 MELO3C009179.2 chr04 33,763,652 33,766,776 3042 1013 + Receptor-kinase, putative 22 MELO3C009177.2 chr04 33,766,795 33,780,875 3231 1076 + Receptor-kinase, putative 23 MELO3C004259.2 chr05 25,752,437 25,757,292 3951 1316 + TMV resistance protein N-like isoform X1 24 MELO3C004288.2 chr05 26,044,574 26,052,361 3171 1056 + TMV resistance protein N-like 25 MELO3C004289.2 chr05 26,065,157 26,071,880 3867 1288 – TMV resistance protein N-like 26 MELO3C004301.2 chr05 26,231,021 26,237,770 4032 1343 – TMV resistance protein N-like isoform X1 27 MELO3C004303.2 chr05 26,239,395 26,244,020 2052 683 – TMV resistance protein N-like 28 MELO3C004309.2 chr05 26,263,738 26,270,641 4134 1377 + TMV resistance protein N-like 29 MELO3C004311.2 chr05 26,280,801 26,299,171 3156 1051 – TMV resistance protein N-like 30 MELO3C004313.2 chr05 26,311,869 26,315,091 2115 704 – TMV resistance protein N-like 31 MELO3C006780.2 chr06 5,898,974 5,902,420 3447 1148 – Disease-resistance protein 32 MELO3C006801.2 chr06 6,106,483 6,109,133 846 281 – Protein enhanced disease resistance 2-like 33 MELO3C016529.2 chr06 27,910,808 27,913,125 504 167 – TMV resistance protein N 34 MELO3C013803.2 chr06 33,588,343 33,599,894 2184 727 + Protein enhanced disease resistance 2 35 MELO3C017700.2 chr07 26,469,746 26,473,637 3141 1046 – Disease-resistance protein RGA2-like 36 MELO3C017701.2 chr07 26,475,401 26,480,759 3192 1063 + Disease-resistance protein RGA2-like 37 MELO3C017703.2 chr07 26,480,404 26,483,226 2823 940 – Disease-resistance protein RGA2-like 38 MELO3C007354.2 chr08 2,332,143 2,335,108 1806 601 – Cysteine-rich receptor-like protein kinase 29 39 MELO3C007358.2 chr08 2,346,707 2,353,661 4296 1431 – Receptor-like protein kinase 40 MELO3C007360.2 chr08 23,53,788 2,361,267 4088 1395 – Receptor-like protein kinase 41 MELO3C007367.2 chr08 2,372,510 2,386,472 4656 1551 – Receptor-like kinase 42 MELO3C022157.2 chr09 665,753 668,864 2025 674 – TMV resistance protein N-like isoform X1 Islam et al. Gene structures, domain organizations, and motif distribution of R-genes in melon Number Position on chromosome CDS- length (bp) Peptide length (AA) Strand Description Start End 43 MELO3C022154.2 chr09 681,564 689,908 3432 1143 – TMV resistance protein N-like 44 MELO3C022152.2 chr09 700,743 713,705 4173 1390 + TMV resistance protein N-like 45 MELO3C022146.2 chr09 762,107 767,613 2274 757 – TMV resistance protein N-like 46 MELO3C022145.2 chr09 768,255 784,265 3807 1268 + TMV resistance protein N-like 47 MELO3C022144.2 chr09 784,629 792,999 4902 1633 – TMV resistance protein N-like 48 MELO3C025516.2 chr09 6,632,514 6,659,697 4371 1,456 – TMV resistance protein N-like 49 MELO3C025519.2 chr09 6,674,960 6,677,738 762 253 – Disease-resistance protein RGA2-like 50 MELO3C025518.2 chr09 6,675,092 6,676,395 648 215 – Disease-resistance protein RGA2-like 51 MELO3C005450.2 chr09 21,691,401 21,694,271 2790 929 – LRR receptor-like kinase family protein 52 MELO3C005451.2 chr09 21,699,468 21,702,467 3000 999 – LRR receptor-like kinase 53 MELO3C005452.2 chr09 21,708,265 21,711,353 28,17 938 – LRR receptor-like kinase 54 MELO3C012268.2 chr10 1,574,521 1,579,615 1800 599 + Leaf rust 10 disease-resistance locus receptor-like protein kinase-like 1.2 isoform X4 55 MELO3C012049.2 chr10 2,989,020 2,990,934 1869 622 + Leaf rust 10 disease-resistance locus receptor-like protein kinase-like 1.5 56 MELO3C012045.2 chr10 3,007,893 3,014,091 1503 500 – Protein enhanced disease resistance 2 57 MELO3C034399.2 chr10 15,627,727 15,627,921 195 64 + Disease-resistance protein At4g27190-like 58 MELO3C022580.2 chr10 16,222,411 16,222,859 447 148 – Disease-resistance protein RGA2-like 59 MELO3C022447.2 chr11 33,758,671 33,762,610 3030 1009 – Receptor-like protein 60 MELO3C022449.2 chr11 33,770,307 33,772,966 2145 714 – Receptor-like protein 61 MELO3C002671.2 chr12 22,199,381 22,201,102 1350 449 + LRR receptor-like kinase 62 MELO3C002667.2 chr12 22,209,961 22,215,123 3279 1092 + LRR receptor-like kinase 63 MELO3C002666.2 chr12 22,219,699 22,226,478 3114 1037 + LRR receptor-like kinase 64 MELO3C002506.2 chr12 23,598,469 23,607,646 2040 679 – Receptor-like protein kinase 65 MELO3C002504.2 chr12 23,611,543 23,620,880 3870 1289 – Cysteine-rich receptor-like protein kinase 28 66 MELO3C002501.2 chr12 23,633,920 23,636,908 1617 538 + Cysteine-rich receptor-like protein kinase 26 isoform X1 67 MELO3C002394.2 chr12 24,343,418 4,346,595 2385 794 – LRR receptor-like kinase family protein 68 MELO3C002393.2 chr12 24,352,898 4,355,087 2190 729 – LRR receptor-like kinase 69 MELO3C002392.2 chr12 24,358,807 24,361,890 3084 1027 – LRR receptor-like serine/threonine- protein kinase GSO1 70 MELO3C002389.2 chr12 24,376,328 24,380,811 3786 1261 + LRR receptor-like serine/threonine-protein kinase GSO1 aGenomic information based on the reference Genome of Melon (DHL92) v3.6.1 retrieved from the Cucurbit Genomics Database (http://cucurbitgenomics.org) MELO3C022145, MELO3C022144, MELO3C004309, MELO3C004259, and MELO3C004301. A list of the genes and a description of their domains is provided in Table 2. Gene structures, domain organizations, and motif distribution of R-genes in melon the presence of the following conserved domains in their encoded proteins: (i) LRR, (ii) NBS-LRR, (iii) TIR, (iv) TIR-NBS-LRR, (v) NB-ARC, (vi) CC, (vii) RLK, and (viii) DUF (Table 2 and Additional file 1: Fig. S2). Thirty-seven genes encoded proteins with only LRR domains, seven encoded proteins with NB- ARC domains, two encoded proteins with TIR do- mains, and only one encodes a protein with a CC do- main (Table 2). Twelve genes encoded three domains (TIR, NBS, and LRR), including MELO3C004288, MELO3C004289, MELO3C004311, MELO3C004313, MELO3C022154, MELO3C022152, MELO3C022146, MELO3C022145, MELO3C022144, MELO3C004309, MELO3C004259, and MELO3C004301. A list of the genes and a description of their domains is provided in Table 2. the presence of the following conserved domains in their encoded proteins: (i) LRR, (ii) NBS-LRR, (iii) TIR, (iv) TIR-NBS-LRR, (v) NB-ARC, (vi) CC, (vii) RLK, and (viii) DUF (Table 2 and Additional file 1: Fig. S2). Thirty-seven genes encoded proteins with only LRR domains, seven encoded proteins with NB- ARC domains, two encoded proteins with TIR do- mains, and only one encodes a protein with a CC do- main (Table 2). Twelve genes encoded three domains (TIR, NBS, and LRR), including MELO3C004288, MELO3C004289, MELO3C004311, MELO3C004313, MELO3C022154, MELO3C022152, MELO3C022146, We analyzed the conserved motifs of these 70 R-genes using the MEME Suite (http://meme-uite.org/tools/ meme). A total of 20 conserved motifs were detected in these 70 R-genes, each comprising more than 14 amino acids. The greatest number of motifs was identified in the LRR domain-encoding gene MELO3C002394, whereas the fewest were detected in MELO3C029505, MELO3C023580, and MELO3C006801, which are LRR-, Islam et al. BMC Genetics (2020) 21:80 Page 5 of 13 Fig. 1 Chromosomal distribution of R-genes in melon genome-wide. The map was drawn using MapChart (v2.32) Fig. 1 Chromosomal distribution of R-genes in melon genome-wide. The map was drawn using MapChart (v2.32) CC-, and DUF-domain-encoding genes, respectively. The distribution of these conserved motifs, along with the motif sequences, is described in Fig. 2. susceptible accession at different time points. A general trend of low expression for these genes was observed in the susceptible accession (Fig. 4). On the contrary, most of the genes were significantly induced within 12 h of A. citrulli infection in the resistant accession and showed a general increase in expression in this accession. By con- trast, in the susceptible accession, the expression of these genes fluctuated, with little or no expression at the 12 h time point. Gene structures, domain organizations, and motif distribution of R-genes in melon Heatmap analysis of the expression data iden- tified a sub-cluster of six genes (MELO3C023441, MELO3C016529, MELO3C022157, MELO3C022146, MELO3C025518, and MELO3C004303) that showed con- trasting trends of expression in the resistant vs. susceptible accession, with progressively increasing expression after inoculation with A. citrulli in the resistant but not the sus- ceptible accession (Fig. 4). Extensive analysis of these six genes indicated that the expression of four genes (MELO3C023441, MELO3C004303, MELO3C022146, and MELO3C025518) increased in the resistant accession with increasing time after inoculation with A. citrulli (Fig. 5). In the susceptible accession, the expressions of these genes were very low in the initial hours after inocu- lation and did not show significant increase over time after Microsynteny of melon R-genes with genes in the watermelon and cucumber genomes We analyzed the microsyntenic relationships of the 70 R-genes from melon (Cucumis melo) with genes in the watermelon (Citrullus lanatus) and cucumber (Cucumis sativus) genomes using the Circos tool. Most R-genes from melon were homologous to R-genes from water- melon and cucumber. However, watermelon R-genes on chromosomes 11 and 12 lacked homologues in melon (Fig. 3). By contrast, all 70 R-genes in melon had homo- logues in all chromosomes of cucumber. MELO3C022157.2, MELO3C016529.2 MELO3C007354.2, MELO3C007358.2, MELO3C007360.2, MELO3C002506.2, MELO3C012268.2, MELO3C012049.2, MELO3C002504.2, MELO3C002501.2 8 Domain of unknown function (DUF) 8 Domain of unknown function (DUF) Protein enhanced disease resistance 2-like [53, 54] MELO3C006801.2, MELO3C012045.2 domain, an N-terminal TIR or CC domain, or an RLK/ RLP domain [29, 57]. A meta-analysis of the 314 cloned plant R-genes revealed that 191 (61%) such genes are NBS-LRR genes and 60 (19%) genes are RLKs/RLPs [58]. NBS domains bind to and hydrolyze adenosine triphos- phate (ATP) or guanosine triphosphate (GTP) and are involved in signaling; LRR domains are highly adaptable structural domains that are responsible for protein–pro- tein interactions and play an important role in plant– pathogen recognition [59]; TIR domains provide patho- gen specificity and plant defense responses, while CC domains are involved in pathogen recognition and sig- naling; and RLK domains play roles in signaling and plant defense responses. inoculation. In the resistant accession, the expression of these four genes (MELO3C023441, MELO3C004303, MELO3C022146, and MELO3C025518) peaked at 6 d after inoculation, with levels approximately 8-, 8-, 10-, and 7-fold those of the control samples, respectively. In the susceptible accession, the expression of two of these genes did not increase in response to A. citrulli inoculation, whereas the expression of two genes (MELO3C022157 and MELO3C016529) generally increased in response to inoculation, but to a lesser extent than in the resistant ac- cession. The expression of these two genes increased until 3 d after inoculation (5-fold in MELO3C016529 and 2.5- fold in MELO3C022157), followed by a decrease to their lowest levels at 6 d post inoculation (Fig. 5). In melon, four resistance gene homologue sequences were previously reported that contained 14 TIR-NBS- LRR genes [60, 61]. A study of the first complete gen- ome sequence of melon identified 411 putative R-genes, including 161 RLKs, 110 RLP (receptor-like proteins) genes, 19 RLK-GNK2 (kinases containing an additional antifungal protein ginkbilobin-2 domain) genes, and 81 genes containing canonical resistance domains, such as NBS, LRR, and TIR domains [1]. Among these genes, 25 were homologous to Pto genes from tomato and 15 were homologous to Mlo genes from barley [62, 63]. After Expression profiles of melon R-genes in response to A. citrulli inoculation We investigated the expression patterns of the 70 melon R-genes in the leaves of resistant and susceptible melon seedlings at 12 h, 1 d, 3 d, and 6 d of inoculation with A. citrulli strain NIHHS15–280 via qRT-PCR. Several genes showed differential expression in the resistant vs. Page 6 of 13 Islam et al. BMC Genetics (2020) 21:80 Table 2 R-genes throughout the melon genome categorized based on functional disease resistance-related domains Sl. Domain Function Gene ID 1 Leucine-rich repeat (LRR) Recognition of pathogen and Plant Defense [29, 46] MELO3C023577.2, MELO3C023579.2, MELO3C015353.2, MELO3C015354.2, MELO3C017700.2, MELO3C017701.2, MELO3C025518.2, MELO3C009695.2, MELO3C006780.2, MELO3C023441.2, MELO3C023437.2, MELO3C023440.2, MELO3C023439.2, MELO3C023438.2, MELO3C004303.2, MELO3C025516.2, MELO3C010346.2, MELO3C005450.2, MELO3C002394.2, MELO3C005451.2, MELO3C005452.2, MELO3C002671.2, MELO3C002667.2, MELO3C022447.2, MELO3C022449.2, MELO3C002392.2, MELO3C002389.2, MELO3C002393.2, MELO3C029505.2, MELO3C034399.2, MELO3C010827.2, MELO3C010826.2, MELO3C010825.2, MELO3C009179.2, MELO3C009177.2, MELO3C007367.2, MELO3C002666.2 2 Nucleotide-binding site leucine-rich repeat (NBS-LRR) Resistance protein Signaling and Plant Defense [27, 33, 47] MELO3C029319.2 3 Toll/interleukin-1 receptor homology (TIR) TMV resistance protein N [34, 46] MELO3C022157.2, MELO3C016529.2 4 Toll/interleukin-1 receptor homology nucleotide-binding site leucine-rich repeat (TIR-NBS-LRR) Pathogen specificity and defense [34, 46, 48] {Nandety, 2013 #111} MELO3C004288.2, MELO3C004289.2, MELO3C004311.2, MELO3C004313.2, MELO3C022154.2, MELO3C022152.2, MELO3C022146.2, MELO3C022145.2, MELO3C022144.2, MELO3C004309.2, MELO3C004259.2, MELO3C004301.2, 5 Nucleotide-binding adaptor shared by APAF-1, R proteins and CED-4 (NB-ARC) Molecular switch in activating defenses [28, 31] MELO3C017703.2, MELO3C025519.2, MELO3C022580.2, MELO3C023578.2, MELO3C009694.2, MELO3C009693.2, MELO3C013803.2 6 Coiled-coil domain (CC) Pathogen recognition and signaling [31, 32, 49] MELO3C023580.2 7 Protein kinase (RLK) Signaling and plant defense [35, 50–52] MELO3C007354.2, MELO3C007358.2, MELO3C007360.2, MELO3C002506.2, MELO3C012268.2, MELO3C012049.2, MELO3C002504.2, MELO3C002501.2 8 Domain of unknown function (DUF) Protein enhanced disease resistance 2-like [53, 54] MELO3C006801.2, MELO3C012045.2 Table 2 R-genes throughout the melon genome categorized based on functional disease resistance-related domains Sl. Domain Function Gene ID 1 Leucine-rich repeat (LRR) Recognition of pathogen and Plant Defense [29, 46] MELO3C023577.2, MELO3C023579.2, MELO3C015353.2, MELO3C015354.2, MELO3C017700.2, MELO3C017701.2, MELO3C025518.2, MELO3C009695.2, MELO3C006780.2, MELO3C023441 2 MELO3C023437 2 MELO3C023440 2 Discussion Here, we identified R-genes with putative roles in resist- ance to BFB disease in melon by profiling the genome- wide expression patterns of R-genes from melon in re- sponse to inoculation with A. citrulli. Disease resistance in plants involves the interaction between specific dis- ease resistance (R)-genes in plants and avirulence (avr) genes of the pathogen which is known as gene-for-gene model [55, 56]. Most plant R-genes belong to a super- family of genes encoding proteins with an NBS or LRR Islam et al. BMC Genetics (2020) 21:80 Page 7 of 13 Fig. 2 Conserved motifs in the R-genes of melon. Motifs are indicated by different colored rectangles. Motif sequences are provided in the legend ved motifs in the R-genes of melon. Motifs are indicated by different colored rectangles. Motif sequences are provided in th Fig. 2 Conserved motifs in the R-genes of melon. Motifs are indicated by different colored rectangles. Motif sequences are provided in the legend further improvements in the assembly and annotation of the melon (Cucumis melo L.) reference genome, 70 R- genes were ultimately identified in melon [44]. 1 Mb region on chromosome five contains the highest density of R-genes [69]. In addition, a cluster of 13 TNL genes is located in the same region as the melon Vat re- sistance gene [70], and another cluster of 7 TNL genes is located in the region harboring the Fom-1 resistance gene [71]. The Vat locus encodes a CC-NBS-LRR protein that confers resistance to aphid and aphid-mediated viruses in melon. The loss of two highly conserved LRRs is linked with susceptibility to these viruses [72]. In addition, the Fusarium wilt resistance locus Fom-2 is a TIR-NBS-LRR gene [73]. Expression patterns of the genome-wide R-genes are thus studied to identify any potential candidate genes against A. citrulli. Our comprehensive in-silico analysis of the 70 melon R-genes revealed that they encode proteins with several disease resistance-related domains, including LRR, NBS, TIR, NB-ARC, CC, RLK, and DUF domains (Table 2). These genes are distributed across all melon chromo- somes, and some are clustered in the telomeric regions of a few chromosomes (Fig. 1). The clustering of R-genes is an evolutionarily conserved defense mechanism in plants wherein recombination in closely located genes creates new motif combinations, which generates novel resistance specificities and broadens plant resistance to different dis- eases [42, 64]. Discussion R-gene clusters that provide resistance to multiple diseases have been reported for angular leaf spot, downy mildew, and anthracnose diseases in cucumber [65] and for blackleg, sclerotinia stem rot, and clubroot diseases in B. napus [66–68] and B. rapa [66]. In melon, a Six melon genes were highly expressed in the BFB- resistant accession. Of these genes, three (MELO3C016529, MELO3C022157, and MELO3C022146) are TNL genes, two (MELO3C023441 and MELO3C025518) are LRR genes, and one (MELO3C005452) is an NBS-LRR gene (Table 2). These genes were highly expressed at 6 d after Page 8 of 13 Islam et al. BMC Genetics (2020) 21:80 Fig. 3 Microsynteny analysis of all 70 melon R-genes with those of watermelon and cucumber. Melon, watermelon, and cucumber chromosomes are shown in orange, blue, and green, respectively. The diagram was drawn using the web-based tool Circos (http://circos.ca/software/download/) circos-0.69-9.tgz Fig. 3 Microsynteny analysis of all 70 melon R-genes with those of watermelon and cucumber. Melon, watermelon, and cucumber chromosomes are shown in orange, blue, and green, respectively. The diagram was drawn using the web-based tool Circos (http://circos.ca/software/download/) circos-0.69-9.tgz at different time points in Phytophthora capsici-inocu- lated tissues of resistant and susceptible melon geno- types provided a basis for identifying candidate resistant genes [78]. Comparative transcriptome ana- lysis identified ten genes that were differentially expressed in resistant and susceptible cultivars of melon in response to powdery mildew [79]. In addition, a study of the MLO (mildew resistance locus o) gene family in melon revealed candidate genes that might play roles in susceptibility to powdery mildew [80]. In watermelon, six NBS-encoding R-genes were identified as candidates for gummy stem blight (GSB) resistance [81, 82]. Finally, markers have been developed for detect- ing both GSB and BFB resistance in melon based on the sequence polymorphism in the TIR-NBS-LRR gene MELO3C022157 [81, 83]. Notably, all six candidate R- genes identified in the current study have corresponding homologues in watermelon and cucumber (Fig. 3). The inoculation (Fig. 5), which is consistent with our observa- tion that BFB symptoms first appeared at 6–7 d in a sus- ceptible accession [74]. Expression analysis upon infection with A. citrulli indi- cated a general trend of low expression for most R-genes in susceptible accession. By contrast, a set of genes includ- ing MELO3C023441, MELO3C004303, MELO3C022146, and MELO3C025518 were expressed at much higher levels, and MELO3C022157 and MELO3C016529 were expressed at relatively higher levels, (Fig. Discussion 5) in the resistant accession. Such higher expression in response to A. citrulli in the resistant accession indicates the potential involve- ment of these R-genes in BFB resistance in melon. Several comparative transcriptomic studies have been reported in melon [75–77], but few studies have fo- cused exclusively on expression profiling of R-genes against phytopathogenic agents in melon. For example, RNA-seq assessment of the changes in transcript levels Islam et al. BMC Genetics (2020) 21:80 Page 9 of 13 Fig. 4 Heat map of the expression patterns of melon R-genes determined by qRT-PCR in BFB-resistant and -susceptible melon accessions at the indicated time points after inoculation with A. citrulli. The expression levels were normalized to that Actin (the expression levels of the Actin gene are shown in Additional file 1: Fig. S3). The values were obtained from the means of three biological replicates. Red and green represent the minimum and maximum values, respectively. The IDs of six putative R-genes are shown in pink on the right side of the figure. MELO3C002671 and MELO3C022447 were not expressed and are therefore not shown in the heatmap. The heat map was generated using the online tool Heatmapper (http://www.heatmapper.ca/expression/) Fig. 4 Heat map of the expression patterns of melon R-genes determined by qRT-PCR in BFB-resistant and -susceptible melon accessions at the indicated time points after inoculation with A. citrulli. The expression levels were normalized to that Actin (the expression levels of the Actin gene are shown in Additional file 1: Fig. S3). The values were obtained from the means of three biological replicates. Red and green represent the minimum and maximum values, respectively. The IDs of six putative R-genes are shown in pink on the right side of the figure. MELO3C002671 and MELO3C022447 were not expressed and are therefore not shown in the heatmap. The heat map was generated using the online tool Heatmapper (http://www.heatmapper.ca/expression/) Fig. 4 Heat map of the expression patterns of melon R-genes determined by qRT-PCR in BFB-resistant and -susceptible melon accessions at the indicated time points after inoculation with A. citrulli. The expression levels were normalized to that Actin (the expression levels of the Actin gene are shown in Additional file 1: Fig. S3). The values were obtained from the means of three biological replicates. Red and green represent the minimum and maximum values, respectively. Discussion The IDs of six putative R-genes are shown in pink on the right side of the figure. MELO3C002671 and MELO3C022447 were not expressed and are therefore not shown in the heatmap. The heat map was generated using the online tool Heatmapper (http://www.heatmapper.ca/expression/) Fig. 5 Relative expression levels of six candidate R-genes in resistant and susceptible melon accessions at the indicated time points after inoculation with A. citrulli. Error bars represent standard errors of three individual observations. Different letters above the bars indicate significant differences, as determined by Tukey’s pairwise comparison. Ct-control, h- hour, and d- day Fig. 5 Relative expression levels of six candidate R-genes in resistant and susceptible melon accessions at the indicated time points after inoculation with A. citrulli. Error bars represent standard errors of three individual observations. Different letters above the bars indicate significant differences, as determined by Tukey’s pairwise comparison. Ct-control, h- hour, and d- day Page 10 of 13 Islam et al. BMC Genetics (2020) 21:80 roles of these genes in BFB resistance in these two crops remain to be investigated. Fig. 6 Disease symptoms in the leaves of susceptible (PI 614596) and resistant (PI 353814) melon accessions at 12 d after inoculation with A. citrulli. All leaves were detached from the plants immediately before they were photographed roles of these genes in BFB resistance in these two crops remain to be investigated. Conclusions We identified six putative candidate genes that might play roles in resistance to BFB in melon. This is the first report of candidate genes for BFB resistance in melon. Our findings provide a basis for further functional stud- ies to validate the exact roles of these genes. In addition, causal sequence polymorphisms could be identified in these genes, leading to the development of markers for BFB resistance. Our findings will thus be useful for im- proving the BFB resistance trait in melon. Identification and in silico analysis of melon R-genes Identification and in silico analysis of melon R genes Genomic information for all 70 R-genes, as reported in the improved assembly and annotated genome of melon [44], was retrieved from the cucurbit genomic database (http://cucurbitgenomics.org) (Additional file 1: Table S1). The genes were subjected to a series of in silico ana- lyses such as exon–intron structure, motif distribution, domain organization, chromosomal mapping, and microsynteny analyses (for specific analytical tools, see the Results section). Two weeks after germination, the plants were trans- ferred to plastic pots and grown in a greenhouse at 24 ± 2 °C with a relative humidity of 90% where the plants were inoculated with A. citrulli. The resistance status of the accessions was reconfirmed via bioassay (Fig. 6) as previously reported with minor modifications [86]. Plants at the 3–5 true-leaf stage (4–5 weeks old) were sprayed with bacterial suspensions until runoff in a greenhouse at 22 ± 2 °C with a relative humidity of 96%. Plants were re-inoculated 3 d after the first inoculation to ensure that no plants had avoided inoculation and to eliminate false positives. Leaf samples from three Plant materials, growth conditions, and bioassays Plant materials, growth conditions, and bioassays The BFB-resistant (PI 353814) and -susceptible (PI 614596) melon accessions [74, 85] used in this study were obtained from the U.S. National Plant Germplasm System (https://npgsweb.ars-grin.gov/gringlobal/search. aspx), USDA, USA. The seeds were sown in a commer- cial nursery soil mixture in 32-cell trays and grown in a controlled plant growth chamber at 25 ± 2 °C, 16 h day length, relative humidity of 60%, and a light intensity of 440 μmoles/m2/s at bench level. After 3 weeks, the plants were transferred to a greenhouse. Total RNA isolation and cDNA synthesis Total RNA isolation and cDNA synthesis The melon leaves were ground to a powder in liquid nitrogen, and 100 mg of each sample with three bio- logical replicates was subjected to total RNA extrac- tion using the RNeasy Mini kit (Qiagen, Valencia, CA) following the manufacturer’s instructions. First- strand cDNA was synthesized from total RNA with a SuperScript III First-Strand Synthesis System kit (Invi- trogen, Gaithersburg, MD). Methods A. citrulli: collection, culture, and inoculum preparation A. citrulli strain NIHHS15–280 was obtained from the National Institute of Horticultural and Herbal Science (NIHHS), South Korea. The bacterium was cultured on Petri plates containing 20 ml King’s B (KB) medium sup- plemented with 100 μg ml−1 ampicillin for 36–48 h at 28 °C [84] until bacterial colonies formed. For all inocu- lations, a bacterial suspension was prepared by covering the culture plates with 5 ml of sterile, double distilled (DD) water and gently scraping the surface of the KB medium using an L-shaped rubber spreader to an optical density (OD) of 1.0 at 600 nm, as measured using a NanoDrop ND-1000 Spectrophotometer. The bacterial suspension was diluted to a final concentration of ~ 1 × 106 colony forming units (cfu) mL−1. Fig. 6 Disease symptoms in the leaves of susceptible (PI 614596) and resistant (PI 353814) melon accessions at 12 d after inoculation with A. citrulli. All leaves were detached from the plants immediately before they were photographed biological replicates were collected at different time points (0 h, 12 h, 1 d, 3 d, and 6 d), immediately immersed in liquid nitrogen, and stored at −80 °C for RNA extraction and cDNA synthesis. Author details 1 1Department of Horticulture, Sunchon National University, Suncheon, Jeonnam 57922, Republic of Korea. 2Department of Biotechnology, Sher-e-Bangla Agricultural University, Dhaka 1207, Bangladesh. 3Department of Genetics and Plant Breeding, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh. Ethics approval and consent to participate The authors declare that this study conforms with the current laws of the countries in which the experiments were performed. Funding This study was supported by the Golden Seed Project (Grant No. 213007–05- 4-CG100) of the Ministry of Agriculture, Food and Rural Affairs (MAFRA), Republic of Korea. Acknowledgments h k h S We thank the U.S. National Plant Germplasm System, USA, Korean Agricultural Culture Collection, Korea, and National Institute of Horticultural and Herbal Science, Korea for providing the melon seeds and bacterial strains. 8. Kong Q, Yuan J, Niu P, Xie J, Jiang W, Huang Y, Bie Z. Screening suitable reference genes for normalization in reverse transcription quantitative real- time PCR analysis in melon. PLoS One. 2014;9(1):e87197. 9. Schaad NW, Postnikova E, Sechler A, Claflin LE, Vidaver AK, Jones JB, Agarkova I, Ignatov A, Dickstein E, Ramundo BA. Reclassification of subspecies of Acidovorax avenae as A. avenae (Manns 1905) Emend., A. cattleyae (Pavarino, 1911) comb. nov., A. citrulli Schaad et al., 1978 comb. nov., and proposal of A. oryzae sp. nov. Syst Appl Microbiol. 2008;31(6–8): 434–46. Statistical analysis All of the authors of this manuscript give their consent to publish the findings in BMC Genetics. Analysis of variance (ANOVA) and significance tests were carried out using the normalized gene expression values with MINITAB17 software (Minitab Inc., State College, PA, USA). Tukey’s pairwise comparison test was employed to determine the mean separation of ex- pression values. p values indicate statistically significant variations of expression. Competing interests h h d l h The authors declare that there are no conflicts of interest to publish in this journal. Abbreviations 5. Lester GE, Crosby KM. Ascorbic acid, folic acid, and potassium content in postharvest green-flesh honeydew muskmelons: influence of cultivar, fruit size, soil type, and year. J Am Soc Hortic Sci. 2002;127(5):843–7. A. citrulli: Acidovorax citrulli; BFB: Bacterial fruit blotch; R-genes: Resistance genes; avr: Avirulence; Fig: Figure; Chr: Chromosome; CDS: Coding Sequence; bp: Base pair; AA: Amino Acid; LRR: Leucine-rich repeat; NBS-LRR: Nucleotide- binding site leucine-rich repeat; TIR: Toll/interleukin-1 receptor homology; TIR: NBS-LRR- Toll/interleukin-1 receptor homology nucleotide-binding site leucine-rich repeat; NB-ARC: Nucleotide-binding adaptor shared by APAF-1, R proteins and CED-4; CC: Coiled-coil domain; RLK: Protein kinase; DUF: Domain of unknown function 6. Lester GE. Antioxidant, sugar, mineral, and phytonutrient concentrations across edible fruit tissues of orange-fleshed honeydew melon (Cucumis melo L.). J Agric Food Chem. 2008;56(10):3694–8. 7. Mascarell-Creus A, Cañizares J, Vilarrasa-Blasi J, Mora-García S, Blanca J, Gonzalez-Ibeas D, Saladié M, Roig C, Deleu W, Picó-Silvent B. An oligo-based microarray offers novel transcriptomic approaches for the analysis of pathogen resistance and fruit quality traits in melon (Cucumis melo L.). BMC Genomics. 2009;10(1):467. 7. Mascarell-Creus A, Cañizares J, Vilarrasa-Blasi J, Mora-García S, Blanca J, Gonzalez-Ibeas D, Saladié M, Roig C, Deleu W, Picó-Silvent B. An oligo-based microarray offers novel transcriptomic approaches for the analysis of pathogen resistance and fruit quality traits in melon (Cucumis melo L.). BMC Genomics. 2009;10(1):467. References 1. Garcia-Mas J, Benjak A, Sanseverino W, Bourgeois M, Mir G, González VM, Hénaff E, Câmara F, Cozzuto L, Lowy E. The genome of melon (Cucumis melo L.). PNAS. 2012;109(29):11872–7. 1. Garcia-Mas J, Benjak A, Sanseverino W, Bourgeois M, Mir G, González VM, Hénaff E, Câmara F, Cozzuto L, Lowy E. The genome of melon (Cucumis melo L.). PNAS. 2012;109(29):11872–7. 2. Kong Q, Gao L, Cao L, Liu Y, Saba H, Huang Y, Bie Z. Assessment of suitable reference genes for quantitative gene expression studies in melon fruits. Front Plant Sci. 2016;7:1178. 2. Kong Q, Gao L, Cao L, Liu Y, Saba H, Huang Y, Bie Z. Assessment of suitable reference genes for quantitative gene expression studies in melon fruits. Front Plant Sci. 2016;7:1178. 3. Fita A, Bowen HC, Hayden RM, Nuez F, Pico B, Hammond JP. Diversity in expression of phosphorus (P) responsive genes in Cucumis melo L. PLoS One. 2012;7(4):e35387. 3. Fita A, Bowen HC, Hayden RM, Nuez F, Pico B, Hammond JP. Diversity in expression of phosphorus (P) responsive genes in Cucumis melo L. PLoS One. 2012;7(4):e35387. 4. Lester G, Eischen F. Beta-carotene content of postharvest orange-fleshed muskmelon fruit: effect of cultivar, growing location and fruit size. Plant Food Hum Nutr. 1996;49(3):191–7. Received: 25 February 2020 Accepted: 12 July 2020 Additional file 1: Table S1. Details of the primers designed for expression profiling of melon R-genes. Figure S1. Exon–intron structures of R-genes in melon genome-wide. Light red rectangles and black lines indicate exons and introns, respectively. Figure S2. Domain structures of the 70 R-genes in melon. The conserved domains were identified using the NCBI Conserved Domain Database (CDD) (https://www.ncbi.nlm.nih. gov/Structure/bwrpsb/bwrpsb.cgi). Detailed descriptions of these do- mains are provided in Table 2. Specific domains in each protein are shown in the diagram. Figure S3. Gene expression profiles of resistant and susceptible melon accessions at different time points normalized to melon Actin expression (CmACT7, 149 bp), as determined by qRT-PCR analysis. Primer design and quantitative RT-PCR analysis Gene-specific primers for quantitative RT-PCR (qRT- PCR) were designed using Primer3Plus (https://primer3 plus.com/cgibin/dev/primer3plus.cgi) (Table 1). The Page 11 of 13 Islam et al. BMC Genetics (2020) 21:80 Page 11 of 13 Page 11 of 13 Islam et al. BMC Genetics (2020) 21:80 expression patterns of the R-genes were analyzed by qRT-PCR in a LightCycler® instrument (Roche, Mann- heim, Germany) following the manufacturer’s instruc- tions. The reactions were performed in a 10 μL volume consisting of 5 μL of 2x qPCRBIO SyGreen Mix Lo-ROX (PCR Biosystems, London, UK), 5 pmol of primers, and cDNA template diluted to the appropriate concentra- tions. The PCR conditions were as follows: 5 min at 95 °C, followed by 3-step amplifications at 95 °C for 15 s, 56 °C for 15 s and 72 °C for 20 s for 45 cycles. The mean expression levels of relevant genes were calculated by the 2–ΔΔ Ct method [87] using the average value of three reference genes [2, 8, 88] as internal control. revised and finalized the manuscript. I.-S.N., H.-T.K. and J.-I.P. conceived and supervised the project. All authors read and approved the final draft of the manuscript. revised and finalized the manuscript. I.-S.N., H.-T.K. and J.-I.P. conceived and supervised the project. All authors read and approved the final draft of the manuscript. Availability of data and materials We declare that the dataset(s) supporting the conclusions of this article are encompassed within the article (and its additional file(s). Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12863-020-00885-9. Received: 25 February 2020 Accepted: 12 July 2020 Received: 25 February 2020 Accepted: 12 July 2020 Authors’ contributions Toll and interleukin-1 receptor (TIR) domain-containing proteins in plants: a genomic perspective. Trends Plant Sci. 2002;7(9):388–91. 22. Hopkins DL, Thompson CM, Hilgren J, Lovic B. Wet seed treatment with peroxyacetic acid for the control of bacterial fruit blotch and other seedborne diseases of watermelon. Plant Dis. 2003;87(12):1495–9. 23. Stephens D, Schneider R, Walcott R, Johnson C. A procedure, based on exposure to chlorine gas, for disinfesting watermelon seeds. In: Phytopathology: 2008: Amer Phytopathological Soc 3340 Pilot Knob Road, St Paul, MN 55121 USA; 2008. p. S150–1. 49. Rairdan GJ, Collier SM, Sacco MA, Baldwin TT, Boettrich T, Moffett P. The coiled-coil and nucleotide binding domains of the potato Rx disease resistance protein function in pathogen recognition and signaling. Plant Cell. 2008;20(3):739–51. 24. Feng J, Li J, Randhawa P, Bonde M, Schaad NW. Evaluation of seed treatments for the eradication of Acidovorax avenae subsp. citrulli from melon and watermelon seeds. Can J Plant Pathol. 2009;31(2):180–5. 50. Goff KE, Ramonell KM. The role and regulation of receptor-like kinases in plant defense. Gene Regul Syst Biol. 2007;1:117762500700100015. 51. Gómez-Gómez L, Boller T. FLS2: an LRR receptor–like kinase involved in the perception of the bacterial elicitor flagellin in Arabidopsis. Mol Cell. 2000; 5(6):1003–11. 25. Wan H, Yuan W, Bo K, Shen J, Pang X, Chen J. Genome-wide analysis of NBS-encoding disease resistance genes in Cucumis sativus and phylogenetic study of NBS-encoding genes in Cucurbitaceae crops. BMC Genomics. 2013; 14(1):109. 52. Shiu S-H, Bleecker AB. Plant receptor-like kinase gene family: diversity, function, and signaling. Sci STKE. 2001;2001(113):re22. 26. Jones JD. Plant disease resistance genes: structure, function and evolution. Curr Opin Biotechnol. 1996;7(2):155–60. 53. Gu Y, Innes RW. The KEEP ON GOING protein of Arabidopsis recruits the ENHANCED DISEASE RESISTANCE1 protein to trans-Golgi network/early endosome vesicles. Plant Physiol. 2011;155(4):1827–38. 27. DeYoung BJ, Innes RW. Plant NBS-LRR proteins in pathogen sensing and host defense. Nat Immunol. 2006;7(12):1243. 54. Yang Q, Niu X, Tian X, Zhang X, Cong J, Wang R, Zhang G, Li G. Comprehensive genomic analysis of the DUF4228 gene family in land plants and expression profiling of ATDUF4228 under abiotic stresses. BMC Genomics. 2020;21(1):1–16. 28. Dangl JL, Jones JD. Plant pathogens and integrated defence responses to infection. Nature. 2001;411(6839):826. 29. Ellis J, Dodds P, Pryor T. Structure, function and evolution of plant disease resistance genes. Curr Opin Plant Biol. 2000;3(4):278–84. 55. Authors’ contributions M.R.I. designed and conducted the entire experiment, DNA extraction, performed wet lab experiments, analyzed the qRT-PCR data, interpreted the results, and wrote the first draft of the manuscript. D.M.I.J. helped with the in silico analysis, performed the bioinformatics analysis, and constructed the Figs. H.-J.J. assisted with the bioassay and PCR assays. M.R.H. comprehensively 10. Silva GM, Souza RM, Yan L, Júnior RS, Medeiros FH, Walcott RR. Strains of the group I lineage of Acidovorax citrulli, the causal agent of bacterial fruit Page 12 of 13 Page 12 of 13 Page 12 of 13 Islam et al. BMC Genetics (2020) 21:80 36. Guo S, Zhang J, Sun H, Salse J, Lucas WJ, Zhang H, Zheng Y, Mao L, Ren Y, Wang Z. The draft genome of watermelon (Citrullus lanatus) and resequencing of 20 diverse accessions. Nat Genet. 2013;45(1):51. blotch of cucurbitaceous crops, are predominant in Brazil. Phytopathology. 2016;106(12):1486–94. blotch of cucurbitaceous crops, are predominant in Brazil. Phytopathology. 2016;106(12):1486–94. 11. Burdman S, Kots N, Kritzman G, Kopelowitz J. Molecular, physiological, and host-range characterization of Acidovorax avenae subsp. citrulli isolates from watermelon and melon in Israel. Plant Dis. 2005;89(12):1339–47. 37. Zhou T, Wang Y, Chen J-Q, Araki H, Jing Z, Jiang K, Shen J, Tian D. Genome- wide identification of NBS genes in japonica rice reveals significant expansion of divergent non-TIR NBS-LRR genes. Mol Gen Genomics. 2004;271(4):402–15. 12. Burdman S, Walcott R. Acidovorax citrulli: generating basic and applied knowledge to tackle a global threat to the cucurbit industry. Mol Plant Pathol. 2012;13(8):805–15. 38. Monosi B, Wisser R, Pennill L, Hulbert S. Full-genome analysis of resistance gene homologues in rice. Theor Appl Genet. 2004;109(7):1434–47. 13. Bahar O, Burdman S. Bacterial fruit blotch: a threat to the cucurbit industry. Isr J Plant Sci. 2010;58(1):19–31. 39. Lv S, Changwei Z, Tang J, Li Y, Wang Z, Jiang D, Hou X. Genome-wide analysis and identification of TIR-NBS-LRR genes in Chinese cabbage (Brassica rapa ssp. pekinensis) reveal expression patterns to TuMV infection. Physiol Mol Plant Pathol. 2015;90:89–97. 14. Noh J, Kim J-H, Lim JH, Kim TB, Seong MH, Jung GT, Kim JM, Cheong S-S, Oh NK, Lee W-H. Occurrence of diseases and case of clinical diagnosis on watermelon in South Korea, 2008-2012. Res Plant Dis. 2014;20(1):8–14. 40. Song W, Wang B, Li X, Wei J, Chen L, Zhang D, Zhang W, Li R. Identification of immune related LRR-containing genes in maize (Zea mays L.) by genome-wide sequence analysis. Authors’ contributions Int J Genomics. 2015;2015:231358. 15. O'brien R, Martin H. Bacterial blotch of melons caused by strains of Acidovorax avenae subsp. citrulli. Aust J Exp Agr. 1999;39(4):479–85. 41. Bouktila D, Khalfallah Y, Habachi-Houimli Y, Mezghani-Khemakhem M, Makni M, Makni H. Large-scale analysis of NBS domain-encoding resistance gene analogs in Triticeae. Genet Mol Biol. 2014;37(3):598–610. 16. Islam M, Hossain MR, Kim H-T, Jesse DMI, Abuyusuf M, Jung H-J, Park J-I, Nou I-S. Development of molecular markers for detection of Acidovorax citrulli strains causing bacterial fruit blotch disease in melon. Int J Mol Sci. 2019;20(11):2715. 42. Meyers BC, Kozik A, Griego A, Kuang H, Michelmore RW. Genome-wide analysis of NBS-LRR–encoding genes in Arabidopsis. Plant Cell. 2003;15(4):809–34. 17. Carvalho FC, Santos LA, Dias RC, Mariano RL, Souza EB. Selection of watermelon genotypes for resistance to bacterial fruit blotch. Euphytica. 2013;190(2):169–80. 43. Lee S-Y, Seo J-S, Rodriguez-Lanetty M, Lee D-H. Comparative analysis of superfamilies of NBS-encoding disease resistance gene analogs in cultivated and wild apple species. Mol Gen Genomics. 2003;269(1):101–8. 18. Conceição CS, Felix KCS, Mariano RL, Medeiros EV, Souza EB. Combined effect of yeast and silicon on the control of bacterial fruit blotch in melon. Sci Hortic. 2014;174:164–70. 44. Ruggieri V, Alexiou KG, Morata J, Argyris J, Pujol M, Yano R, Nonaka S, Ezura H, Latrasse D, Boualem A. An improved assembly and annotation of the melon (Cucumis melo L.) reference genome. Sci Rep. 2018;8(1):8088. 19. de Melo EA, Rosa de Lima RM, Laranjeira D, dos Santos LA, de Omena Gusmão L, de Souza EB. Efficacy of yeast in the biocontrol of bacterial fruit blotch in melon plants. Trop Plant Pathol. 2015;40(1):56–64. 45. Natarajan S, Kim H-T, Thamilarasan SK, Veerappan K, Park J-I, Nou I-S. Whole genome re-sequencing and characterization of powdery mildew disease- associated allelic variation in melon. PLoS One. 2016;11(6):e0157524. 20. Schaad NW, Sowell G Jr, Goth R, Colwell R, Webb R. Pseudomonas pseudoalcaligenes subsp. citrulli subsp. nov. Int J Syst Evol Microbiol. 1978; 28(1):117–25. 46. Ng A, Xavier RJ. Leucine-rich repeat (LRR) proteins: integrators of pattern recognition and signaling in immunity. Autophagy. 2011;7(9):1082–4. 47. Bai J, Pennill LA, Ning J, Lee SW, Ramalingam J, Webb CA, Zhao B, Sun Q, Nelson JC, Leach JE. Diversity in nucleotide binding site–leucine-rich repeat genes in cereals. Genome Res. 2002;12(12):1871–84. 21. Hopkins D. The hypothetical exam question becomes reality. Plant Dis. 1995;79:761–5. 48. Jebanathirajah JA, Peri S, Pandey A. Authors’ contributions melo and Arabidopsis thaliana. Plant Mol. Biol. 2003;51(5):703–18. 61. Garcia-Mas J, van Leeuwen H, Monfort A, de Vicente MC, Puigdomènech P, Arús P. Cloning and mapping of resistance gene homologues in melon. Plant Sci. 2001;161(1):165–72. 82. Whitham S, Dinesh-Kumar S, Choi D, Hehl R, Corr C, Baker B. The product of the tobacco mosaic virus resistance gene N: similarity to toll and the interleukin-1 receptor. Cell. 1994;78(6):1101–15. 62. Büschges R, Hollricher K, Panstruga R, Simons G, Wolter M, Frijters A, van Daelen R, van der Lee T, Diergaarde P, Groenendijk J. The barley Mlo gene: a novel control element of plant pathogen resistance. Cell. 1997;88(5):695–705. 83. Islam M, Hossain MR, Jesse DMI, Jung H-J, Kim H-T, Park J-I, Nou I-S. Development of Molecular Marker Linked with Bacterial Fruit Blotch Resistance in Melon (Cucumis melo L.). Genes. 2020;11(2):220. 63. Loh Y-T, Martin GB. The disease-resistance gene Pto and the fenthion- sensitivity gene fen encode closely related functional protein kinases. PNAS. 1995;92(10):4181–4. 84. Schaad N, Jones J, Chun W. Laboratory guide for the identification of plant pathogenic bacteria. Paul, MN: Amer Phytopathol. Soc; 2001. 85. Wechter WP, Levi A, Ling K-S, Kousik C, Block CC. Identification of resistance to Acidovorax avenae subsp. citrulli among melon (Cucumis spp.) plant introductions. Hortscience. 2011;46(2):207–12. 64. Hulbert SH, Webb CA, Smith SM, Sun Q. Resistance gene complexes: evolution and utilization. Annu Rev Phytopathol. 2001;39(1):285–312. 65. Wang Y, Tan J, Wu Z, VandenLangenberg K, Wehner TC, Wen C, Zheng X, Owens K, Thornton A, Bang HH. STAYGREEN, STAY HEALTHY: a loss-of- susceptibility mutation in the STAYGREEN gene provides durable, broad- spectrum disease resistances for over 50 years of US cucumber production. New Phytol. 2019;221(1):415–30. 86. Bahar O, Kritzman G, Burdman S. Bacterial fruit blotch of melon: screens for disease tolerance and role of seed transmission in pathogenicity. Eur J Plant Pathol. 2009;123(1):71–83. 87. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods. 2001;25(4): 402–8. 66. Kato T, Hatakeyama K, Fukino N, Matsumoto S. Fine mapping of the clubroot resistance gene CRb and development of a useful selectable marker in Brassica rapa. Breed.Sci. 2013;63(1):116–24. 88. Rachmawati Y, Daryono BS, Aristya GR. Molecular characters of melon (Cucumis melo L.“Tacapa”) in response to karst critical land. In: AIP Conference Proceedings: 2017: AIP Publishing LLC; 2017. p. 020030. 67. Authors’ contributions Li L, Luo Y, Chen B, Xu K, Zhang F, Li H, Huang Q, Xiao X, Zhang T, Hu J. A genome-wide association study reveals new loci for resistance to clubroot disease in Brassica napus. Front Plant Sci. 2016;7:1483. Publisher’s Note 68. Delourme R, Laperche A, Bouchet A-S, Jubault M, Paillard S, Manzanares- Dauleux M-J, Nesi N. Genes and quantitative trait loci mapping for major agronomic traits in Brassica napus L. In: The Brassica napus genome. In: Springer; 2018. p. 41–85. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 69. González VM, Aventín N, Centeno E, Puigdomènech P. Interspecific and intraspecific gene variability in a 1-Mb region containing the highest density of NBS-LRR genes found in the melon genome. BMC Genomics. 2014;15(1):1131. 70. Lecoq H, Pitrat M:Effect on cucumber mosaic virus incidence of the cultivation of partially resistant muskmelon cultivars. In: IV Conference on recent Advances in Vegetable Virus Research 127: 1982; 1982: 137–146. 71. Oumouloud A, Arnedo-Andres MS, Gonzalez-Torres R, Alvarez JM. Development of molecular markers linked to the Fom-1 locus for resistance to Fusarium race 2 in melon. Euphytica. 2008;164(2):347–56. 72. Dogimont C, Chovelon V, Pauquet J, Boualem A, Bendahmane A. The V at locus encodes for a CC-NBS-LRR protein that confers resistance to Aphis gossypii infestation and A. gossypii-mediated virus resistance. Plant J. 2014; 80(6):993–1004. 73. Wang Y-H, Choi W, Thomas CE, Dean RA. Cloning of disease-resistance homologues in end sequences of BAC clones linked to Fom-2, a gene conferring resistance to Fusarium wilt in melon (Cucumis melo L.). Genome. 2002;45(3):473–80. 74. Islam MR, Hossain MR, Kim H-T, Nath UK, Abuyusuf M, Jung H-J, Park J-I, Nou I-S. Molecular characterization of Acidovorax citrulli strain NIHHS15-280 causing bacterial fruit blotch disease in Korea and screening of resistance sources in melon. Hortic Environ Biotechnol. 2019;61(1):1–12. 75. Hong C, Wei-Ping K, Jun-Feng L. Comparative transcriptome analysis of powdery mildew resistance between two melon (Cucumis melo L) with different thickness Peel. BioRxiv. 2019;541391. 76. Polonio Á, Pineda M, Bautista R, Martínez-Cruz J, Pérez-Bueno ML, Barón M, Pérez-García A. RNA-seq analysis and fluorescence imaging of melon powdery mildew disease reveal an orchestrated reprogramming of host physiology. Sci Rep. 2019;9(1):7978. 77. Silvia Sebastiani M, Bagnaresi P, Sestili S, Biselli C, Zechini A, Orrù L, Cattivelli L, Ficcadenti N. Transcriptome analysis of the melon-Fusarium oxysporum f. sp. melonis race 1.2 pathosystem in susceptible and resistant plants. Front Plant Sci. 2017;8:362. 78. Authors’ contributions Slootweg E, Koropacka K, Roosien J, Dees R, Overmars H, Lankhorst RK, van Schaik C, Pomp R, Bouwman L, Helder J. Sequence exchange between homologous NB-LRR genes converts virus resistance into nematode resistance, and vice versa. Plant Physiol. 2017;175(1):498–510. 30. Harris CJ, Slootweg EJ, Goverse A, Baulcombe DC. Stepwise artificial evolution of a plant disease resistance gene. PNAS. 2013;110(52):21189–94. 31. Van Ooijen G, Mayr G, Kasiem MM, Albrecht M, Cornelissen BJ, Takken FL. Structure–function analysis of the NB-ARC domain of plant disease resistance proteins. J Exp Bot. 2008;59(6):1383–97. 56. Bonas U, Van den Ackerveken G. Gene-for-gene interactions: bacterial avirulence proteins specify plant disease resistance. Curr Opin Microbiol. 1999;2(1):94–8. 32. Qi D, DeYoung BJ, Innes RW. Structure-function analysis of the coiled-coil and leucine-rich repeat domains of the RPS5 disease resistance protein. Plant Physiol. 2012;158(4):1819–32. 57. van Leeuwen H, Garcia-Mas J, Coca M, Puigdoménech P, Monfort A. Analysis of the melon genome in regions encompassing TIR-NBS-LRR resistance genes. Mol Gen Genomics. 2005;273(3):240–51. 33. McHale L, Tan X, Koehl P, Michelmore RW. Plant NBS-LRR proteins: adaptable guards. Genome Biol. 2006;7(4):212. 58. Kourelis J, van der Hoorn RA. Defended to the nines: 25 years of resistance gene cloning identifies nine mechanisms for R protein function. Plant Cell. 2018;30(2):285–99. 34. Nandety RS, Caplan JL, Cavanaugh K, Perroud B, Wroblewski T, Michelmore RW, Meyers BC. The role of TIR-NBS and TIR-X proteins in plant basal defense responses. Plant Physiol. 2013;162(3):1459–72. 59. Marone D, Russo MA, Laidò G, De Leonardis AM, Mastrangelo AM. Plant nucleotide binding site–leucine-rich repeat (NBS-LRR) genes: active guardians in host defense responses. Int J Mol Sci. 2013;14(4):7302–26. 35. Wu Y, Zhou JM. Receptor-L ike kinases in plant innate immunity. J Integr Plant Biol. 2013;55(12):1271–86. Page 13 of 13 Page 13 of 13 Page 13 of 13 Islam et al. BMC Genetics (2020) 21:80 60. Van Leeuwen H, Monfort A, Zhang H-B, Puigdomènech P. Identification and characterisation of a melon genomic region containing a resistance gene cluster from a constructed BAC library. Microcolinearity between Cucumis melo and Arabidopsis thaliana. Plant Mol. Biol. 2003;51(5):703–18. 80. Zhou S, Jing Z, Shi J. Genome-wide identification, characterization, and expression analysis of the MLO gene family in Cucumis sativus. Genet Mol Res. 2013;12(4):6565–78. 81. Hassan M, Rahim M, Natarajan S, Robin A, Kim H-T, Park J-I, Nou I-S. Gummy stem blight resistance in melon: inheritance pattern and development of molecular markers. Int J Mol Sci. 2018;19(10):2914. Publisher’s Note Wang P, Wu H, Zhao G, He Y, Kong W, Zhang J, Liu S, Liu M, Hu K, Liu L. Transcriptome analysis clarified genes involved in resistance to Phytophthora capsici in melon. PLoS One. 2020;15(2):e0227284. 79. Zhu Q, Gao P, Wan Y, Cui H, Fan C, Liu S, Luan F. Comparative transcriptome profiling of genes and pathways related to resistance against powdery mildew in two contrasting melon genotypes. Sci Hortic. 2018;227: 169–80. 79. Zhu Q, Gao P, Wan Y, Cui H, Fan C, Liu S, Luan F. Comparative transcriptome profiling of genes and pathways related to resistance against powdery mildew in two contrasting melon genotypes. Sci Hortic. 2018;227: 169–80.
https://openalex.org/W4368404637	https://www.mdpi.com/2073-4433/14/5/822/pdf?version=1683270670	English	null	Spatial and Temporal Variation of Aerosol Optical Depth in Huaihai Economic Zone from 1982 to 2021	Atmosphere	2,023	cc-by	13,592	Article Spatial and Temporal Variation of Aerosol Optical Depth in Huaihai Economic Zone from 1982 to 2021 g Xue 1,3,, Yuxin Sun 1 , Chunlin Jin 1, Minghao Zhang 1 , Xingxing Jiang 1 and Xi Lu 1 Shuhui Wu 1,2, Yong Xue 1,3,, Yuxin Sun 1 , Chunlin Jin 1, Minghao Zhang 1 , Xingxing Jian 1 School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China 2 Artiﬁcial Intelligence Research Institute, China University of Mining and Technology, Xuzhou 221116, China 3 School of Computing and Mathematics, College of Science and Engineering, University of Derby, Kedleston Road, Derby DE22 1GB, UK * Correspondence: yxue@derbyac uk 1 School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China 2 Artiﬁcial Intelligence Research Institute, China University of Mining and Technology, Xuzhou 221116, China 3 School of Computing and Mathematics, College of Science and Engineering, University of Derby, Kedleston Road, Derby DE22 1GB, UK * Correspondence: y.xue@derby.ac.uk 1 School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China 1 School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China 2 Artiﬁcial Intelligence Research Institute, China University of Mining and Technology, Xuzhou 221116, China 3 School of Computing and Mathematics, College of Science and Engineering, University of Derby, Kedleston Road, Derby DE22 1GB, UK Abstract: Aerosol Optical Depth (AOD), quantifying the amount of aerosol in the atmosphere, is widely regarded as a crucial indicator for research on atmospheric physics and regional air quality. At present, the inversion of AOD from observation of satellite remote sensing sensors has become the main technology for large-scale monitoring of aerosol load. The Huaihai Economic Zone is the connecting belt of two key areas of atmospheric governance (the Yangtze River Delta and the Beijing-Tianjin-Hebei region, China), and it has been suffering from air pollution for many years and few studies of AOD focus on this region. Therefore, the spatial and temporal characteristics of the AOD are explored using MODIS AOD data and AVHRR AOD data in this region during the period from 1982 to 2021 in this study. The correlation coefﬁcients between the AOD of satellite observation and actual air pollution were analyzed by combining PM2.5 pollutant concentration and air quality index (AQI) data. The results showed that the AOD is higher in the northwest than in the southeast, and it is different from season to season. Citation: Wu, S.; Xue, Y.; Sun, Y.; Jin, C.; Zhang, M.; Jiang, X.; Lu, X. Spatial and Temporal Variation of Aerosol Optical Depth in Huaihai Economic Zone from 1982 to 2021. Atmosphere 2023, 14, 822. https://doi.org/ 10.3390/atmos14050822 Academic Editors: Alexandros Papayannis and Dean Venables Keywords: aerosol; PM2.5; AQI; AOD; AVHRR; MODIS Keywords: aerosol; PM2.5; AQI; AOD; AVHRR; MODIS Keywords: aerosol; PM2.5; AQI; AOD; AVHRR; MODIS Received: 28 March 2023 Revised: 25 April 2023 Accepted: 30 April 2023 Published: 2 May 2023 atmosphere atmosphere Article Spatial and Temporal Variation of Aerosol Optical Depth in Huaihai Economic Zone from 1982 to 2021 The annual variation of AOD in the Huaihai Economic Zone is a W-shaped trend from 1982 to 2011, while the trend of annual AOD is decreasing after 2011. In terms of seasons, the whole differences in AOD are evident, exhibiting AOD values in summer > those in spring > those in autumn > and those in winter. Furthermore, it indicated that the quarterly and monthly variation of the AOD tends to be ﬂat in recent years. Since 2015, the concentration of PM2.5 has continued to decline, the same as that of AQI. Meanwhile, the quarterly and monthly differences in PM2.5 are still obvious, with higher PM2.5 in winter and lower PM2.5 in summer. However, it also represented that PM2.5 is signiﬁcantly higher in spring than in autumn from 2015 to 2018, which is the opposite for 2019 to 2021. Lastly, the correlation between AOD and PM2.5/AQI is also given; i.e., the correlation coefﬁcients of AOD with PM2.5/AQI are 0.84/0.82, with the highest correlation coefﬁcient in autumn (R = 0.86/0.91) and the lowest in winter (R = 0.46/0.48). Keywords: aerosol; PM2.5; AQI; AOD; AVHRR; MODIS atmosphere atmosphere 1. Introduction Aerosols are a collection of liquid and solid particles suspended in the air, which directly or indirectly affect the climate [1–3]. Moreover, their accumulation near the ground can cause a reduction in atmospheric visibility and deterioration of air quality, which can seriously affect the normal life of human beings and even endanger their health [4–8]. Aerosol pollution shows regionalization; therefore, it is important to study the physical and chemical properties of aerosols and monitor their spatiotemporal variability on a regional scale [9–12]. Aerosol optical depth (AOD) is an important indicator and key parameter for understanding atmospheric physics and regional air quality by quantifying the aerosol load in the atmosphere [13–15]. Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). There are two main methods to measure AOD in the atmosphere, which are ground- based observations using sunphotometer measurements on the ground and remote sensing https://www.mdpi.com/journal/atmosphere Atmosphere 2023, 14, 822. https://doi.org/10.3390/atmos14050822 Atmosphere 2023, 14, 822 2 of 17 inversion, which relies on sensor data from the satellite platform. At present, several ground-based monitoring networks have been established to provide AOD data in the world, such as AERosol RObotic NETwork (AERONET) [16], Multiﬁlter Rotating Shadow Band Radiometer (MFRSR) [17], Chinese Sun Hazemeter Network (CSHNET) [18], Chi- nese Aerosol Remote Sensing NETwork (CARSNET) [19], European Brewer Network (EuBrewNet) [20,21], Sun-sky radiometer Observation Network (SONET) [22], Sky Ra- diometer Network (SKYNET) [23], etc. However, these networks are overly dependent on the density of station deployment, which leads to the fact that it is difﬁcult to ensure the full integration of the regional scope and results in high construction and maintenance costs. With the continuous optimization and updating of satellite sensors and inversion algorithms, AOD measurements taken by remote sensors have become more and more accurate, and the calculation of AOD from satellite remote sensing data has become the main technique for the large-scale monitoring of aerosol loads [24]. There are many sen- sors that can monitor AOD, such as the Moderate Resolution Imaging Spectroradiometer (MODIS) [23], the Visible Infrared Imaging Radiometer (VIIRS) [25], the Advanced Very High Resolution Radiometer (AVHRR) [26], the Ozone Monitoring Instrument (OMI) [27], the Cloud-Aerosol Lidar and Infrared Pathﬁnder Satellite Observations (CALIPSO) [28], the Polarization and Anisotropy of Reﬂectances for Atmospheric Sciences coupled with Observations from a Lidar (PARASOL) [29], etc. Among them, the AVHRR, which is carried on the NOAA series of satellites, has been continuously conducting Earth observation missions since 1979, so it has an accumulation of data for more than 40 years and a great potential for applications. At the same time, the MODIS, which is developed by NASA and carried on the TERRA and AQUA satellites, is widely used for studying regional AOD due to its advantages, such as easy access and excellent spatial and temporal resolution. f h l d b f O d h l h Accurate information on the spatiotemporal distribution of AOD and the relationship between AOD and atmospheric environmental pollution is important for understanding Atmosphere 2023, 14, 822 3 of 17 onship the extent of atmospheric pollution and promoting inter-regional aerosol transport decom- pression government decisions. In this study, based on the AOD data from MODIS and AVHRR, the spatiotemporal variation characteristics of AOD in the Huaihai Economic Zone during 1982–2021 are investigated. We also combine PM2.5 pollutant concentration data and air quality index (AQI) data to analyze the correlation between the AOD observed by satellite and the actual air pollution, which provides effective data support for the implementation of regional coordinated development strategy among provinces and cities in the Huaihai Economic Zone and the establishment of a more effective new mechanism for regional synergistic pollution reduction. p p p g g p compression government decisions. In this study, based on the AOD data from MODIS and AVHRR, the spatiotemporal variation characteristics of AOD in the Huaihai Economic Zone during 1982–2021 are investigated. We also combine PM2.5 pollutant concentration data and air quality index (AQI) data to analyze the correlation between the AOD ob- served by satellite and the actual air pollution, which provides eﬀective data support for the implementation of regional coordinated development strategy among provinces and cities in the Huaihai Economic Zone and the establishment of a more eﬀective new mech- anism for regional synergistic pollution reduction. g y p p Many studies have shown that there is a correlation between AOD and PM2.5 concentrations [30–33], and many scholars have estimated near-ground atmospheric par- ticulate matter concentrations based on satellite remote sensing AOD products [34–38]. However, the correlation between AOD and PM2.5 concentrations on a time scale with peri- odic characteristics is ignored. In particular, it is also possible that the correlation between AOD and PM2.5 concentrations is different in various regions due to their geographical location and meteorological environment. Current domestic and international studies on the regional level of air pollution in China mainly focus on economically developed regions such as Beijing-Tianjin-Hebei [39,40], Yangtze River Delta [41–45], Pearl River Delta [46,47] and other key regions, such as the Fenwei Plain [48], and there are fewer studies on other regions. The Huaihai Economic Zone, established in 1986, is one of the earliest groups of regional economic cooperation organizations in China and is located in the combined zone of Beijing-Tianjin-Hebei, Central Plains Economic Zone, and Yangtze River Delta in China. It is a typical inter-provincial border cooperation region in China, covering the junction area of Jiangsu, Shandong, Henan, and Anhui provinces [49,50]. In the region, there are geopolitical connections, humanistic connections, abundant resources, and obvi- ous transportation location advantages. At the same time, it is also a key area for China’s national industrial transformation and the conversion of old and new dynamics, national food security, a new national economic development support belt, and an important coal and energy base in east China. In the overall pattern of national regional development, it resides in the strategic position of connecting the north and the south and carrying the east and the west and is also the key to speeding up the high-quality development of the four provinces of Jiangsu, Shandong, Henan, and Anhui. In addition, the region is located in the north–south climate transition zone of China, and as the connection zone between two key regions of air management around the Yangtze River Delta and Beijing-Tianjin-Hebei, it presents signiﬁcant composite pollution characteristics and becomes one of the most polluted areas in the country; thus, it becomes a key control area of the Ministry of Ecology and Environment. 2. Data and Methods 2.1. Study Area 2. Data and Methods 2.1. Study Area Th l tit d d l The latitude and longitude range of the study area is approximately 113–121◦E, 32–37◦ N, with a temperate continental monsoon climate. In this paper, 10 cities in the Huaihai Economic Zone are selected for exploration and analysis, namely, Xuzhou, Lianyungang, and Suqian in Jiangsu Province; Huaibei and Suzhou in Anhui Province; Zaozhuang, Jining, Heze and Linyi in Shandong Province; and Shangqiu in Henan Province. The Huaihai Economic Zone has a resident population of about 59.32 million, as shown in Figure 1, with several resource-based cities, and the 10 cities in the region still have a large proportion of industrial activity. The latitude and longitude range of the study area is approximately 113–121°E, 32– 37°N, with a temperate continental monsoon climate. In this paper, 10 cities in the Huaihai Economic Zone are selected for exploration and analysis, namely, Xuzhou, Lianyungang, and Suqian in Jiangsu Province; Huaibei and Suzhou in Anhui Province; Zaozhuang, Jining, Heze and Linyi in Shandong Province; and Shangqiu in Henan Province. The Huaihai Economic Zone has a resident population of about 59.32 million, as shown in Figure 1, with several resource-based cities, and the 10 cities in the region still have a large proportion of industrial activity. Figure 1. The population density in China and Huaihai Economic Zone. Figure 1. The population density in China and Huaihai Economic Zone. Figure 1. The population density in China and Huaihai Economic Zone. Figure 1. The population density in China and Huaihai Economic Zone. 2.2. Data Source and Description 2.2. Data Source and Description 2.2.1. Satellite AOD Data 2.2.1. Satellite AOD Data In this study, we mainly used the AVHRR AOD data product from Jin et al. [51] and the MCD19A2 V6 data product (https://ladsweb.modaps.eosdis.nasa.gov/search/; ac- cessed on 7 January 2023). Jin et al. [51] have retrieved and published the 1982–2016 AOD datasets in China based on AVHRR data with a spatial resolution of 5 km × 10 km. It has been veriﬁed that the correlation coeﬃcient between this AOD data product and the AER- ONET and CARSNET ground-measured AOD is 0.78, and 63.31% of the points are within the Expected Error (EE) range of ± (0.05 + 0.25τ). Moreover, the Root Mean Squared Error (RMSE) of this data is 0.26, which is better than AVHRR Deep Blue (DB) AOD in mainland China. The MCD19A2 V6 data product is MODIS Multi-angle Implementation of Atmos- pheric Correction (MAIAC) Land AOD gridded Level 2 product produced daily, and it is a global AOD product released by NASA with 1 km spatial resolution. Some traditional In this study, we mainly used the AVHRR AOD data product from Jin et al. [51] and the MCD19A2 V6 data product (https://ladsweb.modaps.eosdis.nasa.gov/search/; accessed on 7 January 2023). Jin et al. [51] have retrieved and published the 1982–2016 AOD datasets in China based on AVHRR data with a spatial resolution of 5 km × 10 km. It has been veriﬁed that the correlation coefﬁcient between this AOD data product and the AERONET and CARSNET ground-measured AOD is 0.78, and 63.31% of the points are within the Expected Error (EE) range of ± (0.05 + 0.25τ). Moreover, the Root Mean Squared Error (RMSE) of this data is 0.26, which is better than AVHRR Deep Blue (DB) AOD in mainland China. The MCD19A2 V6 data product is MODIS Multi-angle Implementation of Atmospheric Correction (MAIAC) Land AOD gridded Level 2 product produced daily, and it is a global AOD product released by NASA with 1 km spatial resolution. Some traditional AOD inversion algorithms, such as Dark Target (DT), assume that the land surface is uniform in the inversion process, but in fact the land surface is often non-uniform, especially in bright surface areas such as cities or deserts. The high reﬂectance and high heterogeneity of the land surface make AOD inversion extremely complex. 2.2.2. Ground-Based Monitoring Data The AOD ground-based monitoring data from the data of AERONET provided by the National Aeronautics and Space Administration (NASA) and the Centre National de la Recherche Scientiﬁque (CNRS) (https://aeronet.gsfc.nasa.gov/; accessed on 7 January 2023) were selected. The automatic tracking and scanning sun photometer (CE-318), developed and manufactured by CIMEL, is one of the key observing instruments of the AERONET observing network and is one of the most effective ways to observe the physical and optical properties of atmospheric aerosols, mainly for measuring the direct sun irradiance over multiple observation channels in the 340–1640 nm spectral range from the visible to the near infrared. In addition, the ground-based monitoring data for PM2.5 and AQI from the National Urban Real-Time Air Quality Release Platform (http://113.108.142.147:20035/; accessed on 7 January 2023) also are selected. It is an ofﬁcial data distribution platform based on the Ambient Air Quality Standard (GB 3095–2012), which uses monitoring sites across the country to collect data including hourly monitoring values of PM2.5 concentration and AQI. 2.2.1. Satellite AOD Data 2.2.1. Satellite AOD Data The MAIAC algorithm, which uses time series data to decouple aerosol and surface contributions, Atmosphere 2023, 14, 822 4 of 17 4 of 17 assumes that the surface state is stable in a short period of time and has spatial heterogeneity. At the same time, the inﬂuence of bidirectional surface reﬂectance is also considered in the proposed algorithm, and the combination of time series and spatial analysis helps to improve the quality of cloud and snow detection. Thus, it has higher data accuracy on heterogeneous surfaces due to its improved cloud detection and surface characterization using the MAIAC algorithm that includes time series and spatial processing. So, this data can well meet the requirements of different applications such as the analysis of atmospheric particulate pollution and aerosol environmental effects. Among them, the AVHRR AOD data from 2002–2016 were used for validation only. For the time series analysis, we used the AVHRR AOD for the period 1982–2002 and the Aqua MODIS AOD for the period 2002–2021. In addition, we retained their original resolution for the study. 2.3. Methods and Data Processing The pre-processing of MCD19A2 data mainly includes geo-correction, raster projection, mosaic averaging, outlier rejection, band extraction, and cropping. The MCD19A2 product uses a 10◦(longitude) × 10◦(latitude) (1200 km × 1200 km) tile approach to provide hierarchical data format (HDF), and this study area involves data with row numbers 26, 27, and 28 and column number 5. Since the data set includes aerosol parameters such as 550 nm AOD, percentage of ﬁne particles, and uncertainty factor, we extracted AOD and time band data from them for the study. After that, the AOD data were excluded and cropped, and only the AOD data in the study area were retained. Finally, the values of AOD at the points of interest were extracted and information such as the average value of AOD for each time period was calculated. The spatial scale of AERONET data is different from that of satellite data, and the AOD was processed in order to maintain spectral, spatial, and temporal consistency. The AOD of AERONET data at 550 nm was obtained according to the Ångström formula [52] and so was the interpolation of AOD at 440 nm and 870 nm following Equation (1). It can be simpliﬁed as following Equation (2). τ550 = τ500 exp −α × ln 500 550 (1) τ550 = τ500 × (500 550)α (2) (1) (2) ln() is the logarithmic function of ‘’; τ500 is the AOD value at 500 nm; α is the Ångström index at 440–870 nm; and τ550 is the AOD value at 550 nm obtained by interpolation. In this paper, we acquired the yearly average, quarterly average, monthly average, and weekly average synthetic images of MCD19A2 550 nm AOD data and AVHRR 550 nm Atmosphere 2023, 14, 822 5 of 17 5 of 17 AOD data, respectively, and analyzed the spatiotemporal characteristics of aerosols in the junction area of Jiangsu, Anhui, Shandong, and Henan at different scales, mainly includes the spatiotemporal dependence of weather and pollution, the characteristics of annual/seasonal/monthly variation and spatial distribution. At the same time, the correlative analysis of the measured data of PM2.5 and AQI with AOD was conducted. In the analysis process, we mainly used Pearson’s correlation (R) for the correlative analysis among the parameters, and the RMSE was used to determine the uncertainty of the data following Equations (2) and (3). 2.3. Methods and Data Processing In addition, we mainly use the standard deviation (STD) to quantify the ﬂuctuation between data following Equation (4). R = ∑n i=1(xi −x)(yi −y) q ∑n i=1(xi −x)2 × ∑n i=1(yi −y)2 (3) RMSE = s 1 n n ∑ i=1 (xi −yi)2 (4) STD = s 1 n −1 n ∑ i=1 (xi −x)2 (5) (3) (4) STD = s 1 n −1 n ∑ i=1 (xi −x)2 (5) (5) n denotes the number, xi and yi are the individual sample values of the two parameters, and x and y are the mean values of x and y, respectively. 3.2. Spatial Patterns of AOD 3.2. Spatial Patterns of AOD 3.2. Spatial Patterns of AOD 3.2. Spatial Patterns of AOD pt f In this paper, we used every ﬁve years as a time point, but we used AVHRR AOD data to show the changes during 1982–2002 and Aqua MODIS AOD data to show the changes during 2003–2021. Due to the quality control problem of the MODIS data, the data are abnormal at the 35°N boundary. Therefore, the data close to this location are dis- continuous. This phenomenon has little eﬀect on our research. Speciﬁcally, we divided the period of 1997–2006 into two periods, 1997–2002 and 2003–2006, because the resolution of the two data are diﬀerent. The results of annual and seasonal means of AOD for 1982– 2021 are shown in Figures 4 and 5. There are diﬀerences in the pillar industries, land use classiﬁcation, and population distribution of each city, all of which can lead to regional diﬀerences in AOD distribution. The annual means of AVHRR AOD show that the AOD of cities in the Huaihai Economic Zone (MEAN in 0.4–0.45) is rather uniformly distributed during the period before 2002. It can also be seen that the AOD in the northeastern part of the study area is lower than that in the southwestern part. In addition, the areas of heavy pollution were diﬀerent in each area; for example, there was a prominent high value in In this paper, we used every ﬁve years as a time point, but we used AVHRR AOD data to show the changes during 1982–2002 and Aqua MODIS AOD data to show the changes during 2003–2021. Due to the quality control problem of the MODIS data, the data are abnormal at the 35◦N boundary. Therefore, the data close to this location are discontinuous. This phenomenon has little effect on our research. Speciﬁcally, we divided the period of 1997–2006 into two periods, 1997–2002 and 2003–2006, because the resolution of the two data are different. The results of annual and seasonal means of AOD for 1982–2021 are shown in Figures 4 and 5. There are differences in the pillar industries, land use classiﬁcation, and population distribution of each city, all of which can lead to regional differences in AOD distribution. The annual means of AVHRR AOD show that the AOD of cities in the Huaihai Economic Zone (MEAN in 0.4–0.45) is rather uniformly distributed during the period before 2002. 3.1. Evaluations of AVHRR AOD and MODIS AOD Based on AERONET data within a ‘±1 h time window’ from the AVHRR/AQUA local time, the satellite data products in the study area during the study period are veriﬁed, respectively. Figure 2 shows that the AVHRR AOD data products and MODIS AOD data products have good accuracy, and the correlation coefﬁcients with AERONET reach 0.900 and 0.910, respectively. There is also a remarkable consistency in the trends (Figure 2). Then, the correlative analysis of AVHRR AOD data and MODIS AOD data is carried out, and the results show that they also show good spatiotemporal consistency in Figure 3. Therefore, it is feasible and reasonable to analyze the spatiotemporal distribution of the Huaihai Economic Zone in the past 40 years based on two satellite AOD data products. 14, 822 6 of 17 Figure 2. The trend consistency test of AVHRR (a) and MODIS (b) with AERONET AOD data, and the correlation validation of AVHRR (c) and MODIS (d) with AERONET AOD data. Figure 2. The trend consistency test of AVHRR (a) and MODIS (b) with AERONET AOD data, and the correlation validation of AVHRR (c) and MODIS (d) with AERONET AOD data. Figure 2. The trend consistency test of AVHRR (a) and MODIS (b) with AERONET AOD data, and the correlation validation of AVHRR (c) and MODIS (d) with AERONET AOD data. Figure 2. The trend consistency test of AVHRR (a) and MODIS (b) with AERONET AOD data, and the correlation validation of AVHRR (c) and MODIS (d) with AERONET AOD data. Atmosphere 2023, 14, 822 6 of 17 ta, and Figure 3. The trend consistency test of monthly AOD data between AVHRR and MODIS. Figure 3. The trend consistency test of monthly AOD data between AVHRR and MODIS. Figure 3. The trend consistency test of monthly AOD data between AVHRR and MODIS. Figure 3. The trend consistency test of monthly AOD data between AVHRR and MODIS. Figure 3. The trend consistency test of monthly AOD data between AVHRR and MODIS. Figure 3. The trend consistency test of monthly AOD data between AVHRR and MODIS. 3.2. Spatial Patterns of AOD 3.2. Spatial Patterns of AOD It can also be seen that the AOD in the northeastern part of the study area is lower than that in the southwestern part. In addition, the areas of heavy pollution were different in each area; for example, there was a prominent high value in the urban area of Heze City in the period of 1992–1996, while in the period of 1997–2002, the signiﬁcant high values were found in the urban parts of the mineral-rich city of Huaibei and Shangqiu. In addition, the annual means of MODIS AOD show that there are obvious regional differences in the study area, with decreasing direction from northwest to southeast. Among them, Heze City (MEAN = 0.626) has the highest AOD value, followed by Jining City (MEAN = 0.622). Lianyungang City (MEAN = 0.502) has lower AOD values. The spatial distribution of AOD also varies slightly among seasons, which may be related to meteorological factors, elevation, population, and so on [53]. For example, the eastern coastal area represented has high values of AOD in summer, which is due to the monsoon transported sea salt aerosol and water vapor from the sea, and the hygroscopic particulate matter in the air is easy to combine with water vapor to form a new aerosol, resulting in a large increase in AOD. In contrast, in autumn and winter, the contribution of water vapor to AOD decreases signiﬁcantly as the monsoon winds change direction to blow from land to sea, while the contribution of the particulate matter in the air to AOD increases, but the overall trend is slowly decreasing. The above reasons lead to the overall decline of AOD from northwest to southeast in the autumn and winter. However, it exhibits lower values of AOD in the southwestern region and higher values of AOD in the northern and southeastern regions in summer (Figure 5). On the other hand, as shown in Figure 6, the seasonal variations in AOD are distinct for different cities. For example, the difference in the average AOD of Shangqiu City (STD = 0.017), Suzhou City (STD = 0.021), and Huaiibei City (STD = 0.032) are slight in each quarter, while that of Jining City (STD = 0.106), Lianyungang city (STD = 0.102) and Suqian city (STD = 0.087) are very variable in all seasons. 3.2. Spatial Patterns of AOD 3.2. Spatial Patterns of AOD 7 of 17 ian- Atmosphere 2023, 14, 822 sphere 2023, 14, 822 7 City (STD 0.032) are slight in each quarter, while that of Jining City (STD 0.106), Lian yungang city (STD = 0.102) and Suqian city (STD = 0.087) are very variable in all seasons. 1992–1996 1997–2002 1982–1986 1987–1991 Atmosphere 2023, 14, 822 8 of 17 Figure 4. The annual spatial distribution of AOD from 1982 to 2021 over Huaihai Economic Zone. 2012–2016 2017–2021 2003–2006 2007–2011 Figure 4. The annual spatial distribution of AOD from 1982 to 2021 over Huaihai Economic Zon 1987–1991 1982–1986 1997–2002 8 of 1992–1996 mosphere 2023, 14, 822 2003–2006 2007–2011 2017–2021 2012–2016 Figure 4. The annual spatial distribution of AOD from 1982 to 2021 over Huaihai Economic Zone. Figure 4. The annual spatial distribution of AOD from 1982 to 2021 over Huaihai Economic Zone. Atmosphere 2023, 14, 822 8 of 17 Z Figure 4. The annual spatial distribution of AOD from 1982 to 2021 over Huaihai Economic Zone. Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. Autumn Winter Spring Summer Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. phere 2023, 14, 822 9 of Figure 6. The mean of seasonal and annual AOD of ten cities in Huaihai Economic Zone dur 1982–2021. 3.3. Temporal Variability of AOD The mean and standard deviation of AOD for each year in the Huaihai Econom Zone over the last four decades are shown in Figure 7. Overall, the AOD was higher d Figure 6. The mean of seasonal and annual AOD of ten cities in Huaihai Economic Zone during 1982–2021. Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. Autumn Winter Spring Summer Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. ere 2023, 14, 822 9 o Spring Summer Winter Autumn re 2023, 14, 822 Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. Figure 5. The seasonal spatial distribution of MODIS AOD over Huaihai Economic Zone from 1982 to 2021. to 2021. to 2021. to 2021. to 2021. to 2021. Figure 6. 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades The mean and standard deviation of AOD for each year in the Huaihai Economic Zone over the last four decades are shown in Figure 7. Overall, the AOD was higher during the period of 2003–2015, with an annual mean greater than 0.5. In addition, the annual average AOD was even greater than 0.6 during 2007 and 2010–2012. This suggests that the atmospheric turbidity was very high during this period, which may be the result of the intense industrial development that was going on [54]. Furthermore, the annual average AOD remained consistently and steadily below 0.5 for the past six years, and the annual average AOD was below 0.4 in 2018 (MEAN = 0.396) and 2021 (MEAN = 0.386). one over the last four decades are shown in Figure 7. Overall, the AOD was higher d g the period of 2003–2015, with an annual mean greater than 0.5. In addition, the ann verage AOD was even greater than 0.6 during 2007 and 2010–2012. This suggests that mospheric turbidity was very high during this period, which may be the result of tense industrial development that was going on [54]. Furthermore, the annual aver OD remained consistently and steadily below 0.5 for the past six years, and the ann verage AOD was below 0.4 in 2018 (MEAN = 0.396) and 2021 (MEAN = 0.386). average AOD was below 0.4 in 2018 (MEAN 0.396) and 2021 (MEAN 0.386). Fi 7 St ti ti f d t d d d i ti f AOD i H ih i E i Z Figure 7. Statistics of mean and standard deviation of AOD in Huaihai Economic Zone during 1 2021 Figure 7. Statistics of mean and standard deviation of AOD in Huaihai Economic Zone during 1982–2021. gure 7. Statistics of mean and standard deviation of AOD in Huaihai Economic Zone during 1 21 Figure 7. Statistics of mean and standard deviation of AOD in Huaihai Economic Zone during 1982–2021. During 1982–2001, the average AOD values in the study area showed a W-sha end. After that, between 2002 and 2011, the Huaihai Economic Zone showed two sta increasing trends in AOD, which coincided with the rapid industrialization of Ch ter 2000. The highest value of AOD was reached in 2011 (MEAN = 0.673). However, a 11, the AOD in the Huaihai Economic Zone showed a signiﬁcant decreasing tre hich is consistent with the results of Zhao et al. 3.2. Spatial Patterns of AOD 3.2. Spatial Patterns of AOD The mean of seasonal and annual AOD of ten cities in Huaihai Economic Zone du 1982–2021. Figure 6. The mean of seasonal and annual AOD of ten cities in Huaihai Economic Zone during 1982–2021. gure 6. The mean of seasonal and annual AOD of ten cities in Huaihai Economic Zone du 82–2021. Figure 6. The mean of seasonal and annual AOD of ten cities in Huaihai Economic Zone during 1982–2021. 9 of 17 9 of 17 Atmosphere 2023, 14, 822 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades Statistics of seasonal mean of AOD in Huaihai Economic Zone from 1982 to 2021 Figure 8. Statistics of seasonal mean of AOD in Huaihai Economic Zone from 1982 to 2021. Figure 8. Statistics of seasonal mean of AOD in Huaihai Economic Zone from 1982 to 2021. In terms of AOD diﬀerences between months, the trend in the average AOD is nearly the same across months (Figure 9). June (MEAN = 0.568) and July (MEAN = 0.553) had the highest mean values of AOD with more than 0.5 in 1982–2021, they were followed by Au- gust (MEAN = 0.499), April (MEAN = 0.480), March (MEAN = 0.472), May (MEAN = 0.466), October (MEAN = 0.428), February (MEAN = 0.426), September (MEAN = 0.399), Novem- ber (MEAN = 0.375), January (MEAN = 0.373) and December (MEAN = 0.399). This is con- sistent with the above inter-seasonal AOD diﬀerences. At the same time, we can ﬁnd that the monthly mean AOD trends have gradually leveled oﬀ in recent years, which may be due to the increasing intensity of local pollution control and the control of targeted sub- regional and sub-period pollution In terms of AOD differences between months, the trend in the average AOD is nearly the same across months (Figure 9). June (MEAN = 0.568) and July (MEAN = 0.553) had the highest mean values of AOD with more than 0.5 in 1982–2021, they were followed by Au- gust (MEAN = 0.499), April (MEAN = 0.480), March (MEAN = 0.472), May (MEAN = 0.466), October (MEAN = 0.428), February (MEAN = 0.426), September (MEAN = 0.399), Novem- ber (MEAN = 0.375), January (MEAN = 0.373) and December (MEAN = 0.399). This is consistent with the above inter-seasonal AOD differences. At the same time, we can ﬁnd that the monthly mean AOD trends have gradually leveled off in recent years, which may be due to the increasing intensity of local pollution control and the control of targeted sub-regional and sub-period pollution. 11 of 17 regional and sub period pollution. Furthermore, it can be noted from Figure 9 that a very high AOD is observed in May (MEAN = 1.419) and June (MEAN = 1.057) in 1996 speciﬁcally, which resulted in an aver- age AOD of 0.198 and 0.196 higher in that year than in 1995 and 1997, respectively. 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades [54]. This phenomenon can be relate During 1982–2001, the average AOD values in the study area showed a W-shaped trend. After that, between 2002 and 2011, the Huaihai Economic Zone showed two stages of increasing trends in AOD, which coincided with the rapid industrialization of China after 2000. The highest value of AOD was reached in 2011 (MEAN = 0.673). However, after 2011, the AOD in the Huaihai Economic Zone showed a signiﬁcant decreasing trend, which is consistent with the results of Zhao et al. [54]. This phenomenon can be related to the promulgation of a series of national air pollution control policies and the implementation of measures on straw burning, ﬁreworks discharge, and the improvement of automobile emission standards [55]. In addition, we can see that the AOD ﬂuctuates greatly in 1996, with the highest standard deviation (0.240) of AOD. The standard deviation of AOD is also relatively high at around 0.15 in 2010–2015, which reﬂects the large variation of regional AOD in these years. Overall, seasonal differences in AOD are signiﬁcant in Figure 8, displaying sum- mer (MEAN = 0.535) > spring (MEAN = 0.471) > autumn (MEAN = 0.400) > winter (MEAN = 0.377). This phenomenon is correlated with meteorological and other factors. For example, the high temperature and humidity in summer provide environmental con- ditions for photochemical reactions, which promotes aerosol formation and facilitates the hygroscopic expansion of water-soluble aerosols, leading to an increase in the AOD content [53,56,57]. On the other hand, spring is relatively dry, and it is the seeding season, which leads to more particulate matter in the air [58,59]. This is a rather unexpected ﬁnding, as the highest AOD for all seasons occurred in winter in 2013. Moreover, the average AOD in winter was higher than in autumn after 2013 with the exception of 2018, and the average AOD in summer was no longer prominently high. This also means that the authorities have achieved good results in controlling AOD in the summer. 10 of 17 AOD in es have Atmosphere 2023, 14, 822 Figure 8. Statistics of seasonal mean of AOD in Huaihai Economic Zone from 1982 to 2021. Figure 8. Statistics of seasonal mean of AOD in Huaihai Economic Zone from 1982 to 2021. Figure 8 Statistics of seasonal mean of AOD in Huaihai Economic Zone from 1982 to 2021 Figure 8. 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades Statistics of monthly mean of AOD in Huaihai Economic Zone from 1982 to 2021. Figure 9. Statistics of monthly mean of AOD in Huaihai Economic Zone from 1982 to 2021. Figure 9. Statistics of monthly mean of AOD in Huaihai Economic Zone from 1982 to 2021. Figure 9. Statistics of monthly mean of AOD in Huaihai Economic Zone from 1982 to 2021. 3.4. Relationship between AOD and PM2.5/AQI According to ambient air quality standards (GB 3095-2012) of China, the ambient air functional zones are divided into nature reserves and other areas requiring special pro- tection (Class I) and residential areas, commercial traﬃc and residential mixed areas, cul- tural areas, industrial areas, and rural areas (Class II). Since air quality stations are mainly located in Class II areas, the secondary concentration limit of PM2.5 (daily average is 75µg/m3 and annual average is 35µg/m3) is adopted in this study as the criterion to judge hethe the PM o e t atio i thi a ea e eed the ta da d We o du ted tati ti Furthermore, it can be noted from Figure 9 that a very high AOD is observed in May (MEAN = 1.419) and June (MEAN = 1.057) in 1996 speciﬁcally, which resulted in an average AOD of 0.198 and 0.196 higher in that year than in 1995 and 1997, respectively. Similarly, a surprisingly high average AOD was recorded in August (MEAN = 1.228) 2011, resulting in the highest average AOD for that year in 40 years. Therefore, the occurrence of high values in some years is often due to anomalously high values resulting from the transit of high pollution in a speciﬁc month. 3.4. Relationship between AOD and PM2.5/AQI According to ambient air quality standards (GB 3095-2012) of China, the ambient air functional zones are divided into nature reserves and other areas requiring special pro- tection (Class I) and residential areas, commercial traﬃc and residential mixed areas, cul- tural areas, industrial areas, and rural areas (Class II). 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades Simi- larly, a surprisingly high average AOD was recorded in August (MEAN = 1.228) 2011, resulting in the highest average AOD for that year in 40 years. Therefore, the occurrence of high values in some years is often due to anomalously high values resulting from the transit of high pollution in a speciﬁc month. Figure 9. Statistics of monthly mean of AOD in Huaihai Economic Zone from 1982 to 2021. 3.4. Relationship between AOD and PM2.5/AQI According to ambient air quality standards (GB 3095-2012) of China, the ambient air functional zones are divided into nature reserves and other areas requiring special pro- tection (Class I) and residential areas, commercial traﬃc and residential mixed areas, cul- tural areas, industrial areas, and rural areas (Class II). Since air quality stations are mainly located in Class II areas, the secondary concentration limit of PM2.5 (daily average is 75µg/m3 and annual average is 35µg/m3) is adopted in this study as the criterion to judge whether the PM2.5 concentration in this area exceeds the standard. We conducted statistics Figure 9. Statistics of monthly mean of AOD in Huaihai Economic Zone from 1982 to 2021. Furthermore, it can be noted from Figure 9 that a very high AOD is observed in May (MEAN = 1.419) and June (MEAN = 1.057) in 1996 speciﬁcally, which resulted in an average AOD of 0.198 and 0.196 higher in that year than in 1995 and 1997, respectively. Similarly, a surprisingly high average AOD was recorded in August (MEAN = 1.228) 2011, resulting in the highest average AOD for that year in 40 years. Therefore, the occurrence of high values in some years is often due to anomalously high values resulting from the transit of high pollution in a speciﬁc month. Furthermore, it can be noted from Figure 9 that a very high AOD is observed in M (MEAN = 1.419) and June (MEAN = 1.057) in 1996 speciﬁcally, which resulted in an a age AOD of 0.198 and 0.196 higher in that year than in 1995 and 1997, respectively. S larly, a surprisingly high average AOD was recorded in August (MEAN = 1.228) 2 resulting in the highest average AOD for that year in 40 years. Therefore, the occurre of high values in some years is often due to anomalously high values resulting from transit of high pollution in a speciﬁc month. Figure 9. 3.4. Relationship between AOD and PM2.5/AQI 3.4. Relationship between AOD and PM2.5/AQI According to ambient air quality standards (GB 3095-2012) of China, the ambient air functional zones are divided into nature reserves and other areas requiring special protection (Class I) and residential areas, commercial trafﬁc and residential mixed areas, cultural areas, industrial areas, and rural areas (Class II). Since air quality stations are mainly located in Class II areas, the secondary concentration limit of PM2.5 (daily average is 75 µg/m3 and annual average is 35 µg/m3) is adopted in this study as the criterion to judge whether the PM2.5 concentration in this area exceeds the standard. We conducted statistics and analysis of PM2.5 and AQI data from ground-based air quality monitoring stations, and the annual average values and PM2.5 concentration exceedances are shown in Table 1. It is very obvious that the annual average AQI and PM2.5 concentrations are decreasing since 2015. The frequency of PM2.5 concentration exceedances is also declining, which means that heavy pollution is decreasing, and air quality is improving. There is slightly a difference in 2019, in which the average value of pollution is lower than the previous year, but the number of PM2.5 concentration exceedances of the standard is somewhat higher. Table 1. The statistics of AQI annual average, PM2.5 concentration annual average, the number of days when the number of stations with PM2.5 concentration exceeding the standard accounted for more than 30%, and PM2.5 annual exceedance rate for Huaihai Economic Zone from 2015 to 2021. Table 1. The statistics of AQI annual average, PM2.5 concentration annual average, the number of days when the number of stations with PM2.5 concentration exceeding the standard accounted for more than 30%, and PM2.5 annual exceedance rate for Huaihai Economic Zone from 2015 to 2021. Year AQI PM2.5 (µg/m3) Days over the Limit (d) Rate of PM2.5 Exceeds the Standard 2015 116.94 82.37 236 47.03% 2016 107.67 74.21 173 38.49% 2017 103.42 69.61 164 36.60% 2018 95.51 62.48 133 29.68% 2019 93.53 60.62 135 29.79% 2020 82.99 53.36 109 23.71% 2021 79.77 47.07 100 20.10% Atmosphere 2023, 14, 822 12 of 17 The seasonal PM2.5 exceedances for each year were calculated, and it can be seen in Figure 10 that PM2.5 exceedances were mainly concentrated in winter, while the exceedance rate in summer was still low, relatively. Surprisingly, the exceedances were explicitly higher in spring than in autumn until 2018, which was reversed after 2019. 3.3. Temporal Variability of AOD The mean and standard de one over the last four decades Since air quality stations are mainly located in Class II areas, the secondary concentration limit of PM2.5 (daily average is 75µg/m3 and annual average is 35µg/m3) is adopted in this study as the criterion to judge h h h PM i i hi d h d d W d d i i Furthermore, it can be noted from Figure 9 that a very high AOD is observed in May (MEAN = 1.419) and June (MEAN = 1.057) in 1996 speciﬁcally, which resulted in an average AOD of 0.198 and 0.196 higher in that year than in 1995 and 1997, respectively. Similarly, a surprisingly high average AOD was recorded in August (MEAN = 1.228) 2011, resulting in the highest average AOD for that year in 40 years. Therefore, the occurrence of high values in some years is often due to anomalously high values resulting from the transit of high pollution in a speciﬁc month. Atmosphere 2023, 14, 822 11 of 17 11 of 17 3.4. Relationship between AOD and PM2.5/AQI In addition, the exceedance rate suddenly picks up in the winter of 2019 and gradually decreases thereafter. In the spring, the exceedance rate is on a downward trend from 2015 to 2020 and only rises in 2021, but the rise is not high and remains smaller than in 2019 and before. On the other hand, it is gratifying to ﬁnd that the PM2.5 exceedance rate has been decreasing in the autumn and summer. exceedance rate in summer was still low, relatively. Surprisingly, the exceedances were explicitly higher in spring than in autumn until 2018, which was reversed after 2019. In addition, the exceedance rate suddenly picks up in the winter of 2019 and gradually de- creases thereafter. In the spring, the exceedance rate is on a downward trend from 2015 to 2020 and only rises in 2021, but the rise is not high and remains smaller than in 2019 and before. On the other hand, it is gratifying to ﬁnd that the PM2.5 exceedance rate has been decreasing in the autumn and summer. Figure 10. The statistics of PM2.5 exceeding standard rate in each quarter of Huaihai Economic Zone from 2015 to 2021. Figure 10. The statistics of PM2.5 exceeding standard rate in each quarter of Huaihai Economic Zone from 2015 to 2021. Figure 10. The statistics of PM2.5 exceeding standard rate in each quarter of Huaihai Economic Zone from 2015 to 2021. Figure 10. The statistics of PM2.5 exceeding standard rate in each quarter of Huaihai Economic Zone from 2015 to 2021. Atmosphere 2023, 14, 822 12 of 17 ic Zone 12 of 17 ic Zone Subsequently, we have also statistics on the monthly situation of the Huaihai Eco- nomic Zone, and the data show that the monthly trend is relatively consistent for each year, showing a U-shape (Figure 11). Among them, July (MEAN = 7.78%) and August (MEAN = 6.23%) have small exceedance rates, both less than 8%. However, the exceedances in January (MEAN = 67.09%), February (MEAN = 53.09%), and December (MEAN = 59.97%) were very serious, all greater than 50%. At the same time, it can be clearly seen that the vari- ation of each month in each year does not exactly coincide with the overall variation of each year. 3.4. Relationship between AOD and PM2.5/AQI For example, the exceedances in November and December in 2016 (MEAN11 = 56.53%, MEAN12 = 74.48%) were higher than in 2015 (MEAN11 = 49.89%, MEAN12 = 67.92%), al- though the annual exceedance rate in 2015 (MEAN = 47.03%) was higher than in 2016 (MEAN = 38.49%). Moreover, the highest value of the exceedance rate for each month from 2015–2021 was found in January 2020, reaching 75.28%. This reminds us that the number of heavily polluted days does not necessarily decrease when annual pollution is decreasing, especially in the months with the high frequency of pollution. Subsequently, we have also statistics on the monthly situation of the Huaihai Eco- nomic Zone, and the data show that the monthly trend is relatively consistent for each year, showing a U-shape (Figure 11). Among them, July (MEAN = 7.78%) and August (MEAN = 6.23%) have small exceedance rates, both less than 8%. However, the exceed- ances in January (MEAN = 67.09%), February (MEAN = 53.09%), and December (MEAN = 59.97%) were very serious, all greater than 50%. At the same time, it can be clearly seen that the variation of each month in each year does not exactly coincide with the overall variation of each year. For example, the exceedances in November and December in 2016 (MEAN11 = 56.53%, MEAN12 = 74.48%) were higher than in 2015 (MEAN11 = 49.89%, MEAN12 = 67.92%), although the annual exceedance rate in 2015 (MEAN = 47.03%) was higher than in 2016 (MEAN = 38.49%). Moreover, the highest value of the exceedance rate for each month from 2015–2021 was found in January 2020, reaching 75.28%. This reminds us that the number of heavily polluted days does not necessarily decrease when annual pollution is decreasing, especially in the months with the high frequency of pollution. Figure 11. The statistics of PM2.5 exceeding standard rate in each month of Huaihai Economic Zone from 2015 to 2021. Figure 11. The statistics of PM2.5 exceeding standard rate in each month of Huaihai Economic Zone from 2015 to 2021. In order to further understand the air pollution in the Huaihai Economic Zone, we conducted statistics and analysis on the average PM2.5 concentration and AQI of ten cities. According to the data, Heze has the most serious air pollution, with the highest PM2.5 concentration (MEAN = 61.52 µg/m3) and AQI (MEAN = 100.04) among the ten cities, which are far higher than other cities. 3.4. Relationship between AOD and PM2.5/AQI This is due to the fact that the concentration of PM2.5 is related to the source of pollution, atmospheric deposition, diffusion degree, vegetation adsorption, and other factors. In the winter and spring, the vegetation coverage is low and the wind speed is high, which makes it easy to produce dusty weather. Moreover, coal-burning heating in northern areas also releases a lot of soot, which leads to an increase in PM2.5 concentration. In summer, there are many rainy days, and the washing of rain is beneﬁcial to the deposition of PM2.5, thus reducing the PM2.5 concentration [61]. The Huaihai Economic zone belongs to the temperate continental monsoon climate, which has the characteristics of mild, humid, rain, and heat at the same time. There is a large part of plain terrain with low topography, and the vegetation is mainly deciduous broad-leaved vegetation. Aerosol is a general term for liquid or solid particles suspended in the atmosphere, so the concentration of aerosol is positively correlated with the water vapor content in the atmosphere, in summer, there are many clouds in the sky during the rainy season, and hygroscopic particles are easy to form new aerosols with water vapor, so the AOD value is large [62]. In winter and spring, the weather is sunny and the air is dry, so the contribution of water vapor to AOD decreases signiﬁcantly, while the contribution of airborne particles to AOD increases, but the overall AOD shows a decreasing trend. 14 of 17 Figure 12. Annual AOD, AQI, and PM2.5 concentrations in Huaihai Economic Zone from 2015 to 2021 Figure 12. Annual AOD, AQI, and PM2.5 concentrations in Huaihai Economic Zone from 2015 to 2021. Figure 12. Annual AOD, AQI, and PM2.5 concentrations in Huaihai Economic Zone from 2015 to 2021 Figure 12. Annual AOD, AQI, and PM2.5 concentrations in Huaihai Economic Zone from 2015 to 2021. 3.4. Relationship between AOD and PM2.5/AQI Lianyungang has the cleanest air, with a PM2.5 concentration of 43.16 µg/m3 and an AQI of 73.77. This is due to the high air humidity in the Lianyungang coastal area, which is not conducive to the growth of particulate matter. According to PM2.5 concentrations in descending order, the other cities are Jining (MEAN = 54.80 µg/m3), Suzhou (MEAN = 54.70 µg/m3), zaozhuang (MEAN = 53.87 µg/m3), Shangqiu (MEAN = 51.73 µg/m3), Xuzhou (MEAN = 49.17 µg/m3), Linyi (MEAN = 47.80 µg/m3), Suqian (MEAN = 47.68 µg/m3), and Huaibei (MEAN = 47.63 µg/m3). This is closely related to the industrial structure and regional geography of each city. If we follow the AQI values, the ranking between cities differs somewhat from the above, in order of zaozhuang (MEAN = 90.71), Jining (MEAN = 89.50), Shangqiu (MEAN = 86.45), Linyi (MEAN = 85.72), Suzhou (MEAN = 84.52), Xuzhou (MEAN = 84.30), Huaibei (MEAN = 79.56) and Suqian (MEAN = 78.14). Finally, we studied the correlation coefﬁcient between AOD and PM2.5 concentra- tions/AQI for the period 2015 to 2021 based on MODIS AOD data, PM2.5 concentration data, and AQI data. The results can be seen in Figure 12, and the annual trends of the three are generally consistent. After extensive data calculations, the correlation between AOD and PM2.5 was 0.84, and the correlation between AOD and AQI was 0.82. Because of the signiﬁcant differences in the seasonal changes in AOD and PM2.5, we also calculated Atmosphere 2023, 14, 822 13 of 17 13 of 17 the correlation coefﬁcients of the three by season. The results showed that the correlation coefﬁcients between AOD and PM2.5 and AQI were the highest in autumn, at 0.86 and 0.91, respectively. In winter, the correlation coefﬁcients between AOD and PM2.5 and AQI were the lowest, at 0.46 and 0.48, respectively. This result is consistent with the ﬁndings of many existing studies [34,60]. It also means that the correlation between AOD and AQI was higher than that of PM2.5 in autumn and winter. However, this was not the case in spring and autumn. In spring, the correlation coefﬁcients of AOD with PM2.5 and AQI were 0.79 and 0.71, respectively. In summer, the correlation coefﬁcients of AOD with PM2.5 and AQI were 0.66 and 0.65, respectively. In addition, we also noticed the variations and the totally reversed seasonal characteristics between AOD and PM2.5. 2021. 4. Conclusions 4. Conclusions AOD is closely related to meteorological climate and human life and property safety. In this paper, we analyzed the spatial and temporal evolution characteristics of the AOD in the Huaihai Economic Zone over the past four decades based on daily AOD products of AVHRR and MODIS. The correlation coeﬃcient between AOD and PM2.5/AQI is ex- plored by combining PM2.5 concentration data and AQI data from air quality monitoring stations. The results show that the geographical diﬀerences of AOD gradually increase with the intensiﬁcation of industrial production activities and the development of urban- ization. Due to the diﬀerences in the meteorological environment and geography of the four seasons, the distribution of AOD in the study area is not exactly the same in all four seasons. In terms of temporal variation, with the development of industrialization, the AOD is closely related to meteorological climate and human life and property safety. In this paper, we analyzed the spatial and temporal evolution characteristics of the AOD in the Huaihai Economic Zone over the past four decades based on daily AOD products of AVHRR and MODIS. The correlation coefﬁcient between AOD and PM2.5/AQI is explored by combining PM2.5 concentration data and AQI data from air quality monitoring stations. The results show that the geographical differences of AOD gradually increase with the intensiﬁcation of industrial production activities and the development of urbanization. Due to the differences in the meteorological environment and geography of the four seasons, the distribution of AOD in the study area is not exactly the same in all four seasons. In terms of temporal variation, with the development of industrialization, the AOD shows an upward trend until it peaked in 2011. There is a signiﬁcant downward trend after the Atmosphere 2023, 14, 822 14 of 17 14 of 17 aggressive implementation of local clean air policies. In recent years, the monthly and quarterly differences in AOD have gradually decreased, which means that the targeted pollution reduction measures in the Huaihai Economic Zone have played a good role. The data from the air quality site showed that the air quality in the Huaihai Economic Zone has improved signiﬁcantly in recent years, and PM2.5 air pollution has been controlled effectively. At the same time, the PM2.5/AQI showed a signiﬁcant correlation with AOD and the highest correlation coefﬁcient in autumn. 2021. 4. Conclusions This study provides data support for atmospheric environmental monitoring, pollu- tion assessment, and traceability, and helps local authorities to better utilize the coordination and linkage mechanism of the Huaihai Economic Zone to develop holistic management initiatives. The “integration” to solve the “fragmentation” problem, which can promote the development of the overall regional ecological environment, and further improve the atmospheric quality of the Huaihai Economic Zone. This is of great signiﬁcance to improve regional strategic coordination, integrated development, regional cooperation and mutual assistance, and inter-regional beneﬁt compensation. However, it is not enough to grasp the spatial and temporal distribution of AOD at present. We need to understand the sources of pollution as well as explore the reasons for the changes in AOD, which is the next step in our research. Author Contributions: Conceptualization, S.W. and Y.X.; methodology, S.W.; validation, S.W., C.J. and Y.X.; formal analysis, S.W.; investigation, S.W.; resources, S.W. and C.J.; data curation, S.W. and Y.S.; writing—original draft preparation, S.W.; writing—review and editing, S.W., M.Z. and Y.X.; visualization, S.W.; supervision, X.J. and X.L.; funding acquisition, Y.X. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Fundamental Research Funds for the Central Universities, grant number 2022XSCX17. Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data sharing is not applicable to this article, as no new data were created or analyzed in this study. Data Availability Statement: Data sharing is not applicable to this article, as no new data were created or analyzed in this study. Acknowledgments: The authors are very grateful for the comments and remarks of the reviewers who helped to improve the manuscript. Thanks to the editors for all their work on this manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Kaufman, Y.J.; Tanré, D.; Boucher, O. A Satellite View of Aerosols in the Climate System. Nature 2002, 419, 215–223. [CrossRef] [PubMed] 2. Rosenfeld, D.; Lohmann, U.; Raga, G.B.; O’Dowd, C.D.; Kulmala, M.; Fuzzi, S.; Reissell, A.; Andreae, M.O. Flood or Drought: How Do Aerosols Affect Precipitation? Science 2008, 321, 1309–1313. [CrossRef] [PubMed] p G.; Highwood, E.J. Local and Remote Impacts of Aerosol Species on Indian Summer Monsoon Rainfall in a 29, 6937–6955. [CrossRef] 3. Guo, L.; Turner, A.G.; Highwood, E.J. Local and Remote Impacts of Aerosol Species on Indian Sum GCM. J. Clim. 2016, 29, 6937–6955. [CrossRef] 4. Kloog, I.; Koutrakis, P.; Coull, B.A.; Lee, H.J.; Schwartz, J. Assessing Temporally and Spatially Resolved PM2.5 Exposures for Epidemiological Studies Using Satellite Aerosol Optical Depth Measurements. Atmos. Environ. 2011, 45, 6267–6275. [CrossRef] 4. Kloog, I.; Koutrakis, P.; Coull, B.A.; Lee, H.J.; Schwartz, J. Assessing Temporally and Spatially Resolved PM2.5 Exposures for Epidemiological Studies Using Satellite Aerosol Optical Depth Measurements. Atmos. Environ. 2011, 45, 6267–6275. [CrossRef] 5. Evans, J.; van Donkelaar, A.; Martin, R.V.; Burnett, R.; Rainham, D.G.; Birkett, N.J.; Krewski, D. Estimates of Global Mortality Attributable to Particulate Air Pollution Using Satellite Imagery. Environ. Res. 2013, 120, 33–42. [CrossRef] g J J g p y p y p Epidemiological Studies Using Satellite Aerosol Optical Depth Measurements. Atmos. Environ. 2011, 45, 6267–6275. [CrossRef] 5. Evans, J.; van Donkelaar, A.; Martin, R.V.; Burnett, R.; Rainham, D.G.; Birkett, N.J.; Krewski, D. Estimates of Global Mortality p e o og c S es Us g S e e e oso Op c ep e s e e s 0 , , 6 6 6 5 [C oss e ] 5. Evans, J.; van Donkelaar, A.; Martin, R.V.; Burnett, R.; Rainham, D.G.; Birkett, N.J.; Krewski, D. Estimates of Global Mortality Att ib t bl t P ti l t Ai P ll ti U i S t llit I E i R 2013 120 33 42 [C R f] p g g p p , , [ ] 5. Evans, J.; van Donkelaar, A.; Martin, R.V.; Burnett, R.; Rainham, D.G.; Birkett, N.J.; Krewski, D. Estimates of Global Mortality Attributable to Particulate Air Pollution Using Satellite Imagery. Environ. Res. 2013, 120, 33–42. [CrossRef] g g y 6. Kumar, M.; Singh, R.S.; Banerjee, T. Associating Airborne Particulates and Human Health: Exploring Possibilities. Environ. Int. 2015, 84, 201–202. [CrossRef] 7. Wang, Y.; Jiang, J.H.; Su, H. 9. Chen, S.-P.; Lu, C.-H.; Davies, J.E.; Ou-Yang, C.-F.; Lin, N.-H.; Huff, A.K.; Pierce, B.R.; Kondragunta, S.; Wang, J.-L. Infusing Satellite Data into Aerosol Forecast for near Real-Time Episode Detection and Diagnosis in East Asia. Sci. Total Environ. 2023, 856, 158797. [CrossRef] 8. Banerjee, T.; Kumar, M.; Mall, R.K.; Singh, R.S. Airing “clean Air” in Clean India Mission. Environ. Sci. Pollut. Res. 2017, 24, 6399–6413. [CrossRef] References Atmospheric Responses to the Redistribution of Anthropogenic Aerosols. J. Geophys. Res. 2015, 120, 9625–9641. [CrossRef] 8. Banerjee, T.; Kumar, M.; Mall, R.K.; Singh, R.S. Airing “clean Air” in Clean India Mission. Environ. Sci. Pollut. Res. 2017, 24, 6399–6413. [CrossRef] 9. Chen, S.-P.; Lu, C.-H.; Davies, J.E.; Ou-Yang, C.-F.; Lin, N.-H.; Huff, A.K.; Pierce, B.R.; Kondragunta, S.; Wang, J.-L. Infusing Satellite Data into Aerosol Forecast for near Real-Time Episode Detection and Diagnosis in East Asia. Sci. Total Environ. 2023, 856, 158797. [CrossRef] 15 of 17 15 of 17 Atmosphere 2023, 14, 822 10. Sogacheva, L.; Rodriguez, E.; Kolmonen, P.; Virtanen, T.H.; Saponaro, G.; Leeuw, D.; Georgoulias, A.K. Spatial and Seasonal Variations of Aerosols over China from Two Decades of Multi-Satellite Observations. Part II: AOD Time Series for 1995–2017 Combined from ATSR ADV and MODIS C6.1 for AOD Tendencies Estimation. Atmos. Chem. Phys. Discuss. 2018, 18, 16631–16652. [CrossRef] 11. Wei, J.; Peng, Y.; Mahmood, R.; Sun, L.; Guo, J. Intercomparison in Spatial Distributions and Temporal Trends Derived from Multi-Source Satellite Aerosol Products. Atmos. Chem. Phys. 2019, 19, 7183–7207. [CrossRef] 12. Xia, X. Variability of Aerosol Optical Depth and Angstrom Wavelength Exponent Derived from AERONET Observations in Recent Decades. Environ. Res. Lett. 2011, 6, 044011. [CrossRef] 13. Bhatia, N.; Tolpekin, V.A.; Stein, A.; Reusen, I. Estimation of AOD under Uncertainty: An Approach for Hyperspectral Airborne Data. Remote Sens. 2018, 10, 947. [CrossRef] 14. Filonchyk, M.; Yan, H.; Zhang, Z.; Yang, S.; Li, W.; Li, Y. Combined Use of Satellite and Surface Ob Optical Depth in Different Regions of China. Sci. Rep. 2019, 9, 18650. [CrossRef] [PubMed] 15. Li, Y.; Xue, Y.; Guang, J.; de Leeuw, G.; Self, R.; She, L.; Fan, C.; Xie, Y.; Chen, G. Spatial and Temporal Distribution Characteristics of Haze Days and Associated Factors in China from 1973 to 2017. Atmos. Environ. 2019, 214, 116862. [CrossRef] 16. Holben, B.N.; Eck, T.F.; Slutsker, I.; Tanré, D.; Buis, J.P.; Setzer, A.; Vermote, E.; Reagan, J.A.; Kaufman, Y.J.; Nakajima, T.; et al. AERONET-A Federated Instrument Network and Data Archive for Aerosol Characterization. Remote Sens. Environ. 1998, 66, 1–16. [CrossRef] [ ] 17. Slater, J.F.; Dibb, J.E.; Campbell, J.W.; Moore, T.S. Physical and Chemical Properties of Surface and Column Aerosols at a Rural New England Site during MODIS Overpass. Remote Sens. Environ. 2004, 92, 173–180. [CrossRef] g g p 18. References Long Term Variations and Co Properties Based on MODIS and Ground-Based Data in Thailand. Atmos. Environ. 2022, 286, 11921 rties Based on MODIS and Ground-Based Data in Thailand. Atmos. Environ. 2022, 286, 119218. [CrossRef] 24. Wei, X.; Chang, N.-B.; Bai, K.; Gao, W. Satellite Remote Sensing of Aerosol Optical Depth: Advances, Challenges, and Perspectives. Crit. Rev. Environ. Sci. Technol. 2020, 50, 1640–1725. [CrossRef] 25. He, L.; Wang, L.; Lin, A.; Zhang, M.; Bilal, M.; Wei, J. Performance of the NPP-VIIRS and Aqua-MODIS Aerosol Optical Depth Products over the Yangtze River Basin. Remote Sens. 2018, 10, 117. [CrossRef] Y.; He, X.; de Leeuw, G.; Mei, L.; Che, Y.; Rippin, W.; Guang, J.; Hu, Y. Long-Time Series Aerosol Optical D 26. Xue, Y.; He, X.; de Leeuw, G.; Mei, L.; Che, Y.; Rippin, W.; Guang, J.; Hu, Y. Long-Time Series Aerosol Optical Depth Retrieval from AVHRR Data over Land in North China and Central Europe. Remote Sens. Environ. 2017, 198, 471–489. [CrossRef] S O Al i h h i l i l Cl d A l d S f di d d S 26. Xue, Y.; He, X.; de Leeuw, G.; Mei, L.; Che, Y.; Rippin, W.; Guang, J.; Hu, Y. Long Time Series Aerosol Optical Depth Retrieval from AVHRR Data over Land in North China and Central Europe. Remote Sens. Environ. 2017, 198, 471–489. [CrossRef] 27. Stammes, P. OMI Algorithm Theoretical Basis Document Volume III Clouds, Aerosols, and Surface UV Irradiance, 2nd ed.; Stammes, P., d S G dd d S l h C G b l S l 27. Stammes, P. OMI Algorithm Theoretical Basis Document Volume III Clouds, Aerosols, and Surface UV Irradiance, 2nd ed.; Stammes, P., Ed.; NASA Goddard Space Flight Center: Greenbelt, MD, USA, 2002; Volume 3, pp. 1–114. 27. Stammes, P. OMI Algorithm Theoretical Basis Document Volume III Clouds, Aerosols, and Surface UV Irradiance, 2nd ed.; Stammes, P., Ed.; NASA Goddard Space Flight Center: Greenbelt, MD, USA, 2002; Volume 3, pp. 1–114. 27. Stammes, P. OMI Algorithm Theoretical Basis Document Volume III Clouds, Aerosols, and Surface UV Ir Ed.; NASA Goddard Space Flight Center: Greenbelt, MD, USA, 2002; Volume 3, pp. 1–114. g f Ed.; NASA Goddard Space Flight Center: Greenbelt, MD, USA, 2002; Volume 3, pp. 1–114. p g pp N.; Anshumali; Solanki, R. Evaluation and Utilization of MODIS and CALIPSO Aerosol Retrievals over a Himalaya. Remote Sens. Environ. 2018, 206, 139–155. [CrossRef] 28. References Xin, J.; Wang, Y.; Li, Z.; Wang, P.; Hao, W.M.; Nordgren, B.L.; Wang, S.; Liu, G.; Wang, L.; Wen, T.; et al. Aerosol Optical Depth (AOD) and Angstrom Exponent of Aerosols Observed by the Chinese Sun Hazemeter Network from August 2004 to September 2005. J. Geophys. Res.-Atmos. 2007, 112, D05203. [CrossRef] p y 19. Che, H.; Zhang, X.; Chen, H.; Damiri, B.; Goloub, P.; Li, Z.; Zhang, X.; Wei, Y.; Zhou, H.; Dong, F.; et al. Instrument Calibration and Aerosol Optical Depth Validation of the China Aerosol Remote Sensing Network. J. Geophys. Res. Atmos. 2009, 114, D03206. [CrossRef] 20. Lopez-Solano, J.; Redondas, A.; Carlund, T.; Rodriguez-Franco, J.J.; Diemoz, H.; Leon-Luis, S.F.; Hernandez-Cruz, B.; Guirado- Fuentes, C.; Kouremeti, N.; Grobner, J.; et al. Aerosol Optical Depth in the European Brewer Network. Atmos. Chem. Phys. 2018, 18, 3885–3902. [CrossRef] 21. Rimmer, J.S.; Redondas, A.; Karppinen, T. EuBrewNet-A European Brewer Network (COST Action ES1207), an Overview. Atmos. Chem. Phys. 2018, 18, 10347–10353. [CrossRef] y 22. Ma, Y.; Li, Z.; Li, Z.; Xie, Y.; Fu, Q.; Li, D.; Zhang, Y.; Xu, H.; Li, K. Validation of MODIS Aerosol Opti 22. Ma, Y.; Li, Z.; Li, Z.; Xie, Y.; Fu, Q.; Li, D.; Zhang, Y.; Xu, H.; Li, K. Validation of MODIS Aerosol Optical Depth Retrieval over Mountains in Central China Based on a Sun-Sky Radiometer Site of SONET. Remote Sens. 2016, 8, 111. [CrossRef] 22. Ma, Y.; Li, Z.; Li, Z.; Xie, Y.; Fu, Q.; Li, D.; Zhang, Y.; Xu, H.; Li, K. Validation of MODIS Aerosol Optical Depth Retrieval over Mountains in Central China Based on a Sun-Sky Radiometer Site of SONET. Remote Sens. 2016, 8, 111. [CrossRef] 23 Pilahome O ; Ninssawan W ; Jankondee Y; Janjai S ; Kumharn W Long-Term Variations and Comparison of Aerosol Optical ountains in Central China Based on a Sun-Sky Radiometer Site of SONET. Remote Sens. 2016, 8, 111. [CrossR lahome, O.; Ninssawan, W.; Jankondee, Y.; Janjai, S.; Kumharn, W. Long-Term Variations and Comparison o y ome, O.; Ninssawan, W.; Jankondee, Y.; Janjai, S.; Kumharn, W. Long-Term Variations and Comparison of A 23. Pilahome, O.; Ninssawan, W.; Jankondee, Y.; Janjai, S.; Kumharn, W. Long-Term Variations and Co Properties Based on MODIS and Ground Based Data in Thailand Atmos Environ 2022 286 11921 23. Pilahome, O.; Ninssawan, W.; Jankondee, Y.; Janjai, S.; Kumharn, W. p p 34. Kim, S.-M.; Koo, J.-H.; Lee, H.; Mok, J.; Choi, M.; Go, S.; Lee, S.; Cho, Y.; Hong, J.; Seo, S.; et al. Co Korea Estimated from the Various Ground-Based and Satellite AOD. Appl. Sci.-Basel 2021, 11, 10755 References Wei, J.; Huang, W.; Li, Z.; Xue, W.; Peng, Y.; Sun, L.; Cribb, M. Estimating 1-Km-Resolution PM2.5 Con Using the Space-Time Random Forest Approach. Remote Sens. Environ. 2019, 231, 111221. [CrossRef] .; Huang, W.; Li, Z.; Xue, W.; Peng, Y.; Sun, L.; Cribb, M. Estimating 1-Km-Resolution PM2.5 Concentration ng, W.; Li, Z.; Xue, W.; Peng, Y.; Sun, L.; Cribb, M. Estimating 1-Km-Resolution PM2.5 Concentrations acros ace-Time Random Forest Approach. Remote Sens. Environ. 2019, 231, 111221. [CrossRef] , J ; g, ; , ; , ; g, ; , ; , g sing the Space-Time Random Forest Approach. Remote Sens. Environ. 2019, 231, 111221. [CrossRef] 37. Wei, J.; Li, Z.; Lyapustin, A.; Sun, L.; Peng, Y.; Xue, W.; Su, T.; Cribb, M. Reconstructing 1-Km-Resolution High-Quality PM2.5 Data Records from 2000 to 2018 in China: Spatiotemporal Variations and Policy Implications. Remote Sens. Environ. 2021, 252, 112136. [CrossRef] 38. Zhang, H.; Kondragunta, S. Daily and Hourly Surface PM2.5 Estimation from Satellite AOD. Earth Space Sci. 2021, 8, e2020EA001599. [CrossRef] 38. Zhang, H.; Kondragunta, S. Daily and Hourly Surface PM2.5 Estimation from Satellite AOD. Earth Space Sci. 2021, 8, e2020EA001599. [CrossRef] 39 S L Li R B Ti X P W i J A l i f th T l d S ti l V i ti f A l i th B iji Ti ji H b i R i 39. Sun, L.; Li, R.-B.; Tian, X.-P.; Wei, J. Analysis of the Temporal and Spatial Variation of Aerosols in the Beij with a 1 Km AOD Product. Aerosol Air Qual. Res. 2017, 17, 923–935. [CrossRef] 40. Zhao, C.; Wang, Q.; Ban, J.; Liu, Z.; Zhang, Y.; Ma, R.; Li, S.; Li, T. Estimating the Daily PM2.5 Concentration in the Beijing-Tianjin- Hebei Region Using a Random Forest Model with a 0.01 Degrees x 0.01 Degrees Spatial Resolution. Environ. Int. 2020, 134, 105297. [CrossRef] Xu, J.; Yuan, L.; Li, D.; Jin, Z.; Zhang, K. Aerosol Vertical Distribution and Sources Estimation at a Site of 41. Fan, W.; Qin, K.; Xu, J.; Yuan, L.; Li, D.; Jin, Z.; Zhang, K. Aerosol Vertical Distribution and Sourc Yangtze River Delta Region of China. Atmos. Res. 2019, 217, 128–136. [CrossRef] 42. Liu, Y.; de Leeuw, G.; Kerminen, V.-M.; Zhang, J.; Zhou, P.; Nie, W.; Qi, X.; Hong, J.; Wang, Y.; Ding, A.; et al. References Analysis of Aerosol Effects on Warm Clouds over the Yangtze River Delta from Multi-Sensor Satellite Observations. Atmos. Chem. Phys. 2017, 17, 5623–5641. [CrossRef] 43. Sun, E.; Che, H.; Xu, X.; Wang, Z.; Lu, C.; Gui, K.; Zhao, H.; Zheng, Y.; Wang, Y.; Wang, H.; et al. Variat Optical Depth over the Yangtze River Delta from 1980 to 2016. Theor. Appl. Climatol. 2019, 136, 363–375 44. Wang, X.; Guo, Z.; Wang, Y.; Chen, Y.; Zheng, X.; Xu, X. Monitoring Temporal-Spatial Variations of AOD over the Yangtze River Delta, China. Stoch. Environ. Res. Risk Assess. 2018, 32, 2619–2632. [CrossRef] 45. Xiao, Q.; Wang, Y.; Chang, H.H.; Meng, X.; Geng, G.; Lyapustin, A.; Liu, Y. Full-Coverage High-Resolution Daily PM2.5 Estimation Using MAIAC AOD in the Yangtze River Delta of China. Remote Sens. Environ. 2017, 199, 437–446. [CrossRef] g g 46. Song, W.; Jia, H.; Huang, J.; Zhang, Y. A Satellite-Based Geographically Weighted Regression Model for Regional PM2.5 Estimation over the Pearl River Delta Region in China. Remote Sens. Environ. 2014, 154, 1–7. [CrossRef] 47. Zhang, M.; Wang, Y.; Ma, Y.; Wang, L.; Gong, W.; Liu, B. Spatial Distribution and Temporal Variation of Aerosol Optical Depth and Radiative Effect in South China and Its Adjacent Area. Atmos. Environ. 2018, 188, 120–128. [CrossRef] 48. Li, X.; Xie, P.; Li, A.; Xu, J.; Ren, H.; Ren, B.; Li, Y.; Li, J. Study of Aerosol Characteristics and Sources Using MAX-DOAS Measurement during Haze at an Urban Site in the Fenwei Plain. J. Environ. Sci. 2021, 107, 1–13. [CrossRef] 49 S C il f h P l ’ R bli f Chi Th A l f h S C il h O ll U b Pl i f X h 48. Li, X.; Xie, P.; Li, A.; Xu, J.; Ren, H.; Ren, B.; Li, Y.; Li, J. Study of Aerosol Characteristics and Sources Using MAX DOAS Measurement during Haze at an Urban Site in the Fenwei Plain. J. Environ. Sci. 2021, 107, 1–13. [CrossRef] 49 State Council of the People’s Republic of China The Approval of the State Council on the Overall Urban Planning of Xuzhou g J , , [ ] Council of the People’s Republic of China. The Approval of the State Council on the Overall Urban Pla 49. State Council of the People’s Republic of China. The Approval of the State Council on the Overall Ur Available online: http://www.gov.cn/zhengce/content/2017-06/23/content_5204776.htm (accessed o 50. References National Development and Reform Commission. Notice of the National Development and Reform Commission concerning Printing and Distributing the Huaihe Ecological Ecomnmic Belt Development Plan. Available online: https://www.ndrc.gov.cn/ xxgk/zcfb/ghwb/201811/t20181107_962252_ext.html (accessed on 7 January 2023). 50. National Development and Reform Commission. Notice of the National Development and Reform Commission concerning Printing and Distributing the Huaihe Ecological Ecomnmic Belt Development Plan. Available online: https://www.ndrc.gov.cn/ xxgk/zcfb/ghwb/201811/t20181107_962252_ext.html (accessed on 7 January 2023). g g y 51. Jin, C.; Xue, Y.; Jiang, X.; Wu, S.; Sun, Y. Retrieval and Validation of Long-Term Aerosol Optical De China. Int. J. Digit. Earth 2022, 15, 1817–1832. [CrossRef] Å g 52. Ångström, A. The Parameters of Atmospheric Turbidity. Tellus 1964, 16, 64–75. [CrossRef] 52. Ångström, A. The Parameters of Atmospheric Turbidity. Tellus 1964, 16, 64–75. [CrossRef] 53. He, Q.; Zhang, M.; Huang, B. Spatio-Temporal Variation and Impact Factors Analysis of Satellite-Based Aerosol Optical Depth over China from 2002 to 2015. Atmos. Environ. 2016, 129, 79–90. [CrossRef] 54. Zhao, B.; Jiang, J.H.; Gu, Y.; Diner, D.; Worden, J.; Liou, K.-N.; Su, H.; Xing, J.; Garay, M.; Huang Regional Aerosol Particle Properties and Their Linkage to Emission Changes. Environ. Res. Lett. 20 Regional Aerosol Particle Properties and Their Linkage to Emission Changes. Environ. Res. Lett. 2017, 12, 054021. [CrossRef] 55. Mehta, M.; Singh, R.; Singh, A.; Singh, N.; Anshumali. Recent Global Aerosol Optical Depth Variations and Trends-A Comparative Study Using MODIS and MISR Level 3 Datasets Remote Sens Environ 2016 181 137 150 [CrossRef] g p g g [ ] 55. Mehta, M.; Singh, R.; Singh, A.; Singh, N.; Anshumali. Recent Global Aerosol Optical Depth Variations and Trends-A Comparative Study Using MODIS and MISR Level 3 Datasets. Remote Sens. Environ. 2016, 181, 137–150. [CrossRef] y g 56. Bian, Y.X.; Zhao, C.S.; Ma, N.; Chen, J.; Xu, W.Y. A Study of Aerosol Liquid Water Content Based on Hygroscopicity Measurements t Hi h R l ti H idit i th N th Chi Pl i At Ch Ph 2014 14 6417 6426 [C R f] y g 56. Bian, Y.X.; Zhao, C.S.; Ma, N.; Chen, J.; Xu, W.Y. A Study of Aerosol Liquid Water Content Based on Hygroscopicity Measurements at High Relative Humidity in the North China Plain. Atmos. Chem. Phys. 2014, 14, 6417–6426. [CrossRef] 57. Zhang, L.; Sun, J.Y.; Shen, X.J.; Zhang, Y.M.; Che, H.; Ma, Q.L.; Zhang, Y.W.; Zhang, X.Y.; Ogren, J.A. 60. Green, M.; Kondragunta, S.; Ciren, P.; Xu, C. Comparison of GOES and MODIS Aerosol Optical Depth (AOD) to Aerosol Robotic Network (AERONET) AOD and IMPROVE PM2.5 Mass at Bondville, Illinois. J. Air Waste Manag. Assoc. 2009, 59, 1082–1091. [CrossRef] References Kumar, A.; Singh, N.; Anshumali; Solanki, R. Evaluation and Utilization of MODIS and CALIPS Complex Terrain in Himalaya. Remote Sens. Environ. 2018, 206, 139–155. [CrossRef] 29. Formenti, P.; Mbemba Kabuiku, L.; Chiapello, I.; Ducos, F.; Dulac, F.; Tanré, D. Aerosol Optical Properties Derived from POLDER- 3/PARASOL (2005-2013) over the Western Mediterranean Sea- Part 1: Quality Assessment with AERONET and in Situ Airborne Observations. Atmos. Meas. Tech. 2018, 11, 6761–6784. [CrossRef] 30. Chudnovsky, A.; Tang, C.; Lyapustin, A.; Wang, Y.; Schwartz, J.; Koutrakis, P. A Critical Assessment of High-Resolution Aerosol Optical Depth Retrievals for Fine Particulate Matter Predictions. Atmos. Chem. Phys. 2013, 13, 10907–10917. [CrossRef] 31. Xin, J.; Gong, C.; Liu, Z.; Cong, Z.; Gao, W.; Song, T.; Pan, Y.; Sun, Y.; Ji, D.; Wang, L.; et al. The Obse between PM2.5 and AOD over China. J. Geophys. Res.-Atmos. 2016, 121, 10701–10716. [CrossRef] 32. Xu, Q.; Chen, X.; Yang, S.; Tang, L.; Dong, J. Spatiotemporal Relationship between Himawari-8 Hourly Columnar Aerosol Optical Depth (AOD) and Ground-Level PM2.5 Mass Concentration in Mainland China. Sci. Total Environ. 2021, 765, 144241. [CrossRef] [PubMed] 33. Yang, Q.; Yuan, Q.; Yue, L.; Li, T.; Shen, H.; Zhang, L. The Relationships between PM2.5 and Aero Mainland China: About and behind the Spatio-Temporal Variations. Environ. Pollut. 2019, 248, 526 33. Yang, Q.; Yuan, Q.; Yue, L.; Li, T.; Shen, H.; Zhang, L. The Relationships between PM2.5 and Aerosol Optical Depth (AOD) in Mainland China: About and behind the Spatio-Temporal Variations. Environ. Pollut. 2019, 248, 526–535. [CrossRef] [PubMed] 34. Kim, S.-M.; Koo, J.-H.; Lee, H.; Mok, J.; Choi, M.; Go, S.; Lee, S.; Cho, Y.; Hong, J.; Seo, S.; et al. Comparison of PM2.5 in Seoul, Korea Estimated from the Various Ground-Based and Satellite AOD. Appl. Sci.-Basel 2021, 11, 10755. [CrossRef] p p 34. Kim, S.-M.; Koo, J.-H.; Lee, H.; Mok, J.; Choi, M.; Go, S.; Lee, S.; Cho, Y.; Hong, J.; Seo, S.; et al. Comparison of PM2.5 in Seoul, Korea Estimated from the Various Ground-Based and Satellite AOD. Appl. Sci.-Basel 2021, 11, 10755. [CrossRef] Atmosphere 2023, 14, 822 16 of 17 35. Park, S.; Lee, J.; Im, J.; Song, C.-K.; Choi, M.; Kim, J.; Lee, S.; Park, R.; Kim, S.-M.; Yoon, J.; et al. Estimation of Spatially Continuous Daytime Particulate Matter Concentrations under All Sky Conditions through the Synergistic Use of Satellite-Based AOD and Numerical Models. Sci. Total Environ. 2020, 713, 136516. [CrossRef] [PubMed] 36. 58. Cheng, T.; Xu, C.; Duan, J.; Wang, Y.; Leng, C.; Tao, J.; Che, H.; He, Q.; Wu, Y.; Zhang, R.; et al. Seasonal Variation and Difference of Aerosol Optical Properties in Columnar and Surface Atmospheres over Shanghai. Atmos. Environ. 2015, 123, 315–326. [CrossRef] 59. Kang, N.; Kumar, K.R.; Hu, K.; Yu, X.; Yin, Y. Long-Term (2002–2014) Evolution and Trend in Collection 5.1 Level-2 Aerosol Products Derived from the MODIS and MISR Sensors over the Chinese Yangtze River Delta. Atmos. Res. 2016, 181, 29–43. [CrossRef] p p p g 59. Kang, N.; Kumar, K.R.; Hu, K.; Yu, X.; Yin, Y. Long-Term (2002–2014) Evolution and Trend in Collection 5.1 Level-2 Aerosol Products Derived from the MODIS and MISR Sensors over the Chinese Yangtze River Delta. Atmos. Res. 2016, 181, 29–43. [CrossRef] 61. Jin, J.; Liu, S.; Wang, L.; Wu, S.; Zhao, W. Fractional Vegetation Cover and Spatiotemporal Variations of PM2.5 Concentrations in the Beijing-Tianjin-Hebei Region of China. Atmosphere 2022, 13, 1850. [CrossRef] j g j g p , , [ ] 62. Qi, L.; Zheng, H.; Ding, D.; Ye, D.; Wang, S. Effects of Meteorology Changes on Inter-Annual Variations of Aerosol Optical Depth and Surface PM2.5 in China—Implications for PM2.5 Remote Sensing. Remote Sens. 2022, 14, 2762. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References Observations of Relative Humidity Effects on Aerosol Light Scattering in the Yangtze River Delta of China. Atmos. Chem. Phys. 2015, 15, 8439–8454. [CrossRef] 58. Cheng, T.; Xu, C.; Duan, J.; Wang, Y.; Leng, C.; Tao, J.; Che, H.; He, Q.; Wu, Y.; Zhang, R.; et al. Seasonal Variation and Difference of Aerosol Optical Properties in Columnar and Surface Atmospheres over Shanghai. Atmos. Environ. 2015, 123, 315–326. [CrossRef] ( ) l d d C ll l l 58. Cheng, T.; Xu, C.; Duan, J.; Wang, Y.; Leng, C.; Tao, J.; Che, H.; He, Q.; Wu, Y.; Zhang, R.; et al. Seasonal Variation and Difference of Aerosol Optical Properties in Columnar and Surface Atmospheres over Shanghai. Atmos. Environ. 2015, 123, 315–326. [CrossRef] 59. Kang, N.; Kumar, K.R.; Hu, K.; Yu, X.; Yin, Y. Long-Term (2002–2014) Evolution and Trend in Collection 5.1 Level-2 Aerosol Products Derived from the MODIS and MISR Sensors over the Chinese Yangtze River Delta. Atmos. Res. 2016, 181, 29–43. [CrossRef] 59. Kang, N.; Kumar, K.R.; Hu, K.; Yu, X.; Yin, Y. Long-Term (2002–2014) Evolution and Trend in Collection 5.1 Level-2 Aerosol Products Derived from the MODIS and MISR Sensors over the Chinese Yangtze River Delta. Atmos. Res. 2016, 181, 29–43. [CrossRef] 60. Green, M.; Kondragunta, S.; Ciren, P.; Xu, C. Comparison of GOES and MODIS Aerosol Optical Depth (AOD) to Aerosol Robotic Network (AERONET) AOD and IMPROVE PM2.5 Mass at Bondville, Illinois. J. Air Waste Manag. Assoc. 2009, 59, 1082–1091. [CrossRef] Atmosphere 2023, 14, 822 17 of 17 17 of 17 Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W1754962947	https://bmccancer.biomedcentral.com/counter/pdf/10.1186/1471-2407-5-118	English	null	Pre-exenterative chemotherapy, a novel therapeutic approach for patients with persistent or recurrent cervical cancer	BMC cancer	2,005	cc-by	7,044	BioMed Central BioMed Central BioMed Central Open Ac Research article Pre-exenterative chemotherapy, a novel therapeutic approach for patients with persistent or recurrent cervical cancer Carlos Lopez-Graniel1,5, Rigoberto Dolores1, Lucely Cetina2, Aaron Gonzalez1, David Cantu1, Jose Chanona3, Jesus Uribe2, Myrna Candelaria2, Rocio Brom4, Jaime de la Garza2 and Alfonso Duenas- Gonzalez2 Address: 1Division of Surgery, Instituto Nacional de Cancerología, Mexico, 2Division of Clinical Research, Instituto Nacional de Cancerología, Mexico, 3Department of Pathology, Instituto Nacional de Cancerología, Mexico, 4Department of CT scan, Instituto Nacional de Cancerología, Mexico and 5Unidad de Investigación Biomédica en Cáncer. Instituto Nacional de Cancerología/Instituto de Investigaciones Biomédicas, UNAM, Mexico mail: Carlos Lopez-Graniel - clopezg@incan.edu.mx; Rigoberto Dolores - doloresrv@hotmail.com; Email: Carlos Lopez-Graniel - clopezg@incan.edu.mx; Rigoberto Dolores - doloresrv@hotmail.com; Lucely Cetina - micuentalucely@yahoo.com; Aaron Gonzalez - agonzaleze@incan.edu.mx; David Cantu - dcantu3@ex Jose Chanona - jchanonav@incan.edu.mx; Jesus Uribe - maria_de_jesus@hotmail.com; Myrna Candelaria - myrnac@p Rocio Brom - rocitla@hotmail com; Jaime de la Garza - jdelagarza1@prodigy net mx; Alfonso Duenas- Email: Carlos Lopez-Graniel - clopezg@incan.edu.mx; Rigoberto Dolores - doloresrv@hotmail.com; Lucely Cetina - micuentalucely@yahoo.com; Aaron Gonzalez - agonzaleze@incan.edu.mx; David Cantu - dcantu3@excite.com; Jose Chanona - jchanonav@incan.edu.mx; Jesus Uribe - maria_de_jesus@hotmail.com; Myrna Candelaria - myrnac@prodigy.net.mx; Rocio Brom - rocitla@hotmail.com; Jaime de la Garza - jdelagarza1@prodigy.net.mx; Alfonso Duenas- Gonzalez - alfonso_duenasg@yahoo.com * Corresponding author Received: 12 January 2005 Accepted: 19 September 2005 Published: 19 September 2005 Received: 12 January 2005 Accepted: 19 September 2005 BMC Cancer 2005, 5:118 doi:10.1186/1471-2407-5-118 This article is available from: http://www.biomedcentral.com/1471-2407/5/118 © 2005 Lopez-Graniel et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. BMC Cancer Open Access Background pelvic side wall. This consists in extending the lateral resection plane of pelvic exenteration to the medial aspects of the lumbosacral plexus, sacrospinous ligament, acetabulum, and obturator membrane to allow for resec- tion with disease-free margins [11]. With this salvage approach, they have reported a 5-year survival probability of 46% for those patients considered only for palliation with current treatment options. Although these results are highly encouraging, severe postoperative complications occur in almost half of patients and the procedure is lim- ited to tumors sized <5 cm with a recurrence-free interval from primary radiation treatment of >5 months, and to recurrences that do not involve the larger sciatic foramen; all forms of parietal pelvic side wall disease are not suited for this procedure [12]. g Cervical cancer continues to be an important health bur- den with a yearly incidence of almost half a million new cases in the world and a mortality rate of about 50% [1]. Currently, locally advanced disease is treated with concur- rent cisplatin-based chemoradiation [2]. However, approximately in 25% of all patients treated for cervical carcinoma, the tumor will progress or recur locally [3,4], being the most common site of recurrence the pelvis. Thus, local relapse continues to be a significant problem for these patients, as tumor persistence or local recurrence in an irradiated pelvis indicates a very dismal prognosis [5,6]. Recurrent disease can be treated by a) chemoradiation if the primary disease was approached with surgery; b) palliative chemotherapy if recurrence is considered unresectable and the primary disease was treated with radiation or chemo- radiation; and c) pelvic exenteration for selected cases with small, central disease even if primarily treated with chem- oradiation or radiation. Pelvic exenteration involves en bloc resection of bladder, genital tract, and rectum; it was first described by Brunschwig in 1948 [7]. This procedure has curative potential in almost half of patients undergo- ing this procedure [6] and it is commonly reserved for only the small subgroup of recurrent disease patients who meet the "standard" criteria for exenteration (small, cen- tral tumors). However, most pelvic recurrences do show a diffuse growth pattern fixed to one or both pelvic side walls. These fixed recurrences are felt at physical examina- tion as "pelvic fibrosis" with or without a dominant mass. Thus, pelvic fibrosis, is an ominous finding significantly related to nodal disease and fixation to pelvic side wall [8]. Background Currently, a combination of cisplatin and paclitaxel has shown better response rate and progression- free survival than single agent cisplatin hence, combination chemo- therapy as been regarded as the standard of care in patients to be treated with systemic palliative chemother- apy [13]. Chemotherapy however, as a definitive treat- ment for recurrent cervical cancer has solely a palliative role, with responses that are at best partial and of short duration, as a consequence, almost all patients eventually show progression and die from their disease. Because objective responses are seen in almost a third of these patients, we reasoned that a "local" consolidation would potentially render some of these responding patients free of disease. These observations prompted us to evaluate in a pilot study, a treatment modality we have called "pre- exenterative chemotherapy" in patients with "fixed" pel- vic recurrence in the aimed to shrinking the pelvic recur- rent tumor to then attempt, then, a "standard" pelvic exenteration. Due to these facts, the vast majority of recurrent cervical cancer patients are left with no curative options, therefore it is important to search for other therapeutic alternatives in patients that are not "standardly" considered for the classical exenterative procedure. The introduction of high- dose-rate intraoperative radiation therapy (HDR-IORT) combined with radical surgical resection has widened the scope of patients who may be offered surgery [9], how- ever, this form of radiation delivery is not widely availa- ble. In addition, despite this modality of treatment provides a reasonable local control rate in patients who have failed prior surgery and/or definitive radiation, only those with complete gross resection at completion of sur- gery appear to benefit from this radical approach in the salvage setting [10]. Abstract Background: Most cervical cancer patients with pelvic recurrent or persistent disease are not candidates for exenteration, therefore, they only receive palliative chemotherapy. Here we report the results of a novel treatment modality for these patients pre-exenterative chemotherapy- under the rational that the shrinking of the pelvic tumor would allow its resection. Methods: Patients with recurrent or persistent disease and no evidence of systemic disease, considered not be candidates for pelvic exenteration because of the extent of pelvic tumor, received 3-courses of platinum-based chemotherapy. Response was evaluated by CT scan and bimanual pelvic examination; however the decision to perform exenteration relied on the physical findings. Toxicity to chemotherapy was evaluated with standard criteria. Survival was analyzed with the Kaplan-Meier method. Results: Seventeen patients were studied. The median number of chemotherapy courses was 4. There were 9 patients who responded to chemotherapy, evaluated by bimanual examination and underwent pelvic exenteration. Four of them had pathological complete response. Eight patients did not respond and were not subjected to surgery. One patient died due to exenteration complications. At a median follow- up of 11 months, the median survival for the whole group was 11 months, 3 months in the non-operated and 32 months in those subjected to exenteration. Conclusion: Pre-exenterative chemotherapy is an alternative for cervical cancer patients that are no candidates for exenteration because of the extent of the pelvic disease. Its place in the management of recurrent disease needs to be investigated in randomized studies, however, its value for offering long-term survival in some of these patients with no other option than palliative care must be stressed. Page 1 of 10 (page number not for citation purposes) BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 Pre-exenterative chemotherapy Chemotherapy was administered in an outpatient setting. Diverse chemotherapy schedules based on cisplatin or carboplatin were used as follows: Carboplatin AUC 5, d1, paclitaxel 135 mg/m2, d1 and gemcitabine 800 mg/m2 d1&8 (2 patients); carboplatin AUC 5–6, d1, and paclit- axel 135 mg/m2, d1 (3 patients); carboplatin AUC 5–6, d1, and 5FU 1 gr/m2 d1-5, (3 patients); cisplatin 100 mg/ m2 d1, and 5FU 1 g/m2 d1-5, (4 patients); and cisplatin 100 mg/m2 d1 and gemcitabine 1 g/m2 d1&d8 (5 patients). Courses were administered every three weeks for a maximum of 6 courses. Conventional antiemetic therapy and ancillary medications were used during drug treatment. Chemotherapy was stopped in cases of disease progression or prohibitive toxicity. * Radiation: 50Gy of external radiation plus brachytherapy to achieve at least 85Gy to point A. ** Six weekly applications of cisplatin at 40 mg/m2 during external radiation. sis of the persistent or recurrent disease until the date of death of last visit. Pelvic exenteration After pre-exenterative chemotherapy, patients were evalu- ated by the same team of gynecologists through pelvic examination (CT scan was not mandatory). However, the decision to proceed or not with the surgical procedure relied only on pelvic examination and was based on whether there was or not response as evaluated with above described criteria. The other criterion for no exenteration was a worsening of the general clinical con- dition of the patient regardless of the pelvic examination. Patients were followed every three months after comple- tion of all treatment. All patients had histological confirmation of their recur- rent disease. The clinical status at entering the study is shown in Table 2. All patients complained of pelvic pain. At physical pelvic exam the disease was felt as fixed to the pelvic wall in all cases, 5 (29%) unilaterally and 12 (71%) Methods All patients had histologically proven persistent or recur- rent cervical carcinoma to primary radiation or chemora- diation. At the pelvic examination -under no anesthesia-, these patients were felt to have pelvic fibrosis and diag- nosed as having a recurrence of diffuse infiltrative growth pattern (with or without a dominant mass) fixed or not to one or both pelvic side walls. Consequently, these patients were considered by the gynecologist team of our Institution (C L-G, A G-E, GM) to be unsuitable for pelvic exenteration regardless of the CT scan findings. Patients also had to meet the following inclusion criteria: Aged between 18 and 70 years; ECOG performance 0–1; ade- quate hematological, hepatic and renal functions as deter- mined by: hemoglobin equal or higher than 10 g/L, leukocyte count higher than 4000/mm3, and a platelet count of at least 100 000/mm3, total bilirubin less than 1.5 times the normal upper limit (NUL), transaminases less than 1.5 times NUL, and normal levels of creatinine Hokel et al., have recently described the laterally extended endopelvic resection (LEER) as a novel surgical salvage therapy to a selected subset of patients with locally advanced and recurrent cervical carcinoma involving the Page 2 of 10 (page number not for citation purposes) Page 2 of 10 (page number not for citation purposes) BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 http://www.biomedcentral.com/1471-2407/5/118 Table 1: Characteristics of patients Number 17 Age 43.3 (29–55) Histology Squamous 15 Adenocarcinoma 2 FIGO Stage at diagnosis 1B1 1 1B2 4 IIB 5 IIIB 7 Primary Treatment RT alone* 9 RT alone + Adj Hyst 4 Chemoradiation** 4 Response to Primary Treatment Complete response 13 Persistence 3 Progression 1 Months to treatment failure Recurrence (13 pts) 16 (9–120) progression (4 pts) 4 (2–7) * Radiation: 50Gy of external radiation plus brachytherapy to achieve at least 85Gy to point A. ** Six weekly applications of cisplatin at 40 mg/m2 during external radiation. Table 1: Characteristics of patients in serum; a normal posteroanterior chest X-ray as well as having the correspondent informed consent. The exclu- sion criteria included: severe systemic or uncontrolled dis- ease (infection, central nervous system, metabolic, etc) that precluded the use of chemotherapy and further exenteration; concomitant treatment with any other experimental drug; mental illness and previous or con- comitant malignancies except non-melanoma skin can- cer. The study was approved by the Institutional Regulatory Boards. Results From May 1999 to March 2003, 17 patients were studied in this pilot trial. Baseline characteristics of patients (at diagnosis of their primary disease) are shown in Table 1. The mean age of patients was 43.3 years and all, but two, were squamous histology. FIGO stage distribution was as follows: one patient was IB1, four were stage IB2, five IIB, and seven, stage IIIB. Nine received radiation alone as the definitive treatment of their primary disease, four were treated with radiation plus extrafacial complementary hysterectomy, and four patients received chemoradiation with weekly cisplatin. A complete clinical response was achieved in 13 patients after the primary treatment, three had persistent disease and one progressed. All cases accrued in this study had local pelvic relapse and the median time to progression after primary treatment was 16 months (9–120) in the 13 cases that had complete response, whereas the time to progression for the persist- ent or progressive disease cases was 4 months (range 2–7 months). Response and toxicity to pre-exenterative chemotherapy Response and toxicity to pre-exenterative chemotherapy Objective evaluation of response to chemotherapy using standard response criteria was not the primary objective of this study as this would have required that all patients had a well-defined and measurable mass. Instead a response to chemotherapy was defined when the pelvic disease was felt less fixed and/or the "fibrosis" was felt softer by the same team of gynecologists that performed the pre-chem- otherapy evaluation. Toxicity to chemotherapy was evalu- ated according to the NCI Common Toxicity Criteria. Survival This was accompanied by unilateral leg edema in six cases, hydronephrosis in three (18%) and both find- ings: edema and hydronephrosis in three cases (18%). bilaterally. This was accompanied by unilateral leg edema in six cases, hydronephrosis in three (18%) and both find- ings: edema and hydronephrosis in three cases (18%). Table 3 depicts the overall treatment received by the patients. The median number of cycles delivered was four (range 2–6 cycles). Evaluation of response following the aforementioned subjective criteria, performed by biman- ual pelvic examination, was achieved in nine patients and these underwent the exenterative procedure. Among the eight patients not exenterated, three showed progression alone, one had clinical deterioration with no change at pelvic examination and four had progression and clinical deterioration. Objective response was also evaluated using classical criteria in measurable disease (complete, no evidence of disease, partial, >50 reduction in the prod- uct of the two longest perpendicular diameters of the measurable lesion; no change or stable, <50% decrease or <25% increase, and progressive disease >25% increase). According to this, within the 8 non-operated patients, only four had pre and post-chemotherapy CT scans, three had no response and one had progression. These data cor- related well with that registered in the physical examina- tion. On the contrary, in the nine operated patients, five patients had pre and post-chemotherapy evaluation, and all five had partial response. This, also correlates with that perceived in the clinical examination. It is remarkable that within the operable patients, in no case an objective com- plete response was observed despite four cases had a path- ological complete response. Figure 1 shows that in the three cases with pathological complete response that had pre and post chemotherapy CT scan there was residual tumor after chemotherapy. All patients not subjected to exenteration showed disease progression and died within the ensuing months, being the median survival of only 3 months. Survival Overall survival was evaluated using the Kaplan-Meier method and was considered from the date of the diagno- Page 3 of 10 (page number not for citation purposes) http://www.biomedcentral.com/1471-2407/5/118 http://www.biomedcentral.com/1471-2407/5/118 BMC Cancer 2005, 5:118 Table 2: Clinical status of patients at entering the study Sign/Symptom Number (%) Pelvic pain 17 100 Fixation to pelvic side wall* Unilateral 5 29 Bilateral 12 71 Ipsilateral leg edema 6 36 Hydronefrosis 3 18 Leg edema/hydronephrosis 3 18 * As determined by bimanual pelvic examination. Table 3: Overall treatment Median number of cycles 4 Exenterated 9 No exenterated 8 Reason for no exenteration Progression 3 Clinical deterioration 1 Both 4 Table 2: Clinical status of patients at entering the study * As determined by bimanual pelvic examination. In regard to the exenterative procedure, a total infraeleva- tor exenteration was done in eight cases and one had ante- rior supraelevator exenteration. This patient was the one with positive surgical margins in the vaginal border. It is worth mentioning that, in this case, the transoperative frozen section of the vaginal margin was reported nega- tive; however, the definitive histological analysis showed disease. The definitive histological analysis of the surgical specimens showed a complete pathological response in 4 cases, a residual disease ≤2 cm in four cases, and one case with a residual measuring 8 cm. In seven patients, the uri- nary diversion consisted of an ileocolonic conduit and an ileal conduit in two cases. Colostomy was done in the eight cases undergoing total exenteration, (Table 5). Regarding surgical morbidity, the mean surgical time was 6.3 hours (range 4.3–8); the mean of bleeding was 1860 mL (range 600–6000 mL). All patients required at least one unit of red blood cells being the mean number of units 3.4 (range 1–6). The mean hospital stay was 11.7 days (range 6–41), and the mean stay in the intensive care unit was 1.8 days (0–12 days). Among the perioperative and post-operative complications, one patient (11%) pre- sented intestinal occlusion that resolved with non-opera- tive measures, one had massive bleeding during the surgery (11%), there was one case with urinary fistula (11%) and two cases showed a perineal infection (22%). One patient (11%) died at day 120 post-exenteration due to sepsis. This patient was one of the four with a patholog- ical complete response (Table 6). bilaterally. Page 4 of 10 (page number not for citation purposes) Survival The status of the operated patients is as follows: patient 1: Path CR, alive without disease at 62 months, patient 2: Path CR, alive without disease 59 months; patient 3: Residual of 2 cm, local and regional recurrence at 7 months post-exentera- tion, patient 4: Residual of 2 cm, alive without disease 52 months, patient 5: residual of 2 cm, local recurrence at 10 months, patient 6: residual of 2 cm, local recurrence at 7 months, patient 7: Path CR died at four months from sur- gical complications, patient 8: Path CR, died at 20 months from liver recurrence; patient 9: Residual of 8 cm, alive without disease 13 months. Thus, four of the nine oper- Chemotherapy was well tolerated. The most common side effect were nausea/vomiting grade 1 and 2, mild to mod- erate anemia was present in half of patients; all patients presented leukopenia and neutropenia which were grade 3 in five and three patients respectively. There were no epi- sodes of infection or bleeding (Table 4). Page 4 of 10 (page number not for citation purposes) Page 4 of 10 (page number not for citation purposes) BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 Patients with complete pathological response post-chemotherapy Figure 1 Patients with complete pathological response post-chemotherapy. CT scans of 3 patients showing residual pelvic mass after chemotherapy. Images a,c,e show CT scans pre-treatment, and images b,d,f are the post-treatment control studies. Notably, a patient (images e-f) shows a residual post-chemotherapy pelvic mass measuring 9 × 5 cm. Patients with complete pathological response post chemotherapy Figure 1 Patients with complete pathological response post-chemotherapy. CT scans of 3 patients showing residual pelvic mass after chemotherapy. Images a,c,e show CT scans pre-treatment, and images b,d,f are the post-treatment control studies. Notably, a patient (images e-f) shows a residual post-chemotherapy pelvic mass measuring 9 × 5 cm. Table 4: Toxicity to chemotherapy. (expressed by patient). Toxicity Grades (%) 0 1 2 3 4 Nausea/Vomiting 0 10 6 1 0 Diarrhea 13 2 2 0 0 Neuropathy 15 1 1 0 0 Anemia 6 4 4 3 0 Leukopenia 0 3 9 5 0 Granulocytopenia 0 6 8 3 0 Thrombopenia 12 2 0 3 0 Table 4: Toxicity to chemotherapy. (expressed by patient). ated patients are alive without disease. Survival Median survival in the intention to treat was 11 months, being 3 versus 32 in the non-operated versus those that underwent exentera- tion (Figures 2 and 3). http://www.biomedcentral.com/1471-2407/5/118 Table 5: Surgical data and pathological response Exenterated 9 Total infraelevator 8 Anterior supraelevator 1 Pathological Response Complete 4 Partial 5 ≤ 2 cm residual 4 8 cm residual 1 Surgical margins Negative 8 Positive 1 Ileocolonic conduit 7 Ileal conduit 2 Colostomy 8 Table 6: Surgical morbidity Mean Range Surgical time 6.3 4.3–8 Bleeding 1860 mL 600–1600 Units transfused 3.4 1–6 Hospital stay 11-7 6–41 Intensive Care Unit stay 1.8 0–12 Complication (events) Intestinal occlusion 1 (11%) Massive bleeding 1 (11%) Urinary fistula 1 (11%) Perineal infection 2 (22%) Death 1 (11%)* The patient died (fistula and infection) at 4 months post-exenteration. This patient had pathological complete response. Table 5: Surgical data and pathological response Table 5: Surgical data and pathological response Exenterated 9 Total infraelevator 8 Anterior supraelevator 1 Pathological Response Complete 4 Partial 5 ≤ 2 cm residual 4 8 cm residual 1 Surgical margins Negative 8 Positive 1 Ileocolonic conduit 7 Ileal conduit 2 Colostomy 8 The clinical characteristics of the patients included in this study are remarkable in the sense that all of them were considered no suitable for pelvic exenteration according to standard criteria by the team of gynecologists of our Institution. This special subgroup of patients with recur- rent disease is better defined if we look at their clinical characteristics: 13 of them relapsed at a median time of only 16 months whereas four were refractory to primary treatment and progressed within two to seven months All of them complained of pelvic pain, five had unilateral leg edema, three presented hydronephrosis and three cases had both signs. Both the short disease-free interval and the presence of one or more of the typical triad of signs and symptoms are either contraindications or factors pre- dicting a very poor outcome after exenteration in most of the reported series [15-20]. Selecting the true candidates for pelvic exenteration is a difficult clinical dilemma in patients with recurrent cervi- cal cancer after radiation therapy. Despite very thorough preoperative investigation, inoperable disease is discov- ered at the time of laparotomy in up to 50% of cases [21]. CT scanning is still one of the most extensively used diagnostic tool, however it may be difficult to differentiate recurrence from postoperative and post-radiation fibrosis [22,23]. http://www.biomedcentral.com/1471-2407/5/118 MRI has been regarded superior to CT scan in vis- ualization of the tumor and parametrial invasion in pri- mary tumors [24]; dynamic contrast-enhanced subtraction MRI may differentiate between recurrent tumor and benign conditions [25]. However, when MRI has been used for determining surgical elegibility for pel- vic exenteration its accuracy has been of 83% [26]. The difficulties encountered by the common imaging meth- ods for evaluating the extent of disease, such as CT scan and MRI [27], have led to propose laparoscopy to select candidates to undergo the procedure [28,29], which proved to be effective as it may spare unnecessary laparot- omy in half of the candidates patients [28]. Table 6: Surgical morbidity Mean Range Surgical time 6.3 4.3–8 Bleeding 1860 mL 600–1600 Units transfused 3.4 1–6 Hospital stay 11-7 6–41 Intensive Care Unit stay 1.8 0–12 Complication (events) Intestinal occlusion 1 (11%) Massive bleeding 1 (11%) Urinary fistula 1 (11%) Perineal infection 2 (22%) Death 1 (11%)* The patient died (fistula and infection) at 4 months post-exenteration. This patient had pathological complete response. The patient died (fistula and infection) at 4 months post-exenteration. This patient had pathological complete response. tional support, and routine postoperative monitoring, have reduced the morbidity from pelvic exenteration [14]. tional support, and routine postoperative monitoring, have reduced the morbidity from pelvic exenteration [14]. It must be stressed, however, that all these imaging and laparoscopy efforts to predict resectability and avoid aborted exenterations are done in the setting of "classic indications" of pelvic exenterations, where the ultimate goals are increasing the efficacy of the procedure in terms of disease control and decreased morbidity and mortality. However, our approach in aimed at increasing the propor- tion of patients in whom this salvage therapy could be attempted under the rationale that systemic chemother- apy would allow obtaining free surgical margins in situa- tions where the extent of pelvic disease predicts that negative surgical margins would unlikely be obtained and therefore exenteration could not be offered to these patients. In order to increase the proportion of patients in whom this salvage therapy could be attempted, we developed the modality of "pre-exenterative chemotherapy" under the rationale that systemic chemotherapy would allow the obtaining free surgical margins in patients undergoing the "standard" supra or infraelevator pelvic exenteration operation in situations where the extent of pelvic disease predicts that negative surgical margins would unlikely be obtained. Page 6 of 10 (page number not for citation purposes) Discussion Although pelvic exenteration plays a definitive role in the management of recurrent cervical carcinoma, its impact in terms of the proportion of cervical cancer patients who benefit from such radical procedure has remained unchanged because it continues to be indicated in only very selected patients with small central pelvic recur- rences. This fact, along with better medical support such as routine use of prophylactic heparin, antibiotics, nutri- Page 5 of 10 (page number not for citation purposes) Page 5 of 10 (page number not for citation purposes) BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 http://www.biomedcentral.com/1471-2407/5/118 The results of this pilot study demonstrate the feasibility of this approach, as nine (53%) out of the 17 patients included in this trial underwent pelvic exentera- tion obtaining disease-free margins in all but one case; four of them are alive without disease. Page 6 of 10 (page number not for citation purposes) BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 Overall survival in the intention to treat; four out of 17 are alive for a median survival of 11 months Figure 2 Overall survival in the intention to treat; four out of 17 are alive for a median survival of 11 months. http://www.biomedcentral.com/1471-2407/5/118 BMC Cancer 2005, 5:118 Page 8 of 10 (page number not for citation purposes) This is shown in figure 1, standing out patient 3 (Figure 1e–f) in whom the residual mass after Page 7 of 10 (page number not for citation purposes) Page 7 of 10 (page number not for citation purposes) http://www.biomedcentral.com/1471-2407/5/118 http://www.biomedcentral.com/1471-2407/5/118 BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 Survival in the operated and non-operated patients; median survival was 3 versus 32 months respectively Figure 3 Survival in the operated and non-operated patients; median survival was 3 versus 32 months respectively. This difference is highly significant. BMC Cancer 2005, 5:118 Overall su Figure 2 Overall survival in the intention to treat; four out of 17 are alive for a median survival of 11 months Figure 2 Overall survival in the intention to treat; four out of 17 are alive for a median survival of 11 months. Overall survival in the intention to treat; four out of 17 are alive for a median survival of 11 month Figure 2 Overall survival in the intention to treat; four out of 17 are alive for a median survival of 11 month These considerations led us to rely on bimanual pelvic examination, which is a subjective test, as our principal criterion for deciding to perform the exenteration (as long as there was no regional or systemic disease evaluated by CT scan). We acknowledge that it would had been very valuable to have an objective pre and post-chemotherapy evaluation of the response by a CT scan, RMI, and/or PET scan in all the cases, however, only nine cases had CT scan pre and post therapy. Notwithstanding it is interesting to notice that although all our patients were "felt" by physi- cal exam to have side wall fixation, the CT scan confirmed it only in six cases based on the criterion of having a less than 3 mm separation of the tumor from the pelvic mus- cles and/or vascular encasement [30]. It is important also to notice that in the four cases that were not candidates for exenteration after chemotherapy and had pre and post- chemotherapy CT scans, there was no response in three and progression in one, matching closely the findings of bimanual pelvic examinations. In contrast, the five cases that underwent exenteration and were felt to have responded by physical examination, with the criteria used, had partial responses according to the standard WHO criteria suggesting that, after all, CT scan can be a reliable method for evaluating the response to chemother- apy in the setting of pelvic recurrences in a previously irra- diated site. Nevertheless, an important observation is the fact that in three out of the four cases that achieved a pathological complete response, the CT scan was clearly positive for the presence of tumor. Survival in Figure 3 Survival in the operated and non-operated patients; median survival was 3 versus 32 months respectively Figure 3 Survival in the operated and non-operated patients; median survival was 3 versus 32 months respectively. This difference is highly significant. Survival in the operated and non operated patients; median survival was 3 versus 32 months respectively Figure 3 Survival in the operated and non-operated patients; median survival was 3 versus 32 months respectively. This difference is highly significant. ed and non operated patients; median survival was 3 versus 32 months respectively ed and non-operated patients; median survival was 3 versus 32 months respectively. This difference is atin versus cisplatin paclitaxel, the response rate in the subgroup with pelvic disease revealed onjective responses in 14(21%) of 66 patients treated with cisplatin alone and in 17(33%) of 52 patients treated with the combination, however, median survival was the same, 8.8 months and 9.7 months, respectively [13]. The response reported here using a platinum-based scheme was 55% (partial responses) in the nine patients with pre and post-chemo- therapy evaluation by CT scan, however it was only 29% taking into account the 17 patients evaluated. This response rate as well as the observed toxicity, is within the range expected but no assumptions can be made on the efficacy of any of the schemes used. chemotherapy (1e) measured 9 × 5 cm. This finding might suggest that exenteration could be useful after any degree of response to chemotherapy because the actual response to chemotherapy could be of greater magnitude than pre- dicted by CT scan. The use of chemotherapy in the palliative setting of per- sistent or recurrent pelvic disease, particularly in a patient who has received definitive radiation or chemoradiation treatment has very limited value. In a review on results of 190 advanced or recurrent disease patients treated with 14 different chemotherapy protocols, the overall response rate was 20.0% (4.2% complete response; 15.8% partial response), with a median response duration of 4.8 months [31]. In a recent phase III study comparing cispl- Page 8 of 10 (page number not for citation purposes) Page 8 of 10 (page number not for citation purposes) BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 http://www.biomedcentral.com/1471-2407/5/118 So far there is information of the efficacy of chemotherapy in terms of pathological response in recurrent or advanced cervical cancer because chemotherapy is only used as a palliative measure and surgery is commonly not affered after chemotherapy. Authors' contributions R D-V, J U-D, JC and RB participated in data collection and analysis; LC, DC, C L-G, A G-E and A D-G managed the patients; MC, and J de la G critically read and partici- pated in the manuscript; additionally C L-G and A D-G conceived and wrote the manuscript. Survival in Figure 3 Here we demonstrate a pathological complete response rate of 44% in the nine patients treated (23.5% taking into the 17 patients), which is higher than obtained in neoadjuvant trials in locally advanced cervical cancer utilizing platinum-based schemes with newer drugs, such as gemcitabine [32,33], vinorelbine [34], pacl- itaxel [35,36], and irinotecan [37]. It is worthwhile notic- ing that the neoadjuvant trials with lower complete response rates were those that subjected more patients to surgery. The two studies with the lowest complete responses, 16% and 17% operated 89% and 95% of patients, respectively [35,36], whereas in the trial with 37.5% of complete response, the surgery rate was only 52% [34]. These data may explain our 44% of pathologi- cal complete response rates, since only 52.9% of our patients underwent surgery. References Miller B, Morris M, Rutledge F, Mitchell MF, Atkinson EN, Burke TW, Wharton JT: Aborted exenterative procedures in recurrent cervical cancer. Gynecol Oncol 1993, 50:94-99. y , 9. del Carmen MG, McIntyre JF, Goodman A: The role of intraoper- ative radiation therapy (IORT) in the treatment of locally advanced gynecologic malignancies. Oncologist 2000, 5:18-25. g g g g 10. Gemignani ML, Alektiar KM, Leitao M, Mychalczak B, Chi D, Venkat- raman E, Barakat RR, Curtin JP: Radical surgical resection and high-dose intraoperative radiation therapy (HDR-IORT) in patients with recurrent gynecologic cancers. Int J Radiat Oncol Biol Phys 2001, 50:687-694. y 11. Hockel M: Laterally extended endopelvic resection: Surgical treatment of infrailiac pelvic wall recurrences of gynecology malignancies. Am J Obstet Gynecol 1999, 180:306-312. g J y 12. Hockel M: Laterally extended endopelvic resection. Novel surgical treatment of locally recurrent cervical carcinoma involving the pelvic side wall. Gynecol Oncol 2003, 9:369-377. g p y 13. Moore DH, Blessing JA, McQuellon RP, Haler HT, Cella D, Benda J, Miller DS, Olt G, King S, Boggess JF, Rocereto TF: Phase III study of cisplatin with or without paclitaxel in stage IVB, recur- rent, or persistent squamous cell carcinoma of the cervix: a gynecologic oncology group study. J Clin Oncol 2004, 22:3113-3119. 14. Goldberg JM, Piver MS, Hempling RE, Aiduk C, Blumenson L, Recio FO: Improvements in pelvic exenteration: factors responsi- ble for reducing morbidity and mortality. Ann Surg Oncol 1998, 5:399-406. Competing interests The author(s) declare that they have no competing interests. References 1. Parkin DM, Bray F, Ferlay J, Pisani P: Estimating the world cancer burden: Globocan 2000. Int J Cancer 2001, 4:153-156. 2. Dueñas-Gonzalez A, Cetina L, Mariscal I, de la Garza J: Modern management of locally advanced cervical carcinoma. Cancer Treat Rev 2003, 29:389-399. 3. Lopez MJ, Spratt JS: Exenterative pelvic surgery. J Surg Oncol 1999, 72:102-114. 3. Lopez MJ, Spratt JS: Exenterative pelvic surgery. J Surg Oncol 1999, 72:102-114. 4. Hockel M, Knapstein P: The combined operative and radiother- apeutic tratment of recurrent tumors infiltrating the pelvic wall. Gynecol Oncol 1992, 46:20-28. 5. Leitao MM, Chi DS: Recurrent cervical cancer. Curr Treat Options Oncol 2002, 3:105-111. 4. Hockel M, Knapstein P: The combined operative and radiother- apeutic tratment of recurrent tumors infiltrating the pelvic wall. Gynecol Oncol 1992, 46:20-28. A noticeable finding of the present report is that the median survival of 11 months compares favorably with studies using systemic chemotherapy in the palliative set- ting, ranging from 6 to 10 months [38,39]; however, we must stress that half of patients (the non-operated) had a median survival of only 3 months which suggest that the patient population of patients had indeed very unfavora- ble clinical characteristics. Of outmost importance is the fact that from the operated patients four achieved patho- logical complete response despite having tomographic evidence of residual tumor and the median survival for these 9 patients taken to exenterative surgery was 32 months. The fact that there was "gross persistent" disease in the CT scan after chemotherapy support our view that the overestimation of pelvic disease, either by pelvic examination and/or imaging methods, hinder offering a potentially curative surgery to a huge proportion of patients with pelvic recurrence of cervical carcinoma. On these bases, we are just to start a randomized study com- paring pre-exenterative chemotherapy and exenteration versus palliative chemotherapy alone in patients with pel- vic disease that do not meet the criteria for pelvic exenteration. y 5. Leitao MM, Chi DS: Recurrent cervical cancer. Curr Treat Options Oncol 2002, 3:105-111. 6. Friedlander M: Guidelines for the treatment of recurrent and metastatic cervical cancer. Oncologist 2002, 7:342-347. g 7. Brunschwig A: Complete excision of pelvic viscera for advanced carcinoma. Cancer 1948, 1:177-183. 8. Miller B, Morris M, Rutledge F, Mitchell MF, Atkinson EN, Burke TW, Wh t JT Ab t d t ti d i t 8. Acknowledgements We thank Elizabeth Robles for her support in the execution of the study. Pre-publication history 20. Torres Lobaton A, Bastida Blanco A, Marquez Acosta G, Hernandez Aten D, Roman Bassaure E, Rojo Herrera G: Pelvic exenteration for cancer of the uterine cervix (prognostic factors). Ginecol Obstet Mex 1994, 62:189-193. p y The pre-publication history for this paper can be accessed here: 21. Haas T, Buchsbaum HJ, Lifshitz S: Nonresectable recurrent pelvic neoplasm. Outcome in patients explored for pelvic exenteration. Gynecol Oncol 1980, 9:177-181. http://www.biomedcentral.com/1471-2407/5/118/pre pub y 22. Halpin TF, Frick HC, Munnell EW: Critical points of failure in the therapy of cancer of the cervix: a reappraisal. Am J Obstet Gynecol 1972, 114:755-764. y 23. Walsh JW, Amendola MA, Hall DJ, Tisnado J, Goplerud DR: Recur- rent carcinoma of the cervix: CT diagnosis. AJR 1981, 136:117-1222. 24. Kim SH, Choi BI, Lee HP, Kang SB, Choi YM, Han MC, Kim CW: Uterine cervical carcinoma; comparison of CT and MR findings. Radiology 1990, 175:45-51. 25. Kinkel K, Ariche M, Tardivon AA, Spatz A, Castaigne D, Lhomme C, Vanel D: Differentiation between recurrent tumor and benign conditions after treatment of gynecologic pelvic carcinoma: value of dynamic contrast-enhanced subtraction MR imaging. Radiology 1997, 204:55-63. g g gy 26. Popovich MJ, Hricak H, Sugimura K, Stern JL: The role of MR imagin in determining surgical elegibility for pelvic exenteration. AJR Am J Roentgenol 1993, 160:525-31. J J g 27. Jeong YY, Kang HK, Chung TW, Seo JJ, Park JG: Uterine cervical carcinoma after therapy: CT and MR imaging findings. Radi- ographics 2003, 23:969-981. g p 28. Zeisler H, Joura EA, Moeschl P, Maier U, Koelbl H: Preoperative evaluation of tumor extension in patients with recurrent cervical cancer. Acta Obstet Gynecol Scand 1997, 76:474-477. y 29. Kohler C, Tozzi R, Possover M, Schneider A: Explorative laparos- copy prior to exenterative surgery. Gynecol Oncol 2002, 86:311-315. 30. Pannu HK, Corl FM, Fishman EK: Evaluation of cervical cancer: Spectrum of disease. Radiographics 2001, 21:1155-1168. 31. Brader KR, Morris M, Levenback C, Levy L, Lucas KR, Gershenson DM: Chemotherapy for cervical carcinoma: Factors deter- mining response and implications for clinical trial design. J Clin Oncol 1998, 16:1879-1884. 32. Dueñas-Gonzalez A, Gonzalez EA, Lopez-Graniel C, Reyes M, Mota A, Munoz D, Solorza G, Hinojosa LM, Guadarrama R, Florentino R, Mohar A, Melendez J, Maldonado V, Chanona J, Robles E, de la Garza J: A phase II study of gemcitabine and cisplatin combination as induction chemotherapy for untreated locally advanced cervical carcinoma. Pre-publication history Ann Oncol 2001, 12:541-547. 33. Duenas-Gonzalez A, Lopez-Graniel C, Gonzalez A, Gomez E, Rivera L, Mohar A, Chanona G, Trejo-Becerril C, de la Garza J: Induction chemotherapy with gemcitabine and oxiplatin for locally advanced cervical carcinoma. Am J Clin Oncol 2003, 26:22-25. J 34. Pignata S, Silvestro G, Ferrari E, Selvaggi L, Perrone F, Maffeo A, Frezza P, Di Vagno G, Casella G, Ricchi P, Cormio G, Gallo C, Iodice F, Romeo F, Fiorentino R, Fortuna G, Tramontana S: Phase II study of cisplatin and vinorelbine as first-line chemotherapy in patients with carcinoma of the uterine cervix. J Clin Oncol 1999, 17:756-760. 35. Zanetta G, Lissoni A, Pellegrino A, Sessa C, Colombo N, Gueli-Alletti D, Mangioni C: Neoadjuvant chemotherapy with cisplatin, ifos- famide and paclitaxel for locally advanced squamous-cell cervical cancer. Ann Oncol 1998, 9:997-980. 36. Duenas Gonzalez A, Lopez-Graniel C, Gonzalez Enciso A, Cetina L, Rivera L, Mariscal I, Montalvo G, Gomez E, de la Garza J, Chanona G, Mohar A: A phase II study of multimodality treatment for locally advanced cervical cancer: neoadjuvant carboplatin and paclitaxel followed by radical hysterectomy and adju- vant cisplatin chemoradiation. Ann Oncol 2003, 14:1278-84. Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 10 of 10 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge 37. Conclusion 15. Rutledge FN, Smith JP, Wharton JT, O'Quinn AG: Pelvic exentera- tion: analysis of 296 patients. Am J Obstet Gynecol 1977, 129:881-892. The therapeutic modality here reported, called pre- exenterative chemotherapy, is a therapeutic alternative for cervical cancer patients with recurrent or persistent disease limited to the pelvis not usually considered candidates for "classical" pelvic exenteration. Its value in the manage- ment of recurrent disease needs to be confirmed in a ran- domized phase III study. 16. Symmonds RE, Pratt JH, Webb MJ: Exenterative operations: experience with 198 patients. Am J Obstet Gynecol 1975, 121:907-918. 17. Stanhope CR, Symmonds RE: Palliative exenteration – what, when and why? Am J Obstet Gynecol 1985, 152:12-16. y J y 18. Matthews CM, Morris M, Burke TW, Gershenson DM, Wharton JT, Rutledge FN: Pelvic exenteration in the elderly patient. Obstet Gynecol 1992, 79:773-777. y 19. Estape R, Angioli R: Surgical management of advanced and recurrent cervical cancer. Semin Surg Oncol 1999, 16:236-241. Page 9 of 10 (page number not for citation purposes) Page 9 of 10 (page number not for citation purposes) http://www.biomedcentral.com/1471-2407/5/118 BMC Cancer 2005, 5:118 http://www.biomedcentral.com/1471-2407/5/118 http://www.biomedcentral.com/1471-2407/5/118 Pre-publication history Sugiyama T, Nishida T, Kumagai S, Fujiyoshi K, Okura N, Yakushiji M, Umesaki N: Combination therapy with irinotecan and cispla- tin as neoadjuvant chemotherapy in locally advanced cervi- cal cancer. Br J Cancer 1999, 81:95-98. J 38. Hogg R, Friedlander M: Role of systemic chemotherapy in met- astatic cervical cancer. Expert Rev Anticancer Ther 2003, 3:234-240. 39. Tambaro R, Scambia G, Di Maio M, Pisano C, Barletta E, Iaffaioli VR, Pignata S: The role of chemotherapy in locally advanced, met- astatic and recurrent cervical cancer. Crit Rev Oncol Hematol 2004, 52:33-44.
https://openalex.org/W4362538491	https://aacr.figshare.com/articles/journal_contribution/Supplementary_Table_2_from_Fibroblast_Growth_Factor-2_Is_an_Important_Factor_that_Maintains_Cellular_Immaturity_and_Contributes_to_Aggressiveness_of_Osteosarcoma/22517212/1/files/39979807.pdf	English	null	Supplementary Table 2 from Fibroblast Growth Factor-2 Is an Important Factor that Maintains Cellular Immaturity and Contributes to Aggressiveness of Osteosarcoma	null	2,023	cc-by	426	Supplementary Table 2. Selected genes of liquid factors expressed in AX cells sorted from tumors. N.D. : not hybridized probably because of the extremely low expression Gene Symbol Accession Number Normalized Value Gene Symbol Accession Number Normalized Value Tgfa NM_031199 N.D. Bmp6 NM_007556 91.0 Tgfb3 NM_009368 51.0 Bmp10 NM_009756 40.6 Tgfb1 NM_011577 171.6 Bmp2 NM_007553 22.9 Ctgf NM_010217 78.3 Bmp15 NM_009757 N.D. Tgfb2 NM_009367 138.8 Bmp2k NM_080708 17.9 Ecgf1 NM_138302 142.1 Bmp5 NM_007555 24.7 Tdgf1 NM_011562 16.4 Bmp1 NM_009755 69.8 Igf2 NM_010514.3 5.0 Bmp3 NM_173404 18.4 Fgf1 NM_010197 3.5 Il13 NM_008355 14.2 Pdgfa NM_008808.3 628.5 Il17d NM_145837 16.6 Pgf NM_008827 25.6 Il21 NM_021782 4.5 Ngfb NM_013609 5.9 Il31ra NM_139299 3.0 Fgf2 NM_008006 9.4 Nfil3 NM_017373 398.8 Igf1 NM_010512;NM_184052 9.1 Il17a NM_010552 36.9 Egf NM_010113 13.0 Il19 XM_283649 1.7 Pdgfc NM_019971 316.4 Il9 NM_008373 3.6 Fgf2 NM_008006 1.4 Il27 NM_145636 6.5 Vegfa NM_009505 71.1 Il12b NM_008352 570.5 Vegfb NM_011697 727.6 Il3 NM_010556 128.0 Hdgf NM_008231 1855.5 Il16 NM_010551 3.1 Hgf NM_010427 6.2 Il6 NM_031168 4.0 Pdgfd NM_027924 67.7 Sfrp5 NM_018780 2.9 Lif NM_008501 4.5 Sfrp1 NM_013834 18.2 Bmp8a NM_007558 40.1 Sfrp2 NM_009144.2 378.7 Bmp7 NM_007557 12.0 Sfrp4 NM_016687 43.1 Bmp8b NM_007559 2.7 Tnf NM_013693 N.D. Bmp4 NM_007554 439.5 Osm XM_137493 21.3 Supplementary Table 2. Selected genes of liquid factors expressed in AX cells sorted from tumors. N.D. : not hybridized probably because of the extremely low expression Gene Symbol Accession Number Normalized Value Gene Symbol Accession Number Normalized Value Tgfa NM_031199 N.D. Bmp6 NM_007556 91.0 Tgfb3 NM_009368 51.0 Bmp10 NM_009756 40.6 Tgfb1 NM_011577 171.6 Bmp2 NM_007553 22.9 Ctgf NM_010217 78.3 Bmp15 NM_009757 N.D. Tgfb2 NM_009367 138.8 Bmp2k NM_080708 17.9 Ecgf1 NM_138302 142.1 Bmp5 NM_007555 24.7 Tdgf1 NM_011562 16.4 Bmp1 NM_009755 69.8 Igf2 NM_010514.3 5.0 Bmp3 NM_173404 18.4 Fgf1 NM_010197 3.5 Il13 NM_008355 14.2 Pdgfa NM_008808.3 628.5 Il17d NM_145837 16.6 Pgf NM_008827 25.6 Il21 NM_021782 4.5 Ngfb NM_013609 5.9 Il31ra NM_139299 3.0 Fgf2 NM_008006 9.4 Nfil3 NM_017373 398.8 Igf1 NM_010512;NM_184052 9.1 Il17a NM_010552 36.9 Egf NM_010113 13.0 Il19 XM_283649 1.7 Pdgfc NM_019971 316.4 Il9 NM_008373 3.6 Fgf2 NM_008006 1.4 Il27 NM_145636 6.5 Vegfa NM_009505 71.1 Il12b NM_008352 570.5 Vegfb NM_011697 727.6 Il3 NM_010556 128.0 Hdgf NM_008231 1855.5 Il16 NM_010551 3.1 Hgf NM_010427 6.2 Il6 NM_031168 4.0 Pdgfd NM_027924 67.7 Sfrp5 NM_018780 2.9 Lif NM_008501 4.5 Sfrp1 NM_013834 18.2 Bmp8a NM_007558 40.1 Sfrp2 NM_009144.2 378.7 Bmp7 NM_007557 12.0 Sfrp4 NM_016687 43.1 Bmp8b NM_007559 2.7 Tnf NM_013693 N.D. Bmp4 NM_007554 439.5 Osm XM_137493 21.3 Supplementary Table 2. Selected genes of liquid factors expressed in AX cells sorted from tumors. N.D. : not hybridized probably because of the extremely low expression
https://openalex.org/W1544465843	https://www.intechopen.com/citation-pdf-url/42130	English	null	Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester	InTech eBooks	2,013	cc-by	9,347	1. Introduction Cancer is a multistage disease involving a series of events and generally occurs over an extended period. During this period, accumulation of genetic and epigenetic alterations leads to the progressive transformation of a normal cell into a malignant cell. Cancer cells acquire several abilities that most healthy cells do not possess: they become resistant to growth inhibition, proliferate without dependence on growth factors, replicate without limit, evade apoptosis, and invade, metastasize and support angiogenesis [1]. Unlike heart disease, death rates for cancer remained approximately the same in the United States from 1975 through 2002. Indeed, it is predicted that by 2020 approximately 15 million new cancer cases will be diagnosed worldwide and 12 million cancer patients will die [2]. © 2013 El-Refaei and Mady, licensee InTech. This is an open access chapter di the Creative Commons Attribution License (http://creativecommons.org/license unrestricted use, distribution, and reproduction in any medium, provided the o cases will be diagnosed worldwide and 12 million cancer patients will die Cancer is a disease characterized by uncontrolled growth and division o cells and its emergence requires several elements, including self-suf signals, insensitivity to growth-inhibitory signals, evasion of apoptosis, potential, tissue invasion and metastasis, and sustained angiogenesi thought to evolve along a multi-step process. Cancer cells are the desce cell in which some kind of internal or external stress causes a change in it event is said to initiate the cell to a precancerous state. In a second stage cell divides in response to a promoting agent to produce daughter cells, cells divide to produce more daughter cells, and so on. The genetic instab through the generations finally result in one cell that no longer requires t to stimulate its proliferation. A cancer cell is thus born with the ability to as growth factors that stimulate proliferation. Finally in the third stage progression, this cancer cell divides to produce daughter cells, which al there is a population of cancer cells with the ability to invade and metasta Cancer is a disease characterized by uncontrolled growth and division of genetically altered cells and its emergence requires several elements, including self-sufficiency in growth signals, insensitivity to growth-inhibitory signals, evasion of apoptosis, limitless replicative potential, tissue invasion and metastasis, and sustained angiogenesis [1, 3]. Cancer is thought to evolve along a multi-step process. http://dx.doi.org/10.5772/55275 http://dx.doi.org/10.5772/55275 © 2013 El-Refaei and Mady, licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Selection of our books indexed in the Book Citation Index in Web of Science™ Core Collection (BKCI) Interested in publishing with us? Contact book.department@intechopen.com Numbers displayed above are based on latest data collected. For more information visit www.intechopen.com Open access books available Countries delivered to Contributors from top 500 universities International authors and editors Our authors are among the most cited scientists Downloads We are IntechOpen, the world’s leading publisher of Open Access books Built by scientists, for scientists 12.2% 191,000 210M TOP 1% 154 7,200 Chapter 14 Additional information is available at the end of the chapter http://dx.doi.org/10.5772/55275 Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester Mohamed F. El-Refaei and Essam A. Mady the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 1. Introduction Cancer cells are the descendants of a normal cell in which some kind of internal or external stress causes a change in its genetic code. This event is said to initiate the cell to a precancerous state. In a second stage, this precancerous cell divides in response to a promoting agent to produce daughter cells, and these daughter cells divide to produce more daughter cells, and so on. The genetic instabilities passed down through the generations finally result in one cell that no longer requires the promoting agent to stimulate its proliferation. A cancer cell is thus born with the ability to make proteins such as growth factors that stimulate proliferation. Finally in the third stage of carcinogenesis, progression, this cancer cell divides to produce daughter cells, which also divide, and soon there is a population of cancer cells with the ability to invade and metastasize [4]. 298 Carcinogenesis It is now clear that cancer phenotypes result from the dysregulation of more than 500 genes at multiple steps in cell signaling pathways. This indicates that inhibition of a single gene product or cell signaling pathway is unlikely to prevent or treat cancer. However, most current anticancer therapies are based on the modulation of a single target [5, 6]. One of the most important findings to have emerged during the past three decades is that cancer is a largely preventable disease. Thus, people need to be educated about the risk factors for cancer and those that prevent the disease. As many as 90% of all cancers have been shown to be due to environmental/acquired factors such as tobacco, diet, radiation and infectious organisms, etc., and only the remaining 5–10% of cases are caused by internal factors such as inherited mutations, hormones, and immune conditions [7]. The ineffective, unsafe, and expensive monotargeted therapies have led to a lack of faith in these approaches. Therefore, the current paradigm for cancer treatment is either to combine several monotargeted drugs or to design drugs that modulate multiple targets. As a result, pharmaceutical companies have been increasingly interested in developing multitargeted therapies. Many plant-derived dietary agents, called nutraceuticals, have multitargeting properties. In addition, these products are less expensive, safer, and more readily available than are synthetic agents [5]. Prevention is better than cure and this is very true in case of cancer. 1. Introduction Chemoprevention was defined as the administration of agents to prevent induction, to inhibit or to delay the progression of cancer [8], or as the inhibition or reversal of carcinogenesis at a premalignant stage [9]. Chemoprevention involves the use of synthetic or natural compounds to inhibit, slow, or reverse carcinogenesis. It is based on the hypothesis that the disruption of biological events involved in carcinogenesis will inhibit this process and can be applied to any stage of carcinogenesis. Chemoprevention utilizes appropriate pharmacological agents [10,11] or dietary agents, consumed in diverse forms like macronutrients, micronutrients, or nonnutritive phytochemicals [12–14]. It is estimated that from 10 to 80 percent of cancer patients use some form of natural compounds as a part of complementary medicine as part of their overall therapy without any real guidance. This explains the growing interest in using the natural compounds properly in the treatment of cancer. Phytochemicals are one wide class of nutraceuticals found in plants, which are extensively researched by scientists for their health-promoting potential. Honey has a wide range of phytochemicals including polyphenols which act as antioxidants. Polyphenols and phenolic acids found in the honey vary according to the geographical and climatic conditions. Some of them were reported as a specific marker for the botanical origin of the honey. Considerable differences in both composition and content of phenolic compounds have been found in different unifloral honeys [15]. Terpenes, benzyl alcohol, 3, 5-dimethoxy-4- hydroxybenzoic acid (syringic acid), methyl 3, 5-dimethoxy-4-hydroxybenzoate (methyl syringate), 3, 4, 5-trimethoxybenzoic acid, 2-hydroxy-3- phenylpropionic acid, 2- hydroxybenzoic acid and 1, 4- dihydroxybenzene are some of the phytochemicals a ascribed for the antimicrobial activity of honey [16]. Among these phytochemicals, polyphenols were reported to have antiproliferative potential. Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 299 Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 2.1. Regulation of inflammatory pathways by CAPE Inflammation is a localized reaction of tissue to infection, irritation, or other injury. Inflammation is a necessary response to clear bacterial and viral infections, repair tissue insults, and suppress tumor initiation/progression. However, when inflammation persists or control mechanisms are dysregulated, diseases such as cancer can develop. Interestingly, inflammation functions at all stages of tumor development: initiation, promotion and progression including metastasis. During the initiation phase, inflammation induces the release of a variety of cytokines and chemokines that promote the activation of inflammatory cells and associated factors. This causes further oxidative damage, DNA mutations, and other changes in the tissue microenvironment, making it more conducive to cell transformation, increased survival, and proliferation [24]. At the molecular level, inflammation, transformation, survival and proliferation are regulated by the proinflammatory transcription factor Nuclear Factor-κB (NF-κB), a family of ubiquitously expressed transcription factors. NF-κB regulates the expression of genes involved in the transformation, survival, proliferation, invasion, angiogenesis and metastasis of tumor cells [25]. TNF-α is also one of the prime signals that induces apoptosis in many different types of cells. Whereas acute activation of NF-κB may be therapeutic, chronic activation may lead to the development of chronic inflammation, cancer and other chronic diseases. There is a strong association between chronic inflammatory conditions and cancer specific to the organ. Epidemiological evidence points to a connection between inflammation and a predisposition for the development of cancer, i.e., long-term inflammation leads to the development of dysplasia. Various factors are known to induce chronic inflammatory responses that further cause cancer. These include bacterial, viral, and parasitic infections (e.g., Helicobacter pylori, Epstein-Barr virus, human immunodeficiency virus, flukes, schistosomes) and chemical irritants (i.e., tumor promoters). Active NF-κB has now been identified in tissues of most cancer patients, including those with leukemia and lymphoma and cancers of the prostate, breast, oral cavity, liver, pancreas, colon and ovary [26]. In the resting stage, NF-κB resides in the cytoplasm as a heterotrimer consisting of p50, p65, and the inhibitory subunit IκBα. On activation, the IκBα protein undergoes phosphorylation, ubiquitination, and degradation. p50 and p65 are then released, are translocated to the nucleus, bind specific DNA sequences present in the promoters of various genes, and initiate their transcription. A number of proteins are involved in the NF- κB signaling pathway. Because of the relevance of the NF-κB signaling pathway in cancer, this pathway has been proven to be an attractive target for therapeutic development. 2. Active compounds in propolis Polyphenolic compounds are widely distributed in the plant kingdom and display a variety of biological activities, including chemoprevention and tumor growth inhibition. Propolis and honey have been known to mankind from the remotest of ancient times and have been widely used by many cultures for different purposes. Propolis is a complex resinous mixture gathered from plants and used by honeybees in their hives as a general-purpose sealer and antibiotic. It is made up of a variety of polyphenolic compounds. Some of the isolated compounds have shown anti-inflammatory activity, carcinostatic, anti-carcinogenic activity and induction of apoptosis. Caffeic acid (CA) and caffeic acid phenethyl ester (CAPE) are members of the polyphenolic compounds and present in high concentrations in medicinal plants and propolis. CAPE showed a wide variety of biological activities at non- toxic concentrations. It has shown antibacterial, anti-inflammatory, antioxidant, antitumor and anti-proliferative activities [17]. CAPE [2-propenoic acid, 3-(3,4-dihydroxy phenyl)-,2-phenethyl ester] (Fig.1) is an active component of propolis with a variety of biological activities. CAPE has been used in folk medicine as a potent antibacterial, anti-inflammatory, antioxidant, antitumor and antiproliferative with a wide variety of biological and pharmacological activities at non- toxic concentrations in a mammal's organs [18]. Figure 1. Structure of caffeic acid phenethyl ester (CAPE) Figure 1. Structure of caffeic acid phenethyl ester (CAPE) CAPE is chemopreventive against intestinal, colon and skin cancer, and also has been shown to decreases the formation of preneoplastic hepatic lesions when is administrated in a rat model of liver carcinogenesis [19-21], but the mechanism of these properties is not completely understood. Recently, CAPE, in a concentration dependent fashion, was shown to inhibit MCF-7 (hormone receptor positive, HR+) and MDA-MB-231 (a model of triple negative BC (TNBC)) tumor growth, either in vitro or in vivo without much effect on normal mammary cells [22]. At the same time, CAPE was found to cause pronounced changes in bCSC characteristics manifested by inhibition of self renewal, progenitor formation, clonal growth in soft agar, and concurrent significant decrease in CD44 content, all signs of decreased malignancy potential [23]. Besides CAPE, other caffeic acid esters in propolis may have biological effects. 300 Carcinogenesis Here we will focus on CAPE and its biological effects against cancer in vitro and in vivo. Specifically, we will discuss how CAPE can modulate inflammatory pathways and thus affect the survival, proliferation, invasion, angiogenesis, and metastasis of the tumor. 2.1. Regulation of inflammatory pathways by CAPE Active NF-κB complexes can contribute to tumorigenesis by regulating genes that promote the growth and survival of cancer cells, during the cell cycle. NF-κB has ability to regulate 301 Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 30 the G1-phase expression of key proto-oncogenes is subject to regulation by the integrated activity of IkappaB kinase (IKK)alpha, IKKbeta, Akt and Chk1. The coordinated binding of NF-κB subunits to the Cyclin D1, c-Myc and Skp2 promoters is dynamic with distinct changes in promoter occupancy and RelA(p65) phosphorylation occurring through G1, S and G2 phases, concomitant with a switch from coactivator to corepressor recruitment. Akt activity is required for IKK-dependent phosphorylation of NF-κB subunits in G1 and G2 phases, where Chk1 is inactive. However, in S-phase, Akt is inactivated, while Chk1 phosphorylates RelA and associates with IKKalpha, inhibiting the processing of the p100 (NF-κB2) subunit, which also plays a critical role in the regulation of these genes. This reveals a complex regulatory network integrating NF- κB with the DNA-replication checkpoint and the expression of critical regulators of cell proliferation [27]. Thus, its inhibition could be a novel approach to breaking the vicious cycle of tumor cell proliferation [28-30] More than 700 inhibitors of the NF-κB activation pathway have been reported, including antioxidants, peptides, small RNA/DNA, microbial and viral proteins, small molecules, and engineered dominant-negative or constitutively active polypeptides [31,32]. Caffeic acid phenethyl ester has been shown to suppress NF-κB activation by suppressing the binding of the p50–p65 complex directly to DNA [33-35]. The molecular basis of CAPE action was elucidated by Natarajan et al. [36]. Since NF-κB has a role in these activities, they examined the effect of CAPE on this transcription factor in an exhaustive manner. They preincubated the U-937 cells with CAPE at various concentrations for 2 hours before treating with TNF (0.1 nM) for 15 minutes. CAPE inhibited the TNF-dependent activation of NF-κB in a dose-dependent manner with maximum effect occurring at 25 μg/mL. NF- κB activation induced by the phorbol ester, phorbol-12- myristate 13-acetate (PMA), ceramide, okadaic acid and hydrogen peroxide was also inhibited by CAPE. It prevented the translocation of the p65 subunit of NF-κB to the nucleus without affecting the TNF-induced IκBα degradation. It did not show any inhibitory effects on the other transcription factors like AP-1, TFIID and oct-1. 2.1. Regulation of inflammatory pathways by CAPE With these findings they concluded that CAPE is a potent and a specific inhibitor of NF-κB activation and this may provide the molecular basis for its multiple immunomodulatory and antiinflammatory activities of CAPE. Abdel-Latif et al.[37] have demonstrated for the first time that CAPE is a major component of propolis, modulates H. pylori-induced NF-κB, AP-1 DNA binding activity and COX-2 expression in gastric epithelial cells. In addition, CAPE reduced TNF-α and IL-8 levels and suppressed the proliferative response of AGS cells to H. pylori. They found also that pretreatment of gastric epithelial cells with CAPE upregulated IkB-a levels and prevented nuclear translocation of NF-κB/ p65 in H. pylori-treated AGS cells. NF-κB is present in the cytosol in an inactive state bound to the inhibitory IkB protein. H. pylori infection of gastric epithelial cells results in phosphorylation and degradation of the IkB, thus allowing nuclear translocation of NF-κB. It is mechanistically proven that inflammation produces reactive oxygen species (ROS) and reactive nitrogen species (RNS). In particular, ROS and RNS lead to oxidative damage and nitration of DNA bases, which increases the risk of DNA mutations and further leads to cancer. Nitric oxide (NO) is associated with inflammatory reaction and is produced by 302 Carcinogenesis inducible nitric oxide synthase (iNOS) in certain cells activated by various proinflammatory agents. NO acts as a host defense by damaging membranes of pathogenic bacteria and as a regulatory molecule with homeostatic activities. However, excessive production of NO is pathogenic for host tissue itself because NO, as a reactive radical directly damages functions of normal tissue. Thus, effective inhibition of NO accumulation by inflammatory stimuli represents a beneficial therapeutic strategy [38,39]. Nagaoka et al. [40] reported that CAPE possesses potent NO inhibitory activities and suggested that the NO inhibitory effect can directly correlate with anti-inflammatory properties of the Netherlands propolis. They suggested that the active principles of the Netherlands propolis, i.e., CAPE and its analogues, should block the activation of iNOS through the suppression of NF-kB activation and resulted in potent NO inhibition. We investigated the anti-inflammatory and antioxidant potential of CAPE on a tumor cell line (ZR-75-1). We found that CAPE at the concentration of 15μM inhibited NO production by (› 47%) compared to NO level of untreated tumor cells (P‹ 0.05). In addition, superoxide dismutase (SOD) was at the highest level in the maintained basal tumor culture cell supernatant ( 231.9± 4.2 μU/L). 2.1. Regulation of inflammatory pathways by CAPE This level reduced intensively to (169.3 ± 3.7 μU/L), in the CAPE-treated culture cells, which was significantly less than the level in untreated cells ( P‹0.001). On the other hand, malonaldehyde (MDA) level, which is considered to be an important parameter for the oxidative damage determination, was inhibited to ( 17.3 ± 2.3 μmol/L) in the CAPE-treated cells when compared with untreated tumor cells (23.8 ± 2.5 μmol/L), but was not statistically significant (Table 1). MDA (μ mol/L) SOD ( μ U/ L ) NO· (μ mol/L) Group Parameters 23.8 ± 2.5 231.9 ± 4.2 19.6 ± 1.9 I 17.3 ± 2.3 169.3 ± 3.7 10.2 ± 1.3 II N.S ‹ 0.001 ‹ 0.05 P (I). Untreated tumor cells , (II). Treated tumor cells at 15μM CAPE. Table 1. Nitric oxide (NO·), Superoxide dismutase (SOD) and Malondialdehyde (MDA) values in culture supernatant. Table 1. Nitric oxide (NO·), Superoxide dismutase (SOD) and Malondialdehyde (MDA) values in culture supernatant. 2.2. Regulation of tumor cell development by CAPE Under normal physiological conditions, the human body maintains homeostasis by eliminating unwanted, damaged, aged, and misplaced cells. Homeostasis is carried out in a genetically programmed manner by a process referred to as apoptosis (programmed cell death). Cancer cells are able to evade apoptosis and grow in a rapid and uncontrolled manner. One of the most important ways by which cancer cells have gained this ability is through mutation in the p53 tumor suppressor gene. Without a functional p53 gene, cells lack the DNA-damage-sensing capability that would normally induce the apoptotic cascade [41-43]. 303 Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester A complex set of proteins, including caspases, proapoptotic and antiapoptotic B cell lymphoma (Bcl)-2 family proteins, cytochrome c, and apoptotic protease activating factor (Apaf)-1, execute apoptosis either by an intrinsic or extrinsic pathway. The intrinsic pathway is mitochondria dependent, whereas the extrinsic pathway is triggered by death receptors (DRs). Some antiapoptotic proteins such as Bcl-2 and B cell lymphoma extra large (Bcl-xL) and survivin are overexpressed in a wide variety of cancers. Therefore, selective downregulation of antiapoptotic proteins and upregulation of proapoptotic proteins and p53 in cancer cells offer promising therapeutic interventions for cancer treatment [44,45]. We tested the effect of CAPE on the viability of human breast cancer ZR-75-1 cells derived from a malignant ascitic effusion in a 63 year-old, white female with infiltrating ductal carcinoma [46]. CAPE induced a significant inhibitory effect on the growth and viability of tumor cells in vitro. We observed that this inhibitory action was highly dosage and time- dependent. The maximum inhibitory action was obtained at 15μM (Figure 2A) on culture media. Figure 2. CAPE induced ZR-75-1 cell death (A); cell viability was observed in tumor cells treated with different CAPE concentrations (5, 10, 15 and 20μM), while (B) cells showed maximum alteration of viability at 48 hrs. of incubation, indicating that CAPE induced cell death in a dose and time dependent manner. The results shown in the histogram were the mean ± S.D. Assay was performed in 12 – well plate (2 x 106 ) cells/well, using trypan blue dye for viability detection. e 2. CAPE induced ZR-75-1 cell death (A); cell viability was observed in tumor cells treated with Figure 2. 2.2. Regulation of tumor cell development by CAPE CAPE induced ZR-75-1 cell death (A); cell viability was observed in tumor cells treated with different CAPE concentrations (5, 10, 15 and 20μM), while (B) cells showed maximum alteration of viability at 48 hrs. of incubation, indicating that CAPE induced cell death in a dose and time dependent manner. The results shown in the histogram were the mean ± S.D. Assay was performed in 12 – well plate (2 x 106 ) cells/well, using trypan blue dye for viability detection. Figure 2. CAPE induced ZR-75-1 cell death (A); cell viability was observed in tumor cells treated with different CAPE concentrations (5, 10, 15 and 20μM), while (B) cells showed maximum alteration of viability at 48 hrs. of incubation, indicating that CAPE induced cell death in a dose and time dependent manner. The results shown in the histogram were the mean ± S.D. Assay was performed in 12 – well plate (2 x 106 ) cells/well, using trypan blue dye for viability detection. On the other hand, after 48 hours of administration, the percentage of cell death increased significantly to 49.6±6.9 (Figure 1B). No further changes were observed after 72 hours of treatment. To investigate the induced effect of apoptosis on cell viability, we analyzed the DNA fragmentation using DAPI. The nuclear structure exhibited condensation and fragmentation of some nuclei that were caused by 15μM CAPE at 48 hours. Apoptotic cells count significantly increased by 27% as compared to control tumor cells (P ‹ 0.01) (Figure 3B) [47]. Additionally, the microscopic examination did not reveal any signs of morphological changes after 12 hours of administration. However, a scattered retraction of the monolayer, vacuoles, and the granulation of the cytoplasm were observed after 24 hours. The 304 Carcinogenesis alterations were further aggravated after 48 hours. The cells were rounded up. Later, they became phase-dense and formed floating aggregates, which gradually increased in size and most of the cells were detached from the flasks. The cell membranes burst, which was followed by a gradual decrease of the cell count. These changes were not reversible. Transferring the cells into a fresh medium did not alter their state (Figure 4) [47]. Figure 3. Effect of CAPE on expression of apoptotic cells. (A) ZR-75-1 cells without treatment. (B) Cells were treated with 15μM CAPE for the indicated time of 48 hrs. Condensation and apoptotic bodies were examined by immunofluorescence microscopy. Magnification, X20. Figure 3. 2.2. Regulation of tumor cell development by CAPE Effect of CAPE on expression of apoptotic cells. (A) ZR-75-1 cells without treatment. (B) Cells were treated with 15μM CAPE for the indicated time of 48 hrs. Condensation and apoptotic bodies were examined by immunofluorescence microscopy. Magnification, X20. Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 305 Figure 4. CAPE induced cell detaching (A) ZR-75-1 cells kept without treatment during experiments. (B) 15μM CAPE treatment after 48 hrs. Morphology changes in cells were examined by phase-contrast microscopy. Magnification, X40. Figure 4. CAPE induced cell detaching (A) ZR-75-1 cells kept without treatment during experiments. (B) 15μM CAPE treatment after 48 hrs. Morphology changes in cells were examined by phase-contrast microscopy. Magnification, X40. 2.3. Tumor growth inhibition in vivo by CAPE The effects of CAPE on the survival of mice bearing tumor are shown in figure 5. The median survival time for the untreated group of mice was 21 days. On the other hand, the group of mice bearing tumor and treated with 10mg/kg S.C/ every 5 days had a median survival time of 29 days. Two mice were completely cured. The median survival time of the group treated with 15mg/kg S.C/ every 5 days was found to be 43 days and 3 mice were completely cured. However, those mice treated with 5mg /kg S.C/ every 5 days did not show any remarkable changes in their survival percentage. It has also been observed that giving treatment more than once a week caused sores and increased irritability in the mice, vehicle (1:1/DMSO: NaCl). We studied the effect of CAPE (15mg/kg S.C/ every 5 days) on the growth of transplanted Ehrlich carcinoma into Swiss mice. The solid tumor volume showed a reduced rate in CAPE-treated mice and appreciably smaller volume (1.9 ± 0.46) mm³, with respect to the untreated group (3.7 ± 0.82) mm³. This value was significant (P‹ 0.01). The difference was observed from the beginning of tumor measurement i.e. since the 6th day after tumor implantation to the host, and was maintained until the end of observation (Figure 6). This finding is one of the characteristic effects of anti-tumor drugs. It is also in accordance with 306 Carcinogenesis the other findings, which suggest that subcutaneous administration of an aqueous crude water-soluble propolis (CWSP) resulted in marked regression of transplanted tumors [48]. Figure 5. Effect of CAPE treatment on the time of survival for mice bearing solid tumor at different doses compared with untreated group. Figure 5. Effect of CAPE treatment on the time of survival for mice bearing solid tumor at different doses compared with untreated group. gure 5. Effect of CAPE treatment on the time of survival for mice bearing solid tumor at different oses compared with untreated group. Figure 6. Effect of CAPE (15mg/kg S.C), once/week and twice/week treatment on tumor volume against untreated group of mice. Data represented as mean ± S.D. Figure 6. Effect of CAPE (15mg/kg S.C), once/week and twice/week treatment on tumor volume against untreated group of mice. Data represented as mean ± S.D. The percentage of apoptotic cells with hypodiploid DNA content was determined from DNA histograms. 2.3. Tumor growth inhibition in vivo by CAPE Untreated tumor-bearing mice showed a peak pattern which represented proliferative and high metastatic tumor activity (Figure 7A). However the mice which had their tumor treated at 15mg/kg S.C did not alter the relative size of the peak, but induced a significant parallel shift to less-intense fluorescence (D- area). This decrease in the intensity and shift may be termed as nuclear apoptosis and fragmentation (Figure 7B). Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 307 Figure 7. The value in the tumor apoptosis histogram was higher in mice-bearing tumors, which were treated with CAPE (15mg/kg/S.C). (B) as compared to mice-bearing tumors untreated with CAPE (A). The apoptosis was identified by PI staining (D-area) with increased DNA fragmentation [47]. Figure 7. The value in the tumor apoptosis histogram was higher in mice-bearing tumors, which were treated with CAPE (15mg/kg/S.C). (B) as compared to mice-bearing tumors untreated with CAPE (A). The apoptosis was identified by PI staining (D-area) with increased DNA fragmentation [47]. Chung et al., [49] showed that both CA and CAPE selectively inhibit matrix metalloproteinases (MMP) 2 and 9. CAPE inhibited strongly MMPs 2 and 9 with IC50 of 2– 5μM, while CA required 10–20 μM for such inhibition. In contrast, MMPs 1, 3, 7 and Cathepsin-K were not completely inhibited by either of them. CA and CAPE had a dose- dependent inhibitory effect on the proliferation of HEPG2 cells. In HepG2 cells, CA at the concentration of 200 μg/mL reduced the viability to 61% , while CAPE, at 10 times lower concentration, inhibited the viability to 72% as compared to the respective controls. CAPE and CA suppressed the MMP 9 expression in HepG cells exposed to phorbol 12-myristate 13-acetate (PMA). They also confirmed that CA (20 mg/kg) and CAPE (5mg/kg) repressed the growth of HepG2 tumor xenografts in nude mice as well as liver metastasis when administered subcutaneous or orally. Finally they concluded their observation that CA and its derivative CAPE: (1) inhibited the enzymatic activity of MMP-9 that plays an important role in cancer invasion and metastasis, (2) blocked the invasive potential through the suppression of MMP-9 gene transcription by inhibiting NF-κB function in PMA-stimulated HepG2 cells and (3) suppressed the growth of HepG2 cell xenografts in nude mice. 2.4. Regulation of tumor cell proliferation by CAPE Dysregulated proliferation is one of the major characteristics of tumorigenesis. In normal cells, proliferation is regulated by a delicate balance between growth signals and antigrowth signals. Cancer cells, however, acquire the ability to generate their own growth signals and become insensitive to antigrowth signals [51]. Their growth is controlled by cell cycle regulators at the G1/S-phase boundary, in the S phase, and during the G2/M phases of the cell cycle. A precise set of proteins called cyclins and cyclin-dependent kinases (CDKs) control the progression of cell cycle events. Whereas cyclin binding is required for CDK activity, CDK inhibitors (CKIs) such as p21 and p27 prevent CDK activity and prevent cell cycle progression. The G1-to-S-phase transition also requires cellular v-myc myelocytomatosis viral oncogene homolog (c-Myc), and inhibition of c-Myc expression leads to growth arrest [52,53]. The expression of c-Myc in turn is regulated by cdc25, a phosphatase that activates CDKs. The well-characterized tumor suppressor p53 has been implicated in controlling the G1-to- S-phase transition and in blocking cell cycle progression at the G1 phase in response to DNA damage [54]. A number of genes controlling cell cycle progression, including the CKI p21, are transcribed in a p53-dependent manner [55,56]. Rb is a tumor suppressor retinoblastoma protein that, like p53, functions as a negative regulator of cell growth [57]. Rb inactivation or deletion has been found in many cancers, including retinoblastomas and carcinomas of the lung, breast, bladder, and prostate. By binding to and inhibiting transcription factors such as elongation 2 factor (E2F), which are necessary for S-phase entry, Rb is believed to inhibit cell cycle progression [58]. On the other hand, phosphorylation of Rb (pRb) by CDK/cyclin complexes results in the release of active E2F species to stimulate the transcription of genes involved in DNA synthesis and S-phase progression [59,60]. Currently, a number of inhibitors based on cell cycle regulators, including nutraceuticals, are being developed as therapeutic intervention for cancer prevention. Nutraceuticals have been shown to have potential in cancer prevention for halting cell cycle progression by targeting one or more steps in the cell cycle. Most nutraceuticals prevent the transition of cancer cells from the G1 to S phase. Some of these nutraceuticals act through p53 and some through Rb. Acetyl-keto-beta-boswellic acid was shown to arrest colon cancer cells at the G1 phase, which was associated with decreases in cyclin-D1, cyclin-E, CDK-2, CDK-4, and pRb and an increase in p21[61]. 2.3. Tumor growth inhibition in vivo by CAPE Therefore, these two drugs were reported as strong candidates for treatment of cancer and metastasis via dual mechanisms (dual inhibition of metastasis-specific enzyme activity and gene transcription) [50]. 308 Carcinogenesis 2.4. Regulation of tumor cell proliferation by CAPE In Ehrlich ascites tumor cells, acetoxychavicol acetate was shown to stimulate the accumulation of tumor cells in the G1 phase of the cell cycle, which was accompanied by a decrease in pRb and an increase in Rb [62]. β-Escin, a triterpene saponin, induced cell cycle arrest at the G1/S phase by inducing p21 and reducing pRb in a p53-independent manner in HT-29 human colon cancer cells [63]. In gastric cancer cells, curcumin was shown to suppress the transition of cells from the G1 to S phase, which was accompanied by a decrease in cyclin-D1 and p21-activated kinase 1 activity [64]. Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 30 The therapeutic goal of cancer treatment is to induce apoptotic death of cancer cells rather than necrosis due to the deleterious consequences of the latter, which include leakage of lysosomal enzymes to the extracellular media and spawning a substantial inflammatory reaction. Several investigators have demonstrated that CAPE has an anti-proliferative effect and an apoptosis inducing effect against various tumor cell lines. Cavaliere et al., showed that CAPE treatment increased the percentage of cells in G0/G1 and decreased the percentage of cells in S and G2/M phase in addition to its ability to inhibit DNA, RNA, and protein synthesis , thus delaying cell cycle progression to G2/M phase. CAPE also caused high levels of apoptotic cell death of 77.1%, with no signs of significant necrosis in PL104 cells [65]. Also, Wang et al., demonstrated that CAPE treatment was associated with a strong inhibition of proliferation in a dose- and time-dependent manner, along with induction of G0/G1 arrest and apoptosis in HCT116 cells [66]. The therapeutic goal of cancer treatment is to induce apoptotic death of cancer cells rather than necrosis due to the deleterious consequences of the latter, which include leakage of lysosomal enzymes to the extracellular media and spawning a substantial inflammatory reaction. Several investigators have demonstrated that CAPE has an anti-proliferative effect and an apoptosis inducing effect against various tumor cell lines. Cavaliere et al., showed that CAPE treatment increased the percentage of cells in G0/G1 and decreased the percentage of cells in S and G2/M phase in addition to its ability to inhibit DNA, RNA, and protein synthesis , thus delaying cell cycle progression to G2/M phase. 2.4. Regulation of tumor cell proliferation by CAPE CAPE also caused high levels of apoptotic cell death of 77.1%, with no signs of significant necrosis in PL104 cells [65]. Also, Wang et al., demonstrated that CAPE treatment was associated with a strong inhibition of proliferation in a dose- and time-dependent manner, along with induction of G0/G1 arrest and apoptosis in HCT116 cells [66]. 2.6. Regulation of tumor cell angiogenesis by CAPE Angiogenesis, the process during which new blood vessels are formed from preexisting ones, can be classified as either physiological or pathological. Physiological angiogenesis provides a driving force for organ development in ontogeny, is necessary for ovulation, and is a prerequisite for wound healing; pathological angiogenesis occurs during tumor growth at primary and metastatic sites [85]. The angiogenic cascade during tumor development consists of the release of angiogenic factors, binding of angiogenic factors to receptors on endothelial cells (ECs), EC activation, degradation of the basement membrane by proteases, and migration and proliferation of ECs. Adhesion molecules then help to pull the sprouting blood vessels forward, and ECs are finally organized into a network of new blood vessels [86]. The signaling pathway governing tumor angiogenesis is exceedingly complex, involving various angiogenic mediators. The major signaling mediators include VEGF, platelet- derived growth factor, fibroblast growth factors (FGFs), epidermal growth factor, ephrins, angiopoietins, endothelins, integrins, cadherins, and notch [87]. P2 P1 Treated Group n = 10 Tumor Group n = 10 Normal Group n = 10 Group Parameters P‹ 0.01 P‹ 0.001 142.1 84.5 – 196.3 181.9 88.2 – 216.1 130.9 79.6 – 166.4 MMP-9 (ng/ml ) Median Range P‹ 0. 01 P = 0.01 1.9 1.3 – 9.3 1.4 1.0 – 5.4 2.2 1.1 – 6.3 Endostatin (ng/ml) Median Range P1 Tumor group Vs Normal group P2 Treated group Vs Tumor group Table 2. Levels of MMP-9 and Endostatin serum in normal mice group, tumor bearing group and treated tumor group. Table 2. Levels of MMP-9 and Endostatin serum in normal mice group, tumor bearing group and treated tumor group. Table 2. Levels of MMP-9 and Endostatin serum in normal mice group, tumor bearing group and treated tumor group. We found that the untreated mice bearing Ehrlich tumor elicited a highly significant increase of serum MMP-9 level (181.9 ng/ml), which was reduced (142.1 ng/ml) in mice treated with CAPE at a dose of 15mg/kg (P‹ 0.01) close to the normal mice serum level. However, in the untreated mice bearing tumor serum endostatin (sE) was significantly lowered (1.4 ng/ml) compared with the normal mice. In CAPE-treated mice serum endostatin level was significantly higher (1.9 ng/ml) than the serum level in the untreated group (Table 2)[47]. 2.5. Regulation of tumor cell invasion by CAPE Tumor cell invasion and metastasis are interrelated processes involving cell growth, cell adhesion, cell migration, and proteolytic degradation of tissue barriers such as the extracellular matrix and basement membrane. Several proteolytic enzymes, including MMPs (chiefly MMP-2 and MMP-9) [67,68] and intercellular adhesion molecule (ICAM; chiefly ICAM-1), participate in the degradation of these barriers [69,70]. A number of studies in lung, colon, breast, and pancreatic carcinomas have demonstrated overexpression of MMPs in malignant tissues compared with adjacent normal tissues [71-78]. Apart from MMPs, cysteine proteases [79] and serine proteases [80] such as urokinase-type plasminogen activator (u-PA) have also been involved in the invasion and metastasis of cancer cells. Since both u-PA and u-PA receptor (u-PAR) contain binding sites for NF-κB and activator protein (AP)-1 in their promoter regions [81-83], inhibition of these transcription factors will eventually result in the inhibition of u-PA–u-PAR complex and subsequent suppression of invasive behavior. Furthermore, Hwang et al. [84] investigated the effect of CAPE on tumor invasion and metastasis in HT 1080 fibrosarcoma cells by determining the regulation of matrix metalloproteinases (MMPs). HT 1080 cells were treated with increasing concentration of CAPE and the m-RNA transcripts of MMP-2 and MMP-9 were analyzed using semi- quantitative RT-PCR. Both MMP- 2 and 9 proteins levels were significantly suppressed in a dose dependent manner. Gelatin zymography also indicated constitutively expressed MMP-2 and 9 proteins in HT 1080 cells, which gradually reduced after treating with CAPE. To further corroborate the down regulation of MMP-2, activation studies of pro- MMP2 were performed using organomercuric compound, 4-aminophenylmercuric acetate (APMA), and the result indicated the down regulation of MMP-2 by CAPE. It has been shown that mRNA levels of tissue inhibitor of matrix metalloproteinases (TIMPs) and membrane type-matrix metalloproteinases (MT-1 MMPs) were also reduced significantly. CAPE also inhibited the cell invasion, cell migration and colony formation of tumor cells. Thus CAPE acts as a vital antimetastatic agent, by inhibiting the metastatic and invasive potential of malignant cells. 310 Carcinogenesis 310 Carcinogenesis 3. Conclusion These findings obtained suggest that CAPE is a potent agent, which has antioxidant properties. In vitro findings support that CAPE could be potentially useful in the control of tumor cell proliferation as well as, an apoptotic-inducing agent. Furthermore, CAPE exhibits anti-metastatic and anti-angiogenic properties. CAPE could be potentially useful in the control of tumor growth in experimental models. Its action is accompanied by the shifting and elevating of the angiostatic and inhibiting angiogenic factors. Finally, it has been demonstrated that CAPE has many biological and pharmacological properties with predictive future applications in human clinical trials. Author details Mohamed F. El-Refaei* Engineering and Biotechnology Institute, Menoufiya University, Sadat City, Egypt Dept. of Molecular Biology, Institute of Genetic, Menoufia University, Egypt Faculty of Medicine, Al-Baha University, Al-Baha Province, KSA Mohamed F. El-Refaei* Engineering and Biotechnology Institute, Menoufiya University, Sadat City, Egypt Dept. of Molecular Biology, Institute of Genetic, Menoufia University, Egypt Faculty of Medicine, Al-Baha University, Al-Baha Province, KSA Mohamed F. El-Refaei* Engineering and Biotechnology Institute, Menoufiya University, Sadat City, Egypt Dept. of Molecular Biology, Institute of Genetic, Menoufia University, Egypt Faculty of Medicine, Al-Baha University, Al-Baha Province, KSA Mohamed F. El-Refaei Engineering and Biotechnology Institute, Menoufiya University, Sadat City, Egypt Dept. of Molecular Biology, Institute of Genetic, Menoufia University, Egypt Faculty of Medicine, Al-Baha University, Al-Baha Province, KSA Essam A. Mady Dept. of Biochemistry, Faculty of Science, Ain Shams University, Egypt Faculty of Medicine, Al-Baha University, Al-Baha Province, KSA Essam A. Mady Dept. of Biochemistry, Faculty of Science, Ain Shams University, Egypt Faculty of Medicine, Al-Baha University, Al-Baha Province, KSA 2.6. Regulation of tumor cell angiogenesis by CAPE On the other hand, there was a negative correlation between (sMMP-9 and sE) and the total white blood cells (WBCs), hemoglobin (HB) and the platelet count of mice (Table 3)[47]. Based on these findings, we concluded that; the endogenous inhibitor of the angiogenic serum (endostatin) has been shown to be overexpressed, significantly higher in treated mice compared to untreated mice (1.9 ng/ml), and nearly to the value of normal serum mice (Table 2). These findings may help to utilize endostatin itself in the therapy. It also indicates that 311 Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 31 CAPE has the potential of an anti-metastatic agent. It may mediate CAPE effects by inhibiting the cell proliferation. The findings of this study are in accordance with Schuch et al. [88] who, claimed that endostatin microbeads significantly inhibit the growth of subcutaneous choloromas in SCID mice as compared to control mice. On the other hand, MMP-9 and endostatin did not correlate with the white blood cells (WBCs), hemoglobin (HB) and platelet count (Table 3). It may be concluded that MMP-9 and endostatin are independent factors. HB WBCs Count Platelets Count MMP-9 R= 0.11, ›0.05 R= 0.19, ›0.05 R= 0.24, ›0.05 Endostatin R= 0.09, ›0.05 R= 0.1, ›0.05 R= 0.04, ›0.05 Table 3. Correlation between investigated angiogenic factors and, hemoglobin (HB), white blood cells count (WBCs) and platelets count in tumor treated mice. Table 3. Correlation between investigated angiogenic factors and, hemoglobin (HB), white blood cells count (WBCs) and platelets count in tumor treated mice. Table 3. Correlation between investigated angiogenic factors and, hemoglobin (HB), white blood cells count (WBCs) and platelets count in tumor treated mice. * Corresponding Author 4. References [1] Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell. J.100(1) :57–70. [2] Bray F, Moller B (2006) Predicting the future burden of cancer. Nat Rev Can.J.6: 6 [3] Hanahan D, Weinberg RA (2011) The hallmarks of cancer: the next generation.. Cell. J . 4;144(5):646-674. [4] Vincent TL, Gatenby RA (2008) An evolutionary model for initiation, promotion, and progression in carcinogenesis. Int J Oncol 32(4): 729-737. [5] Anand P, Sundaram C, Jhurani S, Kunnumakkara AB, Aggarwal BB(2008) Curcumin and cancer: An “old-age” disease with an “age-old” solution. CanLet J. 267:133–164. [6] Vogelstein B, Kinzler KW (2004) Cancer genes and the pathways they control. Nat Med. J .10:789–799. [7] Aggarwal BB, Vijayalekshmi RV, Sung B (2009) Targeting inflammatory pathways for prevention and therapy of cancer: short-term friend, longterm foe. Clin Can Res. J .15: 425–430. [8] Sporn MB, Newton DL (1979) “Chemoprevention of cancer with retinoids,” Feder Proce. J . 38(11) 2528–2534. [9] Kelloff GJ (1999) “Perspectives on cancer chemoprevention research and drug development” Adva in CanRes.J.78:320–334. [10] Kelloff, GJ, Boone CW (1994) Eds., “Cancer chemopreventive agents: drug development status and future prospects Cell Biochem. J . 20:1–303. [11] Kelloff GJ, Hawk ET, Sigman CC (2004) Eds., Cancer Chemoprevention: Prom Can Chemop Ag vol. 1, Humana Press, Totowa, NJ, USA. [12] Ferguson LR (1994) “Antimutagens as cancer chemopreventive agents in the diet,” Mut Res.j. 307(1) 395–410. [13] Ferguson LR, Philpott M, Karunasinghe N (2004) “Dietary cancer and prevention using antimutagens,” Toxico. J .198(1–3):147–159. [14] Surh YJ, (2003) “Cancer chemoprevention with dietary phytochemicals,”Nat Rev Can. J .3(10):768–780. [15] Amiot MJ, Aubert S, Gonnet M, Tacchini M, (1989) “Les compos´es ph´enoliques des miels: ´etude pr´eliminaire sur l’identifi cation et la quantifi cation par familles,” Apidologie. J .20:115–125. [16] World Wide Wounds, (2002) “Honey as a topical antibacterial agent for treatment of infected wounds,”http://www.worldwidewounds.com/2001/november/Molan/ honey-as- topical agent.html. p g [17] Orsolic N, Terzic S, Mihaljevic Z, Sver L, basic I (2005) Effects of local administration of propolis and its polyphenolic compounds on tumor formation and growth. Biol Pharm Bull. J . 28:1928-1933. [18] Carrasco-Legleu CE, Márquez-Rosado L, Fattel-Fazenda S, Arce-Popoca E et al., (2004) Chemoprotective effect of caffeic acid phenethyl ester on promotion in a medium-term rat hepatocarcinogenesis assay. Int J Can.. 10;108(4):488-92. Acknowledgement The authors wish to express gratitude for Al-Baha University for their support. Deepest gratitude is also due to the Dean and the staff members, faculty of Medicine, Al-Baha University, KSA. Carcinogenesis 312 Carcinogenesis 4. References p g y [19] Mahmoud NN, Carothers AM, Grunberger D, Bilinski RT, Churchill MR, Martucci C, et al.,(2000) Plant phenolics decrease intestinal tumors in an animal model of familial adenomatous polyposis.Carcinog. 21(5):921-927. Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 3 Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 31 [20] Huang MT, Ma W, Yen P, Xie JG, Han J, Frenkel K, Grunberger D, Conney AH (1996) Inhibitory effects of caffeic acid phenethyl ester (CAPE) on 12-O-tetradecanoylphorbol- 13-acetate-induced tumor promotion in mouse skin and the synthesis of DNA, RNA and protein in HeLa cells. Carcinog .17(4):761-765. [21] Carrasco-Legleu CE, Sánchez-Pérez Y, Márquez-Rosado L et al., (2006) A single dose of caffeic acid phenethyl ester prevents initiation in a medium-term rat hepatocarcinogenesis model. W Gastroen . J .14;12(42):6779-85. [22] Wu J, Omene C, Karkoszka J, Bosland M, Eckard J, Klein CB, Frenkel K (2011) Caffeic acid phenethyl ester (CAPE), derived from a honeybee product propolis, exhibits a diversity of anti-tumor effects in pre-clinical models of human breast cancer.Cancer Lett. 1;308(1):43-53. [23] Omene CO, Wu J, Frenkel K (2012) Caffeic Acid Phenethyl Ester (CAPE) derived from propolis, a honeybee product, inhibits growth of breast cancer stem cells. Invest. New D. J.30(4):1279-1288. [24] Subash C,Gupta Ji, Hye Kim, Sahdeo Prasad, Bharat B, Aggarwal BB (2010) Regulation of survival, proliferation, invasion, angiogenesis, and metastasis of tumor cells through modulation of inflammatory pathways by nutraceuticals. Can. Meta. J .29(3): 405–434. y p y y [25] Aggarwal BB, Gehlot P (2009) Inflammation and cancer: How friendly is the relationship for cancer patients? Current Opin. in Pharm.. J . 9:351–369. [26] Coussens LM, Werb Z (2002) Inflammation and cancer. Nat.. J . 420: 860–867. [27] Dey A, Tergaonkar V, Lane DP. (2008) Double-edged swords as cancer therapeutics: simultaneously targeting p53 and NF-kappaB pathways. Nat. Rev. D. Disc. 7, 1031-1040. [28] Kabrun N, Enrietto PJ (1994) The Rel family of proteins in oncogenesis and differentiation. Seminars in Can. Biol. 5:103–112. [29] Ghosh S, Karin M (2002) Missing pieces in the NF-kappaB puzzle. Cell. J . 109(Suppl):S81–S96. [30] Hussain SP, Hofseth LJ, Harris CC (2003) Radical causes of cancer. Nat. Rev. J .3: 276– 285 [31] Gupta SC, Sundaram C, Reuter S, Aggarwal BB (2010) . Inhibiting NF- kappaB activation by small molecules as a therapeutic strategy. Bioch. et Bioph. Act.. J .1799(10- 12):775-787. 4. References [32] de Visser KE, Eichten A, Coussens LM (2006) Paradoxical roles of the immune system during cancer development. Nat. Rev Can.. J .6: 24–37. [33] Chaturvedi MM, Kumar A, Darnay BG, Chainy GB, Agarwal S, Aggarwal BB(1997) Sanguinarine (pseudochelerythrine) is a potent inhibitor of NF-kappaB activation, IkappaBalpha phosphorylation, and degradation. Biol. Chem.. J .272:30129–30134. [34] Kumar A, Dhawan S, Aggarwal BB. (1998) Emodin (3-methyl-1,6,8- trihydroxyanthraquinone) inhibits TNF-induced NF-kappaB activation, IkappaB degradation, and expression of cell surface adhesion proteins in human vascular endothelial cells. Onco. ;17:913–918. [35] Jing Y, Yang J, Wang Y, Li H, Chen Y, Hu Q, et al. (2006) Alteration of subcellular redox equilibrium and the consequent oxidative modification of nuclear factor kappaB are critical for anticancer cytotoxicity by emodin, a reactive oxygen species-producing agent. Free Radic. Biol. & Med.. J .40:2183–2197. Carcinogenesis 314 Carcinogenesis 314 [36] Natarajan K, Singh S, Burke T Jr TR, Grunberger D Aggarwal BB (1996) “Caffeic acid phenethyl ester is a potent and specific inhibitor of activation of nuclear transcription factor NF-κB,” Proce. of the Nati.Acad. of Sci.of theUSA. 93(17)9090–9095. [37] Abdel-Latif MM, Windle HJ, El Homasany BS, Sabra K, Kelleher D (2005). Caffeic acid phenethyl ester modulates Helicobacter pylori-induced nuclear factor-kappa B and activator protein-1 expression in gastric epithelial cells. B J. of Pharm .146: 1139–1147. [38] Moncada S, Palmer RMJ, Higgs EA (1991) Nitric oxide: physiology, pathophysiology, and pharmacology. Pharm. Rev J . 43:109—142. p gy [39] Kou PC, Schroeder RA (1995) The emerging multifacated roles of nitric oxide.Annal. Sur. J .221:220—235. [40] Nagaoka T, Banskota AH, Tezuka Y, Midorikawa K, Matsushige K, Kadota S (2003) Caffeic acid phenethyl ester (CAPE) analogues: potent nitric oxide inhibitors from the Netherlands propolis. Biol. Pharm. Bull. J . 26: (4) 487—491. p p ( ) [41] Steller H (1995) Mechanisms and genes of cellular suicide. Sci.. J . 267:1445–1449. [42] Green DR (2000) Apoptotic pathways: Paper wraps stone blunts scissors. Cell. J.1 [43] Meier P, Finch A, Evan G (2000) Apoptosis in development. Nat.. J . 407:796–801. [44] Wang S, Yang D, Lippman ME (2003) Targeting Bcl-2 and Bcl-XL with nonpeptidic small-molecule antagonists. Seminars in Onco.. 30:133–142. [45] Ambrosini G, Adida C, Altieri DC (1997) A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nat. Med..J. .3:917–921. [46] Engel L, Young N (1978) Human breast carcinoma cells in continuous culture: a review. Can. Res. 38: 4327-4339. 4. References [47] El-Refaei M, El-Naa M (2010) Inhibitory effect of caffeic acid phenethyl ester on mice bearing tumor involving angiostatic and apoptotic activities. Chemi Biologi Interact 186:152-156. [48] Suzuki I, Hayashi I, Takaki T, Groveman D, Fujimiya Y (2002) Antitumor and Anticytopenic Effects of Aqueous Extracts of Propolis in Combination with Chemotherapeutic Agents. Can. Bioth.y & Radiopharmac., 17:553-562. [48] Chung TW, Moon SK, Chang YC et al., (2004) “Novel and therapeutic effect of caffeic acid and caffeic acid phenyl ester on hepatocarcinoma cells: complete regression of hepatoma growth and metastasis by dual mechanism,” The FASEB. J .18(14):1670–1681. [49] Ryan KM, Birnie GD (1987) Deregulated expression of c-Myc has been implicated in a number of human malignancies. Myc oncogenes: The enigmatic family. Nat.. J . 328:445–449. [50] Hanahan D, Weinberg RA (2000) The hallmarks of cancer. Cell J;100:57–70. [51] Heikkila R, Schwab G, Wickstrom E, Loke SL, Pluznik DH et al., ( 1987) A c-myc antisense oligodeoxynucleotide inhibits entry into S phase but not progress from G0 to G1 Nat. J. . 5;328(6129):445-449. [52] Evan GI, Littlewood TD (1993) The role of c-myc in cell growth. Curr. Opin. in Gen. & Dev..3:44–49. [53] Kuerbitz SJ, Plunkett BS, Walsh WV, Kastan MB (1992) Wild-type p53 is a cell cycle checkpoint determinant following irradiation. Proc. of the Nat. Acad. of Sci. of the USA. 89:7491–7495. [54] Dulic V, Kaufmann WK, Wilson SJ, Tlsty TD, Lees E, Harper JW, et al.(1994) p53- dependent inhibition of cyclin-dependent kinase activities in human fibroblasts during radiation-induced G1 arrest. Cell. J . 76:1013–1023. Regulation of Apoptosis, Invasion and Angiogenesis of Tumor Cells by Caffeic Acid Phenethyl Ester 315 [55] El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM (1993) WAF1, a potential mediator of p53 tumor suppression. Cell. J . 75:817–825. [56] Weinberg RA (1995) The retinoblastoma protein and cell cycle control. Cell .J. 81:323–330. [57] King KL, Cidlowski JA (1998) Cell cycle regulation and apoptosis. Ann. Rev. of Phys. 60:601–617. [58] Hiebert SW (1993) Regions of the retinoblastoma gene product required for its interaction with the E2F transcription factor are necessary for E2 promoter repression and pRb-mediated growth suppression. Mol.and Cell. Biol. 13:3384–3391. [59] Qian Y, Luckey C, Horton L, Esser M, Templeton DJ (1992) Biological function of the retinoblastoma protein requires distinct domains for hyperphosphorylation and transcription factor binding. Mol. Cell Biol. J. 12(12): 5363–5372. 4. References [60] Liu JJ, Huang B, Hooi SC (2006) Acetyl-keto-beta-boswellic acid inhibits cellular proliferation through a p21-dependent pathway in colon cancer cells. B. J. of Pharm.. J. 148:1099– 1107. [61] Xu S, Kojima-Yuasa A, Azuma H, Huang X, Norikura T, Kennedy DO, et al.(2008) (1’S)- Acetoxychavicol acetate and its enantiomer inhibit tumor cells proliferation via different mechanisms. Chem. Biol. Interact.172:216–223. [62] Patlolla JM, Raju J, Swamy MV, Rao CV (2006) Beta-escin inhibits colonic aberrant crypt foci formation in rats and regulates the cell cycle growth by inducing p21(waf1/cip1) in colon cancer cells Mol. Can. Therapeu.. 5:1459–1466. [63] Cai XZ, Wang J, Li XD, Wang GL, Liu FN, Cheng MS, et al.(2009) Curcumin suppresses proliferation and invasion in human gastric cancer cells by downregulation of PAK1 activity and cyclin D1 expression. Can. Biol. & Ther 8:1360–1368. [64] Cavaliere V, Papademetrio DL, Lorenzetti M, Valva P, Preciado MV, Gargallo P, Larripa I, Monreal MB, et al. (2009) Caffeic Acid Phenylethyl Ester and MG-132 Have Apoptotic and Antiproliferative Effects on Leukemic Cells But Not on Normal Mononuclear Cells. Transl Oncol. ;2(1):46-58. [65] Wang D, Xiang DB, He YJ, Li ZP, Wu XH, Mou JH, Xiao HL, Zhang QH (2005) Effect of caffeic acid phenethyl ester on proliferation and apoptosis of colorectal cancer cells in vitro. Wor. J Gastro. 14;11(26):4008-4012 [66] Sternlicht MD, Werb Z (2001) How matrix metalloproteinases regulate cell behavior. Annual Review of Cell and Dev. Biol.. 17:463–516. [67] Jiang MC, Liao CF, Lee PH (2001) Aspirin inhibits matrix metalloproteinase-2 activity, increases Ecadherin production, and inhibits in vitro invasion of tumor cells. Bioch. and Bioph. Res. Commu.. 282:671–677. p [68] Aimes RT, Quigley JP (1995) Matrix metalloproteinase-2 is an interstitial collagenase. Inhibitor-free enzyme catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific 3/4- and 1/4-length fragments. Biol.Chem. J . 270:5872– 5876. [69] Kleiner DE Jr, Stetler-Stevenson WG (1993) Structural biochemistry and activation of matrix metalloproteases. Current Opinion in Cell Bio. ;5:891–897. [70] Lochter A, Bissell MJ (1999) An odyssey from breast to bone: Multi-step control of mammary metastases and osteolysis by matrix metalloproteinases. APMIS 107:128–136. [71] Davidson B, Goldberg I, Liokumovich P, Kopolovic J, Gotlieb WH et al., (1998) Expression of metalloproteinases and their inhibitors in adenocarcinoma of the uterine cervix.Intern. J. of Gyn. Path..17:295–301. 4. References Carcinogenesis 316 Carcinogenesis 316 [72] Kugler A, Hemmerlein B, Thelen P, Kallerhoff M, Radzun HJ, Ringert RH(1998) Expression of metalloproteinase 2 and 9 and their inhibitors in renal cell carcinoma. J d’Urologie.160:1914–1918. [73] Hashimoto K, Kihira Y, Matuo Y, Usui T (1998) Expression of matrix metalloproteinase- 7 and tissue inhibitor of metalloproteinase-1 in human prostate. J d’Urologie. 160:1872– 1876. [74] Sutinen M, Kainulainen T, Hurskainen T, Vesterlund E, Alexander JP, Overall CM, et al.(1998) Expression of matrix metalloproteinases (MMP-1 and -2) and their inhibitors (TIMP-1, -2 and -3) in oral lichen planus, dysplasia, squamous cell carcinoma and lymph node metastasis. B. J. of Can.. 77:2239–2245. [75] Gonzalez-Avila G, Iturria C, Vadillo F, Teran L, Selman M, Perez-Tamayo R (1998) 72- kD (MMP-2) and 92-kD (MMP-9) type IV collagenase production and activity in different histologic types of lung cancer cells. Pathob. J. . 66:5–16. [76] Nawrocki B, Polette M, Marchand V, Monteau M, Gillery P, Tournier JM, et al. (1997) Expression of matrix metalloproteinases and their inhibitors in human bronchopulmonary carcinomas: Quantificative and morphological analyses. Intern. J.of Can.. 72:556–564. [77] Bramhall SR (1997) The matrix metalloproteinases and their inhibitors in pancreatic cancer. From molecular science to a clinical application. International Journal of Pancreatology 21:1–12. gy [78] Chapman HA, Riese RJ, Shi GP (1997) Emerging roles for cysteine proteases in human biology. Ann. Rev. of Phys.. 59:63–88. [79] Andreasen PA, Kjoller L, Christensen L, Duffy MJ (1997) The urokinase-type plasminogen activator system in cancer metastasis: A review. Intern.J. of Can.. 72:1–22. [80] Nerlov C, Rorth P, Blasi F, Johnsen M (1991) Essential AP-1 and PEA3 binding elements in the human urokinase enhancer display cell type-specific activity. Onco.6:1583–1592. [81] Lengyel E, Gum R, Stepp E, Juarez J, Wang H, Boyd D (1996) Regulation of urokinase- type plasminogen activator expression by an ERK1-dependent signaling pathway in a squamous cell carcinoma cell line. J. of Cell.Bio. 61:430–443. [82] Wang Y (2001) The role and regulation of urokinase-type plasminogen activator receptor gene expression in cancer invasion and metastasis. Medi. Res. Rev.. 21:146–170. [83] Hwang HJ, Park HJ, Chung HJ, et al. (2006) “Inhibitory effects of caffeic acid phenethyl ester on cancer cell metastasis mediated by the down-regulation of matrix metalloproteinase expression in human HT1080 fibrosarcoma cells,” J. of Nut. Bio..17(5): 356–362. [84] Folkman J (2007) Angiogenesis: An organizing principle for drug discovery? Nature Reviews. Drug Disc. 6:273–286. 4. References [85] Fan TP, Yeh JC, Leung KW, Yue PY, Wong RN (2006) Angiogenesis: From plants to blood vessels. Trends in Pharma. Sci.. 27:297–309. [86] Gordon MS, Mendelson DS, Kato G (2010) Tumor angiogenesis and novel antiangiogenic strategies. Intern. J. of Can.. 126:1777–1787. [87] Schuch G, Oliveira-Ferrer L, Loges S, Laack E, Bokemeyer C, Hossfeld DK, Fiedler W, Ergun S. (2005) Antiangiogenic treatment with endostatin inhibits progression of AML in vivo.Leuk.;19(8):1312-1317.
https://openalex.org/W3158440466	https://misinforeview.hks.harvard.edu/wp-content/uploads/2021/04/literat_youth_media_literacy_social_media_20210428.pdf	English	null	Research note: Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign on social media	null	2,021	cc-by	5,724	Research note: Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign on social media To better understand youth attitudes towards media literacy education on social media, and the opportunities and challenges inherent in such initiatives, we conducted a large-scale analysis of user responses to a recent media literacy campaign on TikTok. We found that reactions to the campaign were mixed, and highly political in nature. While young people appreciated the urgency of media literacy education and understood its relevance to their social media participation, many displayed a sarcastic attitude, criticizing both the content and the dissemination of the campaign. Based on these responses, we identify key takeaways and recommendations that can valuably inform future media literacy campaigns on social media. Authors: Ioana Literat (1), Abubakr Abdelbagi (1), Nicola YL Law (1), Marcus Y-Y Cheung (1), Rongwei Tang (1) Affiliations: (1) Teachers College, Columbia University, USA How to cite: Literat, I., Abdelbagi, A., Law, N. YL., Cheung, M. Y-Y., & Tang, R. (2021). Research note: Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign on social media. Harvard Kennedy School (HKS) Misinformation Review, 2(3). How to cite: Literat, I., Abdelbagi, A., Law, N. YL., Cheung, M. Y-Y., & Tang, R. (2021). Research note: Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign on social media. Harvard Kennedy School (HKS) Misinformation Review, 2(3). Received: January 27th, 2021. Accepted: March 22nd, 2021. Published: April 28th, 2021. Received: January 27th, 2021. Accepted: March 22nd, 2021. Published: April 28th, 2021. Research questions q • How do users respond to a platform-initiated media literacy education campaign on youth- oriented social media? • How do users respond to a platform-initiated media literacy education campaign on youth- oriented social media? • What lessons can we draw from these responses in terms of targeted media literacy education initiatives for youth? • What lessons can we draw from these responses in terms of targeted media literacy education initiatives for youth? 1 A publication of the Shorenstein Center for Media, Politics and Public Policy at Harvard University, John F. Kennedy School of Government. Harvard Kennedy School Misinformation Review1 April 2021, Volume 2, Issue 3 Creative Commons Attribution 4.0 International (CC BY 4.0) Reprints and permissions: misinforeview@hks.harvard.edu DOI: https://doi.org/10.37016/mr-2020-67 Website: misinforeview.hks.harvard.edu Harvard Kennedy School Misinformation Review1 April 2021, Volume 2, Issue 3 Creative Commons Attribution 4.0 International (CC BY 4.0) Reprints and permissions: misinforeview@hks.harvard.edu DOI: https://doi.org/10.37016/mr-2020-67 Website: misinforeview.hks.harvard.edu Research Note g y How to cite: Literat, I., Abdelbagi, A., Law, N. YL., Cheung, M. Y-Y., & Tang, R. (2021). Research note: Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign on social media. Harvard Kennedy School (HKS) Misinformation Review, 2(3). h d h Implications Amidst an alarming rise in online misinformation (McGinty & Gyenes, 2020; Pasquetto et al., 2020), there has been a growing interest in media literacy education as a way to combat the rapid spread of misinformation (Bulger & Davison, 2018; Chang et al., 2020; Roozenbeek & Van Der Linden, 2020; Tully, et al., 2020). However, the focus of both research and practice has been on the integration of media literacy instruction into young people’s various educational experiences, rather than implementing and evaluating such initiatives in situ, on social media (Bulger & Davison, 2018; Pasquetto et al., 2020)—which is known to be their central news source (Common Sense Media, 2019). Indeed, often due to the lack of necessary data (see Pasquetto et al., 2020), there has been insufficient attention devoted to assessing media literacy initiatives on social media; furthermore, existing studies focus on older platforms like Facebook or Twitter (e.g., Guess et al., 2019; Vraga & Tully, 2021), and we lack a nuanced understanding of these dynamics on the platforms most popular with youth today. Given this context, our study aimed to facilitate a bottom-up and naturalistic study of user responses to a media literacy education campaign on TikTok, an increasingly popular youth-oriented social media platform. Based on a large-scale qualitative analysis of user responses to this media literacy education campaign, we find that user reactions were mixed, and highly political in nature. On the positive side, our findings show that young people appreciate the urgency of media literacy education and understand its relevance to social media and to their own online habits; this attitude can be seen as an opening for future initiatives and a solid foundation to build off of. While critical and sarcastic comments abounded, this is to be expected on social media in general (Muresan et al., 2016; Zappavigna, 2012), and even more so when it comes to educational topics (Literat, 2021). Significantly, the more negative responses to this campaign can serve as rich feedback for the design of future media literacy initiatives for youth, particularly on social media. Below, we tease out a few of these directions. Research note summary • This study analyzed how users respond to a platform-initiated media literacy education campaign on youth-oriented social media, with a focus on the lessons that we might draw from these responses in terms of targeted media literacy education initiatives for youth. • This study analyzed how users respond to a platform-initiated media literacy education campaign on youth-oriented social media, with a focus on the lessons that we might draw from these responses in terms of targeted media literacy education initiatives for youth. • Our large-scale qualitative content analysis—which included 11,449 public comments posted on 5 TikTok videos—facilitated a naturalistic study of young people’s attitudes towards such initiatives. • Our large-scale qualitative content analysis—which included 11,449 public comments posted on 5 TikTok videos—facilitated a naturalistic study of young people’s attitudes towards such initiatives. Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign 2 2 • While many users appreciated the media literacy campaign and TikTok’s role in implementing it, there was also resistance to the campaign, due to factors related to video content, dissemination approach, and the very identity of TikTok as a platform. Across the board, comments were highly political, illustrating the politicized nature of media literacy education today. • While many users appreciated the media literacy campaign and TikTok’s role in implementing it, there was also resistance to the campaign, due to factors related to video content, dissemination approach, and the very identity of TikTok as a platform. Across the board, comments were highly political, illustrating the politicized nature of media literacy education today. p g p y y • These reactions serve as rich feedback that can usefully inform future media literacy campaigns, and we outline a set of recommendations in this respect. • These reactions serve as rich feedback that can usefully inform future media literacy campaigns, and we outline a set of recommendations in this respect. Key takeaways and recommendations First, it is clear that youth were keenly aware of TikTok’s “rap sheet” in regard to curbing misinformation (or not), and that this perception—shaped by both micro-level and macro-level factors (e.g., personal experience with reporting/moderation on the platform, mass media reports, political rhetoric like Trump’s anti-TikTok discourse)—had a strong impact on their attitude towards the campaign. Commenters in this study openly rebuked TikTok for sharing educational materials addressing fake news while failing to curb the spread of misinformation on their platform, and this perceived hypocrisy generated a negative response to the campaign. This sense of discontent with content moderation on TikTok resulted in a diminished level of trust, which negatively affected user engagement with the campaign videos. As Brunk and colleagues (2019) found, providing explicit reasoning around content moderation practices promotes a higher level of user trust, and might thus alleviate any communicative hindrances to learning. Therefore, in order to ensure the effectiveness of such educational initiatives, we recommend that campaign creators address any perceived inconsistencies between action and rhetoric. In the case of platform-run initiatives, Literat; Abdelbagi; Law; Cheung; Tang 3 this might translate to increased transparency in communicating how decisions are made regarding user reports of misinformation or content violations. Much of the negative feedback to the campaign can be attributed to the fact that it was perceived— in both promotion and design/content—as being at odds with the ethos and function of TikTok in youth lives. Youth engage on social networks such as TikTok with specific expectations related to both the aesthetic and social nature of their participation on the platform and are quick to sniff out content they perceive as inauthentic (boyd, 2014). While TikTok’s campaign attempted to meet these expectations (e.g., by using humor and leveraging TikTok microcelebrities to act out these video skits), users did not connect with the language and style employed in the videos. This might be understood in relation to the shift in TikTok users’ aesthetic appetite, which is moving away from “visual theatrics” and towards more relatable content that taps into the current trends and latest memes (Abidin, 2021). A key way to preempt this pitfall and facilitate a sense of relevance and nowness is to include the main stakeholders, youth themselves, in the design of such initiatives. Key takeaways and recommendations Participatory design allows participants to enact their identities, providing valuable insight into the characteristics, perceptions, and expectations of the target population (Coenraad et al., 2019); significantly, in the media literacy sphere, it has been shown to be an effective strategy for the design of news literacy initiatives for youth (see Literat et al., 2020). Regarding campaign promotion and dissemination, we found that youth were extremely frustrated with how the videos were being pushed, often repeatedly, onto their feeds. This essentially begs the question of how—or whether—platforms can promote such educational content without it being negatively perceived or dismissed for its top-down, forceful, unsolicited approach? In line with our previous recommendation, and as a way to facilitate a more organic reach, we recommend that platforms consider utilizing more participatory, community-oriented channels. This could take the form, for instance, of initiating participatory activities (e.g., TikTok challenges), thereby distributing the ownership of the campaign and encouraging more user buy-in, and/or engaging influencers in the promotion of content. Indeed, recent initiatives leveraging TikTok influencers in the promotion of educational campaigns, including around COVID-19 messaging, highlight “the efficacy of influencers in information dissemination and control in the online space” (Abidin et al., 2021, p. 124). Another key takeaway from this research is the highly political nature of user responses to this campaign. Certainly, this political focus was also influenced by the timing of the videos, given the prominence of the Black Lives Matter movement and the presidential election in this timeframe. However, it is a powerful illustration of the politicized nature of media literacy education in the current sociocultural landscape. Designers of future media literacy initiatives—on social media and elsewhere—need to keep in mind that, in this landscape, even apolitical content might be politicized and framed in polarizing terms by users. More broadly, these findings also reinforce the link between media literacy and civic education (Mihailidis, 2018; Mihailidis & Thevenin, 2013), and the fact that, particularly in this political and technological context, each is a prerequisite for the other. “ok but this tip is awesome because it encourages critical thinking” Positive responses: Understanding and embracing the educational aims of the videos On the positive side, users expressed appreciation for the aims of this campaign, and for TikTok taking this initiative. Many users seemed to understand and appreciate the educational aims of the videos, and complimented the videos for being “helpful,” “cool,” “educational,” “timely,” “engaging,” and “necessary:” “ok but this tip is awesome because it encourages critical thinking” Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign 4 “basically telling you to do your own research! that's refreshing to hear from a social media platform” Significantly, there was evidence of self-reflection and perceived usefulness among these responses: “wow you're right, my media has always been left leaning, i see all my favorite celebrities pushing a left leaning agenda, maybe that's a scary bias!” "Me ready to send this to my transphobic family" Users also understood the urgency and significance of media literacy education more largely. They often thanked TikTok for taking this initiative, sometimes as part of a positive comparison with other social media platforms: "Yeahhhhhhh this is so necessary. Thank you tik tok for doing this." “this a very important topic thank you tiktok for doing more than most social media cites” “Facebook should learn a thing or two from you. I deleted my FB account” Other comments implied that these videos on TikTok were just as effective or even more effective than the media literacy instruction they received in school: “Ok but im in college and why am I finally understanding this ” “Who needs school when TikTok teaches us something new every day” "This short video was more effective than my mandatory 5 hour “unconscious bias” training " “Why can’t my history teachers give me this instead of an Intro To Trusted Sourcing lesson” Negative responses: Confusion, annoyance, disconnection and perceived hypocrisy However, negative comments—often sarcastic in tone—far outweighed the more positive takes above. Indeed, the corpus included about twice as many negative comments than positive ones—though, as mentioned above, this balance should be taken with a grain of salt, given the nature of social media expression, compounded by the educational focus of the initiative. These negative reactions were grounded in: confusion about the aim of the videos; annoyance at videos being pushed on users’ “for you page” (FYP);2 disliking the video content and aesthetics; and perceived hypocrisy due to TikTok’s own actions and policies. 2 TikTok’s “For You Page” is an algorithmically curated feed that functions as the platform’s homepage. "I swear if I see one more video from this acc I'll just delete tiktok all together- " Trust in the effectiveness of the reporting feature and the enforcement of community guidelines seemed low: Yes, except reporting TikToks does no good because y’all pick and choose what guidelines to withho “Yes, except reporting TikToks does no good because y’all pick and choose what guidelines to withhold.” “i reported a video of a dude throwing stuff at homeless people and laughing and they said they found nothing wrong with it” “[The reporting] feature is functionally worthless. Doesn’t matter what you report - it goes through a bot and gets spat right back out” “[The reporting] feature is functionally worthless. Doesn’t matter what you report - it goes through a bot and gets spat right back out” “Also, when you report false or misleading information nothing happens because it doesn’t violate their community guidelines.” “Also, when you report false or misleading information nothing happens because it doesn’t violate their community guidelines.” Along the same lines, others brought up TikTok’s practice of over-/mis-censoring or shadowbanning users, an accusation that was often politicized along partisan or racial lines (e.g., “I’m more worried about tiktok continuing to silence black creators and uplift white creators who are appropriating black culture...”; “and why is BLM [Black Lives Matter] showdown ban [shadowbanned]”3). Finally, users also referenced TikTok’s business practices and connection to China—a government associated with heavy media censorship and misinformation. The latter perception was undoubtedly shaped by Donald Trump’s recent discourse and efforts to ban TikTok. "I swear if I see one more video from this acc I'll just delete tiktok all together- " Others disliked the approach that the videos took, for the following reasons: 1) use of outdated language and aesthetics (e.g., “are these targeted at millennials because I'm getting 90s nostalgia”; “What kind of early 2000s commercial is this”); 2) videos were perceived as too didactic, and incongruent with the perceived function of TikTok in youth lives (e.g., “This is TIKTOK not school so bye”; “ok i’m on my phone during class so i don’t have to do class and yet tik tok is showing me how to identify the source”); 3) the educational content was seen as too basic (e.g., “I learned this is 3rd grade”; “Why is this here? Kindergarten lesson”); and 4) videos were perceived as too boring (e.g., “I- I just wanna know what level of bored”; “There’s no way that was only a minute I feel like I just sat through orientation ”). Others disliked the approach that the videos took, for the following reasons: 1) use of outdated language and aesthetics (e.g., “are these targeted at millennials because I'm getting 90s nostalgia”; “What kind of early 2000s commercial is this”); 2) videos were perceived as too didactic, and incongruent with the perceived function of TikTok in youth lives (e.g., “This is TIKTOK not school so bye”; “ok i’m on my phone during class so i don’t have to do class and yet tik tok is showing me how to identify the source”); 3) the educational content was seen as too basic (e.g., “I learned this is 3rd grade”; “Why is this here? Kindergarten lesson”); and 4) videos were perceived as too boring (e.g., “I- I just wanna know what level of bored”; “There’s no way that was only a minute I feel like I just sat through orientation ”). Interestingly, some users were resistant to this media literacy campaign just based on their existing perceptions of TikTok’s actions regarding media literacy. In particular, they called out TikTok’s failure to control the spread of misinformation on its platform; here, many users brought up stories of reporting questionable content, with TikTok taking no action about it. e also Rosenblatt (2021) for the wider context of race-based shadowbanning and content suppression on TikTo “i swear if u put this on my fyp 9272836292539352935936 times-” “i swear if u put this on my fyp 9272836292539352935936 times-” Positive responses: Understanding and embracing the educational aims of the videos Users were confused as to why they appeared on their FYP, as this kind of content did not seem to match their interests and regular FYP content. Some thought the videos were ads; interestingly, this pattern was most prominent on the videos about distinguishing facts from opinions and, respectively, about questioning graphics, as both of these videos used humorous ads for fake products as a way to anchor their narratives. A large number of users expressed their annoyance at being unable to block these videos, and being forced to watch them multiple times in a short period of time: Literat; Abdelbagi; Law; Cheung; Tang 5 “This is 4 times In a row I see these adds” The political nature of responses A key finding from this research study was that—whether positive, negative, or neutral—a large proportion of comments (roughly a third of the data corpus) were political, i.e., making reference to political actors, groups, or issues. Political comments were posted on all videos, but were especially prominent on the video about distinguishing facts from opinions, which proved to be a very politicized issue. Key topics of discussion included Black Lives Matter, partisanship, the presidential election, as well as feminist, LGBTQ, racial discrimination issues. Many commenters explicitly expressed a particular Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign 6 political identity (Democrat / Republican, pro/anti-Trump, liberal/conservative, left/right, etc.). Comments often included political slogans (e.g., “Trump 2020”, “BIDEN 2020”) and were not afraid to express political stances or challenging others’ views. A prevalent pattern was mentioning the opposing party or ideology in comments and saying that they need to watch these videos—thus implying that the other party/ideology is lacking media literacy skills: political identity (Democrat / Republican, pro/anti-Trump, liberal/conservative, left/right, etc.). Comments often included political slogans (e.g., “Trump 2020”, “BIDEN 2020”) and were not afraid to express political stances or challenging others’ views. A prevalent pattern was mentioning the opposing party or ideology in comments and saying that they need to watch these videos—thus implying that the other party/ideology is lacking media literacy skills: “@ all trump supporters should see this” “Liberals take notes” “Democrats seeing this 🦯 🦯 🦯 🦯 🦯” "people who watch fox news needs this" “the left ain't gonna like this one” “@ all trump supporters should see this” Significantly, the rationale behind the videos was politicized too, framing media literacy education as an inherently political endeavor: “Bias education? This is political.” “Bias education? This is political.” “No need to make it political” “I literally went to the comments and counted to 5 knowing I would find someone that brings up politics before I finished counting” At the same time, political comments usually got more responses (e.g., likes, replies) than non-political ones, and exchanges were longer and more engaged, though often impolite and uncivil. Methods This study aimed to analyze how users respond to a platform-initiated media literacy education campaign on youth-oriented social media, with a focus on the lessons that we might draw from these responses in terms of targeted media literacy education initiatives for youth. While there is broad societal consensus about the urgency and significance of media literacy education, especially for young people, we still lack a grounded, bottom-up understanding of youth attitudes towards such initiatives. Moreover, trying to capture these attitudes through self-report methods like surveys or interviews runs the risk of producing biased results, due to social desirability effects (especially strong when it comes to the topic of media literacy and misinformation; see Tully et al., 2020), as well as the normative pressures that shape the way youth might respond to adult questions about their educational experiences. Here, using qualitative Literat; Abdelbagi; Law; Cheung; Tang 7 content analysis on a large corpus of youth reactions to a media literacy campaign on social media facilitates a naturalistic study of young people’s attitudes towards such initiatives in situ, “in the wild.” Research site content analysis on a large corpus of youth reactions to a media literacy campaign on social media facilitates a naturalistic study of young people’s attitudes towards such initiatives in situ, “in the wild.” Research site Our research site, TikTok, is a social media platform where users can create, share and browse short looping videos. Users can edit their videos by adding music or soundtracks, visual effects, and filters. They can follow other users, like and comment on videos, and communicate via direct messages. Users can discover content by searching for specific tags, users or soundtracks, but are also exposed to a custom feed of recommended videos on their For You Page. The popularity of TikTok has grown significantly since its parent company, ByteDance, acquired its predecessor, musical.ly, in late 2018. In 2020, TikTok reached 45.4 million users in the United States (eMarketer, 2020a), and the audience growth rate is predicted to reach 15.2% in 2021 (eMarketer, 2020b). The app is particularly popular with youth, who make up 60% of its demographic in the U.S. (App Ape, 2020). The cultural impact of TikTok on youth lives is significant, as the platform shapes youth culture, crowns grassroots celebrities, and popularizes new music, dances, and cultural trends (Herrman, 2020; The Youth Lab, 2019). 4 In view of the data available and our methodological approach, we cannot ascertain the specific ages of users represented in this corpus of comments. However, based on the age demographics of TikTok and the content of the comments themselves (e.g., references to school, use of youth lingo, references to youth culture), we can infer that the vast majority of commenters are youth. Similarly, while the exact geographic location of commenters is unknown, the vast majority were likely from the United States, in view of the aims and distribution of the campaign (i.e., produced by TikTok US and targeting American TikTok users) as well as the content of comments (making reference to US politics, US education, etc.). 5 In coding for sarcasm, we maintained a bottom-up, grounded approach that considered both the content of each comment and its larger context of expression (including relationship to the video, to other user replies, and to relevant real-life events). We paid Data corpus This study relied on the qualitative content analysis of 11,449 public comments posted in response to five videos published by an official TikTok account (specific handle hidden for anonymization purposes). Created by TikTok US in partnership with external experts, the five videos were part of a strategic media literacy campaign designed to highlight—via the use of humorous narratives acted out by TikTok microcelebrities—skills like fact-checking, assessing the credibility of information and visuals, understanding bias, and distinguishing facts from opinions. The videos were very similar in terms of their comedic approach and aesthetics, making for a congruent campaign. They were extremely popular on the platform, accumulating a total of 53.8m views, 1.63m likes, and 11k shares; at the same time, it is worth noting that the high view count was likely boosted by their being disseminated from an official TikTok account and pushed onto users’ For You Pages.4 Methods In the United States, TikTok has been a key venue for youth political expression (see Herrman, 2020) and has drawn the ire of former President Donald Trump, who unsuccessfully tried to ban it in 2020. Data analysis The data corpus was analyzed qualitatively, using a thematic analysis approach (Braun & Clarke, 2006) in order to identify patterns and nuance in the data. Four researchers (authors 2-5 of this study) participated in the qualitative coding process. The first analytical step involved identifying comments that were related to the content and/or educational aims of the videos, and were thus relevant to understanding youth responses to this media literacy initiative. These comments were labeled with simple codes (e.g., “positive,” “confused,” “sarcastic,”5 “comparison with school,” “comparison with other social media,” Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign 8 “political,” etc.) that facilitated a birds-eye understanding of the data corpus. Then, in a second round of analysis, we read the comments again and identified subthemes in the data; finally, these subthemes were further synthesized and coalesced into the key findings presented above. “political,” etc.) that facilitated a birds-eye understanding of the data corpus. Then, in a second round of analysis, we read the comments again and identified subthemes in the data; finally, these subthemes were further synthesized and coalesced into the key findings presented above. attention to both the text of the comment, and any emojis or hashtags that might encourage a sarcastic reading. At the same time, we acknowledge the difficulties in objectively and definitively detecting sarcasm online (see, e.g., Muresan et al., 2015) and the inescapable subjectivity inherent in this process. Bibliography Abidin, C. (2021). Mapping internet celebrity on TikTok: Exploring attention economies and visibility labours. Cultural Science Journal, 12(1), 77–103. http://doi.org/10.5334/csci.140 Abidin, C. (2021). Mapping internet celebrity on TikTok: Exploring attention economies and visibility labours. Cultural Science Journal, 12(1), 77–103. http://doi.org/10.5334/csci.140 Abidin, C., Lee, J., Barbetta, T., & Miao, W. S. (2021, September 26). Influencers and COVID-19: Reviewing key issues in press coverage across Australia, China, Japan, and South Korea. Media International Australia, 178(1), 114–135. https://doi.org/10.1177/1329878X20959838 Abidin, C., Lee, J., Barbetta, T., & Miao, W. S. (2021, September 26). Influencers and COVID-19: Reviewing key issues in press coverage across Australia, China, Japan, and South Korea. Media International Australia, 178(1), 114–135. https://doi.org/10.1177/1329878X20959838 App Ape. (2020, July 22). Distribution of TikTok users in the United States as of June 2020, by age group. Statista. https://www.statista.com/statistics/1095186/tiktok-us-users-age/ boyd, d. (2014). It’s complicated: The social lives of networked teens. Yale University Press. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. http://doi.org/10.1191/1478088706qp063oa Brunk, J., Mattern, J., & Riehle, D.M. (2019). Effect of transparency and trust on acceptance of automatic online comment moderation systems. In IEEE 21st Conference on Business Informatics (CBI) (Vol. 1, pp. 429–435). IEEE. https://doi.org/10.1109/CBI.2019.00056 Bulger, M., & Davison, P. (2018). The promises, challenges, and futures of media literacy. Journal of Media Literacy Education, 10(1), 1–21. https://doi.org/10.23860/JMLE-2018-10-1-1 Bulger, M., & Davison, P. (2018). The promises, challenges, and futures of media literacy. Journal of Media Literacy Education, 10(1), 1–21. https://doi.org/10.23860/JMLE-2018-10-1-1 Chang, Y. K., Literat, I., Price, C., Eisman, J., Gardner, J., Chapman, A., & Truss, A. (2020). News literacy education in a polarized political climate: How games can teach youth to spot misinformation. Harvard Kennedy School (HKS) Misinformation Review. https://doi.org/10.37016/mr-2020-020 Chang, Y. K., Literat, I., Price, C., Eisman, J., Gardner, J., Chapman, A., & Truss, A. (2020). News literacy education in a polarized political climate: How games can teach youth to spot misinformation. Harvard Kennedy School (HKS) Misinformation Review. https://doi.org/10.37016/mr-2020-020 Coenraad, M., Palmer, J., Franklin, D., & Weintrop, D. (2019). Enacting identities: Participatory design as a context for youth to reflect, project, and apply their emerging identities. Proceedings of the 18th ACM International Conference on Interaction Design and Children. https://doi.org/10.1145/3311927.3323148 Coenraad, M., Palmer, J., Franklin, D., & Weintrop, D. (2019). Enacting identities: Participatory design as a context for youth to reflect, project, and apply their emerging identities. Bibliography Proceedings of the 18th ACM International Conference on Interaction Design and Children. https://doi org/10 1145/3311927 3323148 Common Sense Media. (2019). Teen news engagement: Key findings and toplines. https://www.commonsensemedia.org/sites/default/files/uploads/pdfs/2019_cs- sm_summarytoplines_release.pdf Common Sense Media. (2019). Teen news engagement: Key findings and toplines. https://www.commonsensemedia.org/sites/default/files/uploads/pdfs/2019_cs- sm_summarytoplines_release.pdf eMarketer. (2020a, February 26). Number of TikTok users in the United States from 2019 to 2024 (in millions) [Graph]. Statista. https://www.statista.com/statistics/1100836/number-of-us-tiktok- users/ eMarketer. (2020a, February 26). Number of TikTok users in the United States from 2019 to 2024 (in millions) [Graph]. Statista. https://www.statista.com/statistics/1100836/number-of-us-tiktok- users/ eMarketer. (2020b, February 1). Annual TikTok user growth in the United States from 2019 to 2024 [Graph]. Statista. https://www.statista.com/statistics/1100842/tiktok-us-user-growth/ eMarketer. (2020b, February 1). Annual TikTok user growth in the United States from 2019 to 2024 [Graph]. Statista. https://www.statista.com/statistics/1100842/tiktok-us-user-growth/ Guess, A., Nagler, J., & Tucker, J. (2019). Less than you think: Prevalence and predictors of fake news dissemination on Facebook. Science Advances, 5(1), eaau4586. https://doi.org/10.1126/sciadv.aau4586 Guess, A., Nagler, J., & Tucker, J. (2019). Less than you think: Prevalence and predictors of fake news dissemination on Facebook. Science Advances, 5(1), eaau4586. https://doi.org/10.1126/sciadv.aau4586 Herrman, J. (2020, June 28). TikTok is shaping politics. But how? The New York Times. https://www.nytimes.com/2020/06/28/style/tiktok-teen-politics-gen-z.html attention to both the text of the comment, and any emojis or hashtags that might encourage a sarcastic reading. At the same time, we acknowledge the difficulties in objectively and definitively detecting sarcasm online (see, e.g., Muresan et al., 2015) and the inescapable subjectivity inherent in this process. Literat; Abdelbagi; Law; Cheung; Tang 9 Literat, I. (2021, February 25). “Teachers act like we’re robots”: TikTok as a window into youth experiences of online learning during COVID-19. AERA Open, 7(1). https://doi.org/10.1177/2332858421995537 Literat, I., Chang, Y. K., & Hsu, S. Y. (2020, June 2). Gamifying fake news: Engaging youth in the participatory design of news literacy games. Convergence: The International Journal of Research into New Media Technologies, 26(3), 503–516. https://doi.org/10.1177%2F1354856520925732 McGinty, M., & Gyenes, N. (2020, May 12). A dangerous misinfodemic spreads alongside the SARS-COV- 2 pandemic [Editorial]. Harvard Kennedy School (HKS) Misinformation Review. https://misinforeview.hks.harvard.edu/article/a-misinfodemic-as-dangerous-as-sars-cov-2- pandemic-itself/ Mihailidis, P. (2018, February 6). Civic media literacies: Re-imagining engagement for civic intentionality. Learning, Media and Technology, 43(2), 152–164. Mihailidis, P. (2018, February 6). Civic media literacies: Re-imagining engagement for civic intentionality. Learning, Media and Technology, 43(2), 152–164. https://doi.org/10.1080/17439884.2018.1428623 Mihailidis, P., & Thevenin, B. (2013, May 31). Media literacy as a core competency for engaged citizenship in participatory democracy. American Behavioral Scientist, 57(11), 1611–1622. Bibliography https://doi.org/10.1177/0002764213489015 Muresan, S., Gonzalez‐Ibanez, R., Ghosh, D., & Wacholder, N. (2016). Identification of nonliteral language in social media: A case study on sarcasm. Journal of the Association for Information Science and Technology, 67(11), 2725–2737. https://doi.org/10.1002/asi.23624. Pasquetto, I., Swire-Thompson, B., Amazeen, M. A., Benevenuto, F., Brashier, N. M., Bond, R. M., Bozarth, L. C., Budak, C., Ecker, U. K. H., Fazio, L. K., Ferrara, E., Flanagin, A. J., Flammini, A., Freelon, D., Grinberg, N., Hertwig, R., Jamieson, K. H., Joseph, K., Jones, J. J. … Yang, K. C. (2020). Tackling misinformation: What researchers could do with social media data. Harvard Kennedy School (HKS) Misinformation Review. https://doi.org/10.37016/mr-2020-49 Roozenbeek, J., & Van Der Linden, S. (2020, November 6). Breaking Harmony Square: A game that “inoculates” against political misinformation. Harvard Kennedy School (HKS) Misinformation Review, 1(8). https://doi.org/10.37016/mr-2020-47 Roozenbeek, J., & Van Der Linden, S. (2020, November 6). Breaking Harmony Square: A game that “inoculates” against political misinformation. Harvard Kennedy School (HKS) Misinformation Review, 1(8). https://doi.org/10.37016/mr-2020-47 Rosenblatt, K. (2021, February 9). Months after TikTok apologized to Black creators, many say little has changed. NBC News. https://www.nbcnews.com/pop-culture/pop-culture-news/months-after- tiktok-apologized-black-creators-many-say-little-has-n1256726 Rosenblatt, K. (2021, February 9). Months after TikTok apologized to Black creators, many say little has changed. NBC News. https://www.nbcnews.com/pop-culture/pop-culture-news/months-after- tiktok-apologized-black-creators-many-say-little-has-n1256726 The Youth Lab. (2019). TikTok through the eyes of a 16 year old. https://www.theyouthlab.com/insights/tiktok-through-the-eyes-of-a-16-year-old The Youth Lab. (2019). TikTok through the eyes of a 16 year old. https://www.theyouthlab.com/insights/tiktok-through-the-eyes-of-a-16-year-old Tully, M., Vraga, E. K., & Bode, L. (2020). Designing and testing news literacy messages for social media. Mass Communication and Society, 23(1), 22–46. https://doi.org/10.1080/15205436.2019.1604970 Tully, M., Vraga, E. K., & Bode, L. (2020). Designing and testing news literacy messages for social media. Mass Communication and Society, 23(1), 22–46. https://doi.org/10.1080/15205436.2019.1604970 Vraga, E. K., & Tully, M. (2021). News literacy, social media behaviors, and skepticism toward information on social media. Information, Communication & Society, 24(2), 150–166. https://doi.org/10.1080/1369118X.2019.1637445 Vraga, E. K., & Tully, M. (2021). News literacy, social media behaviors, and skepticism toward information on social media. Information, Communication & Society, 24(2), 150–166. https://doi.org/10.1080/1369118X.2019.1637445 Zappavigna, M. (2012). Discourse of Twitter and social media: How we use language to create affiliation on the web. Bloomsbury Academic. Zappavigna, M. (2012). Discourse of Twitter and social media: How we use language to create affiliation on the web. Bloomsbury Academic. Likes, sarcasm and politics: Youth responses to a platform-initiated media literacy campaign 10 Funding There are no funding sources to declare. Copyright This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided that the original author and source are properly credited. Competing interests Co pe g e es s The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Ethics Ethics Due to the focus on youth, this study was submitted for ethics board review and was approved as Exempt (Category 4) by the Institutional Review Board of Teachers College, Columbia University (Protocol #21- 099). Data Availability Statement The raw data cannot be made available publicly as a downloadable corpus due to both the terms of our approved IRB Protocol, and the terms of service of TikTok.
https://openalex.org/W4321267423	https://khrife-journal.org/index.php/journal/article/download/521/588	Ukrainian	null	Peculiarities of Determining Affecting Properties of Small-bore Bullets After Ricochet	Teorìâ ta praktika sudovoï ekspertizi ì krimìnalìstiki	2,022	cc-by	8,243	Дослідницька робота Дослідницька робота 1 За правилами української мови від глагола «уражати» (у значенні ранити, убивати; по- рушувати життєдіяльність — про хвороби) дієприкметники утворити не можна, варіант «уражаючий» є калькою з російської: суфікси -уч-, -юч- узагалі не характерні для україн- ської мови. Доречніше послуговуватися прикметником «уразливий» (у значенні болю- чий). Проте, зважаючи на стале застосування в чинних нормативно-правових актах, що регулюють судово-експертну діяльність, згаданих помилкових словоформ, ми змушені оперувати у цій праці саме ними.— Ред. Постановка наукової проблеми рикошету кулі й хаотичністю фрагмен- тації кулі внаслідок її ударно-контакт- ної взаємодії із поверхнею перешкоди. У зв’язку із цим виникає потреба за- стосувати альтернативний критерій оцінювання уражаючих властивостей куль після рикошету та запровадити граничне значення цього параметра. Окрім того, у практиці судової балісти- ки необхідно започаткувати розрахун- ковий метод оцінювання уражаючих властивостей куль після рикошету, завдяки чому стане можливим спрог нозувати ступінь тяжкості шкоди, завданої життю та здоров’ю людини. Криміналістичне дослідження особ- ливостей ударно-контактної взаємодії куль, фрагментів куль чи окремих елементів конструкції після їх рико- шету від поверхонь відносно щільних перешкод із біологічними об’єктами є доволі складним завданням. Основ- ні труднощі виникають під час дослі дження уражаючих 1 властивостей цих об’єктів та оцінювання їх спромож- ності завдавати тілу людини тілесних ушкоджень різного ступеня тяжкості, оскільки відсутні науково обґрунтовані методики таких досліджень, а сучасна енергетична концепція ураження цілі, яку використовують у судовій балісти- ці, у таких випадках має доволі обмеже- не застосування. Передусім це зумовле- но неможливістю визначити початкові умови влучання дестабілізованої після Особливості визначення уражаючих властивостей малокаліберних куль після рикошету Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal влучання як самих куль, так і їх окремих фрагментів. Здобуті результати експериментальних досліджень і натурних ви пробувань дали змогу сформувати альтернативний критерій оцінювання ступеня тяжкості тілесних ушкоджень, яких лю- дина зазнала внаслідок влучання свинцевих малокаліберних куль після рикошету. Ним є довжина ранового каналу, спричи- неного кулею чи її фрагментами внаслідок їх влучання у тіло людини після рикошету. Запропоновано розрахунковий метод визначення довжини спричиненого ранового каналу, який мож- на застосовувати в експертній практиці для визначення по- тенційної шкоди життю та здоров’ю людини. влучання як самих куль, так і їх окремих фрагментів. Здобуті результати експериментальних досліджень і натурних ви пробувань дали змогу сформувати альтернативний критерій оцінювання ступеня тяжкості тілесних ушкоджень, яких лю- дина зазнала внаслідок влучання свинцевих малокаліберних куль після рикошету. Ним є довжина ранового каналу, спричи- неного кулею чи її фрагментами внаслідок їх влучання у тіло людини після рикошету. Запропоновано розрахунковий метод визначення довжини спричиненого ранового каналу, який мож- на застосовувати в експертній практиці для визначення по- тенційної шкоди життю та здоров’ю людини. Ключові слова: енергетична концепція ураження цілі; роз- рахунковий метод; уражаючі властивості; альтернативний критерій оцінювання; довжина ранового каналу; імітатор біо- логічних тканин; малокаліберна куля; рикошет. Постановка наукової проблеми Постановка наукової проблеми Особливості визначення уражаючих властивостей малокаліберних куль після рикошету Олександр Коломійцев * a, Віктор Сапєлкін b, Павел Ґіверц * c, Олександр Герман **** d * Канд. техн. наук, провідн. наук. співробітн., ННЦ «ІСЕ ім. Засл. проф. М. С. Бокаріуса», м. Харків, Україна, ORCID: https://orcid.org/0000-0003-1932-1034, e-mail: sashagun@ukr.net * Канд. техн. наук, провідн. наук. співробітн., ННЦ «ІСЕ ім. Засл. проф. М. С. Бокаріуса», м. Харків, Україна, ORCID: https://orcid.org/0000-0003-1932-1034, e-mail: sashagun@ukr.net g Канд. мед. наук, Харківська медична академія післядипломної освіти, м. Харків, Україна, ORCID: https://orcid.org/0000-0002-9795-8463, e-mail: bodydoc666@gmail.com Україна, ORCID: https://orcid.org/0000-0002-9795-8463, e-mail: bodydoc666@gmail.com * PhD з права, криміналістики і судової експертизи, Балістична лабораторія поліції Ізраїля, м. Єрусалим, Ізраїль, ORCID: https://orcid.org/0000-0002-4019-5055, e-mail: pavel.giverts@gmail.com p g @g ** Ст. наук. співробітн., ННЦ «ІСЕ ім. Засл. проф. М. С. Бокаріуса», м. Харків, Україна, ORCID: https://orcid.org/0000-0001-6522-9381, e-mail: herman30121966@gmail.com p g @g ** Ст. наук. співробітн., ННЦ «ІСЕ ім. Засл. проф. М. С. Бокаріуса», м. Харків, Україна, ORCID: https://orcid.org/0000-0001-6522-9381, e-mail: herman30121966@gmail.com a Методологія, концепція, написання оригінального проєкту. a Методологія, концепція, написання оригінального проєкту. b Кураторство даними, формальний аналіз. с Ресурси. с Ресурси. d Адміністрування проєкту, нагляд. d Адміністрування проєкту, нагляд. DOI: 10.32353/khrife.2.2022.05 УДК 343.983 Надійшло 07.06.2022 / Рецензовано 07.06.2022 / Прийнято до друку 17.06.2022 / Доступно онлайн 30.06.2022 Автори мали на меті узагальнити експериментальні дослідження уражаючих властивостей безоболонкових куль калібру .22 Long Rifle після їх рикошету від різних видів пере шкод. Запропоновано альтернативний критерій оцінювання уражаючих властивостей куль, за допомогою якого можливо прогнозувати ступінь і тяжкість тілесних ушкоджень після рикошету. Виявлено особливості й умови, за яких відбувається фрагментація безоболонкових малокаліберних куль, а також визначено швидкість польоту куль і їх фрагментів після удар- но-контактної взаємодії з перешкодами різних видів. Особли- ву увагу приділено дослідженню характеру взаємодії куль і їх фрагментів після рикошету із в’язко-пружним середовищем, що імітує біологічні тканини тіла людини. Установлено, що енергетична концепція ураження біологічної цілі та критерії її оцінювання під час дослідження уражаючих властивостей куль після рикошету мають обмежене застосування, оскіль- ки не завжди можна достеменно визначити початкові умови © 2022 Автор(и). Видано Національним науковим центром «Інститут судових експертиз ім. Засл. проф. М. С. Бокаріуса» та Національним юридичним університетом імені Ярослава Мудрого. Це стаття з відкритим доступом, розповсюджена на умовах Ліцензії атрибуції Creative Commons (CC_BY_4.0), що дає змогу необмежено використовувати, розповсюджувати й відтворювати на будь-якому носії за умови посилання на оригінального автора та джерела. Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Аналіз основних досліджень і публікацій 7 (3). Pp. 44—51 ; Haag M., Haag L. Shooting incident reconstruction. 2020. 560 p. DOI: 10.1016/C2018-0-03137-0 (дата звернення: 11.05.2022). p ( р ) 3 Rathman G. А. Bullet ricochet and associated phenomena. AFTE Journal. 1987. Vol. 19 (4). Pp. 374—381. 3 Rathman G. А. Bullet ricochet and associated phenomena. AFTE Journal. 1987. Vol. 19 (4). Pp. 374—381. p 4 Kotas Z., Reno C. E. A Case Study in the use of bullet ricochet experimentation to provide evidence in a homicide. Ibid. 2014. Vol. 46 (3). Pp. 211—217. 4 Kotas Z., Reno C. E. A Case Study in the use of bullet ricochet experimentation to provide evidence in a homicide. Ibid. 2014. Vol. 46 (3). Pp. 211—217. 5 Кофанов А. В. Теоретичні та практичні аспекти моделювання рикошетування вогнепаль- ного снаряда при проведені експертного експерименту. Криміналістика і судова експер- тиза : міжвідом. наук.-метод. зб. 2018. Вип. 63 (1). С. 310—322. URL: http://nbuv.gov.ua/UJRN/ krise_2018_63%281%29__36 (дата звернення: 11.05.2022). 5 Кофанов А. В. Теоретичні та практичні аспекти моделювання рикошетування вогнепаль- ного снаряда при проведені експертного експерименту. Криміналістика і судова експер- тиза : міжвідом. наук.-метод. зб. 2018. Вип. 63 (1). С. 310—322. URL: http://nbuv.gov.ua/UJRN/ krise_2018_63%281%29__36 (дата звернення: 11.05.2022). ( р ) 6 Thompson E., Wyant R. Asphalt skip shooting reconstruction. AFTE Journal. 2003. Vol. 35 (1). Pp. 74—758. 7 Коломійцев О. В. Метод розрахункового та експериментального визначення вигляду сна- ряда спеціального призначення із заданими аеродинамічними властивостями : дис. … канд. техн. наук. Харків, 2005. 234 с. 8 Гусенцов А. О., Артюшин А. А., Данилова В. А. Теория и практика создания модели ри- кошета огнестрельного снаряда в условиях лабораторного эксперимента. Судово-баліс- тичні дослідження та суміжні галузі знань : перш. міжнар. наук.-практ. семінар. Київ, 2009. С. 30—35. 9 Shepitko V., Shepitko M., Simakova-Yefremian E., Kolomiitsev A. The Determination of the Bullet Velocity After Ricochet Using Computer Modeling. AFTE Journal. 2020. Vol. 52 (2). Pp. 98—102. p 10 Moxnes J. F., Frøyland Ø., Skriudalen S. et al. On the study of ricochet and penetration in sand, water and gelatin by spheres, 7.62 mm APM2, and 25 mm projectiles. Defence Technology. 2016. Vol. 12. Pp. 159—170. DOI: 10.1016/j.dt.2015.12.004 (дата звернення: 29.11.2021). 11 Эйдлин Л. М. Огнестрельные повреждения. Врачебное и криминалистическое распозна- вание и оценка. Ташкент, 1963. С. 31—109 ; Попов В. Л., Шигеев В. Б., Кузнецов Л. Е. Су- дебно-медицинская баллистика. Санкт-Петербург, 2002. 656 с. URL: https://djvu.online/file/ hkakzjDMC5I3P (дата звернення: 29.11.2021). 12 Озерецковский Л. Б., Гуманенко Е. Аналіз основних досліджень і публікацій Аналіз основних досліджень і публікацій Дослідженню рикошету приділили доволі значну увагу сучасні фахів- ці в галузі судової балістики, зокрема 1 За правилами української мови від глагола «уражати» (у значенні ранити, убивати; по- рушувати життєдіяльність — про хвороби) дієприкметники утворити не можна, варіант «уражаючий» є калькою з російської: суфікси -уч-, -юч- узагалі не характерні для україн- ської мови. Доречніше послуговуватися прикметником «уразливий» (у значенні болю- чий). Проте, зважаючи на стале застосування в чинних нормативно-правових актах, що регулюють судово-експертну діяльність, згаданих помилкових словоформ, ми змушені оперувати у цій праці саме ними.— Ред. 60 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 M. та L. Haag 2, G. А. Rathman 3, Z. Kotas і C. E. Reno 4, А. В. Кофанов 5 та ін. Проте вони мають переважно практичну спря- мованість, пов’язану з експерименталь- ним відтворенням умов рикошетування куль, або розглядають тільки наслідки реальних випадків рикошету куль від поверхонь різних видів перешкод (зокре- ма, E. Thompson і R. Wyant 6, О. В. Коломій- цев 7, А. О. Гусенцов, А. А. Артюшин і В. А. Данилова 8). Незначну кількість наукових робот присвячено визначен- ню аеробалістичних характеристик куль після рикошету 9, а також комп’ютерно- му моделюванню фізичних процесів, що відбуваються під час рикошету різних видів уражаючих елементів 10. Окремим напрямом є дослідження наслідків рикошету куль, що викладено у працях фахівців судової медицини 11. Провідною темою цих робіт є визначен- ня комплексу ознак вогнепальних по- ранень, характерних для випадків ри- кошету куль від різних видів перешкод. Іншою галуззю досліджень рикошету куль різних калібрів є ранова балісти- ка 12, що визначає бойову ефективність дії по живій цілі. Значну частину на укових праць присвячено дослідженню фізичних процесів, що відбуваються під 2 Haag L. C. Bullet Ricochet: An empirical study and a device for measuring ricochet angle. AFTE Journal. 1975. Vol. 7 (3). Pp. 44—51 ; Haag M., Haag L. Shooting incident reconstruction. 2020. 560 p. DOI: 10.1016/C2018-0-03137-0 (дата звернення: 11.05.2022). 2 Haag L. C. Bullet Ricochet: An empirical study and a device for measuring ricochet angle. AFTE Journal. 1975. Vol. 7 (3). Pp. 44—51 ; Haag M., Haag L. Shooting incident reconstruction. 2020. 560 p. DOI: 10.1016/C2018-0-03137-0 (дата звернення: 11.05.2022). 2 Haag L. C. Bullet Ricochet: An empirical study and a device for measuring ricochet angle. AFTE Journal. 1975. Vol. 2 Haag L. C. Bullet Ricochet: An empirical study and a device for measuring ricochet angle. AFTE Journal. 1975. Vol. 7 (3). Pp. 44—51 ; Haag M., Haag L. Shooting incident reconstruction. 2020. 560 p. DOI: 10.1016/C2018-0-03137-0 (дата звернення: 11.05.2022). Викладення основного матеріалу дослідження Викладення основного матеріалу дослідження Більшість згаданих наукових праць присвячено дослідженню оболонкових куль зі сталевим чи свинцевим осердям. Однак уражаючим властивостям куль малокаліберних патронів кільцевого запалення (зокрема, .22 Short і .22 Long Rifle, далі — .22 LR), приділено недостат- ньо уваги, незважаючи на те, що вони є найбільш поширеними і найчастіше стають об’єктами судово-балістичної й судово-медичної експертиз. Для експериментальних досліджень використано 5,6 мм спортивні патрони кільцевого запалення (.22 LR), споря джені безоболонковою свинцевою ку- лею, та 5,6 мм пістолети моделей Grand Power К22 S і К22 X-TRIM (див. рис. 1—3). Рис. 1. 5,6 мм спортивні патрони кільцевого запалення Мета статті Експериментальне дослідження ура- жаючих властивостей безоболонкових куль калібру .22 LR після їх рикошету від поверхонь різних видів перешкод. На Рис. 1. 5,6 мм спортивні патрони кільцевого запалення 13 Jauhari M. Bullet ricochet from metal plates. The Journal of Criminal Law, Criminology, and Police Science. 1969. Vol. 60 (3). Pp. 387—394. DOI: 10.2307/1141995 (дата звернення: 29.11.2021) ; Burke Т. W., Rowe W. F. Bullet ricochet: A comprehensive revive. Journal of Forensic Sciences. 1992. Vol. 37 (5). Pp. 1254—1260. DOI: 10.1520/JFS13312J (дата звернення: 29.11.2021) ; Karget B., Joosten U. A case of «boomerang» bullet ricochet. International Journal of Legal Medicine. 2001. Vol. 115 (2). Pp. 70—71. DOI: 10.1007/s004140000148 (дата звернення: 29.11.2021) ; Білен- ко О. І., Кириченко О. О., Павлов Д. В. Дослідження впливу матеріалу метального еле- менту на характеристики рикошету. Збірник наукових праць Національної гвардії України. 2017. Вип. 2 (30). С. 15—21. URL: http://nbuv.gov.ua/UJRN/znpavs_2017_2_5 (дата звернення: 29.11.2021) ; Гиверц П. В. Криминалистическое исследование винтовок AIRSOFT, модер- низированных для стрельбы патронами калибра 5,56 × 45 мм. Теорія та практика судової експертизи і криміналістики. 2018. Вип. 18. С. 301—308. DOI: 10.32353/khrife.2018.34 (дата звернення: 29.11.2021). 13 Jauhari M. Bullet ricochet from metal plates. The Journal of Criminal Law, Criminology, and Police Science. 1969. Vol. 60 (3). Pp. 387—394. DOI: 10.2307/1141995 (дата звернення: 29.11.2021) ; Burke Т. W., Rowe W. F. Bullet ricochet: A comprehensive revive. Journal of Forensic Sciences. 1992. Vol. 37 (5). Pp. 1254—1260. DOI: 10.1520/JFS13312J (дата звернення: 29.11.2021) ; Karget B., Joosten U. A case of «boomerang» bullet ricochet. International Journal of Legal Medicine. 2001. Vol. 115 (2). Pp. 70—71. DOI: 10.1007/s004140000148 (дата звернення: 29.11.2021) ; Білен- ко О. І., Кириченко О. О., Павлов Д. В. Дослідження впливу матеріалу метального еле- менту на характеристики рикошету. Збірник наукових праць Національної гвардії України. 2017. Вип. 2 (30). С. 15—21. URL: http://nbuv.gov.ua/UJRN/znpavs_2017_2_5 (дата звернення: 29.11.2021) ; Гиверц П. В. Криминалистическое исследование винтовок AIRSOFT, модер- низированных для стрельбы патронами калибра 5,56 × 45 мм. Теорія та практика судової експертизи і криміналістики. 2018. Вип. 18. С. 301—308. DOI: 10.32353/khrife.2018.34 (дата звернення: 29.11.2021). 14 Петрученко О. С. Щодо вивчення взаємодії кулі чи осколка снаряда з перешкодою. Збірник наукових праць Військової академії (м. Одеса). Технічні науки. 2015. Вип. 2 (4). С. 74—81. URL: http://nbuv.gov.ua/UJRN/zbnpva_2015_2_13 (дата звернення: 29.11.2021) ; Петрученко О. С., Флюд О. В., Білаш О. В., Величко Л. Д. Дослідження співудару твердих тіл. VII Науково- практична конференція : зб. тез допов. Харків, 2017. С. 125 ; Біленко О. р 14 Петрученко О. С. Щодо вивчення взаємодії кулі чи осколка снаряда з перешкодою. Збірник наукових праць Військової академії (м. Одеса). Технічні науки. 2015. Вип. 2 (4). С. 74—81. URL: http://nbuv.gov.ua/UJRN/zbnpva_2015_2_13 (дата звернення: 29.11.2021) ; Петрученко О. С., Флюд О. В., Білаш О. В., Величко Л. Д. Дослідження співудару твердих тіл. VII Науково- практична конференція : зб. тез допов. Харків, 2017. С. 125 ; Біленко О. І., Кириченко О. О., Павлов Д. В. Методика зниження енергетичних характеристик поражаючого елемента після відбиття від перешкоди. Science Rice. 2018. Vol. 4 (45). P. 14—18. DOI: 10.15587/2313- 8416.2018.129326 (дата звернення: 29.11.2021). 13 Jauhari M. Bullet ricochet from metal plates. The Journal of Criminal Law, Criminology, and Police Science. 1969. Vol. 60 (3). Pp. 387—394. DOI: 10.2307/1141995 (дата звернення: 29.11.2021) ; Burke Т. W., Rowe W. F. Bullet ricochet: A comprehensive revive. Journal of Forensic Sciences. 1992. Vol. 37 (5). Pp. 1254—1260. DOI: 10.1520/JFS13312J (дата звернення: 29.11.2021) ; Karget B., Joosten U. A case of «boomerang» bullet ricochet. International Journal of Legal Medicine. 2001. Vol. 115 (2). Pp. 70—71. DOI: 10.1007/s004140000148 (дата звернення: 29.11.2021) ; Білен- ко О. І., Кириченко О. О., Павлов Д. В. Дослідження впливу матеріалу метального еле- менту на характеристики рикошету. Збірник наукових праць Національної гвардії України. 2017. Вип. 2 (30). С. 15—21. URL: http://nbuv.gov.ua/UJRN/znpavs_2017_2_5 (дата звернення: 29.11.2021) ; Гиверц П. В. Криминалистическое исследование винтовок AIRSOFT, модер- низированных для стрельбы патронами калибра 5,56 × 45 мм. Теорія та практика судової експертизи і криміналістики. 2018. Вип. 18. С. 301—308. DOI: 10.32353/khrife.2018.34 (дата звернення: 29.11.2021). Аналіз основних досліджень і публікацій К., Бояринцев В. В. Раневая баллистика. Санкт-Петер- бург, 2006. 374 с. ; Денисов А. В., Тюрин М. В., Сохранов М. В. и др. Особенности пораже- ния живых целей в зоне рикошета пуль при стрельбе по твёрдым преградам. Вестник Российской военно-медицинской академии. 2014. № 1 (45). С. 179—184. 61 Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal час ударно-контактної взаємодії куль із поверхнями перешкод (здебільшого ме- талевих 13). Водночас дослідженню ура- жаючих характеристик як самих куль, так і їх фрагментів після рикошету при- ділено вкрай недостатньо уваги: лише окремі роботи пов’язані з дослідженням енергетичних характеристик уража ючих елементів після рикошету 14. основі здобутих результатів натурних випробувань запропонувати альтерна- тивний критерій оцінювання їх ура- жаючих властивостей, що дасть змогу прогнозувати ступінь і тяжкість тілес- них ушкоджень, завданих людині кулею або її фрагментами внаслідок рикошету. час ударно-контактної взаємодії куль із поверхнями перешкод (здебільшого ме- талевих 13). Водночас дослідженню ура- жаючих характеристик як самих куль, так і їх фрагментів після рикошету при- ділено вкрай недостатньо уваги: лише окремі роботи пов’язані з дослідженням енергетичних характеристик уража ючих елементів після рикошету 14. Мета статті І., Кириченко О. О., Павлов Д. В. Методика зниження енергетичних характеристик поражаючого елемента після відбиття від перешкоди. Science Rice. 2018. Vol. 4 (45). P. 14—18. DOI: 10.15587/2313- 8416.2018.129326 (дата звернення: 29.11.2021). 62 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Рис. 2. Правий і лівий боки 5,6 мм пістолета Grand Power К22 S Рис. 2. Правий і лівий боки 5,6 мм пістолета Grand Power К22 S Рис. 3. Правий і лівий боки 5,6 мм пістолета Grand Power К22 X-TRIM Рис. 3. Правий і лівий боки 5,6 мм пістолета Grand Power К22 X-TRIM Експериментальні стрільби серіями по 25 пострілів здійснювали у бетонний блок бордюру, білу силікатну цеглу й ме- талеву плиту, кут зустрічі куль із поверх- нею перешкод — 0—30°. Завдяки відносно низькій швидкості куль у момент влучан- ня вдалося здобути дані щодо енергетич- них характеристик куль і їх фрагментів після рикошету. Результати досліджень наведено у табл. 1—3 та на рис. 4—6. Таблиця 1 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по бетонному блоку патронами кільцевого запалення калібру 5,6 мм із пістолета Grand Power К22 S (тільки залікові постріли) № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 1 285,9 189,6 2 296,0 40,5 Таблиця 1 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по бетонному блоку патронами кільцевого запалення калібру 5,6 мм із пістолета Grand Power К22 S (тільки залікові постріли) № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 1 285,9 189,6 2 296,0 40,5 Таблиця 1 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по бетонному блоку патронами кільцевого запалення калібру 5,6 мм із пістолета Grand Power К22 S (тільки залікові постріли) № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 1 285,9 189,6 2 296,0 40,5 63 Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Мета статті Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 3 296,9 212,3 4 279,0 201,5 5 282,0 270,8 6 285,0 76,7 7 284,6 201,3 8 303,1 251,9 9 283,2 244,2 10 267,1 233,4 11 303,4 268,4 12 296,8 263,8 Таблиця 2 Таблиця 2 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по силікатній цеглі патронами кільцевого запалення калібру 5,6 мм із пістолета Grand Power К22 S (тільки залікові постріли) № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 1 296,6 232,5 2 294,4 234,2 3 297,6 48,5 4 297,3 246,7 5 301,2 246,5 6 286,1 225,5 7 297,2 232,2 8 280,0 232,1 9 292,8 101,4 10 301,4 226,2 Таблиця 3 лістичні характеристики уражаючих елементів до та після рикошету Балістичні характеристики уражаючих елементів до та після рикоше стрільби по силікатній цеглі патронами кільцевого запалення калібру із пістолета Grand Power К22 S (тільки залікові постріли) ц арактеристики уражаючих елементів до та після рикошету під час Балістичні характеристики уражаючих елементів до та після рикоше Балістичні характеристики уражаючих елементів до та після рикоше стрільби по металевій плиті патронами кільцевого запалення калібр із пістолета Grand Power К22 X-TRIM (тільки залікові постріли ц арактеристики уражаючих елементів до та після рикошету під час ц Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по металевій плиті патронами кільцевого запалення калібру 5,6 мм із пістолета Grand Power К22 X-TRIM (тільки залікові постріли) № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 1 287,7 244,4 2 285,8 246,2 3 298,8 265,5 4 295,8 262,5 5 290,3 208,8 по металевій плиті патронами кільцевого запалення калібру 5,6 мм пістолета Grand Power К22 X-TRIM (тільки залікові постріли) 64 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 № пострілу Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с 6 292,3 259,5 7 285,4 195,4 8 286,8 237,8 9 295,8 244,0 10 291,3 195,8 11 293,4 252,3 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Рис. 6. Пошкодження білої силікатної цегли, спричинені рикошетом 5,6 мм свинцевих куль Швидкість польоту куль до та після рикошету вимірювали оптоелектрон- ним вимірювальним комплексом ИБХ- 731.4. За мішені слугував балістичний пластилін, що є імітатором біологічних тканин тіла людини. Мета статті Рис. 4. Пошкодження бетонного блока, спричинені рикошетом 5,6 мм свинцевих куль Рис. 5. Сліди рикошету на поверхні металевої плити, спричинені влучанням 5,6 мм свинцевих куль Швидкість польоту куль до та після рикошету вимірювали оптоелектрон- ним вимірювальним комплексом ИБХ- 731.4. За мішені слугував балістичний пластилін, що є імітатором біологічних тканин тіла людини. Рис. 4. Пошкодження бетонного блока, спричинені рикошетом 5,6 мм свинцевих куль Рис. 4. Пошкодження бетонного блока, спричинені рикошетом 5,6 мм свинцевих куль Рис. 6. Пошкодження білої силікатної цегли, спричинені рикошетом 5,6 мм свинцевих куль Рис. 4. Пошкодження бетонного блока, спричинені рикошетом 5,6 мм свинцевих куль Рис. 4. Пошкодження бетонного блока, спричинені рикошетом 5,6 мм свинцевих куль Дослідженням встановлено, що фрагментування 5,6 мм свинцевих куль спортивно-мисливських патронів кіль- цевого запалення майже завжди відбу- вається під кутами зустрічі з поверхнею перешкоди від 5 до 30°. Це насамперед зумовлено значною первинною пло- щею контакту з перешкодою, фізико-ме- ханічні характеристики матеріалу якої дорівнюють або перевищують відповід- ні параметри матеріалу куль. Величина сил реакції опору матеріалу перешкоди (з урахуванням контактного тиску) знач- но перевищує межу міцності матері алу кулі, що завдає руйнівних наслідків (див. рис. 7—10). 5,6 мм свинцевих куль Рис. 5. Сліди рикошету на поверхні металевої плити, спричинені влучанням 5,6 мм свинцевих куль Рис. 5. Сліди рикошету на поверхні металевої плити, спричинені влучанням 5,6 мм свинцевих куль 65 Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal алістики. 2022. Випуск 2 (27) e-journal.org/index.php/journal кошету становила 2,55—2,60 г, а після рикошету — 2,37—2,44 г із масою фраг- ментів 0,25—1,75 г. Рис. 9. Розсіювання 5,6 мм свинцевих куль і їх фрагментів після рикошету Рис. 10. Пошкодження манекена 5,6 мм свинцевими кулями та їх фрагментами після рикошету Можливість визначати швидкість і масу куль та їх фрагментів дала змогу розрахувати енергетичні характерис к оре рез а і р ко е кошету становила 2,55—2,60 г, а після рикошету — 2,37—2,44 г із масою фраг- ментів 0,25—1,75 г. Рис. 9. Розсіювання 5,6 мм свинцевих куль і їх фрагментів після рикошету Рис. 9. Розсіювання 5,6 мм свинцевих куль і їх фрагментів після рикошету Рис. 9. Розсіювання 5,6 мм свинцевих куль і їх фрагментів після рикошету Рис. 7. Фрагменти куль, утворені внаслідок рикошету (один бік фрагментів; кут зустрічі — 5—30°) Рис. 7. Фрагменти куль, утворені внаслідок рикошету (один бік фрагментів; кут зустрічі — 5—30°) Рис. 7. Мета статті Фрагменти куль, утворені внаслідок рикошету (один бік фрагментів; кут зустрічі — 5—30°) а б Рис. 8. 5,6 мм свинцева куля під кутом зустрічі з поверхнею бетонного блока, наближеним до 0° (а — один бік кулі; б — зворотний бік кулі зі слідами рикошету) К і і б і Рис. 10. Пошкодження манекена 5,6 мм свинцевими кулями та їх фрагментами після рикошету а б а Рис. 8. 5,6 мм свинцева куля під кутом зустрічі з поверхнею бетонного блока, наближеним до 0° (а — один бік кулі; б — зворотний бік кулі зі слідами рикошету) Рис. 8. 5,6 мм свинцева куля під кутом зустрічі з поверхнею бетонного блока, Рис. 10. Пошкодження манекена 5,6 мм свинцевими кулями та їх фрагментами після рикошету Можливість визначати швидкість і масу куль та їх фрагментів дала змогу розрахувати енергетичні характерис тики утворених у результаті рикошету вторинних уражаючих елементів. Для цього поза оптоелектронним приладом для вимірювання швидкості уражаючих Коли кути зустрічі наближені до діапазону 0—5°, свинцеві кулі калібру 5,6 мм під час контактної швидкості до 300 м/с завдають менших пошкоджень (див. рис. 8). Зважуванням встановлено, що маса 5,6 мм свинцевих куль до ри- 66 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 елементів встановили блок баліс тичного пластиліну розміром 150 × 210 × 140 мм, відстань від блока до перешкоди (від поверхні якої зрикоше- тило) — 1000 мм. Результати досліджень наведено у табл. 4 та на рис. 11—17. опору середовища під час руху у ньо- му кінетичного снаряда (див. рис. 8). До того ж після деформування основний фрагмент кулі після рикошету набуває доволі складної просторової форми з пе- ремінними параметрами руху, він як уражаючий елемент має непередбачу- вану (квазібалістичну) траєкторію руху, що унеможливлює визначення форми та площі його первинного контакту із блоком балістичного пластиліну. Варто зауважити, що більша части- на маси кулі зберігається (див. табл. Мета статті 4), але форма кулі доволі сильно змінюєть- ся, що суттєво ускладнює розрахунки, пов’язані з визначенням коефіцієнта сил 67 Таблиця 4 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по блоку балістичного пластиліну 5,6 мм патронами кільцевого запалення з пістолета Grand Power К22 S (тільки залікові постріли) № пострілу Маса фрагментів після рикошету, г Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с Довжина каналу пошкодження, мм 1 - 287,4 245,6 - 2 - 300,6 206,8 - 3 - 289,4 254,9 72* 4 2,43 292,2 254,2 58 5 2,36 259,6 226,7 89* 6 2,05 302,3 256,9 98** 7 2,38 290,8 242,4 50*** 8 - 298,2 266,3 - Примітка: * — пошкодження, позначене як «1»; — пошкодження, позначене як «2»; * — пошкодження, позначене як «3»; ** — пошкодження, позначене як «4»; *** — пошкодження, позначене як «5». Рис. 11. Фрагменти 5,6 мм куль після рикошету (один бік) Таблиця 4 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по блоку балістичного пластиліну 5,6 мм патронами кільцевого запалення з пістолета Grand Power К22 S (тільки залікові постріли) № пострілу Маса фрагментів після рикошету, г Швидкість кулі до рикошету, м/с Швидкість кулі (фрагментів кулі) після рикошету, м/с Довжина каналу пошкодження, мм 1 - 287,4 245,6 - 2 - 300,6 206,8 - 3 - 289,4 254,9 72* 4 2,43 292,2 254,2 58 5 2,36 259,6 226,7 89* 6 2,05 302,3 256,9 98** 7 2,38 290,8 242,4 50* 8 - 298,2 266,3 - і * Таблиця 4 Балістичні характеристики уражаючих елементів до та після рикошету під час стрільби по блоку балістичного пластиліну 5,6 мм патронами кільцевого запалення з пістолета Grand Power К22 S (тільки залікові постріли) * — пошкодження, позначене як «3»; — пошкодження, позначене як «4»; *** — пошкодження, позначене як «5». Примітка: 67 Рис. 11. Фрагменти 5,6 мм куль після рикошету (один бік) Рис. 11. Фрагменти 5,6 мм куль після рикошету (один бік) 67 алістики. 2022. Випуск 2 (27) e-journal.org/index.php/journal Рис. 15. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «3» Рис. 16. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «4» Рис. 17. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «5» Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) SSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Рис. 12. Мета статті Пошкодження блока балістичного пластиліну фрагментами 5,6 мм куль після рикошету Рис. 13. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «1» Рис. 14. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «2» Рис. 15. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «3» Рис. 16. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «4» Рис. 17. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «5» Зважаючи на результати проведе- них досліджень, найтиповішою фор- мою куль, позначених як «2», «3» та «5» (див. рис. 14, 15, 17), після їх деформу- вання внаслідок рикошету, можна вва- жати форму, наближену до овальної. Якщо, рухаючись у товщі перешкоди, Рис. 15. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «3» Рис. 12. Пошкодження блока алістичного пластиліну фрагментами 5 6 мм куль після рикошету Рис. 15. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «3» пластиліну, позначеного як «3» Рис. 16. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «4» Рис. 12. Пошкодження блока балістичного пластиліну фрагментами 5,6 мм куль після рикошету Рис. 12. Пошкодження блока балістичного пластиліну фрагментам 5,6 мм куль після рикошету 5,6 мм куль після рикошету Рис. 13. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «1» 5,6 мм куль після рикошету Рис. 13. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «1» Рис. 16. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «4» Рис. 16. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «4» Рис. 17. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «5» Рис. 17. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «5» Рис. 13. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «1» Рис. 14. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «2» Рис. 17. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «5» Зважаючи на результати проведе- них досліджень, найтиповішою фор- мою куль, позначених як «2», «3» та «5» (див. рис. 14, 15, 17), після їх деформу- вання внаслідок рикошету, можна вва- жати форму, наближену до овальної. Якщо, рухаючись у товщі перешкоди, Рис. 14. Поздовжній перетин пошкодження блока балістичного пластиліну, позначеного як «2» 68 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 вони не змінювали свого положення, то площа контакту деформованих куль із в’язко-пружним середовищем бло- ка балістичного пластиліну становила 141,14—176,15 мм2. Для таких куль складено табл. 5 і 6 із параметрами, необхідними для по- дальших розрахунків (зокрема, для визначення коефіцієнта сил опору сере- довища). Мета статті Таблиця 5 Параметри фрагментів куль для визначення коефіцієнта сил опору середовища № фрагмен та Маса фрагмента кулі, г Розрахункова площа контакту, мм2 Швидкість фрагмента кулі в момент влучання після рикошету, м/с Глибина каналу пошкодження, мм 2 2,43 160,42 254,2 58 3 2,36 141,14 226,7 89 5 2,38 176,15 242,4 50 Таблиця 6 Енергетичні параметри фрагментів куль після рикошету № фрагмента Кінетична енергія, Дж Питома кінетична енергія, Дж/мм2 2 78,51 0,489 3 60,64 0,430 5 69,92 0,396 Таблиця 5 Параметри фрагментів куль для визначення коефіцієнта сил опору середовища № фрагмен та Маса фрагмента кулі, г Розрахункова площа контакту, мм2 Швидкість фрагмента кулі в момент влучання після рикошету, м/с Глибина каналу пошкодження, мм 2 2,43 160,42 254,2 58 3 2,36 141,14 226,7 89 5 2,38 176,15 242,4 50 Т араметри фрагментів куль для визначення коефіцієнта сил опору серед Таблиця 5 Параметри фрагментів куль для визначення коефіцієнта сил опору середовища № М Р Ш д і ф Г б ц Енергетичні параметри фрагментів куль після рикошету № фрагмента Кінетична енергія, Дж Питома кінетична енергія, Дж/мм2 2 78,51 0,489 3 60,64 0,430 5 69,92 0,396 Енергетичні параметри фрагментів куль після рикошету Отже, вважаємо, що всю кінетичну енер- гію окремого фрагмента кулі витрачено на подолання сил опору в’язко-пружно- го середовища. Тоді: Аналіз даних табл. 5 і 6 свідчить: хоча фрагменти кулі й мають доволі високу контактну швидкість у момент влучан- ня у блок балістичного пластиліну, але для спричинених ними пошкоджень характерні доволі неглибокі канали, що зумовлено значною контактною пло- щею деформованого фрагмента кулі. Незважаючи на це, на основі гранично- го значення довжини ранового каналу 50 мм можна констатувати, що в разі влучання в тіло людини досліджувані фрагменти куль обов’язково завдадуть проникного поранення зі значним обся- гом пошкодження біологічних тканин. (1), де: А — робота, яку здійснив фрагмент кулі у товщі перешкоди, Дж; у у р , ; Еkin — кінетична енергія фрагмента кулі у момент влучання, Дж; m — вага фрагмента кулі, кг; Vc — швидкість фрагмента кулі у мо- мент влучання, м/с; Fc — сила опору середовища, Н; Оскільки всю кінетичну енергія фрагменти куль передали блоку баліс тичного пластиліну, то в першому на- ближенні можна знехтувати втратами енергії на нагрів і подолання сил тертя. 15 Озерецковский Л. Б., Гуманенко Е. К., Бояринцев В. В. Указ. соч. ; Коломійцев О. В., Остро- пілець В. Р., Нікітюк В. Г., Герман О. М., Добривченко Б. Д. Розробка методичних рекомен- дацій з визначення уражаючих властивостей куль після рикошету. Рукопис. № держре єстр. 0118U004606. Харків, 2019. 110 с. l — довжина каналу пошкоджен- ня, м. l — довжина каналу пошкоджен- ня, м. l — довжина каналу пошкоджен- ня, м. Перетворивши емпіричну залеж- ність (1) щодо сили опору Fc і вра- ховуючи емпіричну залежність для 69 Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal розрахунку сили опору, можна визначи- ти коефіцієнт сили опору в’язко-пруж- ного середовища: розрахунку сили опору, можна визначи- ти коефіцієнт сили опору в’язко-пруж- ного середовища: Зменшення контактної швидкості за меншої площині контакту збільшує гли- бину завданого пошкодження. р р (2), У розрахунках щільність середови- ща становила 1040 кг/м3. Здобуті дані дають змогу обчислити граничну швид- кість фрагмента кулі після рикошету у в’язко-пружному середовищі (Vgr), яке визначають за формулою 15: (2), (2), де: ρ — щільність середовища, кг/м3; So — площа контакту фрагмента кулі в момент влучання, м2. Результати розрахунків наведе- но в табл. 7. Аналіз результатів свід- чить, що чим більша площа контакту та швидкість фрагмента кулі в момент влучання, тим більшою є сила опору се- редовища і тим менший рановий канал завдовжки. Vgr = a lnVc — b, (3), (3), де: Vс — швидкість фрагмента кулі у мо- мент влучання, м/с; a, b — коефіцієнти кореляції. у , / ; a, b — коефіцієнти кореляції. Таблиця 7 Та Коефіцієнт опору середовища фрагментів куль після рикошету Коефіцієнт опору середовища фрагментів куль після рикошету № фраг мента Глибина каналу пошкодження, мм Розрахункова площа контакту, мм2 Кінетична енергія, Дж Сила опору середовища, Н Коефіцієнт сил опору середовища 2 58 160,42 78,51 1353,62 0,251 3 89 141,14 60,64 681,35 0,181 5 50 176,15 69,92 1398,40 0,260 Граничне значення швидкості Vgr визначають на основі квадратичного закону спротиву, згідно з яким швид- кість уражаючого елемента в разі так званих сліпих вогнепальних поранень (пошкоджень) не може дорівнювати 0: його визначають із отриманих функціо- нальних залежностей для конкретного уражаючого елемента за допущень, що l = 0, до того ж ln Vc = ln Vgr. На підставі наведених у табл. 7 да- них визначено, що для досліджуваних фрагментів куль після рикошету значен- ня коефіцієнтів кореляції становлять: a = 0,349; b = 1,977. Значення коефіцієн- та опору для досліджуваних фрагментів куль після рикошету для в’язко-пружно- го середовища — 0,181—0,260. Отже, якщо вважати за граничне значення довжину ранового каналу 70 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 50 мм, то розрахункове граничне зна- чення швидкості такого уражаючого елемента, як деформований фрагмент 5,6 мм свинцевої кулі після рикошету, становитиме 250 м/с. Тому, використо- вуючи емпіричне вираження (3), можна обчислити довжину ранових каналів для експериментально визначених зна- чень контактної швидкості фрагментів куль після рикошету. Результати розрахунків наведено у табл. 8. розрахунку сили опору, можна визначи- ти коефіцієнт сили опору в’язко-пруж- ного середовища: Таблиця 8 Розрахункові значення довжини каналів пошкоджень № фраг мента Швидкість фрагмента кулі в момент влучання після рикошету, м/с Експериментально визначена глибина каналу пошкодження, мм Розрахункове значення глибини каналу пошкодження, мм Погрішність, % 2 254,2 58 52 10,3 3 226,7 89 33 62,9 5 242,4 50 47 6,0 а за більших — збільшується, хоча прове- дені експерименти довели зворотну за- лежність, яка має фізичне пояснення: зі зростанням швидкості сила опору збіль- шується у квадратичній залежності, що за інших рівних умов призводить до зменшення довжини ранового каналу. Наприклад, для фрагмента кулі «3» за ін- ших рівних умов (із контактними швид- костями 200, 250 і 300 м/с) розрахункові значення довжини каналу пошкоджень становитимуть 10,4; 50,0 і 82,4 мм від- повідно, хоча на практиці має місце зво- ротна тенденція. а за більших — збільшується, хоча прове- дені експерименти довели зворотну за- лежність, яка має фізичне пояснення: зі зростанням швидкості сила опору збіль- шується у квадратичній залежності, що за інших рівних умов призводить до зменшення довжини ранового каналу. Наприклад, для фрагмента кулі «3» за ін- ших рівних умов (із контактними швид- костями 200, 250 і 300 м/с) розрахункові значення довжини каналу пошкоджень становитимуть 10,4; 50,0 і 82,4 мм від- повідно, хоча на практиці має місце зво- ротна тенденція. Аналіз наведених у табл. 8 даних свідчить про наявність доволі високих розбіжностей у значеннях глибини ка- налів пошкоджень, здобутих експери- ментально та шляхом розрахунків, у разі значень контактної швидкості, менших за граничне. Пояснити це можна лише недостатнім обсягом експерименталь- них даних, необхідних для їх статистич- ного оброблення. Водночас значну кіль- кість експериментальних досліджень доволі складно статистично обробити. Отже, визначені емпіричні залежності для обчислення глибини ранового ка- налу фрагментів 5,6 мм свинцевих куль після рикошету мають обмежене засто- сування, оскільки їх можна використа- ти для випадків, коли контактні швид- кості близькі до граничного значення (250 м/с). За менших контактних швид- костей розрахункове значення довжини ранового каналу суттєво зменшується, Згідно з енергетичною концепцією ураження цілі основним критерієм є гранична величина питомої кінетич- ної енергії (0,5 Дж/мм2), тому для до- сліджуваних фрагментів куль можна обчислити відповідні значення контакт- ної швидкості. Результати розрахунків викладено в табл. 9. 71 Теорія та практика судової експертизи і криміналістики. 2022. розрахунку сили опору, можна визначи- ти коефіцієнт сили опору в’язко-пруж- ного середовища: Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Таблиця 9 Швидкість фрагментів 5,6 мм свинцевих куль після рикошету (граничне значення питомої кінетичної енергії 0,5 Дж/мм2) № фрагмента Граничне значення питомої кінетичної енергії, Дж/мм2 Відповідне значення швидкості, м/с 2 0,5 256,9 3 0,5 244,6 5 0,5 272,1 Т Швидкість фрагментів 5,6 мм свинцевих куль після рикошету (граничне значення питомої кінетичної енергії 0,5 Дж/мм2) поверхні уражаючих елементів під час їх просування в товщі імітатора біоло- гічних тканин), за умови розрахування довжини ранового каналу на підставі наявних емпіричних залежностей,— можливе зниження ступеня достовір- ності результатів. Однак, у першому на- ближенні такий розрахунковий метод можна застосовувати для визначення уражаючих властивостей куль і їх фраг- ментів після рикошету та визначення ймовірної шкоди, завданої тілу люди- ни. До того ж проведені дослідження свідчать, що такий спосіб доцільний для визначення уражаючих властиво- стей малокаліберних куль також інших калібрів після рикошету. поверхні уражаючих елементів під час їх просування в товщі імітатора біоло- гічних тканин), за умови розрахування довжини ранового каналу на підставі наявних емпіричних залежностей,— можливе зниження ступеня достовір- ності результатів. Однак, у першому на- ближенні такий розрахунковий метод можна застосовувати для визначення уражаючих властивостей куль і їх фраг- ментів після рикошету та визначення ймовірної шкоди, завданої тілу люди- ни. До того ж проведені дослідження свідчать, що такий спосіб доцільний для визначення уражаючих властиво- стей малокаліберних куль також інших калібрів після рикошету. Середнє значення граничної швид- кості — 257,8 м/с, що наближене до визначеного раніше граничного значен- ня (250 м/с). Проте ці значення актуальні лише для фрагментів, які близькі за ма- сою (не менше ніж 90—95 % від первин- ної маси куль) і мають відповідну площу контакту, схожу за формою та ступенем деформування кулі після рикошету. За інших параметрів форми, розмірів і маси фрагментів куль їх уражаючі властивості після рикошету потрібно визначати індивідуально. Висновки Результати проведених досліджень да- ють підстави стверджувати, що (не- зважаючи на складність проведення експериментів і наявність неповної кореляції результатів цих досліджень із результатами розрахунків) запрова дження у практику судово-балістичної й судово-медичної експертиз альтерна- тивного критерію оцінювання уража ючих властивостей куль і їх фрагментів після рикошету, який ґрунтується на довжині ранового каналу з відповідним його граничним значенням, є доціль- ним. Певні недоліки такого підходу (без використання комп’ютерного мо- делювання фізичних процесів, які від буваються під час рикошету, і без більш точного обчислення значень контактної Окремо варто зазначити, що під час експериментів уперше здобуто дані щодо швидкості свинцевих мало- каліберних куль і їх фрагментів після рикошету, що дало змогу розрахувати й визначити емпіричну залежність дов- жини ранового каналу від контактної швидкості уражаючого елемента.f f Properties of Small-bore Bullets After Ricochet t Oleksandr Kolomiitsev, Viktor Sapielkin, Pavel Giverts, Oleksandr Herman Authors aimed to summarize experimental researches of affecting properties of monolithic .22 Long Rifle bullets after their ricochet from various types of obstacles. Alternative criterion 72 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 for evaluating the affecting properties of bullets is proposed using this criterion it is possible to forecast the degree and severity of bodily injuries after ricochet. Specifics and conditions under which fragmentation of monolithic small-bore bullets occurs are revealed, as well as flight speed of bullets and their fragments after impact-contact interaction with obstacles of various types is determined. Special attention is paid to research on interaction nature of bullets and their fragments with a viscoelastic environment imitating biological human body tissues after ricocheting. It has been determined that energy concept of affection to biological target and criteria for its evaluation research on affecting properties of bullets after ricochet have limited application, since it is not always possible to accurately determine initial conditions of hitting as the bullets themselves as their individual fragments. Obtained results of experimental research and field tests made possible to form an alternative criterion for assessing the severity of bodily injuries that a person suffered as a result of being hit by small-bore lead bullets after ricocheting. It is the length of the wound channel caused by a bullet or its fragments as a result of their impact on a person’s body after ricocheting. A calculation method for determining the length of the caused wound channel is proposed that can be used in forensic expert practice to determine potential injury to human life and health. for evaluating the affecting properties of bullets is proposed using this criterion it is possible to forecast the degree and severity of bodily injuries after ricochet. Specifics and conditions under which fragmentation of monolithic small-bore bullets occurs are revealed, as well as flight speed of bullets and their fragments after impact-contact interaction with obstacles of various types is determined. Special attention is paid to research on interaction nature of bullets and their fragments with a viscoelastic environment imitating biological human body tissues after ricocheting. f Properties of Small-bore Bullets After Ricochet It has been determined that energy concept of affection to biological target and criteria for its evaluation research on affecting properties of bullets after ricochet have limited application, since it is not always possible to accurately determine initial conditions of hitting as the bullets themselves as their individual fragments. Obtained results of experimental research and field tests made possible to form an alternative criterion for assessing the severity of bodily injuries that a person suffered as a result of being hit by small-bore lead bullets after ricocheting. It is the length of the wound channel caused by a bullet or its fragments as a result of their impact on a person’s body after ricocheting. A calculation method for determining the length of the caused wound channel is proposed that can be used in forensic expert practice to determine potential injury to human life and health. й аналізуванні даних, рішенні про пу- блікацію чи підготовку рукопису. References Bilenko, O. I., Kyrychenko, O. O., Pavlov, D. V. (2017). Doslidzhennia vplyvu materialu metalnoho elementu na kharakterystyky rykoshetu [Methods of reducing the energy characteristics of affecting element after reflection from obstacle]. Zbirnyk naukovykh prats Natsionalnoi hvardii Ukrainy. Vyp. 2 (30). URL: http:// nbuv.gov.ua/UJRN/znpavs_2017_2_5 [in Ukrainian]. Bilenko, O. I., Kyrychenko, O. O., Pavlov, D. V. (2018). Metodyka znyzhennia enerhetychnykh kharakterystyk porazhaiuchoho elementa pislia vidbyttia vid pereshkody [Research on influence of metal element material on ricochet characteristics]. Science Rice. Vol. 4 (45). DOI: 10.15587/2313-8416.2018.129326 [in Ukrainian]. Keywords: energy concept of target hitting; calculation method; affecting properties; alternative evaluation criterion; length of the wound canal; imitator of biological tissues; small-bore bullet; ricochet. Burke, Т. W., Rowe, W. F. (1992). Bullet ricochet: A comprehensive revive. Journal of Forensic Sciences. Vol. 37 (5). DOI: 10.1520/JFS13312J. Denisov, A. V., Tiurin, M. V., Sokhranov, M. V. i dr. (2014). Osobennosti porazheniia zhivykh tselei v zone rikosheta pul pri strelbe po tverdym pregradam [Specificsof defeating live targets in bullet ricochet zone while firing at solid obstacles]. Vestnik Rossiiskoi voenno-meditsinskoi akademii. № 1 (45) [in Russian]. Учасники Учасники Автори внесли свій внесок винятково в інтелектуальну дискусію, що є осно- вою цього документа, дослідження су- дової практики, написання та редагу- вання, і беруть на себе відповідальність за її зміст і тлумачення. Декларація щодо конфлікту інтересів Автори заявляють, що у них відсутній конфлікт інтересів. References Фінансування Це дослідження не отримало жодно- го спеціального гранту від фінансових установ у державному, комерційному чи некомерційному секторах. Відмова від відповідальності Відмова від відповідальності Засновники не грали жодної ролі у роз робленні дослідження, добиранні д о а д д о да ос Засновники не грали жодної ролі у роз робленні дослідження, добиранні Засновники не грали жодної ролі у роз робленні дослідження, добиранні Ehidlin, L. M. (1963). Ognestrelnye povrezhdeniia. Vrachebnoe i kriminalisticheskoe raspo Ehidlin, L. M. (1963). Ognestrelnye povrezhdeniia. Vrachebnoe i kriminalisticheskoe raspo 73 Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Kolomiitsev, O. V. (2005). Metod rozrakhunkovoho ta eksperymentalnoho vyznachennia vyhliadu snariada spetsialnoho pryznachennia iz zadanymy aerodynamichnymy vlastyvostiamy [Calculation method of and experimental determination of appearance of a special-purpose projectile with given aerodynamic properties] : dys. … kand. tekhn. nauk. Kharkiv [in Ukrainian]. znavanie i otsenka [Gunshot damage. Medical and forensic recognition and evaluation]. Tashkent [in Russian]. Giverts, P. V. (2018). Kriminalisticheskoe issledovanie vintovok AIRSOFT, moder nizirovannykh dlia strelby patronami kalibra 5,56 × 45 mm [Criminalistic research on airsoft rifles modified for shooting cartridges of the caliber 5.56 × 45 mm]. Teoriia ta praktyka sudovoi ekspertyzy i kryminalistyky. Vyp. 18. DOI: 10.32353/khrife.2018.34 [in Russian]. Kolomiitsev, O. V., Ostropilets, V. R., Nikitiuk, V. H., Herman, O. M., Dobryvchenko, B. D. (2019). Rozrobka metodychnykh rekomendatsii z vyznachennia urazhaiuchykh vlastyvostei kul pislia rykoshetu [Development of methodological recommendations for determining the affecting properties of bullets after ricochet]. Rukopys. № derzhreiestr. 0118U004606. Kharkiv [in Ukrainian]. Gusentsov, A. O., Artiushin, A. A., Danilo va, V. A. (2009). Teoriia i praktika sozdaniia modeli rikosheta ognestrelnogo snariada v usloviiakh laboratornogo ehksperimenta [Theory and practice of creating a ricochet model of firearm projectile under conditions of a laboratory experiment]. Sudovo-balistychni doslidzhennia ta sumizhni haluzi znan : persh. mizhnar. nauk.-prakt. seminar. Kyiv [in Russian]. Kotas, Z., Reno, C. E. (2014). A Case Study in the use of bullet ricochet experimentation to provide evidence in a homicide. AFTE Journal. Vol. 46 (3). y Haag, L. C. (1975). Bullet Ricochet: An empirical study and a device for measuring ricochet angle. AFTE Journal. Vol. 7 (3). Moxnes, J. F., Frøyland, Ø., Skriudalen, S. et al. (2016). On the study of ricochet and penetration in sand, water and gelatin by spheres, 7.62 mm APM2, and 25 mm projectiles. Defence Technology. Vol. Теорія та практика судової експертизи і криміналістики. 2022. Випуск 2 (27) ISSN 1993-0917 e-ISSN 2708-5171 https://khrife-journal.org/index.php/journal Коломійцев, О., Сапєлкін, В., Ґіверц, П., Герман, О. (2022). Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. Теорія та практика судової експертизи і криміналістики. Вип. 2 (27). С. 59—75. DOI: 10.32353/khrife.2.2022.05. Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Фінансування 12. DOI: 10.1016/j.dt.2015.12.004. Haag, M., Haag, L. (2020). Shooting incident reconstruction. DOI: 10.1016/C2018-0- 03137-0. Jauhari, M. (1969). Bullet ricochet from metal plates. The Journal of Criminal Law, Criminology, and Police Science. Vol. 60 (3). DOI: 10.2307/1141995. Ozeretskovskii, L. B., Gumanenko, E. K., Boiarintsev, V. V. (2006). Ranevaia ballistika [Wound ballistics]. Sankt-Peterburg [in Russian]. Karget, B, Joosten, U. (2001). A case of «boomerang» bullet ricochet. International Journal of Legal Medicine. Vol. 115 (2). DOI: 10.1007/s004140000148. Petruchenko, O. S. (2015). Shchodo vyvchennia vzaiemodii kuli chy oskolka snariada z pereshkodoiu [Regarding the study of the interaction of a bullet or a fragment of a projectile with an obstacle]. Zbirnyk naukovykh prats Viiskovoi akademii (m. Odesa). Tekhnichni nauky. Vyp. 2 (4). URL: http://nbuv.gov.ua/UJRN/ zbnpva_2015_2_13 [in Ukrainian]. Kofanov, A. V. (2018). Teoretychni ta praktychni aspekty modeliuvannia rykoshetuvannia vohnepalnoho snariada pry provedeni ekspertnoho eksperymentu [Theoretical and practical aspects of modeling of firearm projectile ricochet while forensic expert experiment]. Kryminalistyka i sudova ekspertyza. Vyp. 63 (1). URL: http://nbuv.gov.ua/ UJRN/krise_2018_63%281%29__36 [in Ukrainian]. Petruchenko, O. S., Fliud, O. V., Bilash, O. V., Velychko, L. D. (2017). Doslidzhennia spivudaru tverdykh til [Research on collision of solid bodies]. VII Naukovo- praktychna konferentsiia : zb. tez dopov. Kharkiv [in Ukrainian]. 74 Олександр Коломійцев, Віктор Сапєлкін, Павел Ґіверц, Олександр Герман. Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. DOI: 10.32353/khrife.2.2022.05 Shepitko, V., Shepitko, M., Simakova- Yefremian, E., Kolomiitsev, A. (2020). The Determination of the Bullet Velocity After Ricochet Using Computer Modeling. AFTE Journal. Vol. 52 (2). Popov, V. L., Shigeev, V. B., Kuznetsov, L. E. (2002). Sudebno-meditsinskaia ballistika [Forensic ballistics]. Sankt- Peterburg. URL: https://djvu.online/file/ hkakzjDMC5I3P [in Russian]. Thompson, E., Wyant, R. (2003). Asphalt skip shooting reconstruction. AFTE Journal. Vol. 35 (1). Rathman, G. А. (1987). Bullet ricochet and associated phenomena. AFTE Journal. Vol. 19 (4). Коломійцев, О., Сапєлкін, В., Ґіверц, П., Герман, О. (2022). Особливості визначення уражаючих властивостей малокаліберних куль після рикошету. Теорія та практика судової експертизи і криміналістики. Вип. 2 (27). С. 59—75. DOI: 10.32353/khrife.2.2022.05. 75 75
W4391235603.txt	https://journals.um.si/index.php/geography/article/download/3567/2517	sl		Navodila za pripravo člankov za Revijo za geografijo	Revija za geografijo	2,023	cc-by	1,087	Revija za geografijo – Journal for Geography, 18-1, 2023, str. 115–118 Navodila za pripravo člankov za Revijo za geografijo 1 Sestavine članka Članki imajo naslednje sestavine: glavni naslov članka, ime in priimek avtorja, avtorjev elektronski naslov in ORCiD, ustanova (primer: Univerza v Mariboru, Filozofska fakulteta, Oddelek za geografijo; Maribor, Slovenija), besedilo članka (Izvleček, Uvod, Metodologija, Rezultati, Sklep, Literatura; glavna poglavja je možno razdeliti na podpoglavja, nadaljnje členitve ni). V primeru, da je članek napisan v slovenskem jeziku, obvezno pripravimo še: title (angleški prevod naslova), abstract (angleški prevod izvlečka, skupaj do 120 besed), keywords (angleški prevod ključnih besed, skupaj do 8 besed), summary (angleški povzetek članka, skupaj s presledki do 8000 znakov). V primeru, da je članek napisan v angleškem jeziku, obvezno pripravimo še: naslov (slovenski prevod naslova), izvleček (slovenski prevod abstract, skupaj do 120 besed), ključne besede (slovenski prevod keywords, skupaj do 8 besed), povzetek (slovenski povzetek članka, skupaj s presledki do 8000 znakov). 2 Tehnična navodila Članek oddamo v digitalni obliki (.docx format). Za vnos uporabimo šablono, ki je dostopna na portalu: Zapis besedila je enostaven, brez oblikovanja (Normal style). Uporabimo pisavo Verdana, velikost pisave je 9, razmik med vrsticami je enojen. Označimo izključno krepki (naslovi in podnaslovi) ter ležeči (tujke) tisk. Besedilo je v celoti izpisano z malimi črkami, razen velikih začetnic in kratic. Slikovno gradivo je umeščeno v besedilo na primerno mesto v samem dokumentu. Karte in grafikoni so v .ai, .cdr ali .svg formatu, fotografije in druge grafične priloge v .png ali .jpg formatu z ločljivostjo vsaj 300 dpi. Posebnosti: Ko je krajšava, okrajšava in kratica omenjena prvič, je v oklepaju naveden celoten naziv. Podobno v primeru, ko je pojem znan v angleškem jeziku, je naveden njegov prevod v oklepaju (primer: energetska mešanica (ang. energy mix)). Decimalna števila so zapisana z vejico (primer: 54,6), tisočice in milijonice so ločene s piko (primer: 94.370, 2.079.797). Številčno merilo karte je zapisano nestično (primer: 1 : 25.000). Med številkami in enotami je presledek (primer: 2.864 m, 5,4 %), pri oznakah v potenci ni presledka (primer: 20.273 km², 9,2° C). Dokument s člankom je poimenovan s priimkom prvega avtorja in skrajšanim naslovom članka (primer: Novak Vegetacija na Pohorju.docx). Pravopis in lektura je odgovornost avtorja. 115 Poziv za prispevke 3 Slike in tabele Napisi slikovnega gradiva in tabel je pod elementom. Napis vsebuje zaporedno številko in naslov z virom (primer: Slika 1: Spreminjanje števila prebivalstva v Sloveniji v obdobju med leti 2010 in 2020 (Vir podatkov: SURS, 2023)). Napise pripravimo ročno, ne uporabljamo funkcije za avtomatsko označevanje in oštevilčevanje. Slike zasedajo največ pol strani, z izjemo večjih kart. Tabele so brez posebne oblike, podobno kot preostalo besedilo. Karte ne vsebujejo naslova. Oblika besedila na karti je enaka obliki besedila v članku. Velikost pisave v legendi je 8, v kolofonu pa 6. V kolofonu so navedeni Viri podatkov, Kartografska podlaga in Avtor. Na karti uporabimo grafično merilo. Za grafične priloge, za katere nimamo avtorskih pravic, pridobimo dovoljenje za objavo od lastnika avtorskih pravic. Pri tem v podnapisu dopišemo tudi avtorja slike. Ob oddaji prispevka priložimo dokazilo dovoljenja za objavo. 4 Citiranje v članku V prispevkih sledimo načinu citiranja oz. navajanja literature po APA oz. APA standard. Podrobnosti tega načina citiranja najdemo v priročniku Publication manual of the American Psychological Association. Pri pripravi članka uporabimo funkcijo za vnos citatov in samodejno urejanje seznama literature v Word programu (Reference – Citati). a) Citiranje med besedilom En avtor Primer: Čepič (2018) ali (Čepič, 2018) Dva avtorja Med oba priimka vstavimo veznik »in«. Primer: Avsec in Ferk Savec (2019) ali (Avsec in Ferk Savec, 2019) Trije avtorji ali več avtorjev Vsakič citiramo (tudi ob prvem navajanju) le prvega avtorja in dodamo »idr.« Primer: (Razpotnik idr., 2019) ali Razpotnik idr. (2019) Če imamo več virov, kjer so trije, ali več avtorjev, ki imajo na začetku enake priimke, moramo navajati vse avtorje do vključno prvega različnega, da se izognemo dvoumnosti. Primer: Kapoor, Bloom, Montez idr. (2017) Kapoor, Bloom, Zucker idr. (2017) Če je isto misel ali idejo podalo več avtorjev in želimo navesti vse, potem jih med seboj ločimo s podpičjem. Uredimo jih po abecednem vrstnem redu glede na priimek prvega avtorja. 116 Revija za geografijo – Journal for Geography, 18-1, 2023, str. 115–118 Primer: (Bagiati in Evangelou, 2015; Hu idr., 2017; Lee in Kemple, 2014; Zaranis in Oikonomidis, 2016) Če je isti avtor ali ista skupina avtorjev v enem (istem) letu izdala več del, letnicam dodamo črke a, b, c …. Iste oznake uporabimo tudi v seznamu virov na koncu. Primer: Kostanjevec (2004a, 2004b) ali (Kostanjevec, 2004a, 2004b) Če navajamo več del enega avtorja, dela razvrstimo po kronološkem vrstnem redu. Primer: (Foucault, 2008b, 2009, 2010). b) Navajanje virov Seznam oblikujemo po abecedi. Virov ne številčimo in pred njimi ne dodajamo označevalcev (bullets). Na koncu vsakega vira je pika, razen če gre za elektronski vir. Pri tem uporabimo funkcijo za samodejni vnos seznama literature v Word programu (Reference – Bibliografija). znanstveni članek: Ivajnšič, D., Kaligarič, M., in Žiberna, I. (2014). Geographically weighted regression of the urban heat island of a small city. Applied Geography, 53, 341-353. monografija: Kumer, P. (2019). Lastniki gozdov v Sloveniji. Založba ZRC. monografija z uredniki namesto avtorjev: Perko, D., Ciglič, R. in Zorn, M. (ur.). (2020). The geography of Slovenia: small but diverse (pp. 227-243). Springer. poglavje v monografiji: Pipan, P. in Zorn, M. (2020). Public participation in earthquake recovery in the border region between Italy and Slovenia. Participatory Research and Planning in Practice, 147-167. članek iz spletnega časopisa: Rokavec, Ž. (14. 2. 2017). Program Erasmus+: več možnosti za mednarodne izkušnje za študente iz socialno šibkih družin. Dnevnik. https://www.dnevnik.si/1042762647/lokalno/ljubljana/program-erasmus-vecmoznosti-za-mednarodne-izkusnje-za-studente-iz-socialno-sibkih-druzin članek iz tiskane revije: Gaberšek, M. in Kocijančič, S. (2019). Prvi koraki v Arduino – sporočilo za stisko SOS. Tim: revija za tehnično in znanstveno izobrazbo mladine, 58(1), 22–25. zakon iz Uradnega lista: Zakon o usmerjanju otrok s posebnimi potrebami (ZUOPP-1). (2011). Uradni list RS, št. 58/11, 40/12 – ZUJF, 90/12 in 41/17 – ZOPOPP. http://pisrs.si/Pis.web/pregledPredpisa?id=ZAKO5896# 117 Poziv za prispevke 5 Recenziranje člankov Članki se recenzirajo po pristopu dvojne slepe recenzije. Recenzijo opravijo znanstveniki in strokovnjaki z ožjega področja, ki ga obravnava članek. Če recenziji ne zahtevata popravka ali dopolnitve članka, se avtorju članka recenzij ne pošlje. Avtor po prejemu recenzije lahko članek dopolni in ob nalaganju dopolnjene različice priloži recenzijski obrazec, v katerem so označene predlagane in vnesene spremembe. Uredniški odbor lahko na predlog urednika ali recenzenta zavrne objavo prispevka. Prispevke oddamo prek sistema Open Journal https://journals.um.si/index.php/geography/about/submissions. V primeru težav lahko kontaktiramo uredništvo: Peter Kumer Oddelek za geografijo Filozofska fakulteta Univerza v Mariboru Koroška 160 2000 Maribor e-pošta: peter.kumer@um.si 118 Systems:
https://openalex.org/W2114851799	https://www.scielo.br/j/rsp/a/stV6VCY7WxWMJ4xgWgTMrzq/?lang=pt&format=pdf	Portuguese	null	Estudos sorológicos para pesquisa de anticorpos de arbovírus em população humana da região do Vale do Ribeira: III - inquérito em coabitantes com casos de encefalite por Flavivirus Rocio	Revista de saúde pública/Revista de Saúde Pública	1,982	cc-by	4,672	*Do Departamento de Epidemiologia da Faculdade de Saúde Pública da Universidade de São Paulo — Av. Dr. Arnaldo, 715 — 01255 — São Paulo, SP — Brasil e da Secretaria de Estado da Saúde de São Paulo — Av. Dr. Arnaldo, 451— 01246 — São Paulo, SP — Brasil. O trabalho de campo foi realizado com o auxílio parcial financeiro da Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP). Processo 04 — Biológicas 79/1495. Do Instituto Evandro Chagas. Fundação de Serviços de Saúde Pública — Av. Almirante1' Barroso, 492 — 66000 — Belém, PA — Brasil. Da Universidade Federal da Paraíba — Campus Universitário — 58000 — João Pessoa, PB — Brasil. INTRODUÇÃO ram que 75% dos casos (349/465) ocupa- vam moradias nas quais nenhum outro caso da moléstia foi assinalado. Em publicações anteriores, em que se estu- daram aspectos relacionados com a extensa epidemia de encefalite por arbovírus Rocio, ocorrida recentemente na região sudeste do Brasil, aventou-se a hipótese de que, em alguns casos, poderia ter se verificado a transmissão domiciliar do agente etiológico, seja por mosquitos hematófagos, seja por outros mecanismos (Tiriba 2 2, 1975, Forat- tini e col.7 1978, Iversson 9, 1980). Porém, escassa informação existe sobre a freqüência de casos sub-clínicos ou clíni- cos sem comprometimento do sistema ner- voso central. Em outras encefalites por arbo- vírus essa freqüência tem sido calculada. Entre os Alphavirus, Casals e Clarke4 (1965) mencionam uma proporção de um caso de encefalite para 10 de infecção pelo vírus da encefalite eqüina do leste entre crianças, um para 20 entre pessoas idosas e um para 50 ou mais casos em adultos de meia idade. Martin e col.11 (1972), investigando em Costa Rica a encefalomielite venezuelana, em 13 casos da infecção identificados soro- logicamente, encontraram 5 assintomáticos. Em relação aos Flavivirus, Brody e col.2 (1959) e Monath 15 (1979) verificaram em epidemias de encefalite de St. Louis, EUA, relações de 1:64 e 1:355 entre infecções clinicamente aparentes e infecções assinto- máticas causadas pelo referido vírus. Sou- tham 20 (1956) estimou que em Tokio, Japão, pelo menos 500 e possivelmente 1.000 casos de infecção inaparente por vírus da ence- falite japonesa ocorreram para um caso clinicamente aparente. Halstead e Grosz8 (1962), em uma epidemia dessa mesma moléstia em uma base aérea da Coréia, verificaram uma proporção de um caso de encefalite para 25 casos de infecção inapa- rente. A análise dos caracteres epidemiológicos dessa moléstia na região do Vale do Ribeira, mostrou que os indivíduos adultos, do sexo masculino, que exerciam atividades no ambiente silvestre e/ou no ambiente modificado, junto aos núcleos rurais e urbanos, apresentaram maior risco de adquirir a moléstia. No entanto, entre os doentes existiam um adulto epilético e 22 crianças menores de quatro anos de idade que não haviam se afastado das moradias desde 30 dias antes do início dos primeiros sintomas, o que permitiu supor que também pudesse ter havido transmissão domiciliar da doença (Iversson9, 1980). ESTUDOS SOROLÓGICOS PARA PESQUISA DE ANTICORPOS DE ARBOVÍRUS EM POPULAÇÃO HUMANA DA REGIÃO DO VALE DO RIBEIRA III INQUÉRITO EM COABITANTES COM CASOS DE ENCEFALITE POR FLAVIVIRUS ROCIO Lygia Busch Iversson * Amélia P.A. Travassos d Jorge Travassos da Rosa Clodoaldo da Silveira Cost IVERSSON, L.B. et al. Estudos sorológicos para pesquisa de anticorpos de arbovírus em população humana da região do Vale do Ribeira. III — Inquérito em coabi- tantes com casos de encefalite por Flavivirus Rocio. Rev. Saúde públ., S. Paulo, 16: 160-70, 1982. RESUMO: Foi realizado inquérito sorológico para pesquisa de anticorpos de 17 arbovírus existentes no país, em coabitantes com doentes de encefalite por Rocio, residentes em zona urbana da região do Vale do Ribeira, São Paulo (Brasil), onde ocorreu recentemente uma extensa epidemia dessa molés- tia. Não se verificou maior prevalência de anticorpos IH para vírus Rocio nessas pessoas quando comparadas com indivíduos que não coabitavam com doentes de encefalite. Foram observados e discutidos alguns aspectos já verificados em outros grupos populacionais estudados anteriormente: maior prevalência de anticorpos IH de arbovírus em homens, particularmente pes- cadores; aumento dessa prevalência com a idade e presença de pessoa com antecedente de encefalite que apresentou, exclusivamente anticorpos neutra- lizantes para o Alphavirus EEL, o qual até agora não tem sido responsa- bilizado por moléstia na região. Encontrou-se baixa proporção de indivíduos com anticorpos para Rocio e Flavivirus em geral, fato este estranhável con- siderando a recente epidemia. UNITERMOS: Arboviroses. Inquéritos sorológicos. Encefalite, epidemia. Flavívirus Rocio. UNITERMOS: Arboviroses. Inquéritos sorológicos. Encefalite, epidemia. Flavívirus Rocio. INTRODUÇÃO Pesquisas entomológicas realizadas na área epidêmica, assinalando a presença nos domicílios de culicídeos de hábitos silvestres, tornou pos- sível associá-la a ocorrência de infecção no grupo etário de baixa idade (Forattini e col.7, 1978). Havia interesse em se verificar o papel que a moradia teria representado na cadeia de transmissão da arbovirose à população humana. Os dados epidemiológicos indica- vam que raramente ocorreu mais de um caso clínico na mesma residência. Entre os 414 doentes da região do Vale do Ribeira, em que foi realizada a investigação domiciliar, só foram encontradas 16 famílias onde esse fato se registrou. Em 12 delas observou-se a presença de 2 casos e em 4, de 3 casos (Iversson9, 1980). Em 1975, na área de maior morbidade por encefalite, abrangendo também municípios não pertencentes ao Vale do Ribeira, Lopes e col.13 (1978) verifica- A inexistência de confirmação sorológica para a maioria dos casos de encefalite noti- ficados na região do Vaie do Ribeira e as limitações da reação de inibição de hema- glutinação para identificação específica dos arbovírus dificultam a estimativa da razão entre infecção inaparente: doença para a encefalite por Rocio. No entanto, é possível uma comparação entre os dados sorológicos relativos a Rocio, de residentes ou não, na mesma casa de doentes de encefalite por Rocio, objetivando verificar se haveria maior concentração de pessoas com anticorpos para esse vírus nessas casas. Dois inquéritos sorológicos ralizados na região do Vale do Ribeira em pessoas sem antecedente de encefalite, que não coabitavam com doentes dessa moléstia, mostraram proporções de 4,2% (21/502) e 6,0% (5/83) de indivíduos com anticorpos inibidores de hemaglutinação para o Flavi- virus Rocio (Iversson e col.11,12, 1980, 1981). nessas casas. Dois inquéritos sorológicos ralizados na região do Vale do Ribeira em pessoas sem antecedente de encefalite, que não coabitavam com doentes dessa moléstia, mostraram proporções de 4,2% (21/502) e 6,0% (5/83) de indivíduos com anticorpos inibidores de hemaglutinação para o Flavi- virus Rocio (Iversson e col.11,12, 1980, 1981). táveis de anticorpos. Selecionou-se para este estudo, entre os 73 doentes, os residentes na zona urbana e pesquisou-se as pessoas que com eles coabitavam. Foram realizados testes de inibição de hemaglutinação para Rocio e para outros arbovírus existentes no país em 82 comunicantes de 24 doentes, residentes na zona urbana de 4 cidades da área. * Dados fornecidos pela Comissão de Atividades referentes a Arboviroses da Secretaria de Estado da Saúde do Estado de São Paulo. INTRODUÇÃO A distribuição desses comunicantes investigados, segundo as residências, foi a seguinte: 9 e 7 em uma casa, 6 em 2, 4 em 7, 3 em 3, 2 em 7 e 1 em 3. Programou-se o presente estudo soro- lógico em residentes na mesma moradia de doente de encefalite por Rocio, confirmada sorologicamente. Evidentemente, os dados seriam analisados levando em consideração as atividades exercidas por esses comuni- cantes, pois nas familias em que todos os membros igualmente se expuseram à arbo- virose fora do domicílio seria plausível o encontro de grande número de infectados, mesmo sem a ocorrência de infecção intra- domiciliar. Dos 82 pesquisados, 56 (68,3%) tinham 15 ou mais anos, 46 (56,1%) eram do sexo feminino, 80 (97,6%) haviam nascido na região e entre estes 68 sempre residiram no local de nascimento. Os 2 não naturais moravam no Vale do Ribeira há 6 anos. Três pessoas referiram antecedente de encefalite diagnosticada em hospital local. Duas pessoas mencionaram vacinação contra febre amarela. Foi realizado teste de neutra- lização em camundongos para o Alphavirus da encefalite eqüina do leste (EEL) em um comunicante com antecedente de ence- falite e que apresentou no teste de inibição de hemaglutinação (IH) reação monotípica para EEL. Procurou-se também, na pesquisa, veri- ficar a proporção e as características dos indivíduos com sorologia positiva para outros arbovírus, a fim de observar se fatos evidenciados em outros grupos populacionais se repetiam nesse grupo, residente em zona urbana. As características da área e da população do Vale do Ribeira encontram-se descritas em trabalhos anteriores (Forattini e col.6, 1978, Iversson9,10, 1977, 1980). Os dados do inquérito sorológico anterior- mente realizado em pacientes do Hospital Regional de Pariquera-Açú, citados na Tabela 1, referem-se às 502 pessoas desse grupo que não tinham antecedente de ence- falite e não coabitavam com doente da mo- léstia. MATERIAL E MÉTODOS Entre os 821 casos de encefalite notificados na região do Vale do Ribeira no período de janeiro de 1975 a julho de 1978, 178 dis- punham dos resultados de testes sorológicos para pesquisa de anticorpos de arbovírus Rocio , dos quais 73 foram positivos, ou seja, apresentaram variação de 4 vezes ou mais no título de anticorpos entre 2 amostras de soro, colhidas nas fase aguda e na con- valescença, ou mostraram títulos altos e es- O sangue foi coletado por punção venosa. Os soros foram conservados a — 20°C até a realização dos testes sorológicos. A reação de inibição de hemaglutinação (IH) foi realizada para 18 antígenos de arbovírus isolados no Brasil, dos gêneros Alphavirus, Flavivirus, Bunyavirus e outros (Berge 1, 1975): encefalite eqüina do leste (EEL), encefalite eqüina do oeste (EEO), Mayaro, Mucamho, febre amarela (H 111, amostra silvestre e 17 D, vacina) Bussu- quara, Ilhéus, St. Louis, Rocio, Caraparu, Guaroa, Oropouche, Utinga, Icoaraci, Ta- caiuma e duas possíveis novas amostras, AN 327600 (Flavivirus) e AR 314206 (Bunyavirus). Foi utilizada microtécnica descrita por Shope21 (1963), os antígenos tendo sido preparados de cérebro de ca- mundongo recém-nascido, pela técnica da extração, com sucrose-acetona ou de soro de camundongo ou hamster por extração com acetona; os soros foram tratados por acetona e absorvidos por hemácias de ganso. amostra silvestre e 17 D, vacina) Bussu- quara, Ilhéus, St. Louis, Rocio, Caraparu, Guaroa, Oropouche, Utinga, Icoaraci, Ta- caiuma e duas possíveis novas amostras, AN 327600 (Flavivirus) e AR 314206 (Bunyavirus). Foi utilizada microtécnica descrita por Shope21 (1963), os antígenos tendo sido preparados de cérebro de ca- mundongo recém-nascido, pela técnica da extração, com sucrose-acetona ou de soro de camundongo ou hamster por extração com acetona; os soros foram tratados por acetona e absorvidos por hemácias de ganso. inoculadas intracerebralmente em camun- dongos albinos. O soro com um logarítmo do índice de neutralização > 1,7 foi consi- derado positivo. RESULTADOS De acordo com o mencionado na Tabela 1, a proporção de pessoas investigadas nesta pesquisa com anticorpos para o Flavivirus Rocio, 5/82, não difere significantemente (x2 = 0,603 < 3,841 para a = 0,05) da encontrada, 21/502, em pacientes do Hospi- tal Regional de Pariquera-Açú. Também não difere da observada, 5/83, em homens resi- dentes em ambiente silvestre. No teste de neutralização para EEL em- pregou-se técnica de soro constante (diluição final 1:8) com variação na diluição do vírus de 10 vezes (Casals5, 1967). As misturas, incubadas a 37°C por uma hora, foram O que chama a atenção nessa Tabela 1 é o fato dos moradores das zonas urbana e rural, comunicantes ou não de doentes de encefalite por Rocio, apresentarem no teste de IH reações de maior especificidade para Rocio do que os residentes em ambiente silvestre. famílias, em metade das quais foram exa- minados de três a sete comunicantes, onde não se observou a presença de pessoas com anticorpos para Rocio. Por outro lado, alguns dados, a seguir apresentados, referentes às famílias em que os comunicantes mostraram anticorpos para Rocio, sugerem que há maior probabilidade de a infecção, nesses casos, ter sido adqui- rida no local de trabalho ou no peridomi- cílio do que no domicílio. A distribuição nas residências dos comu- nicantes com anticorpos para Rocio (Tabela 2) também não sugere concentração de infectados em casa de doente. Quatro das cinco pessoas com anticorpos para Rocio pertencem a famílias diferentes. Há 20 - Família do doente P.D., estudante de 16 anos de idade, nascido em zona rural do município e residente há 3 anos em Ca- nanéia, zona urbana. Costumava acompanhar o pai em pescarias e caçadas. Foram exami- nados os soros de 6 irmãos de 4, 5, 8, 10, 14, 18 anos de idade, da avó de 77 anos e dos pais de 45 e 48 anos. Os 7 primeiros foram negativos. O pai, pescador profis- sional, e a mãe, prendas domésticas, nas- cidos em zona rural do município de Ca- nanéia e residentes há 13 anos na zona urbana, apresentaram anticorpos IH para arbovírus: o pai, reação sorológica mono- típica para o Flavivirus Ilhéus (título 40), e a mãe reação monotípica para o Flavi- virus Rocio (título 40). A casa dista menos de 50m do rio e mais de 100m da mata. anticorpos para arbovírus nos soros dos 2 irmãos. Não há entre os comunicantes investigados crianças abaixo de 4 anos cuja movi- mentação em geral se faz em torno do domicílio, mas cumpre lembrar que os me- nores de 8 anos em 1976, época que os familiares tiveram a encefalite, pertenciam àquele grupo etário. Os dados apresentados mencionam anticorpos para Rocio em um comunicante de 8 anos e em um doente de 7 anos, residentes no bairro do Rocio, zona urbana de Iguape. Nesse local, situado junto a orla marítima, foi grande o número de pessoas acometidas pela encefalite, in- cluindo o doente de onde se isolou pela primeira vez o Flavivirus Rocio. O bairro apresenta características rurais, com um casario de padrão simples, cercado por vege- tação arbustiva e ruas sem calçamento onde há quase constantemente água estagnada. Família do O.D., 15 anos de idade, sexo feminino, sempre residente no bairro do Rocio que, embora situado dentro dos limites urbanos da cidade litorânea de Iguape, apresenta características rurais. Essa doente participava de pescarias e de excursões à mata próxima para recolhi- mento de lenha. Dos 6 comunicantes exami- nados de 8, 10, 15, 16, 33 e 45 anos de idade, 3 apresentaram anticorpos para ar- bovírus: o de 8 anos mostrou anticorpos para Rocio (título 20), o de 15 anos, pes- cador profissional, anticorpos para ence- falite eqüina do leste (título 40) e o de 45 anos, também pescador profissional, reação múltipla para os Flavivirus Rocio (título 40), Ilhéus (título 20), febre amarela, amostra silvestre e 17 D (título 20). A casa dista menos de 50m do rio e menos de 100m da mata. Em relação aos anticorpos de outros arbovírus testados, foram observados em pessoas de 10 das 24 famílias estudadas, na freqüência expressa na Tabela 3. — Família de L.T., pedreiro e pescador, de 35 anos de idade, residente na zona urbana de Cananéia desde o nascimento. Dos 4 comunicantes pesquisados, 3 eram mulheres de 13, 45 e 60 anos e um homem de 74 anos de idade, aposentado. Só o último, residente no município, zona rural e urbana, desde o nascimento, apresentou anticorpos para os Flavivirus Ilhéus (título 20) e Rocio (título 80). — Família da doente C.R.F. de 7 anos de idade, nascida e sempre residente no bairro do Rocio, anteriormente mencionado. A criança costumava nadar em rio junto à casa e brincar no peridomicílio, onde existem valetas de água estagnada. Foram pesquisados dois irmãos de 13 e 23 anos de idade e os pais de 42 e 53 anos. O pai, trabalhador braçal, apresentou reação múl- tipla para os Flavivirus Rocio (título 20), Ilhéus (título 40) e da febre amarela, amostra 17D (título 20). A mãe, prendas domésticas, apresentou anticorpos para Caraparu (título 20). Não se encontraram Entre as três pessoas que referiram ante- cedente de encefalite em 1976, só uma apresentou anticorpos para arbovírus ELL (reação monotípica). Entre as duas vaci- nadas contra febre amarela uma mostrou anticorpos para a amostra 17 D. rito somente anticorpos para EEL (título 20). O teste de neutralização em camundongo para esse Alphavirus foi positivo (log IN = 3,16). Os indivíduos do sexo masculino mos- traram maior proporção de anticorpos para arbovírus do que os do sexo femi- nino. Assim, 9 em 36 pessoas do sexo masculino (25%) e 6 em 46 do sexo feminino (13%) tiveram sorologia posi- tiva (Tabela 4). Outro objetivo deste levantamento soro- lógico foi observar se no grupo populacional estudado, residente em zona urbana, se repetiam fatos verificados em estudo ante- rior em que foram investigados sorologica- mente 516 residentes em zona urbana e rural da região (Iversson e col.12, 1981). Isto realmente se verificou: 3. 3. O número de pessoas com anticorpos para arbovírus aumentou com a idade (Tabela 4). 1. Um dos comunicantes do doente V.C., pedreiro de 25 anos de idade, nascido em zona rural, residente há 8 anos em zona urbana, referiu antecedente de encefalite em 1976, diagnosticada em hospital da região, por critério clíni- co-epidemiológico, como encefalite por Rocio. Apresentou no presente inqué- Anticorpos do Bunyavirus Caraparu foram evidenciados em 3 pessoas: duas crianças de 9 a 10 anos e uma mulher de 45 anos, prendas domésticas. Essas pessoas não apresentaram anticorpos para outros arbovírus. Chama a atenção COMENTÁRIOS E CONCLUSÕES O presente trabalho não invalida a ocor- rência de transmissão domiciliar da ence- falite por Rocio, sugerida por estudos ba- seados nos dados clínicos e entomológicos (Tiriba 2 2, 1975; Forattini e col.7 1978; Iversson 9, 1980), mas faz crer que na zona urbana esta ocorrência, se existiu, não foi freqüente. Outro aspecto a ser comentado é a pre- sença de um comunicante com antecedente de encefalite, que apresentou no presente inquérito anticorpos só para EEL, o que obviamente não faz diagnóstico pregresso da moléstia. Esse achado, no entanto, também já observado em outro grupo populacional pesquisado (Iversson e col.12. 1981), traz à tona novamente a hipótese que alguns dos casos de encefalite atribuídos ao vírus Rocio possam ter sido causados por outros arbo- vírus. O Alphavirus EEL já foi recentemente isolado na área de pools de Culex (Melano- conion) sp, (Calisher e col.3, 1980), culi- cídeo de hábitos silvestres, encontrado em densidade significante em domicílios da região (Forattini e col.7, 1978). Até o presente esse arbovirus não tinha sido res- ponsabilizado por moléstia humana ou eqüina na região. Levando em conta que os pescadores se apresentaram infectados em alta proporção por Alphavirus e Flavivirus, informação já evidenciada em inquérito sorológico anterior (Iversson e col.12, 1981), delinea-se a hipó- tese que, para esses homens, a transmissão das arboviroses esteja se fazendo fora do ambiente florestal. De fato, as pessoas re- feridas como pescadores nesses inquéritos sorológicos são as que exercem profissional- mente só esta atividade. Não foram incluídos os lavradores ou outros profissionais que, para aumentar sua renda, trabalham também em pesca, em determinados períodos do ano. O cultivo da lavoura de mandioca ou cereais por pescadores é feito apenas para consumo próprio. A incursão desses homens no ambiente silvestre é rara, realizada apenas para coleta de taquara (Bambusa vulgaris) para a manufatura de "cercos", armadilhas para captura de peixes, ou para eventual caça de animais para sua alimen- tação. Sua atividade diária é exercida no rio Ribeira de Iguape e afluentes, no Mar Pequeno, canal entre o continente e a ex- tensa Ilha Comprida ou no mar aberto. Em geral, iniciam seu trabalho à tarde e o terminam no início da manhã. colocam no fim da tarde retornando para a coleta do peixe pela manhã. São portanto pessoas que se expõem diariamente à picada de insetos hematófagos no horário mencio- nado, fora da mata. o fato de que indivíduos do sexo mas- culino, que exercem maior atividade tora do domicílio, não tivessem anti- corpos para este vírus. 5. Entre os pescadores investigados, uma proporção de 60% apresentou anti- corpos para arbovírus (Tabela 5), mais especificamente para EEL, Rocio e Ilhéus. 5. Entre os pescadores investigados, uma proporção de 60% apresentou anti- corpos para arbovírus (Tabela 5), mais especificamente para EEL, Rocio e Ilhéus. 5. Entre os pescadores investigados, uma proporção de 60% apresentou anti- corpos para arbovírus (Tabela 5), mais especificamente para EEL, Rocio e Ilhéus. Como o número de pescadores profis- sionais investigados foi baixo, um estudo dirigido só para este grupo poderia escla- recer o assunto. Há interesse em assinalar o comporta- mento dos anticorpos do Bunyavirus Cara- paru (Grupo C) que, ou não são encontra- dos nos homens ou quando aparecem o fazem na mesma proporção de que nas mulheres (Iversson e col.12, 1981), sugerindo que a transmissão desse vírus se faça no peri- domicílio ou domicílio. COMENTÁRIOS E CONCLUSÕES Quando utilizam as chamadas "redes de espera", as Porém uma das informações mais rele- vantes deste, como de outros inquéritos anteriores (Iversson e col.11,12, 1980, 1981) é a baixa proporção de pessoas na região que apresentaram anticorpos para o Flavi- virus Rocio (4,2% a 6,1%) assim como Flavivirus em geral (9,8% a 13,2%), tendo em vista a extensa epidemia ocorrida. É bem verdade que os grupos pesquisados não constituem, cada um deles isolado, uma amostra representativa da população do Vale do Ribeira. Mas, em conjunto, consi- derando que reúnem pessoas residentes em todos os municípios atingidos pela epidemia, zonas urbana, rural e ambiente silvestre, mostram pelo menos uma tendência da fre- qüência de anticorpos para Rocio e Flavi- virus em geral. -se em observações próprias ou de outros pesquisadores, consideram, entre outros fatores, o papel desempenhado pelos ventos no transporte passivo entre áreas geográ- ficas, algumas vezes distantes, dos vetores de virus responsáveis por moléstias no homem e em outros animais. Esses aspectos não foram estudados na região do Vale do Ribeira, assim como também não existem informações que per- mitam saber se ocorreu mudança na viru- lência ou no neurotropismo dos agentes etiológicos, Rocio ou possíveis outros arbo- vírus. De acordo com os dados obtidos, não se pode raciocinar que a propagação em onda (Iversson10, 1977) e o término da epidemia estivessem ligados a um esgota- mento de susceptíveis na população humana. Estariam esses fatos ligados só a dinâmica populacional dos vetores e/ou dos reserva- tórios? O combate às formas larvárias e aladas dos possíveis vetores foi realizado no interior e em torno dos núcleos urbanos. Na extensa zona rural foi realizado com- bate às formas aladas só no local das resi- dências de doentes conhecidos (SUCEN ). Que mecanismos naturais, alterações climá- ticas e de cobertura vegetal, ou outros, teriam atuado sobre os reservatórios e os vetores? Por exemplo, Sellers e colabora- dores em seus trabalhos (Sellers e col.17,18,19 1977, 1978, 1979; Sellers16, 1980) apoiando- São questões ainda sem resposta que sugerem a necessidade de estudos continua- dos na área. * Informação pessoal do Dr. George Kenge Ishihata, Superintendente da SUCEN (Superintendência de Controle de Endemias). IVERSSON, L.B. et al. [Serological studies for research of arbovirus antibodies in human population of the Ribeira Valley. III — Survey among persons cohabiting with encephalitis cases by Flavivirus Rocio]. Rev. Saúde públ., S. Paulo, 16: 160-70, 1982. REFERÊNCIAS BIBLIOGRÁFICAS Hemorrágicas e Arbovírus dos Trópicos, Belém, 1980]. 1. BERGE, T.O., ed. International catalogue of arboviruses. 2nd ed. Washington, D.C.. US Dept. Health. 1975. (Publ. 75-8301). 12. IVERSSON, L.B. et al. Estudos sorológicos para pesquisa de anticorpos de arbovírus em população humana da região do Vale do Ribeira. II — Inquérito em pacientes do Hospital Regional de Pariquera-Açú. Rev. Saúde públ., S. Paulo, 15:557-86, 1981. 2. BRODY, J.A. et al. Apparent and inappa- rent attach rates for St. Louis encepha- litis in a selected population. N. Eng. J. Med., 261:644-6, 1959. 3. CALISHER, C.H. et al. Isolations of new Alpha and Bunyaviruses of Southern Brazil: proposed reclassification of sero- groups. [Trabalho apresentado ao Sim- pósio Internacional de Febres Hemor- rágicas e Arbovírus dos Trópicos, Belém, 1980]. 13. LOPES, O. de S. et al. Emergence of a new arbovirus disease in Brazil. II — Epidemiologic studies on 1975 epidemic. Amer. J. Epidem., 108:394-401, 1978. 14. MARTIN, D.H. et al. An epidemiologic study of Venezuelan equine encephalo- myelitis in Costa Rica. Amer. J. Epidem., 95:565-78, 1972. 4. CASALS, J. & CLARKE, D.H. Arbovirus. Group A. In: Horsfall, F.L. & Tamm, I., ed. Viral and richettsial inflections of man. Philadelphia, Lippincott, 1965. p. 583-605. 15. MONATH, T.P. Arthropod-borne encepha- litides in the Americas. Bull. Wld Hlth Org., 57:513-33, 1979. 5. CASALS, J. Immunological techniques for animal viruses. In: Maramorosh, K. & Koprowski, H. Methods in virology. New York, Academic Press, 1967. v. 3. p. 175-81. 16. SELLERS, R.F. Weather host and vector — their interplay in the spread of insect-borne animal virus diseases. J. Hyg., Cambridge, 85:65-102, 1980. 6. FORATTINI, O.P. et al. Estudos ecológicos sobre mosquitos Culicidae no Sistema da Serra do Mar, Brasil. 1 — Observações no ambiente extradomiciliar. Rev. Saúde públ., S. Paulo, 12:297-325, 1978. 17. SELLERS, R.F. et al. Possible origin of the bluetongue epidemic in Cyprus, August 1977. J. Hyg., Cambridge, 83:547- -55, 1979. 7. FORATTINI, O.P. et al. Estudos ecológicos sobre mosquitos Culicidae no Sistema da Serra do Mar, Brasil. 2 — Observações no ambiente domiciliar. Rev. Saúde públ., S. Paulo, 12:476-96, 1978. 18. SELLERS, R.F. et al. Possible spread of African horse sickness on the wind. J. Hyg., Cambridge, 79:279-98, 1977. 19. SELLERS, R.F. et al. Possible windborne spread of bluetongue to Portugal, June- -July 1956. J. Hyg., Cambridge, 81:189- -96, 1978. 8. HALSTEAD, S.B. & GROSZ, C.R. Subcli- nical Japanese encephalitis. AGRADECIMENTO Ao Dr. Geraldo Henrique Pinto, Diretor da Divisão Regional de Saúde do Vale do Ribeira (DEVALE), pelo valioso auxílio proporcionando facilidades durante o tra- balho de campo. IVERSSON, L.B. et al. [Serological studies for research of arbovirus antibodies in human population of the Ribeira Valley. III — Survey among persons cohabiting with encephalitis cases by Flavivirus Rocio]. Rev. Saúde públ., S. Paulo, 16: 160-70, 1982. ABSTRACT: A serological survey for hemagglutination-inhibition anti- bodies to 17 arbovirus was carried out in households with cases of Rocio encephalitis, in the urban zone of four cities of the Ribeira Valley, Brazil, where an epidemic of Rocio encephalitis occurred recently. Among those households the prevalence of Rocio antibodies was not higher than in house- holds without cases of encephalitis. Some facts, which were reported before, were again observed: a large prevalence of antibodies in men, particulary fishermen, an increase of antibodies with age and the presence of one past case of encephalitis that presented only neutralizing antibodies against EEE. That Alphavirus has never been responsible for human disease in the area. There is also a very small proportion of people with Rocio and Flavivirus antibodies which, in view of the recent epidemic, was surprising. UNITERMS: Arboviruses. Serological surveys. Encephalitis, epidemic. Flavivirus Rocio. UNITERMS: Arboviruses. Serological surveys. Encephalitis, epidemic. Flavivirus Rocio. REFERÊNCIAS BIBLIOGRÁFICAS I — Infection of Americans with limited residence in Korea. Amer. J. Hyg., 75:190-201, 1962. 20. SOUTHAM, C.H. Serological studies of en- cephalitis in Japan. II — Inapparent in- fection by Japonese B encephalitis virus. J. infect. Dis., 99:163-9, 1956. 9. IVERSSON, L.B. Aspectos da epidemia de encefalite por arbovírus na região do Vale do Ribeira, São Paulo, Brasil, no período de 1975 a 1978. Rev. Saúde públ., S. Paulo, 14:9-35, 1980. 21. SHOPE, R.E. The use of micro-hemagglu- tination-inhibition test to follow antibody response after arthropodborne virus infection in a community of forest animals. An. Microbiol., 11:167-71, 1963. 10. IVERSSON, L.B. Epidemia de encefalite por arbovírus na região Sul do Estado de São Paulo, Brasil, em 1975 e 1976: aspectos da distribuição cronológica e geográfica dos casos. Rev. Saúde públ., S. Paulo, 11:375-88, 1977. 22. TIRIBA, A. de C. Epidemia de encefalite atribuída a arbovírus ocorrida no litoral Sul do Estado de São Paulo, em 1975; contribuição para o estudo clínico. São Paulo, 1975. [Tese de Livre Docência —- Escola Paulista de Medicina]. 11. IVERSSON, L.B. et al. Estudos sorológicos para pesquisa de anticorpos de arbovírus em população humana da região do Vale do Ribeira. I — Seguimento sorológico de grupo populacional residente em ambiente silvestre. [Trabalho apresentado ao Simpósio Internacional de Febres Recebido para publicação em 18/01/1982 Aprovado para publicação em 14/04/1982
https://openalex.org/W3135500412	https://zenodo.org/records/4670173/files/1-s2.0-S1361920921000687-main.pdf	English	null	The climate change mitigation effects of daily active travel in cities	Transportation research. Part D, Transport and environment	2,021	cc-by	18,022	The climate change mitigation effects of daily active travel in cities Christian Brand a,b,, Evi Dons c,d, Esther Anaya-Boig e, Ione Avila-Palencia f,g, Anna Clark h, Audrey de Nazelle e, Mireia Gascon f,i,j, Mailin Gaupp-Berghausen k, Regine Gerike l, Thomas G¨otschi m, Francesco Iacorossi n, Sonja Kahlmeier o,p, Michelle Laeremans c,t, Mark J Nieuwenhuijsen f,i,j, Juan Pablo Orjuela a,e, Francesca Racioppi q, Elisabeth Raser u, David Rojas-Rueda f,s, Arnout Standaert c, Erik Stigell h, Simona Sulikova a, Sandra Wegener r, Luc Int Panis c,d,t a Transport Studies Unit, University of Oxford, Oxford, United Kingdom b Centre for Research on Energy Demand Solutions, University of Oxford, Oxford, United Kingdom c Flemish Institute for Technological Research (VITO), Mol, Belgium e Centre for Environmental Policy, Imperial College London, London, United Kingdom g Urban Health Collaborative, Dornsife School of Public Health, Drexel University, Philadelphia, USA h Trivector Traffic Stockholm Sweden h Trivector Traffic, Stockholm, Sweden i i Universitat Pompeu Fabra (UPF), Barcelona, Spain Corresponding author at: Transport Studies Unit, University of Oxford, South Parks Road, Oxford OX1 3QY, United Kingdom. E-mail address: christian.brand@ouce.ox.ac.uk (C. Brand). Available online 27 February 2021 1361-9209/© 2021 Published by Elsevier Ltd. https://doi.org/10.1016/j.trd.2021.102764 Transportation Research Part D 93 (2021) 102764 Transportation Research Part D 93 (2021) 102764 1. Introduction location, socio-economic status, car ownership, type of car, bike access, perceptions related to the safety, convenience and social status associated with active travel), land use and built environment factors (which impact journey lengths and trip rates), accessibility to public transport, jobs and services, and meteorological conditions (Adams, 2010; Alvanides, 2014; Anable and Brand, 2019; Bearman and Singleton, 2014; Brand and Preston, 2010; Buehler, 2011; Cameron et al., 2003; Carlsson-Kanyama and Linden, 1999; Javaid et al., 2020; Ko et al., 2011; Nicolas and David, 2009; Stead, 1999; Timmermans et al., 2003). However, to answer the above question on the role of active travel in this, it is important to understand why, where, when and how far people travel by active and motorized modes of travel. Many studies do not dig that deep and across different contexts, thus presenting us with a number of shortcomings. First, many studies are based on analyses of the potential for emissions mitigation by developing hypothetical mitigation scenarios (see e.g. Goodman et al., 2019; Lovelace et al., 2011; Tainio et al., 2017; Woodcock et al., 2018; Yang et al., 2018). While this is useful in its own right, it sometimes lacks the empirical evidence needed to underpin the hy potheses. Second, the majority of studies are focusing on a single city, region or country (Brand et al., 2014; Neves and Brand, 2019), thus limiting the potential for generalizability and transferability to other populations and contexts. Third, the majority of active travel studies tackle carbon emissions from end-use only (Goodman et al., 2019), thus omitting the full life cycle impacts of (active and motorized) travel. It is worth noting that, on a life cycle emissions basis, cycling cannot be considered a ‘zero-carbon emissions’ mode of transport. The manufacture, maintenance and disposal of bikes (plus batteries and motors in the case of e-bikes), as well as potential dietary changes due to more physical activity, produce carbon emissions further up- or downstream. Overall, however, life cycle emissions from cycling can be more than ten times lower per passenger-km travelled than those from passenger cars (ECF, 2011). 1. Introduction Transport has been one of the most challenging sectors for reducing its significant impacts of fossil energy use and associated greenhouse gas (GHG) emissions since the 1990s (Sims et al., 2014). In Europe, GHG emissions decreased in the majority of sectors between 1990 and 2017, with the exception of transport (EEA, 2019). Modal shifts away from carbon-intensive to low-carbon modes of travel hold considerable potential to mitigate carbon emissions (Cuenot et al., 2012). Given the urgency of moving to a ‘net zero’ carbon emissions economy, there is growing consensus that technological substitution via electrification will not be sufficient or fast enough to transform the transport system (Creutzig et al., 2018; IPCC, 2018). Beyond a net reduction in travel demand, one of the more promising ways to reduce transport carbon dioxide (CO2) emissions1 is to promote and invest in ‘active travel’ (that is, walking, cycling, or e-biking for transport) while ‘demoting’ motorized modes that rely on fossil energy sources (Bearman and Singleton, 2014; Castro et al., 2019; de Nazelle et al., 2010; ECF, 2011; Frank et al., 2010; Goodman et al., 2012; Keall et al., 2018; Neves and Brand, 2019; Quarmby et al., 2019; Sælensminde, 2004; Scheepers et al., 2014; Tainio et al., 2017; Woodcock et al., 2018). This could reduce CO2 emissions from road transport more quickly than technological measures alone, particularly in urban areas (Beckx et al., 2013; Creutzig et al., 2018; Graham-Rowe et al., 2011; Neves and Brand, 2019). This may become even more relevant considering the vast economic effects of the COVID-19 pandemic, which may result in reduced capacities of individuals and organizations to renew the rolling stock of vehicles in the short and medium period, and of governments to provide incentives to fleet renewal. l So how much carbon can be saved – overall – by active travel? The complex relationships between carbon emissions and transport in general have been investigated for many years. From previous research we know that mobility-related carbon emissions are determined by transport mode choice and usage, which in turn are influenced by journey purpose (e.g. commuting, visiting friends and family, shopping), individual and household characteristics (e.g. A R T I C L E I N F O Keywords: CO2 emissions Active mobility Walking Cycling Climate change mitigation Sustainable urban transport Active travel (walking or cycling for transport) is considered the most sustainable form of per sonal transport. Yet its net effects on mobility-related CO2 emissions are complex and under- researched. Here we collected travel activity data in seven European cities and derived life cycle CO2 emissions across modes and purposes. Daily mobility-related life cycle CO2 emissions were 3.2 kgCO2 per person, with car travel contributing 70% and cycling 1%. Cyclists had 84% lower life cycle CO2 emissions than non-cyclists. Life cycle CO2 emissions decreased by −14% per additional cycling trip and decreased by −62% for each avoided car trip. An average person who ‘shifted travel modes’ from car to bike decreased life cycle CO2 emissions by 3.2 kgCO2/day. Available online 27 February 2021 1361-9209/© 2021 Published by Elsevier Ltd. https://doi.org/10.1016/j.trd.2021.102764 Available online 27 February 2021 1361-9209/© 2021 Published by Elsevier Ltd. https://doi.org/10.1016/j.trd.2021.102764 C. Brand et al. Transportation Research Part D 93 (2021) 102764 Promoting active travel should be a cornerstone of strategies to meet net zero carbon targets, particularly in urban areas, while also improving public health and quality of urban life. 1 For transport, CO2 is by far the most important greenhouse gas, comprising approximately 99% of direct greenhouse gas emissions. Surface transport is still dominated by vehicles with internal combustion engines running on petrol (gasoline) and diesel fuels. These propulsion systems emit relatively small amounts of the non-CO2 greenhouse gases methane (CH4) and nitrous oxide (N2O), adding approximately 1% to total greenhouse gas emissions over and above CO2. 0.6 0.2 0.1 EA/AN ares of l fleet all trip bus/ra ormal 1. Introduction y g p p g p g ( , ) To address these shortcomings and better understand the carbon-reduction impacts of active travel, it is thus important to assess the key determinants of travel carbon emissions across a wide range of contexts and include a detailed, comparative analysis of the distribution and composition of emissions by transport mode (e.g. bike, car, van, public transport, e-bike) and emissions source (e.g. vehicle use, energy supply, vehicle manufacturing). For most journey purposes active travel covers short to medium trips – typically 2 km for walking, 5 km for cycling and 10 km for e-biking (Castro et al., 2019). Typically, the majority of trips in this range is made by car (Beckx et al., 2013; JRC, 2013; Keall et al., 2018; Neves and Brand, 2019; U.S. Department of Transportation, 2017), with short trips contributing disproportionately to emissions because of ‘cold starts’, especially in colder climates (Beckx et al., 2010; de Nazelle et al., 2010). On the other hand, these short trips, which represent the majority of trips undertaken by car within cities, would be amenable to at least a partial modal shift towards active travel (Beckx et al., 2013; Carse et al., 2013; de Nazelle et al., 2010; Goodman et al., 2014; Keall et al., 2018; Neves and Brand, 2019; Vagane, 2007). To investigate these issues, we included seven European cities with different travel activity patterns, transport mode shares, infrastructure provisions, climates, mobility cultures and socio-economic makeups. To the best of our knowledge no international multicenter study on the associations of daily active and motorized travel and carbon emissions has been reported. 1 For transport, CO2 is by far the most important greenhouse gas, comprising approximately 99% of direct greenhouse gas emissions. Surface transport is still dominated by vehicles with internal combustion engines running on petrol (gasoline) and diesel fuels. These propulsion systems emit relatively small amounts of the non-CO2 greenhouse gases methane (CH4) and nitrous oxide (N2O), adding approximately 1% to total greenhouse gas emissions over and above CO2. 2 2 Pub gCO mea 2.9 3.0 3.6 4.2 3.1 2.8 3.4 MT, 20 uplift w buses w stance o tp://ww Pu gC m 2. 3. 3. 4. 3. 2. 3. 014–2016 average ransport † Bicy km gCO SD mea 0.5 0.11 0.6 0.17 0.6 0.23 0.1 0.01 0.6 0.20 0.2 0.08 0.1 0.00 ACEA/ANFAC, 201 shares of the vehic onal fleet mix of 38 or all trip purpose a bus/rail share b normal bike. 201 c tran /pkm (ACE el sha tiona 4 for me a b % no C. Brand et al. This study also addresses a number of practical needs. First, there is a lack of standardized definitions and measurements (self- reported or measured) to identify groups within a population who use a ‘main’ mode of transport (e.g. based on distance, duration or frequency over a given time period), or who may be classified as ‘frequent cyclists’, ‘occasional walkers’ – or simply ‘cycling’ (yes/no). These should be split as much as possible as there may be different effects on overall CO2 emissions. Second, given the dominance of travel by car and public transport, active modes must be included to the extent possible by oversampling people using these modes. And finally, instead of focusing on the commute journey only, as with many studies that rely on Census data, trips for a wider range of journey purposes should be considered. This paper aims to investigate to what extent active travel is associated with lower carbon emissions from daily travel activity. Using primary data collected in a large European multicenter study of transport, environment and health, the paper first describes how total life cycle CO2 emissions from daily travel activity were derived at the individual and population levels, considering urban transport modes, trip stages, trip purposes and emissions categories. The core analysis then identifies the main determinants and models the effects of mode choice and usage on life cycle carbon emissions. Further analysis identifies and compares differences in life cycle carbon emissions between ‘groups of transport users’, including by ‘main’ mode of transport and different categories of cycling frequency. By doing so, the paper provides a detailed and nuanced assessment of the benefits of active travel in reducing total life cycle carbon emissions in cities. 2.1. Study design and population This study used longitudinal data from the ‘Physical Activity through Sustainable Transport Approaches’ (PASTA) project (Dons et al., 2015; Gerike et al., 2016). The analytical framework of PASTA distinguished hierarchical levels for various factors (i.e. city, individual, and trips), and four main domains that influence mobility behavior, namely factors relating to transport mode choice and use, socio-demographic factors, socio-geographical factors, and socio-psychological factors (Dons et al., 2015; G¨otschi et al., 2017). Seven European cities (Antwerp, Barcelona, London, Orebro, Rome, Vienna, and Zurich) were selected to provide a good represen tativeness of urban environments in terms of size, built environment, transport provision, modal split and ambition to increase levels of active travel (Raser et al., 2018). To ensure sufficiently large sample sizes for different transport modes, users of less common transport modes such as cycling were oversampled (Raser et al., 2018). Participants were recruited opportunistically on a rolling basis following a standardized guidance for all cities and also some city-specific approaches. A comprehensive user engagement strategy was applied to minimize attrition over the two-year timeframe. Further details on the recruitment strategy are given elsewhere (Gaupp-Berghausen et al., 2019). A total of 10,722 participants entered the study on a rolling basis between November 2014 and November 2016 by completing a baseline questionnaire (BLQ). Participants provided detailed information on general travel behavior, daily travel activity, geolocations (home, work, education), vehicle ownership (private motorized, bicycle, etc.), public transport accessibility and socio-demographic characteristics. Follow-up questionnaires were distributed every two weeks: every third of these follow-up questionnaires also included a one-day travel diary, henceforth labelled a ‘long follow-up’ (long FUQ) (Dons et al., 2015). All valid travel diaries were included in the analyses, with participants providing multiple diary data at different time points. Using longitudinal data of repeated measurements of daily travel activity has been shown to improve measurement of ‘typical’ travel behavior (Branion-Calles et al., 2019). Participants had to be 18 years of age (16 years in Zurich) or older, and had to give informed consent at registration. Data handling and ethical considerations regarding confidentiality and privacy of the information collected were reported in the study protocol (Dons et al., 2015). Table S2 in the Supplementary Information provides an excerpt of the PASTA BLQ, including travel diary data. 1. Introduction MT, 2 uplift buses tance p://w P gC m 2 3 3 4 3 2 3 MT, 2 uplift buses stance p://w P g m 2 3 3 4 3 2 3 MT, uplif buse stanc p:// ng 2014–2016 averages to match study period). Vehicle supply blic transport † Bicycle, e-bike # Car, van, motorcycle * O2/pkm gCO2/pkm gCO2/pkm an SD mean SD mean SD 9 0.5 0.111 0.003 12.9 1.9 5 0.6 0.171 0.004 10.6 1.6 5 0.6 0.236 0.004 12.9 1.9 7 0.1 0.011 0.000 13.2 2.1 9 0.6 0.200 0.004 11.5 1.8 6 0.2 0.087 0.002 13.3 2.0 0.1 0.002 0.000 12.5 1.8 l ts (ACEA/ANFAC, 2014; DEFRA/DECC, 2016; DfT, 2015; SMMT, fuel shares of the vehicle fleet, and cold start emissions. A 22% uplif national fleet mix of 38% gasoline, 61% diesel and 1% electric; bus 54 for all trip purposes. Cold/hot ratio of 1.3 and cold trip distan ume a bus/rail share based on EPOMM Modal Split Tool. (http:// .5% normal bike. † Bicy gCO mea 0.1 0.17 0.2 0.0 0.20 0.08 0.00 NFAC, 201 f the vehic mix of 38 p purpose ail share b bike. Transportation Research Part D 93 (2021) 102764 C. Brand et al. C. Brand et al. 2.2. Exposure: transport mode choice and use The primary exposure variables were daily trip frequencies obtained from the travel diaries, for each of the main modes: walking; cycling; e-biking; motorcycle or moped; public transport; and car or van. The most common metric used by local and national ad ministrations across the world is mode share (or split) by trip frequency, not by distance (EPOMM, 2020; U.S. Department of Transportation, 2017); hence the results of the primary exposure analysis may be used to estimate life cycle CO2 emissions directly from trip mode share data. Due to low counts of e-biking and motorcycle trips, e-biking was merged with cycling, with indirect emissions derived from observed bike/e-bike shares (see also footnote of Table 1). Also, motorcycle was merged with car as reported CO2 emission rates for motorcycles are comparable to cars on a per passenger-km basis (BEIS, 2019). Participants provided information on each trip made on the previous day, including start time, location of origin, transport mode, trip purpose, location of destination, end time and duration (Supplementary Table S2). The diary was based on the established KONTIV-Design (Br¨og et al., 2009; Socialdata, 2009), with some adaptations for online use. 5623 participants provided a valid travel diary in either the BLQ or the long FUQ; out of those 3836 participants completed valid baseline surveys and travel diaries. In the travel diary, trip purpose, duration and location were self-reported. Total trip duration was also derived as the difference between start and end time, while trip distance was obtained retrospectively feeding origin and destination coordinates to the Google Maps Application Programming Interfaces (API), which returned the fastest route per mode between origin and destination. Three secondary exposure variables were developed to explore differences between groups of individuals. First, participants were categorized as using a ‘main mode’ of travel based on furthest daily distance (levels: walking, cycling, car, public transport). Further Transportation Research Part D 93 (2021) 102764 C. Brand et al. Table 2 Summary statistics of outcomes, exposures and other covariates. 2.3. Outcome variables: carbon dioxide emissions 2.3. Outcome variables: carbon dioxide emissions The primary outcome (or dependent variable) of interest was daily life cycle CO2 emissions (mass of carbon dioxide in gram or kilogram per day) attributable to passenger travel. Life cycle CO2 emissions categories considered were operational emissions, energy supply emissions and vehicle production emissions. First, operational emissions were derived for each trip based on trip distance (computed from travel diary data), ‘hot’ carbon emissions factors, emissions from ‘cold starts’ (for cars only) and vehicle occupancy rates (passengers/vehicle) that varied by trip purpose. The method for cars and vans considered mean trip speeds (derived from the travel diaries), location-specific vehicle fleet compositions (taking into account the types of vehicle operating in the vehicle fleets during the study period) and the effect of ‘real world driving’ (adding 22% to carbon emissions derived from ‘real world’ test data based on ICCT (2017)) to calculate the so called ‘hot’ emission of CO2 emitted per car-km. For motorcycle, bus and rail, operational emissions per vehicle-km were taken from BEIS (2019). The use of data from a UK study is a reasonable assumption as vehicle technology does not vary significantly across Europe, given the international nature of the vehicle market which is dominated by a relatively small number of manufacturers. An average urban bus in 2016 used about 34.2 L of diesel per 100 km and emitted about 1.24 kgCO2/vehicle-km. The growing share of hybrid ICE (up to 20%, depending on the city) and battery electric (only small at present) buses in the fleets is gradually improving these figures, which is taken into account in the study. Vehicle fuel type shares (gasoline, diesel, electric, other) and average occupancy rates for each city were based on international databases, including IIASA’s Greenhouse gas Air pollution INteractions and Synergies (GAINS) and Eurostat’s transport and energy data (Eurostat, 2016, 2020; IEA, 2015; IIASA, 2014). During the study period, buses were mainly powered by diesel powertrains; motorcycles were 100% gasoline; and urban rail was 100% electric. Public transport occupancy rates varied somewhat across cities. For instance for city buses, we assumed 28 passengers/vehicle in Barcelona, 21 in London, 17 in Rome, and 9 in ¨Orebro (Adra et al., 2004; BEIS, 2019; EEA, 2016). Together with bus/rail split in trip rates obtained from EPOMM (2020), this translated into city-specific per passenger-km emission rates, as shown in Table 1. 2.2. Exposure: transport mode choice and use Total study sample (n = 9858) and mean (SD) values CO2 emissions All modes, life cycle 3.18 (7.68) (kg per day) Car, life cycle 2.23 (7.25) Public transport, life cycle 0.93 (2.90) Bike, life cycle 0.03 (0.05) Walk, life cycle 0 (–) Al modes, direct only ^ 2.22 (5.62) All modes, indirect only ^ 0.96 (2.20) Transport mode usage Car 0.69 (1.29) (trips per day) Public transport 0.90 (1.24) Bike 1.05 (1.58) Walk 0.82 (1.36) All modes 3.47 (1.83) Average distance travelled Car 14.61 (50.32) (km per day) Public transport 15.51 (43.62) Bike 5.06 (9.71) Walk 0.88 (2.08) All modes 36.06 (63.51) Average travel time (min/day) All modes 87.84 (70.45) Age (years) All 39.19 (11.16) BMI (kg/m2) All 23.66 (3.83) Sub samples/groups and mean(SD) values of main outcome measure Exposures Life cycle CO2 (mean (SD)), in kg/day n (%) Main mode Car 9.139 (12.532) 2307 (23%) (based on distance) Public transport 2.746 (5.292) 3546 (36%) Bike 0.169 (0.468) 3012 (31%) Walk 0.031 (0.159) 993 (10%) Cycling category Non-cyclist (none) 4.438 (8.892) 6031 (61%) (based on trips per day) Occasional cyclist (once or twice) 1.517 (5.552) 2329 (24%) Frequent cyclist (thrice or more) 0.708 (2.343) 1498 (15%) Cycling (yes/no) Not cycling on the day 4.438 (8.892) 6031 (61%) Cycling on the day 1.201 (4.589) 3827 (39%) City Antwerp 3.487 (7.763) 1713 (17%) Barcelona 2.468 (5.792) 1806 (18%) London 3.209 (7.788) 1027 (10%) Oerebro 4.559 (9.451) 607 (6%) Rome 3.929 (10.012) 1061 (11%) Vienna 2.651 (6.153) 1752 (18%) Zurich 3.199 (8.16) 1892 (19%) Sex Male 3.305 (8.043) 5061 (51%) Female 3.051 (7.282) 4797 (49%) Age (for sensitivity analysis) Age <35 years 2.903 (6.398) 4199 (43%) Age >=35 years 3.387 (8.507) 5659 (57%) Age >55 years 3.807 (9.551) 981 (10%) Self-rated health Excellent 3.197 (7.857) 1036 (10%) Very good 3.074 (7.854) 4221 (43%) Good 3.331 (7.575) 3839 (39%) Fair or poor 3.001 (6.998) 762 (8%) BMI (for sensitivity analysis) Healthy BMI (18.5<=BMI<25) 3.019 (7.307) 6927 (70.3%) Overweight (BMI>=25) 3.599 (8.649) 2599 (26.4%) Household income Low income (Less than €25k) 2.884 (7.436) 2713 (28%) Middle income (€25–€75k) 3.176 (7.449) 5535 (56%) High income (€75k or more) 3.699 (8.503) 1610 (16%) Employment status Working (full-time or part-time) 3.241 (7.761) 8404 (85%) Not working (retired/student/etc.) 2.838 (7.208) 1454 (15%) Education level Higher education or degree 3.124 (7.261) 7814 (79%) No higher education or degree 3.401 (9.118) 2044 (21%) Household composition HH two or more adults, no kids 3.156 (7.462) 4788 (49%) Single HH, no kids 2.778 (6.133) 1750 (18%) HH with kids 3.431 (8.662) 3320 (34%) Car accessibility Always or sometimes 3.561 (8.093) 7755 (79%) Never 1.781 (5.719) 2103 (21%) ^ Di il i i i ^ I di ll k (f l/ d i ) l hi l f BMI b d 5 C. 2.4. Covariates A number of covariates (independent variables) were hypothesized to confound the association between carbon emissions and transport mode choice and use (e.g. Brand et al., 2013; Büchs and Schnepf, 2013; Goodman et al., 2019). Demographic and socio- economic covariates considered in the analyses were age, sex, employment status, household income, educational level, and house hold composition (e.g. single occupancy, or having children or not). Vehicle ownership covariates considered were car accessibility, having a valid driving license, and bicycle accessibility. Health covariates considered were self-rated health status and Body Mass Index (BMI), which have been shown to influence motorized travel and transport CO2 emissions (Goodman et al., 2012). The perceived walking times to the nearest bus stop, tram stop or railway station were included as public transport accessibility measures. All of the covariates were self-reported. BMI was derived from self-reported weight and height as weight(kg)/height(m)2 (Dons et al., 2018). 2.3. Outcome variables: carbon dioxide emissions l i For cars, ‘cold start’ excess emissions were added to ‘hot’ emissions based on the vehcile fleet composition, ambient temperatures (see Supplementary Table S13) and trip distances observed in each city: across the seven cities, cold start emissions averaged 126 (SD 42) gCO2 per car trip, with the trip share of a car operating with a ‘cold’ engine averaging 13 (SD 8) percent. Cold start emissions were higher-than-average in colder climates such as Orebro and Zurich, and lower in warmer ones such as Barcelona. Second, carbon emissions from energy supply considered upstream emissions from the extraction, production, generation and distribution of energy supply, with values taken from international databases for fossil fuel emissions (2016; JEC, 2014; Odeh et al., 2013) and emissions from electricity generation and supply (Ecometrica, 2011). Third, vehicle life cycle emissions considered emissions from the manufacture of vehicles, with aggregate carbon values per vehicle type (cars, motorcycles, bikes and public transport vehicles) derived assuming typical lifetime mileages, mass body weights, material composition and material-specific emissions and energy use factors. The main functional relationships and data are provided in the Supplementary Information. The derived emissions rates are shown in Table 1 for each city, disaggregated by emissions category and transport mode. Second, carbon emissions from energy supply considered upstream emissions from the extraction, production, generation and distribution of energy supply, with values taken from international databases for fossil fuel emissions (2016; JEC, 2014; Odeh et al., Total daily emissions were calculated as the sum of emissions for each trip, mode and purpose (e.g. the sum of 4 trips on a given day = trip 1: home to work by car, trip 2: work to shop by bike, trip 3: shop to work by bike; and trip 4: work to home by car). Secondary outcomes of interest were total life cycle CO2 emissions for four aggregated journey purposes: (1) work or education/school trips; (2) business trips; (3) social or recreational trips; and (4) shopping, personal business, escort or ‘other’ trips. 2.2. Exposure: transport mode choice and use Brand et al. Transportation Research Part D 93 (2021) 102764 categorizations based on cycling frequency included a dichotomous variable of ‘cycling’ on the diary day (yes/no) as well as a trichotomous variable characterizing participants as ‘frequent cyclist’ (three or more times a day), ‘occasional cyclist’ (once or twice a day), or ‘non-cyclist’ (none). Table 2 shows sample sizes and mean (SD) values of the primary outcome variable for each group. categorizations based on cycling frequency included a dichotomous variable of ‘cycling’ on the diary day (yes/no) as well as a trichotomous variable characterizing participants as ‘frequent cyclist’ (three or more times a day), ‘occasional cyclist’ (once or twice a day), or ‘non-cyclist’ (none). Table 2 shows sample sizes and mean (SD) values of the primary outcome variable for each group. 2.3. Outcome variables: carbon dioxide emissions 2.5. Statistical analysis and modelling In a first step, bivariate analyses were performed to assess the association between transport-related CO2 emissions, the exposure variables, and the potential covariates. Only covariates with p-value < 0.1 were included in further analysis. In a second step, dif ferences in CO2 emissions between the different transport mode users were identified by using linear mixed-effects models (Baltagi, 2008) with city as a random effect (to take account of correlation among responses from the same city). The analysis used multiple data points for each individual, obtained on different weekdays. This justified using a multi-level modelling approach (Bryk and Rau denbush, 2002); we therefore included ‘individual ID’ and ‘weekday’ (level 1) as random effects, nested within city (level 2). Transportation Research Part D 93 (2021) 102764 C. Brand et al. C. Brand et al. Because CO2 emissions were heavily skewed towards the right (see also Fig. 1), we applied the transformation ‘ln([x/mean(x)] + 0.01)’ (adding 0.01 to avoid turning zeros into missing values) in the comparative analysis. This improved our regression diagnostics, with residuals closer to a normal distribution and their variance less heteroscedastic. Note a log transformation changes the focus from absolute to relative or percentage change; therefore, any regression coefficient β is a mean difference of the log outcome comparing adjacent units of a predictor. This is practically useless, so we exponentiate the parameter eβ and interpret this value as a geometric mean difference (Vittinghoff et al., 2012). iii g Three regression models were fitted. First, ‘model 0′ fitted an unadjusted (exposure only) fixed effects model; second, ‘model 1′ fitted a hierarchical (two levels) mixed effects model (Bryk and Raudenbush, 2002) partially adjusted by socio-demographic covariates of sex, age, education level, employment status, household income, and household composition; and third, model 2 fitted a hierar chical (two levels) mixed effects model fully adjusted by all covariates from model 1 and additionally other covariates of interest (those found to be statistically significant in previous literature described earlier): holding a valid driving license, access to a car or van, bicycle ownership, self-rated health, BMI, walking-time accessibility to the nearest bus stop, and walking-time accessibility to the nearest train station. Age was included as a continuous variable as a proxy for time. As described earlier, city, participant and weekday were included as random effects in the mixed effects models (models 1 and 2). 2.5. Statistical analysis and modelling The same set of models were fitted for each of the four journey purposes. Equation (1) provides the general formula for the fully-adjusted model of the main exposure variable (daily transport mode usage), log-transformed CO2 emissions as the dependent variable, and a set of covariates with fixed effects (βi) and random effects (ρj). For model definitions of the three secondary exposures, replace ‘transport mode usage’ with categorical variables ‘main mode of travel’, ‘cycling frequency’, and ‘cyclist/non-cyclist’. CO2* = β1 × fcycle + β2 × fcar + β3 × fPT + β4 × fwalk + β5 × sex + β6 × age + β7 × educ + β8 × employ + β9 × HHinc + β10 × HHcomp + β11 × license + β12 × caraccess + β13 × bikeaccess + β14 × health + β15 × BMI + β16 × PTbus + β17 × PTrail + ρ1 × weekday + ρ2 × participant + ρ3 × city + ∊ (1) × weekday + ρ2 × participant + ρ3 × city + ∊ (1) where Dependent variable: CO2* = ln([CO2/mean(CO2)] + 0.01); Exposure variables: fmode.x = trips per day by mode x; Covariates: sex, age, educ (highest level of education), employ (employment status), HHinc (household income), HHcomp (household composition), licence (driving licence), caraccess (car access), bikeaccess (bike access), health (self-rated health status), BMI (body mass index), PTbus (walking time to nearest bus stop), PTrail (walking time to nearest train station), weekday, participant, city; i p y vector of fixed effects; ρj = vector of random effects. ∊ = vector of random errors. p p y βI = vector of fixed effects; ρj = vector of random effects. ∊ = vector of random errors. Potential interaction by sex, employment status, income, car access, BMI and city were investigated with Type II Wald chisquare tests in the fully-adjusted models. We observed significant interactions for some transport modes (e.g. use of all modes and car access; public transport use and gender; car use and income); therefore, all models’ sensitivity to different levels of the above factors were tested. We also tested the models’ sensitivity to a number of other factors: age (‘<35 years’), working status (‘working’), car access (‘not having access to a car’), body weight (‘being overweight’), household income (‘high income’) and city (Table 2). 3.2. Levels and sources of life cycle CO2 emissions Life cycle CO2 emissions from all travel activity were 3.18 (SD 7.68) kilograms of CO2 (kgCO2) per person per day, with the majority from car travel at 2.23 (SD 7.25) kgCO2/day – i.e. 70% of the daily total (see Table 2). In contrast, life cycle emissions from cycling (which included a 4.5% share of e-biking across the sample) amounted for only 0.03 (SD 0.05) kgCO2/day. Direct (operational) emissions from all travel activity made up the majority (70%) of total life cycle emissions. While travel to work or place of education produced the largest share of CO2 emissions (37%), there were also considerable contributions from social and recreational trips (34%), business trips (11%) and travel for shopping or personal business (17%). Fig. 1 shows a highly unequal distribution of emis sions. It also shows that the top decile of emitters were responsible for 59% (all purposes), 47% (work or education), 78% (business), 67% (social or recreational) and 58% (shopping, personal business, escort or other) of the respective life cycle CO2 emissions. Those in the top decile were more likely to be male, have higher household incomes, holding a driving license and always have access to a car, be in full-time employment, have higher BMI, have poor bus or train accessibility and live in Orebro, Antwerp or Rome. In contrast, those in the bottom decile of emitters were more likely to be female, economically inactive or a student, living in a household without kids, be on lower household incomes, not to hold a driving license, without access to a car, own a bike, have lower BMI, live nearer to train stations, and live in Barcelona or London. To explain this it is worth highlighting that while Antwerp and Orebro had signifi cantly2 higher cycling trip shares amongst the case study cities, they also had higher car shares (together with Rome) and low walking shares (also together with Rome). On the contrary, Barcelona and London had lower car trip shares (like Vienna and Zurich) and higher walking shares (Supplementary Table S3). 2.5. Statistical analysis and modelling Participants 0.01 0.1 1 10 100 1000 10000 1 2 3 4 5 6 7 8 9 10 CO2 [grams/day] emissions deciles all purposes work or education business social or recreational shopping, PB, escort or other Fig. 1. Distributions of mean life cycle CO2 emissions by travel emissions decile, subdivided by journey type (log-normal plot, error bars are 95% CIs). Fig. 1. Distributions of mean life cycle CO2 emissions by travel emissions decile, subdivided by journey type (log-normal plot, error bars are 95% CIs). Fig. 1. Distributions of mean life cycle CO2 emissions by travel emissions decile, subdivided by journey type (log-normal plot, error bars are 95% CIs). 7 Transportation Research Part D 93 (2021) 102764 C. Brand et al. were also ranked according to their CO2 emissions (all travel and by trip purpose) then split into ten emissions deciles. Chi-square tests were performed on selected covariates to profile the ‘bottom’ and ‘top’ deciles. Possible mediation of the effect of transport mode use on CO2 emissions was assessed for three potential mediators: total daily distance travelled, BMI and self-rated health (VanderWeele, 2016; Wanner et al., 2012). Only observations without missing data were included. R statistical software v3.6.1 was used for all analyses. were also ranked according to their CO2 emissions (all travel and by trip purpose) then split into ten emissions deciles. Chi-square tests were performed on selected covariates to profile the ‘bottom’ and ‘top’ deciles. Possible mediation of the effect of transport mode use on CO2 emissions was assessed for three potential mediators: total daily distance travelled, BMI and self-rated health (VanderWeele, 2016; Wanner et al., 2012). Only observations without missing data were included. R statistical software v3.6.1 was used for all analyses. 3.2. Levels and sources of life cycle CO2 emissions i In our sample, respondents in Orebro and Rome produced significantly higher-than-average CO2 emissions (mean 4.56 kgCO2/day and 3.93 kgCO2/day, respectively) due to the higher car mode shares, while those in Barcelona and Vienna produced lower emissions (mean 2.47 kgCO2/day and 2.65 kgCO2/day, respectively) due to higher share of walking (Barcelona) and a combination of lower car and higher public transport shares (Vienna) (Table 2 and Supplementary Table S3). Those in Antwerp had the highest active travel share, but also higher (than sample average) car and lower public transport shares, resulting in higher than average CO2 emissions overall (mean 3.49 kgCO2/day). These figures are generally in line with regional per capita CO2 emissions estimates. Differences between cities can partially be explained by differences in sample demographics, socio-economics and observed mode shares (Sup plementary Table S1 and Table S3). 3.1. Sample description and summary statistics The final sample included 3836 participants who had completed 9858 one-day travel diaries reporting 34,203 trips (Table 2). The sample was well balanced between male and female, and between the seven cities. Participants were highly educated with 79% of the participants having at least a secondary or higher education degree. Aged between 16 and 91, the majority of participants were employed full-time (66%), with 72% on middle to high household incomes (i.e. >€25 k) and 34% reported to have children living at home. The share of participants without access to a car was 21%. While cycling and public transport were the most frequent transport modes among our participants, people travelled furthest by public transport and car. Transport mode usage was similar between sexes, with a slightly higher prevalence of male cyclists and drivers vs. female walkers and public transport users. Participants reported an average of 3.47 (SD 1.83) trips per day ranging from 3.10 (SD 1.63) trips per day in Rome to 3.75 (SD 2.0) trips per day in Antwerp (Table 2). The observed cycling trip share for our sample was between 17% in Barcelona and 54% in Antwerp (Supplementary Table S1), i.e. somewhat higher than cycling shares reported for the cities (EPOMM, 2020) and a direct result of purposively over sampling cyclists. Reported trip durations and distances varied between subjects and cities, with respondents travelling on average 36.1 (SD 63.5) km a day and for 87.8 (SD 70.4) min a day. Average trip lengths across the cities were 1.1 (SD 1.6) km for walking, 5.0 (SD 5.3) km for cycling, 20.5 (SD 45.9) km for driving and 16.7 (SD 33.6) km for public transport. These are comparable to travel statistics for European cities (JRC, 2013). Further results for each city are given in Supplementary Table S3. 2 Comparing deciles with chi-square tests of independence. Table 3 Table 3 Results from the linear fixed-effects and mixed-effects models for the four exposures (n = 9858). Full models are presented in the Supplementary Information. Table 3 Results from the linear fixed-effects and mixed-effects models for the four exposures (n = 9858). Full models are presented in the Supplementary Information. Model 0: unadjusted (fixed effects) Model 1: partly adjusted (mixed effects) † Model 2: fully adjusted (mixed effects) # Coefficient (95% CI) p-value Coefficient (95% CI) p-value Coefficient (95% CI) p-value (a) Association between log-transformed life cycle CO2 emissions and transport mode usage (trips/day) (full model in Table S4) Bike −0.154 (−0.172 to −0.137) <0.0001 −0.16 (−0.179 to −0.142) <0.0001 −0.151 (−0.17 to −0.132) <0.0001 Car 0.997 (0.976–1.017) <0.0001 0.974 (0.953–0.996) <0.0001 0.962 (0.94–0.983) <0.0001 Public transport 0.741 (0.719 to 0.763) <0.0001 0.737 (0.714–0.76) <0.0001 0.748 (0.724–0.771) <0.0001 Walk −0.287 (−0.305 to −0.269) <0.0001 −0.278 (−0.297 to −0.259) <0.0001 −0.273 (−0.292 to −0.254) <0.0001 (b) Association between log-transformed life cycle CO2 emissions and main transport mode categories (full model in Table S6) Bike 0 – 0 – 0 – Car 3.89 (3.84–3.939) <0.0001 3.881 (3.829–3.932) <0.0001 3.866 (3.813–3.919) <0.0001 Public transport 2.599 (2.554–2.643) <0.0001 2.624 (2.575–2.673) <0.0001 2.635 (2.586–2.684) <0.0001 Walk −1.071 (−1.137 to −1.005) <0.0001 −0.956 (−1.023 to −0.888) <0.0001 −0.931 (−0.999 to −0.862) <0.0001 (c) Association between log-transformed life cycle CO2 emissions and cycling frequency categories (full model in Table S7) None 0 – 0 – 0 – Once or twice a day −1.697 (−1.781 to −1.612) <0.0001 −1.768 (−1.855 to −1.681) <0.0001 −1.747 (−1.835 to −1.659) <0.0001 Three + times a day −2.016 (−2.116 to −1.916) <0.0001 −2.071 (−2.177 to −1.966) <0.0001 −2.038 (−2.145 to −1.932) <0.0001 (d) Association between log-transformed life cycle CO2 emissions and cycling (yes/no) (full model in Table S8) Not cycling 0 – 0 – 0 – Cycling −1.822 (−1.893 to −1.75) <0.0001 −1.875 (−1.952 to −1.797) <0.0001 −1.848 (−1.927 to −1.769) <0.0001 † Model 1 adjusted for sex, age, education level, employment status, household income, household composition; city and person as random effects. # Model 2 adjusted for sex, age, education level, employment status, household income, household composition, driver license, car access, bike access, self-rated health, BMI, bus accessibility, rail accessibility; city, person and day of the week as random effects. † Model 1 adjusted for sex, age, education level, employment status, household income, household composition; city and person as random effects. Table 3 The ‘economically inactive’ (home duties, retired, unemployed, etc.) showed significantly higher emissions for social, recreational, shopping and personal business purposes, with lower emissions from work or educational trips, as expected. Those with children living i A series of sensitivity analyses largely confirmed our results (Fig. 3a): excluding participants older than 35 or on lower incomes did not change our conclusions; and differences between those ‘working’ and ‘not working’ and those being ‘overweight’ (above 25 kg/m2) and ‘healthy weight’ were small. For people who did not have access to a car the effects were larger for motorized travel and smaller for active travel, suggesting that active travel for non-car owning households may substitute for public transport and other active travel. The associations between log-transformed life cycle CO2 emissions for the four trip purposes (secondary outcomes) and transport mode usage were also largely significant (Table 4a and Supplementary Fig. S3a). Cycling frequency had larger effects on emissions from commuting to work or place of education than on emissions from all purposes (primary outcome models). Motorized transport mode use showed larger effects on life cycle CO2 emissions from social, shopping and recreational travel than for work/business travel. The ‘economically inactive’ (home duties, retired, unemployed, etc.) showed significantly higher emissions for social, recreational, shopping and personal business purposes, with lower emissions from work or educational trips, as expected. Those with children living at home had significantly lower business, social and recreational emissions, while emissions from shopping, personal business and escort trips were higher. Poor bus accessibility and better car access meant higher emissions from work or educational trips. Table 3 # Model 2 adjusted for sex, age, education level, employment status, household income, household composition, driver license, car access, bike access, self-rated health, BMI, bus accessibility, rail accessibility; city, person and day of the week as random effects. emissions decreased by a factor of 0.15 (95%CI 0.13–0.17) for each additional cycling trip. They also decreased by a factor of 0.96 (95%CI 0.94–0.98) for one less car trip. When converting the log-transformed results into physical units, we find that for each avoided car trip daily life cycle CO2 emissions from transport decreased by 62% (95%CI 61–63%) while for each additional bike trip life cycle CO2 emission decreased by 14% (95%CI 12–16%). While this may be confusing at first, it is the result of converting log-transformed outcomes to physical units, as is common practice in such regression models (Vittinghoff et al., 2012). To derive percentage changes of the untransformed outcome, we exponentiated the regression coefficient, subtracted 1 and multiplied by 100. For example, the coefficient of −0.962 for one less car trip (see also Table 3), we first exponentiate (=0.382), then subtract 1 (=−0.618) and multiply by 100 (=61.8) to obtain a 62% decrease. Those who made one less car trip and one more bike trip a day (a proxy for mode shift from car to bike) decreased life cycle CO2 emissions from transport by 67% (95%CI 66–68%). emissions decreased by a factor of 0.15 (95%CI 0.13–0.17) for each additional cycling trip. They also decreased by a factor of 0.96 (95%CI 0.94–0.98) for one less car trip. When converting the log-transformed results into physical units, we find that for each avoided car trip daily life cycle CO2 emissions from transport decreased by 62% (95%CI 61–63%) while for each additional bike trip life cycle CO2 emission decreased by 14% (95%CI 12–16%). While this may be confusing at first, it is the result of converting log-transformed outcomes to physical units, as is common practice in such regression models (Vittinghoff et al., 2012). To derive percentage changes of the untransformed outcome, we exponentiated the regression coefficient, subtracted 1 and multiplied by 100. For example, the coefficient of −0.962 for one less car trip (see also Table 3), we first exponentiate (=0.382), then subtract 1 (=−0.618) and multiply by 100 (=61.8) to obtain a 62% decrease. Table 3 Those who made one less car trip and one more bike trip a day (a proxy for mode shift from car to bike) decreased life cycle CO2 emissions from transport by 67% (95%CI 66–68%). Adjusting for demographic, socio-economic, other individual variables as well as including random effects parameters (city, in dividual, weekday) only slightly changed the estimates in the partly and the fully adjusted models (model 1 and model 2) compared to the unadjusted model (model 0). The addition of car availability and bus station accessibility in the fully adjusted model (model 2) slightly lowered the estimate for car but increased the estimate for public transport use compared to the unadjusted (0) and partly adjusted models (1). The effects of transport mode use on transformed carbon emissions was partially mediated via total distance travelled (see Fig. 2): the indirect effects of total distance travelled were +0.13 for car use (13% mediated), −0.02 for cycling (14% mediated), +0.10 for public transport use (13% mediated), and −0.05 for walking (18% mediated). Neither BMI nor health status mediated this association. i the indirect effects of total distance travelled were +0.13 for car use (13% mediated), −0.02 for cycling (14% mediated), +0.10 for public transport use (13% mediated), and −0.05 for walking (18% mediated). Neither BMI nor health status mediated this association. A series of sensitivity analyses largely confirmed our results (Fig. 3a): excluding participants older than 35 or on lower incomes did not change our conclusions; and differences between those ‘working’ and ‘not working’ and those being ‘overweight’ (above 25 kg/m2) and ‘healthy weight’ were small. For people who did not have access to a car the effects were larger for motorized travel and smaller for active travel, suggesting that active travel for non-car owning households may substitute for public transport and other active travel. The associations between log-transformed life cycle CO2 emissions for the four trip purposes (secondary outcomes) and transport mode usage were also largely significant (Table 4a and Supplementary Fig. S3a). Cycling frequency had larger effects on emissions from commuting to work or place of education than on emissions from all purposes (primary outcome models). Motorized transport mode use showed larger effects on life cycle CO2 emissions from social, shopping and recreational travel than for work/business travel. 3.3. Transport mode usage We found statistically significant associations between life cycle CO2 emissions and transport mode usage across all modes of travel (Table 3a): those who cycled or walked more had lower daily mobility-related CO2 emissions, while those who drove more or used public transport more had higher daily total CO2 emissions. In the fully-adjusted model (model 2), log-transformed life cycle carbon Transportation Research Part D 93 (2021) 102764 C. Brand et al. Table 3 Results from the linear fixed-effects and mixed-effects models for the four exposures (n = 9858). Full models are presented in the Supplementary Information. 3.4. Main mode of travel (by daily distance travelled) We also found statistically significant associations between life cycle CO2 emissions and the main modes of travel according to daily distance travelled (Table 3b): when compared to using a bike as the main mode, using the car or public transport increased CO2 while walking decreased CO2. In the fully adjusted model (model 2) CO2 emissions were 98 (95%CI 98–98) percent higher for using a car or van as the main mode than for using a bike. An average person using a car or van as the main mode had 7.1 kgCO2/day higher life cycle 9 9 Transportation Research Part D 93 (2021) 102764 C. Brand et al. p ( ) A Car: 11.11* Bike: -1.90* Public transport: 8.58* Walk: -4.17* Total distance Transport mode frequency Lifecycle CO2 Transport mode frequency Lifecycle CO2 Direct eﬀects: Car: 0.85* Bike: -0.13* Public transport: 0.67* Walk: -0.22* B 0.0113* Total eﬀects: Car: 0.98* Bike: -0.15* Public transport: 0.77* Walk: -0.27*** Indirect eﬀects (AB): Car: 0.13 Bike: -0.02* Public transport: 0.10* Walk: -0.05* Fig. 2. Mediation of total daily travel distance in the association between transport mode frequency and log-transformed lifecycle CO2 emissions. The mediation models were adjusted for sex, age, education level, employment status, household income, household composition, driver license, car access, bike access, self-rated health, BMI, bus accessibility, rail accessibility, day of the week; city was included as a random effect. The numbers presented in the figure are the regression coefficients (‘*’ denotes level of statistical significance at p < 0.001). Transport mode frequency Lifecycle CO2 A Total distance Transport mode frequency Lifecycle CO2 Fig. 2. Mediation of total daily travel distance in the association between transport mode frequency and log-transformed lifecycle CO2 emissions. The mediation models were adjusted for sex, age, education level, employment status, household income, household composition, driver license, car access, bike access, self-rated health, BMI, bus accessibility, rail accessibility, day of the week; city was included as a random effect. The numbers presented in the figure are the regression coefficients (‘*’ denotes level of statistical significance at p < 0.001). Fig. 3. Sensitivity analyses. Exposure variables are: transport mode usage in panel (a), main mode of travel (by distance) in panel (b), cycling frequency in panel (c), and cycling (no/yes) in panel (d). The dots are the beta coefficients and indicate differences in log-transformed CO2 emissions (error bars are 95% CIs). Fig. 3. Sensitivity analyses. 3.4. Main mode of travel (by daily distance travelled) Exposure variables are: transport mode usage in panel (a), main mode of travel (by distance) in panel (b), cycling frequency in panel (c), and cycling (no/yes) in panel (d). The dots are the beta coefficients and indicate differences in log-transformed CO2 emissions (error bars are 95% CIs). Fig. 3. Sensitivity analyses. Exposure variables are: transport mode usage in panel (a), main mode of travel (by distance) in panel (b), cycling frequency in panel (c), and cycling (no/yes) in panel (d). The dots are the beta coefficients and indicate differences in log-transformed CO2 emissions (error bars are 95% CIs). 10 e. Social or recreational # Shopping, per p-value Coefficient (95% CI) p-value Coefficient (9 ay) (full model in Table S9) 0.174 0.062 (0.031–0.094) <0.001 0.158 (0.126– <0.001 0.725 (0.689–0.761) <0.001 0.826 (0.79–0 <0.001 0.475 (0.436–0.514) <0.001 0.393 (0.354– <0.001 −0.092 (−0.123 to −0.061) <0.001 −0.025 (−0.0 es (full model in Table S10) – 0 – <0.001 2.281 (2.164–2.397) <0.001 1.987 (1.867– <0.001 1.002 (0.895–1.108) <0.001 0.677 (0.566– 0.033 −0.784 (−0.934 to −0.634) <0.001 −0.462 (−0.6 ull model in Table S11) – 0 – < 0.001 −0.865 (−0.977 to −0.754) < 0.001 −0.77 (−0.88 < 0.001 −0.756 (−0.89 to −0.622) < 0.001 −0.344 (−0.4 n Table S12) – 0 – <0.001 −0.826 (−0.925 to −0.727) <0.001 −0.617 (−0.7 nt status, household income, household composition, driver license, car ac . g, per nt (9 126– 79–0 354– −0.0 867– 566– −0.6 −0.88 −0.4 −0.7 p-value ) (full mo 0.174 <0.001 <0.001 <0.001 (full mod – <0.001 <0.001 0.033 l model i – < 0.001 < 0.001 Table S12 – <0.001 status, h 3.5. Cycling frequency and cycling vs not cycling Associations between life cycle CO2 emissions and cycling frequency were all highly significant. Table 3c shows that in the fully adjusted model (model 2) life cycle CO2 emissions were 83 (95%CI 81–84) percent lower for ‘occasional cyclists’ (i.e. those cycling ‘once or twice a day’) than for those who did not cycle; and they were even lower for ‘frequent cyclists’ (those cycling ‘three or more times a day’) with 87 (95%CI 86–88) percent lower daily life cycle CO2. The sensitivity analysis (Fig. 3c) generally confirmed our results, with slightly higher effects for high earners and lower effects if you were younger or without access to a car. Regular cycling was also associated with reduced life cycle CO2 emissions for all the four trip purposes, with the strongest effect observed for commuting and social trips (Supplementary Table S11): cycling three or more times a day decreased life cycle CO2 emissions for work or education by 78 (95%CI 75–80) percent, for social or recreational trips by 53 (95%CI 46–59) percent, for shopping and personal business by 29 (95%CI 19–38) percent, and for business travel by 20 (95%CI 10–28) percent. As expected, the binary cyclist/non-cyclist analysis showed similar effect sizes and correlations to the analysis of cycling frequency for both primary and secondary outcome measures. ‘Cyclists’ had 84 (95%CI 83–85) percent lower life cycle CO2 emissions than ‘non- cyclists’ (Table 2d and Supplementary Table S12); this translated into 0.97 (95%CI 0.54–1.74) kgCO2/day lower life cycle CO2 emissions for an average person who cycled. The sensitivity analysis showed that the effects were lower for the younger respondents and those without access to a car, and higher for those on higher incomes (Fig. 3d). As expected, the binary cyclist/non-cyclist analysis showed similar effect sizes and correlations to the analysis of cycling frequency for both primary and secondary outcome measures. ‘Cyclists’ had 84 (95%CI 83–85) percent lower life cycle CO2 emissions than ‘non- cyclists’ (Table 2d and Supplementary Table S12); this translated into 0.97 (95%CI 0.54–1.74) kgCO2/day lower life cycle CO2 emissions for an average person who cycled. The sensitivity analysis showed that the effects were lower for the younger respondents and those without access to a car, and higher for those on higher incomes (Fig. 3d). Fig. 4. Effect sizes from the fully adjusted model and sensitivity analyses (city stratification). Transportation Research Part D 93 (2021) 102764 C. Brand et al. CO2 emissions than someone using a bike as their main mode of transport. A comparison with the results of the non-transformed model suggested that using a car or van increased emissions by 8.9 kgCO2/day when compared to cycling as the main mode (Supplementary Table S5 and Fig. S2) – suggesting the linear model slightly overestimated differences in emissions by main mode when compared to the (statistically superior) log-linear model. Those using public transport as the main mode had 71 (95%CI 71–71) percent lower emissions than those mainly using a car, van or motorcycle; for an average person this difference equated to 5.1 kgCO2/day. i Again, the sensitivity analysis (Fig. 3b) largely confirmed our results. Total distance travelled partially (12%) mediated the effects of main mode (by daily distance) on transformed life cycle CO2 emissions. The associations for log-transformed CO2 emissions by journey purpose were also all significant (Supplementary Table S10 and Fig. S3), with the strongest effects for mainly using public transport for work or education and car for social and shopping trips. Women, those with a degree or no access to a car had significantly lower work or education emissions. As expected, the economically inactive had significantly higher social, recreational and shopping/ personal business emissions, yet lower work/education emissions. 4.2. Comparison with previous studies Mean total CO2 emissions of 3.18 kgCO2/day were much higher than the median (0.81 kgCO2/day) and near the upper end of the derived interquartile range (0.07–3.27 kgCO2 per day), confirming a positively skewed distribution of emissions. In other words, a relatively small share of individuals was responsible for the vast majority of carbon emissions, a finding that is very much in line with the evidence on unequal carbon emissions distributions (Brand and Boardman, 2008; Büchs and Schnepf, 2013; Ko et al., 2011; Preston et al., 2013; Susilo and Stead, 2009). Our findings that the likelihood of being in top or bottom emissions decile depended on de mographic, socio-economic, car availability, health, public transport accessibility and contextual factors further support the growing evidence on travel emissions inequalities (Banister, 2018; Bel and Rosell, 2017; Brand, 2008; Ko et al., 2011). The analysis of transport mode use as the main exposure showed that each additional cycling trip reduced life cycle CO2 emissions from all travel activity by about 14% when compared to baseline emissions. On average, those who did one less trip by car and one more by bike or public transport decreased emissions by 67% and 19% respectively. To further aid interpretation of the factorial results we need to apply the percentage changes to baseline (or mean) levels of our measured outcomes. For example, an average person ‘shifting modes’ from car (from 3 to 2 trips a day) to bike (from 0 to 1 trip a day) decreased emissions by 3.2 (95%CI 2.0–5.2) kgCO2/ day. Similarly, a person ‘shifting modes’ from car (from 3 to 2 trips a day) to public transport (from 0 to 1 trip a day) decreased emissions by 0.9 (95%CI 0.6–1.5) kgCO2/day. If 10% of the population were changing travel behavior this way, emissions would be expected to decrease by about 10% (car → bike) and 3% (car → public transport). The size and direction of emissions changes are in line with some of the few empirical (Brand et al., 2013; Goodman et al., 2012) and scenario/modelling (Goodman et al., 2019; Rabl and de Nazelle, 2012; Tainio et al., 2017; Woodcock et al., 2018) studies in this area. i The differences in emissions between people using different modes for the majority (defined by max. distance travelled) of their daily travel were also highly significant, although the effects were partially (12%) mediated by total daily distance travelled. 4.1. Summary of results This paper started on the premise that we still do not know very much about how much carbon from passenger transport is saved – overall – by active travel. In this multi-city study with thousands of participants providing nearly 10,000 valid person-days of travel activity, we found highly significant associations between transport mode choice and total life cycle CO2 emissions. We showed that cyclists had significantly lower total CO2 emissions than non-cyclists. More cycling or walking decreased mobility-related life cycle CO2 emissions – suggesting that active travel indeed substitutes for motorized travel (i.e. this was not just additional travel over and above motorized travel). This means that even if not all car trips could be substituted by bicycle trips the potential for decreasing emissions is very high. A number of sensitivity analyses that tested for potential interactions and biases in the study sample confirmed our main results and provided new insights into differences of emission levels and exposures by city, journey purpose and key covariates. The differences in mean emissions and effect sizes in the seven cities may be explained by contextual factors such as differences in modal shares, mode trip lengths, and the provision (or not) of good public transport services and active travel infra structure – it may also be due to differences in sampling (Raser et al., 2018). The analysis of emissions for each trip purpose highlighted the relative importance of emissions from non-work/business trips, particularly trips for social and shopping purposes. 3.6. Sensitivity: city effects The random intercepts of city explained 2.2% (a: transport mode usage), 5.4% (b: main mode of transport), 2.6% (c: cycling frequency) and 2.5% (d: cycling yes/no) of the residual variance in the fully adjusted models. Mean CO2 emissions were significantly lower in Barcelona and Vienna and higher in Orebro and Rome (Table 2). Further sensitivity analyses of the fully adjusted models stratified by city showed that the effect estimates for cycling were generally the lowest in Barcelona and highest in Orebro and Rome (see Fig. 4). By comparison, CO2 effects for car travel were highest in Barcelona (and Vienna to some extent) and lowest in London and Rome. 3.5. Cycling frequency and cycling vs not cycling Exposure variables: transport mode usage in panel a; main mode of transport (by distance) in panel b; cycling frequency in panel c; and cycling/not cycling in panel d. The dots indicate differences in CO2 emissions and the error bars indicate 95% CIs. Fig. 4. Effect sizes from the fully adjusted model and sensitivity analyses (city stratification). Exposure variables: transport mode usage in panel a; main mode of transport (by distance) in panel b; cycling frequency in panel c; and cycling/not cycling in panel d. The dots indicate differences in CO2 emissions and the error bars indicate 95% CIs. 12 Transportation Research Part D 93 (2021) 102764 Transportation Research Part D 93 (2021) 102764 C. Brand et al. 3.6. Sensitivity: city effects 4.2. Comparison with previous studies Our finding that, on average, using a bike as the main mode decreased life cycle CO2 emissions by about 7.1 kgCO2/day when compared to using a car or van suggests that making more sustainable choices on to how we get from A to B has significant carbon benefits. Similarly, our finding that doing at least one trip a day by bike significantly decreased mobility-related life cycle CO2 emissions provides further evidence of mode substitution away from motorized travel. Much of the research in this area has focused on travel activity and associated carbon emissions from work and business travel (Bearman and Singleton, 2014; Clark et al., 2016). In our study, commuting, education and business travel emissions represented ‘only’ about half (49%) of total emissions, ranging from 39% in Antwerp to 59% in London and Rome. The findings that life cycle CO2 emissions from social, shopping, personal business and recreational journeys were more strongly associated to car and, to some extent, public transport use suggest for research and policy to go beyond commuting and business travel and consider the whole range of journey purposes when investigating mode shift away from motorized to active travel (Brand et al., 2013). This seems to be partic ularly important with the growing shares of the elderly in the population. Shopping and personal business trips were found to be significantly shorter, therefore increasing the potential for mode shift to active travel. i The mediation analysis by distance travelled indicated that lower carbon emissions for cyclists was unlikely to be entirely caused by increased bike usage. The remaining emissions difference might be explained by distance-related factors that influence mode choice 13 C. Brand et al. Transportation Research Part D 93 (2021) 102764 such as urban form and location of housing, services and jobs (Banister et al., 1997; Beenackers et al., 2012; Curtis, 1996; Welch, 2013). i such as urban form and location of housing, services and jobs (Banister et al., 1997; Beenackers et al 2013). i such as urban form and location of housing, services and jobs (Banister et al., 1997; Beenackers et al., 2012; Curtis, 1996; Welch, 2013). While focusing on cycling above we also found that using public transport was more beneficial than private motorized transport across all exposure measures, thus confirming findings from the large body of literature that already exists in this area (see e.g. 4.3. Potential limitations of the study, and how we addressed them In interpreting these findings we need to bear in mind the study’s limitations. First, the recruitment and sampling strategy means that our study sample of adults living in urban areas cannot be assumed to be representative of the general population. Orebro was the lone city that made a concerted effort for random sampling, whereas in other cities an opportunistic recruitment strategy was followed (Dons et al., 2015). However, by oversampling some of the less frequent transport modes, we had a sufficiently large sample of cyclists in all cities to find statistically significant associations. Furthermore, we have made a concerted attempt to account for any potential bias in the study sample through application of multi-level modelling, adjusting for the key socio-demographic predictors/control variables, and conducting extensive sensitivity analyses of potential biases and interactions. We have confidence, therefore, that the findings are generalisable in the context of the adult population in European cities. i Second, recall bias and participant burden of a substantive survey instrument may have impacted the travel diary reporting, which may have reduced the number of reported trips. However, the observed trip frequencies (e.g. 3.47 trips per person per day on average) and mode shares (e.g. significantly higher cycling shares in Antwerp, lower cycling shares in Barcelona, higher public transport shares in London, Vienna and Zurich) were in line with figures reported for the cities (Raser et al., 2018). While trip distances were derived from Google API data, trip durations were self-reported. Trip durations from self-completion travel diaries are known to be over- reported (Kelly et al., 2013), so mean speeds may have been lower than actual speeds leading to increased emissions rates in urban areas. However, further investigation of mean speeds by mode of transport showed that the derived mean speeds of 4.8 kph for walking, 15.6 kph for cycling/e-biking, 39.9 kph for driving a car or van, and 17.9 kph for urban public transport were in line with figures reported elsewhere (Raser et al., 2018). Note these are daily averages not just peak-time speeds (as often reported). i Third, outcome and exposure variables were reported at different time points and days of the week – this was taken into account in the mixed effect models by including ‘day of the week’ and person ID as random (intercept) variables. 4.3. Potential limitations of the study, and how we addressed them Other periodic effects cannot be excluded and we tried to cover for that as much as possible by including relevant time-varying covariates (such as participant age) and factors influencing outcomes such as ambient temperature (for ‘cold start’ emissions). l Fourth, our analysis is cross-sectional, meaning that the direction of causality (if any) behind many of the observed associations is unclear. A longitudinal analysis of change in emissions by change in exposures has been completed and published in Brand et al. (2021). l Fifth, while we accounted for several influencing factors that were often not available in previous studies, such as trip data by mode and purpose, accessibility and health status, our regression models did not account for more than 78% of the variation in the popu lation (see Supplementary results). This suggests that travel choices and associated CO2 emissions are also influenced by other factors such as other built environment factors or lifestyle and socio-cultural factors (Brand et al., 2019; Panter et al., 2013; Weber and Perrels, 2000). We initially explored and added more ‘objective’, GIS based data at both home and work locations to the analysis, including street density, building density, richness of facilities, home-work distance, and public transport availability (timetables, frequency) (Gascon et al., 2019). However, none of these factors improved the models significantly, and the main findings were unchanged. ii Sixth, we excluded carbon emissions from dietary intake as the evidence is not strong on whether day-to-day active travel (as opposed to performance/sport activity) significantly increases overall dietary intake when compared to motorized travel (Tainio et al., 2017). Finally, the interdisciplinary breadth of the PASTA study meant that we measured daily travel behavior, individual and spatial- environmental characteristics using briefer survey tools than might have been feasible in a single-discipline study. This may have introduced some measurement error that could have attenuated our effect sizes. However, the multi-city approach in different countries with different travel patterns, built environments, public transport accessibility levels and active mobility use adds value to the analysis, which clearly showed additional insights compared to smaller, single-location studies. 4.2. Comparison with previous studies Banister, 2008; Graham-Rowe et al., 2011; Nieuwenhuijsen, 2020; Woodcock et al., 2009). ii Last but not least, it is important to highlight that increased active and reduced motorized travel have significant benefits beyond climate change mitigation, particularly with regards to improving population health through increased physical activity and reduced local air pollution (Nieuwenhuijsen, 2020; Shaw et al., 2014). The relative importance of these impacts depends on the context, but the evidence is convincing (G¨ossling et al., 2019; G¨otschi et al., 2020; Rodrigues et al., 2020). Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The authors alone are responsible for the views expressed in this article and they do not necessarily represent the views, decisions or policies of the institutions with which they are affiliated. Acknowledgements This work was supported by the European project Physical Activity through Sustainable Transportation Approaches (PASTA). PASTA (http://www.pastaproject.eu/) was a four-year project funded by the European Union’s Seventh Framework Program (EU FP7) under European Commission - Grant Agreement No. 602624. CB is also supported by UK Research and Innovation (UKRI) under the Centre for Research on Energy Demand Solutions (CREDS, Grant agreement number EP/R035288/1). ED is also supported by a postdoctoral scholarship from FWO – Research Foundation Flanders. ML held a joint PASTA/VITO PhD scholarship. SS is supported by the Martin Filko Scholarship from the Ministry of Education in Slovakia. CRediT authorship contribution statement Christian Brand: Conceptualization, Methodology, Data curation, Validation, Formal analysis, Writing - original draft, Writing - review & editing, Funding acquisition, Investigation, Visualization, Supervision, Project administration. Evi Dons: Conceptualization, Data curation, Formal analysis, Writing - review & editing. Esther Anaya-Boig: Data curation, Writing - review & editing. Ione Avila- Palencia: Data curation, Writing - review & editing. Anna Clark: Data curation, Writing - review & editing. Audrey Nazelle: Methodology, Writing - review & editing, Funding acquisition, Investigation. Mireia Gascon: Data curation, Writing - review & editing. Mailin Gaupp-Berghausen: Data curation, Writing - review & editing. Regine Gerike: Methodology, Writing - review & editing, Funding acquisition, Investigation. Thomas G¨otschi: Conceptualization, Data curation, Writing - review & editing, Funding acquisition, Investigation. Francesco Iacorossi: Data curation, Writing - review & editing. Sonja Kahlmeier: Data curation, Writing - review & editing, Funding acquisition, Investigation. Michelle Laeremans: Data curation, Writing - review & editing. Mark J Nieuwenhuijsen: Conceptualization, Methodology, Writing - review & editing, Funding acquisition, Investigation. Juan Pablo Orjuela: Data curation, Writing - review & editing. Francesca Racioppi: Methodology, Writing - review & editing, Funding acqui sition, Investigation. Elisabeth Raser: Conceptualization, Methodology, Writing - review & editing, Funding acquisition, Investiga tion. David Rojas-Rueda: Methodology, Data curation, Writing - review & editing. Arnout Standaert: Data curation, Writing - review & editing. Erik Stigell: Writing - review & editing. Simona Sulikova: Methodology, Data curation, Writing - review & editing. Sandra Wegener: Methodology, Writing - review & editing. Luc Int Panis: Methodology, Data curation, Writing - review & editing, Funding acquisition, Investigation. 5. Conclusions This paper started on the premise that we still do not know very much about how much carbon from passenger transport is saved – overall – by travelling actively. It investigated to what extent active travel is associated with lower mobility-related life cycle CO2 emissions by using primary data collected in a large European multicentre study to derive total and purpose-specific life cycle CO2 emissions from travel activity at the individual and population levels. The analysis of a sample of thousands of participants and nearly 10,000 person-days of daily travel across the seven sites provided quantitative estimates of the life cycle carbon benefits of active travel using a variety of metrics that could be used in other European cities and beyond. Active travel has attributes of social distancing that are likely to be desirable for some time (Kissler et al., 2020). It could help to cut Active travel has attributes of social distancing that are likely to be desirable for some time (Kissler et 14 Transportation Research Part D 93 (2021) 102764 C. Brand et al. back transportation energy use, CO2 emissions and air pollution while improving population health (Nieuwenhuijsen, 2020; Shaw et al., 2014) as confinement is eased. Therefore, locking in, investing in and promoting active travel should be a cornerstone of sustainability strategies, policies and planning (Andor et al., 2020; Creutzig et al., 2016; Creutzig et al., 2020) to meet our very challenging sustainable development goals that are unlikely to be met without significant mode shift to sustainable transport (Creutzig et al., 2018). Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.trd.2021.102764. Alvanides, S., 2014. Active transport: why and where do people (not) walk or cycle? J. Transport Health 1 (4), 211–213. Anable, J., Brand, C., 2019. Energy, pollution and climate change. In: Docherty, I., Shaw, J. (Eds.), Transport Matters. Policy Press, Bristol, p. 452. Andor, M.A., Gerster, A., Gillingham, K.T., Horvath, M., 2020. Running a car costs much more than people think — stalling the uptake of green travel. Nature 580, 453–455. Banister, D., Watson, S., Wood, C., 1997. Sustainable cities, transport, energy and urban form. Environ. Plan. B: Plan. Des. 24 (1), 125–143. Bearman, N., Singleton, A.D., 2014. Modelling the potential impact on CO2 emissions of an increased uptake of active travel for the home to school commute using individual level data. J. Transport Health 1 (4), 295–304. Adams, J., 2010. Prevalence and socio-demographic correlates of “active transport” in the UK: analysis of the UK time use survey 2005. Prev. Med. 50 (4), 199–203. Adra, N., Michaux, J.-L., Andre, M., 2004. Analysis of the load factor and the empty running rate for road transport. Artemis - assessment and reliability of transport emission models and inventory systems, ARTEMIS - Assessment and reliability of transport emission models and inventory systems, Report INRETS-LTE 0419 INRETS, Bron, FR. Banister, D., 2018. Inequality in Transport. Alexandrine Press, Marcham, UK. Baltagi, B.H., 2008. Econometric Analysis of Panel Data, fourth ed. Wiley, New York. Banister, D., 2008. The sustainable mobility paradigm. Transp. Policy 15 (2), 73–80. CEA/ANFAC, 2014. European Motor Vehicle Parc 2014: Vehicles in Use (2009–2014). ANFAC/ACEA, Madrid. d l d d h l f h l f h References Associations of individual, household and environmental characteristics with carbon dioxide emissions from motorised passenger travel. Appl. Energy 104, 158–169. Brand, C., G¨otschi, T., Dons, E., Gerike, R., Anaya-Boig, E., Avila-Palencia, I., de Nazelle, A., Gascon, M., Gaupp-Berghausen, M., Iacorossi, F., Kahlmeier, S., Int Panis, L., Racioppi, F., Rojas-Rueda, D., Standaert, A., Stigell, E., Sulikova, S., Wegener, S., Nieuwenhuijsen, M.J., 2021. The climate change mitigation impacts of active travel: evidence from a longitudinal panel study in seven European cities. Global Environ. Change. Brand C Preston J M 2010 ’60-20 emission’ The unequal distribution of greenhouse gas emissions from personal non-business travel in the UK Transp Policy 17 rand, C., G¨otschi, T., Dons, E., Gerike, R., Anaya-Boig, E., Avila-Palencia, I., de Nazelle, A., Gascon, M., Gaupp-Berghausen, M., Ia Panis, L., Racioppi, F., Rojas-Rueda, D., Standaert, A., Stigell, E., Sulikova, S., Wegener, S., Nieuwenhuijsen, M.J., 2021. The clima active travel: evidence from a longitudinal panel study in seven European cities. Global Environ. Change. Brand, C., G¨otschi, T., Dons, E., Gerike, R., Anaya-Boig, E., Avila-Palencia, I., de Nazelle, A., Gascon, M., Gaupp-Berghausen, M., Iacorossi, F., Kahlmeier, S., Int Panis, L., Racioppi, F., Rojas-Rueda, D., Standaert, A., Stigell, E., Sulikova, S., Wegener, S., Nieuwenhuijsen, M.J., 2021. The climate change mitigation impacts of active travel: evidence from a longitudinal panel study in seven European cities. Global Environ. Change. Brand, C., Preston, J.M., 2010. ’60-20 emission’–The unequal distribution of greenhouse gas emissions from personal, non-business travel in the UK. Transp. Policy 17 (1), 9–19. rand, C., Preston, J.M., 2010. ’60-20 emission’–The unequal distribution of greenhouse gas emissions from personal, non-business tra (1), 9–19. (1), 9 19. Branion-Calles, M., Winters, M., Nelson, T., de Nazelle, A., Panis, L.I., Avila-Palencia, I., Anaya-Boig, E., Rojas-Rueda, D., Dons, E., G¨otschi, T., 2019. Impacts of study design on sample size, participation bias, and outcome measurement: a case study from bicycling research. J. Transport Health 15, 100651. Winters, M., Nelson, T., de Nazelle, A., Panis, L.I., Avila-Palencia, I., Anaya-Boig, E., Rojas-Rueda, D., Dons, E., G¨otschi, T., 2019. Im ple size, participation bias, and outcome measurement: a case study from bicycling research. J. Transport Health 15, 100651. I R l J W ll R 2009 E l ti f l t t l b h i h i f th ti t T P li 16 Br¨og, W., Erl, E., Ker, I., Ryle, J., Wall, R., 2009. References Creutzig, F., Agoston, P., Minx, J.C., Canadell, J.G., Andrew, R.M., Qu´er´e, C.L., Peters, G.P., Sharifi, A., Yamagata, Y., Dh structure climate solutions. Nat. Clim. Change 6 (12), 1054–1056. Creutzig, F., Javaid, A., Koch, N., Knopf, B., Mattioli, G., Edenhofer, O., 2020. Adjust urban and rural road pricing for fair mobility. Nat. Clim. Change. Creutzig, F., Roy, J., Lamb, W.F., Azevedo, I.M.L., Bruine de Bruin, W., Dalkmann, H., Edelenbosch, O.Y., Geels, F.W., Grubler, A., Hepburn, C., Hertwich, E.G., Khosla, R., Mattauch, L., Minx, J.C., Ramakrishnan, A., Rao, N.D., Steinberger, J.K., Tavoni, M., Ürge-Vorsatz, D., Weber, E.U., 2018. Towards demand-side Creutzig, F., Roy, J., Lamb, W.F., Azevedo, I.M.L., Bruine de Bruin, W., Dalkmann, H., Edelenbosch, O.Y., Geels, F.W., Grubler, A., Hepburn, C., Hertwich, E.G., Khosla, R., Mattauch, L., Minx, J.C., Ramakrishnan, A., Rao, N.D., Steinberger, J.K., Tavoni, M., Ürge-Vorsatz, D., Weber, E.U., 2018. Towards demand-side solutions for mitigating climate change. Nat. Clim. Change 8 (4), 268–271. Cuenot F Fulton L Staub J 2012 The prospect for modal shifts in passenger transport worldwide and impacts on energy use and CO2 Energy Policy 41 98–106 solutions for mitigating climate change. Nat. Clim. Change 8 (4), 268–271. Cuenot, F., Fulton, L., Staub, J., 2012. The prospect for modal shifts in passenger transport worldwide and impacts on energy use and CO2. Energy Policy 41, 98–106. Curtis C 1996 Can strategic planning contribute to a reduction in car based travel? Transp Policy 3 (1 2) 55 65 solutions for mitigating climate change. Nat. Clim. Change 8 (4), 268 271. Cuenot, F., Fulton, L., Staub, J., 2012. The prospect for modal shifts in passenger transport worldwide and impacts on energy use and CO2. Energy Policy 41, 98–106. Curtis, C., 1996. Can strategic planning contribute to a reduction in car-based travel? Transp. Policy 3 (1–2), 55–65. de Nazelle, A., Morton, B.J., Jerrett, M., Crawford-Brown, D., 2010. Short trips: an opportunity for reducing mobile-source emissions? Transp. Res.: Part D: Transport g g g g L., Staub, J., 2012. The prospect for modal shifts in passenger transport worldwide and impacts on energy use and CO2. Energy Polic an strategic planning contribute to a reduction in car-based travel? Transp. Policy 3 (1–2), 55–65. Cuenot, F., Fulton, L., Staub, J., 2012. The prospect for modal shifts in passenger transport worldwide and impacts on energy use and CO2. Energy Policy 41, 98–106. Curtis, C., 1996. References Applying activity-travel data for the assessment of vehicle exhaust emissions: collection tool. Transp. Res.: Part D: Transport Environ. 15 (2), 117–122. Beenackers, M.A., Foster, S., Kamphuis, C.B., Titze, S., Divitini, M., Knuiman, M., van Lenthe, F.J., Giles-Corti, B., 2012. Taking up cycling after residential relocation: built environment factors. Am. J. Prev. Med. 42 (6), 610–615. Beenackers, M.A., Foster, S., Kamphuis, C.B., Titze, S., Divitini, M., Knuiman, M., van Lenthe, F.J., Giles-Corti, B., 2012. Taking up cycling after residential relocation: built environment factors. Am. J. Prev. Med. 42 (6), 610–615. ( ), BEIS, 2019. Greenhouse gas reporting: conversion factors 2019, accessed at https://www.gov.uk/government/publications/greenhouse-gas-reporting-conversion- factors-2019 on 12 Nov 2019. Department for Business, Energy & Industrial Strategy, London. BEIS, 2019. Greenhouse gas reporting: conversion factors 2019, accessed at https://www.gov.uk/government/publications/greenhouse-gas-reporting-conversion- factors-2019 on 12 Nov 2019. Department for Business, Energy & Industrial Strategy, London. BEIS, 2019. Greenhouse gas reporting: conversion factors 2019, accessed at https://www.gov.uk/government/publications/greenhouse-gas-reporting-conversion- factors-2019 on 12 Nov 2019. Department for Business, Energy & Industrial Strategy, London. Bel, G., Rosell, J., 2017. The impact of socioeconomic characteristics on CO2 emissions associated with urban mobility: inequality across individuals. Energy Econ. 64, 251–261. i p gy gy act of socioeconomic characteristics on CO2 emissions associated with urban mobility: inequality across individuals. Energy Econ. 64 Bel, G., Rosell, J., 2017. The impact of socioeconomic characteristics on CO2 emissions associated with urban mobility: inequality 251–261. Brand, C., 2008. Personal Travel and Climate Change - Exploring Climate Change Emissions from Personal Travel Activity of Individuals and Households, first ed. Verlag Dr. Müller (VDM), Saarbrücken. ii Brand, C., 2008. Personal Travel and Climate Change - Exploring Climate Change Emissions from Personal Travel Activity of Individuals and Households, first ed. Verlag Dr. Müller (VDM), Saarbrücken. ii Brand, C., Anable, J., Morton, C., 2019. Lifestyle, efficiency and limits: modelling transport energy and emissions using a socio-technical approach. Energ. Effi. 12 (1), 187–207. Taming of the few - The unequal distribution of greenhouse gas emissions from personal travel in the UK. Energy Policy 36 (1), Brand, C., Boardman, B., 2008. Taming of the few - The unequal distribution of greenhouse gas emissions from personal travel 224–238. Brand, C., Goodman, A., Ogilvie, D., 2014. Evaluating the impacts of new walking and cycling infrastructure on carbon dioxide emissions from motorized travel: a controlled longitudinal study. Appl. Energy 128, 284–295. Brand, C., Goodman, A., Rutter, H., Song, Y., Ogilvie, D., 2013. References ACEA/ANFAC, 2014. European Motor Vehicle Parc 2014: Vehicles in Use (2009–2014). ANFAC/ACEA, Madrid. d l d i d hi l f i i h l i f h i Adams, J., 2010. Prevalence and socio-demographic correlates of “active transport” in the UK: analysis of the UK time use survey 2005. Prev. Med. 50 (4), 199–203. Adra, N., Michaux, J.-L., Andre, M., 2004. Analysis of the load factor and the empty running rate for road transport. Artemis - assessment and reliability of transport emission models and inventory systems, ARTEMIS - Assessment and reliability of transport emission models and inventory systems, Report INRETS-LTE 0419 INRETS Bron FR Alvanides, S., 2014. Active transport: why and where do people (not) walk or cycle? J. Transport Health 1 (4), 211–213. Anable, J., Brand, C., 2019. Energy, pollution and climate change. In: Docherty, I., Shaw, J. (Eds.), Transport Matters. Policy Press, Bristol, p. 452. Andor, M.A., Gerster, A., Gillingham, K.T., Horvath, M., 2020. Running a car costs much more than people think — stalling the uptake of green travel. Nature 580, 453–455. Baltagi, B.H., 2008. Econometric Analysis of Panel Data, fourth ed. Wiley, New York. Banister, D., 2008. The sustainable mobility paradigm. Transp. Policy 15 (2), 73–80. Banister, D., 2018. Inequality in Transport. Alexandrine Press, Marcham, UK. Banister D Watson S Wood C 1997 Sustainable cities transport energy and Banister, D., 2018. Inequality in Transport. Alexandrine Press, Marcham, UK. Banister, D., Watson, S., Wood, C., 1997. Sustainable cities, transport, energy and urban form. Environ. Plan. B: Plan. Des. 24 (1), 125–143. Bearman, N., Singleton, A.D., 2014. Modelling the potential impact on CO2 emissions of an increased uptake of active travel for the home to school commute using individual level data. J. Transport Health 1 (4), 295–304. 15 15 Transportation Research Part D 93 (2021) 102764 C. Brand et al. Beckx, C., Broekx, S., Degraeuwe, B., Beusen, B., Int Panis, L., 2013. Limits to active transport substitution of short car trips. Transp. Res.: Part D: Transport Environ. 22, 10–13. Beckx, C., Broekx, S., Degraeuwe, B., Beusen, B., Int Panis, L., 2013. Limits to active transport substitution of short car trips. Transp. Res.: Part D: Transport Environ. 22, 10–13. I., Janssens, D., Wets, G., 2010. Applying activity-travel data for the assessment of vehicle exhaust emissions: application of a GPS-e Transp. Res.: Part D: Transport Environ. 15 (2), 117–122. Beckx, C., Panis, L.I., Janssens, D., Wets, G., 2010. References Evaluation of voluntary travel behaviour change: experiences from three continents. Transp. Policy 16 (6), 281–292. Bryk, S.W., Raudenbush, A.S., 2002. Hierarchical Linear Models: Applications and Data Analysis Methods, second ed. Sage Publications, Thousand Oaks, CA. Büchs, M., Schnepf, S.V., 2013. Who emits most? Associations between socio-economic factors and UK households’ home energy, transport, indirect and total CO2 emissions. Ecol. Econ. 90, 114–123. y , , , , pp y , g , , Büchs, M., Schnepf, S.V., 2013. Who emits most? Associations between socio-economic factors and UK households’ home energy, transport, indirect and total CO2 emissions. Ecol. Econ. 90, 114–123. er, R., 2011. Determinants of transport mode choice: a comparison of Germany and the USA. J. Transp. Geogr. 19 (4), 644–657. , , p p y p g ( ), Cameron, I., Kenworthy, J.R., Lyons, T.J., 2003. Understanding and predicting private motorised urban mobility. Transp. Res.: Part D: Transport Environ. 8 (4), 267–283. I., Kenworthy, J.R., Lyons, T.J., 2003. Understanding and predicting private motorised urban mobility. Transp. Res.: Part D: Trans 83. Carlsson-Kanyama, A., Linden, A.-L., 1999. Travel patterns and environmental effects now and in the future: implications of differences in energy consumption among socio-economic groups. Ecol. Econ. 30 (3), 405–417. l g p ( ), dman, A., Mackett, R.L., Panter, J., Ogilvie, D., 2013. The factors influencing car use in a cycle-friendly city: the case of Cambridge. J. T Castro, A., Gaupp-Berhausen, M., Dons, E., Standaert, A., Laeremans, M., Clark, A., Anaya, E., Cole-Hunter, T., Avila-Palencia, I., Rojas-Rueda, D., Nieuwenhuijsen, M., Gerike, R., Panis, L.I., de Nazelle, A., Brand, C., Raser, E., Kahlmeier, S., G¨otschi, T., 2019. Physical activity of electric bicycle users compared to conventional bicycle users and non-cyclists: Insights based on health and transport data from an online survey in seven European cities. Transp. Res. Interdis. Perspect. 100017. e, K., Melia, S., 2016. Changes to commute mode: The role of life events, spatial context and environmental attitude. Transp. Res. 105. i p Clark, B., Chatterjee, K., Melia, S., 2016. Changes to commute mode: The role of life events, spatial context and environmental Pract. 89, 89–105. i Creutzig, F., Agoston, P., Minx, J.C., Canadell, J.G., Andrew, R.M., Qu´er´e, C.L., Peters, G.P., Sharifi, A., Yamagata, Y., Dhakal, S., 2016. Urban infrastructure choices structure climate solutions. Nat. Clim. Change 6 (12), 1054–1056. References Can strategic planning contribute to a reduction in car-based travel? Transp. Policy 3 (1–2), 55–65. Curtis, C., 1996. Can strategic planning contribute to a reduction in car-based travel? Transp. Policy 3 (1–2), 55–65. de Nazelle, A., Morton, B.J., Jerrett, M., Crawford-Brown, D., 2010. Short trips: an opportunity for reducing mobile-source emissi E i 15 (8) 451 457 Curtis, C., 1996. Can strategic planning contribute to a reduction in car based travel? Transp. Policy 3 (1 2), 55 65. de Nazelle, A., Morton, B.J., Jerrett, M., Crawford-Brown, D., 2010. Short trips: an opportunity for reducing mobile-source emissi Environ. 15 (8), 451–457. g p y , rett, M., Crawford-Brown, D., 2010. Short trips: an opportunity for reducing mobile-source emissions? Transp. Res.: Part D: Transport DEFRA/DECC, 2016. UK Government conversion factors for Company Reporting, full 2016 dataset. Department for the Environment, Food and Rural Affairs and Department for Energy and Climate Change, London. p gy g DfT, 2015. Transport Statistics Great Britain: 2015 Edition. Department for Transport, London. nsport Statistics Great Britain: 2015 Edition. Department for Transport, London. Dons, E., Gotschi, T., Nieuwenhuijsen, M., de Nazelle, A., Anaya, E., Avila-Palencia, I., Brand, C., Cole-Hunter, T., Gaupp-Berghausen, M., Kahlmeier, S., Laeremans, M., Mueller, N., Orjuela, J.P., Raser, E., Rojas-Rueda, D., Standaert, A., Stigell, E., Uhlmann, T., Gerike, R., Int Panis, L., 2015. Physical Activity through Sustainable Transport Approaches (PASTA): protocol for a multi-centre, longitudinal study. BMC Public Health 15 (1), 1126. ons, E., Gotschi, T., Nieuwenhuijsen, M., de Nazelle, A., Anaya, E., Avila-Palencia, I., Brand, C., Cole-Hunter, T., Gaupp-Berghause Laeremans, M., Mueller, N., Orjuela, J.P., Raser, E., Rojas-Rueda, D., Standaert, A., Stigell, E., Uhlmann, T., Gerike, R., Int Pani through Sustainable Transport Approaches (PASTA): protocol for a multi-centre, longitudinal study. BMC Public Health 15 (1), mans, M., Mueller, N., Orjuela, J.P., Raser, E., Rojas-Rueda, D., Standaert, A., Stigell, E., Uhlmann, T., Gerike, R., Int Panis, L., 20 gh Sustainable Transport Approaches (PASTA): protocol for a multi-centre, longitudinal study. BMC Public Health 15 (1), 1126. Dons, E., Rojas-Rueda, D., Anaya-Boig, E., Avila-Palencia, I., Brand, C., Cole-Hunter, T., de Nazelle, A., Eriksson, U., Gaupp-Berghausen, M., Gerike, R., Kahlmeier, S., Laeremans, M., Mueller, N., Nawrot, T., Nieuwenhuijsen, M.J., Orjuela, J.P., Racioppi, F., Raser, E., Standaert, A., Int Panis, L., G¨otschi, T., 2018. Transport mode choice and body mass index: cross-sectional and longitudinal evidence from a European-wide study. Environ. Int. 119, 109–116. References choice and body mass index: cross-sectional and longitudinal evidence from a European-wide study. Environ. Int. 119, 109 l f l d h l f f l l d ECF, 2011. Cycle more Often 2 cool down the planet! - Quantifying CO2 savings of Cycling. European Cyclists’ Federation (ECF), Brussels. Ecometrica 2011 Electricity specific emission factors for grid electricity Ecometrica ECF, 2011. Cycle more Often 2 cool down the planet! - Quantifying CO2 savings of Cycling. European Cyclists’ Federation (ECF), Brussels. Ecometrica, 2011. Electricity-specific emission factors for grid electricity. Ecometrica. i EEA, 2012. COPERT 4 (COmputer Programme to calculate Emissions from Road Transport), last accessed at http://emisia.com/content/copert-documentation on 20/ 02/2018. European Environment Agency, Copenhagen. EEA, 2016. Carbon dioxide emissions from passenger transport, TERM 027 indicator and TRACCS database. European Environment Agency, Copenhagen. EEA, 2019. Total greenhouse gas emission trends and projections in Europe, accessed at https://www.eea.europa.eu/data-and-maps/indicators/greenhouse-gas- emission-trends-6/assessment-3 on 30/03/2020. European Environment Agency, Copenhagen. p g y, p g P/EEA air pollutant emission inventory guidebook 2016, Technical guidance to prepare national emission inventories. European Env en. MEP/EEA, 2016. EMEP/EEA air pollutant emission inventory guidebook 2016, Technical guidance to prepare national emission inve Agency, Copenhagen. EPOMM, 2020. TEMS - The EPOMM Modal Split Tool, accessed at http://www.epomm.eu/tems/index.phtml on 21/3/2020. European Platform on Mobility Management (EPOMM), Leuven, BE. Eurostat, 2016. Transport data database, October 2015 update. Last accessed at http://ec.europa.eu/eurostat/web/transport/data/database on 10/03/2017. Eurostat, European Commission, Brussels. nsport and environment statistics: 2020 edition, accessed at https://ec.europa.eu/eurostat/documents/3217494/11478276/KS-DK- d-1745-76b5-838e-013524781340 Eurostat, European Union, Brussels. urostat, 2020. Energy, transport and environment statistics: 2020 edition, accessed at https://ec.europa.eu/eurostat/documents/32 001-EN-N.pdf/06ddaf8d-1745-76b5-838e-013524781340 Eurostat, European Union, Brussels. 16 C. Brand et al. Transportation Research Part D 93 (2021) 102764 Frank, L.D., Greenwald, M.J., Winkelman, S., Chapman, J., Kavage, S., 2010. Carbonless footprints: promoting health and climate stabilization through active transportation. Prev. Med. 50 (Suppl 1), S99–S105. Frank, L.D., Greenwald, M.J., Winkelman, S., Chapman, J., Kavage, S., 2010. Carbonless footprints: promoting health and climate stabilization through active transportation. Prev. Med. 50 (Suppl 1), S99–S105. p pp Gascon, M., G¨otschi, T., Nazelle, A.d., Gracia, E., Ambr`os, A., M´arquez, S., Marquet, O., Avila-Palencia, I., Brand, C., Iacorossi, F., Raser, E., Gaupp-Berghausen, M., Dons, E., Laeremans, M., Kahlmeier, S., S´anchez, J., Gerike, R., Anaya-Boig, E., Panis, L.I., Nieuwenhuijsen, M., 2019. Correlates of Walking for Travel in Seven European Cities: The PASTA Project. Environ. Health Perspect. 127(9), 097003. References Gaupp-Berghausen, M., Raser, E., Anaya-Boig, E., Avila-Palencia, I., de Nazelle, A., Dons, E., Franzen, H., Gerike, R., G¨otschi, T., Iacorossi, F., H¨ossinger, R., Nieuwenhuijsen, M., Rojas-Rueda, D., Sanchez, J., Smeds, E., Deforth, M., Standaert, A., Stigell, E., Cole-Hunter, T., Int Panis, L., 2019. Evaluation of different recruitment methods: longitudinal, web-based, pan-European physical activity through sustainable transport approaches (PASTA) project. J. Med. Internet Res. 21 (5), e11492. , Gerike, R., de Nazelle, A., Nieuwenhuijsen, M., Panis, L.I., Anaya, E., Avila-Palencia, I., Boschetti, F., Brand, C., Cole-Hunter, T., Dons, E., Eriksson, U., Gaupp- Berghausen, M., Kahlmeier, S., Laeremans, M., Mueller, N., Orjuela, J.P., Racioppi, F., Raser, E., Rojas-Rueda, D., Schweizer, C., Standaert, A., Uhlmann, T., Wegener, S., G¨otschi, T., 2016. Physical Activity through Sustainable Transport Approaches (PASTA): a study protocol for a multicentre project. BMJ Open 6 (1), e009924. Goodman, A., Brand, C., Ogilvie, D., 2012. Associations of health, physical activity and weight status with motorised travel and transport carbon dioxide emissions: a cross-sectional, observational study. Environ. Health 11 (1), 52. Goodman, A., Rojas, I.F., Woodcock, J., Aldred, R., Berkoff, N., Morgan, M., Abbas, A., Lovelace, R., 2019. Scenarios of cycling to school in England, and associated health and carbon impacts: application of the ‘Propensity to Cycle Tool’. J. Transport Health 12, 263–278. Goodman, A., Sahlqvist, S., Ogilvie, D., 2014. New walking and cycling routes and increased physical activity: one- and 2-year findings from the UK iConnect study. qvist, S., Ogilvie, D., 2014. New walking and cycling routes and increased physical activity: one- and 2-year findings from the UK iC Health e1–e9. Am. J. Public Health e1–e9. G¨ossling, S., Choi, A., Dekker, K., Metzler, D., 2019. The social cost of automobility, cycling and walking in the European Union. Ecol. Econ. 158, 65–74. G¨otschi, T., de Nazelle, A., Brand, C., Gerike, R., 2017. Towards a comprehensive conceptual framework of active travel behavior: a review and synthesis of published G¨ossling, S., Choi, A., Dekker, K., Metzler, D., 2019. The social cost of automobility, cycling and walking in the European Union. Ecol. Econ. 158, 65–74. G¨otschi, T., de Nazelle, A., Brand, C., Gerike, R., 2017. Towards a comprehensive conceptual framework of active travel behavior: a review and synthesis of published frameworks. Curr. Environ. Health Rep. 4 (3), 286–295. G¨otschi, T., Kahlmeier, S., Castro, A., Brand, C., Cavill, N., Kelly, P., Lieb, C., Rojas-Rueda, D., Woodcock, J., Racioppi, F., 2020. Sims, R., Schaeffer, R., Creutzig, F., Cruz-Nú˜nez, X., D’Agosto, M., Dimitriu, D., Meza, M.J.F., Fulton, L., Kobayashi, S., O., L., McKinnon, A., Newman, P., Ouyang, M., Schauer, J.J., Sperling, D., Tiwari, G., 2014. Transport. In: Edenhofer, O., Pichs-Madruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier, P., Kriemann, B., Savolainen, J., Schl¨omer, S., Stechow, C.v., Zwickel, T., Minx, J.C. (Eds.), Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. SMMT, 2016. UK new car market starts 2016 on a high with best January in 11 years, http://www.smmt.co.uk/2016/02/uk-new-car-market-starts-2016-on-a-high- with-best-january-in-11-years/ [last accessed on 18/01/2016]. SMMT, London. References Integrated impact assessment of active travel: expanding the scope of the health economic assessment tool (HEAT) for walking and cycling. Int. J. Environ. Res. Public. Health 17 (20). Graham-Rowe, E., Skippon, S., Gardner, B., Abraham, C., 2011. Can we reduce car use and if so, how? A review of available evidence. Transp. Res.: Part A: Pol. Practice 45 (5), 401–418. ( ), ICCT, 2017. Road tested: Comparative overview of real-world versus type-approval NOX and CO2 emissions from diesel cars in Europe, ICCT White Paper. Last accessed at https://www.theicct.org/sites/default/files/publications/ICCT_RoadTested_201709.pdf on 18/04/2018. International Council on Clean Transportation, Berlin. p IEA, 2015. World energy outlook 2015. International Energy Agency (IEA), Paris. IIASA 2014 IIASA GAINS model scenario WPE 2014 CLE: the updated ‘current legislation’ (after the bilateral consultations in 2014) of the PRIMES 2013 IEA, 2015. World energy outlook 2015. International Energy Agency (IEA), Paris. IIASA, 2014. IIASA GAINS model, scenario WPE_2014_CLE: the updated ‘current legislation’ (after the bilateral consultations in REFERENCE activity projection. IIASA, Laxenburg, Austria. IPCC, 2018. Global Warming of 1.5◦C, Special Report. Last accessed in October 2018 at: http://www.ipcc.ch/report/sr15/. Intergovernmental Panel on Climate Change, Geneva. Javaid, A., Creutzig, F., Bamberg, S., 2020. Determinants of low-carbon transport mode adoption: systematic review of reviews. Environ. Res. Lett. 15 (10), 103002. JEC, 2014. JEC Well-To-Wheels Analysis, Report EUR 26237 EN - 2014. Last accessed at http://iet.jrc.ec.europa.eu/about-jec/sites/iet.jrc.ec.europa.eu.about-jec/ files/documents/report_2014/wtt_report_v4a.pdf on 10/03/2017. JEC - Joint Research Centre-EUCAR-CONCAWE collaboration, Brussels. JRC 2013 Analysis of National Travel Statistics in Europe European Commission Joint Research Centre ISBN: 978 92 79 32358 4 Seville Javaid, A., Creutzig, F., Bamberg, S., 2020. Determinants of low-carbon transport mode adoption: systematic review of reviews. E JEC, 2014. JEC Well-To-Wheels Analysis, Report EUR 26237 EN - 2014. Last accessed at http://iet.jrc.ec.europa.eu/about-jec/s files/documents/report_2014/wtt_report_v4a.pdf on 10/03/2017. JEC - Joint Research Centre-EUCAR-CONCAWE collabora s of National Travel Statistics in Europe. European Commission, Joint Research Centre. ISBN: 978-92-79-32358-4, Seville. C Ch R H d Ch P 2018 R d i i b di id i i f i i li . Analysis of National Travel Statistics in Europe. European Commission, Joint Research Centre. ISBN: 978-92-79-32358-4, Seville ., Shaw, C., Chapman, R., Howden-Chapman, P., 2018. Reductions in carbon dioxide emissions from an intervention to promote cycli JRC, 2013. Analysis of National Travel Statistics in Europe. European Commission, Joint Research Centre. ISBN: 978 92 79 32358 4, Seville. Keall, M.D., Shaw, C., Chapman, R., Howden-Chapman, P., 2018. References Reductions in carbon dioxide emissions from an intervention to promote cycling and walking: a case study from New Zealand Transp Res : Part D: Transport Environ 65 687–696 Keall, M.D., Shaw, C., Chapman, R., Howden-Chapman, P., 2018. Reductions in carbon dioxide emissions from an intervention to promote cycling and walking: a case study from New Zealand. Transp. Res.: Part D: Transport Environ. 65, 687–696. ., Shaw, C., Chapman, R., Howden-Chapman, P., 2018. Reductions in carbon dioxide emissions from an intervention to promote cycli from New Zealand. Transp. Res.: Part D: Transport Environ. 65, 687–696. Keall, M.D., Shaw, C., Chapman, R., Howden-Chapman, P., 2018. Reductions in carbon dioxide emissions from an intervention to p study from New Zealand. Transp. Res.: Part D: Transport Environ. 65, 687–696. ll h f h d ff b lf d d l b l Kelly, P., Krenn, P., Titze, S., Stopher, P., Foster, C., 2013. Quantifying the difference between self-reported and global positioning systems-measured journey durations: a systematic review. Transport Rev. 33 (4), 443–459. Kelly, P., Krenn, P., Titze, S., Stopher, P., Foster, C., 2013. Quantifying the difference between self-reported and global positioning systems-measured journey durations: a systematic review. Transport Rev. 33 (4), 443–459. Ki l S M T dij C G ld i E G d Y H Li i h M 2020 P j i h i i d i f SARS C V 2 h h h d i i d Ko, J., Park, D., Lim, H., Hwang, I.C., 2011. Who produces the most CO2 emissions for trips in the Seoul metropolis area? Transp. Res.: Part D: Transport Environ. 16 (5), 358–364. i R., Beck, S.B.M., Watson, M., Wild, A., 2011. Assessing the energy implications of replacing car trips with bicycle trips in Sheffield, UK –2087. on, M., Wild, A., 2011. Assessing the energy implications of replacing car trips with bicycle trips in Sheffield, UK. Energy Policy 39 (4) Lovelace, R., Beck, S.B.M., Watson, M., Wild, A., 2011. Assessing the energy implications of replacing car trips with bicycle trips in Sheffield, UK. Energy Policy 39 (4), 2075–2087. , , , , , , , , g gy p p g p y pi , gy y ( ), 2075–2087. Neves A Brand C 2019 Assessing the potential for carbon emissions savings from replacing short car trips with walking and cycling using a mixed GPS-travel diary 2075 2087. Neves, A., Brand, C., 2019. References E G B h M D E A B i E A il P l i I B d C C t A Cl k A E ik U G¨t hi Quarmby, S., Santos, G., Mathias, M., 2019. Air quality strategies and technologies: a rapid review of the international evidence. Sustainability 11 (10). Rabl, A., de Nazelle, A., 2012. Benefits of shift from car to active transport. Transport Policy 19 (1), 121–131. Raser, E., Gaupp-Berghausen, M., Dons, E., Anaya-Boig, E., Avila-Palencia, I., Brand, C., Castro, A., Clark, A., Eriksson, U., G¨otschi, T., Int Panis, L., Kahlmeier, S., Raser, E., Gaupp-Berghausen, M., Dons, E., Anaya-Boig, E., Avila-Palencia, I., Brand, C., Castro, A., Clark, A., Eriksson, U., G¨otschi, T., Int Panis, L., Kahlmeier, S., Laeremans, M., Mueller, N., Nieuwenhuijsen, M., Orjuela, J.P., Rojas-Rueda, D., Standaert, A., Stigell, E., Gerike, R., 2018. European cyclists’ travel behavior: Differences and similarities between seven European (PASTA) cities. J. Transport Health 9, 244–252. Laeremans, M., Mueller, N., Nieuwenhuijsen, M., Orjuela, J.P., Rojas-Rueda, D., Standaert, A., Stigell, E., Gerike, R., 2018. European cyclists’ travel behavior: Differences and similarities between seven European (PASTA) cities. J. Transport Health 9, 244–252. Rodrigues P F Alvim Ferraz M C M Martins F G Saldiva P S´a T H Sousa S I V 2020 Health economic assessment of a shift to active transport Environ p p az, M.C.M., Martins, F.G., Saldiva, P., S´a, T.H., Sousa, S.I.V., 2020. Health economic assessment of a shift to active transport. Envi Sælensminde, K., 2004. Cost–benefit analyses of walking and cycling track networks taking into account insecurity, health effects and external costs of motorized traffic. Transp. Res.: Part A: Pol. Pract. 38 (8), 593–606. i Scheepers, C.E., Wendel-Vos, G.C.W., den Broeder, J.M., van Kempen, E.E.M.M., van Wesemael, P.J.V., Schuit, A.J., 2014. Shifting from car to active transport: a systematic review of the effectiveness of interventions. Transp. Res.: Part A: Pol. Pract. 70, 264–280. i y p , Shaw, C., Hales, S., Howden-Chapman, P., Edwards, R., 2014. Health co-benefits of climate change mitigation policies in the transp 427–433. pman, P., Edwards, R., 2014. Health co-benefits of climate change mitigation policies in the transport sector. Nat. Clim. Change 4 (6) Sims, R., Schaeffer, R., Creutzig, F., Cruz-Nú˜nez, X., D’Agosto, M., Dimitriu, D., Meza, M.J.F., Fulton, L., Kobayashi, S., O., L., McKinnon, A., Newman, P., Ouyang, M., Schauer, J.J., Sperling, D., Tiwari, G., 2014. Transport. References Assessing the potential for carbon emissions savings from replacing short car trips with walking and cycling using a mixed GPS-travel diary approach. Transp. Res.: Part A: Pol. Pract. 123, 130–146. Neves, A., Brand, C., 2019. Assessing the potential for carbon emissions savings from replacing short car trips with walking and cyc approach. Transp. Res.: Part A: Pol. Pract. 123, 130–146. pp p , Nicolas, J.-P., David, D., 2009. Passenger transport and CO2 emissions: what does the French transport survey tell us? Atmos. Environ. 43 (5), 1015–1020. Nieuwenhuijsen, M.J., 2020. Urban and transport planning pathways to carbon neutral, liveable and healthy cities; A review of the current evidence. Environ. Int. 105661. 105661. Odeh, N., Hill, N., Forster, D., 2013. Current and Future Lifecycle Emissions of Key “Low Carbon” Technologies and Alternatives, Final Report. Ricardo AEA for the Committee on Climate Change, Harwell, UK. i g , Jones, A., van Sluijs, E., 2013. Individual, socio-cultural and environmental predictors of uptake and maintenance of active gitudinal results from the SPEEDY study. Int. J. Behav. Nutr. Phys. Act. 10 (1), 83. Panter, J., Corder, K., Griffin, S., Jones, A., van Sluijs, E., 2013. Individual, socio cultural and environmental predictors of uptake and maintenance of active commuting in children: longitudinal results from the SPEEDY study. Int. J. Behav. Nutr. Phys. Act. 10 (1), 83. Preston, I., White, V., Thumim, J., Bridgeman, T., Brand, C., 2013. Distribution of Carbon Emissions in the UK: Implications for Domestic Energy Policy. Joseph Rowntree Foundation, London. g g y y ( ), Preston, I., White, V., Thumim, J., Bridgeman, T., Brand, C., 2013. Distribution of Carbon Emissions in the UK: Implications for Domestic Energy Policy. Joseph Rowntree Foundation, London. J., Bridgeman, T., Brand, C., 2013. Distribution of Carbon Emissions in the UK: Implications for Domestic Energy Policy. Joseph on. , Quarmby, S., Santos, G., Mathias, M., 2019. Air quality strategies and technologies: a rapid review of the international evidence. Sustainability 11 (10). Rabl, A., de Nazelle, A., 2012. Benefits of shift from car to active transport. Transport Policy 19 (1), 121–131. uarmby, S., Santos, G., Mathias, M., 2019. Air quality strategies and technologies: a rapid review of the international evidence. Su abl, A., de Nazelle, A., 2012. Benefits of shift from car to active transport. Transport Policy 19 (1), 121–131. land transport. Lancet 374. Yang, Y., Wang, C., Liu, W., 2018. Urban daily travel carbon emissions accounting and mitigation potential analysis using surveyed individual data. J. Cleaner Prod. 192, 821–834. References In: Edenhofer, O., Pichs-Madruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier, P., Kriemann, B., Savolainen, J., Schl¨omer, S., Stechow, C.v., Zwickel, T., Minx, J.C. (Eds.), Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. SMMT, 2016. UK new car market starts 2016 on a high with best January in 11 years, http://www.smmt.co.uk/2016/02/uk-new-car-market-starts-2016-on-a-high- with-best-january-in-11-years/ [last accessed on 18/01/2016]. SMMT, London. SMMT, 2016. UK new car market starts 2016 on a high with best January in 11 years with-best-january-in-11-years/ [last accessed on 18/01/2016]. SMMT, London. 17 Transportation Research Part D 93 (2021) 102764 C. Brand et al. Socialdata, 2009. The New KONTIV-Design (NKD), accessed at http://www.socialdata.de/info/KONTIV_engl.pdf on 8 September 2019 Socialdata GmbH, Munich. Stead, D., 1999. Relationships between transport emissions and travel patterns in Britain. Transp. Policy 6 (4), 247–258. Socialdata, 2009. The New KONTIV-Design (NKD), accessed at http://www.socialdata.de/info/KONTIV_engl.pdf on 8 September 2019 Socialdata GmbH, Munich. Stead, D., 1999. Relationships between transport emissions and travel patterns in Britain. Transp. Policy 6 (4), 247–258. il d di id l b di id i i d i l f d i i h h l d d h i d i d Socialdata, 2009. The New KONTIV-Design (NKD), accessed at http://www.socialdata.de/info/KONTIV_engl.pdf on 8 September 2019 Socialdata GmbH, Munich. Stead, D., 1999. Relationships between transport emissions and travel patterns in Britain. Transp. Policy 6 (4), 247–258. Susilo, Y.O., Stead, D., 2009. Individual carbon dioxide emissions and potential for reduction in the Netherlands and the United Kingdom. Transp. Res. Rec. 2139, 142–152. Socialdata, 2009. The New KONTIV-Design (NKD), accessed at http://www.socialdata.de/info/KONTIV_engl.pdf on 8 Septemb Stead D 1999 Relationships between transport emissions and travel patterns in Britain Transp Policy 6 (4) 247–258 Susilo, Y.O., Stead, D., 2009. Individual carbon dioxide emissions and potential for reduction in the Netherlands and the Unite 142–152. Tainio, M., Monsivais, P., Jones, N.R., Brand, C., Woodcock, J., 2017. Mortality, greenhouse gas emissions and consumer cost impacts of combined diet and physical activity scenarios: a health impact assessment study. BMJ Open 7 (2). Tainio, M., Monsivais, P., Jones, N.R., Brand, C., Woodcock, J., 2017. Mortality, greenhouse gas emissions and consumer cost impacts of combined diet and physical activity scenarios: a health impact assessment study. BMJ Open 7 (2). References y p y p Timmermans, H., van der Waerden, P., Alves, M., Polak, J., Ellis, S., Harvey, A.S., Kurose, S., Zandee, R., 2003. Spatial context and the complexity of daily travel patterns: an international comparison. J. Transp. Geogr. 11 (1), 37–46. Timmermans, H., van der Waerden, P., Alves, M., Polak, J., Ellis, S., Harvey, A.S., Kurose, S., Zandee, R., 2003. Spatial context and the complexity of daily travel patterns: an international comparison. J. Transp. Geogr. 11 (1), 37–46. Timmermans, H., van der Waerden, P., Alves, M., Polak, J., Ellis, S., Harvey, A.S., Kurose, S., Zandee, R., 2003. Spatia patterns: an international comparison. J. Transp. Geogr. 11 (1), 37–46. p p g f Transportation, 2017. National Household Travel Survey: Vehicle Trips, accessed at https://nhts.ornl.gov/vehicle-trips on 20/03/ Transportation, Federal Highway Administration, Washington, DC. p p p g U.S. Department of Transportation, 2017. National Household Travel Survey: Vehicle Trips, accessed at https://nhts.ornl.gov/v Department of Transportation, Federal Highway Administration, Washington, DC. Vagane, L., 2007. Short car trips in Norway: is there a potential for modal shift? Proceedings of the European Transport Conference (ETC) 2007 held 17-19 October 2007, Leiden, The Netherlands. Vagane, L., 2007. Short car trips in Norway: is there a potential for modal shift? Proceedings of the European Transport Conference (ETC) 2007 held 17-19 October 2007, Leiden, The Netherlands. Vagane, L., 2007. Short car trips in Norway: is there a potential for modal shift? Proceedings of the European Transpor 2007, Leiden, The Netherlands. VanderWeele, T.J., 2016. Mediation analysis: a practitioner’s guide. Annu. Rev. Public Health 37 (1), 17–32. y p g Vittinghoff, E., Glidden, D.V., Shiboski, S.C., McCulloch, C.E., 2012. Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models, second ed. Springer, Boston, MA. Vittinghoff, E., Glidden, D.V., Shiboski, S.C., McCulloch, C.E., 2012. Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models, second ed. Springer, Boston, MA. G¨ hi i i hl i S i 2012 A i h i l i i d b d i h i d l A d 42 ( ) Wanner, M., G¨otschi, T., Martin-Diener, E., Kahlmeier, S., Martin, B.W., 2012. Active transport, physical activity, and body weight in adults. Am. J. Prev. Med. 42 (5), 493–502. rels, A., 2000. Modelling lifestyle effects on energy demand and related emissions. Energy Policy 28 (8), 549–566. References 013 E i i Th di ib i f i d i i ff d bl h i i T Weber, C., Perrels, A., 2000. Modelling lifestyle effects on energy demand and related emissions. Energy Policy 28 (8), 549–566. Welch, T.F., 2013. Equity in transport: The distribution of transit access and connectivity among affordable housing units. Transp. P Welch, T.F., 2013. Equity in transport: The distribution of transit access and connectivity among affordable housing units. Transp. Policy 30, 283–293. Woodcock, J., Abbas, A., Ullrich, A., Tainio, M., Lovelace, R., S´a, T.H., Westgate, K., Goodman, A., 2018. Development of the Impacts of Cycling Tool (ICT): A Welch, T.F., 2013. Equity in transport: The distribution of transit access and connectivity among affordable housing units. Transp. P Woodcock, J., Abbas, A., Ullrich, A., Tainio, M., Lovelace, R., S´a, T.H., Westgate, K., Goodman, A., 2018. Development of the Impa d ll d d b l f l h l h d l f l k d ( ) Woodcock, J., Abbas, A., Ullrich, A., Tainio, M., Lovelace, R., S´a, T.H., Westgate, K., Goodman, A., 2018. Development of the Impacts of Cycling Tool (ICT): A d lli t d d b t l f l ti h lth d i t l i t f li t k PL S M d 15 (7) 1002622 p Yang, Y., Wang, C., Liu, W., 2018. Urban daily travel carbon emissions accounting and mitigation potential analysis using surveyed individual data. J. Cleaner Prod. 192, 821–834. 18 18
https://openalex.org/W3148642024	https://upcommons.upc.edu/bitstream/2117/335424/1/TPAUdF1de1.pdf	Latin	null	BEST : Bézier-Enhanced Shell Triangle : a new rotation-free thin shell finite element	null	2,023	public-domain	119,650	Pere-Andreu Ubach de Fuentes ADVERTIMENT La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents condicions d'ús: La difusió d’aquesta tesi per mitjà del repositori institucional UPCommons (http://upcommons.upc.edu/tesis) i el repositori cooperatiu TDX ( h t t p : / / w w w . t d x . c a t / ) ha estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei UPCommons o TDX. No s’autoritza la presentació del seu contingut en una finestra o marc aliè a UPCommons (framing). Aquesta reserva de drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita de parts de la tesi és obligat indicar el nom de la persona autora. ADVERTENCIA La consulta de esta tesis queda condicionada a la aceptación de las siguientes condiciones de uso: La difusión de esta tesis por medio del repositorio institucional UPCommons (http://upcommons.upc.edu/tesis) y el repositorio cooperativo TDR (http://www.tdx.cat/?locale- attribute=es) ha sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio UPCommons No se autoriza la presentación de su contenido en una ventana o marco ajeno a UPCommons (framing). Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la persona autora. WARNING On having consulted this thesis you’re accepting the following use conditions: Spreading this thesis by the institutional repository UPCommons (http://upcommons.upc.edu/tesis) and the cooperative repository TDX (http://www.tdx.cat/?locale- attribute=en) has been authorized by the titular of the intellectual property rights only for private uses placed in investigation and teaching activities. Reproduction with lucrative aims is not authorized neither its spreading nor availability from a site foreign to the UPCommons service. Introducing its content in a window or frame foreign to the UPCommons service is not authorized (framing). These rights affect to the presentation summary of the thesis as well as to its contents. Pere-Andreu Ubach de Fuentes In the using or citation of parts of the thesis it’s obliged to indicate the name of the author. Abstract Se presenta un nuevo elemento ﬁnito de lámina delga- da. Este nuevo elemento no usa rotaciones como gra- dos de libertad. En su lugar, para satisfacer el requi- sito de mantener continuidad C1 entre elementos, el autor recurre a mejorar la descripción geométrica de los triángulos planos de una malla formada por trián- gulos lineales, por medio de polinomios de Bernstein y particiones triangulares de Bernstein-Bézier. A new thin shell ﬁnite element is presented. This new element doesn’t have rotational degrees of freedom. Instead, in order to overcome the C1 continuity requirement across elements, the au- thor resorts to enhance the geometric description of the ﬂat triangles of a mesh made out of linear triangles, by means of Bernstein polynomials and triangular Bernstein-Bézier patches. Para deﬁnir dichas particiones de Bernstein- Bézier, el autor realiza una estimación de las norma- les a la superﬁcie en los nodos de una malla de trián- gulos. Ubach, Estruch y García-Espinosa realizaron una comparación estadística exhaustiva entre distin- tos factores de ponderación, incluyendo otros usados previamente en la literatura. La conclusión de dicho trabajo (que se incluye como anejo de esta tesis) con- duce a usar como factor de ponderación para calcular una media de los vectores normales de los triángulos: el inverso del área de la circunferencia circunscrita al triángulo y el ángulo interno del triángulo en el nodo considerado. Usando este nuevo factor de pon- deración, se reduce en aproximadamente un 10 % el error medio cuadrático cometido en la estimación de las normales de superﬁcies generadas aleatoriamente, respecto del mejor factor de ponderación usado previa- mente en la literatura. The author estimates the surface normals at the nodes of a mesh of triangles, in order to use them to deﬁne the Bernstein-Bézier patches. Ubach, Estruch and García-Espinosa performed a comprehensive statistical comparison of differ- ent weighting factors, including other weights previously used in the literature. The conclu- sion of that work (included as an appendix to this thesis) is that the inverse of the area of the cir- cumscribed circle to the triangle and the inter- nal angle of the triangle at the node considered, should be used as weighting factor. Using this new weighting factor, we reduce by about 10% the root mean square error in the estimation of normals of randomly generated surfaces with re- spect to the previous best weighting factor found in the literature. PhD program in Structural Analysis BEST: Bézier-Enhanced Shell Triangle A NEW ROTATION-FREE THIN SHELL FINITE ELEMENT A thesis submitted in partial fulﬁllment of the requirements for the degree of PhD in Civil and Environmental Engineering by the Polytechnic University of Catalonia This thesis was partially supported by the International Center for Numerical Methods in Engineering Doctoral thesis by: Pere-Andreu Ubach de Fuentes Thesis Advisors: Eugenio Oñate Ibáñez de Navarra Julio García-Espinosa Departament d’Enginyeria Civil i Ambiental Departament de Resistència de Materials i Estructures en Enginyeria Barcelona, December 2019 All rights reserved This thesis as partially supported by he International Center for Numerical Methods in Engineering Barcelona, December 2019 All rights reserved ABSTRACT Se presenta un nuevo paradigma que consiste en reconstruir la geometría de un elemento triangular cúbico. Esta reconstrucción geométrica aprovecha las propiedades de las funciones cúbicas B-spline (trián- gulo cúbico de Bézier). De esta forma, el autor crea un elemento de lámina conforme. A new paradigm is presented consisting on the reconstruction of the geometry of a cubic tri- angular element. This geometric reconstruction exploits the properties of cubic B-spline functions (cubic Bézier triangle). This way, the author builds a conforming shell ﬁnite element. El elemento BEST trata de resolver el problema de la dependencia de la precisión de los elementos de lá- mina sin rotaciones con respecto a la distorsión de la malla. El elemento BEST deﬁne la curvatura como un tensor 2D. Esta deﬁnición es consistente con la deﬁni- ción de la curvatura obtenida de la geometría diferen- cial. The BEST element tackles the problem of rotation-free shell elements’ precision depen- dency with respect to mesh distortion. The BEST element deﬁnes the curvature as a 2D tensor. This deﬁnition is consistent with the deﬁnition of curvature obtained by differential geometry. The construction of the cubic Bézier triangle requires ﬁxing 30 parameters (3 coordinates for each of the 10 control points). Therefore it needs to apply 30 independent conditions. 15 of these conditions are given directly by the positions of the 3 vertices of the triangle and the orientations of the normal vectors at the 3 vertices. With these 15 conditions the triangles maintain C1 continu- ity at the vertices. La construcción del triángulo cúbico de Bézier re- quiere determinar 30 parámetros (3 coordenadas pa- ra cada uno de los 10 puntos de control). Para ello es necesario aplicar 30 condiciones independientes. 15 de estas condiciones se deducen de manera directa de la posición de los 3 vértices del triángulo y de las orientaciones de los vectores normales en los 3 vértices. Con estas 15 condiciones se asegura que los triángulos mantienen continuidad C1 en los vértices. 8 of the remaining conditions are imposed introducing energy minimization considerations. These energy minimization considerations serve also to deﬁne a well-posed element. A set of re- duced problems, which are sufﬁciently represen- tative and general, is deﬁned in order to avoid imposing the energy minimization considerations globally. Abstract The author uses the information of the nor- mal vectors at the nodes and the triangular Bernstein-Bézier patches to build cubic Bézier triangles. These cubic Bézier triangles are sur- face interpolants; C1 continuous at the nodes and C0 continuous across the edges. Owing to this approach, the new element is called Bézier- enhanced shell triangle (BEST). Con la información de los vectores normales en los nodos y las particiones triangulares de Bernstein- Bézier, el autor construye triángulos cúbicos de Bézier. Estos triángulos cúbicos de Bézier interpolan la su- perﬁcie; con continuidad C1 en los nodos y C0 a través de las aristas. Debido a este planteamiento, el nuevo elemento recibe el nombre de BEST. El elemento BEST aprovecha todas las conectivi- dades de los nodos de cada triángulo de la malla. El cálculo de los vectores normales en los nodos no de- pende del número de triángulos que rodean cada nodo de la malla. El elemento BEST es independiente de la topología de la malla. The BEST element takes advantage of all the nodes’ connectivities in each triangle of the mesh. The computation of the normal vectors at the nodes doesn’t depend on the number of triangles surrounding each node of the mesh. The BEST element is independent from the mesh topology. i ii ABSTRACT More speciﬁcally, the author deﬁnes 3 different reduced problems for the 3 different shell deformation modes: bending deformation, membrane (in-plane extension) deformation and in-plane shear (drilling rotation) deformation. De las otras 15 condiciones que quedan por impo- ner, 8 de ellas se obtienen a partir de criterios de mi- nimización de la energía interna de deformación del elemento. Estos criterios de minimización de la ener- gía interna del elemento sirven para construir un ele- mento bien planteado. Para no tener que imponer las consideraciones de minimización de la energía sobre el problema global, el autor desarrolla un conjunto de problemas reducidos que son lo suﬁcientemente re- presentativos y generales. Concretamente se deﬁnen 3 problemas reducidos para los 3 modos de deformación de la lámina: deformación de ﬂexión, deformación de membrana (extensión en el plano) y deformación de cortante en el plano (rotación de taladro). Built in this manner, the only degrees of free- dom of the BEST element are the vertices’ coor- dinates (9 variables). The remaining 21 parame- ters are solved internally. In order to ﬁx the val- ues of these 21 internal parameters, each BEST element solves 9 systems of linear equations of rank 3. El elemento BEST construido de esta manera mantiene únicamente como incógnitas las posiciones de los vértices (9 variables). Los 21 parámetros res- tantes se resuelven internamente. La determinación de estos 21 parámetros internos del elemento implica la resolución de 9 sistemas de ecuaciones lineales de ran- go 3 para cada elemento BEST. The BEST element is successfully applied to the analysis of thin shells in linear and geo- metrically non-linear regimes using an implicit method. The non-linearity is solved using a Total Lagrangian formulation. Se ha conseguido aplicar con éxito el elemento BEST al cálculo de láminas delgadas en régimen li- neal y geométricamente no-lineal con un método im- plícito. La no-linealidad se plantea con una formula- ción Lagrangiana total. The author shows how to perform a de- composition of the description of the deforma- tion (including all the higher order terms) and the matrices of change of coordinates; grant- ing a computationally efﬁcient and geometrically El autor demuestra cómo descomponer los térmi- nos de la descripción de la deformación y de las ma- iii trices de cambio de coordenadas; permitiendo así una pre-integración en el espesor computacionalmente eﬁ- ciente y geométricamente precisa. Agradecimientos Quiero agradecer en primer lugar a mis profesores Eugenio Oñate y Juan Miquel el que me desvelaran el mundo de la investigación y me abrieran sus puertas invitándome a perseguir mis estudios de doctorado. De no ser por ellos y por su oferta, mi vida hubiera seguido un rumbo muy distinto. Tengo que decir que la vida de investigador me atraía de pequeño, pero nunca había soñado con que me la ofrecieran con tanta generosidad. Quiero dedicar un agradecimiento aparte y muy especial a mi tutor y mentor Don Eugenio Oñate. Eugenio, como familiarmente le llamamos en CIMNE, ha mantenido una conﬁanza en mis capacidades que según como se mire, rayaría la temeridad. Su paciencia para conmigo, en lo que a la elaboración de mi tesis se reﬁere, no ha tenido límites; y ha sabido en- tender en todo momento cómo manifestarme ese estímulo que me pudiera faltar para reemprender y proseguir el esfuerzo. Además, su ejemplo de dedicación y trabajo generoso ha sido una de las mejores enseñanzas que he recibido en estos años de trabajo juntos. A MIS PADRES vaya un GRACIAS en mayúsculas por haber querido compartir conmigo la aventura que decidí emprender al tomar la de- cisión de completar mis estudios de doctorado. Ellos han sido al tiempo, y durante muchos años: un soporte incondicional, la infraestructura de mi actividad (permitiendo multiplicar mi rendimiento), y los interlocutores ideales con quien reﬂexionar y compartir mis logros y preocupaciones. Todo esto no es en modo alguno una tarea sencilla y ellos la han resuelto con la mejor nota posible. A A Julio García, quien ha ejercido de tutor de mi tesis en lo que se re- ﬁere a la parte naval y todo aquello relacionado con la mecánica de ﬂuidos, quiero reconocer también la inﬁnita paciencia que ha tenido con mis tími- dos avances cuando mi dedicación era mayoritariamente a tareas ajenas a mi doctorado. Sus aportaciones e ideas han sido valiosísimas para hacer que mi trabajo tuviera mayor rendimiento; y he tratado humildemente de desarrollarlas en la medida de mis posibilidades. La mayor parte de las aportaciones que esta tesis doctoral incluye en lo relacionado con la mecánica de ﬂuidos y la interacción ﬂuido-estructura, cabe atribuírselas a él. ABSTRACT Solo es preciso eva- luar las integrales a través del espesor una vez: en la conﬁguración de referencia. Las integrales en el espe- sor se han reducido a 14 integrales escalares distintas para cada punto de Gauss. accurate through-the-thickness pre-integration. The through-the-thickness integrals are evalu- ated just once: at the reference conﬁguration. There are just 14 through-the-thickness scalar integrals to perform for each Gauss point. trices de cambio de coordenadas; permitiendo así una pre-integración en el espesor computacionalmente eﬁ- ciente y geométricamente precisa. Solo es preciso eva- luar las integrales a través del espesor una vez: en la conﬁguración de referencia. Las integrales en el espe- sor se han reducido a 14 integrales escalares distintas para cada punto de Gauss. The numerical examples results show that the BEST element has the potential to achieve cu- bic convergence. Although they also cast doubts on the possibility of reproducing this result for a wide range of problems. For in-plane shear domi- nated problems, the formulation used in this the- sis only achieves linear convergence. For mem- brane oriented tests with curvature, the conver- gence is quadratic. Los resultados de los ejemplos numéricos mues- tran que el elemento BEST tiene potencial para alcan- zar convergencia cúbica. Pero al mismo tiempo tam- bién existen dudas sobre la posibilidad de reproducir de manera consistente este resultado para un amplio rango de problemas. Para problemas dominados por la deformación de cortante en el plano, la formula- ción utilizada en esta tesis solo alcanza convergencia lineal. Para ejemplos orientados a la deformación de membrana que incluyen curvatura, la convergencia es cuadrática. The BEST element exhibits membrane lock- ing behavior. The author suggests exploiting fur- ther the drilling rotations kinematics in order to solve membrane locking. El elemento BEST presenta problemas de bloqueo por membrana. El autor sugiere desarrollar más pro- fundamente la cinemática de las rotaciones de taladro para resolver el bloqueo por membrana. v v Agradecimientos También quiero reservar un espacio en este apartado de agradecimien- tos a los doctores Ramón Ribó y [de nuevo a] Julio García por poner a mi disposición las herramientas que habían elaborado ellos previamente: RAMSeries y Tdyn; y en el marco de las cuales yo he elaborado mis partes de programa que implementan los desarrollos teóricos expuestos en la tesis. Este agradecimiento es por haber compartido los mismos ide- vii viii AGRADECIMIENTOS ales que yo considero necesarios para revolucionar el conocimiento; y que pasan necesariamente por la colaboración entre todos aquellos actores que pueden aportar su trabajo y su conocimiento. Quiero manifestar mi agradecimiento a Carles Estruch por su impli- cación en el trabajo sobre estimación de las normales en los nodos de las mallas de triángulos planos. Ha sido un privilegio contar con su colab- oración y rigor para la realización de las varias decenas de millones de casos que hubo que evaluar para demostrar de manera convincente una de las principales aportaciones de esta tesis. Pero su participación no fue meramente de ejecutor. La búsqueda de la excelencia de Carles queda perfectamente reﬂejada en la presentación de resultados del artículo que publicamos conjuntamente y que se incluye como anejo de esta tesis. Finalmente quiero agradecer a todo el personal de CIMNE, profesores adscritos del Departamento de Resistencia de Materiales y Estructuras en Ingeniería de la UPC, personal de administración y compañeros de estu- dios por la camaradería mostrada en todo momento. La experiencia de CIMNE fue extraordinaria desde el primer momento en que me invitaron y todavía maniﬁesto esa sensación del primer día en que si te lo propones puedes conseguir tus objetivos. Muy especialmente quiero manifestar mi agradecimiento a: • Pooyan Dadvand por introducirme al mundo de la programación del Método de los elementos Finitos. Junto a Pooyan, empecé mis primeros meses en CIMNE ideando y programando el entorno de de- sarrollo de los solvers de KRATOS, la obra maestra de Pooyan. • Miquel Pasenau por aguantar todas mis preguntas de programación en C++. Si bien no fue el único, Miquel me facilitó muchísimo la curva de aprendizaje del lenguaje de programación en el que he implemen- tado los desarrollos teóricos y algoritmos expuestos en esta tesis. • Enrique Escolano y Aleix Valls por compartir conmigo los análisis a fondo en innumerables descansos vespertinos de mis teorías ge- ométricas y matemáticas. Contents Abstract i Agradecimientos vii List of Figures xv List of Tables xxi Acronyms xxiii Notation xxv Foreword xxix 1 Motivation 1 1.1 The importance of shell analysis for sailboats . . . . . . . . 1 1.2 The hull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Spars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Sails . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.5 Shells vs Membranes . . . . . . . . . . . . . . . . . . . . . . 4 1.5.1 The shell-like nature of sails . . . . . . . . . . . . . . 5 1.5.2 Spinnakers: a different case . . . . . . . . . . . . . . 6 1.6 Justiﬁcation of the use of high technology in sailing . . . . 7 1.6.1 The competition rules between boats of diferent kinds 7 2 Objectives 11 2.1 Narrowing down the objective . . . . . . . . . . . . . . . . . 13 2.2 Rotation-Free Thin Shell Element . . . . . . . . . . . . . . 15 2.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3 State of the Art 19 3.1 Academic approaches to computer simulation of sails . . . 19 3.2 Thin shell ﬁnite elements . . . . . . . . . . . . . . . . . . . 21 3.2.1 Mechanic response of a shell . . . . . . . . . . . . . . 21 3.2.2 A glimpse on the development of thin shell ﬁnite elements . . . . . . . . . . . . . Agradecimientos • El equipo de desarrollo de GiD por poner a nuestra disposición una muy potente herramienta que nos facilita desarrollar día a día nuevos métodos numéricos y algoritmos cada vez más potentes. • Javier Mora por saber alentarme en los momentos de frustración y hacerme ver el mérito en aquello que a veces he considerado acceso- rio. Suele decirse que los individuos incorporamos a nuestra personalidad facetas o rasgos de aquellos otros con quienes interactuamos. Yo me siento orgulloso de haber interaccionado con todos ellos y sería un honor identi- ﬁcar en mí aquello que he valorado en ellos. Contents . . . . . . . . . . . . 22 4 Ideas for a new rotation-free shell element 27 4.1 Design requirements for the new shell element . . . . . . . 28 4.1.1 Going rotation-free . . . . . . . . . . . . . . . . . . . 28 ix ix CONTENTS x 4.1.2 Implicit vs Explicit time integration . . . . . . . . . 32 4.1.3 Advantages of the total Lagrangian formulation . . 33 4.2 Computing curvatures without second derivatives . . . . . 33 4.3 Raw approach and why it doesn’t work . . . . . . . . . . . . 35 4.3.1 A historical remark . . . . . . . . . . . . . . . . . . . 38 4.4 Attempts to add energy modes and stabilize the element . 39 4.5 Evolving to a higher order description of the triangle . . . 40 4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 5 Using Bézier triangles 43 5.1 Selecting the normals . . . . . . . . . . . . . . . . . . . . . 44 5.2 Shape Functions . . . . . . . . . . . . . . . . . . . . . . . . . 46 5.3 Geometric construction of the Bézier triangle . . . . . . . . 49 5.4 Determining shape through energy minimization . . . . . 54 5.4.1 Reduction of the problem . . . . . . . . . . . . . . . . 56 5.4.2 Location of the central control point . . . . . . . . . 60 5.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 6 Implementation of a continuum-based formulation for the Bézier-enhanced shell triangle 65 6.1 Thickness change due to the Poisson effect and the mild taper assumption . . . . . . . . . . Contents . . . . . . . . . . . . . . 66 6.2 Deﬁnition of local axes . . . . . . . . . . . . . . . . . . . . . 67 6.3 Description of the deformation . . . . . . . . . . . . . . . . 69 6.4 Decomposition into thickness-independent tensors . . . . . 71 6.5 Principle of virtual work . . . . . . . . . . . . . . . . . . . . 73 6.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 7 Construction of the tangent stiffness matrix 77 7.1 Material and geometric stiffness matrices . . . . . . . . . . 78 7.1.1 Derivatives of the control points’ coordinates . . . . 80 7.2 Through-the-thickness pre-integration . . . . . . . . . . . . 81 7.3 Tangent loads matrix . . . . . . . . . . . . . . . . . . . . . . 85 7.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 8 On the integration order of the element 89 8.1 Bending dominant cases . . . . . . . . . . . . . . . . . . . . 90 8.1.1 Slender beam . . . . . . . . . . . . . . . . . . . . . . 90 8.1.2 Hemisphere with point loads . . . . . . . . . . . . . 90 8.2 Membrane dominant cases . . . . . . . . . . . . . . . . . . . 92 8.2.1 Cylinder subject to internal pressure . . . . . . . . . 92 8.2.2 Parabolic roof . . . . . . . . . . . . . . . . . . . . . . 93 8.3 In-plane shear dominant cases . . . . . CONTENTS xi 9 Membrane locking of thin shells: a study on how this affects the BEST element and how to solve it 101 9.1 On membrane locking of thin shells . . . . . . . . . . . . . 101 9.2 Improving the BEST element kinematics by emulation of drilling rotations . . . . . . . . . . . . . . . . . . . . . . . . 102 9.2.1 Abanico analogy . . . . . . . . . . . . . . . . . . . . . 102 9.2.2 Modiﬁed kinematics using the abanico analogy . . . 104 9.2.3 This is not an incompatible mode method . . . . . . 106 9.3 Energy minimization for the in-plane shear deformation mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 9.3.1 Reduction of the problem . . . . . . . . . . . . . . . . 108 9.4 Combining the Ψϕ and Ψθ expressions . . . . . . . . . . . . 109 9.4.1 Solution of the indeterminate limit . . . . . . . . . . 111 9.5 Fine-tuning the kinematics associated to the drilling rota- tions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 9.5.1 A temporary ﬁx . . . . . . . . . . . . . . . . . . . . . 118 9.5.2 The real deal . . . . . . . . . . . . . . . . . . . . . . . 119 9.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 10 How to apply Dirichlet boundary conditions on rota- tions and their application to kinking and branching conﬁgurations 125 10.1 Fully clamped boundary condition . . . . . . . . . . . . . . CONTENTS 126 10.2 Continuous simply supported shell in the tangent plane direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 10.3 Boundary clamped in the perpendicular direction . . . . . 128 10.4 Continuous simply supported shell in the normal direction 130 10.5 Symmetry boundary condition . . . . . . . . . . . . . . . . . 132 10.6 Hinge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 10.7 Kinked shell . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 10.8 Branching conﬁgurations . . . . . . . . . . . . . . . . . . . . 139 10.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 11 Numerical examples 143 11.1 In-plane shear oriented examples . . . . . . . . . . . . . . . 143 11.1.1 Thick beam . . . . . . . . . . . . . . . . . . . . . . . . 143 11.1.2 Torsion of tube . . . . . . . . . . . . . . . . . . . . . . 146 11.2 Membrane oriented tests . . . . . . . . . . . . . . . . . . . . 150 11.2.1 Cylinder under internal pressure . . . . . . . . . . 150 11.2.2 Parabolic roof . . . . . . . . . . . . . . . . . . . . . . 151 11.3 Bending oriented tests . . . . . . . . . . . . . . . . . . . . . 154 11.3.1 Slender beam . . . . . . . . . . . . . . . . . . Contents . . . . . . . . . . . 94 8.3.1 Thick beam . . . . . . . . . . . . . . . . . . . . . . . . 94 8.3.2 Cylinder under torsion . . . . . . . . . . . . . . . . . 95 8.4 Effect of the mesh . . . . . . . . . . . . . . . . . . . . . . . . 95 8.4.1 Structured mesh with symmetrical triangles . . . . 96 8.4.2 Structured mesh with non-symmetrical triangles . 96 8.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 CONTENTS CONTENTS . . . . 155 11.3.2 Circular plate . . . . . . . . . . . . . . . . . . . . . . 160 11.4 Shell Obstacle Course . . . . . . . . . . . . . . . . . . . . . . 164 11.4.1 Scordelis-Lo Roof . . . . . . . . . . . . . . . . . . . . 165 11.4.2 Pinched Cylinder . . . . . . . . . . . . . . . . . . . . 167 11.4.3 Pinched Sphere / Hemispherical Shell . . . . . . . . 169 11.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 12 Future work and further research 175 CONTENTS xii 12.1 Optimize the programming . . . . . . . . . . . . . . . . . . 176 12.2 Nodal vs elemental assembly: edge-based assembly? . . . . 176 12.3 Solve the membrane locking . . . . . . . . . . . . . . . . . . 177 12.4 Implement Dirichlet boundary conditions . . . . . . . . . . 178 12.5 Develop adequate representation techniques for the BEST element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 12.6 Try a different curvature operator . . . . . . . . . . . . . . 179 12.7 What about composite and layered shells? . . . . . . . . . . 181 12.8 Practical applications . . . . . . . . . . . . . . . . . . . . . . 182 13 Conclusions 185 13.1 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . 185 13.1.1 A new thin shell ﬁnite element . . . . . . . . . . . . CONTENTS 185 13.1.2 Optimal estimation of the surface normal from a faceted mesh . . . . . . . . . . . . . . . . . . . . . . . 187 13.1.3 New paradigm for reconstructing a cubic shell el- ement using the information from the neighboring elements . . . . . . . . . . . . . . . . . . . . . . . . . 188 13.1.4 Successful element construction based on energy minimization principles . . . . . . . . . . . . . . . . 189 13.1.5 Cubic convergence using linear information . . . . . 191 13.1.6 Efﬁcient and geometrically accurate through-the- thickness pre-integration . . . . . . . . . . . . . . . 192 13.2 Lessons learned . . . . . . . . . . . . . . . . . . . . . . . . . 195 13.2.1 The quantum nature of research work . . . . . . . . 195 13.2.2 I stand up for libraries . . . . . . . . . . . . . . . . . 195 13.2.3 Goals vs Objectives . . . . . . . . . . . . . . . . . . . 197 13.2.4 Non-linearity is a lifeline, not a hurdle . . . . . . . . 198 A The use of numerical methods in sailboats design 199 A.1 The America’s Cup as a sports and technological pinnacle . 200 A.2 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 B Published paper: On the interpolation of normal vectors for triangle meshes 209 C Minimization of the membrane (axial) energy of a 2D cu- bic Bernouilli beam 233 D Analytic derivatives of some complex expressions 235 D.1 Derivatives of the element normal . . . . . . . . . . . . . . 235 D.2 Derivatives of the system matrix and independent vector to compute the control point locations . . . . . . . . . . . . 237 D.2.1 Boundary control points . . . . . . CONTENTS xiii 249 CONTENTS . . . . . . . . . . 237 D.2.2 Central control point candidates . . . . . . . . . . . 239 D.3 Derivatives of the normal at the node . . . . . . . . . . . . 240 D.4 Derivatives of dij . . . . . . . . . . . . . . . . . . . . . . . . 244 D.5 Derivatives of θij . . . . . . . . . . . . . . . . . . . . . . . . 245 D.6 Derivatives of ϕij . . . . . . . . . . . . . . . . . . . . . . . . 247 List of Figures 4.1 Representation of the global and parametric coordinates used to deﬁne the positions of each of the nodes of the tri- angle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 4.2 In order to compute the curvature inside the central trian- gle, we use the information of the patch of elements that surround each node. The shading patterns indicate which triangles contribute to the determination of each normal. . 36 4.3 Numerical example showing the displacements along the diagonals of a simply supported square with central point load. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.4 Figure depicting the concept of the antimetric bending en- ergy mode to stabilize the element. . . . . . . . . . . . . . . 40 4.5 Figure depicting the concept for gently forcing the normals toward the perpendicular of the element. . . . . . . . . . . . 40 5.1 Example of a mesh approaching a surface S in the vicinity of a node i. The normal at node i is estimated using the normals at each of the surrounding triangles. ri is the total number of triangles surrounding node i. . . . . . . . . . . . 45 5.2 Left: barycentric representation of the Bézier triangle along with its shape functions corresponding to each of the control points of the net. Right: equivalent representation using parametric coordinates. . . . . . . . . . . . . . . . . . . . . . 47 5.3 Representation of the shape functions corresponding to 5 different control points. . . . . . . . . . . . . . . . . . . . . . 48 5.4 The geometry of the cubic Bézier triangle is determined by the position of the 10 control points. Notation used to iden- tify each of the control points in a triangle. . . . . . . . . . . List of Figures 50 5.5 Representation of the net of control points for a cubic Bézier triangle constructed using the nodal positions and normals. Three planes deﬁne the position of a control point of the contour. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 5.6 Two adjacent Bézier triangles are G1 continuous if all their adjacent control triangles are co-planar. . . . . . . . . . . . . 53 5.7 Energy minimization. Representation of the 2D problem simpliﬁcation. . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 5.8 Regions of start and endpoint angles of a cubic spline for which spirals are generated. . . . . . . . . . . . . . . . . . . . 58 xv xvi LIST OF FIGURES 5.9 Graphical solution of equation (5.25). Comparison with the solution obtained using equation (5.21). . . . . . . . . . . . . 61 5.10 Example of the enhancement provided by the procedure presented in this chapter to model the geometry obtained from a mesh of linear triangles. The BEST mesh (c) has the same degrees of freedom as the linear mesh (b). . . . . . . . 62 6.1 Representation of the curved shell element identifying the curvilinear coordinates (ξ, η, ζ) and the effect of the mild taper assumption. The reference midsurface is identiﬁed. . 66 6.2 A set of orthogonal local axes (t1, t2, t3) is deﬁned in the reference conﬁguration in order to establish the principal directions of the material over the curved surface of the el- ement. The user deﬁnes the material local axes (e′ 1, e′ 2, ˆy) on the ﬂat triangle geometry. The vector t3 is deﬁned in the same way as n in the current conﬁguration. . . . . . . . . . 68 8.1 Comparison of h-convergence using structured meshes and different numerical quadratures for a simply supported beam. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of Figures 91 8.2 Comparison of h-convergence using structured meshes and different numerical quadratures for a pinched hemisphere. 91 8.3 Comparison of h-convergence using unstructured meshes and different numerical quadratures for a pinched hemi- sphere. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 8.4 Comparison of h-convergence using non-symmetric struc- tured meshes and different numerical quadratures for a thin-walled cylinder. . . . . . . . . . . . . . . . . . . . . . . . 92 8.5 Comparison of h-convergence using non-symmetric struc- tured meshes and different numerical quadratures for a pa- rabolic roof. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 8.6 Comparison of h-convergence using unstructured meshes and different numerical quadratures for a parabolic roof. . 93 8.7 Comparison of h-convergence using structured meshes and different numerical quadratures for a thick beam. . . . . . 94 8.8 Comparison of h-convergence using structured meshes and different numerical quadratures for a cylinder under tor- sion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 8.9 Example of a cylinder meshed using a structured mesh of symmetric triangles. . . . . . . . . . . . . . . . . . . . . . . . 96 8.10 Qualitative comparison of the results obtained using struc- tured meshes of symmetric triangles and different numer- ical quadratures for a thin-walled cylinder. Displacements in the x-direction as viewed on the y-z plane. . . . . . . . . 97 8.11 Example of a cylinder meshed with a non-symmetrical structured triangles mesh. All the diagonals are oriented in the same direction. This is the mesh used to obtain the results shown in ﬁgure 8.12. . . . . . . . . . . . . . . . . . . 98 LIST OF FIGURES xvii 8.12 Comparison of the results obtained using different numer- ical quadratures and non-symmetrical structured triangle meshes for a thin-walled cylinder. List of Figures Displacements in the x- direction as viewed on the y-z plane. The mesh used is also displayed in ﬁgure 8.11 as a reference to indicate the effect of the skewed triangles. . . . . . . . . . . . . . . . . . . . . . 98 9.1 An abanico made in the XIX century. Painted ivory with a Cupid on the frame and with delicate chantilly lace. From the collection donated by Gloria Trueba Gómez in 1997 to the city of Seville and exhibited at the Abanicos’ Room of the Reales Alcázares of Seville (Spain). . . . . . . . . . . . . 102 9.2 Scheme depicting the abanico analogy. . . . . . . . . . . . . . 103 9.3 Representation of the net of control points (dashed lines) for a cubic Bézier triangle constructed using the nodal positions and normals. This ﬁgure differs from ﬁgure 5.5 in that the drilling rotations θij are used here. . . . . . . . . . . . . . . . 105 9.4 Two adjacent Bézier triangles are G1 continuous if all their adjacent control triangles are co-planar. This construction is similar to the one proposed in ﬁgure 5.6, but the present one is more general and can be used with non-ﬂat curved edges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 9.5 Problem reduction to minimize the in-plane shear deforma- tion energy. The author analyzes a ﬂat equilateral triangle subjected to a periodic torsion θ in all three corner nodes. . . 109 9.6 Comparison of the minimization solutions for the Ψij θ values for three different values of the Poisson coefﬁcient ν. The red curve represents an adjustment for small values of θ ∈ [−π 10, π 10]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 9.7 Comparison of the minimization functions Ψij ϕ (dashed) and Ψij θ (continuous) in the range of the variables ϕij, θij ∈ [0, π 10]. . . . . . . . . . . . . . List of Figures . . . . . . . . . . . . . . . . . . 111 9.8 Comparison of the variation of the total strain energy with the drilling rotation angle to verify that it follows a quadra- tic variation. The example is performed with a rectangle triangle subjected to an in-plane shear deformation in tor- sion mode as deﬁned in ﬁgure 9.5 (red line), and a simple quadratic function (blue line). . . . . . . . . . . . . . . . . . 112 9.9 Geometric interpretation of the change of variables applied in equations (9.12) and (9.13) at the limit when ρ2 →0. . . 113 9.10 Study of the effect of including drilling rotations on the thick beam example. . . . . . . . . . . . . . . . . . . . . . . . 117 9.11 Geometric interpretation of the deﬁnition of the drilling ro- tation provided by equation (9.6) and how it affects the ac- tual deformation of the edges D-E-F in the mesh. . . . . . . 118 9.12 Geometric interpretation of the deﬁnition of the drilling ro- tation provided initially by equation (9.6) in red, compared to the effect caused by the new deﬁnition of equation (9.33) in blue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 xviii LIST OF FIGURES 9.13 Scheme depicting the process to ﬁnd the value of θij using equation (9.35). . . . . . . . . . . . . . . . . . . . . . . . . . . 121 10.1 Figure showing the meaning of a fully clamped boundary condition as a combination of two different conditions. . . . 127 10.2 Figure showing the meaning of the boundary condition of a continuous simply supported shell edge in the tangent plane direction. . . . . . . . . . . . . . . . . . . . . . . . . . . 128 10.3 Figure showing the meaning of the boundary clamped in the perpendicular direction. A graphical interpretation of the meaning of equation (10.9) is also presented. List of Figures That is, the vector ni can rotate only about the vector t⊥. . . . . . . 129 10.4 Figure showing the meaning of the boundary condition of a continuous simply supported shell edge in the direction normal to the tangent plane. A graphical interpretation of the meaning of equation (10.13) is also presented. That is, the vector ni can rotate only about the vector t∥. . . . . . . . 131 10.5 Figure representing the conditions that conﬁgure a symme- try boundary condition. . . . . . . . . . . . . . . . . . . . . . 133 10.6 Whenever two surface entities meet at an angle, their nor- mals shall be distinguished at the nodes along the disconti- nuity line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 10.7 Representation of two curved elements meeting at a hinge line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 10.8 Representation of two curved elements meeting at a kink. . 138 11.1 Thick beam. Comparison of results with other elements from the literature. . . . . . . . . . . . . . . . . . . . . . . . . 144 11.2 Calculation meshes used for the thick cantilever beam ex- ample. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 11.3 Geometry and conditions of the problem. Elevation and lat- eral view. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 11.4 Torsion of tube. Comparison of results with 2 different val- ues for θij: using a symmetric deﬁnition of θij according to equation (9.33), or setting θij = 0. . . . . . . . . . . . . . . . 148 11.5 Torsion of tube. Results obtained using an unstructured mesh with 1268 nodes. List of Figures These results correspond to a value of θij = 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 11.6 Cylinder under internal pressure. Results obtained for two possible deﬁnitions of the drilling rotations: θij = 0 and θij = −θji. The reference solution for the variation of the diameter is 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 11.7 Transverse shrinkage due to the Poisson modulus. The ref- erence value of the displacement at the edges is 0.3. Result obtained using a mesh with 16 divisions around the perime- ter and 10 divisions along the lengh. . . . . . . . . . . . . . 151 11.8 Parabolic roof problem. . . . . . . . . . . . . . . . . . . . . . . 151 11.9 The example of the parabolic roof is inspired in catenary roofs built around the world. . . . . . . . . . . . . . . . . . . 152 11.10 Results of the parabolic roof example. . . . . . . . . . . . . . 153 LIST OF FIGURES xix 11.11 Effects of mesh bias on the results. As the mesh is reﬁned the effect of the mesh bias diminishes. Displacements in the direction of the supported edges of the parabolic roof example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 11.12 Geometry, boundary conditions and property materials for the slender beam problem. . . . . . . . . . . . . . . . . . . . . 155 11.13 Samples of meshes for the slender beam problem. . . . . . . 156 11.14 Slender beam. Convergence of the deﬂection of the free edge’s midpoint. Series of results of the linear problem. . . . 157 11.15 Detail of an anisotropic mesh using symmetric divisions. Divisions with a size ratio of 5 : 1. List of Figures The resulting triangles have circumscribed circles of very different sizes. The larger circle has a radius 4.5 times that of the smaller circle; and an area 20 times larger. . . . . . . . . . . . . . . . . . . . . . 157 11.16 Slender beam. Convergence of the deﬂection of the free edge’s midpoint. Series of results of the non-linear problem. 159 11.17 Comparison of the effect of the biased meshes on the result of the displacements in the Z-direction. . . . . . . . . . . . . 160 11.18 Geometry, boundary conditions and material properties of the circular plate problem. Plan and cross section. . . . . . 161 11.19 Sample mesh of the circular plate problem with 10 divisions in the radial direction. . . . . . . . . . . . . . . . . . . . . . . 161 11.20 Convergence of the deﬂection of the central point in the cir- cular plate problem. Linear solution. . . . . . . . . . . . . . . 162 11.21 Evolution of the convergence of the BEST element as the thickness decreases. Example of a simply supported circu- lar plate. Non-linear solution. . . . . . . . . . . . . . . . . . . 163 11.22 Convergence of the deﬂection of the central point in the cir- cular plate problem. Non-linear solution. . . . . . . . . . . . 164 11.23 Scordelis-Lo problem. . . . . . . . . . . . . . . . . . . . . . . . 165 11.24 Comparison of the convergence between different shell ele- ments for the Scordelis-Lo problem. . . . . . . . . . . . . . . 166 11.25 Field of vertical displacements of the Scordelis-Lo roof prob- lem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 11.26 Pinched cylinder problem. . . . . . . . . . . . . . . . . . . . . List of Figures 167 11.27 Comparison of the convergence between different shell ele- ments for the pinched cylinder problem. . . . . . . . . . . . . 168 11.28 Field of displacements (modulus) of the pinched cylinder problem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 11.29 Pinched sphere problem. . . . . . . . . . . . . . . . . . . . . . 170 11.30 Comparison of the convergence between different shell ele- ments for the hemispherical shell problem. . . . . . . . . . . 170 11.31 Field of vertical displacements of the hemispherical shell problem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 C.1 2D cubic Bézier curve representing the edge of a cubic Bézier triangle. . . . . . . . . . . . . . . . . . . . . . . . . . . 233 D.1 This ﬁgure is a modiﬁcation of ﬁgure 5.1. A new notation is used for the neighboring nodes. . . . . . . . . . . . . . . . . . 241 List of Tables 2.1 Summary of tools and technologies used in the industry to simulate sailboats. . . . . . . . . . . . . . . . . . . . . . . . . . 14 4.1 Main advantages and disadvantages of the different rotation- free ﬁnite element technologies in the literature. . . . . . . . 31 5.1 Pascal’s triangle of monomials for bi-variate cubic functions. 50 7.1 Weights and coordinates for fourth order integration using Gaussian quadrature over a triangle. . . . . . . . . . . . . . . 82 xxi LIST OF TABLES xxii ξ ﬁrst one of the parametric coordinates of the element. It can also be referred to as ξ1. xvi, 34, 35, 46–50, 66, 68–73, 78–81, 83–86, 235– 237 Acronyms BEST This is the shell triangle element developed in this PhD thesis. And stands for Bézier-enhanced shell triangle. i–iii, xvi, xix, 62, 63, 74, 75, 77, 87, 99, 101–104, 106, 116, 118–120, 123, 125, 140, 143, 145, 147, 150, 151, 154–156, 158–161, 163–165, 167–169, 171–173, 176–179, 183, 185–192 BPT basic plate triangle. 29 BPT basic plate triangle. 29 BSN basic shell node. 33, 41, 186 BST basic shell triangle. 28, 29, 33, 165, 186 BST basic shell triangle. 28, 29, 33, 165, 186 CAD computer aided design. 30, 31, 44 CAD computer aided design. 30, 31, 44 CAE computer aided engineering. 30 CAE computer aided engineering. 30 CAGD computer aided geometric design. 30 CAGD computer aided geometric design. 30 CIMNE International Center for Numerical Methods in Engineering us- ing the initials of the catalan name. xxix, 1, 4, 16, 204, 206 EBST enhanced basic shell triangle. 165, 168, 169, 197 FEM ﬁnite element method. 43, 47, 57 FEM ﬁnite element method. 43, 47, 57 IGA Isogeometric Analysis. 31, 43, 44, 62 NURBS non-uniform rational B-Spline. 30, 31, 43, 44 xxiii ϕ function that deﬁnes a surface in R3. Not to be confused with ϕ. 34, see ϕ Notation Ψ fraction of the cordal lenght of an element’s edge used to position the boundary control points minimizing the deformation energy of the element. xvii, 51, 52, 55–61, 105–114, 116, 135–137, 139, 156, 159, 162, 172, 189, 190, 234, 238, 239 δ Kronecker delta. Its value is 1 if the indices are repeated. 0 in any other case. The indices can be subscripts or superscripts. Not to be confused with the symbol δ used to identify the variational of a magnitude. 72, 79, 80, 83, 85, 235, 236, 238, 242–247 κ curvature tensor. Deﬁned in terms of differential geometry of surfaces. See references [67, 122]. 34 λ thickness stretch factor of the shell considering the Poisson effect of the material. A subscript can refer to a speciﬁc layer. 67, 69, 70, 73, 78, 79, 83, 84 ν Poisson coefﬁcient. xvii, 108, 110 θ angle of drilling rotation of a node. This angle is emulated and not in- cluded as a variable of the system. The rotation is measured with respect a given edge of the mesh. The ﬁrst superindex indicates the node whose drilling rotation is being measured. The second su- perindex indicates which of all the edges converging to that node is taken as a reference. For example, if the superindices are θij, then it refers to the angle about the normal at i (ni) and with respect to the edge xi −xj. xvii, xviii, 104–116, 118–121, 127, 135–137, 139, 140, 143, 147, 149–151, 154, 156, 159, 162, 164, 165, 171, 172, 244, 245 ϕ angle of inclination of the normal vector at the node with respect to the normal of the ﬂat triangle. The angle is measured between the normal at the node and the plane perpendicular to the edge being considered. For example, if the superindices are ϕij, then it refers to the angle between ni and the plane perpendicular to the edge xi−xj. xvii, 55–57, 59–61, 105, 107, 110–116, 135–137, 139, 156, 162, 234, 247 ϕ function that deﬁnes a surface in R3. Not to be confused with ϕ. 34, see ϕ ϕ function that deﬁnes a surface in R3. Not to be confused with ϕ. 34, see xxv xxvi NOTATION NOTATION η second one of the parametric coordinates of the element. It can also be referred to as ξ2. Notation xvi, 34, 35, 46–50, 66, 68–73, 78–80, 83–86, 235–237 ζ third one of the parametric coordinates of the element. It can also be referred to as ξ3. xvi, 34, 35, 47, 66, 69–73, 78, 79, 82–86, 194 A matrix of change of coordinates from the local axes in the reference conﬁguration to the parametric coordinates of the element. 70–72, 85, 194 B matrix containing the partial derivatives of g with respect to the coor- dinates of the nodes that deﬁne the shape of the element. 74, 75, 78, 79, 83, 85, 127, 155, 156 C Cauchy-Green deformation tensor. 35, 37, 69, 70 C0 mathematic classiﬁcation of functions which means that the functions belonging to this group are continuous in a given domain. i, 23, 51, 105, 134, 137 C1 mathematic classiﬁcation of functions which means that the functions belonging to this group are continuous and their ﬁrst derivatives are also continuous in a given domain. i, ii, 23, 25–29, 31, 33, 52–54, 63, 107, 188, 189 D fourth order constitutive tensor, or its corresponding second order ten- sor in Voigt notation taking advantage of the major and minor sym- metries. 35, 37, 38, 67, 70, 74, 75, 78, 83–85 E Green-Lagrange strain tensor, or its corresponding vector in Voigt no- tation. 35, 37, 38, 67, 69, 70, 72–74, 78, 120–123 F deformation gradient tensor. 35, 69, 70, 122, 123 G inverse of the jacobian tensor in the reference conﬁguration (J−1). 35, 37, 70, 71, see J NOTATION h thickness of the shell in the reference conﬁguration. A subscript can refer to a speciﬁc layer. 67, 69–73, 78, 79, 84, 86 I identity matrix of the convenient dimension, or its corresponding vector in Voigt notation. 35, 37, 38, 68–70, 72–75, 78, 84, 104, 106, 120–123, 129–132, 136, 139 I identity matrix of the convenient dimension, or its corresponding vector in Voigt notation. 35, 37, 38, 68–70, 72–75, 78, 84, 104, 106, 120–123, 129–132, 136, 139 J jacobian tensor of the change of coordinates function from parametric coordinates to global coordinates in the reference conﬁguration. xxvi, xxvii, 35, 69–71, 84 J determinant of the reference jacobian tensor (J). 71, 78, 84–86 j jacobian tensor of the change of coordinates function from parametric coordinates to global coordinates in the current conﬁguration. 35, 69, 70 K tangent stiffness matrix of the element. A subscript may specify whether it corresponds to the material stiffness matrix (KM), or to the geometric stiffness matrix (KG). 78, 83 L derivative of the shape function (N) corresponding to the node of the element indicated by the superscript. It is derived with respect to the coordinate indicated in the subscript. 35, 48, 49, 69–73, 84, 235– 237 N normal vector to the surface of the shell in the reference conﬁguration. 49, 69, 71, 73, 84, 126, 128–132 N shape function corresponding to the node of the element indicated by the superscript. xxvii, 34, 35, 46–48, 69, 81, 86 n normal vector to a surface (in general refers to the current conﬁgura- tion). xvi, xviii, xxv, 34, 35, 45, 46, 49, 51, 54, 66, 68–70, 73, 80, 86, 104–107, 113, 120, 126–132, 134–140, 235, 236, 238, 240, 246 P tensor containing the coordinates of the 10 control points of the cubic Bézier triangle in the reference conﬁguration. 49, 69, 71–73, 84 p tensor containing the coordinates of the 10 control points of the cubic Bézier triangle. The tensor is arranged as an array of 10 column vectors. Each column contains the 3 coordinates of each control point. When affected by a superindex, it refers to the coordinates of the indicated control point. 47, 49, 51, 53, 54, 69, 70, 73, 78–81, 86, 106, 136, 137, 235–237, 239 Q matrix of change of coordinates, for tensors expressed as vectors in Voigt notation, from the parametric coordinates of the element to the local axes in the reference conﬁguration. S second Piola-Kirchhoff stress tensor, or its corresponding vector in Voigt notation. 35, 37, 38, 70, 73, 74 G Shear modulus. 146 g pseudo deformation tensor. This tensor contains the deformation met- rics of the current (deformed) conﬁguration. It cannot be named a deformation tensor because it lacks the information of the reference conﬁguration. xxvi, 70, 72–75, 78, 79, 83, 85, 194 G1 mathematic classiﬁcation of functions which means that the functions belonging to this group are geometrically continuous and their ﬁrst derivatives are also geometrically continuous in a given domain. xv, xvii, 23, 44, 52, 53, 107, 136, 137 H matrix containing the linearized external loads. A subscript may spec- ify whether it corresponds to the loads deﬁned over the volume in the reference conﬁguration (HV 0), or the loads deﬁned over the sur- face in the reference or current conﬁgurations (HS0 and HS, respec- tively), or the loads deﬁned over the boundary in the reference or current conﬁgurations (HΓ0 and HΓ, respectively). 85, 86 xxvii x position vector of a material point in the current conﬁguration. When affected by a tilde sign (˜), it indicates a collection of position vectors for different material points. In this case, the author refers to each coordinate of the composite vector using a superindex to refer to the material point, and a subindex to refer to each of the three coordi- nates. xxv, 34, 35, 38, 47, 49, 51, 54, 66, 69, 70, 74, 75, 77–81, 83, 85, 86, 104, 106, 113, 114, 120, 121, 123, 126–129, 131, 132, 136–139, 155, 156, 238, 242–244, 247 Foreword D ESDE HACE MÁS DE 25 AÑOS el Centro Inter- nacional de Métodos Numéricos en Ingeniería (CIMNE)1 ha incorporado en sus líneas de investi- gación el interés por el mundo naval; y en particu- lar, por los barcos de vela. Por CIMNE han pasado una cantidad importante de estudiantes y cientíﬁ- cos de prestigio con una vocación especial en este ámbito (ingenieros navales, regatistas, capitanes de yate, etc.) y ello ha dejado una notable impronta en CIMNE. Toda esta gente se acercó a CIMNE con la visión de desarrollar técnicas numéricas que facili- taran la comprensión acerca del funcionamiento de los barcos de vela. F OR OVER 25 YEARS the International Cen- ter for Numerical Methods in Engineering (Centre Internacional de Mètodes Numèrics en Enginyeria (CIMNE))1 has grown a genuine in- terest for naval issues and sailboats in partic- ular. Many students and world-renowned re- searchers fond of the topic (naval engineers, sailors, captains, etc.) have worked and studied at CIMNE. This has left a mark in the institu- tion. All these people approached CIMNE with the vision of developing the numerical tech- niques that could allow a better understanding on the functioning of sailboats. D F Como fruto de este historial CIMNE es hoy en día uno de los centros de investigación en métodos numéricos con una mejor preparación y mayor grado de conocimiento de la problemática de los barcos de vela de cuantos existen en el mundo. No es arriesgado aﬁrmar que, como máquina, un barco de vela tiene uno de los funcionamientos más complejos de cuantas ha concebido el hombre a lo largo de la historia. Esto es debido al altísimo gra- do de interacción que tienen los distintos elementos que la componen y de lo sensible que es el rendi- miento del barco a dicha interacción. A su vez, todos estos componentes son de muy distinta naturaleza; y cada uno de ellos está sometido a gran disparidad de regímenes en la casuística de navegación de un barco a vela. Resulting from this historical background, nowadays CIMNE has become one of the re- search centers on numerical methods better prepared and with a greater knowledge on the issues affecting sailboats in the world. It isn’t bold to say that a sailboat encloses one of the trickiest functionings amongst all the ma- chines conceived by mankind in history. 1The acronym corresponds to the original name in cata- lan ”Centre Internacional de Mètodes Numèrics en Enginy- eria”. www.cimne.com 1Las siglas corresponden al nombre original en catalán “Centre Internacional de Mètodes Numèrics en Enginyeria”. www.cimne.com NOTATION 72–75, 78, 84, 194 R residual vector resulting from the difference between the external forces and the internal forces of the structure. 75, 77 xxviii xxviii NOTATION NOTATION T matrix of change of coordinates from local axis in the reference conﬁgu- ration to global axis. 68, 70, 71 T matrix of change of coordinates from local axis in the reference conﬁgu- ration to global axis. 68, 70, 71 T matrix of change of coordinates from local axis in the reference conﬁgu- ration to global axis. 68, 70, 71 X position vector of a material point in the reference conﬁguration. 34, 35, 49, 68, 69, 123, 128 X position vector of a material point in the reference conﬁguration. 34, 35, 49, 68, 69, 123, 128 x position vector of a material point in the current conﬁguration. When affected by a tilde sign (˜), it indicates a collection of position vectors for different material points. In this case, the author refers to each coordinate of the composite vector using a superindex to refer to the material point, and a subindex to refer to each of the three coordi- nates. xxv, 34, 35, 38, 47, 49, 51, 54, 66, 69, 70, 74, 75, 77–81, 83, 85, 86, 104, 106, 113, 114, 120, 121, 123, 126–129, 131, 132, 136–139, 155, 156, 238, 242–244, 247 x position vector of a material point in the current conﬁguration. When affected by a tilde sign (˜), it indicates a collection of position vectors for different material points. In this case, the author refers to each coordinate of the composite vector using a superindex to refer to the material point, and a subindex to refer to each of the three coordi- nates. xxv, 34, 35, 38, 47, 49, 51, 54, 66, 69, 70, 74, 75, 77–81, 83, 85, 86, 104, 106, 113, 114, 120, 121, 123, 126–129, 131, 132, 136–139, 155, 156, 238, 242–244, 247 Foreword This is because the very tight degree of interaction that exists between the different elements that make a sailboat and the high sensibility of the performance of the boat to this interaction. At the same time all these elements are of very different nature and each of them is subject to a wide range of working regimes depending on the sailing scenarios of a sailboat. I can’t think of another research center bet- ter suited in which to develop this PhD the- sis because the topic ﬁts perfectly within the goals pursued by CIMNE and in various of its most prominent research lines. For example, I mention a few: the research line on Multi- physics Simulation, the research line on Struc- No se me ocurre otro centro de investigación me- jor indicado en el que desarrollar mi tesis doctoral, pues el tema se enmarca plenamente en los objetivos perseguidos por CIMNE y en varias de sus líneas de investigación más destacadas. Se nombran a modo de ejemplo: la línea de investigación en Simulación Multifísica, la línea de investigación en Análisis de xxix Motivation A L TERMINAR MIS ESTUDIOS DE INGENIERÍA en el año 2001 tuve la ocasión de combinar por primera vez mis conocimientos de cálculo de estructuras con mi gran aﬁción al deporte de la vela. El objeto de aquél estudio era explorar las opciones que ofrecía el análisis de estructuras de membrana para el cálculo de velas. En los meses previos a mi estudio, se había llevado a cabo en el Centro Internacional de Métodos Numéricos en Ingeniería (CIMNE) un proyecto relacionado con la fabricación de velas de barcos en el que éstas se habían estudiado como estructuras tipo lámina. El tema despertó mi interés de manera inmediata. En el apartado 1.4 resumo algunos aspectos de los inicios de mi investigación. W HEN I FINISHED MY ENGINEERING STUDIES in 2001 I had the opportunity to combine for the ﬁrst time my knowledge on structural analysis with my passion for the sport of sailing. The goal of that study was to explore the options of using membrane analysis for the calculation of sails. During the months that preceded my study, a project related to sails manufacturing had been conducted at the International Center for Numerical Methods in Engineering (CIMNE). For that project, the sails were studied as shells. The topic raised my interest immediately. In section 1.4 I sum- marize some of the aspects of the beginnings of my research. W A L After many considerations and after realiz- ing the validity of the problem still today in the scientiﬁc community and in the industry, I de- cided to relaunch the study of the analysis of sailboats. Después de numerosas reﬂexiones y tras consta- tar la vigencia que todavía tiene el problema en la comunidad cientíﬁca y en el ámbito industrial, deci- dí relanzar el estudio del análisis de barcos de vela. FOREWORD xxx tural Shell Analysis, the research line on Com- posite Materials and the research line on Fluid Mechanics. This thesis has the objective of de- veloping a New Rotation-Free Thin Shell Ele- ment. The advances achieved with this devel- opment shall enable signiﬁcant improvements in the computations performed nowadays and which are so necessaray to target in the end the full simulation, in three dimensions, and in transient dynamic regime, of a sailboat. That challenge is yet to be solved efﬁciently in or- der to make a leap in the improvement of the analysis and design tools for sailboats. Solv- ing this challenge will in turn enable substan- tial beneﬁts to the performance of sailboats as it will reduce the number of non-trivial simpli- ﬁcations made for their simulation and analy- sis. Chapter 1 includes a detailed explanation on why the calculation of shells is relevant for the analysis of sailboats. Láminas Estructurales, la línea de investigación en el Estudio de Materiales Compuestos y la línea de investigación en Mecánica de Fluidos. La presente tesis doctoral tiene por objetivo el desarrollo de un Nuevo Elemento Finito de Lámina Delgada Sin Ro- taciones. Los avances obtenidos con este desarrollo deben permitir signiﬁcativas mejoras en los cálcu- los que se realizan hoy en día; y son en buena parte necesarios para abordar en última instancia la si- mulación completa, en tres dimensiones, y en régi- men dinámico transitorio, de un barco de vela. Este es un desafío pendiente de resolver de manera eﬁ- ciente para que se puedan mejorar las herramien- tas de análisis y de diseño de barcos de vela; y sin embargo se considera que podrá aportar ventajas sustanciales al rendimiento de los barcos de vela por cuanto reduzca el número de simpliﬁcaciones no inocuas a la hora de simularlos. En el capítu- lo 1 se explica en detalle la importancia que tiene el cálculo de láminas para el análisis de barcos de vela. I hope that along the reading of this PhD thesis I can share with you the good expe- riences, ﬁndings and intelectual satisfactions that my dedication to it has provided me all the time. Confío que con la lectura de esta tesis doctoral puedan compartir las agradables experiencias, ha- llazgos, y satisfacciones intelectuales que mi dedi- cación a ella me han supuesto en todo momento. 1.2 The hull Si nos ﬁjamos en la estructura del casco de un barco de vela podemos observar que su tecnología de cons- trucción actual se diferencia bastante de la utiliza- da hace apenas medio siglo. Si en el pasado la es- tructura estaba compuesta principalmente por ele- mentos lineales (baos y cuadernas) entrecruzados para componer un emparrillado que actuase de so- porte de la piel exterior del casco, hoy en día la apli- cación de los plásticos, y más concretamente de los plásticos reforzados con ﬁbras (FRP según sus si- glas en inglés) en sándwich o en laminado han per- mitido combinar la función estructural con la de la piel del casco en un único elemento de característi- cas inherentemente bidimensionales. La curvatura con que se diseñan los cascos de barcos para opti- mizar su hidrodinámica así como la esbeltez que se requiere para optimizar el uso de los materiales con- ducen necesariamente a considerar el casco de un barco moderno como una lámina delgada. When looking at the structure of a modern sailboat, we can realize that the technology of its construction differs a lot from that used just half a century ago. While in the past the structure was composed mainly of linear elements (frames and planks) crisscrossed in order to conﬁgure a frame that could sup- port the external skin of the boat, nowadays the use of plastics and more speciﬁcally the ﬁber reinforced plastics (FRP) in sandwich or laminate have enabled combining the struc- tural and skin functions in a single element with the characteristic of being inherently bi- dimensional. The curvature used to design the boats in order to optimize the hydrodynamics as well as the slenderness required in order to optimize the use of materials, lead inevitably to consider the hull of a modern boat as a thin shell. CHAPTER 1. MOTIVATION 2 1.1 The importance of shell analysis for sailboats El estudio del comportamiento de un barco de vela requiere de la comprensión en detalle de su respues- ta ante las acciones y fuerzas a las que está some- tido. Para ello es preciso analizar la dinámica de los ﬂuidos en los que está inmerso el barco (aire y agua) al mismo tiempo que considerar el equilibrio estático y dinámico del conjunto estructural. En los siguientes apartados se describen cada uno de los principales elementos que conﬁguran la estructura de un barco de vela y el papel que en cada elemento juegan las estructuras de lámina. Studying the behavior of a sailboat requires de- tailed understanding of its response to the ac- tions and forces acting on it. Therefore it is necessary to analyze the dynamics of the ﬂuids in which the boat is immersed (air and water) and at the same time the static and dynamic equilibrium of the structural ensemble. In the following sections the main elements that con- form the structure of a sailboat are described. And the role played by shell structures in each of these elements is speciﬁed. 1 CHAPTER 1. MOTIVATION 1.4 Sails Para completar el conjunto de un barco de vela es indispensable contemplar las velas! Mi trabajo de investigación empezó precisamente estudiando la forma funicular de una vela sometida a un conjun- to de cargas.[133] En aquél estudio la principal hi- pótesis de trabajo era que el comportamiento de la vela se podía asemejar estructuralmente al de una membrana. Es decir, que la energía de deformación por ﬂexión es despreciable. Lo que se pretendía era obtener una aproximación inicial de la forma de la vela que permitiese acelerar el cálculo de la respues- ta estructural de la vela. A sailboat requires sails! My research work started studying the funicular shape of a sail under a set of loads.[133] In that study the main hypothesis was that the behavior of a sail could be represented structurally by a mem- brane. This implies that the bending energy is negligible. The goal was to obtain an initial ap- proximation to the the shape of the sail allow- ing to accelerate the computation of the struc- tural response of the sail. One of the conclusions reached during that study was that calculating a membrane is not in strictu sensu the calculation of a structure. The argument lies in that lacking bending stiff- ness, the membrane turns into a mechanism with inﬁnite swivels. In order to tackle the problem posed (from the structural and numer- ical point of view [133]) by the fact that a mem- brane is a mechanism, it is required to apply strategies to add energy and complete the sys- tem to full rank. Una de las conclusiones obtenidas de aquél estu- dio fue la certeza de que el cálculo de una membra- na no es en strictu sensu el cálculo de una estructu- ra. El argumento reside en que al carecer de rigidez a ﬂexión, la membrana se convierte en un mecanis- mo con múltiples rótulas. Para atajar el problema que supone (desde el punto de vista estructural y nu- mérico [133]) el hecho de que la membrana sea un mecanismo es preciso aplicar técnicas que añadan energía para que la matriz del sistema de ecuacio- nes sea de rango completo. Many authors have applied artiﬁcial bend- ing stiffnesses to the swivels between elements (either at the vertices or at the edges). In a pre- vious work by R.L. 1.3 Spars Un barco de vela, a diferencia de otros tipos de bar- cos, tiene la peculiaridad de que su estructura no ﬁnaliza en el casco (o los cascos en el caso de mul- ticascos); sino que a éste hay que añadir uno o más mástiles, botavaras y tangones que sirven de soporte para multitud de velas. Por si fuera poco, la enver- gadura que adquieren los mástiles contemporáneos (alturas de hasta 40 metros o más) requiere en ge- neral que éstos estén atirantados por un complejo sistema de cables llamado jarcia y que está com- puesto por estayes, obenques, burdas, crucetas, etc. El conjunto del mástil o mástiles con la jarcia que los acompaña recibe el nombre de arboladura. Unlike other kinds of boats, a sailboat’s struc- ture continues above the deck of the hull (or hulls in the case of multihulls). Sailboats in- clude one or more masts, booms and poles which are used to support and deploy mul- tiple sails. Moreover the span of contem- porary yachts’ masts (up to 40 meters tall or more) requires them to be cable-stayed by a sophisticated system of cables and spread- ers named standing rigging which comprises: stays, shrouds, back-stays, spreaders, etc. Technology has also affected how the masts are made. Solid wood masts has been replaced by hollow and thin walled sections of extruded metal (generally aluminum alloys), and more recently of FRP (generally epoxy resin rein- forced with carbon ﬁbers). It’s with the use of advanced composites that masts have reached spectacular proportions thanks also by the de- sign of the laminates. Even though the mast is an inherently linear piece, a detailed study of the mast requires considering the piece as a cylindric or conic shell with sections of variable geometry. También en el caso de los mástiles la tecnolo- gía de construcción ha permitido sustituir el mate- rial con el que se construyen: pasando de la madera maciza a secciones huecas de paredes delgadas for- madas por metal extruido en primer lugar (general- mente aleaciones de aluminio), y más recientemente por plásticos reforzados con ﬁbras (generalmente re- sina epoxi reforzada con ﬁbra de carbono). Es en es- te último caso donde los mástiles están alcanzando su máxima expresión, con arboladuras espectacula- res por su altura y donde sus prestaciones vienen fa- vorecidas por la construcción en laminado. Si bien 1.4. 1.3 Spars SAILS 3 la pieza elástica del mástil es inherentemente lineal, un estudio detallado del mismo requiere considerar la pieza como una lámina cilíndrica o cónica con sección de geometría variable. CHAPTER 1. MOTIVATION 4 rizes my thinking is as follows. rizes my thinking is as follows. rizes my thinking is as follows. reemprendí mis estudios de doctorado en 2006. El pensamiento que resume mis reﬂexiones es el que si- gue. reemprendí mis estudios de doctorado en 2006. El pensamiento que resume mis reﬂexiones es el que si- gue. 1. If the transient dynamic solution leads to a static stationary result, then such static stationary result exists and should be reachable by means of a static calcu- lation. Note that in [123], when the solu- tion converges towards the static solution, the dynamic conditions of the calculation are dropped and a ﬁnal static step is per- formed. 1. Si la solución dinámica y transitoria condu- ce a un resultado estacionario, es que dicho resultado estacionario existe y se debe poder alcanzar mediante un cálculo estático. Nóte- se que en [123], cuando la solución transitoria converge y se aproxima a la solución estática, se relajan las condiciones del cálculo dinámi- co y se realiza un último cálculo estático. 2. If there is a need to add modes or en- ergy to complete the system to full rank in order to reach the static result, it’s be- cause such energy actually exists and it has been stripped in our classical hypoth- esis. The main of which is considering the sail as a membrane. 2. Si para obtener el resultado estático es preciso añadir modos o energía que completen el ran- go del sistema, es porque dicha energía existe en realidad y la estamos quitando en nuestras hipótesis. La principal de todas ellas es consi- derar la vela como una membrana. ¿Y si la vela efectivamente es capaz de resistir ﬂe- xiones? ¿Y si en deﬁnitiva la energía de ﬂexión no es despreciable? ¿Porqué añadir energía de ﬂexión ar- tiﬁcialmente en lugar de considerarla naturalmen- te? Estas eran preguntas que una y otra vez resur- gían y me inquietaban. ¿Y si la vela efectivamente es capaz de resistir ﬂe- xiones? ¿Y si en deﬁnitiva la energía de ﬂexión no es despreciable? ¿Porqué añadir energía de ﬂexión ar- tiﬁcialmente en lugar de considerarla naturalmen- te? Estas eran preguntas que una y otra vez resur- gían y me inquietaban. Can the sail indeed resist bending? What if the bending energy is not negligible? Why adding artiﬁcial bending energy instead of taking it into account naturally? 1The decision was also affected by the trend set by North Sails with the program MemBrain. See ap- pendix A.1. 1La decisión también estuvo inﬂuenciada por la tendencia iniciada por North Sails con su programa MemBrain. Ver el apéndice A.1. CHAPTER 1. MOTIVATION These were questions that kept pounding me. In section 1.5 the author attempts to pro- vide an answer to these questions from an in- tuitive and engineering point of view. En el apartado 1.5 se trata de dar respuesta a estas preguntas desde un punto de vista intuitivo e ingenieril. 1.4 Sails Taylor [123] he uses another technique to add the necessary energy by con- sidering the dynamic effects into the problem. That is, by avoiding to solve the solution to the static problem the authors acknowledge the ex- istence of a transient state during which the membrane makes the transition from a state out of equilibrium to the conﬁguration that counterbalances the loads applied upon it. By means of including the effects of the inertia and damping forces, we end up adding enough en- ergy to the problem to avoid the numerical is- sues caused by the mechanism and hence solve the problem. Muchos autores han recurrido a la asignación de rigideces artiﬁciales en las rótulas de unión en- tre elementos (ya sea en los vértices o en las aristas). En un trabajo contemporáneo de R.L. Taylor [123] se utiliza otra técnica para añadir la energía nece- saria mediante la introducción de la dinámica en el problema. Es decir, huyendo de buscar directamen- te la solución estática del problema, se reconoce la existencia de un estado transitorio en el que la mem- brana transita dinámicamente de un estado en des- equilibrio a la conﬁguración estable que equilibra las cargas aplicadas sobre ella. Mediante la consi- deración de las fuerzas de inercia y de amortigua- miento en el problema, asignamos suﬁciente energía al problema para evitar los inconvenientes numéri- cos del mecanismo y poder así resolver el problema. This discussion kept my interest during the years past after I completed my undergraduate studies in 2001 and until I restarted my PhD studies in 2006. The reasoning that summa- Toda esta problemática ha mantenido despier- to mi interés durante los años que han transcurri- do desde que terminé la carrera en 2001 hasta que CHAPTER 1. MOTIVATION La naturaleza laminar de las velas En este capítulo se ha mencionado reiteradamente como la introducción de nuevos materiales y nuevas tecnologías de construcción han afectado el desarro- llo de los barcos. El caso de las velas no es distin- to. Si bien primitivamente las velas se confecciona- ban a partir de tejidos exclusivamente —en [57] Ic- kert ilustra como todavía en los años 50 del siglo XX las velas se confeccionaban a partir de tejido de algodón— y esta tecnología invitaba a la considera- ción de que en efecto existían un sinfín de rótulas en el dis-continuo del material que componía la ve- la; más recientemente los materiales reforzados con ﬁbras también han hecho su irrupción en la tecnolo- gía de fabricación de velas. Inicialmente se aplica- ban tratamientos que impermeabilizaban el tejido, permitiendo aumentar su eﬁciencia aerodinámica al evitar pérdidas del gradiente de presión que las velas generan cuando deﬂectan el ﬂujo de aire. La técnica se fue depurando y hoy en día las matrices de resina que se utilizan permiten que el entrama- do de ﬁbras no tenga que ser tupido y mucho menos que tenga que ser tejido. Hasta el extremo de que las velas en los casos más exigentes ya no se fabrican mediante paños unidos (cosidos primero y soldados después) sino que se fabrican disponiendo las ﬁbras una a una sobre el molde que da la forma a la vela y se laminan ambas caras con la matriz de resina elegida. El avance más popular de los últimos años en lo que a tecnología de fabricación de velas con- cierne es precisamente el último descrito, y que fue objeto de la patente denominada 3DL [7]. During this chapter I have repeatedly men- tioned how the introduction of new materi- als and new construction technologies have af- fected the development of boats. Sails are not different. While in the old times sails were manufactured using textiles only —in [57] Ick- ert illustrates how cotton was used was still used to make sails in mid of the 20th century— and this technology determines making the consideration that there are indeed an inﬁnite number of swivels in the dis-continuum mate- rial of the sail; more recently FRP have dis- rupted the sail’s manufacturing technology. 1.5. SHELLS VS MEMBRANES 5 lo eran muy inferiores a los esperados debido a que otros modos de energía absorbían toda la energía de deformación al modo de energía por deformación de membrana. Por ello se pretendía partir de una con- ﬁguración geométrica más cercana a la ﬁnal con el cálculo de la forma funicular de las cargas. placements obtained by calculation were far smaller than the displacements expected be- cause other deformation modes where absorb- ing all the energy instead of allocating the en- ergy adequately to the membrane deformation mode. That’s why my undergraduate thesis tried to start with a geometric conﬁguration closer to the ﬁnal shape by means of the funic- ular shape of the loads. 1.5 Shells vs Membranes Revisitemos la decisión de modelar las velas como membranas. Aunque esta reﬂexión no quedase do- cumentada en mi tesina de ﬁnal de carrera [133], lo cierto es que la principal motivación para conside- rar las velas como membranas es que se había hecho sin demasiado éxito como láminas en un trabajo an- terior de Lara Pellegrini [99] también en CIMNE.1 Tuve la ocasión en su momento de comentar con la Sra. Pellegrini sus experiencias del estudio de ve- las y una de ellas era que los elementos de lámi- na utilizados no ofrecían resultados suﬁcientemente satisfactorios. El elemento de lámina utilizado pa- ra aquellos cálculos era el elemento DKT. El prin- cipal motivo era el bloqueo de la solución. Es decir, que los desplazamientos obtenidos mediante cálcu- Let’s revisit the decision to model sails as mem- branes. Despite this consideration was not documented in my undergraduate thesis [133] truth is that the main reason behind consider- ing the sails as membranes was because they had been modeled as shells without much suc- cess in a previous work by Lara Pellegrini [99] also at CIMNE.1 At the time I had the chance to discuss with Ms. Pellegrini her experiences on the study of sails, and one of them were that the shell elements used did not provide sat- isfactory results. The shell element used for those simulations was the DKT element. The main reason behind this lack of performance was locking of the elements. That is, the dis- 1.5.2 Spinnakers: a different case Merecen una mención aparte una familia de velas que por sus características y prestaciones se diferen- cian del resto de velas que enarbola un barco de ve- la. Los spinnaker (simétricos o asimétricos) son una familia de velas diseñadas para empujar la embar- cación en rumbos portantes, esto es, favorecidos por la dirección del viento. Debido a esto, su uso no sue- le causar peligrosas fuerzas laterales de escora en el barco y estas velas alcanzan proporciones gigan- tescas. Dado el enorme metraje de estas colosales velas es imprescindible minimizar su gramaje to- do lo posible. —De hecho no es extraño ver “explo- tar” estas velas cuando la presión del viento vence su resistencia estructural—. Esta restricción de di- seño ha supuesto hasta ahora una barrera a la in- troducción de la tecnología de fabricación lamina- da en los spinnaker. No únicamente por considera- ciones de peso sino también de cariz más práctico: cuando se tienen que plegar estas velas es necesa- rio que ocupen muy poco espacio una vez estibadas y por ello tienen que poder permitir ser replegadas en desmesura. Los delicados materiales utilizados para laminar velas no permiten un tratamiento tan desconsiderado sin afectar a la durabilidad del ma- terial y ello ha permitido a la tecnología textil so- brevivir en la fabricación de los spinnaker. Está por ver la evolución futura de esta situación, ya que la aparición de los novedosos y extremos Códigos Cero [43] abre una brecha para que se empiece a pensar en la fabricación de spinnakers a base de materiales compuestos. There is a kind of sails that deserves a special consideration that given their characteristics and behavior are different from the other sails hoisted in a sailboat. The spinnakers (symmet- ric or asymmetric) are a family of sails designed to propel the boat when the wind blows in fa- vor. This very circumstance makes their use uncompromising for the stability of the boat be- cause they don’t generate large lateral forces that could tip the boat over. As a result, these sails reach gigantic dimensions. Given the enormous area of these sails it becomes neces- sary to minimize their surface density as much as possible. La naturaleza laminar de las velas In the beginning the textile material was treated to make it airtight and thus increase its aero- dynamic efﬁciency as it decreases the losses of the pressure gradient generated by the sails as they deﬂect the airﬂow. This technique has evolved and nowadays the resin matrices used allow the use of a coarser fabric or even the use of unwoven ﬁbers. In the limit, for the more demanding situations, the sails are no longer manufactured by joining together differ- ent panels (sewn at ﬁrst and glued later) but instead by laying the ﬁbers one by one onto a mold that provides the shape to the sail and laminating both sides with the resin matrix chosen. Precisely, the technique described con- stitutes the most popular advance in the tech- nology of sail-making in the last decades. This was object of a controverted patent named 3DL [7]. Así pues, la respuesta al dilema entre láminas y membranas nos la da la tecnología utilizada para fabricar las velas. A la vista de la evolución tecnoló- gica, ya no existen diferencias fundamentales entre el modo como se fabrican el casco, el mástil, o las velas, y por lo tanto tampoco debería haberlas en el modo como las concebimos estructuralmente. Therefore, the technology used to manu- facture the sails provides the answer to the dilemma between shells and membranes. Look- ing at the technological evolution there aren’t fundamental differences between the manufac- turing methods used for hulls, masts, or sails. CHAPTER 1. MOTIVATION 6 And thus there shouldn’t be differences ei- ther between the ways we conceive them struc- turally. 1.6 Justiﬁcation of the use of high technology in sailing La justiﬁcación que voy a exponer tiene motiva- ciones principalmente económicas, y no se entende- ría que un trabajo cientíﬁco de carácter ingenieril se justiﬁcase de inicio por motivos económicos — cuando por añadidura no es exacto—. Pero sí que es cierto que el autor no es insensible a los argumentos que presenta a continuación. The discussion I am about to present has mainly economic reasons, and the reader might not understand that a scientiﬁc work with an engineering tone is justiﬁed by economic moti- vations —besides, this is not exactly the case—. But it is true that the author is not indifferent to the arguments set forth below. 1.6. JUSTIFICATION OF THE USE OF HIGH TECHNOLOGY IN SAILING 7 que decanta la balanza claramente en favor de los elementos de lámina sin rotaciones en frente de los elementos de membrana. que decanta la balanza claramente en favor de los elementos de lámina sin rotaciones en frente de los elementos de membrana. ligible. But while the membrane model will inevitably require a dynamic transient analy- sis (or another analogous technique), the shell model can tackle the static problem straight away. This is, in my opinion, a clear advan- tage that tips the balance clearly in favor of the rotation-free shell elements with respect to the membrane elements. 1.5.2 Spinnakers: a different case —In fact, it is not uncommon to see these sails “explode” when the wind pres- sure overcomes their structural resistance.— This design restriction has thus far implied a de facto barrier to the introduction of the lam- inate manufacturing technology in spinnakers. Not just because of weight considerations, but also because of another more pragmatic con- sideration: in order to stack these sails they need to take as little volume as possible and therefore they have to withstand being folded and pressed with little regard. The delicate materials used to manufacture laminated sails cannot withstand such careless manipulation without affecting the durability of the mate- rial. This consideration has allowed the woven textile manufacturing technology to survive for the case of spinnakers. The future evolution of this situation is uncertain. The recent ap- pearance of new and radical sail designs such as Code Zero [43] opens an opportunity to start thinking on the manufacturing of spinnakers using composite materials. Por ahora, sin embargo, no es incorrecto concebir el análisis de un spinnaker como una membrana. Sin embargo, tampoco sería muy incorrecto anali- zar un spinnaker como una lámina delgada. Dada la extrema esbeltez de la vela, si esta puede ser re- producida por el modelo de lámina delgada, las di- ferencias en los resultados analizando con un mo- delo u otro serán prácticamente inexistentes. Con la diferencia de que el modelo de membrana requeri- rá en todo caso un enfoque dinámico transitorio (u otra técnica equivalente), mientras que el modelo de lámina podrá abordar directamente el problema es- tático. Esta es bajo mi punto de vista una ventaja However it be, it is not incorrect to con- sider the analysis of a spinnaker sail as a mem- brane. But it wouldn’t be erroneous neither to analyze it as a thin shell. If the extreme slenderness of the sail can be taken into ac- count and accurately represented by a thin shell model, the differences between the results using one model or the other should be neg- 1.6. JUSTIFICATION OF THE USE OF HIGH TECHNOLOGY IN SAILING CHAPTER 1. MOTIVATION 8 que las reglas no siempre han conseguido promo- ver una evolución tecnológica positiva de los diseños de los barcos. Al contrario, en numerosos casos, bar- cos con peores características pueden resultar ven- cedores frente a mejores barcos debido a los defectos de las reglas2. A pesar de ello, las fuerzas del mer- cado siempre terminan triunfando y los diseños — tomados en promedio— han ido evolucionando po- sitivamente con el tiempo; lo que ha ido obligando a revisiones de las reglas para tener en cuenta las in- novaciones en los diseños —véanse las recientes mo- diﬁcaciones a la última versión del programa VPP (Velocity Prediction Program) para la regla ORC International [91]—. Más aún, con el crecimiento de lo que podríamos llamar el mercado de la com- petición de la vela a nivel mundial, varias regiones del mundo fueron desarrollando reglas especíﬁcas. Hasta el punto que hoy en día, las reglas son regula- das y administradas por las autoridades nacionales de cada país o conjunto de estados. No es descabe- llado pensar en el extremo de que cada club tenga su propia regla de competición. always promoted a positive technological evo- lution of the sailboats. On the contrary, in many cases boats with worse characteristics can result winners in front of better boats be- cause of the errors in the rules2. Neverthe- less, market drivers tend to prevail and the de- signs —taken in average— have evolved pos- itively with time; which has forced a review of the rules in order to take into account the design innovations —see the recent modiﬁca- tions to the latest version of the VPP (Velocity Prediction Program) used for the International ORC rule [91]—. Moreover, as the global sail- ing competition market has grown, different regions of the planet have developed speciﬁc rules. To the point that nowadays the rules are established and administrated by the national authorities of each country or group of coun- tries. It is not completely ridiculous to think that each club might end up with its speciﬁc competition rule. There is another explanation for this prolif- eration of organisms that manage a particular competition rule. It consists in the acceptance of the obligation to possess a certiﬁcate comply- ing with a given rule in order to compete in a regatta administered by that speciﬁc rule. 2DeBord et al. [59, p. 4] explican de manera concisa cómo se puede producir esta paradoja en su artículo sobre los diseños en la Copa del América y la regla IACC. 2DeBord et al. [59, p. 4] explain very eloquently how this paradox can occur in their paper on the America’s Cup designs and the IACC rule. Las reglas de competición entre barcos de vela distintos Despite how strange it may sound, the creation of the Chesapeake Sailing Yacht Symposium (CSYS) doesn’t happen in response the the interests of the scientiﬁc community around the America’s Cup as the pinnacle of sailing. When reviewing the topics presented during the ﬁrst editions of the symposium, it’s reveal- ing to realize that the large majority discusses about the rules that allow comparing differ- ent boats competing together. At that moment, the rule that had been created shortly be- fore was the International Offshore Rule (IOR). From a market standpoint there is a great eco- nomic interest in enabling different boats to compete together between them, grouped by classes or similarities; and hence this focus on the rules. Enabling this possibility implies that more yacht owners can compete with their gen- eral purpose yacht. And the economic value of this market is huge. Por extraño que pueda parecer, el nacimiento del Chesapeake Sailing Yacht Symposium (CSYS) no responde a los intereses de la comunidad cientíﬁca alrededor de la Copa del América como pináculo del deporte de la vela. Si repasamos los asuntos trata- dos en sus primeras ediciones resulta revelador que en su mayoría se hable de las reglas que permiten comparar distintos barcos que compiten entre sí. En ese momento, la regla que se había creado reciente- mente era la International Offshore Rule (IOR). Es- te énfasis en las reglas se entiende fácilmente por el gran interés económico y de mercado que tiene per- mitir que distintos barcos puedan competir entre sí; agrupados por clases/similitud. Abrir esa posibili- dad implica que más propietarios de barcos puedan competir con sus utilitarios y por lo tanto tengan un mayor incentivo para poseer una embarcación. Y el valor económico que tiene este mercado es enorme. Por lo tanto tiene interés repasar aunque sea muy someramente cuál ha sido la evolución de es- tas reglas de comparación para que distintos bar- cos puedan competir entre sí. Las primeras reglas dominaban internacionalmente, pues estos esfuer- zos primitivos eran escasos. Sin embargo las re- glas eran imperfectas, permitiendo a los diseñado- res explotar los defectos de las reglas. Esto implica Therefore it’s interesting to review, albeit brieﬂy, how these rules have evolved. The ﬁrst rules dominated globally because the ﬁrst ef- forts were scarce. However the rules were im- perfect, allowing the designers to exploit their defects. This implies that the rules haven’t CHAPTER 1. Las reglas de competición entre barcos de vela distintos MOTIVATION CHAPTER 1. MOTIVATION This certiﬁcate has a cost that has to be renewed pe- riodically even if the boat has not undergone any change. This paradigm is reaching unde- sirable situations because the last decade has seen a drastic decline in the number of boats competing in regattas open to boats different between them. Thus loosing a great market op- portunity. Existe otro motivo que explica esta proliferación de organismos que se erigen como administradores de una regla de competición particular. Y consiste en la aceptación de que para poder participar en una regata administrada bajo una determinada re- gla de competición cada barco participante debe es- tar en posesión de un certiﬁcado de acuerdo con esa determinada regla. Certiﬁcado que tiene un coste y que es preciso renovar periódicamente, aunque el barco no haya sufrido ninguna modiﬁcación. Esta situación está llegando a extremos indeseables, pues se ha visto en la última década una drástica reduc- ción en el número de embarcaciones participantes en regatas abiertas a barcos distintos entre sí. Por lo que se está perdiendo una gran oportunidad de mercado. The hope brought by science is the possi- bility of aligning the interests of yacht own- ers willing to compete between them despite having different boats and the unstoppable technological advance of the newer designs. This objective shall be achieved without gen- erating absurd scenarios as science’s goal can never be the absurdity. Therefore, unlike the large majority of sports where the use of high-end technology is only justiﬁed for high performance competition; high technology in La esperanza que trae la ciencia es poder unir los intereses de los propietarios de barcos que quie- ren competir entre sí aunque tengan barcos distin- tos, y el irrefrenable avance tecnológico de los dise- ños de barcos. Conseguir este objetivo es hacerlo sin por ello generar oportunidad para situaciones ab- surdas, pues no es nunca el objetivo de la ciencia producir un absurdo. Por lo tanto, a diferencia de 1.6. JUSTIFICATION OF THE USE OF HIGH TECHNOLOGY IN SAILING 1.6. JUSTIFICATION OF THE USE OF HIGH TECHNOLOGY IN SAILING 9 la gran mayoría de deportes, en los que la aplica- ción de alta tecnología solo se justiﬁca para la prác- tica de la competición de élite; en el deporte de la vela la alta tecnología —además de aplicarse en la competición de élite— es imprescindible para poder practicar la competición a nivel amateur. Esta pecu- liaridad es debida a que los barcos (el equipamiento necesario para la práctica de la vela) tienen un coste económico que supera en varios órdenes de magni- tud al equipamiento necesario en la mayoría de los deportes. La ciencia nos ofrece la promesa de hacer comparables barcos que rinden distintamente y por lo tanto poder valorar al deportista minimizando las diferencias debidas al equipamiento. La presen- te tesis pretende aportar solo algunas de las piezas que pueden ser necesarias para que la ciencia cum- pla su promesa. sailing —besides being applied to the elite competition— is absolutely necessary in order to compete at amateur level. This peculiar- ity happens because the yachts (the required equipment to practice the sport) has an eco- nomic cost that surpasses in several orders of magnitude the cost of the equipment necessary in the majority of sports. Science makes the promise to make comparable boats that per- form differently and therefore make it possi- ble to evaluate the sportsman minimizing the differences established by the equipment. The present thesis attempts to bring just some of the elements that can be necessary in order for science to fulﬁll its promise. • Simulación dinámica de membranas en 3D • Simulación dinámica de membranas en 3D • Dynamic simulation of membranes in 3D • Dynamic simulation of membranes in 3D – Teoría no lineal con grandes desplazamien- tos. – Large displacements non-linear theory. – Orthotropic materials model. – Modelo de materiales ortótropos. • Dynamic simulation of shells in 3D • Simulación dinámica de láminas en 3D – Large displacements non-linear theory. – Teoría no lineal con grandes desplazamien- tos. – Large strains non-linear theory. – Anisotropic laminated composite mate- rials model. – Teoría no lineal con grandes deformaciones. – Modelo de materiales laminares compues- tos anisótropos. – Assign the principal directions of the material properties. – Asignación de direcciones principales. • Dynamic simulation of beams in 3D • Dynamic simulation of beams in 3D • Simulación dinámica de vigas en 3D Objectives I N ORDER TO RESPOND TO THE MOTIVATIONS set forth in chapter 1 and state the objectives of this thesis, let’s analyze which are the com- putational tools (capabilities) needed to increase the level of detail in the simulation and enable a further understanding and analysis of the response of sailboats. An enumeration of some of them follows: P A P ARA PODER DAR RESPUESTA A LAS MOTIVACIONES expuestas en el capítulo 1 y enunciar los objetivos de la presente tesis, veamos qué herramientas compu- tacionales (capacidades) se precisan para simular con un nivel de detalle que permita dar un paso más allá en el análisis y comprensión del comportamiento de barcos de vela. A continuación se enumeran algunas de ellas: I N • Cálculo en paralelo • Sailboats singularities • Sailboats singularities – Preferiblemente en memoria compartida (protocolo OpenMP). – The boat is a prestressed structure, and thus it is necessary to reproduce the different states of prestress of the structure imposing the compatibility of uncoupled relative displacements. • Interacción Fluido-Estructura – Coupling interfaces in 3D. – Large deformations of the ﬂuid mesh. – Acoplamiento fuerte Aire ⇄Barco ⇄Agua. – Interfases de acoplamiento en 3D. – Acoplamiento fuerte Aire ⇄Barco ⇄Agua. • Dynamic control of the boat • Dynamic control of the boat – Grandes deformaciones de la malla del ﬂui- do. – Equilibrium with several degrees of freedom to satisfy both static or dy- namic conditions. • Control dinámico del barco • Parallel computing • Parallel computing – Equilibrio de varios grados de libertad pa- ra satisfacer restricciones estáticas o diná- micas. – Preferably implemented in shared memory (OpenMP protocol). – Modelo de materiales no lineales. – Free surface problem. – Free surface problem. • Simulación dinámica de ﬂuidos en 3D • Fluid-Structure interaction • Fluid-Structure interaction – Fluidos newtonianos no compresibles. – Problema de superﬁcie libre. – Strong coupling Air ⇄Boat ⇄Water. – Strong coupling Air ⇄Boat ⇄Water. – Coupling interfaces in 3D. L d f ti f th ﬂid h CHAPTER 2. OBJECTIVES – Non compressible Newtonian ﬂuids. – Modelo de materiales no lineales. • Simulación dinámica de vigas en 3D – Large displacements non-linear theory. – Teoría no lineal con grandes desplazamien- tos. – Large strains non-linear theory. – Anisotropic laminated composite mate- rials model. – Teoría no lineal con grandes deformaciones. – Modelo de materiales compuestos lamina- res anisótropos. • Dynamic simulation of cables in 3D – Large displacements non-linear theory. • Simulación dinámica de cables en 3D – Non-linear materials model. – Teoría no lineal con grandes desplazamien- tos. • Dynamic simulation of 3D ﬂuids 11 CHAPTER 2. OBJECTIVES 12 Focalizando el objetivo In view of the introductory list of the necessary capabilities and the review of the technologies used nowadays by sailboat designers, there is a need to concentrate the efforts in order to develop those which are particularly relevant to improve the calculation tools for sailboats. Therefore let’s divide the complex system which is a sailboat into 3 large subsystems from the computational me- chanics point of view: A la vista de la enumeración anterior de capacida- des necesarias, es preciso concentrar los esfuerzos para desarrollar aquellas que resultan especialmente rele- vantes para mejorar las herramientas de cálculo de barcos de vela. Es por ello que vamos a descomponer el sistema complejo que representa un barco de vela en 3 grandes subsistemas desde el punto de vista de la mecánica computacional: Estructura: que comprende todos los elementos só- lidos resistentes en el barco, incluyendo el cas- co, los apéndices, la arboladura, el velamen y el aparejo. Structure: comprising all the solid resisting el- ements in the boat, including the hull, the appendages, the spars, the sails and the standing rigging. Fluido: que contempla tanto el medio líquido (agua) en el que se soporta la embarcación como el me- dio gaseoso que lo propulsa (aire). Fluido: que contempla tanto el medio líquido (agua) en el que se soporta la embarcación como el me- dio gaseoso que lo propulsa (aire). Fluid: comprising both the liquid medium (wa- ter) on which the boat ﬂoats and the gaseous medium which propels it (air). Fluid: comprising both the liquid medium (wa- ter) on which the boat ﬂoats and the gaseous medium which propels it (air). Interfase Fluido-Estructura: es la parte que es preciso tener en cuenta desde un punto de vista de la mecánica computacional cuando se resuel- ve el sistema abordando sus partes separadas. Esta técnica es por lo general más eﬁciente, pero tiene algunos inconvenientes que es preciso resol- ver. Fluid-Structure interface: is the part that needs to be accounted for from a compu- tational mechanics point of view when the system is solved tackling its parts sepa- rately. This technique is more efﬁcient in general, but implies some difﬁculties that need to be solved. Se puede apreciar que en esta clasiﬁcación hemos omitido expresamente algunas de las capacidades lis- tadas en un principio como por ejemplo el control di- námico del barco. 2.1. NARROWING DOWN THE OBJECTIVE 13 • Singularidades de los barcos de vela • Singularidades de los barcos de vela – El barco es una estructura pretensada, y por ello es preciso reproducir los distin- tos estados de pretensado de la estructura mediante la compatibilización de desplaza- mientos relativos desacoplados. * Loading of the spars and stand- ing rigging (tensioning of stays and shrouds). * Setting the battens (compatibility of the batten geometry with that of its bag in the sail). * Puesta en carga de la jarcia (tensado de estayes y obenques). * Colocación de sables forzados (compati- bilización de la geometría del sable con la de su funda en la vela). * Hoisting the sails (compatibility of the geometry of the sail with that of the spars). * Izado de las velas (compatibilización de la geometría de la vela con la geometría de la arboladura). * Trimming the rig: sheets, vang, outhaul, downhaul, etc. * Trimado del aparejo. Cazado de esco- tas, contra, pajarín, cunningham, etc. Appendix A presents succinctly the evolution of the use of numerical methods in the calculation and design of high performance sailboats; while also signaling the principal tools used by the in- dustry. Undoubtedly the argument of this histor- ical review uses the America’s Cup as a backdrop, although the author uses also other top level ref- erences to enrich the argument with a particu- lar point of view like the ocean going round-the- world regattas and the advances of the industrial leaders in the ﬁeld. El apéndice A presenta sucintamente cómo ha evo- lucionado la presencia de los métodos numéricos en el cálculo y diseño de veleros de competición; al tiem- po que indica las principales herramientas usadas en la industria. El hilo conductor de esta revisión es sin lugar a dudas la Copa América, aunque también se usan otras referencias de primer nivel que enriquecen el punto de vista como las competiciones oceánicas de vuelta al mundo y los avances de los líderes industria- les del sector. 2.1. NARROWING DOWN THE OBJECTIVE Focalizando el objetivo Las capacidades listadas en la in- troducción de este capítulo representan sin duda al- guna el conjunto de mis ambiciones cientíﬁcas en el campo de la mecánica computacional en el momen- to de abordar mi carrera académica (sin limitarme a ellas), pero sería poco realista asumir que todas ellas son alcanzables en el marco de una tesis doctoral. De hecho, confío poder continuar mi carrera académica resolviendo de manera consecutiva los distintos retos planteados en esta introducción. The reader will notice that the list above ex- pressly omits some of the capabilities listed in the beginning; like for example the dynamic control of the boat. The capabilities included in the list in the introduction of the present chapter represent beyond any doubt the ensemble of my scientiﬁc ambitions in the ﬁeld of computational mechan- ics at the moment of starting my academic career (without becoming limited by them), but it would be unrealistic to imagine that all of them are at- tainable in the framework of a doctoral thesis. In fact, I look forward at progressing in my career solving consecutively the various challenges set forth in this introduction. Especíﬁcamente en esta tesis me he centrado úni- camente en el primero de los tres grandes subsistemas descritos en el punto anterior: la estructura. Speciﬁcally, in this thesis I have set the fo- cus on the ﬁrst of the three great subsystems de- scribed in the previous section: the structure. El motivo para dejar a un lado el subsistema que representa el ﬂuido es que se trata de un campo en muy rápido desarrollo en la comunidad de la mecá- nica computacional a nivel mundial, e incorporar es- ta temática en una tesis doctoral implicaría inmedia- tamente dedicar la totalidad de la tesis a este tema. The reason for setting aside the subsystem representing the ﬂuid is that it is a ﬁeld under very quick development within the international computational mechanics community. Including this topic in a doctoral thesis would imply imme- CHAPTER 2. OBJECTIVES 14 Table 2.1: Summary of tools and technologies used in the industry to simulate sailboats. Subsystem Subsistema Part Parte Capabilities needed Capacidades requeridas Computational tools used Herramientas computacionales utilizadas Structure Hull Static and dynamic simulation of shells. Small displacements. Linear theory. Fairly well solved using Reissner-Mindlin-type elements. This implies using rotational degrees of freedom. 2.2 Rotation-Free Thin Shell Element Structure-wise, in the previous pages the author has pinpointed different kinds of structural typol- ogy: membranes, shells, beams and cables. Out of these types of structural typologies this thesis focuses on shells; more precisely on thin shells. The argument for this election is two-fold. On the one side, the one-dimensional typologies such as beams and cables as well as the typology of membranes have been continuously thoroughly studied and it’s fair to say that satisfactory solu- tions have been achieved in all cases. The imple- mentations considering the non-linear theory of large strains/displacements or anisotropic mate- rials aren’t anything more than that: implemen- tations. Their corresponding methodology has been widely developed and its inclusion in a doc- toral research work would not be justiﬁed. On the other side, shells are prominently used in the construction technology of sailboats. A detailed justiﬁcation of this statement is presented in sec- tion 1.1. Therefore, the selection of the shells structural typology as the object of study of the present thesis is supported by both scientiﬁc ob- jectives and technological objectives. Por lo que respecta a la estructura, en las páginas an- teriores se han distinguido varias tipologías estructu- rales: membranas, láminas, vigas y cables. De todas ellas, esta tesis se centra en la tipología estructural de láminas; y más concretamente en las láminas delga- das. El motivo de esta elección es doble. Por un lado las tipologías unidimensionales de viga y cable así como la tipología de membranas han sido objeto intenso de estudio de manera continuada y se puede considerar que se han alcanzado soluciones muy satisfactorias en todos los casos. Las implementaciones con teoría no lineal de grandes deformaciones/desplazamientos o con materiales no isótropos no son más que eso: im- plementaciones. La metodología para las cuales ha si- do ampliamente desarrollada y no se justiﬁcaría su inclusión en un trabajo de investigación doctoral. Por otro lado las láminas tienen una destacada presen- cia en la tecnología de construcción de barcos a vela. Desarrollo una justiﬁcación detallada de esta aﬁrma- ción en el apartado 1.1. Por lo tanto, se puede aﬁrmar que la elección de la tipología de láminas como obje- to de estudio especíﬁco de esta tesis es a la vez una elección motivada por objetivos cientíﬁcos y objetivos tecnológicos. The selection of the thin shells subtypology re- sponds to the challenge presented by the sails of boats. 2.2. ROTATION-FREE THIN SHELL ELEMENT 2.2. ROTATION-FREE THIN SHELL ELEMENT bles en la mejora de los algoritmos de simulación en los que la suavidad geométrica puede ser capital para la correcta resolución de los problemas de interacción ﬂuido-estructura. some aspects that can be used to improve the sim- ulation algorithms where the geometry smooth- ness is important for a correct solution of the ﬂuid-structure interaction smoothness. Focalizando el objetivo Rotation-free elements haven’t reached yet the same level of maturity as elements with rotations. Spars Static and dynamic simulation of beams and cables. Small displacements. Linear theory. Sails Static and dynamic simulation of shells and membranes. Large displacements. Non-linear theory. MemBrainTM (only for membranes and only dynamic). There aren’t computational tools to simulate sails as shells. Fluid Air Potential ﬂow for upwind, Navier-Stokes for other cases. FlowTM, FLUENTTM, etc. Very good existing solvers from the aeronautical industry. Water Navier-Stokes equations, free surface. Panel solvers, FEM solvers (potential ﬂow and RANS). For example: FANSTM, FLUENTTM, SplashTM, etc. Fluid- Structure interaction Air ⇵ Boat Strong coupling and mesh deformation. In order to improve coupling it is desirable to use the same kinematic variables =⇒no rotations. Structure decomposition: rig separated from hull and considered ﬁx in air ﬂow. Boat ⇵ Water Structure decomposition: hull separated from rig and considered ﬁx in the water. Air ⇵ Boat ⇵ Water Results from Air⇄Boat and Boat⇄Water tabulated and entered as input for VPP model and equilibrium obtained for 3 or 4 degrees of freedom. able 2.1: Summary of tools and technologies used in the industry to simulate sailboats. diately concentrate all the efforts to this topic. Moreover, the thesis would necessarily concen- trate on some very speciﬁc aspect of ﬂuid com- putational mechanics, and this is not the spirit of this research work. For that reason, on anything related to the solution of the ﬂuid domain of the problem, I will rely on previous methods already implemented in existing computational codes. Además, la tesis se enfocaría necesariamente en algún aspecto muy particular de la mecánica computacional de ﬂuidos, cosa que no es el espíritu de este trabajo. Por ello, en todo lo que hace referencia a la resolución del dominio ﬂuido del problema, me apoyaré en métodos previos y ya implementados en códigos computaciona- les existentes. Del mismo modo, la mejora de los algoritmos de in- teracción ﬂuido-estructura requiere de un trabajo de investigación y desarrollo que queda fuera del alcan- ce de la presente tesis. Sin embargo, la investigación realizada incluye aspectos que pueden ser aprovecha- Likewise, improving the ﬂuid-structure inter- action algorithms requires a research work that falls outside of the scope of this thesis. However, the research conducted within the thesis includes 15 CHAPTER 2. OBJECTIVES la tipología de los elementos de lámina sin rotaciones. Dicho elemento de lámina delgada sin rotaciones de- berá poder incorporar los avances que se están reali- zando en paralelo por otros investigadores de CIMNE (el profesor Eugenio Oñate y el profesor Francisco Zá- rate [145]) en el desarrollo de un marco general para complementar elementos de lámina sin rotaciones con las variables de rotación que les permita reproducir la teoría de láminas gruesas de Reissner y Mindlin. shell elements kind. This new rotation-free thin shell element shall permit to include the par- allel advances being made by other CIMNE re- searchers (professor Eugenio Oñate and profes- sor Francisco Zárate [145]) on a new general framework to complement the rotation-free shell elements with the rotation variables enabling the reproduction of the thick shell theory by Reissner and Mindlin. Una de las ventajas que presentan los elementos de lámina sin rotaciones es que se pueden combinar de manera natural con elementos de sólido. Además, las rotaciones carecen de la propiedad aditiva, por lo que su tratamiento en teoría no-lineal con grandes de- formaciones y grandes desplazamientos se hace suma- mente complicado. One of the main advantages of the rotation- free shell elements is their natural combination with solid elements. Moreover, the rotation vari- ables lack the additive property, which makes their use in non-linear theory with large strains and large displacements very difﬁcult. The present thesis will attempt to solve the following challenges posed by rotation-free thin shell elements: La presente tesis pretende resolver los siguientes retos que presentan los elementos de lámina sin rota- ciones: • To guarantee the accuracy of the membrane response of the element. • Garantizar la precisión de la respuesta membra- nal del elemento. • To prevent the dependency of curvature computation on mesh distortion. • Evitar la dependencia del cálculo de la curvatu- ra respecto de la distorsión de la malla. • To design the element as simple as possible in order to limit the stiffness matrix band- width. • Diseñar el elemento lo más simple posible para limitar el ancho de banda de la matriz. • To simplify the application of boundary con- ditions. • Simpliﬁcar la aplicación de las condiciones de contorno. 2.2 Rotation-Free Thin Shell Element In many cases involving composite lami- nated materials, there is agreement that the shell needs to be analyzed under the Reissner-Mindlin assumptions (see section 3.2.2). In other words, the shear strain is not negligible. This is why it is questionable to use the thin shells theory to analyze the structural response of the other structural elements of the boat built with com- posite laminated materials. There will be cases where the Kirchhoff-Love hypotheses sufﬁce, and there will be cases where the Reissner-Mindlin hypotheses are needed. La elección de la subtipología de láminas delgadas responde al reto que supone el cálculo de las velas de los barcos. En el caso de materiales compuestos lami- nados, en muchos casos se considera que es necesario analizar las láminas bajo las hipótesis de Reissner- Mindlin (ver apartado 3.2.2). Es decir, que la deforma- ción por cortante no es despreciable. Por ello, la cons- trucción de los otros elementos estructurales del barco con laminados de materiales compuestos pone en du- da la idoneidad de la teoría de láminas delgadas para analizar la respuesta estructural de dichos elementos. Habrá casos en los que las hipótesis de Kirchhoff-Love sean suﬁcientes, y habrá otros casos en los que sea pre- ciso recurrir a las hipótesis de Reissner-Mindlin. Therefore, a speciﬁc objective of this thesis consists in developing a versatile thin shell el- ement. The author focuses on the rotation-free Por todo lo anterior, un objetivo especíﬁco de esta tesis doctoral consiste en el desarrollo de un elemen- to de lámina delgada versátil. Nos concentraremos en CHAPTER 2. OBJECTIVES 16 2.3 Summary Este capítulo ha enumerado la lista de las herramien- tas computacionales (capacidades) necesarias para si- mular barcos de vela con un nivel de detalle que per- mita dar un paso más allá en el análisis y compren- sión de su comportamiento. A continuación cierra el foco para centrarse únicamente en los aspectos relati- vos a los dominios de la mecánica computacional del problema del barco de vela (estructura, ﬂuido e inter- fase ﬂuido-estructura). Y procede a descomponer los componentes del problema mecánico. El autor apro- vecha esta visión segmentada del barco de vela para determinar qué herramientas de análisis usa actual- mente la industria y si existe margen para mejorarlas. Se puede encontrar una explicación sobre el estado de la técnica de todas estas herramientas en el apéndi- ce A. El autor justiﬁca que es pertinente centrar los This chapter has listed the computational tools (capabilities) needed to increase the level of de- tail in the simulation and enable a further un- derstanding of the response of sailboats. It then narrows down on the computational mechanics domains affected by the sailboat problem (struc- ture, ﬂuid and ﬂuid-structure interface). And starts taking apart the components of the me- chanical problem. The author takes advantage of this segmented viewpoint of the boat to analyze which are the current tools in use by the industry and whether there is room for improvement. The reader can ﬁnd an explanation on the evolution and actual use of all these tools in appendix A. The author justiﬁes that it is worthwhile to focus the efforts of the thesis on the structure system 2.3. SUMMARY 17 of the boat. And more speciﬁcally on developing a new rotation-free thin shell element. The reasons are two-fold: esfuerzos de la tesis en el sistema de la estructura del barco de vela. Y en particular en desarrollar un nue- vo elemento ﬁnito de lámina delgada sin rotaciones. Para ello se esgrimen dos motivos: Firstly, because there’s a need to analyze sails as shells under non-linear assumptions of large strains/displacements. And secondly, because the use of this technology can subsequently be used broadly in the structural analysis of the sailboat. En primer lugar, por la necesidad de analizar las velas como láminas en condiciones de no-linealidad y asumiendo grandes deformaciones/desplazamientos. 2.3 Summary Y en segundo lugar, porque el uso de esta tecnología se puede aplicar posteriormente al análisis general de la estructura del barco de vela. En primer lugar, por la necesidad de analizar las velas como láminas en condiciones de no-linealidad y asumiendo grandes deformaciones/desplazamientos. Y en segundo lugar, porque el uso de esta tecnología se puede aplicar posteriormente al análisis general de la estructura del barco de vela. The new rotation-free thin shell ﬁnite element shall comply with the following requirements: El nuevo elemento ﬁnito de lámina delgada sin ro- taciones debería cumplir los siguientes requisitos: • Non-linear theory for large displacements • Non-linear theory for large strains • Teoría no-lineal para grandes desplazamientos. • Transient dynamic analysis • Teoría no-lineal para grandes deformaciones. • Account for the intense use of anisotropic materials • Análisis dinámico transitorio. • Análisis dinámico transitorio. • Considerar el uso intenso de materiales anisó- tropos. • Interact naturally with the CFD analysis • Interactuar naturalmente con los análisis CFD. Estado del Arte E N EL PRESENTE CAPÍTULO revisaremos el esta- do del arte en cuanto a la tecnología de cálculo de estructuras laminares delgadas que como hemos justiﬁcado en la apartado 1.1 es la más relevante para el cálculo estructural de los componentes de un velero. Ya hemos revisado también en la apéndice A el estado de la práctica relativo al uso de programas en la industria para el cálculo de veleros, tanto en lo que concierne a las velas como en lo que concierne a los cascos. I N THIS CHAPTER we will review the state of the art regarding the calculation technolo- gies for thin shell structures. Thin shell struc- tures are, as we have justiﬁed in section 1.1 the most relevant type for the analysis of the struc- tural behavior of the elements of a sailboat. We have already reviewed in appendix A the state of practice regarding the use of computer pro- grams in the industry to calculate sailboats; both on what relates the sails and on what re- lates the hulls. E N I N El apartado 3.1 presenta los enfoques que abordan el cálculo de velas usando distintas estrategias computacionales. Basándose en una bre- vísima descripción de las características mecánicas de las estructuras de lámina, en el apartado 3.2 se incluye una breve relación de las ventajas y des- ventajas de los elementos basados en la teoría de Kirchhoff-Love frente a aquellos basados en la teo- ría de Reissner-Mindlin para analizar láminas del- gadas y se comentan las principales líneas de in- vestigación recientes que emplean elementos sin ro- taciones para analizar láminas delgadas bajo las hipótesis de Kirchhoff-Love. Section 3.1 introduces the different ap- proaches employed to tackle the computation of sails using various computational strategies. Section 3.2 uses a very brief description of the mechanical characteristics of thin shells to sup- port a short list of advantages and disadvan- tages of the elements based on the Kirchhoff- Love theory with respect those based on the Reissner-Mindlin theory for the analysis of thin shells. The main recent research lines based on the use of rotation-free elements to analyze thin shells under the Kirchhoff-Love assump- tions are also introduced. CHAPTER 3. STATE OF THE ART CHAPTER 3. STATE OF THE ART 20 ments in the mesh. Thus modeling them as elastic chords articulated in every node. aristas de los elementos de la malla; modelizándo- las como cuerdas elásticas articuladas en todos los nodos. An evolution of this method presented by Charvet, Hauville and Huberson [23] consists on dividing the sail response in 2 different steps. The ﬁrst step is used to ﬁnd a start- ing geometry of the sail that balances the wind pressures. In this ﬁrst step there is the as- sumption that the sail tissue has no bending resistance and cannot stretch. The structural response is computed in the second step obtain- ing the sail deformations using a shell model. This method is very similar to the one devel- oped by Ubach in [133] and is summarized in section 1.4 on page 3. Una evolución de este método y propuesto por Charvet, Hauville y Huberson [23] consiste en divi- dir la respuesta de la vela en 2 partes diferenciadas. En un primer paso se calcula una forma inicial de la geometría de la vela que equilibra las presiones ejercidas por el viento. En este primer paso se con- sidera que el tejido de la vela no tiene resistencia a ﬂexión y que es inextensible. En un segundo paso se calcula la respuesta estructural en deformacio- nes de la vela a partir de un modelo de lámina. Este enfoque es muy similar al que desarrolló Ubach en [133] que se resume en el apartado 1.4 en la pági- na 3. An original contribution for sail analysis was made by Pérez del Castillo and García in [21]. Their method consists on considering the structural response of a sail modeled as a mem- brane [123] as an energy minimization prob- lem. And for that they solve the system of equa- tions using an optimization method. This ap- proach is also used by Ortigosa in her PhD the- sis (see [5] and its references) and in the paper by García, Ortigosa and Fernandez [46]. Una propuesta original para el cálculo de velas la hacen Pérez del Castillo y García en [21]. La pro- puesta consiste en considerar la respuesta estructu- ral de la vela modelizada como membrana [123] co- mo un problema de minimización de energía y para ello resolver las ecuaciones mediante un método de optimización. CHAPTER 3. STATE OF THE ART Este enfoque también lo utiliza Orti- gosa en su tesis doctoral (véase [5] y sus referencias) y en el artículo publicado por García, Ortigosa y Fernández [46]. The problem of membrane simulation, and in particular the phenomenon of wrinkling, is studied in depth by Rossi, Lazzari, Vitaliani and Oñate in [111]. However, wrinkling mod- eling in membranes results in a smooth simu- lated geometry instead of a wrinkled one. That is because the model simulates the effect of the wrinkle on the overall shape of the membrane as argue Razenbach and Xu [106]. Another kind of sails, solar sails, do require a detailed analysis of the wrinkles formed in the mirror. Tessler, Sleight and Wang [125] proceed to sim- ulate solar sails modeling them as shells in or- der to capture their wrinkling behavior. El problema de resolución del cálculo de mem- branas, y en particular del tratamiento de las arru- gas, está tratado en profundidad por Rossi, Lazzari, Vitaliani y Oñate en [111]. A pesar de todo, la mode- lización de las arrugas en una membrana conduce a una solución suave de la geometría en vez de una geometría arrugada. Ello se debe a que el modelo de arrugas simula el efecto que tiene la arruga so- bre la geometría global de la membrana, tal y como explican Razenbach y Xu [106]. Tenemos que irnos a otro tipo de velas, las velas solares, para encontrar simulaciones que capturen el detalle de las arrugas que se forman en el espejo. Para conseguirlo, Tess- ler, Sleight y Wang [125] analizan las velas solares como láminas. Finally, scientists from Southampton Uni- versity and INRIA (France) have also sug- gested recently the need to model the sails of boats as thin shells [128]. Finalmente, cientíﬁcos en la Universidad de Southampton y en el INRIA (Francia) han propues- to también recientemente la necesidad de calcular las velas de barcos directamente como láminas del- gadas [128]. With respect to shell analysis, a short re- view of the calculation methods of both shells and membranes was written by Valdés in [134]. For more thorough reviews the author recom- mends the books by Zienkiewicz and Taylor [144] and by Oñate [93]. 3.1 Academic approaches to computer simulation of sails It’s important referring to the works of the dif- ferent schools of thought that have tackled the problem of simulating sails. The most simple modeling of sails is presented by Le Maître, Souza de Cursi and Huberson in [78]. It con- sists on ﬁrst meshing the sail surface and then taking into account only the edges of the ele- Es importante hacer referencia al trabajo de las dis- tintas escuelas de pensamiento que han abordado el problema de simular las velas de los barcos. La mo- delización más simpliﬁcada de velas de barcos la presentan Le Maître, Souza de Cursi y Huberson en [78], y consiste en mallar primero la superﬁcie de la vela y a continuación considerar únicamente las 19 Respuesta mecánica de una lámina Las láminas representan un tipo de estructura que se diferencia sustancialmente del resto de estructu- ras que estamos acostumbrados a ver (y por consi- guiente a entender). Ramm las llama la primadon- na de las estructuras [103, 104], y es que las lámi- nas combinan de manera única la posibilidad de resistir tensiones en su plano y momentos ﬂectores; al tiempo que cubren superﬁcies con formas arbitra- rias. Existen varios tratados clásicos sobre la me- cánica y teoría de láminas. Sin embargo no encaja en el ámbito de esta tesis cubrir dichos contenidos. Simplemente señalaré los aspectos principales que son necesarios para entender la mecánica y cinemá- tica de las láminas para así poder diseñar un ele- mento ﬁnito con éxito. Shells are a structural typology which dif- fers signiﬁcantly from the most common struc- tures we are used to see (and thus to under- stand). The primadonna of structures as called by Ramm [103, 104], shells combine in a unique way the possibility of resisting axial stresses as well as bending moments while spanning sur- faces of arbitrary shape. There are many clas- sical treaties on the mechanics and theory of shells. Therefore, it is not in the scope of this thesis to cover those contents. I will here just pinpoint the main traits that are relevant for understanding the mechanics and kinematics of shells in order to design a ﬁnite element suc- cessfully. Una lámina se deﬁne como una estructura en la que una dimensión es mucho menor a las otras 2 y que puede manifestar curvatura [doble]. Si ana- lizamos la sección de la superﬁcie que representa la lámina, ésta puede resistir cargas en su plano (mecanismo resistente tipo-membrana). Es decir, la lámina ofrece resistencia a fuerzas externas que ac- túan alineadas a la superﬁcie media de la lámina. Pero la lámina también puede resistir momentos ex- ternos actuando en la superﬁcie media de la lámi- na gracias a una distribución no-uniforme de las tensiones a través del espesor (ﬂexión/mecanismo resistente tipo-placa). En lo que atañe a momentos A shell can be deﬁned as a structure in which one dimension is much smaller than the other two and which can exhibit [double] cur- vature. Considering the section of the surface, a shell can exhibit a membrane-like load carry- ing mechanism. Elementos ﬁnitos de lámina delgada En los capítulos 1 y 2 ya se trata en detalle la necesi- dad de modelizar las velas como estructuras de lá- mina delgada. A continuación vamos a revisar bre- vemente los métodos de elementos ﬁnitos desarro- llados para realizar cálculos de este tipo, pero antes el lector debe entender la naturaleza de las láminas en tanto que estructuras y como éstas responden a cargas externas. The need for modeling sails as thin shell struc- tures has been discussed in detail in chapters 1 and 2. Next we will review brieﬂy what are the ﬁnite element methods developed to per- form this kind of calculations, but before that the reader needs to understand the nature of shells as structural elements and how they re- spond to external loads. 3.2. THIN SHELL FINITE ELEMENTS the case of Stolarski et al. [121], MacNeal [73], Yang et al. [141] and Gal and Levy [45]. the case of Stolarski et al. [121], MacNeal [73], Yang et al. [141] and Gal and Levy [45]. the case of Stolarski et al. [121], MacNeal [73], Yang et al. [141] and Gal and Levy [45]. miendan los libros de Zienkiewicz y Taylor [144] y de Oñate [93]. Autores que han realizado distintas revisiones desde una perspectiva histórica son Sto- larski et al. [121], MacNeal [73], Yang et al. [141] y Gal y Levy [45]. perpendiculares a la superﬁcie media de la lámina, perpendiculares a la superﬁcie media de la lámina, CHAPTER 3. STATE OF THE ART Some authors have done reviews from a historical viewpoint; that’s En lo que respecta al cálculo de láminas, una breve revisión de métodos de cálculo tanto de lámi- nas como de membranas fue realizada por Valdés en [134]. Para revisiones más exhaustivas se reco- 21 3.2. THIN SHELL FINITE ELEMENTS CHAPTER 3. STATE OF THE ART This encyclopædic effort cannot be thanked enough by those students who want to dig in the ﬁeld of thin shell structures and use a single point of reference to start with. Flügge [44]: Este es otro libro excelente escrito por uno de los ingenieros más relevantes de la magníﬁca escuela alemana. Ventsel y Krauthammer [135]: Para una publi- cación más contemporánea, enfocada especíﬁ- camente en láminas delgadas y que contiene una cantidad ingente de información y refe- rencias para la seguir la evolución histórica de la comprensión de las láminas, recomien- do sin ninguna reserva esta obra. Este esfuer- zo enciclopédico no se puede agradecer lo su- ﬁciente por parte de aquellos estudiantes que deseen profundizar en el campo de las lámi- nas delgadas y quieran usar un único punto de referencia para empezar. Ventsel y Krauthammer [135]: Para una publi- cación más contemporánea, enfocada especíﬁ- camente en láminas delgadas y que contiene una cantidad ingente de información y refe- rencias para la seguir la evolución histórica de la comprensión de las láminas, recomien- do sin ninguna reserva esta obra. Este esfuer- zo enciclopédico no se puede agradecer lo su- ﬁciente por parte de aquellos estudiantes que deseen profundizar en el campo de las lámi- nas delgadas y quieran usar un único punto de referencia para empezar. Respuesta mecánica de una lámina That is, the shell will oppose a resistance to exterior forces acting aligned to the midsurface of the shell. Also, it can resist external moments acting on the shell’s midsur- face via a non-uniform distribution of stresses across the thickness (bending/plate-like load carrying mechanism). As for external moments orthogonal to the shell’s midsurface, the shell resists exhibiting in-plane shear stresses. Fi- nally, regarding forces perpendicular to the perpendiculares a la superﬁcie media de la lámina, CHAPTER 3. STATE OF THE ART CHAPTER 3. STATE OF THE ART 22 la lámina los resiste en la forma de tensiones de cor- tante en el plano. Y por último, en el caso de fuerzas perpendiculares a la superﬁcie media de la lámina, estas fuerzas generarán una combinación de tensio- nes axiales en el plano de la lámina (uniformes y no-uniformes) —de manera que se activan los mo- dos de membrana y de ﬂexión de la lámina— según las características de curvatura de la superﬁcie de la lámina. Son todos estos mecanismos resistentes —y particularmente sus combinaciones—, los que conﬁeren a las láminas una gran rigidez aún a pe- sar de su esbeltez. Esta es una característica suma- mente apreciada por los ingenieros de estructuras. shell’s midsurface, these will generate a combi- nation of in-plane axial stresses (uniform and non-uniform) —activating both the membrane and bending modes— according to the curva- ture characteristics of the shell. All these re- sisting mechanisms —and in particular their combinations— a shell is capable of, confer them with their highly regarded stiffness yet lightness so valuable for structural engineers. For a complete review (if anyone can ever claim completeness in any given ﬁeld of knowl- edge) the reader can refer to the following ref- erences: Para una revisión completa (si es que alguien puede hablar de completitud en cualquier campo del conocimiento) el lector puede acudir a las si- guientes referencias: Timoshenko [127]: Probably the most cited book for classic analysis of plates and shells. Flügge [44]: This is another excellent book by one of the most relevant engineers of the ﬁne tradition of German scholars. Timoshenko [127]: Se trata probablemente del li- bro de mecánica clásica de placas y láminas más citado de cuantos existen. Ventsel and Krauthammer [135]: For a more contemporary publication focused speciﬁcally on thin shells and containing a wealth of information and references to the historic understanding of shells, I unreservedly recommend this work. This encyclopædic effort cannot be thanked enough by those students who want to dig in the ﬁeld of thin shell structures and use a single point of reference to start with. Ventsel and Krauthammer [135]: For a more contemporary publication focused speciﬁcally on thin shells and containing a wealth of information and references to the historic understanding of shells, I unreservedly recommend this work. láminas delgadas son las siguientes: láminas delgadas son las siguientes: • El espesor de la lámina se mantiene constante después de la deformación. • A straight ﬁber perpendicular to the shell midsurface remains straight after the de- formation. • Las ﬁbras rectas y perpendiculares a la super- ﬁcie media de la lámina se mantienen rectas después de la deformación. • A straight ﬁber perpendicular to the shell midsurface remains perpendicular after the deformation. • Las ﬁbras rectas y perpendiculares a la super- ﬁcie media de la lámina se mantienen perpen- diculares después de la deformación. The ﬁrst ﬁnite elements based on the Kirchhoff-Love hypotheses for thin shells had to face the challenge of maintaining C1 con- tinuity across different elements. This condi- tion arises from the need to compute the curva- ture of the shell’s midsurface as this is essential in order to reproduce the bending mode of the shell. More precisely the challenge is to main- tain G1 continuity, which is a relaxation over C1, but still hard for ﬁnite elements to comply across element boundaries. This condition re- sulted in overly complex elements. A good ex- ample is the tri-cubic triangle by Clough and Tocher [27, 37, 73]. Los primeros desarrollos de elementos ﬁnitos de lámina delgada basados en las hipótesis de Kirchhoff-Love tenían que enfrentarse al desafío de mantener continuidad C1 entre los elementos. Este condicionante surge de la necesidad de calcular la curvatura de la superﬁcie media de la lámina; ya que esta magnitud es esencial para reproducir el modo de ﬂexión de la lámina. Siendo más exactos, el desafío consiste en mantener continuidad G1, que es una relajación con respecto a la continuidad C1, pero aún complicada de imponer por los elementos ﬁnitos entre elementos. De hecho, este condicionante dio lugar a elementos enormemente complejos. Un buen ejemplo de ello es el elemento triangular tri- cúbico de Clough y Tocher [27, 37, 73]. The use of the Reissner-Mindlin hypothe- ses [81, 107, 108] soon gathered popularity among those scholars developing new shell ﬁ- nite elements. Reissner developed a theory for plates and shells dropping the third hypothesis of those stated above. The objective is to try to increase the range of validity of the theory also for thick shells. It was clear that as the thick- ness of the shell increases, the third hypoth- esis is no longer valid. 1The reader can ﬁnd a full chapter devoted to the dis- cussion of shear locking and the advantages/disadvantages of Reissner-Mindlin elements in [93]. Un vistazo rápido al desarrollo de elementos ﬁnitos de lámina delgada Modern theories for thin shells are based on Kirchhoff’s hypothesis [65] for the kinematics of plate bending and translated into shells by Love [72]. Many theories reﬁning or improving those hypotheses have been formulated to solve classical problems of shells. The Kirchhoff-Love kinematic hypotheses for thin shells are: Las teorías modernas de láminas delgadas se desa- rrollan a partir de las hipótesis de Kirchhoff [65] sobre la cinemática de la ﬂexión de placas y apli- cadas por Love para láminas [72]. Han sido mu- chos los autores que han aportado diversas teorías que reﬁnan y mejoran en uno u otro aspecto las teo- rías para resolver problemas clásicos de láminas. Las hipótesis cinemáticas de Kirchhoff-Love para 3.2. THIN SHELL FINITE ELEMENTS 23 • The thickness of the shell remains con- stant after the deformation. láminas delgadas son las siguientes: CHAPTER 3. STATE OF THE ART 24 que conduce a lo que se denomina bloqueo por cor- tante1. Algunas de las estrategias que se aplicaron para resolver este problema es la integración redu- cida [selectiva] o los campos de deformación condi- cionados. A pesar de que en un inicio se pensó que estas técnicas servían para resolver los problemas de bloqueo por cortante, hoy en día se sabe que sólo sirven para aliviar el problema parcialmente [35]. A parte, existen muchas otras diﬁcultades asociadas. Para una revisión histórica completa de los desafíos confrontados y las diﬁcultades encontradas por los cientíﬁcos que desarrollan elementos ﬁnitos de lá- mina, el lector puede acudir a las siguientes referen- cias y las contenidas en ellas: [73, 141]. Es oportuno recordar que han pasado 50 años desde el desarro- llo de los primeros elementos ﬁnitos de lámina. que conduce a lo que se denomina bloqueo por cor- tante1. Algunas de las estrategias que se aplicaron para resolver este problema es la integración redu- cida [selectiva] o los campos de deformación condi- cionados. A pesar de que en un inicio se pensó que estas técnicas servían para resolver los problemas de bloqueo por cortante, hoy en día se sabe que sólo sirven para aliviar el problema parcialmente [35]. A parte, existen muchas otras diﬁcultades asociadas. Para una revisión histórica completa de los desafíos confrontados y las diﬁcultades encontradas por los cientíﬁcos que desarrollan elementos ﬁnitos de lá- mina, el lector puede acudir a las siguientes referen- cias y las contenidas en ellas: [73, 141]. Es oportuno recordar que han pasado 50 años desde el desarro- llo de los primeros elementos ﬁnitos de lámina. the strategies developed to overcome this prob- lem are reduced [selective] integration and as- sumed strain ﬁelds. Although these techniques were thought initially to solve the shear lock- ing problems, nowadays it is known that they only partially alleviate the problem [35]. Be- sides, there are many more challenges associ- ated. For a complete historic review of the chal- lenges and difﬁculties faced by scholars in the development of shell ﬁnite elements, the follow- ing references and the ones contained therein are recommended: [73, 141]. It is important to note that 50 years have passed by since the development of the ﬁrst shell ﬁnite elements. 1El lector encontrará un capítulo entero dedicado al tema del bloqueo por cortante y las ventajas en inconvenientes de los elementos de Reissner-Mindlin en [93]. láminas delgadas son las siguientes: Translated into ﬁnite elements, the Reissner-Mindlin hypotheses re- quire to take into account the rotations of the transverse ﬁbers as additional degrees of free- dom. But maybe the main reason for the popu- larity of Reissner-Mindlin elements is that only C0 continuity is required. However, the schol- ars soon found out that using the Reissner- Mindlin hypothesis had numerical drawbacks. By including the rotations as independent de- grees of freedom, the elements did not have control over how much energy the shear mode absorbs as the thickness tends to 0; leading to what is known as shear locking1. Some of Ante esta diﬁcultad los cientíﬁcos que desarro- llaban nuevos elementos de lámina popularizaron rápidamente el uso de las hipótesis de Reissner- Mindlin [81, 107, 108]. Reissner desarrolló una teo- ría de placas y láminas en las que obviaba la tercera de las hipótesis indicadas anteriormente. La inten- ción de Reissner era aumentar el rango de validez de la teoría para poder incluir láminas gruesas. Re- sultaba evidente que a medida que aumentaba el espesor de la lámina, la tercera hipótesis dejaba de ser válida. Al trasladarlas a los elementos ﬁnitos, las hipótesis de Reissner-Mindlin implican tomar en cuenta las rotaciones de las ﬁbras transversales como nuevos grados de libertad. Pero la principal razón por la que los elementos de Reissner-Mindlin obtuvieron tanta popularidad fue seguramente por- que ya no se requería continuidad C0. Sin embargo los cientíﬁcos pronto se percataron de que usar las hipótesis de Reissner-Mindlin tenía efectos numéri- cos indeseados. Al incluir las rotaciones como gra- dos de libertad independientes, los elementos dejan de discernir cuanta energía debe absorber el modo de cortante a medida que el espesor tiende a 0; lo Rotation-free elements as a resort to develop thin shell elements Recientemente, el interés por la resolución adecua- da de las láminas delgadas modelizadas con la teo- ría de Kirchhoff-Love ha experimentado un resurgi- miento con la aparición de los elementos sin rotacio- nes. Gärdsback y Tibert ofrecen una comparación de varios de estos métodos en [49]. En dicho artículo se presentan las ventajas que ofrecen los elementos de lámina delgada sin rotaciones frente a otras fami- lias de elementos. De entre ellas destaco las siguien- tes: Recently, the interest for using Kirchhoff-Love hypotheses has picked-up with the advent of rotation-free elements. Gärdsback and Tibert offer a comparison of various elements of this kind [49]. In their paper they discuss the ad- vantages offered by the rotation-free shell el- ements with respect to other families of thin shell elements. The following are highlighted: • Reduced number of degrees of freedom in the model. • Menos grados de libertad en el modelo. simplifying the application of boundary conditions, which may be cumber- some in some cases [142]). • La conexión de los elementos de lámina sin rotaciones a otras tipologías de elementos no siempre es trivial. En la presente tesis se procura resolver o minimizar los inconvenientes encontrados por Gärdsback y Ti- bert y otros inconvenientes encontrados en trabajos anteriores, como por ejemplo simpliﬁcar la aplica- ción de las condiciones de contorno, que puede ser rebuscado en algunos casos [142]). Aside from the family of rotation-free shell elements developed by Oñate et al. [41, 95, 96] other element families are cited in the ref- erences of [49]. In fact, there are not many rotation-free shell elements in the literature. And very few have been applied for geomet- rically nonlinear problems. The reader is re- ferred on the one side to the works by Gärds- back and Tibert [49], Zhou and Sze [142], Lin- hard et al. [71], and Oñate, Flores and Zárate [41, 95, 96] and references therein. On the other side, Cirak, Ortiz et al. [24] have devel- oped a new paradigm to simulate rotation-free shells. This paradigm is based on the applica- tion of developments in the ﬁeld of computer graphics, so that mesh subdivision algorithms are used to obtain new meshes suitable to inter- polate a cubic function in regular patches of el- ements. Finally, a new trend with a broad sup- port has emerged from the group led by T.J.R. Hughes based on the paradigm of isogeometric analysis [56]. The ﬁrst application of the iso- geometric analysis to the simulation of shells is the one by Benson [11] using the Reissner- Mindlin hypotheses. However, a great advan- tage of the isogeometric analysis is the possibil- ity of representing the geometry with C∞conti- nuity. Therefore, it should now be possible to achieve C1 continuity in the geometric repre- sentation of shells and use the Kirchhoff-Love Aparte de la familia de elementos de lámina sin rotaciones desarrollada por Oñate y colaboradores [41, 95, 96] se pueden encontrar otras familias de elementos en las referencias de [49]. Aunque de he- cho no existen muchos elementos de lámina sin ro- taciones en la literatura. Y de ellos, muy pocos se han llevado a problemas geométricamente no linea- les. • Menos grados de libertad en el modelo. • Menos grados de libertad en el modelo. • As the span/thickness ratio increases, us- ing rotation-free elements results in sys- tem matrices with a lower condition num- ber than if regular shell elements with ro- tations were used. • A medida que la relación de esbeltez de la lá- mina se acentúa, la aplicación de los elemen- tos de lámina sin rotaciones resulta en siste- mas matriciales mejor condicionados que si aplicamos elementos con grados de libertad de rotaciones. • The same applies when the meshes are reﬁned. • Sucede lo mismo a medida que se reﬁnan las mallas de cálculo. • 3D rotations don’t have the commutative property. Moreover, rotations are difﬁcult to derive for large displacements applica- tions. • Las rotaciones carecen de la propiedad con- mutativa en el espacio 3D. Más aún, en aplica- ciones con grandes desplazamientos, las gran- des rotaciones son difíciles de derivar. With respect to the disadvantages mentioned in [49]: • The precision of the membrane response of the element becomes limited. This is because the elements reviewed are using 3.2. THIN SHELL FINITE ELEMENTS 25 low order interpolations to describe the membrane kinematics. low order interpolations to describe the membrane kinematics. low order interpolations to describe the membrane kinematics. En cuanto a las desventajas que se anuncian en [49]: low order interpolations to describe the membrane kinematics. En cuanto a las desventajas que se anuncian en [49]: • Limitación de la precisión a la precisión de la parte membranal del elemento. Esto se debe a que los elementos revisados usan interpolacio- nes de bajo orden para describir la cinemática asociada al modo de membrana. • The computation of the curvature of the element is sensible to mesh distortion. • The bandwidth of the stiffness matrix in- creases because now the connectivities are increased in order to compute the cur- vatures. • Sensibilidad a la distorsión de la malla para la parte de ﬂexión del elemento. • Connecting rotation-free elements to other element typologies is not always trivial. • Aumenta el ancho de banda de la matriz de ri- gidez debido a que para calcular las curvatu- ras es preciso incrementar las conectividades. The present thesis tackles and tries to solve the problems encountered by Gärdsback and Tib- ert, and other additional issues found in pre- vious efforts (i.e. • Menos grados de libertad en el modelo. El lector puede acudir por un lado a los traba- jos de: Gärdsback y Tibert [49], Zhou y Sze [142], Linhard, Wüchner y Bletzinger [71], Oñate, Flores y Zárate [41, 95, 96] y las referencias contenidas en ellos. Y por otro lado Cirak, Ortiz y otros colabora- dores han desarrollado un nuevo paradigma para el cálculo de láminas sin rotaciones [24]. Este pa- radigma se basa en aplicar desarrollos propios del campo de la gráﬁca computacional, de modo que usando algoritmos de subdivisión de mallas se ob- tienen nuevas mallas que son útiles para interpo- lar una función cúbica sobre parcelas regulares de elementos. Finalmente, otra nueva tendencia surge desde el grupo liderado por Tomas J.R. Hughes y que cuenta con un seguimiento muy amplio. La idea consiste en el paradigma del análisis isogeométri- co [56]. La primera aplicación del análisis isogeo- métrico a la simulación de lámina es la realizada CHAPTER 3. STATE OF THE ART 26 por Benson [11] utilizando las hipótesis de Reissner- Mindlin. No obstante, una de las grandes ventajas del análisis isogeométrico consiste en la posibilidad de representar la geometría de cálculo con continui- dad C∞. Por ello, debería ser posible alcanzar conti- nuidad C1 en la representación geométrica de las lá- minas y usar las hipótesis de Kirchhoff-Love. El pri- mer grupo en conseguirlo ha sido Kiendl, Bletzinger et al. [63]. Sin embargo, todavía existen aspectos sin zanjar; como por ejemplo la asignación de condicio- nes de contorno a bordes deﬁnidos por superﬁcies trimadas [102]. Una revisión más profunda de los elementos de lámina sin rotaciones con sus ventajas e inconvenientes se encuentra en el apartado 4.1.1. assumptions. The ﬁrst group to achieve it are Kiendl, Bletzinger et al. [63]. However, philo- sophical issues with isogeometric analysis re- main still unsettled. For example, there is not a direct way to apply boundary conditions on boundaries deﬁned by trimmed surfaces [102]. A more in depth review of the rotation-free el- ements with their advantages and disadvan- tages is provided in section 4.1.1. Chapter 4 Ideas for a new rotation-free shell element I N THIS CHAPTER I will introduce the core ideas on which the new ele- ment is based. I will provide evidences of the ﬁrst failures as well as explanations for those. The objective of this chapter is to introduce the reader into the difﬁculties entrenched and provide the foundations to un- derstand the solution adopted in chapter 5. I N The following sections are based on the work reported in [132] pub- lished by Ubach and Oñate, and provide a broader introduction to the work than the one reported in the paper itself. In the following pages the author presents the original ideas that sup- port the development of the new shell element formulation. The triangle presented in this chapter is ﬂawed: it exhibits hourglassing. That is, spu- rious or zero energy modes that affect the solution. In spite of it, this chapter serves to present the fundamental ideas that support the advan- tages of the ﬁnal formulation: simplicity and generality. Simplicity, be- cause the formulation is free from rotation degrees of freedom. Generality, because the triangle here presented can be used regardless of the mesh topology, thus generality is conserved for any mesh-represented surface (see the topology requirements by other previous rotation-free shell ele- ments in the literature in section 4.1.1). The element uses the neighboring elements (sharing the nodes) in order to enrich the information about the normals, but it does not require them nor a minimum number of them to perform the computations. Section 4.1 presents the design requirements for the new element and the different strategic decisions made to build the new element. These de- cisions will hold throughout this thesis development. Section 4.2 presents the geometric principle that allows computing the curvature as a prod- uct of ﬁrst derivatives while deﬁning the normal orientation of the shell surface in a continuous way, and thus opening a door to overcome the C1 condition. Section 4.3 develops this idea into a ﬁrst concept triangular element. This development already shows the use of the design guide- lines introduced in section 4.1. In section 4.4 the author reports several attempts to control the spurious energy modes detected in the previous section. Finally, in section 4.5 the author reﬂects on the root cause for the 27 CHAPTER 4. Chapter 4 Ideas for a new rotation-free shell element IDEAS NEW SHELL ELEMENT 28 zero energy modes and presents the path that will ultimately lead to the successful development of a new rotation-free shell element. Clough-Tocher interpolant This element is included here for being the ﬁrst thin shell element to sat- isfy the C1 continuity condition. It was presented in 1965 [27]. It is a complicated element because it is a triangle split into three sub-triangles. Each sub-triangle is cubic and some degrees of freedom are restrained in order to maintain C1 continuity across the sub-triangles. In order to achieve slope continuity across the macro triangles, it uses the rotations at the nodes as degrees of freedom, but also the cross boundary derivatives at the center of the sides. 4.1 Design requirements for the new shell element It is common to specify a number of design requirements for any new shell element. It is no different in this case. The motivation already establishes some hard conditions. First of all, the new element should combine eas- ily with other elements solving different physics like ﬂuids, but also other structural typologies like beams or solids. Also, it is required that the ele- ment solves geometrically nonlinear problems. The dynamics of the shell should also be taken into account. Finally, many modern shell structures are made out of composite materials. This renders them highly anisotropic properties. 4.1.1 Going rotation-free The ﬁrst and most important design decision made early on has been to opt for a rotation-free element. Avoiding the use of rotations ensures parting from the possibility of suffering shear locking. It also has the advantage of not having to deal with the difﬁcult topic of ﬁnite rotations in the geomet- rically nonlinear regime. Chapter 3 provides an overview of the different rotation-free shell element approaches adopted to the date. Let’s analyze them in more detail in order to draw some conclusions for this new devel- opment. We classify the different methods in three main families. We will only include in this review the most contemporary developments, except for the historic case of the Clough-Tocher interpolant. Subdivision paradigm This is an original development proposed by Cirak, Ortiz et al. [24–26]. They propose to take advantage of subdivision algorithms developed for computer graphics applications to generate regular meshes. They start with a coarse mesh generated using conventional methods. Then, they use subdivision algorithms for surfaces that guarantee C1 continuity in the limit for all the set of points generated. The practical application for using these subdivision algorithms is that all the new points gener- ated are regular points. A point in a mesh of triangles is regular if it is surrounded exactly by 6 triangles. Using this property, the authors gen- erate cubic representations of the surface using patches of 13 elements (12 nodes). Following this strategy the descriptions for two adjacent tri- angles are C1 continuous. This approach, however, has some drawbacks. First, those vertices of the original mesh which are irregular, remain ir- regular throughout the subdivision process. Therefore, there are regions of the surface for which the new paradigm does not provide a solution. It is true, though, that those regions can be made arbitrarily small by means of successive subdivision steps. Nevertheless, this constitutes a very strong topology requirement. Furthermore, the subdivision process requires the application of edge detection algorithms in order to prevent smoothing of the structure’s surface. One of the main contributions of these works is to show that it is possible to generate cubic representations of a surface represented by a tessellation of linear triangles. Most importantly, these works open the path for ﬁnally attaining C1 continuity. 4 element patches This family involves all those elements that overcome the C1 continuity condition by computing the curvature over an element using a patch of 4 elements (the main element and the 3 elements sharing one edge). This family includes the basic shell triangle (BST) elements developed by Oñate et al. [41, 95, 96], the formulations by Phaal and Calladine [100], Zhou and Sze [142], Sabourin and Brunet [112] and many other formulations re- viewed in depth in the works by Gärdsback and Tibert [49] and by Oñate in [92]. These formulations perform a computation of the average cur- vature in the element. Either they compute a constant two-dimensional 4.1. DESIGN REQUIREMENTS 29 curvature taking into account all 4 elements in the patch, or more usu- ally, they compute a direction-wise curvature for every pair of triangles. It must be noted that most of these shell elements are an evolution of their thin plate formulation counterparts to which a membrane formulation has been added in order to analyze shell structures. This is the case for ex- ample with the BST, which evolved from the basic plate triangle (BPT) designed by Oñate and Cervera [94]. Indeed, this is a simpliﬁcation and can work when the curvature is small or when the mesh is sufﬁciently ﬁne. All these formulations represent a sincere effort to make a rotation- free element as simple as possible. But in many cases it can lead to errors as reported. There is also a topology requirement in that an element must be surrounded by 3 others. Most authors propose to drop this requirement at the boundary, but [142] maintains it and creates ghost slave elements to fulﬁll the condition. Some membrane locking issues are also reported which are partially alleviated when using a quadratic interpolation for the membrane behavior of the element and when the mesh is reﬁned. 1The author considers prof. Bazilevs a scholar of ICES although his afﬁliation in this work is the University of California. Discontinuous Galerkin methods This methodology has become popular in a number of ﬁelds in computa- tional mechanics and also in ﬂuid dynamics problems. The method con- sists on imposing the continuity of the unknown variables in a weak form. Thus, some authors have found it suitable for enforcing the C1 continuity condition across elements and avoid the need to use rotational degrees of CHAPTER 4. IDEAS NEW SHELL ELEMENT 30 freedom. Güzey, Stolarsky, Cockburn and Tamma [50] propose a discon- tinuous Galerkin formulation to build a new bilinear quadrilateral shell element. This method also depends on the selection of a set of arbitrary constants that modify the behavior of the function jumps in the discontin- uous ﬁelds. Later, in separate works by Dung and Wells [34] and by Noels and Radovitzky [86] have proposed a discontinuous Galerkin approach to build quadratic and bicubic thin shell elements. The problem with the dis- continuous Galerkin method is that in general, by relaxing the continuity requirements, they lose accuracy. It is also cumbersome to implement and computationally expensive because both surface integrals and boundary integrals are required—the latter in order to apply the weak form of the continuity condition—. Isogeometric analysis (IGA) The last approach is the one proposed by Hughes, Cottrell and Bazilevs [56] and implemented for thin shells by Kiendl, Bletzinger et al. [63]. This approach proposes the use of the non-uniform rational B-Spline (NURBS) description of surfaces that is commonly used in computer aided geomet- ric design (CAGD). This approach has many advantages over conventional ﬁnite element formulations. The ﬁrst one is to avoid the meshing step, which is usually complex and time consuming. Then, the use of NURBS functions allows to attain solutions with C∞continuity, attaining the high- est orders of convergence with the least number of elements used. How- ever, this approach also has some drawbacks. The ﬁrst one is having to deal with trimmed surfaces in the geometric description. This is not a big issue in computer aided design (CAD) because trimming NURBS surfaces does not have an impact on visualization. For computer aided engineering (CAE), though, it is paramount to solve this issue. Schmidt, Wüchner and Bletzinger [114] have proposed a solution for this issue. Let’s review for a moment the efforts made to overcome the chal- lenge posed by trimmed surfaces to measure the magnitude and impor- tance of the problem. This has been addressed separately by the group at ICES (University of Texas at Austin)1 in collaboration with Sederberg reformulating Isogeometric Analysis using T-Splines [8]. Also the group at Technische Universität München (Germany) have formulated a clever approach proposed by Schmidt, Wüchner and Bletzinger [114] to embed trimmed surfaces into the Isogeometric Analysis formulation. This topic had also been addressed previously by Kim, Seo and Youn from KAIST (Korea) splitting trimmed elements using triangles [64]. Besides, other drawbacks of the isogeometric approach are that it de- parts from the traditional conception of structural analysis in which the points of analysis are material points in the structure. Instead, here the points used for the analysis are control points of the geometry. This re- quires an important effort regarding the pre- and post-processing tools. Furthermore, dealing with NURBS functions, has intrinsic smoothing prop- erties that brings some precision advantages, but which are also more com- plex to deal with than the traditional Lagrangian shape functions used 4.1. DESIGN REQUIREMENTS 31 traditionally in isoparametric analysis. 4.1.2 Implicit vs Explicit time integration Another of the important design decisions was about the time integration scheme. On the one side, it is common for new ﬁnite elements to opt for an explicit time integration scheme. This avoids the difﬁcult task of having to derive the tangent stiffness matrix of the element and the tangent ma- trix of the loads on the element, and invert them. Explicit ﬁnite elements solve ﬁrst the accelerations of the system taking into account the state of equilibrium of the previous time step and the inertiæ (mass matrix) of the system. The positions are computed by forward integration. Then, the new equilibrium is calculated and a new step starts. The explicit time integra- tion scheme is conditionally stable and therefore very small time steps are required. The advantage of this scheme is that usually a lumped mass matrix is used, which turns the system of equations matrix into a diagonal matrix, and thus trivial to solve and very fast. They are also very robust, albeit their accuracy is conditional. The main disadvantage though, is that only dynamic problems can be solved. The static problems are treated solv- ing the quasi-static problem and considering it solved when the velocities and accelerations are close to zero. On the other side, implicit integration schemes offer unconditional sta- bility, which means that the time step can be arbitrarily set as large as wanted. Usually to the time scale resolution of the problem at hand. When the time scale resolution of interest is similar to the conditional time step of the explicit time integration scheme, an implicit scheme is not competi- tive. However for most engineering problems this is not the case and im- plicit schemes are very competitive for most cases. This scheme requires to invert the tangent stiffness matrix of the element and therefore solving a nontrivial system of equations. In an implicit scheme the mass matrix needs not be anymore lumped (as in the case with explicit time integra- tion schemes) because we already face the challenge of solving a nontrivial system of equations. This offers another opportunity for maintaining ac- curacy in the analysis. Another advantage usually claimed in favor of explicit time integration schemes is that they are more accurate when solving problems involving shock waves. In fact, what happens is that the discretization inherent in the ﬁnite element method introduces numerical diffusion in the solution. Isogeometric analysis (IGA) Finally, by directly using the CAD geometry for performing the mechanical analysis, the method is subjected to the CAD deﬁnition of the geometry, which may not be well suited for simulation purposes (i.e. it may happen that the geometry features far more geometry patches than actually necessary for the simulation, which can become a burden). At this point, it is relevant to cite the pioneering work of Rho and Cho [109, 110]. Rho and Cho proposed originally to blend the CAD description of the geometry with the ﬁnite element analysis for shell structures. Their work is independent of the developments in Isogeometric Analysis (IGA) but has not gained the momentum that isogeometric analysis has created. Table 4.1: Main advantages and disadvantages of the different rotation- free ﬁnite element technologies in the literature. Table 4.1: Main advantages and disadvantages of the different rotation free ﬁnite element technologies in the literature. Rotation-free element family Advantages Disadvantages Clough-Tocher interpolant •Solves C1 continuity. •Very sophisticated. •Needs the transverse derivatives at the edges. 4 element patches •Simplicity. •Linear elements are used. •Boundary conditions sometimes complicated. •Membrane locking. •Precision issues. Subdivision paradigm •C1 continuity achieved. •Linear elements are used for input. •Strong topological re- quirement. •Complex process. Discontinuous Galerkin •C1 continuity is weakly achieved. •Tuning required. •Requires surface and boundary integrals. Isogeometric analysis •Continuity achieved natu- rally. •Inherits good approxima- tion properties of NURBS functions. •Does not need to generate a mesh. •Complexity. •Has to deal with trimmed functions. •Dependency on CAD de- scription. Table 4.1 presents a summary of the advantages and disadvantages brought by each of the approaches reported. One advantage that is com- mon to all rotation-free shell elements is that by only having displacement degrees of freedom, they are reducing the number of degrees of freedom in the system. Therefore, it will be a priority not to undermine this advantage by adding degrees of freedom to the element unless absolutely necessary. It has been shown that each approach has merits and drawbacks. The effort undertaken in the present thesis is to try to gather the experience of all these previous works and combine their strengths while at the same time avoid or provide solutions to tackle the drawbacks. CHAPTER 4. IDEAS NEW SHELL ELEMENT 32 2It was reported in [96] that the nodal implementation of the BST, dubbed basic shell node (BSN), was marginally more precise than the elemental implementation. This is taken as an advice and a new way to better exploit the enriched information is sought. 4.1.3 Advantages of the total Lagrangian formulation The last main design decision was to develop all the formulation in the total Lagrangian framework. The total lagrangian framework describes the deformation with respect to the reference conﬁguration. Whereas the updated lagrangian framework describes the deformation with respect to the previous equilibrium conﬁguration in the analysis. The total Lagrangian framework equations are very similar to their counterparts in the updated Lagrangian framework. The main advantage of the total Lagrangian framework is that it can yield some computational savings if there are magnitudes that can be precomputed in the reference conﬁguration and stored throughout the analysis. On the other hand, the updated Lagrangian framework allows over- coming a problem encountered by the total Lagrangian framework in anal- ysis involving large rotations when angles become larger than 2π. In [85] Mäkinen also claims that the updated Lagrangian framework yields a fully symmetric tangent matrix for conservative loads. The decision was to develop the formulation in the total Lagrangian framework. The main reasons are to take advantage of the precomputa- tion of magnitudes in the reference conﬁguration and the lack of rotational degrees of freedom in the formulation as explained in section 4.1.1. With respect to the symmetry/asymmetry of the tangent matrix, the author did not consider this a deciding factor, since a wide range of loadings are to be applied to the element, both conservative and non-conservative. A clear example of the beneﬁt of using the total Lagrangian framework is shown in section 7.2, where the author shows the possibility of prein- tegrating the expression of the element’s Jacobian —maintaining all the terms of the polynomial— through-the-thickness of the shell. This is an original contribution for rotation-free shell elements, it is also the ﬁrst time it is applied to a shell element in a total Lagrangian formulation and had only been done before in the updated Lagrangian formulation by Stan- ley [120]. 4.1.2 Implicit vs Explicit time integration This effect is more severe in implicit time integration schemes than in explicit schemes. In this case both methods beneﬁt from the use of higher order formulations because they enrich the solution space. For this development, the author decided to opt for an implicit time integration scheme. Having weighted the advantages and disadvantages of both approaches, it seems that the implicit scheme represents a short term investment (in terms of development effort) that will yield many long term beneﬁts. In addition, it is always possible to easily develop the ex- plicit version of the element for those problems whose time scale resolution comes close to the conditional time step requirements. Another advantage in terms of development is that the implicit scheme forces to perform many more checks than the explicit scheme. It would have been harder to iden- tify many of the problems encountered in the development of this research work if the tangent stiffness matrix had not been taken into account and COMPUTING CURVATURES 33 4.2. only dynamic analysis were considered. 4.2 Computing curvatures without second derivatives Several attempts have been made to overcome the C1 condition that thin- shell ﬁnite elements need to satisfy in order to avoid spurious modes [24, 56, 63, 73, 93, 121, 143]. This work builds on the original proposal for rotation-free shell elements reported in [41, 93, 95, 96], and extends those principles in order to gather richer geometrical information from the patch of all the elements—not only those sharing one side—surrounding a given triangle in the mesh.2 This allows to avoid the need for computing an av- eraged curvature for the element and instead we can obtain precise values CHAPTER 4. IDEAS NEW SHELL ELEMENT 34 for the curvature at any point in the triangle, in particular at the Gauss points. The original approach is based on a linear interpolation (i.e. using lin- ear basis functions). The curvatures are computed using the gradient of the surface normal at each element. In all cases simplicity is of the utmost importance when dealing with boundary conditions. Boundary conditions are managed very easily, both for Dirichlet and Neumann conditions. The normal vector to the surface at every node accounts for this simplicity without the need to use additional degrees of freedom nor virtual nodes. In the following we write the basic principles on which the formulation is based. Let [67, 122]: Let [67, 122]: X ∈S ⊂R3 \| X = (x, y, z)T = ϕ(u, v) = ϕ (4.1) (4.1) ∇ϕ deﬁnes the tangent plane to S as follows: ∇ϕ deﬁnes the tangent plane to S as follows: (∇ϕ)1 ≡ϕ′1 = ∂ϕ ∂u , (∇ϕ)2 ≡ϕ′2 = ∂ϕ ∂v (4.2) (4.2) We deﬁne: We deﬁne: n = ϕ′1 × ϕ′2 (4.3) (4.3) then: n ⊥ϕ′1 , n ⊥ϕ′2 ⇒∇(n · ∇ϕ) = 0 (4.4) ∇(n · ∇ϕ) = ∇n · ∇ϕ + n · ∇∇ϕ = 0 (4.5) ∇n · ∇ϕ = κ = −n · ∇∇ϕ (4.6) (4.4) (4.5) (4.6) (4.6) The means to accomplish the objective of computing the curvature (κ) at speciﬁc locations—using the ﬁrst equality given by equation (4.6)—is to compute the values of the normal directions at each of the three nodes of the triangle (this process is described in section 5.1). Using the values of the normal at the three nodes of the triangle we can build the following vector ﬁeld of normal directions for the triangle and obtain its gradient (∇n): n(ξ, η) = P3 i=1 N i(ξ, η) · ni ∥P3 i=1 N i(ξ, η) · ni∥ (4.7) (4.7) Using this ﬁeld of normal vectors we can express the position of any point of the shell as: x(ξ, η, ζ) = 3 X i=1 N i(ξ, η) · xi + hζ · n(ξ, η) (4.8) (4.8) where the shape functions are represented by N and h stands for the thick- ness of the shell. It must be noted that the shape functions used here are linear Lagrangian functions. 4.3. RAW APPROACH 35 4.3 Raw approach and why it doesn’t work The development of the formulæ follows a continuum-based approach (very similar to the one presented in [123]), which yields the following expres- sion for the Cauchy-Green strain tensor: C = F T · F = j · J−1T · j · J−1 = GT · g · G (4.9) G = J−1 (4.10) g = jT · j (4.11) j = ï∂x ∂ξ , ∂x ∂η , ∂x ∂ζ ò (4.12) J = ï∂X ∂ξ , ∂X ∂η , ∂X ∂ζ ò (4.13) where G is the Jacobian matrix of the inverse function of change of coor- dinates from global (in the reference conﬁguration) to parametric of the element, and g contains the information of the deformed conﬁguration. The reminding symbols are standard in the literature and the reader can refer to the Notation on page xxv. The terms of g can be expressed as: The terms of g can be expressed as: gαβ = 3 X i=1 Li αxiT · 3 X i=1 Li βxi + 2λζ 3 X i=1 Li αxiT · U · 3 X i=1 Li βni + λ2ζ2 3 X i=1 Li αniT · U · U · 3 X i=1 Li βni ∀α, β = ξ, η (4.14) U = 1 ∥P3 i=1 N ini∥·  I − ÄP3 i=1 N iniä ⊗ ÄP3 i=1 N iniäT ∥P3 i=1 N ini∥2   (4.15) (4.14) U = 1 ∥P3 i=1 N ini∥·  I − ÄP3 i=1 N iniä ⊗ ÄP3 i=1 N iniäT ∥P3 i=1 N ini∥2   (4.15) (4.15) where Li α refers to the derivative of the shape function corresponding to node i with respect to the coordinate α, and U contains the information regarding the normalization of the normal vectors. The three terms of equation (4.14) are respectively: the membranal deformation, the bending deformation and the so-called second order deformation terms. Figures 4.1 and 4.2 depict respectively the global and parametric coor- dinates that identify the nodes of the triangles, and the relation between the normals at the nodes and the patch of triangles that surround each node. The relation between strains and stresses in a linear elastic material is established by the constitutive tensor: S′ = D: E′ (4.16) (4.16) which is a fourth order tensor. 4.3 Raw approach and why it doesn’t work But taking advantage of the major and minor symmetries corresponding to the symmetry of the stress and strain which is a fourth order tensor. But taking advantage of the major and minor symmetries corresponding to the symmetry of the stress and strain 36 CHAPTER 4. IDEAS NEW SHELL ELEMENT Figure 4.1: Representation of the global and parametric coordinates used to deﬁne the positions of each of the nodes of the triangle. Figure 4.1: Representation of the global and parametric coordinates used to deﬁne the positions of each of the nodes of the triangle. Figure 4.2: In order to compute the curvature inside the central triangle, we use the information of the patch of elements that surround each node. The shading patterns indicate which triangles contribute to the determi- nation of each normal. Figure 4.2: In order to compute the curvature inside the central triangle, we use the information of the patch of elements that surround each node. The shading patterns indicate which triangles contribute to the determi- nation of each normal. 4.3. RAW APPROACH 4.3. RAW APPROACH 37 tensors—and using Voigt notation—, it can be expressed as a second order tensor. E is the Green-Lagrange strain tensor. S is the second Piola- Kirchhoff stress tensor. And the ′s indicate that the strain and stress ten- sors must be expressed in material local coordinates, since that is the base used by the elasticity tensor D. E = 1 2 (C −I) = 1 2 (GT · g · G −I) (4.17) E′ = T T · E · T = 1 2 (T T · GT · g · G · T −I) = 1 2 (AT · g · A −I) (4.18) A = G · T (4.19) δE′ = 1 2 AT · δg · A (4.20) (4.17) where T is the matrix of change of coordinates from local to global (in the reference conﬁguration). Thus, A results in the matrix of change of coordinates from local of the material to parametric of the element—in any conﬁguration! The expression for the internal virtual work can be expressed as: δΠint = ZZZ V δE′ IJ · S′ IJ · dV = ZZZ V δE′ : S′ · dV (4.21) (4.21) Taking advantage of Voigt’s notation we can write the following expres- sions: δE′ IJ · S′ IJ = 1 2 AT Ii · δgij · AjJ · S′ IJ = 1 2 δgij · AiI · AjJ · S′ IJ =1 2 δgij · sij = 1 2 δgT V oigt · sV oigt (4.22) gV oigt =   g11 g22 g12 + g21  Not = g (4.23) sV oigt =   s11 s22 s12  Not = s (4.24) IV oigt =   1 1 0  Not = I (4.25)   s11 s22 s12  =   A2 11 A2 12 2A11A12 A2 21 A2 22 2A21A22 A11A21 A12A22 A11A22 + A12A21   \| {z } QT ·   S′ 11 S′ 22 S′ 12   (4.26) From this point onward, unless speciﬁcally indicated by subscripts, Voigt notation will be used for the engineering strains and stresses. 3Sabourin and Brunet [112] also report a similar simultaneity of independent research publications on the same topic in 1993. where Q is deﬁned in equation (4.26). where Q is deﬁned in equation (4.26). The development of this ﬁrst and raw approach does not lead to satis- factory results. The reason behind it is that zero energy modes affect the solution. The ﬁrst thing that raises our suspicion is that the element as described above is non-conforming. That is, the normal at each point of the element is not necessarily perpendicular to the surface of the element (see equations (4.7) and (4.8)). We have kept the geometric description of the el- ement intentionally as simple as possible by using linear shape functions. As a consequence, it is possible for the mesh of triangles to fold like an ac- cordion and not develop signiﬁcant deformation energy. Take for example a simply supported square plate subjected to a point load at its center (ﬁg- ure 4.3). Note that because of the near-symmetry of the folds, the normals barely change direction making their gradient very close to zero. Thus the computed curvature using the ﬁrst equality in equation (4.6) might be accurate, even if the plane of the triangle differs a lot from the direction of the normals at the nodes. But this does not prevent the system from assigning totally erroneous positions to the nodes. 4.3. RAW APPROACH Thus, the δE′ IJ · S′ IJ = 1 2 AT Ii · δgij · AjJ · S′ IJ = 1 2 δgij · AiI · AjJ · S′ IJ =1 2 δgij · sij = 1 2 δgT V oigt · sV oigt (4.22) gV oigt =   g11 g22 g12 + g21  Not = g (4.23) sV oigt =   s11 s22 s12  Not = s (4.24) IV oigt =   1 1 0  Not = I (4.25)   s11 s22 s12  =   A2 11 A2 12 2A11A12 A2 21 A2 22 2A21A22 A11A21 A12A22 A11A22 + A12A21   \| {z } QT ·   S′ 11 S′ 22 S′ 12   (4.26) (4.22) (4.23) (4.24) From this point onward, unless speciﬁcally indicated by subscripts, Voigt notation will be used for the engineering strains and stresses. Thus, the 8 CHAPTER 4. IDEAS NEW SHELL ELEMENT CHAPTER 4. IDEAS NEW SHELL ELEMENT 38 V oigt subscript will be dropped from the notation. V oigt subscript will be dropped from the notation. V oigt subscript will be dropped from the notation. s = QT · S′ (4.27) δE′ IJ · S′ IJ = 1 2 δgij · sij = 1 2 δgT · QT · S′ = 1 2 δgT · QT · D · E′ (4.28) E′ = 1 2 (Q · g −I) (4.29) δg = ∂g ∂x · δx = B · δx (4.30) int Y = 1 4 ZZ S h 2 Z −h 2 δxT · BT · QT · D · (Q · g −I) · dz′ · dA (4.31) where Q is deﬁned in equation (4.26). where Q is deﬁned in equation (4.26). 4.3.1 A historical remark The current approach was ﬁrst introduced by Ubach and Oñate in [130]. This early proposal was followed by the current framework also by Ubach and Oñate in [131]. However, it must be said that, an almost identical and independent work (with identical results) was presented at the same congress by Linhard et al. [70]. They later published their work in [71]. This coincidence emphasizes the interest that the computation of thin- shells using rotation-free elements enjoys within the computational me- chanics community3. But it also underscores the difﬁculty in developing new and original strategies not explored before. 4.4. ATTEMPTS TO ADD ENERGY 39 Figure 4.3: Numerical example showing the displacements along the di- agonals (discontinuous line and dots) of a simply supported square with central point load. The results are compared to the corresponding solution given by the DKT element (solid line). Figure 4.3: Numerical example showing the displacements along the di- agonals (discontinuous line and dots) of a simply supported square with central point load. The results are compared to the corresponding solution given by the DKT element (solid line). 4.4 Attempts to add energy modes and stabilize the element We have explored different strategies to neutralize the instabilities caused by the lack of conformity in the description of the element: • The simplest strategy consisted in making an analogy that assumed that the deviation of the normal from the theoretic perpendicular to the plane at the center of the element represented a rotation due to shear deformation. • Another strategy consisted in assuming that the missing energy mode in the elements was an antimetric bending mode. Again, the devia- tion of the normal from the theoretic perpendicular at the center was used to estimate an increased deformation energy (ﬁgure 4.4). • Yet another strategy was devised. In this case the normal at the center of the element was prescribed to remain perpendicular to the element. To achieve this the triangle was further subdivided into three sub-triangles (ﬁgure 4.5). In all cases the resulting ﬁnite elements were stable and did not show zero energy modes. However the strategies had secondary effects. The ﬁrst ef- fect was an increased stiffness of the shell, yielding smaller displacements CHAPTER 4. IDEAS NEW SHELL ELEMENT 40 Figure 4.4: Figure depicting the concept of the antimetric bending energy mode to stabilize the element. Figure 4.4: Figure depicting the concept of the antimetric bending energy mode to stabilize the element. Figure 4.5: Figure depicting the concept for gently forcing the normals toward the perpendicular of the element. Figure 4.5: Figure depicting the concept for gently forcing the normals toward the perpendicular of the element. than the theoretic ones. The second observation was that the convergence, unlike that of conventional ﬁnite element formulations, approached the solution from less stiff solutions. This is suspected to be caused by the underlying non-conforming formulation. Of the three strategies described, the one that showed better results was the analogy with an antimetric bending mode. However, those results were achieved at the cost of excessive sophistication of the model and yet the errors were still inadmissible. 4.5 Evolving to a higher order description of the triangle In order to neutralize the instabilities caused by the lack of conformity in the description of the element, we have shifted towards a strategy that con- sists on increasing the order of the geometrical description so that all the 41 4.6. SUMMARY modes of deformation can be represented and thus avoiding the problems associated with the non-conforming formulation. The triangle of lowest order that can interpolate at the same time the positions and the normals at the nodes is a cubic triangle. However, in order to determine unambigu- ously the ten parameters of a cubic triangle there is not enough data with the three nodes and the three normals. Therefore, choices must be made. 4.6 Summary The design guidelines for the new rotation-free shell element have been presented. The design guidelines are: • Use only displacement degrees of freedom. Do not add any additional degree of freedom to the element if possible. • Develop the implicit formulation version of the element. This ensures being able to solve the widest possible range of engineering problems. An explicit version is easy to develop once the implicit is working. • Use a total Lagrangian framework to exploit its advantages for the computational implementation. • Develop the formulation for a continuum-based approach. This pro- vides rigor to the formulation. Using these guidelines and using the left-hand-side of equation (4.6) to compute the curvature on a surface S ⊂R3 a ﬁrst formulation is pre- sented. This work takes as starting point the nodal implementation of a basic triangle shell element (BSN; see Oñate and Zárate [96]). In order to use an exact formula for the curvature, the normal directions at each node and the way to characterize them are proposed. This ﬁrst development did not result to be a conforming element. Dif- ferent strategies were sought to solve this problem and control the spu- rious modes. Finally the decision was made to enhance the geometrical description of the element while keeping the main design features. Chapter 5 Using Bézier triangles A T THE TIME WHEN THE IDEAS PRESENTED IN THIS CHAPTER were be- ing developed, the Isogeometric Analysis was starting to boom. The author faced the decision of whether to join this new current or stick within a more traditional mesh-based Finite Element Method approach. The de- cision was to blend some of the advantages brought by both methods us- ing Bézier shape functions, which are of polynomial nature —just as the Lagrange functions used in the ﬁnite element method (FEM)—, but also serve as the basis to generate the NURBS functions used in IGA, and also use the versatility provided by the meshes of triangles. A number of chal- lenges have emerged as a consequence of this decision. Chapter 5 presents the framework provided by the Bézier shape functions and explains the geometric construction of the element. Chapter 6 develops all the under- lying formulation to obtain the equations that describe the deformation of the element and sets the basis to apply the variational principles used in continuum mechanics. Chapter 7 is devoted to instantiate a general expression of the tangential stiffness matrix of the element considering a linear elastic material, and developing all the expressions which are of par- ticular interest attending the inherent non-linear characteristics of shells’ geometries. Chapter 8 determines the most adequate order of integration of the element, while chapter 9 deals with the phenomenon of membrane locking. A T Chapter 4 showed the need for enhancing the geometric description of the element. The reasons are the following: • The idea of computing averaged normals at the nodes and using them to enrich the geometrical description of the element seems a good one. During his presentation at the 9th USNCCM Linhard [70] did report excellent accuracy in the computation of the curvatures using the scheme explained in section 4.2. • If the shell is to be computed using a conforming representation, then a linear description will not do. We have to increase the polynomial degree of the modeling. • The Bézier framework fulﬁlls the two requirements above: it al- lows to use the information provided by the normals and does so 43 44 CHAPTER 5. USING BÉZIER TRIANGLES using higher order polynomials. Furthermore, Bézier shape func- tions share the same variation diminishing and monotone variation properties that seem so valuable in IGA which uses NURBS shape functions [56, p. 4149]. Chapter 5 Using Bézier triangles As stated in section 4.1, it is a priority not to increase the number of degrees of freedom unless absolutely necessary. Therefore, we will use the simplest possible splines. The minimal spline that interpolates the nodes and the endpoint derivatives of a curve is a cubic. The same happens for a triangle (interpolating vertices and vertex normals). Furthermore, by interpolating the normals at the nodes, G1 continuity is attained automat- ically at the nodes (not overall G1 continuity though). The challenge then becomes to enhance the geometric description of the triangle without increasing the number of degrees of freedom in the system. A way to precisely deﬁne the geometrical parameters for the cubic triangle without increasing the number of degrees of freedom is adopting the concept of Bézier triangles ﬁrst deﬁned by de Casteljau [19, 20][37, Ch. 17, pp. 309–333]. The different sections in this chapter are devoted to understand this process. Since one of the main inputs that characterize the geometry of the tri- angles are the vertex normals, it is of paramount importance to obtain precise normals at the nodes of the mesh. This process is explained in section 5.1. Section 5.2 provides a detailed explanation of the shape func- tions used in Bézier triangles and their transformation from natural or barycentric coordinates to parametric coordinates. Section 5.3 provides the fundamentals to construct the geometry of the cubic Bézier triangle based on geometric considerations. Up to this point, only geometrical con- siderations are taken into account for the construction of the enhanced triangle. But a cubic Bézier triangle has more unknowns than data can be obtained from local geometric constraints. Therefore we will also take into account energy minimization in the geometric construction. This process is explained in section 5.4. 5.1 Selecting the normals The computation of the normals at the nodes is an arbitrary choice since this is unprescribed data and we must estimate it from the mesh informa- tion. Some might argue that this could be retrieved as exact data from the CAD model. However, that would result in an inconsistency in the process of analysis. There is no way to obtain the information on the normals from CAD in the deformed conﬁguration. If the normals are obtained using dif- ferent methods in the current and deformed conﬁgurations, it can lead to uncontrolled errors in the analysis. We choose to compute the normal at every node as a weighted average of the normals of all the triangles that surround the node (see ﬁgure 5.1). The weights chosen have been determined so that the error is minimized for a wide range of possible surface shapes. The paper [129] by Ubach, Estruch and García-Espinosa presents a speciﬁc study conducted to deter- mine those weights. Here the author will only present the conclusions of that paper. 5.1. SELECTING THE NORMALS 45 Figure 5.1: Example of a mesh approaching a surface S in the vicinity of a node i. The normal at node i is estimated using the normals at each of the surrounding triangles. ri is the total number of triangles surrounding node i. Figure 5.1: Example of a mesh approaching a surface S in the vicinity of a node i. The normal at node i is estimated using the normals at each of the surrounding triangles. ri is the total number of triangles surrounding node i. The goal of the study is to ﬁnd the weighting factors: The goal of the study is to ﬁnd the weighting factors: wk ∈R, k = 1 ÷ ri \| ˆni = ri X k=1 wkˆyk ri X k=1 wkˆyk + ⃗ O (5.1) (5.1) where ˆni is the actual unit normal vector of S at i and ⃗ O represents the error made; so that ⃗ O is minimized. See ﬁgure 5.1 for a graphical repre- sentation. The study proposes a framework for determining the weights that better characterize the representativity of each triangle’s normal vec- tor: Let’s consider the surface S and its local Taylor expansion series at the vertex [i]. Then, by truncating this series at the second order terms, we will obtain a quadric. The study concludes that: The study concludes that: In order to approximate the normal vectors at the nodes of a tri- angle mesh using a weighted average rule (see equation (5.1)), the weight consisting on the interior angle of the triangle at the node considered divided by the area of the circumscribed circle to the triangle (wα/A◦) is recommended. The mathematical ex- pression of this weighting factor for the node A as a function of the coordinates of the nodes of the Ï ABC triangle is shown in equation (5.2). (Ubach et al. [129], p. 267) wα/A◦= Ç ∥−−→ AB × −→ AC∥ ∥−−→ AB∥· ∥−→ AC∥· ∥−−→ BC∥ å2 · arccos Ç −−→ AB · −→ AC ∥−−→ AB∥· ∥−→ AC∥ å (5.2) (5.2) Therefore, the normal at each node is computed using the following formula: ˆni = ri X k=1 wk α/A◦ˆyk ri X k=1 wk α/A◦ˆyk (5.3) (5.3) In this case ˆni is the approximation to the actual unit normal vector of the surface S, unlike in equation (5.1) where the error term had not been truncated. CHAPTER 5. USING BÉZIER TRIANGLES We will still make one more assumption. This is that the conic resulting from the intersection with the quadric is a closed curve. This may seem like a very strong assumption. But in fact it is no more limiting than the restriction which represent the triangles themselves. What this assumption implies is that the section of the quadric is bounded; just like the triangles are. In no way, are we implying in the current argument that this framework will produce approximations with an error bounded by O ∝h2. What this analysis provides us is a framework to enrich the information provided by the set of [ˆyk] and their cor- responding triangles. (Ubach et al. [129], p. 249) The study concludes that: 5.1 Selecting the normals So now we have two different approximations to S at [i], namely: • the set of triangles with a vertex at [i], • the set of triangles with a vertex at [i], • and the quadric. • and the quadric. If we now make the assumption that the quadric interpolates all the vertices of the set of triangles, the error we are making is of the order of h2. Being h a measure of the size of the triangles. This allows us to reinterpret the relationship between [ˆyk] and the surface S. The vector [ˆyk] can be thought of as the normal direction of a plane section of the quadric. 46 5.2 Shape Functions Figure 5.2 shows a barycentric representation of the cubic Bézier trian- gle along with the corresponding parametric representation. The respec- tive shape functions in barycentric and parametric coordinates are as well written in the ﬁgure. Next we write explicitly the ten shape functions for the element: N 1 = (1 −ξ −η)3 N 2 = ξ3 N 3 = η3 N 4 = 3 · ξ · (1 −ξ −η)2 N 5 = 3 · η · (1 −ξ −η)2 N 6 = 3 · ξ2 · η N 7 = 3 · ξ2 · (1 −ξ −η) N 8 = 3 · η2 · (1 −ξ −η) N 9 = 3 · ξ · η2 N 10 = 6 · ξ · η · (1 −ξ −η) (5.4) (5.4) 5.2. SHAPE FUNCTIONS 47 Figure 5.2: Left: barycentric representation of the Bézier triangle along with its shape functions corresponding to each of the control points of the net. The capital Bs use the Bernstein polynomials notation [38]. Right: equivalent representation using parametric coordinates. The capital Ns refer to the more traditional FEM notation for the shape functions. The arrow in the center speciﬁes the transformation used from barycentric to parametric coordinates. Figure 5.2: Left: barycentric representation of the Bézier triangle along with its shape functions corresponding to each of the control points of the net. The capital Bs use the Bernstein polynomials notation [38]. Right: equivalent representation using parametric coordinates. The capital Ns refer to the more traditional FEM notation for the shape functions. The arrow in the center speciﬁes the transformation used from barycentric to parametric coordinates. These shape functions are represented in ﬁgure 5.3 for the control points along the diagonal edge of the parametric triangle and the central control point. Notice the change of terminology: the points which describe the geometry of the triangle are no longer called nodes, and instead they are called control points (p). As we will see later this doesn’t mean that we part away from the concepts of the isoparametric formulations. Even though we only use the vertices of the triangle as mesh and system variables, both the geometric description and the deformation description are done using the same shape functions together with this set of 10 control points. Using these shape functions the points on the shell’s midsurface are computed using equation (5.5). 5.2 Shape Functions ¯x(ξ, η) = x(ξ, η)\|ζ=0 = 10 X i=1 pi · N i(ξ, η) = piN i = p · N (5.5) p = p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 (5.6) (5.5) (5.6) CHAPTER 5. USING BÉZIER TRIANGLES 48 (a) (b) (b) (a) (b) (b) (a) (a) (c) (d) (d) (c) (c) (c) (d) (e) (e) Figure 5.3: Representation of the shape functions corresponding to 5 dif- ferent control points. Graphs (a), (b), (c), (d) and (e) depict respectively shape functions: N 2, N 3, N 6, N 9 and N 10. The vertical axes indicate the maximum value for each function. Figure 5.3: Representation of the shape functions corresponding to 5 dif- ferent control points. Graphs (a), (b), (c), (d) and (e) depict respectively shape functions: N 2, N 3, N 6, N 9 and N 10. The vertical axes indicate the maximum value for each function. The ﬁrst derivatives of the shape functions are: The ﬁrst derivatives of the shape functions are: The ﬁrst derivatives of the shape functions are: Lξ =   −3(1 −ξ −η)2 3ξ2 0 3(1 −ξ −η)(1 −3ξ −η) −6η(1 −ξ −η) 6ξη 3ξ(2 −3ξ −2η) −3η2 3η2 6η(1 −2ξ −η)   Lη =   −3(1 −ξ −η)2 0 3η2 −6ξ(1 −ξ −η) 3(1 −ξ −η)(1 −ξ −3η) 3ξ2 −3ξ2 3η(2 −2ξ −3η) 6ξη 6η(1 −ξ −2η)   (5.7) (5.7) 5.3. GEOMETRIC CONSTRUCTION 49 5.3. GEOMETRIC CONSTRUCTION 49 and the second derivatives are: Lξ,ξ =   6(1 −ξ −η)2 6ξ2 0 −6(2 −3ξ −2η) 6η 6η 6(1 −3ξ −η) 0 0 −12η   Lξ,η =   6(1 −ξ −η)2 0 0 −6(1 −2ξ −η) −6(1 −ξ −2η) 6ξ −6ξ −6η 6η 6(1 −2ξ −2η)   (5.8) Lη,ξ =   6(1 −ξ −η) 0 0 −6(1 −2ξ −η) −6(1 −ξ −2η) 6ξ −6ξ −6η 6η 6(1 −2ξ −2η)   Lη,η =   6(1 −ξ −η) 0 6η 6ξ −6(2 −2ξ −3η) 0 0 6(1 −1ξ −3η) 6ξ −12ξ   (5.9) and the second derivatives are: Lξ,ξ =   6(1 −ξ −η)2 6ξ2 0 −6(2 −3ξ −2η) 6η 6η 6(1 −3ξ −η) 0 0 −12η   Lξ,η =   6(1 −ξ −η)2 0 0 −6(1 −2ξ −η) −6(1 −ξ −2η) 6ξ −6ξ −6η 6η 6(1 −2ξ −2η)   (5.8) Lη,ξ =   6(1 −ξ −η) 0 0 −6(1 −2ξ −η) −6(1 −ξ −2η) 6ξ −6ξ −6η 6η 6(1 −2ξ −2η)   Lη,η =   6(1 −ξ −η) 0 6η 6ξ −6(2 −2ξ −3η) 0 0 6(1 −1ξ −3η) 6ξ −12ξ   (5.9) (5.8) (5.9) And therefore, the curvilinear coordinates are expressed as:1 ¯x,ξ (ξ, η) = ∂¯x ∂ξ = 10 X i=1 pi · Li ξ(ξ, η) = piLi ξ = p · Lξ (5.10) ¯x,η (ξ, η) = ∂¯x ∂η = 10 X i=1 pi · Li η(ξ, η) = piLi η = p · Lη (5.11) (5.10) (5.11) Using these expressions, the normal vector to the midsurface can be ob- tained. n(ξ, η) = ¯x,ξ ×¯x,η ∥¯x,ξ ×¯x,η ∥= piLi ξ × piLi η ∥piLi ξ × piLiη∥= (p · Lξ) × (p · Lη) ∥(p · Lξ) × (p · Lη)∥ (5.12) (5.12) For the reference conﬁguration, the normal vector N can also be referred to as t3. The ﬁrst derivatives of the shape functions are: t3 ≡N (ξ, η) = ¯X,ξ × ¯X,η ∥¯X,ξ × ¯X,η ∥= (P · Lξ) × (P · Lη) ∥(P · Lξ) × (P · Lη)∥ (5.13) (5.13) See ﬁgure 6.2 on page 68 for a graphical representation. See ﬁgure 6.2 on page 68 for a graphical representation. 1Notice the use of the Einstein convention for the summation over repeated indices. 2See section 5.4 for a detailed explanation of this issue. 5.3 Geometric construction of the Bézier triangle On one side, the author uses a barycentric-like notation to deﬁne the position of the control points. In order to refer to each of the control points in a cubic Bézier triangle, the author uses a dual notation (see ﬁgure 5.4). On one side, the author uses a barycentric-like notation to deﬁne the position of the control points. On the other side, once the control points have been deﬁned, the author uses a natural notation to use them in conjunction with the shape func- tions deﬁned in section 5.2. Since this section is focused on the geometric construction of the Bézier triangle, the barycentric-like notation will be used. y p p On the other side, once the control points have been deﬁned, the author uses a natural notation to use them in conjunction with the shape func- tions deﬁned in section 5.2. Since this section is focused on the geometric construction of the Bézier triangle, the barycentric-like notation will be used. The vertex control points are straight forward to evaluate: pi = xi ∀i = 1 ÷ 3 (5.14) (5.14) The net’s remaining control point coordinates can be evaluated for each point computing the intersection of three planes. For the contour points, the planes are the ones displayed shaded or darkened in ﬁgure 5.5: The net’s remaining control point coordinates can be evaluated for each point computing the intersection of three planes. For the contour points, the planes are the ones displayed shaded or darkened in ﬁgure 5.5: 1. The plane perpendicular to the normal at the vertex —this is a nec- essary and sufﬁcient condition to interpolate the normals—. 1. The plane perpendicular to the normal at the vertex —this is a nec- essary and sufﬁcient condition to interpolate the normals—. 2. The plane that contains the curve of the triangle’s contour. In order to maintain C0 continuity, this has to be a symmetric condition for two adjacent triangles. The selection of the plane is so that one of the directors is the edge of the ﬂat triangle and the other director is the average of the 2 normals at the nodes. In section 9.2 the author will show a variation on this condition to enhance the versatility of the element. 3. And a plane perpendicular to the edge of the ﬂat triangle. the exact location of this plane will be explained in section 5.4. 3This deﬁnition will change when the author introduces the emulated drilling rotations in chapter 9. Refer to equation (9.7) on page 106. 5.3 Geometric construction of the Bézier triangle The mathematical expression of a cubic triangle contains the complete monomial basis up to 3rd order: i.e. 10 monomials (notice the geometric 1Notice the use of the Einstein convention for the summation over repeated indices. CHAPTER 5. USING BÉZIER TRIANGLES 50 Figure 5.4: The geometry of the cubic Bézier triangle is determined by the position of the 10 control points. Notation used to identify each of the control points in a triangle. Left: barycentric-like notation. Right: natural notation. Notice that there exists a bijective relationship between the two notations. Figure 5.4: The geometry of the cubic Bézier triangle is determined by the position of the 10 control points. Notation used to identify each of the control points in a triangle. Left: barycentric-like notation. Right: natural notation. Notice that there exists a bijective relationship between the two notations. analogy between Pascal’s triangle in table 5.1 and the barycentric repre- sentation of the Bernstein polynomials in ﬁgure 5.2, which indicates that both sets of polynomial basis span the same polynomial space). Therefore, in order to obtain the expression for a given triangle we have to determine the 10 coefﬁcients for the complete polynomial in the 3 spatial coordinates: 30 in total. There is clearly not enough information in the position of the 3 vertices (3 × 3 = 9 conditions) and the orientation of the 3 vertex nor- mals (3 × 2 = 6 conditions) to uniquely determine these 30 coefﬁcients. I could have opted to simply discard 5 terms of the polynomial and reduce the order of description of the cubic triangle to an incomplete quadratic triangle. Table 5.1: Pascal’s triangle of monomials for bi-variate cubic functions. 1 ξ η ξ2 ξη η2 ξ3 ξ2η ξη2 η3 Instead, I seized the opportunity that Bézier triangles provide for con- structing higher order triangular surfaces. The bulk of this thesis work consists on ﬁnding the geometrical relations that allow to complete the cu- bic polynomial basis for a Bézier triangle aimed at performing non-linear thin shell analysis. Next I explain the process I have followed and which generates a space of solutions that includes the solution of the discrete problem. Other arbitrary choices do not, in general, generate such a space of solutions2. 5.3. GEOMETRIC CONSTRUCTION 51 In order to refer to each of the control points in a cubic Bézier triangle, the author uses a dual notation (see ﬁgure 5.4). 5.3 Geometric construction of the Bézier triangle Sufﬁce it to say, that the criterion to position the plane is based on energy minimiza- tion and not on geometric considerations. The mathematical expression of these three planes and their intersec- tion is presented in equation (5.15):   niT dijT (xi −xj)T   \| {z } Aij p ·pij =   niT · xi dijT · xi (xi −xj)T · (1 −Ψij)xi + Ψijxj   \| {z } bij p ∀i, j = 1 ÷ 3, i ̸= j (5.15) (5.15) ∀i, j = 1 ÷ 3, i ̸= j So the control points on the contour are computed by solving the system of equations: Aij p · pij = bij p (5.16) (5.16) where dij represents the vector perpendicular to the plane indicated by item 2 and represented light gray in ﬁgure 5.53; and the parameter Ψij CHAPTER 5. USING BÉZIER TRIANGLES 52 Figure 5.5: Representation of the net of control points (dashed lines) for a cubic Bézier triangle constructed using the nodal positions and normals. Three planes—shaded, light gray and dark gray—deﬁne the position of a control point A of the contour. Question marks indicate that the position of the dark gray planes has not been determined yet. Their position will be solved in section 5.4. Figure 5.5: Representation of the net of control points (dashed lines) for a cubic Bézier triangle constructed using the nodal positions and normals. Three planes—shaded, light gray and dark gray—deﬁne the position of a control point A of the contour. Question marks indicate that the position of the dark gray planes has not been determined yet. Their position will be solved in section 5.4. speciﬁes the location of the plane indicated by item 3 on page 51 and rep- resented dark gray in ﬁgure 5.5. These deﬁnitions (dij and Ψij) will change in chapter 9. The central control point is determined by the intersection of the three mid-edge control triangles. There is a geometric condition that can be used. If all the control triangles sharing an edge across two triangles are co-planar, then the two triangles have G1 continuity [37, pp. 368–371]. This is very desirable as has been discussed in sections 3.2.2 and 4.1.1 on page 22 and on page 28. 5.3 Geometric construction of the Bézier triangle Because the control triangles at the corners of each Bézier triangle are orthogonal to the normal vectors at the nodes (see item 1 on page 51), then all the corner control triangles at a given node have already been deﬁned as co-planar. Then, we would only need to enforce co-planarity of the mid- edge control triangles to achieve C1 continuity (see ﬁgure 5.6). However this is a double-edged sword and not always possible. Being a very ap- pealing opportunity to construct a fully C1 continuous triangle, it would be, alas, a very ill-conditioned equation system. Take for example the case of a ﬂat triangle shape. In this case, all three mid-edge control triangles would be co-planar, and their intersection indeﬁnite. Another undesirable case would be when the three mid-edge control triangles intersect far away from the triangle’s domain, or don’t intersect at all. In order to work around the ill-conditioning of the above condition, let’s perform a trick. Instead of computing the intersection of the three mid- edge control triangles, I will estimate a likely position for this intersection. 5.3. GEOMETRIC CONSTRUCTION 53 Figure 5.6: Two adjacent Bézier triangles are G1 continuous if all their ad- jacent control triangles are co-planar. This condition could be used to en- force C1 continuity and determine the position of the central control point. The control triangles at the corners are always co-planar since they are perpendicular to the normal at the node. The mid-edge control triangles are colored dark gray. Figure 5.6: Two adjacent Bézier triangles are G1 continuous if all their ad- jacent control triangles are co-planar. This condition could be used to en- force C1 continuity and determine the position of the central control point. The control triangles at the corners are always co-planar since they are perpendicular to the normal at the node. The mid-edge control triangles are colored dark gray. It is likely that the central control point will be located near the vertical line that runs through the barycenter of the triangle. To be more pre- cise, let’s use the barycenter of the 6 control points in the perimeter of the Bézier triangle and the line perpendicular to the ﬂat triangle running through it. 5.3 Geometric construction of the Bézier triangle Where njk refers to the average of the normals nj and nk. njk = nj + nk 2 (5.20) (5.20) The resulting system of equations above is well posed, and thus a ro- bust method to compute the central control point. Constructing the cen- tral control point in this way does not ensure C1 continuity, but comes close enough. In any case, neighboring triangles are C1 continuous at the nodes, but not—in general—along the edges. 5.3 Geometric construction of the Bézier triangle Then, let’s compute the candidate position for the central con- trol point corresponding to an edge (p0i), by intersecting this line with the plane perpendicular to the plane described by item 2 on page 51 and pass- ing through the intermediate control points of the edge. The barycenter of these three candidate positions is taken as the position for the central control point. p123 = 1 3 3 X i=1 p0i (5.17) (5.17) Thus, the central point of the net is obtained as an average of the three candidate points for this position. Each of the three candidate points is computed solving a system of the form: Thus, the central point of the net is obtained as an average of the three candidate points for this position. Each of the three candidate points is computed solving a system of the form: A0i p · p0i = b0i p (5.18) (5.18) The deﬁnition of A0i p and b0i p is given by equation (5.19). This is an analo- gous procedure to the one followed for the control points of the contour. So the intersection of three planes is solved: 54 CHAPTER 5. USING BÉZIER TRIANGLES 1. A plane perpendicular to the one described by item 2 on page 51. This is a symmetric condition and thus the same plane will be used for adjacent triangles. 2. A plane perpendicular to one of the edges of the ﬂat triangle and passing through the barycenter of the 6 control points of the perime- ter. 3. And a plane perpendicular to another edge of the ﬂat triangle and passing through the barycenter of the 6 control points of the perime- ter.   njk × (pjk −pkj) × (pjk −pkj) (xi −xj)T (xi −xk)T   \| {z } A0i p ·p0i = =   njk × (pjk −pkj) × (pjk −pkj) · pjk (xi −xj) · 1 6(pij + pji + pik + pki + pjk + pkj) (xi −xk) · 1 6(pij + pji + pik + pki + pjk + pkj)   \| {z } b0i p ∀i = 1 ÷ 3, (j, k) = combination{1, 2, 3} −{i} (5.19) ∀i = 1 ÷ 3, (j, k) = combination{1, 2, 3} −{i} (5.19) Where njk refers to the average of the normals nj and nk. Where njk refers to the average of the normals nj and nk. 5.4 Determining shape through energy minimization As reported in [132] by Ubach and Oñate, the construction of the Bézier triangle did not account initially for energy aspects. Some arbitrary de- cisions were adopted in that work. For example, ﬁxing the boundaries of the triangle to lie in a plane causes the boundaries to be planar curves. Clearly, this does not exploit all the potential of a cubic formulation. This aspect will be treated by the author in section 9.2. There are, though, two relevant differences between the constructions reported in section 5.3 and in [132]. The ﬁrst difference is the distance at which the plane speciﬁed by item 3 on page 51 is located. The second one is the decision to use the barycenter of the 6 contour control points instead of the triangle’s barycenter as the reference to locate the central control point. 5.4. ENERGY MINIMIZATION 55 Concerning the ﬁrst case, Ubach and Oñate stated that the value Ψ = 1 3 was applied always. No issues were detected using this constraint when solving linear problems. A lack of precision was identiﬁed, though, for what should be expected in a cubic triangle. Consequently Ubach and Oñate did specify in [132, Section 5. Conclusions]: We need to improve the capability of the element to represent constant curvatures not only in the limit of the element’s size becoming zero, but for the more general case. becoming zero, but for the more general case. At that time the author was working behind the idea of making the value of the parameter Ψ depend on the angle ϕ between the edge’s chord and the tangent of the curved boundary at the vertex.—The deﬁnition of ϕ is given in ﬁgure 5.7.—The aim was to create boundary curves of approximately constant curvature. This thinking changed when the author tried to use the construction in [132] to solve non-linear problems. The element formulated in that way was incapable of converging to any solution. This posed a great challenge. The author struggled for several months trying to ﬁnd the root cause of the problem. A full revision of the code was undertaken to make sure no bugs were present in the software code. And a thorough veriﬁcation of every derivative was performed numerically. Finally the author got a hint at the problem when performing trials with different numerical quadratures. 5.4 Determining shape through energy minimization The element seemed to exhibit a somewhat better convergence behavior (never converging totally) when only one quadrature point was used at the center of the element. This was a very bizarre behavior. So something had to be going on regarding the geometry of the element in the points located far from the center. The author’s suspicion was that the decision to assign a value of 1 3 to the parameter Ψ was playing a role. Furthermore, the fact that even using a Newton-Raphson nonlinear solving scheme it was impossible to obtain convergence, made the author suspect that the space of solutions created did not include the solution to the problem. Let’s go deeper into this idea for a second. In the Finite Element Method, the continuous domain is modeled by a discrete domain. There- fore, the space of solutions generated by the mesh of ﬁnite elements is a reduced version of the space in which the solution for the continuous problem lies. Furthermore, by including arbitrary constrains to the con- struction of the Bézier triangle, we had not checked for the existence of the solution in the constrained space of ﬁnite elements we had generated. Indeed, by constraining the geometric construction of the element with the arbitrary decision of placing the dark grayed plane in ﬁgure 5.5 at 1 3 of the edge’s chordal length, we were modifying the shape of the deformed tri- angle away from the solution that minimizes the deformation energy with respect to the nodal forces. So it was as if we were imposing virtual forces and/or moments in the interior of the element. These virtual forces and moments did not belong to the discrete model of the problem and therefore it was impossible for the element to obtain a converged solution. To put it in another way, the non-linear problem was not well-posed. It was at this point that the author acknowledged the need to take into account the energy associated with the deformation of the element CHAPTER 5. USING BÉZIER TRIANGLES 56 if the element had to solve non-linear problems. In particular, in order to obtain the conditions to determine the location of the Bézier control points, those conditions which cannot be determined via geometric considerations, should be obtained considering an energy minimization problem. 5.4.1 Reduction of the problem In order to tackle the energy minimization problem associated to the value of Ψ (i.e. the location of the dark grayed plane in ﬁgure 5.5), the author had to simplify the general 3D problem to something more manageable. The general problem can be solved with optimization algorithms. However this approach has two drawbacks. On the one side, optimization algorithms are computationally expensive. On the other side, they do not provide information on the derivatives of the solution, which are needed to solve the non-linear problem. Since the general 3D solution cannot be obtained analytically, the author performed a reduction of the problem in order to obtain an analytic solution that could be used to approximate the general solution. First, instead of ﬁnding an analytic solution concerning the whole tri- angle, only the edge of the triangle was considered. So that the bend- ing and membrane deformation energies of the triangle were assimilated to the bending and axial (membrane) deformation energies of a 2D line (Bernoulli beam). Having previously considered that the edges of the tri- angle were ﬂat curves came in handy. Figure 5.7: Representation of the 2D problem simpliﬁcation. The edge of the triangle between nodes i and j is represented as a ﬂat curve de- ﬁned by the parametrization ⃗r(t). The angle ϕij is deﬁned here for ﬂat curved edges. For the more general case of 3D curved edges, refer to the deﬁnitions provided in ﬁgure 9.3 and equation (9.18) on page 105 and on page 113. Figure 5.7: Representation of the 2D problem simpliﬁcation. The edge of the triangle between nodes i and j is represented as a ﬂat curve de- ﬁned by the parametrization ⃗r(t). The angle ϕij is deﬁned here for ﬂat curved edges. For the more general case of 3D curved edges, refer to the deﬁnitions provided in ﬁgure 9.3 and equation (9.18) on page 105 and on page 113. Second, the problem of minimizing the total energy was split into the two separate problems: minimization of the energy associated to bending deformation; and minimization of the energy associated to axial (mem- brane) deformation. We know from the works of Chapelle and Bathe [22] and Briassoulis [15–17], that the asymptotic behavior of shell structures as their thickness tends to zero can be membrane dominant, shell domi- nant, or mixed; depending on the geometry, the loads and the boundary 5.4. ENERGY MINIMIZATION 57 conditions. 4The author is knowledgeable of the bad press for adjustments in the FEM literature. Fear not. The process followed is fully rigorous and maintains the objective of minimizing the overall deformation energy. However, taking into account compatibility considerations, this problem is converted into a locally-deﬁned energy-adjustment problem. 5.4.1 Reduction of the problem Therefore, by considering the two energy modes separately we create the opportunity for taking into account the different combinations of bending energy and membrane energy modes in our study. Clearly, by simplifying the problem it is impossible to account for every particular case and for the different material properties and distributions in the shell. Nevertheless, the solutions provided are only local and there- fore they do not reduce the validity of the discretization strategy using the FEM. On the other hand, by taking into account separately the bending and axial (membrane) energy, we are effectively decoupling the effect that layered composite materials could have on the ratio of these energies in the shell laminate. Finally, by changing the overall minimization prob- lem into the analysis of a reduced problem, we have to carefully analyze and adjust, not minimize, the deformation energy of the shell.4 It is the author’s conviction that this approach yields a sufﬁciently good approxi- mation for every particular case. Bending energy The problem of bending energy minimization (adjustment) was the easiest of the two and was tackled ﬁrst. The study started by acknowledging that a necessary condition for the absence of any virtual moments along the 2D curve of the edge was that the resulting shape of the curve shall be a spi- ral. That is, a 2D beam with only endpoint moments has a law of moments which varies monotonically (strictly speaking the law of moments varies linearly in the absence of axial forces). Considering that the effect of mo- ments induced by axial forces should not affect the monotonicity of the law of moments if the shell is shallow enough, this condition is accepted. Since for a 2D Bernoulli beam the curvature at any point is propor- tional to the bending moment, a monotonic law of bending moments results in a monotonic law of curvatures along the beam. Therefore, the construc- tion of the cubic Bézier curve shall result in all cases in a spiral. Figure 5.8 shows in red the region of pairs of angles at the endpoints of a 2D cubic Bézier curve for which spirals are obtained ﬁxing the value Ψ = 1 3. 3 At this point the author implemented the idea of improving the capa- bility of the element to represent constant curvatures. The idea was that the analytic solution of the value of Ψ for representing constant curvatures was a better approximation than ﬁxing this value to 1 3; and thus a solution candidate. This value is found for a symmetric curve (the start and end point angles are the same) by forcing the center point of the cubic Bézier spline to coincide with the circular arc spanning the same chord with the same start and endpoint tangents. The value of Ψ that satisﬁes this con- dition for any given angle ϕ is speciﬁed in equation (5.21). Ψij ϕ (ϕij) = 1 3 · 2 · cos ϕij 1 + cos ϕij (5.21) (5.21) 4The author is knowledgeable of the bad press for adjustments in the FEM literature. Fear not. The process followed is fully rigorous and maintains the objective of minimizing the overall deformation energy. However, taking into account compatibility considerations, this problem is converted into a locally-deﬁned energy-adjustment problem. 58 CHAPTER 5. 5.4. ENERGY MINIMIZATION 59 With this formula we are ensuring to comply with the patch test and therefore to satisfy the accurate representation of constant curvatures. How about non-symmetric curves? Figure 5.8 shows in blue the the re- gion of pairs of angles at the endpoints of a 2D cubic Bézier curve for which spirals are obtained plugging the value of Ψij ϕ of equation (5.21) into equation (5.15). The improvement is clear in all the quadrants of the do- main. The improvement is more signiﬁcant when the sign of the curvature doesn’t change than when the sign of the curvature does change. Taking into account this result, the author considers that equation (5.21) constitutes a good approximate solution for the minimization of the bend- ing energy. Furthermore, this is a very simple equation that is easy to differentiate and thus does not add extra complexity to the problem. Bending energy USING BÉZIER TRIANGLES (a) (b) Figure 5.8: Regions of start and endpoint angles of a cubic spline for which spirals are generated. The red line indicates the region corresponding to a construction of the cubic Bézier spline using Ψ = 1 3. The blue line indicates the region corresponding to a construction of the cubic Bézier spline using equation (5.21). (a) Variables in the axis expressed in terms of the angle of the normal with respect to the chord. (b) Variables in the axis expressed in terms of the sinus of the angles represented in (a). (b) (b) Figure 5.8: Regions of start and endpoint angles of a cubic spline for which spirals are generated. The red line indicates the region corresponding to a construction of the cubic Bézier spline using Ψ = 1 3. The blue line indicates the region corresponding to a construction of the cubic Bézier spline using equation (5.21). (a) Variables in the axis expressed in terms of the angle of the normal with respect to the chord. (b) Variables in the axis expressed in terms of the sinus of the angles represented in (a). 5.4. ENERGY MINIMIZATION Axial (membrane) energy While the bending problem was the easiest to solve, the membrane energy problem is the least evident or intuitive to lay out. Following the same simpliﬁcation procedure explained in section 5.4.1 on page 56, the author developed a thorough analytic solution for this reduced problem. The com- plete development can be followed in appendix C. The key aspect of this development is to understand that we are not so interested in minimizing the total strain energy as in minimizing the variation of the density of strain energy along the beam. The main reason that justiﬁes this approach is that, in order to satisfy the patch test, the element has to be able to represent constant states of deformation. There- fore, we are interested in designing an element which represents constant states of deformation and thus letting the energy be what it ought to be. Let’s make an analogy: the analogy shall be between the axial strain of the Bernoulli beam and the celerity (c) of the curve parameter (t). Let, ⃗r(t) = (x(t), y(t)) (5.22) (5.22) be the equation of a parametric curve in the Cartesian coordinates (x, y). ⃗r(t) represents the generatrix of the 2D Bernoulli beam. Then, c is deﬁned as: ⃗r′(t) = ∂⃗r(t) ∂t (5.23) c = ∥⃗r′(t)∥ (5.24) (5.23) (5.24) Following this analogy, the density of axial deformation energy is pro- portional to c2. Then, we are interested in ﬁnding the values of Ψ which minimize the variation of c2 along the curve. In equation (5.25) we take the square of the value to minimize in order to avoid the effect of the sign. ∂ ∂Ψij Z 1 0 Å∂c2 ∂t ã2 dt = 0 ∂ ∂Ψji Z 1 0 Å∂c2 ∂t ã2 dt = 0          (5.25) (5.25) The system in equation (5.25) is a system of 2 cubic equations, with two unknowns: Ψij and Ψji; and taking the angles ϕij and ϕji as variables (see 60 CHAPTER 5. USING BÉZIER TRIANGLES appendix C). The author has programmed this system of equations into the Maple® symbolic manipulation software. The analytic solution obtained by Maple® has over 1 million characters. This makes it almost impossible to verify and very hard to program efﬁciently. But most importantly, its evaluation would be very costly computationally. Since the analytic solution is incomprehensible, the author sought a graphical solution of the problem. Axial (membrane) energy Each of the two equations in equa- tion (5.25) represents a 2D curve in the coordinate space (Ψij, Ψji). There- fore, the intersection of these two curves is the solution of the problem. The author plotted these intersections using Maple® for different pairs of angles (ϕij, ϕji). The results are shown in ﬁgure 5.9. (ϕ ϕ ) Figure 5.9 deserves some comments. First of all, the reader must notice how close the solution provided by Ψϕ in equation (5.21) is to the symmet- ric cases, even for values of ϕ up to 27°. This is a very encouraging result, because it opens the door to use a very simple and uncoupled expression (equation (5.21)), and obtain a good approximation of a very complex and coupled equation (equation (5.25)). Secondly, the reader can also observe that the exact solution exhibits strong coupling for the non-symmetric cases. This limits effectively the applicability of the approximation pro- vided by Ψϕ to small values of ϕ where the coupling effect is not so severe. Combining both observations, the author concludes that the formula pro- vided by equation (5.21) can be used to solve also the membrane energy aspects of the element as long as the values of ϕ do not exceed 18°. This is consistent and well within the limits of the validity of the formula re- garding the construction of spirals shown in ﬁgure 5.8. After all 18° is not so small, as it allows an element to conﬁdently span an arc of up to 36°; which is very interesting. 5.4.2 Location of the central control point The decision to locate the central control point near the the barycenter of the 6 control points of the contour also responds to an energy minimiza- tion criterion. However in this case the reasoning is heuristic. The author observed that the solution for both the bending and axial energy mini- mization problems in the 2D Bernoulli beam tends to make Bézier splines which have control segments of similar length. That is, the higher the value of ϕ, the smaller the corresponding value of Ψϕ, to compensate for the extra length of the control path. Also, by understanding the parametric construction of a Bézier curve and a Bézier triangle that is well analyzed in the analogy used to pose equation (5.25), the author gains this heuristic knowledge that having the control points equally spaced favors a correct distribution of the deformation energy density. This is why the author decides to position the central control point in a line that passes through the barycenter of the 6 contour control points. A more sophisticated approach would have been to compute the plane that minimizes the square distances to the 6 contour control points and posi- tion the central control point in the line perpendicular to that plane and that passes through the barycenter of these 6 control points. Instead, the author selects another line far easier to compute which is the line perpen- dicular to the ﬂat shape of the triangle. 5.4. ENERGY MINIMIZATION 61 (9, −9) (18, −18) (27, −27) (0, 0) 1 −Ψji Ψij (36, −36) (9, 0) (18, 0) (27, 0) (36, 0) (45, 0) (9, 9) (18, 9) (27, 9) (36, 9) (45, 9) (18, −9) (18, 18) (27, 18) (36, 18) (45, 18) (27, −9) (27, −18) (27, 27) (36, 27) (45, 27) (36, −9) (36, −18) (36, −27) (36, 36) (45, 36) (45, −9) (45, −18) (45, −27) (45, −36) (45, 45) Figure 5.9: Graphical solution of equation (5.25). Each plot solves the system of equations for a given pair of angles (ϕij, ϕji) indicated at the top right corner of each plot. The vertical axis represents the value of Ψij and it varies from 0 at the top and 0.5 at the bottom. The horizontal axis represents the value of 1 −Ψji, taking values between 0.5 and 1. The two black curves represent each of the two expressions in equation (5.25). CHAPTER 5. USING BÉZIER TRIANGLES 62 5.4.2 Location of the central control point So their intersection is the exact solution of the system. The red dot in each plot represents the values of Ψij and Ψji obtained using equation (5.21). Figure 5.9: Graphical solution of equation (5.25). Each plot solves the system of equations for a given pair of angles (ϕij, ϕji) indicated at the top right corner of each plot. The vertical axis represents the value of Ψij and it varies from 0 at the top and 0.5 at the bottom. The horizontal axis represents the value of 1 −Ψji, taking values between 0.5 and 1. The two black curves represent each of the two expressions in equation (5.25). So their intersection is the exact solution of the system. The red dot in each plot represents the values of Ψij and Ψji obtained using equation (5.21). CHAPTER 5. USING BÉZIER TRIANGLES 5.5 Summary The process of geometric construction of the cubic Bézier triangle has been presented. First, the need for obtaining accurate normals at the nodes is explained and the best averaging weights have been obtained for a very wide variety of surfaces. This is an original contribution of this thesis. The shape functions based on the Bernstein polynomials for a cubic triangle have been presented. These shape functions offer very attrac- tive properties like the variation-diminishing property and the monotone- variation property that IGA enjoys and exploits. The reader is encouraged to consult the references for a thorough understanding of the properties of this polynomial basis. (a) Geometry of a cylinder. (b) Coarse mesh used to discretize a cylinder. (c) Geometric representation of the BEST el- ements obtained from the coarse mesh (left) to discretize the cylinder. (a) Geometry of a cylinder. (a) Geometry of a cylinder. (a) Geometry of a cylinder. (b) Coarse mesh used to discretize a cylinder. (c) Geometric representation of the BEST el- ements obtained from the coarse mesh (left) to discretize the cylinder. (b) Coarse mesh used to discretize a cylinder. (c) Geometric representation of the BEST el- ements obtained from the coarse mesh (left) to discretize the cylinder. (b) Coarse mesh used to discretize a cylinder. Figure 5.10: Example of the enhancement provided by the procedure pre- sented in this chapter to model the geometry obtained from a mesh of lin- ear triangles. The BEST mesh (c) has the same degrees of freedom as the linear mesh (b). Taking into account these shape functions the geometric construction of the cubic Bézier triangle is performed; control point per control point. First, the corners. Then, the contour. And ﬁnally the central control point. 63 5.5. SUMMARY 5.5. SUMMARY The construction presented is C1 continuous at the nodes. Following these geometric rules the author has developed a new way to describe a cubic triangle using only linear unknowns corresponding to the coordinates of the vertices of the triangle. The coordinates of the control points are de- termined taking into account the coordinates of the neighboring triangles’ vertices. This geometric construction for a cubic shell triangle is an origi- nal contribution of this thesis. In order to obtain non-linear convergence, a detailed energy minimiza- tion analysis is performed on the geometric construction of the cubic Bézier triangle. The study is performed after a reduction of the dimensional com- plexity of the problem from 3D to 2D, and uncoupling the bending and membrane energy modes of the shell. This study concludes the determi- nation of the remaining parameters that could not be determined through geometric conditions alone. Using the simple formula (see equation (5.21)) obtained in this study, the author has achieved excellent non-linear conver- gence in all cases. This analysis is an original contribution of this thesis. Owing to the construction process followed to create this new ﬁnite element, the author decided to name it Bézier-Enhanced Shell Triangle (BEST). Figure 5.10 provides an example showing how using the Bézier enhancement process affects the geometric representation of a curved shell. Taking into account the number of original contributions included in this chapter, the reader will understand that this is a central chapter of the thesis. 1Cauchy stresses are suggested for engineering analysis. The development presented uses the second Piola-Kirchhoff stress tensor for the computation of elastic energy, since this is the conjugate of the Green-Lagrange strain tensor. 2Notice the difference in terminology with respect to the geometrically exact term coined 1Cauchy stresses are suggested for engineering analysis. The development presented uses the second Piola-Kirchhoff stress tensor for the computation of elastic energy, since this is the conjugate of the Green-Lagrange strain tensor. 2Notice the difference in terminology with respect to the geometrically exact term coined Chapter 6 U SING THE BÉZIER TRIANGLES DESCRIBED in sections 5.1 to 5.4 we can now develop a conforming formulation based on the continuum- based approach. Continuum-based shell formulations—a term proposed by Stanley [120]—have become popular because of their simplicity. Gen- eral curved shell theories are complex and difﬁcult to implement, while a continuum-based framework beneﬁts from the general 3D continuum the- ory of elasticity. There are a number of frameworks that differ in name, but different scholars group them together [9, p. 536][12, p. 76]. This is the case of the degenerated solid approach, ﬁrst proposed by Ahmad, Irons and Zienkiewicz [2], and that of the geometrically exact shell model [118]. The author shares the preference stated by Belytschko et al. [9] and refers to this as the continuum-based approach. U The formulation here presented is based on the same principles already applied in section 4.3. Here we will develop them further. The author pro- poses the use of Green-Lagrangian strains and Cauchy stresses1 instead of generalized strains and generalized stresses because the former are bet- ter suited to evaluate the state of strain and stress of the material. In the case of a laminated shell, each layer will be subject to different stress states and evaluating these values at each layer allows to evaluate failure of the structure in a more precise manner than using generalized stresses and strains (for example in the case of evaluating principal strains and stresses and comparing their directions to the directions of the material ﬁbers), which only provide information on the pre-integrated section char- acteristics and thus cannot account for all the cases. Having said that, the author has applied an efﬁcient and geometrically accurate2 pre-integration 65 CHAPTER 6. CONTINUUM-BASED FORMULATION 66 scheme in order to avoid extra computations associated to the volume in- tegrals. 3Notice that combined with equation (6.7) and equation (4.16), equation (6.8) deﬁnes an implicit second order equation for λ. The solution to this equation includes a fraction and a square root with terms that belong to the reference conﬁguration and the deformed conﬁguration and cannot be factored. As a result, the expression of λ cannot be linearized like all the other terms. This is why it is much more effective and efﬁcient to evaluate λ explicitly instead of implicitly. 6.1 Thickness change due to the Poisson effect and the mild taper assumption Let’s thus deﬁne the plane stress state constitutive tensor of a linear elastic orthotropic material:   S′ 11 S′ 22 S′ 12  = 1 1 −ν12 · ν21   E1 ν12 · E1 0 ν21 · E2 E2 0 0 0 (1 −ν12 · ν21) · G12   \| {z } D ·   ε1 ε2 γ12   (6.4) 1 G12 ≃1 + ν21 E1 + 1 + ν12 E2 (6.5)   S′ 11 S′ 22 S′ 12  = 1 1 −ν12 · ν21   E1 ν12 · E1 0 ν21 · E2 E2 0 0 0 (1 −ν12 · ν21) · G12   \| {z } D ·   ε1 ε2 γ12   (6.4) 1 G12 ≃1 + ν21 E1 + 1 + ν12 E2 (6.5) E′ =   E′ 11 E′ 22 E′ 12 + E′ 21  =   ε1 ε2 γ12   (6.6) ε3 = −ν31 · S′ 11 + ν32 · S′ 22 (6.7)   S′ 11 S′ 22 S′ 12  = 1 1 −ν12 · ν21   E1 ν12 · E1 0 ν21 · E2 E2 0 0 0 (1 −ν12 · ν21) · G12   \| {z } D ·   ε1 ε2 γ12   (6 4) (6.4) 1 G12 ≃1 + ν21 E1 + 1 + ν12 E2 (6.5) E′ =   E′ 11 E′ 22 E′ 12 + E′ 21  =   ε1 ε2 γ12   (6.6) ε3 = −ν31 · S′ 11 + ν32 · S′ 22 E3 (6.7) λ = 1 + ε3 (6.8) (6.5) (6.7) (6.8) Where E1 and E2, ν12 and ν21, and G12 are respectively the Young moduli, the Poisson ratios and the shear modulus of the material in the principal directions of the plane; and E3, ν31 and ν32 are respectively the Young mod- ulus and Poisson ratios associated with the principal direction orthogonal to the plane. p The value of λ—evaluated after each time-step—can be used to update the thickness of the material.3 Thus, the total thickness in the deformed conﬁguration will be simply expressed as: ♯layers X l=1 λlhl (6.9) (6.9) Where h is the thickness of the material in the reference conﬁguration. Where h is the thickness of the material in the reference conﬁguration. 6.1 Thickness change due to the Poisson effect and the mild taper assumption The mild taper assumption for the shell thickness is adopted. This as- sumption implies that in the case of a shell with varying thickness, where the curvilinear coordinate ζ in the normal direction changes orientation to adapt to each layer of the shell, it is assumed that its direction remains constant through the thickness and coincident with the normal vector to the midsurface (see ﬁgure 6.1). The mild taper assumption is expressed mathematically in equation (6.1). In the case of a shell with steeply vary- ing thickness, this will be considered by changing the nominal thickness from element to element. The mesh can be adaptively reﬁned if necessary. Figure 6.1: Representation of the curved shell element identifying the curvilinear coordinates (ξ, η, ζ) and the effect of the mild taper assump- tion. The reference midsurface is identiﬁed. Figure 6.1: Representation of the curved shell element identifying the curvilinear coordinates (ξ, η, ζ) and the effect of the mild taper assump- tion. The reference midsurface is identiﬁed. x,ζ (ξ, η, ζ) = ∂x ∂ζ ≃∂x ∂ζ ζ=0 = ∂¯x ∂ζ ∥n(ξ, η) ⇒ x,ζ (ξ, η, ζ) ≃∥n(ξ, η) (6.1) (6.1) However, thickness change is considered accounting for the Poisson ef- fect. This does not imply that the zero vertical stress assumption is elimi- nated. The author wants to include the effect that stretching in the shell’s surface has on the thickness of the shell. The classical Kirchhoff-Love kinematic hypotheses have been stated in section 3.2.2 on page 22. We will drop the ﬁrst one and keep the other two. But unlike other authors that propose 6 and 7 parameter models to account for the full three-dimensional stress state of the shell, the author will simply allow that ε3 ̸= 0 (6.2) (6.2) ε3 ̸= 0 by Simo and Fox in [118]. DEFINITION OF LOCAL AXES 6.2. 67 but maintaining a plane stress state of the material: but maintaining a plane stress state of the material: σ3 = 0 (6.3) (6.3) By allowing this stretching effect that can affect some thin shells made of hyper-elastic materials due to the Poisson effect, the author increases the range of applicability of the formulation to some particular thin-shell structures like balloons. 6.2 Deﬁnition of local axes Since the original mesh is deﬁned with linear triangles, the local axes used to deﬁne the principal directions of the material e′ 1 and e′ 2 are uniform CHAPTER 6. CONTINUUM-BASED FORMULATION 68 over the surface of the ﬂat triangle (see ﬁgure 6.2). Therefore a rotation is needed to obtain the material local axes t1 and t2. To that end, the author uses the normal vectors ˆy and t3. The rotation vector and the rotation angle are deﬁned as follows: ω = ˆy × t3 ∥ˆy × t3∥ (6.10) α = arcsin∥ˆy × t3∥ (6.11) (6.10) (6.11) where t3 has been deﬁned in equation (5.13) on page 49, and ˆy = (Xj −Xi) × (Xk −Xi) ∥(Xj −Xi) × (Xk −Xi)∥ (6.12) (6.12) Figure 6.2: A set of orthogonal local axes (t1, t2, t3) is deﬁned in the ref- erence conﬁguration in order to establish the principal directions of the material over the curved surface of the element. The user deﬁnes the ma- terial local axes (e′ 1, e′ 2, ˆy) on the ﬂat triangle geometry. The vector t3 is deﬁned in the same way as n in the current conﬁguration. Figure 6.2: A set of orthogonal local axes (t1, t2, t3) is deﬁned in the ref- erence conﬁguration in order to establish the principal directions of the material over the curved surface of the element. The user deﬁnes the ma- terial local axes (e′ 1, e′ 2, ˆy) on the ﬂat triangle geometry. The vector t3 is deﬁned in the same way as n in the current conﬁguration. Then, using the Rodrigues formula, we can easily compute the other two local vectors on the curved surface. Let R be the rotation matrix de- ﬁned by the vector ω and the angle α. R(ω, α) = cos α · (I −ω ⊗ω) + (ω ⊗ω) + sin α · eijkωj (6.13) (6.13) Where eijk stands for the permutation tensor. Then, Where eijk stands for the permutation tensor. Then, t1(ξ, η) = R · e′ 1 (6.14) t2(ξ, η) = R · e′ 2 (6.15) (6.14) (6.15) and the matrix of change of coordinates from local axes to global axes is T = [t1 t2 t3] (6.16) (6.16) 6.3. DESCRIPTION OF THE DEFORMATION 69 6.3 Description of the deformation In this section the equations and concepts from section 4.3 will be adapted to the Bézier description of the triangle. A point with parametric coordi- nates (ξ,η,ζ) in the element has the global coordinates given by the position of the midsurface and the orientation of the normal vector to the midsur- face. x(ξ, η, ζ) = ¯x(ξ, η) + λ h ζ n(ξ, η) = p · N + λ h ζ n (6.17) X(ξ, η, ζ) = ¯X(ξ, η) + h ζ N (ξ, η) = P · N + h ζ N (6.18) (6.17) (6.18) (6.18) Equations (4.9) to (4.13) remain unchanged. Let’s develop the expression for the components of equation (4.12). Equations (4.9) to (4.13) remain unchanged. Let’s develop the expression for the components of equation (4.12). ∂x ∂ξ = p · Lξ + λ h ζ n,ξ (6.19) ∂x ∂η = p · Lη + λ h ζ n,η (6.20) ∂x ∂ζ = λ h n (6.21) (6.19) (6.20) (6.21) where n,ξ = ∂n ∂ξ = (p · Lξ,ξ) × (p · Lη) + (p · Lξ) × (p · Lη,ξ) ∥(p · Lξ) × (p · Lη)∥ · (I −n ⊗n) (6.22) n,η = ∂n ∂η = (p · Lξ,η) × (p · Lη) + (p · Lξ) × (p · Lη,η) ∥(p · Lξ) × (p · Lη)∥ · (I −n ⊗n) (6.23) Following the thought expressed in footnote 3 on page 67, the value of λ has been considered constant throughout the domain of the element. This is because it is very difﬁcult to obtain a derivative expression of λ. But in addition, its value is expected to change very little; except for the step vari- ations across the layers of the laminate. These can be taken into account when computing the stiffness matrix of the element (see chapter 7). Most authors consider that the second order terms of the Green-Lagrange deformation tensor (E) are negligible. It is evident that by making that as- sumption they are making an error that adds to all the other assumptions. The author intends to make as little assumptions as possible in order to reduce the errors in the modeling. Therefore, the second order terms will be fully considered. The author will show that in the current framework, doing so does not increase the complexity of the element. 6.3 Description of the deformation Let’s recall the deﬁnition of some tensors: J = ï∂X ∂ξ ∂X ∂η ∂X ∂ζ ò (6.24) j = ï∂x ∂ξ ∂x ∂η ∂x ∂ζ ò (6.25) F = ∂x ∂X = j · J−1 (6.26) C = F T · F = j · J−1T · j · J−1 (6.27) J = ï∂X ∂ξ ∂X ∂η ∂X ∂ζ ò (6.24) j = ï∂x ∂ξ ∂x ∂η ∂x ∂ζ ò (6.25) F = ∂x ∂X = j · J−1 (6.26) C = F T · F = j · J−1T · j · J−1 (6.27) (6.24) CHAPTER 6. CONTINUUM-BASED FORMULATION 70 E = 1 2 (C −I) (6.28) E′ = T T · E · T (6.29) S′ = D: E′ (6.30) (6.28) (6.29) (6.30) (6.30) Now, similarly to the procedure followed in section 4.3, let’s deﬁne some useful tensors: Now, similarly to the procedure followed in section 4.3, let’s deﬁne some useful tensors: G = J−1 (6.31) A = G · T (6.32) g = jT · j (6.33) (6.31) (6.32) (6.33) (6.32) (6.33) Notice that G and A are deﬁned in the reference conﬁguration, while g is deﬁned in the current conﬁguration. Using these tensors we can rewrite equations (6.27) to (6.29): C = F T · F = j · J−1T · j · J−1 = GT · g · G (6.34) E = 1 2 (C −I) = 1 2 (GT · g · G −I) (6.35) E′ = T T · E · T = 1 2 (T T · GT · g · G · T −I) = 1 2 (AT · g · A −I) (6.36) (6.34) (6.35) The tensor g is a symmetric tensor. Furthermore, since x,ξ ⊥x,ζ and x,η ⊥x,ζ ⇒g13 = g23 = g31 = g32 = 0 (6.37) then g is of the form x,ξ ⊥x,ζ and x,η ⊥x,ζ ⇒g13 = g23 = g31 = g32 = 0 (6.37) (6.37) g =   g11 g12 0 g22 0 sym. 6.3 Description of the deformation g33   (6.38) (6.38) The components of g can be expressed in the following way: The components of g can be expressed in the following way: gαβ = LT ξα·pT ·p·Lξβ+λ h ζ (LT ξα·pT ·n,ξβ +LT ξβ·pT ·n,ξα )+λ2 h2 ζ2 n,T ξα ·n,ξβ ∀α, β = 1 ÷ 2 (6.39) where Lξ1 ≡Lξ Lξ2 ≡Lη (6.40) (6.40) and n,ξ1 ≡n,ξ n,ξ2 ≡n,η (6.41) (6.41) The explicit value of g33 is not of interest because given the structure of A it is only meaningful towards the determination of ε3. In fact, the value of ε3 can only be obtained using equation (6.7), because it does not generate any work in the plane stress assumption. 6.4. DECOMPOSITION INTO THICKNESS-INDEPENDENT TENSORS 71 6.4 Decomposition into thickness-independent tensors This section provides the necessary proofs to decompose the tensors that govern the deformation of the shell element into sums of tensors that are constant in the thickness direction and which eventually may be affected by scalar components that provide the variation in the thickness direc- tion. This is useful in order to subsequently pre-integrate through-the- thickness, which is invaluable in order to reduce the computational cost of the calculations. It can be proved that A is of the form It can be proved that A is of the form A =   A11 A12 0 A21 A22 0 0 0 A33   (6.42) (6.42) Let’s proceed with the demonstration. In order to analyze A we will use it’s inverse A−1: A−1 = (G · T )−1 = T −1 · G−1 = T T · J = = Ñ ↑ ↑ ↑ t1 t2 t3 ↓ ↓ ↓ éT · Ñ ↑ ↑ ↑ P · Lξ + hζN ,ξ P · Lη + hζN ,η h t3 ↓ ↓ ↓ é = = Ñt1 · (P · Lξ + h ζ N ,ξ ) t1 · (P · Lη + h ζ N ,η ) 0 t2 · (P · Lξ + h ζ N ,ξ ) t2 · (P · Lη + h ζ N ,η ) 0 0 0 h é (6.43) A−1 1 1 2 2 := Åt1 · (P · Lξ + h ζ N ,ξ ) t1 · (P · Lη + h ζ N ,η ) t2 · (P · Lξ + h ζ N ,ξ ) t2 · (P · Lη + h ζ N ,η ) ã (6.44) det A−1 = det(T T · J) = det T T det J = det J := J (6.45) b t l d t A−1 d t A−1 d t A−1 h d t A−1 (6 46) A−1 1 1 2 2 := Åt1 · (P · Lξ + h ζ N ,ξ ) t1 · (P · Lη + h ζ N ,η ) t2 · (P · Lξ + h ζ N ,ξ ) t2 · (P · Lη + h ζ N ,η ) ã (6.44) det A−1 = det(T T · J) = det T T det J = det J := J (6.45) but also . . . 6.4 Decomposition into thickness-independent tensors det A−1 = det A−1 1 1 2 2 det A−1 33 = h det A−1 1 1 2 2 (6.46) then: det A−1 1 1 2 2 = J h (6.47) det A−1 = det(T T · J) = det T T det J = det J := J (6.45) but also . . . det A−1 = det A−1 1 1 2 2 det A−1 33 = h det A−1 1 1 2 2 (6.46) J then: det A−1 1 1 2 2 = J h (6.47) A 1 1 2 2 = Å A−1 1 1 2 2 ã−1 = h J (A0 + h ζ A1) = = h J Å t2 · (P · Lη + h ζ N ,η ) −t1 · (P · Lη + h ζ N ,η ) −t2 · (P · Lξ + h ζ N ,ξ ) t1 · (P · Lξ + h ζ N ,ξ ) ã (6.48) A0 = Å tT 2 · P · Lη −tT 1 · P · Lη −tT 2 · P · Lξ tT 1 · P · Lξ ã (6.49) A1 = Å t2 · N ,η −t1 · N ,η −t2 · N ,ξ t1 · N ,ξ ã (6.50) A33 = 1 h (6.51) = h J Å t2 · (P · Lη + h ζ N ,η ) −t1 · (P · Lη + h ζ N ,η ) −t2 · (P · Lξ + h ζ N ,ξ ) t1 · (P · Lξ + h ζ N ,ξ ) ã (6.48) A0 = Å tT 2 · P · Lη −tT 1 · P · Lη −tT 2 · P · Lξ tT 1 · P · Lξ ã (6.49) A1 = Å t2 · N ,η −t1 · N ,η −t2 · N ,ξ t1 · N ,ξ ã (6.50) A 1 (6 51) A33 = 1 h (6.51) (6.51) This concludes the demonstration. This demonstration is going to proof very useful in the coming sections because not only have we provided the explicit expression for A, but we have also found a decomposition with matrices that are constant in the thickness direction. 72 CHAPTER 6. CONTINUUM-BASED FORMULATION By rearranging the terms of equation (6.36) we can deﬁne a new tensor Q that permits the use of Voigt notation. 6.4 Decomposition into thickness-independent tensors E′ IJ = 1 2(AT Ii gij AjJ −δIJ) = 1 2(AiI AjJ gij −δIJ) (6.52) E′ V oigt =   E′ 11 E′ 22 E′ 12 + E′ 21  = 1 2(Q · gV oigt −IV oigt) not = 1 2(Q · g −I) = = 1 2 â   A2 11 A2 21 A11 A21 A2 12 A2 22 A12 A22 2 A11 A12 2 A22 A21 A11 A22 + A12 A21   \| {z } Q ·   g11 g22 g12 + g21  −   1 1 0   ì (6.53) (6.52) ì (6.53) However, since Q contains only the components of A 1 1 2 2 , we can make use of its decomposition in equation (6.48) to also decompose Q: E′ 1 1 2 2 = 1 2 Å AT 1 1 2 2 · g 1 1 2 2 · A 1 1 2 2 −I 1 1 2 2 ã = = 1 2 ï h2 J2 (A0 + h ζ A1)T · g 1 1 2 2 · (A0 + h ζ A1) −I 1 1 2 2 ò = = 1 2 ï h2 J2 AT 0 · g 1 1 2 2 · A0+ + h3 ζ J2 AT 0 · g 1 1 2 2 · A1 + AT 1 · g 1 1 2 2 · A0 + + h4 ζ2 J2 AT 1 · g 1 1 2 2 · A1 −I 1 1 2 2 ò (6.54) AT 0 · g 1 1 2 2 · A0 V oigt = Q0 · g (6.55) AT 0 · g 1 1 2 2 · A1 + AT 1 · g 1 1 2 2 · A0 V oigt = Q1 · g (6.56) AT 1 · g 1 1 2 2 · A1 V oigt = Q2 · g (6.57) E′ = 1 2 ï h2 J2 Q0 + h ζ Q1 + h2 ζ2 Q2 · g −I ò (6.58) Q = h2 J2 Q0 + h3 ζ J2 Q1 + h4 ζ2 J2 Q2 (6.59) E′ 1 1 2 2 = 1 2 Å AT 1 1 2 2 · g 1 1 2 2 · A 1 1 2 2 −I 1 1 2 2 ã = = 1 2 ï h2 J2 (A0 + h ζ A1)T · g 1 1 2 2 · (A0 + h ζ A1) −I 1 1 2 2 ò = = 1 2 ï h2 J2 AT 0 · g 1 1 2 2 · A0+ + h3 ζ J2 AT 0 · g 1 1 2 2 · A1 + AT 1 · g 1 1 2 2 · A0 + (6.54) 1 2 ï h2 J2 AT 0 · g 1 1 2 2 · A0+ (6.54) (6.54) (6.55) AT 0 · g 1 1 2 2 · A1 + AT 1 · g 1 1 2 2 · A0 V oigt = Q1 · g (6.56) AT 1 · g 1 1 2 2 · A1 V oigt = Q2 · g (6.57) E′ = 1 2 ï h2 J2 Q0 + h ζ Q1 + h2 ζ2 Q2 · g −I ò (6.58) Q = h2 J2 Q0 + h3 ζ J2 Q1 + h4 ζ2 J2 Q2 (6.59) Q0 =   (tT 2 · P · Lη)2 (tT 2 · P · Lξ)2 −(tT 2 · P · Lη)(tT 2 · P · Lξ) (tT 1 · P · Lη)2 (tT 1 · P · Lξ)2 −(tT 1 · P · Lη)(tT 1 · P · Lξ) −2(tT 2 · P · Lη)· −2(tT 2 · P · Lξ)· (tT 2 · P · Lη)(tT 1 · P · Lξ)+ ·(tT 1 · P · Lη) ·(tT 1 · P · Lξ) +(tT 1 · P · Lη)(tT 2 · P · Lξ)   (6.60) Q0 =   (tT 2 · P · Lη)2 (tT 2 · P · Lξ)2 −(tT 2 · P · Lη)(tT 2 · P · Lξ) (tT 1 · P · Lη)2 (tT 1 · P · Lξ)2 −(tT 1 · P · Lη)(tT 1 · P · Lξ) −2(tT 2 · P · Lη)· −2(tT 2 · P · Lξ)· (tT 2 · P · Lη)(tT 1 · P · Lξ)+ ·(tT 1 · P · Lη) ·(tT 1 · P · Lξ) +(tT 1 · P · Lη)(tT 2 · P · Lξ)   (6.60) Q0 =   (tT 2 · P · Lη)2 (tT 2 · P · Lξ)2 −(tT 2 · P · Lη)(tT 2 · P · Lξ) (tT 1 · P · Lη)2 (tT 1 · P · Lξ)2 −(tT 1 · P · Lη)(tT 1 · P · Lξ) −2(tT 2 · P · Lη)· −2(tT 2 · P · Lξ)· (tT 2 · P · Lη)(tT 1 · P · Lξ)+ ·(tT 1 · P · Lη) ·(tT 1 · P · Lξ) +(tT 1 · P · Lη)(tT 2 · P · Lξ)   (6.60) 6.5. 6.4 Decomposition into thickness-independent tensors Finally, by combining equation (6.63) into equation (6.58), we obtain a fully decomposed expression for the Green- Lagrange strain tensor in local coordinates of the material. All the vectors and tensors in equation (6.67) are constant in the thickness direction. We shall remark again that the Green-Lagrange strain tensor (E’) is deﬁned in the reference conﬁguration. E′ = 1 2 ï h2 J2 Q0 + h ζ Q1 + h2 ζ2 Q2 · (gm + 2 λ h ζ gb + λ2 h2 ζ2 gn) −I ò (6.67) ò (6.67) 6.4 Decomposition into thickness-independent tensors PRINCIPLE OF VIRTUAL WORK 73 Q1 =   2(tT 2 · P · Lη)· 2(tT 2 · P · Lξ)· −[(tT 2 · P · Lη)(t2 · N ,ξ )+ ·(t2 · N ,η ) ·(t2 · N ,ξ ) +(tT 2 · P · Lξ)(t2 · N ,η )] 2(tT 1 · P · Lη)· 2(tT 1 · P · Lξ)· −[(tT 1 · P · Lη)(t1 · N ,ξ )+ ·(t1 · N ,η ) ·(t1 · N ,ξ ) +(tT 1 · P · Lξ)(t1 · N ,η )] −2[(tT 2 · P · Lη)· −2[(tT 2 · P · Lξ)· [(tT 2 · P · Lη)(t1 · N ,ξ )+ ·(t1 · N ,η )+ ·(t1 · N ,ξ )+ +(tT 1 · P · Lξ)(t2 · N ,η )+ +(tT 1 · P · Lη)· +(tT 1 · P · Lξ)· +(tT 1 · P · Lη)(t2 · N ,ξ )+ ·(t2 · N ,η )] ·(t2 · N ,ξ )] +(tT 2 · P · Lξ)(t1 · N ,η )]   (6.61) Q1 =   2(tT 2 · P · Lη)· 2(tT 2 · P · Lξ)· −[(tT 2 · P · Lη)(t2 · N ,ξ )+ ·(t2 · N ,η ) ·(t2 · N ,ξ ) +(tT 2 · P · Lξ)(t2 · N ,η )] 2(tT 1 · P · Lη)· 2(tT 1 · P · Lξ)· −[(tT 1 · P · Lη)(t1 · N ,ξ )+ ·(t1 · N ,η ) ·(t1 · N ,ξ ) +(tT 1 · P · Lξ)(t1 · N ,η )] −2[(tT 2 · P · Lη)· −2[(tT 2 · P · Lξ)· [(tT 2 · P · Lη)(t1 · N ,ξ )+ ·(t1 · N ,η )+ ·(t1 · N ,ξ )+ +(tT 1 · P · Lξ)(t2 · N ,η )+ +(tT 1 · P · Lη)· +(tT 1 · P · Lξ)· +(tT 1 · P · Lη)(t2 · N ,ξ )+ ·(t2 · N ,η )] ·(t2 · N ,ξ )] +(tT 2 · P · Lξ)(t1 · N ,η )]   (6.61) Q2 =   (t2 · N ,η )2 (t2 · N ,ξ )2 −(t2 · N ,η )(t2 · N ,ξ ) (t1 · N ,η )2 (t1 · N ,ξ )2 −(t1 · N ,η )(t1 · N ,ξ ) −2(t2 · N ,η )· −2(t2 · N ,ξ )· (t2 · N ,η )(t1 · N ,ξ )+ ·(t1 · N ,η ) ·(t1 · N ,ξ ) +(t1 · N ,η )(t2 · N ,ξ )   (6.62) (6.62) But g can also be decomposed and equation (6.39) is rewritten as: g = gm + 2 λ h ζ gb + λ2 h2 ζ2 gn (6.63) gm =   LT ξ · pT · p · Lξ LT η · pT · p · Lη 2 LT ξ · pT · p · Lη   (6.64) gb =   LT ξ · pT · nξ LT η · pT · nη LT ξ · pT · nη + LT η · pT · nξ   (6.65) gn =   nξ · nξ nη · nη 2 nξ · nη   (6.66) (6.65) (6.66) where gm, gb and gn contain respectively the membrane, bending and second order terms of the deformation. 6.5 Principle of virtual work The second Piola-Kirchhoff stress tensor (S) is work-conjugate of the Green- Lagrange strain tensor (E). So the density of work produced by the defor- CHAPTER 6. CONTINUUM-BASED FORMULATION 74 mation is expressed as: mation is expressed as: U = E : S = E′ : S′ (6.68) (6.68) The expression of the virtual work principle for the internal virtual work of an element is: The expression of the virtual work principle for the internal virtual work of an element is: δW int e = ZZZ Ω0e δE′ : S′ dΩ0 = ZZZ Ω0e δE′T · D · E′ dΩ0 = = 1 4 ZZZ Ω0e δgT · QT · D · (Q · g −I) dΩ0 (6.69 (6.69) and the external virtual work is: 6.6. SUMMARY 75 B is a matrix with 3 rows and three times as many columns as nodes exist in the patch of the element. Usually, in the literature, other authors use this notation for the matrix relating the nodal displacements with the de- formation tensor. The author is not using a formulation based on displace- ments but based on nodal coordinates, so this possibility is forfeited in this work. Besides, in terms of derivatives, the product of Q and B would be equivalent to the matrix B used in the literature. The author prefers to keep them separated in order to simplify the decomposed expression with respect to the thickness coordinate. Now we can rewrite equations (6.69) and (6.70) using the expressions above: δW int e = 1 4δ˜xrT ZZZ Ω0e BT · QT · D · (Q · g −I) dΩ0 (6.74) δW ext e = δ˜xrT ñZZZ Ve Å ∂x ∂˜xr ãT · qV dV + ZZ Se Å ∂x ∂˜xr ãT · qS dS+ + Z Γe Å ∂x ∂˜xr ãT · qΓ dΓ + ˜qr ô (6.75) (6.74) (6.75) Equaling the expression for the internal virtual work and the external virtual work we establish a condition for equilibrium. Since the equality must hold for any arbitrary value of δ˜xr, then we can write: 1 4 ZZZ Ω0e BT · QT · D · (Q · g −I) dΩ0 = ZZZ Ve Å ∂x ∂˜xr ãT · qV dV + + ZZ Se Å ∂x ∂˜xr ãT · qS dS + Z Γe Å ∂x ∂˜xr ãT · qΓ dΓ + ˜qr (6.76) (6.76) The left hand side contains what is commonly referred to as the vector of equivalent nodal loads to the internal stresses of the element (F int e ). The right hand side contains what is commonly referred to as the vector of equivalent external nodal loads (F ext e ). Equation (6.76) is a nonlinear equation. In order to solve it, we use the Newton-Raphson method. Rk e = F int e (˜xr)k −F ext e (˜xr)k (6.77) Rk+1 e ≈∂Rk e ∂˜xr · ∆˜xr = 0 (6.78) (˜xr)k+1 = (˜xr)k + ∆˜xr (6.79) (6.77) and the external virtual work is: and the external virtual work is: δW ext e = ZZZ Ve δx · qV dV + ZZ Se δx · qS dS + Z Γe δx · qΓ dΓ + rtotal X i=1 δxi · qi (6 70 (6.70) (6.70) where qV , qS, qΓ and qi refer to the volume loads, surface loads, linear loads and nodal loads, respectively. Notice that the integrals in equa- tion (6.70) do not specify whether the integral is deﬁned in the reference or in the current conﬁguration. That is because some loads will be deﬁned in the reference conﬁguration (e.g. self weight) and some other loads will be deﬁned in the deformed conﬁguration (e.g. a follower pressure load). Because the BEST element uses neighboring nodes to deﬁne the geom- etry, the author refers to the set of nodes that surround the triangle as ˜xr. This is a column vector which has the different vector coordinates of each of the surrounding nodes stacked one after the other (with the coordinates of the three nodes of the triangle taking the ﬁrst 9 positions). ˜xr :=   x1 1 x1 2 x1 3 ... xk 1 xk 2 xk 3 ... xrtotal 1 xrtotal 2 xrtotal 3   (6.71) ˜xr :=   x1 1 x1 2 x1 3 ... xk 1 xk 2 xk 3 ... xrtotal 1 xrtotal 2 xrtotal 3   (6.71) (6.71) Where rtotal is the total number of nodes in the patch of elements that surrounds the triangle. For a regular mesh, rtotal = 12. Where rtotal is the total number of nodes in the patch of elements that surrounds the triangle. For a regular mesh, rtotal = 12. When taking derivatives of the different magnitudes with respect to the coordinates of the nodes of the BEST element, all the coordinates of the neighboring nodes need to be accounted for. For example: δg = ∂g ∂˜xr δ˜xr = B · δ˜xr (6.72) δx = ∂x ∂˜xr δ˜xr (6.73) (6.72) (6.73) 6.6. SUMMARY 6.6 Summary This chapter presents the description of the deformation of the BEST ele- ment for a total Lagrangian formulation and for a modiﬁed version of the Kirchhoff-Love hypothesis of the shell mechanics in which the constraint on the thickness change has been relaxed while assuming a plane stress state of the material. Under the above conditions the author has obtained a decomposition of the Green-Lagrange strain tensor in local coordinates (see equation (6.58)), CHAPTER 6. CONTINUUM-BASED FORMULATION 76 where the components are independent through the thickness except for scalar coefﬁcients. This will prove very valuable in the next chapter in or- der to perform the efﬁcient and geometrically accurate through-the-thickness pre-integration of the stiffness matrix in the total Lagrangian framework. where the components are independent through the thickness except for scalar coefﬁcients. This will prove very valuable in the next chapter in or- der to perform the efﬁcient and geometrically accurate through-the-thickness pre-integration of the stiffness matrix in the total Lagrangian framework. 7.1 Material and geometric stiffness matrices The stiffness matrix stems from the linearization of the equivalent inter- nal nodal loads. ∂F int e (˜xr) ∂˜xr = 1 4 ∂ ∂˜xr ZZZ Ω0e BT · QT · D · (Q · g −I) dΩ0 = = 1 4 ZZZ Ω0e BT · QT · D · Q · B dΩ0 \| {z } KM + + 1 2 ZZZ Ω0e ∂BT ∂˜xr · QT · D · E′ dΩ0 \| {z } KG (7.2) {z KG Where KM is the material stiffness matrix, and KG is the geometric stiff- ness matrix. Let’s proceed with the details of the computations in the parametric space of coordinates. KM = 1 4 ZZZ Ω0e BT · QT · D · Q · B dΩ0 = = 1 4 Z 1 0 Z 1−ξ 0 Z 1 2 −1 2 BT · QT · D · Q · B · J dξ dη dζ (7.3) KG = 1 2 ZZZ Ω0e ∂BT ∂˜xr · QT · D · E′ dΩ0 = 1 2 Z 1 0 Z 1−ξ 0 Z 1 2 −1 2 ∂BT ∂˜xr · QT · D · E′ · J dξ dη dζ (7.4) (7.3) (7.4) The matrix B and the tensor in 3 dimensions ∂B ∂˜xr have not been thor- oughly deﬁned, yet. Let’s recall the ﬁrst time B has been introduced in equation (6.72) and the decomposition of g in equation (6.63): B = ∂g ∂˜xr = ∂gm ∂˜xr + 2 λ h ζ ∂gb ∂˜xr + λ2 h2 ζ2 ∂gn ∂˜xr = = Bm + 2 λ h ζ Bb + λ2 h2 ζ2 Bn (7.5) Bm = ∂gm ∂˜xr = ∂gm ∂p : ∂p ∂˜xr (7.6) Bb = ∂gb ∂˜xr = ∂gb ∂p : ∂p ∂˜xr (7.7) Bn = ∂gn ∂˜xr = ∂gn ∂p : ∂p ∂˜xr (7.8) (7.5) (7.7) (7.8) For the sake of brevity, the author will request from the reader an effort of abstraction. The formulæ will try to be presented in condensed form when- ever possible. For example, the above expressions can be jointly presented as: For the sake of brevity, the author will request from the reader an effort of abstraction. The formulæ will try to be presented in condensed form when- ever possible. Chapter 7 Construction of the tangent stiffness matrix T HIS CHAPTER PRESENTS THE CONSTRUCTION of the tangent stiffness matrix of the BEST element as required per equation (6.78), which implies the solution of a linear system of equations. On the one side, the linearization of the expression of the equivalent internal nodal loads leads to obtaining the classical stiffness matrix of the element. On the other side, the linearization of the equivalent external nodal loads leads to the tangent matrix of the nodal loads. In many cases, the external loads are constant, but it is not infrequent that an engineer faces a structure with varying loads. T H ∂Re ∂˜xr = ∂F int e (˜xr) ∂˜xr −∂F ext e (˜xr) ∂˜xr (7.1) (7.1) Section 7.1 presents the computation of the material and the geometric stiffness matrices of the element. The geometric stiffness matrix appears only in geometrically nonlinear problems. This is the case of shells. This is accentuated in the case of very thin shells whose structures can undergo very large displacements. An efﬁcient integration of the stiffness matrix is necessary in order to keep the computational cost competitive with other ﬁnite elements. This issue is presented in section 7.2. Finally, the compu- tation of the tangent loads matrix is presented in section 7.3. The BEST element is also different in this aspect, because the shape of the element depends not only on the element’s nodal positions, but also on the positions of the patch of nodes surrounding the element. 77 CHAPTER 7. TANGENT STIFFNESS MATRIX 78 7.1 Material and geometric stiffness matrices For example, the above expressions can be jointly presented as: Bχ = ∂gχ ∂˜xr = ∂gχ ∂p : ∂p ∂˜xr ∀χ = {m, b, n} (7.9) (7.9) 7.1. MATERIAL AND GEOMETRIC STIFFNESS MATRICES 79 And the terms included in the following expression And the terms included in the following expression ∂B ∂˜xr = ∂2g (∂˜xr)2 = ∂2gm (∂˜xr)2 + 2 λ h ζ ∂2gb (∂˜xr)2 + λ2 h2 ζ2 ∂2gn (∂˜xr)2 = = ∂Bm ∂˜xr + 2 λ h ζ ∂Bb ∂˜xr + λ2 h2 ζ2 ∂Bn ∂˜xr (7.10) (7.10) can be further expanded as follows: can be further expanded as follows: ∂Bχ ∂˜xr = ∂2gχ (∂˜xr)2 = Å ∂2gχ (∂p)2 : ∂p ∂˜xr ã : ∂p ∂˜xr + ∂gχ ∂p : ∂2p (∂˜xr)2 = = Ç ∂2gχ(i) ∂p(jq) ∂p(ls) : ∂p(ls) ∂˜xr (w) å : ∂p(jq) ∂˜xr (v) + ∂gχ(i) ∂p(jq) : ∂2p(jq) ∂˜xr (v) ∂˜xr (w) = = ∂Bχ(iv) ∂˜xr (w) ∀χ = {m, b, n} ∀i = 1 ÷ 3 ∀v, w = 1 ÷ 3 rtotal (7.11) (7.11) Since these matrices start to get rather large in the number of dimensions, the author prefers to express some of the deﬁnitions using indicial notation for further clarity. Indices are written in parentheses to distinguish them from subscripts. 7.1 Material and geometric stiffness matrices ∂gχ ∂p = ∂gχ(k) ∂p(lv) ∀χ = {m, b, n} ∀k, l = 1 ÷ 3 ∀v = 1 ÷ 10 (7.12) ∂gm(1) ∂p(lv) = 2 p(lj)Lξ(j)Lξ(v) (7.13) ∂gm(2) ∂p(lv) = 2 p(lj)Lη(j)Lη(v) (7.14) ∂gm(3) ∂p(lv) = 2 (p(lj)Lξ(j)Lη(v) + p(lj)Lη(j)Lξ(v)) (7.15) (7.12) (7.15) Again, the author makes an abuse of notation and switches indistinctively between tensorial notation and Voigt notation for the components of g (see equation (4.23) on page 37); in order to condense the above expressions in the following manner: ∂gm(αβ) ∂p(lv) = 2 p(lj)Lξα(j)Lξβ(v) ∀α, β = 1, 2 (7.16) ∂gb(αβ) ∂p(lv) = n,ξα(l) Lξβ(v) + p(ji)Lξβ(i) ∂n,ξα(j) ∂p(lv) ∀α, β = 1, 2 (7.17) ∂gn(αβ) ∂p(lv) = 2 n,ξα(i) ∂n,ξβ(i) ∂p(lv) ∀α, β = 1, 2 (7.18) ∂2gm(αβ) ∂p(jq) ∂p(ls) = 2δ(jl)Lξα(q)Lξβ(s) ∀α, β = 1, 2 (7.19) ∂2gb(αβ) ∂p(jq) ∂p(ls) = ∂n,ξα(j) ∂p(ls) Lξβ(q) + ∂n,ξβ(l) ∂p(jq) Lξα(s) + p(jk)Lξα(k) ∂2n,ξβ(j) ∂p(jq) ∂p(ls) ∀α, β = 1, 2 (7.20) ∂2gm(αβ) ∂p(jq) ∂p(ls) = 2δ(jl)Lξα(q)Lξβ(s) ∀α, β = 1, 2 (7.19) ∂2gb(αβ) ∂ ∂ = ∂n,ξα(j) ∂ Lξβ(q) + ∂n,ξβ(l) ∂ Lξα(s) + p(jk)Lξα(k) ∂2n,ξβ(j) ∂ ∂ ∂2gm(αβ) ∂p(jq) ∂p(ls) = 2δ(jl)Lξα(q)Lξβ(s) ∀α, β = 1, 2 (7.19) ∂2gb(αβ) ∂p(jq) ∂p(ls) = ∂n,ξα(j) ∂p(ls) Lξβ(q) + ∂n,ξβ(l) ∂p(jq) Lξα(s) + p(jk)Lξα(k) ∂2n,ξβ(j) ∂p(jq) ∂p(ls) ∀α, β = 1, 2 (7.20) CHAPTER 7. TANGENT STIFFNESS MATRIX 80 ∂2gn(αβ) ∂p(jq) ∂p(ls) = 2 Ç ∂n,ξα(i) ∂p(jq) ∂n,ξβ(i) ∂p(ls) + n,ξβ(i) ∂2n,ξα(i) ∂p(jq) ∂p(ls) å ∀α, β = 1, 2 (7.21) (7.21) The expressions for ∂n,ξ ∂p , ∂n,η ∂p , ∂2n,ξ (∂p)2 and ∂2n,η (∂p)2 become rather complex. F thi th th f t t th i di D ( ) ( ) For this reason the author prefers to present them in appendix D. 7.1.1 Derivatives of the control points’ coordinates Similarly to the procedure followed in section 5.3, the derivatives of the control points’ coordinates will be determined by groups. First, the corner control points, which are coincident with the triangle’s nodes. Second, the control points in the boundaries of the Bézier triangle. And third, the central control point. Using equation (5.14), it is straight forward to ﬁnd the derivatives of the corner control points. ∂pi ∂˜xr = ∂xi ∂˜xr = ∂xi (l) ∂xh (v) = δihδ(lv) ∀i, l, v = 1 ÷ 3 ∀h = 1 ÷ rtotal (7.22) (7.22) ∂2pi (∂˜xr)2 = 0 (7.23) (7.23) The control points of the Bézier triangle’s boundary are determined by solving a linear system of 3 equations. In order to reduce the computa- tional cost, we will take advantage of the fact that the system matrix has to be inverted. We will not invert any other system matrix for the determi- nation of the derivatives. So deriving by parts equation (5.16) and solving for ∂pij ∂˜xr we obtain: for ∂p j ∂˜xr we obtain: ∂pij ∂˜xr = ∂pij (m) ∂xh (v) = Aij−1 p(ml) ∂bij p(l) ∂xh (v) − ∂Aij p(ln) ∂xh (v) pij (n) ! ∀i, j = 1 ÷ 3 \| i ̸= j ∀l, m, n, v = 1 ÷ 3 ∀h = 1 ÷ rtotal (7.24) Aij p · ∂pij ∂˜xr = ∂bij p ∂˜xr −∂Aij p ∂˜xr · pij (7.25) ∂pij ∂˜xr = ∂pij (m) ∂xh (v) = Aij−1 p(ml) ∂bij p(l) ∂xh (v) − ∂Aij p(ln) ∂xh (v) pij (n) ! (7.24) ∀i, j = 1 ÷ 3 \| i ̸= j ∀l, m, n, v = 1 ÷ 3 ∀h = 1 ÷ rtotal Aij p · ∂pij ∂˜xr = ∂bij p ∂˜xr −∂Aij p ∂˜xr · pij (7.25) (7.25) Deriving again by parts equation (7.25) and solving for ∂2pij (∂˜xr)2 we obtain: ∂2pij (∂˜xr)2 = ∂2pij (m) ∂xh (v) ∂xs (w) = Aij−1 p(ml) ∂2bij p(l) ∂xh (v) ∂xs (w) − ∂2Aij p(ln) ∂xh (v) ∂xs (w) pij (n) − ∂Aij p(ln) ∂xh (v) ∂pij (n) ∂xs (w) − ∂Aij p(ln) ∂xs (w) ∂pij (n) ∂xh (v) ! 7.2. THROUGH-THE-THICKNESS PRE-INTEGRATION 7.2. THROUGH-THE-THICKNESS PRE-INTEGRATION 81 The central control point of the Bézier triangle is computed as an av- erage of three candidate points (see equation (5.17)). The derivatives of the central control point will then be the average of the candidate points’ derivatives. ∂p123 ∂˜xr = 1 3 3 X i=1 ∂p0i ∂˜xr (7.27) ∂2p123 (∂˜xr)2 = 1 3 3 X i=1 ∂2p0i (∂˜xr)2 (7.28) (7.27) (7.27) (7.28) Each of the candidate points is determined by solving a linear system of 3 equations. To compute their derivatives, an analogous procedure as the one used for the control points in the contour can be used. Therefore, we will derive by parts equation (5.18) and solve for ∂p0i ∂˜xr to obtain: Each of the candidate points is determined by solving a linear system of 3 equations. To compute their derivatives, an analogous procedure as the one used for the control points in the contour can be used. Therefore, we will derive by parts equation (5.18) and solve for ∂p0i ∂˜xr to obtain: ∂p0i ∂˜xr = ∂p0i (m) ∂xh (v) = A0i−1 p(ml) Ç∂b0i p(l) ∂xh (v) − ∂A0i p(ln) ∂xh (v) p0i (n) å ∀i, l, m, n, v = 1 ÷ 3 ∀h = 1 ÷ rtotal (7.29) A0i p · ∂p0i ∂˜xr = ∂b0i p ∂˜xr −∂A0i p ∂˜xr · p0i (7.30) (7.29) (7.30) Deriving again by parts equation (7.30) and solving for ∂2p0i (∂˜xr)2 we obtain: 2p0i ˜xr)2 = ∂2p0i (m) ∂xh (v) ∂xs (w) = A0i−1 p(ml) Ç ∂2b0i p(l) ∂xh (v) ∂xs (w) − ∂2A0i p(ln) ∂xh (v) ∂xs (w) p0i (n) − ∂A0i p(ln) ∂xh (v) ∂p0i (n) ∂xs (w) − ∂A0i p(ln) ∂xs (w) ∂p0i (n) ∂xh (v) å ∀i, l, m, n, v, w = 1 ÷ 3 ∀h, s = 1 ÷ rtotal (7.31) (7.31) ∀i, l, m, n, v, w = 1 ÷ 3 ∀h, s = 1 ÷ rtotal or ∂A0i p ∂˜xr , ∂2A0i p (∂˜xr)2 , ∂b0i p ∂˜xr and ∂2b0i p (∂˜xr)2 become rather com- The expressions for ∂A0i p ∂˜xr , ∂2A0i p (∂˜xr)2 , ∂b0i p ∂˜xr and ∂2b0i p (∂˜xr)2 become rather com- plex. For this reason the author prefers to present them in appendix D. ( ) ( ) plex. For this reason the author prefers to present them in appendix D. 7.1.1 Derivatives of the control points’ coordinates ∀i, j = 1 ÷ 3 \| i ̸= j ∀l, m, n, v, w = 1 ÷ 3 ∀h, s = 1 ÷ rtotal (7.26) (7.26) The expressions for ∂Aij p ∂˜xr , ∂2Aij p (∂˜xr)2 , ∂bij p ∂˜xr and ∂2bij p (∂˜xr)2 become rather com- plex. For this reason the author prefers to present them in appendix D. The expressions for ∂Aij p ∂˜xr , ∂2Aij p (∂˜xr)2 , ∂bij p ∂˜xr and ∂2bij p (∂˜xr)2 become rather com- plex. For this reason the author prefers to present them in appendix D. 1Stanley is regarded as the ﬁrst scholar to have successfully introduced this kind of de- composition in the through-the-thickness direction in order to pre-integrate the stiffness ma- trix. Oñate also cites different authors who have followed this approach, see [93, p. 640] and references therein. 7.2 Efﬁcient and geometrically accurate through-the-thickness pre-integration Equations (7.3) and (7.4) are to be integrated numerically. The author proposes to decouple the through-the-thickness integration ( R 1 2 −1 2 ) from the area integration ( R 1 0 R 1−ξ 0 ). 0 0 Regarding the integration over the area of the triangle, the ﬁrst thing to ﬁgure out is the order of integration necessary. Since the shape functions (N) are cubic, the deformation has a quadratic description. Therefore, the internal virtual energy is being described with fourth order functions. The numerical quadrature that needs the least number of integration points 2Yang et al. [141] explain that many shell elements are based on moderate rotation as- sumptions. Since rotations lead to rational expressions that imply complicated derivatives, researchers have sought for ways to circumvent those rational expressions and use aproxi- mated polynomial expressions instead [9, pp. 545–549]. In addition, in order to compensate for the errors introduced by the approximations of the large rotations, most ﬁnite element developers favored the Updated Lagrangian formulation, this was also the reason behind the development of the co-rotational shell elements [138]. CHAPTER 7. TANGENT STIFFNESS MATRIX 82 CHAPTER 7. TANGENT STIFFNESS MATRIX for a fourth order polynomial is a Gaussian quadrature, which needs 6 integration points over the triangle. The author uses the quadrature ob- tained by Dunavant in [33]. There are other authors who have also ob- tained fourth order quadratures for the triangle. The reader can refer to the references in [29] for a complete review on numerical quadratures on different domains. The beauty of Dunavant’s quadrature is that it is symmetric, all the weights are positive and all the points are inside the triangle (see ta- ble 7.1). Table 7.1: Weights and coordinates for fourth order integration using Gaussian quadrature over a triangle. Table 7.1: Weights and coordinates for fourth order integration using Gaussian quadrature over a triangle. Gauss point Weight Parametric coordinates Gauss point Weight Parametric coordinates 1 w1 (α1, β1) 2 w1 (β1, α1) 3 w1 (β1, β1) 4 w2 (α2, β2) 5 w2 (β2, α2) 6 w2 (β2, β2) w1 = 0.223381589678011 α1 = 0.108103018168070 β1 = 0.445948490915965 w2 = 0.109951743655322 α2 = 0.816847572980459 β2 = 0.091576213509771 Regarding the integration through the thickness of the triangle, there are two options. The ﬁrst one, is to also perform a Gaussian integration through the thickness and multiply each point in the triangle quadrature by the number of evaluation points through the thickness. The problem with this approach is that at least 5 evaluation points in the thickness di- rection are necessary. This implies a total of not less than 30 evaluation points. The number of evaluation points is a direct measure of the cost of building the stiffness matrix of the element. Fortunately, the decision for developing a total Lagrangian formulation comes in very handy. As it has been shown in equations (6.59), (6.63), (6.67), (7.5) and (7.10); the tensors that conform the material stiffness matrix and the geometric stiff- ness matrix in equations (7.3) and (7.4) can be decomposed into constant tensors in the thickness direction and scalar factors that are affected by the thickness coordinate ζ. Furthermore, these scalar factors that have variation in the thickness direction, are deﬁned in the reference conﬁgura- tion and do not change in the current conﬁguration. Therefore, the author follows what other scholars have done for other elements [120]1 and pro- poses a through-the-thickness pre-integration of all the terms, which can 7.2. THROUGH-THE-THICKNESS PRE-INTEGRATION 83 . THROUGH-THE-THICKNESS PRE-INTEGRATION ˆD l mm = h 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j ( ˆJ)3 dζ å (7.39) ˆD l mb = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+1 ( ˆJ)3 dζ å (7.40) ˆD l mn = h3 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+2 ( ˆJ)3 dζ å (7.41) ˆD l bb = 4 ˆDl mn (7.42) ˆD l bn = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+3 ( ˆJ)3 dζ å (7.43) ˆD l nn = h5 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+4 ( ˆJ)3 dζ å (7.44) ˆD l m = h 4 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi ˆJ dζ å (7.45) ˆD l b = h2 2 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi+1 ˆJ dζ å (7.46) ˆD l n = h3 4 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi+2 ˆJ dζ å (7.47) J = det J = h J0 + h2 ζ J1 + h3 ζ2 J2 (7.48) ˆJ = J h = J0 + h ζ J1 + h2 ζ2 J2 (7.49) J0 = ↑ ↑ ↑ P · Lξ P · Lη N ↓ ↓ ↓ (7.50) J1 = ↑ ↑ ↑ P · Lξ N ,η N ↓ ↓ ↓ + ↑ ↑ ↑ P · Lη N ,ξ N ↓ ↓ ↓ (7.51) J2 = ↑ ↑ ↑ N ,ξ N ,η N ↓ ↓ ↓ (7.52) Wh Dl d t th tit ti t d ﬁ d i ti (6 4) f ˆD l mm = h 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j ( ˆJ)3 dζ å (7.39) ˆD l mb = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+1 ( ˆJ)3 dζ å (7.40) ˆD l mn = h3 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+2 ( ˆJ)3 dζ å (7.41) ˆD l bb = 4 ˆDl mn (7.42) ˆD l bn = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+3 ( ˆJ)3 dζ å (7.43) ˆD l nn = h5 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+4 ( ˆJ)3 dζ å (7.44) ˆD l m = h 4 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi ˆJ dζ å (7.45) ˆD l b = h2 2 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi+1 ˆJ dζ å (7.46) ˆD l n = h3 4 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi+2 ˆJ dζ å (7.47) J = det J = h J0 + h2 ζ J1 + h3 ζ2 J2 (7.48) ˆJ = J h = J0 + h ζ J1 + h2 ζ2 J2 (7.49) J0 = ↑ ↑ ↑ P · Lξ P · Lη N ↓ ↓ ↓ (7.50) J1 = ↑ ↑ ↑ P · Lξ N ,η N ↓ ↓ ↓ + ↑ ↑ ↑ P · Lη N ,ξ N ↓ ↓ ↓ (7.51) J2 = ↑ ↑ ↑ N ,ξ N ,η N ↓ ↓ ↓ (7.52) ˆD l mm = h 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j ( ˆJ)3 dζ å (7.39) ˆD l mb = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+1 ( ˆJ)3 dζ å (7.40) ˆD l mn = h3 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+2 ( ˆJ)3 dζ å (7.41) ˆD l bb = 4 ˆDl mn (7.42) ˆD l bn = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+3 ( ˆJ)3 dζ å (7.43) ˆD l nn = h5 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+4 ( ˆJ)3 dζ å (7.44) ˆD l m = h 4 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi ˆJ dζ å (7.45) ˆD l b = h2 2 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi+1 ˆJ dζ å (7.46) ˆD l n = h3 4 2 X i=0 hi QT i · Dl · I · ÇZ ζl ζl−1 ζi+2 ˆJ dζ å (7.47) J = det J = h J0 + h2 ζ J1 + h3 ζ2 J2 (7.48) ˆJ = J h = J0 + h ζ J1 + h2 ζ2 J2 (7.49) J0 = ↑ ↑ ↑ P · Lξ P · Lη N ↓ ↓ ↓ (7.50) J1 = ↑ ↑ ↑ P · Lξ N ,η N ↓ ↓ ↓ + ↑ ↑ ↑ P · Lη N ,ξ N ↓ ↓ ↓ (7.51) J2 = ↑ ↑ ↑ N ,ξ N ,η N ↓ ↓ ↓ (7.52) Wh Dl d t th tit ti t d ﬁ d i ti (6 4) f ˆD l mm = h 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j ( ˆJ)3 dζ å (7.39) ˆD l mb = h2 2 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+1 ( ˆJ)3 dζ å (7.40) ˆD l mn = h3 4 2 X i=0 2 X j=0 hi+j QT i · Dl · Qj · ÇZ ζl ζl−1 ζi+j+2 ( ˆJ)3 dζ å (7.41) ˆ l ˆ l (7.39) (7.52) Where Dl denotes the constitutive tensor deﬁned in equation (6.4) for the material layer l. . THROUGH-THE-THICKNESS PRE-INTEGRATION 7.2. be pre-computed in the reference conﬁguration. To the author’s knowledge, however, nobody has performed this through-the-thickness pre-integration taking into account all the higher order terms of the Jacobian and the de- formation tensors for a rotation-free shell element2. This approach has a signifﬁcant computational advantage over the Updated Lagrangian formu- lations, as the preintegration is performed only once, instead of performing it at every time-step or load-step. Performing a little bit of arithmetic, all of this results in the following expressions for the material and geometric stiffness matrices. KM = X ∀α={m,b,n} X ∀β={m,b,n} KMαβ (7.32) KG = X ∀α={m,b,n} X ∀β={m,b,n} KGαβ (7.33) (7.32) KMαβ = Z 1 0 Z 1−ξ 0 BT α · ˆDαβ · Bβ dξ dη (7.34) KGαβ = Z 1 0 Z 1−ξ 0 ∂BT α ∂˜xr · ( ˆDαβ · gβ −δαβ · ˆDβ) dξ dη (7.35) ˆDαβ = ♯layers X l=1 (λl)κ+χ ˆD l αβ (7.36) ˆDβ = ♯layers X l=1 (λl)χ ˆD l β (7.37) κ =      0 if α = m, 1 if α = b, 2 if α = n. χ =      0 if β = m, 1 if β = b, 2 if β = n. (7.38) (7.38) Notice that in equations (7.34) and (7.35), all the terms that do not de- pend on the thickness coordinate ζ have been taken out of the through-the- thickness integral. All the other terms, including those that are constant in the current conﬁguration have been condensed into the ˆDl αβ tensors and the ˆDl β vectors; which can be computed in the reference conﬁguration once and for all in the simulation/analysis if the material is linear elastic. In the case of non-linear behavior of the materials (e.g. plasticity, damage, etc.) these tensors need to be reevaluated in the current conﬁguration; but the integrals are not reevaluated. CHAPTER 7. TANGENT STIFFNESS MATRIX 84 Another important consideration has been introduced. Since there can be different materials in the thickness direction, the through-the-thickness integral has been split into a summation over the different layers of the laminate. The stretch factor (λ) is taken into account evaluating it once per each material layer. So the author considers that it is sufﬁcient to consider a piece-wise constant variation of the stretch factor through the thickness. . THROUGH-THE-THICKNESS PRE-INTEGRATION All the other superindices in the above equations 7.3. TANGENT LOADS MATRIX 85 mean exponents. The limits of the thickness integrals are the bottom and top coordinates of each layer l. Thus, ζ0 = −1 2, and ζ♯layers = 1 2. 2 ♯ 2 Now, the advantage of these expressions is that all the through-the- thickness integrals contain only scalar values and are completely deﬁned in the reference conﬁguration. There are in total 14 (9 + 5) scalar inte- grals to perform. The integrands are rational, because the expression of the Jacobian (J) is polynomic. However, most authors perform a Taylor series expansion on the jacobian J in the thickness direction3 —dropping the quadratic terms in the thickness direction4— and take only into ac- count the maximum exponent in the numerators to evaluate the polyno- mial order of the integrands. The maximum exponent in the numerator is 8. This is why the author suggests using a 5 points Gaussian quadrature to numerically evaluate the integrals of the shell in the thickness direction. However, because these expressions are so cheap, it is possible to select an arbitrary precision for the thickness integrals at a marginal increment of the computational cost. The author suggests to use at least 5 integration points per material layer in the laminate. We can now take advantage of all this notation and perform also a through-the-thickness pre-integration of the equivalent internal nodal loads: F int e = X ∀α={m,b,n} X ∀β={m,b,n} F int αβ (7.53) F int αβ = Z 1 0 Z 1−ξ 0 BT α · ( ˆDαβ · gβ −δαβ · ˆDβ) dξ dη (7.54) (7.53) 4Büchter and Ramm in [18, pp. 44–46] describe the different proposed approaches to perform the through-the-thickness integration in continuum based (degenerated) shell ele- ments. Bischoff et al. in [12, p. 101] further explain that using an exact shifter (A in our case) yields better results than using an aproximated shifter that facilitates performing an analytic integration through-the-thickness. 7.3 Tangent loads matrix The tangent loads matrix stems from the linearization of the equivalent external nodal loads. ∂F ext e ∂˜xr = ∂ ∂˜xr ñZZZ Ve Å ∂x ∂˜xr ãT · qV dV + ZZ Se Å ∂x ∂˜xr ãT · qS dS+ + Z Γe Å ∂x ∂˜xr ãT · qΓ dΓ + ˜qr ô = = ∂ ∂˜xr ZZZ Ω0 ∂xT ∂˜xr · q0 V dΩ0 \| {z } HV 0 + ∂ ∂˜xr ZZZ Ω ∂xT ∂˜xr · qV dΩ \| {z } HV + + ∂ ∂˜xr ZZ ∂Ω0 ∂xT ∂˜xr · q0 S dS0 \| {z } HS0 + ∂ ∂˜xr ZZ ∂Ω ∂xT ∂˜xr · qS dS \| {z } HS + ∂F ext e ∂˜xr = ∂ ∂˜xr ñZZZ Ve Å ∂x ∂˜xr ãT · qV dV + ZZ Se Å ∂x ∂˜xr ãT · qS dS+ + Z Γe Å ∂x ∂˜xr ãT · qΓ dΓ + ˜qr ô = = ∂ ∂˜xr ZZZ Ω0 ∂xT ∂˜xr · q0 V dΩ0 \| {z } HV 0 + ∂ ∂˜xr ZZZ Ω ∂xT ∂˜xr · qV dΩ \| {z } HV + + ∂ ∂˜xr ZZ ∂Ω0 ∂xT ∂˜xr · q0 S dS0 \| {z } HS0 + ∂ ∂˜xr ZZ ∂Ω ∂xT ∂˜xr · qS dS \| {z } HS + {z HS0 {z HS0 3Also a Taylor series expansion or other approximations of the shell shifter are common in the literature on shell ﬁnite elements [12, p. 94] 4 3Also a Taylor series expansion or other approximations of the shell shifter are common in the literature on shell ﬁnite elements [12, p. 94] CHAPTER 7. TANGENT STIFFNESS MATRIX 86 + ∂ ∂˜xr Z ∂2Ω0 ∂xT ∂˜xr · q0 Γ dΓ 0 \| {z } HΓ0 + ∂ ∂˜xr Z ∂2Ω ∂2xT ∂˜xr · qΓ dΓ \| {z } HΓ (7.55) {z HΓ Each of the tangent matrices above can be developed for each partic- ular case considered. Here, the author will only develop as an example the self-weight loads (equation (7.56)), which are a case of loads deﬁned over the volume in the reference conﬁguration, and then two different sur- face loads. One for dead loads (equation (7.57)), which are also deﬁned in the reference conﬁguration. And another one for follower pressure loads (equation (7.58)), which are deﬁned in the deformed conﬁguration and the direction changes with the geometry. 7.3 Tangent loads matrix SUMMARY 7.4. SUMMARY 87 7.3 Tangent loads matrix HV 0 = ∂ ∂˜xr ZZZ Ω0 ρ g eT 3 · ∂x ∂˜xr dΩ0 = Z 1 0 Z 1−ξ 0 Z 1 2 −1 2 ρ g eT 3 · ∂2x (∂˜xr)2 J dξ dη dζ = = Z 1 0 Z 1−ξ 0 g eT 3 · ∂2p (∂˜xr)2 · N 2 X i=0 " hi+1J0 i ♯layers X l=1 Ç ρl Z ζl ζl−1 ζi dζ å# dξ dη+ + Z 1 0 Z 1−ξ 0 g eT 3 · ∂2n (∂˜xr)2 2 X i=0 " hi+2J0 i ♯layers X l=1 Ç λlρl Z ζl ζl−1 ζi+1 dζ å# dξ dη (7.56) HS0 = ∂ ∂˜xr ZZ ∂Ω0 ∂¯xT ∂˜xr · q0 S dS0 = Z 1 0 Z 1−ξ 0 ∂2¯xT (∂˜xr)2 · q0 SJ0 dξ dη = = Z 1 0 Z 1−ξ 0 N T · ∂2pT (∂˜xr)2 · q0 SJ0 dξ dη (7.57) HS = ∂ ∂˜xr ZZ ∂Ω p∂¯xT ∂˜xr · n dS = = Z 1 0 Z 1−ξ 0 p Ü ∂2¯xT (∂˜xr)2 · n + ∂¯xT ∂˜xr · ∂n ∂˜xr \| {z } non-symmetric ê J0 dξ dη (7.58) HS0 = ∂ ∂˜xr ZZ ∂Ω0 ∂¯xT ∂˜xr · q0 S dS0 = Z 1 0 Z 1−ξ 0 ∂2¯xT (∂˜xr)2 · q0 SJ0 dξ dη = = Z 1 0 Z 1−ξ 0 N T · ∂2pT (∂˜xr)2 · q0 SJ0 dξ dη (7.57) HS0 = ∂ ∂˜xr ZZ ∂Ω0 ∂¯xT ∂˜xr · q0 S dS0 = Z 1 0 Z 1−ξ 0 ∂2¯xT (∂˜xr)2 · q0 SJ0 dξ dη = Z 1 Z 1−ξ N T ∂2pT 0 J dξ d (7.57) (7.58) The corresponding equivalent external nodal loads are: The corresponding equivalent external nodal loads are: F ext V 0 = Z 1 0 Z 1−ξ 0 g eT 3 · ∂p ∂˜xr · N 2 X i=0 " hi+1J0 i ♯layers X l=1 Ç ρl Z ζl ζl−1 ζi dζ å# dξ dη+ + Z 1 0 Z 1−ξ 0 g eT 3 · ∂n ∂˜xr 2 X i=0 " hi+2J0 i ♯layers X l=1 Ç λlρl Z ζl ζl−1 ζi+1 dζ å# dξ dη (7.59) F ext S0 = Z 1 0 Z 1−ξ 0 (q0 S)T · ∂p ∂˜xr · N J0 dξ dη (7.60) F ext S = Z 1 0 Z 1−ξ 0 p nT · ∂¯x ∂˜xr J0 dξ dη (7.61) F ext V 0 = Z 1 0 Z 1−ξ 0 g eT 3 · ∂p ∂˜xr · N 2 X i=0 " hi+1J0 i ♯layers X l=1 Ç ρl Z ζl ζl−1 ζi dζ å# dξ dη+ + Z 1 0 Z 1−ξ 0 g eT 3 · ∂n ∂˜xr 2 X i=0 " hi+2J0 i ♯layers X l=1 Ç λlρl Z ζl ζl−1 ζi+1 dζ å# dξ dη (7.59) F ext S0 = Z 1 0 Z 1−ξ 0 (q0 S)T · ∂p ∂˜xr · N J0 dξ dη (7.60) F ext S = Z 1 0 Z 1−ξ 0 p nT · ∂¯x ∂˜xr J0 dξ dη (7.61) 7.4. 7.4 Summary This chapter presents the explicit expressions to construct the tangent stiffness matrix of the BEST element. Taking advantage of the decompo- sition of the Green-Lagrange tensor presented in the previous chapter and the Jacobian of the element, the author introduces a through-the-thickness pre-integration of the terms of the stiffness matrix (including the higher order terms). The formulæ are valid also in the case of a multi-layered shell. To the author’s knowledge this is a novel contribution of this thesis. g The expressions for some of the possible tangent loads matrices are also presented (self-weight loads, dead loads, and follower pressure loads). The equations presented in this chapter are complemented —for the sake of completeness— with the expressions needed of all the derivatives in appendix D. 8.1.1 Slender beam Firstly, we show the analysis of a simple slender beam. In linear analysis, this conﬁguration only generates bending stresses. The dimensions and the magnitude of the load are deﬁned in such way that the deﬂection at the center is 1. L = 10, b = 1, t = 0.01, E = 2.5 · 1011, ν = 0.3, q = 160. The author does not recall big differences amongst the different quadra- tures. The order of convergence is O ∝h1/2 in all cases. This result will be further discussed in the Summary in section 8.5. 90 CHAPTER 8. ON THE INTEGRATION ORDER OF THE ELEMENT 3 Gauss points: This quadrature corresponds to a quadratic integration. This would be the right choice if the functions that describe the de- formation of the element contain linear information. As a result, the integrand would be a quadratic function. 3 Gauss points: This quadrature corresponds to a quadratic integration. This would be the right choice if the functions that describe the de- formation of the element contain linear information. As a result, the integrand would be a quadratic function. 4 Gauss points: This quadrature corresponds to a cubic integration. This is a very rare quadrature in triangular ﬁnite elements because it’s order is not even. Nevertheless, the peculiarities of the present de- velopment make it worth considering it in the study. Furthermore, in the case of resulting adequate, it is the quadrature that marginally requires the least number of additional evaluations with respect to the lower order quadratures. 4 Gauss points: This quadrature corresponds to a cubic integration. This is a very rare quadrature in triangular ﬁnite elements because it’s order is not even. Nevertheless, the peculiarities of the present de- velopment make it worth considering it in the study. Furthermore, in the case of resulting adequate, it is the quadrature that marginally requires the least number of additional evaluations with respect to the lower order quadratures. 6 Gauss points: This quadrature corresponds to a quartic integration. This is the last option to consider because it would imply that the information contained in the element’s shape functions is cubic at all effects. As a consequence, the functions describing the deformation would be quadratic and the integrands would be quartic functions. 8.1 Bending dominant cases This section includes 2 cases of structures under such loading that the main deformation energy corresponds to the bending deformation. The ﬁrst case is that of a simply supported slender beam modeled as a shell. The second is a classic veriﬁcation example for shell elements: a hemi- sphere subjected to opposite point loads. Chapter 8 On the integration order of the element S ELDOM A DEVELOPER QUESTIONS the integration order of a ﬁnite el- ement. This is because the knowledge of isoparametric elements is quite deep and the integration criteria of the stiffness matrix are clear. Nevertheless, given the speciﬁcities of the formulation in the present de- velopment it is reasonable to ponder about the integration order required for the element. Let’s consider the different options: S E • The input information is linear. Linear integration? • The input information is linear. Linear integration? • The approximation order of the improved geometry is cubic. Quartic integration? • The functions stemming from the linearization of the normalization of the normal vectors are rational. What order shall be applied? In order to determine the corresponding integration order we shall de- velop the Taylor series of the expression of the stiffness matrix and determine how many terms of the series are relevant. But the Tay- lor series expansions are only local and writing the expansion would be a very complex task. Even if we did, determining which are the relevant terms wouldn’t be an easy call. Therefore, the need arises to determine how many Gauss points are required when performing the numerical integrations over the element. Using too many points will result in an increased computational cost of the computations for the element without increasing the precision. Using too few points can lead to unpredictable results (i.e. loosing the full rank of the stiffness matrix) or simply a loss in precision that makes the development of the new element totally worthless. The methodology to determine the order of the quadrature has been to identify the different options available and evaluate them using a series of representative cases. The quadratures considered are: 1 Gauss point: This quadrature corresponds to a linear integration con- sidering that the information of the shape functions —after deriving them to describe the element’s deformation— is barely constant. 89 8.2 Membrane dominant cases Next we present 2 cases of structures subject to such loading that the main deformation energy corresponds to membrane deformation. The ﬁrst case is that of a very thin-walled cylinder under internal pressure. The second case is that of a roof in the form of a parabolic cylinder simply supported along its two generatrices and subjected to self-weight loading. 8.1.2 Hemisphere with point loads The second case consists on the already classical example of a hemispheric shell with an 18° hole at the center. The hemisphere is subjected to two pairs of diametrically opposed loads. The displacement of the points loaded is 0.093 for the following values: φ = 20, t = 0.04, E = 6.825 · 107, ν = 0.3, P = 2. It appears that using fewer Gauss points improves convergence. How- ever, in all cases the convergence order is O ∝h. Notwithstanding this fact, the author recalls the apparent better precision obtained with a quadra- ture of a single Gauss point. In fact, the precision achieved with a single Gauss point is an order of magnitude better than that obtained with any other quadrature considered. 8.1. BENDING DOMINANT CASES 91 Figure 8.1: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a simply supported beam. Figure 8.1: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a simply supported beam. Figure 8.2: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a pinched hemisphere. Figure 8.3: Comparison of h-convergence using unstructured meshes and different numerical quadratures for a pinched hemisphere. Figure 8.2: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a pinched hemisphere. Figure 8.2: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a pinched hemisphere. Figure 8.3: Comparison of h-convergence using unstructured meshes and different numerical quadratures for a pinched hemisphere. Figure 8.3: Comparison of h-convergence using unstructured meshes and different numerical quadratures for a pinched hemisphere. 92 CHAPTER 8. ON THE INTEGRATION ORDER OF THE ELEMENT 8.2.1 Cylinder subject to internal pressure The motivation for choosing this case stems from the desire to observe the virtues of the BEST element. That is, a case where the description using the cubic Bézier functions for the element’s geometry plays an im- portant role in the out-of-plane deformation of the element. For this rea- son it is important to choose an example with curved geometry yet sub- ject solely to membrane stresses and not to bending stresses. The cylin- der is a very well suited geometry to study using structured meshes. In order to avoid activating the bending energy that would experiment the cylinder, its thickness is restricted to an extremely small value. Simi- larly to the case of the simply supported beam, the parameters of the ex- ample have been selected in such way that the radial enlargement is 1. L = 20, φ = 10, t = 0.0005, E = 108, ν = 0.3, p = 1000. Figure 8.4: Comparison of h-convergence using non-symmetric structured meshes and different numerical quadratures for a thin-walled cylinder. Figure 8.4: Comparison of h-convergence using non-symmetric structured meshes and different numerical quadratures for a thin-walled cylinder. The observation of the results of this case shows it takes full advantage of the cubic formulation of the BEST element. Using the quartic quadra- ture of 6 points, corresponding to the theoretical quadrature for an element with cubic shape functions, yields convergence of order O ∝h3. Using a cubic quadrature of 4 points yields convergence of order O ∝h2. And using the quadratures quadratic and linear, of 3 and 1 point respectively, yields convergence of order O ∝h. This is a very important result, as it demonstrates it is possible to achieve cubic convergence out of linear information. The mechanism to accomplish it, unlike other methods based on the construction of macro- 2. MEMBRANE DOMINANT CASES 93 8.2. elements, uses the information of the neighboring elements in order to build an element of greater order. CHAPTER 8. ON THE INTEGRATION ORDER OF THE ELEMENT 94 order O ∝h5/2, while the other quadratures show convergence rates of order O ∝h3/2. In the case of unstructured meshes the convergence prop- erties become reversed and now the quadrature with 1 Gauss point has convergence of order O ∝h2 while all the other cases improve and now ex- hibit convergences of order O ∝h5/2. Nevertheless, the quadrature with 1 Gauss point seems to have more precision. The quadrature using 4 Gauss points also obtains notable results with respect to the other quadratures that do not use the triangle’s barycenter as an evaluation point in the quadrature. Therefore, it seems as if the barycenter of the triangle has some special property when calculating the bending of the element. order O ∝h5/2, while the other quadratures show convergence rates of order O ∝h3/2. In the case of unstructured meshes the convergence prop- erties become reversed and now the quadrature with 1 Gauss point has convergence of order O ∝h2 while all the other cases improve and now ex- hibit convergences of order O ∝h5/2. Nevertheless, the quadrature with 1 Gauss point seems to have more precision. The quadrature using 4 Gauss points also obtains notable results with respect to the other quadratures that do not use the triangle’s barycenter as an evaluation point in the quadrature. Therefore, it seems as if the barycenter of the triangle has some special property when calculating the bending of the element. 8.3 In-plane shear dominant cases Finally, this section presents 2 cases of structures whose loading makes in-plane shear the main deformation mode. The ﬁrst case is that of a thick beam modeled with shell elements. The second case is that of a cylinder under uniform torsion. 8.2.2 Parabolic roof This case differs slightly from the idealism of the previous case. Now, the roof’s self-weight makes the structure experience tractions mainly. How- ever, the real funicular shape of the load shall be a catenary instead of a parabola. Therefore some bending stresses appear, although the main de- formation mode is membrane. L = 50, a = 20, c = 10, t = 0.05, E = 108, ν = 0.0, q = 1000. Figure 8.5: Comparison of h-convergence using non-symmetric structured meshes and different numerical quadratures for a parabolic roof. Figure 8.5: Comparison of h-convergence using non-symmetric structured meshes and different numerical quadratures for a parabolic roof. Figure 8.6: Comparison of h-convergence using unstructured meshes and different numerical quadratures for a parabolic roof. Figure 8.6: Comparison of h-convergence using unstructured meshes and different numerical quadratures for a parabolic roof. This example shows that when introducing some bending, the excellent convergence properties veriﬁed in the previous example are lost, and the quadrature with 1 Gauss point is again the best one. When using struc- tured meshes, the quadrature with 1 Gauss point exhibits convergence of 8.3.1 Thick beam This is the most trivial case to activate the element’s in-plane shear. Con- sists on a cantilevered deep beam subject to a uniform force along its free end. The reference solution is 0.35533. Figure 8.7: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a thick beam. Figure 8.7: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a thick beam. In this example the differences between the numerical quadratures aren’t signiﬁcant. Indeed, the results are numerically identic. That’s why the different curves of ﬁgure 8.7 are overlapped. The convergence is in all cases of order O ∝h. This result is not surprising because the formula- tion of the BEST element has been designed to take advantage of the cubic Bézier description in the curvature of the element. In a case like this the element behaves in its plane and the cubic description of the Bézier func- tions doesn’t play any role, thus behaving like a vulgar linear element. 8.4. EFFECT OF THE MESH 95 8.3.2 Cylinder under torsion This case consists on a cylinder with one base clamped and 2 pairs of tan- gential forces at the free edge generating a torque. Unlike the case of the cylinder under internal pressure, in this case we set a very thick wall for the cylinder with the objective of activating solely shear deformations and avoid bending the elements. Figure 8.8: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a cylinder under torsion. Figure 8.8: Comparison of h-convergence using structured meshes and dif- ferent numerical quadratures for a cylinder under torsion. The convergence obtained in this case is very low. All the quadratures achieve a convergence rate of order only O ∝h1/2. It seems again that the quadrature with a single Gauss point has some advantage albeit only marginal with respect to the other quadratures. This result will be further discussed in the Summary in section 8.5. 8.4.1 Structured mesh with symmetrical triangles Generating structured triangle meshes usually becomes a problem in curved surfaces. For example in the case of cylindrical surfaces. Structured tri- angle meshes are generated based on the corresponding structured quad- rangles mesh. For the quadrangles, meshing a cylindrical surface using a structured mesh is a well deﬁned problem and the solution is very sat- isfactory. But when switching to triangles it doesn’t hold. A cylindrical surface has 2 main directions that deﬁne it and the quadrangles adapt very well to this space covering it in a Cartesian manner. However the triangles don’t ﬁt well in that paradigm, and they should follow a different strategy. Trying to ﬁt triangles using the mesh deﬁned by the quadrangles often results in ugly meshes. In particular, subdividing the quadrangles into symmetric triangles (4 triangles per quadrangle) generates a mesh which does not keep some of the basic properties of the original surface. For example, convexity. Figure 8.9: Example of a cylinder meshed using a structured mesh of sym- metric triangles. Figure 8.9: Example of a cylinder meshed using a structured mesh of sym- metric triangles. Figure 8.9 shows an example of a structured mesh using symmetric triangles. Few engineers would choose such a mesh for their computations. Both because of the mesh coarseness and because of its quality. The reason for showing this mesh is to illustrate the kind of geometric defect I am arguing. In ﬁgure 8.10 it’s easy to see the effect caused by a lower order numeri- cal quadrature when performing the integrals on the results obtained. The main feature to observe is a lack of continuity in the results, exacerbated when using a single Gauss point. Continuity is recovered for the 4 and 6 Gauss points quadratures; cubic and quartic, respectively. We shall recall that a key aspect in the design of the BEST element is precisely the con- struction of a geometry with greater inter-elemental continuity. Loosing that characteristic because of the numerical quadrature is unacceptable. 8.4 Effect of the mesh One of the main goals for designing this new rotation-free thin shell ele- ment was to avoid the mesh dependence that the elements developed with the BST technology suffered. It is known that those elements exhibit a clear dependence of the precision of the element on the mesh topology. Thus, in cases of markedly anisotropic meshes, the operator to calculate the curvature of the BST element rapidly looses precision. This is an effect we try to avoid at all costs in the design of the new rotation-free element. Along the present study on the order of integration we have been able to detect the beneﬁts of a higher order of integration in the inﬂuence of the mesh on the results. The author has observed clearly how a reduced order of integration exposes the element to the negative effects of a mesh inad- equate to the simulation. On the contrary, when using a quadrature with more Gauss points we can completely cancel out the effect of a mesh that could undermine the result. Next the case of the cylinder under internal pressure is presented. 96 CHAPTER 8. ON THE INTEGRATION ORDER OF THE ELEMENT 8.4.2 Structured mesh with non-symmetrical triangles Non-symmetrical meshes adapt better to the convexity of the surface. For that, every quadrangle is divided into two coplanar triangles. This setup, 8.4. EFFECT OF THE MESH 97 (a) 1 Gauss point. (b) 3 Gauss points. (c) 4 Gauss points. (d) 6 Gauss points. Figure 8.10: Qualitative comparison of the results obtained using struc- tured meshes of symmetric triangles and different numerical quadratures for a thin-walled cylinder. Displacements in the x-direction as viewed on the y-z plane. (b) 3 Gauss points. (d) 6 Gauss points. (a) 1 Gauss point. (c) 4 Gauss points. 1 Gauss point. (c) 4 Gauss points. Figure 8.10: Qualitative comparison of the results obtained using struc- tured meshes of symmetric triangles and different numerical quadratures for a thin-walled cylinder. Displacements in the x-direction as viewed on the y-z plane. besides avoiding to increase the number of nodes, takes advantage of the goodness of structured quadrangle meshes. However, it has the disadvan- tage that unless the mesh is generated with careful attention to details, the result can be a strongly biased mesh. Usually the mesh generator will set all the diagonals that divide the quadrangles in the same orientation. This is what causes the mesh to exhibit a markedly anisotropic behavior that affects the results of the structural analysis. As if the mesh was rein- forced in the direction of the diagonals. Figure 8.11 shows an example of a non-symmetrical structured trian- gles mesh. Unlike the example shown in ﬁgure 8.9 this case maintains the convexity of the geometry, but instead exhibits the diagonals biased as they are all oriented in the same direction. Once again, ﬁgure 8.12 shows clearly the effect produced in the results when using a numerical quadrature of lower order. The mesh effect can be seen very clearly in the results for the cases of low order quadratures. In theory the result should exhibit radial symmetry. Therefore, the iso- lines of the results should be aligned along the generatrices of the cylinder. The correct orientation of the isolines is recovered for the cases that use quadratures of 4 and 6 Gauss points; cubic and quartic, respectively. With these two cases we have clearly shown the need to use an ade- quate integration order to avoid the effect of the mesh on the calculations. 8.5 Summary A number of cases representing different deformation modes have been tested. Each case has been solved in the linear regime using four different quadratures corresponding to different orders of integration: linear, qua- dratic, cubic and quartic. For the BEST element, full integration requires a quartic order of integration. All the other quadratures tested are there- fore instances of reduced quadratures. The deformation modes included in the analysis are the bending mode, the membrane mode and the in-plane shear mode. In most cases, the results obtained with only one quadrature point are the most accurate. This seems to indicate that reduced integration repre- sents an advantage for the element. The reason for this behavior could be that the BEST element is affected by some sort of locking behavior. Choi et al. [31] assert that “membrane locking only occurs in non-inhibited thin shells” (p. 131), and that it consists in an inadequacy of the ﬁnite elements to describe pure bending deformations (i.e. bending of the mid- surface without extensional deformations). The non-inhibited cases tested are the two corresponding to the bending deformation mode: the simply supported beam and the pinched sphere. Actually, the sphere is a non- developable surface and therefore, there is some inhibition caused by the geometry of the shell (not the boundary conditions). It can be observed that the simply supported beam suffers a more severe degradation of the convergence O ∝h1/2. In both cases, however, the reduced integration in- creases the precision, but it does not affect the order of convergence. The same occurs in the case of the cylinder under torsion. This is consistent with the results reported in [31] where the authors explain that reduced integration does not eliminate membrane locking. The possible cause for the reduced order of convergence of the BEST el- ement in most cases might be related to another relevant result obtained in section 8.3.1. The thick beam example shows that the BEST element does not show any difference when different quadratures are used in the in-plane shear deformation mode. This result—combined with the result in section 8.2.1 where each quadrature provides the corresponding theo- retical order of convergence in the solution—gives an answer to the ques- tions posed at the beginning of the chapter. That is: it is indeed possible to obtain cubic convergence from linear information provided that the ade- quate geometric construction is built. 8.4.2 Structured mesh with non-symmetrical triangles At the same time, these results demonstrate qualitatively the good proper- ties of the BEST element to obtain results independent of the mesh topol- ogy. 98 CHAPTER 8. ON THE INTEGRATION ORDER OF THE ELEMENT Figure 8.11: Example of a cylinder meshed with a non-symmetrical struc- tured triangles mesh. All the diagonals are oriented in the same direction. This is the mesh used to obtain the results shown in ﬁgure 8.12. Figure 8.11: Example of a cylinder meshed with a non-symmetrical struc- tured triangles mesh. All the diagonals are oriented in the same direction. This is the mesh used to obtain the results shown in ﬁgure 8.12. (a) 1 Gauss point. (b) 3 Gauss points. (c) 4 Gauss points. (d) 6 Gauss points. Figure 8.12: Comparison of the results obtained using different numerical quadratures and non-symmetrical structured triangle meshes for a thin- walled cylinder. Displacements in the x-direction as viewed on the y-z plane. The mesh used is also displayed in ﬁgure 8.11 as a reference to indicate the effect of the skewed triangles. (a) 1 Gauss point. (c) 4 Gauss points. (b) 3 Gauss points. (d) 6 Gauss points. (a) 1 Gauss point. (b) 3 Gauss points. (b) 3 Gauss points. (d) 6 Gauss points. ( ) p (c) 4 Gauss points. (d) 6 Gauss points. c) 4 Gauss points. Figure 8.12: Comparison of the results obtained using different numerical quadratures and non-symmetrical structured triangle meshes for a thin- walled cylinder. Displacements in the x-direction as viewed on the y-z plane. The mesh used is also displayed in ﬁgure 8.11 as a reference to indicate the effect of the skewed triangles. 8.5. SUMMARY 8.5. SUMMARY 99 8.5 Summary The information extracted from the neighboring nodes is paramount. But, since the author has not tackled the speciﬁc in-plane kinematics of the Bézier-enhanced triangle, the informa- tion from the neighboring nodes is not being used adequately. This issue will be studied in detail in chapter 9. To dispel any doubts on the possible beneﬁts for using reduced inte- gration, the author presents in the last section of the chapter a study on the effects of the mesh when reduced integration is used. As expected, re- duced integration causes hourglass modes to appear, or mesh dependence issues to arise. Furthermore, when using full integration, the study shows that the BEST element exhibits an excellent behavior with respect to mesh topology dependence. Therefore, it is highly unadvised to use reduced in- tegration in the BEST element. Chapter 9 Membrane locking of thin shells: a study on how this affects the BEST element and how to solve it L O L OOKING AT THE RESULTS OF THE CASES TESTED in the previous chap- ter, the reader could come to the conclusion that the BEST element suffers from membrane locking. The author agrees with this conclusion, but this is only half the picture. Let’s recall the geometric construction of the Bézier-enhanced triangle presented in chapter 5. Particularly item 2 on page 51, where the contour of the triangle is deﬁned by plane curved edges. This decision limits the space of shapes that can be constructed for the Bézier triangle. In particular, it limits the in-plane kinematics of the element. This chapter will review that decision and its consequences. It will also suggest ways to alleviate the poor convergence for those cases where in-plane shear is at play. CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT Following these results stems that the BEST element should not suf- fer membrane locking if the full potential of the cubic shape functions is deployed. Precisely, the in-plane cubic kinematic description of the BEST element is limited to linear because of the constraint imposed that the tri- angle edges shall remain plane curves. This constraint downgrades the cubic description of the cubic Bézier triangle to a linear description of the in-plane kinematics; which determines the membrane deformations of the shell. Therefore, it is imperative to recover the full cubic description of the element boundaries in order to avoid the locking behavior experimented by the BEST element. The author will suggest some strategies to fulﬁll this objective. These strategies shall take into account both the kinematics of the shell triangle but also take into account the energy involved in those kinematics in a similar way to the one described in section 5.4. In order to enrich the in-plane kinematics of the BEST element, it is necessary to allow relative in-plane displacements of the triangle edges. The ﬁrst idea to describe this kind of kinematics is to emulate the in-plane rotations of the vertices of the triangle. The in-plane rotations of a shell element are commonly referred to as drilling rotations. Felippa [39]—for triangles—, and Wisniewski and Turska [140]—for quadrangles— have reported that including the enriched kinematics provided by the drilling rotations improves signiﬁcantly the precision of the shell elements. 9.2 Improving the BEST element kinematics by emulation of drilling rotations 9.2.1 Abanico analogy Figure 9.1: An abanico made in the XIX century. Painted ivory with a Cupid on the frame and with delicate chantilly lace. From the collection donated by Gloria Trueba Gómez in 1997 to the city of Seville and exhib- ited at the Abanicos’ Room of the Reales Alcázares of Seville (Spain). 9.1 On membrane locking of thin shells Reviewing some relevant works in the literature about the topic of mem- brane locking in shell elements the author draws two main conclusions. On the one side, Choi et al. [31] afﬁrm that membrane locking only occurs in non-inhibited shells when the element is incapable of deforming under bending without also experimenting membrane deformations of the mid- surface (pure bending deformation). On the other side, both Hakula, Leino and Pitkäranta [51] and Choi et al. [31] conclude that using higher order polynomials reduces the effect of membrane locking in the ﬁnite element framework; in particular cubic and higher. The author has veriﬁed that the convergence issues of the BEST ele- ment are related to membrane locking. By performing selective reduced integration on a non-inhibited shell structure, the results improve much more when reduced integration is applied to the membrane deformation mode than when reduced integration is applied to the bending deformation mode. This technique is suggested by Oñate in [93, pp. 550–551, 591–593]. 101 102 9.2.1 Abanico analogy Figure 9.1: An abanico made in the XIX century. Painted ivory with a Cupid on the frame and with delicate chantilly lace. From the collection donated by Gloria Trueba Gómez in 1997 to the city of Seville and exhib- ited at the Abanicos’ Room of the Reales Alcázares of Seville (Spain). 9.2. EMULATION OF DRILLING ROTATIONS 103 9.2. EMULATION OF DRILLING ROTATIONS An abanico is a type of folding hand-fan that was invented in China and introduced in Europe during the XVII century. The author uses this delicate object (see ﬁgure 9.1) to illustrate the construction used to emulate the drilling degrees of freedom at the nodes. Since the construction of the BEST element avoids the use of rotation degrees of freedom, the drilling rotations need to be emulated. The author proposes the following analogy: given a node in a mesh, the node shall represent the pivot of the folding hand-fan (abanico), and each of the edges of the mesh converging at that node shall represent the slats of the abanico. The abanico analogy uses the assumption that—as the abanico opens and folds—the pivot of the abanico experiments the same rotation as the average of the rotations of the individual slats of the abanico. Therefore, it is possible to compute the change in relative orientation of the pivot with respect to each of the slats in the deformed and reference conﬁgurations. Figure 9.2: Scheme depicting the abanico analogy. For any mesh node i, and for any neighboring node j, all the βjk i angles are measured for every other neighboring k node. The gray area represents the plane perpendicu- lar to ni at i. And eij and eik are the projections of the i–j and i–k edges onto that plane, respectively. Figure 9.2: Scheme depicting the abanico analogy. For any mesh node i, and for any neighboring node j, all the βjk i angles are measured for every other neighboring k node. The gray area represents the plane perpendicu- lar to ni at i. And eij and eik are the projections of the i–j and i–k edges onto that plane, respectively. For every node i, and for every pair of neighboring nodes j and k, the au- thor deﬁnes the angle βjk i (see ﬁgure 9.2) which measures the angle formed CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 104 between the i–j and i–k edges, projected on the plane perpendicular to ni. 9.2. EMULATION OF DRILLING ROTATIONS βjk i =                    π 2 if eij · eik = 0 ∧(eij × eik) · ni > 0, 3π 2 if eij · eik = 0 ∧(eij × eik) · ni < 0, arctan \|eijeikni\| eij·eik if eij · eik > 0 ∧(eij × eik) · ni > 0, 2π + arctan \|eijeikni\| eij·eik if eij · eik > 0 ∧(eij × eik) · ni < 0, π + arctan \|eijeikni\| eij·eik if eij · eik < 0. (9.1) Deﬁned in this way, the angle βjk i is a continuous function with continuous derivatives. The angle βjk i takes values in the interval (0, 2π) and is equal to 0 when k = j. The vectors eij and eik are deﬁned in ﬁgure 9.2 and are computed according to equations (9.2) and (9.3). Deﬁned in this way, the angle βjk i is a continuous function with continuous derivatives. The angle βjk i takes values in the interval (0, 2π) and is equal to 0 when k = j. The vectors eij and eik are deﬁned in ﬁgure 9.2 and are computed according to equations (9.2) and (9.3). eij = (I −ni ⊗niT ) · (xj −xi) (9.2) eik = (I −ni ⊗niT ) · (xk −xi) (9.3) (9.2) (9.3) (9.2) (9.3) The abanico analogy above, can be written mathematically as: αij = 1 zi zi X k=1 k̸=j βjk i (9.4) (9.4) where αij is the average of the relative angles of the zi edges (slats) sur- rounding node i with respect to the edge i–j. This angle is measured coun- terclockwise around the normal vector ni. And by computing the difference between this angle in the reference and deformed conﬁgurations, the rotation of the node i with respect to the edge i–j is found: ∆αij = αij −αij 0 (9.5) (9.5) The angle ∆αij provides a measure of the drilling rotation of the node. The angle ∆αij provides a measure of the drilling rotation of the node. 9.2. EMULATION OF DRILLING ROTATIONS 105 2. The contour is not anymore a plane curve and we rotate the contour plane deﬁned in item 2 on page 51 in order to impose the drilling rota- tion on the edge ﬁber as it approaches the node. In order to maintain C0 continuity, this has to be a symmetric condition for two adjacent triangles. The selection of the plane is such that one of the directors is the edge of the ﬂat triangle rotated an angle θij and the other di- rector is the normal at the node ni. dij is the director vector of this plane. 3. And a plane perpendicular to the edge of the ﬂat triangle. the exact location of this plane will be explained in section 9.3. Sufﬁce it to say, that the criterion to position the plane is again based on energy minimization and not on geometric considerations. Figure 9.3: Representation of the net of control points (dashed lines) for a cubic Bézier triangle constructed using the nodal positions and normals. This ﬁgure differs from ﬁgure 5.5 in that the drilling rotations θij are used here. Three planes deﬁne the position of a control point A of the contour. The plane that was represented shaded in ﬁgure 5.5 is not drawn here for clarity, but remains unchanged. The two dark gray planes are parallel to each other and perpendicular to the i–j straight edge. They are placed at a distance Ψij which will be determined in section 9.3. The light gray plane has been split and rotated an angle θij about ni to account for the drilling rotation experimented by the corner node. The angles ϕij and θij are not drawn exactly as deﬁned in ﬁgures 5.7 and 9.5, respectively. Figure 9.3: Representation of the net of control points (dashed lines) for a cubic Bézier triangle constructed using the nodal positions and normals. This ﬁgure differs from ﬁgure 5.5 in that the drilling rotations θij are used here. Three planes deﬁne the position of a control point A of the contour. The plane that was represented shaded in ﬁgure 5.5 is not drawn here for clarity, but remains unchanged. The two dark gray planes are parallel to each other and perpendicular to the i–j straight edge. They are placed at a distance Ψij which will be determined in section 9.3. 9.2.2 Modiﬁed kinematics using the abanico analogy Using the drilling rotation deﬁned with the help of the abanico analogy the author proposes the following modiﬁcation of the kinematics of the BEST element. In the construction below, the following assumption is made: θij = ∆αij (9.6) (9.6) The construction used to determine the locations of the contour control points is changed as follows. Still three planes are intersected to ﬁnd the location of the control point. The planes are the ones displayed in ﬁg- ure 9.3: 1. The plane perpendicular to the normal at the vertex —this is a nec- essary and sufﬁcient condition to interpolate the normals—. This is the same plane already used in item 1 on page 51. 1. The plane perpendicular to the normal at the vertex —this is a nec- essary and sufﬁcient condition to interpolate the normals—. This is the same plane already used in item 1 on page 51. 9.2. EMULATION OF DRILLING ROTATIONS 9.2.3 This is not an incompatible mode method Some scholars may see similarities relating this enhancement to the method of incompatible modes [139]. However, this strategy is quite different and totally unrelated [55, 116, 117]. In the case of the incompatible modes method, the authors in the references above try to solve the locking prob- lems stemming from the low order description of the element kinematics by adding speciﬁc internal variables to the kinematic description of the element. In the present case, the trick is to unleash the potential of the cubic formulation of the element. So, it is not necessary to add any man- ufactured kinematic description to the element. The cubic nature of the element already has the inherent capabilities to represent the different deformation modes without causing locking mechanisms. The locking oc- curs due to the restrictive rules imposed for the geometric construction. These restrictions were so severe because there were no real means (un- til the emulation of the drilling degrees of freedom) to build a meaningful and consistent construction of the element kinematics without the drilling degrees of freedom. 9.2. EMULATION OF DRILLING ROTATIONS The light gray plane has been split and rotated an angle θij about ni to account for the drilling rotation experimented by the corner node. The angles ϕij and θij are not drawn exactly as deﬁned in ﬁgures 5.7 and 9.5, respectively. The mathematical formulas to compute the intersection of these three planes have been already provided in equations (5.15) and (5.16) on page 51. CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 106 However, the drilling rotation θij requires to redeﬁne dij. dij =ni × R(ni, θij) · (xi −xj) = =ni × (xi −xj) · cos θij + ni × [ni × (xi −xj)] · sin θij = =ni × (xi −xj) · cos θij −(I −ni ⊗niT ) · (xi −xj) · sin θij (9.7) (9.7) In the above expression, R(ni, θij) represents the rotation matrix deﬁned by a vector ni and an angle θij. And its expression is obtained using Ro- drigues’ formula. The variations in the determination of the contour control points also imply a subtle change on how the central control point shall be obtained. It is still obtained as an average of three candidate points. Since no longer a single plane along the ﬂat edge is used to compute the contour control points, it cannot be used either as an input for computing the candidate central control points (see ﬁgure 5.6). In the computation of the candidate central control points, the author proposes to slightly change the plane or- thogonal to the plane deﬁned in item 2 on page 51 by another plane which is orthogonal to another plane that also contains nij and the two interme- diate contour control points (see ﬁgure 9.4). Notice how the expressions of A0i p and b0i p in equations (5.18) and (5.19) are still correct. 9.3 Energy minimization for the in-plane shear deformation mode The introduction of emulated drilling rotations and the corresponding en- hancement of the in-plane shear deformation kinematics for the BEST el- ement means that we need to review the deﬁnition of Ψij. Let’s recall that Ψij was introduced in section 5.4 to determine the precise shape of the Bézier triangle which minimizes the elastic energy of the shell. However, 9.3. IN-PLANE SHEAR ENERGY MINIMIZATION 107 Figure 9.4: Two adjacent Bézier triangles are G1 continuous if all their ad- jacent control triangles are co-planar. This condition could be used to en- force C1 continuity and determine the position of the central control point. The light gray plane contains the average of the two corner normals nij and passes through the two mid edge control points. The dark gray trian- gles are orthogonal to the light gray plane. The candidate central control point corresponding to the edge marked with a thick gray line is repre- sented with a circle and is determined intersecting the dark gray plane with the line that passes through the barycenter of the 6 contour control points and is perpendicular to the ﬂat triangle. This construction is simi- lar to the one proposed in ﬁgure 5.6, but the present one is more general and can be used with non-ﬂat curved edges. Figure 9.4: Two adjacent Bézier triangles are G1 continuous if all their ad- jacent control triangles are co-planar. This condition could be used to en- force C1 continuity and determine the position of the central control point. The light gray plane contains the average of the two corner normals nij and passes through the two mid edge control points. The dark gray trian- gles are orthogonal to the light gray plane. The candidate central control point corresponding to the edge marked with a thick gray line is repre- sented with a circle and is determined intersecting the dark gray plane with the line that passes through the barycenter of the 6 contour control points and is perpendicular to the ﬂat triangle. This construction is simi- lar to the one proposed in ﬁgure 5.6, but the present one is more general and can be used with non-ﬂat curved edges. this energy minimization did not take into account the kinematics asso- ciated with the in-plane shear that becomes activated with the use of the drilling rotation. 9.3.1 Reduction of the problem Likewise to the procedure followed in section 5.4.1, the author proposes to analyze a 2D case representative of the problem at hand. In this case, the 2D reduction consists on neglecting the effect that curvature has on the in-plane shear deformation. Therefore, a 2D analysis can be performed considering the full triangle. This simpliﬁes the process and analytic solu- tions can be obtained. In [39] Felippa explains the process for determining optimal free pa- rameters of the ANDES template model for isotropic and non-isotropic materials. That process is analogous to the present energy minimization study. Felippa argues that the modes of deformation used in the study were initially the 3 in-plane bending modes, but that these are not lin- early independent and a fourth torsion mode was needed to avoid rank deﬁciency in the stiffness matrix. The author here proposes to study ex- clusively the torsion mode. The reason behind this decision is simplicity. Considering all the deformation modes implies ﬁnding the function for all the six different Ψαβ values depending on the six different θαβ angles in the triangle (∀α, β = 1 ÷ 3, α ̸= β). This is a set of 6 coupled problems in 6 dimensions. Instead, studying exclusively the torsion mode allows the author to consider a periodic case and to reduce the set of problems to a single problem in one dimension. This approach implies assuming that the value of each Ψij depends only on θij and is uncoupled from the other θαβ values in the triangle. 08 CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 108 9.3 Energy minimization for the in-plane shear deformation mode Ψij is the only parameter that does not depend strictly on the geomet- ric considerations of the triangle. Instead, Ψij depends on two other geo- metric parameters—the in-plane shear deformation and the out of plane membrane and bending deformation—but with the objective to minimize the overall elastic energy of the element. In section 5.4 the author obtains an expression of Ψij which depends on the value of ϕij (see Ψij ϕ in equa- tion (5.21)). Presumably, the value of Ψij will depend differently on θij than it does on ϕij. The author presents next how to obtain this dependency and then how to combine the two kinematic modes and a unique expression of Ψij depending on the variables ϕij and θij. Torsion energy In order to ﬁnd an expression of Ψij θ which minimizes the in-plane shear deformation energy, the author has modeled the equations represented by the drawing in ﬁgure 9.5 into the Maple® symbolic manipulation software. The analytic solution obtained by Maple® appears very complex at ﬁrst, but when plotted: it is not (see ﬁgure 9.6. The problem reduction stated in section 9.3.1 and depicted in ﬁgure 9.5 does not follow the guidelines of the parcel test. That is, the deformation imposed does not correspond to a constant deformation. Instead, the deformation corresponds to a periodic strain state throughout the triangle’s surface. The author does not know how to impose a constant in plane shear strain in a triangle using only drilling rotations at the corner nodes. So the author takes this torsion mode as the best possible to work with. In fact, in order to verify that only the in-plane shear deformation mode is activated, the author compares the results of the optimized expressions of Ψij θ for three different values of the Poisson coefﬁcient (ν = 0, 0.25, 0.5). The results are plotted in ﬁgure 9.6 and they are almost undistinguishable. This veriﬁes that only in-plane shear is at play and the other in-plane axial strains are residual in this problem set. Taking a close look at ﬁgure 9.6, the author notices that the solution exhibits a smooth kink of the plot for values of θ ≈± π 10. This kink is what makes the analytic expression very complex, and clearly distinguishes two different regimes in the solution. A highly non-linear regime for small val- ues of θ and an almost linear regime for higher values of θ. For extremely 109 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS Figure 9.5: Problem reduction to minimize the in-plane shear deforma- tion energy. The author analyzes a ﬂat equilateral triangle subjected to a periodic torsion θ in all three corner nodes. The dash lines represent the reference conﬁguration, while the thick curved lines represent the de- formed conﬁguration. The thin lines represent the control polygon of the deformed conﬁguration. Figure 9.5: Problem reduction to minimize the in-plane shear deforma- tion energy. The author analyzes a ﬂat equilateral triangle subjected to a periodic torsion θ in all three corner nodes. The dash lines represent the reference conﬁguration, while the thick curved lines represent the de- formed conﬁguration. Torsion energy The thin lines represent the control polygon of the deformed conﬁguration. high values of θ the author can describe a third regime, but it is so ex- treme, that it has no real engineering value. It would be possible to adjust the whole curve quite nicely for the full range of values of θ, but doing so with a simple expression means that the kink would be sharp instead of smooth, and continuity of the derivatives would be lost at that point; which is highly undesirable. Therefore, the author decides to settle for an adjustment that captures very well the solution for small values of θ. The reasoning is that in very few cases the deformation will imply values of θ larger than π 10. This adjustment is presented in mathematical form in equation (9.9) and plotted in ﬁgure 9.6. Ψij θ = 1 3 cos6 θ (9.8) (9.8) Using the assumption that this result can be uncoupled for each of the edges of the triangle, we can write the more general expression: Ψij θ (θij) = 1 3 cos6 θij (9.9) (9.9) case of a rectangle triangle. case of a rectangle triangle. As a conclusion to all these thoughts, the author deﬁnes the following function that combines the expressions of equations (5.21) and (9.9) into one: Ψij = Ψij(ϕij, θij) = ϕij2 · Ψij ϕ + θij2 · Ψij θ ϕij2 + θij2 (9.10) (9.10) Notice that the above expression has an indetermination when ϕij2 + θij2 →0 (9.11) (9.11) 9.4 Combining the Ψϕ and Ψθ expressions Both equations (5.21) and (9.9) are used to deﬁne the same magnitude in the construction of the control polygon that determines the shape of the 110 CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT Figure 9.6: Comparison of the minimization solutions for the Ψij θ values for three different values of the Poisson coefﬁcient ν. The red curve represents an adjustment for small values of θ ∈[−π 10, π 10]. Figure 9.6: Comparison of the minimization solutions for the Ψij θ values for three different values of the Poisson coefﬁcient ν. The red curve represents an adjustment for small values of θ ∈[−π 10, π 10]. cubic Bézier triangle. Since these two expressions are not coincident, the author needs to ﬁnd a way to merge them together into a single expression Ψij(ϕij, θij). The good news is that for ϕij = 0 and θij = 0 the two functions have the same value and even the same derivative. Taking a look at ﬁg- ure 9.7 the reader can observe that the two functions diverge signiﬁcantly as the variables increase their value. Therefore the merging proposal shall take into account how to take this fact into account. The most simple proposal is to use a weighted average of the two seed expressions. The weighting factor shall depend on the main variable for each expression. It makes sense that the function whose variable exhibits a bigger value also has more inﬂuence on the overall function that deﬁnes the merged expression. But how much inﬂuence? Well, this is an easy one to answer. Since the Ψ functions are deﬁned with the aim of minimiz- ing the deformation energy of the triangle, the resulting merged function should maintain this objective intact. In this sense, in general, the author assumes that the deformation energy varies quadratically with the kine- matic variables. So the square of the variables should be a good weighting factor. For a very quick veriﬁcation the author has compared on one hand how the total deformation energy of the torsion mode deﬁned in section 9.3 and using the expression in equation (9.9) varies with respect the variable θ, and on the other hand a simple quadratic expression of θ, and I can say it matches pretty well. 9.4 Combining the Ψϕ and Ψθ expressions The result in ﬁgure 9.8 also serves to verify that the solution obtained for an equilateral triangle is not too far off in the 111 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS Figure 9.7: Comparison of the minimization functions Ψij ϕ (dashed) and Ψij θ (continuous) in the range of the variables ϕij, θij ∈[0, π 10]. Figure 9.7: Comparison of the minimization functions Ψij ϕ (dashed) and Ψij θ (continuous) in the range of the variables ϕij, θij ∈[0, π 10]. Figure 9.7: Comparison of the minimization functions Ψij ϕ (dashed) and Ψij θ (continuous) in the range of the variables ϕij, θij ∈[0, π 10]. case of a rectangle triangle. 9.4.1 Solution of the indeterminate limit This indetermination can be worked out using the following change of vari- able: ϕij = ρ cos ω (9.12) θij = ρ sin ω (9.13) 112 CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT Figure 9.8: Comparison of the variation of the total strain energy with the drilling rotation angle to verify that it follows a quadratic variation. The example is performed with a rectangle triangle subjected to an in-plane shear deformation in torsion mode as deﬁned in ﬁgure 9.5 (red line), and a simple quadratic function (blue line). Figure 9.8: Comparison of the variation of the total strain energy with the drilling rotation angle to verify that it follows a quadratic variation. The example is performed with a rectangle triangle subjected to an in-plane shear deformation in torsion mode as deﬁned in ﬁgure 9.5 (red line), and a simple quadratic function (blue line). then ϕij2 + θij2 →0 ⇔ϕij →0 ∧θij →0 ⇔ρ2 →0 ⇔ρ →0 (9.14) Ψij = cos2 ω · Ψij ϕ + sin2 ω · Ψij θ (9.15) lim ρ2→0Ψij = lim ϕij→0 θij→0 Ψij = cos2 ω · 1 3 + sin2 ω · 1 3 = 1 3 (9.16) Now we can rewrite equation (9.10) pointing out the solution of the indetermination. Ψij = Ψij(ϕij, θij) =        ϕij2 · Ψij ϕ + θij2 · Ψij θ ϕij2 + θij2 if ϕij2 + θij2 ̸= 0, 1 3 if ϕij2 + θij2 = 0. (9.17) (9.17) It is also important to point out that the value of ϕij needs to be com- 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS 113 Figure 9.9: Geometric interpretation of the change of variables applied in equations (9.12) and (9.13) at the limit when ρ2 →0. Figure 9.9: Geometric interpretation of the change of variables applied in equations (9.12) and (9.13) at the limit when ρ2 →0. puted as1 puted as1 ϕij = 2 · arctan Å∥∥xi −xj∥ni −∥ni∥(xi −xj)∥ ∥∥xi −xj∥ni + ∥ni∥(xi −xj)∥ ã −π 2 (9.18) (9.18) First derivative The solution of the indetermination in the case of the ﬁrst derivative can be obtained quite straightforwardly using the same change of variables indicated in equations (9.12) and (9.13). ∂Ψij ∂˜xr = ∂Ψij ∂xh (v) = 2 sin ω · cos ω · ∂ω ∂xh (v) (Ψθ −Ψϕ) + cos2 ω ∂Ψϕ ∂x(v) + sin2 ω ∂Ψθ ∂x(v) (9.19) 1Equation (9.18) always delivers a correct value of ϕij ∈[0, π]. And as reported by Kahan in [61, 62] this formula is much more accurate than the other trigonometric formulæ. Indeed, the author has traced back the cause of the numerical loss of accuracy in the calculation of some examples to an inadecuate deﬁnition of ϕij using other trigonometric formulæ. CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 114 lim ϕij→0 θij→0 ∂Ψij ∂xh (v) =2 sin ω · cos ω · ∂ω ∂xh (v) · 0 + cos2 ω ∂Ψϕ ∂x(v) ϕij=0 + sin2 ω ∂Ψθ ∂x(v) θij=0 = = cos2 ω ∂Ψϕ ∂ϕij ϕij=0 ∂ϕij ∂x(v) + sin2 ω ∂Ψθ ∂θij θij=0 ∂θij ∂x(v) = = cos2 ω · 0 · ∂ϕij ∂x(v) + sin2 ω · 0 · ∂θij ∂x(v) = 0 (9 20) (9.20) We can then write: We can then write: ∂Ψij ∂˜xr = ∂Ψij ∂xh (v) =                                    1 ϕij2 + θij2 ß1 3 ïÅ 4ϕij cos ϕij 1 + cos ϕij − −2ϕij2 sin ϕij (1 + cos ϕij)2 ã ∂ϕij ∂xh (v) + + (2θij cos6 θij− −6θij2 cos5 θij sin θij) ∂θij ∂xh (v) ô − −2Ψij Ç ϕij ∂ϕij ∂xh (v) + θij ∂θij ∂xh (v) å´ if ϕij2 + θij2 ̸= 0, 0 if ϕij2 + θij2 = 0. (9.21) (9.21) The expressions for ∂ϕij ∂xh (v) and ∂θij ∂xh (v) can be found in appendix D. 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS 115 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS In order to ﬁnd a workaround, let’s remind us the need for having this second derivative. The second derivative is required by the Newton- Raphson algorithm to solve the non-linear system of equations (equation (6.7 The second derivative appears because the geometric non-linearities al- ready use ﬁrst derivatives of the kinematic expressions. Therefore, the second derivative is used by the Newton-Raphson algorithm to make an approximation of the solution in the iterative process. Then, the expres- sion needed to overcome the indetermination at the point ρ2 →0 shall preserve the equilibrium if the equilibrium is reached at that point; and should push the approximation out of the indetermination point if that is not the solution. Once the approximation is out of the indetermination, full convergence properties of the Newton-Raphson algorithm are recovered. Let’s take a look at ﬁgure 9.9. using this construction we can establish that: lim ϕij→0 θij→0 cos2 ω = dϕij2 dϕij2 + dθij2 (9.24) lim ϕij→0 θij→0 sin2 ω = dθij2 dϕij2 + dθij2 (9.25) (9.24) (9.25) On the other hand, since we are in the context of deriving with respect to ∂xh (v) and ∂xs (w): these are the only two variables that affect dϕij and dθij. Therefore, in this context and only in this context: On the other hand, since we are in the context of deriving with respect to ∂xh (v) and ∂xs (w): these are the only two variables that affect dϕij and dθij. Therefore, in this context and only in this context: dϕij = ∂ϕij ∂xh (v) dxh (v) + ∂ϕij ∂xs (w) dxs (w) (9.26) dθij = ∂θij ∂xh (v) dxh (v) + ∂θij ∂xs (w) dxs (w) (9.27) (9.26) (9.27) It is very tempting to impose the condition It is very tempting to impose the condition dxh (v) = dxs (w) (9.28) (9.28) However, this is an arbitrary condition that does not respond to any math- ematical reality. The author decides to impose it nevertheless because it allows to obtain a determinate expression of equation (9.23). Second derivative Following the same process and using the change of variables speciﬁed in equations (9.12) and (9.13), the second derivative of equation (9.10) can be written as: ∂2Ψij ∂xh (v)∂xs (w) = cos2 ω ∂2Ψϕ ∂xh (v)∂xs (w) + sin2 ω ∂2Ψθ ∂xh (v)∂xs (w) + +2 sin ω · cos ω ñ ∂ω ∂xh (v) Ç ∂Ψθ ∂xs (w) −∂Ψϕ ∂xs (w) å + ∂ω ∂xs (w) Ç ∂Ψθ ∂xh (v) −∂Ψϕ ∂xh (v) åô + +2 ñ (cos2 ω −sin2 ω) ∂ω ∂xh (v) ∂ω ∂xs (w) + sin ω · cos ω ∂2ω ∂xh (v)∂xs (w) ô (Ψθ −Ψϕ) (9.22) lim ϕij→0 θij→0 ∂2Ψij ∂xh (v)∂xs (w) = lim ϕij→0 θij→0 cos2 ω ∂2Ψϕ ∂xh (v)∂xs (w) + lim ϕij→0 θij→0 sin2 ω ∂2Ψθ ∂xh (v)∂xs (w) (9.23) The author concludes that, in light of equation (9.23), the second deriva- tive does not have a solution to the indetermination when ρ2 →0. This is so because the second derivatives for Ψij ϕ and Ψij θ are different at the indetermination point. And as a consequence the second derivative of Ψij(ϕij, θij) depends on the direction (ω) considered. 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS ∂2Ψij ∂xh (v)∂xs (w) = 1 ϕij2 + θij2 ® 1 3 ñ −2ϕij2 (1 + cos ϕij)2 Ç sin ϕij ∂2ϕij ∂xh (v)∂xs (w) + +(2 −cos ϕij) ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) å − 8ϕij sin ϕij (1 + cos ϕij)2 ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + + 4 cos ϕij 1 + cos ϕij Ç ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + ϕij ∂2ϕij ∂xh (v)∂xs (w) å + 2 cos4 θij(cos2 θij− −12θij cos θij sin θij + 15θij2 sin2 θij −3θij2 cos2 θij) ∂θij ∂θij + ∂2Ψij ∂xh (v)∂xs (w) = 1 ϕij2 + θij2 ® 1 3 ñ −2ϕij2 (1 + cos ϕij)2 Ç sin ϕij ∂2ϕij ∂xh (v)∂xs (w) + +(2 −cos ϕij) ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) å − 8ϕij sin ϕij (1 + cos ϕij)2 ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + + 4 cos ϕij 1 + cos ϕij Ç ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + ϕij ∂2ϕij ∂xh (v)∂xs (w) å + 2 cos4 θij(cos2 θij− −12θij cos θij sin θij + 15θij2 sin2 θij −3θij2 cos2 θij) ∂θij ∂xh (v) ∂θij ∂xs (w) + +2 cos5 θij(θij cos θij −3θij2 sin θij) ∂2θij ∂xh (v)∂xs (w) ô − −2 ∂Ψij ∂xh (v) Ç ϕij ∂ϕij ∂xs (w) + θij ∂θij ∂xs (w) å −2 ∂Ψij ∂xs (w) Ç ϕij ∂ϕij ∂xh (v) + θij ∂θij ∂xh (v) å − −2Ψij Ç ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + ϕij ∂2ϕij ∂xh (v)∂xs (w) + ∂θij ∂xh (v) ∂θij ∂xs (w) + θij ∂2θij ∂xh (v)∂xs (w) å´ (9.32) The expressions for ∂2ϕij ∂xh (v)∂xs (w) and ∂2θij ∂xh (v)∂xs (w) can be found in appendix D. 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS ∂2Ψij ∂xh (v)∂xs (w) = 1 ϕij2 + θij2 ® 1 3 ñ −2ϕij2 (1 + cos ϕij)2 Ç sin ϕij ∂2ϕij ∂xh (v)∂xs (w) + +(2 −cos ϕij) ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) å − 8ϕij sin ϕij (1 + cos ϕij)2 ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + +2 cos5 θij(θij cos θij −3θij2 sin θij) ∂2θij ∂xh (v)∂xs (w) ô − −2 ∂Ψij ∂xh (v) Ç ϕij ∂ϕij ∂xs (w) + θij ∂θij ∂xs (w) å −2 ∂Ψij ∂xs (w) Ç ϕij ∂ϕij ∂xh (v) + θij ∂θij ∂xh (v) å − −2Ψij Ç ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + ϕij ∂2ϕij ∂xh (v)∂xs (w) + ∂θij ∂xh (v) ∂θij ∂xs (w) + θij ∂2θij ∂xh (v)∂xs (w) å´ (9.32) 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS MEMBRANE LOCKING OF THE BEST ELEMENT 116 and ﬁnally obtains (under all the previous conditions) lim ϕij→0 θij→0 ∂2Ψij ∂xh (v)∂xs (w) = −1 6 Ç ∂ϕij ∂xh (v) + ∂ϕij ∂xs (w) å2 ∂ϕij ∂xh (v) · ∂ϕij ∂xs (w) Ç ∂ϕij ∂xh (v) + ∂ϕij ∂xs (w) å2 + Ç ∂θij ∂xh (v) + ∂θij ∂xs (w) å2 − −2 Ç ∂θij ∂xh (v) + ∂θij ∂xs (w) å2 ∂θij ∂xh (v) · ∂θij ∂xs (w) Ç ∂ϕij ∂xh (v) + ∂ϕij ∂xs (w) å2 + Ç ∂θij ∂xh (v) + ∂θij ∂xs (w) å2 (9.31) (9.31) For the more general case—away from the indetermination—the expres- sion without change of variables is: For the more general case—away from the indetermination—the expres- sion without change of variables is: ∂2Ψij ∂xh (v)∂xs (w) = 1 ϕij2 + θij2 ® 1 3 ñ −2ϕij2 (1 + cos ϕij)2 Ç sin ϕij ∂2ϕij ∂xh (v)∂xs (w) + +(2 −cos ϕij) ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) å − 8ϕij sin ϕij (1 + cos ϕij)2 ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + + 4 cos ϕij 1 + cos ϕij Ç ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + ϕij ∂2ϕij ∂xh (v)∂xs (w) å + 2 cos4 θij(cos2 θij− −12θij cos θij sin θij + 15θij2 sin2 θij −3θij2 cos2 θij) ∂θij ∂xh (v) ∂θij ∂xs (w) + +2 cos5 θij(θij cos θij −3θij2 sin θij) ∂2θij ∂xh (v)∂xs (w) ô − −2 ∂Ψij ∂xh (v) Ç ϕij ∂ϕij ∂xs (w) + θij ∂θij ∂xs (w) å −2 ∂Ψij ∂xs (w) Ç ϕij ∂ϕij ∂xh (v) + θij ∂θij ∂xh (v) å − −2Ψij Ç ∂ϕij ∂xh (v) ∂ϕij ∂xs (w) + ϕij ∂2ϕij ∂xh (v)∂xs (w) + ∂θij ∂xh (v) ∂θij ∂xs (w) + θij ∂2θij ∂xh (v)∂xs (w) å´ (9.32) The expressions for ∂2ϕij ∂xh (v)∂xs (w) and ∂2θij ∂xh (v)∂xs (w) can be found in appendix D. 9.4. COMBINING THE Ψϕ AND Ψθ EXPRESSIONS Apply- ing equation (9.28) and substituting equations (9.26) and (9.27) into equa- tions (9.24) and (9.25), the author now rewrites the latter: lim ϕij→0 θij→0 cos2 ω = Ç ∂ϕij ∂xh (v) + ∂ϕij ∂xs (w) å2 Ç ∂ϕij ∂xh (v) + ∂ϕij ∂xs (w) å2 + Ç ∂θij ∂xh (v) + ∂θij ∂xs (w) å2 (9.29) lim ϕij→0 θij→0 sin2 ω = Ç ∂θij ∂xh (v) + ∂θij ∂xs (w) å2 Ç ∂ϕij ∂xh (v) + ∂ϕij ∂xs (w) å2 + Ç ∂θij ∂xh (v) + ∂θij ∂xs (w) å2 (9.30) (9.29) (9.30) CHAPTER 9. 9.5.1 A temporary ﬁx Despite the apparent improvement in the precision of the element for coarse meshes, there is something wrong in the new formulation: the el- ement does not converge to the right result. It seems that the element is too stiff. The author suspects that the kinematics associated to the drilling rotations are not well adjusted mesh-wise. Let’s analyze a simple example. Figure 9.11: Geometric interpretation of the deﬁnition of the drilling rota- tion provided by equation (9.6) and how it affects the actual deformation of the edges D-E-F in the mesh. Figure 9.11: Geometric interpretation of the deﬁnition of the drilling rota- tion provided by equation (9.6) and how it affects the actual deformation of the edges D-E-F in the mesh. Let’s consider a Cartesian mesh (for the sake of simplicity we will only consider the edges) subjected to a constant in-plane shear deformation (see ﬁgure 9.11). The dashed lines represent the orientation of the straight edges in the deformed conﬁguration as the A, B, C, G, H and I points shift position to their A’, B’, C’, G’, H’, and I’ counterparts. According to the the- ory of continuum mechanics we know that the actual position of the edges in the deformed conﬁguration is the one marked by the dashed lines. How- ever, because of the emulation of the drilling rotations of the nodes given by the abanico analogy and because of how we have applied them to the rotation of the edges at the node by virtue of equation (9.6), the orientation of the edges at the nodes is the one described by the short pointed lines. Indeed! By applying the relative rotation of the node with respect to the edge back to each edge, we recover the original relative angle of each of the edges with respect to the node, but at a rotated orientation! Of course, by doing this but forcing the edges to preserve their relative angles at the reference conﬁguration, the deformed edges become so crooked that they take deformation energy in excess. After realizing this mistake, the author sought a quick ﬁx and proposes to change the deﬁnition of θij. θij = ∆αij −∆αji = −θji (9.33) (9.33) Figure 9.12 provides an interpretation of the effect of this new deﬁni- tion. It clearly improves the behavior in the sense that the deformed edges do not take as much deformation energy. 118 CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 118 9.5 Fine-tuning the kinematics associated to the drilling rotations 9.5 Fine-tuning the kinematics associated to the drilling rotations The author reports that using the enhancement enabled by the drilling degrees of freedom as detailed above produces a signiﬁcant improvement in the precision of the BEST element. This can be observed in the example of the thick beam (see ﬁgure 9.10 which includes also another solution with a deﬁnition of symmetric drilling rotations that will be introduced in section 9.5.1). This example has been selected to test the improvements brought by the emulated drilling rotations because it is the one that best characterizes the in-plane shear locking behavior of the element. 9.5. FINE-TUNING THE DRILLING ROTATIONS 117 (a) Geometry and properties of the problem. (b) Displacement of the free tip of the beam. Results normalised using as reference solution 0,35533. (a) Geometry and properties of the problem. (b) Displacement of the free tip of the beam. Results normalised using as reference solution 0,35533. (c) Error convergence with different meshes. (c) Error convergence with different meshes. Figure 9.10: Study of the effect of including drilling rotations on the thick beam example. The results are scaled with respect to the reference so- lution published in [82]. The continuous line corresponds to the solution when no drilling rotations are used and the boundaries are constrained to remain ﬂat. The discontinuous line corresponds to the solution when drilling rotations are used and the formula presented in equation (9.6) is applied. The pointed line corresponds to the solution when drilling rota- tions are used and the formula presented in equation (9.33) is applied. 118 CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 9.5.1 A temporary ﬁx In fact, using equation (9.33) the BEST element converges to the right result. However, it still does so at 9.5. FINE-TUNING THE DRILLING ROTATIONS 119 Figure 9.12: Geometric interpretation of the deﬁnition of the drilling ro- tation provided initially by equation (9.6) in red, compared to the effect caused by the new deﬁnition of equation (9.33) in blue. Figure 9.12: Geometric interpretation of the deﬁnition of the drilling ro- tation provided initially by equation (9.6) in red, compared to the effect caused by the new deﬁnition of equation (9.33) in blue. the same mere linear rate of convergence reported earlier. This has many reasons. Firstly, because the new deﬁnition of θij no longer uses local in- formation. Instead, it is using a larger set of information without using it to have better precision. Secondly, because by using a symmetric deﬁnition of θij and θji a degree of freedom is lost in the kinematic description of the in-plane shear of the BEST element. So this represents a step backwards with respect everything discussed in this chapter. the same mere linear rate of convergence reported earlier. This has many reasons. Firstly, because the new deﬁnition of θij no longer uses local in- formation. Instead, it is using a larger set of information without using it to have better precision. Secondly, because by using a symmetric deﬁnition of θij and θji a degree of freedom is lost in the kinematic description of the in-plane shear of the BEST element. So this represents a step backwards with respect everything discussed in this chapter. For all the above reasons, the author also proposes another possible value for θij. θij = 0 (9.34) (9.34) This value is proposed considering that it can be as good as any other ar- bitrary value of θ. But with the caveat that unlike the developments pre- sented in this thesis previous to introducing the drilling rotations, now the kinematics of the element have been changed and therefore, even with a value of the drilling rotations set to zero, the boundaries of the element are no longer restricted to remain in a ﬂat curve. Therefore there is interest in knowing how this change affects the BEST element. 9.5.2 The real deal After the analysis included above, the author realizes that the formula in equation (9.6) is incomplete. The net effect of equation (9.6) is to apply a solid rigid rotation to all the edges converging on the node i, but preserving the relative angles of the reference conﬁguration. Notwithstanding, the CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 120 element should represent the deformed conﬁguration wholly. Being able to capture the solid rigid rotation of the node is a good accomplishment and it’s what the abanico analogy was all about (see section 9.2.1). But in order to really take advantage of the advanced kinematics brought in by the drilling rotations, it is absolutely necessary to add on top of the solid rigid rotation, the in-plane shear deformation; which is what the BEST element really needs. The drilling rotation shall be composed of two terms: one corresponding to the solid rigid rotation of the node and another one corresponding to the in-plane shear deformation of the element. θij = ∆αij \| {z } solid rigid rotation + γij \|{z} in-plane shear (9.35) (9.35) The difﬁculty now lies in that γij shall be determined consistently for all the edges converging at the node. The data available, which corre- sponds to the deformation of each of the neighboring elements, does not need to correspond to a unique value of the local shear deformation at the node. The idea is therefore to use the data provided by the neighboring nodes and approximate the in-plane shear deformation that occurs at the node. This characterization shall then serve to impose the correct direc- tion of the edges’ tangents at node i so that the in-plane shear deformation is compliant and consistent. Or put in other words: this construction shall enforce continuity at the nodes of the in-plane shear deformation. The following paragraphs attempt to provide a systematic way to characterize the local in-plane shear deformation at the node based on the information provided by the surrounding nodes. That is, ﬁnding a good approximation to a local value, based on global information. By proceeding this way, we will also satisfy the Patch Test. Because in the case of a constant deforma- tion of the mesh, there will be no difference between the local deformation at the node and the deformation obtained by averaging the information gathered from the nodes surrounding the node of interest. 9.5.2 The real deal Let’s consider the plane deﬁned in ﬁgure 9.2 and perpendicular to the normal vector at the node (ni). This is the plane used in ﬁgure 9.13. The author is interested in characterizing the shear deformation that happens on that plane. That is, at the tangent plane to the surface at i. Or in other words: in the limit when the area considered degenerates into the material point xi (see ﬁgure 9.13). The formula that determines the relative angle in the deformed con- ﬁguration between two given ﬁbers deﬁned at the reference conﬁguration is: cos βjk i = ˘eijT ∥˘eij∥· (I + 2 ˘E) · ˘eik ∥˘eik∥ s 1 + 2 ˘eijT ∥˘eij∥· ˘E · ˘eij ∥˘eij∥ · s 1 + 2 ˘eikT ∥˘eik∥ · ˘E · ˘eik ∥˘eik∥ (9.36) Where we maintain the notation provided in ﬁgures 9.2 and 9.13. The vec- tors ˘eij and ˘eik correspond to the eij and eik vectors after being rotated an angle ∆αij and ∆αik, respectively. After this rotation, the vectors recover cos βjk i = ˘eijT ∥˘eij∥· (I + 2 ˘E) · ˘eik ∥˘eik∥ s 1 + 2 ˘eijT ∥˘eij∥· ˘E · ˘eij ∥˘eij∥ · s 1 + 2 ˘eikT ∥˘eik∥ · ˘E · ˘eik ∥˘eik∥ (9.36) Where we maintain the notation provided in ﬁgures 9.2 and 9.13. The vec- tors ˘eij and ˘eik correspond to the eij and eik vectors after being rotated an angle ∆αij and ∆αik, respectively. After this rotation, the vectors recover Where we maintain the notation provided in ﬁgures 9.2 and 9.13. The vec- tors ˘eij and ˘eik correspond to the eij and eik vectors after being rotated an angle ∆αij and ∆αik, respectively. After this rotation, the vectors recover 9.5. FINE-TUNING THE DRILLING ROTATIONS 121 Figure 9.13: Scheme depicting the process to ﬁnd the value of θij using equation (9.35). The vectors eij 0 and eik 0 are not really in the plane of the ﬁgure, but they are represented anyway to indicate that their angle βjk i0 is the same as the angle between ˘eij and ˘eik. Figure 9.13: Scheme depicting the process to ﬁnd the value of θij using equation (9.35). The vectors eij 0 and eik 0 are not really in the plane of the ﬁgure, but they are represented anyway to indicate that their angle βjk i0 is the same as the angle between ˘eij and ˘eik. 9.5.2 The real deal Figure 9.13: Scheme depicting the process to ﬁnd the value of θij using equation (9.35). The vectors eij 0 and eik 0 are not really in the plane of the ﬁgure, but they are represented anyway to indicate that their angle βjk i0 is the same as the angle between ˘eij and ˘eik. the same relative angles as those of the reference conﬁguration. ˘E is a Green-Lagrange strain tensor, but differs from the tensors deﬁned in sec- tion 4.3 and chapter 6 in that this tensor is deﬁned only in two dimensions and serves only to characterize the shear deformation at the node. Now, by conﬁning the tensors to the two dimensions of the plane and deﬁning a local reference frame ˘T centered at xi and with the direction of the ﬁrst versor coincident with ˘eij, we can rewrite equation (9.36) as: cos βjk i = Ç 1 0 åT · (I + 2 ˘E[ ˘T ]) · Ç cos βjk i0 sin βjk i0 å Ã 1 + 2 Ç 1 0 åT · ˘E[ ˘T ] · Ç 1 0 å · Ã 1 + 2 Ç cos βjk i0 sin βjk i0 åT · ˘E[ ˘T ] · Ç cos βjk i0 sin βjk i0 å (9.37) (9.37) Taking into account that we only care about the in-plane shear deforma- tion produced between the reference and current conﬁgurations, it is safe to say that the tensor ˘E is a deviatoric tensor and has the following general expression: Taking into account that we only care about the in-plane shear deforma- tion produced between the reference and current conﬁgurations, it is safe to say that the tensor ˘E is a deviatoric tensor and has the following general expression: ˘E[ ˘T ] = ˘Eh [ ˘T ] \|{z} hydrostatic + ˘Ed [ ˘T ] \|{z} deviatoric = ˘Ed [ ˘T ] = ï Ei d γi/2 γi/2 −Ei d ò (9.38) (9.38) 122 CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT The (·)i superindices serve to emphasize the message that the ˘E tensor is deﬁned and has meaning only at the node i. The reference frame ˘T is deﬁned in ﬁgure 9.13 by the orthonormal vectors t′ 1 and t′ 2. 9.5.2 The real deal 1 2 Substituting equation (9.38) into equation (9.37) and rearranging terms we obtain the following non-linear equation: Ä 2 cos βjk i0 sin βjk i0 ä · Ç Ei d γi å = −cos βjk i0 + + cos βjk i » 1 + 2Ei d » 1 + 2Ei d · cos 2βjk i0 + γi · sin 2βjk i0 (9.39) (9.39) where Ei d and γi are the unknowns. A system of equations can be built with every instance of equation (9.39) for each edge i–k different than i–j. A number of cases can be considered: • In the case when there is only one triangle having node i as a vertex, the system of equations will have only one equation and cannot be solved. In this case the author chooses to let the edge’s tangents at the node follow the direction dictated by the mesh. And the value of θij in equation (9.35) is set to zero. So that node will not have any emulated drilling rotation. • In the case when there is only one triangle having node i as a vertex, the system of equations will have only one equation and cannot be solved. In this case the author chooses to let the edge’s tangents at the node follow the direction dictated by the mesh. And the value of θij in equation (9.35) is set to zero. So that node will not have any emulated drilling rotation. • In the case when there are two edges other than i–j converging at i, the system of equations will have exactly two equations and can be solved in an ordinary way. • In the case when there are three or more edges other than i–j, the system of equations will be over-determined. In this case, a least squares solution can be worked out. Assuming solved the system of equations—and ˘E obtained—, there is still another necessary step in order to obtain the value of γij in equa- tion (9.35). The Green-Lagrange deformation tensor provides information about the relative position of material points in the current conﬁguration based on their relative position in the reference conﬁguration. But does not provide any information about the relative position of a material point in the current conﬁguration with respect to its position in the reference conﬁguration. The tensor that provides that information is the tensor gra- dient of deformation ˘F . 9.5.2 The real deal I + 2 ˘E[ ˘T ] = ˘F T [ ˘T ] · ˘F [ ˘T ] (9.40) (9.40) In the general 2D case, ˘F has 4 independent components, so it is not possible to obtain them from this equation because this is the equation of a symmetric tensor and therefore 2 components are linearly dependent (identical). But we can take advantage of the fact that we have the in- formation on the solid rigid rotation provided by the abanico analogy, and beneﬁt from the polar decomposition of the deformation gradient tensor: ˘F [ ˘T ] = R \|{z} rotation · U \|{z} deformation (9.41) (9.41) 123 9.6. SUMMARY 9.6. SUMMARY where R is an orthogonal tensor and U is a symmetric tensor. Substituting equation (9.41) into equation (9.40) the following is obtained: where R is an orthogonal tensor and U is a symmetric tensor. Substituting equation (9.41) into equation (9.40) the following is obtained: I + 2 ˘E[ ˘T ] = U T · RT · R · U = U · U = U 2 (9.42) (9.42) The ﬁrst thing that equation (9.42) teaches us is that the deformation de- scribed by ˘E is independent of solid rigid rotations. Therefore it can be interpreted that the value of ˘E is obtained based on the relative angles of the edges converging on i in the reference conﬁguration (βjk i0 ), or the same relative angles after being rotated by the solid rigid rotation of the node and evaluated using the abanico analogy (see ﬁgure 9.13). The author chooses the latter. Therefore, the author assumes that there is no rigid body rotation after the relative orientations of the edges in the reference conﬁguration are recovered using the abanico analogy. 9.5.2 The real deal And writes: ˘F [ ˘T ] = R · U = I · U = U = q I + 2 ˘E[ ˘T ] (9.43) (9.43) Finally, γij can be computed using the following development: dx[ ˘T ] = ˘F [ ˘T ] · dX[ ˘T ] (9.44) ds · Ç cos(βjk i0 + γik) sin(βjk i0 + γik) å = dS · ˘F [ ˘T ] · Ç cos βjk i0 sin βjk i0 å (9.45) ik = arctan à F12 cos βjk i0 + F22 sin βjk i0 F11 cos βjk i0 + F12 sin βjk i0 −tan βjk i0 1 + tan βjk i0 F12 cos βjk i0 + F22 sin βjk i0 F11 cos βjk i0 + F12 sin βjk i0 í (9.46) (9.44) and in particular, when k = j and in particular, when k = j and in particular, when k = j γij = arctan ÅF12 F11 ã (9.47) 9.6 Summary The topic of membrane locking in thin shells has been discussed. The au- thor presents solid arguments existing in the bibliography to support the idea that a fully cubic BEST element should not suffer from membrane locking. However, it does because not all the potential of the cubic descrip- tion of the BEST element is unleashed. in order to ﬁx it, the author pro- poses to emulate the concept of drilling degrees of freedom and therefore provide the information needed to complete the geometric construction of the cubic triangle. The geometric construction of the element taking into account the cor- ner drilling rotations again brings about the issue of energy minimization. The author solves the problem by uncoupling the problem and reducing it to a 2D periodic torsion mode. The function that minimizes the energy for the in-plane shear deformation differs from the function that also mini- mizes the energy for the bending and out-of-plane membrane deformation CHAPTER 9. MEMBRANE LOCKING OF THE BEST ELEMENT 124 reported in section 5.4. This issue is solved by constructing a weighted average function. The indetermination of this weighted average function is also presented and solved for the primitive and its derivatives. Finally, the author reports a problem in the deﬁnition of the corner drilling rotation based on the rotations of the neighboring edges. The ini- tial deﬁnition results in overly crooked elements, and thus an excessively stiff response of the shell. In order to correct this issue a simple ﬁx is pre- sented, although represents a step backwards and makes all the effort of little advantage. g It is necessary a more thorough theoretic work on the deﬁnition of the corner drilling rotations based on the rotations of the neighboring trian- gles. The author presents an idea focused on sound continuum mechanics concepts about the deformation at a point. In particular, the author pro- poses that the solution should enforce the continuity of the in-plane shear deformation between elements at the nodes. This in-plane shear is char- acterized uniquely for each node but means that the in-plane shear angle depends on the orientation of the ﬁbers in the tangent plane. This idea has not been implemented numerically. Numerical results cannot be reported to verify the validity and goodness of the idea. CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS to the boundary, or in the direction parallel to the boundary. As in sec- tion 10.2, the boundary condition which inhibits out-of-plane rotations in the direction parallel to the boundary is presented as a continuous simple support along the boundary because it is completely equivalent. In all cases, the strategy used to constrain the rotations is focused on the construction process of the element’s geometry. That is, when comput- ing the location of the Bézier control points of the element, instead of using the information provided by the neighboring elements to compute the nor- mal direction, the ﬁxed normal will be used. And the corresponding equa- tion will be identiﬁed and swapped by a new one that applies the speciﬁc boundary condition. Note that the normal direction must not be directly imposed. Instead, its rotations are to be set. This is an important caveat, as doing otherwise would certainly generate undesired deformations and introduce unwanted deformation energy with respect to the reference con- ﬁguration. A boundary condition of particular interest to the ﬁnite element analyst is the symmetry boundary condition. This is a condition that does not correspond to a physical reality. But its effect is very real and moreover, it is useful. Because it allows the ﬁnite element analyst to signiﬁcantly reduce the computational cost of a simulation where this condition can be applied. Section 10.5 explains how to obtain this particular boundary condition combining the conditions presented in sections 10.2 and 10.3. Then, the chapter takes a turn and proceeds explaining how to solve the more advanced conﬁgurations involving internal hinges or kinks in the shell. These are detailed in sections 10.6 and 10.7. Finally, the author provides guidance into how to evolve to the more complex conﬁgurations involving the convergence of three or more shells using combinations of the boundary conditions already explained. This is presented in section 10.8. Chapter 10 A S HAS BEEN DISCUSSED in section 4.1.1, rotation-free elements have inherent difﬁculties in imposing boundary conditions over the rota- tions because they don’t have rotational degrees of freedom associated. The author has considered different alternatives like using Lagrange mul- tipliers, the penalty method or even applying the boundary condition in other weak form using Nitsche’s method. But none of these options are of the author’s liking because they have uninteresting algorithmic or stabil- ity implications. A S Somewhat surprisingly, the application of the Dirichlet boundary con- ditions on rotations is straightforward for the BEST element. This chapter will explain how and why. The original idea in the BEST element is to impose the rotation boundary conditions not on displacements but on the normals... because they are explicitly deﬁned! By doing this, the shape functions of the element are indeed modiﬁed in order to satisfy the bound- ary conditions. And this has the advantages that the size of the system matrix does not increase, there is no need to deal with esoteric coefﬁcients and furthermore: the symmetric and positive deﬁnite character of the ma- trix is conserved. This chapter is organized assigning each section to a speciﬁc boundary condition or combination of them. Section 10.1 presents the case where all the out-of-plane rotations are inhibited. This case is presented ﬁrst be- cause of its pedagogic value. Next, section 10.2 explains the boundary con- dition which inhibits the in-plane rotations. It is presented as a continuous simple support along the boundary because it is completely equivalent. Sections 10.3 and 10.4 present the decomposition of the boundary con- dition presented in section 10.1 according to the local axes at the boundary. That is: whether the rotations are inhibited in the direction perpendicular 125 126 10.2. CONTINUOUS SUPPORT 127 The reader shall not be confused by this expression. The interpretation of the above derivatives is not a free movement of the normal direction at the node independent of the nodal displacements. This is prevented by equation (10.1). No, instead equations (10.2) and (10.3) are implicitly and subtly modifying the derivatives of the pseudo-deformation tensor B and ∂B ∂˜x in a way that provides the necessary information to the stiffness matrix to account for the particular way the element will deform maintaining the normal direction ni still. Notice that this condition also results as a combination of the condi- tions described in sections 10.3 and 10.4 (see ﬁgure 10.1). Figure 10.1: Figure showing the meaning of a fully clamped boundary condition as a combination of two different conditions. See ﬁgures 10.3 and 10.4. Figure 10.1: Figure showing the meaning of a fully clamped boundary condition as a combination of two different conditions. See ﬁgures 10.3 and 10.4. 10.1 Fully clamped boundary condition This is the easiest condition and will be presented ﬁrst for pedagogic pur- poses. Fully clamped means that nothing moves. This means that all three displacements and rotations are ﬁxed. Let’s talk about the rotations as the process for the displacements is standard in the literature. That the rotations are ﬁxed is equivalent to the following statement: ni = N i ∀t > t0 (10.1) (10.1) Equation (10.1) substitutes equation (5.3) for those nodes having the con- dition. As a result, equations (D.28) and (D.29) are substituted by equa- tions (10.2) and (10.3), respectively for those nodes having the fully clamped condition. Equation (10.1) substitutes equation (5.3) for those nodes having the con- dition. As a result, equations (D.28) and (D.29) are substituted by equa- tions (10.2) and (10.3), respectively for those nodes having the fully clamped condition. ∂ni ∂˜x = 0 (10.2) ∂2ni (∂˜x)2 = 0 (10.3) (10.2) (10.3) 10.2. CONTINUOUS SUPPORT 10.2 Continuous simply supported shell in the tangent plane direction A simple support would usually be applied directly to the displacement degrees of freedom. However, in order to properly apply a continuously supported boundary condition, attention to the rotations must be given as well. In this sense, if the edge of a triangle spanning nodes i and j is set to have a simply supported boundary condition in the tangent plane (see ﬁgure 10.2), this means that the drilling rotations associated to the i–j edge will be zero. θij = θji = 0 (10.4) (10.4) CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS 128 Figure 10.2: Figure showing the meaning of the boundary condition of a continuous simply supported shell edge in the tangent plane direction. Figure 10.2: Figure showing the meaning of the boundary condition of a continuous simply supported shell edge in the tangent plane direction. And therefore equation (9.7) becomes: dij = N i × (Xi −Xj) (10.5) dji = N j × (Xj −Xi) (10.6) (10.5) (10.6) (10.6) for each segment affected by the condition. And consequently the following also applies, substituting equations (D.42) and (D.43). ∂dij ∂˜x = 0 (10.7) ∂2dij (∂˜x)2 = 0 (10.8) (10.7) 10.3 Boundary clamped in the perpendicular direction Similarly to the case presented in section 10.1, in this case a condition is established on the normal ni. The key is to deﬁne the unit vector rep- resenting the direction about which the rotation of ni is not constrained. Let’s name that vector t⊥(see ﬁgure 10.3). Then, the following condition must be satisﬁed: ni · t⊥= N i · t⊥ (10.9) (10.9) In order to satisfy this condition, equation (5.3) is modiﬁed in the fol- 10.3. BOUNDARY CLAMPED IN THE PERPENDICULAR DIRECTION 129 lowing way: lowing way: lowing way: lowing way: ni = » 1 −(N i · t⊥)2 · (I −t⊥⊗t⊥) · ri P k=1 wk α/A◦· ˆyk (I −t⊥⊗t⊥) · ri P k=1 wk α/A◦· ˆyk + (N i · t⊥) · t⊥ (10.10 (10.10) For the general case it is necessary to explicitly deﬁne the direction of t⊥by the user. Notice that this directly allows to specify a clamped condition allowing rotation about an axis not necessarily contained in the tangent plane of the shell. g p Deriving equation (10.10) we obtain: ∂ni ∂˜xr = » 1 −(N i · t⊥)2 · (I −t⊥· t⊥T ) (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk ·        ri X k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä − − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 wk α/A◦ˆyk          (10.11) Figure 10.3: Figure showing the meaning of the boundary clamped in the perpendicular direction. A graphical interpretation of the meaning of equation (10.9) is also presented. That is, the vector ni can rotate only about the vector t⊥. However, even if this condition is met at the two nodes bounding an edge, it cannot be enforced along all the edge. Doing so would require imposing conditions on the triangle’s central control point. 10.3 Boundary clamped in the perpendicular direction BOUNDARY CONDITIONS ON ROTATIONS 130 and and ∂2ni ∂xh (v)∂xs (w) = » 1 −(N i · t⊥)2 · (I −t⊥· t⊥T ) (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk ·        ri X k=1 ∂2 Ä wk α/A◦ˆykä ∂xh (v)∂xs (w) + + 3 Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂xh (v) Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 · · Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 wk α/A◦ˆyk− − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂xh (v) Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä − − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 ∂ ∂xh (v) Ä wk α/A◦ˆykä − −   Ç ri P k=1 ∂ ∂xh (v) wk α/A◦ˆyk åT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 + + Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂2 Ä wk α/A◦ˆykä ∂xh (v)∂xs (w) (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2   · ri X k=1 wk α/A◦ˆyk              (10.12) which substitute equations (D.28) and (D.29), respectively. which substitute equations (D.28) and (D.29), respectively. 10.3 Boundary clamped in the perpendicular direction Deriving equation (10.10) we obtain: Deriving equation (10.10) we obtain: ∂ni ∂˜xr = » 1 −(N i · t⊥)2 · (I −t⊥· t⊥T ) (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk ·        ri X k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä − − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t⊥· t⊥T ) · ri P k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä (I −t⊥· t⊥T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 wk α/A◦ˆyk          (10.11) Figure 10.3: Figure showing the meaning of the boundary clamped in the perpendicular direction. A graphical interpretation of the meaning of equation (10.9) is also presented. That is, the vector ni can rotate only about the vector t⊥. However, even if this condition is met at the two nodes bounding an edge, it cannot be enforced along all the edge. Doing so would require imposing conditions on the triangle’s central control point. CHAPTER 10. 10.4 Continuous simply supported shell in the normal direction If the edge of a triangle containing node i is set to have a simply supported boundary condition in the normal direction, this means that the curvature parallel to the boundary is constrained. This condition is imposed by deﬁn- ing a unit vector parallel to the boundary. Let’s name that vector t∥(see 10.4. CONTINUOUS SIMPLY SUPPORTED SHELL IN THE NORMAL DIRECTION 131 10.4. CONTINUOUS SIMPLY SUPPORTED SHELL IN THE NORMAL DIRECTION 131 In this case, the direction of t⊥corresponds to  (10.15) and 10.4. CONTINUOUS SIMPLY SUPPORTED SHELL IN THE NORMAL DIRECTION 131 ﬁgure 10.4). Then, the following condition must be satisﬁed: ni · t∥= N i · t∥ (10.13) (10.13) Which translates into the following variation of equation (5.3): ni = » 1 −(N i · t∥)2 · (I −t∥· t∥ T ) · ri P k=1 wk α/A◦· ˆyk (I −t∥· t∥ T ) · ri P k=1 wk α/A◦· ˆyk + (N i · t∥) · t∥ (10.1 (10.14) For the general case it is necessary to explicitly deﬁne the direction of t∥by the user. In particular, it must be deﬁned by the pre-processor, as the direction tangent to the boundary at the node cannot be obtained from the mesh and needs to be provided by the actual geometry of the shell. p y g y Deriving equation (10.14) we obtain the following expressions to sub- stitute equations (D.28) and (D.29): ∂ni ∂˜xr = » 1 −N i · t∥ 2 · (I −t∥· t∥ T ) (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk ·        ri X k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä − ∂ni ∂˜xr = » 1 −N i · t∥ 2 · (I −t∥· t∥ T ) (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk ·        ri X k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä − Figure 10.4: Figure showing the meaning of the boundary condition of a continuous simply supported shell edge in the direction normal to the tan- gent plane. A graphical interpretation of the meaning of equation (10.13) is also presented. That is, the vector ni can rotate only about the vector t∥. When this condition is met at the two nodes bounding an edge, then it is satisﬁed along all the edge. Figure 10.4: Figure showing the meaning of the boundary condition of a continuous simply supported shell edge in the direction normal to the tan- gent plane. A graphical interpretation of the meaning of equation (10.13) is also presented. That is, the vector ni can rotate only about the vector t∥. When this condition is met at the two nodes bounding an edge, then it is satisﬁed along all the edge. CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS 132 132 CHAPTER 10. 10.4. CONTINUOUS SIMPLY SUPPORTED SHELL IN THE NORMAL DIRECTION 131 BOUNDARY CONDITIONS ON ROTATIONS − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t∥· t∥ T ) · ri P k=1 ∂ ∂˜xr Ä wk α/A◦ˆykä (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 wk α/A◦ˆyk          (10.15) and ∂2ni ∂xh (v)∂xs (w) = » 1 −N i · t∥ 2 · (I −t∥· t∥ T ) (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk ·        ri X k=1 ∂2 Ä wk α/A◦ˆykä ∂xh (v)∂xs (w) + + 3 Å ri P k=1 wk α/A◦ˆyk ãT · (I −t∥· t∥ T ) · ri P k=1 ∂ ∂xh (v) Ä wk α/A◦ˆykä (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2 · · Å ri P k=1 wk α/A◦ˆyk ãT · (I −t∥· t∥ T ) · ri P k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 wk α/A◦ˆyk− − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t∥· t∥ T ) · ri P k=1 ∂ ∂xh (v) Ä wk α/A◦ˆykä (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä − − Å ri P k=1 wk α/A◦ˆyk ãT · (I −t∥· t∥ T ) · ri P k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2 · ri X k=1 ∂ ∂xh (v) Ä wk α/A◦ˆykä − −   Ç ri P k=1 ∂ ∂xh (v) wk α/A◦ˆyk åT · (I −t∥· t∥ T ) · ri P k=1 ∂ ∂xs (w) Ä wk α/A◦ˆykä (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2 + + Å ri P k=1 wk α/A◦ˆyk ãT · (I −t∥· t∥ T ) · ri P k=1 ∂2 Ä wk α/A◦ˆykä ∂xh (v)∂xs (w) (I −t∥· t∥ T ) · ri P k=1 wk α/A◦ˆyk 2   · ri X k=1 wk α/A◦ˆyk              (10.16) 10.5 Symmetry boundary condition This condition is obtained combining the conditions described in sections 10.2 and 10.3 (see ﬁgure 10.5). 10.5 Symmetry boundary condition This condition is obtained combining the conditions described in sections 10.2 and 10.3 (see ﬁgure 10.5). In this case, the direction of t⊥corresponds to 133 10.6. HINGE the director vector of the plane of symmetry. Figure 10.5: Figure representing the conditions that conﬁgure a symmetry boundary condition. Figure 10.5: Figure representing the conditions that conﬁgure a symmetry boundary condition. 10.6 Hinge With this case the author starts the cases where the shell presents two distinct areas that do not present continuity in the normal direction. This can be because the shell is folded, or hinged, or branched. The case of the hinge is treated ﬁrst for pedagogic purposes. Whenever a region of the shell presents discontinuity in the normal direction, the modeler using the BEST ﬁnite element shall instruct the pre-processor to identify the surface entities at each side of the discontinu- ity line. This can be established as a special boundary condition. Doing so allows the computer program to compute two different normal directions for one same node located on the discontinuity line: one for each surface entity (see ﬁgure 10.6). By discriminating the elements, the computation of the normals for one surface entity is independent of the elements at the other side of the hinge, which allows the two sides to have the free rotation intended by the hinge. For notation purposes, let’s label the magnitudes relative to the surface entity on the same side of the discontinuity line as the element being considered using a bullet and a vertical line (•\|) as a sub- script. The magnitudes relative to the surface entity on the opposite side of the discontinuity line to the element being considered will be identiﬁed 134 CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS using a vertical line and a circle (\|◦) as subscript. using a vertical line and a circle (\|◦) as subscript. ni •\| = ri X •\| k=1 wk α/A◦ˆyk ri X •\| k=1 wk α/A◦ˆyk (10.17) ni \|◦= ri X \|◦ k=1 wk α/A◦ˆyk ri X \|◦ k=1 wk α/A◦ˆyk (10.18) Figure 10.6: Whenever two surface entities meet at an angle, their nor- mals shall be distinguished at the nodes along the discontinuity line. Here the dark surface has the normal vector ni •\| at i. And the light surface has the normal vector ni \|◦at i. (10.17) (10.18) Figure 10.6: Whenever two surface entities meet at an angle, their nor- mals shall be distinguished at the nodes along the discontinuity line. Here the dark surface has the normal vector ni •\| at i. And the light surface has the normal vector ni \|◦at i. A hinge causes the two regions at each side of the hinge to have only C0 continuity at the boundary deﬁned by the hinge line. 2. The plane perpendicular to the other normal at the vertex (ni \|◦). 10.6 Hinge In order to enforce C0 continuity it is necessary not only to share the corner nodes, but also that the boundary control points along the hinge are coincident. So far, the construction of the BEST elements accomplishes this, as long as the normal vectors at the nodes are coincident for all the elements sharing the node. As soon as the normals become dissociated along a hinge line, the geometry of the intersection between the two elements at each surface en- tity becomes indeterminate. As a result, the author suggests the following approach. 135 10.6. HINGE Figure 10.7: Representation of two curved elements meeting at a hinge line. The geometry of the intersection is determined by intersecting the planes perpendicular to ni •\| (horizontal stripes) and ni \|◦(vertical stripes), respectively. Finally, the precise location of the intermediate control points along the hinge line is determined by the intersection of an additional plane (shaded dark gray), which is located at the distance speciﬁed by the parameter Ψ (equation (10.20)). Figure 10.7: Representation of two curved elements meeting at a hinge line. The geometry of the intersection is determined by intersecting the planes perpendicular to ni •\| (horizontal stripes) and ni \|◦(vertical stripes), respectively. Finally, the precise location of the intermediate control points along the hinge line is determined by the intersection of an additional plane (shaded dark gray), which is located at the distance speciﬁed by the parameter Ψ (equation (10.20)). The intersection line between the two elements needs to be computed (see ﬁgure 10.7). The location of the control points along the hinge de- termines the tangent direction of the hinge at the nodes. The tangent direction of the intersection between two surfaces can be computed as the intersection of the two surfaces’ tangent planes. Once the hinge’s tangent line has been obtained, the distance at which the control points shall be located needs to be determined. This distance will be different for the two connecting triangles because they are no longer co-planar. Therefore the author proposes averaging the values of Ψ(ϕ, θ) obtained for the triangles at each side of the hinge line. Using all the above, the contour control points of an element having one side at a hinge are determined intersecting the following three planes: 1. The plane perpendicular to the normal at the vertex (ni = ni •\|). 136 CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS 3. 10.6 Hinge The plane perpendicular to the edge of the ﬂat triangle considered. The exact location of this plane will be deﬁned by the magnitude Ψij. The mathematical expression of these three planes and their intersec- tion is presented in equation (10.19):   ni •\| T ni \|◦ T (xi −xj)T  · pij =   ni •\| T · xi ni \|◦ T · xi (xi −xj)T · (1 −Ψij)xi + Ψijxj   (10.19)   ni •\| T ni \|◦ T (xi −xj)T  · pij =   ni •\| T · xi ni \|◦ T · xi (xi −xj)T · (1 −Ψij)xi + Ψijxj   (10.19) (10.19) It’s obvious to say that when ni •\| and ni \|◦are coincident, then this sys- tem is indeterminate. But this would most probably be a result of a design error, since there is no point in setting a hinge in a shell only to result in a perfectly G1 continuous shape. The virtue of the hinge is that it allows the two sides to be at an angle and therefore increases the axial carrying load without overloading the bending capacity of the shell. If the two sides are G1 continuous, it means that there wouldn’t be a signiﬁcant bending mo- ment across the hinge line and therefore the hinge is totally unnecessary. Furthermore, a structural designer will use a hinge wherever large point loads are likely to appear in the transverse direction of the shell midsur- face, and would otherwise cause a very large and local bending moment. In order to better resist these loads, the structural designer will shape the shell parts at an angle to the load direction in order to better resist the loads with the phenomenal axial carrying capacity of the shell. Usually, the hinge will result in smaller bending moments in the shell. If however the displacements are an issue of concern, the structural designer will in- stead dispose a kink in the shell that will form an even stiffer structure. This conﬁguration will be covered in section 10.7. 10.6 Hinge Note also that dissociating the normals along the hinge line for each surface entity implies that the sets of nodes used to compute ni •\| and ni \|◦ are each a subset of all the nodes that would form the patch around the node in the absence of the hinge line (see equations (10.17) and (10.18)). But because the two magnitudes are used in the formulas, all the nodes in the patch are needed in order to compute the stiffness matrix of the \| \| are each a subset of all the nodes that would form the patch around the node in the absence of the hinge line (see equations (10.17) and (10.18)). But because the two magnitudes are used in the formulas, all the nodes in the patch are needed in order to compute the stiffness matrix of the element; regardless of whether they belong to a different surface entity. The equations (5.16), (7.25) and (7.26) are used changing the expres- sions for Aij p and bij p according to equation (10.19). The derivatives of these expressions as well as the derivatives of equation (10.23) can be found in appendix D. 10.7 Kinked shell The term kink is synonymous of fold or bend and implies a sharp corner of an otherwise smooth geometry. When a shell presents a kink, the ﬁnite element analyst can no longer assume G1 continuity; unless a very reﬁned mesh is used. For these cases it is common to assume that the angle be- tween the normal vectors at the edge of the elements on each side of the fold remains constant throughout the deformation process [42]. This as- sumption is consistent with maintaining the continuity of the shell’s trans- verse ﬁbers at the kink. C0 continuity is achieved using the same approach used for hinges in section 10.6. The location of the control points along the boundary of the elements at both sides of the kink is determined by solving the intersection of the two neighboring elements at each side of the kink or mesh set. How- ever, in this case, unlike in section 10.6, the relative angle of the normals is constrained and must be kept constant throughout the deformation. The author proposes to apply the same approach already explored in section 9.2.1 to maintain a rigid angle at the kink. The abanico analogy showed that by imposing a rotation equal to the average of the relative rotations of the remaining vectors, a rigid rotation of all the vectors is recovered (see section 9.5.1). The normal vector ni is obtained by conveniently rotating the average of the normals of the surrounding triangles on the same side of the kink line (ni •\|). In order to maintain the relative angle of ni with respect to the other normal at the node ⋆ni (see ﬁgure 10.8), the latter is also obtained by conveniently rotating the normal computed as an average of the normals of the surrounding triangles on the opposite side of the kink line (ni \|◦). The normals ni •\| and ni \|◦have been deﬁned in equations (10.17) and (10.18). •\| \|◦ First, let’s deﬁne the relative angle between the two normals as τ ij •\|◦= ( + arccos(ni •\| · ni \|◦) if (ni •\| × ni \|◦) · (xi −xj) > 0, −arccos(ni •\| · ni \|◦) if (ni •\| × ni \|◦) · (xi −xj) < 0. 10.6 Hinge Using the deﬁnition of equation (9.17), the value of Ψij is redeﬁned for the hinged sides of an element in the following way: Ψij = Ψ(ϕij •\|, θij •\|) + Ψ(ϕij \|◦, θij \|◦) 2 (10.20) (10.20) and where ϕij •\| and ϕij \|◦are deﬁned by extension of equation (9.18) ϕij •\| = arcsin ni •\| · (xi −xj) ∥xi −xj∥ (10.21) ϕij \|◦= arcsin ni \|◦· (xi −xj) ∥xi −xj∥ (10.22) θij •\| = arcsin sin ϕij •\| · (ni •\| · ni \|◦) −sin ϕij \|◦ cos ϕij •\| · ∥(I −ni •\| ⊗niT •\| ) · ni \|◦∥ (10.23) θij \|◦= arcsin sin ϕij \|◦· (ni \|◦· ni •\|) −sin ϕij •\| cos ϕij \|◦· ∥(I −ni \|◦⊗niT \|◦) · ni •\|∥ (10.24) ϕij •\| = arcsin ni •\| · (xi −xj) ∥xi −xj∥ (10.21) ϕij \|◦= arcsin ni \|◦· (xi −xj) ∥xi −xj∥ (10.22) (10.21) θij •\| = arcsin sin ϕij •\| · (ni •\| · ni \|◦) −sin ϕij \|◦ cos ϕij •\| · ∥(I −ni •\| ⊗niT •\| ) · ni \|◦∥ (10.23) θij \|◦= arcsin sin ϕij \|◦· (ni \|◦· ni •\|) −sin ϕij •\| cos ϕij \|◦· ∥(I −ni \|◦⊗niT \|◦) · ni •\|∥ (10.24) (10.24) Note that the above deﬁnitions of θij •\| and θij \|◦serve the only purpose of computing Ψij and they have no direct effect on the determination of the 10.7. KINKED SHELL 137 geometry of the element edge. Therefore, we do not care about the actual sign of θij •\| and θij \|◦because the function Ψij(ϕij, θij) is an even function with respect to both variables. Note also that dissociating the normals along the hinge line for each surface entity implies that the sets of nodes used to compute ni •\| and ni \|◦ Note also that dissociating the normals along the hinge line for each surface entity implies that the sets of nodes used to compute ni •\| and ni \|◦ are each a subset of all the nodes that would form the patch around the node in the absence of the hinge line (see equations (10.17) and (10.18)). 10.7 Kinked shell (10.25) τ ij •\|◦= ( + arccos(ni •\| · ni \|◦) if (ni •\| × ni \|◦) · (xi −xj) > 0, −arccos(ni •\| · ni \|◦) if (ni •\| × ni \|◦) · (xi −xj) < 0. (10.25) (10.25) CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS 138 Figure 10.8: Representation of two curved elements meeting at a kink. In this case the reference conﬁguration is represented (doted lines). This is used to explain that the relative angles between the two surfaces converg- ing at the kink are maintained in the deformed conﬁguration. The normal vector at each side of the fold line is obtained by conveniently rotating the normals of the adjacent elements (nj •\| and nj \|◦) until the relative angle be- tween the normals at each side of the kink line (T j •\|◦) is recovered. The ge- ometry of the intersection is determined following the procedure depicted in ﬁgure 10.7. Figure 10.8: Representation of two curved elements meeting at a kink. In this case the reference conﬁguration is represented (doted lines). This is used to explain that the relative angles between the two surfaces converg- ing at the kink are maintained in the deformed conﬁguration. The normal vector at each side of the fold line is obtained by conveniently rotating the normals of the adjacent elements (nj •\| and nj \|◦) until the relative angle be- tween the normals at each side of the kink line (T j •\|◦) is recovered. The ge- ometry of the intersection is determined following the procedure depicted in ﬁgure 10.7. It’s straightforward to infer that It’s straightforward to infer that It’s straightforward to infer that τ ij ◦\|• = −τ ij •\|◦ (10.26) τ ij ◦\|• = −τ ij •\|◦ (10.26) (10.26) The author deﬁnes an associated unit rotation vector aligned with the tan- gent of the kink line or fold (ˆf ij). The author deﬁnes an associated unit rotation vector aligned with the tan- gent of the kink line or fold (ˆf ij). ˆf ij =          + ni •\| × ni \|◦ ∥ni •\| × ni \|◦∥ if (ni •\| × ni \|◦) · (xi −xj) > 0, − ni •\| × ni \|◦ ∥ni •\| × ni \|◦∥ if (ni •\| × ni \|◦) · (xi −xj) < 0. (10.27) (10.27) 10.8. BRANCHING CONFIGURATIONS 139 Note that the functions deﬁning τ ij •\|◦and ˆf ij are not deﬁned when Note that the functions deﬁning τ ij •\|◦and ˆf ij are not deﬁned when (ni •\| × ni \|◦) · (xi −xj) = 0 This should not be a problem because this only happens if: This should not be a problem because this only happens if: • ni •\|, ni \|◦and (xi −xj) are co-planar (impossible); or • ni •\|, ni \|◦and (xi −xj) are co-planar (impossible); or • if ni •\| and ni \|◦are parallel, which is contradictory with the deﬁnition of kink. Finally equation (5.3) is modiﬁed and the normal vector is corrected ap- plying the Rodrigues’ formula and the abanico analogy. Also the normal of the other surface entity is corrected in order to determine the intersection between the two surface entities. ni = R ˆf ij, τ ij •\|◦−T ij •\|◦ 2 ! · ni •\| (10.28) ⋆ni = R ˆf ij, τ ij ◦\|• −T ij ◦\|• 2 ! · ni \|◦= RT ˆf ij, τ ij •\|◦−T ij •\|◦ 2 ! · ni \|◦ (10.29) (10.28) T ij •\|◦corresponds to the value of τ ij •\|◦in the reference conﬁguration. \| \| Equation (10.19) holds for determining the geometry of the intersection of the kink because the intersection of the planes perpendicular to ni and ⋆ni is the same as the intersection of the planes perpendicular to ni •\| and ni \|◦. The only thing that changes is the value of Ψij. It’s straightforward to infer that Ψij = Ψ(ϕij, θij) + Ψ( ⋆ϕij, ⋆ θij) 2 (10.30) (10.30) and where ϕij was deﬁned in equation (9.18), and ⋆ϕij is deﬁned by exten- sion ⋆ϕij = arcsin ⋆ni · (xi −xj) ∥xi −xj∥ (10.31) θij = arcsin sin ϕij · (ni · ⋆ni) −sin ⋆ϕij cos ϕij · ∥(I −ni ⊗niT ) · ⋆ni∥ (10.32) ⋆ θij = arcsin sin ⋆ϕij · ( ⋆ni · ni) −sin ϕij cos ⋆ϕij · ∥(I − ⋆ni ⊗ ⋆niT ) · ni∥ (10.33) 10.9 Summary In the BEST element, only the application of Dirichlet boundary conditions on the displacements of discrete points is straightforward. For the application of Dirichlet boundary conditions on displacements of lines it does not sufﬁce ﬁxing the value of the displacement degrees of freedom of the corner nodes. The displacement of the control points at the boundary also needs to be ﬁxed. And for that, the user needs to impose conditions either on the direction of the normal n at the nodes of the boundary or on the drilling rotations θ of the edges on the boundary. Likewise, the application of Dirichlet boundary conditions on the rota- tions of discrete points or lines also requires to impose conditions either on the direction of the normal n at the nodes or on the drilling rotations θ of the edges on the line. In doing this, the shape functions of the element are indeed modiﬁed to satisfy the boundary conditions. The advantages of this method are that the system matrix does not increase in size (as could have been if we were to impose additional constrains on the relation amongst the free variables and then applying the penalty method), and that there is no need to deal with esoteric coefﬁcients. But more importantly the symmetric and positive deﬁnite character of the matrix is conserved. The simplicity of applying those conditions on internal variables of the BEST element is a clear advantage with respect to other rotation-free shell elements that require the use of additional nodes to apply Dirichlet bound- ary conditions on the rotations of the element [100, 142]. However, in order to implement these boundary conditions in the over- all workﬂow, the pre-processor has to share the information about the con- ditions on lines with the solver. In this thesis, the author has used the GiD pre- and post-processor to run the examples. The solver has been pro- grammed using the RamSeries solver framework for the Tdyn solver suite of multi-physics problems. The problemtype deﬁnition for the RamSeries solver framework in GiD does not provide by default the information set on the line boundaries to the edges of the mesh. It transfers the information of the conditions of lines to the nodes on those lines. Therefore, it would be required to adapt the RamSeries problemtype to the requirements of the element. This thesis stops short of doing that adaptation. CHAPTER 10. BOUNDARY CONDITIONS ON ROTATIONS One possible approach to this scenario could be to compute an average intersection tangent. But the author is rather inclined to combine differ- ent boundary conditions. In this case, for the third and onward incoming surfaces to a kink or a hinge, an additional boundary condition should be applied. That would be a boundary condition equivalent to a continuous simply supported shell (both in the tangent and normal directions). Re- view sections 10.2 and 10.4. 10.8 Branching conﬁgurations In the previous two sections the author has presented general approaches to solve hinges and kinks in shells. The approach presented takes advan- tage that only two different surfaces converge at each interface. Thanks to this, the tangent direction of the intersection of both surfaces is eas- ily determined. When the number of converging surfaces grows to three or more, the determination of the intersection tangent is not unique, in general. 140 10.9 Summary And all the numeric examples are run without the correct application of the boundary conditions on lines. The ease of the BEST element to apply Dirichlet boundary conditions on the rotations is equivalent to that of other a rotation-free shell ele- 141 10.9. SUMMARY 10.9. SUMMARY 141 ments, and satisﬁes one of the main design goals set out for the element in section 2.2. Chapter 11 Numerical examples T HIS CHAPTER PRESENTS A SERIES OF BENCHMARK TESTS using the BEST element developed in the present thesis. A series of tests have been performed on the element to evaluate its performance and also to ﬁnd out the goodness of the two possible values of θij proposed in section 9.5.1. The tests have been selected to evaluate the speciﬁc performance of the BEST element under the different deformation modes of a shell: mem- brane shear, membrane extension, and bending. Finally, the shell obstacle course is used to compare the performance of the BEST element with re- spect to other shell elements in the literature. T H 11.1.1 Thick beam This example consists on a cantilever thick beam subject to a point load at its tip. The purpose of this test is to conﬁrm whether the in-plane shear mode is preventing the element from achieving fast convergence. The au- thor had the intuition that shear deformation, which had not been dealt with speciﬁcally, was responsible for the slow convergence of the element to the solution of the different problems. The reference solution for the displacement of the tip, as reported in [82], is 0,35533. Both translation degrees of freedom are restrained at the root nodes. The properties of the example are shown in ﬁgure 11.1. And the calculation meshes used are shown in ﬁgure 11.2. This problem has already been used to evaluate the improvement of the BEST element when drilling degrees of freedom are added to the kinematics of the element. But further comments are de- served than those provided in chapter 9. This example tests the plane defromation properties of the shell ele- ment. This includes the membrane stretch kinematics and in-plane shear kinematics. But the bending mechanism is non-existent because the out- of-plane displacements are restrained. This is a very well suited example to evaluate the improvements brought in by the drilling degrees of freedom in shell and plane deformation elements. The reader can observe that the BEST element with symmetric drilling degrees of freedom achieves similar precision to that obtained with other 143 CHAPTER 11. NUMERICAL EXAMPLES 144 (a) Geometry and conditions of the problem. (b) Vertical displacement and its error of the free tip of the beam. Results normalised using as reference solution 0,35533. (a) Geometry and conditions of the problem. (b) Vertical displacement and its error of the free tip of the beam. Results normalised using as reference solution 0,35533. Figure 11.1: Thick beam. Comparison of results with other elements from the literature. The data from other elements is obtained from [82]. elements in the literature. Although there are elements with signifﬁcantly better precision. Unlike the next example of the cylinder under torsion, in this example the conﬁguration of the element setting the angle θij = 0 does not improve the results over what had been presented in ﬁgure 8.7 on page 94. 11.1.1 Thick beam Indeed, the results are identical because in this problem with a ﬂat geometry, the modiﬁcation on the computation of the control points introduced in sec- tion 9.2.2 (see ﬁgures 5.5 and 9.3 on page 52 and on page 105) does not produce any difference in the computation of the control points and their kinematics if θij is set to 0. Nevertheless, in all the cases the error converges linearly. Therefore, 145 11.1.2 Torsion of tube This example consists on a cylindrical shell subject to pure torsion. In this case the torque is applied using 4 point loads on one of its bases, while the other base is ﬁxed (see ﬁgure 11.3). This example was devised by the author to test the the BEST element subject to in-plane shear and comple- ment the results obtained in the “thick beam” example. In the thick beam example the geometry is ﬂat and therefore the BEST elements do not have any curvature. In this example, the deformation remains strictly as in- plane shear, but in this case the geometry of the BEST elements is curved, therefore the advantages of the kinematics implemented in section 9.2.2 (see ﬁgures 5.5 and 9.3 on page 52 and on page 105) may have an effect. So this simple example using Saint-Venant’s theory for pure torsion of hol- low tubes is used. In order to prevent an excessive effect of the localised loads, a thick wall is used in this example. And the reference measure- ments are taken in the midle of the cylinder to stay away from those local deformations caused by the point loads. Figure 11.3: Geometry and conditions of the problem. Elevation and lat- eral view. Figure 11.3: Geometry and conditions of the problem. Elevation and lat- eral view. Under Saint-Venant’s hypotheses, which are valid for a straight cylin- der under uniform torque, the rate of angular torsion is constant along the length of the tube and there is no warping of the cross-sections of the cílinder. This allows to ﬁnd a very simple solution to the problem. α = Mtor · l G · I0 (11.1) (11.1) Where α is the total rotation of the tube from end to end, Mtor is the torque applied, l is the distance from the ﬁxed base of the cylinder to the mea- surement point, and I0 is the polar moment of the tube’s section. For a thin walled tube the polar moment of inertia can be calculated as I0 = 2 · π · R3 · h (11.2) (11.2) where h is the wall thickness. Then, the tangential displacement of the measurement point is equal to the angle α times the radius. Under these assumptions, the displacement of a point A in the middle of the cylinder should be 6, 621 · 10−4. CHAPTER 11. NUMERICAL EXAMPLES 146 11.1. IN-PLANE SHEAR ORIENTED EXAMPLES if the improvement suggested in section 9.5.2 were to be be successful and the element achieved the full cubic convergence enabled by the potential of the cubic nature of the description of its kinematics, then the BEST element would prove to be a superior element to the other elements in the literature with similar characteristics. (a) Mesh with 20 degrees of freedom. (b) Mesh with 28 degrees of freedom. (c) Mesh with 86 degrees of freedom. (d) Mesh with 298 degrees of freedom. Figure 11.2: Calculation meshes used for the thick cantilever beam exam- ple. (a) Mesh with 20 degrees of freedom. (a) Mesh with 20 degrees of freedom. (a) Mesh with 20 degrees of freedom. (b) Mesh with 28 degrees of freedom. (b) Mesh with 28 degrees of freedom. (b) Mesh with 28 degrees of freedom. (c) Mesh with 86 degrees of freedom. (c) Mesh with 86 degrees of freedom. (d) Mesh with 298 degrees of freedom. (d) Mesh with 298 degrees of freedom. Figure 11.2: Calculation meshes used for the thick cantilever beam exam- ple. CHAPTER 11. NUMERICAL EXAMPLES 11.1.2 Torsion of tube But instead, the numerical solution for the average of the mid section seems to converge towards a value of 6, 6046·10−4, which 11.1. IN-PLANE SHEAR ORIENTED EXAMPLES 147 is within a 0, 5% margin of error. Certainly the localized application of the torque load leads to localized deformations which in turn leaves less deformation energy available to the torsion mechanism. The ﬁrst thing to notice when looking at the results in ﬁgure 11.4 is that for the averaged measurement of the tangential displacement of the mid section of the tube, the solutions obtained using a value of θij = −θji equation (9.33), differ from the solutions obtained using a value of θij = 0 equation (9.34). More so: while with θij = 0 the solution converges rapidly, with θij = −θji the convergence is much slower. The second thing to no- tice is that for the tangential displacement of point A, and when using θij = −θji, the structured and unstructured meshes yield different results, whereas when using θij = 0 the results with structured and unstructured meshes converge to the same value. This leads the author to suspect about the validity of the results obtained with θij = −θji, which is a very arbi- trary condition. The results obtained with θij = 0 converge to a value of 6.566 · 10−4 for the tangential displacement of point A. Figure 11.5 shows an unstructured mesh using 1268 nodes and the corresponding converged results. The results are as expected. Half of the cylinder exhibits the effects of the localised loads. This can be observed in the displacements in the direction of the cylinder’s axis, which although they are restricted to vey low values (of the order of 10−5), they pinpoint clearly the position of the point loads that provide the torque. The other half of the cylinder does not exhibit any displacement in the direction of the axis. This is consistent with the theory. The displacements in the X and Z directions are conjugate of each other, as they represent the two directions in the cross section of the cylinder. They also reﬂect the position of the point loads. It is best then to analyze the modulus of the displacement in order to evaluate the results in the cross section of the cylinder, because we have already determined that the displacement in the axial direction is 0. 11.1.2 Torsion of tube Taking a look at the modulus of the displacements we can conclude that the tangential displacement varies linearly along the ﬁrst half of the cylinder, which is also consistent with the theory. But the most interesting detail to notice is that, contrary to what could be expected, the convergence obtained in this example with θij = 0 is no- ticeably faster than what was achieved in section 8.3.1 on page 94. In order to introduce the drilling rotations, the construction of the BEST element has been changed (see section 9.2.2). This result is a deﬁnitive proof that the improved kinematics of the BEST element to emulate drilling rota- tions, which unleash the cubic nature of the element for the in-plane shear deformations, are a step in the right direction. Even with the drilling rotations locked —with θij = 0—, when the element has curvature, the boundary of the element is no longer conﬁned to a ﬂat curve, and instead is allowed to warp. Even without fully unleashing the potential that the drilling rotations provide, just this slight modiﬁcation represents a great improvement for the accuracy of the element. Remains to test whether the proposed value by the author for θij as a sum of the rigid body rotation and the rotation caused by the in-plane shear (see equation (9.35) on page 120), can yield even better convergence properties. 148 CHAPTER 11. NUMERICAL EXAMPLES (a) Average tangential displacement of the central section of the cylinder. The average of a quarter of the section is computed. Results obtained using structured meshes. (a) Average tangential displacement of the central section of the cylinder. The average of a quarter of the section is computed. Results obtained using structured meshes. (b) Tangential displacement of point A (along the same generatrix of a loaded point). Results obtained using structured and unstructured meshes. (b) Tangential displacement of point A (along the same generatrix of a loaded point). Results obtained using structured and unstructured meshes. Figure 11.4: Torsion of tube. Comparison of results with 2 different values for θij: using a symmetric deﬁnition of θij according to equation (9.33), or setting θij = 0. 11.1. IN-PLANE SHEAR ORIENTED EXAMPLES 149 (a) Unstructured mesh with 1268 nodes. (b) Displacements in the axial direction of the cylinder. (c) Displacements in the X direction. (d) Displacements in the Z direction. (e) Modulus of displacements. 11.1.2 Torsion of tube (f) Tangential displacements of the cross section at point A. Figure 11.5: Torsion of tube. Results obtained using an unstructured mesh with 1268 nodes. These results correspond to a value of θij = 0. (b) Displacements in the axial direction of the cylinder. (a) Unstructured mesh with 1268 nodes. (b) Displacements in the axial direction of the cylinder. (a) Unstructured mesh with 1268 nodes. (c) Displacements in the X direction. (d) Displacements in the Z direction. (d) Displacements in the Z direction. (c) Displacements in the X direction. (e) Modulus of displacements. (f) Tangential displacements of the cross section at point A. Figure 11.5: Torsion of tube. Results obtained using an unstructured mesh with 1268 nodes. These results correspond to a value of θij = 0. 150 CHAPTER 11. NUMERICAL EXAMPLES 11.2. MEMBRANE ORIENTED TESTS 151 Figure 11.6c shows that results computed using both values of θij con- verge very quickly, although only quadratic convergence is obtained and not cubic. It appears that for the kinematics deﬁned in section 9.2.2, nei- ther θij = 0 nor θij = −θji are good enough to ensure optimal convergence. Nevertheless, both values do converge to the correct value and do so at a quadratic rate of convergence. Figure 11.7 shows no spurious effects of the mesh. Figure 11.7: Transverse shrinkage due to the Poisson modulus. The refer- ence value of the displacement at the edges is 0.3. Result obtained using a mesh with 16 divisions around the perimeter and 10 divisions along the lengh. Figure 11.7: Transverse shrinkage due to the Poisson modulus. The refer- ence value of the displacement at the edges is 0.3. Result obtained using a mesh with 16 divisions around the perimeter and 10 divisions along the lengh. 11.2.1 Cylinder under internal pressure This test was selected by the author in section 8.2.1 to determine whether using full quartic integration for the BEST element was worth it. Indeed, ﬁgure 8.4 made the case for using full quartic integration for the BEST ele- ment. In the present chapter, the author runs the test again after applying the changes described in chapter 9. (a) Geometry and properties of the problem. (b) Original and deformed mesh for a mesh with 16 divisions around the perimeter and 10 divisions along the length of the cylinder. (c) Diameter variation at point A. Results obtained using structured biased meshes. Figure 11.6: Cylinder under internal pressure. Results obtained for two possible deﬁnitions of the drilling rotations: θij = 0 and θij = −θji. The reference solution for the variation of the diameter is 1. (a) Geometry and properties of the problem. (a) Geometry and properties of the problem. (b) Original and deformed mesh for a mesh with 16 divisions around the perimeter and 10 divisions along the length of the cylinder. (a) Geometry and properties of the problem. (b) Original and deformed mesh for (a) Geometry and properties of the problem. (b) Original and deformed mesh for a mesh with 16 divisions around the perimeter and 10 divisions along the length of the cylinder. (b) Original and deformed mesh for a mesh with 16 divisions around the perimeter and 10 divisions along the length of the cylinder. (a) Geometry and properties of the problem. (c) Diameter variation at point A. Results obtained using structured biased meshes. Figure 11.6: Cylinder under internal pressure. Results obtained for two possible deﬁnitions of the drilling rotations: θij = 0 and θij = −θji. The reference solution for the variation of the diameter is 1. (c) Diameter variation at point A. Results obtained using structured biased meshes. Figure 11.6: Cylinder under internal pressure. Results obtained for two possible deﬁnitions of the drilling rotations: θij = 0 and θij = −θji. The reference solution for the variation of the diameter is 1. Figure 11.6: Cylinder under internal pressure. Results obtained for two possible deﬁnitions of the drilling rotations: θij = 0 and θij = −θji. The reference solution for the variation of the diameter is 1. 11.2.2 Parabolic roof This example was also used in chapter 8, section 8.2.2 to test the mem- brane capabilities of the BEST element. This test takes inspiration from the roof structure of the Dulles International Airport. Other buildings that have also used a similar roof typology are shown in ﬁgure 11.9. (a) (b) Figure 11.8: Parabolic roof problem. (a) Geometry and material properties of the problem. (b) Sample structured mesh with biased triangles. Mesh with 231 nodes and 400 elements. (b) (a) (b) (a) Figure 11.8: Parabolic roof problem. (a) Geometry and material properties of the problem. (b) Sample structured mesh with biased triangles. Mesh with 231 nodes and 400 elements. The parabolic roof example aproaches the geometry of the catenary. But is not equal. Therefore, under self-weight load it will develop some CHAPTER 11. NUMERICAL EXAMPLES 152 (a) The roof of the Dulles International Airport was designed by Eero Saarinen as a catenary. (b) The roof of the Caracas hipodrome was designed by Arthur Froehlich as a series of catenaries. (c) The canopy of the Bellvitge gas station (Barcelona) was also designed as a series of catenaries. (a) The roof of the Dulles International Airport was designed by Eero Saarinen as a catenary. (a) The roof of the Dulles International Airport was designed by Eero Saarinen as a catenary. (a) The roof of the Dulles International Airport was designed by Eero Saarinen as a catenary. (c) The canopy of the Bellvitge gas station (Barcelona) was also designed as a series of catenaries. (c) The canopy of the Bellvitge gas station (Barcelona) was also designed as a series of catenaries. (b) The roof of the Caracas hipodrome was designed by Arthur Froehlich as a series of catenaries. (b) The roof of the Caracas hipodrome was designed by Arthur Froehlich as a series of catenaries. (b) The roof of the Caracas hipodrome was designed by Arthur Froehlich as a series of catenaries. (c) The canopy of the Bellvitge gas station (Barcelona) was also designed as a series of catenaries. (d) The canopy of the Portuguese pavilion for the Expo’98 in Lisbon is a spectacular catenary conceived by Alvaro Siza Vieira and designed by Cecil Balmond. (d) The canopy of the Portuguese pavilion for the Expo’98 in Lisbon is a spectacular catenary conceived by Alvaro Siza Vieira and designed by Cecil Balmond. 11.2.2 Parabolic roof Figure 11.9: The example of the parabolic roof is inspired in catenary roofs built around the world. The parabolic shape is the funicular of a uniform distributed load. While the catenary shape is the funicular of the self- weight load. (d) The canopy of the Portuguese pavilion for the Expo’98 in Lisbon is a spectacular catenary conceived by Alvaro Siza Vieira and designed by Cecil Balmond. Figure 11.9: The example of the parabolic roof is inspired in catenary roofs built around the world. The parabolic shape is the funicular of a uniform distributed load. While the catenary shape is the funicular of the self- weight load. Figure 11.9: The example of the parabolic roof is inspired in catenary roofs built around the world. The parabolic shape is the funicular of a uniform distributed load. While the catenary shape is the funicular of the self- weight load. 11.2. MEMBRANE ORIENTED TESTS 153 11.2. MEMBRANE ORIENTED TESTS bending stresses (just as the cylinder under internal pressure does due to the decrease in curvature when the radius increases). The deformation of the roof will approach the catenary. The shape of a catenary is shallower at the trough and steeper at the abutments than the parabola. Therefore, the shell will rise at the trough and descend at the ﬂanks to approach the catenary (see ﬁgure 11.10). We will measure how much the shell elevates at the center of the free edge. The problem has been solved using both structured and structured meshes. The full model has been meshed. In the case of the structured meshes, non-symmetric meshes have been used (see ﬁgure 11.8b). (a) Elevation of the free edge’s midpoint. Averaged values of the two free edges. Convergence of the error. (a) Elevation of the free edge’s midpoint. Averaged values of the two free edges. Convergence of the error. (b) Lateral view of the parabolic roof. Comparison of the reference mesh and the deformed mesh. Solution obtained with a structured mesh of 10000 elements and 5151 nodes. Figure 11.10: Results of the parabolic roof example. (b) Lateral view of the parabolic roof. Comparison of the reference mesh and the deformed mesh. Solution obtained with a structured mesh of 10000 elements and 5151 nodes. Figure 11.10: Results of the parabolic roof example. Figure 11.10: Results of the parabolic roof example. 154 CHAPTER 11. NUMERICAL EXAMPLES Figure 11.10 shows the convergence of the different meshes with the two different settings of θij. In all cases the convergence is similar and the results converge quadratically with the number of nodes in the mesh. Reﬁnement has been analogous in the two main directions of the mesh. A converged solution has been obtained using an extra-reﬁned mesh in the direction of the parabola (with 160 divisions in the direction of the parabola and 150 divisions length-wise), and a value of 0,1882 has been used as a reference elevation at A. Reﬁnement has been analogous in the two main directions of the mesh. A converged solution has been obtained using an extra-reﬁned mesh in the direction of the parabola (with 160 divisions in the direction of the parabola and 150 divisions length-wise), and a value of 0,1882 has been used as a reference elevation at A. 11.2. MEMBRANE ORIENTED TESTS The effects of the biased mesh can be seen in the y-displacements ﬁeld, see ﬁgure 11.11. As the displacements along the y-axis direction ought to be 0, plotting the ﬁeld of displacements in the direction of the y-axis re- veals the magnitude of the effect that the biased mesh has on the numeric results. As ﬁgure 11.11 shows, the effect of mesh bias is some orders of magnitude smaller than the displacement values in the z- or x-axis. And its value diminishes as the mesh is reﬁned. Therefore, we can conclude that the BEST element is safe from the effects of mesh bias. (a) Mesh with 3600 elements and 1891 nodes. (b) Mesh with 24000 elements and 12231 nodes. Figure 11.11: Effects of mesh bias on the results. As the mesh is reﬁned the effect of the mesh bias diminishes. Displacements in the direction of the supported edges of the parabolic roof example. (a) Mesh with 3600 elements and 1891 nodes. (a) Mesh with 3600 elements and 1891 nodes. (b) Mesh with 24000 elements and 12231 nodes. (b) Mesh with 24000 elements and 12231 nodes. (b) Mesh with 24000 elements and 12231 nodes. Figure 11.11: Effects of mesh bias on the results. As the mesh is reﬁned the effect of the mesh bias diminishes. Displacements in the direction of the supported edges of the parabolic roof example. Figure 11.11: Effects of mesh bias on the results. As the mesh is reﬁned the effect of the mesh bias diminishes. Displacements in the direction of the supported edges of the parabolic roof example. 11.3 Bending oriented tests Shells have the property of resisting deformations through sheer mem- brane rigidity. That’s what confers them their unique advantage. In order 11.3. BENDING ORIENTED TESTS 155 to evaluate the bending properties of the BEST element, the author pro- poses using examples with a ﬂat geometry. In the linear regime, the BEST element cannot activate its membrane rigidity if the geometry is ﬂat and the loads are exclusively oriented per- pendicular to the shell’s surface. Under these circumstances, the shell acts as a plate because in the reference conﬁguration the shell has no curvature to transfer the loads to the membrane load bearing mechanisms. However, as the shell deforms due to bending, the current conﬁguration presents curvature and the membrane load bearing mechanisms are activated. g The author takes advantage of this property in order to evaluate the bending accuracy of the BEST element with 2 simple examples. Linear analysis The problem is set so that large deﬂections develop. Actually, the analyt- ical solution —as a beam— for the midpoint deﬂection in this particular case is wmax = 1 ⇒wmax L = 1 10. This is a value well beyond what is consid- ered small deﬂections ( w L ≈ 1 1000 ∨w < h). Nevertheless, we carry on with the linear analysis to continue the characterization of the BEST element. The most striking result when looking at ﬁgure 11.14 is that the sym- metric and unsymmetric meshes converge to different results. And the other obvious result is that both settings with θij = 0 and θij = −θji are identical (only the results with θij = 0 are presented). The second observa- tion stems from the fact previously explained that Bm ⊥∆˜xr. Therefore, the value of θij —which for a ﬂat geometry only affects Bm— is irrelevant. Let’s discuss into more detail the ﬁrst observation. In this case, the convergence deteriorates signifﬁcantly. The author ar- gues that the cause for this phenomenon, which occurs in a ﬂat geometry of the shell —but not when the shell has curvature— is in the deﬁnition of Ψij ϕ . We shall recall that the deﬁnition of Ψij ϕ in equation (5.21) on page 57 obeys to a very strong hypothesis: that of assuming that the curve de- scribed by the edge of the triangle is a ﬂat curve and also that this curve is a circular arch with symmetric angles with respect to the ﬂat triangle at the vertices. This determines a function Ψij ϕ (ϕ) which has a derivative equal to 0 when the geometry of the shell is ﬂat (ϕ = 0) See ﬁgure 9.7 on page 111. This derivative equal to 0 makes the element overly stiff in this particular example. Furthermore, in this ﬂat geometry the element becomes overly sensi- tive to mesh anisotropy. For the symmetric series of meshes, the author has created meshes with divisions of an equal size for the two main direc- tions of the beam; with 2 exceptions. In one case, the author has divided the width in 8 divisions while dividing the length in 96 divisions. This makes divisions with a size ratio of 1.2 : 1 for the two main directions of the mesh; and the result doesn’t ﬁt in the convergence curve. 11.3.1 Slender beam This example consists on a rectangular plate simply supported in the two far sides, acting as a simply supported ﬂat beam. The plate has a uniform load. See ﬁgure 11.12. Figure 11.12: Geometry, boundary conditions and property materials for the slender beam problem. Both short edges of the plate are simlpy sup- ported. Plan and elevation. Figure 11.12: Geometry, boundary conditions and property materials for the slender beam problem. Both short edges of the plate are simlpy sup- ported. Plan and elevation. Under linear analysis, this problem degenerates to a plate. The shell does not develop membrane strains because under linear analysis the loads perpendicular to the midsurface of the shell cause only displacements also perpendicular to the midsurface of the shell. And because of that orthog- onality, the tensor that relates the membrane deformations to the nodal coordinates (Bm) is also orthogonal to the nodal displacements (∆˜xr), and the multiplication ∆˜xr · Bm = 0. That’s why linear analysis is only valid for small deﬂections. The author has used 2 series of structured meshes to analyze this prob- lem. A set of symmetrically divided quadrangles and another set of unsym- mmetrically divided quadrangles have been used for the computations (see ﬁgure 11.13). CHAPTER 11. NUMERICAL EXAMPLES 156 (a) Symmetric mesh. Quadrangles divided symmetrically. (b) Unsymmetric/biased mesh. Quadrangles divided assymetrically. Figure 11.13: Samples of meshes for the slender beam problem. Top: ex- ample for the series of symmetric meshes. Bottom: example for the series of unsymmetric/biased meshes. (b) Unsymmetric/biased mesh. Quadrangles divided assymetrically. Figure 11.13: Samples of meshes for the slender beam problem. Top: ex- ample for the series of symmetric meshes. Bottom: example for the series of unsymmetric/biased meshes. Figure 11.13: Samples of meshes for the slender beam problem. Top: ex- ample for the series of symmetric meshes. Bottom: example for the series of unsymmetric/biased meshes. Linear analysis Instead, in a mesh with fewer nodes (8 divisions across and 80 divisions length-wise) the result is much more accurate and also ﬁts better in the convergence graph. 11.3. BENDING ORIENTED TESTS 157 Figure 11.14: Convergence of the deﬂection of the free edge’s midpoint. Series of results of the linear problem obtained with regular meshes of 1:1 sized cells (continuous and discontinuous lines); both symmetrically and unsymmetrically divided (see ﬁgure 11.13). Individual results for meshes of different size ratios are also plotted (see ﬁgure 11.15 for an example of a mesh with a 5:1 size ratio of the cells symmetrically divided). Figure 11.14: Convergence of the deﬂection of the free edge’s midpoint. Series of results of the linear problem obtained with regular meshes of 1:1 sized cells (continuous and discontinuous lines); both symmetrically and unsymmetrically divided (see ﬁgure 11.13). Individual results for meshes of different size ratios are also plotted (see ﬁgure 11.15 for an example of a mesh with a 5:1 size ratio of the cells symmetrically divided). Figure 11.15: Detail of an anisotropic mesh using symmetric divisions. Di- visions with a size ratio of 5 : 1. The resulting triangles have circumscribed circles of very different sizes. The larger circle has a radius 4.5 times that of the smaller circle; and an area 20 times larger. Figure 11.15: Detail of an anisotropic mesh using symmetric divisions. Di- visions with a size ratio of 5 : 1. The resulting triangles have circumscribed circles of very different sizes. The larger circle has a radius 4.5 times that of the smaller circle; and an area 20 times larger. 158 CHAPTER 11. NUMERICAL EXAMPLES Using a larger size ratio only exacerbates the problem (see ﬁgure 11.14). Using a larger size ratio only exacerbates the problem (see ﬁgure 11.14). The reason for this behavior stems from the deﬁnition of the normal vector at the nodes. The normal vector at the nodes is deﬁned assuming that the mesh approximates a quadric surface; and that each of the trian- gles in the mesh represents a section of that quadric surface (see [129]). This assumption determines the weights that ultimately deﬁne the direc- tion of the normal at the node. In the case of the ﬂat surface, there is no doubt about which is the normal vector. But the problem lies in its deriva- tive as the midsurface of the shell deforms. Linear analysis The relative shapes and sizes of the triangles surrounding the node determine the behavior of the deriva- tive of the normal. There is no geometric information to characterize the normal derivative because the surface has no curvature. Therefore, the characterization of the normal derivative depends only on the mesh topol- ogy. Then, if the relative sizes and shapes of the triangles surrounding the nodes of the ﬂat surface are uneven, most probably the information on the normal derivatives will behave oddly. This is for example the case for an extra-reﬁned mesh in the length-wise direction; with 500 elements length- wise and only 10 elements across. So the triangles have a ratio of 5 : 1 for the two main directions of the mesh (see ﬁgure 11.15). With a sym- metrically divided mesh, the differences between the triangles are very pronounced and the result is completely wrong in magnitude. However, with a biased mesh, the triangles are all equal (all equally skewed). And in this case the result is much closer to the correct one (although overstiff). For these two reasons —which also cause the mismatch of the conver- gence of the two mesh sets— the author concludes that the BEST element isn’t suitable for linear analysis of plate problems. And that in those cases it has to be used always in the non-linear regime in order to overcome the shortcomings described above. Those two shortcomings are corrected once the geometry has some curvature, which is the deﬁning characteristic of shells and what the BEST element has been designed for. Therefore, it becomes important to perform a non-linear analysis of shells under large-displacements assumptions in order to accurately char- acterize the exact elastic response of the shell to the loading applied. Non-linear analysis When switching to the non-linear analysis, we are turning this problem, which was conceived as a bending dominated problem, into a membrane dominated problem. Because, now, as the supported edges of the plate are restricted from moving toward each other, the shell will undergo important axial stresses which not only will dwarf the bending stresses of the shell, but which will also reduce the bending moments in the shell. Timoshenko and Woinowsky-Krieger [127] present the analytical solu- tion for the general problem of a rectangular plate simply supported on two oposite sides and with an uniform load. The analytical solution is 3, 0000 · 10−2. The difference with respect to the numerical results may be due to the fact that the analytical solution is for a strip of the plate sufﬁciently separated from the free edges of the plate. And not only this problem is too narrow to have any point sufﬁciently separated from the free edges, but in addition, the numerical result is measured at the edge. 11.3. BENDING ORIENTED TESTS 159 Figure 11.16: Convergence of the deﬂection of the free edge’s midpoint. Series of results of the non-linear problem obtained with regular meshes of 1:1 sized cells; both symmetrically and unsymmetricaly divided (see ﬁg- ure 11.13). Only the formulation with θij = 0 converges to a result. Deﬁn- ing θij = −θji does not yield convergence. Figure 11.16: Convergence of the deﬂection of the free edge’s midpoint. Series of results of the non-linear problem obtained with regular meshes of 1:1 sized cells; both symmetrically and unsymmetricaly divided (see ﬁg- ure 11.13). Only the formulation with θij = 0 converges to a result. Deﬁn- ing θij = −θji does not yield convergence. Nevertheless, the numerical result is sufﬁciently close to the analytical result to accept that the numerical result is correct. But the most relevant result of all is that in trying to solve the non- linear problem, the author has faced the same problem reported earlier in this thesis in section 5.4. When trying to solve the non-linear problem following the Newton-Raphson method and having deﬁned θij = −θji, the program cannot reach convergence. When the residual seems to be reduc- ing, all of a sudden it jumps in value again. The cause of the problem now and then is very similar. In [132], the problem was the deﬁnition of the kinematic parameter Ψ as ﬁxed. Non-linear analysis The author concluded that the space of solutions created by the way the Bézier triangle was being constructed, did not include the solution to the non-linear problem. With the present result we may reach the same conclusion. Either way, what causes this issue is that the problem is not well posed. Notwithstanding the previous negative result, the author reports ex- cellent convergence for the non-linear problem using the formula in equa- tion (9.34). That is: θij = 0. As explained in section 9.5.1, this is just a temporary ﬁx. And a more rich and comprehensive deﬁnition of θij is desirable in order to continue improving the BEST element. The author has proposed a framework to ﬁnd that comprehensive deﬁnition of θij in section 9.5.2. An in depth discussion will ensue in chapter 12. CHAPTER 11. NUMERICAL EXAMPLES 160 (a) Linear solution. The effect of the biased mesh can be noticed if paying close attention to the curved isolines. (b) Non-linear solution. The effect of the biased mesh is completely eliminated. Figure 11.17: Comparison of the effect of the biased meshes on the result of the displacements in the Z-direction. Results obtained with a mesh of 124 nodes. (b) Non-linear solution. The effect of the biased mesh is completely eliminated. Figure 11.17: Comparison of the effect of the biased meshes on the result of the displacements in the Z-direction. Results obtained with a mesh of 124 nodes. Be as it may, the BEST element shows linear convergence for the non- linear problem. Also important is the fact that convergence is reached with much coarser meshes than for the linear problem. The issue of the different converging values for the two different sets of meshes (symmet- ric and biased) is completely alleviated (see ﬁgure 11.16). The slight effect of the biased meshes on the results is mostly vanished (see ﬁgure 11.17). The only issue that remains is the effect of the meshes with uneven ele- ment sizes for the two main directions. Albeit even that problem has been signifﬁcantly reduced when solving the problem in the non-linear regime. 11.3.2 Circular plate This is also a classic problem that has the same characteristics as the slen- der beam problem above. However, in this case, because of the axisym- metric symmetry we can use a different mesh topology. And evaluate the performance of the BEST element in these conditions. Linear analysis The analytical solution for the linear analysis of a circular plates under uniform load is provided by Timoshenko and Woinowsky-Krieger [127] and by Ventsel and Krauthammer [135]. The maximum deﬂection at its center is calculated with the following simple formula: wmax = 3qR4(1 −ν2) 16Eh3 Å5 + ν 1 + ν ã (11.3) (11.3) In the case deﬁned in ﬁgure 11.18 the value for the maximum deﬂection at the center is 8, 28125 · 10−2. The results show that the convergence graph crosses the result. This produces a peculiar pattern in the error graph. The error graph in ﬁg- ure 11.20 shows two distinctive branches. Because the convergence graph 11.3. BENDING ORIENTED TESTS 1Shell elements don’t lock when solving plate problems because under linear analysis they do not develop membrane stresses. See section 11.3.1. 11.3. BENDING ORIENTED TESTS 11.3. BENDING ORIENTED TESTS 161 Figure 11.18: Geometry, boundary conditions and material properties of the circular plate problem. Plan and cross section. 161 Figure 11.18: Geometry, boundary conditions and material properties of the circular plate problem. Plan and cross section. Figure 11.19: Sample mesh of the circular plate problem with 10 divisions in the radial direction. Figure 11.19: Sample mesh of the circular plate problem with 10 divisions in the radial direction. crosses the exact result, only the later part of the error graph can be in- terpreted. Convergence in this problem is very slow. The cause of this apparently poor performance is that the author has not implemented in the program the boundary conditions for continuous supports described in chapter 10. Particularly the continuous simply supported shell in the tan- gent plane direction condition (see section 10.2) and the continuous simply supported shell in the normal direction condition (see section 10.4). Both conditions are needed to adequately model this problem. Without them, the continuous support of the boundary is only modeled discretely at the vertices of the triangles. So convergence to the right solution occurs much slower than it should because the model needs that many more nodes at the boundary. Another cause for the poor convergence performance of the BEST el- 162 CHAPTER 11. NUMERICAL EXAMPLES Figure 11.20: Convergence of the deﬂection of the central point in the circular plate problem. Linear solution. Results obtained with regular meshes (see ﬁgure 11.19). Figure 11.20: Convergence of the deﬂection of the central point in the circular plate problem. Linear solution. Results obtained with regular meshes (see ﬁgure 11.19). ement in this problem has already been discussed in section 11.3.1 and consists on the role played by the deﬁnition of the Ψϕ function and its derivative equal to 0 when ϕ = 0. Non-linear analysis Let’s swtich again the problem into the non-linear regime. Again, when doing so we are turning this problem, which was conceived as a bending dominated problem, into a membrane dominated problem. Timoshenko and Woinowsky-Krieger [127] present the analytical solu- tion for the general problem of a circular plate simply supported under uniform load for the particular case of ν = 0.3. They also provide a series of graphics for the case of ν = 0.25. This case is aproaching the very thin shell problem. And therefore is prone to membrane locking. As should have been expected, membrane locking occurs in this problem. It did not happen for the linear analysis, because shell elements do not lock when solving plate problems1 (Choi, Palma, Sanchez-Palencia and Vilariño [31]). However, this example shows the problem caused by not having found a comprehensive solution for the drilling rotations. Fixing the value of θij to 0 limits the degree of the poly- nomial expression of the in-plane kinematics of the element. Choi et al. prove in [31, p. 139] that representing the in-plane kinematics of the shell 11.3. BENDING ORIENTED TESTS 163 Figure 11.21: Evolution of the convergence of the BEST element as the thickness decreases. Example of a simply supported circular plate. Non- linear solution. Results obtained with regular meshes (see ﬁgure 11.19). Figure 11.21: Evolution of the convergence of the BEST element as the thickness decreases. Example of a simply supported circular plate. Non- linear solution. Results obtained with regular meshes (see ﬁgure 11.19). element with polynomials of lower order than the order used to represent the out-of-plane kinematics of the element leads inevitably to membrane locking. While Hakula, Leino and Pitkäranta demonstrate that even for ﬁnite elements with the same ﬁnite element representation in all the com- ponents of the displacement ﬁeld [51, p. 161], locking may still occur then the polynomial degree of such representation is low (1 or 2). Hakula et al. also demonstrate numerically that for such elements using polynomial ex- pressions of degree 3 and higher, the elements should not show signifﬁcant locking problems. In ﬁgure 11.21 the author presents the evolution of the convergence curves as the ratio thickness/radius becomes smaller and smaller. We can observe how the numerical solution using the BEST element progressively diverges from the exact solution. Non-linear analysis Even the pattern of the convergence curves changes at a given point, when it becomes apparent that even in 164 CHAPTER 11. NUMERICAL EXAMPLES Figure 11.22: Convergence of the deﬂection of the central point in the cir- cular plate problem. Non-linear solution. Results obtained with very re- ﬁned regular meshes (see ﬁgure 11.19). Particular case for h=0.005. Figure 11.22: Convergence of the deﬂection of the central point in the cir- cular plate problem. Non-linear solution. Results obtained with very re- ﬁned regular meshes (see ﬁgure 11.19). Particular case for h=0.005. the limit of the size of the elements approaching 0 the numerical solution will not reach convergence to the exact solution. For the case of h=0.005, the author presents in ﬁgure 11.22 an extended convergence curve that shows how as the mesh becomes very reﬁned it converges to the correct result, albeit very slowly. 11.4 Shell Obstacle Course Next, I present the comparison of the performance of the BEST element with other elements in the literature using three well known test exam- ples. These three examples are commonly refered to as the shell obstacle course [10] because “together they are a very discriminating set of prob- lems”. After having tested the BEST element with the previous set of exam- ples, the author has reached the conclusion that the variation of the for- mulation using θij = −θji according to equation (9.33) does not offer any advantage performance-wise nor precision-wise with respect to the basic solution θij = 0 proposed in equation (9.34). Furthermore, it has been demonstrated that using equation (9.33) does not produce convergence in non-linear problems, and therefore it leads to a non-well-posed problem; while equation (9.34) does yield convergence for non-linear problems. Be- 11.4. SHELL OBSTACLE COURSE 165 casue all of this, the tests of the shell obstacle course will be presented using only θij = 0. The comparison will be presented against a wide range of shell ele- ments to have an accurate view of how does the BEST element rank with respect other elements. For comparison with another family of rotation- free ﬁnite elements I have chosen the BST/enhanced basic shell triangle (EBST) family by Oñate, Zárate and Flores [95]. For comparison with other elements with rotations, I have taken the linear triangle S3 from ABAQUS [1] and a quadratic element family: the T6/T6H elements im- plemented in ANSYS [4]. Also the versions of the EBST shell element with rotational degrees of freedom called EBST+ and EBST+1 [145] are in- cluded in the comparison. Finally, also the quadrangular elements MITC9 [6], Q8H [54] and S4 (and its reduced integration version S4R) also from ABAQUS are included. The results published in [68] have been used to benchmark the BEST element with respect to the following elements: Q8H, MITC9, T6, T6H, S3, S4 and S4R. The author expresses his gratitude to professor Laulusa for gratiously providing his published results in tabulated form. For the BST/EBST/EBST+/EBST+1 family of elements, the results for comparison are taken from [95, 145] 11.4.1 Scordelis-Lo Roof This problem consists on a cylindrical roof under a uniform pressure load. The roof is simply supported on rigid diafragms on both ends (the di- afragms are rigid in their plane but ﬂexible out of plane). The structure is loaded with a uniform dead weight q = 90. (a) (b) Figure 11.23: Scordelis-Lo problem. (a) Geometric deﬁnition and material properties. (b) Mesh used for the computations. A structured mesh is used with biased triangles. The full domain is computed. (a) (b) (b) (a) Figure 11.23: Scordelis-Lo problem. (a) Geometric deﬁnition and material properties. (b) Mesh used for the computations. A structured mesh is used with biased triangles. The full domain is computed. The results presented are all normalized and the unit of comparison of the meshes is the total number of degrees of freedom used for 1 4 of the geometry. Figure 11.24 presents a comparison of the performance between a se- ries of shell elements. This problem exempliﬁes the virtues of shells to CHAPTER 11. NUMERICAL EXAMPLES 166 166 CHAPTER 11. NUMERICAL EXAMPLES Figure 11.24: Comparison of the convergence between different shell el- ements for the Scordelis-Lo problem. The reference solution is taken as 0.3024. Figure 11.25: Field of vertical displacements of the Scordelis-Lo roof prob- lem. Results obtained using a mesh with 64 divisions for half the span and half the arch. Representation of the deformation of the structure magni- ﬁed x10. Figure 11.24: Comparison of the convergence between different shell el- ements for the Scordelis-Lo problem. The reference solution is taken as 0.3024. Figure 11.24: Comparison of the convergence between different shell el- ements for the Scordelis-Lo problem. The reference solution is taken as 0.3024. Figure 11.25: Field of vertical displacements of the Scordelis-Lo roof prob- lem. Results obtained using a mesh with 64 divisions for half the span and half the arch. Representation of the deformation of the structure magni- ﬁed x10. Figure 11.25: Field of vertical displacements of the Scordelis-Lo roof prob- lem. Results obtained using a mesh with 64 divisions for half the span and half the arch. Representation of the deformation of the structure magni- ﬁed x10. 167 11.4. SHELL OBSTACLE COURSE resist loads through form and membrane stresses. Therefore, in this exam- ple the roof undergoes all sorts of membrane deformations while minimiz- ing bending. In the words of Belytschko et al. [10, p. 11.4.1 Scordelis-Lo Roof 239]: “A substantial part of the strain energy is membrane strain energy”. This implies also the presence of in-plane membrane shear in the shell; just as in the case of the circular plate problem. And they further predict: “[. . . ] inadequacies in membrane stress accuracy will severely inhibit convergence.” We already know that the BEST element lacks accuracy to represent in-plane shear strains and as a result its performance is clearly subpar when compared with all the other elements. Nevertheless, the author has also reported the reason for this defect (see section 11.3.2) and the belief that it can be very much alleviated (see chapter 9). Figure 11.25 displays the deformed structure magniﬁed. 11.4.2 Pinched Cylinder (a) (b) (c) Figure 11.26: Pinched cylinder problem. (a) Geometric deﬁnition and ma- terial properties. (b) Mesh used for the computations. A structured mesh is used with biased triangles. The full domain is computed. (c) Axial view of the deformation of the cylinder. Magniﬁcation factor: x500. 11.4.2 Pinched Cylinder 11.4.2 Pinched Cylinder (a) (b) (a) (b) (b) (a) (c) (c) (c) Figure 11.26: Pinched cylinder problem. (a) Geometric deﬁnition and ma- terial properties. (b) Mesh used for the computations. A structured mesh is used with biased triangles. The full domain is computed. (c) Axial view of the deformation of the cylinder. Magniﬁcation factor: x500. 168 CHAPTER 11. NUMERICAL EXAMPLES Figure 11.27: Comparison of the convergence between different shell ele- ments for the pinched cylinder problem. The reference solution is taken as 0.0018249. Figure 11.27: Comparison of the convergence between different shell ele- ments for the pinched cylinder problem. The reference solution is taken as 0.0018249. This example consists on a cylinder subjected to two diametrically op- posed puntual loads, see ﬁgure 11.26. The ends of the cylinder are simply supported on rigid diafragms (the diafragms are rigid in their plane but ﬂexible out of plane). It is (according to Belytschko et al. [10, p. 239]) “one of the most severe tests for both inextensional bending modes and com- plex membrane states”. However, this statement is challenged by Chapelle and Bathe [22] who state that in this problem “pure bending is inhibited”. Therefore it does not test the elements for inextensional bending modes to the degree claimed by Belytschko et al. The author agrees with Chapelle and Bathe. 11.4.1 Scordelis-Lo Roof As we will see in section 11.4.3, the hemispherical shell prob- lem really poses a problem to those elements which are not able to repro- duce inextensional bending modes. g The results of the Pinched Cylinder in ﬁgure 11.27 show the excellent behavior of the BEST element. It clearly outperforms the EBST family of elements (with and without rotations). It also outperforms all the other triangular elements in the set; both linear and quadratic (S3, T6 and T6H). It is on par with the quadrilateral elements S4 and S4R. The only elements in the set that perform better than the BEST element in this problem are the MITC9 and the Q8H elements. Nevertheless, the author believes that if the in-plane shear problems of the BEST element are solved, there is still room for improvement even for this excellent result. Figure 11.26c and ﬁgure 11.28 display the deformed structure magni- 169 11.4. SHELL OBSTACLE COURSE ﬁed. Figure 11.28: Field of displacements (modulus) of the pinched cylinder problem. Results obtained using a mesh with 96 divisions for a quarter of the circumference and 64 divisions for half the length of the cylinder. Representation of the deformation of the structure magniﬁed x1000. ﬁed. Figure 11.28: Field of displacements (modulus) of the pinched cylinder problem. Results obtained using a mesh with 96 divisions for a quarter of the circumference and 64 divisions for half the length of the cylinder. Representation of the deformation of the structure magniﬁed x1000. 11.4.3 Pinched Sphere / Hemispherical Shell The pinched sphere example (also called hemispherical shell) is also a clas- sic benchmark example for shell elements. It consists on a hemispherical shell truncated at the top according to an 18º hole. The shell is subject to diametral opposed forces on the 2 main axis. One pair of forces pulls the shell while the other pair of forces pinches the shell. This is an example of what Chapelle and Bathe [22] call non-hinhibited bending. Therefore, the problem is bending dominated and as Belytschko et al. report: “it exhibits almost no membrane strains”. Although, as Chapelle and Bathe point out, the geometry of the problem is very sensitive to the boundary conditions and could very easily become an inhibited bending problem. We shall ex- pect the BEST element to have difﬁculties solving this problem. As expected, ﬁgure 11.30 shows the issues of the BEST element with inextensional bending modes. The EBST family of elements performs bet- ter than the BEST element, although not by much. The author is sur- prised about the results reported by Laulusa et al. [68] for the S4 and S4R elements. Being all of them linear elements, it is remarkable that in this problem so prone to membrane locking they outperform other higher order CHAPTER 11. NUMERICAL EXAMPLES 170 (a) (b) Figure 11.29: Pinched sphere problem. (a) Geometric deﬁnition and mate- rial properties. (b) Mesh used for the computations. A structured mesh is used with symmetric triangles. The full domain is computed. (b) (a) Figure 11.29: Pinched sphere problem. (a) Geometric deﬁnition and mate- rial properties. (b) Mesh used for the computations. A structured mesh is used with symmetric triangles. The full domain is computed. Figure 11.30: Comparison of the convergence between different shell ele- ments for the hemispherical shell problem. The reference solution is taken as 0.094. Figure 11.30: Comparison of the convergence between different shell ele- ments for the hemispherical shell problem. The reference solution is taken as 0.094. 171 11.5. SUMMARY Figure 11.31: Field of vertical displacements of the hemispherical shell problem. Results obtained using a mesh with 32 divisions for a quarter of the equator and for each meridian. Deformations magniﬁcation factor x50. Figure 11.31: Field of vertical displacements of the hemispherical shell problem. Results obtained using a mesh with 32 divisions for a quarter of the equator and for each meridian. Deformations magniﬁcation factor x50. 11.4.3 Pinched Sphere / Hemispherical Shell elements such as the MITC9 and the Q8H; and the triangular S3 is not far behind. In the opinion of the author, the result obtained by Laulusa et al. in [68, Fig. 12] contradicts the results obtained by Hakula et al. in [51]. Figure 11.31 displays the contour ﬁll for the vertical displacements ﬁeld of the shell. 11.5 Summary Four sets of tests have been presented. The ﬁrst three sets have tested the BEST element convergence properties for the 3 different deformation modes of a shell: in-plane shear, axial strain (membrane) and bending. Also, for these ﬁrst three sets, the two different formulæ for the value of the drilling rotations —equations (9.33) and (9.34)— have been tested. The results obtained for the thick beam problem (section 11.1.1) demon- strate the beneﬁts of enriching the kinematics of the element through the value of θij. However the improvement is minimal because the value as- signed in equation (9.33) is not a sufﬁciently good approximation. On the other hand, the results obtained for the torsion of a tube give a ﬁrst hint at the problems that the equation (9.33) might introduce that surpass the advantages. This ﬁrst hint is a lack of consistency in the re- sults between structured and unstructured meshes. With respect to the membrane oriented tests, there is almost no differ- ence between the results obtained by the use of equation (9.33) or equa- tion (9.34). And in both problems, the cylinder under internal pressure, and the parabolic roof, the convergence is better than linear. There has been a loss of convergence speed in the problem of the cylinder under inter- nal pressure with respect to the results already reported in section 8.2.1. 172 CHAPTER 11. NUMERICAL EXAMPLES The difference between the two results can only be caused by the changes in the construction of the element to enable the θ drilling rotations (see the differences between ﬁgures 5.5 and 9.3). So, cubic convergence has been lost for this problem, but if a correct value of θ is found, it should be recoverable. Finally, in order to test the bending deformation mode of the BEST ele- ment, the author uses two plate examples: a rectangular plate conﬁgured as a slender beam and a circular plate. These examples are chosen because mathematically, in the linear case, the shell elements do not develop mem- brane strains. Therefore the analyst can focus exclusively on the bending mode response of the element. However, since there is no membrane strain and the geometry is ﬂat, this means that there is no difference in the re- sults between the use of equation (9.33) or equation (9.34). Nevertheless, a wealth of information is extracted from these two examples. 11.5 Summary The ﬁrst interesting result is that for the slender beam problem, dis- cretizing the domain with biased meshes or with symmetrically divided meshes, yields different results —although very close to the correct result. This result leads the author to conclude that the BEST element isn’t well suited for plate problems, because it needs the information provided by the curvature of the surfaces. The curvature of the surface embeds informa- tion onto the mesh to determine not just the direction of the normals, but also their derivatives when the surface deforms. With zero initial curva- ture, the mesh has zero initial information from the surface on how the normals will change as the surface deforms. And therefore that informa- tion becomes mesh-dependant. Since the plate doesn’t have initial curva- ture, the computation becomes overly sensitive to the mesh discretization. The author demonstrates these statements by turning the problem into geometrically non-linear. In this non-linear conﬁguration the BEST ele- ment recovers the consistency between the sets of meshes (biased trian- gles and symetrically divided triangles). But the author also discovers that using the equation (9.33) for θij, the BEST element does not con- verge in the non-linear solving alogrithm. The author reports that the behavior is similar to the one already reported in section 5.4 —ﬁxing the value of Ψ = 1 3 made it impossible to obtain convergence in the non-linear Newton-Raphson algorithm—. This result is conclusive to discard the use of equation (9.33) in the construction of the BEST element; as it produces an ill-posed problem. On the contrary, using equation (9.34) the BEST element converges optimally in the Newton-Raphson algorithm. The circular plate problem provides another opportunity to further char- acterize the BEST element. While for the linear problem the BEST el- ement converges to the correct result (albeit very slowly), when turning it into geometrically non-linear the author found that the BEST element wasn’t converging to the correct result. This result was shocking at ﬁrst. Certainly, the mechanisms of the linear and the non-linear problems are different. While for the linear problem the response is only in bending mode, for the non-linear problem the response is mostly in membrane mode. But an error of over 7% was completely unexpected. The portion of the deformation energy absorbed by the membrane mode becomes larger as the thickness becomes small. Futuros trabajos y propuestas de mejora A NY RESEARCH WORK departs from one or many starting points and makes develop- ments based on that basis; and in most cases without reaching a closed result. Usually re- search efforts serve rather to open windows than to close doors, and I believe that this thesis is a good example. I am sincerely satisﬁed with my work, but I can’t refrain from expressing some annoyance because there is still job to do. The ex- ercise of starting a research work, exploring the knowledge, pursuing the endless frontier (borrow- ing the words from Chuck Vest [136]); is an in- trinsically bold and ambitious endeavor. As a con- sequence, as it usually happens, the author did not measure with precision the effort required to achieve the initial objectives; because it is impos- sible to evaluate the difﬁculty of solving the chal- lenges and overturn the adversities that such a discovery endeavor entails. A N T ODO TRABAJO DE INVESTIGACIÓN toma uno o va- rios puntos de partida y realiza unos desarrollos a partir de ellos; en la inmensa mayoría de casos sin llegar a un resultado acabado. Normalmente los tra- bajos de investigación sirven más para abrir puertas que para cerrarlas, y creo que esta tesis es un ejemplo de ello. Estoy sinceramente satisfecho con el trabajo que he hecho, pero no puedo dejar de manifestar una cierto desasosiego por aquello que queda por hacer. El ejercicio de emprender un trabajo de investigación, de exploración del conocimiento, de perseguir la fronte- ra inalcanzable (en palabras de Chuck Vest [136]); es un ejercicio inherentemente atrevido y ambicioso. Es natural por consiguiente que el autor no midiera con precisión el esfuerzo requerido para cumplir los obje- tivos iniciales planteados, porque era imposible medir la diﬁcultad de resolver los desafíos y sortear los obs- táculos que se encontrarían en ese ejercicio de descu- brimiento. T This chapter lists those aspects of the re- search that are affected because eventually we all have to call it a day in order to move on. The the- sis includes considerations of varying degree to shine some light on the paths that are left to ex- plore. There is even two full chapters of the thesis devoted to expose clearly the opinion and recom- mendation of the author to improve the results reached by the thesis. 11.5 Summary This fact led the author to test the same problem in the non-linear regime for different thicknesses, ﬁnding a very 173 11.5. SUMMARY enlightening result. Figure 11.21 shows how the BEST element progres- sively becomes locked as the thickness of the plate becomes smaller. This result is consistent with the explanation published by Choi et al. [31]. p p y The last set of problems is the shell obstacle course. In this set of prob- lems the most remarcable result is obtained for the pinched cylinder prob- lem. For the pinched cylinder problem, the BEST element exhibits excel- lent performance, proving to be superior to all the other elements used for comparison except for the more sophisticated MITC9 and Q8H elements which are quadratic quadrangular elements with rotation degrees of free- dom. For the Scordelis-Lo roof problem and the hemispherical shell prob- lem the results obtained are consistent with the results already discussed for the slender beam, the circular plate and the thick beam tests. The advantage of using examples that test the speciﬁc deformation modes of the element is that it allows to better identify the defects of the element. Whereas the advantages of using the examples in the shell obstacle course is that it provides a consistent base to compare against other elements in the literature. 12.1 Optimize the programming The BEST element presented in this thesis has been programmed and implemented in a com- puter code. The implementation done by the au- thor makes an abuse of the tensors of many di- mensions (up to 5 dimensions) for the storage of variables. The code also makes an abuse of nested loops in order to process all those vari- ables. In reality most of the components of these tensors are zero. This abuse of tensors and nested loops generates a great inefﬁciency in the element’s system matrix build routine. There- fore, the current implementation of the code does many superﬂuous operations and is using a lot of unneeded memory. El código que ha programado el autor para implemen- tar el elemento BEST desarrollado en esta tesis hace un abuso de los tensores de varias dimensiones (hasta 5 dimensiones) para el almacenamiento de variables. Asimismo también abusa de los bucles anidados pa- ra procesar todas esas variables. En realidad muchos de los componentes de esos tensores son cero. Ese abu- so de tensores y de bucles anidados causa una gran ineﬁciencia en la rutina de construcción de la matriz del sistema del elemento. Puesto que con la implemen- tación actual del código se están realizando muchas operaciones superﬂuas y se está usando espacio de me- moria de manera innecesaria. Una evidente y necesaria mejora de la implemen- tación del código del elemento BEST sería la optimi- zación y uso de librerías especializadas para la ma- nipulación de tensores de múltiples dimensiones emi- nentemente sparse y los correspondientes algoritmos para operar con ellos. Sin dicha optimización no es ni siquiera planteable un análisis en profundidad de la eﬁciencia computacional del elemento BEST. It is necessary, indeed, to improve the imple- mentation of the programming code of the BEST element by optimizing and using specialized li- braries for the manipulation of tensors of many dimensions which are essentially sparse and the corresponding algorithms to operate with them. Without that optimization it is pointless to ana- lyze the computational efﬁciency of the BEST el- ement in detail. Futuros trabajos y propuestas de mejora Este capítulo enumera de manera sucinta aquellos aspectos de la tesis que por motivos va- rios se han quedado en el tintero. La tesis incluye re- ﬂexiones de distinta profundidad sobre la manera de abordar los caminos que quedan por transitar. Inclu- so dos capítulos enteros de la tesis están dedicados a detallar la resolución que el autor propone a estos re- tos. Sirvan pues los contenidos de este capítulo para exponer de manera clara cual es la opinión y la reco- mendación del autor para mejorar los resultados al- canzados en esta tesis. 175 CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH 176 Resolver el bloqueo por membrana Como se ha mostrado en el capítulo 9 el elemento BEST tal y como se ha deﬁnido cinemáticamente pa- dece de bloqueo por membrana. Este problema tam- bién se ha demostrado con más severidad en el ejem- plo de la placa circular (véase el apartado 11.3.2). Di- cho bloqueo impide que el elemento pueda converger cúbicamente en aquellos casos en los que exista defor- mación por cortante en el plano del elemento en algu- na medida, y en su lugar sólo exhibe convergencia li- neal —o peor—. Es altamente deseable poder corregir este comportamiento, pues ello terminaría de justiﬁcar la estrategia desarrollada en esta tesis. No obstante, se han alcanzado resultados parciales que corroboran que la estrategia desarrollada es acertada; si bien es muy deseable una evolución. El autor ha señalado en el capítulo 9 el defecto en la construcción de la cine- mática del elemento que causa este bloqueo. Y ha pro- puesto resolver ese defecto liberando los llamados gra- dos de libertad de taladro. El elemento BEST permite emular el efecto de estos grados de libertad de taladro, y por ello no es preciso añadirlos a la descripción del elemento. Lo que se precisa es adaptar la cinemática de la construcción del elemento para que éste pueda reproducir su efecto. Chapter 9 has shown that the BEST element suf- fers membrane locking because of the way it has been kinematically built. This problem has also been reproduced with more severity in the exam- ple of the circular plate (see section 11.3.2). Such locking inhibits the element from converging cu- bically wherever there may be any in-plane shear deformation, and instead exhibits only linear con- vergence —or worse—. It is highly desirable to ﬁx this behavior, as that would conclusively jus- tify the strategy developed in this thesis. Never- theless, the results obtained, albeit partial, con- ﬁrm that the strategy developed is correct; al- though an evolution is highly desirable. The au- thor has also set forth in chapter 9 the exist- ing defect in the element’s kinematic construc- tion which causes that locking behavior. And has come up with an idea to solve this issue liber- ating the so called drilling degrees of freedom. The BEST element allows to emulate the effect of these drilling degrees of freedom, that’s why it isn’t necessary to include them in the descrip- tion of the element. 12.3. SOLVE THE MEMBRANE LOCKING 177 again by all the elements that share that node. Los únicos datos del elemento que requieren de la información simultánea de los 3 nodos del elemento son los relativos al cálculo del punto de control cen- tral del elemento. Como la información para calcular el punto de control central toma como punto de parti- da la información de las tres aristas, podría ser venta- joso estudiar una formulación basada en las aristas. De hecho, recientemente varios autores han desarro- llado adaptaciones de elementos ﬁnitos de lámina a formulaciones basadas en las aristas aplicando técni- cas de suavizado [30, 69, 101]. The only values of the element requiring the simultaneous information from the 3 nodes of the element are those related to the central control point of the element. Since the information to compute the central control point requires infor- mation from the three edges, an edge-based for- mulation can be considered worth studying. In fact, different authors have performed adapta- tions of shell ﬁnite elements into an edge-based smoothed formulations recently [30, 69, 101]. 12.2 Nodal vs elemental assembly: Ensamblaje nodal vs ensamblaje elemental: edge-based assembly? ¿ensamblaje por aristas? En [96] Oñate y Zárate introducen la consideración de usar elementos de lámina sin rotaciones basados en los vértices (BSN). En su caso, la mejora de la preci- sión de la formulación basada en los vértices (sobre la formulación basada en los elementos) se debe al mayor número de conectividades logrado por la formulación basada en los vértices. In [96] Oñate and Zárate make a point for using vertex-based rotation-free shell elements (BSN). In their case, the precision improvement of the vertex-based formulation over the element-based formulation stems from the increased number of connectivities achieved by the vertex-based for- mulation. El elemento BEST realiza casi la totalidad de las operaciones tomando como referencia los nodos. De tal manera que en cada operación se puede estructurar la información conteniéndola en el nodo en lugar de en el elemento. Por consiguiente es oportuno plantear si una programación del elemento ensamblando por nodos puede ser más rápida y eﬁciente que una pro- gramación ensamblando por elementos. En la progra- mación ensamblando por nodos hay muchas posibili- dades de reducir la repetición de operaciones que en el ensamblaje por elementos se producen dado que las variables asociadas a cada nodo son recalculadas una y otra vez por todos los elementos que lo comparten. The BEST element performs most of the cal- culations using the nodes as the reference to ac- cess the information. So much so, that for each calculation the information can be structured and contained associated to the node instead of the element. Therefore it is appropriate to consider whether programming the element using nodal assembly could be faster and more efﬁcient than using an elemental assembly algorithm. In a nodal assembly program there are lots of possi- bilities to reduce repeated operations that hap- pen in elemental assembly because the variables associated to each node are recomputed once and 12.3. SOLVE THE MEMBRANE LOCKING CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH 178 formation, and lacks sophistication. The author proposes to improve the description of the de- formation of the element associated to the kine- matics of the drilling rotations. This description of the deformation shall be consistent with the continuum mechanics theories in order to avoid adding an artiﬁcial stiffness to the element of any kind; which is the cause for the membrane lock- ing behavior in our case. las rotaciones de taladro. Esta descripción de la de- formación debería ser consistente con los postulados de la mecánica del continuo para evitar introducir en el elemento algún tipo de rigidez artiﬁciosa, que es lo que en deﬁnitiva está activando el bloqueo por mem- brana del elemento. En el apartado 9.5.2 el autor avanza un marco de trabajo que debe permitir esa descripción precisa de la deformación a partir de la cinemática de las rota- ciones de taladro. Dicho marco de trabajo establece que la cinemática del elemento debe contemplar que para cada arista del elemento, la rotación de taladro se descompone en un giro de sólido rígido y un giro de deformación de cortante en el plano. Y por otro la- do se explica como el tensor gradiente de la deforma- ción asociado al plano tangente en el vértice —donde se conceptualizan las rotaciones de taladro— también se descompone a su vez en una componente de rotación (de sólido rígido) y otra de deformación de cortante (en el plano). Y se propone una manera de relacionar la ci- nemática de las rotaciones de taladro con la descrip- ción de la deformación, derivando las correspondien- tes fórmulas. No obstante, el desarrollo completo de dichas fórmulas y la adecuación del resto de la cine- mática del elemento a lo que estas fórmulas implican es un trabajo que queda para el futuro. g In section 9.5.2 the author sets forth a frame- work that leads to a precise description of the deformation based on the drilling rotations kine- matics. That framework establishes that the el- ement’s edge kinematics shall account for a solid rigid rotation and a rotation associated to the in- plane shear deformation. And on the other side, it explains that the deformation gradient tensor associated to the tangent plane at the vertex — where the drilling rotations are conceptualized— is also decomposed into a (solid rigid) rotation tensor and a shear deformation tensor (in-plane). CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH And therefore proposes to relate the drilling ro- tations kinematics to the description of the defor- mation, by deriving the corresponding formulæ. However, the full development of those formulæ and the implications on the rest of the element’s kinematics is an adaptation left to be done in a future work. Resolver el bloqueo por membrana What the element needs is to modify the kinematics of its construction to be able to accurately reproduce their effects. El autor ha demostrado con éxito la manera de li- berar la cinemática de las rotaciones de taladro en el elemento. Y con ello muestra la mejora que experimen- ta la convergencia el elemento. Si bien, los resultados sólo muestran una mejora marginal, pues aún es nece- sario traducir la cinemática de las rotaciones de tala- dro en una descripción de la deformación del elemento precisa. La descripción que se ha implementado en el código programado por el autor es todavía muy primi- tiva. Por ello se propone mejorar la descripción de la deformación del elemento asociada a la cinemática de The author has successfully shown how to un- leash the kinematics of the drilling rotations in the element. And in doing so he shows that the el- ement improves its convergence properties. How- ever, the results only display a marginal improve- ment, because the implementation still does not translate the kinematics of the drilling rotations into a precise description of the deformation of the element. The author has implemented in the computer code a very basic description of the de- 12.4 Implement Dirichlet boundary conditions The BEST element presented in this thesis has been programmed and implemented in a com- puter code. The implementation done by the au- thor lacks the complementary routines to apply the Dirichlet boundary conditions on the models. In order to extend the use of the BEST element to a wide range of academic and engineering prob- lems it is necessary to implement these boundary conditions (see chapter 10). Nevertheless, the ex- amples shown in this thesis allow to reach sufﬁ- ciently founded conclusions regarding the devel- opments of the thesis. El código que ha programado el autor para implemen- tar el elemento BEST desarrollado en esta tesis no in- cluye las rutinas necesarias para aplicar las condicio- nes de contorno de Dirichlet en los modelos. Si bien la implementación de dichas condiciones de contorno (véase el capítulo 10) es necesaria para poder exten- der el uso del elemento BEST a una amplia casuística de problemas académicos e ingenieriles, los ejemplos expuestos en esta tesis permiten alcanzar conclusio- nes suﬁcientemente fundamentadas en relación a los desarrollos de la tesis. De hecho, las conclusiones de esta tesis indican claramente la necesidad de mejorar el rendimiento del elemento BEST frente al fenómeno de bloqueo por membrana. Es la opinión del autor que carece de in- terés evaluar el elemento BEST frente a un conjunto de ejemplos más amplio que los presentados en esta Indeed, the conclusions of this thesis clearly express the need to improve the performance of the BEST element with respect to the membrane locking phenomenon. In the opinion of the au- thor, it is worthless evaluating the BEST element 12.6. TRY A DIFFERENT CURVATURE OPERATOR 179 tesis sin previamente mejorar el elemento BEST fren- te al fenómeno de bloqueo por membrana. Dicho es- to, es altamente deseable la inclusión en el código del programa que implementa el elemento BEST la posi- bilidad de aplicar las condiciones de contorno que se detallan en el capítulo 10 para entonces evaluar en to- da su extensión las capacidades del elemento BEST y comparar sus virtudes y defectos con otras formula- ciones existentes en la literatura usando una variedad más completa de los ejemplos académicos normalmen- te utilizados comúnmente. using a broader set of examples than those pre- sented in this thesis without previously improv- ing the behavior of the BEST element to pre- vent membrane locking. 12.5 Develop adequate representation techniques for the BEST element Hubiera sido muy deseable para el desarrollo del tra- bajo de esta tesis haber dispuesto de un sistema de re- presentación gráﬁca de la geometría del elemento de Bézier reconstruido. La única representación incluida en esta tesis se encuentra en la ﬁgura 5.10c y se ge- neró programando los elementos de esa malla uno a uno en una hoja de cálculo de Maple. Sin la posibi- lidad de visualizar la construcción geométrica de Bé- zier resultante, ha sido muy complicado para el autor diagnosticar los problemas que se iba encontrando. The development of this research would have beneﬁted from having a graphical representation system of the element’s Bézier-enhanced geome- try. The only such representation included in this thesis is in ﬁgure 5.10c. And it was generated af- ter programming the elements of that mesh one by one on a Maple sheet. Lacking the means to visualize the resulting Bézier constructions has been a handicap in diagnosing the problems en- countered. It is thus highly advisable —in any further de- velopment to improve the BEST element— to ar- range a geometric representation system which enables the visualization of the resulting Bézier geometric reconstruction. Es por lo tanto altamente recomendable que —en futuros desarrollos orientados a mejorar el elemento BEST— se cuente con un sistema de representación de la geometría que permita visualizar los resultados de la reconstrucción geométrica de Bézier. Este desarrollo tendría un efecto añadido beneﬁ- cioso para el post-proceso de los resultados. Ya que ac- tualmente los resultados se están representando sobre la malla original de triángulos lineales. Esto supone una evidente pérdida de resolución de los resultados realmente obtenidos, que tienen carácter cúbico. Such work would yield an additional advan- tage when post-processing the results. As the results could then be plotted on the geometric representation of the Bézier-enhanced elements. Currently the results are being represented on the original mesh of linear triangles. This im- plies an obvious loss of resolution of the results actually computed, which are of cubic nature. 12.4 Implement Dirichlet boundary conditions Nonetheless, it is highly desirable to include in the code of the program that implements the BEST element the feature to apply those boundary conditions described in chapter 10 in order to benchmark the BEST ele- ment capabilities thoroughly and compare them to other existing formulations in the literature and using a wider variety of benchmarking ex- amples commonly used. • métodos de estimación del tensor de curvatura. • métodos de estimación del tensor de curvatura. En el artículo se consideran métodos que usan los no- dos del primer anillo de vecinos, del segundo anillo de vecinos o incluso de hasta el tercer anillo de vecinos. The paper includes methods that use the nodes of the ﬁrst ring of neighboring nodes, the sec- ond ring and up to the third ring of neighboring nodes. En esta tesis no se ha explorado la posibilidad de obtener la métrica de la curvatura en el elemento a partir de cualquiera de estos métodos. Puede resultar interesante analizar la idoneidad del uso de cualquie- ra de estos métodos y comparar tanto su precisión y coste computacional con la precisión y coste compu- tacional del método de cálculo de la curvatura del ele- mento desarrollado en esta tesis. En cualquier caso, se pueden hacer algunas consideraciones apriorísticas. Los métodos basados en la interpolación de superﬁ- cies de creciente orden polinómico padecen de varios defectos. Entre ellos, la sensibilidad a las distorsio- nes de malla debido al mal condicionamiento de la matriz del sistema que es preciso resolver para obte- ner las coeﬁcientes de la superﬁcie interpolante y por consiguiente también el elevado coste computacional de la resolución de esa matriz por métodos iterativos. Los métodos discretos son los más eﬁcientes desde el punto de vista computacional, pero ofrecen resultados poco útiles para el cálculo de láminas porque no ofre- cen la información completa del tensor de curvatura; lo que conduciría a una pérdida signiﬁcativa de pre- cisión en el cálculo del elemento. Los métodos de es- timación del tensor de curvatura son los que pueden ofrecer algún interés práctico. Aunque los resultados de Gatzke y Grimm no son muy halagüeños en lo que se reﬁere a la sensibilidad de estos métodos respecto de la regularidad de la malla. Dado que la robustez del elemento desarrollado con respecto a la irregularidad de la malla era un aspecto que se buscaba explícita- mente en esta tesis, esta es una línea de investigación que no se ha explorado. This thesis hasn’t explored the possibility of ﬁnding the metric of the curvature of the element based on any of these methods. It is very ap- pealing to analyze the convenience of using any of these methods and compare both their preci- sion and their computational cost with those of the ﬁnite element developed in this thesis. 12.6 Try a different curvature operator During the research work to ﬁnd an optimal op- erator to approximate the normal direction at the nodes of the mesh (see section 5.1 [129]), an in- teresting reference on the approximation of cur- vatures in triangle meshes arose [48]. That pa- per evaluates the relative precision and compu- Durante los trabajos de obtención de un operador óp- timo para aproximar la dirección de la normal en los nodos de la malla (ver apartado 5.1 [129]), apareció una referencia interesante sobre la aproximación de curvaturas en mallas de triángulos [48]. En dicho ar- tículo se evalúa la relativa precisión y coste compu- CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH 180 tational cost of different approximation methods of triangle meshes’ curvature. This is a funda- mental operator in shell analysis, and has been a central aspect of the development of the present thesis. Gatzke and Grimm distinguish 3 different families of operators: tacional de diversos métodos de aproximación de la curvatura de mallas de triángulos. Este es un opera- dor fundamental en el cálculo de láminas y ha sido un aspecto central del desarrollo de esta tesis. De acuerdo con Gatzke y Grimm, se pueden distinguir 3 familias de operadores: • methods that interpolate a surface to the nodes of the mesh, • métodos de interpolación de superﬁcies a los no- dos de la malla, • discrete methods (which attempt to obtain partial metrics related to the curvature based on different hypotheses), and • métodos discretos (que pretenden obtener métri- cas parciales relacionadas con la curvatura a partir de distintas hipótesis), y • methods of estimation of the curvature ten- sor. 12.7. WHAT ABOUT COMPOSITE AND LAYERED SHELLS? 181 Podría resultar interesante explorar la posibilidad de desarrollar un elemento ﬁnito de lámina delgada sin rotaciones a partir del uso de un operador de apro- ximación de la curvatura en los nodos de la malla y evaluar si ello implica ventajas computacionales. En ese sentido se tendría que realizar un estudio compa- rativo sobre la precisión en el cálculo de la curvatu- ra entre los métodos basados en operadores como los indicados en el artículo citado [48] y el método desa- rrollado en la presente tesis. Aunque desde el punto de vista del autor, la especialización del método desarro- llado en la presente tesis debería implicar una venta- ja en cuanto a la relación precisión/coste computacio- nal. El método desarrollado en la presente tesis usa un anillo de vecinos (para cada nodo del triángulo). Por consiguiente debería ser más eﬁciente computacional- mente que los métodos que usan 2 anillos de vecinos. Y la especialización del tratamiento de los datos debe- ría dar resultados mejores que los métodos que usan 2 o incluso hasta el tercer anillo de vecinos. Pero sería interesante dilucidar todas estas conjeturas. It could be of some interest to explore the pos- sibility of developing a rotation-free thin shell ﬁnite element based on the use of a curva- ture approximation operator at the nodes of the mesh, and evaluate whether it entails any computational advantage. A comparative study on the precision in the calculation of the cur- vature between the methods based on opera- tors as those cited in the paper [48] and the method developed in the present thesis ought to be done. However, from the author’s stand- point, the level of specialization achieved in the present thesis should provide an edge regard- ing the precision/computational cost relationship. The method used in this thesis uses 1 ring of neighboring nodes (for every node of the triangle). As a result, it should be more efﬁcient than the methods that use 2 rings of neighboring nodes. And the specialization of the treatment of data should provide better results than the methods using 2 or even a third ring of neighbors. But dispelling these guesses could be an interesting exercise. • métodos de estimación del tensor de curvatura. Nev- ertheless, the author makes some aprioristic con- siderations. The methods based on the interpo- lation of surfaces of increasing polynomial order suffer from various defects. Amongst them, their sensibility to mesh distortions caused by the ill- conditioning of the system matrix that needs to be solved in order to obtain the coefﬁcients of the interpolating surface. As a result, the computa- tional cost of solving that matrix is affected if it- erative methods are used. Discrete methods are the most efﬁcient from the point of view of their computational cost, but they give unusable re- sults for shell analysis, because they do not pro- vide the complete information of the curvature tensor; which would cause a signiﬁcant loss of precision in the calculation of the element. The methods that estimate the curvature tensor offer some practical interest. Although the results pro- vided by Gatzke and Grimm aren’t promising on what relates to their sensibility to mesh regular- ity. As element robustness with respect to mesh distortion and irregularity was an aspect speciﬁ- cally sought by this thesis development, this is a research topic that has not been explored. 12.7. WHAT ABOUT COMPOSITE AND LAYERED SHELLS? Continuum-based formulation La preocupación por abordar el asunto de las láminas hechas de materiales compuestos y laminadas es uno de los motivos para escoger una formulación basada en el continuo (también llamada formulación de só- lido degenerado) en vez de un planteamiento basado en la teoría de láminas. Otro motivo muy importante para la esta elección es la ﬁrme base matemática que proporciona la formulación basada en el continuo y que deﬁne el estado de tensión y deformación en ca- da punto material de la lámina con precisión. Esto es gracias a que maneja con naturalidad todos los tér- minos de alto orden. En cambio, en un elemento de lá- mina basado en resultantes de esfuerzos, o en un mo- delo de teoría de láminas, normalmente los términos de alto orden son más complicados de manipular— particularmente en el régimen no-lineal—y a menudo se obvian. The concern for addressing the issue of composite and layered shells is one of the reasons to opt for a continuum-based approach (also called degen- erated solid approach) instead of a shell theory approach. Besides providing a sound mathemati- cal foundation to the formulation, the continuum- based model deﬁnes the stress and strain state of each material point in the shell with precision. This is because, it handles with ease all the high order terms. Instead, in a resultant based shell element, or in a shell theory model, usually the high order terms are more difﬁcult to deal with —particularly in the non-linear regime— and are often dropped. Additionally, it is the view of the author, that for engineering purposes, in the case of layered shells made of anisotropic materials it is much more interesting to provide the analyst with in- formation on the stress and strain states of each material layer, rather than providing him with generalized (integrated through the thickness) measures of strains and stresses. Even more so, when in the non-linear regime, the higher order terms might become relevant and their corresponding generalized measures (called bi- moments) don’t have an easy interpretation to the engineer [13]. Furthermore, if required, these generalized measures can be computed as a post- process. ¿Cómo se resuelven las láminas de materiales compuestos y laminadas? One of the requirements of the new element was the possibility to solve for composite and layered shells. However, it has been decided not to in- clude this development in the scope of this the- sis. Certainly the use of a rotation-free element affects the possibilities of simulating a compos- ite structure. In his book [93] Oñate explains in detail the reasons why composite beams/shells even if slender, need to account for shear defor- mations. Without the rotation degrees of freedom it is harder to reproduce the Reissner-Mindlin hy- potheses, which are better suited to account for shear deformations than the Kirchhoff-Love hy- potheses. Nevertheless, for the case of layered shells, there are other theories like the reﬁned zig-zag theory which can solve this issue [124, 126], or hierarchical split of the displacements method which has been applied successfully but has not yet been extended to geometrically non- linear problems [87]. Despite the fact that this thesis’ scope doesn’t include ﬁnally the consider- ation for composite and layered shells, this re- Uno de los requisitos para el nuevo elemento era la po- sibilidad de resolver láminas de materiales compues- tos y laminadas. Sin embargo, se ha optado por no incluir este desarrollo en la tesis. Ciertamente el uso de un elemento sin rotaciones afecta las posibilidades de simular una estructura de materiales compuestos. En este libro [93] Oñate explica en detalle las razones por las que incluso las vigas y láminas de materiales compuestos delgadas necesitan contemplar las defor- maciones de cortante. Sin los grados de libertad sin rotaciones es más complicado reproducir las hipótesis de Reissner-Mindlin, que son más indicadas para te- ner en cuenta las deformaciones de cortante que las hipótesis de Kirchhoff-Love. No obstante, para el ca- so de láminas laminadas existen otras teorías como la del zig-zag reﬁnado que permite resolver este aspecto [124, 126], o el método del desacoplamiento jerárquico de los desplazamientos que ha sido aplicado exitosa- mente aunque no se ha extendido todavía a problemas geométricamente no-lineales [87]. A pesar de que ﬁ- nalmente el alcance de la tesis no incluye la conside- ración de láminas de materiales compuestas y lami- CHAPTER 12. FUTURE WORK AND FURTHER RESEARCH 182 quirement has been taken into account so that it doesn’t hamper future efforts to implement this capability. The reason is explained below. Continuum-based formulation Por otro lado, es la opinión del autor que para pro- pósitos ingenieriles, en el caso de láminas laminadas formadas por materiales anisótropos es mucho más adecuado proveer al calculista la información de los estados de tensión y deformación de cada capa de ma- terial; en lugar de proveer al calculista los esfuerzos generalizados (integrados en el espesor) de tensiones y deformaciones. Por añadidura, en el régimen no-lineal los términos de alto orden pueden resultan relevantes y sus correspondientes métricas generalizadas (llama- das bi-momentos) no tienen una interpretación senci- lla para el ingeniero [13]. Finalmente, si es preciso, estas métricas generalizadas se pueden calcular como un post-proceso. ¿Cómo se resuelven las láminas de materiales compuestos y laminadas? nadas, se ha tenido en cuenta este requisito para que no obstaculice futuros desarrollos para implementar esta capacidad. La razón se explica a continuación. Aplicaciones prácticas A la vista de todos los puntos incluidos previamente en este capítulo, y teniendo en cuenta la motivación expuesta en el capítulo 1, resulta evidente que me he quedado lejos de las metas ambicionadas inicialmen- te. Y particularmente en lo que respecta a la justiﬁca- ción de la selección del objetivo de esta tesis; explicada en el apartado 2.3: In view of all the items included previously in this chapter, and taking into account the moti- vation set forth in chapter 1, it appears clear that I fell short of the expectations coveted initially. And particularly on what relates the justiﬁcation for the selection of the objective of this thesis; ex- plained in section 2.3: “Firstly, because there’s a need to analyze sails as shells under non-linear assumptions of large strains/displacements. And secondly, be- “En primer lugar, por la necesidad de analizar las velas como láminas en condiciones de no-linealidad y asumiendo grandes deformaciones/desplazamientos. 12.8. PRACTICAL APPLICATIONS 183 Y en segundo lugar, porque el uso de esta tecnología se puede aplicar posteriormente al análisis general de la estructura del barco de vela.” Y en segundo lugar, porque el uso de esta tecnología se puede aplicar posteriormente al análisis general de la estructura del barco de vela.” cause the use of this technology can subsequently be used broadly in the structural analysis of the sailboat.” En esta tesis no he llegado a realizar un caso prác- tico de análisis de una vela. Y tampoco he aplicado el elemento BEST al análisis general de la estructura de un barco de vela. Para poder poner en práctica el elemento BEST, es necesario resolver los retos plan- teados en los apartados 12.1 a 12.7. Una vez se ha- yan cumplido esos hitos, es preciso aplicar el elemento BEST al cálculo práctico de un barco de vela. In this thesis I haven’t been able to perform a sail’s practical analysis. Nor have I applied the BEST element to the general analysis of a sail- boat’s structure. In order to put the BEST ele- ment into practice, the challenges set forth in sec- tions 12.1 to 12.7 ought to be solved. Once these milestones are accomplished, the BEST element shall be applied to the practical case of a sailboat analysis. Para ello, en primer lugar será preciso hacer que el elemento BEST esté acoplado con un código de resolu- ción de ﬂuidos. Aplicaciones prácticas Esto debería ser relativamente sencillo pues, como ya se ha explicado en el apartado 10.9, el elemento BEST se ha integrado en el programa Ram- Series que a su vez pertenece a la suite Tdyn. Tdyn es una excelente suite de solvers multifísica que ya tiene bien resuelta la interacción ﬂuido-estructura e incluye un solver de dinámica de ﬂuidos de la máxima solven- cia (como lo pone de maniﬁesto la referencia [28]). In order to do it, ﬁrst the BEST element has to be coupled with a CFD code. This should be rel- atively simple since, as I have explained in sec- tion 10.9, the BEST element has been integrated in the RamSeries program; which in turn belongs to the Tdyn suite. Tdyn is an excellent multi- physics solvers suite that has sorted out ﬂuid- structure interaction and includes a top notch CFD solver (as credited by the reference [28]). Y en segundo lugar habrá que deﬁnir una serie de tests de validación del funcionamiento del elemento BEST. Es decir, para el análisis de una vela hay que elegir un caso de benchmark que incluya: And secondly a set of tests have to be deﬁned to validate the BEST element. Id est, for the analysis of a sail, a benchmark case ought to be chosen including: 1. la deﬁnición geométrica del diseño del barco, 1. la deﬁnición geométrica del diseño del barco, 1. the geometric deﬁnition of the boat’s design, 2. the experiment’s conditions, and 2. las condiciones de ensayo, y 1. la deﬁnición geométrica del diseño de una vela, 1. la deﬁnición geométrica del diseño de una vela, 1. the geometric deﬁnition of the sail’s design, 1. the geometric deﬁnition of the sail’s design, 2. las condiciones de contorno y de ﬂujo de viento en las que se analiza, y 2. the boundary and wind ﬂow conditions for the analysis, and 3. the geometric deﬁnition of the deformed ﬂy- ing shape; 3. la deﬁnición geométrica de su forma deformada en vuelo; para así comparar los resultados del análisis del ele- mento BEST con los del ensayo. Y para el análisis de un barco de vela, hay que elegir un caso de benchmark que incluya: in order to compare the results of the BEST ele- ment analysis with those of the experiment. And for the analysis of a sailboat, another benchmark case ought to be chosen including: in order to compare the results of the BEST ele- ment analysis with those of the experiment. And for the analysis of a sailboat, another benchmark case ought to be chosen including: 3. los resultados del ensayo. 3. the experiment’s results. 3. los resultados del ensayo. Con toda probabilidad, para poder reproducir ade- cuadamente el benchmark de una vela, será necesario combinar el elemento BEST con otros elementos es- tructurales de biga y de cables, así como añadir la ca- pacidad al programa de pretensar convenientemente la estructura de la lámina. In all likelihood, in order to adequately repro- duce the sail’s benchmark, it will be necessary to combine the BEST element with other structural cable and beam elements, as well as adding to the program the capability to prestress appropriately the shell structure. Conclusiones L A ELABORACIÓN DE ESTA TESIS ha supuesto un camino de aprendizaje. Seguramente con todo lo aprendido haría muchas cosas de manera distinta si tuviera que volver a hacerlas. Pero precisamente en eso consiste también el trabajo de investigación de un doctorado. Un doctorado no consiste únicamen- te en descubrir o en generar nuevo conocimiento, que también. Además es un camino de aprendizaje (y re- aprendizaje) de conocimiento ya existente. He tratado y trataré de reﬂejar en las siguientes líneas en qué ha consistido ese aprendizaje en sus aspectos más desta- cables así como señalar cuáles son las aportaciones de esta tesis. L A T HE WORK PRESENTED IN THIS THESIS has implied a learning path. After the lessons learned I would most certainly do things differ- ently if I were to redo them again. And that’s precisely the purpose of the research work in a doctorate. A doctorate consists on generating new knowledge and ﬁnding new discoveries, but not only. It is also a path for learning (and re-learning) pre-existing knowledge. I will try to reﬂect on the following lines what are the most notable aspects of that learning path and signal the main contributions of this thesis. T CHAPTER 13. CONCLUSIONS 186 la dependencia de la precisión de los elementos de lá- mina sin rotaciones con respecto a la distorsión de la malla. En [49] Gärdsback y Tibert explican que los elementos de lámina sin rotaciones que calculan la curvatura como un tensor 2D obtienen mayor ro- bustez respecto a mallas irregulares que los elementos de lámina sin rotaciones que contemplan la curvatu- ra de la superﬁcie como una superposición de curva- turas unidimensionales. En este sentido, el elemento BEST adopta el planteamiento de deﬁnir la curvatu- ra como un tensor 2D (véase la ecuación (4.6) donde se muestra claramente este planteamiento; y la ecua- ción (6.65) donde se describe la pseudo-deformación asociada a la ﬂexión en notación de Voigt aprovechan- do la simetría del tensor). The BEST element tackles the problem of rotation-free shell elements’ precision depen- dency with respect to mesh distortion. Gärdsback and Tibert explain in [49] that those rotation- free shell elements which compute curvature as a 2D tensor, achieve better consistency in their accuracy with respect to irregular meshes than rotation-free shell elements which consider the curvature of the surface as a superposition of unidimensional curvatures. Accordingly, the BEST element adopts the approach of deﬁning the curvature as a 2D tensor (see equation (4.6) where this approach is clearly shown; and equa- tion (6.65) where the pseudo-deformation associ- ated to bending is represented in Voigt notation taking advantage of the symmetry of the tensor). Además, existe otro aspecto que mejora la precisión de los elementos de lámina sin rotaciones, en lo que respecta al cálculo de la curvatura, en relación con la malla usada. En [96] Oñate y Zárate explican que el elemento BSN alcanza mejor precisión que el elemen- to BST. La diferencia entre la implementación nodal (BSN) y la implementación elemental (BST) consis- te en usar todas las conectividades alrededor de cada nodo en el primer caso, o solamente los triángulos ad- yacentes en el segundo caso. Cirak, Ortiz et al. [24-26] proponen el paradigma de las superﬁcies subdividi- das mediante el que también aprovechan todas las co- nectividades de los nodos del triángulo situado en una parcela regular. El elemento BEST hace lo propio, y también aprovecha todas las conectividades de los no- dos de cada triángulo de la malla (ver la ﬁgura 4.2). Un nuevo elemento ﬁnito de lámina delgada There are lots of shell ﬁnite elements in the sub- ject’s literature. To review them all is a task for the very few. That’s why it is important to under- score the references cited in the last paragraph of section 3.1 and the references therein which, albeit they do not encompass all the different el- ements that exist, they represent a genuine ef- fort to include comprehensively the ensemble of the developments on shell ﬁnite elements. This diversity of shell elements showcases the inher- ent difﬁculty of ﬁnding an overall solution to the problem. The BEST element doesn’t achieve it ei- ther, but constitutes another element in the fam- ily of shell ﬁnite elements and, in particular, in the family of rotation-free thin shell ﬁnite ele- ments, solving some of their shortcomings. Existen muchos elementos ﬁnitos de lámina en la lite- ratura de la materia. Es una tarea al alcance de muy pocas personas el revisarlos todos. Por ello es preciso destacar las referencias citadas en el último párrafo del apartado 3.1 y las referencias contenidas en ellos, que aunque no los revisan absolutamente todos, sí ha- cen un esfuerzo por abarcar de manera exhaustiva el conjunto de trabajos para desarrollar elementos ﬁni- tos de lámina. Esta diversidad de elementos pone de maniﬁesto la diﬁcultad de alcanzar una solución que cierre el problema. El elemento BEST tampoco lo con- sigue, pero representa un nuevo elemento en la familia de los elementos ﬁnitos de lámina, y en particular en la de los elementos ﬁnitos de lámina delgada sin rota- ciones que resuelve algunas de sus deﬁciencias. El elemento BEST trata de resolver el problema de 185 CHAPTER 13. CONCLUSIONS Estimación óptima de la normal a una superﬁcie a partir de una malla de triángulos Un aspecto clave del desarrollo del elemento BEST es la estimación precisa de las normales a la superﬁcie en los nodos. Al ser un paso intermedio del cálculo, me preocupaba especialmente que cualquier impreci- sión se propagase a la subsiguiente construcción del elemento. Por consiguiente, he querido asegurarme de minimizar el error incurrido al condensar la informa- ción de los triángulos que rodean cada nodo, para es- timar el vector normal a la superﬁcie en el nodo. A key development aspect of the BEST element is the precise approximation of the surface nor- mals at the nodes. Being an intermediate step in the calculations, I was particularly worried about carrying errors on to the construction of the ele- ment. Therefore, I wanted to make sure I could minimize the error incurred in the estimation of the surface normal at the node, when condensing the information of the triangles surrounding each node. p ﬁ Por ese motivo he desarrollado un marco teórico para reinterpretar el signiﬁcado de los vectores nor- males de cada triángulo de una malla de triángulos. En este marco teórico, he considerado que una cuádri- ca es una aproximación de segundo orden a una su- perﬁcie suave. Y que los triángulos de la malla repre- sentan secciones planas de dicha superﬁcie cuádrica. Por consiguiente, el vector normal de cada triángulo de la malla es representativo de un vector normal a la superﬁcie, en la medida que la sección representada por el triángulo es representativa del punto de la su- perﬁcie que se quiere aproximar. Junto con Estruch y García-Espinosa, en [129] he desarrollado dicho mar- co teórico y he justiﬁcado el uso de distintas medidas para ponderar las normales de los triángulos de la malla, para estimar la dirección de la normal en cada nodo de la malla. For this reason I have created a framework to reinterpret each triangle’s normal vector mean- ing, in a mesh of triangles. In this framework I have considered that a quadric is a second order approximation to a smooth surface. And that the triangles of the mesh represent planar sections of that quadric surface. Therefore, each triangle’s normal vector represents a normal vector of the surface to the same extent as the section of the quadric represented by the triangle represents the point of interest in the surface. CHAPTER 13. CONCLUSIONS Para evitar la restricción de usar mallas regulares de triángulos, en el elemento BEST se realiza el paso in- termedio de calcular los vectores normales en los no- dos, condensando así la información de los triángulos que rodean cada nodo. g g y y There is another aspect that improves the pre- cision of the rotation-free shell elements, regard- ing the computation of curvature, with respect to the mesh used. Oñate and Zárate explain in [96] that the BSN element achieves better accuracy than the BST element. The difference between the nodal implementation (BSN) and the elemen- tal implementation (BST) consists in using all the connectivities surrounding each node for the for- mer, or just the adjacent triangles for the latter. Cirak, Ortiz et al. [24–26] propose the subdivi- sion surfaces paradigm to also take advantage of all the connectivities of the nodes of the triangle situated in a regular patch. Likewise, the BEST element takes advantage of all the nodes’ con- nectivities in each triangle of the mesh (see ﬁg- ure 4.2). In order to avoid the restriction of hav- ing to use regular meshes of triangles, the BEST element performs an intermediate step calculat- ing the normal vectors at the nodes, and thus condensing the information of the triangles sur- rounding each node. Por lo tanto, BEST es un nuevo elemento ﬁnito de lámina delgada sin rotaciones desarrollado como un elemento basado en el continuo, y que utiliza una for- mulación Lagrangiana total para la resolución del problema no-lineal. El elemento BEST resuelve co- rrectamente problemas de lámina delgada bajo las hi- pótesis de Kirchhoff-Love, aunque adolece de bloqueo por membrana que se acentúa a medida que la lámi- na se hace más y más delgada (ver capítulo 11). El nombre del elemento BEST resulta del acrónimo de lámina triangular mejorada con Bézier (en inglés). BEST is therefore a new rotation-free thin shell ﬁnite element developed as a continuum- based ﬁnite element, which uses a total La- grangian formulation for the resolution of the non-linear problem. The BEST element solves correctly thin shell problems under the Kirchhoff- Love hypothesis, although it suffers from mem- brane locking, which becomes more severe as the shell becomes thinner (see chapter 11). The name of the BEST element is obtained as the acronym of Bézier-Enhanced Shell Triangle. 13.1. CONTRIBUTIONS 187 13.1.3 New paradigm to reconstruct a cubic shell element using the neighboring elements El elemento BEST aprovecha la información sobre la geometría de la superﬁcie facilitada por los triángu- los que rodean cada uno de los triángulos de la ma- lla. Con esta información genera geometrías cúbicas a partir de una malla de triángulos planos. Para conse- guirlo, se calcula una aproximación del vector normal a la superﬁcie en cada uno de los nodos del triángulo a partir de la información de los triángulos que ro- dean cada nodo. El cálculo de los vectores normales en los nodos no depende del número de triángulos que rodean cada nodo de la malla. De esta manera el ele- mento BEST es independiente de la topología de la malla, a diferencia del paradigma de las superﬁcies subdivididas desarrollado por Cirak, Ortiz et al. The BEST element takes advantage of the sur- face’s geometric information provided by the tri- angles around each of the mesh nodes. With this information it builds cubic geometries starting from a mesh of ﬂat triangles. To accomplish it, approximated surface normals are computed at each node of the triangle using the information of the neighboring triangles. The computation of the normal vectors at the nodes doesn’t depend on the number of triangles surrounding each node of the mesh. Unlike the subdivision surfaces paradigm by Cirak, Ortiz et al., the BEST ele- ment is independent from the mesh topology. Using the information of the normal vectors at the nodes, I have developed a new paradigm consisting on reconstructing the geometry of a cu- bic triangular element exploiting the properties of cubic B-spline functions (cubic Bézier trian- gle). This way, I build a conforming shell ﬁnite element which overcomes the unsatisfactory re- sult presented in section 4.3. This approach is an original contribution of this thesis. A partir de la información de los vectores normales en los nodos he desarrollado un nuevo paradigma que consiste en reconstruir la geometría de un elemento triangular cúbico usando funciones cúbicas B-spline (triángulo cúbico de Bézier). De esta manera constru- yo un elemento ﬁnito de lámina que es conforme y que sortea el resultado insatisfactorio presentado en el apartado 4.3. Este planteamiento es una contribu- ción original de la presente tesis. The construction of the cubic Bézier triangle requires ﬁxing 30 parameters. Therefore it needs to apply 30 independent conditions. Estimación óptima de la normal a una superﬁcie a partir de una malla de triángulos Along with Estruch and García-Espinosa, in [129] I have de- veloped this framework and I have justiﬁed the use of different metrics as averaging weights of the triangles’ normal vectors, for the estimation of the surface’s normal direction at each node of the mesh. Ubach, Estruch y García-Espinosa realizaron una comparación estadística exhaustiva de distintos fac- tores de ponderación, incluyendo otros usados previa- mente en la literatura. La conclusión de dicho trabajo conduce a usar como factor de ponderación para cal- cular una media de los vectores normales de los trián- gulos: el inverso del área de la circunferencia circuns- crita al triángulo y el ángulo interno del triángulo en el nodo considerado (ver ecuaciones (5.2) y (5.3) en el apartado 5.1). Este resultado es una aportación origi- nal del artículo citado, pero forma parte integral del trabajo de investigación de esta tesis. Usando este nue- vo factor de ponderación, se reduce en aproximada- mente un 10 % el error medio cuadrático cometido en la estimación de las normales de superﬁcies generadas aleatoriamente, respecto del mejor factor de pondera- ción usado previamente en la literatura y propuesto por Max [77] (ver [129, Tabla I]). Ubach, Estruch and García-Espinosa per- formed a comprehensive statistical comparison of different weighting factors including other weights previously used in the literature. The conclusion of that work is that the inverse of the area of the circumscribed circle to the trian- gle and the internal angle of the triangle at the node considered should be used as weighting fac- tor (see equations (5.2) and (5.3) in section 5.1). This result is an original contribution of the cited paper, but it is also an integral part of this thesis’ research. Using this new weighting factor, we re- duce by about 10% the root mean square error in the estimation of normals of randomly generated surfaces with respect to the previous best weight- ing factor found in the literature and proposed by Max [77] (see [129, Table I]). CHAPTER 13. CONCLUSIONS 188 CHAPTER 13. CONCLUSIONS 13.1. CONTRIBUTIONS 13.1. CONTRIBUTIONS 189 a (5.19). Cada una de las tres estimaciones aproxima- das está deﬁnida también extendiendo los conceptos de minimización de la energía de deformación toman- do como referencia el baricentro de los seis puntos de control intermedios del perímetro, y partiendo del ob- jetivo de que los elementos aproximen la continuidad C1 a lo largo de las aristas (ver la ﬁgura 5.6). a (5.19). Cada una de las tres estimaciones aproxima- das está deﬁnida también extendiendo los conceptos de minimización de la energía de deformación toman- do como referencia el baricentro de los seis puntos de control intermedios del perímetro, y partiendo del ob- jetivo de que los elementos aproximen la continuidad C1 a lo largo de las aristas (ver la ﬁgura 5.6). energy by taking the barycenter of the six inter- mediate control points as a reference, and using the goal of approaching C1 continuity across the elements along the edges (see ﬁgure 5.6). The 6 remaining conditions are obtained from the drilling rotations for each of the edges that converge on each node (see section 9.2). The re- sults obtained in this thesis ﬁnally lead to estab- lish the value of the drilling rotation as 0 in equa- tion (9.34). But I propose as future work the mod- iﬁcation of the element to establish drilling ro- tations different than 0 based on the description of the in-plane shear deformation —on the plane tangent to the surface at each node— consistently with the theory of deformation of the continuum (see section 9.5.2). Las 6 condiciones restantes las obtengo de las ro- taciones de taladro para cada una de las aristas que convergen en cada nodo (ver el apartado 9.2). Los re- sultados obtenidos en esta tesis ﬁnalmente establecen el valor de las rotaciones de taladro en 0 en la ecua- ción (9.34). Pero propongo para un trabajo futuro la modiﬁcación del elemento para poder establecer rota- ciones de taladro distintas de 0 basadas en la descrip- ción de la deformación de cortante en el plano —según el plano tangente a la superﬁcie en cada nodo— de manera consistente con la teoría de la deformación del continuo (véase el apartado 9.5.2). Built in this manner, the only variables of the BEST element are the vertices positions (9 vari- ables). And it solves internally the remaining pa- rameters corresponding to the geometric deﬁni- tion (21 parameters). 13.1. CONTRIBUTIONS 13 of these parameters are determined based on geometric considerations, strictly speaking: 6 out of the normal orienta- tions at the vertices, 6 out of the drilling rota- tions at the vertices and 1 out of the central con- trol point elevation with respect to the barycen- ter. Whereas the other 8 parameters are deter- mined based on mechanic energy minimization considerations: 6 Ψ values and the 2 coordinates which deﬁne the position of the barycenter. In or- der to ﬁx the values of these 21 internal parame- ters, each BEST element needs to solve 9 systems of linear equations of rank 3. p El elemento BEST construido de esta manera mantiene únicamente como incógnitas las posiciones de los vértices (9 variables). Y resuelve internamente los parámetros correspondientes al resto de la deﬁni- ción geométrica (21 parámetros). 13 de estos paráme- tros quedan determinados en base a consideraciones estrictamente geométricas: 6 de las orientaciones de las normales en los vértices, 6 de las rotaciones de ta- ladro en los vértices y 1 de la elevación del punto de control central respecto del baricentro. Mientras que los 8 parámetros restantes se determinan a partir de consideraciones de minimización de la energía mecá- nica del elemento: 6 valores de Ψ y las 2 coordenadas que deﬁnen la posición del baricentro. La determina- ción de estos 21 parámetros internos del elemento im- plica la resolución de 9 sistemas de ecuaciones lineales de rango 3 para cada elemento BEST. Taking all this into account, the BEST ele- ment can also be understood as a reduced order cubic shell element; for which the 30 variables of the cubic element have been condensed into just 9. This represents a new paradigm in which the model is not order-reduced as a whole, but through the order-reduction of each of the model’s elements. Por todo ello también se puede reinterpretar el ele- mento BEST como un elemento de lámina cúbico de orden reducido, en el que se han condensado las 30 variables del elemento cúbico en únicamente 9. Esto representa un nuevo paradigma en el que no se reduce el orden del modelo en su conjunto sino a través de la reducción de cada uno de los elementos del modelo. 13.1.3 New paradigm to reconstruct a cubic shell element using the neighboring elements 15 of these conditions are given directly by the positions of the 3 vertices of the triangle and the orientations of the normal vectors at the 3 vertices (see ﬁg- ure 5.5). These 15 conditions ensure that the tri- angles maintain C1 continuity at the vertices. La construcción del triángulo cúbico de Bézier re- quiere determinar 30 parámetros. Para ello es necesa- rio aplicar 30 condiciones independientes. 15 de estas condiciones se deducen de manera directa de la posi- ción de los 3 vértices del triángulo y de las orientacio- nes de los vectores normales en los 3 vértices (ver la ﬁgura 5.5). Con estas 15 condiciones se asegura que los triángulos mantienen continuidad C1 en los vérti- ces. 6 of the remaining 15 conditions are obtained using criteria of internal energy minimization of the element. I have deduced these internal en- ergy minimization criteria of the element by solv- ing reduced problems on the geometric construc- tion of the Bézier triangle (see the sections 5.4, 9.3 and 9.4). De las otras 15 condiciones que quedan por impo- ner, 6 de ellas las obtengo a partir de criterios de mi- nimización de la energía interna de deformación del elemento. Estos criterios de minimización de la ener- gía interna del elemento los deduzco a base de solucio- nar problemas reducidos de la construcción geométri- ca del triángulo de Bézier (ver los apartados 5.4, 9.3 y 9.4). In order to avoid the deﬁnition of an ill- conditioned system of equations (or even incom- patible) for ﬁnding the position of the element’s central control point, I have opted for a solution of the 3 coordinates of the central control point by averaging three approximate estimations; see the equations (5.17) to (5.19). Each of these three ap- proximated estimations is deﬁned by extending the concepts of minimization of the deformation Para evitar el planteamiento de un sistema de ecuaciones mal condicionado (o incluso incompatible) en la determinación del punto de control central del elemento, he optado por una solución de las 3 coorde- nadas del punto de control central promediando tres estimaciones aproximadas; ver las ecuaciones (5.17) CHAPTER 13. CONCLUSIONS 190 binando información geométrica de la malla y con- sideraciones de minimización de la energía. Llegué a la conclusión de que era necesario introducir con- sideraciones de índole energética en la construcción geométrica del elemento, al intentar resolver ejem- plos en régimen no-lineal. Previamente, ante la infra- determinación de los 30 parámetros necesarios para deﬁnir la geometría del elemento, me vi impulsado a tomar decisiones arbitrarias pero plausibles. La más determinante inicialmente fue suponer que el valor del parámetro Ψ era igual a 1 3 (ver el apartado 5.4). Al comprobar que con este valor ﬁjo de Ψ, el elemento re- sultaba mal planteado —y al llegar a esta conclusión por un razonamiento de desequilibrio de la energía interna— decidí que la solución al mal planteamiento del elemento la tenía que obtener introduciendo aspec- tos de minimización de la energía en la construcción geométrica del elemento. combining, geometric information from the mesh and energy minimization considerations. I drew the conclusion that it is necessary to introduce energy considerations in the geometric construc- tion of the element when I tried to solve exam- ples in the non-linear regime. Before that, faced with the under-determination of the 30 neces- sary parameters to deﬁne the element’s geome- try, I felt compelled to make some arbitrary but plausible choices. The most determining of them was assuming that the value of the Ψ parame- ter was equal to 1 3 (see section 5.4). When I re- alized that if I used this ﬁxed value for Ψ, the element was not well-posed —and reaching this conclusion by pondering on an internal energy disequilibrium— I became convinced that I would ﬁnd the solution to the ill-posedness of the ele- ment by introducing energy minimization consid- erations in the element’s geometric construction. g Estos aspectos de minimización de la energía se concretan en este caso en la parametrización de la construcción de la superﬁcie del triángulo cúbico de Bézier. De modo que establezco una relación entre la deﬁnición paramétrica de la superﬁcie del triángulo de Bézier —que depende de la posición de los puntos de control— y la energía interna de deformación del elemento. CHAPTER 13. CONCLUSIONS Como al resolver un problema mecánico por el método de los elementos ﬁnitos, en el fondo estoy re- solviendo un problema de minimización de la energía con respecto a las variables del problema; y como en el paradigma del elemento BEST las posiciones de los puntos de control interiores del triángulo no forman parte de las variables del problema: la determinación de las posiciones de los puntos de control estaba que- dando al margen de la solución del problema de mi- nimización de la energía mecánica de todo el mode- lo. Por consiguiente he desarrollado un conjunto de problemas reducidos para poder determinar el valor de los parámetros internos del elemento que dependen de consideraciones energéticas y que son lo suﬁcien- temente representativos y generales para no tener que resolver sus valores de manera global; ni siquiera a nivel de todo el elemento. Concretamente he planteado 3 problemas reducidos para los 3 modos de deforma- ción de la lámina: These energy minimization aspects are fo- cused on the parametrization of the Bézier cubic triangle’s surface construction. This way I estab- lish a relationship between the parametric deﬁ- nition of the Bézier triangle’s surface —which is governed by the position of the control points— and the element’s internal deformation energy. Since the resolution of a mechanic problem us- ing the ﬁnite element method, implies in essence the resolution of an energy minimization problem with respect to the problem’s variables; and since in the paradigm of the BEST element, the inter- nal control points’ positions do not belong to the set of the problem’s variables, then: the determi- nation of the control points’ positions was being put aside from the resolution of the overall me- chanic energy minimization problem. Therefore, I have deﬁned a set of reduced problems in or- der to determine the value of the element’s in- ternal parameters depending on energy consider- ations and which are sufﬁciently representative and general in order to avoid the need to solve their values globally; not even for the element as a whole. More speciﬁcally, I have deﬁned 3 dif- ferent reduced problems for the 3 different shell deformation modes: 1. Problema reducido de la deformación de ﬂexión, o variación de la curvatura (ver el apartado Ben- ding energy en la página 57). 1. Reduced problem for the bending deforma- tion, or curvature variation (see the section titled Bending energy on page 57). 13.1.4 Successful element construction based on energy minimization principles Un aspecto original de esta tesis consiste en la deﬁni- ción geométrica del triángulo cúbico de Bézier, com- An original aspect of this thesis consists on the geometric deﬁnition of the cubic Bézier triangle CHAPTER 13. CONCLUSIONS Axial (membrane) energy en la página 59). Axial (membrane) energy en la página 59). 3. Problema reducido de la deformación de cortan- te en el plano, o rotación de taladro (ver el apar- tado Energy minimization for the in-plane shear deformation mode en la página 106). 3. Reduced problem for the in-plane shear de- formation, or drilling rotation (see the sec- tion titled Energy minimization for the in- plane shear deformation mode on page 106). La solución del primer problema reducido genera un resultado muy aproximado al de la solución del se- gundo problema reducido (ver la ﬁgura 5.9). En cam- bio la solución del tercer problema reducido genera un resultado distinto a los dos primeros (ver la ﬁgu- ra 9.7). He resuelto la discrepancia entre ambas so- luciones realizando una media ponderada de las dos soluciones según la ecuación (9.10). The solution to the ﬁrst reduced problem yields a result that approximates very well the solution of the second reduced problem (see ﬁgure 5.9). How- ever, the third reduced problem yields a different result from the ﬁrst two (see ﬁgure 9.7). I have solved the disagreement between the two solu- tions by averaging them with weights following equation (9.10). g El segundo y tercer problema reducidos contienen de manera implícita la hipótesis de que el material es elástico lineal (ver ecuación (5.25) y ﬁgura 9.8). Adi- cionalmente, el tercer problema reducido también in- cluye de manera implícita la hipótesis de material isó- tropo (ver la ﬁgura 9.5). Y además, la ponderación en- tre las soluciones del primero y tercer problemas redu- cidos también contiene de manera implícita la hipóte- sis de que el material es elástico lineal. Es fácil cam- biar la ponderación de la ecuación (9.10) para mate- riales no-lineales. Sería conveniente comprobar si las hipótesis incluidas de manera implícita sobre las pro- piedades elásticas del material, inﬂuyen de manera excesivamente negativa para el caso de materiales an- isótropos y/o no-lineales. The second and third reduced problems in- clude implicitly the hypothesis of linear elastic material (see equation (5.25) and ﬁgure 9.8). The third reduced problem also includes implicitly the hypothesis of isotropic material (see ﬁgure 9.5). Moreover, the weighted average between the so- lutions of the ﬁrst and third reduced problems also implies the hypothesis of linear elastic ma- terial. It’s easy to change the weighting factor in equation (9.10) for non-linear materials. Axial (membrane) energy en la página 59). In the case of anisotropic and/or non-linear materials it would be advisable to check whether the hypothe- sis on the elastic properties of the material, intro- duced implicitly, affect too negatively the results. Con la implementación de las soluciones a estos problemas reducidos, he conseguido aplicar con éxi- to el elemento BEST al cálculo de láminas delgadas en régimen lineal y geométricamente no-lineal con un método implícito (ver la ﬁgura 11.22). By implementing the solutions of these re- duced problems, I have successfully applied the BEST element to the analysis of thin shells in linear and geometrically non-linear regimes us- ing an implicit method (see ﬁgure 11.22). CHAPTER 13. CONCLUSIONS 2. Problema reducido de la deformación de mem- brana, o extensión en el plano (ver el apartado 2. Reduced problem for the membrane defor- 13.1. CONTRIBUTIONS 191 mation, or in-plane extension (see the sec- tion titled Axial (membrane) energy on page 59). mation, or in-plane extension (see the sec- tion titled Axial (membrane) energy on page 59). 13.1.5 Cubic convergence using linear information I had serious doubts about the element’s order of convergence because the approach to develop the BEST element is non-orthodox (see chapter 8). And this cause of concern was certainly justi- ﬁed, because there are cases in which conver- gence is not even linear (specially when the el- ement suffers from membrane locking), but there are also cases where convergence is linear (see ﬁgure 11.1b), other cases in which the conver- gence is quadratic (see ﬁgures 11.6c and 11.10a), Al adoptar un enfoque no ortodoxo para el desarro- llo del elemento BEST, tenía serias dudas sobre el or- den de convergencia del elemento (ver el capítulo 8). Y esta preocupación estaba ciertamente justiﬁcada, ya que hay casos en los que ni siquiera se obtiene conver- gencia lineal (especialmente cuando el elemento sufre por bloqueo de membrana), pero también hay casos en los que se obtiene convergencia lineal (ver la ﬁgu- ra 11.1b), otros en los que se obtiene convergencia cua- drática (ver las ﬁguras 11.6c y 11.10a), e incluso he CHAPTER 13. CONCLUSIONS 192 and I have even obtained a result with cubic con- vergence (see ﬁgure 8.4). obtenido un resultado con convergencia cúbica (ver la ﬁgura 8.4). This last result shows that the BEST element has the potential to achieve cubic convergence. But at the same time, the fact that this order of convergence depends on the value given to the drilling rotations, and that with the value ﬁnally used in this particular case (Cylinder under inter- nal pressure) I have only achieved quadratic con- vergence, casts doubts on the possibility of repro- ducing this result for a wide range of problems. Este último resultado demuestra que el elemento BEST tiene potencial para alcanzar convergencia cú- bica. Pero al mismo tiempo, el hecho de que este orden de convergencia dependa del valor de las rotaciones de taladro, y que con el valor adoptado ﬁnalmente en es- te caso concreto solo se haya conseguido convergencia cuadrática hace que existan dudas sobre la posibili- dad de reproducir de manera consistente este resulta- do para un amplio rango de problemas. I have also demonstrated that for in-plane shear dominated problems, the formulation used in this thesis only achieves linear convergence. También se ha demostrado que para problemas do- minados por la deformación de cortante en el plano, la formulación utilizada en esta tesis solo alcanza con- vergencia lineal. 13.1.5 Cubic convergence using linear information This thesis does not tackle the issue of the computational cost-beneﬁt analysis of the BEST element compared to other classical elements. That is, whether the precision improvement pro- vided by a quadratic or cubic order of conver- gence, compensates the extra cost added by the resolution of the systems of linear equations of rank 3 for the element’s cubic geometric re- construction. Besides, the cost-beneﬁt analysis should also take into account that the resolution of the global system of equations is not equiva- lent to that of a linear element —despite the total number of variables is equivalent to that of a lin- ear element—. In the BEST element, the global system’s bandwidth is larger than those of other linear rotation-free shell elements; because, as I have said earlier, more connectivities from neigh- boring elements are used. En esta tesis no se ha abordado el análisis de coste- beneﬁcio computacional del elemento BEST respecto a elementos clásicos. Es decir, si el coste añadido que su- pone la resolución de los sistemas lineales de ecuacio- nes de rango 3 para la reconstrucción de la geometría cúbica del elemento BEST, queda compensado por la mejora de precisión que representa una convergencia de orden cuadrático o cúbico. Por otro lado, también se tendría que tener en cuenta que el coste de resolución del sistema de ecuaciones global no es equivalente al de un elemento lineal —aunque el número total de va- riables sí sea equivalente al de un elemento lineal— ya que en el elemento BEST el ancho de banda del siste- ma global está ampliado con respecto a otros elemen- tos de lámina sin rotaciones lineales, dado que como he explicado anteriormente, se usan más conectivida- des de elementos vecinos. Be as it may, I have demonstrated numeri- cally for different problems, that the BEST ele- ment achieves a better-than-linear order of con- vergence (quadratic or cubic). This is a very rele- vant result, because it means a potential advan- tage of the paradigm developed in this thesis. Sea como fuere, se ha demostrado numéricamente que para distintos problemas el elemento BEST alcan- za un orden de convergencia más-que-lineal (cuadrá- tico o cúbico). Este es un resultado muy relevante, por- que representa una ventaja potencial del paradigma desarrollado en esta tesis. 13.1. CONTRIBUTIONS 13.1. CONTRIBUTIONS 193 Lagrangiana total. Como dice el propio Stanley [120, REMARK 3.4]: a su entender “serían necesarias hipó- tesis más severas [que las que Stanley utiliza] para lo- grar el mismo efecto de desacoplamiento en el espesor con una descripción Lagrangiana total (. . . ) debido a la no-linealidad explícita que se engendraría.” grangian formulation. Citing Stanley [120, RE- MARK 3.4]: “It appears [to him] that harsher assumptions would be required to achieve the same thickness-decoupling effect with a Total La- grange description (. . . ) due to the explicit non- linearity thereby engendered.” p q g En realidad el problema no subyace en la no- linealidad que conlleva la formulación Lagrangiana total, sino en las simpliﬁcaciones que Stanley [120, p. 34] y numerosos otros autores [9, pp. 545–549] apli- can en la formulación Lagrangiana actualizada al li- nealizar los desplazamientos asociados a las rotacio- nes de la ﬁbra perpendicular a la superﬁcie media de la lámina. Ya que, por un lado, con hipótesis idénti- cas —a efectos del desacoplamiento en el espesor— a las que usa Stanley, yo también he logrado el efecto de desacoplamiento en el espesor; incluyendo todos los términos de alto orden de la descripción de la defor- mación. Y por el otro lado las formulaciones Lagran- giana actualizada y Lagrangiana total son equivalen- tes. Los distintos autores que usan la formulación La- grangiana actualizada para problemas no-lineales la preﬁeren porque al actualizar la conﬁguración en ca- da paso de tiempo o incremento de carga de la resolu- ción del problema no-lineal, se limita el error que se comete al descartar los términos de alto orden que se desprecian al linealizar las ecuaciones. Mientras que en la formulación Lagrangiana total ese error se acu- mularía; en particular el relacionado con las grandes rotaciones. Actually, the problem doesn’t lie in the non- linearity associated with a Total Lagrangian for- mulation, but in the simpliﬁcations that Stanley [120, p. 34] and many other scholars [9, pp. 545– 549] apply, in the Updated Lagrangian formula- tion, when they linearize the rotation-associated displacements of the ﬁber perpendicular to the shell’s midsurface. Because, on the one hand, using identical hypothesis —on what relates to the thickness decoupling— to those used by Stan- ley, I have also accomplished the thickness de- coupling result; including all the higher order terms of the deformation description. Preintegración en el espesor eﬁciente y con precisión geométrica This is without doubt the most difﬁcult con- clusion to write down. It strikes me that Stanley proposed the through-the-thickness pre- integration for continuum-based shell elements using an Updated Lagrangian formulation, to re- duce the computational cost in the building of the shell elements, back in 1985; and since then, this solution has not been yet applied for a Total La- Esta es seguramente la conclusión más difícil de re- dactar. Me resulta sorprendente que en 1985 Stan- ley propusiera la preintegración en el espesor para elementos de lámina basados en el continuo con for- mulación Lagrangiana actualizada, para economizar el coste computacional de la construcción de los ele- mentos de lámina; y que desde entonces no se haya aplicado esta misma solución para una formulación CHAPTER 13. CONCLUSIONS CHAPTER 13. CONCLUSIONS 194 tualizada todavía es preciso reevaluar las integrales en el espesor en cada paso de tiempo o incremento de carga, es decir: en cada actualización de la conﬁgura- ción; mientras que para la formulación Lagrangiana total sólo es necesario evaluar las integrales en el espe- sor una sola vez: para la conﬁguración de referencia. Es verdad que tal y como se ha explicado en la pági- na 85, las integrales en el espesor se han reducido a 14 integrales distintas para cada punto de Gauss, con lo que su coste computacional no es signiﬁcativo. Por consiguiente, se puede evaluar el incremento de coste computacional entre las formulaciones Lagrangiana total y Lagrangiana actualizada para cada paso de tiempo o paso de carga. En cualquier caso, el avance conceptual es notable. ing all the conceptual and formal power of the continuum-based shell elements. However, in the Updated Lagrangian formulation there is still the need to reevaluate the through-the-thickness integrals at each time-step or load-step, that is: for each conﬁguration update. Whereas in the Total Lagrangian formulation, the through- the-thickness integrals are evaluated just once: at the reference conﬁguration. The cost of the through-the-thickness integrals after decoupling the thickness terms is relatively cheap. There are just 14 scalar integrals to perform for each Gauss point. Therefore the difference in computational cost between the Total Lagrangian and the Up- dated Lagrangian formulations can be evaluated for each time-step or load-step. Anyway, the con- ceptual advantage is signiﬁcant. En el desarrollo realizado se han descompuesto el tensor de pseudo-deformaciones —g— y la matriz de cambio de coordenadas de locales del material a pa- ramétricas del elemento —A en notación tensorial y Q en notación de Voigt— (un concepto parecido al de shifter que se usa en la bibliografía pero que no es exactamente el mismo), en tres componentes en fun- ción del exponente {0,1,2} que afecta a la coordenada del espesor ζ presente en el numerador de las ecuacio- nes (6.48), (6.59) y (6.63) en las páginas 71-73. En el caso del tensor de deformaciones, éste se ha descom- puesto en la componente de membrana, de ﬂexión y de los términos no-lineales. En el caso de la matriz de cambio de coordenadas, ésta se ha descompuesto en los términos constante en el espesor y lineal —para A— y en los términos constante, lineal y cuadrático —para Q. CHAPTER 13. CONCLUSIONS Con ello se consigue transformar las ecua- ciones (7.3) y (7.4) en la página 78 en las expresiones de las ecuaciones (7.32) y (7.33) y las ecuaciones (7.34) y (7.35) en la página 83; donde las integrales en el espesor quedan condensadas en integrales de valores escalares en las expresiones de las ecuaciones (7.36) a (7.47) en las páginas 83-84. p g g In the development of the present thesis I have decomposed the pseudo-deformation ten- sor —g— and the matrix of change of coordi- nates from local of the material to parametric of the element —A in tensor notation and Q in Voigt notation— (a concept similar to the shifter used by other scholars but not exactly equal), into three components depending on the expo- nent {0,1,2} affecting the thickness coordinate ζ present in the numerator of the equations (6.48), (6.59) and (6.63) on pages 71–73. The deforma- tion tensor has been decomposed into the: mem- brane, bending and non-linear terms. The matrix of change of coordinates has been decomposed into the constant and linear terms —for A— and into the: constant, linear and quadratic terms — for Q. This way, the equations (7.3) and (7.4) on page 78 are transformed into the expressions of equations (7.32) and (7.33) and equations (7.34) and (7.35) on page 83; where the thickness inte- grals are turned into the scalar integrals shown in equations (7.36) to (7.47) on pages 83–84. Así pues, en este apartado concluyo que es posible realizar una descomposición de los términos de la des- cripción de la deformación y de las matrices de cambio de coordenadas que permite pre-integrar en el espesor del elemento de manera eﬁciente y geométricamente precisa. Esta conclusión es equivalente a la de Stan- ley [120]. Sin embargo, a diferencia de Stanley, en es- te caso he logrado aplicar esta pre-integración en el espesor para una formulación Lagrangiana total. El haber desarrollado un elemento de lámina sin rota- ciones es un factor que sin duda ha contribuido a este resultado al no tener que tratar con la complejidad de To sum up this section, I conclude that it is possible to perform a decomposition of the description of the deformation terms and the matrices of change of coordinates, that grants an efﬁcient and geometrically accurate through- the-thickness pre-integration. This conclusion is equivalent to that provided by Stanley [120]. 13.1. CONTRIBUTIONS And, on the other hand, the Updated Lagrangian and To- tal Lagrangian formulations are equivalent. Dif- ferent scholars that use the Updated Lagrangian formulation for non-linear problems, prefer it be- cause by updating the conﬁguration at each time- step or load-step of the non-linear resolution pro- cess, the error incurred by discarding the higher order terms of the linearization of the equations is limited. Whereas in the Total Lagrangian for- mulation this error would add-up; particularly the one associated to large rotations. Así pues, estoy sorprendido de no haber encon- trado referencias en la bibliografía revisada (que es más extensa que la citada) del desacoplamiento y pre- integración en el espesor de los términos de la des- cripción de la deformación y del jacobiano sin aplicar simpliﬁcaciones adicionales para elementos de lámi- na con formulación Lagrangiana total. Este desaco- plamiento y pre-integración en el espesor también son una novedad para los elementos de lámina delgada sin rotaciones. Therefore, I am stricken by not having found in all the bibliography reviewed (which is broader than the cited bibliography) any references of the thickness decoupling and through-the-thickness pre-integration of the description of the defor- mation and Jacobian expressions without apply- ing further simpliﬁcations, for shell elements with Total Lagrangian formulation. This thick- ness decoupling and through-the-thickness pre- integration are also a novelty for thin shell rotation-free elements. Stanley [120] ya describía las ventajas que conlle- van este desacoplamiento y pre-integración en el espe- sor para los elementos de lámina basados en el conti- nuo. Y es que lo que permite esta preintegración en el espesor, es hacer que la tecnología de elementos de lá- mina basados en el continuo sea equivalente en coste a los elementos basados en resultantes o en teorías de lá- mina; al tiempo que se conserva toda la potencia con- ceptual y formal de los elementos basados en el conti- nuo. Sin embargo, en la formulación Lagrangiana ac- Stanley [120] already described how the continuum-based shell elements beneﬁted from the advantages associated with this thick- ness decoupling and through-the-thickness pre- integration. The through-the-thickness pre- integration grants to make the technology of continuum-based shell elements competitive in cost with the resultant-based shell elements or the elements based on shell theories; while keep- CHAPTER 13. CONCLUSIONS 13.2.1 The quantum nature of research work He realizado el trabajo de investigación para esta tesis en paralelo a otros desempeños profesionales. Si bien el trabajo ordinario se puede linealizar e interrum- pir prácticamente a voluntad en multitud de desempe- ños, en lo que concierne a la investigación, esto es con- traproducente. Llegué a esta conclusión relativamente temprano. Y por ello concentré mis esfuerzos para tra- bajar en la tesis durante los meses de verano (y espe- cialmente los meses de agosto) para conseguir avan- ces. Porque franquear la frontera del conocimiento y empujarla más allá, requiere una comprensión deta- llada de las zonas de penumbra que no tienen la ilu- minación de los focos del conocimiento establecido. El tránsito por estas zonas de penumbra se necesita hacer desde multitud de puntos de vista, teniendo en cuenta multitud de consideraciones. Esto requiere un esfuerzo de abstracción muy intenso y que se tiene que mante- ner hasta que se alcanza la comprensión del problema que se está abordando en cada caso. La interrupción de dicho esfuerzo de abstracción por periodos de tiem- po prolongados obliga inevitablemente a que el inves- tigador regrese a la casilla de salida cuando retoma el trabajo de investigación. Por este motivo hablo de la naturaleza cuántica del trabajo de investigación. Para alcanzar nuevos niveles de conocimiento se re- quieren umbrales cuánticos de esfuerzo. La combina- ción de esfuerzos que individualmente no alcancen el umbral cuántico de esfuerzo necesario no conducirán a alcanzar el nivel de conocimiento perseguido; inclu- so si la suma de todos ellos supera con creces el valor de umbral cuántico. I have done this research thesis in parallel to other professional assignments. While many or- dinary work tasks can be distributed linearly and interrupted almost whenever desired; on what concerns research, this practice is counterproduc- tive. I reached this conclusion relatively early. And this is why I concentrated my efforts to work on the thesis during the summers (and spe- cially during the months of August) in order to attain progresses. Because pursuing the knowl- edge frontier and pushing beyond it, requires a detailed understanding of the shadowy areas that remain away from the established knowledge lights. Transiting through these shadowy areas needs careful consideration from many points of view. This requires a very intense abstraction ex- ercise that has to be sustained, until the prob- lem at hand is fully comprehended. 13.2.1 The quantum nature of research work Interrupt- ing this abstraction effort for long periods of time forces the researcher to retreat back to square one when the research work is resumed. That’s the reason why I speak about the quantum na- ture of research work. A quantum work thresh- old is required in order to reach new knowledge grounds. The combination of smaller efforts that, taken individually, do not reach that quantum work threshold, will not lead to the achievement of the knowledge grounds being pursued; even if all of them combined exceeds largely that quan- tum threshold value. 13.2. LESSONS LEARNED las grandes rotaciones en el marco de una formulación Lagrangiana total. element has certainly contributed to this result, by not having to deal with the complexity of the large rotations in the framework of a Total La- grangian formulation. CHAPTER 13. CONCLUSIONS However, unlike Stanley, in this case I have been able to apply this through-the-thickness pre-integration for a Total Lagrangian formula- tion. The development of a rotation-free shell 195 13.2. LESSONS LEARNED CHAPTER 13. CONCLUSIONS CHAPTER 13. CONCLUSIONS CHAPTER 13. CONCLUSIONS 196 published scientiﬁc production. published scientiﬁc production. chas respuestas es la producción cientíﬁca publicada. chas respuestas es la producción cientíﬁca publicada. Las bibliotecas universitarias que facilitan los me- canismos de búsqueda y acceso a la producción cien- tíﬁca publicada, son un instrumento valiosísimo pa- ra los investigadores; e igual de imprescindible que el papel, el lápiz o el ordenador. Las tecnologías de la in- formación y la suscripción que pagan las bibliotecas universitarias a los agregadores de revistas cientíﬁ- cas: hacen posible el trabajo del investigador, acortan los tiempos necesarios para consultar referencias bi- bliográﬁcas al vuelo y reducen también la interacción del investigador con la infraestructura física de la bi- blioteca. Pero todavía hay recursos que sólo se pueden consultar en los ediﬁcios de las bibliotecas y hay ar- tículos que sólo se pueden obtener solicitándolos expre- samente; cosa que requiere la intervención de personal de la biblioteca. The university libraries facilitate the mecha- nisms to search and access the published scien- tiﬁc production. And in doing so, they become an invaluable instrument for researchers; equally essential as a pen, paper or a computer. The advent of information technologies and the sub- scriptions that university libraries pay to scien- tiﬁc journal aggregators: facilitates the work of researchers, shortens the times to consult bibli- ographic references on the spot, and also reduce the interaction of the researcher with the phys- ical infrastructure of the library. But there are still many resources which can only be consulted at the libraries’ buildings and there are papers which can only be obtained by requesting them speciﬁcally. These circumstances require the li- brary staff to intervene. Buena parte del trabajo de investigación de esta te- sis se enmarca en el marco temporal de la crisis econó- mica que estalló en el año 2008. Las universidades no han quedado indemnes de esta crisis económica y han aplicado una serie de ajustes en sus presupuestos. Una de las decisiones que ha tomado la Universitat Politèc- nica de Catalunya ha sido el cierre de las bibliotecas durante el periodo no lectivo de verano. La decisión de la Universitat Politècnica de Catalunya de cerrar sus bibliotecas durante los periodos no-lectivos de ve- rano es una decisión con impacto sobre la producción cientíﬁca de la propia universidad. Porque un periodo no-lectivo no implica que sea no hábil para la investi- gación. CHAPTER 13. CONCLUSIONS Lo contrario en cambio es cierto: los periodos no-lectivos en la universidad son los más productivos para la investigación. Y para sostener esta aﬁrmación me remito al apartado 13.2.1. Me he encontrado en más de una ocasión con la necesidad de consultar un recurso que no he podido solicitar hasta la reapertura de la biblioteca. Perdiendo así una ventana de opor- tunidad para aprovechar al máximo ese umbral de esfuerzo cuántico de que disponía durante el mes de agosto. A substantial part of this thesis’ research work takes place in the time frame of the eco- nomic crisis that unfolded in the year 2008. Uni- versities haven’t come out unscathed from this economic crisis, and have applied several budget cuts. One of such decisions made by the Poly- technic University of Catalonia has been to close the libraries during the summer vacations pe- riod. The decision made by the Polytechnic Uni- versity of Catalonia to close its libraries during the summer vacations period has an impact on its own scientiﬁc output, because a non-academic period does not imply necessarily a non-working period; particularly for research. The opposite is true: the non-academic periods are the most pro- ductive for research. And to sustain this state- ment I refer to section 13.2.1. I have found my- self in many occasions in need of consulting a given resource, that I haven’t been able to request until the libraries reopened. This circumstance made me lose a window of opportunity to take full advantage of of that quantum work thresh- old I could invest during those months of August. Es fundamental que una universidad que aspira a generar producción cientíﬁca mantenga abiertas sus bibliotecas durante todo el año. Ello es una declara- ción de principios que ninguna crisis económica pue- de perturbar. Porque aceptar sin rechistar el cierre de las bibliotecas, es aceptar la mutación de la naturale- za misma y razón de ser de la Universitat Politècnica de Catalunya. It’s paramount, for a university that aspires to generate scientiﬁc output, to preserve its li- braries open all year round. This is a statement of principles that no economic crisis should un- dermine. Because accepting without a protest the shutdown of libraries, is accepting the mutation of the very nature and mission of the Polytechnic University of Catalonia. Rompo una lanza a favor de las bibliotecas Como decía en la introducción de este capítulo: en el trabajo de doctorado hay mucho de aprendizaje. Pero este aprendizaje no es ordenado. Hay mucho de au- toaprendizaje. El investigador o la investigadora tie- ne que encontrar respuestas a medida que se formu- la preguntas. Y la principal fuente para encontrar di- As I wrote at the introduction of this chapter: there’s a lot of learning in a doctoral work. But this learning isn’t necessarily ordered. There’s a lot of self-learning. The researcher must ﬁnd answers as he or she asks questions. The main source of knowledge to ﬁnd those answers is the CHAPTER 13. CONCLUSIONS Esta tesis se ha beneﬁciado en gran medida de los trabajos de muchos otros cientíﬁcos que han publica- This thesis has largely beneﬁted from the 13.2. LESSONS LEARNED 197 do su trabajo. Y hubiera sido imposible alcanzar los resultados que aquí se incluyen, junto con sus explica- ciones claras y concisas, sin que el autor hubiera po- dido consultar las referencias que se indican. Aún así, no todas las referencias incluidas en esta tesis se han obtenido por la vía de la biblioteca. Algunas las he descargado de internet. Y algunas de las referencias incluidas ni siquiera las he podido consultar. Pero es- tán referenciadas por la relevancia de la contribución que contenían esas publicaciones y cuyos resultados se han perpetuado a lo largo del tiempo para acreditar a los autores. Así pues, es necesario seguir reforzando los recursos de las bibliotecas para asegurar el acceso al conocimiento publicado por parte de la comunidad cientíﬁca. do su trabajo. Y hubiera sido imposible alcanzar los resultados que aquí se incluyen, junto con sus explica- ciones claras y concisas, sin que el autor hubiera po- dido consultar las referencias que se indican. Aún así, no todas las referencias incluidas en esta tesis se han obtenido por la vía de la biblioteca. Algunas las he descargado de internet. Y algunas de las referencias incluidas ni siquiera las he podido consultar. Pero es- tán referenciadas por la relevancia de la contribución que contenían esas publicaciones y cuyos resultados se han perpetuado a lo largo del tiempo para acreditar a los autores. Así pues, es necesario seguir reforzando los recursos de las bibliotecas para asegurar el acceso al conocimiento publicado por parte de la comunidad cientíﬁca. work of many other scientists that have pub- lished their results. And it would have been im- possible to achieve the results included in this thesis, along with the clear and concise explana- tions within, without the possibility for the au- thor to consult the references cited in the bibli- ography. Nevertheless, not all the references in- cluded in this thesis have been obtained using the libraries’ services. Some of them have been down- loaded from the internet. And I haven’t been able to check at all a few of the references included; but they are included because of the relevance of the contribution they provide, and whose results have perpetuated over time to the credit of their authors. 13.2.3 Goals vs Objectives The third profound lesson learned in this thesis is that goals and objectives shall not be confused. That is, the objectives stated in chapter 2 must be understood as goals. The goals shall serve to keep in mind which is the direction in which I want to progress, and why. But the objectives are measurable advances in the path towards those goals. In this sense, my thesis advisor proposed me in a clear way the objective for a thesis (my thesis): “can you improve the performance of the EBST [41] element?” El tercer aprendizaje de calado de esta tesis es que no se deben confundir las metas con los objetivos. Es decir, los objetivos expresados en el capítulo 2 deben entenderse en realidad como metas. Las metas deben servir para no perder de vista la dirección en la que se quiere avanzar y porqué. Pero los objetivos son avan- ces medibles en el camino hacia esas metas. En es- te sentido, mi director de tesis me planteó de mane- ra muy nítida el objetivo para una tesis (mi tesis): “a ver si consigues mejorar el rendimiento del elemento EBST [41]”. Yo he combinado el objetivo que me marcaba mi director de tesis con mis motivaciones. Y ello me ha llevado a perder el enfoque del objetivo. Ello también ha sido fruto de haber percibido un posible exceso de optimismo respecto de mis capacidades como investi- gador. Esa distracción entre metas y objetivos (tam- bién inducida por las necesidades de los proyectos de investigación que ﬁnancian la actividad investigado- ra) ha hecho que no haya orientado suﬁcientemente el trabajo hacia la consecución del objetivo. I have mixed the objective stated by my thesis advisor with my motivations. And this has led me to lose the focus on the objective. This has also been the result of having perceived some overop- timism regarding my research abilities. This dis- traction between goals and objectives (also in- duced by the burdens of research projects that ﬁnance research) has somewhat distracted me from my endeavor to achieve the objective. If I had more ﬁrmly held the focus of my re- search work towards achieving the objective, I would have probably noticed earlier the need to take notice of the membrane locking issues in shell elements. This is an aspect of the shell ﬁnite elements technology I ignored in my engineering training. CHAPTER 13. CONCLUSIONS Therefore, it is necessary to continue strengthening the libraries’ resources to guaran- tee the access to published knowledge by the re- search community. 13.2.4 Non-linearity is a lifeline, not a hurdle En el desarrollo de un nuevo elemento ﬁnito estruc- tural, se puede percibir el tratamiento del problema de la no-linealidad como una diﬁcultad añadida. Sin embargo, el trabajo añadido que conlleva resolver el problema no-lineal desde el punto de vista de formula- ción y programación es muy pequeño en comparación a los beneﬁcios que aporta. Dealing with the issue of non-linearity may be perceived as an added complexity when develop- ing a new structural ﬁnite element. However, the beneﬁts obtained by solving the non-linear prob- lem, require little additional work from the point of view of equation writing and programming. In my particular case, testing how the ele- ment worked in the non-linear regime, allowed me to detect problems caused by the hypothesis used in the formulation. Despite these hypothe- sis may not raise suspicions beforehand, putting them to the test using a non-linear example, al- lows sorting out quickly whether the hypothesis applied lead to a well-posed problem or, on the contrary, the problem is actually not well-posed and therefore does not respond to the physical re- ality being modeled. En mi caso, la realización de pruebas sobre el fun- cionamiento del elemento en régimen no-lineal me ha permitido detectar problemas con las hipótesis aplica- das en la formulación. Pese a ser hipótesis que de an- temano no despierten sospechas, el ponerlas a prueba con un ejemplo no-lineal permite detectar enseguida si las hipótesis que se están aplicando conducen a un problema bien planteado o si por el contrario el proble- ma que resulta está mal planteado y por consiguiente no responde a la realidad física que se quiere modelar. For this very reason, any researcher intend- ing to develop a new structural ﬁnite element, should quickly put the new formulation to the test against a non-linear problem in order to check the validity of the approach. Por este motivo, cualquier investigador que preten- da desarrollar un nuevo elemento ﬁnito estructural, debería tener cierta prisa por someter la nueva formu- lación al test de un problema no-lineal para veriﬁcar la bondad del planteamiento. I have beneﬁted twice from this trait of the non-linear setting. And in both occasions I have regretted not having performed the test earlier on. Hence, we must view non-linear tests as a lifeline to avoid making serious mistakes in our development’s hypothesis; not as an additional hurdle. CHAPTER 13. CONCLUSIONS CHAPTER 13. CONCLUSIONS 198 ﬂected by section 3.2.2, in the review of the state of the art I did not pay sufﬁcient attention to the issue of membrane locking. And therefore I was not sufﬁciently on guard against this type of prob- lems presented in continuum-based shell ﬁnite elements which, however, the references cited in this thesis comment on. As Bischoff, Wall, Blet- zinger and Ramm explain in [12, p. 123]: “Al- though literature in ﬁnite element technology in general is extremely rich, the problem of mem- brane locking has enjoyed less attention than, for instance, transverse shear locking in plates and shells.”, it should have served me as a clear warn- ing and I should have taken adequate measures early on. durante el doctorado debería haber adquirido hacia el principio y no hacia el ﬁnal. De hecho, como lo de- muestra el apartado 3.2.2, en la revisión del estado del arte no presté suﬁciente atención a la cuestión del bloqueo por membrana. Y por consiguiente no estaba prevenido frente a este tipo de problemas en los ele- mentos ﬁnitos basados en la teoría del continuo que sin embargo las referencias que incluyo en esta tesis comentan. Como explican Bischoff, Wall, Bletzinger y Ramm en [12, p. 123]: “Aunque la literatura en tecno- logía de elementos ﬁnitos es en general muy rica, el problema del bloqueo por membrana ha recibido me- nos atención que, por ejemplo, el problema del bloqueo por cortante en placas y láminas.”, me debiera haber servido de advertencia y yo debía haber tomado las debidas precauciones desde un primer momento. 13.2.3 Goals vs Objectives But I should have become aware of this issue towards the beginning rather than towards the end of the learning process. Indeed, as re- De haber mantenido con más ﬁrmeza la orienta- ción de la labor de investigación hacia la consecución del objetivo, seguramente me habría percatado más rápidamente de la necesidad de prestar atención a los problemas de bloqueo por membrana de los elemen- tos de lámina. Este era un aspecto de la tecnología de elementos ﬁnitos de lámina que desconocía en mi formación como ingeniero. Pero que en mi recorrido CHAPTER 13. CONCLUSIONS El cálculo por métodos numéricos en el mundo de los veleros P ARA TRATAR EL USO de los métodos numéricos en el mundo de los veleros es preciso referirse a los ámbitos de la práctica del deporte de la vela, donde la competitividad es máxima y por lo tanto se requieren los mejores y últimos avances para obtener esas décimas extras de velocidad que se conviertan en una ventaja competitiva deﬁnitiva. La competi- ción que reﬂeja este espíritu a la perfección y como ninguna otra es la Copa del América. Pero si bien la competición ha sido el laboratorio de ensayos para el desarrollo de nuevas y maravillosas tecnologías; éstas, una vez comprobada su validez no se quedan en el ámbito de la competición y son aplicadas una y otra vez siempre que haya oportunidad económica para ello. I N ORDER TO DISCUSS THE USE of numeri- cal methods in sailboats design it is com- pulsory to refer to the practices of the sail- ing sport where competitivity is maximum and therefore the best and latest advances are re- quired in order to obtain those tiny bits of ex- tra speed and turn them into the winning com- petitive advantage. The competition which reﬂects that spirit the best and beyond any doubt is the America’s Cup. While the com- petition has become the testing laboratory for new and wonderful technologies, once their va- lidity has been veriﬁed they are applied once and again beyond the competitive setting if there is an economic chance. P A I Por ejemplo, el mercado de superyates está cre- ciendo aún a pesar de la crisis económica [66] en par- te gracias a que la tecnología permite fabricar velas cada vez más grandes sin que ello repercuta en un peso de las velas tan desmesurado que sea imposi- ble moverlas a bordo (aunque a menudo se necesitan grúas para ello). Esto es posible por la utilización de avanzados materiales compuestos, pero también a las herramientas numéricas que calculan sus ten- siones para así poder realmente optimizar el uso del material. For example, the superyacht market is growing despite the economic crises [66] partly because technology enables making ever larger sails without increasing their weight in such a way that it would become im- possible to move them onboard (despite cranes are needed often). This is possible because of the use of advanced composite materials. 13.2.4 Non-linearity is a lifeline, not a hurdle Yo me he beneﬁciado en 2 ocasiones de esta virtud del planteamiento no-lineal. Y en ambas ocasiones me he arrepentido de no haber hecho la prueba antes. Así pues, debemos ver las pruebas no-lineales como una tabla de salvación para evitar cometer errores de bulto con las hipótesis de nuestros desarrollos; no como un obstáculo añadido. La Copa del América como referente tecnológico y deportivo The America’s Cup hardly needs any intro- duction. I will simply make the compul- sory remark that it is the oldest continuously awarded trophy in international sport. A win- ning strike of 132 years by the United States only increases the legend of this competition. It increases its prestige because it was pre- cisely a technological advance what permited Australia II to take the cup from the ameri- cans in 1983. It was thanks to a revolutionary keel design concept by Van Oossanen. Indeed, as Oossanen himself explains in the paper by Spurr [119], numerical methods and CFD sim- ulation played a vital role in the development of the design of the Australia II. They used a simulation program created for airplane wings at the National Aerospace Laboratory in Am- sterdam that allowed calculating the viscous resistance and was modiﬁed by Joop Slooff to include the wave making resistance effect. This is a decisive moment in sailing history; a sport historically dominated by tradition and folklore. Apenas hace falta presentar la Copa del Améri- ca. Simplemente haré la obligada reseña recordan- do que es la competición deportiva más antigua de cuantas se continúan celebrando. La leyenda engen- drada tras una racha de victorias de 132 años por Estados Unidos ensalza aún más si cabe el presti- gio de esta competición. Lo ensalza porque fue pre- cisamente un adelanto tecnológico el que permitió al Australia II arrebatar la copa a los norteameri- canos en 1983 gracias a un revolucionario concepto en el diseño de la quilla realizado por Van Oossa- nen. De hecho, tal y como explica Van Oossanen en el artículo de Spurr [119], los métodos numéricos y la simulación CFD jugaron un papel muy importante en el desarrollo del diseño de la quilla del Australia II. Se utilizó un programa de simulación para alas de avión del Laboratorio Nacional Aeroespacial de Amsterdam que permitía calcular la resistencia vis- cosa y que fue modiﬁcado por Joop Slooff para in- cluir los efectos de la resistencia por formación de olas. Este fue un momento decisivo en la historia de la vela; un deporte dominado históricamente por la tradición y las costumbres. Los norteamericanos se dieron cuenta de la ne- cesidad de aplicar la tecnología más puntera si pre- tendían recuperar la Copa del América (y con ella parte del orgullo perdido). APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN 200 part on the summer Olympic Games. In these regattas the competition is so close that some teams use the most advanced technologies to accelerate the learning curve and pull ahead of their competitors and themselves. So far, the use of technology for accelerated learning is limited [74], but the use of numerical methods to accelerate the learning curve is certainly in the path to the future. los Juegos Olímpicos de verano. En estas regatas la competencia es tan igualada que algunos equipos re- curren a las técnicas más avanzadas para poder ad- quirir conocimientos de manera acelerada y así su- perar a sus contrincantes y a sí mismos. Por ahora el uso de la tecnología para el aprendizaje acelerado es todavía limitado [74], pero ciertamente el uso de mé- todos numéricos para continuar el aprendizaje está en el camino hacia el futuro. El cálculo por métodos numéricos en el mundo de los veleros But also because of the numerical tools and pro- grams that compute their stresses and opti- mize every gram of material used. Otro ejemplo se encuentra en las competicio- nes de barcos que por su tamaño difícilmente dispondrían de suﬁciente presupuesto para justiﬁ- car semejantes desarrollos. Estamos hablando de los barcos de vela ligera que regularmente participan en We ﬁnd another example in the competi- tions with sailboats that, given their size and overall budget, can hardly count on the re- sources required to beneﬁt from those develop- ments. I am refering to the dinghies that take 199 APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN La Copa del América como referente tecnológico y deportivo Y así fue, en la siguiente edición de 1987 el Stars & Stripes logró recuperar la copa tras un esfuerzo en investigación y desarrollo sin precedentes [113]. The americans realized the need to apply cutting edge technology if they were to recover the America’s Cup (and part of the pride lost with it). And that’s how it went. In the next edition in 1987 Stars & Stripes achieved win- ning back the cup after an unprecedented re- search and development effort [113]. La historia anterior de la Copa del América tie- ne poca importancia en cuanto a su relación con los métodos numéricos. Sin embargo la historia más re- ciente habla por sí sola. El esfuerzo realizado por Es- tados Unidos en la campaña de 1987 lo mantuvieron y lo incrementaron si cabe defendiendo con éxito la The prior history of the America’s Cup has little importance regarding numerical meth- ods. But the more recent history talks by itself. The efforts the United States made for the 1987 campaign were sustained and increased A.1. THE AMERICA’S CUP AS A SPORTS AND TECHNOLOGICAL PINNACLE 201 copa en 19921 con America3. copa en 19921 con America3. copa en 19921 con America3. defending successfully the cup in 19921 with America3. Pero el resto de naciones ya habían tomado buena nota de que la inversión en tecnología les podía valer la conquista de un trofeo con la historia y el presti- gio de la Copa del América. Así, de nuevo en 1995 los norteamericanos fueron derrotados, esta vez por Black Magic de Nueva Zelanda. Este equipo rebosa- ba talento deportivo2, pero demostró asimismo que gracias a la inversión en tecnología se podía estar a un nivel competitivo que permitía aspirar a todo. But the other nations had already noticed the possibility of winning a trophy with the prestige and history of the America’s Cup by investing in technology. And thus, again in 1995 the americans werre defeated. This time the winner was Black Magic from New Zealand. This team was awash with talent2, but also demonstrated that thanks to invest- ing in technology they could achieve a compet- itive level enabling them to succeed. 1I am intentionally ommitting the 1988 regatta where many technological advances were applied, particularly brought from aeronautics, but the development time was far too short to allow any substantial advance and of pro- longed effect. 2Two of the key team members of that newzealander syndicate: Russell Coutts and Brad Butterworth, have won in all the America’s Cup editions in which they have taken part of since that regatta of 1995 until they faced off (then only one could win). In the 2010 regatta BMW Or- acle with Russell Coutts defeated Alinghi with Brad But- terworth. 1En este pasaje obvio intencionadamente la regata celebrada en 1988, pues si bien se aplicaron numerosos avances tecnológi- cos sobretodo importados desde el campo de la aeronáutica, los tiempos de desarrollo fueron excesivamente cortos para permitir un avance sustancial y de efecto prolongado. 2Dos de los miembros clave de aquél equipo neozelandés: Russell Coutts y Brad Butterworth, han vencido en todas las edi- ciones de la Copa del América en las que han participado desde aquella regata de 1995 hasta que se enfrentaron (sólo uno podia ganar). En aquella regata de 2010 el BMW Oracle con Russell Coutts venció al Alinghi con Brad Butterworth. 1En este pasaje obvio intencionadamente la regata celebrada en 1988, pues si bien se aplicaron numerosos avances tecnológi- cos sobretodo importados desde el campo de la aeronáutica, los tiempos de desarrollo fueron excesivamente cortos para permitir un avance sustancial y de efecto prolongado. 2Dos de los miembros clave de aquél equipo neozelandés: Russell Coutts y Brad Butterworth, han vencido en todas las edi- ciones de la Copa del América en las que han participado desde aquella regata de 1995 hasta que se enfrentaron (sólo uno podia ganar). En aquella regata de 2010 el BMW Oracle con Russell Coutts venció al Alinghi con Brad Butterworth. 3Nótese el contraste entre los enfoques adoptados para me- jorar la velocidad de los barcos. Si previamente el peso de los desarrollos había recaído principalmente en minimizar el freno que supone el avance del casco y sus apéndices a través del agua, ahora se incluía además maximizar la potencia y la eﬁciencia que se podía obtener de la propulsión de las velas. 4Es lo que se comúnmente se conoce como efecto twist. Es precisamente este efecto, combinado con las diﬁcultades de los ensayos a escala en túnel de viento en el estudio de las velas, que llevó al desarrollo del llamado Sailing Dynamometer [60] en 1988 por un estudiante de J. Milgram en MIT. 3Notice the change in the different tactics for increas- ing the boat speed. While the blunt of the developments had remained previously on minimizing the hull resis- tance and that of its appendages through the water, now the effort included maximizing the power and efﬁciency that could be obtained from the sails. 3Nótese el contraste entre los enfoques adoptados para me- jorar la velocidad de los barcos. Si previamente el peso de los desarrollos había recaído principalmente en minimizar el freno que supone el avance del casco y sus apéndices a través del agua, ahora se incluía además maximizar la potencia y la eﬁciencia que se podía obtener de la propulsión de las velas. La Copa del América como referente tecnológico y deportivo Si hasta ese momento los desarrollos y avances tecnológicos se habían centrado principalmente en la reducción de la resistencia al avance del casco a través del agua, los neozelandeses cogieron por sor- presa a sus rivales al poner en práctica un ambicio- so programa de investigación y desarrollo en velas; es decir, la propulsión de los barcos.3 El estudio de las velas entraña algunas diﬁcultades añadidas al estudio de la dinámica de los cascos en el agua. Si bien una componente del problema de los cascos en el agua —que no encontramos en el estudio de las velas— es la alteración que estos producen en la su- perﬁcie libre al moverse, los cascos se consideran en general como rígidos y que el ﬂujo de agua a su re- dedor tiene una dirección dominante uniforme. Sin embargo las velas cambian de forma debido a su ex- trema esbeltez y el ﬂujo de aire a su rededor varia en celeridad y en dirección dominante con la altu- ra debido al efecto de la capa límite atmosférica y a su combinación vectorial con la velocidad de avan- ce del velero.4 El programa de investigación y desa- While until then the technological ad- vances had focused mainly in the reduction of the resistance to the hull advance through the water, the newzealanders caught their rivals by surprise when they set off an ambitious re- search and development program in sails; that is, the power source of the boats.3 The study of sails entails some added challenges to that of the study of of hull dynamics in the water. While an aspect of the hulls in the water prob- lem —not found in the study of sails— is the variation they cause to the free surface when they move, the hulls are generally considered as rigid and the water ﬂow around them has a main and uniform direction. Instead, the sails change their shape provided their ex- treme slenderness. And the air ﬂow around them changes in velocity and dominant direc- tion with height because of the atmospheric boundary layer and its vectorial combination 3Notice the change in the different tactics for increas- ing the boat speed. 1I am intentionally ommitting the 1988 regatta where many technological advances were applied, particularly brought from aeronautics, but the development time was far too short to allow any substantial advance and of pro- longed effect. 2Two of the key team members of that newzealander syndicate: Russell Coutts and Brad Butterworth, have won in all the America’s Cup editions in which they have taken part of since that regatta of 1995 until they faced off (then only one could win). In the 2010 regatta BMW Or- acle with Russell Coutts defeated Alinghi with Brad But- terworth. 3Notice the change in the different tactics for increas- ing the boat speed. While the blunt of the developments had remained previously on minimizing the hull resis- tance and that of its appendages through the water, now the effort included maximizing the power and efﬁciency that could be obtained from the sails. 5A pesar de una exhaustiva búsqueda bibliográﬁca no he encontrado ninguna referencia especíﬁca detallando el funcio- namiento del programa MemBrainTM. Sin embargo sí conozco, mediante numerosas entrevistas con diseñadores de North Sails (aunque no con el propio Michael Richelsen) que el programa MemBrain es un programa de elementos ﬁnitos que utiliza un modelo de membranas para calcular los esfuerzos y las deforma- ciones de las velas. El modelo analiza las velas de manera esta- cionaria aun a pesar de modelarlas como membranas (véase la discusión en el apartado 1.4). Al no existir evidencias documen- tales acerca del funcionamiento de MemBrain, lo consideraremos como estado de la práctica y no como estado del arte. 5Despite an exhaustive bibliographic research I haven’t found any reference with the detailed function- ing of the MembrainTM program. However, I do know, through a number of interviews with North Sails design- ers (Mickey Ickert and Sandro Benini, although not with Michael Richelsen hihmself) that the MemBrain program is a ﬁnite element program which uses a membrane model to calculate the strains and stresses of the sails. The pro- gram performs a static analysis of the sails; despite model- ing them as membranes (see the discussion in section 1.4). As I can’t provide documental evidences of the working of MemBrain, I consider it state of practice and not state of the art. 4This is what is commonly known as the twist effect. Precisely this effect, combined with the complication of conducting scale tests in wind tunnel to study sail shapes, is what encouraged the development of the Sailing Dy- namometer [60] in 1988 by a pupil of J. Milgram at MIT. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN 202 with the sailboat’s forward speed.4 The re- search and development program of the New Zealand team for sails design focused on var- ious aspects. Burns Fallow [36] explains that the new technology developed by North Sails with the 3DL sails [7] opened the design space for the sail designers. Also, these new sails needed new structural analysis tools. That’s why North Sails assigned Michael Richelsen (one of its employees) to develop a software program called MemBrain TM5. At the same time and provided the greater importance of the downwind courses for the 1995 edition of the America’s Cup with respect to the previous editions the team from New Zealand, in col- laboration with North Sails New Zealand and the University of Auckland, decided to build a twisted ﬂow wind tunnel. This wind tunnel [40] —it still remains a world reference in its class— permited the team of New Zealand to pull ahead of their competitors in the knowl- edge of how these sails work. The combination of this knowledge, the effort made to exploit as much as possible the advantages offered by the 3DL sails, and the new calculation tools Black Magic won the 1995 regatta with a blow- ing 5–0 in front of a powerless Young America crewed by the crew of Stars & Stripes. rrollo para el diseño de velas en el equipo de Nue- va Zelanda se centró en varios aspectos. Tal y como comenta Burns Fallow en [36], la nueva tecnología desarrollada por North Sails con las velas 3DL [7] abría el campo de juego para los diseñadores de ve- las. Asimismo, este tipo de velas requería de nuevas herramientas de análisis estructural. Es por ello que North Sails encargó a Michael Richelsen (uno de sus empleados) el desarrollo de un programa de cálculo llamado MemBrainTM5. Al mismo tiempo, y dada la mayor importancia que tenían los rumbos de empo- pada en la edición de la Copa del América de 1995 con respecto a ediciones anteriores, el equipo de Nue- va Zelanda, en colaboración con North Sails Nueva Zelanda y la Universidad de Auckland, decidió cons- truir un túnel de viento de ﬂujo retorcido. Este túnel de viento [40] —que todavía es una referencia mun- dial en su clase— permitió al equipo de Nueva Ze- landa avanzar a sus competidores en el conocimien- to y funcionamiento de las velas. La Copa del América como referente tecnológico y deportivo While the blunt of the developments had remained previously on minimizing the hull resis- tance and that of its appendages through the water, now the effort included maximizing the power and efﬁciency that could be obtained from the sails. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN The computer revolution If there had not occured signifﬁcant advances in the application of numerical methods to the sailing industry in general and high perfor- mance in particular before it’s mainly because of the very high computational cost of the nu- merical tools or simply the lack of resources to make the computations before the 1980’s. It’s somewhat awkward to use terms like com- putational cost nowadays that a smartphone weighing less than 200 grams and costing around $100 has the same computer power as a supercomputer 30 years ago which occupied an entire room and costed several thousand $. That’s the main reason why numerical meth- ods still lacked validation and the scientiﬁc community relied heavily on laboratory tests in order to make their designs and technolog- ical advances. Take for example the paper by Jerome Milgram [79] on the different technolo- gies developed and applied in the design of the boats of the America3 syndicate. Within that paper and the ensuing discussion Milgram ex- plains the controversy still existing in 1992!!! between numerical and experimental tools. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN Combinando este conocimiento con el esfuerzo realizado para aprove- char al máximo las capacidades ofrecidas por las ve- las 3DL y las nuevas herramientas de cálculo, Black Magic ganó la regata de 1995 por un contundente 5– 0 ante un impotente Young America tripulado por la tripulación del Stars & Stripes. No fue hasta dos ediciones posteriores, en 2003, cuando nuevamente la tecnología (respaldada por supuesto por un gran talento a bordo de la embarca- ción) permitió en este caso a una nación sin apenas tradición náutica, como es Suiza, conquistar la Copa del América con el sindicato Alinghi. Para entonces los métodos numéricos ya habían cobrado gran acep- tación como herramienta de diseño en un entorno tan exigente y los suizos demostraron gran habili- dad en utilizarla en sus estudios para el diseño de unas máquinas que se demostraron netamente supe- riores a las de sus contrincantes tanto en 2003 como en 2007[3]. En su artículo de 2005, Parolini y Quar- New Zealand held the cup for another edi- tion until in 2003 technology again helped an- other nation win the America’s Cup. Switzer- land, being a landlocked country, doesn’t have a nautical tradition. Nevertheless, its Alinghi sindicate did count with great talent onboard. By that time, numerical methods were already A.1. THE AMERICA’S CUP AS A SPORTS AND TECHNOLOGICAL PINNACLE 203 teroni [98] detallan parte del abanico de herramien- tas numéricas utilizadas para el análisis y el diseño de los barcos del sindicato Alinghi. Queda claro, a la vista de esta evolución, que los ensayos a escala en canal y a escala real han dejado paso a los méto- dos numéricos y las simulaciones por ordenador. widely accepted as a design tool in such a demanding environment. The swiss showed their ability in using it to design a two boat program that proved to be superior to their ri- vals’ both in 2003 and in 2007[3]. In their 2005 paper, Parolini and Quarteroni [98] explain some of the numerical tools used in the design and analysis of the Alinghi syndicate boats. Watching this evolution it becomes clear that reduced and full scale tow tank tests have given way to numerical methods and computer simulations. Así pues, ¿cuáles son los desarrollos tecnológicos de análisis numérico que han tenido lugar en las edi- ciones recientes de la Copa del América? Los enume- ro a continuación de manera muy resumida. La revolución computacional p El principal motivo por el que previamente no se ha- bían producido avances signiﬁcativos en materia de métodos numéricos aplicados al mundo de la vela en general, y en particular a la máxima competición, es el altísimo coste computacional que tenían los méto- dos numéricos o directamente la inexistencia de re- cursos para realizar los cálculos con anterioridad a la década de 1980. Resulta extraño hablar en estos términos de coste computacional en la actualidad en que un teléfono móvil que pesa apenas 200 gramos y cuesta alrededor de $100 tiene la misma capaci- dad computacional que un superordenador de ha- ce 30 años, ocupaba una sala entera y costaba unos cuantos miles de $. Por este mismo motivo, los méto- dos numéricos estaban todavía poco validados y la comunidad cientíﬁca todavía se apoyaba principal- mente en ensayos de laboratorio para realizar sus di- seños y avances tecnológicos. Valga como referencia el artículo [79] de Jerome Milgram en el que descri- be las diversas tecnologías desarrolladas y aplicadas en el diseño de los barcos del sindicato America3. En dicho artículo y en la discusión que lo sucede, Mil- gram explica la controvertida relación existente to- davía en 1992!!! entre las técnicas numéricas y las experimentales. So then, which have been the technological advances in numerical analysis that have oc- cured in the more recent editions of the Amer- ica’s Cup? I’ll describe them succintly below. Los VPP U d l Una de las aportaciones de Milgram en el artícu- lo mencionado es el papel central que adoptaron en aquella época (y todavía hasta hoy) los programas de predicción de velocidad o VPP (en sus siglas en inglés). Los VPP son sencillos programas de orde- nador que resuelven las ecuaciones de equilibrio de fuerzas en algunos de los ejes del espacio para un velero. Sostengo que son sencillos porque resuelven pequeños sistemas de ecuaciones —en algunos casos hasta 6 ecuaciones—. Los VPP entrañan un cierto grado de diﬁcultad debido a que las ecuaciones son 6Para una explicación detallada acerca del funcionamiento de los programas VPP véase el artículo [90] de Van Oossanen. Para una revisión de su utilización conjuntamente con las distintas técnicas de estimación de las fuerzas que actúan en un barco de vela, véase el artículo de DeBord et al. [59]. 7El profesor Idelsohn alterna su dedicación a CIMNE y al CIMEC en Santa Fe, Argentina. 6A detailed explanation of VPP programs can be found in the paper [90] by Van Oossanen. A review of its usage along with an estimation of the forces acting on a sailboat can be found in the paper [59] by DeBord et al. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN 204 a sailboat. I make the statement that these programs are simple because they solve small systems of equations —at most 6 equations—. VPPs entail some degree of difﬁculty because the equations are non-linear and the terms used to ﬁll the different components of the forces are in many cases semi-empirical, and this might cause some inconsistencies, which in turn leads to lack of convergence or an ex- cess of solutions in the design space [79].6 Ac- cording to Milgram, VPPs facilitated work- ing out different designs and evauate which performed better and which performed worse. However, the key for a good VPP is still the estimation of the forces —and their different components— acting on a boat. This is a sci- ence still under development today. no-lineales y los términos que sirven para alimentar las distintas componentes de las fuerzas son en mu- chos casos semi-empíricas, por lo que puede llegar a producirse alguna inconsistencia, lo que conduce a veces a una ausencia de convergencia o a un exce- so de soluciones en el espacio de diseño [79].6 Según explica Milgram, los VPP permitían barajar con ra- pidez varios diseños y evaluar cuáles eran mejores y cuáles peores. Sin embargo, la clave para un buen VPP sigue siendo la evaluación de las fuerzas —y sus distintas componentes— que actúan sobre una embarcación y esa es una ciencia que sigue avanzan- do en la actualidad. Avances en mecánica de ﬂuidos ﬂ Las herramientas disponibles hoy en día superan en gran medida las capacidades de las disponibles hace 20 años. En gran medida los avances han sido posi- bles por el incremento de la capacidad computacio- nal y la accesibilidad a la misma. Pero por el otro la- do, las técnicas numéricas también han experimen- tado mejoras signiﬁcativas. Baste ver los resultados publicados por cientíﬁcos de CIMNE; algunas de cu- yas herramientas se utilizaron en el desarrollo de los barcos del sindicato Alinghi [28]. En este sentido nos limitamos a citar un documento recopilatorio de la intensa actividad de los cientíﬁcos de CIMNE7 en el ámbito del CFD [97]. También es pertinente citar los recientes resultados publicados por Julio García en materia de validación de los métodos numéricos CFD frente a resultados experimentales en canales de ensayo para cascos de embarcaciones [47]. Si bien los avances realizados por los cientíﬁcos son muy no- tables en la última década, es pertinente comentar que los equipos que compiten en la Copa del América se han decantado progresivamente por el uso de he- rramientas comerciales que aportan al mercado las técnicas más avanzadas posible pero manteniendo la robustez necesaria para no comprometer los tiempos de desarrollo necesarios para el diseño de las embar- caciones. Ilustraremos esta aﬁrmación con dos refe- rencias. Si en 2001 Jones y Korpus [58] reportaban el uso por primera vez en la Copa del América de un 7Professor Idelsohn alternated his dedication between CIMNE and CIMEC in Santa Fe, Argentina. VPPs Amongst Milgram’s contributions in the refer- enced paper there is the pivotal role of veloc- ity prediction programs or VPPs in that epoch (and still today). VPPs are simple computer programs that solve the equilibrium equations in some of the degrees of freedom in space of APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN 6A detailed explanation of VPP programs can be found in the paper [90] by Van Oossanen. A review of its usage along with an estimation of the forces acting on a sailboat can be found in the paper [59] by DeBord et al. 7Professor Idelsohn alternated his dedication between CIMNE and CIMEC in Santa Fe, Argentina. 8Reynolds-Averaged Navier-Stokes 9La herramienta desarrollada por Michael Richelsen para North Sails recibe el nombre de FlowTM. Igual a como ocurre con el programa MemBrain, no he conseguido obtener referen- cias que describan el funcionamiento de Flow, aunque mediante entrevistas con diseñadores de North Sails he podido conﬁrmar que se trata de un software que resuelve las ecuaciones de ﬂujo potencial utilizando un método de paneles. De modo similar a co- mo sucede con MemBrain, al no existir evidencia documentada sobre el funcionamiento de Flow, no se puede considerar estado del arte y lo consideraremos estado de la práctica. 10ANSYS ha sustituido la herramienta GAMBIT por el pro- ducto Design Modeler. 8Reynolds-Averaged Navier-Stokes 9The tool developed by Michael Richelsen for North Sails is called FlowTM. Likewise MemBrain, I haven’t obtained references describing the way Flow works, but through interviews with North Sails’ designers I have con- ﬁrmed that it is a software program that solves the poten- tial ﬂow equation using a panel method. Therefore, as with MemBrain, in lacking written evidence on the func- tioning of Flow, it can’t be considered state of the art and I will consider it state of practice. 10ANSYS has changed the tool GAMBIT by the product Design Modeler. Simulación de velas en rumbos de ceñida y portantes y p El desarrollo del programa MemBrain por parte de North Sails abrió la puerta a subsiguientes desarro- llos para mejorar el análisis del rendimiento de las velas. La situación práctica de monopolio de North Sails propicia que todos estos desarrollos se hayan desarrollado o bien en el seno de North Sails o bien en colaboración con otras entidades de investiga- ción universitarias. Los primeros desarrollos tuvie- ron por objetivo la simulación de las velas y sus ar- boladuras en rumbos de ceñida. Esto es debido a que la tecnología CFD madura y disponible para ello fue en primer lugar la de simulación de ﬂujo potencial.9 Asimismo, en la década de 1990 surgieron múlti- ples técnicas de interacción ﬂuido-estructura. Aun- que se han ido mejorando, estas técnicas ya permi- tían obtener resultados muy útiles para los diseña- dores. En [106] Razenbach (del Grupo de Diseño de Velas Quantum) y Xu explican el proceso de simula- ción combinando estas técnicas para rumbos de ce- ñida. North Sails ha promovido el desarrollo de lo que han denominado como Túnel de Viento Virtual, que no es más que el acoplamiento de MemBrain con Fluent facilitado por la herramienta GAMBIT10 pa- ra poder analizar el rendimiento de las velas en rum- bos portantes [53]. Sails simulation in upwind and downwind courses The development of the MemBrain program by North Sails opened the door to ensuing de- velopments to improve the performance anal- ysis of sails. The practical monopoly of North Sails in the America’s Cup facilitates that all of these developments happened either within North Sails or in collaboration with other re- search universities. First, the developments aimed at the simulation of the rigs upwind. This is because the earliest available and ma- ture CFD techology was the simulation of po- tential ﬂow using the panel method.9 Also, in the 1990s several ﬂuid-structure interac- tion techniques emerged. Although they have improved, these techniques were already use- ful to designers. In [106] Razenbach (from the Quantum Sail Design Group) and Xu de- scribe the simulation procedure using these techniques for upwind courses. North Sails has promoted a so called Virtual Wind Tunnel, which is nothing else than coupling MemBrain with Fluent using the tool GAMBIT10 in order to analyze sails performance downwind [53]. 10ANSYS ha sustituido la herramienta GAMBIT por el pro- ducto Design Modeler. Advances in ﬂuid mechanics Today’s technologies are superior to those available 20 years ago. The advances have been possible largely thanks to the improve- ment of the computational capabilities and the accessibility thereof. Besides, numerical tech- niques have also improved signifﬁcantly. It sufﬁces taking a look at the results published by CIMNE researchers; some of their tools were used in the development of the boats of the Alinghi syndicate [28]. I will just cite a compilatory reference of the intense activ- ity of CIMNE scientists7 in the ﬁeld of CFD [97]. It is also convenient citing recent re- sults published by Julio Garcia in the topic of validation of CFD tools comparing with tow tank test results of boat hulls [47]. While the advances achieved by scientists in the last decade are notable, it is appropriate saying that the teams competing in the America’s Cup have progressively used more and more those commercial tools that bring to the mar- ket the most advanced techniques while main- taining the robustness required to avoid com- promising the short development times of the boats’ development. Let’s use two references to A.1. THE AMERICA’S CUP AS A SPORTS AND TECHNOLOGICAL PINNACLE 205 modelo RANS8 para cálculos CFD; Viola et al. [137] informan que en 2003 todavía eran pocos los equi- pos que utilizaron este tipo de simulaciones, y que en 2007 casi todos los equipos lo hacían. Y lo hicieron utilizando software comercial como Fluent, a dife- rencia del código FANS que en el año 2000 tenía un carácter marcadamente más experimental. further discuss this idea. While in 2001 Jones and Korpus [58] documented the use for the ﬁrst time in the America’s Cup of a RANS8 model for CFD calculations, Viola et al. [137] report that in 2003 there were still few teams using these simulations, and that in 2007 al- most all of them did. And they did so using a commercial software like Fluent, unlike the FANS code that in the year 2000 had a marked experimental character. 8Reynolds-Averaged Navier-Stokes 9 10ANSYS has changed the tool GAMBIT by the product Design Modeler. 11Even though this section uses the America’s Cup as the main reference in the development of sailing sport, on what relates structural analysis another competition takes precedence. This is because the structural integrity of the boats is fundamental when the competitors ﬁnd themselves thousands of miles away of any other external help when racing around the world. 11Aunque esta sección utiliza la Copa del América como refe- rente fundamental del desarrollo en el deporte de la vela, en este apartado referente al cálculo estructural cobra mayor importan- cia otra competición como referente principal. Esto es debido a que la ﬁabilidad estructural de los barcos es primordial cuando los competidores se encuentran a miles de kilómetros de cual- quier fuente de ayuda externa mientras navegan alrededor del mundo. Structural analysis La adopción de los métodos numéricos en la Copa del América tuvo lugar antes en el campo del cálculo Adoption of numerical methods in the Amer- ica’s Cup occurred earlier applied to the struc- APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN Para una revi- sión exhaustiva sobre métodos de cálculo estructura- les en veleros, es recomendable el material publicado por el 17º International Ship and Offshore Structu- res Congress [115]. En este documento se detalla de manera exhaustiva el estado del arte relativo a las técnicas de fabricación y de análisis de: cascos, más- tiles y apéndices de barcos de vela. Sin embargo, a pesar de que a la vista de este informe parece que los métodos de análisis y cálculo de las estructuras de los veleros están muy consolidadas, sorprende com- probar que los barcos de competición diseñados para dar la vuelta al mundo como en la Volvo Ocean Ra- ce11 siguen sufriendo fallos catastróﬁcos [14, 32, 83, 84]. De hecho, una de las características comunes de los barcos vencedores en las máximas competiciones es su ﬁabilidad en competición [76]. Es decir, la au- sencia de roturas que les obliguen a retirarse de la competición. Este es un fenómeno que no es exclusivo del deporte de la vela, sino que está presente en cual- quier deporte. Por ello, cobra máxima importancia la capacidad de analizar estructuralmente la respues- ta de los barcos de vela de competición, pues alige- rarlos es una necesidad imperiosa para hacerlos ve- loces; pero obra directamente en contra de la ﬁabili- dad. Encontrar el punto óptimo de dimensionamien- to es fundamental. Cientíﬁcos de CIMNE también han realizado aportaciones relevantes [88, 89, 105] que podrían permitir mejorar los cálculos de las es- tructuras de materiales compuestos que se utilizan en la construcción de los barcos de vela. estructural de los cascos de las embarcaciones que en el resto de áreas comentadas anteriormente (cálculo CFD de casco y velas, y análisis estructural de ve- las). Este fenómeno tuvo lugar de manera paralela a la incorporación del uso de los plásticos reforza- dos con ﬁbras en la fabricación de los barcos en la Copa América. En este caso se adoptaron tanto los métodos de fabricación como de cálculo utilizados en la industria aeronáutica [52, 75]. Para una revi- sión exhaustiva sobre métodos de cálculo estructura- les en veleros, es recomendable el material publicado por el 17º International Ship and Offshore Structu- res Congress [115]. En este documento se detalla de manera exhaustiva el estado del arte relativo a las técnicas de fabricación y de análisis de: cascos, más- tiles y apéndices de barcos de vela. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN Sin embargo, a pesar de que a la vista de este informe parece que los métodos de análisis y cálculo de las estructuras de los veleros están muy consolidadas, sorprende com- probar que los barcos de competición diseñados para dar la vuelta al mundo como en la Volvo Ocean Ra- ce11 siguen sufriendo fallos catastróﬁcos [14, 32, 83, 84]. De hecho, una de las características comunes de los barcos vencedores en las máximas competiciones es su ﬁabilidad en competición [76]. Es decir, la au- sencia de roturas que les obliguen a retirarse de la competición. Este es un fenómeno que no es exclusivo del deporte de la vela, sino que está presente en cual- quier deporte. Por ello, cobra máxima importancia la capacidad de analizar estructuralmente la respues- ta de los barcos de vela de competición, pues alige- rarlos es una necesidad imperiosa para hacerlos ve- loces; pero obra directamente en contra de la ﬁabili- dad. Encontrar el punto óptimo de dimensionamien- to es fundamental. Cientíﬁcos de CIMNE también han realizado aportaciones relevantes [88, 89, 105] que podrían permitir mejorar los cálculos de las es- tructuras de materiales compuestos que se utilizan en la construcción de los barcos de vela. tural calculation of the hulls of the boats than any other aspect commented before (CFD cal- culation of hulls and sails, or structural analy- sis of sails). This phenomenon happened along the transition towards the use of ﬁber rein- forced plastics in the construction of the Amer- ica’s Cup boats. For that, the manufactur- ing and calculation methods stemming from the aeronatical industry were used [52, 75]. For an in depth review on structural calcula- tion methods used in sailboats the author rec- comends the material published by the 17th International Ship and Offshore Structures Congress [115]. This document describes with great deal of detail the state of the art re- garding the manufacturing and analysis tech- niques of: hulls, masts and appendages of sail- boats. Nevertheless, even though according to that report it might seem that the meth- ods to calculate and analyze sailboat’s struc- tures are well established, it’s surprising real- ize that the high performance boats designed to race around the world in regattas such as the Volvo Ocean Race11 keep suffering catas- trophic failures [14, 32, 83, 84]. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN 206 tural calculation of the hulls of the boats than any other aspect commented before (CFD cal- culation of hulls and sails, or structural analy- sis of sails). This phenomenon happened along the transition towards the use of ﬁber rein- forced plastics in the construction of the Amer- ica’s Cup boats. For that, the manufactur- ing and calculation methods stemming from the aeronatical industry were used [52, 75]. For an in depth review on structural calcula- tion methods used in sailboats the author rec- comends the material published by the 17th International Ship and Offshore Structures Congress [115]. This document describes with great deal of detail the state of the art re- garding the manufacturing and analysis tech- niques of: hulls, masts and appendages of sail- boats. Nevertheless, even though according to that report it might seem that the meth- ods to calculate and analyze sailboat’s struc- tures are well established, it’s surprising real- ize that the high performance boats designed to race around the world in regattas such as the Volvo Ocean Race11 keep suffering catas- trophic failures [14, 32, 83, 84]. Indeed, one of the common facts of the winning boats in the top competitions is their reliability in com- petition [76]. That is, the lack of breakages forcing them to retire from the race. This phe- nomenon is not exclusive of sailing. It happens in any discipline. That’s why it is so impor- tant the capability to analyze the structural response of high performance sailboats. Mak- ing them light is an ever increasing need to make them fast, but it works directly against their reliability. Finding the optimal compro- mise in design is fundamental. CIMNE re- searchers have also made signiﬁcant contri- butions in this ﬁeld [88, 89, 105] that could improve the analysis of composite materials structures used in the construction of sail- boats. estructural de los cascos de las embarcaciones que en el resto de áreas comentadas anteriormente (cálculo CFD de casco y velas, y análisis estructural de ve- las). Este fenómeno tuvo lugar de manera paralela a la incorporación del uso de los plásticos reforza- dos con ﬁbras en la fabricación de los barcos en la Copa América. En este caso se adoptaron tanto los métodos de fabricación como de cálculo utilizados en la industria aeronáutica [52, 75]. Sumario It’s very telling in light of the review per- formed in the present chapter that the con- cepts used nowadays for the analysis on the performance of a sail boat (VPP programs, de- composition of the water resistance to a sail- boat into different components, analysis of sails as membranes, stationary simulations, etc.) are the same that were proposed more than 20 years ago. This is just the evidence of the burden of heritage and the importance it exerts on the natural process of humans to acquire new knowledge. But once this under- standing has been achieved, it would consti- tute an enormous mistake to consider that the big problem is the sum of the different com- ponents into which the mind has broken down the problem and built artiﬁcially. In reality the big problem has the added complexity of understanding how all the little components interact. Resulta muy reveladora la revisión realizada y ver que los conceptos con los que se trabaja actualmente en materia de análisis del rendimiento de un velero (programas VPP, descomposición de la resistencia al avance de un buque en distintas componentes, aná- lisis de las velas como membranas, simulaciones es- tacionarias, etc.) son los mismos que ya se proponían hace más de 20 años. Esto no es más que una mues- tra del peso que tiene la herencia y de la importancia que tiene para el conocimiento humano el dividir los problemas grandes en partes más pequeñas. Forma parte del proceso natural del ser humano para ad- quirir conocimiento, el dividir un problema grande en problemas más pequeños para poder comprender mejor su complejidad. Pero una vez alcanzada esa comprensión sería un error considerar que el proble- ma grande es la suma de las pequeñas componen- tes que la mente ha construido artiﬁcialmente. En realidad el problema grande tiene una complejidad añadida que es cómo todas esas componentes inter- accionan entre sí. The problem of analyzing a sailboat is not an exception, as Milgram points out [80, pp. 618–619]. This author stands ﬁrmly on that statement and radicalizes it, if that’s pos- sible, extrapolating it beyond the ﬁeld of hy- drodynamics and the hull. The goal of this point of view is to effectively take into account the interaction between all the boat parts and the external forces: both aerodynamic and hy- drodynamic. APPENDIX A. THE USE OF NUMERICAL METHODS IN SAILBOATS DESIGN Indeed, one of the common facts of the winning boats in the top competitions is their reliability in com- petition [76]. That is, the lack of breakages forcing them to retire from the race. This phe- nomenon is not exclusive of sailing. It happens in any discipline. That’s why it is so impor- tant the capability to analyze the structural response of high performance sailboats. Mak- ing them light is an ever increasing need to make them fast, but it works directly against their reliability. Finding the optimal compro- mise in design is fundamental. CIMNE re- searchers have also made signiﬁcant contri- butions in this ﬁeld [88, 89, 105] that could improve the analysis of composite materials structures used in the construction of sail- boats. A.2. SUMMARY 207 Sumario The ﬁnal result of this point of view would provide us with the possibility of analyzing the whole boat as a structure inte- grated in its environment and stop assuming that the hull is a rigid solid in transit through the water and the waves, or that the spars and the sails have a stationary position while the boat sails. El problema del estudio y análisis de un barco de vela no es una excepción y así lo apunta Milgram en [80, pp. 618–619]. Esta tesis se posiciona ﬁrme- mente sobre esta aﬁrmación y la radicaliza, si cabe, extrapolándola más allá del campo hidrodinámico que afecta al casco de la embarcación. El objetivo de este enfoque es considerar de manera efectiva la in- teracción de todas las partes del barco entre ellas, así como con las fuerzas externas: tanto las aerodi- námicas como las hidrodinámicas. El culmen de es- te enfoque nos llevaría a analizar todo el barco como una estructura integrada en su entorno, y no asu- mir que el casco es un sólido rígido en su tránsito a través del agua y de las olas, o que la arboladura y las velas tienen una posición estacionaria mientras el barco navega. On the interpolation of normal vectors for triangle meshes Internationa Journal for Numerical Methods in Engineering ATTENTION¡¡ Pages 210 to 232 of the thesis, which contain the article, are available on the editor’s web https://onlinelibrary.wiley.com/doi/abs/10.1002/nme.4567 209 The coordinates of the control points are: The coordinates of the control points are: The coordinates of the control points are: ⃗b0 = (x0, y0) = (0, 0) ⃗b1 = (x1, y1) = (Ψij, Ψij · tan ϕij) ⃗b2 = (x2, y2) = (1 −Ψji, Ψji · tan ϕji) ⃗b3 = (x3, y3) = (1, 0) (C.5) (C.5) Using the de Casteljau algorithm and Bernstein functions, the derivative of ⃗r can be expressed in the following way as a function of the curve pa- rameter t. Using the de Casteljau algorithm and Bernstein functions, the derivative of ⃗r can be expressed in the following way as a function of the curve pa- rameter t. ⃗r′(t) = (r′ x, r′ y) = (⃗b1 −⃗b0) · (1 −t)2 + (⃗b2 −⃗b1) · 2t(1 −t) + (⃗b3 −⃗b2) · t2 (C.6) ⃗r′(t) = (r′ x, r′ y) = (⃗b1 −⃗b0) · (1 −t)2 + (⃗b2 −⃗b1) · 2t(1 −t) + (⃗b3 −⃗b2) · t2 (C (C.6) r′ x = x1 · (1 −t)2 + (x2 −x1) · 2t(1 −t) + (1 −x2) · t2 r′ y = y1 · (1 −t)2 + (y2 −y1) · 2t(1 −t) + y2 · t2 (C.7) c2 = r′2 x + r′2 y (C.8) (C.7) (C.8) Considering the values of ϕij and ϕji ﬁxed. And maintaining Ψij and Ψji as variables, then c2 = c2(Ψij, Ψji, t) (C.9) (C.9) The total membrane elastic energy in the curve can be expressed as: E = Z 1 0 U dt ⇒E ∝ Z 1 0 c2 dt (C.10) (C.10) The author is interested in minimizing the variation of the energy density along the curve. In order to do that, it is useful to minimize the square of the variation of the energy density along the curve: The author is interested in minimizing the variation of the energy density along the curve. Minimization of the membrane (axial) energy of a 2D cubic Bernouilli beam T HE PRESENT APPENDIX DEVELOPS THE FORMULÆ introduced in sec- tion 5.4.1 on page 59 to discuss the problem of minimizing the vari- ation of membrane energy of the edge of the curved shell triangle in the construction process. T H Figure C.1: 2D cubic Bézier curve representing the edge of a cubic Bézier triangle. For the purposes of the exercise and without loss of generality, the distance between the endpoints of the curve will be 1. Figure C.1: 2D cubic Bézier curve representing the edge of a cubic Bézier triangle. For the purposes of the exercise and without loss of generality, the distance between the endpoints of the curve will be 1. Let’s state again the problem: if the membrane strain (ε) of the curve depicted in ﬁgure C.1 is proportional to the celerity (c) of the curve param- eter (t), then the energy density (U) is proportional to the square of the celerity (c2). ⃗r(t) = (x(t), y(t)) (C.1) ⃗r′(t) = ∂⃗r(t) ∂t (C.2) c = ∥⃗r′(t)∥ (C.3) ε ∝c ⇒U ∝ε2 ∝c2 (C.4) The above relationship assumes that the material is linear elastic. The above relationship assumes that the material is linear elastic. 233 APPENDIX C. MINIMIZATION OF THE MEMBRANE ENERGY 234 The coordinates of the control points are: The coordinates of the control points are: In order to do that, it is useful to minimize the square of the variation of the energy density along the curve: Z 1 0 Å∂U ∂t ã2 dt (C.11) (C.11) The minimization is carried out by differentiating the above expression with respect to the two variables of the problem Ψij and Ψji: ∂ ∂Ψij Z 1 0 Å∂U ∂t ã2 dt = 0 ∂ ∂Ψji Z 1 0 Å∂U ∂t ã2 dt = 0          (C.12) (C.12) Since U ∝c2, the roots of the above system of equations are the same as the roots of ∂ ∂Ψij Z 1 0 Å∂c2 ∂t ã2 dt = 0 ∂ ∂Ψji Z 1 0 Å∂c2 ∂t ã2 dt = 0          (C.13) (C.13) The expressions in equation (C.13) are polynomials of 4th order with re- spect to the independent variables Ψij and Ψji. Their graphical solution is presented in ﬁgure 5.9 on page 61. The expressions in equation (C.13) are polynomials of 4th order with re- spect to the independent variables Ψij and Ψji. Their graphical solution is presented in ﬁgure 5.9 on page 61. Appendix D Analytic derivatives of some complex expressions T HIS APPENDIX INCLUDES THE ANALYTIC EXPRESSIONS that are not relevant for the understanding of the development of the formulæ in the thesis, or too cumbersome to be included in the main body of the text. Note that Einstein’s indicial notation is in full effect. T H D.1 Derivatives of the element normal The deﬁnition of n can be found in equation (5.12) on page 49. Its deriva- tives with respect to the control points are presented next. ∂n ∂p = ∂n(i) ∂p(mn) = (e(ikm) −n(i) · e(jkm) · n(j)) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n)) ∥(p · Lξ) × (p · Lη)∥ (D.1) ∂n ∂p = ∂n(i) ∂p(mn) = (e(ikm) −n(i) · e(jkm) · n(j)) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n)) ∥(p · Lξ) × (p · Lη)∥ (D.1) ∂2 (D.1) ∂2n(i) ∂p(mn)∂p(vw) = 1 ∥(p · Lξ) × (p · Lη)∥· (e(imv) + n(i) · e(mjv) · n(j))· ·(Lξ(n) · Lη(w) −Lη(n) · Lξ(w))− −∂n(i) ∂p(mn) · e(jkv) · n(j) · p(kl) · (Lξ(l) · Lη(w) −Lη(l) · Lξ(w))− −∂n(i) ∂p(vw) · e(jkm) · n(j) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n))− − n(i) · p(jl) · p(ko) ∥(p · Lξ) × (p · Lη)∥· (e(tjm) · e(tkv) + e(tjm) · n(t) · e(xkv) · n(x))· ·(Lξ(l) · Lη(n) −Lη(l) · Lξ(n)) · (Lξ(o) · Lη(w) −Lη(o) · Lξ(w)) (D.2) Also, the higher derivatives of the expressions deﬁned in equa- tions (6.22) and (6.23) on page 69 are presented below. ∂n,ξα ∂p = ∂n,ξα(i) ∂p(mn) = (δ(ij) −n(i) · n(j)) · Ç ∂Lpξ,ξα(j) ∂p(mn) + ∂Lpη,ξα(j) ∂p(mn) å − − Ç ∂n(i) ∂p(mn) · n(j) + n(i) · ∂n(j) ∂p(mn) å · (Lpξ,ξα(j) + Lpη,ξα(j)) ∀α = {1, 2} (D.3) (D.3) 235 APPENDIX D. ANALYTIC DERIVATIVES 236 ∂2n,ξα (∂p)2 = ∂2n,ξα(i) ∂p(mn)∂p(vw) = = (δ(ij) −n(i) · n(j)) · Ç ∂2Lpξ,ξα(j) ∂p(mn)∂p(vw) + ∂2Lpη,ξα(j) ∂p(mn)∂p(vw) å − − Ç ∂2n(i) ∂p(mn)∂p(vw) · n(i) + ∂n(i) ∂p(mn) · ∂n(j) ∂p(vw) + + ∂n(i) ∂p(vw) · ∂n(j) ∂p(mn) + n(i) · ∂2n(j) ∂p(mn)∂p(vw) å · Lpξ,ξα(j) + Lpη,ξα(j) − − Ç ∂n(i) ∂p(vw) · n(j) + n(i) · ∂n(j) ∂p(vw) å · Ç ∂Lpξ,ξα(j) ∂p(mn) + ∂Lpη,ξα(j) ∂p(mn) å − − Ç ∂n(i) ∂p(mn) · n(j) + n(i) · ∂n(j) ∂p(mn) å · Ç ∂Lpξ,ξα(j) ∂p(vw) + ∂Lpη,ξα(j) ∂p(vw) å ∀α = {1, 2} (D.4) (D.4) (D.4) (D.4) These expressions are used in equations (7.17) to (7.21) on pages 79–80. These expressions are used in equations (7.17) to (7.21) on pages 79–80. D.1 Derivatives of the element normal ∂2Lpξ,ξα(i) ∂p(mn)∂p(vw) = 1 ∥(p · Lξ) × (p · Lη)∥· · e(imv) · (Lξ,ξα(n) · Lη(w) −Lη(n) · Lξ,ξα(w))− −∂Lpξ,ξα(i) ∂p(mn) · n(j) · e(jkv) · p(kl) · (Lξ(l) · Lη(w) −Lη(l) · Lξ(w))− −∂Lpξ,ξα(i) ∂p(vw) · n(j) · e(jkm) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n))− −Lpξ,ξα(i) · ∂n(j) ∂p(mn) · e(jkv) · p(kl) · (Lξ(l) · Lη(w) −Lη(l) · Lξ(w))− −Lpξ,ξα(i) · n(j) · e(jmv) · (Lξ(n) · Lη(w) −Lη(n) · Lξ(w)) ∀α = {1, 2} (D.9) −Lpξ,ξα(i) · ∂n(j) ∂p(mn) · e(jkv) · p(kl) · (Lξ(l) · Lη(w) −Lη(l) · Lξ(w))− Lp n e (L L L L ) ∀α {1 2} } (D.9) ∂2Lpη,ξα(i) ∂p(mn)∂p(vw) = 1 ∥(p · Lξ) × (p · Lη)∥· · e(imv) · (Lξ(n) · Lη,ξα(w) −Lη,ξα(n) · Lξ(w))− −∂Lpη,ξα(i) ∂p(mn) · n(j) · e(jkv) · p(kl) · (Lξ(l) · Lη(w) −Lη(l) · Lξ(w))− −∂Lpη,ξα(i) ∂p(vw) · n(j) · e(jkm) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n))− −Lpη,ξα(i) · ∂n(j) ∂p(mn) · e(jkv) · p(kl) · (Lξ(l) · Lη(w) −Lη(l) · Lξ(w))− −Lpη,ξα(i) · n(j) · e(jmv) · (Lξ(n) · Lη(w) −Lη(n) · Lξ(w)) ∀α = {1, 2} (D.10) D.1 Derivatives of the element normal Where Lpξ,ξα = (p · Lξ,ξα) × (p · Lη) ∥(p · Lξ) × (p · Lη)∥= e(ijk) · p(jl) · Lξ,ξα(l) · p(km) · Lη(m) ∥(p · Lξ) × (p · Lη)∥ = Lpξ,ξα(i) ∀α = {1, 2} (D.5) Lpη,ξα = (p · Lξ) × (p · Lη,ξα) ∥(p · Lξ) × (p · Lη)∥= e(ijk) · p(jl) · Lξ(l) · p(km) · Lη,ξα(m) ∥(p · Lξ) × (p · Lη)∥ = Lpη,ξα(i) ∀α = {1, 2} (D.6) (D.5) their ﬁrst derivatives are expressed in the following way, their ﬁrst derivatives are expressed in the following way, ∂Lpξ,ξα(i) ∂p(mn) = e(ijm) · p(jl) · (Lξ,ξα(l) · Lη(n) −Lη(l) · Lξ,ξα(n)) ∥(p · Lξ) × (p · Lη)∥ − −Lpξ,ξα(i) · e(jkm) · n(j) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n)) ∥(p · Lξ) × (p · Lη)∥ ∀α = {1, 2} (D.7) ∂Lpη,ξα(i) ∂p(mn) = e(ijm) · p(jl) · (Lξ(l) · Lη,ξα(n) −Lη,ξα(l) · Lξ(n)) ∥(p · Lξ) × (p · Lη)∥ − −Lpη,ξα(i) · e(jkm) · n(j) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n)) ∥(p · Lξ) × (p · Lη)∥ ∀α = {1, 2} (D.8) ∂p(mn) ∥(p Lξ) × (p Lη)∥ −Lpη,ξα(i) · e(jkm) · n(j) · p(kl) · (Lξ(l) · Lη(n) −Lη(l) · Lξ(n)) ∥(p · Lξ) × (p · Lη)∥ ∀α = {1 2} (D.8) (D.8) ∀α = {1, 2} D.2. DERIVATIVES OF THE SYSTEM MATRIX AND INDEPENDENT VECTOR TO COMPUTE THE CONTROL POINT LOCATIONS 237 and the expressions of their second derivatives are as follows. and the expressions of their second derivatives are as follows. D.2 Derivatives of the system matrix and independent vector to compute the control point locations The location of the control points is computed by solving a system of equa- tions for each one. These include the control points at the boundaries and the central control point. D.2.1 Boundary control points The general expression of the system of equations to determine the control points of the boundaries is presented in equation (5.15) on page 51. The expression of the derivatives of the system matrix Aij p is presented row by row. ∂Aij p(1l) ∂xh (v) = ∂ni (l) ∂xh (v) (D.11) ∂Aij p(2l) ∂xh (v) = ∂dij (l) ∂xh (v) (D.12) (D.11) (D.12) APPENDIX D. ANALYTIC DERIVATIVES 238 ∂Aij p(3l) ∂xh (v) = (δih−δjh) · δ(lv) (D.13) (D.13) The expression of the derivatives of the independent vector is also pre- sented component-wise. ∂bij p(1) ∂xh (v) = xi (l) · ∂ni (l) ∂xh (v) + δih · ni (v) (D.14) ∂bij p(2) ∂xh (v) = ∂dij (l) ∂xh (v) · xi (l) + δih · dij (v) (D.15) ∂bij p(3) ∂xh (v) = δih · (xi (v) −xj (v)) + (δih−δjh) · xi (v)− −∥xi −xj∥2 · ∂Ψij ∂xh (v) −2(δih−δjh) · (xi (v) −xj (v)) · Ψij (D.16) (D.14) (D.15) (D.16) (D.16) And the second derivatives are presented below. And the second derivatives are presented below. ∂2Aij p(1l) ∂xh (v)∂xs (w) = ∂2ni (l) ∂xh (v)∂xs (w) (D.17) ∂2Aij p(2l) ∂xh (v)∂xs (w) = ∂2dij (l) ∂xh (v)∂xs (w) (D.18) ∂2Aij p(2l) ∂xh (v)∂xs (w) = 0 (D.19) ∂2bij p(1) ∂xh (v)∂xs (w) = xi (l) · ∂2ni (l) ∂xh (v)∂xs (w) + δih · ∂ni (v) ∂xs (w) + δis · ∂ni (w) ∂xh (v) (D.20) ∂2bij p(2) ∂xh (v)∂xs (w) = ∂2dij (l) ∂xh (v)∂xs (w) · xi (l) + δis · ∂dij (w) ∂xh (v) + δih · ∂dij (v) ∂xs (w) (D.21) ∂2bij p(3) ∂xh (v)∂xs (w) = δ(vw) · [δih · (δis−δjs) + δis · (δih−δjh)− −2 · (δih−δjh) · (δis−δjs) · Ψij] −∥xi −xj∥2 · ∂2Ψij ∂xh (v)∂xs (w) − −2 · (δih−δjh) · (xi (v) −xj (v)) · ∂Ψij ∂xs (w) − −2 · (δis−δjs) · (xi (w) −xj (w)) · ∂Ψij ∂xh (v) (D 22) (D.22) (v) (D.22) The reader will notice that in order to implement these equations we need to develop the expressions of the derivatives of ni and dij. We will D.2. DERIVATIVES OF THE SYSTEM MATRIX AND INDEPENDENT VECTOR TO COMPUTE THE CONTROL POINT LOCATIONS 239 239 use the deﬁnitions provided in equations (5.3) and (9.7) on page 46 and on page 106, respectively. The expressions for Ψij and its derivatives are provided in equa- tions (9.17) to (9.32) on pages 112–116. D.2.2 Central control point candidates + +(pjk (m) −pkj (m)) · ∂2pjk (m) ∂xh (v)∂xs (w) − ∂2pkj (m) ∂xh (v)∂xs (w) !# · njk (l) (D.24) A0i p(2l) = Aij p(3l) ⇒ ∂A0i p(2l) ∂xh (v) = ∂Aij p(3l) ∂xh (v) ⇒ ∂2A0i p(2l) ∂xh (v)∂xs (w) = ∂2Aij p(3l) ∂xh (v)∂xs (w) (D.25) A0i p(3l) = Aik p(3l) ⇒ ∂A0i p(3l) ∂xh (v) = ∂Aik p(3l) ∂xh (v) ⇒ ∂2A0i p(3l) ∂xh (v)∂xs (w) = ∂2Aik p(3l) ∂xh (v)∂xs (w) (D.26) D.2.2 Central control point candidates The general expression of the system of equations to determine the cen- tral control point candidates is presented in equation (5.19) on page 54. The expression of the ﬁrst and second derivatives of the system matrix A0i p is presented row by row. However, only the ﬁrst row presents a chal- lenge, because the expressions for the second and third rows have already been solved in the case of the boundary control points; see equations (D.25) and (D.26). ∂A0i p(1l) ∂xh (v) = " ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · njk (m) + (pjk (m) −pkj (m)) · ∂njk (m) ∂xh (v) # · (pjk (l) −pkj (l))+ + [(pjk (m) −pkj (m)) · njk (m) · ∂pjk (l) ∂xh (v) − ∂pkj (l) ∂xh (v) ! − −2 · " (pjk (m) −pkj (m)) · ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) !# njk (l)− −(pjk (m) −pkj (m))2 · ∂njk (l) ∂xh (v) (D 23) (D.23) APPENDIX D. ANALYTIC DERIVATIVES 240 ∂2A0i p(1l) ∂xh (v)∂xs (w) = " ∂2pjk (m) ∂xh (v)∂xs (w) − ∂2pkj (m) ∂xh (v)∂xs (w) ! · njk (m) + ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · ∂njk (m) ∂xs (w) + + ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) ! · ∂njk (m) ∂xh (v) + (pjk (m) −pkj (m)) · ∂2njk (m) ∂xh (v)∂xs (w) # · (pjk (l) −pkj (l))+ +[(pjk (m)−pkj (m))·njk (m)]· ∂2pjk (l) ∂xh (v)∂xs (w) − ∂2pkj (l) ∂xh (v)∂xs (w) ! −(pjk (m)−pkj (m))2· ∂2njk (l) ∂xh (v)∂xs (w) + + " ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · njk (m) + (pjk (m) −pkj (m)) · ∂njk (m) ∂xh (v) # · ∂pjk (l) ∂xs (w) − ∂pkj (l) ∂xs (w) ! + + " ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) ! · njk (m) + (pjk (m) −pkj (m)) · ∂njk (m) ∂xs (w) # · ∂pjk (l) ∂xh (v) − ∂pkj (l) ∂xh (v) ! D.2.2 Central control point candidates − −2 · " (pjk (m) −pkj (m)) · ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) !# · ∂njk (l) ∂xs (w) − −2 · " (pjk (m) −pkj (m)) · ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) !# · ∂njk (l) ∂xh (v) − −2 · " ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) ! + +(pjk (m) −pkj (m)) · ∂2pjk (m) ∂xh (v)∂xs (w) − ∂2pkj (m) ∂xh (v)∂xs (w) !# · njk (l) (D.24) A0i p(2l) = Aij p(3l) ⇒ ∂A0i p(2l) ∂xh (v) = ∂Aij p(3l) ∂xh (v) ⇒ ∂2A0i p(2l) ∂xh (v)∂xs (w) = ∂2Aij p(3l) ∂xh (v)∂xs (w) (D.25) A0i p(3l) = Aik p(3l) ⇒ ∂A0i p(3l) ∂xh (v) = ∂Aik p(3l) ∂xh (v) ⇒ ∂2A0i p(3l) ∂xh (v)∂xs (w) = ∂2Aik p(3l) ∂xh (v)∂xs (w) (D.26) ∂2A0i p(1l) ∂xh (v)∂xs (w) = " ∂2pjk (m) ∂xh (v)∂xs (w) − ∂2pkj (m) ∂xh (v)∂xs (w) ! · njk (m) + ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · ∂njk (m) ∂xs (w) + + ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) ! · ∂njk (m) ∂xh (v) + (pjk (m) −pkj (m)) · ∂2njk (m) ∂xh (v)∂xs (w) # · (pjk (l) −pkj (l))+ +[(pjk (m)−pkj (m))·njk (m)]· ∂2pjk (l) ∂xh (v)∂xs (w) − ∂2pkj (l) ∂xh (v)∂xs (w) ! −(pjk (m)−pkj (m))2· ∂2njk (l) ∂xh (v)∂xs (w) + + " ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · njk (m) + (pjk (m) −pkj (m)) · ∂njk (m) ∂xh (v) # · ∂pjk (l) ∂xs (w) − ∂pkj (l) ∂xs (w) ! + + " ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) ! · njk (m) + (pjk (m) −pkj (m)) · ∂njk (m) ∂xs (w) # · ∂pjk (l) ∂xh (v) − ∂pkj (l) ∂xh (v) ! − −2 · " (pjk (m) −pkj (m)) · ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) !# · ∂njk (l) ∂xs (w) − −2 · " (pjk (m) −pkj (m)) · ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) !# · ∂njk (l) ∂xh (v) − −2 · " ∂pjk (m) ∂xh (v) − ∂pkj (m) ∂xh (v) ! · ∂pjk (m) ∂xs (w) − ∂pkj (m) ∂xs (w) ! D.3 Derivatives of the normal at the node Let’s use a general expression for a weighted average normal vector at the node. ni = ri X k=1 wk · ˆyk ri X k=1 wk · ˆyk (D.27) ni = ri X k=1 wk · ˆyk ri X k=1 wk · ˆyk (D.27) (D.27) D.3. DERIVATIVES OF THE NORMAL AT THE NODE 241 Then, the general expression for the ﬁrst derivative is Then, the general expression for the ﬁrst derivative is Then, the general expression for the ﬁrst derivative is = ri X k=1 Ç ∂wk ∂xh (v) ˆyk (j) + wk ∂ˆyk (j) ∂xh (v) å −ni (j) ni (l) · ri X k=1 Ç ∂wk ∂xh (v) ˆyk (l) + wk ∂ˆyk (l) ∂xh (v) å ri X k=1 wk · ˆyk (D 28) (D.28) (D.28) and for the second derivative and for the second derivative and for the second derivative ∂2ni (j) ∂xh (v)∂xs (w) = 1 ri X k=1 wk · ˆyk " ri X k=1 Ç ∂2wk ∂xh (v)∂xs (w) ˆyk (j) + ∂wk ∂xh (v) ∂ˆyk (j) ∂xs (w) + + ∂wk ∂xs (w) ∂ˆyk (j) ∂xh (v) + wk ∂2ˆyk (j) ∂xh (v)∂xs (w) å −ni (j) ∂ni (l) ∂xs (w) ri X k=1 Ç ∂wk ∂xh (v) ˆyk (l) + wk ∂ˆyk (l) ∂xh (v) å − − ∂ni (j) ∂xs (w) ni (l) ri X k=1 Ç ∂wk ∂xh (v) ˆyk (l) + wk ∂ˆyk (l) ∂xh (v) å − ∂ni (j) ∂xh (v) ni (l) ri X k=1 Ç ∂wk ∂xs (w) ˆyk (l) + wk ∂ˆyk (l) ∂xs (w) å − −ni (j) ni (l) ri X k=1 Ç ∂2wk ∂xh (v)∂xs (w) ˆyk (l) + ∂wk ∂xh (v) ∂ˆyk (l) ∂xs (w) + ∂wk ∂xs (w) ∂ˆyk (l) ∂xh (v) + wk ∂2ˆyk (l) ∂xh (v)∂xs (w) å (D.29) Figure D.1: This ﬁgure is a modiﬁcation of ﬁgure 5.1. A new notation is used for the neighboring nodes. Each node has a speciﬁc notation depend- ing to the triangle considered and the direction of its normal. Figure D.1: This ﬁgure is a modiﬁcation of ﬁgure 5.1. A new notation is used for the neighboring nodes. Each node has a speciﬁc notation depend- ing to the triangle considered and the direction of its normal. In the case of the weight proposed in equation (5.2) on page 46, we will write the derivatives after rewriting the formula using the nodal coordi- 242 APPENDIX D. ANALYTIC DERIVATIVES nates of the neighboring nodes. But ﬁrst we need to redeﬁne some nota- tion regarding the neighboring nodes. For each triangle k that surrounds node i, the two neighboring nodes associated will be named k+ and k− corresponding to looping them in a counterclockwise order as the normal points towards the observer (see ﬁgure D.1 which substitutes ﬁgure 5.1 on page 45). wk α/A◦= ∥vk∥2 (xk−−xi)2(xk+−xi)2(xk+−xk−)2 · arccos(cα) (D.30) (D.30) The ﬁrst derivative is presented below. The ﬁrst derivative is presented below. and for the second derivative ∂wk α/A◦ ∂xh (v) = −2·wk α/A◦· " (xk− (v) −xi (v)) (δk−h−δih) (xk−−xi)2 + (xk+ (v) −xi (v)) (δk+h−δih) (xk+−xi)2 + + (xk+ (v) −xk− (v)) (δk+h−δk−h) (xk+−xk−)2 # + 1 (xk−−xi)2 (xk+−xi)2 (xk+−xk−)2 · · ® 2 · vk (j) · ∂vk (j) ∂xh (v) · arccos(cα) − ∥vk∥2 p 1 −c2α · ∂cα ∂xh (v) ´ (D.31) And the second derivative is as follows. ∂2wk α/◦ ∂xh (v)∂xs (w) = 1 (xk−−xi)2 (xk+−xi)2 (xk+−xk−)2 · ñ 2 · Ç ∂vk (j) ∂xh (v) · ∂vk (j) ∂xs (w) + +vk (j) · ∂2vk (j) ∂xs (w) å · arccos(cα) − 2 · vk (j) √1 −cα · Ç ∂vk (j) ∂xh (v) · ∂cα ∂xs (w) + ∂vk (j) ∂xs (w) · ∂cα ∂xh (v) å − −∥vk∥2 p 1 −c2α · Ç ∂2cα ∂xh (v)∂xs (w) + cα 1 −c2α · ∂cα ∂xh (v) · ∂cα ∂xs (w) åô − −2·wk α/A◦· ñ δ(vw) · (δk−h−δih) · (δk−s−δis) (xk−−xi)2 + δ(vw) · (δk+h−δih) · (δk+s−δis) (xk+−xi)2 + + δ(vw) · (δk+h−δk−h) · (δk+s−δk−s) (xk+−xk−)2 + + 2 · (xk− (v) −xi (v)) · (δk−h−δih) · (xk+ (w) −xi (w)) · (δk+s−δis) (xk−−xi)2 · (xk+−xi)2 + + 2 · (xk− (v) −xi (v)) · (δk−h−δih) · (xk+ (w) −xk− (w)) · (δk+s−δk−s) (xk−−xi)2 · (xk+−xk−)2 + + 2 · (xk+ (v) −xi (v)) · (δk+h−δih) · (xk− (w) −xi (w)) · (δk−s−δis) (xk−−xi)2 · (xk+−xi)2 + + 2 · (xk+ (v) −xi (v)) · (δk+h−δih) · (xk+ (w) −xk− (w)) · (δk+s−δk−s) (xk+−xi)2 · (xk+−xk−)2 + + 2 · (xk+ (v) −xk− (v)) · (δk+h−δk−h) · (xk− (w) −xi (w)) · (δk−s−δis) (xk−−xi)2 · (xk+−xk−)2 + ∂2wk α/◦ ∂xh (v)∂xs (w) = 1 (xk−−xi)2 (xk+−xi)2 (xk+−xk−)2 · ñ 2 · Ç ∂vk (j) ∂xh (v) · ∂vk (j) ∂xs (w) + +vk (j) · ∂2vk (j) ∂xs (w) å · arccos(cα) − 2 · vk (j) √1 −cα · Ç ∂vk (j) ∂xh (v) · ∂cα ∂xs (w) + ∂vk (j) ∂xs (w) · ∂cα ∂xh (v) å − −∥vk∥2 p 1 −c2α · Ç ∂2cα ∂xh (v)∂xs (w) + cα 1 −c2α · ∂cα ∂xh (v) · ∂cα ∂xs (w) åô − −2·wk α/A◦· ñ δ(vw) · (δk−h−δih) · (δk−s−δis) (xk−−xi)2 + δ(vw) · (δk+h−δih) · (δk+s−δis) (xk+−xi)2 + + δ(vw) · (δk+h−δk−h) · (δk+s−δk−s) (xk+−xk−)2 + D.3. and for the second derivative DERIVATIVES OF THE NORMAL AT THE NODE 243 + 2 · (xk+ (v) −xk− (v)) · (δk+h−δk−h) · (xk+ (w) −xi (w)) · (δk+s−δis) (xk+−xi)2 · (xk+−xk−)2 # − −2 ∂wk α/A◦ ∂xs (w) · " (xk− (v) −xi (v)) (δk−h−δih) (xk−−xi)2 + (xk+ (v) −xi (v)) (δk+h−δih) (xk+−xi)2 + + (xk+ (v) −xk− (v)) (δk+h−δk−h) (xk+−xk−)2 # −2 ∂wk α/A◦ ∂xh (v) · " (xk− (w) −xi (w)) (δk−s−δis) (xk−−xi)2 + + (xk+ (w) −xi (w)) (δk+s−δis) (xk+−xi)2 + (xk+ (w) −xk− (w)) (δk+s−δk−s) (xk+−xk−)2 # (D.32) (D.32) Where cα corresponds to the cosine of the angle formed by the sides of the triangle at the vertex i (see ﬁgure D.1). cα = (xk−−xi) · (xk+−xi) ∥xk−−xi∥· ∥xk+−xi∥ (D.33) (D.33) ∂cα ∂xh (v) = (xk− (v) −xi (v)) · (δk+h−δih) + (xk+ (v) −xi (v)) · (δk−h−δih) ∥xk−−xi∥· ∥xk+−xi∥ − −cα · " (xk− (v) −xi (v)) · (δk−h−δih) (xk−−xi)2 + (xk+ (v) −xi (v)) · (δk+h−δih) (xk+−xi)2 # (D.34) (D.34) ∂2cα ∂xh (v)∂xs (w) = δ(vw) · (δk+h−δih) · (δk−s−δis) + (δk−h−δih) · (δk+s−δis) ∥xk−−xi∥· ∥xk+−xi∥ − −∂cα ∂xh (v) · " (xk− (w) −xi (w)) · (δk−s−δis) (xk−−xi)2 + (xk+ (w) −xi (w)) · (δk+s−δis) (xk+−xi)2 # − −∂cα ∂xs (w) · " (xk− (v) −xi (v)) · (δk−h−δih) (xk−−xi)2 + (xk+ (v) −xi (v)) · (δk+h−δih) (xk+−xi)2 # − cα · ® δ(vw) · ñ (δk−h−δih) · (δk−s−δis) (xk−−xi)2 + (δk+h−δih) · (δk+s−δis) (xk+−xi)2 ô + + (xk− (v) −xi (v)) · (δk−h−δih) · (xk+ (w) −xi (w)) · (δk+s−δis) (xk−−xi)2 · (xk+−xi)2 + + (xk+ (v) −xi (v)) · (δk+h−δih) · (xk− (w) −xi (w)) · (δk−s−δis) (xk−−xi)2 · (xk+−xi)2 − − (xk− (v) −xi (v)) · (δk−h−δih) · (xk− (w) −xi (w)) · (δk−s−δis) (xk−−xi)3/2 − − (xk+ (v) −xi (v)) · (δk+h−δih) · (xk+ (w) −xi (w)) · (δk+s−δis) (xk+−xi)3/2 ) (D.35 244 APPENDIX D. ANALYTIC DERIVATIVES And vk is the cross product of the vectors that link the sides of the triangle. ∂vk (j) ∂xh (v) = e(jmv) · î δk−h(xi (m) −xk+ (m)) + δih(xk+ m −xk− (m)) + δk+h(xk− (m) −xi (m)) ó (D.37) ∂2vk (j) ∂xh (v)∂xs (w) = e(jvw) · î δk−h(δk+s−δis) + δih(δk−s−δk+s) + δk+h(δis−δk−s) ó (D.38) (D.38) Then we conclude with the deﬁnition of ˆyk and its derivatives. and for the second derivative ˆyk = vk ∥vk∥ (D.39) ˆyk (j) ∂xh (v) = ∥vk∥−1 · Ç ∂vk (j) ∂xh (v) −ˆyk (j) · ˆyk (l) · ∂vk (l) ∂xh (v) å (D.40) ∂2ˆyk (j) ∂xh (v)∂xs (w) = ∥vk∥−1 · Ç ∂2vk (j) ∂xh (v)∂xs (w) − ∂ˆyk (j) ∂xh (v) · yk (l) · ∂vk (l) ∂xs (w) − − ∂ˆyk (j) ∂xs (w) · ˆyk (l) · ∂vk (l) ∂xh (v) −ˆyk (j) · ∂ˆyk (l) ∂xs (w) · ∂vk (l) ∂xh (v) −ˆyk (j) · ˆyk (l) · ∂2vk (l) ∂xh (v)∂xs (w) å (D.41) (D.39) D.4 Derivatives of dij DERIVATIVES OF θij 245 −e(lmk) ∂ni (m) ∂xs (w) (xi (k) −xj (k)) sin θij ∂θij ∂xh (v) −(δih−δjh) e(lmv) ni (m) sin θij ∂θij ∂xs (w) − −(δis−δjs) e(lmw) ni (m) sin θij ∂θij ∂xh (v) −e(lmk) ni (m) (xi (k)−xj (k)) cos θij ∂θij ∂xh (v) ∂θij ∂xs (w) − −e(lmk) ni (m) (xi (k)−xj (k)) sin θij ∂2θij ∂xh (v)∂xs (w) +(xi (m)−xj (m)) sin θij Ç ∂ni (l) ∂xh (v) ∂ni (m) ∂xs (w) + + ∂ni (l) ∂xs (w) ∂ni (m) ∂xh (v) + ni (l) ∂2ni (m) ∂xh (v)∂xs (w) + ∂2ni (l) ∂xh (v)∂xs (w) ni (m) å + + (δis−δjs) ∂ni (l) ∂xh (v) ni (w) sin θij + (δih−δjh) ∂ni (l) ∂xs (w) ni (v) sin θij+ + ni (m) (xi (m) −xj (m)) cos θij Ç ∂ni (l) ∂xh (v) ∂θij ∂xs (w) + ∂ni (l) ∂xs (w) ∂θij ∂xh (v) å + + (δis−δjs) ni (l) ∂ni (w) ∂xh (v) sin θij + (δih−δjh) ni (l) ∂ni (v) ∂xs (w) sin θij+ + ni (l) (xi (m) −xj (m)) cos θij Ç∂ni (m) ∂xh (v) ∂θij ∂xs (w) + ∂ni (m) ∂xs (w) ∂θij ∂xh (v) å − −(δih−δjh) (δ(lv) −ni (l) ni (v)) cos θij ∂θij ∂xs (w) − −(δis−δjs) (δ(lw) −ni (l) ni (w)) cos θij ∂θij ∂xh (v) + + (δ(lm) −ni (l) ni (m)) (xi (m) −xj (m)) sin θij Ç ∂θij ∂xh (v) ∂θij ∂xs (w) − ∂2θij ∂xh (v)∂xs (w) å (D.43) −e(lmk) ∂ni (m) ∂xs (w) (xi (k) −xj (k)) sin θij ∂θij ∂xh (v) −(δih−δjh) e(lmv) ni (m) sin θij ∂θij ∂xs (w) − −(δis−δjs) e(lmw) ni (m) sin θij ∂θij ∂xh (v) −e(lmk) ni (m) (xi (k)−xj (k)) cos θij ∂θij ∂xh (v) ∂θij ∂xs (w) − −e(lmk) ni (m) (xi (k)−xj (k)) sin θij ∂2θij ∂xh (v)∂xs (w) +(xi (m)−xj (m)) sin θij Ç ∂ni (l) ∂xh (v) ∂ni (m) ∂xs (w) + + ∂ni (l) ∂xs (w) ∂ni (m) ∂xh (v) + ni (l) ∂2ni (m) ∂xh (v)∂xs (w) + ∂2ni (l) ∂xh (v)∂xs (w) ni (m) å + + (δis−δjs) ∂ni (l) ∂xh (v) ni (w) sin θij + (δih−δjh) ∂ni (l) ∂xs (w) ni (v) sin θij+ + ni (m) (xi (m) −xj (m)) cos θij Ç ∂ni (l) ∂xh (v) ∂θij ∂xs (w) + ∂ni (l) ∂xs (w) ∂θij ∂xh (v) å + + (δis−δjs) ni (l) ∂ni (w) ∂xh (v) sin θij + (δih−δjh) ni (l) ∂ni (v) ∂xs (w) sin θij+ + ni (l) (xi (m) −xj (m)) cos θij Ç∂ni (m) ∂xh (v) ∂θij ∂xs (w) + ∂ni (m) ∂xs (w) ∂θij ∂xh (v) å − −(δih−δjh) (δ(lv) −ni (l) ni (v)) cos θij ∂θij ∂xs (w) − −(δis−δjs) (δ(lw) −ni (l) ni (w)) cos θij ∂θij ∂xh (v) + + (δ(lm) −ni (l) ni (m)) (xi (m) −xj (m)) sin θij Ç ∂θij ∂xh (v) ∂θij ∂xs (w) − ∂2θij ∂xh (v)∂xs (w) å (D.43) −e(lmk) ∂ni (m) ∂xs (w) (xi (k) −xj (k)) sin θij ∂θij ∂xh (v) −(δih−δjh) e(lmv) ni (m) sin θij ∂θij ∂xs (w) − −(δis−δjs) e(lmw) ni (m) sin θij ∂θij ∂xh (v) −e(lmk) ni (m) (xi (k)−xj (k)) cos θij ∂θij ∂xh (v) ∂θij ∂xs (w) − Ç ∂ i ∂ i ( ) ∂x(v) ( ) ( ) ( ) ∂x(v) ∂x(w) −e(lmk) ni (m) (xi (k)−xj (k)) sin θij ∂2θij ∂xh (v)∂xs (w) +(xi (m)−xj (m)) sin θij Ç ∂ni (l) ∂xh (v) ∂ni (m) ∂xs (w) + −e(lmk) ni (m) (xi (k)−xj (k)) sin θij ∂2θij ∂xh (v)∂xs (w) +(xi (m)−xj (m)) sin θij Ç ∂ni (l) ∂xh (v) ∂ni (m) ∂xs (w) + D.4 Derivatives of dij The vector dij is the director vector of the plane that bounds the edge of the cubic triangle at the corner nodes. It is deﬁned in item 2 on page 105 and in equation (9.7). Its ﬁrst derivative is presented below. ∂dij (l) ∂xh (v) = e(lmk) · ∂ni (m) ∂xh (v) ·(xi (k) −xj (k))·cos θij +(δih−δjh)·e(lmv) ·ni (m) ·cos θij− −e(lmk) ·ni (k) ·(xi (k) −xj (k))·sin θij · ∂θij ∂xh (v) + ∂ni (l) ∂xh (v) ·ni (m) ·(xi (m) −xj (m))·sin θij+ + ni (l) · ∂ni (m) ∂xh (v) · (xi (m) −xj (m)) · sin θij −(δih−δjh) · (δ(lv) −ni (l) · ni (v)) · sin θij− −(δ(lm) −ni (l) · ni (m)) · (xi (m) −xj (m)) · cos θij ∂θij ∂xh (v) (D.42) And the expression of the second derivative is included next. ∂2dij (l) h (v)∂xs (w) = e(lmk) (xi (k)−xj (k)) Ç ∂2ni (m) ∂xh (v)∂xs (w) cos θij − ∂ni (m) ∂xh (v) sin θij ∂θij ∂xs (w) å + + (δis−δjs) e(lmw) ∂ni (m) ∂xh (v) cos θij + (δih−δjh) e(lmv) ∂ni (m) ∂xs (w) cos θij− D.5. D.5 Derivatives of θij θij is the angle of the drilling rotation as deﬁned in equation (9.33). The ﬁrst and second derivatives are presented below. ∂θij ∂xh (v) = ∂αij ∂xh (v) −∂αji ∂xh (v) = 1 zi zi X k=1 k̸=j ∂βjk i ∂xh (v) −1 zj zj X k=1 k̸=i ∂βik j ∂xh (v) (D.44) ∂2θij ∂xh (v)∂xs (w) = ∂2αij ∂xh (v)∂xs (w) − ∂2αji ∂xh (v)∂xs (w) = = 1 zi zi X k=1 k̸=j ∂2βjk i ∂xh (v)∂xs (w) −1 zj zj X k=1 k̸=i ∂2βik j ∂xh (v)∂xs (w) (D.45) (D.45) 246 APPENDIX D. ANALYTIC DERIVATIVES APPENDIX D. ANALYTIC DERIVATIVES 246 And βjk i has been deﬁned in equation (9.1). And βjk i has been deﬁned in equation (9.1). ∂βjk i ∂xh (v) = (eij · eik) · ∂\|eijeikni\| ∂xh (v) −\|eijeikni\| · ∂(eij · eik) ∂xh (v) (eij · eik)2 + \|eijeikni\|2 (D.46) ∂2βjk i ∂xh (v)∂xs (w) = ® −2 · ∂βjk i ∂xh (v) · ñ (eij · eik) · ∂(eij · eik) ∂xs (w) + +\|eijeikni\| · ∂\|eijeikni\| ∂xs (w) ô + ∂(eij · eik) ∂xs (w) · ∂\|eijeikni\| ∂xh (v) − −∂(eij · eik) ∂xh (v) · ∂\|eijeikni\| ∂xs (w) + (eij · eik) · ∂2\|eijeikni\| ∂xh (v)∂xs (w) − −\|eijeikni\| · ∂2(eij · eik) ´ · (eij · eik)2 + \|eijeikni\|2−1 The vectors eij and eik have been deﬁned in equations (9.2) and (9.3); see ﬁgure 9.2 on page 103. The vectors eij and eik have been deﬁned in equations (9.2) and (9.3); see ﬁgure 9.2 on page 103. D.5 Derivatives of θij ∂\|eijeikni\| ∂xh (v) = ∂eij ∂xh (v) eik ni + eij ∂eik ∂xh (v) ni (D.48) ∂(eij · eik) ∂xh (v) = ∂eij (l) ∂xh (v) · eik (l) + eij (l) · ∂eik (l) ∂xh (v) (D.49) (D.48) ∂eij (l) ∂xh (v) = (δjh −δih) · (δ(lv) −ni (l) · ni (v))− − Ç ∂ni (l) ∂xh (v) · ni (m) + ni (l) · ∂ni (m) ∂xh (v) å · (xj (m) −xi (m)) (D.50) ∂2\|eijeikni\| ∂xh (v)∂xs (w) = ∂2eij ∂xh (v)∂xs (w) eik ni + ∂eij ∂xh (v) ∂eik ∂xs (w) ni + ∂eij ∂xh (v) eik ∂ni ∂xs (w) + + eij ∂2eik ∂xh (v)∂xs (w) ni + ∂eij ∂xs (w) ∂eik ∂xh (v) ni + eij ∂eik ∂xs (w) ∂ni ∂xh (v) (D.51) ij ∂2(eij · eik) ∂xh (v)∂xs (w) = ∂2eij (l) ∂xh (v)∂xs (w) · eik (l) + eij (l) · ∂2eik (l) ∂xh (v)∂xs (w) + + ∂eij (l) ∂xh (v) · ∂eik (l) ∂xs (w) + ∂eij (l) ∂xs (w) · ∂eik (l) ∂xh (v) (D.52) D.6. DERIVATIVES OF ϕij 247 ∂2eij (l) ∂xh (v)∂xs (w) = −(δjh −δih) · Ç ∂ni (l) ∂xs (w) · ni (v) + ni (l) · ∂ni (v) ∂xs (w) å − −(δjs −δis) · Ç ∂ni (l) ∂xh (v) · ni (w) + ni (l) · ∂ni (w) ∂xh (v) å − − Ç ∂2ni (l) ∂xh (v)∂xs (w) · ni (m) + ∂ni (l) ∂xh (v) · ∂ni (m) ∂xs (w) + + ∂ni (l) ∂xs (w) · ∂ni (m) ∂xh (v) + ni (l) · ∂2ni (m) ∂xh (v)∂xs (w) å · (xj (m) −xi (m)) (D.53) (D.53) (D.53) D.6 Derivatives of ϕij ϕij is the angle of the inclination of the tangent of the curved edge at the node with respect to its chord as deﬁned in equation (9.18). The ﬁrst and second derivatives are presented below. ∂ϕij ∂xh (v) = − tan ϕij ∥xi −xj∥2 · (δih −δjh) · (xi (v) −xj (v))− − sec ϕij ∥xi −xj∥· ñ (δij −δjh) · ni (v) + (xi (k) −xj (k)) ∂ni (k) ∂xh (v) ô (D.54) ∂2ϕij ∂xh (v)∂xs (w) = tan ϕij · ® ∂ϕij ∂xh (v) · ∂ϕij ∂xs (w) + (δih −δjh)(δis −δjs) ∥xi −xj∥2 · · " (xi (v) −xj (v))(xi (w) −xj (w)) ∥xi −xj∥2 −δ(vw) #) −∂ϕij ∂xh (v) · (δis −δjs)(xi (w) −xj (w)) ∥xi −xj∥2 − −∂ϕij ∂xs (w) · (δih −δjh)(xi (v) −xj (v)) ∥xi −xj∥2 − sec ϕij ∥xi −xj∥· ñ (δih −δjh) · ∂ni (v) ∂xs (w) + +(δis −δjs) · ∂ni (w) ∂xh (v) + (xi (k) −xj (k)) ∂2ni (k) ∂xh (v)∂xs (w) ô (D.55) Bibliography [1] "ABAQUS Theory Manual, ABAQUS/Standard User’s Manual". vol. II. ABAQUS. 1998 (cit. on p. 165). [2] Sohrabuddin Ahmad, Bruce M. Irons, and O. C. Zienkiewicz. “Anal- ysis of thick and thin shell structures by curved ﬁnite elements”. In: International Journal for Numerical Methods in Engineering 2.3 (July 1970), pp. 419–451. ISSN: 0029-5981. DOI: 10.1002/nme. 1620020310 (cit. on p. 65). [3] Alinghi et al. “The Simulation Race for America’s Cup. Yacht de- signers used engineering simulation in a variety of applications to edge out the competition”. In: ANSYS Advantage 1 (2 2007), pp. 3– 6. URL: http://www.ansys.com/magazine/issues/1- 2- 2007 - biomedical / 03 - americas - cup . pdf (cit. on pp. 202, 203). [4] "ANSYS User’s Manual". Report Version 5.5. ANSYS. 1998 (cit. on p. 165). [5] Inma Ortigosa Barragán. “Development of a decision support sys- tem for the design and adjustment of sailboat rigging”. Direc- tor: Julio García-Espinosa. PhD thesis. Universitat Politècnica de Catalunya, 2012. URL: http :/ / cataleg. upc .edu /record = b1418440~S1cat (cit. on p. 20). [6] Klaus-Jurgen Bathe, Phil-Seung Lee, and Jean-François Hiller. “"Towards improving the MITC9 shell element"”. In: "Computers and Structures" 81 (2003), pp. 477–489. DOI: 10.1016/S0045- 7949(02)00483-2 (cit. on p. 165). [7] Jean-Pierre Baudet. “Sail of one piece three dimensional laminated fabric having uninterrupted load bearing yarns. A one piece sail is made on a mold and includes an outer impervious layer, such as ﬁlm, and an inner layer of continuous and uninterrupted yarns disposed in continuous trajectories from one edge of the sail to the other and carrying the majority of the load imposed on the sail.” Pat. 5,097,784 (United States). North Sails Group Inc. Aug. 21, 1990. URL: http://patft.uspto.gov/netacgi/nph-Parser? Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml% 2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=5,097,784.PN. &OS=PN/5,097,784&RS=PN/5,097,784 (cit. on pp. 5, 202). 249 250 BIBLIOGRAPHY [8] Y Bazilevs et al. “Isogeometric analysis using T-splines”. In: Com- puter Methods in Applied Mechanics and Engineering 199.5-8 (Jan. 2010), pp. 229–263. ISSN: 00457825. DOI: 10.1016/j.cma.2009. 02.036 (cit. on p. 30). [9] Ted Belytschko, Wing Kam Liu, and Brian Moran. Nonlinear Fi- nite Elements for Continua and Structures. ﬁrst. Wiley, 2000. ISBN: 978-0471-98774-1. URL: https://www.wiley.com/en- us / Nonlinear + Finite + Elements + for + Continua + and + Structures % 2C + 2nd + Edition - p - 9781118700082 (cit. on pp. 65, 83, 193). [10] Ted Belytschko et al. “Stress projection for membrane and shear locking in shell ﬁnite elements”. Bibliography In: "Computer Methods in Ap- plied Mechanics and Engineering" 51 (1985), pp. 221–258. DOI: 10. 1016/0045-7825(85)90035-0 (cit. on pp. 164, 167, 168). [11] D.J. Benson, Y. Bazilevs, and T.J.R. Hughes. “Isogeometric shell analysis: The Reissner–Mindlin shell”. In: Computer Methods in Applied Mechanics and Engineering 199 (5-8 2010): Computational Geometry and Analysis, pp. 276–289. DOI: 10.1016/j.cma.2009. 05.011 (cit. on pp. 25, 26). [12] M Bischoff, W A Wall, and E Ramm. “Chapter 3: Models and Finite Elements for Thin-walled Structures”. In: Encyclopedia of Compu- tational Mechanics. Volume 2: Solids and Structures. Ed. by Erwin Stein, René de Borst, and Thomas J. R. Hughes. 1921. Chichester, UK: John Wiley & Sons, Ltd, Nov. 2004. Chap. Chapter 3: p. 79. ISBN: 0470846992. DOI: 10.1002/0470091355 (cit. on pp. 65, 85, 198). [13] Manfred Bischoff and Ekkehard Ramm. “On the physical sig- niﬁcance of higher order kinematic and static variables in a three-dimensional shell formulation”. In: International Journal of Solids and Structures 37.46-47 (Nov. 2000), pp. 6933–6960. ISSN: 00207683. DOI: 10 . 1016 / S0020 - 7683(99 ) 00321 - 2 (cit. on p. 182). [14] BOWLER ON MOVISTAR. Farr Yacht Design’s Russel Bowler talks about his ﬁrst assessment of the damage that resulted in movistar being abandoned. Volvo Ocean Race 2005-2006. May 25, 2006. URL: http://www.volvooceanrace.com/news/article/2006/ may/bowler/index.aspx (cit. on p. 206). [15] D. Briassoulis. “Testing the asymptotic behaviour of shell elements. Part I: the classical benchmark tests”. In: International Journal for Numerical Methods in Engineering 54.3 (May 2002), pp. 421–452. ISSN: 0029-5981. DOI: 10.1002/nme.436 (cit. on p. 56). [16] D. Briassoulis. “Testing the asymptotic behaviour of shell elements. Part II: New limit tests: analytical solutions and the RFNS ele- ment case”. In: International Journal for Numerical Methods in Engineering 54.5 (June 2002), pp. 631–670. ISSN: 0029-5981. DOI: 10.1002/nme.435 (cit. on p. 56). 251 BIBLIOGRAPHY [17] Demetres Briassoulis. “Asymptotic deformation modes of bench- mark problems suitable for evaluating shell elements”. In: Com- puter Methods in Applied Mechanics and Engineering 194.21-24 (June 2005), pp. 2385–2405. ISSN: 00457825. DOI: 10.1016/j. cma.2004.07.038 (cit. on p. 56). [18] Norbert Büchter and Ekkerhard Ramm. “Shell theory vs degeneration—A comparison in large rotation ﬁnite element analy- sis”. In: International Journal for Numerical Methods in Engineer- ing 34 (1992), pp. 39–59. DOI: 10.1002/nme.1620340105 (cit. on p. 85). [19] P. de Casteljau. Outillages méthodes calcul. Tech. rep. Paris: A. Cit- roën, 1959 (cit. on p. 44). [20] P. de Casteljau. Bibliography Courbes et surfaces à pôles. Tech. rep. Paris: A. Citroën, 1963 (cit. on p. 44). [21] Pablo Pérez del Castillo and Julio García. Un modelo numérico para el análisis de velas. Informe Técnico IT-466. Barcelona: CIMNE, Dec. 2005. 143 pp. (cit. on p. 20). [22] D. Chapelle and K.J. Bathe. “Fundamental considerations for the ﬁnite element analysis of shell structures”. In: Computers & Struc- tures 66.1 (Jan. 1998), pp. 19–36. ISSN: 00457949. DOI: 10.1016/ S0045-7949(97)00078-3 (cit. on pp. 56, 168, 169). [23] T. Charvet, F. Hauville, and S. Huberson. “Numerical simulation of the ﬂow over sails in real sailing conditions”. In: Journal of Wind Engineering and Industrial Aerodynamics 63 (1996): Special is- sue on sail aerodynamics, pp. 111–129. DOI: 10 . 1016 / S0167 - 6105(96)00072-4 (cit. on p. 20). [24] F. Cirak, M. Ortiz, and P. Schröder. “Subdivision surfaces: a new paradigm for thin shell ﬁnite-element analysis.” In: Interna- tional Journal for Numerical Methods in Engineering 47.12 (2000), pp. 2039–2072. DOI: 10.1002/(SICI)1097- 0207(20000430) 47:12%3C2039::AID-NME872%3E3.0.CO;2-1 (cit. on pp. 25, 29, 33, 186). [25] Fehmi Cirak and Michael Ortiz. “Fully C 1 -conforming subdivision elements for ﬁnite deformation thin-shell analysis”. In: Interna- tional Journal for Numerical Methods in Engineering 51.7 (2001), pp. 813–833. DOI: 10.1002/nme.182 (cit. on pp. 29, 186). [26] Fehmi Cirak et al. “Integrated modeling, ﬁnite-element analysis, and engineering design for thin-shell structures using subdivision”. In: Computer-Aided Design 34.2 (Feb. 2002), pp. 137–148. ISSN: 00104485. DOI: 10 . 1016 / S0010 - 4485(01 ) 00061 - 6 (cit. on pp. 29, 186). [27] R. Clough and J. Tocher. “Finite element stiffness matrices for anal- ysis of plates in bending”. In: Proceedings of Conference on Matrix Methods in Structural Analysis. 1965 (cit. on pp. 23, 28). [28] Compass participa en la Copa América de la mano de Alinghi y Great Britain. 2003. URL: http : / / www . compassis . com / compass/Compass/Prensa/Copa+america (cit. on pp. 183, 204). 252 BIBLIOGRAPHY [29] Ronald Cools and Philip Rabinowitz. “Monomial cubature rules since "Stroud": a compilation”. In: Journal of Computational and Applied Mathematics 48 (1993), pp. 309–326. DOI: 10.1016/0377- 0427(93)90027-9 (cit. on p. 82). [30] Xiangyang Cui et al. “Analysis of plates and shells using an edge- based smoothed ﬁnite element method”. In: Computational Me- chanics 45 (2010), pp. 141–156. DOI: 10 . 1007 / s00466 - 009 - 0429-9 (cit. on p. 177). [31] Choi D et al. Bibliography “Membrane locking in the ﬁnite element computa- tion of very thin elastic shells”. In: Modelisation mathematique et analyse numerique 32.2 (1998), pp. 131–152. URL: http://www. numdam.org/item?id=M2AN\_1998\_\_32\_2\_131\_0 (cit. on pp. 99, 101, 162, 173). [32] Rick Deppe. Puma breaks boom. Volvo Ocean Race 2008-2009. Jan. 24, 2009. URL: http : / / www . youtube . com / watch ? v = s9S60G47Ikg (cit. on p. 206). [33] D.A. Dunavant. “High degree efﬁcient symmetrical Gaussian quadrature rules for the triangle”. In: International Journal for Numerical Methods in Engineering 21 (1985), pp. 1129–1148. DOI: 10.1002/nme.1620210612 (cit. on p. 82). [34] Nguyen Tien Dung and Garth N. Wells. “Geometrically nonlinear formulation for thin shells without rotation degrees of freedom”. In: Computer Methods in Applied Mechanics and Engineering 197.33- 40 (June 2008), pp. 2778–2788. ISSN: 00457825. DOI: 10.1016/j. cma.2008.01.001 (cit. on p. 30). [35] Element types used in structural analysis. Moldﬂow Flex. Autodesk. 2016. URL: http : / / help . autodesk . com / view / MFIWS / 2016 / ENU / ?guid = GUID - C6040A5F - 6941 - 450C - B147 - FBC16693CC13 (cit. on p. 24). [36] J. Burns Fallow. “America’s Cup sail design”. In: Journal of Wind Engineering and Industrial Aerodynamics 63 (1996): Special is- sue on sail aerodynamics, pp. 183–192. DOI: 10 . 1016 / S0167 - 6105(96)00075-X (cit. on p. 202). [37] G. Farin. Curves and Surfaces for CAGD. A practical guide. 5th ed. 340 Pine Street, Sixth Floor, San Francisco, CA 94104-3205, USA: Morgan Kaufmann Publishers, 2002. URL: https : / / www . elsevier.com/books/curves- and- surfaces- for- cagd/ farin/978-1-55860-737-8 (cit. on pp. 23, 44, 52). [38] Gerald Farin. “Triangular Bernstein-Bézier patches”. In: Com- puter Aided Geometric Design 3.2 (Aug. 1986), pp. 83–127. ISSN: 01678396. DOI: 10.1016/0167-8396(86)90016-6 (cit. on p. 47). [39] Carlos A. Felippa. “A study of optimal membrane triangles with drilling freedoms”. In: Computer Methods in Applied Mechan- ics and Engineering 192.16-18 (Apr. 2003), pp. 2125–2168. ISSN: 00457825. DOI: 10 . 1016 / S0045 - 7825(03 ) 00253 - 6 (cit. on pp. 102, 108). 253 BIBLIOGRAPHY [40] Richard G.J. Flay. “A twisted ﬂow wind tunnel for testing yacht sails”. In: Journal of Wind Engineering and Industrial Aerodynam- ics 63 (1996): Special issue on sail aerodynamics, pp. 171–182. DOI: 10.1016/S0167-6105(96)00080-3 (cit. on p. 202). [41] F. G. Flores and E. Oñate. Bibliography “Improvements in the membrane be- haviour of the three node rotation-free BST shell triangle using an assumed strain approach.” In: Computer Methods in Applied Mechanics and Engineering 194.6-8 (2005), pp. 907–932. DOI: 10. 1016/j.cma.2003.08.012 (cit. on pp. 25, 28, 33, 197). [42] Fernando G Flores and Eugenio Oñate. “A rotation-free shell tri- angle for the analysis of kinked and branching shells”. In: Inter- national Journal for Numerical Methods in Engineering 69.7 (Feb. 2007), pp. 1521–1551. ISSN: 00295981. DOI: 10.1002/nme.1823 (cit. on p. 137). [43] David Flynn. “Code Zero—The new way to close reach”. In: Sailing World (Mar. 2001), pp. 62–64. URL: http://www.quantumsails. co.za/downloads/Code_Zero.pdf (cit. on p. 6). [44] Wilhelm Flügge. Statique et dynamique des coques. Ed. by Eyrolles. Paris: Eyrolles, 1960. URL: http://cataleg.upc.edu/record= b1036939~S1cat (cit. on p. 22). [45] Erez Gal and Robert Levy. “Geometrically Nonlinear Analysis of Shell Structures Using a Flat Triangular Shell Finite Element”. In: Archives of Computational Methods in Engineering 13.3 (2006), pp. 331–388. DOI: 10.1007/BF02736397 (cit. on p. 21). [46] J. García-Espinosa, I. Ortigosa, and A. Fernández. “Desarrollo de un sistema de apoyo a la decisión para optimizar el comportamiento de barcos de vela”. In: Revista Internacional de Métodos Numéricos para Cálculo y Diseño en Ingeniería 31.3 (July 2015), pp. 146–153. ISSN: 02131315. DOI: 10.1016/j.rimni.2014.04.002 (cit. on p. 20). [47] Julio García-Espinosa, Abel Coll, and Eugenio Oñate. “Advances in the development of an unstructured FEM solver for ﬂuid-structure interaction problems in marine engineering”. In: Computational Methods in Marine Engineering III. (Trondheim, Norway). Ed. by T. Kvamsdal et al. Barcelona, Spain: CIMNE, 2009, pp. 155–158. ISBN: 978-84-96736-66-5 (cit. on p. 204). [48] Timothy D. Gatzke and Cindy M. Grimm. “Estimating curvature on triangular meshes”. In: International Journal of Shape Modeling 12.1 (2006), pp. 1–28. DOI: 10.1142/S0218654306000810 (cit. on pp. 179, 181). [49] Mattias Gärdsback and Gunnar Tibert. “A comparison of rotation- free triangular shell elements for unstructured meshes”. In: Com- puter Methods in Applied Mechanics and Engineering 196 (49-52 2007), pp. 5001–5015. DOI: 10.1016/j.cma.2007.06.017 (cit. on pp. 24, 25, 28, 186). 254 BIBLIOGRAPHY [50] S. Güzey et al. “Design and development of a discontinuous Galerkin method for shells”. In: Computer Methods in Applied Me- chanics and Engineering 195.25-28 (May 2006), pp. 3528–3548. ISSN: 00457825. DOI: 10.1016/j.cma.2005.08.001 (cit. on p. 30). [51] H Hakula, Y Leino, and J. Pitkäranta. “Scale resolution, lock- ing, and high-order ﬁnite element modelling of shells”. Bibliography In: Com- puter Methods in Applied Mechanics and Engineering 133 (1996), pp. 157–182. DOI: 10.1016/0045- 7825(95)00939- 6 (cit. on pp. 101, 163, 171). [52] John G. Hamilton and Jocelyn M. Patterson. “Structural Design and Analysis of an America’s Cup Yacht”. In: SAMPE Journal 28 (6 1992), pp. 9–13 (cit. on p. 206). [53] Kevin C. Horrigan. Bringing a Virtual Wind Tunnel to Market. Tech. rep. Sponsored by North Sails Performance Resource Group. Thayer School of Engineering, Dartmouth College, Sept. 2003 (cit. on p. 205). [54] Thomas J.R. Hughes and Wing Kam Liu. “"Nonlinear ﬁnite element analysis of shells: Part I. three-dimensional shells"”. In: "Com- puter Methods in Applied Mechanics and Engineering" 26.3 (1981), pp. 331–362. DOI: 10.1016/0045- 7825(81)90121- 3 (cit. on p. 165). [55] T.J.R Hughes. The Finite Element Method; Linear Static and Dy- namic Finite Element Analysis. Englewood Cliffs, New Jersey: Prentice Hall, 1987 (cit. on p. 106). [56] T.J.R. Hughes, J.A. Cottrell, and Y. Bazilevs. “Isogeometric Anal- ysis: CAD, ﬁnite elements, NURBS, exact geometry and mesh re- ﬁnement.” In: Computer Methods in Applied Mechanics and Engi- neering 194.39-41 (2005), pp. 4135–4195. DOI: 10.1016/j.cma. 2004.10.008 (cit. on pp. 25, 30, 33, 44). [57] Mickey Ickert. “FINNatics - The History and Techniques of Finn Sailing”. In: International Finn Association, 1999. Chap. 32. Plas- tic Fantastic - a new chapter in sail development, pp. 162–163. URL: http://www.finnclass.org/The%20Finn/Finnatics/ %2032_Plasticfantasic_162-163.pdf (cit. on p. 5). [58] Paul Jones and Rich Korpus. “International America’s Cup Class Yacht Design Using Viscous Flow CFD”. In: The 15th Cheasa- peake Sailing Yacht Symposium. (Annapolis, MD). Ed. by CSYS. The Cheasapeake Sailing Yacht Symposium. 2001. URL: http:// appliedfluidtech.com/publications- pdfs/CSYS_2001_ Jones.pdf (cit. on pp. 204, 205). [59] Frank DeBord Jr. et al. “Design Optimization for the International America’s Cup Class”. In: SNAME Transactions (2002) (cit. on pp. 8, 204). 255 BIBLIOGRAPHY [60] James Stackpole Herman Jr. “A Sail Force Dynamometer: De- sign, implementation and Data Handling”. Master of Science in Naval Architecture and Marine Engineering. English. Master The- sis. Cambridge, MA: Massachusetts Institute of Technology (MIT), 1988. URL: http://hdl.handle.net/1721.1/14491 (cit. on pp. 201, 202). [61] William Kahan. “"How Futile are Mindless Assessments of Round- off in Floating-Point Computation?"” Work in progress. Jan. 2006. URL: http : / / people . eecs . berkeley . edu / ~wkahan / Mindless.pdf (cit. on p. 113). [62] William Kahan. “"Miscalculating Area and Angles of a Needle-like Triangle"”. Bibliography from Lecture Notes for Introductory Numerical Analy- sis Classes. Sept. 2014. URL: https://people.eecs.berkeley. edu/~wkahan/Triangle.pdf (cit. on p. 113). [63] J Kiendl et al. “Isogeometric shell analysis with Kirchhoff–Love elements”. In: Computer Methods in Applied Mechanics and En- gineering 198.49-52 (2009), pp. 3902–3914. ISSN: 0045-7825. DOI: 10.1016/j.cma.2009.08.013 (cit. on pp. 26, 30, 33). [64] Hyun-Jung Kim, Yu-Deok Seo, and Sung-Kie Youn. “Isogeometric analysis for trimmed CAD surfaces”. In: Computer Methods in Ap- plied Mechanics and Engineering 198.37-40 (Aug. 2009), pp. 2982– 2995. ISSN: 00457825. DOI: 10.1016/j.cma.2009.05.004 (cit. on p. 30). [65] G.R. Kirchhoff. “"Über das Gleichgewicki und die Bewegung einer elastishem Scheibe"”. In: "Journal für die Reine und Angewandte Mathematik" 40 (1850), pp. 51–88 (cit. on p. 22). [66] Patricia Kranz. “Measuring wealth by the foot”. In: The New York Times (Mar. 2008). URL: http://www.nytimes.com/2008/03/ 16/business/16drop.html?_r=1&scp=1&sq=measuring+ wealth+by+the+foot&st=nyt (cit. on p. 199). [67] E. Kreyszig. Differential Geometry. Dover, 1991. URL: https:// store . doverpublications . com / 0486667219 . html (cit. on pp. xxv, 34). [68] A. Laulusa et al. “Evaluation of some shear deformable shell el- ements”. In: International Journal of Solids and Structures 43 (2006), pp. 5033–5054. DOI: 10.1016/j.ijsolstr.2005.08.006 (cit. on pp. 165, 169, 171). [69] Leonardo Leonetti and H. Nguyen-Xuan. “"A mixed edge-based smoothed solid-shell ﬁnite element method (MES-FEM) for lam- inated shell structures"”. In: Composite Structures 208 (2018), pp. 168–179. DOI: 10.1016/j.compstruct.2018.09.077 (cit. on p. 177). [70] J Linhard, K.-U. Bletzinger, and M. Firl. “Upgrading Membranes to Shells - The CEG Rotation Free Shell Element and its Applica- tions”. In: 9th US National Congress on Computational Mechanics. Vol. CD-ROM. University of California at Berkeley, USACM. San Francisco, California, USA, July 2007 (cit. on pp. 38, 43). 256 BIBLIOGRAPHY [71] Johannes Linhard, Roland Wüchner, and Kai-Uwe Bletzinger. ““Upgrading” membranes to shells—The CEG rotation free shell element and its application in structural analysis”. In: Finite Ele- ments in Analysis and Design 44.1–2 (2007), pp. 63 –74. ISSN: 0168- 874X. DOI: 10.1016/j.finel.2007.09.001 (cit. on pp. 25, 38). [72] A. E. H. Love. “On the small free vibrations and deformations of elastic shells”. In: Philosophical Transactions of the Royal Society of London. A 179 (1888), pp. 491–546. URL: http://www.jstor. org/stable/90527 (cit. on p. 22). [73] Richard H. MacNeal. “Perspective on ﬁnite elements for shell anal- ysis”. In: Finite Elements in Analysis and Design 30 (3 1998), pp. 175–186. DOI: 10.1016/S0168-874X(98)00005-5 (cit. on pp. 21, 23, 24, 33). [74] Jacquelin Magnay. Bibliography “London 2012 Olympics: Team GB making waves with new technology”. In: Telegraph.co.uk (Mar. 2010). URL: http : / / www . telegraph . co . uk / sport / othersports / sailing / 7535609 / London - 2012 - Olympics - Team - GB - making-waves-with-new-technology.html (cit. on p. 200). [75] George Marsh. “America’s Cup – pushing materials to their limits”. In: Reinforced plastics 45 (10 2001), pp. 48–50. DOI: 10 . 1016 / S0034-3617(01)80389-1 (cit. on p. 206). [76] George Marsh. “This carbon sail-race winner is ‘diabolical’”. In: Re- inforced plastics 50 (9 2006), pp. 20–27. DOI: 10.1016/S0034- 3617(06)71126-2 (cit. on p. 206). [77] Nelson Max. “Weights for computing vertex normals from facet nor- mals”. In: Journal of graphics, gpu and game tools 4.2 (1999), pp. 1– 6. DOI: 10.1080/10867651.1999.10487501 (cit. on p. 187). [78] O. Le Maître, J.E. Souza de Cursi, and S. Huberson. “Large dis- placement analysis of ideally ﬂexible sails”. In: European Journal of Mechanics - A/Solids 17.4 (1998), pp. 619–636. DOI: 10.1016/ S0997-7538(99)80025-4 (cit. on p. 19). [79] Jerome H. Milgram. “Naval Architecture Technology used in Win- ning the 1992 America’s Cup Match”. In: SNAME Transactions 101 (1993), pp. 399–436 (cit. on pp. 203, 204). [80] Jerome H. Milgram. “Fluid mechanics for sailing vessel design”. In: Annu. Rev. Fluid Mech. 30 (1998), pp. 613–653. DOI: 10.1146/ annurev.fluid.30.1.613 (cit. on p. 207). [81] R.D. Mindlin. “Inﬂuence of rotary inertia and shear on ﬂexural mo- tions of isotropic elastic plates”. In: Journal of Applied Mechanics 18 (1951) (cit. on p. 23). [82] R.a.S. Moreira and J. Dias Rodrigues. “A non-conforming plate facet-shell ﬁnite element with drilling stiffness”. In: Finite Ele- ments in Analysis and Design 47.9 (Sept. 2011), pp. 973–981. ISSN: 0168874X. DOI: 10.1016/j.finel.2011.03.013 (cit. on pp. 117, 143, 144). 257 BIBLIOGRAPHY [83] Gustav Morin. ERICSSON 3 EMAIL: ’A FOUR-METER CRACK AND AN OPEN HOLE IN THE HULL’. Volvo Ocean Race 2008- 2009. Jan. 27, 2009. URL: http://www.volvooceanrace.com/ news/article/2009/January/EMAIL-ERT3-L4-D10-0952/ (cit. on p. 206). [84] Movistar abandons ship. Volvo Ocean Race 2005-2006. May 21, 2006. URL: http : / / www . volvooceanrace . com / news / article/2006/may/abandon/index.aspx (cit. on p. 206). [85] Jari Mäkinen. “A Formulation for Flexible Multibody Mechan- ics”. PhD thesis. Tampere University of Technology, 2004. ISBN: 9521512881. URL: http://butler.cc.tut.fi/~jmamakin/ vk.pdfhttp://butler.cc.tut.fi/~jmamakin/tutkimus/ Raportti.htm (cit. on p. 33). [86] L. Noels and R. Radovitzky. “A new discontinuous Galerkin method for Kirchhoff–Love shells”. Bibliography In: Computer Methods in Applied Me- chanics and Engineering 197.33-40 (June 2008), pp. 2901–2929. ISSN: 00457825. DOI: 10.1016/j.cma.2008.01.018 (cit. on p. 30). [87] Bastian Oesterle, Ekkerhard Ramm, and Manfred Bischoff. “A shear deformable, rotation-free isogeometric shear formulation”. In: Computer Methods in Applied Mechanics and Engineering 307 (Aug. 2016), pp. 235–255. DOI: 10.1016/j.cma.2016.04.015 (cit. on p. 181). [88] S. Oller. Análisis y Cálculo de Estructuras de Materiales Com- puestos. Ref. L60. Barcelona: CIMNE, 2002. 576 pp. ISBN: 84- 95999-06-4. URL: http://www.cimne.upc.es/tiendaCIMNE/ ProductosCon.asp?id_prod=960 (cit. on p. 206). [89] S. Oller. Simulación Numérica del Comportamiento Mecánico de los Materiales Compuestos. Ref. M74. Barcelona: CIMNE, 2003. 193 pp. ISBN: 84-95999-24-2. URL: http : / / www . cimne . upc . es/tiendaCIMNE/ProductosCon.asp?id_prod=994 (cit. on p. 206). [90] Peter van Oossanen. “Predicting the Speed of Sailing Yachts”. In: SNAME Transactions 101 (1993), pp. 337–397 (cit. on p. 204). [91] ORC VPP Documentation 2009. Version 1.3. Offshore Racing Congress. 2009. URL: http : / / www . orc . org / rules / ORC % 20VPP%20Documentation%202009.pdf (cit. on p. 8). [92] E. Oñate. “A Review of some Finite Element Families for Thick and Thin Plante and Shell Analysis”. In: Recent Developments in Finite element Analysis. A book dedicated to Robert L. Taylor. Ed. by T.J.R. Hughes, E. Oñate, and O.C. Zienkiewicz. Barcelona: CIMNE, 1994, pp. 98–111. ISBN: 84-87867-45-6. URL: https://books.cimne. com / shop / recent - developments - in - finite - element - analysis-a-book-dedicated-to-robert-l-taylor/ (cit. on p. 28). 258 BIBLIOGRAPHY [93] E. Oñate. Structural Analysis with the Finite Element Method. Vol. 2 Plates and Shells. CIMNE-Springer, 2009. URL: https : //books.cimne.com/shop/structural- analysis- with- the-finite-element-method-linear-statics-volume-2- beams-plates-and-shells/ (cit. on pp. 20, 21, 23, 24, 33, 82, 101, 181). [94] E. Oñate and M. Cervera. “Derivation of Thin Plate Bending Ele- ments with One Degree of Freedom per Node: A Simple Three Node Triangle”. In: Engineering Computations 10.6 (1993), pp. 543–561. ISSN: 0264-4401. DOI: 10.1108/eb023924 (cit. on p. 29). [95] E. Oñate and F. Flores. “Advances in the formulation of the rotation-free basic shell triangle.” In: Computer Methods in Applied Mechanics and Engineering 194.21-24 (2005), pp. 2406–2443. DOI: 10.1016/j.cma.2004.07.039 (cit. on pp. 25, 28, 33, 165). [96] E. Oñate and F. Zárate. “Rotation-free triangular plate and shell elements.” In: International Journal for Numerical Methods in En- gineering 47 (2000), pp. 557–603. DOI: 10.1002/(SICI)1097- 0207(20000110/30)47:1/3%3C557::AID-NME784%3E3.0.CO; 2-9 (cit. on pp. 25, 28, 33, 41, 176, 186). Bibliography [97] Eugenio Oñate, Julio García, and Sergio Idelsohn. “Ship Hydrody- namics”. In: Encyclopedia Of Computational Mechanics. Ed. by Er- win Stein, René de Borst, and Thomas J.R. Hughes. Vol. 3 Fluids. Chichester: John Wiley & Sons, Ltd., 2004. Chap. 18, pp. 579–610. DOI: 10.1002/0470091355.ecm070 (cit. on p. 204). [98] Nicola Parolini and Alﬁo Quarteroni. “Mathematical models and numerical simulations for the America’s Cup”. In: Computa- tional Methods in Applied Mechanics and Engineering (194 2005), pp. 1001–1026. DOI: 10 . 1016 / j . cma . 2004 . 06 . 020 (cit. on p. 203). [99] Lara Pellegrini. “Desarrollo y validación de un sistema para el cál- culo de velas de barcos mediante el método de los elementos ﬁni- tos”. Máster en Métodos Numéricos para Cálculo y Diseño en In- geniería. Spanish. Tesis de Máster. Barcelona: Universitat Politèc- nica de Catalunya (UPC), 2000 (cit. on p. 4). [100] R. Phaal and C. R. Calladine. “Simple class of ﬁnite elements for plate and shell problems. II: An element for thin shells, with only translational degrees of freedom”. In: International Journal for Nu- merical Methods in Engineering 35.5 (1992), pp. 979–996. ISSN: 00295981. URL: http://www.scopus.com/inward/record. url?eid=2-s2.0-0026925349\&partnerID=tZOtx3y1 (cit. on pp. 28, 140). [101] Quoc-Hoa Pham et al. “"Geometrically nonlinear analysis of func- tionally graded shells using an edge-based smoothed MITC3 (ES- MITC3) ﬁnite elements"”. In: Engineering with Computers (2019), pp. 1–14. DOI: 10.1007/s00366-019-00750-z (cit. on p. 177). 259 BIBLIOGRAPHY [102] B. Philipp et al. “Integrated design and analysis of structural membranes using the Isogeometric B-Rep Analysis”. In: Com- puter Methods in Applied Mechanics and Engineering 303 (2016), pp. 312–340. ISSN: 00457825. DOI: 10.1016/j.cma.2016.02. 003 (cit. on p. 26). [103] E. Ramm and W. A. Wall. “Shell structures—a sensitive interrela- tion between physics and numerics”. In: International Journal for Numerical Methods in Engineering 60.1 (2004), pp. 381–427. ISSN: 1097-0207. DOI: 10.1002/nme.967 (cit. on p. 21). [104] Ekkehard Ramm. “Heinz Isler Schalen”. In: ed. by E. Ramm and E. Schunck. Verlag, 1986. Chap. Form und Tragverhalten, pp. 29–34 (cit. on p. 21). [105] Fernando Rastellini et al. “Composite materials non-linear mod- elling for long ﬁbre-reinforced laminates: Continuum basis, compu- tational aspects and validations”. In: Computers & Structures 86 (9 2008), pp. 879–896. DOI: 10.1016/j.compstruc.2007.04.009 (cit. on p. 206). [106] Robert Razenbach and Zhenlong Xu. “Sail Aero-Structures: Study- ing Primary Load Paths and Distortion”. In: The 17th Cheasapeake Sailing Yacht Symposium. Ed. by CSYS. Bibliography The Cheasapeake Sailing Yacht Symposium. Mar. 2005, pp. 193–204 (cit. on pp. 20, 205). [107] E. Reissner. “On the theory of bending of elastic plates”. In: Journal of Mathematical Physics 23 (1944), pp. 184–191 (cit. on p. 23). [108] E. Reissner. “The effect of transverse shear deformation on the bending of elastic plates”. In: Journal of Applied Mechanics 12 (1945), A69–A77 (cit. on p. 23). [109] Hee Yuel Roh and Maenghyo Cho. “The application of geometri- cally exact shell elements to B-spline surfaces”. In: Computer Meth- ods in Applied Mechanics and Engineering 193.23-26 (June 2004), pp. 2261–2299. ISSN: 00457825. DOI: 10.1016/j.cma.2004.01. 019 (cit. on p. 31). [110] Hee Yuel Roh and Maenghyo Cho. “Integration of geometric design and mechanical analysis using B-spline functions on surface”. In: International Journal for Numerical Methods in Engineering 62.14 (Apr. 2005), pp. 1927–1949. ISSN: 0029-5981. DOI: 10.1002/nme. 1254 (cit. on p. 31). [111] Riccardo Rossi et al. “Simulation of light-weight membrane struc- tures by wrinkling model”. In: International Journal for Numeri- cal Methods in Engineering 62 (15 2005), pp. 2127–2153. DOI: 10. 1002/nme.1266 (cit. on p. 20). [112] Francis Sabourin and Michel Brunet. “Detailed formulation of the rotation-free triangular element “S3” for general purpose shell analysis”. In: Engineering Computations 23.5 (2006), pp. 469–502. ISSN: 0264-4401. DOI: 10 . 1108 / 02644400610671090 (cit. on pp. 28, 38). 260 BIBLIOGRAPHY Box 17, 3300 AA Dordrecht, The Nederlands: Springer, 2005. Chap. Finite Element Analysis of Membrane Structures, pp. 47–68. DOI: 10.1007/1- 4020-3317-6 (cit. on pp. 3, 4, 20, 35). [123] R.L. Taylor, E. Oñate, and P.-A. Ubach. “Textile Composites and Inﬂatable Structures”. In: ed. by E. Oñate and B. Kröplin. Vol. 3. Computational Methods in Applied Sciences. P.O. Box 17, 3300 AA Dordrecht, The Nederlands: Springer, 2005. Chap. Finite Element Analysis of Membrane Structures, pp. 47–68. DOI: 10.1007/1- 4020-3317-6 (cit. on pp. 3, 4, 20, 35). [124] Alexander Tessler, Marco Di Sciuva, and Marco Gherlone. “A con- sistent reﬁnement of ﬁrst-order shear deformation theory for lami- nated composite and sandwich plates using improved zigzag kine- matics”. In: Journal of Mechanics of Materials and Structures 5.2 (Aug. 2010), pp. 341–367. ISSN: 1559-3959. DOI: 10.2140/jomms. 2010.5.341 (cit. on p. 181). [125] Alexander Tessler, David W. Sleight, and John T. Wang. “Effective Modelling and Nonlinear Shell Analysis of Thin Membranes Ex- hibiting Structural Wrinkling”. In: Journal of Spacecraft and Rock- ets 42.2 (2005), pp. 287–298. DOI: 10.2514/1.3915 (cit. on p. 20). [126] Alexander Tessler et al. Reﬁnement of Timoshenko Beam The- ory for Composite and Sandwich Beams Using Zigzag Kinematics. Tech. rep. NASA/TP-2007-215086. Hampton, VA: NASA Langley Research Center, Nov. 2007, p. 45. URL: http : / / ntrs . nasa . gov/search.jsp?R=20070035078 (cit. on p. 181). [127] Stephen Timoshenko and S. Woinowsky-Krieger. Teoria de pla- cas y láminas. Ed. by Urmo. Bilbao: Urmo, 1975. URL: http:// cataleg.upc.edu/record=b1030550~S1*cat (cit. on pp. 22, 158, 160, 162). [128] Daniele Trimarchi et al. “Fluid-structure interactions of anisotropic thin composite materials for application to sail aerodynamics of a yacht in waves”. In: 12th Numerical Towing Tank Symposium 2009 (NuTTS’09). (Cortona, Italy). Numerical Towing Tank Symposium. Curran Associates, Inc., 2009. ISBN: 9781615676149. URL: http: //eprints.soton.ac.uk/69141/ (cit. on p. 20). [129] P.-A. Ubach, C. Estruch, and J. García-Espinosa. “On the inter- polation of normal vectors for triangle meshes”. In: International Journal for Numerical Methods in Engineering 96.4 (Sept. 2013), pp. 247–268. ISSN: 00295981. DOI: 10.1002/nme.4567 (cit. on pp. 44, 46, 158, 179, 187). [130] P.-A. Ubach and E. Oñate. “New Rotation-Free Composite Shell Triangle using Accurate Geometrical Data”. In: WCCM VII. 7th World Congress on Computational Mechanics. Vol. CD-ROM. UCLA, iacm, Northwestern University. Los Angeles, California, USA, July 2006 (cit. on p. 38). [131] P.-A. Ubach and E. Oñate. BIBLIOGRAPHY BIBLIOGRAPHY [113] Nils Salvesen, Martin Fritts, and Michael Meinhold. “Technology and Design for Stars & Stripes”. In: Finite elements in Analysis and Design 4 (1 1988): Special Issue on Uniﬁcation of Finite Ele- ment Methods in Theory and Test, pp. 79–82. DOI: 10.1016/0168- 874X(88)90025-X (cit. on p. 200). [114] Robert Schmidt, Roland Wüchner, and Kai-Uwe Bletzinger. “Iso- geometric analysis of trimmed NURBS geometries”. In: Computer Methods in Applied Mechanics and Engineering 241-244 (Oct. 2012), pp. 93–111. ISSN: 00457825. DOI: 10.1016/j.cma.2012. 05.021 (cit. on p. 30). [115] A. Shenoi et al., eds. Sailing Yacht Design. Vol. 2: 17th Interna- tional Ship and Offshore Structures Congress. (Seoul, Korea). ISSC. 2009. Chap. Committee V.8, pp. 433–493. URL: http://139.30. 101 . 246 / ISSC2012 / Material / 2009 / SAILING % 20YACHT % 20DESIGN.pdf (cit. on p. 206). [116] J. C. Simo and M. S. Rifai. “A class of assumed strain methods and the method of incompatible modes”. In: International Journal for Numerical Methods in Engineering 29 (1990), pp. 1595–1638. DOI: 10.1002/nme.1620290802 (cit. on p. 106). [117] J. C. Simo, R. L. Taylor, and K. s. Pister. “Variational and Projec- tion Methods for the volume constraint in ﬁnite deformation elasto- plasticity”. In: Computer Methods in Applied Mechanics and Engi- neering 51 (1985), pp. 177–208. DOI: 10.1016/0045-7825(85) 90033-7 (cit. on p. 106). [118] J.C. Simo and D.D. Fox. “On a stress resultant geometrically ex- act shell model. Part I: Formulation and optimal parametrization”. In: Computer Methods in Applied Mechanics and Engineering 72.3 (Mar. 1989), pp. 267–304. ISSN: 00457825. DOI: 10.1016/0045- 7825(89)90002-9 (cit. on pp. 65, 66). [119] Dan Spurr. “The Hydrodynamicist. Winged Victory”. In: Profes- sional Boatbuilder 121 (2009), pp. 50–61. URL: https : / / www . proboat.com/2009/09/winged-victory/ (cit. on p. 200). [120] Gary Mitchell Stanley. “Continuum-based shell elements”. PhD thesis. Stanford University, 1985, p. 515. URL: http : / / searchworks.stanford.edu/view/1190339 (cit. on pp. 33, 65, 82, 193, 194). [121] Hanryk Stolarski, Ted Belytschko, and Sang-Ho Lee. “A review of shell ﬁnite elements and corotational theories”. In: Computational Mechanics Advances 2.2 (1995), pp. 125–212. ISSN: "0927-7951" (cit. on pp. 21, 33). [122] D.J. Struik. Lectures on Classical Differential Geometry. 2nd ed. Dover, 1988. URL: https://store.doverpublications.com/ 0486656098.html (cit. on pp. xxv, 34). 261 BIBLIOGRAPHY [123] R.L. Taylor, E. Oñate, and P.-A. Ubach. “Textile Composites and Inﬂatable Structures”. In: ed. by E. Oñate and B. Kröplin. Vol. 3. Computational Methods in Applied Sciences. P.O. BIBLIOGRAPHY “Advances of the New Rotation-Free Fi- nite Element Shell Triangle Using Accurate Geometrical Data”. In: 9th US National Congress on Computational Mechanics. Vol. CD- ROM. University of California at Berkeley, USACM. San Francisco, California, USA, July 2007 (cit. on p. 38). 262 BIBLIOGRAPHY [132] P.-A. Ubach and E. Oñate. “New Rotation-Free Finite Element Shell Triangle Accurately Using Geometrical Data”. In: Computer Methods in Applied Mechanics and Engineering 199 (5-8 2009): Computational Geometry and Analysis, pp. 383–391. DOI: 10 . 1016/j.cma.2009.01.006 (cit. on pp. 27, 54, 55, 159). [133] Pere-Andreu Ubach. “Análisis y cálculo de membranas mediante el método de los elementos ﬁnitos. Búsqueda de la aproximación ini- cial.” Tesina Final de Carrera. Escola Tècnica Superior d’Enginyers de Camins, Canals i Ports de Barcelona, Sept. 2001. URL: http: //bibliotecnica.upc.es/cddecma/colleccions/tesines/ Divulgacio/Tesines/F2001/EC/Resums/Ubach_Fuentes. htm (cit. on pp. 3, 4, 20). [134] J.G. Valdés, E. Oñate, and J. Miquel. Nonlinear Analysis of Orthotropic Membrane and Shell Structures Including Fluid- Structure Interaction. Monograph 107. Barcelona: CIMNE, Dec. 2007. 216 pp. ISRN: 978-84-96736-37-5. URL: https://books. cimne.com/shop/nonlinear-analysis-of-orthotropic- membrane - and - shell - structures - including - fluid - structure-interaction/ (cit. on p. 20). [135] Eduard Ventsel and Theodor Krauthammer. Thin Plates and Shells: Theory, Analysis, and Applications. Ed. by Pennsylva- nia State University. New York: Taylor & Francis, 2001. ISBN: 0824705750. DOI: 10.1201/9780203908723 (cit. on pp. 22, 160). [136] Charles M. Vest. "Pursuing the Endless Frontier: Essays on MIT and the Role of Research Universities". "The MIT Press", 2005. ISBN: 0-262-22072-5. URL: https : / / mitpress . mit . edu / books/pursuing-endless-frontier (cit. on p. 175). [137] Ignazio Maria Viola, Raffaele Ponzini, and Giuseppe Passoni. “Sail- ing Past a Billion”. In: ANSYS Advantage 3 (2 2009), pp. 47–48. URL: http://www.ansys.com/magazine/issues/vol3-iss2- 2009/cilea.pdf (cit. on p. 205). [138] Gerald Wempner. “Finite elements, ﬁnite rotations and small strains of ﬂexible shells”. In: International Journal of Solids and Structures 5.2 (1969), pp. 117–153. DOI: 10 . 1016 / 0020 - 7683(69)90025-0 (cit. on p. 83). [139] E. L. Wilson et al. “Incompatible displacement models”. In: Numeri- cal and Computer Methods in Structural Mechanics (1973), pp. 43– 57. DOI: 10.1016/B978- 0- 12- 253250- 4.50008- 7 (cit. on p. 106). [140] K. Wisniewski and E. Turska. “Enhanced Allman quadrilateral for ﬁnite drilling rotations”. In: Computer Methods in Applied Me- chanics and Engineering 195.44-47 (Sept. 2006), pp. 6086–6109. ISSN: 00457825. DOI: 10 . 1016 / j . cma . 2005 . 11 . BIBLIOGRAPHY 003. URL: http : / / linkinghub . elsevier . com / retrieve / pii / S0045782505004913 (cit. on p. 102). 263 BIBLIOGRAPHY [141] Henry T Y Yang et al. “A survey of recent shell ﬁnite elements”. In: International Journal for Numerical Methods in Engineering 47.1-3 (Jan. 2000), pp. 101–127. ISSN: 0029-5981. DOI: 10.1002/(SICI) 1097-0207(20000110/30)47:1/3<101::AID-NME763>3.0. CO;2-C (cit. on pp. 21, 24, 83). [142] Y.X. Zhou and K.Y. Sze. “A geometric nonlinear rotation-free tri- angle and its application to drape simulation”. In: International Journal for Numerical Methods in Engineering 89.4 (Jan. 2012), pp. 509–536. ISSN: 00295981. DOI: 10.1002/nme.3250 (cit. on pp. 25, 28, 29, 140). [143] O.C. Zienkiewicz and R.L. Taylor. Finite Element Method. 5th ed. Vol. 2. Oxford, UK: Butterworth-Heinemann, 2000. URL: https:// www.elsevier.com/books/the-finite-element-method- for - solid - and - structural - mechanics / zienkiewicz / 978-0-08-045558-7 (cit. on p. 33). [144] O.C. Zienkiewicz and R.L. Taylor. El Método de los Elementos Fini- tos. Ed. by E. Oñate. 5th ed. Vol. 2. CIMNE, 2004. URL: https: //books.cimne.com/shop/el-metodo-de-los-elementos- finitos-vol-2-mecanica-de-solidos/ (cit. on pp. 20, 21). [145] Francisco Zárate and Eugenio Oñate. “Extended rotation-free shell triangles with transverse shear deformation effects”. In: Computa- tional Mechanics 49.4 (Nov. 2011), pp. 487–503. ISSN: 0178-7675. DOI: 10.1007/s00466-011-0653-y (cit. on pp. 16, 165).
https://openalex.org/W2686808192	https://europepmc.org/articles/pmc5479541?pdf=render	English	null	Comparison study of exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method	PloS one	2,017	cc-by	7,430	RESEARCH ARTICLE OPEN ACCESS Citation: Lee KH (2017) Comparison study of exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method. PLoS ONE 12(6): e0179351. https://doi.org/10.1371/journal.pone.0179351 Editor: Xiao-Dong Wang, North China Electric Power University, CHINA Editor: Xiao-Dong Wang, North China Electric Power University, CHINA Received: March 20, 2017 Accepted: May 26, 2017 Published: June 21, 2017 Copyright: © 2017 Kyun Ho Lee. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Comparison study of exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Kyun Ho Lee* Department of Aerospace Engineering, Sejong University, Seoul, Republic of Korea Kyun Ho Lee* Department of Aerospace Engineering, Sejong University, Seoul, Republic of Korea a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * khlee0406@sejong.ac.kr Abstract A space propulsion system is important for the normal mission operations of a spacecraft by adjusting its attitude and maneuver. Generally, a mono- and a bipropellant thruster have been mainly used for low thrust liquid rocket engines. But as the plume gas expelled from these small thrusters diffuses freely in a vacuum space along all directions, unwanted effects due to the plume collision onto the spacecraft surfaces can dramatically cause a deterioration of the function and performance of a spacecraft. Thus, aim of the present study is to investigate and compare the major differences of the plume gas impingement effects quantitatively between the small mono- and bipropellant thrusters using the compu- tational fluid dynamics (CFD). For an efficiency of the numerical calculations, the whole calculation domain is divided into two different flow regimes depending on the flow charac- teristics, and then Navier-Stokes equations and parallelized Direct Simulation Monte Carlo (DSMC) method are adopted for each flow regime. From the present analysis, thermal and mass influences of the plume gas impingements on the spacecraft were analyzed for the mono- and the bipropellant thrusters. As a result, it is concluded that a careful understanding on the plume impingement effects depending on the chemical characteristics of different propellants are necessary for the efficient design of the spacecraft. Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method on the characteristics of the liquid propellants, the low thrust liquid rocket engines can be divided into two main types as seen in Fig 1 [1]: a monopropellant thruster which needs one single fuel which decomposes into hot gas when in contact with the solid catalyst, and a bipro- pellant thruster which requires a fuel and oxidizer separately to generate a high-pressure hot combustion gas [1]. As the exhaust plume gas flow expands freely in a vacuum space environ- ment along all directions and collides with spacecraft surfaces directly shown in Fig 2 [2], the plume gas impingement can cause adverse influences during mission operations of the space- craft, such as a disturbing force/torque, severe thermal loading, and critical contamination of sensitive components and sensors [2]. Since several failures of the function and performance of the spacecraft due to the plume effects have been reported, an accurate assessment and a minimization of undesirable plume impingement effects are a requisite verification process for the spacecraft design when using the low thrust liquid rocket engines [2]. For an efficient pre- diction, various numerical methods have been developed recently to simulate the physical characteristics of the plume gas flowfields rather than experimental approaches [2]. Among such numerical methods, the Direct Simulation Monte Carlo (DSMC) method [3,4] has been majorly used to analyze the plume gas flow because it can predict an accurate flowfields in a rarefied transition and a free molecular regime under a vacuum environment space by solving the Boltzmann equation statistically. Thus, various studies have investigated the exhaust plume gas flow behaviors of the small thrusters using the DSMC method [5–16]. Abbreviations: dv1!, molecules of class with velocity v1; dΩ, elementary solid angle; F~, external force vector; f, probability density function of molecules; M, molecular mass [g/mol]; m, mass [kg]; n, number density [1/m3]; p, pressure [N/m2]; R, gas constant [8314 J/kmol K]; r, radial direction distance [m]; r~, position vector of molecules; T, temperature [K]; t, time [sec]; v~, velocity vector of molecules; vr, relative velocity of molecules; x, x direction distance [m]; Y, mass fraction; y, y direction distance [m]; z, z direction distance [m]; ρ, density [kg/m3]; σ, collision cross section; , post-collision value; i, ith chemical species. Introduction Funding: This research is supported by the Space Core Technology Development Program of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (MSIP) (NRF-2013M1A3A3A02042426). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Generally, a space propulsion system generates a reaction force (thrust) to perform normal mission operations of a spacecraft through the ejection of a specific amount of a high speed propellant gas flow into a vacuum space environment. Among the various system types, a low thrust liquid rocket engine, or a small thruster, provides a precise thrust or/and impulse bit for attitude control, drag make-up and orbit transfer maneuvers of the spacecraft by generating the kinetic energy of the propellant gas flow through a nozzle expansion from the chemical reaction energy of the liquid propellants in the combustion chamber. This exhaust combustion gas flow in high-temperature and high-pressure is defined as the plume gas flow. Depending Competing interests: The author has declared that no competing interests exist. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 1 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 However, the mono- and bipropellant thrusters have been employed as the representative space propulsion systems for the various spacecraft, quantitatively comparison studies of the exhaust plume flow impingement effects of small mono- and bipropellant thrusters have not been investigated yet based on a literature review. Actually, major selection criteria of the pro- pellant and its operating system have been usually focused on the performance parameters such as a thrust level and a specific impulse, etc. rather than the exhaust plume effects. But as the plume gas flows generated from these two propellants have different characteristics in terms of chemical, thermal and fluid properties, it would be anticipated that their collisions consequently cause different impingement effects on the spacecraft. Because the evaluation of Fig 1. Examples of small mono- and bipropellant thrusters. (A) Monopropellant thruster (Hydrazine propellant) [1]. (B) Bipropellant thruster (MMH-NTO propellant) [1]. https://doi.org/10.1371/journal.pone.0179351.g001 Fig 1. Examples of small mono- and bipropellant thrusters. (A) Monopropellant thruster (Hydrazine propellant) [1]. (B) Bipropellant thruster (MMH-NTO propellant) [1]. Fig 1. Examples of small mono- and bipropellant thrusters. (A) Monopropellant thruster (Hydrazine propellant) [1]. (B) Bipropellant thruster (MMH-NTO propellant) [1]. https://doi.org/10.1371/journal.pone.0179351.g001 2 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 2. Plume impingement concept of small thruster [2]. https://doi.org/10.1371/journal.pone.0179351.g002 Fig 2. Plume impingement concept of small thruster [2]. https://doi.org/10.1371/journal.pone.0179351.g002 https://doi.org/10.1371/journal.pone.0179351.g002 the plume gas flow influences is one of the essential design verification tasks, understanding fundamental characteristics of the overall plume impingement effects depending on the differ- ent propellant types can provide useful information for reducing a development cost and time at the initial design phase. the plume gas flow influences is one of the essential design verification tasks, understanding fundamental characteristics of the overall plume impingement effects depending on the differ- ent propellant types can provide useful information for reducing a development cost and time at the initial design phase. Thus, the aim of the present study is to investigate and compare the major differences of the plume gas flow impingement effects quantitatively between small mono- and bipropellant thrusters using the computational fluid dynamics (CFD). For an efficiency of the numerical calculations, the whole calculation domain is divided into two different flow regimes depend- ing on the flow characteristics, and then Navier-Stokes equations and parallelized Direct Simu- lation Monte Carlo (DSMC) method are adopted for each flow regime sequentially so that individually calculated results could be combined as boundary conditions for other methods. Through the present analysis results, thermal and mass influences of the plume gas impinge- ments on the spacecraft including number flux and heat flux were investigated between the mono- and the bipropellant thrusters. As a result, it is anticipated that the present study could provide practically useful information to the related engineers on determining the proper pro- pulsion system type and evaluating the spacecraft system design through investigations of the plume impingement effects of the mono- and bipropellant thrusters. Numerical methodology Parallel Direct Simulation Monte Carlo method PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Parallel Direct Simulation Monte Carlo method The flow regimes can be generally characterized by the Knudsen number (Kn), which depicts a comparison between an actual characteristic length and a mean free path of gas particles. When the Knudsen number of a specific flow is greater than a unity, the mean free path of flow molecules can be comparable to an actual length of the physical problem. Therefore, a flow model based on a continuum assumption such as the N-S equation cannot be available for a good approximation any more in case of the high Knudsen number flow. Thus, a kind of molecular dynamic simulations with statistical methods should be employed for the rarefied flow regime to deal with the Boltzmann equation in a nonlinear form of Eq (1) [3,4] @ @t ðnf Þ þ~v @ @~r ðnf Þ þ ~F @ @~v ðnf Þ ¼ R 1 1 R 4p 0 n2ðf f 1 ff1ÞvrsdO dv1 ! ð1Þ ð1Þ In Eq (1), n,~v,~r, f and vr are number density, velocity vector, position vector, probability PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 3 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method density function and relative velocity of molecules, respectively. Also, ~F, d~v1, dO and σ are external force vector, molecules of class with velocity v1, elementary solid angle and collision cross section [3]. Among various molecular dynamic simulation methods, the Direct Simula- tion Monte Carlo (DSMC) method proposed by Bird [3] is regarded as the most effective tech- nique for solving the Boltzmann equation to deal with the rarefied flow. The DSMC is a statistical particle simulation method based on kinetic theory, which uses the representative particles to trace in space and time to simulate the physical behaviors of the real gas [3]. But the DSMC method has some drawbacks because it should evaluate numerous intermolecular interactions between all the simulated particles. For example, it is generally found that the DSMC method required a much longer computational time than the conventional continuum flow models because it considered all the interactions of every simulated molecules sequen- tially. As the present study intends to analyze the plume impinging influences on a complex three dimensional unstructured grids of the actual satellite configuration, a parallelized DSMC code was employed to increase computational efficiencies such as a faster calculation time with 28 CPU cores. Results and discussion For efficiency and accuracy of the numerical calculations, the calculation domain was com- posed of two different flow regimes depending on the flow characteristics, which were a 2-D axisymmetric continuum flow domain inside the nozzle and a 3-D rarefied plume gas flow domain under the vacuum environment, respectively. By doing this, appropriate numerical methods could be combined and applied to each subdomain sequentially to use numerical solutions obtained from one method as the boundary conditions for others. Parallel Direct Simulation Monte Carlo method The variable hard sphere (VHS) model [3] is used as the intermolecular- collision model and the no-time counter (NTC) method is for the collision sampling technique [3]. The Larsen-Borgnakke model [17] is employed to redistribute the translational and the internal energy exchange between the gas molecules. In addition, the unstructured three dimensional grid system of the computation domain was modeled using commercial software, GRIDGEN that adopts the Delaunay technique [18]. The parallel processing was made by dividing the computational domain into several subdomains by using the MeTiS library, which is based on the k-way, n-partitioning technique by Karypis [19]. After each time step of Δt = 1×10−7 sec, information about the particles and their properties was exchanged through the subdomain boundary by using the Message Passing Interface (MPI) library [20]. Also, flow properties were averaged over a large number of sampling steps about 10,000 to minimize sta- tistical scatter. A steady state was typically completed after 500 transient steps, and additional 5,000 sampling steps were conducted to obtain the time averaged flow properties. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Inlet and boundary conditions of plume simulation As the plume gas behaviors in the vacuum are dominantly affected by the continuum nozzle flow inside the thruster, an accurate estimation of the plume flowfields was required to define the inlet boundary condition at the nozzle exit plane for the plume analysis. Thus, the N–S equations were solved numerically to predict continuum gas flow distributions inside the thruster nozzle. The small thruster configuration considered in this study is illustrated in Fig 3. It has a conical nozzle with an expansion ratio of 50:1 and it assumed to be provide a five new- ton reaction force at the stagnated chamber pressure (pc) = 1.45 MPa [16]. For the boundary conditions, the stagnation conditions inside the chamber, such as mole fractions of the combustion gas species, molecular masses of gas mixture, and adiabatic flame temperatures, were calculated as given in Table 1 by solving the chemical equilibrium reaction equations and then they were specified as an inlet condition of the nozzle [16]. Also, ten kinds PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 4 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 3. Small thruster configuration [10,16]. https://doi.org/10.1371/journal.pone.0179351.g003 Fig 3. Small thruster configuration [10,16]. https://doi.org/10.1371/journal.pone.0179351.g003 https://doi.org/10.1371/journal.pone.0179351.g003 of product gas species in Table 1 were regarded as a mixture of perfect gases in Eq (2), and their compositions were treated as a chemically frozen flow during the nozzle expansion pro- cess [16]. p ¼ rRT X N i¼1 Yi Mi ð2Þ ð2Þ Here, Yi and Mi are the mass fraction and the molecular mass [g/mol] of the product gases, and R is the universal gas constant (8314.41 J/kmol K), respectively. As a result, the propellant gas flowfields at the nozzle exit plane can be plotted as Fig 4 including the two velocity components, density, and temperature [16]. First, the temperature of the bipropellant gas is observed much higher than that of the monopropellant because the adiabatic temperature of MMH-NTO reaches about 3,087 K in the chamber while the decom- position temperature of the hydrazine is only about 1,343 K as seen in Table 1. For example, the gas temperatures at the center of the nozzle exit plane were estimated roughly as 260 K for hydrazine and 690 K for MMH-NTO propellant, respectively. As a result, higher velocities were also estimated for the MMH-NTO thruster because the exhaust velocity tends to increase Table 1. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Inlet and boundary conditions of plume simulation Chemical equilibrium reaction results of hydrazine and MMH-NTO [16]. Results Hydrazine MMH-NTO Mole fractions of combustion gas species H2 0.35761 0.15657 N2 0.27152 0.30513 NH3 0.37087 - H2O - 0.32741 CO - 0.13145 CO2 - 0.03628 H - 0.02133 NO - 0.00235 O - 0.00131 OH - 0.01709 O2 - 0.00108 Molecular mass of gas mixture [g/mol] 14.62 20.46 Adiabatic ﬂame temperature [K] 1342.8 3087.4 https://doi.org/10.1371/journal.pone.0179351.t001 Table 1. Chemical equilibrium reaction results of hydrazine and MMH-NTO [16]. Results Hydrazine MMH-NTO Mole fractions of combustion gas species H2 0.35761 0.15657 N2 0.27152 0.30513 NH3 0.37087 - H2O - 0.32741 CO - 0.13145 CO2 - 0.03628 H - 0.02133 NO - 0.00235 O - 0.00131 OH - 0.01709 O2 - 0.00108 Molecular mass of gas mixture [g/mol] 14.62 20.46 Adiabatic ﬂame temperature [K] 1342.8 3087.4 https://doi.org/10.1371/journal.pone.0179351.t001 Table 1. Chemical equilibrium reaction results of hydrazine and MMH-NTO [16]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 5 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 4. Exhaust plume flowfields at the thruster nozzle exit plane by N-S equations [16]. https://doi.org/10.1371/journal.pone.0179351.g004 Fig 4. Exhaust plume flowfields at the thruster nozzle exit plane by N-S equations [16]. Fig 4. Exhaust plume flowfields at the thruster nozzle exit plane by N-S equations [16]. https://doi org/10 1371/journal pone 0179351 g004 Fig 4. Exhaust plume flowfields at the thruster nozzle exit plane by N-S equations [16]. https://doi.org/10.1371/journal.pone.0179351.g004 proportional to the chamber temperature based on the general rocket performance equation. However, a gas flow with a higher density was produced for the hydrazine decomposition opposite to the temperature distribution. The reason is that the internal nozzle flow was assumed to obey the perfect gas law in Eq (2), which is defined as an inverse relation between the density and the temperature at a given pressure. Moreover, some drastic variations of the flowfields were found in the given profiles nearby the nozzle wall due to the boundary layer effect. For the three dimensional DSMC simulation of the present study, the nozzle exit plane is chosen as the breakdown face between the continuum flow and the rarefied flow domains, and this assumption is suitable for most nozzles when the exit-to-throat area ratio are not very large [9]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Plume flow impingement effects on spacecraft The overall configuration of the satellite considered in the present study is shown in Fig 5A. A large sized solar panel was installed onto the side of the satellite following a longitudinal axis direction to generate higher electrical power from the sun. A numerical grid system for the bottom platform of the satellite and the major components is given in Fig 5B. One S-band antenna and four reaction control thrusters were located on the bottom platform, which were surrounded by a launch vehicle interface ring. Additionally, two antenna modules for a preci- sion orbit determination (POD) were modeled simply adjacent to the thrusters. Part of the solar panel was attached vertically to the bottom platform of the satellite outside the ring with a 1 m height. The nozzle exit plane of each thruster was canted outward to the vacuum space to provide the necessary thrust and momentum for efficient attitude control as shown in Fig 5C. The three-dimensional unstructured geometry was generated with the GRIDGEN com- mercial software [18] using about 24,000 nodes and 120,000 tetrahedron cells. The cell size at the vicinity of the thruster exit was generated sufficiently smaller than the local mean free path to guarantee good accuracy. To simulate the vacuum boundary condition, the computational domain was extended to 3.2 m in radius and 2.0 m in height from the center of the bottom platform of the satellite. Because the height of the solar panel is higher than the interface ring and two kinds of antennas were located beside the thrusters, all the components including the satellite structures have a strong possibility of being exposed directly to the exhausted plume gas flow. To reflect the maximum plume impingement effects, the present study considered a simultaneous firing condition of all four thrusters as a representative case. As a first outcome, the overall plume flow behaviors in the vacuum space, such as the veloc- ity streamline, number density and overall temperature, were compared for both propellants. The velocity streamlines of the mono- and bipropellant plume gases at the center cross section of the platform are shown in Fig 6. Both hydrazine and MMH-NTO plume gases expanded with similar patterns as the simulated plume particles, which were initially injected from the four nozzle exits of the firing thrusters, spread primarily outward to the vacuum space far from the satellite. impinging plume particles. The remaining boundaries were assumed to be a particle sink to simulate a vacuum condition. impinging plume particles. The remaining boundaries were assumed to be a particle sink to simulate a vacuum condition. impinging plume particles. The remaining boundaries were assumed to be a particle sink to simulate a vacuum condition. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Inlet and boundary conditions of plume simulation Then the continuum flowfield results at the nozzle exit in Fig 4, which were obtained by the N-S equations, were applied directly as the inflow boundary condition for the DSMC method together with the gas mixture compositions given in Table 1 using the Maxwell distri- bution function based on the previous researches [5–14]. In the present study, a steady state flow is considered and the plume flow is assumed as a mixture of single-phase ideal gases with- out any solid/liquid particulates. Also, a nozzle outside space is assumed as a vacuum condition while the gas pressures at the nozzle exit plane is much higher, which means there is no inverse flux of a free stream into the nozzle inside through the nozzle exit plane. And most of the exhaust flow at the nozzle exit plane is a supersonic condition while a subsonic flow region is very limited. As a result, one-way coupling approach is employed between the N-S equations and the DSMC interface based on the previous researches [5–14]. For the three dimensional DSMC simulation of the present study, the continuum flowfield results at the nozzle exit in Fig 4 were applied directly as inlet conditions together with the gas mixture compositions given in Table 1. In the case of the solid boundaries, the satellite struc- ture and components were all modeled as a diffusely reflected surface with complete energy accommodation. Additionally, the coldest surface temperatures of each component and the structure during satellite operation listed in Table 2, which were obtained from an orbit simu- lation of the satellite, were used as a solid wall boundary temperature in the present DSMC analysis to estimate heat fluxes on the satellite surfaces and components caused by the PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 6 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Table 2. Coldest Temperature of satellite components and structure. Satellite components Temperature [K] Thruster 158.0 Solar panel 200.0 POD antenna 155.0 S-band antenna 155.0 Bottom platform 142.0 Launch vehicle adapter ring 151.0 https://doi.org/10.1371/journal.pone.0179351.t002 Table 2. Coldest Temperature of satellite components and structure. https://doi.org/10.1371/journal.pone.0179351.t002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Plume flow impingement effects on spacecraft Because the plume particles expanded simultaneously, some portion of them merged together and formed a main flow stream at the center of the bottom platform for both propellants. Then, a considerable amount of the plume particles directly collided with the solar panel because a large portion of the panel area was exposed directly to a main stream of the exhaust plume flow. Thus, undesirable plume effects including disturbance force/torque, heat load, and chemical species deposition could become critical at the solar panel due to severe plume gas impingement. Moreover, it was predicted that a small amount of plume backflow, PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 7 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 5. Calculation domain of plume impingement for DSMC method. (A) Example of the satellite configuration. (B) Computational grid with satellite bottom platform. (C) Detailed grid inside bottom platform. https://doi.org/10.1371/journal.pone.0179351.g005 Fig 5. Calculation domain of plume impingement for DSMC method. (A) Example of the satellite configuration. (B) Computational grid with satellite bottom platform. (C) Detailed grid inside bottom platform. https://doi.org/10.1371/journal.pone.0179351.g005 https://doi.org/10.1371/journal.pone.0179351.g005 which was directed to the POD and S-band antenna components inside the interface ring, could cause additional impingement effects. Regarding to the density distribution of the plume gas, Fig 7A shows that the monopropel- lant hydrazine plume gas flow spread more densely all over the calculation domain including the backflow regions inside the interface ring and outside the solar panel than that of the bipropellant MMH-NTO gas shown in Fig 7B because a higher density profile was initially applied at the inflow condition of the DSMC method based on the continuum flow results inside the thruster nozzle. This can be confirmed more clearly when converted into the num- ber density distributions shown in Fig 8. It was estimated that the number density of the hydrazine plume particles ranged roughly between 1.0E+18 1/m3 and 3.0E+19 1/m3 near the antenna components and inside the interface ring shown in Fig 8A, while it was less than 1.0E +19 1/m3 for the MMH-NTO plume gas from Fig 8B. Additionally, the order of the number PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 8 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 6. Velocity streamlines of plume gas flow. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Plume flow impingement effects on spacecraft (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g006 Fig 6. Velocity streamlines of plume gas flow. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g006 Fig 6. Velocity streamlines of plume gas flow. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g006 https://doi.org/10.1371/journal.pone.0179351.g006 density at the core region of the main plume stream was over 1.0E+20 1/m3 for the hydrazine gas and 5.0E+19 1/m3 for the MMH-NTO gas, respectively. As a result, a larger number of hydrazine plume particles were found to be distributed in the proximity of the solar panel and in the backflow regions inside the interface ring. Thus, it was anticipated that the high number density of the hydrazine plume particles could cause a significant increase in the collision pos- sibility of the plume particles onto the spacecraft components and structure including the solar panel rather compared to the MMH-NTO plume gas. In addition, the difference in the overall temperatures between the hydrazine and MMH-NTO plume flow gases were compared in Fig 9. In contrast to the number density, a higher temperature plume gas for the MMH-NTO propellant was observed to be distributed over the whole calculation domain according to the continuum flow temperature profile at the nozzle exit plane shown in Fig 4. The MMH-NTO plume particles ranged between 600 K ~ 1,000 K around the antenna components inside the interface ring from Fig 9B, whereas lower temperatures were calculated for the hydrazine propellant that ranged between 200 K ~ 500 K PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 9 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 7. Density distributions of plume gas flow [kg/m3]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g007 Fig 7. Density distributions of plume gas flow [kg/m3]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. Fig 7. Density distributions of plume gas flow [kg/m3]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g007 https://doi.org/10.1371/journal.pone.0179351.g007 https://doi.org/10.1371/journal.pone.0179351.g007 as seen in Fig 9A. Moreover, high temperature regions were observed in the vicinity of the upper portion of the solar panel because a considerable amount of thermal energy from the plume gas molecules was converted from their reduced kinetic energy which was proportional to the increased collision of the plume particles on the surface of the solar panel. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Plume flow impingement effects on spacecraft Thus, the PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 10 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 8. Number density distributions of plume gas flow [1/m3]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g008 Fig 8. Number density distributions of plume gas flow [1/m3]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g008 https://doi.org/10.1371/journal.pone.0179351.g008 plume gas temperature varied gradually from 300 K to 800 K for the hydrazine and from 400 K to 1,200 K for the MMH-NTO following the height of the solar panel. These results indicate that the plume gas temperature was greatly dependent on the amount of thermal energy released from the chemical reactions of the propellant. Thus, it was anticipated that plume gas PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 11 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 9. Overall temperature distributions of plume gas flow [K]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi org/10 1371/journal pone 0179351 g009 Fig 9. Overall temperature distributions of plume gas flow [K]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g009 impingement could exert a severe thermal loading influence on the spacecraft components and structure depending on the type of propellant used. impingement could exert a severe thermal loading influence on the spacecraft components and structure depending on the type of propellant used. As a second outcome, distributions of the number flux and heat flux of the two plume gases on the satellite components and structure were investigated to compare the plume impinge- ment influences. Fig 10 clearly shows that a highly intensive gas particle distribution greater PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 12 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 10. Surface number flux distributions of plume gas flow [1/m2s]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g010 Fig 10. Surface number flux distributions of plume gas flow [1/m2s]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. Plume flow impingement effects on spacecraft https://doi.org/10.1371/journal.pone.0179351.g010 https://doi.org/10.1371/journal.pone.0179351.g010 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 13 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method than 1E+22 1/m2s was found on the upper regions of the solar panel and the S-band antenna because they were directly exposed to the plume flow. Especially, a larger number of the hydra- zine plume gas particles were extensively distributed over the surfaces of the antenna compo- nents inside the interface ring and the solar panel compared to that of the MMH-NTO gas because a more dense combustion gas of the hydrazine propellant was exhausted from the thruster and collided onto the satellite as predicted from the streamline and density behaviors of the plume particles. Additionally, the surface heat flux on the solar panel was similar to the num- ber flux result shown in Fig 11 because its distribution is directly proportional to the number of plume particles impinged on the surfaces of the satellite. However, an intense heat flux over 2,000 W/m2 was generated by the MMH-NTO plume gas on the upper area of the solar panel whereas the maximum value of the hydrazine gas was predicted to be less than about 1,500 W/ m2 which was not relatively high compared with the solar constant of qsol = 1353 W/m2 [21]. Therefore, it was anticipated that a more excessive thermal loading can be transferred to the spacecraft if the MMH-NTO thruster is used because of the higher chemical energy inherent in the bipropellant itself which is similar to the overall temperature results of the plume particles. To consider the influence of the plume backflow in more detail, surface distributions of H2 species were investigated for both the propellants shown in Fig 12. Because it has the lightest molecular weight among the various gas compositions, previous studies revealed that H2 sepa- rated strongly from the main flow stream compared to other gas species, and thus, it was a main gas ingredient of the hydrazine plume backflow [10,12]. Plume flow impingement effects on spacecraft In the case of the hydrazine pro- pellant, Fig 12A shows that H2 species was distributed more spaciously over not only the solar panel but also over the bottom platform region of the satellite inside the interface ring com- pared to the MMH-NTO propellant in Fig 12B because a larger mole numbers of H2 was pro- duced initially from the chemical reaction process of the hydrazine propellant inside the thruster chamber as given in Table 1. Accordingly, this chemical separation of H2 with the higher mole fraction contained in the hydrazine plume backflow also transferred the local heat flux over a more extensive area inside the interface ring compared to the MMH-NTO propel- lant from Fig 11B. As a consequence, it is predicted that the higher collisions of the hydrazine plume particles can cause more critical impacts on the spacecraft in terms of a disturbance force/torque and a contamination of the chemical species because a more dense decomposi- tion gas is produced from the monopropellant hydrazine propellant. Moreover, the compo- nents and structures located adjacent to the hydrazine thruster will be influenced greatly by the H2 molecules because a relatively larger amount of H2 is separated from the main plume stream and included in the backflow as a major species. Thus, surface contamination by the deposition of the plume particles including a considerable amount of H2 onto the spacecraft could be a significant problem for sensitive equipment such as solar cells, optical equipment and GaAs microwave devices when the monopropellant hydrazine thruster is used. For a quantitative comparison, the number flux and the surface heat flux variations of the plume gas were measured at the center of the solar panel following the z-axis for both the pro- pellants and plotted in Fig 13. Similar to the previously examined number density and overall temperature results, these surface flux variations of the two plume gases also increased gradu- ally from the bottom to the top of the panel, while a higher level of the number flux was main- tained for the hydrazine propellant and vice versa for the surface heat flux. As a final result, the disturbance forces and torques induced by the thruster plume impinge- ment was evaluated in Table 3 as relative values to absolute nominal ones. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Plume flow impingement effects on spacecraft From the previous plume streamline and surface number flux distributions of the two propellants, the larger dis- turbance force and torque were predicted to be caused by the hydrazine plume impingement rather than by the MMH-NTO. From Table 3, it could be inferred that the collision between the plume backflow and the satellite structure was more dominant for the hydrazine propellant PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 14 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method because the disturbance forces of the hydrazine propellant were greater in the x and y axes du Fig 11. Surface heat flux distributions of plume gas flow [W/m2]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g011 Fig 11. Surface heat flux distributions of plume gas flow [W/m2]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g011 because the disturbance forces of the hydrazine propellant were greater in the x and y axes due to the considerable amount of exhaust plume particles in the backflow regime with the separa- tion of H2. However, a main stream of the MMT-NTO plume gas with a lower density tended to flow straight following the axial direction of the nozzle rather than diffuse in the radial 15 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method direction, which caused less plume backflow impingement on the satellite. Additionally, a l b f h h d i l i l i d d h l l f di Fig 12. Surface number flux distributions of H2 species [1/m2s]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g012 Fig 12. Surface number flux distributions of H2 species [1/m2s]. (A) Monopropellant hydrazine. (B) Bipropellant MMH-NTO. https://doi.org/10.1371/journal.pone.0179351.g012 direction, which caused less plume backflow impingement on the satellite. Additionally, a larger number of the hydrazine plume particles induced a much more severe level of distur- bance torques in the x and y axes than that of the MMT-NTO gas because of a direct plume impingement effect on the solar panel. Plume flow impingement effects on spacecraft 16 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Fig 13. Surface flux distributions of the plume gas flow at the center of solar panel in z-axis. (A) Surface number flux [1/m2s]. (B) Surface heat flux [W/m2]. https://doi.org/10.1371/journal.pone.0179351.g013 Fig 13. Surface flux distributions of the plume gas flow at the center of solar panel in z-axis. (A) Surface number flux [1/m2s]. (B) Surface heat flux [W/m2]. https://doi.org/10.1371/journal.pone.0179351.g013 Conclusions The present study conducted numerical analysis to investigate and compare major differences of the plume gas flow impingement effects quantitatively between the small mono- and bipro- pellant thrusters. To increase an efficiency of the numerical calculations, the computational 17 / 20 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Table 3. Predictions of relative disturbance force and torque values of two plume gases. Disturbance Force x y z Absolute Nominal Value [N] 0.000 0.000 19.404 Monopropellant Hydrazine [%] 0.103 42.753 0.269 Bipropellant MMH-NTO [%] 0.010 35.549 1.248 Disturbance Torque x y z Absolute Nominal Value [N m] 0.087 0.034 0.000 Monopropellant Hydrazine [%] 1079.057 17.416 0.974 Bipropellant MMH-NTO [%] 756.457 0.751 0.081 https://doi.org/10.1371/journal.pone.0179351.t003 Table 3. Predictions of relative disturbance force and torque values of two plume gases. fluid dynamics (CFD) methods with the N–S equations and the parallelized DSMC method were employed for the different calculation flow regimes depending on the flow conditions. Major differences of the plume gas impingement effects between the two propellants were summarized from the present results. fluid dynamics (CFD) methods with the N–S equations and the parallelized DSMC method were employed for the different calculation flow regimes depending on the flow conditions. Major differences of the plume gas impingement effects between the two propellants were summarized from the present results. fluid dynamics (CFD) methods with the N–S equations and the parallelized DSMC method were employed for the different calculation flow regimes depending on the flow conditions. Major differences of the plume gas impingement effects between the two propellants were summarized from the present results. 1. The monopropellant hydrazine plume gas flow spread more densely all over the calculation domain including the backflow region than that of the MMH-NTO gas because a combus- tion gas with a higher density was produced inside the chamber. Consequently, the higher collisions of the hydrazine plume particles caused more critical impacts for the spacecraft in terms of a disturbance force/torque and a contamination of the chemical species compared to the MMH-NTO propellant. 2. Components and structures located adjacent to the hydrazine thruster was influenced greatly by the H2 molecules because a relatively larger amount of H2 was separated from the main plume stream and was included in the backflow as a major species. Conclusions Thus, surface con- tamination by the plume molecule deposition including a considerable amount of H2 onto the spacecraft could be a significant problem for the any H2 sensitive equipment when the monopropellant hydrazine thruster is used. 3. A much more excessive heat flux was transferred to the spacecraft when the MMH-NTO thruster was used because of a higher chemical energy inherent in the bipropellant itself. As the plume gas temperature was greatly dependent on the amount of thermal energy released from the chemical reactions of the propellant, it was anticipated that plume gas impinge- ment could exert a severe thermal loading influence on the spacecraft components and structure depending on the type of propellant used. From the present analysis results, it could be found that the impingement influences of the plume gas on the spacecraft components and structure were highly correlated with the chemi- cal reaction characteristics of the propellant used. Thus, if small mono- and/or bipropellant thrusters are to be used, then a careful understanding of not only the performance characteris- tics of these two propellants but also the impinging effects of their plume gases is required to design the spacecraft efficiently. Consequently, it is anticipated that the present study could provide practically useful data to the related engineers on the optimized design philosophy of the spacecraft and the selection of the proper propulsion system through detailed investiga- tions of the plume impingement effects of the mono- and bipropellant thrusters, which can finally lead to reduction of development costs and time from the initial design phase. Author Contributions Conceptualization: KHL. References 1. Ley W, Wittmann K, Hallmann W, Handbook of space technology, West Sussex: John Wiley & Sons; 2009. 2. Dettleff G. Plume flow and plume impingement in space technology. Prog Aerosp Sci. 1991; 28: 1–71 https://doi.org/10.1016/0376-0421(91)90008-R 3. Bird GA. Molecular gas dynamics and the direct simulation of gas flows. Oxford: Clarendon Press; 1992. 4. Balaja M, Roohia E, Akhlaghia H, Myong RS, Investigation of convective heat transfer through constant wall heat flux micro/nano channels using DSMC. Int J Heat Mass Tran. 2014; 71: 633–638 https://doi. org/10.1016/j.ijheatmasstransfer.2013.12.053 5. Giordano D, Ivanov M, Kashkovsky A, Markelov G, Tumino G, Koppenwallner G, Application of numeri- cal multizone approach to the study of satellite thruster plumes. J Spacecraft Rockets. 1998; 35: 502– 508 https://doi.org/10.2514/2.3359 6. Ivanov MS, Markelov GN, Gerasimov YI, Krylov AN, Mishina LV, Sokolov EI, Free-flight experiment and numerical simulation for cold thruster plume. J Propul Power. 1999; 15: 417–423 https://doi.org/10. 2514/2.5460 7. Sholes B, Zeller C, Matheson B, Hydrazine thruster plume contamination analysis for the Kepler pho- tometer. The 3rd European Workshop on Hydrazine, Sardinia, Italy: 2004. 8. Park JH, Baek SW, Kim JS, Direct Simulation Monte Carlo analysis of thruster plumes/satellite base region interaction. AIAA J. 2004; 42: 1622–1632 https://doi.org/10.2514/1.1975 9. Xiao Z, Cheng H, Zhou H, Plume interaction in parallel multi-thrusters propulsion system and the effect on backflow. 9th AIAA/ASME Joint Thermophysics and Heat Transfer Conference, Fluid Dynamics and Co-located Conferences, San Francisco, California; 2006 10.2514/6.2006–3599. 10. Lee KH, Lee SN, Yu MJ, Kim SK, Baek SW. Combined analysis of thruster plume behavior in rarefied region by preconditioned Navier-Stokes and DSMC methods. Trans Jpn Soc Aeronaut Space Sci. 2009; 52: 135–143 https://doi.org/10.2322/tjsass.52.135 11. Wartelski M, Scremin G, Theroude C, Astrium tools for plume-spacecraft interaction analysis. Space Propulsion 2010 Conference, San Sebastian, Spain: 2010. 12. Lee KH, Choi SW. Interaction effect analysis of thruster plume on LEO satellite surface using parallel DSMC method. Comput Fluids 2013; 80: 333–341 13. Yim JT, Sibe F, Ierado N, Plume impingement analysis for the European service module propulsion sys- tem. 50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference, Cleveland, Ohio; 2014 10.2514/ 6.2014–3883. 14. Jeon W, Baek S, Park J, Ha D, Rocket Plume Analysis with DSMC Method. J Korean Soc Propuls Eng. 2014; 18: 54–61 https://doi.org/10.6108/KSPE.2014.18.5.054 15. Kashkovsky AV, Vashchenkov PV, Banyai T, Modeling of reentry space vehicle aerodynamics with con- trol thruster plume—free-stream interaction. Thermophys Aeromech+. 2015; 21: 719–728 https://doi. org/10.1134/S0869864314060067 16. Lee KH. Supervision: KHL. Supervision: KHL. Visualization: KHL. Visualization: KHL. Writing – original draft: KHL. Writing – original draft: KHL. Writing – original draft: KHL. Formal analysis: KHL. Funding acquisition: KHL. Investigation: KHL. Methodology: KHL. Methodology: KHL. Conceptualization: KHL. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 18 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method Project administration: KHL. Software: KHL. 17. Borgnakke C, Larsen PS. Statistical collision model for Monte Carlo simulation of polyatomic gas mix- ture. J Comput Phys. 1975; 18: 405–420 https://doi.org/10.1016/0021-9991(75)90094-7 20. Gropp W, Lusk E, Doss N, Skjellum A, A high-performance, portable implementation of the MPI mes- sage passing interface standard. Parallel Comput. 1996; 22; 789–828 https://doi.org/10.1016/0167- 8191(96)00024-5 19. Karypis G, Kumar V, Multilevel k-way partitioning scheme for irregular graphs. J Parallel Distrib Comput. 1998; 48: 96–129 https://doi.org/10.1006/jpdc.1997.1404 18. Pointwise Inc.. Gridgen user manual. Pointwise Inc.; 1997. 21. Modest MF, Radiative heat transfer, New York: McGraw–Hill; 2003. PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 21. Modest MF, Radiative heat transfer, New York: McGraw–Hill; 2003. References Numerical comparison of exhaust plume flow behaviors of small monopropellant and bipropel- lant thrusters. PLoS One. 2017; 12(5): e0176423 https://doi.org/10.1371/journal.pone.0176423 PMID: 28481892 PLOS ONE \| https://doi.org/10.1371/journal.pone.0179351 June 21, 2017 19 / 20 Exhaust plume impingement effects of small mono- and bipropellant thrusters using parallelized DSMC method 20 / 20
https://openalex.org/W2544628411	http://www.journalijar.com/uploads/734_IJAR-12810.pdf	English	null	CHARACTERISTICS OF THE ISRAEL'S SOCIALISTIC ECONOMY.	International journal of advanced research	2,016	cc-by	5,183	Economic Aspects in the Collective Identity:- The economic content world of the collective identity includes fundamental traits of the economic action, which are related, directly or indirectly, to the Zionist ideology and the ideological goals in light of which Jewish society in the Land of Israel developed, and to constraints dictated by the reality in the Land of Israel. Therefore, the discussion of the economic content world of the Zionist collective identity should begin with the presentation of the three characteristics that accompanied the Zionist economic activity for a long period of time. First, this activity was dictated by the national goals, and the good of the nation was the supreme consideration. Second, the economic feasibility of this activity was most slight. Third, this activity was connected in a close and multifaceted manner to politics. Indeed, many researchers will agree with the statement that the main trait of the economic sector in Israel from its first days was its considerable subordination to the national-political imperatives and to the organizations and institutions that represented these imperatives, namely the parties and the pan-national institutions. This combination of characteristics created tension between economic considerations that derived from the market imperatives and its conditions and political considerations that derived from the aspiration to realize the national ideology. This tension was the fundamental problem of the economy of the Jewish community in the Land of Israel and of the State of Israel for many years, and it is apparent in a number of dimensions of economic activity2. CHARACTERISTICS OF THE ISRAEL'S SOCIALISTIC ECONOMY. Copy Right, IJAR, 2016,. All rights reserved. …………………………………………………………………………………………………….... When Jewish settlements began to be established in the Land of Israel in large numbers, with the goal of building a future State, at the end of the 19th century and at the beginning of the 20th century, most of the immigrants and the political leaders were from the Soviet countries in Eastern Europe. Consequently, a collectivist social opinion began to form that encourages the subordination of personal needs in favor of collective needs for the building and development of a new Jewish State1. Corresponding Author:-Daniel Galily. Address:-Dean of student's affairs, Los Angeles University, United States. 594 Corresponding Author:-Daniel Galily. Address:-Dean of student's affairs, Los Angeles University, United States. ISSN: 2320-5407 ISSN: 2320-5407 Int. J. Adv. Res. 4(10), 594-600 Journal Homepage: -www.journalijar.com Article DOI:10.21474/IJAR01/1835 DOI URL: http://dx.doi.org/10.21474/IJAR01/1835 INTERNATIONAL JOURNAL OF ADVANCED RESEARCH (IJAR) Journal Homepage: -www.journalijar.com Article DOI:10.21474/IJAR01/1835 DOI URL: http://dx.doi.org/10.21474/IJAR01/1835 INTERNATIONAL JOURNAL OF ADVANCED RESEARCH (IJAR) ISSN: 2320-5407 Int. J. Adv. Res. 4(10), 594-600 ISSN: 2320-5407 ISSN: 2320-5407 Most of the Zionist goals had economic implications, whether direct or indirect. One economic implication of the goal to establish a new Jewish society in the Land of Israel, which would be different from the Jewish society in the Diaspora, was the social-economic ‘rehabilitation’ of the Jews: the establishment of a productive occupational structure that means the referral of most of the immigrants to manual labor and the manufacturing of products with economic worth. A second implication was the construction of a modern economic sector, from the foundations, since in the Ottoman Land of Israel there barely was an infrastructure for modern economic activity. The technical and administrative infrastructure was largely built by the British Mandate government, according to the Mandate given to Britain. However, the Jewish settlers strived to establish a productive economy and a system of services, and in this, too, they had to begin from the foundations, since the economy of the old Jewish community was primarily based on support from the outside and there was very little independent productive activity, and in any event the new arrivals sought to disconnect themselves from the old community3. The economic significance of the Jewish settlement in the Land of Israel was the extensive investments in the purchase and preparation of lands, in the construction of infrastructures for the communities (building homes, planting orchards, etc.), and support of the communities until they developed a productive economy of their own. The immigration of Jews to the Land of Israel also had economic significance, and primarily for the investment in the absorption of the immigrants who came without their own capital: assurance of occupation and housing, professional training, so as to integrate the immigrants in the productive economic activity, and different services – including those that constitute processes for the socialization of the immigrant, for example language instruction. However, in the period that preceded the establishment of the State of Israel immigrants with capital also came and could finance their absorption, but the lack of appropriate central planning prevented the intelligent use of this capital and induced crises. Most of this economic activity was not profitable, since the investments led to national-political benefit but not financial profit. 3Plessner, Y. (1994). The Political Economy of Israel: From Ideology to Stagnation, p.72. Google Books. Retrieved on September 8. 2011. 4Kimmerling, B. (2001). The End of the Ashkenazi Hegemony. Jerusalem: Keter Press. (Hebrew) 5Mark, C. (2004). "Israel: US Foreign Assistance”, Congressional Research Service, July 12, 2004. Retrieved July19, 2012. The Tension between the Market Imperatives and the National Imperatives:- 594 1Smith, B. J. (Ed.) (1993). The Roots of Separatism in Palestine: British Economic Policy, 1920-1929. 2Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (Hebrew) . Corresponding Author:-Daniel Galily. Address:-Dean of student's affairs, Los Angeles University, United States. 1Smith, B. J. (Ed.) (1993). The Roots of Separatism in Palestine: British Economic Policy, 1920-1929. 2Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (Hebrew) . 594 Corresponding Author:-Daniel Galily. Address:-Dean of student's affairs, Los Angeles University, United States. 594 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 ISSN: 2320-5407 ISSN: 2320-5407 ISSN: 2320-5407 The issue of the Jewish labor was a topic of severe struggles. Moreover, the meaning of immigration in this context was the incessant entry of workers who were inexperienced and who therefore were expensive in the employers’ opinion, into the job market. In other words, the Zionist objective itself – to encourage as much immigration as possible, at a pace and scope not necessarily dictated by economic considerations – influenced the economic situation. Therefore, the question of the economic regulation of the immigration arose: to create a balance between the contribution of the immigration in terms of the human capital and the need for the investments of economic capital that would facilitate the absorption of the human capital and the production of benefit from it. A third example of the economic weakpoint of the realization of the Zionist ideologies is the question of the investments in defense and security, created as a result of the national conflict on the Jewish settlement in the Land of Israel. These investments were intended to establish, arm, and train military organizations, and from the capital invested directly in defense it is possible to note on the loss side the expense of the workforce from the productive work circle. The investments in security steadily increased over the years, without there being any profit. In essence, such a profit was produced only with the development of the arms industry in the State of Israel. Since the choice of the regions of settlement were also influenced by the national and security considerations, according to the availability of the lands for settlement in the period of the British Mandate, settlements were founded in distant regions, without any economic justification for their establishment. The tension between the national imperatives and the economic imperatives originated, therefore, in the placement of the economic activity in the service of the ideological objectives. The economic activity, which by nature was supposed to be held in the economic markets according to considerations of cost-benefit and supply-demand, was subordinated to ideological imperatives that did not take into account economic needs, under conditions that were definitely not financial6. Kimmerling7 maintains that the ideological solution to this tension was given through the development of a pioneering ethos that emphasized the sacrifice of the individual for the good of the collective and for a better future. 6Fischer, S. (1987). "The Israeli Stabilization Program, 1985-86". The American Economic Review. American Economic Association. 77 (2) 7Kimmerling, B. (1993). Theory and Criticism of Militarism in Israeli Society, Fall, pp. 4, 123-140. 8Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in Hebrew) (Eds.) ISSN: 2320-5407 The economic activity that was guided by the national imperatives was therefore ‘pioneering economics’. Already in 1927, Haim Orlozorov defined this type of economic activity as ‘heroic economics’, economics in which the economic institution does not act for purely profit goals but undertakes the burden of the construction of the national home. In essence, Orlozorov intended to define a new type of economic adventure and criticized the economic activity of the Histadrut (the Workers Union) in the first half of the 1920s. Since then the Histadrut changed some of its ways of economic activity, although not radically. For many years, even after the establishment of the State, the national-ideological imperatives continued to dictate the economic ways of actions, for instance, the opening of the country to unrestricted immigration and the assignment of the responsibility to the State to absorb the immigrants. Last, the pioneering ethos had an additional implication, which originates in the norms of ‘making do with a little’.These norms dictated a simple life style and the avoidance of conspicuous consumption, even when the individuals had the means for this8. The Struggle between Socialism and Capitalism in the EconomicActivity in Israel:- The recognition of the national importance of an economy was shared by all the components of the political spectrum in the Jewish settlement in the Land of Israel and the argument about it revolved around the question of the appropriate way of developing it. The main outlooks in this argument were the socialist outlook and the capitalist- liberal outlook, although the camps that held these outlooks were not uniform. ISSN: 2320-5407 In the preface to his book on the economic side of the realization of the Zionist ideology, Kimmerling4notes three examples of the lack of economic worthwhileness. One significant example is the issue of the purchase of land. As Kimmerling shows elsewhere, in the years 1910-1944 the prices of land in Israel soared and were extremely high, and moreover, the moment lands were transferred to Jewish hands, they became national lands and could not be sold. The tremendous investment in the purchase of land therefore had political benefit but no economic profit. Another example is the cost of labor. Most of the immigrants came without any training for productive work but the primary available work in the first decades of the Zionist enterprise was manual labor – in agriculture, in the preparation of infrastructures, or in craftsmanship. However, many of the immigrants aspired to engage in productive work out of purely ideological motives but their output was low and it remained that way for a considerableperiod of time, until they acquired training and experience. In contrast, the standard of living to which they were accustomed in their countries of origin was high in comparison to the standard of living of the simple workers in the Land of Israel in this period. Therefore, they demanded a relatively high wage. The available solution for employers was to turn to cheap Arab labor, but this solution was not commensurate with the national ideology5. 595 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 The Struggle between Socialism and Capitalism in the EconomicActivity in Israel:- The socialist outlook was represented by the Labor Movement, when the main parties in it were HaPoelHaTsair (the Young Worker), which was not avowedly socialist, and AchdutHaAvoda (Unity of Work), which over the years 596 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 ISSN: 2320-5407 ISSN: 2320-5407 Int. J. Adv. Res. 4(10), 594-600 became more moderate in the socialist enthusiasm that characterized it at its beginning. The question of the extent to which the Israeli Labor Movement was a socialist movement is at the center of the dispute among the researchers of the history of Israeli society. Some maintain that the Labor Movement stopped being socialist already in an early stage of its existence. However, the economic regime that formed when its leaders were the leaders of the Jewish community in the land of Israel was very centralized. While this centralization was a relic of the socialistic perception of economic-political management, this trait of the Israeli economy indicated the desire to accumulate power, more than ideological adherence. Others hold that the Labor Movement remained socialist to some degree or another, but along with the socialist paradigm it adopted the national paradigm. The contemporary rhetoric indicates without a doubt that the heads of the movement and its activists were convinced that their outlooks and deeds were socialist outlooks and deeds. However, research studies show that the Labor Movement failed in its attempts to implement socialist principles such as equality, class solidarity, and social justice9. The liberal capitalist outlook was represented by three political camps – the general Zionists, the Mizrachi movement, and the Revisionist movement. The ‘standard bearer’ of this outlook was the general Zionist camp, which saw it to be the main component in its ideology. This political camp represented most of the private entrepreneurs and capital owners in the Jewish community. In the framework of the discussions on the ideological attitudes of the general Zionists there are two approaches, one conservative and one liberal. The conservative approach adopted uncompromising capitalist attitudes, while the liberal approach adopted more moderate capitalist attitudes, which recognized the national need for centralized leadership of the Zionist economy and cooperation with the middle class and the labor movement. For the Mizrachi movement and the Revisionist movement, the capitalist outlook was a component in the ideology but not the main one. 9Bank of Israel (2012). "Israel's International Investment Position (IIP), June 2012". September 19, 2012. Retrieved October 15, 2012. 10Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in Hebrew) The Struggle between Socialism and Capitalism in the EconomicActivity in Israel:- However, in the framework of the political rivalry between it and the labor movement, the Revisionist movement emphasized its position for a free economy and against worker struggles and strikes. Therefore, in the ideological and rhetorical dimension socialism and capitalism existed as two different and even rivalling social-economic outlooks. The political power was in the hands of the Labor Movement, which saw itself as having socialist outlooks and as having a large part of the economic activity in its hands, but the economic power was primarily in the hands of private entrepreneurs who held the liberal capitalist outlook. However, in both sides there were factors that necessarily recognized the need to cooperate so as to achieve the national goals and therefore recognize the legitimacy of the two outlooks and the inability to force on the Jewish community a comprehensive and inclusive social-economic outlook of any type. y yp Thus, two sectors of economic activity existed side by side – the public sector, operated by the Labor Movement, as the dominant political organization in the Jewish community, and the private sector. Each one of the sectors held its own outlook regarding the ideal model of economic activity, but in actuality the private entrepreneurs, who had the economic power, agreed to the centralized management of the community economy by the political leadership while the Labor Movement, with the political power, agreed to the existence of private initiative and a free economy market outside of its own sector. The model of economic activity in the Jewish community was not a socialist or capitalist model, but a mixed model, and the faithful expression of this model was the dual sectorial structure of the economy10. The Dual Structure of the Economy – Private Capital and Public Capital:- The Dual Structure of the Economy – Private Capital and Public Capital:- The division of the capital that was brought to the Land of Israel into public-national capital and private capital created the structural division of the economy of the Jewish society to the public-national13 sector and to the private sector14. These two sectors are different not only in size – as aforementioned, the private sector was always considerably larger than the public-national sector – but also in the directions of their activity, the economic perceptions that characterized them, and in the degree of political power that each one relied upon, since the private sector did not succeed in translating its economic power into political power. These two sectors existed side by side and recognized the importance of the other sector, primarily after the Labor Movement became the dominant political organization in the national institutions at the beginning of the 1930s. The division into the two sectors exists today as well, but it has experienced changes. After the establishment of the State of Israel, the country became owners of a large part of the sector. First, the State appropriated the lands of Arabs who had fled or been expelled during the War of Independence, the lands of the Mandate government, and the lands of the JNF. Second, with the establishment of a mechanism of bureaucratic control the State became the largest employer. In the first decades after the establishment of the State, the public-national sector greatly increased. However, afterwards public property was sold only to the private sector. The import of capital remained a main characteristic of the Israeli economy, although the dominance of one-sided capital from global Jewry15was replaced by the dominance of two-sided capital from other countries and international organizations, through the State, and private investors. The division into two sectors expresses not only the two types of ownership of capital and property but also two separate perceptions of economic activity. The Zionist movement held the pluralistic perception in all that pertains to economic activity, but one of the implications of the argument between Weizmann and Brandeis was the preference of the national and public capital and the centralized management of the economic activity since the private capital was less subordinate than the public national capital to the good of the nation. Import of Capital:- When the products of the communities of the First Immigration to the Land of Israel are examined in the terms of this discussion, it is possible to say that the economic and financial hardships are related largely to the attempt to realize national goals through economic activity that makes use of private capital and relies on considerations of economic feasibility, without taking into account the tension between the market imperatives and the national imperatives.While in its first years the Zionist movement did not form an unequivocal position on economic questions, immediately with the establishment of the World Zionist Organization it began an institutional effort to 597 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 ISSN: 2320-5407 ISSN: 2320-5407 recruit capital from worldwide Jewry. This capital, which was consideredthe capital of the national movement, was used to finance the achievement of the national goals, alongside the private capital11. recruit capital from worldwide Jewry. This capital, which was consideredthe capital of the national movement, was used to finance the achievement of the national goals, alongside the private capital11. There is a distinction between national capital, which was recruited using the organizations of the World Zionist Organization, primarily the Jewish National Fund (JNF, or in Hebrew Keren KayemetLeYisrael, KKL) and the United Israel Appeal (Keren HaYesod), as well as organizations like Hadassah, WIZO, and Aid Fund for the Land of Israel, and public capital, which was recruited through non-Zionist organizations, primarily PICA (Palestine Jewish Colonization Association), the Joint, and KAPAI (Hebrew acronym for the Fund for the Workers of the Land of Israel). However, in the summative presentation of the data of the capital, they are addressedgenerally as one unit – national and public capital. In the period of the Ottoman government and the British military government, there was the allocation of the entire national capital and of some of the public capital by the Israeli Land Office and then by the Zionist Executive in Jerusalem and then by the management of the Jewish Agency, which was established primarily to recruit funds from Jews around the world. The allocation of public capital was undertaken by political organizations that recruited this capital, for instance, the Labor Movement, which commanded KAPAI. 11Benchimol, J. (2016). "Money and Monetary Policy in Israel during the Last Decade". Journal of Policy Modeling 38 (1): 103–124. 12Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in Hebrew) 13In essence comes from donations and thus is a one-sided import. 14Comes from investments and thus is two-way import. 15Through the State and the Jewish Agency Import of Capital:- The private capital was brought by immigrants with capital, and the Mandate government even encouraged their coming so as to increase the investments in the economic development of the Land of Israel. The rate of private capital brought in the period of the pre-State settlement of the Land of Israel was always considerable larger than the rate of the public-national capital12. 14Comes from investments and thus is two-way import. 15 15Through the State and the Jewish Agency 13In essence comes from donations and thus is a one-sided import. 14 Politicization of the Economic Activity:- y The roots of the politicization of the economic activity can be found in two factors. The first is the main role of the Zionist Executive and the management of the Jewish Agency in the allocation of the capital recruited from the Jews around the world and in the allocation of the immigration permits. The second is the entry of parties into economic activity. As aforementioned, the management of the economy by the national movement was a lesson learned following the economic hardships of the communities of the First Immigration and the necessity to enforce ‘national logic’ on the economic activity, to achieved national-economic goals. However, the Zionist Executive and the Jewish Agency management were political organizations whose members represented the political-ideological groups in Zionism and saw their status, as responsible for the allocation of economic resources, as a position of political power. Thus, there was an opening not only for the dominance of national considerations over the economic activity but also for the development of the power struggles and for the transformation of the recruitment and allocation of financial resources into a mechanism of the implementation of authority by the national institutions as a partial substitute for the absence of sovereignty. The economic activity became a field of struggle between the political- ideological groups and the between them and the national institutions. In this struggle, the Labor Movement had advantages from the beginning, in comparison to the other groups, since the parties of the Labor Movement began their activity already in the first decade of the 20th century and were better organized than every other political-ideological in the community. Because of the circumstances of their appearance they fit into a wide variety of areas of activity, including economic activity, and established an elaborate institutional structure, the significant expression of which was the mechanism of the Histadrut. Because of this institutional structure, the parties of the Labor Movement could absorb new members and ensure their loyalty. In addition, their leaders were possessed of political awareness, initiative, and drive and could control and direct the party members. Therefore, the institutions of the Labor Movement could act considerably in the economic field, for the achievement of the national goals. ISSN: 2320-5407 Int. J. Adv. Res. 4(10), 594-600 ISSN: 2320-5407 foremost by considerations of profit and loss, as opposed to the public and national capital, for which the considerations in the investment were secondary. The political dominance of the Labor Movement tilted the scales even further in favor of the public national capital and the centralized perception. This perception dominated the management of Israel economy for many years and its structural implications are still manifest today16. 16Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in Hebrew) 17Such as Arthur Rupin and Haim Weizmann. 18Melnick, R., &Mealemen, Y. (2009). Israel's Economy: 1986–2008, June 2009. & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel The Dual Structure of the Economy – Private Capital and Public Capital:- Even when the private capital was directed to investments commensurate with the national interests, the use of it was dictated first and 598 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 ISSN: 2320-5407 17Such as Arthur Rupin and Haim Weizmann. 18Melnick, R., &Mealemen, Y. (2009). Israel's Economy: 1986–2008, June 2009. Politicization of the Economic Activity:- Moreover, the parties of the Labor Movement were perceived by themselves and by others prominent in the Zionist Movement leadership17 as a primary power that could realize the national goals and therefore as deserving of support. In contrast to these parties, the political and organizational weakness of the organizations of the political center and the right’s excessive focus on the political on the political dimension were very prominent. The preferred position of the Labor Movement was expressed in the respectable segment of the public-national capital and immigration licenses that it received, and these enabled it to develop its own economic sector and to influence the development of the economy of the entire Jewish community. The desire to obtain resources from the Zionist Executive led eventually the parties of the Labor Movement to focus their efforts on achieving a majority in the Zionist Executive. Their victory in the elections to the management of the World Zionist Organization and to the management of the Jewish Agency in the first half of the 1930s established the position of their hegemony not only in the field of the determination of the political stances of the Jewish settlement, but also in all that pertained to the allocation of the economic resources.From this time onward, the Labor Movement became the organization in the Jewish community in the land of Israel that most influenced the establishment of a centralized and collectivist economic policy18. As aforementioned, the second factor of the politicization of the economy of the Jewish settlement was the economic activity of the parties of the Labor Movement. The roots of this activity should be sought in the period of the growth of these parties and in the tradition of the political culture influencing them. At first the parties of the 599 ISSN: 2320-5407 ISSN: 2320-5407 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Int. J. Adv. Res. 4(10), 594-600 Labor Movement and the non-party worker organizations supplied the material lacks of their members, and then the economic activity of the Youth Movement was formed in the framework of the Histadrut, through the development of existing economic institutions19 and through the establishment of other economic institutions20. All these institutions were collected in the framework of the Workers’ Company and over the years this became an economic giant. Politicization of the Economic Activity:- The extensive economic activity of the Histadrutwas integrated in political activity and served as an instrument for the accumulation of political power, both in the period of pre-statehood settlement and in the period of the country21. The mix of politics and economics in the Workers’ Companyis presented as follows. It should be emphasized that one of the goals for the establishment of a company of workers was from the beginning to support the Histadrut economically and politically. In the Workers’ Company politics and economics were always been combined. The politicization and ‘nationalization’ of the economic activity has most important significance in the economic content world of the collective identity, since they can explain the considerable degree of intervention of the government in the State of Israel in the economic activity. In the Israel economy the public-national sector was constructed on a far larger scale than the private sector, so as to absorb immigration. However, consequently, the political weakness of the supporters of the private market caused the situation in which they were forced to accept the dictates of the State’s centralized economy, which saw the construction of the State economy to be the main issue and not the laws of the free economy such as supply and demand. p y 20HaPoalim Bank, HaSneh Insurance Company, Tnuva, KupatCholim (a healthcare system), cooperatives for production and services, factories and industrial plants, Funds for the recruitment of money, loans, and credit. 21The collectivist ideology of the heads of the Labor Movement is what dictated the establishment and management of economic organizations by the political organization and the control of the economic resources enabled them to make their movement the dominant one in the Jewish community in the Land of Israel. 22Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in 19Company for Public Works and the Mashbir. 22Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in Hebrew) (Eds.) 19Company for Public Works and the Mashbir. 20HaPoalim Bank, HaSneh Insurance Company, Tnuva, KupatCholim (a healthcare system), cooperatives for production and services, factories and industrial plants, Funds for the recruitment of money, loans, and credit. 21The collectivist ideology of the heads of the Labor Movement is what dictated the establishment and management of economic organizations by the political organization and the control of the economic resources enabled them to make their movement the dominant one in the Jewish community in the Land of Israel. 22Yaar, E., & Shavit, Z. (Eds.). (2001). Trends in Israeli Society (Vol. 1). Tel Aviv: Open University of Israel. (in Hebrew) (Eds.) 19Company for Public Works and the Mashbir. 20 Conclusion:- To conclude, the transition from settlement to State, which occurred without political and structural shocks, allowed the situation that reigned in the settlement economy to be copied to the state economy, although the economic roles of the national institutions were transferred to the government of Israel, and the power relations that exist between the centralized-national economic perception and the perception of the free market and non-government intervention to be preserved. The source of the tradition of intervention lies also in the socialist influence on the LaborMovement. The economic activity of the Labor Movement tended clearly to centralized planning and supervision, for instance, in the issue of the equalization of the salary or in questions of employment policy. Another expression of the socialist influence can be seen in the very use of economic power, for example the power built in the framework of a workers’ company, so as to fortify political interests and political hegemony22. 600

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.